Antiinsulin Receptor Autoantibodies Induce Insulin Receptors to ...

0021-972X/99/$03.00/0 The Journal of Clinical Endocrinology & Metabolism Copyright © 1999 by The Endocrine Society

Vol. 84, No. 9 Printed in U.S.A.

Antiinsulin Receptor Autoantibodies Induce Insulin Receptors to Constitutively Associate with Insulin Receptor Substrate-1 and -2 and Cause Severe Cell Resistance to Both Insulin and Insulin-Like Growth Factor I* ` LE DESBOIS-MOUTHON, MARTINE AUCLAIR, CORINNE VIGOUROUX, CHRISTE ¨ ` LE CHERQUI, JOELLE DEIBENER, PIERRE KAMINSKI, OLIVIER LASCOLS, GISE JACQUELINE CAPEAU, AND MARTINE CARON INSERM Unite´ 402, Faculte´ de Me´decine Saint-Antoine (M.A., C.V., C.D.-M., O.L., G.C., J.C., M.C.), 75571 Paris Cedex 12; Service de Biochimie, Hoˆpital Rothschild (C.V., J.C.), 75571 Paris Cedex 12; and De´partement de Me´decine J, Hoˆpital Brabois (J.D., P.K.), 54500, Vandoeuvre les Nancy, France ABSTRACT We report here that antiinsulin receptor (anti-IR) autoantibodies (AIRs) from a newly diagnosed patient with type B syndrome of insulin resistance induced cellular resistance not only to insulin but also to insulin-like growth factor I (IGF-I) for the stimulation of phosphatidylinositol 3-kinase and mitogen-activated protein kinase activities and of glycogen and DNA syntheses. The molecular mechanisms of this dual resistance were investigated. Patient AIRs bound the IR at the insulin-binding site and caused insulin resistance at the IR level by inducing a 50% decrease in cell surface IRs and a severe defect in the tyrosine kinase activity of the residual IRs, manifested by a loss of insulin-stimulated IR autophosphorylation and IR substrate-1 (IRS-1)/IRS-2 phosphorylation. In contrast, cell resistance to IGF-I occurred at a step distal to IGF-I receptors (IGF-IRs), as AIRs

I

NSULIN AND insulin-like growth factor I (IGF-I) exert a variety of biological effects that are mediated by their cognate transmembrane receptors. Although these receptors have a high degree of sequence similarity, the insulin receptor (IR) is predominantly involved in metabolic signaling, whereas the IGF-I receptor (IGF-IR) largely functions as a mitogenic receptor (1, 2). Both IR and IGF-IR belong to the large family of tyrosine kinase receptors and share common intracellular pathways for signal transduction. They autophosphorylate upon ligand binding and phosphorylate on tyrosine residues several endogenous substrates, including the IR substrate (IRS) family and Shc (3). These proteins serve as docking molecules by connecting the activated receptors to the phosphatidylinositol 3-kinase (PI 3-K) and mitogenactivated protein kinase (MAPK) signaling pathways, which, in turn, culminate in the biological effects of insulin and IGF-I. IR and IGF-IR can bind the heterologous ligand, albeit

Received March 11, 1999. Revision received May 20, 1999. Accepted May 25, 1999. Address all correspondence and requests for reprints to: Dr. Martine Caron, INSERM Unit 402, Faculte´ de Me´decine Saint-Antoine, 27 rue Chaligny, 75571 Paris Cedex 12, France. E-mail: caron@st-antoine. inserm.fr. * This work was supported by grants from INSERM.

altered neither IGF-I binding nor IGF-I-induced IGF-IR autophosphorylation, but inhibited the ability of IGF-IRs to mediate tyrosine phosphorylation of IRS-1 and IRS-2 in response to IGF-I. Coimmunoprecipitation assays showed that in AIR-treated cells, IRs, but not IGF-IRs, were constitutively associated with IRS-1 and IRS-2, strongly suggesting that AIR-desensitized IRs impeded IGF-I action by sequestering IRS-1 and IRS-2. Accordingly, AIRs had no effect on the stimulation of mitogen-activated protein kinase activity or DNA synthesis by vanadyl sulfate, FCS, epidermal growth factor, or platelet-derived growth factor, all of which activate signaling pathways independent of IRS-1/IRS-2. Thus, AIRs induced cell resistance to both insulin and IGF-I through a novel mechanism involving a constitutive and stable association of IRS-1 and IRS-2 with the IR. (J Clin Endocrinol Metab 84: 3197–3206, 1999)

with a lower affinity (2), and insulin signaling has been shown to be mediated at least in part by IGF-IRs in cells with defective IRs (4, 5). However, defective IRs may also exert dominant negative effects on IGF-IR signaling, as IGF-I-mediated biological activities are sometimes blunted in fibroblasts from insulin-resistant patients (6) and in transfected cells overexpressing mutant IRs (7–9). The type B syndrome of extreme insulin resistance is a rare clinical disorder characterized by the presence of circulating autoantibodies directed against the IR (AIRs) (10, 11). It is usually associated with hyperinsulinemia, hyperglycemia, and acanthosis nigricans. Most AIRs are polyclonal IgGs that compete with insulin for binding to the a-subunit of the IR. Although in some cases autoantibodies from type B patients were shown to be directed against an antigenic site found in both IR and IGF-IR or to contain subpopulations of antibodies directed against each receptor (12, 13), in most cases AIRs are specific of the IR. In cultured cells, AIRs display both short term insulin-like properties (14) and long term antagonizing effects on insulin action (15–17), the latter being associated in some cases with a progressive loss of cellular IRs (18 –20). To our knowledge, the effect of AIRs on IGF-I signaling has never been studied. In the present study, we report the case of a 68-yr-old

3197

3198

AUCLAIR ET AL.

woman with clinical signs of autoimmune disease and extreme insulin resistance due to AIRs. In short term experiments, patient AIRs blocked insulin binding and mimicked insulin action in cultured CHO cells overexpressing human IRs, but failed to alter [125I]IGF-I binding. Most important, when used in long term experiments, patient AIRs induced severe cell resistance not only to insulin, but also to IGF-I, although they affected neither the IGF-IR expression level nor IGF-IR autophosphorylation. The mechanisms underlying this molecular resistance were investigated. Our results provide the first evidence that AIR-desensitized IRs exerted a dominant negative effect on IGF-I signaling at a step distal to IGF-IRs through a mechanism involving the constitutive association of the IRs with IRS-1 and IRS-2. Subjects and Methods Patient The patient, a 68-yr-old French woman, developed in about 3 months many clinical and biological symptoms indicative of immune dysfunction. Several criteria allowed the diagnosis of systemic lupus erythematosus with renal involvement. She was treated with prednisone (0.5 mg/kgzday) and an iv bolus of cyclophosphamide (500 mg/m2zmonth). This treatment did not improve her clinical state, and 3 months later she displayed massive polyuria and polydipsia due to a severe insulinresistant nonketotic diabetes. Despite iv insulin treatment (6 IU/kgzday), hyperglycemia remained elevated (;25 mmol/L), and in a few weeks the patient developed histologically proven acanthosis nigricans. As described below, AIRs were found in the patient’s serum that defined a type B syndrome of insulin resistance. At this time, angioimmunoblastic lymphoadenopathy with dysproteinemia was diagnosed on the basis of typical histological findings from a cervical node. A high dose of prednisone (2 mg/kgzday) was required to improve the clinical and biological features of autoimmunity. Within 2 weeks, the titer of AIRs decreased by about 6-fold, and 10 days later insulin treatment was no longer required. At that time, circulating AIRs were not detected. Three months later the patient was in remission with normal glucose tolerance.

Chinese hamster ovary (CHO) cells Untransfected CHO cells expressing 2 3 103 endogenous IRs and 4 3 10 endogenous IGF-IRs and CHO cells expressing 5 3 105 human IRs (CHO-IR cells) were cultured as previously described (21). For insulin and IGF-I stimulation, porcine insulin (Novo, Copenhagen, Denmark) or recombinant human IGF-I (PreproTech, London, UK) was added at 50 or 100 nmol/L as indicated. Binding of insulin or IGF-I was performed in the presence of either 19 pmol/L [125I]insulin (3-[125I]iodotyrosylA14 insulin; 2000 Ci/mmol; Amersham Pharmacia Biotech France SA, Les Ulis, France) or 30 pmol/L [125I]IGF-I (3-[125I]iodotyrosyl IGF-I; 2000 Ci/mmol; Amersham Pharmacia Biotech France SA) following standard procedures (22). 4

Characterization of AIRs in the patient’s serum Patient and control serum samples were used either directly or after treatment with 33% ammonium sulfate at 4 C for 4 h to obtain IgGenriched fractions (17). Increasing dilutions of patient serum collected when hyperglycemia was maximal inhibited the binding of [125I]insulin to CHO-IR cells (20.9 6 4.0 fmol/mg protein), with a half-maximal effect at a 1:500 dilution. The inhibitory activities were retained in the IgG fraction of the patient’s serum, with maximal and half-maximal effects at 250 and 15 mg/mL, respectively. Patient serum samples collected after normalization of glucose tolerance as well as the related IgGs did not inhibit insulin binding. Patient IgGs did not modify the ability of [125I]IGF-I to bind the IGF-IR at 22 C for 90 min (4.90 6 1.05 and 4.34 6 9.08 fmol/mg protein in CHO-IR cells preincubated with control and patient IgGs, respectively). These results were consistent with the presence in the patient serum of autoantibodies directed against the IR, but not against the IGF-IR.

JCE & M • 1999 Vol 84 • No 9

Cell desensitization by long term treatment with patient AIRs The ability of long term treatment with AIRs to induce cell resistance to insulin and IGF-I was evaluated by incubating serum-free CHO-IR cells with a maximally effective concentration of patient IgGs (250 mg/ mL) for 8 –16 h at 37 C. After two washes with ice-cold PBS, AIRs were dissociated from the cell surface by a 3-min acid wash at 4 C in a buffer containing 100 mmol/L HEPES, 120 mmol/L NaCl, 1.2 mmol/L MgSO4, 5 mmol/L KCl, and 15 mmol/L sodium acetate, pH 3.5, as described previously (17). Washed cells were then tested for their ability to respond to insulin and IGF-I for various metabolic and mitogenic processes, as described below. The acid wash altered neither cell integrity nor sensitivity to insulin and IGF-I, as evaluated by cell number, protein content, [125I]insulin binding capacity, and insulin and IGF-I stimulation of IRS-1 tyrosine phosphorylation and MAPK activity, which were similar in unwashed and acid-washed cells. The efficiency of the acid wash to dissociate AIRs from the cell surface was assessed by verifying that 1) the lysates from AIR-treated cells that were subjected to the acid wash did not contain immunoreactive IgGs; immunodetection was performed by blotting Protein A/G Plus-agarose (Santa Cruz Biotechnology, Inc., Santa Cruz, CA) immunoprecipitates with protein A/horseradish peroxidase (Transduction Laboratories); and 2) the acid wash reversed the short term inhibition of insulin binding induced by AIRs; this was shown by measuring [125I]insulin binding at 15 C for 5 h in CHO cells treated with patient IgGs for 1 h at 22 C that were subjected, or not, to the acid wash.

Glycogen and DNA syntheses Cell monolayers (5 3 105 cells) were incubated for 2 h with insulin or IGF-I and 2 mCi/mL d-[U-14C]glucose (300 mCi/mmol; Amersham Pharmacia Biotech). Glycogen was then extracted as described previously (6). The results are given as a percentage of the basal value, which is expressed as picomoles of [14C]glucose incorporated into glycogen per mg protein/h. Subconfluent CHO-IR cells or cultured human fibroblasts (105 cells) were maintained for 24 h in serum-free DMEM, treated or not with IgGs for 8 h, and subjected to the acid wash. Cells were then incubated for 16 h with the indicated ligand followed by the addition of 0.5 mCi [methyl3 H]thymidine (5 Ci/mmol; Amersham Pharmacia Biotech) for 4 h. DNA was extracted as described previously (6). The results are given as a percentage of basal value, which is expressed as disintegrations per min/mg protein.

Western blotting studies Cell monolayers (2 3 106 cells) were lysed at 4 C for 30 min in a buffer containing 50 mmol/L HEPES, 50 mmol/L NaF, 100 mmol/L NaCl, 5 mmol/L ethylenediamine tetraacetate (EDTA), 5 mmol/L ethyleneglycol-bis-(b-aminoethyl ether)-N,N,N9,N9-tetraacetic acid, 0.2 mmol/L Na3VO4, 1 mg/mL leupeptin, 0.2 mmol/L PMSF, and 1% Triton X-100, pH 7.4. The lysate was clarified by centrifugation, and equal protein amounts (500 mg) were incubated with the indicated antibody for 2 h at 4 C. Protein A/G Plus-agarose (25 mL) was then added for 16 h. Aliquots of the immunoprecipitates (corresponding to 150 mg cell lysate) were washed three times in lysis buffer containing 0.1% Triton X-100 and processed for SDS-PAGE and Western blotting. Immune complexes were visualized by chemiluminescence (enhanced chemiluminescence detection kit, Amersham Pharmacia Biotech). The antibodies used in these study were obtained from Santa Cruz Biotechnology, Inc. (insulin Rb C-19, IGF-I Ra 2C8, IGF-I Rb C-20, IRS-1 C-20), Upstate Biotechnology, Inc. (Lake Placid, NY; IRS-2 06 –506), and Transduction Laboratories (Lexington, KY; antiphosphotyrosine antibody RC20H). When indicated, cell monolayers (7 3 105 cells) were solubilized in Laemmli sample buffer containing 100 mmol/L dithiothreitol, and aliquots of whole cell lysates (8 mg) were processed for SDS-PAGE and Western blotting.

MAPK assay Cell monolayers (7 3 105 cells) were lysed in a buffer containing 50 mmol/L Tris, 50 mmol/L NaF, 100 mmol/L NaCl, 5 mmol/L EDTA, 40 mmol/L b-glycerophosphate, 0.2 mmol/L Na3VO4, 1 mg/mL leupeptin,

CELL DESENSITIZATION BY INSULIN RECEPTOR ANTIBODIES

3199

0.2 mmol/L PMSF, and 1% Triton X-100, pH 7.5. Aliquots of cell lysates (8 mg) were subjected to SDS-PAGE and Western blotting with an antibody directed against the activated forms of extracellularly regulated kinases (anti-active MAPK polyclonal antibody, Promega Corp., Madison, WI).

PI 3-K assay CHO-IR cells (2 3 106 cells) were lysed for 20 min at 4 C in a buffer containing 20 mmol/L Tris, 0.137 mmol/L NaCl, 1 mmol/L MgCl2, 1 mmol/L CaCl2, 0.1 mmol/L Na3VO4, 1% Nonidet P-40, 10% glycerol, and 1 mmol/L PMSF, pH 8.0. Lysates were clarified by centrifugation, and the supernatants (500 mg) were immunoprecipitated with an anti-PY antibody (PY 99, Santa Cruz Biotechnology, Inc.) in the presence of 25 mL Protein A/G Plus-agarose. After successive washes at 4 C with PBS; 100 mmol/L Tris, 500 mmol/L LiCl (pH 7.4); 10 mmol/L Tris, 100 mmol/L NaCl, and 1 mmol/L EDTA (pH 7.4), pellets were resuspended in the latter buffer (final volume, 50 mL) containing 1 mmol/L MgCl2, 20 mg l-aphosphatidylinositol (sonicated in 5 mmol/L HEPES, pH 7.4), 50 mmol/L ATP, and 12 mCi [g-32P]ATP (2000 Ci/mmol; Amersham Pharmacia Biotech). The reaction was stopped 30 min later by the addition of 100 mL 1 mol/L HCl and 200 mL CHCl3/methanol (1:1). Samples were centrifuged, and the lower organic phase was recovered, washed with 80 mL HCl/ methanol (1:1), and applied to a silica gel TLC plate (Merck KGaA, Darmstadt, Germany) coated with 1% potassium oxalate. TLC plates were developed in CHCl3/CH3OH/H2O/NH4OH (60:47:11.3:2) and dried, and labeled lipids were visualized by autoradiography. Standard lipids were run in parallel and colored with iodine vapor.

Data analysis Chemiluminescent and radioactive signals were quantified by scanning densitometry using NIH Image 1.5 software. Results are expressed as the mean 6 sem for the indicated number of experiments.

Results Short term treatment with patient AIRs mimicked insulin action

As a result of AIR interaction with the IR, we observed short term stimulation of several insulin-related events (Fig. 1). Patient IgGs (lanes 4) enhanced IR b-subunit and IRS-1 tyrosine phosphorylation (Fig. 1A) and increased the level of activated ERK1 and ERK2 (Fig. 1B) to the same extent as did a maximally effective concentration of insulin (lanes 2). In contrast, control IgGs were ineffective (lanes 3). The stimulations by patient IgGs were dose dependent and specific, as they were abolished by 30-min pretreatment with a 1:25 dilution of anti-human IgG serum (data not shown). The maximal effects of insulin and patient IgGs were not additive (lanes 5). Long term treatment with patient AIRs induced cell resistance to insulin and IGF-I metabolic and mitogenic signaling

The next set of experiments was designed 1) to examine the ability of patient AIRs to induce insulin desensitization in vitro in cultured CHO cells and 2) to check whether these AIRs could also induce cell resistance to IGF-I, a hormone that shares with insulin several biological activities (1, 2). As shown in Table 1, long term treatment of CHO-IR cells with patient IgGs (250 mg/mL) severely decreased insulin and IGF-I (50 nmol/L) stimulation of glycogen synthesis compared to that in untreated cells without significantly modifying the basal level. The effect of patient IgGs was specific, as treatment of the cells with control IgGs (300 mg/mL) did not modify the ability of insulin and IGF-I to stimulate gly-

FIG. 1. Short term treatment with patient AIRs increased IR b-subunit and IRS-1 tyrosine phosphorylation (A) and activated MAPK (B) in CHO-IR cells. FCS-free CHO-IR cells were stimulated without or with insulin (100 nmol/L) and control (300 mg/mL) or patient (250 mg/mL) IgGs for 10 (A) or 15 min (B) at 37 C. Cells were lysed in Laemmli sample buffer, 100 mmol/L dithiothreitol (A), or MAPK lysis buffer (B). Aliquots of cell lysates (10 or 8 mg) were immunoblotted with an antiphosphotyrosine (A) or an antiactivated MAPK (B) antibody. Immune complexes were visualized by chemiluminescence. A representative immunoblot from three (A) and four (B) separate experiments is shown.

cogen synthesis. AIR-treated cells also failed to respond to insulin and IGF-I (100 nmol/L) for the stimulation of DNA synthesis. It must be noted that long term treatment with AIRs weakly increased basal DNA synthesis, an effect that was also observed in cells preexposed to control IgGs. These results correlated with the inability of AIR-pretreated cells to respond to both insulin and IGF-I for the activation of PI 3-K and MAPK (Fig. 2), two kinases involved in the stimulation of glycogen and DNA syntheses. Indeed, the stimulation of PI 3-K by insulin or IGF-I (50 nmol/L) was observed in cells treated with control IgGs (Fig. 2A, lanes 5 and 6), but was lost in cells chronically treated with patient IgGs (lanes 8 and 9), which displayed a slight increase in basal PI 3-K activity. Patient AIRs also caused cell resistance to insulin and IGF-I for the activation of MAPK, as evaluated by Western blot analysis with an antibody specific for the phosphorylated forms of ERK1 and ERK2 (Fig. 2B, compare lanes 8 and 9 with lanes 2 and 3 or 5 and 6). Taken together these findings showed that patient AIRs could induce severe insulin resistance in CHO-IR cells, and that this defect was associated with a severe resistance of the cells to IGF-I. Long term treatment with patient AIRs altered IR, but not IGF-IR, level and function

To explore the mechanisms involved in cell resistance to insulin and IGF-I induced by patient AIRs, we searched for

3200

JCE & M • 1999 Vol 84 • No 9

AUCLAIR ET AL.

TABLE 1. Patient AIRs severely inhibited insulin and IGF-I stimulation of glycogen and DNA syntheses in CHO-IR cells Glycogen synthesis (% of basal)

Basal Insulin IGF-I

DNA synthesis (% of basal)

None

Control IgGs

Patient IgGs

None

Control IgGs

Patient IgGs

100 299 6 37 367 6 58

110 6 5 294 6 65 357 6 36

122 6 16 148 6 30 174 6 39

100 403 6 45 398 6 43

131 6 12 353 6 44 367 6 46

128 6 15 120 6 18 128 6 6

Glycogen synthesis: CHO-IR cells were incubated without or with control (300 mg/mL) or patient IgGs (250 mg/mL) for 16 h at 37 C, followed by the acid wash. Cells were then incubated with the hormones for 2 h in the presence of 1 mCi D-[U-14C]glucose. The results are the mean 6 SEM of four separate experiments and are expressed as a percentage of the basal level, which accounted for 349 6 54 pmol/mg protein z h. DNA synthesis: FCS-free CHO-IR cells were incubated for 8 h without or with control (300 mg/mL) or patient (250 mg/mL) IgGs, followed by the acid wash. Insulin or IGF-I was then added for 16 h, followed by addition of 0.5 mCi [methyl-3H]thymidine for 4 h. The results are the mean 6 SEM of four separate experiments and were expressed as a percentage of the basal level, which accounted for 262 6 54 3 103 dpm/mg protein.

showing that long term treatment with AIRs could induce IR down-regulation (18 –20). In contrast, cell treatment with patient AIRs altered neither the level of immunoreactive IGFIRs (Fig. 3A, right panel, lane 3) nor the binding of [125I]IGF-I to CHO-IR cells (6.25 6 1.31 and 5.92 6 1.38 fmol/mg protein in control and patient IgG-treated cells, respectively). Long term treatment of CHO-IR cells with patient IgGs (250 mg/mL) also blocked insulin stimulation of IR b-subunit tyrosine phosphorylation (Fig. 3B). This was observed in anti-IR b-subunit (left panel, lane 6) and anti-PY (right panel, lane 4) immunoprecipitates. However, patient IgGs failed to modify the stimulatory effect of IGF-I on IGF-IR autophosphorylation (Fig. 3C) regardless of whether detection was performed after immunoprecipitation of cell lysates with an anti-IGF-IR a-subunit (left panel, lane 4) or an anti-PY antibody (right panel, lane 4). Taken as a whole, the results showed that patient AIRs caused cell resistance to insulin signaling at the IR level by two mechanisms: down-regulation of IR number and loss of activation of IR autophosphorylation. In contrast, AIRs failed to modify the expression level of IGF-IRs or their ability to autophosphorylate in response to IGF-I, indicating that they induced cell resistance to IGF-I at a step distal to IGF-I-R activation.

FIG. 2. Long term treatment with patient AIRs blocked insulin and IGF-I stimulation of PI 3-K (A) and MAPK (B) in CHO-IR cells. FCS-free CHO-IR cells were incubated without or with control (300 mg/mL) or patient (250 mg/mL) IgGs for 16 h at 37 C, followed by the acid wash. Cells were then stimulated, or not, for 30 min with insulin or IGF-I at 50 nmol/L, and the activation of PI 3-K (A) and MAPK (B) was evaluated as described in Materials and Methods. The autoradiograph (A) is representative of three separate experiments, and the immunoblot (B) is representative of four separate experiments.

possible defects at the IR and IGF-IR levels. Long term treatment of CHO-IR cells with patient IgGs (250 mg/mL) decreased the IR expression level by 50%, as detected by immunoblotting anti-IR immunoprecipitates with an anti-IR b-subunit antibody (Fig. 3A, left panel, lane 3). In accordance, equilibrium binding studies performed at 15 C for 5 h indicated that [125I]insulin binding was 50% lower in AIR-treated cells than in untreated cells (9.75 6 0.86 and 20.3 6 1.6 fmol/mg protein, respectively); this was confirmed by Scatchard analysis of the data, which indicated a decreased IR number without an appreciable change in IR affinity (not shown). These results are consistent with previous studies

Long term treatment with patient AIRs blocked insulin and IGF-I stimulation of IRS-1 and IRS-2 tyrosine phosphorylation and caused IRS-1 and IRS-2 constitutive association with IRs

An early postreceptor step shared by both IRs and IGF-IRs is tyrosine phosphorylation of IRS-1 (2, 3). We thus tested whether AIRs modified IRS-1 tyrosine phosphorylation in response to these hormones (Fig. 4A). We initially observed by immunoblotting anti-IRS-1 immunoprecipitates with an anti-PY antibody that desensitized IRs did not mediate insulin stimulation of IRS-1 tyrosine phosphorylation (lane 8). The AIR-induced defect was specific, as it could not be produced by control IgGs (lane 5), and it could not result from a decreased level of IRS-1 protein (Fig. 4B, lanes 7–9). Alternatively, despite the ability of IGF-IRs to autophosphorylate in response to IGF-I (Fig. 3C), they could not transduce IRS-1 tyrosine phosphorylation in cells treated with patient IgGs (Fig. 4A, lane 9). As overexpression of human IRs in CHO cells can lead to the formation of insulin/IGF-I receptor hybrids (23), it was important to determine whether the defect in IGF-I action induced by AIRs could be due to the presence


3201

FIG. 3. Long term treatment with patient AIRs induced IR, but not IGF-IR, down-regulation and blocked insulininduced IR autophosphorylation, but not IGF-I-induced IGF-IR autophosphorylation. FCS-free CHO-IR cells were incubated for 16 h at 37 C without or with control (300 mg/mL) or patient IgGs (250 mg/mL), followed by the acid wash. Cells were then stimulated, or not, with insulin or IGF-I at 50 nmol/L for 20 min and solubilized in Triton X-100 lysis buffer, and lysates (500 mg proteins) were immunoprecipitated with the indicated antibody. A, Aliquots of anti-IR b immunoprecipitates were blotted with an anti-IR b antibody (left panel), and aliquots of anti-IGF-IR a immunoprecipitates were blotted with an anti-IGF-IR b antibody (right panel). B, Left panel, Anti-IR b immunoprecipitates were probed with an anti-PY antibody; right panel, anti-PY immunoprecipitates were probed with an anti-IR b antibody. C, Same experiment as that in B, except that in the left panel cell lysates were immunoprecipitated with an anti-IGF-IR a antibody, and in the right panel anti-PY immunoprecipitates were blotted with an anti-IGF-IR b antibody. We have checked in these experiments that the anti-IR b antibody did not immunoprecipitate the IGF-IR and that the anti-IGF-IR a antibody did not immunoprecipitate the IR. A representative immunoblot from four (A) and three (B and C) separate experiments is shown.

of such hybrids in CHO-IR cells. To examine this issue, the above experiments were repeated in parental CHO cells (Fig. 4C), and again, we observed resistance to IGF-I (and insulin) in cells treated with patient IgGs (lanes 6 and 5) compared to that in cells treated with control IgGs (lanes 3 and 2). These results strongly suggested that the blockade of IGF-I signaling induced by patient AIRs occurred at an early postreceptor step, distal to IGF-R phosphorylation but proximal to IRS-1 phosphorylation, by a mechanism that presumably involved a defect in IRS-1 assembly with IGF-IRs.

To compare the ability of IRS-1 to associate with IGF-IRs in cells treated with control or patient IgGs (Fig. 5A), we performed coimmunoprecipitation assays with an anti-IRS-1 (left panel) and an anti-IGF-IR a-subunit (right panel) antibody. We observed that in CHO-IR cells treated with control IgGs, IGF-IRs were complexed with IRS-1 20 min after IGF-I addition (lanes 2) under conditions where maximal tyrosine phosphorylation of both IGF-IR (Fig. 3C, lanes 2) and IRS-1 (Fig. 4A, lane 6) occurred (24). In contrast, IGF-IR.IRS-1 complexes were hardly detected in AIR-treated cells stimulated

3202

AUCLAIR ET AL.

FIG. 4. Long term treatment with patient AIRs blocked the stimulation by insulin and IGF-I of IRS-1 tyrosine phosphorylation in CHO cells. FCS-free CHO-IR cells (A and B) or CHO parental cells (C) were incubated for 16 h at 37 C without or with control (300 mg/mL) or patient (250 mg/mL) IgGs, followed by the acid wash. Cells were then stimulated, or not, with insulin or IGF-I at 50 nmol/L for 20 min and solubilized in Triton X-100 lysis buffer, and lysates (500 mg protein) were immunoprecipitated with an anti-IRS-1 antibody. Aliquots of immunoprecipitates were blotted with an antibody directed against phosphotyrosine (A and C) or IRS-1 (B). A representative immunoblot from four (A) and three (B and C) separate experiments is shown.

with IGF-I (Fig. 5A, lanes 4), suggesting that the defect of IGF-I signaling resulted from the inability of activated IGF-IRs to form a complex with IRS-1 in AIR-treated cells. In view of this finding, we searched for a potential role of desensitized IRs in this defect. Coimmunoprecipitation assays (Fig. 5B) with an anti-IRS-1 (left panel) or an anti-IR b-subunit (right panel) antibody indicated that the association

JCE & M • 1999 Vol 84 • No 9

of IR with IRS-1 was strictly different in AIR-treated cells (left panel, lanes 5 and 6, and right panel, lanes 3 and 4) and in cells treated or not with control IgGs (left panel, lanes 1 and 2, and lanes 3 and 4) regardless of whether they were stimulated with insulin for 20 min. In fact, AIR treatment profoundly altered the kinetics of IR.IRS-1 association (Fig. 5C); in control cells stimulated by insulin (left panel), IR and IRS-1 formed a transient complex that was observed at 2 min, persisted up for 10 min (not shown), and disappeared at 20 min, a time at which IR (Fig. 3B, lane 2) and IRS-1 (Fig. 4A, lane 2) tyrosine phosphorylation was maximal. These results were consistent with previous studies that showed that phosphorylated IRS-1 rapidly dissociates from the IR after insulin stimulation (3, 25–27). In contrast, in AIR-treated cells (Fig. 5C, right panel), IR-IRS-1 complexes were formed constitutively in the absence of insulin and were unmodified by the hormone. In these cells, sequestration of IRS-1 by the IR b-subunit correlated with impaired IR (Fig. 3B, lane 6) and IRS-1 (Fig. 4A, lane 8) tyrosine phosphorylation. Similar assays conducted in parental CHO cells (Fig. 5D) showed that AIRs also induced constitutive association of IRs with IRS-1 in untransfected cells (lane 4) regardless of whether insulin (lane 5) and IGF-I (lane 6) were present. This rules out the possibility that the AIR effect seen in CHO-IR cells reflected a forced mechanism due to IR overexpression. To check whether the bulk of IRS-1 was sequestered by IRs in AIR-treated cells (Fig. 5E), supernatants from either the anti-IR b-subunit (right panel) or the anti-IRS-1 (left panel) immunoprecipitate were subjected to a second immunoprecipitation with either an anti-IRS-1 or an anti-IR b-subunit antibody. The results indicated the absence of immunoreactive IRS-1 (right panel, lanes 3 and 4) or IR b-subunit (left panel, lanes 3 and 4) in the supernatants of the anti-IR b-subunit and the anti-IRS-1 immunoprecipitates from cells treated with patient IgGs. As IRS-2, the alternate substrate of IRS-1 (28, 29), could compensate for disrupted insulin and IGF-I signaling through IRS-1 in AIR-treated cells, we repeated the above experiments with an anti-IRS-2 antibody (Fig. 6). We observed that insulin and IGF-I also failed to increase IRS-2 tyrosine phosphorylation in AIR-treated cells (Fig. 6A, left panel, lanes 4 – 6), which could not be explained by a decreased level of IRS-2 protein (right panel, lanes 5 and 6), but correlated with sequestration of the substrate by desensitized IRs (Fig. 6B, left and right panels, lanes 3 and 4). Taken as a whole the results are consistent with the idea that persistent interaction of AIRs with the IR had profoundly modified the IR conformation, so that it constitutively sequestered IRS-1 and IRS-2. The unavailability of these signaling molecules for association with activated IGF-IRs was presumably responsible for the defect in IGF-I signaling observed in AIR-treated cells. Patient AIRs failed to block the effects of agents that activate MAPK and mitogenesis independently of IRS-1

The results described above showed that treatment of CHO-IR cells with patient AIRs inhibited the ability of insulin and IGF-I to increase tyrosine phosphorylation of IRS-1 and IRS-2 and to activate MAPK and DNA synthesis. These inhibitory effects of the AIRs could be linked or not. To


3203

FIG. 5. Long term treatment with patient AIRs prevented IRS-1 association with the IGF-IR, but promoted its constitutive and stable association with the IRs in CHO cells. FCS-free CHO-IR cells (A–C and E) or CHO parental cells (D) were incubated for 16 h at 37 C without or with control (300 mg/mL) or patient (250 mg/mL) IgGs, followed by the acid wash. Cells were then stimulated, or not, with insulin or IGF-I at 50 nmol/L for 20 min (A, B, D, and E) or the indicated time (C). Cell lysates (500 mg protein) were immunoprecipitated, and aliquots were blotted with the indicated antibody. E, The supernatants of antiIRS-1 (left panel) or anti-IR b (right panel) immunoprecipitates were subjected to a second immunoprecipitation with an anti-IR b or an anti-IRS-1 antibody, respectively, followed by immunoblotting with the indicated antibody. In each case, a representative immunoblot from at least four separate experiments is shown.

distinguish between these possibilities, we examined whether the MAPK pathway remained functional in AIRtreated cells by using vanadyl sulfate (VS) and FCS, which activate these kinases through a pathway independent of IRS-1. Indeed, Fig. 7A shows that VS (100 mmol/L) and FCS (10%) markedly enhanced MAPK activation in CHO-IR cells treated with control IgGs (lanes 3 and 4), whereas they did not increase IR b-subunit and IRS-1 tyrosine phosphorylation (Fig. 7B, lanes 3 and 4). Whereas cell treatment with patient IgGs almost blocked insulin activation of MAPK (Fig. 7A, lane 6), it failed to modify the VS- or FCS-mediated activation of MAPK in these cells (lanes 7 and 8). We then compared the mitogenic effects of insulin and

IGF-I to those of FCS, epidermal growth factor (EGF), and platelet-derived growth factor (PDGF), which act as potent activators of DNA synthesis in cultured human skin fibroblasts (30, 31). Table 2 shows that whereas the action of insulin and IGF-I on DNA synthesis was severely impaired in cultured fibroblasts pretreated with patient IgGs, the stimulatory effect of FCS, EGF, and PDGF was unchanged. These results showing that AIRs did not impair the stimulation of mitogenic signaling by agents that acted independently of IRS-1 tyrosine phosphorylation argue for a close relationship between the inhibitory effects exerted by AIRs on tyrosine phosphorylation of IRS-1 and activation of the mitogenic pathway.

3204

AUCLAIR ET AL.

JCE & M • 1999 Vol 84 • No 9

FIG. 6. Long term treatment with patient AIRs blocked insulin or IGF-I stimulation of IRS-2 tyrosine phosphorylation and induced its constitutive and stable association with the IRs in CHO cells. FCS-free CHO-IR cells were incubated for 16 h at 37 C without or with control (300 mg/mL) or patient (250 mg/ mL) IgGs, followed by the acid wash. Cells were then stimulated, or not, with insulin or IGF-I at 50 nmol/L for 20 min. Cell lysates (500 mg protein) were immunoprecipitated with the indicated antibody. Aliquots of immunoprecipitates were blotted with the indicated antibody. We checked that anti-IRS-2 antibody did not immunoprecipitate IRS-1. A representative immunoblot from at least four separate experiments is shown.

Discussion

We report here that AIRs from a newly diagnosed patient with the type B syndrome of insulin resistance induced cell resistance not only to insulin but also to IGF-I for both metabolic and mitogenic signaling despite the fact that AIRs exhibited a restricted specificity for the IRs. Indeed, patient AIRs impaired the binding of insulin, but not that of IGF-I, and immunoprecipitated the IRs, but not the IGF-IRs (not shown). Moreover, patient AIRs induced IR, but not IGF-IR, down-regulation and severely affected ligand-induced autophosphorylation of IRs, but not of IGF-IRs. However, both IRs and IGF-IRs were unable to mediate IRS-1 and IRS-2 tyrosine phosphorylation in AIR-treated cells. These findings indicate that insulin resistance resulted from altered IR functioning and that IGF-I resistance was secondary to a postreceptor defect in IGF-I signaling, presumably due to the presence of unfunctional IRs. Interestingly, IGF-I resistance was previously reported in several situations in which IR activation was impaired (7–9, 32, 33). In these cases, the mechanisms by which defective IRs altered IGF-I signaling were not fully elucidated, although explanations based either on the formation of hybrids between IGF-IRs and overexpressed IRs and/or on the competition for common substrates were suggested. In the present study, the hypothesis that IGF-IRs could not mediate the IGF-I signal due to their engagement in inactive hybrids with the IRs is unlikely, because 1) AIRs impeded neither [125I]IGF-I binding nor IGF-IR autophosphorylation in response to IGF-I, which would be the case for hybrid receptors; and 2) AIRs altered IGF-I postreceptor signaling not only in CHO-IR cells but also in parental CHO cells and cultured human fibroblasts (not shown), in which there is no enforced IR expression. In contrast, the hypothesis invoking the substrate competition to account for the blocking of IGF-I signaling in AIR-

treated cells provides an explanation for our results as developed below. Coimmunoprecipitation assays with antibodies directed against IR, IGF-IR, and IRS-1 showed that in AIR-treated cells, the bulk of cellular IRS-1 was constitutively sequestered in a stable complex by desensitized IRs, which prevented IRS-1 association with ligand-activated IGF-IRs. Sequestration of IRS-1 by kinase-defective IRs was reported in previous studies (25, 34) and was suggested to result from altered conformation of the IR tyrosine kinase domain. From crystallographic data, a model for IRS-1 assembly with the IR was proposed (27, 35) in which insulin binding to control cells alters the spatial conformation of the IR and promotes IR autophosphorylation. This allows unrestricted access of IRS-1 to the IR substrate-binding pocket and IRS-1 binding to Tyr960 in the juxtamembrane of the IR through its phosphotyrosine binding domain (3). Bound IRS-1 is then tyrosine phosphorylated by the activated IR kinase, and once SH2-proteins are engaged with IRS-1, it dissociates from the IR. In light of this model, a possible explanation for the present data is that the persistent anchoring of AIRs at or in a close proximity to the insulinbinding site disrupted the spatial conformation of the IRs so that they were stabilized in a pseudoactivated conformation, with an opened substrate-binding pocket easily accessible for IRS-1. However, due to defective kinase activity of desensitized IRs, bound IRS-1 could not be phosphorylated in response to insulin, remaining stably associated with the IRs and hence unavailable for engagement with activated IGF-IRs. Our investigations also showed that AIRs induced sequestration of IRS-2 by the IRs and defective IRS-2 phosphorylation in response to insulin or IGF-I. This finding may be surprising, because it was shown that, in contrast to IRS-1, which can associate with unphosphorylated IRs (25, 36),


3205

TABLE 2. Patient AIRs had no influence on the mitogenic effect of FCS, EGF, or PDGF in cultured human fibroblasts DNA synthesis (% of basal) Growth factor

None

Control IgGs

Patient IgGs

Basal Insulin (100 nmol/L) IGF-I (100 nmol/L) FCS (10%) EGF (100 nmol/L) PDGF (5 nmol/L)

100 320 6 50 317 6 42 543 6 84 293 6 48 354 6 39

105 6 8 384 6 44 320 6 35 483 6 44 393 6 34 304 6 40

131 6 18 105 6 12 115 6 13 474 6 48 388 6 36 290 6 33

FCS-free human fibroblasts (105 cells) were incubated without or with control (300 mg/mL) or patient (250 mg/mL) IgGs for 8 h at 37 C, followed by the acid wash. Cells were incubated, or not, for 16 h with the indicated concentration of the agent, and DNA synthesis was evaluated as described in Materials and Methods. The results are the mean 6 SEM of four separate experiments and were expressed as a percentage of the basal level, which accounted for 146 6 26 3 103 dpm/mg protein.

FIG. 7. Long term treatment with patient AIRs did not alter the stimulatory effect of VS and FCS on MAPK activity (A), two agents that did not increase IRS-1 and IR b-subunit tyrosine phosphorylation in CHO-IR cells (B). FCS-free CHO-IR cells were incubated without or with control (300 mg/mL) or patient IgGs (250 mg/mL) for 16 h at 37 C, followed by the acid wash. Cells were then incubated, or not, for 30 min with insulin (50 nmol/L), VS (100 mmol/L), or FCS (10%) and lysed in the Triton X-100 lysis buffer. A, MAPK activity was evaluated by blotting cell lysates (8 mg protein) with an antiactivated MAPK antibody. B, IR and IRS-1 tyrosine phosphorylation was evaluated by blotting cell lysates (8 mg protein) with an anti-PY antibody. The immunoblots are representative of at least three separate experiments.

IRS-2 only associates with tyrosine-phosphorylated IRs (37). However it must be underlined that IRs from AIR-treated cells exhibited a low level of tyrosine phosphorylation, which could presumably be sufficient to allow IRS-2 (and presumably IRS-1) engagement by the IRs, but not to promote their phosphorylation, an event that is essential for IRS dissociation from the IRs (3, 37). It is thus possible that IRS-1 and IRS-2 are stably linked to the IRs by a mechanism involving Tyr960 of the IR. Taken as a whole, these data show that IRS-1 and IRS-2 are sequestered by desensitized IRs and that none of them could transduce the insulin signal or serve as a substrate for ligand-activated IGF-IRs in AIR-treated cells. Interestingly, AIRs also blunted insulin and IGF-I activation of MAPK and DNA syntheses, which are classically initiated by the engagement and tyrosine phosphorylation of Shc (24, 38 – 40). Although both IRS-1 and Shc bound Tyr960 of the IR through their phosphotyrosine binding domain, their time course of association and their mode of interaction are different (41). IRS-1 binds Tyr960 regardless of whether it is phosphorylated (36, 42), whereas Shc binding is slower and requires phosphorylation of both Tyr960 in the juxtamembrane domain and Tyr1316 and Tyr1322 in the carboxyl-ter-

minal domain of the IR (36, 41, 43). Thus, Shc signaling of MAPK and DNA synthesis in response to insulin and IGF-I is presumably inhibited in AIR-treated cells due to the misactivation and engagement of the IRs in stable complexes with IRS proteins. The failure of IGF-I to signal MAPK activation and mitogenesis in AIR-treated cells may suggest that kinase-competent IGF-IRs are unable to engage Shc. However, the relative contributions of Shc and IRS-1 in the activation of the MAPK signaling cascade are complex and not clearly defined (3, 44). MAPK activation may also be initiated by the engagement of IRS-1 with PI 3-K (45) or Grb-2-SOS (46, 47) or by the formation of a large signaling complex including IRS-1 and Shc (38, 48, 49). In light of these data, the loss of MAPK activation by IGF-I in AIR-treated cells may be explained by the inability of IRS-1 to associate with Grb-2-SOS, PI 3-K, or Shc. Consistent with this explanation, we provide evidence that AIR treatment affected neither the stimulation of MAPK and/or mitogenesis by vanadate, FCS, and PDGF, all of which act through an IRS-1-independent pathway (30, 50), nor the stimulation of DNA synthesis by EGF, a growth factor that signals mitogenesis through an Shc-dependent pathway (51). These latter findings further argue for IRS-1 dysfunction being responsible for the resistance of AIRtreated cells to insulin and IGF-I mitogenic signaling. When comparing the state of insulin resistance achieved by chronic treatment with either insulin or AIRs, it appears that both insulin-resistant states correlated with IR downregulation and defective tyrosine kinase activation (Refs. 2, 11, and 52 and our results). It must be underlined that a mechanism by which insulin promotes cellular resistance to the hormone consisted of increased serine phosphorylation of IRS-1 (53) and IRs. However, chronic treatment of the cells with AIRs failed to alter IR and IRS-1 serine phosphorylation (data not shown), whereas it generated a misactivated IR that sequestered IRS proteins. In conclusion, this study analyzed the molecular basis of the severe insulin resistance induced by AIRs in a newly diagnosed patient with type B syndrome. This analysis performed at the cellular level provides evidence for the ability of AIRs to induce a dual resistance to insulin and IGF-I

3206

AUCLAIR ET AL.

through a novel mechanism involving the formation of stable and inactive complexes between IRs and IRSs, which act as negative signals for insulin and IGF-I actions.

25.

Acknowledgments

26.

We thank Prof. E. Clauser for kindly providing the human IR complementary DNA. We are grateful to Dr. C. Brahimi-Horn for useful critical review of the manuscript, and to B. Jacquin for expert secretarial assistance.

27.

References

29.

1. Prager D, Melmed S. 1993 Insulin and insulin-like growth factor I receptors: are there functional distinctions? Endocrinology. 132:1419 –1420. 2. Van Obberghen E. 1994 Signalling through the insulin receptor and the insulin-like growth factor-1 receptor. Diabetologia. 37:S125–S134. 3. White MF. 1998 The IRS-signalling system: a network of docking proteins that mediate insulin action. Mol Cell Biochem. 182:3–11. 4. Backeljauw PF, Alves C, Eidson M, et al. 1994 Effect of intravenous insulinlike growth factor 1 in two patients with leprechaunism. Pediatr Res. 36:749 –754. 5. DiCola G, Cool MH, Accili D. 1997 Hypoglycemic effect of insulin-like growth factor-1 in mice lacking insulin receptors. J Clin Invest. 99:2538 –2544. 6. Desbois-Mouthon C, Danan C, Amselem S, et al. 1996 Severe resistance to insulin and insulin-like growth factor-1 in cells from a patient with leprechaunism as a result of two mutations in the tyrosine kinase domain of the insulin receptor. Metabolism. 45:1493–1500. 7. McClain DA, Maegawa H, Thies RS, Olefsky JM. 1990 Dissection of the growth vs. metabolic effects of insulin and insulin-like growth factor-1 in transfected cells expressing kinase-defective human insulin receptors. J Biol Chem. 265:1678 –1682. 8. Jhun BH, Haruta T, Meinkoth JL, et al. 1995 Signal transduction pathways leading to insulin-induced early gene induction. Biochemistry. 34:7996 – 8004. 9. Takata Y, Imamura T, Haruta T, et al. 1996 The dominant negative effect of a kinase-defective insulin receptor on insulin-like growth factor-1-stimulated signaling in rat-1 fibroblasts. Metabolism. 45:1474 –1482. 10. Flier JS, Kahn CR, Roth J, Bar RS. 1975 Antibodies that impair insulin receptor binding in an unusual diabetic syndrome with severe insulin resistance. Science. 190:63– 65. 11. Tritos NA, Mantzoros CS. 1998 Syndromes of severe insulin resistance. J Clin Endocrinol Metab. 83:3025–3030. 12. Kasuga M, Sasaki N, Kahn CR, Nissley SP, Rechler MM. 1983 Antireceptor antibodies as probes of insulinlike growth factor receptor structure. J Clin Invest. 72:1459 –1469. 13. Tappy L, Fujita-Yamagushi Y, LeBon TR, Boden G. 1988 Antibodies to insulin-like growth factor 1 receptors in diabetes and other disorders. Diabetes. 37:1708 –1714. 14. Kahn CR, Baird K, Flier JS, Jarrett DB. 1977 Effects of autoantibodies to the insulin receptor on isolated adipocytes. Studies of insulin binding and insulin action. J Clin Invest. 60:1094 –1106. 15. Karlsson FA, Van Obberghen E, Grunfeld C, Kahn CR. 1979 Desensitization of the insulin receptor at an early post-receptor step by prolonged exposure to antireceptor antibody. Proc Natl Acad Sci USA. 76:809 – 813. 16. Grunfeld C, Van Obberghen E, Karlsson F, Kahn CR. 1980 Antibody-induced desensitization of the insulin receptor: studies of the mechanism of desensitization in 3T3–L1 fatty fibroblasts. J Clin Invest. 66:1124 –1134. 17. Grunfeld C, Jones DS, Shigenaga JK. 1985 Autoantibodies against the insulin receptor. Dissociation of the acute effects of the antibodies from the desensitization seen with prolonged exposure. Diabetes. 34:205–211. 18. Grunfeld C. 1984 Antibody against the insulin receptor causes disappearance of insulin receptors in 3T3–L1 cells: a possible explanation of antibody-induced insulin resistance. Proc Natl Acad Sci USA. 81:2508 –2511. 19. Taylor SI, Marcus-Samuels B. 1984 Anti-receptor antibodies mimic the effect of insulin to down-regulate insulin receptors in cultured human lymphoblastoid (IM-9) cells. J Clin Endocrinol Metab. 58:182–186. 20. Shimoyama R, Shelmet JJ, Savage CR, Boden G. 1986 Effects of anti-insulin receptor antibodies (AIRA) on downregulation and turnover of insulin receptors on cultured hepatocytes. Diabetes. 35:28 –32. 21. Desbois-Mouthon C, Blivet-Van Eggelpoel M-J, Auclair M, et al. 1998 Insulin differentially regulates SAPKs/JNKs and ERKs in CHO cells overexpressing human insulin receptors. Biochem Biophys Res Commun. 243:765–770. 22. Desbois-Mouthon C, Sert-Langeron C, Magre´ J, et al. 1996 Deletion of Asn281 in the a-subunit of the human insulin receptor causes constitutive activation of the receptor and insulin desensitization. J Clin Endocrinol Metab. 81:719 –727. 23. Frattali AL, Treadway JL, Pessin JE. 1992 Insulin/IGF-I hybrid receptors: implications for the dominant-negative phenotype in syndromes of insulin resistance. J Cell Biochem. 48:43–50. 24. Sasaoka T, Ishiki M, Sawa T, et al. 1996 Comparison of the insulin and

28.

30. 31. 32. 33. 34. 35. 36. 37.

38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49.

50. 51. 52. 53.

JCE & M • 1999 Vol 84 • No 9

insulin-like growth factor 1 mitogenic intracellular signaling pathways. Endocrinology. 137:4427– 4434. Wilden PA, Rosiva I, Broadway DE. 1996 Insulin receptor structural requirements for the formation of a ternary complex with IRS-1 and PI 3-kinase. Mol Cell Endocrinol. 122:131–140. Yenush L, White MF. 1997 The IRS-signalling system during insulin and cytokine action. BioEssays. 19:491–500. Hubbard SR. 1997 Crystal structure of the activated insulin receptor tyrosine kinase in complex with peptide substrate and ATP analog. EMBO J. 16:5572–5581. Araki E, Lipes MA, Patti M-E, et al. 1994 Alternative pathway of insulin signalling in mice with targeted disruption of the IRS-1 gene. Nature. 372:186–190. Tobe K, Tamemoto H, Yamauchi T, et al. 1995 Identification of a 190-kDa protein as a novel substrate for the insulin receptor kinase functionally similar to insulin receptor substrate-1. J Biol Chem. 270:5698 –5701. Cleasson-Welsh L. 1994 Platelet-derived growth factor receptor signals. J Biol Chem. 269:32023–32026. Sasaoka T, Langlois WJ, Leitner JW, Draznin B, Olefsky JM. 1994 The signaling pathway coupling epidermal growth factor receptors to activation of p21ras. J Biol Chem. 269:32621–32625. Jacob RJ, Sherwin RS, Greenawalt K, Shulman GI. 1992 Simultaneous insulinlike growth factor I and insulin resistance in obese Zucker rat. Diabetes. 41:691– 697. Conover CA, Clarkson JT, Bale LK. 1994 Physiological concentrations of insulin induce cellular desensitization to the mitogenic effects of insulin-like growth factor 1. Diabetes. 43:1130 –1137. Levy-Toledano R, Caro LHP, Accili D, Taylor SI. 1994 Investigation of the mechanism of the dominant negative effect of mutations in the tyrosine kinase domain of the insulin receptor. EMBO J. 13:835– 842. Hubbard SR, Wei L, Ellis L, Hendrickson WA. 1994 Crystal structure of the tyrosine kinase domain of the human insulin receptor. Nature. 372:746 –753. Paz K, Voliovitch H, Hadari YR, et al. 1996 Interaction between the insulin receptor and its downstream effectors. Use of individually expressed receptor domains for structure/function analysis. J Biol Chem. 271:6998 –7003. Sawka-Verhelle D, Tartare-Deckert S, White MF, Van Obberghen E. 1996 Insulin receptor substrate-2 binds to the insulin receptor through its phosphotyrosine-binding domain and through a newly identified domain comprising amino acids 591–786. J Biol Chem. 271:5980 –5983. Giorgetti S, Pelicci PG, Pelicci G, Van Obberghen E. 1994 Involvement of Src-homology/collagen (SHC) proteins in signaling through the insulin receptor and the insulin-like growth factor receptor. Eur J Biochem. 223:195–202. Jhun BH, Meinkoth JL, Leitner JW, Draznin B, Olefsky JM. 1994 Insulin and insulin-like growth factor-1 signal transduction requires p21ras. J Biol Chem. 269:5699 –5704. Myers MG, White MF. 1995 New frontiers in insulin receptor substrate signaling. Trends Endocrinol Metab. 6:209 –215. He W, O’Neill TJ, Gustafson TA. 1995 Distinct modes of interaction of SHC and insulin receptor substrate-1 with the insulin receptor NPEY region via non-SH2 domains. J Biol Chem. 270:23258 –23262. White MF. 1997 The insulin signalling system and the IRS proteins. Diabetologia. 40:S2–S17. Isakoff SJ, Yu Y-P, Su Y-C, et al. 1996 Interaction between the phosphotyrosine binding domain of Shc and the insulin receptor is required for Shc phosphorylation by insulin in vivo. J Biol Chem. 721:3959 –3962. Myers MG, Sun XJ, Cheatham B, et al. 1993 IRS-1 is a common element in insulin and insulin-like growth factor-1 signaling to phosphatidylinositol 39kinase. Endocrinology. 132:1421–1430. Yamauchi K, Holt K, Pessin JE. 1993 Phosphatidylinositol 3-kinase functions upstream of Ras and Raf in mediating insulin stimulation of c-fos transcription. J Biol Chem. 268:14597–14600. Skolnik EY, Batzer C-H, Vincentini LM, et al. 1993 The SH2/SH3 domain containing protein GRB2 interacts with tyrosine-phosphorylated IRS-1 and SHC: implications for insulin control of ras signalling. EMBO J. 12:1929 –1936. Myers MG, Wang L-M, Sun XJ, et al. 1994 Role of IRS-1-Grb-2 complexes in insulin signaling. Mol Cell Biol. 14:3577–3587. Kasus-Jacobi A, Perdereau D, Tartare-Deckert S, et al. 1997 Evidence for a direct interaction between insulin receptor substrate-1 and Shc. J Biol Chem. 272:17166 –17170. Dupont J, Derouet M, Simon J, Taouis M. 1998 Effect of nutritional state on the formation of a complex involving insulin receptor IRS-1, the 52 kDa Src homology/collagen protein (Shc) isoform and phosphatidylinositol 39-kinase activity. Biochem J. 335:293–300. Hei YJ, Farahbakhshian S, Chen X, Battell ML, McNeill JH. 1998 Stimulation of MAP kinase and S6 kinase by vanadium and selenium in rat adipocytes. Mol Cell Biochem. 178:367–375. Pronk GJ, DeVries-Smits AMM, Buday L, et al. 1994 Involvement of Shc in insulin- and epidermal growth factor-induced activation of p21ras. Mol. Cell. Biol. 14:1575–1581. Kahn CR. 1995 Causes of insulin resistance. Nature. 373:384 –385. Qiao LY, Goldberg JL, Russell JC, Sun XJ. 1999 Identification of enhanced serine kinase activity in insulin resistance. J Biol Chem. 274:10625–10632.