R 154

Sonja Grilc, Vlasta Kuster and Robert Zorec

MODELLING OF CYTOSOLIC CALCIUM OSCILLATIONS UNDER CONSIDERATION OF THE ER TRANSMEMBRANE POTENTIAL, CYTOSOLIC Ca2+ BINDING PROTEINS AND MITOCHONDRIA

Rat melanotrophs from the pituitary pars intermedia synthesize a number

Marko Marhla, Milan Brumena,b, Stefan SchusteF, Reinhart Heinrichd

of peptide hormones deriving from pro-opiomelanocortin, which are

Numerous mathematical models of cytosolic Ca2+ oscillations have been developed. In all of these models, intracellular stores such as the ER were shown to be a necessary prerequisite for the generation of self-sustained Ca2+ oscillations. Here we include in addition, the potential difference across the ER membrane, two classes of Ca2+ binding proteins in the cytosol (the so-called signalling and buffering proteins), and the exchange of Ca2+ between the cytosol and mitochondria. Our model predictions are in good agreement with the experimental observations. Amplitudes of calcium oscillations are in the range from 0.5 to 1.0 pM and oscillation frequencies fiom some seconds to several minutes. Although the absolute changes of the cytosolic calcium concentrations are rather small, large amounts of intracellular Ca2+ are translocated across the ER membrane. This is due to the Ca2+ binding to cytosolic proteins and due to the Ca2+ uptake and release by mitochondria. Model predictions show that the Ca2+ translocation across the ER in the period of one calcium spike is large enough to cause the changes in ER transmembrane potential in a range of some millivolts. In general, we conclude that the Ca2+ binding to cytosolic proteins as well as the Ca2+ exchange between the cytosol and mitochondria reduce the cytosolic Ca2+ concentration. In addition, the mitochondria play a specific role of maintaining the constant Ca2+ amplitudes. The property of Ca2+ amplitudes to be nearly independent of oscillation frequency is likely to allow the cell to encode information almost exclusively in the frequency.

AN RT-PCR ANALYSIS OF SYNAPTOTAGMINS (I to VII) IN RAT PARS INTERMEDIA CELLS

released in a calcium-dependent manner. In neurones, release of neurotransmitter involves Synaptotagmin I, a Ca2+-binding protein of synaptic vesicles that is believed to serve as a Ca2' sensor. It is likely that synaptotagmin I plays a role in regulated exocytosis of neuroendcrine cells. However, eleven synaptotagmin isoforms have been reported in mammmals. Therefore, we investigated which of the synaptotagmin isoforms are expressed in rat melanotrophs. Using the RT-PCR analysis we looked at the expression of rat synaptotagmin isoforms I to VII. The presence of PCR products for the synaptotagmin isoforms was compared to the PCR analysis of mRNA samples obtained from hippocampus, spinal cord, skeletal muscle, liver and cerebellum. Here we report that Synaptotagmin I, IV, V and VII are abundantly expressed, whereas the signals reporting the expression of synaptotagmin I11 and VI are weak in pars intermedia cells.

Synaptotagmin I1

was not detected in rat

melanotrophs in our study. The expression of the synaptotagmin isoforms is discussed in terms of the apparent high calcium affinity of secretory apparatus of rat melanotrophs. Lab.

Neuroendocrinology-Molecular Cell

Physiolgy,

Institute

of

Pathophysiology, Medical School, pp 221 1 1104, Ljubljana, Slovenia

Wniversity of Maribor, Faculty of Education, KoroSka cesta 160, 2000 Maribor, Slovenia; bJosef Stefan Institute, Jamova 39, 1000 Ljubljana, Slovenia; CMax Delbriick Center, Robert-Rossle-Str. 10, 13125 BerlinBuch, Germany; dHumboldt University, Institute of Biology, Invaliden Str. 42, 10115 Berlin, Germany.

THE EFFECTS OF CALCIUM CHANNEL ANTAGONIST (NICARDIPINE) ON SMOOTH MUSCLE - A COMPARISON OF TWO PROTOCOLS WITH TWO DIFFERENT ISOLATED ORGANS

PROTEINS AND LIPIDS MEMBRANE FUSION

Irena JapeQ, Vladimir Ilievski, Metka V. Budihna

To elucidate stages in the dense-core vesicle (DCV) exocytotic pathway and the biochemical mechanisms involved in membrane fusion, we have reconstituted ATP-dependent, Ca2'-activated norepinephrine secretion in mechanically-permeabilized PC 12 neuroendocrine cells and in exocytosiscompetent membrane preparations docked DCVs. In spite of the fact that DCVs are docked at the plasme membrane, ATP-dependent steps are required to prime DCVs for Ca2'-triggered fusion. ATP-dependent priming requires NSF and SNAP proteins that act on SNARE proteins. SNARE proteins are essential for late stages of exocytosis where they may be regulated through interactions with synaptotagmin, a vesicle Ca2'-binding protein. In addition to essential membrane proteins, several cytosolic proteins are required for ATP-dependent priming and Ca"-triggered fusion of DCVs. A priming mechanism requiring the synthesis of PtdIns(4,5)2 was identified by the characterization of PtdIns transfer protein and PtdIns(4)P 5-kinase as cytosolic priming factor. A cytosolic factor for late triggering steps, CAPS, is novel PH domain-containing protein that may function by binding DCV and plasma membranes as well as PtdIns(4,5)2. CAPS is specifically required for regulated DCV exocytosis but not regulated synaptic vesicle exocytosis. Hence, the late stages of DCV exocytosis require parallel ATP- and Ca2'-dependent reactions catalyzed by cytosolic factors that operate on integral proteins and phospholipids.

For the in vitro studies of the effects of calcium channel antagonists (CCA) on smooth muscles, different isolated organs and different protocols with different ways of calculation of the potencies are used. In the present study we compared the effects of three isomeric forms (S(+), R-(-) enantiomer and racemic mixture) of nicardipine (NC), a CCA, on two different isolated organs using two different protocols. For the first protocol we used the isolated guinea-pig taenia caeci contracted with cumulatively added CaC12 (0.1 - 100 mM) in a solution containing 40 mM KCl. As a measure of potency we used pD$. For the second protocol we used the isolated porcine coronary artery contracted with cumulatively added KCl (5 - 90 mM) in a solution containing 2.5 mM CaCl2. As a measure of potency we used Ic50 (calculated at 60 mM KCl). All three forms of NC concentration-dependently (10 nM - 0.1 pM) inhibited the contractions of taenia caeci induced by CaC12. pD2' for S-(+), R-(-) enantiomer and for racemic mixture of NC were 8.1, 7.9 and 8.0, respectively. All three forms of NC concentration-dependently (10 pM-0.1 mM) also inhibited the contractions of the isolated porcine coronary artery induced by KCl. ICS0 (calculated at 60 mM KCI) for S-(+), R-(-) enantiomer and for racemic mixture of NC were 6.6 nM, 31.8 nM and 10.9 nM, respectively. The order of potencies of NC forms was the same in the two models. S(+) enantiomer was practically as potent as R-(-) enantiomer in inhibiting the contractions of taenia caeci induced by CaC12 and 4.8 times more potent in inhibiting the contractions of the isolated porcine coronary artery induced by KC1. The two protocols, thus, give the same order of the potencies, although with significantly higher potency ratio in the protocol with isolated porcine coronary artery. Dept. of Pharmacol. & Exp. Toxicol., Medical Faculty, University of Ljubljana, Korytkova 2, 1000 Ljubljana, Slovenia; [email protected]

REQUIRED

FOR

REGULATED

T. F. J.Martin

Dept. of Biochemistry, University of Wisconsin, Madison, W1 53706

R 155

SPATIO-TEMPORAL ASPECTS OF CA2+SIGNALLING: A STUDY AT THE SUBCELLULAR LEVEL USING SELECTIVELY TARGETED RECOMBINANT AEQUORINS

MODULATION OF UNITARY EXOCYTOTIC EVENTS BY CYCLIC AMP Sujit K.Sikdar*, Marko Kreft and Robert Zorec

Tullio Pozzan Over the last years overwhelming evidence has accumulated indicating the importance of the spatio-temporal aspects in the signalling pathway that depends on Ca” changes within the cytoplasm. Typical examples are the localised Ca2’ increases controlling secretion and ion channel activity and the selective modulation of gene expression by specific patterns of Ca” oscillations. Methodologies that allow direct monitoring in intact living cells of Ca” concentrations in selected cytoplasmic regions or within organelles represent key tools for unraveling these important characteristics of the Ca2’ signalling code. In this presentation I discuss the approach we have undertaken for studying Ca2+ homeostasis in selected part of the cells. The procedure is based on the recombinant expression of chimeric Ca” sensitive aequorins, selectively targeted to specific cellular compartments. Using similar strategies we have also constructed chimeric proteins of GFP mutants, with different spectral properties, to follow in parallel the physical interaction between organelles in living cells. I will briefly review the rationale of the methodology and address a few general concepts that emerged from this approach. I will then concentrate on some recent results obtained with aequorin and GFP chimeras, targeted to the outer surface of the inner mitochondrial membrane, to the matrix of this organelle and to the lumen of endoplasmic reticulum. I will present data indicating that mitochondria in living cells are to a large extent represented by a lumenally continuous tubular network that undergoes continuous remodeling; ii) close apposition of mitochondria and endoplasmic reticulum are often observed; iii) upon generation of IP3 by plasma membrane receptor stimulation the mitochondrial surface is transiently exposed to a Ca” concentration higher than that of the rest of the cytosol. These results provide conclusive evidence in favor of the hypothesis that the rapid Ca” uptake by mitochondria and the activation of the key dehydrogenases in their matrix is dependent on localized microdomains of high Ca” concentration generated during cell activation in close proximity of IPSgated channels. Dept. Biomedical Sciences, University of Padova, Via G. Colombo 3, 35121 Padova, Italy. Email: [email protected]

BOTULINUM NEUROTOXINS AND FAST Ca2+DEPENDENT SECRETORY ACTMTY IN SINGLE RAT MELANOTROPHS M Rupnik’, M Kreftl, SK Sikda?, TFJ Martin3and R Zorec’

Flash photolysis of caged intracellular calcium (M-EGTA) in single rat melanotrophs results in a biphasic secretory response, exhibiting a fast exocytic burst and a slower component of exocytosis (Thomas et al., 1993) suggesting at least two distinct functional states of secretory vesicles. The molecular machinery underlying this late dockinghsion even& is believed to be built around a complex of in+racellularproteins termed the SNARE complex (Sollner et al., Nature, 1993, 362:318; Rothman and Warren, Curr Biol, 1994, 4:220, Weber et al, Cell, 1998, 92759-772). Tbe sensitivity of SNARE proteins (SynaptobrevinNAMP, SNAP-25, syntaxin) to botuhm neurotoxins makes these toxins powerfultools for dissecting the machinery mediating neuroendocrine secretion. The aim of our study was to investigate the effects of hotulinum toxin types A and E (cleaving SNAP25) and B (cleaving SynaptobreviuWAMP)on secretory response of single rat melanotrophs triggered by flash photolyss. We combined the whole-cell patch-clamp to monitor fast changes in membrane capacitance (C+ a measure of membrane surface area) and Furaptra based measurements of cytosolic Ca’ changes induced by the flash photolysis. Cells were isolated to a primary culture (Rupnik & Zorec, FEBS Letts, 1992, 303:221) and neurotoxins were microinjected (-7 ng/Fl) together with a dextran-rhodamin marker. Identified cells were examined electrophysiologically 1 to 3 hours after incubation at 37°C. Both kinetic components of Catdependent exocytosis were blocked by hotulinum toxins B and E, whereas toxin A did not affect secretory responses under these conditions. Our results indicate that SNARES (synaptohrevin, SNAP-25, syntaxin) play a role in the regulation of both Ca“ dependent components of secretion in rat melanotrophs. In contrast to results of the similar study (Xu et al, Nature Neurosci, 1998, 1; 192-299) the extent ofblockage by B and E t o m depended on the time of incubation of injected cells with complete blockage obtained 1-3 hours after injection. This suggests that the assembly/dissasemhly of the SNARE complex IS not a fast process, which is consistent with the biochemical stability of the SNARE complexes in virro (Hayashi et al, EMBO J, 1995, 14:2317-25). The work was supported by the project J3-6027-381 by the Ministry of Science and Technology of Slovenia and Joint US-SLO project 95-471. ‘Laboratory of Neuroendocrinology-MolecularCell Physiology, Institute of Pathophvsioloay, School of Medicine, University of Ljubljana, 1000 Ljubljana, Slovenia, 2MolecuiarBiGhysics Unit, Indian Institute of Science; B&galore-560012, India, 3Department of Biochemistry, University of Wisconsin, Madison, Wisconsin 53706, USA

Calcium dependent secretion of POMC derived hormones (such as Qmelanocyte stimulating hormone or P-endorphin) from rat pituitary pars intermedia melanotrophs is enhanced by cyclic AMP (CAMP). Using whole-cell patch-clamp membrane capacitance (C,) measurements we monitored calcium-dependent secretory activity of single melanotrophs in the presence and absence of CAMP. Cytosol dialysis with a patch-pipette solution containing 450 nM [Ca”] and CAMP resulted in a 3 fold increase in membrane capacitance in comparison to controls. Similar results were obtained previously on melanotrophs and lactotrophs, indicating that CAMP may affect the exocytic machinery. To investigate whether CAMP directly modulates the secretory apparatus, CAMP was dialyzed into cultured rat melanotrophs and the interaction of a single secretory granule with the plasmalema was monitored by high-resolution membrane capacitance measurements. Changes in membrane capacitance reflect changes in membrane area, which fluctuate due to exo- and endocytosis. The signal-tonoise ratio of recordings was improved by preincubation of cells with a hydrophobic anion dipictylamine. We found that the amplitude of discrete “on” steps in C,, a parameter correlated to the size of exocytosing secretory granules, increases in the presence of CAMP. On the basis of this result we propose that CAMP-dependent mechanisms mediate a differential fusion of larger granules with the plasmalemma, a process physiologically relevant since the larger granules contain fully processed peptide hormones. CAMP may selectively mediate the fusion of larger granules or/and may indirectly stimulate the growth of secretory granules by stimulating intergranule fusion (detected by electron microscopy in these cells), hence increasing the population of larger secretory granules. Inst. of Lab. Neuroendocrinology-Molecular Cell Physiology, Pathophysiology, Medical School, pp 221 1, 1001 Ljubljana, Slovenija, and *Mol. Biophysics Unit, Indian Inst. of Science, Bangalore-560012, India.

CALCIUM HOMEOSTASIS AND SECRETION FROM NEURONS

SM Smith’ z, Y Namkung3, D Foletti’ *, J KlingauP, HS Shin3, RH Scheller’ z, RW Tsien’ Voltage dependent calcium channels open in response to membrane depolarization and allow the passage of Ca” into the cell without significant flux of any other ion. The resulting increase in [Ca”] triggers vesicle fusion, which leads to the release of neurotransmitter. In contrast to the well-characterized role of the main subunit of voltage-gated Ca” channels, the pore-forming, antagonist-binding alpha1 subunit, less is known about how beta subunits contribute to neuronal Ca2’ channel function. We used a gene targeting strategy to study the role of the Ca2+ channel beta3 subunit, the major Ca2’ channel beta subunit in neurons. The beta3 deficient (ccb3-/-) animals were indistinguishable from the wild type (wt). However, in sympathetic ccb3-/- neurons, the N-type Ca” channels made up a smaller proportion of the total Ca2’ current than in the wt. This difference was due to a lower N-type current density. Norepinephrine inhibited wt and ccb3-/- neurons similarly but the voltage sensitive component was greater for N-type than P/Q-type Ca” channels. Activation curves of P/Q-type Ca2’ channels were described by two Boltzmann components with different voltage-dependence, analogous to the “reluctant” and ”willing“ states reported for N-type channels. The absence of the b3 subunit was associated with a hyperpolarizing shift of the “reluctant” component of activation. The reduction in the expression of Ntype Ca” current in the b3-/- mice may be expected to impair Ca” entry and therefore synaptic transmission in these animals. This effect may be of P/Q channels reversed, at least in part, by the increase in the proportion - . activated at more hyperpolarizing voltages. Department of Molecular and Cellular Physiology’& Howard Hughes Medical Institute2, Stanford University, Palo Alto CA 94305-5345, USA and Department of Life Science‘, Pohang University of Science & Technology, San-3 1, Hyoja-Dong, Pohang, 790-784, Republic of Korea

R 156

A NEW APPROACH COMBINING C, MEASUREMENTS, NPEGTA FLASH PHOTOLYSIS AND [ca2+li MEASUREMENTS IN SINGLE RAT MELANOTROPHS Gregor ZupanEiE’ and Robert Zorecz The goal of our research is understanding the processes involved in fast exocytosis occuring in neurons and neuroendocrine cells l i e melanotropbs. One way of monitoring exocytosis is by membrane capacitance (C,) measurments. Due to very fast nature of the process this requires very high time (1 ms) and amplitude (1-10 E)resolution. In addition it requires triggering of exocytosisby rapid [Ca“], elevation by 1 to 2 orders of magnitude, usually achieved by flash photolysis of caged calcium (Ca-NP-EGTA). This in tum requires a very higb energy UV flash. In addition [CaTi has to he measured simultaneously at high time resolution over a very wide magnitude range to allow subsequent analysis of experimental data. To achieve these goals we devised a new approach to do these experiments. First we increased amplitude and time resolution of C, records. The amplitude resolution was increased by increasing the amplitude of the sine voltage, used in C, measurements, from usual 1-10 mV, to 111 mV,. This meant a proportional increase in the signal-to-noiseratio but caused problems with time resolution. The time resolution limit in C, measurements is limited by the frequency of the sine voltage as well as it’s amplitude. In our experiments on single rat melanotrophs we used a frequency of 1.6 kHz.To achieve desired time resolution of 1 kHz we had to use very high oversampling (50 liHz) and very high order (300) FIR digital filtering followed by downsampling.In this way we were able to get almost ideal DC to 1% signal bandwidth. Secondly we increased UV flash intensity, to achieve sufficient NP-EGTA photolysis and subsequent increase in [Ca”li, by focusing the flash light from a HI-TECH” Xe flashlamp through the x40 fluor 1.4 NA oil immersion objective, using the epifluorescent illumination of the Nikon Diaphot microscope. We thus achieved an estimated 1200 1500 d / c m 2 energy flux (within bandwidth of 340-390 nm), enough to photolyze estimated 55 - 70% of all Np-EGTA molecules in the focus with a single flash. Thirdly we simultaneously measured [Ca”li with high time and amplitude resolution. We used a fluorescent Ca” dye furaptra and excited it at 420 nm according to the method by Carter and Ogden (1994, Pfliigers Archiv, 428: 476-484). We chose fiuaptra because of it’s high dynamic range, it’s 48 pM Ca” & and because it allows single wavelength excitation. The flash and the furaptra excitation light were combined by a 390 nm dichroic filter and then sent through the objective via 430 nm dichroic. In this way we were able to measure [CaZA]; in the range from 1 to 100 pM with 5 ms time resolution using a photon counting photomultiplier system. We conclude that the performance of this approach is better than with similar systems used so far (Thomas et al. 1993, EMBO J. 12: 303-306; Heinemann et al. 1994, Biophys. J. 67: 2546-2557; Parsons et al. 1996, Cell Calcium 19: 185-192; Kasai et al. 1996, J Physiol. 494 53-65). University of Ljubljana, ‘Biotechnical Faculty, Department of Biology, VeEna pot 11 1, Ljubljana and z’Iustitute of Pathophysiology, Medical School, ZaloSka 4, Ljubljana,Dept. of Pharmacol. & Exp. Toxicol., Medical Faculty, University of Ljubljana