- Email: [email protected]
5-HT4 Receptors contribute to the motor stimulating effect of levosulpiride in the guinea-pig gastrointestinal tract
Digestive and Liver Disease 35 (2003) 244–250 www.elsevier.com / locate / dld
Alimentary Tract
5-HT 4 Receptors contribute to the motor stimulating effect of levosulpiride in the guinea-pig gastrointestinal tract M. Tonini a , *, R. De Giorgio b , V. Spelta a , G. Bassotti c , A. Di Nucci d , L. Anselmi a , B. Balestra d , F. De Ponti e a
Department of Physiological and Pharmacological Sciences, University of Pavia, Piazza Botta 10 /11, 27100 Pavia, Italy b Department of Internal Medicine and Gastroenterology, University of Bologna, Bologna, Italy c Department of Clinical and Experimental Medicine, University of Perugia, Perugia, Italy d Department of Internal Medicine and Therapeutics, University of Pavia, Pavia, Italy e Department of Pharmacology, University of Bologna, Bologna, Italy Received 2 October 2002; accepted 12 November 2002
Abstract Background. The dopamine D 2 receptor antagonist levosulpiride is a substituted benzamide derivative, whose gastrokinetic properties are exploited clinically for the management of functional dyspepsia. However, for other benzamide derivatives, such as cisapride and mosapride, agonism towards serotonin 5-HT 4 receptors is considered the main mechanism leading to gastrointestinal prokinesia. Aims. To assess whether levosulpiride is able to activate 5-HT 4 receptors in the guinea-pig isolated gastrointestinal tract. Materials and methods. Circular muscle strips from gastric antrum, and colonic longitudinal muscle strips were used to detect electrically stimulated neurogenic contractions. The effect of levosulpiride was assessed in the absence and presence of GR125487, a selective 5-HT 4 receptor antagonist. Furthermore, potential interaction of levosulpiride with 5-HT 3 receptors and tissue cholinesterases was assessed in unstimulated ileal longitudinal muscle-myenteric plexus preparations. Results. Antral and colonic strip contractions were cholinergic / tachykinergic in nature. Micromolar concentrations of levosulpiride potentiated submaximal responses, through a mechanism competitively antagonized by GR125487 (pKB 59.4). In LMMPs, levosulpiride slightly affected contractions caused by the 5-HT 3 receptor agonist 2-methyl-5-HT, and had no effect on contractions to exogenous acetylcholine. Conclusions. Our results indicate that levosulpiride acts as a moderate agonist at the 5-HT 4 receptor. This property, together with antagonism at D 2 receptors, may contribute to its gastrointestinal prokinetic effect. 2003 Editrice Gastroenerological Italiana S.r.l. Published by Elsevier Science Ireland Ltd. All rights reserved. Keywords: Guinea-pig gastrointestinal tract; 5-HT 4 receptors; Levosulpiride
1. Introduction The substituted benzamide derivative levosulpiride is a selective dopamine D 2 receptor antagonist used for the management of upper gastrointestinal tract disorders associated with functional dyspepsia [1–3]. The rationale for proposing antidopaminergic compounds as prokinetic agents was based on the evidence that dopamine is present in the gastrointestinal wall of several mammals [4,5], where it potently inhibits motility [6,7] and release of *Corresponding author. Tel.: 139-0382-506-363; fax: 139-0382-506419. E-mail address: [email protected] (M. Tonini).
excitatory neurotransmitters (e.g., acetylcholine (ACh)) [8– 10]. Substituted benzamide gastrointestinal prokinetics belong to a chemical class with a complex pharmacological profile, since they may (metoclopramide) or may not (cisapride, mosapride) possess antidopaminergic properties together with agonism at the serotonin 5-HT 4 receptor (metoclopramide, cisapride, mosapride) and antagonism at the 5-HT 3 receptor (metoclopramide, cisapride and a metabolite of mosapride) [11,12]. Furthermore, some of these drugs have moderate anticholinesterase activity [13,14]. In the gastrointestinal tract, 5-HT 4 receptors are located on myenteric neurons, enterocytes and smooth muscle cells. In the enteric nervous system, the presence of 5-HT 4 receptors has been demonstrated on several func-
1590-8658 / 03 / $30 2003 Editrice Gastroenerological Italiana S.r.l. Published by Elsevier Science Ireland Ltd. All rights reserved. doi:10.1016 / S1590-8658(03)00061-6
M. Tonini et al. / Digestive and Liver Disease 35 (2003) 244–250
tionally distinct subtypes of neural cells including intrinsic primary afferent neurons, ascending cholinergic / tachykinergic excitatory pathways and descending non-adrenergic, non-cholinergic inhibitory pathways. Activation of 5-HT 4 receptors is associated with release of transmitters involved in enteric reflexes and peristalsis, and with water and electrolyte secretion from enterocytes [15,16]. In smooth muscle cells of oesophagus and colon, 5-HT 4 receptors cause cAMP-dependent relaxations [17,18]. It is widely accepted, therefore, that the prokinetic action of several marketed benzamide derivatives depend, at least in part, on the activation of 5-HT 4 receptors [16]. To date, no information on possible interaction of levosulpiride with serotonin receptors and tissue cholinesterases is available. This study was designed to re-assess the pharmacological profile of levosulpiride using isolated tissues from the guinea-pig gastrointestinal tract.
2. Materials and methods Male Dunkin–Hartley guinea-pigs (250–400 g; Bettinardi S.r.l., Momo, Novara, Italy) were used. Animals were housed in standard animal facilities, providing constant temperature (2161 8C), relative humidity (50–55%), alternating 12-h light and dark cycles, and were fed standard laboratory food (Laboratori Dr Piccioni, Gessate, Milan, Italy) and tap water ad libitum.
2.1. Tissue preparation Segments of gastric antrum, distal ileum and distal colon were carefully removed, cleared of their intraluminal contents and placed in a Petri dish containing prewarmed Tyrode solution. All tissues prepared from the above segments (strips of gastric antrum, longitudinal musclemyenteric plexus, and distal colon) reached equilibrium in Tyrode solution maintained at 37 8C and aerated with a mixture of 95% O 2 15% CO 2 for at least 60 min before experiments were begun.
245
transmitters involved in the contractile response were investigated using muscarinic (hyoscine) and tachykinergic antagonists acting at NK1 (SR140333) and NK2 receptors (MEN11420) left in contact with the tissue for 20–40 min before stimulation. Contractions were referred as percentage of the maximal response obtained at 30 Hz. Noncumulative concentration–response curves to levosulpiride (i.e., each drug concentration was removed from the bath by changing the medium at 3-min intervals on three occasions) in potentiating submaximal contractions induced by 5 Hz were constructed in the absence and presence of the 5-HT 4 receptor antagonist GR125487 (incubation time: 20 min). The apparent pKB value (a measure of antagonist affinity) for GR125487 was evaluated using the Gaddum equation [19].
2.3. Longitudinal muscle-myenteric plexus preparations from distal ileum According to the method of Paton and Zar [20], strips of longitudinal muscle with the myenteric plexus attached (LMMPs) were separated from the underlying circular muscle by stroking tangentially away from the mesenteric attachment with a moistened cotton wad. LMMPs were set up isometrically (tension: 5 mN) in a 5-ml organ bath. In unstimulated preparations, non-cumulative concentration– response curves to the contractile effect of 2-methyl-5-HT (0.3–10 mM), a 5-HT 3 receptor agonist, were obtained in the absence and presence of 10 mM levosulpiride (incubation time: 20 min). Agonist responses were referred as a percentage of the response induced by the maximal 2methyl-5-HT concentration (10 mM). Antagonist affinity for levosulpiride was evaluated as apparent pKB value. In a separate set of experiments, cumulative concentration– response curves to the contractile effects of ACh (i.e., each agonist concentration was added to the bath consecutively, once the previous drug administration produced a plateau response) were obtained in the absence and presence of levosulpiride (100 mM) or eserine (30 nM).
2.4. Longitudinal muscle strips of distal colon 2.2. Circular muscle strips of gastric antrum Segments of gastric antrum were pinned flat on a sylastic holder with the mucosa layer facing upwards. After removal of the mucosa, circular muscle strips (2.5–3 cm long and 0.5 cm wide) were mounted isometrically (tension: 8 mN) in a 5-ml organ bath containing Tyrode solution maintained at 37 8C and aerated with a mixture of 95% O 2 15% CO 2 . Electrical field stimulation was delivered by means of two platinum electrodes placed at the top and bottom of the chamber, respectively. Parameters of stimulation were: trains of pulses delivered at 1-min intervals, 60 V, 1–30 Hz frequency, 0.5 ms duration, for 5 s. In electrically-stimulated preparations, the excitatory
Segments of distal colon were cut open along the mesenteric axis and pinned flat on a sylastic holder with the mucosa layer facing upwards. After removal of the mucosa, longitudinal muscle strips (2.5–3 cm long and 0.5 cm wide) were mounted isometrically (tension: 8 mN) in a 5-ml organ bath containing Tyrode solution maintained at 37 8C and aerated with a mixture of 95% O 2 15% CO 2 . Parameters of electrical field stimulation were: trains of pulses delivered at 1-min intervals, 60 V, 1–30 Hz frequency, 0.5 ms duration, for 5 s. In electrically-stimulated preparations, the excitatory transmitters involved in the contractile response were studied using muscarinic (hyoscine) and tachykinergic antagonists (SR140333 and MEN11420) left in contact with the tissue for 20–40 min
246
M. Tonini et al. / Digestive and Liver Disease 35 (2003) 244–250
before stimulation. Contractions were referred as percentage of the maximal response obtained at 30 Hz. Noncumulative concentration–response curves to levosulpiride in potentiating submaximal contractions induced by 5 Hz were constructed in the absence and presence of the 5-HT 4 receptor antagonist GR125487 (incubation time: 20 min).
2.5. Statistical analysis Data are expressed as means6standard error of the mean (SEM). Differences between means (multiple comparisons) were analysed by one-way analysis of variance (ANOVA). Values of p,0.05 were taken as statistically significant.
2.6. Solutions and drugs Tyrode solution (pH 7.4) had the following composition (mM): NaCl 136.9, KCl 2.7, MgCl 2 1.04, NaHCO 3 11.9, NaH 2 PO 4 0.4, glucose 5.5. The following drugs were used: SR140333 (a selective NK 1 antagonist kindly donated by Dr T. Croci, Sanofi Research, Milan, Italy), MEN11420 (a selective NK 2 antagonist kindly donated by Dr C.A. Maggi, Menarini Research, Florence, Italy), GR125487 (a selective 5-HT 4 antagonist kindly supplied by Dr S.G. Lister, GlaxoSmithKline, Hertfordshire, UK), 2-methyl-5-hydroxytryptamine maleate, S(2)-sulpiride [levosulpiride], R(1)-sulpiride [dextrosulpiride] (Research Biochemical International, Natick, USA), acetylcholine (ACh) chloride, tetrodotoxin (Sigma Diagnostics, St Louis, MO, USA), eserine hemisulphate (ICN Pharmaceuticals Inc., Cleveland, USA), hyoscine hydrochloride [(2)scopolamine] (Sigma-Chemie, Deisenhofen, Germany). All drugs were dissolved in distilled water, with the exception of SR140333, which was previously dissolved in dimethyl sulphoxide (DMSO) (stock solution: 10 mM) and then diluted in distilled water, and levosulpiride, which was dissolved in 0.1 N HCl (stock solution: 100 mM) and then diluted in distilled water. Drugs were administered in volumes not exceeding 1% v / v of the bath volume. The DMSO / water and HCl / water solutions were without effect on gastrointestinal mechanical activity.
3. Results
Fig. 1. Frequency-dependent contractions to electrical field stimulation (1–30 Hz) in isolated circular muscle strips of guinea-pig gastric antrum. Responses are referred as contractions in the absence of pharmacological treatment (control) or in the presence of NK 1 (SR140333: 1 mM) and NK 2 (MEN11420: 1 mM), or NK 1 , NK 2 and muscarinic receptor blockade by hyoscine (1 mM). Values are expressed as percent of the response to 30 Hz stimulation. Any pharmacological treatment caused a significant reduction ( p,0.05 or better: ANOVA for continuous-by-class effects) of the frequency-response curve. Each point represents the mean6SEM of 4–8 determinations.
tion are mediated by acetylcholine activating postjunctional muscarinic receptors and, to a lesser extent, by tachykinins acting at NK 1 and NK 2 receptors [21]. Levosulpiride (3–1000 mM) concentration-dependently enhanced the amplitude of submaximal contractions elicited at 5 Hz without affecting the basal tone. The agonist potency (pD 2 value) was 4.6. The potentiating effect of levosulpiride was competitively antagonized by 3 nM GR125487 (Fig. 2). The apparent pKB value for GR125487 was 9.460.2. Conversely, dextrosulpiride (3– 100 mM) was virtually inactive, although at 100 mM, the drug enhanced twitch contractions by 35% in one out of seven preparations (data not shown). This evidence suggests that levosulpiride, and not dextrosulpiride, possesses moderate agonist properties at the 5-HT 4 receptor in the guinea-pig gastric antrum.
3.1. Circular muscle strips of gastric antrum 3.2. Longitudinal muscle strips of distal colon Electrical field stimulation elicited reproducible twitch contractions in a frequency-dependent (1–30 Hz) manner (Fig. 1). A combination of SR140333 (1 mM) and MEN11420 (1 mM) partially inhibited electrically-induced contractions. Administration of hyoscine (1 mM) in the presence of tachykinin antagonists abolished any contractile response (Fig. 1). These findings indicate that in the gastric antrum, contractile responses to electrical stimula-
Electrical field stimulation elicited reproducible twitch contractions in a frequency-dependent (1–30 Hz) manner (Fig. 3). Contractions were sensitive to a combination of SR140333 (1 mM), MEN11420 (1 mM) and hyoscine (1 mM), which reduced the response by approximately 80%. The administration of tetrodotoxin (TTX: 1 mM) abolished any residual response. These findings indicate that in the
M. Tonini et al. / Digestive and Liver Disease 35 (2003) 244–250
Fig. 2. Concentration–response curves for the potentiating effect of levosulpiride on electrically stimulated (5 Hz) submaximal contractions of antral strips in the absence or presence of 0.3 nM GR125487, a selective 5-HT 4 receptor antagonist. Values are expressed as percent increase in control contractions. Each point represents the mean6SEM of 5–8 determinations.
distal colon, contractile responses to electrical stimulation are mediated by ACh activating postjunctional muscarinic receptors, by tachykinins acting at NK 1 and NK 2 receptors, and by an unknown excitatory transmitter responsible for approximately 20% of the response. Levosulpiride (3–300 mM) concentration-dependently enhanced the amplitude of submaximal contractions elicited at 5 Hz (Fig. 4) without affecting the basal tone. The agonist potency (pD 2 value) was 5.3. The potentiating effect of levosulpiride was markedly antagonized by 3 and 10 nM GR125487, suggesting also that in colonic longitudinal strips, levosulpiride behaves as a moderate agonist at the 5-HT 4 receptor. The apparent pKB value for GR125487 evaluated at 3 nM was 9.560.1.
3.3. LMMPs from distal ileum In unstimulated LMMPs, 2-methyl-5-HT (0.3–10 mM) induced neurogenic contractile responses, which were slightly antagonized by 10 mM levosulpiride (Fig. 5) with 20% depression of the maximum agonist response. The apparent pKB value was 4.660.3 (n58), suggesting a very slight antagonism at the 5-HT 3 receptor, which appears partial or non-surmountable in nature. In a separate set of experiments, cumulative administration of ACh (0.1 nM– 100 mM) caused the preparation to contract. Levosulpiride (100 mM) had no effect on the concentration–response curve to applied ACh (Fig. 6A). Conversely, the contrac-
247
Fig. 3. Frequency-dependent contractions to electrical field stimulation (1–30 Hz) in isolated longitudinal muscle strips of guinea-pig distal colon. Responses are referred to as contractions in the absence of pharmacological treatment (control), or in the presence of NK 1 (SR140333: 1 mM) and NK 2 (MEN11420: 1 mM), or NK 1 , NK 2 and muscarinic receptor blockade by hyoscine (1 mM). Contractions were completely suppressed by tetrodotoxin (TTX: 1 mM). Values are expressed as percent of the response to 30 Hz stimulation. Any pharmacological treatment caused a significant reduction ( p,0.05 or better: ANOVA for continuous-by-class effects) of the frequency-response curve. Each point represents the mean6SEM of five determinations.
tile response curve to ACh was shifted to the left and in a parallel fashion by 30 nM eserine (Fig. 6B). These findings indicate that the potentiating effect of levosulpiride on twitch contractions does not involve inhibition of endogenous cholinesterases.
4. Discussion The results of the present study indicate that in two distinct regions of the guinea-pig gastrointestinal tract, namely the gastric antrum and the distal colon, levosulpiride potentiates electrically-induced excitatory neuromuscular transmission, which is partly mediated by release of ACh acting at postjunctional muscarinic receptors (M 3 ), and tachykinins that activate postjunctional NK 1 and NK 2 receptors. In the gastric antrum, levosulpiride-induced enhancement of submaximal contractile responses to 5 Hz stimulation was stereospecific, since the enantiomer dextrosulpiride was virtually inactive, and was counteracted by the selective 5-HT 4 receptor antagonist GR125487 with an
248
M. Tonini et al. / Digestive and Liver Disease 35 (2003) 244–250
Fig. 4. Concentration–response curves for the potentiating effect of levosulpiride on electrically stimulated (5 Hz) submaximal contractions of colonic longitudinal muscle strips in the absence or presence of 3 and 10 nM GR125487, a selective 5-HT 4 receptor antagonist. Values are expressed as percent increase in control contractions. Each point represents the mean6SEM of 5–8 determinations.
Fig. 5. Concentration–response curves for the contractile response induced by 2-methyl-5-HT in unstimulated longitudinal muscle-myenteric plexus preparations from guinea-pig distal ileum, in the absence and presence of 10 mM levosulpiride. Values are expressed as percent of the response induced by 10 mM 2-methyl-5-HT. Note that levosulpiride caused a slight but significant parallel shift to the right of the 2-methyl-5HT curve ( p,0.05: ANOVA for continuous-by-class effects). Each point represents the mean6SEM of 4–8 determinations.
estimated pKB value of 9.4. This affinity value is similar to that previously reported for GR125487 acting at 5-HT 4 receptors in intestinal and extraintestinal preparations [22,23]. These findings indicate that the enhanced neurogenic excitatory drive to antral circular smooth muscle cells induced by levosulpiride is mediated by 5-HT 4 receptors. Therefore this benzamide derivative possesses 5-HT 4 agonist properties that are qualitatively similar to those of other compounds of the same chemical class, such as metoclopramide, cisapride and mosapride [11,16]. However, compared to cisapride and mosapride, the agonist potency of levosulpiride at the 5-HT 4 receptor is moderate, since this molecule is effective in the micromolar range of concentrations. Previous studies have documented the presence of 5-HT 4 receptors on cholinergic nerves in the guinea-pig stomach where they facilitate the release of ACh as observed in antral, corporal and fundic strips [24–27]. This effect has been postulated to be involved in the regulation of gastric motility. Recently, prucalopride, a benzofuran derivative with potent 5-HT 4 receptor agonist properties, was found to enhance nerve-mediated (cholinergic) contractility of canine proximal stomach both in vitro and in vivo [28]. The important role of 5-HT 4 receptors in mediating the gastroprokinetic action of benzamide derivatives is inferred from studies carried out in different animal species. As an example, benzamide derivatives were found to enhance gastric emptying in dogs and rats [29–33]. Cholinergically mediated enhancement of gastric emptying and gastroduodenal motility induced by benzamide derivatives through 5-HT 4 receptors stimulation were also documented in man [34]. The presence of 5-HT 4 receptors on cholinergic nerves supplying colonic longitudinal muscle has been identified in several mammalian species including humans [28]. Longitudinal muscle preparations of the guinea-pig distal colon should be considered one of the most sensitive in vitro systems to investigate the effects of 5-HT 4 receptor on the enteric excitatory innervation [35]. Indeed, in these preparations, 5-HT 4 receptor activation was found to enhance the release of ACh [35–37] and tachykinins [37,38]. With regard to the functional role of 5-HT 4 receptor activation, agonists such as prucalopride and the amino-guanidine-indole derivative, tegaserod (a partial 5HT 4 agonist), were found to facilitate the release of both ACh and tachykinins leading to the potentiation of distension-evoked standing reflexes and peristalsis [39,40]. Our results in electrically stimulated guinea-pig longitudinal muscle preparations are consistent with the release of ACh and tachykinins, since the administration of hyoscine plus SR140333 and MEN11420 reduced to 20% the amplitude of submaximal contractions to 5 Hz stimulation. In this model system, levosulpiride potentiated in a concentration-dependent manner the submaximal contractile response induced by electrical stimulation through a mechanism competitively antagonized by GR125487 (pKB 59.5). The potency estimate (pD 2 ) for levosulpiride
M. Tonini et al. / Digestive and Liver Disease 35 (2003) 244–250
249
Fig. 6. Concentration–response curves for the contractile effects induced by exogenous ACh in unstimulated longitudinal muscle-myenteric plexus, in the absence and in the presence of 100 mM levosulpiride (A) or 30 nM eserine (B). Note the inefficacy of levosulpiride on the contractile response evoked by ACh. Conversely, in (B), the curve profile of ACh in the presence of eserine is statistically different ( p,0.01: ANOVA for continuous-by-class effects) from the control curve. Each point represents the mean6SEM of 5–8 determinations.
was 5.3, a value significantly lower than that obtained with the full agonist, 5-methoxytryptamine (5-MeOT), in the guinea-pig isolated colon (pD 2 56.9) [38]. This, together with the evidence that the potentiating effect of levosulpiride was lower than that produced by 5-MeOT, strongly indicate that levosulpiride behaves as a partial 5-HT 4 receptor agonist in the guinea-pig gastrointestinal tract. Other well-known properties of substituted benzamide derivatives are the mild / moderate antagonism at the 5-HT 3 receptor [11] and the inhibition of tissue cholinesterases [13,14]. The 5-HT 3 antagonist properties of compounds such as granisetron, tropisetron and ondansetron are used in clinical practice to prevent chemotherapy-evoked vomiting. Furthermore, 5-HT 3 receptor antagonists have recently been proposed for the treatment of diarrhoea-predominant irritable bowel syndrome on account of their anti-propulsive effects in the small and large bowel [16]. With regard to the inhibition of cholinesterases, this effect may potentiate the procholinergic action of some benzamide derivatives (i.e., metoclopramide and cisapride) [13,14], thus contributing to their gastrointestinal prokinetic effects. In our hands, levosulpiride was found to antagonize with low affinity (pKB 54.6) contractile responses induced by the selective 5-HT 3 receptor agonist 2-methyl-5-HT in unstimulated longitudinal muscle and myenteric plexus preparations. This implies that levosulpiride has very mild antagonist properties (with characteristics of partial or non-surmountable antagonism) at the 5-HT 3 receptor. In the same preparations, levosulpiride was ineffective on contractions induced by exogenous ACh even at high concentrations (up to 100 mM), whereas the cholinesterase inhibitor 30 nM eserine significantly shifted the ACh curve to the left.
In conclusion, our results suggest that the gastrointestinal prokinetic agent levosulpiride possesses moderate partial agonist properties at the serotonin 5-HT 4 receptor and very mild antagonist property at the 5-HT 3 receptor, while it has no anticholinesterase activity. It is well recognized that the prokinetic efficacy of 5-HT 4 receptor agonists (e.g. cisapride) is superior to that of pure antidopaminergic compounds (e.g., domperidone) [12]. In the case of levosulpiride, however, agonism at the 5-HT 4 receptor may contribute together with antagonism at D 2 receptors to the overall gastrointestinal prokinetic profile of the molecule in vivo.
Conflict of interest statement None declared.
List of abbreviations ACh, acetylcholine; DMSO, dimethyl sulphoxide; 5-HT 4 , 5-hydroxytryptamine4; LMMPs, longitudinal muscle-myenteric plexus preparations.
References [1] Corazza GR, Biagi F, Albano O. Levosulpiride in functional dispepsia: a multicentric, double-blind, controlled trial. Ital J Gastroenterol 1996;28:317–23.
250
M. Tonini et al. / Digestive and Liver Disease 35 (2003) 244–250
[2] Corazza GR, Tonini M. Levosulpiride for dispepsia and emesis. A review of its pharmacology, efficacy and tolerability. Clin Drug Invest 2000;19:151–62. [3] Mansi C, Borro P, Giacomini M, Biagini R, Mele MR, Pandolfo N, et al. Comparative effects of levosulpiride and cisapride on gastric emptying and symptoms in patients with functional dyspepsia and gastroparesis. Aliment Pharmacol Ther 2000;14:561–9. [4] Shichijo K, Sakurai-Yamashita Y, Semine I, Taniyama K. Neuronal release of endogenous dopamine from corpus of guinea pig stomach. Am J Physiol 1997;273:G1044–50. ˚ ˚ [5] Eisenhofer G, Aneman A, Friberg P, Hooper D, Fandriks L, Lonroth H, et al. Substantial production of dopamine in the human gastrointestinal tract. J Clin Endocrinol Metab 1997;82:3864–71. [6] Schuurkes JAJ, Van Nueten JM. Is dopamine an inhibitory modulator of gastrointestinal motility? Scand J Gastroenterol 1981;16:33–6. [7] Willems JL, Buylaert WA, Lefebvre RA, Bogaert MG. Neuronal dopamine receptors on autonomic ganglia and sympathetic nerves and dopamine receptors in the gastrointestinal system. Pharmacol Rev 1985;37:165–216. [8] Kusunoki M, Taniyama K, Tanaka C. Dopamine regulation of [ 3 H]acetylcholine release from guinea pig stomach. J Pharmacol Exp Ther 1985;284:713–9. [9] Takeda K, Taniyama K, Kuno T, Sano I, Ishikawa T, Ohmura I, et al. Clebopride enhances contractility of the guinea pig stomach by blocking peripheral D 2 dopamine receptor and alpha-2 adrenoceptor. J Pharmacol Exp Ther 1991;257:806–11. [10] Takahashi T, Kurosawa S, Wiley JW, Owyang C. Mechanism for the gastrokinetic action of domperidone. In vitro studies in guinea pig. Gastroenterology 1991;101:703–10. [11] Tonini M. Recent advances in the pharmacology of gastrointestinal prokinetics. Pharmacol Res 1996;33:217–26. [12] De Ponti F, Malagelada JR. Functional gut disorders: from motility to sensitivity disorders. A review of current and investigational drugs for their management. Pharmacol Ther 1998;80:49–88. [13] Briejer MR, Akkermans LM, Schuurkes JA. Gastrointestinal prokinetic benzamides: the pharmacology underlying stimulation of motility. Pharmacol Rev 1995;47:631–51. [14] Chemnitius JM, Haselmeyer KH, Gonska BD, Kreuzer H, Zech R. Indirect parasympathomimetic activity of metoclopramide: reversible inhibition of cholinesterases from human central nervous system and blood. Pharmacol Res 1996;34:65–72. [15] Foxx-Orenstein AE, Kuemmerle JF, Grider JR. Distinct 5-HT receptors mediate the peristaltic reflex induced by mucosal stimuli in human and guinea pig intestine. Gastroenterology 1996;111:1281– 90. [16] De Ponti F, Tonini M. Irritable bowel syndrome. New agents targeting serotonin receptor subtypes. Drugs 2001;63:317–32. [17] McLean PG, Coupar IM, Molenaar P. A comparative study of functional 5-HT 4 receptors in human colon, rat oesophagus and rat ileum. Br J Pharmacol 1995;115:47–56. [18] McLean PG, Coupar IM. Stimulation of cyclic AMP formation in the circular smooth muscle of human colon by activation of 5-HT 4 like receptors. Br J Pharmacol 1996;117:238–9. [19] Gaddum JH. Series of drug antagonism. Pharmacol Rev 1957;9:211–8. [20] Paton WD, Zar MA. The origin of acetylcholine released from guinea-pig intestine and longitudinal muscle strips. J Physiol 1968;194:13–33. [21] Zagorodnyuk V, Maggi CA. Tachykinin NK 1 and NK 2 receptors mediate non-adrenergic non-cholinergic excitatory neuromuscular transmission in the guinea-pig stomach. Neuroscience 1997;80:625– 34. [22] Gale JD, Grossman CJ, Darton J, Bunce KT, Whitehead JWF, Knight J, et al. GR 125487: a selective and high affinity 5-HT 4 receptor antagonist. Br J Pharmacol 1994;113:120P.
[23] Candura SM, Messori E, Franceschetti GP, D’Agostino G, Vicini D, Tagliani M, et al. Neural 5-HT 4 receptors in the human isolated detrusor muscle: effects of indole, benzimidazolone and substituted benzamide agonists and antagonists. Br J Pharmacol 1996;118:1965–70. [24] Matsuyama S, Sakiyama H, Nei K, Tanaka C. Identification of putative 5-hydroxytryptamine(4) (5-HT 4 ) receptors in guinea pig stomach: The effect of TKS159, a novel agonist, on gastric motility and acetylcholine release. J Pharmacol Exp Ther 1996;276:989–95. [25] Tamura T, Sano I, Satoh M, Mizumoto A, Itoh Z. Pharmacological characterization of 5-hydroxytryptamine-induced motor activity (in vitro) in the guinea pig gastric antrum and corpus. Eur J Pharmacol 1996;308:315–24. [26] Taniyama K, Makimoto N, Furuichi A, Sakurai-Yamashita Y, Nagase Y, Kaibara M, et al. Functions of peripheral 5-hydroxytryptamine receptors, especially 5-hydroxytryptamine4 receptor, in gastrointestinal motility. J Gastroenterol 2000;35:575–82. [27] Takeda M, Tsukamoto K, Sakurai-Yamashita Y, Suzuki T, Taniyama K. Facilitation of acetylcholine release by SK-951, a benzofuran derivative, via the 5-hydroxytryptamine4 receptor in guinea pig stomach. Jpn J Pharmacol 2000;82:138–43. [28] Prins NH, van Der Grijn A, Lefebvre RA, Akkermans LM, Schuurkes JA. 5-HT 4 receptors mediating enhancement of contractility in canine stomach; an in vitro and in vivo study. Br J Pharmacol 2001;132:1941–7. [29] Costall B, Naylor RJ, Tan CC. A central site of dopamine agonist action to modify gastric secretion in the rat. Eur J Pharmacol 1985;117:61–9. [30] Schiavone A, Volonte M, Micheletti R. The gastrointestinal motor effect of benzamide derivatives is unrelated to 5-HT 3 receptor blockade. Eur J Pharmacol 1990;187:323–9. [31] Gullikson GW, Virina MA, Loeffler RF, Yang DC, Goldstin B, Wang SX, et al. SC-49518 enhances gastric emptying of solid and liquid meals and stimulates gastrointestinal motility in dogs by a 5hydroxytryptamine4 receptor mechanism. J Pharmacol Exp Ther 1993;264:240–8. [32] Rizzi CA, Sagrada A, Schiavone A, Schiantarelli P, Cesana R, Schiavi GB, et al. Gastroprokinetic properties of the benzimidazolone derivative BIMU 1, an agonist at 5-hydroxytryptamine4 and antagonist at 5-hydroxytryptamine3 receptors. Naunyn Schmiedebergs Arch Pharmacol 1994;349:338–45. [33] Hegde SS, Wong AG, Perry MR, Ku P, Moy TM, Loeb M, et al. 5-HT4 receptor mediated stimulation of gastric emptying in rats. Naunyn Schmiedebergs Arch Pharmacol 1995;351:589–95. [34] Wiseman LR, Faulds D. Cisapride. An updated review of its pharmacology and therapeutic efficacy as a prokinetic agent in gastrointestinal motility disorders. Drugs 1994;47:116–52. [35] Wardle KA, Sanger GJ. The guinea-pig distal colon—a sensitive preparation for the investigation of 5-HT 4 receptor-mediated contractions. Br J Pharmacol 1993;110:1593–9. [36] Leung E, Pulido-Rios MT, Bonhaus DW, Pekins LA, Zeitung KD, Hsu SA, et al. Comparison of 5-HT 4 receptors in guinea-pig colon and rat oesophagus: effects of novel agonists and antagonists. Naunyn Schmiedebergs Arch Pharmacol 1996;354:145–56. [37] Briejer MR, Schuurkes JA. 5-HT 3 and 5-HT 4 receptors and cholinergic and tachykinergic neurotransmission in the guinea-pig proximal colon. Eur J Pharmacol 1996;308:173–80. [38] Kojima S, Shimo Y. Investigation into the 5-hydroxytryptamineinduced atropine-resistant neurogenic contraction of guinea-pig proximal colon. Br J Pharmacol 1996;117:1613–8. [39] Grider JR, Foxx-Orenstein AE, Jin JG. 5-Hydroxytryptamine4 receptor agonists initiate the peristaltic reflex in human, rat, and guinea pig intestine. Gastroenterology 1998;115:370–80. [40] Jin JG, Foxx-Orenstein AE, Grider JR. Propulsion in guinea pig colon induced by 5-hydroxytryptamine (HT) via 5-HT 4 and 5-HT 3 receptors. J Pharmacol Exp Ther 1999;288:93–7.
Alimentary Tract
5-HT 4 Receptors contribute to the motor stimulating effect of levosulpiride in the guinea-pig gastrointestinal tract M. Tonini a , *, R. De Giorgio b , V. Spelta a , G. Bassotti c , A. Di Nucci d , L. Anselmi a , B. Balestra d , F. De Ponti e a
Department of Physiological and Pharmacological Sciences, University of Pavia, Piazza Botta 10 /11, 27100 Pavia, Italy b Department of Internal Medicine and Gastroenterology, University of Bologna, Bologna, Italy c Department of Clinical and Experimental Medicine, University of Perugia, Perugia, Italy d Department of Internal Medicine and Therapeutics, University of Pavia, Pavia, Italy e Department of Pharmacology, University of Bologna, Bologna, Italy Received 2 October 2002; accepted 12 November 2002
Abstract Background. The dopamine D 2 receptor antagonist levosulpiride is a substituted benzamide derivative, whose gastrokinetic properties are exploited clinically for the management of functional dyspepsia. However, for other benzamide derivatives, such as cisapride and mosapride, agonism towards serotonin 5-HT 4 receptors is considered the main mechanism leading to gastrointestinal prokinesia. Aims. To assess whether levosulpiride is able to activate 5-HT 4 receptors in the guinea-pig isolated gastrointestinal tract. Materials and methods. Circular muscle strips from gastric antrum, and colonic longitudinal muscle strips were used to detect electrically stimulated neurogenic contractions. The effect of levosulpiride was assessed in the absence and presence of GR125487, a selective 5-HT 4 receptor antagonist. Furthermore, potential interaction of levosulpiride with 5-HT 3 receptors and tissue cholinesterases was assessed in unstimulated ileal longitudinal muscle-myenteric plexus preparations. Results. Antral and colonic strip contractions were cholinergic / tachykinergic in nature. Micromolar concentrations of levosulpiride potentiated submaximal responses, through a mechanism competitively antagonized by GR125487 (pKB 59.4). In LMMPs, levosulpiride slightly affected contractions caused by the 5-HT 3 receptor agonist 2-methyl-5-HT, and had no effect on contractions to exogenous acetylcholine. Conclusions. Our results indicate that levosulpiride acts as a moderate agonist at the 5-HT 4 receptor. This property, together with antagonism at D 2 receptors, may contribute to its gastrointestinal prokinetic effect. 2003 Editrice Gastroenerological Italiana S.r.l. Published by Elsevier Science Ireland Ltd. All rights reserved. Keywords: Guinea-pig gastrointestinal tract; 5-HT 4 receptors; Levosulpiride
1. Introduction The substituted benzamide derivative levosulpiride is a selective dopamine D 2 receptor antagonist used for the management of upper gastrointestinal tract disorders associated with functional dyspepsia [1–3]. The rationale for proposing antidopaminergic compounds as prokinetic agents was based on the evidence that dopamine is present in the gastrointestinal wall of several mammals [4,5], where it potently inhibits motility [6,7] and release of *Corresponding author. Tel.: 139-0382-506-363; fax: 139-0382-506419. E-mail address: [email protected] (M. Tonini).
excitatory neurotransmitters (e.g., acetylcholine (ACh)) [8– 10]. Substituted benzamide gastrointestinal prokinetics belong to a chemical class with a complex pharmacological profile, since they may (metoclopramide) or may not (cisapride, mosapride) possess antidopaminergic properties together with agonism at the serotonin 5-HT 4 receptor (metoclopramide, cisapride, mosapride) and antagonism at the 5-HT 3 receptor (metoclopramide, cisapride and a metabolite of mosapride) [11,12]. Furthermore, some of these drugs have moderate anticholinesterase activity [13,14]. In the gastrointestinal tract, 5-HT 4 receptors are located on myenteric neurons, enterocytes and smooth muscle cells. In the enteric nervous system, the presence of 5-HT 4 receptors has been demonstrated on several func-
1590-8658 / 03 / $30 2003 Editrice Gastroenerological Italiana S.r.l. Published by Elsevier Science Ireland Ltd. All rights reserved. doi:10.1016 / S1590-8658(03)00061-6
M. Tonini et al. / Digestive and Liver Disease 35 (2003) 244–250
tionally distinct subtypes of neural cells including intrinsic primary afferent neurons, ascending cholinergic / tachykinergic excitatory pathways and descending non-adrenergic, non-cholinergic inhibitory pathways. Activation of 5-HT 4 receptors is associated with release of transmitters involved in enteric reflexes and peristalsis, and with water and electrolyte secretion from enterocytes [15,16]. In smooth muscle cells of oesophagus and colon, 5-HT 4 receptors cause cAMP-dependent relaxations [17,18]. It is widely accepted, therefore, that the prokinetic action of several marketed benzamide derivatives depend, at least in part, on the activation of 5-HT 4 receptors [16]. To date, no information on possible interaction of levosulpiride with serotonin receptors and tissue cholinesterases is available. This study was designed to re-assess the pharmacological profile of levosulpiride using isolated tissues from the guinea-pig gastrointestinal tract.
2. Materials and methods Male Dunkin–Hartley guinea-pigs (250–400 g; Bettinardi S.r.l., Momo, Novara, Italy) were used. Animals were housed in standard animal facilities, providing constant temperature (2161 8C), relative humidity (50–55%), alternating 12-h light and dark cycles, and were fed standard laboratory food (Laboratori Dr Piccioni, Gessate, Milan, Italy) and tap water ad libitum.
2.1. Tissue preparation Segments of gastric antrum, distal ileum and distal colon were carefully removed, cleared of their intraluminal contents and placed in a Petri dish containing prewarmed Tyrode solution. All tissues prepared from the above segments (strips of gastric antrum, longitudinal musclemyenteric plexus, and distal colon) reached equilibrium in Tyrode solution maintained at 37 8C and aerated with a mixture of 95% O 2 15% CO 2 for at least 60 min before experiments were begun.
245
transmitters involved in the contractile response were investigated using muscarinic (hyoscine) and tachykinergic antagonists acting at NK1 (SR140333) and NK2 receptors (MEN11420) left in contact with the tissue for 20–40 min before stimulation. Contractions were referred as percentage of the maximal response obtained at 30 Hz. Noncumulative concentration–response curves to levosulpiride (i.e., each drug concentration was removed from the bath by changing the medium at 3-min intervals on three occasions) in potentiating submaximal contractions induced by 5 Hz were constructed in the absence and presence of the 5-HT 4 receptor antagonist GR125487 (incubation time: 20 min). The apparent pKB value (a measure of antagonist affinity) for GR125487 was evaluated using the Gaddum equation [19].
2.3. Longitudinal muscle-myenteric plexus preparations from distal ileum According to the method of Paton and Zar [20], strips of longitudinal muscle with the myenteric plexus attached (LMMPs) were separated from the underlying circular muscle by stroking tangentially away from the mesenteric attachment with a moistened cotton wad. LMMPs were set up isometrically (tension: 5 mN) in a 5-ml organ bath. In unstimulated preparations, non-cumulative concentration– response curves to the contractile effect of 2-methyl-5-HT (0.3–10 mM), a 5-HT 3 receptor agonist, were obtained in the absence and presence of 10 mM levosulpiride (incubation time: 20 min). Agonist responses were referred as a percentage of the response induced by the maximal 2methyl-5-HT concentration (10 mM). Antagonist affinity for levosulpiride was evaluated as apparent pKB value. In a separate set of experiments, cumulative concentration– response curves to the contractile effects of ACh (i.e., each agonist concentration was added to the bath consecutively, once the previous drug administration produced a plateau response) were obtained in the absence and presence of levosulpiride (100 mM) or eserine (30 nM).
2.4. Longitudinal muscle strips of distal colon 2.2. Circular muscle strips of gastric antrum Segments of gastric antrum were pinned flat on a sylastic holder with the mucosa layer facing upwards. After removal of the mucosa, circular muscle strips (2.5–3 cm long and 0.5 cm wide) were mounted isometrically (tension: 8 mN) in a 5-ml organ bath containing Tyrode solution maintained at 37 8C and aerated with a mixture of 95% O 2 15% CO 2 . Electrical field stimulation was delivered by means of two platinum electrodes placed at the top and bottom of the chamber, respectively. Parameters of stimulation were: trains of pulses delivered at 1-min intervals, 60 V, 1–30 Hz frequency, 0.5 ms duration, for 5 s. In electrically-stimulated preparations, the excitatory
Segments of distal colon were cut open along the mesenteric axis and pinned flat on a sylastic holder with the mucosa layer facing upwards. After removal of the mucosa, longitudinal muscle strips (2.5–3 cm long and 0.5 cm wide) were mounted isometrically (tension: 8 mN) in a 5-ml organ bath containing Tyrode solution maintained at 37 8C and aerated with a mixture of 95% O 2 15% CO 2 . Parameters of electrical field stimulation were: trains of pulses delivered at 1-min intervals, 60 V, 1–30 Hz frequency, 0.5 ms duration, for 5 s. In electrically-stimulated preparations, the excitatory transmitters involved in the contractile response were studied using muscarinic (hyoscine) and tachykinergic antagonists (SR140333 and MEN11420) left in contact with the tissue for 20–40 min
246
M. Tonini et al. / Digestive and Liver Disease 35 (2003) 244–250
before stimulation. Contractions were referred as percentage of the maximal response obtained at 30 Hz. Noncumulative concentration–response curves to levosulpiride in potentiating submaximal contractions induced by 5 Hz were constructed in the absence and presence of the 5-HT 4 receptor antagonist GR125487 (incubation time: 20 min).
2.5. Statistical analysis Data are expressed as means6standard error of the mean (SEM). Differences between means (multiple comparisons) were analysed by one-way analysis of variance (ANOVA). Values of p,0.05 were taken as statistically significant.
2.6. Solutions and drugs Tyrode solution (pH 7.4) had the following composition (mM): NaCl 136.9, KCl 2.7, MgCl 2 1.04, NaHCO 3 11.9, NaH 2 PO 4 0.4, glucose 5.5. The following drugs were used: SR140333 (a selective NK 1 antagonist kindly donated by Dr T. Croci, Sanofi Research, Milan, Italy), MEN11420 (a selective NK 2 antagonist kindly donated by Dr C.A. Maggi, Menarini Research, Florence, Italy), GR125487 (a selective 5-HT 4 antagonist kindly supplied by Dr S.G. Lister, GlaxoSmithKline, Hertfordshire, UK), 2-methyl-5-hydroxytryptamine maleate, S(2)-sulpiride [levosulpiride], R(1)-sulpiride [dextrosulpiride] (Research Biochemical International, Natick, USA), acetylcholine (ACh) chloride, tetrodotoxin (Sigma Diagnostics, St Louis, MO, USA), eserine hemisulphate (ICN Pharmaceuticals Inc., Cleveland, USA), hyoscine hydrochloride [(2)scopolamine] (Sigma-Chemie, Deisenhofen, Germany). All drugs were dissolved in distilled water, with the exception of SR140333, which was previously dissolved in dimethyl sulphoxide (DMSO) (stock solution: 10 mM) and then diluted in distilled water, and levosulpiride, which was dissolved in 0.1 N HCl (stock solution: 100 mM) and then diluted in distilled water. Drugs were administered in volumes not exceeding 1% v / v of the bath volume. The DMSO / water and HCl / water solutions were without effect on gastrointestinal mechanical activity.
3. Results
Fig. 1. Frequency-dependent contractions to electrical field stimulation (1–30 Hz) in isolated circular muscle strips of guinea-pig gastric antrum. Responses are referred as contractions in the absence of pharmacological treatment (control) or in the presence of NK 1 (SR140333: 1 mM) and NK 2 (MEN11420: 1 mM), or NK 1 , NK 2 and muscarinic receptor blockade by hyoscine (1 mM). Values are expressed as percent of the response to 30 Hz stimulation. Any pharmacological treatment caused a significant reduction ( p,0.05 or better: ANOVA for continuous-by-class effects) of the frequency-response curve. Each point represents the mean6SEM of 4–8 determinations.
tion are mediated by acetylcholine activating postjunctional muscarinic receptors and, to a lesser extent, by tachykinins acting at NK 1 and NK 2 receptors [21]. Levosulpiride (3–1000 mM) concentration-dependently enhanced the amplitude of submaximal contractions elicited at 5 Hz without affecting the basal tone. The agonist potency (pD 2 value) was 4.6. The potentiating effect of levosulpiride was competitively antagonized by 3 nM GR125487 (Fig. 2). The apparent pKB value for GR125487 was 9.460.2. Conversely, dextrosulpiride (3– 100 mM) was virtually inactive, although at 100 mM, the drug enhanced twitch contractions by 35% in one out of seven preparations (data not shown). This evidence suggests that levosulpiride, and not dextrosulpiride, possesses moderate agonist properties at the 5-HT 4 receptor in the guinea-pig gastric antrum.
3.1. Circular muscle strips of gastric antrum 3.2. Longitudinal muscle strips of distal colon Electrical field stimulation elicited reproducible twitch contractions in a frequency-dependent (1–30 Hz) manner (Fig. 1). A combination of SR140333 (1 mM) and MEN11420 (1 mM) partially inhibited electrically-induced contractions. Administration of hyoscine (1 mM) in the presence of tachykinin antagonists abolished any contractile response (Fig. 1). These findings indicate that in the gastric antrum, contractile responses to electrical stimula-
Electrical field stimulation elicited reproducible twitch contractions in a frequency-dependent (1–30 Hz) manner (Fig. 3). Contractions were sensitive to a combination of SR140333 (1 mM), MEN11420 (1 mM) and hyoscine (1 mM), which reduced the response by approximately 80%. The administration of tetrodotoxin (TTX: 1 mM) abolished any residual response. These findings indicate that in the
M. Tonini et al. / Digestive and Liver Disease 35 (2003) 244–250
Fig. 2. Concentration–response curves for the potentiating effect of levosulpiride on electrically stimulated (5 Hz) submaximal contractions of antral strips in the absence or presence of 0.3 nM GR125487, a selective 5-HT 4 receptor antagonist. Values are expressed as percent increase in control contractions. Each point represents the mean6SEM of 5–8 determinations.
distal colon, contractile responses to electrical stimulation are mediated by ACh activating postjunctional muscarinic receptors, by tachykinins acting at NK 1 and NK 2 receptors, and by an unknown excitatory transmitter responsible for approximately 20% of the response. Levosulpiride (3–300 mM) concentration-dependently enhanced the amplitude of submaximal contractions elicited at 5 Hz (Fig. 4) without affecting the basal tone. The agonist potency (pD 2 value) was 5.3. The potentiating effect of levosulpiride was markedly antagonized by 3 and 10 nM GR125487, suggesting also that in colonic longitudinal strips, levosulpiride behaves as a moderate agonist at the 5-HT 4 receptor. The apparent pKB value for GR125487 evaluated at 3 nM was 9.560.1.
3.3. LMMPs from distal ileum In unstimulated LMMPs, 2-methyl-5-HT (0.3–10 mM) induced neurogenic contractile responses, which were slightly antagonized by 10 mM levosulpiride (Fig. 5) with 20% depression of the maximum agonist response. The apparent pKB value was 4.660.3 (n58), suggesting a very slight antagonism at the 5-HT 3 receptor, which appears partial or non-surmountable in nature. In a separate set of experiments, cumulative administration of ACh (0.1 nM– 100 mM) caused the preparation to contract. Levosulpiride (100 mM) had no effect on the concentration–response curve to applied ACh (Fig. 6A). Conversely, the contrac-
247
Fig. 3. Frequency-dependent contractions to electrical field stimulation (1–30 Hz) in isolated longitudinal muscle strips of guinea-pig distal colon. Responses are referred to as contractions in the absence of pharmacological treatment (control), or in the presence of NK 1 (SR140333: 1 mM) and NK 2 (MEN11420: 1 mM), or NK 1 , NK 2 and muscarinic receptor blockade by hyoscine (1 mM). Contractions were completely suppressed by tetrodotoxin (TTX: 1 mM). Values are expressed as percent of the response to 30 Hz stimulation. Any pharmacological treatment caused a significant reduction ( p,0.05 or better: ANOVA for continuous-by-class effects) of the frequency-response curve. Each point represents the mean6SEM of five determinations.
tile response curve to ACh was shifted to the left and in a parallel fashion by 30 nM eserine (Fig. 6B). These findings indicate that the potentiating effect of levosulpiride on twitch contractions does not involve inhibition of endogenous cholinesterases.
4. Discussion The results of the present study indicate that in two distinct regions of the guinea-pig gastrointestinal tract, namely the gastric antrum and the distal colon, levosulpiride potentiates electrically-induced excitatory neuromuscular transmission, which is partly mediated by release of ACh acting at postjunctional muscarinic receptors (M 3 ), and tachykinins that activate postjunctional NK 1 and NK 2 receptors. In the gastric antrum, levosulpiride-induced enhancement of submaximal contractile responses to 5 Hz stimulation was stereospecific, since the enantiomer dextrosulpiride was virtually inactive, and was counteracted by the selective 5-HT 4 receptor antagonist GR125487 with an
248
M. Tonini et al. / Digestive and Liver Disease 35 (2003) 244–250
Fig. 4. Concentration–response curves for the potentiating effect of levosulpiride on electrically stimulated (5 Hz) submaximal contractions of colonic longitudinal muscle strips in the absence or presence of 3 and 10 nM GR125487, a selective 5-HT 4 receptor antagonist. Values are expressed as percent increase in control contractions. Each point represents the mean6SEM of 5–8 determinations.
Fig. 5. Concentration–response curves for the contractile response induced by 2-methyl-5-HT in unstimulated longitudinal muscle-myenteric plexus preparations from guinea-pig distal ileum, in the absence and presence of 10 mM levosulpiride. Values are expressed as percent of the response induced by 10 mM 2-methyl-5-HT. Note that levosulpiride caused a slight but significant parallel shift to the right of the 2-methyl-5HT curve ( p,0.05: ANOVA for continuous-by-class effects). Each point represents the mean6SEM of 4–8 determinations.
estimated pKB value of 9.4. This affinity value is similar to that previously reported for GR125487 acting at 5-HT 4 receptors in intestinal and extraintestinal preparations [22,23]. These findings indicate that the enhanced neurogenic excitatory drive to antral circular smooth muscle cells induced by levosulpiride is mediated by 5-HT 4 receptors. Therefore this benzamide derivative possesses 5-HT 4 agonist properties that are qualitatively similar to those of other compounds of the same chemical class, such as metoclopramide, cisapride and mosapride [11,16]. However, compared to cisapride and mosapride, the agonist potency of levosulpiride at the 5-HT 4 receptor is moderate, since this molecule is effective in the micromolar range of concentrations. Previous studies have documented the presence of 5-HT 4 receptors on cholinergic nerves in the guinea-pig stomach where they facilitate the release of ACh as observed in antral, corporal and fundic strips [24–27]. This effect has been postulated to be involved in the regulation of gastric motility. Recently, prucalopride, a benzofuran derivative with potent 5-HT 4 receptor agonist properties, was found to enhance nerve-mediated (cholinergic) contractility of canine proximal stomach both in vitro and in vivo [28]. The important role of 5-HT 4 receptors in mediating the gastroprokinetic action of benzamide derivatives is inferred from studies carried out in different animal species. As an example, benzamide derivatives were found to enhance gastric emptying in dogs and rats [29–33]. Cholinergically mediated enhancement of gastric emptying and gastroduodenal motility induced by benzamide derivatives through 5-HT 4 receptors stimulation were also documented in man [34]. The presence of 5-HT 4 receptors on cholinergic nerves supplying colonic longitudinal muscle has been identified in several mammalian species including humans [28]. Longitudinal muscle preparations of the guinea-pig distal colon should be considered one of the most sensitive in vitro systems to investigate the effects of 5-HT 4 receptor on the enteric excitatory innervation [35]. Indeed, in these preparations, 5-HT 4 receptor activation was found to enhance the release of ACh [35–37] and tachykinins [37,38]. With regard to the functional role of 5-HT 4 receptor activation, agonists such as prucalopride and the amino-guanidine-indole derivative, tegaserod (a partial 5HT 4 agonist), were found to facilitate the release of both ACh and tachykinins leading to the potentiation of distension-evoked standing reflexes and peristalsis [39,40]. Our results in electrically stimulated guinea-pig longitudinal muscle preparations are consistent with the release of ACh and tachykinins, since the administration of hyoscine plus SR140333 and MEN11420 reduced to 20% the amplitude of submaximal contractions to 5 Hz stimulation. In this model system, levosulpiride potentiated in a concentration-dependent manner the submaximal contractile response induced by electrical stimulation through a mechanism competitively antagonized by GR125487 (pKB 59.5). The potency estimate (pD 2 ) for levosulpiride
M. Tonini et al. / Digestive and Liver Disease 35 (2003) 244–250
249
Fig. 6. Concentration–response curves for the contractile effects induced by exogenous ACh in unstimulated longitudinal muscle-myenteric plexus, in the absence and in the presence of 100 mM levosulpiride (A) or 30 nM eserine (B). Note the inefficacy of levosulpiride on the contractile response evoked by ACh. Conversely, in (B), the curve profile of ACh in the presence of eserine is statistically different ( p,0.01: ANOVA for continuous-by-class effects) from the control curve. Each point represents the mean6SEM of 5–8 determinations.
was 5.3, a value significantly lower than that obtained with the full agonist, 5-methoxytryptamine (5-MeOT), in the guinea-pig isolated colon (pD 2 56.9) [38]. This, together with the evidence that the potentiating effect of levosulpiride was lower than that produced by 5-MeOT, strongly indicate that levosulpiride behaves as a partial 5-HT 4 receptor agonist in the guinea-pig gastrointestinal tract. Other well-known properties of substituted benzamide derivatives are the mild / moderate antagonism at the 5-HT 3 receptor [11] and the inhibition of tissue cholinesterases [13,14]. The 5-HT 3 antagonist properties of compounds such as granisetron, tropisetron and ondansetron are used in clinical practice to prevent chemotherapy-evoked vomiting. Furthermore, 5-HT 3 receptor antagonists have recently been proposed for the treatment of diarrhoea-predominant irritable bowel syndrome on account of their anti-propulsive effects in the small and large bowel [16]. With regard to the inhibition of cholinesterases, this effect may potentiate the procholinergic action of some benzamide derivatives (i.e., metoclopramide and cisapride) [13,14], thus contributing to their gastrointestinal prokinetic effects. In our hands, levosulpiride was found to antagonize with low affinity (pKB 54.6) contractile responses induced by the selective 5-HT 3 receptor agonist 2-methyl-5-HT in unstimulated longitudinal muscle and myenteric plexus preparations. This implies that levosulpiride has very mild antagonist properties (with characteristics of partial or non-surmountable antagonism) at the 5-HT 3 receptor. In the same preparations, levosulpiride was ineffective on contractions induced by exogenous ACh even at high concentrations (up to 100 mM), whereas the cholinesterase inhibitor 30 nM eserine significantly shifted the ACh curve to the left.
In conclusion, our results suggest that the gastrointestinal prokinetic agent levosulpiride possesses moderate partial agonist properties at the serotonin 5-HT 4 receptor and very mild antagonist property at the 5-HT 3 receptor, while it has no anticholinesterase activity. It is well recognized that the prokinetic efficacy of 5-HT 4 receptor agonists (e.g. cisapride) is superior to that of pure antidopaminergic compounds (e.g., domperidone) [12]. In the case of levosulpiride, however, agonism at the 5-HT 4 receptor may contribute together with antagonism at D 2 receptors to the overall gastrointestinal prokinetic profile of the molecule in vivo.
Conflict of interest statement None declared.
List of abbreviations ACh, acetylcholine; DMSO, dimethyl sulphoxide; 5-HT 4 , 5-hydroxytryptamine4; LMMPs, longitudinal muscle-myenteric plexus preparations.
References [1] Corazza GR, Biagi F, Albano O. Levosulpiride in functional dispepsia: a multicentric, double-blind, controlled trial. Ital J Gastroenterol 1996;28:317–23.
250
M. Tonini et al. / Digestive and Liver Disease 35 (2003) 244–250
[2] Corazza GR, Tonini M. Levosulpiride for dispepsia and emesis. A review of its pharmacology, efficacy and tolerability. Clin Drug Invest 2000;19:151–62. [3] Mansi C, Borro P, Giacomini M, Biagini R, Mele MR, Pandolfo N, et al. Comparative effects of levosulpiride and cisapride on gastric emptying and symptoms in patients with functional dyspepsia and gastroparesis. Aliment Pharmacol Ther 2000;14:561–9. [4] Shichijo K, Sakurai-Yamashita Y, Semine I, Taniyama K. Neuronal release of endogenous dopamine from corpus of guinea pig stomach. Am J Physiol 1997;273:G1044–50. ˚ ˚ [5] Eisenhofer G, Aneman A, Friberg P, Hooper D, Fandriks L, Lonroth H, et al. Substantial production of dopamine in the human gastrointestinal tract. J Clin Endocrinol Metab 1997;82:3864–71. [6] Schuurkes JAJ, Van Nueten JM. Is dopamine an inhibitory modulator of gastrointestinal motility? Scand J Gastroenterol 1981;16:33–6. [7] Willems JL, Buylaert WA, Lefebvre RA, Bogaert MG. Neuronal dopamine receptors on autonomic ganglia and sympathetic nerves and dopamine receptors in the gastrointestinal system. Pharmacol Rev 1985;37:165–216. [8] Kusunoki M, Taniyama K, Tanaka C. Dopamine regulation of [ 3 H]acetylcholine release from guinea pig stomach. J Pharmacol Exp Ther 1985;284:713–9. [9] Takeda K, Taniyama K, Kuno T, Sano I, Ishikawa T, Ohmura I, et al. Clebopride enhances contractility of the guinea pig stomach by blocking peripheral D 2 dopamine receptor and alpha-2 adrenoceptor. J Pharmacol Exp Ther 1991;257:806–11. [10] Takahashi T, Kurosawa S, Wiley JW, Owyang C. Mechanism for the gastrokinetic action of domperidone. In vitro studies in guinea pig. Gastroenterology 1991;101:703–10. [11] Tonini M. Recent advances in the pharmacology of gastrointestinal prokinetics. Pharmacol Res 1996;33:217–26. [12] De Ponti F, Malagelada JR. Functional gut disorders: from motility to sensitivity disorders. A review of current and investigational drugs for their management. Pharmacol Ther 1998;80:49–88. [13] Briejer MR, Akkermans LM, Schuurkes JA. Gastrointestinal prokinetic benzamides: the pharmacology underlying stimulation of motility. Pharmacol Rev 1995;47:631–51. [14] Chemnitius JM, Haselmeyer KH, Gonska BD, Kreuzer H, Zech R. Indirect parasympathomimetic activity of metoclopramide: reversible inhibition of cholinesterases from human central nervous system and blood. Pharmacol Res 1996;34:65–72. [15] Foxx-Orenstein AE, Kuemmerle JF, Grider JR. Distinct 5-HT receptors mediate the peristaltic reflex induced by mucosal stimuli in human and guinea pig intestine. Gastroenterology 1996;111:1281– 90. [16] De Ponti F, Tonini M. Irritable bowel syndrome. New agents targeting serotonin receptor subtypes. Drugs 2001;63:317–32. [17] McLean PG, Coupar IM, Molenaar P. A comparative study of functional 5-HT 4 receptors in human colon, rat oesophagus and rat ileum. Br J Pharmacol 1995;115:47–56. [18] McLean PG, Coupar IM. Stimulation of cyclic AMP formation in the circular smooth muscle of human colon by activation of 5-HT 4 like receptors. Br J Pharmacol 1996;117:238–9. [19] Gaddum JH. Series of drug antagonism. Pharmacol Rev 1957;9:211–8. [20] Paton WD, Zar MA. The origin of acetylcholine released from guinea-pig intestine and longitudinal muscle strips. J Physiol 1968;194:13–33. [21] Zagorodnyuk V, Maggi CA. Tachykinin NK 1 and NK 2 receptors mediate non-adrenergic non-cholinergic excitatory neuromuscular transmission in the guinea-pig stomach. Neuroscience 1997;80:625– 34. [22] Gale JD, Grossman CJ, Darton J, Bunce KT, Whitehead JWF, Knight J, et al. GR 125487: a selective and high affinity 5-HT 4 receptor antagonist. Br J Pharmacol 1994;113:120P.
[23] Candura SM, Messori E, Franceschetti GP, D’Agostino G, Vicini D, Tagliani M, et al. Neural 5-HT 4 receptors in the human isolated detrusor muscle: effects of indole, benzimidazolone and substituted benzamide agonists and antagonists. Br J Pharmacol 1996;118:1965–70. [24] Matsuyama S, Sakiyama H, Nei K, Tanaka C. Identification of putative 5-hydroxytryptamine(4) (5-HT 4 ) receptors in guinea pig stomach: The effect of TKS159, a novel agonist, on gastric motility and acetylcholine release. J Pharmacol Exp Ther 1996;276:989–95. [25] Tamura T, Sano I, Satoh M, Mizumoto A, Itoh Z. Pharmacological characterization of 5-hydroxytryptamine-induced motor activity (in vitro) in the guinea pig gastric antrum and corpus. Eur J Pharmacol 1996;308:315–24. [26] Taniyama K, Makimoto N, Furuichi A, Sakurai-Yamashita Y, Nagase Y, Kaibara M, et al. Functions of peripheral 5-hydroxytryptamine receptors, especially 5-hydroxytryptamine4 receptor, in gastrointestinal motility. J Gastroenterol 2000;35:575–82. [27] Takeda M, Tsukamoto K, Sakurai-Yamashita Y, Suzuki T, Taniyama K. Facilitation of acetylcholine release by SK-951, a benzofuran derivative, via the 5-hydroxytryptamine4 receptor in guinea pig stomach. Jpn J Pharmacol 2000;82:138–43. [28] Prins NH, van Der Grijn A, Lefebvre RA, Akkermans LM, Schuurkes JA. 5-HT 4 receptors mediating enhancement of contractility in canine stomach; an in vitro and in vivo study. Br J Pharmacol 2001;132:1941–7. [29] Costall B, Naylor RJ, Tan CC. A central site of dopamine agonist action to modify gastric secretion in the rat. Eur J Pharmacol 1985;117:61–9. [30] Schiavone A, Volonte M, Micheletti R. The gastrointestinal motor effect of benzamide derivatives is unrelated to 5-HT 3 receptor blockade. Eur J Pharmacol 1990;187:323–9. [31] Gullikson GW, Virina MA, Loeffler RF, Yang DC, Goldstin B, Wang SX, et al. SC-49518 enhances gastric emptying of solid and liquid meals and stimulates gastrointestinal motility in dogs by a 5hydroxytryptamine4 receptor mechanism. J Pharmacol Exp Ther 1993;264:240–8. [32] Rizzi CA, Sagrada A, Schiavone A, Schiantarelli P, Cesana R, Schiavi GB, et al. Gastroprokinetic properties of the benzimidazolone derivative BIMU 1, an agonist at 5-hydroxytryptamine4 and antagonist at 5-hydroxytryptamine3 receptors. Naunyn Schmiedebergs Arch Pharmacol 1994;349:338–45. [33] Hegde SS, Wong AG, Perry MR, Ku P, Moy TM, Loeb M, et al. 5-HT4 receptor mediated stimulation of gastric emptying in rats. Naunyn Schmiedebergs Arch Pharmacol 1995;351:589–95. [34] Wiseman LR, Faulds D. Cisapride. An updated review of its pharmacology and therapeutic efficacy as a prokinetic agent in gastrointestinal motility disorders. Drugs 1994;47:116–52. [35] Wardle KA, Sanger GJ. The guinea-pig distal colon—a sensitive preparation for the investigation of 5-HT 4 receptor-mediated contractions. Br J Pharmacol 1993;110:1593–9. [36] Leung E, Pulido-Rios MT, Bonhaus DW, Pekins LA, Zeitung KD, Hsu SA, et al. Comparison of 5-HT 4 receptors in guinea-pig colon and rat oesophagus: effects of novel agonists and antagonists. Naunyn Schmiedebergs Arch Pharmacol 1996;354:145–56. [37] Briejer MR, Schuurkes JA. 5-HT 3 and 5-HT 4 receptors and cholinergic and tachykinergic neurotransmission in the guinea-pig proximal colon. Eur J Pharmacol 1996;308:173–80. [38] Kojima S, Shimo Y. Investigation into the 5-hydroxytryptamineinduced atropine-resistant neurogenic contraction of guinea-pig proximal colon. Br J Pharmacol 1996;117:1613–8. [39] Grider JR, Foxx-Orenstein AE, Jin JG. 5-Hydroxytryptamine4 receptor agonists initiate the peristaltic reflex in human, rat, and guinea pig intestine. Gastroenterology 1998;115:370–80. [40] Jin JG, Foxx-Orenstein AE, Grider JR. Propulsion in guinea pig colon induced by 5-hydroxytryptamine (HT) via 5-HT 4 and 5-HT 3 receptors. J Pharmacol Exp Ther 1999;288:93–7.