- Email: [email protected]
Enhancement of endocrine pancreatic secretions by essential fatty acids
JOURNAL
OF SURGICAL
RESEARCH
48,329-332
Enhancement EMMANUEL Departments
(1990)
of Endocrine Pancreatic by Essential Fatty Acids
C. OPARA, PH.D.,
Secretions
WARNER M. BURCH, M.D.,* VAN S. HUBBARD, AND ONYE E. AKWARI, M.D.
M.D.,
PH.D.,?
of Surgery and *Medicine, Duke University Medical Center, Durham, North Carolina 27710, and tNationa1 of Diabetes and Digestive and Kidney Diseases, National Institutes of Health, Bethesda, Marykznd 20892
Presented at the Annual Meeting of the Association
for Academic Surgery, Louisville,
Recent studies have suggested the beneficial effects of essential fatty acids in postoperative patients receiving total pare&era1 nutrition. While there is abundant information on the role of glucose and amino acids on insulin release, the effect of essential fatty acids on endocrine pancreatic secretions is not clear. Since linoleic and linolenic acids are constituents of TPN solutions as well as dietary fat, our aim was to examine their effect on the endocrine pancreatic function, using isolated islets. In each experiment, six islets microdissected from three mice were preperifused at the rate of 1 ml/min with Krebs-Ringer bicarbonate (KRB) buffer pH 7.4 containing 2% bovine albumin and 5.5 mM glucose (basal) with continuous supply of 95%/5%, Oz/COz for 1 hr, after which basal samples were collected on ice every minute. The perifusion was continued for 20 min after the addition of a mixture of 10 mM linoleic acid and 5 mM linolenic acid to the KRB. During each perifusion phase, e&tent samples were also collected for insulin and glucagon assay. The mean integrated area under the curve/ 20 min showed an increase in both insulin and glucagon secretions with the addition of fatty acids. Hence insulin increased from a basal 3154.8 f 953.7 to 8393.0 + 2073.1 pg (P < 0.025, n = 6) and glucagon increased from 193.7 +- 46.9 to 1566.1 + 411.2 pg (P < 0.0025, n = 5). The fatty-acid-induced insulin but not glucagon secretion was blocked by the addition of 2 mMpalmoxirate an inhibitor of fatty acid oxidation. In conclusion our data clearly show that linoleic acid and linolenic acid simultaneously enhance insulin and glucagon secretions and suggest different mechanisms for their insulinotropic 0 isso Academic POW, 1"~. and glucagonotropic effects.
November
15-18, 1989
resultant increase in plasma insulin level is necessary for enhancing the synthesis and storage of carbohydrate, fat, and protein [4]. Although, the stimulatory effects of glucose and amino acids on insulin output have been welldocumented, the role of fatty acids on insulin secretion has not been adequately evaluated [4,5]. Previous studies have suggested that while short chain fatty acids did not augment insulin release, medium chain fatty acids appeared to be insulinotropic [5]. Only recent reports from in viva studies have indicated that long chain fatty acids may stimulate insulin secretion [6, 71. A direct in vitro effect of these long chain fatty acids, however, remains to be established. Even more scanty are data to show the direct effect of fatty acids on glucagon secretion, an important glucose counterregulatory hormone to insulin [8]. It is known that pancreatic glucagon secretion increases after a mixed meal. This has been attributed primarily to the combined stimulatory effects of amino acids and gut hormones overcoming the suppressive effects of increased glucose concentrations [4]. The effect of free fatty acids on glucagon secretion therefore deserves to be examined, in order to assess the overall metabolic implications of dietary polyunsaturated fatty acids, which have recently received tremendous attention because of their probable beneficial effects in health and disease [9]. In the present study we have examined in vitro the direct effects of the essential fatty acids linoleic and linolenic on insulin and glucagon secretions, using concentrations of the fatty acids similar to those measured in mixed meals [6]. METHODS
Isolation INTRODUCTION
Kentucky,
Institute
of Islets
For each experiment, three adult CD1 female mice (Charles River Laboratories) were fasted for at least 4 hr but were allowed free access to drinking water. The animals, stunned with a blow to the head, were decapitated and their pancreata dissected out into a flat-bottomed plastic dish into which a mixture of wax and Norit A charcoal (American Drug and Chemical Company, Inc.,
Recent studies have shown that the use of essential fatty acids is of benefit to postoperative patients receiving total parenteral nutrition [l, 21 and in the nutritional management of some disease states such as cystic fibrosis [3]. Insulin secretion is stimulated by food intake and the 329
All
oozz-4804/90 $1.50 Copyright 0 1990 by Academic Press, Inc. rights of reproduction in any form reserved.
330
JOURNAL
OF SURGICAL
RESEARCH:
VOL. 48, NO. 4, APRIL
1990
by radioimmunoassay. The assay for insulin was based on the original dextran-coated charcoal procedure of separating bound from free insulin, using pork insulin as standard [ 121. Glucagon assay was performed using pancreatic-specific antiserum 30K (obtained from Dr. Roger Unger), with pork glucagon as standard [13]. Data Analysis 0
0'
' 0
'
'
20
'
40
'
'
'
60 Minutes
Basal
'
'
Fatlty Acids
'
'
'
80
FIG. 1. Fatty-acid-induced stimulation of insulin secretion. Microdissected islets were preperifused for 1 h at 37°C and after collection of basal samples, a mixture of linoleic and linolenic acids was incorporated into the perifusion, and effluent samples were also collected. Following the withdrawal of fatty acids, samples were again collected during another period of perifusion with basal glucose buffer, pH 7.4. A representation of the dynamics of insulin secretion is shown and the mean data for all experiments have been analyzed as AUC/PO min and provided as an inset.
New York) had been melted to create a smooth surface. Two islets were then isolated from each animal by a modification of previously reported microdissection techniques [lo, 111, using an operating microscope (Carl Zeiss 30184, Germany). Perifusion The six islets thus isolated were pooled into a plastic flowthrough chamber [ll] in which they were perifused at a constant rate of 1 ml/min with a modified KrebsRinger bicarbonate (KRB) solution, pH 7.4, containing 5.5 mM glucose, for 1 hr at 37°C. The KRB buffer, which was made up with 120 mM NaCl, 5 mM KCl, 1.1 mM MgC&, 2.5 mM CAClz, 25 mM NaHC03, 100 KIU/ml trasylol, and 2% bovine serum albumin (Armour Pharmaceuticals, Kankakee, IL), was gassed continuously with a 95%/5% Oz/COz gas mixture. Aliquots of perifusate were collected every minute for at least the last 5 min of this basal perifusion, depending on the experiment performed. Thereafter the perifusion and 2 ml of effluent sampling were continued for an additional 20 min with the addition of a mixture of 10 mM linoleic acid and 5 mM linolenic acid with or without the fatty acid oxidation inhibitor, palmoxirate. The fatty acids were then withdrawn and sampling was continued during a final 20 min of basal perifusion. The solution changes were easily accomplished using a system of stopcocks. The effluent samples were collected on ice and frozen until they were assayed for hormone content.
The amounts of hormones secreted during each perifusion phase were analyzed as areas under the curve [ 141 for period of sample collection. Statistical evaluation was done by Student’s t test, and duplicates of RIA data were considered separately. RESULTS
Fatty-Acid-Induced Stimulation Glucagon Secretions
of Insulin
and
The incorporation of the mixture of linoleic and linolenic acids into the 5.5 mM glucose perifusion caused an enhancement of insulin secretion as shown in Fig. 1. Hence, there was an increase in the mean total integrated area under the curve (AUC/20 min) from a basal of 3154.8 + 953.7 to 8383.0 + 2073.1 pg (P < 0.025) with the addition of the fatty acids. Figure 2 shows that these fatty acids also caused a simultaneous stimulation of glucagon secretion and the AUC/BO min increased from a basal of 193.7 + 46.9 to 1566.1 &- 411.2 pg (P < 0.0025). Time Course of Stimulatory Glucagon to Fatty Acids
Response of Insulin
and
As shown in Figs. 1 and 2 when the time course of action of the fatty acids on glucagon and insulin release was evaluated, it was found that the response of glucagon to the fatty acid challenge was more rapid than the response of insulin. Thus, by about 5 min of the perifusion of the islets with fat, pronounced glucagon secretion was
150
2500 T
.E E 22
100
50 Basal I
0
I
Fatty Acids I
t
60 Minutes
Measurement
of Insulin
and Glucagon
The amount of insulin and glucagon secreted by the perifused islets into each efhuent sample was measured
FIG. 2. Fatty-acid-induced stimulation of glucagon secretion. The amount of glucagon released during an individual experiment has been presented while the mean data have been analyzed as AUC/ZO min as in Fig. 1.
OPARA
ET AL.: EFFECTS
OF POLYUNSATURATED
FATTY
evident, with a maximum response occurring by 12 min (Fig. 2). Following the withdrawal of fatty acids from the perifusion, recovery of basal glucagon secretion was immediate and occurred in less than 5 min. In contrast, the insulin response of islets to the presence of the fatty acids was slow and at least 10 min of fatty acid perifusion was necessary before a significant insulin stimulation was obtained. The decay of the fat-induced effect on insulin secretion was also slow, and insulin output remained elevated above basal even after 20 min of withdrawal of the fatty acids (Fig. 1). Effect of Palmoxirate on Fatty-Acid-Induced Glucagon Secretions
Insulin
and
The addition of 2.0 mMpalmoxirate (PMX) to the fatty acid perifusion profoundly inhibited the fat-induced stimulation of insulin secretion (Fig. 3). This inhibition of the fatty-acid-induced effect on insulin by PMX was followed by an “off response” in basal insulin secretion, almost immediately after the withdrawal of the fatty acid mixture with PMX. Incorporation of PMX to 5.5 mM glucose perifusion had no effect on basal insulin secretion as apparent in Fig. 3. On the other hand, the fatty acid mixture enhanced glucagon secretion despite the presence of PMX (Fig. 4). As already noted, switching the perifusion back to 5.5 mM glucose caused a decrease of glucagon secretion to basal values. DISCUSSION
Our data clearly show that linoleic and linolenic acids stimulate the release of both insulin and glucagon. That these fatty acids should potentiate the simultaneous secretion of insulin and glucagon in the presence of low (basal) glucose is of particular physiologic importance be-
IOmM Linoleic ,
.;
2mM Palmoxirate
5.5mM Glucose ,
800
2
0'
' 0
' 10
'
' 20
'
' 30
'
' 40
'
' 50
'
1 60
Minutes inFIG. 3. Effect of palmoxirate (PMX) on fatty-acid-stimulated sulin secretion. 2.0 mM PMX (maximum concentration emulsifiable in our buffer) was incorporated into the perifusion 10 min before the addition of the fatty acids. Effluent samples were collected during all phases of perifusion. Data represent means + SE, n = 3. In each experiment, six islets microdissected from three animals were used.
ACIDS
ON INSULIN
AND
GLUCAGON
SECRETIONS
1OmM Linoleic 5mM Linolenic
120
I
331
5.5mM Glucose I
2mM Palmoxirate
100
.G E 2
80 60 40 20 0
0
10
20
30
40
50
60
Minutes FIG. 4. Effect of palmoxirate on fatty acid-stimulated glucagon secretion. The amount of glucagon present was also measured in the samples collected during the experiment described in Fig. 3. Data represent means + SE. Missing points denote glucagon amount below detection.
cause in an in vivo setting the effects of glucagon upon hepatic glucose output would prevent hypoglycemia [4]. There appears to be a distinction between the mechanism that regulate fatty-acid-stimulated insulin secretion and glucagon secretion. While the oxidation of fatty acids appears to be necessary for the stimulatory effect on insulin release, this does not seem to be the case for glucagon. This observation suggests different mechanisms for the effect of the fatty acids on (Ycells secreting glucagon and the p cells producing insulin, consistent with previous work [ 151. It is also interesting to find in our study that the inhibition of fatty-acid-stimulated insulin release by palmoxirate is followed by an off response in basal insulin output, after withdrawal of the inhibitor. This postinhibitory off response characteristic of the p cell secretion has previously been noted [ 161. Since glucagon release stimulated by the fatty acids was still obtained despite the presence of PMX, it is tempting to speculate that the observed postinhibitory off response in insulin secretion may be induced by the preceding, elevated glucagon level. The involvement of oxidation of fatty acids in their stimulatory effect on insulin secretion is consistent with the current hypothesis of the mechanism of insulin release. The oxidation of fatty acids would generate ATP leading to closure of K+ channels in the plasma membrane. The resulting decrease in K+ permeability causes membrane depolarization with activation of voltage-dependent Cazf channels. An increase in Ca2+ influx ensues, which raises the cytoplasmic concentration of free Ca2+ and ultimately triggers insulin secretion [ 171. Although calcium has been shown to play a role in glucagon secretion as well [18, 191, the mechanisms of fatty-acid-stimulated glucagon release is unknown. On the basis of the faster dynamics and the probable lack of involvement of fatty acid oxidation in stimulating glucagon secretion, one can speculate that the fatty acid effect derives essentially from a message transmitted from the (Ycell plasma membrane. Thus, such a mechanism would involve the CAMP
332
JOURNAL
OF SURGICAL
RESEARCH.
system interacting with calcium in the (Ycell as previously proposed [20, 211. In conclusion we have shown that a mixture of linoleic and linolenic acids simultaneously enhance insulin and glucagon secretions. In addition our data suggest different mechanisms for their observed insulinotropic and glucagonotropic effects.
VOL. 48, NO. 4, APRIL
1990
8. Gerich, J., Langlois, M., Noacco, C., Karam, J., and Forsham, P. s, 10.
Lack of glucagon response to hypoglycemia in diabetes: Evidence for an intrinsic alpha cell defect. Science 182: 171, 1973. Axelrod, L. Omega-3 fatty acids in diabetes mellitus: Gift from the sea? Diabetes 38: 539, 1989. Opara, E. C., Atwater, I., and Go, V. L. W. Characterization and control of pulsaltile secretion of insulin and glucagon. Pancreas 3:
484,1988. ACKNOWLEDGMENTS The authors thank Dr. Ronald Chance of Eli Lilly for the generous supply of crystalline insulin and glucagon, the R. W. Johnson Pharmaceutical Research Institute for a gift of palmoxirate, Drs. John Miles and Gordon Weir for helpful suggestions, Spencer Bridges and Lula Copeland for skillful technical assistance, and Pamela McAuley for superb editorial assistance.
REFERENCES 1.
Meguid, M. M., Akahoshi, M. P., Jeffers, S., Hayashi, R. J., and Hammond, W. G. Amelioration of metabolic complications of conventional Total Parenteral Nutrition: A prospective randomized study. Arch. Surg. 119: 1294, 1984.
2. Meguid, M. M., Kurzer, M., Hayashi,
R. J., and Akahosi, M. P. Short-term effects of fat emulsion on serum lipids in postoperative patients. JPEN 13: 77, 1989.
3. Hubbard, V. S. What is the association of essential fatty acid status with cystic fibrosis? Eur. J. Pediutr. 141: 68, 1983. S., and Weir, G. C. The Islets of Langerhans and 4. Bonner-Weir, Diabetes Mel&us. 21.
Kalamazoo,
MI: The Upjohn Co., 1986. Pp. 19-
5. Laube, H., and Pfeiffer,
E. F. Insulin secretion and the role of nutritional factors. In (H. M. Katzen and R. J. Mahler, Eds.), Diabetes, Obesity and Vascular Disease-Metabolic and Molecular Interrelationships. New York: Wiley, 1978. Part 2, pp. 398-425.
6. Lardinois,
C. K., Starich, G. H., Mazzaferri, E. L., and DeLett, A. Polyunsaturated fatty acids augment insulin secretion. J. Amer. Coil. Nutr. 6: 507, 1987.
7. Vara, E., and Tamarit-Rodriguez,
J. Glucose stimulation of insulin secretion in islets of fed and starved rats and its dependence on lipid metabolism. Metabolism 36: 266,1986.
11.
Scott, A. M., Atwater, I., and Rojas, E. A method for the simultaneous measurement of insulin release and B cell membrane potential in single mouse islets of Langerhans. Diubetologia 21: 470, 1981. 12. Herbert, V., Lau, K. S., Gottlieb, C. W., and Bleicher, S. Coated charcoal immunoassay of insulin. J. Clin. Endocrirwl. Metab. 25: 1375,1965. 13. Unger, R. H., Eisentraut, A. M., McCall, M. S., and Madison, L. L. Glucagon antibodies and immunoassay for glucagon. J. Clin. Znuest. 40: 1280, 1961. 14. Shuster, L. T., Go, V. L. W., Rizza, R. A., O’Brien, P. C., and Service, F. J. Incretin effect due to increased secretion and decreased clearance of insulin in normal humans. Diabetes 37: 200, 1988. 15. Pek, S., Santiago, J. C., and Tai, T. Y. L-Leucine-induced secretion of glucagon and insulin, and the “off-response” to L-leucine in vitro. I. Characterization of the dynamics of secretion. Endocrinology 103: 1208, 1978. 16. Moltz, J. H., and McDonald, J. K. Neuropeptide Y: Direct and indirect action on insulin secretion in the rat. Peptides 6: 1155, 1985. 17. Henquin, J. C. Regulation of insulin release by ionic and electrical events in B cells. Horm. Res. 27: 168, 1987. 18. Gerich, J. E., Frankel, B. J., Fanska, R., West, L., Forsham, P. H., and Grodsky, G. M. Calcium dependency of glucagon secretion from the in vitro perfused rat pancreas. Endocrinology 94: 1381, 1974. 19. Leclercq-Meyer, V., Marchand, J., and Malaisse, W. J. The role of calcium in glucagon release, interactions between glucose and calcium. Diabetologia 12: 531, 1976. 20. Howell, S. L., and Whitfield, M. Cytochemical localization of adenyl cyclase activity in rat islets of Langerhans. J. Hi&o&m. Cytochem.
20:873,1972. 21. Howell, S. L., Edwards, J. C., and Montague,
W. Regulation of adenylate cyclase and cyclic-AMP protein kinase activities in Ascell rich guinea-pig islets of Langerhans. Horm. Metab. Res. 6: 49, 1974.
OF SURGICAL
RESEARCH
48,329-332
Enhancement EMMANUEL Departments
(1990)
of Endocrine Pancreatic by Essential Fatty Acids
C. OPARA, PH.D.,
Secretions
WARNER M. BURCH, M.D.,* VAN S. HUBBARD, AND ONYE E. AKWARI, M.D.
M.D.,
PH.D.,?
of Surgery and *Medicine, Duke University Medical Center, Durham, North Carolina 27710, and tNationa1 of Diabetes and Digestive and Kidney Diseases, National Institutes of Health, Bethesda, Marykznd 20892
Presented at the Annual Meeting of the Association
for Academic Surgery, Louisville,
Recent studies have suggested the beneficial effects of essential fatty acids in postoperative patients receiving total pare&era1 nutrition. While there is abundant information on the role of glucose and amino acids on insulin release, the effect of essential fatty acids on endocrine pancreatic secretions is not clear. Since linoleic and linolenic acids are constituents of TPN solutions as well as dietary fat, our aim was to examine their effect on the endocrine pancreatic function, using isolated islets. In each experiment, six islets microdissected from three mice were preperifused at the rate of 1 ml/min with Krebs-Ringer bicarbonate (KRB) buffer pH 7.4 containing 2% bovine albumin and 5.5 mM glucose (basal) with continuous supply of 95%/5%, Oz/COz for 1 hr, after which basal samples were collected on ice every minute. The perifusion was continued for 20 min after the addition of a mixture of 10 mM linoleic acid and 5 mM linolenic acid to the KRB. During each perifusion phase, e&tent samples were also collected for insulin and glucagon assay. The mean integrated area under the curve/ 20 min showed an increase in both insulin and glucagon secretions with the addition of fatty acids. Hence insulin increased from a basal 3154.8 f 953.7 to 8393.0 + 2073.1 pg (P < 0.025, n = 6) and glucagon increased from 193.7 +- 46.9 to 1566.1 + 411.2 pg (P < 0.0025, n = 5). The fatty-acid-induced insulin but not glucagon secretion was blocked by the addition of 2 mMpalmoxirate an inhibitor of fatty acid oxidation. In conclusion our data clearly show that linoleic acid and linolenic acid simultaneously enhance insulin and glucagon secretions and suggest different mechanisms for their insulinotropic 0 isso Academic POW, 1"~. and glucagonotropic effects.
November
15-18, 1989
resultant increase in plasma insulin level is necessary for enhancing the synthesis and storage of carbohydrate, fat, and protein [4]. Although, the stimulatory effects of glucose and amino acids on insulin output have been welldocumented, the role of fatty acids on insulin secretion has not been adequately evaluated [4,5]. Previous studies have suggested that while short chain fatty acids did not augment insulin release, medium chain fatty acids appeared to be insulinotropic [5]. Only recent reports from in viva studies have indicated that long chain fatty acids may stimulate insulin secretion [6, 71. A direct in vitro effect of these long chain fatty acids, however, remains to be established. Even more scanty are data to show the direct effect of fatty acids on glucagon secretion, an important glucose counterregulatory hormone to insulin [8]. It is known that pancreatic glucagon secretion increases after a mixed meal. This has been attributed primarily to the combined stimulatory effects of amino acids and gut hormones overcoming the suppressive effects of increased glucose concentrations [4]. The effect of free fatty acids on glucagon secretion therefore deserves to be examined, in order to assess the overall metabolic implications of dietary polyunsaturated fatty acids, which have recently received tremendous attention because of their probable beneficial effects in health and disease [9]. In the present study we have examined in vitro the direct effects of the essential fatty acids linoleic and linolenic on insulin and glucagon secretions, using concentrations of the fatty acids similar to those measured in mixed meals [6]. METHODS
Isolation INTRODUCTION
Kentucky,
Institute
of Islets
For each experiment, three adult CD1 female mice (Charles River Laboratories) were fasted for at least 4 hr but were allowed free access to drinking water. The animals, stunned with a blow to the head, were decapitated and their pancreata dissected out into a flat-bottomed plastic dish into which a mixture of wax and Norit A charcoal (American Drug and Chemical Company, Inc.,
Recent studies have shown that the use of essential fatty acids is of benefit to postoperative patients receiving total parenteral nutrition [l, 21 and in the nutritional management of some disease states such as cystic fibrosis [3]. Insulin secretion is stimulated by food intake and the 329
All
oozz-4804/90 $1.50 Copyright 0 1990 by Academic Press, Inc. rights of reproduction in any form reserved.
330
JOURNAL
OF SURGICAL
RESEARCH:
VOL. 48, NO. 4, APRIL
1990
by radioimmunoassay. The assay for insulin was based on the original dextran-coated charcoal procedure of separating bound from free insulin, using pork insulin as standard [ 121. Glucagon assay was performed using pancreatic-specific antiserum 30K (obtained from Dr. Roger Unger), with pork glucagon as standard [13]. Data Analysis 0
0'
' 0
'
'
20
'
40
'
'
'
60 Minutes
Basal
'
'
Fatlty Acids
'
'
'
80
FIG. 1. Fatty-acid-induced stimulation of insulin secretion. Microdissected islets were preperifused for 1 h at 37°C and after collection of basal samples, a mixture of linoleic and linolenic acids was incorporated into the perifusion, and effluent samples were also collected. Following the withdrawal of fatty acids, samples were again collected during another period of perifusion with basal glucose buffer, pH 7.4. A representation of the dynamics of insulin secretion is shown and the mean data for all experiments have been analyzed as AUC/PO min and provided as an inset.
New York) had been melted to create a smooth surface. Two islets were then isolated from each animal by a modification of previously reported microdissection techniques [lo, 111, using an operating microscope (Carl Zeiss 30184, Germany). Perifusion The six islets thus isolated were pooled into a plastic flowthrough chamber [ll] in which they were perifused at a constant rate of 1 ml/min with a modified KrebsRinger bicarbonate (KRB) solution, pH 7.4, containing 5.5 mM glucose, for 1 hr at 37°C. The KRB buffer, which was made up with 120 mM NaCl, 5 mM KCl, 1.1 mM MgC&, 2.5 mM CAClz, 25 mM NaHC03, 100 KIU/ml trasylol, and 2% bovine serum albumin (Armour Pharmaceuticals, Kankakee, IL), was gassed continuously with a 95%/5% Oz/COz gas mixture. Aliquots of perifusate were collected every minute for at least the last 5 min of this basal perifusion, depending on the experiment performed. Thereafter the perifusion and 2 ml of effluent sampling were continued for an additional 20 min with the addition of a mixture of 10 mM linoleic acid and 5 mM linolenic acid with or without the fatty acid oxidation inhibitor, palmoxirate. The fatty acids were then withdrawn and sampling was continued during a final 20 min of basal perifusion. The solution changes were easily accomplished using a system of stopcocks. The effluent samples were collected on ice and frozen until they were assayed for hormone content.
The amounts of hormones secreted during each perifusion phase were analyzed as areas under the curve [ 141 for period of sample collection. Statistical evaluation was done by Student’s t test, and duplicates of RIA data were considered separately. RESULTS
Fatty-Acid-Induced Stimulation Glucagon Secretions
of Insulin
and
The incorporation of the mixture of linoleic and linolenic acids into the 5.5 mM glucose perifusion caused an enhancement of insulin secretion as shown in Fig. 1. Hence, there was an increase in the mean total integrated area under the curve (AUC/20 min) from a basal of 3154.8 + 953.7 to 8383.0 + 2073.1 pg (P < 0.025) with the addition of the fatty acids. Figure 2 shows that these fatty acids also caused a simultaneous stimulation of glucagon secretion and the AUC/BO min increased from a basal of 193.7 + 46.9 to 1566.1 &- 411.2 pg (P < 0.0025). Time Course of Stimulatory Glucagon to Fatty Acids
Response of Insulin
and
As shown in Figs. 1 and 2 when the time course of action of the fatty acids on glucagon and insulin release was evaluated, it was found that the response of glucagon to the fatty acid challenge was more rapid than the response of insulin. Thus, by about 5 min of the perifusion of the islets with fat, pronounced glucagon secretion was
150
2500 T
.E E 22
100
50 Basal I
0
I
Fatty Acids I
t
60 Minutes
Measurement
of Insulin
and Glucagon
The amount of insulin and glucagon secreted by the perifused islets into each efhuent sample was measured
FIG. 2. Fatty-acid-induced stimulation of glucagon secretion. The amount of glucagon released during an individual experiment has been presented while the mean data have been analyzed as AUC/ZO min as in Fig. 1.
OPARA
ET AL.: EFFECTS
OF POLYUNSATURATED
FATTY
evident, with a maximum response occurring by 12 min (Fig. 2). Following the withdrawal of fatty acids from the perifusion, recovery of basal glucagon secretion was immediate and occurred in less than 5 min. In contrast, the insulin response of islets to the presence of the fatty acids was slow and at least 10 min of fatty acid perifusion was necessary before a significant insulin stimulation was obtained. The decay of the fat-induced effect on insulin secretion was also slow, and insulin output remained elevated above basal even after 20 min of withdrawal of the fatty acids (Fig. 1). Effect of Palmoxirate on Fatty-Acid-Induced Glucagon Secretions
Insulin
and
The addition of 2.0 mMpalmoxirate (PMX) to the fatty acid perifusion profoundly inhibited the fat-induced stimulation of insulin secretion (Fig. 3). This inhibition of the fatty-acid-induced effect on insulin by PMX was followed by an “off response” in basal insulin secretion, almost immediately after the withdrawal of the fatty acid mixture with PMX. Incorporation of PMX to 5.5 mM glucose perifusion had no effect on basal insulin secretion as apparent in Fig. 3. On the other hand, the fatty acid mixture enhanced glucagon secretion despite the presence of PMX (Fig. 4). As already noted, switching the perifusion back to 5.5 mM glucose caused a decrease of glucagon secretion to basal values. DISCUSSION
Our data clearly show that linoleic and linolenic acids stimulate the release of both insulin and glucagon. That these fatty acids should potentiate the simultaneous secretion of insulin and glucagon in the presence of low (basal) glucose is of particular physiologic importance be-
IOmM Linoleic ,
.;
2mM Palmoxirate
5.5mM Glucose ,
800
2
0'
' 0
' 10
'
' 20
'
' 30
'
' 40
'
' 50
'
1 60
Minutes inFIG. 3. Effect of palmoxirate (PMX) on fatty-acid-stimulated sulin secretion. 2.0 mM PMX (maximum concentration emulsifiable in our buffer) was incorporated into the perifusion 10 min before the addition of the fatty acids. Effluent samples were collected during all phases of perifusion. Data represent means + SE, n = 3. In each experiment, six islets microdissected from three animals were used.
ACIDS
ON INSULIN
AND
GLUCAGON
SECRETIONS
1OmM Linoleic 5mM Linolenic
120
I
331
5.5mM Glucose I
2mM Palmoxirate
100
.G E 2
80 60 40 20 0
0
10
20
30
40
50
60
Minutes FIG. 4. Effect of palmoxirate on fatty acid-stimulated glucagon secretion. The amount of glucagon present was also measured in the samples collected during the experiment described in Fig. 3. Data represent means + SE. Missing points denote glucagon amount below detection.
cause in an in vivo setting the effects of glucagon upon hepatic glucose output would prevent hypoglycemia [4]. There appears to be a distinction between the mechanism that regulate fatty-acid-stimulated insulin secretion and glucagon secretion. While the oxidation of fatty acids appears to be necessary for the stimulatory effect on insulin release, this does not seem to be the case for glucagon. This observation suggests different mechanisms for the effect of the fatty acids on (Ycells secreting glucagon and the p cells producing insulin, consistent with previous work [ 151. It is also interesting to find in our study that the inhibition of fatty-acid-stimulated insulin release by palmoxirate is followed by an off response in basal insulin output, after withdrawal of the inhibitor. This postinhibitory off response characteristic of the p cell secretion has previously been noted [ 161. Since glucagon release stimulated by the fatty acids was still obtained despite the presence of PMX, it is tempting to speculate that the observed postinhibitory off response in insulin secretion may be induced by the preceding, elevated glucagon level. The involvement of oxidation of fatty acids in their stimulatory effect on insulin secretion is consistent with the current hypothesis of the mechanism of insulin release. The oxidation of fatty acids would generate ATP leading to closure of K+ channels in the plasma membrane. The resulting decrease in K+ permeability causes membrane depolarization with activation of voltage-dependent Cazf channels. An increase in Ca2+ influx ensues, which raises the cytoplasmic concentration of free Ca2+ and ultimately triggers insulin secretion [ 171. Although calcium has been shown to play a role in glucagon secretion as well [18, 191, the mechanisms of fatty-acid-stimulated glucagon release is unknown. On the basis of the faster dynamics and the probable lack of involvement of fatty acid oxidation in stimulating glucagon secretion, one can speculate that the fatty acid effect derives essentially from a message transmitted from the (Ycell plasma membrane. Thus, such a mechanism would involve the CAMP
332
JOURNAL
OF SURGICAL
RESEARCH.
system interacting with calcium in the (Ycell as previously proposed [20, 211. In conclusion we have shown that a mixture of linoleic and linolenic acids simultaneously enhance insulin and glucagon secretions. In addition our data suggest different mechanisms for their observed insulinotropic and glucagonotropic effects.
VOL. 48, NO. 4, APRIL
1990
8. Gerich, J., Langlois, M., Noacco, C., Karam, J., and Forsham, P. s, 10.
Lack of glucagon response to hypoglycemia in diabetes: Evidence for an intrinsic alpha cell defect. Science 182: 171, 1973. Axelrod, L. Omega-3 fatty acids in diabetes mellitus: Gift from the sea? Diabetes 38: 539, 1989. Opara, E. C., Atwater, I., and Go, V. L. W. Characterization and control of pulsaltile secretion of insulin and glucagon. Pancreas 3:
484,1988. ACKNOWLEDGMENTS The authors thank Dr. Ronald Chance of Eli Lilly for the generous supply of crystalline insulin and glucagon, the R. W. Johnson Pharmaceutical Research Institute for a gift of palmoxirate, Drs. John Miles and Gordon Weir for helpful suggestions, Spencer Bridges and Lula Copeland for skillful technical assistance, and Pamela McAuley for superb editorial assistance.
REFERENCES 1.
Meguid, M. M., Akahoshi, M. P., Jeffers, S., Hayashi, R. J., and Hammond, W. G. Amelioration of metabolic complications of conventional Total Parenteral Nutrition: A prospective randomized study. Arch. Surg. 119: 1294, 1984.
2. Meguid, M. M., Kurzer, M., Hayashi,
R. J., and Akahosi, M. P. Short-term effects of fat emulsion on serum lipids in postoperative patients. JPEN 13: 77, 1989.
3. Hubbard, V. S. What is the association of essential fatty acid status with cystic fibrosis? Eur. J. Pediutr. 141: 68, 1983. S., and Weir, G. C. The Islets of Langerhans and 4. Bonner-Weir, Diabetes Mel&us. 21.
Kalamazoo,
MI: The Upjohn Co., 1986. Pp. 19-
5. Laube, H., and Pfeiffer,
E. F. Insulin secretion and the role of nutritional factors. In (H. M. Katzen and R. J. Mahler, Eds.), Diabetes, Obesity and Vascular Disease-Metabolic and Molecular Interrelationships. New York: Wiley, 1978. Part 2, pp. 398-425.
6. Lardinois,
C. K., Starich, G. H., Mazzaferri, E. L., and DeLett, A. Polyunsaturated fatty acids augment insulin secretion. J. Amer. Coil. Nutr. 6: 507, 1987.
7. Vara, E., and Tamarit-Rodriguez,
J. Glucose stimulation of insulin secretion in islets of fed and starved rats and its dependence on lipid metabolism. Metabolism 36: 266,1986.
11.
Scott, A. M., Atwater, I., and Rojas, E. A method for the simultaneous measurement of insulin release and B cell membrane potential in single mouse islets of Langerhans. Diubetologia 21: 470, 1981. 12. Herbert, V., Lau, K. S., Gottlieb, C. W., and Bleicher, S. Coated charcoal immunoassay of insulin. J. Clin. Endocrirwl. Metab. 25: 1375,1965. 13. Unger, R. H., Eisentraut, A. M., McCall, M. S., and Madison, L. L. Glucagon antibodies and immunoassay for glucagon. J. Clin. Znuest. 40: 1280, 1961. 14. Shuster, L. T., Go, V. L. W., Rizza, R. A., O’Brien, P. C., and Service, F. J. Incretin effect due to increased secretion and decreased clearance of insulin in normal humans. Diabetes 37: 200, 1988. 15. Pek, S., Santiago, J. C., and Tai, T. Y. L-Leucine-induced secretion of glucagon and insulin, and the “off-response” to L-leucine in vitro. I. Characterization of the dynamics of secretion. Endocrinology 103: 1208, 1978. 16. Moltz, J. H., and McDonald, J. K. Neuropeptide Y: Direct and indirect action on insulin secretion in the rat. Peptides 6: 1155, 1985. 17. Henquin, J. C. Regulation of insulin release by ionic and electrical events in B cells. Horm. Res. 27: 168, 1987. 18. Gerich, J. E., Frankel, B. J., Fanska, R., West, L., Forsham, P. H., and Grodsky, G. M. Calcium dependency of glucagon secretion from the in vitro perfused rat pancreas. Endocrinology 94: 1381, 1974. 19. Leclercq-Meyer, V., Marchand, J., and Malaisse, W. J. The role of calcium in glucagon release, interactions between glucose and calcium. Diabetologia 12: 531, 1976. 20. Howell, S. L., and Whitfield, M. Cytochemical localization of adenyl cyclase activity in rat islets of Langerhans. J. Hi&o&m. Cytochem.
20:873,1972. 21. Howell, S. L., Edwards, J. C., and Montague,
W. Regulation of adenylate cyclase and cyclic-AMP protein kinase activities in Ascell rich guinea-pig islets of Langerhans. Horm. Metab. Res. 6: 49, 1974.