0013.7227/92/1302-0657$03,00/O Endocrinology Copyright 0 1992 by The Endocrine
Characterization Polyunsaturated EMMANUEL ONYE
AND
C. OPARA, E. AKWARI
Vol. 130, No. 2 Printed in U.S.A.
Society
of the Insulinotropic Fatty Acids* VAN
S. HUBBARD,
WARNER
Departments of Surgery and Medicine, Duke University Medical and the National Institute of Diabetes and Digestive and Kidney (V.S.H.), Bethesda, Maryland 20892
Potency
of
M. BURCH,
Center, Durham, North Carolina 27710; Diseases, National Institutes of Health
ABSTRACT. In this study we have assessed the individual abilities of the essential fatty acids, linoleic and linolenic acids, to release insulin and compared their insulinotropic potencies with those of the more established nutrient insulin secretagogues, glucose and arginine. In each experiment, a total of six islets microdissected from three mice were preperifused at the rate of 1 ml/min with Krebs-Ringer bicarbonate buffer, DH 7.4. containing 2% bovine albumin and 5.5 mM glucose (basal) with a continuous supply of 95% O,-5% CO1 at 37 C for 1 h. After collecting basal samples, the effects of 27.7 mM glucose, 20 mM arginine, 10 mM linoleic acid (l&2,&), and 5 mM linolenic acid (18:3,w3) were tested using a sandwich protocol that entails 20min alternating periods of stimulation with a secretagogue and a washout with basal perifusion. These nutrient concentrations were selected from initial experiments performed to characterize their dose-response effects on insulin secretion. Effluent samples were collected throughout each experiment for measurement of insulin by RIA. In one series of experiments, islets were challenged three times with 27.7 mM glucose, 10 mM linoleic acid, and 5 mM linolenic acid. In another set of experiments, islets were perifused with 20 mM arginine, 27.7 mM glucose, and 10 mM linoleic acid. All of these nutrients stimulated insulin release
in a dose-dependent manner. In comparing the insulinotropic potencies of these secretagogues, we assessed insulin secretion as the integrated areas under the curve during 20 min of perifusion with a given nutrient. Thus, the mean integrated area under the curve per 20 min above basal in the presence of 27.7 mM glucose was 6,516 + 1,435 pg, which was not significantly different from the value of 4,772 + 866 pg obtained during arginine perifusion. However, the area under the curve during 20 min above basal obtained in the presence of linoleate and linolenic acid (8,712 + 1,949 and 10,506 + 1,490 pg, respectively) were significantly different (P C 0.05) from those calculated during arginine and glucose perifusions. There was no statistically significant difference between the effects of these two fatty acids at the concentrations tested. In conclusion, our data suggest that linoleic acid and linolenic acid may be, at least in this murine islet preparation, as effective in stimulating insulin release as glucose and arginine, hitherto used to assess the abilities of nutrients to stimulate insulin secretion. However, it remains to be seen whether the efficacy of these polyunsaturated fatty acids in insulin release by murine islets will be obtained in experiments performed on human islets. (Endocrinology 130: 657-662, 1992)
P
REVIOUS work resulted in the establishment of the relationship between diets devoid of polyunsaturated fatty acids and the development of symptoms of essential fatty acid deficiency in experimental animals and humans (l-4). Presently, there has been a great deal of emphasis on the provision of these fatty acids because of their efficacy in a variety of situations, including total parenteral nutrition (5,6), cystic fibrosis (7-g), and other diseases (10). After an earlier report from experiments performed in viuo which suggested that polyunsaturated fatty acids may enhance insulin release (ll), a direct
effect of mixtures of these fatty acids on insulin secretion has been demonstrated (12, 13). However, the abilities of the individual fatty acids to stimulate insulin output from pancreatic islets remains to be determined. In the present study, therefore, we have evaluated the insulinotropic potency of linoleic and linolenic acids compared to that of the more established nutrient insulin secretagogues, glucose and arginine (14, 15). Materials
and Methods
Adult female CD-l albino mice were obtained from Charles River (Raleigh, NC). The procurement and method of use of these animals in this study were approved by the Duke University Medical Center review board for the welfare of animals. BSA (fraction V) free of fatty acids and insulin-like activity was purchased from Armour Pharmaceuticals (Kankakee, IL), and monoiodinated [9]insulin was obtained from New England Nuclear (Boston, MA). All nutrients, Trasylol, and other
Received September 3, 1991. Address all correspondence and requests for reprints to: Dr. Emmanuel C. Opara, Department of Surgery, Box 3076, Duke University Medical Center, Durham, North Carolina 27710. *This paper was presented in part at a minisymposium of the American Institute of Nutrition during the 75th Annual Meeting of the Federation of American Societies for Experimental Biology, Atlanta, GA, April 25, 1991. This work was supported in part by a grant from the Cystic Fibrosis Foundation (to E.C.O.).
657
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 01 May 2015. at 15:48 For personal use only. No other uses without permission. . All rights reserved.
658
EFFECTS
chemicals were purchased Louis, MO).
OF LINOLEIC
AND
from Sigma Chemical
LINOLENIC Co. (St.
Perifused islet preparation For each experiment, three animals fasting for at least 4 h, but allowed free access to water, were killed. The pancreas of each mouse was taken into a dissecting dish, and two large islets were isolated from the tail region by microdissection (16), with the aid of an operating microscope (model 30184, Carl Zeiss, Germany). All six islets were then pooled in a plastic flow-through perifusion miniature chamber and preperifused at a constant rate of 1 ml/min with a modified Krebs-Ringer bicarbonate buffer containing 5.5 mM glucose (basal) for 1 h at 37 C. This buffer was maintained at pH 7.4 by continuous gassing with a mixture of 95% O,-5% CO,. The Krebs-Ringer bicarbonate solution was comprised of 120 mM NaCl, 5 mM KCl, 1.1 mM MgC&, 2.5 mM CaCl*, and 25 mM NaHC03, and in addition contained 100 kallikrein inhibitor units/ml Trasylol and 2% albumin. This albumin concentration was sufficient to bring into solution the concentrations of the fatty acids examined. After the preperifusion, basal effluent samples were routinely collected for 20 min. To compare the effects of the different nutrients, the perifusion was performed for 20 min in the presence of 27.7 mM glucose, 20 mM arginine, 10 mM linoleic acid (18:2,w6) or 5 mM linolenic acid (18:3,w3). The nutrient perifusions were separated by 20-min periods of washout with basal glucose. To take into consideration the possible effect of priming in the second position during multiple perifusions of secretagogues on the same islet preparation (17, 18), the position of each nutrient in the perifusion protocol was randomly altered during these experiments. These nutrient concentrations were determined as maximally effective from dose-response experiments.
ACIDS
ON INSULIN
SECRETION
Endo. 1992 Voll30 -No 2
in the buffer solution. It was, therefore, first dissolved in methanol before incorporation into the perifusate to form an emulsified palmitate solution. The final methanol concentration in this fatty acid perifusate was 0.5%, which had no effect on basal insulin release. Data analysis Insulin secretion, assessed as the integrated area under the curve (AUC), was calculated using a modification (20) of the kinetic model (21). Statistical evaluation was performed using a one-way analysis of variance, and depending on the outcome of the analysis of variance, the Bonferroni method was used to assessthe significance of differences between groups. P < 0.05 was consideredsignificant. All values are expressedasthe mean+ SEM.
Results Dose-response characteristics for linoleic acid and glucose-stimulated insulin secretion The two protocols performed in assessing the doseresponse effects of nutrients yielded qualitatively similar results. Insulin secretion was stimulated in response to increasing concentrations of linoleate. When the data from the experiments were pooled and assessedas mean integrated insulin area under the curve (AUC/20 min) above basal, as shown in Fig. 1, insulin secretion stimulated by 1, 2, 5, and 10 mM linoleic acid was 487.5 + 128.9, 2085 + 131, 2842 -+ 417.7, and 7421.2 f 980 pg,
Dose-response effects of nutrients In experiments to characterize the dose-response effects of the different nutrients, two types of protocols were examined. In one series of experiments, after the preperifusion and taking of basal samples, the glucose concentration was raised, or a given concentration of arginine or fatty acid was added to the basal glucose buffer, and the perifusion was continued for 20 min, with sample collection. In another group of experiments, after the preperifusion, the perifusion was continued in 20-min cycles, with the addition of increasing concentrations of a given nutrient. Solutions were changed using a stopcock, and effluent perifusate was collected on ice at 2-min intervals and stored frozen until RIA for insulin (19), using pork insulin as standard.
8000
.-
T
8000
E z z!
4000
2000
Specificity of the insulinotropic effect of fatty acids A control experiment to check the specificity of the effects of the polyunsaturated fatty acids on insulin secretion was performed. In this experiment, the response of islets to perifusion with 10 mM of each of the linoleate and palmitate perifusions separated by 40 min of washout was examined. Linoleic acid was obtained in the form of oil and was readily solubilized in the presence of 2% albumin. In contrast, palmitate was purchased as a solid powder and could not be solubilized directly
0 1mM
2mM
Linoleic
5mM
1OmM
Acid
FIG. 1. Dose-response effect of linoleate on insulin secretion, assessed as AUC/SO min. In separate experiments, islets were perifused with different concentrations of the fatty acid immediately after preperifusion. The data were pooled and assessed as the mean integrated AUC -t SEM (n = 5).
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 01 May 2015. at 15:48 For personal use only. No other uses without permission. . All rights reserved.
EFFECTS
OF LINOLEIC
AND
LINOLENIC
respectively. Although insulin AUC/20 min secreted in the presence of 1 mM linoleic acid was not significantly different from that obtained under basal glucose conditions, all insulin release values at the other concentrations of fatty acid were statistically significantly different from basal with at least P < 0.05. However, a continuously elevated rate of insulin secretion was only obtained in the presence of 10 mM linoleate. A continuously elevated output of insulin was also observed in the presence of 5 mM linolenic acid (data not shown). These two concentrations were, therefore, chosen to perform experiments comparing the insulinotropic potencies of the nutrients. Insulin output from islets was stimulated with increments in the ambient glucose concentrations. The mean integrated insulin AUC/SO min increased from a mean basal of 2526 + 155 to 3924 k 626 and 7487 f 685 pg (P < 0.05) when the glucose concentration was raised to 11.1 and 22.2 mM, respectively. No statistically significant increase in insulin secretion was observed when the islets were perifused with higher glucose concentrations (27.7 and 33.3 mM), although 27.7 InM glucose was used in .experiments to compare the potencies of the secretagogues.
ACIDS
ON INSULIN
Evaluation of the insulinotropic secretagogues (Figs. 3 and 4)
and linoleate on
5.5mM
12000 , g
I
potencies of nutrient
ARG 5.5mM
GLU
In a separate set of control experiments, shown in Fig. 2, the addition of 10 mM palmitate to basal glucose perifusate had no effect on basal insulin output, as the basal mean integrated insulin AUC/20 min values were 3351 k 311 pg without and 3128 fi 233 pg with palmitate. In contrast, the addition of linoleate to the basal glucose perifusate enhanced insulin output to 6669 + 708 pg (P < 0.0001) above basal (Fig. 2).
659
Figure 3 is a diagrammatic representation of the perifusion protocols used to compare the insulinotropic potencies of the different nutrients in a series of nine experiments in which the position of each nutrient in the perifusion profile was randomly altered. These two separate protocols, shown in Fig. 3, were used in these experiments to avoid exposure of islets to 27.7 mM glucose after a preceding polyunsaturated fatty acid challenge. The mean data were assessed as integrated insulin AUC/BO min above basal, and as shown in Fig. 4, there was no statistically significant difference between the islet responses to 20 mM arginine and 27.7 mM glucose. However, 10 mM linoleate was more potent than 20 mM arginine in stimulating insulin release (P < 0.05), although the value of this response was not significantly different from that observed in the presence of 27.7 mM glucose. The stimulatory effect of 5 mM linolenic acid on insulin secretion observed was more potent than those of 27.7 mM glucose (P < 0.05) and 20 mM arginine (P < O.Ol), but was not statistically different from that of linoleic acid (Fig. 4). The arginine concentration (20 PUFA
Comparison of the effects of palmitate insulin secretion (Fig. 2)
SECRETION
5.5mM
PUFA 5.5mM
ARG 5.5mM
5.5mM glucose PUFA
5.5mM
5.5mM glucose
FIG. 3. Experimental
protocols used for comparing insulinotropic potencies of nutrients. The data shown in Fig. 4 were generated using these two protocols in which islets underwent a 20-min washout with basal perifusion (5.5 mM glucose) before another secretagogue perifusion. The protocols involved the perifusion of the same islet preparation with three different nutrients. The order of perifusion shown in the diagram was randomly switched during these experiments. PUFA, Polyunsaturated fatty acid; ARG, arginine; GLU, glucose.
10000
Ez 6 8000
10000
E -E 8000 3 6000
.-E 6000 8 3 4000
2
4000
Glucose 5.5mM Glucose
+lOmM Palmitate
l lOmM Linoleate
FIG. 2. Comparison of the effects of palmitate and linoleate on insulin secretion. Islets were perifused with equimolar concentrations of palmitate and linoleate, separated by a washout with 5.5 mM glucose perifusate. Insulin secretion was assessed as the mean integrated area under the curve/20 min for six experiments.
27.7mM
Arginine 20mM
Linoleic Linolenic Acid 1OmM Acid 5mM
FIG. 4. Effect of nutrient secretagogues on insulin secretion. In a series of nine separate experiments, the insulin response to maximally effective concentrations of glucose, arginine, linoleic acid, and linolenic acid were compared. The data were assessed as the mean integrated AUC/ 20 min and compared as follows: 1 us. 4, 2 us. 3, P < 0.05; 2 us. 4, P < 0.01.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 01 May 2015. at 15:48 For personal use only. No other uses without permission. . All rights reserved.
660
EFFECTS
OF LINOLEIC
AND
LINOLENIC
mM) used in this study was shown in a recent report (18) to be maximally stimulatory of insulin release.
Discussion We have clearly shown in the present study that individually, each of the polyunsaturated fatty acids used in our preceding report (12) is capable of stimulating insulin secretion to at least the same extent as glucose and arginine, which are presently used as the standards for evaluating the insulinotropic potencies of various nutrients (14, 15, 22). It is of interest that 5 mM linolenic acid and 10 mM linoleic acid were equipotent in stimulating insulin release. The insulinotropic abilities of these polyunsaturated fatty acids may be dependent upon their enhanced entry into the P-cell mitochondria (23-25), since it has been suggested that the oxidation of fatty acids may be involved in their stimulatory effect on insulin secretion (12, 26, 27). It is, therefore, not surprising that palmitate failed to stimulate insulin secretion, since long chain saturated fatty acids have only a limited ability to cross the inner mitochondrial membrane, as their entry has to be stimulated by carnitine (28). It has to be emphasized that the insulinotropic effects of polyunsaturated fatty acids and arginine were obtained in the presence of a basal (5.5 mM) glucose concentration. Attempts to perifuse the islets with the fatty acids alone (in the absence of basal glucose) resulted in poor islet survival and insulin response. Therefore, the presence of basal glucose in the fatty acid perifusate was essential to their stimulatory effect on insulin secretion. The link between polyunsaturated fatty acid oxidation and insulin release, which has recently been described (12, 13, 18), is consistent with one hypothesis on the mechanism of insulin secretion. The oxidation of these fatty acids would generate abundant ATP, leading to the closure of K+ channels on the P-cell plasma membrane. The resulting decrease in K+ permeability causes membrane depolarization, with activation of voltage-dependent Ca’+ channels. An increase in Ca2+ influx ensues, which raises the cytosolic concentration of free Ca2+ and ultimately triggers insulin release (29). However, there are other mechanisms by which polyunsaturated fatty acids may stimulate insulin release. It has been shown that glucose may stimulate insulin secretion via a mechanism involving the mediation of arachidonic acid, a metabolite of linoleic acid, as a second messenger (30, 31). It is, therefore, conceivable that the elongation and desaturation of linoleic acid to arachidonate would result in a stimulatory effect on insulin secretion. Furthermore, it has recently been reported that some intermediates of the tricarboxylic acid cycle (32) as well as malonyl-coenzyme-A and long chain acyl-coenzyme-A (33) may serve as metabolic coupling factors in signal transduction when islets are
ACIDS
ON INSULIN
SECRETION
Endo. 1992 Vol 130. No 2
stimulated with high glucose or glucose combined with other fuels. Although not all products of the Krebs cycle are capable of stimulating insulin release, it has been shown that esters of succinic acid are insulinotropic (32). Consequently, nutrients whose metabolism would generate this intermediate may release insulin via this pathway. Since the metabolism of these essential fatty acids would generate this and other products, such as malonlycoenzyme-A, it is certainly possible that these alternative mechanisms may be involved in the stimulatory effects of these fatty acids on insulin secretion. It should be pointed out that the insulin response to nutrients of murine islets, as apparent from this and previous studies (34-36), is monophasic and diminished in magnitude compared to that of islets from rats and humans. The reason for this diminished monophasic insulin response is unclear, but may be related to the fact that pancreatic islets in the mouse are exposed to higher basal blood glucose levels (37,38). It is conceivable that the chronic exposure of murine islets to such elevated basal glucose levels could impair the influence of these nutrients on insulin secretion (39-43). It would be appropriate to evaluate the insulinotropic potencies of these polyunsaturated fatty acids on islets with the classic biphasic insulin response to nutrients. However, it is encouraging that mixtures of these essential fatty acids have been shown to be potent insulin secretagogues in both man and rat (11,13). Although, the fatty acids that we have examined are mostly of benefit in hyperalimentation (5), in somewhat unphysiological situations they occur in the normal human diet (11) and could contribute postprandially to the stimulatory effect of a mixed meal on insulin secretion. In the presence of high concentrations of fatty acids, an artefactual increase in insulin secretion could be caused by a detergent effect of the fatty acids on pancreatic P-cells, but when we measured the insulin concentrations of the fatty acid-treated islets using an extraction procedure for peptides (44), it was found that less than 10% of the total insulin content of the islets was released during these experiments. Furthermore, a high concentration (10 mM) of palmitate failed to stimulate insulin output during these experiments. These observations would argue against a fatty acid-induced toxicity that would cause the islets to lyse and, thus, release most of their hormone contents. In addition, a recent report showed that human plasma fatty acid variations and what was previously thought unphysiological fatty acid levels occur in relationship to dietary intake (45). Besides, plasma levels of free fatty acids may not be an accurate index of their biologically effective concentrations, since the presence of lipoprotein lipase on the endothelial surface of blood capillaries, causing rapid hydrolysis of triglycerides (46), would increase the free acid concentrations to which the cells are
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 01 May 2015. at 15:48 For personal use only. No other uses without permission. . All rights reserved.
EFFECTS
OF LINOLEIC
AND
LINOLENIC
exposed. We recently observed that during multiple glucose perifusions on the same islet preparation, a possible priming effect occurred during a second challenge subsequent to the initial exposure (18). To take into account the possible influence of this phenomenon on the data from a protocol involving multiple challenges of the same islet preparation with different nutrient secretagogues, the positions of the various nutrients in the perifusion profile were randomly altered. It was necessary to have two versions of the perifusion protocol, so as to avoid a desensitization effect of a preceding polyunsaturated fatty acid challenge on islet response to glucose stimulation, which has been recently described (13,X3). It was suggested in these reports that products of polyunsaturated fatty acid metabolism may impair glucose metabolism by inhibiting glycolysis and, consequently, abolish glucose-stimulated insulin secretion. This phenomenon of polyunsaturated fatty acid-induced desensitization of islets to glucose response, therefore, prevented an additive stimulatory effect of glucose and these fatty acids on insulin secretion. In conclusion, our data suggest that linoleic acid and linolenic acid may be at least as potent, if not moreso, in releasing insulin in isolated murine islets as glucose and arginine, which were hitherto used as the standards in assessing the abilities of nutrients to stimulate the release of insulin. Acknowledgments The authors would like to thank Spencer Bridges, Lula Copeland, and Eleanor Matthews for expert technical help, and Pamela McAuley for superb editorial assistance. References
5.
6. 7. 8.
9.
10.
Burr GO, Burr MM 1930 On the nature and role of the fatty acids essential to nutrition. J Biol Chem 86587-621 Holman RT 1964 Nutritional and metabolic interrelationships between fattv acids. Fed Proc 23:1062-1067 Holman RT-1986 Essential fatty acids, prostaglandins and leukotrienes. Prog Lipid Res 25:19-47 Kinsella JE, Broughton KS, Whelan JW 1990 Dietary unsaturated fatty acids: interactions and possible needs in relation to eicosanoid synthesis. J Nutr Biochem 1:123-141 Meguid MM, Kurzer M, Hayashi RJ, Akahoshi MP 1989 Shortterm effects of fat emulsion on serum lipids in postoperative patients. JPEN 13:77-80 Rosner M. Grant JP 1987 Intravenous liuid emulsion. Nutr Clin Pratt 2:96-107 Hubbard VS 1983 What is the association of essential fatty acid status with cvstic fibrosis? Eur J Pediatr 141:68-70 Hubbard VS; McKenna MC 1987 Absorption of safflower oil and structured lipid preparation in patients with cystic fibrosis. Lipids 22:424-428 Parsons HG, O’Loughlin EV, Forbes D, Cooper D, Gall DG 1988 Supplemental calories improve essential fatty acid deficiency in cystic fibrosis patients. Pediatr Res 24:253-256 Broughton KS, Whelan J, Hardardottir I, Kinsella JE 1991 Effect of increasing the dietary (n-3) to (n-6) polyunsaturated fatty acid
ACIDS
11.
12.
13.
14.
15.
16. 17.
18.
19. 20.
21.
22. 23. 24.
25.
26.
27 28 29. 30.
31.
32.
33.
34.
ON INSULIN
SECRETION
661
ratio on murine liver and peritoneal cell fatty acids and eicosanoid formation. J Nutr 121:155-164 Lardinois CK, Starich GH, Mazzaferri EL, DeLett A 1987 Polyunsaturated fatty acids augment insulin secretion. J Am Co11 Nutr 6:507-515 Opara EC, Burch WM, Hubbard VS, Akwari OE 1990 Enhancement of endocrine pancreatic secretions by essential fatty acids. J Surg Res 48329-332 Sako Y, Grill VE 1990 A 48-hour lipid infusion in the rat timedependently inhibits glucose-induced insulin secretion and beta cell oxidation through a process likely coupled to fatty acid oxidation. Endocrinologv 127:1580-1589 Gerich JE, Charles-MA, Grodsky GM 1974 Characterization of the effects of arginine and glucose on glucagon and insulin release from the perfused rat pancreas. J Clin Invest 54:833-841 van Haeften TW, Voetberg GA, Gerich JE, van der Veen E 1989 Dose-response characteristics for arginine stimulated insulin secretion in man and influence of hyperglycemia. J Clin Endocrinol Metab 69:1059-1064 Opara EC, Atwater I, GO VLW 1988 Characterization and control of pulsatile secretion of insulin and glucagon. Pancreas 3:484-487 Turk J, Wolf BA, McDaniel ML 1987 The role of phospholipidderived mediators including arachidonic acid, its metabolite and inositol triphosphate and of intracellular Ca++ in glucose induced secretion by pancreatic islets. Prog Lipid Res 26:125-181 Opara EC, Hubbard VS, Burch WM, Akwari OE 1991 Homologous desensitization of pancreatic beta cells to glucose response by polyunsaturated fatty acids. J Nutr Biochem 2:424-429 Herbert V, Lau KS, Gottlieb GW, Blecher S 1965 Coated charcoal immunoassay of insulin. J Clin Endocrinol Metab 25:1375-1384 Morishima T, Pye S, Polonsky KS, Radziuk J 1986 The measurement and validation of the non-steady state rates of C-peptide appearance in the dog. Diabetologia 29:440-446 Eaton RF, Allen RC, Schade DS, Erickson KM, Standefer J 1980 Prehepatic insulin production in man: kinetic analysis using peripheral connecting peptide behavior. J Clin Endocrinol Metab 51:520-528 Malaisse WJ, Malaisse-Lagae F 1968 Stimulation of insulin secretion by noncarbohydrate metabolites. J Lab Clin Med 72:438-448 Lynn WS, Brown RH 1959 Oxidation and activation of unsaturated fatty acids. Arch Bichem Biophys 81:353-362 Jones PH. Pencharz PB. Clandinin MT 1985 Whole bodv oxidation of dietary’fatty acids: implications for energy utilization: Am J Clin Nutr 42:769-777 Leyton J, Drury PJ, Crawford MA 1987 Differential oxidation of saturated and unsaturated fatty acids in uiuo in the rat. Br J Nutr 57:383-393 Malaisse WJ, Malaisse-Lagae F, Sener A, Hellerstrom C 1985 Participation of endogenous fatty acids in the secretory activitv of the pancreatic B-cell.Biochem J 227:995-1002 Malaisse WJ 1989 Dual role of liuids in the stimulus-secretion coupling for insulin release. Bioche-m Sot Trans 17:59-66 Lehninger AL 1975 Biochemistry, chapt 20. Worth, New York, pp 543-547 Henquin JC 1987 Regulation of insulin release by ionic and electrical events in B-cells. Horm Res 27:168-178 Turk J, Colca JR, Kotagal N, McDaniel ML 1984 Arachidonic acid metabolism in isolated pancreatic islets. II. The effects of glucose and of inhibitors of arachidonic acid metabolism on insulin secretion and metabolite synthesis. Biochim Biophys Acta 794:125-136 Walsh MF, Pek SB 1984 Possible role of endogenous arachidonic acid metabolites in stimulated release of insulin and glucagon from the isolated perfused rat pancreas. Diabetes 33:929-936 McDonald MJ, Fahien LA, Mertz RJ, Rana RS 1989 Effect of esters of succinic acid and other citric acid cycle intermediates on insulin release and inositol phosphate formation by pancreatic islets. Arch Biochem Biophys 269:400-406 Liang Y, Matschinsky FM 1991 Content of CoA-esters in perifused rat islets stimulated by glucose and other fuels. Diabetes 40: 327-333 Scott AM, Atwater I, Rojas E 1981 A method for the simultaneous measurement of insulin release and b-cell membrane potential in
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 01 May 2015. at 15:48 For personal use only. No other uses without permission. . All rights reserved.
662
EFFECTS
OF LINOLEIC
AND
LINOLENIC
single mouse islets of Langerhans. Diabetologia 21:470-475 35. Atwater I, Carroll PB, Li MX 1989 Electrophysiology of the pancreatic B-cell. In: Draznin B (ed) Insulin Secretion. Liss, pp 49-68 36. deMegue1 R, Tamagawa T, Schmeer W, Nenquin M, Henquin JC 1987 Effect of acute sodium omission on insulin release, ionic flux, and membrane potential in mouse pancreatic B-cells. Biochim Biophys Acta 969:198-207 37. LeMarchand-Brustel Y, Balloti R, van Obberghen E 1985 Insulin receptor tyrosine kinase is defective in skeletal muscle of insulinresistant obese mice. Nature 315:676-679 38. Lebrun P, Atwater I 1985 Chaotic and irregular bursting electrical activity in mouse pancreatic B-cells. Biophys J 48:529-531 39. Leahy JL, Cooper HE, Deal DA, Weir GC 1986 Chronic hyperglycemia is associated with impaired glucose influence on insulin secretion. J Clin Invest 77:908-915 40. Robertson RP 1989 Perspectives in diabetes: type II diabetes, glucose “non-sense,” and islet desensitization. Diabetes 38: 1501-1505
ACIDS
ON
INSULIN
SECRETION
Endo. 1992 Voll30. No 2
41. Bolaffi JL, Bruno L, Heldt A, Grodsky GM 1988 Characteristics of desensitization of insulin secretion in fully in vitro systems. Endocrinology 122:1801-1809 42. Zawalich WS, Zawalich KC, Shulman GI, Rossetti L 1990 Chronic in uiuo hyperglycemia impairs phosphoinositide hydrolysis and insulin release in isolated perifused rat islets. Endocrinology 126:253-260 43. Sako Y, Grill VE 1990 Coupling of B-cell desensitization by hyperglycemia to excessive stimulation and circulating insulin in glucoseinfused rats. Diabetes 39:1580-1583 44. Koch TR, Carney JA, Go VLW 1987 Distribution and quantitation of gut neuropeptides in normal intestine and inflammatory bowel diseases. Dig Dis Sci 32:369-376 45. Lopes SM, Trimbo SL, Mascioli EA, Blackburn GL 1991 Human plasma fatty acid variations and how they are related to dietary intake. Am J Clin Nutr 53:628-637 46. Have1 RJ 1982 Symposium on lipid disorders: approach to the patient with hyperlipidemia. Med Clin North Am 66:319-333
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 01 May 2015. at 15:48 For personal use only. No other uses without permission. . All rights reserved.
Characterization Polyunsaturated EMMANUEL ONYE
AND
C. OPARA, E. AKWARI
Vol. 130, No. 2 Printed in U.S.A.
Society
of the Insulinotropic Fatty Acids* VAN
S. HUBBARD,
WARNER
Departments of Surgery and Medicine, Duke University Medical and the National Institute of Diabetes and Digestive and Kidney (V.S.H.), Bethesda, Maryland 20892
Potency
of
M. BURCH,
Center, Durham, North Carolina 27710; Diseases, National Institutes of Health
ABSTRACT. In this study we have assessed the individual abilities of the essential fatty acids, linoleic and linolenic acids, to release insulin and compared their insulinotropic potencies with those of the more established nutrient insulin secretagogues, glucose and arginine. In each experiment, a total of six islets microdissected from three mice were preperifused at the rate of 1 ml/min with Krebs-Ringer bicarbonate buffer, DH 7.4. containing 2% bovine albumin and 5.5 mM glucose (basal) with a continuous supply of 95% O,-5% CO1 at 37 C for 1 h. After collecting basal samples, the effects of 27.7 mM glucose, 20 mM arginine, 10 mM linoleic acid (l&2,&), and 5 mM linolenic acid (18:3,w3) were tested using a sandwich protocol that entails 20min alternating periods of stimulation with a secretagogue and a washout with basal perifusion. These nutrient concentrations were selected from initial experiments performed to characterize their dose-response effects on insulin secretion. Effluent samples were collected throughout each experiment for measurement of insulin by RIA. In one series of experiments, islets were challenged three times with 27.7 mM glucose, 10 mM linoleic acid, and 5 mM linolenic acid. In another set of experiments, islets were perifused with 20 mM arginine, 27.7 mM glucose, and 10 mM linoleic acid. All of these nutrients stimulated insulin release
in a dose-dependent manner. In comparing the insulinotropic potencies of these secretagogues, we assessed insulin secretion as the integrated areas under the curve during 20 min of perifusion with a given nutrient. Thus, the mean integrated area under the curve per 20 min above basal in the presence of 27.7 mM glucose was 6,516 + 1,435 pg, which was not significantly different from the value of 4,772 + 866 pg obtained during arginine perifusion. However, the area under the curve during 20 min above basal obtained in the presence of linoleate and linolenic acid (8,712 + 1,949 and 10,506 + 1,490 pg, respectively) were significantly different (P C 0.05) from those calculated during arginine and glucose perifusions. There was no statistically significant difference between the effects of these two fatty acids at the concentrations tested. In conclusion, our data suggest that linoleic acid and linolenic acid may be, at least in this murine islet preparation, as effective in stimulating insulin release as glucose and arginine, hitherto used to assess the abilities of nutrients to stimulate insulin secretion. However, it remains to be seen whether the efficacy of these polyunsaturated fatty acids in insulin release by murine islets will be obtained in experiments performed on human islets. (Endocrinology 130: 657-662, 1992)
P
REVIOUS work resulted in the establishment of the relationship between diets devoid of polyunsaturated fatty acids and the development of symptoms of essential fatty acid deficiency in experimental animals and humans (l-4). Presently, there has been a great deal of emphasis on the provision of these fatty acids because of their efficacy in a variety of situations, including total parenteral nutrition (5,6), cystic fibrosis (7-g), and other diseases (10). After an earlier report from experiments performed in viuo which suggested that polyunsaturated fatty acids may enhance insulin release (ll), a direct
effect of mixtures of these fatty acids on insulin secretion has been demonstrated (12, 13). However, the abilities of the individual fatty acids to stimulate insulin output from pancreatic islets remains to be determined. In the present study, therefore, we have evaluated the insulinotropic potency of linoleic and linolenic acids compared to that of the more established nutrient insulin secretagogues, glucose and arginine (14, 15). Materials
and Methods
Adult female CD-l albino mice were obtained from Charles River (Raleigh, NC). The procurement and method of use of these animals in this study were approved by the Duke University Medical Center review board for the welfare of animals. BSA (fraction V) free of fatty acids and insulin-like activity was purchased from Armour Pharmaceuticals (Kankakee, IL), and monoiodinated [9]insulin was obtained from New England Nuclear (Boston, MA). All nutrients, Trasylol, and other
Received September 3, 1991. Address all correspondence and requests for reprints to: Dr. Emmanuel C. Opara, Department of Surgery, Box 3076, Duke University Medical Center, Durham, North Carolina 27710. *This paper was presented in part at a minisymposium of the American Institute of Nutrition during the 75th Annual Meeting of the Federation of American Societies for Experimental Biology, Atlanta, GA, April 25, 1991. This work was supported in part by a grant from the Cystic Fibrosis Foundation (to E.C.O.).
657
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 01 May 2015. at 15:48 For personal use only. No other uses without permission. . All rights reserved.
658
EFFECTS
chemicals were purchased Louis, MO).
OF LINOLEIC
AND
from Sigma Chemical
LINOLENIC Co. (St.
Perifused islet preparation For each experiment, three animals fasting for at least 4 h, but allowed free access to water, were killed. The pancreas of each mouse was taken into a dissecting dish, and two large islets were isolated from the tail region by microdissection (16), with the aid of an operating microscope (model 30184, Carl Zeiss, Germany). All six islets were then pooled in a plastic flow-through perifusion miniature chamber and preperifused at a constant rate of 1 ml/min with a modified Krebs-Ringer bicarbonate buffer containing 5.5 mM glucose (basal) for 1 h at 37 C. This buffer was maintained at pH 7.4 by continuous gassing with a mixture of 95% O,-5% CO,. The Krebs-Ringer bicarbonate solution was comprised of 120 mM NaCl, 5 mM KCl, 1.1 mM MgC&, 2.5 mM CaCl*, and 25 mM NaHC03, and in addition contained 100 kallikrein inhibitor units/ml Trasylol and 2% albumin. This albumin concentration was sufficient to bring into solution the concentrations of the fatty acids examined. After the preperifusion, basal effluent samples were routinely collected for 20 min. To compare the effects of the different nutrients, the perifusion was performed for 20 min in the presence of 27.7 mM glucose, 20 mM arginine, 10 mM linoleic acid (18:2,w6) or 5 mM linolenic acid (18:3,w3). The nutrient perifusions were separated by 20-min periods of washout with basal glucose. To take into consideration the possible effect of priming in the second position during multiple perifusions of secretagogues on the same islet preparation (17, 18), the position of each nutrient in the perifusion protocol was randomly altered during these experiments. These nutrient concentrations were determined as maximally effective from dose-response experiments.
ACIDS
ON INSULIN
SECRETION
Endo. 1992 Voll30 -No 2
in the buffer solution. It was, therefore, first dissolved in methanol before incorporation into the perifusate to form an emulsified palmitate solution. The final methanol concentration in this fatty acid perifusate was 0.5%, which had no effect on basal insulin release. Data analysis Insulin secretion, assessed as the integrated area under the curve (AUC), was calculated using a modification (20) of the kinetic model (21). Statistical evaluation was performed using a one-way analysis of variance, and depending on the outcome of the analysis of variance, the Bonferroni method was used to assessthe significance of differences between groups. P < 0.05 was consideredsignificant. All values are expressedasthe mean+ SEM.
Results Dose-response characteristics for linoleic acid and glucose-stimulated insulin secretion The two protocols performed in assessing the doseresponse effects of nutrients yielded qualitatively similar results. Insulin secretion was stimulated in response to increasing concentrations of linoleate. When the data from the experiments were pooled and assessedas mean integrated insulin area under the curve (AUC/20 min) above basal, as shown in Fig. 1, insulin secretion stimulated by 1, 2, 5, and 10 mM linoleic acid was 487.5 + 128.9, 2085 + 131, 2842 -+ 417.7, and 7421.2 f 980 pg,
Dose-response effects of nutrients In experiments to characterize the dose-response effects of the different nutrients, two types of protocols were examined. In one series of experiments, after the preperifusion and taking of basal samples, the glucose concentration was raised, or a given concentration of arginine or fatty acid was added to the basal glucose buffer, and the perifusion was continued for 20 min, with sample collection. In another group of experiments, after the preperifusion, the perifusion was continued in 20-min cycles, with the addition of increasing concentrations of a given nutrient. Solutions were changed using a stopcock, and effluent perifusate was collected on ice at 2-min intervals and stored frozen until RIA for insulin (19), using pork insulin as standard.
8000
.-
T
8000
E z z!
4000
2000
Specificity of the insulinotropic effect of fatty acids A control experiment to check the specificity of the effects of the polyunsaturated fatty acids on insulin secretion was performed. In this experiment, the response of islets to perifusion with 10 mM of each of the linoleate and palmitate perifusions separated by 40 min of washout was examined. Linoleic acid was obtained in the form of oil and was readily solubilized in the presence of 2% albumin. In contrast, palmitate was purchased as a solid powder and could not be solubilized directly
0 1mM
2mM
Linoleic
5mM
1OmM
Acid
FIG. 1. Dose-response effect of linoleate on insulin secretion, assessed as AUC/SO min. In separate experiments, islets were perifused with different concentrations of the fatty acid immediately after preperifusion. The data were pooled and assessed as the mean integrated AUC -t SEM (n = 5).
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 01 May 2015. at 15:48 For personal use only. No other uses without permission. . All rights reserved.
EFFECTS
OF LINOLEIC
AND
LINOLENIC
respectively. Although insulin AUC/20 min secreted in the presence of 1 mM linoleic acid was not significantly different from that obtained under basal glucose conditions, all insulin release values at the other concentrations of fatty acid were statistically significantly different from basal with at least P < 0.05. However, a continuously elevated rate of insulin secretion was only obtained in the presence of 10 mM linoleate. A continuously elevated output of insulin was also observed in the presence of 5 mM linolenic acid (data not shown). These two concentrations were, therefore, chosen to perform experiments comparing the insulinotropic potencies of the nutrients. Insulin output from islets was stimulated with increments in the ambient glucose concentrations. The mean integrated insulin AUC/SO min increased from a mean basal of 2526 + 155 to 3924 k 626 and 7487 f 685 pg (P < 0.05) when the glucose concentration was raised to 11.1 and 22.2 mM, respectively. No statistically significant increase in insulin secretion was observed when the islets were perifused with higher glucose concentrations (27.7 and 33.3 mM), although 27.7 InM glucose was used in .experiments to compare the potencies of the secretagogues.
ACIDS
ON INSULIN
Evaluation of the insulinotropic secretagogues (Figs. 3 and 4)
and linoleate on
5.5mM
12000 , g
I
potencies of nutrient
ARG 5.5mM
GLU
In a separate set of control experiments, shown in Fig. 2, the addition of 10 mM palmitate to basal glucose perifusate had no effect on basal insulin output, as the basal mean integrated insulin AUC/20 min values were 3351 k 311 pg without and 3128 fi 233 pg with palmitate. In contrast, the addition of linoleate to the basal glucose perifusate enhanced insulin output to 6669 + 708 pg (P < 0.0001) above basal (Fig. 2).
659
Figure 3 is a diagrammatic representation of the perifusion protocols used to compare the insulinotropic potencies of the different nutrients in a series of nine experiments in which the position of each nutrient in the perifusion profile was randomly altered. These two separate protocols, shown in Fig. 3, were used in these experiments to avoid exposure of islets to 27.7 mM glucose after a preceding polyunsaturated fatty acid challenge. The mean data were assessed as integrated insulin AUC/BO min above basal, and as shown in Fig. 4, there was no statistically significant difference between the islet responses to 20 mM arginine and 27.7 mM glucose. However, 10 mM linoleate was more potent than 20 mM arginine in stimulating insulin release (P < 0.05), although the value of this response was not significantly different from that observed in the presence of 27.7 mM glucose. The stimulatory effect of 5 mM linolenic acid on insulin secretion observed was more potent than those of 27.7 mM glucose (P < 0.05) and 20 mM arginine (P < O.Ol), but was not statistically different from that of linoleic acid (Fig. 4). The arginine concentration (20 PUFA
Comparison of the effects of palmitate insulin secretion (Fig. 2)
SECRETION
5.5mM
PUFA 5.5mM
ARG 5.5mM
5.5mM glucose PUFA
5.5mM
5.5mM glucose
FIG. 3. Experimental
protocols used for comparing insulinotropic potencies of nutrients. The data shown in Fig. 4 were generated using these two protocols in which islets underwent a 20-min washout with basal perifusion (5.5 mM glucose) before another secretagogue perifusion. The protocols involved the perifusion of the same islet preparation with three different nutrients. The order of perifusion shown in the diagram was randomly switched during these experiments. PUFA, Polyunsaturated fatty acid; ARG, arginine; GLU, glucose.
10000
Ez 6 8000
10000
E -E 8000 3 6000
.-E 6000 8 3 4000
2
4000
Glucose 5.5mM Glucose
+lOmM Palmitate
l lOmM Linoleate
FIG. 2. Comparison of the effects of palmitate and linoleate on insulin secretion. Islets were perifused with equimolar concentrations of palmitate and linoleate, separated by a washout with 5.5 mM glucose perifusate. Insulin secretion was assessed as the mean integrated area under the curve/20 min for six experiments.
27.7mM
Arginine 20mM
Linoleic Linolenic Acid 1OmM Acid 5mM
FIG. 4. Effect of nutrient secretagogues on insulin secretion. In a series of nine separate experiments, the insulin response to maximally effective concentrations of glucose, arginine, linoleic acid, and linolenic acid were compared. The data were assessed as the mean integrated AUC/ 20 min and compared as follows: 1 us. 4, 2 us. 3, P < 0.05; 2 us. 4, P < 0.01.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 01 May 2015. at 15:48 For personal use only. No other uses without permission. . All rights reserved.
660
EFFECTS
OF LINOLEIC
AND
LINOLENIC
mM) used in this study was shown in a recent report (18) to be maximally stimulatory of insulin release.
Discussion We have clearly shown in the present study that individually, each of the polyunsaturated fatty acids used in our preceding report (12) is capable of stimulating insulin secretion to at least the same extent as glucose and arginine, which are presently used as the standards for evaluating the insulinotropic potencies of various nutrients (14, 15, 22). It is of interest that 5 mM linolenic acid and 10 mM linoleic acid were equipotent in stimulating insulin release. The insulinotropic abilities of these polyunsaturated fatty acids may be dependent upon their enhanced entry into the P-cell mitochondria (23-25), since it has been suggested that the oxidation of fatty acids may be involved in their stimulatory effect on insulin secretion (12, 26, 27). It is, therefore, not surprising that palmitate failed to stimulate insulin secretion, since long chain saturated fatty acids have only a limited ability to cross the inner mitochondrial membrane, as their entry has to be stimulated by carnitine (28). It has to be emphasized that the insulinotropic effects of polyunsaturated fatty acids and arginine were obtained in the presence of a basal (5.5 mM) glucose concentration. Attempts to perifuse the islets with the fatty acids alone (in the absence of basal glucose) resulted in poor islet survival and insulin response. Therefore, the presence of basal glucose in the fatty acid perifusate was essential to their stimulatory effect on insulin secretion. The link between polyunsaturated fatty acid oxidation and insulin release, which has recently been described (12, 13, 18), is consistent with one hypothesis on the mechanism of insulin secretion. The oxidation of these fatty acids would generate abundant ATP, leading to the closure of K+ channels on the P-cell plasma membrane. The resulting decrease in K+ permeability causes membrane depolarization, with activation of voltage-dependent Ca’+ channels. An increase in Ca2+ influx ensues, which raises the cytosolic concentration of free Ca2+ and ultimately triggers insulin release (29). However, there are other mechanisms by which polyunsaturated fatty acids may stimulate insulin release. It has been shown that glucose may stimulate insulin secretion via a mechanism involving the mediation of arachidonic acid, a metabolite of linoleic acid, as a second messenger (30, 31). It is, therefore, conceivable that the elongation and desaturation of linoleic acid to arachidonate would result in a stimulatory effect on insulin secretion. Furthermore, it has recently been reported that some intermediates of the tricarboxylic acid cycle (32) as well as malonyl-coenzyme-A and long chain acyl-coenzyme-A (33) may serve as metabolic coupling factors in signal transduction when islets are
ACIDS
ON INSULIN
SECRETION
Endo. 1992 Vol 130. No 2
stimulated with high glucose or glucose combined with other fuels. Although not all products of the Krebs cycle are capable of stimulating insulin release, it has been shown that esters of succinic acid are insulinotropic (32). Consequently, nutrients whose metabolism would generate this intermediate may release insulin via this pathway. Since the metabolism of these essential fatty acids would generate this and other products, such as malonlycoenzyme-A, it is certainly possible that these alternative mechanisms may be involved in the stimulatory effects of these fatty acids on insulin secretion. It should be pointed out that the insulin response to nutrients of murine islets, as apparent from this and previous studies (34-36), is monophasic and diminished in magnitude compared to that of islets from rats and humans. The reason for this diminished monophasic insulin response is unclear, but may be related to the fact that pancreatic islets in the mouse are exposed to higher basal blood glucose levels (37,38). It is conceivable that the chronic exposure of murine islets to such elevated basal glucose levels could impair the influence of these nutrients on insulin secretion (39-43). It would be appropriate to evaluate the insulinotropic potencies of these polyunsaturated fatty acids on islets with the classic biphasic insulin response to nutrients. However, it is encouraging that mixtures of these essential fatty acids have been shown to be potent insulin secretagogues in both man and rat (11,13). Although, the fatty acids that we have examined are mostly of benefit in hyperalimentation (5), in somewhat unphysiological situations they occur in the normal human diet (11) and could contribute postprandially to the stimulatory effect of a mixed meal on insulin secretion. In the presence of high concentrations of fatty acids, an artefactual increase in insulin secretion could be caused by a detergent effect of the fatty acids on pancreatic P-cells, but when we measured the insulin concentrations of the fatty acid-treated islets using an extraction procedure for peptides (44), it was found that less than 10% of the total insulin content of the islets was released during these experiments. Furthermore, a high concentration (10 mM) of palmitate failed to stimulate insulin output during these experiments. These observations would argue against a fatty acid-induced toxicity that would cause the islets to lyse and, thus, release most of their hormone contents. In addition, a recent report showed that human plasma fatty acid variations and what was previously thought unphysiological fatty acid levels occur in relationship to dietary intake (45). Besides, plasma levels of free fatty acids may not be an accurate index of their biologically effective concentrations, since the presence of lipoprotein lipase on the endothelial surface of blood capillaries, causing rapid hydrolysis of triglycerides (46), would increase the free acid concentrations to which the cells are
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 01 May 2015. at 15:48 For personal use only. No other uses without permission. . All rights reserved.
EFFECTS
OF LINOLEIC
AND
LINOLENIC
exposed. We recently observed that during multiple glucose perifusions on the same islet preparation, a possible priming effect occurred during a second challenge subsequent to the initial exposure (18). To take into account the possible influence of this phenomenon on the data from a protocol involving multiple challenges of the same islet preparation with different nutrient secretagogues, the positions of the various nutrients in the perifusion profile were randomly altered. It was necessary to have two versions of the perifusion protocol, so as to avoid a desensitization effect of a preceding polyunsaturated fatty acid challenge on islet response to glucose stimulation, which has been recently described (13,X3). It was suggested in these reports that products of polyunsaturated fatty acid metabolism may impair glucose metabolism by inhibiting glycolysis and, consequently, abolish glucose-stimulated insulin secretion. This phenomenon of polyunsaturated fatty acid-induced desensitization of islets to glucose response, therefore, prevented an additive stimulatory effect of glucose and these fatty acids on insulin secretion. In conclusion, our data suggest that linoleic acid and linolenic acid may be at least as potent, if not moreso, in releasing insulin in isolated murine islets as glucose and arginine, which were hitherto used as the standards in assessing the abilities of nutrients to stimulate the release of insulin. Acknowledgments The authors would like to thank Spencer Bridges, Lula Copeland, and Eleanor Matthews for expert technical help, and Pamela McAuley for superb editorial assistance. References
5.
6. 7. 8.
9.
10.
Burr GO, Burr MM 1930 On the nature and role of the fatty acids essential to nutrition. J Biol Chem 86587-621 Holman RT 1964 Nutritional and metabolic interrelationships between fattv acids. Fed Proc 23:1062-1067 Holman RT-1986 Essential fatty acids, prostaglandins and leukotrienes. Prog Lipid Res 25:19-47 Kinsella JE, Broughton KS, Whelan JW 1990 Dietary unsaturated fatty acids: interactions and possible needs in relation to eicosanoid synthesis. J Nutr Biochem 1:123-141 Meguid MM, Kurzer M, Hayashi RJ, Akahoshi MP 1989 Shortterm effects of fat emulsion on serum lipids in postoperative patients. JPEN 13:77-80 Rosner M. Grant JP 1987 Intravenous liuid emulsion. Nutr Clin Pratt 2:96-107 Hubbard VS 1983 What is the association of essential fatty acid status with cvstic fibrosis? Eur J Pediatr 141:68-70 Hubbard VS; McKenna MC 1987 Absorption of safflower oil and structured lipid preparation in patients with cystic fibrosis. Lipids 22:424-428 Parsons HG, O’Loughlin EV, Forbes D, Cooper D, Gall DG 1988 Supplemental calories improve essential fatty acid deficiency in cystic fibrosis patients. Pediatr Res 24:253-256 Broughton KS, Whelan J, Hardardottir I, Kinsella JE 1991 Effect of increasing the dietary (n-3) to (n-6) polyunsaturated fatty acid
ACIDS
11.
12.
13.
14.
15.
16. 17.
18.
19. 20.
21.
22. 23. 24.
25.
26.
27 28 29. 30.
31.
32.
33.
34.
ON INSULIN
SECRETION
661
ratio on murine liver and peritoneal cell fatty acids and eicosanoid formation. J Nutr 121:155-164 Lardinois CK, Starich GH, Mazzaferri EL, DeLett A 1987 Polyunsaturated fatty acids augment insulin secretion. J Am Co11 Nutr 6:507-515 Opara EC, Burch WM, Hubbard VS, Akwari OE 1990 Enhancement of endocrine pancreatic secretions by essential fatty acids. J Surg Res 48329-332 Sako Y, Grill VE 1990 A 48-hour lipid infusion in the rat timedependently inhibits glucose-induced insulin secretion and beta cell oxidation through a process likely coupled to fatty acid oxidation. Endocrinologv 127:1580-1589 Gerich JE, Charles-MA, Grodsky GM 1974 Characterization of the effects of arginine and glucose on glucagon and insulin release from the perfused rat pancreas. J Clin Invest 54:833-841 van Haeften TW, Voetberg GA, Gerich JE, van der Veen E 1989 Dose-response characteristics for arginine stimulated insulin secretion in man and influence of hyperglycemia. J Clin Endocrinol Metab 69:1059-1064 Opara EC, Atwater I, GO VLW 1988 Characterization and control of pulsatile secretion of insulin and glucagon. Pancreas 3:484-487 Turk J, Wolf BA, McDaniel ML 1987 The role of phospholipidderived mediators including arachidonic acid, its metabolite and inositol triphosphate and of intracellular Ca++ in glucose induced secretion by pancreatic islets. Prog Lipid Res 26:125-181 Opara EC, Hubbard VS, Burch WM, Akwari OE 1991 Homologous desensitization of pancreatic beta cells to glucose response by polyunsaturated fatty acids. J Nutr Biochem 2:424-429 Herbert V, Lau KS, Gottlieb GW, Blecher S 1965 Coated charcoal immunoassay of insulin. J Clin Endocrinol Metab 25:1375-1384 Morishima T, Pye S, Polonsky KS, Radziuk J 1986 The measurement and validation of the non-steady state rates of C-peptide appearance in the dog. Diabetologia 29:440-446 Eaton RF, Allen RC, Schade DS, Erickson KM, Standefer J 1980 Prehepatic insulin production in man: kinetic analysis using peripheral connecting peptide behavior. J Clin Endocrinol Metab 51:520-528 Malaisse WJ, Malaisse-Lagae F 1968 Stimulation of insulin secretion by noncarbohydrate metabolites. J Lab Clin Med 72:438-448 Lynn WS, Brown RH 1959 Oxidation and activation of unsaturated fatty acids. Arch Bichem Biophys 81:353-362 Jones PH. Pencharz PB. Clandinin MT 1985 Whole bodv oxidation of dietary’fatty acids: implications for energy utilization: Am J Clin Nutr 42:769-777 Leyton J, Drury PJ, Crawford MA 1987 Differential oxidation of saturated and unsaturated fatty acids in uiuo in the rat. Br J Nutr 57:383-393 Malaisse WJ, Malaisse-Lagae F, Sener A, Hellerstrom C 1985 Participation of endogenous fatty acids in the secretory activitv of the pancreatic B-cell.Biochem J 227:995-1002 Malaisse WJ 1989 Dual role of liuids in the stimulus-secretion coupling for insulin release. Bioche-m Sot Trans 17:59-66 Lehninger AL 1975 Biochemistry, chapt 20. Worth, New York, pp 543-547 Henquin JC 1987 Regulation of insulin release by ionic and electrical events in B-cells. Horm Res 27:168-178 Turk J, Colca JR, Kotagal N, McDaniel ML 1984 Arachidonic acid metabolism in isolated pancreatic islets. II. The effects of glucose and of inhibitors of arachidonic acid metabolism on insulin secretion and metabolite synthesis. Biochim Biophys Acta 794:125-136 Walsh MF, Pek SB 1984 Possible role of endogenous arachidonic acid metabolites in stimulated release of insulin and glucagon from the isolated perfused rat pancreas. Diabetes 33:929-936 McDonald MJ, Fahien LA, Mertz RJ, Rana RS 1989 Effect of esters of succinic acid and other citric acid cycle intermediates on insulin release and inositol phosphate formation by pancreatic islets. Arch Biochem Biophys 269:400-406 Liang Y, Matschinsky FM 1991 Content of CoA-esters in perifused rat islets stimulated by glucose and other fuels. Diabetes 40: 327-333 Scott AM, Atwater I, Rojas E 1981 A method for the simultaneous measurement of insulin release and b-cell membrane potential in
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 01 May 2015. at 15:48 For personal use only. No other uses without permission. . All rights reserved.
662
EFFECTS
OF LINOLEIC
AND
LINOLENIC
single mouse islets of Langerhans. Diabetologia 21:470-475 35. Atwater I, Carroll PB, Li MX 1989 Electrophysiology of the pancreatic B-cell. In: Draznin B (ed) Insulin Secretion. Liss, pp 49-68 36. deMegue1 R, Tamagawa T, Schmeer W, Nenquin M, Henquin JC 1987 Effect of acute sodium omission on insulin release, ionic flux, and membrane potential in mouse pancreatic B-cells. Biochim Biophys Acta 969:198-207 37. LeMarchand-Brustel Y, Balloti R, van Obberghen E 1985 Insulin receptor tyrosine kinase is defective in skeletal muscle of insulinresistant obese mice. Nature 315:676-679 38. Lebrun P, Atwater I 1985 Chaotic and irregular bursting electrical activity in mouse pancreatic B-cells. Biophys J 48:529-531 39. Leahy JL, Cooper HE, Deal DA, Weir GC 1986 Chronic hyperglycemia is associated with impaired glucose influence on insulin secretion. J Clin Invest 77:908-915 40. Robertson RP 1989 Perspectives in diabetes: type II diabetes, glucose “non-sense,” and islet desensitization. Diabetes 38: 1501-1505
ACIDS
ON
INSULIN
SECRETION
Endo. 1992 Voll30. No 2
41. Bolaffi JL, Bruno L, Heldt A, Grodsky GM 1988 Characteristics of desensitization of insulin secretion in fully in vitro systems. Endocrinology 122:1801-1809 42. Zawalich WS, Zawalich KC, Shulman GI, Rossetti L 1990 Chronic in uiuo hyperglycemia impairs phosphoinositide hydrolysis and insulin release in isolated perifused rat islets. Endocrinology 126:253-260 43. Sako Y, Grill VE 1990 Coupling of B-cell desensitization by hyperglycemia to excessive stimulation and circulating insulin in glucoseinfused rats. Diabetes 39:1580-1583 44. Koch TR, Carney JA, Go VLW 1987 Distribution and quantitation of gut neuropeptides in normal intestine and inflammatory bowel diseases. Dig Dis Sci 32:369-376 45. Lopes SM, Trimbo SL, Mascioli EA, Blackburn GL 1991 Human plasma fatty acid variations and how they are related to dietary intake. Am J Clin Nutr 53:628-637 46. Have1 RJ 1982 Symposium on lipid disorders: approach to the patient with hyperlipidemia. Med Clin North Am 66:319-333
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 01 May 2015. at 15:48 For personal use only. No other uses without permission. . All rights reserved.
