Diacylglycerols Modulate Insulin Action in Rat Adipocytes Maria L. Terry, J. Levy and G. Grunberger Diabetes Section, Department of Internal Medicine, Wayne State University School of Medicine, Detroit, Michigan, U. S. A.
Effect of 1,2-diacylglycerols on the insulin receptor function and insulin action in rat adipocytes was studied. 1,2-dioctanoylglycerol (100 ug/ml) did not alter insulin binding but it did stimulate phosphorylation of the (B-subunit of the insulin receptor as well as its tyrosine kinase activity. However, dioctanoylglycerol inhibited insulin-stimulated receptor autophosphorylation. This concentration of dioctanoylglycerol inhibited insulin-stimulated CO2 metabolism, lipogenesis and 3-O-methyl-glucose transport in a dose-dependent manner but did not alter any of these bioeffects in absence of insulin. While there was no direct link between diacylglycerol effect on tyrosine kinase activity of the insulin receptor and insulin action in rat adipocytes, the parallel inhibition of insulin-stimulated receptor autophosphorylation and insulin bioeffects by dioctanoylglycerol suggests its direct or indirect role in insulin signalling in rat fat cells. Key words Insulin Action — Insulin Receptor — Diacylglycerol — Protein Tyrosine Kinase
Introduction Transmission of the insulin signal from the initial binding of the hormone to its specific membrane receptor to the ultimate effects in the cell interior results from an interplay of several signalling mechanisms (Brautigan and Kuplic 1988). The serine- and threonine-specific protein kinase C (Kikkawa, Minakuchi, Takai and Nishizuka 1983) could modulate and/or terminate the insulin receptor tyrosine kinase signal (Duronio and Jacobs 1990). Tumor promoting phorbol diesters (e. g. TPA) and sn-1,2-diacylglycerols activate protein kinase C. Specific receptors for TPA copurify with protein kinase C and DAG's compete with TPA for binding to the phorbol ester ester receptors {Parker, Coussens, Totty, Rhee, Young, Chen, Stabel, Waterfleld and Ullrich 1986; Sharkey, Leach and Blumberg 1984). In addition to these classic effects of TPA and DAG's, these compounds modulated the function of the insulin receptors in cultured human mononuclear cells (Grunberger, Zick, Taylor and Gorden 1984a; Grunberger, lacopetta, Carpentier and Gorden 1986) which suggested a direct interaction of TPA and DAG's with the insulin receptor. In this report exogenous DAG stimulated the Horm.metab.Res. 23 (1991)266-270 © Georg Thieme Verlag Stuttgart -New York
tyrosine kinase activity (TKA) of the rat adipocyte insulin receptor but inhibited insulin-stimulated receptor autophosphorylation as well as several bioeffects of insulin. These results differed somewhat from past work (Christensen, Shade, Graves and McDonald 1987; Standaert, Farese, Cooper and Pollet 1988) where DAG's stimulated glucose uptake in absence of insulin. Materials and Methods Rat adipocytes were obtained from the epididymal fat pads of 6-8 male Sprague-Dawley rats (180-220 g). They were pooled and digested with collagenase (Rodbell 1964). Cell viability was assessed by the trypan blue exclusion. The cells were washed at 37 °C in a Krebs-Ringer bicarbonate buffer (pH 7.4; 30 mM HEPES with 1 % untreated BSA) and reduced to 10 mM HCO3. After evaporation of the DAG-chloroform solution under N2, the DAG's were resuspended in methanol (final concentration < 0.1 %; at this concentration methanol did not affect any of the assays described) before the addition of assay buffer (Grunberger et al. 1986). Specific 125I-insulin binding to intact cells was done as in Sonne and Gliemann (1980). Cells were incubated with 125I-insulin (0.1 nM) for 5,15,30,45 and 60 min at 37 °C in the presence or absence of unlabeled insulin (Gliemann, Osterlind, Vinten and Gammeltoft 1972). "Nonspecific" binding (with 16.7 uM insulin) was subtracted from total binding to give specific insulin binding. Plasma membranes were prepared (Belsham, Denton and Tanner 1980) with modifications. SBEM (0.25 M sucrose/10 mM Tris-HCl/2 mM EGTA, 10 mM sodium fluoride, pH 7.4) in the presence of PMSF (4 mM) and aprotinin (2 jig/ml), was added to washed cells. After homogenization the cell lysate was centrifuged (5 min at 1,000 x g). The infranate and pellet were aspirated from below the fat layer and centrifuged (30 min at 30,000 x g); the pellet was suspended in SBEM. Plasma membranes were obtained by centrifugation in the presence of a self-forming gradient of Percoll. The plasma membrane band was removed and washed two times (15 min at 10,000 x g) with a 5-fold dilution of a buffer (0.15 M NaCl/10 mM TrisHC1/1 mM EGTA, pH 7.4). The pellet was suspended in the buffer, solubilized in 1 % Triton X-100 in presence of protease inhibitors and applied to a wheat-germ agglutinin/agarose column. The receptor-enriched preparation was eluted with 0.3 M N-acetyl-D-glucosamine and 10% glycerol. Protein content of the solubilized material was determined {Bradford 1976). Autophosphorylation receptor B-subunit
of the
Volumes of solubilized partially-enriched adipocyte insulin receptors were adjusted to equal specific insulin binding capacity and preincubated in the absence or presence of insulin (0.1 uM), DiC8 (100 ug/ml) or both for 30 min at 22 °C as in Grunberger et al. (1984a). The phosphorylation reaction done in a final concentration of 3 mM Mn acetate, 20 mM MgCl2, 50 nM [y-32P ]-ATP, and 50 Received: 20 Apr. 1990
Accepted: 5 Oct. 1990 after revision
Downloaded by: Boston University. Copyrighted material.
Summary
Diacylglycerob and Insulin Action
Horm. metab. Res. 23 (1991)
267
Table 1 Effect of DiC8 on phosphorylation of the rat adipocyte insulin receptor. 32
Basal + DiC8 (100 ug/ml) + insulin (0.1 u.M) + DiC8 and insulin
P Content of p-subunit 0.80 ±0.08 2.81 ±0.14 15.71 ±0.34 8.27 ±0.64
Percent of basal
Tyrosine kinase activity
Percent of basal
100 351 1964 1034
2.23 ±0.76 6.04 ±0.73 15.13±1.05 15.70 ±0.81
100 271 678 704
32
P content of the receptor B-subunit (in arbitrary density units), obtained from scanning densitometry of autoradiograms, is expressed as mean±SEM of duplicates from two experiments. Tyrosine kinase activity is expressed in pmole 32P/mg (Glu80Tyr20 )/10 min/receptor as mean±SEM of quadruplicates from four experiments. See the Methods section for details.
Tyrosine kinase activity of the insulin receptor 80
20
Table 2 Insulin and DiC8-stimulated 3-O-methyl glucose transport.
Basal + DiC8 (100 ug/ml) + insulin (0.1 uM) + DiC8 and insulin
fmole/cell/min
Percent of basal
0.098 ±0.006 0.091 ±0.005 4.538 ±0.235 3.734 ±0.176*
100 93 4630 3810
Data represent the mean ± SEM of triplicates of three separate experiments. *The result compared to insulin alone was significantly different at P < 0.01 level.
Poly (Glu Tyr ) was phosphorylated in the absence or presence of DiC8 (100 |xg/ml), insulin (0.1 uM) or both (Grunberger, Comi, Taylor and Gorden 1984b). Results were expressed in tyrosine kinase units [pmol 32P/mg poly(Glu80Tyr20) at 22 °C in 10 min by a unit of specific insulin binding activity defined as the and phosphorylation of the B-subunit of the insulin receptor bound/free ratio from parallel specific 125I-insulin binding assays]. was measured. Insulin, a positive control, increased autophosphorylation of the receptor's B-subunit ~ 20-fold over the basal state (Table 1). DiC8 alone stimulated the phosphorylaGlucose oxidation tion of the insulin receptor ~ 3.5-fold. However, simultaneous Adipocytes in the KRBH buffer with 3 % BSA were presence of the D A G and insulin decreased receptor autoplaced in a shaking waterbath (40 osc/min) at 37 °C and pre-inphosphorylation compared to insulin alone (Table 1). cubated with insulin and/or DAG for 15 min prior to a 60 min incubation with 0.1 mM glucose and 0.1 uCi glucose-l-14C/ml (Salans and Insulin (0.1 uM) stimulated phosphorylation Dougherty 1971). Incubation with DAG's did not affect the cell viability or cause cell leakage compared to control incubations. of an artificial substrate, poly (Glu 80 Tyr20 ) by the partially purified insulin receptor preparations 7-fold (Table 1). DiC8 (100 ug/ml) also increased TKA of these lectin-purified prepLipogenesis arations, even though always less than insulin. Insulin and Lipogenesis was measured by the method of Moody, DiC8 together increased TKA as much as insulin alone. The Stan, Stan and Gliemann (1974). Fat cells (10 ) were incubated in 1 ml optimal effect of DiC8 on poly(Glu80Tyr20) phosphorylation Krebs-Ringer phosphate buffer with 1% BSA, [3-3H]glucose (0.1 was at 100 ug/ml, after a 15 min preincubation at room temuCi), 5 mM glucose and insulin (0-16.67 nM) in the absence or presperature (not shown). ence of 100 u.g/ml of DiC8. D-Glucose
transport
The rate of 3-O-methylglucose transport was determined as in Foley, Cushman and Salans (1978) and Karnieli, Zarnowski, Hissin, Simpson, Salans and Cushman (1981). Adipocytes (~ 5 x 16) in the KRBH buffer with 3 % BSA were preincubated with insulin (6.67 nM) and/or DAG (100 ug/ml) for 30 min. The cell suspension was pulsed with [14C]-3-0-methylglucose (0.1 mM, 58.2 uCi/u,mol) and [3H]-L-glucose (0.1 mM, 11.6 uCi/umol). Uptake was stopped by cytochalasin B (0.4 mM) (Cushman and Wardzala 1980). Statistical analyses of the differences between experimental conditions were done using t-test; significance was accepted as P < 0.05. Results DiC8 or OAG did not consistently alter any I-insulin binding parameters in intact rat adipocytes. Solubilized lectin-purified fat cell insulin receptors were then preincubated with DiC8 (100 ug/ml), insulin (0.1 uJVl) or both
We next assessed the effect of DiC8 on bioeffects of insulin. Glucose oxidation was measured by 14C-CO2 production by the adipocytes. Insulin (6.67 nM) stimulated CO2 metabolism 20.6 ± 2.6-fold. Both exogenous DAG's (1,2-0AG and l,2-DiC8) significantly decreased the insulin effect. In presence of DiC8 (100 (ug/ml) insulin increased glucose oxidation only by 9.0+ 1.2-fold. This inhibition of insulin-stimulated glucose oxidation was largely prevented when DAG's were washed away prior to addition of insulin. DiC8, a membrane-permeable synthetic glycerolipid resembling endogenous DAG, was more potent than OAG in inhibiting the insulin effect. DiC8 (5—1000 ug/ml) by itself had no effect on glucose oxidation (Figure 1). In the presence of insulin, 50 Ug/ml DiC8 significantly inhibited insulin action. The maximum ( ~ 60%) inhibition was obtained with 100 and 200 ug/ml of DiC8. DiC8 (100 ug/ml) did not alter basal glucose oxidation for up to an hour-long incubation (Figure 2). In presence of insulin, DiC8 inhibited CO2 metabolism at every time-point tested between 10 and 60 min to the same degree. No effect of DiC8 on insulin-stimulated CO2 metabolism,
Downloaded by: Boston University. Copyrighted material.
mM HEPES (pH 7.6). After incubation for 15 min at 4 °C the reaction was stopped, the phosphorylated material was immunoprecipitated with human anti-receptor antibody B-10 (1:100), phosphoproteins were reduced and supernatants analyzed by SDS-PAGE and autoradiography (Grunberger et al. 1984a). Scanning densitometry of the autoradiograms was analyzed.
Maria L. Terry, J. Levy and G. Grunberger
Fig. 1 Concentration dependence of the DiC8 effect on glucose metabolism. Glucose oxidation (see Methods) by isolated rat adipocytes was measured in the presence of DiC8 alone ( • ) or after insulin (6.67 nM) stimulation ( • ) . Illustrated are data from a representative experiment. Concentration of DiC8 greater than 50 ug/ml was required for a significant inhibition of insulin-stimulated glucose oxidation. DiC8 did not change glucose oxidation at any concentration.
Fig. 2 Time dependence of the effect of DiC8 on insulin-stimulated glucose oxidation by rat adipocytes. Time course of insulin-stimulated (A) glucose oxidation is compared to DiC8 (100 ug/ml) plus insulin (A) or DiC8 alone ( • ) . In a separate series of experiments there was no DiC8 effect on glucose oxidation over the initial 5 min. The data represent the mean±SEM of triplicates of two experiments.
Fig. 3 Effect of insulin on DiC8 inhibition of the insulin-stimulated glucose oxidation by rat adipocytes. An insulin dose response of glucose oxidation was assessed in the absence (O) and presence ( • ) of DiC8 (100 ug/ml) with various concentrations of unlabeled insulin (0.33-66.7 nM). The data represent the mean±SEM of triplicates of two experiments.
Fig. 4 Effect of DiC8 on insulin-stimulated lipogenesis by rat adipocytes. Lipogenesis was assessed in the absence (O) or presence ( • ) of DiC8 (100 ug/ml) over a spectrum of concentrations of unlabeled insulin (0-16.67 nM). The data are depicted as percent of basal (in absence of insulin) lipogenesis and represent the mean±SEM of duplicates of nine experiments.
however, was seen over the initial 10 min. The DiC8 inhibited the insulin-stimulated glucose oxidation at all concentrations of the hormone tested (0.33-66.7 nM). The inhibition (~ 50 %) of insulin-stimulated CO2 release was constant at all insulin concentrations (Figure 3).
effect (25 % inhibition) was smaller than its effect on either insulin-stimulated glucose oxidation or lipogenesis.
Specificity of the DiC8 effect on insulin action was next studied. DiC8 did not affect lipogenesis in the absence of insulin. However, DiC8 significantly inhibited the insulin-stimulated (0.0167—16.7 nM) lipogenesis at all insulin concentrations (Figure 4); the maximum inhibition (by 75 %) occurred with 16.7 nM insulin. Finally, DiC8 (100 ug/ml) alone did not affect 3-O-methylglucose transport but did inhibit the insulin-stimulated glucose transport (Table 2). The magnitude of this DiC8
Discussion If DAG's mimic insulin action, and activation of the receptor tyrosine kinase is essential for insulin effects, then DAG's should stimulate TKA of those receptors. If, however, the DAG effect is mediated by activation of protein kinase C or other postreceptor pathway (Standaert et al. 1988), DAG's could indirectly inhibit the TKA of insulin receptors. DiC8 stimulated phosphorylation of the insulin receptor's |3-subunit as well as TKA of the solubilized lectin-purified rat adipocyte plasma membranes. These findings were analogous to our observations with the mononuclear cell
Downloaded by: Boston University. Copyrighted material.
268 Horm. metab. Res. 23 (1991)
Horm. metab. Res. 23 (1991)
Diacylglycerols and Insulin Action
DiC8 stimulation of the TKA of receptor-enriched preparations argues against a direct involvement of protein kinase C. However, these results can not be automatically extrapolated to the bioassays carried out with intact cells. Direct effect of exogenous diacylglycerols on insulin-regulated bioeffects was studied (Christensen et al. 1987; Standaert et al. 1988). Two of the reasons for discrepancies in previous reports have been different manner of preparation of reagents and the methods of mixing labeled sugar with cell suspensions (Martz, Mookerjee and Jung 1986). To rule out these possibilities, we mimicked the methods used by Christensen et al. (1987) and Standaert et al. (1988), suspending DAG's in chloroform and assay buffer or dimethylsulfoxide, respectively. There were no quantitative differences in the DAG- and insulin-stimulated glucose oxidation. Another uncertainty in these experiments is the actual amount of exogenous DAG incorporated into the plasma membrane. Stralfors (1988) used sodium taurodeoxycholate to enhance the transfer of DAG to the plasma membrane. DiC8, even suspended in the bile salt, did not alter glucose uptake but it did stimulate phosphorylation of a 40-kD protein in intact adipocytes. Dissociation between effects of DAG on glucose uptake and on kinase activation was therefore implicated. DAG can alter lipid composition of membrane (Das and Rand 1984) and thus its viscosity and fluidity (Shinitzky and Henkart 1979). DiC8 effects in our study could be the result of such changes in membrane fluidity. The inhibition of insulin-stimulated receptor autophosphorylation by DiC8 was mirrored by inhibition of insulin-activated glucose oxidation, transport and lipogenesis. It remains unclear whether activation of protein kinase C by DiC8 was indirectly responsible for our observations. DAG's might play multiple roles in modulating insulin actions. They might be involved in both propagating and terminating of insulin signals depending on the ratio of "active" vs. "inactive" insulin receptors and relative abundance of other essential mediators (calcium, phospholipids, phosphorylated inositol-glycans, etc.). Abbreviations TPA DAG DiC8 OAG SBEM HEPES BSA
= = = = = = =
12-0-tetradecanoyl-P-phorbol-13-acetate diacylglycerol sn-l,2-dioctanoylglycerol sn-l-oleoyl-2-acetylglycerol sucrose-based extraction medium 4-(2-hydroxyethyl)-l-piperazineethane sulfonic acid bovine serum albumin
KRBH = Krebs-Ringer bicarbonate/HEPES buffer SDS-PAGE= sodium dodecyl sulfate-polyacrylamide gel electrophoresis TKA = tyrosine kinase activity Acknowledgements Portions of this work were presented at the 1986 Annual Meeting of the Midwestern Region of the American Federation for Clinical Research (AFCR) (Clin. Res. 34: 927A, 1986) and the 1987 National Annual Meeting of the AFCR (Clin. Res. 35: 518A, 1987). We greatly appreciate the secretarial assistance of Ms. Andrell Sturdivant. Part of this work was supported by a grant S07RR05384 (G. G.) from the National Institutes of Health. References Belsham, G. J., R. M. Denton, J. A. Tanner: Use of a novel rapid preparation of fat-cell plasma membranes employing Percoll to investigate the effects of insulin and adrenaline on membrane protein phosphorylation within intact fat cells. Biochem. J. 192: 457-467 (1980) Bradford, M. M.: A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of proteindye binding. Anal. Biochem. 72:248-254 (1976) Brautigan, D. L., J. D. Kuplic: Proposal for a pathway to mediate the metabolic effects of insulin. Int. J. Biochem. 20: 349-356 (1988) Christensen, R. L., D. L. Shade, C. B. Graves, J. M. McDonald: Evidence that protein kinase C is involved in regulating glucose transport in the adipocyte. Int. J. Biochem. 19:259-265 (1987) Cushman, S. W., L. J. Wardzala: Potential mechanism of insulin action on glucose transport in the isolated rat adipose cell. J. Biol. Chem. 255:4758-4762(1980) Das, S., R. P. Rand: Diacylglycerol causes major structural transitions in phospholipid bilayer membranes. Biochem. Biophys. Res. Commun. 124:491-496(1984) Duronio, V., S. Jacobs: The effect of protein kinase C inhibition on insulin receptor phosphorylation. Endocrinology 127: 481-487 (1990) Foley, J. E., S. W. Cushman, L. B. Salans: Glucose transport in isolated rat adipocytes with measurements of L-arabinose uptake. Am. J.Physiol. 234:E112-E119(1978) Gliemann, J., K. Osterlind, J. Vinten, S. Gammeltoft: A procedure for measurement of distribution spaces in isolated fat cells. Biochim. Biophys.Acta 286:1-9(1972) Grunberger, G., R. Comi, S. I. Taylor, P. Gorden: Tyrosine kinase activity of the insulin receptor of patients with Type A extreme insulin resistance: Studies with circulating mononuclear cells and cultured lymphocytes. J. Clin. Endocrinol. Metab. 59: 1152-1158 (1984b) Grunberger, G., B. Iacopetta, J.-L. Carpentier, P. Gorden: Tumor-promoting phorbol ester (TPA) stimulates tyrosine phosphorylation in U-937 monocytes. Diabetes 34:1364-1370 (1986) Grunberger, G., J. Levy: Diacylglycerols modulate phosphorylation of the insulin receptor from human mononuclear cells. Eur. J. Biochem. 187:191-198(1990) Grunberger, G,, Y. Tick, S. I. Taylor, P. Gorden: Tumor-promoting phorbol ester (TPA) stimulates tyrosine phosphorylation in U-937 monocytes. Proc. Natl. Acad. Sci. USA 81:2762-2766 (1984a) Karnieli, E., M. J. Zarnowski, P. J. Hissin, I. A. Simpson, L. B. Salans, S. W. Cushman: A possible mechanism of insulin resistance in the rat adipose cell in streptozotocin-induced diabetes mellitus, J. Biol. Chem.256:4772-4777(1981) Kikkawa, U., R. Minakuchi, Y. Takai, Y. Nishizuka: Calcium-activated, phospholipid-dependent protein kinase (protein kinase C) from rat brain. Meth. Enzymol. 99:288-298 (1983) Martz, A., B. K. Mookerjee, C. Y. Jung: Insulin and phorbol esters affect the maximum velocity rather than the half-saturation constant
Downloaded by: Boston University. Copyrighted material.
insulin receptors where predominant effect was on phosphorylation of tyrosine (Grunberger and Levy 1990). DAG and insulin combined decreased receptor autophosphorylation but not poly(Glu80Tyr20) phosphorylation compared to insulin alone, indicating substrate specificity. Glucose oxidation, lipogenesis and glucose transport yielded similar qualitative results: no effect of l,2-DiC8 in the absence of insulin but a significant inhibition of insulin-stimulated activity. This inhibitory effect of DiC8 was thus analogous to that on B-subunit autophosphorylation in the presence of insulin. However, this analogy does not imply a cause-and-effect relationship. DiC8 in this high concentration could have directly affected the bioassays surveyed as a detergent and not by modulating the receptor kinase activity.
269
Horm. metab. Res. 23 (1991) of 3-O-methylglucose transport in rat adipocytes. J. Biol. Chem. 261:13606-13609(1986) Moody, A. J., M. A. Stan, M. Stan, J. Gliemann: A simple free fat cell bioassay for insulin. Horm. Metab. Res. 6:12—16 (1974) Parker, P. J., L. Coussens, N. Totty, L. Rhee, S. Young, E. Chen, S. Stabel, M. D. Waterfleld, A. Ullrich: The complete primary structure of protein kinase C — the major phorbol ester receptor. Science 233:853-859(1986) Rodbell, M.: Metabolism of isolated fat cells. J. Biol. Chem. 239: 375380(1964) Salans, L. B., J. W. Dougherty: The effect of insulin upon glucose metabolism by adipose cells of different size. J.Clin. Invest. 50:1399— 1410(1971) Sharkey, N. A., K. L. Leach, P. M. Blumberg: Competitive inhibition by diacylglycerol of specific phorbol ester binding. Proc. Natl. Acad.Sci.USA81:607-610(1984) Shinitzky, M., P. Henkart: Fluidity of cell membranes - current concepts and trends. Int. Rev. Cytol. 60:121 -147 (1979)
Maria L. Terry, J. Levy and G. Grunberger Sonne, O., J. Gliemann: Insulin receptors of cultured human lymphocytes (IM-9): Lack of receptor-mediated degradation. J. Biol. Chem. 255:7449-7454 (1980) Standaert, M. L., R. V. Farese, D. R. Cooper, R. J. Pollet: Insulin-induced glycerolipid mediators and the stimulation of glucose transport in Bc3H-l myocytes. J. Biol. Chem. 263:8696-8705 (1988) Stralfors, P.: Insulin stimulation of glucose uptake can be mediated by diacylglycerol in adipocytes. Nature 335: 554—556 (1988)
Requests for reprints should be addressed to: George Grunberger, M. D. Diabetes Section UHC-4H4201 St. Antoine Detroit, MI 48201 (U. S. A.)
Downloaded by: Boston University. Copyrighted material.
270
Effect of 1,2-diacylglycerols on the insulin receptor function and insulin action in rat adipocytes was studied. 1,2-dioctanoylglycerol (100 ug/ml) did not alter insulin binding but it did stimulate phosphorylation of the (B-subunit of the insulin receptor as well as its tyrosine kinase activity. However, dioctanoylglycerol inhibited insulin-stimulated receptor autophosphorylation. This concentration of dioctanoylglycerol inhibited insulin-stimulated CO2 metabolism, lipogenesis and 3-O-methyl-glucose transport in a dose-dependent manner but did not alter any of these bioeffects in absence of insulin. While there was no direct link between diacylglycerol effect on tyrosine kinase activity of the insulin receptor and insulin action in rat adipocytes, the parallel inhibition of insulin-stimulated receptor autophosphorylation and insulin bioeffects by dioctanoylglycerol suggests its direct or indirect role in insulin signalling in rat fat cells. Key words Insulin Action — Insulin Receptor — Diacylglycerol — Protein Tyrosine Kinase
Introduction Transmission of the insulin signal from the initial binding of the hormone to its specific membrane receptor to the ultimate effects in the cell interior results from an interplay of several signalling mechanisms (Brautigan and Kuplic 1988). The serine- and threonine-specific protein kinase C (Kikkawa, Minakuchi, Takai and Nishizuka 1983) could modulate and/or terminate the insulin receptor tyrosine kinase signal (Duronio and Jacobs 1990). Tumor promoting phorbol diesters (e. g. TPA) and sn-1,2-diacylglycerols activate protein kinase C. Specific receptors for TPA copurify with protein kinase C and DAG's compete with TPA for binding to the phorbol ester ester receptors {Parker, Coussens, Totty, Rhee, Young, Chen, Stabel, Waterfleld and Ullrich 1986; Sharkey, Leach and Blumberg 1984). In addition to these classic effects of TPA and DAG's, these compounds modulated the function of the insulin receptors in cultured human mononuclear cells (Grunberger, Zick, Taylor and Gorden 1984a; Grunberger, lacopetta, Carpentier and Gorden 1986) which suggested a direct interaction of TPA and DAG's with the insulin receptor. In this report exogenous DAG stimulated the Horm.metab.Res. 23 (1991)266-270 © Georg Thieme Verlag Stuttgart -New York
tyrosine kinase activity (TKA) of the rat adipocyte insulin receptor but inhibited insulin-stimulated receptor autophosphorylation as well as several bioeffects of insulin. These results differed somewhat from past work (Christensen, Shade, Graves and McDonald 1987; Standaert, Farese, Cooper and Pollet 1988) where DAG's stimulated glucose uptake in absence of insulin. Materials and Methods Rat adipocytes were obtained from the epididymal fat pads of 6-8 male Sprague-Dawley rats (180-220 g). They were pooled and digested with collagenase (Rodbell 1964). Cell viability was assessed by the trypan blue exclusion. The cells were washed at 37 °C in a Krebs-Ringer bicarbonate buffer (pH 7.4; 30 mM HEPES with 1 % untreated BSA) and reduced to 10 mM HCO3. After evaporation of the DAG-chloroform solution under N2, the DAG's were resuspended in methanol (final concentration < 0.1 %; at this concentration methanol did not affect any of the assays described) before the addition of assay buffer (Grunberger et al. 1986). Specific 125I-insulin binding to intact cells was done as in Sonne and Gliemann (1980). Cells were incubated with 125I-insulin (0.1 nM) for 5,15,30,45 and 60 min at 37 °C in the presence or absence of unlabeled insulin (Gliemann, Osterlind, Vinten and Gammeltoft 1972). "Nonspecific" binding (with 16.7 uM insulin) was subtracted from total binding to give specific insulin binding. Plasma membranes were prepared (Belsham, Denton and Tanner 1980) with modifications. SBEM (0.25 M sucrose/10 mM Tris-HCl/2 mM EGTA, 10 mM sodium fluoride, pH 7.4) in the presence of PMSF (4 mM) and aprotinin (2 jig/ml), was added to washed cells. After homogenization the cell lysate was centrifuged (5 min at 1,000 x g). The infranate and pellet were aspirated from below the fat layer and centrifuged (30 min at 30,000 x g); the pellet was suspended in SBEM. Plasma membranes were obtained by centrifugation in the presence of a self-forming gradient of Percoll. The plasma membrane band was removed and washed two times (15 min at 10,000 x g) with a 5-fold dilution of a buffer (0.15 M NaCl/10 mM TrisHC1/1 mM EGTA, pH 7.4). The pellet was suspended in the buffer, solubilized in 1 % Triton X-100 in presence of protease inhibitors and applied to a wheat-germ agglutinin/agarose column. The receptor-enriched preparation was eluted with 0.3 M N-acetyl-D-glucosamine and 10% glycerol. Protein content of the solubilized material was determined {Bradford 1976). Autophosphorylation receptor B-subunit
of the
Volumes of solubilized partially-enriched adipocyte insulin receptors were adjusted to equal specific insulin binding capacity and preincubated in the absence or presence of insulin (0.1 uM), DiC8 (100 ug/ml) or both for 30 min at 22 °C as in Grunberger et al. (1984a). The phosphorylation reaction done in a final concentration of 3 mM Mn acetate, 20 mM MgCl2, 50 nM [y-32P ]-ATP, and 50 Received: 20 Apr. 1990
Accepted: 5 Oct. 1990 after revision
Downloaded by: Boston University. Copyrighted material.
Summary
Diacylglycerob and Insulin Action
Horm. metab. Res. 23 (1991)
267
Table 1 Effect of DiC8 on phosphorylation of the rat adipocyte insulin receptor. 32
Basal + DiC8 (100 ug/ml) + insulin (0.1 u.M) + DiC8 and insulin
P Content of p-subunit 0.80 ±0.08 2.81 ±0.14 15.71 ±0.34 8.27 ±0.64
Percent of basal
Tyrosine kinase activity
Percent of basal
100 351 1964 1034
2.23 ±0.76 6.04 ±0.73 15.13±1.05 15.70 ±0.81
100 271 678 704
32
P content of the receptor B-subunit (in arbitrary density units), obtained from scanning densitometry of autoradiograms, is expressed as mean±SEM of duplicates from two experiments. Tyrosine kinase activity is expressed in pmole 32P/mg (Glu80Tyr20 )/10 min/receptor as mean±SEM of quadruplicates from four experiments. See the Methods section for details.
Tyrosine kinase activity of the insulin receptor 80
20
Table 2 Insulin and DiC8-stimulated 3-O-methyl glucose transport.
Basal + DiC8 (100 ug/ml) + insulin (0.1 uM) + DiC8 and insulin
fmole/cell/min
Percent of basal
0.098 ±0.006 0.091 ±0.005 4.538 ±0.235 3.734 ±0.176*
100 93 4630 3810
Data represent the mean ± SEM of triplicates of three separate experiments. *The result compared to insulin alone was significantly different at P < 0.01 level.
Poly (Glu Tyr ) was phosphorylated in the absence or presence of DiC8 (100 |xg/ml), insulin (0.1 uM) or both (Grunberger, Comi, Taylor and Gorden 1984b). Results were expressed in tyrosine kinase units [pmol 32P/mg poly(Glu80Tyr20) at 22 °C in 10 min by a unit of specific insulin binding activity defined as the and phosphorylation of the B-subunit of the insulin receptor bound/free ratio from parallel specific 125I-insulin binding assays]. was measured. Insulin, a positive control, increased autophosphorylation of the receptor's B-subunit ~ 20-fold over the basal state (Table 1). DiC8 alone stimulated the phosphorylaGlucose oxidation tion of the insulin receptor ~ 3.5-fold. However, simultaneous Adipocytes in the KRBH buffer with 3 % BSA were presence of the D A G and insulin decreased receptor autoplaced in a shaking waterbath (40 osc/min) at 37 °C and pre-inphosphorylation compared to insulin alone (Table 1). cubated with insulin and/or DAG for 15 min prior to a 60 min incubation with 0.1 mM glucose and 0.1 uCi glucose-l-14C/ml (Salans and Insulin (0.1 uM) stimulated phosphorylation Dougherty 1971). Incubation with DAG's did not affect the cell viability or cause cell leakage compared to control incubations. of an artificial substrate, poly (Glu 80 Tyr20 ) by the partially purified insulin receptor preparations 7-fold (Table 1). DiC8 (100 ug/ml) also increased TKA of these lectin-purified prepLipogenesis arations, even though always less than insulin. Insulin and Lipogenesis was measured by the method of Moody, DiC8 together increased TKA as much as insulin alone. The Stan, Stan and Gliemann (1974). Fat cells (10 ) were incubated in 1 ml optimal effect of DiC8 on poly(Glu80Tyr20) phosphorylation Krebs-Ringer phosphate buffer with 1% BSA, [3-3H]glucose (0.1 was at 100 ug/ml, after a 15 min preincubation at room temuCi), 5 mM glucose and insulin (0-16.67 nM) in the absence or presperature (not shown). ence of 100 u.g/ml of DiC8. D-Glucose
transport
The rate of 3-O-methylglucose transport was determined as in Foley, Cushman and Salans (1978) and Karnieli, Zarnowski, Hissin, Simpson, Salans and Cushman (1981). Adipocytes (~ 5 x 16) in the KRBH buffer with 3 % BSA were preincubated with insulin (6.67 nM) and/or DAG (100 ug/ml) for 30 min. The cell suspension was pulsed with [14C]-3-0-methylglucose (0.1 mM, 58.2 uCi/u,mol) and [3H]-L-glucose (0.1 mM, 11.6 uCi/umol). Uptake was stopped by cytochalasin B (0.4 mM) (Cushman and Wardzala 1980). Statistical analyses of the differences between experimental conditions were done using t-test; significance was accepted as P < 0.05. Results DiC8 or OAG did not consistently alter any I-insulin binding parameters in intact rat adipocytes. Solubilized lectin-purified fat cell insulin receptors were then preincubated with DiC8 (100 ug/ml), insulin (0.1 uJVl) or both
We next assessed the effect of DiC8 on bioeffects of insulin. Glucose oxidation was measured by 14C-CO2 production by the adipocytes. Insulin (6.67 nM) stimulated CO2 metabolism 20.6 ± 2.6-fold. Both exogenous DAG's (1,2-0AG and l,2-DiC8) significantly decreased the insulin effect. In presence of DiC8 (100 (ug/ml) insulin increased glucose oxidation only by 9.0+ 1.2-fold. This inhibition of insulin-stimulated glucose oxidation was largely prevented when DAG's were washed away prior to addition of insulin. DiC8, a membrane-permeable synthetic glycerolipid resembling endogenous DAG, was more potent than OAG in inhibiting the insulin effect. DiC8 (5—1000 ug/ml) by itself had no effect on glucose oxidation (Figure 1). In the presence of insulin, 50 Ug/ml DiC8 significantly inhibited insulin action. The maximum ( ~ 60%) inhibition was obtained with 100 and 200 ug/ml of DiC8. DiC8 (100 ug/ml) did not alter basal glucose oxidation for up to an hour-long incubation (Figure 2). In presence of insulin, DiC8 inhibited CO2 metabolism at every time-point tested between 10 and 60 min to the same degree. No effect of DiC8 on insulin-stimulated CO2 metabolism,
Downloaded by: Boston University. Copyrighted material.
mM HEPES (pH 7.6). After incubation for 15 min at 4 °C the reaction was stopped, the phosphorylated material was immunoprecipitated with human anti-receptor antibody B-10 (1:100), phosphoproteins were reduced and supernatants analyzed by SDS-PAGE and autoradiography (Grunberger et al. 1984a). Scanning densitometry of the autoradiograms was analyzed.
Maria L. Terry, J. Levy and G. Grunberger
Fig. 1 Concentration dependence of the DiC8 effect on glucose metabolism. Glucose oxidation (see Methods) by isolated rat adipocytes was measured in the presence of DiC8 alone ( • ) or after insulin (6.67 nM) stimulation ( • ) . Illustrated are data from a representative experiment. Concentration of DiC8 greater than 50 ug/ml was required for a significant inhibition of insulin-stimulated glucose oxidation. DiC8 did not change glucose oxidation at any concentration.
Fig. 2 Time dependence of the effect of DiC8 on insulin-stimulated glucose oxidation by rat adipocytes. Time course of insulin-stimulated (A) glucose oxidation is compared to DiC8 (100 ug/ml) plus insulin (A) or DiC8 alone ( • ) . In a separate series of experiments there was no DiC8 effect on glucose oxidation over the initial 5 min. The data represent the mean±SEM of triplicates of two experiments.
Fig. 3 Effect of insulin on DiC8 inhibition of the insulin-stimulated glucose oxidation by rat adipocytes. An insulin dose response of glucose oxidation was assessed in the absence (O) and presence ( • ) of DiC8 (100 ug/ml) with various concentrations of unlabeled insulin (0.33-66.7 nM). The data represent the mean±SEM of triplicates of two experiments.
Fig. 4 Effect of DiC8 on insulin-stimulated lipogenesis by rat adipocytes. Lipogenesis was assessed in the absence (O) or presence ( • ) of DiC8 (100 ug/ml) over a spectrum of concentrations of unlabeled insulin (0-16.67 nM). The data are depicted as percent of basal (in absence of insulin) lipogenesis and represent the mean±SEM of duplicates of nine experiments.
however, was seen over the initial 10 min. The DiC8 inhibited the insulin-stimulated glucose oxidation at all concentrations of the hormone tested (0.33-66.7 nM). The inhibition (~ 50 %) of insulin-stimulated CO2 release was constant at all insulin concentrations (Figure 3).
effect (25 % inhibition) was smaller than its effect on either insulin-stimulated glucose oxidation or lipogenesis.
Specificity of the DiC8 effect on insulin action was next studied. DiC8 did not affect lipogenesis in the absence of insulin. However, DiC8 significantly inhibited the insulin-stimulated (0.0167—16.7 nM) lipogenesis at all insulin concentrations (Figure 4); the maximum inhibition (by 75 %) occurred with 16.7 nM insulin. Finally, DiC8 (100 ug/ml) alone did not affect 3-O-methylglucose transport but did inhibit the insulin-stimulated glucose transport (Table 2). The magnitude of this DiC8
Discussion If DAG's mimic insulin action, and activation of the receptor tyrosine kinase is essential for insulin effects, then DAG's should stimulate TKA of those receptors. If, however, the DAG effect is mediated by activation of protein kinase C or other postreceptor pathway (Standaert et al. 1988), DAG's could indirectly inhibit the TKA of insulin receptors. DiC8 stimulated phosphorylation of the insulin receptor's |3-subunit as well as TKA of the solubilized lectin-purified rat adipocyte plasma membranes. These findings were analogous to our observations with the mononuclear cell
Downloaded by: Boston University. Copyrighted material.
268 Horm. metab. Res. 23 (1991)
Horm. metab. Res. 23 (1991)
Diacylglycerols and Insulin Action
DiC8 stimulation of the TKA of receptor-enriched preparations argues against a direct involvement of protein kinase C. However, these results can not be automatically extrapolated to the bioassays carried out with intact cells. Direct effect of exogenous diacylglycerols on insulin-regulated bioeffects was studied (Christensen et al. 1987; Standaert et al. 1988). Two of the reasons for discrepancies in previous reports have been different manner of preparation of reagents and the methods of mixing labeled sugar with cell suspensions (Martz, Mookerjee and Jung 1986). To rule out these possibilities, we mimicked the methods used by Christensen et al. (1987) and Standaert et al. (1988), suspending DAG's in chloroform and assay buffer or dimethylsulfoxide, respectively. There were no quantitative differences in the DAG- and insulin-stimulated glucose oxidation. Another uncertainty in these experiments is the actual amount of exogenous DAG incorporated into the plasma membrane. Stralfors (1988) used sodium taurodeoxycholate to enhance the transfer of DAG to the plasma membrane. DiC8, even suspended in the bile salt, did not alter glucose uptake but it did stimulate phosphorylation of a 40-kD protein in intact adipocytes. Dissociation between effects of DAG on glucose uptake and on kinase activation was therefore implicated. DAG can alter lipid composition of membrane (Das and Rand 1984) and thus its viscosity and fluidity (Shinitzky and Henkart 1979). DiC8 effects in our study could be the result of such changes in membrane fluidity. The inhibition of insulin-stimulated receptor autophosphorylation by DiC8 was mirrored by inhibition of insulin-activated glucose oxidation, transport and lipogenesis. It remains unclear whether activation of protein kinase C by DiC8 was indirectly responsible for our observations. DAG's might play multiple roles in modulating insulin actions. They might be involved in both propagating and terminating of insulin signals depending on the ratio of "active" vs. "inactive" insulin receptors and relative abundance of other essential mediators (calcium, phospholipids, phosphorylated inositol-glycans, etc.). Abbreviations TPA DAG DiC8 OAG SBEM HEPES BSA
= = = = = = =
12-0-tetradecanoyl-P-phorbol-13-acetate diacylglycerol sn-l,2-dioctanoylglycerol sn-l-oleoyl-2-acetylglycerol sucrose-based extraction medium 4-(2-hydroxyethyl)-l-piperazineethane sulfonic acid bovine serum albumin
KRBH = Krebs-Ringer bicarbonate/HEPES buffer SDS-PAGE= sodium dodecyl sulfate-polyacrylamide gel electrophoresis TKA = tyrosine kinase activity Acknowledgements Portions of this work were presented at the 1986 Annual Meeting of the Midwestern Region of the American Federation for Clinical Research (AFCR) (Clin. Res. 34: 927A, 1986) and the 1987 National Annual Meeting of the AFCR (Clin. Res. 35: 518A, 1987). We greatly appreciate the secretarial assistance of Ms. Andrell Sturdivant. Part of this work was supported by a grant S07RR05384 (G. G.) from the National Institutes of Health. References Belsham, G. J., R. M. Denton, J. A. Tanner: Use of a novel rapid preparation of fat-cell plasma membranes employing Percoll to investigate the effects of insulin and adrenaline on membrane protein phosphorylation within intact fat cells. Biochem. J. 192: 457-467 (1980) Bradford, M. M.: A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of proteindye binding. Anal. Biochem. 72:248-254 (1976) Brautigan, D. L., J. D. Kuplic: Proposal for a pathway to mediate the metabolic effects of insulin. Int. J. Biochem. 20: 349-356 (1988) Christensen, R. L., D. L. Shade, C. B. Graves, J. M. McDonald: Evidence that protein kinase C is involved in regulating glucose transport in the adipocyte. Int. J. Biochem. 19:259-265 (1987) Cushman, S. W., L. J. Wardzala: Potential mechanism of insulin action on glucose transport in the isolated rat adipose cell. J. Biol. Chem. 255:4758-4762(1980) Das, S., R. P. Rand: Diacylglycerol causes major structural transitions in phospholipid bilayer membranes. Biochem. Biophys. Res. Commun. 124:491-496(1984) Duronio, V., S. Jacobs: The effect of protein kinase C inhibition on insulin receptor phosphorylation. Endocrinology 127: 481-487 (1990) Foley, J. E., S. W. Cushman, L. B. Salans: Glucose transport in isolated rat adipocytes with measurements of L-arabinose uptake. Am. J.Physiol. 234:E112-E119(1978) Gliemann, J., K. Osterlind, J. Vinten, S. Gammeltoft: A procedure for measurement of distribution spaces in isolated fat cells. Biochim. Biophys.Acta 286:1-9(1972) Grunberger, G., R. Comi, S. I. Taylor, P. Gorden: Tyrosine kinase activity of the insulin receptor of patients with Type A extreme insulin resistance: Studies with circulating mononuclear cells and cultured lymphocytes. J. Clin. Endocrinol. Metab. 59: 1152-1158 (1984b) Grunberger, G., B. Iacopetta, J.-L. Carpentier, P. Gorden: Tumor-promoting phorbol ester (TPA) stimulates tyrosine phosphorylation in U-937 monocytes. Diabetes 34:1364-1370 (1986) Grunberger, G., J. Levy: Diacylglycerols modulate phosphorylation of the insulin receptor from human mononuclear cells. Eur. J. Biochem. 187:191-198(1990) Grunberger, G,, Y. Tick, S. I. Taylor, P. Gorden: Tumor-promoting phorbol ester (TPA) stimulates tyrosine phosphorylation in U-937 monocytes. Proc. Natl. Acad. Sci. USA 81:2762-2766 (1984a) Karnieli, E., M. J. Zarnowski, P. J. Hissin, I. A. Simpson, L. B. Salans, S. W. Cushman: A possible mechanism of insulin resistance in the rat adipose cell in streptozotocin-induced diabetes mellitus, J. Biol. Chem.256:4772-4777(1981) Kikkawa, U., R. Minakuchi, Y. Takai, Y. Nishizuka: Calcium-activated, phospholipid-dependent protein kinase (protein kinase C) from rat brain. Meth. Enzymol. 99:288-298 (1983) Martz, A., B. K. Mookerjee, C. Y. Jung: Insulin and phorbol esters affect the maximum velocity rather than the half-saturation constant
Downloaded by: Boston University. Copyrighted material.
insulin receptors where predominant effect was on phosphorylation of tyrosine (Grunberger and Levy 1990). DAG and insulin combined decreased receptor autophosphorylation but not poly(Glu80Tyr20) phosphorylation compared to insulin alone, indicating substrate specificity. Glucose oxidation, lipogenesis and glucose transport yielded similar qualitative results: no effect of l,2-DiC8 in the absence of insulin but a significant inhibition of insulin-stimulated activity. This inhibitory effect of DiC8 was thus analogous to that on B-subunit autophosphorylation in the presence of insulin. However, this analogy does not imply a cause-and-effect relationship. DiC8 in this high concentration could have directly affected the bioassays surveyed as a detergent and not by modulating the receptor kinase activity.
269
Horm. metab. Res. 23 (1991) of 3-O-methylglucose transport in rat adipocytes. J. Biol. Chem. 261:13606-13609(1986) Moody, A. J., M. A. Stan, M. Stan, J. Gliemann: A simple free fat cell bioassay for insulin. Horm. Metab. Res. 6:12—16 (1974) Parker, P. J., L. Coussens, N. Totty, L. Rhee, S. Young, E. Chen, S. Stabel, M. D. Waterfleld, A. Ullrich: The complete primary structure of protein kinase C — the major phorbol ester receptor. Science 233:853-859(1986) Rodbell, M.: Metabolism of isolated fat cells. J. Biol. Chem. 239: 375380(1964) Salans, L. B., J. W. Dougherty: The effect of insulin upon glucose metabolism by adipose cells of different size. J.Clin. Invest. 50:1399— 1410(1971) Sharkey, N. A., K. L. Leach, P. M. Blumberg: Competitive inhibition by diacylglycerol of specific phorbol ester binding. Proc. Natl. Acad.Sci.USA81:607-610(1984) Shinitzky, M., P. Henkart: Fluidity of cell membranes - current concepts and trends. Int. Rev. Cytol. 60:121 -147 (1979)
Maria L. Terry, J. Levy and G. Grunberger Sonne, O., J. Gliemann: Insulin receptors of cultured human lymphocytes (IM-9): Lack of receptor-mediated degradation. J. Biol. Chem. 255:7449-7454 (1980) Standaert, M. L., R. V. Farese, D. R. Cooper, R. J. Pollet: Insulin-induced glycerolipid mediators and the stimulation of glucose transport in Bc3H-l myocytes. J. Biol. Chem. 263:8696-8705 (1988) Stralfors, P.: Insulin stimulation of glucose uptake can be mediated by diacylglycerol in adipocytes. Nature 335: 554—556 (1988)
Requests for reprints should be addressed to: George Grunberger, M. D. Diabetes Section UHC-4H4201 St. Antoine Detroit, MI 48201 (U. S. A.)
Downloaded by: Boston University. Copyrighted material.
270
