Pituitary Adenylate Cyclase-Activating Polypeptide Relaxes Rat Gastrointestinal Smooth Muscle Z. MUNGAN, A. ARIMURA, A. ERTAN, W. J. ROSSOWSKI & D. H. COY
Gastroenterology Section and Peptide Research Laboratories, Dept. of Medicine, Tulane University School of Medicine, New Orleans, and US-Japan Biomedical Research Laboratories, Tulane University Herbert Center, Belle Chasse, Louisiana, USA
Scand J Gastroenterol Downloaded from informahealthcare.com by UB Giessen on 10/30/14 For personal use only.
Mungan Z, Arimura A, Ertan A, Rossowski WJ, Coy DH. Pituitary adenylate cyclase-activating polypeptide relaxes rat gastrointestinal smooth muscle. Scand J Gastroenterol 1992;27:375-380.
Pituitary adenylate cyclase-activating polypeptide (PACAP) is a new member of the secretin/glucagon peptides family, being most homologous to vasoactive intestinal peptide (VIP). The present study was designed to investigate a possible effect of PACAP on the rat gastrointestinal smooth muscle in vitro. We demonstrated that 1) PACAP reduced basal smooth-muscle contractions in all portions of the gastrointestinal tract, but the effect of VIP was region-specific. The inhibitory effect of PACAP in midcolon was approximately 100 times greater than that of VIP. 2) PACAP significantly inhibited smoothmuscle contractions induced by acetylcholine or carbachol. The inhibitory effect of PACAP was not affected by hexamethonium and was additive to the inhibitory effect of atropine and pirenzepine. 3) PACAP inhibited smooth-muscle contractions induced by substance P, cholecystokinin, and galanin, even after atropine treatment. Although the exact mechanism of the inhibitory action of PACAP remains to be clarified, PACAP appears to exert its effect in the rat at a site other than muscarinic receptors, probably through a direct effect on gastrointestinal smooth muscle in vitro. Key words: Pituitary adenylate cyclase-activatingpolypeptide; rat GI tract; smooth-muscle contraction; vasoactive intestinal polypeptide Atilla Ertan, M . D . , Tulane University Medical School, 1430 Tulane A v e . , New Orleans, LA 70112, U S A
Pituitary adenylate cyclase-activating polypeptide (PACAP) with 38 residues was originally isolated from ovine hypothalamic tissues (1). The primary structure indicates that PACAP is a new member of the glucagon/secretin peptides family but the most homologous to vasoactive intestinal peptide (VIP). A shorter form with 27 residues (PACAP27) was subsequently isolated from the side fractions of hypothalamic extracts which were used for isolation of PACAP-38 (2). PACAP-27 and PACAP-38 have a comparable adenylate cyclase-stimulating activity (ACSA), as tested in rat pituitary cell cultures, but their ACSA is at least 1000 times greater than that of VIP (1). The rat pituitary, brain, adrenal medulla, and testis have a high-affinity specific binding site for PACAP which is not shared with VIP (3). On the other hand, lung, liver, duodenum, and other peripheral organs have a high-affinity binding site for PACAP which is shared with VIP (3). Therefore, there seem to be at least two subtypes of PACAP binding sites. Our preliminary immunohistochemical study with ovine digestive tract indicated that PACAP immunoreactive nerve fibers were present in smooth-muscle layers in all portions of intestine. PACAP fibers exist around Brunner's glands in submucosa of duodenum, and PACAP cell bodies were found in the myenteric plexus (4).This finding prompted us to investigate the effect of PACAP on the smooth muscle of the digestive tract. The present study was designed to investigate a possible effect of PACAP on contraction of smooth muscle of rat gastrointestinal tract in vitro. The interaction of acetyl-
choline (ACh) or other gut peptides with PACAP on smooth-muscle contractions was also investigated. MATERIALS AND METHODS Animals All experiments in this study were approved by the Advisory Committee for Animal Resources of Tulane University School of Medicine. Male rats of CD strain (230-290g body weight) from Charles River Laboratories (Wilmington, Mass., USA) were kept in animal quarters with controlled air, temperature, and light (on at 0600 h and off at 1800 h). They were allowed access to food and water ad lib. The animals were killed by decapitation, and stomach and various portions of intestine were rapidly removed. Segments of antrum and fundus of stomach, duodenum, jejunum, ileum, cecum, mid-colon, and descending colon were used in this study. Each segment was cleaned and placed in KrebsHenseleit buffer ( p H 7.4) containing in 118 mM NaCl, 4.75 mM KCl, 1.19 mM MgS04, 25 mM NaHCO,, 2.54 mM CaCl,, 1.19mM KH2P04, and 11mM glucose, and then gassed with 95% O2 + 5% COz at 37.5"C. Tissue strips were prepared as described previously (5). Each tissue strip was attached vertically in a longitudinal or circular orientation to a force transducer (Narco Bio-System Myograph-Transducers, F-60, Houston, Tex., USA) and was placed in a siliconized 25-ml organ bath (Harvard Apparatus, South Natick, Mass., USA) containing the Krebs-Henseleit buffer.
376
Z. Mungan et al.
4
C
4 Scand J Gastroenterol Downloaded from informahealthcare.com by UB Giessen on 10/30/14 For personal use only.
d
e
t f Fig. 1. Effect of pituitary adenylate cyclase-activating polypeptide with 27 residues (PACAP-27) (arrow, M) on basal contractions in different parts of the rat gastrointestinal tract: a) fundus, circular; b) antrum, circular; c) antrum, longitudinal; d) duodenum, longitudenal; e) jejunum,longitudinal; f) ilcum, longitudinal; g) cecum, circular; h) cecum, longitudinal; i) mid-colon, circular; and j ) mid-colon, longitudinal.
The temperature was maintained at 37.5"C throughout the experiment. Isometric contractions were recorded with a Narcotrace 40 (Narco Bio-System, Houston, Tex., USA). Base-line tension was set at 0.5 g and was equilibrated for 40-60min. Three to four doses of each peptide or agent were tested with a single tissue strip. After each test the tissue was rinsed twice and equilibrated for 15-20 min before the next sample was added. The viability of each tissue strip was tested with ACh at 5 x 10 -'M, at the beginning and at the end of each experiment. If a tissue strip did not show an appropriate response to ACh, the tissue was not used for further tests. Each experiment was repeated four to six times using different tissue strips. The mean 2 SE of responses to each test sample was calculated from the results obtained using these different strips. Peptides and agent Ovine PACAP-27 was synthesized by Dr. Chieko Kitada, Takeda Chemical Industries, Inc., Tsukuba Research Laboratories, Japan, and ovine PACAP-38 was synthesized by Drs. Yoshihiko Suitani and Janos Varga, American Peptide Co., Santa Clara, Calif., USA. VIP, substance P (SP), and galanin (Gal) were synthesized at our Peptide Research Laboratories. Cholecystokinin octapeptide (CCK-8) was purchased from Bachem Inc., Torrance, Calif., USA. ACh, carbachol, atropine sulfate, and hexamethonium were purchased from Sigma, St Louis, Mo., USA. Gastrozepine (pirenzepine, 2HCI (PZ)) was a gift from Boehringer Ingelheim Ltd., Ridgefield, Conn., USA. Each of these
peptides and agents was freshly dissolved in 250 pi deionized water in various concentrations and was then added to the solution bathing the tissue strip. Evaluation of effect of peptides and agents The response of smooth muscle of the tissue was determined by the change in the tonus and number of contractions recorded on the Narcotrace 40. ACh was used as a standard contractile agent at 5 x M in all experiments. The magnitude of contraction of the tissue strip induced by 5x M ACh was considered N O % , and any change in the ACh-induced contractions by the test was expressed in percentages. Statistical analysis Data of dose responses were analyzed using multiple linear regression. Group membership was indicated with dummy variables. The log transformation was applied to dose. Before the transformation, scores were multiplied by lo1'.
RESULTS Effect on basal smooth-muscle contractions In all gut segments tested (longitudinal and circular smooth muscle) except for circular smooth muscle of small intestine, both PACAP-27 and PACAP-38 reduced the basal contractions of smooth muscle at a dose range from lo-' to lo-' M. The basal contractions of circular smooth muscles of the small intestine were too small to demonstrate the
PACAP as GI Smooth-Muscle Relaxant Table I. Effect of pituitary adenylate cyclase-activating polypeptide with 27 residues (PACAP-27) and vasoactive intestinal polypeptide (VIP) on basal smooth-muscle contractions in different parts of rat gastrointestinal tract
to 10-4M did not affect the spontaneous smooth-muscle contractions of all segments tested.
inhibitory effect of peptide. The inhibitory effect of PACAP27 on basal smooth-muscle contractions in different parts of rat gastrointestinal tract is shown in Fig. 1. In our experiments VIP at lo-' to lo-' M also suppressed the basal contractions of circular smooth muscles of fundus and antrum, and both circular and longitudinal smooth muscles of cecum, mid-colon, and descending colon. However, in contrast to antrum, the inhibitory effect of VIP on circular smooth muscle of fundus was inconsistent. The relaxant effect of VIP was generally more prominent at the distal parts of rat gastrointestinal tract. VIP had various effects on the longitudinal smooth muscle of different segments o f the small intestine: contraction of duodenum, relaxation of jejunum, and relaxation or contraction of ileum (Table I). The relaxant activity of VIP in the basal contraction of circular muscle of mid-colon was compared with that of PACAP-27. The relaxant effect of PACAP-27 was found to be approximately 100 times greater than that of VIP (Table 11). The infective dose, 50% (ID5") value for PACAP-27 was 3.13 2 0.73nM, and for VIP it was 258.5 rt 80.2 nM (?SE). Atropine at to M and hexamethonium at
Effect of PACAP on induced contraction of smooth muscle ACh (10-6-10-5 M) induced contractions of smooth muscle in a dose- and time-dependent manner. Carbachol, M induced contractions an M2 and M3 agonist, at 5 x M that were about 60% of that induced by ACh at 5 x when tested using jejunum. PACAP-27 and PACAP-38 caused a comparable and dose-dependent inhibition of ACh-induced contractions of longitudinal smooth muscle of duodenum (Table 111). Atropine also showed dose-dependent inhibition of ACh-induced contractions (Table 111). The ID50(LSE) value for PACAP27 was 5.37 2 0.98 nM, for PACAP-38 was 9.33 3.70 nM, and for atropine was 8.30 ? 0.16 nM. Simultaneous application of atropine and PACAP-27 showed an additive inhibitory effect (Fig. 2A-C). Statistical analysis indicated that the inhibitory effect of PACAP-27 and atropine was similar, and both were more effective than PACAP-38 in inhibition ( p = 0.02 and p < 0.01, respectively). When PACAP-27, M, and ACh, M, were added to the bath solution simultaneously, the contractile activity of ACh was completely prevented (data not shown), as was the case with atropine and ACh. Hexamethonium had no effect on the ACh-induced contractions, indicating that ACh-induced contractions did not involve nicotinic receptors. PZ, an antagonist of the MI receptor, showed a dosedependent inhibition in the dose range 10-7-10-5 M. The maximum inhibition of ACh-induced contractions by PZ did not exceed 50% even at the highest concentration of M (data not shown). PZ-resistant contractions were further inhibited by M PACAP-27 (Fig. 2D). Conversely, PACAP-27-resistant contractions were inhibited by P Z (Fig. 2E). When PACAP-27 and PZ were added simultaneously, the inhibition was complete (Fig. 2F). These findings indicate that the inhibitory effects of P Z and PACAP-27 on ACh-induced contractions of smooth muscle were additive. M) induced smooth-muscle contracCarbachol ( 5 x tions to 60% that of ACh-induced contractions at the same
Table 11. Inhibitory effect of different doses of pituitary adenylate cyclase-activating polypeptide with 27 residues (PACAP-27) and vasoactive intestinal polypeptide (VIP) on basal smooth-muscle contractions on circular muscles of mid-colon (numbers represent means SE of four to six experiments, as a percentage)
Table 111. Inhibitory effect of different doses of pituitary adenylate cyclase-activating polypeptide with 27 and 38 residues (PACAP-27, PACAP-38) and atropine on acetylcholine-induced smooth-muscle contractions on duodenal longitudinal muscle (numbers represent mean SE of four to six experiments, as a percentage)
PACAP-27 Long. Fundus Antrum Duodenum Jejunum Ileum Cecum Mid-colon Descending colon Scand J Gastroenterol Downloaded from informahealthcare.com by UB Giessen on 10/30/14 For personal use only.
377
-
Circ.
Long.
Circ.
0 0
-
-
? ? ?
~
-
-*
+
? ? ?
-
-
VIP
*
-
-
-
-
-
-
-
-
-
-
Long. = longitudinal muscle; Circ. = circular muscle; - = relaxation; + = contraction; 0 = no effect; * = inconsistent; ? = could not be evaluated because of very small basal contractions.
*
Doses (M) 10-9
5x 10 10-7 5 x 10-7 10-6 5 x 10-6
PACAP-27 0 31.8 f 1.6 50.3 t 1.7 93.3 f 2.7
VIP 0
0 37.0 2 3.4 57.0 4.1 79.4 5 4.4 93.0 2 3.5
*
*
*
Doses (M)
PACAP-27
PACAP-38
Atropine
10-10
029.0 5 0.8 35.5 2 2.9 NT* 52.2 ? 3.2 70.2 2 2.3 NT
0 0 28.7 i 0.3 NT 32.4 2 3.1 58.4 i 3.6 NT
0 0 14.2 2 1.3 42.6 5 5.2 66.0 t 2 NT 100 +. 0
2.5 X 5x 7.5 x 10-9 5x 10-7
* N T = not tested.
Z . Mungan et a1
318
Atropine
PACAP-27
PACAP-27
Atropine
'"1 0
ACh
f
A ACh
f
E ACh
f
C
lm,fi &
Scand J Gastroenterol Downloaded from informahealthcare.com by UB Giessen on 10/30/14 For personal use only.
PAC AP-27
PZ
PACAP-27
0
E
PZ
50
0
ACh
f
-
ACh
f
ACh
P
F
1 mln
Fig. 2. Effect of pituitary adenylate cyclase-activating polypeptide with 27 residues (PACAP-27) (lo-' M), atropine (10 ' M), and pirenzepine (PZ) (1W6M) on acetylcholine (ACh) ( 5 X lO-'M)-contracted longitudinal smooth muscle of jejunum.
concentration. PACAP-27 (10 M) completely inhibited carbachol-induced contractions (Fig. 3). In contrast to PACAP, VIP did not show any inhibitory effect on AChinduced contractions of jejunal smooth muscle even at a very high concentration, such as M. PACAP-27 (loW8 M) also inhibited contractions of longitudinal smooth muscle of jejunum by SP M), CCK-8 (5 x lO-'M), and Gal (10-7M) (Fig. 3). Atropine did not show any inhibitory effect on SP-, CCK-8-, and Gal-induced contractions, and pretreatment with atropine did not block the inhibitory effect of PACAP-27 on these induced contractions. VIP, M, induced contractions of longitudinal smooth muscle of duodenum. Despite considerable homology with VIP, PACAP-27, lo-", also inhibited VIPinduced smooth-muscle contractions (Fig. 3). DISCUSSION These results clearly showed that PACAP suppressed both spontaneous and induced contractions of smooth muscle of the rat gastrointestinal tract as tested in vitro. In our experiments, we could not show a clear-cut inhibitory effect of VIP on gastric fundus, which was well reported (6, 7). This discrepancy may be related to methods and species of animals. In spite of considerable homology of structure the effect of VIP was region-specific. Suppression of AChinduced contractions by atropine or PZ, but not by hexamethonium, indicates that the contractions may involve mus-
carinic receptors. However, P Z did not block ACh-induced contractions completely, and these PZ-resistant contractions were further inhibited by PACAP. Therefore, PACAP may exert its inhibitory effect at a site different from the M I muscarinic binding site. Furthermore, PACAP appears to suppress smooth-muscle contractions induced by other gut peptides, and this effect continued after atropine pretreatment. It seems that PACAP suppresses induced contractions regardless of the contractile substance. It is therefore likely that PACAP exerts its inhibitory action through a specific binding site on its target cell and that the inhibitory action may affect a common step that is directly linked with relaxation of smooth muscle. Whether the intestinal smoothmuscle cell contains a specific binding site for PACAP remains to be clarified. Previous studies from our laboratories (3) indicated that rat duodenum membrane preparation contains specific, highaffinity PACAP-binding sites, although the number of binding sites was not as abundant as in the pituitary and hypothalamus. In contrast to the binding sites for PACAP in the pituitary and brain, the binding sites in the duodenum and other peripheral tissues appear to be shared with VIP (3). It is therefore very intriguing that VIP-induced contractions of duodenum were inhibited by PACAP, despite VIP's considerable homology with PACAP. Both VIP and PACAP stimulate adenylase cyclase in various tissues (1, 8-10), and in rat pituitary cell cultures the effect of PACAP is approximately 1000 times greater than that of VIP (1). VIP has
PACAP as GI Smooth-Muscle Relaxant
er-4
4
bP? 4
4
0
ACli
VIP
Scand J Gastroenterol Downloaded from informahealthcare.com by UB Giessen on 10/30/14 For personal use only.
4
0
4 b
0 4 0
4 0
CCK-I
Carbacliol
a
4
4 b
4
s I’
0
min 1
4
0
(;alanin
Fig. 3. Effect of pituitary adenylate cyclase-activating polypeptide with 27 residues (open arrow, M) on 1) acetylcholine (ACh) (5 X 10 6M)-, 2) carbachol (5 X M)-, 3) substance P (SP) (lo-’ M)-, 4) vasoactive intestinal polypeptide (VIP) (loT6M)-, 5 ) cholecystokinin-8(CCK-8) ( 5 x M)-, and 6) galanin (lo-’ M)induced contractions of longitudinal smooth muscle of jejunum and/ or duodenum. Responses to contractile agents were identical in duodenum and jejunum except VIP (see Table I).
mostly inhibitory effects on gastrointestinal smooth muscle (6, 8, l l ) , and this effect may be mediated through an increase of intracellular CAMP (8). VIP may act as a peptidergic neurotransmitter capable of directly inducing smoothmuscle relaxation (8). It is difficult to reconcile that the inhibition of VIP-induced contraction of duodenal smooth muscle by PACAP is mediated by further increase in intracellular CAMP. Although VIP causes relaxation of muscle cells from isolated longitudinal muscle layer, it causes atropine-sensitive contraction of muscle strips of the longitudinal muscle layer, implying the presence of VIP receptors on cholinergic neurons that mediate release of ACh (12). Recent immunohistochemical studies demonstrate that the distribution of VIP- and PACAP-immunoreactive nerve fibers in the small intestine was similar but that the network of VIP fibers was generally denser than that of PACAP. VIPimmunoreactive fibers were found in all layers of intestine,
379
whereas PACAP-positive fibers were few and very fine in the mucosa. In smooth-muscle layers both circular and longitudinal muscles were well innervated by both PACAP and VIP fibers. Brunner’s glands in the submucosa of the duodenum were surrounded by both VIP- and PACAP-immunoreactive fibers (4). Furthermore, the walls of blood vessels in the subserosal layer were well innervated by PACAP (4), suggesting that PACAP acts as a vasoregulatory substance. The findings in the present study, along with immunohistochemical studies, indicate that both VIP and PACAP may play a regulatory role in intestinal smooth muscle. However, the effects of these peptides on smooth muscle were different. It seems that VIP and PACAP act in opposite directions in some portions of intestine, and in the same direction in other portions. It is possible that these homologous peptides act through different receptors specific to each peptide under certain conditions and that they act through a common receptor under other conditions. The structure of PACAP appears to be well preserved in various species. cDNA for human (13) and rat (14) PACAP precursor has been cloned. The structure of rat and human PACAP was found to be identical to that of sheep PACAP. By means of an antiserum against PACAP-27, immunoreactive PACAP neurons and fibers were stained in the human brain (15). As in the ovine, PACAP immunoreactivity has also been observed in the human, porcine, feline, and murine gut wall (16). It is probable that PACAP released from the terminals of these nerve fibers may act directly on smooth-muscle cells through its specific binding site, thereby regulating motility of the gastrointestinal tract of the rat. ACKNOWLEDGEMENTS This study was supported in part by NIH grants DK-09094 and DK-30167, a grant from Takeda Chemical Industries, Ltd, and Gastroenterology Section Research Fund 530569. The authors thank Dr. Janet C . Rice for her help with statistical analysis and Lisa M. Baylor and Sebnem C. Salt for their ‘excellent technical assistance. REFERENCES 1 . Miyata A, Arimura A, Dahl RR, et al. Isolation of a novel 38 residue-hypothalamic polypeptide which stimulates adenylate cyclase in pituitary cells. Biochem Biophys Res Commun 1989$64567-74. 2. Miyata A, Jiang L, Dahl RR, et al. Isolation of a neuropeptide corresponding to the N-terminal 27 residues of the pituitary adenylate cyclase activating polypeptide with 38 residues (PACAP38). Biochem Biophys Res Commun 1990;170: 643-8. 3. Gottschall PE, Tatsuno I, Miyata A, Arimura A. Characterization and distribution of binding sites for the hypothalamic
peptide, pituitary adenylate cyclase activating polypeptide. Endocrinology 1990;127:272-7. 4. Koves K, Arimura A. Immunohistochemical demonstration of pituitary adenylate cyclase activating polypeptide in ovine gut [abstract]. Clin Res 1990;38:969A.
Scand J Gastroenterol Downloaded from informahealthcare.com by UB Giessen on 10/30/14 For personal use only.
380
Z. Mungun et al.
5. Nieber K , Milenov K, Bergmann J, Oehme P. Contractile response of longitudinal and circular smooth muscle. Acta Biol Med 1981;40:209-16. 6. Grider JR, Rivier JR. Vasoactive intestinal peptide (VIP) as transmitter of inhibitory motor neurons of the gut: evidence from the use of sclective VIP antagonist and VIP antiserum. J Pharmacol Exp Ther 1990;253:738-42. 7. De Buerme FA, Lefebvre RA. Influence of a-chymotripsin and trypsin on the non-adrenergic non-cholinergic relaxation in the rat gastric fundus. Br J Pharmacol 1987;91:171-7. 8. Makhlouf GM. Role of the VIP in the function of the gut. In: Said SI, editor. Vasoactive intestinal peptide. New York: Raven Press, 1982;425-44. 9. Buscail L, Gourlet P, Cauvin A, et al. Presence of highly selective receptors for PACAP (pituitary adenylate cyclase activating peptide) in membranes from the rat pancreatic-acinar cell line AR 4-2J. FEBS 1990;262:77-81. 10. Robbcrecht P, Gourlet P; Cauvin A, et al. PACAP and VIP receptors in rat liver membranes. Am J Physiol 1991;260:G97102. 11. Bodanszky M, Klausncr YS, Said SI. Biological activities of synthetic peptides corresponding to fragments of and to the Received 26 August 1991 Accepted 16 December 1991
entire sequence of the vasoactive intestinal peptide. Proc Natl Acad Sci 1973;70:382-4. 12. Makhlouf GM, Grider JR. Receptors for gut peptides on smooth muscle cells of the gut. In: Handbook of physiology, Section 6 . The gastrointestinal system, Vol 2. Bethesda, Md.: American Phvsioloev Societv. 1989:281-9. 13. Kimura ud, Ohkuia S, Ogi K, et al. A novel peptide which stimulates adenylate cyclase: molecular cloning and characterization of the ovine and human cDNAs. Biochem Biophys Res Commun 1990;166:81-9. 14. Ogi K, Kimura C, Onda H, Arimura A, Fujino M. Molecular cloning and characterization of rat pituitary adenylate cyclase activating polypeptide (PACAP). Biochem Biophys Res Commun 1990;173:127 1-9. 15. Vigh S, Arimura A, Koves K, Somogyvari-Ugh A, Sitton J, Fermin CD. Immunohistochemical localization of the neuropeptide, pituitary adenylate cyclase activating polypeptide (PACAP), in human and primate hypothalamus. Peptides 1991 ;l2:313-8. 16. Sundler F, Ekblad E, HIkanson R, Koves K, Arimura A. Pituitary adenylate cyclase activating polypeptide (PACAP), a novel VIP-like gut neuropeptide [abstract]. Digestion 1990;46 Suppl 1:111.
Gastroenterology Section and Peptide Research Laboratories, Dept. of Medicine, Tulane University School of Medicine, New Orleans, and US-Japan Biomedical Research Laboratories, Tulane University Herbert Center, Belle Chasse, Louisiana, USA
Scand J Gastroenterol Downloaded from informahealthcare.com by UB Giessen on 10/30/14 For personal use only.
Mungan Z, Arimura A, Ertan A, Rossowski WJ, Coy DH. Pituitary adenylate cyclase-activating polypeptide relaxes rat gastrointestinal smooth muscle. Scand J Gastroenterol 1992;27:375-380.
Pituitary adenylate cyclase-activating polypeptide (PACAP) is a new member of the secretin/glucagon peptides family, being most homologous to vasoactive intestinal peptide (VIP). The present study was designed to investigate a possible effect of PACAP on the rat gastrointestinal smooth muscle in vitro. We demonstrated that 1) PACAP reduced basal smooth-muscle contractions in all portions of the gastrointestinal tract, but the effect of VIP was region-specific. The inhibitory effect of PACAP in midcolon was approximately 100 times greater than that of VIP. 2) PACAP significantly inhibited smoothmuscle contractions induced by acetylcholine or carbachol. The inhibitory effect of PACAP was not affected by hexamethonium and was additive to the inhibitory effect of atropine and pirenzepine. 3) PACAP inhibited smooth-muscle contractions induced by substance P, cholecystokinin, and galanin, even after atropine treatment. Although the exact mechanism of the inhibitory action of PACAP remains to be clarified, PACAP appears to exert its effect in the rat at a site other than muscarinic receptors, probably through a direct effect on gastrointestinal smooth muscle in vitro. Key words: Pituitary adenylate cyclase-activatingpolypeptide; rat GI tract; smooth-muscle contraction; vasoactive intestinal polypeptide Atilla Ertan, M . D . , Tulane University Medical School, 1430 Tulane A v e . , New Orleans, LA 70112, U S A
Pituitary adenylate cyclase-activating polypeptide (PACAP) with 38 residues was originally isolated from ovine hypothalamic tissues (1). The primary structure indicates that PACAP is a new member of the glucagon/secretin peptides family but the most homologous to vasoactive intestinal peptide (VIP). A shorter form with 27 residues (PACAP27) was subsequently isolated from the side fractions of hypothalamic extracts which were used for isolation of PACAP-38 (2). PACAP-27 and PACAP-38 have a comparable adenylate cyclase-stimulating activity (ACSA), as tested in rat pituitary cell cultures, but their ACSA is at least 1000 times greater than that of VIP (1). The rat pituitary, brain, adrenal medulla, and testis have a high-affinity specific binding site for PACAP which is not shared with VIP (3). On the other hand, lung, liver, duodenum, and other peripheral organs have a high-affinity binding site for PACAP which is shared with VIP (3). Therefore, there seem to be at least two subtypes of PACAP binding sites. Our preliminary immunohistochemical study with ovine digestive tract indicated that PACAP immunoreactive nerve fibers were present in smooth-muscle layers in all portions of intestine. PACAP fibers exist around Brunner's glands in submucosa of duodenum, and PACAP cell bodies were found in the myenteric plexus (4).This finding prompted us to investigate the effect of PACAP on the smooth muscle of the digestive tract. The present study was designed to investigate a possible effect of PACAP on contraction of smooth muscle of rat gastrointestinal tract in vitro. The interaction of acetyl-
choline (ACh) or other gut peptides with PACAP on smooth-muscle contractions was also investigated. MATERIALS AND METHODS Animals All experiments in this study were approved by the Advisory Committee for Animal Resources of Tulane University School of Medicine. Male rats of CD strain (230-290g body weight) from Charles River Laboratories (Wilmington, Mass., USA) were kept in animal quarters with controlled air, temperature, and light (on at 0600 h and off at 1800 h). They were allowed access to food and water ad lib. The animals were killed by decapitation, and stomach and various portions of intestine were rapidly removed. Segments of antrum and fundus of stomach, duodenum, jejunum, ileum, cecum, mid-colon, and descending colon were used in this study. Each segment was cleaned and placed in KrebsHenseleit buffer ( p H 7.4) containing in 118 mM NaCl, 4.75 mM KCl, 1.19 mM MgS04, 25 mM NaHCO,, 2.54 mM CaCl,, 1.19mM KH2P04, and 11mM glucose, and then gassed with 95% O2 + 5% COz at 37.5"C. Tissue strips were prepared as described previously (5). Each tissue strip was attached vertically in a longitudinal or circular orientation to a force transducer (Narco Bio-System Myograph-Transducers, F-60, Houston, Tex., USA) and was placed in a siliconized 25-ml organ bath (Harvard Apparatus, South Natick, Mass., USA) containing the Krebs-Henseleit buffer.
376
Z. Mungan et al.
4
C
4 Scand J Gastroenterol Downloaded from informahealthcare.com by UB Giessen on 10/30/14 For personal use only.
d
e
t f Fig. 1. Effect of pituitary adenylate cyclase-activating polypeptide with 27 residues (PACAP-27) (arrow, M) on basal contractions in different parts of the rat gastrointestinal tract: a) fundus, circular; b) antrum, circular; c) antrum, longitudinal; d) duodenum, longitudenal; e) jejunum,longitudinal; f) ilcum, longitudinal; g) cecum, circular; h) cecum, longitudinal; i) mid-colon, circular; and j ) mid-colon, longitudinal.
The temperature was maintained at 37.5"C throughout the experiment. Isometric contractions were recorded with a Narcotrace 40 (Narco Bio-System, Houston, Tex., USA). Base-line tension was set at 0.5 g and was equilibrated for 40-60min. Three to four doses of each peptide or agent were tested with a single tissue strip. After each test the tissue was rinsed twice and equilibrated for 15-20 min before the next sample was added. The viability of each tissue strip was tested with ACh at 5 x 10 -'M, at the beginning and at the end of each experiment. If a tissue strip did not show an appropriate response to ACh, the tissue was not used for further tests. Each experiment was repeated four to six times using different tissue strips. The mean 2 SE of responses to each test sample was calculated from the results obtained using these different strips. Peptides and agent Ovine PACAP-27 was synthesized by Dr. Chieko Kitada, Takeda Chemical Industries, Inc., Tsukuba Research Laboratories, Japan, and ovine PACAP-38 was synthesized by Drs. Yoshihiko Suitani and Janos Varga, American Peptide Co., Santa Clara, Calif., USA. VIP, substance P (SP), and galanin (Gal) were synthesized at our Peptide Research Laboratories. Cholecystokinin octapeptide (CCK-8) was purchased from Bachem Inc., Torrance, Calif., USA. ACh, carbachol, atropine sulfate, and hexamethonium were purchased from Sigma, St Louis, Mo., USA. Gastrozepine (pirenzepine, 2HCI (PZ)) was a gift from Boehringer Ingelheim Ltd., Ridgefield, Conn., USA. Each of these
peptides and agents was freshly dissolved in 250 pi deionized water in various concentrations and was then added to the solution bathing the tissue strip. Evaluation of effect of peptides and agents The response of smooth muscle of the tissue was determined by the change in the tonus and number of contractions recorded on the Narcotrace 40. ACh was used as a standard contractile agent at 5 x M in all experiments. The magnitude of contraction of the tissue strip induced by 5x M ACh was considered N O % , and any change in the ACh-induced contractions by the test was expressed in percentages. Statistical analysis Data of dose responses were analyzed using multiple linear regression. Group membership was indicated with dummy variables. The log transformation was applied to dose. Before the transformation, scores were multiplied by lo1'.
RESULTS Effect on basal smooth-muscle contractions In all gut segments tested (longitudinal and circular smooth muscle) except for circular smooth muscle of small intestine, both PACAP-27 and PACAP-38 reduced the basal contractions of smooth muscle at a dose range from lo-' to lo-' M. The basal contractions of circular smooth muscles of the small intestine were too small to demonstrate the
PACAP as GI Smooth-Muscle Relaxant Table I. Effect of pituitary adenylate cyclase-activating polypeptide with 27 residues (PACAP-27) and vasoactive intestinal polypeptide (VIP) on basal smooth-muscle contractions in different parts of rat gastrointestinal tract
to 10-4M did not affect the spontaneous smooth-muscle contractions of all segments tested.
inhibitory effect of peptide. The inhibitory effect of PACAP27 on basal smooth-muscle contractions in different parts of rat gastrointestinal tract is shown in Fig. 1. In our experiments VIP at lo-' to lo-' M also suppressed the basal contractions of circular smooth muscles of fundus and antrum, and both circular and longitudinal smooth muscles of cecum, mid-colon, and descending colon. However, in contrast to antrum, the inhibitory effect of VIP on circular smooth muscle of fundus was inconsistent. The relaxant effect of VIP was generally more prominent at the distal parts of rat gastrointestinal tract. VIP had various effects on the longitudinal smooth muscle of different segments o f the small intestine: contraction of duodenum, relaxation of jejunum, and relaxation or contraction of ileum (Table I). The relaxant activity of VIP in the basal contraction of circular muscle of mid-colon was compared with that of PACAP-27. The relaxant effect of PACAP-27 was found to be approximately 100 times greater than that of VIP (Table 11). The infective dose, 50% (ID5") value for PACAP-27 was 3.13 2 0.73nM, and for VIP it was 258.5 rt 80.2 nM (?SE). Atropine at to M and hexamethonium at
Effect of PACAP on induced contraction of smooth muscle ACh (10-6-10-5 M) induced contractions of smooth muscle in a dose- and time-dependent manner. Carbachol, M induced contractions an M2 and M3 agonist, at 5 x M that were about 60% of that induced by ACh at 5 x when tested using jejunum. PACAP-27 and PACAP-38 caused a comparable and dose-dependent inhibition of ACh-induced contractions of longitudinal smooth muscle of duodenum (Table 111). Atropine also showed dose-dependent inhibition of ACh-induced contractions (Table 111). The ID50(LSE) value for PACAP27 was 5.37 2 0.98 nM, for PACAP-38 was 9.33 3.70 nM, and for atropine was 8.30 ? 0.16 nM. Simultaneous application of atropine and PACAP-27 showed an additive inhibitory effect (Fig. 2A-C). Statistical analysis indicated that the inhibitory effect of PACAP-27 and atropine was similar, and both were more effective than PACAP-38 in inhibition ( p = 0.02 and p < 0.01, respectively). When PACAP-27, M, and ACh, M, were added to the bath solution simultaneously, the contractile activity of ACh was completely prevented (data not shown), as was the case with atropine and ACh. Hexamethonium had no effect on the ACh-induced contractions, indicating that ACh-induced contractions did not involve nicotinic receptors. PZ, an antagonist of the MI receptor, showed a dosedependent inhibition in the dose range 10-7-10-5 M. The maximum inhibition of ACh-induced contractions by PZ did not exceed 50% even at the highest concentration of M (data not shown). PZ-resistant contractions were further inhibited by M PACAP-27 (Fig. 2D). Conversely, PACAP-27-resistant contractions were inhibited by P Z (Fig. 2E). When PACAP-27 and PZ were added simultaneously, the inhibition was complete (Fig. 2F). These findings indicate that the inhibitory effects of P Z and PACAP-27 on ACh-induced contractions of smooth muscle were additive. M) induced smooth-muscle contracCarbachol ( 5 x tions to 60% that of ACh-induced contractions at the same
Table 11. Inhibitory effect of different doses of pituitary adenylate cyclase-activating polypeptide with 27 residues (PACAP-27) and vasoactive intestinal polypeptide (VIP) on basal smooth-muscle contractions on circular muscles of mid-colon (numbers represent means SE of four to six experiments, as a percentage)
Table 111. Inhibitory effect of different doses of pituitary adenylate cyclase-activating polypeptide with 27 and 38 residues (PACAP-27, PACAP-38) and atropine on acetylcholine-induced smooth-muscle contractions on duodenal longitudinal muscle (numbers represent mean SE of four to six experiments, as a percentage)
PACAP-27 Long. Fundus Antrum Duodenum Jejunum Ileum Cecum Mid-colon Descending colon Scand J Gastroenterol Downloaded from informahealthcare.com by UB Giessen on 10/30/14 For personal use only.
377
-
Circ.
Long.
Circ.
0 0
-
-
? ? ?
~
-
-*
+
? ? ?
-
-
VIP
*
-
-
-
-
-
-
-
-
-
-
Long. = longitudinal muscle; Circ. = circular muscle; - = relaxation; + = contraction; 0 = no effect; * = inconsistent; ? = could not be evaluated because of very small basal contractions.
*
Doses (M) 10-9
5x 10 10-7 5 x 10-7 10-6 5 x 10-6
PACAP-27 0 31.8 f 1.6 50.3 t 1.7 93.3 f 2.7
VIP 0
0 37.0 2 3.4 57.0 4.1 79.4 5 4.4 93.0 2 3.5
*
*
*
Doses (M)
PACAP-27
PACAP-38
Atropine
10-10
029.0 5 0.8 35.5 2 2.9 NT* 52.2 ? 3.2 70.2 2 2.3 NT
0 0 28.7 i 0.3 NT 32.4 2 3.1 58.4 i 3.6 NT
0 0 14.2 2 1.3 42.6 5 5.2 66.0 t 2 NT 100 +. 0
2.5 X 5x 7.5 x 10-9 5x 10-7
* N T = not tested.
Z . Mungan et a1
318
Atropine
PACAP-27
PACAP-27
Atropine
'"1 0
ACh
f
A ACh
f
E ACh
f
C
lm,fi &
Scand J Gastroenterol Downloaded from informahealthcare.com by UB Giessen on 10/30/14 For personal use only.
PAC AP-27
PZ
PACAP-27
0
E
PZ
50
0
ACh
f
-
ACh
f
ACh
P
F
1 mln
Fig. 2. Effect of pituitary adenylate cyclase-activating polypeptide with 27 residues (PACAP-27) (lo-' M), atropine (10 ' M), and pirenzepine (PZ) (1W6M) on acetylcholine (ACh) ( 5 X lO-'M)-contracted longitudinal smooth muscle of jejunum.
concentration. PACAP-27 (10 M) completely inhibited carbachol-induced contractions (Fig. 3). In contrast to PACAP, VIP did not show any inhibitory effect on AChinduced contractions of jejunal smooth muscle even at a very high concentration, such as M. PACAP-27 (loW8 M) also inhibited contractions of longitudinal smooth muscle of jejunum by SP M), CCK-8 (5 x lO-'M), and Gal (10-7M) (Fig. 3). Atropine did not show any inhibitory effect on SP-, CCK-8-, and Gal-induced contractions, and pretreatment with atropine did not block the inhibitory effect of PACAP-27 on these induced contractions. VIP, M, induced contractions of longitudinal smooth muscle of duodenum. Despite considerable homology with VIP, PACAP-27, lo-", also inhibited VIPinduced smooth-muscle contractions (Fig. 3). DISCUSSION These results clearly showed that PACAP suppressed both spontaneous and induced contractions of smooth muscle of the rat gastrointestinal tract as tested in vitro. In our experiments, we could not show a clear-cut inhibitory effect of VIP on gastric fundus, which was well reported (6, 7). This discrepancy may be related to methods and species of animals. In spite of considerable homology of structure the effect of VIP was region-specific. Suppression of AChinduced contractions by atropine or PZ, but not by hexamethonium, indicates that the contractions may involve mus-
carinic receptors. However, P Z did not block ACh-induced contractions completely, and these PZ-resistant contractions were further inhibited by PACAP. Therefore, PACAP may exert its inhibitory effect at a site different from the M I muscarinic binding site. Furthermore, PACAP appears to suppress smooth-muscle contractions induced by other gut peptides, and this effect continued after atropine pretreatment. It seems that PACAP suppresses induced contractions regardless of the contractile substance. It is therefore likely that PACAP exerts its inhibitory action through a specific binding site on its target cell and that the inhibitory action may affect a common step that is directly linked with relaxation of smooth muscle. Whether the intestinal smoothmuscle cell contains a specific binding site for PACAP remains to be clarified. Previous studies from our laboratories (3) indicated that rat duodenum membrane preparation contains specific, highaffinity PACAP-binding sites, although the number of binding sites was not as abundant as in the pituitary and hypothalamus. In contrast to the binding sites for PACAP in the pituitary and brain, the binding sites in the duodenum and other peripheral tissues appear to be shared with VIP (3). It is therefore very intriguing that VIP-induced contractions of duodenum were inhibited by PACAP, despite VIP's considerable homology with PACAP. Both VIP and PACAP stimulate adenylase cyclase in various tissues (1, 8-10), and in rat pituitary cell cultures the effect of PACAP is approximately 1000 times greater than that of VIP (1). VIP has
PACAP as GI Smooth-Muscle Relaxant
er-4
4
bP? 4
4
0
ACli
VIP
Scand J Gastroenterol Downloaded from informahealthcare.com by UB Giessen on 10/30/14 For personal use only.
4
0
4 b
0 4 0
4 0
CCK-I
Carbacliol
a
4
4 b
4
s I’
0
min 1
4
0
(;alanin
Fig. 3. Effect of pituitary adenylate cyclase-activating polypeptide with 27 residues (open arrow, M) on 1) acetylcholine (ACh) (5 X 10 6M)-, 2) carbachol (5 X M)-, 3) substance P (SP) (lo-’ M)-, 4) vasoactive intestinal polypeptide (VIP) (loT6M)-, 5 ) cholecystokinin-8(CCK-8) ( 5 x M)-, and 6) galanin (lo-’ M)induced contractions of longitudinal smooth muscle of jejunum and/ or duodenum. Responses to contractile agents were identical in duodenum and jejunum except VIP (see Table I).
mostly inhibitory effects on gastrointestinal smooth muscle (6, 8, l l ) , and this effect may be mediated through an increase of intracellular CAMP (8). VIP may act as a peptidergic neurotransmitter capable of directly inducing smoothmuscle relaxation (8). It is difficult to reconcile that the inhibition of VIP-induced contraction of duodenal smooth muscle by PACAP is mediated by further increase in intracellular CAMP. Although VIP causes relaxation of muscle cells from isolated longitudinal muscle layer, it causes atropine-sensitive contraction of muscle strips of the longitudinal muscle layer, implying the presence of VIP receptors on cholinergic neurons that mediate release of ACh (12). Recent immunohistochemical studies demonstrate that the distribution of VIP- and PACAP-immunoreactive nerve fibers in the small intestine was similar but that the network of VIP fibers was generally denser than that of PACAP. VIPimmunoreactive fibers were found in all layers of intestine,
379
whereas PACAP-positive fibers were few and very fine in the mucosa. In smooth-muscle layers both circular and longitudinal muscles were well innervated by both PACAP and VIP fibers. Brunner’s glands in the submucosa of the duodenum were surrounded by both VIP- and PACAP-immunoreactive fibers (4). Furthermore, the walls of blood vessels in the subserosal layer were well innervated by PACAP (4), suggesting that PACAP acts as a vasoregulatory substance. The findings in the present study, along with immunohistochemical studies, indicate that both VIP and PACAP may play a regulatory role in intestinal smooth muscle. However, the effects of these peptides on smooth muscle were different. It seems that VIP and PACAP act in opposite directions in some portions of intestine, and in the same direction in other portions. It is possible that these homologous peptides act through different receptors specific to each peptide under certain conditions and that they act through a common receptor under other conditions. The structure of PACAP appears to be well preserved in various species. cDNA for human (13) and rat (14) PACAP precursor has been cloned. The structure of rat and human PACAP was found to be identical to that of sheep PACAP. By means of an antiserum against PACAP-27, immunoreactive PACAP neurons and fibers were stained in the human brain (15). As in the ovine, PACAP immunoreactivity has also been observed in the human, porcine, feline, and murine gut wall (16). It is probable that PACAP released from the terminals of these nerve fibers may act directly on smooth-muscle cells through its specific binding site, thereby regulating motility of the gastrointestinal tract of the rat. ACKNOWLEDGEMENTS This study was supported in part by NIH grants DK-09094 and DK-30167, a grant from Takeda Chemical Industries, Ltd, and Gastroenterology Section Research Fund 530569. The authors thank Dr. Janet C . Rice for her help with statistical analysis and Lisa M. Baylor and Sebnem C. Salt for their ‘excellent technical assistance. REFERENCES 1 . Miyata A, Arimura A, Dahl RR, et al. Isolation of a novel 38 residue-hypothalamic polypeptide which stimulates adenylate cyclase in pituitary cells. Biochem Biophys Res Commun 1989$64567-74. 2. Miyata A, Jiang L, Dahl RR, et al. Isolation of a neuropeptide corresponding to the N-terminal 27 residues of the pituitary adenylate cyclase activating polypeptide with 38 residues (PACAP38). Biochem Biophys Res Commun 1990;170: 643-8. 3. Gottschall PE, Tatsuno I, Miyata A, Arimura A. Characterization and distribution of binding sites for the hypothalamic
peptide, pituitary adenylate cyclase activating polypeptide. Endocrinology 1990;127:272-7. 4. Koves K, Arimura A. Immunohistochemical demonstration of pituitary adenylate cyclase activating polypeptide in ovine gut [abstract]. Clin Res 1990;38:969A.
Scand J Gastroenterol Downloaded from informahealthcare.com by UB Giessen on 10/30/14 For personal use only.
380
Z. Mungun et al.
5. Nieber K , Milenov K, Bergmann J, Oehme P. Contractile response of longitudinal and circular smooth muscle. Acta Biol Med 1981;40:209-16. 6. Grider JR, Rivier JR. Vasoactive intestinal peptide (VIP) as transmitter of inhibitory motor neurons of the gut: evidence from the use of sclective VIP antagonist and VIP antiserum. J Pharmacol Exp Ther 1990;253:738-42. 7. De Buerme FA, Lefebvre RA. Influence of a-chymotripsin and trypsin on the non-adrenergic non-cholinergic relaxation in the rat gastric fundus. Br J Pharmacol 1987;91:171-7. 8. Makhlouf GM. Role of the VIP in the function of the gut. In: Said SI, editor. Vasoactive intestinal peptide. New York: Raven Press, 1982;425-44. 9. Buscail L, Gourlet P, Cauvin A, et al. Presence of highly selective receptors for PACAP (pituitary adenylate cyclase activating peptide) in membranes from the rat pancreatic-acinar cell line AR 4-2J. FEBS 1990;262:77-81. 10. Robbcrecht P, Gourlet P; Cauvin A, et al. PACAP and VIP receptors in rat liver membranes. Am J Physiol 1991;260:G97102. 11. Bodanszky M, Klausncr YS, Said SI. Biological activities of synthetic peptides corresponding to fragments of and to the Received 26 August 1991 Accepted 16 December 1991
entire sequence of the vasoactive intestinal peptide. Proc Natl Acad Sci 1973;70:382-4. 12. Makhlouf GM, Grider JR. Receptors for gut peptides on smooth muscle cells of the gut. In: Handbook of physiology, Section 6 . The gastrointestinal system, Vol 2. Bethesda, Md.: American Phvsioloev Societv. 1989:281-9. 13. Kimura ud, Ohkuia S, Ogi K, et al. A novel peptide which stimulates adenylate cyclase: molecular cloning and characterization of the ovine and human cDNAs. Biochem Biophys Res Commun 1990;166:81-9. 14. Ogi K, Kimura C, Onda H, Arimura A, Fujino M. Molecular cloning and characterization of rat pituitary adenylate cyclase activating polypeptide (PACAP). Biochem Biophys Res Commun 1990;173:127 1-9. 15. Vigh S, Arimura A, Koves K, Somogyvari-Ugh A, Sitton J, Fermin CD. Immunohistochemical localization of the neuropeptide, pituitary adenylate cyclase activating polypeptide (PACAP), in human and primate hypothalamus. Peptides 1991 ;l2:313-8. 16. Sundler F, Ekblad E, HIkanson R, Koves K, Arimura A. Pituitary adenylate cyclase activating polypeptide (PACAP), a novel VIP-like gut neuropeptide [abstract]. Digestion 1990;46 Suppl 1:111.
