Naunyn-Schmiedeberg's Arch. Pharmacol. 290, 357--373 (1975) 9 by Springer-Verlag 1975

Intraaxonal and Extraaxonal Transport of 125I-Tetanus Toxin in Early Local Tetanus G. E r d m a n n , H . W i e g a n d , a n d H . H . W e l l h S u e r Pharmakologisches Institut der Justus Liebig-Universitgt Giel]en Received May 29 / Accepted June 30, 1975

Summary. 1. The distribution of radioactivity in the sciatic nerve, the spinal ganglia, the ventral roots and the spinal cord was studied by means of histoautoradiography after injection of 125I-labelled tetanus toxin into gastrocnemius muscles of cats. 2. In the sciatic nerve the major part of the radioactivity was found in the epineurium, but some axons also contained radioactivity. 3. In the ventral root the radioactivity was strictly confined to a few axons; no radioactivity was found in other parts of the ventral root. 4. In the spinal cord the radioactivity was confined to a few motoneurones where it was found in the soma as well as in the dendrites. 5. Transient cooling of the ventral roots prevented the ascent of radioactivity into the spinal cord. 6. Colchicine and vinblastine, after local application to the sciatic nerve, reduced the amount of radioactivity found in the ventral roots and in the spinal cord. However, the same effect was also obtained but to a lesser degree with lumieolehicine. 7. I t is concluded that the intraaxonal compartment is involved in the neural ascent of tetanus toxin into the spinal cord. Key words: Tetanus Toxin -- Retrograde Axonal Transport -- Colchicine, Vinblastine -- Motoneurones. A n e u r a l a s c e n t o f n a t i v e t e t a n u s t o x i n in local t e t a n u s has b e e n s u g g e s t e d or d e m o n s t r a t e d b y m a n y a u t h o r s (for r e v i e w see W r i g h t , 1955; L a u r e n c e a n d W e b s t e r , 1963; K r y z h a n o v s k y , 1966, 1967, 1973; A d a m s et al., 1969; L a m a n n a a n d C a r l 1967; F e d i n e c , 1967; Z a c k s a n d Sheff, 1970; v a n H e y n i n g e n a n d M e l l a n b y , 1971; B i z z i n i et aI., 1974). T h e q u a n t i t a t i v e e v a l u a t i o n of t h e s e e x p e r i m e n t s m e t c e r t a i n difficulties (discussed e l s e w h e r e : W e l l h 6 n e r et al., 1973a) w h i c h w e r e o v e r c o m e

Send o/[print requests to: H. H. WellhSner, Pharmakologisches Institut der Justus Liebig-Universitgt, D-6300 GieBen, Frankfurter StraBe 107, Federal Republic of Germany. This work was supported by a grant from the Deutsche Forschungsgemeinschaft to H. H. W. A communication was presented at the Fourth International Conference on Tetanus, Dakar, April 6--12, 1975.

358

G. Erdmann et al.

when highly purified a n d labelled t o x i n became available (3H-toxin: Fedinec, 1967; R o n n e v i et al., 1973; ~3~I-toxin: K r y z h a n o v s k y et al., 1970; 12~I-toxin: H a b e r m a n n , 1970, 1972; Stoeckel et al., 1975).--There is agreement t h a t t e t a n u s t o x i n reaches the spinal cord b y n e u r a l ascent via the v e n t r a l roots. A l t h o u g h t e t a n u s t o x i n has also been f o u n d i n a variable a m o u n t i n the dorsal roots of some a n i m a l species, a contrib u t i o n to the a c c u m u l a t i o n of t e t a n u s t o x i n i n the spinal cord b y a flow of t o x i n t h r o u g h the dorsal roots has n e v e r been d e m o n s t r a t e d a n d the m a j o r i t y of a u t h o r s do n o t consider such a flow to be i m p o r t a n t . However, no direct i n f o r m a t i o n exists on the tissue c o m p a r t m e n t i n the v e n t r a l root which is concerned with the ascent of t o x i n i n t o the spinal cord a n d this has been i n v e s t i g a t e d i n the work presented below. To o b t a i n f u r t h e r i n f o r m a t i o n we also looked a t the peripheral nerve, the spinal ganglion a n d the spinal cord. Materials and Methods

General Experiments were performed on 24 cats of either sex weighing 1.8--3.3 kg. Tetanus toxin was obtained from Drs. Bizzini and Turpin, Institut Pasteur and was labelled with 1~5Iin our department by Drs. Habermann and R~iker according to methods previously described (ttabermann, 1970, 1972). With the calculation described elsewhere (WellhSner et aI., 1973a) the characteristic values of the labelled batches were in the range of one minimal lethal dose per kg cat (1 MLD/kg) equal to 4.5 Izg l~sI-tetanus toxin and corresponding to 3.8 • 106 cpm (windows of the gamma spectrometer at 48 and 100 KeV) or 4.75 • 10-6 Ci. For injection, lesI-toxin was dissolved in saline containing 0.1~ bovine serum albumin. The cats were anesthetized with ether to allow the injection of 0.5--2.7 MLD l~sI-toxin into one or into either gastrocnemius muscle.

Histoautoradiography In preliminary experiments the reliability of the standard tissue embedding technique was tested by searching for signs of dislocation of radioactivity after aldehyde fixation. Twenty-four hrs after injection of 125I-toxin, samples of neural tissue were taken from non-perfused animals, their wet weights and radioactivities were determined, and they were transferred into Karnovsky's fixative for 12 hrs. They were then homogenized in 20 volumes of phosphate-buffered saline first in a Potter glass homogenizer and then in an ultrasonic device. :Each homogenate was centrifuged at 8000 g for 10 rain and the radioactivity present in the pellets and supernatants was determined. In the main experiments, 0.5--2.7 MLD 125I-toxin/kg cat was injected into tile gastrocnemius muscle under ether anesthesia. Twenty-four hrs later, the cats were anesthetized again with pentobarbital, one jugular vein was canulated for the injection of heparin prior to perfusion, the blood pressure was monitored from one carotid artery, the posterior vena cava was exposed, the horizontal branch of a T-shaped cannula was inserted into the abdominal aorta, and a perfusion system was connected, tteparin (5 mg/kg) was injected intravenously, 1 rain later the posterior vena cava was opened and the perfusion started with 1 1 of Ringer's solution and continued with 2.5 1 of Karnovsky's fixative. :By use of a pneumatic device the external perfusion pressure was adjusted to give an intracarotid fluid pressure of

Intraaxonal and Extraaxonal Transport of Tetanus Toxin

359

a t least 80 m m Hg. We considered it i m p o r t a n t to measure the intraarterial pressure during the perfusion and to avoid values below 60 m m Hg. The sciatic nerves were t h e n quickly exposed a n d covered with gauze soaked in Karnovsky's fixative. After laminectomy a n d removal of the dura m a t e r the spinal cord was also covered with K a r n o v s k y ' s solution. The cord, the root a n d the nerves were removed, tissue samples were prepared, weighed a n d counted as described earlier (WellhSncr et al., 1973a) with the modification t h a t the tissue samples were fixed for 6--12 hrs in K a r n o v s k y ' s fixative before counting a n d were counted in vessels filled with eaeodylate buffer. The samples were t h e n dehydrated in aleohol, transferred into methyl benzoate a n d benzene and embedded in Paraplast e. Sections of 7 ~m thickness were cut and dewaxed in xylene. The autoradiographs were prepared using Ilford L4 emulsion (diluted 1:2 with distilled water) and K o d a k D 19b developer (2.5 min a t 20 ~ C). The sections were exposed for 5--25 weeks at 4 ~ C; the average exposure time was 14 weeks. Sections stained with cresyI violet or hematoxylin as well as unstained sections were photographed using a Leitz Ortholux microscope a n d a Leitz O r t h o m a t camera.

Cooling the Dorsal Roots A laminectomy was performed under pentobarbital anesthesia and the ventral roots were frozen b y the method of Ochs a n d J o h n s o n (1969). Briefly, the ventral root $1 was lifted with a t h r e a d and laid on an aluminium hook precooled to --70 ~ C. The root was t h e n left on the hook until freezing of the loaded p a r t was complete, which took approximately 8 - - 1 0 sec a n d could be clearly recognized. The freezing hook was placed near the e n t r y of the root into the spinal cord. Care h a d to be t a k e n n o t to touch other roots or the spinal cord with the hook. After freezing was complete, saline a t room temperature was poured over the frozen p a r t to allow removal of the hook. The procedure was t h e n repeated on the ventral roots L7 a n d L6. On the contralateral side, the ventral roots were lifted and placed for 10 sec on the hook which was, however, a t room temperature. Subsequently m~I-labelled tetanus toxin was injected in a dose of 0.6--0.9 MLD/kg into either gastrocnemius muscle. For the next 24 hrs intensive care was given to the cats as described b y WellhTner et al. (1973b) after which the cats were killed b y exsanguination a n d tissue samples were t a k e n and processed as described previously (WellhSner et al., 1973a).

Local Exposure o] the Sciatic Nerves to VinbIastine, Colchicine and Lumicolchicine Axonal t r a n s p o r t in b o t h directions can be inhibited with colchicine (Kristensson a n d Sj5strand, 1972; Kristcnsson et al., 1971; Abe et al., 1974; H e n d r y et al., 1974) and vinblastine (Edstr5m and Hanson, 1973). The procedure described by DahlstrSm (1968) a n d HSkfelt a n d DahlstrSm (1971) for rats was adapted for cats in the following way: The cats were anesthetized with ether. The sciatic nerves on b o t h sides were exposed in the hollow of the knee. On one side cotton pellets soaked in solutions of vinblastine (Velbe * E. Lilly, GieBen), colchicine (Merck, Darmstadt) or lumicolchicine were placed around the exposed part of the nerves. On the other side cotton pellets soaked in saline were used. The pellets were removed after 30 rain, sodium penicillin was instilled into the cavity, the wounds were sutured, and the application of ether was terminated. Twenty-four hrs later, the cats were anesthetized with ether once more and 12SI-labelled tetanus toxin was injected in a dose of 0.5--0.9 MLD/kg into both gastroenemius m uscles.After another 24 hrs the cats were killed b y exsanguination a n d samples were processed as described (WellhSner et al., 1973 a).

G. Erdmann et al.

360

Vinblastine has been found to be effective in a concentration of 10-~ M in cultures of Itela cells (Creswell, cited by Wilson et at., 1974), but concentrations up to 0.01 M have been used by HSkfelt and DahlstrSm (1971). Experiments were therefore performed to find the lowest concentrations of vinblastine and colchicine still affecting the ascent of labelled tetanus toxin. Lumicolchieine was prepared according to Price (1974). The completeness of conversion from colchicine to lumicolchicine was ascertained by ultraviolet spectroscopy. Lumicolchicine and colchicine were used on the nerves in the same molar concentrations.

Results

Tissue Fixation and Dislocation o/Radioactivity W h e n freshly o b t a i n e d unfixed n e r v o u s tissue c o n t a i n i n g radioa c t i v i t y was homogenized a n d centrifuged, a considerable a m o u n t of r a d i o a c t i v i t y was f o u n d in the s u p e r n a t a n t . However, when the n e r v o u s tissue was fixed i n K a r n o v s k y ' s fixative prior to homogenization, no r a d i o a c t i v i t y was detectable in the s u p e r n a t a n t . A n example is given in T a b l e 1.

Table 1. Influence of Karnovsky's fixative on the mobility of radioactive material in the sciatic nerve. All counts were made for 20 rain

Wet weight of sample Radioactivity of sample l~adioactivity of supernatant l~adioactivity of pellet

Sciatic nerve, non-fixed

Sciatic nerve, fixed

28.0 mg 38 cpm 10 cpm 27 cpm

26.6 mg 29 cpm 0 cpm 26 cpm

Ventral Roots, Autoradiographie Studies Autoradiographs of the v e n t r a l roots showed t h a t the r a d i o a c t i v i t y was strictly confined to a few axons (Fig. l). Since these axons h a d occasional sections deficient i n labelling it was difficult to get a n idea of the t o t a l n u m b e r of radioactive fibres in a n i n d i v i d u a l v e n t r a l root. This difficulty was more or less overcome b y application of ligatures to the v e n t r a l roots near their e n t r y into the spinal cord. This p r e v e n t e d the flow of radioactivity, resulting i n a drastic increase i n a x o n a l labelling over relatively long c o n t i n u o u s stretches. E v e n t h e n r a d i o a c t i v i t y was f o u n d only in axons (Fig. 2), while the e n d o n e u r i u m a n d the fiat mesot h e l i u m covering the v e n t r a l roots showed no radioactivity. V a r y i n g n u m b e r s of labelled axons were f o u n d in different v e n t r a l roots. However, i n the same ligatured root the v a r i a t i o n b e t w e e n n u m b e r s o b t a i n e d from consecutive sections was small. F o r instance, from the v e n t r a l root SI shown i n Fig. 2 the following n u m b e r s of radioactive axons were o b t a i n e d i n five consecutive sections : 33, 32, 29, 31, 31.

Intraaxonal and Extraaxonal Transport of Tetanus Toxin

1

361

2

Fig. 1. Autoradiograph of the right ventral root S1 showing one labelled axon 24 hrs after injection of 1.1 • l06 cpm 125I-tetanus toxin/kg body weight into the right gastroenemius muscle. Activity in the right ventral root S1 : 1236 =k36 cpm/g wet weight. Free passage of radioactivity through the ventral root was allowed during the time of toxin ascent. Unstained section. • 930 Fig. 2. Autoradiograph of the left ventral root S1 showing one labelled axon 24 hrs after injection of 7 • l06 epm 12~I-tetanus toxin/kg body weight into the left gastrocnemius muscle. Ten hrs after toxin injection a ligature was placed on the ventral root near its entry into the spinal cord. Activity in the left ventral root S1 : 1733 =k 156 cpm/g wet weight. Note that neither the endoneurium nor the mesothelium covering the root contains radioactivity. Cresyl violet. • 930

Spinal Cord, A utoradiographic Studies I n the spinal cord the r a d i o a c t i v i t y was confined to a few motoneurones i n the half segments supplying the injected g a s t r o c n e m i u s muscle (Fig. 3). The a c t i v i t y was considerable i n the p e r i k a r y o n as welI as i n the dendrites, while the cell nucleus had only low activity. T w e n t y four hrs after i.m. i n j e c t i o n of 125I-labelled t e t a n u s t o x i n i n t o the gast r o c n e m i u s muscle a transfer of r a d i o a c t i v i t y from labelled moton e u r o n e s to small i n t e r n e u r o n e s was n o t obvious. I n other experiments, i n which labelled t e t a n u s t o x i n was injected i n t o intercostal muscles, again a few m o t o n e u r o n e s became labelled, b u t 24 hrs after t o x i n i n j e c t i o n there was no obvious labelling of cells in t h e lateral c o l u m n of the spinal cord giving rise to efferent s y m p a t h e t i c fibres.

G. Erdmann et al.

362

Fig. 3. Autoradiograph demonstrating label distribution in the left spinal cord half segment L7 24 hrs after injection of 5 • 10Gcpm 12sI-tetanus toxin/kg body weight into the left gastrocnemius muscle. Activity in the left half of the spinal cord segment L7: 2475 =~39 cpm/g wet weight. Radioactivity is seen in the perikarya of three motoneurones and in a dendrite (inset). Cresyl violet. • 320

Cooling the Ventral Roots, Gross Counting o/ Spinal Cord Freezing of the v e n t r a l roots according to Ochs a n d J o h n s o n (1969) interferes with a x o n a l flow. I n our experiments, freezing of the v e n t r a l roots resulted in a reduced r a d i o a c t i v i t y in the corresponding spinal cord half segments as compared to the c o n t r a l a t e r a l half segments where the inflow of r a d i o a c t i v i t y was left u n i m p a i r e d (Table 2). Freezing a p p a r e n t l y i n d u c e d a n a c c u m u l a t i o n of r a d i o a c t i v i t y in the v e n t r a l roots.

Table 2. Radioactivity (cpm/g wet weight) in the ventral roots and the spinal cord of a cat 24 hrs after injection of 2.37 • Gcpm t25I-tetanus toxin into either gastrocncmius muscle. Prior to toxin injection the ventral roots of the left side were transiently frozen as described by Ochs and Johnson (1969) Right side (control)

Left side (freezing) Ventral root

Spinal cord

Spinal cord

Ventral root

L6

325 2~ 62

94 ~- 11

151 ~ 12

159 -4- 50

L7

1519 zt= 46

138 =~ 18

341 ~= 20

544 -t- 42

Intraaxonal and Extraaxonal Transport of Tetanus Toxin

363

Fig. 4. Autoradiograph of the left L7 spinal ganglion demonstrating radioactivity in a pseudounipolar cell 24 hrs after injection of 1.5 • 106 cpm 12aI-tetanus toxin/kg body weight into the left gastrocnemius muscle. Hematoxylin. • 465

.Spinal Ganglia, Autoradiographie Studies After injection of l~I-tetanus toxin into the gastrocnemius muscle radioactivity was also found in a few pseudounipolar cells of the corresponding spinal ganglia. These cells displayed a smaller intensity of labelling than the motoneurones (Fig4).

Peripheral Nerve, Autoradiographic Studies Autoradiographs prepared from cross sections taken from the distal segment of the sciatic nerve showed that the bulk of radioactivity was located in the epineurium (Fig. 5). Far fewer grains were present in the perineurium. Radioactivity in the fascicles enclosed by the perineurium was poor on the average with the exception of 1--3 fascicles. I n these fascicles radioactivity was elevated in small areas (Fig. 6). I n most cases, it was hardly possible to attribute this labelling to any particular structure. This difficulty may have arisen partly from poor autoradiographic resolution due to a comparatively long range of radiation which is evident from Fig. 2. IIowever, in a few cases, a definite localization was possible and labelling could be ascribed to the endoneurium or to the axon (inset Fig. 6). Preferential labelling of the intraaxonal space was much less prominent in the peripheral nerve than in the ventral roots.

364

G. E r d m a n n et al.

Fig.5. Autoradiograph of the left sciatic nerve 24 hrs after injection of 5 • 106 cpm lssI-tetanus toxin/kg body weight into the left gastroenemius muscle. The section was t a k e n from a site 3 cm above the entry of the gastrocnemius nerve into the muscle. The bulk of radioactivity is seen in the epineurium. Hematoxylin. • 465

Fig. 6. Autoradiograph of the left sciatic nerve showing the distribution of radioactivity within a fascicle 24 hrs after injection of 5 X 106 cpm l~sI-tetanus toxin/kg body weight into the left gastrocnemius muscle. I n a few axons the radioactivity is slightly elevated (one axon in inset), ttematoxylin. • 465

106 (pMmaris longus)

106 (gastrocnemius)

142

vinblastine } 112 ~M on sciatic nerve

vinblastine ] 11.2 mM on / sciatic nerve

vinblastine ) 11.2 m M o n ~ sciatic nerve J

SC = Spinal cord; V R = Ventral root.

l0 s (gastrocnemius)

saline

0.66 • 106 (gastrocnemius) spinMized cat

98

143

saline

1.33 • 106 (gastrocnemius)

92

vinblastine ) 11.2 m M o n / sciatic nerve

106 (gastrocnemius)

91

vinblastine ] !1,2 m3/!on / sciatic nerve

Drug on the left

106 (gastrocnemius)

cpm l~aI-toxin/kg into either M.

87

Cat no.

saline

vinblastine 112 fzM on sciatic nerve

/ ~ J

vinblastine ] 1.12 mM on / median nerve

saline

saline

saline

saline

Drug on t h e r i g h t

SC L6 SC L7

SC L7 SC $1 VRL7 VRS1

SC C7 SC TH1

SC L7 SC S1

SC L6 SC L7 s c s1

SC L6 SC L7 SC S1

SC L7 SC St

Tissue; Segment

~: ~: ~: ~

12 21 41 72 120 i 9 496 -b 16

274 405 1094 954

177 ~ 10 178 ~ 12

46 • 19 52 :t: 10

48 • 12 100 :t: 13 79 4- 30

49 J : 11 161 i 14 131 ~ 28

85 ~ 6 I01 :[_ 9

drug side

• -j: i ~:

12 23 53 66 266 :L 10 1308 • 22

401 872 1738 1818

429 :t: 10 1009 :~: 16

78 • 19 138 :t- 11

172 • 11 1061 • 25 101 • 25

79 :t: 10 395 :~ 16 257 ~: 32

610 :[: 10 1035 i 16

control side

cpm/g wet weight on

Table 3. l~adioactivity in ventral roots a n d half segments of spinal cord 48 hrs after injection of leaI-tetanus toxin into the gastrocnemius or palmuris longus muscles a n d 24 hrs after topical application of vinblastine to t h e sciatic or median nerves

r

3"

O N

9z

~z N 9

N

O

eolehieine } 220 IxM on median nerve

eolehicine 220 tzM on sciatic nerve

lumieolchicine } 220 lxM on median nerve

1.5 X 10 s (palmaris longus)

2 • 106 (gastrocnemius)

2 • 10 ~ (palmaris longus)

144

145

SC = Spinal cord; VK = Ventral root.

colchicine 1.25 mM on median nerve

0.83 • 106 (palmaris longus) }

lumicolchicine } 1 . 2 5 mM o n sciatic nerve

0.83 X 10r (gastroenemius)

150

lumicolchicine } 125 vM on sciatic nerve

1.3 • 106 (gastrocnemius)

147

}

Drug on the left

cpm l~I-toxin/kg into either M.

Cat no. on

the right

saline

saline

eolchieine 125 txM on sciatic nerve

saline

saline

saline

Drug

SC C6 SC C7

SC S1 Vl~ $1

SC L6 SC L7 SC $1

SC C5 SC C6

SC L6 SC L7 SC $1 VRL6 V/~L7 VRS1

SC C5 SC C6 SC C7

Tissue; Segment

=E 14 ~ 26 i 33 j= 93 i 92 ~c i23

126 ~ 7 173 J_ 7

395 • 13 892 i 42

203 ~- 9 416 ~ 12 591 ~ 18

728 =t: 16 1375 • 19

288 1734 1496 917 3929 3565

272 J- 8 459 ~_ 9 560 ~_ 11

drug side

i 15 =~ 29 ~ 45 ~ 113 • 91 ~- 135

151 :~ 7 316 ~ 9

595 ~c 16 1204 =[_ 42

168 ~ 9 328 ~ 12 707 ~ 18

1211 i 19 1701 =~ 20

647 2202 2803 2285 4450 6263

550 J- 9 938 =~ 13 687 ~ 12

control side

cpm/g wet weight on

Table 4. l~adioactivity in ventral roots and half segments of spinal cord 48 hrs after injection of msI-tetanus toxin into the gastrocnemins or pMmaris longus muscles and 24 hrs after topical application of colchicine or lumicolehicine to the sciatic or median nerves

~

Intraaxonal and Extraaxonal Transport of Tetanus Toxin

367

Local Application o/ Vinblastine, Colchicine and Lumicolchicine, Gross Counting o~ Spinal Cord and Ventral Roots A number of cats were used to evaluate the minimal concentrations of vinblastine and colchicine still effective in interfering with the neural ascent of labelled toxin. A summary of these experiments is given in Tables 3 and 4. The minimal concentration was found to be in the range of 10 -4 M for vinblastine, for colchicine and also for lumicolchicine. I t was surprising to see that conversion of colchicine to lumicolchicine did not reduce to any appreciable degree the property of the alkaloid molecule to inhibit the neural ascent of radioactivity after injection of 125I-toxin.

Discussion As the findings in the ventral roots are the basis for our conclusions, we start the discussion by asking : Does the confinement of radioactivity to a few axons in the ventral root really indicate that toxin (i.e. material that is still toxic) reaches the spinal cord by the same route ? - - I t should be pointed out that the challenges and arguments given in the following pages refer to the stage of early local tetanus, i.e. for the stage of intoxication 24 hrs after i.m. injection of native or labelled tetanus toxin. At this time motor signs of local tetanus are already well developed after injection of only one MLD (cat) of toxin. Challenge (1). Tetanus toxin reaching the spinal cord via the dorsal roots contributes to the accumulation of toxin in the spinal cord in early local tetanus and to spinal symptoms of local tetanus. Reply: This argument holds neither for native toxin nor for labelled toxin. Although in some animal species considerable amounts of toxin may be found in the dorsal roots (for instance in the donkey: Kryzhanovsky et al., 1961a), the same authors (Kryzhanovsky et al., 1961b) found native tetanus toxin only occasionally in the dorsal roots of cats and make the following statement : "The fact that in rare cases it could be found in cats in posterior roots cannot be regarded as a regular and characteristic phenomenon for these animals as statistical evaluation of the finding shows that . . . the detection of toxin in the posterior roots in the above cases must be due to individual features of some of the animals." With l~aI-labelled tetanus toxin in cats, essentially the same results were obtained as with native toxin (WellhSner et al., 1973a). On the basis of extensive experiments in mice, rats, guinea pigs, rabbits, cats, dogs, monkeys and donkeys Kryzhanovsky (1967) states with respect to the dorsal roots: " . . . such way can not be considered as having essential pathogenic importance." One of the main arguments in favour of this conclusion is that after injection of native tetanus toxin, local tetanus still develops after ligature of the dorsal roots but not after 24

Xaunyn-Schmiedeberg's Arch. Pharmacol., Vol. 290

368

G. Erdm~nn et al.

ligature of the ventral roots. Essentially the same result was obtained with 12hi-tetanus toxin in cats by Wellhhner et al. (1973a). Recently Stoeckel et al. (1975) also failed to find labelled tetanus toxin in the dorsal roots of rats. The existence of tetanus dolorosus is no counterargument, as working hypotheses can be offered to explain the symptoms in other ways. The first has been given by K r y z h a n o v s k y who introduced the concept of the "universal dispatch station". Others are : toxin m a y pass from motoneurones to spinal interneurones b y interneuronal transfer (Schubert and Kreutzberg, 1974, see also for other references); toxin m a y act directly on the pseudounipolar cells in the spinal ganglia - - i t has been detected in the perikarya of these cells b y Stoeckel et al. (1975) as well as in the present work; toxin m a y even pass in late local tetanus via the dorsal roots to the spinal cord after having broken down the functional barrier described in more detail by Stoeckel et al. (1975) for the pseudounipolar afferent neurones. Challenge (2) : Radioactivity has been removed from the spinal roots during the histological treatment. Reply: The results of our dislocation tests exclude this possibility. I n these experiments the animals were not perfused and the tissue samples were first fixed and then thoroughly homogenized in phosphate-buffered saline. In spite of this, after centrifugation the radioactivity was found exclusively in the pellet and not in the supernatant or in the fixative. Moreover, no sign of dislocation was detected in the autoradiographs of the ventral roots. Challenge (3) : The radioactivity found in the intraaxonal spaces does not represent toxin, but degradation products. Toxic material migrates between the axons. Reply : Even after the radioactivity had reached the spinal cord, on SDS-gel filtration almost all the radioactivity still appeared in the eluates in one peak which coincided with an external toxin standard and no free 125I was found (Habermann et al., 1973). I f we admit t h a t labelled tetanus toxin reaches the spinal cord via the ventral roots we then have to admit that the percentage of l~hI-tetanus toxin in the ventral roots must be at least as high as in the spinal cord. However, this labelled material is found exclusively within the axons of the ventral roots. To escape the conclusion t h a t the intraaxonal space is the essential p a t h w a y for the ascent of toxin (leading to early local tetanus) one must postulate t h a t only native toxin has the ability to ascend in the endoneural compartment, i.e. extraaxonally. However, even this hypothesis is against existing evidence. After i.m. injection of native toxin ehromatolysis and other morphological changes have been found (Kryzhanovsky et al., 1973; Tarlov et al., 1973; Tarlov, 1974) in motoneurones. Tarlov et al. emphasize that in early tetanus striking changes in morphology are found only in a small fraction (1:64) of motoneurones and t h a t these motoneurones are located in the half

Intraaxonal and Extraaxonal Transport of Tetanus Toxin

369

segments of the spinal cord giving rise to the innervation of the injected muscle. The chromatolysis was also observed by Dimpfel and Habermann (1973) after i.m. injection not of native but of 125I-toxin; and they found the radioactivity exactly over the chromatolytie motoneurones. We find it extremely hard to imagine that non-labelled tetanus toxin ascends extraneuronally into the spinal cord and there selects only a few motoneurones to enter and that native toxin should select just these motoneurones which became radioactive from intraneuronal ascent of labelled material, when labelled toxin had been injected. Challenge (4): The phenomena of tetanus ascendens and descendens are difficult to explain on the basis of a predominantly intraaxonal transport of toxin. Reply: Tetanus ascendens (and descendens) is observed not in early local tetanus but in later stages of the disease. For these stages, working hypotheses compatible with an intraaxonal transport can be offered. The concept of universal dispatch and the concept of interneuronal transfer have already been mentioned above. Challenge (5) : Assuming an intraaxonal transport of tetanus toxin it is difficult to explain the beneficial effects of intrathecal injections of antitoxin (Kryzhanovsky and Krasnova, 1971). Reply: Even a speculative a t t e m p t to explain the action of intratheeal antitoxin should be based on the pharmacokinetics of tetanus antitoxin after intrathecal injection. While a first report has been given (Bolot et al., 1975) sufficient details are not available as yet. We would like to emphasize t h a t the working hypotheses mentioned above have been quoted merely because one of the authors has been asked for such speculations at the Fourth International Conference on Tetanus. I t is not our intention to suggest that we favour any one of these working hypotheses, or to anticipate the results of further experimental work. As to the ventral roots we merely conclude at the present time that in early local tetanus 12aI-labelled tetanus toxin reaches the spinal cord via intraaxonal ascent through the ventral roots and with the available experimental facts we do not find any experiment demonstrating t h a t native tetanus toxin as compared with 12SI-labelled tetanus toxin has qualitatively different pharmaeokineties in the ventral roots in early local tetanus. Our findings of radioactivity in the motoneurones are in line with the results obtained by Dimpfel and H a b e r m a n n (1973) and by Stoeekel et al. (1975). Radioactivity in pseudounipolar cells of the spinal ganglia has been found, by Stoeckel et al. as well as b y us. However, there is a quantitative difference between their results and ours. While in the experiments of Stoeckel et al. m a n y unipolar cells were found to be labelled after injection of 125I-tetanus toxin into the forepaw, in our experiments only a few cells became labelled following i.m. injection of l~I-tetanus 24*

370

G. Erdmann et aI.

toxin. As an explanation we suggest that after i.m. injection toxin also ascended through muscular afferents and that the corresponding unipolar cells became labelled. Since the number of afferents originating in the gastrocncmius muscle is smaller than the number of afferents originating in the densely innervated skin of the forepaw, this might explain the quantitative difference between our results and those of Stoeekel et al. As to the peripheral nerve we agree with Gardner and Fedinec (1975) who used an immunohistochemical technique. We found the bulk of radioactivity in the epineurium. This also confirms King and Fedinee (1973), who found the bulk of toxicity in the epineurium. However, for the ascent of toxin into the spinal cord the epineural compartment is a 'dead-end' station. Even small protein molecules, e.g. horseradish peroxidase (Klemm, 1970), can not penetrate the perineurium; this is true even in inflammation (S6derfeldt, 1974). In addition the epineurium has no intradural continuation. A detailed morphological description has been given for the cat by Andres (1967). Our experiments with vinblastine and colehieine gave the expected results if one assumes t h a t these drugs inhibit axonal transport in both directions. However, this assumption seems to be doubtful in the light of the experiments reported above. Firstly, the concentrations of colehieine and vinblastine interfering with the ascent of 125I-tetanus toxin are still very high as compared with concentrations effective in other systems (Fink et al., 1973). Secondly, while the conversion of colchicine to lumicolehieine was complete (as judged by the UV-spectrum), the irradiation product of colchicine still interfered with the ascent of tetanus toxin, although to a lesser degree than eolehicine. Comparative investigations of the action of purified lumieolchicine, colchieine and vinblastine on retrograde axonal transport and pharmaeokinetic studies on the distribution of these drugs after topical application are required. Addendum. Additional evidence for a retrograde transport of 125I-tetanus toxin in axons of crushed peripheral nerves has recently been presented by D. L. Price, J. Griffin, Ann Young, K. Peck and Adelaine Stocks in Science 188, 945--947 (1975). Acknowledgement. Tetanus toxin was obtained from Drs. Bizzini and Turpin, Institut Pasteur (Paris). It was labelled by Prof. tIabermann, Pharmakologisches Institut (Giegen). Prof. Thoenen, Biozentrum (Basel), placed at our disposal a manuscript of a paper submitted for publication. Valuable suggestions were made by Prof. Oksche and Prof. MSller, Anatomisches Institut (Giefien). Mrs. B~rbel Panz provided technical assistance. Our thanks are due to all these people.

References

Abe, T., Haga, T., Kurokawa, M.: P~ctrograde axoplasmic transport: Its continuation as anterograde transport. FEBS Lett. 47, 272--275 (1974)

Intraaxonal and Extraaxonal Transloort of Tetanus Toxin

371

Adams, E.B., Laurence, D. 1~., Smith, J. W.C.: Tetanus. Oxford-Edinburgh: Blaekwell Scientific Publications 1969 Andres, K. H. : Uber die Feinstruktur der Arachnoidea und Dura mater yon Mammalia. Z. ZellForsch. 79, 272--295 (1967) Bizzini, B., Turpin, A., Raynaud, M. : Immunochimie et m6canisme d'action de la toxine t6tanique. Bull. Inst. Pasteur 72, 177--219 (1974) Bolot, J . F . , Arnand, Ph., Charmeau, A., Fournier, G. : Etude isotopique du metabolisme des immunoglobulines antitetaniques d'origine humaine. Abstract, Fourth International Conference on Tetanus. Dakar. April 6--12, 1975 DahlstrSm, A.: Effect Of colchicine on transport of amine storage granules in sympathetic :nerves of rat. Europ. J. Pharmacol. 5, 111--113 (1968) Dimpfel, W., Habermann, E.: Histoautoradiographic localisation of mSI-labeled tetanus toxin in rat spinal cord. Naunyn-Schmiedeberg's Arch. Pharmacol. I~80, 177--182 (1973) EdstrSm, A., Hanson, M. : Retrograde axonal transport of proteins in vitro in frog sciatic nerves. Brain Res. 61, 311--320 (1973) Fedinec, A. A.: Absorption and distribution of tetanus toxin in experimental animals. In: Principles on tetanus, L. Eckmalm, ed., pp. 168--175. BernStuttgart: H. Huber 1967 Fink, B. R., Byers, M. R., Middaugh, M. E.: Dynamics of colchicine effects on rapid axonal transport and axonal morphology. Brain Res. 56, 299--311 (1973) Gardner, D. P., Fedinec, A. A. : Localization of tetanus toxin in the rat sciatic nerve. Abstract, Fourth International Conference on Tetanus. Dakar. April 6--12, 1975 Habermann, E. : Pharmakokinetische Besonderheiten des Tetanustoxins und ihre Beziehungen zur Pathogenese des lokalen bzw. generalisierten Tetanus. NaunynSchmiedebergs Arch. Pharmak. 267, 1--19 (1970) Habermann, E. : Distribution of msI-tetanus toxin and ~25Ltoxoid in rats with local tetanus as influenced by antitoxin. Naunyn-Schmiedeberg's Arch. Pharmacol. 272, 75--88 (1972) Habermann, E., Dimpfel, W., Rgker, K. 0. : Interaction of labelled tetanus toxin with substructures of rat spinal cord in vivo. Naunyn-Schmicdeberg's Arch. Pharmaeol. 276, 361--374 (1973) Hendry, I.A., StSckel, K., Thoenen, H., Iversen, L. L.: The retrograde axonal transport of nerve growth factor. Brain Res. 68, 103--121 (1974) van Heyningen, W.E., Mellanby, J. H. : Tetanus toxin. In: Microbial toxins, S. Kadis, T. 0. Montie and S. J. Ajl, eds., Vol. IIA, pp. 69--108. New YorkLondon: Academic Press 1971 HSkfelt, T., DahlstrSm, A.: Effects of two mitosis inhibitors (colchicine and vinblastine) on the distribution and axonal transport of norach'enaline storage particles, studied by fluorescence and electron microscopy. Z. Zellforsch. 119, 460--482 (1971) King, L. E., Fedinec, A. A. : Distribution of tetanus toxin in rat peripheral nerve trunks during the development of local tetanus. Acta neuropath. (Berl.) 24, 244--255 (1973) Klemm, I-I.: Das Perineurium als Diffusionsbarriere gegenfiber Peroxydase bei epi- und endoneuraler Applikation. Z. Zellforsch. 108, 431--445 (1970) Kristensson, K., Lyeke, E., SjSstrand, J.: Spread of Herpes simplex virus in peripheral nerves. Acta neuropath. (Ber].) 17, 44--53 (1971)

372

G. Erdmann et al.

Kristensson, K., Sj5strand, J. : Retrograde transport of protein tracer in the rabbit hypoglossal nerve during regeneration. Brain Res. 45, 175--182 (1972) Kryzhanovsky, G. N. : Tetanus. Moscow: State Publish. House "Medicine", 1966 (in Russian) Kryzhanovsky, G. N.: The neural pathway of toxin. In: Principles on tetanus, L. Eckmann, ed., pp. 155--168. Bern-Stuttgart: H. Huber 1967 Kryzhanovsky, G. N. : The mechanism of action of tetanus toxin: Effect on synaptic processes and some particular features of toxin binding by the nervous tissue. Naunyn-Schmiedeberg's Arch. Pharmacol. 276, 247--270 (1973) Kryzhanovsky, G. N., Alekseev, L. P., Rozanov, A. Ja. : Antitoxin neutralization of tetanus toxin bound with the cerebral substance. Bull. exp. Biol. IVied. (Russ.) 9, 63--65 (1970) Kryzhanovsky, G. iN., Grafova, V.N., Tumanov, V.P., Vtyurin, B. V. : Ultrastructural changes in the spinal cord in tetanus intoxication. Bull. exp. Biol. Med. (Russ.) 75, 100--105 (1973) Kryzhanovsky, G.N., Krasnova, N. M. : Intracisternal introduction of tetanus antitoxin in experimental tetanus intoxication. Bull. exp. Biol. Med. (Russ.) 5, 38--42 (1971) Kryzhanovsky, G. N., Pevnitsky, L. A., Grafova, V. N., Polgar, A. A. : Routes of entrance of tetanus toxin into the central nervous system and certain questions in the pathogenesis of experimental tetanus. IV. On the pathogenesis of ascending and descending tetanus. Bull. exp. Biol. Med. (Engl. Transl.) 52, 1370--1377 (1961a) Kryzhanovsky, G. N., Pevnitsky, L. A., Grafova, V. N., Polgar, A. A. : Routes of entrance of tetanus toxin into the central nervous system and some problems connected with the pathogenesis of experimental tetanus. II. Experiments on mice, guinea pigs, rabbits, and cats. Bull. exp. Biol. Med. (Engl. Transl.) 52, 894--898 (1961b) Lamanna, C., Carr, C. J.: The botulinal, tetanal and enterostaphylococcal toxins: A review. Clin. Pharmacol. Ther. 8, 286--332 (1967) Laurence, D. R., Webster, R. A. : Pathologic physiology, pharmacology and therapeutics of tetanus. Clin. Pharmacol. Ther. 4, 36--72 (1963) Ochs, S., Johnson, J. : Fast and slow phases of axoplasmic flow in ventral root nerve fibres. J. ~eurochem. 16, 845-853 (1969) Price, M. T.: The effects of colchicine and lumicolchicine on the rapid phase of axonal transport in the rabbit visual system. Brain Res. 77, 497--501 (1974) Ronnevi, L.-O., BystrSm, J., Eriksson, E., Ottova, L. : Quantitative distribution of tetanus toxin in peripheral nerves and central nervous system. Europ. Surg. Res. 5, 401--413 (1973) Schubert, P., Kreutzberg, G.W.: Axonal transport of adenosine and uridine derivatives and transfer to postsynaptic neurons. Brain Res. 76, 526--530 (1974) SSderfeldt, B. : The perineurium as a diffusion barrier to protein tracers. Acta neuropath. (Berl.) 27, 55--60 (1974) Stoeckel, K., Schwab, M., Thoenen, H.: Comparison between retrograde axonal transport of nerve growth factor and tetanus toxin in motor, sensory and adrenergic neurons. Brain Res. (in press, 1975) Tarlov, J. M. : Rigidity and primary motoneuron damage in tetanus. Exp. Neurol. 44, 246--254 (1974)

Intraaxonal and Extraaxonal Transport of Tetanus Toxin

373

Tarlov, J. ~I., Ling, H., Gamada, H.: Neuronal pathology in experimental local tetanus: Clinical implications. Neurology (Minneap.) 23, 580--591 (1973) WellhSner, H. It., Hensel, B., Seib, U.C.: Local tetanus in eats: Neuropharmaeokinetics of 125I-tetanus toxin. Naunyn-Schmiedeberg's Arch. Pharmaeol. 276, 375--386 (1973a) WellhSner, H. H., Seib, U. C., tIensel, B.: Local tetanus in cats: The influence of neuromuscular activity on spinal distribution of 125I-labelled tetanus toxin. Naunyn-Schmiedeberg's Arch. Pharmacol. 276, 387--394 (1973b) Wilson, L., Bamburg, J. R., Mizet, S. B., Grisham, L. M., Creswell, K. ~ . : Interaction of drugs with microtubule proteins. Fed. Proc. 33, 158--166 (1974) Wright, G. P.: The neurotoxins of clostridium botulinum and clostridium tetani. Pharmacol. Rev. 7, 413--465 (1955) Zacks, S.J., Sheff, M. F.: Pathobiological aspects of the action of tetanal toxin in the nervous system and skeletal muscle. In: Neurosciences Research, S. Ehrenpreis and O. C. Solnitzky, eds., Vol. III, pp. 210--287. New York: Academic Press 1970