Effects of the Octapeptide of Cholecystokinin on Insulin and Glucagon Secretion in the Dog CAROLYN M. FRAME, MAYER B. DAVIDSON, AND RICHARD A. L. STURDEVANT Departments of Physiology and Medicine, UCLA School of Medicine, and Research and Medical Services, VA Wads worth Hospital Center, Los Angeles, California enhance insulin and glucagon responses to intravenous infusion of amino acids. The results suggest that insulin- and glucagon-releasing actions of porcine cholecystokinin preparations should not be attributed entirely to gastric inhibitory polypeptide or other impurities contained in these preparations since the synthetic active fragment of cholecystokinin alone increases insulin and glucagon concentrations in peripheral plasma. (Endocrinology 97: 549, 1975)
ABSTRACT. The effects of intravenous infusion of synthetic C-terminal octapeptide of cholecystokinin (OP-CCK) on concentrations of insulin and glucagon in peripheral venous plasma of conscious dogs were studied. Both hormones increased in response to 160 and 480 ng/kg/h of OP-CCK. The increases to 480 ng/kg/h were larger than those to 160 ng/kg/h. Peripheral venous concentrations of glucose and intestinal glucagon-like immunoreactivity were not altered by OP-CCK. OP-CCK, 160 ng/kg/h, did not
C
HOLECYSTOKININ (CCK) has been reported to stimulate insulin secretion in man (1,2,3), monkey (4), dog (5-9), rat (4,10,11), and mouse (12). In most of these species it has been shown to stimulate glucagon secretion as well (1,6-10). Recently, however, gastric inhibitory polypeptide (GIP), which was present in the impure CCK preparation used for these studies, was also found to stimulate insulin and glucagon release in vivo (13-15). When comparisons were made in humans and rats between matching doses of 10% pure CCK, highly purified CCK and GIP, most of the insulinogenic and all of the glucagonogenic effects of the 10% pure CCK could be accounted for by the GIP in it (13-15). GIP contamination, however, cannot explain reports that caerulein, a chemical analogue of CCK possessing many of its biological actions, stimulates glucagon and insulin release in the dog (16,17) and human (18,19). The present study was undertaken to help resolve these conflicting findings. The synthetic Cterminal octapeptide of CCK (OP-CCK) was selected as the stimulating agent because it has all of the gastro-intestinal actions of the whole molecule which have been investigated to date (20) and because it contains no contaminating peptides. Received November 25, 1974.
Materials and Methods Conscious mongrel dogs weighing 20 to 30 kg were studied after an overnight fast. The dogs were restrained by slings in an upright position. Three cannulas were inserted in leg veins, one for collecting blood samples and two for receiving infusions. When the effects of OP-CCK alone were studied, a 30 min basal period was followed by a 30 min test period and then a 30 min recovery period. During the basal and recovery periods, 0.9% saline was infused through both cannulas at the rate of 0.5 ml/min by a peristaltic pump. During the test period, OP-CCK (Lot #UTA-000H/TA kindly supplied by Dr. James Knill, Squibb Institute for Medical Research, Princeton, New Jersey) dissolved in normal saline replaced one of the saline infusions. The concentration of OPCCK was adjusted so that a dose of 160 or 480 ng/kg/h was delivered. (One mg of OP-CCK has a gallbladder contraction potency of 20,000 Ivy dog units and a pancreatic enzyme stimulation potency of 16,000 Ivy dog units in the anesthetized dog (20). Blood samples for glucose, glucagon and insulin determinations were obtained at 15 min intervals during the basal period, at 2.5, 5, 7.5, 10, 15, and 30 min after the start of the OP-CCK infusion, and at 15 min intervals during the recovery period. When the effects of OP-CCK during hyperaminoacidemia were studied, an amino acid solution was substituted for saline in one cannula at the end of the basal period. Fifteen minutes later, a blood sample was drawn and the OP-CCK infusion begun in the second cannula. The re-
549
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 04 August 2016. at 02:55 For personal use only. No other uses without permission. . All rights reserved.
550
FRAME, DAVIDSON AND STURDEVANT
mainder of the experiment was as before, the amino acid infusion being terminated at the same time as the OP-CCK infusion. The amino acids infused were L-glycine, L-serine, L-leucine, and L-threonine, combined in equimolar amounts in normal saline and brought to pH 7.4 with sodium hydroxide. The concentration was adjusted so tliat a dose of 0.04 mmol/kg-min of combined amino acids was delivered at an infusion rate of 2 ml/min. Four milliliter blood samples were drawn, divided into 2 ml aliquots, placed in iced 12 x 75 mm glass tubes containing either 0.1 ml of 2.4% EDTA and 0.1 ml of Trasylol (FBA Pharmaceuticals, New York, New York) or EDTA alone and centrifuged at 4 C. The resulting plasma was stored at - 2 0 C until analyzed. Plasma glucose was measured by the ferricyanide method of Hoffman (21) as modified by Technicon autoanalyzer methodology N2b. Plasma insulin was measured by a double antibody radioimmunoassay method (22) using pork insulin as standards, an antiserum to pork insulin raised in guinea pigs and iodinated ox insulin. Plasma glucagon was measured by radioimmunoassay (23) using pancreatic-specific antiserum 30K (kindly supplied by Dr. Roger Unger, The University of Texas Southwestern Medical School at Dallas, Dallas, Texas) at a final dilution of 1:40,000. Free and antibody-bound glucagon were separated with dextran-coated charcoal (24). Beef-pork recrystallized glucagon (a gift of Dr. Mary Root, The Lilly Research Laboratories, Indianapolis, Indiana) was used as a standard and for preparation of [125I]iodoglucagon by the chloraniine T method of Greenwood and Hunter (25). Total glucagon-like immunoreactivity (GLI), which is predominantly of intestinal origin (26), was measured as for glucagon except that a nonspecific glucagon antiserum (Antiserum 98J from Dr. Roger Unger) was used at a final dilution of 1:25,000. To minimize the effects of inter-animal variation in plasma hormone concentrations, hormone responses are expressed as changes in concentration from each animal's mean concentration during the basal period. Mean basal concentrations of each substance measured were not significantly different prior to each type of infusion. The mean (±SEM) basal concentrations for all experiments were: glucose 89 ± 2 mg/100 ml, insulin 8 ± 0.6 /i,U/ml, glucagon 93 ± 3 pg/ml, total GLI 554 ± 25 pg/ml. The statistical signif-
Endo • 1975 Vol 97 • No 3
icance of responses to treatments was determined by Student's t test for paired values applied to the difference between each animal's mean log concentration during the basal period and its mean log concentration during the period in which the response occurred. The insulin response period was the first 10 min of the test period. The glucagon response period was the entire 30 min of the test period. Logarithmic transformations of the data were performed to better meet the assumption of a Gaussian distribution of values required for parametric statistical analyses. Results The effects of a 30 min iv infusion of two doses of OP-CCK on plasma levels of gluclose, insulin, glucagon, and total GLI are shown in Fig. 1. OP-CCK at: 480 ng/kg/h caused a transient increase in plasma insulin during the first 10 min of the infusion which was statistically significant (F
ABSTRACT. The effects of intravenous infusion of synthetic C-terminal octapeptide of cholecystokinin (OP-CCK) on concentrations of insulin and glucagon in peripheral venous plasma of conscious dogs were studied. Both hormones increased in response to 160 and 480 ng/kg/h of OP-CCK. The increases to 480 ng/kg/h were larger than those to 160 ng/kg/h. Peripheral venous concentrations of glucose and intestinal glucagon-like immunoreactivity were not altered by OP-CCK. OP-CCK, 160 ng/kg/h, did not
C
HOLECYSTOKININ (CCK) has been reported to stimulate insulin secretion in man (1,2,3), monkey (4), dog (5-9), rat (4,10,11), and mouse (12). In most of these species it has been shown to stimulate glucagon secretion as well (1,6-10). Recently, however, gastric inhibitory polypeptide (GIP), which was present in the impure CCK preparation used for these studies, was also found to stimulate insulin and glucagon release in vivo (13-15). When comparisons were made in humans and rats between matching doses of 10% pure CCK, highly purified CCK and GIP, most of the insulinogenic and all of the glucagonogenic effects of the 10% pure CCK could be accounted for by the GIP in it (13-15). GIP contamination, however, cannot explain reports that caerulein, a chemical analogue of CCK possessing many of its biological actions, stimulates glucagon and insulin release in the dog (16,17) and human (18,19). The present study was undertaken to help resolve these conflicting findings. The synthetic Cterminal octapeptide of CCK (OP-CCK) was selected as the stimulating agent because it has all of the gastro-intestinal actions of the whole molecule which have been investigated to date (20) and because it contains no contaminating peptides. Received November 25, 1974.
Materials and Methods Conscious mongrel dogs weighing 20 to 30 kg were studied after an overnight fast. The dogs were restrained by slings in an upright position. Three cannulas were inserted in leg veins, one for collecting blood samples and two for receiving infusions. When the effects of OP-CCK alone were studied, a 30 min basal period was followed by a 30 min test period and then a 30 min recovery period. During the basal and recovery periods, 0.9% saline was infused through both cannulas at the rate of 0.5 ml/min by a peristaltic pump. During the test period, OP-CCK (Lot #UTA-000H/TA kindly supplied by Dr. James Knill, Squibb Institute for Medical Research, Princeton, New Jersey) dissolved in normal saline replaced one of the saline infusions. The concentration of OPCCK was adjusted so that a dose of 160 or 480 ng/kg/h was delivered. (One mg of OP-CCK has a gallbladder contraction potency of 20,000 Ivy dog units and a pancreatic enzyme stimulation potency of 16,000 Ivy dog units in the anesthetized dog (20). Blood samples for glucose, glucagon and insulin determinations were obtained at 15 min intervals during the basal period, at 2.5, 5, 7.5, 10, 15, and 30 min after the start of the OP-CCK infusion, and at 15 min intervals during the recovery period. When the effects of OP-CCK during hyperaminoacidemia were studied, an amino acid solution was substituted for saline in one cannula at the end of the basal period. Fifteen minutes later, a blood sample was drawn and the OP-CCK infusion begun in the second cannula. The re-
549
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 04 August 2016. at 02:55 For personal use only. No other uses without permission. . All rights reserved.
550
FRAME, DAVIDSON AND STURDEVANT
mainder of the experiment was as before, the amino acid infusion being terminated at the same time as the OP-CCK infusion. The amino acids infused were L-glycine, L-serine, L-leucine, and L-threonine, combined in equimolar amounts in normal saline and brought to pH 7.4 with sodium hydroxide. The concentration was adjusted so tliat a dose of 0.04 mmol/kg-min of combined amino acids was delivered at an infusion rate of 2 ml/min. Four milliliter blood samples were drawn, divided into 2 ml aliquots, placed in iced 12 x 75 mm glass tubes containing either 0.1 ml of 2.4% EDTA and 0.1 ml of Trasylol (FBA Pharmaceuticals, New York, New York) or EDTA alone and centrifuged at 4 C. The resulting plasma was stored at - 2 0 C until analyzed. Plasma glucose was measured by the ferricyanide method of Hoffman (21) as modified by Technicon autoanalyzer methodology N2b. Plasma insulin was measured by a double antibody radioimmunoassay method (22) using pork insulin as standards, an antiserum to pork insulin raised in guinea pigs and iodinated ox insulin. Plasma glucagon was measured by radioimmunoassay (23) using pancreatic-specific antiserum 30K (kindly supplied by Dr. Roger Unger, The University of Texas Southwestern Medical School at Dallas, Dallas, Texas) at a final dilution of 1:40,000. Free and antibody-bound glucagon were separated with dextran-coated charcoal (24). Beef-pork recrystallized glucagon (a gift of Dr. Mary Root, The Lilly Research Laboratories, Indianapolis, Indiana) was used as a standard and for preparation of [125I]iodoglucagon by the chloraniine T method of Greenwood and Hunter (25). Total glucagon-like immunoreactivity (GLI), which is predominantly of intestinal origin (26), was measured as for glucagon except that a nonspecific glucagon antiserum (Antiserum 98J from Dr. Roger Unger) was used at a final dilution of 1:25,000. To minimize the effects of inter-animal variation in plasma hormone concentrations, hormone responses are expressed as changes in concentration from each animal's mean concentration during the basal period. Mean basal concentrations of each substance measured were not significantly different prior to each type of infusion. The mean (±SEM) basal concentrations for all experiments were: glucose 89 ± 2 mg/100 ml, insulin 8 ± 0.6 /i,U/ml, glucagon 93 ± 3 pg/ml, total GLI 554 ± 25 pg/ml. The statistical signif-
Endo • 1975 Vol 97 • No 3
icance of responses to treatments was determined by Student's t test for paired values applied to the difference between each animal's mean log concentration during the basal period and its mean log concentration during the period in which the response occurred. The insulin response period was the first 10 min of the test period. The glucagon response period was the entire 30 min of the test period. Logarithmic transformations of the data were performed to better meet the assumption of a Gaussian distribution of values required for parametric statistical analyses. Results The effects of a 30 min iv infusion of two doses of OP-CCK on plasma levels of gluclose, insulin, glucagon, and total GLI are shown in Fig. 1. OP-CCK at: 480 ng/kg/h caused a transient increase in plasma insulin during the first 10 min of the infusion which was statistically significant (F
