- Email: [email protected]
Central somatostatin diminished inhibitory action of central CGRP on pancreatic basal secretion in conscious rats
Journal of the Autonomic Nervous System 73 Ž1998. 143–148
Central somatostatin diminished inhibitory action of central CGRP on pancreatic basal secretion in conscious rats Kyoko Miyasaka a
a,)
, Setsuko Kanai a , Masao Masuda a , Akihiro Funakoshi
b
Department of Clinical Physiology, Tokyo Metropolitan Institute of Gerontology, 35-2 Sakaecho, Itabashiku, Tokyo-1730015, Japan b DiÕision of Gastroenterology, National Kyushu Cancer Center, Fukuoka-8111347, Japan Received 3 June 1998; revised 3 July 1998; accepted 24 July 1998
Abstract We examined whether central somatostatin prevents an inhibitory effect of central calcitonin-gene related peptide ŽCGRP. on pancreatic secretion in conscious male Wistar rats Ž330–330 g.. Rats were prepared with separate cannulas for draining bile and pancreatic juice and with a duodenal cannula and an extrajugular vein cannula. In addition, another cannula was stereotactically implanted into the left lateral cerebral ventricle. Rats were placed in restraint cages and experiments were conducted 4 days after the operation without anesthesia. An injection of CGRP Ž0.1, 1.0 nmolr10 ml. into the left lateral cerebral ventricle Ži.c.v.. inhibited pancreatic secretion dose-dependently. To confirm the inhibitory effect of CGRP Ži.c.v.. was mediated via sympathetic nerves, phentolamine was injected intravenously Ži.v.. bolus Ž0.5 mg kgy1 . 0.5-h before CGRP Ži.c.v.., followed by continuous infusion of 0.2 mg kgy1 hy1. Phentolamine Ži.v.. reversed the inhibition produced by CGRP Ži.c.v... An injection of 4 nmolr10 ml somatostatin Ži.c.v.. 5 min prior to CGRP injection diminished the inhibitory effect of CGRP Ži.c.v... It is concluded that centrally administered somatostatin diminished the inhibitory action of CGRP Ži.c.v.. on pancreatic secretion, probably via inhibiting autonomic Žsympathetic. nerve excitation at the central site. q 1998 Elsevier Science B.V. All rights reserved. Keywords: Somatostatin; Intracerebroventricular administration; Calcitonin-gene related peptide; CGRP; Pancreatic secretion
1. Introduction As pancreatic exocrine secretion is controlled by both autonomic nerve and hormonal signals, the regulation of these signals concerning this secretion in the central nervous system ŽCNS. is therefore important ŽMessmer et al., 1993; Masuda et al., 1995, 1997, 1998.. The input from the abdominal cavity runs through vagal and spinal afferent fibers ŽRenehan et al., 1995., and the vagal afferent fibers project to interneurons in the nucleus of the solitary tract ŽNTS. ŽLoewy and Haxhiu, 1993.. Calcitonin-gene related peptide ŽCGRP. has been known to be one of major neurotransmitters of splanchnic afferent fibers, and CGRP-containing terminals are also present in the NTS ŽHolzer, 1988.. Concerning the neural output from the
) Corresponding author. Tel.: q81 813 39643241 ext. 3089; fax: q81 813 35794776; e-mail: [email protected]
CNS, the excitation of vagal efferent fibers originating from the dorsal vagal complex ŽDVC. causes pancreatic hypersecretion, while excitation of the sympathetic nerve inhibits pancreatic secretion ŽNelson et al., 1993; Rogers et al., 1995.. It has been reported ŽKato and Kanno, 1983; Messmer et al., 1993; Rogers et al., 1995; Masuda et al., 1997. that central administration of thyrotropin releasing hormone ŽTRH. produces vagal efferent nerve excitation, resulting in pancreatic hypersecretion, and also that centrally administered CGRP inhibits pancreatic secretion via sympathetic nerve excitation, because the inhibitory effect is not affected by vagotomy, but abolished by peripheral administration of phentolamine Ž a-adrenergic receptor blocker.. Somatostatin, originally identified from ovine hypothalamus as a physiological regulator of growth hormone release from the anterior pituitary ŽReisine, 1995.. Somatostatin and its receptors are widely distributed in the CNS, and five different somatostatin receptors were cloned and type 1 and 2 were distributed with highest levels in
0165-1838r98r$ - see front matter q 1998 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 5 - 1 8 3 8 Ž 9 8 . 0 0 1 3 0 - 1
144
K. Miyasaka et al.r Journal of the Autonomic NerÕous System 73 (1998) 143–148
hippocampus, hypothalamus, cortex, and amygdala ŽFinley et al., 1981; Uhl et al., 1985; Reisine, 1995.. Although several behavioral studies have suggested that somatostatin in the brain modulates locomotor activity and cognitive functions ŽReisine, 1995., the multiple functions of central somatostatin have not been fully elucidated. It has been shown that both intracerebroventricular injection of TRH and intravenous infusion of 2-deoxy-D-glucose Ž2DG. increase pancreatic secretion via excitation of the vagal efferent nerve because the stimulatory effects are completely eliminated by vagotomy ŽRoze ´ et al., 1977; Kato and Kanno, 1983; Tache´ and Yang, 1990.. We have found ŽMasuda et al., 1997. that somatostatin injected into the cerebral ventricle abolishes the stimulatory effect of centrally administered TRH, but not of peripheral administration of 2DG. Thus, we suggest that central somatostatin might affect autonomic nerve excitation produced by peptidergic ŽTRH. neurons. In the present study, we examined whether somatostatin injected into the cerebral ventricle prevents the inhibitory action of central CGRP on pancreatic secretion.
2. Materials and methods 2.1. Materials CGRP and somatostatin-14 were purchased from the Peptide Institute, Osaka, Japan. Phentolamine was purchased from Sigma, St. Louis, MO. The cannulae used in this study were Silastic Medical Grade Tubing ŽDow-Corning, Midland, MI.; 0.025 in. i.d., 0.037 in. o.d... 2.2. Animal preparation Male Wistar rats ŽNippon SLC, Shizuoka, Japan. weighing 300–328 g were used. The operative procedure has been described in detail previously ŽMiyasaka et al., 1986.. Briefly, after intraperitoneal anesthesia with sodium pentobarbital Ž15 mgr300 g body weight., a cannula was inserted into the right jugular vein. For intracerebroventricular injection of CGRP and somatostatin, a metal cannula Ža modification of a 23-gauge needle. was implanted stereotactically into the left lateral ventricle ŽAP—1.0 mm, L—1.3 mm, DV—4.0 mm. ŽMasuda et al., 1995, 1997, 1998.. A midline abdominal incision was then made, and a cannula was inserted into the common bile duct proximal to the ampulla of Vater. The common bile duct was ligated proximally to the pancreas near the liver, and a second cannula was inserted above the ligation below the bifurcation of the bile duct. Thus, pure bile and pure pancreatic juice were collected separately. A third cannula was inserted into the duodenum to return bile and pancreatic juice, its outlet tip located near the ampulla of Vater. All cannulae were initially brought into the abdominal cavity through a subcutaneous channel in the back near the tail.
After the operation, the rats were placed in modified Bollman-type restraint cages under conditions of controlled temperature Ž248C. and illumination Ž12-h lightrdark cycle starting at 5 AM., and given free access to food and water. Bile and pancreatic juice were continuously returned to the intestine via the duodenal cannula. Experiments were conducted on day 4 after the operation, after withholding food for 5 h ŽMiyasaka et al., 1986.. At the end of each experiment, rats were sacrificed, and black ink was injected into the lateral ventricle in the rats. Successful cannulation was verified by the visualization. 2.3. Measurement of pancreatic fluid and protein output Bile and pancreatic juice were collected separately for 30-min periods, and the volume of pancreatic juice was measured with a Hamilton syringe. Samples of 15 ml of pancreatic juice were used for determination of protein concentrations, and protein concentration in the pancreatic juice was determined from the optical density at 280 nm ŽKeller et al., 1958. of samples diluted 200-fold with 0.04 M Tris buffer, pH 7.8. The remainder was mixed with the bile and infused into the duodenum with a syringe pump ŽCompact Infusion Pump, Harvard Apparatus, Southnatick, MA. over the next 30 min. The pooled bile and pancreatic juice were infused during the first 30 min of the experiment. Bile and pancreatic juice were continuously returned to the intestine throughout the experimental period. 2.4. Effect and mechanism of intracerebroÕentricular administration of CGRP on pancreatic secretion CGRP was dissolved in 1% BSA. After a 90-min basal collection with return of bile and pancreatic juice, CGRP Ž0.1 or 1 nmolr10 ml. or vehicle was injected into the left lateral ventricle with a Hamilton microsyringe, and pancreatic secretions were collected for the following 2 h. To confirm the inhibitory mechanism of central CGRP via sympathetic Ž a-adrenergic. nerves, phentolamine was injected intravenously bolus Ž0.5 mg kgy1 . 30 min before intracerebroventricular injection of CGRP, followed by continuous infusion of 0.2 mg kgy1 hy1 . Since CGRP may leak from the cerebrospinal fluid into the systemic circulation and exert biological actions, the effects of CGRP Ž1 nmol mly1 . given intravenously on the stimulation of pancreatic secretion were examined. 2.5. Effect of intracerebroÕentricular injection of somatostatin A 4 nmolr10 ml dose of somatostatin was injected into the cerebral ventricle 5 min prior to CGRP injection Ž1 nmolr10 ml.. The changes in pancreatic secretion were monitored as described above. Based on intracerebroventricular and intravenous injection of somatostatin, it has been verified in a previous study ŽMasuda et al., 1995. that
K. Miyasaka et al.r Journal of the Autonomic NerÕous System 73 (1998) 143–148
145
peated measures followed by Scheffe’s multiple comparison test. A P - 0.05 was considered significant.
3. Results 3.1. Effect and mechanism of intracerebroÕentricular administration of CGRP on basal pancreatic secretion The changes in pancreatic fluid and protein outputs were monitored. Under the present experimental condition, intracerebroventricular injection of isotonic saline Ž10 ml, injected into the left lateral ventricle. did not affect either fluid or protein output significantly. The results of protein output are shown in Fig. 1A.
Fig. 1. Effects of intracerebroventricular administration of CGRP Ž1.0 nmol. on basal pancreatic protein secretion ŽA.. The results of the statistical analysis are shown in the text. An arrow indicates the injection of CGRP. ), significantly different from the values of vehicle injection. Decremental changes during 2 h after intracerebroventricular administration of CGRP Ž0.1, 1.0 nmol. ŽB.. Values are significantly different Ž F s 4.58, P - 0.03.. †, significantly different from the other two values. Values are means"SE. Number of rats is indicated in parentheses.
somatostatin alone injected intracerebroventricularly does not affect basal pancreatic fluid and protein secretion. Moreover, although this dose of somatostatin increased circulating plasma somatostatin levels from a basal level of 27.0 " 7.0 pM to 83.3 " 8.6 pM, the circulating level of somatostatin at 66.2 " 15.1 pM produced by intravenous infusion of 100 ng kgy1 of somatostatin did not affect pancreatic secretion ŽMasuda et al., 1995.. 2.6. Statistical analysis All results are expressed as means " SE. Results were analyzed by a One-way analysis of variance ŽANOVA. or by a multiple analysis of variance ŽMANOVA. with re-
Fig. 2. The decremental changes in protein ŽA. and fluid secretion ŽB. during the 2 h after intracerebroventricular administration of CGRP Ž1.0 nmol. with or without phentolamine. The levels of protein secretion are significantly different Ž F s 4.3, P - 0.04. but the levels of fluid secretion are not Ž F s 3.5, P ) 0.06.. †, significantly different from the other two values. Number of rats is indicated in parentheses.
146
K. Miyasaka et al.r Journal of the Autonomic NerÕous System 73 (1998) 143–148
Fig. 3. The effect of intracerebroventricular administration of somatostatin on the inhibitory effects of CGRP Ž1.0 nmol.. The changes in protein secretion are significantly different Ž F s 7.99, P - 0.01.. ), significantly different from the corresponding value of somatostatinq vehicle; †, significantly different from the other corresponding values. The number of animals is five for somatostatinq vehicle, four for somatostatinq CGRP, and six for CGRP alone.
Intracerebroventricular injection of CGRP Ž0.1, 1.0 nmolr10 ml. inhibited pancreatic fluid and protein outputs ŽFig. 1A.. Although changes in fluid and protein outputs were essentially paralleled to each other, the changes in protein secretion were more apparent than those in fluid secretion, therefore, the results of protein output are shown in the figures. Protein secretion immediately decreased by injection of CGRP at 30 min and continued for 90 min ŽFig. 1A.. When analyzed by MANOVA with repeated measures including three treatments ŽCGRP; 0.1, 1.0 nmolr10 ml, and vehicle., the changes in protein secretion were significantly different with respect to doses of CGRP Ž F s 3.97, P - 0.05. and to time Ž F s 2.64, P - 0.04.. However, the changes in pancreatic fluid secretion were similar to those in pancreatic protein secretion, but not statistically significant Ž F s 1.03, P ) 0.3 for treatment, F s 1.00, P ) 0.4 for time.. The decremental changes during 2 h after intracerebroventricular injection of CGRP are shown in Fig. 1B. The inhibition of protein output produced by CGRP was dose-dependent. The decremental changes in fluid secretion during 2 h were not statistically significant Ž F s 3.27, P ) 0.07., although the mean values were lower in CGRP than in vehicle administration. Intravenous injection of CGRP slightly, but not significantly, decreased pancreatic secretion Ž F values with respect to time were 0.56, P ) 0.7 for protein and 0.40, P ) 0.8 for fluid secretion, n s 5.. Intravenous infusion of phentolamine abolished the inhibitory effect of intracerebroventricular injection of 1.0 nmol of CGRP, although the changes in fluid secretion were not statistically significant, as described above ŽFig. 2.. Infusion of phentolamine alone did not significantly affect pancreatic secretion ŽMasuda et al., 1995..
3.2. Effect of intracerebroÕentricular injection of somatostatin Intracerebroventricular injection of somatostatin Ž4.0 nmolr10 ml. 5 min before CGRP injection diminished the decrease in pancreatic responses to intracerebroventricular injection of CGRP Ž1.0 nmolr10 ml.. Fig. 3A shows the changes in protein secretion after 1.0 nmolr10 ml of CGRP with or without somatostatin. Administration of somatostatin with CGRP Žclosed triangles. diminished the inhibition produced by CGRP alone Žclosed circles., but the mean values of somatostatinq CGRP were still lower, but not statistically significant, than those treated with somatostatinq vehicle Žopen triangles.. The changes in fluid secretions were not statistically significant among treatments.
4. Discussion Administration of CGRP into the left lateral ventricle inhibited pancreatic fluid and protein secretion, although the inhibitory effect on fluid secretion was not statistically significant. The inhibitory effect was reversed by peripheral administration of the a-adrenergic receptor blocker, phentolamine. This observation is compatible with the previous report by Messmer et al. Ž1993.. Intravenous infusion of CGRP Ž1.0 nmol. did not affect pancreatic secretion significantly as observed previously ŽLi and Owyang, 1993b.. Therefore, it is confirmed that centrally administered CGRP stimulates sympathetic nerves, resulting in the inhibition of basal pancreatic secretion. The present study shows that intracerebroventricular injection of somatostatin diminishes the inhibitory effect
K. Miyasaka et al.r Journal of the Autonomic NerÕous System 73 (1998) 143–148
induced by intracerebroventricular administration of CGRP on pancreatic protein secretion. We have recently observed ŽMasuda et al., 1997. that somatostatin injected into the cerebral ventricle eliminates the stimulatory effect of centrally administered TRH, which produces vagal efferent nerve excitation. Thus, we have proposed that centrally administered somatostatin might prevent vagal efferent nerve excitation, probably via somatostatin receptors on the peptidergic neurons, because central somatostatin does not affect the stimulatory effect of peripheral administration of 2-deoxy-D-glucose, which has been shown to stimulates the vagal efferent nerve ŽRogers et al., 1995.. However, it is clear that the acting site of centrally administered CGRP is not on the vagal neurons but on sympathetic neurons because the inhibitory effect on pancreatic secretion is not affected by vagatomy ŽMessmer et al., 1993. and reversed by phentolamine ŽMessmer et al., 1993 and present study.. Therefore, it is suggested that centrally administered somatostatin may inhibit the sympathetic nerve excitation produced by centrally administered CGRP at the central site. In a previous study ŽMasuda et al., 1995., we have reported that centrally administered somatostatin inhibits only CCK-stimulated pancreatic secretion via sympathetic nerve excitation and that the inhibitory action is abolished by the intravenous administration of phentolamine. This inhibitory mechanism is similar to that of central CGRP observed in the present study. However, intracerebroventricular injection of somatostatin did not inhibit basal secretion ŽMasuda et al., 1995., while CGRP does. CCK has been shown to be a neurotransmitter of the vagal afferent nerve ŽRaybould et al., 1988; Moran et al., 1990; Rogers et al., 1995., and Li and Owyang Ž1993a. have reported that the excitation of the vagal afferent nerve results in pancreatic hypersecretion via vagal efferent nerve excitation. Somatostatin receptors are widely distributed in the CNS including the cortex, hypothalamus, hippocampus, striatum, medulla, and pons, and they are localized in the NTS and DVC ŽUhl et al., 1985.. The brain siteŽs. of the action of somatostatin was not determined in this study. However, as CGRP was injected into the left lateral ventricle Žnot into the fourth ventricle., it is suggested that CGRP may act at the hypothalamus, which is a center of autonomic nervous system, and inhibit sympathetic nerve excitation. This interpretation is consistent with previous neurohistochemical and electrochemical studies ŽFinley et al., 1981; Uhl et al., 1985; Jacquin et al., 1988.. Furthermore, as centrally administered somatostatin does not inhibit basal Žnon-stimulated. secretion, somatostatin may be effective only when some neural function is modified by such as TRH or CGRP. Intracerebroventricular injection of CGRP has been reported to inhibit gastric acid secretion ŽTache´ and Yang, 1990.. However, the basal protein and fluid secretions in conscious rats are primarily regulated by cholinergic tone and by circulating secretin, respectively ŽMiyasaka and
147
Green, 1983; Guan et al., 1991., and gastric acids are of less importance in pancreatic secretion ŽGreen, 1990.. Therefore, the inhibitory action of central CGRP on gastric acid secretion may be less relevant to the inhibition of pancreatic secretion in this case. In conclusion, our present results indicate that central somatostatin acts as an inhibitor to autonomic nerve excitation produced by peptidergic neuronal activity related to central TRH and CGRP.
Acknowledgements This study was supported in part by grants from the Ministry of Education, Science and Culture, the Life Science Foundation of Japan, Sandoz Foundation for Gerontological Research, Sasagawa Medical Research Foundation, and the Research Foundation of Pancreatic Disease of Japan.
References Finley, J.C.W., Maderdrut, J.L., Roger, L.J., Petrusz, P., 1981. The immunocytochemical localization of somatostatin-containing neurons in the rat central nervous system. Neuroscience 6, 2173–2192. Green, G.M., 1990. Role of gastric juice in feedback regulation of rat pancreatic secretion by luminal proteases. Pancreas 5, 445–451. Guan, D., Spannagel, A., Ohta, H., Nakano, I., Chey, W.Y., Green, G.M., 1991. Role of secretin in basal and fat-stimulated pancreatic secretion in conscious rats. Endocrinology 128, 979–982. Holzer, P., 1988. Local effector functions of capsaicin-sensitive sensory nerve endings: involvement of tachykinins, calcitonin gene-related peptide and other neuropeptides. Neuroscience 24, 739–768. Jacquin, T., Champagnat, J., Madamba, S., Denavit-Saubie, ´ M., Siggins, G.R., 1988. Somatostatin depresses excitability in neurons of the solitary tract complex through hyperpolarization and augmentation of I M , a non-inactivating voltage-dependent outward current blocked by muscarinic agonists. Proc. Natl. Acad. Sci. U.S.A. 85, 948–952. Kato, Y., Kanno, T., 1983. Thyrotropin-releasing hormone injected intracerebroventricularly in rats stimulates exocrine pancreatic secretion via the vagus nerve. Regul. Pept. 7, 347–356. Keller, P.J., Cohen, E., Neurath, H., 1958. The proteins of bovine pancreatic juice. J. Biol. Chem. 233, 344–349. Loewy, A.D., Haxhiu, M.A., 1993. CNS cell groups projecting to pancreatic parasympathetic preganglionic neurons. Brain Res. 620, 323–330. Li, Y., Owyang, C., 1993a. Vagal afferent pathway mediates physiological action of cholecystokinin on pancreatic enzyme secretion. J. Clin. Invest. 92, 418–424. Li, Y., Owyang, C., 1993b. Mechanism of action of calcitonin gene-related peptide in inhibiting pancreatic enzyme secretion in rats. Gastroenterology 105, 194–201. Masuda, M., Kanai, S., Miyasaka, K., Funakoshi, A., 1995. Somatostatin inhibits pancreatic exocrine secretion centrally via sympathetic nerves in conscious rats. J. Auton. Nerv. Syst. 56, 31–37. Masuda, M., Kanai, S., Miyasaka, K., 1997. Central somatostatin prevents vagal efferent nerve excitation produced by TRH but not by 2-deoxy-D-glucose. Am. J. Physiol. 272, G351–G356. Masuda, M., Kanai, S., Miyasaka, K., 1998. Inhibitory effect of central dopamine on basal pancreatic secretion in conscious rats. Am. J. Physiol. 274, G29–G34.
148
K. Miyasaka et al.r Journal of the Autonomic NerÕous System 73 (1998) 143–148
Messmer, B., Zimmerman, F.G., Lenz, H.J., 1993. Regulation of exocrine pancreatic secretion by cerebral TRH and CGRP: role of VIP, muscarinic, and adrenergic pathways. Am. J. Physiol. 264, G237–G242. Miyasaka, K., Green, G., 1983. The effect of atropine on rat basal pancreatic secretion during return or diversion of bile pancreatic juice. Proc. Soc. Exp. Biol. Med. l74, 187–192. Miyasaka, K., Kitani, K., Green, G., 1986. The sequential changes in pancreatic exocrine function after abdominal surgery in the rat. Pancreas 1, 347–353. Moran, T.H., Norgeren, R., Crosby, R.J., McHugh, P.R., 1990. Central and peripheral vagal transport of cholecystokinin binding sites occurs in afferent fibers. Brain Res. 526, 95–102. Nelson, M.T., Debas, H.T., Mulvihill, S.J., 1993. Vagal stimulation of rat exocrine pancreatic secretion occurs via multiple mediators. Gastroenterology 105, 221–228. Raybould, H.E., Gayton, R.J., Dockray, G.J., 1988. Mechanisms of action of peripherally administered cholecystokinin octapeptide on brain stem neurons in the rat. J. Neurosci. 8, 3018–3024.
Reisine, T., 1995. Somatostatin receptors. Am. J. Physiol. 269, G813–G820 Žeditorial review.. Renehan, W.E., Zhang, X., Beierwaltes, W.H., Fogel, R., 1995. Neurons in the dorsal motor nucleus of the vagus may integrate vagal and spinal information from the GI tract. Am. J. Physiol. 268, G780–G790. Rogers, R.C., McTigue, D.M., Hermann, G.E., 1995. Vagovagal reflex control of digestion: afferent modulation by neural and ‘endoneurocrine’ factors. Am. J. Physiol. 268, G1–G10. Roze, ´ C., La Tour, J., Chariot, J., Souchard, M., Vaille, C., Debray, C., 1977. Pancreatic stimulation by 2-deoxy-D-glucose in the rat: a study of neural and hormonal mechanisms. Gastroenterol. Clin. Biol. 1, 435–446. Tache, ´ Y., Yang, H., 1990. Brain regulation of gastric acid secretion by peptides. Ann. NY Acad. Sci. 597, 128–145. Uhl, G.R., Tran, V., Snyder, S.H., Martin, J.B., 1985. Somatostatin receptors: distribution in rat central nervous system and human frontal cortex. J. Comp. Neurol. 240, 288–304.
Central somatostatin diminished inhibitory action of central CGRP on pancreatic basal secretion in conscious rats Kyoko Miyasaka a
a,)
, Setsuko Kanai a , Masao Masuda a , Akihiro Funakoshi
b
Department of Clinical Physiology, Tokyo Metropolitan Institute of Gerontology, 35-2 Sakaecho, Itabashiku, Tokyo-1730015, Japan b DiÕision of Gastroenterology, National Kyushu Cancer Center, Fukuoka-8111347, Japan Received 3 June 1998; revised 3 July 1998; accepted 24 July 1998
Abstract We examined whether central somatostatin prevents an inhibitory effect of central calcitonin-gene related peptide ŽCGRP. on pancreatic secretion in conscious male Wistar rats Ž330–330 g.. Rats were prepared with separate cannulas for draining bile and pancreatic juice and with a duodenal cannula and an extrajugular vein cannula. In addition, another cannula was stereotactically implanted into the left lateral cerebral ventricle. Rats were placed in restraint cages and experiments were conducted 4 days after the operation without anesthesia. An injection of CGRP Ž0.1, 1.0 nmolr10 ml. into the left lateral cerebral ventricle Ži.c.v.. inhibited pancreatic secretion dose-dependently. To confirm the inhibitory effect of CGRP Ži.c.v.. was mediated via sympathetic nerves, phentolamine was injected intravenously Ži.v.. bolus Ž0.5 mg kgy1 . 0.5-h before CGRP Ži.c.v.., followed by continuous infusion of 0.2 mg kgy1 hy1. Phentolamine Ži.v.. reversed the inhibition produced by CGRP Ži.c.v... An injection of 4 nmolr10 ml somatostatin Ži.c.v.. 5 min prior to CGRP injection diminished the inhibitory effect of CGRP Ži.c.v... It is concluded that centrally administered somatostatin diminished the inhibitory action of CGRP Ži.c.v.. on pancreatic secretion, probably via inhibiting autonomic Žsympathetic. nerve excitation at the central site. q 1998 Elsevier Science B.V. All rights reserved. Keywords: Somatostatin; Intracerebroventricular administration; Calcitonin-gene related peptide; CGRP; Pancreatic secretion
1. Introduction As pancreatic exocrine secretion is controlled by both autonomic nerve and hormonal signals, the regulation of these signals concerning this secretion in the central nervous system ŽCNS. is therefore important ŽMessmer et al., 1993; Masuda et al., 1995, 1997, 1998.. The input from the abdominal cavity runs through vagal and spinal afferent fibers ŽRenehan et al., 1995., and the vagal afferent fibers project to interneurons in the nucleus of the solitary tract ŽNTS. ŽLoewy and Haxhiu, 1993.. Calcitonin-gene related peptide ŽCGRP. has been known to be one of major neurotransmitters of splanchnic afferent fibers, and CGRP-containing terminals are also present in the NTS ŽHolzer, 1988.. Concerning the neural output from the
) Corresponding author. Tel.: q81 813 39643241 ext. 3089; fax: q81 813 35794776; e-mail: [email protected]
CNS, the excitation of vagal efferent fibers originating from the dorsal vagal complex ŽDVC. causes pancreatic hypersecretion, while excitation of the sympathetic nerve inhibits pancreatic secretion ŽNelson et al., 1993; Rogers et al., 1995.. It has been reported ŽKato and Kanno, 1983; Messmer et al., 1993; Rogers et al., 1995; Masuda et al., 1997. that central administration of thyrotropin releasing hormone ŽTRH. produces vagal efferent nerve excitation, resulting in pancreatic hypersecretion, and also that centrally administered CGRP inhibits pancreatic secretion via sympathetic nerve excitation, because the inhibitory effect is not affected by vagotomy, but abolished by peripheral administration of phentolamine Ž a-adrenergic receptor blocker.. Somatostatin, originally identified from ovine hypothalamus as a physiological regulator of growth hormone release from the anterior pituitary ŽReisine, 1995.. Somatostatin and its receptors are widely distributed in the CNS, and five different somatostatin receptors were cloned and type 1 and 2 were distributed with highest levels in
0165-1838r98r$ - see front matter q 1998 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 5 - 1 8 3 8 Ž 9 8 . 0 0 1 3 0 - 1
144
K. Miyasaka et al.r Journal of the Autonomic NerÕous System 73 (1998) 143–148
hippocampus, hypothalamus, cortex, and amygdala ŽFinley et al., 1981; Uhl et al., 1985; Reisine, 1995.. Although several behavioral studies have suggested that somatostatin in the brain modulates locomotor activity and cognitive functions ŽReisine, 1995., the multiple functions of central somatostatin have not been fully elucidated. It has been shown that both intracerebroventricular injection of TRH and intravenous infusion of 2-deoxy-D-glucose Ž2DG. increase pancreatic secretion via excitation of the vagal efferent nerve because the stimulatory effects are completely eliminated by vagotomy ŽRoze ´ et al., 1977; Kato and Kanno, 1983; Tache´ and Yang, 1990.. We have found ŽMasuda et al., 1997. that somatostatin injected into the cerebral ventricle abolishes the stimulatory effect of centrally administered TRH, but not of peripheral administration of 2DG. Thus, we suggest that central somatostatin might affect autonomic nerve excitation produced by peptidergic ŽTRH. neurons. In the present study, we examined whether somatostatin injected into the cerebral ventricle prevents the inhibitory action of central CGRP on pancreatic secretion.
2. Materials and methods 2.1. Materials CGRP and somatostatin-14 were purchased from the Peptide Institute, Osaka, Japan. Phentolamine was purchased from Sigma, St. Louis, MO. The cannulae used in this study were Silastic Medical Grade Tubing ŽDow-Corning, Midland, MI.; 0.025 in. i.d., 0.037 in. o.d... 2.2. Animal preparation Male Wistar rats ŽNippon SLC, Shizuoka, Japan. weighing 300–328 g were used. The operative procedure has been described in detail previously ŽMiyasaka et al., 1986.. Briefly, after intraperitoneal anesthesia with sodium pentobarbital Ž15 mgr300 g body weight., a cannula was inserted into the right jugular vein. For intracerebroventricular injection of CGRP and somatostatin, a metal cannula Ža modification of a 23-gauge needle. was implanted stereotactically into the left lateral ventricle ŽAP—1.0 mm, L—1.3 mm, DV—4.0 mm. ŽMasuda et al., 1995, 1997, 1998.. A midline abdominal incision was then made, and a cannula was inserted into the common bile duct proximal to the ampulla of Vater. The common bile duct was ligated proximally to the pancreas near the liver, and a second cannula was inserted above the ligation below the bifurcation of the bile duct. Thus, pure bile and pure pancreatic juice were collected separately. A third cannula was inserted into the duodenum to return bile and pancreatic juice, its outlet tip located near the ampulla of Vater. All cannulae were initially brought into the abdominal cavity through a subcutaneous channel in the back near the tail.
After the operation, the rats were placed in modified Bollman-type restraint cages under conditions of controlled temperature Ž248C. and illumination Ž12-h lightrdark cycle starting at 5 AM., and given free access to food and water. Bile and pancreatic juice were continuously returned to the intestine via the duodenal cannula. Experiments were conducted on day 4 after the operation, after withholding food for 5 h ŽMiyasaka et al., 1986.. At the end of each experiment, rats were sacrificed, and black ink was injected into the lateral ventricle in the rats. Successful cannulation was verified by the visualization. 2.3. Measurement of pancreatic fluid and protein output Bile and pancreatic juice were collected separately for 30-min periods, and the volume of pancreatic juice was measured with a Hamilton syringe. Samples of 15 ml of pancreatic juice were used for determination of protein concentrations, and protein concentration in the pancreatic juice was determined from the optical density at 280 nm ŽKeller et al., 1958. of samples diluted 200-fold with 0.04 M Tris buffer, pH 7.8. The remainder was mixed with the bile and infused into the duodenum with a syringe pump ŽCompact Infusion Pump, Harvard Apparatus, Southnatick, MA. over the next 30 min. The pooled bile and pancreatic juice were infused during the first 30 min of the experiment. Bile and pancreatic juice were continuously returned to the intestine throughout the experimental period. 2.4. Effect and mechanism of intracerebroÕentricular administration of CGRP on pancreatic secretion CGRP was dissolved in 1% BSA. After a 90-min basal collection with return of bile and pancreatic juice, CGRP Ž0.1 or 1 nmolr10 ml. or vehicle was injected into the left lateral ventricle with a Hamilton microsyringe, and pancreatic secretions were collected for the following 2 h. To confirm the inhibitory mechanism of central CGRP via sympathetic Ž a-adrenergic. nerves, phentolamine was injected intravenously bolus Ž0.5 mg kgy1 . 30 min before intracerebroventricular injection of CGRP, followed by continuous infusion of 0.2 mg kgy1 hy1 . Since CGRP may leak from the cerebrospinal fluid into the systemic circulation and exert biological actions, the effects of CGRP Ž1 nmol mly1 . given intravenously on the stimulation of pancreatic secretion were examined. 2.5. Effect of intracerebroÕentricular injection of somatostatin A 4 nmolr10 ml dose of somatostatin was injected into the cerebral ventricle 5 min prior to CGRP injection Ž1 nmolr10 ml.. The changes in pancreatic secretion were monitored as described above. Based on intracerebroventricular and intravenous injection of somatostatin, it has been verified in a previous study ŽMasuda et al., 1995. that
K. Miyasaka et al.r Journal of the Autonomic NerÕous System 73 (1998) 143–148
145
peated measures followed by Scheffe’s multiple comparison test. A P - 0.05 was considered significant.
3. Results 3.1. Effect and mechanism of intracerebroÕentricular administration of CGRP on basal pancreatic secretion The changes in pancreatic fluid and protein outputs were monitored. Under the present experimental condition, intracerebroventricular injection of isotonic saline Ž10 ml, injected into the left lateral ventricle. did not affect either fluid or protein output significantly. The results of protein output are shown in Fig. 1A.
Fig. 1. Effects of intracerebroventricular administration of CGRP Ž1.0 nmol. on basal pancreatic protein secretion ŽA.. The results of the statistical analysis are shown in the text. An arrow indicates the injection of CGRP. ), significantly different from the values of vehicle injection. Decremental changes during 2 h after intracerebroventricular administration of CGRP Ž0.1, 1.0 nmol. ŽB.. Values are significantly different Ž F s 4.58, P - 0.03.. †, significantly different from the other two values. Values are means"SE. Number of rats is indicated in parentheses.
somatostatin alone injected intracerebroventricularly does not affect basal pancreatic fluid and protein secretion. Moreover, although this dose of somatostatin increased circulating plasma somatostatin levels from a basal level of 27.0 " 7.0 pM to 83.3 " 8.6 pM, the circulating level of somatostatin at 66.2 " 15.1 pM produced by intravenous infusion of 100 ng kgy1 of somatostatin did not affect pancreatic secretion ŽMasuda et al., 1995.. 2.6. Statistical analysis All results are expressed as means " SE. Results were analyzed by a One-way analysis of variance ŽANOVA. or by a multiple analysis of variance ŽMANOVA. with re-
Fig. 2. The decremental changes in protein ŽA. and fluid secretion ŽB. during the 2 h after intracerebroventricular administration of CGRP Ž1.0 nmol. with or without phentolamine. The levels of protein secretion are significantly different Ž F s 4.3, P - 0.04. but the levels of fluid secretion are not Ž F s 3.5, P ) 0.06.. †, significantly different from the other two values. Number of rats is indicated in parentheses.
146
K. Miyasaka et al.r Journal of the Autonomic NerÕous System 73 (1998) 143–148
Fig. 3. The effect of intracerebroventricular administration of somatostatin on the inhibitory effects of CGRP Ž1.0 nmol.. The changes in protein secretion are significantly different Ž F s 7.99, P - 0.01.. ), significantly different from the corresponding value of somatostatinq vehicle; †, significantly different from the other corresponding values. The number of animals is five for somatostatinq vehicle, four for somatostatinq CGRP, and six for CGRP alone.
Intracerebroventricular injection of CGRP Ž0.1, 1.0 nmolr10 ml. inhibited pancreatic fluid and protein outputs ŽFig. 1A.. Although changes in fluid and protein outputs were essentially paralleled to each other, the changes in protein secretion were more apparent than those in fluid secretion, therefore, the results of protein output are shown in the figures. Protein secretion immediately decreased by injection of CGRP at 30 min and continued for 90 min ŽFig. 1A.. When analyzed by MANOVA with repeated measures including three treatments ŽCGRP; 0.1, 1.0 nmolr10 ml, and vehicle., the changes in protein secretion were significantly different with respect to doses of CGRP Ž F s 3.97, P - 0.05. and to time Ž F s 2.64, P - 0.04.. However, the changes in pancreatic fluid secretion were similar to those in pancreatic protein secretion, but not statistically significant Ž F s 1.03, P ) 0.3 for treatment, F s 1.00, P ) 0.4 for time.. The decremental changes during 2 h after intracerebroventricular injection of CGRP are shown in Fig. 1B. The inhibition of protein output produced by CGRP was dose-dependent. The decremental changes in fluid secretion during 2 h were not statistically significant Ž F s 3.27, P ) 0.07., although the mean values were lower in CGRP than in vehicle administration. Intravenous injection of CGRP slightly, but not significantly, decreased pancreatic secretion Ž F values with respect to time were 0.56, P ) 0.7 for protein and 0.40, P ) 0.8 for fluid secretion, n s 5.. Intravenous infusion of phentolamine abolished the inhibitory effect of intracerebroventricular injection of 1.0 nmol of CGRP, although the changes in fluid secretion were not statistically significant, as described above ŽFig. 2.. Infusion of phentolamine alone did not significantly affect pancreatic secretion ŽMasuda et al., 1995..
3.2. Effect of intracerebroÕentricular injection of somatostatin Intracerebroventricular injection of somatostatin Ž4.0 nmolr10 ml. 5 min before CGRP injection diminished the decrease in pancreatic responses to intracerebroventricular injection of CGRP Ž1.0 nmolr10 ml.. Fig. 3A shows the changes in protein secretion after 1.0 nmolr10 ml of CGRP with or without somatostatin. Administration of somatostatin with CGRP Žclosed triangles. diminished the inhibition produced by CGRP alone Žclosed circles., but the mean values of somatostatinq CGRP were still lower, but not statistically significant, than those treated with somatostatinq vehicle Žopen triangles.. The changes in fluid secretions were not statistically significant among treatments.
4. Discussion Administration of CGRP into the left lateral ventricle inhibited pancreatic fluid and protein secretion, although the inhibitory effect on fluid secretion was not statistically significant. The inhibitory effect was reversed by peripheral administration of the a-adrenergic receptor blocker, phentolamine. This observation is compatible with the previous report by Messmer et al. Ž1993.. Intravenous infusion of CGRP Ž1.0 nmol. did not affect pancreatic secretion significantly as observed previously ŽLi and Owyang, 1993b.. Therefore, it is confirmed that centrally administered CGRP stimulates sympathetic nerves, resulting in the inhibition of basal pancreatic secretion. The present study shows that intracerebroventricular injection of somatostatin diminishes the inhibitory effect
K. Miyasaka et al.r Journal of the Autonomic NerÕous System 73 (1998) 143–148
induced by intracerebroventricular administration of CGRP on pancreatic protein secretion. We have recently observed ŽMasuda et al., 1997. that somatostatin injected into the cerebral ventricle eliminates the stimulatory effect of centrally administered TRH, which produces vagal efferent nerve excitation. Thus, we have proposed that centrally administered somatostatin might prevent vagal efferent nerve excitation, probably via somatostatin receptors on the peptidergic neurons, because central somatostatin does not affect the stimulatory effect of peripheral administration of 2-deoxy-D-glucose, which has been shown to stimulates the vagal efferent nerve ŽRogers et al., 1995.. However, it is clear that the acting site of centrally administered CGRP is not on the vagal neurons but on sympathetic neurons because the inhibitory effect on pancreatic secretion is not affected by vagatomy ŽMessmer et al., 1993. and reversed by phentolamine ŽMessmer et al., 1993 and present study.. Therefore, it is suggested that centrally administered somatostatin may inhibit the sympathetic nerve excitation produced by centrally administered CGRP at the central site. In a previous study ŽMasuda et al., 1995., we have reported that centrally administered somatostatin inhibits only CCK-stimulated pancreatic secretion via sympathetic nerve excitation and that the inhibitory action is abolished by the intravenous administration of phentolamine. This inhibitory mechanism is similar to that of central CGRP observed in the present study. However, intracerebroventricular injection of somatostatin did not inhibit basal secretion ŽMasuda et al., 1995., while CGRP does. CCK has been shown to be a neurotransmitter of the vagal afferent nerve ŽRaybould et al., 1988; Moran et al., 1990; Rogers et al., 1995., and Li and Owyang Ž1993a. have reported that the excitation of the vagal afferent nerve results in pancreatic hypersecretion via vagal efferent nerve excitation. Somatostatin receptors are widely distributed in the CNS including the cortex, hypothalamus, hippocampus, striatum, medulla, and pons, and they are localized in the NTS and DVC ŽUhl et al., 1985.. The brain siteŽs. of the action of somatostatin was not determined in this study. However, as CGRP was injected into the left lateral ventricle Žnot into the fourth ventricle., it is suggested that CGRP may act at the hypothalamus, which is a center of autonomic nervous system, and inhibit sympathetic nerve excitation. This interpretation is consistent with previous neurohistochemical and electrochemical studies ŽFinley et al., 1981; Uhl et al., 1985; Jacquin et al., 1988.. Furthermore, as centrally administered somatostatin does not inhibit basal Žnon-stimulated. secretion, somatostatin may be effective only when some neural function is modified by such as TRH or CGRP. Intracerebroventricular injection of CGRP has been reported to inhibit gastric acid secretion ŽTache´ and Yang, 1990.. However, the basal protein and fluid secretions in conscious rats are primarily regulated by cholinergic tone and by circulating secretin, respectively ŽMiyasaka and
147
Green, 1983; Guan et al., 1991., and gastric acids are of less importance in pancreatic secretion ŽGreen, 1990.. Therefore, the inhibitory action of central CGRP on gastric acid secretion may be less relevant to the inhibition of pancreatic secretion in this case. In conclusion, our present results indicate that central somatostatin acts as an inhibitor to autonomic nerve excitation produced by peptidergic neuronal activity related to central TRH and CGRP.
Acknowledgements This study was supported in part by grants from the Ministry of Education, Science and Culture, the Life Science Foundation of Japan, Sandoz Foundation for Gerontological Research, Sasagawa Medical Research Foundation, and the Research Foundation of Pancreatic Disease of Japan.
References Finley, J.C.W., Maderdrut, J.L., Roger, L.J., Petrusz, P., 1981. The immunocytochemical localization of somatostatin-containing neurons in the rat central nervous system. Neuroscience 6, 2173–2192. Green, G.M., 1990. Role of gastric juice in feedback regulation of rat pancreatic secretion by luminal proteases. Pancreas 5, 445–451. Guan, D., Spannagel, A., Ohta, H., Nakano, I., Chey, W.Y., Green, G.M., 1991. Role of secretin in basal and fat-stimulated pancreatic secretion in conscious rats. Endocrinology 128, 979–982. Holzer, P., 1988. Local effector functions of capsaicin-sensitive sensory nerve endings: involvement of tachykinins, calcitonin gene-related peptide and other neuropeptides. Neuroscience 24, 739–768. Jacquin, T., Champagnat, J., Madamba, S., Denavit-Saubie, ´ M., Siggins, G.R., 1988. Somatostatin depresses excitability in neurons of the solitary tract complex through hyperpolarization and augmentation of I M , a non-inactivating voltage-dependent outward current blocked by muscarinic agonists. Proc. Natl. Acad. Sci. U.S.A. 85, 948–952. Kato, Y., Kanno, T., 1983. Thyrotropin-releasing hormone injected intracerebroventricularly in rats stimulates exocrine pancreatic secretion via the vagus nerve. Regul. Pept. 7, 347–356. Keller, P.J., Cohen, E., Neurath, H., 1958. The proteins of bovine pancreatic juice. J. Biol. Chem. 233, 344–349. Loewy, A.D., Haxhiu, M.A., 1993. CNS cell groups projecting to pancreatic parasympathetic preganglionic neurons. Brain Res. 620, 323–330. Li, Y., Owyang, C., 1993a. Vagal afferent pathway mediates physiological action of cholecystokinin on pancreatic enzyme secretion. J. Clin. Invest. 92, 418–424. Li, Y., Owyang, C., 1993b. Mechanism of action of calcitonin gene-related peptide in inhibiting pancreatic enzyme secretion in rats. Gastroenterology 105, 194–201. Masuda, M., Kanai, S., Miyasaka, K., Funakoshi, A., 1995. Somatostatin inhibits pancreatic exocrine secretion centrally via sympathetic nerves in conscious rats. J. Auton. Nerv. Syst. 56, 31–37. Masuda, M., Kanai, S., Miyasaka, K., 1997. Central somatostatin prevents vagal efferent nerve excitation produced by TRH but not by 2-deoxy-D-glucose. Am. J. Physiol. 272, G351–G356. Masuda, M., Kanai, S., Miyasaka, K., 1998. Inhibitory effect of central dopamine on basal pancreatic secretion in conscious rats. Am. J. Physiol. 274, G29–G34.
148
K. Miyasaka et al.r Journal of the Autonomic NerÕous System 73 (1998) 143–148
Messmer, B., Zimmerman, F.G., Lenz, H.J., 1993. Regulation of exocrine pancreatic secretion by cerebral TRH and CGRP: role of VIP, muscarinic, and adrenergic pathways. Am. J. Physiol. 264, G237–G242. Miyasaka, K., Green, G., 1983. The effect of atropine on rat basal pancreatic secretion during return or diversion of bile pancreatic juice. Proc. Soc. Exp. Biol. Med. l74, 187–192. Miyasaka, K., Kitani, K., Green, G., 1986. The sequential changes in pancreatic exocrine function after abdominal surgery in the rat. Pancreas 1, 347–353. Moran, T.H., Norgeren, R., Crosby, R.J., McHugh, P.R., 1990. Central and peripheral vagal transport of cholecystokinin binding sites occurs in afferent fibers. Brain Res. 526, 95–102. Nelson, M.T., Debas, H.T., Mulvihill, S.J., 1993. Vagal stimulation of rat exocrine pancreatic secretion occurs via multiple mediators. Gastroenterology 105, 221–228. Raybould, H.E., Gayton, R.J., Dockray, G.J., 1988. Mechanisms of action of peripherally administered cholecystokinin octapeptide on brain stem neurons in the rat. J. Neurosci. 8, 3018–3024.
Reisine, T., 1995. Somatostatin receptors. Am. J. Physiol. 269, G813–G820 Žeditorial review.. Renehan, W.E., Zhang, X., Beierwaltes, W.H., Fogel, R., 1995. Neurons in the dorsal motor nucleus of the vagus may integrate vagal and spinal information from the GI tract. Am. J. Physiol. 268, G780–G790. Rogers, R.C., McTigue, D.M., Hermann, G.E., 1995. Vagovagal reflex control of digestion: afferent modulation by neural and ‘endoneurocrine’ factors. Am. J. Physiol. 268, G1–G10. Roze, ´ C., La Tour, J., Chariot, J., Souchard, M., Vaille, C., Debray, C., 1977. Pancreatic stimulation by 2-deoxy-D-glucose in the rat: a study of neural and hormonal mechanisms. Gastroenterol. Clin. Biol. 1, 435–446. Tache, ´ Y., Yang, H., 1990. Brain regulation of gastric acid secretion by peptides. Ann. NY Acad. Sci. 597, 128–145. Uhl, G.R., Tran, V., Snyder, S.H., Martin, J.B., 1985. Somatostatin receptors: distribution in rat central nervous system and human frontal cortex. J. Comp. Neurol. 240, 288–304.