zyxwvutsrqponml zyxwvutsrqponm zyxwvutsrqpo Journal of Neurochemistry Lippincott Williams & Wilkins, Inc., Philadelphia 0 1999 International Society for Neurochemistry A Nitric OxideKyclic GMP-Dependent Protein Kinase Pathway Alters Transmitter Release and Inhibition by Somatostatin at a Site Downstream of Calcium Entry zyxwvutsrqpon zyxwv zyxwv D. Bruce Gray, *Luis Polo-Parada, *Guillermo R. Pilar, Peau Eang, ?Ryan R. Metzger, ?Eric Klann, and ?Stephen D. Meriney Department of Biology, Simmons College, Boston, Massachusetts; *Department of Neuroscience, Case Western Reserve University School of Medicine, Cleveland, Ohio; and ?Department of Neuroscience, University of Pittsburgh, Pittsburgh, Pennsylvania, U.S.A. Abstract: We have examined the somatostatin-mediated modulation of acetylcholine release from intact chick embryo choroid tissue and compared these data with those obtained using acutely dissociated neuronal cell bodies from the chick ciliary ganglion. Acetylcholine release, evoked in a calcium-dependent manner by a high potassium (55 mM KCI) stimulation in both preparations, was inhibited almost completely by 100 nM somatostatin. Measurement of intracellular calcium in these neurons revealed that somatostatin blocked the large calcium transient that was observed in control neurons following KCI exposure. The modulatory effect of somatostatin on transmitter release was significantly attenuated by pretreatment with pharmacologic agents that selectively block cyclic GMP (cGMP)-dependent protein kinase (PKG) or nitric oxide (NO) synthase. It is interesting that this prevention of somatostatin-mediated acetylcholine release inhibition occurred without reversal of the somatostatin-mediated block of the KCI-evoked calcium transient. Furthermore, a NO donor or cGMP analogue could block KCI-evoked acetylcholine release, but only cGMP could reduce the KCI-evoked calcium transient. Although cGMP could reduce the KCI-evoked calcium transient, a cGMP analogue was shown to reduce calcium ionophore-evoked transmitter release. Thus, somatostatin reduces acetylcholine release by modulating calcium influx, but the NO-PKG pathway can inhibit acetylcholine release, and alter somatostatin-mediated inhibition, by affecting transmitter release at some point after calcium entry. Key Words: Acetylcholine-Calcium-Cyclic GMP-dependent protein kinase-Nitric oxide. J. Neurochem. 72, 1981-1 990 (1999). Saggau, 1997; Dolphin, 1998). Previously, Gray et al. (1989, 1990) studied somatostatin-mediated modulation of acetylcholine (ACh) release at the junction of parasympathetic choroid neurons with vascular smooth muscle in the choroid coat isolated from the hatchling chick eye. These studies led to the hypothesis that somatostatin-mediated G protein activation causes a decrease in presynaptic Ca2+ current resulting in decreased ACh release. Electrophysiological recordings from the soma of these neurons in culture confirmed that somatostatin strongly inhibited Ca2+ current in a pertussis toxinsensitive manner (Dryer et al., 1991; Meriney et al., 1994). Furthermore, one of these previous reports implicated a cytoplasmic cyclic GMP (cGMP)-dependentprotein kinase (PKG) pathway that altered the characteristics of the membrane-delimited somatostatin-mediated inhibition of Ca2+ current (Meriney et al., 1994). In this report, we demonstrate that the PKG pathway interacts with the somatostatin-mediated modulation of ACh release from ciliary ganglion neurons and hypothesize that this occurs at a site after Ca2+ entry. zyxwvut zyxwvutsrq zy Received September 25, 1998; revised manuscript received December 16, 1998; accepted December 18, 1998. Address correspondence and reprint requests to Dr. D. B. Gray at Department of Biology, Simmons College, 300 The Fenway, Boston, MA 021 15, U.S.A. The present address of Dr. R. R. Metzger is Department of Pharmacology and Toxicology, 112 Skaggs Hall, University of Utah, Salt Lake City, UT 84112, U.S.A. Abbreviations used: ACh, acetylcholine; 8-Br-cGMP, 8-bromoguanosine-3',5'-cyclic monophosphate; cGMP, cyclic GMP; DMSO, dimethyl sulfoxide; NAME, N"-nitroarginine methyl ester; L-NMMA, L-NG-monomethylarginine; NO, nitric oxide; g-pCPT-cGMP, 8-(4chloropheny1thio)guanosine-3',5'-cyclic monophosphate; PKG, cyclic GMP-dependent protein kinase; Rp-8-Br-PET-cGMPS. 0-phenyl1,N2-etheno-X-bromoguanosine-3',5'-cyclic monophosphorothioate, Rp-isomer; Rp-8-pCPT-cGMPS, 8-(4-~hlorophenylthio)guanosine3 ' 3 '-cyclic monophosphorothioate, Rp-isomer; SNP, sodium nitroprusside; ST,stage. Neuropeptide-mediated modulation of transmitter release occurs via G protein-coupled signal transduction pathways that alter the gating of calcium (Ca2+) channels. In many cases, this inhibition appears to be membrane-delimited and not to require the participation of any known protein kinases or other soluble intracellular components (for reviews, see Hille, 1992, 1994; Wu and 1981 I982 zyxwvutsrqp zyxwv D. B. GRAY ET AL. The subpopulation of ciliary ganglion neurons that express both ACh and somatostatin have been identified as the choroid neurons that innervate the vascular bed of the choroid coat in the chick eye (Epstein e t al., 1988; Gray et al., 1989, 1990; De Stefan0 et al., 1993). Because nitric oxide (NO) is known to be a potent modulator of vascular tone (for review, see Garthwaite and Boulton, 1995), and N O synthase has been localized to ciliary ganglion neurons (Nichol et al., 1995), we were prompted to investigate a hypothesized role for NO and PKG in the somatostatin modulation of evoked ACh release from choroid nerve terminals onto vascular smooth muscle. NO has been shown to modulate evoked transmitter release in other preparations; however, the mechanisms of action are unknown (Kilbinger, 1996). In these and previous experiments addressing the effects of somatostatin on ACh release from hatchling animals, intact choroid nerve terminal preparations have been used. It is difficult to compare these data directly with the modulation of Ca” currents in acutely dissociated embryonic ciliary ganglion cell somata (Meriney et al., 1994; Pilar e t al., 1996). Therefore, w e have extended previous observations from intact nerve terminals of hatchling chickens (Gray et al., 1989, 1990) to an embryonic age [stage (ST) 401 at which we could perform experiments with both intact nerve terminals and dissociated neuronal somata using freshly dissociated ciliary ganglion neurons in vitro to study somatostatin-mediated modulation of ACh release. By comparing potassiumevoked ACh release with photometric measurements of Ca” influx in acutely isolated neurons, we have hypothesized that a cytoplasmic NO-PKG pathway directly affects Ca2+-dependent secretion downstream of Ca2+ entry. Furthermore, the effects of this NO-PKG pathway on secretion interact with the functional effect on secretion of G protein-mediated modulation of Ca2+ channels. A preliminary account of some of these findings has been reported (Pilar et al., 1996). MATERIALS AND METHODS Preparation of tissue polyomithine solution in sodium borate, dried overnight, and washed in distilled water. This enzymatic approach (20-min incubation in 0.08% trypsin) is optimized for the isolation of the ciliary ganglion cell bodies, without regard for the preganglionic nerve terminals that appear to be destroyed during trituration (G. R. Pilar, unpublished observations). For ACh release experiments, neuronal density averaged 5,000-10,000 per well (1.5 cm’). Experiments were performed 3-6 h after the neurons were plated. zyxw Determinations of ACh release Choroid tissue wedges or dissociated ciliary ganglion neurons were preincubated in oxygenated Tyrode’s solution containing [3H]choline (final specific activity, 8.1 Ci/mol; total choline concentration, 1 pkl) for 70 min and washed 10-15 times by centrifugation in zero Ca’+ Tyrode’s solution to remove extracellular label (Gray et al., 1989, 1990). Prestimulation or basal release of [3H]choline was measured in 5-min collection periods in normal Tyrode’s solution and then a second basal 5-min collection period with or without drug additions. Drugs or peptides were always added in the last 5-min basal collection period before potassium stimulation to allow sufficient time for equilibration either with presynaptic endings within the intact choroid tissue or with dissociated cells (Gray et al., 1989). To evoke ACh release, tissues were incubated in 55 mM KCl Tyrode’s solution (equimolar substitution for NaCl) with or without somatostatin for 3-5-min collection periods (as cited in figure legends). Release values were converted to disintegrations per minute (dpm) for each collection period and expressed as percent stimulation over basal release where basal release is defined as zero (see Figs. 1-3). [,H]ACh was separated from [3H]choline by high-voltage paper electrophoresis and measured by scintillation counting (Vaca and Pilar, 1979; Gray et al., 1989). All experiments were performed at 37°C. zyxwvut zyxwv zyxwvuts For measurements from intact choroid tissue, eyes were removed from 14-day (ST 40; Hamburger and Hamilton, 1951) chick embryos in oxygenated avian Tyrode’s solution (134 mM NaCI, 3 mM KCl, 20.5 mM NaHCO,, 3 mM CaCl,, 1 mM MgCl,, 12 mM glucose, pH 7.2). A wedge-shaped section of choroid tissue, bounded by the ciliary nerves on one side and the ciliary body surrounding the iris on the other, was removed intact. The vitreous humor and retina were peeled away from the interior surface of the wedge and discarded. The choroid layer, containing a high density of cholinergic ciliary ganglion presynaptic varicosities along with the pigment epithelium layer, was removed from the underlying sclera and maintained in oxygenated avian Tyrode’s solution. For measurements from dissociated cells, ciliary ganglia were removed from 14-day (ST 40) chick embryos, dissociated (as described by Crean et al., 1982; Gray and Tuttle, 1987), and plated with Dulbecco’s hybridoma medium plus 10% chick embryo extract (modified from Crean et al., 1982) onto 24-well Linbro tissue culture plates that had been coated with 10% Measurements of cytosolic Ca2+ Ciliary ganglion neurons were isolated as described above and plated onto polyomithine-coated 35-mm dishes. Intracellular Ca’+ was measured in individual neurons using the Ca2+sensitive dye fura-2. Neurons were loaded with fura-2 by adding 2.5 pA4 of the acetoxymethyl ester form of the dye to the culture medium (without serum) for 30 min at 37°C. Fluorescence changes were detected using photometry on an RM-R system (Photon Technology International, Princeton, NJ, U.S.A.), and cytosolic Ca’+ was measured using the ratio of the emission detected following excitation for 50 ms at two wavelengths (340 and 380 nm) alternating at 6 Hz. Fura-2 solutions were calibrated using known concentrations of Ca’+ and fits to the curve described by the Grynkiewicz equation (Grynkiewicz et al., 1985). Neurons were bathed in normal Tyrode’s saline, and experimental solutions and reagents were added directly to the cell under study by using a large diameter (100-200 pm) pipette. The analog output from the photometer was routed through an A/D converter (LabMaster, Scientific Solutions, Solon, OH, U.S.A.) to a Pentium-based computer for analysis using a custom program. cGMP radioimmunoassay Ciliary ganglion neurons were isolated as described above, except that cells were plated onto polyornithine-coated 24-well culture plates at a higher density (10,000-15,000 per well). After a 3-6-h incubation, the cells were bathed in normal chick Tyrode’s solution in the presence of the phosphodiesterase inhibitor 3-isobutyl-1-methylxanthine(1 mM). After a 1-h in- zyxwvutsrqpo J. Neurochem., Vol. 72, No. 5, 1999 zyxwvu zyxwvutsr zy zyxwvutsrqpo NO-PKG ALTERATION OF SOMATOSTATIN INHIBITION I983 zyxwvutsrq zyxwvutsrqp zyxwvutsrq FIG. 1. Pharmacological blockade of PKG A 2007 or NO synthase reverses the somatostatin1501 r mediated inhibition of ACh release from choroid nerve terminals isolated from the ST 40 embryonic chick eye. A ACh release expressed as a % increase in 3H-labeled ACh release over basal levels when evoked with a 3-min 55 mM potassium stimulation. Control tissue is compared with tissue treated with 100 nM somatostatin (SOM), the PKG inhibitor 10 pA4 Rp-8-pCPTcGMPS (Rp-cGMPS), and a coapplication Cont SOM SOM+ RpcGMPS con1 cGMP SNP of both somatostatin and Rp-8-pCPTRo-CGMPS cGMPS (SOM + Rp-cGMPS). B Same experimental protocol as in A, except that the B 2ooj T active NO synthase inhibitor 10 pM L-NAME or inactive enantiomer 10 pM D-NAMEwas used. C: Same experimental protocol as in A, except that the PKG agonist 8-Br-cGMP (cGMP, 100 clnn) or the NO donor SNP (300 clnn) was used. *Significantly different from control ( p < 0.05) using a one-way analysis of variance with Dunnett’s test. D Doseresponse curve for the effects of hemoglobin (Hb) on somatostatin-mediated inhibi0 500 1000 tion of ACh release. All points represent the Con1 SOM SOMI SOM+ L-NAME Hb concentration(pM) L-NAME D-NAME mean effect of somatostatin plus varying concentrations (0-1,000 LL~M)of Hb. The mean level of control potassium-evoked release in these experiments is represented by the solid line. At concentrations of 100 f l or greater, Hb reversed the somatostatin inhibition of ACh release by -50%. All values are means t SEM with n 2 4. C T zyxw zyxw cubation in this solution, experimental wells were treated with varying doses of somatostatin (1, 10, or 100 nM) or 300 pM sodium nitroprusside (SNP) for 1 min. Incubations were terminated by adding trichloroacetic acid (at a final concentration of 6%). cGMP concentrations were determined by previously published procedures, with minor modifications (Chetkovich et al., 1993). In brief, cellular material was harvested and centrifuged at 4°C for 15 min to separate the soluble and particulate fractions. The particulate fraction was set aside for protein determinations, and the supernatant fraction from different experiments was kept frozen at -80°C for cGMP determination at a later date. Harvested material from each experiment was assayed with an ‘251-labeled cGMP radioimmunoassay kit (New England Nuclear). All cGMP measurements were performed using the same kit to avoid variations inherent to the technique. Aliquots of the particulate fraction were processed for protein determination using the Bradford method, and the cGMP levels were normalized using these determinations for each sample. Materials log (La Jolla, CA, U.S.A.) and solubilized in Tyrode’s solution. Fura-2 acetoxymethyl ester (Molecular Probes, Eugene, OR, U.S.A.) stock solutions (1 mM) were made in DMSO and diluted in Tyrode’s solution to a final concentration of 2.5 p M (0.25% DMSO). Somatostatin (Sigma) was dissolved directly into Tyrode’s solution to a final concentration of 100 nM. All other reagents were obtained from Sigma and dissolved in Tyrode’s solution. RESULTS Somatostatin-mediatedmodulation of ACh release from intact nerve terminals Evoked release of labeled ACh was measured from intact ST 40 choroid tissue, which contains axons, synaptic terminals, and vascular target tissue. Following the washout of the extracellular [3H]ACh that had accumulated in the choroid tissue during loading, a 3-min exposure to 55 mM KC1 Tyrode’s saline elicited an increase (100-180%) in [‘HIACh release (Fig. 1). This potassium-evoked release has been shown previously to be &*+-dependent (Gray et al., 1992). Incubation with 100 nM somatostatin blocked the evoked release of [3H]ACh (Fig. 1A and B). The effects of inhibitors of the NO-PKG signal transduction system were tested in ST 40 intact choroid tissue. An inhibitor of PKG (Rp-8-pCPT-cGMPS) reversed the effect of somatostatin on ACh release (Fig. 1A), but Rp-8-pCPT-cGMPS had no effect on control evoked release. In another set of experiments, the somatostatinmediated inhibition of ACh release was also reversed by an inhibitor of NO synthase (L-NAME), but not by the inactive enantiomer D-NAME(Fig. IB). When adminis- zyxwvutsrqp Protein kinase inhibitors KT5823, KT5720, and calphostin C were purchased from Calbiochem (San Diego, CA, U.S.A.) and solubilized in dimethyl sulfoxide (DMSO; final vehicle conmonophoscentration, 0.1%). 8-Bromoguanosine-3’,5’-cyclic phate (8-Br-cGMP) was purchased from Sigma (St. Louis, MO, U.S.A.) and solubilized in DMSO as described above. 3-Isobutyl- 1-methylxanthine, L- and D-N”-nitroarginine methyl ester (L- and D-NAME), and L-NG-monomethylarginine (L-NMMA) were obtained from Sigma and solubilized in ethanol (final vehicle concentration, 0.1%). 8-(4-Chlorophenylthio)guanosine-3’,5’-cyclic monophosphate (8-pCPT-cGMP), 8-(4chlorophenylthio)guanosine-3’,5’-cyclic monophosphorothioate, Rp isomer (Rp-8-pCPT-cGMPS), and P-phenyl-1,N2etheno-8-bromoguanosine-3’,5’-cyclicmonophosphorothioate, Rp isomer (Rp-8-Br-PET-cGMPS) were purchased from Bio- zy J. Neurochem., Vol. 72, No. 5, 1999 1984 zyxwv zyxwvutsrqpo zyxwvutsrqpo D. B. GRAY ET AL. zyxwvutsrq zyxwvutsrq zyxwvutsrqp T FIG. 2. Pharmacological blockade of PKG or NO synthase relsol verses the somatostatin-mediated inhibition of ACh release 20001 from acutely dissociated ST 40 ciliary ganglion neurons. A Repg1500]\ E resentativewashout of extracellu55 mM KCI lar PH]ACh in superfusate following a 30-min incubation in [3H]choline followed by stimulation of PH]ACh release using 55 mM potassium (horizontal bar). Open circles represent release in Cont SOM SOM+ SOM+ L-NAME 0 5 10 15 20 25 zero Ca2+ Tyrode’s solution, and L-NAME D-NAME Time (min) filled circles represent release in Tyrode’s solution containing 3 mM Ca2+. 6: ACh release expressed as a % increase in 3Hlabeled ACh release over basal levels when evoked with a 5-min 55 mM potassium stimulation. Control tissue is compared with tissue treated with 100 nM somatostatin (SOM), the PKG inhibitor 10 f l Rp-8-pCPT-cGMPS (RpcGMPS), and a coapplication of both somatostatin and Rp-8pCPT-cGMPS (SOM + RpCon1 cGMP SNP Con1 SOM SOM+ Rp-cGMPS cGMPS). C: Same experimental Rp-cGMPS protocol as in B, except that the active NO synthase inhibitor 10 f l L-NAME or the inactive enantiomer 10 pM D-NAME was used. D: Same experimental protocol as in 6,except that the PKG agonist 10 f l 8-pCPT-cGMP (cGMP) or the NO donor 300 f l SNP was used. All values are means 2 SEM with n 2 4. *Significantly different from control ( p < 0.05) using a one-way analysis of variance with Dunnett’s test. c T zyxwvutsr zyx tered alone, L-NAME had no effect on control evoked ACh release. These data suggest that a NO-PKG pathway interacts with the somatostatin-mediatedmodulation of ACh release in intact nerve terminals. It is interesting that although inhibitors of NO synthase or PKG had no effect on evoked ACh release when applied alone, an NO generator (SNP) or an analogue of cGMP (8-Br-cGMP) reduced potassium-evoked ACh release significantly (Fig. IC). Thus, activators of the NO-PKG pathway can, themselves, inhibit transmitter release, whereas inhibitors of the NO-PKG pathway can reverse somatostatinmediated inhibition of transmitter release. These data suggest that the NO-PKG pathway directly affects the transmitter release process and interacts with somatostatin-mediated effects on transmitter release. We tested the possibility that the endogenous NO necessary for maintaining the activity of the NO-PKG pathway was derived in part through transmembrane cell-cell diffusion within the intact choroid tissue. NO permeates cell membranes and is often able to diffuse between cells in a local environment. The source of this NO could include choroid nerve terminals and the vascular tissue of the choroid coat. Oxyhemoglobin, a relatively large protein that remains extracellular when added exogenously, is known to bind NO with high affinity and thus buffer diffusible NO effects. To test if NO that may have been produced in the choroid tissue had transcellular effects, hemoglobin was added to the incubation medium in concentrations ranging from 50 to 1,000 pM and somatostatin-mediated inhibition of ACh release was measured. Maximal effects were observed using 200 pkl hemoglobin, which reversed -50% of the somatostatin-mediated inhibition of ACh release (Fig. 1D). Thus, NO derived from extracellular tissue surrounding choroid nerve terminals contributed partially to the maintenance of the activity of the endogenous NO-PKG pathway. Somatostatin modulation of ACh release from acutely dissociated somata in vitro Previously, the effect of somatostatin on ACh release from intact choroid nerve terminals was hypothesized to be due to the modulation of the Ca2+ channels that regulate ACh release (Gray et al., 1989, 1990). As Ca2+ influx is difficult to study directly in intact choroid nerve terminals, direct comparisons between ACh release measurements and Ca2+ influx are performed more easily in a model system. To extend ACh release measurements from intact choroid nerve terminals (Gray et al., 1989, 1990, 1992; Fig. 1) to a preparation with which one could compare ACh release data with somatostatin modulation of Ca2’ influx, KC1-evoked release of [3H]ACh was measured from acutely isolated ST 40 ciliary ganglion neuron cell bodies. Figure 2A shows a characteristic release profile of [3H]ACh from dissociated ST 40 ciliary ganglion neurons. As in intact choroid, the elevated release during initial washes is thought to be due to synthesis and release of [3H]ACh into extracellular spaces during the 70-min incubation with [3H]choline. After the level of [3H]ACh reached a plateau, exposure zyxwvutsr J. Neurochem., Vol. 72, No. 5, 1999 zyxwvuts zyxw zyxwvut zyxwvuts zyxwvut NO-PKG ALTERATION OF SOMATOSTATIN INHIBITION to 55 mM KC1 Tyrode’s saline for 5 min elicited a oneto twofold increase in [3H]ACh release. Exposure to zero Ca2+ Tyrode’s solution decreased basal release slightly and blocked KC1-evoked [3H]ACh release (Fig. 2A). As with intact choroid tissue, incubation with 100 nM somatostatin blocked the evoked release of [3H]ACh (Fig. 2B and C). Furthermore, an inhibitor of PKG (Rp-8pCPT-cGMPS) had no effect on evoked release when presented alone, but could reverse the effect of somatostatin (Fig. 2B). In another set of experiments, an inhibitor of NO synthase (L-NAME),but not the inactive enantiomer (D-NAME),reversed the effect of somatostatin (Fig. 2C). As described above for intact choroid nerve terminals, an NO generator (SNP) or an analogue of cGMP (8-pCPT-cGMP) reduced significantly potassiumevoked ACh release (Fig. 2D), whereas inhibitors of NO synthase or PKG had no effect on evoked ACh release when applied alone. To rule out a complex cascade involving protein kinase C or cyclic AMP-dependent protein kinase (Nestler and Greengard, 1984) and provide stronger evidence that PKG was involved in the somatostatin modulation of ACh release, the effects of three different selective protein kinase inhibitors were tested. The selective PKG inhibitor KT5823 (at 0.2-1 pM) had no effect on control release, but could reverse the effect of somatostatin (Fig. 3A). Inhibitors selective for protein kinase C (1 pM calphostin C) and cyclic AMP-dependent protein kinase (1 pM KT5720) were without effect on control release and had no significant effect on somatostatin-mediated modulation of ACh release (Fig. 3B and C). The results obtained using KT5823 are similar to the effects of Rp-8-pCPT-cGMPS reported in Figs. 1A and 2B. Thus, somatostatin modulation of ACh release from ciliary ganglion cell bodies is modulated by NO and PKG, as has been shown for intact choroid nerve terminals. Cont SOM 1985 SOM+ KT5823 KT5823 B T140 T Con! SOM SOM+ KT5720 KT5720 zyxwvutsrq zyxwvu zy Effects of cGMP on transmitter release can occur downstream of Ca2+ influx It is known that somatostatin modulates transmitter release by acting on the Ca2+ channels that regulate vesicle fusion (Gray et al., 1989, 1990); however, the mechanism by which cGMP might inhibit KC1-evoked transmitter release in this system is not known. To determine if cGMP could inhibit ACh release downstream of Ca2+ entry, we evoked transmitter release using a Caz+ ionophore (10 pA4 A23187 for 10 min) and tested its sensitivity to a cGMP analogue (100 pA4 8-Br-cGMP). In these experiments, transmitter release from acutely dissociated ciliary ganglion neurons was measured as described above. Ionophore-mediated release is Ca’+-dependent, because removal of Ca2+ from the extracellular saline blocked ionophore-mediated release (data not shown). Figure 4 shows that 100 p M 8-Br-cGMP significantly reduced ionophore-evoked transmitter release by -50%. Thus, we hypothesize that a cGMP analogue can reduce Ca2+-dependent transmitter release in this system by affecting some aspect of the release process that occurs after Ca2+ entry. This hy- Con! SOM SOM+ Calphostin Calphoslin FIG. 3. Specificity of the effects of protein kinase blockers on somatostatin-mediatedinhibition of potassium-evoked PH]ACh release from acutely dissociated ST 40 ciliary ganglion neurons. A ACh release expressed as a % increase in 3H-labeled ACh release over basal levels when evoked with a 5-min 55 mM potassium stimulation. Control tissue is compared with tissue treated with 100 nM somatostatin (SOM), 1 pM KT5823 (a PKG inhibitor), and a coapplication of both somatostatin and KT5823. 6: Same experimental protocol as in A, except that 1 pM KT5720 (a cyclic AMP-dependent protein kinase inhibitor) was used. C: Same experimental protocol as in A, except that 1 pM calphostin C (a protein kinase C inhibitor) was used. Only the PKG inhibitor is able to reverse the effect of SOM on potassiumevoked PH]ACH release. All values are means ? SEM with n ? 4. *Significantly different from control ( p < 0.05) using a one-way analysis of variance with Dunnett’s test. pothesis was investigated further in the next series of experiments by directly measuring Ca2+ influx. Ca2+ influx We measured photometrically the Ca2’ influx in ST 40 ciliary ganglion neurons in response to high potassium stimulation to determine if the experimental perturbations described above altered intracellular Ca2’. Figure 5 shows representative examples of the effects of these perturbations. Exposure of control neurons to 55 mM KC1 elevated intracellular Ca2+ from a control value of 105 5 11 nM (mean t- SEM) to a peak value of 1,577 zyxw J. Neurochem., Vol. 72, No. 5, 1999 1986 zyxwvuts zyxwv zyxwvutsrqpo D. B. GRAY ET AL. affect the KC1-evoked Ca2+ transient or the somatostatin-mediated blockade of the potassium-evoked Ca2+ transient. As we demonstrated that an NO generator or cGMP analogue could reduce KC1-evoked transmitter release (Figs. 1C and 2D), we examined the KC1-evoked Ca2+ transient under these conditions. Pretreatment with an NO generator (300 pA4 SNP) did not significantly reduce the KC1-evoked Ca” transient (1,538 % 160 nM in the presence of SNP versus 1,860 2 246 nM during control KC1 exposure; n = 19; Fig. 5E). However, treatment with a cGMP agonist (100 FM 8-pCPT-cGMP) did significantly reduce the KC1-evoked Ca2+ transient beyond that normally observed with repeated exposure (699 2 57 nM in the presence of 8-pCPT-cGMP versus 1,022 ? 82 nM during control KC1 exposure; n = 10; p < 0.05; Fig. SF). An effect of a cGMP agonist on Ca2’ influx is expected based on the previous report of decreases in Ca2+ current in these neurons (Meriney et al., 1994). zyxwvutsrqpon zyxwvutsrqp zyxwvutsrq zyxwv cant cGMP FIG. 4. ACh release evoked by a Ca2+ ionophore is inhibited by 8-Br-cGMP. ACh release is expressed as a % increase over basal levels when evoked by a 10-min exposure to the Ca2+ ionophore A23187. Control cells are compared with cells treated with 100 f l 8-Br-cGMP (cGMP). Values plotted are means 5 SEM with n 2 4. *Significantly different from control ( p < 0.05) using Student’s t test. zy zyxwvutsrqpo ? 134 nM (n = 41; Fig. 5A). With several minutes of continuous 55 mM potassium stimulation, the transient rise in Ca2+ decayed to a plateau at 300-500 nM Ca2+ that was maintained until the KC1 stimulus was removed (Fig. 5A). Two 55 mM KCl stimuli presented in succession resulted in a small (1 1.7%), but significant, reduction in the size of the transient (1,895 2 348 nM following the first KCl application versus 1,646 ? 289 nM following the second KC1 application; n = 13). This reduction in peak Ca2+ influx with repeated KC1 exposure in control cells was compared with that in treatment groups presented below (using a one-way analysis of variance with Dunnett’s test). The large Ca2+ transient that was observed following 55 mM KC1 application was completely abolished by application of 100 nM somatostatin immediately before KC1 exposure. Figure 5B shows a representative example in which, following somatostatin application, KC1 consistently failed to elicit a detectable rise in intracellular Ca2+. To examine the effects of the NO-PKG pathway on the Ca2+ transients elicited by KC1 and the blocking effects of somatostatin on those transients, we examined the KC1-evoked Ca2+ influx following exposure to antagonists and agonists of the NO-PKG pathway. A 5-10min exposure to an inhibitor of NO synthase (L-NMMA) did not significantly alter the KC1-evoked Ca2+ transient beyond that observed with repeated KCI exposure in control cells. The Ca2+ transient recorded following KCl exposure was only slightly smaller in the presence of L-NMMA (1,248 2 215 nM in the presence of L-NMMA versus 1,330 247 nM before L-NMMA application; n = 6; Fig. 5C). In addition, the blocking effect of somatostatin was not affected significantly by the presence of L-NMMA (Fig. SC). Similarly, the presence of an inhibitor of PKG (Rp-8-Br-PET-cGMPS) did not result in a significant reduction in the KC1-evoked Ca2+ transient (1,550 2 238 nM in the presence of Rp-8-Br-PETcGMPS versus 1,852 2 178 nM before Rp-8-Br-PETcGMPS application; n = 6). In addition, somatostatinmediated blockade of Ca2+ influx was not altered by exposure to Rp-8-Br-PET-cGMPS (Fig. 5D). Thus, inhibitors of NO synthase or PKG do not significantly + Intracellular cGMP levels following the application of somatostatin or NO Because it appeared that a NO-PKG signal transduction cascade interacted with the somatostatin-mediated inhibition of potassium-evoked transmitter release, we investigated whether somatostatin receptor occupation activated this NO-PKG pathway. Meriney et al. (1994) had hypothesized previously that somatostatin receptor occupation might initiate two signaling cascades that converge on the Ca2+ channel: a membrane-delimited, and a cytoplasmic pathway involving PKG. Thus, we were interested in understanding the conditions under which the PKG pathway could be activated. We directly measured cGMP production using a radioimmunoassay following application of SNP (at 300 pM, a NO generator) or varying concentrations of somatostatin. Incubation of ciliary ganglion cells for 1 min with somatostatin at 1, 10, or 100 nM did not increase cGMP above control values (Fig. 6). However, a 1-min exposure to SNP resulted in an 10-fold increase in cGMP levels. Thus, although NO generation leads to a robust increase in cGMP, we could not detect any effect of somatostatin on cGMP levels. - DISCUSSION Ciliary ganglion neurons as a model for studying somatostatin modulation of ACh release Direct recording from most presynaptic nerve terminals is difficult because of their small size (Stanley and Goping, 1991; Borst and Sakmann, 1996; Yazejian et al., 1997). Therefore, cell preparations where the cell body releases transmitter in a Ca2+-dependent manner are useful for studying the modulation of transmitter release. The neuropeptide somatostatin has been shown to be a potent inhibitor of Ca2+ current in a number of neuronal cells (Ikeda and Schofield, 1989; Sah, 1990; Surprenant et al., 1990; Wang et al., 1990; Golard and Siegelbaum, 1993; Viana and Hille, 1996), including parasympathetic zyxwvuts zyxwvuts J. Neurochem., Vol. 72, No. 5, 1999 zyxwvutsr zyxw zy NO-PKG ALTERATION OF SOMATOSTATIN INHIBITION 1987 zyxwvutsrqpon zyxwvutsrq FIG. 5. Representativeexamples of the photometric measurements of Ca2+ transients elicited by exposure of ciliary ganglion neurons to elevated extracellular potassium (KCI). Bars above each plot represent the time when KCI or reagents (as indicated) were applied to the neuron under study. A The response of control cells to 25 or 55 mM KCI exposure. A prolonged exposure to 55 mM KCI resulted in a brief Caz+ transient, followed by a plateau of elevated Ca2+ lasting as long as the exposure to 55 mM KCI. In all subsequent panels, 55 mM KCI was used. B The Ca2+ transient observed following KCI exposure was eliminated by coapplication of 100 nM somatostatin. C An inhibitor of NO synthase (L-NMMA) did not affect the KCIevoked Ca2+ transient and did not reverse the somatostatin-mediated blockade of the KCI-evokedCa2+transient. D An inhibitor of PKG, Rp-8-pCPT-cGMPS (Rp-cGMPS), did not reverse the somatostatin blockade of the KCI-evoked Ca2+transient. E: A NO donor, SNP, did not affect the KCI-evoked Ca2+ transient. F A cGMP agonist, 8-pCPTcGMP (cGMP), slightly, but significantly, reduced the KCI-evoked Ca2+transient. I 0 5 10 15 20 Time (min) 25 0 E " 5 10 Time ( m n ) 5 10 Time (min) " , " 15 15 zyxwvuts KCI KCI I I SNP I 1500 0 1200 zyxwvutsrq 300 10 Time (min) 15 - , 0 5 10 15 20 Time (min) zyxwvut ciliary ganglion neurons (Dryer et al., 1991; Meriney et al., 1994; White et al., 1997). These neurons express somatostatin and synthesize the transmitter ACh (Epstein et al., 1988; Gray et al., 1990; De Stefan0 et al., 1993), and also release endogenous somatostatin preferentially under periods of high-frequency stimulation (D. B. Gray and G. R. Pilar, unpublished observations; see Hokfelt, 1991). When released, somatostatin is a potent inhibitor of ACh release serving a negative feedback role at the nerve terminal (Gray et al., 1989, 1990). ACh release induced by a Ca2+ ionophore in these terminals is not sensitive to somatostatin, indicating that the site of somatostatin modulation may be transmembrane Ca2+ flux through voltage-dependent Ca2+ channels (Gray et al., 1989). Previously reported effects of somatostatin on Ca2+ currents in acutely dissociated ciliary ganglion neurons support the hypothesis that somatostatin is acting to inhibit release by decreasing Ca" influx (Meriney et al., 1994). Therefore, we have used freshly dissociated ciliary ganglion neurons in vitro as a model to study somatostatin-mediated modulation of ACh release and Ca2+ influx. This acutely dissociated cell is a unique system because ciliary ganglion neurons have been shown to release transmitter from their cell bodies (Johnson and Pilar, 1980). These ganglionic neurons provide the opportunity to study the coupling of somatostatin receptors to Ca2+ influx and the inhibition of transmitter release in parasympathetic neurons, as well as compare results obtained here with those that have been reported elsewhere. We describe the alteration by a NO-PKG pathway of the somatostatin-mediated inhibition of Ca2+ influx and transmitter release. NO modulation of evoked ACh release has been demonstrated in other preparations (Meulemans et al., 1995; Kilbinger, 1996; Meng et al., 1996; Sawada and Ichinose, 1996; Suzuki et al., 1997) J. Neurochem., Vol. 72, No. 5, 1999 I988 zyxwvutsr zyxwvu zyxwv D. B. GRAY ET AL. Con1 surprising that somatostatin reduces Ca2+ current by -50%, while reducing transmitter release to a much greater extent. The KC1-evoked Ca2+ signal we recorded in ciliary ganglion neuron cell bodies is completely blocked by somatostatin. Perhaps the roughly 50% reduction in Ca2+ current mediated by somatostatin reduces Ca2+ influx such that local intracellular buffering near the plasma membrane prevents this smaller influx from raising the volume-averaged Ca2+ concentration in the cell body. Alternatively, the KC1-evoked Ca" signal may be generated, in part, by Ca2+ from sources other than the voltage-gated Ca2+ channel, and somatostatin may block the elevation of Ca2+ generated by these sources. High potassium stimulation causes a rise in intracellular Ca2+ that is mediated by several types of voltage-gated Ca2+ channels (Momiyama and Takahashi, 1994) and may also be affected by release from intracellular stores (Friel and Tsien, 1992; Budd and Nicholls, 1996). Under these stimulus conditions, it is likely that transmitter release is sensitive to Ca" from these varied sources. In this respect, KC1 stimulation is different from electrical stimulation of transmitter release. Although we may not photometrically measure selectively the subcompartment of Ca2+ that triggers transmitter release from these neuron somata, it is striking that somatostatin completely blocks the KC1-evoked Ca2' transient, and this effect is not altered by manipulations of the NO-PKG pathway. zyxwvutsrqponmlkjihgf zyxwvutsrqponmlkj zyxwvutsrqp zyxwvutsrq 1 nM lOnM 10OnM SNP Somatostatin FIG. 6. cGMP levels (fmol/pg of protein) measured from acutely dissociated ST 40 ciliary ganglion neurons. Somatostatin (1, 10, or 100 nM) did not significantly increase cGMP above control (Cont) levels, whereas 300 pM SNP caused a significant increase. All values are means ? SEM with n 2 4. *Significantly different from control ( p < 0.05) using a one-way analysis of variance with Dunnett's test. and, in some cases, has been shown to involve cGMP (Kilbinger, 1996; Wu et al., 1997). However, these effects on ACh release are distinct from the NO-PKG regulation of the effectiveness of G protein-mediated modulation of ACh release reported here. This is the first demonstration of a convergent effect of a NO-PKG pathway onto peptide-mediated inhibition of transmitter release in neurons. zyxwvutsrq Somatostatin modulation of ACh release Somatostatin is a potent inhibitor of ACh release from both intact choroid nerve terminals (Fig. 1) and ciliary ganglion cell bodies at ST 40 (Figs. 2 and 3). Somatostatin often inhibits KC1-evoked release almost completely in both the intact choroid tissue and isolated neuronal somata. This is in contrast to the roughly 50% inhibition of Ca2+ current observed using patch-clamp recording techniques from the cell bodies (Meriney et al., 1994; White et al., 1997). The reason for this difference in the magnitude of effects in the cell bodies where the preparations are identical may be due to three factors. First, not all types of Ca2+ channels in the cell body are modulated by somatostatin. ST 40 ciliary ganglion neurons express N- and L-type channels, but at this stage, only the N-type is modulated (White et al., 1997). Second, Ca2+ channels that are modulated by somatostatin are not expected to be completely blocked. Under conditions in which modulated current is associated with kinetic slowing, recent single-channel recordings have revealed a delayed first latency to opening for modulated channels (Carabelli et al., 1996; Patil et al., 1996) and, in some cases, a decreased probability of opening (Carabelli et al., 1996). Lastly, because the relationship between Ca2+ entry and transmitter release has been shown to be very nonlinear (Dodge and Rahamimoff, 1967; Augustine and Charlton, 1986; Borst and Sakmann, 1996; Takahashi et al., 1996), a modest reduction in Ca2' current can have very pronounced effects on transmitter release. Thus, it is not J. Neurochem., Vol. 72, No. 5, 1999 Potential mechanisms mediating the alteration by a NO-PKG pathway of the somatostatin modulation of ACh release The ability of inhibitors of NO synthase or PKG to override somatostatin-mediated inhibition of transmitter release (Figs. 1-3) occurs without any apparent change in the somatostatin-mediated block of the photometrically measured Ca2+ transient (Fig. 5). The more subtle effect of PKG inhibitors on the gating of modulated current previously reported (Meriney et al., 1994) is not likely to be detected using photometric measures of cytosolic Ca2+ following exposure to elevated potassium for several minutes. As such, these data do not address the previously reported effects of PKG on the characteristics of modulated Ca2+ current studied using the patchclamp electrophysiological technique (Meriney et al., 1994). It is interesting that an NO generator could decrease potassium-evoked transmitter release without significantly altering the Ca2' transient (Fig. 5E). Although a cGMP analogue can reduce Ca2+ influx (Fig. 5F), this effect on Ca2+ channels may be in addition to direct effects of cGMP downstream of Ca2' entry, because a cGMP analogue can reduce the transmitter release activated by a Ca2+ ionophore (Fig. 4). These data suggest that the NO-PKG pathway can inhibit transmitter release at a point after Ca2+ entry. This is in contrast to the established effects of somatostatin on transmitter release that have been shown to be due to a reduction in Ca2' entry (Gray et al., 1989, 1990). In considering the po- NO-PKG ALTERATION OF SOMATOSTATIN INHIBITION tential mechanisms whereby somatostatin-mediated modulation of the Ca2+ channels that regulate transmitter release could interact with a NO-PKG pathway that can directly affect the release process, we have limited our discussion to one hypothesis that seems most congruous with the data. We propose that the NO-PKG pathway modulates the sensitivity of the release apparatus to Ca2+. When active, the NO-PKG pathway is hypothesized to reduce the Ca2+ sensitivity and, when pharmacologically blocked, to increase the Ca2+ sensitivity. Individual proteins associated with the secretory machinery have been shown to be phosphorylated by a variety of kinases, and some of these phosphorylation events have been shown to alter the Ca2+ dependence of transmitter release (Trudeau et al., 1996, 1998; Rettig et al., 1997; Yokoyama et al., 1997). Furthermore, intracellular signaling cascades can alter transmitter release downstream of Ca2' entry (Dale and Kandel, 1989; Man-Son-Hing et al., 1989; Przywara et a]., 1991). Our hypothesis explains the ability of NO-PKG blockers to override the somatostatin-mediated inhibition of transmitter release without reversing the somatostatinmediated reduction in Ca2+ influx. The inability of NO synthase inhibitors or PKG inhibitors to increase potassium-evoked transmitter release may be due to saturation of the transmitter release apparatus with Ca" during the 5-min potassium challenge such that an increase in responsiveness or sensitivity to Ca2+ cannot further increase transmitter release. Although this hypothesis is consistent with our data, it is also possible that the NO-PKG pathway alters the local microdomain of Ca2+ that triggers transmitter release, and we were unable to detect these changes. Endogenously, somatostatin is most likely to be released during periods of high-frequency action potential activity, which significantly increases intracellular Ca2+. This same stimulus, rather than somatostatin receptor occupation, may activate the NO-PKG pathway as well. Thus, high levels of excitation of these cholinergic neurons may lead to release of somatostatin and an increase in the activity of the NO-PKG pathway. When active together, the NO-PKG pathway may amplify the effects of somatostatin, leading to a feedback inhibition of ACh release. Both somatostatin and NO (from ciliary ganglion neurons and/or neighboring vascular tissue) may spread this amplification by diffusion to surrounding ACh release sites in a paracrine fashion. Independent of the specific mechanisms involved, the convergence of somatostatin-mediated, membrane-delimited inhibition of Ca2+ channels with an active NOPKG intracellular signal transduction cascade appears to act as a coincidence detector (Bourne and Nicoll, 1993). When active together, these two signal transduction pathways assure that the somatostatin inhibition of Ca2+ channels and ACh release will be robust. The presence of a NO-PKG cascade provides another level of complexity to the mechanisms that neurons possess to fine-tune synaptic activity. It is possible that high-frequency action potential activity serves both to release endogenous so- zyxw zy 1989 matostatin from choroid nerve terminals and to activate the NO-cGMP pathway as a result of the elevated intraterminal Ca2+. Furthermore, the NO-cGMP pathway could be activated following NO generation in neighboring choroid vascular tissue. In this fashion, somatostatin modulatory effects on presynaptic nerve terminal Ca2+ channels and subsequent transmitter release can be controlled at several points by regulating the release of somatostatin from choroid nerve terminals and the activation of the NO-cGMP pathway at several sites within the choroid tissue. The endogenous control over this balance in the activation and effectiveness of the somatostatin modulation of vascular tone in vivo remains a subject for future inquiry. zyxw Acknowledgment:This work was supported by The Patrick and Catherine Donaghue Medical Research Award no. 93-047 and The Simmons Fund for Research no. 377 (D.B.G.), NIH NS 10338 (G.R.P.), NIH NS 32345 (S.D.M.), and a Grant-inAid from the American Heart Association (S.D.M.). We thank Debra Artim, James Dilmore, John Pattillo, Robert Poage, and James Simples, Jr., for many helpful discussions and critical evaluation of the manuscript. zyxwvu zy zyxwv zyxwvu REFERENCES Augustine G. J. and Charlton M. P. (1986) Calcium dependence of presynaptic calcium current and post-synaptic response at the squid giant synapse. J. Physiol. (Lond.) 381, 619-640. Borst J. G. G. and Sakmann B. (1996) Calcium influx and transmitter release in a fast CNS synapse. Nature 383, 431-434. Bourne H. R. and Nicoll R. (1993) Molecular machines integrate coincident synaptic signals. Cell 72, 65-75. Budd S. L. and Nicholls D. G. (1996) A reevaluation of the role of mitochondria in neuronal Ca2+ homeostasis. J. Neurochem. 66, 403-41 1. Carabelli V., Lovallo M., Magnelli V., Zucker H., and Carbone E. (1996) Voltage-dependent modulation of single N-type CaZ+ channel kinetics by receptor agonists in IMR32 cells. Biophys. J. 70, 2144-2154. Chetkovich D. M., Klann E., and Sweatt J. D. (1993) Nitric oxide synthase-independent long-term potentiation in area CA 1 of hippocampus. Neuroreport 4, 919-922. Crean G., Pilar G., Tuttle J. B., and Vaca K. (1982) Enhanced chemosensitivity of chick parasympathetic neurones in co-culture with myotubes. J. Physiol. (Lond). 331, 97-104. Dale N. and Kandel E. R. (1990) Facilitatory and inhibitory transmitters modulate spontaneous transmitter release at cultured Aplysia sensorimotor synapses. J. Physiol. (Lond.) 421, 203-229. De Stefan0 M. E., Luzzato A., and Mugnaini E. (1993) Neuronal ultrastructure and somatostatin immunolocalization in the ciliary ganglion of chicken and quail. J. Neurocytol. 22, 868-892. Dodge F. A. and Rahamimoff R. (1967) Co-operative action of calcium ions in transmitter release at the neuromuscular junction. J. Physiol. (Lond.)193, 419-432. Dolphin A. C. (1998) Mechanisms of modulation of voltage-dependent calcium channels by G proteins. J. Physiol. (Lond.) 506, 3-1 1. Dryer S. E., Dourado M. M., and Wisgirda M. E. (1991) Properties of Caz+ currents in acutely dissociated neurons of the chick ciliary ganglion: inhibition by somatostatin-14and somatostatin-28.Neuroscience 44, 663-672. Epstein M. L., Davis J. P., Gellman L. E., Lamb J. R., and Dahl J. L. (1988) Cholinergic neurons of the chicken ciliary ganglion contain somatostatin. Neuroscience 25, 1053-1060. Friel D. D. and Tsien R. W. (1992) Phase-dependentcontributionsfrom Ca2+ entry and Ca2' release to caffeine-induced [Ca2+],oscillations in bullfrog sympathetic neurons. Neuron 8, 1109-1 125. J. Neurochem., Vol. 72, No. 5, 1999 I990 zyxwvu zyxwv zyxwvutsr zyxwvutsrqp D. B. GRAY ET AL. Garthwaite J. and Boulton C. L. (1995) Nitric oxide signaling in the central nervous system. Annu. Rev. Physiol. 57, 683-706. Golard A. and Siegelbaum S. A. (1993) Kinetic basis for the voltagedependent inhibition of N-type calcium current by somatostatin and norepinephrine in chick sympathetic neurons. J. Neurosci. 13, 3884 -3894. Gray D. B. and Tuttle J. B. (1987) ACh synthesis in cultured ciliary ganglion neurons: effects of myotube membranes. Dev. Biol. 119, 290 -298. Gray D. B., Pilar G. R., and Ford M. J. (1989) Opiate and peptide inhibition of transmitter release in parasympathetic nerve terminals. J. Neurosci. 9, 1683-1692. Gray D. B., Zelazny D., Manthay N., and Pilar G. (1990) Endogenous modulation of ACh release by somatostatin and the differential roles of calcium channels. J. Neurosci. 10, 2687-2698. Gray D. B., Brusts J. L., and Pilar G. R. (1992) Developmental switch in the pharmacology of Ca2+ channels coupled to acetylcholine release. Neuron 8, 715-724. Grynkiewicz G., Poenie M., and Tsien R. Y. (1985) A new generation of Ca2+ indicators with greatly improved fluorescence properties. J . Biol. Chem. 260, 3440-3450. Hamburger V. and Hamilton H. L. (1951) A series of normal stages in the development of the chick embryo. J. Morphol. 88, 49-92. Hille B. (1992) G protein-coupled mechanisms and nervous signaling. Neuron 9, 187-195. Hille B. (1994) Modulation of ion channels by G-protein coupled receptors. Trends Neurosci. 17, 531-535. Hokfelt T. (1991) Neuropeptides in perspective: the last ten years. Neuron 7, 867-879. Ikeda S. R. and Schofield G. G. (1989) Somatostatin blocks a calcium current in rat sympathetic ganglion neurons. J. Physiol. (Lond.) 409, 221-240. Johnson D. A. and Pilar G. (1980) The release of acetylcholine from postganglionic cell bodies in response to depolarization. J. Physiol. (Lond.) 299, 605-619. Kilbinger H. (1996) Modulation of acetylcholine release by nitric oxide. Prog. Brain Res. 109, 219-224. Man-Son-Hing H., Zoran M. J., Lukowiak K., and Haydon P. G. (1989) A neuromodulator of synaptic transmission acts on the secretory apparatus as well as on ion channels. Nature 341, 237-239. Meng W., Ma J., Ayata C., Hara H., Huang P. L., Fishman M. C., and Moskowitz M. A. (1996) ACh dilates pial arterioles in endothelial and neuronal NOS knockout mice by NO-dependent mechanisms. Am. J. Physiol. 271, HI 145-HI 150. Meriney S. D., Gray D. B., and Pilar G. R. (1994) Somatostatininduced inhibition of neuronal Ca2+ current modulated by cGMPdependent protein kinase. Nature 369, 336-339. Meulemans A,, Mothet J. P., Schirar A,, Fossier P., Tauc L., and Baux G. (1995) A nitric oxide synthase activity is involved in the modulation of acetylcholine release in Aplysia ganglion neurons: a histological, voltammetric and electrophysiological study. Neuroscience 69, 985-995. Momiyama A. and Takahashi T. (1994) Calcium channels responsible for potassium-induced transmitter release at rat cerebellar synapses. J. Physiol. (Land.) 476, 197-202. Nestler E. J. and Greengard P. (1984) Neuron-specific phosphoproteins in mammalian brain. Adv. Cyclic Nucleotide Protein Phosphorylation Res. 17,483-488. Nichol K. A., Chan N., Davey D. F., and Bennett M. R. (1995) Location of nitric oxide synthase in developing avian ciliary ganglion. J. Auton. Nerv. Syst. 51, 91-102. Patil P. G., de Leon M., Reed R. R., Dubel S., Snutch T. P., and Yue D. T. (1996) Elementary events underlying voltage-dependent G protein inhibition of N-type calcium channels. Biophys. J. 71, 2509-2521. Pilar G., Gray D. B., and Meriney S. D. (1996) Membrane delimited and intracellular soluble pathways in the somatostatin modulation of ACh release. Life Sci. 58, 1979-1996. Przywara D. A,, Bhave S. V., Bhave A., Wakade T. D., and Wakade A. R. (1991) Dissociation between intracellular Caz+ and modulation of [3H]noradrenaline release in chick sympathetic neurons. 3. Physiol. (Lond.) 437, 201-220. Rettig J., Heinemann C., Ashery U., Sheng Z.-H., Yokoyama C. T., Catterall W. A,, and Neher E. (1997) Alteration of Ca2+ dependence of neurotransmitter release by disruption of Ca2+ channel/ syntaxin interaction. J. Neurosci. 17, 6647-6656. Sah D. W. (1990) Neurotransmitter modulation of calcium current in rat spinal cord neurons. J. Neurosci. 10, 136-141. Sawada M. and Ichinose M. (1996) Nitric oxide donor sodium nitroprusside inhibits the acetylcholine-induced K+ current in identified Aplysia neurons. J. Neurosci. Res. 44, 21-26. Stanley E. F. and Goping G. (1991) Characterization of a calcium current in a vertebrate cholinergic presynaptic nerve terminal. J. Neurosci. 11, 985-993. Surprenant A,, Shen K. Z . , North R. A,, and Tatsumi H. (1990) Inhibition of calcium currents by noradrenaline, somatostatin and opioids in guinea-pig submucosal neurones. J. Physiol. (Lond.) 431, 585-608. Suzuki T., Nakajima K., Fujimoto K., Fujii T., and Kawashima K. (1997) Nitric oxide increases stimulation-evoked acetylcholine release from rat hippocampal slices by a cyclic GMP-independent mechanism. Brain Res. 760, 158-162. Takahashi T., Forsythe I. A,, Tsujimoto T., Barnes-Davis M., and Onodera K. (1996) Presynaptic calcium current modulation by a metabotropic glutamate receptor. Science 274, 594-597. Trudeau L.-E., Emery D. G., and Haydon P. G . (1996) Direct modulation of the secretory machinery underlies PKA-dependent synpatic facilitation in hippocampal neurons. Neuron 17, 789-797. Trudeau L.-E., Fang Y., and Haydon P. G. (1998) Modulation of an early step in the secretory machinery in hippocampal nerve terminals. Proc. Natl. Acad. Sci. USA 95, 7163-7168. Vaca K. and Pilar G. (1979) Mechanisms controlling choline transport and acetylcholine synthesis in motor nerve terminals during electrical stimulation. J. Gen. Physiol. 73, 605-628. Viana F. and Hille B. (1996) Modulation of high voltage-activated calcium channels by somatostatin in acutely isolated rat amygdaloid neurons. J. Neurosci. 16, 6000-6011. Wang H. L., Reisine T., and Dichter M. (1990) Somatostatin-14 and somatostatin-28 inhibit calcium currents in rat neocortical neurons. Neuroscience 38, 335-342. White M. G., Crumling M. A., and Meriney S. D. (1997) Developmental changes in calcium current pharmacology and somatostatin inhibition in chick parasympathetic neurons. J. Neurosci. 17, 6302-6313. Wu L. G. and Saggau P. (1997) Presynaptic inhibition of elicited neurotransmitter release. Trends Neurosci. 20, 204-212. Wu S. Y., Dun S. L., Forstermann U., and Dun N. J. (1997) Nitric oxide and excitatory postsynaptic currents in immature rat sympathetic preganglionic neurons in vitro. Neuroscience 79, 237-245. Yazejian B., DiGregorio D. A,, Vergara J. L., Poage R. E., Meriney S. D., and Grinnell A. D. (1997) Direct measurements of presynaptic calcium and calcium-activated potassium currents regulating neurotransmitter release at cultured Xenopus nerve-muscle synapses. J. Neurosci. 17, 2990-3001. Yokoyama C. T., Sheng Z . H., and Catterall W. A. (1997) Phosphorylation of the synaptic protein interaction site on N-type calcium channels inhibits interactions with SNARE proteins. J . Neurosci. 17, 6929-6938. zyxwvutsrqpo zyxwvutsrqpon zyxwvutsrqpon J. Neurochem., Vol. 72, No. 5, 1999