INHIBITION OF GASTRIC EMPTYING IS A PHYSIOLOGICAL ACTION OF CHOLECYSTOKININ

Transcription

1 GASTROENTEROLOGY68: , 1975 Copyright 1975 by The Williams & Wilkins Co. Vol. 68, No. 5, Part 1 Printed in U.S.A. NHBTON OF GASTRC EMPTYNG S A PHYSOLOGCAL ACTON OF CHOLECYSTOKNN HALE T. DEBAS, M.D., MAR F AROOQ, M.D., AND MoRTON. GROSSMAN, M.D., PH.D. VA Wadsworth Hospital Center and UCLA School of Medicine, Los Angeles, California This study was designed to determine whether cholecystokinin (CCK) plays a physiological role in the inhibition of gastric emptying. Physiological conditions were simulated by giving CCK by continuous intravenous infusion rather than by bolus injection, by using doses known to be distinctly submaximal for pancreatic protein secretion and for gallbladder contraction, and by releasing endogenous CCK. The rate of gastric emptying was determined in 4 dogs with gastric fistulas by measuring the volume of fluid remaining in the stomach 1 min after instillation of 3 ml of.15 M NaCl. Rate of emptying was studied during intravenous infusion of saline (control) and of different doses of 98% pure CCK, commerically available 2% pure CCK, synthetic COOH-terminal octapeptide of CCK (OP-CCK), pentagastrin, and heptadecapeptide gastrin. The effect of endogenously released CCK was studied by measuring the rate of emptying of solutions in which different concentrations of tryptophan replaced equiosmolar amounts of NaCl. The D 5 's of 2% pure CCK (3 U kg- 1 hr- 1 ) and of OP-CCK (125 ng kg- 1 hr- 1 ) for inhibition of gastric emptying were about the same as their D 5 's for cholecystokinetic and pancreoyminic actions. By contrast, although both pentagastrin and heptadecapeptide gastrin inhibited gastric emptying, the doses required for this action were much higher than the D 5 's required for stimulation of gastric acid secretion. The effectiveness of OP-CCK indicates that inhibition of gastric emptying is attributable to CCK itself and not to an impurity in the CCK preparation. We have confirmed this directly by showing that pure CCK is a potent inhibitor of gastric emptying. Tryptophan also inhibited gastric emptying. n other dogs pancreatic protein secretion and gallbladder contraction were shown to be stimulated during the time tryptophan was inhibiting gastric emptying. This evidence supports the view that inhibition of gastric emptying is one of the physiological actions of CCK, but in the case of gastrin it must be regarded as a pharmacological action. Cholecystokinin (CCK) has been shown previously to inhibit gastric motor activity in both dog and man. 2-5 n all of these studies 1% pure CCK was used and ad- Received August 9, Accepted November 5, A part of this work was presented at the Annual Meeting of the American Gastroenterological Association, San Francisco, California, May 1973, and a preliminary report appeared in abstract form. 1 This study was supported by a Veterans Administration Senior Medical nvestigatorship (Dr. Grossman) and by grants from the National nstitute of 1211 ministered intravenously by bolus injection. Recently, however, Brown 6 isolated gastric inhibitory peptide (GP) from 1% pure CCK and showed that it is an inhibitor of both antral and fundic motor activ- Arthritis, Metabolism and Digestive Diseases (AM 8354) and the National Science Foundation (GB 437X). Dr. Debas held a fellowship from the Canadian Medical Research Council and has now returned to the University of British Columbia. The authors are grateful to Mr. John Washington for expert technical assistance.

2 1212 DEBAS ETAL. Vol. 68, No. 5, Part 1 ity. Therefore, these earlier studies failed to establish whether inhibition was caused by CCK itself or by another constituent of the crude extract and whether physiological amounts of CCK can produce the effect. This study had three aims: (1) to test whether CCK that is essentially free of GP and synthetic COOH-terminal octapeptide of CCK (OP-CCK) can inhibit gastric emptying; (2) to determine whether this action is physiological; and (3) to compare the potencies of CCK and gastrin for this action. Our findings support the view that inhibition of gastric emptying is one of the physiological actions of CCK. By contrast, gastrin is effective in inhibiting gastric emptying only in doses far above the D 5 for its primary action, stimulation of gastric acid secretion. Methods and Materials Animal preparations. Three types of surgical preparations were used in this study. A gastric fistula was constructed in each of 4 dogs by inserting a Thomas cannula into the most dependent portion of the body of the stomach. Two other dogs were prepared with both gastric fistulas and chronic pancreatic fistulas using a modified Herrera technique. 7 n an additional2 dogs, a similar technique was employed to construct chronic bile fistulas and gastric fistulas. n the latter dogs the duodeunum was transected just beyond the pylorus and just proximal to the main pancreatic duct so that a small duodenal pouch containing the ampulla of Vater could be made. Duodenoduodenostomy restored intestinal continuity. The cannula was placed with its lateral limb in the pouch and the intestinal limb in the duodenum distal to the entrance of the main pancreatic duct. The same principle for preparing a biliary fistula had been used in earlier studies. 8 Simulation of physiological conditions. Physiological conditions were simulated in three ways. 1. The peptides used in this study were given by continuous intravenous infusion rather than by bolus injection. 2. Doses of CCK and OP-CCK studied were distinctly submaximal for the pancreoyminic and cholecystokinetic actions of these peptides. n earlier studies we established that the 5 for pancreoyminic action was 3 U kg- 1 hr- 1 for 2% CCK and 13 ng kg-' hr- 1 for OP-CCK. n unpublished studies in dogs with the kind of biliary fistula described above, we found that the D 5 's of these peptides for gallbladder contraction did not differ significantly from their D 5 's for pancreoyminic action. The doses we selected for study were the 5 and one dose below (.5 5 ) and one dose above (2 5 ). We employed a similar range of doses for pentagastrin and synthetic human heptadecapeptide gastrin (SHG-17-1), whose o 5 's for stimulation of gastric acid secretion in dogs are about.8 f.lg kg- 1 hr- 1 and 16 pmoles kg-' hr- 1, respectively However, inhibition of gastric emptying was not seen in this dose range. and we resorted to doses 8 to 16 times the The action of endogenous CCK was studied. L-Tryptophan (TRP) was selected for this purpose because it is a potent releaser of CCK without having any significant effect on the release of gastrin or secretin. The rates of gastric emptying of meals of TRP of different molarities were determined in all the dogs. n the dogs with pancreatic and bile fistulas, pancreatic enyme response and gallbladder contraction (bilirubin output) were estimated simultaneously with the measurement of gastric emptying. n this way, the release of endogenous CCK could be bioassayed during studies of gastric emptying of tryptophan. The volume of pancreatic juice or bile secreted during the 1-min period when the tryptophan meal remained in the stomach was measured to the nearest.1 ml. Total protein concentration in the pancreatic juice was estimated by measuring absorbance at 28 nm, using an autoanalyer with a Beckman DB spectrophotometer and employing bovine serum albumin as standard. Bilirubin concentration in the bile was estimated colorimetrically by making a 1:1 dilution with water and reading absorbance at 45 nm using water as blank. Determination of rate of gastric emptying. The rate of gastric emptying was determined by measuring the volume of fluid remaining in the stomach 1 min after instillation via gastric fistula of 3 ml of.15 M NaCl containing phenol red (6 mg/liter) as indicator. The rate of emptying was determined with the dogs receiving only saline intravenously (control) and while different doses of 2% CCK were infused intravenously. nfusion of these peptides was started 2 min before the meal and continued during the 1-min period the meal remained in the stomach. The concentration of phenol red was determined in both the meal instilled and the fluid drained from the stomach at the end of the 1 -min period. A 1.-ml sam-

3 May 1975 CCK NHBTON OF GASTRC EMPTYNG 1213 ple and 2. ml of a solution of Na 2 PO. (27.5 g/ liter) were pipetted into a 1-ml volumetric flask, which was then filled to the mark with distilled water. After thorough mixing the OD of the solution was read at 55 nm against water blank in a Coleman Junior spectrophotometer. Rate of gastric emptying was calculated from the formula: Rate of emptying (ml/1 min) (V,P.) - (V,P,) (P, + P,)/2 where V, and V, are the volumes of fluid instilled and recovered, respectively; P, and P, are the concentrations of phenol red in the fluid instilled and recovered, respectively. The assumption is made that the concentration of phenol red in fluid leaving the stomach is the mean of the initial and final concentrations. n the studies on the effect of pentagastrin and synthetic human hepatadecapeptide gastrin (SHG-17-) on gastric emptying, it was necessary to use.1 M of NaHC 3 as the test meal in order to maintain the gastric ph above 5., thus obviating any possibility of secretin release from the duodenum. For release of endogenous CCK, meals of TRP solutions were prepared in which different concentrations of the amino acid replaced an equiosmolar amount of NaCl. Pure CCK was the kind gift of Professor Viktor Mutt of the Karolinska nstitute. Twenty per cent pure CCK was purchased from the Gastrointestinal Hormone Research Unit of the Karolinska nstitute in Stockholm. Synthetic OP-CCK was the kind gift of Dr. Miguel Ondetti of the Squibb nstitute for Medical Research, Princeton, New Jersey. Pentagastrin and SHG-17- were the kind gifts of Dr. J. S. Morley of mperial Chemical ndustries, Cheshire, England. L-Tryptophan, grade A, was purchased from Calbiochem, San Diego, California. Results Effects of CCK and OP-CCK on gastric emptying. The effectiveness of both pure and 2% CCK and OP-CCK on gastric empyting can be seen from table 1. Graded concentrations of these peptides caused graded inhibition of gastric emptying. The D 5 for this action can be estimated as 3 U kg- 1 hr- 1 for 2% CCK and 125 ng- 1 kg- 1 hr- 1 for OP-CCK (fig. 1). We have insufficient information to estimate the D 5 o for pure CCK. Rate of emptying of TRP meals. Meals of TRP of graded molarity caused graded inhibition of gastric emptying (table 1 and TABLE 1. Effect of intravenous infusion of 2% cholecystokinin (CCK) and synthetic COOH-terminal octapeptide cholecystokinin (OP-CCK) on gastric emptying of a saline meal (3 ml), and comparison of emptying of saline meal and equisomolar -tryptophan meals of various molarities" Rate of gastric Significance of Dose nhibition difference from control Test emptying substance Test Control mg/lomin % p Pure CCK.75 Ukg- 1 hr ± ± ± 16. < U kg- 1 hr ± ± ± 9. <.1 3.Ukg- 1 1\'l' 57 ± ± ± 7 <.1 2%CCK 1.5 U kg- 1 hr ± ± ± 4.1 <.1 3. U kg- 1 hr ± ± ± 6.3 <.1 6. U kg- 1 hr ± ± ± 5. <.1 OP-CCK 63 ng kg- 1 hr ± ± ± 1.6 < ng kg- 1 hr ± ± ± 7.1 <.1 25 ng kg- 1 hr ± ± ± 5.9 <.1 TRP 1 mmoles/liter 172 ± ± ± 4. <.1 2 mmoles/liter 12 ± ± ± 8. <.1 4 mmoles/liter 54± ± ± 7. <.1 "CCK, cholecystokinin; OP-CCK, synthetic C-terminal octapeptide cholecystokinin; TRP; L-tryptophan. Each number represents the mean of at least 2 experiments on each of 4 dogs. Statistical significance was calculated by the t-test for paired values.

4 1214 DEBAS ETAL. Vol. 68, No. 5, Part 1 fig. 1). nhibition of emptying by the 4 mm TRP meal was about the same as that produced by either 6 U kg- 1 hr- 1 of 2% CCK or 25 ng kg- 1 hr- 1 of OP-CCK. Simultaneous measurements of gastric emptying and gallbladder contraction. Figure 2 shows percentage of inhibition of gastric emptying caused by meals of TRP of different molarities and gallbladder contraction, as measured by bilirubin output from the bile fistulas during the 1-min period when gastric emptying was being measured. n these dogs the emptying rate was more markedly inhibited by TRP than in the dogs with gastric fistulas alone, such that even the lowest concentration of TRP caused near maximal inhibition. Graded concentrations of TRP produced graded degrees of gallbladder contraction. Simultaneous measurements of gastric emptying and pancreatic protein secretion. Again, in the dogs with gastric and pancreatic fistulas, graded concentrations of TRP caused graded inhibition of gastric emptying accompanied by graded pancreatic protein secretion (fig. 3). Effect of gastrin and pentagastrin on gastric emptying. Figure 4 shows the effect of intravenous infusion of graded doses of 1 C) 8 >- li: ::=; w 6 u a: f- (f) 4 Q f- ai 2,e B % CCK OP-CCK 6. 2."/oCCK U lojhr"' 25 OPCCK ngkg- 1 hr 4 TRP mmol/1 FG. 1. Mean ( ±SE) per cent inhibition of rate of gastric emptying of a saline meal caused by intravenous infusion of graded doses of 2% cholecystokinin (CCK) or synthetic COOH-terminal octapeptide cholecystokinin (OP-CCK) or by addition of tryptophan (TRP) to the saline meal in graded concentrations. Each point is the mean of 2 experiments on each of 4 dogs. SHG-17- and pentagastrin on emptying from the stomach of saline meals. nhibition of emptying comparable to that seen with 2% CCK and OP-CCK occurred only with doses of the gastrin peptides, which are near maximal for gastric acid secretion: 16 pmoles kg- 1 hr- 1 for SHG-17- and 8 1 GASTRC EMPTYNG -- T "' 2. c 8 r e 1.6 T/1 1 E GALL-BLADDER CONTRACTON Q f- ll. >- Cll :::> :: 4.8 iii :::>,e :: 2 ;;;!.4 "' ,.--...,------, BO 2 4 BO 2 4 mmol/liter TRYPTOPHAN mmol/liter FG. 2. Mean (±SE) per cent inhibition of rate of gastric emptying of a saline meal caused by addition of graded amounts of tryptophan to the meal (left) and mean ( ±SE) of bilirubin output from a bile fistula during the 1-min study period of gastric emptying (right). Each point is the mean of 2 experiments on each of 2 dogs. NHBTON OF GASTRC EMPTYNG 1 2 c E Q 8 16 r/ d 6 >- iii u 1/1 "' :::12 :;,;4 w " <n /1 2 4 u "' ;;; STMULATON OF PANCRfATC PROTEN r mmol/!iter mmol/liter TRYPTOPHAN FG. 3. Mean (±SE) per cent inhibition of rate of gastric emptying of a saline meal caused by addition of graded amounts of tryptophan to the meal (left) and mean (±SE) of protein output from a pancreatic fistula during the 1-min study period of gastric emptying (right). Each point is the mean of 2 experiments on each of 2 dogs.

5 May1975 CCK NHBTON OF GASTRC EMPTYNG 1215 to 16 pg kg- 1 hr- 1 for pentagastrin. The response to lower doses was erratic. n the lowest two doses studied, both SHG-17-1 and pentagastrin frequently accelerated gastric emptying, although the mean data do not differ from control. n figure 5, we compare gastrin and CCK with respect to their action on gastric emptying. nhibition of gastric emptying is plotted against multiples of the D 5 for the primary action of these hormones, assuming gallbladder contraction and pancreatic enyme secretion to be the primary action of CCK, and gastric acid secretion of gastrin. While distinctly submaximal doses (2 times D 5 ) of CCK and OP-CCK inhibited gastric emptying by 6 to 7%, comparable inhibition with gastrin and pentagastrin was seen only with doses near maximal for stimulation of acid secretion (8 to 1 times D 5 ). Discussion This study has shown that 2% CCK inhibits gastric emptying of a saline meal. Furthermore, the synthetic COOH-terminal octapeptide of CCK is as effective in this action as the 2% CCK. This indicates that the active fragment of the CCK molecule possesses the ability to inhibit gastric emptying and that the activity seen with impure CCK preparations cannot be as- (!) >= 1 ti:: 8 ::::!: w u ie (/l < g, 4 (!) i= iii 2_ SHG-17-1 B O B G PENTAGASTRN. ).19 kg-1 t"r S HG p mol kg-1 tr -1 FG. 4. Mean (±SE) per cent inhibition of rate of gastric emptying of a saline meal caused by intravenous infusion of graded doses of SHG-17-1 or pentagastrin. Each point is the mean of 2 experiments on each of 4 dogs. (!) 8 ll. ::::!: w 6 fl: 1-- (/l < ',- t1 6',. ' -f f' PENTA GASTRN, MULTPLES OF 5 FOR PRMARY ACTON FG. 5. Comparison of per cent inhibition of rate of gastric emptying of a saline meal caused by intravenous infusion of cholecystokinin (CCK), synthetic C-terminal octapeptide cholecystokinin (OP-CCK), SHG-17-1, or pentagstrin. The doses of each peptide are expressed as multiples of the D 5 for the primary action of that peptide. Each point represents the mean of 2 experiments on each of 4 dogs. cribed solely to contaminants. n a limited number of experiments we have also shown that 98% pure CCK is a potent inhibitor of gastric emptying. The main aim of this study was to see whether inhibition of gastric emptying can be considered one of the physiological actions of CCK. Each gastrointestinal hormone has now been shown to act on many target organs. For any individual hormone, some of these actions are clearly physiological, occurring during normal digestion and capable of being evoked with doses of the exogenous hormone that are equivalent to endogenously released levels. Some other actions are clearly pharmacological and are only elicited with doses of the hormone that cannot possibly be attained during normal alimentation. Rigid criteria cannot be set for deciding whether an action is physiological. deally, the following set of circumstances would enable us to make such a decision: (1) ability to release the hormone in question without releasing other hormones that would interact with it to influence the particular action in question; (2) a reliable method to accurately

6 1216 DEBAS ETAL. Vol. 68, No. 5, Part 1 measure the concentration and molecular form of the hormone in the circulation or, better still, at the receptor sites on the target organ; and (3) availability of the pure hormone so that it could be given by continuous, intravenous infusion and not by bolus injection to produce blood concentrations comparable to those achieved by endogenous release. Unfortunately, the first of these criteria is probably unattainable since there is no way to release one endogenous hormone without at the same time initiating other hormonal and neural events. Tryptophan was used because it does not release either gastrin or secretin. We do not know whether it releases other hormones. For CCK there is as yet no established, reliable radioimmunoassay system, and we resorted to bioassay in this study. Completely pure natural or synthetic CCK is not available, and we have used 2% CCK and synthetic OP-CCK. Within the limitations enumerated above, our study has shown that: (1) tryptophan, a known releaser of CCK, inhibited gastric emptying while at the same time it caused the gallbladder to contract and the pancreas to secrete enymes; and (2) the same degree of gastric inhibition could be produced by intravenous infusion of 2% CCK and OP-CCK in doses that were distinctly submaximal for the primary action of these peptides and that produced pancreatic and gallbladder responses comparable to those seen with tryptophan. We believe this constitutes evidence that inhibition of gastric emptying is a physiological action of CCK. Both gastrin 12 and pentagastrin have been shown to delay emptying of liquids from the stomach. Cooke et al. 14 found that doses of pentagastrin of 1 and 4 J.lg kg- 1 hr- 1 did inhibit gastric emptying. These findings differ from ours. The differences are due to the method of calculation of gastric emptying. They used for their calculation the volume of test meal passing the pylorus. We believe that the total volume of fluid passing the pylorus (i.e., test meal and volume of gastric secretion) is a better index. A measurement that takes into account only the meal and not added gastric secretion underestimates gastric emptying and therefore ascribes to gastrin too great an apparent inhibitory potency. The two methods of calculation give essentially the same results in tests where high rates of gastric secretion are not stimulated. Our findings indicate that, by contrast to CCK, much higher doses of gastrin are required to inhibit gastric emptying. While CCK and OP-CCK caused 6 to 7% inhibition in doses twice the D 5 for their primary action, with SHG-17-1 and pentagastrin 8 to 1 times their D 5 was required to cause a comparable degree of inhibition. Blood levels achieved by infusion of these doses are much higher than those that occur during normal digestion We conclude, therefore, that inhibition of gastric emptying of liquids is probably not an action of gastrin that occurs under physiological conditions. t is possible, however, that gastrin may inhibit emptying as a result of the acid secreted in response to it. Acid could act by releasing secretin, and secretin has been shown to inhibit gastric emptying, 5 16 although it is not known whether this is a physiological action. n addition, Cooke 17 has recently suggested that acid bathing of the first part of the duodenum may release a hormone other than secretin, or activate a neural reflex, or do both to inhibit gastric emptying. ntestinal acidification is also known to stimulc: :e pancreatic enyme secretion, 18 suggesting that CCK may be released. Fats are powerful releasers of CCK. 18 However, in addition to CCK, fats also release from the intestine another "enterogastrone" or "enterogastrones" one of which may be GP. t is likely that the other "enterogastrone" and perhaps neural reflexes, as well as CCK, mediate the action of fats. Since the pancreatic protein secretion in response to a meal is equivalent to more than a D 5 of CCK, 19 it seems likely that the amount of CCK released by a meal would have a significant effect in retarding gastric emptying. REFERENCES 1. De bas HT, Farooq, Grossman M: nhibition of gastric emptying is a physiological action of

7 May 1975 CCK NHBTON OF GASTRC EMPTYNG 1217 cholecystokin (abstr). Gastroenterology, 66:873, Johnson LP, Magee DF: Cholecystokinin-pancreoymin extracts and gastric motor inhibition. Surg Gynecol Obstet 121: , Johnson LP, Brown JC, Magee DF: Effect of secretin and cholecystokinin-pancreoymin extracts on gastric motility in man. Gut 7:52-57, Dinoso V Jr., Chey WY, Hendricks J, et a!.: ntestinal mucosal hormones and motor function on the stomach in man. J Appl Physiol 26: , Chey WY, Hitanant S, Hendricks J, eta!: Effect of secretin and cholecystokinin on gastric emptying and gastric secretion in man. Gastroenterology 58:82-827, Brown JC, Pederson RA, Jorpes J, eta!: Preparation of highly active enterogastrone. Can J Physiol Pharmacal 4 7: , Herrera E, Kemp DR, Tsukamoto M, eta!: A new cannula for the study of pancreatic function. J Appl Physiol 25:27-29, Nahrwold DL, Grossman M: Effect of cholecystectomy on bile flow and composition in response to food. Am J Surg 119:3-34, Debas HT, Grossman M: Pure cholecystokinin: pancreatic protein and bicarbonate response. Digestion 9: , Johnson LR, Grossman M: Analysis of inhibition of acid secretion by cholecystokinin in dogs. Am J Physiol 218:55-554, Walsh JH, Debas HT, Grossman M: Pure human big gastrin: mmunochemical properties, half life and acid stimulating action in dogs. J Clin nvest 54: , Hunt JN, Ramsbottom N: Effect of gastrin-11 on gastric emptying and secretion during a test meal. Br Med J 4: , Doois RR, Kelly KA: Effect of a gastrin pentapeptide on canine gastric emptying of liquids. Am J Physiol 221: , Cooke AR, Chvasta TE, Weisbrodt NW: Effect of pentagastrin on emptying and electrical and motor activity of the dog stomach. Am J Physiol 223: , Walsh JH, Csendes A, Grossman M: Effect of truncal vagotomy on gastrin release and Heidenhain pouch acid secretion in response to feeding in dogs. Gastroenterology 63:593-6, Chvasta TE, Cooke AR: Secretin-gastric emptying and motor activity: natural versus synthetic secretin. Proc Soc Exp Bioi Med 142: , Cooke AR: Duodenal acidification: role of 1st part of duodenum in gastric emptying and secretion in dogs. Gastroenterology 67:85-92, Meyer JH: Release of secretin and cholecystokinin. n Endocrinology of the Gut. Edited by WY Chey and FP Brooks. Thorofare, NJ, Charles B Slack, nc, 1974, p Grossman M, Konturek SJ: Gastric acid does drive pancreatic bicarbonate secretion. Scand J Gastroenterol 9:299-32, 1974