Naunyn-Schmiedeberg´s Arch Pharmacol (2009) 379 (Suppl 1):1–100 DOI 10.1007/s00210-009-0404-1 50th Annual Meeting Deutsche Gesellschaft für Experimentelle und Klinische Pharmakologie und Toxikologie Mainz, March 10 – 12, 2009 Abstract Numbers Pharmacology 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26

Transport Growth Factors Molecular Receptor Pharmacology Adrenergic Mechanisms Serotoninergic Mechanisms Glutamatergic Mechanisms Purinergic Mechanisms Peptidergic Mechanisms, Opioids NO- and CO-mediated Mechanisms Oxidative Stress Heterotrimeric G-Proteins Small G-Proteins Cyclic Nucleotides Protein Kinases, Tyrosine Kinases Calcium Channel Modulators, Calcium, Calmodulin Potassium Channel Modulators Renin-Angiotensin-System Endocrine Pharmacology Immunopharmacology, Cytokines Cytostatics/Apoptosis Central Nervous System Heart Circulation Kidney/Urinary Bladder Pharmacokinetics and Drug Metabolism in Animals Miscellaneous

1 14 18 42 49 51 53 60 62 74 85 90 103 115 127 143 151 156 167 199 208 233 264 276

– 13 – 17 – 41 – 48 – 50 – 52 – 59 – 61 – 73 – 84 – 89 – 102 – 114 – 126 – 142 – 150 – 155 – 166 – 198 – 207 – 232 – 263 – 275 – 279 280 281 – 290a

Toxicology 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46

Drug Metabolizing Enzymes Drug Metabolism/Toxicokinetics Organ Toxicity Toxicological Aspects of Drugs Toxic Metals Toxicological Aspects of Natural Compounds Apoptosis DNA Damage DNA Repair Mutagenesis/Clastogenesis Carcinogenesis Signal Transduction Genomics/Proteomics Transport Mechanisms Oxidative Stress In Vivo/In Vitro Test Systems Screening Systems Food Toxicology Risk Assessment Regulatory Toxicology

291 302 312 336 341 346 356 362 371 386 391 396 411 415 421 428 436 443 454 456

– – – – – – – – – – – – – – – – – – –

458 468 478 481 494 502

– – – – – –

301 311 335 340 345 355 361 370 385 390 395 410 414 420 427 435 442 453 455 457

Clinical Pharmacology 47 48 49 50 51 52

Clinical Studies Pharmacogenomics and Pharmacoepidemiology Pharmacokinetics and PK-PD-Modelling Molecular Clinical Pharmacology and Disease Models Clinical Pharmacy Other Areas of Clinical Pharmacology

467 477 480 493 501 508

2

AUTHOR INDEX Figures refer to the abstract number Aasland D. 384 Abdel-Rahman M.S. 280 Abdelaziz A.M. 479 Abel J. 292, 293, 325, 410, 452 Abu-Taha I.H. 86 Abuharbeid S. 14 Ackermann F. 122 Ackermann H. 482 Ackermann S. 143 Adams V. 258 Adamska M. 296 Agouri S. 152 Aguilar-Bryan L. 144 Agundez J.A.G. 299 Ahles A. 26 Ahmad M. 326, 333 Ahnert-Hilger G. 93 Ahr H.-J. 440 Ahuja V. 316, 317, 318 Aigner A. 14 Akool E.S. 181 Aktories K. 85, 89, 96, 360, 361, 408, 409 Al Hadithy A.F.Y. 469 Al-Rawi M. 420 Alaraby M.A. 479 Alber J. 7 Albrecht A. 453 Albrecht C. 322, 327, 328, 427 Altenhöfer S. 62, 75, 76 Altevogt P. 279 Altis K. 462 Alves F. 147, 203 Ammer H. 61, 165 Amr D. 500 Andrae U. 413 Angioni C. 481, 482, 486, 506 Anke T. 179 Appel K.-E. 291, 391 Appenroth D. 221 Appl H. 215 Arand M. 294, 295, 296 Arend M. 493 Armbrust E. 324 Arndt P. 100 Arnold B. 394 Arsic M. 372 Art J. 62, 174, 175, 179 Artamonov N. 337 Artati A. 311 Assi A.A. 32, 280 Atanasov A.G. 285 Atzler D. 493 Auge D. 502 Aurbek N. 339 Aykac V.A. 458 Baarsma H.A. 22 Baba H.A. 258 Babu P.P. 224 Bach F.W. 341, 342 Bachmakov I. 491 Bachmann H.S. 468 Bachmann M. 177, 184 Bader M. 49 Baechler S. 422 Bäumer W. 192, 341, 342 Bakiri L. 195 Balda M.S. 229 Balszuweit F. 439 Balz V. 152 Bankstahl J.P. 232 Bankstahl M. 226, 227, 232 Baqi Y. 54 Baraka A.L. 159 Barbara A. 398 Barchanski A. 112 Barreto-Pereira H.F. 243 Barth H. 407, 416, 417, 418, 419 Bartholomé A. 401 Bartoszyk G. 198 Basara N. 460 Bashammakh S. 49 Batke M. 457 Bau M. 468 Bauer M. 377 Bauer T. 80 Baum M. 443 Baumann F. 466 Baumgärtner W. 229 Bayer M. 180 Bechmann I. 231 Beck K.-F. 63

Becker C. 109 Becker H. 377 Becker W. 115, 119, 503 Beermann A. 402 Beermann S. 191 Beetz N 46 Beetz N.B. 239 Behm C. 305, 448 Behrendt H. 319, 330 Beil W. 120 Bellinghausen I. 173 Bellmann B. 366 Belyi Y. 360 Beneke S. 371 Benndorf R.A. 461 Berberich N. 178 Berdelle N. 384 Berezin V. 212 Berger F.I. 443 Bergermann A. 351, 355 Bernauer U. 291 Bernshausen T. 307, 438 Bert B. 50 Besche V. 171, 174 Beuerlein K. 118, 176 Beyer M. 148 Beyer S.R. 3 Beyer T. 187 Beyersdorf F.B. 239 Biel M. 209 Bielek H.S. 405 Bien S. 13, 253 Bierbach U. 460 Binder U. 417 Birkner R. 383 Birmili B. 427 Birnbaumer L. 130, 141 Birod K. 481, 482, 486 Bitsch A. 457 Bitzer M. 358 Blaich A. 127, 248 Blaszkewicz M. 299, 305, 447 Bloch W. 71 Blömeke B. 297, 298, 399, 432 Blume U. 434 Bock E. 212 Bock H.H. 210 Bockamp E. 394 Bode-Böger S.M. 464 Boeckmann S. 206 Böddeker S. 355 Böer U.B. 225 Boege F. 274 Böger R.H. 461, 493 Böhm A. 193, 484 Boehm A. 17 Boehme K. 411 Böhn S.N. 307, 438 Boekhoff I. 122, 282 Börchers S. 164 Börner C. 167, 168 Bösch-Saadatmandi C. 324 Bohle R.M. 199 Boissel J.-P. 2, 3 Boivin V. 43 Boknik P. 260, 262 Bollmann B. 245 Bolt H.M. 387 Bonifas J. 298 Boosen M. 63 Boots A. 327, 328 Boots A.W. 322 Boronowski D. 45 Borowitzki G. 362 Borth H. 122, 282 Bosbach H. 349 Bouche E. 210 Bovensiepen C. 173 Braeter M. 277 Braeuning A. 301, 391 Brand E. 233, 235 Brand-Herrmann S.M. 233, 235 Brandes R.P. 189 Brandner S. 413 Brandt C. 226, 227 Braszko J.J. 275 Brauer-Dewor N. 395 Braun A. 247 Braun M. 290a Braunbeck T. 441 Brawek B. 52 Brecht B. 467, 472, 501

Breddemann A. 495 Breit A. , 60, 142 Breitenkamp A.F.S. 138 Brenneis C. 503 Brixius K. 71, 465 Broll H. 444 Bros M. 171, 172, 173, 174 Broschard T. 441 Brosnihan B.K. 463 Brouwers J.R.B.J. 469 Brown R. 366 Brückel B. 455 Brüning T. 362 Bruggeman R. 469 Brummer S. 126 Bryan J. 144 Bubik M.F. 165 Bucher M. 278 Buchmann A. 301, 358 Budde T. 253 Büch T.R.H. 142 Büchele B. 182, 183, 202, 205 Bührmann S. 495 Bünemann M. 88, 103, 107 Bünger J. 386 Bürkle A. 371, 372 Buhrke T. 445 Bujok V. 18 Bulashevska S. 224 Burhenne H. 112 Burkart J. 422 Burkhardt B. 449 Bussek A. 254 Buters J.T.M. 319, 330 Cabezas-Wallscheid N. 394 Camacho Londoño J.E. 128, 130, 141 Cambien F. 235 Carrier L. 236 Cascorbi I. 9, 10, 271 Cassee F.R. 328 Censarek P. 186 Chaitanya G.V. 224 Charron P. 236 Chatterjee S.S. 51 Chen X. 135 Chien K.R. 236 Chisari F. 199 Chourdakis M. 219 Chovolou Y. 348, 349, 350, 351, 352, 353, 354, 355, 400, 402, 403 Christ T. 244, 289 Christen U. 180 Christian F. 116 Christmann M. 357, 382, 383, 384 Chubanov V. 135, 136 Chudziak D. 396 Class P. 366 Clauss T. 478 Closhen D. 231 Closs E.I. 2, 3, 77 Constantin M.C. 359 Cordasic N. 493 Coste O. 115, 503 Creutzenberg O. 366 Cronin A. 294, 295, 296 Croxford A. 231 Csanády Gy.A. 310 Culman J. 217, 220 Culmsee C. 39, 74, 83, 84 Czubayko F. 14 Dai G. 166, 203, 274 Daiber A. 77, 78 Damm E. 60 Damm G. 303 Dangwal S. 266 Danowski N. 234 Dao V.T-V. 82 Dao V.T. 152 De Amici M. 21 Decker M. 40, 221 Decker M.M.E. 296 Dees C. 88 Degen G.H. 305, 344, 387, 447, 448 Dekant W. 456 Deml K.F. 28 Dempe J.S. 346 Deng S. 336 Desch M. 68 Deufel T. 221 Deuschle K. 95

3

Dhein S. 240, 241 Diabaté S. 329 Diel P. 344, 387 Diemert S. 84 Diesel B. 185 Dietrich A. 130, 132, 136, 141 Dietrich C. 369 Dietrich S. 330 Dietze A. 241 Dietze S. 50 Digweed M. 380 Dilanian G. 236 Dirsch V.M. 81, 207, 285 Dißman J. P. 156 Dittrich-Breiholz O. 118, 194 Dizayee S. 139 Djannatian J.R. 211 Do Thanh P. 286 Doberstein K. 279 Dobrev D. 245, 246 Doehring A. 471, 476, 477 Döll F. 121 Dördelmann C. 235 Döring B. 4 Dörrie A. 176 Dohna-Schwake C. 495 Dolga A. 74 Dolga A.M. 39 Doller A. 153, 181 Domes K. 127, 248 Dominiak P. 151, 155 Dorn S.B. 387 Dreger S.C. 93, 98 Dreikhausen U.E. 169 Drewes G. 385 Drews G. 144, 146, 150 Driessen J. 484 Drzewiecki K. 262 Dubrovska G. 136 Düfer M. 144, 146, 150 Dukart I. 416 Dullin C. 147, 203 Dupuis M. 247 Ebenfeld S.E. 508 Eberhardt K. 323 Eberhardt W. 153, 181 Eckers A. 404 Eckhardt A. 365 Eder A. 237 Egerer M. 408 Ehlers A. 444 Eich M. 380 Eichele K 204 Eichwald V. 394 Eisel U.L.M. 39 Eisele T. 41 Eisenbrand G. 443 Eisinger D.A. 61 El-Armouche A. 237, 261, 262 Ellrich J 55 Eloranta J.J. 490 Elsenhans B. 323 Emmler T. 288 Enderlein J. 41 Endig S. 244 Engelhardt S. 26, 242, 247, 256 Ennen J. 287 Enzenmüller S. 419 Enzmann H. 15, 16 Epe B. 363, 364, 367, 368, 385 Erdorf M. 56 Erkel G. 179 Ermler S. 17, 484 Eschenhagen T. 236, 237, 261 Escher S. 457 Eshkind L. 394 Esselen M. 347 Esser C. 312, 313, 314 Essin K. 136 Fabian E. 307, 437 Fabian E.J. 438 Fabritz L. 262 Fährmann M. 120 Fahrer J. 417, 419 Fang L. 40, 221 Faust D. 369 Fazeli G. 423 Fehr M. 422, 450 Feil R. 68, 108, 109, 269 Feil S. 68, 108, 109, 269 Feistel B. 161 Feld J. 443 Feldmann I. 393 Fernandes M. 430 Ferrario C.M. 463 Feser K. 342

Fester I. 85, 89 Fésüs G. 136 Feuerstein T.J . 52, 214 Filser J.G. 310, 311 Fink H. 50, 218 Finol-Urdaneta R.K. 145 Fischer J. 233 Fischer J.W. 17, 154, 166, 203, 264, 270, 274, 484 Fischer S. 127 Fischer W. 57 Fischer-Posovszky P. 30 Fleck C. 221 Flockerzi V. 128, 130, 141, 238 Flockerzie K. 147 Föllmann W. 305, 448 Förster A. 288 Förster I. 292, 322 Förstermann U. 69, 73, 75, 76, 77, 78 Fokkens P. 328 Fortmueller K. 315 Fortmüller L. 262 Foth H. 326, 333 Frank S. 156 Franke H. 57 Fraunholz M. 11, 12 Frebel K. 42 Freeman D.J. 5 Freichel M. 128, 130, 141, 238 Freidel K. 186 Freiman T. 214 Freiman T.M. 52 French R.J. 145 Frenker J. 249 Freudenberger J. 308 Freudenberger T. 166, 203, 264 Frey U.H. 234 Freyberger A. 440 Freynhagen R. 471, 477 Fricker G. 8 Friebe A. 66, 70 Friedel C. 13 Friedrich F. 236 Friedrich S. 67 Fritsche E. 293, 312, 325, 452 Fritz G. 98, 201, 340, 356, 379, 392, 406 Fritz J. 347 Frobel A.-K. 496 Fromm M. F. 470, 489, 491, 502 Fromme H. 330, 451 Frotscher M. 210 Frühmesser A. 321 Fruehwald J. 134 Fürst R. 165, 178 Fuest C. 212, 229 Fuhrmann R. 15, 16 Fußer M. 368, 385 Gäbler R. 71 Gaestel M. 98 Galal O. 160 Gamer A.O. 331 Garcia-Martin E. 299 Gasnier B. 2 Gassmann K. 452 Gauer S. 153 Geduhn J. 105, 106, 110 Gehrke H. 321, 345 Geiger H. 153 Geisslinger G. 115, 198, 462, 471, 476, 477, 481, 482, 483, 486, 503, 505, 506 Gekle M. 260, 263 Genisyürek S. 408 Genth H. 91, 92, 94, 98 Gerberick G.F. 454 Gergs U. 249, 257, 258, 260, 263, 478 Gerhard R. 97, 99 Gerlach M. 337 Gerling A. 108, 109 Gerloff K. 322, 327, 328 Gerlofs-Nijland M. 328 Gerstmayr N. 365 Geyer J. 4 Ghaly H. 163 Ghoneim M.T. 159 Giegold O. 188 Gier B. 144 Giera S. 301 Gierschik P. 30, 95, 100, 101, 102 Giese B. 11, 12 Gilsbach R 46, 243 Glänzel M. 54 Glaeser H. 470, 489, 502 Glahn F. 333 Glaschik S. 95 Glatz T. 220 Glaubitz C. 486 Gminski R. 389, 455 Gnana Oli R. 424

Gnügge R. 108 Goebel C. 454 Göbel P. 256 Göckler N. 119 Gödtel-Armbrust U. 336, 470, 488 Göthert M. 275 Göttle M. 105 Göttlicher M. 359 Gogakos A. 219 Gohlke P. 220 Golka K. 299, 305 Gollasch M. 136 Golombek M. 169 Gonder S. 334, 436 Goren I. 156 Gori T. 78 Gorny M. 497 Goschke A. 20 Gosens R. 22 Grabbe S. 171, 172, 173, 174 Graff J. 476 Grandoch M. 18 Grasmück T. 350 Graziani A. 131 Grebhardt S. 6 Gregor I. 41 Grether-Beck S. 410 Griesbach R. 269 Griessinger N. 471, 477 Grittner D. 149, 265 Grösch S. 481, 482, 486 Gröticke I. 213 Grohm J. 83, 84 Groneberg D. 66 Groschner K. 131 Grosser G. 4 Großmann C. 260, 263 Grube M. 11, 12, 13, 507 Gruhler A. 412 Gschrei C. 421 Gudermann T. 60, 122, 132, 135, 136, 142, 282 Gülden M. 324 Guenther B. 281 Günther G. 428, 429 Günther J. 47 Günther S. 258 Guttenberg G. 408 Gutwein P. 279 Haag R. 1, 284 Haarmann-Stemmann T. 292 Haasbach E. 223 Haberger M. 413 Habermeier A. 2, 3, 77 Haberthür D. 147 Haberzettl R. 50 Hadamek K 46 Haefeli W.E. 6 Haenisch S. 9, 10, 271 Hagedorn C. 235 Hagedorn I. 264 Hagen A. 240, 241 Hagenbusch J. 14 Hagn M. 187 Halayko A.J. 22 Hallier E. 386 Halpert J.R. 487 Halwachs S. 415 Hambrock A. 143 Hamscher G. 446 Hansen A. 237 Hansen F. 237 Hansen J.B. 149, 265 Harder S. 476 Harders A. 333 Harrison R. 427 Hartig C. 406 Hartmannsgruber R. 84 Hasenkamp S. 235 Hatlapatka K. 157 Hatzfeld M. 257 Haubold M. 44 Hauck P. 43 Haunhorst E. 446 Hauptmann S. 257, 258 Hausmann R. 41 Hausmann R.H. 58 Heck R. 394 Heijink I.H. 22 Heil S. 62 Heim H.-K. 166 Heimes K. 161 Hein L. 44, 46, 243 Hein L.H. 239 Heine K. 371, 407 Heinrich A. 439 Heinrich A.H. 225 Heinrich Gräfe U. 472, 474, 499, 501

4

Heise T. 428, 429 Heiss A. 61 Heiss E. 285 Heiss E.H. 81 Helbig L. 379 Held B. 238 Hellack B. 427 Hengstler J.G. 428, 429 Hennen J. 297 Henry M. 129 Hensel M. 419 Herberg F.W. 116 Herbst E. 324 Herbst K. 397 Herlyn H. 487 Hermes M. 111, 276 Herold M. 437 Herrero San Juan M. 189 Herrmann S. 255 Herz J. 210 Herzig S. 138, 139, 140, 283 Herzog M. 392 Hesse S. 467 Hessel S. 300, 306 Hettwer M. 430 Hewitt P. 412 Hewitt P.G. 315, 414 Hickel R. 302, 309 Hieke B.M. 114 Hildebrand A. 304 Hildebrand D. 361 Hildebrandt F. 132 Hilger H. 407 Hilgers K.F. 493 Hillenbrand M. 108 Hiller K.-A. 365 Hiller S. 143 Hilmenyuk T. 173 Hintermann E. 180, 181 Hintzsche H. 388 Hinz B 204, 206 Hippe H.-J. 86 Hippius M. 475 Ho M.S.P. 129 Höcherl K. 153, 278 Höfer N. 344 Hoehn M. 74 Höhn Y. 172 Hoek H.W. 469 Höllt V. 20, 23, 24, 27, 167, 168 Hölter S.M. 109 Höltje M. 93 Hörth K. 146 Hoffmann C. 19, 33, 36 Hoffmann K. 54, 213 Hoffmann L.S. 65 Hoffmann M. 486 Hoffmann R. 207 Hofinger A. 426 Hofmann F. 68, 70, 93, 125, 126, 127, 209, 248 Hofmann G. 478 Hofmann T. 135 Hohlfeld T. 31, 485 Hohmann N. 73 Hokema F. 466 Holdener M. 180 Holland S. 115 Hollender J. 432 Holmdahl R. 189 Holzberg D. 176 Holzgrabe U. 21, 37, 163 Homey B. 203 Hommers L.G. 88 Honig D. 308, 453 Honisch S. 120 Honscha K. 415 Honscha W. 415 Hopfer C. 421 Hoppstädter J. 185 Horke S. 75, 76 Horn A. 229 Horn K. 184 Howlett A. 32 Hsien L. 498 Huber E. 359 Hübenthal U. 293 Hübinger J. 312 Hübler N. 441 Hübner M. 110 Hülsenbeck J. 92, 98, 201, 340, 379, 406 Huener H.-A. 307 Hullin R. 140 Hummel M. 444 Hundsrucker C. 116 Hunfeld H.-P. 184 Hunt S.P. 217 Huppert J. 231 Hutter M. 347

Huwiler A. 35, 121, 190, 198, 288 Ibrahim S.S. 500 Idelevich E. 463, 467, 474 Ihlefeld D. 478 Iken M. 430 Iliev A.I. 211 Illes P. 57 Isnard R. 236 Jackisch R. 52 Jacob L. 24 Jacobi J. 493 Jacobsen C. 413 Jäger R. 104 Jäger S. 308, 390 Jaenecke I. 2 Jahns R. 43 Jahreis G. 453 Jahrsdörfer B. 187 Jakob H.G. 234 Jakubowski J.A. 484 Jakubowski N. 393 Janasch K. 203 Jangsangthong W. 140 Jansen T. 78 Janßen N. 38 Jastrow C. 388 Jedlitschky G. 11, 12, 507 Jentzsch C. 242 Jermann E. 427 Jess A. 324 Jessen S. 108 Joa H. 285 Jobi K. 79, 266 Jöst E. 356 John A. 306 Jones L.R. 258 Joshua- Tor L. 281 Jozwiak J. 241 Jürgens U.R. 48 Juhr D. 236 Jumpertz S. 40 Jung N. 80 Just I. 91, 92, 93, 94, 97, 98, 99 Jux B. 313 Kabagema C. 147 Kaber G. 273 Kadow S. 313 Käfferlein H.U. 362 Kaelble S. 281 Kämpf K. 337 Kästner L. 149 Kaestner S. 139 Kaever V. 112, 191 Kahl E. 27 Kahl R. 348, 349, 350, 351, 352, 353, 354, 355, 400, 402, 403 Kaina B. 356, 357, 369, 373, 374, 375, 376, 377, 378, 380, 381, 382, 383, 384 Kaiser E. 416, 418 Kaisers U. 466 Kalberlah F. 456 Kalmes M. 297, 399 Kalwa H. 132, 136, 142 Kammerer M. 52 Kamp H.G. 437, 438 Kampkötter A. 348, 349, 350, 351, 352, 353, 354, 355, 400, 402, 403 Kamuf J. 78 Kandulski A. 464 Kapp M.-D. 433 Karlberg A. 432 Kattner L. 261 Katus H. 237 Kaufel D. 37 Kaufmann M. 482 Kaufmann T. 438 Kebig A. 21, 38 Kehe K. 343, 439 Keim Y. 65 Keller S. 453 Kelm M. 71 Kendzia B. 362 Kern P. 454 Kessler W. 311 Kettner-Buhrow D. 176, 195, 281 Khaled H. 500 Khan I.F.Y. 140 Khandoga A.G. 178 Khobta A. 363, 364, 367 Kiedron O. 376 Kiehntopf M. 221 Kiemer A.K. 185, 199 Kietzmann M. 192, 287, 290a, 341, 342 Kilian B. 504 Kim M. 216

Kim R.B. 5 Kindla J. 491 Kirch W. 463, 467, 472, 474, 499, 501, 504 Kirch W.K. 458 Kirchhof P. 262 Kirkpatrick C.J. 343 Kitsera N. 364, 367 Klamp T. 186 Klein D. 310 Klein J.K. 222 Klein K. 111, 470, 490 Kleinbongard P. 71 Kleine-Ostmann T. 388 Kleinert H. 62, 174, 175, 179 Klenk J.C. 25, 36 Kleppisch T. 209 Kleuser B. 284 Klimpel C. 164 Kloor D. 111, 276 Klotz L.-O. 338, 401, 404 Klugbauer N. 123, 124, 137, 405 Kluge M. 459 Klussmann E. 116 Knaut M. 244 Knepel W.K. 225 Knobloch J. 338 Knop T. 214 Knüchel-Clarke R. 500 Knüpfer H. 492 Knust Z. 96 Kocgirli O. 152 Koch A. 35 Koch M. 438 Koch T. 20, 23, 24, 27 Kocoski V. 43 Köberle B. 373, 374 Köck K. 11, 12, 13, 507 Köhle C. 301, 442 Koenen A. 11, 12 König B. 105, 110 König J. 491, 502 König P. 66 Körholz D. 460 Körner R.K. 508 Koesling D. 66, 67, 104 Köster G. 87 Köstner K. 127 Kojda G. 72, 82, 152 Kolb M. 332 Konietzko J. 27 Korkowski M. 292 Kostenis E. 21 Kostopanagiotou G. 250 Kottenberg-Assenmacher E. 234 Kouvelas D. 219 Kovacs F.E. 208, 214 Kovalskyy D. 51 Kracht M. 118, 176, 194, 195, 281 Kraft M.E. 502 Kraiczy P. 184 Kraus J. 167, 168 Krause C. 341, 342 Krause E. 391 Kreimeyer I. 99 Krenn M. 131 Krennrich G. 437 Kriete K. 163 Krippeit-Drews P. 144, 146, 150 Krishtal O. 51 Kristiansen G. 279 Kröger A. 354 Kroemer H.K. 11, 12, 13, 253, 507 Kroll C. 418 Krombach F. 178 Kronich P. 195 Kruegel J. 117 Kruse F.E. 502 Kubatzky K. 361, 409 Kucerova D. 260 Kuczka K. 462 Küchler S. 1 Küper J. 45 Kuhlmann C.R.W. 231 Kuhn U.D. 475 Kullak-Ublick G.A. 490 Kumar S. 487 Kuntner C. 232 Kupis S. 483 Kurig B. 90 Kurz C.M. 47, 275 Kusche T. 258 Kuzmenkina E. 140 Kynast K. 483 Laakmann S. 262 Ladage D. 465 Lächelt S. 9, 10 Läer S. 473, 495, 496, 497, 498

5

Lambert S. 133 Lampen A. 300, 306, 429, 444, 445 Lamy E. 370 Lamyel F. 48 Landfester K. 183 Landsiedel R. 307, 331, 433, 438 Lang D.L. 222 Lang T. 470 Lange T. 460 Langer C. 309 Langer O. 232 Langer T. 102 Laszlo S. 276 Laub R. 467 Laws J. 495 Layh B. 255 Lazik A. 468 Lee E.-H. 362 Leemhuis J 46, 210 Leffke A. 444 Lehmann J. 221 Lehmann K. 445 Lehmann L. 167, 308, 390, 453 Lehmann S. 372 Leibold E. 437 Leierseder S. 242 Leiss V. 70 Leist M. 435 Lemke T. 248 Lemoine H . 45, 149, 265 Leurs R. 192 Li H. 69, 73, 77, 78, 368 Li Q. 310 Li X. 170 Lica M. 92 Lichey N. 386 Licht O. 457 Lichtmannegger J. 413 Limper C. 353, 355 Lindner S. 415 Linek B. 381 Lingg T. 363, 364 Link S. 238 Linke B. 503 Linker K. 62 Liu G. 469 Loch S. 133 Löscher W. 213, 226, 227, 232 Lösle M. 143 Lötsch J. 462, 471, 476, 477, 480, 482 Loga F. 126 Lohmann H. 254 Lohmüller B. 137 Lohse M.J. 19, 25, 26, 33, 36, 43, 103, 107, 251 Looser R. 437 Lopez J.J. 486 López-Antón N. 207 Loppnow H. 263 Lorenz K. 251 Lu Q. 69, 77 Ludwig A. 209, 255 Luebberstedt M. 444 Lüpertz R. 402, 403 Luhmann H.J. 231 Lukowski R. 70, 125, 126 Lunemann S. 314 Lunov O. 182, 183, 202 Lupp A. 221, 320, 475 Lutz S. 86, 87 Lv Q. 368 Ma-Hock L. 331 Maarsingh H. 64 Maas R. 461, 493 Mäthner F. 78 Mahran L. 479, 500 Mahringer A. 8 Maier J. 187 Maier T.J. 486 Malinowska B. 275 Mallmann R. 3 Mally A. 412 Mandery K. 489, 502 Mangelsdorf I. 457 Mangerich A. 372 Mangold W. 257 Mannebach S. 133, 134 Manns M.P. 180 Manthey I. 234 Marckscheffel A. 99 Marczynski B. 362 Marko D. 321, 345, 347, 422, 450 Markwardt F. 59 Marquardt C. 329 Martinez C. 299 Maser E. 324 Mathäs M. 488 Matroos G. 469

Matschke K. 245, 246 Matt L. 209 Matta M.M. 159 Matthes J. 139, 140, 283 Matthes S. 49 Matthias M. 253 Matus M. 42, 252 Maximyuk O. 51 May M. 94 Mayer B.A. 165, 178 Mayer P. 15, 16, 158, 166 Maywald U. 472, 474, 501 Mederos y Schnitzler M. 135, 136 Medvedeva Y.V. 216 Meerkamp K. 4 Meis K. 53, 196 Meissner M. 128, 130, 141, 238 Melber C. 457 Melching-Kollmuss S. 456 Melchior-Becker A. 154, 274 Mellert W. 437 Meloh J. 120 Mergia E. 67 Mering S. 158 Merkle S. 247 Mersch-Sundermann V. 370, 389, 455 Meszaros J. 139 Metzler M. 303, 304, 346, 449 Meurs H. 22, 64 Meyer D. 282 Meyer D.K. 123, 124, 210 Meyer P. 138 Meyer zu Heringdorf D. 197 Meyer zu Schwabedissen H.E. 5 Meyer-Kirchrath J. 31, 193 Michael C. 460 Michalakis S. 209 Mielke C. 274 Miesel A 151 Milde M. 113 Miller D.S. 8 Mintert E. 246 Miosge N. 117 Mitterecker F. 294 Mönkemüller K. 464 Moepps B. 30, 100, 101, 102 Möser C. 483 Mohr F.W. 240, 241 Mohr K. 21, 37, 38, 40 Mohr-Andrä M. 21, 38 Molderings G.J. 53, 196 Molkentin J.D. 128 Moosmang S. 127, 248 Morad S.A.F. 202, 205 Moreth K. 63 Mou T.C. 110 Mourouzis I. 250 Mückter H. 421 Mühl H. 177, 184 Mühlenstädt C. 93 Mühlstedt S. 247 Müller C.E. 54, 276 Müller C.J. 213 Müller C.M. 508 Müller D. 320, 475 Müller E. 283 Müller F.U. 42, 252, 259, 260, 262, 268, 272 Müller H 151, 155, 176 Müller O.J. 237 Müller R. 123, 124 Mueller S.O. 315, 411 Münzel T. 77, 78 Muhammad S. 223 Muñoz K. 447 Murr A. 307 Nagel D. 435 Nagel G. 382 Nagy N. 154 Napiwotzki J. 45 Nass R.D. 138 Nau H. 430 Naumann S.C. 376 Nem D. 488 Neuber C. 237 Neuhuber W. 147 Neumann D. 191 Neumann H. 464 Neumann J. 249, 257, 258, 260, 263, 478 Nevo-Koch E. 395 Nicken P. 446 Niederberger E. 483 Niederwieser D. 460 Niedorf F. 287, 341 Nienhaus G.U. 95 Nieß R. 479, 500 Niestroy J.L. 398 Nieswandt B. 247, 264

Nijholt I.M. 39 Nikolaev V.O. 43, 103, 107 Nikolova T. 376, 378, 381, 382 Nilius B. 128 Noack K. 150 Nörenberg W. 57 Noss A. 438 Nürnberg B. 90, 139, 264 Nürnberg P. 138 Numfor A. 393 Nunes F. 42 Oberemm A. 428, 429 Oberstadt M. 13 Oberwinkler J. 133, 134 Oeder S. 330 Oelze M. 77, 78 Oertel R. 504 Oertel R.O. 458 Oesch F. 369, 396 Oetjen E. 117, 164, 286 Oetjen E.O. 225 Oezguen N. 487 Özuyaman B. 71 Ohngemach S. 394 Oli R.G. 337 Olling A. 97 Oppermann M. 72, 152 Orth J. 85, 89 Orth J.H.C. 361 Osman B. 181 Osswald H. 276 Oßwald H. 111 Ostrau C. 201, 340 Oswald S. 10 Ott M. 430 Pahlke G. 422, 450 Panas A. 329 Panten U. 162 Pantos C. 250 Papadopoulos C. 119 Papatheodorou P. 408 Papritz M. 343 Paul M. 233, 235 Pautz A. 62, 174, 175, 179 Pekcec A. 212, 228, 229 Pelka J. 345 Perimenis P. 250 Pesch B. 362 Peters J. 234 Petzinger E. 7 Peuschel H. 410 Pexa K. 264 Pfeiffer E. 303, 304, 346, 449 Pfeiffer R. 293 Pfeilschifter J. 35, 63, 121, 153, 156, 177, 181, 184, 188, 189, 190, 197, 198, 279, 288 Pflöger A. 163 Pfreimer M. 100, 101 Philipp M. 136 Philipp S.E. 133, 134 Piée-Staffa A. 373, 374 Piekorz R. 139, 264 Pieper B. 148 Pierre S. 503 Pierre S.C. 115 Piesch C. 111, 276 Pietsch U.C. 466 Pigla U. 467 Pitterle K. 170, 202, 205 Planz O. 223 Platt K.L. 396 Platzek T. 317, 318 Pleskova M. 63 Plesnila N. 74 Plunien R. 417 Pönicke K. 478 Pohl C. 343 Polak J. 461 Polascheck N. 227 Pongratz S. 365 Popoff M. 407, 418 Poscher E. 203 Poteser M. 131 Potschka H. 212, 228, 229 Pott L. 246 Poulet C. 289 Pourzitaki Chr. 219 Prajwal P. 90 Preiss R. 460, 466, 492 Preuß I. 85, 89, 361 Prokoudine A. 437 Proksch P. 285, 348, 353, 354, 355, 400 Propping S. 277 Prutner W. 446 Pütz C. 311 Punkt K. 258

6

Pusch G. 319 Qiu H. 487, 488 Quack I. 67 Quante T. 195 Quell T. 433 Quiros S. 375, 376 Raasch W. 145, 151, 155 Racké K. 15, 16, 48, 158 Radeke H.H. 188, 189, 190, 197, 198 Radowski M. 1, 284 Radtke C. 93 Rady M. 500 Ramanujam D. 247 Ramer R 204, 206 Rankovic M. 23, 24 Rauch B.H. 17, 79, 193, 266, 267, 484, 485 Rauch K. 112 Rauch S.J. 193 Raulf-Heimsoth M. 362 Rauschkolb P. 62, 179 Rauwolf T. 244 Ravens U. 244, 245, 246, 254, 277, 289 Regenthal R. 466 Reiche I. 464 Reichl F.X. 302, 309 Reidenbach C. 71, 465 Reimann K. 338, 404 Reinauer C. 186 Reiner S. 33 Reinsberg L. 288 Reipschläger S.M. 409 Reitz M 55 Reitzenstein U. 15, 16, 158 Remmler C. 271 Resch P. 451 Reske-Kunz A.B. 171, 172, 173, 174 Rex A. 218 Richter C. 189 Richter K. 190 Richter N. 311 Ridder D.A. 224 Rieckeheer E. 71, 465 Riefenstahl A. 507 Riehle R. 111, 276 Riemann K. 200 Rignall B. 391 Rimmbach C. 253 Ristow M. 385 Rochais F. 26, 256 Röck K. 270 Röcker C. 95 Rödelsperger K. 455 Röhl C. 324 Rösen R. 80 Röser C 46 Roggenkamp D. 98 Rohde A. 206 Rohde J. 500 Rohr-Udilova N. 426 Rohrig R. 348, 351, 353, 355 Romanos M. 337 Rood A. 149, 265 Roos P.H. 393, 397, 398 Roos W.P. 356, 375, 376, 380 Rosen M. 270 Rosenau T. 426 Rosenkranz A.C. 17, 79, 193, 266, 267, 484, 485 Rosenthal W. 116 Rosenwald E. 358 Roßbach K. 192 Rosskopf D. 253 Rossner M.J . 230 Roth B.R. 508 Rothe H. 454 Ruckhäberle E. 482 Rudy A. 207 Rühmkorf A. 10 Rüßmann C. 233 Rüther U. 338 Ruhl S. 348, 351, 353, 355 Rump L.C. 67 Russwurm C. 104 Russwurm M. 104 Rustenbeck I. 157, 162, 163 Ruth P. 108, 147 Ryan C. 454 Rylski B.R. 239 Rzeczkowski K. 176 Sabisch K. 389 Sabri O. 467 Sacher B. 326 Sack M. 349 Sader R. 156 Sadik C.D. 177, 184 Sahin S. 163

Saklatvala J. 176 Salamat S. 476 Saloga J. 173 Sander K. 192 Sandner J. 481, 482 Sandrock K. 405 Saranteas T. 250 Sarioglu H. 413 Sass S. 372 Saur D. 125 Sausbier M. 147 Sausbier U. 147 Savani R.C. 203 Schaafsma D. 64 Schachner D. 81 Schäfer E.A.M. 142 Schäfer L. 63 Schaefer S. 65 Schäfer-Korting M. 1, 284 Schall C. 111 Scharffetter-Kochanek K. 182 Scharmach E. 300 Scheerle R.K. 346 Scheitza S. 298 Scherbart A. 327, 328 Scherl-Mostageer M. 500 Schiefelbein D. 156 Schiffmann S. 481, 482, 486, 506 Schildknecht S. 435 Schindler C. 463, 467, 474, 501 Schinner E. 68 Schins R.P.F. 322, 327, 328, 427 Schittny J. C. 147 Schlicker E. 47, 275, 290 Schlipp A. 43 Schlötzer-Schrehardt U. 502 Schlossmann J. 68, 125 Schlute J.S. 252 Schmalbach K. 390 Schmall A. 9 Schmalz G. 365 Schmalzing G. 41, 58, 59 Schmerwitz U.K. 178 Schmid K.W. 200, 468 Schmidt C. 278 Schmidt D. 473 Schmidt G. 96, 98, 405, 409 Schmidt H. 198, 481, 482, 503, 505, 506 Schmidt H.H.H.W. 65 Schmidt M. 18, 22, 132, 202, 476 Schmidt N. 62, 175, 179, 218 Schmidt P.M. 65 Schmidt R. 481 Schmidtko A. 462 Schmitt C.S. 414 Schmitt J. P. 251 Schmitteckert E. M. 251 Schmitz B. 233, 235 Schmitz J. 37 Schmitz R. 351, 355 Schmitz S. 370 Schmitz W. 42, 252, 259, 260, 262, 268, 272 Schnattinger C. 304 Schneider E. 29 Schneider E.H. 34 Schneider H. 118, 194 Schneider J 46 Schneider S. 396 Schneider T. 129 Schnell D. 29 Schnizler K. 216 Schober W. 319, 330 Schönegge A. 109 Scholich K. 115, 503 Schrader T. 388 Schreck I. 396 Schreiber T. 325 Schreier B. 260, 263 Schröder H. 20 Schroeder H.W.S. 13 Schröder J. 472, 474, 499, 501 Schröder M. 197, 198 Schrör K. 17, 31, 79, 186, 193, 266, 267, 273, 484, 485 Schroer K. 166 Schüssler P. 459 Schütz A. 49 Schuetz C. 315 Schütz E. 475 Schug M. 428, 429 Schughart K. 213 Schuhholz J. 100, 101 Schuhmacher S. 78 Schulte J.S. 42 Schulte K. 47 Schulz D. 14 Schulz E. 78 Schulz F. 91, 92 Schulz L. 384

Schulz M. 148, 433 Schulz N. 260 Schumacher S. 341 Schupp N. 423 Schwalm S. 121 Schwan C. 210 Schwaninger M. 223, 224, 230 Schwanstecher C. 148 Schwanstecher M. 148 Schwarz C. 450 Schwarz M. 301, 358, 391, 442 Schwarz U.I. 5 Schwedhelm E. 461, 493 Schweikl H. 365 Schwinger R.H.G. 71, 465 Scupp N. 424 Seeger T. 334, 436 Seehase S. 97 Seel S. 451 Seidel A. 306, 319, 396 Seidl M.D. 42, 252, 268, 272 Seidler U. 125 Seifert A. 20, 23 Seifert R. 28, 29, 34, 56, 105, 106, 110, 112, 169, 191, 215 Seiss M. 302, 309 Seithel A. 491 Seitz J.M. 341 Seitz O. 156 Selinski S. 299 Seshadri S. 461 Settels E. 291 Settmacher U. 475 Seymour A. 8 Shaat E.A. 159 Shi F.D. 199 Sickmann A. 256 Sieber A. 19 Sieber M. 412 Siebert J. 290a Sied U. 373, 374 Siegert J. 463, 467, 472, 474, 499, 501 Siegmund W. 10 Siepmann M.S. 458 Siepmann T.S. 458 Sies H. 401 Siffert W. 200, 234, 468 Sigl K. 68 Silber R.E. 257 Simetska N. 457 Simm A. 257 Simmet T. 170, 182, 183, 187, 202, 205 Simon S. 411 Sing V.J. 469 Sinzig J. 138 Sittl R. 471, 477 Skroblin P. 116 Smrcka A. 132 Snieder H. 469 Soares-da-Silva P. 228 Sobottka H. 57 Soerensen J. 228 Solinski H.J. 60 Sonnenburg A. 316, 431 Sontheimer K. 187 Sotoud H. 261 Spahn-Langguth H. 479, 500 Spanier G. 77 Spiessberger B. 125 Spiller C. 359 Spitzlei P. 53, 196 Staab A. 42 Stahl J. 287, 290a, 341 Stahl M. 276, 360 Stahlmann R. 316, 317, 318, 431 Stangl M. 310 Stark H. 192, 197, 198 Stasch J.P. 65 Staszewski S. 480 Stegbauer J. 67 Steger G. 186 Steiger A. 459 Steinberg P. 385, 395, 430, 434, 446 Steingräber A.K. 268, 272 Steinhilber D. 486 Stöck I. 220 Stoecklein N. 203 Stohr S. 142 Stolze K. 426 Stopper H. 337, 388, 423, 424 Storch U. 132, 135, 136 Storm D. 428, 429 Stosch C. 283 Strack S. 420 Straßer A. 28 Straub I. 133 Strauss V. 437 Strauß V. 331

7

Strobel J. 493 Struck H. 129 Stümpel F. 42, 259 Su Z. 102 Subkowiak E. 153 Suckrau I. 446 Sugidachi A. 484 Sullivan L.M. 461 Sur T.M.H. 252, 268, 272 Suvorava T. 72, 82, 152 Swidergall M. 352 Sydlik U. 410 Syrovets T. 170, 182, 183, 187, 202, 205 Szabo B 46, 208, 214 Szamel M. 169 Taeger D. 386 Taha H. 106 Tan E.M. 199 Tang T. 455 Tanner S. 352 Tatge H. 97, 99 Tauer U. 217 Taylor P.M. 3 Tegeder I. 462, 483 Teiber J.F. 75 Teichert J. 460 Telgmann R. 233, 235 Tenner K. 49 Terlau H. 145 Teschner C. 265 Tesseromatis C. 250 Thamsen M. 172 Theile D. 6 Thevis M. 387 Thiefes A. 176 Thierbach R. 385, 434 Thiermann H. 334, 339, 343, 436, 439 Thurmond R. 34 Tierling S. 199 Tietze C. 432 Tigges J. 293 Tirona R.G. 5 Tobaben S. 74 Todkar A. 269 Tönnies M. 291 Toliat M. 470 Tomicic M.T. 357, 384 Torabi S. 332 Torky A. 326, 333 Tot E. 495 Track N. 302 Tränkle C. 40 Trauner M. 490 Trausch A. 246 Treml M. 31 Treumann S. 331 Tröger U. 464 Trojandt S. 171, 173 Truisi G.L. 414 Twarock S. 203 Tybl E. 199 Uibel F. 442 Ulrych T. 485 Unfried K. 410 Unkrüer B. 229 Urbanski M 46 Urbanski M.J. 208, 214 Usachev Y.M. 216 Usanova S. 373, 374 Vahlsing S. 118 Valcaccia S. 82 Valtcheva N. 108 van Berlo D. 327, 328 van Harten P.N. 469 van Ravenzwaay B. 307, 331, 433, 437, 438 van Schooten F.J. 328 van Schouwenburg M.R. 64 Varga-Szabo D. 264 Vasan R.S. 461

Verlohner A. 307 Verspohl E.J. 160, 161 Vetter T. 25 Vierzig A.V. 508 Vilardaga J.P. 25 Villard E. 236 Völk H. 419 Vogt A. 87 Vogt W. 498 Voigt N. 245, 246 Vollmar A.M. 165, 178, 207 von Hayn K. 103, 107 von Hentig N. 471, 476, 477, 480 von Herrath M.G. 180 von Kügelgen I. 53, 54, 196 von Landenberg F. 441 Vormbrock I. 273 Vosshans B. 362 Wabitsch M. 30 Wätjen W. 348, 349, 350, 351, 352, 353, 354, 355, 400, 402, 403 Wagner E.F. 195 Wagner M. 490 Wagner T.F.J. 133 Waisman A. 231 Wait R. 176 Walijew A. 412 Walitza S. 337 Walk T. 437 Wallbach M. 117 Walliser C. 95 Walter J. 199 Walter Y. 315 Walther S.C. 451 Walther U.I. 332 Wang H. 182 Wang J. 472 Wang X. 20 Wang Y. 353 Wanner R. 316, 317, 318, 431 Ware J.A. 5 Warnke A. 337 Warnken M. 15, 16, 48, 158 Wasiliew P. 194 Waßermann L. 415 Weber A. 194 Weber A.-A. 186 Weber A.A. 18 Weber M. 82 Wegener J. 68 Wehmeyer A. 229 Weichenmeier I. 330 Weiergräber M. 129 Weighardt H. 292 Weigt J. 464 Weigt S. 441 Weindl G. 284 Weindl K. 109 Weinmeister P. 125 Weinshenker D 46 Weiss C. 329, 396, 420 Weiß D. 120 Weiss J. 6 Weiss M. 478 Weiß S. 243 Weissgerber P. 128 Welge P. 362 Welge-Lüßen U. 502 Welling A. 127, 248 Wenzel P. 78 Werner C. 125 Werner F. 263 Werner L. 230 Werthmann R.C. 103, 107 Wessa P. 438 Wessels A. 327, 328, 427 Wessels J. 147 Westphal G.A. 386 Wettwer E. 244,245, 254, 289 Wick G. 427 Widmayer P. 282

Wiechmann N. 171, 172, 173, 174 Wieker G. 200 Wieland S. 199 Wieland T. 86, 87, 101 Wielgat P. 275 Wiemer J. 437 Wiench K. 331 Wiese J. 333 Wilffert B. 469 Wilgenbus P. 75, 76 Wilhelm B. 122 Wilhelm M. 344 Wilhelmi V. 327 Willenborg M. 157, 162 Wilmes T. 123, 124 Wirth M.P. 277 Wirth S. 337 Witte I. 75, 76 Wittek F. 332 Wittig K. 87 Wittköpper K. 262 Wittmann H.J. 28 Wittsiepe J. 344 Wobst I. 481, 482 Woitkowiak C. 307 Wojnowski L. 336, 470, 487, 488 Wolf A. 118 Wolf N.B. 284 Wolf N.M. 86 Wolf P.A. 461 Wolfrum K. 385 Wolter S. 176, 195 Woltersdorf R. 59 Worek F. 334, 339, 436 Wormke M. 413 Wouters F.S. 211 Wu D.F. 20, 23 Wübbeke C. 343 Wuest M. 277 Wunder F. 113 Wurst W. 109 Xanthakis V. 461 Xia N. 69, 73, 77 Yakout D.Y. 159 Yan T. 336 Yang L.Q. 20 Yassouridis A. 459 Yatsenko N. 51 Ye Y. 43 Zabel U. 19, 36 Zaghloul A.S. 500 Zagon J. 444 Zahedi R. 256 Zaher H. 5 Zahler S. 165, 178 Zahner D. 7 Zakrzeska A. 275 Zanger U.M. 470 Zdzienicka M.Z. 376 Zeilinger K. 444 Zenn H.M. 116 Zettner M.A. 308 Zhang D.E. 394 Zhang Y. 40, 221 Zhao Y. 217, 220 Zheng W. 125 Ziebell S. 505 Ziegler N. 19 Ziemann C. 366 Zimmermann M. 471, 477 Zimmermann N. 262 Zimmermann S. 359 Zischka H. 425 Zitranski N. 122 Zolk O. 502 Zschauer T.-C. 452 Zünkler B.J. 163 Zürn A. 19, 36

8

1 Nanoparticles for skin penetration enhancement – a comparison of dendritic coremultishell nanotransporters and solid lipid nanoparticles Küchler S. (1), Radowski M. (2), Haag R. (2), Schäfer-Korting M. (1) Due to the lower risk of systemic side effects topical treatment of skin diseases appears favourable, but is often limited by poor skin penetration of the drug. Thus, nanoparticulate carrier systems (e.g. liposomes, lipid nanoparticles) gained interest and were proven to improve the dermal penetration. In particular solid lipid nanoparticles (SLN, 150 – 200 nm) appear very promising due to its good stability and biocompatibility. New types of nanocarriers are dendritic core-multishell (CMS) nanotransporters based on polymeric polyglycerol cores surrounded by a doublelayered shell consisting of a C18-alkyl chain and of monomethoxy poly(ethylene glycol). Above a critical concentration, single CMS nanotransporters with sizes about 6 to 9 nm aggregate and form larger particles with diameters of 20 to 50 nm that can encapsulate lipophilic as well as hydrophilic agents and transport them to polar as well as nonpolar environments. Aiming to unravel the influences of the nature and size of nanocarriers on dermal uptake of the loaded agent, we compared the skin penetration of the lipophilic model dye nile red when loaded onto CMS nanotransporters and SLN. Conventional base cream served as control. Furthermore we investigated the feasibility of these carrier systems for topical drug delivery of hydrophilic substances. Rhodamin B served as model drug. Data evaluation showed a significantly increased amount of the dyes in the stratum corneum (up to 8-fold), viable epidermis (up to 13-fold) and dermis (up to 10-fold) as compared to the conventional cream. Furthermore, CMS nanotransporters were clearly superior to SLN, respectively, for both lipophilic and hydrophilic dye. The dermal and ocular safety is an essential precondition for topical application. Thus, the EPISKIN® skin irritation and HET-CAM test were performed with SLN and CMS nanotransporters, respectively. Results predict no skin irritating potential according EU classification R38 and no eye irritating potential. In conclusion SLN and CMS nanotransporters are promising systems to increase effectiveness and tolerability of the local treatment of skin diseases. 1. Dept. of Pharmacology and Toxicology, Freie Universität, Berlin, Germany, 2. Dept. of Organic Chemistry and Biochemistry, Freie Universität, Berlin, Germany

2 Does SLC7A14 correspond to the lysosomal transporter for cationic amino acids named system c? Jaenecke I. (1), Boissel J.-P. (1), Habermeier A. (1), Gasnier B. (2), Closs E.I. (1) The solute carrier family SLC7 comprises two distantly related subfamilies of plasma membrane transporters, the cationic amino acid transporters (CATs) and the heteromeric amino acid transporters (HATs). We asked if SLC7A14, an orphan protein assigned to the SLC7 family due to sequence homology (42-45% identity with CATs and 16-20% with HATs), is also an amino acid transporter. Overexpression of SLC7A14 in Xenopus laevis oocytes or mammalian cells did not lead to any increase in the influx of cationic amino acids. Fusion proteins between SLC7A14 and EGFP localized to intracellular vesicles, co-staining with the lysosomal markers lysotracker and Lamp1 (lysosomal associated membrane protein). To test transport activity of SLC7A14, we tried to bring SLC7A14 to the plasma membrane by mutating putative lysosomal targeting motives. However, all our attempts were without success. We then created chimeras carrying either the backbone of CAT-1 or CAT-2 and the so-called “functional domain” of SLC7A14 corresponding to a protein stretch of 81 amino acids that determines the transport activity of the CAT proteins. The chimera with the CAT-2, but not the CAT-1 backbone exhibited transport activity, which was similar, but not identical to the transport activity of CAT-2A (the low affinity splice variant of the CAT-2 gene). Noticeably, arginine transport by the CAT-2/SLC7A14 chimera was pH-dependent, trans-stimulated and inhibited by epsilon-three methyl-L-lysine, properties assigned to the lysosomal transport system c in human skin fibroblasts. Expression analysis showed strong expression of SLC7A14 mRNA in these cells. Our results indicate that SLC7A14 may indeed correspond to lysosomal system c. Further characterization of this transporter is important as it provides a salvage pathway for export of cystine from lysosomes in patients with a defect in the lysosomal cystine transporter. 1. Dept. of Pharmacology, Johannes Gutenberg University, Mainz, Germany, 2. Institut de Biologie Physico-Chimique, Paris, France

3 Two cysteine residues in the human cationic amino acid transporter hCAT-2A are necessary for its inhibition by N-ethylmaleimide (NEM) Beyer S.R. (1), Mallmann R. (1), Habermeier A. (1), Taylor P.M. (2), Boissel J.P. (1), Closs E.I. (1) Three members of the cationic amino acid transporter family (CAT-1, CAT-2B and CAT3) exhibit transport properties of the so-called system y+, shown to be inhibited by a short term treatment (10 min) of human erythrocytes with the thiol-modifying agent Nethylmaleimide (NEM, 0.2 mM) (Deves et al., J Physiol 468: 753, 1993). In contrast, system y+L, also present in these cells, is insensitive to NEM inhibition. We first asked the question if sensitivity to NEM distinguishes generally all CAT and y+LAT isoforms. Transport assays in Xenopus laevis oocytes expressing each transporter individually, established that indeed all human CAT isoforms (including the low affinity hCAT-2A) are inhibited by NEM and that both y+LAT isoforms are NEM insensitive. Of the 13 cysteine residues in hCAT-2A (the isoform with the least cysteine residues amongst the CATs), 7 residues are conserved between all CATs. Individual mutation of each of these conserved residues to alanine did not lead to NEM insensitivity. However, three mutants (C33A, C85A, and C273A) needed higher NEM concentration for inhibition and/or were not completely inhibited even at the highest NEM concentration (1mM). Therefore, we made three double mutants, each carrying two of the three mutations at a time. Of these, only the C33A/C273A mutant was insensitive to NEM inhibition. However, this mutant had a very low transport activity. This suggests that the cysteine residues 33 and 273 are important for transport function and that their modification either by mutation or by NEM leads to loss of transport activity. We then constructed a mutant where all but the two cysteine residues 33 and 273 were mutated to alanine. This mutant exhibited a good transport activity and sensitivity to NEM. Taken together, our data identify the two cysteine residues 33 and 273 as important entities for the function of CAT proteins.

According to the 14 TM model, both residues are located at the intracellular side in protein regions so far not recognized as important for transport function. 1. Dep. of Pharmacology, Johannes Gutenberg University, Germany, 2. Division of Molecular Physiology, School of Life Sciences, University of Dundee, UK

4 Differences in the substrate pattern of the SLC10 carriers NTCP, ASBT, and SOAT Grosser G. (1), Döring B. (1), Meerkamp K. (1), Geyer J. (1) The solute carrier family SLC10 comprising seven members formerly was known as the family of “sodium bile acid cotransporters”. The founding member of this family, the Na+/taurocholate cotransporting polypeptide (NTCP; SLC10A1) is mainly expressed in the liver and transports bile acids, sulfoconjugated bile acids, sulfoconjugated steroid hormones, and a few further organic compounds. The apical sodium-dependent bile acid transporter (ASBT; SLC10A2) is expressed in the brush border membrane of ileocytes, in the apical domain of the proximal tubules in the kidney, and the apical membrane of cholangiocytes in the liver. Its substrate pattern is restricted to bile acids. Both transporters are important factors for the maintenance of the enterohepatic circulation of bile acids between the liver and the gut. We compared the substrate specificities and kinetic parameters of NTCP and ASBT with a novel member of the SLC10 carrier family, the sodium-dependent organic anion transporter (SOAT; SLC10A6). Although SOAT showed the highest phylogenetic relationship to ASBT, we found no transport activity for bile acids but a specific transport of the sulfoconjugated steroid hormones dehydroepiandrosterone-sulfate, estrone-3-sulfate, pregnenolonesulfate, and the sulfoconjugated bile acid taurolithocholic acid-3-sulfate. In contrast to the other transporters of the SLC10 family, SOAT is highly expressed in human testis and placenta and the SOAT promoter contains a highly conserved sequence for an androgen receptor binding site suggesting a hormonal dependent function of SOAT. In conclusion: Gene expression and functional properties are in general quite different between the SLC10 carriers NTCP, ASBT, and SOAT. However, the high structural and phylogenetic relationship, in particular between ASBT and SOAT, and on the other hand the clear differences in transport function requires further analysis of the structureactivity relationships and the location of the substrate binding pockets for bile acids and steroid sulfates in the respective carriers. 1. Institute of Pharmacology and Toxikology, Justus-Liebig-University of Gießen, Giessen, Germany

5 Targeted deletion of murine OATP1B2 transporter: loss of hepatic thyroid hormone uptake results in aberrant glucose homeostasis Meyer zu Schwabedissen H.E. (1,2), Ware J.A. (3,4), Zaher H. (3,5), Schwarz U.I. (2), Freeman D.J. (2), Tirona R.G. (2), Kim R.B. (2) Organic anion transporting polypeptide transporters (OATPs) mediate the cellular uptake of numerous endogenous and exogenous compounds. Oatp1b2 is the murine orthologue of human OATP1B1 and OATP1B3, hepatic transporters with broad substrate specificity which includes several drugs in clinical use, bile acids, and hormones. Recently we have shown that the Slco1b2-/- mouse is a useful tool to study the function of OATP1B1 and OATP1B3 in vivo. In order to more fully understand the physiologic relevance of these hepatic OATP1B transporters, we utilized the mouse model in which Slco1b2 had been disrupted. Absence of Oatp1b2 expression was confirmed by real-time PCR, Western blot analysis, and immunohistochemistry. AffyMetrix-chip based global mRNA expression analysis comparing wildtype and Slco1b2-/- mice revealed differential regulation of genes involved in glucose homeostasis in liver. Subsequent performed oral glucose tolerance testing showed a significant delay in the time to normalization of plasma glucose levels, elevated hepatic glycogen storage, and normal insulin levels in the Slco1b2-/- mice. We also noted a trend to higher plasma thyroxin (T4) and triiodothyronine (T3) levels in the Slco1b2-/- mice while hepatic expression of T4 and T3 target genes including 5’Deiodinase Type 1, Cyp7a1 and Phosphoenolcarboxykinase was paradoxically lower, thus suggesting an important role of Oatp1b2 in the hepatic uptake and function of T4 and T3. In fact, using cell based reporter assays we noted that activation of the thyroid hormone receptor (TR) was markedly greater in cells overexpressing Oatp1b2. In summary, we identified an unexpected phenotype in which failure of adequate hepatic thyroid hormone uptake mediated by Oatp1b2 resulted in elevated plasma glucose level due to reduced liver TR activation of genes involved in glucose metabolism. Our findings add important new insights to the role of transporters to hormone receptor activation and the etiology of chronic diseases such as diabetes. 1. Department of Pharmacology, Ernst-Moritz-Arndt-University, Greifswald, Germany, 2. Department of Medicine, University of Western Ontario, London, Ontario, Canada, 3. Pfizer Global Research and Development, Ann Arbor, Michigan, USA, 4. Genentech, Inc. San Francisco, California,USA , 5. Drug Discovery Support Boehringer Ingelheim Pharmaceuticals, Ridgefield, Connecticut, USA

6 Characterisation of the mechanism of synergistic action of FOLFOX chemotherapy Theile D. (1), Grebhardt S. (1), Haefeli W.E. (1), Weiss J. (1) FOLFOX is a cytostatic drug combination for adjuvant treatment of colorectal cancer (CRC) consisting of folinic acid (FA), fluorouracil (FU), and oxaliplatin (L-OHP). Several cellular mechanisms have been proposed as determinants of drug response to single LOHP or single FU such as glutathione (GSH) and metallothionein content, or expression of thymidylate synthase, respectively. However, a comprehensive understanding of the mechanisms leading to efficacy or failure of FOLFOX combination is still lacking. Drug transporters are also known modulators of drug efficacy and sites of drug interactions frequently leading to adverse drug effects. For FOLFOX little is known concerning the role of drug transporters and the impact of the L-OHP metabolite oxalate in this drug combination. We used LS180 cells as a CRC model and performed proliferation assays (determination of IC50) and qRT-PCR to elucidate the transcriptional and functional alterations accompanying the particular agent. 72 h preincubation with FU led to a

9

significant sensitisation towards L-OHP. Concurrently ATP7B, a copper-transporting Ptype ATPase and known efflux transporter of platinum (Pt) drugs, was down regulated by 50% and several multidrug resistance-associated proteins (MRPs) were induced (25-fold). To further clarify these findings, we determined the IC50 of FA, FU, the L-OHP metabolite oxalate, the respective remnant diaminocyclohexane-Pt (DACH-Pt), and in vitro reconstituted L-OHP in MDCK cells respectively overexpressing P-glycoprotein, breast cancer resistance protein, MRP1, MRP2, or MRP3. Combination index analysis revealed that overexpression of MRPs leads to synergistic effects of oxalate and its carrier DACH-Pt and might therefore be a beneficial factor for the efficacy of L-OHP containing cancer drug therapy. In addition, this study revealed that MRP overexpressing cells contain less GSH than the corresponding parental cell line and that GSH content clearly correlates with the IC50 of DACH-Pt. However, there was no such correlation when DACH-Pt plus oxalate (= reconstituted oxaliplatin) or oxaliplatin alone were applied. In conclusion, our results indicate a new mechanism of synergism of FOLFOX chemotherapy including reciprocal down- and up-regulations of transport proteins both favouring synergistic drug effects in FOLFOX chemotherapy. Additionally, oxalate positively modulated oxaliplatin’s action. 1. Dept. Internal Medicine VI, Clinical Pharmacology and Pharmacoepidemiology

to the expression under non-stimulated conditions was determined by quantitative rtPCR. Each experiment was conducted three times under identical conditions. Results: After IL1β incubation up to 12 hours, a time dependent increase of the ABCC2 mRNA expression was observed. The ratios compared to non-exposed cells was 2.59 ± 0.24 at 0.012 n/ml, 1.88 ± 0.39 at 1 ng/ ml, and 1.91 ± 0.16, at 10 ng/ml, respectively. Interestingly, after 48 hours of IL-1β incubation, the ABCC2 mRNA expression approximates again the baseline. No significant changes of the ABCC2 mRNA expression were detected under any concentration and at any time point for IL6, TNFα or LPS treatment in SK-Hep1 cells. Conclusions: These results in an in-vitro model suggest that IL1β, but not IL6, TNFα or LPS stimulates the mRNA expression of the ABC membrane transporter ABCC2 in the hepatic cell line SK-Hep1. The stimulation is strongly dependent on the duration and concentration of the inflammatory stimulus with a maximum at 12 hours and a further decline up to 48 hours, indicating a possible timedependent modulation of mitogen activated protein kinase pathways. 1. Dept. of Experimental and Clinical Pharmacology, UKSH, Campus Kiel, Germany

10 7 P-glycoprotein of the sheep cloning and functional expression in MDCK-cells Zahner D. (1), Alber J. (1), Petzinger E. (1) P-glycoprotein (P-gp) plays a crucial role in multidrug resistance of pathogenic helminths in sheep (Ovis aries) as well as in pharmacokinetics of antiparasitic drugs in the host. We cloned the cDNA of sheep P-gp and expressed it stably in MDCK cells. The open reading frame consists of 3858 base coding for a 1285 amino acid containing protein. The sequence shows great homology to the orthologues of other mammalian species, especially cattle. Both ruminant DNA sequences show a 9 bp insertion that was lacking in all other investigated sequences. Expressed in MDCK cells the protein shows a size of 170 kDa in Western analysis. Transfection of MDCK cell with sheep P-gp resulted in 10 to 50 fold resistance to cytotoxic P-gp substrates colchicin and daunorubicin and reduced digoxin accumulation. 1. Inst. of Pharmacology and Toxicology, Justus-Liebig-University, Gießen, Germany

8 Arylhydrocarbon receptor-dependent regulation of the ABC transporters BCRP, MRP2 and P-gp in kidney tubules from killifish (Fundulus heteroclitus) Mahringer A. (1,2), Seymour A. (1,3), Miller D.S. (1,3), Fricker G. (1,2) Today, humans are increasingly exposed towards environmental pollutants from herbicides, plastic materials, adhesives or toxic food contaminants from charbroiled meat or cigarette smoke. Halogenated and polycyclic aromatic hydrocarbons are classical ligands of the Arylhydrocarbon (Ah) receptor that plays a pivotal role in the induction of metabolizing enzymes, like cytochrome 1a1, 1a2 and 1b1, and hence, in a protective detoxification mechanism. However, the Ah receptor-dependent regulation of the efflux transporters Breast Cancer Resistance Protein (BCRP), Multidrug Resistance Protein 2 (MRP2) and P-glycoprotein (P-gp) is still not determined. Apart from their expression in other metabolic barriers like intestine, liver or blood-brain barrier BCRP, MRP2 and P-gp are expressed in the luminal membrane of kidney tubules, where they are – together with organic anion (OATs), cation (OCTs) and peptide transporters (PEPTs) - responsible for the excretion and reabsorption of nutritients and xenobiotics. Here, we show for the first time the Ah receptor-dependent regulation of BCRP, MRP2 and P-gp function and protein expression in kidney tubules from killifish (Fundulus heteroclitus). Incubation of freshly isolated intact tubules with 0.5µM β-naphthoflavone, a typical Ah receptor agonist, for 3h resulted in an increased efflux transport of the particular specific fluorescent substrates mitoxantrone (BCRP), fluorescein-methotrexate (MRP2) and NBD-Cyclosporine A (P-gp). Additionally, we determined an increase in the amount of transporter protein expression upon stimulation of the Ah receptor which was reversable by inhibition with the Ah receptor antagonist resveratrol. None of these effects was observed for OATs. 2,4’-DDT (2,4’-dichlorodiphenylchlorethane), used as a representative for chemical pollutants, mimicked the induction of transporter function and was abolished by incubation with resveratrol suggesting that insecticides might also function via the Ah receptor cascade. Finally, it remains to be determined whether MAPK, JNK or ERK might be involved in the Ah receptor-dependent signaling of ABC transporters in kidney tubules. 1. Mount Desert Island Biological Laboratory, Salisbury Cove, Maine, 2. Dept. of Pharmaceutical Technology and Biopharmacy, Ruprecht-Karls University Heidelberg, Germany, 3. NIEHS, National Institute of Environmental Health Sciences, Research Triangle Park, North Carolina

9 Time and concentration depending influence of IL1β, IL6, TNFα, and LPS on transcriptional gene regulation of ABCC2 in-vitro Haenisch S. (1), Lächelt S. (1), Schmall A. (1), Cascorbi I. (1) Background: The transmembrane transporter ABCC2 (MRP2) contributes to the hepatic, intestinal and renal efflux of conjugated and unconjugated endogenous as well as exogenous compounds. In the past decade several in vivo and in vitro studies reported inflammation-mediated changes in the expression and activity of membrane transporters which may contribute to alterations in the absorption, distribution and clearance of xenobiotics in inflammatory diseases. The observed effects however have not been completely consistent. Differences in mRNA versus protein expressions as well as species and tissue specific differences are described. Methods: To clarify the inflammatory progress on transcriptional level in a time and concentration dependent manner we conducted a systematically in vitro approach by stimulating SK-Hep1 cells with different concentrations of interleukin-1β (IL1β, 0.012 ng/ml, 1 ng/ml, 10 ng/ml), interleukin 6 (IL6, 0.8 ng/ml, 1 ng/ml, 10 ng/ml), tumor necrosis factor α (TNFα, 0.05 ng/ml, 1 ng/ml, 10 ng/ml) and lipopolysaccharide from E. coli) (LPS, 0.1 ng/ml, 0.5 ng/ml, 1 ng/ml, 2.5 ng/ml) up to 48 hours. At seven different time points (0, 1, 3, 6, 12, 24 and 48 hours) total RNA was isolated and the ABCC2 mRNA expression normalized

Influence of ABCC2 haplotypes on transcriptional and posttranscriptional gene regulation Lächelt S. (1), Haenisch S. (1), Rühmkorf A. (1), Oswald S. (2), Siegmund W. (2), Cascorbi I. (1) Background: The polymorphic multidrug resistance-associated protein 2 (MRP2/ ABCC2) serves as an efflux pump of conjugated and unconjugated organic anions and is widely expressed in barriers of mammalian tissues. There is upcoming evidence from in vitro and clinical studies that a C>T exchange at the position -24 in the 5´-UTR which is strongly linked with the silent SNP 3972C>T impairs the function of ABCC2. In contrast, the missense variant 1249G>A (Val417Ile) is discussed to increase the activity of the transporter. Methods: In order to proof the molecular background on transcriptional level, we firstly investigated by electromobility shift assays DNA-proteinbinding interactions using radioactive labelled probes containing either -24C or T allele and nuclear extracts of human liver (SK-Hep 1) and kidney (A498) cell lines. Secondly, the mRNA-stability of these cell lines heterozygote for -24C>T was tested by allelespecific quantitative real-time PCR and cDNA-specific pyrosequencing after treatment with the RNA-polymerase-inhibitor actinomycin D. To investigate the influence of the genetic variants on posttranscriptional level, ABCC2 full-length cDNA linked to a flag-tag sequence was cloned into a pIRES2-ZsGreen1 vector. Common haplotypes containing the SNPs -24C>T, 1249G>A and 3972C>T, were constructed by introducing point mutations into the wild type ABCC2 vector construct and controlled for protein expression by western blotting and immunofluorescence. The functional activity was investigated in HEK293T/17 cells transfected with ABCC2 haplotypes by determining the accumulation of ABCC2 substrates. Results: ABCC2 -24C>T had no effect on DNAprotein interactions, mRNA expression and mRNA-stability. However, on posttranscriptional level we could confirm a haplotype depending differential protein expression. The -24C/1249A/3972C haplotype showed an increased protein expression as compared to the wild type construct (-24C/1249G/3972C) whereas 24T/1249G/3972T exhibited a decreased protein expression. The currently ongoing accumulation experiments will give further information about the ABCC2 mediated efflux either depending on different protein expression levels of ABCC2 or on an additional effect on the activity of the transporter. 1. Dept. of Experimental and Clinical Pharmacology, UKSH, Campus Kiel, Germany, 2. Dept. of Pharmacology, Ernst-Moritz-Arndt University, Greifswald, Germany

11 A dileucine motif in OATP2B1 affects apical and basolateral sorting Köck K. (1), Grube M. (1), Koenen A. (1), Giese B. (2), Fraunholz M. (2), Jedlitschky G. (1), Kroemer H.K. (1) Introduction: OATP2B1 is a member of the organic anion transporting polypeptides (OATP/SLCO) and shows a wide tissue distribution including intestinum, liver and placenta. In these organs it might be involved in the uptake of endogenous or exogenous compounds, like steroid conjugates or statins. While a basolateral localization has been shown in most of the organs, an apical localization was demonstrated in enterocytes, suggesting that OATP2B1 is involved in the intestinal uptake of substances. To date, only little is known about the underlying mechanisms of this differential sorting in polarized cells. However, it is known that C-terminal domains of transport proteins often contain putative sorting signals and PDZ domains. Methods and Results: We analyzed the effect of the C-terminal region of OATP2B1 on trafficking of this protein using a GFP-tagged OATP2B1. The introduction of GFP did not result in altered trafficking or changes of estrone sulfate transport compared to the untagged OATP2B1 (Km 6.7 ± 4.2 µM and 7.0 ± 2.8 µM; respectively). To study the sorting effect of the C-terminal region, we employed GFP-tagged OATP2B1 variants lacking four (∆4), eleven (∆11) or seventeen (∆17) amino acids and compared the localization to the wildtype variant. While ∆4 and ∆11 showed a strong basolateral staining, the ∆17 mutant demonstrated an exclusive apical expression suggesting that this domain contains a putative sorting signal necessary for basolateral sorting. We could further identify a dileucine motive (EQLLV) in the respective region and substitution of the leucine at position 696 but not 695 by alanine lead to enhanced apical localization of the OATP2B1 protein. Compared to the wild type protein, either of the dileucine mutants revealed differences estrone sulfate transport as determined by Km values. Conclusion: We could identify a dileucine motif (EQLLV) in the C-terminal region of OATP2B1, which is necessary for basolateral sorting. Interaction of adapter proteins with this motif might give an explanation for differential sorting in polarized cells. 1. Dept. of Pharmacology, Ernst Moritz Arndt University, Greifswald, Germany, 2. Competence Center for Functional Genomics, Ernst Moritz Arndt University, Greifswald, Germany

10

12 Influence of PKC phosphorylation sites on OATP2B1 function Koenen A.B. (1), Köck K. (1), Grube M. (1), Giese B. (2), Fraunholz M. (2), Jedlitschky G. (1), Kroemer H.K. (1) Background: The organic anion transporting polypeptide OATP2B1 is an important member of the solute carrier family SLCO. It is an uptake transporter with a wide tissue distribution but narrow substrate specificity compared to other OATPs. Common substrates are steroid hormone conjugates such as estrone-3-sulfate (E1S) or dehydroepiandrosteronesulfate (DHEAS) and xenobiotics like atorvastatin, glibenclamide or benzylpenicillin. Hitherto little is known regarding the regulation of OATP2B1. Posttranslational modification processes like phosphorylation by protein kinases are possible regulatory mechanisms affecting the transport activity of this protein. Results: In initial experiments we found that preincubation with the PKC activator phorbol-12-myristate-13-acetate (PMA) significantly inhibited the [3H]DHEAS and [3H]E1S uptake into OATP2B1-MDCKII cells in a time and concentration dependent manner. This effect could be reversed by coincubation with the PKC inhibitor BIM I (1 µM). Kinetic analyses of the E1S uptake revealed a decrease of the maximal transport velocity from 288 (DMSO) to 165 pmol/min/mg protein (100 nM PMA), while the substrate-affinity remained unchanged (9.5 vs. 13.2 µM). Using immunofluorescence we could demonstrate rapid redistribution into vesicle-like intracellular compartments. Using in silico analyses based on three different phosphorylation site prediction programs we identified two putative protein kinase C (PKC) recognition motifs at the amino acid position S337 and T338 of OATP2B1, which have been predicted by all programs. We therefore investigated the role of these PKC sites on OATP2B1 function by mutating S337 and T338 to alanine by the use of site-directed-mutagenesis. However, these substitutions had no influence on the PMA effect described before. Furthermore, no effect on PMA-mediated internalization was observed. Conclusion: Taken together, these data suggest the existence of other, e.g. atypical phosphorylation sites in the amino acid sequence of OATP2B1 which serve as recognition sites for protein kinase C or other factors and adaptor proteins that regulate internalization. 1. Dept. of Pharmacology, E.-M.-A.-University, Greifswald, Germany, 2. Competence Center for Functional Genomics, E.-M.-A.-University, Greifswald, Germany

13 OATP2B1-mediated transport of cytostatic drugs used in the treatment of glioma Oberstadt M. (1), Bien S. (1), Köck K. (1), Grube M. (1), Friedel C. (2), Schroeder H.W.S. (2), Kroemer H.K. (1) The transport of cytostatics through the blood-brain barrier or into glioma cells is one prerequisite for their antitumoral efficacy. Particularly in the treatment resistant glioblastoma multiforme, the influx characteristics of drugs is widely unknown which is in contrast to well-characterized export transporters. One of the influx transporters detected in brain and glioma samples is OATP2B1. Therefore, our aim was to characterize the influx transporter OATP2B1 in glioblastoma tissue and glioma cell lines and to investigate whether cytostatics commonly used for glioma therapy are substrates of OATP2B1.OATP2B1 was detected on mRNA level via semiquantitative RT-PCR in the tested glioblastoma samples and in control samples of cerebellum. In addition, OATP2B1 mRNA was detected in the glioma cell lines LN-18 (glioblastoma multiforme) and in U87-MG (astrocytoma grade III) arguing for an OATP2B1 mRNA expression not only in microvessels of the glioma tissue but also in the tumor cells itself. Moreover, OATP2B1 protein was quantified via immunoblot and immunofluorescence in samples from patients with glioblastoma multiforme and in healthy cerebellum. Immunoblot analysis revealed a difference in glycosylation of the protein OATP2B1 between the glioblastoma and cerebellar samples. The glioblastoma tissue expressed more of the glycosylated variant. In contrast, the unglycosylated form was more abundant in the cerebellum. OATP2B1 protein could also be detected in the glioma cell lines LN-18 and U87-MG.Immunofluorescence microscopy showed OATP2B1 signals restricted to membrane of the both tumor cells and microvessels of glioblastoma samples. Finally, we investigated the effect of the cytostatics etoposide, teniposide, carmustine, lomustine, temozolomide, doxorubicin and vincristine on OATP2B1-mediated transport of estrone-3-sulfate, a known substrate of OATP2B1. In contrast to the other cytostatics, etoposide and teniposide showed a significantly decreased OATP2B1-mediated estron3-sulfat transport (p ≤ 0.01) with IC50 values of 28 µM for etoposide and 2,1 µM for teniposide arguing for a role of etoposide and teniposide as substrates and/or inhibitors. In conclusion, the individual expression pattern or function of OATP2B1 may influence the uptake of these cytostatics. 1. Dept. of Pharmacology, Ernst Moritz Arndt University, Greifswald, Germany, 2. Dept. of Neurosurgery, Ernst Moritz Arndt University, Greifswald, Germany

14 Complex biochemical properties of a fibroblast growth factor-binding protein (FGF-BP) Schulz D. (1), Hagenbusch J. (1), Abuharbeid S. (1), Czubayko F. (1), Aigner A. (1) Several members of the fibroblast growth factor (FGFs) family are upregulated in solid tumors and have been shown to be critical for tumor cell proliferation and tumor angiogenesis. However, upon secretion some FGFs are tightly bound to the extracellular matrix and this immobilization prevents them from reaching their receptors. One mechanism of bioactivation involves an FGF-binding protein (FGF-BP), which has been shown to bind to FGFs, leading to their release from the extracellular matrix. Thus, FGFBP expression can be rate-limiting for tumor growth and tumor angiogenesis, and may serve as an 'angiogenic switch molecule'. FGF-BP has been shown to be upregulated in various tumors, and here we demonstrate enhanced expression levels in ovarian carcinoma and colon carcinoma tissue samples or tumor cells. Beyond its extracellular mode of action, FGF-BP properties suggest additional functions. By confocal microscopy, we demonstrate a nuclear localization of FGF-BP in cells stably transfected with an FGF-BP/YFP fusion protein. To further analyse possible functions of FGF-BP in tumor cells, we generated inducible shRNA expression vectors for conditional RNAimediated FGF-BP knockdown. Upon stable transfection of colon carcinoma cells with the shRNA expression vectors, this system required a surprisingly extensive clonal selection to obtain cells with inducible FGF-BP downregulation and proved to be not very robust with regard to reproducibility. These results were confirmed in a luciferase

knockdown model and were thus independent of the target gene. For further studies, recombinant FGF-BP was expressed in bacteria and yeast (P. pastoris) and revealed proliferation-stimulating as well as potentially toxic effects of the protein. FGF-BP expression in P. pastoris could only be maintained under constant G418 selection, and purified FGF-BP was rapidly degraded under native conditions indicating protein instability. Taken together, this suggests complex functions of FGF-BP beyond its already described extracellular role and a tight regulation of cellular FGF-BP levels. 1. Dept. of Pharmacology, Philipps-University, School of Medicine, Marburg, Germany

15 Characterization of proliferative effects of Insulin in human lung fibroblasts Reitzenstein U. (1), Fuhrmann R. (1), Mayer P. (2), Enzmann H. (2), Racké K. (1), Warnken M. (1) Because insulin has the potential to act as growth factor, there are concerns about long term safety of insulin and its analogues used in the treatment of diabetes. Recently, we demonstrated proliferative effects of insulin and insulin-like growth factor (IGF) in primary human lung fibroblasts and several human lung fibroblast cell lines (HEL299, IMR90, MRC5). The present study aimed to further characterize these proliferative effects.Proliferation of HEL299 fibroblasts was studied by determination of 3H-thymidine incorporation. IGF (EC50 3nM) and insulin glargine (EC50 6nM) showed to be more potent and more effective (maximum increase by 155 and 140%, resp.) in stimulating proliferation than insulin (EC50 22nM) or insulin detemir (EC50 110nM, maximum increase by 80%). The stimulatory effects of maximally effective concentrations of insulin (1µM) and IGF (10nM) were not additive. In contrast, the proliferative effect of insulin in combination with IGF was significantly smaller than that of IGF alone. The proliferative effects of insulin and IGF were largely inhibited by the tyrosine receptor kinase inhibitor tyrphostin AG1024 (300nM), whereas inhibition of PI3 kinase by wortmannin (300nM), mTOR by rapamycin (300nM), p38 MAP kinase by SB203580 (10µM) or ERK MAP kinase by PD98059 (30µM) did not attenuate proliferative effects of insulin and IGF, although inhibition of these pathways resulted in reduction in the spontaneous rate of proliferation by 60-80%.The pharmacological profile suggests that proliferative effects of insulin and its analogues in HEL299 fibroblasts are mediated primarily via IGF receptors at which insulin appears to act as partial agonist. The proliferative effects are mediated via receptor tyrosine kinase activity, but signaling further down-stream remains to be illuminated. 1. Inst. of Pharmacology & Toxicology, University of Bonn, Bonn, Germany, 2. Federal Institute for Drugs and Medical Devices, Bonn, Germany

16 Characterization of proliferative effects of insulin and insulin-like growth factor in human airway epithelial cells Warnken M. (1), Reitzenstein U. (1), Fuhrmann R. (1), Mayer P. (2), Enzmann H. (2), Racké K. (1) Because insulin has the potential to act as growth factor, there are concerns about long term safety of insulin and its analogues used in the treatment of diabetes. Therefore, expression of insulin and insulin-like growth factor (IGF) receptors, and possible proliferative effects of insulin and IGF were studied in human airway epithelial cells. Using RT-PCR, mRNA for IGF-1 receptors and both splicing variants of insulin receptors was detected in H292 human airway epithelial cells. Using 3H-thymidine incorporation as a measure of proliferation, marked stimulatory effects of IGF (EC50 0.6 nM, maximum increase by 114±8%, n=52), insulin (EC50 6 nM, maximum increase by 119±8%, mean±SEM, n=45) and its analogues insulin glargine (EC50 189 nM, maximum increase by 218±23%, mean±SEM, n=12) and insulin detemir (EC50 1000 nM, maximum increase by 189±15%, mean±SEM, n=12) were observed without displaying significant differences in maximum effects. The stimulatory effects of maximally effective concentrations of insulin (1 µM) and IGF (3 nM) were not additive. The proliferative effect of insulin and IGF were largely inhibited by the tyrosine receptor kinase inhibitor tyrphostin AG 1024 (300 nM), whereas inhibition of PI3 kinase by wortmannin (300 nM), p38 MAP kinase by SB 203580 (10 µM) or ERK MAP kinase by PD 98059 (30 µM) did not attenuate the proliferative effects of insulin and IGF, although inhibition of these pathways resulted in a reduction of the spontaneous rate of proliferation by about 4050%. In conclsusion, in human airway epithelial cells insulin and IGF act as potent and equally effective stimulators of proliferation, whereas in Hel299 human lung fibroblasts (see Reitzenstein et al, this meeting) insulin glargine and IGF showed to be more potent and more effective than insulin in stimulating proliferation. The proliferative effects of insulin and IGF are mediated via receptor tyrosine kinase activity, but signaling further down-stream remains to be illuminated. 1. Dept. of Pharmacology & Toxicology, University Bonn, Bonn, Germany, 2. Federal Institute for Drugs and Medical Devices, Bonn, Germany

17 Factor-Xa induces expression of sphingosine kinase-1 in human vascular smooth muscle cells Ermler S. (1), Boehm A. (1), Rosenkranz A.C. (1), Fischer J.W. (1), Schrör K. (1), Rauch B.H. (1) Sphingosine kinase-1 (SPHK1) generates the lipid signaling molecule sphingosine-1phosphate (S1P), an important mediator of cellular functions, which also regulates endothelial permeability. In the present study we demonstrate for the first time that the activated clotting factor X (FXa) induces expression of SPHK1 in human vascular smooth muscle cells (SMC), thereby inducing S1P synthesis. This mechanism enables FXa to antagonize thrombin-induced endothelial cell (EC) layer damage. Human aortic SMC and saphenous vein EC were cultured separately or in double-layer co-culture. EC permeability was determined by transcellular passage of fluorescently-labelled bovine serum albumin (BSA). SPHK1 expression was determined by semiquantitative/real-time PCR and by Western blotting. S1P was measured by thin-layer chromatography. Immunostaining was performed in aortic plaques from ApoE-deficient mice after 16 weeks on a high-fed diet.FXa induced a time-dependent (within 3 to 16 hours) and concentration-dependent (10 to 100 nM) upregulation of SPHK1 mRNA (3-fold) and

11

protein (2-fold) expression in human SMC. This was accompanied by an increased production of S1P by human SMC. In co-culture experiments, EC permeability to BSA was time- and concentration-dependently increased by thrombin (1 - 10 U/ml) within 2 hours of incubation. This thrombin-induced EC permeability was attenuated when SMC were pre-treated with FXa for 24 hours. This effect of FXa was abolished by coincubation of SMC with a selective SPHK1 inhibitor. In atherosclerotic plaques from ApoE-deficient mice subendothelial and intraplaque FX/FXa was observed and colocalized with expression of SPHK-1. This may suggest transendothelial diffusion of FX and activation to FXa which leads to increased SPHK1 expression in atherosclerotic plaques. These data suggest that FXa antagonizes thrombin-enhanced EC permeability via induction of SPHK1 expression and S1P production in human SMC. This activity may be relevant for endothelial barrier protection in atherosclerotic tissues in vivo. 1. Institut für Pharmakologie und Klinische Pharmakologie, Universitätsklinikum Düsseldorf

18 Soluble adenylyl cyclase in B lymphoma cells Grandoch M. (1), Bujok V. (1), Weber A.A. (1), Schmidt M. (2) We have shown in the murine immature B lymphoma cell line WEHI-231 that activation of extracellular signal-regulated kinases ERK1/2 and the phosphoinositid 3-kinase effector protein kinase B (PKB)/Akt by the B cell receptor (BCR) was cyclic AMP (cAMP)-dependent. Interestingly, the well-known cAMP-target protein kinase A (PKA) was not involved in these signaling events but cAMP-regulated Epac (exchange protein directly activated by cAMP). We were interested, if these effects on BCR responses can also be observed in human B lymphocytes. Here we report on the effects of cAMP and Epac on BCR-induced activation of ERK1/2 in the human B lymphoma cell line Raji. As cAMP and Epac act as mediators in this signaling cascade in both cell lines, we aimed next to identify the adenylyl cyclase (AC) being responsible for the BCR-induced cAMP production in B lymphocytes (transmembrane AC (tmAC) versus soluble AC (sAC)). We first confirmed the expression of sAC by immunostaining with a specific antibody directed against mammalian sAC using western blot and immunofluorescence. Furthermore, we used the specific sAC inhibitor KH7 to distinguish between sAC and tmAC-mediated effects. Preincubation of both cell types, WEHI-231 and Raji, with the inhibitor significantly diminished BCR-induced intracellular signaling, e.g. ERK1/2 phosphorylation. Taken together, our results indicate an involvement of sAC in BCRinduced cAMP-production and activation of ERK1/2 in murine as well as in human B lymphocytes. 1. Institut für Pharmakologie, Universitätsklinikum Essen, Essen, Germany, 2. Department of Molecular Pharmacology, University of Groningen, Groningen, The Netherlands

19 Distinct conformational changes of the human M3 acetlycholine receptor in living cells Hoffmann C. (1), Ziegler N. (1), Sieber A. (1), Zürn A. (1), Zabel U. (1), Lohse M.J. (1) The activation of G protein-coupled receptors (GPCRs) is the initial step in the signaling cascade which is triggered by binding of an agonist to a GPCR. Several lines of evidence suggest that GPCRs can adopt different active conformations but the evidence for these distinct conformations is still indirect. Using a fluorescence resonance energy transfer (FRET) based approach we have studied conformational changes in differently labeled muscarinic acetylcholine receptor (M3AChR). The receptors were labeled with cyan fluorescent protein (CFP) at their C-terminus, and with fluorescein arsenical hairpin binder (FlAsH) via tetra-cysteine tags inserted at two different sites in the third intracellular loop: N-terminally, i.e. close to transmembrane domain V or C-terminally, i.e. close to transmembrane domain VI. A set of structurally different agonists were investigated with respect to their potential to induce changes in FRET and were compared to the endogenous agonist acetylcholine. Several compounds structurally derived from oxotremorine or arecoline showed equally reduced FRET-changes for both constructs. However, compounds that were structurally related to muscarine show similar FRET-change as acetylcholine for the constructs labeled close to transmembrane domain VI while they exhibited significantly reduced changes in FRET for the second construct labeled close to transmembrane domain V. A correlation of the obtained data would be consistent with the induction of different receptor conformations for different ligands in living cells. 1. Dept of Pharmacology, University of Würzburg, Würzburg, Germany

20 Role of PLD2-derived DAG in agonist-induced opioid receptor endocytosis Seifert A. (1), Yang L.Q. (1), Wu D.F. (1), Goschke A. (1), Wang X. (2), Schröder H. (1), Höllt V. (1), Koch T. (1) Internalization of opioid receptors after agonist activation via clathrin-dependent pathway is one important regulatory event of opioid receptor signal transduction. Previously our group has reported that activation of Phosphoplipase D2 (PLD2), which hydrolyzes phosphatidylcholine to phosphatidic acid (PA) and choline, is required for agonistinduced µ-opioid receptor endocytosis. PA can be converted to diacylglycerol (DAG) by PA phosphohydrolase (PPH) and vice versa by DAG kinase (DGK), respectively. Here, we analyze the role of PLD2-derived DAG agonist-induced opioid receptor endocytosis. Agonist-induced µ-opioid receptor (MOR) and d-opioid receptor (DOR) endocytosis was significantly decreased by inhibition of the conversion from PA to DAG by the PPH inhibitor propranolol. Conversely, inhibition of PA formation from DAG by R59949 increased agonist-induced MOR/DOR endocytosis indicating the significant role of PLD2-derived DAG in opioid receptor endocytosis. RNA interference studies showed that reduction of PLD2 and PPH expression in agonist-treated MOR- or DOR-expressing cells resulted in a decrease in endocytosis. Furthermore, PLD2-derived DAG caused phosphorylation of MAPK p38 which is known to be involved in the induction of opioid receptor endocytosis. The enhanced p38 phosphorylation in agonist-treated MOR- or DOR-expressing cells was blocked by reduction of PLD2 and PPH expression via RNA

interference. Further experiments need to be done to investigate how DAG mediates the phosphorylation of p38. 1. Institute of Pharmacology and Toxicology, Otto von Guericke University, Magdeburg, Germany, 2. Institute of Immunology, Otto von Guericke University, Magdeburg, Germany

21 Cellular dynamic mass redistribution reveals biased signaling of a novel class of GPCR activators Kebig A. (1), Holzgrabe U. (2), De Amici M. (3), Kostenis E. (4), Mohr K. (1), Mohr-Andrä M. (1) Dualsteric activators of muscarinic receptors combine high potency receptor activation via the orthosteric transmitter binding site with M2 subtype selective binding via the allosteric site [1]. Stimulation of the M2 receptor by conventional agonists such as acetylcholine is known to activate the Gi/o pathway and additionally the Gs pathway. In order to gain insight into signaling pathway activation by dualsteric agonists we applied the EpicTM real-time biophysical whole cell assay that measures cell activationdependent relocation of cellular constituents. The resulting local change of optical density affects cellular light diffraction properties. This was detected by means of the EpicTM system in CHO cells stably transfected with the hM2 gene and cultivated in microplates that contain resonant waveguide grating optical biosensors in each well. Dynamic mass redistribution (DMR) is recorded as a shift of wavelength of the reflected light. Upon orthosteric and dualsteric agonist addition a DMR signal evolved instantly, peaked after approximately 100 seconds and then declined slowly. Cells having been pretreated with the irreversible Gi/o inhibitor pertussis toxin (PTX) responded to agonist administration with a DMR deflection into the opposite direction. The downward deflected DMR signal is compatible with Gs pathway activation as direct stimulation of adenylyl cyclase by forskolin yields a similar signal. Measurement of M2 receptor mediated G-protein activation in membranes from PTX-pretreated hM2-CHO cells using a [35S]GTPγS-assay revealed an effect of orthosteric agonists such as acetylcholine but a complete lack of action of dualsteric agonists. Thus, whole cell DMR measurements give valid insight into receptor/G protein-interactions. In conclusion, signaling induced by dualsteric agonists is selective for the Gi/o pathway in contrast to promiscuous signaling induced by conventional agonists. [1] Antony et al. (2009) FASEB J. in press Supported by the DFG (GRK 677) Acknowledgment: We thank CorningTM (New York, USA) for providing us with the EpicTM System. 1. Pharmacology & Toxicology, Institute of Pharmacy, University of Bonn, Bonn, Germany, 2. Pharmaceutical Chemistry, Institute of Pharmacy, University of Würzburg, Würzburg, Germany, 3. Institute of Medicinal and Toxicological Chemistry, University of Milan, Milan, Italy, 4. Institute of Pharmaceutical Biology, University of Bonn, Bonn, Germany

22 A MEK / β-catenin signaling pathway regulates airway smooth muscle growth Gosens R. (1), Baarsma H.A. (1), Heijink I.H. (2), Halayko A.J. (3), Meurs H. (1), Schmidt M. (1) We recently identified β-catenin, a component of adherens junctions that acts as a transcriptional co-activator when liberated from cell contacts, as an important signaling intermediate that regulates airway smooth muscle growth. β-Catenin-specific siRNA reduced serum and PDGF-induced Rb phosphorylation, DNA synthesis and airway smooth muscle cell proliferation. In the current study, we aimed to investigate this mechanism in more detail. β-Catenin activation is generally the result of reduced glycogen synthase kinase 3 (GSK3) activity, a kinase that in its unphosphorylated form induces cytoplasmic β-catenin protein degradation. GSK3 inactivation by phosphorylation thus allows for β-catenin cytoplasmic stabilisation and nuclear translocation and subsequent gene transcription. However, although levels of β-catenin in whole cell lysates and nuclear fractions were clearly increased after serum treatment, β-catenin was not liberated from the plasma membrane. Rather, de novo β-catenin formation was involved, as actinomycin D and cycloheximide both abrogated the accumulation of β-catenin in whole cell lysates and in nuclear extracts. Moreover, realtime PCR indicated a clear induction of β-catenin mRNA. Although pharmacological inhibition of GSK3 using SB216763 also increased β-catenin protein in whole cell lysates and nuclear extracts, it failed to induce β-catenin mRNA, confirming that seruminduced β-catenin mRNA expression depends on pathways distinct from GSK3. Indeed, inhibition of MEK using U0126 or expression of a dominant-negative H-Ras (N17Ras) attenuated serum-induced β-catenin protein accumulation and reduced β-catenin mRNA, whereas GSK3 phosphorylation was unaffected. Collectively, these data indicate that β-catenin is an important signaling intermediate in airway smooth muscle growth, and that its accumulation requires de novo protein formation effected via H-Ras and MEK. 1. Dept. of Molecular Pharmacology, University of Groningen, Groningen, the Netherlands, 2. Dept. of Pulmonology, University Medical Center Groningen, Groningen, the Netherlands, 3. Dept. of Physiology, University of Manitoba, Winnipeg, Canada

23 Role of phospholipase D2 in the functional selectivity of mu-opioid receptor ligands Koch T. (1), Seifert A. (1), Wu D.F. (1), Rankovic M. (1), Höllt V. (1) The action of most clinically important opiate drugs, e.g. morphine, is mediated by the mu-opioid receptor (MOPr) which belongs to the G-protein coupled receptor (GPCR) family. Activation of opioid receptors leads to neuronal inhibition caused by multiple effectors, including inhibition of intracellular cAMP production and voltage gated Ca2+ channels, as well as the stimulation of inwardly rectifying K+ channels. However, upon activation with the mu-agonists DAMGO or ß-endorphin the MOPr undergoes rapid redistribution from the plasma membrane into endocytotic vesicles whereas the receptor does not internalize after exposure to morphine or buprenorphine. We recently demonstrated that the agonist-selective trafficking of MOPr is affected by the interaction with the phospholipase D2 (PLD2), a phospholipid-specific phosphodiesterase located in

12

the plasma membrane. We further showed that in MOPr-PLD2-coexpressing HEK293 cells, treatment with DAMGO or ß-endorphin led to an increase in PLD2 activity followed by an induction of receptor endocytosis, whereas morphine and buprenorphine, which do not induce opioid receptor endocytosis failed to activate PLD2. Therefore, we tested here the functional selectivity of opioids in receptor-mediated PLD2 signaling pathways such as ROS (reactive oxygen species)-synthesis and p38 MAPK-activation. We found that receptor-internalizing agonists (like DAMGO, ß-endorphin, Sufentanil, Fentanyl, and Etonitazene) strongly induce ROS-synthesis via PLD-dependent pathways, whereas agonists that do not induce MOPr endocytosis and PLD2-activation (like morphine, buprenorphine, hydromorphone, and oxycodone) failed to activate ROS-synthesis in transfected HEK293 cells. Moreover we demonstrate that the opioid-selective activation of PLD2 is a requirement for the activation of p38 MAPK, which is involved in the induction of receptor endocytosis. 1. Dept. of Pharmacology and Toxicology, Otto-von-Guericke-University, Magdeburg, Germany

24 Regulation of mu-opioid receptor activity by ADP-ribosylation factor 6 Jacob L. (1), Rankovic M. (1), Koch T. (1), Höllt V. (1) Agonist-induced µ-opioid receptor (MOPr) phosphorylation, desensitization and subsequent internalization are important regulatory mechanisms in the development of cellular opioid tolerance during chronic opioid treatment. Numerous studies have demonstrated that desensitized and internalized MOPr are rapidly dephosphorylated and recycled to the cell surface in a reactivated state, thus counteracting receptor desensitization and opioid tolerance. Further studies revealed agonist-selective differences in the ability to induce opioid receptor internalization and that the endocytotic efficacy of opioids is negatively correlated with the induced receptor desensitization and opioid tolerance. We have recently shown that the MOPr-mediated and ARF-dependent activation of phospholipase D2 (PLD2) is essential for the induction of MOPr endocytosis and recycling. In the present study we investigated whether the opioidinduced PLD2 activation is mediated via ADP-ribosylation factor 6 (ARF 6) or ARF 1. We demonstrate here, that overexpression of ARF6 /N48I mutant, which is incapable of activation of PLD leads to a reduced internalization, a faster desensitization and a slower dephosphorylation/resensitization of MOPr after treatment with the receptorinternalizing agonist [D-Ale2-MePhe4-Gly-ol]enkephalin (DAMGO). However, in cells treated with hydromorphon, an opioid that cannot activate PLD2 and thus does not induce MOPr endocytosis, the receptor desensitization and dephosphorylation / resensitization was not affected by the coexpression of ARF6 /N48I mutant. On the other hand, coexpression of MOPr with ARF6/T157N – a fast cycling mutant that can induce PLD activation resulted in a stronger MOPr endocytosis, a faster dephosphorylation/resensitization and a reduced desensitization after treatment with the agonist hydromorphone. ARF1 dominant negative mutant did not show any effect in any of the experiments. Taken together these results suggest that ARF6 and not ARF1 plays an important role in regulation of µ-opioid receptor activity and that this effect is PLD2mediated. 1. Dept. of Pharmacology and Toxicology, Otto-von-Guericke-University, Magdeburg, Germany

based approach. We generated βAR FRET sensors with a yellow fluorescent protein (YFP) inserted into the third intracellular loop and a cyan fluorescent protein (CFP or Cerulean) fused to the C-terminal tail. Upon stimulation of the sensors we determined the activation characteristics of the polymorphic receptors in real time and in living cells. The FRET sensors retained the pharmacological properties (i.e. ligand affinities) of the native receptors, and also reflected the functionality of the wild-type receptors since they were able to produce cAMP to the same extent. Comparing the activation kinetics of the different receptor variants upon a single agonist stimulation did not reveal any significant differences. In contrast, we found that receptor polymorphisms critically determine the kinetics of receptor activation upon repetitive stimulation by agonists. Current experiments address the hypothesis that cellular mechanisms determine receptor conformation kinetics in dependence on receptor polymorphisms. Specifically, the role of G protein-coupled receptor kinases is addressed. Our data suggest that naturally occurring receptor polymorphisms critically determine the activation kinetics of βadrenergic receptors, and thus possibly affect their signalling properties. 1. Rudolf Virchow Center, DFG-Research Center for Experimental Biomedicine, JuliusMaximilians University, Würzburg, Germany, 2. Developmental Biology Institute, Université de la Méditerranée, Marseille, France, 3. Institute for Pharmacology and Toxicology, Julius-Maximilians University, Würzburg, Germany, 4. IPT - Institute for Pharmacology and Toxicology, Technical University (TUM), Munich, Germany

27 Role of receptor endocytosis in the development of cross-tolerance between opioids Konietzko J. (1), Kahl E. (1), Höllt V. (1), Koch T. (1) Opioids are important drugs in clinical pain therapy, but the long term use of opioids is often limited by the development of opioid tolerance. On the molecular level tolerance development after chronic opioid treatment depends on a receptor phosphorylation and G-protein uncoupling (desensitization) followed by receptor internalization. Receptor internalization has long been considered to directly contribute to the tolerance development by decreasing the number of signaling receptors. However, we have previously shown, that the endocytotic efficacies of opioids are negatively correlated to the induced µ-opioid receptor desensitization and opioid tolerance. Our findings indicate that receptor endocytosis is an important mechanism to ensure that desensitized and internalized receptors are rapidly recycled to the cell surface in an active form maintaining receptor signaling and reducing receptor desensitization and tolerance development. Clinically, patients often demonstrate incomplete cross-tolerance between opioid analgesics. Therefore, we tested whether agonist-induced receptor endocytosis and recycling affect the development of cross-tolerance to opioids. In a HEK293 cell model we found that chronic treatment with endocytotic drugs (such as etonitazene, etorphine or sufentanyl) resulted in a strong cross-desensitization among numerous opioids, whereas chronic treatment with non-endocytotic drugs (such as morphine, oxycodone, or hydromorphone) induced only partial, incomplete cross-desensitization to other opioids. Further resensitization studies implicate that recycled and reactivated opioid receptors might be functionally different from "naive" (non-endocytosed) receptors. An explanation for the observed functional differences between "naive" and "used" (recycled) receptors might be that opioid receptors can remain persistently phosphorylated after recycling. 1. Dept. of Pharmacology and Toxicology, Otto-von-Guericke University, Magdeburg, Germany

25 NHERF1 forms a ternary complex with parathyroid hormone receptor 1 and betaarrestin2 Klenk J.C. (1), Vetter T. (1), Vilardaga J.P. (2), Lohse M.J. (1) Na+/H+ exchanger regulatory factor 1 (NHERF1) and βArrestin-2 both are scaffolding proteins of the parathyroid hormone receptor 1 (PTHR1). NHERF1 is 50 kDa phosphoprotein consisting of two tandem PDZ domains and one Ezrin-Radixin-Moesin (ERM)-binding domain. NHERF1 interacts with the carboxyl terminus of the PTHR at one of the PDZ domains thereby tethering the receptor via the ERM-domain to the actin cytoskeleton and recruiting signaling complexes such as phospholipases to the PTHR1 via the second PDZ domain. βArrestin-2 serves as a multifunctional adaptor protein to many G protein-coupled receptors. It is rapidly recruited to ligand-activated PTHR1 and provokes its uncoupling from G-proteins and subsequent internalization. Moreover, recent studies indicate that βArrestin-2 can link the PTHR1 to the MAPK-signaling pathway by recruiting several kinases to the receptor.Here we demonstrate that NHERF1 affects βArrestin-2 recruitment of the PTHR. Using Fluorescent Resonance Energy Transfer (FRET) measurements in living cells we determined the association kinetics between βArrestin-2 and the PTHR after receptor activation. In the presence of NHERF1 βArrestin-2 recruitment of the PTHR was twofold faster compared to cells not expressing NHERF. We further show that under basal conditions NHERF1 constitutively binds to PTHR and to βArrestin-2 without a direct interaction between βArrestin-2 and PTHR1. Upon receptor activation with parathyroid hormone analogs PTH(1-14) and PTH(1-34) we detected the formation of a stable ternary complex between PTHR1, NHERF1 and βArrestin-2. These data suggest a new role for NHERF1 serving as an adaptor for βArrestin-2 and bringing it into close proximity to the PTHR1 thereby enhancing the speed of βArrestin-2 recruitment by the receptor. 1. Inst. of Pharmacology, Julius Maximilians University, Würzburg, Germany, 2. Dept. of Pharmacology & Chemical Biology, University of Pittsburgh, Pittsburgh, PA, USA

26 Impact of polymorphisms on β-adrenergic receptor activation Ahles A. (1,3,4), Rochais F. (2), Lohse M.J. (3), Engelhardt S. (1,3,4) β1-and β2-adrenergic receptors (βARs) play a key role in the sympathetic nervous system mediating crucial catecholamine effects in the organism. They are important pharmaceutical targets for pulmonary and cardiovascular diseases with variable treatment success. These interindividual differences in drug response might be due to frequently occurring polymorphisms within both G protein-coupled receptor subtypes. The most important polymorphisms are Ser49Gly and Gly389Arg for the β1AR, Arg16Gly and Gln27Glu for the β2AR. To investigate the impact of these polymorphisms on receptor conformation we used a fluorescence resonance energy transfer (FRET)

28 Dual histamine H1- and H4-receptor ligands Straßer A. (1), Wittmann H.J. (2), Deml K.F. (3), Seifert R. (4) Histamine H1-receptor (H1R) antagonists are clinically important for treatment of allergic diseases, and thehistamine H4-receptor (H4R) is discussed as a new therapeutic target for inflammation. The aim of this study was to analyze if H1R antagonists also show affinity to the H4R. Thus, we co-expressed the human (h) H1R and RGS4 on the one hand and the hH4R-GAIP fusion protein with Gαi2 and Gβ1g2 on the other hand in Sf9 insect cells and performed [3H]mepyramine (H1R) and [3H]histamine (H4R) competition binding assays and steady-state GTPase assays with a large number of H1R ligands. Doxepine (1), a potent hH1R antagonist acts as an inverse agonist at the hH4R. However, compared to the hH1R, the affinity is significantly decreased at hH4R. Additionally, we characterized typical H1R partial agonists, the phenylhistamines and histaprodifens, at the hH4R. Analogous to the hH1R, phenylhistamines show partial agonism at the hH4R. However, in contrast to the hH1R, most of the histaprodifens act as inverse agonists at the hH4R. A comparable affinity at the hH1R and hH4R was found for phenylhistamines with an additional histamine moiety (2) and suprahistaprodifen (3). cpd. hH1R hH4R 8.9 ± 0.1 4.6 ± 0.1 1 5.4 ± 0.1 5.8 ± 0.1 2 6.3 ± 0.1 5.8 ± 0.1 3 These results are a good starting point for developing dual H1R and H4R antagonists. The project is supported by the Deutsche Forschungsgemeinschaft (GRK 760) 1. Department of Pharmaceutical and Medicinal Chemistry I, University of Regensburg, 2. Faculty of Chemistry and Pharmacy, University of Regensburg, 3. Faculty of Medicine, University of Regensburg, 4. Institute of Pharmacology, Medical School of Hannover

13

29 Species homologs as tools for the molecular analysis of constitutive H4-receptor activity Schnell D. (1), Schneider E. (1), Seifert R. (2) The histamine H4-receptor (H4R) is the most recently identified G protein-coupled receptor (GPCR) for histamine and plays an important role in the immune system [1]. Although other histamine receptors share a high degree of homology across species, this is not true for the H4R for which the homology ranges from 65-72% at the protein level. Those structural differences also translate into substantial pharmacological differences [2]. The human H4R possesses constitutive, i.e. agonist-independent, activity in transfected cells, whereas the rodent H4R does not. These data open the possibility that inverse agonists may have different in vivo properties than neutral antagonists. The molecular basis and a possible (patho)physiological function of the high constitutive activity, however, are not yet known. In this study, we use various H4R species homologs to investigate their constitutive activity. After expression of H4Rs in Sf9 cell membranes, we show differences in constitutive activity using proximal readouts such as high-affinity GTPase activity. The human H4R displays very high basal signalling, whereas in rodent receptors, constitutive activity is nearly absent. In fact, this is reflected by a lower affinity of the rodent H4R homologs for agonists, a lower potency and efficacy of partial agonists and no intrinsic activity of inverse agonists. Moreover, some ligands previously identified to be neutral antagonists at human H4R turned out to be partial agonists at rodent H4R. Interestingly, constitutive H4R activity is also evident in living cells. When expressed in HEK293 cells, human H4R is constitutively internalized, as reflected by the higher fraction of intracellularly localized receptors. To identify the domains responsible for those differences, we follow a chimeric approach and perform site-directed mutagenesis to unmask the contribution of single amino acid residues. [1] Thurmond RL et al., Nat Rev Drug Discov, 2008, Jan; 7(1):41-53. [2] Liu C et al., J Pharmacol Exp Ther, 2001, Oct; 299(1):121-30. (Supported by Deutsche Forschungsgemeinschaft - GRK 760) 1. Dept. of Pharmacology and Toxicology, University of Regensburg, Regensburg, Germany, 2. Institute for Pharmacology, Hanover Medical School, Hanover, Germany

30 Chemotactic formyl peptide receptors in human adipocytes: novel mediators of obesity-related inflammation Moepps B. (1), Roth N. (1), Fischer-Posovszky P. (2), Wabitsch M. (2), Gierschik P. (1) Accumulation of neutrophils and macrophages at sites of inflammation are of critical importance in the pathogenesis of inflammatory diseases. Chemoattractant receptors, including chemokine receptors, complement fragment C5a receptors, and members of the N-formyl peptide receptor family (FPR) have been reported to regulate various functions of neutrophils and macrophages, e.g. directed migration, respiratory burst, and phagocytosis. Some of these responses are also mediated by FPR receptors, including FPR (the classical N-formyl peptide receptor), and the two FRR-like receptors, FPRL1 and FPRL2. While FPR1 has high affinity to N-formyl peptides typically derived from bacteria and/or mitochondria, FPRL1 has low and FPRL2 no affinity for these proinflammatory chemotactic agents. Most intriguingly, the receptors also interact with and are activated by several other endogenous ligands, including annexin-1, lipoxin A4, and chemokine fragments. Interstingly, some of the latter have anti-inflammatory actions. Thus, FPRs may regulate the balance between pro- and anti-inflammatory mechanisms in inflamed tissue, including the adipose tissue in the obese state. Adipocytes, preadipocytes, and other cell types within adipose tissue are known to produce a plethora of factors including classical adipokines as well as pro-inflammatory cytokines and chemokines. The latter may contribute to the infiltration of neutrophils and macrophages into adipose tissue observed in humans and laboratory animals on a highfat diet. To investigate the potential role of FPR receptors in the pathogenesis of obesityrelated inflammation, we set out to characterize the expression of and the cellular responses elicited by FPR receptors in cultured human SGBS (Simpson-Golabi-Behmel syndrome) preadipocytes, which serve as an excellent in vitro model of human adipocyte differentiation and thus adipose tissue formation. The results showed that (i) both FRP and FRL1 are expressed at the mRNA level in SGBS preadipocates and (ii) treatment of the cells with N-formyl-methionine-leucine-phenylalanine induced a robust concentration-dependent rapid and transient rise in the concentration of cytoplasmic free calcium. Taken together, the data indicate that two members of the FPR receptor family are expressed and functionally active in cultured human preadipocytes. 1. Institute of Pharmacology and Toxicology, University of Ulm Medical Center, Ulm, Germany, 2. Pediatric Endocrinology, Diabetes and Obesity Unit, Department of Pediatrics and Adolescent Medicine, University of Ulm, Ulm, Germany

31 Prostaglandin EP4 receptor stimulation attenuates hypoxia-induced cell death in H9c2 cardiomyoblasts Treml M. (1), Schrör K. (1), Hohlfeld T. (1), Meyer-Kirchrath J. (1) Prostaglandin E2 (PGE2) is involved in numerous physiological and pathological processes such as vasodilation, cell growth and inflammation. In the heart, the generation of PGE2 is significantly increased in acute myocardial ischemia and is regarded as an intrinsic mechanism of protection against ischemia/reperfusion injury. PGE2 exerts its multiple actions through the specific EP receptor-subtypes EP1, EP2, EP3 and EP4, all of which belong to the family of G-protein coupled receptors but activate different signaling pathways. The involvement of both EP3 and EP4 in antiischemic PGE2- actions has been assumed. To study specifically the role of Gscoupled EP4 in PGE2-mediated cell survival we generated H9c2 rat cardiomyoblasts lentivirally overexpressing the human EP4 receptor. The cells were starved and subjected to 24 h of hypoxia with subsequent reoxygenation for 60 min. Initial EP4 stimulation with the specific agonist L-000902688 (10 nM) markedly decreased the number of dead cells by 45 ± 11 % (p<0.001) as determined by SYTOX® Green fluorescence. Measurement of LDH activity in the supernatant confirmed this result. LDH was reduced from 124 ± 10 to 42 ± 8 U/L (p<0.001) after EP4 stimulation. This effect could not be abrogated by specific inhibition of either PI3 kinase or MAP kinase (ERK1/2). Moreover, there was no detectable activation of GSK-3b and AKT kinase subsequent to EP4 stimulation. Considering exchange protein directly activated by

cAMP (Epac) as an alternative target of cAMP mediated signaling, we verified its presence in the cells and detected EP4-dependent activation of the downstream effector Rap1. However, the application of a specific and direct activator of Epac (8-pCPT-2'OMe-cAMP, 250 µM) had no effect on cell survival. A systematic search for survivalrelated autocrine secondary mediators was performed by microarray analysis and revealed the EP4 dependent upregulation of the EGF-like growth factor amphiregulin. We conclude that EP4 receptor signaling improves cell survival in cardiomyocytes under hypoxic conditions, possibly involving amphiregulin-mediated activation of the EGF receptor. 1. Dept. of Pharmacology and clinical Pharmacology, Heinrich-Heine-University, Düsseldorf, Germany

32 Cannabinoid agonists and cellular signal transduction mechanisms Assi A.A. (1), Howlett A. (2) It is known that CB1 cannabinoid receptors in the brain are coupled to pertussis toxinsensitive G proteins of the Gi/Go family. Their cellular signaling mechanism, however, are poorly understood. CB1 cannabinoid receptors are expressed on N18TG2 neuroblastoma and C6 glioma cells. Extracellular signal-regulated kinases 1 and 2 (ERK1/2) are a group of kinases that play an important role in cell proliferation and differentiation. Using N18TG2 neuroblastoma and C6 glioma cells, we investigated the influence of CB1 cannabinoid agonists on the ERK pathway. N18TG2 cells C6 glioma cells were serum-starved (12-24 h) and treated with the CB1 cannabinoid receptor antagonist SR141716 prior to stimulation. Treatment of cells with CP55940, Win55212-2 or methanandamide produced a concentration-dependent phosphorylation of ERK as quantitated by densitometry of bands on Western blots of cell lysates. The potency ratios were: CP55940 > Win55212-2 > methanandamide. All drugs produced a concentration-dependent increase in the level of the phosphorylated forms of ERK. Studying the time-course for such stimulation revealed that the effect was obtained 2-10 minutes of treatment with different CB1 agonists. Pretreating cells with 100 ng/ml pertussis toxin attenuated the cannabinoid agonist-induced ERK activation, implicating Gi and/or Go in this signal transduction pathway. These data suggest that activation of mitogen-activated protein kinase is an important cellular pathway of CB1 cannabinoid agonists. These studies elucidate a signal transduction relay between CB1 receptor activation of G proteins and the ERK pathway in cells. Supported by NIDA U24DA12385 to develop the Neuroscience/Drug Abuse Research Program, R01-DA03690 and K05DA00182 to AH and NINDS F05-NS11110 to AA. 1. Dept. of Pharmacology, College of Medicine, Assiut University, Assiut, EGYPT, 2. Neuroscience/Drug Abuse Research Program, JL Chambers Biomedical/Biotechnology Research Institute, North Carolina Central University, Durham, NC USA

33 Receptor fingerprinting of the endogenous β2-adrenergic receptor agonist endogenous and nor-epinephrine Reiner S. (1), Lohse M.J. (1), Hoffmann C. (1) The novel concept of functional selectivity suggests that a ligand can have distinct efficacies with regard to different signaling pathways associated with that receptor. There are several examples of G-protein coupled receptors, including the β2AR, that appear to behave according to this theory. However, the possibility that the two endogenous β-receptor agonists nor-epinephrine and epinephrine can lead to the selective activation of diverse effectors has not been examined. In the present study we investigated the influence of epinephrine and nor-epinephrine, as well as fenoterol and isoproterenol on G-protein activation, β2AR internalization and β-arrestin2 membrane translocation in living cells. Although the endogenous agonists epinephrine and norepinephrine behave both as full agonists concerning the G-protein activation, norepinephrine causes a slower and significantly decreased receptor internalization compared to epinephrine. In contrast, fenoterol exhibited less G-protein activation and cAMP production than the endogenous β-receptor agonists, but arrestin translocation and receptor internalization were similar to epinephrine. These findings suggest that the tested ligands each exhibit a special efficacy profile towards the functional endpoints considered and thus behave ”biased”. Moreover, our study shows that the agonists not only differ quantitatively in receptor internalization and arrestin translocation but also in terms of kinetics. 1. Dept. of Pharmacology, University of Würzburg, Würzburg, Germany

14

34 Functional characterization of inverse agonists at the histamine H4 receptor Schneider E.H. (1), Thurmond R. (2), Seifert R. (3) The histamine H4 receptor (hH4R) is expressed on cells of the immune system like eosinophils and mast cells as well as neuronal cells [1,2] and is involved in the pathogenesis of pruritus [1-3]. Accordingly, hH4R antagonists may become important drugs for the therapy of pruritus [3]. When co-expressed with Gαi2 and Gβ1γ2 in Sf9 insect cells, the hH4R shows extraordinarily high and sodium-insensitive constitutive activity in the steady-state GTPase assay. This suggests that inverse agonists rather than neutral antagonists should be used to treat H4R-associated diseases. We investigated a series of hH4R antagonistic indoles (a, X = CH) benzimidazoles (a, X = N) and thienopyrroles (b) that are derived from the prototypical H4R antagonist JNJ-7777120 (1-[(5-Chloro-1Hindol-2-yl)carbonyl]-4-methyl-piperazine) [4]. We performed steady-state GTPase assays with Sf9 cell membranes co-expressing the hH4R with Gαi2 and Gβ1γ2. Thioperamide, the most efficacious inverse hH4R agonist known, was used as reference. Most compounds behaved as partial inverse agonists. The efficacies varied in a range of a 20-80 % of the thioperamide effect. Interestingly, N R1 substitution of R4 by –Br in the indole compounds (R = H: JNJ-7781111) or of R1-3 by –CH3 in the R2 N X thienopyrrole derivatives (R1 = H: JNJ-18056545, R1 = Cl: JNJ-19409637) resulted in neutral antagonism. N O R3 H Surprisingly, the Kb values of most inverse agonistic compounds determined in the presence of histamine R4 (100 nM) were much lower than the EC50 values X = CH X=N determined in the absence of histamine. This indicates that the histamine-stabilized active hH4R N state shows a higher affinity to inverse agonists than b the constitutively active agonist-free hH4R state. Our N S results provide first hints for structure-activity R1 relationships of inverse agonists at the hH4R and N O show that the hH4R adopts distinct active states that H can be differentiated from each other by their different R2 affinity to inverse agonists. [1] Nat Rev Drug Discov. 7:41-53 (2008), [2] Brain Res. doi:10.1016/j.brainres.2008.11.018, in press (2008), [3] J Allergy Clin Immunol. 119:176183 (2007), [4] J Med Chem. 48:8289-8298 (2005) 1. Dept. of Pharmacology and Toxicology, University of Regensburg, Regensburg, Germany, 2. Johnson & Johnson Pharmaceutical Research and Development, L.L.C., San Diego, CA, USA, 3. Institute for Pharmacology, Hannover Medical School, Hannover, Germany

35 Effects of thiazolidinediones on S1P receptor expression in renal mesangial cells Koch A. (1), Pfeilschifter J. (1), Huwiler A. (2) Since peroxisome proliferator-activated receptors (PPARs) were discovered, they have evolved from another member of the nuclear hormone receptor family to an extremely important set of targets for drug discovery. Three isoforms of PPAR, namely PPARα, PPARβ/δ and PPARγ, are characterized. PPARγ is a ligand-induced transcription factor that can be activated by thiazolidinediones, fatty acids and eicosanoids and regulates the expression of target genes by binding to DNA sequence elements as a heterodimer with the 9-cis retinoic acid receptor. PPARγ has been shown to play an essential role in the regulation of adipogenesis, immune response, insulin sensitivity, and glucose homeostasis. Of particular interest, PPARγ agonists (thiazolidinediones) are not only able to ameliorate glomerulosclerosis and kidney dysfunctions in diabetic nephropathy but also exert beneficial actions in nondiabetic chronic kidney disease. In addition, S1P receptor signalling in mesangial cells can be associated with progression of chronic kidney disease. Furthermore, renal mesangial cells play a central role in most pathological processes of the renal glomerulus. To evaluate a possible correlation between PPARγ activation and S1P receptor signalling in association with the development of chronic kidney diseases, we investigated the effect of PPARγ ligands (troglitazone and rosiglitazone) on the expression and relevant downstream signalling effects of S1P receptors in mesangial cells. In this study, we show for the first time that thiazolidinediones are able to increase the relative mRNA levels of S1P receptors in rat mesangial cells, indicating a possible relationship between PPARγ activation and S1P receptor expression. 1. Pharmazentrum Frankfurt/ZAFES, Klinikum der Goethe-Universität, Frankfurt, Germany, 2. Institut für Pharmakologie, Universität Bern, Bern, Switzerland

36 Site-specific two color labeling of proteins with FlAsH and ReAsH living cells Zürn A (1), Klenk J.C. (1), Zabel U. (1), Lohse M.J. (1), Hoffmann C. (1) Specific labeling of two proteins simultaneously in living cells with two different molecular probes would be an important further development for multiparameter imaging of cellular functions. Here we report a strategy to selectively label two different proteins in living cells with two different fluorophores, FlAsH and ReAsH. Recently improved tetracysteine binding motifs have been described to selectively bind FlAsH or ReAsH. We compared the six amino acid motif CCPGCC and the twelve amino acid motif FLNCCPGCCMEP with respect to their affinity for FlAsH and ReAsH. For both fluorophores, we observed a higher affinity for the FLNCCPGCCMEP motif. For both target sequences, FlAsH showed more stable interactions than ReAsH. To demonstrate selective labeling of different proteins in the same cell, we used two target proteins that are localized in different cellular compartments. As model proteins we chose a plasma membrane-localized G protein-coupled receptor for PTH (PTH-receptor), which was Cterminally modified with the FLNCCPGCCMEP motif, and the cytosolic β-arrestin-2 protein, which was C-terminally modified with the CCPGCC motif. Both proteins were co-expressed in HEK cells and labeled using the following procedure. First, cells were incubated with ReAsH and both tetra-cysteine binding motifs were labeled. Next, using a wash protocol employing 2,3-dimercaptopropanol (BAL), we were able to selectively remove ReAsH from the low affinity six amino acid motif while leaving the labeling to the

high affinity twelve amino acid motif intact. In a second labeling step, cells were incubated with FlAsH, which could now selectively bind to the vacant six amino acid motif. Dual imaging using confocal microscopy confirmed the ReAsH labeling of plasma membrane localized PTH receptors and FlAsH labeling of cytosolic β-arrestin-2. Upon agonist stimulation of the PTH receptor, β-arrestin-2 translocates to the plasma membrane. This assay was used to confirm the specific labeling. Upon application of 1 µM PTH, β-arrestin-2 was visibly recruited to the plasma membrane. Fluorescence intensity as measured by cross sectioning of the cells prior and after agonist application further proves the translocation of the FlAsH label to the plasma membrane. Taken together our data demonstrate that FlAsH and ReAsH can be used for orthogonal labeling to different binding motifs fused to different target proteins in living cells. 1. Dept. of Pharmacology, University of Würzburg, Würzburg, Germany

37 Muscarinic receptor interaction of novel allosteric/orthosteric antagonist ligands Kaufel D. (1), Schmitz J. (2), Holzgrabe U. (2), Mohr K. (1) The ligand binding pocket of muscarinic acetylcholine receptor subtypes contains a less well conserved allosteric domain that is located extracellular to the highly conserved orthosteric binding site. Archetypal allosteric modulators such as bis(ammonio)alkane-type compounds display highest binding affinity for the M2-subtype and lowest for M5. In an attempt to exploit the allosteric site for subtype-selectivity and the orthosteric site for high affinity, allosteric/orthosteric hybrid compounds were designed. In the case of receptor activating hybrids, this strategy proved to be successful [1]. Here, we checked whether the dualsteric ligand concept is likewise applicable for the design of subtype-selective antagonists. A newly synthesized hybrid was used in which the orthosteric agonist moiety was replaced by the inverse agonist N-methylscopolamine. Receptor binding of the antagonist hybrid was studied in wild-type and point-mutated human muscarinic receptors that were expressed in CHO cells (orthosteric radioligand [3H]Nmethylscopolamine, 0.2 nM; buffer composition 10mM HEPES, 10mM MgCl2, 100mM NaCl, pH 7.4, 30°C). In contrast to the corresponding agonist hybrid, the antagonist hybrid did not display M2 over M5 binding selectivity for free receptors. Correspondingly, binding of the antagonist hybrid was not sensitive to the allosteric double mutant M2177Tyr→Gln + 423Thr→His, in which M2 selectivity-providing amino acids are replaced by their corresponding M5 counterparts. With the orthosteric receptor mutant M2104Tyr→Ala, binding affinity of the antagonist hybrid was significantly reduced relative to wild-type M2. In conclusion, although sharing an identical allosteric building block with the agonist hybrid, the antagonist hybrid fails to exploit M2 selectivityproviding amino acids of the allosteric site. [1] Antony et al. (2009) FASEB J. in press 1. Pharmacology & Toxicology, Institute of Pharmacy, University of Bonn, 53121 Bonn, Germany, 2. Institute of Pharmaceutical Chemistry, University of Würzburg, 97074 Würzburg, Germany

38 Role of the muscarinic allosteric site for receptor activation: the conserved epitope Trp 7.35 is critical for the action of acetylcholine at the hM4 subtype Janßen N. (1), Kebig A. (1), Mohr-Andrä M. (1), Mohr K. (1) In addition to the orthosteric acetylcholine binding site muscarinic receptors contain an allosteric binding site that is located in the extracellular entrance of the ligand binding pocket. Archetypal muscarinic allosteric modulators are subtype-selective with a preference for M2 and M4. Previously, we have shown that the conserved amino acid Trp 7.35 at the beginning of TM 7 provides subtype-independent baseline affinity for various allosteric agents [1]. In addition we found for the hM2 receptor that this epitope (M2 422 Trp) is critical for the potency of the orthosteric endogenous agonist acetylcholine [2]. In the present study we aimed at elucidating the role of Trp 7.35 for acetylcholine action on the M4 subtype that shares the Gi/o coupling preference with the M2 subtype. CHO cells were stably transfected either with the hM4 wild type receptor gene or with the point-mutated receptor gene hM4435Trp→Ala. Receptor mediated G protein-activation was measured in CHO cell membranes using a [35S]GTPyS-binding assay (0.07nM [35S]GTPγS, 10 µM GDP, 10 mM HEPES, 10 mM MgCl2, 100 mM NaCl, pH 7.4, 30°C). Incubation was terminated after 1 h by vacuum filtration and [35S]GTPγS-binding was measured. The potency of acetylcholine for G protein-activation amounted to pEC50 M4: 6.29 ± 0.05 and to pEC50 M4435Trp→Ala: 5.17 ± 0.11 (mean values ± S.E., n = 4). At the M2 subtype, a similar potency shift of about 1.15 decades was found. In both receptor pairs, the Trp 7.35→Ala mutation did not diminish the efficacy of acetylcholine relative to the respective wild-type receptor.In conclusion, the role of the allosteric epitope Trp 7.35 for physiological receptor activation is conserved among the M4- and the M2-subtype of the muscarinic acetylcholine receptor.[1] Prilla et al., (2006) Mol. Pharmacol. 70: 181196[2] Jäger et al., (2007) J. Biol. Chem. 282: 34968-34976 1. Institute of Pharmacy, Friedrich-Wilhelms University, Bonn, Germany

39 Activation of small conductance KCa channels regulates the [Ca2+]i influx and provides neuroprotection against glutamate-induced neurotoxicity Dolga A.M. (1), Nijholt I.M. (2), Eisel U.L.M. (2), Culmsee C. (1) Previous studies have shown that tumor necrosis factor-alpha (TNF-α) induces neuroprotection against excitotoxic damage in primary cortical neurons via sustained nuclear factor-kappa B (NF-κB) activation. The transcription factor NF-κB can regulate the expression of small conductance calcium-activated potassium (KCa) channels. Small conductance KCa channels are functionally coupled with NMDA receptors and their activity modulates the shape of excitatory postsynaptic potentials (EPSPs). By regulating after-hyperpolarization currents and reducing NMDA-receptor activity in a

15

variety of neurons, small conductance KCa channels may reduce neuronal excitability and as such may yield neuroprotection against neuronal overstimulation. In the present study we investigated whether activation of small conductance KCa channels induces cellular survival under stress-related circumstances. In addition we studied whether activation of TNF-α-mediated neuroprotective signaling is inducing changes in the expression of small conductance KCa channels. Primary cortical neurons treated with different activators of small conductance KCa channels were analyzed by single-cell fluorescence imaging. Changes in intracellular [Ca2+] influx were recorded for a period of 20 minutes. Calcium measurements analysis showed that activation of small conductance KCa channels by 6,7-dichloro-1H-indole-2,3-dione 3-oxime (NS309) reduced the glutamate-induced [Ca2+]i influx. Interestingly, this drug operated at the level of the second peak of glutamate-induce increase in [Ca2+]i. Dampening the [Ca2+]i influx, small conductance KCa channel activators led to neuroprotection against the glutamate challenge. The expression of small conductance KCa2.2 channel was upregulated by TNF-α treatment in a time-dependent manner. The increase in small conductance KCa2.2 channel expression after TNF-α treatment was shown to be dependent on TNF-R2. Furthermore, treatment with the small conductance KCa channel blocker apamin reverted the neuroprotective effect elicited by TNF-α. We conclude that treatment of primary cortical neurons with TNF-α increases KCa2.2 channel expression, which renders neurons more resistant to excitotoxic cell death. Small conductance KCa channels-mediated neuroprotection effect place small conductance KCa channels as possible targets for the treatment of disorders linked to neuronal hyperexcitability. 1. Klinische Pharmazie, Institut fuer Pharmakologie und Toxikologie, PhilippsUniversitaet Marburg, Germany, 2. Department of Molecular Neurobiology, University of Groningen, Haren, The Netherlands

42 Transcription factor CREB is critically implicated in β-adrenergic receptormediated detrimental cardiac effects Staab A. (1), Frebel K. (1), Matus M. (1), Nunes F. (1), Schulte J.S. (1), Seidl M.D. (1), Stümpel F. (1), Schmitz W. (1), Müller F.U. (1) The transcription factor cAMP response element (CRE)-binding protein (CREB) plays acritical role in the regulation of gene expression in response to activation of the cAMPdependent signalling pathway. A chronic stimulation of this pathway, e.g. by elevated plasma catecholamines, is regarded as a pivotal step in the pathogenesis of heart failure. Mice with a cardiomyocyte-specific inactivation of CREB (hsCREB-KO) were treated with the β-adrenoceptor agonist isoprenaline (iso; 10mg per kg body weight and day) to test whether CREB is critical for detrimental β-adrenoceptor-mediated cardiac changes. Differences in left-ventricular (LV) function were observed, measured by LV catheterization after seven days of stimulation. Cardiac output, heart rate and rate of contraction were reduced in WTiso in contrast to KOiso animals (e.g. rate of contraction at maximal acute isoprenaline stimulation; in mmHg/s, mean±SEM: WTvehicle 10638±605, n=9; hsCREB-KOvehicle 9110±1260, n=7; WTiso: 5974±729, n=8; hsCREB-KOiso 10214±449*, n=9; *p<0,05 vs. WTiso). In order to search for proteins up- or downregulated along with the functional rescue in isoprenaline-treated mice, we performed a DIGE (differential in gel electrophoresis) analysis.Three proteins were identified by MALDI-TOF/mass spectrometry which were upregulated in isolated adult cardiomyocytes from KO versus WT mice.In conclusion, the transcription factor CREB is critical for cardiac deterioration after β-adrenergic stimulation, suggesting CREB as a transcription factor implicated in the pathophysiology of heart failure. (Supported by the DFG) 1. Institute of Pharmacology and Toxicology, University Münster, Germany

40 Allosteric muscarinic M1 receptor action of hybrids related to the M1 agonist xanomeline and to tacrine Jumpertz S. (1), Fang L. (2), Decker M. (3), Zhang Y. (2), Mohr K. (1), Tränkle C. (1) A dimer of the muscarinic antagonist and atypical modulator tacrine showed increased allosteric potency compared to tacrine while still fitting into the allosteric binding area of muscarinic M2-receptors [1]. Here, we tested in M1 receptors the binding of novel hybrid compounds (cf. Table1) in which moieties related to the M1 agonist xanomeline were attached to tacrine. Allosteric b inding was measured in membranes of chinese hamster ovary O HN

N

[ ]m N H

[ ]n

O

O

mn cpd 1 8 1 cpd 2 8 5 cpd 3 3 2

N S N

N

(means ± SEM, n = 3-4)

compound xanomeline tacrine 1 2 3

-log EC50,diss 4.95±0.03 5.18±0.02 6.35±0.03 7.08±0.03 8.03±0.03

nH -1.10±0.09 -1.91±0.14* -1.43±0.16* -1.94±0.25* -1.41±0.13*

cells stably expressing M1 receptors (4 mM Na2HPO4, 1 mM KH2PO4, pH 7.4, 23 °C), whose orthosteric site was blocked by [3H]N-methylscopolamine ([3H]NMS). Retardation of [3H]NMS dissociation indicated allosteric binding of the test compounds; –log EC50,diss corresponds to the concentration for a half maximal reduction of the [3H]NMS off-rate as a measure of affinity. The -logEC50,diss values of the hybrids clearly exceeded those of xanomeline and tacrine up to 1000fold (cf. Table). The corresponding curve slopes nH were significantly greater than unity (*F-test, P<0.05) suggesting an atypical allosteric behaviour of the hybrids. Taken together, the increase in allosteric binding affinity by hybrid formation indicates that the allosteric site of the M1 receptor may favourably accommodate xanomeline-like moieties in addition to tacrine. [1] Tränkle et al. (2005) Mol Pharmacol. 68:1597-1610 1. Pharmacology & Toxicology, Institute of Pharmacy, University of Bonn, Bonn, Germany, 2. Center of Drug Discovery, China Pharm. University, Nanjing, PRC, 3. School of Pharmacy, Queen's University Belfast, Belfast, UK

41 Subunit stoichiometry of heterologously expressed P2X receptors analyzed by single-molecule spectroscopy Eisele T. (1), Hausmann R. (1), Gregor I. (2), Enderlein J. (3), Schmalzing G. (1) P2X receptors assemble from a repertoire of seven subunit isoforms to form six homotrimeric channels composed of P2X1 - P2X5 or P2X7 subunits. Moreover, at least seven heterotrimeric receptor channels have been identified including P2X1+2 (Aschrafi et al., JMB 342:333-43, 2004), P2X1+4, P2X1+5, P2X2+6, P2X4+6, P2X2+3, and P2X4+7. For the P2X2+3 heteromer, a fixed subunit stoichiometry of one P2X2 subunit in a complex with two P2X3 subunits has been deduced from electrophysiological experiments (Jiang et al., J Neurosci. 23:8903-10, 2003). In contrast, analysis of the P2X2+6 heteromer by atomic force microscopy suggested that the subunit stoichiometry varied with the relative subunit expression level (Barrera et al., Biophys J. 93:505-12, 2007). Here, we assessed the subunit stoichiometry of heterotrimeric P2X1+2 receptors by two approaches. In the first method, we affinity-purified heterotrimeric P2X1+2 receptors from X. laevis oocytes and quantified the relative amounts of co-assembled plasma membrane-bound P2X1 and P2X2 subunits in SDS-PAGE gels. The data obtained are consistent with a subunit stoichiometry of one P2X2 subunit and two P2X1 subunits in the trimeric complex. This subunit stoichiometry was independent of the expression level. A 1:2 (P2X2:P2X3) subunit stoichiometry was also obtained for the P2X2+3 receptor analyzed as a positive control. As a second method, we used single molecule fluorescence spectroscopy to determine the reduction of the CFP lifetime within a FRET pair consisting of heterotrimeric P2X1-ECFP / P2X2-EYFP or P2X1-EYFP / P2X2-ECFP receptors. The deduced FRET efficiencies were also consistent with a fixed subunit stoichiometry of 1:2 of both, the P2X2/P2X1 heterotrimer and the P2X2/P2X3 heterotrimer. The work was financially supported by grants of the Deutsche Forschungsgemeinschaft (FOR450, TP11 and FOR748, TP 3). 1. Molecular Pharmacology, RWTH Aachen University, Aachen, Germany, 2. Dept. Molecular Neurosensorics, caesar research center, Bonn, Germany, 3. Institute of Physical and Theoretical Chemistry, Eberhard Karls University, Tübingen, Germany

43 Diagnostics of cardio-stimulatory beta1-receptor antibodies and evaluation of their blockade by novel cyclopeptides using FRET-microscopy Schlipp A. (1), Boivin V. (1), Nikolaev V.O. (1), Hauck P. (1), Ye Y. (2), Kocoski V. (3), Lohse M.J. (1), Jahns R. (2) Background: Dilated cardiomyopathy (DCM) is one of the main causes of severe heart failure in young adults, leading to progressive dilatation and pump failure of the heart. Up to 50% of DCM-patients develop agonist-like autoantibodies against the second extracellular loop of the β1-adrenergic receptor (β1-ECII-abs). These functionally active antibodies cause DCM in the rat. Here we attempt to block their harmful stimulatory effect in vitro by different cyclopeptides mimicking β1-ECII. Methods: Recently, we developed a novel diagnostic method to detect functionally active β1-ECII-abs using HEK-293 cells stably expressing the human β1-adrenergic receptor and an Epac1based cAMP-sensor. Upon β1-ECII-ab-induced stimulation, intracellular cAMP increases. Generated cAMP binds to the sensor resulting in conformational changes and thereby decreases fluorescence resonance energy transfer (FRET) between its chromophores CFP and YFP. The β1-ECII loop contains 3 cysteine residues. Two of them form an intraloop cysteine-bridge, one builds a cysteine-bridge between ECII and the third transmembrane domain (TM3) of the receptor. Two different ECII homologous 18aa (amino-acid) cyclopeptide-mutants (18aaCys/Cys/Ser and 18aaCys/Ser/Cys) have been tested for their capability of blocking activating antibodies. Results: 18aaCys/Cys/Ser efficiently blocked β1-ECII-abs in n=81/85 immunized rats. Only 4 animals exhibited β1-ECII-abs that were blocked by either 18aaCys/Cys/Ser or 18aaCys/Ser/Cys, as determined by ELISA. In most animals with the Cys/Cys/Sercyclopeptide (CP) ELISA signals were blocked by 62±3% compared to only 2±1% with the Cys/Ser/Cys-CP. Regarding functional activity, 18aaCys/Cys/Ser-CP also efficiently blocked the stimulatory effect of polyclonal rabbit and monoclonal mouse β1-ECII-abs in our novel FRET assay, whereas a scrambled peptide (containing the same aa in a randomized order) had no significant antibody-blocking effect. Conclusion: 18aaCys/Cys/Ser-cyclopeptides, containing an intraloop-bridge between the first and second cysteine of β1-ECII, exhibit a markedly superior blocking efficacy of functional β1-ECII-abs compared to 18aaCys/Ser/Cys-CP. Our results fit very well with the recently published crystal structure of the β1 adrenergic receptor. Our data suggests that functionally active β1-ECII-abs recognize conformational epitopes within the β1-ECII loop that can be mimicked by specific cyclopeptides in order to scavenge such harmful antibodies. 1. Institut für Pharmakologie und Toxikologie, Julius-Maximilians-Universität, Würzburg, Germany, 2. Medizinische Klinik und Poliklinik I / Kardiologie, Universitätsklinikum Würzburg, Würzburg, Germany, 3. Institut für Virologie und Immunbiologie, JuliusMaximilians-Universität, Würzburg, Germany

44 Defects in lung development of α2B-adrenoceptor-deficient mice Haubold M. (1), Hein L. (1) α2-adrenoceptors are not only essential presynaptic regulators of norepinephrine release from sympathetic nerves but also influence developmental signaling pathways and angiogenesis. Previous studies in gene-targeted mice have demonstrated that α2Badrenoceptors are essential for development of the placental vascular system between embryonic days E10.5 and E12.5. Thus this study was initiated to determine the significance of α2B-adrenoceptors for embryonic and perinatal development in mice. α2Bdeficient mice which survived the embryonic period were lost during the first 24 hours after birth. These α2B-/- mice were characterized by reduced body weight and dwarfism. Immediately postnatally breathing rate, drinking behavior, cardiac rhythm and heart rate were not altered in α2B-/- vs. α2B+/+ mice. Within the first hours after birth, α2B-/- mice rapidly developed cyanosis. Hepatic synthesis of hemoglobin chains did not differ between genotypes. Thorough investigation of heart, liver, gastrointestinal system and kidney from α2B-/- newborn mice did not reveal any morphological defects. Postnatal closure rates of the ductus arteriosus did not differ between genotypes. However, microscopical anatomy of the lung was severely altered in α2B-/- newborn mice. Alveolar spaces were significantly reduced and alveolar septi were abnormally thickened, indicating that the final phase of pulmonary development was delayed in α2B-/- mice. Several key factors required for lung development were differentially expressed between newborn α2B-/- and α2B+/+ mice. α-smooth muscle actin and sonic hedgehog (shh) were

16

significantly upregulated in α2B-deficient lung specimens. These results demonstrate that α2B-adrenoceptors play an important role for late stage embryonic lung development. 1. Institute of Experimental and Clinical Pharmacology and Toxicology, University of Freiburg

45 ß2-Sympathomimetics (ß2SYM) inhibit restitution of intracellular pH after acid load of HEK 293 cells transfected with ß2-adrenoceptors (ß2AR) and type II adenylyl cyclases Boronowski D. (1), Napiwotzki J. (1), Küper J. (1), Lemoine H. (1) ß2SYM used in the therapy of bronchial asthma act via ß2AR and type II adenylyl cyclase (ACII) and mediate relaxation of airway smooth muscle. In patients with acute asthma the pH of exhaled airway vapor condensate falls down up to pH 5.2 (Hunt et al., Am J Respir Crit Care Med 2000; 161:694); this acidification causes an efficacy loss of ß2SYM to activate AC and bronchial relaxation (Napiwotzki et al., this journal 2002; 365: R27). To investigate the effects of ß2SYM on the recovery of low intracellular pH (pHi), HEK 293 cells stably transfected with ß2AR and ACII (Bösel and Pfeuffer 1998; FEBS 422: 209) were cultivated on 12-well strips (96-well formate), preloaded with 2 µM BCECF (Molecular Probes B1150) and tested in recently developed multichannel(12) fluorescence detectors using 505/435 nm for excitation (ratio technique) and a long pass (530 nm) emission filter (Boronowski et al. 2008; this journal 377: 53). Acid load of HEK 293 cells was achieved with the NH4Cl (20 mM) prepulse technique using a standard physiological salt solution buffered with 20 mM HEPES in the absence of bicarbonate, a condition commonly used to study the Na+/H+ exchanger (NHE). Surprisingly, ß2SYM inhibited the recovery from low pHi of cells. In the presence of isoprenaline (ISO) the rate of pH extrusion decreased from 0.10-0.13 to 0.03-0.05 pH/min. This inhibiting effect occured with low ISO concentrations (pIC50=9.73±0.04) corresponding to a high effectiveness of ISO to stimulate AC (pEC50=7.95±0.02) and could be antagonised by 1 µM bupranolol. Receptor-saturating concentrations of formoterol, fenoterol and salbutamol induced a similar inhibition of pHi-recovery (see table) compared to maximum NHE-inhibition with cariporide. ßAR-ligand µM inhibition (%) &antagonist H+-extrusion For conclusion: A vicious cycle of pursueing low pHi Isoprenaline 20 61.6 ± 5.5 by blockade of H+-extrusion followed by an significant &Bupranolol 1.0 13.0 ± 2.5 attenuation of AC-activity and airway relaxation limits Formoterol 2.0 57.5 ± 5.5 the therapeutical benefit of ß2SYM during severe Salbutamol 100 54.0 ± 2.3 episodes of bronchial asthma. Fenoterol 100 54.3 ± 5.1 Cariporide 10 85.9 ± 2.7 1. Mol. Drug Research, Lasermedicine, Heinrich-Heine-Universität, Düsseldorf, Germany

46 Genetic dissection of α2-adrenoceptor functions in adrenergic versus nonadrenergic cells Gilsbach R. (1), Röser C. (2), Beetz N. (1), Hadamek K. (2), Leemhuis J. (1), Schneider J. (1), Urbanski M. (1), Szabo B. (1), Weinshenker D. (3), Hein L. (1) Previous studies revealed distinct physiological functions of the α2-adrenoceptor subtypes (α2A,B,CAR). Less is known whether these functions are mediated by α2AR in adrenergic neurons or by α2AR in non-adrenergic neurons. To clarify this issue, we used the dopamine β-hydroxylase (Dbh) promoter to drive expression of α2A-adrenoceptors exclusively in noradrenergic and adrenergic cells of transgenic mice. These mice were backcrossed with α2A/α2C-AR deficient mice to generate lines with selective expression of α2A receptors in adrenergic cells (A-/-C-/-Tg). These mice were subjected to a comprehensive phenotyping analysis and compared to wild-type mice, which express α2A- and α2C-receptors in both adrenergic and non-adrenergic cells, and α2A/α2Cdeficient mice, which do not express these receptors in any cell type. Expression and function of the Dbh-α2ATg was confirmed by quantitative RT-PCR in locus coeruleus and sympathetic ganglia, by immunohistochemistry, by electrophysiology and [3H]norepinephrine release experiments. Surprisingly, only a few functions previously ascribed to α2-adrenoceptors were mediated by receptors on adrenergic neurons, including feedback inhibition of norepinephrine release from sympathetic nerves and spontaneous locomotor activity. Other agonist effects including analgesia, hypothermia, sedation and anesthetic-sparing were mediated by α2-receptors in non-adrenergic cells. In dopamine β-hydroxylase knockout mice lacking norepinephrine, the α2-agonist medetomidine still induced a loss of the righting reflex, confirming that the sedative effect of α2-adrenoceptor stimulation is not mediated via autoreceptor-mediated inhibition of norepinephrine release. The present study paves the way for a revision of the current view of the α2-adrenoceptor function. 1. Institute of Exp. and Clin. Pharmacology and Toxicology, University of Freiburg, 2. Institute of Pharmacology and Toxicology, University of Würzburg, 3. Dept. of Human Genetics, Emory University, Atlanta, USA

sulprostone was used instead and L-826,266 was administered at 0.05, 0.32 and 2 µM, a pA2 of 7.6, based on a Schild plot, was obtained. Sulprostone also inhibited the electrically (3 Hz) evoked tritium overflow from cortex slices preincubated with 3Hserotonin and from vas deferens pieces preincubated with 3H-noradrenaline. Again, L826,266 0.32 µM shifted to the right the concentration-response curves of the EP3 agonist, yielding an apparent pA2 value of 7.7 in either paradigm. In a separate series of experiments on tissues superfused in the absence of naproxen, prostaglandin E2 1 µM did not affect the electrically (3 Hz) evoked tritium overflow from rat cortical, striatal and hippocampal slices preincubated with 3H-choline (representing quasi-physiological acetylcholine release). Prostaglandin E2 1 µM also failed to influence the electrically (0.3 Hz) evoked tritium overflow from guinea-pig retinal discs preincubated with 3Hnoradrenaline (representing quasi-physiological dopamine release). In conclusion, experiments based on the competitive EP3 receptor antagonist L-826,266 prove that the inhibitory effect of prostaglandins of the E series on monoamine release in the rat brain and on noradrenaline release in the rat vas deferens involves the EP3 receptor. Receptors for prostaglandins of the E series do not appear to influence acetylcholine release in the rat brain and dopamine release in the guinea-pig retina. 1. Institut für Pharmakologie und Toxikologie, Universität Bonn, Bonn, Deutschland

48 Characterization of expression and functional role of ß2-adrenoceptors in human lung fibroblasts Lamyel F. (1), Racké K. (1), Jürgens U.R. (2), Warnken M. (1) Fibrotic alterations are part of the airway remodelling processes observed in asthma and COPD. There is increasing evidence that in addition to acute bronchodilatory effects, classical anti-obstructive drugs such as ß-adrenoceptor agonists may also modulate long term remodelling processes. The present study aimed to explore ß-adrenergic mechanisms in lung fibroblasts. Primary human lung fibroblasts and the human lung fibroblast cell line MRC-5 were cultured as described (Matthiesen et al. AJRCMB 2006, 35:621). Using RT-PCR mRNA encoding ß2-adrenoceptors was clearly detectable in primary human lung fibroblasts as well as in MRC-5 cells, whereas in both cell types mRNA for ß1- and ß3-adrenoceptors could not be detected. Expression of ß2adrenoceptors was confirmed at protein level by Western blot analysis. After 24 h exposure to dexamethasone (1 µM) or budesonide (100 nM) ß2-adrenoceptor mRNA was increased by about 60%, whereas 24 h treatment with TGF-ß (5 ng/ml) caused an decrease by about 70%, an effect not significantly affected by 1 µM dexamethasone. Proliferation of MRC-5 cells as determined by 3H-thymidine incorporation was inhibited by isoprenaline (IC50 3 nM) and formoterol (IC50 57 pM), maximally by about 30-40%. The concentration response curve of formoterol was not affected by the ß1-selective antagonist GCP 20712 (3 µM), but markedly shifted to the right by the ß2-selective antagonist ICI 118,551 (0.3 and 3 µM) (mean apparent pA2 value: 9.6). Furthermore, collagen synthesis as determined by 3H-proline incorporation was also inhibited in a concentration-dependent manner by formoterol (IC50 91 pM) and isoprenaline (IC50 1 nM), maximally by about 30 %. Under basal culture conditions, MRC-5 show significant expression of α-smooth muscle actin, a marker of myo-fibroblast differentiation (determined by Western blot analysis). Basal α-smooth muscle actin expression was reduced following 24 h exposure to 1 and 100 nM formoterol by 48±4 and 46±3% (means±SEM), respectively. In conclusion, human lung fibroblasts express ß2-adrenoceptors which mediate inhibitory effects on different pro-fibrotic features. 1. Dept. of Pharmacology & Toxicology, University Bonn, Bonn, Germany, 2. Dept. of Pulmology, Univ. Hospital Bonn, Bonn, Germany

49 Tryptophan hydroxylases as drug targets Matthes S. (1), Tenner K. (1), Schütz A. (2), Bashammakh S. (1), Bader M. (1) Tryptophan hydroxylases (TPH) are the rate-limiting enzymes in serotonin (5-HT) synthesis. Two isoforms exist, TPH1 and TPH2. TPH2 is mainly expressed in the brain and is responsible for the local synthesis of 5-HT, which has important functions and is a major drug target, e.g. for depressive disorders. In contrast, TPH1 is mainly expressed in the gut and generates 5-HT taken up into platelets and transported and released in the blood stream. Platelet serotonin is involved, e.g., in hemostasis and in growth and regeneration processes in peripheral organs. Thus, the two TPH isoforms are interesting drug targets for so different diseases as depression and thrombosis. We have developed two high-throughput screening methods to find substances which either activate TPH2 or inhibit TPH1. One is based on the different fluorescence characteristics of the TPH substrate tryptophan and the product 5-hydroxytryptophan. The second is a cell-based assay relying on a suicide drug, 7-hydroxytryptophan, which kills cells depending on their intrinsic TPH activity. With these assays we have already performed screens of 37.000 chemicals and discovered several classes of substances for both purposes. These substances will be further developed and tested for their efficiencies in animal models including knockout mice for both TPH isoforms to ensure specificity. 1. Max-Delbrück-Center for Molecular Medicine (MDC), Berlin-Buch, Germany, 2. Helmholtz-PSPF, MDC, Berlin-Buch, Germany

47 The EP3 receptor is involved in the inhibitory effect of prostaglandins of the E series on monoamine release in rat tissues: proof with L-826,266 Günther J. (1), Schulte K. (1), Kurz C.M. (1), Schlicker E. (1) The inhibitory effect of prostaglandins of the E series on nordrenaline release (i) in the brain and (ii) in sympathetically innervated peripheral tissues and (iii) on serotonin release in the brain of the rat has been known for many years. The aim of the present study was to study whether EP3 receptors are implicated, using the selective antagonist L-826,266 (compound 10b in the paper by Juteau et al., Bioorg Med Chem 9, 1977, 2001). Rat tissues were preincubated with 3H-noradrenaline (cerebral cortical slices, vas deferens pieces) or 3H-serotonin (cerebral cortical slices) and then superfused with medium containing the cyclooxygenase inhibitor naproxen 10 µM. The electrically (0.3 Hz) evoked overflow from cortex slices preincubated with 3H-noradrenaline was inhibited by prostaglandin E2 and its concentration-response curve was shifted to the right by L826,266 0.1 µM, yielding an apparent pA2 of 7.7. When the EP3 receptor agonist

50 The role of the postsynaptic 5-HT1A-receptor for the regulation of body temperature - a radio telemetry study Bert B. (1), Haberzettl R. (1), Dietze S. (2), Fink H. (1) The serotonin1A (5-HT1A) -receptor appears somatodendritically on serotonergic neurons and postynaptically on subsequent neurons. So far, the functionality of both receptor locations for the regulation of body temperature has not been fully clarified. For men and rats a postsynaptic mechanism has been assumed whereas for mice presynaptic 5HT1A-receptors seem to be more important for thermoregulation. At our institute exists a transgenic mouse line with a distinct overexpression of the 5-HT1A-receptor in cortex and hippocampus, both projection areas of serotonergic neurons. A previous study using a rectal probe has shown that male transgenic mice exhibit a slightly lower baseline body temperature and respond with an exaggerated hypothermic effect to the

17

administration of the 5-HT1A-receptor agonist 8-OH-DPAT (Bert et al. 2006, Behav Brain Res 167:321-41), suggesting a dominant role of the postsynaptic receptor. In the present study we used radio telemetry to further clarify the involvement of the postsynaptic 5-HT1A-receptor. We investigated whether the hypothermic effect of 8-OHDPAT (i.p.) could be abolished by the silent 5-HT1A-receptor antagonist WAY100635 (1 mg/kg, i.p.) in male transgenic and wild-type mice. Additionally we tested the effect of the selective serotonin reuptake inhibitor fluoxetine (5, 10, 20 mg/kg, i.p.) and we expected that the response of transgenic mice would be more pronounced. Our results show that in wild-type mice 1 mg/kg 8-OH-DPAT did not affect body temperature when measured by radio telemetry, whereas a clear hypothermic effect was previously observed by rectal recordings. In transgenic mice 8-OH-DPAT induced a hypothermic effect measured by radio telemetry, but surprisingly the double dose (1 mg/kg) was needed than the one used for the rectal reading (0.5 mg/kg). This hypothermic effect was antagonised by 1 mg/kg WAY100635. The administration of 5 mg/kg fluoxetine was ineffective in both genotypes. In transgenic mice 10 mg/kg fluoxetine induced hypothermia, whereas only 20 mg/kg were effective in wild-type mice. In summary, our data show that recording the body temperature in freely moving animals by radio telemetry might lead to different outcomes of drug effects compared to rectal readings, especially concerning the dose-response relationship. Additionally, the higher sensitivity of transgenic mice towards 8-OH-DPAT assessed by both methods and fluoxetine argues for an involvement of postsynaptic 5-HT1A-receptors in the regulation of body temperature in mice. 1. Institute of Pharmacology and Toxicology, Freie Universität Berlin, Germany, 2. CCR/Institute of Pharmacology, Charité-Universitätsmedizin Berlin, Germany

51 Phloroglucinol and its derivatives as therapeutically interesting ion-channel modulators Chatterjee S.S. (1), Yatsenko N. (2), Maximyuk O. (2), Kovalskyy D. (3), Krishtal O. (2) Structure-activity studies conducted with hyperforin and some other structurally analogous bio-active plant metabolites strongly suggested that the condensed 1,3,5trihydroxybenzene-moiety could be a common essential pharmacophore for their inhibitory effects on NMDA receptor-gated and other ion channels. Therefore, it was of interest to define such activities of the parent molecule phloroglucinol, which is also a secondary plant metabolite with demonstrated efficacy in patients suffering from dyspeptic disorders. Unlike its other naturally occuring derivatives and analogues studied to date, incubation of acutely isolated rat brain neurones for about 5 minutes with phloroglucinol enhanced the NMDA receptor-gated ionic current. Its observed effects persisted after prolonged incubation or after several thorough washings. Its effective concentration was 1µM or lower, after which it had no observable effects on ten other voltage- and receptor-gated ion channels tested. These and other observations, made during screening of numerous synthetic derivatives of the parent molecule, led us construct several phloroglucinol based chemical libraries with desired spectrums of ion cannel modulating activities. Safety. oral bio-availability and CNS-function modulating activity profiles of several synthesized phloroglucinol derivatives indicate that the assembled chemical libraries could be a starting point for obtaining structurally and functionally novel psychoactive drugs, hits and leads. 1. Karlsruhe, Germany, 2. Bogomoletz Institute of Physiology, Kiev, Ukraine, 3. Research Department Enamine, Kiev, Ukraine

52 Species differences between the effects of antiepileptic drugs on glutamate release in rat and human neocortical synaptosomes Kammerer M. (1), Brawek B. (1), Freiman T.M. (2), Jackisch R. (3), Feuerstein T.J . (1) Carbamazepine (CBZ), lamotrigine (LTG), phenytoin (PHT), gabapentin (GBP) and pregabalin (PGB) are common antiepileptic drugs (AEDs). In epileptic brain regions there is an increase in the release of the excitatory amino acid neurotransmitter glutamate. Therefore, many AEDs aim at decreasing glutamate concentrations in the synaptic cleft, e.g. via blocking voltage-gated ion channels in glutamatergic neurons. The present study investigated and compared the effects of AEDs on [3H]-glutamate release from both rat and human neocortical synaptosomes preincubated with [3H]glutamate. Release was evoked by (1) the Na+-channel activator veratridine (10 µM) and (2) by elevation of [K+] to 15 mM, which mainly leads to Ca2+-dependent exocytosis. (1) Veratridine-evoked [3H]-glutamate release from rat neocortical synaptosomes was significantly diminished by AED concentrations of 100 µM (CBZ: -39%, LTG: -37%, PHT: -46%), whereas CBZ, LTG, PHT at 10 µM, as well as PGB and GBP at 100 µM were ineffective. Veratridine-evoked [3H]-glutamate release from rat neocortical synaptosomes was higher (20% of tissue-³H) than that from human synaptosomes (9% of tissue-³H, p < 0.001). The effects of the AEDs, however, were similar: at 100 µM, CBZ, LTG, PHT diminished the evoked [3H]-glutamate release by approximately 38%, whereas GBP and PGB were ineffective. At a lower concentration of 10 µM, CBZ, LTG and PHT had no significant effect. (2) At AED concentrations of 100 µM, K+-evoked [3H]-glutamate release from rat neocortical synaptosomes was diminished by CBZ (22%), LTG (-21%), PHT (-41%) and PGB (-13%), whereas at 10 µM CBZ, LTG, PHT as well as GBP at 100 µM were ineffective. In contrast, none of the tested AEDs had a significant effect on K+-evoked [3H]-glutamate release in human neocortical synaptosomes. The amount of K+-evoked [3H]-glutamate release was higher in rat (4% of tissue tritium) than that in human synaptosomes (3% of tissue tritium, p < 0.02) Our results show a surprising species difference in the effects of the Na+-channel blockers (CBZ, LTG, PHT) on K+-evoked release of glutamate, while the effects of these drugs on that induced by veratridine were quite similar. Since it has been reported that veratridine-induced glutamate release is mainly mediated by a reversal of Na+dependent glutamate carriers, our data suggest an involvement of this mechanism in the effects of CBZ, LTG and PHT, as well as species differences in the amount of Na+ influx following K+ stimulation. 1. Section of Clinical Neuropharmacology of the Department of Neurosurgery, Neurozentrum, Albert-Ludwigs University, Freiburg, Germany, 2. Department of Neurosurgery, Neurozentrum, Albert-Ludwigs University, Freiburg, Germany, 3. Department of Experimental and Clinical Pharmacology and Toxicology, Albert-Ludwigs University, Freiburg, Germany

53 Adenosine A2B-receptor-mediated activation of the transcription factors NFkappaB and AP1 in human mast cells von Kügelgen I. (1), Meis K. (1), Spitzlei P. (1), Molderings G.J. (2) The activation of adenosine A2A-receptors has been shown to promote antiinflammatory effects in human mast cells, whereas the activation of A2B-receptors mediates pro-inflammatory responses including the production and secretion of cytokines such as interleukin-8 (IL-8). In the present study, we searched for changes in transcription factors due to the activation of A2A- and A2B-receptors in a human mast cell (HMC1) line. Receptor-mediated changes in transcription factors were assessed by a reporter gene (luciferase) assay in cultured HMC1 cells transiently transfected with either the pAP1 (activator protein-1)-luc, the pCRE (cAMP response element)-luc, the pNFAT (nuclear factor of activated t cells)-luc or the pNFkappaB (nuclear factor kappa B)-luc vectors. Adenosine (10 to 30 µM) and its analogue NECA (N-ethylcarboxamidoadenosine; 0.3 to 10 µM) increased the AP1-, CRE-, NFAT and NFkappaB-controlled luciferase expression about 10-fold, 50-fold, 5-fold and 2-fold, respectively. The A2Breceptor antagonist MRS 1754 (8-(4-[{(4-cyanophenyl)carbamoylmethyl}oxy]phenyl)-1,3di(n-propyl)xanthine, 1 µM) markedly reduced (CRE, NFAT) or almost abolished (AP1, NFkappaB) the responses to NECA. The A2A-receptor agonist CGS21680 (0.3 to 3 µM) increased the CRE-directed luciferase expression about 30-fold and the NFAT-directed luciferase expression about 2-fold, but had no effects on the AP1- and NFkappaBdirected luciferase expression. Moreover, the A1-receptor agonist cyclopentyladenosine (CPA, 1 µM) and the A3-receptor agonist N6-(3-iodobenzyl)adenosine-5'-Nmethylcarboxamide (IB-MECA, 0.1 µM) failed to cause changes in the NFkappaB- and AP1-directed luciferase expression. In addition to its effects on transcription factors, NECA (0.3 to 3 µM) markedly increased the IL-8 production; this response was markedly reduced by MRS1754 (1 µM) and the NFkappaB-inhibitor curcumin (10 µM). The results indicate that adenosine mediates stimulatory effects on the transcription factors NFkappaB and AP1 in human mast cells by activation of A2B-receptors, but not by A2A-, A1- or A3-receptors. Moreover, the data suggest an involvement of the NFkappaB- and AP1-pathways in pro-inflamatory responses of human mast cells to activation by adenosine. 1. Dept. of Pharmacology, University of Bonn, Bonn, Germany, 2. Insitute of Human Genetics, University of Bonn, Bonn, Germany

54 Identification of the site of interaction of new reactive blue 2-derived antagonists at the human platelet P2Y12-receptor Hoffmann K. (1), Baqi Y. (2), Glänzel M. (3), Müller C.E. (2), von Kügelgen Î. (1) The P2Y12-receptor plays an important role in platelet aggregation. In the present study, we searched for amino acid residues of the human P2Y12-receptor involved in the binding of non-nucleotide antagonists. Receptor function was assessed by measuring the cAMP response element (CRE) directed luciferase expression in Chinese Hamster Ovary cells (Hoffmann et al., 2008, Biochemical Pharmacol 76:1201-1213). The cellular cAMP production was accelerated by forskolin. 2-Methylthio-ADP was used to activate the recombinant human P2Y12-receptor or mutant constructs. In cells expressing wild type receptors 2-methylthio-ADP inhibited the CRE-dependent luciferase expression with a half-maximal concentration of about 1 nM. The anthraquinone derivative reactive blue 2 (RB 2) used at increasing concentrations shifted the concentration-responsecurve of 2-methylthio-ADP to the right in a manner compatible with competitive antagonism (pA2-value 7.4). Its novel analogue PSB-0739 (lacking a sulfonic acid residue at ring F) showed a markedly higher antagonistic potency with a pA2-value of 9.8. In cells expressing the R256A-mutant receptor, the potencies of both RB 2 (apparent pKB 5.9) and PSB-0739 (apparent pKB 9.1) were decreased. The same difference was observed for the pure structural isomers RB 2 meta, RB 2 para and cibacron blue3-GA (ring F ortho-isomer of RB 2; for chemical structures see Glänzel et al., 2005, Eur J Med Chem 40:1262-1276), as well as for PSB-0801 (an analogue of PSB-0739): all compounds showed decreased pKB-values at the R256A-mutant construct when compared to those determined at the wild type receptor. In contrast, the analogue PSB-0826 (lacking a sulfonic acid residue at ring D) showed similar apparent pKB-values at wild-type (8.4) and R256A-mutant receptors (8.3). The results demonstrate that PSB-0739 is the most potent competitive non-nucleotide antagonist at the human P2Y12-receptor described so far. A sulfonic acid residue at ring D is likely to be involved in the interaction of antagonists derived from RB 2 with the residue Arg256 of the human P2Y12-receptor. 1. Dept. of Pharmacology, University of Bonn, Germany, 2. Pharmaceutical Chemistry I, University of Bonn, Germany, 3. Dept. of Pharmacology, University of Freiburg, Germany

55 Antipodal effects of ATP on neck muscle nociception mediated by purinergic P2X and P2Y receptors Ellrich J (1), Reitz M (1) Purinergic mechanisms play an important role in muscle nociception. Studies on muscle pain typically apply α,β-meATP as noxious agent due to its narrow receptor profile (P2X1, P2X3, P2X2/3) and sustained stability in tissue. In contrast, native ATP is quickly degraded to metabolites and interacts with excitatory P2X and inhibitory P2Y receptors. The experimental pharmacological study compares effects of both molecules in myofascial nociception in a model of neck muscle pain. Noxious stimulation of semispinal neck muscles was performed by intramuscular (i.m., 20 µl) bilateral injection of α,β-meATP (100 nM, 1 µM) or native ATP (100 nM, 1 µM, 7.6 mM) in anesthetized C57BL/6N mice (n=65). The impact of neck muscle noxious input on brainstem sensory processing was monitored by the jaw-opening reflex elicited via electrical tongue stimulation. Neuronal excitability of the brainstem reflex network was assessed by calculation of the reflex integral. The P2Y1 receptor antagonist MRS2179 (1 µM, 20 µl) or the P2Y1 receptor agonist 2-MeSADP (1 µM, 20 µl) were i.m. administered 20 min before ATP or two hours after α,β-meATP application, respectively. Injection of α,βmeATP increased the reflex integral within two hours in a dose-dependent manner (100 nM: +98%, 1 µM: +171%; p<0.05). In contrast, native ATP injection evoked facilitation with low dosage (100 nM: +30%) but not with high dosage (1 µM, 7.6 mM). Preceding

18

blockade of P2Y1 receptors by MRS2179 (1 µM) revealed reflex facilitation even under high dosage of native ATP (1 µM: +83%). Ongoing reflex facilitation after injection of α,β-meATP (1 µM) was reversed back to baseline by subsequent activation of P2Y1 receptors via 2-MeSADP (1µM). Whereas α,β-meATP induced brainstem reflex facilitation in a dose-dependent manner, native ATP was only effective with low dosage. P2Y1 blockade revealed excitatory ATP effects on P2X receptors. P2Y1 activation counteracted P2X3 excitation by α,β-meATP. Results demonstrate opposing excitatory P2X3 and inhibitory P2Y1 effects of ATP in neck muscle nociceptive processing in mice. These antipodal ATP effects may be due to facilitation of P2X3 receptor-channel desensitization by P2Y1 receptors as shown under in-vitro conditions (Br J Pharmacol 151: 226-36, 2007). The mechanisms may be involved in the pathophysiology in man and could indicate new options in the treatment of headache and neck pain. 1. Medical Physiology & Experimental Pharmacology Group, Aalborg University, Denmark

56 Differences of ATP-induced increases in intracellular Ca2+ in a Lesch-Nyhan disease cell model Erdorf M. (1), Seifert R. (2) Lesch-Nyhan disease (LND) is a rare disorder caused by the lack of the purine salvage enzyme hypoxanthine guanine phosphoribosyl transferase (HPRT) [1]. The link between the metabolic defect and the severe neuropsychiatric symptoms like self-injurious behavior, dystonia and mental retardation is still unknown. In HPRT-deficient rat B103 neuroblastoma cells abnormalities in purine and pyrimidine nucleotide content are found [2]. Hence, we asked the question whether there are also abnormalities in the regulation of receptors for extracellular purine and pyrimidine nucleotides. To analyze cellular signal transduction pathways, cultured cells were loaded with the fluorescent dye fura-2 acetoxymethylester (AM) and stimulated with different nucleotide triphosphates (NTPs) and diphosphates (NDPs). The resulting increase in intracellular Ca2+ ([Ca2+]i) was measured using a TECAN fluorescence plate reader [3]. Stimulation by ATP, ADP, GTP and GDP (100 µM each) yielded significant differences in [Ca2+]i between control- and HPRT-deficient cells up to 50 %. Kinetic [Ca2+]i studies showed cell line-specific curves. In control cells calcium responses were clearly faster than in HPRT-deficient cells, but the signal duration was shorter in control cells. The maximum increase in [Ca2+]i resulting from stimulation of cells with 100 µM ATP was 6.5 ± 0.75 µM for control cells, whereas in HPRT-cells the maximum [Ca2+]i was significantly reduced and reached only 2.5 ± 0.3 µM. After removing extracellular calcium, the maximum increase in [Ca2+]i was significantly lower, but the difference between control cells (0.66 ± 0.1 µM) and HPRTcells (0.24 ± 0.04 µM) was still evident. These findings suggest that ATP induces intracellular Ca2+ mobilization and extracellular Ca2+ influx in B103 cells and that HPRTdeficient cells are less responsive than control cells. Future studies will be directed towards the identification of the specific receptors involved in the calcium responses. [1] Seegmiller et al. (1967); Science 155:1682-4. [2] Zoref-Shani E. et al. (1993); J. Neurochem. 61:457-63. [3] Lin, Sadée, Quillan (1999); BioTechniques 26:318-26. 1. Dept. of Pharmacology, University of Regensburg, Regensburg, Germany, 2. Inst. for Pharmacology, Hannover Medical School, Hannover, Germany

57 Evidence of functional P2Y but not P2X nucleotide receptors in cultured cerebrocortical neurons of the rat Fischer W. (1), Franke H. (1), Sobottka H. (1), Illes P. (1), Nörenberg W. (1) There is growing interest in the characterization of P2 nucleotide receptors implicated in important neuronal cell functions and intercellular signalling. Whether or not various subtypes of P2X (ionotropic) or P2Y (metabotropic) receptors are present in the CNS is still a matter of controversy. In the present study, the expression and functionality of P2X/P2Y receptors were investigated in multipolar (non-pyramidal) neurons of mixed cerebrocortical cell cultures using immunocytochemistry, fura-2 microfluorimetry and patch-clamp techniques. The vast majority of neurons were GABA-immunopositive and expressed predominantly P2X2,4,6 and P2Y1,2 subtypes. P2X1 and P2X7 receptor immunoreactivity was found on thin axon-like processes and presynaptic structures. External ATP caused a small concentration-dependent increase in intracellular Ca2+ concentration ([Ca2+]i) in most investigated neurons, whereas only about the half of these cells responded to BzATP, ADPbS, 2MeSADP, 2-MeSATP and to UTP. In contrast, a,b-meATP, UDP and UDP-glucose failed to produce any [Ca2+]i increase. The response to ATP itself was inhibited by suramin (300 µM), PPADS, MRS2179, as well as by a cyclopiazonic acid-induced depletion of intracellular Ca2+ stores. Various blockers of voltage-sensitive Ca2+ channels and the gap junction inhibitor carbenoxolone did not alter the effect of ATP, whereas a combination of the ionotropic glutamate receptor antagonists D-AP5 and CNQX decreased it. Specific cross-desensitization experiments between ADPbS or UTP and ATP indicated the co-existence of P2Y1 and P2Y2 receptors. Prolonged superfusion with high BzATP or ATP (300 µM each for 60 min) did not induce pore formation or marked cell death. During current-clamp recordings, neurons responded to depolarizing current injection with firing patterns comparable to those already described in neocortical fast-spiking- and regularly-spiking non-pyramidal interneurons, respectively. Since, whole-cell patch-clamp recordings excluded the presence of functional P2X receptors in these cells, we suggest that in cultured interneuron-like cells the ATP-induced increase in [Ca2+]i is mediated chiefly by P2Y1,2 receptors. However, additional mechanisms such as interactions with surrounding astroglial cells, the release of glutamate or the entry of Ca2+ via storeoperated Ca2+ channels must also be considered. 1. Rudolf-Boehm-Institute of Pharmacology and Toxicology, University of Leipzig, Germany

58 Expression level dependence of ligand potencies is a common feature of P2X receptors Hausmann R.H. (1), Schmalzing G. (1) P2X receptors are ATP-gated cation channels consisting of three identical or homologous membrane-bound subunits that form the cation-conducting pore along the central axis. Distinct pharmacological phenotypes originate from multiple combinations of the repertoire of seven subunit isoforms belonging to the P2X family. For the nondesensitizing homotrimeric P2X2 receptor, ATP potency has been observed to increase with the expression level (Fujiwara & Kubo. J Physiol 558,31-43, 2004) and may account for significant lab-to-lab differences of reported EC50 values. Here, we systematically examined whether receptor abundance also affects ATP potency at other P2X subtypes or the inhibitory potency of the non-selective P2X receptor antagonist suramin. To this end, P2X1, P2X2 and P2X3 receptors were expressed in various levels in X. laevis oocytes by varying the amount of injected cRNA. A 8-fold variation in the ATP-gated current amplitude was associated with a 52-fold variation in the abundance of rP2X2 receptors at the plasma membrane. ATP concentration response curves and suramin concentration inhibition curves were shifted ~2.4-fold to the left and ~3-fold to the right, respectively, by this 52–fold increase in the cell surface abundance. This indicates that not only agonist potency but also antagonist potency is rP2X2 receptor expression level-dependent. A positive correlation with P2X2 receptor abundance persisted for inhibition constants (Ki) calculated according to the Cheng-Prusoff equation. This indicates that the expression-dependent variation of ATP EC50 values alone does not fully account for the observed variation of suramin IC50 values. To enable steady state recordings of otherwise fast desensitizing P2X1 and P2X3 receptor currents, non-desensitizing receptors chimeras were used consisting of a short N terminal domain of the P2X2 subunit fused in frame to the complementary portion of the P2X1 or P2X3 subunit. For both chimeras, rP2X2-X1 and rP2X2-X3, a ~2-fold left shift of the ATP concentration response curve was observed at the lowest and highest achievable expression level. Altogether, receptor abundance appears to affect ATP and antagonist potency at P2X receptors and should therefore considered as an important source of lab-to-lab variation of EC50 and IC50 values. 1. Molecular Pharmacology, RWTH Aachen University, Aachen, Germany

59 Human pannexin 1: quarternary structure and interaction with the human P2X7 receptor Woltersdorf R. (1), Markwardt F. (2), Schmalzing G. (1) Pannexin 1 (Panx1) is a member of the pannexin family of hemichannels. Panx1 is abundantly expressed in neurons and shares some structural features with the gap junction forming proteins. In erythrocytes Panx1 acts as an ATP release channel. In immune cells such as macrophages, Panx1 has been reported to be required for caspase-1 activation and interleukin-1ß release after activation of P2X7 receptors (P2X7Rs) by ATP, consistent with the established role of P2X7Rs in inflammatory responses. Sustained ATP activation of the P2X7R has been further shown to result in the induction of a large cytolytic pore permeable to organic cations. Cytolytic pore formation has first been thought to be formed by the P2X7 receptor itself through progressive pore dilatation. Recently, the view has been advocated that cytolytic pore formation is a secondary phenomenon and that pannexin-1 represents the dye uptake pathway opened by P2X7R activation. Here, we purified recombinant human Panx1 (hPanx1) from X. laevis oocytes by Ni-NTA chromatography and analyzed its oligomeric state by blue native PAGE and, after chemical cross-linking with the cleavable crosslinker DTSSP, by SDS-PAGE. Both techniques showed plasma membrane-bound hPanx1 to exist as a defined oligomer with a mass of a hexamer or larger. To unravel the true quaternary structure of hPanx1, concatamers consisting of two to seven contiguous copies of wt hPanx1 were constructed and analyzed in the fully glycosylated and natively deglycosylated state by BN-PAGE. All the concatamers appeared as intact proteins of the expected mass at the plasma membrane of X. laevis oocytes. Lower order concatamers comprising up to three Panx1 monomers assembled to higher order oligomers smaller or larger than the oligomer assembled solely from monomeric wt hPanx1. Also the concatamer data did not allow to discriminate between a hexameric or larger oligomeric state. To investigate a possible physical interaction between hPanx1 and hP2X7R, we co-expressed non-StrepII-tagged hPanx1 with the StrepII-tagged P2X7R. Co-purification by Strep-Tactin chromatography revealed an interaction of immature forms of hPanx1 with the hP2X7R on whole protein level. Interactions between mature forms of both proteins as present at the plasma membrane could not be detected. We conclude from these data that hPanx1 exists as a hexamer or larger oligomer in the plasma membrane. A physically stable interaction between hPanx1 and hP2X7R is unlikely to exist. 1. Molecular Pharmacology, RWTH Aachen University, Aachen, Germany, 2. JuliusBernstein-Instiute for Physiology, Martin Luther University, Halle/Saale, Germany

60 Human MrgX1 and murine MrgC11 receptors show differences in agonistpromoted receptor desensitization Solinski H.J. (1), Damm E. (1), Gudermann T. (1), Breit A. (1) Recently, the Mas-related gene (Mrg) family of receptors has been shown to belong to the superfamily of G protein-coupled receptors (GPCR). In humans six isoforms belonging to the MrgX subfamily exist, whereas in mice more than 50 Mrg receptors subdivided in seven subfamilies (A-G) have been found. All Mrg receptors tested so far show similar signal transduction as they activate PLC and subsequently release calcium from intracellular stores. Interestingly, human MrgX and murine MrgC11 receptors have been shown to be exclusively expressed in somatosensory dorsal root ganglion (DRG) neurons. In line with this restricted expression profile Mrg receptors have been shown to modulate pain perception in rats thus, Mrg receptors might represent promising targets for analgesic therapies in the future. The vast majority of GPCR (e.g. opioid receptors) suffer decreased signaling after prolonged agonist stimulation (agonist-promoted desensitization and endocytosis) however a few members have been shown to be resistant against this negative feedback mechanism. Despite the importance of receptor desensitization for the understanding of GPCR signaling, up to now, little is known about

19

agonist-promoted regulation of the MrgX receptor family. Therefore, the goal of the present study was to analyze the regulation of the human MrgX1 receptor after prolonged agonist stimulation when expressed in HEK293 and DRG-like F11 cells. Further, we compared the human variant with its murine counterpart, the MrgC11 subtype. Surprisingly, whereas the human MrgX1 receptor is resistant to agonistpromoted endocytosis the MrgC11 receptor is prone to this receptor activity modulating process. In line with these species-specific differences, calcium transients induced by the MrgC11 but not by the MrgX1 receptor desensitized after prolonged agonist stimulation. Therefore, we classify the human MrgX1 receptor as one of the few members within the GPCR superfamily that does not undergo agonist-promoted endocytosis, and additionally, we describe species-specific differences in regulation of closely related receptor subtypes. To our knowledge antithetic behavior in receptor regulation depending on the species is unique within the GPCR superfamily. Furthermore, these new findings might help to understand differences in the nociception of humans and rodents and, thus, could facilitate the development of future analgesic drugs regulating MrgX1 activity. 1. Walther-Straub-Institute for Pharmacology and Toxicology, Ludwig-MaximiliansUniversity, Munich, Germany

61 Stimulation of PI3K/Akt kinase by delta-opioid receptors is mediated by transactivation of IGF-1 receptors in NG108-15 cells Eisinger D.A. (1), Heiss A. (1), Ammer H. (1) It is well established that δ-opioid receptor (DOR) activation results in stimulation of the Akt kinase pathway, a regulatory mechanism that protects the cell from apoptosis. The aim of the present study is to investigate the signal transduction mechanism by which opioids bring about Akt kinase activation in DOR carrying neuroblastoma x glioma hybrid (NG108-15) cells. Cells were stimulated either with the δ-opioid receptor selective agonist [D-Pen2,5]enkephalin (DPDPE) or the non-selective opiate etorphine for 5 min before reactions were stopped and Akt kinase autophosphorylation was determined by Western blot technique after solubilisation of the cells. Short-term exposure of the cells to maximum effective concentrations of both agonists strongly stimulated Akt kinase phosphorylation in a naloxone sensitive manner. Opioid-induced stimulation of Akt kinase activity is mediated by inhibitory G proteins and PI3K, because pre-treatment of the cells with pertussis toxin (PTX) and the PI3K inhibitors wortmannin and LY294002 prevented this effect. Activation of PI3K may either be facilitated by Gβγ subunits released from activated G proteins or by transactivation of receptor tyrosine kinases (RTK). In order to discern between both mechanisms, NG108-15 cells were first transiently transfected with the Gβγ scavenger phosducin. However, disruption of Gβγ signaling by phosducin failed to impair DOR-stimulated Akt kinase phosphorylation. NG108-15 cells contain two potential RTK species which might be involved in DOR coupling to the PI3K/Akt kinase pathway, i.e. the insulin like growth factor (IGF-1) receptor and the platelet-derived growth factor (PDGF) receptor. Whereas pre-treatment of the cells with the PDGF receptor inhibitor AG1296 was ineffective, the IGF-1 receptor blocker AG1024 largely attenuated both DPDPE- and etorphine-induced Akt kinase phosphorylation. These results demonstrate that the signal transduction pathway by which DORs stimulate Akt kinase in NG108-15 hybrid cells is composed of Gi/o proteindependent, IGF-1 receptor-stimulated activation of PI3K. 1. Institute of Pharmacology, Toxicology and Pharmacy, LMU Muenchen, Muenchen, Germany

62 AUF1 regulates human iNOS expression by modulation of mRNA stability Pautz A. (1), Schmidt N. (1), Art J. (1), Rauschkolb P. (1), Linker K. (1), Altenhöfer S. (1), Heil S. (1), Kleinert H. (1) Human iNOS-expression is controlled by transcriptional and post-transcriptional mechanisms. Post-transcriptional modulation of gene expression is mediated mainly by regulation of mRNA stability. AU-rich elements (ARE) are critical cis-acting elements in the 3’-untranslated region (3’-UTR) of many inherently unstable mRNAs and they are targets for trans-acting proteins regulating mRNA stability. The human iNOS 3’-UTR, which is able to destabilize heterologous mRNAs, contains five AU-rich elements. The RNA binding protein AUF1 (AU binding factor 1) is an important mediator of rapid AREmediated mRNA decay. AUF1 is a family of related ARE-binding mRNA-destabilizing factors, consisting of four isoforms (p37-, p40-, p42- and p45-AUF1) generated by alternative splicing of one encoding mRNA. In recent years, several studies have provided conflicting results suggesting that certain AUF1 isoforms might also be stabilizers of ARE-mRNAs. In this study, we analyzed in which way the four AUF1 isoforms regulate human iNOS expression. In DLD-1 cells overexpressing each isoform as a fusion protein with the enhanced green fluorescent protein (EGFP) cytokineinduced iNOS expression was markedly reduced. Accordingly, siRNA-mediated downregulation of the expression of the different AUF1 isoforms resulted in enhanced cytokine-induced iNOS expression in all cases. Inhibition of RNA Polymerase IIdependent transcription clearly demonstrated that all AUF1 isoforms decrease iNOS mRNA half-life to the same extent. UV-cross linking experiments showed that the different isoforms interacted with the same AREs in the human iNOS mRNA 3’-UTR. Upon cytokine stimulation the intracellular binding of AUF1 to the human iNOS mRNA was reduced. This may explain the stabilization of iNOS mRNA under these conditions. In conclusion, our data indicate that AUF1 decreases iNOS expression by destabilization of the iNOS mRNA. Thereby, all four isoforms seem to act in a similar way. 1. Dept. of Pharmacology, Johannes Gutenberg University, Mainz, Germany

63 Cytokines and NO upregulate prolyl hydroxylase 3 (PHD-3) mRNA and protein expression in rat renal mesangial cells Boosen M. (1), Pleskova M. (2), Moreth K. (1), Schäfer L. (1), Pfeilschifter J. (1), Beck K.-F. (1) The hypoxia-inducible transcription factors HIF-1 and HIF-2 are expressed constitutively but, under normoxic conditions, HIFs become hydroxylated by HIF prolyl hydroxylases (PHDs) followed by a rapid degradation by the proteasome. Recent reports also suggest a role for HIF in normoxic signalling as induced by inflammatory cytokines, NO and growth factors indicating a high impact of HIF in tumour growth and inflammatory processes. Rat renal mesangial cells are regarded as the main players in inflammatory glomerular diseases. To investigate a role of hypoxia signalling in glomerular inflammation, we stimulated primary rat mesangial cells with the cytokine interleukin 1b (IL-1β) and analysed the mRNA expression of the PHDs and factor inhibiting HIF 1 (FIH1) by RT-PCR and/or Northern blotting. IL-1β markedly induced PHD-3 mRNA expression but did not affect PHD-1, PHD-2 and FIH-2 mRNA levels. The time and dose dependent effect of IL-1β on PHD-3 mRNA was partly reduced by coadministration of the NO-synthase inhibitor L-NMMA and enhanced by the NO donor DETA-NO indicating that endogenous NO levels contributed at least in part to cytokine-driven PHD-3 expression. Analysis of PHD-3 protein by immunoblotting revealed that the cytokineeffect on PHD-3 mRNA expression is also transduced to PHD-3 protein levels. Remarkably, blockage of the NFκB signalling pathway by an IkB kinase inhibitor or by silencing the expression of the NFκB subunit p65 completely blocked IL-1β-evoked PHD-3 expression indicating that cytokine-induced PHD-3 expression is mediated via the NFκB signalling pathway. In a rat model for mesangio-proliferative glomerulonephritis we could demonstrate that PHD-3 expression is also enhanced in the course of glomerular disease. Our data clearly demonstrate a cytokine/NO-induced upregulation of PHD-3 but not of other hydroxylases in vitro and in vivo. 1. Dept. of Pharmacology and Toxicology, Johan Wolfgang Goethe-University, Frankfurt am Main, Germany, 2. Laboratory for Materials-Biology Interactions, EMPA, St. Gallen, Switzerland

64 Interactive role of Rho-kinase and constitutive NO synthase in airway responsiveness to methacholine Maarsingh H. (1), Schaafsma D. (1,2), van Schouwenburg M.R. (1), Meurs H. (1) The Rho/Rho-kinase signalling pathway causes calcium sensitization in airway smooth muscle (ASM) and increased Rho-kinase activity contributes to the development of airway hyperresponsiveness in allergic asthma. In addition to calcium sensitization, Rhokinase could regulate airway responsiveness by attenuating the production of bronchodilating nitric oxide (NO) in the airway epithelium, since in the vasculature it has been shown that Rho-kinase may inhibit endothelial NO synthase (eNOS) activity via inhibition of PI3-kinase and Akt. Conversely, eNOS-derived NO has been shown to inhibit the Rho/Rho-kinase pathway. In the present study, we investigated the interactive role of Rho-kinase and epithelial constitutive NOS (cNOS) isoforms in regulating airway responsiveness to methacholine using perfused guinea pig tracheal preparations. In line with previous findings, intraluminal (IL) application of the NOS inhibitor L-NAME (0.1 mM) increased maximal responsiveness (Emax) to methacholine by 1.8-fold, clearly demonstrating the bronchoprotective role of cNOS-derived NO. Incubation with the selective Rho-kinase inhibitor Y-27632 (1 µM), either intra- or extraluminally applied, reduced the Emax as well as the sensitivity (pEC50) towards methacholine. The inhibitory effect on Emax by IL Y-27632 was dose-dependently reversed by L-NAME. Moreover, in the presence of Y-27632 a higher dose of L-NAME (1 mM) was needed to obtain the 1.8-fold increase in Emax, supporting that Y-27632 decreases airway responsiveness by increasing cNOS-derived NO production. Incubation with Y-27632 did not affect airway relaxations elicited by the NO donor SNP in precontracted guinea pig tracheal ring preparations, indicating that NO-mediated airway smooth muscle relaxation by itself is independent of Rho-kinase. In conclusion, in addition to increasing the calcium sensitivity of ASM towards contractile agonists, Rho-kinase mediates airway responsiveness by attenuating the production of cNOS-derived NO in the airway epithelium. 1. Dept. of Molecular Pharmacology, University of Groningen, Groningen, The Netherlands, 2. Dept. of Physiology, University of Manitoba, Winnipeg, Manitoba, Canada

65 Protection against oxidation-induced degradation of soluble guanylate cyclase Hoffmann L.S. (1,2), Schmidt P.M. (3), Keim Y. (1), Schaefer S. (1), Schmidt H.H.H.W. (4), Stasch J.P. (1,2) Soluble guanylate cyclase (sGC) is an α/β haem protein catalyzing the conversion of GTP to the second messenger cGMP. Basal enzyme activity is several-fold increased upon binding of nitric oxide (NO) to a reduced Fe2+ haem group proximally ligated to H105 of sGC β subunit. In endothelial dysfunction and vascular disease, NO signalling is impaired due to the oxidation and subsequent loss of the sGC haem. In addition to its ability to activate the haem-free sGC the haem-independent sGC activator, cinaciguat (BAY 58-2667), also protects sGC from ubiquitin-triggered degradation. Here we investigate whether this protection is a unique feature of cinaciguat or a general characteristic of haem-site ligands such as the structurally unrelated sGC activator ataciguat (HMR 1766), the haem-mimetic Zn-protoporphyrin IX (Zn-PPIX), or extends even to the haem-dependent sGC stimulator BAY 41-2272. The sGC inhibitor 1H(1,2,4)-oxadiazole[4,3-a]quinoxalin-1-one (ODQ) was used to induce oxidation-induced degradation of sGC. Activity and protein levels of sGC were measured in a Chinese hamster ovary (CHO) cell line as well as in primary porcine endothelial cells expressing wildtype (WT) sGC. Cells expressing mutant sGC were used to shed light onto the molecular mechanism underlying the observed effects. Oxidation-induced sGC degradation was prevented by cinaciguat and Zn-PPIX in both cell types resulting in increased sGC protein levels. In contrast, the sGC activator, ataciguat, and the sGC stimulator, BAY 41-2272, were devoid of any protection. Similarly, the haem-free b1H105F mutant mimicked haem oxidation and was stabilized by cinaciguat and ZnPPIX. Thus, distinct abilities to mimic the spatial structure of the haem group and the

20

resulting occupation of the enzyme’s haem pocket differentiate whether a sGC activator like cinaciguat also stabilizes sGC upon haem oxidation-induced degradation and represents a unique example of bimodal target interaction. 1. Pharma Research Center, Bayer HealthCare, Aprather Weg 18a, Wuppertal, Germany, 2. Martin-Luther-University, School of Pharmacy, Wolfgang-Langenbeck-Str. 4, Halle, Germany, 3. CSIRO Molecular Health Technologies, 343 Royal Parade, Parkville, Victoria 3052, Australia, 4. Department of Pharmacology & Centre for Vascular Health, Monash University, Melbourne, Clayton, Victoria 3800, Australia

66 The nitric oxide receptor guanylyl cyclase in gastrointestinal smooth muscle is dispensable for gastrointestinal motility and survival in mice Groneberg D. (1), König P. (2), Koesling D. (1), Friebe A. (1) The enteric nervous system controls the motor function of the gastrointestinal tract, containing adrenergic, cholinergic and non-adrenergic non-cholinergic (NANC) neurons. NANC neurons express neuronal NO synthases which produce NO. NO has been demonstrated to be the most important physiological mediator of NANC relaxation of gastrointestinal smooth muscle, besides vasoactive intestinal peptide (VIP) and ATP. The main target of the messenger molecule NO is NO-sensitive guanylyl cyclase (NOGC). Activation of NO-GC leads to production of guanosine 3´,5´-cyclic monophosphate (cGMP), which exerts its effects via cGMP-dependent kinases, channels, or phosphodiesterases. This study aimed to identify the consequences of general and smooth muscle-specific deletion of NO-GC for gastrointestinal motility. We used isometric force studies with two distinct NO sources, endogenous NO production induced by electric field stimulation and pharmacological NO release using NO donors. Total gut transit time was measured to study the consequences of NO-GC deletion in vivo. General lack of NO-GC abolishes NO-dependent relaxation of gastrointestinal smooth muscle. Neither physiological concentrations of NO donors nor electric field stimulation at low frequencies led to relaxation of gastric fundus strips or sphincters. The whole gut transit time was increased approximately three times compared with WT. This dysfunction leads to early mortality starting around day 18. Mortality was reduced by omission of dietary fiber. Surprisingly, in smooth muscle-specific knockout mice (SMKO) the NO-dependent relaxation is reduced, but still intact. Total gut transit time shows no difference and survival of SMKO animals was similar to that of WT mice. In conclusion the NO receptor guanylyl cyclase in gastrointestinal smooth muscle is dispensable for gastrointestinal motility and survival. 1. Dept. of Pharmacology, Ruhr University Bochum, Germany, 2. Dept. of Anatomy, University Lübeck, Germany

67 Renovascular hypertension: negative feedback regulation within NO/cGMP pathway attenuates vasodilatory response Mergia E. (1), Stegbauer J. (2), Friedrich S. (2), Quack I. (2), Rump L.C. (2), Koesling D. (1) Hypertension is the leading risk factor for cardiovascular mortality and has been associated with alterations in endothelium- and smooth muscle-dependent vascular relaxation. One of the major pathways that mediate vascular relaxation is the NO/cGMP signaling cascade, in which the NO receptor guanylyl cyclase (NO-GC) holds a key position by translating the NO signal into cGMP formation. There are two isoforms of the NO receptor GC. Both isoforms consist of an identical b subunit which is dimerized to either one of two a subunits (α1 or α2) and will be referred to as NO-GC1 (so far α1β1 heterodimer) and NO-GC2 (so far α2β1 heterodimer), respectively. Knockout (KO) mice deficient in either one of the NO-GCs, NO-GC1 or NO-GC2, revealed that both NO-GCs are capable to mediate vascular relaxation. The NO-GC1 appears to be the major isoform, particularly in the aorta, where NO-GC1 represents approximately 90% of total NO-GC content. Deletion of the NO-GC1 resulted in reduced endothelium-dependent relaxation and reduced vasodilatory response to exogenous NO, which is mediated by the NO receptor GC2 in NO-GC1-deficient mice. However, despite the reduced vascular relaxation NO-GC1 KO mice are normotensive. In the present study, we used the Goldblatt model of renovascular hypertension (2 kidney 1 clip [2K1C] operation) to induce hypertension in the NO-GC1 KO mice and investigated the impact of renovascular hypertension in aorta and renal vascular system of 2K1C-operated NOGC1-deficient mice. 1. Dept. of Pharmacology, Ruhr University, Bochum, Germany, 2. Dept. of Nephrology, Heinrich Heine University, Duesseldorf, Germany

68 Cardiovascular and gastrointestinal function of IRAG Desch M. (1), Schinner E. (1), Sigl K. (2), Wegener J. (2), Feil R. (3), Feil S. (3), Hofmann F. (2), Schlossmann J. (1) Signalling by NO / cGMP is important for regulation of vascular tone, gastrointestinal motility and for the activation state of platelets. One important signalling pathway of cGMP-dependent protein kinase type I (cGKI) is mediated by IRAG (IP3 Receptor Associated cGKI substrate) which is highly expressed in smooth muscle tissues and in platelets. In order to elucidate the physiological role of IRAG we generated IRAG knockout mice by targeted deletion of exon 3. These mice show an enlarged gastrointestinal tract including pylorus stenosis. Exogenous and endogenous NO and cGMP stimulation relaxes hormone induced contraction of wild type aortic vessels as well as gastrointestinal tissues of colon and jejunum. This effect is significantly reduced in IRAG knockout animals. The expression levels of other cGKI substrates and the NOinduced cGMP synthesis are not affected in these tissues from IRAG KO mice. The NO / cGMP-inhibitory effect of platelet aggregation induced by various agonists is clearly diminished in IRAG deficient mice. Furthermore the inhibition of ATP secretion from platelet granula by NO / cGMP is reduced in IRAG KO mice. These results indicate that IRAG signalling is essentiaI for smooth muscle relaxation, inhibition of platelet aggregation and ATP secretion by NO / cGMP.

1. Institut für Pharmakologie & Toxikologie, Univeristät Regensburg, Germany, 2. Institut für Pharmakologie & Toxikologie, TU München, Germany, 3. Interfakultäres Institut für Biochemie, Universität Tübingen, Germany

69 Betulinic acid protects against cerebral ischemia/reperfusion injury in mice Lu Q. (1), Xia N. (1), Förstermann U. (1), Li H. (1) Increased production of reactive oxygen species and reactive nitrogen species following cerebral ischemia/reperfusion is a major cause for neuronal injury. In atherosclerotic apolipoprotein E knockout (ApoE-KO) mice, 2 h of middle cerebral artery occlusion (MCAO) followed by 22 h of reperfusion led to an enhanced expression of NADPH oxidase subunits (NOX2, NOX4 and p22phox) and isoforms of nitric oxide synthase (neuronal nNOS and inducible iNOS) in the ischemic hemisphere compared with the non-ischemic collateral hemisphere. This was associated with elevated levels of 3nitrotyrosine, an indicator of peroxynitrite-mediated oxidative protein modification. Pretreatment with betulinic acid (50 mg/kg/day for 7 days via gavage) prior MCAO prevented the ischemia/reperfusion-induced upregulation of NOX2, nNOS and iNOS. In parallel, betulinic acid reduced the levels of 3-nitrotyrosine. In addition, treatment with betulinic acid enhanced the expression of endothelial eNOS, both in the ischemic and non-ischemic hemispheres. Finally, betulinic acid reduced infarct volume and ameliorated the neurological deficit in this mouse stroke model. In conclusion, betulinic acid protects against cerebral ischemia/reperfusion injury in mice. The reduced oxidative stress (by downregulation of NOX2) and nitrosative stress (by reduction of nNOS and iNOS), and enhanced blood flow (by upregulation of eNOS) may play an important role. 1. Dept. of Pharmacology, Johannes Gutenberg University, Mainz, Germany

70 Islet cGKI is involved in glucose homeostasis Leiss V. (1), Friebe A. (2), Hofmann F. (1), Lukowski R. (1,3) The islets of Langerhans act as a glucose sensor, adjusting hormone production/output to the prevailing blood glucose levels. The two main hormones released are the functional antagonists insulin and glucagon. B-cells are activated to secrete insulin at hyperglycaemia, whereas glucagon is released from A-cells during hypoglycaemia. Nitric oxide (NO) has an important role for the incidence and development of type-1 diabetes [1]. The molecular pathway whereby NO affects islet function in health or disease has not been identified. We hypothesized that cGMP-dependent kinase I (cGKI) is an effector of NO in islets and might be involved in the endocrine control of blood glucose levels. The expression pattern of islet cGKI was determined in A- and B-cell specific cGKI knockout (A-KO and B-KO) mice, conventional cGKI-KOs and in cGKIα rescue animals expressing the Iα isozyme only in smooth muscle cells [2]. The combined analysis of the different gene-targeted mouse lines clearly demonstrated that endogenous cGKI localizes to glucagon-producing A-cells, hence, B-KOs showed no changes of islet cGKI levels, whereas the cGKI protein was either absent or diminished in conventional cGKI-KOs, Iα-rescue mice or the A-KOs. In addition, the beta subunit of the soluble guanylyl cyclase and the cGMP-degrading phosphodiesterase-5 were identified in the same cell type indicating that the pathway might have a definite role in islets. A functional analysis of fasted animals revealed that the basal blood glucose levels and serum glucagon levels were significantly increased in cGKIα rescue animals compared to age- and littermate controls. These in vivo findings were supported by hormone release experiments of isolated islets as a glucose-induced suppression of glucagon at low (physiological) concentrations (6 mM) was abolished in islets that lacked cGKI.In conclusion, our genetic approach identifies an important new role of cGMP-cGKI signaling for the function of A-cells. The loss of islet cGKI activity leads to an abnormal glucagon secretion causing a hyperglycaemic fasting phenotype indicative for a pre-diabetic syndrome. [1] Kröncke at al. Biochem Biophys Res Commun 175: 752758 (1991).[2] Weber at al. Circ Res 101: 1096-1103 (2007). 1. FOR 923 at Institut für Pharmakologie und Toxikologie, TU München, Germany, 2. Institut für Pharmakologie und Toxikologie, Ruhr-Universität Bochum, Germany, 3. Institut für Pharmakologie und Toxikologie, TU München, Germany

71 Nebivolol induced NO release in erythrocytes Reidenbach C. (1), Rieckeheer E. (1), Schwinger R.H.G. (2), Gäbler R. (3), Kleinbongard P. (4), Özuyaman B. (5), Kelm M. (5), Bloch W. (1), Brixius K. (1) Nebivolol is a ß1-adrenoceptor selective blocker with additional vasodilatory properties, which are mediated by the stimulation of ß3-adrenoceptors. Only previously, it has been described that the endotheial nitric oxide synthase (eNOS) is also expressed in erythrocytes and can be stimulated by various interventions (e.g. application of insulin). In addition, it has been described that the ß3-adrenoceptor is also present in erythrocytes. Therefore, we investigated whether nebivolol (10 µM) may be able to activate the erythrocyte eNOS immunohistochemical detection of eNOS phosphorylation at serine 1177) and whether this phosphorylation may go along with an increased release of nitric oxide (Faraday-rotation-spektroskopy), which may affect erythrocyte aggregation (Laser-assisted Optical Rotational Cell Analyzer (LORCA)). For comparison, we investigated the ß1-adrenoceptor selective blocker metoprolol as well as the ß3-adrenoceptor agonist BRL 37344 (10 µM) each in comparison to control (DMSO). Only nebivolol and BRL were able to induce a significant and time-dependent release of nitric oxide via phophorylation of eNOS at serine1177. Nebivolol and metoprolol did not alter the maximal aggregation capacity of the erythrocytes. Metoprolol but not nebivolol seems to decrease erythrocyte deformability. In conclusion, nebivolol increases the activation of the erythrocyte eNOS system. This additional eNOS activation provides a additional NO-pool, which may be of importance or the treatment of vascular disease. 1. Abt. für Molekulare und Zelluläre Sportmedizin, Deutsche Sporthochschule Köln, Köln,Deutschland, 2. Klinik II, Klinikum Weiden, Weiden i.d. OPf., Deutschland, 3. Invivo GmbH - Institute for trace gas technology, St. Augustin, Deutschland, 4. Institut für Pathophysiologie, Universitätsklinikum Essen, Essen, 5. Medizinische Klinik I, RWTH Aachen, Aachen

21

72

75

S-nitrosylation of soluble guanylyl cyclase is increased by endothelial nitric oxide synthase overexpression in mice Oppermann M. (1), Suvorava T. (1), Kojda G. (1) Vascular soluble guanylyl cyclase (sGC) activity has been shown to be regulated in-vitro by nitric oxide (NO) via a) downregulation of the expression and b) a post-translational S-nitrosylation mechanism. However, we were previously able to show that sGC-activity was not affected in-vivo in nitrate-treated animals. In this study, we investigated the expression, activity, and S-nitrosylation of sGC in mice with an endothelial-specific overexpression of eNOS. eNOS++-mice with a 2.8-fold higher aortic eNOS-expression and a decreased systolic blood pressure were generated using the endothelium-specific Tie-2 promotor, transgene negative littermates served as controls (eNOSn). Westernblots for both α1- and β1-subunits of sGC were performed in mouse lungs using polyclonal rabbit antibodies and showed similar protein levels of sGC-α1 (103.4±10.3%) and sGC-β1 (109.7±28.6%, P>0.05 each) in eNOS++ vs eNOSn. Phosphorylation of vasodilator-stimulated phosphoprotein (VASP) was increased by approximately 2-fold (P<0.05) as determined by westernblot with a P-ser239-VASP monoclonal antibody. Conversion of [α-32P]-GTP to 32P-cGMP was used to measure sGC activity stimulated with S-nitroso-N-acetyl-penicillamin (SNAP, 1 mmol/L). The activities of isolated sGC were decreased in lungs of eNOS++ (800±250 vs. 1562±250 pmol/mg/min, n=5, P<0.05). S-Nitrosylation of β1-sGC was semiquantitatively evaluated by labeling S-nitrosylated residues with biotin and performing westernblot after separation on neutravidin-agarose (biotin switch assay). There was an increase of nitrosylated/total sGC-β1 subunit of 158±15.4% in eNOS++ (100% in eNOSn, n=4, P<0.05). Aortic vasorelaxation by acetylcholine and SNAP was analyzed in organ bath experiments. Concentrationdependent vasodilation to endogenous and exogenous NO were similar. While expression of sGC does not seem to be affected by an increased bioavailability of nitric oxide, activity of isolated sGC is decreased by 2-fold. The increased ratio of nitrosylated sGC vs. total sGC suggests an endogenous regulation mechanism as was described invitro. However, the signaling function of the pathway is unaffected as indicated in vasorelaxation experiments and by increased VASP-phosphorylation in eNOS++. 1. Institute for Pharmacology and Clinical Pharmacology, Heinrich Heine University, Duesseldorf, Germany

One way to escape mammalian host defence: the Pseudomonas aeruginosa quorum-sensing signal 3OC12 inactivates human paraoxonase-2 Horke S. (1), Altenhöfer S. (1), Witte I. (1), Wilgenbus P. (1), Förstermann U. (1), Teiber J.F. (2) Pseudomonas aeruginosa is a common nosocomial pathogen of immunocompromised patients, i.e. those with cystic fibrosis, chronic obstructive pulmonary disease, pneumonia etc. P. aeruginosa produces several virulence factors that are induced by quorum-sensing, a known bacterial communication system sensing bacterial density and regulating gene expression. Among those virulence factors are pyocyanin and the lactone 3OC12 (N-(3-oxododecanoyl)-L-homoserine lactone). Pyocyanin is a redoxactive compound that decreases mammalian glutathione and catalase levels, inactivates the antimicrobial defense system, and initiates oxidative stress (i.e. superoxide) in the infected host. The lactone 3OC12 is a key component of the bacterial quorum-sensing suspected to cause Ca2+-influx in host cells. We recently showed that the human enzyme paraoxonase-2 (PON2), which has enzymatic lactonase activity, plays a dominant role in inactivating the lactone 3OC12 by hydrolysis; however, PON2 expression negatively correlates with intracellular calcium levels, prompting us to determine host-pathogen interactions of 3OC12, pyocyanin and PON2. Here we show that 3OC12 indeed caused a significant cytosolic Ca2+-influx, e.g. in lung epithelial A549 cells, within 0.25-2 minutes. Interestingly, inactivation of PON2 lactonase activity occurred in a similar time frame (~90% activity lost in 10min). Further, 3OC12 instigated degradation of both PON2 mRNA and protein. Because inactivation of PON2 enzymatic activity was observed within minutes, whereas decay of PON2 mRNA and protein occurred within hours, we propose that 3OC12 causes rapid PON2 inactivation by a post-translational mechanism. As a result, we show that 3OC12 also ablates PON2 antioxidative activity, since the Pyocyanin-induced superoxide formation, which otherwise is alleviated by PON2 overexpression, is enhanced after 3OC12 co-administration. Similarly, 3OC12 disables the anti-apoptotic effect of PON2, rendering cells susceptible to 3OC12-induced apoptosis. Collectively, our study shows that 3OC12 potentiates the impact of pyocyanin by disabling PON2, thus representing a cooperative and effective mechanism of P. aeruginosa virulence factors in weakening mammalian host defense. 1. Dept. of Pharmacology, Johannes Gutenberg University, Mainz, Germany, 2. Dept. of Internal Medicine, University of Texas Southwestern Medical Center, Dallas, USA

73 Betulinic acid stimulates eNOS activity by activating PI3K pathway Hohmann N. (1), Xia N. (1), Förstermann U. (1), Li H. (1) Betulinic acid is a naturally occurring pentacyclic triterpenoid which has antiinflammatory and anticancer properties. We have recently demonstrated that betulinic acid (10 µM, 18 h) upregulates the expression of endothelial nitric oxide synthase (eNOS) and downregulates the expression of NADPH oxidase subunits NOX4 and p22phox in human endothelial cells. As a result, endothelial NO production is enhanced and superoxide generation reduced. We report in this study that betulinic acid also possesses an acute effect on eNOS activity. A short-term (5-60 min) incubation of human EA.hy 926 endothelial cells with betulinic acid (1-100 µM) resulted in phosphorylation of eNOS at serine 1177 and dephosphorylation of eNOS at threonine 495 residues in a concentration- and time-dependent manner. This was associated with enhanced production of bioactive NO (RFL-6 reporter cell assay). Pretreatment with inhibitors of phosphatidylinositol 3-OH kinase (PI3K), wortmannin or LY 294002, reduced the betulinic acid-induced eNOS phosphorylation at serine 1177. These results indicate that PI3K pathway is involved in betulinic acid-induced eNOS activation. 1. Dept. of Pharmacology, Johannes Gutenberg University, Mainz, Germany

74 12/15-Lipoxygenases play a key role in AIF-dependent cell death after glutathione depletion in HT-22 cells Tobaben S. (1), Hoehn M. (2), Dolga A. (1), Plesnila N. (3), Culmsee C. (1) Oxidative stress plays a key role in neuronal degeneration and death in Alzheimer’s disease and Parkinson’s disease. Here, we investigated the role of lipoxygenases in ROS formation and cell death mechanisms in a neuronal cell line (HT-22). These immortalized hippocampal neurons were exposed to glutamate (2-5 mM) which blocks the xCT-mediated cysteine import followed by a decline of intracellular glutathione levels and cytotoxic ROS formation. In addition, the cells were exposed to 4-Hydroxynonenal (HNE, 2-10 µM) an established second messenger of free radicals that plays an important role in neuropathological processes. After exposure to glutamate we detected a nearly two-fold increase of ROS formation within 8 h measured by the fluorescent dyes DCF (Dichlorodihydrofluoresceine diacetate) or Bodipy (4,4-difluoro-5-(4-phenyl1,3-butadienyl)-4-bora-3a,4a-diaza-s-indacene-3-undecanoic acid). This first increase in ROS formation was followed by a boost of free radical accumulation at 18 h. Glutamateinduced ROS formation and cell death were significantly reduced by the LOX inhibitors PD146176 and AA861. Lipoxygenase-dependent cell death further involved Bid translocation to the mitochondria, loss of mitochondrial membrane potential and release of mitochondrial proteins like AIF. Bid inhibition (Bid-inhibitor BI-6C9, Bid-siRNA) and AIF-siRNA also prevented HT-22 neurons from glutamate-induced cell death whereas caspase inhibitors showed no effect. It is important to note that the cytotoxic effects of tBid were neither prevented by LOX-inhibition nor by the vitamin E analog Trolox suggesting that lipid peroxidation occurred upstream of Bid activation and the related mitochondrial death signaling pathways. Preliminary results indicate that inhibition of LOX or Bid also reduce HNE-induced cell death. In conclusion, our data suggest that glutathione depletion leads to an increase in 12/15-LOX activity and enhanced formation of lipid peroxides, followed by activation of Bid, mitochondrial AIF release and cell death. Therefore, inhibition of 12/15-LOX is proposed as a promising therapeutic approach to prevent neuronal cell death associated with oxidative stress in acute and chronic neurological diseases. 1. Klinische Pharmazie, Institut fuer Pharmakologie und Toxikologie, PhilippsUniversitaet Marburg, Germany, 2. Pharmazeutische Biologie-Biotechnologie, LudwigMaximilians-Universitaet, Muenchen, Germany, 3. Dept. of Neurodegeneration & Dept. of Physiology, Royal College of Surgeons in Ireland (RCSI), Dublin, Ireland

76 When the protector becomes the traitor: PON2 weakens pro-apoptotic stimuli by its anti-oxidative mechanism Witte I. (1), Altenhöfer S. (1), Wilgenbus P. (1), Förstermann U. (1), Horke S. (1) Paraoxonase-2 (PON2) is a ubiquitously expressed protein at the nucleus, mitochondria and the endoplasmic reticulum (ER), which is found upregulated in different tumour tissues. Though known as a lactonase, PON2 is able to decrease cellular oxidative stress. Moreover, we recently demonstrated anti-apoptotic effects of PON2, as its overexpression attenuated ER stress-induced apoptosis. Here we show that PON2 did not alter activation of ER stress mediators ATF6, IRE1 and PERK, but decreased induction of the ER stress triggered pro-apoptotic CHOP protein. This effect was independent from its enzymatic activity, but could be mimicked by antioxidant treatment, suggesting a major role for the anti-oxidative function of PON2. However, because PON2 also protected cells from other, ER stress unrelated apoptotic stimuli, alteration of CHOP induction by PON2 insufficiently explained its anti-apoptotic effect. Therefore, we reasoned that its key function resides on the mitochondria. Indeed, while PON2 did not protect against extrinsic stimuli such as TRAIL or TNFa-induced apoptosis, it readily reduced intrinsic pro-apoptotic signalling (e.g. by ActinomycinD, Staurosporine or Doxorubicin). This was evident by decreased activation of caspases 8, 9, and 3, and attenuated cell death by PON2 overexpression. Addressing the underlying mechanism, we found that PON2 diminished mitochondrial cytochrome C release in a caspaseindependent fashion, strongly arguing for the specific mitochondrial function of PON2. While independent from its enzymatic activity, decrease of apoptosis by PON2 was found to rely on its anti-oxidative function. As a concept, we propose that reduction of oxidative stress, e.g. by upregulating PON2, represents an efficient strategy of tumours to evade intrinsic apoptotic stimuli. 1. Dept. of Pharmacology, Johannes Gutenberg University, Mainz, Germany

77 SIRT1 as a regulator of cardiovascular oxidative stress Xia N. (1), Daiber A. (2), Habermeier A. (1), Spanier G. (1), Lu Q. (1), Oelze M. (2), Closs E.I. (1), Münzel T. (2), Förstermann U. (1), Li H. (1) SIRT1 is a NAD+-dependent class III histone deacetylase. Recent studies point to SIRT1 as a key regulator of angiogenesis, vascular tone and endothelial function. The aim of the present study was to investigate the role of SIRT1 in regulating oxidative stress in the cardiovascular system. Atherosclerotic apolipoprotein E knockout (ApoEKO) mice were treated with a SIRT1 activator, resveratrol (30 or 100 mg/kg/day for 7 days via gavage). Resveratrol treatment significantly reduced the levels of superoxide in the heart, which was associated with an upregulation of superoxide dismutase isoforms (SOD1-3), glutathione peroxidase 1 (GPx1), and catalase. In parallel, mRNA expression of NADPH oxidase subunits NOX2 and NOX4 was reduced. Resveratrol also inhibited the translocation of p47phox and Rac1 and decreased the activity of NADPH oxidase complex in heart membrane fraction. The cofactor of endothelial nitric oxide synthase (eNOS), (6R)-5,6,7,8-tetrahydro-L-biopterin (BH4), is synthesized from guanosine 5'triphosphate (GTP) via a de novo pathway by the rate-limiting enzyme GTP cyclohydrolase I (GCH1), and essential for eNOS functionality. ApoE-KO mice are associated with oxidative stress and eNOS-mediated superoxide production (i.e. eNOS uncoupling). Treatment of ApoE-KO mice with resveratrol enhanced the expression of GCH1 and increased the cardiac levels of BH4. This was associated with reduced superoxide production from eNOS. Also in human endothelial EA.hy 926 cells, resveratrol (100 µM) upregulated SOD1-3, GPx1 and catalase, and downregulated NOX4. SIRT1 inhibition with sirtinol or SIRT1 knockdown by siRNA blocked part of the effects of resveratrol. These data indicate that SIRT1 is a key regulator of cardiovascular oxidative stress.

22

1. Dept. of Pharmacology, Johannes Gutenberg University, Mainz, Germany, 2. II. Medizinische Klinik, Johannes Gutenberg University, Mainz, Germany

1. Department of Pharmacology, University of Cologne, Cologne, Germany., 2. Department of Internal Medicine I, University of Cologne, Cologne, Germany.

78

81

The organic nitrate pentaerithrityl tetranitrate improves vascular oxidative stress and dysfunction in experimental hypertension Wenzel P. (1), Schuhmacher S. (1), Schulz E. (1), Oelze M. (1), Gori T. (1), Jansen T. (1), Kamuf J. (1), Mäthner F. (1), Li H. (2), Förstermann U. (2), Münzel T. (1), Daiber A. (1) Objectives: With the present study we tested in two animal models of experimental hypertension, whether chronic treatment with pentaerithrityl tetranitrate (PETN) or isosorbide mononitrate (ISMN) improves hypertension-associated vascular oxidative stress and dysfunction. Background: In contrast to other organic nitrates PETNtreatment induces neither nitrate tolerance nor cross-tolerance, which has been attributed to the induction of the expression of the antioxidant enzyme heme oxygenase1 (HO-1). Methods: Hypertension was induced pharmacologically by treatment of rats with angiotensin-II (AT-II, 1mg/kg/d) or genetically by using spontaneously hypertensive rats (SHRs). Co-treatment with PETN (15mg/kg/d) or ISMN (75mg/kg/d) was performed via subcutaneous infusion for 7d. Vascular function was assessed by isometric tension studies and reactive oxygen species (ROS) were detected by chemiluminescence, HPLC or fluorescent microtopography. Protein and mRNA expression was measured by Western blotting and RT-PCR, respectively. Results: Hypertensive rats showed impaired endothelial and smooth muscle function and increased vascular and cardiac ROS production (mitochondria, NADPH oxidase activity and uncoupled eNOS). In contrast to ISMN, co-treatment with PETN improved these parameters probably by induction of antioxidative HO-1 and of tetrahydrobiopterin synthesizing enzymes such as GTPcyclohydrolase-I and dihydrofolate reductase, which prevents eNOS uncoupling. Conclusions: This is the first demonstration that an organic nitrate improves rather than worsens endothelial dysfunction and oxidative stress in animal models of hypertension, suggesting that PETN treatment in addition to antianginal actions in patients with CAD may also slow down the progress of atherosclerosis. 1. II. Med. Klinik, Universitätsmedizin, Mainz, Germany, 2. Institut für Pharmakologie, Universitätsmedizin, Mainz, Germany

Activation of Nrf2 may prevent eNOS uncoupling under redox stress by transient downregulation of eNOS protein Heiss E.H. (1), Schachner D. (1), Dirsch V.M. (1) The synthetic triterpenoid 2-cyano-3,12-dioxooleana-1,9-dien-28-oic imidazolide (CDDO-IM) is known as potent activator of the redox-and xenobiotic-sensitive transcription factor NF-E2-related factor (Nrf2). We administered CDDO-IM to primary human umbilical vein endothelial cells in order to study the impact of Nrf2 activation on endothelial redox homeostasis and the endothelial nitric oxide synthase (eNOS) system. Activation of Nrf2 by CDDO-IM elicited a dose- and time-dependent reduction of intracellular reactive oxygen species (ROS) with optimal response after 24 hrs with a concentration of 100 nM. Consistently with lowered ROS the amount of bioavailable NO, a major contributor to vascular homeostasis, was increased presumably due to reduced inactivation of NO by superoxide to peroxynitrite. Interestingly and in apparent contrast to elevated NO levels, eNOS protein expression was transiently decreased. De novo protein synthesis of heme oxygenase 1 (HO-1), and subsequent heme deficiency was identified as cause for the Nrf2-dependent reduction of eNOS To reconcile our findings in a physiological context, we hypothesize that under redox stress when the availability of reduced BH4, a pivotal stoichiometric eNOS cofactor is limited, activation of Nrf2 could lead a) to reduction of ROS and b) transient reduction of eNOS protein levels that ensures the proper ratio of eNOS /BH4 to avoid eNOS uncoupling and endothelial dysfunction until original BH4 levels are restored by the antioxidant cellular response to active Nrf2. 1. Dept. of Pharmacognosy, University of Vienna, Vienna, Austria

79 Pro-oxidant action of activated factor X (FXa) in human vascular smooth muscle cells Jobi K. (1), Rauch B.H. (1), Schrör K. (1), Rosenkranz A.C. (1) Background: Oxidative stress is a hallmark of diabetes, atherosclerosis and vascular remodelling. Activated factor X (FXa) can elicit coagulation-independent effects such as vascular smooth muscle cell (SMC) proliferation and migration via protease-activated receptors PAR-1 and PAR-2. We investigated if FXa also induces oxidative stress in human vascular SMC. Methods: Human saphenous SMC were serum-deprived (48-72h) prior to stimulation with FXa (30nM). Intracellular oxidant stress was determined by dichlorofluorescein fluorescence. Total protein, mRNA and cell-surface expression levels of PARs and NADPH oxidase subunits were measured by western blotting, realtime PCR and FACS respectively. Results: FXa elicited a time-dependent induction (approximately 3-4-fold) of the NADPH oxidase subunit NOX-1 at both the mRNA (maximally at 1h) and protein level (6-24h, both n=4, P<0.05). Other NADPH oxidase subunits (NOX4, p47phox) remained unchanged. The increase in NOX-1 expression was validated by immunofluorescence and was accompanied by a significant increase in intracellular oxidant stress (n=5, P<0.05). NADPH oxidase inhibition (diphenyliodinium plus apocynin) completely prevented the oxidative response to FXa in preliminary studies. Given that oxidative stress regulates many genes associated with VSMC proliferation and inflammation, the influence of FXa on expression of the inflammatory cytokine interleukin-6, as well as of its own receptors PAR-1 and PAR-2, was examined. FXa time-dependently upregulated both IL-6 and PAR-2 mRNA (both n=6), while PAR-1 was not affected (n=6). PAR-2 total protein (n=4) and cell-surface expression (n=2) were also markedly induced over 6-24h, possibly indicating a redox-dependent feedback regulation. Conclusion: The coagulation factor FXa elicts oxidative stress in human vascular SMC by stimulating NADPH oxidase expression and activity. This is accompanied by upregulation of its receptor PAR-2 and of IL-6. Such actions are likely to support proliferative, migratory and inflammatory processes after vascular injury and contribute to vascular remodelling in vivo. 1. Institut für Pharmakologie und Klinische Pharmakologie, Universitätsklinikum Düsseldorf, Germany

82 Generation of animal model for increased vascular oxidative stress Suvorava T. (1), Weber M. (2), Valcaccia S. (1), Dao V.T-V. (1), Kojda G. (1) Purpose: Compelling evidence suggest that numerous pathological conditions are associated with increased vascular oxidative stress. Whether elevated vascular levels of reactive oxygen species promote cardiovascular pathology or are the consequences of the disease state is an important but ananswered question. Methods: We have generated a mutant of bovine eNOS in which cysteine 101 was replaced by alanine (C101A) resulting in uncoupling of eNOS. The effects of C101A mutation have been characterized in cultured human embryonic kidney cells (HEK) transfected with either mutant eNOS (eNOSmu) or wild type eNOS (eNOSwt). Transgenic mice carrying eNOSmu have been generated on a C57BL/6 background using the endothelium-specific Tie-2 promotor. By breeding these mice with eNOS knockouts (eNOS-/-), mice that only express eNOSmu (eNOS-/-/eNOSmu+) or no eNOS at all (eNOS-/-/eNOSmu-) were obtained. Results: Studies in HEK cells transfected with eNOSmu revealed decreased eNOS activity, increased superoxide generation inhibited by L-NAME treatment, less NO production (45.8±6.2 %), strongly elevated protein oxidation level and decreased eNOS dimer stability in native SDS-gels as compared to eNOSwt suggesting that eNOSmu has the typical features of uncoupled eNOS. Organ bath experiments showed increased aortic sensitivity of eNOS-/-/eNOSmu+ to NO-donor S-nitroso-N-acetyl-penicillamine. In striking contrast, there was a significant rightward shift in acetylcholine dose-response curve in eNOS-/-/eNOSmu+ (p=0.0062 vs C57Bl/6, n=11-13). eNOS-/-/eNOSmu+ mice have elevated systolic blood pressure as compared to C57Bl/6 (138±2.3 vs 118.4±3.1 mmHg in C57BL/6, n=6-8, p<0.05), but not to eNOS-/-/eNOSmu- (135.5±0.8, n=8). Western blot analysis confirmed the expression of eNOS in eNOS-/-/eNOSmu+ and revealed an increased phosphorylation of eNOS at Ser1179 in aorta, heart and muscle of eNOS-//eNOSmu+ as compared to C57Bl/6 (p<0.05, n=4-8). Amount of f circulating stem cells with endothelial progenitor capacity (EPCs) defined as CD34/Flk-1 or as Sca-1/Flk-1 double positive cells was significantly lower in peripheral blood of eNOS-/-/eNOSmu+ as compared to eNOS-/-/eNOSmu- (n=7-8, p<0.05) strongly suggesting a role of uncoupled eNOS and increased vascular oxidative stress in regulation of circulating EPCs. Conclusions: Endothelial specific overexpression of uncoupled eNOS and increased vascular oxidative stress decrease the number of circulating stem cells with endothelial progenitor capacity. 1. Dept. of. Pharmacology, Heinrich-heine University, Düsseldorf, Germany, 2. Division of Cardiology, Emory University School of Medicine, Atlanta, USA

83 80 p-Glycoprotein and its interaction with the ROS indicator dihydrorhodamine Bauer T. (1), Jung N. (2), Rösen R. (1) Introduction: Dihydrorhodamine 123 (DHR123) is used as indicator for oxidative stress. DHR123 passively diffuses across the cell membrane and is oxidized by ROS to the fluorescent dye rhodamine 123 (R123) which should accumulate intracellularly. On the other hand R123 is used as a substrate for p-Glycoproteine (pGP) to characterize its transport effectiveness. We tested the hypothesis that pGP influences rhodaminefluorescence by active R123 transport. Methods: Confluent human endothelial cells (EaHY.926) were incubated with DHR123. In the absence or presence of cyclosporine A, a pGP-inhibitor, oxidative stress was induced by glycerol trinitrate (GTN). R123 was quantified using fluorescence microscopy or after extraction with an organic solvent by HPLC. R123 efflux was measured by a FACS-based assay. Results: In EaHy.296 an efflux of R123 was observed which was inhibitable by cyclosporine A. After incubation of EaHy.926 with GTN an increase in ROS generation was observed after 5 min. To our surprise the detected intracellular fluorescence decreased with longer incubation time. In the presence of cyclosporine A this effect was abolished. Conclusion: When using DHR123 as a tool for ROS detection it must be considered that intracellular concentrations of R123 may be altered by pGP. To exclude this possibility we recommended either to determine intra- and extracellular R123 concentrations or inhibit the transport proteins.

A small molecule inhibitor of Bid prevents mitochondrial fission and cell death signaling in neurons exposed to oxidative stress Grohm J. (1), Culmsee C. (1) Mitochondria are dynamic organelles that undergo permanent fission and fusion under physiological conditions. Oxidative stress, which is proposed as a key trigger of neurodegenerative processes in neurological diseases such as Alzheimer's disease or Parkinson's disease, has been recently linked to pathological mitochondrial fission. In the present study we investigated the effects of the small molecule inhibitor of Bid, BI6c9, in models of oxidative stress in immortalized hippocampal neurons (HT-22 cells) or glutamate-induced excitotoxicity in primary embryonic neurons, respectively. Mitochondrial morphology was analyzed either in cells transfected with a mitochondriatargeted GFP vector (Mito-GFP) or by using fluorescent Mitotracker dyes. In HT-22 cells, oxidative stress resulted in pronounced fragmentation of the mitochondria which appeared as short tubules and round fragments in contrast to long tubular structures in untreated controls. Moreover, mitochondrial fission was accompanied by loss of mitochondrial membrane integrity as evaluated by the JC-1 assay. Bid translocation to mitochondria preceded mitochondrial fission, membrane permeability and AIF translocation to the nucleus, as evaluated in cells expressing Bid-red fusion protein and AIF-GFP, respectively. Notably, overexpression of full-length Bid neither affected mitochondrial dynamics nor cell viability. The Bid-inhibitor BI-6c9 prevented Bid translocation to mitochondria and mitochondrial fission, and protected the cells from mitochondrial demise and AIF-dependent cell death. Moreover, pharmacological

23

inhibition of Drp1, a key regulator of mitochondrial fusion/fission-machinery, also attenuated mitochondrial fission and prevented HT-22 cells and primary neurons from glutamate toxicity. In conclusion, neuronal cell death induced by oxidative stress or glutamate-induced excitotoxicity involved activation and mitochondrial translocation of Bid thereby causing mitochondrial fission and AIF-mediated cell death. In addition to interference with Bid activation, inhibition of Drp-1 is proposed as a novel strategy to prevent mitochondrial fragmentation and neuronal cell death in these model systems of neurodegenerative diseases. 1. Klinische Pharmazie, Institut fuer Pharmakologie und Toxikologie, PhilippsUniversitaet Marburg, Germany

84 Inhibition of p53 preserves mitochondrial morphology and function, and prevents glutamate-induced cell death in neurons Diemert S. (1), Grohm J. (1), Hartmannsgruber R. (1), Culmsee C. (1) Activation of p53 by oxidative stress or DNA damage has been identified as a key mediator of programmed cell death in various model systems of neurological diseases, including cerebral ischemia or brain trauma. Enhanced p53 transcriptional activity may increase the expression of pro-apoptotic bcl-2 family proteins such as Bax and concomitantly represses neuroprotective activities of the transcription factor NF-kB. Yet another emerging pro-apoptotic effect of p53 may result from its direct translocation to mitochondria, where it disrupts mitochondrial membrane integrity, thereby inducing the release of pro-apoptotic factors and the generation of reactive oxygen species. Here, we investigated the potential role of p53 in mitochondrial demise in models of oxidative stress or DNA damage induced in immortalized HT-22 neurons by glutamate or camptothecin, respectively. After exposure to glutamate p53 levels increased in HT-22 cells, and translocation of p53 to mitochondria was detected within 6 h and up to 18 h after induction of oxidative stress. The p53 inhibitor pifithrin-alpha (PFT) enhanced mitochondrial fusion and significantly reduced mitochondrial fission and cell death induced by the glutamate challenge. Similar protective effects were observed in a paradigm of camptothecin-induced DNA damage. Inhibition of NF-kB blocked the neuroprotective effects of PFT whereas activation of NF-kB correlated well with protective effects of novel PFT derivates, suggesting an involvement of the NF-kB transcriptional activity in the detected protective effects. Furthermore, PFT significantly attenuated mitochondrial ROS formation and preserved mitochondrial membrane potential in synaptosomal preparations exposed to oxidative stress induced by Fe2+, suggesting that at least part of the pro-apoptotic activity of p53 was mediated directly at the mitochondrial level and is therefore independent of transcriptional activity. In conclusion, inhibition of p53 preserved mitochondrial morphology and mitochondrial membrane integrity, and enhanced neuronal survival following oxidative or genotoxic stress. These protective effects of PFT may result from interference with p53 transcriptional activity, enhanced activation of NF-kB, and inhibition of p53 activities at the level of mitochondria. 1. Klinische Pharmazie, Institut fuer Pharmakologie und Toxikologie, PhilippsUniversitaet Marburg, Germany

85 Pasteurella multocida toxin activates heterotrimeric G proteins by a covalent modification Orth J. (1), Preuß I. (1), Fester I. (1), Aktories K. (1) PMT is a major virulence factor of Pasteurella multocida causing pasteurellosis in man and animals and is responsible for atrophic rhinitis in pigs. PMT modulates various signaling pathways by acting on the heterotrimeric G proteins Gαq and Gα12/13. PMT activates Gq but not G11 to increase inositol phosphate production via phospholipase Cβ and alteration of gene expression via the JAK/STAT pathway. The toxin also activates RhoA via Gαq and Gα12/13 family proteins, resulting in formation of stress fibers and focal adhesions. Recently, we reported that PMT is a powerful activator of Gi protein. Here, we show that PMT causes a covalent modification, which leads to constitutive activation of the G protein. A similar modification is also evident for Gαq, but not for the closely related Gα11, which is not a substrate of PMT. Our findings identify the α-subunits of heterotrimeric G proteins as the direct molecular target of PMT and show the type of covalent modification. 1. Dept. of Pharmacology and Toxicology, Albert-Ludwigs-University, Freiburg, Germany

86 Nucleoside diphosphate kinase C modulates the membrane content and function of heterotrimeric G proteins Abu-Taha I.H. (1), Wolf N.M. (2), Lutz S. (1), Hippe H.-J. (2), Wieland T. (1) The formation of complexes of heterotrimeric G proteins with nucleoside diphosphate kinases (NDPK), particularly NDPK B, has been shown to be involved in receptorindependent G protein activation, but the mechanisms that regulate the complex formation and membrane attachment of NDPKs remain elusive. NDPK C is a member of the NDPK family which differs from other NDPK isoforms by a hydrophobic N-terminal sequence forming a membrane insertion domain. We therefore investigated the role of NDPK C in G protein complex formation and function. Using far blots and coimmunoprecipitation assays, we demonstrated a direct protein-protein interaction between NDPK C and heterotrimeric G proteins in vitro and in living cells. In addition, adenoviral overexpression of NDPK C in mouse embryonic fibroblasts enhanced the content of G protein αs- and αq/11-subunits as well as β1- and β2-subunits at the membrane. The increase in membranous Gas content was accompanied by enhanced basal cAMP synthesis (≈ 200%) in NDPK C over-expressing cells. Vice versa, siRNAmediated knockdown of endogenous NDPK C in mouse embryonic fibroblasts decreased cAMP levels by about 50%. Our data therefore indicate that the hydrophobic NDPK C isoform might be a crucial component in the complex formation of NDPKs with heterotrimeric G proteins. It apparently modulates the G protein membrane content and contributes to basal G protein function.

1. Department of Experimental and Clinical Pharmacology and Toxicology, University of Heidelberg, Mannheim, Germany, 2. Department of Internal Medicine III, Cardiology, Angiology, Pulmonology, University of Heidelberg, Heidelberg, Germany

87 Regulator of G protein signaling 3 expression gradually shifts Gi/o-induced activation of RhoGTPases dependent on phosphorylation by AKT/PKB Wittig K. (1), Vogt A. (1), Köster G. (1), Lutz S. (1), Wieland T. (1) The 519 aa isoform of regulator of G protein signaling 3 (RGS3L), an abundant RGS protein in the heart, has recently been shown to regulate RhoGTPase activation by recombinantly expressed M2 muscarinic receptors (mAChR). As the expression of RGS3 in neonatal rat cardiomyocytes (NRCM) is under control of fibroblast growth factor-2 (FGF-2), a known cardioprotective stimulus, we studied the influence of FGF-2 and RGS3L expression on carbachol-induced activation of Rac1 and RhoA in NRCM. In control NRCM, a prominent mAChR-induced Rac1 activation was accompanied by a minor RhoA activation. FGF-2 treatment caused a marked increase in RGS3L expression which was accompanied by a very weak carbachol-induced activation of Rac1 and a strong mAChR-induced activation of RhoA. To test wether RGS3L is of importance in the mAChR-induced shift from Rac1 to RhoA activation we adenovirally overexpressed RGS3L. Increasing amounts of RGS3L led to incrementally enhanced RhoA and suppressed carbachol-induced Rac1 activation. Vice versa, siRNA-induced depletion of endogenous RGS3L abolished mAChR-induced RhoA activation without affecting carbachol-stimulated Rac1 activation. Both, carbachol-induced Rac1 and RhoA activation were sensitive the phosphatidylinositol 3-kinase (PI3K) inhibitor LY294002 in control as well as FGF-2 treated NRCM. RGS3L contains a putative phosphorylation site for AKT/PKB at S264. As AKT/PKB activity is regulated by PI3K, we tested whether mutation of S264 to alanine or glutamate alters the influence of RGS3L on the activation of RhoGTPases by recombinantly expressed mAChRs. Both, expression of the phosphorylation resistant S264A as well as the phosphomimetic S264E mutant suppressed the carbachol-induced activation of Rac1 comparable to the non mutated control. In contrast, RGS3L-S264A was no longer able to propagate carbachol-induced activation of RhoA whereas RGS3L-S264E was even more potent than the wildtype protein. We conclude that the FGF-2 controlled expression level of RGS3L alters the activation pattern of mAChR for RhoGTPases in NRCM. Apparently, the phosphorylation of RGS3L at S264 is required to induce this shift. 1. Institute of experimental and clinical pharmacology and toxicology, Medical Fakulty Mannheim, University of Heidelberg, Mannheim, Germany

88 Active alpha2a-receptors inactivate GTPgammaS-preactivated Go proteins Hommers L.G. (1), Dees C. (1), Bünemann M. (1) G protein-coupled receptors catalyze the exchange of GDP by GTP at the Gα subunit by stabilizing a nucleotide-free intermediate and thereby activate G proteins. We aimed to investigate for the model system α2A-receptors, Gi/o signal cascade, whether a “reversed G protein activation” may exist: catalysis of GTP- or GTPγS from preactivated G proteins by active receptors. Monitoring agonist-induced G protein activation and receptor/G protein interaction in single living cells has become possible by means of FRET between differently fluorescence-tagged receptors and G protein subunits. In order to conduct these experiments, control of the intracellular nucleotide composition was gained by permeabilizing the cell membrane of transiently transfected HEK293 cells via reversible addition of saponine.The agonist-induced amplitude of FRET between tagged receptors and G proteins was significantly stronger in the presence of low concentration of GDP, GTP and GTPγS compared to corresponding high concentrations of each nucleotide. Superfusion of cells with GTPγS in the presence of agonist was monitored to result in maximal activation of G proteins by means of FRET between tagged G protein subunits. When omitting GTPγS from the superfusion, the FRET signal recorded over 5 minutes corresponded to a deactivation of G proteins with kinetics depending on the concentration of superfused agonist. As a control experiment, nonlabeled GTPγS could displace radioactively labeled GTPγ35S in an agonist dependent manner. Monitoring G protein activation by means of FRET in the presence of low concentrations of GTPγS suggested a reduced G protein activation during superfusion of cells with saturating compared to non-saturating concentration of agonist. Paralleling agonist-induced receptor/G protein interaction, this effect was antagonized by the presence of high concentration of GTPγS.Taken together the results suggest (I) interaction of active α2A-receptors with GTPγS-preactivated Go proteins, (II) loss of GTPγS upon this interaction and (III) sequestration of nucleotide-free G proteins by active receptors. This mechanism sufficiently explains known agonist-induced inhibition of preactivated effectors such as G protein activated inwardly rectifying potassium channels (GIRK). 1. Dept. of Pharmacology, Julius Maximilians University, Wuerzburg, Germany

89 Pasteurella multocida toxin affects GTPase activity of heterotrimeric G proteins Fester I. (1), Preuß I. (1), Orth J. (1), Aktories K. (1) Pasteurella multocida toxin (PMT) is one of the strongest activators of Gαq-dependent phospholipase C (PLC) β to induce inositoltrisphosphate production, Ca2+ mobilization and formation of diacylglycerol. The toxin also activates the small GTPase RhoA, resulting in formation of stress fibers and focal adhesions. The activation of RhoA depends on Gαq and Gα12/13. Interestingly, Gα11, a member of the Gq family is not stimulated by PMT. In addition, PMT induces MAP kinase and STAT activation. Additionally, PMT is a powerful activator of Gi protein. PMT decreases basal isoproterenol and forskolin-stimulated cAMP accumulation in intact Swiss 3T3 cells and inhibits adenylyl cyclase activity in cell membrane preparations. Although the effects of PMT are not inhibited by pertussis toxin (PTx), PMT blocks PTx-catalyzed ADPribosylation of Gi. PMT also inhibits steady-state GTPase activity and GTP binding of Gi in Swiss 3T3 cell membranes stimulated by lysophosphatidic acid and single cycle GTPase activity of recombinant Gαi. The data indicate that PMT is a novel activator of Gi, modulating its GTPase activity and converting it into a PTx-insensitive state.

24

1. Dept. of Pharmacology and Toxicology, Albert-Ludwigs-University, Freiburg, Germany

90 Impact of H-Ras on activation of monomeric and heterodimeric phosphoinositide 3 kinase γ Kurig B. (1), Prajwal P. (1), Nürnberg B. (1) Class I phosphoinositide 3-kinases (PI3Ks) regulate fundamental cellular responses like proliferation, survival and migration. Activation of all class I PI3Ks leads to the formation of PtdIns-3,4,5-P3, which results in plasma membrane recruitment and subsequent activation of proteins containing a PIP3-sensitive pleckstrin homology domain. Specificity of these responses is supported by the existence of multiple PI3K-isoforms and their differential regulation. Class IB PI3Kγ is regulated by G-protein-coupledreceptors (GPCRs) via Gbγ-dependent recruitment and activation. Despite this well characterised regulatory mechanism, GTP-bound H-Ras is able to bind directly to and activate p110γ. Initial data have suggested a higher impact of H-Ras on the regulation of PI3Kγ, as assumed before. But information is lacking, concerning the exact contribution of Ras on receptor induced activation of PI3Kγ. Furthermore, the role of the noncatalytic subunits in H-Ras-dependent activation of PI3Kγ is unknown. To address these questions we have employed confocal laser scanning microscopy, allowing us to visualize fluorescently labeled proteins in living cells. Surprisingly, and in contrast to cytosolic p110γ, membrane bound p110γ-CAAX could be activated by wild type H-Ras, indicating that membrane localisation is an important prerequisite not only for activation of PI3Kγ via Gbγ, but also by H-Ras. We therefore asked, whether H-Ras is able to directly recruit p110γ to membranes. Based on previously published in vitro and ex vivo experiments H-Ras induced recruitment was denied. However, analysing intact, living cells revealed accumulation of p110γ at the plasma membrane upon cotransfection of constitutive active, but not of inactive H-Ras. Colocalisation of p110γ and H-Ras-V12 was confirmed by immunofluorescence analysis. Furthermore, H-Ras dependent recruitment was able to substitute for Gbγ-p101-dependent recruitment, allowing GPCRinduced activation of PI3Kγ in the absence of p101. These findings indicate firstly that the role of Ras in the stimulation process of PI3Kγ is not only to activate, but also to recruit the lipid kinase to membranes. Secondly, comparison of the activation of monomeric and heterodimeric PI3Kγ by H-Ras revealed, that the presence of a noncatalytic subunit increased the impact of H-Ras on the activity of p110γ. We therefore conclude, that p87 and p101 exhibit an auxiliary effect on Ras-dependent activity of p110γ. 1. Institut für Experimentelle und Klinische Pharmakologie und Toxikologie, Klinikum der Eberhard-Karls-Universität, Tübingen, Germany

91 C. sordellii lethal toxin-induced apoptotic cell death is based on inhibition of phosphoinositide-3 kinase Akt signalling Schulz F. (1), Just I. (1), Genth H. (1) Clostridium sordellii causes myonecrosis in obstetric patients and necrotizing fasciitis in injection drug users, in severe cases accompanied by a toxic shock syndrome. Lethal Toxin (TcsL) is regarded as the major pathogenicity factor. TcsL inactivates low molecular weight GTP-binding proteins of the Rho- and Ras-subfamily by monoglucosylation. Inactivation of Rho/Ras proteins leads to actin-reorganization (“cytopathic effect”) and apoptotic cell death (“cytotoxic effect”). Inactivation of Rho/Ras-proteins by TcsL causes an inhibition of the phosphoinositide-3 kinase (PI3K)/Akt pro-survival pathway, resulting in activation of the mitochondrial cell death pathway. In this study, the role of PI3K/Akt signaling in the TcsL-induced cytotoxic effect was analyzed using tauroursodeoxycholic acid (TUDCA), an anti-apoptotic agent that activates PI3K/Akt. TcsL induced the cytotoxic effect in macrophage-like J774A.1 cells as analyzed in terms of phosphatidylserine exposure (AnnexinV-staining), reduction of metabolic activity (WST-1 assay), and activation of caspase-3. Pre-treatment of J774A.1 cells with TUDCA prevented the cytotoxic effects of TcsL. The specific PI3K-inhibitor LY294002 as well as the Akt inhibitor abolished the protective effect of TUDCA, showing that the cytotoxic effects of TcsL depended on inhibition of PI3K/Akt signaling. TcsL is highly related to Toxin B (TcdB) from Clostridium difficile. TcdB (together with Toxin A) is the causative agent of the C. difficile-associated diarrhea and its severe form, the pseudomembranous colitis. TcdB exclusively inactivates Rho proteins (not Ras proteins). In contrast to TcsL, TUDCA failed to prevent the cytotoxic effects of TcdB. This finding showed that inhibition of PI3K/Akt signaling is critical for TcsL- but not for TcdB-induced apoptosis. TcsL and TcdB thus initiate apoptotic cell death by inhibiting distinct survival pathways. 1. Institut für Toxikologie, Medizinische Hochschule Hannover, Hannover, Germany

92 Difference in the cytotoxic effects of Toxin B from Clostridium difficile in proliferating and non-proliferating cells Lica M. (1), Schulz F. (1), Huelsenbeck J. (2), Just I. (1), Genth H. (1) Toxin A (TcdA) and Toxin B (TcdB) from Clostridium difficile are the causative agents of the C. difficile-associated diarrhoea (CDAD). The CDAD is characterized by a loss of mucosal barrier function, secretory diarrhoea, and colonic inflammation. TcdA and TcdB both glucosylate and thereby inactivate low molecular weight GTP-binding proteins of the Rho subfamily. In cultured cell lines, TcdB induces actin re-organization (“cytopathic effect”) and cell death (“cytotoxic effect”). The onset of the execution of apoptotic cell death is asynchronous across a population of proliferating cells. Therefore, cells in distinct phases of the cell cycle may exhibit differential susceptibility to TcdB-induced apoptotic cell death. This hypothesis was corroborated in Hela cells synchronized in early S-phase using the thymidine double block technique. The cells were released from the block and treated with TcdB either in S-phase (after 1h) or in G2/M phase (after 6h) or in G0/G1-phase (after 12h), as determined by FACS analysis. Apoptosis was analysed in terms of phosphatidylserine exposure (Annexin V-staining), reduction of cellular viability (WST-1 test), and DNA fragmentation (analysis of the sub-G1 population).S-phase cells turned out to be most sensitive, while cells in G1-phase were almost completely insensitive to TcdB-induced apoptosis. To confirm this finding, we

further analysed TcdB-induced apoptosis in human colonic HT29 cells synchronized in G1/G0-phase through density inhibition. G1/G0-arrested cells were almost insensitive towards the cytotoxic effect, while a proliferating population of cells was sensitive to TcdB-induced apoptosis. HT29 cells synchronized in G1/G0-phase through density inhibition may represent a model for non-proliferating enterocytes. TcdB apparently was not capable of inducing apoptosis in such cells. Regarding the pathology of CDAD, TcdB seems not to contribute by causing apoptosis of colonic cells. One may assume rather necrosis of these cells to be more important for the development of the disease, as necrosis is accompanied by a release of cytoplasmic contents, leading to local inflammation. 1. Institut für Toxikologie, Medizinische Hochschule Hannover, Hannover, Germany, 2. Institut für Toxikologie, Johannes Gutenberg Universität, Mainz, Germany

93 Neurotrophic effect of Clostridium botulinum C3 Dreger S.C. (1), Hofmann F. (1), Mühlenstädt C. (1), Radtke C. (2), Höltje M. (3), Ahnert-Hilger G. (3), Just I. (1) C3bot from Clostridium botulinum is the prototype of the family of C3-like exoenzymes, which inactivates the Rho family GTP-binding proteins RhoA/B/C by ADP-ribosylation. Based on its substrate specificity, C3 is an established pharmacological tool for the analysis of cellular functions of Rho proteins. Rho proteins play versatile roles in axonal growth and branching. Incubation of neurons with C3bot prevents growth cone collapse and induces axonal growth. These effects have been attributed to C3bot-induced inactivation of axonal RhoA. Based on this neurotrophic activity, cell permeable C3bot is currently applied as drug to treat acute spinal cord injury. We found, however, that exclusively the isoform C3bot (from C. botulinum) exhibited neurotrophic activity, while other isoforms of C3 failed to do so. Surprisingly, enzyme-deficient C3bot was as effective as the enzyme-competent wild type C3bot. Finally, a C-terminal region covering 29 amino acid residues was identified to be sufficient for the neurotrophic activity. Thus, this activity of C3bot is not based on its intrinsic enzymatic activity but is mediated by a restricted surface exposed region. The C3bot peptide (29 aa) acted on primary murine hippocampal neurons but also on terminally differentiated human NT2-N neurons. Its activity on neurons was characterised by an increase in axonal and dendritic growth and branching as well as in synaptic connectivity. The activity of C3bot peptide was restricted to neurons, in contrast to enzyme-competent C3bot, which additionally acted on astrocytes, enhancing migration and glutamate release. On microglia, it triggered the release of pro-inflammatory cytokines. Both effects strictly depended on the ADP-ribosylation of Rho. Thus, the C3bot peptide acted selectively trophic on neurons, making it a promising compound to treat neuronal lesions. In order to extend our findings from the cellular system to an animal model, we compared the neurotrophic activities of C3bot peptide and enzyme-competent C3bot in the rat sciatic nerve lesion model. The C3bot peptide turned out to be superior to enzyme-competent C3 in the recovery rate of the damaged sciatic nerve. 1. Institut für Toxikologie, Medizinische Hochschule Hannover, Germany, 2. Klinik für Plastische-, Hand und Wiederherstellungschirurgie, Medizinische Hochschule Hannover, Germany, 3. Centrum für Anatomie, Charité-Universitätsmedizin Berlin, Germany

94 Inhibition of cytokinesis by Toxin B from Clostridium difficile May M. (1), Just I. (1), Genth H. (1) Low molecular weight GTP-binding proteins of the Rho subfamily control many cellular processes, including the dynamics of the actin cytoskeleton, cell-cell and cell-matrix adhesion, cell polarity, gene expression, and cell cycle progression. In cytokinesis, RhoA activity is critical for contractile ring formation in cytokinesis. Rac1 and Cdc42 regulate entry into mitosis through activation of PAK and Aurora A kinase. RhoA, Rac1 and Cdc42 are the intracellular targets of Toxin A (TcdA) and Toxin B (TcdB) from C. difficile, the causative agents of C. difficile-associated diarrhoea (CDAD). These toxins glucosylate and thereby inactivate RhoA, Rac1, and Cdc42. In this study, we analyzed the effects of TcdB on mitosis entry and cytokinesis. Hela cells were synchronized by the thymidine double block in early S-phase. Cells were then allowed to release from the block. After 8 h, the cells were in G2-phase and treated with TcdB. Inactivation of the Rho proteins caused a delay of 2h in G2-M transition, as analyzed in terms of Aurora A and histone H3 phosphorylation. Treatment with TcdB further prevented contracile ring formation, as visualized by rhodamin-phalloid-staining of fixed cells by fluorescence microscopy. Cytokinesis inhibition of TcdB resulted in the formation of multinucleated cells. Multinucleation was also observed in primary mouse embryonic fibroblasts, excluding that TcdB-induced multi-nucleation was present only in immortalized cell lines. A Toxin B isoform (TcdBF) glucosylates Rac1 and Cdc42 but not RhoA. Contracile ring formation and cytokinesis but not multi-nucleation were observed in TcdBF-treated cells, confirming that RhoA (but not Rac1 or Cdc42) activity was critical for cell division. Progression through G2-M was delayed in TcdBF-treated cells, based on impaired (Rac1 / Cdc42-dependent) centrosomal activation of Aurora A. Delayed cell cycle progression and inhibition of cytokinesis inhibition by Toxin B isoforms has not yet been discussion with respect to CDAD. Permanent renewal of gut epithelial surfaces is one key mechanism against bacterial colonization. Renewal of mucosal surfaces in colon depends on the activity of epithelial progenitor cells within the Lieberkuehn crypts. It is tempting to speculate that delayed cell cycle progression or inhibited cytokinesis in these progenitor cells prevents epithelial renewal, which in turn promotes colonization of C. difficile. 1. Institut für Toxikologie, Medizinische Hochschule Hannover, Hannover, Germany

25

95

TcdE in colonization of C. difficile in the host intestine and in releasing its pathogenic factors causing diarrhea and pseudomembranous colitis. 1. Dept. of Toxicology, Hannover Medical School, Hannover, Germany

Regulation of phospholipases C-gamma1 and gamma2 by the B cell antigen receptor in DT40 B cells Walliser C. (1), Deuschle K. (2), Glaschik S. (2), Röcker C. (2), Nienhaus G.U. (2), Gierschik P. (1) We have recently found that PLCγ2, but not PLCγ1, is activated in intact cells by active mutants of Rac. The mechanism of PLCγ2 activation by Rac GTPases was independent of tyrosine phosphorylation and enhanced formation of phosphatidylinositol 3,4,5trisphosphate. PLCγ2 is an important regulator of many functions of hematopoietic cells, e.g. platelets, neutrophils, macrophages, osteoclasts, mast cells, and B lymphocytes. In murine platelets, activation of PLCγ2 by glycoprotein VI and CLEC-2 (C-type lectin-like) receptors has been shown to be mediated by Rac1, indicating that this novel pathway is critical for thrombus formation in vivo. In B cells, PLCγ2 couples the B cell antigen receptor (BCR) to proliferation, differentiation, and cell death, to establish complex immunological responses such as tolerance and memory. The relative importance of PLCγ1 in mediating BCR signals in B cells is currently unclear. Both PLCγ1 and PLCγ2 have been shown to be expressed in B cells and to be substrates of BCR-associated tyrosine kinases. Furthermore, the mobilization of intracellular Ca2+ has been shown to be impaired in a PLCγ1-deficient B cell line and reduced expression of PLCγ1 in the absence of PLCγ2 impeded early B-cell development at the pro-B- to pre-B-cell transition, indicating defective pre-BCR signaling. Using cultured chicken DT40 B lymphoblastoid cells and analysis of the intracellular Ca2+ concentration in single cells by spinning disc confocal laser scanning microscopy, we now show that wild-type DT40 cells display a marked and sustained oscillating increase of the intracellular Ca2+ concentration in response to BCR ligation in the absence of extracellular Ca2+ and a further increase due to influx across the plasma membrane upon addition of Ca2+ to the medium. Both responses are completely absent in PLCγ2-/- cells. Expression of PLCγ2, but not of PLCγ1, in PLCγ2-/- cells rescued the oscillating response of intracellular Ca2+ to BCR activation. These results indicate that PLCγ2, but not PLCγ1, is specifically coupled to BCR in DT40 B cells and that this cellular system is well suited for structure-functionanalyses of the role of Rac GTPases in regulating this receptor-mediated PLCγ2 activation. 1. Institute for Pharmacology and Toxicology, University of Ulm, Germany, 2. Institute for Biophysics, University of Ulm, Germany

Role of p38 MAP kinase in the up-regulation of the immediate-early gene product RhoB Roggenkamp D. (1), Dreger S.C. (1), Gaestel M. (2), Huelsenbeck J. (3), Fritz G. (3), Schmidt G. (4), Just I. (1), Genth H. (1) The low molecular weight GTP-binding protein RhoB is an immediate-early gene product that is up-regulated in response to growth factors as well as to genotoxic (alkylating compounds, anti-metabolites, ionizing radiation) and cytotoxic treatment (bacterial protein toxins, farnesyl-/geranylgeranyl-transferase inhibitors, statins). RhoB is suggested to be an important regulatory element, which participates in the decision of a damaged cell to undergo apoptotic cell death or to activate protective mechanisms. Those pathways regulating RhoB up-regulation remain to be identified but stressdependent protein kinases are likely involved. In this study, we investigated the possible role of p38 MAP kinase in RhoB up-regulation. Treatment of murine fibroblasts with the cytotoxic necrotizing factor 1 (CNF1) from Escherichia coli causes a strong activation of p38 MAP kinase and up-regulation of rhoB mRNA and RhoB protein. RhoB upregulation also was enhanced in cells pre-treated with SB203580, an inhibitor of p38 MAP kinase. Accordingly, RhoB up-regulation was more pronounced in p38 (-/-) fibroblasts compared to the wild-type cell line, although the half life period of RhoB protein and rhoB mRNA were comparable in p38(-/-) and wildtype cells. These findings suggest a negative role of p38 MAP kinase in RhoB up-regulation. Consistently, CNF1induced activation of rhoB promoter was more pronounced in p38 (-/-) than in wild-type cells. In sum, p38 influences the RhoB level by suppression of the rhoB promoter rather than by regulating rhoB mRNA stability. 1. Institut für Toxikologie, Medizinische Hochschule Hannover, Hannover, Germany, 2. Institut für Physiologische Chemie, Medizinische Hochschule Hannover, Hannover, Germany, 3. Institut für Toxikologie, Johannes Gutenberg Universität, Mainz, Germany, 4. Institut für Pharmakologie and Toxikologie, Albert-Ludwigs-Universität, Freiburg, Germany

96

99

Cleavage of the Escherichia coli cytotoxic necrotizing factor 1 (CNF1) is required for full biologic activity Knust Z. (1), Aktories K. (1), Schmidt G. (1) The cytotoxic necrotizing factor 1 (CNF1) is a protein toxin mainly produced by uropathogenic Escherichia coli (UPEC) strains. It belongs to the family of Rho-activating bacterial protein toxins. Other members of this family are the E. coli toxins CNF2 and CNF3, the Y. pseudotuberculosis CNFy and the dermonecrotic toxin (DNT) produced by B. pertussis. The activation of Rho GTPases occurs by deamidation (transglutamination by DNT) of a specific glutamine, which is crucial for GTP hydrolysis, resulting in a constitutively active protein. This causes extensive stress fiber formation and multinucleation of cultured cells. CNF1 is taken up into mammalian cells by receptormediated endocytosis in a clathrin and caveolin independent manner, and is delivered from late endosomes into the cytosol. Up to date, it was generally accepted that the toxin reaches the cytosol as a full-length toxin. By immunoprecipitation with an antibody against the catalytic portion, we could identify a fragment of CNF1. This C-terminal fragment is only present in the cytosol and shows deamidation activity. Its processing requires an acidic pH and insertion of the toxin into the endosomal membrane. Furthermore we show that this cleavage is performed by a serine protease which is either localized in the endosomes or within the toxin itself. In addition, we define the region of toxin-cleavage by mutational analysis. 1. Institut für Experimentelle und Klinische Pharmakologie und Toxikologie, AlbertLudwigs-Universität, Freiburg, Germany

Autoproteolytic cleavage of Clostridium difficile Toxin A is not essential for its function Gerhard R. (1), Kreimeyer I. (1), Tatge H. (1), Marckscheffel A. (1), Just I. (1) Toxin A and toxin B from Clostridium difficile are single chain proteins and the causative agents of the antibiotic-associated pseudomembranous colitis. They are of an A/B-type of structure and possess autoproteolytic cleavage activity to release their glucosyltransferase domain to the cytoplasm of target cells. In the present study we show that extracellular cleavage of the transferase domain leads to functional inactivation of these toxins. In contrast to TcdB which is almost quantitatively cleaved by addition of Inostol-hexakisphosphate, TcdA shows poor extracellular autoproteolytic cleavage. MALDI-MS analysis identified the cleavage site of the glucosyltransferase domain of TcdA between Leu542 and Ser543. We mutated the cleavage site of TcdA to generate a non-cleavable TcdA that is completely resistant to autocleavage but still exhibited cytopathic and cytotoxic properties. Although the specific glucosyltransferase activity of mutant TcdA was identical to that of wild type TcdA, the EC50 regarding the cytopathic effect was about 75-fold reduced. The cytotoxic effect, i.e. activation of caspase-3, however, was identical. Thus, extracellular cleavage inactivates TcdA and TcdB whereas intracellular cleavage is not a prerequisite for the biological activity of TcdA. In addition, we found that conformational changes induced by acidic pH (5.0), accelerates the poor autoproteolytic cleavage of full length TcdA. In contrast, the cysteine protease activity (when present in the toxin fragment aa 1-1065), is enhanced at pH 7.4 but blocked at pH 5. From these data it can be concluded that the pHgoverned TcdA translocation from the endosomes also contributes to autoproteolytic release of the glucosyltransferase domain. Furthermore, cleavage is not essential for the toxic effect of TcdA, it seems to modulate only the potency. 1. Institut für Toxikologie, Medizinische Hochschule, Hannover, Germany

97 Structure-function analysis of the holin-like TcdE from Clostridium difficile Olling A. (1), Seehase S. (1), Tatge H. (1), Just I. (1), Gerhard R. (1) The pathogenic toxins A (TcdA) and B (TcdB) produced by Clostridium difficile monoglucosylate, and thereby inactivate small GTPases of the Rho subfamily. The tcdE ORF encoding the 19 kDa protein TcdE is located between the genes for TcdA/B. TcdE shows significant sequence homology to bacteriophage-encoded holins which are lytic proteins causing cell death of bacteria. This finding led to the hypothesis that TcdE participates in the release of TcdA/B into the extracellular environment of C. difficile. Since expression of TcdE in E. coli leads to bacterial cell lysis growth profiles of E. coli expressing different deletion mutants of TcdE were monitored. Additionally, site-directed mutagenesis was performed to investigate a putative dual-start-motif which might account for TcdE protein of different length and function, as it is known for holins. The expression of full length TcdE and deletion mutants lacking either the N- or C-terminus or both in E. coli resulted in inhibition of bacterial growth. The expression pattern of wild type TcdE reflects a 19 kDa and a 16 kDa protein. Mutant TcdE that exclusively codes for full length protein (19 kDa) by lacking an alternative start codon mediates a bacteriostatic effect. In contrast, truncated TcdE (16 kDa) induces cell lysis, as shown by a decrease in culture turbidity. Mutations of the alternative ribosome binding site within the 5’-region of the coding sequence shifted the ratio of full length to truncated protein from 1:10 to 1:1, however, without affecting bacterial cell lysis. The observed phenomenon of bacteria lysis was transient and a recovery of the optical density was observed 5 h after induction. This reversal in growth inhibition was due to functional inactivation of the tcdE DNA by base insertions which can be construed as a bacterial regulation against TcdE-induced toxicity. Thus, expression of the 16 kDa protein reflects a stronger selection force and a more potent impact of the truncated TcdE. Interestingly, the kinetic of recovery of bacterial growth depended on the amount of full length TcdE (19 kDa). These data verify a dual-start-motif regulating the ratio of full length (19 kDa) to truncated (16 kDa) TcdE. Truncated TcdE is proposed to be the active bacteriolytic protein. Knocking out the tcdE gene in C. difficile may help to understand the role of

98

100 Structural requirements for the regulation of pertussis-toxin-insensitive signaling of human CC chemokine receptors CCR2a and b Schuhholz J. (1), Pfreimer M. (1), Arndt P. (1), Gierschik P. (1), Moepps B. (1) Chemokines constitute a growing superfamily of small cytokines involved in regulating a wide array of leukocyte functions, including chemotaxis, adhesion, and transendothelial migration. Transmembrane signaling of chemokines is mediated by chemokine receptors, members of the G protein-coupled receptor (GPCR) family, which are coupled to pertussis- toxin-sensitive and, in certain cases, also pertussis-toxininsensitive heterotrimeric G protein(s). We have previously reported that stimulation with CCL2 of the two human CC chemokine receptors CCR2a and CCR2b transiently expressed in COS-7 cells resulted in enhanced inositol phosphate formation and in serum response factor (SRF)-dependent transcriptional activation of a luciferase reporter gene. In addition, we showed that CCL2-dependent induction of SRF activity was specifically mediated by Gαq and Gα14, but not by Gα11 and Gα16 in these cells. Interestingly, Gαq and Gα11 differ in only 11 % of their amino acid sequences and the proteins are completely identical in their carboxyl-terminal ends. The amino- and carboxyl-terminal-most portions of the Gα proteins are important for the Gα/receptor interaction. On the CCR2 receptor side, we identified a juxtamembrane octapeptide representing the amino-terminal portion of a putative 'eighth helix', which might be critically involved in Gαq protein activation and/or intracellular trafficking of the receptor polypeptides. To delineate the structural requirements for Gαq/CCR2 receptor interaction, we constructed chimeric Gαq/11 subunits and CCR2b receptor mutants, the latter carrying alanine replacements of specific amino acids within the octapeptide. The wild-type and chimeric Gα subunits were functionally characterized by determining their ability to activate SRF in COS-7 cells expressing wild-type or mutant CCR2b receptor.

26

The results showed, that (i) amino acids of the helical and/or GTPase domains, but not of the amino-terminal portion of Gaq are important for CCR2 receptor activation and (ii) that three amino acids within the putative 'eighth helix' of CCR2 are specifically required for the CCL2-and CCR2-dependent, Gaq-mediated induction of SRF. 1. Institute for Pharmacology and Toxicology,University of Ulm Medical Center, Ulm, Germany

101 Role of Galphaq-specific RhoGEFs in the activation of Rho GTPases by the human CC chemokine receptors CCR2a and b Pfreimer M. (1), Schuhholz J. (1), Wieland T. (2), Gierschik P. (1), Moepps B. (1) Chemokines constitute a growing superfamily of small cytokines involved in regulating a wide array of leukocyte functions, including chemotaxis, adhesion, and transendothelial migration. Transmembrane signaling of chemokines is mediated by chemokine receptors, members of the G-protein-coupled receptor (GPCR) family, which are coupled to pertussis- toxin-sensitive and in certain cases, also to pertussis-toxininsensitive G protein(s). As previously reported, expression of the human CC chemokine receptors CCR2a and CCR2b enabled COS-7 cells to respond to CCL2 with an increase in inositol phosphate formation, which was mediated in part by G proteins of the Gq subfamily. It is well known that activation of Gq proteins by cell surface receptors is linked to activation of the RhoGTPase RhoA, which in turn increases the activity of the transcription factor serum response factor (SRF). The marked activation of Gq proteins by CCR2a and CCR2b prompted us to investigate the roles of the Gαq-regulated and Gαq/12/13-regulated RhoGEFs p63RhoGEF and LARG, respectively, in CCR2a- and CCR2b-mediated signaling. To this end, we expressed the RhoGEFs alone, together with Gαq and/or together with CCR2a or CCR2b in COS-7 cells and analyzed receptorinduced SRF activity. While coexpression of both RhoGEFs with Gαq in the presence of the CCR2 receptors synergistically enhanced SRF activity, the two RhoGEFs displayed differences in their ability to induce SRF activity in the absence of the receptors. Thus, expression of p63RhoGEF and Gαq led to a synergistical enhancement of SRF activity even in the absence of the receptors, while expression of LARG together with Gaq had little effect in this respect. To identify the structural elements of LARG required for mediating the stimulatory effect of receptors on SRF activity, several mutants of LARG were produced and functionally characterized. One mutant lacks the amino-terminally located PDZ domain, a second was composed of the PDZ domain alone, and a third contained the PDZ and RGS domains of LARG. The LARG mutants were expressed alone or together with Gaq and the CCR2 receptors, and SRF activity was measured. While coexpression of the PDZ domain or the PDZ-RGS fragment had no effect on CCR2-receptor-mediated SRF-activation, expression of the mutant lacking the PDZ domain had an inhibitory effect. The latter finding indicates that CCR2-induced activation of LARG depends on the presence of the LARG PDZ domain and implies that a direct interaction occurs between the receptors and the RhoGEF. 1. Institute of Pharmacology and Toxicology, University of Ulm Medical Center, Ulm, Germany, 2. Department of Pharmacology and Toxicology, Faculty of Clinical Medicine, University of Heidelberg, Mannheim, Germany.

102 Merlin-1 and Ezrin(T567D) differentially influence the inhibitory effect of RhoGDIalpha in Dbs- or mDia1-mediated activation of serum response factor Su Z. (1), Moepps B. (1), Langer T. (1), Gierschik P. (1) Small GTP-binding proteins of the Rho family play many important roles in regulating intracellular functions. For example, several Rho GTPases are able to activate the transcription factor serum response factor (SRF), which is mainly involved in regulating cell proliferation, morphology, and organization of the actin cytoskeleton. The activation and the subcellular distribution of Rho GTPases are regulated by Rho guaninenucleotide exchange factors (RhoGEFs) and Rho GDP dissociatioin inhibitors (RhoGDIs). The regulation of SRF also involves mammalian diaphanous-related forming_1 (mDia1), which is participating in the actin remodelling process and located downstream of activated Rho GTPases. RhoGDIs had been reported to interact with merlin, a tumor suppressor and the protein product of neurofibraomatosis type 2 gene (NF2). However the functional significance of this interaction is not understood. Like ezrin, radixin, and moesin, merlin (moesin, ezrin, and radixin-like protein) belongs to the 4.1-protein superfamily. We have previously found that merlin-1 enhanced the inhibitory effects of RhoGDIa on G-protein-coupled-receptor-mediated SRF activity and ezrin(T567D) decreased this inhibition. Ezrin(T567D) carries an phosphomimetic, activating mutation in position T567. The current study was undertaken to investigate the influences of merlin-1 and ezrin(T567D) on the effects of RhoGDIa on Dbs (the RhoGEF Dbl’s big sister) and mDia1-mediated activation of SRF in transfected COS-7 cells. Our results showed that expression of exogenous RhoGDIa caused a significant inhibition of Dbs- or mDia1-mediated SRF activation. Coexpression of Merlin-1 did not enhance the inhibitory effect of RhoGDIa on Dbs-mediated activation of SRF, unlike was the case with GPCR-mediated SRF activation. Ezrin(T567D) enhanced Dbs-mediated activation of SRF, but there was no obvious change on the inhibitory effect of RhoGDIa on this function. Furthermore, the coexpression of merlin-1 caused an increase, but ezrin(T567D) caused a reduced the inhibitory effect of RhoGDIa on mDia1-mediated activation of SRF. The results suggest that the proteins of 4.1-protein superfamily such as merlin-1 and ezrin(T567D) differentially influence the inhibitory effect of RhoGDIa on Dbs- or mDia1-mediated activation of SRF. 1. Institut fuer Pharmakologie und Toxikologie, Universitaetsklinikum, Ulm

103 Purinergic inhibition of β-adrenergic induced cAMP-elevation in vascular smooth muscle cells von Hayn K. (1), Werthmann R.C. (1), Nikolaev V.O. (1), Lohse M.J. (1), Bünemann M. (1) Vascular smooth muscle tone is an important factor for the regulation of blood pressure. Two counteracting physiological regulators of smooth muscle tone are cyclic AMP and Ca2+, released for instance after activation of Gq-coupled purinergic-P2 receptors. While stimulation of Gq-coupled purinergic receptors contracts, cAMP relaxes vascular smooth muscle cells. In addition intracellular released Ca2+ was reported to regulate synthesis of cAMP by activating or inhibiting adenylylcylases or phosphodiesterases. To assess the influence of purinergic stimulation on cAMP-concentrations in single living vascular smooth muscle cells, we used the fluorescence resonance energy transfer (FRET) – based sensor Epac1-camps (Nikolaev et al, JBC, 2004). Ca2+-release due to purinergic stimulation was measured by monitoring changes in ratiometric Fura2-fluorescence. Submaximal cAMP production was induced by stimulating cells with isoproterenol. Upon additional stimulation of Gq-coupled purinergic receptors by UDP of ATP, which evoke intracellular Ca2+-release, a decrease of cAMP-concentration was monitored. We aimed to test, which mechanism leads to the cAMP decrease. Inhibition of PDE1, the only phosphodiesterase known to be regulated by Ca2+, and therefore the only one influenceable due to purinergic stimulation, with 8-MM-IBMX, did not alter the effect. Overexpression of adenylyl cyclase 4, which is not regulated by Ca2+, antagonized the purinergic-evoked cAMP-decrease. These results suggested, that not PDEs but adenylyl cyclases, inhibited by activation of the Gq-pathway, may be responsible for the effect. Adenylyl cylcases 5 and 6 are the only cyclases reported to be inhibited by Ca2+. However, overexpression of adenylylcylase 6 did not significantly strengthen the UDPevoked decrease of cAMP-concentration compared to control cells. Silencing adenylyl cyclases 5 and 6 by si-RNAs significantly reduced the decrease of cAMP-levles upon stimulation of purinergic receptors compared to control cells transfected with nontargeting-siRNA. We suggest, that purinergic stimulation of vascular smooth muscle cells, by evoking intracellular Ca2+-release, leads to an inhibition of adenylyl cyclases 5 and 6, resulting in an attenuation of adenylyl-cyclase-induced elevation of cAMP. 1. Dept. of Pharmacology and Toxicology, University of Würzburg, Würzburg, Germany

104 Synthetic ligands to study GAF-dependent activation of phosphodiesterases 2 and 5 Jäger R. (1), Russwurm C. (1), Koesling D. (1), Russwurm M. (1) Cyclic nucleotides play an important role in intracellular signalling and mediate physiological processes like smooth muscle relaxation and inhibition of platelet aggregation. Degradation by phosphodiesterases (PDEs) has an important impact on cyclic nucleotide signals. Currently 11 PDE families have been identified differing in substrate selectivity and regulation. All eleven PDEs share a common domain structure with a conserved C-terminal catalytic domain and an N-terminal regulatory domain which differs considerably between the families. The N-terminal regulatory domains of five PDEs contain so called GAF domains, cGMP or cAMP binding domains regulating catalytic activity. Here, we focussed on two of these GAF-domain containing PDE families, PDE2 and PDE5. PDE2 is a dual substrate PDE, i.e. hydrolyzes cAMP and cGMP, whereas PDE5 is highly selective for cGMP. PDE2 has been initially described as "cGMP-stimulated PDE", because cGMP stimulated cAMP hydrolysis, later this cGMP effect was shown to be GAF domain-mediated. In PDE5, the stimulatory effect on catalytic activity has been overlooked and the enzyme has been described as "cGMPbinding cGMP-specific PDE". Partly, this was due to the instability of the enzyme that adopted the stimulated conformation upon purification and storage, but difficulties to analyse an allosteric stimulatory effect of an enzyme's substrate further delayed the recognition of this important regulation. Here we set out to identify ligands that specifically bind to the GAF domains without strong interference with the catalytic domain. We used fusion proteins of the GAF domains and two fluorescent proteins and analyzed changes in fluorescence resonance energy transfer caused by conformational changes upon binding of a range of different cyclic nucleotide analogs. Thereby, we identified several analogs with sub-micromolar affinity, some of which specifically bound to either the GAF domains of PDE2 or PDE5. Next, we analyzed the effects of these GAF ligands on the PDE2 and PDE5 holoenzymes and identified activators of PDE2 and PDE5 leading to 30 and 20-fold stimulation of cGMP hydrolysis, respectively. 1. Institut für Pharmakologie und Toxikologie, Ruhr-Universität Bochum, Germany

105 Cytidylyl cyclase activity and uridylyl cyclase activity of bacterial adenylyl cyclase toxins Göttle M. (1), Geduhn J. (2), König B. (2), Seifert R. (3) Bacillus anthracis and Bordetella pertussis, the causative agents of anthrax disease and whooping cough, respectively, secrete the adenylyl cyclase (AC) exotoxins edema factor (EF) and CyaA. The toxins generate supraphysiological cAMP levels in host immune cells, weakening immune response and, thereby, promoting the pathogenesis of the infections [1, 2]. We have developed adenylyl cyclase toxin inhibitors with high potency and selectivity with respect to mammalian adenylyl cyclase isoforms. Moreover, we found that besides the ability to form cAMP from ATP, both AC toxins also possess cytidylyl cyclase activity, resulting in the conversion of CTP to cCMP. AC toxin enzyme activity on CTP, UTP and ITP is characterized to elucidate the role of uncommon cyclic nucleotides in suppression of human immune responses by bacterial virulence factors. In our cytidylyl cyclase (CC) activity assay, the radioactively labeled substrate [a32 P]CTP is converted to [32P]cCMP which is quantified by liquid scintillation [3]. Upon incubation of the AC toxins with [α-32P]CTP, [32P]cCMP is produced in a linear manner over time. As [32P]cCMP production depends on the endogenous AC toxin activator protein calmodulin, as physiological pH promotes the reaction and as standard AC inhibitors show high potency on EF and CyaA, catalysis results from specific CC activity. Michaelis-Menten kinetics of EF CC activity yielded Km = 13 ± 3 µM and Vmax = 8 ± 1 s1 . A signal was also obtained when UTP was applied, indicative for [32P]cUMP production. [32P]UTP yielded Km = 135 ± 23 µM and Vmax = 1.3 ± 0.1 s-1. As an alternative approach, the NTP consumption and cNMP formation were monitored by

27

HPLC [4]. CTP (100 µM) were converted to cCMP by 20 nM EF within 1 h. Not only CTP, but also UTP and ITP were converted to the corresponding cNMPs. In the presence of heat-inactivated enzyme, no cNMP was formed. The identity of cCMP, cUMP and cIMP was confirmed by mass spectrometry. Based on these findings and the fact that cCMP inhibits host immune responses, we propose that bacterial ACs generate multiple cNMPs, increasing infection severity. The molecular targets and the pathophysiological role of uncommon cyclic nucleotides have to be investigated in order to provide new starting points for the development of inhibitors against bacterial cyclase toxins. [1] Stubbs MT (2002); TIPS 23:539-541. [2] Leppla SH (1982); Biochemistry 79:3162-3166. [3] Alvarez R (1990); Anal Biochem 187:98-103. [4] Witters E (1997); J Chromatogr B 694:55-63. 1. Dept. of Pharmacology and Toxicology, University of Regensburg, Regensburg, Germany, 2. Dept. of Organic Chemistry, University of Regensburg, Regensburg, Germany, 3. Dept. of Pharmacology, Hannover Medical School, Hannover, Germany

106 Molecular analysis of the interaction of anthrax adenylyl cyclase toxin, edema factor, with 2',3'-O-bis-(N-(methyl)anthraniloyl)-substituted purin and pyrimidine nucleotides Taha H. (1), Geduhn J. (2), Seifert R. (3) Bacillus anthracis causes anthrax disease and exerts its deleterious effects by the release of three exotoxins, i.e. lethal factor, protective antigen and edema factor (EF), a highly active calmodulin-dependent adenylyl cyclase (AC). However, conventional antibiotic treatment is ineffective against either toxemia or antibiotic-resistant strains. Thus, more effective drugs for anthrax treatment are needed. Our previous studies showed that EF is differentially inhibited by various purine and pyrimidine nucleotides modified with N-methylanthraniloyl (MANT)- or anthraniloyl (ANT) groups at the 2’(3’)-Oribosyl position, with the unique preference for the base cytosine. MANT-CTP was the most potent EF inhibitor (Ki, 100 nM) among 16 compounds studied. Here, we examined the interaction of EF with 2’,3’-O-Bis-(M)ANT-substituted nucleotides. Bis-MANT-ATP and Bis-MANT-CTP were the most potent EF inhibitors among nucleotides studied (Ki, 210 nM for both nucleotides). Bis-MANT-nucleotides inhibited EF competitively. Activation of EF by calmodulin resulted in effective fluorescence resonance energy transfer (FRET) from tryptophan and tyrosine residues located in the vicinity of the catalytic site to Bis-MANT-ATP, but FRET to Bis-MANT-CTP was only small. Mutagenesis studies revealed that F586 is crucial for FRET to Bis-MANT-ATP and BisMANT-CTP and that the mutations N583Q, K353A and K353R differentially alter the inhibitory potencies of Bis-MANT-ATP and Bis-MANT-CTP. We conclude that the nucleotide binding site of EF is spacious and readily accommodates bulky Bis-(M)ANTsubstituted purine and pyrimidine nucleotides. These data provide a solid basis for future structure/activity relationship studies aiming at the development of potent EF inhibitors with high selectivity relative to mammalian ACs. 1. Dept. Pharmacology, University of Regensburg, Regensburg, Germany, 2. Institute of Organic chemistry, University of Regensburg, Regensburg, Germany, 3. Institute of pharmacology, Hannover medical school, Hannover, Germany

107 Thrombin-mediated increase in [Ca2+] causes a transient decrease in cAMP levels in endothelial cells Werthmann R.C. (1), von Hayn K. (1), Nikolaev V.O. (1), Lohse M.J. (1), Bünemann M. (1) The passage of macromolecules and fluid between blood and interstitial tissue is controlled by the barrier function of endothelial cells. The processes that are involved in regulation of the barrier mostly rely on Ca2+ and cAMP signals. The inflammatory agonist thrombin is known to enhance permeability of endothelial cells, whereas a rise in membranous cAMP is discussed to enhance barrier function. As Ca2+ and cAMP signals exhibit cross talks, we investigated the effect of thrombin on elevated cAMP levels in endothelial cells. The FRET-based cAMP sensor Epac1-camps was utilised to monitor changes in cAMP in real time in living cells. Transfected Human Umbilical Vein Endothelial Cells (HUVECs) were stimulated with 10nM isoproterenol to elevate cAMP levels. Additional stimulation with thrombin (10U/ml) lead to a transient increase in FRET ratio, that is consistent with a transient decrease in [cAMP]. This effect was not PTX sensitive, but rather Ca2+ dependent. Ca2+ is known to directly inhibit adenylylcyclases (Adcy) 5 and 6, and Adcy6 is expressed in endothelial cells. Downregulation of Adcy5 and Adcy6 in HUVECs completely abolished inhibitory effects of thrombin on cAMP. As Epac1-camps represents a monomolecular sensor for cAMP, it allows to determine concentrations of cAMP dependent on changes in ratiometric FRET. Therefore, we performed an in vitro calibration of Epac1-camps based on [cAMP] and ∆FRET. FRET experiments in living HUVECs revealed that ratiometric FRET was stable under basal conditions. Upon stimulation of cells with 1µM isoproterenol the FRET ratio reached a stable minimum with kinetics indicative of sensor saturation. Therefore we assumed that [cAMP] is less than 10nM under resting conditions and elevated to more than 100µM at maximal concentrations of isoproterenol, according to the in vitro calibration of Epac1camps. Based on this in vivo calibration we calculated thrombin-mediated changes in [cAMP]. Stimulation of HUVECs with 10nM isoproterenol resulted in an increase of [cAMP] to 3.2±0.5µM and thrombin caused a subsequent decrease in [cAMP] by 1.0±0.2µM. Thus, we conclude that thrombin, via increasing [Ca2+]i, mediates a robust inhibition of prestimulated Adcy5 and 6, resulting in a decrease in [cAMP]. 1. Dept. of Pharmacology and Toxicology, Julius-Maximilians-University, Würzburg, Germany

108 Influence of oxidative stress on the structure and function of the cGMP-dependent protein kinase type I Gnügge R. (1), Jessen S. (1), Valtcheva N. (1), Hillenbrand M. (1), Gerling A. (1), Feil S. (1), Ruth P. (2), Feil R. (1) The cGMP-dependent protein kinase type I (cGKI) is an important mediator of cGMP signaling. In mammals one gene encodes for two cGKI isoforms (termed α and β) that

differ only in their N-terminal ~100 amino acids. Both isoforms are homodimers. Despite their high similarity, cGKIα and cGKIβ display different susceptibilities to cGMP activation and are thought to interact with distinct anchoring and substrate proteins. However, the specific cellular functions of each cGKI isozyme are not well understood. Recent findings indicate a novel isoform-specific role of cGKIα in redox sensing. It is assumed to be mediated by a reactive cysteine residue in the N-terminal region of cGKIα, which results in the formation of an intermolecular disulfide bond between the protomers of the homodimeric enzyme. Notably, cGKIβ lacks this cysteine residue. Here, we have analyzed the effect of oxidative stress induced by micromolar concentrations of hydrogen peroxide (H2O2) on the structure and function of cGKI in vitro and in intact cells. Transgenic cell models expressing or lacking cGKI isoforms were used to control the specificity of the observed H2O2 effects. In both mouse embryonic fibroblasts and primary vascular smooth muscle cells, H2O2 induced the formation of an intermolecular disulfide bond in cGKIα but not in cGKIβ. Interestingly, incubation of intact cells with H2O2 led to a slightly increased phosphorylation of the cGKI substrate, vasodilator-stimulated phosphoprotein (VASP). In contrast, H2O2 attenuated the 8-BrcGMP-induced VASP phosphorylation. In line with these findings, preliminary data indicate that H2O2 increases the basal (cGMP-independent) activity of purified cGKIα, whereas it inhibits cGMP activation of the enzyme. Taken together, these results show that oxidative stress influences specifically the structure and activity of cGKIα but not cGKIβ. 1. Interfakultäres Institut für Biochemie, Universität Tübingen, Germany, 2. Pharmazeutisches Institut, Universität Tübingen, Germany

109 In vivo analysis of cGMP/cGKI signaling in learning and cognition Gerling A. (1), Becker C. (1), Schönegge A. (1), Feil S. (1), Weindl K. (2), Hölter S.M. (2), Wurst W. (2), Feil R. (1) The cGMP-dependent protein kinase type I (cGKI) is expressed in various regions of the nervous system indicating a functional role of this enzyme in neuronal processes. In vitro studies have shown that cGMP signaling via cGKI contributes to long-term potentiation (LTP) of synaptic transmission. However, convincing in vivo proof for an important role of this protein kinase in learning and memory is still missing. Since conventional cGKI null mutants suffer from severe smooth muscle dysfunction and have a limited lifespan of only 5-6 weeks, they are not an appropriate model to study behavioral phenotypes. Mouse mutants that lack the protein specifically in the hippocampus are healthy, but showed no defects in spatial learning. The absence of a behavioral phenotype in hippocampus-specific cGKI mutants could be due to the presence of cGKI in other brain regions. Here, we have investigated two new mouse models lacking cGKI in the entire nervous system: neuron-specific cGKI knockout mice generated via Cre/lox technology, and cGKI-smooth muscle rescue mice in which cGKI function has been restored only in smooth muscle cells. These mouse mutants were tested for spatial learning in a discriminatory watermaze task. No apparent differences between mutant and control mice were observed, thus, providing further in vivo evidence that cGKI is not centrally involved in spatial learning. However, tests addressing social discrimination and object recognition revealed a significant role of cGKI in other cognitive functions. 1. Interfakultäres Institut für Biochemie, Universität Tübingen, Germany, 2. Institute of Developmental Genetics, GSF - National Research Centre for Environment and Health, Neuherberg, Germany

110 Crystal structure of MANT-ITP – membranous adenylyl cyclase complex Hübner M. (1), Mou T.C. (2), Geduhn J. (3), König B. (3), Seifert R. (4) Membrane-bound mammalian adenylyl cyclases (mAC) catalyze the formation of the second messenger cAMP from ATP. mACs play a major role in G-protein coupled receptor signaling and the search for potent and selective AC subtype V inhibitors could lead to the development of novel drugs for cardiovascular disorders [1,2]. 2´(3´)Methylanthraniloyl (MANT) - nucleotides are potent mAC inhibitors that bind to the catalytic center of the enzyme and exhibit Ki values in the nanomolar range [3]. In this study, x-ray crystallography was used to investigate the interaction of mAC with MANTinosine 5’-triphosphate (MANT-ITP), the most potent mAC inhibitor known so far (Ki 0.7nM). The catalytic domains C1a of mAC isoform V and C2a of mAC isoform II were purified and activated by forskolin and Gsα-GTPγS. The crystal structure of this complex, bound to MANT-ITP:Mn2+, was determined and compared to the published crystal structure of mAC with MANT-GTP, which exhibits more than 20-fold lower potency for the AC complex than MANT-ITP. Superimposing both structures revealed strong similarities of the inhibitors in the catalytic center. There are subtle differences in binding in the purine binding pocket, as well as in the MANT binding pocket, but the phosphate binding and the position of the ribosyl group are not altered. It is possible that MANT-ITP exhibits a higher potency due to less binding constraint of the hypoxanthine ring in the purine binding pocket. Interestingly, those small differences in binding can lead to large differences in fluorescence properties of both nucleotides. It was observed that MANT-ITP exhibits a very large basal FRET and direct fluorescence signal. These data show that MANT-ITP promotes C1/C2 assembly already under basal conditions. In conclusion, we have resolved the structure of mAC with the most potent enzyme inhibitor known so far, providing a solid basis for future inhibitor domain. [1] Yan et al., Cell 2007, 130, 247-58. [2] Rottländer et al., J.Pharmacol.Exp.Ther. 2007, 321, 608-15. [3] Gille et al., J. Biol. Chem. 2003, 278, 12672-12679. [4] Mou et al., J. Biol. Chem. 2005, 280, 7253-61. (M.H. is the recipient of a predoctoral scholarship of the Bayerische Eliteförderung) 1. Department of Pharmacology and Toxicology, University of Regensburg, Regensburg, Germany, 2. Center for Biomolecular Structure and Dynamics, University of Montana, Missoula, USA, 3. Institute of Organic Chemistry, University of Regensburg, Regensburg, Germany, 4. Institute of Pharmacology, Medical School Hannover, Hannover, Germany

28

111 Site-directed mutagenesis of the putative cAMP binding site of human placental S-adenosyl-L-homocysteine hydrolase Hermes M. (1), Klein K. (1), Schall C. (2), Riehle R. (1), Piesch C. (1), Oßwald H. (1), Kloor D. (1) S-Adenosyl-L-homocysteine (AdoHcy) hydrolase (AdoHcyase) catalyses the hydrolysis of AdoHcy. AdoHcyase activity is stimulated by cAMP suggesting that cAMP keeps AdoHcy tissue content at a low level, which corresponds to a high cellular methylation capacity. To gain further insight into the mechanism of cAMP binding and regulation of AdoHcyase activity, we performed site-directed mutagenesis studies. The cAMP binding site residue W310, recently identified by photoaffinity labeling, was mutated to glycine (W310G), alanine (W310A), leucine (W310L), or phenylalanine (W310F). Wild-type and mutant AdoHcyase were overexpressed in E. coli JM109 and HEK-293 cells. AdoHcyase protein expression was detected by immuno-blotting and activity was determined photometrically at 292 nm. For heat inactivation experiments, recombinant protein was incubated for 15 min at various temperatures ranging from 22-61°C. Saturation binding experiments were performed with AdoHcyase, purified by chromatographical techniques, and [3H]-cAMP or [3H]-adenosine at a final ligand concentration of 1-250 nM. W310A, W310G, and W310L mutants showed a dramatic reduction of AdoHcyase protein expression and, correspondingly, a decreased enzymatic activity in crude cell extracts of E. coli and HEK-293 cells cultured at 37°C, whereas the W310F mutant resulted in a slightly decreased protein expression and enzymatic activity. Kinetics of thermal inactivation showed that impaired protein expression and activity is due to thermo-sensitivity of mutant enzymes. Thus, further experiments were performed with wild-type, W310G, and W310F AdoHcyase purified from E. coli cultured at an appropriate temperature (25°C). Enzyme kinetics revealed that purified W310F and W310G AdoHcyase retained a substantial amount of the overall catalytic activity (66% respectively 75% of the wild-type). While both, W310F and W310G mutations, reduced cAMP binding by 25% and 40%, respectively, adenosine binding was unchanged in W310F mutant protein and 1.5-fold increased in the W310G mutant AdoHcyase. In conclusion, our results suggest that the large rigid aromatic group of W310, which tends to orient towards the interior of the folded protein molecule, is important for protein folding and stability and cAMP and adenosine binding. 1. Institut für Experimentelle und Klinische Pharmakologie und Toxikologie, EberhardKarls Universität Tübingen, Germany, 2. Interfakultäres Institut für Biochemie, EberhardKarls Universität Tübingen, Germany

112 Hydrophilic interaction liquid chromatography coupled to tandem mass spectrometry as method for sensitive and specific quantitation of nucleotides (nucleotidomics) Kaever V. (1), Rauch K. (1), Barchanski A. (1), Burhenne H. (1), Seifert R. (1) During the last years high performance liquid chromatography (LC) coupled to tandem mass spectrometry (MS/MS) has become the preferred method for sensitive and specific determination of drugs and endogenous metabolites. However, reverse phase LC materials traditionally applied for LC-MS/MS analysis of polar, low molecular weight analytes are not well suited because of their low retention capacities for such substances. Therefore, we have established new analytical methods based on hydrophilic interaction liquid chromatography (HILIC) coupled to MS/MS. As one example, simultaneous quantitation of various cyclic nucleotides is presented. The experimental conditions were as follows: LC 1100 (Agilent); API 3200QTrap (Applied Biosystems); ZIC-HILIC, 2.1 x 10 cm, 3.5 µm (SeQuant); Solvent A: ACN/H2O/100 mM NH4Oac, pH 5.8 (90/5/5); Solvent B: ACN/H2O/100 mM NH4Oac, pH 5.8 (50/45/5); A/B = 70/30; flow 0.2 ml/min; negative ion mode (MRM). All cyclic nucleotides (cAMP, cGMP, cCMP, cUMP, cIMP, cXMP, cTMP) could be separated within a total analysis time of 10 min. Detection limits of about 30 fmol could be achieved for all analytes. This method can be used as non-radioactive alternative for enzymatic nucleotidyl cyclase assays as well as quantitation of various nucleotides in intact cells and tissues. In summary, HILIC-MS/MS represents a sophisticated method for the simultaneous quantitation of different cyclic nucleotides. Further enhancement of the assay sensitivity (comparable to immunoassay) may be achieved by additional derivatization steps in order to gain optimal chromatographic and MS/MS ionization conditions. 1. Inst. of Pharmacology, Hannover Medical School, Hannover, Germany

113 Pharmacological characterization of receptor guanylyl cyclase reporter cell lines Wunder F. (1), Milde M. (1) Receptor guanylyl cyclases play essential roles in cardiovascular (patho-)physiology, reproduction biology, cell proliferation, bone growth and sensory signal transduction and, therefore, are important pharmacological targets. Known receptor ligands include natriuretic peptides and members of the guanylin peptide family. We report here the generation and pharmacological characterization of three receptor guanylyl cyclase reporter cell lines. Plasmid constructs encoding the natriuretic peptide receptors GC-A and GC-B, and the guanylin receptor GC-C, were stably transfected in a parental reporter cell line expressing the cyclic nucleotide-gated (CNG) cation channel CNGA2, acting as the biosensor for intracellular cGMP. In our reporter cell lines, receptor guanylyl cyclase activity can be monitored in real-time via aequorin luminescence stimulated by calcium influx through the CNG channel. By using different natural as well as synthetic receptor ligands, we could show that our reporter assay monitors receptor activation with very high sensitivity. On the GC-A reporter cell line, the natriuretic peptides ANP and BNP stimulated luminescence signals with very high potency, whereas CNP was less potent. In contrast, on the GC-B receptor cell line CNP showed higher potency compared to ANP and BNP. E. coli heat-stable enterotoxin, uroguanylin and guanylin where without activity on both the GC-A and GC-B reporter cell lines, however, activated luminescence signals on the GC-C receptor cell line. ANP receptor antagonists were also tested on our GC-A reporter cell line and, unexpectedly, were characterized as partial agonists. The results imply that our novel guanylyl cyclase reporter cell lines are well suited for the characterization of receptor agonists and antagonists and may also be used for the identification of natural ligands of guanylyl cyclase orphan receptors.

1. Lead Discovery Wuppertal, Bayer HealthCare AG, Wuppertal, Germany

114 Subcellular distribution of cGMP signalling proteins in VSMCs of IRAG KO mice Hieke B.M. (1) Stimulation of cGMP signalling cascade in smooth muscle results in decreased intracellular Ca2+ levels and thus muscle relaxation. Inositoltrisphosphate Receptor 1 (IP3R1) – a Ca2+ release channel located in the membrane of sarcoplasmatic reticulum (SR) - is inhibited by the phosphorylation of IP3R1 associated cGMP kinase dependent substrate (IRAG) via cGMPkinase1β (cGK1β). On the other hand, Ca2+ is pumped back into SR by the Sarco/Endoplasmatic Reticulum Ca2+ ATPase (SERCA), which is inhibited by Phospholamban (PLB). Phosphorylation of PLB by cGK1 suppresses its inhibitory effect on SERCA. To characterize the subcellular distribution of these cGMP signalling proteins upon deletion of IRAG, aortic vascular smooth muscle cells (VSMCs) of IRAG KO mice were immunohistochemically analyzed and compared to VSMCs of littermate wild type animals. Furthermore we investigated the effects of treatment with 8Br-cGMP on the distribution of these proteins in IRAG KO VSMCs versus WT VSMCs. 1. Lehrstuhl für Pharmakologie und Toxikologie, Universität Regensburg, Deutschland

115 PAM (protein associated with myc) regulates maintenance of thermal hyperalgesia through DLK1 and p38 MAPK Holland S. (1), Coste O. (1), Becker W. (1), Pierre S.C. (1), Geisslinger G. (1), Scholich K. (1) Protein associated with Myc (PAM) is an E3 ubiquitin ligase that negatively regulates neuronal growth. Although PAM has been shown to also modulate synaptic strength, the underlying mechanisms are largely unknown. We generated conditional PAM knockout mice that do not express functional PAM in nociceptive and thermoreceptive neurons of the dorsal root ganglia (SNS-PAM-/-). Motor abilities, basal pain thresholds and nociceptive behavior in the acute phase of the formalin test, a model for acute and inflammatory pain in the paw, were not altered in these animals. However, formalininduced thermal hyperalgesia was prolonged in the SNS-PAM-/- mice. p38MAPK, known to maintain thermal hyperalgesia, is constitutively activated in dorsal root ganglia (DRG) of SNS-PAM-/- mice. Also its activator DLK-1, a known substrate for PAM, and the transient receptor potential vanilloid 1 (TRPV1), whose expression is regulated by p38MAPK is increased expressed in DRGs of SNS-PAM-/- mice. Intrathecal application of the p38MAPK-inhibitor SB203580 eliminated the prolonged hyperalgesia in the knockout mice suggesting that PAM located in peripheral sensoric neurons regulates duration of hyperalgesia through the DLK1/p38MAPK pathway. Besides the p38MAPK pathway both the Jnk- and the Erk2-MAPKinases are highly activated in DRGs of PAMdeficient mice. Thus it seems that PAM is involved in the regulation of these three MAPK-pathways. 1. Dept. of clin. Pharmacology, Johann Wolfgang Goethe-Universität, Frankfurt, Germany

116 Identification of GSKIP as an AKAP Skroblin P. (1), Hundsrucker C. (1), Christian F. (1), Zenn H.M. (2), Herberg F.W. (2), Rosenthal W. (1), Klussmann E. (1) A kinase anchoring proteins (AKAPs) comprise a class of scaffolding proteins that target protein kinase A (PKA) and other signalling proteins to cellular compartments and thereby confine the activities of the associated proteins to defined regions within cells. The compartmentalization by AKAPs facilitates local integration of cellular signalling and is key to regulation of many cellular processes including vasopressin-mediated water reabsorption and regulation of cardiomycyte contractility. AKAPs bind PKA directly, facilitating phosphorylation of PKA substrates in close proximity. The interaction is mediated by the dimerization and docking (DD) domain of regulatory subunits of PKA (RI or RII) and the RII-binding domain of AKAPs. Analysis of the interactions of the DD domain with RII-binding domains of several AKAPs yielded an AKAP signature motif. A bioinformatics and peptide array screening approach based on this motif identified GSKIP (GSK3β interacting protein) as an AKAP. GSKIP directly interacts with PKA and glycogen synthase kinase 3β (GSK3β). The RII-binding domain of GSKIP is located in the domain of unknown function 727 (DUF727). This domain is evolutionarily conserved and confers an AKAP function to the orthologues of GSKIP. GSKIP is widely expressed in mammals. Its subcellular localisation and association with PKA substrates are currently being investigated. 1. Leibniz Institute for Molecular Pharmacology, Berlin, Germany, 2. Institute of Biology, University of Kassel, Kassel, Germany

117 Translocation of the mitogen-activated triple kinase DLK to the nucleus Wallbach M. (1), Kruegel J. (2), Miosge N. (2), Oetjen E. (1) Mitogen-activated kinases (MAPK) enable the cell to adapt to environmental demands thereby regulating a variety of cellular functions, including cell proliferation, differentiation and stress response. Thus, these kinases are potential novel drug targets. MAPK transmit extracellular signals via sequential activation of a MAP3K, MAP2K and finally a MAPK. Our previous studies showed that TNFα-induced inhibition of the transcriptional activity of CREB and its co-activator CBP was at least in part dependent on the activity of the dual leucine zipper bearing kinase (DLK). In the present study the subcellular localization of DLK was investigated. Immunocytochemical analysis indicated that treatment of the beta-cell line HIT with TNFα lead to a nuclear translocation of DLK in a time-dependent manner. These findings were confirmed with the help of electron microscopy applying a gold-labeled antibody against DLK. The subcellular localization of over expressed flag epitope tagged DLK and its kinase dead mutant was detected by a gold-labeled antibody against the flag epitope. Upon TNFα-

29

treatment both over expressed proteins translocated to the nucleus, thereby indicating that the enzymatic activity of DLK is dispensable for its translocation. The underlying mechanisms were investigated: By in silico analysis a bipartite nuclear localization signal (NLS) within DLK from amino acids 186 to 200 was identified. The mutation of the basic amino acids to alanine within the individual parts of the NLS was sufficient to diminish the TNFα-induced nuclear translocation of the over expressed DLK mutants as detected by immunocytochemistry. In transient transfections the over expression of the individual NLS-DLK mutants no longer inhibited the transcriptional activity of a luciferase reporter gene under control of four copies of the CRE of the rat somatostatin gene promoter upon membrane depolarization. Using the GAL4 system, neither the transcriptional activity of CREB nor that of its co-activator CBP was inhibited by the NLS mutants. Taken together, our data indicate that DLK contains a functional nuclear localization signal which is required for the kinase’s inhibitory effect on CRE/CREBdirected transcription. Our data thereby suggest that a MAP3K might exert at least some of its effects directly in the nucleus. Thus, the subcellular localization of MAPK might contribute to their regulation of cellular functions. 1. Dept. of Pharmacology, Georg August University, Göttingen, Germany, 2. AG Oral Biology and Tissue Regeneration, Georg August University, Göttingen

118

120 Loss of Raf-1 activity is compensated by c-Src limited CaMKII-dependent activity and apoptosis to normal renal cystogenesis Weiß D. (1), Honisch S. (1), Meloh J. (1), Beil W. (1), Fährmann M. (1) Cysts are initial structures to develop renal tubules. A limited apoptosis is discussed to be involved in renal cystogenesis. However, the precise molecular mechanism of cystogenesis, and significance of apoptosis is unclear. We found that cystogenesis depended at least on the activities of postreceptor protein kinases c-Src, CaMKII and Raf-1. Overexpression of either dominant negative mutant forms of c-Src or Raf-1 or regulation defect mutant forms of CaMKII significantly reduced cystogenesis. Overexpression of constitutively active CaMKII T286D increased apoptosis 3-fold, and restricted cystogenesis. Co-overexpression of CaMKII T286D and dominant-negative Raf-1 reduced CaMKII T286D-dependent apoptosis but totally abolished cystogenesis. Co-overexpression of CaMKII T286D together with c-Src (WT or constitutively active) and dominant-negative Raf-1, however, increased apoptosis only between 1.8- and 2fold, and resulted in normal cystogenesis. In this case, activation of MEK1/2 at S217/S221 was similar to control of non-transfected cells of normal cystogenesis. In vivo activities of either CaMKII or c-Src which were measured by each specific CFP/YFPFRET reporter in confocal live cell imaging of cysts showed that c-Src down regulated CaMKII activity. Conclusively, a c-Src-limited CaMKII activity which was accompanied by a two-fold increase of apoptosis is compensative to lost Raf-1 activity. 1. Institut für Pharmakologie, Medizinische Hochschule Hannover, Germany

Identification of novel TAK1 and p38 MAPK dependent phosphorylation sites in TAB1, a regulatory subunit of the protein kinase TAK1 Wolf A. (1), Vahlsing S. (1), Dittrich-Breiholz O. (2), Schneider H. (2), Beuerlein K. (1), Kracht M. (1) The protein kinase TAK1 and its regulatory subunit TAB1 are important upstream molecules in the pathways activated by IL-1, TNF and toll-like receptors. TAK1 is a member of the MAP3K family and thus can activate the MAP kinases p38 and JNK as well as the NFkB signaling pathway. By mobility shifts upon SDS-PAGE and by mutational studies we detected novel phosphorylation sites of TAB1 located within a Cterminal cluster of six serines, namely aa452/3 and aa456/7. As assessed by newly generated phosphorylation-site specific antibodies and by TAB1 mutants in which serines 452, 453, 456 and 457 were exchanged to alanine or glutamate, these sites are phosphorylated by catalytically active forms of TAK1 as well as by p38 in intact cells and in vitro. Functional studies suggest that these sites are not only targets of p38 (or TAK1) but also modulate activity of activated TAK1 or p38 MAPK suggesting that they are part of an autoregulatory feedback loop. To assess the biological role of TAB1 modifications in more detail we reconstituted TAB1-deficient embryonic fibroblasts with wild type and mutant TAB1 (deltaS452-457) proteins and analysed their effects on mRNA expression and secretion of a number of inflammatory response genes. Deletion of TAB1 in mice is embryonic lethal and little is known about its specific functions. Hence, our results may provide one clue to delineate unknown functions of TAB1 depending on the new phosphorylation sites. 1. Rudolf-Buchheim-Institute of Pharmacology, Justus-Liebig-University Giessen, Giessen, Germany, 2. Institute of Pharmacology, Medical School Hannover, Hannover, Germany

Sphingosine kinase-1 is a hypoxia-regulated gene that stimulates migration of human endothelial cells Schwalm S. (1,2), Döll F. (1,2), Pfeilschifter J.M. (1), Huwiler A. (2) Sphingosine kinases (SK) catalyse the production of sphingosine-1-phosphate which in turn regulates cell responses such as proliferation and migration. Here, we show that exposure of the human endothelial cell line EA.hy 926 to hypoxia stimulates a increased SK-1, but not SK-2, mRNA, protein expression and activity. This effect was due to stimulated SK-1-promoter activity which contains two putative hypoxia-inducible factorresponsive-elements (HRE). By deletion of one of the two HREs, hypoxia-induced promoter activation was abrogated. Furthermore, hypoxia upregulated the expression of HIF-1α and HIF-2α, and both contributed to SK-1 gene transcription as shown by selective depletion of HIF-1α or HIF-2α by siRNA. The hypoxia-stimulated SK-1 upregulation was functionally coupled to increased migration since the selective depletion of SK-1, but not of SK-2, by siRNAs abolished the migratory response. In summary, these data show that hypoxia upregulates SK-1 activity and results in an accelerated migratory capacity of endothelial cells. SK-1 may thus serve as an attractive therapeutic target to treat diseases associated with increased endothelial migration and angiogenesis such as cancer growth and progression. 1. pharmazentrum Frankfurt/ZAFES, Goethe-University Frankfurt am Main, Germany, 2. Institute of Pharmacology, University of Bern, Switzerland

119

122

Characterization of harmine as a potent and specific inhibitor of the protein kinase DYRK1A Göckler N. (1), Papadopoulos C. (1), Becker W. (1) DYRK1A is a dual specificity protein kinase that autophosphorylates on a conserved tyrosine residue (Tyr321) in the activation loop of the catalytic domain but phosphorylates exogenous substrates only on serine or threonine residues. DYRK1A has been proposed to play a role in neurodegenerative alterations, because the enhanced phosphorylation of the DYRK1A substrates tau, APP (amyloid precursor protein) and a-synuclein has been linked to Alzheimer’s disease and Parkinson’s disease, respectively. Recently, the beta-carboline alkaloid harmine has been identified as a potent inhibitor of DYRK1A in vitro (Bain et al. 2007: The selectivity of protein kinase inhibitors: a further update. Biochem J. 408:297-315). Using an optimized peptide (DYRKtide) as a substrate, we determined that harmine inhibits DYRK1A with an IC50 value of 33 nM. The closely related kinase DYRK1B (85% of identical amino acids in the catalytic domain) was inhibited with an IC50 of 165 nM, and the IC50 values for DYRK2 and DYRK4 were more than 10-fold higher (2 µM and 74 µM). In HeLa cells, harmine inhibited the phosphorylation of Thr434 in the DYRK1A substrate, splicing factor 3b1 (SF3b1), with an IC50 value of 48 nM, indicating that harmine can be used to inhibit DYRK1A in cultured cells. We used an in vitro-translation system to assess whether harmine also inhibits the inceptive tyrosine autophosphorylation of DYRK1A. This reaction takes place cotranslationally and is thought to be essential for the one-off activation of the kinase. Tyrosine autophosphorylation of the in vitro-translated DYRK1A protein was half-maximally inhibited by 1 µM harmine, indicating that this reaction is less sensitive to the inhibitor than the phosphorylation of substrates by the mature kinase. Notably, the inhibition of the activating tyrosine autophosphorylation is expected to inhibit DYRK1A irreversibly, because the mature kinase is incapable to phosphorylate Tyr321 in the activation loop. However, treatment with 4 µM harmine had no irreversible effect on the catalytic activity of DYRK1Aa and did not reduce the phosphotyrosine content of DYRK1A in HEK293 cells. We conclude from our results that harmine potently and specifically inhibits DYRK1A in vitro and in cell lines and will serve as a valuable tool for the analysis of DYRK1A effects in cultured cells. 1. Institute of Pharmacology and Toxicology, RWTH Aachen University, Aachen, Germany

Calcium/calmodulin kinase II, a binding partner of the multi PDZ domain protein MUPP1 in mammalian spermatozoa regulates acrosomal exocytosis Ackermann F. (1), Zitranski N. (3), Borth H. (3), Wilhelm B. (2), Gudermann T. (3), Boekhoff I. (3) Upon adhesion to the zona pellucida of the oocyte, mammalian sperm undergo acrosomal exocytosis, a process resembling Ca2+-regulated exocytosis in neurons. We have recently observed that the Multi PDZ domain protein MUPP1, which contains 13 potential protein binding motifs, is expressed in spermatozoa of different mammalian species, where it is exclusively localized to the acrosomal region. In addition, using a functional exocytosis assay which was combined with a Ca2+ chelator strategy, we found that MUPP1 is functionally involved in recruiting molecules controlling the initial tethering and/or docking between the acrosomal vesicle and a specific site of the target plasma membrane whereas a role in the final SNARE-controlled fusion process can be excluded. Here we present data which show that MUPP1 plays a functional role in orchestrating signaling molecules operative in acrosomal secretion by controlling a Ca2+dependent assembly and disassembly of specific binding partners. The results revealed that CaMKII, known to contribute to the tethering of synaptic vesicles to the active zone, shows a striking co-localization with MUPP1 in the acrosomal region as well as in detergent resistant membrane fractions of isolated sperm. GST-pull down experiments with solubilized sperm preparations and a panel of various GST-fusion proteins of nonoverlapping MUPP1-PDZ domains revealed that the activated form of CaMKII binds to the PDZ domain 10-11 whereas other MUPP1 constructs did not show any binding. To assess whether CaMKII is functionally active in acrosomal secretion, we pretreated isolated epididymal mouse sperm either with the membrane permeable Calmodulin inhibitor W7 or the CaMKII inhibitor KN93 and subsequently quantified the spontaneous rate of acrosome reaction. The results showed that both inhibitors led to a dosedependent potentiation of the rate of unstimulated acrosomal exocytosis in mammalian spermatozoa, thus suggesting that CaMKII is involved in preventing spontaneous acrosomal secretion. 1. Institut für Pharmakologie und Toxikologie, Philipps-Universität Marburg, 2. Institut für Anatomie und Zellbiologie, Philipps-Universität Marburg, 3. Walther-Straub-Institut für Pharmakologie und Toxikologie, Ludwig Maximilians Universität München

121

30

123 The cellular matrix determines the regulation of the cell cycle by phosphoinositide-3-kinases Müller R. (1), Wilmes T. (1), Klugbauer N. (1), Meyer D.K. (1) Of the 3 classes of phosphatidylinositide-3-kinases (PI3-Ks) only the isoforms of class I are involved in cell attachment, proliferation and motility. Cultured neocortical astrocytes express only the isoforms p110a and p110b. Using dominant negative p110a (dnp110a) and dn-p110b as well as the pharmacological inhibitors AS-252424 and TGX221 to inactivate p110a and p110b, respectively, we found that both isoforms contribute to the regulation of proliferation of astroglial cells grown on a vitronectin matrix. PI3-Ks and growth factor receptors can interact with integrins in a matrix-dependent manner. Therefore, we have investigated in the present study the role of both isoforms in the proliferation of neocortical astroglial cells grown on poly-L-ornithin (pLO) matrix. In astrocytes grown on vitronectin matrix and synchronized by transient serum withdrawal for 24h, dn-p110a or dn-p110b as well as AS-252424 or TGX-221 reduced the nuclear uptake of BrdU by approximately 50%. In contrast, in cells grown on pLO matrix only the combination of AS-252424 and TGX-221 showed a significant reduction of approximately 40%. In astrocytes grown on vitronectin, both PI3-K-isoforms induced the phosphorylation and inactivation of glycogen synthase kinase-3b (GSK-3b) and thereby facilitated the nuclear localization of cyclin D1, which is essential for DNA synthesis. In cells grown on pLO matrix, however, only the inhibition of the PI3-K a-isoform reduced the nuclear localization of cyclin D1 by approximately 35%. This effect was independent of GSK-3b phosphorylation. The protein level of the CDK inhibitor p27Kip1 was suppressed only by p110a in astroglial cells grown on vitronectin matrix, whereas in cells cultivated on pLO matrix both p110a and p110b decreased p27Kip1 levels. Our data show that the extracellular matrix determines the PI3-K-dependent regulation of proliferation in neocortical astrocytes. 1. Institute of Experimental and Clinical Pharmacology and Toxicology, Albert-LudwigsUniversity, 79104 Freiburg, Germany

124 The role of class I phosphatidylinositide-3-kinase isoforms in astroglial cell migration Wilmes T. (1), Müller R. (1), Klugbauer N. (1), Meyer D.K. (1) Astroglial cells migrate during brain development as well as after brain injury. Phosphatidylinositide-3-kinases (PI3-Ks) of class I contribute to cell motility and migration. In cultured rat neocortical astroglial cells, only the PI3-K isoforms p110a and p110b are expressed but not 110d or 110g (unpublished data). In the present study, we have investigated whether both isoforms are involved in the regulation of astroglial migration. PI3-K activity was inhibited by the PI3-K pan-inhibitor Wortmannin as well as by AS-252424 and TGX-221, which inhibit p110a and p110b, respectively. A wound healing assay was used as read-out system. In the presence of serum, Wortmannin (100 nM) applied for 24 h reduced by approximately 75% the migration of neocortical astrocytes into the inner wound area. Only the combined application of AS-252424 (10 µM) and TGX-221 (100 nM) significantly reduced cell migration. After scratching, the astrocytes first extended protrusions for about 8-10 h before they started the actual migration. Wortmannin or the combination of AS-252424 and TGX-221 significantly reduced the length of the protrusions. Taken together, these results suggested redundant roles for p110a and p110b in serum-induced protrusion formation and migration. Protrusions length in the absence of serum was further increased by 27% after addition of insulin (5µg/ml). The effect of insulin could be blocked by AS-252424 applied alone, whereas TGX-221 was without any effect, indicating that protrusion formation caused by insulin was mediated only by p110a. Cofilin plays a major role in cell migration by reorganising the actin cytoskeleton. The phosphatase slingshot (SSH) dephosphorylates and thereby activates cofilin. Since PI3-Ks can enhance SSH activity, we investigated whether PI3-K inhibition increased the levels of phospho-cofilin in migrating astrocytes. After application of Wortmannin immunocytochemistry showed more phospho-cofilin in cell protrusions than in control cells. Compared to controls, Wortmannin increased the level of phospho-cofilin in whole cell lysates of migrating cells as shown by Western blots. These results suggests that cofilin reorganises the actin cytoskeleton downstream of PI3Ks. According to these results in astrocytes, PI3K inhibition may perturb protrusion formation and subsequent migration by reducing the activity of cofilin. 1. Institute of Experimental and Clinical Pharmacology and Toxicology, Albert-LudwigsUniversity, Freiburg, Germany

125 Analysing the role of cGMP-cGKI-signaling for duodenal bicarbonate secretion Spiessberger B. (1), Weinmeister P. (1), Hofmann F. (1), Werner C. (1), Saur D. (2), Seidler U. (3), Zheng W. (3), Schlossmann J. (4), Lukowski R. (1,5) Bicarbonate secretion is one of the most important defence mechanisms protecting the duodenal mucosa from deleterious effects of gastric acid. It is well known, that the mucosal bicarbonate secretion is increased in response to hydrochloride acid in the gastric juice. However, in addition to luminal acid alkaline secretion is influenced by the central and enteric nervous systems, by the local production of eicosanoids (e.g. Prostaglandin E2), and some hormones such as glucagon, secretin and the gastric inhibitory peptide. In the present study, the possible role of the cGMP-cGKI pathway for the duodenal bicarbonate secretion was investigated in conventional cGKI knockouts and in rescue mice that express either the cGKIa or Ib isoform in SM22a positive smooth muscle cells [1]. The different gene-targeted cGKI mice showed a strongly reduced basal secretion rate of bicarbonate in comparison to their respective littermate controls. Furthermore, the acid induced bicarbonate secretion was nearly absent in all cGKI mutants, whereas pH measurements demonstrated that the hydrochloride acid production of the stomach was similar to control mice. We detected severe gastrointestinal bleedings in cGKIa or Ib-rescue, and cGKI-KO animals, which were caused by age-dependent aggravation of an epithelial ulceration that localized to the papilla Vateri. In conclusion, analysis of cGKI-KO and cGKI-rescue mice indicates that a cGMP-cGKI-dependent pathway is present in non-smooth muscle cells of the duodenum that is involved in the basal and acid-induced secretion of bicarbonate. The

inability to secrete adequate amounts of bicarbonate ultimately leads to the death of cGKI-KO, cGKIa, and Ib rescue mice. [1] Weber at al. Circ Res 101: 1096-1103 (2007). 1. For 923 at Institut für Pharmakologie und Toxikologie, TU München, Germany, 2. Department of Internal Medicine II, Klinikum Rechts der Isar, TU München, Germany, 3. Department of Gastroenterology, Hepatology and Endocrinology, Hannover Medical School, Germany, 4. Department of Pharmacology and Toxicology, Universität Regensburg, Germany , 5. Institut für Pharmakologie und Toxikologie, TU München, Germany

126 cGMP-cGKI signaling in neurohormonal-induced heart hypertrophy Lukowski R. (1, 2), Loga F. (1, 2), Brummer S. (1), Hofmann F. (2) Pathological heart hypertrophy is characterized by the maladaptive growth of cardiomyocytes, ventricular fibrosis, and re-activation of a fetal cardiac gene program ultimately leading to heart failure and sudden death. Analysis of transgenic mice with increased natriuretic peptide receptor signaling or chronic inhibition of the cGMPdegrading phosphodiesterase-5 by sildenafil indicated that cGMP can blunt hypertrophic signals from pressure load and neurohormonal stress perhaps via activation of cardiac cGMP-dependent kinase I (cGKI). To test whether the deletion of cGKI in the heart results in an altered cardiac remodeling, we used the recently generated cGKIβ rescue mice [1] since the cGKI protein is absent from the myocardium of these animals. We studied hypertrophy in response to chronic β-adrenoreceptor stimulation by isoproterenol (ISO) and detected significantly reduced expression levels of the β1adrenergic receptor in the hearts of gene-targeted cGKI mice as compared to littermate controls. However, ISO (30 mg/kg/d) increased the heart/body weight ratios (HW/BW) to a similar extent in both groups. Up-regulation of fetal gene markers (Myh7, PI-16) and down-regulation of genes which are normally expressed in the adult heart (Myh6, Serca2), both indicative for pathological hypertrophy, were comparable between the two genotypes. Interestingly, an increase of the brain natriuretic peptide (NPPB), a hallmark of pathological cardiac remodeling, was not detected in hearts of Iβ-rescue mice. In conclusion, the data indicate that endogenous cGKI signaling does not protect the heart from hypertrophy induced by chronic β-adrenergic stimulation. [1] Weber at al. Circ Res 101: 1096-1103 (2007). 1. Institut für Pharmakologie und Toxikologie, TU München, Germany, 2. FOR923 at Institut für Pharmakologie und Toxikologie, TU München, Germany

127 Facilitation of L-type calcium channels – new insights from a knock-in mouse line Köstner K. (1,2), Blaich A. (2), Domes K. (2), Fischer S. (1,2), Moosmang S. (1,2), Hofmann F. (2), Welling A. (1,2) The voltage-gated L-type Ca2+ channel CaV1.2 is the predominant Ca2+ channel in the cardiovascular system. In the heart, the influx of Ca2+ through CaV1.2 determines the plateau-phase of the cardiac muscle action potential and is essential for excitationcontraction coupling. The availability of the CaV1.2 current is modulated by Ca2+ in a dual-purpose way. As Ca2+ enters the cell it leads to inactivation of the Ca2+ channels, a phenomenon known as Ca2+-dependent inactivation. A train of repetitive depolarisations or a strong depolarising prepulse increases the L-type Ca2+ current. This phenomenon is known as Ca2+-dependent facilitation. Various mechanisms have been proposed for facilitation, including a switch between gating modes of channel activity and voltagedependent phosphorylation of the channel. Ca2+/calmodulin-dependent protein kinase II (CaMKII) has been reported to play a key role in this process. We identified serines 1512 and 1570 in the CaV1.2 carboxyterminus as CaMKII phosphorylation sites [1]. To elucidate the molecular details of the interaction between CaMKII and the carboxyterminus of the CaV1.2 alpha subunit in vivo, a mouse line was established using a knock-in (Ki) approach with targeted mutation of serines 1512 and 1570 of the channel to alanine. Mice were viable and fertile and showed no gross abnormalities in behavioural tests. The basal Ca2+ current (ICa) of control (ctr) and Ki cardiomyocytes had a similar current amplitude and current density. Voltage-dependence of activation was unchanged. Ca2+ -dependent inactivation was slightly increased by the mutation. Depolarising prepulses facilitated ICa in ctr cardiomyocytes. As expected, this facilitation was significantly diminished in Ki cardiomyocytes. A train of slow repetitive depolarisations (0.5 Hz) increased the L-type Ca2+ current in ctr cells. The increase of ICa was significantly enhanced in the Ki cardiomyocytes. Thus voltage- and frequencydependent facilitation is differently affected by serines 1512 and 1570 in the carboxyterminus of the CaV1.2 alpha subunit. [1 ] T.-S. Lee, R. Karl, S. Moosmang, P. Lenhardt, N. Klugbauer, F. Hofmann, T. Kleppisch. A. Welling. J Biol Chem. 281: 255607, (2006) 1. Institut für Pharmakologie und Toxikologie, TU München, Germany, 2. FOR 923 at Institut für Pharmakologie und Toxikologie, TU München, Germany

128 Temporally regulated inactivation of the loxP-targeted CaVβ2 gene in the adult mouse heart Meissner• M. (1), Weissgerber• P. (1), Camacho Londoño J.E. (1), Molkentin J.D. (2), Nilius B. (3), Flockerzi V. (1), Freichel M. (1) The L-type Ca2+ channels in heart are predominantly formed by α1C (CaV1.2) and CaVβ2 (β2). Disruption of the β2 gene resulted in diminished L-type Ca currents in cardiomyocytes of embryonic day (E) 9.5 (1). This led to a functionally compromised heart, causing defective remodeling of intra- and extra-embryonic blood vessels and embryonic death following E10.5 (1). The embryonic death and the defects in vascular remodeling were also observed when the β2 gene was selectively targeted in cardiomyocytes, demonstrating that they are secondary to cardiac failure (1). To manipulate β2-gene expression in the adult mouse heart, the mice carrying floxed β2 alleles were crossed with transgenic mice carrying the Cre recombinase fused to a mutant estrogen receptor (Mer) under the control of the cardiac specific a-MHC promoter (2). Tamoxifen-induced Cre-mediated recombination was monitored by β2 protein expression using three independent antibodies for β2. No change did occur in the absence of tamoxifen, but β2 protein expression was reduced to 17.6 ± 3.1 %

31

compared with vehicle treated animals as controls (n=5 series of experiments with independent animal cohorts). In myocytes from tamoxifen-treated mice L-type Ca current densities were reduced by ~30% in the absence and presence of Bayk 8644, whereas -V1/2 (mV) was not significantly different. CaVβ1, CaVβ3 and CaVβ4 did not compensate the more than 82% loss of β2 protein and protein levels of CaV1.2 and α2-δ were not changed. The data indicate that in the adult heart a small fraction of the β2 protein is sufficient for L-type Ca channel activity and that an additional significant fraction may serve functions, which have yet to be identified. • contributed equally 1 Weißgerber et al. (2006) Circ Res. 99:749-757.2 Sohal et al. (2001) Circ Res. 89:20-25. 1. Institut für Experimentelle und Klinische Pharmakologie und Toxikologie, Universität des Saarlandes, Homburg, Germany, 2. Department of Pediatrics, University of Cincinnati, Cincinnati, USA, 3. Departement Moleculaire Celbiologie, Katholieke Universiteit Leuven, Leuven, Belgium

129 Disturbances of thalamocortical rhythmicity in Cav2.3 deficient mice Weiergräber M. (1,2), Struck H. (1), Henry M. (1), Ho M.S.P. (3), Schneider T. (1,2) Voltage-gated calcium channels (VGCCs) regulate neuronal excitability and are important factors in epileptogenesis and neurodegeneration. Recent findings in an animal model of experimentally induced epilepsy suggest a novel, important proictogenic and proneuroapoptotic role of the Cav2.3 E/R-type VGCCs in convulsive generalized tonic-clonic and hippocampal seizures (Weiergräber et al., 2007). Though Cav2.3 is expressed in key structures of the thalamocortical circuitry, its functional relevance in non-convulsive absence seizure activity remains unknown. To this end, we investigated absence specific spike-wave discharge (SWD) susceptibility in control and Cav2.3-deficient mice by systemic administration of g-hydroxybutyrolactone (GBL, 70 mg/kg i.p.), followed by electrocorticographic radiotelemetric recordings, behavioral analyses and histomorphological characterizations (Weiergräber et al., 2008). Based on motoric studies, SWD and power spectrum density (PSD) analyses, our results demonstrate that Cav2.3-/- mice exhibit increased absence seizure susceptibility and altered absence seizure architecture compared to control animals. This study provides evidence for the first time that Cav2.3 E/R-type Ca2+ channels are important in modulating thalamocortical hyperoscillation by exerting anti-epileptogenic effects in nonconvulsive absence seizures. Weiergräber M, Henry M, Ho MSP, Struck H, Hescheler J, Schneider T (2008), Mol Cell Neurosci, 39: 605-618. Weiergräber M, Henry M, Radhakrishnan K, Hescheler J, Schneider T (2007). J Neurophysiology, 97: 3660-3669. 1. Inst. Neurophysiologie, Uniklinik Köln, Germany, 2. Center for Molecular Medicine Cologne (CMMC), 3. Inst. Biochemie, Uniklinik Köln, Germany

VE-cadherin-coated microbeads. In summary, we demonstrate that the function of endothelial TRPC4 is strictly dependent on the cells phenotype. We suggest recruitment of TRPC4 proteins into cell-cell contacts as a key mechanism for control of endothelial Ca2+ signalling along with endothelial phenotype switching. 1. Inst. of Pharmaceutical Sciences, Pharmacology and Toxicology, University of Graz, Austria

132 Phospholipase Cε (PLCε)-mediated activation of classical transient receptor potential 6 (TRPC6) increases barrier function of glomerular podocytes Dietrich A. (1), Kalwa H. (1), Storch U. (2), Schmidt M. (3), Smrcka A. (4), Hildebrandt F. (5), Gudermann T. (2) The classical transient receptor potential channel 6 (TRPC6) is activated by direct exposure to diacylglycerol (DAG) produced by isoforms of the phospholipase (PLC) family. Mutations in TRPC6 identified in patients with nephrotic syndrome of the kidney link this channel to structural components of the glomerular slit diaphragma in podocytes (reviewed in Gudermann, Nat. Genet. 37: 663-664 (2005)), while another publication (Hinkes et al., Nat. Genet. 38: 1397-1405 (2006)) identified mutations in PLCe in patients with similar kidney disorders. For that reason we investigated, if there is a functional interaction between these two important mediators of cell signalling cascades in podocytes. We were able to co-immunoprecipitate PLCε and TRPC6 in primary murine podocytes. The importance of this interaction was further emphasized by the fact that murine podocytes showed an angiotensin II (ATII)-induced increase in barrier function, which was mimicked by OAG, a membrane permeable DAG-analogue. Most interestingly, TRPC6-/- podocytes isolated from TRPC6-deficient mice displayed a markedly decreased barrier function in comparison to wild-type cells. Moreover, actin fiber formation by ATII was significantly reduced in TRPC6-/- podocytes. Similar results were obtained with PLCε-/- podocytes. These data suggest an important role of PLCε in the regulation of TRPC6 activity in glomerular podocytes, and may provide inroads into the understanding of AT1 receptor blocker treatment in nephrotic syndrome. Supported by a grant from the Fritz-Thyssen-Stiftung. 1. Dept. of Pharmacology and Toxicology, Philipps University, Marburg, Germany, 2. Dept. of Pharmacology and Toxicology, Ludwig Maximilians University, Munich, Germany, 3. Dept. of Molecular Pharmacology, University of Groningen, The Netherlands, 4. Dept. of Pharmacology and Physiology, University of Rochester, Rochester, NY, USA, 5. Dept. of Pediatrics, University of Michigan, Ann Arbor, MI, USA

133 130 Analysis of cardiac hypertrophy induced by isoproterenol or angiotensin II in TRPC-deficient mice Camacho Londoño J.E. (1), Meissner M. (1), Dietrich A. (2), Birnbaumer L. (3), Flockerzi V. (1), Freichel M. (1) Cardiac hypertrophy is an adaptive response of the heart to mechanical load abnormalities. The hypertrophic response can be induced after stimulation of G protein coupled receptors such as β adrenoceptors or Angiotensin II receptor I (AT1). It is characterized by enhanced cellular growth and elevated intracellular diastolic Ca2+ levels which initiates Ca2+-dependent signalling cascades and reactivation of fetal genes through the calcineurin-NFAT pathway. The mechanisms leading to the impaired intracellular Ca2+ homeostasis are still incompletely understood, but receptor-operated cation channels of the TRPC family might be activated by neurohumoral hypertrophic stimuli. Recent reports showed that cardiomyocyte specific overexpression of either TRPC3 or TRPC6 in mice leads to an increase of cardiomyocyte size, cardiac mass and heart failure. We currently evaluate this concept by analysing the development of cardiac hypertrophy induced by continuous infusion of hypertrophic agents in mouse lines lacking either TRPC1, TRPC3, TRPC4, TRPC5 or TRPC6 as well as in TRPC compound knockout mice derived thereof. Up to now we provide evidence that in TRPC1/TRPC4 (-/-)2 mice the isoproterenol-induced hypertrophy response is reduced, whereas angiotensin II-induced hypertrophic response is increased in compound TRPC3/TRPC6 (-/-)2 mice. Mouse lines in which only one TRPC gene is inactivated are currently systematically analysed.supported by the Klinische Forschergruppe 196 (“Signaltransduktion bei adaptativen und maladaptativen kardialen RemodelingProzessen”) and GK 1326. 1. Experimentelle und Klinische Pharmakologie und Toxikologie, Universität des Saarlandes, Homburg, Germany, 2. Pharmakologie und Toxikologie, Philipps-Universität Marburg, Marburg, Germany, 3. Transmembrane Signaling Group, National Institute of Environmental Health Sciences (NIEHS), Research Triangle Park, NC, USA

131 Phenotype-dependent cellular targeting and function of TRPC4 in human vascular endothelial cells Graziani A. (1), Krenn M. (1), Poteser M. (1), Groschner K. (1) TRPC4 (transient receptor potential channel, canonical 4) has originally been suggested as a store-operated channel that governs endothelial Ca2+ homeostasis. However, this concept was recently questioned by siRNA knock down experiments in human endothelial cells (Abdullaev et al. 2008; Circ. Res.103:1289). As TRPC4 has been specifically implicated in the control of endothelial barrier function, we set out to characterize cellular localization and function of TRPC4 in both proliferative and quiescent, barrier-forming human microvascular endothelial cells (HMEC). Membrane targeting and surface presentation of TRPC4 was strongly promoted by the formation of cell-cell adhesions. Moreover, both basal- and epidermal growth factor-stimulated Ca2+ influx was substantially enhanced in barrier-forming, confluent cells. Endothelial TRPC4 was found to co-precipitate with VE-cadherin, an essential component of junctional complexes, and the two membrane proteins co-localized upon expression in HEK293 cells, displaying distinct targeting into cell-cell contacts. TRPC4-mediated Ca2+ signals and membrane currents in HEK293 cells were significantly promoted by the formation of either VE-cadherin-mediated cell-cell contacts or artificial contacts between cells and

The TRPM3 agonist pregnenolone sulphate enhances insulin secretion Wagner T.F.J. (1), Lambert S. (1), Straub I. (2), Loch S. (1), Mannebach S. (1), Philipp S.E. (1), Oberwinkler J. (1) TRPM3 is a poorly characterized member of the large family of TRP ion channels. Overexpression of TRPM3 in HEK293 cells induces the expression of highly Ca2+ permeable channels, which can be activated by two chemically unrelated substances: the endogenous steroid pregnenolone sulphate and the dihydropyridine nifedipine. We found that pregnenolone sulphate, and nifedipine as well, lead to influx of Ca2+ in INS1 cells, a cell line derived from rat pancreatic β cells. Furthermore, pregnenolone sulphate induced Ca2+ signals in whole pancreatic islets and in pancreatic islet cells in primary culture. When tested in whole-cell patch-clamp recordings, the endogenous steroid activated channels in these cells showed the key biophysical properties of recombinant TRPM3 channels. TRPM3 expression in INS1 cells and pancreatic islets was evaluated by Northern and Western blotting. We demonstrated with combined Ca2+-imaging and immunohistochemistry that pregnenolone sulphate induced Ca2+ signals exclusively occur in insulin producing pancreatic β cells, but not in glucagon secreting α cells. Application of pregnenolone sulphate increased the insulin secretion from whole pancreatic islets. This increase in secretion was directly correlated to the pregnenolone sulphate induced increase of the intracellular Ca2+ concentration in cells of whole pancreatic islets. Our results establish that TRPM3 is an essential component of an ionotropic steroid receptor in pancreatic β cells enabling unanticipated crosstalk between steroidal and insulin-signaling endocrine systems. 1. Dept. of experimental und clinical Pharmacology and Toxicology, University of the Saarland, Homburg, Germany, 2. Rudolf-Boehm-Institute for Pharmacology and Toxicology, University of Leipzig, Leipzig, Germany

134 Identification of regulatory domains in TRPM3 steroid receptors Fruehwald J. (1), Mannebach S. (1), Oberwinkler J. (1), Philipp S.E. (1) The TRPM3 gene encodes a large number of variants which arise by alternative splicing (1) and we recently showed that TRPM3 proteins built ionotropic receptors in pancreatic beta cells allowing Ca2+-entry in response to neurosteroids like pregnenolone sulphate (2). Analyzing the Ca2+ dependent regulation of TRPM3, we found that calmodulin binds to regions common to all isoforms of the channel protein. Within a minimal calmodulin binding site we identified those residues, which are essential for binding. Accordingly, steroid induced Ca2+-entry in stably TRPM3 expressing HEK293 cells as well as in the pancreatic beta cell line Ins-1 was strongly affected after introduction of recombinant calmodulin. All splice variants TRPM3alpha2, alpha3, alpha4, alpha5 and alpha6, which differ in short stretches of 10 to 25 amino acids encoded by exons 8, 15 and 17, were able to mediate steroid-induced Ca2+-entry when expressed in HEK293 cells. In clear contrast, a splice variant TRPM3alpha7 lacking 18 amino acid residues encoded by exon 13 did not show any Ca2+-signal after application of pregnenolone sulphate. TRPM3alpha7 proteins were still transported to the plasma membrane and coimmunoprecipitated with other TRPM3subunits. Consistently, introduction of recombinant TRPM3alpha7 into TRPM3alpha2 expressing cells reduced steroid activated Ca2+-entry. Furthermore, introduction of TRPM3alpha7 exerted dominant negative effects on endogenous steroid activated Ca2+-entry channels in pancreatic Ins1 cells. Alignments with other members of the TRPM subfamily showed strong conservation of corresponding domains. Deletion of this domain in TRPM8 resulted in a complete loss of Ca2+-entry in response to menthol. Our observations do not only

32

pinpoint a domain necessary for Ca2+/calmodulin-dependent regulation of TRPM3 but also a region which might be essential in general for TRPM dependent channel function. Furthermore our findings indicate that splicing of 18 amino acid residues provides a mechanism to reduce the activity of TRPM3 channel complexes. 1.Oberwinkler,J., et al. Alternative Splicing Switches the Divalent Cation Selectivity of TRPM3 Channels. J. Biol. Chem. 280, 22540-22548 (2005). 2.Wagner,T.F. et al. Transient receptor potential M3 channels are ionotropic steroid receptors in pancreatic beta cells. Nat. Cell Biol. in press (2008). 1. Dept. of Pharmacology, University of Saarland, Homburg, Germany

135 Phenotypical divergence between mice and humans carrying loss-of-function alleles of TRPM6 Chubanov V. (1), Storch U. (1), Chen X. (2), Mederos y Schnitzler M. (2), Hofmann T. (2), Gudermann T. (1) Divalent cation-selective outwardly rectifying currents, which are induced upon removal of intracellular Mg2+, have been described in virtually all mammalian cells examined so far. Accordingly, these currents were referred to as magnesium-inhibited currents (MIC) or magnesium nucleotide-regulated metal ion currents (MagNuM). TRPM6 and TRPM7 (melastatin-related members of the transient receptor potential gene family) were found to be molecular candidates for MIC/MagNuM. Recent studies revealed that TRPM7 is essential for the cell viability, embryonic development and Mg2+ homeostasis. Loss-offunction mutations in the human TRPM6 gene result in hypomagnesemia with secondary hypocalcemia (HSH). HSH is an autosomal recessive disorder characterized by low serum Mg2+ and Ca2+ levels. It was reported that TRPM6 is specifically expressed in the intestinal epithelium and the distal convoluted tubule in the kidney. Together, these findings underlined a concept that TRPM6 can directly participate in Mg2+ uptake by renal and intestinal epithelial cells. In order to rigorously investigate the role of TRPM6 in the pathomechanism of HSH, we initiated a phenotypic analysis of TRPM6 gene deficient mice carrying a LacZ reporter sequence. Unexpectedly from what we know about the clinical picture of HSH, homozygous -/- TRPM6 mice died at a midgestational stage. Tracking the LacZ expression and using TRPM6-specific antibody, we observed an expression profile of TRPM6 much broader than that a previously assumed, supporting the notion that the phenotype of HSH patients only partially depicts the physiological relevance of TRPM6. 1. Walther Straub Institut für Pharmakologie und Toxikologie, Ludwig Maximilians Universität München, Germany, 2. Institut für Pharmakologie und Toxikologie, Philipps Universität Marburg, Germany

136 TRPC1 suppresses the calcium permeability in heteromeric channel complexes with TRPC4/5 without effects on arterial tone Mederos y Schnitzler M. (1), Storch U. (2), Essin K. (3), Philipp M. (1), Fésüs G. (3), Dubrovska G. (3), Kalwa H. (1), Chubanov V. (2), Dietrich A. (1), Gollasch M. (3), Gudermann T. (2) Among the classical transient receptor potential (TRPC) subfamily, TRPC1, 3, 4, 5, 6 and 7 are expressed in vascular smooth muscle cells (VSMCs). However, their specific roles in arterial smooth muscle (ASM) function are still elusive. We tested the hypothesis that TRPC1 may form functional heteromeric TRPC channels with other TRPC subunits to regulate vascular tone via store- or receptor-operated Ca2+ influx. We studied TRPC1 proteins in a heterologous expression system and TRPC1 gene-deficient mice (TRPC1-/). In patch-clamp recordings, we found that recombinant TRPC1 is not able to build functional homomeric channels in HEK293 cells. Instead, TRPC1 was able to form functional heteromeric channel complexes with TRPC4 and 5, but not with TRPC3, 6 and 7. In both TRPC1/4 and TRPC1/5 heteromers, TRPC1 subunits significantly decreased Ca2+ permeation. Changes of the amino acids in the putative pore forming region of TRPC1 further reduced the Ca2+ permeability. On the cellular level in enzymatically isolated cerebral smooth muscle cells, we obtained no significant differences in cation currents activated by hypoosmotic swelling and positive pipette pressure between TRPC1-/- and wild-type cells. Even in intact mesenteric arteries, short intravascular pressure pulses provoked the same vessel dilatations in both cases, indicating that under physiological conditions with unaffected membrane proteins and an intact extracellular matrix TRPC1 is not mechanosensitive. Compared to wild-type mice, thoracic aortas and mesenteric arteries of TRPC1-/- mice showed no change in storeoperated cation influx induced by caffeine and cyclopiazonic acid. Furthermore, deficiency of TRPC1 neither affected α1-adrenoceptor-induced, inositol-1,4,5 trisphosphate-dependent contractions of mesenteric arteries and aorta rings, nor Ca2+ influx in isolated VSMCs induced by high external [Ca2+] or Ca2+ store depletion using cyclopiazonic acid or caffeine. Ca2+ sparks and spontaneous transient outward currents were similar in TRPC1-/- and wild-type VSMCs. We conclude that TRPC1 is able to form heteromeric TRP channels with TRPC4 and 5 with a reduced Ca2+ permeability. However, TRPC1 alone or in combination with TRPC4 and 5 is not an obligatory component of store and receptor-operated Ca2+ influx in ASM. 1. Institut für Pharmakologie und Toxikologie, Philipps-Universität, Marburg, Deutschland, 2. Walther-Straub-Institut für Pharmakologie und Toxikologie, LudwigMaximilians-Universität, München, Deutschland, 3. Charité Campus Virchow, Medizinische Klinik, m.S. Nephrologie und Internistische Intensivmedizin und ECRC, Berlin, Deutschland

137 Identification of the interaction site between the L-type Cav1.2 calcium and BKCa channel Lohmüller B. (1), Klugbauer N. (1) L-type calcium channels are involved in a large number of physiological functions such as contraction of cardiac and smooth muscle, the release of neurotransmitters and hormones and the regulation of enzymatic activities. In neurons, voltage-activated calcium channels appear to form macromolecular complexes with the unique large conductance calcium- and voltage-activated K+ channels (BKCa). The BKCa channel co-

purified with the calcium channels Cav1.2, Cav2.1, Cav2.2, but not with Cav2.3. Such colocalization ensures the selective and rapid activation of the BKCa channels by the local increase of cytosolic calcium. Together these channels contribute to the action potential repolarization, to the fast phase of afterhyperpolarization and to the formation of a feedback loop that regulates the release of hormones and transmitters. In order to demonstrate a direct interaction between the BKCa and voltage-activated calcium channels and to identify the precise interaction site, we used the split-ubiquitin system, a yeast based genetic screening system. With this novel system the interaction of two proteins can be detected in situ at the cellular membrane. As a positive control we routinely used the calcium channel ß2 subunit, which binds to the intracellular loop connecting domain I and II. To identify the interaction site with the BKCa channel, we performed screenings of BKCa-cDNA fragments with the four different domains of the Cav1.2 a1 subunit as baits. The interaction was found between segment S6 of the BKCa channel and domain I of the Cav1.2 a1 subunit. We narrowed down the interaction site and identified segments IS1 and IS3 of the Cav1.2 channel to be crucial for the binding to the BKCa segment S6. Shortening of segment 6 of the BKCa channel reduced or even abolished the interaction. Furthermore, the exchange of the four residues glycine (376), leucine (377), methionine (379) und phenylalanine (380) to alanine in segment 6 of the BKCa channel abolished the interaction. Here we demonstrate for the first time a direct interaction between transmembrane segments of different ion channels. 1. Institut für Pharmakologie und Toxikologie, Albert-Ludwigs-Universität, Freiburg, Germany

138 Genetic study of calcium channel beta-subunits in autistic patients Breitenkamp A.F.S. (1), Nass R.D. (1), Sinzig J. (2), Meyer P. (3), Nürnberg P. (4), Herzig S. (1) Calcium channels are heteromultimeric protein complexes consisting of a pore-forming Cav (or alpha) subunit and the auxiliary subunits beta, alpha-delta (and gamma).Autistic symptoms are part of the Timothy-syndrome, where a point mutation of the pore-forming calcium channel subunit Cav1.2 leads to incomplete inactivation (Splawski et al., Cell 2004). In addition, private mutations of the Cav3.2 pore subunit were found sporadically among autistic patients (Splawski et al., JBC 2006). Mutations of an auxiliary subunit beta of the L-type calcium channel may be responsible for similar biophysical channel phenotypes. For instance, shortening of the N-terminal structure of the voltage-gated calcium channel beta2-subunit causes defects of inactivation (Herzig et al., FASEB J. 2007). In addressing the complex genetic basis of autism, multipoint linkage analyses provided by “International Molecular Genetic Study of Autism Consortium” revealed a potential role of the chromosomes containing the gene loci of the calcium channel subunits beta1, beta2, beta4 (IMGSAC, Am. J. Hum. Genet. 2001). Mutations in these gene loci may influence neuronal function or development by amino acid exchange, alternative splicing or affected promoter regions. To examine the potential role of beta subunits, we determined the DNA sequence all exons of the entire N-terminal regions of these three beta-subunits in 162 patients with autism spectrum disorder (ASD). We detected several known and new (intronic) variations in each subunit and compared these genotypes with 165 controls. As our results show no coding mutations at present, we are analysing the distribution of genotypes in patients and controls to test for a possible association of beta-subunit genes with autism. According to our findings we will choose one subunit gene for complete sequencing of all exons with flanking intronic regions to screen for ASD-specific exonic mutations or functional intronic variations. 1. Department of Pharmacology, University of Cologne, 2. Department of Child and Youth Psychiatry, University of Cologne, 3. Institute of Molecular Medicine, Munich, 4. Cologne Center of Genomics, University of Cologne

139 Subtype-specific inhibition of cardiac L-type calcium channels by Gαi3 Dizayee S. (1), Kaestner S. (1), Piekorz R. (2), Matthes J. (1), Meszaros J. (1), Nürnberg B. (2), Herzig S. (1) Chronic overexpression of β2-adrenoceptors in transgenic mice depresses the activity of L-type calcium channels (Heubach et al., Br J Pharmacol 2001). Analysis of singlechannel experiments suggests that this effect is due to the activation of Gαi3 but not Gαi2 proteins (Foerster et al., PNAS 2003). Here we examined this idea using myocytes from knockout mice lacking either one of the pertussis toxin-sensitive proteins, Gαi2 or Gαi3. Gene deletion was verified by determining protein expression levels of the highly homologous Gαi isoforms in cardiac tissue by immunoblot and ADP-ribosylation. A marked upregulation (2-3fold) of Gαi3 protein was detected in Gαi2 knockout mice. In ventricular myocytes from Gαi2 knockout mice peak whole-cell calcium current density was decreased (Gαi2: -6.9±0.9 pA/pF, n=7) compared to wildtype (wt: -10.4±0.5 pA/pF, n=17, p <0.05) controls. In addition, we observed kinetic alterations of whole-cell currents in Gαi2 knockout mice: 1. half-maximum voltage of steady state inactivation was shifted to more negative values (Gαi2: -25.1±0.9 mV, n=7, wt: -18.3±0.9 mV, n=10, p < 0.05), 2. time dependent inactivation was accelerated, and 3. the recovery from inactivation was delayed. In contrast, basal current density was increased in myocytes from Gαi3 knockout (Gαi3: -14.3±0.8 pA/pF, n=14, p < 0.05 vs. wt). Here, no apparent changes of kinetic parameters were found. All genotype-related differences in calcium currents were dissipated by incubation of myocytes with pertussis toxin for 3 hours. We conclude that a tonic inhibition of cardiac L-type calcium channel activity is mediated by Gαi3 protein, but not by Gαi2. 1. Dept. of Pharmacology, University of Cologne, Cologne, Germany, 2. Dept. of Biochemistry II, University of Düsseldorf, Düsseldorf, Germany

33

140 Length-dependent modulation of cardiac L-type calcium channels by the Nterminus of a β1a subunit Jangsangthong W. (1,2), Kuzmenkina E. (1), Khan I.F.Y. (1), Matthes J. (1), Hullin R. (3), Herzig S. (1,2) β-subunits of L-type Ca2+ channel (L-VDCC) profoundly modulate properties of L-VDCC. Previously, we demonstrated that the N-terminus of β2-subunit serves as a lengthdependent structural determinant of channel inactivation (Herzig et al., FASEB J. 2007). Here, we tested the role of the N-terminus of β1a-subunit. Three artificial β1a-subunit mutants with different N-terminus lengths, β1aN18, β1aN27 and β1aN51, were generated. Their modulatory functions were investigated in recombinant L-VDCCs in HEK293 cells and compared with the natural full-length isoform, termed β1aN60. In whole-cell patch-clamp measurements, we confirmed functional expression of all β1asubunit isoforms by a marked increase of current density and a leftward shift of activation, as compared to control transfections without any β-subunit. No obvious differences were found among β1a-subunit isoforms regarding peak inward current. However, shortenening of the N-terminus progressively decreased the rate and extent of time-dependent inactivation at all test voltages. Descriptive single-channel analysis (e.g., peak ensemble average current, open probability, availability) revealed similar values among β1a-subunit isoforms, except for small deviations with β1aN51. Intriguingly, the extent of the macroscopic inactivation of ensemble average currents followed the length of the N-terminus (β1aN60>β1aN51>β1aN27>β1aN18). We performed Markov modeling for detailed kinetic analysis of single channel data. Open probability, availability, first-latency, open-time and closed-time histograms were simultaneously fitted. A significant linear correlation was found between the inactivation rates and the N-terminus length. The other rate constants (e.g., close state to close state or close state to open state) did not vary with the N-terminal length of the β1asubunit. Our results suggest that inactivation is under length-dependent control of the Nterminus of L-VDCC β1-subunit. This finding could represent a general mechanism of βsubunit modulation. 1. Dept. of Pharmacology, University of Cologne, Cologne, Germany, 2. Center for Molecular Medicine, University of Cologne, Cologne, Germany, 3. Dept. of Cardiology, University Hospital Bern, Bern, Switzerland

141 Analysis of the role of TRPC proteins for platelet aggregation in mice Camacho Londoño J.E. (1), Meissner M. (1), Dietrich A. (2), Birnbaumer L. (3), Flockerzi V. (1), Freichel M. (1) Platelets play an important role not only in primary hemostasis but also in intravascular thrombus formation after atherosclerotic remodeling of the vascular system. Changes in intracellular Ca2+ concentration due to Ca2+ entry from the extracellular space are critical for platelet activation and subsequent aggregation. Ca2+ entry follows activation of plasma membrane receptors including the Gq-coupled P2Y1 receptors, protease activated receptors 4 (PAR4) and TP receptors, which are activated by agonists such as ADP, thrombin and thromboxane, respectively. Recently it has been proposed that proteins of the TRPC channel subfamily such as TRPC1, TRPC3, TRPC4, TRPC5 and TRPC6 are expressed in platelets and might contribute to platelet activation as constituents of agonist-activated Ca2+ entry channels; however, the analysis of TRPC channels in platelets is hampered by the lack of appropriate channel blockers and agonists as well as suitable antibodies to identify TRPC proteins in primary cells. To address this issue, we have analysed the expression of different TRP proteins in mouse platelets using platelets isolated from TRP-deficient mice as negative controls. Using antibody 861 directed against mouse TRPC6 we identify the 100 kDa TRPC6 proteins in platelets from wild type but not from TRPC6-/- mice. Additionally, we have been evaluating platelet aggregation in vitro using washed platelets isolated from different TRPC deficient mouse lines, including mouse lines lacking either TRPC1, TRPC3, TRPC4, TRPC5 or TRPC6 as well as from TRPC compound knockout mice derived thereof. Until now, we found a reduced ADP-induced aggregation in platelets from TRPC1/C6 (-/-)2 mice, but no differences in response to thrombin and collagen.supported by the Klinische Forschergruppe 196 (“Signaltransduktion bei adaptativen und maladaptativen kardialen Remodeling-Prozessen”) and GK 1326. 1. Experimentelle und Klinische Pharmakologie und Toxikologie, Universität des Saarlandes, 66421 Homburg, Germany, 2. Pharmakologie und Toxikologie, PhilippsUniversität Marburg, 35043 Marburg, Germany, 3. Transmembrane Signaling Group, National Institute of Environmental Health Sciences (NIEHS), Research Triangle Park, North Carolina 27709, USA

142 Cellular responses of small cell lung cancer cells upon allyl isothiocyanate stimulation Schäfer E.A.M. (1), Stohr S. (1), Kalwa H. (2), Breit A. (1), Büch T.R.H. (1), Gudermann T. (1) Isothiocyanates (ITCs), naturally occurring in cruciferous vegetables such as mustard, brussels sprouts or horseradish were suggested to have a chemopreventive activity against lung carcinogenesis. Since ITCs are known activators of the TRPA1 cation channel, we determined the expression of TRPA1 in classic small cell lung cancer (SCLC) cell lines (H69, H510, H146) and their cellular responses upon stimulation with allyl isothiocyanate (AITC). Regarding the mRNA expression TRPA1 was found to be highly expressed in H146 cells and moderately expressed in H510 and H69 cells. In comparison, there was no TRPA1 expression in the pancreatic carcinoma cell line 8988t representing a negative control in this study. Furthermore, we tested whether AITC stimulation leads to a rise of intracellular calcium, using cell lines stably expressing the calcium sensor protein aequorin. H69Aeq cells as well as H510Aeq cells showed a transient increase of the intracellular calcium concentration [Ca2+]i after application of AITC, whereas 8988tAeq cells showed no reaction. Two further TRPA1-activators (cinnamaldehyde and URB597), as well as a TRP-channel blocker (ruthenium red) were tested. A transient increase of the [Ca2+]i was also observed following application of cinnamaldehyde and URB597. In addition, the AITC mediated Ca2+-signal was completely blocked after incubation with ruthenium red indicating a potential role of TRPA1 in SCLC-cell lines, a hitherto unknown fact. Finally, application of AITC caused a

time dependent activation of p44/42 MAPK in the SCLC cell lines H69, H510 and H146, whereas in the TRPA1-negative cell line 8988t no effect was detected. Thus, the investigated cell lines may represent an excellent tool to distinguish TPRA1-dependent from -independent effects of AITC in tumour cells. 1. Walther-Straub-Institut für Pharmakologie und Toxikologie, Ludwig-MaximiliansUniversität, München, Deutschland, 2. Institut für Pharmakologie und Toxikologie, Philipps-Universität, Marburg, Deutschland

143 Influence of the mutation R1379C in NBF2 of sulfonylurea receptor 1 (SUR1) on the interaction of SUR with 17β-estradiol Lösle M. (1), Hiller S. (1), Ackermann S. (1), Hambrock A. (1) Sulfonylurea receptor (SUR) 1 forms the regulatory subunit of the pancreatic ATPsensitive K+ channel (KATP channel), which plays a key role in triggering of insulin secretion in the β-cell. Previously, we have shown that SUR, in addition to regulating electrical activity of KATP channels, is also specifically involved in the modulation of βcell apoptosis after exposure to 17β-estradiol and that this SUR1-specific effect is abolished by the mutation M1289T in transmembrane helix 17 of SUR. Different SUR1 mutations at position 1379 in nucleotide binding fold 2 (NBF2), e.g. the mutation R1379C, have been shown to result in an increased ATPase activity of SUR1 and have been detected in patients with transient neonatal diabetes (de Wet et al., 2007, Proc Natl Acad Sci 104, 18988-18992; Babenko et al., 2006, N Engl J Med 355, 456-466, Vaxillaire et al., 2007, Diabetes 56, 737-741). By analysis of different apoptotic parameters, we now show that the mutation R1379C significantly enhances SUR1specific 17β-estradiol-induced apoptosis in recombinant HEK293 cells. In addition, the interaction of 17β-estradiol with SUR1 and SUR1(R1379C) was compared in binding studies using the sulfonylurea 3H-glibenclamide as the radioligand. These assays, performed in the absence and in the presence of MgATP, revealed no significant differences between SUR1-wildtype and the mutant. Our data suggest that the mutation R1379C not directly effects estradiol binding to SUR. On the other hand, the observation that cells expressing the mutant SUR1(R1379C) are characterized by an increased susceptibility to 17β-estradiol-treatment indicates that NBF2 might plays an important role in the induction of apoptosis. These results may suggest that single nucleotide polymorphisms in the SUR1 (ABCC8) gene can not only confer an increased risk for the development of diabetes via affecting KATP channel function but possibly also via inducing pathophysiological changes in β-cell mass. 1. Institut für Experimentelle und Klinische Pharmakologie und Toxikologie, Medizinische Fakultät der Universität Tübingen, Tübingen

144 Suppression of KATP channel activity protects pancreatic beta cells against oxidative stress by upregulation of antioxidant enzymes Gier B. (1), Krippeit-Drews P. (1), Aguilar-Bryan L. (2), Bryan J. (2), Drews G. (1), Düfer M. (1) Oxidative stress and inflammation crucially contribute to the loss of beta cell function and mass during the development of type 2 diabetes mellitus. We have shown previously that genetic ablation of KATP channels renders pancreatic beta cells less sensitve to oxidative stress. ROS-induced impairment of beta cell function and viability is less pronounced in islet cells from mice lacking functional KATP channels (SUR1-KO) vs wildtype (WT) cells. The aim of the study was to elucidate if this protective effect could be mimicked by pharmalogical inhibition of KATP channels as well as to identify the underlying protective mechanisms. In WT beta cells one hour of incubation with 10 µM H2O2 increased the number of caspase3-positive beta cells from 9.9±2.0% to 28.2±4.1% (n=3, P<0.01). In SUR1-KO beta cells basal apoptosis was increased compared to WT cells, but cell viability was not affected by the addition of 10 µM H2O2. Preincubating WT islet cells with 25 µM tolbutamide for 4 hours augmented basal apoptosis similar to the situation in SUR1-KO cells (20.0±1.3% vs 21.2±1.9%, n=3), but prevented the deleterious effect of 10 µM H2O2. To evaluate potential protective mechanism, the activity of antioxidant enzymes in WT vs SUR1-KO islets was determined. The activities of the most important antioxidant enzymes were increased in SUR1-KO and WT islets. The superoxide dismutase (SOD) activity was 36±4 vs 24±3 U/mg protein (n=4, P<0.05) in SUR1-KO vs WT islets. Furthermore, glutathione peroxidase and catalase activities were approximately doubled in SUR1-KO vs WT islets (0.37±0.02 vs 0.2±0.02 U/mg protein, n=5, P<0.05; 25±1 vs 14±1 mU/mg protein, n=3, P<0.001). Importantly, treatment of WT islets with tolbutamide (100 µM, 4 hours) also augmented SOD activity vs controls (32±3 vs 24±2 U/mg protein, n=6, P<0.05). These data show that limiting KATP channel activity upregulates the activity of antioxidant enzymes thereby protecting beta cells from an oxidative insult. This is of great relevance due to the fact that antioxidant defence mechanisms are poorly pronounced in pancreatic beta cells. Consequently, KATP channel inhibition may be an interesting approach to protect beta cells against oxidative stress, thus delaying or preventing the development of diabetes mellitus. 1. Pharmazeutisches Institut, Eberhard-Karls-Universität, Tübingen, Germany, 2. Pacific Northwest Diabetes Research Institute, Seattle, WA, United States

145 Modulation of glucose stimulated insulin release by blockade of voltage gated K channels Raasch W. (1), Finol-Urdaneta R.K. (2), French R.J. (2), Terlau H. (1) The amount of insulin secretion of pancreatic beta-cells depends on the electrical activity of these cells. Upon a glucose stimulus ATP sensitive potassium channels (KATP) are closed, which results in a depolarization, calcium entry and insulin secretion with an amount of insulin that is directly coupled to beta-cell electric activity. In addition to ATPsensitive currents, which are the molecular target of common antidiabetic drugs such as sulfonylureas calcium-activated (KCa) and voltage activated (Kv) potassium currents are also known to modulate electrical activity of beta-cells and thereby insulin secretion. Although several Kv channels have been identified to be present in islets and/or pancreatic beta-cells the molecular identities of the Kv channels involved in the

34

regulation of glucose-dependent insulin secretion are only partially known. Here we report that a peptide from a cone snail venom identified to block Kv channels modulates glucose induced insulin secretion in vitro and in vivo. The peptide was injected i.v. into rats prior to an oral glucose tolerance test (OGTT) and compared to a preincubation with glibencamide, a KATP antagonist interacting with sulfonylurea receptor subunits of the channel complex. Upon glucose stimulation the peptide pretreatment resulted in a transient increase of insulin release followed by a transient reduction in the blood glucose increase. Glibenclamide also resulted in a reduction of the blood glucose increase but also hypoglycemia under control condition bfore the glucose stimulus was observed. To exclude a central nervous induced regulation or adaptation during the peptide treatment glucose was continuously applied to decerebrated rats and the blood glucose and insulin was measured. During these “glucose clamp” experiments the glucose induced glucose increase for control and peptide treated animals is identical for about the first 15 min of the glucose stimulus. Nevertheless, the steady state level of the blood glucose is significantly reduced for the peptide treated animals. These results indicate that a modulation of Kv channels can result in a modulation of glucose stimulated insulin release that can be an alternative strategy to treat metabolic disorders like Type-II diabetes with a low risk of hypoglycemia. 1. Dept. of Pharmacology and Toxicology, University of Luebeck, Germany, 2. Dept. of Physiology and Biophysics, University of Calgary, Canada

148 Mechanism of allosteric SUR1 Kir6.2 interaction Schwanstecher M. (1), Pieper B. (1), Beyer M. (1), Schulz M. (1), Schwanstecher C. (1) Nucleotide control of SUR1/Kir6.2 channels has been shown to be critically affected by point mutations within the sulfonylurea binding region of the second transmembrane domain of SUR1. The aim of this study was to further analyze this effect. For this purpose SUR1 and Kir6.2 mutants were generated, sequenced to verify PCR fidelity, transiently expressed in COS1-cells and analyzed using binding studies and the insideout mode of the patch-clamp technique. First, we assessed the impact of a series of point mutations on the negative allosteric interaction between the sites for sulfonylureas and potassium channel openers. Then, the effect of these mutations was tested on closure/activation of SUR1/Kir6.2 channels through ATP or MgADP, respectively. Results showed a close correlation between the mutants´ influ-ence on the negative allosteric interaction between the sites for sulfonylureas and potassium channel openers and their potency to either weaken ATP-induced channel closure or enhance MgADPinduced channel activation. These data provide further support for a complex model of allosteric channel control through at least two non-catalytic nucleotide sites on SUR1. The study suggests close local association of these sites with the re-ceptor sites for channel opening and closing drugs. 1. Molecular Pharmacology and Toxicology, Technische Universität Braunschweig, Germany

146 Influence of bile acids on stimulus-secretion coupling in pancreatic beta cells Hörth K. (1), Krippeit-Drews P. (1), Düfer M. (1), Drews G. (1) Bile acids (BAs) play an important role in digestion and cholesterol metabolism. In addition, they act as signalling molecules in several metabolic pathways. Especially in the liver BAs affect insulin action and thus interfere with glucose homeostasis.To test whether BAs directly influence pancreatic beta cell function, we studied the effects of BAs on stimulus-secretion coupling.Taurochenodeoxycholic acid (Ta) in low concentrations (250-500 nM) led to significant increase of the cytosolic calcium concentration ([Ca2+]c) (239% ± 56%, n=9, p<0.05) in isolated islet cells perifused with 15 mM glucose. The Ta-induced rise in [Ca2+]c was not influenced by the SERCA blocker thapsigargin (n=5), but completely suppressed by the L-type Ca2+ channel blocker nifedipine (n=4). This suggests that Ta activates Ca2+ influx. Furthermore, Ta (10 µM) induced calcium oscillations (n=5) at a sub-threshold glucose concentration (6 mM). To analyse the underlying mechanism, we tested the influence of Ta on the membrane potential (Vm) under similar conditions. Decreasing the glucose concentration from 10 mM to 6 mM hyperpolarized Vm and action potentials ceased. The subsequent addition of Ta (10 µM) depolarized Vm and action potentials reappeared (n=8, p<0.001).As BAs are ligands of the nuclear farnesoid X receptor (FXR), we tested whether the effects of Ta on beta cell stimulus-secretion coupling are mediated by this receptor. The specific FXR agonist GW4064 had very similar effects as Ta and acted in the same concentration range. GW4064 increased [Ca2+]c (197% ± 13%, n=13, p<0.001) at a supra-threshold glucose concentration (15 mM) and induced Ca2+ oscillations and action potentials at the sub-threshold glucose concentration (6 mM). These stimulating effects of Ta and GW4064 could be blocked by the FXR antagonist guggulsterone (n=9 and n=5, respectively).In conclusion, BAs affect the stimulus-secretion coupling in pancreatic beta cells at physiologic concentrations. Our data disclose a new signalling pathway in beta cells possibly mediated by the FXR receptor. 1. Pharmazeutisches Institut, Eberhard-Karls-Universität, Tübingen, Germany

147 Enhanced cathepsin k release from osteoclasts is linked to idiopathic osteoporosis in mice with BK channel ablation Sausbier M. (1), Dullin C. (2,4), Kabagema C. (1), Flockerzie K. (1), Haberthür D. (3), Wessels J. (4), Alves F. (4), Schittny J. C. (3), Neuhuber W. (5), Ruth P. (1), Sausbier U. (1) Osteoporosis is characterized by low bone mass accompanied with an increased bone fragility. This bone loss is due to an imbalance between bone resorption and formation. The osteoclastic bone resorption is based on demineralisation of inorganic bone components by Cathepsin K secretion from osteoclasts. Their activity is regulated by sRANKL and osteoprotegerin (OPG), which are both released from osteoblasts and lymphocytes. Secretion and motility of osteoclasts is also regulated by cytosolic Ca2+. The large conductance, voltage and Ca2+-activated (BK) K+ channel links membrane depolarization and local increases in cytosolic calcium to hyperpolarizing K+ outward currents in many cell types. We found this channel type to be expressed in osteoclasts but not in osteoblasts and osteocytes. To analyze the functional relevance of BK channel in bone and to evaluate its impact in idiopathic osteoporosis, we analyzed juvenile female wild-type (WT) and BK-deficient (BK-/-) mice. Serum levels of sRANKL, OPG, estrogene, Ca2+ and triiod-thyronine were not altered in BK-/- females when compared to WT. Moreover, the relative number of osteoclasts in femur and tibia were not significantly different in BK-/- compared to WT. However, serum Cathepsin K levels in juvenile BK-/- females were elevated two-fold when compared to WT littermates. The increased release of Cathepsin K in BK-/- mice was linked to an osteoporotic phenotype as demonstrated by flat-panel volumetric computed tomography and synchrotron X-ray computer tomography. These results suggest that BK channel is involved in controlling resorptive osteoclast activity. 1. Pharmakologie & Toxikologie, Institut für Pharmazie der Universität, Tübingen, Germany, 2. Diagnostische Radiologie, Uni-Klinikum, Göttingen, Germany, 3. Institut für Anatomie der Universität, Bern, Switzerland, 4. Hämatologie und Onkologie, UniKlinikum, Göttingen, Germany, 5. Institut für Anatomie der Universität, ErlangenNürnberg, Germany

149 The position of a methyl-group in 6-chloro-3-alkylamino-4H-thieno[3,2-e]-1,2,4thiadiazine 1,1-dioxide derivatives determines agonism or antagonism of compounds in calf tracheal KATP-channels Kästner L. (1), Grittner D. (1), Rood A. (1), Hansen J.B. (2), Lemoine H. (1) Nielsen et al. (J Med Chem 45: 4171, 2002) reported on new KATPH H N N R channel openers derived from diazoxide characterized by high Cl affinities (KD: 10 to 1000 nM) for SUR1-type KATP-channels. In rat N S S aortic smooth muscle cells, compounds had turned out to act as O O agonists or antagonists dependent on the structure of the side chain (R) (Lemoine et al., this journal 375: 45, 2007). That is why we were interested in the regulation of tracheal KATP-channels by the compounds. Calf tracheal smooth muscle was dissected under sterile conditions, disaggregated by the use of 0.5 g/l collagenase and 0.05 g/l pronase E; cells were cultivated with DMEM (7% CO2) until they reached confluency. Cells of early passages (P1 and P2) were cultivated on 12-well strips (96well formate, Greiner) and labelled with the membrane potential dye DiBAC4(3) (5 µM; excitation 488 nm, emission > 515 nm) in a salt solution composed of (mM) 120 NaCl, 2.0 KCl, 1.0 MgCl2, 2.0 CaCl2, 5 glucose buffered with 20 HEPES (pH 7.4). Hyperpolarisation indicated agonistic, depolarisation (after prestimulation with 0.1 mM diazoxide) antagonistic effects. Results are listed in the Table. Functionally, compounds can be devided in 2 groups: side chains with a quaternary carbon acted as agonists (maximum hyperpolarisation: 64 to 84 % of the effect of 10 µM rilmakalim), those with a tertiary carbon acted as antagonists (maximum depolarisation: 64 to 88 % of control). Note that NNC55,0411 and NNC55,0153 are skeletal isomers. agonistic activity antagonistic activity NNC NNC pEC50±ASD pIC50±ASD R R 55,... 55,... Max (%) (n) Max (%) (n) 7.04±0.05 6.81±0.04 0411 0153 64.3±2.7 (16) 64.5±2.2 (13) 0462 0177 6.67±0.04 6.89±0.05 (C54) (C50) 79.9±2.4 (15) 74.1±3.0 (21) 0467 0047 6.89±0.06 6.48±0.03 (C55) (C51) 78.9±3.2 (16) 88.2±2.2 (16) 0463 0180 6.16±0.08 7.08±0.05 (C56) (C52) 82.0±5.9 (13) 84.2±2.9 (13) 1. Mol. Drug-Research, Lasermedicine, Heinrich-Heine-Universität, D-40225 Düsseldorf, 2. Novo Nordisk Research, DK 2760, Måløv, Denmark

150 The role of AMPK in beta cell stimulus-secretion coupling Noack K. (1), Krippeit-Drews P. (1), Düfer M. (1), Drews G. (1) The AMP-activated protein kinase (AMPK) is known to be an effective target to improve glucose homeostasis by reducing hepatic gluconeogenesis and inducing peripheral glucose utilization. The aim of this study was to elucidate effects of AMPK activation on pancreatic beta cell function. AICAR (500 µM throughout), a well established AMPK activator, significantly increased glucose-induced (10 mM) insulin secretion of mouse islets in vitro from 0.55±0.13 ng/(islet*h) to 1.37±0.26 ng/(islet*h) (n=10, P<0.05). In 3 mM glucose AICAR was without effect (n=10). The cytosolic Ca2+ concentration ([Ca2+]c) in the presence of 10 mM glucose was augmented by 62±15% during application of AICAR (n=9, P<0.01). At the threshold concentration of 6 mM glucose AICAR induced an increase in [Ca2+]c in 5 out of 10 cells. In the perforated-patch configuration KATP current was virtually zero in 15 mM glucose. A low concentration of diazoxide (20-40 µM) reopened KATP channels to give a current of 5.8±0.7 pA in response to a 10 mV voltage-step. AICAR reduced this current to 4.4±0.9 pA (n=4, P<0.05). The same conditions in the current-clamp mode revealed a membrane potential of -61±3 mV, which depolarized to -57±2 mV during AICAR treatment (n=5, P<0.05). Furthermore, AICAR increased glucose-induced (15 mM) electrical activity (quantified as the fraction of plateau phase) from 49±5 % to 75±9 % (n=5, P<0.05). We suggest that AICAR in beta cells elevates cellular ATP content via activation of AMPK thereby potentiating glucose-induced insulin secretion due to a reduction in KATP channel activity. Thus, specific activation of AMPK is likely to be useful in the treatment of type 2 diabetes mellitus, because it activates stimulus-secretion-coupling in beta cells besides the improvement of insulin sensitivity. 1. Pharmazeutisches Institut, Eberhard-Karls-Universität, Tübingen, Germany

35

151 Blood pressure normalizing doses of an AT1-blocker improve glucose utilization and weight gain in rats with metabolic syndrome as potent as the combination AT1-blocker/ACE-inhibitor Miesel A, (1), Müller H, (1), Dominiak P, (1), Raasch W, (1) Many patients suffer from so-called metabolic syndrome. AT1-blockers and ACE inhibitors are well established in the therapy of high blood pressure and also provide additional clinical benefits by reducing the development of type-2 diabetes. Combination therapies are becoming increasingly important for the treatment of high blood pressure, however, little is know whether double blockade of AT1-receptors and ACE will also exert synergistic metabolic effects. We treated hypertensive rats that were additionally fed with a high caloric diet (cafeteria diet) for 12 weeks with telmisartan (TEL, 8 mg/kg), ramipril (RAM, 4 mg/kg) or the combination (TEL/RAM, 8+4 mg/kg) and investigated changes in glucose homeostasis. The glucose utilization was determined via an oral glucose tolerance test (OGTT) and an insulin tolerance test (ITT). Blood pressure was reduced by telmisartan below normotensive values. When TEL was combined with RAM, no additional blood pressure reduction was observed. RAM alone was less potent. Interestingly, even though the energy intake was slightly increased in the TEL and TEL/RAM groups in comparison to RAM alone or controls, the bodyweight gain during the course of the diet was markedly ameliorated in the TEL and TEL/RAM arms. RAM alone was less potent in preventing bodyweight increases. In addition, plasma leptin levels of TEL- and TEL/RAM-treated rats were reduced during the period of 12 weeks. Baseline levels of glucose were not reduced by TEL, TEL/RAM or RAM within the treatment period. Insulin levels, however, were significantly reduced by TEL and TEL/RAM, but not by RAM. The observed glucosuria was markedly diminished by TEL and TEL/RAM and less by RAM. In response to glucose challenge, insulin levels were comparably reduced by TEL and TEL/RAM but not by RAM in order to maintain plasma glucose almost on control levels. The insulin tolerance test revealed that glucose utilization was equieffectively improved by TEL and TEL/RAM. In conclusion, the combination of TEL and RAM did not reveal better efficacies on bodyweight regulation and glucose homeostasis than telmisartan alone. However, TEL was clearly more effective than RAM in improving metabolic parameters including bodyweight gain and glucose utilization and it is speculated that observed effects may not be enhanced by increasing the dose of RAM. 1. Institute of exp. and clin Pharmacology&Toxicology, University Lübeck, Germany

152 Effect of endogenous NO and pentaerythritol tetranitrate on myocardial AT-2 receptor expression in-vivo Dao V.T. (1), Suvorava T. (1), Kocgirli O. (1), Agouri S. (1), Oppermann M. (1), Balz V. (2), Kojda G. (1) We hypothesized that pentaerythritol tetranitrate (PETN) and endothelial nitric oxide (NO) might impact on the expression of angiotensin (AT) type 1 (AT-1) and type 2 (AT2) receptors. We generated mice with an endothelial-specific overexpression of endothelial NO·-synthase (eNOS) using the Tie-2 promotor and backcrossed these mice to the C57BL/6 background. Two of these lines were characterized by eNOS-western blot analyses and blood pressure measurements in comparison to transgene negative littermates. In addition, C57Bl/6 mice were fed with either 6 or 60 mg PETN/kg body weight/day for 4 weeks. Analysis of line 1 of transgenic eNOS mice (1-eNOS++) showed a 2.3±0.15 fold higher aortic expression of eNOS and a reduction of blood pressure to 109.6±2.0 mmHg (P<0.01, n=4-6). Analysis of line 2 of transgenic eNOS mice (2eNOS++) showed a 3.3±0.3 fold higher aortic expression of eNOS and a reduction of blood pressure to 105.0±3.0 mmHg (n=6, p<0.01). Treatment of 2-eNOS++ with the NOS-inhibitor L-nitroarginine (L-NAME) for 30 days completely inhibited the difference in blood pressure suggesting that the reduction of blood pressure in transgenic mice was caused by an increased bioavailability of endogenous NO. In lungs and left ventricular myocardium of 2-eNOS++ the expression of AT-1-receptors was similar to transgene negative littermates (P>0.05, n=8). In striking contrast, the expression of AT-2 receptors was increased by eNOS overexpression in a gene-dose-dependent manner in the myocardium. In 1-eNOS++ and 2-eNOS++ this increase was significantly higher vs. control (P<0.05 and P<0.01, respectively). In lung tissue the AT-2 receptor expression was significantly increased in both lines vs. control (P<0.05). In addition the AT-2 receptor expression of L-NAME-fed mice was decreased in heart tissue of eNOS transgenic mice (n=4, P<0.05). Furthermore, in-vitro studies with the NO-Donor SNitroso-N-Acetyl-D,L-Penicillamin incubated porcine aortic endothelial cells resulted in significant higher expression levels of the AT-2 receptor (P<0.05). Preliminary experiments with PETN-fed mice showed a significant increase in AT-2-receptor expression in myocardial tissue (P<0.05) whereas the expression of the AT-1 receptors did not change (P>0.05) neither in myocardial nor in aortic tissue. These results show that endogenous NO and NO-Donors can upregulate vascular AT-2 receptor in-vivo. This newly discovered regulation might contribute to vasoprotective effects of NO. 1. Dept. of Pharmacology, Heinrich Heine University, Duesseldorf, Germany, 2. Dept. of Otorhinolaryngology, Heinrich Heine University, Duesseldorf, Germany

153 Posttranscriptional upregulation of COX-II and PAI-1 by angiotensin II involves an increased activity of the mRNA-binding protein HuR in vivo and in vitro: implications for renal fibrosis Doller A. (1), Gauer S. (2), Höcherl K. (3), Subkowiak E. (2), Geiger H. (2), Pfeilschifter J. (1), Eberhardt W. (1) Angiotensin-II (AngII)-induced sclerosis in the kidney is a major complication of chronic hypertension which is mainly accompanied by an upregulation of profibrotic factors including the serine protease inhibitor plasminogen activator inhibitor-1 (PAI-1) and proinflammatory enzymes such as cyclooxygenase-2 (COX-2). Both genes, in addition to a transcriptional regulation, underlay a posttranscriptional control mainly by their AUrich destabilizing elements (AREs), present in the 3´untranslated region of the mRNA. Using a model of AngII-induced hypertensive rats, we demonstrate, that continuous AngII perfusion (400ng/kg/min) between day 7 and 14 caused an accumulation of extracellular matrix (ECM) concomitant with a significant increase in the PAI-1 and COX2 protein -and mRNA levels. Employing cell-fractionation and RNA-pull-down assay

revealed an almost 10-fold and 4-fold increase in the cytoplasmic binding of the ubiquitous RNA-binding protein HuR to mRNAs of COX-2 and PAI-1, respectively. Using actinomycin-D we furthermore demonstrate, an AngII-induced increase in the stability of COX-2 and PAI-1 mRNAs from 2 hours to 8 hours, and from 5 hours to 8 hours, respectively. Mechanistically, AngII induces the export of HuR to the cytoplasm which is strongly abrogated by rottlerin, and valsartan thus indicating the specific involvement of the PKCd and the AngII-type I-(AT1) receptor. Cell fractionation and coimmunoprecipitation furthermore revealed that AngII induces a rapid accumulation of PKCd in the nucleus concomitant with its physical binding to nuclear HuR both, in vitro and in vivo.. HuR-dependent stabilization of ARE-containing mRNAs by PKCd may reflect an important mode of AngII-induced cell responses critically involved in the pathogenesis of hypertensive nephrosclerosis. 1. Dept. of Pharmacology and Toxikology, J. W.Goethe University, Frankfurt, Germany, 2. Dept. of Nephrology, J. W.Goethe University, Frankfurt, Germany, 3. Institute of Physiology, University of Regensburg, Regensburg, Germany

154 Telmisartan inhibits biglycan expression in atherosclerotic plaques in ApoEdeficient mice Nagy N. (1), Melchior-Becker A. (1), Fischer J.W. (1) Biglycan is a small leucine-rich proteoglycan (SLRP) secreted by vascular smooth muscle cells (SMC) into the extracellular matrix (ECM) during remodeling of the artery wall. It has recently been shown that biglycan accumulation in the neointima precedes the retention of lipids and accumulation of macrophages during early atherosclerosis in human coronary arteries. Biglycan is therefore considered a pro-atherogenic proteoglycan that might play a key-role in the “response to retention hypothesis” of atherogenesis. The aim of the present study was to determine whether telmisartan affects biglycan accumulation and general extracellular matrix (ECM) composition in a murine model of accelerated atherosclerosis. ApoE-deficient mice on Western-Diet were chronically treated with telmisartan (10 mg/kg, 12 weeks). Since this dose of telmisartan caused lowering of mean arterial blood pressure, hydralazine (500 mg/dl drinking water) was applied as a control for blood pressure dependent effects. Furthermore, simvastatin (50 mg/kg) and simvastatin plus telmisartan were given. After 12 weeks of treatment animals were sacrificed and aortic plaque score, plaque morphology and extracellular matrix (ECM) as well as cellular plaque composition were analysed in the aortic root. Biglycan accumulation in the aorta and the aortic root was significantly inhibited by telmisartan as determined by Western blotting and immunohistochemistry, respectively. Decorin a closely related SLRP was not affected. Furthermore, telmisartan inhibited aortic plaque score and aortic root plaque size compared to mice receiving hydralazine whereas simvastatin had no effect. The amount of SMC, macrophages and collagen in the plaques was not affected by either treatment as determined by αSM-actin-, MAC2and Sirius red staining. Interestingly, the combination of telmisartan plus simvastatin increased the mRNA expression of tissue transglutaminase-1 and procollagen lysylhydroxylase-1 suggesting increased matrix crosslinking. The current study shows that biglycan accumulation in aortic lesions is responsive to telmisartan treatment which might contribute to the anti-atherogenic effects of telmisartan. 1. Department of Pharmacology, University of Duisburg-Essen, Essen, Germany

155 Long term treatment with angiotensin II does not induce alterations in glucose utilization in obese Zucker rats Müller H. (1), Dominiak P. (1), Raasch W. (1) We recently have found that the reactivity of the hypothalamus-pituitary-adrenal (HPA) stress axis was stimulated in leptin-resistant obese Zucker rats (OZR) by angiotensin II (Ang II) via an adrenal-dependent manner. This HPA-hyperreactivity was associated with an impaired glucose-utilization. This mechanism was suggested to be additionally involved into the potency of the renin-angiotensin II-aldosteron-system to regulate the glucose homeostasis. We aimed to investigate whether long-term treatment with Ang II results in a manifestation of type II diabetes (T2DM) and if yes, whether this finding will be associated with the proposed HPA-dependent mechanism. OZR were treated for 84 d with Ang II (9 µg/h) or saline (control) by using osmotic minipumps. HPA-reactivity was tested by an ACTH-test and the glucose utilization by an oral glucose tolerance test. The efficacy of the Ang II treatment was verified by prominent increases in blood pressure, plasma Ang II and plasma aldosterone. The corticosterone response to the ACTH stimulus was enhanced in Ang II-treated OZR. By no means, long term Ang II treatment induced T2DM. Baseline levels of glucose and insulin were rather reduced. In response to the OGTT, glucose and insulin profiles were not altered in Ang II-treated OZR. However, plasma corticosterone was significant increased after the glucose challenge. We conclude that long term treatment with Ang II did not induce the manifestation of a T2DM even though the stress reactivity is still increased after the 3 months lasting treatment period. Thus, we suggest, that the Ang II stimulus alone is too weak to induce prominent metabolic alterations by enhancing the reactivity of the HPA-axis. The discrepancy to our previous findings is supposed to be related to differences in the duration of the Ang II treatment. 1. Institute of experimental and clinical Pharmacology and Toxicology, Medical University of Lübeck, Germany

156 Keratinocyte-derived VEGF biosynthesis represents a pleiotropic side effect of PPAR agonist troglitazone but not rosiglitazone: implications for diabetesimpaired skin repair Goren I. (1), Dißman J. P. (1), Schiefelbein D. (1), Seitz O. (1), Sader R. (2), Pfeilschifter J. (1), Frank S. (1) The peroxisome proliferator-activated receptors (PPARs) represent pharmacological target molecules to improve insulin resistance in type 2 diabetes mellitus. Here we assessed a functional connection between pharmacological activation of PPAR and vascular endothelial growth factor (VEGF) expression in keratinocytes and during diabetes-impaired acute skin repair in obese/obese (ob/ob) mice. PPARβ/ agonist 4-[3-

36

[4-acetyl-3-hydroxy-2-propylphenoxy)propoxy]phenoxy]acetic acid (L165,041) and PPAR agonists ciglitazone and troglitazone, but not rosiglitazone, potently induced VEGF mRNA and protein expression from cultured keratinocytes. Inhibitor studies revealed a strong functional dependence of troglitazone- and L165,041-induced VEGF expression on p38 and p42/44 mitogen-activated protein kinase (MAPK) activation in keratinocytes. Rosiglitazone also induced activation of p38 MAPK but failed to mediate the activation of p42/44 MAPK in the cells. Functional ablation of PPARβ/ and PPAR from keratinocytes by small interfering RNA did not abrogate L165,041- and troglitazone-induced VEGF biosynthesis and suggested VEGF induction as a pleiotropic, PPAR-independent effect of both drugs in the cells. In accordance with the in vitro situation, we found activated p38 MAPK in wound keratinocytes from acute wounds of rosiglitazone- and troglitazone-treated diabetic obese/obese mice, whereas keratinocyte-specific VEGF protein signals were only prominent upon troglitazone treatment. In summary, our data from cell culture and wound healing experiments suggested p38 MAPK activation as a side effect of thiazolidinediones; however, only troglitazone, but not rosiglitazone, seemed to translate p38 MAPK activation into a PPAR-independent induction of VEGF from keratinocytes. 1. Dept. of Pharmacology, J. W. Goethe University, Frankfurt, Germany, 2. Zentrum der Chirugie, J. W. Goethe University, Frankfurt, Germany

157 Relevance of triggering and amplifying signals for the first phase of insulin secretion Rustenbeck I. (1), Hatlapatka K. (1), Willenborg M. (1) The sequence of KATP channel closure, membrane depolarization and influx of Ca2+ via voltage-dependent Ca2+ channels is widely accepted to initiate insulin secretion (triggering). On the other hand, about 70% of the amount of secreted insulin are estimated to be relased via KATP channel-independent signalling (amplification), mostly during the second phase. Here, the relevance of triggering vs amplification for the full expression of the first phase of insulin secretion was explored. NMRI mouse islets and beta cells were used to measure insulin secretion by perifusion and ELISA, membrane potential by patch clamping and cytosolic free calcium concentration ([Ca2+]i) by microfluorimetry. It was observed that the depolarization by KCl corresponded closely to the theoretical values. In contrast to physiological or pharmacological KATP channel block, KCl depolarization did not induce action potential spiking. The depolarization by 15 mmol/l K+ (21 mV) corresponded to the plateau depolarization by 50 or 500 µM tolbutamide, that by 40 mmol/l K+ (41 mV) corresponded to the peak values of tolbutamide-induced action potentials. Nifedipine and diazoxide abolished action potential spiking but had only a minimal effect on the depolarization by KCl. 40 mmol/l K+ induced a massive secretory response in the presence of 5 mmol/l glucose, whereas 15 mmol/l K+, like 50 µmol/l tolbutamide was only moderately effective, even though an increase in [Ca2+]i could be clearly shown. Raising glucose from 5 to 10 mmol/l in the continued presence of 15 mmol/l K+ revealed an enhanced biphasic response. The depolarization pattern of this combination could be mimicked by combining basal glucose with 15 mmol/l K+ and 50 µmol/l tolbutamide, however, the secretory response was clearly inferior. In conclusion, the depolarization by 15 mmol/l, but not by 40 mmol/l K+ corresponds to that of KATP channel closure and is only moderately effective to initiate insulin secretion. KATP channel-independent signalling may play a major role for the full expression of the first phase. 1. Institute of Pharmacology, University of Braunschweig, Germany

158 Interaction of TGF-ß and insulin/IGF-I stimulation in human bronchial fibroblasts and epithelial cells Mayer P. (1), Mering S. (1,2), Warnken M. (2), Reitzenstein U. (2), Racké K. (2) Insulin may act as a growth factor on certain cell types. This is an ongoing concern for the inhalative use of insulin since several patients were exposed, and some of them developed bronchial tumours. The causality remains unclear. Therefore we have investigated the effects of insulin and IGF-I (the receptor of the latter may also be activated by insulin) on bronchial epithelial cells and bronchial fibroblasts in respect to proliferation, alterations in gene expression and morphological changes. In human bronchial fibroblasts (cell line Hel299) insulin and IGF-I caused a pronounced proliferative response (see also abstracts of Warnken et al. and Reitzenstein et al. at this meeting) which was surprisingly accompanied by only a weak induction of common immediate early genes. Accordingly, late response genes were also hardly altered. In contrast, TGF-ß induced several genes, among them the gene for IGF-I. In the presence of TGF-ß it could be shown that IGF-I itself enhanced its own gene expression in these cells. The human bronchial epithelial cell line H292 also responded to insulin with proliferation, albeit to a markedly lesser extent than the Hel299 fibroblasts. To date no major changes in gene expression were detected in H292 cells in response to insulin alone. However, in response to TGF-ß H292 cells underwent a pronounced morphological change accompanied by down-regulation of chemokine CXCL1 expression and induction of smooth muscle markers. These morphological changes (probably reflecting the switch to a myo-epithelial phenotype) were clearly suppressed by insulin, and the down-regulation of CXCL1 was also concentration-dependently counteracted (although not fully) by insulin in the nanomolar range. This indicates that insulin exerts pronounced and direct effects on bronchial epithelial cells, the cell type which gives rise to bronchial carcinomas. Importantly, these effects only became obvious in conjunction with another biologically active peptide (TGF-ß). Although it is unknown to date how myo-epithelial differentiation and chemokine expression are related to tumour formation our findings indicate that insulin may influence the differentiation state of bronchial epithelial cells when embedded in a wider framework of local signalling peptides in the lung. This has to be considered when investigating the influence of insulin on bronchial carcinogenesis. 1. Institute for Drugs and Medical Devices, Bonn, Germany, 2. Institute for Pharmacology and Toxicology, University of Bonn, Germany

159 Therapeutic and protective effects of lisinopril and candesartan on ceruleininduced chronic pancreatitis and and pancreatic fibrosis in rats Ghoneim M.T. (1), Matta M.M. (1), Baraka A.L. (1), Shaat E.A. (2), Yakout D.Y. (1) In the present study, a rat model of cerulein-induced chronic pancreatitis and fibrosis (CP&F) was used to examine the possible prophylactic and therapeutic effects of the angiotensin converting enzyme inhibitor, lisinopril, and the angiotensin receptor blocker candesartan either alone or in combination, on the development of (CP&F). Chronic pancreatitis and fibrosis (CP&F) was induced by i.p injection of cerulein in a dose of 50 ug/kg three times a week for three weeks. Drugs were given as follows: lisinopril or candesartan were given orally in doses of 2.5 mg/Kg and 1.5 mg/kg respectively daily for 4 weeks starting from the first day of cerulein administration (prophylactic effect) or in other groups they were given in the same dose for two weeks starting from the third week of cerulein administration (therapeutic effect). The effect of combined treatment of losartan and candesartan given in the same schedule and dosing was also assessed. Pancreatic tissue homogenates were used to measure the total pancreatic protein concentration, hydroxyproline (HPO) content and myeloperoxidase activity (MPO). Serum was used to measure Transforming Growth Factor B1 (TGF-b1) concentration and serum amylase activity in all groups. Cerulein-induced (CP&F) revealed a significant increase in pancreatic (MPO) activity, (HPO) content, serum TGF-B1 and amylase activity. The therapeutic and prophylactic treatment of (CP&F) with lisinopril or candesartan produced significant reduction in pancreatic (MPO) activity & (HPO) content, and serum TGF-B1 concentration and amylase activity, while the prophylactic treatment produced a more reduction in the above mentioned parameters. The therapeutic and prophylactic effect of combined treatment with lisinopril and candesartan was more superior than either treatments with any of the drugs alone. Conclusion: Treatment with lisinopril or candesartan can inhibit chronic pancreatic inflammation and fibrotic progression induced by cerulein. Combination of the two drugs is more effective against chronic pancreatic inflammation and fibrosis. 1. Pharmacology Dept., Faculty of Medicine, 51521 El-Messalla, Alexandria Uni, Egypt., 2. Biochemistry Dept., Faculty of Medicine, 51521 El-Messalla, Alexandria Uni, Egypt.

160 The impact of ADMA (asymmetric dimethylarginine) on diabetes Galal O. (1), Verspohl E.J. (1) L-Arginine is a semi essential cationic amino acid. ADMA acts as a competitive inhibitor for NOS (NO synthetase) and inhibits production of NO from L-arginine. The enzymatic activity NOS is regulated by the ratio between the concentrations of L-arginine and ADMA. ADMA levels are increased during diabetes and may be responsible for secondary effects of diabetes such as circulation problems and renal dysfunction. ADMA may also be linked to diabetes directly. In urethane-anaesthetized rats blood glucose (Glucoquant®) and plasma insulin (RIA) were determined. In vitro insulin release from INS-1 cells was determined. A bolus of ADMA (3.8 mg/kg) increases blood glucose and plasma insulin for several hours. ADMA also enhances the ratio blood glucose/plasma insulin for the first 60 min which indicates an increase of insulin resistance and, therefore, a diabetic situation induced by ADMA. L-Arginine increases blood glucose marginally and clearly decreases plasma insulin levels which is not expected from in vitro data of this cationic and therefore depolarizing amino acid. The effect of L-arginine (increase in blood glucose) is reversed by the pretreatment with an ADMA bolus. ADMA increases glucose (8.3 mM)-induced insulin secretion in vitro in a concentrationdependent manner (0.05 to 10 mM). L-arginine is not able to reduce insulin release as may have been expected from in vivo data (shown above), but in contrast, there is a marginal insulinotropic effect of L-arginine probably due to its cationic and therefore depolarizing effect. This lack of an insulin reducing effect may be due to a lack of NOsynthetase in INS-1 cells. ADMA is not able to interfere with the effect of L-arginine in vitro in contrast to the in vivo experiments shown before. Conclusion: ADMA appears to induce a diabetogenic situation (insulin resistance) in vivo although it possesses a direct insulinotropic effect in vitro. NO may be involved in the L-arginine effect since a 15 min pretreatment with ADMA is able to reverse the in vivo L-arginine effect. 1. Dept. of Pharmacology (Inst. Medicinal Chem.), University of Muenster, Germany

161 Impact of the 5-HT3 receptor channel system for insulin secretion and interaction of ginger extracts Verspohl E.J. (1), Feistel B. (2), Heimes K. (1) The relevance of serotonin and in particular that of 5-HT3 receptors is unequivocal with respect to emetic/antiemetic effects, visceral nociception, and contraction/relaxation of smooth muscles, but it is controversial with respect to their impact on insulin and antidiabetic effects. Therefore, the effects of tropisetrone (5-HT3 receptor antagonist) and various ginger extracts (Zingiber officinale; known to interact with the 5-HT3 receptor channel system) were investigated. Insulin was measured in vivo (rats) and in vitro (INS-1 cells) by RIA. The presence of 5-HT3 binding sites was investigated in INS1 cells by using 0.4 nM [3H]GR65630 in competition displacement experiments. The effect on [14C]guanidinium uptake in N1E-115 cells (model of 5-HT3 effects) and relaxant effects on rat ileum was measured. Serotonin (32 to 500 µM) inhibits insulin release from INS-1 cells which is reversed by tropisetrone (10 to 100 µM) and two different ginger extracts (normal spissum extract and an oily extract with high volatile oil and pungent concentration). Their effects are obvious even in the absence of serotonin but are more pronounced in its presence (doubled to tripled). Specific 5-HT3 binding sites are present in INS-1 cells. The in vitro data with respect to ginger are corroborated by in vivo data on glucose-loaded rats showing that blood glucose is decreased (by approx. 35 %) and plasma insulin is increased (by approx. 10 %). Both the spissum extract and the oily ginger extract are effective in two other models: they inhibit [14C]guanidinium uptake into N1E-115 cells and relax rat ileum both directly and as a serotonin antagonistic effect. Other receptors addressed by ginger are 5-HT2 receptors as demonstrated by using methysergide and ketanserin: they weakly antagonize the serotonin effect as well. It may be concluded that serotonin and in particular the 5-HT3 receptor channel system are involved in modulating insulin release and that tropisetrone as well as various ginger extracts known for their 5-HT3 receptor antagonistic effect can be used to improve the diabetic situation.

37

1. Dept. Pharmacology (Inst. Medicinal Chem.), Universität Münster, Germany, 2. Finzelberg GmbH & Co. KG, Andernach, Germany

162 Triggering and amplification of insulin secretion by dimethyl α-ketoglutarate, a membrane permeable α-ketoglutarate analogue Willenborg M. (1), Panten U. (1), Rustenbeck I. (1) Cytosolic α-ketoglutarate is a potential signalling compound at late steps of stimulussecretion-coupling in the course of insulin secretion induced by glucose and other fuels. This hypothesis was mainly based on the insulin-releasing effect of the membrane permeable ester dimethyl α-ketoglutarate (DMKG) which enters the β-cell and is cleaved to produce cytosolic monomethyl α-ketoglutarate and eventually α-ketoglutarate. By measuring insulin release, KATP channel currents, membrane potential, ATP/ADP ratio and fluorescence of NAD(P)H (reduced pyridine nucleotides) this hypothesis was tested using mouse pancreatic islets and primary β-cells. At a substimulatory glucose concentration (5 mM), 15 mM DMKG produced a sustained insulin release, but no change of the islet ATP/ADP ratio and NAD(P)H fluorescence. In the absence of glucose, however, 15 mM DMKG did not stimulate insulin release although it increased the ATP/ADP ratio and NAD(P)H fluorescence. Consequently, DMKG depolarized the plasma membrane of intact β-cells and generated action potentials in the absence of glucose. However, DMKG had also a direct inhibitory effect on KATP channels as became evident from the use of inside-out membrane patches from β-cells. Finally, DMKG strongly amplified the insulin secretion induced by a maximally effective concentration of the KATP channel-blocking sulfonylurea glipizide in the presence of 5 mM glucose and moderately but significantly amplified the glipizide-induced secretion in the absence of glucose. In conclusion, the insulinotropic effect of DMKG is not sufficient to demonstrate a direct signalling role of cytosolic α-ketoglutarate. Rather, DMKG produces a complex pattern of effects by influencing several signalling pathways in the B-cell. The metabolism of α-ketoglutarate generated intracellularly by ester cleavage contributes to the stimulatory effect both by indirect KATP channel inhibition (via activation of ATP production) and by a KATP channel-independent amplifying effect. In the presence of basal glucose, the direct KATP channel inhibition is of critical importance for the insulin-releasing property of DMKG. 1. Inst. of Pharmacology, Technical University of Braunschweig, Braunschweig, Germany

163 The insulinotropic effect of fluoroquinolones Ghaly H. (1), Kriete K. (1), Sahin S. (1), Pflöger A. (1), Holzgrabe U. (2), Zünkler B.J. (3), Rustenbeck I. (1) Antimicrobial fluoroquinolones induce, with strongly varying frequency, life-threatening hypoglycemias, which is explained by their ability to block KATP channels in pancreatic Bcells and thus to initiate insulin secretion. In apparent contradiction to this, we observed that none of the fluoroquinolones in this study (100 µM of gatifloxacin, moxifloxacin, ciprofloxacin, and a number of fluorophenyl-substituted compounds) initiated insulin secretion of perifused mouse islets when the glucose concentration was basal (5 mM). Only when the glucose concentration was stimulatory by itself (10 mM), the fluoroquinolones enhanced secretion. The fluoroquinolones were ineffective on SUR1 Ko islets, which do not have functional KATP channels. All of these fluoroquinolones depolarized the membrane potential of mouse B-cells (patch-clamping in the whole-cell mode). In contrast to the secretory response the depolarization was virtually irreversible and could not be antagonized by diazoxide. Using metabolically intact B-cells (perforated-patch mode) however, 100 µM of gatifloxacin, ciprofloxacin or moxifloxacin were unable to depolarize when the glucose concentration was 5 mM, whereas other KATP channel blockers (tolbutamide and efaroxan) remained effective. Only at a very high concentration (500 µM) gatifloxacin and moxifloxacin, but not ciprofloxacin induced repetitive depolarizations which could be antagonized by diazoxide. In the presence of 10 mM glucose all fluoroquinolones which enhanced secretion markedly elevated cytosolic calcium concentration ([Ca2+]i). In the presence of 5 mM glucose gatifloxacin and moxifloxacin at 500 µM elevated [Ca2+]i. It is concluded that fluoroquinolones in the clinically relevant concentration range are not initiators, but rather enhancers of glucoseinduced insulin secretion. The block of KATP channels appears necessary but not sufficient to explain the hypoglycemic effect of fluoroquinolones. Measurements of the NAD(P)H fluorescence of pancreatic islets suggest that an interference with mitochondrial energetics is an additional factor which influences the insulinotropic effect of the fluoroquinolones. 1. Institute of Pharmacology and Toxicology, Technical University of Braunschweig, Braunschweig, Germany, 2. Institute of Pharmaceutical Chemistry, University of Würzburg, Würzburg, Germany, 3. Federal Institute for Drugs and Medical Devices, Bonn, Germany

164 Induction of beta-cell apoptosis by the proinflammatory cytokine IL-1beta not involving the apoptosis-inducing dual leucine zipper bearing kinase DLK Klimpel C. (1), Börchers S. (1), Oetjen E. (1) The proinflammatory cytokine IL-1β was shown to contribute to the pathogenesis of diabetes mellitus type 2 through the induction of beta cell apoptosis thereby reducing beta cell mass. We recently demonstrated that over expression of the dual leucine zipper bearing kinase DLK resulted in beta cell apoptosis. In the present study we investigated whether activation of DLK contributes to IL-1β-induced beta cell death. Incubation of the insulin producing beta cell line with IL-1β decreased the viability of the cells in a time- and concentration-dependent manner as measured by the MTT test. In an immunocytochemical analysis using an antibody against the cleaved caspase-3 as hallmark for apoptosis IL-1β increased the number of apoptotic beta cells in a time- and concentration-dependent manner. Over expression of DLK enhanced the number of cleaved caspase-3 positive cells as well. However, over expression of a kinase dead DLK mutant did not prevent the apoptosis-inducing effect of IL-1β. The cytokine activated the DLK downstream kinase JNK in a time-dependent way with a threefold

stimulation detected after 30 min treatment as measured by the phosphorylation of JNK using a phospho-specific antibody in immunoblot analysis. The reduction of endogenous DLK by RNAi did not decrease IL-1β-induced JNK phosphorylation. Consistent with this in an in vitro kinase assay the activity of immunoprecipitated DLK was not activated by IL-1β. Immunoblot analysis using a DLK antibody revealed that IL-1β reduced the amount of cellular DLK in a time dependent manner. A reduction by 20% was already seen after 30 min of treatment. Additional treatment with the proteasome inhibitors lactacystin and epoxomicin did not prevent the IL-1β-induced degradation of DLK, suggesting that IL-1β might interfere with the synthesis of the kinase. Indeed, cycloheximide, an inhibitor of protein synthesis, prohibited the IL-1β-induced loss of DLK. Our data show that IL-1β and DLK induce beta cell apoptosis and that IL-1β action, although the cytokine activates the DLK downstream kinase JNK, does not involve DLK itself. 1. Dept. of Pharmacology, Georg August University, Göttingen, Germany

165 ANP protects against endothelial leakage by influencing endothelial cell contraction signaling Mayer B.A. (1), Bubik M.F. (1), Ammer H. (2), Zahler S. (1), Vollmar A.M. (1), Fürst R. (1) Severe pathologies, such as sepsis or atherosclerosis, are associated with vascular leakiness. An adequate pharmacotherapy directly targeting endothelial permeability is still lacking. The cardiovascular hormone atrial natriuretic peptide (ANP) has recently been shown to exert protective effects on endothelial barrier function in vitro. Moreover, we found that ANP inhibits inflammation (histamine)-induced hyperpermeability in vivo. These results prompted us to elucidate the underlying molecular mechanisms. Pretreatment of HUVECs with ANP (1 µM) rescued the decline of transendothelial electrical resistance caused by histamine (1 µM). We investigated the impact of ANP on a crucial regulator of permeability, the myosin light chain 2 (MLC2) as part of the contractile machinery. Using immunocytochemistry and Western blot analysis, we found that ANP reduces the histamine-evoked phosphorylation of MLC2. Since the small GTPases Rac1 and RhoA play important roles in the regulation of the contraction status, we investigated the influence of ANP on RhoA activation, which is important for the regulation of MLC2 phosphorylation: ANP abrogated the histamine-induced RhoA activation (pull down assay). RhoA is known to inhibit myosin light chain phosphatase. By co-immunoprecipitation, we showed that ANP strengthens the binding of MLC2 to the myosin phosphatase targeting unit (MYPT), which leads to an increased dephosphorylation of MLC2. We have previously demonstrated that ANP activates Rac1 (Circ Res. 2005;96:43-53), which might antagonize the histamine-induced contraction. In fact, in the presence of the Rac1 inhibitor NSC23766 (200 µM) ANP lost its ability to inhibit the histamine-evoked MLC2 phosphorylation (Western blot). Upstream of Rac1, we investigated the influence of ANP on intracellular cAMP levels (ELISA) and the cAMP effector PKA: We found ANP to increase cAMP concentrations. Applying the PKA inhibitors PKI (10 µM) and Rp-cAMPS (100 µM) eliminated the impact of ANP on MLC2 phosphorylation. Our results highlight ANP as a barrier-protecting agent that directly targets key regulators of endothelial permeability: ANP protects against histamineinduced hyperpermeability by induction of the cAMP/PKA pathway, activation of Rac1, and inhibition of RhoA, thus leading to a decreased MLC2 phosphorylation and increased dephosphorylation. These data warrant future investigations on how cAMP production is activated by ANP and the link between Rho inhibition and Rac activity. 1. Department of Pharmacy, Pharmaceutical Biology, Ludwig-Maximilians-University, Munich, Germany, 2. Institute of Pharmacology, Toxicology and Pharmacy, LudwigMaximilians-University, Munich, Germany

166 Estradiol supresses hyaluronic acid synthase expression in human vascular smooth muscle ells Freudenberger T. (1), Dai G. (1), Mayer P. (2), Heim H.-K. (2), Schroer K. (3), Fischer JW. (1) Backround: Hyaluronic acid (HA) is a constituent of the extracellular matrix. HA is produced by three hyaluronic acid synthase (HAS) isoforms, namely HAS1, HAS2 and HAS3. It has been noted that platelet-derived growth factor-BB (PDGF-BB) induces HA synthesis in vascular smooth muscle cells (SMCs) and thereby contributes to formation of HA-rich matrices that promote the proliferative phenotype in SMCs. At the same time estradiol reduces the proliferative capacity of SMCs. However, nothing is known about the effects of estradiol on HAS-expression and HA-secretion. Methods: Human SMCs were synchronized by serum withdrawal for 24 hours and were subsequently treated with PDGF-BB (10 ng/ml) alone or in combination with estradiol (1x10-9 M and 0.1x10-6 M). HAS-mRNA-expression, HA-secretion, proliferation and apoptosis were investigated. Results: Estradiol at both concentrations significantly reduced the PDGFinduced expression of HAS1-mRNA to 57 % ± 15 %, n = 7, p < 0,05 (1x10-9 M) and 58 % ± 16 %, n = 7, p < 0,05 (0.1x10-6 M) as well as Hyaluronidase 1 (Hyal 1) mRNA to 77 % ± 9 %, n = 7, p < 0,05 (1x10-9 M) and 78 % ± 6 %, n = 7, p < 0,05 (0.1x10-6 M). Furthermore, PDGF-induced HAS2-mRNA expression was significantly reduced to 79 % ± 6 % with 0.1x10-6 M estradiol (n = 7, p < 0,05). Concomitantly the PDGF-induced HAsecretion into the medium was significantly reduced to less than 38 % (n = 3, p < 0,01) in the presence of estradiol. Furthermore, the proliferative capacity of SMCs treated with PDGF and estradiol was decreased in a concentration dependent manner as evidenced by 3H-thymidine incorporation (PDGF: 100 %, PDGF+1x10-9 M estradiol: 90 % ± 1 %, PDGF+0.1x10-6 M estradiol: 80 % ± 5 %, p < 0,05, n = 3). Evaluation of caspase-3activity showed a non-significant 2.5 fold increase in the ratio of activated/total caspase3 at 1x10-9 M estradiol compared to PDGF alone (n = 4) and a significant 7-fold increase of this ratio at 0.1x10-6 M estradiol (n = 4, p < 0,05). Conclusion: Estradiol reduced the PDGF-BB-induced proliferation of SMCs. Mechanistically this might be related to reduced expression of HAS1, HAS2 and Hyal1 and a thereby severely reduced HA-secretion into the medium. At the same time estradiol considerably increased capase-3-activity. We therefore hypothesize that estradiol might exert antiproliferative and pro-apoptotic effects that might in part be due to down-regulation of HAS-expression and reduction of HA-secretion in SMCs.

38

1. Dept. of Pharmacology, University Clinics Essen, University Duisburg-Essen, Essen, Germany, 2. Bundesinstitut fuer Arzneimittel ud Medizinprodukte, Bonn, Germany, 3. Dept. of Pharmacology, Heinrich-Heine University, Duesseldorf, Germany

167 Epigenetic mechanisms involved in the inducible expression of the mu opioid receptorgene in T cells Kraus J. (1), Börner C. (1), Lehmann L. (1), Höllt V. (1) Mu opioid receptors are expressed constitutively in many neuronal cells. In contrast, in T lymphocytes, the gene is normally silenced. However, the cytokines interleukin-4 (via the transcription factor STAT6) and tumor necrosis factor (via NFkappaB), as well as activation of the T cell receptor (via NFkappaB, AP-1 and NFAT) induce de novo expression of the gene in T cells. Additionally, we observed that trichostatin A, which is a histone deacetylase inhibitor, and 5- aza-2-deoxycytidine, which is an inhibitor for histone methyl transferases, both induced mRNA of the mu opioid receptor gene in Jurkat T cells. An even stronger induction of the gene was observed, when both drugs were applied together. This suggests that expression of mu opioid receptors in T cells is under epigenetic control. Here we present mechanisms underlying the de-novo induction of the mu opioid receptor gene in Jurkat T cells by interleukin-4 and investigated, if epigenetic mechanisms are involved therein. Employing quantitative realtime RT-PCR it was found that the onset of mu opioid receptor transcription in Jurkat T cells starts between three and eight hours after interleukin-4-stimulation of the cells. Using chromatin immunoprecipitation analysis (CHIP), STAT6 binding to the respective mu opioid receptor promoter regulatory element at nt -997 was observed three, five and 16 hours after cytokine-stimulation. In contrast, unstimulated cells and cells stimulated for one and two hours did not contain STAT6/-997DNA complexes. Simultaneously, a decrease in the binding of MeCP2, a protein that binds to methylated DNA, to the mu opioid receptor core promoter between nt -373 and nt -165 was observed within the first five hours of incubation of the cells with interleukin-4. These data indicate that epigenetic mechanisms are likely to be involved in the IL-4-induced expression of the gene in T cells. 1. Dept. of Pharmacology and Toxicology, Magdeburg University, Magdeburg, Germany

168 Opioid- and cannabinoid-mediated cAMP formation in T lymphocytes inhibits T cell receptor signaling and interleukin-2 production Börner C. (1), Höllt V. (1), Kraus J. (1) Opioids and cannabinoids inhibit functions of activated T lymphocytes, in particular the T cell receptor-triggered induction of interleukin-2. In a cAMP/PKA-dependent manner, the drugs inhibit the protein Lck, which is a master switch that controls signaling downstream of the activated T cell receptor. The first step in the drug-T-cell-interactions is an increase in cAMP caused by opioids and cannabinoids. The effects of the opioids morphine and beta-endorphin are mediated by mu-opioid receptors, while cannabinoids inhibit T cell signaling via CB1 and CB2 receptors, as demonstrated with receptorspecific agonists and antagonists. It is known that these receptors are Gi/o-coupled, which is classically associated with decreased cAMP formation. Here we show that the drugs indeed inhibit cAMP formation within the first fifteen minutes of incubation of Jurkat T cells with the drugs. However, incubation of the cells for longer times strongly increased cAMP levels up to six fold over the control levels. Thus, increased cAMP formation caused by opioids was found between one and eight hours of drug incubation with maximal levels between two and four hours. When using the phosphodiesterase inhibitor IBMX, cAMP levels produced by the opioids remained elevated, suggesting that opioid-incubation of the T cells might lead to activation of phosphodiesterases. Interestingly, the effect of the cannabinoids was long lasting and cAMP levels remained elevated for at least 48 hours of incubation. Both phases of the cAMP regulation produced by opioids and cannabinoids, the initial decrease, as well as the subsequent increase in cAMP, were blocked by pertussistoxine, indicating the involvement of Giproteins. When the drugs were washed out and antagonists were added after 24 h of incubation, we observed an overshoot in cAMP production, indicating that the drugs caused superactivation of adenylyl cylases. These findings help to explain molecular mechanisms of opioid- and cannabinoid-mediated immunosuppression. On the other hand, the inhibitory effect of the drugs on T cell responses may be advantageous in the treatment of certain neuroinflammatory diseases such as multiple sclerosis. 1. Dept. of Pharmacology and Toxicology, Otto-von-Guericke University, Magdeburg, Germany

characterize Th3 type Tregs. CD3+ICOS stimulation induced expression of high levels of FoxP3 comparable to those induced by CD28. The H4R-agonist 4-methylhistamine significanly enhanced gene expression of IL-4 and IFN-γ without affecting IL-2 synthesis and secretion. In sharp contrast FoxP3 expression was nearly completely inhibited by 4methylhistamine at concentrations of 10-50 nM. In this range 4-methylhistamine selectively activates the H4R. In conclusion, this work shows that ICOS may deliver signals for regulatory T cell differentiation that are effectively antagonised by the H4R. This work was supported by grants of the DFG Sz-45/5-2 1. Institute of Pharmacology, Medical School Hannover, Hannover, Germany

170 Akt2, but not Akt1, is crucial for the plasmin-induced migration of immature dendritic cells Li X. (1), Syrovets T. (1), Pitterle K. (1), Simmet T. (1) Inflammation is accompanied by contact activation leading to the generation of the serine protease plasmin. Contact activation takes place in any type of chronic inflammation including atherosclerosis. Consistently, in human atherosclerotic lesions the fibrinolytic activity positively correlates with the severity of coronary lesions. Pivotal roles have been proposed for dendritic cells and the adaptive immune response in atherosclerosis. Using immunohistochemistry we observed that both, plasmin and DC, are abundant in human atherosclerotic lesions, where they colocalize. In addition, we identified plasmin as a potent activator of immature dendritic cells. Within 5 min after stimulation of dendritic cells with plasmin a rapid activation of Akt and an Akt-dependent activation of the ERK1/2 MAP kinase occurred. Three Akt isoforms, which share a high degree of homology, are so far known. Quantitative real-time PCR and RT-PCR analysis of the Akt isoforms expressed by immature dendritic cells revealed that the cells express predominantly Akt1 and less Akt2, but not the Akt3 isoform. Among the two Akt isoforms expressed by dendritic cells, plasmin activates exclusively Akt2, but not Akt1 as shown by immunoprecipitation of tyrosine-phosphorylated proteins. Selective downregulation of Akt2 but not Akt1 in dendritic cells by small hairpin RNA plasmids encoding corresponding siRNA inhibited the plasmin-induced ERK1/2 activation and the plasmininduced chemotactic response of immature dendritic cells. This novel mechanism of dendritic cell activation might be relevant for the function of dendritic cells in chronic inflammation and atherosclerosis. Supported by the DFG. 1. Institute of Pharmacology of Natural Products & Clinical Pharmacology, Ulm University, Ulm, Germany

171 Genetically modified murine dendritic cells engineered to overexpress immunomodulatory cytokines display a protolerogenic status Besche V. (1), Wiechmann N. (1), Trojandt S. (1), Grabbe S. (1), Reske-Kunz A.B. (1), Bros M. (1) Dendritic cells (DCs) in their immatur and semi-mature state are essential for the induction and maintenance of peripheral tolerance. Therefore, DCs resemble attractive targets for the treatment of autoimmune and allergic diseases as well as transplant rejections. Vaccination with DNA has been established as an efficient mean to induce a stable tolerogenic status in DCs both in vitro and in vivo. Due to the refractory state of DCs towards transfection in vitro we have optimized a lentiviral transduction protocol for murine myeloid bone marrow derived DCs yielding transduction rates up to 98%. In this study we have evaluated the potential of the immunomodulatory cytokines IL-10, IL-21 and IL-1RA to induce a stable tolerogenic state in DCs upon their (over)expression. Due to autocrine effects these cytokines mediate a protolerogenic DC phenotype largely resistant to maturation with lipopolysaccharid (LPS) and characterized by diminished expression of MHCII and costimulatory molecules (CD40, CD80, CD86, OX40L). DCs transduced with an IL-1RA expression unit showed enhanced IL-10 expression under basal conditions as compared with control DCs. In contrast to the other DC populations, DCs engineered to de novo express IL-21 did not upregulate IL-10 production in response to stimulation with LPS. In accordance with their protolerogenic phenotype all cytokine (over)expressing DC populations exerted diminished allogenic T cell stimulatory potency and modulated the cytokine profile of DC/T cell cocultures differentially. Ongoing work is focussed on eliciting the potential of the differentially tolerized DC populations to induce regulatory T cells. 1. Dept. of Dermatology, Johannes Gutenberg University, Mainz, Germany

172 169 Molecular mechanisms of T-cell activation: Interference of ICOS- and histamine H4-receptor induced signals on the differentiation of human regulatory T cells Szamel M. (1), Golombek M. (1), Dreikhausen U.E. (1), Seifert R. (1) T-cells are fully activated when they receive at least two signals. The first signal is delivered through the T-cell-receptor/CD3 complex after recognition of the antigen.The second one is provided by costimulatory molecules, like CD28 or the inducible costimulator (ICOS). The histamine H4-receptor (H4R) was suggested to modulate T-cell responses. However, the molecular mechanisms are not well established. Aim of this study was to characterize ICOS and H4R-induced signals during T-cell activation. We compared the effects induced by costimulation of CD28, ICOS, or both in CD3-activated cells. Optimal stimulation via CD3+CD28 induced the expression of both CD25 and ICOS, secretion of IL-2, IFN-γ, and TNF-α, as well as low levels of IL-5 and IL-10. The CD3+CD28+ICOS stimulation further increased secretion of IL-2, IL-10, IFN-γ, TNF-α and IL-5. Also, the surface expression of ICOS and CD25 was highly increased under these conditions. Blockade of either IL-2 or IFN-γ by specific antibodies (Ab) strikingly inhibited cell proliferation and ICOS expression. Anti-IL-2 Ab inhibited secretion of all cytokines. Anti-IFN-γ Ab inhibited secretion of TNF-α and IL-2, and increased secretion of IL-5 and IL-10. These data suggest that ICOS favors differentiation of type 1 regulatory T cells (Tr1), which produce high levels of IL-10. To further assess the role of ICOS in regulatory T-cells (Treg) differentiation, we evaluated the effect of ICOS triggering on secretion of TGF-β1 and expression of Forkhead Box P3 (FoxP3), which

Long term treatment of murine dendritic cells mit a cAMP analogue induces a protolerogenic state Höhn Y. (1), Thamsen M. (1), Wiechmann N. (1), Grabbe S. (1), Reske-Kunz A.B. (1), Bros M. (1) In their activated state dendritic cells (DCs) constitute the most potent antigen presenting cells (APCs) of the immune system. In their immature state, however, DCs vitally contribute to the the maintenance of pheripheral tolerance. Concerning therapeutic interventions in case of allergies, autoimmune diseases as well as transplant rejections several protocols aimed to tolerize DCs by treatment with immunomodulatory agents have been developed. For some of these compounds including prostaglandin E2 involvement of the intracellular secondary messenger cAMP has been demonstrated. In order to eluciate the actual contribution of cAMP induced signalling for DC tolerization we have treated DCs with a stable cAMP analogue (db-cAMP, Dibutyryl-cAMP). Upon long term treatment of myeloid DC progenitors db-cAMP exerted an antiproliferative effect concening the number of immature DCs obtained. However, these immature cAMP-DCs themselves were characterized by unaltered proliferation. Compared with control DCs, cAMP-DCs displayed an altered expression pattern concerning CREB and CREM isoforms and a protolerogenic gene expression pattern. Stimulation of cAMPDCs with lipopolysaccharide (LPS) resulted in a diminished upregulation of costimulatory molecules (CD40, CD80, CD86) and of the proinflammatory cytokine IL12, while immunmodulatory molecules as heme oxygenase-1 (HO-1) and IL-10 were expressed at elevated level by cAMP-DCs as compared with stimulated control DCs. In accordance with their protolerogenic expression profile, cAMP-DCs displayed lower

39

APC activity than control DCs at either state of stimulation and mediated an altered cytokin profile of cocultured T cells. 1. Dept. of Dermatology, Johannes Gutenberg University, Mainz, Germany

173 Comparative analysis of phenotypes and antigen presenting properties of differential stimulated human dendritic cells Trojandt S. (1), Wiechmann N. (1), Bovensiepen C. (1), Bellinghausen I. (1), Hilmenyuk T. (1), Saloga J. (1), Grabbe S. (1), Reske-Kunz A.B. (1), Bros M. (1) Immature dendritic cells (DCs) are localized in most tissues and are specialized in the uptake of antigenic material. In their immature and semi-mature state DCs are essential for the establishment and maintenance of peripheral tolerance towards autoantigens and harmless environmental antigens. DCs are activated by pathogen-derived factors as well as proinflammatory mediators. At their fully mature state they constitute potent antigen presenting cells which are capable to stimulate even naïve T cells. In our study we have analyzed the DC activating properties of physiologic proinflammatory mediators (IL-1β, TNF-α, Prostaglandin E2 [PGE2]) which are usually applied as a cocktail to achieve full stimulation of in vitro generated human monocyte-derived DCs. The strongest upregulation of costimulatory molecules (CD80, CD83), proinflammatory cytokines (IL-6, IL12), of the DC maturation marker fascin, but also the tolerance marker Ido was noted upon stimulation of DCs with the maturation cocktail. Concerning the outcome of DC stimulation with cocktail components, only PGE2 mediated a significant upregulation of costimulators, but was incapable to induce the production of proinflammatory cytokines. On the other hand, the various agents inhibited production of the antiinflammatory cytokine IL-10 to similar extent as the maturation cocktail. For several other immunomodulatory molecules we observed distinct modes of regulation in response to treatment of DCs with the various cocktail components. In accordance with the increased expression state of costimulatory molecules and proinflammatory cytokines, cocktail-stimulated DCs exerted the strongest proliferation of allogenic CD4+ T cells, followed by PGE2-stimulated DCs. Moreover, DCs stimulated with TNF-α or IL1β or a combination of both cytokines lacked strong T cell stimulatory capacity. 1. Dept. of Dermatology, Johannes Gutenberg University, Mainz, Germany

174 The RNA binding protein tristetraprolin influences the activation state of murine dendritic cells Bros M. (1), Wiechmann N. (1), Besche V. (1), Art J. (2), Pautz A. (2), Grabbe S. (1), Kleinert H. (2), Reske-Kunz A.B. (1) Under homeostatic conditions immature and semi-mature dendritic cells (DCs) are necessary for the maintenance of peripheral tolerance, but initiate strong immune responses after activation mediated for example by pathogen-derived factors like lipopolysaccharide (LPS). DC activation is accompanied by rapid upregulation of proinflammatory cytokines which were demonstrated to be regulated by both transcriptional and posttranscriptional mechanisms in other cell types. In this study we have analyzed the relevance of the RNA binding protein tristetraprolin (TTP) for DCs. We differentiated myeloid DCs from bone marrow progenitor cells derived from wild type (WT) and TTP knockout (TTPko) mice and analyzed their characteristics in a comparative manner. Unstimulated TTPko DCs expressed lower levels of mRNA species which encode costimulatory molecules, but displayed an elevated expression for other well known TTP target mRNAs. LPS-mediated stimulation of DCs of either genotype resulted in rapid upregulation of all monitored mRNA species to comparable extent. mRNA decay analysis revealed gene specific differences regarding mRNA stability which also depended on the DC activation state. Concerning functional alterations, we noted a lower T cell stimulatory potential of TTPko DCs as compared with WT DCs under basal conditions and induced an altered cytokin pattern in DC/T cell cocultures. With respect to the tolerogenic potential of unstimulated DCs, T cells primed by either DC population were equally refractory to restimulation and suppressed proliferation of naïve T cells at same extent. Taken together, our data indicate that posttranscriptional regulation constitutes an important mechanism involved in the control of the gene expression profile of DCs at either state of activation. TTP is one of the RNA binding proteins which contribute to posttranscriptional gene regulation. 1. Clinical Research Unit Allergy, Dept. of Dermatology, Johannes Gutenberg University, Mainz, Germany, 2. Dept. of Pharmacology, Johannes Gutenberg University, Mainz, Germany

175 The mRNA-binding protein KSRP regulates pro-inflammatory gene expression in human chondrocytes Art J. (1), Pautz A. (1), Schmidt N. (1), Kleinert H. (1) Gene expression is controlled by transcriptional and post-transcriptional mechanisms. A central part of the post-transcriptional modulation of gene expression is mediated by the regulation of mRNA stability. The KH-type splicing protein (KSRP) is a multifunctional protein, which contains four hnRNA K-homology domains and is a member of the family of far upstream sequence binding proteins (FUBP). In addition, KSRP plays an important role in modulation of mRNA stability by interaction with AU-rich elements (AREs) located in the 3´ untranslated region (3´UTR). In this study, we investigated the role of KSRP in regulating pro-inflammatory gene expression in human immortalized chondrocytes (C28-I/2) as a model of inflammatory joint diseases. In KSRP overexpressing C28-I/2 cells, iNOS, RANTES and IL-8 mRNA expression was decreased after cytokine treatment in comparison to C28-I/2 chondrocytes transfected with the empty expression vector. In contrast, cytokine-induced iNOS expression was increased in chondrocytes expressing a dominant-negative KSRP mutant (KSRPdn). Additionally, the endogenous KSRP protein level in KSRP as well as in KSRPdn overexpressing C28/I-2 cells was enhanced after cytokine-stimulation compared to control cells. We suppose a stabilizing effect by KSRP of its own expression. The KSRPdn mutant was generated by deletion of the KH-4 domain, which is essential for mRNA binding. Therefore, a direct binding of KSRP on its mRNA can be excluded. Hence, this effect seems to be mediated by an indirect, still unknown mechanism. Also

we analyzed the regulation of KSRP mRNA expression in DLD-1 cells. In cytokinestimulated as well as in untreated DLD-1 cells transfected with the KSRP 3´UTR a negative effect on the KSRP expression was detected. In DLD-1 cells transfected with a 3 kb fragment of the KSRP promoter the KSRP expression was regulated in a positive manner after cytokine treatment. In summary, the RNA binding protein KSRP plays an important role in the modulation of pro-inflammatory gene expression. Furthermore, KSRP expression seems to be regulated in a cytokine-dependent and a very distinguished manner. 1. Dept. of Pharmacology, Johannes Gutenberg University, Mainz, Germany

176 A novel JNK-regulated ribonucleoprotein complex that responds to cytokines and participates in inflammatory gene expression Rzeczkowski K. (1), Holzberg D. (2), Kettner-Buhrow D. (1), Beuerlein K. (1), Müller H. (1), Wolter S. (2), Thiefes A. (2), Dörrie A. (2), Wait R. (3), Saklatvala J. (3), Kracht M. (1) The proinflammtory cytokine interleukin-1 and stressful stimuli activate the JNK signalling pathway to regulate the expression of secondary proinflammatory genes such as ccl2. This occurs in part by JNK-mediated phosphorylation of the transcription factor c-Jun. However, little is known about other substrates of JNK or its role in postrancriptional regulation. Using tandem-affinity purification we identified argonaute 2 (Ago2) as a novel JNK-binding protein. The Ago2:JNK interaction was confirmed in vitro and by co-immunoprecipitation of ectopically expressed T7-Ago2 and GFP-JNK. Ago2 is the essential catalytic subunit of the RNA-induced silencing complex (RISC) which contains mRNAs and micro(mi) RNAs and which regulates miRNA- and siRNAmediated mRNA silencing. The Ago2:JNK interaction was constitutive and independent of JNK activation, but was disrupted by depletion of cell lysates from mRNA and miRNAs, indicating that Ago2 and JNK are part of a novel ribonucleoprotein (RNP) complex. Out of 157 miRNAs the expression of miR-23b, miR-135a, miR-186, miR-214 as well as miR-335 was altered in JNK-deficient cells. miR-23b, miR-186 and miR-214 were also isolated from immunoprecipitated Ago2:JNK complexes of HEK293IL-1R cells. MEKK1-mediated JNK phosphorylation resulted in decreased binding of miR-23b, miR-186 and miR-210 but not of miR-146a and miR-214 to Ago2:JNK complexes. Furthermore, we found that JNK phosphorylates another putative Ago2-interacting protein, DCP1α at ser315. JNK also bound to DCP1α in Co-IP assays. Strong JNK activators such as IL-1, sorbitol or transfected MAP3Ks altered the number and size of DCP1α containing RNP particles in intact cells. Collectively, these data suggest that we have identified parts of a novel JNK-dependent pathway that regulates mRNA processing by RNP particles. As some of the miRNAs contained in these particles such as miR-23b are predicted to bind inflammatory mRNAs such as ccl2, this pathway likely modulates the overall inflammatory gene response. 1. Rudolf-Buchheim-Institute of Pharmacology, Justus-Liebig-University Giessen, Giessen, Germany, 2. Institute of Pharmacology, Medical School Hannover, Hannover, Germany, 3. The Kennedy Institute of Rheumatology Division, Imperial College London, London, UK

177 Immunomodulation by misplaced U1-snRNA Sadik C.D. (1), Bachmann M. (1), Pfeilschifter J. (1), Mühl H. (1) U1-snRNA is an integral part of the nuclear U1 ribonucleoprotein pivotal for eukaryotic pre-mRNA splicing. Toll-like receptor (TLR) signaling has recently been associated with immunoregulatory capacities of U1-snRNA. Here, we report for the first time on interferon regulatory factor (IRF)-3-dependent signaling in A549 lung epithelial/carcinoma cells initiated by endosomal delivery of U1-snRNA. IRF3 activation was associated with expression of the IRF3-dependent genes interferon-β, IP10/CXCL10, and indolamine 2,3-dioxygenase. Notably, expression of these parameters was suppressed by bafilomycin, an inhibitor of endosomal acidification, implicating endosomal TLR activation. Similar data were obtained using DLD-1 colon epithelial/carcinoma cells. Since resiquimod, an agonist of TLR7/8, failed to stimulate A549 or DLD-1 cells, current data suggest TLR3 as of prime relevance for cellular activation. In order to characterize the overall regulatory potential of U1-snRNAactivated epithelial cells on inflammatory cytokine production, cocultivation with freshlyisolated peripheral blood mononuclear cells (PBMC) was performed. Notably, epithelial cells activated by U1-snRNA enhanced phytohemagglutinin-induced interleukin (IL)-10 release by PBMC but suppressed that of tumor necrosis factor-α, implying immunosuppression by U1-snRNA. Since U1-snRNA is enriched in apoptotic bodies and epithelial cells are capable of performing efferocytosis, present data may in particular relate to immunobiological aspects of apoptosis at host/environment interfaces. 1. pharmazentrum frankfurt, Klinikum dfer Goethe-Universität Frankfurt

178 Novel effects of the CDK inhibitor flavopiridol: restraint of leukocyte-endothelial cell interactions by downregulation of endothelial cell adhesion molecules Schmerwitz U.K. (1), Khandoga A.G. (2), Mayer B.A. (1), Berberich N. (1), Krombach F. (2) , Zahler S. (1), Vollmar A.M. (1), Fürst R. (1) Flavopiridol, a semisynthetic flavone, is a potent inhibitor of cyclin-dependent kinases (CDKs). It is the first CDK-inhibitor that has entered clinical trials and is currently tested for the treatment of different cancer diseases, such as lymphomas, leukemias, or carcinomas. We hypothesized that flavopiridol, besides its well documented anti-tumour action, exerts additional effects on inflammation-activated endothelial cells. Thus, aim of the present study was to elucidate the anti-inflammatory potential of flavopiridol both in vivo and in vitro by assessing leukocyte-endothelial cell interactions. Moreover, we aimed for gaining a first insight into the underlying mechanisms of action. Using intravital microscopy, leukocyte adhesion and transmigration were measured in postcapillary venules of the mouse cremaster muscle. Animals were pretreated with flavopiridol (i.v. bolus to reach 100 nM plasma concentration) 15 min before intrascrotal injection of TNF-α (300 ng, 4 h). We found that flavopiridol effectively diminishes the number of adhering and transmigrating leukocytes. In vitro, flavopiridol lowered TNF-α-induced (10

40

ng/ml, 24 h) leukocyte adhesion to human umbilical vein endothelial cells (HUVECs), as determined by measurement of myeloperoxidase activity. Since endothelial cell adhesion molecules (CAMs) are crucial for leukocyte-endothelial cell interactions, we analyzed the influence of flavopiridol on the TNF-α-induced expression of E-selectin, ICAM-1, and VCAM-1. In vitro, flavopiridol (30 min pretreatment) strongly inhibited TNFα-evoked (10 ng/ml,24 h) upregulation of E-selectin (IC50 ~118 nM), ICAM-1 (IC50 ~27 nM), and VCAM-1 (IC50 ~74 nM) (flow cytometry). Regarding mRNA expression (real time RT-PCR), we could demonstrate a downregulation of ICAM-1 in cremaster tissue samples and in HUVECs. Concerning the mechanisms of CAM downregulation, we found that flavopiridol slightly decreases the activation of the transcription factors AP-1 and NF-kB by TNF-α in vitro (electrophoretic mobility shift assay). Our data show for the first time that the CDK inhibitor flavopiridol inhibits leukocyte-endothelial cell interactions due to a downregulation of endothelial CAMs, both in vivo and in vitro, which warrants further detailed mechanistic investigations. In conclusion, we could highlight flavopiridol as a novel, promising anti-inflammatory agent, which might reach clinical relevance for the treatment of inflammatory diseases. 1. Department of Pharmacy, Pharmaceutical Biology, Ludwig-Maximilians-University, Munich, Germany, 2. Walter Brendel Center of Experimental Medicine, LudwigMaximilians-University, Munich, Germany

179 S-Curvularin inhibits pro-inflammatory gene expression in different models of rheumatoid arthritis Schmidt N. (1), Art J. (1), Pautz A. (1), Rauschkolb P. (1), Erkel G. (2), Anke T. (2), Kleinert H. (1) Dysregulation of the expression of several of pro-inflammatory molecules is involved in the development and progression of chronic inflammatory diseases such as rheumatoid arthritis (RA). For example, enhanced expression of inducible nitric oxide synthase (iNOS) leads to inappropriate NO production and thereby contributes to joint destruction. In former studies we described the isolation of new anti-inflammatory fungal compounds (Sporogen, S14-95 and S-Curvularin), which were able to reduce iNOS promoter activity as well as iNOS mRNA expression and iNOS-related NO production in human alveolar A549/8 cells (Yao et al., 2003, Mol Pharmacol 63, 383-391). In the current study we investigated the effect of S-Curvularin (SC) on inflammatory gene expression in different models of RA. In the in vivo model of collagen-induced arthritis (CIA) in mice, treatment with SC markedly reduced inflammation and swelling of the paws in arthritic mice. As shown by DNA-microarray experiments, SC inhibited the CIA-mediated induction of a wide variety of pro-inflammatory genes. For example, the chemotactic molecule S100A8, which is constitutively expressed by neutrophils, activated monocytes and macrophages, is considered to amplify pro-inflammatory responses through activation of NF-kB and MAP kinase pathways. S100A8 mRNA expression was increased in paws, periphere mononuclear blood cells (PBMCs) and in liver after immunization with chicken collagen. Treatment with SC reduced S100A8 mRNA and protein expression in the paws and PBMCs of the animals. Human C-28/I2 chondrocytes were used as a cellular model. In these cells, cytokine-induced iNOS mRNA and protein expression as well as iNOS-related NO production was decreased by treatment with SC. Surprisingly, the potent glucocorticoid dexamethasone was not able to inhibit cytokine-induced iNOS expression. The glucocorticoid-resistance of C-28/I2 chondrocytes may be due to the absence of the glucocorticoid receptor (GCR) mRNA. In conclusion, SC may be a promising candidate for the future treatment of inflammatory diseases. 1. Dept. of Pharmacology, Johannes Gutenberg University, Mainz, Germany, 2. Dept. of Biotechnology, Technical University, Kaiserslautern, Germany

180 The CYP2D6 mouse model for autoimmune liver disease: Mapping of cytochrome P450 2D6 T cell epitopes in mice with autoimmune hepatitis Holdener M. (1), Hintermann E. (1), Bayer M. (1), Manns M.P. (3), von Herrath M.G. (2), Christen U. (1) The etiology of autoimmune hepatitis (AIH) is poorly understood although the major autoantigen, cytochrome P450 2D6 (CYP2D6), has been identified and immunodominant epitopes mapped. Therefore, we generated an animal model for human AIH using the natural autoantigen CYP2D6. We infected transgenic mice expressing human CYP2D6 in the liver (CYP-2D6 mice) with an Adenovirus-CYP2D6 vector (Ad-2D6) to break self-tolerance. Surprisingly, upon infection with Ad-2D6 not only transgenic CYP2D6 mice but also wildtype FVB mice showed several persistent features characteristic for liver damage associated with AIH. These features included massive hepatic fibrosis, ‘fused’ liver lobules, disorganized architecture, cellular infiltrations and focal to confluent necrosis. Further, all Ad-2D6-infected mice generated high titers of anti-CYP2D6 antibodies. Epitope mapping revealed that such anti-CYP2D6 antibodies predominantly recognized the immunodominant linear epitope recognized by LKM-1 antibodies from AIH patients. In order to analyze the T cell mediated immune response lymphocytes were isolated and stimulated with 61 overlapping 20-mer peptides covering the entire human CYP2D6. T cells produced IFNγ after stimulation with peptides only if isolated from Ad-2D6-infected mice. The frequency of IFNγ producing T cells was more then three times higher in lymphocytes isolated from the liver then from the spleen indicating either an organ specific proliferation or a selective retention process in the liver. In contrast to the Ad-2D6-infected mice, mice infected with a control Adenovirus did neither develop liver damage nor generated anti-CYP2D6 antibodies nor produced IFNγ after ex-vivo stimulation with CYP2D6 peptides. 1. Pharmazentrum Frankfurt / Zentrum für Arzneimittelforschung, Entwicklung und Sicherheit (ZAFES), Klinikum der Goethe Universität, Frankfurt am Main, Germany, 2. La Jolla Institute for Allergy and Immunology, La Jolla, CA, 3. Hannover Medical School, Hannover, Germany

181 The mTOR kinase inhibitor rapamycin propagates profibrotic cell signaling cascades in rat renal mesangial cells Osman B. (1), Doller A. (1), Akool E.S. (1), Hintermann E. (1), Pfeilschifter J. (1), Eberhardt W. (1) The mTOR kinase inhibitor Rapamycin (Sirolimus) is a drug with potent immunosuppressive and antiproliferative properties. Rapamycin is commonly used in combination with the calcineurin inhibitors (CNIs) ciclosporin A (CsA) and tacrolimus (FK506) in transplantaion medicine in order to limit the strong nephrotoxicity of CNIs. In the present study, we demonstrate that Rapamycin by itself is able to induce Smadsignaling cascades in rat renal mesangial cells (MC) as is indicated by the rapid increase in the nuclear contents of phosphorylated Smad proteins Smad-2 and Smad-3, respectively. By EMSA, we demonstrate a Rapamycin-triggered increase in the DNAbinding of both transcription factors to a cognate Smad-binding promoter element (SBE) which is accompanied by an increased expression of tissue growth factor (CTGF) and plasminogen activator inhibitor-1 (PAI-1). The activation of the Smad signaling cascade by Rapamycin is independent of p70 S6 kinase inhibition but interestingly, depends on the intracellular immunophillin FKBP12 as we could demonstrate by a small interference (si)RNA experiments. Mechanistically, the activation of receptor-(R)-Smads by Rapamycin is due to an increased release of latent TGFβ1 as indicated by ELISA and by the strong inhibitory effects of a neutralizing TGFβ-antibody. Using an activin receptor-like kinase (ALK)-5 inhibitor and by siRNA we furthermore could depict the critical involvement of both types of TGFb receptors. Besides SB203580, a specific inhibitor of the p38 MAPK, the reactive oxygen species (ROS) scavanger N-acetylcysteine (NAC) and a cell-permeable superoxide dismutate (SOD) mimetic strongy impaired the stimulatory effects of Rapamycin on Smad signaling indicating a critical involvement of ROS. Moreover, we found a rapid increase in ROS generation by Rapamycin after 30 minutes by the significant rise in dichlorofluorescein (DCF) formation. Collectively, our data demonstrate that Rapamycin via increased ROSformation can activate latent TGFβ and thereby may popagate profibrotic gene expressions in the kidney. 1. Dept. of Pharmacology, Pharmacenter Frankfurt, Johann Wolfgang GoetheUniversity, Frankfurt, Germany

182 Targeting NF-κB in macrophages alleviates skin inflammation in a mouse model of psoriasis Syrovets T. (1), Wang H. (2), Büchele B. (1), Lunov O. (1), Scharffetter-Kochanek K. (2), Simmet T. (1) Psoriasis vulgaris is a frequently occurring inflammatory skin disease with an incidence of 2-3%. The pathogenesis of psoriasis is not understood in detail. Cytokines and other inflammatory mediators contribute to the establishment and maintenance of the pathologic phenotype of psoriasis lesions by promoting leukocyte infiltration and activation, and by directly modulating keratinocyte proliferation and differentiation Activation of nuclear factor κB (NF-κB) has been suggested to play a pathogenic role in psoriasis. However, the exact role of NF-κB in this disease remains undefined. The CD18 hypomorphic (CD18hypo) PL/J mouse is a model of human psoriasis, which is characterized by the reduced expression of CD18, the common chain of β2 integrins (CD11/CD18), to 2–16% of wild type levels. We have previously demonstrated that this form of psoriatic skin inflammation depends on efficient recruitment and activation of macrophages, a major source of proinflammatory TNF-α. Here we report that in lesions of human psoriatic skin and in affected skin of a CD18hypo mouse NF-κB activation in terms of degradation of IκBα and phosphorylation of p65 is predominantly confined to dermal macrophages. Inhibition of NF-κB signaling in macrophages by systemic or topical administration of the NF-κB inhibitor acetyl-11-keto-β-boswellic acid (AKβBA) suppressed the TNF-α production by macrophages, which was accompanied by a profound improvement of the disease score in the CD18hypo mice. Concomitantly, the number of macrophages and lymphocytes dramatically decreased in the previously affected areas to normal values. Our data demonstrate that the activation of NF-κB in macrophages is pivotal for the pathogenesis in the CD18hypo mouse model of psoriasis. Selective targeting of NF-κB in macrophages may therefore provide an effective strategy for the treatment of human psoriasis. Supported by the DFG. 1. Institute of Pharmacology of Natural Products & Clinical Pharmacology, Ulm University, Ulm, Germany, 2. Department of Dermatology & Allergic Diseases, Ulm University, Ulm, Germany

183 Superparamagnetic iron oxide nanoparticles used as contrast agents induce radical oxygen species formation in human macrophages Lunov O. (1), Syrovets T. (1), Büchele B. (1), Landfester K. (2), Simmet T. (1) Nanotechnology and the clinical application of nanoparticles is a rapidly growing field. Superparamagnetic iron oxide nanoparticles are used in magnetic resonance imaging (MRI), for the treatment with hyperthermia, for drug delivery and studies of cell mechanics as well as in the industry in catalytic and separation processes. However, so far our knowledge on the physiological and pathophysiological effects of manufactured nanoparticles remains rather limited. The aim of this study was to investigate effects of two superparamagnetic contrast agents, namely ferucarbotran, Resovist® (SHU 555A) and Supravist® (SHU 555C), on human macrophages. Dynamic laser light scattering analysis revealed that the mean size of the carboxydextran-coated Resovist® particle is approximately 60 nm and of Supravist® about 20 nm. Electron microscopic analysis showed that the size of the iron oxide core of both particles is 5 nm. Addition of both, Resovist® and Supravist®, concentration- and time-dependently impaired the viability of monocyte-derived human macrophages as assessed by the XTT assay. Moreover, both particles triggered the intracellular production of reactive oxygen species (ROS) in macrophages as determined with the oxidation-sensitive fluorescent indicator 5-(and-6)carboxy-2’,7’-dichlorodihydrofluorescein diacetate (carboxy-H2DCFDA). The accumulation of ROS led to sustained Jun N-terminal Kinase (JNK) activation and cell apoptosis as evidenced by annexin V binding indicating phosphatidylserine expression on the outer leaflet of the membrane as well as caspase 3 activation. JNK activation induced by superparamagnetic iron oxide nanoparticles as well as the cytotoxicity as

41

determined by XTT could be inhibited by ROS-scavengers 6-hydroxy-2,5,7,8tetramethylchromane-2-carboxylic acid and N-acetyl-cystein. This study revealed a molecular mechanism for a putative adverse drug reaction and implies beneficial effects of ROS scavengers. Further studies are required to investigate whether such effects might occur in vivo. Supported by the DFG. 1. Institute of Pharmacology of Natural Products & Clinical Pharmacology, Ulm University, Ulm, Germany, 2. Institute of Organic Chemistry III, Ulm University, Ulm, Germany

184 Distinct cytokine profile associated with infection of human peripheral blood mononuclear cells by Borrelia burgdorferi Horn K. (1), Kraiczy P. (2), Bachmann M. (1), Hunfeld H.-P. (2), Sadik C.D. (1), Pfeilschifter J. (1), Mühl H. (1) Lyme disease is a spirochetal infection caused by the Borrelia burgdorferi sensu lato complex that can proceed towards serious inflammatory joint, cardiac, and neuronal manifestations. Early infection is characterized by flu-like symptoms and by erythema migrans developing at the bite site area. Production of inflammatory cytokines orchestrating leukocytic infiltration and local activation is supposed to be key to early pathogenesis of infection. Using human peripheral blood mononuclear cells (PBMC) freshly-infected with Borrelia burgdorferi 297, we set out to assess the cytokine profile initiated early during course of infection with focus on the newly characterized group of Th17-like cytokines. Here, we report for the first time on a distinct set of cytokines being produced by infected PBMC that is characterized by high levels of TNFα, IFNg, IL-22, and IL-23 but insufficient IL-18 release. Most notably, even through being highly activated, PBMC displayed a complete lack of IL-17 production. This rather distinct cytokine profile may impact on the pathophysiology of infection and open the avenue towards novel cytokine-based therapeutic strategies aiming at early Borrelia burgdorferi infections. 1. pharmazentrum frankfurt, Klinikum der Goethe-Universität Frankfurt, 2. Institut für Medizinische Mikrobiologie und Krankenhaushygiene, Klinikum der Goethe-Universität Frankfurt

185 Expression of the anti-inflammatory mediator GILZ (Glucocorticoid-Induced Leucine Zipper) is downregulated by Toll-like receptor activation Hoppstädter J. (1), Diesel B. (1), Kiemer A.K. (1) Toll-like receptors (TLRs) play a key role in the induction of innate immune responses by recognising pathogen-associated molecular patterns (PAMPs). The exposure to PAMPs results in the activation of Nuclear Factor-κB (NF-κB), leading to transcription of proinflammatory mediators, such as cytokines. Glucocorticoid-Induced Leucine Zipper (GILZ) is a recently described protein important for the anti-inflammatory action of glucocorticoids. It interferes with NF-κB-mediated gene transcription in macrophages. Although NF-κB signalling is significantly linked to the pathogenesis of inflammatory diseases like asthma or Chronic Obstructive Pulmonary Disease (COPD), little is known about the regulation of GILZ during inflammation. We therefore aimed to investigate GILZ expression in human macrophages after TLR activation. Treatment of alveolar macrophages or in vitro differentiated macrophages with lipopolysaccharide (LPS, TLR4 ligand) resulted in a time- and dose-dependent decrease of GILZ mRNA and protein levels, as assessed by real time RT-PCR and Western Blot. Due to the rapid decrease of GILZ mRNA levels unpon LPS addition we suggested that GILZ mRNA stability might be affected. Thus, we quantified mRNA levels after treatment with the transcription inhibitor Actinomycin D. Interestingly, GILZ mRNA half-life turned out to be significantly shortened by LPS. Similar effects could be observed after exposure to other TLR agonists like Pam3CSK4 (TLR1/2), Flagellin (TLR5), FSL-1 (TLR2/6), Imiquimod (TLR7), synthetic ssRNA (TLR8), and bacterial DNA (TLR9). TLR3 ligand polyinosinic:polycytidylic acid (Poly(I:C)) displayed no ability to decrease GILZ mRNA levels, although cells were highly activated by Poly(I:C). TLR3 is the only TLR whose signalling pathway does not involve the adaptor molecule MyD88. Therefore, an influence of this adaptor protein in regulation of GILZ expression is suggested. Furthermore, activation of NF-κB seems to be pivotal for the downregulation of GILZ: inhibition of NF-κB by MG-132 or BAY-7085 prior to LPS challenge completely abrogated the diminution of GILZ mRNA. Conclusion: Our data indicate a MyD88- and NF-κB-dependent GILZ downregulation in human (alveolar) macrophages. GILZ suppression seems to be mediated by mRNA destabilization, which might represent a regulatory mechanism in TLR activation. 1. Dept. of Pharmaceutical Biology, Saarland University, Saarbrücken

186 Analysis of cylooxygenase-1b expression on genomic and protein level Reinauer C. (1), Freidel K. (1), Steger G. (2), Klamp T. (3), Weber A.-A. (4), Schrör K. (1), Censarek P. (1) Objectives: Cyclooxygenase-1b (COX-1b, formerly COX-3) differs from COX-1 in the complete retention of intron1. This retention of human intron1 leads to a frame-shift and a hypothetical truncation of the protein. Nevertheless, full-length COX-1b protein expression is detectable in human tissues and was shown to contribute to prostaglandin and thromboxane production in vitro. This study elucidates a potential mechanism of frame-shift repair and functional protein expression. Methods: mRNA was extracted from human stomach and liver samples from 10 individuals, confirmed to be free of genomic DNA and transcribed into cDNA. The region enframing intron1 was amplified and subcloned. Ten independent clones from each individual were scanned for aberrations in intron1 length and sequence. RNA structure analysis utilised the algorithms RNAfold and pknotsRG. A COX-1b clone was synthezised and tagged with a His-Tag for proteinpurification. COX-1b protein was overexpressed in HEK freestyle cells and purified with Ni-NTA resin. Results: All sequenced clones comprised an intron1 sequence of 94 bp, theoretically leading to a frame-shift and protein truncation. No RNA-editing could be detected. Despite single aberrations, the intron1 sequence was consistent with the published sequence. RNA structure algorithms predicted a hypothetical slippery

sequence in combination with a thermostable pseudoknot structure in intron1, which is a premise for ribosomal frame-shifting. COX-1 overexpression and purification led to a single band in SDS page with a electrophoretic mobility slightly over 70 kD, which could be detected by an anti-COX-1- and anti-His-antibody. Conclusions: We found no evidence for RNA-editing, a postulated mechanism for COX-1b protein production. RNA structure analysis provides implications for ribosomal frame-shifting, a mechanism for frame-shift modification during translation. A +1 ribosomal frame-shift would revert the sequence into the open reading frame, resulting in full-length COX-1b protein production. Such a full-length COX-1b protein can be overexpressed in a human cell line and results in the expected protein size. As our former studies showed, full-length COX1b protein is an active COX-1 variant, capable of prostaglandin and thromboxane synthesis1. 1Censarek et al., (2006): Human COX-1b (COX-3) is not the elusive target of acetaminophen. Eur J Pharmacol, 551, 50-53 1. Institut für Pharmakologie und Klinische Pharmakologie, Heinrich-Heine-Universität, Düsseldorf, Germany, 2. Institut für Physikalische Biologie, Heinrich-Heine-Universität, Düsseldorf, Germany, 3. Department of Internal Medicine III, Johannes Gutenberg University, Mainz, Germany, 4. Institut für Pharmakologie, Universität Duisburg-Essen, Universitätsklinikum Essen, Germany

187 JAK1/STAT3-dependent induction of granzyme B in human B cells by interleukin 21 and viral antigens Jahrsdörfer B. (1), Hagn M. (1), Sontheimer K. (1), Maier J. (1), Beyer T. (1), Syrovets T. (1), Simmet T. (1) B cells are not currently known to produce granzyme B (GrB) in (patho-)physiological settings. We recently reported that B-chronic lymphocytic leukemia cells and normal B cells treated with interleukin-21 (IL-21) and anti-B cell receptor antibodies (anti-BCR) produce functional GrB. Here we demonstrate for the first time that viral antigens can also induce specific peripheral B lymphocytes to produce and secrete substantial amounts of active GrB. Using FACS, ELISpot, immunofluoresence and Western blot, we show that B cells from subjects recently vaccinated against tick-borne encephalitis virus (TBEV) but not unvaccinated subjects respond to viral TBEV antigens with GrB secretion in a dose-dependent manner. The response is direct and occurs only in the presence of co-activation with IL-21. GrB production was also seen on the mRNA level, suggesting GrB secretion results from neogenesis rather than from release from preformed vesicles. This is also supported by the observation that the translation inhibitor cycloheximide completely abrogates GrB production by B cells. We further show that the GrB production by B cells involves activation of JAK1 and STAT3. Inhibition of JAK1 by pyridone 6 completely abrogates GrB induction by viral antigens or anti-BCR. In summary our findings suggest B cells respond to IL-21 and viral antigens with a JAK1/STAT3-dependent GrB secretion. Our findings may represent a novel antiviral immune response mechanism. 1. Institute of Pharmacology of Natural Products and Clinical Pharmacology, Ulm University, Helmholtzstr. 20, 89081 Ulm, Germany

188 Cooperative role of CXCR3 and CXCR4 for transmigration of immune cells in chronic inflammation Giegold O. (1), Pfeilschifter J.M. (1), Radeke H.H. (1) Chronic inflammatory diseases like rheumatoid arthritis are characterized by an increased infiltration of TH1 cells into affected tissues. Typical chemokine receptors of TH1 cells are CXCR3 known as inflammatory and CXCR4 as homeostatic receptor. Interestingly, convincing evidence was presented that during chronic inflammation homeostatic CXCR4 due to an interaction with CXCR3 may propagate the inflammatory process. Herein we performed static chemotaxis assays focusing on gradient dependent migration of central memory TH1 cells induced by ligands of CXCR3 (CXCL9, -10, -11) and CXCR4 (CXCL12). Our data revealed that both CXCL9 and CXCL12 augmented chemotaxis of TH1 cells. Pre-stimulation of TH1 cells with CXCL9 for 0.5h led to an increased migration induced by subsequent CXCL12 (heterologous sensitization), whereas longer periods reduced migration. As expected, migration induced by CXCL9 decreased after pre-stimulation with CXCL9 (homologous desensitization). FACS data showed reduced cell surface expression of CXCR3, whereas expression of CXCR4 was not affected. To examine this unusual cooperation of CXCR3/-4 with regard to lymphocyte rolling and adhesion we applied a dynamic flow chamber system. First results with slides coated with E-selectin and ICAM-1 (Fc-chimera) showed an increased adhesion after CXCL12 and CXCL9 comparably to static assays. Using pre-stimulated TH1 cells with CXCL9 for 0.5h exhibited enhanced adhesion in presence of immobilized CXCL12 compared to non pre-stimulated cells. In order to identify responsible mechanisms for these sensitizing effects we performed intracellular calcium determinations and found a correlation to previous findings in the functional assays. Thus, our data confirm a major role of cooperative effects of inflammatory and homeostatic chemokines for extravasation of immune cells. Experiments are in progress with microvascular endothelial cells and different chemokine inhibitors like vMIP-II and AMD3100 to reveal mechanisms in detail. 1. pharmazentrum frankfurt / ZAFES, Clinic of Goethe University, Frankfurt/Main, Germany

189 p47phox as an important modulator of IL-12p70 in dendritic cells and the consequences for the Th1 response Richter C. (1), Brandes R.P. (2), Herrero San Juan M. (1), Holmdahl R. (3), Pfeilschifter J.M. (1), Radeke H.H. (1) Dendritic cells as the sentinels between innate and adaptive immunity have the prime task to decide towards tolerance or immunity upon invasion of pathogens. Preventing excessive responses leading to autoimmune or chronic diseases a fine tuned balance of immune signalling cascades is necessary. We investigated the regulatory function of p47phox, an organizing protein of the NADPH oxidase, in the signalling pathway downstream of TLRs in dendritic cells. Our initial investigations revealed an enhanced

42

TLR9-induced (CpG2216) IL-12p70 expression in p47phox deficient (p47phox-/-) compared to WT spleen cells (C57Bl/6), which -unexpectedly- was independent of the production of reactive oxygen species. In contrast to TLR9 stimulation TLR4-induced IL12p70 secretion was not affected, which hints to a negative feedback regulation of IL12p70 restricted to TLR-MyD88 signalling. In several mouse and rat strains with different mutations of the p47phox gene, resulting in none or less functional p47phox protein, we confirmed the negative feedback regulation of IL-12p70 by p47phox. Since IL-12p70, produced by dendritic cells, is a major cytokine and important 3rd signal for polarizing naïve T cells into Th1 cells we immunized mice with the antigen ovalbumin and CpG2216 as adjuvants and measured the IFNγ response by Elispot assays ex vivo. In the regional lymph nodes from p47phox deficient mice immunized with ovalbumin and CpG2216 we found an increased Th1 response compared to WT mice. Immunization of mice with ovalbumin alone did not result in IFNγ response. These results suggest that independent from its role for NADPH oxidase p47phox is essential for the fine tuning of TLR/MyD88-dependent IL-12p70 response in dendritic cells and this consequently regulates the Th1 response. 1. pharmazentrum frankfurt/ZAFES, Goethe University, Frankfurt am Mainz, Germany, 2. Center of Physiology, Goethe University, Frankfurt am Main, Germany, 3. Section for Medical Inflammation Research, Karolinska Institutet, Stockholm, Sweden

190 The influence of sphingosine kinase activities on murine interleukine-12 expression Richter K. (1), Pfeilschifter J.M. (1), Huwiler A. (2), Radeke H.H. (1) Sphingosine kinases 1 and 2 (SphK) are the key enzymes producing the important intraand extracellular messenger Sphingosine-1-phosphate (S1P). Conflicting data have been published showing that S1P influences the immune response of dendritic cells by reducing IL-12 levels and either decreasing IL-23p19 or enhancing Th17 development. On the other hand Th1 development is supposedly associated with an increased SphK-1 expression. In an experimental approach to specifically analyze the role of SphK 1 or 2 we used spleen cells from SphK 1- and 2-deficient (-/-) mice and determined IL12p70 production following stimulation with LPS, LPS/PGE2 or LPS/TGFß. PGE2 and TGFß are known to reduce IL12p70 assembly of dendritic cells (DC). LPS stimulated SphK2-/exhibited a 25-30% reduction of IL12p70 release compared to WT, whereas the SphK1-/were not different from WT. TGFβ and PGE2 showed the expected down regulation of IL12p70, but interestingly these effects were significantly enhanced in SphK2-/-. To allow determination of small changes of kinase activity in DC dependent immune responses – as opposed to the expression levels of SphK – we established a fluorescence assay employing NBD-labelled sphingosine (Sph) as substrate for SphK. After establishing an assay being non-selective for either SphK1 or SphK2 we could demonstrate that in spleen cells S1P is mainly formed by SphK2 which is in sharp contrast to previous studies by S. Spiegel et al. Thus our preliminary data suggest that not only the S1P formation and subsequent S1P receptor effects, but also other mechanism possibly dependent on the cell-type or cell compartment distribution of SphK1 and 2 enzymes influence the immunological sphingolipid activities. In further studies we are now investigating, whether independent from S1P formation stimulation/inhibition of SphK activity by different stimuli or lentiviral shRNA knockdown affects dendritic and T lymphocyte function. 1. pharmazentrum frankfurt / ZAFES, Clinic of the Goethe University, Frankfurt, Germany, 2. Pharmacology, University of Bern, Switzerland

191 Analysis of the effect of histamine on the interferon-gamma synthesis in mouse spleen and thymus cells Neumann D. (1), Beermann S. (1), Kaever V. (1), Seifert R. (1) The biological function of histamine is mediated by four different receptors, namely histamine H1 receptor (H1R), H2R, H3R, and H4R. During an immune reaction histamine acts as a local proinflammatory mediator and contributes to the polarisation of the adaptive immune reaction by modulating the activity of dendritic cells and T cells. In these cells, histamine may modulate the induced expression of characteristic T cell cytokines such as IFNγ, which plays a central role in a number of autoimmune diseases. The present study was initiated to analyze the involvement of histamine on the induced production of IFNγ by immune cells. Spleen and thymus cells from C57Bl/6 mice were stimulated in vitro by either immobilized anti-CD3 antibodies or CpG-Oligonucleotides (CpG-ODN) in the presence or absence of histamine or 4-methyl-histamine, a H4Rselective agonist. IFNγ production was evaluated by analysis of cell culture supernatants by ELISA. Histamine and 4-methylhistamine concentration-dependently reduced IFNγ production in splenocytes induced by either anti-CD3 antibodies or CpG-ODN. In thymocytes the anti-CD3 antibody induced IFNγ synthesis was reduced by histamine as well. However, it was not affected by 4-methyl-histamine. The histamine effect on the induced production of IFNγ in spleen cells was completely inhibited by the H2R-specific antagonist famotidine, while H4R-, H1R-, and H3/H4R-selective antagonists had no or only moderate effects. Interestingly, the H4R-selective reagent JNJ7777120, which serves as an antagonist on human cells, did not inhibit the histamine-mediated reduction of anti-CD3 induced IFNγ synthesis, but it slightly enhanced the histamine effect. Thus, at the murine H4R, JNJ7777120 may be a partial agonist. We conclude that histamine modulates the induced production of IFNγ by T cells via mainly the H2R and, to a much lesser extend, the H4R, with JNJ7777120 showing unexpected pharmacological properties. 1. Dept. of Pharmacology, Hannover Medical School, Hannover, Germany

192 Role of histamine H4 receptor on wheal and flare reactions in dog skin and on canine mast cell function Roßbach K. (1), Stark H. (2), Sander K. (2), Leurs R. (3), Kietzmann M. (1), Bäumer W. (1) Histamine is a well known mediator of allergic skin diseases and has long been recognised as a classical inducer of itch in humans. With the discovery of the histamine H4 receptor (H4R) in the year 2000, the role of histamine is re-evaluated. The H4R is mainly located on immune cells, in particular mast cells, eosinophils, dendritic cells and T cells and recent findings have shown that the H4R plays a role in multiple functions of these cells. Furthermore, this receptor seems to be involved in inflammatory responses in vivo and in mediating pruritus. In own studies we have shown, that the selective H4R antagonist JNJ7777120 clearly inhibits hapten-induced scratching in a murine model of allergic contact dermatitis. However, there are only limited published data, elucidating the role of the H4R in dogs. In mice, intradermal injection of histamine (0.25 µmol and 2.5 µmol/site) produced itch, whereas beagle dogs (n = 12) treated intradermally with histamine (0.25 µmol and 2.5 µmol/site) reacted with a classic wheal and flare reaction, but no dog showed signs of pruritus. The dogs reacted also with a wheal and flare reaction after intradermal injection of H4R agonist/H3R antagonist clobenpropit (0.1 µmol) and selective H4R agonist VUF8430 (1.5 µmol). But again, no scratching occurred in dogs. Corresponding to this, the selective H4R antagonist JNJ7777120 (1.25 µmol) reduced the histamine-induced wheal reaction in nine out of twelve dogs. To determine, whether canine mast cells are susceptible for H4R-mediated reactions, effects of clobenpropit and VUF8430 were tested in canine mastocytoma cells (C2), a common cell line, which shares morphological and functional characteristics with skin mast cells. Incubation with H4R agonists (up to 10 µmol/l) induced a distinct calcium2+ influx in these cells. C2 cells also responded with enhanced chemotaxis towards histamine, VUF8430 and clobenpropit and this effect was antagonized by JNJ7777120. However, neither VUF8430 nor clobenpropit (up to 10 µmol/l) enhanced histamine liberation from canine mastocytoma cells, whereas the wasp venom peptide mastoparan and concanavalin A used as a positive control, increased histamine concentrations measured in supernatants. Corresponding to this, pre-treatment with JNJ7777120 also did not inhibit histamine release from mastoparan or concanavalin A stimulated C2 cells. In conclusion, H4R antagonists may offer the potential to treat allergic skin disorders in dogs. 1. Dept. of Pharmacology, Toxicology and Pharmacy, University of Veterinary Medicine Hannover, Foundation, Germany, 2. Institute for Pharmaceutical Chemistry, ZAFES/CMP, Johann Wolfgang Goethe-University, Frankfurt/Main, Germany, 3. Leiden/Amsterdam Center for Drug Research, Dept. of Medical Chemistry, Vrije Universiteit Amsterdam, The Netherlands

193 Cholesterol-induced apoptosis is associated with downregulation of the activated leukocyte cell adhesion molecule (ALCAM) in human monocytic cells Böhm A. (1), Rauch S.J. (1), Rosenkranz A.C. (1), Meyer-Kirchrath J. (1), Schrör K. (1), Rauch B.H. (1) Migration of monocytes plays an important role in development and progression of atherosclerosis. The activated leukocyte cell adhesion molecule (ALCAM/CD166) is important for cell migration and leukocyte invasion. The present study investigates the impact of cholesterol-enrichment on the expression of ALCAM in the human monocytic cell line U937. Monocytic U937 cells were enriched with cholesterol by incubation with methyl-beta-cyclodextrin (MbCD)-cholesterol. Expression of adhesion molecules was determined by flow cytometry; apoptosis by Annexin-V FITC/PI (propidium iodide) double-label cytometry. Migration of calcein-AM-loaded U937 cells was quantified in 3µm-chemotaxis chambers by fluorescence of transmigrated cells. Immunohistochemistry for ALCAM was performed in atherosclerotic plaques of ApoE-deficient mice.High expression of ALCAM was observed in atherosclerotic plaques of ApoE-deficient mice. Incubation of monocytic U937 cells with cholesterol (10 - 100 µg/mL) for 18 h induced a concentration-dependent increase in early and late phase apoptosis, while MbCD alone had no effect. Increased apoptosis rate was associated with a reduction of ALCAM expression by >50%. In contrast, expression of VCAM-1 (vascular cell adhesion molecule-1) was strongly increased and ICAM-1 (intercellular cell adhesion molecule-1) levels were not affected by cholesterol loading. Pretreatment with the nonselective caspase/apoptosis inhibitor Q-VD-OPh (100 µM) attenuated cholesterol-induced alteration of adhesion molecule expression. Cholesterol loading was also associated with an 80% reduction of cell migration towards 10% serum. Q-VD-OPh partially rescued the chemotactic response in cholesterol-rich monocytes. This effect was prevented by addition of ALCAM-neutralizing antibodies (10 µg/mL), but not an isotypematched IgG. Cholesterol-induced apoptosis in monocytic cells is accompanied by reduced expression of ALCAM. Loss of ALCAM under conditions of cholesterol-loading and apoptosis attenuates monocyte migration, a mechanism which may expedite vascular lesion formation. 1. Institut für Pharmakologie und Klinische Pharmakologie, Universitätsklinikum Düsseldorf, Düsseldorf

194 A map of the IL-1R signalling network and its response to inflammatory stimuli Weber A. (1), Wasiliew P. (1), Dittrich-Breiholz O. (2), Schneider H. (2), Kracht M. (1) The interleukin-1 (IL-1) signalling pathway represents a paradigm for all IL-1 family cytokines (IL-1α, IL-1β, IL-18, IL-33, IL1F5 to IL1F10) as well as for the toll-like-receptorinduced pathways. More than 20 years of research on IL-1 signalling has generated a large body of experimental data that is scattered across many different publications. No signal transduction map exists that covers comprehensively this huge knowledge on IL1-activated effector molecules. We attempted to improve the publically accessible IL-1 signalling map deposited in Nethpath (http://www.netpath.org/pathways). This map contains 36 molecules and information for a total of 89 reactions including physical interactions, catalytic reactions (e.g. phosphorylations) and subcellular distributions. Netpath uses data from experimental results reported in 46 publications. We imported the Netpath IL-1 signalling map in Pathvisio, Version1.1 and created an updated version that now contains 215 molecules and 408 reactions which were taken from original data

43

contained in 216 selected publications. In addition to visualization of molecules involved in IL-1 signal transduction this map also enables direct access to the scientific evidence that was used to build up the map. By using GenMAPP, Version 2.1 we were able to display mRNA expression levels directly within the IL-1 map to visualize the impact of external stimuli on pathway components. Genome-wide expression profiles obtained from IL-1α, TNF-α, LPS, Virus- or Chlamydia-infected cells showed significant differences in mRNA levels of IL-1 network components, implying that these conditions use multiple ways of modulating the IL-1 response. By analysing data from cells treated with 50µM PD 98059 we were also able to map the part of the IL-1- signalling network that reacts at the level of mRNA to ERK-inhibition. In summary, this type of knowledgebased curated data base serves not only to sort and to store data from biochemical, cell biological and molecular biology experiments but may also helps to depict nodes of signal transduction that can be used to design new hypothesis-driven experiments. 1. Rudolf-Buchheim-Institute of Pharmacology, Giessen, Germany, 2. Institute of Pharmacology, Medical School Hannover, Hannover, Germany

195 AP-1 dimers c-Jun-c-Jun and c-Jun-ATF2 regulate IL-8 transcription differentially Kettner-Buhrow D. (1), Quante T. (2), Kronich P. (1), Wolter S. (2), Bakiri L. (3), Wagner E.F. (3), Kracht M. (1) Activator protein 1 (AP-1) is one of several important transcription factors that regulate the highly inducible chemokine gene Interleukin-8 (IL-8). Putative AP-1 binding sites are found in a large number of genes. AP-1 is a homo- or heterodimer composed of Jun (cJun, JunB, JunD) and Fos (c-Fos, FosB, Fra-1, Fra-2) proteins. These protein members can also dimerize with many other basic leucine zipper proteins (i.e. ATF), increasing the number of the potential AP-1 factors that bind to a given AP-1 site. We have previously characterized an AP-1 site in the IL-8 gene and have shown that various AP1 proteins can bind to this site in response to Interleukin-1 (IL-1). We therefore set up experiments to identify the functional impact of AP-1 dimer combinations on IL-8 transcription. Besides loss of function studies with siRNAs against AP-1 protein members we performed gain of function experiments. We ectopically expressed 11 different AP-1 dimers as single chain proteins, whereby two subunits were fused by a flexible glycine-linker. Dimers composed of c-Jun-ATF-2, c-Jun-Fra-1, c-Jun-c-Fos, JunB-c-Fos strongly activated the IL-8 promoter, whereas c-Jun-c-Jun homodimers and JunD containing dimers were significantly less active. The effects on basal IL-8 transcription were direct because when we used IL-8 reporter gene construct with a mutated AP-1 site, c-Jun-ATF-2 and c-Jun-c-Jun dimers were incapable to activate IL-8 transcription. Moreover, chromatin immunoprecipation experiments revealed binding of both dimers to the ectopically expressed and endogenous IL-8 promoter. In coimmunoprecipitation assays, the homodimer composed of c-JUN-c-JUN but not the cJun-ATF2 heterodimer interacted with histone deacetylase 3 (HDAC3), providing a mechanistic explanation for the compared to other AP-1 dimers weak transcriptional activity of c-JUN-c-JUN. 1. Rudolf-Buchheim-Institute of Pharmacology, Justus-Liebig-University Giessen, Giessen Germany, 2. Institute of Pharmacology, Medical School Hannover, Hannover, Germany, 3. Research Institute of Molecular Pathology, Vienna, Austria

196 Investigation on the molecular mechanisms mediating the inhibitory effect of benzodiazepines on mast cells Meis K. (1), Spitzlei P. (1), von Kügelgen I. (1), Molderings G.J. (2) Benzodiazepines, in particular flunitrazepam, lorazepam and bromazepam, are clinically effective in the treatment of mast cell mediator-induced symptoms in patients with systemic mastocytosis. Since improvement of the symptoms by the benzodiazepines does not seem to be related with their sedative and anxiolytic effects, an involvement of GABA-A receptors in the amelioration of the mast cell mediator-induced symptoms is unlikely. In agreement with this conclusion, mast cells are not endowed with GABA-A receptors. Therefore, the aim of the present study was to identify other potential targets and mechanisms in mast cells at which benzodiazepines may cause an inhibitory effect on mast cell activation. Benzodiazepines may act at peripheral-type benzodiazepine receptors (ptBZR) which are strongly expressed by mast cells. In line with this idea, the selective agonist at ptBZR chlorodiazepam (Ro5-4864) inhibited ionomycin-induced ATP efflux (reflecting mast cell degranulation) from cells of the human mast cell leukemia cell line HMC1. However, at the mRNA level expression of ptBZR in HMC1 cells and mast cells from patients with systemic mastocytosis did not differ from that in mast cells from human healthy volunteers. Sequencing of the cDNA encoding the human ptBZR isolated from mast cells of the individuals of both groups did not show any differences. Next, we investigated in HMC1 cells whether benzodiazepines affect adenosine A2B-receptorinduced changes in transcription factors. Receptor-induced changes in transcription factors were assessed by a reporter gene (luciferase) assay. A2B-receptors were activated by the adenosine analogue NECA (N-ethylcarboxamido-adenosine 0.3-3 µM). Flunitrazepam (30 µM) inhibited the increases in the CRE (cAMP response element)and the NFkappaB (nuclear factor kappa B)-, but not in the NFAT (nuclear factor of activated t cells)-controlled luciferase expression induced by NECA. In addition, flunitrazepam (30 µM) inhibited NECA (1µM)-induced synthesis of interleukin 8 (reflecting mast cell activation by A2B receptors) in HMC1 cells. In chinese ovary cells expressing recombinant human A2B-receptors, flunitrazepam did not change the NECAinduced increases in CRE-controlled luciferase expression excluding a direct effect of flunitrazepam on A2B-receptors. Taken together, our data suggest that the inhibitory effects of benzodiazepines on mast cell activation may be due to an inhibition of the intracellular cAMP- and NFkappaB-pathways. 1. Dept. of Pharmacology, University of Bonn, Bonn, Germany, 2. Institute of Human Genetics, University of Bonn, Bonn, Germany

197 The impact of FTY720 analogs on the intracellular Ca2+ levels in murine Langerhans cells Schröder M. (1), Meyer zu Heringdorf D. (1), Stark H. (2), Pfeilschifter J.M. (1), Radeke H.H. (1) In the last few years there is increasing evidence that endogenous lysophospholipids such as sphingosine-1-phosphate (S1P) may have potent immune modulating properties. The sphingosine analog FTY720 (Fingolimod) is a potent drug which causes lymphopenia and following phosphorylation interacts with all known S1P receptors except S1P2. To avoid the known adverse effects of FTY720, like bradycardia, it would be of high interest to identify analogs which modulate only a smaller set of receptors but are equally immunosuppressive. Herein, we analysed the activity of S1P, FTY720 and seven FTY720 analogs (ST968-ST973) on intracellular Ca2+ levels in the murine Langerhans cell line XS52 with confocal microscopy (live single cell imaging). Recorded dose response curves led to the identification of two different analogs which had a comparable or even 14-fold better EC50 (S1P=46.9nM, FTYP=45.5nM, ST970=3.3nM, ST971=105.5nM). Whether these analogs initiate the Ca2+ release via one or more of the five S1P receptors will be investigated in further experiments with HA-tagged S1P15 receptors and FACS analysis. Taken together the Ca2+-imaging and FACS experiments could provide data to identify new and more selective drugs to address the S1P receptor subtypes being responsible for specific immune modulatory steps. 1. Pharmazentrum Frankfurt / ZAFES, Clinic of the Goethe University, 2. Pharmaceutical / Medicinal Chemistry, Goethe University Frankfurt/Main

198 The impact of sphingolipid enzymes on sphingolipid metabolism and distribution in murine immune cells Schröder M. (1), Schmidt H. (2), Huwiler A. (1), Stark H. (3), Bartoszyk G. (4), Geisslinger G. (2), Pfeilschifter J.M. (1), Radeke H.H. (1) In the last few years there is increasing evidence that endogenous lysophospholipids such as sphingosine-1-phosphate (S1P) may have potent immune modulating properties. To investigate the complex enzymatic network leading to the different sphingolipid levels, it is necessary to clarify the localisation and concentration of those metabolites within the different compartments of the cell. To evaluate a reliable metabolite determination system we used the not endogenously produced S1P analog FTY720. Spleen cells of C57Bl/6 wild type mice were incubated for 15 h with 1 µM FTY720, 15 µM MK571 (suggested ABCC1 transport inhibitor) or 1 µM FTY720 + 15 µM MK571 and then fractionated into supernatant, cytosol, nucleus and membrane. The supernatants of methanol precipitated plasma samples were directly assessed by liquid chromatography-electrospray ionisation-tandem mass spectrometry (LC-ESIMS/MS) and analyzed with HPLC under gradient conditions using a C18 reversed phase column. The LLOQ was 0.175 ng/ml (FTY720) and 0.4 ng/ml (FTY720-P). We were able to recover >80% of the applied FTY720 together with the product FTY720-P, both of which were quite well distributed throughout the cell compartments. Additionally, we observed that application of MK571 reduced the distribution of FTY720-P from cytosol to the supernatant. In future experiments we will also elucidate the compartment levels of endogenous sphingosin/S1P and sphinganin/sphinganin1P in wt samples and compare these with samples of mSK1-/-, mSK2-/-, mS1P-Lyase-/- mutants and S1P-phosphatase 1 and 2 knockdown mutants. Taken together those MS/MS data will give us a detailed insight in the distribution of sphingolipids, the correlating localisation of the enzymes and the overall enzymatic network in immune cells. 1. General Pharmacology, pharmazentrum frankfurt / ZAFES, Clinic of the Goethe University, 2. Clinical Pharmacology, pharmazentrum frankfurt / ZAFES, Clinic of the Goethe University, 3. Pharmaceutical / Medicinal Chemistry, Goethe University Frankfurt/Main, 4. Merck KGaA

199 Tumour-associated autoantigen p62 as a regulator of hepatic IGF2 expression and metabolic dysfunction Tybl E. (1), Shi F.D. (4), Walter J. (2), Tierling S. (2), Bohle R.M. (3), Wieland S. (4), Chisari F. (4), Tan E.M. (4), Kiemer A.K. (1) HCC is one of the most common cancers in the developing countries, but during the last years HCC is also on the advance in the industrial countries (Scharf et al., 2001). The p62 protein belongs to the family of Insulin-like growth factor II (IGF2)-mRNA binding proteins (IMPs) and was first detected as an auto-antigen in malignant cancer cells isolated from patients suffering from HCC (Zhang et al., 1999). In order to study the functions of hepatic p62 expression, transgenic mice were generated showing a liver specific overexpression of the transgene in a doxycycline-regulated manner. Histological staining with Scharlach Red revealed the remarkable phenotype of a fatty liver accompanied by a decrease in glycogen synthesis, proven by Period-Acid-Schiff (PAS) staining, in two out of five p62 transgenic mice. Since it is known that IGF2 contributes to the progression of HCC via metabolic pathways we assessed the impact of the IMP p62 on IGF2 expression. By performing real-time PCR experiments, we could demonstrate that p62 transgenic mice showed a strongly enhanced overexpression of IGF2 and also of H19 mRNA. IGF2 is an imprinted gene and often regulated together with the non-translated H19 mRNA functioning as a tumour suppressor gene. Loss of imprinting (LOI) for those genes has been reported for a number of tumours (Kaneda et al., 2005). In order to find out whether p62 might be a regulator of genomic imprinting, we crossed p62 transgenic mice with SD7 mice (containing the Mus spretus IGF2-H19 region). The results of the single-nucleotide primer extension (SNuPE) assays combined with HPLC (El-Maari et al., 2002) could not verify a biallelic expression and therefore a loss of imprinting (LOI). Focussing on anti-apoptotic downstream targets of IGF2, we aimed to elucidate the expression of the protein kinase B/Akt. Akt phosphorylation is often increased in tumours, whereas simultaneously the expression of the tumor suppressor gene PTEN (phosphatase and tensin homolog) is reduced (Moorehead et al., 2001). We could neither detect any alterations in pAKT nor in PTEN protein and mRNA expression in p62 transgenic mice. As a result that underlines the importance of p62 on IGF2 and H19 regulation also in humans, we showed that in two human hepatoma cell lines (HepG2 and Alexander cells) p62 knock-down by RNA interference (RNAi) was followed by a decrease in IGF2 and H19 mRNA expression levels.

44

Conclusion: The changes in liver metabolism accompanied by the up-regulation of IGF2 and H19 in consequence of p62 overexpression may contribute to HCC development. 1. Dept. of Pharmaceutical Biology, Saarland University, Saarbruecken, Germany, 2. Institute for Genetics, Saarland University, Saarbruecken, Germany, 3. Institute for Pathology, Saarland University Hospital, Homburg/Saar, Germany, 4. Dept. of Molecular & Experimental Medicine, The Scripps Research Institute, La Jolla, USA

200 Common CHK2 promoter polymorphisms do not influence disease course of breast cancer patients Wieker G. (1), Schmid KW. (2), Siffert W. (1), Riemann K. (1) Purpose: Human cells possess a checkpoint kinase 2 (CHK2) dependent mechanism to activate cell cycle arrest in response to DNA damage caused by UV radiation, genotoxic agents, radio- or chemotherapeutics. This provides time for facilitating repair or, if damage is too severe, sent cells into senescence or apoptosis preventing genomic instability and cancer. CHK2 plays an essential role in this signaling cascade by regulating the G1/S-checkpoint and G2/M-checkpoint. A disturbed checkpoint function causes unobstructed proliferation of DNA damaged cells. Mutations in CHK2 are known to be predisposing factors in a variety of cancers, such as breast, colon and prostate cancer. Our aim was to investigate whether common promoter polymorphisms influence breast cancer patients’ disease course. Methods: The promoter region of 10 healthy volunteers was sequenced. Subsequently we focused on the two most frequent polymorphisms, -7161G>A and -7235C>G. For each polymorphism 234 healthy female volunteers and 265 patients were genotyped using Restriction Fragment Length Polymorphism (RFLP). Haplotypes were associated with risk for breast cancer and genotypes with survival, progression and relapse. Functional analysis included reporter assays in HEK293 cells. Results: Survival, progression and relapse of patients with breast cancer were not dependent on either promoter polymorphism. No risk allele could be identified. Additionally, no significant relation could be provided by functional analysis to determine genotype-dependent differences in promoter activity. Conclusions: Taken together, our results indicate that the two most common polymorphisms in the CHK2 promoter region, -7161G>A and -7235C>G, display no functional properties and do not impact on survival, progression and relapse in breast cancer patients. 1. Institute of Pharmacogenetics, University of Duisburg-Essen Medical School, Essen, Germany, 2. Institute of Pathology and Neuropathology, University of Duisburg-Essen Medical School, Essen, Germany

201 Lovastatin attenuates cytotoxic side effects of doxorubicin in vivo Huelsenbeck J. (1), Ostrau C. (1), Fritz G. (1) 3-Hydroxy-3-methylglutaryl-coenzyme A (HMG-CoA) reductase inhibitors (statins) are widely used in the treatment of hypercholesterolemia. They exhibit further beneficial effects on inflammation, platelet aggregation, thrombosis, and endothelial function. Statin treatment causes apoptosis in certain tumour types such as acute myelogenous leukemia. Furthermore, statin cotreatment potentiates the effect of chemotherapeutic drugs (e.g. cisplatin, doxorubicin) on transformed cells and can reverse chemoresistance. Recently, we demonstrated that statin treatment protects human umbilical vein endothelial cells (HUVEC) from the cytotoxic effects of doxorubicin. Here, we extend these findings to the in vivo situation. Balb/C mice were treated with doxorubicin, either alone or in cotreatment with lovastatin. After three weeks, the mice were sacrificed. Analysis of serum parameters revealed that doxorubicin caused an increase of the level of troponin I, as well as GLDH and GPT, indicative of heart and liver damage, respectively. Furthermore, the mRNA level of the fibrosis marker connective tissue growth factor (CTGF) was elevated in heart and liver of doxorubicin treated mice. Cotreatment with lovastatin reduced these effects of doxorubicin significantly. Neither troponin I. nor GLDH nor GPT serum levels were elevated. The CTGF mRNA level was nearly reduced to the level in control mice. Furthermore, mice cotreated with lovastatin and doxorubicin exhibited less loss of weight compared to mice treated with doxorubicin alone. In conclusion, lovastatin attenuated the cytotoxic effects of doxorubicin on non transformed cells in vivo. Regarding cancer treatment using doxorubicin, lovastatin is thus useful in two ways. On the one hand, it either causes death of certain transformed cells or may increase their sensitivity to doxorubicin. On the other hand, lovastatin attenuates some side effects of this chemotherapeutic agent. 1. Dept. of Toxicology, Johannes Gutenberg University, Mainz, Germany

202 Inhibition of Akt induces cell cycle arrest and triggers apoptosis in chemoresistant human prostate cancer cells Morad S.A.F. (1), Schmidt M. (1), Büchele B. (1), Pitterle K. (1), Lunov O. (1), Syrovets T. (1), Simmet T. (1) Akt is a serine/threonine protein kinase that acts as a signal collection node and a downstream target of PI3K. Deregulations of Akt are frequently associated with human diseases including cancer and diabetes. Activation of Akt plays an important role in cancer cell survival and resistance to chemotherapy. It influences cell cycle progression and has, therefore, a critical role in malignant transformation. We have recently identified certain triterpenoids as effective antitumor compounds. Starting from 11-ketoβ-boswellic acid (KBA) isolated from the oleo-gum resin of Boswellia Carterii, we synthesized the semisynthetic derivative 3-cinnamoyl-11-keto-β-boswellic acid (CKBA) by the Steglich reaction. Thin layer chromatography, reversed phase high performance liquid chromatography, mass spectrometry, and one and two dimensional NMR spectroscopy were used for purity control and structural identification. The novel semisynthetic compound CKBA inhibited the kinase activity of AKT1 as shown in an in vitro kinase assay using an active recombinant human AKT1. KBA lacking the cinnamoyl residue did not inhibit the AKT1 kinase activity. Since AKT1 is constitutively active in most androgen-independent prostate cancers, we examined the effect of CKBA on the androgen-independent PC-3 prostate cancer cell line, which is resistant to therapeutic intervention. CKBA exhibited cytotoxic effects on the PC-3 cells, whereas the parent compound KBA showed nearly no effect. This cytotoxicity was due to

apoptosis as analyzed by phosphatidylserine expression on the cell surface and formation of DNA laddering. Moreover, CKBA inhibited proliferation of prostate cancer cells xenografted onto chick chorioallantoic membranes. Using a molecular modeling approach, a binding pocket for CKBA was identified within the pleckstrin homology domain of AKT1. These results revealed that inhibition of Akt activity might be sufficient to trigger apoptosis in prostate cancer cells and indicate that 3-cinnamoyl-11-keto-βboswellic acid could serve as a lead compound for the development of novel chemotherapeutic drugs for the treatment of chemoresistant human prostate cancer. Supported by the Deutsche Krebshilfe. 1. Institute of Pharmacology of Natural Products & Clinical Pharmacology, Ulm University, Ulm, Germany

203 Esophageal cancer cell-derived hyaluronan synthase 3 promotes malignant phenotype and critically regulates tumor progression in vitro and in vivo Twarock S. (1), Freudenberger T. (1), Poscher E. (1), Dai G. (1), Janasch K. (2), Dullin C. (2), Alves F. (2), Stoecklein N. (3), Savani R.C. (4), Homey B. (5), Fischer J.W. (1) Hyaluronic acid is a polysaccharide component of the extracellular matrix in parenchyma and stroma of many cancer types including esophageal cancer. Clinically, esophageal cancer represents a highly aggressive tumor type with poor prognosis resulting in a 5year survival of 5%. In the present study, we demonstrate that HA-synthase 3 (HAS3) is the predominant HAS enzyme in esophageal squamos cell carcinomas (SCC). After xenografting esophageal SCC cells (OSC) in nude mice the small molecule inhibitor of HA synthesis, 4-methylumbelliferone (4-MU), inhibited OSC proliferation, microvascularization and tumor progression. HAS3 was also the predominant isoform in various OSC cell line and knock down of HAS3 using lentiviral expression of shRNA recapitulated effects of systemic 4-MU treatment in the xenograft model. This suggested that HAS3 expressed by OSC cells is critical for tumor progression in vivo rather than the expression of HAS-isoforms by stromal cells. Subsequently, the molecular mechanisms of the tumor inhibitory action of 4-MU and knock down of HAS3 were addressed in OSC cells in vitro. The pericellular HA-matrix surrounding extended actindependent filopodia was inhibited by 4-MU (0.3 mM) and shHAS3 in OSC cells. Furthermore both 4-MU and shHAS3 caused loss of filopodia that was associated with calpain-mediated cleavage of focal adhesion kinase (FAK). These effects were paralleled by reduction of tumor cell proliferation and migration. Furthermore, we showed that not CD44 but RHAMM is the critical receptor to mediate HA-related malignant phenotype of OSC cells. In vitro HAS3 is strongly stimulated by EGF providing a potential target for EGF-inhibitors. In conclusion, HAS3 mediated synthesis of pericellular HA-matrix by OSC cells is required for tumor growth in vivo. The molecular mechanism likely involves the support of a malignant OSC tumor cell phenotype by the pericellular HA-matrix, which is mediated through HAS3 and RHAMM signaling and control of FAK catabolism. Furthermore, it is conceivable that pharmacologic interventions that target either HAS3 mediated HA-synthesis directly or the inducers of HAS3 such as EGF present novel treatment options for esophageal SCC. 1. Institut für Pharmakologie, Universitätsklinikum Essen, Universität Duisburg Essen, Germany, 2. Abteilung Hämatologie/Onkologie, Zentrum Innere Medizin, Universitätsklinikum Göttingen, Göttingen, Germany, 3. Klinik für Allgemein-, Viszeralund Kinderchirurgie, Universitätsklinikum Düsseldorf, Germany, 4. Divisions of Pulmonary & Vascular Biology and Neonatal-Perinatal Medicine, Department of Pediatrics, The University of Texas Southwestern Medical Center, Dallas, Texas, USA., 5. Hautklinik, Universitätsklinikum Düsseldorf, Germany

204 R(+)-methanandamide-induced apoptosis of human cervical cancer cells involves a cyclooxygenase-2-dependent pathway Ramer R. (1), Eichele K. (1), Hinz B. (1) Cannabinoids have received renewed interest due to their antitumorigenic effects. Using human cervical carcinoma cells (HeLa), this study investigates the role of the prostaglandin (PG)-synthesizing enzyme cyclooxygenase-2 (COX-2) in apoptosis elicited by the endocannabinoid analog R(+)-methanandamide (MA). MA led to an induction of COX-2 expression, PGD2 and PGE2 synthesis, accompanied by a substantial decrease of viability and enhanced apoptosis. Cells were significantly less sensitive to MA-induced apoptosis when either COX-2 expression or its activity was suppressed by small interfering (si) RNA and by the selective COX-2 inhibitor NS-398, respectively. COX-2 expression and apoptosis by MA was also prevented by the ceramide synthase inhibitor fumonisin B1, but not by antagonists to cannabinoid receptors and TRPV1. Experiments performed to clarify how COX-2 leads to apoptosis revealed a profound proapoptotic action of PGD2 and its dehydration product, 15-deoxyD12,14 PGJ2 (15d-PGJ2), whereas PGE2 was inactive in this respect. In line with these findings, MA-induced apoptosis was prevented by siRNA targeting lipocalin-type PGD synthase (L-PGDS), which catalyzes the isomerization of PGH2 to PGD2. Moreover, apoptosis by chemotherapeutics, PGD2 and 15d-PGJ2 was suppressed by the peroxisome proliferator-activated receptor γ (PPARγ) antagonist, GW-9662, or by PPARγ siRNA. Finally, a COX-2- and PPARγ-dependent apoptotic mechanism of MA was confirmed in human lung cancer cells (A549) as well as in another cervical carcinoma cell line (C33A). Collectively, this study demonstrates COX-2 induction and synthesis of L-PGDS-derived, PPARγ-activating PGs as a decisive target by which MA induces apoptosis. Supported by DFG (SFB 539, TP BI.6) and Deutsche Krebshilfe e.V. 1. Dept. of Pharmacology and Toxicology, University of Rostock

45

205 Antiproliferative effect of the sesquiterpene lactone arglabin on androgenindependent prostate cancer cells Pitterle K. (1), Morad S.A.F. (1), Büchele B. (1), Syrovets T. (1), Simmet T. (1) In developed countries, prostate cancer is now the most commonly diagnosed male cancer and the third most common cause of death in males. Initially the cancer responds well to androgen deprivation therapy, but most tumors relapse within two years to an incurable androgen-independent state. Here we analyzed the effects of arglabin, a sesquiterpene γ-lactone isolated from Artemisia glabella, a species of wormwood endemic to the Karagandy region of Kazakhstan, on the androgenindependent PC-3 prostate cancer cell line. We demonstrate that arglabin decreases proliferation and viability of the PC-3 cell line as measured by the XTT assay (IC50 = 15.9 µM, 48 h). Treatment with arglabin induced cell apoptosis as revealed by annexin V binding and occurrence of the so-called sub-G1 peak. In addition, arglabin induced accumulation of PC-3 in the G2 phase of the cell cycle. The antiproliferative effect of low arglabin concentrations was further confirmed in the colony formation assay (IC50 = 0.2 µM, 3 weeks) and in vivo in tumors xenografted onto chick chorioallantoic membranes, where it showed no overt systemic toxicity. Arglabin interferes with MAP kinase and Akt signaling pathways, which might represent the molecular mechanisms of the antipropiferative activity of arglabin. Hence, arglabin might be considered as a promising compound for treatment of chemoresistant cancers. 1. Institute of Pharmacology of Natural Products & Clinical Pharmacology, Ulm University, Ulm, Germany

206 Decreased formation of plasminogen activator inhibitor 1 may contribute to the antiinvasive action of cannabidiol on human lung cancer cells Rohde A. (1), Ramer R. (1), Boeckmann S. (1), Hinz B. (1) Although cannabinoids exhibit a broad variety of anticarcinogenic effects, their potential use in cancer therapy may be limited by their psychoactive effects. We therefore evaluated the impact of cannabidiol, a plant-derived non-psychoactive cannabinoid, on the invasiveness of human cancer cells. In human lung cancer cells (A549) cannabidiol caused a time- and concentration-dependent suppression of cell invasion through Matrigel without modifying migration. Inhibition of invasiveness was accompanied by a decrease of plasminogen activator inhibitor-1 (PAI-1) levels in cell culture supernatants. Effects of cannabidiol on either PAI-1 levels and invasion were suppressed by antagonists to CB1 and CB2 receptors as well as to transient receptor potential vanilloid 1 (TRPV1), implying the involvement of cannabinoid receptor and TRPV1-dependent pathways. An essential role of PAI-1 in mediating a proinvasive action on A549 cells was proven by use of recombinant human PAI-1 and PAI-1 siRNA, which led to a concentration-dependent up- and down-regulation of invasiveness, respectively. Finally, the decrease of invasion by cannabidiol was completely blocked by restoration of endogenous PAI-1 levels with recombinant PAI-1 at non-proinvasive concentrations, suggesting a causal link between PAI-1 downregulation and the anti-invasive action of cannabidiol. Altogether, these findings provide evidence for a hitherto unknown cannabinoid receptor and TRPV1-dependent mechanism underlying the anti-invasive action of cannabidiol and imply its use as a possible therapeutic option for the treatment of highly invasive cancers. Supported by DFG (SFB 539, TP BI.6) and Deutsche Krebshilfe e.V. 1. Inst. für Toxikologie und Pharmakologie, Universität Rostock, Rostock, Germany

207 The antiapoptotic protein Bcl-2 is not able to prevent helenalin-induced cell death Hoffmann R. (1), López-Antón N. (1), Dirsch V.M. (2), Vollmar A.M. (1), Rudy A. (1) The antiapoptotic mitochondrial Bcl-2 protein plays a crucial role in controlling apoptosis and enhancing cell survival in response to diverse apoptotic stimuli like chemotherapeutic agents. Interestingly, the sesquiterpene lactone helenalin induces cell death in Bcl-2 overexpressing Jurkat cells (Dirsch, V.M. et al. Cancer Res., 2001). Our aim was to elucidate the character and the signalling pathways of helenalin-induced cell death in this cell line. Thus, we first investigated the effects of helenalin (20 µM) on apoptosis parameters as well as colony formation and protein expression. Western Blot analysis revealed that Bcl-2 is not inactivated by helenalin through phosphorylation and that cell death occurs independently of cytochrome c release. Furthermore, helenalin induced cell death does not depend on apoptosome formation since silencing of Apaf-1 did not abrogate DNA fragmentation. Additionally, the broad spectrum caspase-inhibitor Q-VD-OPh reduced DNA fragmentation after helenalin treatment to only about 50%. In search for other responsible players in helenalin-induced cell death, we focused on Ca2+ and JNK. Pre-treatment with the Ca2+ chelator BAPTA-AM leads to reduced DNA fragmentation caused by helenalin. Furthermore, JNK phosphorylation determined by Western Blot analysis showed that helenalin increased JNK activity after 2 h, which could be reversed by pre-treatment with BAPTA-AM. Our first results indicate JNK as a key target, but its role has to be elucidated in further experiments. Furthermore, it is possible that helenalin induces cell death with participation of autophagy as a general and early protein degradation after helenalin-treatment is observed. Taken together, these combined and complex features make helenalin a promising tool in chemotherapy of highly resistant cancer cells. 1. Department of Pharmacy, Ludwig-Maximilians-University, Munich, Germany, 2. Department of Pharmacognosy, University of Vienna, Austria

208 Purinergic receptor-mediated endocannabinoid production and retrograde signaling in the cerebellar cortex Kovacs F.E. (1), Urbanski M.J. (1), Szabo B. (1) The CB1 cannabinoid receptor is localized on axon terminals and its activation leads to presynaptic inhibition of neurotransmission. During retrograde signaling, the presynaptic CB1 receptor is activated by endogenous cannabinoids (endocannabinoids) synthesized by postsynaptic neurons. The hypothesis of the present work was that activation of

calcium-permeable transmitter-gated P2X purinergic receptors leads to endocannabinoid production and retrograde signaling. Cerebellar slices were prepared from mouse brain and Purkinje cells were patch-clamped. GABAergic inhibitory currents were evoked by electrical stimulation (eIPSCs). Miniature IPSCs were isolated by tetrodotoxin (t-mIPSCs) or cadmium (cd-mIPSCs). Calcium concentrations in Purkinje cells were determined by imaging. A series of experiments indicated that pressure ejection of ATP from a pipette evoked P2X4 receptor-mediated purinergic currents in Purkinje cells. When the purinergic current was >~ 500 pA, calcium spikes appeared additionally, and the intracellular calcium concentration increased strongly. In experiments with calcium spikes, ATP suppressed eIPSCs by 53± 6 %, and this suppression was prevented by the CB1 antagonist rimonabant (10-6 M). The suppression is most likely the consequence of endocannabinoid release from Purkinje cells and subsequent inhibition of GABA release from axon terminals by retrogradely diffusing endocannabinoids. In a further series of experiments with calcium spikes, ATP led to suppression of t-mIPSCs by 30 ± 4 %; this suppression was also sensitive to rimonabant (10-6 M). In these experiments it could not be decided whether calcium influx through P2X4 receptor ion channels or through voltage-gated calcium channels (VGCCs) triggers endocannabinoid production. For answering the question, we blocked VGCCs by cadmium (10-4 M) and recorded cd-mIPSCs. Under this condition, ATP still elicited a significant calcium increase in the Purkinje cells. Simultaneously, ATP suppressed cdmIPSCs by 38 ± 16 %, and this suppression was sensitive to rimonabant (10-6 M). The results show that activation of P2X4 receptors leads to calcium increase in Purkinje cells. This calcium increase evokes endocannabinoid production and endocannabinoidmediated retrograde suppression of GABA release from axon terminals. Activation of VGCCs in the case of strong purinergic currents greatly enhances the increase of intracellular calcium, endocannabinoid release and endocannabinoid-mediated retrograde signaling. 1. Dept. of Pharmacology, Albert-Luwig-University, Freiburg, Germany

209 Inhibition of LTP in the temporoammonic input to CA1 pyramidal cells is modulated by HCN2 in stratum oriens interneurons Matt L. (1), Michalakis S. (2), Biel M. (2), Ludwig A. (3), Hofmann F. (1), Kleppisch T. (1) The temporoammonic (TA) pathway conveying inputs from layer III of the entorhinal cortex directly to distal dendrites of hippocampal pyramidal cells in the stratum lacunosum moleculare is vital for precise firing in CA1 place cell populations. HCN1 channels constrain long-term potentiation (LTP) in this input. We sought to elucidate the role of HCN2 channels, the second major HCN isoform in pyramidal neurons, for hippocampal LTP. Mimicking the phenotype of HCN1-deficient mice, HCN2 null mutants (HCN2-/-) displayed enhanced LTP in the TA pathway compared to the wildtype (WT), while Schaffer collateral LTP was normal. But unlike HCN1-deficient mice, mice with a deletion of the HCN2 gene in principal hippocampal neurons show normal LTP arguing against a pyramidal neuron-associated cause for the phenotype of HCN2 null mutants. Immunohistochemical stainings proved efficient deletion of the HCN2 gene in pyramidal neurons of the conditional knockout and revealed HCN2 immunoreactivity in cells located in the stratum oriens which were co-stained by a somatostatin antibody a marker of oriens-lacunosum moleculare (O-LM) interneurons. We hypothesized that HCN2 channels regulate inhibitory modulation of the TA input by O-LM interneurons axons of which terminate onto distal dendrites of pyramidal neurons. Fittingly, the difference in LTP between HCN2-/- and WT mice was abolished by the GABAA receptor antagonist picrotoxin which selectively increased LTP in the WT to the level observed in the mutants. Further supporting our hypothesis, picrotoxin increased the slope of basal field EPSPs in the TA pathway of the WT, but not in the HCN2-/- mice. HCN2-/- mice exhibited a significantly reduced amplitude of hyperpolarization-activated currents (Ih) in O-LM interneurons resulting in a hyperpolarizing shift of their resting potential, while action potential threshold remained unaltered. Importantly, the frequency of spontaneous inhibitory postsynaptic currents recorded from CA1 pyramidal neurons was significantly decreased in HCN2-/- mice compared to that in the WT. This genotype difference was precluded by a selective blocker of Ih. Our findings suggest that HCN2 channels regulate the excitability of O-LM interneurons and, thus, modulate synaptic plasticity in the TA input onto distal dendrites of CA1 pyramidal neurons via feed-back inhibition. 1. Institut für Pharmakologie und Toxikologie, TU München, München, Germany, 2. Munich Center for Integrated Protein Science and Department of Pharmacy – Center for Drug Research, Ludwig-Maximilians-Universität-München, München, Germany, 3. Institut für Experimentelle und Klinische Pharmakologie und Toxikologie, FAU ErlangenNürnberg, Erlangen, Germany

210 Reelin controls neurite and growth cone-motility in cortical neurons Leemhuis J. (1,2), Bouche E. (2), Schwan C. (1), Frotscher M. (2,3), Herz J. (2,4), Meyer D.K. (1,2), Bock H.H. (2,5) Reelin, a large secreted signaling protein, coordinates the positioning of neurons during brain development by signaling through a conserved signaling pathway, consisting of the very low-density lipoprotein receptor (Vldlr) and apolipoprotein E receptor-2 (ApoER2), the cytoplasmic adaptor protein Disabled-1 (Dab1) and Src family kinases (SFK). Activation of this Reelin signaling pathway is crucial for neuronal positioning and the lamination of the cortex and the cerebellum. By controling the stability and assembly of the actin cytoskeleton phosphatidylinositol 3-Kinases (PI3Ks) and Rho-GTPases regulate neuronal morphogenesis and migration. There is evidence for divergent roles of the two Reelin receptors Vldlr and ApoER2, with Vldlr mediating a stop signal for migrating neurons and ApoER2 being essential for the migration of neocortical neurons. We report here by using several knock-out models of the Reelin signaling pathway and time-lapse-microscopy that Reelin induces growth-cone and neurite motility in stage II cultivated cortical neurons by activation of ApoER2 in a Dab1- and PI3K- dependent manner. Vldlr was not involved. Rho-GTPases effector pulldown assays and FRET imaging techniques showed that Rho-GTPases mediate the effects of Reelin to the actin cytoskeleton. In summary we were able to show that Reelin activates a defined signalling pathway to the actin cytoskeleton, which results in an increased growth conemotility. This signaling pathway, which connects ApoER2 to the actin cytoskeleton, may also be involved in other Reelin induced cellular effects. This work was supported by the

46

Forschungskommission of the University Hospital Freiburg and the DeutscheForschungs-Gemeinschaft 1. Dept of Pharmacology, Albert-Ludwigs-Universität Freiburg, Germany, 2. Center for Neuroscience Albert-Ludwigs-Universität Freiburg, Germany, 3. Institute of Anatomy and Cell Biology Department of Neuroanatomy Albert-Ludwigs-Universität Freiburg, Germany, 4. Department of Molecular Genetics, UT Southwestern Medical Center at Dallas, USA, 5. Department of Medicine II Albert-Ludwigs-Universität Freiburg, Germany

211 Lithium modulates pneumolysin-induced rapid loss of mitochondrial potential Djannatian J.R. (2), Wouters F.S. (2), Iliev A.I. (1) Pneumolysin is a member of the cholesterol-dependent cytolysin toxin group and a major pathogen factor of Streptococcus pneumoniae. It is known to produce rapid cell lysis at higher concentrations and apoptosis at lower concentrations. Here, we show that pneumolysin produces early mitochondrial membrane depolarization, preceding cell permeabilization. Application of lithium at therapeutic concentrations prevented mitochondrial membrane depolarization despite the ongoing macropore formation at the cell membrane. This effect was GSK3beta-independent, but calcium influx-dependent. We speculate that lithium might preserve early mitochondrial changes due to the effect of pore-forming toxin and might prove useful as a therapeutic agent in the course of pneumococcal meningitis. 1. DFG Membrane/Cytoskeleton Interaction Group, Institute of Pharmacology and Toxicology / Rudolf-Virchow Center, Würzburg, Germany, 2. Cell Biophysics, Department of Physiology, Medical Faculty, University of Göttingen, Germany

212 Effect of NCAM and erythropoietin mimetic peptides on neurogenesis in status epilepticus models Fuest C. (1), Pekcec A. (1), Bock E. (2), Berezin V. (2), Potschka H. (1) Purpose: Mimetic peptides such as neural cell adhesion molecule (NCAM)- and erythropoietin-derived peptides have been shown to influence neuronal plasticity in different animal models of CNS diseases. Therefore, we studied the effect of plannexin that mimics NCAM function and of the erythropoietin mimetic peptides epotris and epobis in different animal models of temporal lobe epilepsy. We concentrated on seizure-associated alterations in hippocampal neurogenesis which have been hypothesized to contribute to the development and progression of epilepsy. Methods: Female Wistar or Spraque Dawley rats were treated with either plannexin, epotris, epobis, or vehicle (10 mg/kg s.c.). A status epilepticus (SE) was induced by administration of pilocarpine or by electrical stimulation of the basolateral amygdala on the third day of the treatment protocol. The effects of the mimetic peptides on neuronal progenitor cells and on alterations of hippocampal neurogenesis were investigated by doublecortin (DCX)-immunohistochemistry at different time points following SE. Data were compared with control data from animals without SE. Results: In the pilocarpine model, SE decreased the number of DCX-labeled cells in the dentate gyrus and enhanced the number of persistent basal dendrites 48 h post-pilocarpine administration. In the electrical model, SE increased the number of DCX-labeled cells in the dentate gyrus and enhanced the number of persistent basal dendrites 14 d post stimulationem. Plannexin and epotris treatment attenuated the SE-induced changes of hippocampal neurogenesis. Conclusion: The results indicate that a SE is associated with changes and alterations in hippocampal neurogenesis. Plannexin and epotris partially antagonize the seizure-associated modulation in hippocampal neurogenesis. 1. Inst. of Pharmacology, Toxicology, and Pharmacy, Ludwig-Maximilians-University Munich, Munich, Germany, 2. Protein Laboratory, Inst. of Molecular Pathology, Panum Institute, University of Copenhagen, Copenhagen, Denmark

214 Exogenous and endogenous cannabinoids suppress GABAergic synaptic transmission in the human neocortex Knop T. (1), Urbanski M.J. (1), Kovacs F.E. (1), Freiman T. (2), Feuerstein T.J. (2), Szabo B. (1) The Gαi/o protein-coupled CB1 cannabinoid receptor is found at high density in many brain regions. Activation of CB1 receptors on axon terminals by exogenous cannabinoids (for example, ∆9-tetrahydrocannabinol) and by endogenous cannabinoids (endocannabinoids) leads to presynaptic inhibition of neurotransmission (Szabo & Schlicker, Handb Exp Pharmacol 168:318–56, 2005; Chevaleyre et al., Ann Rev Neurosci 29:37-75, 2006). The endocannabinoids are usually released by postsynaptic neurons and diffuse retrogradely to the presynaptic axon terminals (this process is called endocannabinoid-mediated retrograde signaling). The function of presynaptic CB1 receptors has been frequently studied in rat and mouse brain slices. The aim of the present study was to characterize the effect of exogenous and endogenous cannabinoids on GABAergic synaptic transmission in the human neocortex.Cortical tissue surgically removed in order to gain access to pathological brain areas was used. 250 µm-thick slices were cut with a vibratom and superfused. Spontaneous GABAergic inhibitory postsynaptic currents (sIPSCs) were recorded in neocortical pyramidal neurons with patch-clamp techniques. In order to enhance the activity of cannabinoidsensitive presynaptic axons, muscarinic receptors were continuously stimulated by carbachol (5 x 10-6 M). The synthetic cannabinoid receptor agonist WIN55212-2 ((+)[2,3-dihydro-5-methyl-3-[(morpholinyl)-methyl]-pyrrolo[1,2,3-de]-1,4-benzoxazin-yl]-(1naphthalenyl)-methanone mesylate; 5 x 10-6 M) inhibited sIPSCs by 52 ± 11 %. The CB1 antagonist rimonabant (10-6 M) alone had no effect on sIPSCs, but prevented the effect of WIN55212-2. Endocannabinoids were released from postsynaptic pyramidal neurons by 9 depolarizing pulses (from -70 mV to 0 mV for 100 ms) at a frequency of 1 Hz. This led to a suppression of sIPSCs by 26 ± 5 %; rimonabant (10-6 M) prevented this suppression.This is the first demonstration that an exogenous cannabinoid causes presynaptic inhibition of GABAergic synaptic transmission in the human neocortex. We also show for the first time that endocannabinoids released by postsynaptic neurons lead to retrograde synaptic suppression in the human brain. Retrograde signaling is one form of short-term synaptic plasticity. The interference of exogenous cannabinoid agonists and antagonists with synaptic plasticity in the human neocortex suggests that these compounds will interfere with memory and learning in humans. 1. Dept. of Pharmacology, Albert-Ludwigs-University, Freiburg, Germany, 2. Dept. of Neurosurgery, Albert-Ludwigs-University, Freiburg, Germany

215 Interaction of antipsychotic drugs with the four histamine receptor subtypes Appl H. (1), Seifert R. (2) The local mediator and neurotransmitter histamine plays an important (patho)physiological role in a number of processes by activating H1-, H2-, H3- and H4receptors (HxRs). H1-3Rs are already well examined with potent and selective agonists and antagonists being available. In contrast, the function and pharmacological properties of the H4R are still incompletely understood. Antipsychotics show affinity to HxRs, mostly to the H1R which is known to cause the sedative side effects of these compounds. Hence, we asked the question whether antipsychotic drugs also interact with other HxRs, thereby contributing to potentially desired or unwanted effects. In order to better understand the interactions between these compounds and HxRs, we expressed the different histamine receptor subtypes in Sf9 insect cells. We determined the affinities (Ki-values) of various antipsychotics by performing radioligand binding studies using [³H]mepyramine (H1R), [³H]tiotidine (H2R), [³H]N-α-methylhistamine (H3R) and [³H]histamine (H4R) as radioligands (Tab. 1): hH1R [nM] hH2R [nM] hH3R [nM] hH4R [nM]

1 100 21,000 n.d. 1 27,000 5,700 n.d. Chlorpromazine 2 10,900 7,000 3,000 Chlorprothixene 8 2,300 7,000 2,900 Clozapine 3 1,300 22,500 500 N-Desmethylclozapine 4 2,000 32,000 530 Fluphenazine 5 9,700 8,000 70,000 Loxapine 3 5,400 2,900 n.d. Promethazine 6 1,300 36,000 n.d. Thioridazine 3 320 3,400 8,800 Tab. 1: Ki-values of antipsychotics determined by radioligand binding studies; n.d.=not determined Most of the antipsychotics bound to the H3R and H4R with mo-derate to low affinities (Kivalues from 500 nM to 32 µM), whereas some of them showed higher affinity to H1R and H2R (Ki-values from 1 nM up to 320 nM). At therapeutically relevant concentrations (7302,300 nmol/L), the atypic antipsychotic clozapine antagonized the properties of histamine at H1R (Ki-value 3 nM) and H2R (Ki-value 1,300 nM). Partial H4R agonism of clozapine (Ki-value 500 nM; EC50-value 510 nM, Emax 72%) may contribute to its agranulocytosis-inducing properties. At therapeutically relevant concentrations (360-900 nmol/L and 270-540 nmol/L, respectively), amitriptyline and thioridazine also blocked the H2R. Thus, these properties may be exploited in patients suffering from psychiatric diseases and gastroduodenal ulcer. The lipophilicity of antipsychotics facilitates penetration of the blood-brain barrier. Accordingly, the affinity of amitriptyline and thioridazine to H2Rs in the CNS may contribute to their anti-psychotic effects. 1. Dept. of Pharmacology and Toxicology, University of Regensburg, Regensburg, Germany, 2. Inst. for Pharmacology, Medical School Hannover, Hannover, Germany Amitriptyline Amoxapine

213 Differences in sensitivity to the convulsant pilocarpine in substrains and sublines of C57BL/6 mice Müller C.J. (1,2), Gröticke I. (1,2), Hoffmann K. (1), Schughart K. (3), Löscher W. (1,2) In rodents, the cholinomimetic convulsant pilocarpine is widely used to induce status epilepticus (SE), followed by hippocampal damage and spontaneous recurrent seizures, resembling temporal lobe epilepsy. This model has initially been described in rats, but is increasingly used in mice, including the C57BL/6 (B6) inbred strain. In the present study, we compared the effects of pilocarpine in three B6 substrains (B6JOla; B6NHsd; B6NCrl) that were previously reported to differ in several behavioral and genetic aspects. In B6JOla and B6NHsd, only a small percentage of mice developed SE independently of whether pilocarpine was administered at high bolus doses or with a ramping up dosing protocol, but mortality was high. The reverse was true in B6NCrl, in which a high percentage of mice developed SE, but mortality was much lower compared to the other substrains. However, in subsequent experiments with B6NCrl mice, striking differences in SE induction and mortality were found in sublines of this substrain coming from different barrier rooms of the same vendor. In B6NCrl from Barrier #8, administration of pilocarpine resulted in a high percentage of mice developing SE, but mortality was low, whereas the opposite was found in B6NCrl mice from four other barriers of the same vendor. Breeding experiments strongly indicate an X-chromosome linked genetic variation as the cause of the observed phenotypic alterations. These differences in B6 substraines and sublines offer a unique opportunity to identify the genes and pathways involved and contribute to a better understanding of the underlying molecular mechanisms of seizure susceptibility. 1. Dept. of Pharmacology, Toxicology, and Pharmacy, University of Veterinary Medicine, Hannover, Germany, 2. Center for Systems Neuroscience, Hannover, Germany, 3. Dept. of Experimental Mouse Genetics, Helmholtz Centre for Infection Research & University of Veterinary Medicine Hannover, Braunschweig, Germany

47

216 +

Functional tetrodotoxin-resistant Na channels are expressed presynaptically in rat DRG neurons Medvedeva Y.V. (1), Kim M. (1), Schnizler K. (2), Usachev Y.M. (1) The tetrodotoxin-resistant (TTX-R) voltage-gated Na+ channels Nav1.8 and Nav1.9 are expressed by a subset of primary sensory neurons and have been implicated in various pain states. Although recent studies suggest involvement of TTX-R Na+ channels in sensory synaptic transmission and spinal pain processing, it remains unknown whether TTX-R Na+ channels are expressed and function presynaptically. We examined expression of TTX-R channels at sensory synapses formed between dorsal root ganglion (DRG) and spinal cord (SC) neurons in a DRG/SC co-culture system. Immunostaining showed extensive labeling of presynaptic axonal boutons with Nav1.8and Nav1.9-specific antibodies. Measurements using the fluorescent Na+ indicator SBFI demonstrated action potential-induced presynaptic Na+ entry that was resistant to TTX but was blocked by lidocaine. Furthermore, presynaptic [Ca2+]i elevation in response to a single action potential was not affected by TTX in TTX-resistant DRG neurons. Finally, glutamatergic synaptic transmission was not inhibited by TTX in more than 50 % of synaptic pairs examined; subsequent treatment with lidocaine completely blocked these TTX-resistant excitatory postsynaptic currents (EPSCs). Taken together, these results provide evidence for presynaptic expression of functional TTX-R Na+ channels that may be important for shaping presynaptic action potentials and regulating transmitter release at the first sensory synapse. 1. Dept. of Pharmacology, University of Iowa Carver College of Medicine, Iowa City, IA, USA, 2. Institut für Pharmakologie und Toxikologie, RWTH Aachen, Aachen, Germany

217 NK1 and NK2 tachykinin receptors mediate the central responses to neurokinin A. A behavioural study in mice with genetic disruption of the NK1 receptor Culman J. (1), Tauer U. (1), Hunt S.P. (2), Zhao Y. (1) The NK1 and NK2 tachykinin receptors preferentially interact with substance (SP) and neurokinin A (NKA), respectively. Intracerebroventricular (ICV) or intrahypothalamic injections of SP or NKA induce grooming behaviour including face washing (FW), hindquarter grooming and biting (HG) and wet-dog-shakes (WDS) and increase blood pressure and heart rate. NK2 receptors are not expressed in the adult brain. The cardiovascular and behavioural responses to SP and NKA are identical indicating that NKA in the brain may exert its effects through activation of NK1 receptors. Pharmacological experiments have demonstrated that a small population of NK2 receptors is present in some areas of the adult brain. We used NK1R-/- mice lacking the NK1 receptor and the high-affinity, non-peptide NK1 and NK2 receptor antagonists, RP 67580 and SR 48968, respectively to investigate which type of tachykinin receptors mediate the behavioural effects of ICV administered NKA. ICV cannulae were implanted 5 days before the experiments. Experiment 1: A single dose of SP, NKA (50 pmol) was injected into the lateral ventricle in NK1R-/- and wild-type (WT) mice. The frequency of FW, HG and WDS was assessed over a 20 min period at 15-s interval starting immediately after ICV injection of the peptides. Experiment 2: SP or NKA (50 pmol) were injected ICV in WT mice 10 min after pretreatment either with the selective NK1 receptor antagonist, RP 67580 (2000 pmol) or with the selective NK2 receptor antagonist, SR 48 968 (2000 pmol). The behavioural response to the peptides was recorded over a period of 20 min. In WT mice, SP and NKA induced an intense and identical grooming behaviour. However, no responses to the peptides were observed in NK1R-/- mice, indicating that SP and NKA require the NK1 receptor to induce behavioural effects. The grooming behaviour to SP was abolished by the NK1 receptor antagonist, RP 67580, but not by the NK2 receptor antagonist, SR 48968. The NK1 receptor antagonist, RP 67580, did not affect the behavioural response to NKA which, in turn, was effectively inhibited by the NK2 receptor antagonist, SR 48968. Our data indicates that NKA interacts with the NK1 receptor to induce behavioural effects in the mouse adult brain, but its actions can only be effectively inhibited by a selective blockade of NK2 receptors. We conclude that NKA exerts its central effects through an interaction between the NK1 and NK2 tachykinin receptors in the brain. 1. Inst. of Exper. and Clin. Pharmacology, University Hospital of S-H, Campus Kiel, 2. Dept. of Anatomy, University College London, London, United Kingdom

218 Atomoxetine treatment during periadolescence induces long-lasting effects on physical development, anxiety-related behaviour and central monoamine transmission in adult rats Schmidt N. (1), Fink H. (1), Rex A. (1) Objective: Attention Deficit Hyperactivity Disorder (ADHD) is the most prevalent psychiatric disorder in childhood. Pharmacotherapy of ADHD includes the use of stimulants like methylphenidate and, recently, the nonstimulant selective norepinephrine reuptake inhibitor atomoxetine. Monoamine reuptake inhibitors are known to induce long-lasting changes in anxiety- and depression-like behaviour in adult rats following postnatal treatment (Pharmacol Biochem Behav, 1987:367-9). In addition, atomoxetine augmented apoptotic cell death in rat embryonic cortical and hippocampal neurons in vitro (Nervenarzt, 2007:S2 206-7). Question: Might atomoxetine treatment in periadolescence pose a risk for behavioural changes in later life? Methods: In male and female Sprague-Dawley rats atomoxetine (10; 30 mg/kg) was administered once daily on days 28-43 postnatal (dpn). Body mass was measured weekly and voluntary motor activity in the home cage was controlled using motion detectors. Individual water- and food intake was registered. Anxiety-related behaviour and exploratory behaviour of the animals was tested in the Elevated Plus Maze (EPM, 50 dpn), the Open Field (OF, 64 dpn) and the Free Exploratory Paradigm (FEP, 78+79 dpn). Tissue levels of dopamine, serotonin and noradrenaline were determined in cortex, striatum, hippocampus and raphe region at day 101 pn. Results: Atomoxetine administered during periadolescence had no effect on food intake but increased weight gain in female rats (10 mg/kg). In female rats atomoxetine reduced entries into and time spent in the open arms of the EPM (10 mg/kg) and entries into the inner area of the OF (30 mg/kg). It had no effect in the FEP. In male rats atomoxetine delayed the first escape from the home cage and decreased the number of explorations in the FEP (10 mg/kg) but had no effect in EPM or OF. Atomoxetine did not affect motor activity in the rats. Determination of

neurotransmitter levels revealed an atomoxetine-related decrease of dopamine in the cortex of adult males and females and in the striatum in females as well as a decrease of serotonin in the raphe nuclei of adult males and females.Conclusion: In rats atomoxetine-treatment during periadolescence seems to have gender-specific effects on anxiety related behaviour and enduring effects on monamine levels in drug free adulthood. Based on our results behavioural alterations caused by atomoxetine in man might not be excluded. 1. Institute of Pharmacology and Toxicology, Freie Universität Berlin, Berlin, Germany

219 Metabolic and behavioral effects of rimonabant on rats Gogakos A. (1), Chourdakis M. (1), Pourzitaki Chr. (1), Kouvelas D. (1) Background: Cannabinoid receptors type 1 (CB1) are expressed in organs and tissues that regulate energy homeostasis, food intake, glucose and lipid metabolism as well as in Central Nervous System (CNS) that regulate cognition, memory and emotion. Perturbations of the activity of the endocannabinoid system (ECS) may contribute to the development of obesity, dyslpidemia, impaired glucose metabolism, but also memory, mood and anxiety disorders. Pharmacologic modulation of ECS activity holds therapeutic promise to treat a wide range of diseases and pathological conditions referring to energy balance, metabolism, cognition and emotionality. Rimonabant is the first selective CB1 receptor blocker. Although the role of CB1 receptor blockade in the setting of activated ECS is widely studied, little is known about its role in the setting of inactivated ECS. Aim: The aim of this study is to evaluate the effects of rimonabant at low doses on the metabolic and behavioral status of lean rats.Material and methods: 18 male Wistar rats were devided in two groups; the study group (RMN) received rimonabant (1mg/kg/day, i.p.), while the control group (CTL) received same dose of placebo. Parameters studied included body weight, metabolic risk factors (glucose, tcholesterol, triglycerides, HDL-C, HbA1c). All rats experienced a 24h stay in metabolic cages and were exposed to an elevated plus-maze test (EPMT) and an olfactory social memory test (OSMT) to evaluate their anxiety and cognitive behaviors. Results: Significant improvement in glucose levels (CTL 163,33±4,81-RMN 133,83±3,86 p=0,003) and triglycerides levels (CTL 182,67±48,22-RMN 101,17±6,96 p=0,044) was seen in RMN group, while there were no differences in t-cholesterol, HDL-C and HbA1c levels between the two groups. RMN group demonstrated decreased food intake (CTL 14,42±0,97-RMN 13,19±1,39) and weight gain (CTL 70±11,55-RMN 50±7,75) although not at significant levels. RMN group spent more time in the closed arms in the EPMT (CTL 151,17±27,16-RMN 223,0±9,73 [p<0,05] and demonstrated higher ratio of recognition to acquisition time of the juvenile during the OSMT (CTL 2,9±0,84-RMN 6,62±1,83) than the CTL group. Conclusions: Rimonabant at low doses in the setting of inactivated ECS leads to improvement in a number of markers, indicative of high metabolic risk, as well as increased anxiety-like behavior and moderately impaired memory. 1. Dept. of Pharmacology, School of Medicine, Aristotle University of Thessaloniki, Greece

220 Effects of pioglitazone on interleukin-1β and interleukin-1 receptor antagonist expression in the frontoparietal cortex after focal cerebral ischemia in rats Zhao Y. (1), Stöck I. (1), Glatz T. (1), Gohlke P. (1), Culman J. (1) Interleukin- 1β (IL-1β) has a prominent role in the initiation of neurodegenerative reactions leading to the neuronal death after ischemic stroke. IL-1β exerts its actions via the plasma membrane IL-1 type-I receptor (IL-1RI). The naturally occurring competitive IL-1 receptor antagonist (IL-1ra) antagonises the harmful effects of IL-1β after ischemic or excitotoxic brain injury. A number of studies have demonstrated that treatment with PPARγ ligands counteracts the harmful effects of brain ischemia. To study the neuroprotective mechanisms of the selective PPARγ ligand, pioglitazone, we investigated its role in the regulation of IL-1β, IL-RI and IL-1ra expression after ischemic injury. Pioglitazone, (3 nmol/h) or vehicle (controls) were infused intracerebroventricularly (ICV) via osmotic minipumps in male Wistar rats, over a 5-day period before, and 24 or 48 h after transient middle cerebral artery occlusion (MCAO) for 90 min. The effects of pioglitazone on the induction IL-1β, IL-1RI and IL-1ra in the frontoparietal cortex at the border to the ischemic core were studied by immunohistochemistry, immunofluorescence staining and Western blot analysis at 24 and 48 h after MCAO. Cerebral ischemia increased the production of IL-1β, IL-1RI and, to a lesser extent, IL-1ra at both time points after MCAO. The majority of microglial cells/macrophages and few neurons showed an intense IL-1β immunoreactivity. IL-1RI and IL-1ra was localised in both, activated microglia/macrophages and neurons. ICV treatment with pioglitazone considerably reduced the expression and the numbers of cells positively stained for IL-1β in the peri-focal cortical areas at both time points after MCAO, the induction of the IL-1RI was not affected. In contrast to IL-1β, pioglitazone augmented the production of IL-1ra at 48 h after cerebral ischemia. Experiments carried out in primary neuronal cell cultures have revealed that the pioglitazone-induced increase in IL-1ra production is not mediated by PPARγ. Our data demonstrates that pioglitazone reduces the over-expression of IL-1β, which acts pro-inflammatory and upregulates IL-1ra, which counteracts the harmful effects of IL-1β and acts neuroprotective. PPARg ligands suppress the post-ischemic inflammatory reactions and improve the recovery from ischemic stroke by inhibiting the production of proinflammatory cytokines and by enhancing the expression of anti-inflammatory proteins, such as IL-1ra. 1. Inst. of Exper. and Clin. Pharmacology, University Hospital of S-H, Campus Kiel

48

221 Tacrine/NO hybrid molecules with improved pro-cognitive properties and without hepatotoxicity – a new strategy in Alzheimer’s disease? Appenroth D. (1), Lupp A. (1), Decker M. (4), Fang L. (5), Kiehntopf M. (3), Deufel T. (3), Zhang Y. (5), Lehmann J. (2), Fleck C. (1) The multifactor nature of Alzheimer's disease supports the most current therapeutic approach based on the “one molecule, multiple targets” paradigm. The multiple target approach includes novel tacrine hybrids acting as potent cholinesterase inhibitors and nitric oxide (NO)-donors. The highly potent AChE inhibitor tacrine suffers from hepatotoxicity limiting its therapeutic utility. Some NO-donors have been described as hepatoprotective agents. Therefore, hybrid molecules were synthesized that covalently connect tacrine via an alkylenediamine spacer to organic nitrates. The two hybrids tested (FL16 and FL38) surpassed tacrine’s inhibitory activity at AChE with lower activity at BChE. Vasorelaxant effects of the hybrids were tested on porcine pulmonary arteries exhibiting EC50 values of 3.7 to 22.5 mM. For hepatotoxicity testing ASAT, albumin in serum, protein concentration in liver, hepatic biotransformation capacity and liver morphology were investigated. Tacrine caused increased activity of ASAT as well as reduced levels of albumin and protein. Both FL16 and FL38 did not alter these parameters. Morphological studies revealed pericentral necrosis and fatty degeneration of hepatocytes after tacrine. Neither FL16 nor FL38 caused such histopathological changes. Tacrine as well as FL16 and especially FL38 caused a distinct induction of CYP P450 1A1, 2B1 and 3A2 enzymes expression. Respective CYP activities, however, representing the net effect between toxicity and enzyme induction, were significantly increased after FL38 treatment only. To determine pro-cognitive activity of the hybrids cognition impairment of rats after application of scopolamine was evaluated in an eightarm radial maze. This effect was significantly attenuated after administration of either tacrine or the NO-tacrine hybrids. Additionally, tacrine administration without prior scopolamine administration caused a significant slowdown of the rats’ performance in the maze, whereas this was not the case for FL16 and FL38. In summary, NO-tacrine hybrids show vasorelaxant properties, increased ChE-inhibition, in vivo pro-cognitive activity, and lack the pronounced hepatotoxicity of tacrine. 1. Institute of Pharmacology, Friedrich Schiller University, Jena, Germany, 2. Dept. of Pharmacy, Friedrich Schiller University, Jena, Germany, 3. Dept. of Clinical Cemistry, Friedrich Schiller University, Jena, Germany, 4. School of Pharmacy, Queen's University, Belfast, U.K., 5. Centre of Drug Discovery, Pharmaceutical University Nanjing, PR China

222 Neuroprotective effects of bilobalide in brain ischemia Lang D.L. (1), Klein J.K. (1) Stroke is the third leading cause of death in the industrialized world and the treatment options are very limited. In animal experiments numerous compounds were found to be neuroprotective, but clinical trials yielded disappointing results. Bilobalide, a constituent of Ginkgo biloba leaf extracts, is well known to possess neuroprotective properties in various cell models. In brain ischemia, we showed in earlier experiments that 10 mg/kg bilobalide, given i.p. 1 hour before stroke, reduces infarct area in mice. The purpose of the present study was to test the dose-response relationship of bilobalide. Stroke was induced in mice by permanent middle cerebral artery occlusion (MCAO), an established in vivo model for brain ischemia. Blood flow reduction was >85% as controlled by Laser doppler. Different doses of bilobalide (0.3-10 mg/kg) were injected i.p. in female CD-1 mice 1 h before stroke. 10% DMSO in saline was used as control. 24 hrs after MCAO, the infarcted area was stained by 2,3,5-triphenyltetrazolium chloride (TTC) and quantified together with edema formation by image analysis using ImageJ. In these experiments, control animals (n= 6) sustained a large stroke ranging from the forebrain to the cerebellum. The TTC staining showed a distinct white infarct area which spread from the striatum to the cortex and was separated from the healthy tissue by a clear boundary. These mice had serious problems with coordination and movement one day after stroke induction. In comparison, mice that were treated with bilobalide (n= 4-6 per group) were prominently protected against stroke at the doses of 1, 3 and 10 mg/kg as judged by behaviour and TTC staining. Only two mice in the 10 mg/kg bilobalide group and one mouse each in the 3 and 1 mg/kg dosage group sustained a visible, small stroke in the back of the brain while the other treated mice had no stroke. Moreover, the infarct area in the bilobalide-treated mice was situated only in the striatum and in the hippocampus, but not in the cortex as observed in control mice. The reduction of infracted area was > 50% in the treatment groups with 1, 3 and 10 mg/kg bilobalide. Lower doses of bilobalide (0,3 and 0,1 mg/kg) had limited protective effects, and the stroke area looked similar as in control animals ranging from the forebrain to the cerebellum. Based on these findings, pretreatment with bilobalide has neuroprotective effects in doses as low as 1-3 mg/kg. 1. Department of Pharmacological Sciences, School of Pharmacy, Goethe University, Frankfurt am Main, Germany

223 Influenza infection leads to poor outcome in a mouse model of stroke Muhammad S. (1), Haasbach E. (2), Planz O. (2), Schwaninger M. (1) Stroke is the second most common cause of death and still the leading cause of neurological disability worldwide. Despite intensive research and discovery of potentially effective neuroprotective compounds in animal models of stroke, all the therapeutic possibilities have failed in clinical trials. In contrast to healthy experimental animals, most of the stroke patients are accompanied by risk factors and potential triggers of stroke such as atherosclerosis, obesity and acute infection that lead to a systemic inflammatory state. How the systemic inflammation influences the development of stroke, in which way it affects the outcome after stroke and which mechanisms are involved is unclear. Here, we have investigated the role of systemic inflammation (infection) in the outcome of stroke by subjecting mice to a non lethal infection of influenza A virus (APR 8/34) in a mouse model of cerebral ischemia. The male mice with an age of 9-12 weeks were infected intranasally with influenza A virus. On day 5 after infection the mice were subjected to middle cerebral artery occlusion (MCAO) for 48 hours. The infarct volume was quantified on cryosections of brain after Nissl staining. The lesion size was significantly increased in influenza-infected mice. We further

investigated IgG extravasation to evaluate the blood-brain barrier leakage. Indeed, the brains of influenza-infected mice showed intensive IgG staining after MCAO. The disrupted blood-brain barrier can lead to increased infiltration of peripheral immune cells. However, it is unclear weather the influenza infection alters the invasion of leukocytes into the ischemic brain. Our results demonstrate that the systemic changes due to influenza infection lead to poor outcome after stroke through increased blood brain barrier leakage. 1. Institute of Pharmacology, University of Heidelberg, Heidelberg, Germany, 2. Friedrich-Loeffler-Institut, Institute of Immunology, Tübingen, Germany

224 Discovery of novel transcriptional programs in cerebral ischemia by in silico promoter analysis Ridder D.A. (1), Bulashevska S. (2), Chaitanya G.V. (3), Babu P.P. (3), Schwaninger M. (1) In stroke, gene transcription plays a central role in delayed processes such as neuroinflammation and neuroregeneration. However, our knowledge of transcriptional regulation is incomplete. To predict new transcriptional regulatory mechanisms in cerebral ischemia, we applied a computational approach combining two kinds of information: the results of a microarray analysis in a mouse model of stroke and in silico detection of transcription factor (TF) binding sites in promoter regions of the genes on the array. By using a discriminative logistic regression model, we identified binding sites and hence TFs significantly associated with the up-regulation of genes. Out of 356 TF binding sites defined in TRANSFAC, we could link 33 to gene up-regulation in cerebral ischemia, binding such established players as CREB, NF-κB, and STAT3 as well as other factors that have not been implicated in cerebral ischemia. To evaluate the results further we investigated whether two TFs, C/EBPβ and vitamin D receptor (VDR), are activated as predicted. Immunohistochemistry demonstrated that C/EBPβ translocated to the nucleus in cerebral ischemia and chromatin immunoprecipitation revealed increased binding to the promoter of its target gene saa3, both results supporting the activation of C/EBPβ in cerebral ischemia. Similarly, immunohistochemistry of VDR confirmed that it was activated in cerebral ischemia. According to these data, our in silico analysis provides a valid picture of transcriptional regulation in stroke and suggests several novel transcriptional programs for further exploration. 1. Dept. of Pharmacology, University of Heidelberg, Germany, 2. Division 'Theoretical Bioinformatics', German Cancer Research Center (DKFZ), Heidelberg, Germany, 3. Centre for Biotechnology & Department of Animal Sciences, University of Hyderabad, India

225 The lithium-induced increase of the interaction between CREB and its coactivator TORC1 is not dependent on lysine 290 of CREB Heinrich A.H. (1), Böer U.B. (1), Oetjen E.O. (1), Knepel W.K. (1) Lithium (Li+), used for decades to treat bipolar disorder, was recently shown to enhance the interaction between the transcription factor CREB and its coactivator TORC1 (transducer of regulated CREB 1), thereby enhancing CREB-directed transcriptional activity. Li+ and magnesium (Mg2+) have similar ionic radii and share ligand-binding properties. Indeed, Mg2+ was shown to inhibit the interaction between CREB and TORC1. This effect was counteracted by Li+. The amino acid lysine at position 290 (K290) of CREB is known to be crucial for the coordination of a Mg2+ in the bifurcation of the basic leucine zipper (bZip) being essential for DNA-binding. In the present study we investigated the role of K290 of CREB for the Li+-induced enhancement of the interaction between CREB and TORC1. K290 was mutated to glutamate (E) by sitedirected mutagenesis. For luciferase-reporter gene assays in HIT cells CREB wild type (wt) and CREB-K290E were expressed as GAL4-fusion proteins. The luciferase gene under control of 5 repeats of the GAL4-binding site was cotransfected. Compared to GAL4-CREB-wt, CREB-K290E-dependent transcription was 5-fold higher, and the induction upon treatment with 1 mM cAMP was 3-fold stronger for the mutant. GAL4CREB-K290E showed also a 3-fold enhancement by Li+ of the cAMP-induced transcriptional activity, which was comparable to GAL4-CREB-wt. Using only the bZip as GAL4-fusion protein no inducible activity was detected. Overexpression of TORC1 led to a cAMP-stimulated transcriptional activity of GAL4-bZip-wt, which was further increased by Li+. The same effect was observed for GAL4-bZip-K290E. To investigate effects of Li+ directly on the interaction between CREB and TORC1, in vitro GST-Pull Down-Assays were performed with recombinant GST-CREB-wt and GST-CREB-K290E. TORC1 was radioactively labelled using [35S]methionine. The GST-Pull Down revealed a markedly reduced interaction between GST-CREB-K290E and [35S]TORC1 compared to GSTCREB-wt. Nevertheless, increasing concentrations of Li+ enhanced the amount of [35S]TORC1 bound to GST-CREB-K290E by 1.7-fold, comparable to GST-CREB-wt. In addition, similar results were obtained when K290 was mutated to alanine. The present study showed that CREB-K290E as well as K290A behave like CREB-wt with respect to effects of Li+. These data suggest that the amino acid K290 of CREB is not involved in conferring the effect of Li+ on cAMP-induced CREB-directed transcriptional activity. 1. Dept. of Pharmacology, Georg-August-University, Göttingen, Germany

226 Differences in pilocarpine-induced epileptogenesis and behavioral alterations of Sprague-Dawley rats from two different breeders Bankstahl M. (1), Brandt C. (1), Löscher W. (1, 2) The pilocarpine model of epilepsy is a widely used animal model of partial epilepsy with spontaneous recurrent seizures (SRS) and gains increasing importance in the development of new antiepileptic drugs. Performing this model in our lab for many years, we recently observed changes in pilocarpine sensitivity of Sprague-Dawley (SD) rats. Aim of the present study was to evaluate if there are differences in SD substrains of two main laboratory animal breeders. We compared naïve female rats and pilocarpine treated rats of Charles River and Harlan Laboratories, respectively. Fractionated administration of pilocarpine was used to induce a status epilepticus (SE). Several weeks following SE the animals were continuously video- and EEG- monitored to detect

49

substrain differences in occurrence of SRS. For assessment and comparison of epilepsy-associated behavioral alterations, we performed a battery of behavioral tests including the open field test, the elevated plus maze test, and hyperexcitability tests. To evaluate whether substrain differences in epileptogenesis are combined with varieties in epilepsy-induced loss of neurons, neurodegeneration was evaluated in brain sections using a semiquantitative scoring system. The two substrains differed in their pilocarpine sensitivity and progression of epileptogenesis. SD rats from Charles River showed lower SE incidence. In contrast, the percentage of rats with SRS was significantly higher in SD rats from Charles River, when recorded 7-9 weeks after SE, indicating slower epileptogenesis in SD rats from Harlan. Additionally, epileptic Charles River SD rats demonstrated more distinct behavioral alterations. Severity of neurodegeneration is currently under investigation. The study shows that differences in rat substrains should be taken into account when interpreting results of post-SE models of chronic epilepsy. 1. Dept. of Pharmacology, Toxicology, and Pharmacy, University of Veterinary Medicine, Hannover, Germany, 2. Center for Systems Neuroscience, Hannover, Germany

marker NeuN in rat brain tissue revealed a predominant neuronal expression in the dentate gyrus, the cornu ammonis pyramidal cell layer, layer III of the piriform cortex as well as throughout all layers of the parahippocampal cortex. In the hilus of the hippocampus single neurons expressed YB-1. The neuronal expression pattern was comparable in the hippocampus and parahippocampal cortex of adult macaques and humans. Double-labeling studies with the endothelial cell marker Glut-1, the multidrugtransporter P-glycoprotein, and the astrocytic marker GFAP did not indicate a colocalization. No differences in YB-1 expression were evident between control rats and rats at different time-points post status epilepticus. The data demonstrate that YB-1 is predominantly expressed in neurons in the adult brain of rats, macaques and humans. Lack of a co-localization of Glut-1 and P-glycoprotein in naïve rats and rats following status epilepticus argues against a direct role of YB-1 in the regulation of blood-brain barrier P-glycoprotein. 1. Inst. of Pharmacology, Toxicology, and Pharmacy, Ludwig-Maximilians-University Munich, Munich, Germany, 2. Division of Cell Biology, Inst. of Ophthalmology, University College London, London, UK, 3. Inst. of Anatomy, Ludwig-Maximilians-University Munich, Munich, Germany, 4. Dept. of Pathology, University of Veterinary Medicine Hannover, Hannover, Germany

227 Prophylactic treatment with the COX-2 inhibitor parecoxib after status epilepticus: Effect on epileptogenesis, behavioral alterations, and neuronal damage in rats Polascheck N. (1, 2), Bankstahl M. (1), Brandt C. (1), Löscher W. (1, 2) Epileptogenesis, i.e. the process leading to epilepsy with spontaneous recurrent seizures (SRS), can be initiated by a number of brain damaging insults, including traumatic brain injury, status epilepticus (SE), and stroke. Such acquired epilepsy is often associated with memory impairment and behavioral problems. The mechanisms that transform a normal brain to an epileptic one are not fully understood but seem to involve inflammatory processes, such as expression of pro-inflammatory mediators (COX-2, cytokines such as IL-1ß, IL-6 or TNFα and others), which affects the integrity of the blood-brain barrier, enhances neuronal excitability, decreases seizure threshold and may exacerbate brain injury. Recently, Jung et al. (Neurobiol. Dis. 23, 237-246, 2006) reported that the COX-2 inhibitor celecoxib, administered after a pilocarpine-induced SE in rats, prevented neuronal damage and microglia activation in the hippocampus and decreased the likelihood of developing SRS. If this finding could be reproduced by other groups, it would open new avenues for prophylactic treatment after brain insults such as SE. This prompted us to perform experiments in which we administered the highly selective COX-2 inhibitor parecoxib after a pilocarpine-induced SE in rats. Parecoxib was administered twice daily at a dose of 10 mg/kg i.p. over 18 days. Control rats received vehicle instead of parecoxib after SE. Parecoxib did not alter the incidence of rats with SRS nor the behavioral alterations developing after SE. 1. Dept. of Pharmacology, Toxicology, and Pharmacy, University of Veterinary Medicine, Hannover, Germany, 2. Center for Systems Neuroscience, Hannover, Germany

228 Pharmacology of eslicarbazepine acetate in the amygdala kindling model of temporal lobe epilepsy Soerensen J. (1), Pekcec A. (1), Soares-da-Silva P. (2), Potschka H. (1) Purpose: Eslicarbazepine acetate (ESL, BIA 2-093) is a novel voltage-gated sodium channel blocker that is efficacious as adjunctive therapy in adults with partial-onset seizures with or without secondary generalisation when given once daily [1]. ESL was efficacious in several models predictive of anticonvulsant efficacy [2] and has the potential to retard kindling-induced epileptogenesis [3]. The present study determines the effect of ESL in the mouse amygdala kindling model of temporal lobe epilepsy. Methods: Male NMRI mice were subsequently stimulated once daily via an implanted depth electrode until 10 generalized seizures were elicited. The following dosages were tested in experiments in these fully kindled animals: ESL 100, 200, 300 mg/kg. The drug was administered intraperitoneally 15 min before stimulation. Each drug experiment was preceded by a vehicle control experiment in the same group of animals. Results: ESL dose-dependently increased the focal seizure threshold (ADT), effects being significant (P<0.05) at 200 and 300 mg/kg. In 1 out of 10 animals receiving 200 mg/kg and in 7 out of 13 animals receiving 300 mg/kg no seizure activity was observed until the maximum stimulation current of 1200 µA. In response to 200 and 300 mg/kg ESL, threshold increases reached >289% and >1319%, respectively. Seizure duration and afterdischarge duration recorded at ADT were not significantly altered by eslicarbazepine acetate. Seizure severity was reduced in a dose-dependent manner by ESL with a significant difference to the vehicle control experiment at 200 and 300 mg/kg. Conclusion: The data indicates that ESL inhibits seizure initiation and can protect against focal seizure activity. The effect of ESL on seizure severity gives evidence that the compound interferes with seizure progression by inhibiting propagation of activity from the focus. [1] Elger C, Bialer M, Cramer JA, et al. (2007). Epilepsia 48, 497-504. [2] Almeida L, Soares-da-Silva P. (2007) Neurotherapeutics 4, 88-96. [3] Potschka H, Pekcec A, Soares-da-Silva P. (2007) Proceedings of the 27th International Epilepsy Congress, Singapore, Abstract Book. 1. Inst. of Pharmacology, Toxicology, and Pharmacy, Ludwig-Maximilians-University, Munich, Germany, 2. BIAL-Portela & Ca., S.A., S. Mamede do Coronado, Portugal

229 Cellular localization of Y-box binding protein 1 in brain tissue of rats, macaques, and humans Pekcec A. (1), Unkrüer B. (1), Fuest C. (1), Wehmeyer A. (1), Balda M.S. (2), Horn A. (3), Baumgärtner W. (4), Potschka H. (1) The Y-box binding protein 1 (YB-1) is considered to be one of the key regulators of transcription and translation, e.g. in the regulation of the expression of the blood-brain barrier efflux transporter, P-glycoprotein. However, so far only limited knowledge exits regarding its cellular distribution in the adult brain. Brain tissue of naïve female Wistar rats, macaques and human tissue was used to study the distribution of YB-1 in the adult brain by immunohistochemistry. To analyze temporal differences in the expression of YB-1 with regard to seizure-induced up-regulation of P-glycoprotein, a status epilepticus was induced in female Wistar rats by fractionated pilocarpine administrations (10 mg/kg i.p.). Analysis of YB-1 immunolabeling as well as double-labeling with the neuronal

230 Cell lineage mapping of doublecortin positive cells in the hypothalamus Werner L. (1), Rossner M.J. (2), Schwaninger M. (1) There are two known regions in the adult mammalian brain, where continuous production of neurons take place: the subventricular zone (SVZ) of the lateral ventricles and the dentate gyrus (DG) of the hippocampus. These neurogenic regions contain neural stem and progenitor cells which are positive for glial fibrillary acidic protein (GFAP), nestin, polysialylated neural cell adhesion molecule (PSA-NCAM) and doublecortin (DCX) e.g, depending on their developmental stage. In order to identify the fate of neuronal progenitors we used a BAC-transgenic mouseline carrying the tamoxifen inducible Cre recombinase (CreERT2) under the control of the DCX promoter. DCX is a commonly used marker for neuronal progenitors, which is specifically expressed in neuroblasts. After treatment with tamoxifen the CreERT2 recombinase activates a reportergene. In brain sections of eight weeks old mice double positive for DCXCreERT2 and the lacZ reportergene we found lacZ staining, one week after the last tamoxifen injection in the cerebellum, the cortex, the region of the fourth ventricle, near the aqueduct, in the bulbus olfactorius, as well as in the dentate gyrus. Surprisingly we saw a relatively strong staining in defined nuclei of the hypothalamus. To validate the LacZ staining we did double immunostaining against Cre and DCX. Almost all Cre positive cells also expressed DCX. ß-Gal and DCX double immunostaining of sections of mice, treated with Tamoxifen one week before analysis show that also after one week some ß-Gal positive cells in the hypothalamus are still DCX positive. Double staining of sections from the same mice with NeuN reveals that a few ß-Gal positive cells are mature neurons. This results indicates that the generated mouse line is a useful tool to study the cell lineage of DCX positive cells in the hypothalamus. 1. Dept. of Pharmacology, Ruprecht Karls University, Heidelberg, Germany, 2. MaxPlanck-Institute of Experimental Medicine, Göttingen, Germany

231 Cellular mechanisms of interleukin-17 induced blood-brain-barrier disruption Kuhlmann C.R.W. (1), Closhen D. (1), Huppert J. (1,2), Croxford A. (2), Bechmann I. (3), Waisman A. (2), Luhmann H.J. (1) The proinflammatory cytokine interleukin-17A (IL-17A) was shown to play a role in the progression of multiple sclerosis and experimental autoimmune encephalomyelitis (EAE). Recently T-helper 17 (Th17) cells were demonstrated to disrupt the blood-brainbarrier (BBB) by the mutual action of IL-17A and IL-22. In the present study, we show that IL-17A induces NADPH-/xanthin-oxidase-dependent reactive oxygen species (ROS) production. The resulting oxidative stress activated the endothelial contractile machinery, which was accompanied by downregulation of the tight junction molecule occludin. Blocking either ROS formation or myosin light chain phosphorylation prevented IL-17A induced BBB disruption. Treatment of mice afflicted with EAE using ML-7, an inhibitor of the myosin light chain kinase, resulted in less infiltration of lymphocytes via the BBB, and subsequently reduced clinical characteristics of EAE. These observations indicate that IL-17A accounts for a crucial step in the development of EAE by impairing the integrity of the BBB involving the augmented production of ROS. 1. Inst. of Physiology and Pathophysiology, Johannes Gutenberg University, Mainz, Germany, 2. First Medical Department, Johannes Gutenberg University, Mainz, Germany, 3. Institute for Clinical Neuroanatomy, Johann Wolfgang Goethe-University, Frankfurt, Germany

232 Functional modulation of P-glycoprotein at the rat blood-brain barrier by elacridar and tariquidar studied with (R)-[11C]-verapamil and small animal positron emission tomography Bankstahl J.P. (1), Kuntner C. (2), Bankstahl M. (1), Langer O. (2), Löscher W. (1) Multidrug efflux transporters like P-glycoprotein (Pgp) at the blood-brain barrier (BBB) are believed to play an important role in resistance to central nervous system drug treatment. In epilepsy patients overexpressing Pgp at the BBB, modulation of Pgp by third generation inhibitors like tariquidar (XR9576) or elacridar (GF120918) may be a promising strategy to overcome drug resistance by enhancing the brain penetration of Pgp substrates. Aim of the present study was to quantify functional Pgp expression in the brain using small animal positron emission tomography (PET) as an in-vivo imaging technique and to apply this technique to obtain dose response curves for the above mentioned highly selective Pgp-inhibitors. To study modulation of cerebral Pgp activity in naïve female Sprague-Dawley rats, the Pgp substrate (R)-[11C]-verapamil was utilized. The rats underwent dynamic 60-min (R)-[11C]-verapamil PET scans at 2 hours after intravenous administration of tariquidar (1.5 to 30 mg/kg, n = 12) or elacridar (0.3 to 7.5 mg/kg, n = 12) using femoral vein catheterization. Radioactivity uptake in brain was expressed as the area under the time activity curve (AUC) normalized for injected 11C-

50

activity. A sigmoidal dose-response curve was fitted to the AUC values measured after different inhibitor doses, which provided estimated half-maximum effect doses (ED50) of 3.00 ± 0,24 mg/kg for tariquidar and 1.16 ± 0.12 mg/kg for elacridar, respectively. Our data demonstrate that both tariquidar and elacridar are potent inhibitors of Pgp at the BBB and enhance brain uptake of Pgp substrates. The current data will help to study changes in Pgp-function at the BBB in disease models and pharmacoresistant patients. (Supported by the EU [EURIPIDES]). 1. Department of Pharmacology, Toxicology, and Pharmacy, University of Veterinary Medicine, Hannover, Germany, 2. Department of Radiopharmaceuticals, Austrian Research Centers GmbH - ARC, Seibersdorf, Austria

233 Profiling of biglycan molecular promoter haplotypes Schmitz B. (1), Telgmann R. (1), Brand E. (2), Rüßmann C. (1), Fischer J. (3), Paul M. (4), Brand-Herrmann S.M. (1) Introduction: The small leucine rich proteoglycan biglycan (BGN) is involved in cardiovascular disease (CVD) pathophysiology. The aim of the current study was to identify and functionally characterize the differential usage of BGN gene promoter constellations in different cell lines. Material and Methods: We directly resequenced an 1199 bp promoter portion of the BGN gene in 57 patients with CVD to characterize its variant structure. Molecular haplotypes (MolHaps) were determined by subcloning of individuals’ DNA. MolHaps and promoter deletion constructs were subcloned into reporter gene vector pGL3-basic and transfected into HEK293T, EA.hy926 and THP-1 monocytes. Cells were kept under basal conditions or stimulated with 10-8M PMA or 0.5 mM 8-Br-cAMP for 24 hrs. Transcriptional initiation sites (TIS) were determined by semiquantitative PCR. Results: We identified three SNPs in the BGN 5’-regulatory region (G-848A, G-578A, G-151A), and one in the 5’-UTR (G+94T) leading to the determination of three respective MolHaps: [G-848-G-578-G-151-G+94; wildtype (wt)], [G-848G-578-A-151-T+94] and [G-848-A-578-G-151-G+94]. Under basal as well as stimulatory conditions MolHap 2 was significantly less active (all p values ≤0.05) than wt in HEK293T and EA.hy926. Whereas all cell lines displayed transcripts from the main TIS (position +1), EA.hy926 cells showed strong usage of an alternative TIS positioned ~45 bp upstream of the main TIS. Transfection assays revealed a short promoter deletion fragment spanning from positions -39 to +162 to exert high transcriptional activity, being 20-fold over the longest fragment tested, in EA.hy926 cells (p ≤0.001). Conclusion: We were able to show that (1) BGN MolHaps exert cell type-specific differentially functional properties, (2) promoter usage obviously differs across cell lines, and (3) the 5’-UTR represents an active part of the BGN gene promoter. 1. Leibniz-Institute for Arteriosclerosis Research, University of Münster, Germany, 2. Internal Medicine D, University Clinic of Münster, Germany, 3. Institute of Pharmacology and Clinical Pharmacology, University of Düsseldorf, Germany, 4. Faculty of Health, Medicine, and Life Science, Maastricht University, Maastricht, Netherlands

234 Decreased expression of microRNA miR-133 but not miR-1 is associated with impaired cardiac performance Danowski N. (1), Frey U.H. (2), Kottenberg-Assenmacher E. (2), Jakob H.G. (3), Manthey I. (1), Siffert W. (1), Peters J. (2) MicroRNAs (miRs) post-transcriptionally regulate gene expression. miR-1 and miR-133 can evoke opposing effects on apoptosis in cardiomyocytes with miR-1 acting proapoptotically and miR-133 anti-apoptotically. In 86 patients undergoing coronary artery bypass grafting (CABG), we tested the hypothesis that cardiac miR-1 and miR-133 expression are associated with altered cardiac performance. Cardiac index and vascular pressures were determined during general anesthesia and right atrial appendages were obtained during cannulation for extracorporal circulation. miR expression was quantified by RNase protection assay and by Real-Time PCR. Decreased miR-133 expression was significantly associated with diminished exercise performance as indicated by worsened NYHA functional class (p=0.0045), and higher pulmonary artery occlusion pressures (p=0.0042). Furthermore, patients with NT-proBNP concentrations >1800 pg/ml showed a 60% decrease in miR-133 expression compared to patients with concentrations <300 pg/ml (p=0.036). In contrast, no association could be detected between miR-1 expression and variables of cardiac performance. This study shows for the first time a link between decreased miR- 133 expression and development of cardiac failure in CABG patients, possibly indicating a role of miR-133 in regulating cell fate by inhibiting apoptosis. 1. Institute of Pharmacogenetics, University of Duisburg-Essen, Essen, Germany, 2. Department of Anaesthesiology and Intensive Care Medicine, University of DuisburgEssen, Essen, Germany, 3. Clinic for Cardio-Thoracic Surgery, University of DuisburgEssen, Essen,Germany

235 Identification and functional analyses of molecular haplotypes of the human osteoprotegerin gene promoter Hagedorn C. (1), Telgmann R. (1), Dördelmann C. (1), Schmitz B. (1), Hasenkamp S. (2), Cambien F. (3), Paul M. (4), Brand E. (2), Brand-Herrmann S.M. (1) Osteoprotegerin (OPG) has been reported to be involved in the development of atherosclerotic disease phenotypes. The structural and functional architecture of the OPG promoter locus and its transcriptional regulation are poorly characterized. We identified 1008 bp of the OPG 5’ flanking region to be sufficiently active in osteosarcoma cells, and analyzed this portion for cell type specific transcription start sites (TSS). To identify cis regulatory regions, serial deletion constructs were generated. We identified one novel and corrected the position of another TSS in osteosarcoma cell lines SaOs 2 and U2Os. Direct sequencing of 1008 bp of the OPG promoter portion in 57 patients with cardiovascular disease (CVD) led to the identification of five variants (T-960C [rs3134071], A-946G [rs3102735], G-900A [rs3134070], T-864G [rs3134069], and T-159C [rs2073617]). Individual subcloning procedures revealed the existence of three common molecular haplotypes (MolHaps): [T-960-A-946-G-900-T-864; wild type (wt)], [T-960-G-946-G-900-T-864; MolHap2], [C-960-G-946-A-900-

G-864; MolHap4], which were analyzed by transient transfection in SaOs-2 and U2Os cells. The OPG full length construct displayed a moderate transcriptional activity, whereas activities of MolHaps 2 and 4 were significantly reduced (p = 0.0018). Introduction of the -159C allele reduced transcriptional activities of the respective full length constructs (p = 0.0014), whereas it significantly increased activities of the deletion constructs in both cell lines (p = 0.0005). Bandshift assays revealed specific DNA:protein interactions for the MolHaps with Sp1 and NF-1, co-transfections identified NF-1/A1.1 and NF-1/B2 interacting with the MolHaps. Competition assays and ChIP experiments identified Egr1 interacting exclusively with the -159T allele. In conclusion, we propose new structural features and the existence of genetically differential functional portions within the OPG promoter region. 1. Leibniz-Institute for Arteriosclerosis Research, Münster, 2. University Hospital Münster, Internal Medicine D, 3. INSERM, UMR S 525, Université Pierre et Marie Curie, Paris, France, 4. Maastricht University, Faculty of Health, Medicine, and Life Science, Maastricht, The Netherlands

236 Identification of a Mef2C gain-of-function mutation in the Isl1 transcription factor in patients with familial cardiomyopathies Friedrich F. (1), Dilanian G. (2), Juhr D. (1), Charron P. (2), Isnard R. (3), Chien K.R. (4), Eschenhagen T. (1), Villard E. (2), Carrier L. (1) Purpose. The transcription factor Islet-1 (Isl1) is a marker of cardiovascular progenitors, which contribute to major cardiac structures, and is essential for mammalian cardiogenesis. Disturbances in Isl1 function could lead to alterations in formation, expansion and differentiation of Isl1+ progenitor cells and result in congenital heart disease or cardiomyopathy in adult life. The aim of this project was to screen for ISL1 mutations in patients with familial hypertrophic (HCM), dilated (DCM) or arrhythmogenic right ventricular cardiomyopathy (ARVC). Methods. The 6 exon and intron boundaries of ISL1 were screened for mutations in a large cohort of index cases (295 HCM, 79 DCM, and 26 ARVC) by direct sequencing of PCR products amplified on genomic DNA. Consequences of the identified ISL1 mutation were indirectly determined by Luciferase reporter assay measuring the activity of the Myocyte enhancer factor 2C (Mef2C) promoter in HEK 293 and CHO cells co-transfected with WT or mutant Isl1. The Mef2C promoter is an established Isl1 target containing 2 consensus Isl1-binding sites. Results. Four known SNPs with no clinical association reported were detected and considered as neutral: rs3762977, rs36216897, rs3917084, and rs2303751. Furthermore, 5 unknown variants were found: 2 in the 5’-UTR (-482G>C, FHCM=3/592 and -240G>A, FHCM=4/592) and 3 in exon 4 (EX4+35G>A, FHCM=2/590; EX4+89C>T, FHCM=3/590; EX4+277A>G, FHCM=1/590 and FDCM=2/158). The last one results in an Asn252Ser exchange being conserved through evolution suggesting selection pressure. This mutation was found at the heterozygous state in 1 HCM patient, who also carries a MYH7 mutation, and at the homozygous state in 1 DCM patient. The mutation was not detected in 350 control individuals. We investigated the effect of WT and mutant Isl1 on Mef2C promoter activity. WT and mutant Isl1 increased Mef2C promoter activity by 287% and 488% in HEK cells and by 307% and 463% in CHO cells, respectively. The effect of mutant Isl1 was significantly greater than the WT (5.9±0.4 vs. 3.9±0.3 in HEK cells, and 5.6±0.2 vs. 4.1±0.6 in CHO cells, n=16/8, p<0.001/<0.05, Student t-test). Conclusions. This study describes a new gain-of-function mutation in the human ISL1 gene. The mutation was found at the heterozygous state in 1 HCM patient, and at the homozygous state in 1 DCM patient. This Asn252Ser ISL1 mutation could potentially lead to greater activation of downstream targets involved in cardiac development, remodelling and hypertrophy. 1. Institute of Pharmacology, University Medical Center Hamburg-Eppendorf, Hamburg, Germany, 2. Inserm UMRS 621, Paris, France, 3. Institut de Cardiologie, GH PitiéSalpêtrière, Paris, France, 4. Richard B. Simches Research Center, Boston, MA, USA

237 AAV-mediated shRNA-delivery to knockdown components of the beta-adrenergic signaling pathway in reconstituted heart tissue in vitro and in the murine heart in vivo Neuber C. (1), Müller O.J. (2), Eder A. (1), Hansen A. (1), Hansen F. (1), Katus H. (2), Eschenhagen T. (1), El-Armouche A. (1) Post-transcriptional gene silencing with RNAi is a promising therapeutic strategy in heart disease. Recombinant adeno-associated virus (rAAV)-mediated gene delivery has emerged as a realistic cardiac gene-transfer approach. The long term aim of this project is to study the therapeutic impact of rAAV-mediated RNAi against stimulatory components of the β-adrenergic signaling pathway (e.g. the proximal β1-adrenoceptor [AR] and the distal amplifier molecule inhibitor-1 [I-1]) in vitro and in vivo. The strategy is as follows: (i) 4 different H1-promoter-driven shRNAs against the β1-AR or I-1 were delivered into HEK293 cells and showed 50-90% knockdown efficiency at the mRNA and protein level. (ii) The 2 most efficient H1-I-1/β1-AR-shRNA cassettes and nonsilencing controls (shCTR) were subcloned into a rAAV-6 plasmid bicistronically encoding EGFP. Transfected cells showed >80% EGFP positivity with ~70% reduction in I-1/β1-AR mRNA and ~50% reduction in I-1 protein/β1-AR density (after 72 h) compared to shCTR. (iii) From these constructs rAAV were produced with titers of 0.5 x 1012 vg/ml. (iv) Functional consequences of rAAV-mediated I-1/β1-AR RNAi are currently addressed in vitro in heart tissues reconstituted from neonatal rat myocytes (engineered heart tissue, EHT) as a highly sensitive multi-cellular test system. 12 days after infection with 104 vg/cell transduction efficiency reached ~70%. Preliminary results from a small series of EHTs either non-infected (CTR) or infected with rAAV-6-shb1-AR/shCTR indicate substantial non-specific effects on force generation (TT: 0.95±0.1 mN, n=4 in CTR vs. 0.53±0.1 mN, n=2 in shCTR, p=0.07). Strikingly, rAAV-6-shβ1 infected EHTs did not develop any force (n=4) indicating that an early loss of β1-AR may impair differentiation and/or development in these heart tissues. (v) For the in vivo experiments AAV-6, which is superior in vitro, will be replaced by serotype 9 because of its proven selectivity for the myocardium after intravenous injection. In a pilot experiment, AAV-9 encoding EGFP was delivered into mice via tail vein injection (3x1011 vg/mouse). Two weeks after injection, transduction efficiency amounted to ~30% of cardiomyocytes in the left ventricle with homogeneous transmural distribution. Other tissues including skeletal muscle and liver were negative. In long-term, this technology should be useful

51

to evaluate for therapeutic potential of knocking down different components of the β-AR signaling pathway in the context of heart failure. 1. Dept. of Pharmacology, University Medical Center Hamburg-Eppendorf, Germany, 2. Dept. of Internal Medicine III, University of Heidelberg, Germany

238 CaVβ2 diversity and developmental expression in the murine heart Link S. (1), Meissner M. (1), Held B. (2), Freichel M. (1), Flockerzi V. (1) The L-type Ca channels in heart are predominantly formed by α1C (CaV1.2) and CaVβ2 (β2) and targeted deletion of either gene is lethal at embryonic day 11 (β2) (1) and 14.5 (CaV1.2) (2). Here, we cloned and characterized full length CaV1.2 cDNAs and the cDNAs of the β2 splice variants, which were not available from mouse heart. A β2 antibody detected at least two β2-variants of 68 and 74 kDa in the murine heart, the 74 kDa protein being abundant during embryonic and perinatal stages (E11.5 to postnatal day (P)27) and the 68 kDa protein during adulthood. To identify the underlying β2variants, cDNA libraries were constructed from poly(A)+RNA isolated from hearts of P7 and adult mice and screened for β2 cDNAs. Sixty independent β2 cDNA clones were identified, isolated and sequenced. They code for four β2 variants, which differed in their 5’ sequences covered by the alternatively spliced exons 1A and 2A (β2aN1), exon 2B (β2aN3), exon 2C (β2aN4) and exon 2D (β2aN5). From the number of clones obtained we estimated the frequencies of the four variants in heart in vivo. Reconstitution of the β2 protein patterns observed in embryonic and adult hearts required recombinant β2N1, -N4 and -N5 proteins but not an additional, yet unknown CaVβ2 splice variant. cDNA fragments coding for β2aN1, -N4 and -N5 but not for β2aN3 were amplified from isolated embryonic and adult cardiomyocytes, demonstrating that the palmitoylated β2aN3 is expressed in vascular or neuronal cells rather than in myocytes. For functional expression we cloned the full length cDNA of the CaV1.2 pore: 19 clones obtained by RT-PCR were sequenced. Expression of the CaV1.2 alone did not yield currents distinguishable to currents recorded from non-transfected HEK-cells. Coexpression of CaV1.2a and β2 yielded significant L-type currents which showed CaVβ2-variant specific regulation and modulation; the 74 kDa β2aN1 shifted the activation of Ca currents significantly to more positive potentials compared to the other β2 variants. The data reveal a diversity of L-type Ca channel subunits in mouse heart and show that single β2 splice variants may adapt properties of the L-type Ca channels to changing myocardial requirements during development with the β2aN1-induced changes of current activation being essential for embryonic heart function.1 Weißgerber et al. (2006) Circ Res. 99:749-757.2 Seisenberger et al. (2000) J Biol Chem. 275:39193– 39199. 1. Institute of experimental and clinical pharmacology, Universität des Saarlandes, Homburg, Germany, 2. Frankfurt International Research Graduate School for Translational Biomedicine, Frankfurt, Germany

239 Reversibility of cardiac fibrosis and hypertrophy after chronic pressure overload Beetz N.B. (1), Rylski B.R. (2), Beyersdorf F.B. (2), Hein L.H. (1) For a long time heart failure has been considered to be irreversible. Temporary therapy of patients with a left ventricular assist device (LVAD) has allowed researchers to gain insight into a process of cardiac "reverse remodeling" and functional improvement. However, the mechanisms of cardiac recovery after LVAD implantation are largely unknown. Thus, the aims of this study were (1) to establish a mouse model with reversible transverse aortic constriction (rTAC) to mimic the effects of cardiac unloading after LVAD implantation in heart failure patients and (2) to investigate the mechanisms that govern the recovery phase following removal of the aortic stenosis. Minimally invasive TAC was performed in C57BL6/J mice at the age of eight weeks using a novel tourniquet technique. Sham operated mice without aortic stenosis served as controls (sham). After three weeks of pressure overload, the aortic band was removed in one group of mice with a minimally invasive technique avoiding a second thoracotomy and mice were followed for another 8 weeks (rTAC). The second group was sham operated without removal of the tourniquet (TAC). Banding of the transverse aortic arch for 3 weeks resulted in an increase in flow velocity across the stenosis which was significantly reduced in the rTAC group. At the end of the observation period, systolic aortic pressure was elevated by > 25 % in TAC mice as compared to sham or rTAC mice. Similarly, TAC was accompanied by tachycardia which was not present in rTAC or sham operated animals. TAC caused significant cardiac hypertrophy, indicated by increased ventricle weight to body weight ratio, which was almost normalized in rTAC mice (TAC 159 %, rTAC 114 % of sham). Left ventricular interstitial fibrosis as induced by TAC was only partially alleviated after tourniquet removal (fibrosis score: TAC 3.6 fold, rTAC 2.6 fold of sham). Ventricular atrial natriuretic peptide mRNA expression was strongly induced by TAC but nearly normal in rTAC animals. TAC and rTAC were associated with distinct gene expression profiles as determined by microarray and qPCR analysis. In conclusion, reversible tourniquet banding of the aortic arch in mice led to cardiac hypertrophy, fibrosis and induction of the fetal cardiac gene program which was partially normalized after removal of the stenosis. This model may provide a basis to identify the molecular mechanisms of reverse remodeling and functional improvement in cardiac failure and hypertrophy. 1. Dept. of Clinical and Experimental Pharmacology and Toxicology, Albert-LudwigsUniversity, Freiburg, Germany, 2. Dept. of Cardiovascular Surgery, University Medical Center, Albert-Ludwigs-University, Freiburg, Germany

240 Connexin-isoform-specific effect of the gap junction modulator antiarrhythmic peptide AAP10 on electrical and metabolic coupling Dhein S. (1), Mohr F.W. (1), Hagen A. (2) Aims: Intercellular coupling is maintained by gap junction channels, which are built as dodecameric assembly of connexins. In the heart, 3 main isoforms of connexins are found: Cx43 (ubiquitous), Cx40 (conduction system; atrium), Cx45 (early developmental stages, conduction system). Gap junction channels allow the transmission of current (electrical coupling) and exchange of small molecules (<1000 Da; metabolic coupling).

Antiarrhythmic peptides such as AAP10 or its derivative rotigaptide enhance cardiac gap junction coupling and act antiarrhythmically. However, it is unclear, (a) whether they act on all cardiac connexin-isoforms, and (b) whether they affect both electrical and metabolic coupling. Methods: We determined the effect of 50nM AAP10 (H2N-Gly-AlaGly-4Hyp-Pro-Tyr-CONH2) on macroscopic gap junction conductance (gj) by dual whole cell voltage clamp in human HeLa cells (which do not express connexins endogenously), transfected with either Cx40, Cx43 or Cx45. Moreover, we assessed metabolic coupling in these cells by either dye injection (Lucifer Yellow, LY; for Cx40 and Cx43) or scrape load technique (for Cx45). Results: Under control conditions there was a slight increase in gj with time (0.134±0.030 nS/min) in HeLa Cx43, which was significantly enhanced by 50nmol/l AAP10 (0.267±0.047 nS/min, p<0.05). Similar results were obtained from HeLa Cx45 (control: 0.065±0.026 nS/min; AAP10: 0.245±0,032 nS/min, p<0.05). In contrast, AAP10 had no effect in HeLa Cx40. Regarding metabolic coupling we found significant enhancement of the number of dye coupled cells by AAP10 in HeLa Cx43 cells (control: 22±1.83 coupled cells vs. AAP10 (50 nM, 30 min): 35.87±4.85 coupled cells (p<0.05). AAP10 did not affect dye coupling in HeLa Cx40 (control: 6.625±0.586 coupled cells vs. AAP10: 6.957±0.771 coupled cells). In scrape load experiments performed on HeLa-Cx45-cells the number of cells labeled with dye by diffusion was significantly enhanced by 50 nM AAP10 (controls: 20.02±0.98% vs. AAP10: 29.40±2.39%, p<0.05). Conclusion: Antiarrhythmic peptide AAP10 improves both electrical and metabolic gap junction intercellular coupling in cells coupled by Cx43 and Cx45, but not in Cx40-coupled cells. 1. Clinic for Cardiac Surgery, University of Leipzig, Leipzig, Germany, 2. University Hospital for Children, University of Leipzig, Leipzig, Germany

241 The gap junction modulator antiarrhythmic peptide AAP10 affects human cardiac gap junction coupling, and acts preferentially on uncoupled cells Dhein S. (1), Hagen A. (2), Jozwiak J. (1), Dietze A. (1), Mohr F.W. (1) Aims: Ischemia-induced ventricular fibrillation is one of the most important causes of death in industrial countries. In recent years modification of gap junctional coupling has been established as a new antiarrhythmic principle acting in rabbits, guinea pigs and rats. We now wanted to examine (a) whether the gap junction modulator AAP10 also acts on human cardiac gap junctions, (b) whether the effect might be enhanced in cells uncoupled by acidosis, and (c) whether it acts specifically in ischemic tissue. Methods: We determined influence of 50nM AAP10 (H2N-Gly-Ala-Gly-4Hyp-Pro-Tyr-CONH2) on macroscopic gap junction conductance by dual whole cell voltage clamp in human and rat cardiomyocytes. Cells were partially uncoupled by CO2-mediated acidosis (pH 6.3) or kept at “normal” conditions (pH 7.4, T 36 °C) with or without pre-treatment with AAP10. We furthermore investigated AAP10 effects in isolated rabbit hearts undergoing LAD occlusion for 30 min. Results: In local ischemia, AAP10 prevented from dephosphorylation of the gap junction protein Cx43 and resulted in preserved conduction velocity (control: 0.7±0.2; ischemia: 0.38±0.2; ischemia+AAP10: 0.6±0.2 m/s, p<0.05), and decreased dispersion. The AAP10 effect was confined to the ischemic area, while there was only very minor effect in the non-ischemic area. In isolated human and rat cardiomyocytes pairs AAP10 (50 nM) enhanced gap junctional intercellular coupling under normal conditions (+0.34±0.08 nS/min (n=6, human; +0.36±0.10 nS/min (n=6, rat)), and completely prevented from CO2-acidosis-induced uncoupling (acidosis: 0.76±0.30 nS/min (n=6, human), -1.36±0.59 nS/min (n=6, rat) vs. acidosis+AAP10: +0.96±0.17 nS/min (n=5, human) (p<0.05) and +0.24±0.06 nS/min (n=6, rat) (p<0.05). The coupling effect of AAP10 was significantly enhanced in previously uncoupled cells. There was no effect on I(Na). Conclusion: We conclude that the antiarrhythmic peptide AAP10 improves gap junctional intercellular coupling and prevents uncoupling by acidification in human cardiomyocytes. AAP10 is a new tool for antiarrhythmic therapy of ischemia-related ventricular arrhythmias. 1. Clinic for Cardiac Surgery, University of Leipzig, Leipzig, Germany, 2. University Hospital for Children, University of Leipzig, Leipzig, Germany

242 Identification of microRNAs involved in cardiomyocyte growth control Jentzsch C. (1), Leierseder S. (1), Engelhardt S. (2) Pathological growth of cardiomyocytes is a key event during the development and progression of heart failure. There is increasing evidence that recently discovered small RNA molecules (microRNAs) may play an important role in this process. For the identification of functionally relevant miRNAs we developed a high-throughput screening assay in a 96-well format. This assay enabled us to screen a library of about 500 miRNAs directly on the functional level. To this end primary neonatal rat cardiomyocytes were transfected with specific synthetic precursor miRNAs and stained by a fluorophoreconjugated α-actinin antibody. Computational-based microscopic analysis of cardiomyoctye cell size was conducted in a 96-well format. The in vitro culture of transfected cardiomyocytes under basal and hypertrophy-inducing conditions enabled us to classify the investigated miRNAs according to their function. After the first screening procedure approximately 15% of the investigated miRNAs displayed a phenotypic effect either by promoting or inhibiting cell growth. In a subsequent validation process the phenotypic effect of the majority of candidate miRNAs could be verified. A defined set of miRNAs induced strong promotion (> 2 fold) or inhibition (< 0.5 fold) of cardiomyocyte cell growth. Taken together, this high-throughput screen disclosed the phenotypic effect of a broad range of miRNAs on cardiomyocyte cell growth. 1. Rudolf Virchow Center, DFG-Research Center for Experimental Biomedicine, JuliusMaximilians University, Wuerzburg, Germany, 2. Institute of Pharmacology and Toxicology, TU Munich, Munich, Germany

52

243 Methyl-CpG-binding protein 2 induces cardiac failure in mice Weiß S. (1), Gilsbach R. (1), Barreto-Pereira H.F. (1), Hein L. (1) Background: Methyl-CpG-binding protein 2 (MeCP2) was first identified as a binding partner for specific DNA sequences which are methylated at cytosine residues. MeCP2 is an important epigenetic modulator of gene expression in neurons, but its role in the cardiovascular system is unknown at present. Preliminary experiments have suggested that MeCP2 expression may be altered in human heart failure. Methods and Results: In order to search for a possible function of MeCP2 in the heart, we determined mRNA expression of MeCP2 in wild-type (WT) after induction of cardiac hypertrophy by transverse aortic constriction (TAC) or in mice with increased sympathetic tone due to genetic deletion of presynaptic α2-adrenoceptors (α2ABCKO). In WT mice, TAC increased ventricle/body weight ratio significantly as compared with sham-operated mice after 8 weeks. In WT TAC mice, cardiac MeCP2 mRNA levels were decreased by >90% as compared with the sham group. Cardiac MeCP2 mRNA levels also showed a similar decrease in α2ABCKO mice. In order to investigate the role of MeCP2 for the development of cardiac hypertrophy and failure, transgenic mice overexpressing MeCP2 under the control of the cardiac α-myosin heavy chain gene (αMHC) promoter were generated. MeCP2 founder mice died within 2 months after birth (n=6, p<0.01 KaplanMeier analysis vs. littermates) due to a severe cardiomyopathy, characterized by cardiac hypertrophy and severe interstitial fibrosis. Additionally, markers of the fetal gene program e.g. ANP were differentially regulated. In order to search for the mechanism of MeCP2-induced cardiac failure and possible cardiac MeCP2 target genes, whole genome expression analysis was performed. In total, 185 genes were significantly regulated (>1.5fold, p<0.05) in hearts of MeCP2 transgenic mice. 44 out of 185 regulated genes were previously identified as target genes of MeCP2 in the central nervous system. Conclusion Recognition of methylated DNA sequences by the DNAbinding protein MeCP2 may significantly affect cardiac gene expression profiles and contributes to the development of cardiac hypertrophy and failure. 1. Institute of Exp. and Clin. Pharmacology and Toxicology, University of Freiburg, Germany

244 Unchanged in-vivo conduction velocity and cellular Na+-currents in human chronic atrial fibrillation Christ T. (1), Endig S. (1), Rauwolf T. (2), Knaut M. (3), Wettwer E. (1), Ravens U. (1) Success rate of pharmacological cardioversion of atrial fibrillation (AF) with Na+ channel blockers declines within a few months of AF duration, suggesting alterations in Na+ channel function by atrial remodeling. One possible alteration is impaired Na+ channel sensitivity to pharmacological block. Since little is known about Na+ channel remodeling in AF, we have studied the effect of the widely-used Na+ channel blocker flecainide on Na+ currents (I) in isolated myocytes from patients in AF and in sinus rhythm (SR). Invivo INa function was assessed by measuring conduction velocity in right atrial appendages prior biopsy. Under in-vivo conditions atrial conduction velocity was not different in patients with AF and SR 1.08 ± 0.13 m/s vs. 1.11 ± 0.16 m/s (n = 12 each group, cycle length 500 ms). In isolated myocytes INa density (test pulse potential –30 mV, holding potential –110 mV, extracellular sodium concentration 5 mM, 22°C) was slightly but significantly lower in AF compared to SR (24.8 ± 1.3 pA/pF vs. 30.1 ± 1.2 pA/pF, n = 31/18 vs. 77/28 cells/patients, p < 0.05). Voltage-dependencies of activation and inactivation were not different in AF and SR (V0.5act –80.1 ± 0.16 mV vs. -80.5 ± 0.25 mV and V0.5inact –33.6 ± 0.15 mV vs. -35.4 ± 0.24 mV, n = 34/15 vs. 58/20 and 30/14 vs. 69/21). Maximum upstroke velocity (dV/dtmax) and conduction time measured in intact atrial trabeculae were also not different in AF and SR (dV/dtmax 211 ± 13 V/s vs. 214 ± 20 V/s; n = 14 and 8). Potency and efficacy of flecainide to block INa and to reduce dV/dtmax were in the same order of magnitude in AF and in SR (IC50 for channel block 2.4 µM in AF and 3.0 µM in SR, n =11/5 vs. 8/5; EC50 for reduction of dV/dtmax 11 µM in AF vs. 14 µM in SR, n = 8 and 14). Despite the slight reduction in INa density upstroke velocity and atrial conduction are not slowed in AF. Sensitivity of Na+ channel to flecainide-induced block persists in chronic AF. Therefore the lack of the drug’s efficacy to cardiovert long-standing AF must be due to additional mechanisms most probably induced by ongoing electrical and structural remodeling. 1. Department of Pharmacology and Toxicology, Medical Faculty Carl Gustav Carus, Dresden University of Technology, Germany, 2. Department of Cardiology, Medical Faculty Carl Gustav Carus, Dresden University of Technology, Germany, 3. Department of Cardiac Surgery, Medical Faculty Carl Gustav Carus, Dresden University of Technology, Germany

245 Alpha-adrenergic regulation of IK1 and ACh-gated IK,ACh is impaired in patients with atrial fibrillation Voigt N. (1), Bollmann B. (1), Wettwer E. (1), Matschke K. (2), Ravens U. (1), Dobrev D. (1) Background: In chronic atrial fibrillation (cAF) IK,ACh develops arrhythmogenic AChindependent constitutive activity with limited additional activation via muscarinic receptors. The extent of activity of the inward rectifier currents IK1 and IK,ACh is known to depend on PIP2 membrane content. Here we have studied the effect of membrane PIP2 depletion via alpha-1-adrenoceptor (AR) stimulation on IK,ACh activity in patients with SR and cAF. Methods: IK1 and IK,ACh were measured with whole-cell voltage-clamp in isolated right atrial myocytes from 29 sinus rhythm (SR) and 17 cAF patients. Results: Basal current in the absence of muscarinic-receptor agonists was confirmed to be higher in cAF (-22.3±2.7 pA/pF, n=42/17 vs. -12.7±1,1 pA/pF, n=61/29; p<0.05), whereas carbachol-activated (CCh) IK,ACh was lower (7.2±0.9 pA/pF vs. 11.8±1.0 pA/pF; p<0.05). In SR alpha-1-AR stimulation with phenylephrine (PE, 100 µM) reduced basal current by 51% and CCh-activated IK,ACh by 79%. Neomycin (NEO, 500 µM), a blocker of phospholipase-C (PLC), prevented the PE effect suggesting that PE blunts IK1 and IK,ACh activity via PLC-induced depletion of membrane PIP2 content required for both currents. The PE effects on basal current and agonist-dependent IK,ACh were blunted in cAF suggesting impaired alpha-1-AR channel regulation in cAF. In contrast, beta-AR stimulation with isoprenaline or activation of protein kinase A with forskolin had no effect on basal current and agonist-dependent IK,ACh in both SR and cAF. Conclusions: In SR

activation of alpha-1-AR reduces IK1 and agonist-dependent IK,ACh, whereas beta-AR stimulation has no effect. The alpha-1-AR-mediated attenuation of IK1 and the inhibition of agonist-dependent IK,ACh are impaired in cAF. The blunted sympathetic inhibition of IK1 and IK,ACh in cAF is likely to contribute to AF promotion and maintenance. 1. Dept. of Pharmacology and Toxicology, Dresden University of Technology, Dresden, Germany, 2. Dept. of Cardiosurgery, Dresden University of Technology, Dresden, Germany

246 Impaired sodium-dependent regulation of ACh-activated IK,ACh in patients with chronic atrial fibrillation Trausch A. (1), Voigt N. (1), Mintert E. (2), Pott L. (2), Matschke K. (3), Ravens U. (1), Dobrev D. (1) Background: In chronic atrial fibrillation (cAF) IK,ACh develops agonist-independent, constitutive activity, but exhibits smaller additional activation via muscarinic (M)receptors than in SR. IK,ACh is a heterotetramer composed of Kir3.1 and Kir3.4 subunits. Intracellular Na+ is supposed to enhance IK,ACh activity probably by stronger Gβγ-binding to Kir3.4 channel subunit. Here we tested the hypothesis that impaired Na+ dependence of IK,ACh contributes to cAF-related IK,ACh remodeling. Methods: IK1 and IK,ACh were measured with whole-cell voltage-clamp in isolated right atrial myocytes from 16 sinus rhythm (SR) and 9 cAF patients. Sodium dependence was investigated using intracellular solutions with different Na+ concentrations (0 mM-60 mM). IK,ACh channel subunit proteins were studied with Western blot. Results: With [Na+]i=8 mM, basal current in the absence of M-receptor agonists was confirmed to be higher in cAF (‑17.2±2.0 pA/pF, n=15/8 [myocytes/patients]) than in SR (-9.6±1.3 pA/pF, n=15/7; p<0.05). This current was independent of [Na+]i in either group. With [Na+]i=8 mM CChactivated IK,ACh was lower in cAF (‑2.3±0.5 pA/pF) compared to SR (‑7.2±1.0 pA/pF; p<0.05). In SR this current strongly depended on [Na+]i, being 43% lower in the absence and 19% higher in the presence of 60 mM [Na+]i. In cAF, however, CCh-activated IK,ACh was independent of [Na+]i. Loss of Na+ dependence of IK,ACh in cAF could be due to a shift in subunit expression ratio Kir3.4:Kir3.1. This ratio was 0.25 in cAF compared to 0.80 in SR. Conclusions: [Na+]i modulates human atrial IK,ACh in a concentrationdependent manner, however, the Na+-dependence is blunted in cAF. A possible explanation for this phenomenon is the altered Kir3.4:Kir3.1 channel stoichiometry. These results provide a potential molecular basis for the limited M-receptor-mediated activation of IK,ACh in cAF patients. 1. Dept. of Pharmacology and Toxicology, Dresden University of Technology, Dresden, Germany, 2. Institute of Physiology, Ruhr-University Bochum, Bochum, Germany, 3. Dept. of Cardiosurgery, Dresden University of Technology, Dresden, Germany

247 A role for STIM1 in cardiac calcium regulation Merkle S. (2), Ramanujam D. (2), Mühlstedt S. (1), Dupuis M. (1), Braun A. (1), Nieswandt B. (1), Engelhardt S. (2) Calcium (Ca2+) is an universal intracellular second messenger that regulates a variety of fundamental cellular functions in various cell types. In cardiomyocytes, cytoplasmic Ca2+-levels are involved not only in excitation-contraction coupling but also are known to modulate signal transduction. Increased cytosolic Ca2+-levels can contribute to cardiomyocyte hypertrophy, a key step in the development of heart failure. It has been reported in non-cardiomyocyte cells that depletion of the sarcoplasmic reticulum activates store-operated calcium entry (SOCE) across the plasma membrane. Stromal inteaction molecule 1 (STIM1) and calcium release-activated calcium modulator 1 (ORAI1) are described to be essential for controlling SOCE. STIM1 is considered as the Ca2+-sensor in the endoplasmic reticulum, ORAI1 has been described as the putative Ca2+-channel in the plasma membrane. While this process has been studied intensively in many cell types, the cardiac function of these proteins is currently unknown. Both in a murine heart failure model and in human disease, we found dysregulation of the STIM1 protein. In a series of in vitro experiments, we studied the function and interplay of STIM1 and ORAI1. Using confocal and TIRF microscopy, we found that depletion of intracellular Ca2+-stores induced translocation of fluorophore-tagged STIM1 protein into distinct puncta that appear to be enriched near the cell periphery in neonatal rat cardiomyocytes (NRCM). In contrast, a constitutive active mutant of STIM1 was already prelocalized in puncta independent of store depletion. Overexpression of STIM1 together with ORAI1 leads to a strong amplification in store-operated entry in neonatal cardiomyocytes. Moreover, we identified a significant increase in cardiomyocyte cell size when STIM1 was overexpressed in NRCM. Coexpression of STIM1 together with ORAI1 potentiated this effect (p<0.05). Taken together, our findings suggest a role for STIM1 and ORAI1 in cardiomyocyte SOCE and thus possibly in cardiomyocyte hypertrophy. 1. Rudolf Virchow Center/DFG-Research Center for Experimental Biomedicine, JuliusMaximilians University, Wuerzburg, Germany, 2. Institute of Pharmacology and Toxicology, TU Munich, Munich, Germany

248 Function and regulation of the Cav1.2 calcium channel in the cardio-vascular system Domes K. (1), Blaich A. (1), Lemke T. (2), Welling A. (1,2), Moosmang S. (1,2), Hofmann F. (1) Voltage-gated calcium channels play a central role in regulating the electrical and biochemical properties of the heart, neurons, muscle cells and other tissues. Among the different types of calcium channels, Cav1.2 L-type calcium channel is very important for the cardio-vascular system. The C terminus of Cav1.2 contains several structural elements that regulate the calcium influx. The distal C-terminal fragment of the Cav1.2 channel has been reported to be a transcription factor regulating more than 50 different genes [1]. We generated a mouse line that carries 3 stop codons C-terminal to analyze the Cav1.2 C-terminus’ role in the cardio-vascular system. These stop codons terminate the amino acid sequence after a.a. 1905 deleting a AKAP binding site (“Stop mouse”). Deletion of the C-terminus in the stop mouse results in embryonic lethality between day

53

E18.5 and P0. Remarkably, these mice almost completely lack the Cav1.2 protein at day E18.5 as shown by Western blot. Furthermore, the deletion of the C-terminus leads to a different heart size. Compared to wild type, the hearts of the stop mice have an approximately 10% increase in heart size at days E15.5 and E18.5. In contrast, inactivation of the CACNA1C gene (Cav1.2 gene) [2] does not result in an increased heart size. These findings indicate that deletion of the distal C-terminus affects the cardiac development different than complete deletion of the CACNA1C gene. In conclusion, the heart phenotype of the Stop mice indicates further a high relevance of the distal C-terminus of the Cav1.2 calcium channel for the function of the cardiovascular system. [1] Dolmetsch et al. Cell 127(3): 591-606 (2006) [2] Seisenberger et al. J Biol Chem. 275(50): 39193-39199 (2000) 1. FOR923 at Institut für Pharmakologie und Toxikologie, TU München, Germany, 2. Institut für Pharmakologie und Toxikologie, TU München, Germany

overexpression of ERK2T188D enhanced it. The extent of hypertrophy correlated well with the induction of marker genes of hypertrophy and interstitial fibrosis. Fractional shortening, left ventricular contractility and relaxation were reduced after TAC only in ERK2T188D-mice indicating cardiac dysfunction. To investigate downstream effects of Thr188-phosphorylation that might cause cardiac hypertrophy, we determined the phosphorylation status of ERK1/2-targets in the different ERKT188-mutant mouse lines. TAC increased phosphorylation of the cytosolic proteins p90RSK and p70S6K in all lines to a similar extent. In contrast, TAC massively increased phosphorylation of nuclear targets of ERK1/2 (Elk1, MSK1, c-Myc) in ERK2T188D-hearts, to some extent in ERK2T188T and wt-hearts but very little in ERK2T188S and ERK2T188A hearts, indicating that Thr188 phosphorylation of ERK2 specifically promotes phosphorylation of its nuclear target proteins. In summary, direct interaction of Gβγ and with the MAPK cascade leads to phosphorylation of ERK1/2 at Thr188 and promotes phosphorylation of nuclear target proteins, which leads to the induction of cardiac growth. 1. Dept. of Pharmacology and Toxicology, University of Würzburg, Würzburg, Germany

249 Desensitisation of cardiac serotonin receptors in a transgenic mouse model Frenker J. (1), Gergs U. (1), Neumann J. (1) Serotonin (5HT) exerts positive inotropic effects (PIE) in the human heart mediated by 5HT4-receptors. In order to better understand the function of these receptors, we have generated a transgenic mouse (TG) that overexpresses human 5HT4a-receptors under the control of the α-myosin-heavy-chain promoter in the heart. We have shown previously that this receptor is functional in eliciting inotropic effects in isolated left atrial preparations of TG. Wild type mice do not exhibit any inotropic effect to 5HT and were thus not further studied. Here, we hypothesized that like in cell culture a homologous or heterologous desensitisation of the cardiac 5HT4-receptor in the organ bath could occur. To that end, in isolated electrically driven left atrial preparations from TG, we generated cumulative concentration response curves to 5HT (10-9M to 10-6M). Thereafter, samples were treated for one hour with 600 µM serotonin or isoproterenol (10 µM) or buffer. Initially, 5HT induced PIE with an EC50 value of 1.22 x 10-8M. Treatment with 5HT led to desensitisation of the receptors as apparent from reduced maximal inotropic effect and reduced potency (EC50 value 1.20 x 10-7M). On the other hand, pre-treatment with Iso hardly effected the efficacy or potency of subsequently applied 5HT (EC50 value= 3.19 x 10-8M, n= 6). In summary, we conclude that ex vivo the human 5HT4a-receptor undergoes homologous desensitisation which might be relevant in patients, e.g. suffering from high circulating 5HT levels in thrombosis (supported by the DFG). 1. Institut für Pharmakologie und Toxikologie, Med. Fakultät, Martin-Luther-Universität Halle-Wittenberg, Halle(Saale), Germany

250 Morphine administration at reperfusion limits myocardial injury via increased phosphorylation of HSP27 Mourouzis I. (1), Saranteas T. (2), Perimenis P. (1), Tesseromatis C. (1), Kostopanagiotou G. (2), Pantos C. (1) Background and Objectives: The present study was conducted to investigate the effects of morphine administration on reperfusion injury as well as potential molecular mechanisms underlying this response. Methods: Hearts after stabilization, were subjected to 30 min of zero-flow global ischaemia (I) and 45 min of reperfusion (R), (CONT), n=10. Morphine (10-6M) was administered (in the perfusate) only at R, (MORPH), n=10. Post-ischaemic recoveries of left ventricular (LV) developed pressure were expressed as % of the initial value (LVDP%). At the end of the experimental protocol LDH release in the perfusate was measured and the left ventricle was isolated and used for determination of oxidized actin, MAPKs activation and HSP27 phosphorylation. Results: LVDP% was 50.1% ±6.8 in MORPH and 44.07% ±5.5 in CONT hearts, p>0.05 while LDH release was significantly reduced in MORPH as compared to CONT hearts [7.2±0.3 vs 8.8±0.6, p<0.05 respectively]. Furthermore, LVDP% was negatively correlated to LDH release in CONT hearts (r=-0.8, p=0.006), while in MORPH hearts no correlation was found (r=-0.2, p=0.57). The levels of phosphorylated p38 MAPK, JNKs, ERKs and Akt at 45min of R were similar in both groups, p>0.05. However, 1.5 fold increase in the ratio of phospho-HSP27 to total HSP27 was found in MORPH hearts as compared to CONT hearts, p<0.05. In addition, the ratio of oxidized actin to total actin was found to be 1.9 fold more in MORPH as compared to CONT hearts, p<0.05. Conclusion: Morphine administration at reperfusion limits the extent of myocardial injury and HSP 27 is shown to be involved in this response. 1. Department of Pharmacology, University of Athens, School of Medicine, Athens, Greece, 2. 2nd Department of Anesthesiology and Cardiovascular Intensive Unit, Attikon Hospital, University of Athens School of Medicine, Greece

251 Gβγ-induced ERK2 phosphorylation at Thr188 directs MAPK activation towards cardiac hypertrophy Lorenz K. (1), Schmitt J. P. (1), Schmitteckert E. M. (1), Lohse M. J. (1) Activation of the MAPK-cascade consisting of Raf1, MEK1/2 and ERK1/2 can lead to cardiac hypertrophy. The upstream kinases MEK1/2 phosphorylate ERK1/2 at the TEYmotif. Activated ERK1/2 have multiple cellular target proteins and it is largely unknown how ERK1/2 accomplishes signaling specificity towards cardiac hypertrophy. Coimmunoprecipitation assays with purified proteins showed that ERK2 can directly interact with Gβγ. This interaction of ERK2 with Gβγ resulted in increased ERK2 phosphorylation but at a site other than the canonical phosphorylation site within the TEY-motif. 2D-phosphopeptide mapping revealed a new autophosphorylation site of ERK2 at Thr188, which was verified by the use of Thr188-phosphospecific antibodies. Gβγ mediated Thr188 phosphorylation was found in hypertrophied murine hearts and in human heart failure as well as in cardiomyocytes upon Gαq coupled receptor activation by Ang II or phenyephrine. A causal relationship of Thr188 phosphorylation and cardiac hypertrophy was investigated in ERK2 mutants, which were either phosphorylation deficient (ERK2T188A or ERK2T188S) or simulated Thr188 phosphorylation (ERK2T188D). After transverse aortic constriction (TAC), overexpression of ERK2T188A or ERK2T188S led to a significant decrease in the development of left ventricular hypertrophy whereas

252 Reduced Ik,slow1 contributes to the enhanced cardiac function and calcium cycling alterations in ATF1 knockout mice Matus M. (1), Schlute J.S. (1), Sur T.M.H. (1), Seidl M.D. (1), Schmitz W. (1), Müller F.U. (1) The transcriptional regulation by transcription factors of the CREB/CREM/ATF1 family represents a fundamental mechanism of a cAMP-dependent gene control possibly involved in the pathogenesis of heart failure. Activating transcription factor 1 (ATF1) is expressed ubiquitously, however, its cardiac role remains unknown. The knockout of ATF1 (KO) leads to increased cardiac contractility and output, based upon increased, prolonged calcium transients compared to wildtype animals (WT). Moreover, we recorded action potentials using perforated patch whole cell current clamp. Action potential duration APD90 was significantly increased in KO mice (APD90[ms]: KO (n=25), 106±11*; WT (n=19), 72±5, *p<0.05 vs. WT). No changes in L-typ calcium channel I-V relationship, steady-state activation or inactivation, or in outward potassium peak and steady-state current amplitudes were detected. Separation of Ik,slow1 via 50µM 4‑aminopyridine (4AP) showed a 23-30% reduction of peak current amplitude of Ik,slow1 in KO mice. In calcium transients measurement, the inhibition of Ik,slow1 via 4AP led to an increase in calcium transient amplitude and contractility of WT cells to the level of KO. WT (n=5) KO (n=3) Ctr 4AP Ctr 4AP CaT amplitude 0.07±0.02 0.17±0.03† 0.14±0.01* 0.18±0.01 Cell shortening 1.1±0.4 3.3±1.0† 6.8±0.1 5.9±2.2 [%] In conclusion, the observed increased contractility and calcium transient amplitude might be explained by a prolonged action potential due to a decrease in the 4AP-sensitive current Ik,slow1. (Supported by DFG). 1. Institut für Pharmakologie und Toxikologie, Universität Münster, Universitätsklinikum, Münster, Germany

253 Regulation of methyltransferases following dororubicin treatment in cardiomyoctes Matthias M. (1), Bien S. (2), Budde T. (3), Rimmbach C. (4), Rosskopf D. (5), Kroemer H.K. (6) Doxorubicin is one of the most important anticancer drugs. Nevertheless the exact mechanism of action reamains still unclear. A variety of pathways seems to be responsible for its antitumoral action. Among these are DNA intercalation and inhibition of macromolecular biosynthesis, inhibition of topoisomerase II with DNA strand breaks as well as activation of apoptotic cascades. Moreover, production of reactive oxygen species is discussed. The clinical use of Doxorubicin is limited by its cardiac side effects. The risks of developing such cardiac side effects, including congestive heart failure, dilated cardiomyopathy, and death, dramatically increases with the cumulative dose of Doxorubicin. Using the Affymetrix Genechip technology, we investigated the transcriptom of control and Doxorubicin treated mice (C57 Bl10, n = 5, 5d). We observed, among other effects, a 67 fold induction of the DNA methyltransferases 3B (Dnmt3B). We therefore decided to analyze DNA methyltransferases on the mRNA level via Real-Time -PCR. Using this technology we screened DNA methyltransferases 3B and 3A (Dnmt3A) beeing de Novo methyltransferases, DNA methyltransferase 1 (Dnmt1) and DNA methyltransferase 1 associated protein (Dmap1), which are steady state methyltransferases. As cellular model for cardiomyocytes we used HL-1 cells treated for 72h with DMSO and Doxorubicin. Dnmt3B showed a maximum induction of 3.5 fold after 48h, Dnmt3A 2.9 after 72h (p = 0.0108). Dnmt1 had its peak expression after 48h with 4.8 (p = 0.0329) and Dmap1 after 48h with a 4.2 fold induction. The global DNA methylation was investigated using bisulfite treatment of DNA and simultaneous PCR of multiple DNA repetitive elements (Alu elements). HL-1 cells treated with DMSO or Doxorubicin showed significant differences in methylation pattern. In summary, our results indicate that Doxorubicin causes transcriptional alterations of methyltransferases combined with changes in the gene methylation pattern in cadiomyocytes 1. Dept. of Pharmacology, Ernst-Moritz-Arndt Univerity, Greifswald, Germany

254 Electrophysiological recordings in acute heart slices from adult guinea pigs Bussek A. (1), Lohmann H. (2), Ravens U. (1), Wettwer E. (1) The aim of the study was to establish cardiac tissue slices from adult mammalian hearts as an easy to use, robust model for electrophysiological and pharmacological investigations. In vitro drug testing in intact organs is limited to a single experiment, whereas use of isolated cells allows to perform multiple testing. Acute heart slices take an intermediate position. They have the advantage that the cells remain connected to neighbouring cells in their natural surrounding, which is absent in isolated cell models. In addition several slices can be obtained from a single heart. Method: Guinea-pig hearts

54

were perfused on a Langendorff apparatus with oxygenated high potassium (20mM) Tyrode’s solution. After cessation of spontaneous activity a cube of tissue was cut from the left ventricle and glued at the cutting stage of a vibratome. Vertical transmural slices (450 µm) were cut in cold oxygenated high potassium solution containing 15 mM 2,3butanedione monoxime (BDM). Action potentials from slices were recorded with sharp intracellular micro-electrodes; extracellular recordings were made on a multi-electrode array (MEA) at 37°C, stimulation rate 1 Hz. Results: Under control conditions resting membrane potential (RMP) was –84.8 ± 0.8 mV; action potential amplitude (APA) 120.7 ± 0.8 mV; action potential duration at 90% of repolarisation (APD90) 169.9 ± 4.0 ms (59/22;slices/animals) and increasing stimulation frequency shortened the action potential. Measurements with multi-electrode arrays showed a mean conduction velocity of 0.47 ± 0.15 m/s (n=11). Responses to typical drugs were also tested. E4031 (1µM) prolonged APD90 by 60.2 ± 8.3 ms (IC50 31 nM; n=7), the neuroleptic drug risperidone (10µM) by 38.5 ± 4.7 ms (IC50 0.47 µM; n=11). Conclusion: Our results show that acute slices from mammalian heart are a suitable model for electrophysiological and pharmacological studies in cardiac tissue. This technique allows to measure action potential propagation in the myocardium with largely intact architecture and integrity of cells and intact cell-to-cell contacts. Heart slices are easy to prepare compared to isolated single myocytes (enzyme treatment) or other methods (embedded heart in agarose). This robust model could therefore be a new tool in heart research including risk assessment of drugs. 1. Dept. of Pharmacology & Toxicology, University of Technology Dresden, Germany, 2. Lohmann Neuropharmacological Consulting, Castrop-Rauxel, Germany

255 Sick sinus syndrome - creation of an inducible mouse model Herrmann S. (1), Layh B. (1), Ludwig A. (1) The term sick sinus syndrome or sinus nodal dysfunction (SND) is applied to a broad range of electrophysiological abnormalities including sinus arrest, sinus node exit block, inappropriate sinus bradycardia, impairment in heart rate acceleration and more. Mainly elderly people are affected by this disease and approximately half of all pacemaker implantations in the United States are caused by this syndrome. Although the exact aetiology is not known, it has been suggested that age-dependent degenerative fibrosis of nodal tissue is a main cause. Here we describe the generation of a mouse model which reflects several patho-physiological characteristics of sick sinus syndrome. By the use of two genetically engineered mouse lines, one with a sinoatrial node specific CreERT2 expression (HCN4 KiT) and one with a conditional diphtheria toxin transgene (ROSA26-eGFP-DTA), we achieve a tamoxifen inducible deletion of sinoatrial node (SAN) cells. The deletion results in a degenerative fibrosis of nodal tissue, which mimics the most common intrinsic cause of SND. The degree of the sinoatrial fibrosis can be controlled by the applied dosage of tamoxifen. A low-dose treatment protocol leads to a moderate degenerative fibrosis, whereas a high-dose protocol results in complete ablation of virtually all SAN cells, as demonstrated by histological and immunhistochemical analysis of SAN slices. Surface ECG recordings on sedated mice display several abnormalities like bradycardia, AV-block or impairment in heart rate acceleration. The occurrence of these abnormalities correlates well with the progression of fibrosis and so with the applied dose of tamoxifen. However, surprisingly the complete destruction of the primary pacemaker centre results usually not in a condition that is incompatible with life. 1. Institut für Experimentelle und Klinische Pharmakologie, Universität Erlangen Nürnberg

256 Identification of the ß-adrenergic phosphoproteome Göbel P. (1), Rochais F. (2), Zahedi R. (3), Sickmann A. (3), Engelhardt S. (4) Phosphorylation is an important regulatory key event that allows the modulation of cellular signaling pathways. Many proteins can either change their function, localization or activity depending on their phophorylation status. In the heart, the activation of ßadrenergic receptors is associated with the phosphorylation of several downstream target proteins. Since the sustained activation of these receptors is known to contribute to the maladaptive effects involved in the development of heart failure, the identification of further ß-adrenergic phosphoproteins may lead to a broader understanding of the complex pathways underlying this disease. To address this issue, we employed a mass spectrometry-based approach to assess the cardiac phosphoproteome after ßadrenergic stimulation. For this we used mice that were either treated with a nonselective ß-adrenergic agonist or antagonist. In a first step, two different procedures, namely IMAC (immobilized metal affinity chromatography) and TiO2 were chosen to enrich phosphorylated proteins from murine myocardial tissue. These phosphopeptides were analysed by LC-MS/MS (liquid chromatography tandem mass spectrometry). By subsequent comparative analysis, we found several proteins with at least one exclusive phosphosite under ß-adrenergic stimulated conditions. From these proteins we selected several candidates with a high cardiac expression level and a PKA consensus site for further characterization. Therefore we chose MRM (multiple reaction monitoring), a method that allows the relative quantification of a specific phosphosite by comparing corresponding peak areas of two different samples. In these measurements up to 15fold upregulation of phosphorylation under receptor-activated conditions was found. Taken together, our data significantly expand the array of known downstream targets of the ß-adrenergic receptor. 1. Rudolf Virchow Center/DFG-Research Center for Experimental Biomedicine, JuliusMaximilians University, Wuerzburg, Germany, 2. Developmental Biology Institute, Université de la Méditerranée, Marseille, France, 3. Institute for Analytical Sciences, Dortmund, Germany, 4. Institute of Pharmacology and Toxicology, TU Munich, Munich, Germany

257 Expression of desmosomal proteins in atria of surgical patients Mangold W. (1), Hauptmann S. (2), Hatzfeld M. (3), Simm A. (4), Silber R.E. (4), Gergs U. (1), Neumann J. (1) Structural proteins of the area composita of adhering junctions that connect cardiomyocytes as well as components of the Z discs that form lateral connections between myofibrils to connect the contractile apparatus with the cytoskeleton and the extracellular matrix are known to be important for maintaining structural integrity in cardiac muscle. Mutations or deficiency of many of these proteins including plakoglobin (gamma-catenin), plakophilin 2 and cypher 1 (ZASP) have been linked to heart failure or arrhythmias in patients. Moreover, knock-out animals support a role of these proteins in cardiac functions. We wanted to address the question whether these proteins are expressed in the atrium of the human heart and whether the expression shows an agedependency. To this end, atrial specimens obtained from patients undergoing routine bypass surgery for coronary heart disease were subjected to immunohistology and western-blotting. All patients (n=53) obtained therapy with a β-blocker (metoprolol, bisoprolol or atenolol), most of them were also treated with nitrates and ACE-inhibitors. Patients were in NYHA class 1 or 2 and Canadian angina class 2-3. We noted a detectable expression of plakoglobin, plakophilin 2 and cypher 1 in these atrial preparations on western-blotting and immunohistochemistry. There was a significant increase of cypher 1 expression, but not of plakophilin or plakoglobin with age. The present data extend our knowledge on the expression of anchoring proteins in the atrium of the human heart and indicate an age-dependent expression. It is tempting to speculate that increased expression of ZASP may contribute to mechanical and electrical disturbances in the aging human myocardium. 1. Institut für Pharmakologie und Toxikologie, Med. Fakultät, Martin-Luther-Universität Halle-Wittenberg, Halle (Saale), Germany, 2. Institut für Pathologie, Med. Fakultät, Martin-Luther-Universität Halle-Wittenberg, Halle (Saale), Germany, 3. Institut für Pathophysiologie, Med. Fakultät, Martin-Luther-Universität Halle-Wittenberg, Halle (Saale), Germany, 4. Klinik für Herz- und Thoraxchirurgie, Med. Fakultät, Martin-LutherUniversität Halle-Wittenberg, Halle (Saale), Germany

258 Effect of losartan on cardiac and skeletal muscle functions in a heart failure model Günther S. (1), Baba H.A. (2), Kusche T. (3), Punkt K. (3), Adams V. (4), Jones L.R. (5), Hauptmann S. (6), Gergs U. (1), Neumann J. (1) Cardiac failure is not only accompanied by altered Ca2+ homoeostasis in the heart but also connected with dysfunction and biochemical alterations in skeletal muscle. The present study focuses two questions. Does a genetic model of heart failure lead to impaired function of skeletal muscle? And can alterations in both tissues be treated by the angiotensin-II-receptor antagonist losartan? To this end, transgenic mice with targeted overexpression of calsequestrin in the heart (TG), a Ca2+ binding protein in the lumen of the sarcoplasmic reticulum, were used as a model of heart failure. Cardiac failure was evident at two months of age (relative cardiac weight 6.43±0.19 mg/g in wild type mice [WT] and 11.51±0.37 mg/g in TG, n=5-7, p<0.05) and progressed in five months aged mice (relative heart weight 13.50±1.22 mg/g in TG, n=7), the latest time point of this study. Cardiac hypertrophy was accompanied by an increased degree of fibrosis (from 0.29±0.04 in WT to 0.77±0.06, n=8, p<0.05) quantified by sirius red staining. Cardiac function was greatly impaired in TG as exemplified by reduced pressure development, global edema and cardiac arrhythmias. Moreover, WT and TG were perorally dosed with losartan (5 mg/kg body weight per day) for two or five months. Resorption of losartan and formation of its active metabolite were monitored in plasma and tissue by means of a HPLC method. Losartan led to an improvement of hemodynamic function. Remarkably, the activity of succinate dehydrogenase in skeletal muscle (M. quadriceps femoris) was increased by 275.6% in TG compared to WT (n=4, p<0.05). This increase was greatly attenuated in losartan-treated transgenic mice. We conclude that losartan does not only improve impaired cardiac performance in a genetic model of heart failure but can also reduce increased enzyme activity and can presumably normalize the function in skeletal muscle as well. This present model might be useful to elucidate the mechanism of skeletal muscle dysfunction in heart failure (supported by the DFG). 1. Dept. of Pharmacology, Martin Luther University, Halle/Saale, Germany, 2. Dept. of Pathology and Neuropathology, University of Duisburg-Essen, Essen, Germany, 3. Dept. of Anatomy, University, Leipzig, Germany, 4. Cardiac Center, Leipzig, Germany, 5. Krannert Dept. of Cardiology, Indianapolis, USA, 6. Dept. of Pathology, Martin Luther University, Halle/Saale, Germany

259 Heart-specific overexpression of a dominant negative isoform of the transcription factor AP-2α does not protect from myocardial apoptosis Stümpel F. (1), Schmitz W. (1), Müller F.U. (1) The transcription factor AP-2α (activating protein 2α) plays an important role in mammalian development and in the regulation of proliferation and cell death. AP-2αdeficient mice die at birth from severe cranio-thoraco-abdominal defects. In different adult tissues, AP-2α can promote pro- as well as anti-apoptotic effects. AP-2α is expressed in the human heart and upregulated in idiopathic dilated cardiomyopathy. Furthermore, adenovirus-mediated overexpression of AP-2α triggers apoptosis in rat cardiomyocytes. Here we tested the hypothesis that the inhibition of AP-2α activity in the heart confers resistance against apoptotic stimuli. A mouse model was generated that expresses the human dominant-negative splice variant AP-2αB under control of the heart-specific αMHC promoter, thus decreasing AP-2α activity specifically in cardiomyocytes. Expression was verified by Western Blot analysis. Transgenic (TG) mice were viable and had the same life expectancy as wildtype (WT) animals. Body weights and heart to body weight ratios were unchanged as compared to WT mice. Hearts were excised, retrogradely perfused according to Langendorff, and subjected to normothermic ischaemia (1 h) and reperfusion (4 h). Samples were fixed, embedded, and cut in 5 µm slices, before fragmented DNA was stained using TUNEL (rTdTmediated dUTP Nick End Labelling). TUNEL-positive nuclei were counted in 15 random fields of view per heart at 20x magnification and classified into cardiomyocyte nuclei and

55

non-cardiomyocyte nuclei by morphologic criteria. There was no significant difference between WT (n=11) and TG (n=10) hearts, neither within cardiomyocyte nuclei (WT, 60±13 mm-2 [mean±SEM]; TG, 55±17 mm-2), nor within non-cardiomyocyte nuclei (WT, 134±21 mm-2; TG, 120±19 mm-2). In conclusion, the inhibition of AP-2α activity in myocardium did not protect from an acute (short term) apoptotic stimulus in this transgenic model. 1. Dept. of Pharmacology and Toxicology, Westphalian Wilhelms University, University Hospital, Münster, Germany

260 Protective role of the protein phosphatase 2A against septic stress in the heart Kucerova D. (1), Großmann C. (2), Schreier B. (2), Schulz N. (3), Gergs U. (3), Boknik P. (1), Gekle M. (2), Müller F.U. (1), Neumann J. (3), Schmitz W. (1) Transgenic mice with heart-specific overexpression of the catalytic subunit of protein phosphatase 2A (PP2A) developed cardiac hypertrophy and impaired contractility. These mice also suffered from ventricular dilatation and diminished response to βadrenergic stimulation. We have previously reported that overexpression of PP2A was protective against lipopolysaccharide (LPS)-induced stress with regard to hemodynamic parameters in isolated perfused heart (Schulz et al. 2008 Naunyn-Schmiedeberg´s Arch Pharmacol 377 (Suppl. 1):56). To investigate the protective role of PP2A against septic stress, wild type (WT) and PP2A mice were treated with LPS (30µg/g b.w., 72h). Real time PCR analysis revealed an increase of tumor necrosis factor α (TNFα) expression in WT (from 0.7±0.3 to 4.6±1.7, n=4, p<0.05), but not in PP2A mice (1.3±0.4 vs. 1.5±0.2, n=4) after LPS treatment. To get new insights into the protective role of PP2A at cellular level, we studied effects of LPS (10ng/ml, 3h) in isolated adult ventricular cardiomyocytes. Neither genotype nor LPS-treatment affected cell shortening or amplitude of Ca2+ transients under control conditions. However, relative cell shortening after β-adrenergic (β-AR) stimulation (isoproterenol 1µM, 10min, iso) was decreased after incubation with LPS in WT (control: 1.4±0.6%, iso: 10.6±1.5%†, n=4; LPS: 1.2±0.3%, iso after LPS: 3.7±1.1%*, n=6, p<0.05 vs. control† or iso*). In PP2A cardiomyocytes cell shortening was impaired after β-AR stimulation in absence of LPS (control: 1.0±0.3%, iso: 4.3±1.5%, n=5) but pre-incubation with LPS did not induce further deterioration (LPS: 0.7±0.3%, iso after LPS: 3.8±0.8%, n=8). LPS did not affect the iso-induced effect on the amplitude of Ca2+ transients either in WT (∆ Indo-1 ratio iso: 0.25±0.02, iso after LPS: 0.20±0.01) or in PP2A mice (iso: 0.34±0.05, iso after LPS: 0.30±0.04). In summary, the decrease in cell shortening was not followed by a corresponding decrease of Ca2+ transients amplitude under β-AR stimulation after LPS incubation in WT. This might be due to reduced Ca2+ sensitivity of cardiac contractile apparatus. In addition, our data indicate that despite impaired cardiomyocyte contractility after β-AR stimulation, PP2A overexpression could be protective in septic stress. This might be partly explained by the lack of TNFα increasing-effect of LPS in PP2A mice. 1. Institut für Pharmakologie und Toxikologie, Universität Münster, Münster, Germany, 2. Julius-Bernstein-Institut für Physiologie, Martin-Luther-Universität Halle-Wittenberg, Halle (Saale), Germany, 3. Institut für Pharmakologie und Toxikologie, Martin-LutherUniversität Halle-Wittenberg, Halle (Saale), Germany

261 Colorimetric protein phosphatase inhibition assay for specific detection of phosphatase-inhibitor-1 inhibitory compounds Sotoud H. (1), Kattner L. (2), Eschenhagen T. (1), El-Armouche A. (1) The functional diversity of protein-phosphatase-1 in vivo results from the association of its catalytic subunit (PP1c) with different regulatory or inhibitory subunits, one of which is the inhibitor-1 (I-1) that inhibits PP1c only in its PKA-phosphorylated form. Disruption of the I-1 gene in mice results in protection from catecholamine-induced lethal arrhythmias and cardiac hypertrophy suggesting that pharmacological I-1 blockade may represent a therapeutic strategy in heart failure. Here, we aimed to develop a reliable, cost-efficient in vitro assay system to screen a chemical library for compounds inhibiting I-1. Therefore, we established a colorimetric PP1c inhibition system (recombinant PP1cmediated dephosphorylation of pNPP) in a multi-well-format. This assay was first evaluated with the chemical PP inhibitor cantharidin (Cnt) and the PP1c-proteininhibitor-2 (I-2), which, in contrast to I-1, is constitutively active. Cnt and I-2 inhibited PP1 activity concentration-dependently with the expected IC50 of 550 nM and 2.8 nM, respectively. Similarly, PKA-thiophosphorylated recombinant I-1 inhibited PP1c with an IC50 of ~200 nM, whereas non-thiophosphorylated I-1 was inactive. To further test the feasibility of this system, an 8 residue peptide I-1[7–14] that contains the essential PP1c RVXF-motif for binding and a 10 residue control peptide I-1[18–27] were pre-incubated with thiophosphorylated I-1. Whereas I-1[18–27] was ineffective up to 30 µg, I-1[7–14] concentration-dependently (0.3-30 µg) prevented thiophosphorylated I-1-mediated PP1c inhibition. Thus, an assay for specific detection of I-1 inhibitory compounds has been established and screening of a small molecule library with ~1000 compounds has been started. 1. Dept. of Pharmacology, University Medical Center Hamburg-Eppendorf, Germany, 2. Endotherm GmbH, Saarbrücken, Germany

WT (25 ± 2 fmol/mg protein, n = 8). Furthermore, the effect of forskolin, a direct activator of the adenylyl cyclase (AC), in electrically driven left atria was comparable, e.g. 10 µM forskolin increased force of contraction from 2.1 ± 0.3 mN to 4.1 ± 0.6 mN in TG (n = 9) and from 2.0 ± 0.5 mN to 3.8 ± 0.7 mN in WT (n = 6). In addition, we assessed the cardiac function under in vivo conditions by echocardiography under baseline and after i.p. injection of the β-adrenoceptor agonist isoproterenol (Iso, 2mg/kg). The averaged resting heart rate during isoflurane anesthesia was increased to 502 ± 13 bpm in 12 week old TG (n = 12) compared to 460 ± 10 bpm in age-matched WT littermates (n = 15, p < 0.05). In contrast, the chronotropic effect of Iso was attenuated in TG mice (124 ± 4 % vs. 138 ± 3 %, p < 0.05). The echocardiographic assessment showed an increased diastolic left ventricular diameter (LVDd) in the 12 week old TG mice (3.79 ± 0.1 mm, n = 12 vs. 3.48 ± 0.1 mm, n = 15, p < 0.05). The septal wall thickness during systole (and so the contraction of the wall) was also greater in the TG mice as compared to WT littermates (1.23 ± 0.04 mm vs. 1.14 ± 0.02 mm) under basal conditions but not after Iso injection. In summary, transgenic overexpression of A2A-AR was accompanied by an increased heart rate and contractility. Moreover, in agreement with our previous findings on isolated preparations, the β-adrenergic pathway was attenuated also in vivo. This is apparently not due to alterations of β-adrenoceptor density or AC, but can in part be explained by a decrease of Gs-protein expression. (Supported by the BfArM) 1. Institut für Pharmakologie und Toxikologie, Universität Münster, Universitätsklinikum, Münster, Germany, 2. Medizinische Klinik und Poliklinik C, Universitätsklinikum, Münster, Germany, 3. Institut für Experimentelle und Klinische Pharmakologie und Toxikologie, Universitätsklinikum Hamburg-Eppendorf, Hamburg, Germany, 4. Bundesinstitut für Arzneimittel und Medizinprodukte, Bonn, Germany

263 Involvement of PP5 in cardiac response to LPS-induced sepsis Gergs U. (1), Werner F. (1), Großmann C. (2), Schreier B. (2), Gekle M. (2), Loppnow H. (3), Neumann J. (1) Protein phosphatase 5 (PP5), a Ser/Thr phosphatase, is expressed in all mammalian tissues examined, including the heart, but its (patho)physiological role is still unknown. Putative roles for PP5 include cell cycle regulation and signaling by nuclear receptors, as well as regulation of gene expression. Transgenic (TG) mice with heart-specific overexpression of PP5 were generated, and the influence of PP5 overexpression on myocardial dysfunction and inflammation in a lipopolysaccharide-(LPS)-induced mouse model of sepsis were analyzed. TG and littermate wild type mice (WT) were intraperitoneally injected at day one with 30 µg of LPS per g body weight. As a control, mice were injected with isotonic NaCl solution (n=6 per group). After three days, successful induction of LPS-induced sepsis was indicated by decreased body weight (to about 90%), increased spleen weight (to 165%), and enhanced plasma levels of interleukin-6 (to 2,600 pg/ml; IL-6). Contractility of WT and TG hearts was analyzed in vitro using the isolated work performing heart preparation. We noted a decrease in left ventricular (LV) systolic pressure, maximum rate of LV pressure development (+dP/dt) and decline (-dP/dt) in hearts of LPS-treated mice compared to control. Interestingly, LPS-induced contractile dysfunction was much more pronounced in WT hearts compared to TG hearts (Table). Control LPS ContractileParameters WT WT TG TG

LV pressure (mmHg) +dP/dt (mmHg/s) -dP/dt (mmHg/s)

71 ± 13 2,228 ± 442

66 ± 25 ± 9 2*

36 ± 3*#

1,533 351 699 ± ± ± 63* 105*# 322

-211 -428 ± -1,342 -892 ± 26* 80*# ± 304 ± 170

qPCR TNFα/18s

0.56 ± 0.07

0.18 0.77 ± ± 0.03# 0.06

0.8 ± 0.1

2.8 ± 1.3 ± 0.3 ± 3.4 ± 1.5 0.3 0.1# 1.4 *p<0.05 vs. control, #p<0.05 vs. WT Quantitative PCR revealed that basal cytokine expression (in control hearts) was 3-4 times lower in TG compared to WT hearts (n=5; Table). The degree of IL-6 and tumor necrosis factor alpha (TNFα) mRNA expression following LPS stimulation was much more pronounced in TG as compared to WT. In summary, LPS induced myocardial dysfunction in WT and to lower extent in TG. It is tempting to speculate that increased PP5 activity in TG led to lower cardiac cytokine expression. 1. Institut für Pharmakologie und Toxikologie, Med. Fakultät, Martin-Luther-Universität Halle-Wittenberg, Halle(Saale), Germany, 2. Julius-Bernstein-Institut für Physiologie, Med. Fakultät, Martin-Luther-Universität Halle-Wittenberg, Halle(Saale), Germany, 3. Universitätsklinik und Poliklinik für Innere Medizin III, Med. Fakultät, Martin-LutherUniversität Halle-Wittenberg, Halle(Saale), Germany IL-6/18s

262 β-Adrenergic signalling in A2A-adenosine receptor overexpressing mice Drzewiecki K. (1), Fabritz L. (2), Fortmüller L. (2), Kirchhof P. (2), Laakmann S. (2), ElArmouche A. (3), Wittköpper K. (3), Zimmermann N. (4), Müller F.U. (1), Schmitz W. (1), Boknik P. (1) Adenosine is a protective metabolite that is generated during stress responses in the heart. A2A-adenosine receptor (A2A-AR) up-regulation has been found in failing human hearts, whereas A1-AR and A3-AR expression remained unchanged. To elucidate the (patho)physiological role of increased A2A-AR expression, transgenic mice with heartspecific overexpression of A2A-AR (TG) were generated. Previously, we demonstrated that the β-adrenergic response was diminished in isolated preparations from TG mice, probably due to a decrease of Gs. In the present study, we focused on the mechanism of this desensitization. Thus, the β-adrenoceptor density in membrane fractions was determined. There was no difference between TG (24 ± 1 fmol/mg protein, n = 10) and

264 Altered hemostasis and reduced thrombus stability in mice lacking the pertussis toxin-sensitive G protein G alpha i2 Pexa K. (1), Hagedorn I. (2), Varga-Szabo D. (2), Freudenberger T. (3), Fischer J.W. (3), Piekorz R. (1), Nieswandt B. (2), Nürnberg B. (1,4) The subfamily of pertussis toxin (PTX) sensitive Gαi proteins includes three highly homologous isoforms Gαi1, Gαi2 and Gαi3, of which Gαi2 is predominantly expressed in cardiovascular tissues and thrombocytes. Gαi2 and Gαi3 deficient mice demonstrate specific as well as compensatory functions of these isoforms as indicated by the early embryonic lethality of Gαi2/Gαi3 double deficient animals. In order to analyze the role of Gi proteins in cardiovascular function and hemostasis under conditions when

56

redundancy of Gi isoforms is largely prevented we established an in vivo protocol of PTX treatment of mice. PTX causes a functional uncoupling of Gi proteins from their receptors. Administration of PTX to mice (90 to 150 µg/kg body weight i.p.) resulted in an almost complete inactivation of Gαi proteins in peripheral organs after two days as shown by quantitative 32PADP ribosylation analysis of cellular membranes. Interestingly, PTX treatment in vivo had a pronounced and dose-dependent effect on hemostasis and caused strongly increased tail bleeding times upon tail tip cut. Moreover, aggregation of thrombocytes from PTX-treated mice was impaired upon ADP stimulation, a phenotype similarly observed both in animals lacking the Gi coupled P2Y12 receptor or Gαi2, but not Gαi3. Concomitantly, Gαi2, but not Gαi3 deficiency, was associated with increased tail bleeding times, reduced adhesion of thrombocytes and thrombus formation on collagen coated surface under flow conditions, and impaired platelet aggregation upon collagen stimulation. In contrast, thrombin-dependent platelet aggregation was similar in all genotypes analyzed. Lastly, employing in vivo injury/thrombosis experiments including photothrombotic carotid occlusion and mechanical occlusion of the abdomial aorta we have shown that Gαi2 deficiency also strongly blunts thrombus formation in vivo. Thus, we conclude that Gαi2 is the crucial isoform required for efficient thrombocyte activation and thrombus formation and/or stability. The role of Gαi proteins and their effectors in primary hemostasis and arterial thrombosis requires further analysis. 1. Institut für Biochemie und Molekularbiologie II, Heinrich Heine Universität, Düsseldorf, 2. Rudolf Virchow Zentrum, DFG Forschungsinstitut für Experimentelle Biomedizin, Würzburg, 3. Institut für Pharmakologie, Universität Essen, 4. Institut für Experimentelle und Klinische Pharmakologie und Toxikologie, Eberhard Karls Universität, Tübingen

265 Agonistic and antagonistic effects of 6-chloro-3-alkyl-amino-4H-thieno[3,2-e]1,2,4-thiadiazine 1,1-dioxide derivatives in KATP-channels of rat aortic smooth muscle Teschner C. (1), Grittner D. (1), Rood A. (1), Hansen J.B. (2), Lemoine H . (1) Nielsen et al. (J Med Chem 45: 4171, 2002) reported on new KATPH H N N channel openers derived from diazoxide characterized by high R Cl affinities (KD: 10 to 1000 nM) for SUR1-type KATP-channels. N S S Recently we have reported on cyclic side chains, that O O methylcycloalcyl derivatives act as agonists, whereas cycloalcyl derivatives convert to antagonists (Lemoine et al., this journal 375: 45, 2007). Here we ask the question if a transition of agonistic to antagonistic effects can also be observed with non-cyclic derivatives. Rat aortas were dissected under sterile conditions, disaggregated by the use of collagenase, and cultivated with DMEM (7% CO2) until they reached confluency. Cells of early passages (P1 and P2) were cultivated on 12-well strips (96-well formate, Greiner) and labelled with the membrane potential dye DiBAC4(3) (5 µM; excitation 488 nm, emission > 515 nm) in a salt solution composed of (mM) 120 NaCl, 2.0 KCl, 1.0 MgCl2, 2.0 CaCl2, 5 glucose buffered with 20 HEPES (pH 7.4). Hyperpolarisation indicated agonistic, depolarisation (after prestimulation with 0.2 mM diazoxide) antagonistic effects. Results are listed in the Table. Interestingly, agonistic and antagonistic effects could be observed with pairs of skeletal isomers (0410-0154; 0411-1053). Side chains with a quaternary carbon acted as agonists (maximum hyperpolarisation: 41 to 76 % of the effect of 30 µM bimakalim), those with a tertiary or secondary carbon acted as antagonists (maximum depolarisation: 60 to 91 % of control). agonistic activity antagonistic activity NNC NNC pEC50±ASD pEC50±ASD R R 55,... 55,... Max (%) (n) Max (%) (n) 6.81±0.04 6.75±0.11 0154 0410 91.2±3.7 (10) 41.0±3.7 (15) 7.38±0.05 7.06±0.05 0411 0153 68.4±2.6 (8) 60.0±2.9 (15) 7.86±0.04 6.57±0.03 0470 0145 76.1±2.9 (12) 86.5±2.4 (16) 1. Mol. Drug-Research, Lasermedicine, Heinrich-Heine-Universität, D-40225 Düsseldorf, 2. Novo Nordisk Research, DK 2760, Måløv, Denmark

266 Increased expression of protease-activated receptor-4 in human vascular smooth muscle cells in response to high glucose Dangwal S. (1), Jobi K. (1), Rauch B.H. (1), Schrör K. (1), Rosenkranz A.C. (1) Background and Objective: Elevated plasma glucose levels are associated with vascular injury, inflammation and increased thrombin generation as well as enhanced clotting tendency. Thrombin contributes to vascular remodeling in vivo by stimulating vascular smooth muscle cell (VSMC) mitogenesis and migration. These cellular actions of thrombin are mediated via protease-activated receptors (PARs). We have examined if high glucose influences the expression of the mitogenic thrombin receptors PAR-1, PAR-3 and PAR-4 in human vascular smooth muscle cells (VSMC). Methods: Human saphenous vein SMC maintained in normal glucose (5.5 mmol/L) were serum-deprived prior to stimulation with high glucose (25 mmol/L) ± study drugs. PAR expression levels were determined by quantitative real-time PCR, western blotting and immunofluorescence; promoter activity was measured by luciferase reporter assay. Results: Exposure of human VSMC to high glucose rapidly increased PAR-4 promoter activity (n=4, P<0.05) and PAR-4 mRNA expression (to 3.2±0.7 fold, n=7, P<0.05), which was sustained over 96h (n=7). Neither PAR-1 nor PAR-3 were significantly regulated by high glucose. PAR-4 total protein expression and immunofluorescence were also markedly induced over 48-96h (n=3-4). Accordingly, high glucose pretreatment (48h) enhanced the ability of a PAR-4 activating peptide (GYPGQV, 200µM, 3h) to induce expression of the inflammatory cytokine TNFα (n=5) in human VSMC. The osmolar control mannitol did not influence PAR-4 immunofluorescence, but inhibition of protein kinase C (staurosporine, calphostin-C) completely suppressed high glucose-stimulated PAR-4 mRNA expression. Conclusion: High glucose induces a rapid and sustained increase in PAR-4 expression in human VSMC via protein kinase Cdependent pathways. This might contribute to the enhanced thrombotic and proliferative responses of diabetic vessels after vascular injury.

1. Institut für Pharmakologie Düsseldorf, Germany

&

Klinische

Pharmakologie,

Universitätsklinikum

267 Vasodilatory prostaglandins transciptionally regulates protease-activated receptor-1 in human vascular smooth muscle cells via inhibition of nuclear factor of activated T-cells Rosenkranz A.C. (1), Rauch B.H. (1), Schrör K. (1) Protease-activated receptor (PAR)-1 mediates the mitogenic and inflammatory actions of thrombin in human vascular smooth muscle cells (VSMC). We recently reported that PAR-1 is downregulated by prostacyclin via a cyclic AMP- and protein kinase A (PKA)dependent mechanism. However, the human PAR-1 promoter possesses no functional cyclic AMP response element, suggesting the involvement of other transcription factors. Sequence analysis of the PAR-1 promoter identified a putative binding motif for NFAT (nuclear factor of activated T-cells), a family of transcription factors which are regulated in part via PKA. We therefore examined the role of NFAT2, the predominant isoform in VSMC, in the regulation of PAR-1 in human VSMC. PAR-1 and NFAT2 expression was determined by western blotting and real-time PCR, PAR-1 promoter activity by luciferase reporter assay. NFAT inhibition with cyclosporin A (CsA) or siRNA suppressed PAR-1 mRNA and protein expression and impaired functional responsiveness (interleukin-6 induction; ERK1/2 phosphorylation) to thrombin or PAR-1 activating peptide (TFLLRN). CsA or mutation of the NFAT consensus-site blunted PAR-1 promoter activity, to a comparable degree as the prostacyclin analog iloprost. Both CsA and iloprost induced NFAT translocation from the nucleus to the cytosol, indicating inactivation, and attenuated NFAT/PAR-1 promoter binding interactions in a chromatin immunoprecipitation assay. PKA inhibition reversed the above effects of iloprost. PAR-1 mRNA and protein expression were also suppressed by the EP2 receptor agonist butaprost and phorbol-myristoyl-acetate which induces cyclooxygenase-2 and endogenous prostaglandin generation. Thus transcriptional regulation of PAR-1 appears to be a general property of Gs-coupled prostaglandin receptors. Stimulation of the “effector protein activated by cyclic AMP” (EPAC) did not influence PAR-1 expression. In conclusion, we provide first evidence that vasodilatory prostaglandins regulate PAR-1 thrombin receptors in human VSMC through PKA-dependent inhibition of NFAT transcriptional activity. This mechanism may serve to limit the atherothrombotic and inflammatory actions of thrombus-derived clotting factors to the site of vascular injury. 1. Institut für Pharmakologie & Klinische Pharmakologie, Universitätsklinikum Düsseldorf, Germany

268 The transcription factor ATF1 is involved in the regulation of proliferation and apoptosis in vascular smooth muscle cells Seidl M.D. (1), Sur T.M.H. (1), Steingräber A.K. (1), Schmitz W. (1), Müller F.U. (1) Activating transcription factor 1 (ATF1) a member of the CREB/CREM (cAMP responsive element binding protein and modulator) family of transcription factors and regulates gene expression via binding to a promoter sequence called cAMP-response element (CRE). In response to a variety of growth factors, stress signals, and other agents that elevate intracellular cAMP or Ca2+ levels, activating isoforms of CREB/CREM/ATF1 are phosphorylated and promote the expression of target genes related to cell growth and proliferation. However, the vascular role of ATF1 is unknown. Here, we studied the relevance of ATF1-mediated transcription in the regulation of proliferation and apoptosis in vascular smooth muscle cells (VSMCs), using mice with a global inactivation of ATF1 (KO). The number of proliferative and apoptotic VSMCs was determined in aortic sections as well as in isolated VSMCs. There were no differences between KO and WT mice under basal conditions. Stimulation of isolated VSMCs with the platelet derived growth factor (PDGF) led to a 2-fold increase of proliferating cells in WT, whereas KO VSMCs were not altered in their proliferative activity after PDGF stimulation. Furthermore, the proportion of apoptotic cells was 1.9-fold higher in KO as compared to WT VSMCs in response to oxidative stress induced by H2O2 treatment.In summary, apoptosis was increased and proliferative activity was reduced in KO VSMCs under stimulating conditions. Thus, ATF1 is implicated in the regulation of vascular proliferation and apoptosis and may contribute to the pathogenesis of vascular proliferative disorders. (Supported by the IZKF Münster). 1. Institute of Pharmacology and Toxicology, University of Münster, Germany

269 Smooth muscle cell trans-differentiation during atherogenesis Feil S. (1), Todkar A. (1), Griesbach R. (1), Feil R. (1) The major cause of death in western countries is atherosclerosis, which leads to heart attack or stroke. To understand how different cell types, in particular vascular smooth muscle cells (VSMCs), contribute to the development of atherosclerotic plaques, we analysed atherogenesis in apolipoprotein E-deficient mice via temporally controlled Cre/lox-mediated cell fate mapping. This method allowed us to genetically label VSMCs of the media prior to atherosclerotic plaque formation and to follow their fate during disease progression. We found that medial VSMCs contributed to plaque growth in a clonal manner and that the VSMC-derived plaque cells drastically changed their phenotype as compared to their medial progenitors. To analyse the phenotypic changes in more detail, we evaluated cultured VSMCs, which display a synthetic/de-differentiated phenotype, as a cell culture model of atherogenesis. Both VSMCs in cell culture and VSMC-derived plaque cells lost typical smooth muscle markers (e.g. SM alpha-actin) and instead expressed macrophage markers (e.g. Mac-2) and a number of proteins associated with cellular growth and stress. This study suggests that during atherogenesis medial VSMCs can trans-differentiate to macrophage-like cells with a distinct expression profile. These VSMC-derived cells contribute to plaque growth and might play an important role in vascular diseases. 1. Interfakultäres Institut für Biochemie, Universität Tübingen, Germany

57

270 Insulin like growth factor and tumor necrosis factor alpha induce hyaluronan synthesis in human vascular smooth muscle cells Röck K. (1), Rosen M. (1), Fischer J.W. (1) Introduction Hyaluronan (HA) is a carbohydrate component of the extracellular matrix (ECM) and is synthesized at the plasma membrane by HA synthases 1, -2 and -3 (HAS1-3). HA accumulates in adipose tissue of diabetic mice and is associated with macrophages, suggesting that HA is involved in the inflammatory response during diabetes. Furthermore, HA is abundant in atherosclerotic plaque and might contribute to atherogenesis. The aim of the current study was to analyze whether circulating factors associated with the metabolic syndrome such as adipokines, insulin, growth factors and angiotensin-2 (ang-II) regulate HA synthesis in human vascular smooth muscle cells (VSMC).Methods Cultured human VSMCs were synchronized by serum withdrawal for 24 hours. Subsequently, mRNA levels of HAS1-3 and HA-secretion were determined after stimulation of VSMC with insulin (100 ng/mL), ang-II (0.1 µM), interleukin-6 (IL-6, 10 ng/mL), tumor necrosis factor-α (TNF-α; 10 ng/mL), adiponectin (1 µg/mL), leptin (100 µg/mL) and insulin-like growth factor-1 (IGF-1; 100 ng/mL) plus or minus LY294002 (10 µg/mL) for 6 and 24 hours, respectively. Results HAS3 was rapidly and transiently induced by TNF-α with a maximum after 6 hours (6.2 ± 2.3 fold of control, n=3, p < 0.05), whereas the other HAS-isoforms were not affected. In turn, total amount of secreted HA was markedly increased by TNF-α (1.8 ± 0.4 fold of control, n=5, p < 0.05) after 24 hours. In contrast, after 24 hours HAS2 mRNA was induced in response IGF (2.6 ± 0.4 fold of control, n=3, p < 0.05) and sensitive to LY294002 (n=3, p < 0.05) suggesting the involvement of PI3-kinase. Accordingly, secretion of HA was increased by IGF (1.5 ± 0.3 fold of control, n=3, p < 0.05). No significant alterations of HAS isoform expression was found after a 6 and 24 hour treatment with IL-6, adiponectin, leptin, insulin or ang-II n=3-5.Discussion In conclusion, TNF-α causes a rapid but transient increase of HAS3 whereas IGF resulted in a sustained induction of HAS2 in human VSMC resulting in elevated extracellular concentrations of HA. Thus, these results indicate that factors that are released from inflamed adipose tissue might indeed contribute to the accumulation of HA in atherosclerotic plaques and accelerated atherogenesis during the course of the metabolic syndrome. 1. Institut für Pharmakologie, Universitätsklinikum Essen, Universität Duisburg Essen, Germany

271 Novel insights on the regulation of the endothelin 1 promoter Remmler C. (1), Haenisch S. (1), Cascorbi I. (1) Background: The expression of the vasoconstrictor peptide endothelin-1 (ET-1) is affected by angiotensin II (AngII) and clofibrate (Co). Studies concerning the regulation of preproET-1 have revealed putatively binding sites of GATA-2 (positions -136 to -131) and AP1 (-108 to -102) in the promoter region. Moreover, a frequent adenine insertion (IA) polymorphism, located 138 bp upstream of the translation site in the 5’-UTR, alters significantly the ET-1 expression and mRNA stability. We aimed to investigate the mechanisms of activation of ET-1 transcription in relation to the IA variant in an in vitro model. Methods: The mRNA expression of ppET-1 in endothelial EA.hy926 cells was determined after 2, 6, 12 and 24h using 1 µM AngII, 6, 60 or 600 µM Co. Reporter gene pGL3 basic vector constructs were designed consisting the ppET-1 5’-UTR and parts of the ppET-1 promoter, including the -136/-131 and the -108/-102 promoter region. Further constructs were generated by mutagenesis, receiving the wild type (WT) or IA variant of the 5’-UTR, and mutated -136/-131 rather -108/-102 promoter region. After transfection, luciferase assays were performed in EA.hy926 with or without 1 µM AngII or 600 µM Co. Gel shift assays were performed with nuclear extract of EA.hy926. Results: The mRNA expression was only slightly affected by AngII, 6 and 60 µM Co after 2 hours, whereas 600 µM Co showed clear induction after 2 hours of incubation. The luciferase activity was dependent on the -136/-131 and the -108/-102 promoter region, genotype dependent differences could be observed in all constructs having at least one mutation. AngII did not affect the activity of any construct, while Co enforced the luciferase activity due to functional -136/-131 region. Gel shift assays revealed a binding protein at the 5’-UTR probe for both variants. Comparing GATA and AP1 consensus sequences with the ppET-1 promoter sequence different binding patterns were observed. Conclusions: The influence of the 5’-UTR +138 IA variant on ppET1 expression could be confirmed. In addition, the results suggest that both, the -136/-131 and the -108/-102 promoter region are crucial for activity of ppET-1. However, there was a lack of activation of ppET-1 by AngII, but the activation through Co differed related to the -136/-131 binding site. The results suggest that the ppET-1 regulation via AngII and Co is caused by other transcription factors than GATA-2 and AP1 and additional post-translational mechanisms. 1. Dept. of Experimental and Clinical Pharmacology, UKSH, Campus Kiel, Germany

development was studied in CREM-KO mice on an ApoE-knockout background. Comparison of KOxApoE-KO and WTxApoE-KO revealed no significant differences in plaque formation and dimensions after 20 weeks of atherogenic high-fat-diet.In summary, the inactivation of CREM resulted in a higher degree of vascular remodelling after carotid ligation. This finding might be explained by an increased proliferative activity of isolated KO VSMCs. (Supported by the IZKF Münster). 1. Institute of Pharmacology and Toxicology, University of Münster, Germany

273 Validation of self-made fluorescent cyanine dyes for analysis of thrombin induced changes in the proteome of vSMCs by 2D-DIGE Kaber G. (1), Vormbrock I. (1), Schrör K. (1) The two-dimensional differential gel electrophoresis (2D-DIGE) is the most versatile method for the differential analysis of whole proteomes as well as functional and structural subproteomes by 2D-electrophoresis due to its high sensitivity and reproducibility. Unfortunately, the commercially available fluorescent dyes used for the minimal covalent labeling of protein samples in the DIGE method are extremely expensive, resulting in costs of € 500-600 per gel. To minimize costs and to increase the number of possible experiments, the cyanine dyes (ethyl Cy2, propyl Cy3, and methyl Cy5) were synthesized according to a protocol published by Jung and Kim (Practical syntheses of dyes for difference gel electrophoresis. Bioorg Med Chem. 2006;14(1):9297). Based on this protocol, synthesis is straightforward and can be realized with standard laboratory equipment. In contrast to the protocol of Jung and Kim, the activation of carboxyl groups was carried out by O-(N-Succinimidyl)-1,1,3,3tetramethyluronium hexafluorophosphate (HSTU) and N,N-Diisopropylethylamin (DIEA). After activation to their N-Hydroxysuccinimid (NHS) esters, the dyes were purified on a Lobar silica column (Merck, size B) using a 0-10% methanol gradient in dichloromethane. The purified cyanine dyes were used for the differential analysis of thrombin stimulated and non-stimulated human vascular smooth muscle cells (vSMCs). Protein labeling and separation were performed according to the 2D-DIGE standard protocol. The protein patterns of the DIGE gels were highly reproducible and showed excellent overlapping of differentially labeled protein spots. Based on this experimental approach we were able to detect variations in the expression level of several proteins of interest. The protocol of Jung and Kim describes a simple method for the synthesis of fluorescent cyanine dyes with standard lab equipment. In practical experimental applications the 'home made' cyanine dyes showed fluorescence and labeling properties equivalent to the commercially available products. 1. Institut für Pharmakologie und Klinische Pharmakologie, Heinrich-Heine-Universität, Düsseldorf, Germany

274 Human hyaluronan synthase-3 splicing variant 2 localises to the ER as an active intracellular hyaluronan synthase Dai G. (1), Melchior-Becker A. (1), Mielke C. (2), Boege F. (2), Fischer J.W. (1) Hyaluronan (HA) is an extracellular matrix carbohydrate. HA is synthesized by three HAsynthases (HAS1, HAS2 and HAS3). In humans two variants of HAS3 have been described, HAS3v1 and HAS3v2 (Sayo et al, J Invest Dermatol, 2002). Both HAS3v1 and HAS3v2 are expressed in smooth muscle cells and endothelial cells. HAS3v1 localizes to the cell membrane and plays an important role in HA synthesis. However, it is unknown whether HAS3v2 is enzymatically active and contributes to either extra- or intracellular HA synthesis. The aim of the present study was (i) to study the subcellular localization of HAS3v2 and (ii) to test whether HAS3v2 is engaged in HA-synthesis. Nterminal fusion proteins of yellow fluorescent protein (YFP) and human HAS2, HAS3v1 and HAS3v2 were cloned and stable expression in HEK cells was achieved. YFP-HAS2 and -HAS3v1 localized to the cell membrane as shown by Alexa Fluor 594 wheat germ agglutinin as a membrane marker. Using ER-tracker as a marker for endoplasmic reticulum (ER) and golgi apparatus, only a small portion of YFP-HAS2 and -HAS3v1 were detected in the golgi apparatus. In contrast, a strong signal of YFP-HAS3v2 was found in the ER, whereas HAS3v2 was absent from the cell membrane. YFP-HAS3v2 overexpression increased intracellular HA but not the extracellular and pericellular HA. Next, it was analysed whether HA-HAS3v2 complexes form in YFP-HAS3v2-transfected HEK cells (Kyossev and Weigel, Anal Biochem, 2008). Indeed the HA-HAS3v2 complex was deteced indicating active HA synthesis by this HAS-isoform. Furthermore, lentiviral overexpressing HAS3v2 in human skin fibroblasts led to a pro-proliferative phenotype as evidenced by increased [3H]-thymidine incorporation in response to PDGF-BB. In conclusion, the current data suggest that HAS3v2 localizes to ER and is involved in the synthesis of intracellular HA, which may support proliferation. 1. Institut für Pharmakologie, Universitätsklinikum Essen, Universität Duisburg-Essen, Essen, Germany, 2. Institute of Clinical Chemistry and Laboratory Diagnostics, HeinrichHeine-University, Düsseldorf, Germany

272 Role of transcription factor CREM in the regulation of vascular proliferation Steingräber A.K. (1), Seidl M.D. (1), Sur T.M.H. (1), Schmitz W. (1), Müller F.U. (1) Vascular remodeling is influenced by trauma and proatherogenic factors. The transcription factor CREM, a member of the CREB/CREM/ATF1 (cAMP response element binding protein and modulator, activating transcription factor 1) family, is implicated in the regulation of cell proliferation and apoptosis in vascular smooth muscle cells (VSMC). To elucidate the relevance of CREM for the pathogenesis of vasculoproliferative disorders we analyzed mice with global CREM inactivation (KO) and wildtype controls (WT). Effects of vascular trauma were investigated in a model of carotid ligation. After 3 weeks of recovering the right (ligated) and left (sham) carotid artery of KO and WT mice were prepared before the neointima formation was analyzed by Weigert Resorcin-Fuchsine and Nuclear Fast Red staining. The neointima formation in KO was significantly increased in comparison to WT (p=0.013, n=7-9), whereas the thickness of carotid media was not altered. In isolated VSMCs, the proportion of proliferating cells (anti-Ki-67 staining) was 1.4-fold increased in KO versus WT (p<0.001) after platelet derived growth factor (PDGF) stimulation. In absence of PDGF, there were no differences between the genotypes. Moreover, no significant differences in apoptosis (TdT-mediated dUTP Nick-End Labeling ,TUNEL) were detected in KO versus WT VSMCs after exposition to H2O2. The role of CREM in the atherogenisis

275 Involvement of central NMDA and thromboxane A2 receptors in the pressor effect of anandamide in rats Malinowska B. (1), Zakrzeska A. (1), Kurz C.M. (2), Schlicker E. (2), Wielgat P. (3), Braszko J.J. (3), Göthert M. (1,2) The mechanisms underlying the pressor effect of AEA are complex and not yet fully understood. They involve central (probably in the medulla oblongata) and peripheral (most probably in blood vessels) components (Kwolek et al., Brit. J. Pharmacol. 2005;145:567). The aim of the present study was to determine the central mechanism(s) responsible for the pressor response to anandamide. Experiments were performed in rats anaesthetized with urethane. Intravenous (i.v.) injection of AEA (1 mg/kg i.v.) or its stable analogue MetAEA (0.3 mg/kg i.v.) led to a triphasic alteration of blood pressure (BP), including an increase in BP by about 25% of the basal value. On the other hand, intracerebroventricular (i.c.v.) administration of AEA (10 µg/2µl) caused a pure pressor effect (by about 20% of the basal value) but it was obtained only in the presence of the cannabinoid CB1 receptor antagonist AM 251 (3.0 µmol/kg, i.v. each) and the nonselective TRPV1 receptor antagonist ruthenium red (3.0 µmol/kg, i.v.). The pressor

58

effects of AEA and/or MethAEA (i.v. or i.c.v.) were diminished by about 30% by the thromboxane A2 receptor (TP receptor) antagonists daltroban, sulotroban (10 µmol/kg each, i.v.) and/or SQ 29548 (1 µmol/kg, i.v.). The increase in BP induced by AEA i.v. was also reduced by SQ 29548 (8 µg/2µl i.c.v.) and the rise in BP elicited by AEA i.c.v. was diminished by the non-selective NMDA receptor antagonist dizocilpine (MK-801; 1 µmol/kg i.v.) and by bilateral adrenalectomy by about 70% and 60%, respectively. By contrast, sulotroban (i.v.) failed to affect the pressor response to AEA (1 mg/kg i.v.) in pithed rats and AEA and MetAEA did not bind to TP receptors in rat platelets. Centrally administered NMDA has been demonstrated to evoke the adrenal secretion of catecholamines via a mechanism involving brain TP receptors (Okada et al., Eur. J. Pharmacol. 2008;586:145). We conclude that in rats central thromboxane A2 and NMDA receptors are involved in the anandamide-induced adrenal secretion of catecholamines and its pressor response. AM 251 (N-(piperidin-1-yl)-5-(4-iodophenyl)-1-(2,4dichlorophenyl)-4-methyl-1H-pyrazole-3-carboxamide); NMDA N-methyl-D-aspartate; SQ 29548 [1S-[1a,2a(Z),3a,4a]]-7-[3-[[2[(phenylamino)carbonyl]hydrazine]methyl]-7oxabicyclo[2.2.1]hept-2-yl]-5-heptenoic acid 1. Dept. Exp. Physiol., Med. Univer., Białystok, Poland, 2. Dept. of Pharmacol. & Toxicol., Univer. Bonn, Germany , 3. Dept. Clin. Pharmacol., Med. Univer., Białystok, Poland

276 Inhibition of S-adenosylhomocysteine hydrolase activity in unilateral ureteral obstructed kidneys: role of pH Kloor D. (1), Laszlo S. (1), Hermes M. (1), Stahl M. (2), Riehle R. (1), Piesch C. (1), Müller C.E. (3), Osswald H. (1) S-adenosyl-L-homocysteine (AdoHcy) hydrolase catalyzes the hydrolysis of AdoHcy, which is a product inhibitor of S-adenosyl-L-methionine (AdoMet)-dependent methyltransferases. Therefore the activity of AdoHcy hydrolase regulates intracellular AdoHcy concentration and affects transmethylation reactions. As shown previously AdoHcy hydrolase activity is modulated by ions like phosphate or by cAMP. Since AdoHcy tissue content is increased in unilateral ureteral obstructed (UUO) kidneys in the present study we examined determinants affecting AdoHcy hydrolase activity in UUO kidneys. The left kidney of male Sprague-Dawley rats (200-250g, n=5-8) was subjected to UUO for 24 h. There-after the left kidney was removed and immediately snap-frozen with clamps precooled to the temperature of liquid nitrogen. Sham-treated rats were used as controls. Deproteinized tissue extracts were prepared for measurements of AdoHcy, AdoMet, and adenosine tissue content as well as for mass spectrometry analysis. The enzyme activity of renal AdoHcy hydrolase was determined photometrically in the hydrolytic direction in 50 mM potassium phosphate buffer pH 7.4 based on the reduction of MTT to formazan at 8 578 nm. Activity of purified bovine AdoHcy hydrolase was measured at a pH range of 6.0-9.0. Ex vivo measurement of AdoHcy hydrolase activity shows no difference between UUO- and Sham-kidneys (2.5"0.8 vs. 2.55"0.7 mU/mg). However, incubation of purified bovine AdoHcy hydrolase with deproteinized tissue extracts from UUO kidney reduced enzyme activities by 50% as compared to controls. Mass spectrometry analysis revealed an elevation of benzoic and acetic acid metabolites in the kidney tissue. Interestingly, the activity of purified bovine AdoHcy hydrolase was pH dependent whereby enzymatic activity is greater at alkaline conditions and completely abolished at pH 6.0. Our data suggest that the accumulation of acidic metabolites could result in metabolic acidosis in the 24 h UUO kidneys and contribute, at least in part, to reduced AdoHcy hydrolase activity leading to the elevated renal AdoHcy levels, which in turn inhibit transmethylation reactions. 1. Institut für Experimentelle und Klinische Pharmakologie und Toxikologie, EberhardKarls Universität Tübingen, 2. Zentrum für Molekularbiologie der Pflanzen, EberhardKarls Universität Tübingen, 3. Pharmazeutisches Institut, Pharmazeutische Chemie, Universität Bonn

277 Differences in muscarinic receptor antagonism of cis-/trans-isomers of two propiverine metabolites Propping S. (1), Braeter M. (2), Wirth M.P. (1), Ravens U. (3), Wuest M. (4) The muscarinic receptor antagonist propiverine, used for the therapy of overactive bladder, in vivo undergoes an extensive first pass metabolism leading to several active metabolites, which possess different potencies against muscarinic receptors and L-type Ca2+ channels. Among them, the N-oxide M-5 (2,2-diphenyl-2-propoxy-acetic acid [1methyl-piperid-4-yl]-ester-N-oxide) is the major metabolite found in serum. M-6 (2,2diphenyl-2-hydroxy-acetic acid [1-methyl-piperid-4yl]-ester-N-oxide), also an N-oxide, is generated from M-5 and lacks aliphatic side chain. Both N-oxides can be synthesized as cis-and trans-isomers. Here we have examined the pharmacodynamic effects of M-5cis and M-5trans as well as M-6cis and M-6trans using different models od detrusor contractile function. Detrusor contraction was determined in mucosa-free murine (C57BL6) and porcine urinary bladder strips. Compound effects were measured against electric field stimulated (EFS) detrusor contractions (30 Hz; 1 ms pulse duration every 2 min) as well as against cumulatively generated concentration-effect curves (CEC) for carbachol or a single concentration of KCl (40 mM). EFS-contractions were concentration-dependently reduced by the M-5 and M-6 isomers although to a different extent. M-5cis was slightly more potent than M-5trans, however the differences in log IC50 [M] values did not reach the level of statistical significance (4.78 vs. 4.43 in mouse; 5.17 vs. 5.66 in pig). On the other hand, M-6cis was significantly more potent than M-6trans (7.24 vs. 6.39 in mouse; 7.30 vs. 6.75 in pig). Responses to carbachol were antagonized by M-5 and M-6 isomers due to rightward shifts of the CEC, but only M-5trans also reduced the maximum response in mouse (100 µM >70%). For M-6cis and M-6trans, the respective affinities (pKB values) were 7.13 vs. 6.86 in mouse and 7.75 vs. 6.66 in pig indicating a higher apparent affinity of M-6cis. KCl-induced contractions were not impaired by up to 10 µM of the M-6 isomers, but strongly suppressed by 100 µM of the M-5 isomers. This reveals potential additional effects of both M-5 isomers reducing Ca2+ influx. The isomers of the propiverine metabolites M-5 and M-6 are all biologically active reducing detrusor contraction. While both isomers of M-6 and als M-5cis apparently are surmountable antagonists at muscarinic receptors, M-5trans also possesses a strong non-competitive antagonistic component. 1. Dept. of Urology, Technische Universität Dresden, Germany, 2. APOGEPHA Arzneimittel GmbH Dresden, Germany, 3. Dept. of Pharmacology and Toxicology,

Technische Universität Dresden, Germany, 4. Dept. of Oncology, University of Alberta, Edmonton, Canada

278 Sepsis-induced downregulation of aquaporin-2 and vasopressin V2 receptor expression is attenuated by NF-kappaB inhibition Höcherl K. (1), Schmidt C. (2), Bucher M. (2) Acute renal failure (ARF) is frequently associated with polyuria and urine concentration defects and it is a severe complication of sepsis because it increases the mortality rate. Inhibition of nuclear factor-kappaB (NF-κB) activation has been suggested to provide a useful strategy for the treatment of septic shock. However, the impact on sepsis-induced ARF is still unclear. Therefore, we examined the effect of pyrrolidine dithiocarbamate (PDTC) and of small interfering RNA (siRNA) silencing NF-κB p50/p105 on sepsisinduced downregulation of vasopressin V2 receptors and aquaporin (AQP)-2 channels using a cecal ligation and puncture (CLP) mouse model. CLP caused a time-dependent downregulation of renal vasopressin V2 receptor and of aquaporin-2 (AQP2) expression without alterations in plasma vasopressin levels. Renal activation of NF-κB in response to CLP was attenuated by PDTC pretreatment, which also attenuated the downregulation of V2 receptor and AQP2 expression. Furthermore, a strong nuclear staining for the NF-κB p50 subunit throughout the whole kidney in response to CLP was observed. siRNA against NF-κB p50 attenuated the CLP-induced nuclear translocation of the p50 subunit and the CLP-induced downregulation of V2 receptor and AQP2 expression. Additionally, PDTC and siRNA pretreatment inhibited the CLP-induced increase in renal TNF-α and IL-1β concentration and in NOS-2 mRNA abundance. Moreover, PDTC and siRNA pretreatment ameliorated CLP-induced hypotension and acute renal failure. Our findings suggest that NF-κB activation is of importance for the downregulation of aquaporin-2 channel and vasopressin V2 receptor expression during sepsis. In addition, our data indicate that NF-κB inhibition ameliorates sepsis-induced acute renal failure. 1. Inst. of Physiology, University of Regensburg, Regensburg, Germany, 2. Inst. of Anaesthesiology, University of Regensburg, Regensburg, Germany

279 The role of L1 adhesion molecule in the development of renal cell carcinoma Doberstein K. (1), Altevogt P. (2), Kristiansen G. (3), Pfeilschifter J. (1), Gutwein P. (1) L1 adhesion molecule plays an important role in the development and progression of different types of cancer. The role of L1 adhesion molecule in renal cell carcinoma (RCC) is poorly understood. In our study we analyzed on a tissue microarray with 286 biopsies of RCC patients the expression of L1 adhesion molecule. Interestingly, L1 adhesion molecule was downregulated in 82% of RCC patients. In 7% of the patients we detected L1 expression which was paralleled by poor prognosis of patient survival. In addition we present evidence, that von Hippel Lindau (VHL) Protein can regulate L1 adhesion molecule in renal carcinomas cells. Investigating transcription-factors known to influence L1 adhesion molecule expression we identified PAX-8 as an interesting candidate for the regulation of L1. Importantly, L1 and PAX-8 were co-expressed in collecting duct cells of the normal kidney. Investigating PAX-8 expression in the RCC TMA revealed a downregulation of PAX-8 during RCC progression. Further experiments have to be performed to determine the functional consequences of L1 and PAX-8 downregulation in the progression of RCC. 1. Dept.of Pharmacology, Frankfurt, Germany, 2. German Cancer Research Center, Heidelberg, Germany, 3. University Hospital , Pathology, Zürich, Switzerland

280 The influence of experimental schistosomiasis on the pharmacokinetics and pharmacodynamics of phenobarbital and diphenylhydantoin in mice Assi A.A. (1), Abdel-Rahman M.S. (1) The effect of experimental Schistosoma mansoni (SM) infection on some pharmacokinetic and pharmacodynamic parameters of phenobarbital (PB) and diphenylhydantoin (DPH) was investigated in mice. A single anticonvulsant dose of each of PB (15.6 mg/kg) and DPH (9.8 mg/kg) was given intravenously (i.v.), to normal and SM-infected mice. Two groups of infected mice were used, based on the duration of infection [early infection (45 days after infection) and late infection (90 days after infection)]. The serum and urine levels of their major metabolites [PB: 5-ethyl-5-(4hydroxyphenyl)-barbituric acid; DPH: 5-(4-hydroxyphenyl)-5-phenylhydantoin] were determined in normal and SM-infected mice. In addition, the following pharmacokinetic parameters of PB and DPH: Cmax, Cmin, Cave, AUC0-24, CL, t1/2, and free (nonprotein bound) fraction, were determined. The pharmacodynamics (anticonvulsant effects for PB and DPH and hypnotic effect for PB) of both drugs were also evaluated and correlated to their pharmacokinetics in normal and SM infected animals. Both early and late SM infection significantly decreased their major major metabolites and clearance and increased the unchanged fraction of both drugs, Cmax, AUC0-24 and t1/2. Such effects were, however, greater in mice with late infection. The anticonvulsant effects of both drugs were approximately the same in normal and SM infected animals, in spite of the significant increase in their blood levels. SM infection, however, produced significant prolongation of the phenobarbital sleep time in mice as well as decrease in its hypnotic ED50. In conclusion, experimental Schistosoma mansoni (especially late infection) affects the pharmacokinetics but not the anticonvulsant effects of PB and DPH in mice. It has influence, however, upon the hypnotic effect of PB. Such findings merit further studies in epileptic patients who are infected with SM, to evaluate the efficacy and toxicity in case of their long-term use. 1. Dept. of Pharmacology, Assiut University, Assiut, EGYPT

59

281

284

Measurement of in vivo RISC activity using an in vitro non-radioactive RISC assay Kettner-Buhrow D. (1), Guenther B. (1), Kaelble S. (2), Joshua-Tor L. (3), Kracht M. (1) Argonaute 2 (Ago2) is the core catalytic enzyme of RISC (RNA-induced silencing complex) that mediates RNA cleavage by RNA-interference (RNAi). RNAi is an important gene regulatory mechanism, that depends on small interfering (si)RNAs. The minimal components of RISC are members of the Argonaute family of proteins plus small RNA guides. So far, Ago2 activity was assessed using 32P-cap-labelled in vitro synthesized mRNA targets and ATP as substrates. Cleavage products are detected after purification and separation of reaction mixtures by denaturing Urea-PAGE followed by autoradiography. This assay is time consuming, difficult to handle and only semiquantitative. Recently, we have reported the establishment of a non-radioactive RISC assay in vitro. GST-Ago2 was purified to near homogeneity from E.Coli and incubated with an 180bp or 723bp luciferase mRNA target. Cleavage reaction was started by addition of phosphorylated single stranded complementary RNA oligonucleotides and resulted in complete cleavage of the target RNA. Purified reaction mixtures were separated and visualized using RNA Chips on an Agilent 2100 Bioanalyzer instrument. Here we report, that after optimization with respect to specificity and efficacy it is also possible to use the established in vitro RISC assay to measuring activity of endogenous Ago2 isolated from cells. Thus, the non-radioactive Ago2 assay can be used to identify inhibitors or activators of Ago2 in cell-based assay systems. 1. Rudolf-Buchheim-Institute of Pharmacology, Justus-Liebig-University Giessen, Giessen, Germany, 2. Institute of Pharmacology, Medical School Hannover, Hannover, Germany, 3. Cold Spring Harbor Laboratory, New York, USA

Nanoparticle dispersions of opioids for local pain reduction - influence on wound healing Wolf N.B. (1), Kleuser B. (1), Weindl G. (1), Radowski M. (2), Haag R. (2), SchäferKorting M. (1) Topical applications of opioids can induce effective analgesia for patients with skin wounds, yet optimal doses still have to be investigated. Furthermore, appropriate formulations have to be developed being effective and safe. Most importantly, neither the opioid nor the vehicle should interfere with the process of wound healing. Interestingly, in the keratinocyte cell line HaCaT all tested opioids (morphine, hydromorphone, fentanyl and buprenorphine) enhanced keratinocyte cell migration about 2-fold when applied in a concentration of 100 nM, which is comparable to the efficiency of the positive control TGF-β (1 ng/ml). Measured in a trans-well system, the effect was concentration dependant and could be blocked by the specific opioid receptor antagonist naloxone (10 nM). Thus, opioids enhance wound healing via a direct opioidreceptor related effect. Furthermore the effect was linked to an enhanced production of nitric oxide (NO, measured via the Griess reaction), due to an up-regulation of the expression of the inducible NO-synthase (iNOS) after stimulating the cells with morphine (100 nM). Additionally, the stimulation of migration by morphine (100 nM) was verified studying accelerated closure of a scratch wound that had been caused in a cell monolayer. Next we looked for an appropriate formulation, tolerable even by damaged skin. Solid lipid nanoparticles (SLN) as well as dendritic core-multishell (CMS) nanotransporters are possible carrier candidates [1-2]. In the migration assay unloaded CMS nanotranspoters did not interfere with HaCaT cell migration while unloaded SLN even enhanced it to a similar extent as TGF-β did. References [1] Schäfer-Korting et al., (2007). Adv Drug Deliv Rev, 59, 427-443. [2] Küchler et al., (2008) Nanoparticles for Skin Penetration Enhancement - A Comparison of Dendritic Core-Multishell Nanotransporter and Solid Lipid Nanoparticles. European J Pharm Biopharm, in press 1. Dept. of Pharmacology and Toxicology, Freie Universität, Berlin, Germany, 2. Dept. of Organic Chemistry and Biochemistry, Freie Universität, Berlin, Germany

282 Heterotrimeric G protein subunits and pheromone receptors of vomeronasal neurons are expressed in mouse spermatozoa Borth H. (1), Meyer D. (2), Widmayer P. (3), Gudermann T. (2), Boekhoff I. (1) The efficiency of reproduction in rodents is strongly depending on the recognition of pheromones, mainly realized by neurons of the vomeronasal organ (VNO). Remarkably, sensory neurons of the VNO are able to detect chemically diverse cues, like volatile compounds as well as water soluble peptides and proteins. Since chemoreception is also an absolute prerequisite for sperm cells to find the egg in the aqueous milieu of the female genital tract as well as to recognize and bind the oocyte's glycoprotein matrix, we set out to investigate whether molecular elements involved in the reception of pheromonal compounds are also expressed in mammalian germ cells. Nonameric peptides which typically serve as ligands for major histocompatibility complex (MHC) class I molecules represent chemosensory cues reflecting the individuality of an animal. Since the prototype of vomeronasal receptors of the V2R class, the V2R2, is expressed in MHC peptide responsive neurons in the basal region of the vomeronasal epithelium, we unraveled whether this subtype may also be present in mouse testicular tissue. RTPCR experiments performed with degenerate primers led to the amplification of PCR products for this candidate pheromone receptor from mammalian germ cells. Currently we perform immunohistochemical experiments with subtype specific antibodies in order to verify whether the V2R2 prorein is actually present at detectable levels in testicular tissue. Interestingly, examining the expression of heterotrimeric Gβγ-subtypes in spermatozoa, which in vomeronasal neurons show a typical co-localization with V2Rs, a prominent subcellular expression pattern emerged: The sensory Gβ2γ8 was found to be concentrated mainly in the connecting piece, a small region between the sperm head and tail which recently has been suggested to represent an intracellular calcium store. This observation indicates that molecular components as well as chemosensory cues of vomeronasal neurons may play an important role in ensuring reproductive success. 1. Dept. of Pharmacology and Toxicology, Philipps University, Marburg, Germany, 2. Dept. of Pharmacology, Ludwig Maximilian University, München, Germany, 3. Dept. of Physiology, University Hohenheim, Stuttgart, Germany

285 Tylophorine inhibits PDGF-induced vascular smooth muscle cell proliferation Joa H. (1), Heiss E. (1), Atanasov A.G. (1), Proksch P. (2), Dirsch V.M. (1) Tylophora indica (Asclepiadaceae) is used in ayurvedic medicine to treat various allergic and inflammatory disorders, including bronchial asthma, rhinitis, whooping cough and catarrh; Tylophorine, a phenanthroindolizidine alkaloid, was identified as its main active component. Various analogues of this compound were also shown to inhibit tumor cell proliferation. Thus, the activity profile of Tylophorine might be promising to treat vasculoproliferative disorders, such as restenosis or atherosclerosis. Aim of the study was to investigate the effect of Tylophorine on platelet-derived growth factor (PDGF)-induced rat aortic vascular smooth muscle cell (VSMC) proliferation and to identify the signalling pathways that are affected. We show that Tylophorine at concentrations as low as 100 nM to 1 µM dose-dependently inhibits proliferation of PDGF-activated VSMCs as demonstrated by reduced BrdU-incorporation. Flow cytometric cell cycle analysis indicated an arrest of cells in the G0/G1–phase of the cell cycle, which was confirmed by hypophosphorylated retinoblastoma protein (Rb) and experiments releasing cells from an early S-phase arrest induced by the DNA-polymerase inhibitor Aphidicolin. In order to identify upstream signaling pathways that are affected by tylophorine we performed western blot analyses, which demonstrated that neither the mitogen-activated protein kinases (MAPK) ERK1/2 and p38 nor the protein kinase Akt or the Jak/STAT3 signaling pathways are inhibited by Tylophorine. Although the signaling pathway affected by Tylophorine in PDGF-activated VSMC is not yet identified we consider Tylophorine due to its strong antiproliferative activity an interesting lead that warrants further investigation. 1. University of Vienna, Department of Pharmacognosy, Vienna, Austria, 2. HeinrichHeine-University of Düsseldorf, Department of Pharmaceutical Biology and Biotechnology, Germany

283 Is e-learning what counts? Acceptance, use and effects of e-books in a course on Basic Pharmacology Matthes J. (1), Müller E. (2), Herzig S. (1), Stosch C. (3) Computer-based Learning (CbL) is still matter of debate. Interestingly yet sophisticated e-learning tools are no guarantee for improved exam results (Hahne et al., Med Educ. 2005; 39: 935-43). Currently, the distribution of textbooks as PDF copies increases. Since less might be more we took advantage from classical textbooks being available as PDF (portable document format) in our faculty’s intranet. In a prospective, randomised trial over two summer terms we tested for acceptance and use of these e-books by medical students attending our course on basic pharmacology. Furthermore we looked at attitudes towards CbL and test results in the final written exam at the end of the course. In our course on basic pharmacology (a total of n=269 students during summer terms of 2007 and 2008) half of the small groups were randomly assigned to use the ebook offer while the other half was asked to do not. Tracking data revealed a significant use of subject-related e-books during the course (weeks 1 to 6) only. At the end of the course (week 6 of each summer term) and during the written exam (week 13 of each summer term) students were asked to evaluate their use of e-books and there attitude towards CbL in general. Response rates ranged between 58% and 72%. User and nonuser did not differ regarding their attitudes towards CbL (mean values at 5-point Likertscales, week 6: 2.9±0.1 vs. 2.8±0.1; week 13: 3.0±0.1 vs. 3.0±0.1). Use of e-books was quite low and did not differ when comparing the period during the course and the period after the end of the course putatively spent on exam preparation (≤1h per week for both periods). Interestingly, in both groups, users and non-users, significantly more time was spent with printed books (≥3h per week). There was no difference between users and non-users regarding their results in the final written exam (user: 27.1±1.0 vs. non-users: 26.9±1.0). Taken together the use of textbooks available via intranet as portable document format was low and did not show an effect on students’ attitude towards CbL nor on their results in a written exam in a course on basic pharmacology. 1. Univeristät zu Köln, Institut für Pharmakologie, Köln, Deutschland, 2. Deutsche Zentralbibliothek für Medizin, Köln, Deutschland, 3. Universität zu Köln, Medizinischen Fakultät, Studiendekanat, Köln, Deutschland

286 The mitogen-activated protein kinase DLK inhibits the CREB co-activator TORC Do Thanh P. (1), Oetjen E. (1) Recently, three isoforms of Transducer of Regulated CREB (TORC) were identified as additional CREB co-activators. Whereas the co-activator CBP interacts with on Ser-119 phosphorylated CREB, TORC1 and TORC2 after dephosphorylation of Ser-167 and Ser-171, respectively, translocate into the nucleus and interact with the dimerized leucine zipper of CREB. TORC translocation is inhibited by the immunosuppressive drugs cyclosporin and tacrolimus. Our previous studies showed that both drugs activate the dual leucine zipper bearing kinase DLK. In the present study the effect of DLK on TORC was investigated. A luciferase reporter gene under control of 5 copies of the GAL4 binding-site was transiently cotransfected with expression vector for GAL4-TORC fusion proteins into the beta cell line HIT. Over expression of DLK inhibited TORC1 and TORC2 transcriptional activity by 80 and 40%, respectively. This reduction was less pronounced using the DLK kinase dead mutant (DLK-K185A), whereas a DLK mutant unable to homodimerize (DLK-PP) showed no inhibitory effect on TORC activity. The mutation of Ser-167 or Ser-171 in TORC1 and 2, respectively, to Ala prevented the inhibitory effect of DLK on TORC transcriptional activity. The interaction between TORC and DLK was investigated by an in vitro pull down assay. Bacterially expressed Histagged full length TORC1 or GST-tagged TORC1 (1-44 amino acids) interacted with [35S]-labeled DLK and with DLK-K185A to the same extent. Deletion of the first 44 amino acids of TORC1 prevented this interaction. In addition, the interaction between DLK-PP and TORC1 was reduced by 40%. Our data show that DLK inhibits TORC activity, presumably by phosphorylation of Ser-167 or Ser-171 in TORC1 and TORC2, respectively, thus preventing the nuclear translocation of TORC. In addition, DLK interacts with TORC1. This interaction depends on the N-terminal amino acids of TORC1 and the ability of DLK to dimerize. Thus, DLK might inhibit TORC through an interaction with and by the phosphorylation of TORC, thereby preventing the nuclear translocation of TORC and ultimately the stimulus-induced CREB transcriptional activity. 1. Dept. of Pharmacology, Georg August University, Göttingen, Germany

60

287 The pH dependent partition in In-vitro percutaneous permeation experiments Ennen J. (1), Niedorf F. (1), Stahl J. (1), Kietzmann M. (1) The process of skin permeation is determined by substance partition, diffusion and solubility. These variables can be modified if formulations need to be optimised, either to promote or retard absorption. Substance solubility in the applied formulation and its partition into the skin is dependent on the pH of the formulation. Since the skin’s stratum corneum behaves as a lipophilic membrane only the unionized forms of drugs are able to permeate. This pH dependent partition is well documented for the general absorption of ionisable drugs across the gastrointestinal tract but it is less well described in the dermal and transdermal delivery of drugs. The purpose of the present study was to determine if drugs conform to pH dependent partition in percutaneous permeation experiments and if the interpretation of results can help to predict and modify skin permeability. Therefore the effect of pH on the permeation of three non-steroidal antiinflammatory drugs with different physicochemical properties (AAP: acetaminophen, SA: salicylic acid, FFA: flufenamic acid) has been examined in Franz type diffusion cells using bovine udder skin and silicone membranes. Furthermore, the octanol/water partition coefficient log P was determined in relation to the pH of the donor phase (log D). As expected, the steady-state flux and apparent permeation coefficient (Papp) of SA and FFA depend upon the vehicle pH, and permeation can be related to the log D, which points to a definable relationship between the fraction unionised and Papp. Surprisingly, permeation of AAP is not influenced by the pH, although the octanol/water partition coefficient is clearly pH dependent. These results show that SA and FFA conform to pH dependent partition, whereas AAP does not. Further investigation will be done on the pH-behaviour of AAP, the modulation and route of permeation of ionisable compounds through skin. 1. Institute of Pharmacology, Toxicology and Pharmacy, University of Veterinary Medicine Hannover, Foundation, Hannover, Germany

288 Effects of glucocorticoids on the regulation of sphingolipid metabolizing enzymes in renal mesangial cells Förster A. (1), Emmler T. (1), Reinsberg L. (1), Pfeilschifter J. (1), Huwiler A. (2) Neutral ceramidase (NCDase) and sphingosine kinases (SphK) 1 and 2 are key regulatory enzymes in the sphingolipid metabolism. NCDase deacylates the proapoptotic lipid second messenger ceramide to produce sphingosine, which can be further processed by SphK to sphingosine-1-phosphate (S1P) possessing cell protective effects. Thus, the cellular balance between ceramide and S1P is a crucial determinant for the cell either to undergo apoptosis or proliferation. Apoptosis and proliferation are important responses of renal mesangial cells during renal diseases. In this study, we show that triggering apoptosis of rat renal mesangial cells (rMCs) with TNFa/cycloheximid, staurosporine or the SphK inhibitor SKI II was significantly reduced by the glucocorticoid dexamethasone. Interestingly, dexamethasone increased cellular S1P concentrations suggesting an activation of one of the sphingolipid metabolizing enzymes NCDase and SphK. We found that mRNA, protein expression, and activity of SphK1 were increased by dexamethasone. Furthermore, mRNA and protein expression levels of NCDase were also increased although to a lesser extent. We further cloned the 5’ untranslated region (5’UTR) of the rat NCDase gene to study NCDase promoter activation. Normal rat kidney (NRK) fibroblasts and rMCs were transfected with four different NCDase promoter fragments with sizes of 550 bp, 1126 bp, 1724 bp, and 2285 bp coupled to a luciferase reporter gene. Stimulation of cells with dexamethasone led to an induction of the NCDase promoter activity. Highest activity was seen for the 550 bp promoter construct. Additionally, co-stimulation with the glucocorticoid receptor antagonist RU-486 completely abolished the dexamethasone effect. Furthermore, mutation of two putative glucocorticoid receptor responsive elements (GREs) within the NCDase promoter region reduced the dexamethasone effect drastically. Thus, by upregulating NCDase and SphK1 expression, dexamethasone alters the sphingolipid rheostat and lead to increased cellular S1P levels that inhibit apoptotic cell processes. 1. Pharmazentrum Frankfurt/ZAFES, Klinikum der Johann-Wolfgang-Goethe Universität, Frankfurt am Main, Germany, 2. Institut für Pharmakologie, Universität Bern, Switzerland

289 Electrophysiological properties of skeletal muscle-derived stem cells Poulet C. (1), Christ T. (2), Wettwer E. (3), Ravens U. (4) Recently, Winitsky et al (2005)1 and Arsic et al (2008)2 reported the isolation of a floating population of skeletal muscle-derived stem cells that exhibited cardiomyogenic properties. The aim of our study was to investigate the eletrophysiological behaviour of these cells and compare them to cardiac and skeletal muscle electrophysiological properties. Using a similar isolation protocol, we isolated cells that proliferated in suspension and could generate action potentials (APs) after 5 days of culture. These APs were significantly shorter than ventricular myocytes APs but had a similar amplitude and upstroke velocity (table1). As ventricular myocytes APs, they were not altered by 100nM of Tetrodotoxine (TTX) but completely blocked by 3µM of TTX, suggesting the presence of TTX-resistant sodium channels. The complete block of APs by 500µM Cadmium supported the presence of the cardiac specific sodium channel Nav1.5. We further tested the sensitivity of the sodium current initiating the APs to various concentrations of TTX. By fitting the data with a two-binding site function, two distinct populations of Na+ channels could be clearly separated. The TTX-sensitive fraction (IC50 ~ 6.5nM) represented 28% of the total Na+ channel fraction whereas the TTX resistant fraction (IC50 ~ 2.1µM) amounted to 70% of the total fraction. These values are consistent with the IC50 values of the skeletal muscle channel Nav1.4 and the cardiac specific channel Nav1.5. The expression of transcripts for Nav1.4 and Nav1.5 was confirmed by RT-PCR. We conclude that a cell population with electrophysiological properties compatible with cardiac function can be isolated from skeletal muscle and proliferate in vitro. This cell population might be a good candidate for therapeutic support of myocardial infarction.

Table 1. Action potentials characteristics Ventricular cardiomyocytes (n=13) Skeletal muscle-derived stem cells (n=15)

APD 90% (ms) 34.7 ± 5.1 19.7 ± 2.5*

Amplitude (mv) 108.9 ± 3.6 109.4 ± 3.0

Upstroke velocity (V/s) 167.2 ± 18.8 146.8 ± 11.3

Temperature: 37°C, stimulation rate: 0.5Hz; results are given as mean ± SE. Statistical analysis: Two-tailed Student’s unpaired t-test, * P<0.05 indicates a significant difference from ventricular cardiomyocytes. 1Winitsky et al, PLos Biology 2005; 3(4): 662-71 2 Arsic et al, Exp Cell Res 2008; 314(6):1266-80 1. Dept. of Pharmacology and Toxicology, Technical University, Dresden, Germany

290 Poems as a means of kindling students´ interest in tedious pharmacology? Schlicker E. (1) Medical students in Germany have to attend the basic course in pharmacology and toxicology in the fifth or sixt term. At this time point, they have passed the examination in disciplines like anatomy, physiology and biochemistry only recently and are only gradually approaching the clinical disciplines. The course shall provide students with a systematic and thorough overview of pharmacology and toxicology. Some students complain that the course is too theoretical or in other words drug- rather than disease- or patient-directed. This criticism is justified and poses a challenge upon docents of pharmacology and toxicology to motivate students anyway. I am teaching two groups of 20 students each in twelve 90-min units each per term and have written poems to attract their interest in the admittedly tedious discipline. One poem is presented in each unit. The poems are dedicated to the nomenclature, mechanism of action, pharmacokinetic properties, side effects, contraindications and/or costs of the drugs under consideration. The poems consist of no more than 14 lines and include limericks, ottava rimas and sonnets. An example of an ottava rima (along with the English translation) is given below. The experience with the poems gathered within one term is so far limited but I feel that some students really like them. ZEHN DICKMACHER Sulfonylharnstoffe, Glitazone, Insulin sind Pharmaka, die das Gewicht erhöhen können. Hier sind nebst ßBlockern weiterhin Glucocorticoide zu erwähnen und sechstens Cyproheptadin, siebtens Trizyklika zu nennen. Gilt achtens für fast jedes Neuroleptikum, neuntens für Valproat, zehntens für Lithium. English translation: Ten things making fat. The following single (or groups of) drugs can increase body weight: (i) sulphonyl ureas, (ii) glitazones, (iii) insulin, (iv) ß-blockers, (v) glucocorticoids, (vi) cyproheptadine, (vii) tricyclics, (viii) neuroleptics (with exceptions), (ix) valproate and (x) lithium. 1. Institut für Pharmakologie und Toxikologie, Universität Bonn, Deutschland

290a The effect of a combination of octenidine and 2-phenoxyethanol on wound healing in pigs in-vivo and its in-vitro percutaneous permeation through physiological and barrier disrupted pig skin Stahl J. (1), Braun M. (2), Siebert J. (2), Kietzmann M. (1) A combination of octenidine and 2-phenoxyethanol (octenisept®) is a commonly used disinfectant in human medicine. Since pig skin represents an adequate model for human skin, the effect of the test compound on wound healing was studied in pigs in comparison to another disinfectant based on PVP-iodine (Vetsept®). Furthermore, the in-vitro percutaneous permeation of the test substances was studied through physiological and barrier disrupted pig skin. The effect of the test formulations on wound healing was studied under occlusive conditions in 6 pigs over a time period of 11 days. Wounds were treated daily with 210 µl of the test substances (pure, 1:5, 1:10, and Ringer solution). Wound healing was examined by planimetrical analysis of the wound areas. The percutaneous permeation of octenidine and 2-phenoxyethanol was studied in-vitro in Franz-type diffusion cells with porcine split skin as well as split skin without stratum corneum after tape stripping. For this purpose, 2 ml of the test compound were applied onto a skin surface of 1.77 cm². Samples of the acceptor medium were taken up to 28 hours. The in-vivo experiment shows that the treatment of wounds with the test formulation (up to a dilution of 1:10) compared with PVP-iodine, vehicle treatment and untreated control (occlusive covered and non occlusive covered) has no influence on the healing rate in pigs and does not induce retardation of wound healing. The in-vitro diffusion experiment reveals that after topical application of 2 ml of the test compound no octenidine was found in the acceptor chamber in the physiologically intact pig skin, while 0.4 % of the applied octenidine permeated through the barrier disrupted skin. 2phenoxyethanol permeated through physiological pig skin in amounts of 6.3 % of the applied 2-phenoxyethanol and through barrier disrupted skin in amounts of 25.4 %. 1. Institute of Pharmacology, Toxicology and Pharmacy, University of Veterinary Medicine Hannover, Foundation, Hannover, Germany, 2. Schülke & Mayr GmbH, Norderstedt, Germany

291 Detection of CYP2F1-associated enzyme activities in human lung microsomes Bernauer U. (1), Settels E. (1), Tönnies M. (2), Appel K.-E. (1) Background: CYP2F enzymes are predominantly expressed in the lung. They are involved in the metabolic activation of a variety of industrially important chemicals, such as styrene or naphthalene. In laboratory animals, CYP2F dependent activation has been regarded as a causative step for the development of lung cancer after long-term inhalation exposure to chemicals (e.g. styrene). Up to now it has not been demonstrated, whether CYP2F dependent metabolic activation also takes place in the human lung. This would require the presence of functional active CYP2F enzyme in human lung tissue. To our knowledge, CYP2F1 in human lung tissue has been determined at the mRNA and at the protein level only. As there is no information about the functionally active enzyme it was our aim to establish a methodology which allows the determination and quantification of CYP2F1 associated enzyme activities in human lung tissue. Methods: Human lung tissue was obtained after informed consent and microsomes were prepared. By using microsomes from bovine lung, a HPLC method has been elaborated in order to quantify 3-[N-acetyl-L-cysteine-S-yl) methyl]indol, which is generated after CYP2F-dependent dehydrogenation of 3-methylindol followed by trapping with N-acetyl-cysteine (NAC). After optimization of relevant parameters (e.g. sample purification, incubation time, substrate concentration, chromatographic

61

conditions), human lung microsomes were investigated by using this method. The specificity of this reaction was investigated by applying the CYP2F-specific inhibitor 5phenyl-1-pentyne. Results: After incubation with 3-methylindol and NAC, 3-[N-acetyl-Lcysteine-S-yl) methyl]indol and other metabolites could clearly be identified in human lung microsomes. By using 5-phenyl-1-pentyne as a specific chemical inhibitor, formation of 3-[N-acetyl-L-cysteine-S-yl) methyl]indol was reduced by at least 40%. The results demonstrate for the first time, that human lung has the capability to activate chemicals which represent CYP2F substrates. These findings can contribute to a refinement and improvement of the risk assessment of industrial chemicals which represent CYP2F substrates. However, further studies regarding the activation and inactivation of the respective chemicals are required in order to be able to extrapolate from animal findings to the human situation. 1. Federal Institute for Risk Assessment / Bundesinstitut für Risikobewertung (BfR), Berlin, Germany, 2. Helios Klinikum Emil von Behring, Lungenklinik Heckeshorn, Berlin, Germany

292 Regulation and function of the arylhydrocarbon receptor repressor (AhRR) in vivo Weighardt H. (1), Haarmann-Stemmann T. (2), Korkowski M. (1), Abel J. (2), Förster I. (1) The arylhydrocarbon receptor repressor (AhRR) is a bHLH-PAS transcription factor highly homologous to AhR and Arnt. The AhRR was shown to compete for AhR function by sequestering Arnt. It is presently unclear, however, whether the AhRR exerts an essential antagonistic function in vivo, either under physiological conditions or in response to exogenous AhR activation. In order to get insight in AhRR expression and function we generated AhRR-deficient mice which additionally allow to monitor expression of the AhRR by a fluorescence approach. Therefore we replaced the coding region of the AhRR gene by an enhanced green fluorescence (EGFP)-cassette. Expression analysis in naive AHRRgfp/+ mice revealed constitutive AhRR expression in the skin in both, dermis and epidermis. Whereas in the epidermis MHCII positive Langerhans cells stain positive for the AhRR, in the dermis both dermal MHCII positive DC as well as CD90.2 positive cells, probably dermal fibroblasts, express the AhRR. AhRR expression could also be detected in the lamina propria of the gut. Further, we will analyze the induction of typical AhR responsive genes in AHRR-deficient mice (AhRRgfp/gfp) in the presence or absence of AhR ligands and will study the in vivo responsiveness to xenobiotics. The mice will also be examined for developmental defects and fertility, as the AhR is known to induce estrogen production. 1. Molecular Immunology, Institut für Umweltmedizinische Forschung, University Düsseldorf, Germany, 2. Molecular Toxicology, Institut für Umweltmedizinische Forschung, University Düsseldorf, Germany

293 Arylhydrocarbon receptor activation increases the expression of Bmal1 in HaCaT keratinocytes Pfeiffer R. (1), Tigges J. (1), Hübenthal U. (1), Fritsche E. (1), Abel J. (1) We recently identified the AhR as a primary target for UVB irradiation mediating parts of the cellular UVB stress response through the intracellular generation of 6formylindolo[3,2-b]carbazole FICZ. Because the molecular players of circadian rhythms, the proteins Per, Clock and Bmal1 belong to the PAS protein family, we hypothesize that the ligand-activated AhR might interfere with the circadian machinery through interaction with one of its players. Circadian gene expression is investigated by real time RT-PCR from HaCaT cell lysates harvested every four hours. For a stably reproducible circadian rhythm we established a protocol which synchronizes the circadian rhythm in culture. This was achieved by starvation of the cells for 24 h followed by treatment with 50% serum for 2 h (= serum shock). This serum shock caused a synchronization of the circadian rhythm of HaCaT cells for up to 72 h. Exposure of these serum shocked cells to 100 nM FICZ in different phases of the rhythm (24 and 36 h after serum shock) caused an increased expression of Bmal1. This gene induction is phase dependent with the largest effect seen when treated 36 h after serum shock. Because we did not observe Bmal1 induction after FICZ exposure in AhR deficient HaCaT cells, we assume that this effect is AhR dependent. Compared to FICZ 3-methylcholanthrene shows a different pattern of influence on the Bmal1 expression because of a prolonged activation of the Ah receptor. AhR activation increases Bmal1, Cyp1A1 and Cox2 but not Per1, Per2 or Clock expression. Proteasome inhibitors (Mg 132 and PSI) act synergistic on Bmal1 and Cox2 expression and inhibit the constitutive expression of Cyp1A1. Cycloheximide treatment enhances constitutively the Per1 expression, acts synergistic on Cox2 expression and superinduces the Cyp1A1 expression. Phosphorylation (scr kinase) is required for constitutive and induceable expression of Bmal1 and Cox2. These inhibitor studies suggest of a complex postranscritpional regulation of the circadian rhythm by the AhR. The molecular mechanisms causing Bmal1 induction after FICZ exposure in HaCaT cells are currently elucidated. 1. Institute of preventive medicine, Heinrich Heine University, Düsseldorf, Germany

294 The role of soluble epoxide hydrolase in protein prenylation Mitterecker F. (1), Arand M. (1), Cronin A. (1) Mammalian soluble epoxide hydrolase (sEH, EC 3.1.3.76; EC 3.3.2.10) is a homodimeric enzyme that combines two catalytic activities in each monomer. While the C-terminal domain functions as epoxide hydrolase, its smaller N-terminal domain belongs to a distinct family of haloacid dehalogenases (HAD) with novel phosphatase activity. Most of the established sEH functions are currently assigned to the epoxide hydrolase activity. The main function of sEH is the metabolism of endogenous arachidonic acid derived signalling molecules like epoxyeicosatrienoic acids (EETs) to the corresponding diols (DHETs). EETs have diverse physiological functions, which is why the sEH has evolved as a target for the therapy of cardiovascular and inflammatory diseases. Recently EETs have also been shown to promote cell proliferation and angiogenesis, which represents yet another branch of EET actions. The sEH phosphatase so far has been shown to accept generic substrates (4-

nitrophenylphosphate), some hydroxy lipid phosphates, as well as isoprenoid phosphates. Isoprenoid phosphates such as farnesyl and geranylgeranyl pyrophosphates are intermediates of the cholesterol biosynthesis pathway. They are also substrates for post translational modification of proteins by lipids. Isoprenylation is a critical process in the correct membrane association and activation of many cellular signalling proteins regulating cell proliferation and survival, such as the small GTPases Rho and Ras, which often play a role in malignant transformation. Presently we are investigating the influence of the sEH phosphatase activity on protein isoprenylation in mammalian V79 and COS-7 cells, which were stably transfected with expression constructs for either wild type sEH or inactive mutants of the sEH phosphatase and/or epoxide hydrolase domain. Generated cell lines were examined for the subcellular localization of sEH proteins by immunofluorescence analysis and intermediate metabolites of the prenylation pathway were analysed by LC/MS-MS. At present we are investigating alterations in the general protein prenylation pattern as well as modifications of specific isoprenylated protein targets (Ras, Rho) by immunocytochemistry and western blot. The physiological role of the sEH phosphatase activity remains uncertain to date, but our investigation will answer the question whether the sEH phosphatase is of biological relevance for post translational protein prenylation. 1. Dept.of Pharmacology and Toxicology, University of Zurich, Zurich, Switzerland

295 Mammalian soluble epoxide hydrolase efficiently hydrolyzes hepoxilins Cronin A. (1), Arand M. (1) The bifunctional human soluble epoxide hydrolase (sEH, EC 3.1.3.76; EC 3.3.2.10) is besides its function as xenobiotic metabolizing enzyme - further implicated in diverse physiological processes due to the brakedown of arachidonic acid derived signalling molecules like epoxyeicosatrienoic acids (EETs) by its epoxide hydrolase domain. Both cis- and trans-epoxy fatty acids have been shown to be sEH substrates, although most enzymatic derived endogenous lipid epoxides are of cis-configuration. Here, we report that hepoxilin A3 and B3 (HxA3 and HxB3), which represent endogenous trans-hydroxyepoxy lipids, are efficiently converted to the corresponding trihydroxy metabolites (trioxilins) by mammalian sEH. The effect of purified sEH on hepoxilin metabolism was determined using LC-MS/MS analysis, followed by kinetic evaluation. HxA3 is hydrolysed by rat sEH with a Vmax of 1.8 mmol/mg/min and a Km of 8 mM, which is in the activity range of the best endogenous substrates for mammalian sEH. Both HxA3 and HxB3 are substrates for rat and human sEH. This finding is of interest as hepoxilins have been shown to play a role in inflammatory events in neutrophils and in the skin, as well as in calcium mobilization, bronchoconstriction, and insulin signalling. These lipid mediators are formed in various organs like liver, brain and skin and rapidly transformed to trioxilins which are regarded degradation products. The mammalian hepoxilin epoxide hydrolase was - although only incompletely characterized – suggested to be distinct from mammalian sEH. However, sEH turns over hepoxilins orders of magnitudes more efficiently then partly purified Hepoxilin epoxide hydrolase (reported specific activity: 0.2 nmol/mg/min). Further, sEH is expressed in hepoxilin producing tissue. Our results suggest a biological relevance of the sEH - rather than Hepoxilin epoxide hydrolase - in hepoxilin metabolism, which opens a new branch in the numerous physiological functions of sEH. Only a throughout comparison of mammalian sEH and Hepoxilin epoxide hydrolase regarding substrate and inhibitor profile and expression pattern will distinguish the physiological role of these two enzymes. 1. Dept.of Pharmacology and Toxicology, University of Zurich, Zurich, Switzerlan

296 The novel epoxide hydrolase EH3 is affected by AUDA - a potent soluble epoxide hydrolase inhibitor Decker M.M.E. (1), Adamska M. (1), Cronin A. (1), Arand M. (1) Soluble epoxide hydrolase was originally found in the liver and thus thought to be involved in the metabolism of xenobiotic compounds, but is now known to be expressed in various other organs and tissues, including vasculature. Besides xenobiotic epoxides sEH likewise hydrolyses endogenous epoxides, mainly epoxyeicosatrienoic acids (EETs), to their mostly less active metabolites, the corresponding dihydroxyeicosatrienoic acids (DHETs). EETs are Cytochrom P450 derived epoxides of arachidonic acid that exhibit various autocrine and paracrine functions in the cardiovascular and renal circulation. The most intensively studied functions of EETs concern their activity as modulators of blood pressure and vasodilatatory effects, regulated by sEH. Breakdown of sEH activity by either administration of specific sEH inhibitors (sEHI) or gene deletion has been shown to decrease blood pressure in spontaneously hypertensive rats and male knock out mice. Hence the development of specific sEHIs is a promising therapeutic approach towards the treatment of hypertension and other cardiovascular diseases, by increasing the EET/DHET ratio and therefore promoting the beneficial effect of EETs. Here we report the inhibitory effect of 12-(3-adamantan-yl-ureido)dodecanoic acid (AUDA) on the enzymatic activity of EH3, a novel human epoxide hydrolase which recently has been identified by our group. AUDA is a urea derived sEHI and has been shown to lower blood pressure and to increase the urinary EET/DHET ratio in rats with salt-sensitive hypertension. We could show a concentration dependent inhibition of EH3 - recombinantly produced in a Sf9 baculovirus expression system - enzymatic activity towards three different EET regioisomers. Preliminary results indicate that AUDA efficiently inhibits enzymatic activity of EH3 with an estimated IC50 value of 50 to100 nM. Our results show that the potent sEHI AUDA is not specific but rather affects other epoxide hydrolases - expressed in non-target tissue in the potential therapeutic range. This may mediate additional or undesired effects of sEHIs and demonstrates the difficulty in the use and development of pharmacological inhibitors. 1. Dept. of Pharmacology and Toxicology, University of Zurich, Zurich, Switzerland

62

297

300

Expression of major phase 1 enzymes in HaCaT cells Hennen J. (1), Kalmes M. (1), Blömeke B. (1) A delayed-type hypersensitivity reaction induced by small reactive chemicals may be mediated by metabolic conversion of the compounds. Skin cells especially keratinocytes play a major role in this process. A total replacement of animal experiments for the endpoint skin sensitization will only be possible when we have well characterized in vitro test systems developed. Thus, it is necessary to understand the metabolic capacities of relevant cells under in vitro conditions. This knowledge will help elucidating whether in vitro metabolism adequately represent human skin metabolism in predicting toxicity. Thus, we evaluated the human keratinocyte cell line HaCaT regarding their metabolic potentials. In this study we studied the expression of major cytochrome P450 (CYP) enzymes by RT-PCR and activities were evaluated for CYP1A1. Interestingly, HaCaT showed comparable basal activities as found for primary keratinocytes, and chemicalinduced enzyme activities were even slightly higher compared to in vitro cultured primary keratinocytes (1.5-3.0 fold). These findings demonstrate that a permanent epithelial cell line from adult human skin exhibits CYP expression and CYP1A1 enzyme activities. In sum, HaCaT cells provide a promising tool for studying metabolism of common chemicals whose exposure may occur via skin. 1. Department of Environmental Toxicology, University Trier, Trier, Germany

Glutathione S-transferase isoforms expressed in human intestinal Caco-2 cells during differentiation Scharmach E. (1), Hessel S. (1), Lampen A. (1) The human colon adenocarcinoma cell line Caco-2 is frequently used for studying human intestinal metabolism and transport of xenobiotica. Previous studies have shown that both Caco-2 cells and human colon cells constitutively express the multigene family of detoxifying enzymes glutathione S-transferases (GSTs), particularly GST alpha and GST pi. The gut fermentation product butyrate can modulate the potential of detoxification. The aim of this study was to investigate the basal expression of further cytosolic GSTs in Caco-2 cells during cell differentiation compared with expression levels in MCF-7 and HepG2 cells and to determine the butyrate-mediated modulation of gene and protein expression by real-time PCR and western blot analysis. In Caco-2 cells, mRNA and protein expression levels of GST alpha increased during cell differentiation. GSTO1 and GSTP1 were constantly high expressed. Expression of GSTM5 and GSTT1 were not detectable. HepG2 cells expressed GSTO1 and MCF-7 cells GSTZ1 most intensive. GSTA5, GSTM5 and GSTP1 expressions were not detected in both cell lines. Incubation of Caco-2 cells with butyrate (5 mM) significantly induced GSTA1 and GSTM2 in proliferating Caco-2 cells. In differentiated cells, butyrate tended to increase GSTO1 and GSTP1. The results of this study suggest that a differentiation-dependent expression of GSTs in human Caco-2 cells may reflect the in vivo situation and indicate the potential of butyrate to modify the intestinal metabolism. 1. Department of Food Safety, Federal Institute of Risk Assessment, Berlin, Germany

298 N-Acetyltransferase 1 enzyme activities in subtypes of HaCaT cells compared to primary keratinocytes Bonifas J. (1), Scheitza S. (1), Blömeke B. (1) Testing the sensitizing potential is an essential aspect in toxicological risk assessment of substances. In light of recent regulatory changes, namely 7th Amendment to the Cosmetic Directive and REACH, non-animal approaches are required as alternative in vitro safety testing methods. Keratinocytes participate in dendritic cell activation by danger signalling as well as via biotransformation of non-reactive chemicals. In order to avoid donor dependencies and limitations caused by cell isolation, the replacement of human primary keratinocytes by a keratinocyte cell line may be one adaption to improve such model systems. In the context of this background we analyzed primary cultures of neonatal human keratinocytes (NHEK) from different donors and HaCaT cells obtained from three different sources regarding their N-acetylation capacities. We analyzed the cells for the presence of promoter-specific NAT-1 mRNAs, estimated enzyme activities using para-Aminobenzoic acid (PABA) as a classical NAT-1 substrate, paraphenylenediamine (PPD), a contact allergen which is known to be acetylated by primary keratinocytes, and 2-aminofluorene (AF) which is known to be acetylated by both NAT isozymes. As expected, acetylation activities varied between the different NAT-1 substrates, whereas N-acetylation activities were highest for PABA acetylation and lowest for AF. Regarding subtypes we found that PABA-acetylation activities (per min and mg total protein) for HaCaT (type A) cell lysates were obviously much higher (47.1±2.1) than those obtained for HaCaT type B and C with 17.8±1.3 and 13.1±0.4, respectively. Subtype A showed very effective acetylation; in this type PABA-acetylation was actually higher than PABA-acetylation of primary keratinocytes (7.3±0.3). Overall, measured acetylation activities in HaCaT cells compared favorably with primary keratinocytes. These data demonstrate that the HaCaT cell model represents a suitable in vitro assay for addressing skin related exposures to aromatic amines. 1. Dept. of Environmental Toxicology, University Trier, Trier, Germany

299 Definite identification of N-acetyltransferase 2 genotype Agundez J.A.G. (1), Golka K. (2), Martinez C. (1), Selinski S. (3), Blaszkewicz M. (2), Garcia-Martin E. (4) Arylamine N-acetyltransferase 2 (CoASAc; NAT2, EC 2.3.1.5) is a drug metabolizing enzyme that displays common polymorphisms leading to impaired drug metabolism and adverse drug effects. Determination of the NAT2 genotype in clinical practice is hampered by the occurrence of ambiguous haplotype combinations that may lead to patient misclassification. We determined the frequencies for ambiguous NAT2 haplotypes and diplotypes in a Caucasian population, and the capacity of PHASE v2.1.1 to clarify this ambiguity and to classify individuals according to their acetylation status. By means of allele-specific haplotype mapping and sequencing, we determined the haplotypes for seven common single nucleotide polymorphisms in the NAT2 gene (n=2,624 haplotypes). To test the performance of PHASE, actual genotypes were deconstructed and then reconstructed again by haplotype prediction. We identified 21 NAT2 allelic variants, including a new variant allele that combines the SNPs rs1801279 rs1799929 and rs1208. In contrast, the previously described variant alleles *5G, *5J, *6E, *7A, *11A, *11B and *14B were not identified in the population study. Ambiguous haplotypes were observed in 98 alleles (3.7%) and ambiguous diplotypes were observed in 64 individuals (4.9%). Eleven individuals (0.8%) were misclassified by the use of haplotype prediction. Ambiguous NAT2 genotyping data are common. Actual NAT2 genotypes cannot be fully determined by haplotype prediction techniques though the reconstruction was correct for >99% of all individuals. This study provides real haplotype data that can be used as a guide to convert NAT2 haplotypes and diplotypes into actual genotypes in Caucasians. 1. Department of Pharmacology, Medical School, University of Extremadura, Badajoz, Spain, 2. Leibniz Research Centre for Working Environment and Human Factors (IfADo) at the TU Dortmund University, Dortmund, Germany, 3. Faculty of Statistics, TU Dortmund University, Dortmund, Germany, 4. Department of Biochemistry & Molecular Biology, School of Biological Sciences, University of Extremadura, Badajoz, Spain

301 The Wnt/β-catenin signaling pathway contributes to the regulation of expression of glutathione S-transferases Braeuning A. (1), Giera S. (1), Köhle C. (1), Buchmann A. (1), Schwarz M. (1) Enzymes of the so-called phase II of drug metabolism, such as the glutathione Stransferases (GSTs), play a decisive role in the defense against electrophilic compounds and oxidative stress. Transcription of various GST isoforms, particularly of members of the GSTa and GSTm subfamilies, is controlled by different transcription factors such as the antioxidant-activated Nrf2 or the constitutive androstane receptor CAR. Here we show that the Wnt/β-catenin pathway is also involved in the transcriptional regulation of GSTs. This finding is based on several lines of evidence: (i) GSTs are overexpressed in mouse liver tumors with activating mutations in the Ctnnb1 gene encoding β‑catenin. Inversely, GST expression is inhibited upon Ras activation. (ii) GSTs are overexpressed in hepatocytes from transgenic mice with liver-specific expression of an activated version of β-catenin. (iii) GST expression is markedly reduced in mice with liver-specific knockout of Ctnnb1. (iv) Wnt3A-mediated activation of β-catenin-dependent signaling in primary hepatocytes stimulates GST expression. (v) Activation of β-catenin signaling in mouse hepatoma cells activates a luciferase reporter driven by a fragment of the 5’regulatory sequence of the mouse GSTm3 gene via a retinoid-X-receptor binding site. In synopsis with previous results describing the ability of β-catenin signaling to induce the expression of several cytochrome P450 isoforms, it now appears that the Wnt/β-catenin pathway functions as a master regulator of the expression of both phase I and phase II drug-metabolizing enzymes in perivenous hepatocytes from mouse liver. 1. Inst. of Pharmacology and Toxicology, University of Tübingen, Tübingen, Germany

302 In vitro stability of methylmethacrylic acid, TEGDMA and HEMA exposed to esterases Track N. (1), Seiss M. (1,2), Hickel R. (2), Reichl F.X. (1,2) Objectives: Comonomers used in dental restorative materials, e.g. triethylenglycoledimethacrylate (TEGDMA), hydroxyethylmethacrylate (HEMA) and methylmethacylic acid (MMA) are methacrylic acid esters. Because of the chemical structure of these esters in earlier studies two different pathways were suggested. The present study was performed in order to analyze which of these chemical pathways are possible in vitro: a) saponification of TEGDMA, HEMA, and MMA leading to free methacrylic acid (MA) as an ionic intermediate or b) enzymatically epoxidation leading to lipophilic intermediates. Methods: Experiments have been performed in an isolated system with pig liver esterase (PLE) in phosphate buffer using MMA, TEGDMA or HEMA, respectively. The reaction of tested comonomers was terminated by the use of sodium hydroxide solution, sulfuric acid or saturated NaCl solution, respectively. The samples were analyzed by use of headspace gas chromatography-mass spectrometry. Results: In all samples a decomposition of comonomers by the use of PLE was observed. Due to the high rate constant k of MMA (e.g. kMMA > 1.3 mg/(l sec)) epoxidation of non-cleaved molecules of this substance can be excluded. Compared to MMA the decomposition of TEGDMA and HEMA is significantly slower (e.g. k(TEGDMA, PLE = 7 units/ml) = 0.004 mg/(l sec) and k(HEMA, PLE = 7 units/ml) = 0.00013 mg/(l sec)). Thus in case of a low liver enzyme concentration the epoxidation of non-cleaved molecules of TEGDMA and HEMA can not be excluded. Significance: In the metabolic pathway of TEGDMA and HEMA the probability of an auxiliary chemical pathway was demonstrated. In case of MMA the formation of epoxidated metabolites can be excluded. Thus the chemical pathway for TEGDMA and HEMA might lead to lipophilic intermediates which can be saved in fatty tissue and may cause long lasting effects. 1. Walther-Straub-Institute of Pharmacology and Toxicology, Ludwig-MaximiliansUniversity of Munich, Munich, Germany, 2. Department of Operative Dentistry and Periodontology, Ludwig-Maximilians-University of Munich, Munich, Germany

63

303 Metabolism of zearalenone in precision-cut rat liver slices Damm G. (1), Pfeiffer E. (1), Metzler M. (1) Zearalenone (ZEN) is a mycotoxin formed by several Fusarium species infesting mostly corn but also other grains, crops and feeding stuff. Until recently, the only reported metabolites of ZEN resulted from reduction of the 7-keto group and 11,12-double bond. Our laboratory has reported that rat liver microsomes biotransform ZEN to several monohydroxylated metabolites of ZEN and alpha-zearalenol (α-ZEL). Two of the major oxidative ZEN metabolites have now been identified as 15-hydroxy-ZEN and 15hydroxy-α-ZEL. These catechols are chemically unstable but can be easily methylated to monomethyl ethers by cytosolic catechol-O-methyl transferase. The aim of the present study was to investigate the oxidative metabolism of ZEN, in particularly the formation of catechols and their methyl ethers, under in vivo-like conditions in precisioncut tissue slices from rat liver. Liver slices contain all cells of the liver in their natural tissue architecture and are therefore able to conduct phase I, II and III metabolism. OH

O

OH

16

HO

O

16

15

1

O

6

9

11 10

15

5

12

13

4

3

Zearalenone

HO

7 8

O

1

O

13

4

3 5

12

6

9

11 10

alpha-Zearalenol

for detection of NAT1 activity [4]. The enzyme, predominantly found in extrahepatic tissues, can also catalyze the acetylation of p-phenylenediamine (PPD), an important component in hair dyes [5]. In this study, constitutive NAT1 activity was determined in two different cell culture systems, i.e. the human urothelial cell line 5637 and primary cultured porcine urinary bladder epithelial cells (PUBEC). The acetylation of PABA and PPD was determined by HPLC analysis in cell culture extracts upon incubation with various substrate levels for up to 60 min. Compared to the 5637 cell line acetylation rates were higher in the primary PUBEC for both compounds, based on PABA conversion to its acetylated metabolite and formation of monoacetylated (MA-PPD) and diacetylated (DA-PPD) PPD. The acetylation rates were lower in 5637 cells compared to PUBEC, but the human cells produced MA-PPD and DA-PPD in sufficient amounts. Thus, we conclude that 5637 cells represent an adequate human-derived cell culture model which can be used for in vitro studies on the metabolism of xenobiotics which are detoxified by N-acetylation such as PPD. [1] Alexander et al., 2008, Toxicol Environ Health A 71:915-922, [2] Degen et al., 2004, Toxicol Lett 151:135-162, [3] Roos et al., 2006, Arch Toxicol 80:45-52, [4] Kawamura A et al., 2005, Biochem Pharmacol 69:347359, [5] Bolt & Golka, 2007, Crit Rev Toxicol 37: 521-536 1. Leibniz Research Centre for Working Environment and Human Factors (IfADo), TU Dortmund, Dortmund, Germany

7 8

OH

Precision-cut tissue slices were prepared from the livers of male Sprague-Dawley rats using a VITRON tissue slicer. Single slices of 200 µm thickness were incubated in 1.7 mL tissue culture medium containing 200 µM ZEN under an oxygen-containing atmosphere at 37°C for 24 h. The incubation media were extracted before and after cleavage with beta-glucuronidase and/or arylsulfatase and extracts analyzed by HPLC and GC-MS. The incubation media were also analyzed directly with LC-MS. The major oxidative ZEN metabolites formed with rat liver microsomes were also detected with rat liver slices. Metabolites were mainly present as glucuronides and sulfates together with small amounts of unconjugated material. One of the main metabolites of liver slices was identified as monomethyl ether of the catechol 15-hydroxy-ZEN, predominantly present as a sulfate conjugate. The reduced form of this metabolite, i.e. 15-hydroxy-α-ZEL, was also found as monomethyl ether, which was present in equal proportions as glucuronide and sulfate. Our results demonstrate that oxidative metabolism of ZEN, especially formation of catechols, occurs in liver slices under in vivo-like conditions. This finding strongly suggests that catechols of ZEN and α-ZEL are also formed in vivo, which may be of toxicological significance. Supported by Deutsche Forschungsgemeinschaft (Grant ME 574/32-1) and “Food and Health” of KIT. (1) Karlsruhe Institute of Technology (KIT), Chair of Food Chemistry, Karlsruhe, Germany

304 Catechol formation is a major metabolic pathway of the growth promotor αzearalanol Hildebrand A. (1), Schnattinger C. (1), Pfeiffer E. (1), Metzler M. (1) α-Zearalanol (α-ZAL) is extensively used as an estrogenic growth promotor (Zeranol, Ralgro®) in several non-EU countries, but little is known about its metabolic fate. We have therefore studied the metabolism of α-ZAL in human liver microsomes (HLM) and in hepatic microsomes from several other species. When HLM were incubated with 100 µM α-ZAL in the presence of an NADPH-regenerating system at 37°C for 40 min, the formation of zearalanone (ZAN) and of two major monohydroxylation products of α-ZAL and ZAN was observed by LC-MS. When ZAN was incubated with HLM under the same conditions, the same pattern of metabolites was found as with α-ZAL, although the amounts of the individual metabolites were different. In order to probe for the formation of catechols, α-ZAL was specifically labeled with deuterium at the two hydrogen-carrying positions of the aromatic ring. Each of the two major monohydroxylated metabolites of this 13,15-D2-α-ZAL had lost one deuterium according to LC-MS analysis, providing unambiguous evidence for the formation of the catechols 13-hydroxy-α-ZAL and 15hydroxy-α-ZAL. The same loss of deuterium was observed for the two major hydroxylation products of ZAN. When supersomes containing the recombinant human cytochrome P450 (CYP) isoforms 1A1, 1A2, 1B1, 2A6, 2B6, 2C8, 2C9, 2C19, 2D6 and 3A4 were incubated with α-ZAL, CYP1A2 had by far the highest activity (49.3 ± 3.0 pmol min-1 pmol CYP-1) and generated exclusively the two catechols of α-ZAL. CYP3A4 (5.8 ± 0.8), 2C19 (0.58 ± 0.29), 2D6 (0.26 ± 0.02) and 1B1 (0.20 ± 0.08) gave rise to the formation of catechols and several minor metabolites. A very similar metabolic pattern as observed with HLM was obtained when bovine, porcine and rat liver microsomes were incubated with α-ZAL or ZAN. This study shows for the first time that the growth promotor α-ZAL and its metabolite ZAN are prone to hydroxylation at the aromatic ring. Catechol formation represents the major pathway in the oxidative metabolism of α-ZAL or ZAN in human as well as other mammalian species in vitro. As catechol metabolites are often associated with adverse effects, the toxicological sequelae of this pathway and its relevance for the in vivo disposition of α-ZAL should now be studied. Supported by Deutsche Forschungsgemeinschaft (Grant ME 574/32-1) and „Food and Health“ of KIT. 1. Chair of Food Chemistry, Karlsruhe Institute of Technology (KIT), Karlsruhe, Germany

306 MRPs, but not BCRP, transport the glutathione conjugates of the ultimate carcinogenic metabolite of benzo[a]pyrene Hessel S. (1), John A. (2), Seidel A. (2), Lampen A. (1) The human gastrointestinal tract represents the main portal for the entry of xenobiotics. The epithelium of the small intestine plays an important role in the detoxification of xenobiotics due to the expression of a number of phase I and phase II xenobioticmetabolizing enzymes (XMEs) as well as several transport proteins of the ATP-binding cassette (ABC) superfamily. In the present study, differentiated human intestinal Caco-2 cells were used as a model for the human small intestine to investigate the metabolism and transport of the pro-carcinogenic food-contaminant benzo[a]pyrene (B[a]P). Previous studies revealed that the B[a]P phenols are excreted as B[a]P phase II metabolites including B[a]P-glucuronides and B[a]P-sulfates is mediated by breast cancer resistance protein (BCRP) (Ebert et al., 2005). In contrast, the detoxification of the ultimate carcinogenic phase I B[a]P metabolite anti-B[a]P-7,8-diol-9,10-epoxide (BPDE) can occur only as glutathione conjugates formed by glutathione-S-transferases (GSTs). This study demonstrates that these glutathione conjugates of BPDE are excreted overall to the basolateral side of the polarized Caco-2 monolayer. Furthermore, it was shown by induction and inhibition studies that the multidrug resistance-associated proteins (MRPs), but not BCRP, are involved in this transport of the BPDE glutathione conjugates. The transport rate is increased by pre-treatment with butyrate, oltipraz and quercetin, which stimulate GST and transport protein expression. The specific BCRP inhibitor Ko143 had no influence on the transport of the glutathione conjugates at all, but a complete inhibition of the transport was achieved by using the unspecific MRP inhibitor MK571, indicating that MRPs are involved in the basolateral and apical transport of the glutathione conjugates of BPDE. Specific MRP inhibitors are not available so far. To identify the specific MRPs that are responsible for the excretion of the glutathione conjugates of BPDE the generation of stable MRP knockdown clones of Caco 2 cells are under way for future studies. 1. Department of Food Safety, Federal Institute for Risk Assessment, Berlin, Germany, 2. Biochemical Institute for Environmental Carcinogens, Prof. Dr. Gernot Grimmer Foundation, Grosshansdorf, Germany

307 Effects of in vitro metabolism on the estrogen activity of DES derivates in yeastbased reporter gene assay Woitkowiak C. (1), Böhn S.N. (1), Bernshausen T. (1), Fabian E. (1), Huener H.-A. (1), Murr A. (1), Verlohner A. (1), van Ravenzwaay B. (1), Landsiedel R. (1) Test substances can be screened for potential endocrine activities using in vitro assays such as the yeast-based reporter gene assays. The xenobiotic metabolic capacity of yeast cells, however, is not well investigated. Using DES derivates as model compounds we evaluated whether rat liver microsomes could compensate this deficiency of the YES assay (yeast estrogen screen). DME-DES (the non-estrogenic dimethylether of the estrogenic diethylstilbestrol) and DES-DP (DES-dipropionate) were extensively metabolized by microsomes under the chosen incubation conditions. Incubations of the DES derivates with active microsomes (AI) as well as heat-desactivated microsomes (HDC) were transferred into the YES assay. In the YES assay DME-DES and its HDC showed no estrogenic activity, whereas its AI showed an estrogenic response, demonstrating that the activation of the estrogenic activity of DME-DES can be detected with this in vitro system. In contrast, DES-DP, its HDC as well as its AI showed estrogenic activity in the YES assay. However, LC/MS-analysis of the medium of the YES assay, incubated with DES-DP resulted in the detection of DES, but not of DESDP. Therewith, it could be demonstrated that the esterases of yeast cells are able to saponify DES-DP forming estrogenic DES. In conclusion, our investigations demonstrated a successful integration of metabolic capacity in the YES assay, allowing a more precise assessment of the estrogen potential under special respect of their potential metabolic activation or deactivation. 1. BASF SE, Experimental Toxicology and Ecology, 67056 Ludwigshafen, Germany

305 Acetylation of p-aminobenzoic acid and the hair dye component pphenylenediamine by N-acetyltransferase 1 in a human urothelial cell line and in primary porcine urinary bladder epithelial cells Föllmann W. (1), Blaszkewicz M. (1), Degen G.H. (1), Behm C. (1), Golka K. (1) It has been shown in previous studies that the bladder epithelium is competent in xenobiotic metabolism [1-3]. In this context, acetylation by N-acetyltransferases is regarded as detoxification reaction since it decreases or prevents bioactivation of carcinogenic arylamines via N-hydroxylation. Cell culture models used to study the process of detoxification in vitro have to be characterized first for their competence in Nacetylation. For this purpose, p-aminobenzoic acid (PABA) can be used as a substrate

308 Intracellular metabolism of 17β-estradiol in cultured transgenic V79 cells: metabolite profile and impact on cellular stress response Zettner M.A. (1), Freudenberger J. (1), Jäger S. (1), Honig D. (1), Lehmann L. (1) The female sex hormone 17β-estradiol (E2) is associated with the incidence of breast cancer. Formation of the E2-catechols 2- and 4-hydroxy (2- and 4-HO)-E2 by breast specific cytochrome P450-dependent monooxygenase (CYP) isoenzymes 1A1 and 1B1 may result in genotoxic stress if detoxification e.g. by catechol-O-methyltransferase (COMT) is insufficient. Therefore the metabolite profile and the influence of E2 on

64

cellular genotoxic stress in cultured transgenic V79-fibroblasts with specific expression of either human (h) CYP1A1 or 1B1-isoenzymes in comparison to the CYP-free parent cell line (PC) was investigated. Metabolite profile was determined by means of GC/MS/MS. PC and cell lines expressing hCYP were characterized with respect to 7ethoxyresorufin-O-deethylase (EROD) and COMT activity revealing no EROD activity in PC. In contrast, COMT activity was detectable in all cell lines. Intracellular glutathione (GSH) level is a sensitive indicator for genotoxic stress since GSH biosynthesis is mediated by various stress signal depending transcription factors. Thus, intracellular GSH level was determined in all cell lines after treatment with E2 by means of flow cytometry. In PC, treatment with E2 resulted neither in free nor in methylated metabolites. In contrast, the formation of 2-HO-E2 (hCYP1A1) and 2-, 4- and 16-HO-E2 (hCYP1B1) as well as the corresponding methylated catechol estrogens was observed. The amount of methylation products exceeded that of free catechols. A significant increase of the GSH level in a dose dependent manner was only observed in hCYPexpressing cells with concentrations resulting in no or slight reduction of viable cells. The metabolism of E2 by breast specific CYPs generates catechol estrogens. Despite extensive deactivation by COMT a significant cellular stress response is induced. Supported by Deutsche Forschungsgemeinschaft (DFG) Le13/29-7. 1. Institute of Applied Biosciences, Section of Food Chemistry, University of Karlsruhe, Germany

309 Quantitative determination of main components in eluates from polymerized resin-based dental restorative materials by use of GC/MS Seiss M. (1,2), Langer C. (1), Hickel R. (2), Reichl F.X. (1,2) This study investigated the leaching of ingredients from several commercial dental composite resins cured with LED, and immersed in methanol or water for 24 h, respectively. The composites used were: Admira Dentin (VOCO), Artemis Enamel (Ivoclar Vivadent), els extra low shrinkage (Saremco Dental), Filtek Supreme XT Dentin (3M ESPE), Gradia Direct (GC), Venus & Venus flow (Heraeus Kulzer), and XRV Herculite Prodigy Enamel (Kerr). From each dental composite 8 specimens with defined structure and 100 mg net weight were made. After the polymerization process according to manufacturers’ instructions, 4 of the specimens were immersed in 1 ml water or 1 ml methanol, respectively and incubated at 37 °C for 24 h in an incubator. Eluted ingredients triethyleneglycoldimethacrylate (TEGDMA), 2,6-di-tert-butyl-4-methylphenol (BHT), and 4-N,N-dimethylaminobenzoicacidethylester (DMABEE) were detected and quantified using gas chromatography-mass spectrometry (GC-MS). The amounts of detected analytes from 100 mg polymerized composites ranged between the following values: TEGDMA: 0-0.51±0.14 mg (water), 0-1.6±0.2 mg (methanol); BHT: not found in water, 0-0.11±0.08 mg (methanol); and DMABEE: 0-0.11±0.06 mg (water), 0-1.4±0.7 mg (methanol). The elution rates into methanol and water differ singificantly. Furthermore it is concluded that even the highest determined amounts eluting from the composites are significantly below toxic relevant concentrations. 1. Walther-Straub-Institute of Pharmacology and Toxicology, Ludwig-MaximiliansUniversity of Munich, Munich, Germany, 2. Department of Operative Dentistry and Periodontology, Ludwig-Maximilians-University of Munich, Munich, Germany

310 Metabolism of ethylene oxide in microsomes and cytosol from livers and lungs of B6C3F1 mice, Fischer 344 rats, and humans Li Q. (1), Csanády Gy.A. (1), Stangl M. (2), Klein D. (3), Filser J.G. (1) The gaseous compound ethylene oxide (EO) is an important industrial chemical used as fumigant and as a precursor for the production of ethylene glycol. The epoxide forms hydroxyethyl adducts with macromolecules such as hemoglobin and DNA and is mutagenic in vivo and in vitro and tumorigenic in experimental animals. Metabolism of EO is assumed to be catalyzed by epoxide hydrolase (EH) and by glutathione Stransferase (GST). The aim of the present work was to investigate the metabolism of EO in microsomes and cytosol from livers and lungs of mice, rats and humans. Pooled human liver microsomes and cytosol were obtained from Becton Dickinson (Heidelberg). All EO incubations were carried out at 37°C in closed headspace vials. The protein concentrations were 5 and 3 mg/ml in microsomal and cytosolic incubations, respectively. To the cytosolic incubations, the coenzyme glutathione was added (15 mmol/l). In addition, some incubations were carried out with heat-inactivated microsomes and cytosol. EO was monitored in the headspace using GC/FID.In rat and mouse liver microsomes, the EH activity was comparable with spontaneous hydrolysis (0.001 [1/min]). In human liver microsomes, EH activities exceeded 3-5 times the rate of the spontaneous hydrolysis. In contrast, microsomal EH from rodent livers hydrolyzed propylene oxide (a homolog of EO) about 50 times faster than EO. In rat and mouse lungs no microsomal EH activity was detectable. Consequently, EO is a very poor substrate for microsomal EH, if any. In all incubations with cytosol, EO disappeared from the gas phase according to first order kinetics. The GST activity [nmol/min/mg] measured at 100 ppm EO was 8.9 (mouse), 2.0 (rat) and 0.21 (human). In lung cytosol, GST activity was substantially lower, 1.9 (mouse) and 0.30 (rat). One has to conclude that the major pathway for EO elimination is mediated by GST.The in-vitro derived rate constants will be used to improve a physiological toxicokinetic model in order to describe the toxicokinetics of ET and EO in mice, rats and humans. Financially supported by the American Chemistry Council. 1. Institut für Toxikologie, Helmholtz Zentrum München, Neuherberg, Deutschland, 2. Chirurgische Klinik und Poliklinik der TU-München, Klinikum rechts der Isar, München, Deutschland, 3. Institut für Toxikologie und Umwelthygiene, Technische Universität München, Deutschland

311 Toxicokinetics of inhaled ethylene and ethylene oxide in mice Artati A. (1), Kessler W. (1), Richter N. (1), Pütz C. (1), Filser J.G. (1) Ethylene (ET) is ubiquitous in the environment and is a most important industrial chemical. In mammals, ET is metabolized to ethylene oxide (EO), an epoxide that was tumorigenic in rats and mice. ET, however, was negative in a carcinogenicity study in rats. From toxicokinetic data, it became evident that the EO burden in the ET exposed rats had been too low to lead to an increased tumor incidence. In mice, no long-term study was conducted with ET. There is also no toxicokinetic data on ET and limited data on EO in this species. In the present study, toxicokinetics of ET and EO were investigated in male B6C3F1 mice by means of gas uptake studies. Initial ET and EO concentrations in closed exposure chambers ranged from 1 to 10000 and 1 to 3000 ppm, respectively. In addition, exhaled EO was measured in chamber air during constant ET exposures between 1 and 10000 ppm. Concentration-time courses were monitored by gas chromatography. Toxicokinetic data analysis was carried out using a two-compartment model. ET metabolism followed saturation kinetics, Vmax being 15 µmol/h/kg. Vmax/2 was reached at 105 ppm ET. The thermodynamic partition coefficient ET-in-animal / ET-in-air was 0.7. Metabolism of EO followed first-order kinetics within an exposure range of up to about 200 ppm. Due to its fast metabolism, the enrichment of EO in the body at steady-state conditions was 3.4, being much lower than the thermodynamic partition coefficient EO-in-animal / EO-in-air of 57. At EO exposure concentrations above 200 ppm, kinetics deviated from linearity, most probably due to depletion of reduced glutathione, the cofactor of the EO eliminating glutathione Stransferase. The areas under the EO concentration-time curves monitored during the exposures to constant ET concentrations reached their maximum value at about 1500 ppm ET. From the kinetic parameters of ET and EO, the internal EO burden in mice exposed under Vmax conditions to ET was calculated to be the same as upon an exposure to 4.4 ppm EO. From the outcome of the EO long-term inhalation study in mice, it can be estimated this EO concentration to be too low to result in an observable increase in the tumor incidence. Therefore, we conclude that a study on the carcinogenicity of ET in mice would become negative, as it was the case in the corresponding rat study. Financially supported by the American Chemistry Council. 1. Inst. of Toxicology, Helmholtz Zentrum München, Neuherberg, Germany

312 The role of the arylhydrocarbon receptor in cytokine production by keratinocytes in response to UVB irradiation Hübinger J. (1), Fritsche E. (2), Esser C. (1) Keratinocytes (KC) make up about 80% of the epidermal cells. By their barrier functions they protect the body from environmental influences, such as mechanical and chemical stress or pathogens. KC are also immunologically active cells and secrete various cytokines. A major environmental insult on the skin is solar UVB radiation. UVB can lead inflammatory responses and to local as well as systemic immunosuppression. In both situations, soluble factors like cytokines play a major role. The arylhydrocarbon receptor (AhR) is a transcription factor responsive to low molecular weight chemicals. In a highly cell-specific manner, the AhR mediates a range of effects by gene-modulation, including stimulation of xenobiotic metabolism, cell proliferation, and cytokine secretion. KC express the AhR constitutively. Recently, 6-formylindolo[3,2-b]carbazole (FICZ) was identified as an AhR-Ligand. FICZ is a photoproduct generated by UVB; it is active in KC. We asked whether a FICZ-AhR axis contributes to the regulation and secretion of cytokines by KC. We analyzed GMCSF, TNFα, and Cox-2 in cultivated primary KC from AhR deficient mice. All factors are pro-inflammatory agents and involved in the UVB skin response. Interestingly, the constitutive mRNA levels of TNFα, GM-CSF, and Cox-2 were significantly lower in KC from AhR-deficient mice. Two doses of 10mJ/cm2 UVB induced mRNA expression of TNFα, GM-CSF 48 hours after the first irradiation; at this time-point Cox-2 was not affected, possibly because mouse KC do not produce IL1ß, a relevant Cox-2 inducer. Higher transcription by UVB irradiation was not abrogated in AhR-deficient KC. ELISA assays confirmed UVB-mediated GM-CSF induction in wildtype mice, but GM-CSF secretion remained below the limit of detection in AhR-ko mice. Extending the studies to human KC, we found IL1β, IL8 and Cox-2 transiently induced by UVB or 50nM FICZ application in the HaCaT cell line. AhR-negative HaCaT cells (siRNA knock-down) inhibited the FICZ-induced ILlβ and Cox-2 mRNA induction. In contrast, IL-8 FICZ-mediated induction was not AhR-dependent. We conclude that in skin epidermal cells, i.e. KC, the AhR is involved in constitutive cytokine expression as well as induction of certain cytokines in response to UVB light. 1. Molecular Immunology, Institut für Umweltmedizinische Forschung (IUF), Düsseldorf, Germany, 2. Molecular Toxicology, Institut für Umweltmedizinische Forschung (IUF), Düsseldorf, Germany

313 Contact allergy is suppressed by arylhydrocarbon receptor over-activation and deficiency by different mechanisms Kadow S. (1), Jux B. (1), Esser C. (1) The arylhydrocarbon receptor is a transcription factor, which responds to low molecular weight chemicals. Persistent and strong activation by, e.g., TCDD, results in AhRdependent immunosuppression of many systemic humoral, cellular and innate immune responses. The skin as a barrier organ initiates immune responses against low molecular weight chemicals (LMWC), such as fragrances in cosmetics, drugs, chemicals from the work-environment, or environmental pollutants. We asked whether allergic skin immune responses are AhR-dependent, using contact hypersensitivity (CHS) as experimental model. CHS is a T cell dominated immune response against LWMC. The fluorescent hapten FITC was applied onto the skin, and the allergic response measured as ear thickness. C57BL/6 mice injected with 10µg/kg TCDD body weight or AhRdeficient (AhRtm1Bra) mice were used. As expected, TCDD exposure resulted in significant suppression of ear thickness (50% reduction), compared to solvent injected control. Surprisingly, AhR deficiency did not abrogate this suppression. The CHS response of AhR-deficient mice was comparable to the untreated control AhRtm1Bra mice. This is in striking contrast to the published systemic humoral and cellular responses, by antigen-injection. Langerhans cells (LC) in the epidermis and dermal dendritic cells (dDC) in the dermis are the antigen-presenting cells in the skin. Upon

65

antigen contact they move to the lymph nodes and activate T cells. Previous results had shown that LC of AhR-deficient mice were defective in maturation. We therefore determined the ability of LC and dDC to migrate into the skin draining lymph nodes (sdLN) 24h after induction of CHS by flow cytometry. In TCDD-treated mice, fewer LC and significantly fewer dDC reached the sdLN, presumably resulting in a lower T cell activation capacity. However, there was no difference between the numbers of LC and dDC in the sdLN of AhR-deficient mice, and AhR-positive mice, i.e. migration appeared to be efficient. CD69 expression, a marker of T cell activation, was equal in both AhRdeficient and –positive mice, indicating that CHS immunosuppression was not a failure of basic T-cell activation. Thus, a functional AhR is needed to mediate CHS. Albeit AhR activation and AhR deficiency both result in diminished CHS, the underlying mechanisms are different, and warrant further investigation. Transcriptional reprogramming, the hall-mark of AhR function, will play a major role. 1. Institut für umweltmedizinische Forschung an der Heinrich-Heine Universität Düsseldorf, Germany

314 Senescent arylhydrocarbon receptor deficient mice are leukopenic, but have more monocytes and less C8+ T-memory cells in the blood Lunemann S. (1), Esser C. (1) The aryl hydrocarbon receptor (AhR) coordinates cellular responses to many small molecular weight chemicals, and is triggered by endogenous substances, environmental pollutants, and plant derived substances found in food. AhR signalling is used by the immune system, and growing evidence suggests its involvement in inflammatory and autoimmune responses. We asked, whether and how lifelong absence of the AhR affects the immune system in old animals, where the immune system is in general less effective. We analyzed parameters of the immune system in old AhR-deficient mice (AhRneg-old; >14 months), and compared them to young (AhRneg-young; <4 months) mice, and to wild-type C57BL/6 mice (AhRwt-old, AhRwt-young). First, leukocyte and lymphocyte count was lower for AhRneg-old mice for AhRwt-old mice (5000/µl compared to 8000/µl, and 2000/µl vs. 5000/µl, respectively). Thus, old AhR-deficient mice are leukopenic compared to age-matched wild-type controls. The monocyte frequency, but not the neutrophil frequency was significantly increased in AhRneg-old mice to appr. 10% of all leukocytes. We subtyped T, B, and NK blood lymphocyte subpopulations flow cytometrically. Neither T cells, B cells, nor NK cells differed in frequency and absolute number between AhRneg-old and AhRwt-old mice. Exceptional were only subsets of CD8+ memory T cells (CD8+/CD62L+ and CD8+/CD62L-), which decreased appr. 3-fold in frequency and absolute number in AhRneg-old mice compared to controls. This difference was not detectable in young mice of either genotype, albeit as expected, memory T cells were rarer in young mice. The numbers of B cells were not affected in AhRneg-old mice. We measured total IgM, IgG1, IgG2a and IgA titers. IgG1 was significantly decreased to 1200µg/ml in AhRneg-old compared to 3000µg/ml in control AhRwt-old. No differences were detectable for IgM, IgG2a and IgA. We immunized mice with ovalbumin (OVA). Interestingly, OVA-specific IgG1 – a memory isotype controlled by IL4 production – appeared earlier, at approximately day 9 after immunisation, and was produced at higher titers in AhRneg-old mice than in AhRwt-old mice. Young AhRneg and AhRwt mice were equal regarding anti-OVA antibody kinetics, which was both faster and produced higher titers than in old mice. These results suggest a role of the AhR in immune homeostasis, competence, and age-related development of the immune system. 1. Molecular Immunology, Institut für umweltmedizinische Forschung, Düsseldorf, Germany

315 Peripheral blood mononuclear cells (PBMCs) - a model for cytotoxicity and immunotoxicity testing Schuetz C. (1), Fortmueller K. (1), Walter Y. (1), Hewitt P.G. (1), Mueller S.O. (1) With the advent of immunomodulatory pharmaceuticals for the treatment of cancer or auto-immune diseases, there is the urgent need for an early (preclinical) detection of immunotoxicity. Since the TeGenero case in 2006 this has become very obvious. The treatment with TGN1412 in the first phase I clinical trial caused a systematic inflammatory response, characterized by a dramatic induction of pro-inflammatory cytokines (cytokine storm). Besides the unintended induction of an immune response (immunostimulation), immunotoxicity may also result from the impaired functionality of the immune system (immunosupression) after exposure to drugs. In this study we used purified PBMCs from human volunteers and cynomolgus monkeys as an in vitro model for the detection of immuno- and cytotoxicity. Two drug development candidates that have shown anti-tumor activity were analyzed and compared to recombinant IL-2 and the positive control ConcanavalinA/Pokeweed mitogen in PHA-stimulated or unstimulated PBMCs. The measurement of pro- and anti- inflammatory cytokines by Luminex® technology in culture supernatants harvested after 2 and 24 h revealed a stronger response to compound B and the positive controls than to compound A accompanied by increased proliferation in both species. This supports preclinical findings where compound A showed reduced immunotoxic side effects caused by T-cell activation and demonstrates that PBMCs may be a relevant screening tool for the early detection of immunotoxicity. We then extended our work to measure cytotoxic effects on PBMCs in comparison to the human histiocytic lymphoma cell line U937. The cells were treated with metformine as negative control, the cytotoxic troglitazone and the sensitizer 2,4,6-trinitrobenzene sulfonic acid (TNBS). Cytotoxicity was tested by CellTiter-Gloassay after 24h, 48h and 72h and showed massive cytotoxic effects after treatment with troglitazone. The cytotoxicity data indicated that U937 cell line was more sensitive to TNBS than the PBMCs. We therefore recommend PBMCs and U937 as an early screening tool for measuring immunomodulation events and immunotoxicity/cytotoxicity, respectively. 1. Early and Explanatory Tox, Institute of Toxicology, Merck KGaA, 64271 Darmstadt, Germany

316 Assessment of the sensitising potential of textile disperse dyes and some of their metabolites by loose-fit coculture-based sensitisation assay (LCSA) Sonnenburg A. (1), Ahuja V. (1), Stahlmann R. (1), Wanner R. (1) Introduction: Certain textile disperse dyes are known to cause allergic reactions of the human skin, such as allergic contact dermatitis and contact urticaria. However, by now only few quantitative data on the sensitising potential of these dyes exist. We have tested 3 disperse azo dyes (Disperse blue 124 - DB124, disperse red 1 – DR1, Disperse yellow 3 – DY3), 3 products of azo-cleavage of these dyes (ANT - 2-amino-5nitrothiazole (DB124), AAA - p-aminoacetanilide and ApC - 2-amino-p-cresol (both DY3)) and one disperse antraquinone dye (Disperse blue 1 – DB1) to achieve data on their sensitising and irritative potential. Therefore we used a loose-fit coculture-based sensitisation assay (LCSA) of primary human keratinocytes and of allogenic DC related cells to emulate the in vivo situation of the human skin. Sensitisation was determined by analysing the expression of the DC maturation marker CD86 by flow cytometry. Estimation of the concentration required to cause a half-maximal increase in CD86expression allowed quantitative risk assessment. Furthermore we used 7-AAD (7amino-actinomycin D)-staining to achieve data on cell viability and thus the irritative potential of the tested substances. The dyes were categorised as weak or strong irritating substances by estimation of the concentration required to devitalize 50 % of the examined cells compared to a zero control. Results: DB1, ANT and AAA were tested up to concentrations of 100, 200 and 300 µmol/l, respectively, and showed no sensitising potential. All other substances were categorised as extreme sensitisers. DB124 showed the strongest sensitising potential, followed by DY3, DR1 and ApC. The irritative potential correlated with the sensitising potential. We observed most pronounced cytotoxic effects for DB124. DY3, DR1 and ApC also turned out to be highly cytotoxic substances, whereas ANT and DB1 showed only weak irritative potential. AAA did not show any cytotoxic effect at the concentration range tested. Conclusion: The LSCA proved to give adequate results for the sensitising potential assessment of coloured substances. In addition we were also able to achieve data on the irritative potential in the same series of tests. Hence the LCSA provides a stable test system to simultaneously analyse two crucial properties of substances relevant for allergy induction. 1. Institute of Clinical Pharmacology and Toxicology, Dept. of Toxicology, Charité Medical University, Berlin, Germany

317 Assessing the sensitising potential of textile dyes using a sensitive biphasic local lymph node assay Ahuja V. (1), Wanner R. (1), Platzek T. (2), Stahlmann R. (1) Reports of dye-related contact sensitization include exposure to dyes in a variety of consumer products. A modified biphasic protocol of the murine “local lymph node assay” (LLNA) was used to study the sensitising potential of various textile dyes. The textile dyes studied included Disperse yellow 3, Disperse blue 124, Disperse blue 106, Disperse orange 37, Disperse red 1, Disperse blue 35, Disperse orange 3, Disperse blue 1. All the dyes were studied at 30% concentration except for Disperse blue 124 and Disperse blue 1, which were studied at 10% concentration being insoluble at higher concentration. The test substances were applied in a biphasic manner, i.e. first on the shaved skin of the back (1-3 d) followed by application on the dorsal side of the ears (15-17 d) after two weeks. The end-points analysed included thickness and weight of an ear biopsy, weight and cell number of the draining lymph node, and lymphocyte cell surface markers (CD4+, CD8+, CD45+, CD19+, CD45+/1A+, CD69+/4+) analysed by flow cytometry. All the studied dyes, except Disperse yellow 3 and Disperse orange 3, caused an increase in the cell count and lymph node weight in the treated animals as compared to the control animals. The phenotypic markers studied also showed similar results with significant modulation in all the cases except for Disperse yellow 3. Disperse blue 124 and Disperse blue 106 showed most pronounced effects, pointing towards potent sensitisation potential among the studied substances. We conclude that a biphasic, modified protocol of the “local lymph node assay” is a suitable and sensitive approach to study the sensitisation potential of the studied compounds. Further studies with lower concentrations are ongoing to study differences in potency. 1. Institute of Clinical Pharmacology and Toxicology, Dept. of Toxicology, Charité Medical University, Berlin, Germany, 2. Federal Institute for Risk Assessment, Berlin, Germany

318 Appraisal of the sensitising potential of orally and dermally administered mercaptobenzothiazol by a biphasic protocol of the local lymph node assay Ahuja V. (1), Wanner R. (1), Platzek T. (2), Stahlmann R. (1) Mercaptobenzothiazole (MBT) is present in natural rubber products and has been reported to cause allergic contact dermatitis in many clinical cases. We studied the allergenic potential of MBT following dermal and oral routes of exposure, using a modified biphasic protocol of local lymph node assay (LLNA). Female Balb/c mice were treated with MBT (dermally - 3, 10, 30% concentrations in DMSO; orally - 1, 10, 100 mg/kg doses in corn oil) on the shaved skin of the back (dermal study) or through oral administration (oral study) on days 1-3 followed by auricular application of 3, 10 and 30% concentrations, respectively, on days 15-17 of the experiment. End points determined on day 19 included ear thickness, ear punch weight, lymph node weight, lymph node cell count, and various lymphocyte subpopulations (CD4+, CD8+, CD45+). After dermal application of a 3% solution, a significant increase in cell count and lymph node weight was observed as compared to the vehicle control group, while the CD8+ cells were found to decrease significantly. Following initial dermal application of a 10% solution, significant augmentation was seen in the lymph node weight and cell count along with decline in CD8+ cells in the MBT-treated group as compared to the vehicle control. Initial dermal application of the 30% solution resulted in a non-significant increase in cell count and lymph node weight. After initial oral administration of 1 mg/kg, we noticed a significant amplification in cell count compared to the vehicle control. Following oral administration of 10 mg/kg, we observed a similar increase in cell count and lymph node weight. These changes were less pronounced after the highest dose. The results of our study show that MBT exhibits a sensitising potential following dermal

66

and oral application, and that the modified biphasic LLNA protocol can be used to study the sensitising potential of a compound also following the oral route of exposure. 1. Institute of Clinical Pharmacology and Toxicology, Dept. of Toxicology, Charité Medical University, Berlin, Germany, 2. Federal Institute for Risk Assessment, Berlin, Germany

319 Cytochrome P450 metabolism abrogates adjuvant effects of phenanthrene in allergic inflammation Schober W. (1), Pusch G. (1), Seidel A. (2), Behrendt H. (1), Buters J.T.M. (1) Background: Diesel exhaust particles (DEPs) act as adjuvants in the immune system and contribute to the increased prevalence and morbidity of asthma and allergic rhinitis. Polycyclic aromatic hydrocarbons (PAHs) are major components of DEPs, which are involved in the induction and enhancement of proallergic processes. PAH metabolism is well documented, but its impact on initiation and aggravation of allergic diseases is still a neglected area. In this study, we explored the role of cytochrome P450 (CYP) metabolism in modulating adjuvant effects of phenanthrene (Phe) on activation parameters of mast cells, fostering allergic inflammation through the release of preformed or granule-derived mediators. Methods: Using V79 cell lines stably expressing CYP1A1, 1A2, 1B1, or 2E1 we found that Phe-1,2-, Phe-3,4-, and Phe-9,10dihydrodiol are main metabolites in man and mouse. The impact of Phe and its diols on anti-IgE induced degranulation was examined with L138.8A cells, an IL-3 dependent murine mast cell line constitutively expressing FcεR1. Supernatants were assayed for LTB4 and LTC4 secretion and β-hexosaminidase release by means of ELISA. In addition, human polymorphonuclear leukocytes were obtained from atopic and nonatopic donors and chemotactic activity of supernatants was evaluated, measuring the neutrophil migration through 3 µm pore polyethylene filter in 24-well transwell chambers. Results: At environmental relevant exposure levels Phe augmented the release of β-hexosaminidase from IgE/anti-IgE activated mast cells by 32%. Moreover, Phe significantly enhanced secretion of LTB4 and LTC4 up to 90%. Both leukotrienes are involved in the pathophysiology of asthmatic and allergic reactions, causing or potentiating symptoms such as bronchoconstriction or infiltration of inflammatory cells in the airway wall. In this study, enriched neutrophils revealed capacity to migrate in response to supernatants from Phe-treated mast cells, which indicates relevance of our findings with murine mast cells for allergic inflammation in humans. In contrast, all main metabolites of Phe had no effect on β-hexosaminidase release or LTB4 and LTC4 secretion, pointing to a modulating role of CYP metabolism on DEP-PAHs’ enhancement of allergic responses. Conclusion: Our results suggest that not only the catalytic activity of various glutathione-S-transferases, but also of CYP isoforms can abrogate adjuvant effects of organic air pollutants in the IgE dependent immune response. 1. Division of Environmental Dermatology and Allergy, Helmholtz Zentrum München/TUM, ZAUM – Center for Allergy and Environment, Technische Universität München, Munich, Germany, 2. Biochemical Institute for Environmental Carcinogens, Prof. Dr. Gernot Grimmer-Foundation, Großhansdorf, Germany

320 Lipopolysaccharide-induced changes of liver function and oxidative stress in rats: influence of pretreatment with gadolinium chloride Lupp A. (1), Müller D. (1) Sepsis, septic shock and subsequent multiorgan failure remain an important cause of morbidity and mortality despite recent advances in critical care. About 25% of the patients with severe inflammatory reaction display an impaired liver function, e.g. with respect to biotransformation capacity and transporter activity, which to a great extent seems to be mediated by cytokines liberated by Kupffer cells. As yet unsolved problem, no specific therapy exists so far for the treatment of liver failure in sepsis. Thus, using a rat model of endotoxinemia, the aim of the present study was to elucidate if Kupffer cell suppression by means of gadolinium chloride (GdCl3) can serve as a possible therapeutic strategy. For this purpose, adult male rats were i.v. pretreated either with saline or with 5, 10 or 20 mg/kg body weight GdCl3 24 h prior to i.p. administration of either saline or of 1 mg/kg body weight lipopolysaccharide (LPS). 16 h after LPS treatment indocyanine green (ICG) clearance experiments were performed. Animals were then sacrificed and different parameters representing liver function or indicating oxidative stress were assessed. LPS treatment caused a distinct decrease in liver cytochrome P450 (CYP) isoforms expression and activities. Bile flow, biliary ICG and bile acid secretion and hepatic bile acid uptake were reduced, whereas both total, direct and indirect bilirubin values in bile were increased. Indirect bilirubin values were also slightly elevated in serum. LPS did not affect hepatic levels of reduced and oxidized glutathione and of lipid peroxidation products. Heme oxygenase expression and activities, on the other hand, were strongly increased. At the highest concentration tested GdCl3 given alone only decreased CYP isoforms expression and activities. Pretreatment with GdCl3 prior to LPS administration either did not affect or even aggravated the LPS-induced changes in CYP isoforms and in heme oxygenase expression and activities. Additionally, a decrease in hepatic glutathione content and an increase in lipid peroxidation products were seen after combined treatment with GdCl3 and LPS. On the other hand, effects of LPS on bile flow, on biliary ICG and bile acid excretion and on hepatic bile acid uptake were reduced by GdCl3 pretreatment. Thus, despite some positive effects on transporter function, Kupffer cell suppression by means of GdCl3 does not seems to be a valuable therapeutic strategy to ameliorate LPSinduced liver injury. 1. Institute of Pharmacology and Toxicology, Friedrich Schiller University, Jena, Germany

321 Influence of engineered silica nanoparticles on human colon carcinoma cells Gehrke H. (1), Frühmesser A. (1), Marko D. (1) The use of nanostructured materials is no longer restricted to technological applications but is rather finding its way into multiple areas of daily life. The use of silica (SiO2) nanoparticles has been extended to biomedical and biotechnological fields, for example in cancer therapy, DNA or drug delivery. Furthermore, as a non-metal oxide, SiO2 has found extensive applications as additives to cosmetics, varnishes and food. Thus, the environmental and health impact of nanoscaled SiO2 is of great interest. A number of studies have been published, showing controversy effects of silica nanoparticles. Studies in human bronchoalveolar carcinoma cells showed a dose-dependent cytotoxicity of SiO2 associated with increased oxidative stress [Lin et al., 2006]. On the other hand a multitude of data has been reported showing no adverse effects of SiO2 nanoparticles. The toxicological relevance of silica nanoparticles focused so far primarily on inhalative intake. In contrast, little is known about the toxicological effects of these nanoparticles in the gastrointestinal tract after oral uptake. Thus, our studies on the toxicological relevance of SiO2 nanoparticles focused on cytotoxic effects and the modulation of the cellular redox status in human colon carcinoma cells (HT29). The investigated particles were commercially available with a primary particle size of 12 nm. Cytotoxicity of SiO2 was determined by the release of the cytosolic LDH. Furthermore, the proliferative effect after SiO2 incubation was determined by the sulforhodamine-B assay. Investigations on the cellular redox status were performed by measuring the total GSH level. Our studies showed a significant proliferative stimulus at a concentration of 250 µg/ml, enhancing with the incubation time from 24h to 72h. At a concentration of 500 µg/ml no proliferative effect was observed. In addition, the LDH assay exhibited slightly cytotoxic effects, occuring only in the highest concentration of 500 µg/ml. Furthermore, the cellular tGSH level was enhanced up to 160 % after an incubation time of 24h at a concentration of 500 µg/ml. In conclusion, the investigated SiO2 nanoparticles stimulate the proliferation of HT29 cells up to a concentration of 250 µg/ml, depending on the incubation time. At higher concentrations, however, they exhibit marginal cytotoxic effects. Furthermore, the investigated particles lead to an increase of the glutathione biosynthesis, indicating a higher metabolic ground state. 1. Section of Food Toxicology, Universität Karlsruhe (TH), Germany

322 Nanoparticle-activated neutrophils cause oxidative DNA damage in human colon epithelial cells Gerloff K. (1), Boots A.W. (1), Förster I. (2), Albrecht C. (1), Schins R.P.F. (1) Engineered nanoparticles (ENP) offer a range of new features that are of interest for optimisation of food and food packaging, thus in the future they will be increasingly used for these purposes. The ability of inhaled nanosized particles to induce inflammatory diseases in the respiratory tract and their association with an increased risk of cancer have been described in previous studies. In contrast, very little is presently known about the potential toxic effects of ENP within the gastrointestinal tract and their possible impact on colon cancer or inflammatory diseases, such as ulcerative colitis or Crohn´s disease. The aim of the present study was to investigate the (oxidative) DNA damage induced by a panel of ENP on the human colon epithelial cell line Caco-2 using the fpgmodified comet assay. To determine the potential effects of ENP in the inflamed colon, we also investigated their ability to activate human neutrophils via lucigenin-enhanced chemiluminescence. Activated human and murine neutrophils were used in a coincubation model to examine their potential for inducing oxidative DNA damage in the Caco-2 cells via superoxide generation. Furthermore, the role of the NADPH oxidase complex was investigated by comparing the response to activated neutrophils from p47phox -/- mice. Our current results revealed significant DNA strand breakage and oxidative DNA damage induction by ZnO in the Caco-2 cells. Most of the particles were also found to activate human neutrophils, which themselves were shown to enhance strand breaks and oxidative damage in the Caco-2 cells. The induced DNA damage appeared to depend on a functional NADPH-oxidase complex. Supported by DFGGK1427 1. Particle Research, Institut für umweltmedizinische Forschung (IUF), Düsseldorf, Germany, 2. Molecular Immunology, Institut für umweltmedizinische Forschung (IUF), Düsseldorf, Germany

323 Absorption enhancers and jejunal uptake kinetics in vitro of methyl α-D-glucoside Eberhardt K. (1), Elsenhans B. (1) So-called absorption enhancers (AEs) are proposed to increase small-intestinal absorption of poorly available drugs. While some of them inhibit the mucosal metabolism or active efflux of drugs, others increase drug permeability by impairing mucosal barrier function. Thus, this latter type of AE, e.g., deoxycholate (DOC), polyethyleneimine (PEI), decanoate (DEC), and ethylenediamminetetraacetate (EDTA) may impair sodiumdependent active transport processes, due to their different structures and properties not necessarily by the same mechanism, however. Therefore, these AEs were compared for their effect on the kinetic parameters of intestinal methyl α-D-[14C]glucoside uptake. KM and Vmax were determined in vitro by means of 5-min incubations with everted rings of rat jejunum and by analysing the concentration-dependent tissue uptake according to Lineweaver-Burk (for results see Table). control DOC PEI DEC EDTA KM (mmol/l) 4.8 ± 0.9 6.9 ± 1.2* 6.6 ± 0.9* 4.5 ± 0.5 4.7 ± 0.5 Vmax (mmol/l/min) 11.3 ± 2.3 8.2 ± 1.0* 8.0 ± 1.2* 6.3 ± 0.4* 7.0 ± 0.8* M ± SD, N = 6; * p < 0.05 (ANOVA) While all AEs employed reduced the maximal uptake capacity (Vmax) of the jejunal mucosa as compared to the control, only DOC and PEI increased KM, i.e., reduced the affinity of the glucose analogue for its transport carrier. Since unstirred water layers are unlikely to be changed under the present conditions, membrane interactions are considered to be responsible for the KM effect, yet the mode of action of the lipophilic DOC may differ from that of the hydrophilic PEI.

67

1. Walther Straub-Institut für Pharmakologie und Toxikologie der LMU München, München

324 Microglia-mediated increase of heme oxygenase-1 (HO-1), superoxide dismutase2 (Sod-2) and glutathione and the effect of microglia on peroxide-induced generation of radical species and lipid peroxidation in astrocytes in vitro Röhl C. (1,2), Armbrust E. (1,2), Herbst E. (2), Jess A. (1), Gülden M. (1), Maser E. (1), Bösch-Saadatmandi C. (3) Oxidative stress as well as inflammation are common features of many neurological disorders, in which microglia and astroglia are cellular key players as they are involved in the generation of radicals, anti- and proinflammatory substances and in the protection against oxidative stress. In previous studies we have shown that microglia activated by lipopolysaccharides (LPS) modulate astroglial enzymes involved in oxidative and inflammatory stress and increase the resistance of astrocytes to oxidative stress induced by hydrogen peroxide (H2O2). Here, we examined further effects of activated microglia on astrocytes: a) on selected antioxidative systems, b) on the elimination of hydrogen peroxide, c) on the cytotoxicity of the organic cumene hydroperoxide (CHP) and d) on peroxide-induced lipid peroxidation and radical generation. Therefore, purified cell cultures of astroglia from neonatal rat brains were treated for three days with medium conditioned by purified quiescent (MCM[–]) or LPS-activated (MCM[+]) rat microglia before measurements were performed. While HO-1 expression was increased by treatment with microglial medium (MCM) irrespective of the microglial activation, Sod2 activity and total glutathione in astrocytes was solely increased by MCM[+]. Sod-2 mRNA expression was enhanced but not that of other selected antioxidative enzymes. Whereas the Sod-2 expression, in parallel to its activity, increased over time of MCM[+] incubation, reaching a plateau after 24 hrs, HO-1 expression showed a single peak after 4 hrs. The ability of astrocytes to eliminate H2O2 was not effected by incubation with MCM. Nevertheless, generation of radical species due to H2O2 treatment was slightly decreased. Finally, pre-treatment of astrocytes with MCM[+] significantly increased their resistance to CHP induced oxidative damage. CHP, in contrast to H2O2, induced lipid peroxidation in astrocytes, which was decreased by pre-treatment with MCM.Taken together, the changes of HO-1 expression, of lipid peroxidation by CHP and of radical generation in astrocytes could not explain the better protection of astrocytes by MCM[+] compared to MCM[–]. Nevertheless, the specific increase of Sod-2 activity and glutathione only in MCM[+] treated astrocytes and their higher resistance also to CHP induced oxidative stress further indicates an important role of microglial-astroglial cell interactions during inflammatory processes. 1. Inst. of Toxikology and Pharmacology for Natural Scientists, Christian-AlbrechtsUniversity, Kiel, Germany, 2. Inst. of Anatomy, Christian-Albrechts-University, Kiel, Germany, 3. Inst. of Human Nutrition and Food Science, Christian-Albrechts-University, Kiel, Germany

325 Influence of polybrominated diphenylethers on neural development Schreiber T. (1), Abel J. (1), Fritsche E. (1) Polybrominated diphenylethers (PBDEs) are persistent and bioaccumulative flame retardants which are present in textiles, electronics, plastics and furniture. They are of concern because they are ubiquitous, potentially toxic and have been found at increasing levels in humans during the past few decades. Animal studies have shown that PBDEs can lead to cancer in high dose studies. Moreover, they are reproductively and developmentally toxic, lead to endocrine disruption and cause central nervous system effects. For investigating effects of exogenous factors on neurodevelopment, we have established a human in vitro system for studying developmental neurotoxicity (DNT) in human primary neural progenitor cells. With this system, influences of chemicals on viability, proliferation, migration and differentiation can be analyzed. To investigate the influence of PBDEs on human neural development, human neurospheres were treated with different non-cytotoxic concentrations of the tetrabrominated BDE-47 or the pentabrominated BDE-99. While there was no effect on proliferation of neurospheres, there were significant reductions of progenitor cell migration after 48 hours with both congeners at all concentrations tested. Furthermore, we observed a strong decrease in number of formed neurons and oligodendrocytes after seven days of differentiation. Taken together, PBDEs cause reduced migration and neural differentiation of neural progenitor cells in vitro. Moreover, the effects of PBDE-99 were stronger than the ones observed by PBDE-47. This is in concert with the literature which reports stronger toxicities of the higher brominated PBDEs compared to the lower brominated ones. We suggest that the effects of PBDEs might be mediated through endocrine disruption of the thyroid hormone system, because (i) PBDEs and thyroid hormone (TH) share structural similarities, (ii) TH contributes to normal brain cell migration and (iii) TH plays a key role in oligodendrocyte development. The precise mechanisms how PBDEs disturb neurodevelopment further have to be elucidated. 1. Institut für umweltmedizinische Forschung, Düsseldorf, Germany

326 Effect of cytokines on poly(ADP-ribose)polymerase-1 (PARP-1) expression and activity in primary human lung cells Sacher B. (1), Torky A. (1), Ahmad M. (1), Foth H. (1) Exposure to ambient air pollution has been associated with lung diseases and cancer. Ambient air contains a complex mixture of toxicants. The mechanisms of foreign substance-induced pulmonary health effects are believed to involve inflammation and oxidative stress. Oxidative stress can lead to activation of the nuclear enzyme poly (ADP-ribose) polymerase (PARP), with subsequent loss of cellular functions and cell death. In addition, PARP enhances the expression of various pro-inflammatory mediators, via activation of NF-kappaB, MAP kinase and AP-1 and other signal transduction pathways. Our study was directed to investigate a possible feed-back regulatory effect of the pro-inflammatory mediators, IL6, IL-1ß, and TNF-α at physiological concentrations (25ng/ml, 0.5ng/ml, 0.5ng/ml, respectively; 24h and 72h) on PARP1 expression and activity in both normal human bronchial epithelial (NHBEC) and

peripheral lung cells (PLC). We cultivated both normal human lung cells from bronchial as well as peripheral lung tissue explants to expand the cultures of NHBEC and PLC. In cultivated NHBEC, treatment with IL1ß and TNF-α for 24h tended to decrease PARP1 expression down to 80% compared to the control. IL6 (24h) induced the expression by 1.2 fold only in one sample. PARP1 expression was slightly decreased (≤ 20% relative to control) upon treatment with IL-6 and TNF-α for 72h in all samples. In PLC the expression of PARP1 showed slight down-regulation (20-25%) with TNF-α and IL-6 (24h), while IL-1ß had almost no effect on protein expression. The results showed in general no effect of all tested cytokines on PARP1 expression in the same sample after 72h of incubation. On the other hand TNF-α (24h) and IL6 (72h) up-regulated the expression by 1.4 fold in one of the samples. The next objective was to measure the PARP1 activity using fluorescence functional assay, because this may be independent from protein content. All used samples showed significant initial suppression reaction of the activity in 24 h cultures followed by reversal of the activity into the stimulated spectrum in the 72 cultures. In summary, the variability of individual response to inflammatory stimuli is the determining factor for the expression of PARP1. Furthermore the response of the PARP1 activity upon treatment with inflammatory mediators is not correlated to the changes in protein expression. 1. Institute of Environmental Toxikology, Faculty of Medicine, Martin Luther University, Halle-Wittenberg

327 In vivo and in vitro investigations of the contribution of neutrophils to the genotoxicity of quartz particles Wessels A. (1), van Berlo D. (1), Wilhelmi V. (1), Scherbart A. (1), Gerloff K. (1), Boots A. (1), Albrecht C. (1), Schins R.P.F. (1) The carcinogenicity of inhaled crystalline silica (quartz) particles is considered to be driven by excessive formation of reactive oxygen species (ROS) generated during inflammation.In this study we investigated the contribution of neutrophil-derived ROS to genotoxicity upon quartz exposure in p47phox-/- mice, which are characterised by an impaired superoxide generation from phagocytes. The in vivo study involved evaluation of inflammation, oxidative stress and DNA damage responses in the lungs of wildtype (WT) or p47phox-/- mice, 24h after pharyngeal aspiration of DQ12 quartz (100 mg/kg b.w.). In parallel in vitro investigations the genotoxicity of DQ12 was evaluated in cocultures of A549 human lung epithelial cells and freshly isolated bone marrow neutrophils from WT or p47phox-/- mice.In the WT as well as in the p47phox-/- mice, DQ12 caused a marked pulmonary inflammation, characterised by the influx of inflammatory cells measured in bronchoalveolar lavage as well as an enhanced mRNA expression of the pro-inflammatory cytokine tumour necrosis factor-α in whole lung tissue. Heme oxygenase-1 mRNA levels were induced in DQ12-treated WT mice, but to significantly lower extent in the p47phox-/- mice, indicative of reduced oxidative stress in the knock out animals. After DQ12 treatment, the DNA base excision repair enzymes Ogg1, Ape-1, and Xrcc1 showed a significantly reduced mRNA expression in whole lung tissue homogenates of p47phox-/- mice but not in the WT mice. No differences in oxidative DNA damage levels, as measured by fpg-comet assay, were detected in the whole lung tissue of p47phox-/- or WT mice treated with quartz in comparison to control animals. In contrast to these in vivo findings, the role of neutrophil-derived ROS in the genotoxicity of quartz could be demonstrated in the in vitro model. Firstly, WTneutrophils that were co-cultured with A549 cells and quartz particles induced more oxidative DNA damage to the A549 cells, than quartz treatment alone. Secondly, this enhancement was not observed when p47phox-/- neutrophils were used. Supporting these findings, the impaired superoxide generation of p47phox-/- neutrophils after quartz treatment in comparison to WT neutrophils could be demonstrated by electron paramagnetic resonance measurements. Taken together, this shows that activated neutrophils may enhance the DNA damaging potency of quartz in a NADPH-oxidase dependent manner. 1. Institut für Umweltmedizinische Forschung (IUF), Düsseldorf, Germany

328 Analysis of oxidative stress, inflammation and oxidative DNA damage in mice after short-term inhalation of spark-generated carbon nanoparticles van Berlo D. (1), Wessels A. (1), Boots A. (1), Gerloff K. (1), Scherbart A. (1), Cassee F.R. (2), Gerlofs-Nijland M. (2), Fokkens P. (2), van Schooten F.J. (3), Albrecht C. (1), Schins R.P.F. (1) Exposure to particulate matter (PM) has been associated with a broad range of adverse health effects. In particular, the ultrafine carbonaceous fraction of this mixture has been implicated in the induction of oxidative stress and inflammation, considered primary mechanisms involved in PM-induced disease. We have investigated the effects of acute exposure to carbon nanoparticles on inflammation, oxidative stress and oxidative DNA damage in mice after short-term inhalation. Female C57Bl/6 mice were exposed noseonly to carbon nanoparticles generated on-site by spark-discharge (140µg/m3, 57.6nm) or to filtered air, once for 4h, or for 4h on three consecutive days. After treatment, animals were sacrificed and bronchoalveolar lavage (BAL) was performed to analyse markers of cytotoxicity (total protein, LDH) and inflammation (total cell influx). Also, lung tissue was isolated for the analysis of oxidative DNA damage. No severe pulmonary toxicity and inflammatory responses were detected in the nanoparticle-exposed mice, as indicated from the BAL fluid analysis. However, observed changes in BAL fluid levels of glutathione suggest a mild oxidative stress induction by the nanoparticles. Employing the Formamidopyrimidine glycosylase (fpg)-modified comet assay in whole lung tissue, no increase in oxidative DNA damage could be detected. Current analysis focuses on the determination of inflammatory-, oxidative stress- and DNA repair markers on the protein level using immunohistochemistry and western blot (using lung tissue sections and whole lung respectively) as well as on the mRNA level by performing quantitative real-time RT-PCR. Present results of ongoing investigations suggest that short-term inhalation of carbon nanoparticles induces no, or only minor responses in the mouse lung. 1. Institut für umweltmedizinische Forschung (IUF), Düsseldorf, Germany, 2. National Institute for Public Health and the Environment (RIVM), Bilthoven, the Netherlands, 3. Maastricht University, Maastricht, the Netherlands

68

329 Biological effects of synthetic metal-oxide nanoparticles in lung cells Marquardt C. (1), Panas A. (1), Weiss C. (1), Diabaté S. (1) With the emergence of nanotechnology there is an increased production of engineered nanoparticles (NPs) with novel physical and chemical properties. Whether some synthetic NPs are hazardous is still poorly understood. From epidemiological studies on ambient air pollutants it has been concluded that inhalation of combustion-derived nanoparticles (NPs) induce adverse health effects. The prevailing hypothesis suggests the generation of reactive oxygen species as a key event to explain NP- induced toxicity. The aim of this study is to identify the most relevant properties of engineered NPs that induce biological and potentially adverse responses. As target cells the human alveolar epithelial cell line (A549) and the mouse macrophage cell line (RAW 264.7) were used. Different metal-oxide NPs (Fe2O3 (12 nm), SiO2 (12 nm), TiO2 (32 nm)) were screened for distinct biological endpoints such as viability, proliferation, generation of reactive oxygen species (ROS), induction of signalling pathways and altered expression of antioxidant enzymes and proinflammatory cytokines. Cellular uptake of all NPs (Fe2O3, TiO2, SiO2) was confirmed by transmission electron microscopy. Interestingly, we could not detect toxic effects (LDH assay) in cells treated with TiO2- or Fe2O33- NPs with concentrations up to 300 µg/ml. In contrast, a significant loss in viability could be observed in RAW 264.7 macrophages, but not in A549 epithelial cells, exposed to SiO2NPs already at a concentration of 50 µg/ml. Furthermore, we investigated the ability of the selected NPs to induce reactive oxygen species (ROS) by use of the fluorescence dye H2DCF. Neither Fe2O3- nor SiO2-NPs generated ROS. In contrast, TiO2-NPs generated ROS under normal light conditions but not in the dark. For the most potent SiO2-NPs more detailed analysis of signalling events in RAW 264.7 macrophages were performed. SiO2-NPs induced activation of the MAP kinases ERK, JNK and p38. This was paralleled by an accumulation of the transcription factor Nrf2 and elevated expression of antioxidative enzymes such as HO-1, γGCL and NQO1. Additionally, enhanced expression of the inflammatory genes IL-6, COX-2 and TNFα could be observed. In summary, the generation of ROS by nanoparticles does not correlate well with loss of cell viability and signal transduction. Thus NPs may also initiate signalling, gene expression and toxicity via novel routes independent of ROS production. 1. Institute of Toxicology and Genetics, Forschungszentrum Karlsruhe, EggensteinLeopoldshafen, Germany

330 Elemental composition and cytotoxic effects of indoor PM10 compared to outdoor PM10 Oeder S. (1), Weichenmeier I. (1), Schober W. (1), Dietrich S. (2), Fromme H. (2), Behrendt H. (1), Buters J.T.M. (1) Background: Outdoor particulate matter (PM10) is associated with a wide range of health effects and a European threshold limit was established in 2005. However, most individuals spend at least 85% of their time indoors where particle concentrations are often higher than outdoors. Since children represent a vulnerable group, we investigated the health effects of indoor air PM10 collected in classrooms compared to outdoor air PM10 within the PAMINA (Particulate Matter in Indoor and Ambient Environments) project. Methods: PM10 was collected indoors and outdoors at five schools in Munich during teaching hours. Particles were recovered by sonication, lyophilized and resuspended in water. Cytotoxicity was assayed as a decline of cellular ATP concentration in human primary keratinocytes, human lung epithelial A549 cells and Chinese hamster V79 lung fibroblasts at concentrations from 3ng/ml to 10µg/ml. In addition, toxicity was assayed after metabolic activation in V79 cells expressing human cytochrome P450 1A1, 1A2, 1B1, 2A6, 2B6, 2C9, 2D6, 2E1, 3A4 or 3A5. Elemental composition of inhalable fraction was analyzed with energy dispersive X-ray spectroscopy (EDX, Thermo Inc.). Using a chemical library, which was established by analyzing the first filters in a learn mode, the particles were classified. Results: School PM10 concentrations were 126±48µg/m3 indoors and 27±15µg/m3 outdoors. While in A549 and V79 cells no toxicity was observed, in human primary keratinocytes indoor and outdoor PM10 at a concentration of 10µg/ml caused a slight, but statistically significant decrease in vitality. This cytotoxic effect was also found in V79 cells after metabolic activation by CYP2C9. The composition of classroom PM differed from that of outdoor air PM. Most of classroom particles were silicates (34%), organic (29%) and calcium carbonate particles (12%). In contrast 44% of outdoor PM was calcium sulfate. The major elements were O>Si>Ca>Al for indoor and O>S>Ca>Fe for outdoor PM. The particle composition did not differ in the five analyzed schools. Conclusion: The collected classroom PM composition differed substantially from that of outdoor air PM. Although cytotoxicity and toxicity after metabolic activation by cytochrome P450 isoform 2C9 were statistically significant at a PM10 concentration of 10µg/ml, this exposure was about 10,000 times higher than exposure encountered in classrooms. We therefore expect no toxic effects of these particles in school children. 1. Division of Environmental Dermatology and Allergy, Helmholtz Zentrum München/TUM, ZAUM-Center for Allergy and Environment, Technische Universität München, Munich, Germany, 2. Department of Environmental Health, Bavarian Health and Food Safety Authority, Oberschleißheim, Germany

331 Short-term inhalation tests with eight nano-materials in rats Ma-Hock L. (1), Treumann S. (1), Strauß V. (1), Gamer A.O. (1), Wiench K. (1), van Ravenzwaay B. (1), Landsiedel R. (1) Inhalation exposure is the route of highest concern for nano-particles and nano-objects. BASF has developed a standard short-term inhalation test to examine the toxicity of inhaled aerosols from nanomaterials. In the course of this project, the generation of atmospheres from nanomaterials and the characterization of these atmospheres (by cascade impactor, optical particle counter and scanning mobility particle sizer) have been investigated. To estimate the toxicity of the inhaled aerosols from nanomaterials an extensive set of parameters has been evaluated an most sensitive and predictive parameters have been identified (Ma-Hock 2008 [1]). Applying the standard short-term inhalation test for nanomaterials, we tested eight nano-materials, TiO2, ZnO, CeO2, ZrO2, amorphous SiO2, surface-coated amorphous SiO2, carbon black and a multi-wall carbon nano tubes, in male Wistar rats. To achieve a concentration-effect relation,

groups of rats were exposure to three different concentrations for six hours on five consecutive days. At the end of the exposure, and after a recovery period of two to three weeks, animals were examined for extensive sets of parameters: Organ burdens were estimated in seven tissues. Electron microscopy was carried out to characterize the particles deposited in the lung and to determine possible translocation of the materials. Moreover, the respiratory tract of the animals, as well as all organs and tissues with macroscopic or organ weight changes were examined by histopathology. S-phase response and apoptosis was examined in the lung. The complete set of hematology and clinical chemical parameters in blood were examined. Additionally, an extensive set of cytokines and chemokines as well as standard clinical pathology parameters were analyzed in lung lavage fluid and blood. The lung and body distribution, the lung effects and NOAEC differed among the eight nanomaterials indicating effects of the agglomeration state, the surface area and the surface chemistry on the fate and effect in the lung. [1] Ma-Hock, L, Burkhardt, S., Strauss, V., Gamer, A.O., Wiench, K., van Ravenzwaay, B., Landsiedel, R. 2008. Development of a short-term inhalation test in the rat using nano titanium dioxide as a model substance. Inhal. Tox., in press. 1. Dept. of Product Safety, BASF SE, Ludwigshafen/Rhein, Germany

332 Influence of different cell numbers exposed to oxidative stressors Torabi S. (1), Wittek F. (1), Kolb M. (1), Walther U.I. (1) Glutathione is thought to be the most important small molecular antioxidative substance in cells. Beyond, exported glutathione might have extracellular direct antioxidative properties as well. In investigations typically used peroxides are eliminated in high rates from medium, even without cells, therefore changed antioxidative medium capacity (by different glutathione contents) might have a strong impact on the toxicity resulted. In this work we incubated alveolar epithelial L2 or A549 cells (normally adherent growing cells) in suspension cultures with different cell numbers in order to change the glutathione content of the incubated batches. Cells were incubated with hydrogen peroxide (HP), tert.-butylhydroperoxide (tBHP) or zinc chloride, all known to initiate oxidative stress. Furthermore, glucocorticoid (GC) pretreatment of L2 and A549 cells is known to diminish glutathione content. Therefore the experiments were performed after GC pretreatment of cells as well. Toxicity was assessed by measurement of 35S-methionine incorporation (inhibition). GC pretreatment significantly decreased glutathione content and methionine incorporation in L2 cells. In A549 cells the cell number was decreased, but glutathione derived parameters were unchanged by the GC. In zinc- or tBHP-exposed L2 cells neither a dependence of toxicity towards glutathione content nor to glutathione/protein ratio was found. In addition after GC pretreatment no changes in the toxicity were found. In A549 cells HP toxicity was significantly dependent on glutathione content, but again, after pretreatment of A549 cells with dexamethasone, no changed behaviour was found. Most investigations with decreased glutathione content base on buthionine sulfoximine treatment, a covalently binding glutathione synthesis inhibitor. Therefore these investigations normally test a condition of decreased glutathione synthesis as well. In our experiments toxicity mediated by zinc or tBHP (toxicities of both are described as glutathione dependent) was not dependent on glutathione. HP-mediated toxicity is known to be antagonized by super oxide dismutase and glutathione as well, and this seems to be dependent on glutathione as seen by a cell number inocculum effect, while zinc and tBHP`s behaviour in this manner can be named as classically. 1. Inst. für Toxikologie und Pharmakologie, Universität Rostock, Rostock, Germany

333 How donor and confluence influence the effects of heavy metals in cultures of human lung cells Glahn F. (1), Wiese J. (1), Harders A. (1), Torky A. (1), Ahmad M. (1), Foth H. (1) Human lung is constantly exposed to low doses of heavy metals. Many of these heavy metals generate reactive oxygen species leading to oxidative DNA-damage. Cells can adapt to heavy metal stress by expression of multidrug resistance-associated proteins (MRP). We used the human lung tumor cell line H322 and primary cultures of normal human bronchial epithelial cells (NHBEC) as model system to study the effects of Cu(II), Cd(II), Co(II) and Pb(II). We established appropriate doses for treatment using MTTassay. By means of the alkaline comet-assay genotoxic effects of Cd, Co and Pb, single and combined, on NHBEC were analysed. The cultures derived from different donors showed altered response to metal stress. A decrease of cell viability below 80% was observed at 2.4 mg/l Cd, 1,2 mg/l Co and 6.2 mg/l Pb for 72h. Combined incubation was more toxic than comparable single doses. In one case much lower doses (0.015 mg/l Cd; 0.025 mg/l Co and 0.55 mg/l Pb) reduced viability to 80%. Cultures from individual donors showed different sensitivities for the DNA-damaging properties of the applied metals in the comet assay. Moreover we used real-time RT-PCR to investigate expression of MRP-transporters in lung cell cultures. In H322 cells exposed to Cu (2.5 and 5 µM) for 72h at different states of confluence neither treatment nor confluence influenced expression of MRP1 and 2. However, MRP3 was repressed (0.3 fold) in untreated cells cultured for three days after cells had reached 100% confluence (100%+3d) compared to controls at 80% confluence. MRP4 was increased (2.8 fold) in 100% confluent controls. Cu increased MPR4 (2 fold) in 100%+3d-cultures. MRP5 was induced in 100% and 100%+3d confluent untreated cultures (4.7 and 3 fold respectively). Treatment with Cu repressed MRP5 in 100% confluent cultures (0.2 and 0.4 fold) and in 100%+3d cultures (0.4 fold). We also treated NHBEC with Cu (2.5 and 5 µM) for two times 5 days with a 5 day recovery period in between. We analysed MRPexpression at 100% confluence. Analysing cultures of 3 different donors we could not find an effect of Cu on the expression of MRP1, 3, 4 and 5, neither after 5 d nor after 5 + 5 d in culture. Only MRP1 was increased (1.5 fold) in the controls of one patient after the second 5 d in culture.Summing up our results demonstrate differences between the reactions of primary cultures from different donors on the one hand and between tumor cells on the other hand when being treated with the applied metals. 1. Institute for Environmental Toxicology, Martin-Luther-University Halle-Wittenberg, Halle (Saale), Germany

69

334 Interspecies differences in pharmacotherapy of peripheral respiratory arrest after exposure with tabun Gonder S. (1), Seeger T. (1), Worek F. (1), Thiermann H. (1) Extrapolation of animal data to humans concerning intoxications with organophosphorus compounds (OP), i.e. pesticides or highly toxic nerve agents, are hampered by marked species differences. Hence, comparative studies on interspecies differences in such poisonings are paramount in order to improve the pharmacotherapy in humans. The aim of the present study was to compare the restoration of tabun blocked force generation of diaphragms from different species. Rats and guinea pigs are mostly used for antidote research in OP poisoning and knowledge on possible differences between these species are of utmost relevance. A high-throughput 12-chamber organ bath with force transducers and stop-flow superfusion was used. The tabun induced neuromuscular block and antagonizing effect of HI 6 and obidoxime was investigated with rat and guinea pig diaphragms. Muscle contraction was induced by direct and indirect field stimulation technique. Hereby, indirect field stimulation activates intramuscular nerve branches. Indirect stimulation was proved by using 10 µM pancuronium. Incubation of diaphragms with tabun resulted in complete neuromuscular block that could be restored to various extents by HI 6 or obidoxime. A significant difference between muscular response in various species could be demonstrated. Diaphragm of rats (n=16-17 per group; mean±SE) showed regeneration of muscle force of 11±2% (20 Hz), 8±3% (50 Hz) and 6±2% (100 Hz) following HI 6 and 59±8% (20 Hz), 53±6% (50 Hz) and 48±5% (100 Hz) after obidoxime. In guinea pigs (n=11-15 per group) the measured percentage showed muscle force restoration of 12±4% (20 Hz), 8±3% (50 Hz) and 8±2% (100 Hz) after HI 6 and 140±19% (20 Hz), 118±11% (50 Hz) and 84±8% (100 Hz) after obidoxime.The results of muscle force measurement in the multi organ bath showed significant differences of oxime efficacy after tabun poisoning between the tested species, especially concerning obidoxime. The results obtained with HI 6 differ markedly less. These findings are in line with previous studies demonstrating substantial differences in kinetic properties of human and animal erythrocyte acetylcholine esterase. Hence, in vitro and vivo studies with rats and guinea pigs on antidotal efficacy in OP poisoning can only be extrapolated very carefully to humans. 1. Bundeswehr Institute of Pharmacology and Toxicology, Munich, Germany

335 Withdrawn

prospectively followed), plus 23 healthy controls. Ongoing studies in the USA, as well as continuation of recently published epidemiological cancer incidence analysis should provide additional reassurance for MPH treated ADHD patients. 1. Department of Toxicology, University of Wuerzburg, Wuerzburg, Germany, 2. Department of Child and Adolescent Psychiatry and Psychotherapy, University of Wuerzburg, Wuerzburg, Germany, 3. Department of Child and Adolescent Psychiatry, University of Zuerich, Zuerich, Switzerland

338 Thalidomide resistance is based on the capacity of the glutathione-dependent antioxidant defense Reimann K. (1,2), Knobloch J. (1,3), Klotz L.O. (2), Rüther U. (1) Thalidomide, a potent treatment for multiple myeloma and leprosy, is also a teratogen causing defects in embryogenesis, resulting in defects such as such as microphthalmia or limb truncations (amelia, phocomelia) in humans [1]. Mice, but neither humans nor chicken, are resistant to the teratogenic properties of thalidomide. It was previously demonstrated that thalidomide-induced limb truncations in the chicken model system require the formation of reactive oxygen species (ROS), suppression of Wnt and Akt survival pathways and induction of apoptosis during early embryonic limb development [2, 3]. Here, we describe the role of glutathione-based antioxidant defense in thalidomide-sensitivity. In line with the species specificity of thalidomide teratogenicity, the drug induces apoptosis in embryonic fibroblasts of humans (HEFs) and chickens (CEFs) but not in those of mice (MEFs). MnTBAP, a superoxide dismutase (SOD) mimetic, but not cell- permeant catalase, abrogates thalidomide-induced cell death in HEFs and CEFs, suggesting that superoxide formation is involved in thalidomide toxicity. We sensitized MEFs for thalidomide by pre-treatment with the naphthoquinone derivative juglone or by depletion of cellular glutathione by application of diethyl maleate (DEM). The addition of thalidomide to such pre-treated MEFs induces apoptosis to an extent comparable to that found with thalidomide-treated CEFs and HEFs. In line with either higher levels of ROS generated or with glutathione depletion being an efficient means of sensitizing MEFs for thalidomide, the glutathione levels found in unstimulated cultured MEFs were higher than in thalidomide-sensitive CEFs and HEFs. From these data, we conclude that thalidomide resistance depends on the capacity of the intracellular antioxidant defense. [1] Smithells RW, Newman CG (1992), J Med Genet. 29, 716-23. [2] Knobloch J, Shaughnessy JD Jr, Rüther U (2007), FASEB J. 21, 141021. [3] Knobloch J, Schmitz I, Götz K, Schulze-Osthoff K, Rüther U (2008), Mol Cell Biol. 28, 29-38. 1. Entwicklungs- und Molekularbiologie der Tiere (EMT), Heinrich-Heine-Universität Düsseldorf, 2. Institut für umweltmedizinische Forschung (IUF) an der Heinrich-HeineUniversität Düsseldorf, 3. Department of Pneumology, University of Cologne

336 The cardioprotectant dexrazoxane depletes topoisomerase II beta from cultured cardiomyocytes and mouse heart tissue Yan T. (1), Deng S. (1), Gödtel-Armbrust U. (1), Wojnowski L. (1) Doxorubicin (DOX) is one of the most successful anti-cancer drugs, but its clinical application has been limited by irreversible cardiotoxicity. Dexrazoxane (DRZ) is the only approved drug which verifiably protects cancer patients against DOX-induced cardiotoxicity, but its precise cardioprotective mechanism is incompletely understood. Since both DOX and DRZ are inhibitors of topoisomerases II, we investigated the effect of DRZ on the expression of topoisomerase II alpha (TOP2A) and beta (TOP2B) in a tumor-derived cell line, in cultured cardiomyocytes (H9C2 cells), and in hearts from B6 mice. In tumor cells, which express both isoforms, DRZ depleted TOP2A and TOP2B by distinct mechanisms affecting transcript level and proteasome-mediated degradation, respectively. In cardiomyocytes and in the hearts DRZ depleted TOP2B, the predominant cardiac TOP2 isoform. This is a first demonstration of a DRZ-induced depletion of TOP2B in vivo. These results suggest that TOP2B depletion by DRZ may protect heart by reducing DOX-induced and TOP2B-mediated DNA damage and apoptosis. We are currently investigating the determinants of the proteasome-dependent TOP2B degradation by DRZ using cells expressing chimeric TOP2A-TOP2B proteins. 1. Dept. of Pharmacology, Johannes Gutenberg University, Mainz, Germany

337 No elevated genomic damage in children and adolescents with attention deficit/hyperactivity disorder after methylphenidate therapy Stopper H. (1), Kämpf K. (1), Artamonov N. (1), Romanos M. (2), Oli R.G. (1), Wirth S. (2), Warnke A. (2), Gerlach M. (2), Walitza S. (1, 3) Background and objective: Attention-deficit/hyperactivity disorder (ADHD) is the most frequent psychiatric disorder in children and adolescents and is often treated with methylphenidate (MPH), resulting in MPH exposure in more than 1% of all children in many countries. A 2005 report on cytogenetic effects in peripheral lymphocytes from 12 ADHD children treated for 3 months with MPH raised questions about its genetic toxicity and potential carcinogenicity. In 2007 we described no elevated micronucleus frequency in 21 children after 3 months of MPH-treatment; since the difference between the two studies could not be explained we now enlarged the overall sample size, and added a healthy control group, a new chronically treated group and positive control slides. Furthermore, micronuclei were analyzed in a second tissue, buccal mucosa. Study participants: A healthy control group (23 individuals), a chronically MPH-treated (>12 months) group (21 patients), and a drug naïve group of ADHD-affected children (26 patients), which was analyzed again after 3 months (17 patients) and 6 months (11 patients), provided samples for analysis of micronucleated lymphocytes. With inclusion of 14 previously obtained blood samples, an overall group size of 31 patients was reached for the comparison of the 3 months observation time with before for micronucleated lymphocytes. For buccal mucosa cells, an additional inclusion of ten more chronically treated patients (no lymphocytes donated) yielded sample numbers of 22 (healthy), 17 (chronically treated), 23 (ADHD drug naïve), 14 (3 months) and 11 (6 months). Results: No significant alteration in genomic damage as seen as micronucleus frequency in peripheral lypmphocytes or buccal mucosa cells was detected after MPH treatment. Conclusions: No indication for genomic damage induced by MPH was obtained in this study, as in our previous study. Together with our previous study, our overall number of MPH-treated patients is now 68 (30 chronically treated, 38

339 Kinetic in vitro analysis of the eligibility of plasma butyrylcholinesterase as therapeutic marker in oxime-treated organophosphorus poisoning Aurbek N. (1), Thiermann H. (1), Worek F. (1) Accidental, suicidal or intentional exposure to organophosphorus compounds (OP) requires rapid and effective antidotal treatment. OP cause inhibition of cholinesterases by phosphylation (denotes phosphorylation and phosphonylation) of their active site serine resulting in a generalized cholinergic crisis due to the inability of OP-inhibited acetylcholinesterase (AChE) to hydrolyze acetylcholine and successive accumulation of this neurotransmitter in synaptic clefts. OP-AChE complexes can be subject to different reactions, i.e. spontaneous reactivation, reactivation by nucleophilic substances like oximes and aging by dealkylation of the OP-AChE conjugate. Post-exposure standard antidotal treatment of OP-poisoning includes administration of atropine as muscarinic antagonist for symptomatic and oximes (obidoxime or pralidoxime) as reactivators of OP-inhibited AChE for causal treatment. Limitations of oxime induced reactivation are aging of OP-inhibited AChE, which prevents a further oxime induced reactivation, high poison load which prevents net reactivation of OP-inhibited AChE by reinhibition of reactivated AChE and limited efficacy of oximes in case of poisoning by certain OP. If reactivation of the OP-inhibited AChE is possible oxime administration should be continued until the OP plasma concentration drops under a critical value allowing reactivation, or until aging is complete. For therapeutic monitoring erythrocyte AChE was shown to be an adequate surrogate parameter for synaptic AChE, ideally the determination of erythrocyte AChE activity, reactivatability of OP-inhibited AChE by the administered oxime and the inhibitory potency of patient plasma (cholinesterase status) gives meaningful information for the optimization of oxime therapy. At present, cholinesterase status is not a standard laboratory parameter. Therefore, plasma butyrylcholinesterase (BChE) is in use for monitoring of OP poisoning and the therapeutic benefit of oxime administration. Kinetic analyses of interactions between AChE, BChE, selected OP and therapeutically relevant oximes revealed substantial differences between erythrocyte AChE and plasma BChE. Hence, monitoring of BChE activity as diagnostic parameter in oxime treatment of OP-poisoning is considered to be of limited value. 1. Bundeswehr Institute of Pharmacology and Toxicology, Munich, Germany

340 Establishment of an in vivo system aiming at reducing organ side effects of antineoplastic therapeutics Ostrau C. (1), Hülsenbeck J. (1), Fritz G. (1) Many antineoplastic agents, e.g. doxorubicin, cisplatin, and ionizing radiation (IR), target the transformed as well as the healthy cell. Specific organ side effects include cardiotoxicity (doxorubicin), nephrotoxicity (cisplatin), haematotoxicity, and hepatotoxicity. Doxorubicin is an intercalating agent which also inhibits the topoisomerase II, a key enzyme of DNA synthesis. Cisplatin is an alkylating agent, causing both intra- and interstrand crosslinks in the DNA. IR causes DNA single and double strand breaks. The three agents are used in the treatment of various cancers, e.g. sarcoma (doxorubicin), squamous cell carcinoma (cisplatin). Middle-term aim of our studies is to identify compounds having beneficial impact on the side effects of these

70

anticancer agents. To this end, adequate treatment regimens and readouts have to be identified. Here, we applied doxorubicin, cisplatin, and IR in a subchronic in vivo model. To this end, mice were treated with doxorubicin (3 x 3 mg/kg, i.p.), cisplatin (5 x 5 mg/kg, i.p.), or IR (2 x 2.5 Gy). Three weeks after the first treatment, mice were sacrificed. Blood (haemotoxicity) and serum (cardio-, nephro-, or hepatotoxicity) samples were analyzed. From the heart, liver, and kidney mRNA was extracted to check for the expression of inflammatory and fibrotic markers by real-time PCR. Doxorubicin treated animals exhibited elevated plasma levels of troponin I as well as GPT and GLDH, indicative of heart and liver damages, respectively. The expression of the inflammatory markers IL-6 and C-reactive protein was up-regulated in the heart. The fibrosis marker CTGF was up-regulated in heart and liver. Cisplatin treated animals exhibited a reduced number of blood platelets, indicative of haematotoxicity. Furthermore, the expression levels of TNFα, IL-1α, and IL-10 as well as KIM1 were elevated in the kidney, indicating inflammation and kidney damage, respectively. Animals treated with ionizing radiation exhibited a reduced number of blood platelets. Furthermore, a minor up-regulation of CTGF was observed in the liver. In conclusion, under experimental conditions, the side effects of doxorubicin were cardio- and hepatotoxicity, of cisplatin nephro- and haematotoxicity, and of IR haematotoxicity with minor hepatotoxicity. Further studies will be conducted to determine compounds that ameliorate these side effects. Initial results obtained applying the HMG-CoA-reductase inhibitor lovastatin will be presented and discussed. 1. Dept. of Toxicology, Johannes Gutenberg University, Mainz, Germany

341 In-vitro biocompatibility testing of degradable alloys based on magnesium on murine fibroblasts L929 Stahl J. (1), Niedorf F. (1), Bäumer W. (1), Schumacher S. (1), Krause C. (2), Seitz J.M. (2), Bach F.W. (2), Kietzmann M. (1) In the last decade, many studies were conducted to develop degradable implants for use both in hard and soft tissue. Different investigations lead to conclusion that metal based on magnesium represents an acceptable material, by reason that these alloys show good degradation behaviour and low corrosion resistance in electrolytic, aqueous environments. The goal of the present study was to examine the influence of metal ions released from different magnesium based alloys on cell viability of murine fibroblasts L929: Mg(pure) (100% magnesium), MgCa0.8 (magnesium + 0.8% calcium), AX30 (magnesium + 3% aluminium + <1% calcium), AL30 (magnesium + 3% aluminium + <1% lithium), AL33 (magnesium + 3% aluminium + 3% lithium), AL36 (magnesium + 3% aluminium + 6% lithium). Furthermore, the impact of different metal salt solutions (ingredients of the examined alloys) on cell viability was studied: MgCl2, CaCl2, AlCl3, LiCl. In-vitro biocompatibility of different magnesium based alloys was studied after degradation in 1ml RPMI medium over 5 days. For this purpose, the cells were incubated with the alloy extract for 24 hours. Furthermore, cells were incubated for 24 hours with different concentrations of metal salt solutions (0.01-100 mmol/l) and cell viability was measured using tetrazolium assay XTT and neutral red assay. AL30, AL33 and AL36 lead to a decrease in cell viability of 16-35% compared to control, while AX30 lead to cell viability of 22-25%. Mg(pure) resulted in cell viability of 52-59%, and MgCa0.8 of 84-91%. A good tolerance was found for the metal salt solution up to the following concentrations: MgCl2 and LiCl 10 mmol/l, AlCl3 3 mmol/l, and CaCl2 1 mmol/l. The results of this in-vitro degradation experiment in combination with the cell viability testing of the metal salt solutions may be used for selection of promising magnesium based alloys, while further investigation is needed to study the influence of metal salt combination on cell viability. These studies are supported by the Deutsche Forschungsgemeinschaft (DFG: SFB 599, R1). 1. Institute of Pharmacology, Toxicology and Pharmacy, University of Veterinary Medicine Hannover, Foundation, Hannover, Germany, 2. Institute of Materials Science, Leibniz University of Hannover, Germany, Hannover

342 Effect of magnesium alloys on dendritic cell function Feser K. (1), Kietzmann M. (1), Krause C. (2), Bach F.W. (2), Bäumer W. (1) Degrading magnesium alloys are new materials for implants used in orthopaedic and trauma surgery. The use of degradable implants in the surgery presupposes an unrestricted biocompatibility with complete degradation in the body. That means alloys have to be absolutely inert for the organism. In this context, the immune system plays an important role. Foreign materials which confront the organism are recognized and evaluated by the immune system. Dendritic cells (DC) are the major antigen presenting cells and thus play a pivotal role for the initiation of a specific immune response or the induction of tolerance. The aim of this study was to investigate the influence of degradable magnesium alloys on the cell function of murine bone marrow derived DC. The study was performed with 5 different alloys. MgP (pure magnesium), MgCa 0.6 (0.6% calcium), MgCa 0.8 (0.8% calcium), MgCa 1.0 (1% calcium), MgCa 1.2 (1.2% calcium). Degradation media were made out of the 5 alloys by incubation over three days in cell culture medium. In parallel, DC were incubated with increasing concentrations (0.1-10 mmol/l) of magnesium chloride and calcium chloride, respectively. The alloys with higher calcium portions showed substantial variability in the magnesium concentration of the degradation medium. Read outs for DC function were viability (annexin V/propidium iodide assay), DC migration (transwell system), DC activation (TNFα secretion) and T cell activation (mixed lymphocyte reaction (MLR)). LPS stimulation and incubation with etoposide were performed as positive controls. Incubation of DC with degradation media over 6 days had no influence on cell viability and only marginal influence on DC migration. Also, the production of TNFα was not enhanced by incubation with degraded magnesium alloys. The MLR showed that there was also no increase of the T cell proliferation in comparison to untreated controls. The positive controls led to expected reactions and activation in DC. As the incubation with increasing concentrations of magnesium and calcium chloride did also fail to influence the above mentioned parameters it can be concluded that degradable magnesium alloys do not have influence on the function (impairment or activation) of dendritic cells. Supported by the Deutsche Forschungsgemeinschaft (SFB599, R1). 1. Dept. of Pharmacology, Toxicology and Pharmacy, University of Veterinary Medicine Hannove, Germany, 2. Institute of Materials Science, Leibniz University of Hanover, Germany

343 Cadmium disrupts epithelial barrier of a human alveolar co-culture model from the basal side Papritz M. (1), Pohl C. (2), Wübbeke C. (1), Kehe K. (1), Thiermann H. (1), Kirkpatrick C.J. (2) Cadmium (Cd2+) is an industrial and environmental pollutant that produces a wide variety of cytotoxic and metabolic effects. Beside oxidative stress, inhibiting mitochondrial ATP production, triggering apoptosis and altering gene expression there is one major effect targeting especially epithelial cells. Recent studies have shown that Cd2+ displaces Ca2+ from its extracellular binding sites on cadherins which results in the breakdown of the adhering junctions. This is soon followed by the disruption of the tight junctions, which results in the overall breakdown of the epithelial barrier function. Our current study shows a protective effect of intact epithelial tight juctions against apically applicated cadmium in a human in vitro co-culture model of the alveolo-capillary barrier. The lung epithelial cell line H441, possessing alveolar type II cell characteristics, was cultured with the endothelial cell line IsoHas as bilayer on a 24-well HTS-Transwell filter plate. The model was exposed to varying concentrations of CdCl2 to the apical and for comparison also to the basolateral side. During exposure, structural barrier integrity of co-cultures was monitored by measurement of transbilayer electrial resistance (TER) over a period of 6 h. After that, functional barrier was tested by transport assays followed by the determination of cytosolic adenylate kinase released by necrotic cells.The results markedly differed between the two ways of exposure. While soon after 4 hours the basolateral stimulated samples showed a 50% reduction of TER, the apical stimulation caused this rate of TER reduction only after 6 hours. This effect is also dose dependent, because at lower basolateral concentrations of Cd2+ the TER breakdown began much earlier in comparison to the apical application. Both further experiments, transport assays and immunofluorescence microscopy, support the observation of epithelial barrier breakdown while apoptosis didn’t seem to play a important role in the first hours after Cd2+ exposure.We confirm the previously described effect in porcine renal and in canine kidney epithelial cells of cadmium application at both sides of an epithelial barrier in a human cell culture model. Our results suggest that the epithelial tight junctions may play a major role of protection from inhalated cadmium ions. 1. Bundeswehr Institute of Pharmacology and Toxicology, Munich, Germany, 2. Institute of Pathology, Johannes Gutenberg University, Mainz, Germany

344 The metalloestrogen cadmium exerts dose and route dependent hormonal acticity in female rats Höfer N. (1), Diel P. (1), Wilhelm M. (2), Wittsiepe J. (2), Degen G.H. (3) The toxic heavy metal cadmium (Cd) is now regarded as potential endocrine disruptor: Cd exerts estrogen-like activity in vitro and can elicit typical estrogenic responses, e.g. uterine weight increases in rodents upon intraperitoneal (i.p.) injection [1]. But, estrogenicity of Cd has not been documented so far with other, more relevant routes of exposure, although it is known that Cd absorption and distribution in the body is strongly affected by the application route. Thus, we investigated the hormonal potency of Cd in ovariectomized Wistar rats after per os administration (50 to 4000 µg/kg bw on 3 days by gavage and 400 to 9000 µg/kg bw for 4 weeks in drinking water) or with i.p. injection of Cd (0.05 to 2000 µg/kg bw). Uterus wet weight and modulation of estrogen-regulated gene expression, i.e. uterine complement C3, were determined, and Cd content in uterus and in other organs was measured by atomic absorption spectrometry. The analysis revealed pronounced differences in Cd tissue levels and hormonal potency for the two routes of administration: A single i.p. injection of Cd led to a dose dependent increase in uterine weight, significant at 500 and 2000 µg/kg bw. Interestingly, low doses of Cd (0.05-50 µg/kg bw) down-regulated C3 mRNA level, whereas the highest dose (2000 µg/kg bw) increased C3 expression in the uterus. Other than i.p. injection, oral dosing of Cd (by gavage or in drinking water) did not increase uterine wet weights. But, both 3 day and 4 week oral Cd administration resulted in a dose dependent stimulation of C3 expression in the uterus, significant at and above 500 µg/kg bw/day. In summary, our data confirm an estrogenic effect of Cd on the uterus upon i.p. injection and show that it can affect molecular mechanisms of estrogenicity. Oral short and long-term administration of Cd did not affect uterine weight. But, on the molecular level, an induction of estrogen sensitive genes was seen in the uterus, albeit at dose levels far exceeding those resulting from dietary exposure in humans. [1] Takiguchi M, Yoshihara S (2006) New aspects of cadmium as endocrine disruptor. Environ Sci 13:107-116. We thank the DFG graduate college 1427 for financial support. 1. Inst. für Kreislaufforschung und Sportmedizin, Deutsche Sporthochschule Köln, Köln, Germany, 2. Abt. für Hygiene, Sozial- und Umweltmedizin, Ruhr-Universität Bochum, Bochum, Germany, 3. IfADo, Leibniz Research Centre for Working Environment and Human Factors, Universität Dortmund, Dortmund, Germany

345 Influence of metallic platinum nanoparticles on the cellular redox status of human colon carcinoma cells Pelka J. (1), Gehrke H. (1), Marko D. (1) Most modern cars are equipped with a three-way catalytic converter emitting platinum as Pt-nanoparticles. These nanostructured particles accumulate at traffic hot-spots and might enter the food chain as a contaminant. In contrast to platinum-halogen compounds, little is known about the toxicological relevance of nanostructured metallic Pt-particles. Our studies on the toxicological relevance of Pt-nanoparticles were focused on the modulation of the redox status. The aim of the present study was to answer the question whether metallic platinum particles are able to invade human colon carcinoma cells (HT29) and therefore influence biological molecules. Glutathione (GSH) in its reduced form plays a crucial role in the cellular defence and therefore, the intracellular GSH level is used for the assessment of the cellular redox status. The study was carried out with special emphasis on the dependence of the toxicological effects on the mean size of the respective particles. In this study HT29 cells were incubated with different preparations of platinum particles (Pt > 100 nm, Pt < 100 nm, Pt < 20 nm) for 3h and 24h. The cellular uptake of platinum particles in HT29 cells was analysed by electron microscopy. The potential influence of platinum nanoparticles on cellular GSH levels was investigated according to Tietze. Furthermore, ROS formation within the cells

71

was measured by the fluorescent DCF assay. Pt-particles < 20nm were found to induce the suppression of the intracellular GSH level in a concentration range of 0.1-10 ng/cm2 after 3h of incubation. By elongating the incubation time up to 24h, effects of Pt-particles < 20 nm on the cellular GSH level were strongly enhanced up to a reduction of 80%. In contrast, platinum particles > 100 nm showed no GSH depleting properties in the tested concentration range. All platinum preparations, tested in the DCF assay showed no increase in the relative fluorescence, suggesting no ROS formation within the cell. In summary, we found that platinum nanoparticles are internalised by human colon carcinoma cells (HT29). Hence, they affect the cellular glutathione level. The observed effects correlate with the particle size in an inverse manner such that the cellular effects are enhanced with a decrease in the particle size and an increase of the incubation time. However, the mechanism by which the tGSH level is affected seems not to be related to the generation of reactive oxygen species by the platinum particles. 1. Section of Food Toxicology, Universität Karlsruhe (TH), Germany

346 Absorption and metabolism of curcumin in the Caco-2 Millicell® system Dempe J.S. (1), Scheerle R.K. (1), Pfeiffer E. (1), Metzler M. (1) Curcumin (CUR) is the yellow pigment from the rhizomes of the Asian plant Curcuma longa. In addition to antioxidative and antiproliferative properties, CUR exhibits anticarcinogenic effects, especially in the gastrointestinal tract. However, little is known about the absorption and metabolism of CUR in the intestine. Therefore, we have studied curcumin in the Caco-2 Millicell® system, an established in vitro model for intestinal absorption and metabolism. Caco-2 cells are derived from a human colorectal carcinoma. When cultured for 3 weeks, these cells undergo spontaneous enterocytic differentiation to form a monolayer of highly polarized cells connected by functional tight junctions, with organized microvilli on the apical membrane. This differentiated monolayer, grown in the insert of the Millicell® system, separates the cell culture medium into two compartments: the apical side represents the intestinal lumen and the basolateral side the portal blood. Moreover, several drug-metabolizing enzymes are expressed in the differentiated cells. The Millicell® system with Caco-2 cell monolayers can therefore be used to study the uptake, metabolism and transport of xenobiotics across the intestinal epithelium. When CUR was incubated with differentiated Caco-2 cells in a normal cell culture dish at 37°C for 6 h, little free CUR but significant amounts of CUR-glucuronide, the reductive metabolites hexahydro-CUR and octahydro-CUR, and their glucuronides and sulfates were detected in the medium by HPLC/DAD. When Caco-2 cells grown in the Millicell® system were apically incubated with 50 µM CUR for 1 to 6 h, the glucuronides and sulfates of hexahydro-CUR and octahydro-CUR were observed in the basolateral compartment. In contrast, the unconjugated reductive metabolites were predominantly excreted into the donor compartment. Apical incubation with 50 µM hexahydro-CUR under the same conditions confirmed that the free reductive metabolites moved to the apical side, whereas their conjugates were delivered to the basolateral side. The apparent permeability coefficient, calculated from the appearance of CUR on the receiver side, indicated a low intestinal absorption of CUR. In summary, our study conducted in the Caco-2 Millicell® system suggests that CUR is poorly absorbed but very efficiently metabolized in the intestine in vivo. This behaviour of CUR in the intestine should have a bearing on its biological effects in other organs. 1. Chair of Food Chemistry, Karlsruhe Institute of Technology (KIT), Karlsruhe, Germany

347 Anthocyanins act as catalytic topoisomerase inhibitors in human colon carcinoma cells Esselen M. (1), Hutter M. (1), Fritz J. (1), Marko D. (1) Anthocyanins represent a class of coloured plant constituents, which occur in many fruits and vegetables of the daily diet. We have previously reported that the anthocyanidin delphinidin acts as a catalytic inhibitor, preventing the formation of the transient DNA-topoisomerase intermediate, the so-called cleavable complex in the human colon carcinoma cell line HT29. However, little is known so far about the impact of anthocyanin-enriched food or food supplements on DNA-integrity in vitro. In the present study the question was addressed whether consumer relevant anthocyanin-rich extracts possess inhibitory properties comparable to their free aglycones and whether they affect the stabilisation of the covalent enzyme/DNA intermediate induced by classical topoisomerase I and II poisons. The so called “isolating in vivo complexes of enzyme to DNA” (ICE) bioassay was performed to detect the level of topoisomerase covalently linked to DNA in cellular systems. To support our hypothesis that anthocyanin extracts act as catalytic inhibitors in HT29 cells, we included merbarone (MER) in our testing. Treatment of HT29 cells for 1 h with the topoisomerase I poison camptothecin (CPT, 100 µM) or the topoisomerase II poison doxorubicin (DOX, 10µM) clearly enhanced the level of topoisomerase covalently bound to DNA. In contrast, the catalytic inhibitor MER did not affect the extent of DNA/topoisomerase binding. Furthermore none of the used anthocyanin-rich extracts affected the level of cleavable complexes in HT29 cells. The above mentioned results indicate that anthocyanins target topoisomerases prior to their binding to the DNA. To extend this hypothesis, the ICE bioassay was used for competition studies in cultured HT29 cells. Pre- and co-incubation with a blackberry extract significantly suppressed the level of the CPT- or DOX-induced DNA/topoisomerase intermediates. In contrast, a bilberry extract and a grape seed extract did not affect the CPT-stabilised topoisomerase I/DNA complexes and exhibited only marginal effects on the amount of topoisomerase II complexes induced by DOX. In summary, anthocyanins represent a class of potent catalytic topoisomerase inhibitors, preventing the formation of the DNA-topoisomerase-intermediate in HT29 cells. These results demand further investigations whether an enhanced intake of anthocyanins for example by respective food supplements, might counteract the effectiveness of therapeutic topoisomerase poisons in vivo. 1. Inst. of Applied Biosciences, Universität Karlsruhe (TH), Karlsruhe, Germany

348 Methylation of quercetin enhanced its toxicity in HepG2 cells Ruhl S. (1, 2), Rohrig R. (1), Chovolou Y. (1), Kampkötter A. (1), Kahl R. (1), Proksch P. (2), Wätjen W. (1) The beneficial health effects of a nutrition rich in vegetables have been linked to flavonoids. Therefore, these substances became increasingly popular as food supplements notwithstanding that only little is known about uptake, accumulation, metabolism and bioactivity of high doses of flavonoids so far. We investigated the uptake and biotransformation of natural occurring flavonoids in different cell lines since these effects are essential to examine possible beneficial or adverse effects on human health of these substances. The flavonol quercetin was taken up rapidly in human hepatoma cells (HepG2) to an extent of 2 ng/µg protein. However, even longer incubation times did not significantly increase the amount of intracellular quercetin. In contrast to that finding, the amount of detected metabolites increased by time. The most prominent metabolite isorhamnetin, a methyl derivative of quercetin, accumulated in growth media up to 15% of the given flavonoid. To examine the pharmacological effects of this methylation we performed a comparative analysis. The methyl derivative showed a decreased ability to act as an antioxidant when compared with quercetin (TEACassay). Nevertheless, it was still more potent than the positive control trolox (a synthetic vitamine E derivative) in this assay. On the other hand, the methylated derivative isorhamnetin showed a significant higher toxicity than quercetin in the MTT-assay at 50 µM. This increase in toxicity was at least in part mediated by the activation of caspases. Isorhamnetin showed a four fold higher activation of caspase-3/7-activity (ApoONEassay) in comparison to quercetin. The results of our investigations showed that metabolism of flavonoids could alter their pharmacological properties and should be taken into account when estimating effects of flavonoids from diet or supplement. 1. Inst. of Toxikology, Heinrich-Heine University, Düsseldorf, Germany, 2. Inst. of Pharmaceutical Biology and Biotechnology, Heinrich-Heine University, Düsseldorf, Germany

349 Influence of quercetin on insulin/insulin like growth factor signaling (ILS) and the transcription factor DAF-16 Bosbach H. (1), Sack M. (1), Chovolou Y. (1), Wätjen W. (1), Kahl R. (1), Kampkötter A. (1) Flavonoids are secondary plant metabolites that are present in several herbs, fruits and vegetables. It is discussed that they have a positive influence on human health, especially on age-related diseases. Although many pharmacological properties of flavonoids have been discovered in in vitro studies, the knowledge about the mechanisms underlying the beneficial health effects in vivo is quite limited. The execution of animal experiments or human studies is very expensive and timeconsuming and thus such attempts are not appropriate to examine a large variety of different flavonoids. In order to overcome this limitation we employed the small soilliving nematode Caenorhabditis elegans as a model to investigate the effects of these substances in a complex, multicellular organism. Many key discoveries in bio-medical research resulted from experiments/studies with this worm and it is increasingly accepted as a model in various scientific areas. There are astonishing homologies between C. elegans and mammals as well in biological functions as in molecular processes. In previous studies we demonstrated that quercetin, a main flavonoid found in many foods and beverages of plant origin, is capable to prolong the adult life span of C. elegans and to enhance the resistance to stress. In C. elegans ageing and response to stress as well as metabolism are known to be regulated by different mechanisms, of which the insulin/insulin growth factor like signaling (ILS) is the most prominent one. This phosphorylation cascade keeps the transcription factor DAF-16, the homologue of mammalian FoxO proteins, inactive by retaining it in the cytosol. Amongst others, activation of DAF-16 can occur by absent ILS and the MAP kinase JNK-1. Based on the observation that quercetin caused a translocation of DAF-16 into the nucleus we have started to elucidate its function and the role of upstream pathways in the protective effects of this flavonoid. In addition, we have begun to investigate an involvement of the transcription factor SKN-1 in the beneficial quercetin actions since this homologue of the mammalian Nrf2 is known to function in the p38 MAP kinase pathway to regulate the oxidative stress response and in ILS to influence life span. 1. Institute of Toxicology, Heinrich-Heine University, Düsseldorf, Germany

350 Molecular effects of apigenin in human cancer cells Grasmück T. (1), Kampkötter A. (1), Wätjen W. (1), Kahl R. (1), Chovolou Y. (1) Flavonoids are polyphenolic compounds that occur ubiquitously in various fruits and vegetables. The flavonoid apigenin exhibit diverse biological effects like cell cycle arrest, growth inhibition and apoptosis in various cancer cells. Therefore, apigenin is expected to be a promising antitumor agent. However, the molecular effects of apigenin have not been fully elucidated. In this study we investigated the cellular and molecular effects of apigenin on four different human cancer cell lines. Apigenin up to 100µM exerts only moderate cytotoxicity in human hepatoma cell lines (Huh-7, HepG2) and in human colon cancer cell line (Hct-116). Only in human breast cancer cells (MCF-7) noticeable apigenin cytotoxicity could be measured at concentrations exceeding 50µM. Tumor necrosis factor (TNF)-related apoptosis-inducing ligand (TRAIL) is a member of the TNF family of cytokines, which can induce apoptotic cell death in a variety of tumor cells while sparing most normal cells. However, some cancer cells are resistant to TRAILinduced apoptosis and it is therefore important to find compounds that can sensitize cancer cells to overcome this resistance. We therefore investigated whether apigenin is able to sensitize Hct-116 and MCF-7 cells to TRAIL-induced apoptosis. Cell death was markedly enhanced in both cell lines when treated with apigenin and TRAIL together. Apigenin significantly sensitized Hct-116 and MCF-7 to TRAIL-induced apoptosis as measured by Caspase 3/7 activation. Furthermore, Hct-116 cells were sensitized by apigenin against tumor necrosis factor-alpha (TNF-α) mediated cytotoxicity. In contrast to other reports this sensitising effect of apigenin is not linked to an inhibition of the NFκB signalling pathway. Rather, the sensitizing effect of apigenin is accompanied by a dose-dependent decrease of the MEK1/2 phosphorylation and subsequent inactivation of the mitogen activated protein kinase (ERK) which is linked to proliferation and cell

72

survival. These data suggest that at least parts of the sensitizing effect of apigenin may be mediated by its ability to modulate the MAPK cascade. 1. Institute of Toxicology, Heinrich-Heine-University, Düsseldorf, Germany

351 Modulation of Nrf2-mediated gene expression by dietary flavonoids Rohrig R. (1), Ruhl S. (1,2), Schmitz R: (1), Bergermann A. (1), Chovolou Y. (1), Kampkötter A. (1), Kahl R. (1), Wätjen W. (1) Flavonoids have become increasingly popular in terms of health protection and therefore are used in food supplements at relatively high doses. Although some flavonoids act as powerful antioxidants it was also shown that, in high concentrations, they also can generate reactive oxygen species (ROS). ROS can modulate gene expression of phase II detoxifying enzymes via the antioxidant responsive element (ARE) which is found in the promoter region of these genes. Nrf2, as a member of the bZIP transcription factor family, transcriptionally activates the ARE and thereby protecting cells against oxidative stress. It has been published that some flavonoids induce the expression of the quinone reductase and HO-1 indicating that they may act through the activation of the ARE. We investigated the activation of the ARE (luciferase reporter gene assay) by structurally related flavonoids (quercetin, luteolin, kaempferol and apigenin) in comparison to the well-known inductor tBHQ using Hct116 human colon carcinoma cells. We detected an approximately 20-fold induction of the ARE after incubation with 100µM quercetin for 24 h. This induction was even higher than the effect of tBHQ. To confirm that the activation of the ARE is dependent on Nrf2, we performed western blot analysis and found a high nuclear accumulation of Nrf2 in cells treated with tBHQ compared to control cells.Furtheron we asessed a slight induction of the ARE downstream target HO-1. In summary, this study provides data showing activating effects of distinct flavonoids on the ARE. This may be also important to estimate effects of flavonoids in applied in high doses as food supplements. 1. Institute of Toxicology, Heinrich-Heine-University, Düsseldorf, Germany; 2. Institute of Pharmaceutical Biology and Biotechnology, Heinrich-Heine-University, Düsseldorf, Germany

352 Effects of catechin and its derivates in the model organism C. elegans Tanner S. (1), Chovolou Y. (1), Wätjen W. (1), Kahl R. (1), Swidergall M. (1), Kampkötter A. (1) Catechin and its derivates are present in high doses in green tea as well as in many other foods and beverages of plant origin. There is good evidence that these green tea catechins have beneficial effects on human health by protecting against degenerative diseases and cancers. We employed the small soil-living nematode C. elegans to examine the impact of green tea catechin and to elucidate their molecular mode of action since this well etablished model allows the investigation of effects on organismal level as well as on molecular level in living animals with comparatively little effort. The tested green tea catechins (+)-catechin, (–)-epicatechin, (–)-epicatechin gallate, (–)epigallocatechin and (–)-epigallocatechin gallate prolong the life span of the worms by 10-20%. The extension of life span was accomponied by a delay in the accumulation of the autofluorescent pigment lipofuscin that is regarded as a biomarker for ageing in C. elegans. The resistance against thermal and oxidative stress of the worms is also increased by some of the catechins. In order to examine the molecular mechanisms of these protective impacts we monitored the influence of the catechins on the transcription factors DAF-16 and SKN-1 that are homologs to the mammalian FoxO proteins and Nrf2, respectively. Both transcription factors are known to have crucial functions in the regulation of life span and response to stress. All catechins cause a translocation of DAF-16 from the cytosol into the nucleus – a necessary prerequisite for its transcriptional activity. The subcellular distribution of the stress-responsive transcription factor SKN-1 is also influenced by some of the catechins. Several catechins are capable to partially reverse the translocation of SKN-1 into the nucleus induced by exposure to oxidative stress. Although the catechins have been shown to affect the life span and stress resistance on the one hand and the subcellular distribution of transcription factors on the other hand a causal link between these two observations has not been demonstrated yet. Thus, we want to explore in ongoing studies whether the protective impact of the catechins on organismal level depends on the activity of DAF-16 and SKN1 on molecular level. 1. Institute of Toxikology, Heinrich-Heine University, Düsseldorf

353 Toxic effects of phenolic compounds isolated from the seeds of Psoralea corylifolia Limper C. (1), Wang Y. (2), Chovolou Y. (1), Kampkötter A. (1), Ruhl S. (1,2), Rohrig R. (1), Proksch P. (2), Kahl R. (1), Wätjen W. (1) Use of herbal remedies in the treatment of various widespread diseases has a long tradition in Asia. The legume plant Psoralea corylifolia is widely applied in traditional Ayurvedic and Chinese herbal medicine due to its e.g. diuretic, purgative, stomachic and antipyretic properties. We isolated 10 known phenolic compounds (flavonoids, isoflavonoids, chalcones) from the seeds of Psoralea corylifolia and also 3-(2,4dihydroxy)-phenyl-7-hydroxychromen-2-one, a so far not published structure was found. In this study we investigated the cytotoxic potential of these substances in different cancer cell lines (H4IIE rat hepatoma cells, Hct 116 human colon carcinoma cells, C6 rat glioma cells) to analyse a possible use of the substances as cytostatic drugs. Therefore we first used the MTT-assay to determine cytotoxicity. The substances showed moderate cytotoxicity: bavachromene and isowighteone revealed the highest effects (EC50-values of approximately 20 µM) in these cell lines. We further examined the mechanism of cell death: xanthoangelol, 6-prenylnaringenin, bavachromene, corylin, 3(2,4-dihydroxy)-phenyl-7-hydroxychromen-2-one, wighteone and isowighteone caused activation of caspase-3/7 in H4IIE cells consistent with induction of apoptosis (APOone assay). On the other hand, lactate dehydrogenase (LDH)-assay showed no significant increase of LDH in the medium suggesting that these substances do not cause necrosis in H4IIE hepatoma cells. Analysing the effects of these substances on intracellular

signaling pathways we found no effects on NF-κB (SEAP reporter gene assay after stimulation with TNFα). Further on effects on MAPK signaling and a possible antioxidative potential are investigated. We conclude that especially bavachromene and isowighteone isolated from the seeds of Psoralea corylifolia showed potent cytotoxic activities through induction of caspase 3/7 mediated apoptosis and may lead to development of new lead structures for anticarcinogenic drugs. 1. Inst. of Toxicology, Heinrich Heine University, Düsseldorf, Germany, 2. Inst. of Pharmaceutical Biology and Biotechnology, Heinrich Heine University, Düsseldorf, Germany

354 Dammarenolic acid sensitizes cancer cells to tumor necrosis factor–related apoptosis-inducing ligand (TRAIL) Kröger A. (1), Proksch P. (2), Kampkötter A. (1), Wätjen W. (1), Kahl R. (1), Chovolou Y. (1) Dammarenolic acid ((20S)-hydroxy-3,4-seco-4(28),24-dammaradien-3-carboxylic acid) is a triterpenoid isolated from various Aglaia species, a plant which is mainly occurring in Southeast Asia. In traditional Chinese medicine extracts of Aglaia serve as treatment of cough, cold and inflammation. Several structure related triterpenoids like ursolic acid have been shown to suppress tumorigenesis and to induce apoptosis in a wide variety of cancer cells. Furthermore triterpenes are considered to be promising candidates for sensitizing cancer cells to chemotherapeutics. So far there are only limited data available about the effects of dammarenolic acid on cancer cells. The aim of this work was therefore to characterize the cytotoxic potential of dammarenolic acid on various cancer cell lines and to assess the potential of dammarenolic acid to sensitize cancer cells to TRAIL-mediated toxicity. Human colon cancer cells (Hct-116) and human breast cancer cells (MCF-7) were treated for 24h with various concentrations of dammarenolic acid and cell viability was assessed by neutral red assay and MTT assay. In both cell lines dammarenolic acid exhibited dose-dependant cytotoxicity with EC50 values (µM) of 9 and 11, respectively. The loss of cell viability after dammarenolic acid treatment was characterized by the appearance of cells displaying apoptotic morphology and increased caspase 3/7 activity. TRAIL is a promising candidate for cancer treatment due to its selective toxicity against cancer cells. However, some cancer cells are resistant to TRAIL and it is therefore of great interest to find strategies that can sensitize cancer cells to TRAIL-induced apoptosis. For that reason MCF-7 cells, resistant to TRAILinduced apoptosis, were preincubated with non-toxic concentrations of dammarenolic acid for 4h with a subsequent treatment with various concentrations of TRAIL for 24h. Dammarenolic acid strongly enhanced TRAIL-induced cytotoxicity as measured by MTT assay. These findings suggest that the combined treatment of dammarenolic acid and TRAIL may be a useful strategy for sensitizing cancer cells but further studies are needed to explore the molecular mechanism. 1. Institute of Toxicology, Heinrich-Heine-University, Düsseldorf, Germany, 2. Institute of Pharmaceutical Biology and Biotechnology, Heinrich-Heine-University, Düsseldorf, Germany

355 Effects of marine natural compounds in mammalian cells: Investigation on intracellular signalling pathways, apoptosis and oxidative stress Wätjen W. (1), Rohrig R. (1), Ruhl S. (1), Böddeker S. (1), Limper C. (1), Schmitz R. (1), Bergermann A. (1), Chovolou Y. (1), Kampkötter A. (1), Proksch P. (2), Kahl R. (1) Marine organisms are a rich source of chemically unique secondary metabolites. The pharmacological potential of these compounds is increasingly recognized e.g. for tumor therapy. We investigate toxicological aspects of marine compounds to identify lead structures as potential anticancer drugs. Therefore we analysed different natural compounds derived from marine organisms, e.g. cyclic peptide derivatives named kahalalides from a Sacoglossan mollusc; polyprenyl-1,4-hydroquinone derivates from zoobenthos-inhabiting sponges; brominated pyrrole alkaloids from Stylissa sponges and anthraquinones from marine crinoids. We determinate cellular/molecular targets of these substances using a set of different assays performed in mammalian cell lines. Cellular toxicity of the substances is determined using MTT reduction assay. For example, distinct kahalalides show strong toxicity in C6 and MCF7 cells (IC50-value in nanomolar range), but not in H4IIE cells. We discriminate between apoptotic and necrotic cell death since the mode of cellular death is an important parameter for anti-cancer drugs (prevention of inflammatory responses). Induction of apoptotic cell death was confirmed e.g. for distinct anthraquinone derivatives. We analyse if the substances modulate oxidative stress in the cells systems using the DCF assay. Prooxidant effects of the substances are furthermore detected by analysis of the antioxidant-response-element (ARE)-pathway. Induction of this pathway (activation of Nrf-2, activation of ARE and induction of ARE-mediated gene expression) was analyzed for avarone/avarol derivates as well as for anthraquinones. Since misregulated gene expression is a basic cause of many diseases including cancer, we further identify natural compounds that specific modulate distinct signaling pathways e.g. MAP-kinase signaling pathway using Western blot experiments with phospho-specific antibodies. Further screening of other kinases modulated by marine products was also performed using a Kinase Microarray. In conclusion, analysis of different natural compounds isolated from marine organisms showed that these substances show partly prominent cytotoxic activities against various cancer cell lines, mediated to some extend by induction of apoptotic cell death and accompanied by an induction of oxidative stress and disturbance of intracellular signalling pathways. 1. Inst. of Toxicology, Heinrich-Heine-University, Düsseldorf, Germany, 2. Inst. of Biological Pharmacology and Biotechnolology, Heinrich-Heine-University, Düsseldorf, Germany

73

356 Interferons sensitise malignant melanoma cell lines to the alkylating agent temozolomide by reactivating the death-receptor (Fas/CD95) apoptosis pathway Jöst E. (1), Fritz G. (1), Kaina B. (1), Roos W.P. (1) Patients diagnosed with malignant melanoma have a poor prognosis. First line therapy includes the use of methylating agents like temozolomide (TMZ). To increase the responsiveness of the tumours towards the drugs, IFNs are used as chemo-sensitizers. The mechanism underlying the increased sensitivity of melanomas to TMZ, in combination with IFNs, is still unclear. To elucidate possible mechanisms, different p53 wild-type (wt) and p53 mutant (mt) melanoma cell lines were compared as to their sensitivity towards TMZ in the presence and absence of IFNs. Both, IFN-alpha and -beta significantly increase the apoptotic response, with IFN-beta showing the stronger effect. These results were confirmed by a higher activation of caspase-3 and -7. Transfection with O6-methylguanine-DNA methyltransferase (MGMT) cDNA completely abolished the sensitisation effect of IFN-beta, showing that IFN-beta sensitise cells to the killing effect of the DNA damage O6- methylguanine (O6-MeG). The increased sensitivity could be due to an influence of IFN-beta on the repair or processing of O6-MeG, or IFN-beta could have an effect on the apoptosis efficiency of melanoma cells. As IFN-beta had no influence on the expression pattern of the repair enzyme MGMT, the mismatch repair (MMR) proteins MSH2, MSH6, PMS2 and MLH1 or the homologous recombination (HR) proteins Rad52 and Rad51, nor the amount of DSBs arising from the processing of the O6-MeG lesion, it was concluded that IFN-beta enhances the TMZ mediated apoptotic response. Using DN-FADD (dominant negative Fas-associated death domain protein) expressing clones and a Fas-activating antibody in combination with TMZ treatment we show that the sensitisation effect of IFN-beta is due to reactivation of the death-receptor (Fas/CD95) pathway. This reactivation is caused by an IFN-beta provoked induction of caspase activity which was verified using TRAIL. Collectively, these data show that pretreatment with IFN-beta can sensitise p53 wt melanoma cell lines to TMZ to a greater extend than p53 mt cells. The sensitisation is not due to an influence on DNA damage nor DNA-repair. Rather, the data point to an IFN-beta induced reactivation of the deathreceptor (Fas/CD95) pathway in p53 wt cells by up-regulation of caspase activity. 1. Department of Toxicology, Johannes Gutenberg University, Mainz, Germany

357 Activation of caspase-2 downstream of caspase-9 is a part of the amplification loop in topotecan-triggered apoptosis in p53 deficient mouse fibroblasts Tomicic M.T. (1), Christmann M. (1), Kaina B. (1) In a previous study we reported that p53-/- mouse embryonic fibroblasts (MEFs) are highly sensitive to topotecan (TPT) due to abrogated degradation of topoisomerase I. In contrast, isogenic wt and apaf-1-/- MEFs are more resistant. Here we analyzed the molecular pathway responsible for TPT-induced apoptosis. In p53-/- cells, the abrogated degradation of topoisomerase I was followed by increased formation of stalled replication forks and DNA double-strand breaks as indicated by the phosphorylation of H2AX. Since caspase-8 was not activated in any of the cell lines, we assume that not the receptor/ligand but the mitochondrial damage pathway is activated during exposure of MEFs to TPT. This was also supported by a strong cytochrome c release from mitochondria and by activation of caspase-9 in wt and p53-/- cells. In apaf-1-/- cells neither caspase-9 nor caspase-3 was activated, again underlining the importance of the mitochondrial apoptosis pathway. In addition to activation of caspase-9 and -3, also caspase-2 was significantly activated in p53-/- cells, leading to cleavage of Bid protein. Neither caspase-2 nor Bid was cleaved in apaf-1-/- cells, indicating that caspase-9 is the most apical caspase. This was also proven by the specific caspase-9 inhibitor (zLEHD) that reduced overall TPT-triggered apoptosis in p53-/- cells to 90%. Co-treatment of p53/cells with TPT and the specific caspase-3 inhibitor (zDEVD) completely abolished caspase-2 activation and Bid cleavage, showing that caspase-2 is only a part of the amplification loop of the death signal. Surprisingly, although caspase-9 and caspase-3 were cleaved to a similar extent in wt and p53-/- cells, only caspase-3 in crude extracts of p53-/- cells showed significant activity. This was due to proteasomal degradation of the IAP proteins, XIAP and “survivin”, in p53-/- cells, which normally block active caspase-3. In accordance to this, down-regulation of XIAP and “survivin” using siRNA approach increased TPT-triggered apoptosis in wt cells. The work was supported by DFG. 1. Dept. of Toxicology, Johannes Gutenberg University, Mainz, Germany

was no clear correlation between the β-catenin status and apoptotic response of the various cell types used. Therefore, further studies are required to unravel the molecular mechanisms underlying the anti-apoptotic effect of GSK-3β inhibition and the possible role of β-catenin in this process. This study was supported by the DFG (SFB 773, Tübingen). 1. Abteilung für Toxikologie, Eberhard Karls Universität, Tübingen, Germany, 2. Innere Medizin I, Universitätsklinikum, Tübingen, Germany

359 Reduced intestinal tumorigenesis and induction of Bak-dependent apoptosis upon interference with histone deacetylase 2 Constantin M.C. (1), Huber E. (2), Zimmermann S. (2), Spiller C. (2), Göttlicher M. (2) Histone deacetylases (HDACs) are promising targets in cancer therapy since their inhibition induces cell death and growth arrest in many types of cancer cells. They are easily inhibited by microbial or synthetic small molecules. At the same time, the family of HDACs serves key functions in regulating local chromatin structure, transcriptional repression, and function of non-histone proteins. Thus, the balance between interference with essential physiological functions of HDACs and the beneficial inhibition in cancer cells is critical for successful therapeutic intervention. At present it is poorly understood which HDAC isoforms have to be inhibited to induce cell cycle arrest or apoptosis in cancer cells without substantial side effects on physiologically required functions. We now analyzed HDAC2 deficient mice and cells to assess the impact of HDAC2-deficiency on physiological development, intes-tinal tumorigenesis, and control of apoptosis. HDAC2 deficient mice are viable and fertile but smaller than their wildtype litter mates. Crossing with mice prone to develop intestinal tumors due to the APCmin mutation revealed reduced tumor rates in HDAC2 mutant compared to HDAC2 wildtype mice. Inhibition of HDACs by valproic acid or knockdown of HDAC2 by siRNA in HT29 colon cancer cells induced apoptosis through the intrinsic Bcl2 family-dependent pathway of apoptotic signaling. Induction of Bak mRNA by HDAC inhibition was required for HDAC inhibitor-induced apoptosis and per se sufficient to induce cell death. Thus, survival of colon cancer cells and Bak expression appear to specifically depend on the HDAC2 isoenzyme. HDAC2 might therefore serve as target for the development of isoenzyme selective HDAC in-hibitors and Bak expression to monitor individual responses to therapeutic interventions. Supported by the European Commission (CASCADE NoE and CRESCENDO IP) and the DFG (SFB456) 1. Institute of Toxicology and Environmental Hygiene, Technische Universität München, Germany, 2. Institute of Toxicology, Helmholtzzentrum München, Germany

360 Defining the substrate recognition of eEF1A by Legionella pneumophila glucosyltransferase Lgt1: implications on reaction mechanism of glucosylation Belyi Y. (1), Stahl M. (2), Aktories K. (2) Legionella pneumophila is a facultative intracellular pathogen that replicates in alveolar macrophages and epithelial lung cells thereby causing a severe pneumonia known as Legionnaires’ disease. The bacterium produces a multitude of effector proteins that are secreted to the host via a Type IV secretion system and subvert many host cell processes to allow intracellular replication of the pathogen. A novel virulence factor of Legionella is the recently identified glucosyltransferase Lgt1 that transfers a glucose molecule to the eukaryotic elongation factor 1A (eEF1A) leading to an inhibition of protein synthesis and ultimately death of target cells (Belyi et al. PNAS, 2006). In an effort to characterize the interaction of Lgt1 with its substrate, we shortened eEF1A in successive rounds and could show that most parts of eEF1A are dispensable for substrate recognition. Even very short peptides containing the acceptor serine-53 were efficiently glucosylated by Lgt1. Systematic exchange of single amino acids in the shortest peptide that reacted in the glucosylation assay disclosed distinct amino acid residues that were essential for substrate recognition by Lgt1. A comparison of the kinetics of the glucoslyation revealed that short substrates are glucosylated by Lgt1 at a higher rate than the full length eEF1A. The shortest peptide was used to inhibit the glucosylation of the purified native substrate eEF1A competitively; it also led to a relief of the Lgt1 mediated inhibition of translation in rabbit reticulocyte lysate. 1. Gamaleya Research Institute, Moscow, Russia, 2. Institute of Experimental and Clinical Pharmacology and Toxicology, Albert-Ludwigs-Universität, Freiburg, Germany

358 Does Wnt/β-catenin signalling interfere with cytostatica-induced apoptosis in mouse hepatoma cells? Rosenwald E. (1), Schwarz M. (1), Bitzer M. (2), Buchmann A. (1) Mutations in the β-catenin gene are frequently observed in human and murine liver tumors indicating that activation of Wnt/β-catenin signalling in tumor cells provides a selective growth advantage, e.g. by stimulation of cell division or inhibition of apoptosis. To study the role of β-catenin in regulation of apoptosis, different mouse hepatoma cell lines expressing either wild-type or a constitutively active form of β-catenin were used to compare their apoptotic response after treatment with the cytostatic drugs etoposide and sorafenib. Induction of apoptosis was modulated by incubating cells with the GSK-3β inhibitors lithium and SB 216763 which are commonly used to activate β-catenin signalling, and knock-down experiments with β-catenin siRNA were performed to further clarify the role of β-catenin in the apoptotic process. Apoptosis was analysed by staining of apoptotic nuclei, determination of DNA laddering and measurement of caspase 3/7 activity which was strongly increased after 24 hours of incubation with the cytostatic drugs. If β-catenin signalling was activated by treatment with lithium or SB 216763, the apoptotic effect provoked by etoposide was significantly reduced whereas sorafenibinduced apoptosis was unaffected. Lithium was found to inhibit GSK-3β by inducing serine-9 hyperphosphorylation which in turn led to activation of β-catenin, as demonstrated by reporter gene assays and RT-PCR analysis of selected Wnt/β-catenin target genes. Although these results may suggest that suppression of etoposide-induced apoptosis by the GSK-3β inhibitors is mediated via activation of β-catenin signalling, the role of β-catenin is not clear for the following reasons: (1) the anti-apoptotic effect of lithium was still present after β-catenin knock-down with specific siRNA, (2) suppression of apoptosis was also observed in β-catenin mutated cells in which all GSK-3β phosphorylation sites in the β-catenin protein are eliminated by deletion and (3) there

361 Pasteurella multocida toxin activates anti-apoptotic signaling pathways Preuß I. (1), Hildebrand D. (2), Orth J.H.C. (1), Kubatzky K. (2), Aktories K. (1) The mitogenic Pasteurella multocida toxin (PMT) possesses potential carcinogenic properties. PMT affects several signal transduction pathways related to carcinogenesis by constitutively stimulating heterotrimeric G proteins. The toxin activates the small GTPase RhoA, the MAP kinase ERK and STAT proteins via the stimulation of two G protein families, Gαq and Gα12/13. PMT action also results in an increase in inositol phosphates, which is due to the stimulation of PLCβ via Gαq. Recent studies indicate that PMT additionally activates Gαi to inhibit adenylyl cyclase. PMT stimulates not only Gα-dependent signaling, but also Gβγ signaling, leading to activation of phosphoinositide 3-kinase (PI3K). The PI3K pathway plays a crucial role in cell growth and survival and is activated in various cancers. The stimulation of downstream effectors like the proto-oncogene Akt can cause a disruption of balance between proliferation and apoptosis. Here we show that treatment of HEK293 cells with PMT leads to PI3K-dependent phosphorylation of Akt. A prerequisite of apoptosis are activated caspases. We measured the caspase 3/7 activity as a read-out for apoptosis, which was induced by staurosporine. PMT, but not the inactive mutant toxin, inhibits staurosporine-mediated apoptosis. Therefore PMT-induced signaling pathways may contribute to the potential carcinogenic properties of the toxin. 1. Institut für Experimentelle und Klinische Pharmakologie und Toxikologie, AlbertLudwigs-Universität, Freiburg, Germany, 2. Institut für Medizinische Mikrobiologie und Hygiene, Universitätsklinikum, Heidelberg, Germany

74

362 DNA damage assessed as DNA strand breaks in leucocytes from induced sputum of asphalt workers: a pilot study Marczynski B. (1), Raulf-Heimsoth M. (1), Pesch B. (1), Kendzia B. (1), Vosshans B. (1), Borowitzki G. (1), Lee E.-H. (1), Welge P. (1), Käfferlein H.U. (1), Brüning T. (1) The aim of this pilot study was to assess DNA strand breaks in cells obtained from induced sputum in mastic asphalt workers which were exposed to fumes of bitumen at high temperatures. Additionally, we compared the results to those recently derived in blood lymphocytes and interleukin-8 (IL-8) in induced sputum. Blood and induced sputum samples were collected from 43 workers (age 22 to 58 years) before and after working shift. The cellular profile of these induced sputum samples consisted mostly of epithelial cells (83-96%) and leucocytes [mostly neutrophils (3-12%), but also macrophages]. In this study we focused on DNA strand breaks in leucocytes of induced sputum which were measured by alkaline Comet assay (visual scoring with classifying in the five categories system) and compared with OTM (Olive Tail Moment, alkaline Comet assay) in blood lymphocytes. IL-8 was determined in cell-free supernatant of induced sputum by an immunoassay as a marker of inflammation in the upper airways. In leucocytes from induced sputum of all workers the DNA damage did not change during shift (arbitrary score: 197 pre-shift, 203 post-shift). However, in blood lymphocytes a higher level of DNA strand breaks (P = 0.02) was measured before compared to after shift. Comparable to the blood results, in induced sputum samples of nonsmokers (n = 15) considerable but not significant higher level of DNA strand breaks was quantified before shift (arbitrary score: 190 pre-shift vs. 80 post-shift). A non-significant monotone association (rs = 0.28, P = 0.07) between DNA strand breaks in induced sputum leucocytes and blood lymphocytes was found in post-shift samples. Before as well as after shift there was a significant correlation between DNA strand break level in sputum leucocytes and IL-8 (pre-shift: rs = 0.31, P = 0.047; post-shift: rs = 0.43, P = 0.004). This pilot study demonstrated that the investigation of DNA damage in sputum leucocytes is possible and offers a non-invasive method to assess genotoxic effects of airborne toxicants in the respiratory tract as primary target organ. In induced sputum the comparison of genotoxic and inflammatory markers is possible. To our knowledge, this is the first study that compares genotoxic effects in sputum and blood cells which were investigated simultaneously. The assessment of genotoxic effects in cells of induced sputum would be a promising improvement in the research of lung diseases. 1. BGFA - Research Institute of Occupational Medicine, German Social Accident Insurance, Ruhr University Bochum, Bochum, Germany

363 DNA single strand breaks (SSB) inhibit transcription even if located in the nontranscribed strand Lingg T. (1), Khobta A. (1), Epe B. (1) It is well established that many kinds of DNA lesions interfere with transcription. For example, bulky lesions such as pyrimidine dimers block RNA polymerase II if located in the transcribed strand and then initiate preferentially transcription-coupled repair. DNA single strand breaks (SSB) are expected to interfere with elongation at the transcribed genes as well. However, the effects of SSB in different DNA strands have not been studied. Recently, DNA strand specific nicking endonucleases have been engineered. They provide a convenient tool to investigate the effect of SSB on gene expression. We have constructed reporter plasmids containing a single site for restriction endonuclease Bpu10I and introduced strand specific nicks in either coding or non-coding strand of the green fluorescent protein (GFP) gene with the help of nicking endonucleases Nb.Bpu10I or Nt.Bpu10I, respectively. The plasmids were co-transfected with a reference plasmid encoding for dsRed-Monomer into mouse embryonic fibroblasts (MEF) derived from the wild-type and DNA repair deficient mice. GFP expression was measured after 24 hours and the levels of transcription of the nicked and covalently closed plasmids were compared. In all tested cell lines, the GFP expression levels from the nicked plasmids were at least 30% lower than those of covalently closed plasmids. Interestingly, plasmids containing a nick in the non-transcribed strand always showed an even lower gene expression than the plasmids with a nick in the transcribed strand. The decrease of the gene expression was particularly strong in poly(ADP-ribose)polymerase-1 (PARP1) knockout MEFs that are deficient in SSB repair. We conclude that SSB strongly inhibit transcription, especially when the PARP-1 dependent repair pathway is inactivated. It has to be noted that the effect on transcription does not require the presence of SSB in the transcribed strand. Thus, termination of transcription at the DNA lesion site is not fully responsible for the observed effects. Rather, a down-regulation of the gene expression by damage-induced signalling might account for the decrease of transcription. 1. Institute of Pharmacy, Johannes Gutenberg University of Mainz, Mainz, Germany

364 Impaired transcription at DNA damage sites: taking advantage of constructed DNA lesions Khobta A. (1), Kitsera N. (1), Lingg T. (1), Epe B. (1) A vast variety of environmental and endogenous factors can cause DNA damage in cells and tissues. However, in most cases it is impossible to attribute the observed toxic effects exclusively to the lesions generated in DNA, since various kinds of damage are induced at the same time in different organelles and biomolecules in the cell. Transfection of the damaged DNA into undamaged cells revealed a decreased transcription of the damaged DNA as a potentially toxic effect that specifically arises from the genotoxic lesions, such as oxidised bases. Contrary to the previously established opinion, our data demonstrate that the impaired transcription of the damaged genes cannot be fully explained by the mechanical blocking of elongating RNA polymerases at the lesion sites. First, oxidised guanine bases had influence on the gene expression only under the condition of their intracellular processing; the effect of DNA damage on transcription was not immediate and therefore indirect. Second, transcription of the reporter gene was strongly inhibited in the presence of very small amounts of oxidised guanines or strand breaks, despite the low probability of the damage to occur within the transcribed DNA strand. To address a mechanism for the decreased transcription more specifically, we have next constructed the plasmids incorporating specific DNA lesions, such as transcription blocking single strand breaks (SSB) and a

mutagenic modified base 7,8-dihydro-8-oxoguanine (8-oxoG). These lesions were introduced in defined positions with respect to the transcribed gene. With the help of the constructed DNA lesions, we show that the same type of damage can cause different physiological effects, dependently on location of the damage site with respect to the functional DNA elements. We conclude that site-specific damage in DNA provides a sufficient signal for down-regulation of transcription of the damaged genes, regardless of the damage suffered by the cell as a whole. 1. Institute of Pharmacy, Johannes Gutenberg University, Mainz, Germany

365 DNA oxidation by a dimethacrylate and an epoxy monomer Eckhardt A. (1), Pongratz S. (1), Gerstmayr N. (1), Hiller K.-A. (1), Schmalz G. (1), Schweikl H. (1) Monomers of dental resin composites induce oxidative stress in mammalian cells. Genotoxic effects caused by dental monomers could be the results of this DNA damage. In the present study we tested whether the dimethacrylate monomer triethyleneglycol dimethacrylate (TEGDMA) and the epoxy monomer 3,4-epoxycyclohexylmethyl-3,4epoxycyclohexane-carboxylat (K-126) induce oxidative DNA damage in V79 cells. Therefore, the production of 8-oxoguanine (8-oxoG) was quantified. Cells were incubated for 2, 24, and 48h with TEGDMA (1 and 3mM) or K-126 (0-3mM) with or without addition of 10mM N-acetylcysteine (NAC), a scavenger of reactive oxygen species. The intracellular production of 8-oxoG was measured with a FITC-labelled antibody in a flow cytometer (FACS). Analyses of significance (n=4, medians) were performed using the Mann-Whitney-U test (α=0.05). After 2h of incubation only 3mM K126 induced a 3-fold increase in the production of 8-oxoG compared to untreated controls. The increase after this short period of incubation was reduced by 50% by NAC. K-126 generated a concentration-dependent increase in 8-oxoG after 24h. The amount of 8-oxoG was 15 times higher in cultures treated with 1mM K-126 compared to untreated cells. The effect of K-126 after 48h was weaker than after 24h. However, DNA oxidation induced by K-126 after 24h was not significantly reduced by NAC. Both TEGDMA concentrations induced a 2-3-fold increase in 8-oxoG after 24h. After 48h 1mM TEGDMA resulted in a tenfold increase. NAC inhibited the DNA oxidation in V79 cells after a 24h exposure to 1mM TEGDMA. These results suggest that the genotoxic effects caused by dimethacrylates like TEGDMA and epoxy monomers like K-126 are at least partly a consequence of oxidative DNA damage. This project was supported by the Deutsche Forschungsgemeinschaft (DFG, Schw 431/11-1). 1. Dept. of Operative Dentistry and Periodontology, University of Regensburg, Regensburg, Germany

366 Effects of heating on free crystalline silica content, appearance, and biological activity of devitrified alkaline earth insulation wools Ziemann C. (1), Class P. (2), Bellmann B. (1), Creutzenberg O. (1), Brown R. (3) Alkaline earth silicate wools (AES) as defined by CAS RN 436083-99-7 are available on the market as soluble, low persistence alternative (exonerated from CRM classification under directive 97/69/EC) to refractory ceramic fibers, which were classified as carcinogens under European legislation. During prolonged periods at high temperatures vitreous insulation wools may convert into a number of crystalline products including some types of crystalline silica (CS). In the present study, 3 AES (SW607, SW607 max, SW607 HT; shotted and ball milled) designed for use at different temperatures were heated for various times at temperatures capable of rapidly causing deterioration [Classification temperature (BS EN 1094)] and at 150°C below that to simulate typical maximum use temperature. Types and quantities of CS formed and fiber size distribution and shape were then determined by x-ray diffraction and electronmicroscopy, respectively. Unheated AES contained no free CS, but CS increased depending on heating temperature. In addition, heating resulted in decreased fiber length and changes in AES shape. Subsequently, the genotoxic (alkaline and hOGG1modified comet assay) and membrane-damaging potential (LDH-release) of the AES samples were investigated in primary rat alveolar macrophages, as a first site of contact for fibers in the lung. Cells were incubated in vitro for 2 h with 200 µg/cm2 of the wools. All unheated and heated AES, except one particle-like heat product, slightly increased DNA-damage and significantly induced LDH-release, as compared to controls. DNAstrand break induction was maximal with AES, which had been heated for 4 weeks. LDH-release, and thus the membrane-damaging potential of the wools, decreased with heating and seemed to correlate best with the number of fibers in the fiber length fraction of > 20 µm. Both DNA-damage and LDH-release did not correlate with free CS and were mostly not CS-specific, as judged by the use of aluminum lactate (AL, 10µM), a known quencher of CS-specific biological effects. Slight oxidative DNA-damage was observed with some samples. As expected, the positive control quartz DQ12 significantly induced DNA-damage and LDH-release and this effect was clearly counteracted by AL. In conclusion, heating of AES mediates changes in free CS content and fiber length/shape. For biological activity of AES and their heat-products changes in fiber length/shape seems to be crucial, whereas free CS appears to be of minor relevance. 1. Genetic Toxicology and Inhalation Toxicology, Fraunhofer ITEM, Hannover, Germany, 2. Thermal Ceramics de France S.A., Wissembourg, France, 3. Toxicology Services, Stretton, Oakham, UK

367 Effect of single and clustered 7,8-dihydro-8-oxoguanine residues on the transcription of transfected plasmid DNA Kitsera N. (1), Khobta A. (1), Epe B. (1) 7,8-Dihydro-8-oxoguanine (8-oxoG) is an important premutagenic oxidative lesion that can potentialy mispair with adenine and lead to G·C to T·A transversions. Previously, we have reported that oxidative guanine modifications randomly generated by exposure to photosensitiser MB plus light cause gene inactivation that is enhanced in Csbm/m mouse embryo fibroblasts but attenuated in absence of Ogg1. Now we have studied the effect of single and clustered 8-oxoG in defined positions within the transcribed strand of

75

the green fluorescent protein (GFP) gene. The transcribed strand was nicked by endonuclease Nb.BsrDI and degraded by the exonuclease III. Then the transcribed strand was reconstituted by extension of a modified oligonucleotide containing synthetic 8-oxoG, and the ligated plasmids were purified from agarose gels. Modified plasmids were co-transfected into wild–type and the repair deficient cells together with an undamaged reference plasmid encoding for dsRed-Monomer. The data show that single synthetic 8-oxoG in the transcribed strand do not cause a significant decrease of transcription. The same transcription levels were observed with the 8-oxoG incorporated in two different positions. Furthermore, a cluster of three 8-oxoG residues positioned in the transcribed strand and spaced by two normal bases did not affect transcription. The results demonstrate that 8-oxoG does not constitute a direct transcriptional block. In the same time, we do not exclude a potential effect of 8-oxoG localised in other positions (e.g., non-transcribed strand) within the plasmid DNA. 1. Institute of Pharmacy, Johannes Gutenberg University, Mainz, Germany

368 Resveratrol protects in vivo against endogenous oxidative DNA damage Fußer M. (1), Lv Q. (2), Li H. (2), Epe B. (1) Resveratrol is a grape ingredient with a wide variety of cellular and in-vivo effects. Invivo protection of the cardiovascular system, neuroprotection and an inhibitory potential in different stages of tumor development have been reported. The purpose of this study was to test whether the basal (steady-state) levels of oxidative DNA damage in repair deficient Ogg1-/-/Csb-/- double knockout and wild type mice can be modulated by a longterm dietary supplementation with resveratrol. The Ogg1-/-/Csb-/- mice have been shown to accumulate 7,8-dihydro-8-oxoguanine (8-oxoG) and mutations in their livers in an age-dependent manner and are therefore an appropriate tool to analyze effects on DNA damage without additional application of a stressor. OGG1 is the main glycosylase for the removal of 8-oxoG by the base excision repair pathway and Cockayne syndrome B protein (CSB) is part of the transcription coupled nucleotide excision repair. We fed the different mice strains several months with 0.04% resveratrol ad libitum and measured the steady-state levels 8-oxoG in the livers after 3, 6 and 9 months. Isolation of the hepatocytes was carried out by a two-step collagenase perfusion technique. The cells were subsequently analyzed for the levels of oxidative DNA base modifications by a modified alkaline elution with the bacterial formamidopyrimidine glycosylase as a probe. We also tested the primary hepatocytes for their resistance against oxidative stress either by trypan blue exclusion or by induction of single-strand breaks after hydrogen peroxide. Resveratrol decreases the basal levels of 8-oxoG in the hepatocytes of Ogg1/-/Csb-/- and wild type mice after few months of treatment by approximately 20%. The cells isolated from resveratrol-treated animals were also found to be more resistant to oxidative stress induced by hydrogen peroxide. To test for the reason of the reduced 8oxoG levels by resveratrol we measured mRNA of antioxidant genes with quantitative Real-Time PCR. Several antioxidant response genes (SOD1, SOD2, GPX, HO1) were highly up-regulated, which could be an explanation for the protection of DNA against endogenous generated reactive oxygen species. Altogether the results indicate that dietary supplementation with resveratrol or maybe even a resveratrol rich diet has beneficial effect for health and might protect from DNA damage and mutations. 1. Institute for Pharmacy, Johannes Gutenberg University, Mainz, Germany, 2. Department of Pharmacology, Johannes Gutenberg University, Mainz, Germany

369 Protective effect of the AhR agonist indolo[3,2-b]carbazole against oxidative DNAdamage in Caco2-cells Faust D. (1), Oesch F. (1), Kaina B. (1), Dietrich C. (1) Epidemiological studies suggest that high dietary intake of fruit and vegetables protects against tumour development in many organs including colon. Especially compounds derived from the Brassica genus, e.g. broccoli, cauliflower, brussels spout or Chinese cabbage, appear to be protective against colorectal cancer. Brassica vegetables are rich in glucosinolates which – after physical damage to the plant material - are hydrolysed to a variety of compounds. For instance, during digestion glucobrassicin is hydrolysed to sulforaphane and indole-3-carbinol, which further condensates in the organism to diindolylmethane and indolo[3,2-b]carbazole. In animal studies, a protective effect of indole-3-carbinol against carcinogen-induced tumour development has been demonstrated in various organs including colon. In line with the high instability of indole3-carbinol, these protective effects are probably mediated by the condensation products, e.g. indolo[3,2-b]carbazole (ICZ). However, the underlying molecular mechanisms are not yet known. Interestingly, ICZ is known to be a potent agonist of the aryl hydrocarbon receptor (AhR). In accordance, we could show a strong ICZ-dependent induction of CYP1A1 which could be blocked by downregulation of the AhR by siRNA. Since it is further suggested that the AhR plays a tumour protective role in the gastrointestinal tract, we are interested in the molecular effects of ICZ in the colonic epithelial cell line Caco-2. Using the alkaline Comet assay we could show that pre-incubation with ICZ (1 µM) protects against DNA-damage induced by benzo[a]pyrene (1 and 10 µM) or hydrogen peroxide (30 µM). This suggests that ICZ protects against oxidative stress. In line, pretreatment with ICZ reduces DNA-damage induced by hydrogen peroxide determined as sites sensitive to FPG which mainly detects 8-oxodG. Interestingly, the protein amount of the transcription factor Nrf2, a critical mediator of the anti-oxidant defence mechanisms, is not elevated by ICZ. The underlying molecular mechanisms are currently under investigation. This work was supported by ECNIS, a network of excellence operating within the European Union 6th Framework Program, Contract No 513943. 1. Dept. of Toxicology, Johannes Gutenberg University, Mainz, Germany

370 Mustard intake reduces benzo[a]pyrene diolepoxide- and hydrogen peroxideinduced DNA damage in human blood Lamy E. (1), Schmitz S. (1), Mersch-Sundermann V. (1) Epidemiological studies indicate that lifestyle choices, including high intake of dietary saturated fats, alcohol consumption and smoking play a fundamental role in the aetiology of degenerative diseases such as cancer. On the other hand, there are chemopreventive constituents in food plants such as sulphur-containing isothiocyanates (ITCs) which have been shown to be quite effective in the prevention of tumorigenesis in laboratory animals. Mustard (Sinapis alba) contains ITCs at exceptionally high concentrations in the order of around 1 g/kg fresh weight. However, so far no information is available about the chemopreventive potency of mustard preparations in humans. The effect of mustard intake on ex vivo induced DNA damage was investigated in a pilot human intervention trial. The study was conducted with the approval of the ethics committee of the University of Freiburg and with the informed consent of the participants. Six healthy human volunteers (m = 3; f = 3) were fed 20 g of mustard daily for 4 days. One week before intervention until one week post intervention, participants followed a controlled diet to minimize the influence of other bioactive compounds. Every 12 h, heparinized blood was collected by venipuncture and immediately exposed to 5 µM benzo[a]pyrene diolepoxide (BPDE) or 300 µM hydrogen peroxide (H2O2) for 2 h at 37°C, each in triplicate. Untreated samples were used as control. Plasma samples were prepared by centrifugation at 1000 g for 10 min and analyzed for glutathione Stransferase (GST) alpha activity. Blood chemistries included ALT, AST, APH, cholesterol, HDLC and LDLC. Already 12 h after intervention start, a significant (P < 0.05) antigenotoxic effect against BPDE- and H2O2-induced DNA damage was shown in peripheral blood leucocytes of volunteers who consumed the supplemented diet when compared with blood samples before study begin. No abnormal blood chemistry parameters were observed in volunteers during the whole of the study. A significant increase in GST-alpha activity was seen, however only in plasma of male volunteers after 72 h compared to plasma samples taken before intervention. The results show that intake of ITC-containing vegetables might indeed be associated with reduced cancer risk through a reduction in DNA damage. 1. Dept. of Environmental Health Sciences, University Medical Center Freiburg, Freiburg i. Br., Germany

371 Poly(ADP-ribose) polymerases 1 and 2 are equally sensitive to trans-dominant inhibition by the DNA-binding domain of poly(ADP-ribose) polymerase 1 Beneke S. (1), Heine K. (2), Bürkle A. (1) Poly(ADP-ribosyl)ation is a posttranslational modification carried out by members of the poly(ADP-ribose) polymerase (PARP) family. Using NAD+ as a substrate, these enzymes modify diverse acceptor proteins with this negatively charged polymer (PAR), thus regulating different cellular pathways like replication, transcription, telomere maintenance, vesicle trafficking, DNA repair, and overall genomic stability. Most prominent is PAR formation after DNA damage, executed by PARP-1 and PARP2. PARP1 is in this setting the most active enzyme, responsible for roughly 90% of newly formed PAR. PARP1 and PARP2 are involved in orchestrating base-excision repair, but have also other, partially overlapping cellular functions. Additionally, PARP activity is involved in augmenting ischemia-reperfusion damage. DNA strand breaks induced by this pathophysiological condition over-activate PARP, leading to high consumption of NAD+ and finally energy failure, thus increasing the affected tissue area. Suppressing specifically PARP1 dependent polymer formation may preserve viability of the cells without interfering too much with DNA repair activities. Recent data provided evidence that despite regulating the same DNA repair pathway, DNA binding specificity may be different between PARP1 and 2. Whereas PARP1 recognizes DNA gaps and 5’recessed ends, PARP2 preferentially binds to nicks. We have previously shown that expression of the DNA-binding domain (DBD) of PARP1 suppresses DNA damage induced PARsylation, thus providing a powerful tool to inhibit PARP activity nonpharmacologically. But as the activity of PARP2 is low compared to PARP1, the polymer produced by PARP2 may have escaped detection. In order to test the possibility that DBD expression is selectively inhibiting PARP1 without interfering with PARP2, we used 3T3 fibroblast strains from wild type, PARP1 and PARP2 knockout mice, respectively. We transiently transfected them with a plasmid encoding for the DBD or with a control vector and challenged them with genotoxic agents. We could demonstrate with doubleimmunofluorescence against DBD and polymer that regardless of the genetic status of the cell, DBD expression is able to inhibit PAR formation. Experiments using pan-PARP as well as newly developed more specific inhibitors are currently ongoing. We conclude that despite reported different DNA substrate specificity between PARP1 and PARP2, the DNA-binding domain of PARP1 is able to suppress all DNA strand break dependent PAR formation. 1. Molecular Toxicology, University of Konstanz, Konstanz, 2. Institute of Pharmacology and Toxicology, University of Ulm Medical Center, Ulm

372 Towards a novel mass spectrometric method for the detection of poly(ADPribose) Sass S. (1), Arsic M. (1), Lehmann S. (1), Mangerich A. (1), Bürkle A. (1) Cellular poly(ADP-ribose) (PAR) is a linear or branched biopolymer of various chain lengths with up to 200 ADP-ribose subunits. It is formed by enzymes of the family of poly(ADP-ribose) polymerases (PARP) by using NAD+ as a substrate in a process known as poly(ADP-ribosyl)ation. Poly(ADP-ribosyl)ation represents a posttranslational modification that is involved in the modulation of a variety of cellular functions, such as DNA repair, transcription, and cell death. Most of the current methods for the detection of PAR rely either on radioactive or antibody-based approaches. Yet, radioactive techniques are inconvenient, expensive, and in general limited to in vitro studies. Antibody-based methods are unable to reliably quantify both low and very high levels of cellular PAR. Moreover, these methods give no structural information about PAR chain length and the degree of branching. However, especially such information would be useful to study the physiological role of poly(ADP-ribosyl)ation. Here, we present a potential alternative approach for the quantification and characterization of PAR: PAR

76

was synthesized in vitro and digested by snake venom phosphodiesterase (SVP) and bone alkaline phosphatase (BAP) giving rise to characteristic fragments of ribosyladenosine, diribosyladenosine, and adenosine. Using mass-spectrometry (LCESI-MS), it was possible to detect ribosyladenosine, which is characteristic for linear PAR, with a sensitivity of down to 10 fmol. Moreover, it was possible to trigger PAR formation in cells, and subsequently to isolate the polymer and to detect ribosyladenosine via mass-spectrometry after SVP/BAP digestion. With this study, we provide proof-of-principle for the detection of PAR by mass-spectrometry, which when fully established, would represent a very sensitive and informative method for the characterization of cellular poly(ADP-ribosyl)ation. 1. Molecular Toxicology Group, University of Konstanz, Konstanz, Germany

373 Possible factors for cisplatin sensitivity in testis tumor cells Usanova S. (1), Piée-Staffa A. (1), Sied U. (1), Kaina B. (1), Köberle B. (1) Over 75% of patients with advanced metastatic testis tumors can be cured using cisplatin-based combination chemotherapy. This is unusual as metastatic cancer in adults is usually incurable. Cell lines derived from testis tumors retain this sensitivity to cisplatin in vitro, reflecting the clinical response. For cisplatin, many sensitizing mechanisms have been reported including mechanisms of DNA repair and pathways which promote apoptosis. DNA mismatch repair (MMR) may influence the cytotoxicity of cisplatin. It has been shown that cells with impaired MMR gained resistance towards cisplatin. In vitro studies showed that MMR proteins can bind cisplatin induced DNA lesions. It is assumed that binding of MMR proteins to cisplatin lesions results in futile repair leading to cell death. We therefore investigated the levels of the MMR proteins Msh2 and Msh6 in 3 testis tumor cells (833K, SuSa, GCT27) and 3 bladder cancer cells (MGH-U1, RT112, HT1376). We found higher levels of Msh2 protein in the testis tumor cells compared to the bladder cancer cells. Treatment with cisplatin did not change Msh2 or Msh6 levels in the cancer cell lines. These data suggest that higher intrinsic levels of Msh2 might play a role for cisplatin sensitivity in the testis tumor cells. Clinical data showed that testis tumors rarely exhibit mutations in the p53 gene compared to other types of cancer. We found that our testis tumor cell lines have functional p53. Treatment of the testis tumor cells with cisplatin lead to activation of p53 in a time and dose-dependent manner. To investigate whether there is a correlation between cisplatin sensitivity and p53 status of the testis tumor cells we used siRNA to down-regulate p53 protein. Down-regulation of p53 increased cisplatin resistance in the testis tumor cells suggesting that the sensitivity to cisplatin is due to a propensity to undergo apoptosis. Currently apoptotic factors controlled by p53 are investigated in testis versus bladder cancer cells. Work was supported by DFG grant KO 1732/1-1. 1. Department of Toxicology, Johannes Gutenberg University of Mainz, Germany

374 The DNA repair factor ERCC1-XPF as a possible determinant for repair deficiency and cisplatin sensitivity in testis tumor cells Usanova S. (1), Piée-Staffa A. (1), Sied U. (1), Kaina B. (1), Köberle B. (1) Metastatic testicular germ cell tumors (TGCT) are cured in over 75% of patients using cisplatin-based combination therapy. The reasons underlying the cisplatin sensitivity are not yet known. Cell lines derived from TGCTs retain this cisplatin hypersensitivity in vitro, therefore providing a good model system for investigating the factors controlling cisplatin sensitivity. We previously showed that testis tumor cells have a reduced capacity to remove cisplatin-induced DNA platination suggesting that repair deficiency might be a factor for the observed cisplatin sensitivity. In further studies we found that the nucleotide excision repair (NER) factor ERCC1-XPF is reduced in testis-tumor cell lines indicating a possible role of ERCC1-XPF for repair deficiency and cisplatin sensitivity of testis tumor cells. Cisplatin induces both intrastrand adducts (IA) and interstrand crosslinks (ICLs). To investigate repair of IAs we used the method of DNA slot blotting, repair of ICLs was investigated using the Comet assay. We found that repair of IAs is similar in the testis tumor cell lines 833K and SuSa compared to the cisplatin resistant bladder cancer cell line MGH-U1. Repair of ICLs, however, was significantly reduced in the testis tumor cells compared to the bladder cancer cell line. It is proposed that ICL repair proceeds via the formation of a double strand break, and ERCC1-XPF plays a role in processing ICL-induced DSBs. To visualize DSBs we used immunodetection of histone variant γH2AX. We found that γH2AX foci formed in both bladder and testis tumor cells after cisplatin treatment. However the γH2AX foci persisted in the testis tumor cells confirming impaired ICL repair in the testis tumor cells compared to bladder cancer cells. To analyse the causal role of ERCC1-XPF for repair deficiency and cisplatin sensitivity in testis tumor cells we overexpressed ERCC1-XPF in the testis tumor cell line 833K using a mammalian expression vector. Overexpression of ERCC1-XPF increased ICL repair in 833K cells suggesting that ERCC1-XPF might be rate-limiting for repair of ICLs in testis tumor cells. Investigations about the effect of ERCC1-XPF overexpression on cisplatin sensitivity in testis tumor cells are currently under way. Work was supported by DFG grant KO 1732/1-1. 1. Department of Toxicology, Johannes Gutenberg University of Mainz, Germany

375 Processing of O6-methylguanine adducts to a toxic lesion requires at least two rounds of DNA replication, but cell death can also be induced at subsequent cell cycles Quiros S. (1), Roos W.P. (1), Kaina B. (1) SN1 alkylating agents such as temozolomide and procarbazine are commonly used genotoxic anticancer drugs. The adduct O6-methylguanine (O6MeG) is a critical pre-toxic lesion generated upon exposure of cells to these agents. O6MeG is repaired by the enzyme O6-methylguanine-DNA methyltransferase (MGMT). If not removed from the DNA, O6MeG exerts its toxicity in a mechanism dependent on the mismatch repair system and DNA replication. Nevertheless, the precise mechanism leading to cell death has not yet been fully clarified. Making use of synchronized CHO-9 cells (MGMT negative), we analyzed the events arising from the exposure of the cells to the O6alkylating agent N-methyl-N’-nitro-N-nitrosoguanidine (MNNG), at low SCE inducing

non-toxic dose, and at a high aberration inducing and toxic dose. We could not observe any delay in the progression of cells through the S-phase of the treatment cell cycle in cells exposed to low dose MNNG. At high dose, a delay in the treatment cell cycle Sphase was observed. This delay was O6MeG independent, as it was also evident in cells expressing MGMT. A dose dependent accumulation of cells in G2/M after two DNA replication cycles was observed, which was completely abolished in the MGMT expressing clone. Only at a toxic dose, an O6MeG dependent delay of the progression of cells through the S-phase of the post treatment cell cycle was observed. The emergence of cell death temporally coincided with the beginning of the G2/M block in the second cell cycle following treatment, and continued to increase with time. The apoptotic response was preceded by an O6MeG dependent increase in the number of γH2AX foci per cell, during the S phase of the post treatment cell cycle. These events were also paralleled by a boost in the activation of the ATR-Chk1 signaling cascade. A fraction of the MNNG treated cell population was able to progress through the G2/M block into following cell cycles. Apoptotic cells were detected at the second and subsequent cell cycles. The results support a model were O6MeG lesions have to undergo at least two rounds of DNA replication in order to be converted into a powerful trigger of apoptosis. A fraction of the cells dies from the post treatment cell cycle, but another progress through subsequent cycles, were the accumulation of mis-repaired damage could eventually induce the apoptotic response. 1. Department of Toxicology, Johannes Gutenberg University, Mainz, Germany

376 The role of Brca2 and Xrcc2 dependent homologous recombination in the protection of cells against genotoxicity caused by O6-methylguanine Roos W.P. (1), Nikolova T. (1), Quiros S. (1), Naumann S.C. (1), Kiedron O. (1), Zdzienicka M.Z. (2), Kaina B. (1) O6-methylguanine (O6MeG) is a critical DNA adduct induced by methylating carcinogens and anticancer drugs such as temozolomide, streptozotocine, procarbazine and dacarbazine. O6MeG -triggered genotoxicity is dependent on reversal repair by O6methylguanine-DNA methyltransferase (MGMT) and mismatch repair (MMR) and is executed by a strong induction of apoptosis. One model for O6MeG -induced genotoxicity and apoptosis points to the formation of DNA double-strand breaks (DSBs). If this is true, repair of DSBs would play a key role in the defense against alkylating agent-induced O6MeG lesions. This study was aimed at elucidating the contribution of the main pathways of DSB repair, non-homologous end joining (NHEJ) and homologous recombination (HR), on cellular protection against O6MeG. NHEJ defective cells (DNAPKcs mutated, and Ku80 mutated) showed a marginal increase in apoptosis following O6MeG induction while HR defective cells (Xrcc2 mutated, and Brca2 mutated) showed an 8-11 fold increase in apoptosis compared to wild-type controls. Stable transfection with the repair protein MGMT completely abrogated the apoptotic response in all cell lines used. NHEJ defective cells were able to repair the DSBs arising from the processing of O6MeG lesions while HR defective cells were not. These results also demonstrated that DSBs arising after treatment with alkylating agents is due to O6MeG lesions, as MGMT transfected cells did not display DSB formation. The consequence of the lack of DSB repair in HR defective cells after N-methyl-N’-nitro-N-nitrosoguanidine or temozolomide treatment was a huge increase in chromosomal aberrations compared to wild-type controls, while there was no difference in clastogenicity between NHEJ defective cells and wild-types. Formation of sister chromatid exchanges (SCEs) was dependent on HR and was required for protection against O6MeG-induced genotoxicity. Collectively, the data clearly demonstrate that HR is required for O6MeG tolerance by a mechanism that involves recombinational bypass of O6MeG /T mispaires following genotoxic treatment. 1. Dep. of Toxikology, Johannes Gutenberg University, Mainz, Germany, 2. Dep. of Molecular Cell Genetics, Collegium Medicum in Bydgoszcz, Nicolaus-Copernicus University, Torun, Poland

377 Abrogation of a DNA repair defect in human monocytes during maturation leads to resistance to methylating and oxidizing agents Bauer M. (1), Becker H. (1), Kaina B. (1) Monocytes (Monos), dendritic cells (DCs) and macrophages (Mphs) are key players in the immune response. We studied the sensitivity of Monos and monocyte-derived DCs and Mphs and found Monos to be more sensitive than DCs and Mphs to apoptosis upon exposure to the methylating anticancer drug temozolomide. This hypersensitivity is due to a defect in base excision repair (BER), which removes N-alkylation lesions from DNA. Expression studies of BER proteins revealed that Monos lack the repair proteins XRCC1 and ligase IIIα and also poly(ADP-ribose) polymerase-1 (PARP-1), which binds to and is activated by DNA strand breaks and may be important when BER is initiated from a SSB to protect and trim the ends for repair synthesis. DCs and Mphs express these BER proteins at a high level. An in vitro repair assay showed that in Monos the ligation step in BER is delayed compared to CD34+ stem cells, DCs and Mphs. These findings are confirmed by the human acute monocytic leukemia cell line MM6, which lack XRCC1, ligase IIIα and PARP-1. Treatment of MM6 cells with the cytokines IL-4 and GM-CSF, which induce monocytic maturation into DCs and Mphs in primary cells, causes upregulation of XRCC1, ligase IIIα and PARP-1 and a decrease of sensitivity to methylating agents. The main lesions subjected to BER beside DNA alkylations are oxidised DNA bases, arising spontaneously within the cell, during inflammatory responses, or from exposure to exogenous agents. We examined the sensitivity of Monos, DCs and Mphs to oxidative DNA damage. To investigate this, we treated the cells with tert-butylhydroperoxide (t-BOOH), a short-chain analogue of lipid hydroperoxides that causes oxidative DNA damage and lipid peroxidation in the cell by inducing free-radical formation. We determined a significantly higher apoptotic response in Monos than in DCs and Mphs. Also the amount of DNA SSBs during the treatment of the cells with t-BOOH was much higher in Monos as in DCs and Mphs. The data revealed that human Monos exhibit a BER defect that makes them vulnerable to methylating genotoxins and oxidative DNA damage. This defect becomes abrogated during the maturation of Monos into DCs and Mphs, induced by the two cytokines IL-4 and GM-CSF. Work was supported by SFB 432/B7. 1. Dept. of Toxicology, Johannes Gutenberg University, Mainz, Germany

77

378 Sensitivity of cell lines deficient in homologous recombination or nonhomologous end-joining proteins to the alkylating agent methyl methane sulphonate Nikolova T. (1), Kaina B. (1) Double strand breaks (DSB) are critical lesions in eukaryote cells resulting in cell death or genomic alterations. Eukaryote cells use two main pathways to cope with induced DNA breaks after exposure to DNA damaging agents, e.g. alkylating agents: homologous recombination (HR) and non-homologous end-joining (NHEJ). HR requires members of the Rad51 protein family, Rad52, Rad54, BRCA1, BRCA2, XRCC2 and XRCC3 as well as proteins of the Rad50, NBS1, MRE11 complex. NHEJ requires the DNA-dependent protein kinases, the heterodimer Ku70/Ku80 and the XRCC4/DNA ligase IV. In this study, we investigated how the deficiency for some HR- and NHEJproteins affected the sensitivity of hamster cell lines to the monofunctional SN2 alkylating agent methyl methane sulphonate (MMS). Cell lines deficient for rad51, BRCA2, XRCC2, Ku80 and DNA-PKCs were treated with MMS concentrations in the range of 0.1 to 1 mM and the induced cell death was determined by FACS analysis of subG1 fraction, surviving fraction by colony formation assay, genotoxic effects by analysis of chromosome aberration and sister chromatid exchange frequency. The induced cell cycle delay was analysed by BrdU incorporation. DSB induction was studied by Western blot analysis of phosphorylated HA2X and immunofluorescent analysis of γH2AX foci. Since H2AX phosphorylation is important for recruitment or retention of checkpoint and other repair factors to DSBs, we further studied expression and co-expression of γH2AX with 53BP. Our data showed that cells deficient in BRCA2 activity are extremely sensitive to MMS, followed by XRCC2 and Ku80 deficient cell lines, while rad51 cells show a mild increase in their sensitivity. The importance of the two DSB repair pathways and distinct pathway proteins is discussed. 1. Institute of Toxicology, Johannes Gutenberg University, Mainz, Germany

379 Late activation of stress kinases (SAPK/JNK) by methylating and cross-linking agents in DSB repair defective cells Helbig L. (1), Hülsenbeck J. (1), Fritz G. (1) Activation of c-Jun-N-terminal kinases/stress activated protein kinases (SAPK/JNK) is part of the cellular response to genotoxic stress, finally regulating gene expression and cell survival. The significance of specific DNA lesions and DNA damage-related mechanisms for genotoxin-triggered signalling to SAPK/JNK is fairly unknown. Here, we investigated the effect of DNA methylating (MMS) and cross-linking (cisplatin) agents on the activation of SAPK/JNK in mouse fibroblasts impaired in non-homologous end joining (NHEJ) because of defect in the catalytic subunit of DNA-protein-kinase (DNAPKcs). Measuring the dual phosphorylation of SAPK/JNK (Thr183/Tyr185) by Western blot analysis, we found that cisplatin-induced late (i.e. 16h) SAPK/JNK activation was more pronounced in DNA-PKcs cells as compared to corresponding wild type. Regarding MMS, SAPK/JNK activation was enhanced 6h after treatment in the repair defective cells. Thus, DNA-PKcs-regulated DNA repair mechanisms contribute to SAPK/JNK activation upon MMS and cisplatin exposure. The molecular mechanisms involved are unclear and might include reactive oxygen species (ROS) and endoplasmatic reticulum (ER) stress or MAPK-phosphatase-1 (MKP-1) regulated functions. Quantitative analysis of ROS formation by FACS-based method showed that MMS and cisplatin increase the level of ROS at early time point after exposure in a dose dependent manner. However ROS production was similar in both wild type and DNAPKcs defective cells. MMS but not cisplatin treatment resulted in ROS formation 16h after exposure. Yet, this was independent of the repair status of the cells. Genotoxins may also elicit ER stress, thereby provoking an IRE and ASK-driven activation of SAPK/JNK. A major marker of ER-stress is chaperone protein Grp78. This marker protein, however, was only induced in wild type but not in DNA-PKcs defective cells after cisplatin treatment. Thus, altogether, it appears unlikely that ROS- or ER stress-related mechanisms contribute to a late activation of SAPK/JNK activation upon cisplatin treatment. The impact of MMS on the expression level of Grp78 is currently analyzed. In summary, we show that the genotoxic agents cisplatin and MMS differ from each other with respect to the time kinetic of SAPK/JNK activation in DSB repair defective cells. Signs of ER stress and ROS formation were not observed in DNA-PKcs defective cells at late time point after cisplatin treatment, showing that the pronounced phosphorylation of SAPK/JNK after cisplatin treatment is a ROS and ER-stress independent process. 1. Dept. of Toxicology, Johannes Gutenberg University, Mainz, Germany

380 Role of Nijmegen breakage syndrome protein (NBN) in DNA signaling and cytotoxicity Eich M. (1), Roos W.P. (1), Digweed M. (2), Kaina B. (1) The DNA repair gene NBN encodes for Nibrin, which causes in case of deficiency the autosomal recessive disease Nijmegen breakage syndrome. This protein is a component of the MRN complex, which is essential for DNA double-strand break (DSB) recognition and repair, especially by homologous recombination. It also takes part in apoptotic signaling and in cell cycle checkpoint control via its interaction with the ATM kinase. Our data show that NBN defective human fibroblasts and lymphoblastoid cells are more sensitive to the methylating agent N-methyl-N’-nitro-N-nitrosoguanidine (MNNG) than the corresponding wild-type cells. Interestingly, annexin V/propidium iodide double-staining showed that methylating agent-induced cell death of NBN defective cells occurs mainly via the induction of necrosis, not apoptosis. Since the expression levels of the mismatch repair (MMR) proteins MLH1, PMS2, MSH2 and MSH6 did not differ significantly between wild-type and NBN mutated cells, the differences in cell sensitivity appear to be independent from the MMR status. Similar to human NBN deficient cells, human ATM deficient fibroblasts are hypersensitive to MNNG and show a higher level of necrosis compared to apoptosis. The presence of MGMT leads to an almost complete rescue from cell death after treatment with MNNG in all tested cell lines, meaning that O6-methylguanine is the main cytotoxic lesion. Additionally, NBN defective as well as ATM defective human fibroblasts showed no defect in base excision repair (BER). As shown by γH2AX immunostaining in human fibroblasts, a reason for the greater sensitivity of NBN mutated cells seems to be

impaired DSB recognition. These data suggest that NBN and ATM interact and trigger the apoptotic pathway and, therefore, are involved in the protection of cells against alkylating agent-induced cell death by necrosis, which is independent from activation of PARP-1. 1. Dept. of Toxicology, Johannes Gutenberg University, Mainz, Germany, 2. Dept. of Human genetics, Charité - University of Berlin, Berlin, Germany

381 The effects of the DNA-PK inhibitor vanillin after treatment with methyl methane sulphonate Linek B. (1), Nikolova T. (1), Kaina B. (1) Methyl methane sulphonate (MMS) is a well-known methylating agent and causes double strand breaks (DSB), which are critical for cell survival. In eucaryotes there are two major DSB repair pathways: homologous recombination (HR) and non-homologous end joining (NHEJ). The DNA-dependent protein kinase (DNA-PK) becomes activated upon binding to the ends of the DSB and phosphorylates itself and other NHEJ proteins. Vanillin is described as a selective DNA-PK inhibitor. In this study, we used two hamster cell lines: V79-2 (wild-type) and VC8 (BRCA2-mutant). BRCA2 is involved in homologous recombination in recruitment of RAD51. Survival fraction was determined by colony formation assay, survival by WST assay, induced cell death by FACS analysis of subG1 fraction, FACS analysis of cell cycle phases. The molecular mechanisms were tested by Western blot analysis of SARK/JNK and phosphorylated p38. DSB induction was studied by Western blot analysis of phosphorylated H2AX and immunofluorescent analysis of γH2AX foci. Our data showed that vanillin enhances slightly the effects of MMS in BRCA2-deficient cells. No effect on cell cycle progression was observed. The SARK/JNK pathway was early activated in both cell lines but already reduced at 4h following MMS treatment. Vanillin did not affect the activation of this signalling pathway. In addition our data indicate that vanillin itself increases the number of γH2AX-foci in the mutant cells, but following a combined treatment with MMS, the number of MMSinduced γH2AX-foci was slightly reduced. Western blot confirmed increased phosphorylation of H2AX in the BRCA2 deficient cells after vanillin treatment. 1. Institut für Toxikologie, Johannes Gutenberg Universität, Mainz, Germany

382 Flap endonuclease-1 expression in human tumours Nikolova T. (1), Christmann M. (1), Nagel G. (1), Kaina B. (1) Flap endonuclease-1 (FEN-1) endonuclease is a DNA replication/repair protein with pleiotropic functions. It was shown to play a critical role in initiator RNA/DNA primer removal during lagging-strand DNA synthesis and is an integral part of the long-patch base excision repair. FEN-1 is also involved in double strand break repair (DSB) by nonhomologous end-joining or homologous recombination repair. FEN-1 is important for telomere stability and the preservation of genomic stability. Overexpression of FEN-1 protein has been reported in lung cancer cell lines, and in prostate cancers. We studied the expression of FEN-1 protein in protein extracts from primary seminoma testis, lung tumours, glioblastoma multiforme and astrocytomas by Western Blot analysis. In addition, we were interested how transient down-regulation of FEN-1 by siRNAs would affect the sensitivity of glioblastoma cell lines to some model chemical mutagens and chemotherapeutical agents as determined by FACS analysis of subG1 hypodiploid fraction and WST assay. There was a tendency of FEN-1 overexpression at a high percentage in all types of investigated tumours. Down-regulation of FEN-1 rendered the LN308 glioblastoma cells more sensitive to cell death induced by MMS and cisplatin. The biological role of increased FEN-1 expression in cancer cells is discussed. 1. Institute of Toxicology, Johannes Gutenberg University, Mainz, Germany

383 Induction of Fra1 by topotecan and ACNU in the human glioblastoma cell line LN229 is independent of MAPK cascade and enhances cellular survival Birkner R. (1), Kaina B. (1), Christmann M. (1) Fra1 belongs together with c-Fos and FosB to the family of Fos proteins, which form with members of the MAF, ATF and Jun family the transcription factor AP-1. Previously we showed that c-Fos protects mouse embryonal fibroblasts against the cytotoxic effects of DNA damaging agents by stimulation of DNA repair processes. To analyse the role of Fos proteins in human tumour cells, in which c-Fos is frequently up-regulated, we utilized the glioblastoma cell line LN229 and studied the expression of the Fos proteins upon exposure to the therapeutically used anticancer drugs ACNU and topotecan. Since it was reported that an increased expression of Fra1 correlates to resistance of cancer cells to chemotherapy, we focused on this Fos-related protein. Whereas Fra1 protein was expressed at low level in untreated LN229 cells, the expression was strongly enhanced already 6h after exposure to topotecan and ACNU. Since no induction of the corresponding mRNA was observed, Fra1 expression seems to be regulated posttranscriptional. This is in opposite to c-Fos and FosB, which are transcriptionally induced upon ACNU and topotecan exposure. The induction of Fra1 is independent on the MAPK cascade since the Jun kinase, the p38 kinase and the extracellular signallingregulated kinases 1/2 are activated only at late time points, e.g. 24 h after exposure. In addition, inhibition of these MAP kinases did not block Fra1 induction. To analyse the role of Fra1 in resistance to ACNU and topotecan we inhibited Fra1 by siRNA. As shown by SubG1 measurement and tryptan blue exclusion assay, knock down of Fra1 clearly sensitizes cells against ACNU and topotecan. Work was supported by Stiftung Rheinland-Pfalz and DFG. 1. Dept. of Toxicology, Johannes Gutenberg University, Mainz, Germany

78

384 c-Fos dependent regulation of the three prime exonuclease 1 (Trex1) by genotoxic stress Christmann M. (1), Tomicic M.T. (1), Aasland D. (1), Berdelle N. (1), Schulz L. (1), Kaina B. (1) Cells deficient in c-Fos are hypersensitive to genotoxic stress such as ultraviolet (UV) light and benzo[a]pyrene diolepoxide (B(a)P). Here we analysed the c-Fos dependent regulation of DNA repair genes in mouse fibroblasts upon exposure to genotoxins. Microarray analysis revealed a strong UV-C induced induction of the three prime exonucleases I (Trex1) and II (Trex2), which was confirmed by semi-quantitative PCR and real time PCR. Induction of Trex1 was also verified by protein expression studies. Experiments using the transcriptional inhibitor actinomycin D showed that the induction of Trex1 is caused by de novo RNA synthesis, whereas induction of Trex2 is caused by a post transcriptional mechanism. This was also shown by promoter analysis demonstrating inducibility of the Trex1 promoter by UV light. Promoter dissection experiments showed that a 730 Bp fragment containing a potential AP-1 site is sufficient UV-C inducibility. In contrast, a 517 Bp fragment missing this AP1-binding shows no induction by UV-C. To verify the activity of this AP-1 binding site, EMSA and CHIP experiments were performed, showing recognition of this TREX1 specific AP-1 binding site by c-Fos (AP-1) in vitro and in vivo. In accordance, induction of Trex1 was not observed in c-Fos deficient cells, suggesting a role for c-Fos in the induction of Trex1. Trex1 was also induced upon exposure to different chemical genotoxic stimuli, such as B(a)P. These agents also induce c-Fos expression at high level, suggesting a role for cFos in regulation of Trex1 induction also by this carcinogen. Analysis of the intercellular distribution revealed a cytoplasmic localization of Trex1. However upon UV-C exposure Trex1 translocates into the nucleus. To study the biological role of Trex1, we performed co-immunoprecipitation experiments to demonstrate physical interaction with other proteins. We were able to show that Trex1 interacts with the translesion polymerase eta (PolH). The data strongly suggest a role for Trex1 in the tolerance of UV and B(a)Pinduced DNA damage. Work was supported by Stiftung Rheinland-Pfalz and DFG KA724. 1. Dept. of Toxicology, Johannes Gutenberg University, Mainz, Germany

385 A role for iron-sulfur cluster proteins in DNA repair Thierbach R. (1,2), Drewes G. (2), Fußer M. (3), Wolfrum K. (2), Epe B. (3), Ristow M. (4), Steinberg P. (1,2) Oxidative damage to DNA is thought to contribute to the pathogenesis of various diseases including several types of cancer. Reactive oxygen species are formed as byproducts during normal cellular metabolism, but can also be formed by genotoxic chemicals and ionizing radiation. Cells have several defence systems against DNA damage caused by oxidative stress. One of the protection mechanisms is DNA repair, which removes oxidative DNA damage and restores correct genetic information. Some of these DNA repair enzymes depend on the activity or expression of iron-sulfur (Fe/S) cluster containing proteins. The aim of this work was to find out whether overexpression or knock out of the gene encoding the mitochondrial protein frataxin, which is involved in the maturation of Fe/S cluster containing proteins, leads to an enhanced or reduced DNA repair. V79 Chinese hamster cells overexpressing human frataxin and freshly isolated hepatocytes from mice with frataxin knockout were analyzed by the alkaline elution technique. At the basal level and after light-induced DNA damage both the single-strand DNA breaks and the Fpg-sensitive modifications (formamidopyrimidine glycosylase-sensitive sites consisting mostly of 8-oxo-dG residues) were comparable in V79 control cells (V79-Neo) and frataxin-overexpressing cells (V79-hFX). Four hours after damage induction, however, the frataxin-overexpressing cells showed significantly less unrepaired DNA damage (25%) than the control cells (38%). In the second experimental system freshly isolated wild-type and frataxin knockout hepatocytes with frataxin knock out and control hepatocytes, both from three-week old mice, were analysed. Whereas the number of basal DNA single-strand breaks was similar in the two experimental groups, the alkaline elution assay in combination with the repair glycosylase Fpg showed that more oxidative DNA base modifications occurred in the hepatocytes with frataxin knock out (35%). We used a cleavage assay to check the role of glycosylase activity in both experimental systems, but no significant differences were detected. Taken together, the data suggest that the activity and the expression of Fe/Scluster containing proteins play a crucial role in the repair of oxidative DNA damage. 1. Institute for Food Toxicology and Analytical Chemistry, Dept. of Food Toxicology and Replacement/Complementary Methods of Animal Testing, University of Veterinary Medicine Hannover, Hannover, Germany, 2. Institute for Nutrition Science, University of Potsdam, Nuthetal, Germany, 3. Institute for Pharmacy, University of Mainz, Mainz, Germany, 4. Institute for Nutrition Sciences, Dept. of Human Nutrition, University of Jena, Jena, Germany and German Institute of Human Nutrition, Potsdam-Rehbrücke, Nuthetal, Germany

386 Myeloeperoxidase mediated cytotoxcity of p-benzoquinone Westphal G.A. (1), Bünger J. (1), Lichey N. (2), Taeger D. (1), Hallier E. (2) The generation of carcinogenic benzene metabolites was proposed to involve P450dependent benzene oxidation in the liver and peroxidase mediated generation of the ultimate carcinogenic metabolites. However, direct evidence for this proposal does not exist. If an oxidised intermediate such as hydroquinone (HQ) or p-benzoquinone (p–BQ) was actually the precursor for the ultimate carcinogenic benzene metabolite and further activation proceeds via MPO mediated reactions, it should be possible to activate p–BQ to a genotoxic compound in vitro. Therefore, we investigated p-BQ using human peripheral mononuclear blood cells (PBMC) of three different donors, co-incubated with PMA. To monitor genotoxic effects we applied the cytochalasin B-block micronucleus test. Results: Addition of 20-28 ng/ml PMA caused a significant increase of micronuclei at low and non-cytotoxic p–BQ concentrations between 0.04–0.2 µg/ml (0.37–1.85 µM). With PMA or p-BQ alone no reproducible elevation of micronuclei was seen up to toxic concentrations. Conclusions: PMA and p–BQ induce micronuclei when administered jointly. We hereby show for the first time genotoxic effects of p–BQ at non cytotoxic concentrations. Our results add further support to the hypothesis that MPO is a key

enzyme in the activation of benzene. Furthermore our results will enable the comparative investigation of other benzene metabolites such as HQ or catechol. 1. BGFA - Research Institute of Occupational Medicine, German Social Accident Insurance, Ruhr-University, Bochum, Germany, 2. Department of Occupational and Social Health, Georg-August-University, Göttingen, Germany

387 Induction of micronuclei in vitro by the anabolic doping steroids desoxymethyltestosterone (madol) and 19-norandrostenedione Dorn S.B. (1), Bolt H.M. (1), Thevis M. (2), Diel P. (3), Degen G.H. (1) The abuse of anabolic steroids for doping raises concerns, since many agents have not been examined for toxicological properties. Aside from hormonal (androgenic) activity, anabolic steroids may also exert genotoxic effects as shown previously for nandrolone, tetrahydrogestrinone and trenbolone (Dorn et al., 2008, Arch Toxicol 82: 257-263). In the present study, we determined the potencies of the designer steroid madol (MAD) and the anabolic pro-hormone 19-norandrostenedione (NA) to induce micronuclei in vitro. CREST analysis was used to differentiate between aneugenic and clastogenic mechanisms of micronucleus induction in V79 cells. Cytotoxicity of the steroids and their influence on the cell cycle were assessed in parallel. In addition, the ability of MAD and NA to increase production of reactive oxygen species and to induce apoptosis were studied. Both agents caused a concentration-dependent increase in the rates of micronuclei in V79 cells, exceeding a doubling of the background micronucleus rates of untreated controls, which was evident at 27 µM and 29 µM for MAD and NA, respectively. The steroid-induced micronuclei were predominantly kinetochor (CREST)negative, pointing to a clastogenic mode of action. As cytotoxicity of both compounds was weak (IC20 value of 300 µM for NA and IC10 of 100 µM for MAD), cytotoxicity is unlikely to contribute to their genotoxicity. The observed genotoxicity of MAD and NA was neither due to apoptosis induction nor to production of reactive oxygen species. However, the ability of both steroids to induce micronuclei appears related to their lipophilicity. Therefore, a non-specific chromosomal genotoxicity of MAD and NA, based on hydrophobic interactions, appears likely. This could result in biologically relevant increases in chromosomal damage when critical concentrations of the agents are reached in vivo. Regarding the current misuse of the steroids for doping, the uncontrolled administration of very high doses must be taken into account. Therefore it cannot be ruled out that MAD and NA present genotoxic hazards under current misuse conditions by athletes in sports or in body building. 1. IfADo – Leibniz Research Centre for Working Environment and Human Factors, Dortmund, Germany, 2. Center for Preventive Doping Research – Institute of Biochemistry, German Sport University Cologne, Cologne, Germany, 3. Center for Preventive Doping Research – Institute of Cardiovascular Research and Sports Medicine, German Sport University Cologne, Cologne, Germany

388 Mobile phone radiation did not cause micronuclei induction in human-hamster hybrid AL cells in a standard protocol Hintzsche H. (1), Jastrow C. (2), Kleine-Ostmann T. (2), Schrader T. (2), Stopper H. (1) Mobile phone use is still increasing throughout the whole world. Currently there are more than three billion subscribers. In Western Europe the penetration rate (subscribers / person) exceeds 110%. Health effects of the transmitting radiation (≈900 MHz and ≈1800 MHz) have been discussed and investigated for a long time. Nevertheless there has not been a final answer because researchers found diverse and even contradictory results. In this investigation AL cells, a CHO cell line containing one human chromosome, were exposed to electromagnetic radiation of 900 MHz for 30 minutes. Different field strengths were applied. Two irradiation modes were used, continuous wave and GSM modulation. Sham-exposed cells as negative controls and cells treated with methyl methanesulfonate as positive controls were included in the evaluation. After irradiation cells were treated with cytochalasin B, cultured for 24 hours and fixed in icecold methanol. All trials were performed as independent triple experiments. Micronucleus frequencies, cell proliferation and mitotic and apoptotic indices were determined on coded slides using fluorescence microscopy. For every tested field strength micronucleus frequencies were not elevated compared to sham-expositions neither in continuous wave mode nor in modulation mode. In the same manner apoptotic and mitotic indices as well as cell proliferation were not altered. Methyl methanesulfonate treated cells showed a significant increase in micronuclei and a decrease in cell proliferation. In summary, mobile phone radiation of 900 MHz did not induce micronuclei in AL cells in a standard protocol. Further studies have to reveal whether the same is true for other genotoxicity endpoints and whether effects can be detected in human cells or the whole organism. 1. Institut für Pharmakologie und Toxikologie, Universität Würzburg, 2. PhysikalischTechnische Bundesanstalt, Braunschweig

389 Cytotoxic and genotoxic effects of indoor air related α,β-unsaturated C7-C10 aldehydes in vitro Gminski R. (1), Sabisch K. (1), Mersch-Sundermann V. (1) In recent years, growing attention has been paid to health effects related to volatile organic compounds (VOC) found in indoor air. The air and settled dust in new buildings or following building/redevelopment work is usually contaminated with α,β-unsaturated aldehydes (2-alkenals). Emissions from wood-based building products such as oriented strandboards (OSB) are considered to be a major source of 2-alkenals. Most arise due to an autoxidative, chain-reactive splitting of unsaturated fatty acids contained in pinewood. Another source is the surface interaction of ozone with carpets, which produce several 2-alkenals as secondary emissions. To investigate the toxicological relevance of unsaturated aldehydes, the cytotoxic, genotoxic and chromosome damaging potential of 2-heptenal, 2-octenal, 2-nonenal and 2-decenal was studied in human lung cells in vitro. Cultured carcinoma pulmonary type II-like epithelium cells (A549) were exposed to the 2-alkenals dissolved in dimethyl sulfoxide (DMSO) at final concentrations ranging from 0.3-6000 µmol/L at 37°C for 24 h. Cytotoxicity was

79

assessed by erythrosine B staining. DNA-damage was monitored by the alkaline singlecell gel electrophoresis (comet) assay. The chromosome damaging potential was investigated by the cytokinesis-block micronucleus assay in vitro (CBMNvit). Results showed that all four 2-alkenals tested showed no or only slight cytotoxicity in the erythrosine B assay (viability >70%) and in the CBMNvit (cytotoxicity <20%). Furthermore, all four compounds caused a dose-dependent increase in DNA damage, as measured by the comet assay. A significant increase in micronuclei induction was found for 2-heptenal and 2-octenal at 0.7 µmol/L. 2-Decenal showed a similar chromosome damaging potency, whereas 2-nonenal was somewhat less active (60 µmol/L). Taken together, our results indicate that the four 2-alkenals tested were genotoxic in A549 human lung cells in vitro. Further studies are necessary to evaluate a possible human health risk. 1. Dept. of Environmental Health Sciences, Uniklinik Freiburg, Freiburg, Germany

390 Quantification of intracellular glutathione: method development and correlation with genotoxic stress Schmalbach K. (1), Jäger S. (1), Lehmann L. (1) Glutathione (GSH) biosynthesis is mediated by various transcription factors activated by cellular stress binding to their responsive elements in the promoter region of the key enzyme of GSH biosynthesis, γ-glutamylcysteine synthetase (GCS). Therefore, the intracellular GSH level is a sensitive indicator for cellular stress. Thus, the aim of this work was to develop a fast and practicable method for measuring relative GSH levels in living cells. The method is based on the reaction of the non-fluorescent monochlorobimane (MCB) and GSH forming a fluorescent product. The measurement of intracellular fluorescence intensity (FI) was performed by means of flow cytometry in cultured V79 cells. The modal value of FI was used to quantify the fluorescence signal. The method was optimized with respect to (i) stained cell density (0.7–13 mio cells/ml), (ii) staining time (15–90 min), and (iii) cells to be analyzed by flow cytometry (2.5 x 104 – 2.5 x 105). The FI was not affected by any of the parameters analyzed. Therefore, (i) time-consuming determination of cell number prior to staining could be omitted, (ii) an individual experimental design was possible, and (iii) a measuring time less than 60 s per sample could be realized. Thus, a multiwell application is possible. The method was calibrated using absolute GSH level determined by the method of Ellman. To specifically modify intracellular GSH content cell were treated with GSH ethyl ester and Lbuthionine-sulfoximine known to inhibit GCS. The fluorescence signal was standardized by microsphere calibration beads with defined and stable FI. In order to investigate the correlation between genotoxicity and GSH level, V79 cells expressing human cytochrome P450 isoenzymes 1A1 and 1B1 were treated with benzo[a]pyrene (B[a]P, 0.05-1 µM), estradiol catechols (ECEs, 1–10 µM), and 4-nitroquinoline N-oxide (NQO, 0.5 -1 µM). While B[a]P increased both number of 6-thioguanine resistant colonies and GSH level in a dose-dependent manner mutagenic concentrations of NQO did not affect the GSH level. Furthermore, ECEs induced both micronuclei and an increase in intracellular GSH. Hence, an increased intracellular GSH level could be a hint at the mode of genotoxic action. With this calibrated flow cytometric method to determine the total intracellular GSH content a fast and practicable indicator test for genotoxic stress was established. Supported by Deutsche Forschungsgemeinschaft (Grant Le 13/29-7). 1. Institute of Applied Biosciences, University of Karlsruhe, Karlsruhe, Germany

391 Pleiotropic effects of phenobarbital on protein patterns in mouse liver Rignall B. (1), Krause E. (2), Braeuning A. (1), Appel K.E. (3), Schwarz M. (1) Phenobarbital (PB) is a tumor promoter in rodent liver. The specific mechanism underlying this effect is not clear but is known to be dependent on the activation of the constitutive active/androstane receptor (CAR). PB induces a battery of drug metabolizing enzymes including members of the cytochrome P450 2B family and several phase II detoxifying enzymes. Furthermore, PB is known to affect cellular signaling pathways associated with cell proliferation and death. Here we report on the outcome of a protein expression analysis in liver of mice treated acutely or for prolonged periods with PB. Short-term treatment consisted of a single PB injection of 90mg/kg b.w. followed by a three days treatment with 0.05% PB in the diet. Tissues from long-term treated mice were obtained from an initiation-promotion experiment, where mice received a diet containing 0.05% PB for 32 weeks until sacrifice. Proteome analysis by two-dimensional gel electrophoresis and mass spectrometry revealed a total of 29 significantly deregulated proteins in liver of PB-treated mice, of which the majority were up-regulated. This effect was most pronounced for the phase II detoxifying glutathioneS-transferases. While glutathione peroxidase, peptide methionine sulfoxide reductase 1, and peroxiredoxin-6 expression were increased, the peroxisome-located catalase was down-regulated. This may indicate an altered redox status and/or increase in reactive oxygen detoxification in liver of PB-treated mice. The level of S-adenosyl methionine synthetase 1 was significantly decreased, which as a consequence should lead to a decrease in global DNA methylation, an effect that has been independently described to occur in liver of PB-treated mice. In conclusion, PB exerts pleiotropic effects in mouse liver which may be related to tumor promotion. A clear cut mechanistic link, however, could not be established. 1. Department of Toxicology, Eberhard Karls University, Tübingen, Germany, 2. LeibnizInstitute for Molecular Pharmacology, Berlin Buch, Germany, 3. Center for Experimental Toxicology, Federal Institute for Risk Assessment, Berlin, Germany

392 Effect of ionizing radiation (IR) on tumor cell adhesion in vitro and metastasis in vivo Herzog M. (1), Fritz G. (1) Metastasis is a complex process comprising a series of sequential steps, including extravasation. For these purposes, tumor cells (TC) have to adhere to endothelial cells (EC) via E-selectin-mediated mechanisms. E-selectin expression is NF-κB-dependent, requires Rho GTPases and can be induced by exposure of human umbilical vein endothelial cells (HUVEC) to ionizing radiation (IR) or inflammatory cytokines (e.g.

TNFα). The HMG-CoA reductase inhibitor lovastatin blocks Rho-regulated signalling and, therefore, is believed to influence TC-EC interactions. To investigate pro-adhesive effects of IR on TC-EC adhesion, and the impact of lovastatin on this process, human colon carcinoma cells (HT29) and HUVEC were used as experimental system. Cell adhesion was determined by ELISA-based method. IR exposure of EC increases Eselectin mRNA and protein expression and, furthermore, amplifies TC-EC adhesion. Notably, treatment of tumor cells with IR also promotes TC-EC adhesion. Simultaneous irradiation of TC and HUVEC had additive effects. Statin pretreatment of EC decreases TC-HUVEC adhesion, whereas lovastatin was poorly effective when solely TC were pretreated with lovastatin. mRNA expression analyses showed that the cell-adhesion molecules involved in IR promoted TC-EC interactions differ between TC and EC. To valuate the physiological relevance of our in vitro findings, we investigated the effect of IR on E-selectin expression and extravasation in vivo. We show that IR exposure of BALB/c mice leads to upregulation of E-selectin mRNA expression, which is attenuated by lovastatin. Preliminary results of these currently ongoing in vivo studies will be presented and discussed. 1. Dept. of Toxicology, Johannes Gutenberg University, Mainz, Germany

393 Primary effects of benzo[a]pyrene on urothelial cells with relevance for bladder carcinogenesis Numfor A. (1), Jakubowski N. (2), Feldmann I. (2), Roos P.H. (1) Bladder carcinogenesis is induced by several chemicals such as polycyclic aromatic hydrocarbons (PAH) and aromatic amines. Both types of compounds require enzymatic activation to achieve carcinogenic properties. Cytochromes P450 CYP1A1, CYP1B1 and CYP1A2, are the most relevant enzymes in this activation process. Their expression levels determine the capacity to generate chemically reactive metabolites which finally lead to DNA sequence alterations. The PAH benzo[a]pyrene (BaP) induces CYP1A1 and CYP1B1 in urothelial cells and thus enhances this capacity. The effects of BaP are not restricted to CYP induction but may concern further initial determinants in chemically induced bladder carcinogenesis. Therefore, we study effects in urothelial cells elicited by BaP within the first minutes up to two days after BaP exposure so that fast responses on proteins and their phosphorylation status as well as slower responses on transcriptional processes can be assessed. Methods: Human 5637 urothelial cells are treated with BaP (0.1 to 3µM). For short-term experiments, cells are lysed after 2, 6 and 10 min of BaP exposure and cell protein is prepared. Effects on mRNA and protein levels of CYP enzymes, cell cycle regulators and signal transduction components were determined after 24h of BaP exposure and analyzed by real-time RT-PCR, immunoblotting and LAICP-MS of blotted proteins. Results: Within the first minutes of BaP exposure, substantial alterations in the phosphoproteome are seen by LA-ICP-MS. Specifically, an increase in the phosphorylation of c-jun is observed after 6 and 10 min. This signal disappears after 24h concomitantly with a general decrease in c-jun while the protein level of c-myc remains unaltered. Transcripts of CYP1A1, CYP1B1 and CYP2S1 are elevated by BaP and transcript levels of p53, p21 and Rb are obviously not affected within 24h while bax mRNA is upregulated. We cannot exclude minor effects on AhR, AhRR, ARNT and GADD45 transcripts. Conclusions: Besides induction of AhRregulated CYPs, BaP elicits fast effects on the phosphorylation status of proteins including cell cycle regulators and on transcript levels of signalling proteins. Currently, we study effects of BaP on further proteins/phosphoproteins involved in regulation of cell cycle and apoptosis. 1. IfADo -Institut für Arbeitsforschung an der TU Dortmund, Dortmund, Germany, 2. Institute for Analytical Sciences (ISAS), Dortmund, Germany

394 Expression of the AML1-ETO fusion gene in haematopoietic stem cells does not lead to leukaemia but needs additional genetoxic lesions for the induction of overt AML Bockamp E. (1), Cabezas-Wallscheid N. (1), Eichwald V. (2), Ohngemach S. (1), Heck R. (1), Zhang D.E. (3), Arnold B. (2), Eshkind L. (1) The t(8;21) (q22;q22) translocation fusing the ETO gene on human chromosome 8 with the AML1 gene on chromosome 21 (AML1-ETO), is one of the most frequent chromosomal abnormalities in acute myeloid leukemia (AML). To elucidate the functional role of the AML1-ETO fusion protein for the pathogenesis of AML, we have generated a conditional mouse model capable of expressing AML1-ETO in haematopoietic stem cells (HSCs). To direct AML1-ETO expression to HSCs, we made use of a knock-in mouse line containing the tTA-2S tetracycline transactivator under the control of SCL regulatory elements, which facilitates conditional and reversible expression of transgenes in HSCs (Bockamp et al., 2006; Wilson et al., 2008). Induction of the AML1-ETO transgene alone was not sufficient to induce leukemia in mice after several month of continuous expression thus directly suggesting that a “second hit” is required for the onset of overt AML. Since it is known that exposure to benzene is correlated with a significant incidence of AML1-ETO-associated M2 AML, we have planned to introduce secondary DNA-lesions in AML1-ETO expressing mice. The use of the well defined and clinically relevant t(8;21) (q22;q22) translocation in a mouse model together with the induction of a second genotoxic hit will serve as an important in vivo model for AML in general and will eventually allow to design novel therapeutic strategies and compounds for treating AML1-ETO-associated leukaemias. Bockamp, E., Antunes, C., Maringer, M., Heck, R., Presser, K., Beilke, S., Ohngemach, S., Alt, R., Cross, M., Sprengel, R. et al. (2006), Blood 108, 1533-41; Wilson, A., Oser, G. M., van der Wath, R. C., Blanco-Bose, W., Laurenti, E., Dunant, C., Bockamp, E., Eshkind, L., Liò, P., MacDonald et al. (2008). Cell, in press 1. Institute of Toxicology, Johannes Gutenberg University, Mainz, Germany, 2. Department of Molecular Immunology, German Cancer Research Center, Heidelberg, Germany, 3. Department of Pathology, Division of Biological Sciences and Moores UCSD Cancer Center, University of California, San Diego, USA

80

395 ®

Anti-carcinogenic effects of vineatrol on different tumor cell lines Nevo-Koch E. (1), Brauer-Dewor N. (1), Steinberg P. (1) Vineatrol® is a vine extract and consists of monomeric trans-resveratrol and to a large extent of resveratrol oligomers. Many studies carried out with resveratrol show that this compound exerts a growth-inhibiting effect on a variety of cancer cell lines. However, although resveratrol occurrs in many different plants, only small amounts can be found in food of plant origin. Besides the monomeric trans-resveratrol many different derivatives of resveratrol have come to scientific interest in recent years. In many plants these derivatives, exhibiting a similar effect as resveratrol, occur in higher concentrations than the primary substance. The main source for vineatrol®, which also includes the different resveratrol oligomers, is the graft of vine. In order to verify the assumption that resveratrol oligomers are also able to inhibit the growth of tumor cell lines, the effect of vineatrol® on the human tumor cell lines HT-29 (derived from a colorectal adenocarcinoma), HCT 116 (from a colorectal carcinoma), SW 480 (from a colon adenocarcinoma), A-431 (from an epidermal carcinoma), PC-3 (from a prostate adenocarcinoma), and HCEC (a non-malignant immortalized human colon epithelial cell line) were tested by making use of the Sulforhodamin B-Assay. In this assay total protein is stained and measured by spectrophotometry. The intensity of the absorption is proportional to the number of cells present in the cell culture flasks and thus shows the degree of proliferation or inhibition of the cell cultures following exposure to the compounds tested. First results show that the studied cell lines react differently to vineatrol®. While A431 cells show an inhibition up to 90% after incubation with the highest vineatrol® concentrations, the colon tumor cell lines as well as the prostate tumor cell line only exhibit a maximal inhibition of about 50%. Further investigations are underway to examine the mechanism(s) of vineatrol®-meditated tumor cell growth inhibition and the differential sensitivities of the diverse cell lines towards the abovementioned extract. 1. Department of Food Toxicology and Replacement/Complementary Methods to Animal Testing, University of Veterinary Medicine Hannover, Hannover, Germany

396 Influence of arylhydrocarbon- (Ah) receptor and genotoxins on DNA repair gene expression and cell survival of mouse hepatoma cells Schreck I. (1), Chudziak D. (1), Schneider S. (1), Seidel A. (2), Platt K.L. (3), Oesch F. (3), Weiss C. (1) The aryl hydrocarbon receptor (AhR) mediates toxicity of a variety of environmental pollutants such as polycyclic aromatic hydrocarbons (PAHs) and dioxins. However, the underlying mechanisms and genetic programmes regulated by AhR to mediate adverse effects but also to counteract poisoning are still poorly understood.Here we analyzed the effects of two AhR ligands, benzo[a]pyrene (B[a]P), a DNA damaging tumour initiator and promotor and 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD), a pure tumour promoter, on cell survival and on nucleotide excision repair (NER) gene expression in mouse hepatoma cells (Hepa-c7). NER deals with so called “bulky” adducts including B[a]Pinduced DNA adducts. Therefore, the hypothesis that AhR may enhance NER gene expression to trigger DNA repair in presence of genotoxic AhR ligands was tested. Indeed, the mRNA levels of the two NER genes XP-C and DNA polymerase kappa were increased by B[a]P and TCDD, however, this was not accompanied by an increase of the amount of the respective proteins. Furthermore, we investigated a potential cytoprotective effect of AhR activation by the non-genotoxic ligand TCDD against cell death induced by B[a]P and its metabolite B[a]P-7,8-dihydrodiol-9,10-epoxide (BPDE). Pretreatment of cells with TCDD did not reduce cytotoxicity induced by these genotoxins. Thus, in Hepa-c7 cells AhR has no major effects on the expression of these crucial NER proteins and does not prevent genotoxin-provoked cell death. Additionally, a direct influence on NER gene expression by agents inducing “bulky adducts” was tested. To this end cells were exposed to the B[a]P metabolite BPDE or the anticancer drug cis-platin (CDDP). The genotoxins BPDE and CDDP led to p53 accumulation and induction of its target p21. Interestingly, however, NER gene expression was not enhanced but rather decreased. As two NER genes, XP-C and DNA damage binding protein ddb2, are up-regulated by p53 and ultraviolet radiation in human cells these findings suggest cell type, species or lesion specific actions of p53 on DNA repair gene expression. Importantly, in cells with damaged DNA upregulation of p53 may not suffice to enhance DNA repair gene expression. 1. Institute of Toxicology and Genetics, Forschungszentrum Karlsruhe, EggensteinLeopoldshafen, Germany, 2. Biochemical Institute for Environmental Carcinogens, Lurup 4, Grosshansdorf, Germany, 3. Institute of Toxicology, University of Mainz, Mainz, Germany

397 Synergistic effects of benzo[a]pyrene and 4-aminobiphenyl on AhR-signal transduction in urothelial cells Herbst K. (1), Roos P.H. (1) The carcinogenic effect of benzo[a]pyrene (BaP) on bladder epithelial cells requires its metabolic activation to the ultimate carcinogen. CYP1A1 can be involved in this activation since it is inducible by BaP in urothelial cells. CYP1A1 transcript, protein and activity are increased by BaP in a dose dependent manner in primary porcine bladder cells and in the human 5637 urothelial cell line. 4-Aminobiphenyl (4ABP), another bladder carcinogen, does not induce CYP1A1 in porcine bladder cells but synergistically increases its BaP-dependent induction. To elucidate the mechanistic background of this synergism and its significance for humans, we analyzed effects of BaP and 4ABP on the expression of CYP1A1, CYP1B1 and components of the Ah-receptor pathway in a human urothelial cell line based on the hypothesis that 4ABP affects components of that pathway. Methods: 5637 human urothelial cells were incubated for 24h with BaP (0.1 to 30 µM), 4ABP (2 and 10µM) and combinations of both. Relative transcript levels for CYP1A1, CYP1B1, AhR, AhRR and ARNT were determined by real time RT-PCR. Cell lysates were used for CYP-detection by immunoblots. Results: In the human 5637 cell line, transcripts of CYP1A1 and CYP1B1 are induced by BaP at concentrations of 0.3 µM and reaching a maximum response at about 1 µM BaP. Within this concentration range, clear induction of CYP1A1 protein is also observed. Similar to the effect in porcine bladder cells, there is a synergistic effect of 4ABP on BaP-mediated CYP1A1

induction in 5637 cells. BaP or 4ABP only slightly if at all modulate the expression of AhR, ARNT and AhhRR while a combination of both leads to an increase in AhRRtranscripts. Conclusions: For the human urothelial cell line 5637, a synergistic effect of 4ABP on BaP-mediated CYP1A1 induction is shown. This is relevant for the carcinogenic potential of tobacco smoke which contains both compounds. There are hints that the synergistic effect is related to interactions of 4ABP with components of the AhR-pathway. However, an increase in AhRR is expected to diminish the BaP mediated induction and not to synergistically enhance it. Probably, interactions of 4ABP with proteins of the receptor pathway are more important than alterations at the transcript level. 1. IfADo – Leibniz-Institut für Arbeitsforschung an der TU Dortmund, 44139 Dortmund

398 Effect of benzo[a]pyrene and flavonoids on the AhR-pathway in duodenal cells Niestroy J.L. (1), Barbara A. (1, 2), Roos P.H. (1) The aryl hydrocarbon receptor (AhR)-pathway mediates adaptive or toxic effects elicited by xenobiotics regulating the gene expression of cytochromes P450 in a liganddependent fashion. This process can be modulated by dietary flavonoids which could positively influence the outcome of diseases associated with CYP-mediated formation of reactive metabolites and with oxidative stress (cardiovascular diseases, inflammation and cancer). Here we examine the effect of benzo[a]pyrene (BaP) and flavonoids on components and target genes of the AhR-pathway in duodenal cell lines and in rat duodenum. Methods: The cell lines HuTu-80 (human adenocarcinoma) and IEC-6 (rat normal epithelial cells) were incubated for 48h with BaP (0.01-10µM) or with flavonoids (kaempferol and quercetin 0.5-10µM). Transcripts related to the AhR-pathway were quantified by real-time RT-PCR. The intracellular accumulation of neutral red (measured at λ=540nm) was used for cytotoxicity assays. 12 weeks old male Wistar albino rats were dosed twice after an 8h fasting interval with 50mg BaP/kg body weight and sacrificed 48h after the first fed. Results: IEC-6 cells: Both, CYP1A1 and CYP1B1, but not CYP1A2 are induced by BaP. However, basal and induced transcript levels of CYP1B1 are significantly higher than those of CYP1A1 (about 40x). Transcripts of AhRR and Nrf2 are elevated by BaP 3- and 2-times, respectively. While quercetin induced the expression of CYP1A1, 1B1 and Nrf2, kaempferol exerted no effect on the AhRpathway. HuTu-80 cells: BaP slightly induced the transcript level of CYP1B1 but only low CYP1A1 and no CYP1A2 expression were found. The expression of AhR, ARNT, GCS and CYP1B1 was slightly induced by kaempferol, while quercetin induced the expression of AhR, ARNT, Nrf2 and GCS. Rat duodenum: BaP had an inductive effect only on CYP1A1 expression while transcript levels of receptor and signalling components remained unaltered. Conclusions: 1. IEC-6 cells respond to BaP treatment inducing CYP1A1 and 1B1; 2. HuTu-80 cells did not respond to the tested BaP concentrations; 3. Flavonoids exert an inductive effect on the expression of components of the AhR-pathway in both cell lines but at higher concentrations in comparison to BaP; 4. BaP induced only CYP1A1 in the in vivo experiment. 1. IfADo – Leibniz-Institut für Arbeitsforschung an der TU Dortmund, Dortmund, 2. CICSUMA-IPN. México City

399 Impact of AhR knockdown on cell cycle progression in human keratinocytes (HaCaT) Kalmes M. (1), Blömeke B. (1) While the activation of the arylhydrocarbon receptor (AhR) by exogenous ligands is well investigated, the physiological activation of this receptor is less understood. By extending research in AhR biology, evidence appeared that the AhR interacts with endogenous ligands and generally plays an important role in cell physiology. In skin, up to now little is known about the endogenous functions of the AhR. We could recently show that an AhR knockdown variant of the human keratinocyte cell line HaCaT (siAhR HaCaT) exhibited changes in cell cycle state distribution in comparison to wildtype HaCaT cells in the absence of exogenous ligands. Significantly more cells were detected in the G0/G1 phase of the cell cycle in siAhR HaCaT cells than in wildtype HaCaT cells. In order to expand this knowledge and to reveal the underlying molecular mechanisms in the absence of exogenous ligands, we now compared DNA synthesis rate of both cell lines and determined protein levels of cell cycle regulatory polypeptides involved in the transition from G0/G1 to S phase in both, wildtype HaCaT and siAhR HaCaT cells. Compared to wildtype HaCaT cells, DNA synthesis rate of siAhR HaCaT cells was significantly decreased, whereas the protein level of the cyclin dependent kinase inhibitor (CKI) p27KIP1 was increased and the protein level of the retinoblastoma protein (pRB) was reduced. A decreased protein level of the cyclin dependent kinase (CDK)2 was detected in siAhR HaCaT and protein levels of CDK4 and CDK6 remained unchanged compared to wildtype HaCaT cells. In conclusion, we could show that in the absence of exogenous ligands the AhR promotes cell cycle progression in HaCaT cells and one can speculate that this is one of the physiological functions of the receptor in this cell line. 1. Dept. of Environmental Toxicology, University Trier, Trier, Germany

400 The redox-sensitive transcription factor NF-kappa B as molecular target for bioactive natural compounds Chovolou Y. (1), Proksch P. (2), Kampkötter A. (1), Wätjen W. (1), Kahl R. (1) NF-kappa B is a family of closely related protein dimers that play a key role in several physiological processes like inflammation, oxidative stress response and apoptosis. As a consequence deregulated NF-kappa B activity contributes to human diseases like chronic inflammation, autoimmune disorders or cancer. Furthermore, in response to chemotherapies cancer cells often activate NF-kappa B with subsequent acquisition of resistance to apoptosis. This enhanced activation of NF-kappa B combined with the fact that many cancer cells show increased NF-kappa B activity prior to treatment with chemotherapeutic compounds is one main reason for reduced cancer therapy efficiency and chemoresistance. It is therefore of great interest to identify specific and potent compounds that are capable of modifying this NF-kappa B signaling pathway and to

81

clarify their mechanism of action at the molecular level. For this purpose we developed a set of cell-based assays to identify natural compounds that block NF-kappa B activation cascade directly or by inhibiting the proteasome machinery. Diverse pentacyclic triterpenoids were used to assess their NF-kappa B modulating potential. From this compounds only ursolic acid but not the closely related compound ursolic acid methylester was able to inhibit NF-kappa B activation in response to several known NFkappa B activators like tumor necrosis factor-alpha (TNF-alpha) and phorbol 12myristate 13-acetate (PMA) or interleukin 1beta (IL1-beta). Pretreatment of cells with ursolic acid prevented phosphorylation of the NF-kappa B subunit p65 and inhibited DNA binding of NF-kappa B in response to TNF-alpha. In addition, diverse natural compounds derived from marine organisms were used to assess their NF-kappa B inhibiting potential. Particularly, five avarone-derivatives derived from marine sponges were analyzed for their NF-kappa B inhibiting potential. We found that one of these avarone derivatives showed in micromolar range a noticeable reduction of the TNFinduced NF-kappa B activation in our cell-based assay. However, possible mechanisms underlying the inhibitory action of this avarone derivative are still under investigation. 1. Institute of Toxicology, Heinrich-Heine-University, Düsseldorf, Germany, 2. Institute of Pharmaceutical Biology and Biotechnology, Heinrich-Heine-University, Düsseldorf, Germany

401 (-)-Epigallocatechin-3-gallate stimulates the insulin-responsive PI3K/Akt signaling pathway Bartholomé A. (1,2), Sies H. (2), Klotz L.O. (1,2) The green tea flavonoid (-)-epigallocatechin-3-gallate (EGCG) has been reported to have insulin-like properties in mammals, a beneficial effect that may be useful in alleviating insulin resistance and diabetic complications. We here demonstrate that EGCG exerts such insulin-mimetic effects in terms of activating the insulin-responsive phosphoinositide-3’-kinase (PI3K)/Akt pathway in HFFF2 human skin fibroblasts. Only EGCG, but neither epicatechin (EC), epigallocatechin (EGC) nor epicatechin gallate (ECG), significantly stimulated phosphorylation of Akt in a dose-dependent manner as demonstrated by Western blotting. This phosphorylation was abolished by the structurally unrelated PI3K inhibitors wortmannin and LY2940002. Several insulin effects relying on transcriptional modulation are mediated by transcription factors of the FoxO (forkhead box, class O) family that, upon stimulation of cells with insulin, are phosphorylated and inactivated by Akt. In HFFF2 cells exposed to ECGC, FoxO proteins were phosphorylated at a position known to be targeted by Akt, indicating that Akt activity is indeed elevated in EGCG-exposed cells. Hydrogen peroxide, like EGCG and insulin, is a potent stimulator of the same signaling pathway. EGCG was reported to cause the generation of significant amounts of hydrogen peroxide in culture media. Whether EGCG stimulates the PI3K/Akt pathway directly or via production of hydrogen peroxide is currently being investigated. 1. Institut für umweltmedizinische Forschung (IUF) an der Heinrich-Heine-Universität Düsseldorf gGmbH, Düsseldorf, Germany. 2. Institut für Biochemie und Molekularbiologie I, Heinrich-Heine-Universität, Düsseldorf, Germany.

402 Role of forkhead transcription factors on doxorubicin-mediated cytotoxicity in human hepatoma cells Beermann A. (1), Lüpertz R. (1), Wätjen W. (1), Kampkötter A. (1), Kahl R. (1), Chovolou Y. (1) The transcription factors FOXO1, FOXO3a and FOXO4 belong to a family of transcriptional regulators characterized by a conserved DNA-binding domain termed the forkhead box. The FOXO transcription factors which play major roles in control of cellular proliferation, oxidative stress and apoptosis, are becoming more and more considered as crucial therapeutic targets in cancer. In our previous work we showed that in human colon cancer cells (Hct-116) cells sustained activation of FOXO4 was connected with an enhanced doxorubicin-induced cytotoxicity and apoptosis (Lüpertz et al., Carcinogenesis, 2008). In this study we further investigated the effects of doxorubicin in various cancer cell lines and the contribution of the AKT/FOXO signalling pathway in doxorubicin action and resistance. First we compared the expression FOXO1, FOXO3a and FOXO4 in five different cell lines, two hepatoma cell lines (HepG2, Huh7), two colon cancer cell lines (Caco2, Hct-116) and one breast cancer cell line (MCF7). Interestingly, we found a highly differing expression pattern for the three transcription factors. FOXO1 was expressed in all tested cell lines with the highest expression in Caco2 cells. The highest expression of FOXO3a was found in MCF-7 cells and could be detected to a lesser degree in the other cell lines while FOXO4 was solely expressed in Hct-116 cells. Next we assessed the doxorubicin mediated cytotoxicity in HepG2 and Huh7 cells by neutral red assay and MTT assay. We found that Huh7 cells are more susceptible to doxorubicin-mediated cytotoxicity, compared to HepG2 cells with EC50 values (µM) of 6.5 and 10, respectively after treatment for 3h with doxorubicin following a recovery period of 24h. Moreover some studies found an association of multidrug resistance protein 1 (MDR1) and enhanced FOXO activity. Therefore the expression of MDR1 in doxorubicin treated human hepatoma cells Huh7 and HepG2 was analyzed by RT-PCR. First results implicate no significant changes in MDR1 mRNA expression after treatment with doxorubicin. More effects of doxorubicin on the phosphorylation of FOXO transcription factors and their impact in modulating cellular response to doxorubicin are still under investigation. 1. Institute of Toxicology, Heinrich-Heine-University, Düsseldorf, Germany

403 FOXO3a dependent regulation of the antioxidant enzyme catalase in MCF-7 carcinoma cells Lüpertz R. (1), Chovolou Y. (1), Kampkötter A. (1), Wätjen W. (1), Kahl R. (1) Physiological amounts of intracellular reactive oxygen species (ROS) are of importance to maintain proper function of redox-signalling. Particularly H2O2 serves as a second messenger for regulation of signalling transduction pathways and redox-sensitive transcription factors. Previously we showed that TNF-α treatment of MCF-7 wildtype

(MCF-7wt) cells resulted in downregulation of catalase and induction of MnSOD, which was associated with an accumulation of intracellular H2O2. TNF-α treatment of MCF-7 cells overexpressing the antioxidative enzyme catalase (MCF-7cat) led to enhanced cytotoxicity. We showed that a sufficient H2O2 level was needed to co-activate the redox-sensitive transcription factor NF-κB after TNF-α treatment and that this impairment of NF-κB activation is related to the observed enhanced TNF-α mediated cytotoxicity in MCF-7cat cells (Lüpertz et al., JCB, 2008). FOXO3a, a member of the forkhead superfamily of transcription factors, is another redox-sensitive transcription factor which can be regulated by H2O2. The relevance of intracellular H2O2 in FOXO3a signalling and the correlation between FOXO3a and catalase was investigated in MCF-7 cells. Reduction of intracellular H2O2 levels by overexpression of catalase resulted in reduced nuclear levels of FOXO3a in MCF-7cat cells. Additionally, overexpression of FOXO3a caused an increase in catalase expression which further supported the close connection between FOXO3a activity and catalase expression. TNF-α mediated downregulation of catalase expression and subsequent accumulation of H2O2 in MCF7wt cells was accompanied by nuclear exclusion of FOXO3a. In contrast, MCF-7cat cells with insufficient intracellular amounts of H2O2 showed nuclear import of FOXO3a after TNF-α treatment. These results suggest that regulation of redox-sensitive transcription factor FOXO3a is dependent on intracellular amounts of H2O2 and that intracellular redox state itself can be modulated by a FOXO3a dependent regulation of the antioxidant enzyme catalase. The significance of FOXO3a in TNF-α mediated downregulation of catalase and TNF-α cytotoxicity is still under investigation. 1. Institute of Toxicology, Heinrich-Heine-University, Düsseldorf, Germany

404 Modulation of PI3K/Akt/FoxO signaling by nickel ions Eckers A. (1), Reimann K. (1), Klotz L.O. (1) Nickel compounds may act as carcinogens, affecting both initiation and promotion stages of carcinogenesis due to their capability of both inducing DNA damage and modulating cellular signaling cascades that are known to affect cellular proliferation. We have previously described the stimulation of phosphoinositide 3-kinase (PI3K)/Akt signaling in cells exposed to copper ions, resulting in phosphorylation and nuclear exclusion of FoxO transcription factors. Here, human hepatoma cells were stressed with nickel or copper ions, followed by comparative analysis of PI3K/Akt-dependent signaling. Exposure of hepatoma cells to copper ions resulted in extensive oxidation of cellular glutathione, whereas no such effect was detected with nickel ions. Similarly, copper ions were more than 100-fold more toxic to cells than nickel, as deduced from colony forming abilities. Despite this lack of oxidative and cytotoxic action, exposure of hepatoma cells to Ni2+ resulted in a significant stimulation of Akt that was abrogated by inhibitors of PI3K. Although coincident with a phosphorylation of Akt substrates, such as glycogen synthase kinase-3, activation of Akt did not result in significant nuclear exclusion of FoxO1a; moreover, while Cu2+ caused a strong phosphorylation of FoxO proteins, only a slight and hardly detectable phosphorylation was induced by exposure to Ni2+. Accordingly, no significant modulation of the activity of a FoxO-responsive promoter construct was observed in cells exposed to nickel ions. In summary, exposure of HepG2 human hepatoma cells to nickel ions results in stimulation of the Ser/Thr kinase Akt in a PI3K-dependent manner, activation most likely being independent of oxidative processes. In sharp contrast to copper-induced effects, nickel-induced Akt activation is not propagated further downstream to FoxO-dependent signaling beyond phosphorylation of FoxO1a and 3a. 1. Institut für umweltmedizinische Forschung an der Heinrich-Heine-Universität gGmbH, Düsseldorf

405 The nuclear import of the small RhoGTPase Rac1 is mediated by the direct interaction with karyopherin α2 Sandrock K. (1), Bielek H.S. (2), Schmidt G. (2), Klugbauer N. (2) The small GTPase Rac1 is a member of the Rac subfamily, comprising the subtypes Rac1, Rac2 and Rac3. They act as molecular switches cycling between a GDP-bound inactive and GTP-bound active state. Members of this subfamily appear to regulate a diverse array of cellular events, including the control of cell growth, cytoskeletal reorganization, and the activation of protein kinases. Recent data point out that Rac1 is not only involved in multiple cytosolic functions but also translocates to the nucleus where it may participate in further signaling pathways. In the yeast two-hybrid system, we identified the nuclear import receptor karyopherin α2 (KPNA2) as a direct interaction partner of Rac1. The C-terminal polybasic region of Rac1 contains a nuclear localization signal (NLS), whereas Rac2 and Rac3 lack a functional NLS and do not bind to KPNA2. The presence of the NLS in Rac1 determines the specificity of this interaction and is a prerequisite for the nuclear import. Activation of Rac1 via the cytotoxic necrotizing factor 1 and the concurrent inhibition of its proteasomal degradation are crucial for the nuclear accumulation of Rac1 with KPNA2. Conversely, the reduction of KPNA2 expression inhibits the nuclear import of Rac1. Recently it became evident that the Rac1 downstream effector PAK1 is also involved in nuclear signaling pathway given that three functional NLS sequences were identified in the N-terminal region of PAK1. Here, we demonstrate the nuclear entry of Rac1 via KPNA2 and hypothesize a linkage to signalling pathways involving PAK1 in the nucleus. 1. Universitätsklinikum Freiburg, Germany, 2. Institut für Experimentelle und Klinische Pharmakologie und Toxikologie, Albert-Ludwigs-Universität Freiburg

406 RhoB attenuates NF-κB-, AP1-, SRE- and p53-regulated gene expression and promotes cell survival after UV-C treatment Hartig C. (1), Hülsenbeck J. (1), Fritz G. (1) RhoB belongs to the Rho family of small GTPases. RhoB and RhoA are 88% identical at the amino acid level. In contrast to rhoA, basal rhoB mRNA expression is low and inducible by genotoxic stress, including UV-light. Furthermore, RhoB differs from RhoA regarding its C-terminal isoprenylation. While RhoA is exclusively geranylgeranylated, RhoB is subject to either geranylgeranylation or farnesylation. RhoB regulates cellular

82

actin dynamics, cell proliferation, gene expression, and vesicle trafficking. It has Januslike effects on apoptosis; while it promotes ionizing radiation and alkylating drug-induced cell death, it is reported to protect keratinocytes from UV-induced apoptosis. To further elucidate the impact of RhoB on UV-induced cell death in mouse fibroblasts (MEFs), we performed FACS analysis using RhoB +/- and RhoB -/- MEFs. Both cell lines exhibited a dose dependent increase in cell death fraction in response to UV-C light. Yet, RhoB -/cells showed more pronounced effects. RhoB might promote the survival of UV-C irradiated cells by regulating complex changes in gene expression. To scrutinize this hypothesis, we cotransfected NIH3T3 fibroblasts with NF-κB-, AP1-, SRE- and p53regulated promoter constructs, driving luciferase expression, and a RhoB expression vector. RhoB suppressed both the basal and the UV-C inducible activity of all promoters tested. The suppressive effect of RhoB on p53-driven gene expression was confirmed by use of the promoter of the base excision repair gene mpg which is regulated in a p53dependent manner. To clarify, whether RhoB-mediated gene repression is affected by the type of C-terminal prenylation, cotransfection experiments using RhoB mutants that are either exclusively geranylgeranylated or farnesylated were performed. Both isoforms had comparable effect as the wildtype. Hence, the inhibitory effects of RhoB on gene expression are independent of its C-terminal prenylation status. As p53 is well known to affect cell death upon genotoxic stress, it is tempting to speculate that inhibition of p53regulated pro-apoptotic functions contributes to the protective effect of RhoB against UV-induced cell killing. In conclusion, the small GTPase RhoB protects mouse fibroblasts from UV-C induced apoptosis. The suppression of gene expression, in particular of p53-regulated pro-apoptotic gene expression, by RhoB may contribute to this protective effect. Alternatively, RhoB might increase the repair of UV-induced DNA damage. 1. Dept. of Toxicology, Johannes Gutenberg University, Mainz, Germany

407 Clostridial actin-ADP-ribosylating toxins induce delayed caspase-dependent apoptosis in mammalian cells Hilger H. (1), Heine K. (1), Popoff M. (2), Barth H. (1) The binary toxins C2 from Clostridium botulinum and Iota from Clostridium perfringens are composed of separated enzyme and binding/translocation components. The latter mediate the transport of the enzyme components into the host cell cytosol, where they mono-ADP-ribosylate G-actin. This leads to depolymerization of actin filaments and cell rounding within 3 hours of incubation, however the intoxicated cells stay viable. Here, we investigated the long-term responses of various mammalian cell lines after treatment with these toxins as well as the fate of the enzyme components in the cytosol. The cytopathic effects of both toxins were non-reversible as intoxicated cells stayed round and no unmodified actin re-appeared in the cytosol. No obvious decrease in the overall amount of actin was observed in the cells thus ADP-ribosylation did not trigger an accelerated degradation of actin in mammalian cells, what is in contrast to the situation in invertebrate cells. Both toxins induced a delayed apoptotic cell death, which was detectable firstly at about 15 to 24 h after application of the toxins. The caspase-3 substrate poly-ADP-ribosyltransferase-1 (PARP-1) was cleaved C2- as well as in Iotatreated cells, an indication of caspase-3 activation and a hallmark of apoptosis. By the use of specific caspase inhibitors, we demonstrated that the caspases 8 and 9 were activated in intoxicated cells. The enzyme component of C2 toxin persisted as an active ADP-ribosyltransferase in the host cell cytosol for at least 48 h. Cells, treated with an artificial toxin that acts exactly like C2I but becomes rapidly degraded in the cytosol, rounded up but recoverd after few hours and unmodified actin reappeared in the cytosol. In conclusion, the long-lived nature of the clostridial actin-ADP-ribosylating toxins the cytosol is likely essential for the induction of delayed apoptosis in mammalian cells. 1. Institute of Pharmacology and Toxicology, University of Ulm Medical Center, Ulm, Germany, 2. Department of Anaerobic Bacteria, Pasteur Institute, Paris, France

408 Auto-catalytic processing of Clostridium difficile toxin B - Binding of inositol hexakisphosphate Egerer M (1), Guttenberg G (1), Genisyürek S (1), Papatheodorou P (1), Aktories K (1) Clostridium difficile toxins A and B, which are the causative agents of antibioticassociated diarrhoea and pseudomembranous colitis, inactivate Rho GTPases by glucosylation. To act in the cytosol, translocation and processing of the catalytic glucosyltransferase is required. Processing of the toxin occurs by auto-catalytic cleavage and is activated by inositol hexakisphosphate (InsP6). Here we studied the inherent protease activity in fragments of toxin B and determined the site of toxin B that interacts with InsP6. We report that a fragment of toxin B, comprised of residues 1-955, is cleaved in the presence of InsP6. In contrast, mutants of the catalytic triad of the putative cysteine protease domain did not cleave this fragment. [3H] InsP6 bound to holotoxin B and to the fragment 1-955, but not to a fragment comprising residues 9002366 or the glucosyltransferase domain (residues 1-544). Binding to the putative cysteine protease domain (residues 544-955) was also observed. InsP6-binding was specific and saturable. Lysine600 of toxin B was identified as essential amino acid for InsP6-binding and for InsP6-dependent activation of the protease activity. 1. Inst. f. Experimentelle und Klinische Pharmakologie und Toxikologie, Albert-LudwigsUniversität Freiburg, Freiburg

409 Activation of STAT transcriptional activity by cytotoxic necrotizing factors (CNFs) Reipschläger S.M. (1), Aktories K. (1), Kubatzky K. (2), Schmidt G. (1) The cytotoxic necrotizing factors are AB-type toxins produced by pathogenic Escherichia coli strains (CNF-1 to CNF-3) or by Yersinia pseudotuberculosis (CNF-Y). CNFs constitutively activate small GTPases of the Rho family by deamidation of a crucial glutamine (Gln 63 in RhoA). These toxins are involved in the pathogenesis of urinary tract infections, enteritis, meningitis and are discussed as a cancer risk factor (Hoffmann and Schmidt, 2004). To investigate the putative role for CNFs in the activation of signal transducer and activator of transcription proteins (STATs), we measured transcriptional activity by a dual luciferase reporter assay performed in Hek293 cells. We found that

STAT transcriptional activity is inducible by CNF-Y and CNF-3 whereas CNF-1 even in high concentration could not stimulate the transcription of the STAT-specific reporter plasmid. The induction of STAT transcriptional activity seems to be dependent on Rhoactivation, given that RhoA activity is stronger induced by CNF-Y and CNF-3 than by CNF-1. Moreover, transiently expressed RhoA leads to an increase in STAT transcriptional activity whereas inhibition of the Rho kinase (ROCK) by Y-27632 caused a complete block of the stimulation by CNFs. Using reporter constructs containing response elements for the different STAT proteins we found that both CNF-Y and CNF-3 activate STAT-3 and STAT-5 transcriptional activity. Considering that a number of tumours show an increased activity of STAT proteins, these findings implicate a new mechanism for a possible carcinogenic role of CNFs. Hoffmann, C., Schmidt, G. (2004) Rev. Physiol. Biochem. Pharmacol. 152, 49-63 1. Institut für Experimentelle und Klinische Pharmakologie und Toxikologie, AlbertLudwigs-Universität Freiburg, Freiburg, Germany, 2. Department of Pathology, University of Michigan, BSRB, Ann Arbor, MI, USA

410 Nanoparticle-induced signaling: The role of membrane structures and receptors Peuschel H. (1), Sydlik U. (1), Grether-Beck S. (2), Abel J. (1), Unfried K. (1) Inhaled nanoparticles have been reported to induce adverse health effects in humans. Our previous studies showed that carbonaceous nanoparticles (CNP) induce proliferative signaling in lung epithelial cells via activation of the MAP kinases ERK1/2. The membrane receptors EGFR and β1-integrins are both involved in this CNP-specific signaling. How this signaling is initiated at the level of the cell membrane and how the receptors mediate the signal response is so far not understood. As possible mediators of receptor crosstalk src family kinases (SFK) were investigated in CNP- and sham-treated RLE-6TN (rat lung epithelial) cells. Moreover, in order to study early signaling events at the level of the membrane, lipid rafts as signaling platforms were isolated and analyzed for changes related to nanoparticle treatment. The induction of proliferative signal cascades involving phosphorylation of ERK1/2, Akt and SFK was studied by Western blot analyses. Particle treatment of cells with 10 µg/cm2 CNPs (Printex 90) resulted in an activation of SFK. Preincubation with the SFK-inhibitor PP2 dose-dependently prevented the activation of the CNP-specific proliferative signaling pathway via ERK1/2 and Akt, indicating that SFK activation is a key step in proliferative signaling induced by nanoparticles. As one member of SFK, c-src was investigated by specific siRNA knockdown experiments. Here, knockdown of c-src resulted in a reduction of the CNPinduced Akt activation. In order to investigate interactions between the described receptors and signaling molecules, co-immunoprecipitation studies were performed showing that EGFR associates with SFK to form a complex. This interaction appeared to be influenced by treatment with CNP. First investigations of isolated lipid rafts from CNP- and sham-treated cells showed that EGFR and SFK are located in these microdomains. Treatment with CNP for 5 min had an impact on the localisation of these signaling molecules resulting in dissociation from lipid rafts. These results indicate an important role for SFK in particle-induced proliferative signaling. Moreover, the observed CNP-induced changes in protein composition of lipid rafts may be an early event in particle-cell interaction triggering NP-specific endpoints. 1. Dept. of Molecular Toxicology, IUF Düsseldorf, Germany, 2. Dept. of Cell Biology, IUF Düsseldorf, Germany,

411 Gene expression signatures of endocrine active compounds in human endometrial cancer cells Boehme K. (1), Simon S. (1), Mueller S.O. (1) Several anthropogenous and naturally occurring substances, referred to as endocrine active compounds (EACs), are able to interfere with estrogen receptor signaling. EACs can therefore either cause adverse health effects in humans and wildlife populations or have beneficial effects on estrogen-dependent diseases, warranting a more thorough analysis regarding their mode of action. The aim of this study was to examine global gene expression profiles in estrogen receptor (ER)-proficient Ishikawa plus and ERdeficient Ishikawa minus endometrial cancer cells treated with selected well-known EACs (diethylstilbestrol (DES), genistein (GEN), zearalenone (ZEA), resveratrol (RESV), bisphenol A (BPA) and o,p'-DDT (DDT)). We also investigated the effect of the pure antiestrogen ICI 182,780 (ICI) on the expression patterns caused by these compounds. We characterized Ishikawa cells regarding their ER mRNA and protein expression status by TaqMan PCR and Western blotting, respectively. Global Transcript levels were quantified 24 h after compound treatment using Illumina BeadChip Arrays (23,000 transcripts) and results were verified by quantitative real-time PCR with TaqMan low density arrays. Gene expression analysis revealed 87 genes with similar expression patterns in response to all EAC treatments in Ishikawa plus, whereas ICI lowered the magnitude or reversed the expression of these genes, indicating ER dependent regulation. Apart from estrogenic gene regulation BPA, DDT, ZEA, GEN and RESV displayed similarities to ICI in their expression patterns, suggesting mixed estrogenic/antiestrogenic properties. In particular, the predominant antiestrogenic expression response of RESV can be clearly distinguished from the other test compounds, indicating a distinct mechanism of action. Divergent gene expression patterns of the phytoestrogens as well as weaker estrogenic gene expression regulation determined for the anthropogenous chemicals BPA and DDT compared to the other compounds warrants a careful assessment of potential detrimental and/or beneficial effects of EACs. The characteristic expression fingerprints and the identified subset of putative marker genes can be used for screening chemicals with an unknown mode of action and predicting their potential to exert endocrine disrupting effects. 1. Institut für Toxikologie, Merck Serono, Darmstadt, Germany

83

412 Integration of genomic, proteomic and metabonomic data for hepatotoxicity assessment Walijew A. (1), Gruhler A. (2), Sieber M. (3), Mally A. (3), Hewitt P. (1) The current drug development process is very time and cost intensive. Genomics, proteomics and metabonomics have undoubtedly provided new biological insight and therefore these omics technologies could provide an opportunity to shift compound attrition to earlier stages in drug development. As part of the FP6–PredTox project, we have integrated transcriptomics together with 2D-DIGE and metabonomics data to ascertain whether this multi-omic systems biology approach would result in improved understanding of hepatotoxicity mechanisms and detection of hepatotoxicity earlier than histopathology. Male wistar rats were repeatedly dosed (1 day, 3 days, 14 days) with a known hepatotoxic compound (EMD 335823). The low dose (ld; 15mg/kg) was the NOAEL and the high dose (hd; 350mg/kg) was chosen to ensure significant hepatotoxicity (liver necrosis, fibrosis and bile duct necrosis/hyperplasia). Gene expression analysis showed that the major affects were on fatty acid and lipid metabolism, suggesting that EMD 335823 acted through activation of the PPARα receptor. Specific PPARα-regulated genes (e.g., ACOX 1, CTE1, CYP4a) were significantly up-regulated after treatment. EMD 335823 is an aldose reductase inhibitor and its inhibition results in increased glucose metabolism and consequently in alterations in fatty acid metabolism. Disturbances in lipid metabolism could result in lower energy resources in the liver, contributing to the pathogenesis of liver damage. From all omics platforms there were several clear effects on the bile duct system followed by tissue reorganization and bile duct proliferation. Looking specifically at the 2 animals with the severest findings a specific gene expression pattern relating to histopathology findings was observed. Other ‘Omics' technologies also showed very clear separation of these 2 animals. However, due to the limited numbers of proteins and metabolites identified, these technologies would not have identified the mechanism of toxicity if used alone. The inclusion of omics profiling allowed a more detailed analysis and identification of potential mechanisms of action, improving the overall understanding of the toxicity of EMD 335823. Additionally, there are indications that a compound specific signature was visible earlier, when no histopathological changes were observed (hd). This shows that there is the potential for shortening existing animal studies, with the added benefit of reducing animal usage (3Rs concept). 1. Dept. of Toxicology, Merck Serono, Darmstadt, Germany, 2. Biopharm. Research Unit, Novo Nordisk, Måløv, Denmark, 3. Dept. of Toxicology, Universität Würzburg, Würzburg, Germany

413 VDAC2 is one of several novel targets of dioxin action revealed by quantitative proteomic analysis of TCDD-exposed 5L rat hepatoma cells Sarioglu H. (1), Brandner S. (2), Haberger M. (2), Jacobsen C. (2), Lichtmannegger J. (2), Wormke M. (2), Andrae U. (2) In an effort to contribute to a better understanding of the hepatic toxicity of the ubiquitous environmental pollutant 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD), a comprehensive quantitative proteome analysis using high-resolution two-dimensional gel electrophoresis was performed on 5L rat hepatoma cells. Cells were exposed to DMSO or 1 nM TCDD for 8 hours, and subsequently proteins were separated on groups of 4 different isoelectric focussing strips with immobilized pH gradients covering overlapping, narrow pH ranges between 3 and 11 in the first dimension and SDS polyacrylamide gels in the second dimension. Using conservative parameter settings for spot detection, we found about 4200 protein spots that were quantitated. 78 protein species corresponding to 73 different proteins were identified as up- or down-regulated following exposure of the cells to the dioxin. The corresponding proteins were identified by MALDI TOF/TOF mass spectrometry. There was an overlap of only nine proteins with those detected as altered by TCDD in our recent study employing the non-gel based isotope-coded protein label (ICPL) method (Sarioglu et al., Proteomics 2006, 6, 24072421) indicating a strong complementarity of the two approaches. For the majority of the altered proteins, an effect of TCDD on their abundance or posttranslational modifications had not been known before. Several observations suggest that a significant fraction of the proteins with altered abundance was induced as an adaptive response to TCDDinduced oxidative stress that was demonstrated using the fluorescent probe dihydrorhodamine 123. A prominent group of these proteins comprised various enzymes for which there is evidence that their expression is regulated via the Keap1/Nrf2/ARE pathway. Other proteins included several involved in the maintenance of mitochondrial energy production and the regulation of the mitochondrial apoptotic pathway. A particularly intriguing finding was the apparent up-regulation of the mitochondrial outer membrane pore protein, voltage-dependent anion channel-selective protein 2 (VDAC2), which was dependent on the presence of a functional Ah receptor and which was not reflected at the mRNA level. In view of the reported central role of VDAC2 as an inhibitor of the activation of the proapoptotic protein BAK and the mitochondrial apoptotic pathway, the present data point to a hitherto unrecognized mechanism by which TCDD may affect cellular homeostasis and survival. 1. Department of Protein Science, Helmholtz Zentrum München, Neuherberg, Germany, 2. Institute of Toxicology, Helmholtz Zentrum München, Neuherberg, Germany

414 Protein extraction from FFPE tissues for toxicological evaluation Truisi G. L. (1), Schmitt C. S. (1), Hewitt P. G. (1) For over a century, fixing tissue in formalin and embedding it in paraffin has been a standard procedure in histopathology. In toxicology formalin-fixed, paraffin-embedded (FFPE) tissues are primary used for histological and immunohistochemical analysis, which exploit the crosslinking nature of formalin. These tissues also contain useful genomic and proteomic information which, based on crosslinking and degradation of nucleic acids and proteins, is problematic to obtain. The high numbers of archived FFPE tissues represent a valuable source of information. Therefore, protein expression profiles associated with toxicological studies might help to identify specific safety biomarkers in future. In this study we analysed the effect of various formalin fixation times on yield and quality of proteins extracted from FFPE rat liver tissues. For this purpose, different commercially available protein extraction systems, the Qproteome

FFPE Tissue Kit (QIAGEN), EpitexTM FFPE Tissue Extraction Kit (Aurelium BioPharma Inc.) and ProteoSOLTM Tissue Extraction System (Expression Pathology) have been utilised. These new kits make it possible to extract proteins, which can be evaluated using downstream applications like SDS-PAGE, Western Blot (WB), two-dimensional gel electrophoresis (2D-GE) and Surface-Enhanced Laser Desorption/Ionisation (SELDI). In contrast to a conventional protein extraction protocol, with these specialised kits it was feasible to extract protein from FFPE tissues. The highest protein yield was obtained using a tissue volume of ~3.4 mm³ and a slight modification of the manufacturer’s protocol, adding a drying step of the pellet prior protein extraction. Application of equal amounts of protein from fresh-frozen, one day and one week formalin fixed samples on an SDS-PAGE revealed i) a reduction of protein yield from one week compared to one day formalin fixation and ii) that only proteins smaller than 65 kDa were detectable. Overall, the protein bands profile from FFPE samples was comparable to fresh-frozen. To summarise, the new protein extraction systems are a promising method for protein extraction from FFPE tissues for use in toxicological evaluations. Furthermore the initial findings allow the interim conclusion, that formalin incubation time is an essential aspect during the process of tissue fixation with regard to the impact on proteins. 1. Institute of Toxicology, Merck Serono Research, Darmstadt, Germany

415 Cloning and functional analysis of bovine, caprine and ovine ABCG2 (MXR/BCRP) in the mammary gland Waßermann L. (1), Lindner S. (1), Halwachs S. (1), Honscha K. (2), Honscha W. (1) Transport of xenobiotics via ATP-binding cassette (ABC) transporters in the mammary gland has enormous, toxicological and nutritional consequences for the suckling and the consumers of dairy products. It was recently demonstrated that the ABC-transporter BCRP (Breast Cancer Resistance Protein) is expressed in alveolar epithelial cells of the mammary gland during pregnancy and lactation, and plays a major role for the active secretion of a variety of drugs (Ivermectin, Baytril), toxins (aflatoxin B1), and carcinogens (PhIP: amino-1-methyl-6-phenylimidazo[4,5-b]pyridin; Trp-P-1: Tryptophan-P-1) into milk. Therefore we want to establish an in vitro cell culture model expressing the bovine, caprine and ovine BCRP to characterize their functional transport activity and to identify different xenobiotic substrates as well as different modulators of the carrier. To address this issue total RNA and mRNA was isolated from the bovine, caprine and ovine mammary gland. After the identification of BCRP gene specific primers, full length clones were created using RACE (rapid amplification of cDNA ends) PCR. The full length clones will be subcloned into an appropriate transfection vector and transfected into MCF7 cells to accomplish 96 transwell assays for functional analyses. The efflux ratio of several substrates and modulators will be evaluated by monitoring the basolateral-to-apical as well as the apical-to-basolateral BCRP mediated transport. 1. Dept. of Pharmacology, Pharmacy and Toxicology, Faculty of Veterinary Medicine, University Leipzig, Germany, 2. Dept. of Veterinary-Physiology, Faculty of Veterinary Medicine, University Leipzig, Germany

416 Evidence for an essential role of cyclophilin A during translocation of the Clostridium botulinum C2 toxin from early endosomes into the cytosol of mammalian cells Kaiser E. (1), Dukart I. (1), Barth H. (1) Clostridium botulinum produces several toxins including the binary C2 toxin. This toxin consists of the binding/translocation component C2IIa, and the separate enzymatic component C2I which mono-ADP-ribosylates G-actin in the cytosol of eukaryotic cells. After receptor-mediated endocytosis the C2 toxin is taken up into the cell and C2I escapes from early endosomes into the cytosol after acidification. For this purpose C2IIa forms pores in the endosomal membranes, and C2I translocates into the cytosol of the eukaryotic cell. We showed earlier that C2I needs to be unfolded to pass the C2IIa pore and has to be refolded in the cytosol to fully display its enzymatic activity. We were able to show, that this step is dependent on the activity of the host cell chaperone Hsp90. Our latest experiments show that cyclosporin A (CsA), an inhibitor of peptidyl-prolyl cis/trans isomerase (PPIase) activity of cyclophilins, significantly delayed the intoxication of various cell lines with C2 toxin and prevented the uptake of C2I into the cytosol. We could show that CsA blocked the pH-dependent translocation of C2I-ADPribosyltransferase activity across membranes of intact cells and of partially purified early endosomes in vitro. In this in vitro-assay, HeLa cell cytosol was added to C2 toxinloaded early endosomes and thereby translocation of C2I ADP-ribosyltransferase activity into the cytosol was induced. This step was blocked in the presence of CsA or an antibody against cyclophilin A. These results suggest that the PPIase activity of cyclophilins facilitate translocation of C2I-ADP-ribosyltransferase from early endosomes into the cytosol. Performing pull-down experiments with biotin-C2I or biotin-C2IN as baits, we revealed a specific binding of cyclophilin A from the cytosol of HeLa cells, suggesting that cyclophilin A interacts with C2I in intact cells. Taken together our results proof the direct interaction of C2I with cyclophilin A and suggest an essential role for cyclophilin A during translocation of C2I across endosomal membranes. To our knowledge, this is the first report describing the involvement of peptidyl-prolyl cis/trans isomerases on the uptake of a bacterial toxin into the cytosol of eukaryotic cells. 1. Institute of Pharmacology and Toxicology, University of Ulm Medical Center, Ulm, Germany

417 A novel cell-permeable transport system for biotinlyated molecules based on the binary Clostridium botulinum C2 toxin Fahrer J. (1), Plunien R. (1), Binder U. (1), Barth H. (1) The C2 toxin produced by C. botulinum is a binary toxin comprising the enzyme component C2I and the binding/translocation component C2II, which is activated by proteolysis to C2IIa. C2I interacts via its N-terminal domain (C2IN) with C2IIa, which is essential for its cellular uptake and transport into the host cell cytosol. Streptavidin, a tetrameric protein from Streptomyces avidinii, is well-known due to its extraordinary high affinity for biotin and is widespread in applications ranging from life-sciences to drug

84

delivery. Genetically engineered streptavidins with a decreased biotin-binding affinity are now available, which allow for a reversible biotin-binding. Aim of our project was to generate a fusion protein, consisting of the N-terminal part of C2I and a dimeric streptavidin with a decreased affinity for biotin. This fusion protein should then be used as a novel, non-toxic transporter for biotinylated molecules in mammalian cells. C2INstreptavidin was cloned and overproduced in E. coli as a GST-tagged fusion protein and purified to homogeneity. To test the biotin-binding properties of the fusion protein, an overlay blot technique and a gelshift assay were established, revealing a concentrationdependent binding of biotinylated proteins and a biotin-labeled oligonucleotide. To exclude cytotoxic properties of the novel transporter, cell viability assays were performed and displayed no effects. Time- and concentration-dependent cellular uptake of the fusion construct was monitored by immunoblot analysis of Vero cells pre-treated with C2IN-streptavidin and C2IIa. To confirm intracellular localization, immunofluorescence analysis and confocal laser scanning studies were conducted, clearly showing a successful cellular uptake. Moreover, a local accumulation of internalized C2IN-streptavidin was observed, most likely representing endosomal localization, which was observed earlier using full length C2I. Further studies will demonstrate the uptake of a biotinylated cargo such as DNA or proteins. In summary, we have successfully engineered and characterized the novel carrier C2IN-streptavidin, which will allow for the specific uptake of biotinlyated molecules into the cytosol of living mammalian cells. 1. Institute of Pharmacology and Toxicology, University of Ulm Medical Center, Ulm, Germany

418 Cyclophilins are required for translocation of the Clostridium perfringens Iota toxin into the cytosol of mammalian cells Barth H. (1), Kroll C. (1), Popoff M. (2), Kaiser E. (1) The binary Iota toxin from Clostridium perfringens is composed of two separate proteins: the enzyme component Iota a (Ia) is an ADP-ribosyltransferase and Iota b (Ib) is the binding/translocation component. At first, Ib facilitates binding of Ia to the cell surface and after receptor-mediated endocytosis of the toxin complex, Ia translocates from the lumen of acidified endosomes through pores, which are formed by Ib in the endosomal membranes. Once in the cytosol Ia ADP-ribosylates G-actin leading to depolymerization of the actin cytoskeleton. We have reported earlier that the host cell chaperone Hsp90 is required for translocation of Ia across endosomal membranes. Here, we investigated whether peptidyl-prolyl cis/trans isomerases (PPIases) are also required for the uptake of Ia into the cytosol of mammalian cells. We found that the specific pharmacological inhibitor cyclosporin A (CsA), which inhibits the PPIase activity of cyclophilins, significantly delayed the intoxication of Iota toxin sensitive Vero cells. We could also show that CsA prevented the translocation of Ia into the cytosol. Moreover, cells treated with Iota toxin in the presence of CsA also showed less ADP-ribosylated actin in the cytosol compared to cells treated without CsA, clearly demonstrating that CsA inhibited the uptake of active Ia ADP-ribosyltransferase into the cytosol. We were able to exclude an influence of CsA on the ADP-ribosyltransferase activity of Ia as well as an interference during the binding of the toxin to the cell surface. Our present studies focus on the role of cyclophilins in membrane translocation of Ia. Taken together our results demonstrate that cyclophilins are involved in the uptake of Ia into the cytosol of mammalian cells. This finding is in line with our recent observation that cyclophilin A facilitates translocation of C2 toxin from acidified endosomes into the host cell cytosol and suggests a common role of these PPIases in cellular uptake of binary actin-ADPribosylating toxins. 1. Institute of Pharmacology and Toxicology, University of Ulm Medical Center, Ulm, Germany, 2. Department of Anaerobic Bacteria, Pasteur Institute, Paris, France

419 Characterisation of the ADP-ribosyl-transferase SpvB in three different strains of Salmonella enterica Völk H. (1), Fahrer J. (1), Enzenmüller S. (1), Hensel M. (2), Barth H. (1) The extra-chromosomal virulence plasmid Spv (Salmonella plasmid of virulence) plays an important role in Salmonella enterica pathogenicity, encoding amongst others for the ADP-ribosyl-transferase SpvB. The SpvB protein catalyzes the modification of G-actin with mono-ADP-ribose resulting in the depolymerization of the actin cytoskeleton. In infected cells, SpvB is translocated into the host cell cytosol from intracellular Salmonella-containing vacuoles. To study the effects of SpvB we used three different S. enterica strains: wild type (WT), a SpvB-deficient strain (Mvp 528) and a strain with an additional plasmid carrying a HA-tagged SpvB (p2967). The aim of our project was primarily to characterize SpvB in these strains in terms of expression and activity and to analyze the secretion of SpvB in more detail. First, all three S. enterica strains were characterized with respect to their growth properties and showed a similar exponential growth. In addition, SpvB was detected in strain p2967 by a polyclonal anti-SpvB antibody, whereas strain Mvp revealed no SpvB expression. The presence of HA-tagged SpvB in strain p2967 was confirmed by immunodetection with an anti-HA antibody. To check the biological activity of the expressed SpvB, we performed an in vitro ADPribosylation assay. Modification of G-actin was observed in WT strain and, to a higher extent, in p2967 strain. In contrast, the Mvp strain did not catalyze the modification of Gactin. Afterwards, we established an infection model in J 774A.1 macrophages. Therefore, we tested the suitable multiplicity of infection (MOI) by using MOIs between 0.1 and 100 and observed the morphological changes during 24 h. MOI values higher than 10 were cytotoxic; thus, a MOI of 1 turned out to be optimal to ensure cell survival and Salmonella replication. Next, we showed the successful uptake of Salmonella into the macrophages by plating lysates from infected 774A.1 cells on LB-Agar. To gain more insight into the translocation of SpvB and to address a possible participation of host cell chaperons, we are currently analyzing the fate of SpvB in infected macrophages. 1. Institute of Pharmacology and Toxicology, University of Ulm Medical Center, Ulm, Germany, 2. Institute of Clinical Microbiology and Immunology, Friedrich-AlexanderUniversity, Erlangen, Germany

420 Uptake and localisation of different sized silica nanoparticles in mammalian cells Al-Rawi M. (1), Strack S. (1), Weiss C. (1) Engineered silica nanoparticles (SiO2-NPs) are used in dyes, varnishes, plastics and glue, as well as in pharmaceuticals, cosmetics and food. However, biological and potentially toxic effects of SiO2-NPs are insufficiently understood. In this study, the cellular uptake and signalling activities of SiO2 nanoparticles was investigated. Fluorescently labelled SiO2 nanoparticles of different sizes (70 nm, 200 nm, 500 nm; Postnova analytics) were examined in human cervical carcinoma cells (HeLa) and murine macrophages (RAW264.7). NP uptake and sub-cellular localisation (nuclei, mitochondria, lysosomes, endosomes) were studied by confocal laser scanning- and epifluorescence-microscopy. Furthermore, the activation of mitogen activated protein kinases (MAPKs), which are key regulators of the inflammatory response, was examined in RAW264.7 macrophages after exposure to SiO2 nanoparticles. Physical characterization by transmission electron microscopy and dynamic light scattering of SiO2-NPs revealed a spherical morphology and a monodisperse size distribution. All NPs, irrespective of size, were detected in membrane-enclosed compartments in the cytosol of HeLa cells. No accumulation of NPs in nuclei or mitochondria of HeLa or RAW264.7 cells could be observed. Furthermore the sub-cellular localisation of SiO2NPs differed with their size and the studied cell type. In RAW264.7 cells SiO2-NPs induced activation of MAPKs dependent on the primary particle size. In conclusion, SiO2-NPs accumulate in epithelial cells and macrophages and trigger, dependent on size, MAPK-activation. Moreover, NP size seems also to effect intracellular distribution by yet ill-defined mechanisms. Finally, no evidence could be provided for nuclear or mitochondrial entry of SiO2-NPs down to the size of 70 nm. 1. Institute of Toxicology and Genetics, Forschungszentrum Karlsruhe, EggensteinLeopoldshafen, Germany

421 Does ozone produce DNA strand breaks? Gschrei C. (1), Hopfer C. (1), Mückter H. (1) Ozone (O3) has an inconclusive record of genotoxicity. It has been regarded as a lung carcinogen in mice and a major air pollutant with a detrimental influence on human health. In vitro experiments have accumulated evidence that DNA strand breaks occur after O3 exposure. However, such effects were mostly observed when clear signals of acute cytotoxicity were present. Unfortunately, the high reactivity of O3 makes quantitative estimates of its dose/concentration-dependent effects difficult. Using an invitro exposure apparatus described previously (J Pharmacol Toxicol Meth 40: 63-69, 1998) we exposed A549 and L929 in standard culture flasks (50mL, 25cm2) to steadyflow O3 (0…4000ppm in the gas phase) up to 4h, delivered in Carbogen carrier gas (95%O2, 5%CO2 at 100mL/min). Controls received carrier gas only. Cells were kept in glucose-free Hanks’s balanced salt solution (HBSS). Cytotoxic damage was assessed by measuring dye exclusion (TrypanBlue), genotoxic damage was determined by calculating Olive tail moments (OTM) from DNA comets in single cell gel electropherograms. O3 inflicted cytotoxic damage to the cells in a concentration- and time-dependent manner. Exposure concentrations were more important than exposure intervals. At higher O3 concentrations (>1000ppm) cytotoxicity was significant, but OTM did not exceed 10 and 25 for L929 and A549 cells, respectively, but A549 cells had generally higher OTM. Equipotent H2O2 concentrations – ≤0.5mM, based on OTM – hardly produced any cytotoxicity at all in either cell line. Higher H2O2 concentrations (>0.7mM) raised OTM to >30, but cell viability was still hardly affected. At low O3 concentrations (<500ppm) both cell lines were similarly susceptible with regard to viability, although OTM were lower with A549 cells, roughly equivalent to 0.1mM H2O2. OTM at lower O3 exposure conditions were not different from baseline values nor were cytotoxicity data. It should be noted that with O3 there was no increase of OTM without a decline of viability in contrast to H2O2. Remarkably, even cells that were apparently severely damaged recovered from O3 when provided with fresh medium after 4h of exposure. At >1300ppm O3 damaged the cells almost irreversibly when O3 exposure intervals exceeded 2h. In our experimental setup O3 exhibited little genotoxic potential at low millimolal(!) concentrations in the gas phase, and the occurrence of DNA comets was always accompanied by severe cytotoxicity leaving considerable doubts on the role of O3 as a primary clastogen. 1. Walther-Straub-Institut, Ludwig-Maximilians-Universität, München, Germany

422 Induction of oxidative stress by Alternaria toxins Fehr M. (1), Burkart J. (1), Baechler S. (1), Pahlke G. (1), Marko D. (1) Mycotoxins of the genus Alternaria are widespread contaminants of food and feed comprising a potential risk for health. Extracts as well as single mycotoxins of Alternaria possess genotoxic and mutagenic properties in vitro. Recently, we have reported that alternariol (AOH), a main metabolite of Alternaria, acts as a topoisomerase poison, with preference to the II alpha isoform inducing DNA strand breaks in human tumor cells (Fehr et al., 2009). The aim of our study was to investigate whether oxidative stress might also be involved in the genotoxic properties of AOH and its monomethylether (AME). Xenobiotic phase I metabolism of both toxins potentially generates reactive catechols or hydroquinones, probably arising from aromatic hydroxylation of AOH or AME (Pfeiffer et al., 2007). These metabolites are capable to form reactive semiquinones and quinones and/or undergo redox-cycling inducing the generation of reactive oxygen species (ROS). We observed that AOH and AME induce the generation of intracellular ROS in human colon (HT29), liver (HepG2) and esophagus (KYSE510) carcinoma cells as measured by dichlorofluorescein (DCF) assay. ROS can induce the Nrf2 (erythroid 2p45 (NF-E2)-related factor-2) / ARE (antioxidative responsive element) pathway. As a consequence the transcription factor Nrf2 separates from its cytosolic binding partner keap 1 (kelch-like ECH-associated protein 1), translocates into the nucleus and binds as a heterodimer to the ARE inducing the transcription of metabolic phase II enzymes such as glutathione S-transferases (GSTs). GSTs are known to inactivate carcinogens by catalyzing the conjugation of electrophiles to glutathione. The question whether AOH and AME induce translocation of Nrf2 into the nucleus was investigated in HT29 cells by Western blot analysis. Our results show that AOH and AME cause Nrf2 translocation at DNA strand breaking concentrations, indicative for the

85

activation of the Nrf2 / ARE pathway. In summary, the Alternaria toxins AOH and AME induce oxidative stress in human carcinoma cells potentially contributing to their genotoxic properties. Their impact on the Nrf2 / ARE pathway indicates a potential role of this antioxidative protection system in the detoxification of these mycotoxins. References: Fehr et al., Mol. Nutr. Food Res., 2009, ahead of print; Pfeiffer et al., Mol. Nutr. Food Res., 2007, 51, 307-316. 1. Institute of Applied Biosciences, Section of Food Toxicology, Universität Karlsruhe (TH), Karlsruhe, Germany

stress on mitochondria, we have taken advantage of the preparative and analytical power of a liquid based electrophoretic separation system, i.e. zone electrophoresis in a free flow device (ZE-FFE). This approach enables (i) the preparation of highly pure mitochondria, (ii) the differential analysis of diverse mitochondrial populations and (iii) the analysis of organelles with altered membrane properties which also occur during oxidative stress. Importantly, analyzed mitochondria are amenable to further analysis at the molecular level and alterations of mitochondrial membrane properties can be investigated, e.g. by functional proteomics, thereby enabling the quest for the critical targets in mitochondrial damage. 1. Institut of Toxicology, Helmholtz Center Munich, Neuherberg, Germany

423 Signaling in angiotensin II-induced DNA damage Fazeli G. (1), Stopper H. (1), Schupp N. (1) Enhanced levels of angiotensin II (Ang II) can cause hypertension and kidney disease. We previously showed that Ang II induces DNA damage in several cell lines from kidney analysed by the comet assay and the micronucleus frequency test via Ang II type 1 (AT1) receptor. Also production of reactive oxygen species was detectable by flow cytometry. Among other effects, Ang II is known to increase nitric oxide synthase (NOS) and NAD(P)H oxidase activity. To verify if the NOS family is involved in the Ang IIinduced genotoxicity, L-NAME as a general inhibitor of the family and 1400W, a specific inhibitor of inducible NOS were added to the cells together with Ang II. None of the inhibitors was able to protect the cells against the genotoxic effect of Ang II, implying that NOS do not play a role in the genotoxicity of Ang II. NAD(P)H oxidase was also targeted with the same strategy using diphenyleneiodonium chloride (DPI), an inhibitor of flavoprotein enzymes such as NAD(P)H oxidase. DPI was able to reduce DNA damage in the cells. One potential mechanism to induce NAD(P)H oxidase is via protein kinase C (PKC). Exploring the role of PKC in induction of DNA damage by Ang II, two specific inhibitors of PKC, rottlerin and sphingosine, were used. Both comet assay and micronucleus frequency test showed a decrease in DNA damage after incubation with these inhibitors. Furthermore, western blot analysis showed a higher phosphorylationdependent activation of PKC after Ang II treatment. Involvement of phospholipase C (PLC) as the activator of PKC was investigated by using the inhibitor U-73122, which successfully protected the cells against Ang II induced DNA damage, while U-73343, the inactive analogue of the inhibitor failed to show the similar effect. A decrease in Ang IIinduced DNA damage was observed after inhibition of PLC, PKC and NAD(P)H oxidase, demonstrating the role of these proteins. This leads to the following preliminary model of signaling in Ang II-induced DNA damage: Binding of Ang II to the AT1 receptor activates PLC, resulting in phosphorylation-dependent activation of PKC which in turn, activates NAD(P)H oxidase. NAD(P)H oxidase then produces reactive oxygen species which cause DNA damage. 1. Department of Toxicology, Julius-Maximilian University of Würzburg, Würzburg, Germany

424 Rosuvastatin protects from angiotensin II induced DNA damage in LLC-PK1 cells Gnana Oli R. (1), Stopper H. (1), Scupp N. (1) Rosuvastatin (RSV) is a synthetic 3-hydroxy-3-methylglutaryl coenzyme A (HMG-CoA) reductase inhibitor, widely available for use in the management of hyperlipidaemia. It has a more potent lipid-lowering efficacy than the other available statins, and significantly more patients receiving RSV achieve LDL-C goals. Many clinical trials have shown that statins substantially decrease cardiovascular morbidity and mortality in patients with and without coronary disease. Many of the actions of statins are thought to be mediated by decreasing reactive oxygen species (ROS) formation. Mainly ROS derived from NADPH oxidase play a critical role in Angiotensin II- (Ang II) mediated hypertension, cardiac hypertrophy, fibrosis and remodeling in the heart and vasculature. Renin angiotensin system (RAS) activation and subsequent release of Ang II exert the vasoconstriction, proliferative, profibrotic and proinflammatory actions. Epidemiological studies exploring the connection between hypertension and cancer incidence found higher cancer mortality in hypertensive patients and an increased risk to develop kidney cancer. Recently we have reported that Ang II induces genomic damage in mammalian cells and isolated mouse kidney, most likely via oxidative mechanism. We have also shown that DNA damage induced by unphysiologic oxidants like H2O2 and phorbol myristate acetate (PMA) was reduced to control values by RSV concentrations starting from 10 nM. RSV restored glutathione levels in cells treated with PMA. RT-PCR and activity assays showed that enzymes of the glutathione metabolism were increased by RSV compared to PMA treatment. Now we report that RSV equally protects in vitro against DNA damage induced by Ang II, which is also elevated in patients suffering from hyperlipidemia. Most interestingly it prevents from the DNA damage induced by Ang II also already at the very low concentration of 10 nM, which is comparable to plasma levels in therapy. 1. Institute of Pharmacology and Toxicology, University of Wuerzburg, Wuerzburg, Germany

425 Differential analysis of mitochondria from pathological situations Zischka H. (1) The exposure of mitochondria to chemically or toxin induced (oxidative) stress is a central menace to life. While it is apparently clear that mitochondria are both, the major cellular producer and the main target of oxidative stress in a variety of severe pathologies and biological aging, a molecular knowledge especially of initial steps in these processes remains elusive. One fundamental problem in this aspect is the limited technical ability to correlate measured or observed mitochondrial damage in affected tissues with pathological processes at the molecular level. While electron micrographs of respective tissues reveal, besides affected mitochondria, a (high) background of intact organelles, current protocols for subsequent mitochondrial isolation are not capable of resolving this heterogeneity. In vitro experiments, on the other hand, frequently use conditions which cause massive deterioration of mitochondria, thereby challenging subsequent conclusions. In order to close the gap between the described limitations of current analytical and preparative methods to analyze pathological affection of oxidative

426 Synthesis of different aminocarbonyl- and methylaminocarbonyl-substituted EMPO derivatives Stolze K. (1), Rohr-Udilova N. (2), Hofinger A. (3), Rosenau T. (3) Several reactive oxygen species are playing a major role in the onset of oxidative stress in biological systems, e.g. superoxide, hydroxyl and alkoxyl free radicals which can be detected using the ESR spin trapping technique. In analogy to our previous studies using carbamoyl-substituted EMPO derivatives we synthetized a series of amide and methylamide-type spin traps and tested their spin trapping abilities. Four (methyl)carbamoyl-substituted EMPO derivatives were investigated, namely 5carbamoyl-3,5-dimethyl-pyrroline N-oxide (CADMPO), 5-carbamoyl-3-ethyl-5-methylpyrroline N-oxide (CAEMPO), 3,5-dimethyl-5-methylcarbamoyl-pyrroline N-oxide (DMMCAPO), and 3-ethyl-5-methyl-5-methylcarbamoyl-pyrroline N-oxide (EMMCAPO). The structure of all compounds was confirmed by 1H and 13C-NMR. The main focus of our investigations was the spin trapping behaviour towards superoxide radicals generated by a xanthine/xanthine oxidase system, but we also tested of a series of different carbon-centered radicals derived from methanol, ethanol, and formic acid, generated in the presence of a Fenton type system, towards the novel compounds. 1. Dept. of Biomedical Sciences, Veterinary University, Vienna, Austria, 2. Div. Gastroenterology and Hepatology, Clinic of Internal Medicine III, Medical University, Vienna, Austria, 3. Dept. of Chemistry, Univ. of Natural Resources and Applied Life Sciences, Vienna, Austria

427 Electron paramagnetic resonance spectrometry as a tool to estimate the biological activity of size-fractionated atmospheric particles Wessels A. (1), Birmili B. (2), Hellack B. (1), Jermann E. (1), Wick G. (1), Albrecht C. (1), Harrison R. (3), Schins R.P.F. (1) Epidemiological studies have demonstrated that increased exposure to ambient particulate matter (PM) is associated with increased respiratory and cardiovascular disease as well as lung cancer. Oxidative stress has been suggested to be a key mechanism of injury caused by PM. In the present study, we used size fractionated PM samples (3-7, 1.5-3, 0.95-1.5, 0.5-0.95 and <0.5 µm), collected at four contrasting locations (three urban sites, one remote background) with a Sierra-Anderson high volume cascade impactor. The H2O2-dependent oxidant generation of the samples was determined by Electron Paramagnetic Resonance (EPR) and the spintrap 5,5-dimethyl1-pyrroline N-oxide. In A549 human lung epithelial cells, we determined the toxicity of samples by the LDH assay as well as the release of interleukin-8 (IL-8). Oxidative DNA damage in A549 cells was measured by using the fpg-comet assay. Marked contrasts were observed in oxidant generation, cytotoxicity, IL-8 release and oxidative DNA damage. The samples collected at the remote background locations showed the lowest EPR signals, were neither genotoxic nor cytotoxic and did not increase IL-8 release. For the samples from the three urban sites, effects were found to depend both on the size fraction and on the sampling location. A significant correlation was observed between the oxidant generating potential and cytotoxicity of the samples. Similarly, the EPR measurements were also found to predict oxidative DNA damage potential of the PM samples. In conclusion, our results further establish that the toxicity of PM strongly varies with the sampling location as well as its size fraction. Measurement of the oxidant generating potential by EPR represents a sensitive method to estimate the toxic potential of PM. 1. Institut für Umweltmedizinische Forschung (IUF), Düsseldorf, Germany, 2. Leibniz Institute for Tropospheric Research, Leipzig, Germany, 3. School of Geography, Earth & Environmental Sciences, University of Birmingham, United Kingdom

428 Resolved in vivo/in vitro discrepancy in methapyrilene induced gene alterations in rat liver and in collagen sandwich cultured primary rat hepatocytes Schug M. (1), Heise T. (2), Günther G. (1), Storm D. (2), Oberemm A. (2), Hengstler J.G. (1) Recent studies have presented evidence that in vivo obtained gene expression data can be used for carcinogen classification, for instance to differentiate between genotoxic and non-genotoxic liver carcinogens. This led several groups to study RNA-expression patterns also in cultured hepatocytes. However, these experiments resulted in conflicting data to the in vivo situation. Examples are the genes Gsk3β and Myd116 that have been reported to be deregulated by methapyrilene (MPy) in hepatocytes in vitro but not in rat liver in vivo. To address this discrepancy, we first optimized an in vitro system with cultured hepatocytes. Testing various culture conditions (collagen coated dishes vs. Matrigel vs. sandwich cultures) and medium additives, we finally decided to use an in vitro system with chemically defined serum free culture medium and hepatocytes between two collagen layers (sandwich culture). Interestingly, the reason for the in vivo/in vitro discrepancy could clearly be identified. MPy in rats has a very short half-life of approx. 2.8 h. Therefore, we observed an induction of Gsk3β and Myd116 after 13 and 18 h that decreased to basal levels after 24 h. Obviously, MPy caused only a transient induction. However, in vitro a permanent induction was obtained due to the constant concentration of MPy in the culture medium. In conclusion the reported difference between in vivo and in vitro data was a consequence of pharmacokinetics and not of qualitatively different behaviour of hepatocytes in vitro and in vivo.

86

1. Systemtoxicology, Leibniz Research Centre for Working Environment and Human Factors, Dortmund, Germany, 2. Federal Institute for Risk Assessment, Berlin, Germany

429 Applicability of rat primary hepatocytes as in vitro system for the detection of gene expression alterations induced by genotoxic and non-genotoxic carcinogens in vivo Heise T. (1), Schug M. (2), Storm D. (1), Günther G. (2), Oberemm A. (1), Hengstler J.G. (2), Lampen A. (1) Toxicogenomic approaches represent a potential tool for predictive toxicology and for gathering information about the involved mode and mechanism of action of toxicants. It has been shown recently that alterations of gene expression patterns in rat liver can be used to detect and categorize liver carcinogens. Due to the ethical problems of the use of animal models, their high costs, time requirement and the changed regulatory legislation in the EU, the development of powerful in vitro test systems is enforced. Recently we could demonstrate that cultivation of rat primary hepatocytes in a collagensandwich is most appropriate to reproducibly detect gene expression alterations induced by the non-genotoxic carcinogen methapyrilene (MP). In the present study we investigated whether an in vitro-in vivo correlation for the deregulation of selected genes upon carcinogen exposure is detectable in the established hepatocyte in vitro system. Therefore, we firstly treated primary rat hepatocytes for 24 hours with a high subcytotoxic- and a low -in vivo relevant- concentration of MP, analyzed the expression of 24 selected genes using real-time RT-PCR and compared the gene expression alterations with existing in vivo data. We could demonstrate that treatment with both concentrations of MP in vitro resulted in qualitatively identical deregulations of the majority of analyzed genes as observed in vivo. Secondly, we analyzed the effect of an additional non-genotoxic (piperonylbutoxide - PBO) and two genotoxic (2-nitrofluorene 2NF and aflatoxin B1 - AB1) agents on the expression of selected genes to further confirm the suitability of the in vitro system to detect gene expression signatures related to carcinogenicity. Selected genes are related to networks of DNA damage response and oxidative stress response. Treatment with different concentrations of PBO, 2-NF and AB1 for 24 hours in vitro resulted in qualitatively identical inductions of the analyzed genes compared to results of in vivo short-term studies. In conclusion, a good in vitro in vivo correlation of carcinogen-induced gene expression alterations could be detected in the established hepatic in vitro system for four different carcinogenic compounds. 1. Federal Institute for Risk Assessment, Berlin, Germany, 2. Leibniz Research Centre for Working Environment and Human Factors (IfADo), Dortmund, Germany

430 Metabolic activation of pro-teratogens in vitro and interaction with target cells for toxicity testing Hettwer M. (1), Fernandes M. (1), Iken M. (2), Nau H. (1), Ott M. (2), Steinberg P. (1) A direct teratogenic effect of a test substance can be detected in vitro by using a target cell system like the embryonic stem cell test (EST). The EST is based on the capacity of murine ES cells (cell line D3) to differentiate into contracting myocardial cells under specific cell culture conditions. The appearance of beating cardiomyocytes in embryonic body outgrowths is a toxicological endpoint to assess the embryotoxic potential of a test compound. However, some compounds are of very low toxicity or not teratogenic by themselves, but are toxic after metabolic activation. For example, only a minor effect is observed in vitro when testing substances like cyclophosphamide (CPA) or valpromide (VPD). But in vivo (e.g. mice) metabolic activation of pro-teratogens like CPA or VPD to the teratogens acrolein/phosporamide mustard and valproic acid, respectively, leads to significant reproduction toxicity. In vitro-methods for the metabolic conversion of test substances have to simulate the in vivo-situation as well as possible with all relevant enzymes and pathways as in the case of freshly isolated hepatocytes. The main objective of our project is to combine primary mouse hepatocytes as a metabolic activation system with D3 cells (EST) as a target system. Since a direct co-culture approach with these two systems is not easy to perform, in a first step several approaches with plain hepatocyte supernatant were tested to determine the maximal pre-incubation time without any detrimental effects on D3 differentiation. To test and improve this approach, the pro-teratogen CPA was used as a test substance. Our results show that after pre-incubation of CPA with hepatocytes for 6 hours, a 50% inhibition of D3 differentiation into contracting cardiomyocytes occurs at significant lower concentrations (about 30-fold) than without pre-incubation. Thirty to sixty percent of CPA is metabolized within a 6 hour-incubation period, as determined by GC/MSmeasurement of CPA-concentrations before and after incubation. Therefore, a metabolic activation system and the EST have successfully been combined. Further steps such as the testing of other compounds (e.g. VPD) as well as studies with freshly isolated hepatocytes from other sources (rat, human) will be performed. This project is funded by the EU under the ReProTect Programme. 1. Dept. of Food Toxicology and Replacement/ Complementary Methods to Animal Testing, University of Veterinary Medicine, Hannover, Germany, 2. Centre of Experimental and Clinical Research: Twincore, Hannover Medical School, Germany

431 Coculture-based in-vitro assay for quantitation of sensitizing potential Wanner R. (1), Sonnenburg A. (1), Stahlmann R. (1) Introduction: Restriction of animal experiments for safety assessment intended us to develop a new in-vitro assay for prediction of sensitizing potential. The design of the new assay is based on current immunological knowledge on pathogenesis of contact dermatitis. It integrates dendritic cells (DC) and keratinocytes which are critically involved in allergy development. The read-out system uses molecular responses typically occurring after encounter of the skin with contact allergens. We emulated the in vivo situation by setting up a loose-fit coculture of primary human keratinocytes and of allogenic DC related cells. The new assay was entitled LCSA (loose-fit coculture-based sensitization assay). Results: Sensitization could be determined by analyzing the expression of the DC maturation marker CD86. We have tested 10 known sensitizers of various potency (Brandowski’s base (PPD), toluene-2,5-diamine (PTD), 1-chloro-2,4-

dinitrobenzene (DNCB), trinitrobenzene sulfonate (TNBS), cinnamaldehyde (CA), hexylcinnamaldehyde (HCA), isoeugenol, trimellitic anhydride (TMA), nickel) and 3 nonsensitizers (salicylic acid (SA), sodium lauryl sulphate (SLS), phenol). The new assay differentiated sensitizers from non-sensitizers and irritants correctly. Estimation of the concentration required to cause a half-maximal increase in CD86-expression allowed quantification of sensitizing potential. To evaluate our data, we have compared results with those of the validated animal-based sensitization test, the murine local lymph node assay (LLNA). We found that the LCSA-results correlated closely with LLNA-data. LCSA and LLNA achieved analogous grouping of allergens into categories like weakmoderate-strong. However, the new assay showed an improved capacity to distinguish sensitizers from non-sensitizers and irritants. The presence of keratinocytes increased the sensitivity to allergens of the cocultured DC-related cells as compared to solely cultured DC. This allowed testing at concentration ranges without general toxicity. Conclusion: The LCSA provides dose-response information, allowing prediction of sensitizing potency of a substance. 1. Inst. of Clinical Pharmacology and Toxicology, Charité, Berlin, Germany

432 Ascaridol – an endoperoxide induces activation of monocyte-derived dendritic cells, monocytic THP-1 cells and skin sensitization in the LLNA Tietze C. (1), Hollender J. (2), Karlberg A. (3), Blömeke B. (1) Ascaridol (1,4-epodioxy-p-menth-2-ene) is an asymmetric monoterpene endoperoxide. Naturally, it is found in the leaves of American wormseed fruit (Chenopodium ambrosioides), boldo tree (Peumus boldus) and in leaves of Artemisia molinieri. Intriguingly, ascaridol has been found in old, oxidized tea tree (Melaleuca alternifolia) oil. Tea tree oil (TTO) is generally composed of terpene hydrocarbons, mainly monoterpenes such as α-terpinene, sesquiterpenes, and their associated alcohols. TTO is widely and increasingly used and involved in various effects including allergic contact dermatitis, and ascaridol has been identified among others as an allergen in TTOsensitized patients. In this study, we asked if ascaridol is a sensitization agent. We synthesized ascaridol and investigated the sensitizing capacity of this endoperoxide by analyzing the phenotype and function of monocyte-derived dendritic cells (MoDCs), human myeloid THP-1 cells that are discussed as a surrogate for MoDCs and studied induced pattern of cytokine release following exposure to ascaridol. In light of the above, we found that ascaridol enhanced IL-8 up to tenfold in each donor. Furthermore, ascaridol-treated MoDC up-regulated CD86 and augmented CD40 expression. In addition, CD54 expression, CD80, CD83 and also CD209 protein expression were upregulated in MoDC after ascaridol incubation. To that end, we investigated the potency of ascaridol to induce sensitization in vivo using the local lymph node assay (LLNA). In conclusion, in vitro and in vivo data emphasized that the endoperoxide ascaridol has a clear potential for sensitization. 1. Dept. of Environmental Toxicology, University Trier, Trier, Germany, 2. Swiss Federal Institute of Aquatic Science and Technology, Eawag, Dübendorf, Switzerland, 3. Dept. of Chemistry, Dermatochemistry and Skin Allergy, Göteborg University, Göteborg, Sweden

433 Streptozotocin – a suitable positive control for the Comet assay in vivo in rat kidney cells Schulz M. (1), Kapp M.-D. (1), Quell T. (1), van Ravenzwaay B. (1), Landsiedel R. (1) Most genotoxicity tests are limited to proliferating bacteria or eukaryotic cells in vitro or proliferating cells or tissues in vivo. The Comet assay (single cell gel electrophoresis) offers, however, a simple, rapid and sensitive method to analyze and quantify DNA damage in any tissues from animals treated with a test substance in vivo. The prerequisite for the application of the Comet assay is the isolation of intact single cells from the tissues. This paper describes the detection of gentoxic effects in kidneys of male and female rats using the Comet assay. To establish the Comet assay in kidneys, we used streptozotocin as positive control substance. The animals were administered 10 to 40 mg/kg body weight intravenously and were sacrificed after six and 24 hours. Whole kidneys, free of blood cells, were obtained by two-step in situ perfusion with EGTA and Collagenase H. After perfusion the kidneys were removed, cut in small pieces and digested using again Collagenase H. The single cell suspension was embedded in agarose gel on a glass slide, lysed overnight at 4°C and then treated with denaturizing buffer to unwind the DNA. The DNA was electrophoresed, neutralized and stained with ethidium bromide. The scoring of tail moments was performed using an image analysis system. The preparation of single cell suspension from the kidneys treated with steptozotocin led to at least 60% viable cells indicated by trypanblue dye exclusion. Six hours after streptozotocin administration a distinct increase of tail intensity was detected at 30 mg/kg b.w. in females but already at 10 mg/kg b.w. in males. Higher doses in males led to a high incidence of hedgehog cells, which are comet images of extremely damaged DNA. In contrast, at 24 hours sacrifice interval clearly increased tail intensities were seen in females already at 10 mg/kg b.w. In males a dose of 20 mg/kg b.w. caused a significant increase of tail intensity but no increase of hedgehog cells. Our study shows that streptozotocin’s genotoxic effect in the kidney is dose-, time- and gender-dependent: Male rats were more sensitive than females early after the treatment (6 h) whereas females were more sensitive at a later sacrifice interval (24 h). This result emphasizes the necessity to examine and optimize experimental conditions to obtain valid information on genotoxic effects from the Comet assay in vivo. 1. BASF SE, Experimental Toxikology and Ecology, Ludwigshafen, Germany

434 Combination of two in vitro-systems to identify compounds with cell transforming activity without animal testing Blume U. (1), Steinberg P. (1), Thierbach R. (1) REACH is a new European Community Regulation on chemicals and their safe use (EC 1907/2006), which entered into force on 1 June 2007. The aim of REACH is to improve the protection of human health and the environment through the better and earlier identification of the toxic properties of chemical substances, including their carcinogenic

87

potential. In order to avoid the use of an extremely high number of experimental animals for this purpose, the development of an in vitro-test system to identify carcinogenic compounds is imperative. The cell transformation assay and the soft agar assay, two well established in vitro-test systems, will be combined to allow an easy and quantitative analysis of the transforming potential of chemical carcinogens. In the “classic” cell transformation assay the permanent adherent growing BALB/c-3T3 A31-1-1 cells are used to determine whether the test compound has tumour initiating and/or tumour promoting properties. Cells are incubated for six weeks with the test compound and the colonies are then counted by light microscopy according to the IARC guidelines. The soft agar test system is used to determine whether cells incubated with a given compound show anchorage-independent growth. After one week the colonies growing in soft agar can be quantified by using a fluorometric assay. In both systems colony growth is the experimental endpoint. In the new assay the BALB/c-3T3 A31-1-1 cells are treated for a short period of time in cell dishes and thereafter are transferred into soft agar. If the cells have been malignantly transformed, a strong fluorometric signal will be observed. In a further step and in analogy to the in vivo situation the test system will be coupled to a metabolizing system. By doing so we expect to establish a high-throughput test system to detect compounds with a transforming potential without having to make use of experimental animal. 1. Institute for Food Toxicology and Analytic Chemistry, Dept. of Food Toxicology and Replacement/complementary Methods to Animal Testing, Universtiy of Veterinary Medicine Hanover, Germany

435 The human dopaminergic neuronal cell line LUHMES as in vitro model for Parkinson´s disease Schildknecht S. (1), Nagel D. (1), Leist M. (1) Parkinson´s disease is characterized by a gradual degeneration of dopaminergic neurons in the substantia nigra. Dopaminergic neurons are continuously exposed to elevated oxidative stress conditions due to the unstable neurotransmitter dopamine that can easily undergo oxidation to form superoxide and a quinine-form capable to react with cysteine residues in proteins or with glutathione to form dopamine-conjugates. For investigations on the molecular events occurring under these conditions, as well as for the validation of potential pharmacological interventions, an experimental human in vitro model that closely resembles the characteristics of dopaminergic neurons in vivo was established. LUHMES are human mesencephalic cells conditionally immortalized with a v-myc retroviral vector. In this system, tetracycline shuts down v-myc expression and allows differentiation into dopaminergic neurons. Tetracycline in combination with dbcAMP and GDNF (glial cell-derived neurotrophic factor) leads to a differentiation into dopaminergic cells within 3-4 days, as shown by the expression of specific markers such as tyrosine hydroxylase or dopamine transporter (DAT). LUHMES cells were validated with respect to their response toward the parkinsonian toxin MPP+. A time-dependent degradation of neurites, accompanied by a loss of cellular ATP and GSH, and increased formation of radical species was observed. These effects were only detected in fully differentiated cells, whereas undifferentiated LUHMES demonstrated no significant response to the same toxic insult. The neurodegenerative effects observed were partially prevented or delayed by co-incubation with the mixed lineage kinase inhibitor CEP1347, or by inhibition of poly-ADP-ribose polymerase (PARP). The involvement of dopamine in the neurodegenerative process was further underlined by application of dopamine transporter, or tyrosine hydroxylase inhibitors that significantly protected against MPP+-induced degeneration. The herein introduced human neuronal cell line closely reflects the unique properties of dopaminergic cells in vivo. This model can not only serve for basic research on the events occurring in neurodegenerative diseases, but can also be used as a screening system within neurotoxicological testing programs. 1. Dept. of Biology, University of Konstanz, Konstanz, Germany

436 Biosensor method for the detection of acetylcholinesterase activity in blood of different species Seeger T. (1), Gonder S. (1), Worek F. (1), Thiermann H. (1) The acute toxicity of phosphororganic compounds is characterized by the inhibition of acetylcholinesterase (AChE), which results in accumulation of the neurotransmitter acetylcholine (ACh). The analysis of AChE activities in blood is important to determine the cholinesterase status of poisoned patients. This work describes the determination of AChE activity in blood via ACh-sensitive electrodes and the SURFE2R technology (SURFE2R workstation, IonGate Biosciences, Germany). This technology is based on solid supported membranes (SSM), which were adsorbed to the gold surface of sensors. Electrical measurements of ACh-transport was mediated by cell membranes containing the organic cationic transporter 2 from rat (rOCT2) which were adsorbed to the SSM by centrifugation with 800 x g. Activation of the rOCT2 was started by a fast injection of an activating solution containing ACh, which resulted in a capacitive current in pA range. Measurements with different concentration of ACh in the activating solution determined a calibrating curve with ca. 60 pA/mM ACh. The AChE activity in pig, guinea pig and human erythrocytes was studied. The kinetics of ACh degradation depended on the amount of applied blood (dilution of pig blood 1:10 V0.5 = 15 min, 1:5 V0.5 = 7 min) and differed between the species. Guinea pig blood showed nearly 50% less ACh-activity compared to human blood. Moreover, the inhibition of blood AChE by paraoxon (1 µM) and reactivation by obidoxime (10 µM) to 84 ± 1% signal amplitude (compared to control) could be detected. The described system uses ACh-sensitive electrodes to directly determine ACh in solutions. Furthermore the system allows pharmacological investigation in high throughput manner. 1. Bundeswehr Institute of Pharmacology and Toxicology, Munich, Germany

437 Metabolite profiling, a tool for early detection of hepatotoxins and their toxicological mechanisms in rats Kamp H.G. (1, 2), Fabian E. (1), Herold M. (2), Krennrich G. (1), Leibold E. (1), Looser R. (2), Mellert W. (1), Prokoudine A. (2), Strauss V. (1), Walk T. (2), Wiemer J. (2), van Ravenzwaay B. (1) Metabolite profiling is based on the determination of endogenous metabolites in biological samples. Compared to other ‘omics’ technologies, metabolite profiling has the advantage that 1) changes of metabolites occur as a physiological consequence of induced effects and 2) it can be determined in minimally invasive matrices such as blood or urine. Based on LC/MS and GC/MS-analysis of blood samples from treated rats and controls, BASF has established a metabolite profiling data base containing the plasma metabolite profile of more than 400 chemicals, agrochemicals and pharmaceuticals. The liver is one of the major targets for chemically induced toxicity and carcinogenicity. Therefore, it is of particular interest to recognize potential liver toxicity including its pathogenesis as early as possible during compound development. In the MetaMap®Tox data base, several metabolite changes correlating with different toxicological modes of action in the liver have been consolidated to form to specific patterns. One of the investigated toxicological modes of action, linked to liver carcinogenesis in rats is peroxisome proliferation. A metabolite pattern was established for this mode of action using model-compounds such as clofibrate, bezafibrate and Wy 14,643. With this pattern other compounds inducing peroxisome proliferation can be identified in the database such as diethyl-hexyl-phthalate, dichlorprop, mecoprop-P or MCPA. Other metabolite patterns were established for liver enzyme induction (based on e.g., Aroclor 1254, phenobarbital) and liver cell necrosis (based on e.g. chloroform, N,Ndimethylformamide). It is possible to classify compounds with respect to their liver toxicity potential according to these metabolite patterns. In conclusion, it could be demonstrated that metabolite profiling is a promising new tool in toxicology for the early identification and characterisation of toxicological modes of action in liver toxicity. 1. BASF SE, Ludwigshafen, Germany, 2. metanomics GmbH, Berlin, Germany

438 Metabolic competences of the chicken embryo used in the toxicity screening test (CHEST) as an in vitro test system for teratogenicity and embryotoxicity Bernshausen T. (1), Noss A. (1), Böhn S.N. (1), Fabian E.J. (1), Kamp H.G. (1), Kaufmann T. (1), Koch M. (1), Wessa P. (1), van Ravenzwaay B. (1), Landsiedel R. (1) The chick embryo has been described to provide a useful test system for the determination of teratogenicity and embryotoxicity. The chicken embryotoxicity screening test (CHEST) uses fertilized hen eggs in very early developmental stages which by legislation are not considered animals. In contrast to other embryotoxicity assays (e.g. whole embryo culture), the CHEST is easier to conduct and does not require the use of mammalian embryos (Jelinek R and Marhan O, 1994). Extended testing of more than twenty compounds with known activity in the CHEST has confirmed its potential to serve as a non-animal method for teratogenicity and embryotoxicity screening. Since the toxicity of many compounds is influenced or governed by their metabolic fate in the organism, the metabolic competence of non-animal test systems is crucial (Coecke et al., 2006). To estimate the metabolic activities of chicken embryos at different development stages, we analyzed subcellular fractions of whole embryos 72 h and 7 days after fertilization, as well as liver subcellular fractions of 14 and 20 day old embryos and adult chickens. Cytochrome P450 (CYP) activities were detected by EROD and PROD assays and glutathione S-transferase (GST) activities were detected by CDNB-conjugation. Metabolic activities of CYP and GST were detectable in 72 h old chicken embryo subcellular fractions. Liver subcellular fractions of 14 days old chicken embryos displayed the highest activities of all studied development stages (including adults). EROD activity of 14 days old chicken embryos were 13.5 nM resorufin/min/mg protein and PROD activity were near detection limit with 0.4 nM resorufin/min/mg protein. Whereas GST activity of 14 days old chicken embryos were 639 nM S-(2,4dinitrobenzyl-)glutathione/min/mg protein. In summary, constitutive metabolic activities were detectable 72 h after fertilization and increases during maturation and declines after day 14 of chicken embryo development. Jelinek R and Marhan O. 1994 Functional and Developmental Morphology 4:317-323; Coecke S, Ahr H, Blaauboer BJ, Bremer S, Casati S, Castell J, et al. Altern Lab Anim 2006; 34: 49-84. 1. Experimental Toxicology and Ecology, BASF SE, Ludwigshafen, Germany

439 A “test battery” of combined methods to investigate cytotoxicity of sulfur mustard and cytoprotective effects of PARP-1 inhibitors Balszuweit F. (1), Heinrich A. (1), Thiermann H. (1), Kehe K. (1) Sulfur mustard (SM) is a strong alkylating agent, interacting both with DNA/RNA as well as proteins. Depending on the severity of the toxic insult, apoptotic and/or necrotic cell death is induced. PARP-1 activation has been shown as a result of SM-induced DNA damage and may result in necrotic cell death and subsequent inflammation. Therefore PARP-1 inhibitors are thought to be beneficial in this situation.Our aim in this study was to distinguish apoptotic from necrotic cell death, to quantify cytotoxic events in general and to investigate potential protective effects, in particular of PARP-1-inhibitors. To achieve this, we combined methods to quantify necrosis (ToxiLight-Assay) with the determination of apoptosis (Cell Death Detection ELISA) and the quantification of the inflammation mediators IL-6 and IL-8.As adenylate kinase (AK) is only released in the event of membrane rupture, AK in the supernatant is an indicator for necrotic cell death. As AK activity can also be quantified from the lysate of surviving cells, the ratio of AK found in supernatant and total AK (lysate and supernatant) serves as an indicator of overall cell survival.DNA fragmentatation with subsequent nucleosome formation is associated with apoptosis. The ratio of nucleosomes to the amount of AK determined from the same lysate provides a quantitative estimate of the extent of apoptosis among these cells. A similar ratio can be calculated to assess the amount of inflammation mediators IL-6 and IL-8 in relation to the cell population.However, when these methods were applied to HaCat cells (immortalized keratinocytes), 24 hours after being challenged by 300 µM SM, we were unable find reproducible effects of PARP-1 inhibition as AK is rapidly degraded in the supernatant under cell culture conditions. Thus, we changed our experimental setup to also monitor the release of AK and

88

inflammation mediators in the first 6 hours after SM exposure.We observed that PARP-1 inhibition reduced necrosis within the first 4-6 hours post exposure. Apoptosis was amplified after approx. 24 hours indicating a shift from necrosis to apoptosis. This is consistent with recently published results (Steinritz et al. 2007). Interestingly, the release of proinflammatory mediators was reduced, possibly due to a link between PARP and the NFκB-system. In summary, the described combination of assays provides more detailed, simultaneous information about cell death and release of proinflammatory cytokines after SM exposure. 1. Bundeswehr Institute of Pharmacology and Toxicology, Munich, Germany

maximum concentration of 1 mM. Tested chemicals showed either an upregulation (VA, RA, lithium, hydroxyurea) or downregulation (caffeine, salicylic acid, thalidomide) of the Wnt signaling pathway. Interestingly, VA and its derivatives showed a graduated response according to their in vivo teratogenic potential. We also found a strong influence of the solvent DMSO and, to a weaker extent, ethanol on the Wnt reporter activity. The results demonstrate that the ReProGlo-assay can detect the influence of embryotoxic compounds and may help to elucidate the complexity of signaling pathways during embryonic development. This study was supported by the EU (ReProTect, LSHBCT-2004-503257).VA derivatives were a kind gift of Prof. Dr. Heinz Nau. 1. Institut für Pharmakologie und Toxikologie, Abteilung Toxikologie, Universität Tübingen

440 Prevalidation data of a rat recombinant androgen receptor binding assay Freyberger A. (1), Ahr H.-J. (2) We had recently published a simple androgen receptor (AR) binding assay using a rat recombinant fusion protein to thioredoxin containing both the hinge region and ligand binding domain of the rat AR and being performed in a 96 well plate format (Freyberger & Ahr, Toxicology 195: 113-126, 2004). Prevalidation of this assay was performed within the ReProTect project. At least three independent experiments per compound were performed on different days, and dilutions of test compounds from deep-frozen stocks, solutions of radiolabeled ligand and receptor preparation were prepared fresh for each experiment. Twelve compounds with different affinities for the AR, dihydrotestosterone (DHT), androstenedione, D(–)-norgestrel, progesterone, prochloraz, 17αmethyltestosterone (Me-T), flutamide, norethynodrel, o,p’-DDT, dibutylphthalate (DBP), vinclozolin, and linuron, were studied. Vinclozolin, due to the lack of metabolic capability of the test system and DBP turned out as non-binders, whereas for o,p’-DDT and linuron due to their limited solubility only partial displacement (10 µM o,p’-DDT: appr. 40%, 100 µM linuron: appr. 35%) of ligand from the AR could be observed. The rank order of AR binding affinity for compounds for which an IC50 could be established was DHT >> Me-T > D(–)-norgestrel >> norethynodrel > progesterone = androstenedione > prochloraz = flutamide. Binding affinities relative to DHT were the most robust parameter and typically were ≤ 25%. Our data suggest that the assay readily detected and discriminated compounds with strong, weak or no affinity for the AR with high accuracy. It is in line with animal welfare considerations, as it avoids the use of laboratory animal tissue as enzyme source, and is adjustable to an intermediate/high throughput format and thus is a promising candidate for a full validation. This work was sponsored in part by ReProTect (LSHB-CT-2004-503257) 1. Dept. of Special Toxicology, Bayer HealthCare AG, Wuppertal, Germany, 2. Dept. of Special Toxicology, Bayer HealthCare AG, Wuppertal, Germany

441 Acetaminophen: Phase I metabolic activation causes teratogenic effects in Zebrafish embryos (Danio rerio) Weigt S. (1), Hübler N. (1), von Landenberg F. (1), Braunbeck T. (2), Broschard T. (1) The Danio rerio embryo test with metabolic activation (mDarT) was developed to assess the teratogenic effects of proteratogens. Induced rat liver microsomes (RLM) were used as a mammalian metabolic activation system (MAS), since they contain many Cytochrome P450 (CYP) isoforms in high concentrations. Acetaminophen (APAP), a widely used analgesic and antipyretic drug, is metabolically activated, mainly by CYP 2E1, to N-acetyl-p-benzoquinone imine (NAPQI), which covalently binds to cellular nucleophiles such as DNA, RNA and proteins. NAPQI is usually inactivated by hepatic reduced glutathion (GSH). The zebrafish embryos were incubated in Tris–buffer under moderate agitation at 32 °C with microsomes, NADPH and APAP for 1 h, beginning at the latest 2 hpf (hours post fertilization). The negative control (APAP alone) and the vehicle control (MAS alone) were incubated in parallel and for statistical robustness 20 eggs were used per test group. After incubation, the fish embryos were transferred individually into 24-well plates filled with fish medium and for the next 72 hours kept at 26°C with a 12 hour-light cycle. Teratogenicity was scored at 24, 48 and 72 hpf using standardized morphological endpoints. All the controls fulfilled the acceptance criteria of ≤10% impaired embryos. RLM were able to bioactivate APAP, resulting in statistically significant teratogenic activity of this compound. The severity of malformations increased with increasing concentrations of APAP. However, as APAP is considered not to be teratogenic in humans or rats, the effects observed in zebrafish embryos likely occur due to the lack of appropriate detoxification mechanisms e.g., via GSH. 1. Institute of Toxicology, Merck KGaA, Darmstadt, Germany, 2. Institute of Aquatic Ecology and Toxicology, University of Heidelberg, Heidelberg, Germany

442 In vitro prediction of teratogenic effects using pathway-specific reporters: the ReProGlo assay Uibel F. (1), Köhle C. (1), Schwarz M. (1) Developmental defects leading to severe malformations and dysfunctions can result from alterations in the differentiating embryo caused by exogenous embryotoxic compounds. During early embryonic development five highly conserved signaling pathways are essential for the regulation of differentiation: the Wnt-, TGF-β-, Notch-, Hedgehog- and RTK-signaling pathway. For in vitro assessment of teratogenic effects, alterations of these pathways could be measured using reporter constructs comprising pathway-specific DNA binding elements upstream of a luciferase reporter gene. In the context of REACH, such in vitro methods for the assessment of reprotoxicity are needed more than ever. To develop a cell-based in vitro test system for reprotoxic effects, in a first approach D3 murine embryonic stem cells were transfected with a reporter construct specific for the Wnt signaling pathway. Stable transfectants in their undifferentiated state were treated for 24h with potential embryotoxic chemicals. A combination of a resazurin-based cell viability assay and a luciferase activity assay was used to determine effects on the signaling pathway. For high throughput, the assay was carried out on 96-well plates. Besides some embryotoxic model compounds like retinoic acid (RA) or thalidomide, we tested valproic acid (VA) and several of its derivatives with graduated teratogenic potential in vivo (Eikel et al., Chem. Res. Toxicol. 19, 272-8; 2006). All chemicals were tested in non-cytotoxic dose ranges, in most cases at a

443 Balance of toxification and detoxification of acrylamide in the rat Feld J. (1), Berger F.I. (1), Baum M. (1), Eisenbrand G. (1) Acrylamide (AA), a heat induced dietary contaminant is found in carbohydrate rich foods in substantial amounts. AA is regarded as a genotoxic carcinogen as a consequence of its biotransformation into glycidamide (GA) through CYP2E1 mediated epoxide formation. GA is thought to be the ultimate genotoxic agent, interacting with the DNA primarily by forming adducts with the N7 of guanine. AA and GA are detoxified via conjugation with glutathione. These adducts are further degraded to urinary mercapturic acids. In this study we investigated the fate of AA and GA at an oral dose of 100 µg AA/kg b.w., reflecting dietary worst case situations. AA was given to male SpraqueDawley rats via drinking water or French fries (AA content: 2800 µg/kg). Rats were daily fed over 1-9 days. Urine was collected over 24h after the last feeding. Mercapturic acids from AA (AAMA) and GA (GAMA) were determined in the urine by HPLC-MS/MS. To determine AA and GA plasma levels by HPLC-MS/MS blood was collected at different time points between 30min and 4h after AA uptake. At the end of the treatment animals were sacrificed and blood and liver samples isolated. Adducts of AA and GA with the Nterminal valine of hemoglobin (Hb) isolated from red blood cells were determined by GC/MS. Potential induction of DNA damage was examined in leukocytes and hepatocytes using the comet assay. Additionally, GA adducts with N7 of guanine were analyzed in liver DNA using HPLC-MS/MS. AA-Val adduct levels in Hb reflected the cumulative AA dose. Under all conditions tested, GA-Val never exceeded background levels. Only traces of GA were found in plasma. Thus, the GA level in blood appeared to be too low to influence GA-val significantly. This view is supported by the absence of significant DNA damage in blood and liver cells. The results allow to conclude, that AA and GA at the dose levels tested, are efficiently detoxified via GSH conjugation. Taken together, at AA uptake levels about 20fold the average daily consumer exposure detoxification pathways in the rat appear to predominate. Supported by via AiF, the FEI and the BLL 1. Dept. of Food Chemistry and Toxicology, University of Kaiserslautern, Kaiserslautern, Germany

444 Acrylamide exposure and dose-effect relations on the genome-wide gene expression profile of human cells Ehlers A. (1), Leffke A. (1), Hummel M. (2), Zagon J. (1), Broll H. (1), Zeilinger K. (3), Luebberstedt M. (3), Lampen A. (1) The finding that acrylamide is formed in food during heat processing and preparation of food has aroused a new interest in assessing human health hazards associated with exposure to acrylamide. In rodents acrylamide induces tumours at high doses and it has been classified by the IARC (International Agency for Research on Cancer) as a probable human carcinogen. Recently published studies suggest a weak positive correlation between increased dietary acrylamide intake and the increased risk of endometrial, ovarian, breast and renal cell cancer. However, risk assessment of acrylamide remains difficult because the carcinogenic mechanisms are still unknown and in particular the molecular effects of low level acrylamide exposure as by dietary intake are not understood. Therefore we incubated ovary, endometrial and breast cancer cell lines as well as primary hepatocytes with different concentrations of glycidamide, the reactive metabolite of acrylamide. Using high-density DNA microarrays for the detection of high glycidamide dose effects (1 mM) we found numerous differentially expressed genes between treated and untreated cells. A subsequently pathway based investigation of the achieved differential gene list revealed an impact on the MAPK (mitogen-activated protein kinase) pathway. The identified glycidamide regulated genes involved in the MAPK pathway were further validated with real-time RT PCR analyzing the glycidamide doses effects between 0.001 and 1 mM. Low dose experiments with glycidamid concentrations between 0.001 and 0.1 mM resulted in nearly no expression changes of these genes. We suggest therefore that exposure to high doses of glycidamide could induce tumours via the MAPK pathway, but lower doses which nevertheless still exceed the range of dietary intake seem to have no influence on tumour progression. 1. Federal Institute for Risk Assessment, Department of Food Safety, Berlin, Germany, 2. Institute of Pathology, Campus Benjamin Franklin, Charité - Universitätsmedizin, Berlin, Germany, 3. AG Experimentelle Chirurgie, Campus Virchow Klinikum, Charité Universitätsmedizin, Berlin, Germany

445 Perfluoroalkyl acids (PFAAs) act as activators of peroxisome proliferatoractivated receptors (PPAR) alpha and delta Buhrke T. (1), Lehmann K. (1), Lampen A. (1) Perfluoroalkyl acids (PFAAs) are used for the fabrication of various industrial products. These substances are chemically and biologically inert and they are global contaminants of water and soil. Contaminated foodstuff such as drinking water, fish and eggs are the main sources for oral uptake of PFAAs by humans. PFAAs are readily absorbed and accumulate in the blood serum. These substances are not metabolized, and renal excretion is very slow. Among the increasing number of PFAAs that have been detected in human blood samples so far, perfluorooctanoate (PFOA) is the leading substance of

89

the class of PFAAs. PFOA is toxicologically well characterised. It is known that PFOA acts as an agonist of the peroxisome proliferator-activated receptor alpha (PPARα) and therefore has primarily hepatoxic effects. In this study we focussed on the toxicological characterization of the PFOA homologs with different carbon chain length. PFAAs with a chain length in the range between four and twelve carbon atoms were tested in various toxicological assays. It was shown with a cytotoxicity assay using the human hepatocarcinoma cell line HepG2 that cytotoxicity of the substances increases with increasing carbon chain length. None of the tested PFAAs had genotoxic potential as demonstrated with two independent assays (Ames test and V79 micronucleus test). All PFAAs tested in this study were potent activators of both human and murine PPARα. PFOA (backbone of eight carbon atoms) turned out to be the strongest PPARα agonist, and in general, PFAAs with shorter carbon chain length were more potent agonists of PPARα than their homologs with longer carbon backbone. PPARγ was not activated by any of the tested substances; however, we also observed activation of PPARδ by PFAAs. Activation of murine PPARδ was more pronounced than that of human PPARδ. PFAAs with longer chain length were more potent agonists of PPARδ than their homologs with shorter carbon backbone. Our data suggest that additional, PPARαindependent mechanisms may also contribute to the observed liver toxicity of PFAAs. 1. Dept. of Food Safety, Federal Institute for Risk Assessment, Berlin, Germany

446 Studies on toxicological interactions of multiple pesticide residues occurring in food Prutner W. (1), Nicken P. (1), Suckrau I. (2), Haunhorst E. (2), Steinberg P. (1), Hamscher G. (1) Statistical data of a German food survey show that in 41.6% of all investigated food samples at least two pesticide residues were detected. In consequence, it might be questioned if today’s risk assessment approaches can sufficiently estimate the toxicological hazard resulting from exposure to combinations of pesticide residues in food. Cumulative maximum residue limits have already been established for those compounds acting by the same mechanism (e.g. organophosphates, carbamates) taking into account potential dose-addition effects. Response-addition might be adopted in the case of combinations in which the components act differently not influencing each other’s effects. However, in case of interactions there is still no general concept available that would assess the toxicological risk arising from potential synergistic effects. Multiple pesticide residues are resulting from a large number of possible combinations of pesticide mixtures that are applied to crops used for human consumption or for the production of animal feed. To date, there are, however, no empirical data available that would either prove the occurrence of toxic interactions among pesticide residues in food or rule them out. Therefore, the European Food Safety Authority proposes a case-by-case-approach whenever there is a biological plausibility for an interaction between pesticides at low, non-effective doses. In an ongoing research project, ten active compounds were exemplarily being tested for their ability to interact among each other. The basis for their selection was recent residue data from food control investigations in the state of Lower Saxony, Germany. Another reason for their selection was that all of the studied compounds have a similar mode of action by affecting the steroidal biosynthesis. In our study, their impact on the steroidal metabolism was tested using a steroid-producing adrenal cell line. Both single and combined dose-effect relationships were determined. To ensure that changes in hormone production were only caused by specific pesticide actions, cytotoxicity tests were carried out before steroid concentrations were quantified by enzyme-linked immunosorbent assays. First results of this project are presented. 1. Dept. of Food Toxicology and Replacement/ Complementary Methods to Animal Testing, University of Veterinary Medicine, Hannover, Germany, 2. Lower Saxony Federal State Office for Consumer Protection and Food Safety, Oldenburg, Germany

447 Improved methods for biomonitoring of ochratoxin A Muñoz K. (1), Blaszkewicz M. (1), Degen G.H. (1) The mycotoxin ochratoxin A (OTA) is a frequent contaminant found worldwide in foods and feeds. Biomonitoring methods have been applied to assess the internal exposure resulting from dietary intake and from inhalation of mycotoxin-containing dusts [1]. Measurements of OTA in blood (plasma or serum) provide a reliable way to estimate exposure in the general population and in occupational settings, but blood samples are collected by an invasive procedure that involves medical personal. Since urine samples are easier to obtain, biomonitoring of OTA in this matrix is now being explored. Moreover, OTA can be converted to metabolites such as ochratoxin-alpha (OTα): It is formed upon hydrolysis of the peptide bond, but this metabolite has not been included in measurements in human biological fluids so far. Here we report on the development of improved methods that allow the simultaneous detection of OTA and OTα upon HPLC separation and quantification by fluorescence detection in extracts of human plasma, and/or urine. Validation of a method for breast milk samples is currently underway. The limit of determination (LOD) is at present 0.1 ng/ml in plasma and 0.05 ng/ml in urine. The method was applied in a pilot study with volunteers (6 males, 7 females) that provided blood and urine samples on the same day: OTA and OTα were detected in all samples; the average OTA concentration in males (0.26 ng/ml plasma) was slightly higher than in females (0.23 ng/ml plasma) and average OTα levels in plasma were 0.1 and 0.08 ng/ml in males and females. This indicates that OTα is a major metabolite in humans. The OTA concentrations in urine were much lower than in blood, but still measurable with clean-up by immunoaffinity columns. Further analysis will also include of enzymatic cleavage in sample preparation to account for the presence of phase-2 metabolites of OTA and thus further improve the sensitivity. [1] Degen GH, Blaskewicz M, Lektarau Y, Grüner C (2003). Mycotoxin Research 19 (1): 3-7; K.M. is supported by a stipend from DAAD and CONICYT. 1. Leibniz-Institut für Arbeitsforschung an der TU Dortmund (IfADo), Dortmund

448 Cytotoxicity of combinations of mycotoxins in V79 cell cultures Behm C. (1), Degen G.H. (1), Föllmann W. (1) Infection of crops with fungi is a worldwide problem, and can result in contamination of food and feedstuffs with mycotoxins. When contaminated grains or grain products reach the food chain, risks for humans and animals cannot be excluded. Several fungi are able to produce more than one toxin, and it is also known that different moulds can grow on a crop in the field or during storage. Thus, several mycotoxins can be found simultaneously in harvested grains. There are limited data concerning the effects of combinations of different mycotoxins in the published literature so far. In the present study, we used the neutral red uptake assay to determine the cytotoxicity of binary mycotoxin mixtures in V79 cells. The combinations studied in vitro are: ochratoxin A (OTA) and citrinin (CIT), OTA and deoxynivalenol (DON), and enniatin B (EnnB) and DON. The basis for selecting these mycotoxin pairs was their co-occurrence either reported in the literature or detected recently in an analysis of wheat and maize from Germany. Moreover, these toxins show a wide range of cytotoxic potency, with DON being more potent than EnnB and OTA, and CIT as the least potent one. For combinations of OTA and CIT an additive cytotoxic effect was observed over a range of tested concentrations. For the combination of OTA and DON the combined cytotoxic effect was additive at most and several concentrations showed a less-than additive effect. The most interesting observation was made for mixtures of EnnB and DON that exhibited an antagonistic effect compared to DON alone. The effect was significant at EnnB concentrations above 1 µM combined with increasing DON concentrations. In summary, in in vitro studies on the cytotoxicity of three relevant mycotoxin mixtures the combined effects were not stronger than additive. Interestingly, the combination of EnnB and DON showed an antagonistic effect that has to be further investigated in more detail. This study was supported by a grant from the Ministerium für Innovation, Wissenschaft, Forschung und Technologie des Landes Nordrhein-Westfalen, Germany, for a collaborative project entitled „Multikomponentenanalyse für Mykotoxine in Getreide und Futtermitteln“ within EU-Ziel 2-program, NRW 2000-2006, Technologie- und Innovationsprogramm. 1. Leibniz Research Centre for Working Environment and Human Factors (IfADo), TU Dortmund, Dortmund, Germany

449 Absorption and metabolism of alternariol and alternariol methyl ether in the Caco2 Millicell® system Burkhardt B. (1), Pfeiffer E. (1), Metzler M. (1) Alternariol (AOH) and alternariol methyl ether (AME) are secondary metabolites produced by Alternaria. These fungi infest various food items, e.g. cereals, vegetables and fruits. AOH and AME have been associated with esophageal cancer, and there are reports on their genotoxicity and mutagenicity. We have recently examined the oxidative and conjugative metabolism of AOH and AME. The aim of the present study was to investigate absorption and metabolism of these mycotoxins in Caco-2 cells, which represent a useful in vitro model system to simulate the human small intestine. Cells were cultured in 24-well Millicell® plates. A permeable membrane divides each well into two compartments, representing the intestinal lumen (top, apical) and the blood side (bottom, basolateral). Differentiated Caco-2 cells were incubated with 20 µM AOH or AME at 37°C for 1-6 h. The incubation buffer was then was analyzed by HPLC-DAD. AOH as well as AME were conjugated to two glucuronides and one sulfate. After 6 h of incubation, non-metabolized AOH could no longer be detected in the apical compartment. However, notable amounts of AME were detected in the apical compartment after the same incubation time. AOH sulfate was preferentially transported to the apical side, while the glucuronides could be detected in both compartments in equal amounts. At the basolateral side, notable amounts of non-metabolized AOH were detected. AME sulfate was preferentially transported to the basolateral compartment, where only small amounts of non-metabolized AME could be detected. The present study has shown that AOH and AME are conjugated in Caco-2 cells. Especially with AOH, notable amounts could be detected on the blood side. Therefore, intestinal absorption of these mycotoxins must be expected to occur after consumption of infested food. Supported by “Food and Health”, Karlsruhe Institute of Technology. 1. Chair of Food Chemistry, Karlsruhe Institute of Technology (KIT), Karlsruhe, Germany

450 DNA damaging properties of complex Alternaria alternata extracts in HT29 cells Schwarz C. (1), Fehr M. (1), Pahlke G. (1), Marko D. (1) Fungi of the genus Alternaria produce mycotoxins such as alternariol (AOH), alternariol monomethylether (AME), altenuene (ALT) and tenuazonic acid (TA) which have been described as cytotoxic, genotoxic and mutagenic in vitro and in vivo. Due to their ubiquitous occurrence and their potential to considerable toxin production even under relatively low temperature conditions, Alternaria spp. are associated with spoilage of refrigerated foods and stored feeds. As the most prominent Alternaria toxins show relatively low acute toxicity, rather little is known so far about the toxic/genotoxic potential of complex fungal extracts. The aim of our present study was to establish and optimise culture conditions for different strains of Alternaria alternata to obtain extracts with defined, reproducible mycotoxin profiles. The DNA damaging effects of these extracts were investigated in cultivated human tumour cell lines. The following A. alternata strains were used: DSM 62006, DSM 62010, DSM 12633 and DSM 1102. The fungi were cultivated on moist rice at 25°C in the dark and rice samples were extracted with acidified water/ethyl acetate. Levels of AOH, AME, ALT and TA were determined via HPLC/DAD/UV. Genotoxic effects were determined via single cell gel electrophoresis (comet assay) in human colon carcinoma cells (HT29) after 1 h of incubation. A modification was thereby implemented using formamidopyrimidine DNA glycosylase (fpg) from E. coli in order to detect possible oxidative damage to the DNA. The A. alternata DSM 12633 extract exhibited potent genotoxic properties. Compared to the corresponding concentrations of pure AOH, the A. alternata DSM 12633 extract showed a significantly enhanced rate of DNA strand breaks. Moreover, treatment with fpg resulted in a significant increase in DNA strand breaks, whereas the according concentrations of AOH didn’t differ between fpg treated and untreated samples. In summary, the A. alternata DSM 12633 extracts possesses genotoxic properties which

90

substantially exceeds the DNA damaging effects of the Alternaria toxins identified so far. The extract also leads to a considerable induction of fpg sensitive sites in then DNA that are not observed with respective concentrations of AOH, AME or TA. 1. Section of Food Toxicology, University of Karlsruhe (TH), Germany

451 Cylindrospermopsin in Aphanizomenon flos-aquae containing dietary supplements Walther S.C. (1), Seel S. (1), Resch P. (1), Fromme H. (1) Cylindrospermopsin, an alkaloid toxin is highly hepatotoxic and considered to be potentially carcinogenic. It is formed primarily by the cyanobacteria Cylindrospermopsis raciborkii, but recent research showed that cylindrospermopsin can also be formed by the cyanobacteria Aphanizomenon-flos aquae. The use of cyanobacteria in Europe as dietary supplements is rising continuously. Often assigned are positive nutritional effects and/or numerous medical effects, but the evidence of these effects is lacking. Aphanizomenon flos-aquae is one of the most frequently merchandised cyanobacteria spezies and Aphanizomenon flos-aquae is principally harvested from nature. Naturally growing Aphanizomenon flos-aquae often contain predominantly toxin forming strains. In the present study 58 commercially available dietary supplemental products containing Aphanizomenon flos-aquae were investigated. The analysis was performed with a screening method based on direct competitive enzyme linked immunoassay (ELISA) with a specific antibody against cylindrospermopsin, no cross-reaction with other nonrelated toxins or compounds are known so far. The detection limit for this assay is 0.040 ppb (µg/l). In 12 samples no cylindrospermopsin could be detected, while to all other probes values up to maximally 0.25 ± 0.03 µg cylindrospermopsin/g dietary supplement were found. That means a daily admission of 0.4 µg cylindrospermopsin based on the daily intake of the dietary supplement as recommended by the producer. Assuming a consumption of 2 g dietary supplement for adults (body weight about 60 kg) and a no observed adverse effect level of 0.15 mg/kg body weight per day (Rev Environ Contam Toxicol 2000; 163: 113 pp) we recommend a guideline value of 4.5 µg/g cylindrospermopsin for Aphanizomenon flos-aquae containing dietary supplements. Even if the results presented are far under this value cyanobacteria containing food auxiliary should be controlled regularly. 1. Bavarian Health and Food Safety Authority, Oberschleißheim, Germany

mixture of ChOL and some phytosterol-rich diets. The relative concentrations of oxysterols in the diets corresponded to those in the livers. In conclusion, dietary uptake of oxysterols contaminating ChOL and phytosterol-rich diets is reflected in tissue levels of oxysterols and may thus have an impact on human health. Supported by Deutsche Forschungsgemeinschaft (DFG) Me 574/24-2. 1. Institute of Applied Biosciences, Section for Food Chemistry, University of Karlsruhe, Germany, 2. Institute for Nutrition Sciences, Chair for Nutrition Physiology, University of Jena, Germany

454 Application of quantitative risk assessment (QRA) principles: An approach for skin sensitization to hair dyes Goebel C. (1), Rothe H. (1), Kern P. (2), Ryan C. (2), Gerberick G.F. (2) Detailed understanding of consumer exposure is important to apply quantitative risk assessment principles for the safety of hair dye products regarding their skin sensitizing properties. Systemic exposure to dermally applied products is generally assessed through an in vitro skin penetration method (OECD guideline 428) and allows to compare systemic effects assessed in hazard studies to the relevant exposure conditions. In addition, this in vitro skin penetration method provides direct quantification of external concentrations on and in the outer skin layer, the Stratum corneum (SC), under use conditions. Therefore, the sum of the systemically available concentration and the concentration on and in the SC is a refinement of applied dose calculations considered relevant for the induction of skin sensitization. The data obtained at a distinct use concentration under realistic use conditions are referred to as the measured consumer exposure level (mCEL). For example, the mCEL for p-phenylenediamine was 22 µg/cm² determined after application of a typical hair dye product containing 2% PPD under use condition (30 min exposure time and rinsing with water and shampoo). We conclude that consideration of the experimentally obtained mCEL allows for a more robust and accurate QRA for skin sensitization to hair dyes when compared to the sensitizing potency data obtained in the Local Lymph node assay (LLNA) in mice. 1. CPS, Procter & Gamble Service GmbH, Darmstadt, Germany, 2. CPS, The Procter & Gamble Company, Cincinnati, OH, USA.

455

452 Green tea flavonoids inhibit migration of human neural progenitor cells in vitro Gassmann K. (1), Zschauer T.-C. (1), Abel J. (1), Fritsche E. (1) Flavonoids are a large family of phytochemicals which have been proposed to protect against a multitude of diseases. Due to these beneficial effects, these compounds are increasingly used as dietary supplements in functional food. The daily intake of these compounds can exceed 1 g and plasma levels of 0.3-7.5 µM were measured in humans. Their potential for toxicity, however, is an understudied field of research. Many of the biological actions of flavonoids have been attributed to their antioxidant properties and their abilities to interfere with a large number of cell signaling pathways. This interaction with cellular macromolecules is suspected to contribute to the beneficial effects which flavonoids are accredited to, i.e. their anti-carcinogenic properties. However, it is well described that developmental processes involve many of the same signaling pathways than developing tumors do, e.g. for cellular survival and migration. Therefore, the aim of this study is to assess the effects of flavonoids on normal development in vitro. The model system we used constist of normal human neural progenitor (NHNP) cells which grow as neurospheres in culture. Treatment of NHNP cells with different flavonoids (epigallocatechingallate, epigallocatechin, epicatechin, quercetin, hesperitin, kaempferol) from 1 to 10 µM caused no cytotoxicity. However, neurosphere migration was disrupted or even completely inhibited when cells were exposed to epigallocatechingallate, epigallocatechin or epicatechin in the lower micromolar range. These effects on migration were also seen when not the spheres, but only the chambers were pre-incubated with flavonoids. This result points to an interaction of flavonoids with the extracellular matrix. To further determine if migration or already adhesion of the cells was disturbed by flavonoids, we performed an adhesion assay. Epigallocatechingallate in contrast to epigallocatechin- disrupted cellular adhesion in a concentration-dependent manner. It is known that epigallocatechingallate can interact with cellular adhesion proteins and we could already shown that neural progenitor cell migration but not adhesion is β1-integrin dependent. The elucidation of the exact underlying mechanism is subject of current experiments. 1. Dept. Molecular Toxicology, Institut für umweltmedizinische Forschung at the Heinrich-Heine-University Düsseldorf gGmbH, Düsseldorf, Germany

Investigations on fine and ultrafine particles released from laser printers and photocopiers as indoor air contaminants in German office rooms Tang T. (1), Gminski R. (1), Rödelsperger K. (2), Brückel B. (2), Mersch-Sundermann V. (1) A lively discussion is currently underway in Germany on the possibly negative health effects of particles emitted during operation from hardcopy devices. Against this background, we performed a study in cooperation with the Federal Institute for Risk Assessment (BfR) in Berlin on the release of fine and ultra fine particulate matter from laser printers and photocopiers in German offices. Ambient particulate matter (PM) emitted from hardcopy devices was collected in 63 office rooms. PM (0.23 - 20 µm in 16 fractions) mass and submicrometer particles (10 - 1000 nm) were monitored during standby (SB), printing/photocopying (OP) and during post-printing/photocopying (PP) phases using a laser aerosol spectrometer (LAS) and a condensation particle counter (CPC). Scanning and transmission electron microscopes (SEM/TEM) were used for analysis of particle morphology, geometric size and elemental composition. PM2.5 of SB, OP and PP phase were 13.74±11.42 µg/m3, 16.52±12.20 µg/m3 and 26.57±23.54 µg/m3, respectively. In addition, PM10 of SB, OP and PP phase were 28.93±11.42 µg/m3, 33.58±16.41 µg/m3 and 43.84±26.51 µg/m3, respectively. Submicrometer particle number medians for SB, OP and PP phase were 6503, 18060 and 15539 particles/cm3, respectively. An unexpected immediate increase in the number of ultrafine particles, socalled “initial-burst”, calls for further research. Toner particles were spheric and contained irregular-formed nanoparticles, while air-samples impacted onto nucleopore filters contained irregular-formed aggregation and agglomeration (A+A) with primary particle (PT). The investigated toners contain Fe>>Si>Al, while Si>Al were found in the air samples collected during the OP phase. In comparison to the low particle concentration impacted onto nucleopore filter (300 - 2400 A+A/ml), the levels of submicrometer particles detected using the CPC during the OP phase were much higher (10 - 100x). This difference permits to assume that CPC measurement does not only record solid particles, but counts all particles such as water droplets, condensations of volatile organic compounds and micro-aggregates of various substances (e.g. silicon oil). The exact composition of the particles in the air samples must be further clarified. Presently, it is not possible to give a definite answer to the question as to whether emissions from office equipment might be linked to health disorders. 1. Dept. of Environmental Health Sciences, University Medical Center Freiburg, Germany, 2. Inst. und Poliklinik für Arbeits- und Sozialmedizin, Justus Liebig Universität Giessen, Germany

453 Hepatic profiles and concentrations of oxysterols reflect dietary sterol exposure Albrecht A. (1), Honig D. (1), Keller S. (2), Jahreis G. (2), Lehmann L. (1) Phytosterols such as β-sitosterol (SitOL), campesterol and stigmasterol are constituents of plant cell membranes. They are found at high concentration in functional food designed to reduce the serum cholesterol (ChOL) level. Due to their unsaturated structure both ChOL and SitOL can be oxidized in food as well as within the human body. Due to the cytotoxic and potentially genotoxic effects of oxysterols, their origin and the internal exposure are of concern. Therefore, we investigated the concentration of oxysterols in the livers of female guinea pigs fed with different diets containing SitOL or/and ChOL as well as in the diet. Oxysterols were isolated by lipophilic extraction, cold saponification, purified by solid phase extraction, separated by gas chromatography and detected by mass spectrometry. In order to quantify the oxysterol content, a corresponding deuterium-labeled internal standard was used. In the livers of the guinea pigs 7-hydroxy(HO)-ChOL, 7-keto-ChOL and 5,6-epoxy-ChOL were detected. 7-HOSitOLs were detected as well but amounted to less then 10% of the oxycholesterols (oxy-ChOLs) resulting in the following order: 5,6β-epoxy-ChOL > 7-keto-ChOL > 5,6αepoxy-ChOL > 7-HO-ChOL > 7β-HO-SitOL > 7α-HO-SitOL. The concentration of 7-HOChOL and 7-keto-ChOL in the livers of guinea pigs fed with a diet only supplemented with ChOL was significantly higher than that in the livers of guinea pigs fed with a

456 Toxicological assessment of non-relevant metabolites in ground and drinking water - health based limit values, TTC concept and testing strategy Melching-Kollmuss S. (1), Dekant W. (2), Kalberlah F. (3) Introduction: Plant protection products are strictly regulated and an extensive toxicological testing of each active ingredient is required. In Europe a clear Cut-Off criterion is a concentration of the pesticide or its relevant metabolite in ground and drinking water of above 0.1 µg/L. Limits for allowable concentrations of non-relevant metabolites of plant protection products in drinking water between 0.1 and 10 µg/L are discussed and their regulation is not strictly health-based. Research Work: 1. The Acceptable Daily Intakes (ADI) of 429 plant protection products have been converted to ground/drinking-water concentrations using the WHO formula (10% ADI, 2L water/day, 70 kg body weight). The toxicological properties of 47 non-relevant metabolites and their parents have been studied and compared. 2. The transferability of the threshold of toxicological concern (TTC) concept to the regulation of non relevant metabolites in water has been investigated. The protective effect of TTC levels with respect to relevant toxicological endpoints has been investigated. 3. A testing strategy has been developed,

91

considering aspects of costs, time, reduction of animal testing and protection of public health. Different safety factor approaches have been compared. Results: 1. Only for 5% of the plant protection products - which are not toxic, mutagenic, reprotoxic, or carcinogenic - the toxicologically allowable drinking water concentration are < 10 µg/L. In general the non relevant metabolites are not more toxic than the parent. 2. At concentrations < 4.5 µg/L (assuming an intake of 10% ADI via drinking water) no toxicological testing – except for genotoxicity - is necessary. 3. At concentrations ≥ 4.5 µg/L enhanced 28/90-day studies (reproduction, endocrine disruption) are proposed. Safety factors of 300/1000 are used in chemicals regulation at an outside estimate. Conclusion: It is concluded, that the TTC concept should be applied to non relevant metabolites (no toxicological testing except for genotoxicity in concentrations < 4.5 µg/L), that safety factors of 300/1000 are protective and that a default of 10% ADI allocation is precautious. 1. GUP/PP, BASF SE, Ludwigshafen, Germany, 2. Department of Toxicology, University of Würzburg, Germany, 3. FoBiG, Freiburg, Germany

iv injections at 2200, 2300, 0000 and 0100 of placebo (PL) and of 4 x 50 µg ghrelin. In the males nonREM sleep increased from (mean ± S.E.M.) 238.7 ± 14.0 min alter PL to 315.7 ± 8.5 after ghrelin (p < 0.05). Stage 2 sleep increased from 199.0 ± 19.1 to 230.6 ± 142 min (p < 0.05). Other sleep-EEG variables including wakefulness, SWS and stage REM remained unchanged. In women ghrelin did not affect sleep EEG significantly (nonREM sleep: 265.7 ± 9.5 after PL versus 272.7 ± 10.7 min alter ghrelin). GH and cortisol were stimulated after ghrelin in women and men as well. The GH stimulating effect of ghrelin was more distinct than that of GHRH in elderly subjects in our previous study. Our data show that the sleep-promoting effect of ghrelin in male subjects is preserved in the elderly. Similarly to our findings in young women sleep EEG remained unchanged after ghrelin in elderly women. Obviously the menopause does not modulate the effect of ghrelin on the sleep EEG. In both gender ghrelin stimulates GH and cortisol in young and elderly subjects. Acknowledgment: Supported by a grant from the Deutsche Forschungsgemeinschaft (STE 486/5-4). 1. Max Planck Institute of Psychiatry, Munich, Germany

457

460

Analysis of target organs in repeated dose toxicity studies Escher S. (1), Batke M. (1), Bitsch A. (1), Licht O. (1), Melber C. (1), Simetska N. (1), Mangelsdorf I. (1) The toxicological data requirements within REACH increase with the production volume of the substances. An essential study for assessing human health effects is the repeated toxicity in rodents. In many cases experimental studies are already available. However, these studies have been performed according to old guidelines, or without GLP. Also in some case only limited target organs were investigated. In the EU funded project OSIRIS it was the question during the development of integrated testing strategies, if the NOAEL and LOAELs derived from these studies are valid, or if the studies must be repeated. To answer this question the RepDose database (Bitsch et al., 2006) created by Fraunhofer ITEM with funding from Cefic LRI was analysed. This database contains 1712 studies on the repeated toxicity in rats and mice. Both modern Guideline studies as well as older ones investigating only a limited number of organs are included in the database.As expected the number of target organs and the number of effects per target organ is higher, the better the quality of the studies is. The analysis also revealed that most effects occurred in liver and kidney (histopathology and organ weight) as well as blood (hematologic parameters). Furthermore, clinical symptoms and a reduced body weight were frequently observed. In inhalation studies nose and lung was also found to be affected. All organs/endpoints were affected when looking at any doses tested in the study but also at the LOAEL. In addition, in >80% of the cases already at the LOAEL several organs were affected. For example testis, which is not covered often in old studies, was affected in 7% of all cases at the LOAEL, but is only triggering the LOAEL in 1% of the studies. Based on these findings it can be concluded that in most cases also older studies can be used to derive a reliable NOAEL or LOAEL, if liver, kidney, body weight, clinical symptoms and blood have been investigated (plus respiratory tract in inhalation studies). A repetition of such studies should be considered carefully. Bitsch A, Jacobi S, Wahnschaffe U, Simetska N, Mangelsdorf I (2006) Toxicology and Pharmacology 46, 202-210 1. Dept. Chemical Risk Assessment, Fraunhofer ITEM, Hannover, Germany

Pharmacokinetics of voriconazole in children and adults Michael C. (1), Teichert J. (1), Bierbach U. (2), Lange T. (3), Basara N. (3), Niederwieser D. (3), Körholz D. (4), Preiss R. (1) Objectives: The aim of this study was to investigate the pharmacokinetics of voriconazole (VRC) after intravenous administration in children and adults with hematological malignancies who require treatment for the prevention or therapy of systematic fungal infection. Methods: VRC blood trough levels over 10 days of therapy and regular blood levels for up to 12 hours postdose at day 3 were measured in twelve children (7.3 +/- 2.7 years, 24.7 +/- 7.7 kg) and twelve adult patients (43.8 +/- 10.2 years, 79.8 +/- 16.3 kg) by high pressure liquid chromatography (HPLC) with fluorescence detection as previously described (Michael et. al. 2008). Nine infants were treated with VRC of 7 mg/kg every 12 h i.v. during the entire study period (group 1). All adults and 3 children received two loading doses of 6 mg/kg every 12 h i.v., followed by a maintenance dose of 5 mg/kg (children, group 2) and 4 mg/kg (adults, group 3) every 12 h i.v. for a period of 10 days. Results: A total of 453 plasma voriconazole concentration samples (average, 18.9 per individual) were examined. Noncompartmental methods were used to calculate pharmacokinetic parameters. The area under the concentration-time curve was calculated by the linear trapezoidal rule. Mean AUC was highest in children with a VRC dosage of 7 mg/kg, though this did not achieve a statistically significant difference compared to the other two groups. Mean cmax levels and Cl in children with 7 mg/kg were significantly higher compared to adults (13.7 +/-6.2 vs. 5.6 +/- 1.7 µg/ml; 3.8 +/- 7.7 vs. 1.8 +/- 8.8 ml/min/kg). Mean predose plasma concentrations in children with 5 mg/kg were significantly lower than in children with 7 mg/kg or in adults (1.1 +/- 0.2 vs. 3.8 +/- 3.4 vs. 2.9 +/- 2.0 µg/ml). Through levels were not significant different between group 1 and 3. Conclusion: A VRC dosage of 7 mg/kg in children leads to significant higher plasma peak levels and clearance compared to adults received 4 mg/kg, but through levels, elimination half-life and AUC were not significantly different. VRC was safe and well tolerated in children. 1. Institute of Clinical Pharmacology, University of Leipzig, Leipzig, Germany, 2. Department of Pedriatric Hematology and Oncology, University Children´s Hospital, Leipzig, Germany, 3. Department of Hematology and Oncology, University of Leipzig, Germany, 4. Department of Pediatrics, Martin-Luther-University Halle Wittenberg, Halle, Germany

458 The effects of venlafaxine on cognitive functions and quantitatve EEG in healthy subjects Siepmann T.S. (1), Aykac V.A. (1), Oertel R.O. (1), Siepmann M.S. (1), Kirch W.K. (1) Antidepressants that selectively block serotonin uptake may cause unwanted effects on cognitive functions such as impairment of vigilance and memory. This randomized, double blind placebo-controlled study examined the effects of venlafaxine, a selective serotonin and noradrenalin reuptake inhibitor (SSNRI), on cognitive functions and quantitative EEG (qEEG) in humans. 12 healthy male subjects aged 23 to 32 years (26 ± 3 years; mean ± sd) orally received 37.5 mg venlafaxine b.i.d. for 7 days and subsequently 75 mg b.i.d. for another 7 days. After a 14-day washout phase, placebo was administered to the subjects for 14 days under randomized double-blind cross over conditions. Treatment compliance was confirmed by measurement of serum trough concentrations of venlafaxine and its main metabolite o-desmethyl-venlafaxine (ODV) on day 14 using an HPLC method. Venlafaxine did not influence cognitive functions such as choice reaction, memory, psychomotor performance and subjective mood (p>0.05). Placebo resulted in an increase of slow alpha power (14.6 µV2/Hz on day 14 vs. 10.5 µV2/Hz at baseline; median; p<0.05) whereas venlafaxine had no effect on qEEG. In conclusion, multiple dosing with venlafaxine did not influence cognitive functions in healthy humans to a significant extent. 1. Institute of Clinical Pharmacology, Medical Faculty, Technical University, Dresden, Germany

459 Ghrelin promotes sleep in elderly men, but not in elderly women Steiger A. (1), Kluge M. (1), Schüssler P. (1), Yassouridis A. (1) We showed previously that pulsatile intravenous (iv) administration of the neuropeptide ghrelin increases slow-wave sleep (SWS) in young normal male subjects, whereas it does not modulate sleep EEG in young normal female subjects. In both gender the nocturnal secretion of growth hormone (GH) and cortisol is elevated after ghrelin. During ageing the influence of some peptides on sleep changes. For example the sleeppromoting effect of GH-releasing hormone (GHRH) in male subjects is weaker in elderly subjects than in young men. On the other hand the sleep-impairing effects of corticotropin-releasing hormone and somatostatin increase during ageing. It is unknown so far whether the effect of ghrelin on sleep changes during the life span. Furthermore it is unclear whether in young women estrogens contribute to the gender difference in the effect of ghrelin on sleep EEG. To clarify these issues we examined the effect of ghrelin on sleep-endocrine activity in 10 healthy women and 10 healthy men, 60 to 70 years old. In a randomized single-blind study sleep EEG (2300-0700) and the plasma concentrations of GH and cortisol (2200-0700) were examined simultaneously after four

461 Association of asymmetric dimethylarginine (ADMA) with carotid artery intimal media thickness in the Framingham heart study offspring cohort Benndorf R.A. (1), Maas R. (2), Xanthakis V. (3), Polak J. (4), Schwedhelm E. (1), Wolf P.A. (5), Sullivan L.M. (5), Vasan R.S. (5,6), Böger R.H. (1), Seshadri S. (6) Background: The endogenous nitric oxides synthase (NOS) inhibitor asymmetric dimethylarginine (ADMA) is related to all-cause mortality in the general population. Promotion of endothelial dysfunction as well as subclinical atherosclerosis have been proposed as the underlying mechanisms. It was the aim of the present study to assess the relation of ADMA and subclinical atherosclerotic changes in the carotid artery in a large community-based cohort. Methods: In 2958 Framingham Heart Study participants (mean age of 58 years, 55% women) we determined ADMA by LC-MSMS. Common carotid artery intimal-media thickness (CCA-IMT) and internal carotid (ICA)/bulb-IMT were determined by ultrasound. Results: In unadjusted analyses we observed a positive relation of ADMA to both CCA-IMT (b per SD increment 0.012, p<0.001) and ICA/bulb IMT (b per SD increment 0.059, p<0.001). Adjusting for age, gender, systolic blood pressure, body mass index (BMI), antihypertensive treatment, smoking status, Total to HDL cholesterol ratio, diabetes, log C-reactive protein, and serum creatinine), plasma ADMA was not associated with CCA-IMT (p=0.991), but remained significantly and positively related to ICA/bulb IMT (b per SD increment 0.02456, p=0.002). Moreover, in a multivariable model controlling for the Framingham Risk Score group, ADMA remained independently associated with ICA/bulb-IMT, (p<0.001). Conclusions: An elevated plasma ADMA concentration is independently associated with greater ICA-IMT but not with CCA-IMT. These findings are consistent with the preferential promotion of subclinical atherosclerosis at known sites of arterial vulnerability. 1. Clinical Pharmacology Unit, Institute of Experimental and Clinical Pharmacology, University Medical Center Hamburg-Eppendorf, Germany, 2. Institute for Experimental and Clinical Pharmacology and Toxicology, University Erlangen-Nuremberg, Germany, 3. Department of Biostatistics, Boston University School of Public Health, Boston, MA, USA, 4. Department of Radiology, Tufts-New England Medical Center, Boston, MA, USA, 5. The National Heart, Lung, and Blood Institute's Framingham Heart Study, Framingham, MA, USA, 6. School of Medicine, Boston University, Boston, MA, USA

92

462 Effects of tramadol, pregabalin and ibuprofen in a model of human mentholevoked cold hyperalgesia Schmidtko A. (1, 2), Altis K. (1, 2), Kuczka K. (1), Lötsch J. (1), Geisslinger G. (1), Tegeder I. (1) Human experimental models of pain are important tools for studying profiles of analgesic drugs. Here, we established a model of cold hyperalgesia evoked by repeated menthol applications and investigated the analgesic efficacy of ibuprofen, tramadol and pregabalin in menthol-evoked cold hyperalgesia in a randomized, placebo controlled 4way cross-over experimental pain study in 18 healthy volunteers. Tramadol 100 mg significantly reduced menthol-evoked cold hyperalgesia whereas effects of ibuprofen 600 mg and pregabalin 300 mg were not significant. Analgesic effects of tramadol were associated with plasma concentrations and the occurrence of minor side effects, particularly fatigue and nausea. Analgesic effects of pregabalin and ibuprofen were also associated with mild side effects, fatigue, dizziness and difficulties to concentrate for pregabalin and gastric upset for ibuprofen. Five subjects had a ≥ 50% pain reduction with tramadol, three additionally responded to pregabalin, two additionally to ibuprofen and one subjects to all four medications including placebo. The number needed to treat (NNT ≥ 50%) was 4.5, 9 and 18 for tramadol, pregabalin and ibuprofen. These NNTs are well in line with the efficacy of these drugs in patients with neuropathic pain suggesting that menthol-evoked cold hyperalgesia may represent a valid model for neuropathic pain, particularly cold allodynia. 1. pharmazentrum frankfurt/ZAFES, Dept. of Clinical Pharmacology, Goethe University, Frankfurt / Main, Germany, 2. Both authors contributed equally to this work.

463 Atorvastatin mediates pleiotropic effects in hypercholesterolemic patients via the Renin-Angiotensin-System (RAS) by increasing angiotensin-(1-7) Schindler C. (1), Ferrario C.M. (2), Brosnihan B.K. (2), Idelevich E. (1), Siegert J. (1), Kirch W. (1) Aim: The present study was undertaken to investigate if atorvastatin exerts RASinhibiting effects and modulates vascular responsiveness to different vasoactive agents in veins of severely hypercholesterolemic patients. Irbesartan was used as a positive control. Methods: A 12-week study was conducted in a randomized, double-blind crossover design. 12 severe hypercholesterolemic patients were randomized to receive 40 mg atorvastatin per os daily for 1 month and then, after a washout period of at least 30 days, received 150 mg irbesartan per os daily for another month or vice versa according to a randomisation plan. Venous responses to Ang II, histamine and glyceroltrinitrate were compared before and after each treatment using the dorsal hand vein compliance method. Plasma levels of Ang II and the pleiotropic peptide Ang-(1-7) were determined before and after each treatment using a radioimmunoassay (RIA). Pulse wave velocity was additionally measured as surrogate parameter for arterial elasticity using the AtCorMedical Sphygmocor device. Results: Ang II-induced constriction was significantly reduced to 38±8% basal vein size [BVS] after atorvastatin compared to 29±7% before treatment (p=0.04). Correspondingly, Ang II-induced constriction was 40±11% before versus 64±8% BVS after irbesartan treatment (p=0.07). Ang II plasma levels increased only after irbesartan treatment from 17±6 to 52±12 pg/ml (p=0.048) compared to atorvastatin: 9±1 versus 11±3 pg/ml, whereas plasma levels of the pleiotropic RAS-peptide Ang-(1-7) increased from 24±2 to 32±2 pg/ml after atorvastatin (p=0.023) compared to 18±3 before versus 37±4 pg/ml after irbesartan (p= 0.002). Pulse wave velocity as surrogate marker for arterial elasticity was slightly reduced after atorvastatin (7.0±0.4 before versus 6.8±0.4 m/s; p=n.s.) and significantly reduced from 6.9±0.4 m/s before compared to 6.3±0.3 m/s after irbesartan treatment (p=0.046). Conclusion: Increased plasma levels of the pleiotropic peptide Ang-(1-7) after atorvastatin treatment indicate a RAS-inhibitory effect of the statin in patients with lipometabolic disorders which might contribute to its beneficial antiatherosclerotic effects. 1. Institute of Clinical Pharmacology, Medical Faculty, Technical University Dresden, 2. Hypertension and Vascular Disease Research Center, Wake Forest University Medical School, Winston Salem, NC, USA

464 Stereoselective pharmacokinetic and CYP2C19-genotyping as outcome predictors of omeprazole therapy for gastroesophageal reflux disease (GERD) – a pilot clinical trial (phase IV) Reiche I. (1), Tröger U. (1), Mönkemüller K. (2), Kandulski A. (2), Neumann H. (2), Weigt J. (2), Bode-Böger S.M. (1) Background: Approximately 20% of all patients with GERD do not respond to standard dosing treatment with proton pump inhibitors (20mg/d for omeprazole). Omeprazole is metabolized by the hepatic CYP2C19 and CYP3A4 enzymes. As CYP2C19 exhibits genetic polymorphisms responsible for the presence of poor metabolizers (PM), intermediate metabolizers (IM) and extensive metabolizers (EM), different dosage regimes could be required in these populations. Aim: This study should evaluate whether an optimized individual dosage of proton pump inhibitors, adjusted to CYP2C19 genotype, leads to a better pharmakodynamic effect (pH metric acid suppression). Methods: A prospective, randomized, controlled, open-label trial with parallel-group design was initiated. Study population includes 32 helicobacter pylori negative patients with endoscopically diagnosed GERD, Stadium A and B according to Los Angeles classification. Based on CYP2C19 genotype patients were randomized to control group A (n=8 EM; n=8 IM/PM) or to one of the intervention groups B1 (n=8 IM/PM) and B2 (n=8 EM). The control group as well as the intervention group B1 received the standard dose of 20mg omeprazole once daily. Whereas the intervention group B2 received 40mg omeprazole twice daily. On the first and the fifteenth day of omeprazole treatment 11 blood samples over 24h were taken for enantioselective quantification of omeprazole and its main metabolites. Furthermore dual gastric and oesophageal 24-h pH monitoring was conducted. Results: Preliminary study results are presented. IM (n=4) had a 1,8fold higher R-omeprazole AUC (CYP2C19) and a 1,3-fold higher S-omeprazole AUC (CYP3A4) compared to EM (n=8) after 2 weeks of 20mg omeprazole treatment. Furthermore, at higher omeprazole dosage (group B2, n=8) CYP2C19 is apparently saturated and CYP3A4 becomes the main metabolic pathway with a S-omeprazole/R-

omeprazole ratio about 1. The percentage of time with gastric pH > 4 was similar in IM (1x20mg) and EM with 40mg omeprazole twice daily, whereas EM patients with 20mg omeprazole once daily only achieved lower percentage. Conclusion: In contrast to IM, EM seems to require higher omeprazole dosing to achieve comparable therapeutical effects. Whether a CYP2C19 genotype-adjusted dose regime will improve omeprazole therapy could not conclusively be determined. 1. Institute of Clinical Pharmacology, Otto-von-Guericke-University, Magdeburg, Germany, 2. Department of Gastroenterology, Hepatology and Infectious Diseases, Otto-von-Guericke-University, Magdeburg, Germany

465 Nebivolol in hypertensive diabetic patients in regard to gender and age Brixius K. (1), Ladage D. (1), Reidenbach C. (1), Rieckeheer E. (1), Ladage D. (1), Schwinger R.H.G. (2) The present study investigated the effect of treatment with the 3rd generation ß-blocker Nebivolol in female and male patients of different age. 5031 Patients (2210 female, 2821 male) with mild to moderate hypertension and type II diabetes were treated with a daily dose of 5mg nebivolol for 12 weeks. The patients group was well balanced with regard to gender (43,6% female, 55,6% male). Nebivolol reduced systolic blood pressure (SBP) in both genders to a similar extent (female -21.2+/-0.31; male -21.0+/-0.26 mmHg). With regard to age, most significant reduction in blood pressure over the 12-week treatment period was observed in the group of patients below the age of 40 (-24.8+/-1.5mmHg). With advancing age, the decrease in SBP induced by nebivolol differed increasingly reaching statistic significance at the age of 70 and above (p<0.05 to <40yrs & 40-50yrs). This effect was even stronger among the decennial age groups in regard to diastolic blood pressure (DPD). The effect on DBP was similar in men and women (-11.0 vs. 10.7 mmHg; p=0.273). In addition, we found age and sex dependant beneficial alterations of the metabolic parameters. Body weight was significantly more effective reduced by nebivolol in men than in women (-1,1+/-0,1 vs. -0,71+/-0,06 kg, p<0.05). This finding was reflected in the body mass index as well (male: -0,35+/-0,01 vs. female: -,25+/-0,02; p<0.05). Nebivolol is effective in treating diabetic patients suffering from high blood pressure and metabolic syndrome. The significant less lowering effects on blood pressure in the elderly may be attributed to increasing endothelial dysfunction with advancing age. 1. Dept. of Molecular and Cellular Sport Medicine, German Sport University Cologne, Cologne, Germany, 2. Klinik II, Klinikum Weiden, Weiden i.d. OPf., Germany

466 Plasmakinetics of fentanyl in liver insufficient patients undergoing orthotopic liver transplantation (OLT) Baumann F. (1), Regenthal R. (1), Pietsch U.C. (2), Hokema F. (2), Kaisers U. (2), Preiss R. (1) Liver transplantation is an established procedure for the therapy of endstage states of most different liver diseases. Fentanyl is the opioid analgesic of choice in most liver transplantation centres because fentanyl's pharmacodynamics is characterized by cardiovascular stability, low histamine release, and an efficient sedation effect. Plasma kinetics of fentanyl as a high extraction drug is more dependent on perfusion than metabolic CYP P450 capacity. A main anaesthesiological goal is an optimum patient outcome after operation procedure. Therefore, patients should be extubated as shortly as possible with stable cardiovascular circulation because postoperative mechanical ventilation is associated with higher rates of complications. By continuous fentanyl infusion, a fast saturation with an unfavourable change of the pharmacokinetics takes place (increase in context-sensitive half-life). Variations of elimination half-life for fentanyl have been observed from 1.7 to 4.4 hours in general surgical patients. No pharmacokinetic data are available for the fentanyl kinetics in OLT. The influence of unhepatic condition on pharmacokinetics of fentanyl in patients is so far unknown however can be approximated by means of pharmacokinetic models to be 20 percent on the basis of patients without OLT. We studied pharmacokinetics of fentanyl in 17 liver transplant recipients using two different application concepts. In concept A fentanyl was infused until unhepatic phase and was compared to concept B where fentanyl was continuously applied until reperfusion of new liver. Results show no significant differences until the unhepatic phase in the fentanyl concentration with both applications. After the unhepatic phase the mean plasma fentanyl concentrations in treatment concept B were 30 to 90 percent greater than values in treatment concept A. Results show a greater influence of unhepatic condition than theoretically expected. Furthermore, we determined also the protein binding rates of fentanyl that were found comparable with both applications (range 86-96 %). As postoperative fentanyl concentration should decrease as fast as possible a critical evaluation of fentanyl application modes based on clinical and pharmacokinetic measures is needed. 1. Institute of Clinical Pharmacology, University of Leipzig, Leipzig, Germany, 2. Clinic and Outpatients' Department of Anesthesiology and Intensive Care Medicine, University of Leipzig, Leipzig, Germany

467 Imaging of chronic inflammation with an anti-CD4MAB fragment (EP 1645) – results of an adaptive study (Bayesian approach) in patients with active rheumatoid arthritis Siegert J. (1), Hesse S. (2), Schindler C. (1), Idelevich E.(1), Brecht B. (1), Laub R. (3), Pigla U. (3), Sabri O. (2), Kirch W. (1) Background: Rheumatoid arthritis (RA) is a disease that requires early and effective drug treatment to preserve patients` quality of life and to limit joint destruction. Furthermore the disease has an extensive socio-economic impact on healthcare systems. EP 1645 (a CD4mAb-Fragment to be labeled with 99mTc) is under development as a diagnostic tool for an effective and reliable diagnosis to identify patients with active disease who should be treated with an advantageous risk-benefit ratio with highly effective drugs with the potential of severe side effects. Methods: To demonstrate efficacy, safety and tolerability of i.v. injections of EP 1645, an open singledose, adaptive design (5 to 9 volunteers planned, Bayesian approach), phase I study

93

was performed in otherwise healthy volunteers (50-80 years, 5 volunteers, sufficient to ), who suffered from joint disease due to rheumatoid arthritis. Results: After injection activity is located time-dependently in kidney, liver, spleen and bladder displaying the elimination path of the radiotracer. The scintigraphic scans 1 to 8 h after administration of EP 1645 confirm well the spots of clinical disease. Aside these spots - in two cases activity was displayed in additional joints aside the obvious clinical disease (subclinical inflammation?, inflammation prior to obvious clinical disease?, joints previously clinically active?). EP 1645 was safe and well tolerated. No serious adverse event was observed during the trial. Conclusion: Safety and tolerability of EP 1645 as well as efficacy in patients with more questionable clinical disease have to be confirmed in trials with larger numbers of patients during upcoming clinical development. 1. Institute of Clinical Pharmacology, Medical Faculty, Technical University Dresden, Dresden, Germany, 2. Klinik und Poliklinik for Nuclear Medicine, University of Leipzig, Leipzig, Germany, 3. Biotectid GmbH, Leipzig, Germany

468 FNTB promoter polymorphisms are associated with survival in different cancers Bachmann H.S. (1), Lazik A. (1), Bau M. (1), Schmid K.W. (2), Siffert W. (1) Background: Farnesyltransferase inhibitors (FTIs, Lonafarnib, Tipifarnib) showed promising results in vitro while their clinical usefulness remains to be proven. The identification of biomarkers that could identify cancer patients who benefit from treatment with FTIs would be highly important. Therefore, we searched for polymorphisms in the gene encoding the farnesyltranserase β-subunit (FNTB) which could potentially be associated with treatment response to FTIs. Methods: The gene FNTB including its promoter region were sequenced. Functional properties of detected polymorphisms were evaluated by electrophoretic mobility shift assays (EMSA) and reporter assays. Healthy blood donors and patients with different malignancies (multiple myeloma, breast cancer, renal cell cancer and melanoma) were retrospectively genotyped for the detected polymorphisms. Polymorphisms were analyzed for linkage equilibrium and possible association of genotypes as well as haplotypes with risk for cancer and disease progression. FNTB expression in normal and cancer tissue samples were analyzed by quantitative real-time PCR. Both genotypes and diplotypes were correlated with FNTB expression and the patients’ survival. Results: Three common promoter polymorphisms were detected, -609G/C, -179T/A, and -173delG (minor allelic frequencies 0.11, 0.30 and 0.43, respectively). Polymorphisms were in partial linkage disequilibrium and build five major haplotypes. All 3 polymorphisms showed significantly altered binding of transcription factors in ESMA experiments, significant genotype- and haplotype-dependent differences of promoter activity, and genotype-dependently altered FNTB expression both in tumor tissue as well as in lymphocytes from healthy subjects. Kaplan-Meier curves revealed a significant and independent association of the polymorphisms and diplotypes with survival in all 4 cancers. Conclusions: Functional FNTB promoter polymorphisms determine FNTB expression levels and indicate high risk groups within different cancers. Our results suggest that genotyping of the FNTB polymorphisms could predict responders/non-responders of cancer patients treated with FTIs. 1. Institute of Pharmacogenetics, University of Duisburg-Essen, Essen, Germany, 2. Institute of Pathology and Neuropathology, University of Duisburg-Essen, Essen, Germany

469 DRD2 polymorphisms and orofaciolingual/limb-truncal dyskinesias in AfricanCaribbeans with chronic antipsychotic treatment: evidence for possible differential association Al Hadithy A.F.Y. (1,2,3,4), Liu G. (5), Wilffert B. (1,2), Bruggeman R. (1,4), Snieder H. (5), Brouwers J.R.B.J. (1,2), Sing V.J. (6), Matroos G. (7), Hoek H.W. (4,6), van Harten P.N. (4,7,8) Background: Tardive dyskinesia (TD) is a severe antipsychotic-induced movement disorder that has been associated with polymorphisms of dopamine D2 receptor gene (DRD2). Aim: to study the association between orofaciolingual and limbtruncal TD (TDof and TDlt, respectively) and DRD2 SNPs TaqIA (rs1800497), 957C>T (rs6277), TaqID (rs1800498), TaqIB (rs1079597), and -141CIns/Del (rs1799732) in AfricanCaribbeans from Curaςao with chronic antipsychotic treatment. Methods: TD was assessed with Abnormal Involuntary Movement Scale (AIMS). The presence of TDof and TDlt was established by a cut-off score of ≥ 2 on any of AIMS items 1-4 and 5-7, respectively. Furthermore, the sum of the corresponding items was used for the severity. DNA extraction and genotyping was conducted according to standard protocols. Pairwise linkage disequilibrium (LD) between these SNPs was tested by calculating D' as well as r2 using GOLD. Linear and logistic regression analyses were used to conduct single SNP association analyses. The frequencies and probabilities of haplotype pairs were estimated by PHASE 2.0. Haplotype Trend Regression approach was used to test for associations of statistically inferred haplotypes with TDof/TDlt. The most frequent haplotype was the reference haplotype. Association analyses were performed using Stata 10 software. Results: We included 126 subjects (age 49.2±13.4 years and 99 males). Genotype distributions were concordant with Hardy-Weinberg Equilibrium. Pairwise r2 analyses do not suggest linkage disequilibrium between the SNPs studied. Although single SNP analyses did not suggest any significant association, haplotype analyses demonstrated that haplotype 11122 (frequency 5.2%, SNP sequence as mentioned above) may exhibit a significant (beta=4.89; p=0.028) and a nearly significant (beta=2.77; p=0.076) association with higher TDof severity and risk, respectively, after correction for age and gender. Haplotype 11122, however, did not exhibit an association with TDlt severity or risk. Discussion: If genuine, the present results confirm the previous findings by suggesting an association between DRD2 and TD and may furthermore indicate pharmacogenetic differences between TDof and TDlt. Since TDof and TDlt are two distinct clinical entities, this finding may be plausible. Subject to further replication in larger samples, our findings extend the available knowledge on the pharmacogenetics of TD and DRD2 gene. 1. Dept. of Pharmacotherapy and Pharmaceutical Care, GUIDE, University of Groningen, Groningen, the Netherlands, 2. Dept. of Clinical Pharmacy and Pharmacology, Zorggroep Noorderbreedte and De Tjongerschans, Leeuwarden, the Netherlands, 3. Dept. of Clinical Pharmacology, University Medical Center Groningen, Groningen, the Netherlands, 4. Dept. of Psychiatry, University Medical Center

Groningen, Groningen, the Netherlands, 5. Dept. of Epidemiology, Unit Genetic Epidemiology and Bioinformatics, University Medical Center Groningen, Groningen, the Netherlands, 6. Parnassia Psychiatric Institute, The Hague, the Netherlands, 7. Dr.D.R.Capriles Clinic, Curaςao, 8. Symfora Group Psychiatric Centre, Amersfoort, the Netherlands

470 Interindividual variability of multidrug resistance protein 2 (ABCC2) expression in human liver: impact of ABCC2 polymorphisms Lang T. (1), Toliat M. (2), Gödtel-Armbrust U. (1), Klein K. (3), Zanger U.M. (3), Wojnowski L. (1), Fromm M.F. (4), Glaeser H. (4) Background. The reasons underlying the pronounced inter-individual variability in the hepatic MRP2 expression are incompletely understood, but may include genetic polymorphisms of the encoding gene ABCC2. Methods used. We investigated the variability in MRP2 expression in normal liver samples from a white population in relation to 12 single nucleotide polymorphisms (SNPs). Hepatic MRP2 protein (n=178) expression was determined by Western Blot analysis from nuclear/membrane pellets. Each individual was genotyped for the following polymorphisms in ABCC2: -1774delG, 1549G>A, -1023G>A, -1019A>G, -24C>T, IVS3-49C>T, 1249G>A, 2761G>A, 3107T>C, 3563T>A, 3972C>T, 4544G>A. Results. MRP2 expression varied 155-fold. Four polymorphism (-1549G>A; allele frequency: 37.2%, -1019A>G; 37.2%, 3563T>A; 5.1%, 4544G>A; 5.1%) were significantly associated with higher expression. These results were substantiated by haplotype analysis. Fourteen ABCC2 haplotypes were identified in the studied DNA samples. The presence of the ABCC2*4 (-1549A; -1019G; IVS3-49T; 3972T, haplotype frequency: 12%) ABCC2*6 (-1549A; -1019G; IVS3-49T; 3563A; 4544A, 5%), and ABCC2*7 (-1549A; -1019G; IVS3-49T, 3%) haplotypes was associated with higher hepatic expression of MRP2. Conclusion. Our study shows the existence of ABCC2 gene variants which alter the hepatic level of MRP2. The likely functional variants are -1549G>A, and -1019A>G, since they are common to all haplotypes associated with altered hepatic MRP2 protein expression. 1. Institute of Pharmacology, University of Mainz, Mainz, Germany, 2. Cologne Center for Genomics, University of Cologne, Köln, Germany, 3. Dr. Margarete Fischer-Bosch Institute of Clinical Pharmacology, Stuttgart, and University of Tuebingen, Germany, 4. Institute of Experimental and Clinical Pharmacology and Toxicology, University of Erlangen-Nuremberg, Erlangen, Germany

471 Influence of currently known pharmacogenetic modulators on opioid therapy in outpatient pain centers Lötsch J. (1), von Hentig N. (1), Freynhagen R. (2), Doehring A. (1), Griessinger N. (3), Zimmermann M. (4), Sittl R. (3), Geisslinger G. (1) In a multicenter study conducted in tertiary care outpatient pain centers, 352 patients, treated for 63.4 ± 92.4 months with various opioids for pain of various origins, were genotyped for all the variants in OPRM1, COMT, MC1R, ABCB1 and CYP2D6 genes reportedly modulating pain in well defined cohorts. Daily opioid doses ranged from 4 to 1750 mg oral morphine equivalents (133.4 ± 203.2 mg), differed between pain diagnostic groups and decreased in a gene dose dependent manner with the Pglycoprotein variant ABCB1 3435C>T. Pain was rated on average at 3.7 ± 2.6 of a 0 to 10 ordinal rating scale and increased in a gene dose dependent manner with the muopioid receptor variant OPRM1 118A>G. However, both modulations were modest and side effects depended on opioid dose and age but not genetic factors. In conclusion, a need of outpatient pain therapy guided by the presently known functional genetic variants cannot be convincingly concluded from the present data. 1. pharmazentrum frankfurt/ZAFES, Institute of Clinical Pharmacology, Johann Wolfgang Goethe-University, Frankfurt, Germany, 2. Department of Anesthesiology, Universitätsklinikum Düsseldorf, Düsseldorf, Germany, 3. Department of Anesthesiology, Universitätsklinikum Erlangen, Erlangen, Germany, 4. Department of Anesthesia, Intensive Care and Pain Therapy/ZAFES, Johann Wolfgang GoetheUniversity Hospital, Frankfurt, Germany

472 Impact of Disease-Management-Programs (DMP), a new drug-price-regulation (since 2003/2004) and trends in treatment of diabetes mellitus on diabetes medications use (1993-2007) in Germany Wang J. (1), Siegert J. (1), Brecht B. (1), Heinrich Gräfe U. (1), Schröder J. (1), Maywald U. (2), Kirch W. (1) Background: Diabetes mellitus is common, costly, and increasingly prevalent. In Germany, since 2003 diabetes DMP’s (evidence based guideline and management measures) have been universally implemented in the statutory health insurances (SHI), furthermore, in 2004 the new drug-price-regulation (nDPR) was in force. Methods: The data source were the German Drug-Prescription-Reports (1994 to 2008); diabetes medication (dm) use and cost for German SHIs were analyzed. We apply autoregression analysis as forecasting model to assess the impact of DMP and nDPR on diabetes medication use and cost. Results: The nDPR uniquely led to a decline in prescriptions (prescribed Defined Daily Dose, DDD) and costs of almost all antidiabetic in 2004. After start of DMP (2003), the increasing trend of prescribed biguanide DDD was continuously unchanged (inside the 95% CI), compared to the past the prescribed glibenclamide DDD showed a slower declining trend (inside the 95% CI), the prescribed DDD of human insulins, insulin analogues and glinides showed significant lower increases (outside the lower bound of 95% CI). The mean number (DDD) of dm per prescription increased from 49 in 1993 to 59 in 2002. Insulin use increased from 31% of all prescribed DDD in 1993 to a zenith of 43% in 2004 and then remained relatively constant until 2007. Glibenclamide use decreased from 60% of total prescribed DDD in1993 to 25% in 2007, biguanide use increased from 4% in 1993 to 26% in 2007, use of alpha-glucosidase inhibitor decreased from 8% in 1996 to 1% in 2007, and glitazone use increased from 0.3% in 2000 to 4% in 2007. The mean cost per prescription increased from 60 DM in 1993 to 92 DM in 2000 and 52 € in 2002, and the mean cost per DDD increased from 1.3 DM in 1993 to 1.7 DM in 2000 and 0.9 € in 2003 and then

94

decreasing to 0.86 € constantly. Conclusions: The diabetes DMP led to relatively reasonable evidence based medication therapy. Increasingly complex and costly diabetes treatments are being applied to an increasing population. But will these changes in medication use result in correspondingly improved outcomes? 1. Institut für Klinische Pharmakologie, Med. Fakultät, TU Dresden, Germany, 2. AOK Plus, Bereich Arzneimittel, Dresden, Germany

473 Discrimination of serotonin transporter polymorphism genotypes using real time PCR Schmidt D. (1), Läer S. (1) Background and objective: The serotonin transporter (HTT) is a key drug target for major human diseases affecting the central nervous system and the cardiovascular system. Serotonin transporter polymorphisms may influence pharmacodynamics and the strongest effect has been found in an insertion / deletion of a 44bp DNA fragment within the HTT promoter region (5-HTTLPR), leading to a long (L) or short (S) allele. In latest research an additional single mutation appears within the 44 bp sequence. Hence L allele carriers are subdivided into LA or LG carriers leading to strong differences in mRNA expression and therefore to different HTT levels. However, the genetic discrimination between L and S allele carriers has been established by time consuming gel electrophoresis. Therefore, our aim was to implement a real time PCR method using a relative quantification technique for quick and precise identification of L and S allele carriers. Method: We used the 2-∆∆Ct calculation with one probe synthesized according to Hut et al. (2006) hybridising specifically to the 44bp L insertion sequence (target). As an endogenous control a probe was designed hybridising specifically to a sequence located within the human albumin as a housekeeping gene. Due to the constant and uniform expression in the genetic environment our target gene was normalized to albumin. For validation, 77 healthy volunteers were genotyped according to their S and L allele genetic status. An additional DNA sequencing of all samples was performed for confirmation and also discrimination of LA and LG carriers. Results: Out of 77 volunteers, we identified 23 S/S (30%), 34 L/S (44%) and 20 L/L (26%) carriers. These results were comparable to published data (Nakamura et al 2000, Hu et al 2006). DNA sequencing analyses showed 100% overlap with published genotype sequences and confirmed precision of the identified genotypes. Furthermore all L allele carriers could be subdivided into LA or LG carriers. Conclusion: Using real time PCR technology with relative quantification and albumin as an external standard, a quick and precise identification of L and S allele carriers is possible. Therefore our aim is now to implement a PCR method which also allows us subsequent to L/S, a LA/LG allele differentiation using specific probes. This will be helpful in investigations of drugs targeting the HT transporter and in understanding the influence of HTT polymorphisms in the pharmacodynamic response. 1. Klinische Pharmazie und Pharmakotherapie, Heinrich-Heine-Universität, Düsseldorf, Deutschland

474 Incidence and pattern of enquiries about cardiovascular drugs - experiences of the drug information service for patients at the Institute of Clinical Pharmacology in Dresden Schröder J. (1), Idelevich E. (1), Heinrich-Gräfe U. (1), Siegert J. (1), Schindler C. (1), Maywald U. (2), Kirch W. (1) Background: Cardiovascular diseases are the most common causes of death in Germany. Successful therapy of these diseases depends, among other things, on the cooperation of the patient. Due to the variety of treatment options and the short time available during the physician consultation, many patients have open questions about their drug therapy. At worst, they could stop their treatment for these reasons. Hence, since 2001, a drug information service for patients has been established at the Institute of Clinical Pharmacology in Dresden, Germany, to provide adequate information about drugs. Since 2007 a special service exists for the patients insured with AOK PLUS (one of the largest health insurances in Germany). Methods: The patients who used the service were interviewed concerning their socio-demographic characteristics, reason for enquiry, number and kind of drugs taken, and diseases. The data were recorded in a relational database. In the present evaluation, all enquiries from patients insured with AOK PLUS from January 2007 to October 2008 were analyzed descriptively with a main focus on cardiovascular drugs. Results: In the evaluated period, 488 enquiries were registered in total. Women used the service more frequently (63,7%) than men. The largest percentage of enquiries originated from the age group 60 to 79 years (54.7%). 35.9% of all drugs taken by the patients were cardiovascular drugs, being the leader among the other drug groups. Cardiovascular drugs were also the most common reason for an enquiry (33.7%), followed by drugs for the nervous system (21.3%). Patients’ unmet information needs were mainly related to beta blocking agents (27,2%), agents acting on the renin-angiotensin system (24,4%), cardiac therapy (11,4%) and lipid modifying agents (10,6%). The need for information about possible drug adverse effects was the reason for contacting the service in 46,3% of all enquiries concerning cardiovascular drugs. 38,0% of the patients asked about possible drug interactions and 17,3% of the enquirers wanted information about the costs of a particular cardiovascular drug. Conclusions: Questions about cardiovascular drugs are the most common reason for contacting the drug information service for patients. 1. Institut für Klinische Pharmakologie, Medizinische Fakultät, Technische Universität Dresden, Dresden, Germany, 2. AOK Plus, Bereich Arzneimittel, Dresden, Germany

475 Biotransformation of dextromethorphan in human surgical liver samples: genetic and other disposition factors Kuhn U.D. (1), Schütz E. (2), Settmacher U. (3), Hippius M. (1), Müller D. (2), Lupp A. (2) Dextromethorphan (DM) on one hand is O-demethylated to dextrorphan, mainly via the genetically polymorphic expressed cytochrome P450 (CYP) 2D6, and consecutively Ndemethylated (mostly by CYP3A4) to 3-hydroxymorphinan (3-HM). The alternative

pathway is the primary N-demethylation of DM (chiefly via CYP3A4) to 3methoxymorphinan (3-MM) with following O-demethylation (predominantly via CYP2D6) to 3-HM. In the present study we investigated the impact of the CYP2D6 polymorphism (regarding alleles *3, *4, *5 and *6) in comparison to the influence of other factors on DM biotransformation activities in 94 histologically tumor-free human liver samples obtained from surgical waste. Genotype was assessed from whole blood by PCR/restriction enzyme analysis. DM biotransformation activities in vitro (at 200 µM) were determined in the 9000g supernatants of the livers with 40 min incubation time. With 94 sample donors investigated, 1 (1%) was heterozygous for CYP2D6*3, 61 (65%) were heterozygous and 4 (4%) homozygous for CYP2D6*4, 5 (5%) heterozygous for CYP2D6*5 and 3 (3%) heterozygous for CYP2D6*6, corresponding to 56 (60%) extensive metabolizers, 33 (35%) heterozygous extensive metabolizers and 5 (5%) poor metabolizers. In vitro formation of 3-HM was significantly decreased and the metabolic ratio 3-MM/3-HM increased in livers of donors homozygous or heterozygous for CYP2D6*4 or CYP2D6*5 and correspondingly in livers of heterozygous extensive metabolizers and poor metabolizers in comparison to those of wild type subjects. There was no significant effect of age, gender, body mass index or smoking habit or of diagnosis leading to liver surgery (adenoma, hepatocellular or cholangiocellular carcinoma, metastases of gastrointestinal tract, kidney or breast cancer or of endocrine tumors) on 3-HM formation or 3-MM/3-HM metabolic ratio. Additionally, 3-HM formation as well as 3-MM/3-HM metabolic ratio neither correlated with serum ALAT, ASAT, CHE or GLDH values nor with blood count, blood glucose levels or the inflammation parameter CRP. But there was a weak significant negative correlation with serum parameters indicating cholestasis, as gamma-GT, AP and total bilirubin, and a positive correlation with serum total protein content. Furthermore, history of alcohol consumption was associated with significantly lower 3-HM formation. 1. Dept. of Clinical Pharmacology, Friedrich Schiller University, Jena, Germany, 2. Institute of Pharmacology and Toxicology, Friedrich Schiller University, Jena, Germany, 3. Dept. of General, Visceral and Vascular Surgery, Friedrich Schiller University, Jena, Germany

476 Genetic variants of the dopamine D2 receptor modulate the risk of opiate addiction and methadone dose requirements Doehring A. (1), von Hentig N. (1), Graff J. (1), Salamat S. (1), Schmidt M. (2), Geisslinger G. (1), Harder S. (1), Lötsch J. (1) Aim: Addictive behavior is importantly mediated by mesolimbic dopaminergic signaling. We analyzed the DRD2 gene locus, and in addition the ANKK1 rs1800497C>T single nucleotide polymorphism (SNP), for associations with the risk of opiate addiction and the methadone dosage requirements. Methods: Allelic frequencies of DRD2/ANKK1 polymorphisms were compared between 85 methadone substituted Caucasian patients versus a random sample of 99 healthy Caucasian controls. Within patients, the average and maximum daily methadone dose during the first year of treatment and the time when that maximum dose was reached were analyzed for an association with DRD2/ANKK1 genetics. Results: Compared to the control group, drug users carried more frequently the minor allele of DRD2 SNP rs1076560G>T (p=0.022, odds ratio 2.343) or the ATCT haplotype of DRD2 rs1799978A>G, rs1076560G>T, rs6277C>T, ANKK1 rs1800497C>T (p=0.048, odds ratio 2.23), with similar tendencies for ANKK1 rs1800497C>T (p=0.056, odds ratio 2.12) and the TCCTCTT haplotype of DRD2 rs12364283T>C, rs1799732Cdel, rs4648317C>T, rs1076560G>T, rs6275C>T, rs6277C>T and ANKK1 rs1800497C>T (p= 0.059, odds ratio 2.31). The average and maximum daily methadone doses were significantly associated with the DRD2 rs6275C>T SNP (p=0.016 and p=0.005 for average and maximum dose, respectively). Carriers of the variant rs6275T allele needed higher methadone doses than noncarriers. In addition, this variant was associated with a longer time to reach the maximum methadone dose (p=0.025). Conclusions: Based on an analysis spanning the whole gene locus, from the DRD2 promoter to the ANKK1 rs1800497C>T polymorphism, DRD2 genetic polymorphisms modulate both the risk of opiate addiction, leading to the necessity of methadone substitution therapy, and the course of this therapy in terms of dosage requirements. 1. Institute of Clinical Pharmacology, JW Goethe University, Frankfurt / Main, Germany, 2. Malteser Drogenambulanz Schielestraße, Frankfurt / Main, Germany

477 Genetics in outpatient pain centers: Does a GTP cyclohydrolase 1 genetic polymorphism delay the need of specialized pain therapy? Doehring A. (1), Freynhagen R. (2), Griessinger N. (3), Zimmermann M. (4), Sittl R. (3), von Hentig N. (1), Geisslinger G. (1), Lötsch J. (1) Reduced-function genetic variants of the GTP cyclohydrolase gene (GCH1) have been associated with reduced pain in patients after lumbar surgery and in two independent cohorts of healthy volunteers. To analyze whether consequences of this genetic variant play a general role for pain therapy we assessed its association with the treatment of pain of various origins in a multicenter setting enrolling 424 patients treated for at least one month in outpatient University tertiary pain care centers. We found no indication that a formerly named “pain protective” GCH1 haplotype is rarer among pain patients than in a random sample of 503 healthy controls (allelic frequencies 14.2% and 15.4% respectively, p=0.45). However, the log duration of pain therapy decreased in a genedose dependent manner (p=0.004) without differences in pain origin or treatment between genotype groups. The actual 24-h pain score (p=0.18) and opioid dosing (p=0.096) were only non-significantly decreased in carriers of the variant. The results thus emphasize a prophylactic role of reduced GCH1 function possibly consisting in a slower development of chronic pain but no complete protection from it, leading to a delay of the need for specialized pain therapy. 1. pharmazentrum frankfurt/ZAFES, Institute of Clinical Pharmacology, JW GoetheUniversity, Frankfurt / Main, Germany, 2. Department of Anesthesiology, Universitätsklinikum Düsseldorf, Düsseldorf, Germany, 3. Department of Anesthesiology, Universitätsklinikum Erlangen, Erlangen, Germany, 4. Department of Anesthesia, Intensive Care and Pain Therapy/ZAFES, JWGoethe-University Hospital,Frankfurt / Main, Germany

95

478

481

Pharmacokinetics of ceftriaxon in surgical patients Ihlefeld D. (1), Clauss T. (1), Weiss M. (1), Pönicke K. (1), Hofmann G. (2), Gergs U. (1), Neumann J. (1) Patients undergoing hip or knee replacement therapy are routinely pre-treated with antibiotics before they enter the theatre. This treatment intends to reduce the incidence and severity of peri- or post-surgical infections. It is controversial which antibiotic is the treatment of choice for this purpose. We wanted to study the plasma and bone kinetics of ceftriaxon in orthopedic patients in order to see whether in the spongiosa and the corticalis of the bones of these patients sufficiently high concentrations of ceftriaxon are obtained in order to inhibit usual hospital infections. Therefore, patients (22, mean age: 66.6, male: 16, mean body mass index: 30.4) undergoing routine surgery were treated with ceftriaxon 2g intravenously as a bolus immediately prior to operation. Plasma samples were withdrawn before and at three time points after drug infusion. After replacement of the bones, extracts from spongiosa or corticalis were obtained. Samples were subjected to standard extractions procedures and ceftriaxon, which is hardly metabolized, was quantified using column chromatography. The kinetics of ceftriaxon distribution into bone was analyzed using a population approach (ADAPT5). We noted, that the corticalis contained significantly higher levels of ceftriaxon (15.7 mg/l) than the spongiosa (10.8 mg/l, n=22). The minimal inhibitory concentration (MIC) of ceftriaxon for susceptible staphylococci amounts to 0.2-2 mg/l. It is concluded that the concentration in spongiosa and corticalis should be adequate to protect the patients against the usual nosocomial infections and is substantially lower than the plasma concentration. 1. Institut für Pharmakologie und Toxikologie, Med. Fakultät, Martin-Luther-Universität Halle-Wittenberg, Halle (Saale), Germany, 2. Bergmannstrost, Akademisches Lehrkrankenhaus, Halle (Saale), Germany

The selective COX-2-inhibitor celecoxib modulates sphingolipid synthesis Schiffmann S. (1), Sandner J. (1), Schmidt R. (1), Birod K. (1), Wobst I. (1), Schmidt H. (1), Angioni C. (1), Geisslinger G. (1), Grösch S. (1) Sphingolipids such as ceramides play important roles in cell proliferation, apoptosis and cell cycle regulation. An increased ceramide level is linked to the cytotoxic effects of several chemotherapeutics. Various selective cyclooxygenase-2 (COX-2) inhibitors induce anti-proliferative effects in tumor cells. We addressed the possible interaction of various selective COX-2 inhibitors (coxibs) with the sphingolipid pathway as an explanation of their anti-proliferative effects. Sphingolipids were measured using liquid chromatography tandem mass spectrometry (LC-MS/MS). Treatment of various cancer cell lines with celecoxib significantly increased sphinganine (dhSph), C16:0-, C24:0-, C24:1-dihydroceramide (dhCer) and led to a depletion of C24:0-, C24:1-ceramide (Cer) in a time and concentration dependent manner while other coxibs had no effect. Using 13C,15N-L-serine we demonstrated that the augmented dihydroceramides after celecoxib treatment originate from de novo synthesis. Celecoxib inhibited the dihydroceramide desaturase (DEGS) in vivo with an IC50 of 78.9 ± 1.5 µM and increased total ceramide level about 2 fold, indicating an activation of sphingolipid biosynthesis. Interestingly, inhibition of the sphingolipid biosynthesis by specific inhibitors of L-serine palmitoyltransferase (SPT) diminished the anti-proliferative potency of celecoxib. In conclusion, induction of de novo synthesis of sphingolipids and inhibition of DEGS contributes to the anti-proliferative effects of celecoxib. This work was supported by the Deutsche Forschungsgemeinschaft (DFG) Forschergruppe FG 784 / TP5 and the LOEWE Lipid Signaling Forschungszentrum Frankfurt (LiFF). 1. pharmazentrum frankfurt, Institut für Klinische Pharmakologie, Klinikum der GoetheUniversität Frankfurt am Main

479

482

Active metabolites formed during hepatic first-pass: Simulations featuring their contribution to the overall effect in altered liver clearance Abdelaziz A.M. (1), Alaraby M.A. (1), Mahran L.G. (1), Nieß R. (1), Spahn-Langguth H. (1) Phase-I- and – occasionally – also phase-II metabolites may contribute to the overall effect of a drug. This is not always apparent or revealed, since total concentrations of metabolites may be low in relation to parent drug concentrations. In the case of carvedilol, using a comparison of HPLC and a b1-specific radioligand-receptor-binding assay (RRA), it was possible to demonstrate that regarding b-adrenoceptor blockade the effect appears directly linked to the serum compartment, which indicates instantaneous equilibrium between the blood compartment and the biophase, and that oxidative metabolites significantly contribute to the effect, particularly after oral administration. This drug serves as model compound in different approaches for evaluating the role of formation of active metabolites during first-pass when hepatic clearance varies. A kinetic model, which includes the immediate transformation of a fraction of the dose into active metabolites during first-pass through the liver, i.e., before it reaches the systemic circulation (AM-FP model), was found superior to standard models. For carvedilol, the hepatic extraction ratio is considerable, and the oral availability (calculated assuming complete absorption and no intestinal elimination) amounts to 25 to 35% for parent drug due to extensive first-pass metabolism. And in spite of their low serum levels, the contribution of the metabolites must not be neglected, especially for oral dosing. Elimination of parent drug was found to be rate-limiting, i.e., the kinetics of the metabolites is formation-rate limited, at least in healthy volunteers. The AM-FP model was most suitable, since parent drug and metabolites appear to enter the systemic circulation simultaneously. The new compartmental model is applicable for PK/PD simulation studies including situations, where hepatic clearance is affected. It was used to simulate the contribution of active metabolites formed during first-pass to the overall effect, also under conditions of impaired liver function or reduced metabolic clearance. 1. Faculty of Pharmacy and Biotechnology, German University in Cairo, New Cairo City, Egypt

Ceramide synthases and ceramide levels are upregulated in breast cancer tissue Sandner J. (1), Schiffmann S. (1), Birod K. (1), Wobst I. (1), Angioni C. (1), Ruckhäberle E. (2), Kaufmann M. (2), Ackermann H. (3), Lötsch J. (1), Schmidt H. (1), Geisslinger G. (1), Grösch S. (1) Several in vitro studies have correlated dysfunction of the sphingolipid signalling pathway with promotion of tumor cell growth as well as progression and resistance of tumors to chemotherapeutic agents. As ceramides (Cer) constitute the structural backbones of all sphingolipids we investigated the endogenous ceramide levels in 63 breast tumor biopsies and compared them with those of healthy reference tissues using liquid chromatography tandem mass spectrometry (LC-MS/MS). The total ceramide levels in malignant tumor tissue samples were statistically significantly elevated when compared to reference tissue samples. Upregulation of the total ceramide level averaged 12-fold and 4-fold higher than reference samples, for malignant and benign tumors, respectively. Specifically, the levels of C16:0-Cer, C24:1-Cer and C24:0-Cer were significantly raised in malignant tumors as compared to benign and reference tissue. The augmentation of the various ceramides could be assigned to an increase of the mRNA levels of ceramide synthases (CerS) LASS2 (LAG1 longevity assurance homolog), LASS4 and LASS6. Notably, elevated levels of C16:0-Cer were associated with a positive lymph node status, indicating a metastatic potential for this ceramide. Moreover, the levels of C18:0-Cer and C20:0-Cer were significantly higher in estrogen receptor (ER) positive tumor tissues as compared to ER negative tumor tissues. In conclusion, progression in breast cancer is associated with increased ceramide levels due to an upregulation of specific LASS genes. This work was supported by the Deutsche Forschungsgemeinschaft (DFG) Forschergruppe FG 784 / TP5 and the LOEWE Lipid Signaling Forschungszentrum Frankfurt (LiFF). 1. pharmazentrum frankfurt, Institut für Klinische Pharmakologie, Klinikum der GoetheUniversität Frankfurt am Main, 2. Institut für Gynäkologie, Klinikum der GoetheUniversität Frankfurt am Main, 3. Institut für Biomathematik, Klinikum der GoetheUniversität Frankfurt am Main

480 Co-administered ritonavir and atazanavir but not demographic parameters or the pharmaceutical formulation influence steady-state population pharmacokinetics of saquinavir in HIV-1 infected adults von Hentig N. (1), Staszewski S. (2), Lötsch J. (1) Saquinavir is currently used at a fixed dose of 1000 mg twice a day in combination with 100 mg of ritonavir. Since side effects or the CD4 cell count are associated with high saquina-vir concentrations, factors affecting the saquinavir pharmacokinetics are of clinical relevance. This study aimed to assess the effects of co-variates including the patient’s demographics, pregnancy (n=13) and co-administration of atazanavir (n=49) on 12-h steady-state pharmaco-kinetic profiles of saquinavir obtained in HIV-1 infected adults (n=136) receiving saquina-vir/ritonavir 1000/100 mg twice daily. A population analysis was performed with the NONMEM computer program adapting a onecompartment model with first-order absorption to 952 plasma samples. The covariates included into the final model indicated that the total body clearance of saquinavir of 61.8 l/h was (i) decreased to ⅔ by atazanavir co-administration and was (ii) negatively weakly (r2=0.05) but statistically significantly correlated with the median ritonavir plasma concentration during the 12 h observation period. In contrast, demographic parameters such as sex, weight, age, and alos not the pharmaceutical fo9rmulation of saquina-vir, had no significant on saquinavir pharmacokinetics, and a considerable effect of pregnancy consisting in doubled volume of distribution did not pass the co-variate selection at p=0.01 level during modeling. In conclusion, we show that saquinavir pharmacokinetics are largely unaffected by many standard candidate co-variates including pregnancy and the pharmaceuti-cal formulation, which provides a basis for keeping drug monitoring at an economical mini-mum. On the other hand, saquinavir pharmacokinetics is very sensitive to changes in CYP3A activity, which may become a contraindication or a signal to apply drug monitoring in cases where additional inhibitors are co-administered. 1. pharmazentrum frankfurt/ZAFES, Institute of Clinical Pharmacology, Johann Wolfgang Goethe-University, Frankfurt, Germany, 2. Medical HIV Research and Treatment Unit, Johann Wolfgang Goethe-University, Frankfurt, Germany

483 I-kappaB kinases as novel targets for the pharmacotherapy of pain and inflammation Möser C. (1), Kynast K. (1), Kupis S. (1), Tegeder I. (1), Geisslinger G. (1), Niederberger E. (1) The transcription factor nuclear factor κB (NF-κB) plays a crucial role in immune response, apoptosis and inflammation. Furthermore, several studies indicate that NF-κB is also involved in the development and processing of pathological pain. Accordingly, a pharmacological intervention into this pathway may have antinociceptive effects and could provide novel treatment strategies for acute and chronic pain. The NF-κB activation cascade offers different targets for pharmacological treatment and we are particularly interested on IκB kinases (IKK) which regulate NF-κB activity by phosphorylation of its inhibitory protein IκB. The classical IKK complex is composed of the subunits α, β and γ. Recently two novel IKK-related kinases, IKKε and TANK‑binding kinase 1 (TBK1) were described which might be also involved in NFκB regulation. These novel kinases are focus of our current studies. We found that IKKε as well as TBK1 are constitutively expressed in neuronal mouse tissues that are relevant for nociception. After peripheral nociceptive stimulation these kinases are positively regulated in the mouse spinal cord. In situ hybridization studies showed expression and regulation of IKKε RNA mainly in the pain relevant laminae I and II of the dorsal horn, as well as in dorsal root ganglia (DRG). To characterize the role of IKKε during nociception we analysed the nocicepetive behaviour of IKKε knock-out mice after formalin-induced inflammation. Our data showed that the knock-out of IKKε results in antinociception. These findings were supported by real time PCR analysis of pain relevant NF-κB target genes. So far, our results indicate that IKKε plays a role in nociception and may therefore prove as a novel target for analgesic therapy. The work is supported by the Deutsche Forschungsgemeinschaft (NI 705/2-1) and Graduate School “Biologicals” (GRK 1172). 1. pharmazentrum frankfurt, Institut für Klinische Pharmakologie, Klinikum der GoetheUniversität Frankfurt am Main

96

484 Thrombin upregulated expression of functionally active P2Y12 ADP receptors in human vascular smooth muscle cells Rauch B.H. (1), Driessen J. (1), Ermler S. (1), Böhm A. (1), Rosenkranz A.C. (1), Fischer J.W. (1), Jakubowski J.A. (2), Sugidachi A. (3), Schrör K. (1) The platelet P2Y12 ADP receptor is a well known target of thienopyridine antiplatelet prodrugs. Expression of P2Y12 has recently also been shown in vascular smooth muscle cells (SMC), although no functionality has been established. Here, we describe a transcriptional upregulation of functionally active P2Y12 by thrombin in human SMC. Experiments were carried out in cultured SMC prepared from human saphenous vein. mRNA and protein expression of P2Y12 were determined by RT-PCR/real-time PCR, Western blotting, flow cytometry, and immunohistochemistry; IL-6 expression by realtime PCR. Cell proliferation was assessed by [3H]-thymidine incorporation. Basal expression of P2Y12 mRNA was low, but was markedly enhanced by thrombin (30 nmol/L). Total protein and cell surface expression of P2Y12 were significantly increased within 6 h after thrombin addition and this expression was sustained over 24 h (n=5-8). The P2Y receptor agonist 2-methyl-thio-ADP (MeS-ADP; 1 µmol/L) did not change basal or thrombin-induced mitogenesis if given simultaneously with thrombin. By contrast addition of MeS-ADP after 6 h preincubation with thrombin, resulted in a markedly enhanced mitogenesis which was completely prevented by 10 µmol/L of the selective P2Y12-antagonist R-138727 (n=5), the active metabolite of prasugrel, suggesting that this effect was P2Y12-mediated. In addition, MeS-ADP-induced expression of IL-6 mRNA was enhanced in human SMC after pretreatment with thrombin and this was P2Y12mediated. These data show that thrombin induces transcriptional upregulation of the P2Y12 receptor in human SMC which is followed by an enhanced P2Y12-mediated mitogenic and proinflammatory response. In additional experiments, positive P2Y12 immunostaining was shown in human carotid artery plaques and was found to colocalize with tissue factor, the rate-controlling step for initiation of thrombin formation in vivo. This suggests that P2Y12 receptors are not only important for ADP mediated platelet function but might also be involved in control of vascular events under conditions of enhanced endogenous thrombin formation, such as acute coronary syndromes. 1. Institut für Pharmakologie und Klinische Pharmakologie, Universitätsklinikum Düsseldorf, Germany, 2. Eli Lilly and Company, Indianapolis, USA, 3. Daiichi Sankyo Co., Ltd., Tokyo, Japan

485 Acetylsalicylic acid inhibits the release of sphingosine-1-phosphate from human platelets Ulrych T. (1), Rosenkranz A.C. (1), Hohlfeld T. (1), Schrör K. (1), Rauch B.H. (1) Thrombin or activating-peptides for the protease-activated receptor-1 (PAR1-AP) induce the release of the sphingosine-derived lipid signaling molecule sphingosine-1-phosphate (S1P) from platelets. The present study investigates the effect of acetylsalicylic acid (ASA) on S1P release from human platelets and its functional consequences. S1P synthesis and release were analyzed in platelets from healthy donors spiked with [3H]sphingosine. Lipid were separated by thin-layer chromatography. Migration of human endothelial umbilical vein endothelial cells (HUVEC) and monocytic U937 cells was determined in Boyden chamber assays. Release of S1P from platelets activated by PAR1-AP (100 µM) was suppressed by ASA (30 – 300 µM) and also by reversible COX inhibitors such as ibuprofen and diclofenac. The thromboxane receptor (TXR) agonist U46619 (10 µM) increased S1P release while the TXR antagonist ramatroban (2 µM) inhibited it. Platelet aggregation or CD62P externalization were not inhibited by ASA after PAR-1 activation. Oral ASA (500 mg single dose) also reduced S1P release from platelets ex vivo in healthy volunteers. This effect occurred within 15 minutes after oral ASA application and persisted for 3 days until platelet TX formation started to recover. Supernatants from PAR1-AP-activated platelets pretreated with ASA elicited significantly lower HUVEC and monocyte chemotaxis as compared to supernatants from untreated controls. The use of specific S1P receptor inhibitors suggested this effect was S1Pdependent. These data suggest that release of S1P from human platelets strictly depends on TXR activation after stimulation of PAR-1 and is independent of platelet degranulation. This novel pathway may contribute to anti-inflammatory and antiangiogenic actions of ASA by affecting recruitment of endothelial cells and monocytes to sites of injury. 1. Institut für Pharmakologie und Klinische Pharmakologie, Universitätsklinikum Düsseldorf, Germany

phospholipid membranes of colorectal cancer cells provides a novel molecular basis for the ability of celecoxib to interact with non-COX2 targets in-vivo despite low plasma concentrations. 1. pharmazentrum frankfurt/ZAFES, Institute of Clinical Pharmacology, GoetheUniversity, Frankfurt/Main, Germany, 2. Institute of Pharmaceutical Chemistry/ZAFES, Goethe-University, Frankfurt/Main, Germany, 3. Centre for Biomolecular Magnetic Resonance and Institute for Biophysical Chemistry, Goethe-University, Frankfurt/Main, Germany

487 Expression, protein stability, and ligand diversity of human and chimpanzee CYP3A4 and their putative common ancestor Wojnowski L. (1), Qiu H. (1), Oezguen N. (2), Herlyn H. (3), Halpert J.R. (4), Kumar S. (4) For currently unknown reasons, the evolution of CYP3A4 underwent acceleration in the human lineage following the split from chimpanzee. We investigated the significance of this event by comparing E. coli-expressed CYP3A4 from humans, chimpanzee, and their most recent common ancestor. The expression level of chimpanzee CYP3A4 was ~50% of the human CYP3A4, whereas ancestral CYP3A4 did not express in E. coli. Steadystate kinetic analysis with 7-benzyloxyquinoline, 7-benzyloxy-4-(trifluoromethyl)coumarin (7-BFC), and testosterone showed no significant differences between human and chimpanzee CYP3A4. Upon addition of a-naphthoflavone (25 µM) human CYP3A4 showed a slightly decreased S50, whereas chimpanzee CYP3A4 showed a > 2-fold increase. No significant differences in inhibition/activation were found for a panel of 43 drugs and endogenous compounds. A striking exception was the hepatotoxic secondary bile acid and CYP3A4 substrate lithocholic acid, which at saturation caused a 5-fold increase in 7-BFC debenzylation by human CYP3A4, but not by chimpanzee CYP3A4. Our results suggest that the wide substrate spectrum of human CYP3A4 most likely precedes the human-chimpanzee split. Positive selection on the human CYP3A4 may have been triggered by an increased load of dietary steroids, which led to a novel defense mechanism against cholestasis. Chimpanzee CYP3A4 appears to be an excellent model of its human counterpart. 1. Dept. of Pharmacology, Johannes Gutenberg University, Mainz, Germany, 2. Department of Biochemistry and Molecular Biology and The Sealy Center for Structural Biology and Molecular Biophysics, University of Texas Medical Branch, 3. Dept. of Anthropology, Johannes Gutenberg University, Mainz, Germany, 4. Skaggs School of Pharmacy & Pharmaceutical Sciences, University of California San Diego

488 Phylogenetic analysis of CYP3A regulatory elements Mathäs M. (1), Qiu H. (1), Nem D. (1), Gödtel-Armbrust U. (1), Wojnowski L. (1) The pronounced inter-individual differences in the expression of CYP3A4 complicate therapies with 50% of contemporary drugs and affect the clearance of other xenobiotics, including carcinogens. Gene variants explain only a fraction of CYP3A4 expression variability. We investigated the evolution of CYP3A4 upstream regulatory elements using comparative genomics and functional analysis. The combination of a proximal ER6 element with XREM and CLEM, such as seen in the present-day human CYP3A4, represents the original scheme of Cyp3a regulation which emerged with placental mammals (Eutheria). The pronounced differences between upstream sequences of the human CYP3A4 and of the other three CYP3A paralogs in the primate lineage were shaped by two classes of evolutionary events: 1) Loss of CLEM from all non-CYP3A4 genes, of XREM from CYP3A7 and CYP3A5, and of all three elements from CYP3A43. 2) The increase in the number of CYP3A4 response elements to nuclear receptors, in particular of the ER6 class. Particularly striking is the origin of 5 new ER6 elements in the lineage leading to humans and chimpanzee. These changes have resulted in CYP3A4 having the highest number of regulatory elements potentially responsive to the xenobiotics sensors PXR and CAR. The complexity of the CYP3A4 promoter and the resulting sensitivity to xenobiotics may explain a significant fraction of its expression variability observed in contemporary humans. 1. Dept. of Pharmacology, Johannes Gutenberg University, Mainz, Germany

489 486 Celecoxib accumulates in phospholipid membranes of human colorectal cancer cells Maier T.J. (1,2), Schiffmann S. (1), Birod K. (1), Angioni C. (1), Hoffmann M. (2), Lopez J.J. (3), Glaubitz C. (3), Steinhilber D. (2), Geisslinger G. (1), Grösch S. (1) Celecoxib is a selective cyclooxygenase-2 (COX-2) inhibitor used to treat inflammation and pain. Additionally, celecoxib displays strong antiproliferative effects in cultured tumor cells superior to those of other coxibs and prevents colorectal cancer in patients if administered at high doses, an effect that was attributed to several non-COX-2 targets. However, celecoxib concentrations in human plasma (up to 8 µM) eclipse those to induce antiproliferation in cell culture (up to 100 µM). We therefore speculated that celecoxib might accumulate in human tumor cells, which allows the drug to interact with non-COX-2 enzymes. LC-MS/MS analysis demonstrated 5-10 fold higher intra-cellular concentration for celecoxib and DMC (a non-COX-2-inhibitory celecoxib analogue with similar anti-proliferative potency) compared to the other coxibs (etoricoxib, valdecoxib, lumiracoxib, rofecoxib) in human HCT-116 and HCA-7 colon carcinoma cells. Neither ATP-depletion nor temperature decrease to 4 °C attenuated the accumulation, suggesting a passive diffusion of celecoxib. Subcellular fractionation demonstrated that 80 % of the total intra-cellular celecoxib was located within cellular membrane fractions. In support of this, MAS assisted NOESY-NMR experiments indicate exclusive interactions of celecoxib and DMC with phospholipid acyl chains located inside the leaflets of multilamellar model membranes. As an apparent consequence of intramembranous accumulation, celecoxib disturbed the plasma membrane integrity of HCT116 cells and displayed an increased COX-2-inhibitory potency in HCA-7 cells, as determined by ELISA measurement of PGE2 synthesis. In conclusion, accumulation in

Characterization of a HEK293 cell line stably expressing OATP2A1 and influence of acetylsalicylic acid on the transport activity of OATP2A1 Mandery K. (1), Fromm M. F. (1), Glaeser H. (1) OATP2A1, one of the organic anion transporting polypeptides (OATPs), is involved in the transport of several prostaglandins such as PGE2. It is known that non-selective cyclooxygenase inhibitors such as acetylsalicylic acid (ASA) can cause gastrointestinal or renal side effects due to a decrease of PGE2 formation. There is increasing evidence that OATP2A1 may be involved in the cellular and tissue distribution of PGE2. Therefore we investigated the influence of ASA on the transport activity of OATP2A1. HEK293 cells were transfected in order to generate a cell line stably expressing OATP2A1. The expression was analyzed using quantitative Real-Time-PCR, immunoblot and immunofluorescence analysis. The selected clone was characterized by a significantly increased OATP2A1 mRNA and protein expression compared to the cells transfected with the empty expression vector. The immunofluorescence revealed that OATP2A1 is localized in the plasma membrane of the HEK293 cells. Using the prototypical substrate PGE2, a Km value of 0.5 µM was determined. Using a concentration of 0.5 µM PGE2 the influence of ASA on the transport activity was investigated at a concentration range from 0.1 µM to 1000 µM. ASA led to a concentration-dependent decrease of the OATP2A1 mediated PGE2 transport up to 30 %. On the basis of these data it could be speculated that the modification of the OATP2A1 transport activity may possibly contribute to the already known side effects of non-selective COX-inhibitors. 1. Institute of Experimental and Clinical Pharmacology and Toxicology, FriedrichAlexander-University of Erlangen-Nuremberg, Erlangen, Germany

97

490

493

Regulation of human concentrative nucleoside transporter-1 gene expression Klein K. (1), Eloranta J.J. (1), Wagner M. (2), Trauner M. (2), Kullak-Ublick G.A. (1) Background: The concentrative nucleoside transporter-1 (CNT1) is a member of the solute carrier 28 (SLC28) gene family, and is expressed in the liver, intestine, and kidneys. In addition to naturally occurring pyrimidine nucleosides, CNT1 mediates uptake of nucleoside analogues used in anticancer and antiviral therapy. Expression levels of CNT1 may thus affect the pharmacokinetics of these drugs and the outcome of drug therapy. Aim: To characterize the mechanisms, by which the expression levels of the CNT1/Cnt1 gene are regulated in human cell lines derived from liver, intestine, and kidneys, as well as in primary human hepatocytes and in mouse models. Results: The transcriptional start site of the CNT1 gene was determined by 5`-RACE. In silico analysis revealed the existence of three putative binding sites for the nuclear receptor hepatocyte nuclear factor-4α (HNF-4α) within the CNT1 promoter. A luciferase reporter gene construct containing the CNT1 promoter region was transactivated by HNF-4α in human cell lines derived from the liver, intestine, and kidneys. Consistent with this, we showed in electromobility shift assays that HNF-4α specifically binds to two conserved direct repeat-1 (DR-1) motifs within the proximal CNT1 promoter. In cotransfection experiments, the transcriptional coactivator peroxisome proliferator-activated receptor-γ coactivator-1α (PGC-1α) further increased, whereas the bile acid-inducible corepressor small heterodimer partner (SHP) reduced, HNF-4α-dependent CNT1 promoter activity. Consistent with the latter phenomenon, CNT1 mRNA expression levels were suppressed in primary human hepatocytes upon bile acid treatment. Supporting the physiological relevance and species-conservation of this effect, ileal Cnt1 mRNA expression was decreased upon bile acid feeding and increased upon bile duct ligation in mice. Conclusions: The promoter region of the human CNT1 gene is regulated by the liver-enriched nuclear receptor HNF-4α, the transcriptional coactivator PGC-1α, and bile acids. Detailed knowledge of the regulation of CNT1 gene expression may help us to understand the mechanisms that influence the pharmacokinetics of drugs transported by CNT1, and potentially enable us to modulate CNT1-mediated drug transport in a defined manner. 1. Division of Clinical Pharmacology and Toxicology, University Hospital Zurich, Switzerland, 2. Division of Gastroenterology and Hepatology, Medical University Graz, Austria

Effect of enhanced monomethylarginine metabolism by human dimethylarginine dimethylaminohydrolase 1 on plaque formation in ApoE-deficient mice Atzler D. (1), Jacobi J. (2), Strobel J. (3), Cordasic N. (2), Arend M. (2), Schwedhelm E. (1), Böger R.H. (1), Hilgers K.F. (2), Maas R. (3) Background: L-monomethylarginine (L-NMMA) and asymmetric dimethylarginine (ADMA) are endogenously formed competitive inhibitors of nitric-oxide synthase (NOS). Elevation of L-NMMA and ADMA is associated with cardiovascular disease. It was the aim of the present study to investigate whether overexpression of the ADMA- and LNMMA-degrading enzyme dimethylarginine dimethylaminohydrolase 1 (DDAH1) confers a vasculoprotective effect. Method: We compared plaque formation in the aorta (en face preparations) of male and female ApoE-deficient mice (n=130) and ApoE competent mice (n=55) with or without transgenic overexpressing of the human (h)DDAH1 kept on a standard rodent chow (sacrifice at 12 months, n=112) or on an atherogenic diet (sacrifice at 7 months, n=73). Results: As compared to wildtype mice DDAH1 overexpression resulted in a significant reduction of plasma ADMA (0.46±0.015 vs. 0.76±0.02 µmol/L, p<0.001) and L-NMMA (0.20±0.00 vs. 0.29±0.01 µmol/L, p<0.001) while symmetric dimethylarginine (SDMA) was not different (0.18±0.01 vs. 0.18±0.01 µmol/L, p=0.527). In ApoE deficient mice on normal or on atherogenic chow overexpression of DDAH1 was associated with reduced plaque formation (standard chow, male: 12.4 vs. 7.6%, p=0.002; female: 10.5 vs. 8.8%, p=0.10; atherogenic diet, male: 33.0 vs. 26.1%, p=0.009; female: 24.6 vs. 19.0%, p=0.02). The inhibitory effect of DDAH1 overexpression on plaque formation was more pronounced in male than in female mice, which overall had higher ADMA levels (male vs. female: 0.57±0.02 vs. 0.69±0.03 µmol/L, p=0.0003). Only in male but not female mice the degree of aortic atherosclerosis significantly correlated with plasma ADMA levels (standard chow: r= 0.52, p = 0.004; atherogenic diet: r=0.46, p=0.015). Conclusion: DDAH1 overexpression is associated with reduced plasma ADMA and L-NMMA levels as well as attenuated formation of plaques induced by diet and/or ApoE-deficiency. The underlying mechanism is most likely increased metabolism of the endogenous NOS inhibitors by the hDDAH1 transgene. 1. Institute of Experimental and Clinical Pharmacology and Toxicology, University Medical Center Hamburg-Eppendorf, Germany, 2. Dept. of Nephrology and Hypertension, Friedrich-Alexander University of Erlangen-Nuremberg, Germany, 3. Institute of Experimental and Clinical Pharmacology and Toxicology, Friedrich-Alexander University of Erlangen-Nuremberg, Germany

491 OATP1B1- and OATP1B3-mediated Pravastatin Uptake: Determination of Uptake Kinetics and Influence of Coadministered Drugs Kindla J. (1), Seithel A. (1), Bachmakov I. (1), Fromm M.F. (1), König J. (1) Transporter-mediated uptake of drugs into cells [mediated e.g., by members of the human organic anion transporting polypeptide (OATP) family] is a key determinant of drug disposition and a prerequisite for subsequent metabolism. The liver-specific human OATP family members OATP1B1 (gene symbol SLCO1B1) and OATP1B3 (gene symbol SLCO1B3) mediate the uptake of endogenous substances and of several drugs (e.g., HMG-CoA-reductase inhibitors) into hepatocytes. We determined the kinetics of OATP1B1- and OATP1B3-mediated pravastatin uptake. Furthermore, the potential influence of coadministered macrolides or oral antidiabetic drugs on pravastatin uptake was investigated. Using HEK293 cells stably expressing either OATP1B3 or OATP1B3, our experiments revealed typical saturation kinetics for pravastatin uptake with Km values of 36.5 µM for OAPT1B1 and 99.8 µM for OATP1B3. The Vmax values for OATP1B1- and OATP1B3-mediated pravastatin uptake were 5.3 pmol/mg/min and 9.7 pmol/mg/min, respectively. Alteration of uptake transporter function by concomitantly administered drugs is a potential mechanism of drug-drug interactions. Therefore, the effects of the macrolides azithromycin, clarithromycin, erythromycin, roxithromycin, and telithromycin and of the antidiabetic drugs repaglinide, rosiglitazone, and metformin on the pravastatin uptake were analysed. These studies demonstrated that the OATP1B1and OATP1B3-mediated uptake of pravastatin can be inhibited by increasing concentrations of all macrolides except azithromycin. Metformin did not inhibit the pravastatin uptake, but repaglinide and rosiglitazone showed significant uptake inhibition. These results demonstrated that inhibition of uptake transporter function is a potential mechanism underlying drug-drug interactions with pravastatin. 1. Institute of Experimental and Clinical Pharmacology and Toxicology, FriedrichAlexander-University of Erlangen-Nuremberg, Erlangen, Germany

492 Preliminary results: breast cancer cell lines produce sIL-6R and sgp130 in vitro Knüpfer H. (1), Preiss R. (1) IL-6 is a pleiotropic cytokine with tumor promoting and also inhibitory properties. Recently we found out that high IL-6 levels (and therefore in the broadest sense IL-6 signaling) has a significant impact on breast tumor patients in regard to tumor stages, clinical outcome and survival. Previously we were able to show that breast tumor samples produce IL-6 endogenously. To further clarify the significance of IL-6 signaling in breast tumor cells, three adherently growing breast tumor cell lines (MCF-7, HCC1143 and BT-474) were analysed in regard to the agonistic soluble IL-6 receptor (sIL6R) and the antagonistic soluble gp130 (sgp130) by PCR and ELOSA. After RNA isolation all three cell lines revealed positive PCR results concerning sIL-6R and sgp130. To confirm these results, 2000 cells per well were seeded into 96 well plates and after 24, 48, 120 and 144 h of incubation cell supernatants were harvested and subjected to ELOSA tests. All three cell lines produced measurable sIL-6R protein amounts, with HCC-1143 being the cell line with the highest amount, followed by BT474. sIL-6R protein of MCF-7 cells was measurable to a lesser extend and to a later time point. Concerning sgp130, the earliest sgp130 production could be detected after 120 h of incubation (MCF-7). MCF-7 cells produced the highest amount, followed by HCC-1143 (144h). sgp130 protein of BT-474 cells could only be detected if more cells were seeded (4000 or 8000 cells per well). Taken together, our in vitro experiments show that breast tumor cells produce the agonistic soluble IL-6 receptor on the one hand, and on the other hand also produce the counterpart soluble gp130, but later on and to a lesser extend. 1. Institute of Clinical Pharmacology, University of Leipzig, Germany

494 Withdrawn

495 Validation of physiologically-based simulations for cidofovir pharmacokinetics in critically ill paediatric patients Breddemann A. (1), Laws J. (2), Bührmann S. (3), Dohna-Schwake C. (4), Tot E. (1), Läer S. (1) Background Pharmacokinetic parameters are important to guide antiviral drug response and therefore help to avoid viral resistance. Cidofovir is an antiviral third line drug for critically ill paediatric patients, but no age-appropriate dosing regimen is available simply because pharmacokinetic parameters are lacking. Physiologically-based simulations may fill the pharmacokinetic information gap for children. Methods In a first step, a multicentre clinical trial in paediatric patients was conducted to investigate the agedependent pharmacokinetics. Six serum concentrations within eight hours were measured following intravenous application of 0.92- 5.16 mg/kg cidofovir. Pharmacokinetic parameters were analysed using a non compartmental approach. Then, individual pharmacokinetic profiles of study patients were simulated using a developed physiologically-based model for cidofovir with the PK-Sim® software 4.0 and compared with the non-compartmental analysis. Results Ten pharmacokinetic profiles from patients aged 0.63 to 16.01 years were analysed and compared with the simulation results. The median deviation of measured vs. simulated AUC (µg*h/ml) and bodyweight-normalised clearance (ml/h/kg) was 14% and 15% respectively. In addition, a 1.5 fold higher median bodyweight-normalised clearance in the paediatric population (207 vs. 138 ml/h/kg) was noted compared to published mean adult clearances. Conclusion Paediatric patients may need an age adapted dosing schedule compared to adults to achieve adequate drug response of cidofovir. It can be assumed that taking growth and maturation processes into account, physiologically-based simulations for cidofovir are capable to predict cidofovir pharmacokinetics a priori. That technique can then provide an a priori dose prediction for individual critically ill patients and thus may improve treatment outcome substantially. 1. Dept. of Clinical Pharmacy and Pharmacotherapy, Heinrich-Heine-University, Düsseldorf, Germany, 2. Dept. of Paediatric Haematology and Oncology, HeinrichHeine-University, Düsseldorf, Germany, 3. Hospital Pharmacy, University of EssenDuisburg, Essen, Germany, 4. Dept. of Paediatric Haematology and Oncology, University of Essen-Duisburg, Essen, Germany

496 Identifying bisoprolol as the beta-blocker of choice for a clinical trial in children with congestive heart failure using physiology-based pharmacokinetic simulation Frobel A.-K. (1), Läer S. (1) Background: Beta-adrenoceptor-blockers (BB) have been proven to reduce morbidity and mortality in adult congestive heart failure (CHF). However, this paradigm has been challenged by a trial that showed no difference in efficacy between carvedilol and placebo in paediatric CHF (Shaddy 2007). According to the FDA, carvedilol is not recommended for paediatric patients, leaving the use of BB in children with CHF unclear. Aim: The aim of our work was to analyse the evidence for the use of BB in children with CHF and, in case evidence was insufficient, in a second step to use pharmacokinetic simulation to give a rationale for choosing the most suitable BB for future trials. Methods: We performed a systematic Cochrane Review searching the

98

literature and selecting all randomised controlled clinical trials (RCCTs) investigating the effect of BB on paediatric CHF. For the simulation PK-Sim 4.0 software was used to create a pharmacokinetic (PK) model incorporating physiologicical changes due to maturation and growth. PK data was extracted from literature for bisoprolol, carvedilol and metoprolol. Paediatric PK of carvedilol have been described (Läer 2002, Albers 2007) and show large inter-individual variability. In contrast, paediatric bisoprolol and metoprolol PK data are lacking and models were created for both. Results Our search identified 677 publications; 3 studies were included into the review (Frobel 2009). In two small studies (20/22 children), CHF improved while a study with 161 patients showed no significant effect. However, study populations showed vast differences with regard to age, aetiology and severity of CHF. In none of the studies there was a rationale for dosing. There is not enough evidence to recommend or discourage the use of BB in children with CHF. PK models of bisoprolol and metoprolol were created. First, models were validated for adults, then adapted to children considering developmental changes in physiology, metabolism and clearance. For metoprolol, due to the CYP2D6 polymorphism inter-individual variability of clearance was similarly large as for carvedilol while bisoprolol plasma curves varied in a more narrow range. Simulations of bisoprolol predicted a dynamic change of clearance over age with a maximum of about 200% of the adult clearance in the 1- to 4-year-old. Summary: A further RCCT with child adapted dosing investigating BB in children with CHF is required. From the viewpoint of pharmacokinetics, the drug of choice should be bisoprolol. 1. Clinical Pharmacy and Pharmacotherapy, Heinrich-Heine-University, Düsseldorf, Germany

497 A clinical decision support system for pediatrics: progress of development Gorny M. (1), Läer S. (1) Background: Under- and overdosing are main reasons for ineffective and toxic drug treatment in pediatric patients. Clinical decision support systems (CDSS) in adults have been shown to substantially avoid dosing errors and improve the safety in pharmacotherapy. There is, however, no electronic health record system available on the German market, containing a CDSS for pediatric patients. Therefore, with a stepwise approach we started to develop an electronic database for a pediatric CDSS. The gaps left by the summaries of product characteristics (SPCs) have to be filled with expert knowledge such as textbooks or reference books, until systematic investigations are available. Methods: First we included 203 drugs which had been identified by a pediatric prescription analysis from the Children´s Hospital of the University of Düsseldorf (Hsien L., Pharm WorldSci 2008). For these substances, SPCs from various companies were analyzed concerning indication, age- and weight groups, dosage, application route, duration of treatment and additional application information. To convert the age-classes terms used in the SPCs, a classification based on the recommendations of the EMEA, FDA and WHO was built. Second, we accomplished a market research of literature on the German market concerning the treatment of diseases in childhood. The main criteria were actuality defined as not published prior to 2003, quotation of references for the given dosing information. In addition, all specialties in pediatrics should be included in one book. Results: Profound gaps concerning safe and effective drug treatment for children became evident: Available dosing labels were low and varied substantially. Whereas at least 47% of the anti-infectives were labeled for all age ranges, only 18% of cardiovascular drugs had sufficient dosing information from newborns up to adolescents. The market research of literature resulted in 25 books concerning the treatment of diseases in childhood, 14 textbooks and 11 reference books. Only two of them, one of each book type, fulfilled the established criterion of quotation of reference. Conclusion: A clinical decision support system will be developed based on the dosing information of the SPCs and the two referenced textbooks available on the German market. 1. Dept. of Clinical Pharmacy and Pharmacotherapy, Heinrich-Heine-University, Düsseldorf, Germany

498 Simulation of sildenafil disposition in virtual children by means of a physiologically based pharmacokinetics modelling Hsien L. (1), Vogt W. (1), Läer S. (1) Background: Due to the lack of data about the pharmacokinetic characteristics (PK) of oral sildenafil in pediatrics, there is no age-specific dosage recommendation for the treatment of pulmonary hypertension (PH) in children. This can potentially result in under- or overdosing as well as to non-compliance due to an administration sequence up to six times a day. Therefore, we aimed to estimate the PK characteristics of sildenafil in virtual children as the basis for the development of rational age-adjusted dose strategies for PH in children. Methods: We developed a physiologically based pharmacokinetic (PBPK) model using PK-Sim software. Parameters describing the metabolism capacity of the intestinal cytochromes CYP3A4 and CYP2C9 involved in the biotransformation of sildenafil were estimated regarding their abundance in duodenum, jejunum and ileum and the effective surface area of these segments. We started with a PBPK model for therapeutic doses of sildenafil in populations with demographic characteristics similar to the healthy volunteers in two published data sets. The models were compared with plasma concentrations after intravenous application in a population aged 45 to 58 years (Walker 1999) and oral application in adults with a mean age of 29 years (Jetter 2002). Secondly, we used the PBPK model to scale sildenafil for children with mean ages of 0.1, 1, 6 and 12 years. Results: A PBPK model of therapeutic doses of sildenafil was developed. The simulated mean values versus the experimental data for PK-parameters after a single 25mg intravenous dose were: AUC: 920 vs. 972 ng.h/ml, clearance: 6.5 vs. 6.0 ml/min/kg and terminal half-life (t1/2) 2.8 vs. 2.4 hours (Walker 1999). The model showed the following PK-parameters for the population investigated by Jetter 2002 taking a single oral dose of 50mg sildenafil: AUC: 660 vs. 619 ng.h/ml, Tmax: 0.88 vs. 0.75 hours, Cmax: 227.06 vs. 254.89 ng/ml and t1/2: 3.14 vs. 3.79 hours. In the second step, the PBPK model was scaled down to children giving clearance values of 10.2, 13.0, 11.0 and 8.4 ml/min/kg for 0.1, 1, 6 and 12 year-old children, respectively. Conclusion: The PBPK model indicates that children at every age group have different clearance values compared with one another and with the adult population. It can be assumed that age-specific dosing recommendation help to treat children better and safer by avoiding over- or underdosing. However, this model has to be validated in a paediatric population.

1. Department of Clinical Pharmacy and Pharmacotherapy

499 Sex hormones as drugs – their possible role in human fetal malformations - The Duogynon® experience Heinrich-Gräfe U. (1), Schröder J. (1), Siegert J. (1), Kirch W. (1) Backround: A patient who called our drug information service had made an enquiry on adverse reactions of Duogynon® (Schering, Berlin, today BSH).The drug was marketed in Germany from the early 50’s until October 1980. This drug was used as a therapy of dysfunctional menstruation and as a hormonal pregnancy test. Duogynon® contained a combination of norethisteroneacetate and ethinylestradiol. In the UK it was marketed as Primodos®. Methods: A literature search was carried out in Pubmed, Scopus and Cochrane. Most of the literature was published in the 70’s and 80’s, due to suspicion that Duogynon® might be connected to fetal malformation. In 1967 Gal et al. first published a paper suggesting a possible relationship between hormonal pregnancy tests with Duogynon® and congenital neural tube defects. Due to this publication several studies were set up worldwide to verify possible teratogenic influences of sex hormones taken during pregnancy. Results: No significant relationship could be proven, although malformations like deformed arms and legs, cleft palates and hydrocephalus were more frequent in regions with higher sales of the drug. The results of the studies were contradictory. In West-Germany a cohort study was performed from 1964 to 1976; no significant increase of malformations due to the intake of Duogynon® could be proven. In this study “the observed odds ratios > than 1 and their upper confidence limits were rather high, which could be regarded as being in accordance with positive findings of other studies. The lack of significance could then be interpreted as due to the small number of observations” (Michaelis J. et al: Prospective study of suspected associations between certain drugs administered during early pregnancy and congenital malformations. Teratology 27:57-64, 1983). Conclusions: The use of hormonal pregnancy tests was not significantly associated with an increase of major malformations. Due to remaining uncertainties the marketing of the drug was stopped in many countries like Sweden, Finland, Belgium, Australia, Netherlands and Great Britain in the beginning of the 70’s. In Germany the drug was marketed with a changed indication (deletion of the use as pregnancy test) and a new brand name (Cumorit®) since 1978. Today the ingredients of Duogynon forte® (estradiol benzoate and progesterone) are used in raising calves and may end up in nutrition. 1. Institut für Klinische Pharmakologie, Medizinische Fakultät der TU Dresden, Germany

500 Overexpression of MRP3/ABCC3 mRNA in different types of bladder cancer in Egyptian patients Rady M. (1), Spahn-Langguth H. (1), Rohde J. (1), Mahran L. (1), Amr D. (1), Ibrahim S.S. (1), Zaghloul A.S. (2), Knüchel-Clarke R. (3), Khaled H. (2), Nieß R. (1), ScherlMostageer M. (1) Cancer chemotherapy remains a principal approach for bladder cancers. However, multidrug resistance (MDR) is a critical factor that results in suboptimal outcomes in cancer therapy. One principal mechanism of MDR is the increased expression of multidrug resistance proteins (MRPs), which are ATP-binding cassette (ABC) transporters that actively extrude anticancer drugs out of cells. Among such transporters, MRP3/ABCC3 was shown to contribute to drug resistance when overexpressed in cancer cell lines, while in clinical specimens, expression of MRP3/ABCC3 and its role in clinical MDR are still under investigation. This study aims to measure the mRNA expression of MRP3/ABCC3 in the different types of bladder cancer; transitional cell carcinoma (TCC; European type of bladder cancer) and squamous cell carcinoma (SCC; Schistosoma-induced bladder cancer). Thirty-six biopsies from primary untreated bladder tumors (among them 29 TCC and 7 SCC) were investigated by quantitative real-time reverse transcription-PCR based on SYBR Green chemistry in comparison to normal bladder tissue. Furthermore, the MRP3/ABCC3 mRNA relative expression in TCC versus SCC was compared and its association with tumor stage and grade was investigated. Expression of MRP3/ABCC3 mRNA was increased in bladder cancer at least 16-folds compared to normal bladder tissue (P < 0.0001). Relative expression of MRP3/ABCC3 was significantly higher in TCC than in SCC (P = 0.0002) and in non-muscle invasive than in muscle invasive bladder cancer (P = 0.004). There was no association, however, between MRP3/ABCC3 expression and tumor grade (P = 0.7474). 1. Faculty of Pharmacy and Biotechnology, German University in Cairo, New Cairo City, Egypt, 2. National Cancer Institute, Cairo, Egypt, 3. Institute of Pathology, University Clinics RWTH Aachen, Germany

501 Off label use of the antiepileptic substances phenobarbital and clonazepam Siegert J. (1), Brecht B. (1), Schindler C. (1), Heinrich-Gräfe U. (1), Schröder J. (1), Maywald U. (2), Kirch W. (1) Background: Phenobarbital and clonazepam are two substances listed in section III of the German narcotics control law with the exemption, that these substances are allowed to be handled with some restrictions by regular prescription. Both substances have a marketing authorisation limited to the use as antiepileptics in limited circumstances. Data of a previous study indicated that there might be a significant off label use for both substances. Methods: The use of phenobarbital and clonazepam in Saxony was derived from a survey covering all prescriptions of the Allgemeine Ortskrankenkasse (AOK, General regional health insurance) Sachsen. All patients with prescriptions of drugs with ATC code N03AE01 and N03AA02 (clonazepam and phenobarbital) in 2003 and 2004 were analysed concerning their medical conditions and diagnoses. Results: Less than 50% of the patients with prescriptions for phenobarbital and clonazepam had the ICD diagnosis G40 (epilepsy) on file. The majority of patients received prescriptions for clonazepam and phenobarbital with diagnoses like psychological and behavioural disturbances due to alcoholism, or tobacco, depression or depressive episode, anxiety disorders, hypertension, disturbed development, other psychological disturbances, sleeping disorders, behavioural disturbances. Conclusions: Off label prescriptions for

99

phenobarbital and clonazepam are comparatively more frequent than “on label use”. Furthermore prescribing these substances to patients with pre-existing substance dependencies is critical and should be avoided. It seems necessary to analyze the situation further and to anticipate corrective measures to avoid potential risks. 1. Institute for Clinical Pharmacology, Medical Faculty, Technische Universität Dresden, Dresden, Germany, 2. AOK Plus, Bereich Arzneimittel, Dresden, Germany

502 The prostaglandin transporter OATP2A1 is expressed in human ocular tissues and transports the antiglaucoma prostanoid latanoprost Kraft M.E. (1), Welge-Lüßen U. (2), Schlötzer-Schrehardt U. (2), Kruse F.E. (2), Zolk O. (1), Auge D. (1), Glaeser H. (1), Mandery K. (1), König J. (1), Fromm M.F. (1) Purpose: The mechanisms of ocular uptake and distribution of prostanoids for glaucoma treatment are not well understood. We hypothesized that OATPs are responsible for uptake of clinically used prostanoids into ocular tissues and might contribute to interindividual differences in drug concentrations and effects. Methods and results: Latanoprost (1.5 µg) pharmacokinetics were investigated in 12 patients 91±43 min after topical administration. Intraocular concentrations of the active metabolite latanoprost free acid were determined by LC/MS/MS. Remarkable interindividual variations in the levels of latanoprost free acid (mean 15±14 ng/ml, minimum 1 ng/ml, maximum 44 ng/ml, coefficient of variation 87%) in aqueous humor samples were found. To investigate whether transporters are involved in the intraocular disposition of antiglaucoma prostanoids, mRNA levels of prostaglandin uptake (OATP2A1 and OATP2B1) and efflux transporters (MRP4) were determined in human ocular tissues (ciliary body, choroidea, retina, iris, lens, and cornea, n=2) by real-time RT-PCR. Expression of all transporters was highest in the ciliary body and choroidea, and was much less in retina, iris, lens, and cornea. OATP2A1 and MRP4 mRNA levels in the ciliary body and choroidea were markedly higher (4.0 to 6.3-fold) than in human liver, whereas OATP2B1 mRNA expression in eye tissues was only 1-7% of that in liver. The inhibitory interactions of therapeutically used prostanoids (tafluprost, bimatoprost, latanoprost, travoprost) and their therapeutically active metabolites (the respective free acids) with the uptake of prototypical substrates (bromosulfophthalein and prostaglandin E2) were tested in stably transfected HEK cells overexpressing either OATP2A1 or OATP2B1. IC50 values for inhibition of bromosulfophthalein uptake by OATP2B1 varied from 10-23 µM (travoprost, latanoprost) to 312 µM (tafluprost). IC50 values for inhibition of prostaglandin E2 uptake by OATP2A1 ranged from 0.7-0.9 µM (latanoprost, bimatoprost free acid) to 90 µM (tafluprost). Uptake experiments revealed that the acid of latanoprost was a substrate of OATP2A1 (Km 3.6 µM, Vmax 83.5 pmol/mg protein/min), but was not transported by OATP2B1. Conclusion: Given the functional profile of uptake transporters expressed in vitro and the transporter expression pattern in the human eye, the present results suggest a role of OATP2A1 in the intraocular disposition of therapeutically used prostanoids, such as latanoprost. 1. Institut of Experimental and Clinical Pharmacology and Toxicology, FriedrichAlexander-University, Erlangen, Germany, 2. Department of Ophthalmology, FriedrichAlexander-University, Erlangen, Germany

503

determined. Unchanged denaverine was detected in both intervals. Higher amounts, but less than 0.1 % of the dose, were found in the samples which were mixed with glucuronidase. This means denaverine showed a rarely observed N-glucuronidation. The following Phase I reactions of denaverine were found: N-demethylation, desaminohydroxylation, O-desalkylation with following reduction, C-hydroxylation and formation of aldehyd and carbon acid functions. The C-hydroxylation product MD22 was the main metabolite of denaverine. Only considering the unconjugated metabolites, MD22 amounted 39 % of the excreted mass. In the samples mixed with glucuronidase MD22 reached up to 79 % of the whole excreted mass. MD22 was found in both intervals, much greater amounts in samples mixed with glucuronidase. MD12 is an Noxide and only appeared in low concentrations, but it is remarkable that denaverine is metabolized in this way. MD22 was not found as a glucuronide. Less than 5 % of the administered dose of denaverine was found as metabolites in urine. In most cases Phase II metabolites were formed. The metabolite profiles and the excreted amounts of the six volunteers were similar. 1. Institute of Clinical Pharmacology, Medical Faculty, TU Dresden, Germany

505 LC-MS/MS method for the determination of endocannabinoids in tissue samples Schmidt H. (1), Ziebell S. (1), Geisslinger G. (1) Studies of the endocannabinoid system have demonstrated that endogenous cannabinoids play an important role in pain modulation. We report a liquidchromatography-tandem mass spectrometry (LC-MS/MS) method for simultaneous quantification of endocannabinoids in tissue samples.Pieces of tissue were weighted, homogenized and extracted by liquid-liquid extraction and the reconstituted samples were analyzed for arachidonoyl ethanolamide (anandamide, AEA), palmitoyl ethanolamide (PEA), 2-arachidonoyl glycerol (2-AG), 1-arachidonoyl glycerol (1-AG) and oleoyl ethanolamide (OEA). The respective deuterated analytes AEA-d8, PEA-d4, 2-AGd5, 1-AG-d5 and OEA-d2 were used as internal standards.HPLC analysis was done under gradient conditions using a Luna HST C18(2) column (Phenomenex, Aschaffenburg, Germany). MS and MS/MS analyses were performed on an API 5000 triple quadrupole mass spectrometer with a Turbo V source (Applied Biosystems, Darmstadt, Germany) in the negative ion mode. Precursor-to-product ion transitions of m/z 346→259 for AEA, m/z 354→86 for AEA-d8, m/z 298→268 for PEA, m/z 302→272 for PEA-d4, m/z 377→303 for 2-AG and 1-AG, m/z 382→303 for 2-AG-d5 and 1-AG-d5, m/z 324→86 for OEA, and m/z 326→86 for OEA-d2 were used for the multiple reaction monitoring (MRM) with a dwell time of 75 ms.Concentrations of the calibration standards, quality controls and unknowns were evaluated by Analyst software (version 1.4; Applied Biosystems, Darmstadt, Germany) and normalized to mg protein. Variations in accuracy and intra-day and inter-day precision (n = 6 for each concentration, respectively) were <15% over the range of calibration.The applicability of the method is shown by measuring cannabinoids in various tissue samples. 1. pharmazentrum frankfurt/ZAFES, Institute of Clinical Pharmacology, Goethe University, Frankfurt (Main), Germany

506

Sphingosine-1-phosphate modulates spinal nociceptive processing Linke B. (1), Coste O. (1), Brenneis C. (1), Pierre S. (1), Becker W. (1), Schmidt H. (1), Geisslinger G. (1), Scholich K. (1) The sphingolipid sphingosine-1-phosphate (S1P) is synthesized by phosphorylation of sphingosine by sphingosine kinases (SPHK) in a wide variety of cell types in response to extracellular stimuli such as nerve growth factor or vascular endothelial growth factor. S1P modulates various cellular functions such as apoptosis, cell differentiation and migration. It can act in a paracrine or autocrine way as an agonist for G protein coupled receptors (S1P1-5) or as an intracellular second messenger. Although S1P is an abundant signaling molecule in the central nervous system, very little is known about its role in neuronal functions. We found that S1P concentrations in the cerebrospinal fluid of adult rats were selectively decreased after peripheral nociceptive stimulation. This decrease seems to be due to a decreased activity of spinal SPHKs. Moreover, inhibition of spinal S1P synthesis decreased basal pain thresholds while intrathecal application of S1P reduced the nociceptive behavior in the formalin assay. Furthermore, we showed that S1P does not change calcium influx in spinal cord neurons. S1P and sphinganine-1phosphate (dihydro-S1P) inhibited cyclic AMP (cAMP) synthesis, a key second messenger of spinal nociceptive processing, in spinal cord neurons. By combining fluorescence-resonance-energy-transfer (FRET)-based cAMP measurements with MultiEpitope-Ligand-Cartography (MELC), we found that S1P-responding neurons stained positive for VGLUT-1 and CD56 suggesting excitatory neurons from laminae II and III. Accordingly, intrathecal application of dihydro-S1P abolished the cAMP-dependent phosphorylation of NMDA receptors in the outer laminae of the spinal cord. Taken together the data show that S1P modulates spinal nociceptive processing through inhibition of neuronal cAMP synthesis. 1. Institute of Clinical Pharmacology, Johann Wolfgang von Goethe University, Frankfurt, Germany

Highly sensitive determination of sphingolipids in cells after treatment of various coxibs using LC-MS/MS analysis Schmidt H. (1), Angioni C. (1), Schiffmann S. (1), Geisslinger G. (1) Sphingolipids play an important role in cell cycle regulation, cell proliferation and apoptosis. We report a precise and accurate method for simultaneous quantification of sphingolipids in cells, such as HCT-116, HeLa or HCA-7, using liquid-chromatographytandem mass spectrometry (LC-MS/MS).5x105 cells were incubated for 24h and treated with various coxibs. After counting cells lipids were extracted by liquid-liquid extraction. Internal standards C17:0-ceramide (C17:0-Cer), C17:0-sphingosine (C17:0-Sph) and C17:0-sphingosine-1-phosphat (C17:0-Sph1P) were added prior to extraction. The reconstituted samples were analyzed for sphingolipids such as sphingosine (Sph), C16:0-Ceramide (C16:0-Cer), or sphinganine (dhSph).The compounds were separated using a Luna C18(2) column (Phenomenex, Aschaffenburg, Germany) under gradient conditions and detected with an API 4000 tandem mass spectrometer equipped with a Turbo V source (Applied Biosystems, Darmstadt, Germany) operating in positive ESI mode.Multiple reaction monitoring (MRM) with a dwell time of 15 ms was used for quantification. The mass transitions used were m/z 300→282 for Sph, m/z 566→264 for C16:0-Cer, m/z 302→284 for dhSph, m/z 286→268 for C17:0-Sph, m/z 366→250 for C17:0-Sph1P and m/z 552→534 for C17:0-Cer, respectively. Concentrations of the calibration standards, quality controls and unknowns were evaluated by Analyst software (version 1.4.2, Applied Biosystems, Darmstadt, Germany) and normalized to 106 cells. Variations in accuracy and intra-day and inter-day precision (n = 6 for each concentration in 5 various cell lines (HCT-116, HeLa, HCA-7, MCF-7, HEK293) were less than 15 % over the range of calibration.The applicability of the method is shown by measuring shingolipids in various cell lines after treatment with different selective COX2-inhibitors. 1. pharmazentrum frankfurt/ZAFES, Institute of Clinical Pharmakology, Goethe University, Frankfurt (Main), Germany

504

507

Metabolism of denaverine in human Oertel R. (1), Kilian B. (1), Kirch W. (1) Denaverine is a neurotropic-musculotropic spasmolytic agent with additional analgesic activity. Although denaverine hydrochloride has been used successfully in therapy for more than 30 years, only recently detailed information about its biotransformation in human were published. In 2008 the authors of this study published the analytical methods to determine denaverine and five of its metabolites in human urine and the elucidation of the structure of nine additional metabolites without the possibility of quantification. The method was applied to six volunteers, who got 200 mg denaverine orally. They collected urine in two intervals: 0-6 h and 6-24 h after administration. In a first study all urine samples were analysed without beta-glucuronidase to determine Phase I metabolites. In a second study the sum of Phase I and Phase II metabolites was

Expression and localization of drug transporters in Caco-2 cells Riefenstahl A. (1), Köck K. (1), Grube M. (1), Jedlitschky G. (1), Kroemer H.K. (1) Background: Intestinal absorption is a prerequisite for systemic therapy after oral administration. The Caco-2 cell line is a widely used in vitro model to predict this process as these cells resemble enterocytes both in morphology and drug transporter expression. However, differentiation dependent variations of drug transporter expression have been shown in Caco-2 cells. In addition to this, the localization of these proteins might influence the interpretation of drug transport studies in the Caco-2 cell line. We therefore analysed uptake and efflux protein expression in this cell line. Methods & Results: To verify the effect of differentiation on transporter expression Caco-2 TC7 cells were seeded at a density of 6 x 104cells/cm² and P-glycoprotein (P-gp) and OATP2B1 mRNA and protein levels were investigated as a function of cultivation time. As

100

described in literature, both transporters displayed an increasing mRNA and protein expression up to 24 days post seeding. After verifying this differentiation dependent expression, we analyzed the localization of several drug transporters by immunofluorescence, separation of apical and basolateral membranes by magnesium precipitation as well as biotinylation studies. As expected, we could detect the efflux transporters P-gp, MRP2 and BCRP exclusively in the apical membrane after 21 days of cultivation. The organic anion uptake transporter OATP2B1 however is mainly localized to the basolateral membrane, which is contrary to the previously published apical expression in the intestine and the Caco-2 cell line1. Conclusion: We could demonstrate a differentiation-dependent expression of P-gp and OATP2B1. Furthermore, in our Caco-2 cell line we could show a basolateral localization of the latter protein. As an apical localization of this protein has been described in the small intestine, a basolateral expression might distort the prediction of the intestinal absorption of OATP2B1 substrates like statins. 1 Sai et al. 2006, Predominant Contribution of Organic Anion Transporting Polypeptide OATP-B (OATP2B1) to Apical Uptake of Estrone-3-sulfate by Human Intestinal Caco-2 Cells. Drug Metabolism and Disposition 1. Dept. of Pharmacology, Ernst Moritz Arndt University, Greifswald, Germany

508 Sensitive HPLC-method for the determination of thiamin diphosphate (TDP) in small whole blood samples - a method suitable for pediatric diagnostics Körner R.K. (1), Vierzig A.V. (2), Ebenfeld S.E. (1), Roth B.R. (2), Müller C.M. (1) Vitamin B1 (thiamin) is a watersoluble vitamin, which plays an essential role in carbohydrate metabolism. It is required by the pyruvate dehydrogenase complex for the transformation of pyruvate to acetyl-CoA. More enzyme complexes rely on vitamin B1 as a coenzyme, such as the oxoglutarate dehydrogenase and the transketolase. TDP is the functional coenzyme form of thiamin and takes part in the metabolic pathways mentioned above. Several case reports about life-threatening thiamin deficiency induce the need for a diagnostic tool that can be used to establish the thiamin status in pediatric intensive care units. Assessment of the nutritional status of vitamin B1 in infants and even in preterm infants may indicate the need of a therapeutic drug monitoring. Therefore an improved and easy to use method for the determination of thiamin diphosphate (TDP) in 100µl of whole blood was developed. Sample preparation comprises the extraction of TDP from whole blood by hemolysis, deproteinization with trichloroacetic acid, and subsequent centrifugation. Potassium ferricyanide is used for pre-column derivatization of TDP to its fluorescent thiochrome derivative. Chromatographic separation was carried out using a reversed-phase column and an isocratic elution which consisted of a phosphate buffer and acetonitrile. The mobile phase was delivered at a flow rate of 1.0ml/min. TDP was detected fluorimetrically (excitation wavelength, 375nm; emission wavelength, 430nm) and quantified by external standardization. The lower limit of detection and the lower limit of quantification were 0.2ng/ml and 4ng/ml, respectively. The intra-assay CV was 2.5% for samples processed from the same specimen (N=14) and ranged between 1.5-3.0% for spiked whole blood samples. Inter-assay CV was 4.2% (N=5). Linearity was demonstrated over the expected concentration range of 4-400ng/ml. In conclusion, we present a convenient HPLC method for the determination of TDP, which is precise, sensitive and suitable for pediatric diagnostics. 1. Department of Pharmacology, University of Cologne, Germany, 2. Department of Neonatology and Pediatric Intensive Care, Children's Hospital of the University of Cologne, Germany