Enteric GABA: mode of action and role in the regulation of the peristaltic reflex J. R. GRIDER AND G. M. MAKHLOUF Departments of Physiology and Medicine, Medical Grider, J. R., and G. M. Makhlouf. Enteric GABA: mode of action and role in the regulation of the peristaltic reflex. Am. J. Physiol. 262 (Gastrointest. Liver Physiol. 25): G690-G694, 1992.-The mode of action of y-aminobutyric acid (GABA) and the role of myenteric GABA neurons in the regulation of peristalsis were examined in various preparations of rat colonic muscle. GABA had no contractile, relaxant, or modulatory effect on smooth muscle cells isolated from the circular muscle layer. In innervated circular muscle strips, GABA elicited concentration-dependent relaxation accompanied by release of vasoactive intestinal peptide (VIP). Relaxation and VIP release were inhibited by tetrodotoxin and by the GABA* receptor antagonist bicuculline but not by the GABAB receptor antagonist phaclofen. Relaxation was inhibited by the VIP receptor antagonist VIP-( 10-28) implying that VIP release was coupled to muscle relaxation. Relaxation was augmented by atropine implying that GABA also activated cholinergic neurons causing release of acetylcholine that attenuated the relaxant response. This pharmacological profile was evident when GABA was released from intrinsic GABA neurons during peristalsis induced by radial stretch. Blockade of GABA* receptors with bicuculline inhibited the descending relaxation mediated by VIP motor neurons and the ascending contraction mediated by cholinergic motor neurons. Stimulation of these receptors with exogenous GABA had the opposite effect. We conclude that on release from myenteric neurons, GABA acts via GABA* receptors on cholinergic and VIP motor neurons responsible for the two components of the peristaltic reflex. vasoactive intestinal peptide; acetylcholine; myenteric circular muscle; rat; guinea pig; colon; muscle relaxation
plexus;
AUTORADIOGRAPHIC and immunohistochemical studies have demonstrated the presence of y-aminobutyric acid (GABA) predominantly in neurons of the myenteric plexus (7, 20, 21, 28, 33, 34, 37). These neurons have shapes characteristic of Dogiel type I neurons with several short dendritic processes and one long branching axon that projects within the plexus as well as to the underlying circular muscle layer (7, 31-33). Electrophysiologically, the neurons are S-type excitatory neurons that release GABA upon field stimulation (20,22, 23, 36, 37) The effects of GABA in the gut are mediated by bicuculline-sensitive GABAA and phaclofen-sensitive GABAR receptors. GABAA receptors are excitatory and are located on both cholinergic and noncholinergic neurons (5, 6, 9, 24-26, 29, 32, 37); GABAB receptors are inhibitory and appear to be located on cholinergic neurons only (8, 9, 24, 26, 32). Together these receptors account for the effect of GABA on intestinal muscle strips. GABA causes relaxation in these strips preceded by a transient atropine-sensitive contraction; these effects are mediated by bicuculline-sensitive GABAA receptors. In strips stimulated electrically to produce cholinergically mediated “twitch” contraction, GABA acting via GABAB receptors inhibits contraction. GABA has no G690
0193-1857/92
$2.00 Copyright
College of Virginia,
Richmond,
Virginia
23298-0711
contractile or relaxant effect in nerve-free preparations (25). One or more of these effects is probably responsible for the inhibitory effect of GABA antagonists on intestinal propulsion (22, 27, 32). The aim of the present study was to determine the nature of the transmitter responsible for GABA-mediated relaxation and to examine the participation of GABA in the regulation of the peristaltic reflex. The results show that 1) GABA receptors are not present on smooth muscle cells; 2) GABA-induced relaxation is mediated by release of vasoactive intestinal peptide (VIP) from myenteric motor neurons; and 3) GABA neurons are involved in the regulation of the ascending and descending components of the peristaltic reflex. METHODS Preparation of isolated smooth muscle cells. Smooth muscle cells were isolated from the circular muscle layer of the colon by enzymatic digestion as described previously for gastric and intestinal cells (3, 4, 18). The circular and longitudinal layers were separated by blunt dissection, and the mucosa was removed by scraping with forceps. The segment was cut into lcm long strips and incubated for two successive 45-min periods in 15 ml of medium containing 0.1% collagenase (type II) and 0.01% soybean trypsin inhibitor. The composition of the medium was (in mM) 120 NaCl, 4 KCl, 2.6 KHgP04, 2 CaC&, 0.6 25 N-2-hydroxyethylpiperazine-N’-2-ethanesulfonic M&l,, acid (HEPES), 14 glucose, 2.1% Eagle’s essential amino acid mixture, and 0.1% bovine serum albumin. After the second incubation, the partially digested muscle strips were washed with enzyme-free medium and reincubated in 15 ml of enzymefree medium for an additional 30 min to allow the cells to disperse spontaneously. The cells were harvested by filtration through 500-pm Nitex mesh. Measurement of response of isolated colonic muscle cells. The contractile response of isolated muscle cells was determined as described previously (3, 4, 18). An aliquot of cell suspension (0.5 ml) containing -IO* cells was mixed with the test agonist, and the reaction was terminated at 30 s by addition of 1% acrolein. The relaxant response was determined in cells maximally contracted with cholecystokinin octapeptide (CCK-8; 1 nM) as described previously (4). Putative relaxant agents were added alone or in combination 60 s before CCK-8, and the reaction was terminated after an additional 30 s. The lengths of the first 50 randomly encountered cells were measured by scanning micrometry. Contraction was expressed as percent decrease in cell length from control, and relaxation was expressed as percent inhibition of CCK-induced contraction. Control length was the mean cell length obtained in the absence of test agonist. Preparation of colonic muscZe strips. Circular muscle strips (3 mm wide) were obtained from the midcolon and suspended in 3-ml organ chambers containing Krebs-bicarbonate medium maintained at 37°C and gassed with 95% 02-5% CO,. The composition of the medium was (in mM) 118 NaCl, 4.8 KCl, 1.2 KH2P04, 1.2 MgSO,, 2.5 CaCl,, 25 NaHCOs, 11 glucose, and 0.1 % bovine serum albumin. Atropine (0.5 PM) and
0 1992 the American
Physiological
Society
Downloaded from www.physiology.org/journal/ajpgi by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on August 30, 2018. Copyright © 1992 American Physiological Society. All rights reserved.
ENTERIC
GABA
guanethidine (20 PM) were added to the medium. Strips were equilibrated for 60 min at an initial tension of 0.5 g before measurement of isometric response. Measurement of response in colonic muscle strips. The response to GABA was measured by noncumulative addition of GABA in the range of 0.1 PM to 1 mM. A single concentration of GABA was added, and the response was recorded until the maximal relaxation for that concentration was achieved. This usually occured between 30 s and 1 min. The tissue was then washed at 5-min intervals over the next 20 min, during which the tone recovered to basal levels. The next concentration of GABA was then added and the procedure repeated. Any colonic strip failing to recover basal tone was discarded. The concentration-response curve was repeated in other strips in the presence of one of the following: the sodium conductance blocker, tetrodotoxin (1 PM); the VIP receptor antagonist, VIP(10-28) (10 PM); the GABA* receptor antagonist, bicuculline (20 PM); or the GABAB receptor antagonist, phaclofen (0.2 mM). In some strips, atropine was omitted from the medium to determine its effect on the response to GABA. Concentration-response curves were analyzed and the mean effective concentration (E&) calculated from computer-fitted curves using the P.Fit program (Biosoft, Elsevier, Cambridge, UK). In some strips, samples were obtained for measurement of VIP and substance P. The strips were first incubated for 10 min in Krebs-bicarbonate medium containing 1,000 U/ml aprotinin and 20 PM bacitracin to determine basal release of VIP and substance P. Other samples were obtained after addition of 1 mM GABA alone and in combination with 1 PM tetrodotoxin or 20 PM bicuculline. The samples were stored at -20°C for subsequent radioimmunoassay. Measurement of peristaltic reflex in colonic segments. Colonic segments were prepared as described in detail previously (15). Briefly, a 4- to 6-cm segment from rat midcolon was secured horizontally in the bottom of a lo-ml organ bath. The segment was incubated in Krebs-bicarbonate medium, maintained at 37°C and gassed with 95% O,-5% COZ. The ascending contraction component of the peristaltic reflex was elicited by radial stretch of the caudad end of the segment and the descending relaxation component by radial stretch of the orad end. Stretch was applied in the range of 2-10 g for 15-s periods at intervals of 2 min using a hook-and-pulley assembly, and contraction or relaxation of circular muscle was measured with a force-displacement transducer. Measurements were made in the presence of 1 mM GABA, 20 PM bicuculline, or 0.2 mM phaclofen. In some segments, samples were obtained for measurement of VIP after 1,000 U/ml aprotinin, and 20 PM bacitracin were added to the medium. Samples were first obtained for 15 min to determine basal VIP release. Samples were also obtained during descending relaxation in the absence or presence of 1 mM GABA, 20 PM bicuculline, or 0.2 mM phaclofen. The samples were stored at -20°C for subsequent radioimmunoassay. Radioimmunoassay of VIP. VIP was measured by radioimmunoassay as described in detail previously (13-15) using VIP antibody AC115 (final dilution 1:50,000). The limit of detection of the assay was 5 pg/ml of original sample and the mean inhibitory concentration (I&) was 46 t 5 pg/ml. The intra-assay and interassay variability were 5 and IO%, respectively. Radioimmunoassay of substance P. Substance P was measured by radioimmunoassay as described in detail previously (12) using substance P antiserum AC95 (final dilution of 1:2,500). The limit of detection was 10 pg/ml of original sample and the IC,, was 138 t 29 pg/ml. Measurements of PHJGABA release. The release of GABA was measured in colonic segments preloaded with [3H]GABA. The segments were incubated for 60 min in Krebs-bicarbonate
G691
IN PERISTALSIS
medium containing 50 nM [3H]GABA (sp act 34.8 Ci/mmol). Aminooxyacetic acid (10 PM) was added to prevent degradation of GABA, and P-alanine (1 mM) was added to prevent uptake of [“HIGABA into glial cells. The segments were washed five times in Krebs medium and then used for measurement of descending relaxation. The bathing medium was removed and replaced every 5 min and saved for liquid scintillation spectrometry. Samples were collected both before and during a period of 4g stretch. At the end of the experiment, the tissue was dissolved in Soluene and subjected to scintillation spectrometry. GABA release was expressed as a percentage of total [“HIGABA content. Materials. GABA and phaclofen were obtained from Research Biochemicals (Natick, MA); aminooxyacetic acid, ,8alanine, atropine sulfate, bicuculline, bacitracin, tetrodotoxin were from Sigma Chemical (St. Louis, MO). Collagenase type II was from Worthington Biochemicals (Freehold, NJ), and VIP-(10-28) was from Bachem (Torrance, CA). Aprotinin was from Mobay Chemical (New York, NY); 1251-labeled VIP and [“HIGABA were from New England Nuclear (Boston, MA). VIP antiserum AC115 was obtained from Cambridge Research Biochemicals (Wilmington, DE), and substance P antiserum AC95 and P251-labeled substance P were obtained from Peninsula (Belmont, CA). RESULTS
Effect of GABA on isolated colonic muscle cells. There was no detectable effect of GABA on isolated colonic circular muscle cells. GABA (0.1-100 PM) did not elicit contraction or relaxation of muscle cells maximally precontracted with CCK-8. GABA also did not augment or inhibit relaxation induced by VIP (1 nM). Effect of GABA on colonic muscle strips.. GABA caused a concentrationdependent relaxation of colonic circular muscle strips (E& = 3.0 t 0.2 PM) (Fig. 1). The GABAA antagonist bicuculline (20 PM) inhibited relaxation, shifting the concentration-response curve to the right (inhibitory constant of bicuculline = 0.5 PM) implying that the response was mediated by GABAA receptors (Fig. 1). The GABAB receptor antagonist phaclofen 0.2 Control
7
6
GABA
5
(-lag
4
3
M)
Fig. 1. Concentration-response curves for relaxant effect of GABA on circular muscle strips from rat colon in absence and presence of GABAA receptor antagonist bicuculline (20 FM). *Significant difference from responses with GABA alone. Values are means k SE of 14 experiments with GABA alone and 4 experiments with bicuculline.
Downloaded from www.physiology.org/journal/ajpgi by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on August 30, 2018. Copyright © 1992 American Physiological Society. All rights reserved.
G692
ENTERIC
GABA
mM had no effect. The relaxant effect of GABA was neurally mediated and was abolished by tetrodotoxin (1 PM) at low concentrations and inhibited by 74 t 2% (P < 0.001) at the highest concentration of GABA (Fig. 2). Previous studies (15, 17) had shown that relaxation in rat colon was mediated by release of VIP from myenteric motor neurons, raising the possibility that GABA-induced relaxation was also mediated by release of VIP. This notion was tested with the VIP receptor antagonist VIP-(lO28). Addition of VIP-( 10-28) (10 ,uM) inhibited relaxation induced by all concentrations of GABA: the response to low concentrations was abolished, and the maximal response was inhibited by 71 t 3% (P < 0.001) (Fig. 3). The involvement of VIP was confirmed by measurement of VIP release in response to GABA. Basal VIP release in colonic strips was 16.3 t 3.4 pg. min-’ .mg tissue wet wt-? Addition of GABA (1 mM) increased
100
Control 80
0 I I
I
I
I
1
7
6
5
4
3
GABA
Hog
M)
Fig. 2. Concentration-response curves for relaxant effect of GABA on circular muscle strips from rat colon in absence and presence of 1 PM tetrodotoxin (TTX; open circles, n = 3). *Significant differences from responses with GABA alone. Values are means t SE of 14 experiments with GABA alone and 3 experiments with TTX.
IN PERISTALSIS
VIP release by 13.1 t 1.8 pg. min-’ .mg tissue wet wt-’ or 112 t 28% above basal levels (P c 0.01; n = 23). GABA-induced VIP release was inhibited significantly by bicuculline (4.2 t 1.6 pg. min-’ mg tissue wet wt-l above basal level; P c 0.01 from control VIP release) and tetrodotoxin (3.8 t 1.6 pg min-’ mg tissue wet wt-’ above basal level; P < 0.01 from control VIP release) but not by phaclofen (16.6 t 6.9 pgemin-’ .mg tissue wet wt-’ above basal level; NS from control VIP release). Basal release of substance P (1.32 t 0.20 pgmmin-’ mg tissue wet wt-‘> was not affected by either 1 FM GABA (1.11 k 0.23 pg. min-’ mg tissue wet wt-‘) or 10 PM bicuculline (1.06 t 0.18 pg. min-’ mg tissue wet wt-‘). When atropine, which was normally present with guanethidine, was omitted from the bathing medium, the relaxant response to all but the maximal concentration of GABA decreased significantly (Fig. 4). The decrease in relaxation implied that GABA also caused release of acetylcholine, which acted to attenuate the relaxant response. Because phaclofen had no effect on the relaxant response, the release of acetylcholine, like that of VIP, was mediated by GABAA receptors. In some studies, the effect of GABA alone and in combination with bicuculline, VIP-( lo-28), or tetrodotoxin was examined in circular muscle strips obtained from guinea pig colon. The results were identical to those obtained in muscle strips from rat colon. GABA caused a concentration-dependent relaxation (EC& = 1 PM) that was abolished by tetrodotoxin (1 PM), and inhibited by bicuculline (20 PM) and by VIP-( 10-28) (10 PM). Effect of GABA on peristaltic reflex. As previously shown ( 15), radial stretch of the caudad end of an isolated colonic segment caused ascending con traction, whereas radial stretch of the orad end caused descending relaxation. Both ascending contraction and descending relaxation increased in proportion to the degree of stretch (Fig. 5). Addition of bicuculline (20 PM) to the bathing medium inhibited both ascending contraction and descending l
l
l
l
l
l
100
+ Atropine ffl
80
80
I
I
I
I
1
7
6
5
4
3
GABA
(-log
Ml
Fig. 3. Concentration-response curves for relaxant effect of GABA on circular muscle strips from rat colon in absence and presence of VIP receptor antagonist VIP-( 10-28) (10 PM). *Significant difference from responses with GABA alone. Values are means t SE of 14 experiments with GABA alone and 4 with VIP-(10-28).
0 1q, 7
6
GABA
5
(-log
,
,
4
3
M)
Fig. 4. Concentration-response curves for relaxant effect of GABA on circular muscle strips from rat colon in presence and absence of 1 PM atropine. *Significant difference between responses. Values are means t SE of 5 experiments.
Downloaded from www.physiology.org/journal/ajpgi by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on August 30, 2018. Copyright © 1992 American Physiological Society. All rights reserved.
ENTERIC DESCENDING
RELAXATION
ASCENDING
GABA
CONTRACTION
*
140
ids
iiintroi
*
20
-
*
* o-
* I 2
oI 4
ORAD
I 6
STRETCH
1 10
1 8
(g)
* I 2
I 4
CAUDAD
I 6
STRETCH
I 8
1 10
(g)
Fig. 5. Effect of 1 mM GABA and 20 PM bicuculline on descending relaxation and ascending contraction components of peristaltic reflex in rat colon. *Significant difference from control responses. Values are means t SE of 4 experiments with GABA and 8 experiments with bicuculline.
relaxation (Fig. 5). Ascending contraction was inhibited by 67 t 15% (P < 0.01) at Z-g stretch and by 30 t 2% (P < 0.001) at 10-g stretch, and descending relaxation was inhibited by 68 t 2% (P < 0.001) at 2-g stretch and by 38 t 8% (P < 0.01) at 10-g stretch. The effect of GABA was opposite to that of bicuculline augmenting both ascending contraction and descending relaxation (Fig. 5). Ascending contraction was augmented by 44 & 10% (P < 0.05) at 2-g stretch and by 18 t 13% (NS) at 10-g stretch, and descending relaxation was augmented by 77 t 14% (P < 0.02) at 2-g stretch and 33 t 13% (P < 0.05) at 10-g stretch. The effect of bicuculline implied that intrinsic GABA neurons were involved in the regulation of the ascending contraction and descending relaxation components of the peristaltic reflex and that the effect of neuronal GABA was mediated by GABAA receptors. The activation of GABA neurons was examined during descending relaxation by measurement of GABA release from coionic segments preloaded with [3H] GABA. Basal release of GABA during a 5-min basal period was 0.033 t 0.004% of total GABA content. Application of a 4-g stretch for a period of 5 min increased GABA release significantly to 0.043 t 0.005% (i.e., 26 t 5% above basal level; P c 0.01). In previous studies (11,15), VIP was shown to increase significantly during descending relaxation only. VIP release during a 15-min control basal period was 0.85 t 0.08 pg/mg wet wt increasing to 1.32 t 0.14 pg/mg (P c 0.02) during a 15min period of descending relaxation; during this period, stretch in the range of 2-10 g was applied for 15 s at 2-min intervals. GABA (1 mM) augmented VIP release during descending relaxation by 82 t 31% (P < 0.05), whereas bicuculline (20 PM) inhibited VIP release by 54 t 15% (P c 0.02). DISCUSSION
The present study shows that GABA had no contractile, relaxant, or modulatory effect on isolated colonic muscle cells confirming previous studies in which GABA was shown to have no effect in nerve-free muscle preparations (25). The relaxant effect in innervated circular
IN PERISTALSIS
G693
muscle strips of rat and guinea pig colon was neurally mediated reflecting activation of VIP and cholinergic motor neurons via bicuculline-sensitive GABAA receptors. The relaxant effect mediated by release of VIP from VIP neurons masked the contractile effect mediated by release of acetylcholine from cholinergic neurons. Consistent with this notion, relaxation in rat colonic muscle strips was 1) accompanied by VIP release; 2) inhibited by tetrodotoxin and by the selective VIP antagonist, VIP-(10-28); and 3) augmented in the presence of atropine. Previous studies on intestinal, chiefly colonic, muscle from several species [rat (26, 29), guinea pig (6,9, 25, 27), and dog (5)] have shown that the predominant effect of GABA was a bicucullineand tetrodotoxin-sensitive relaxation. The nature of the neurotransmitter responsible for relaxation was not identified in these studies. In light of previous studies showing that VIP was responsible for neurally mediated relaxation (1, 2, 4, 10, 11, 13-19, 31, 38), it was postulated that relaxation induced by GABA was mediated by VIP. In various species, VIP is released by electrical field stimulation (1, 2, 13, 14, 16, 38) as well as by reflex stimulation (11, 15, 35) of circular muscle. The release of VIP is accompanied by and coupled to relaxation, which can be inhibited by specific VIP antisera or selective VIP antagonists. The present study shows that VIP motor neurons can be activated by putative neurotransmitters, such as GABA, present in neurons of the myenteric plexus. As noted in several autoradiographic (20, 28, 37) and immunohistochemical studies (7, 21, 31, 33, 34), GABA is found mainly in a subpopulation of Dogiel type I neurons in the myenteric plexus; the neurons project to other neurons in the plexus and to the underlying circular muscle. In the present study, these neurons were activated when the peristaltic reflex was induced by radial stretch. This was evident by 1) an increase in GABA release from colonic segments during peristaltic activity and 2) by the inhibitory effect of bicuculline on the ascending contraction and descending relaxation components of the peristaltic reflex. Thus bicuculline inhibited and exogenous GABA augmented descending relaxation and VIP release. The effect of bicuculline implied that the release of GABA from GABA neurons was coupled via GABA, receptors to release of VIP from VIP motor neurons, which as shown previously (11, 15), are responsible for descending relaxation. The inhibitory effect of bicuculline on ascending contraction implied that the release of GABA was coupled also via GABAA receptors to release of acetylcholine from cholinergic motor neurons, which as shown previously (11, 12, 15), are largely responsible for ascending contraction. It should be noted that the inhibitory effects of bicuculline and the stimulatory effects of exogenous GABA on the two components of the peristaltic reflex parallel their pharmacological effects in isolated muscle strips. Previous studies had shown that GABA accelerates and bicuculline slows the rate of propulsion of fecal pellets in isolated segments of guinea pig colon reflecting the participation of GABA neurons in the regulation of peristaltic activity (22, 27, 32). The present study, using a preparation that distinguishes the two components of the peristaltic reflex, defines the site of action of GABA
Downloaded from www.physiology.org/journal/ajpgi by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on August 30, 2018. Copyright © 1992 American Physiological Society. All rights reserved.
G694
ENTERIC
GABA
and the nature of the transmitters released separately by the activity of GABA neurons during each phase of the reflex. This work was supported by National Institute of Diabetes and Digestive and Kidney Diseases Grant DK-34153. Address correspondence to: J. R. Grider, Box 711, MCV Station Medical College of Virginia, Richmond, VA 23298-0711. Received
11 July 1991; accepted in final form 31 October
IN
PERISTALSIS
18. Grider, 19.
20.
1991.
REFERENCES 1. Biancani, P., J. H. Walsh, and J. Behar Vasoactive intestinal polypeptide. A neurotransmitter for lower esophageal sphincter relaxation. J. C/in. Invest. 73: 963-967, 1984. 2. Biancani, P., J. H. Walsh, and J. Behar. Vasoactive intestinal peptide: a transmitter for relaxation of the rabbit internal anal sphincter. Gustroenterology 89: 867-874, 1985. 3. Bitar, K. N., and G. M. Makhlouf. Receptors on smooth muscle cells: characterization by contraction and specific antagonists. Am. J. Physiol. 242 (Gustrointest. Liver Physiol. 5): G40O-6407, 1982. 4. Bitar, K. N., and G. M. Makhlouf. Relaxation of isolated gastric smooth muscle cells by vasoactive intestinal peptide. Science Wash. DC 216: 531-533, 1982. 5. Boeckxstaens, G. E., P. A. Pelckmans, M. Rampart, I. F. Ruytjens, T. J. Verbeuren, A. G. Herman, and Y. M. Van Maercke. GABAA receptor-mediated stimulation of non-adrenergic non-cholinergic neurones in the dog ileo-colonic junction. Br. J. Pharmacol. 101: 460-464, 1990. 6. Frigo, G. M., A. Galli, S. Lecchini, and M. Marcoli. A facilitatory effect of bicuculline on the enteric neurones in the guinea-pig isolated colon. Br. J. PhurmacoZ. 90: 31-41, 1987. J. B., D. C. Trussell, S. Pompolo, J. C. Bornstein, 7. Furness, B. E. Maley, and J. Storm-Mathisen. Shapes and projections of neurons with immunoreactivity for gamma-aminobutyric acid in the guinea pig small intestine. Cell Tissue Res. 256: 293-301, 1989. 8. Giotti, A., S. Luzzi, C. A. Maggi, S. Spagnesi, and L. Zilletti. Modulatory activity of GABAB receptors on cholinergic tone in guinea pig distal colon. Br. J. Pharmucol. 84: 883-895, 1985. 9. Giotti, A., S. Luzzi, S. Spagnesi and L. Zilletti. GABAA and GABAB receptor-mediated effects in guinea-pig ileum. Br. J. Phurmucol. 78: 469-478, 1983. 10. Goyal, R. K., S. Rattan, and S. I. Said. VIP as a possible neurotransmitter of non-cholinergic, non-adrenergic inhibitory neurons. Nature Lond. 288: 378-380,198O. 11. Grider, J. R. Identification of neurotransmitters regulating intestinal peristaltic reflex in humans. Gustroenterology 97: 1414-1419, 1989. 12. Grider, J. R. Tachykinins as transmitters of ascending contractile component of the peristaltic reflex. Am. J. Physiol. 257 (Gustrointest. Liver Physiol. 20): G709-G714, 1989. 13. Grider, J. R., M. B. Cable, K. N. Bitar, S. I. Said, and G. M. Makhlouf. Vasoactive intestinal peptide. Relaxant neurotransmitter in tenai coli of the guinea pig. Gustroenterology 89: 3642, 1985. 14. Grider, J. R., M. B. Cable, S. I. Said, and G. M. Makhlouf. Vasoactive intestinal peptide as a neural mediator of gastric relaxation. Am, J. Physiol. 248 (Gustrointest. Liver Physiol. 11): G73G78,1985. J. R., and G. M. Makhlouf. Colonic peristaltic reflex: 15. Grider, identification of vasoactive intestinal peptide as mediator of descending relaxation. Am. J. Physiol. 251 (Gustrointest. Liver Physiol. 14): G4O-G45, 1986. 16. Grider, J. R., and G. M. Makhlouf. Prejunctional inhibition of vasoactive intestinal peptide release. Am. J. Physiol. 253 (Gustrointest. Liver Physiol. 16): G7-G12, 1987. J. R., and G. M. Makhlouf. Suppression of inhibitory 17. Grider, neural input to colonic circular muscle by opioid peptides. J. Phurmucol. Exp. Ther. 243: 205-210,1987.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
J. R., and G. M. Makhlouf. Contraction mediated by Ca++ release in circular and Ca++ influx in longitudinal muscle cells. J. Phurmucol. Exp. Ther. 244: 432-437, 1988. Grider, J. R., and J. R. Rivier. Vasoactive intestinal peptide (VIP) as transmitter of inhibitory motor neurons of the gut: evidence from the use of selective VIP antagonists and VIP antiserum. J. Phurmucol. Exp. Ther. 253: 738-742, 1990. Jessen, K. R., J. M. Hills, M. E. Dennison, and R. Mirsky. Gamma-aminobutyrate as an autonomic neurotransmitter: release and uptake of [3H]gamma-aminobutyrate in guinea pig large intestine and cultured entericneurons using physiological methods and electron microscopic autoradiography. Neuroscience 10: 1427-1442, 1983. Jessen, K. R., J. M. Hills, and M. J. Saffrey. Immunohistochemical demonstration of GABAergic neurons in the enteric nervous system. J Neurosci. 6: 1628-1634, 1986. Jessen, K. R., R. Mirsky, and J. M. Hills. GABA as an autonomic neurotransmitter: studies on intrinsic GABAergic neurons in the myenteric plexus of the gut. Trends Neurosci. 10: 255262,1987. Kerr, D. I. B., and A. Krantis. Uptake of stimulus-evoked release of [3H]-gamma-aminobutyric acid by myenteric nerves of guinea pig intestine. Br. J. Phurmucol. 78: 2711-276, 1983. Kleinrok, A., and H. Kilbinger. Gamma-aminobutyric acid and cholinergic transmission in the guinea pig ileum. Nuunyn-Schmiedeberg’s. Arch. Phurmucol. 322: 216-220, 1983. Krantis, A., M. Costa, J. B. Furness, and J. Orbach. Gammaaminobutyric acid stimulates intrinsic inhibitory and excitatory nerves in the guinea pig intestine. Eur. J. PhurmucoZ. 67: 461-468, 1980. Krantis, A., and R. K. Harding. GABA-related actions in isolated in vitro preparations of the rat small intestine. Eur. J. PhurmucoZ. 141: 291-298, 1987. Krantis, A., and D. I. B. Kerr. The effect of GABA antagonism on propulsive activity of the guinea pig large intestine. Eur. J. Phurmucol. 76: 111-114, 1981. Krantis, A., and T. Webb. Autoradiographic localization of [“HIgamma-aminobutyric acid in neuronal elements of the rat gastric antrum and intestine. J. Auton. Nerv. Syst. 29: 41-48,1989. Maggi, C. A., S. Manzini, and A. Meli. Evidence that GABAA receptors mediate relaxation of rat duodenum by activating intramural non-adrenergic non-cholinergic neurons. J. Auton. Phurmucol. 4: 77-85, 1984. Messenger, J. P., and J. B. Furness. Projections of chemicallyspecificied neurons in the guinea pig colon. Arch. HistoZ. Cytol. 53: 467-495,199O. Nurko, S., and S. Rattan. Role of vasoactive intestinal polypeptide in the internal anal sphincter relaxation of the opossum. J. Clin. Inuest. 81: 1146-1153, 1988. Ong, J., and D. I. B. Kerr. GABA*and GABAB-receptormediated modification of intestinal motility. Eur. J. Phurmucol. 86: 9-17, 1983. Pompolo, S., and J. B. Furness. Ultrastructure and synaptology of neurons immunoreactive for gamma-aminobutyric acid in the myenteric plexus of the guinea pig small intestine. J. Neurocytol. 19: 539-549,199o. Saito, N., and C. Tanaka. Immunohistochemical demonstration of GABA containing neurons in the guinea pig ileum using purified GABA antiserum. Bruin Res. 376: 78-84, 1986. Sjoqvist, A., and J. Fahrenkrug. Release of vasoactive intestinal polypeptide anally of a local distension of the feline small intestine. Actu Physiol. Scund. 130: 433-438, 1987. Taniyama, K., M. Kusunoki, N. Saito, and C. Tanaka. Release of gamma-aminobutyric acid from cat colon. Science Wash. DC 217: 1038-1040,1982. Taniyama, K., N. Saito, Y. Miki, and C. Tanaka. Enteric gamma-aminobutyric acid-containing neurons and the relevance to motility of the cat colon. Gustroenterology 93: 519-525, 1987. Wiley, J. W., T. M. O’Dorisio, and C. Owyang. Vasoactive intestinal polypeptide mediated CCK-induced relaxation of Sphincter of Oddi, J. Clin. Inuest, 81: 1920-1924, 1988.
Downloaded from www.physiology.org/journal/ajpgi by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on August 30, 2018. Copyright © 1992 American Physiological Society. All rights reserved.
plexus;
AUTORADIOGRAPHIC and immunohistochemical studies have demonstrated the presence of y-aminobutyric acid (GABA) predominantly in neurons of the myenteric plexus (7, 20, 21, 28, 33, 34, 37). These neurons have shapes characteristic of Dogiel type I neurons with several short dendritic processes and one long branching axon that projects within the plexus as well as to the underlying circular muscle layer (7, 31-33). Electrophysiologically, the neurons are S-type excitatory neurons that release GABA upon field stimulation (20,22, 23, 36, 37) The effects of GABA in the gut are mediated by bicuculline-sensitive GABAA and phaclofen-sensitive GABAR receptors. GABAA receptors are excitatory and are located on both cholinergic and noncholinergic neurons (5, 6, 9, 24-26, 29, 32, 37); GABAB receptors are inhibitory and appear to be located on cholinergic neurons only (8, 9, 24, 26, 32). Together these receptors account for the effect of GABA on intestinal muscle strips. GABA causes relaxation in these strips preceded by a transient atropine-sensitive contraction; these effects are mediated by bicuculline-sensitive GABAA receptors. In strips stimulated electrically to produce cholinergically mediated “twitch” contraction, GABA acting via GABAB receptors inhibits contraction. GABA has no G690
0193-1857/92
$2.00 Copyright
College of Virginia,
Richmond,
Virginia
23298-0711
contractile or relaxant effect in nerve-free preparations (25). One or more of these effects is probably responsible for the inhibitory effect of GABA antagonists on intestinal propulsion (22, 27, 32). The aim of the present study was to determine the nature of the transmitter responsible for GABA-mediated relaxation and to examine the participation of GABA in the regulation of the peristaltic reflex. The results show that 1) GABA receptors are not present on smooth muscle cells; 2) GABA-induced relaxation is mediated by release of vasoactive intestinal peptide (VIP) from myenteric motor neurons; and 3) GABA neurons are involved in the regulation of the ascending and descending components of the peristaltic reflex. METHODS Preparation of isolated smooth muscle cells. Smooth muscle cells were isolated from the circular muscle layer of the colon by enzymatic digestion as described previously for gastric and intestinal cells (3, 4, 18). The circular and longitudinal layers were separated by blunt dissection, and the mucosa was removed by scraping with forceps. The segment was cut into lcm long strips and incubated for two successive 45-min periods in 15 ml of medium containing 0.1% collagenase (type II) and 0.01% soybean trypsin inhibitor. The composition of the medium was (in mM) 120 NaCl, 4 KCl, 2.6 KHgP04, 2 CaC&, 0.6 25 N-2-hydroxyethylpiperazine-N’-2-ethanesulfonic M&l,, acid (HEPES), 14 glucose, 2.1% Eagle’s essential amino acid mixture, and 0.1% bovine serum albumin. After the second incubation, the partially digested muscle strips were washed with enzyme-free medium and reincubated in 15 ml of enzymefree medium for an additional 30 min to allow the cells to disperse spontaneously. The cells were harvested by filtration through 500-pm Nitex mesh. Measurement of response of isolated colonic muscle cells. The contractile response of isolated muscle cells was determined as described previously (3, 4, 18). An aliquot of cell suspension (0.5 ml) containing -IO* cells was mixed with the test agonist, and the reaction was terminated at 30 s by addition of 1% acrolein. The relaxant response was determined in cells maximally contracted with cholecystokinin octapeptide (CCK-8; 1 nM) as described previously (4). Putative relaxant agents were added alone or in combination 60 s before CCK-8, and the reaction was terminated after an additional 30 s. The lengths of the first 50 randomly encountered cells were measured by scanning micrometry. Contraction was expressed as percent decrease in cell length from control, and relaxation was expressed as percent inhibition of CCK-induced contraction. Control length was the mean cell length obtained in the absence of test agonist. Preparation of colonic muscZe strips. Circular muscle strips (3 mm wide) were obtained from the midcolon and suspended in 3-ml organ chambers containing Krebs-bicarbonate medium maintained at 37°C and gassed with 95% 02-5% CO,. The composition of the medium was (in mM) 118 NaCl, 4.8 KCl, 1.2 KH2P04, 1.2 MgSO,, 2.5 CaCl,, 25 NaHCOs, 11 glucose, and 0.1 % bovine serum albumin. Atropine (0.5 PM) and
0 1992 the American
Physiological
Society
Downloaded from www.physiology.org/journal/ajpgi by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on August 30, 2018. Copyright © 1992 American Physiological Society. All rights reserved.
ENTERIC
GABA
guanethidine (20 PM) were added to the medium. Strips were equilibrated for 60 min at an initial tension of 0.5 g before measurement of isometric response. Measurement of response in colonic muscle strips. The response to GABA was measured by noncumulative addition of GABA in the range of 0.1 PM to 1 mM. A single concentration of GABA was added, and the response was recorded until the maximal relaxation for that concentration was achieved. This usually occured between 30 s and 1 min. The tissue was then washed at 5-min intervals over the next 20 min, during which the tone recovered to basal levels. The next concentration of GABA was then added and the procedure repeated. Any colonic strip failing to recover basal tone was discarded. The concentration-response curve was repeated in other strips in the presence of one of the following: the sodium conductance blocker, tetrodotoxin (1 PM); the VIP receptor antagonist, VIP(10-28) (10 PM); the GABA* receptor antagonist, bicuculline (20 PM); or the GABAB receptor antagonist, phaclofen (0.2 mM). In some strips, atropine was omitted from the medium to determine its effect on the response to GABA. Concentration-response curves were analyzed and the mean effective concentration (E&) calculated from computer-fitted curves using the P.Fit program (Biosoft, Elsevier, Cambridge, UK). In some strips, samples were obtained for measurement of VIP and substance P. The strips were first incubated for 10 min in Krebs-bicarbonate medium containing 1,000 U/ml aprotinin and 20 PM bacitracin to determine basal release of VIP and substance P. Other samples were obtained after addition of 1 mM GABA alone and in combination with 1 PM tetrodotoxin or 20 PM bicuculline. The samples were stored at -20°C for subsequent radioimmunoassay. Measurement of peristaltic reflex in colonic segments. Colonic segments were prepared as described in detail previously (15). Briefly, a 4- to 6-cm segment from rat midcolon was secured horizontally in the bottom of a lo-ml organ bath. The segment was incubated in Krebs-bicarbonate medium, maintained at 37°C and gassed with 95% O,-5% COZ. The ascending contraction component of the peristaltic reflex was elicited by radial stretch of the caudad end of the segment and the descending relaxation component by radial stretch of the orad end. Stretch was applied in the range of 2-10 g for 15-s periods at intervals of 2 min using a hook-and-pulley assembly, and contraction or relaxation of circular muscle was measured with a force-displacement transducer. Measurements were made in the presence of 1 mM GABA, 20 PM bicuculline, or 0.2 mM phaclofen. In some segments, samples were obtained for measurement of VIP after 1,000 U/ml aprotinin, and 20 PM bacitracin were added to the medium. Samples were first obtained for 15 min to determine basal VIP release. Samples were also obtained during descending relaxation in the absence or presence of 1 mM GABA, 20 PM bicuculline, or 0.2 mM phaclofen. The samples were stored at -20°C for subsequent radioimmunoassay. Radioimmunoassay of VIP. VIP was measured by radioimmunoassay as described in detail previously (13-15) using VIP antibody AC115 (final dilution 1:50,000). The limit of detection of the assay was 5 pg/ml of original sample and the mean inhibitory concentration (I&) was 46 t 5 pg/ml. The intra-assay and interassay variability were 5 and IO%, respectively. Radioimmunoassay of substance P. Substance P was measured by radioimmunoassay as described in detail previously (12) using substance P antiserum AC95 (final dilution of 1:2,500). The limit of detection was 10 pg/ml of original sample and the IC,, was 138 t 29 pg/ml. Measurements of PHJGABA release. The release of GABA was measured in colonic segments preloaded with [3H]GABA. The segments were incubated for 60 min in Krebs-bicarbonate
G691
IN PERISTALSIS
medium containing 50 nM [3H]GABA (sp act 34.8 Ci/mmol). Aminooxyacetic acid (10 PM) was added to prevent degradation of GABA, and P-alanine (1 mM) was added to prevent uptake of [“HIGABA into glial cells. The segments were washed five times in Krebs medium and then used for measurement of descending relaxation. The bathing medium was removed and replaced every 5 min and saved for liquid scintillation spectrometry. Samples were collected both before and during a period of 4g stretch. At the end of the experiment, the tissue was dissolved in Soluene and subjected to scintillation spectrometry. GABA release was expressed as a percentage of total [“HIGABA content. Materials. GABA and phaclofen were obtained from Research Biochemicals (Natick, MA); aminooxyacetic acid, ,8alanine, atropine sulfate, bicuculline, bacitracin, tetrodotoxin were from Sigma Chemical (St. Louis, MO). Collagenase type II was from Worthington Biochemicals (Freehold, NJ), and VIP-(10-28) was from Bachem (Torrance, CA). Aprotinin was from Mobay Chemical (New York, NY); 1251-labeled VIP and [“HIGABA were from New England Nuclear (Boston, MA). VIP antiserum AC115 was obtained from Cambridge Research Biochemicals (Wilmington, DE), and substance P antiserum AC95 and P251-labeled substance P were obtained from Peninsula (Belmont, CA). RESULTS
Effect of GABA on isolated colonic muscle cells. There was no detectable effect of GABA on isolated colonic circular muscle cells. GABA (0.1-100 PM) did not elicit contraction or relaxation of muscle cells maximally precontracted with CCK-8. GABA also did not augment or inhibit relaxation induced by VIP (1 nM). Effect of GABA on colonic muscle strips.. GABA caused a concentrationdependent relaxation of colonic circular muscle strips (E& = 3.0 t 0.2 PM) (Fig. 1). The GABAA antagonist bicuculline (20 PM) inhibited relaxation, shifting the concentration-response curve to the right (inhibitory constant of bicuculline = 0.5 PM) implying that the response was mediated by GABAA receptors (Fig. 1). The GABAB receptor antagonist phaclofen 0.2 Control
7
6
GABA
5
(-lag
4
3
M)
Fig. 1. Concentration-response curves for relaxant effect of GABA on circular muscle strips from rat colon in absence and presence of GABAA receptor antagonist bicuculline (20 FM). *Significant difference from responses with GABA alone. Values are means k SE of 14 experiments with GABA alone and 4 experiments with bicuculline.
Downloaded from www.physiology.org/journal/ajpgi by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on August 30, 2018. Copyright © 1992 American Physiological Society. All rights reserved.
G692
ENTERIC
GABA
mM had no effect. The relaxant effect of GABA was neurally mediated and was abolished by tetrodotoxin (1 PM) at low concentrations and inhibited by 74 t 2% (P < 0.001) at the highest concentration of GABA (Fig. 2). Previous studies (15, 17) had shown that relaxation in rat colon was mediated by release of VIP from myenteric motor neurons, raising the possibility that GABA-induced relaxation was also mediated by release of VIP. This notion was tested with the VIP receptor antagonist VIP-(lO28). Addition of VIP-( 10-28) (10 ,uM) inhibited relaxation induced by all concentrations of GABA: the response to low concentrations was abolished, and the maximal response was inhibited by 71 t 3% (P < 0.001) (Fig. 3). The involvement of VIP was confirmed by measurement of VIP release in response to GABA. Basal VIP release in colonic strips was 16.3 t 3.4 pg. min-’ .mg tissue wet wt-? Addition of GABA (1 mM) increased
100
Control 80
0 I I
I
I
I
1
7
6
5
4
3
GABA
Hog
M)
Fig. 2. Concentration-response curves for relaxant effect of GABA on circular muscle strips from rat colon in absence and presence of 1 PM tetrodotoxin (TTX; open circles, n = 3). *Significant differences from responses with GABA alone. Values are means t SE of 14 experiments with GABA alone and 3 experiments with TTX.
IN PERISTALSIS
VIP release by 13.1 t 1.8 pg. min-’ .mg tissue wet wt-’ or 112 t 28% above basal levels (P c 0.01; n = 23). GABA-induced VIP release was inhibited significantly by bicuculline (4.2 t 1.6 pg. min-’ mg tissue wet wt-l above basal level; P c 0.01 from control VIP release) and tetrodotoxin (3.8 t 1.6 pg min-’ mg tissue wet wt-’ above basal level; P < 0.01 from control VIP release) but not by phaclofen (16.6 t 6.9 pgemin-’ .mg tissue wet wt-’ above basal level; NS from control VIP release). Basal release of substance P (1.32 t 0.20 pgmmin-’ mg tissue wet wt-‘> was not affected by either 1 FM GABA (1.11 k 0.23 pg. min-’ mg tissue wet wt-‘) or 10 PM bicuculline (1.06 t 0.18 pg. min-’ mg tissue wet wt-‘). When atropine, which was normally present with guanethidine, was omitted from the bathing medium, the relaxant response to all but the maximal concentration of GABA decreased significantly (Fig. 4). The decrease in relaxation implied that GABA also caused release of acetylcholine, which acted to attenuate the relaxant response. Because phaclofen had no effect on the relaxant response, the release of acetylcholine, like that of VIP, was mediated by GABAA receptors. In some studies, the effect of GABA alone and in combination with bicuculline, VIP-( lo-28), or tetrodotoxin was examined in circular muscle strips obtained from guinea pig colon. The results were identical to those obtained in muscle strips from rat colon. GABA caused a concentration-dependent relaxation (EC& = 1 PM) that was abolished by tetrodotoxin (1 PM), and inhibited by bicuculline (20 PM) and by VIP-( 10-28) (10 PM). Effect of GABA on peristaltic reflex. As previously shown ( 15), radial stretch of the caudad end of an isolated colonic segment caused ascending con traction, whereas radial stretch of the orad end caused descending relaxation. Both ascending contraction and descending relaxation increased in proportion to the degree of stretch (Fig. 5). Addition of bicuculline (20 PM) to the bathing medium inhibited both ascending contraction and descending l
l
l
l
l
l
100
+ Atropine ffl
80
80
I
I
I
I
1
7
6
5
4
3
GABA
(-log
Ml
Fig. 3. Concentration-response curves for relaxant effect of GABA on circular muscle strips from rat colon in absence and presence of VIP receptor antagonist VIP-( 10-28) (10 PM). *Significant difference from responses with GABA alone. Values are means t SE of 14 experiments with GABA alone and 4 with VIP-(10-28).
0 1q, 7
6
GABA
5
(-log
,
,
4
3
M)
Fig. 4. Concentration-response curves for relaxant effect of GABA on circular muscle strips from rat colon in presence and absence of 1 PM atropine. *Significant difference between responses. Values are means t SE of 5 experiments.
Downloaded from www.physiology.org/journal/ajpgi by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on August 30, 2018. Copyright © 1992 American Physiological Society. All rights reserved.
ENTERIC DESCENDING
RELAXATION
ASCENDING
GABA
CONTRACTION
*
140
ids
iiintroi
*
20
-
*
* o-
* I 2
oI 4
ORAD
I 6
STRETCH
1 10
1 8
(g)
* I 2
I 4
CAUDAD
I 6
STRETCH
I 8
1 10
(g)
Fig. 5. Effect of 1 mM GABA and 20 PM bicuculline on descending relaxation and ascending contraction components of peristaltic reflex in rat colon. *Significant difference from control responses. Values are means t SE of 4 experiments with GABA and 8 experiments with bicuculline.
relaxation (Fig. 5). Ascending contraction was inhibited by 67 t 15% (P < 0.01) at Z-g stretch and by 30 t 2% (P < 0.001) at 10-g stretch, and descending relaxation was inhibited by 68 t 2% (P < 0.001) at 2-g stretch and by 38 t 8% (P < 0.01) at 10-g stretch. The effect of GABA was opposite to that of bicuculline augmenting both ascending contraction and descending relaxation (Fig. 5). Ascending contraction was augmented by 44 & 10% (P < 0.05) at 2-g stretch and by 18 t 13% (NS) at 10-g stretch, and descending relaxation was augmented by 77 t 14% (P < 0.02) at 2-g stretch and 33 t 13% (P < 0.05) at 10-g stretch. The effect of bicuculline implied that intrinsic GABA neurons were involved in the regulation of the ascending contraction and descending relaxation components of the peristaltic reflex and that the effect of neuronal GABA was mediated by GABAA receptors. The activation of GABA neurons was examined during descending relaxation by measurement of GABA release from coionic segments preloaded with [3H] GABA. Basal release of GABA during a 5-min basal period was 0.033 t 0.004% of total GABA content. Application of a 4-g stretch for a period of 5 min increased GABA release significantly to 0.043 t 0.005% (i.e., 26 t 5% above basal level; P c 0.01). In previous studies (11,15), VIP was shown to increase significantly during descending relaxation only. VIP release during a 15-min control basal period was 0.85 t 0.08 pg/mg wet wt increasing to 1.32 t 0.14 pg/mg (P c 0.02) during a 15min period of descending relaxation; during this period, stretch in the range of 2-10 g was applied for 15 s at 2-min intervals. GABA (1 mM) augmented VIP release during descending relaxation by 82 t 31% (P < 0.05), whereas bicuculline (20 PM) inhibited VIP release by 54 t 15% (P c 0.02). DISCUSSION
The present study shows that GABA had no contractile, relaxant, or modulatory effect on isolated colonic muscle cells confirming previous studies in which GABA was shown to have no effect in nerve-free muscle preparations (25). The relaxant effect in innervated circular
IN PERISTALSIS
G693
muscle strips of rat and guinea pig colon was neurally mediated reflecting activation of VIP and cholinergic motor neurons via bicuculline-sensitive GABAA receptors. The relaxant effect mediated by release of VIP from VIP neurons masked the contractile effect mediated by release of acetylcholine from cholinergic neurons. Consistent with this notion, relaxation in rat colonic muscle strips was 1) accompanied by VIP release; 2) inhibited by tetrodotoxin and by the selective VIP antagonist, VIP-(10-28); and 3) augmented in the presence of atropine. Previous studies on intestinal, chiefly colonic, muscle from several species [rat (26, 29), guinea pig (6,9, 25, 27), and dog (5)] have shown that the predominant effect of GABA was a bicucullineand tetrodotoxin-sensitive relaxation. The nature of the neurotransmitter responsible for relaxation was not identified in these studies. In light of previous studies showing that VIP was responsible for neurally mediated relaxation (1, 2, 4, 10, 11, 13-19, 31, 38), it was postulated that relaxation induced by GABA was mediated by VIP. In various species, VIP is released by electrical field stimulation (1, 2, 13, 14, 16, 38) as well as by reflex stimulation (11, 15, 35) of circular muscle. The release of VIP is accompanied by and coupled to relaxation, which can be inhibited by specific VIP antisera or selective VIP antagonists. The present study shows that VIP motor neurons can be activated by putative neurotransmitters, such as GABA, present in neurons of the myenteric plexus. As noted in several autoradiographic (20, 28, 37) and immunohistochemical studies (7, 21, 31, 33, 34), GABA is found mainly in a subpopulation of Dogiel type I neurons in the myenteric plexus; the neurons project to other neurons in the plexus and to the underlying circular muscle. In the present study, these neurons were activated when the peristaltic reflex was induced by radial stretch. This was evident by 1) an increase in GABA release from colonic segments during peristaltic activity and 2) by the inhibitory effect of bicuculline on the ascending contraction and descending relaxation components of the peristaltic reflex. Thus bicuculline inhibited and exogenous GABA augmented descending relaxation and VIP release. The effect of bicuculline implied that the release of GABA from GABA neurons was coupled via GABA, receptors to release of VIP from VIP motor neurons, which as shown previously (11, 15), are responsible for descending relaxation. The inhibitory effect of bicuculline on ascending contraction implied that the release of GABA was coupled also via GABAA receptors to release of acetylcholine from cholinergic motor neurons, which as shown previously (11, 12, 15), are largely responsible for ascending contraction. It should be noted that the inhibitory effects of bicuculline and the stimulatory effects of exogenous GABA on the two components of the peristaltic reflex parallel their pharmacological effects in isolated muscle strips. Previous studies had shown that GABA accelerates and bicuculline slows the rate of propulsion of fecal pellets in isolated segments of guinea pig colon reflecting the participation of GABA neurons in the regulation of peristaltic activity (22, 27, 32). The present study, using a preparation that distinguishes the two components of the peristaltic reflex, defines the site of action of GABA
Downloaded from www.physiology.org/journal/ajpgi by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on August 30, 2018. Copyright © 1992 American Physiological Society. All rights reserved.
G694
ENTERIC
GABA
and the nature of the transmitters released separately by the activity of GABA neurons during each phase of the reflex. This work was supported by National Institute of Diabetes and Digestive and Kidney Diseases Grant DK-34153. Address correspondence to: J. R. Grider, Box 711, MCV Station Medical College of Virginia, Richmond, VA 23298-0711. Received
11 July 1991; accepted in final form 31 October
IN
PERISTALSIS
18. Grider, 19.
20.
1991.
REFERENCES 1. Biancani, P., J. H. Walsh, and J. Behar Vasoactive intestinal polypeptide. A neurotransmitter for lower esophageal sphincter relaxation. J. C/in. Invest. 73: 963-967, 1984. 2. Biancani, P., J. H. Walsh, and J. Behar. Vasoactive intestinal peptide: a transmitter for relaxation of the rabbit internal anal sphincter. Gustroenterology 89: 867-874, 1985. 3. Bitar, K. N., and G. M. Makhlouf. Receptors on smooth muscle cells: characterization by contraction and specific antagonists. Am. J. Physiol. 242 (Gustrointest. Liver Physiol. 5): G40O-6407, 1982. 4. Bitar, K. N., and G. M. Makhlouf. Relaxation of isolated gastric smooth muscle cells by vasoactive intestinal peptide. Science Wash. DC 216: 531-533, 1982. 5. Boeckxstaens, G. E., P. A. Pelckmans, M. Rampart, I. F. Ruytjens, T. J. Verbeuren, A. G. Herman, and Y. M. Van Maercke. GABAA receptor-mediated stimulation of non-adrenergic non-cholinergic neurones in the dog ileo-colonic junction. Br. J. Pharmacol. 101: 460-464, 1990. 6. Frigo, G. M., A. Galli, S. Lecchini, and M. Marcoli. A facilitatory effect of bicuculline on the enteric neurones in the guinea-pig isolated colon. Br. J. PhurmacoZ. 90: 31-41, 1987. J. B., D. C. Trussell, S. Pompolo, J. C. Bornstein, 7. Furness, B. E. Maley, and J. Storm-Mathisen. Shapes and projections of neurons with immunoreactivity for gamma-aminobutyric acid in the guinea pig small intestine. Cell Tissue Res. 256: 293-301, 1989. 8. Giotti, A., S. Luzzi, C. A. Maggi, S. Spagnesi, and L. Zilletti. Modulatory activity of GABAB receptors on cholinergic tone in guinea pig distal colon. Br. J. Pharmucol. 84: 883-895, 1985. 9. Giotti, A., S. Luzzi, S. Spagnesi and L. Zilletti. GABAA and GABAB receptor-mediated effects in guinea-pig ileum. Br. J. Phurmucol. 78: 469-478, 1983. 10. Goyal, R. K., S. Rattan, and S. I. Said. VIP as a possible neurotransmitter of non-cholinergic, non-adrenergic inhibitory neurons. Nature Lond. 288: 378-380,198O. 11. Grider, J. R. Identification of neurotransmitters regulating intestinal peristaltic reflex in humans. Gustroenterology 97: 1414-1419, 1989. 12. Grider, J. R. Tachykinins as transmitters of ascending contractile component of the peristaltic reflex. Am. J. Physiol. 257 (Gustrointest. Liver Physiol. 20): G709-G714, 1989. 13. Grider, J. R., M. B. Cable, K. N. Bitar, S. I. Said, and G. M. Makhlouf. Vasoactive intestinal peptide. Relaxant neurotransmitter in tenai coli of the guinea pig. Gustroenterology 89: 3642, 1985. 14. Grider, J. R., M. B. Cable, S. I. Said, and G. M. Makhlouf. Vasoactive intestinal peptide as a neural mediator of gastric relaxation. Am, J. Physiol. 248 (Gustrointest. Liver Physiol. 11): G73G78,1985. J. R., and G. M. Makhlouf. Colonic peristaltic reflex: 15. Grider, identification of vasoactive intestinal peptide as mediator of descending relaxation. Am. J. Physiol. 251 (Gustrointest. Liver Physiol. 14): G4O-G45, 1986. 16. Grider, J. R., and G. M. Makhlouf. Prejunctional inhibition of vasoactive intestinal peptide release. Am. J. Physiol. 253 (Gustrointest. Liver Physiol. 16): G7-G12, 1987. J. R., and G. M. Makhlouf. Suppression of inhibitory 17. Grider, neural input to colonic circular muscle by opioid peptides. J. Phurmucol. Exp. Ther. 243: 205-210,1987.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
J. R., and G. M. Makhlouf. Contraction mediated by Ca++ release in circular and Ca++ influx in longitudinal muscle cells. J. Phurmucol. Exp. Ther. 244: 432-437, 1988. Grider, J. R., and J. R. Rivier. Vasoactive intestinal peptide (VIP) as transmitter of inhibitory motor neurons of the gut: evidence from the use of selective VIP antagonists and VIP antiserum. J. Phurmucol. Exp. Ther. 253: 738-742, 1990. Jessen, K. R., J. M. Hills, M. E. Dennison, and R. Mirsky. Gamma-aminobutyrate as an autonomic neurotransmitter: release and uptake of [3H]gamma-aminobutyrate in guinea pig large intestine and cultured entericneurons using physiological methods and electron microscopic autoradiography. Neuroscience 10: 1427-1442, 1983. Jessen, K. R., J. M. Hills, and M. J. Saffrey. Immunohistochemical demonstration of GABAergic neurons in the enteric nervous system. J Neurosci. 6: 1628-1634, 1986. Jessen, K. R., R. Mirsky, and J. M. Hills. GABA as an autonomic neurotransmitter: studies on intrinsic GABAergic neurons in the myenteric plexus of the gut. Trends Neurosci. 10: 255262,1987. Kerr, D. I. B., and A. Krantis. Uptake of stimulus-evoked release of [3H]-gamma-aminobutyric acid by myenteric nerves of guinea pig intestine. Br. J. Phurmucol. 78: 2711-276, 1983. Kleinrok, A., and H. Kilbinger. Gamma-aminobutyric acid and cholinergic transmission in the guinea pig ileum. Nuunyn-Schmiedeberg’s. Arch. Phurmucol. 322: 216-220, 1983. Krantis, A., M. Costa, J. B. Furness, and J. Orbach. Gammaaminobutyric acid stimulates intrinsic inhibitory and excitatory nerves in the guinea pig intestine. Eur. J. PhurmucoZ. 67: 461-468, 1980. Krantis, A., and R. K. Harding. GABA-related actions in isolated in vitro preparations of the rat small intestine. Eur. J. PhurmucoZ. 141: 291-298, 1987. Krantis, A., and D. I. B. Kerr. The effect of GABA antagonism on propulsive activity of the guinea pig large intestine. Eur. J. Phurmucol. 76: 111-114, 1981. Krantis, A., and T. Webb. Autoradiographic localization of [“HIgamma-aminobutyric acid in neuronal elements of the rat gastric antrum and intestine. J. Auton. Nerv. Syst. 29: 41-48,1989. Maggi, C. A., S. Manzini, and A. Meli. Evidence that GABAA receptors mediate relaxation of rat duodenum by activating intramural non-adrenergic non-cholinergic neurons. J. Auton. Phurmucol. 4: 77-85, 1984. Messenger, J. P., and J. B. Furness. Projections of chemicallyspecificied neurons in the guinea pig colon. Arch. HistoZ. Cytol. 53: 467-495,199O. Nurko, S., and S. Rattan. Role of vasoactive intestinal polypeptide in the internal anal sphincter relaxation of the opossum. J. Clin. Inuest. 81: 1146-1153, 1988. Ong, J., and D. I. B. Kerr. GABA*and GABAB-receptormediated modification of intestinal motility. Eur. J. Phurmucol. 86: 9-17, 1983. Pompolo, S., and J. B. Furness. Ultrastructure and synaptology of neurons immunoreactive for gamma-aminobutyric acid in the myenteric plexus of the guinea pig small intestine. J. Neurocytol. 19: 539-549,199o. Saito, N., and C. Tanaka. Immunohistochemical demonstration of GABA containing neurons in the guinea pig ileum using purified GABA antiserum. Bruin Res. 376: 78-84, 1986. Sjoqvist, A., and J. Fahrenkrug. Release of vasoactive intestinal polypeptide anally of a local distension of the feline small intestine. Actu Physiol. Scund. 130: 433-438, 1987. Taniyama, K., M. Kusunoki, N. Saito, and C. Tanaka. Release of gamma-aminobutyric acid from cat colon. Science Wash. DC 217: 1038-1040,1982. Taniyama, K., N. Saito, Y. Miki, and C. Tanaka. Enteric gamma-aminobutyric acid-containing neurons and the relevance to motility of the cat colon. Gustroenterology 93: 519-525, 1987. Wiley, J. W., T. M. O’Dorisio, and C. Owyang. Vasoactive intestinal polypeptide mediated CCK-induced relaxation of Sphincter of Oddi, J. Clin. Inuest, 81: 1920-1924, 1988.
Downloaded from www.physiology.org/journal/ajpgi by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on August 30, 2018. Copyright © 1992 American Physiological Society. All rights reserved.
