Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Endothelin (ET) is a family of three vasoactive isopeptides (ET-1, ET-2, and ET-3) of 21 amino acids (Yanagisawa and Masaki, 1989). The peptides exist not only in nonneuronal tissues but also in the nervous system. ET-like immunoreactivity and/or ET mRNA was found to be present in the neurons of the human spinal cord and dorsal root ganglia (Giaid et al., 1989), in the porcine (Shinmi et al., 1989) and rat (Matsumoto et al., 1989) brain, and in the glial cells of rat brain (Cintra et al., 1989). In addition, three distinct human ET-related genes were cloned by screening the genomic DNA library, and the encoded peptides were thus named ET-1 (identical with the original ET), ET-2, and ET-3 (Inoue et al., 1989). ET-1 has also been demonstrated to be the only ET produced by endothelial cells, whereas both ET-1 and ET-3 have been identified in neural tissue (Greenberg et al., 1992; Sakurai et al., 1992). ET-3 was shown to be abundant in the porcine brain as well as the spinal cord (Yanagisawa and Masaki, 1989; Shinmi et al., 1989) and the rat brain (Matsumoto et al., 1989). Furthermore, at least two kinds of ET receptor subtypes, ET_A and ET_B receptors, have been isolated from the cDNA libraries of many types of mammalian tissue (Masaki et al., 1992).

In nerve terminals of the autonomic nervous system, ET was shown to inhibit the nerve-induced contractions and release of [³H]acetylcholine or [³H]norepinephrine in the guinea pig ileum and femoral artery, respectively (Wiklund et al., 1988; 1989), suggesting that ETs seem to act on presynaptic sites and reduced transmitter release. In our previous investigations of canine cardiac sympathetic ganglia (Tsutsumi et al., 1995; Kushiku et al., 1995), ET-1 and ET-3 inhibited the sympathetic ganglionic transmission at postsynaptic sites through activation of ET_A receptor-operated low-conductance Ca²⁺-activated K⁺ channels, and ET-3 also inhibited the ganglionic transmission at presynaptic sites by reducing the output of acetylcholine release through activation of thromboxane A₂ production. These results indicate that ET may play an important role in sympathetic nerve transmission. However, the ET receptor subtype involved in this presynaptic inhibition of acetylcholine release remains to be determined.

On the other hand, nitric oxide (NO) plays an important role in the control of synaptic function in both the peripheral and central nervous system (Schuman and Madison, 1994). NO is synthesized in neurons through the activation of NO synthase (NOS) with Ca²⁺-calmodulin-dependent manner, in the presence of L-arginine as substrate, tetrahydrobiopterin, and NADPH (Bredt et al., 1992). Major effects of NO are the stimulation of soluble guanylate cyclase, leading to formation of intracellular cGMP (Waldman and Murad, 1987) and of cyclooxygenase 1, producing increased metabolism of arachidonic acid and formation of various prostaglandins (Lundberg, 1996). In addition, NO itself can open Ca²⁺-dependent K⁺ channels, K_Ca²⁺, apparently without requiring cGMP formation (Lundberg, 1996). NOS is found in the majority of preganglionic cholinergic neurons of the sympathetic nervous system in the rat (Anderson et al., 1993; Okamura et al., 1995), guinea pig (Höhler et al., 1995), cat (Dun et al., 1993; Anderson et al., 1995), mouse, and monkey (Dun et al., 1993), and preganglionic nerve stimulation leads to an NO-mediated increase of intracellular cGMP in the postganglionic neurons (Briggs, 1992; Sheng et al., 1992).

ET_B receptor activation triggered by ET-1 in the guinea pig perfused lung and rabbit kidney led to a release of thromboxane A₂ and NO (D'Orléans-Juste et al., 1994). In the whole rat adrenal medulla, ET stimulates NO-induced cGMP generation through ET_B receptors; this finding supports the concept that ET may play a role in regulating the functions in the adrenal medulla (Mathison and Israel, 1998).

In the current study, we investigated the possible contribution of the NO system to the ET-induced inhibition of ganglionic transmission, while also specifying the ET receptor subtype involved in this inhibition.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Animal Care. Mongrel adult dogs of either sex, provided by the Fukuoka City Animal Control Center, were kept for about 1 week in the animal laboratory (Fukuoka University Animal Center), and during that time they underwent general medical examination. Each dog was housed in an individual cage in a temperature-controlled room (22°C) which was humidified (50-60%) and maintained on a 12-h/12-h light/dark cycle (8 AM/8 PM). All animals had free access to water and were fed standard solid laboratory food (ED-1, 300 g/dog/day; Sanwa Chemicals Inc., Tokyo, Japan). Only animals in good physical health were used in the experiments. The experimental procedures were carried out under the protocols approved by the Animal Care Committee of Fukuoka University and in accordance with the principles outlined in the National Institutes of Health Guide for the Care and Use of Laboratory Animals.

In Vitro Experiments on the Ganglia. All dogs (6-12 kg) were anesthetized with pentobarbital sodium (30 mg/kg i.v.). The trachea was cannulated, and artificial ventilation was maintained by a Harvard animal respirator (model 613; Harvard, Millis, MA). Both stellate ganglia were removed, along with about 3 cm of the preganglionic sympathetic nerve for the nerve stimulation experiments. For the experiments of acetylcholine assay, the tissue specimens were placed in a dish containing Locke's solution in the presence of physostigmine (10⁶ M) at room temperature and gassed with a mixture of 95% O₂ and 5% CO₂, and the tissue sheath around the ganglia was carefully removed. The experimental procedures were principally performed according to the methods described by Ohjimi et al. (1994). The composition of the Locke's solution was 136 mM NaCl, 5.6 mM KCl, 2.2 mM CaCl₂, 1.2 mM MgCl₂, 1.2 mM NaH₂PO₄, 20.0 mM NaHCO₃, and 11.0 mM glucose.

6

Determination of Acetylcholine. Acetylcholine was measured by HPLC with electrochemical detection as described by Eva et al. (1984) and Potter et al. (1983). A standard mixture of acetylcholine, choline, and isopropylhomocholine chloride at each concentration of 2 ×10⁶ M was prepared daily from the stock solution of 2 ×10³ M stored at 4°C. Isopropylhomocholine chloride (internal standard, 15 µl of 2 ×10⁵ M) was added into 300 µl of sample solution and filtered through a 0.22-µm membrane filter (UFC30GVOO; Millipore, Tokyo, Japan). Aliquots of the standard and the filtered samples, 5 µl and 10 µl, were injected, respectively, into the HPLC system. The temperature of the enzyme column was maintained at about 33°C by a column heater (U-620; Sugai, Tokyo, Japan). The mobile phase consisted of 0.1 M disodium hydrogen phosphate buffered to pH 8.0 with phosphoric acid, containing 0.6 mM tetramethylammonium chloride and 1.2 mM sodium 1-decanesulfonate. The buffer was first prepared and filtered through a 0.45-µm membrane filter (Toyo Roshi, Tokyo, Japan). Tetramethylammonium chloride and sodium 1-decansulfonate were added, and the solution was degassed by bubbling helium gas at a flow rate of 100 ml/min for 30 min. The pumping rate of the mobile phase was 1.0 ml/min.

Acetylcholine Release. In the experiments performed on the isolated ganglia, the output of acetylcholine was collected for a period of 10 min during preganglionic stimulation. The first and second samples were untreated, and thereafter every third sample was exposed to the drugs. Fresh medium containing the corresponding agents at the same concentration was introduced after the preincubation to remove the resting amount of acetylcholine released during the preincubation periods. Five to six samples were taken in the acetylcholine assay experiment.

Measurement of NO₂ and NO₃ Levels. The isolated ganglia, with their preganglionic trunk removed, were placed in a tube containing 0.5 ml of Locke's solution and were oxygenated with 95% O₂/5% CO₂ and incubated at 37°C in a water bath. The output of NO was collected over periods of 30-min incubation. The first through third samples were untreated, and thereafter fourth through sixth samples were exposed to the drugs. The intervals between each successive incubation period was 20 min. The agents were administered after three consecutive untreated samples had been collected as the basal levels. The left ganglia were used for the control groups, and the right for the antagonist-treated groups. Fresh medium containing the corresponding antagonists at the same concentration was introduced after the preincubation to remove the amount of NO released during the preincubation periods of antagonists.

Drugs Used. The drugs used were: ET-3 and Suc-[Glu⁹,Ala^11,15]-ET-1(8-21) (IRL-1620) (Peptide Institute Inc., Osaka, Japan), N-cis-2,6-dimethyl-piperidinocarbonyl-L-g-methyl-Leu-D-1-methoxycarbonyl-Trp-D-Nle (BQ-788) (Peninsula Laboratories Inc., Belmont, CA), 3-bromo-7-nitroindazole, ODQ, and SNAP (Tocris Cookson Ltd., Bristol, UK), 1-[b-[3-(4-methoxyphenyl)propoxy]-4-methoxyphenethyl]-1H-imidazole (SKF-96365; Biomol Research Laboratories Inc., Plymouth Meeting, PA), ethyleneglycol bis(2-aminoethyl ether) tetraacetic acid (EGTA-2Na; Nacalai Tesque Inc., Kyoto, Japan), N-(6-aminohexyl)-5-chloro-1-naphthalenesulfonamide (W-7), physostigmine hemisulfate and 8-bromoguanosine-3,5-cyclic monophosphate (8-bromo-cGMP; Sigma Chemical Co., St. Louis, MO). Cyclo(D-a-aspartylL-prolyl-D-valyl-L-leucyl-D-tryptophyl) (BQ-123), and 4-tert-butyl-N-[6-(2-hydroxy-ethoxy)-5-(2-methoxyphenoxy)-2,2'-bipyrimidin-4-yl]-benzenesulphonamide sodium salt (bosentan) were a generous gift from Banyu Pharmaceutical Co. Ltd. (Tsukuba, Japan) and Hoffmann-La Roche Ltd. (Basel, Switzerland), respectively.

Statistical Analysis. Each value represents the mean ± S.E. The values between more than two mean values of the dose-response run in the same ganglia were evaluated by the Dunnett test. Comparisons of the left-untreated with antagonists and right-treated ganglia in the same animals were performed using the ANOVA, and followed by the Bonferroni t test. Values of p < .05 were considered to be significantly different.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Inhibitory Effects of ET Agonists on Acetylcholine Release. When the preganglionic nerve was stimulated, the amount of acetylcholine release was increased in a frequency-dependent manner, being one-half the maximum at 5 Hz and maximal at 20 Hz (Ohjimi et al., 1994). According to this finding, the preganglionic nerve stimulation was applied at 5 Hz for the following experiments. As shown in Fig. 1, the output of acetylcholine released by preganglionic stimulation for each 10-min period was reduced in a dose-dependent fashion in the presence of ET-3 at concentrations of 10⁹ to 10⁶ M or IRL-1620 at 10⁸ to 10⁵ M; the release was reduced to 60.9% and 66.6% of the control at maximal concentrations of 10⁶ and 10⁵ M, respectively.

View larger version (21K): [in this window] [in a new window]   Fig. 1.   Concentration-response curves for inhibition by endothelin agonists of acetylcholine output elicited by preganglionic stimulation. ET-3 ()- and IRL-1620 (ETB agonist) ()-treated ganglia. The mean basal amount of acetylcholine released by preganglionic stimulation without the agonist treatment was 7.2 ± 2.26 and 8.4 ± 3.67 nmol/10 min/g, respectively. Each point represents the mean of 10 and 8 ganglia, and vertical bars represent the S.E.M. *, : significant difference between the values before and after the agonists in the same ganglion (p < .05 or less).

Effects of ET Antagonists on ET-3-Induced Reduction of Acetylcholine Release. As shown in Fig. 2, the ET-3-induced inhibition of acetylcholine release during preganglionic stellate stimulation was clearly antagonized by incubation with a nonselective ET_A and ET_B receptors antagonist, bosentan, or an ET_B receptor antagonist, BQ-788, each at a concentration of 10⁶ M, whereas the inhibition was not affected by a ET_A receptor antagonist, BQ-123, at 10⁶ M. The acetylcholine release was not affected significantly by the application of either drug alone in the absence of ET-3. The selective ET_A receptor antagonist, BQ-123, competitively inhibits the effect of ET-1 in the human umbilical artery (pA₂, 6.9), and the contraction induced by ET-3 in the vein is inhibited by BQ-788 (pA₂, 7.6) (Bogoni et al., 1996). Therefore, the concentration of BQ-123 and BQ-788 used in the present study, 10⁶ M, is sufficient to antagonize the effects induced by ET_A and ET_B receptor activation, respectively.

View larger version (26K): [in this window] [in a new window]   Fig. 2.   Antagonism by a nonselective endothelin ETA and ETB receptor antagonist, bosentan, and selective ETB receptor antagonist, BQ-788, of the ET-3-induced reduction in acetylcholine output by preganglionic stimulation in untreated () and those in the presence of 106 M bosentan () or 106 M BQ-788 () or BQ-123 (). In six groups of ganglia, the control mean basal amount of acetylcholine release by the first preganglionic stimulation before ET-3 was 8.9 ± 1.27, 10.3 ± 2.03, 9.3 ± 1.76, 6.8 ± 0.83, 5.8 ± 1.18, and 6.1 ± 1.08 nmol/10 min/g, respectively. Each column represents the mean of five ganglia. *, significant difference between the values before and after ET-3 in the same ganglion (p < .05). , significant difference between untreated and ET antagonist-treated ganglia (p < .05). See the legend to Fig. 1 for additional explanation.

Effects of Neuronal NOS (nNOS) Inhibitor on ET Agonist-Induced Reduction of Acetylcholine Release. As shown in Fig. 3, the ET-3- and IRL-1620-induced inhibitions of acetylcholine release during preganglionic stellate stimulation were antagonized by incubation with 10⁵ M 3-bromo-7-nitroindazole. The application of 3-bromo-7-nitroindazole alone in the absence of ET-3 and IRL-1620 did not affect the acetylcholine release significantly (Fig. 3). Long-term potentiation in the rat dentate gyrus is inhibited by the inhibitory response of NOS activity to 3-bromo-7-nitroindazole at dose of 3 × 10⁵ M (Wu et al., 1997).

View larger version (38K): [in this window] [in a new window]   Fig. 3.   Antagonism by a nNOS inhibitor, 3-bromo-7-nitroindazole, of the ET-3- or IRL-1620-induced reduction in acetylcholine output elicited by preganglionic stimulation in untreated () and 3-bromo-7-nitroindazole (105 M)-treated (). In four groups of ganglia, the mean basal amount of acetylcholine released by preganglionic stimulation before drug was 6.1 ± 1.14, 7.3 ± 0.67, 6.5 ± 1.01 and 7.1 ± 1.09 nmol/10 min/g, respectively. *, significant difference between the values before and after ET-3 or IRL-1620 in the same ganglion (p < .05). , significant difference between untreated and 3-bromo-7-nitroindazole-treated ganglia (p < .05). Each column represents the mean of five ganglia.

5

Inhibitory Effects of 8-Bromo-cGMP and S-Nitroso-N-Acetylpenicillamine on Acetylcholine Output. As shown in Fig. 4, 8-bromo-cGMP, a membrane-permeable cGMP analog, and S-nitroso-N-acetylpenicillamine, a NO donor, at concentrations of 10⁸ to 10⁵ M and 10⁶ to 10³ M, respectively, decreased in a dose-dependent manner the acetylcholine release elicited by preganglionic stimulation. The reduction of the acetylcholine output at a maximum concentration of 10⁵ M by 8-bromo-cGMP was 42.9% and of 10³ M by S-nitroso-N-acetylpenicillamine 42.5%, respectively.

View larger version (19K): [in this window] [in a new window]   Fig. 4.   Concentration-response curves for inhibition by a membrane-permeable cGMP, 8-bromo-cGMP, and a NO donor, SNAP, of acetylcholine output elicited by preganglionic stimulation in 8-bromo-cGMP ()- and SNAP ()-treated ganglia. In two groups of ganglia, the mean basal amount of acetylcholine release by first preganglionic stimulation was 8.5 ± 1.96 and 6.5 ± 0.78 nmol/10 min/g, respectively. *, : significant difference between the values before and after the agents in the same ganglion (p < .05). Each point represents the mean of five ganglia. See the legend to Fig. 1 for an additional explanation.

Effects of ODQ on the ET-3-Induced Reduction of Acetylcholine Release. As shown in Fig. 5, the reduction of acetylcholine output induced by ET-3 at concentrations of 10⁹ to 10⁶ M was antagonized by incubation with a soluble guanylyl cyclase inhibitor, ODQ (10⁴ M). The agent, per se, at the concentration used did not influence the acetylcholine output elicited by preganglionic stimulation. ODQ at a dose of 10⁵ M abolishes an elevation of cGMP accumulation induced by sodium nitroprusside in guinea pig trachea (Hwang et al., 1998).

View larger version (52K): [in this window] [in a new window]   Fig. 5.   Antagonism by a selective inhibitor of soluble guanylyl cyclase, ODQ, of the ET-3-induced reduction in acetylcholine output elicited by preganglionic stimulation in untreated () and 104 M ODQ-treated (). In two groups of ganglia, the mean basal amount of acetylcholine released elicited by preganglionic stimulation before drug was 10.7 ± 1.97 and 7.3 ± 1.04 nmol/10 min/g, respectively. *, significant difference between the values before and after ET-3 in the same ganglion (p < .05). , significant difference between untreated and ODQ-treated ganglia (p < .05). Each column represents the mean of five ganglia. See the legend to Fig. 2 for an additional explanation.

Effects of ET Agonists on NO_x Levels. Fig. 6 shows the effects of ET-3 and IRL-1620 on the NO_x levels in the incubation medium. ET-3 and IRL-1620 at concentrations of 10⁵ and 10⁴ M or 10⁴ and 10³ M, respectively, induced a dose-dependent increase in the NO_x levels in the medium. From the result shown in Fig. 6, the ET-3- and IRL-1620-induced NO_x production at concentrations of 10⁴ and 10³ M, was 139.6% and 103.8% over basal levels, respectively.

View larger version (18K): [in this window] [in a new window]   Fig. 6.   Concentration-response curves for increment by ET agonists, ET-3 and IRL-1620, of NO metabolites (NO2 + NO3, NOx) output in ET-3 ()- and IRL-1620 ()-treated ganglia. In two groups of ganglia, the mean basal amount of NOx release for a 30-min incubation period was 273 ± 35.9 and 185 ± 31.4 pmol/20 µl/30 min/g, respectively. *, : significant difference between the values before and after the ET analog in the same ganglion (p < .05). Each point represents the mean of 12 and 9 ganglia.

Effects of BQ-788 on the ET Agonists-Induced NO_x Levels. As shown in Fig. 7, the ET-3- and IRL-1620-induced increases in the NO_x levels were completely antagonized by ET_B receptor antagonist, BQ-788, at a concentration of 10⁵ M, which did not affect the basal NO_x levels in the absence of either ET-3 or IRL-1620.

View larger version (23K): [in this window] [in a new window]   Fig. 7.   Antagonism by an ETB receptor antagonist, BQ-788 of the ET-3 or IRL-1620-induced increase in NO metabolites (NO2 + NO3, NOx) output in untreated () and 105 M BQ-788-treated (). In four groups of ganglia, the mean basal amount of NOx released before drugs was 413 ± 73.6, 572 ± 60.2, 315 ± 52.4, and 205 ± 36.5 pmol/20 µl/30 min/g, respectively. *, significant difference between values before and after ET-3 or IRL-1620 in the same ganglion (p < .05). , significant difference between untreated and BQ-788-treated ganglia (p < .05). Each column represents the mean of seven ganglia.

Effects of SKF-96365 or the Removal of Extracellular Ca²⁺ on ET-3-Induced NO_x Levels. The ET-3-induced increase in the NO_x levels was not only antagonized by a selective receptor-mediated Ca²⁺-entry inhibitor, SKF-96365 (10⁵ M) (Fig. 8) but also lowered the extracellular Ca²⁺ levels by incubating Ca²⁺-free Locke's solution (data not shown). SKF-96365 at a dose of 3 × 10⁵ M eliminates ET agonist-mediated [Ca²⁺]_i increase in bovine corneal epithelial (Tao et al., 1997).

View larger version (28K): [in this window] [in a new window]   Fig. 8.   Antagonism by a selective inhibitor of receptor-mediated Ca2+ entry, SKF-96365 of the ET-3-induced increase in NO metabolites (NO2 + NO3, NOx) output in untreated () and 105 M SKF-96365-treated () ganglia. In two groups of ganglia, the mean basal amount of NOx released before drugs was 421 ± 71.5 and 374 ± 65.3 pmol/20 µl/30 min/g, respectively. *, significant difference between the values before and after ET-3 in the same ganglion (p < .05). , significant difference between untreated and SKF-96365-treated ganglia (p < .05). Each column represents the mean of four ganglia.

Effects of 8-Bromo-7-Nitroindazole and W-7 on ET-3-Induced NO_x Levels. As shown in Fig. 9, the ET-3-induced increase in the NO_x levels was antagonized by a nNOS inhibitor, 8-bromo-7-nitroindazole, at 10⁵ M or a calmodulin antagonist, W-7, at 10⁴ M. These agents, per se, at the concentrations used did not influence the basal NO_x levels in the absence of ET-3. W-7 at doses of 5 × 10⁷ to 5 × 10⁴ M dose-dependently inhibited NO_x synthesis stimulated by ET-3 in cultured bovine endothelial cells (Hirata and Emori, 1993).

View larger version (18K): [in this window] [in a new window]   Fig. 9.   Antagonism by a nNOS inhibitor, 3-bromo-7-nitroindazole or calmodulin antagonist, W-7, of the ET-3-induced increase in the NO metabolites (NO2 + NO3, NOx) output in untreated (), 105 M 3-bromo-7-nitroindazole (), and 104 M W-7-treated () ganglia. In four groups of ganglia, the mean basal amount of NO metabolites (NO2 + NO3, NOx) released before the drugs was 196 ± 46.9, 196 ± 66.5, 170 ± 41.1, and 169 ± 87.6 pmol/20 µl/30 min/g, respectively. *, significant difference between the values before and after ET-3 in the same ganglion (p < .05). , indicates a significant difference between untreated and 3-bromo-7-nitroindazole- or W-7-treated ganglia (p < .05). Each column represents the mean of seven and five ganglia.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

In the current study, the contribution of ET receptor subtype specificity and the NO system to the presynaptic inhibition induced by ET agonists of the dog stellate ganglionic transmission were investigated. According to the experiment using the acetylcholine assay as described previously (Ohjimi et al., 1994; Kushiku et al., 1995), there were no changes in the elicited output of acetylcholine during eight successive preganglionic stimulations with 10-min intervals. Under such experimental conditions, the output of acetylcholine elicited by the preganglionic stimulation was decreased by exposure of the isolated stellate ganglia to ET-3 at 10⁹ to 10⁶ M (Kushiku et al., 1995). In the present study, ET-3 at the same concentrations in as those used in our previous report (Kushiku et al., 1995), and ET_B receptor agonist, IRL-1620, at 10⁸ to 10⁵ M also inhibited the output of acetylcholine from the isolated stellate ganglia. Therefore, the efficacy of IRL-1620 was lower than that of ET-3.

ET has also been suggested to exert an inhibitory action on the neurotransmission of the preganglionic and postganglionic nerve terminals. ET inhibits the nerve stimulation-induced contractile responses and fractional release of [³H]acetylcholine from the guinea pig ileum as well as [³H]norepinephrine from the guinea pig femoral artery (Wiklund et al., 1988; Wiklund et al., 1989). ET-3 also inhibits the sympathetic ganglionic transmission by reducing the acetylcholine release at preganglionic terminals (Kushiku et al., 1995). However, the specificity of the receptor subtype for the ET-induced inhibition of the sympathetic neurotransmission has yet to be determined. In the present study, the nonspecific ET_A and ET_B receptor antagonist, bosentan, and the specific ET_B receptor antagonist, BQ-788, but not the ET_A receptor antagonist, BQ-123, completely antagonized the ET-3- and IRL-1620-induced inhibition of acetylcholine release elicited by preganglionic stellate stimulation. These results indicate the inhibition induced by ET analog to be mediated by the ET_B receptor activation at the ganglion.

NOS-immunoreactivity is distributed in the preganglionic sympathetic neurons of rat pre- and paravertebral (Anderson et al., 1993; Blottner and Baumgarten, 1992) and superior cervical (Okamura et al., 1995) ganglion. In contrast, NOS is also localized in a population of sympathetic postganglionic neurons in the guinea pig paravertebral and inferior mesenteric ganglia (Höhler et al., 1995) and in the cat stellate and lower lumbar ganglia (Anderson et al., 1995). Furthermore, both nNOS and endothelial NOS are constitutively expressed in neuronal tissue (Dinerman et al., 1994). Hypoxia increases the nNOS mRNA expression in the nodose ganglion and cerebellum, respectively, whereas it had no significant effect on the endothelial NOS levels (Prabhakar et al., 1996). Therefore, these results indicate that nNOS is present in the pre- and/or postganglionic sympathetic neurons.

In the rabbit perfused kidney, the increase of perfusion pressure induced by ET-1 was potentiated by BQ-788, a selective ET_B receptor antagonist and by N^G-nitro-L-arginine methyl ester, a NOS inhibitor (D'Orléans-Juste et al., 1994). The inhibition of NOS with N^G-monomethyl-L-arginine completely abolished the renal vasodilation induced by ET-1 and ET-3 and ET-3-displayed diuresis and natriuresis in the dogs (Chou and Porush, 1995). Furthermore, in the whole rat adrenal medulla, ET-1 and ET-3 stimulate NO-induced cGMP generation through ET_B-receptor activation (Mathison and Israel, 1998). In the present experiment on the sympathetic ganglion, the preganglionic inhibition by ET-3 or ET_B receptor agonist, IRL-1620, of the acetylcholine output elicited by preganglionic stimulation was also inhibited by the treatment with 3-bromo-7-nitroindazole, a selective nNOS inhibitor. On the other hand, the exposure of these ET analogs increased the levels of total NO metabolites in the incubation medium, and this increase disappeared after treatment with BQ-788 and 3-bromo-7-nitroindazole. Furthermore, weak efficacy of IRL-1620 than ET-3 on the release of NO metabolites is well correlated with that on affecting the acetylcholine release in the present study. These facts suggest that the increase of NO formation mediated by ET_B receptor activation may participate in the ET-induced preganglionic inhibition in the canine stellate ganglion.

Endogenous NO is produced from L-arginine by NOS, thus resulting in the stoichiometric production of L-citrulline (Mayer et al., 1989; Palmer and Moncada, 1989). The constitutive NOS exists in endothelial cells (Mayer et al., 1989; Busse and Mülsch, 1990) and neuronal cells (Bredt et al., 1990), which is Ca²⁺-calmodulin- and NADPH-dependent (Palmer and Moncada, 1989). NO identified in many tissues affects a soluble guanylyl cyclase (Arnold et al., 1977; Murad et al., 1978), and binds tightly to the heme region of cyclase and causes an increase in the cGMP levels, which thus affects the ion channel (Rascón et al., 1992). In the current study, the preganglionic inhibition by ET agonists of acetylcholine release elicited by preganglionic stellate stimulation was inhibited by treatment with ODQ, a selective inhibitor of soluble guanylyl cyclase, and this inhibition was mimicked by SNAP, an NO donor, and 8-bromo-cGMP, a membrane-permeable cGMP. In addition, ET agonists stimulated the production of NO_x, and this accumulation was inhibited by W-7, a calmodulin antagonist, SKF-96365, a selective inhibitor of receptor-mediated Ca²⁺ entry and incubating Ca²⁺-free medium. SKF-96365 inhibits a receptor-mediated Ca²⁺ entry as well as voltage-gated Ca²⁺ entry without affecting the internal Ca²⁺ release at the concentrations used in the present experiment, which were the same as that used in a previous report (Merritt, 1990). The results indicate that ET agonists stimulate the receptor-mediated Ca²⁺ entry and Ca²⁺-calmodulin-dependent formation of NO metabolites activating the soluble guanylyl cyclase, thus leading to an increased cGMP content in the dog stellate ganglia.

In conclusion, ET inhibits sympathetic ganglionic transmission at presynaptic sites via ET_B receptor by reducing the output of acetylcholine release through the stimulation of the NO production and cGMP pathways.