The physiological role of gastrin-releasing peptide (GRP) and of its cognate receptors in regulating the intestinal peristaltic reflex was examined in a three-compartment flat-sheet preparation of rat colon. Mucosal stimulation applied to the central compartment at high, but not low levels of intensity, induced GRP release in the caudad compartment where descending relaxation was measured, but not into the ascending compartment where ascending contraction was measured or into the central compartment where the stimuli were applied. The selective GRP (BB2) receptor antagonist, [d-Phe6,des-Met14]bombesin6–14, inhibited descending relaxation and VIP release in the caudad compartment induced by high but not by low levels of stimulation applied to the mucosa in the central compartment. The selective neuromedin B (BB1) receptor antagonist, BIM-23127, had no effect on descending relaxation or VIP release. Neither the BB1 nor the BB2 antagonist had any effect on ascending contraction or substance P release in the orad compartment. Consistent with the effects of the antagonists on the peristaltic reflex, the BB2 antagonist but not the BB1 antagonist decreased the velocity of propulsion of artificial fecal pellets through isolated segments of guinea pig distal colon. The results indicate that GRP is selectively released from myenteric neurons in descending pathways during the peristaltic reflex and that it acts via BB2 receptors to augment the descending phase of the peristaltic reflex and propulsion.
- enteric nervous system
- vasoactive intestinal peptide
- neuromedin B
gastrin-releasing peptide (GRP) and neuromedin B (NMB) are mammalian neuropeptides that belong to a family of peptides that are structurally related to the amphibian peptides bombesin (BB), ranatensin, and phyllolitorin. These are grouped as one large family on the basis of similarity in the amino acid sequence of the COOH-terminal end of the molecule (reviewed in Refs. 5, 21, and 31). GRP is a 27-amino acid peptide and is most closely related to BB, sharing all but 1 of the last 10 amino acids at the COOH-terminal end. The COOH-terminal end of GRP (GRP18–27) exists as a separate entity, and it is termed neuromedin C (NMC). NMB is a 32-amino acid peptide and is more closely related to the amphibian peptide ranatensin, sharing all but one of the last seven amino acids of the COOH-terminal end. No mammalian homolog of phyllolitorin has yet been identified. NMB and GRP differ in only 3 of the last 10 COOH-terminal amino acids.
GRP and NMB have been shown to have a wide variety of actions. In the central nervous system, they have been implicated in thermoregulation, satiety, and the regulation of circadian rhythm (1, 26, 31, 32). They have been shown to modulate the activity of the immune system, most notably macrophages (8), and have trophic effects on normal and neoplastic cells, including the autocrine stimulation of proliferation by small-cell lung cancer tumor cells (30). In the gastrointestinal tract, they have been shown to participate in the regulation of pancreatic enzyme and gastric acid secretion, stimulation of smooth muscle contraction, and stimulation of the release of a number of gut hormones, including gastrin (5, 19, 20, 29, 31, 36, 37).
The response to GRP, NMB, and related peptides is mediated by four subtypes of BB receptors designated BB1 through BB4, although BB4 is only present in amphibians. All four receptors have been cloned, and BB1–3 have been shown to share ∼50% homology in amino acid sequence (2, 5, 12, 21, 31, 43). All of the BB receptors have been shown to be G protein-coupled receptors that are coupled via Gq/11 to activation of phospholipase C, generation of IP3 and diacylglycerol, and elevation of intracellular calcium (5, 21, 31, 38). The receptors demonstrate different affinities for the endogenous ligands (3, 5, 21, 31, 41). The BB1 receptor has a higher affinity for NMB than GRP and BB, and it is often referred to as the NMB-preferring BB receptor. The BB2 receptor has a higher affinity for GRP and BB than for NMB and is often referred to as the GRP-preferring BB receptor. No endogenous ligand has yet been identified for the BB3 receptor, and it does not bind GRP or NMB with high affinity, although specific peptide agonists for this receptor have been designed (25). BB1 and BB2 receptors are widely distributed in both the central nervous system and in the periphery, whereas the BB3 receptors have a much more limited distribution. Within the gut, the BB1 receptor has been found to be present mainly on smooth muscle cells of rat esophagus, whereas the BB2 receptor is present on pancreatic acinar cells, epithelial neuroendocrine cells, and smooth muscle cells and enteric neurons of stomach, intestine, and colon (5, 20–22, 31, 35, 42, 43).
GRP and NMB are present in a variety of mammalian tissues including brain, spinal cord, and gastrointestinal tract (5, 21, 31). In addition, they are expressed in developing lung and in a number of neuroendocrine tumors such as small-cell lung carcinoma (30). In general, GRP is found in higher concentrations than NMB in the gut, whereas NMB is found in higher concentrations than GRP in brain and spinal cord (5, 21, 31). At the cellular level, GRP and NMB are present in neuronal cell bodies and neurites and in developing neuroendocrine cells. Within the gut, GRP immunoreactivity is present in neurons of the myenteric plexus and to a lesser extent the submucosal plexus. In the myenteric plexus of small intestine and colon, GPR immunoreactivity is present in three types of neurons: long descending interneurons, long descending circular muscle motor neurons, and viscerofugal neurons that project to preverterbral ganglia (4, 9–11, 18, 23, 24, 27, 28, 33, 34, 44). The descending GRP-immunoreative interneurons and motor neurons also contain VIP and nitric oxide (NO) synthase (NOS); long descending interneurons additionally contain choline acetyltransferase (ChAT), depending on the species and region examined. The long descending interneurons project to other GRP-immunoreactive descending interneurons to form long, anally projecting chains of GRP-imunoreactive interneurons; to short, anally projecting somatostatin-immunoreactive interneurons; to longitudinal and circular muscle motor neurons; and to ascending interneurons (9, 18, 24, 33, 34, 44). Although the physiological function of GPR-immunoreactive neurons is not known, the topography and pattern of projections of these neurons suggest that they are likely to have a role in modulating descending motor reflexes.
In the present study, we have examined the physiological role of GRP and its receptors in mediating the main propulsive motility of the gut, the peristaltic reflex. The release of GRP during the ascending and descending phases of the peristaltic reflex was measured in a three-compartment preparation of rat colon, and selective BB1 and BB2 receptor antagonists were used to identify the role of GRP and BB receptors in the regulation of the reflex. GRP release increased during the descending but not ascending phase of the reflex. The BB2 receptor antagonists reduced descending relaxation and release of VIP from inhibitory motor neurons, whereas the BB1 antagonist had no effect. The BB1 and BB2 receptor antagonists had no effect on ascending contraction or release of substance P (SP) from excitatory motor neurons. Consistent with the effects of the antagonists on the peristaltic reflex, the BB2 but not the BB1 antagonist inhibited the velocity of propulsion of artificial fecal pellets through isolated segments of guinea pig distal colon. The results suggest that GRP is released from descending interneurons during the descending phase of the peristaltic reflex, where it acts on BB2 receptors on inhibitory motor neurons to stimulate VIP release, descending relaxation, and propulsion.
Compartmented flat-sheet preparation of rat colon.
A three-compartment, flat-sheet preparation of rat colon was used to measure the peristaltic reflex. As described in detail previously (14), a 5- to 7-cm segment of middle to distal colon was opened along the mesenteric attachment and pinned to the Sylgard base of an organ bath with the mucosal side up. The dimensions were maintained at the in situ length without applying any additional stretch in the circular or longitudinal direction. The preparation was divided into three compartments by vertical partitions sealed with vacuum grease, and 2 ml of a Krebs-bicarbonate medium was added to each compartment. The composition of the medium was (in mM) 118 NaCl, 4.8 KCl, 1.2 KH2PO4, 2.5 CaCl2, 1.2 MgSO4, 25 NaH2CO3, and 11 glucose, to which was added 0.1% bovine serum albumin, 10 μM amastatin, and 1 μM phosphoramidon.
The peristaltic reflex was elicited by stroking the mucosa with a fine brush (2–8 strokes at a rate of 1 stroke/s). Ascending contraction of circular muscle was measured in the orad peripheral compartment, and descending relaxation was measured in the caudad peripheral compartment by using force-displacement transducers attached to the muscle layers. For measurement of neurotransmitter release, mucosal stimuli were applied to the central compartment five times each during a 15-min period; the medium from the central and both peripheral compartments was collected and frozen for subsequent radioimmunoassay of VIP, SP, and GRP. At the end of each experiment, the colonic segment was cut into three sections corresponding to the tissue in the three compartments, blotted dry, and weighed. The mean wet weight of tissue was similar in each compartment, and the overall mean wet weight of tissue in all compartments was 118 ± 12 mg. Experiments were repeated in the presence of GRP and NMB, BB1 (NMB-preferring) receptor anatagonist BIM-23127, and BB2 (GRP-preferring) antagonists [d-Phe6,des-Met14]bombesin6–14 ester and [d-Phe6,Leu-NHEt13,des-Met14]bombesin6–14. Preliminary studies in smooth muscle cells isolated from the circular muscle layer of distal colon indicated that neither the BB1 nor the BB2 antagonists had any effect on the relaxation induced by the inhibitory motor transmitters VIP, pituitary adenylate cyclase-activating peptide (PACAP), and NO or on the contraction induced by the excitatory motor transmitters ACh and SP.
Measurement of SP, VIP, and GRP.
SP was measured as described previously by using antibody RAS-7451 (13, 14). The limit of detection of the assay was 3 fmol/ml, and the IC50 was 20 ± 6 fmol/ml of original sample. The concentration of SP in the samples ranged from 5 to 65 fmol/ml. The antibody reacts fully with SP but does not cross-react with GRP, NMB, neuropeptide Y, neurokinin A, neurokinin B, somatostatin, VIP, or [Met]enkephalin. VIP was measured as described previously by using antibody RAS-7161 (14, 15). The limit of detection of the assay was 3 fmol/ml, and the IC50 was 19 ± 7 fmol/ml of original sample. The concentration of VIP in the samples ranged from 7 to 58 fmol/ml. The antibody reacts fully with VIP but does not cross-react with GRP, neuromedin B, neuropeptide Y, PACAP, peptide histidine isoleucine, secretin, glucagon, somatostatin, SP, or [Met]enkephalin. GRP was measured using antibody T-4349. The limit of detection of the assay was 5 fmol/ml, and the IC50 was 96 ± 12 fmol/ml of original sample. The concentration of GRP in the samples ranged from 12 to 179 fmol/ml. The antibody does not cross-react with gastrin, NMB, gastric inhibitory peptide, VIP, neuropeptide Y, neurokinin A, SP, somatostatin, or [Met]enkephalin.
For each sample, the fmol/ml of neuropeptide measured by radioimmunoassay was normalized to the wet weight of colonic tissue in each compartment and to the length of the collection period and was expressed as femtomoles per 100 mg tissue wet weight per minute. Release during the peristaltic reflex was calculated as the change in femtomoles per 100 mg per minute from release during a 15-min basal period, and significant differences between basal release and release during the peristaltic reflex were determined with the Student’s t-test. In each experiment, a single colonic segment was prepared from each animal and only a single antagonist was tested in each preparation; therefore, n represents the number of experimental segments and the number of animals.
Measurement of the velocity of propulsion in an isolated whole segment of guinea pig colon.
The velocity of propulsion was measured in an isolated whole segment of guinea pig colon by using artificial clay pellets that mimicked in size, shape, and consistency a fecal pellet, as previously described (10a). Krebs-bicarbonate medium similar to that used for measurement of the peristaltic reflex was continuously perfused at a rate of 0.25 ml/min via a PE-10 catheter introduced via the caudad end of the segment. Basal velocity was measured by inserting a pellet into the orad end of the segment and calculating the time it took for it to traverse a fixed distance of 2 cm. At 5-min intervals, a second and then a third pellet were inserted, and basal velocity was determined from the mean of three values. Following the determination of basal velocity, either the BB1 or BB2 antagonist (1 μM) was added to the bathing medium for 15 min, and the velocity of propulsion was measured again in similar fashion. The velocity of propulsion was calculated as millimeters per second.
SP, VIP, the BB1 antagonist BIM-23127 (H-d-2-Nal-Cys-Tyr-d-Trp-Orn-Val-Cys-2-Nal-NH2), the BB2 antagonist [d-Phe6,Leu-NHEt13,des-Met14]bombesin6–14, SP antibody RAS-7451, VIP antibody RAS-7161, GRP antibody T-4349, 125I-VIP, 125I-SP, and 125I-GRP were purchased from Bachem-Peninsula (Torrance, CA). The BB2 antagonist [d-Phe6,des-Met14]bombesin6–14 ester was a gift from Dr. D. H. Coy (Tulane University, New Orleans, LA). All other chemicals and reagents were purchased from Sigma (St. Louis, MO).
GRP release during the peristaltic reflex.
Basal release of GRP was similar in each of the three compartments (range: 0.57 ± 0.02 to 0.71 ± 0.04 fmol·100 mg−1·min−1). Stimulation of the mucosa in the central compartment by high but not low levels of stimulation caused an increase in the release of GRP into the caudad compartment but not into the central or orad compartment. The release of GPR ranged from 88 ± 11% above basal at 4 strokes (P < 0.01) to 269 ± 18% above basal at 8 strokes (P < 0.001) (Fig. 1).
Effect of BB1 and BB2 receptor antagonists on the peristaltic reflex.
The involvement of endogenous GRP in mediating the peristaltic reflex was examined with selective antagonists of the GRP-preferring (BB2) receptor and antagonists of the NMB-preferring (BB1) receptor. Addition of the BB2 receptor antagonists, [d-Phe6,des-Met14]bombesin6–14 ester or [d-Phe6,Leu-NHEt13,des-Met14]bombesin6–14, to the caudad compartment inhibited the descending relaxation of circular muscle elicited by mucosal stroking in the central compartment. The BB2 antagonist [d-Phe6,des-Met14]bombesin6–14 ester at a concentration of 1 μM had no effect on the descending relaxation elicited by low levels of stimulation (i.e., 2 strokes) but significantly inhibited the relaxation elicited by higher levels of stimulation (Fig. 2). Inhibition ranged from 29 ± 5% (P < 0.01) at 4 strokes to 51 ± 6% (P < 0.001) at 8 strokes. Similarly, the BB2 antagonist [d-Phe6,Leu-NHEt13,des-Met14]-bombesin6–14 at a concentration of 1 μM had no effect on the descending relaxation elicited by two mucosal strokes but significantly inhibited the relaxation elicited by higher levels of stimulation (Fig. 3). Inhibition ranged from 19 ± 5% (P < 0.01) at 4 strokes to 42 ± 4% (P < 0.001) at 8 strokes. Addition of the BB2 receptor antagonists to the orad and central compartments had no effect (Figs. 2 and 3). Addition of the BB1 antagonist BIM-23127 (1 μM) to the orad or caudad peripheral compartment or to the central compartment had no effect on the ascending contraction or descending relaxation phases of the peristaltic reflex elicited by any level of mucosal stimulation (Fig. 2).
Effect of BB1 and BB2 receptor antagonists on excitatory (SP) and inhibitory (VIP) neurotransmitter release during the peristaltic reflex.
The role of GRP-containing neurons in modulating the peristaltic reflex was further examined by measuring the release of VIP as a marker of the activation of inhibitory motor neurons that mediate the descending relaxation phase of the peristaltic reflex and by measuring the release of SP as a marker of the activation of excitatory motor neurons that mediate the ascending phase of the peristaltic reflex. We and others (13, 17) have previously shown that descending relaxation of circular muscle is mediated by the corelease of VIP, PACAP, and NO from inhibitory motor neurons and that ascending contraction of circular muscle is mediated by the corelease of ACh and the tachykinins SP and neurokinin A from excitatory motor neurons.
Basal release of VIP into each compartment was similar, ranging from 1.23 ± 0.16 to 1.31 ± 0.17 fmol·100 mg−1·min−1. The basal release of VIP was not affected by either the BB1 or the BB2 receptor antagonists. As previously shown by us (17), mucosal stroking in the central compartment caused a stimulus-dependent increase in the release of VIP into the caudad peripheral compartment, where the descending relaxation phase of the peristaltic reflex was recorded, but not into the central compartment or orad peripheral compartment. Addition of 1 μM of the BB2 antagonist [d-Phe6,des-Met14]bombesin6–14 ester to the caudad compartment caused a significant inhibition in the release of VIP in response to higher levels of stimulation (i.e., 4–8 strokes) but had no effect on VIP release elicited by the lowest level of stimulation, two strokes. The degree of inhibition ranged from 21 ± 4% (P < 0.01) at four strokes to 39 ± 5% (P < 0.001) at eight strokes (Fig. 4). In contrast, addition of 1 μM of the BB1 antagonist BIM-23127 to the caudad compartment had no effect on the release of VIP elicited by mucosal stroking in the central compartment (Fig. 4).
Basal release of SP into each compartment was similar, ranging from 1.82 ± 0.21 to 1.96 ± 0.13 fmol·100 mg−1·min−1. The basal release of SP was not affected by either the BB1 or the BB2 receptor antagonist. As previously shown by us (14, 17), mucosal stroking in the central compartment caused a stimulus-dependent increase in the release of SP into the orad peripheral compartment where the ascending contraction phase of the peristaltic reflex was recorded but not into the central compartment or caudad peripheral compartment. The release of SP into the orad compartment in response to mucosal stimulation was not affected by either the BB2 antagonist [d-Phe6,des-Met14]bombesin6–14 ester or the BB1 antagonist BIM-23127 (Fig. 5).
Effect of BB1 and BB2 receptor antagonists on propulsive activity.
Propulsive activity was measured in isolated whole segments of guinea pig colon by using artificial pellets inserted at the orad end. The velocity of propulsion was measured as the average time taken for three successive pellets inserted at 5-min intervals into the orad end to traverse a fixed distance. The basal velocity of propulsion was 1.02 ± 0.15 mm/s. Addition of 1 μM of the BB2 antagonist [d-Phe6,des-Met14]bombesin6–14 ester to the medium decreased the velocity of propulsion by 29.2 ± 6.2% (0.72 ± 0.10 mm/s; P < 0.01, n = 6), whereas addition of 1 μM of the BB1 antagonist BIM-23127 had no effect (0.97 ± 0.12 mm/s, n = 6).
The results of this study provide evidence that GRP acts as a neurotransmitter in the colon to regulate the peristaltic reflex. GRP is released selectively during the descending phase of the peristaltic reflex in a stimulus-dependent manner. These studies further indicate that GRP acts as an excitatory neurotransmitter in descending pathways to activate inhibitory VIP/PACAP/NOS motor neurons to circular muscle, thereby leading to enhanced descending relaxation and VIP release during peristalsis. The ability of GRP to stimulate the release to VIP from myenteric neurons is consistent with the ability of bombesin to increase intracellular calcium in myenteric neurons (40). Through the use of selective BB1 and BB2 receptor antagonists, the effect of endogenous GRP was shown to be mediated by the GRP-preferring BB2 receptor subtype.
The release of GRP during descending relaxation is consistent with the topography of GRP-immunoreactive neurons of the intestine and colon. These neurons are anally directed and project for distances of up to 30 mm in guinea pig, rat, and mouse where they either enter the circular muscle as motor neurons or terminate on other neurons and are thus interneurons (4, 9–11, 18, 23, 24, 27, 28, 44).
The long circular muscle motor neurons that contain GRP also contain VIP/PACAP/NOS and, depending on species and region, other neuropeptides, such as neurofilament protein and dynorphin (4, 11, 23, 34). Because these neurons release VIP, PACAP, and NO as motor transmitters to cause relaxation of circular muscle, it is unlikely that GRP acts as motor transmitter in these neurons, although it is possible that GRP could act a neuromodulator regulating the release of these other motor transmitters by a presynaptic action. Numerous studies demonstrate that GRP causes direct contraction of smooth muscle cells and muscle strips derived from the circular muscle layer (5, 7, 20, 22, 29, 39). In the three-compartment preparation of rat colon used in the present study, addition of exogenous GRP to either the orad or caudad peripheral compartment caused contraction, sometimes followed by increased spontaneous activity (unpublished observations). This contractile response interfered with measurement of the peristaltic reflex and illustrates the difference between a response elicited by the physiological release of transmitter from myenteric neurons and the more diffuse response to exogenously added transmitter.
Long, anally projecting, GRP-immunoreactive interneurons also contain VIP and NOS and terminate on other neurons in intestine and colon. Detailed studies of the terminations of these interneurons indicate that they innervate other VIP/NOS/GRP-immunoreactive descending interneurons, somatostatin/ChAT-immunoreactive descending interneurons, VIP/PACAP/NOS-immunoreactive motor neurons that innervate the circular muscle layer, and cholinergic motor neurons that innervate the longitudinal muscle layer (4, 10, 11, 18, 24, 28, 34, 44). In addition, descending GRP-immunoreactive interneurons innervate a small population of unidentified ascending interneurons (34).
The sequentially coupled GRP/VIP/NOS interneurons comprise descending chains of long interneurons, whereas sequentially coupled somatostatin/ChAT interneurons comprise descending chains of short interneurons (24, 44). These two descending chains are similar in that they have collateral branches to each other and that they both innervate circular muscle motor neurons. In fact, in recent studies (24, 44), every inhibitory circular muscle motor neuron examined was innervated by GRP/VIP/NOS descending interneurons, which are often observed to form pericellular baskets around cell bodies of the inhibitory motor neurons. In previous studies (15, 17) we have shown that somatostatin is a critical interneuron in the descending pathway mediating the peristaltic reflex. It is released selectively during the descending phase and acts to enhance the release of the inhibitory motor transmitters VIP, PACAP, and NO. Similarly, cholinergic neurons also play a critical role in descending pathways. The nicotinic antagonist hexamethonium has been shown previously to abolish both ascending and descending phases of the peristaltic reflex, thereby indicating that there is at least one nicotinic cholinergic synapse in each phase of the peristaltic reflex (17). This also implies that the nicotinic cholinergic neuron would be in series with the other neurons in the descending pathway.
In the present study, we show that GRP is released selectively during the descending phase of the peristaltic reflex and that endogenous GRP is involved in the regulation of VIP release during the descending phase of the peristaltic reflex. It is worth noting that the release of GRP is most evident at higher levels of stimulation and that there is no release of GRP in response to the lowest level of stimulation, whereas somatostatin is released by all levels of stimulation (17). Consistent with the differential release of GRP by higher levels of stimulation, blockade of the BB2 GRP-preferring receptor had no effect on the release of VIP induced by the lowest level of stimulation but strongly inhibited release induced by high levels of stimulation, whereas somatostatin antiserum inhibited VIP release induced by all levels of stimulation (17). This suggests that the descending GRP pathways may represent alternative or parallel pathways activated at higher levels to stimulation to enhance the release of relaxant transmitter from the inhibitory motor neurons in the circular muscle. In previous studies, we and others (13, 17) have shown that a similar dual regulation of the ascending excitatory phase of the peristaltic reflex occurs such that cholinergic neurons mediate the response to low levels of stimulation; at higher levels of stimulation, SP and neurokinin A are additionally involved.
In previous studies (15, 16), we have demonstrated that the myenteric circuitry that is activated during the descending phase of the peristaltic reflex leads to stimulation of excitatory motor neurons to the longitudinal muscle, such that during the descending phase the longitudinal muscle contracts as the circular muscle relaxes. In the proposed circuit, somatostatin released from descending interneurons inhibited the release of opioid peptides from myenteric interneurons. The opioid peptides normally restrain VIP/PACAP/NOS interneurons coupled to the longitudinal muscle excitatory motor neurons. Thus the decrease in opioid peptide release during the descending phase of the peristaltic reflex would result in an increase in release of VIP/PACAP/NO from excitatory interneurons, leading to stimulation of longitudinal muscle motor neurons and descending contraction of longitudinal muscle. As noted above, the VIP/PACAP/NOS interneurons also contain GRP, and it is these neurons that innervate the longitudinal muscle motor neurons (4, 11, 24, 34). Thus it is also likely that the release of GRP from these descending interneurons may reflect release from collaterals that innervate the longitudinal muscle motor neurons and mediate the reciprocal coupling of longitudinal muscle contraction mediated by release of ACh and circular muscle relaxation mediated by release of VIP.
The functional coupling of GRP and VIP was examined with selective antagonists of the NMB-preferring or BB1 receptor antagonist, BIM-23127 or (H-d-2-Nal-Cys-Tyr-d-Trp-Orn-Val-Cys-2-Nal-NH2), and the GRP-preferring or BB2 receptor antagonists [d-Phe6,des-Met14]bombesin6–14 ester and [d-Phe6,Leu-NHEt13,des-Met14]bombesin6–14. The BB2 antagonists but not the BB1 inhibited descending relaxation of circular muscle elicited by high but not low levels of stimulation. This is consistent with the effect of the BB2 receptor antagonist on the release of the inhibitory motor transmitter, VIP, discussed above. Neither the BB1 antagonists nor the BB2 receptor antagonists had any effect on ascending contraction of circular muscle, consistent with their lack of effect on release of the excitatory motor transmitter SP. Together, these results indicate that endogenous GRP released from enteric neurons during the descending phase of the peristaltic reflex acts on BB2 receptors on inhibitory motor neurons to enhance the release of VIP and leads to descending relaxation of circular muscle.
The peristaltic reflex underlies peristalsis, the main propulsive activity of the colon. The ability of GRP to enhance the descending phase of the peristaltic reflex would be predicted to enhance propulsion of intraluminal contents. In the present study, this was examined in an isolated whole segment of guinea pig colon by measuring the velocity of propulsion of artificial fecal pellets placed in the orad end of the segment. Addition of the GRP/BB2 antagonist to the bathing medium caused an ∼30% decrease in the velocity of propulsion of fecal pellets, consistent with the effect of this antagonist on the peristaltic reflex. Addition of the NMB/BB1 antagonist had no effect on the velocity of propulsion, consistent with the lack of effect of this antagonist on the peristaltic reflex. These results demonstrate the functional role of endogenous GRP in mediating propulsive activity of the colon. Addition of GRP to the bathing medium did not enhance propulsion but rather caused a generalized contraction throughout the segment (unpublished observations). This is most likely due to the noncoordinated effect of GRP on nerves and smooth muscle as opposed to the coordinated release of GRP in the context of a reflex.
The present finding that the action of endogenous GRP in stimulating VIP release and descending relaxation is mediated by the GRP-preferring BB2 receptor rather than the NMB-preferring BB1 receptor is consistent with the primary role of the BB2 receptors in mediating most of the actions of GRP and NMB in the gut. Studies using a variety of selective antagonists of the BB1 and BB2 show that the ability of GRP and NMB to cause contraction of smooth muscle and to stimulate motility of gallbladder, gastric fundus, and intestine in several species, including human, are mediated by the BB2 receptor (5, 7, 20–22, 29, 31, 32). Direct examination of BB receptor subtypes in rat stomach and human colon using autoradiography demonstrate that the BB2 is present in high concentrations in the myenteric plexus, the longitudinal muscle layer, and the circular muscle layer, especially along the inner margin where the interstitial cells of Cajal are located (26, 35). The use of selective antagonists of BB receptor subtypes also indicates that the BB2 receptor mediates the other physiological actions of GRP in the gut such as stimulation of gastric acid secretion, regulation of gastrin and somatostatin release, and stimulation of pancreatic amylase secretion (5, 22, 31, 36, 37). A developmental role of GRP and the BB2 in villous growth and in colon cancer has also been recently demonstrated (6). The exception to the predominance of the BB2 receptor in the gut is the esophageal smooth muscle of the rat and cat in which the contractile effect of GRP and NMB is mediated by the NMB-preferring BB1 receptor subtype (21, 29, 31, 42).
In conclusion, the results of the present study demonstrate that GRP is released during the descending phase of the peristaltic reflex, most likely from chains of long descending interneurons. GRP acts on BB2 receptors in descending pathways that terminate on inhibitory motor neurons to circular muscle to stimulate the release of the relaxant neurotransmitter VIP and to enhance descending relaxation of circular muscle and propulsion of fecal pellets through the distal colon. This pathway is activated at high but not low levels of stimulation and likely acts in parallel with the descending pathway of short somatostatin-immunoreactive interneurons that also mediates the peristaltic reflex.
This work was supported by National Institute of Diabetes and Digestive and Kidney Diseases Grant DK-34153 .
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2004 the American Physiological Society