Role of Insulin Receptor Phosphorylation in the Insulinomimetic Effects of Hydrogen Peroxide (Insulin Action/Hexose Transport/Tyrosine Kinase) G

Proc. Nadl. Acad. Sci. USA Vol. 84, pp. 8115-8119, November 1987 Medical Sciences

Role of insulin receptor phosphorylation in the insulinomimetic effects of hydrogen peroxide (insulin action/hexose transport/tyrosine kinase) G. R. HAYES AND D. H. LOCKWOOD Endocrine Metabolism Unit, Department of Medicine, University of Rochester School of Medicine and Dentistry, Rochester, NY 14642 Communicated by Pedro Cuatrecasas, July 24, 1987 (received for review June 30, 1987)

ABSTRACT The oxidant H202 has many insulin-like ef- hormone-stimulated lipolysis (11). However, H202 has no fects in rat adipocytes. To determine whether these effects effect on insulin binding (12). This has led to its use as a tool could be mediated by the tyrosine kinase activity of the insulin to evaluate postreceptor alterations in insulin action (13). receptor, the ability of H202 to stimulate receptor phospho- Since tyrosine kinase activity is intrinsic to the insulin rylation in intact adipocytes and partially purified insulin receptor, we have approached the question of the role of this receptors has been examined. Phosphorylation of the I8 subunit activity in insulin action by using this oxidant, which appar- of the insulin receptor was increased 2-fold by treatment of ently does not interact with the insulin-binding subunit ofthe intact cells with 3 mM H202, a concentration that maximally insulin receptor. In this report we have asked whether the stimulates 2-deoxyglucose uptake. Stimulation of receptor insulin-like effects of H202 are mediated by phosphorylation phosphorylation was rapid, reaching maximal levels within 5 of the insulin receptor. To this end, the action of H202 on min, and preceded activation of glucose transport. insulin receptor phosphorylation and tyrosine kinase activity Phosphoamino acid analysis of insulin receptors from H202- in intact adipocytes, cell homogenates, plasma membranes, treated adipocytes showed that 32p incorporation into and partially purified receptors has been examined. phosphotyrosine and phosphoserine residues of the .3 subunit was enhanced. Furthermore, partially purified receptors from H202-treated cells exhibit increased tyrosine kinase activity, as MATERIALS AND METHODS measured by phosphorylation of the peptide Glu8oTyr20. In Preparation of Adipocytes. Adipocytes from epididymal fat contrast, the direct addition of H202 to partially purified pads of male Sprague-Dawley rats (180-220 g, Charles River insulin receptors did not stimulate tyrosine kinase activity or Breeding Laboratories) were prepared by collagenase diges- insulin receptor autophosphorylation. This was not due to tion (14) in Krebs/Ringer phosphate buffer (pH 7.4) contain- breakdown of H202 or oxidation of ATP or the required ing 3% (wt/vol) bovine serum albumin (Sigma) and 1 mg of divalent cations. To define the factors involved in H202's effect, ml we have examined receptor phosphorylation in fat cell collagenase per (Cooper Biomedical, Malverne, PA). homogenates and purified plasma membranes. Although insu- Cells were washed and suspended in phosphate-free Krebs/ lin stimulated receptor phosphorylation in both of these sys- Ringer bicarbonate buffer supplemented with 25 mM Hepes tems, H202 was only effective in the cell homogenates. These (pH 7.4) containing 1% bovine serum albumin. data demonstrate that, under certain conditions, H202 stimu- 12-5-Labeled Insulin (125I-Insulin)-Binding Studies. 125I-in- lates insulin receptor phosphorylation and tyrosine kinase sulin (1 Ci/,mol; 1 Ci = 37 GBq) was prepared by the activity, suggesting that the insulin-like effects of H202 may be chloramine-T method and purified by chromatography on mediated by stimulation of insulin receptor phosphorylation. Sephadex G-50 (15). Partially purified receptors were incu- This does not appear to be a direct effect of H202 on the insulin bated with 1251-insulin (0.1 nM) at 4°C overnight (16). Binding receptor and requires nonplasma membrane cellular constit- was terminated by addition of bovine gamma globulin and uents. receptor-bound insulin was separated from free hormone by precipitation with 10% polyethylene glycol (16). Nonspecific In intact cells (1) and cell-free systems (2) the insulin receptor binding of 125I-insulin was defined as the amount of radioac- is rapidly phosphorylated subsequent to insulin binding. In tivity not displaced by 0.83 ,uM native insulin. cell-free systems phosphorylation occurs exclusively on 2-Deoxyglucose Transport Assay. 2-Deoxyglucose trans- tyrosine residues (2), whereas in intact cells serine is also port was measured using 0.1 mM D-2-deoxy[1-3H]glucose phosphorylated (1). The f subunit of the receptor contains an (10.4 mCi/mmol) and the indicated concentrations of insulin ATP binding site (3) and evidence indicates that the insulin or H202 as described (13), except that incubation was for 90 receptor itself is a tyrosine-specific protein kinase that sec and cytochalasin B (50 ,uM) was used to correct for undergoes autophosphorylation in an insulin-dependent man- extracellular trapping. ner (4). Receptor autophosphorylation enhances the tyrosine In Situ Phosphorylation. Isolated adipocytes (1 ml; 3 X kinase activity toward exogenous substrates (5). The intrinsic 106 cells) were incubated for 2 hr at 37°C with 0.5 mCi of nature of this kinase activity, as well as the insulin concen- 32P04. Cells were then incubated at 37°C in the absence or tration dependence (6) and time course of autophosphoryla- presence of insulin (0.1 ,uM; 5 min) or H202 (3 mM; 10 min). tion (7), has led to the proposal that receptor phosphorylation The buffer was removed and the cells were solubilized in 50 is an early step in coupling hormone binding to insulin action. mM Tris*HCl (pH 7.4) containing 1% Triton X-100, phospha- The oxidant H202 has many insulin-like effects in rat tase, and protease inhibitors (17). Insulin receptors were adipocytes, including stimulation of glucose transport (8, 9), partially purified on wheat germ agglutinin (WGA)-Sepha- stimulation of glucose utilization (10), and inhibition of rose in the presence of phosphatase inhibitors (17), immunoprecipitated, and analyzed by NaDodSO4/PAGE as The publication costs of this article were defrayed in part by page charge described below. payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. §1734 solely to indicate this fact. Abbreviation: WGA, wheat germ agglutinin. 8115 Downloaded by guest on October 1, 2021 8116 Medical Sciences: Hayes and Lockwood Proc. Natl. Acad. Sci. USA 84 (1987)

Receptor Autophosphorylation. Insulin receptors, partially B purified by chromotography on WGA-Sepharose (17), were incubated in the absence or presence of insulin (100 nM) or 1 2 3 P-Ser --- * H202 (0.03-3 mM) for 15 min at 240C. The phosphorylation A assay was initiated by addition of [y-32P]ATP (final concentration, 50,uM; 2.5 ACi/nmol), MgCI2 (10 mM), and MnCl2 (2 mM). After 10 min at 40C the reaction was terminated (17) and insulin receptors were immunoprecipitated and analyzed by NaDodSO4/PAGE as described below. Phosphorylation of a P-Thr polymer of glutamic acid and tyrosine (Glu8OTyr2O, Sigma) was performed using partially purified insulin receptor by the procedure of Braun et al. (18) as described (17). In experiments determining activation of tyrosine kinase activity in 95 + _0 intact cells, insulin receptors were isolated in the absence of P-Tyr -_ ATP but in the presence of the phosphatase inhibitors described above for in situ phosphorylation. Phosphorylation in Adipocyte Homogenates and Isolated Plasma Membranes. Isolated plasma membranes were prepared by the method of McKeel and Jarret (19) as modified by Cushman and Wardzala (20) except that the homogeni- zation buffer contained 100 mM NaF, 0.1 mM sodium orthovanadate, and 1 mM phenylmethylsulfonyl fluoride. The final pellet was suspended in 50 mM Tris HCl (pH 7.4) containing 0.1 mM sodium vanadate. To prepare cell homogenates, adipocytes were suspended in 50mM Tris-HCl (pH 7.4) containing 100 mM NaF, 0.1 mM sodium vanadate, H202 1 mM phenylmethylsulfonyl fluoride, and 1 mg of bacitracin Insulin per ml and homogenized in a glass homogenizer using a Teflon pestle (10 strokes, 2000 rpm). The homogenate was FIG. 1. Effects of H202 on in situ receptor phosphorylation. (A) Rat adipocytes were equilibrated with 32P04 for 2 hr at 37°C. Insulin centrifuged at 100,000 x g for 60 min at 4°C. The fat cake was (100 nM) or H202 (3 mM) was then added and incubation was discarded and the pellet was suspended in the infranatant. continued for 5 or 10 min, respectively. The buffer was removed and Plasma membranes or homogenates were incubated for 5 min cells were solubilized in the presence of5 mM ATP and phosphatase at 20°C with 10 mM MgCl2 and 2 mM MnCl2 in the absence inhibitors. Insulin receptors were partially purified on WGA-Seph- or presence of insulin (1 ,uM) or H202 (0.3-3 mM). Phospho- arose, immunoprecipitated, and analyzed by NaDodSO4/PAGE and rylation was initiated by addition of [32P]ATP (100lO M; 2-8 autoradiography. The arrow indicates the receptor f3 subunit (shown ,Ci/nmol). After 10 min at 20°C reactions were terminated as Mr X 1O-3). (B) Phosphoamino acid analysis of H202-stimulated by solubilization in 50 mM Tris HCl (pH 7.4) containing 1% receptor phosphorylation. 32P-labeled 13 subunit from H202-treated X-100 and phosphatase inhibitors and insulin recep- cells was eluted from the gel and hydrolyzed in 6 M HCl for 1 hr at Triton 110°C. Acid was removed by repeated drying under vacuum. tors were partially purified on WGA-Sepharose. Insulin Unlabeled phosphoserine (P-Ser), phosphothreonine (P-Thr), and receptor phosphorylation was analyzed by immunoprecipi- phosphotyrosine (P-Tyr) were added and phosphoamino acids were tation and gel electrophoresis. separated by high-voltage thin-layer electrophoresis at pH 3.5 (6). Immunoprecipitation and Gel Electrophoresis. Phosphoryl- ated insulin receptors were immunoprecipitated by incubation with a 1:25 dilution of antibody to rat liver insulin dues are found when partially purified insulin receptors undergo receptor for 16 hr at 4°C; this was followed by precipitation insulin-stimulated autophosphorylation, insulin-stimulated re- with Pansorbin (Calbiochem). Immunoprecipitates were ceptor phosphorylation in intact cells occurs on tyrosine, solubilized at 100°C in 80 mM Tris HCl (pH 6.8) containing serine, and, to a lesser extent, threonine residues (1). Therefore, 3.8% NaDodSO4, 5% 2-mercaptoethanol, 7.5% glycerol, and phosphoamino acid analysis of 32P-labeled insulin receptors 0.0025% bromophenol blue. NaDodSO4 gel electrophoresis from H202-treated adipocytes was performed (Fig. 1B). As has was carried out using the discontinuous buffer system de- been shown in other cell types (1), insulin receptors from scribed by Laemmli (21) with a 4.5% acrylamide stacking gel nonstimulated adipocytes were phosphorylated only on serine and a 7.5% acrylamide resolving gel. Molecular weight residues (data not shown). Similar to insulin, stimulation of fat standards, obtained from Bio-Rad, were myosin (Mr cells with 3 mM H202 resulted in the phosphorylation of 200,000), f-galactosidase (Mr 116,250), phosphorylase b (Mr tyrosine residues as well as increased serine phosphorylation. 92,500), bovine serum albumin (Mr 66,200), and ovalbumin Since H202 stimulated tyrosine phosphorylation of the (Mr 45,000). Radioactive bands were detected by autoradi- insulin receptor in intact adipocytes, the relationship be- ography of fixed and stained gels using intensifying screens. tween insulin receptor phosphorylation and stimulation of Densitometric scanning ofautoradiograms was performed on glucose transport by this agent was examined further. We an LKB model 2202 Ultrascan laser densitometer (LKB). first examined the time dependence of H202's effects (Fig. 2A). In agreement with previous studies (9), there was a short RESULTS lag (1-3 min) before H202 stimulated uptake of 2- Addition of 3 mM H202, a maximally effective concentration deoxyglucose. Hexose transport then reached maximal lev- that stimulates glucose transport -50% as effectively as els at 30 min oftreatment. In contrast, stimulation ofreceptor insulin (9), to intact adipocytes equilibrated with 32P04 resulted phosphorylation by H202 was more rapid (Fig. 2 A and B). in a 2-fold increase in phosphorylation of the ,B subunit of the Insulin receptor phosphorylation was enhanced within 1 min insulin receptor (Fig. 1A, lane 2). Similar to H202's effect on of exposure to H202 and reached maximal levels within 5 glucose transport, this increase was 509o of the stimulation min. Stimulation of receptor phosphorylation by insulin was seen when cells are treated with insulin at maximally effective even more rapid, being essentially complete within 1 min concentrations (lane 3). Although only phosphotyrosine resi- (Fig. 2B). Downloaded by guest on October 1, 2021 Medical Sciences: Hayes and Lockwood Proc. Natl. Acad. Sci. USA 84 (1987) 8117

A (D 80 J 200- l17- 2.0 (OO 70- c cl;. < Cr 0:= en 60 .;-- C I17 M :3 >-Z .5 o0 50 0 M Oo -C -.h: LL - 40 CL - 1.0 Z >, U) ;. x 0 0 n 100 30 a) a- a- 0 I 20 - 0.5 ZU CM - -J cr cow< UJl n _L

0 I0 20 30 0- NONE PEROXIDE INSULIN Time (min) TREATMENT OF CELLS B FIG. 4. In situ activation of insulin receptor tyrosine kinase by Min 0 1 3 5 30 1 20 H202. Adipocytes were incubated for 30 min at 370C with no additions (none), 3 mM H202 (peroxide), or 7 nM insulin (data for H202 and insulin are expressed as mean ± SEM). Cells were then 95 + .,. Aaffm4 acid washed (pH 6.3, 100C) and solubilized in the presence of phosphatase inhibitors. Insulin receptors were partially purified and kinase activity was determined by phosphorylation of Glu8oTyr20 in

+ + + the absence of insulin. H202 + _-

Insulin ~- kinase activity of the insulin receptor since incubation of FIG. 2. Comparison of the time dependence of H202's effects on partially purified receptors with anti-insulin receptor anti- insulin receptor phosphorylation and hexose transport. (A) bodies reduced substrate phosphorylation to the same level Adipocytes, equilibrated with 32P04, were incubated for the indicat- in preparations from untreated, H202-treated, and insulin- ed times with H202 (3 mM) or insulin (100 nM). Cells were then treated cells (data not shown). immediately solubilized with Triton X-100 in the presence of 5 mM The effect of H202 on exogenous substrate phosphoryl- ATP and phosphatase inhibitors. Insulin receptors were partially ation and autophosphorylation of partially purified insulin purified, immunoprecipitated, and analyzed by NaDodSO4/PAGE. receptor was determined. Although insulin stimulated the Phosphorylation was quantitated by densitometric scanning of receptor kinase activity 5-fold in these preparations, concen- autoradiograms. In parallel incubations, hexose transport activity trations ofH202 (0.03-3 mM) that stimulated 2-deoxyglucose was measured by uptake of 2-deoxy[3H]glucose (0.1 mM) for 90 sec at 37°C after preincubation for the indicated times with 3 mM H202. uptake and receptor phosphorylation in situ (Fig. 3) had no (B) Portion of the autoradiogram showing 3-subunit phosphoryla- effect on tyrosine kinase activity when added directly to tion. partially purified insulin receptors (data not shown). Simi- larly, insulin stimulated autophosphorylation of the ,8 subunit As is shown in Fig. 3, the concentration dependence of of the insulin receptor (Fig. 5, lane 2), whereas 3 mM H202 H202-stimulated receptor phosphorylation and 2-deoxyglu- had no effect (lane 3). Importantly, H202 did not inhibit cose transport were identical. Maximal stimulation of both insulin-stimulated receptor phosphorylation (lane 4), indicat- activities occurred between 0.3 and 3 mM H202. Neither ing that the lack of stimulation with H202 was not due to transport nor receptor phosphorylation was enhanced by oxidative destruction ofATP or required cofactors. Although treatment of adipocytes with 0.03 mM H202. breakdown of H202 occurred in these preparations this could To assess whether H202-stimulated receptor phosphoryl- not account for the lack of an effect since, at 3 mM H202, ation activated tyrosine kinase activity, insulin receptors degradation did not lower the concentrations below that were isolated from adipocytes incubated in the absence or effective in stimulating insulin action and receptor phospho- presence of H202 (3 mM) or insulin (7 nM). Tyrosine kinase rylation in situ. Furthermore, the use of peroxide-free Triton activity was determined by phosphorylation of the synthetic X-100 or the inclusion of dithiothreitol, as suggested by the peptide Glu80Tyr20 in the absence of insulin. As is shown in work of Chan et al. (22), did not alter these results. Fig. 4, tyrosine kinase activity in preparations from H202- To examine the cellular location of the factors involved in treated cells was 50% greater than from untreated adipocytes, H202 stimulation of receptor phosphorylation, our initial whereas insulin treatment ofcells increased this activity 80%6. experiments have employed total cellular homogenates and This enhanced kinase activity could be attributed to the isolated plasma membranes. Similar to the intact cell, insulin

70 1.4 60 r (n 0o a ) 50 .~_ C Lf)- 1.2 c3 3 in >1 40 o CU >n 0 _ I- ~0 -6 80t)X3 0 30 cn %. a, 0 1.0 0 < FIG. 3. Concentration dependence of 20 H202's effect on receptor phosphorylation and 9. hexose transport. Adipocytes, equilibrated with C-,j 10 32P4, were incubated for 30 min at 37°C with 0.8 the indicated concentrations of H202. In parallel, hexose transport activity was determined by 0 0.1 1.0 10 measurement of 2-deoxyglucose uptake. Recep- tor phosphorylation was quantitated by densi- [ H202] mM tometric scanning of autoradiograms. Downloaded by guest on October 1, 2021 8118 Medical Sciences: Hayes and Lockwood Proc. Natl. Acad. Sci. USA 84 (1987) 12 3 4 Several lines of evidence indicate that insulin receptor phosphorylation and activation ofits intrinsic tyrosine kinase activity is an early step linking hormone binding to action- e.g., glucose transport. The time course of receptor phosphorylation and dephosphorylation in intact cells is rapid and precedes insulin action (7). Microinjection of monoclonal FIG. 5. Effect of H202 on in antibodies that block receptor kinase activity inhibit insulin vitro insulin receptor autophos- 95* _ ^ phorylation. Partially purified in- action (23). Replacement of the tyrosines that become phos- 95'W sulin receptors were incubated for phorylated (24) or a lysine residue in the ATP-binding region 20 min at 24°C with no additions, (25, 26) by site-directed mutagenesis impairs insulin action. 100 nM insulin, 3 mM H202, or Also several insulinomimetic agents stimulate receptor phos- insulin and H202. Phosphoryl- phorylation (27-29). ation was initiated by the addition Our data suggest that the insulin-like effects of H202 on of [32P]ATP. After 20 min at 24°C, glucose transport may be mediated by stimulation ofreceptor the reaction was terminated by phosphorylation. As is the case with insulin, stimulation of addition of unlabeled ATP and receptor phosphorylation by H202 occurs more rapidly than phosphatase inhibitors. Insulin receptors were immunoprecipitated activation of the glucose transport system. The parallel and analyzed by NaDodSO4 gel dependence of these two processes on the concentration of electrophoresis and autoradi- H202 is also compatible with this hypothesis. However, since ography. The arrow indicates the neither stimulation of receptor phosphorylation nor activa- H202 - - 3subunit of the insulin receptor tion of glucose transport reaches the levels seen with maxi- Insulin - - (shown as Mr x 10-s). mal insulin concentrations, receptor phosphorylation may not be the rate-limiting step for H202's insulin-like effects. In and H202 stimulate receptor phosphorylation in adipocyte this regard, Lane and co-workers (30) have suggested that, in homogenates (Fig. 6A). In contrast, although insulin in- 3T3-L1 adipocytes, tyrosine phosphorylation of a small (15 creased receptor phosphorylation in purified plasma mem- kDa) protein may be the rate-limiting step in insulin activa- branes, H202 had no effect (Fig. 6B). tion of hexose uptake. To date, neither the presence of this protein nor the effects of insulin or H202 on its phosphorylation have been established in other cell types. DISCUSSION H202 may not require the entire insulin receptor to exert insulin-like effects. Several insulinomimetic agents that act Our data demonstrate that addition of H202 to intact by extracellular generation of H202 retain their action in adipocytes stimulates phosphorylation ofthe insulin receptor adipocytes rendered insulin insensitive by trypsin treatment, on tyrosine and serine residues. Additionally, partially puri- which destroys insulin binding (12, 31). Furthermore, H202 fied insulin receptor preparations from H202- or insulin- has easily demonstrable insulin-like effects in Madin-Darby treated cells exhibit increased tyrosine kinase activity. This canine kidney cells, which have no insulin-binding activity increase can be attributed to activation of the intrinsic kinase (32), and in fibroblasts from an insulin-resistant patient with activity of the insulin receptor since immunoprecipitation leprechaunism, which have markedly reduced insulin-bind- with anti-receptor antibodies returned the kinase activity to ing capacity (33). In all of the above examples, the status of the level seen with preparations from untreated cells. the insulin receptor was established by insulin binding and

95/E 4 1 2 3 4

.; :.i

95 .- r n;

0.3 3 .0 H202 (mM) 0 0 H20 (mM) 0 0.3 3 Insulin (nM) 0 100 0 0 Insulin (rlM) 0 100 0 0

FIG. 6. Insulin receptor phosphorylation in cell homogenates and plasma membranes. Adipocyte homogenates (A) or purified plasma membranes (B) were incubated with no addition (lane 1), 1 AM insulin (lane 2), 0.3 mM H202 (lane 3), or 3 mM H202 (lane 4). Phosphorylation was initiated by addition of [32P]ATP. Insulin receptors, isolated by WGA-Sepharose chromatography and immunoprecipitation, were analyzed by gel electrophoresis. The arrows indicate the 18 subunit of the insulin receptor (shown as M, x 10-3). Downloaded by guest on October 1, 2021 Medical Sciences: Hayes and Lockwood Proc. Natl. Acad. Sci. USA 84 (1987) 8119 thus may only reflect a measurement of the binding portion 1. Kasuga, M., Zick, Y., Blithe, D. L., Karlsson, F. A., Haring, of the a subunits of the receptor. However, they do not H. U. & Kahn, C. R. (1982) J. Biol. Chem. 257, 9891-9894. exclude the possibility that the /3 subunit, or some fragment 2. Kasuga, M., Zick, Y., Blithe, D. L., Crettaz, M. & Kahn, of it, may be present and mediate H202's effects. Since H202 C. R. (1982) Nature (London) 298, 667-669. has no effect on insulin 3. Roth, R. A. & Cassell, D. J. (1983) Science 219, 299-301. binding (12), it is probable that an a 4. Avruch, J., Nemenof, R. A., Blackshear, P. J., Pierce, M. W. subunit capable of binding insulin may not be required. & Osathanondh, R. (1982) J. Biol. Chem. 257, 15162-15166. Although the effects of insulin and H202 were qualitatively 5. Rosen, 0. M., Herrera, R., Olowe, Y., Petruzzeli, L. M. & similar in intact cells, this was not always the case in broken Cobb, M. H. (1983) Proc. Natl. Acad. Sci. USA 80, 3237-3240. cell systems. H202 had no effect on either autophosphoryla- 6. Yu, K. T. & Czech, M. P. (1986) J. Biol. Chem. 261, tion or tyrosine kinase activity when added directly to 4715-4722. partially purified insulin receptors. These findings are likely 7. Kohanski, R. A., Frost, S. C. & Lane, M. D. (1986) J. Biol. supported by the studies of Haring et al. (34), who found that Chem. 261, 12272-12281. an insulinomimetic agent, spermine, also had no effect on 8. Czech, M. P., Lawrence, J. C., Jr., & Lynn, W. S. (1974) insulin The Proc. Natl. Acad. Sci. USA 71, 4173-4177. partially purified receptors. insulin-like proper- 9. Livingston, J. N., Amatruda, J. M. & Lockwood, D. H. (1978) ties of this polyamine are dependent on the generation of Am. J. Physiol. 234, E484-E488. H202 by an amine oxidase present in most preparations of 10. May, J. M. & de Haen, C. (1979) J. Biol. Chem. 254, bovine serum albumin (35). It is unclear whether sufficient 9017-9021. bovine serum albumin was present to generate H202 in the 11. Little, S. A. & de Haen, C. (1980) J. Biol. Chem. 255, experiments of Haring et al. (34). 10888-10895. The observation that H202 has no effect on partially 12. Czech, M. P. & Fain, J. N. (1970) Endocrinology 87, 191-194. purified insulin receptors suggests that nonreceptor factors or 13. Maloff, B. L., Drake, L., Riedy, D. K. & Lockwood, D. H. kinases that do not copurify with the receptor may be (1984) Eur. J. Pharmacol. 104, 319-326. involved in 14. Rodbell, M. J. (1964) J. Biol. Chem. 239, 375-380. H202 stimulation of phosphorylation in intact 15. Livingston, J. N. & Purvis, B. J. (1980) Am. J. Physiol. 238, cells. In this regard, our initial experiments to define these E267-E275. factors showed that, although insulin stimulates receptor 16. Krupp, M. N. & Livingston, J. N. (1978) Proc. Natl. Acad. phosphorylation in purified plasma membranes, H202 had no Sci. USA 75, 2593-2597. effect. However, insulin and H202 stimulate receptor phos- 17. Hayes, G. R. & Lockwood, D. H. (1986) J. Biol. Chem. 261, phorylation in crude adipocyte homogenates. This may 2791-2798. indicate that nonplasma membrane constituents are involved 18. Braun, S., Raymond, W. E. & Racker, E. (1984) J. Biol. in H202's effect. At this time it is not known whether these Chem. 259, 2051-2054. factors are components of the or reside in an 19. McKeel, D. W. & Jarret, L. (1970) J. Cell Biol. 44, 417-432. cytosol 20. Cushman, S. W. & Wardzala, L. J. (1980) J. Biol. Chem. 255, intracellular membrane. 4758-4762. Recently, Chan et al. (22) have reported that several 21. Laemmli, U. K. (1970) Nature (London) 227, 680-685. oxidants stimulate phosphorylation of Glu80Tyr20 and endog- 22. Chan, T. M., Chen, E., Tatoyan, A., Shargill, N. S., Pleta, M. enous proteins in purified rat liver plasma membranes. & Hochstein, P. (1986) Biochem. Biophys. Res. Commun. 139, Additionally, the oxidant vitamin K5 modestly stimulated 439-445. phosphorylation of a protein of Mr 95,000, believed to be the 23. Morgan, D. O., Ho, L., Korn, L. J. & Roth, R. A. (1986) /3 subunit of the insulin receptor, in detergent extracts of Proc. Natl. Acad. Sci. USA 83, 328-332. these membranes purified by chromatography on WGA. In 24. Ellis, L., Clauser, E., Morgan, D. O., Edery, M., Roth, R. A. both cases, oxidant-stimulated phosphorylation required di- & Rutter, W. J. (1986) Cell 45, 721-732. thiothreitol. In our study, using immunoprecipitation in 25. Ebina, Y., Araki, E., Taira, M., Shimada, F., Mori, M., Craik, addition to WGA-Sepharose chromatography to assess re- C. S., Siddle, K., Pierce, S. B., Roth, R. A. & Rutter, W. J. ceptor phosphorylation, addition of dithiothreitol, in the (1987) Proc. Natl. Acad. Sci. USA 84, 704-708. concentration used by Chan et al. (22), to either partially 26. Chou, C. K., Dull, T. J., Russel, D. S., Gherzi, R., Lebwohl, purified receptors or isolated plasma membranes did not D., Ullrich, A. & Rosen, 0. M. (1987) J. Biol. Chem. 262, result in H202 stimulation of receptor phosphorylation. At 1842-1847. this time the reason for this difference is not known. 27. Roth, R. A., Cassell, D. J., Maddux, B. A. & Goldfine, I. D. In (1983) Biochem. Biophys. Res. Commun. 115, 245-252. conclusion, H202 stimulates insulin receptor phospho- 28. Tamura, S., Brown, T. A., Dubler, R. E. & Larner, J. (1983) rylation in intact adipocytes and crude cell homogenates but Biochem. Biophys. Res. Commun. 113, 80-86. not in plasma membranes and partially purified receptors. 29. Tamura, S., Fujita-Yamaguchi, Y. & Larner, J. (1983) J. Biol. Taken together these findings are consistent with the hypoth- Chem. 258, 14749-14752. esis that the insulin receptor could serve as a substrate for 30. Bernier, M., Laird, D. M. & Lane, M. D. (1987) Proc. Natl. other tyrosine kinase(s) stimulated by oxidants. Further Acad. Sci. USA 84, 1844-1848. studies defining the factors involved in H202's effects by 31. Lockwood, D. H. & East, L. E. (1974) J. Biol. Chem. 249, reconstitution of various cytosolic or membrane fractions 7717-7722. from cell homogenates with either purified plasma mem- 32. Hoffman, C., Crettaz, M., Bruns, P., Hessel, P. & Hadawi, G. branes or partially purified receptors should now be possible. (1985) J. Cell Biochem. 27, 401-414. Identification of these factors may provide insight into the 33. Knight, A. B., Rechler, M. M., Romanus, J. A., Van postreceptor events involved in insulin action. Obberghen-Schilling, E. E. & Nissley, S. P. (1981) Proc. Natl. Acad. Sci. USA 78, 2554-2558. We thank Dr. J. N. Livingston for advice throughout the course of 34. Haring, H. U., Kasuga, M. & Kahn, C. R. (1982) Biochem. this work and Susan Wall for excellent technical assistance. This Biophys. Res. Commun. 108, 1538-1545. work was supported by Public Health Service Grant AM 20129 and 35. Livingston, J. N., Gurney, P. A. & Lockwood, D. H. (1977) J. New York State Health Research Council Grant D3-044. Biol. Chem. 252, 560-562. Downloaded by guest on October 1, 2021