SIMULTANEOUS MEASUREMENT OF THE PANCREATIC AND BILIARY RESPONSE TO CCK AND SECRETIN

Transcription

1 GASTROENTEROLOGY 70: , 1976 Copyright 1976 by The Williams & Wilkins Co. Vol. 70, No. 3 Printed in U.S.A. SIMULTANEOUS MEASUREMENT OF THE PANCREATIC AND BILIARY RESPONSE TO CCK AND SECRETIN Primate biliary physiology XIII BARRY N. GARDINER AND DONALD M. SMALL Department of Medicine, Boston University Medical Center, Boston, Massachusetts A primate model is described that permits separate but simultaneous sampling of both bile and pancreatic fluid without altering the normal enterohepatic circulation. Simultaneous biliary and pancreatic dose response curves are presented to contrast the production of bile with both pancreatic and the secretion of pancreatic protein and bicarbonate. The data show that in the cholecystectomized animal both cholecystokinin and secretin increase bile at doses that produce appropriate submaximal responses from the pancreas. Both these hormones, therefore, are physiologically important in regulating bile. Further, over a wide dose range (0.2 to 2.3 U per kg) secretin has the major effect on bile, pancreatic, and pancreatic bicarbonate secretion, whereas cholyecystokinin has the major effect on pancreatic protein secretion. A target organ for both cholecystokinin (CCK) and secretin is the pancreas. In the bile fistula animal these hormones also stimulate the production of bile. 1,2 However, a bile fistula interrupts the enterohepatic circulation (EHC) of bile salt and therefore interferes with normal bile salt secretion. Since a major proportion of bile normally depends on bile salt secretion, the true physiological response to these hormones cannot be documented in animals with complete bile fistulas. In fact, the choleretic properties of the gastrointestinal hormones have not been quantitatively measured with the EHC intact. Furthermore, simultaneous pancreatic and biliary dose response' curves to CCK or secretin have not been obtained. It is therefore not known if these hormones influence bile at doses that also stimulate the pancreas. Thus, the aim of these experiments was to determine simultaneously the effects of varying doses of these two hormones on pancreatic and hepatic secretion in animals with the EHC intact. To satisfy the requirements of these experiments, an animal model was developed that allowed separate sampling of both bile and pancreatic fluid, without sig- Received March 31, Accepted October 8, Address requests for reprints to D. M. Small, M. D., Department of Medicine, Boston University Medical Center, Boston, Massachusetts This work was supported by United States Public Health Service Grant AM Dr. Gardiner was supported by National Institutes of Health Grant GM Dr. Gardiner's present address is Letterman Army Institute of Research, Department of Surgery, Presidio of San Francisco, California The authors would like to acknowledge the help of Dr. R. N. Redinger and Dr. S. Strasberg for their help, and Ms. Irene Miller for secretarial assistance. 403 nificantly altering the normal EHC. We modified our standard primate preparation 3 so that bile was separated from pancreatic. Each secretion was diverted through a separate photoelectric stream splitter. Small aliquots of each secretion were continously sampled, and the remainder returned immediately to the duodenum. In this way, the simultaneous biliary and pancreatic outputs could be calculated, hourly or daily, without significantly interrupting the normal of either secretion. Simultaneous biliary and pancreatic dose response curves were obtained for each hormone, which constrast the production of bile to both pancreatic and the secretion of pancreatic protein and bicarbonate. Subsequent publications will document the effects of these hormones on the secretion of biliary electrolytes, lipids, and bile salts. Methods Experimental model. The basic techniques of preparing and maintaining the experimental model have been described in detail elsewhere'; however, significant modifications have' been incorporated to satisfy our specific requirements (fig. 1). Before surgery, female Rhesus monkeys (4.0 to 5.5 kg) were conditioned for 2 weeks in primate restraining chairs. At laparotomy the cystic duct was divided between ligatures, and the contents of the gallbladder were aspirated. The distal common bile duct was transected, both ends ligated, and an 8-Fr. silastic T tube (Extracorporeal Medical Specialties, Inc.) placed in the proximal common duct to create a biliary fistula. The pancreatic fistula was prepared from a segment of duodenum containing the pancreatic ampulla. Both ends of this segment were closed to create a small pouch into which a 10-Fr. T tube (Bardex, Inc.) was placed. An end to end duodenal anastomosis was then performed to reestablish gastrointestinal continuity. A lo-fr. latex T tube was inserted into the duodenum to permit return of biliary and pancreatic

2 404 GARDINER AND SMALL Vol. 70, No.3 A BILE FLOW BILE SALT CONC. PANCREATIC FLOW PROTEIN and HCO; CONC. FIG. 1. Experimental model for simultaneously sampling biliary and pancreatic. All of the bile secreted by the liver passes through an electronic stream splitter (A), which samples 5% of the total now, 95(7, being returned to the duodenum. now, secreted into a duodenal pouch, passes through an electronic stream splitter (B), which collects 5% of the total pancreatic now and returns 95% to the duodenum. The stomach is constantly aspirated to remove acid (C). For further explanation see text. secretions. A gastrostomy was performed using a 12 Fr. catheter (Bardex, Inc.) to prevent gastric acid from entering the duodenum and releasing endogenous secretin. All four tubes were brought through separate stab wounds in the anterior abdominal wall, and the abdomen was closed. After surgery both bile and pancreatic fluid were passed through separate photoelectric stream spliters' and set to sample continuously 5% of the bile and 5% of the pancreatic fluid and to return the remaining 95% of each secretion to the duodenum (fig. 1). Dowling et al. have established that 5% collections of bile represent an insignificant interruption of the EHC. 5 This model allows continuous and simultaneous measurements of both biliary and pancreatic outputs without significantly altering their normal into the duodenum. Experimental design. A 3-week recovery period after surgery was allowed for each monkey. The animals were studied only every other day and between experiments were maintained on chronic 5% sample collections of both bile and pancreatic secretions. The hormone to be tested and the dose used were randomized, and only one dose was infused on any given day. No monkeys were studied that failed to maintain a neutral duodenal ph during histamine infusion. Following an overnight fast the gastrostomy was opened and a continuous intravenous infusion of 0.9% NaCl was begun at 15 ml per hr (Havard Apparatus Co., Inc.). After a 2-hr control period either CCK or secretin (Gastrointestinal Hormone Laboratory, Karolinska Institute, Stockholm, Sweden) was infused in 0.9% NaCl at 15 ml per hr for an additional 2 hr. During the control and experimental periods 5% of the pancreatic and 5% of the bile were continuously sampled, and the remaining 95% of each secretion returned to the duodenum. Since only 5% of each secretion was collected the response to each dose was taken as 20 times the amount of bile and pancreatic fluid sampled during the last 60 min of each infusion. The bicarbonate concentration in pancreatic fluid was determined with an autoanalyser, and protein concentration was measured by the techniques of Lowry et al. The secretory rates of pancreatic bicarbonate and protein were calculated by multiplying their respective concentrations by the hourly rate of pancreatic. Linear regression analysis was performed using the actual secretion rates measured in response to the four lowest doses of CCK and secretin. Regression lines were calculated and, a log-dose response was considered established in the presence of a positive slope and a statistically significant correlation coefficient (P < 0.05). The hormones used in these studies were obtained from the Gastrointestinal Research Institute, Karolinska Institute, Stockholm, Sweden. Doses of CCK are expressed in Ivy dog units, and each 10 units of CCK contained clinical units of secretin activity. Results The mean response to each dose of hormone represents at least nine determinations in a total of 5 animals. Both biliary and pancreatic responses to increasing doses of each hormone were simultaneously measured and expressed graphically as per cent change from control on a log-dose scale ranging from 0.2 U per kg to 5.0 U per kg (figs. 2 and 3). The corresponding absolute values are shown in table 1. Secretin Bile. Secretin was an effective choleretic agent increasing bile to 185% above control (fig. 2). A maximum response to secretin was not obtained in these studies, even at 5 U per kg, and doses of 0.2 U per kg were associated with a significant choleretic response. response. In addition to stimulating bile, infusions of secretin simultaneously increased the of pancreatic fluid as well as bicaronate (HC0 3 -) secretion. No significant dose response relationship existed between secretin and pancreatic protein secretion. Reproducibility of response. The reproducibility in the response to a dose of 0.5 U secretin per kg per hr is shown in table 2. Three or more determinations were performed in 2 animals. CCK Bile. CCK was an effective choleretic increasing bile to 70% above control (fig. 3). The maximum response in bile occurred at 2.3 U per kg, whereas a dose of 5 U per kg exhibited strong inhibition. response. In addition to increasing bile, infusions of CCK simultaneously increased pancreatic protein secretion, with the maximum increase to 90% of control occuring at 2.3 U per kg (fig. 3). Like the biliary response for CCK, pancreatic protein secretion was actually inhibited by a dose of 5.0 U per kg. A small amount of secretin-like activity resulted from each infusion of CCK. Both pancreatic and HC0 3 - were mildly stimulated by CCK (fig. 3). Reproducibility of response. The responses to repeated doses of 1.00 CCK per kg per hr are shown in table 2. Three or more separate determinations at this dose were obtained in 2 animals. Discussion In the past few years, it has been repeatedly demonstrated that both CCK and secretin will stimulate bile

3 March 1976 SIMULTANEOUS RESPONSE OF LIVER AND PANCREAS 405 PERCENT CHANGE O! - I ~, ~ -' , r r ~ o i-l,r' ~ ~ - - ~ ~ - - ~ i O! - I, r ' -, ~ ~ , ~ - PANCREATIC PROTEIN SECREllON response occurring at 2.3 U per kg and an inhibitory effect at higher doses of CCK. This response could represent pure autoinhibition produced by CCK alone. This would be the case if CCK acts simultaneously on two separate receptors, as described by Airens and Simonis,8 one stimulating the target organ and the other inhibiting it. It is just as plausible however, that the inhibition seen with high doses of CCK represents an interaction between CCK and secretin or is due to some other hormone contaminating the CCK preparation. Secretin in the lowest dose tested increases bile as well as pancreatic and secretion of HC However, the pancreatic and biliary dose response curves to secretin are somewhat dissimilar. The maximum pancreatic response to secretin occurred at 2.3 U per kg, whereas doses as high as 5 U per kg continued to stimulate bile. Unlike CCK no inhibition occurred with high doses of secretin. The physiological dose range of secretin in these monkeys appears to lie between 0.2 U per kg and 1.0 U per kg, and in this range, secretin is an effective choleretic. The relative potency of the two hormones can be illustrated by plotting the per cent change in response from the control period for identical doses (fig. 4). If at 50 o I-If---! I CONTROL SECRETIN DOSE (U/kg)! I PERCENT CHANGE FIG. 2. The effect of secretin. The effect of different doses of secretin plotted on a log-dose scale are expressed as percentage of change from control. The points represent mean increase and the bars ± 2 SEM. The lines represent linear regression analysis for secretion rates at doses of 0.2, 0.5, 1.0, and 2.3 U per kg. A dose response was considered established in the presence of a positive slope and a statistically significant correlation coefficient. A significant dose response relationship was established in the dose range of 0.2 to 2.3 U per kg for bile f10w (P < 0.01), pancreatic f10w (P < 0.01), and pancreatic bicarbonate secretion (P < 0.01). No significant dose response curve was elicited for pancreatic protein secretion.. 1,2,7 However, these investigations have been performed in animals with total bile fistulas. Since a bile fistula interrupts the EHC, the effects of these hormones on the EHC have not been determined. Nor is it known what influence the EHC may have on the choleresis produced by these hormones. The model used in the present investigation provides a means of simultaneously measuring both biliary and pancreatic responses while maintaining the EHC intact. Both hormones produce a choleresis at doses that lead to appropriate submaximal responses from the pancreas (figs. 2 and 3). These results show that both CCK and secretin function in the physiologically intact animal to stimulate bile. The choleretic response to CCK is parallel to the secretion of protein from the pancreas, with a maximal FIG. 3. The effect of cholecystokinin. Captions are the same as in Figure 2. A significant dose-response curve was obtained for bile f10w (P < 0.01), pancreatic f10w (P < 0.01), pancreatic bicarbonate secretion (P < 0.01), and pancreatic protein secretion (P < 0.05).

4 406 GARDINER AND SMALL Vol. 70, No.3 TABLE 1. Biliary and pancreatic responses to secretin and cholecystokinin (CCK)a Dose Bile HCO, - secretion Protein secretion Ulkglhr mllhr ml/hr meqlhr mglhr Secretin ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± 7.7 CCK ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± 6.8 a Values represent the mean ± SEM. Linear regression analysis was performed for the four lowest doses of cholecystokinin and secretin, and a log-dose response was considered established in the presence of a positive slope and a statistically significant correlation coefficient (P < 0.05). The only parameter not exhibiting a significant log-dose response was the secretion of pancreatic protein in response to secretin. TABLE 2. The reproducibility in response to cholecystokinin (CCK) and secretin a Animal Bile CCK (1.0 U/kg/hr) protein ml/hr mllhr mglhr No No ~ f-- W a:: u w (f) 300 ~ w (f) Z o B> 200 ll:' ;,e DOSE (U/kg) 0.2 A 0.5 o 1.0 o 2.3 Animal Secretin (0.5 U/kg/hr) Bile HCO,- mllhr mllhr meqlhr No No a The reproducibility obtained with this experimental model is illustrated above. Bile, pancreatic, and the secretion of pancreatic HCO. - and protein are shown in individual animals for repeated doses of either cholecystokinin (1.0 U/kg/hr) or secretin (0.5 U/kg/hr). The variation in response within a given animal appears to be less than it is among different animals. the same dose, each hormone increases the response by an equal amount, this will produce a curve falling along a line at 45 0 The per cent change from control for secretin and CCK are plotted in figure 4 for bile, O ~ - ' r - - ~ - ' - - ~ o 200 % RESPONSE TO CCK FIG. 4. Relative potency of secretin and cholecystokinin (CCK). The percentage of response to CCK is plotted against the percentage of response to secretin at four doses: 0.2, 0.5, 1.0, 2.3, U per kg. The straight solid line at 45 would indicate an equal response to both hormones. Secretin has the major effect on pancreatic bicarbonate, pancreatic, and bile, whereas CCK has the major effect on pancreatic protein secretion. pancreatic, pancreatic HCO a - secretion, and pancreatic protein secretion. The points plotted refer to doses of 0.2, 0.5, 1.0, and 2.3 U per kg. The 5.0 U per kg dose has been omitted owing to the inhibitory effects of CCK at this high dose. Secretin has the major effect on pancreatic HCO a -, pancreatic, and bile, whereas CCK has the main effect on pancreatic protein secretion. The ratio of the increase in response owing to secretin to the increase in response owing to CCK is given in table 3. This gives an estimate of the potency of the effect of secretin in relation to an identical dose of CCK. For bile

5 March 1976 SIMULTANEOUS RESPONSE OF LIVER AND PANCREAS 407 TABLE 3. Relative potency of secretin to cholecystokinin (CCK) on bile, pancreatic, pancreatic HCO, - secretion, and pancreatic protein secretion as a function of dose Dose of hormone Ulkg/hr Bile Secretin/CCK Ratio of response HCO,- secretion CCK/se cretin b protein secretion Percentage of increase in response over control for a given dose of secretin divided by percentage of increase in response over control of same dose of CCK. Inverse of a., pancreatic, and pancreatic bicarbonate production, the lowest dose of the hormone shows much larger effect for secretin than for CCK; hence, the response ratio is quite high. At intermediate doses (0.2 to 2.3 U per kg) the ratio is lower, but still secretin is several times as potent as CCK. At the highest dose (5 U per kg) the ratios are again higher owing to the inhibitory effects of high doses of CCK. For pancreatic protein secretion, the ratio of increase in response owing to CCK to increase in response owing to secretin (CCK-secretin response ratio) is highest at the lowest dose and falls progressively as the.90se of hormone is increased (table 2). Since this model allows simultaneous collections of bile, pancreatic fluid, and gastric secretions, it will be useful in defining the biliary responses to endogenously released gastrointestinal hormones. The nature of various stimuli, (CCK, secretin, gastrin, or some combinations) can be documented by measuring the appropriate response from the end organs, the pancreas and stomach, and the response from the liver can be recorded simultaneously. The model also offers the advantage of being able to screen for activity of one hormone while infusing or stimulating the release of another. For example, infusions of secretin did not significantly influence secretion of pancreatic protein. The secretin used in these studies (Gastrointestinal Hormone Laboratory, Karolinska Institute, Stockholm Sweden) was, therefore, shown to be biologically pure, and endogenous CCK was not stimulated during the test infusions. However, infusions of CCK significantly increased both pancreatic and secretion of HCO 3" This could be due either to contamination of the CCK with secretin or a secretin-like substance, to endogenous release of secretin during the test infusions, or to an intrinsic secretin-like activity of the CCK polypeptide itself. This study has introduced a new model for studying biliary and pancreatic physiology. It has also established that exogenous CCK and secretin influence bile in the physiologically intact animal and do so at doses that produce sub maximal responses from the pancreas. The question remains whether these hormones merely stimulate secretion of water and electrolyes from the hepatocyte and/or bile caniliculus, or whether they also influence the EHC of bile salts and therefore affect the lipid composition of bile as well. REFERENCES 1. Jones RS, Grossman MI: Choleretic effects of secretin and histamine in the dog. Am J Physiol 217: , Jones RS, Grossman MI: Choleretic effects of cholecystokinin, gastrin II, and caerulein in the dog. Am J Physiol 219: , Dowling RH, Mack E, Picott, J, et al: Experimental model for the study ofthe enterohepatic circulation of bile in the Rhesus monkey. J Lab Clin Med 72: , Berger J: Fraction collector using ultrasonic technique. IEEE Trans Biomed Eng 21: Dowling RH, Mack E, Small DM: Effects of controlled interruption of the enterohepatic circulation of bile salts by biliary diversion and by ileal resection on bile salt secretion, synthesis and pool size in the Rhesus monkey. J Clin Invest 49: , Lowry OH, Rosebrough NJ, Farr AL, et al: Protein measurement with the folin phenol reagent. J Bioi Chern 193: , Preisig RG, Cooper HL, Wheeler HO: The relationship between taurocholate secretion rate and bile production in the unanesthetized dog during cholinergic blockade and during secretin administration. J Clin Invest 41: , Ariens EJ, Simonis AM : A molecular basis for drug action. J Pharm Pharmacol 16: , 1964