Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

The propulsion of colonic contents is mediated by a propagated peristaltic reflex initiated by muscle stretch or mucosal stimulation. The reflex involves sequential activation of sensory neurons coupled via modulatory interneurons to ascending excitatory and descending inhibitory motor neurons. Direct measurements of neurotransmitter release during the ascending and descending phases of the reflex indicate that excitatory motor neurons release acetylcholine and/or the tachykinins, Substance P, and neurokinin A, whereas inhibitory motor neurons release vasoactive intestinal peptide and its homologs, peptide histidine isoleucine and pituitary adenylate cyclase activating peptide, as well as nitric oxide (Grider and Makhlouf, 1986; Grider, 1989, 1993; Grider et al., 1994). Interneurons organized into a modulatory circuit release somatostatin, opioid peptides (chiefly methionine enkephalin), -aminobutyric acid, and acetylcholine (Grider, 1994b). Recent studies (Grider, 1994a; Grider and Jin, 1994) indicate that calcitonin gene-related peptide (CGRP), a primary marker of sensory pathways, is released by extrinsic sensory neurons activated by muscle stretch and by intrinsic sensory neurons activated by mucosal stimulation.

The effect of mucosal stimuli is mediated by release of 5-hydroxytryptamine (HT) from enterochromaffin cells. Evidence in favor of this notion may be summarized as follows. 1) Mucosal stimulation releases both 5-HT and CGRP, whereas muscle stretch releases only CGRP (Grider et al., 1996; Foxx-Orenstein et al., 1996). 2) The release of CGRP induced by mucosal stimulation is inhibited by 5-HT₄ antagonists in human intestine and rat colon, and by both 5-HT₄ and 5-HT₃ antagonists in guinea pig colon, implying that release of 5-HT is coupled to release of CGRP (Grider et al., 1996; Foxx-Orenstein et al., 1996). 3) Application of 5-HT to the mucosa stimulates CGRP release and induces ascending contraction and descending relaxation in a concentration-dependent fashion. 4) 5-HT-induced CGRP release and both phases of the peristaltic reflex are blocked by 5-HT₄ antagonists in human intestine and rat colon, and by both 5-HT₄ and 5-HT₃ antagonists in guinea pig colon (Grider et al., 1996; Foxx-Orenstein et al., 1996). 5) The effects of 5-HT on CGRP release and both phases of the peristaltic reflex in human, rat, and guinea pig are reproduced by selective 5-HT₄ agonists (Grider et al., 1998). 6) Release of excitatory (Substance P) and inhibitory (vasoactive intestinal peptide) motor neurotransmitters and both phases of the peristaltic reflex induced by mucosal stimulation are inhibited by CGRP antagonists, and by 5-HT₄ antagonists in human intestine and rat colon, and by both 5-HT₄ and 5-HT₃ antagonists in guinea pig colon; release of these neurotransmitters and both phases of the peristaltic reflex induced by muscle stretch are inhibited only by CGRP antagonists (Grider et al., 1996; Foxx-Orenstein et al., 1996).

Taken together, these results indicate that 5-HT release constitutes the earliest step in the transduction of mucosal stimuli that initiate peristaltic activity. Consistent with this notion, Gershon and coworkers (Kirchgessner et al., 1992; Gershon et al., 1994) showed that mucosal stimuli increase cytochrome oxidase activity and c-fos expression in enteric neurons and that these indices of neural activity are blocked by 5-HT₄ and related 5-HT_1p antagonists.

Studies of propulsion in guinea pig colon suggest that both 5-HT₃ and 5-HT₄ receptors are involved in initiating peristaltic activity (Kadowaki et al., 1996; Wade et al., 1996). However, the selective 5-HT₃ antagonist, ondansetron, and the selective 5-HT₄ antagonist, [I-[2-(methylsulfonyl amino)ethyl]-4-piperidinyl]methyl 1-methyl-lH-indole-3-carboxylate, maleate salt (GR 113808A), were not effective in inhibiting propulsive activity when used separately but were effective when used in combination (Kadowaki et al., 1996). High concentrations of the mixed 5-HT₃/5-HT₄ antagonists, tropisetron, and 2-methoxy-4-amino-5-chloro-benzoic acid 2-(diethylamino) ethyl ester (SDZ 205557) were also effective (Kadowaki et al., 1996; Wade et al., 1996). On the other hand, as noted above, 5-HT₃ and 5-HT₄ antagonists caused inhibition of sensory and motor neurotransmitter release and both phases of the peristaltic reflex when used separately, and their effects were additive in combination (Foxx-Orenstein et al., 1996).

In the present study, we used an isolated segment of guinea pig colon to examine the effects of selective 5-HT₄ agonists [5-methoxy-indole-3-carboxaldehyde amino(pentylamino) methylene hydrazone hydrogen maleate (HTF 919) and 4-amino-5-chloro-2,3-dihydro-N-[l-(3-methoxy propyl)-4-piperidinyl]-7-benzofurancarboxamide monohydrochloride (R093877)], and selective 5-HT₃ [1-methyl-N-(8-methyl-8-azabicyclo[3.2.l]-oct-3-yl)-lH-indazole-3-carboxamide maleate (LY 278584)] and 5-HT₄ (GR 113808A) antagonists on propulsive activity. The results show that 5-HT₃ and 5-HT₄ antagonists, separately and additively, inhibit propulsive activity, implying the participation of endogenous 5-HT and its interaction with both 5-HT₃ and 5-HT₄ receptors. Both selective 5-HT₄ agonists, as well as 5-HT in the presence of the 5-HT_2a antagonist, ketanserin, stimulated propulsive activity in a concentration-dependent fashion.

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

The colon of male guinea pigs (weight 150-200 g) was removed and the proximal 3 to 4 cm and distal 1 to 2 cm were discarded. The remaining, pellet-containing portion was incubated at 37°C for 30 min in Krebs-bicarbonate medium to allow spontaneous evacuation of the pellets. The composition of the medium was 118 mM NaCl, 4.8 mM KCl, 1.2 mM KH₂PO₄, 1.2 mM MgSO₄, 2.5 mM CaCl₂, 25 mM NaHCO₃, and 11 mM glucose. The distal colon was then cut into two equal segments; each segment was secured with pins placed at intervals through the attached mesentery as described previously (Grider et al., 1998). Preliminary studies showed no differences between the two segments either in the initial velocity of propulsion or the response to test agents. Each segment was perfused at a rate of 0.25 ml/min for 30 min with oxygenated Krebs-bicarbonate medium using a PE-10 catheter inserted through the caudad end and advanced 2 to 3 cm. Preliminary studies using perfusion rates ranging from 0.1 to 0.5 ml/min indicated that 0.25 ml/min was the maximal rate that did not, by itself, affect the velocity of pellet propulsion. After the 30-min equilibration period, an artificial fecal pellet similar in shape and size to colonic fecal pellets (10-mm long × 4-mm wide) was inserted into the orad end of the segment and the catheter advanced just caudad to the artificial fecal pellet. The pellet was allowed to pass spontaneously until it exited the caudad end of the segment, pushing the catheter out ahead of it. The velocity of propulsion was calculated from the time taken by a pellet, which was easily visible through the translucent wall of the colon, to traverse a 3-cm segment marked with small dissecting pins placed in the mesentery adjacent to the colonic segment. A second fecal pellet was then placed in the orad end, the catheter reinserted through the caudad end and advanced through the lumen of the segment until it was just distal to the fecal pellet, and the measurement of propulsion velocity repeated. The control (basal) velocity of propulsion was first determined using three successive pellets inserted at 5-min intervals. The segments were then allowed to equilibrate again in fresh Krebs-bicarbonate solution for 30 min without luminal perfusion.

The effects of various agents on the velocity of pellet propulsion were examined by adding the agents to the luminal perfusate. The perfusion of 5-HT agonists was begun 2 min before, whereas that of 5-HT antagonists was begun 15 min before, insertion of fecal pellets into the orad end. Separate colonic segments were used to examine each concentration of test agent or combination of test agents.

The following test agents were used: ketanserin, a selective 5-HT_2a receptor antagonist; LY 278584, a selective 5-HT₃ receptor antagonist; SDZ 205557, a preferential 5-HT₄ receptor antagonist; GR 113808A, a selective 5-HT₄ receptor antagonist; HTF 919, a selective 5-HT₄ agonist; R093877, a selective 5-HT₄, agonist; and the endogenous agonist, 5-HT.

Data Analysis. Results were expressed as percentage of control (basal) velocity in millimeters per second. The concentration causing 50% of maximal response (EC₅₀) was calculated for the concentration-response curves using the p.fit 6.0 program (Elsevier, Cambridge). Values are means ± S.E.M. of n experiments, where n represents the number of colonic segments. Separate colonic segments were used for each agonist or antagonist at each separate concentration. Statistical significance was tested by Student's t test for paired and unpaired values.

Materials. LY 278584 was purchased from Research Biochemicals International (Natick, MA); 5-HT, ketanserin, and all other chemicals were obtained from Sigma Chemical Co. (St. Louis, MO.). SDZ 205557 and HTF 919 (ZELMAC) were gifts from Drs. D. Romer and H.-J. Pfannkuche, Novartis Ltd. (Basel, Switzerland). R093877 (Prucalopride) was a gift from Drs. J. Schuurkes and M. Janssen (Janssen Research Foundation, Beerse, Belgium), and GR 113808A was a gift from Drs. G. Kilpatrick and B. Bain, (Glaxo Research and Development, Middlesex, UK).

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

Control Velocity of Pellet Propulsion. The basal velocity of pellet propulsion was constant for segments obtained from the same colon but differed from one colon to another (range 0.5-3.3 mm/s with a mean ± S.E.M. of 1.3 ± 0.1 mm/s; n = 255 pellets in 86 experiments). Intraluminal perfusion with Krebs-bicarbonate medium did not affect the velocity of pellet propulsion (1.5 ± 0.3 mm/s in the absence versus 1.4 ± 0.3 mm/s in the presence of intraluminal perfusion; n = 5).

Decrease in Velocity of Pellet Propulsion Induced by 5-HT₃ and 5-HT₄ Antagonists. Addition of 5-HT itself or selective 5-HT receptor agonists or antagonists to the serosal bathing medium had no effect on the velocity of pellet propulsion (data not shown). In contrast, intraluminal perfusion with either the selective 5-HT₃ receptor antagonist, LY 278584, or the 5-HT₄ receptor antagonist, GR 113808A, caused a concentration-dependent decrease in the velocity of pellet propulsion (EC₅₀: 68.1 ± 1.1 nM and 138.2 ± 5.7 nM for GR 113808A and LY 278584, respectively) (Fig. 1). At the highest concentration tested (10 µM), LY 278584 decreased the velocity of pellet propulsion by 47 ± 1% (control velocity: 1.2 ± 0.2 mm/s; velocity in the presence of antagonist: 0.7 ± 0.1 mm/s) (Fig. 2) and GR 113808A decreased the velocity of pellet propulsion by 39 ± 2% (control velocity: 1.6 ± 0.3 mm/s; velocity in the presence of antagonist: 1.0 ± 0.2 mm/s). At the same concentration (10 µM), the preferential 5HT₄ antagonist, SDZ 205557, which exhibits 5-HT₃ antagonism at higher concentrations, decreased the velocity of pellet propulsion by 43 ± 3% (control velocity: 1.4 ± 2.3 mm/s; velocity in the presence of antagonist: 0.8 ± 0.1 mm/s) (Fig. 2). The combination of the 5-HT₃ antagonist, LY 272584, with a selective or preferential 5-HT₄ antagonist was additive, eliciting 82 ± 3% to 84 ± 4% decrease in the velocity of pellet propulsion (Fig. 2).

View larger version (17K): [in this window] [in a new window]   Fig. 1.   Concentration-dependent decrease in the velocity of pellet propulsion in segments of guinea pig colon induced by 5-HT3 and 5-HT4 antagonists. Three artificial clay pellets were inserted at 5-min intervals into the orad end of an isolated colonic segment perfused with Krebs medium. The initial velocity was calculated from the time taken to traverse a fixed 3-cm length. After a 30-min interval, the selective 5-HT3 antagonist, LY 278584 (0.01-10 µM) or the selective 5-HT4 antagonist, GR 113808A (0.01-10 µM) were added to the perfusate and the velocities of three additional artificial pellets determined. The velocity of pellet propulsion in the presence of each antagonist was expressed as the percentage of the initial velocity. Values are means ± S.E.M. of four experiments for each concentration of antagonist. Asterisks, significant decrease in pellet velocity from initial value (* p < .05; **p < .005).

View larger version (14K): [in this window] [in a new window]   Fig. 2.   Decrease in the velocity of pellet propulsion in segments of guinea pig colon induced by 5-HT4 and 5-HT3 antagonists. A, three artificial clay pellets were inserted at 5-min intervals into the orad end of an isolated colonic segment perfused with Krebs medium followed by intraluminal perfusion with the selective 5-HT3 antagonist, LY 278584 (10 µM) and/or the selective 5-HT4 antagonist, GR 113808A (10 µM) as described in the legend to Fig. 1. In control studies, no test agents was added to the Krebs medium (). B, in other studies, the selective 5-HT3 antagonist, LY 278584 (10 µM), and/or the preferential 5-HT4 antagonist, SDZ 205557 (10 µM) were added to the perfusate. Data are means ± S.E.M. of three to five experiments. Each point was significantly different from control, p < .01. The effect of the combination of antagonists was significantly different from the effect of each antagonist separately; p < 0.01.

Increase in Velocity of Pellet Propulsion Induced by 5-HT and Selective 5-HT₄ Agonists. Intraluminal perfusion with 0.1 µM 5-HT for 2 min before insertion of pellets contracted the segment and prevented pellet propulsion (Fig. 3). However, after intraluminal perfusion with the 5-HT_2a antagonist, ketanserin (1 µM), for 15 min, addition of 5-HT (0.1 µM) to the perfusate for 2 min increased the velocity of pellet propulsion by 48 ± 5% (control velocity: 1.1 ± 0.1 mm/s; velocity in the presence of 5-HT and ketanserin: 1.7 ± 0.2 mm/s; p < 0.01; n = 4). Intraluminal perfusion with ketanserin (1 µM) alone had no effect on the velocity of pellet propulsion (control velocity: 1.1 ± 0.2 mm/s; velocity in the presence of ketanserin: 1.0 ± 0.2 mm/s; N.S.; n = 3). Suppression of propulsion when 5-HT was perfused alone appeared to reflect the direct contractile effect of 5-HT acting via 5-HT_2a receptors on smooth muscle cells (Briejer et al., 1993, 1995; Kuemmerle et al., 1992).

View larger version (17K): [in this window] [in a new window]   Fig. 3.   Increase in the velocity of pellet propulsion in segments of guinea pig colon induced by 5-HT in the presence of a 5-HT2a antagonist. The effect of 5-HT (0.1 µM) alone on the velocity of pellet propulsion was compared with the effect of a combination of 5-HT (0.1 µM) and ketanserin (1 µM). The velocity of propulsion of artificial clay pellets was measured as described in the legend to Fig. 1. Data are means ± S.E.M. of four experiments. Asterisks, significant increase in velocity with 5-HT and ketanserin (p < 0.01). Propulsion was suppressed by 5-HT alone.

View larger version (18K): [in this window] [in a new window]   Fig. 4.   Increase in the velocity of pellet propulsion in segments of guinea pig colon induced by selective 5-HT4 agonists. The velocity of pellet propulsion was measured as described in the legend to Fig. 1. Addition of maximally effective concentrations of the selective 5-HT4 agonists, HTF 919 (0.1 µM) or R093877 (1 µM), to the perfusate increased the velocity of propulsion. Results are means ± S.E.M. of three to four experiments. Asterisks, significant increase above basal velocity, p < .01.

View larger version (20K): [in this window] [in a new window]   Fig. 5.   Concentration-dependent increase in velocity of pellet propulsion elicited by selective 5-HT4agonists and by a combination of 5-HT and 5-HT2 antagonist. The velocity of propulsion of artificial clay pellets was measured as described in the legend to Fig. 1. R093877 was less potent than HTF 919 and appeared to be a partial agonist. Higher concentrations of all agonists did not elicit an increase in velocity. Results are means ± S.E.M. of three to four experiments.

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

This study shows that endogenous 5-HT is involved in the regulation of propulsive activity in guinea pig colon. The effect of 5-HT is mediated separately and additively by 5-HT₃ and 5-HT₄ receptors. Selective 5-HT₃ and 5-HT₄ antagonists decreased the velocity of pellet propulsion in a concentration-dependent fashion when used singly and virtually abolished propulsive activity when used in combination. The participation of both receptor types is consistent with our earlier studies on the regulation of the peristaltic reflex in guinea pig colon (Foxx-Orenstein et al., 1996). Unlike rat or human where the effect of 5-HT on the peristaltic reflex is mediated exclusively by 5-HT₄ receptors, the effect of 5-HT in guinea pig is mediated additively by 5-HT₃ and 5-HT₄ receptors (Grider et al., 1996).

Our findings differ to some extent from those of Kadowaki and coworkers (1996) who applied the antagonists to the serosal side of the isolated colonic segment. Under these conditions, concurrent application of selective 5-HT₃ and 5-HT₄ antagonists abolished propulsive activity, whereas separate application of the antagonists had no effect; high concentrations of mixed 5-HT₃/5-HT₄ antagonists (e.g., tropisetron, SDZ 205557 and FK 1052), however, decreased the velocity of pellet propulsion (Kadowaki et al., 1996; Wade et al., 1996). The difference between our results and those of Kadowaki et al. (1996) probably reflects the accessibility of the antagonists to the site of action of 5-HT on sensory nerve endings in the mucosa. Nonetheless, both this study and that of Kadowaki and coworkers (1996) demonstrated the involvement of both 5-HT₃ and 5-HT₄ receptors in mediating the effect of endogenous 5-HT on propulsive activity in guinea pig colon.

The physiological involvement of 5-HT in regulating propulsive activity identified by the use of selective antagonists was corroborated by pharmacological studies with selective 5-HT₄ agonists; 5-HT₃ agonists with equivalent selectivity are not available. Application of the selective 5-HT₄ agonists to colonic mucosa caused a concentration-dependent increase in the velocity of pellet propulsion. Significant effects were observed at nanomolar concentrations, emphasizing the potency of 5-HT₄ agonists as well as 5-HT when applied to the mucosa. Earlier studies (Grider et al., 1998) had shown that these agents stimulate release of the sensory neurotransmitter, CGRP, and initiate a peristaltic reflex (i.e., ascending contraction and descending relaxation) when applied in similar concentrations to the mucosa of compartmented, flat-sheet preparations of rat or guinea pig colon and human small intestine. R093877 was less potent than HTF 919 and appeared to be a partial agonist in initiating the peristaltic reflex and increasing propulsive activity.

When applied to the mucosa, 5-HT caused contraction of the colonic segment and abolished pellet propulsion. This was attributed to the direct contractile effect of 5-HT on smooth muscle cells of the circular layer resulting in closure of the lumen (Kuemmerle et al., 1992). When this effect was suppressed by preapplication of the 5-HT_2a antagonist ketanserin to the mucosa, 5-HT caused a concentration-dependent increase in the velocity of pellet propulsion. Higher concentrations of 5-HT₄ agonists or 5-HT in the presence of ketanserin did not cause an increase in the velocity of propulsion. It is possible that at higher concentrations of 5-HT and 5-HT₄ agonists, an increase in the velocity of propulsion was offset by the direct relaxant effect of these agents on smooth muscle (Kuemmerle et al., 1992). An alternative but not exclusive possibility is the rapid desensitization of 5-HT₄ receptors located on sensory neurons (Wade et al., 1996). Our preliminary studies suggest, in effect, that these receptors are rapidly desensitized by HTF 919, implying that the 5-HT₄ receptor subtype mediating the effect of 5-HT on peristalsis is the rapidly desensitizing, long-splice variant of the 5-HT₄ receptor (Gerald et al., 1995; Grider, 1998). It is worth noting that the same rapidly desensitizing, long-splice variant of the 5-HT₄ receptor mediates the direct relaxant effect of 5-HT on dispersed intestinal smooth muscle cells in human and guinea pig (Kuemmerle et al., 1996); the relaxant effect is usually masked by the predominant contractile effect mediated by 5-HT_2a receptors (Kuemmerle et al., 1992; Briejer et al., 1995).

As noted above, 5-HT is involved in mediating the peristaltic reflex induced by mucosal stimulation, but not that induced by muscle stretch. Propulsive activity initiated by the presence of artificial pellets in the colonic lumen appeared to reflect activation of mucosal sensory pathways, because propulsion of these pellets was abolished by a combination of 5-HT₃ and 5-HT₄ antagonists.

The two selective 5-HT₄ agonists used in the present study have been shown to accelerate transit or stimulate the frequency of defecation in vivo. Briejer and coworkers (1997a, b, c; 1998) showed that R093877 increased the frequency of defecation in conscious dogs and cats; in dogs, this was associated with increase in the frequency of colonic giant migrating complexes. HTF 919 accelerated transit in dog colon (measured with indium¹¹¹-labeled amberlite pellets) (Nguyen et al., 1997) and human colon (measured with radiopaque markers) (Appel et al., 1997); the effect in dogs was evident at low doses and limited to the first hour. The selective 5-HT₄ agonist, RS67506, accelerated transit in the lower gut of mouse (Nagakura et al., 1997a). When 5-HT was used to stimulate transit in rat and mouse, its effect was blocked by a combination of 5-HT₃ and 5-HT₄ antagonists (Nagakura et al., 1997b).

In summary, the presence of fecal pellets in guinea pig colon triggers the release of 5-HT, which acts via both 5-HT₄ and 5-HT₃ receptors to regulate propulsive activity. The results obtained with this preparation confirm those obtained in compartmented preparations by direct measurement of 5-HT and neurotransmitter release and the mechanical components of the peristaltic reflex evoked by mucosal stimulation.