- Email: [email protected]
Investigation of nicotine-induced relaxation of circular smooth muscle of the guinea-pig gastric fundus
European JournalofPharmacology, 241 (1993) 171-175 © 1993 Elsevier Science Publishers B.V. AI1 rights reserved0014-2999/93/$06.00
171
EJP 53265
Investigation of nicotine-induced relaxation of circular smooth muscle of the guinea-pig gastric fundus Shu-ichi Kojima, R i k a Ishizaki and Y a s u o S h i m o Department of Pharmacology, Dokkyo University School of Medicine, Mibu, Tochigi321-02, Japan
Received 18 March 1993,revised MS received 16 June 1993, accepted 29 June 1993
A possible mechanism for the nicotine-induced relaxation of circular muscle strips of the guinea-pig gastric fundus was investigated. In the presence of atropine (0.2 /~M), nicotine produced concentration-dependent relaxation with a maximum effect at 100 p.M (mean pECs0 value, 4.60). The maximum relaxation due to nicotine was greatly reduced by pretreatment with tetrodotoxin (0.3/zM) or hexamethonium (10/xM), but not with metitepine (0.3/~M). Combined pretreatment with timolol (0.3 /zM) and phentolamine (0.3 /zM) or chemical sympathectomy by 6-hydroxydopamine pretreatment partially inhibited the nicotine-induced relaxation, a-Chymotrypsin (2 u/ml) which abolished the equivalent relaxation induced by vasoactive intestinal polypeptide (VIP) had no effect on nicotine-induced relaxation. NG-Nitro-L-arginine (L-NNA) and NG-nitro-L-arginine methyl ester (L-NAME) caused a concentration-dependent inhibition of the nicotine-induced relaxation (98% inhibition at 10/~M of L-NNA), but had no effect on sodium nitroprusside- or noradrenaline-induced relaxation. The inhibitory effect of L-NNA or L-NAME was reversed completely by L-arginine (3 mM), but not by D-arginine (3 mM). From these results, we concluded that nicotine-induced relaxation of the guinea-pig gastric fundus is mediated largely by the release of nitric oxide or a related substance and partially by the release of noradrenaline. Possible contributions of 5-hydroxytryptamine or VIP to the nicotine-induced relaxation appear to be negligible. Nicotine-induced relaxation; Nitric oxide (NO); Noradrenaline; Gastric fundus (guinea-pig)
1. Introduction
The gastric fundus has an inhibitory non-adrenergic non-cholinergic (NANC) innervation and there is some evidence suggesting that the NANC neurotransmitter may be vasoactive intestinal polypeptide (VIP) in the guinea-pig (Grider et al., 1985), rat (De Beurme and Lefebvre, 1988) and cat (D'Amato et al., 1988). However, incomplete blockade of NANC-mediated relaxation by VIP antiserum suggests that other components may be involved (De Beurme and Lefebvre, 1988). There is evidence to support the idea that nitric oxide (NO) or a related substance of the L-argin i n e / N O pathway functions as an NANC neurotransmitter in the gastric fundus of the rat (Boeckxstaens et al., 1991) and guinea-pig (Lefebvre et al., 1992). As another candidate, however, 5-hydroxytryptamine (5HT) might also be a neurotransmitter of NANC nerves (Burnstock, 1986). Nicotine has been reported to cause relaxation of gastrointestinal smooth muscle through stimulation of
Correspondence to: Shu-ichi Kojima, Department of Pharmacology, Dokkyo University School of Medicine, Mibu, Tochigi 321-02 Japan.
nicotinic receptors located on NANC inhibitory nerves (Daniel, 1982), but exact mechanisms of the response have not been determined. We have recently shown that exogenously applied 5-HT produces relaxations of circular muscles of the guinea-pig gastric fundus by stimulating muscular 5 - H T r l i k e receptors (Kojima et al., 1992). It was thus of interest to investigate whether the relaxation induced by nicotine in the guinea-pig gastric fundus is mediated by NO, VIP or 5-HT. We have now examined the effects of potent NO synthase inhibitors, NG-nitro-L-arginine (L-NNA) and NG-nitro-L-arginine methylester (L-NAME) (Moncada et al., 1991), VIP degrading enzyme, a-chymotrypsin, and the 5-HTl-like receptor antagonist, metitepine (Kojima et al., 1992), on the nicotine-induced relaxation.
2. Materials and methods 2.1. Tissue preparation
Male guinea-pigs, weighing 300-600 g, were killed by a blow on the head and the stomach was rapidly
172
taken out and placed in a physiological salt solution (PSS) of the following composition (mM): NaCI 136.8; KC1, 2.7; CaC12, 1.8; MgC12, 1.05; NaH2PO4, 0.42; NaHCO3, 11.9; glucose 5.56; ascorbic acid 0.12 (pH 7.4). Mucosa-free strips (approximately 2.0 × 0.5 cm) of circular muscles were dissected from the guinea-pig gastric fundus. The strips were then suspended under a 0.5-g load in 15-ml organ baths filled with a PSS at 37°C and bubbled with 5% CO2-95% 0 2. The PSS always contained atropine (0.2 ttM) for the elimination of cholinergically mediated responses. The tissues were allowed to equilibrate for at least 90 min before the start of experiments. Changes in mechanical activity of the tissue were recorded isotonically by means of an isotonic transducer (Nihon Kohden, TD-112S) and a Nihon Kohden recticoder (RJG3006).
2.2. Experimental protocols After an equilibration period of 90 min, a given concentration of nicotine was added to the organ bath for a contact time of 5 min followed by repeated washing with fresh PSS. After the 60 to 70-min interval which was required to prevent desensitisation, the same concentration of nicotine was added again in the absence or presence of antagonists. In experiments when cumulative concentration-response curves for the relaxation response to sodium nitroprusside or noradrenaline were obtained, cumulatively increasing concentrations of these agonists were added to the organ bath. After the 40-min interval, cumulative concentration-response curves for each agonist were again obtained in the presence of antagonists. When the curves for noradrenaline were made, the tissue was always pretreated with yohimbine (1 /zM) to eliminate the a2-adrenoceptor-mediated contraction (Sahyoun et al., 1982). The effects of antagonists on agonist-induced relaxation were evaluated by comparing the response before and after the addition of antagonists in the same preparation. Antagonists were incubated for at least 30 min prior to the addition of agonists. The relaxation response was expressed as a percentage of the maximal relaxation in response to papaverine (30 /zM) which was observed at the end of the experiments. The percent relaxations were plotted as mean values to obtain log concentration-response curves. For chemical sympathectomy, some guinea-pigs were pretreated intraperitoneally (i.p.) injected with 6-hydroxydopamine 100 mg/kg at 48 h and 250 mg/kg at 24 h before the animals were killing (Kirkpatrick and Burnstock, 1987). A batch-matched control group received an i.p. injection of the same volume of vehicle for 6-hydroxydopamine (ascorbic acid, 1 mg/ml) at the same intervals.
2.3. Data analysis The negative log molar concentrations of agonists causing 50% of the maximum response (pECs0) were determined from each concentration-response curve according to the method of Van Rossum (1963). Means + S.E.M. of n experiments are given throughout the paper. The paired t-test was used to compare mean values, and a probability (P) of less than 5% was considered to be significant.
2. 4. Drugs The drugs used were atropine sulphate, hexamethonium chloride dihydrate, papaverine hydrochloride and sodium nitroprusside (Wako, Osaka); a-chymotrypsin (Type II), 6-hydroxydopamine hydrobromide, N Gnitro-L-arginine, NG-nitro-L-arginine methyl ester, noradrenaline bitartrate, timolol maleate and yohimbine hydrochloride (Sigma Chemical Co. Ct. Louis); Larginine, D-arginine and nicotine tartrate (Nacalai, Kyoto); metitepine maleate (Nihon Roche, Tokyo); phentolamine hydrochloride (Nihon Ciba-Geigy, Takarazuka); tetrodotoxin (Sankyo, Tokyo); vasoactive intestinal polypeptide (Peptide Institute, Osaka). The molar concentrations of drugs described in this paper refer to the final bath concentration.
3. Results
3.1. Effects of nicotine The circular muscle strip of the guinea-pig gastric fundus has sufficient resting tone to produce relaxation (Kojima et al., 1992). In the presence of atropine (0.2 /zM), nicotine (10-100 /zM) produced concentrationdependent relaxation with a mean pECs0 value of 4.60 and a mean maximum effect (Em~x) at 100 /~M, amounting to 53.3 + 5.1% (n = 8) of the papaverine-induced maximal relaxation. A higher concentration (300 ~M) of nicotine produced no further relaxation. Nicotine 100 ~M evoked a reproducible and rapid relaxation response which, after reaching a peak, returned spontaneously to the basal tone (fig. 1A). None of the antagonists investigated had a significant influence on basal tone. As shown in table 1, the relaxant response to nicotine (100 /zM) was not affected by metitepine (0.3/zM), but was significantly reduced by tetrodotoxin (0.3 IzM, 84.2 + 3.5% inhibition), hexamethonium (10/xM, 83.8 + 3.4% inhibition) and the combination of timolol (0.3 t~M) and phentolamine (0.3 tzM) (37.8 + 4.1% inhibition). a-Chymotrypsin (2 u / m l ) completely abolished the relaxant response to VIP (0.1/zM) which amounted to 48.1 + 5.9% (n = 6) of that of papaverine (30/xM), but
173
did not significantly influence the nicotine-induced relaxation. As shown in table 2, L-NNA and L-NAME significantly inhibited the nicotine (100 /~M)-induced relaxation in a concentration-dependent manner, and at the higher concentrations used, L-NNA (10 /zM) and L-NAME (100 ~M), inhibited the response by 98.1 + 0.9% and 86.5 + 5.2% respectively. Neither Larginine (3 mM) nor D-arginine (3 mM) affected the nicotine-induced relaxation (n = 3, results not shown). L-Arginine (3 mM), but not D-arginine (3 mM), completely prevented the inhibitory effect of L-NNA (10
NC 1 0 0
TABLE 1 Effects of antagonists on the relaxant response to nicotine (100/~M) of circular muscle strips of the guinea-pig gastric fundus. The data are means + S.E.M. from (n) of experiments. Atropine (0.2 /~M) was present in all experiments. Antagonists
Saline Tetrodotoxin Hexamethonium Timolol/phentolamine Metitepine a-Chymotrypsin
n
0.3/zM 10 /zM 0.3/zM 0.3 p.M 2.0 u / m l
5 6 6 6 6 5
Nicotine-induced relaxation (%) a Before treatment
After treatment
47.5+ 3.1 48.6+4.5 60.1+ 2.9 54.4 + 5.2 56.0 + 4.0 47.1_+3.0
50.0 + 3.0 8.2+2.0 a 10.1 -I-2.6 b 34.7 + 5.3 c 50.0 + 4.3 37.4 + 5.9
a Relaxations relative to those induced by papaverine 30 ~M. Significantly different from control (before treatment), b p < 0.05, c P < 0.001.
/~M) and L-NAME (10 /zM) on the nicotine-induced relaxation (fig. 1 and table 2). In fundic strips from 6-hydroxydopamine-pretreated animals, 100 /~M nicotine evoked relaxation amounting to 35.7 _+ 4.1% (n = 7) of that produced by papaverine (30 /.tM). The response was significantly smaller than that in the preparation from vehicle-pretreated animals (48.3 + 3.3%, n -- 7, P < 0.05).
Saline NC 1OO
3.2. Effects of sodium nitroprusside and noradrenaline
m
L-NNA 10
The NO donor, sodium nitroprusside (3-1000 nM) produced a concentration-dependent relaxation of the fundic strips. The response was not affected by prein-
NC 1 0 0
TABLE 2 Effects of N°-nitro-L-arginine (L-NNA) and NG-nitro-L-argininemethyl ester (L-NAME) on the relaxant response to nicotine (100 /~M) of circular muscle strips of the guinea-pig gastric fundus in the absence or presence of L- or D-arginine (3 mM).
k-Arg 3000/L-NNA NC 1 0 0
The data are means + S.E.M. from n experiments. Atropine (0.2/zM) was present in all experiments. Compound
n
D-Arg 30OO/L-NNA
I.| 5mln Fig. 1. Typical tracings of the relaxation in response to nicotine (100 ~M) of circular muscle strips of the guinea-pig gastric fundus in the absence (left panels) and presence (right panels) of vehicle (A), N G-nitro-L-arginine (L-NNA 10/.~M, B), L-NNA + L-arginine (L-Arg 3 mM, C) or L-NNA+ D-arginine (D-Arg 3 mM, D). L-Arginine, but not D-arginine, completely reversed the inhibitory effect of L-NNA on the nicotine-induced relaxation.
L-NNA 1/zM L-NNA 10 ~M L-NNA 10/.t M/L-arginine L-NNA 10/~M/D-arginine L-NAME 10/~M L-NAME 100/.~M L-NAME 10/z M/L-arginine L-NAME 10 p.M/D-arginine
6 6 6 6 6 6 6 6
Nicotine-induced relaxation (%) a Before treatment
After treatment
50.5+4.7 53.0+4.6 58.9 + 3.4 59.2 + 7.3 54.4+3.4 51.5+2.9 52.4 + 3.6 51.5 + 5.7
18.9+7.4 b 0.9+0.5 ¢ 55.1 + 3.1 2.9 + 1.7 ¢ 10.8+4.1 c 6.4+2.2 ¢ 49.1 + 4.9 5.7 + 2.3
a Relaxations relative to those induced by papaverine 30/zM. Significantly different from control (before treatment), b p < 0.01, c p < 0.001.
174 TABLE 3 Effects of antagonists on the relaxant response to sodium nitroprusside (3-1000 nM) of circular muscle strips of the guinea-pig gastric fundus. Emax values (percentage for the effect of papaverine, 30 ~M) and pECs0 values ( - log M) are listed. The data are means5:S.E.M. from n experiments. Atropine (0.2/~M) was present in all experiments. Antagonists Control L-NNA Timolol/phentolamine Metitepine
n 16 100 /~M 6 0.3/zM 6 0.3 p.M 4
pECs0 7.19+0.09 7.30+0.14 7.055:0.08 7.555:0.21
Emax (%)
85.4_+2.5 93.65:1.9 91.85:2.6 90.45:4.3
cubation of the tissue with L-NNA (100 /zM), timolol (0.3 /xM)/phentolamine (0.3 /~M) or metitepine (0.3 /~M), as shown by the identical Emax and pECs0 values (table 3). The Emax and pECs0 values for sodium nitroprusside in the fundic strips from 6-hydroxydopamine-pretreated animals were 83.3 + 3.2% and 7.25 + 0.14 (n = 7), which was comparable with the control values, 88.2 ± 3.0% and 7.31 + 0.12 (n = 7). 6Hydroxydopamine pretreatment had no significant effect on the relaxant response to sodium nitroprusside. The involvement of NO or a related substance in the relaxation induced by noradrenaline was also examined. In the presence of yohimbine (1 /~M), noradrenaline ( 0 . 1 - 3 0 / x M ) produced a concentration-dependent relaxation with a mean pECs0 value of 6.57 + 0.23 and E m a x a t 1 /~M, amounting to 68.5 + 5.1% (n = 4) of the effect of papaverine (30/xM). The Emax and pECs0 values were not affected by the pretreatment with L-NNA (100/~M) (n = 4).
4. Discussion
Nicotine produces a concentration-dependent relaxation of circular muscle strips of the guinea-pig gastric fundus under resting tone. The response is probably mediated by the stimulation of nicotinic cholinoceptors located in enteric inhibitory neurons, because it was inhibited by both hexamethonium or tetrodotoxin. However, these antagonists did not completely abolish the response. Hence, a non-neuronal component might be partially involved in the nicotine-induced relaxation. Although the exact nature of the latter components remains unknown, nicotine is known to elicit a T r X resistant response in several peripheral organs (Takayanagi et al., 1984; Hisayama et al., 1988). There are two types of inhibitory neurons that are known to affect the smooth muscle motility of the gastrointestinal tract, noradrenergic neurons and NANC neurons. In the present study, one of the inhibitory neurotransmitters responsible for the nicotine-induced relaxation was likely to be noradrenaline,
because the relaxation was partially reduced either by the combination of timolol, a /3-adrenoceptor antagonist and phentolamine, an a-adrenoceptor antagonist, or by chemical sympathectomy with 6-hydroxydopamine. A similar result was observed by Costall et al. (1983) who showed that the relaxation of circular muscle strips of the cardia and body of the guinea-pig stomach induced by electrical field stimulation were partially antagonized by pretreatment with reserpine, propranolol or phentolamine. T h e r e is now good evidence that NO-like substance(s) and VIP might be inhibitory NANC neurotransmitters in the mammalian gastric fundus (De Beurme and Lefebvre, 1988; Li and Rand, 1990; Boeckxstaens et al., 1991; Lefebvre et al., 1992). In the present study, L-NNA and L-NAME, inhibitors of NO synthase (Moncada, et al 1991), inhibited the nicotineinduced relaxation in a concentration-dependent manner, and at 10 tzM L-NNA abolished the response. The inhibitory effects of L-NNA and L-NAME were stereospecifically prevented by L-arginine, a substrate for NO synthase. These findings indicate that the relaxant action of nicotine on the guinea-pig gastric fundus is mediated entirely by the release of NO or a related substance from the L - a r g i n i n e / N O pathway. Our resuits are comparable with the finding by Irie et al. (1991) that NO or a related substance is involved in the NANC relaxation induced by nicotine in the rat duodenum. Yamamoto et al. (1993) recently demonstrated that L-NNA reduces the noradrenaline overflow in response to electrical field stimulation of the isolated rat mesenteric vasculature, suggesting that NO or a related substance enhances the release of noradrenaline from perivascular adrenergic nerves. This may account for our present finding that the relaxant action of nicotine was partially inhibited by chemical sympathectomy or a- and fl-adrenoceptor blockade. However, we cannot exclude the possibility that nicotine stimulates not only inhibitory nerves but also excitatory NANC nerves in myenteric plexus, since the guinea-pig stomach is innervated by intramural nerves containg 5-HT, VIP, substance P, calcitonin gene-related peptide and neuropeptide Y (Mawe et al., 1989). The elimination of adrenergic components by adrenoceptor antagonists or chemical sympathectomy would enhance the release of excitatory neurotransmitters, which in turn may reduce the relaxation mediated by NO. Further studies will be required to test this possibility. It is unlikely that noradrenaline released by nicotine could enhance the NO production since the relaxation in response to exogenously applied noradrenaline was not affected by LNNA. The possibility that VIP plays an important role in the relaxation due to nicotine is also unlikely, because a-chymotrypsin, at a concentration which abolished the
175
response to exogenous VIP, did not significantly influence the response. In a previous report, we demonstrated that exogenously applied 5-HT produces relaxation of circular muscles of the guinea-pig gastric fundus by stimulating 5-HTl-like receptors located on smooth muscles (Kojima et al., 1992), and 5-HT has also been proposed as a putative NANC neurotransmitter (Burnstock, 1986). However, it is also unlikely that 5-HT plays an important role in the relaxation in response to nicotine, because the potent 5-HTl-like receptor antagonist, metitepine (Kojima et al., 1992), failed to inhibit the response. In conclusion, nicotine produced relaxation of circular muscle strips of the guinea-pig gastric fundus by two mechanism: mainly by the release of NO or a related substance from the L-arginine/NO pathway, and partially by the release of noradrenaline from adrenergic nerves. Further contributions of 5-HT or VIP to nicotine-induced relaxation are apparently negligible. References Boeckxstaens, G.E., P.A. Pelckmans, J.J. Bogers, H. Bult, J.G. De Man, L. Oosterbosch, A.J. Herman and Y.M. Van Maercke, 1991, Release of nitric oxide upon stimulation of nonadrenergic non-cholinergic nerves in the rat gastric fundus, J. Pharmacol. Exp. Ther. 256, 441. Burnstock, G., 1986, The non-adrenergic non-cholinergic nervous system, Arch. Int. Pharmacodyn. 280 (Suppl.), 1. Costall, B., R.J. Naylor and C.C.W. Tan, 1983, Field stimulation-induced responses of circular smooth muscle from guinea-pig stomach, Naunyn-Schmied. Arch. Pharmacol. 323, 155. D'Amato, M., F.A. De Beurme and R.A. Lefebvre, 1988, Comparison of the effect of vasoactive intestinal polypeptide and nonadrenergic non-cholinergic neurone stimulation in the cat gastric fundus, Eur. J. Pharmacol. 152, 71. Daniel, E.E, 1982, Pharmacology of adrenergic, cholinergic, and drugs acting on other receptors in gastrointestinal muscle, in: Mediators and Drugs in Gastrointestinal Motility, II, ed. G. Bertaccini (Springer-Verlag, Berlin, Heidelberg, New York) p. 249.
De Beurme, F.A. and R.A. Lefebvre, 1988, Vasoactive intestinal polypeptide as possible mediator of relaxation in the rat gastric fundus, J. Pharm. Pharmacol. 40, 711. Grider, J.R., M.B. Cable, 5.1. Said and M. Makhlouf, 1985, Vasoactive intestinal peptide as a neural mediator of gastric relaxation, Am. J. Physiol. 248, G73. Hisayama, T., M. Shinkai, I. Takayanagi and T. Toyoda, 1988, Mechanisms of action of nicotine in isolated urinary bladder of guinea-pig, Br. J. Pharmacol. 95, 465. Irie, K., T. Muraki, K. Furukawa and T. Nomoto, 1991, L-NG-Nitro L-arginine inhibits nicotine-induced relaxation of isolated rat duodenum, Eur, J. Pharmacol. 202, 285. Kirkpatrick, K and G. Burnstock, 1987, Sympathetic nerve-mediated release of ATP from the guinea-pig vas deferens is unaffected by reserpine, Eur. J. Pharmacol. 138, 207. Kojima, S., R. lshizaki and Y. Shimo, 1992, Investigation into the 5-hydroxytryptamine-induced relaxation of the circular smooth muscle of guinea-pig stomach fundus, Eur. J. Pharmacol. 224, 45. Lefebvre, R.A., E. Baert and A.J. Barbier, 1992, Influence of N ~nitro-L-arginine on non-adrenergic non-cholinergic relaxation in the guinea-pig gastric fundus, Br. J. Pharmacol. 106, 173. Li, C.C and M.J. Rand, 1990, Nitric oxide and vasoactive intestinal polypeptide mediate non-adrenergic, non-cholinergic inhibitory transmission to smooth muscle of the rat gastric fundus, Eur. J. Pharmacol. 191,303. Mawe, G.M., M. Schemann, J.D. Wood and M.D. Gershon, 1989, Immunocytochemical analysis of potential neurotransmitters present in the myenteric plexus and muscular layers of the corpus of the guinea-pig stomach, Anat. Rec. 224, 431. Moncada, S., R.M.J. Palmer and E.A. Higgs, 1991, Nitric oxide: Physiology, pathophysioiogy and pharmacology, Pharmacol. Rev. 43, 109. Sahyoun, H.A., B. Costall and R.J. Naylor, 1982, On the ability ot domperidone to selectively inhibit catecholamine-induced relaxation of circular smooth muscle of guinea-pig stomach, J. Pharm.Pharmacol. 34, 27. Takayanagi, I., Y. Kizawa and T. Hiruta, 1984, Tetrodotoxin-resistant response to nicotine in rabbit bronchial preparation, Eur. J. Pharmacol. 104, 351. Van Rossum, J.M, 1963, Cumulative dose-response curves. II. Technique for the making of dose-response curves in isolated organs and the evaluation of drug parameters, Arch. Int. Pharmacodyn. 143, 299. Yamamoto, R., A. Wada, Y. Asada, H. Niina and A. Sumiyoshi, 1993, NC-Nitro-L-arginine, an inhibitor of nitric oxide synthesis, decreases noradrenaline outflow in rat isolated perfused mesenteric vasculature, Naunyn-Schmied. Arch. Pharmacol. 347, 238.
171
EJP 53265
Investigation of nicotine-induced relaxation of circular smooth muscle of the guinea-pig gastric fundus Shu-ichi Kojima, R i k a Ishizaki and Y a s u o S h i m o Department of Pharmacology, Dokkyo University School of Medicine, Mibu, Tochigi321-02, Japan
Received 18 March 1993,revised MS received 16 June 1993, accepted 29 June 1993
A possible mechanism for the nicotine-induced relaxation of circular muscle strips of the guinea-pig gastric fundus was investigated. In the presence of atropine (0.2 /~M), nicotine produced concentration-dependent relaxation with a maximum effect at 100 p.M (mean pECs0 value, 4.60). The maximum relaxation due to nicotine was greatly reduced by pretreatment with tetrodotoxin (0.3/zM) or hexamethonium (10/xM), but not with metitepine (0.3/~M). Combined pretreatment with timolol (0.3 /zM) and phentolamine (0.3 /zM) or chemical sympathectomy by 6-hydroxydopamine pretreatment partially inhibited the nicotine-induced relaxation, a-Chymotrypsin (2 u/ml) which abolished the equivalent relaxation induced by vasoactive intestinal polypeptide (VIP) had no effect on nicotine-induced relaxation. NG-Nitro-L-arginine (L-NNA) and NG-nitro-L-arginine methyl ester (L-NAME) caused a concentration-dependent inhibition of the nicotine-induced relaxation (98% inhibition at 10/~M of L-NNA), but had no effect on sodium nitroprusside- or noradrenaline-induced relaxation. The inhibitory effect of L-NNA or L-NAME was reversed completely by L-arginine (3 mM), but not by D-arginine (3 mM). From these results, we concluded that nicotine-induced relaxation of the guinea-pig gastric fundus is mediated largely by the release of nitric oxide or a related substance and partially by the release of noradrenaline. Possible contributions of 5-hydroxytryptamine or VIP to the nicotine-induced relaxation appear to be negligible. Nicotine-induced relaxation; Nitric oxide (NO); Noradrenaline; Gastric fundus (guinea-pig)
1. Introduction
The gastric fundus has an inhibitory non-adrenergic non-cholinergic (NANC) innervation and there is some evidence suggesting that the NANC neurotransmitter may be vasoactive intestinal polypeptide (VIP) in the guinea-pig (Grider et al., 1985), rat (De Beurme and Lefebvre, 1988) and cat (D'Amato et al., 1988). However, incomplete blockade of NANC-mediated relaxation by VIP antiserum suggests that other components may be involved (De Beurme and Lefebvre, 1988). There is evidence to support the idea that nitric oxide (NO) or a related substance of the L-argin i n e / N O pathway functions as an NANC neurotransmitter in the gastric fundus of the rat (Boeckxstaens et al., 1991) and guinea-pig (Lefebvre et al., 1992). As another candidate, however, 5-hydroxytryptamine (5HT) might also be a neurotransmitter of NANC nerves (Burnstock, 1986). Nicotine has been reported to cause relaxation of gastrointestinal smooth muscle through stimulation of
Correspondence to: Shu-ichi Kojima, Department of Pharmacology, Dokkyo University School of Medicine, Mibu, Tochigi 321-02 Japan.
nicotinic receptors located on NANC inhibitory nerves (Daniel, 1982), but exact mechanisms of the response have not been determined. We have recently shown that exogenously applied 5-HT produces relaxations of circular muscles of the guinea-pig gastric fundus by stimulating muscular 5 - H T r l i k e receptors (Kojima et al., 1992). It was thus of interest to investigate whether the relaxation induced by nicotine in the guinea-pig gastric fundus is mediated by NO, VIP or 5-HT. We have now examined the effects of potent NO synthase inhibitors, NG-nitro-L-arginine (L-NNA) and NG-nitro-L-arginine methylester (L-NAME) (Moncada et al., 1991), VIP degrading enzyme, a-chymotrypsin, and the 5-HTl-like receptor antagonist, metitepine (Kojima et al., 1992), on the nicotine-induced relaxation.
2. Materials and methods 2.1. Tissue preparation
Male guinea-pigs, weighing 300-600 g, were killed by a blow on the head and the stomach was rapidly
172
taken out and placed in a physiological salt solution (PSS) of the following composition (mM): NaCI 136.8; KC1, 2.7; CaC12, 1.8; MgC12, 1.05; NaH2PO4, 0.42; NaHCO3, 11.9; glucose 5.56; ascorbic acid 0.12 (pH 7.4). Mucosa-free strips (approximately 2.0 × 0.5 cm) of circular muscles were dissected from the guinea-pig gastric fundus. The strips were then suspended under a 0.5-g load in 15-ml organ baths filled with a PSS at 37°C and bubbled with 5% CO2-95% 0 2. The PSS always contained atropine (0.2 ttM) for the elimination of cholinergically mediated responses. The tissues were allowed to equilibrate for at least 90 min before the start of experiments. Changes in mechanical activity of the tissue were recorded isotonically by means of an isotonic transducer (Nihon Kohden, TD-112S) and a Nihon Kohden recticoder (RJG3006).
2.2. Experimental protocols After an equilibration period of 90 min, a given concentration of nicotine was added to the organ bath for a contact time of 5 min followed by repeated washing with fresh PSS. After the 60 to 70-min interval which was required to prevent desensitisation, the same concentration of nicotine was added again in the absence or presence of antagonists. In experiments when cumulative concentration-response curves for the relaxation response to sodium nitroprusside or noradrenaline were obtained, cumulatively increasing concentrations of these agonists were added to the organ bath. After the 40-min interval, cumulative concentration-response curves for each agonist were again obtained in the presence of antagonists. When the curves for noradrenaline were made, the tissue was always pretreated with yohimbine (1 /zM) to eliminate the a2-adrenoceptor-mediated contraction (Sahyoun et al., 1982). The effects of antagonists on agonist-induced relaxation were evaluated by comparing the response before and after the addition of antagonists in the same preparation. Antagonists were incubated for at least 30 min prior to the addition of agonists. The relaxation response was expressed as a percentage of the maximal relaxation in response to papaverine (30 /zM) which was observed at the end of the experiments. The percent relaxations were plotted as mean values to obtain log concentration-response curves. For chemical sympathectomy, some guinea-pigs were pretreated intraperitoneally (i.p.) injected with 6-hydroxydopamine 100 mg/kg at 48 h and 250 mg/kg at 24 h before the animals were killing (Kirkpatrick and Burnstock, 1987). A batch-matched control group received an i.p. injection of the same volume of vehicle for 6-hydroxydopamine (ascorbic acid, 1 mg/ml) at the same intervals.
2.3. Data analysis The negative log molar concentrations of agonists causing 50% of the maximum response (pECs0) were determined from each concentration-response curve according to the method of Van Rossum (1963). Means + S.E.M. of n experiments are given throughout the paper. The paired t-test was used to compare mean values, and a probability (P) of less than 5% was considered to be significant.
2. 4. Drugs The drugs used were atropine sulphate, hexamethonium chloride dihydrate, papaverine hydrochloride and sodium nitroprusside (Wako, Osaka); a-chymotrypsin (Type II), 6-hydroxydopamine hydrobromide, N Gnitro-L-arginine, NG-nitro-L-arginine methyl ester, noradrenaline bitartrate, timolol maleate and yohimbine hydrochloride (Sigma Chemical Co. Ct. Louis); Larginine, D-arginine and nicotine tartrate (Nacalai, Kyoto); metitepine maleate (Nihon Roche, Tokyo); phentolamine hydrochloride (Nihon Ciba-Geigy, Takarazuka); tetrodotoxin (Sankyo, Tokyo); vasoactive intestinal polypeptide (Peptide Institute, Osaka). The molar concentrations of drugs described in this paper refer to the final bath concentration.
3. Results
3.1. Effects of nicotine The circular muscle strip of the guinea-pig gastric fundus has sufficient resting tone to produce relaxation (Kojima et al., 1992). In the presence of atropine (0.2 /zM), nicotine (10-100 /zM) produced concentrationdependent relaxation with a mean pECs0 value of 4.60 and a mean maximum effect (Em~x) at 100 /~M, amounting to 53.3 + 5.1% (n = 8) of the papaverine-induced maximal relaxation. A higher concentration (300 ~M) of nicotine produced no further relaxation. Nicotine 100 ~M evoked a reproducible and rapid relaxation response which, after reaching a peak, returned spontaneously to the basal tone (fig. 1A). None of the antagonists investigated had a significant influence on basal tone. As shown in table 1, the relaxant response to nicotine (100 /zM) was not affected by metitepine (0.3/zM), but was significantly reduced by tetrodotoxin (0.3 IzM, 84.2 + 3.5% inhibition), hexamethonium (10/xM, 83.8 + 3.4% inhibition) and the combination of timolol (0.3 t~M) and phentolamine (0.3 tzM) (37.8 + 4.1% inhibition). a-Chymotrypsin (2 u / m l ) completely abolished the relaxant response to VIP (0.1/zM) which amounted to 48.1 + 5.9% (n = 6) of that of papaverine (30/xM), but
173
did not significantly influence the nicotine-induced relaxation. As shown in table 2, L-NNA and L-NAME significantly inhibited the nicotine (100 /~M)-induced relaxation in a concentration-dependent manner, and at the higher concentrations used, L-NNA (10 /zM) and L-NAME (100 ~M), inhibited the response by 98.1 + 0.9% and 86.5 + 5.2% respectively. Neither Larginine (3 mM) nor D-arginine (3 mM) affected the nicotine-induced relaxation (n = 3, results not shown). L-Arginine (3 mM), but not D-arginine (3 mM), completely prevented the inhibitory effect of L-NNA (10
NC 1 0 0
TABLE 1 Effects of antagonists on the relaxant response to nicotine (100/~M) of circular muscle strips of the guinea-pig gastric fundus. The data are means + S.E.M. from (n) of experiments. Atropine (0.2 /~M) was present in all experiments. Antagonists
Saline Tetrodotoxin Hexamethonium Timolol/phentolamine Metitepine a-Chymotrypsin
n
0.3/zM 10 /zM 0.3/zM 0.3 p.M 2.0 u / m l
5 6 6 6 6 5
Nicotine-induced relaxation (%) a Before treatment
After treatment
47.5+ 3.1 48.6+4.5 60.1+ 2.9 54.4 + 5.2 56.0 + 4.0 47.1_+3.0
50.0 + 3.0 8.2+2.0 a 10.1 -I-2.6 b 34.7 + 5.3 c 50.0 + 4.3 37.4 + 5.9
a Relaxations relative to those induced by papaverine 30 ~M. Significantly different from control (before treatment), b p < 0.05, c P < 0.001.
/~M) and L-NAME (10 /zM) on the nicotine-induced relaxation (fig. 1 and table 2). In fundic strips from 6-hydroxydopamine-pretreated animals, 100 /~M nicotine evoked relaxation amounting to 35.7 _+ 4.1% (n = 7) of that produced by papaverine (30 /.tM). The response was significantly smaller than that in the preparation from vehicle-pretreated animals (48.3 + 3.3%, n -- 7, P < 0.05).
Saline NC 1OO
3.2. Effects of sodium nitroprusside and noradrenaline
m
L-NNA 10
The NO donor, sodium nitroprusside (3-1000 nM) produced a concentration-dependent relaxation of the fundic strips. The response was not affected by prein-
NC 1 0 0
TABLE 2 Effects of N°-nitro-L-arginine (L-NNA) and NG-nitro-L-argininemethyl ester (L-NAME) on the relaxant response to nicotine (100 /~M) of circular muscle strips of the guinea-pig gastric fundus in the absence or presence of L- or D-arginine (3 mM).
k-Arg 3000/L-NNA NC 1 0 0
The data are means + S.E.M. from n experiments. Atropine (0.2/zM) was present in all experiments. Compound
n
D-Arg 30OO/L-NNA
I.| 5mln Fig. 1. Typical tracings of the relaxation in response to nicotine (100 ~M) of circular muscle strips of the guinea-pig gastric fundus in the absence (left panels) and presence (right panels) of vehicle (A), N G-nitro-L-arginine (L-NNA 10/.~M, B), L-NNA + L-arginine (L-Arg 3 mM, C) or L-NNA+ D-arginine (D-Arg 3 mM, D). L-Arginine, but not D-arginine, completely reversed the inhibitory effect of L-NNA on the nicotine-induced relaxation.
L-NNA 1/zM L-NNA 10 ~M L-NNA 10/.t M/L-arginine L-NNA 10/~M/D-arginine L-NAME 10/~M L-NAME 100/.~M L-NAME 10/z M/L-arginine L-NAME 10 p.M/D-arginine
6 6 6 6 6 6 6 6
Nicotine-induced relaxation (%) a Before treatment
After treatment
50.5+4.7 53.0+4.6 58.9 + 3.4 59.2 + 7.3 54.4+3.4 51.5+2.9 52.4 + 3.6 51.5 + 5.7
18.9+7.4 b 0.9+0.5 ¢ 55.1 + 3.1 2.9 + 1.7 ¢ 10.8+4.1 c 6.4+2.2 ¢ 49.1 + 4.9 5.7 + 2.3
a Relaxations relative to those induced by papaverine 30/zM. Significantly different from control (before treatment), b p < 0.01, c p < 0.001.
174 TABLE 3 Effects of antagonists on the relaxant response to sodium nitroprusside (3-1000 nM) of circular muscle strips of the guinea-pig gastric fundus. Emax values (percentage for the effect of papaverine, 30 ~M) and pECs0 values ( - log M) are listed. The data are means5:S.E.M. from n experiments. Atropine (0.2/~M) was present in all experiments. Antagonists Control L-NNA Timolol/phentolamine Metitepine
n 16 100 /~M 6 0.3/zM 6 0.3 p.M 4
pECs0 7.19+0.09 7.30+0.14 7.055:0.08 7.555:0.21
Emax (%)
85.4_+2.5 93.65:1.9 91.85:2.6 90.45:4.3
cubation of the tissue with L-NNA (100 /zM), timolol (0.3 /xM)/phentolamine (0.3 /~M) or metitepine (0.3 /~M), as shown by the identical Emax and pECs0 values (table 3). The Emax and pECs0 values for sodium nitroprusside in the fundic strips from 6-hydroxydopamine-pretreated animals were 83.3 + 3.2% and 7.25 + 0.14 (n = 7), which was comparable with the control values, 88.2 ± 3.0% and 7.31 + 0.12 (n = 7). 6Hydroxydopamine pretreatment had no significant effect on the relaxant response to sodium nitroprusside. The involvement of NO or a related substance in the relaxation induced by noradrenaline was also examined. In the presence of yohimbine (1 /~M), noradrenaline ( 0 . 1 - 3 0 / x M ) produced a concentration-dependent relaxation with a mean pECs0 value of 6.57 + 0.23 and E m a x a t 1 /~M, amounting to 68.5 + 5.1% (n = 4) of the effect of papaverine (30/xM). The Emax and pECs0 values were not affected by the pretreatment with L-NNA (100/~M) (n = 4).
4. Discussion
Nicotine produces a concentration-dependent relaxation of circular muscle strips of the guinea-pig gastric fundus under resting tone. The response is probably mediated by the stimulation of nicotinic cholinoceptors located in enteric inhibitory neurons, because it was inhibited by both hexamethonium or tetrodotoxin. However, these antagonists did not completely abolish the response. Hence, a non-neuronal component might be partially involved in the nicotine-induced relaxation. Although the exact nature of the latter components remains unknown, nicotine is known to elicit a T r X resistant response in several peripheral organs (Takayanagi et al., 1984; Hisayama et al., 1988). There are two types of inhibitory neurons that are known to affect the smooth muscle motility of the gastrointestinal tract, noradrenergic neurons and NANC neurons. In the present study, one of the inhibitory neurotransmitters responsible for the nicotine-induced relaxation was likely to be noradrenaline,
because the relaxation was partially reduced either by the combination of timolol, a /3-adrenoceptor antagonist and phentolamine, an a-adrenoceptor antagonist, or by chemical sympathectomy with 6-hydroxydopamine. A similar result was observed by Costall et al. (1983) who showed that the relaxation of circular muscle strips of the cardia and body of the guinea-pig stomach induced by electrical field stimulation were partially antagonized by pretreatment with reserpine, propranolol or phentolamine. T h e r e is now good evidence that NO-like substance(s) and VIP might be inhibitory NANC neurotransmitters in the mammalian gastric fundus (De Beurme and Lefebvre, 1988; Li and Rand, 1990; Boeckxstaens et al., 1991; Lefebvre et al., 1992). In the present study, L-NNA and L-NAME, inhibitors of NO synthase (Moncada, et al 1991), inhibited the nicotineinduced relaxation in a concentration-dependent manner, and at 10 tzM L-NNA abolished the response. The inhibitory effects of L-NNA and L-NAME were stereospecifically prevented by L-arginine, a substrate for NO synthase. These findings indicate that the relaxant action of nicotine on the guinea-pig gastric fundus is mediated entirely by the release of NO or a related substance from the L - a r g i n i n e / N O pathway. Our resuits are comparable with the finding by Irie et al. (1991) that NO or a related substance is involved in the NANC relaxation induced by nicotine in the rat duodenum. Yamamoto et al. (1993) recently demonstrated that L-NNA reduces the noradrenaline overflow in response to electrical field stimulation of the isolated rat mesenteric vasculature, suggesting that NO or a related substance enhances the release of noradrenaline from perivascular adrenergic nerves. This may account for our present finding that the relaxant action of nicotine was partially inhibited by chemical sympathectomy or a- and fl-adrenoceptor blockade. However, we cannot exclude the possibility that nicotine stimulates not only inhibitory nerves but also excitatory NANC nerves in myenteric plexus, since the guinea-pig stomach is innervated by intramural nerves containg 5-HT, VIP, substance P, calcitonin gene-related peptide and neuropeptide Y (Mawe et al., 1989). The elimination of adrenergic components by adrenoceptor antagonists or chemical sympathectomy would enhance the release of excitatory neurotransmitters, which in turn may reduce the relaxation mediated by NO. Further studies will be required to test this possibility. It is unlikely that noradrenaline released by nicotine could enhance the NO production since the relaxation in response to exogenously applied noradrenaline was not affected by LNNA. The possibility that VIP plays an important role in the relaxation due to nicotine is also unlikely, because a-chymotrypsin, at a concentration which abolished the
175
response to exogenous VIP, did not significantly influence the response. In a previous report, we demonstrated that exogenously applied 5-HT produces relaxation of circular muscles of the guinea-pig gastric fundus by stimulating 5-HTl-like receptors located on smooth muscles (Kojima et al., 1992), and 5-HT has also been proposed as a putative NANC neurotransmitter (Burnstock, 1986). However, it is also unlikely that 5-HT plays an important role in the relaxation in response to nicotine, because the potent 5-HTl-like receptor antagonist, metitepine (Kojima et al., 1992), failed to inhibit the response. In conclusion, nicotine produced relaxation of circular muscle strips of the guinea-pig gastric fundus by two mechanism: mainly by the release of NO or a related substance from the L-arginine/NO pathway, and partially by the release of noradrenaline from adrenergic nerves. Further contributions of 5-HT or VIP to nicotine-induced relaxation are apparently negligible. References Boeckxstaens, G.E., P.A. Pelckmans, J.J. Bogers, H. Bult, J.G. De Man, L. Oosterbosch, A.J. Herman and Y.M. Van Maercke, 1991, Release of nitric oxide upon stimulation of nonadrenergic non-cholinergic nerves in the rat gastric fundus, J. Pharmacol. Exp. Ther. 256, 441. Burnstock, G., 1986, The non-adrenergic non-cholinergic nervous system, Arch. Int. Pharmacodyn. 280 (Suppl.), 1. Costall, B., R.J. Naylor and C.C.W. Tan, 1983, Field stimulation-induced responses of circular smooth muscle from guinea-pig stomach, Naunyn-Schmied. Arch. Pharmacol. 323, 155. D'Amato, M., F.A. De Beurme and R.A. Lefebvre, 1988, Comparison of the effect of vasoactive intestinal polypeptide and nonadrenergic non-cholinergic neurone stimulation in the cat gastric fundus, Eur. J. Pharmacol. 152, 71. Daniel, E.E, 1982, Pharmacology of adrenergic, cholinergic, and drugs acting on other receptors in gastrointestinal muscle, in: Mediators and Drugs in Gastrointestinal Motility, II, ed. G. Bertaccini (Springer-Verlag, Berlin, Heidelberg, New York) p. 249.
De Beurme, F.A. and R.A. Lefebvre, 1988, Vasoactive intestinal polypeptide as possible mediator of relaxation in the rat gastric fundus, J. Pharm. Pharmacol. 40, 711. Grider, J.R., M.B. Cable, 5.1. Said and M. Makhlouf, 1985, Vasoactive intestinal peptide as a neural mediator of gastric relaxation, Am. J. Physiol. 248, G73. Hisayama, T., M. Shinkai, I. Takayanagi and T. Toyoda, 1988, Mechanisms of action of nicotine in isolated urinary bladder of guinea-pig, Br. J. Pharmacol. 95, 465. Irie, K., T. Muraki, K. Furukawa and T. Nomoto, 1991, L-NG-Nitro L-arginine inhibits nicotine-induced relaxation of isolated rat duodenum, Eur, J. Pharmacol. 202, 285. Kirkpatrick, K and G. Burnstock, 1987, Sympathetic nerve-mediated release of ATP from the guinea-pig vas deferens is unaffected by reserpine, Eur. J. Pharmacol. 138, 207. Kojima, S., R. lshizaki and Y. Shimo, 1992, Investigation into the 5-hydroxytryptamine-induced relaxation of the circular smooth muscle of guinea-pig stomach fundus, Eur. J. Pharmacol. 224, 45. Lefebvre, R.A., E. Baert and A.J. Barbier, 1992, Influence of N ~nitro-L-arginine on non-adrenergic non-cholinergic relaxation in the guinea-pig gastric fundus, Br. J. Pharmacol. 106, 173. Li, C.C and M.J. Rand, 1990, Nitric oxide and vasoactive intestinal polypeptide mediate non-adrenergic, non-cholinergic inhibitory transmission to smooth muscle of the rat gastric fundus, Eur. J. Pharmacol. 191,303. Mawe, G.M., M. Schemann, J.D. Wood and M.D. Gershon, 1989, Immunocytochemical analysis of potential neurotransmitters present in the myenteric plexus and muscular layers of the corpus of the guinea-pig stomach, Anat. Rec. 224, 431. Moncada, S., R.M.J. Palmer and E.A. Higgs, 1991, Nitric oxide: Physiology, pathophysioiogy and pharmacology, Pharmacol. Rev. 43, 109. Sahyoun, H.A., B. Costall and R.J. Naylor, 1982, On the ability ot domperidone to selectively inhibit catecholamine-induced relaxation of circular smooth muscle of guinea-pig stomach, J. Pharm.Pharmacol. 34, 27. Takayanagi, I., Y. Kizawa and T. Hiruta, 1984, Tetrodotoxin-resistant response to nicotine in rabbit bronchial preparation, Eur. J. Pharmacol. 104, 351. Van Rossum, J.M, 1963, Cumulative dose-response curves. II. Technique for the making of dose-response curves in isolated organs and the evaluation of drug parameters, Arch. Int. Pharmacodyn. 143, 299. Yamamoto, R., A. Wada, Y. Asada, H. Niina and A. Sumiyoshi, 1993, NC-Nitro-L-arginine, an inhibitor of nitric oxide synthesis, decreases noradrenaline outflow in rat isolated perfused mesenteric vasculature, Naunyn-Schmied. Arch. Pharmacol. 347, 238.