Contributed by Daniel E. Koshland, Jr., April 23, 1986. ABSTRACT ... protein kinase C does not affect its insulin-binding activity but does inhibit by 65% the receptor's ..... Cooper, J. A.,Sefton, B. M. & Hunter, T. (1983) Methods. Enzymol.
Proc. Natl. Acad. Sci. USA Vol. 83, pp. 5822-5824, August 1986 Biochemistry
Protein kinase C directly phosphorylates the insulin receptor in vitro and reduces its protein-tyrosine kinase activity (signal transduction/second messengers/phorbol esters/diabetes)
GIDEON E. BOLLAG*, RICHARD A. ROTHt, JACQUELINE BEAUDOINt, DARIA MOCHLY-ROSEN*, AND DANIEL E. KOSHLAND, JR.* *Department of Biochemistry, University of California, Berkeley, Berkeley, CA 94720; and tDepartment of Pharmacology, Stanford University School of Medicine, Stanford, CA 94305
Contributed by Daniel E. Koshland, Jr., April 23, 1986
The (3 subunit of purified insulin receptor is ABSTRACT phosphorylated on a serine residue by purified preparations of protein kinase C (ATP: protein phosphotransferase, EC 2.7.1.37). This phosphorylation is inhibited by antibodies to protein kinase C and stimulated by phospholipids, diacylglycerol, and Ca2+. The phosphorylation of the receptor by protein kinase C does not affect its insulin-binding activity but does inhibit by 65% the receptor's intrinsic tyrosine-specific protein kinase activity (ATP: protein-tyrosine O-phosphotransferase, EC 2.7.1.112). These results indicate that activators of protein kinase C, such as phorbol esters, desensitize cells to insulin by direct protein kinase C action on the insulin receptor.
sulfate fractionation, and AcA34 Ultrogel chromatography (11) and had a specific activity of 20-30 nmol of phosphate incorporated into H1 histone per min/mg of protein at 30'C (units/mg). In some experiments, protein kinase C was more extensively purified on DEAE-cellulose, phenyl-Sepharose, and phosphatidylserine columns (12). Monoclonal antibodies to the insulin receptor (24B7, inhibiting; MC51, noninhibiting) (13) and protein kinase C were purified by protein A-Sephadex chromatography. Phosphoamino acids and poly(Glu4,Tyr) (random copolymer, sodium glutamate/tyrosine, 4:1) were from Sigma. Methods. Phosphorylations were carried out as described in the figure legends. Samples were electrophoresed on NaDodSO4/polyacrylamide gels and visualized by autoradiography. Phosphoamino acid analysis was carried out as described (14). Insulin binding and protein-tyrosine kinase assays were as described in the figure legends. The monoclonal antibody (15B5) used in the insulin-binding experiments has been shown to recognize both phosphorylated and nonphosphorylated receptors (13).
The transduction of extracellular signals via membrane receptors often involves second messengers such as cAMP, Ca2+ fluxes, and inositol phospholipid turnover. The simplistic notion that a receptor is activated and then transmits a signal to the output response of the cell is no longer valid. It is becoming clear that complex interactions among multiple signalling systems can result in synergistic or antagonistic responses (1, 2). An example is the interaction between the response to insulin, whose receptor possesses tyrosinespecific protein kinase activity (ATP: protein-tyrosine 0phosphotransferase, EC 2.7.1.112) (3), and the protein kinase C system (ATP: protein phosphotransferase, EC 2.7.1.37), which can be activated by phorbol esters (1). In vivo experiments indicate that phorbol ester activation of protein kinase C stimulates the phosphorylation of serine in the insulin receptor (4-6) and antagonizes the ability of cells to respond to insulin. However, attempts to demonstrate in vitro interactions between protein kinase C and the insulin receptor have met with failure (7, 8). To clarify this question, we have investigated whether direct phosphorylation of the insulin receptor by protein kinase C can occur. Demonstration of an in vitro reaction would be important because it would indicate that direct modification by protein kinase C can occur in the absence of additional enzyme systems. Since more intricate responses to phorbol esters have also been observed (9), this direct reaction may allow the clarification of indirect effects of protein kinase C. An in vitro reaction provides a tool to study the effects of activators, inhibitors, and cofactors without the ambiguities that complex systems present.
RESULTS In the phosphorylation experiments, purified protein kinase C was incubated with purified insulin receptor in the presence of adenosine 5'-[y-32P]triphosphate under a variety of conditions. The reactions were then analyzed by NaDodSO4 gel electrophoresis and autoradiography. Whenever both proteins were included in the reaction mixture, phosphorylation of a 95-kDa protein and an 80-kDa protein was observed (Fig. 1). The 95-kDa protein was identified as the insulin receptor P subunit by immunoprecipitation with monoclonal antibodies to the receptor. The phosphorylation of protein kinase C (11) was also observed in these reactions, as evidenced by the labeled 80-kDa band. Since the insulin receptor is also a protein kinase that can autophosphorylate its 8 subunit (3), a number of tests were performed to test whether additional phosphorylation of the 95-kDa , subunit was due to protein kinase C. First, when the reaction was performed under conditions that inhibited insulin receptor autophosphorylation, the phosphorylation of the 95-kDa protein was still observed. This occurred (i) in the absence of insulin and Mn2+ (Fig. 1 Right), (ii) after pretreating the receptor at 37°C to inactivate its kinase activity, (iii) after pretreating the receptor with unlabeled ATP, and (iv) in the presence of monoclonal antibodies that inhibit the insulin receptor kinase activity (Fig. 1 Left). The 95-kDa band was not observed when protein kinase C was phosphorylated in the absence of insulin receptor (Fig. 1 Right). Second, phosphorylation of the 95-kDa protein was stimulated by
MATERIALS AND METHODS Materials. Insulin receptor was purified on a monoclonal antibody affinity column (10). In most experiments, protein kinase C was purified by DEAE chromatography, ammonium The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. §1734 solely to indicate this fact.
Abbreviation: EGF, epidermal growth factor.
5822
Biochemistry: Boflag et al. PKC Ca2+/PL Anti-InsR Anti-PKC InsR-PKC
+
+
+ +
+ -
+ -
-
-
-
a
U
+ + -
--
+
+
+
+ +
Proc. Natl. Acad. Sci. USA 83 (1986) + + + + + +
a
-
''I
Thr(P)-* Ser(P) af b c d e f
g-
InsRPKC-
Tyr(P)
a
5823
g
-CaK
9
'I
h
a
b
c
d
e f
InsR/auto InsR/PKC PKC/auto
FIG. 1. (Left) Phosphorylation of the insulin receptor (InsR) by protein kinase C (PKC) and inhibition by monoclonal antibodies. One microgram of InsR and, where indicated, 2 x 10-3 units of PKC were incubated for 30 min at 30'C with -0.5 Ag each of anti-InsR and anti-PKC antibodies. Where indicated, phospholipid (PL) (phosphatidylserine at 0.2 mg/ml and diolein at 40 ,g/ml) and 2 mM CaCl2 were added to 30 A.l (final volume) containing 20 mM MgCl2, 7 A.M [y-32P]ATP (20 Ci/mmol; 1 Ci = 37 GBq), 0.4 mg of ovalbumin per ml, 0.005% Triton X-100, 3 mM Tris HCl (pH 7.5), 0.1 mM EDTA, 0.1 mM EGTA, and 7 mM 2-mercaptoethanol, and the reaction mixtures were incubated 20 min at 30°C. A control lane with autophosphorylation of PKC in the absence of InsR did not reveal a band comigrating with the InsR band of lane c. In the grid above, a plus signifies addition of inhibiting antibody, while a minus indicates addition of noninhibiting antibody. (Center) Phosphoamino acid analysis of labeled proteins. The indicated protein bands were cut out of the dried gel (separate experiment, 4 x 10-2 units of PKC), rehydrated, homogenized, eluted, and precipitated. Protein was hydrolyzed for 1 hr in 6 M HCl at 110°C before resolving the labeled amino acids by thin-layer chromatography on silica gel (0.25 mm, 20 x 20 cm, aluminum support) in isobutyric acid/0.5 M NH40H, 5:3 (vol/vol), and visualizing by autoradiography. Unlabeled phosphoamino acids also were run and stained with ninhydrin as markers. InsR/auto, InsR autophosphorylation (Left, lane a); InsR/PKC, InsR phosphorylated by PKC (Left, lane c, top band); PKC/auto, PKC autophosphorylation (Left, lane c, lower band). (Right) Lack of phosphorylation of the InsR by Ca2+/calmodulin-dependent protein kinase (CaK). InsR was incubated with PKC (lane b), PKC plus 3 mM Ca2+ and PL (lane c), alone (lane d), and in the absence (lane e) or presence (lane f) of Ca2 , calmodulin, and CaK. PKC phosphorylation in the absence of InsR (control) was also performed (lane a). This PKC was purified on DEAE-cellulose, phenyl-Sepharose and phosphatidylserine columns (specific activity of 18 Mmol/min per mg of protein). Phosphorylations were performed for 1 hr at 24°C in 25 ,ul (total volume) of buffer A (0.02% Triton X-100/10 mM Hepes, pH 7.6/30 mM NaCl/15 mM MgC12/10% glycerol/2 mM Tris HCl/0.1 mM EDTA/5 mM 2-mercaptoethanol) containing 10 ,uM [y-32P]ATP (20 Ci/mmol). Autophosphorylated CaK (lane f) is at 55 kDa.
phospholipids and Ca2", which are activators of protein kinase C (Fig. 1). Third, the reaction was inhibited by monoclonal antibodies that inhibit protein kinase C activity. Finally, the phosphorylation of the 95-kDa protein was shown to be primarily on serine (Fig. 1 Center). Since the insulin receptor only phosphorylates tyrosine residues (ref. 3; Fig. 1 Center, left lane), whereas protein kinase C phosphorylates proteins on serine and threonine (ref. 11; Fig. 1 Center, right lane), the phosphorylation must be due to protein kinase C. Phosphorylation of the insulin receptor by protein kinase C was relatively specific. For example, a Ca2+/calmodulindependent kinase (15) did not phosphorylate the receptor (Fig. 1 Right). The phosphorylation did not depend on specific protein preparations, as we observed in more than 20 different experiments with three different protein kinase C preparations and three different receptor preparations, including receptor from both human placenta as well as rat liver. In addition, the phosphorylation reaction was essentially stoichiometric, with phosphorylation of almost every receptor molecule (0.5-1.5 phosphates incorporated per receptor). Finally, the same results were obtained in both the presence and absence of insulin. We then examined the effect of protein kinase C phosphorylation on the receptor's activities-namely, insulin binding and tyrosine phosphorylation. Protein kinase C modification of the receptor was found to have no effect on its ability to bind 125I-labeled insulin (Fig. 2), either at low or high insulin concentrations. In contrast, phosphorylation by protein kinase C did inhibit the receptor's ability to phosphorylate an exogenous substrate, poly(Glu4,Tyr). A 65% decrease in the
0
20
x
E0. 0
z
0 m z
10
-J C6
z
N
V-
0 0
" 10-10
10-9
10-8
10-7
INSULIN (M) FIG. 2. Effect of protein kinase C phosphorylation on the insulin-binding activity of the receptor. Insulin receptor (-1 ug) was incubated for 30 min at 24°C with (e) or without (o) protein kinase C in 25 1.d (total volume) of buffer A containing 3 mM CaC12, 50 ,uM ATP, 0.2 mg of phosphatidylserine per ml, and 40 ,g of diolein per ml. The reaction was stopped by adding 150 ,ul of 50 mM Hepes, pH 7.6/150 mM NaCl/0.05% Triton X-100/1 mM N-ethylmaleimide. Aliquots (25 Ml) were adsorbed to microtiter wells coated with a monoclonal antibody (15B5) to the receptor. After 2 hr at 24°C, the wells were washed three times, and receptor adsorbed to the wells was incubated with 20,000 cpm of '251-labeled insulin ("25I-insulin) in the presence of the indicated concentrations of unlabeled insulin. After an additional 2 hr at 24°C, wells were washed three times and assayed for radioactivity. Results shown are averages of three
experiments.
Biochemistry: BoRag et al.
5824
rate of phosphorylation was observed in both the presence and absence of insulin (Fig. 3), indicating that this effect was
not due to an inhibition of insulin binding.
DISCUSSION The present studies show that protein kinase C can act directly on the insulin receptor to phosphorylate it on serine residues and reduce its kinase activity. These findings indicate that the in vivo action of phorbol esters to desensitize cells to insulin (5, 6) probably occurs via the same mechanism, a consequence of protein kinase C modification of the receptor. The phosphorylation acts to reduce the signal emanating from the receptor and apparently does not inhibit insulin binding. [Although a few reports have claimed that phorbol esters decrease insulin binding to intact cells (17, 24), most have not found such an effect (5, 6, 8)]. Modification by protein kinase C has also been reported for the receptors of epidermal growth factor (EGF) (18, 19), transferrin (20), and catecholamines (21); however, the effects are clearly different. Unlike the insulin receptor, the modified EGF receptor apparently loses affinity for its ligand, EGF. In addition, the modified residue appears to lie between the membrane and the EGF receptor active site (7). Since an antibody that blocks the intrinsic insulin receptor kinase activity (13) does not inhibit the protein kinase C phosphorylation (Fig. 1 Left), the modified residue in the insulin receptor is distal to this antibody binding site. While the protein-tyrosine kinase active sites for the insulin and EGF receptors seem to share homologies (7), protein kinase C apparently modulates their actions in alternate ways. Fur0
x
T-
E
0.
w
UJ z
w
z
0
TIME (min) FIG. 3. Effect of protein kinase C phosphorylation on the protein-tyrosine kinase activity of the insulin receptor. The insulin receptor was phosphorylated by protein kinase C and adsorbed to microtiter wells as described in the legend to Fig. 2. The kinase activity of the adsorbed receptor was assayed by adding 3 mM MnCl2, 5 mg of poly(Glu4,Tyr) per ml, 10 ,uM [y-32P]ATP (0.5 Ci/mmol), and, where indicated, 1 AM insulin. After 10, 20, or 40 min at 240C, 5-ILI aliquots of the supernatants from the wells were spotted onto Whatman paper and precipitated with 10% trichloroacetic acid. The papers were washed and assayed for radioactivity. The same result was obtained by electrophoresing the samples, cutting out the poly(Glu4,Tyr) bands and assaying them for radioactivity. The experiment was repeated twice. A, Insulin-receptor; o, insulin receptor with insulin; e, insulin receptor with insulin and protein kinase C; A, insulin receptor with protein kinase C (no insulin).
Proc. Natl. Acad. Sci. USA 83 (1986) thermore, phorbol esters activate internalization of transferrin receptors, suggesting that protein kinase C may also induce receptor sequestration. Therefore, it seems there is no universal mechanism by which protein kinase C modulates signal transduction. The effect of protein kinase C on insulin action is particularly intriguing, since tissues from animals and patients with diabetes mellitus have characteristics similar to those of phorbol ester-treated cells. Whereas these diabetic tissues exhibit little or no change in insulin binding, the metabolic responses to insulin and the insulin receptor protein-tyrosine kinase activity are diminished (16, 22, 23). Since a variety of hormones can activate protein kinase C (1), it may be that the desensitization associated with diabetes involves protein kinase C activation by one of these agents. The availability of an in vitro system mimicking desensitization should provide a useful tool for clarifying the roles of insulin and protein kinase C in this process. We thank H. Schulman and C. Csernansky for gifts of Ca2+/calmodulin-dependent kinase and protein kinase C. This work was supported by research grants from the National Institutes of Health and the National Science Foundation, a National Institutes of Health Research Career Development Award to R.A.R., and a National Science Foundation Predoctoral Fellowship to G.E.B.
Nishizuka, Y. (1984) Nature (London) 308, 693-698. Berridge, M. J. (1985) Sci. Am. 253, 142-152. Kahn, C. R. (1985) Annu. Rev. Med. 36, 429-451. Jacobs, S., Sahyoun, N. E., Saltiel, A. R. & Cuatrecasas, P. (1983) Proc. Nati. Acad. Sci. USA 80, 6211-6213. 5. Takayama, S., White, M. F., Lauris, V. & Kahn, C. R. (1984) Proc. Natl. Acad. Sci. USA 81, 7797-7801. 6. Van de Werve, G., Proietto, J. & Jeanrenaud, B. (1985)
1. 2. 3. 4.
Biochem. J. 225, 523-527. 7. Hunter, T. (1985) Nature (London) 313, 740-741. 8. Jacobs, S., May, S., Watson, S., Lapetina, E. & Cuatrecasas, P. (1985) in Cancer Cells 3: Growth Factors and Transformation, eds. Feramisco, J., Ozanne, B. & Stiles, C. (Cold Spring Harbor Laboratory, Cold Spring Harbor, NY), pp. 139-143. 9. Koontz, J. W. & Goodman, S. (1985) Biochem. Biophys. Res. Commun. 131, 815-820. 10. Roth, R. A. & Cassell, D. J. (1983) Science 219, 299-301. 11. Kikkawa, U., Takai, Y., Minakuchi, R., Inohara, S. & Nishizuka, Y. (1982) J. Biol. Chem. 257, 13341-13348. 12. Gould, K. L., Woodgett, J. R., Cooper, J. A., Buss, J. E., Shalloway, D. & Hunter, T. (1985) Cell 42, 849-857. 13. Morgan, D. O., Ho, L., Korn, L. J. & Roth, R. A. (1986) Proc. Natl. Acad. Sci. USA 83, 328-332. 14. Cooper, J. A., Sefton, B. M. & Hunter, T. (1983) Methods Enzymol. 99, 387-402. 15. Kuret, J. & Schulman, H. (1984) Biochemistry 23, 5495-5504. 16. Reaven, G. M., Chen, Y.-D. I., Donner, C. C., Fraze, E. & Hollenbeck, C. B. (1985) J. Clin. Endocrinol. Metab. 61, 32-36. 17. Thomopoulos, P., Testa, U., Gourdin, M.-F., Hervy, C., Titeux, M. & Vainchenker, W. (1982) Eur. J. Biochem. 129, 389-393. 18. Cochet, C., Gill, G. N., Meisenhelder, J., Cooper, J. A. & Hunter, T. (1984) J. Biol. Chem. 259, 2553-2558. 19. Downward, J., Waterfield, M. D. & Parker, P. J. (1985) J. Biol. Chem. 260, 14538-14546. 20. May, W. S., Jacobs, S. & Cuatrecasas, P. (1984) Proc. Natl. Acad. Sci. USA 81, 2016-2020. 21. Sibley, D. R. & Lefkowitz, R. J. (1985) Nature (London) 317, 124-129. 22. Le Marchand-Brustel, Y., Grdmeaux, T., Ballotti, R. & Van Obberghen, E. (1985) Nature (London) 315, 676-678. 23. Kadowaki, T., Kasuga, M., Akanuma, Y., Ezaki, 0. & Takaku, F. (1984) J. Biol. Chem. 259, 14208-14216. 24. Grunberger, G. & Gorden, P. (1982) Am. J. Physiol. 243, E319-E324.
