Molecular mechanisms of ATP and insulin synergistic stimulation of coronary artery smooth muscle growth.

Coronary artery disease (CAD) is the major cause of death in diabetics. Abnormal proliferation of coronary artery smooth muscle cells (CASMC) leads to intimal thickening in CAD. We examined signaling mechanisms involved in the mitogenic effect of ATP and insulin on CASMC. ATP and insulin individually stimulated DNA synthesis by 4- and 2-fold, respectively; however, they acted synergistically to stimulate an increase of 17-fold over basal. A similar synergistic stimulation of extracellular signal-regulated kinase (ERK) and mitogen-activated protein or ERK kinase activities was observed (ATP, 7-fold; insulin, 2-fold; and ATP + insulin, 16-fold over basal). However, the combination of ATP and insulin stimulated only an additive activation of Raf (ATP, 5-fold; insulin, <2-fold; and ATP + insulin, 8-fold over basal) and Ras (ATP, 5-fold; insulin, 2-fold; and ATP + insulin, 8-fold over basal). Thus convergence of ATP and insulin signals appears to be at the level of Ras and Raf. In addition, insulin stimulated activation of Akt (also known as protein kinase B) (10-fold over basal), whereas ATP had little effect. However, when ATP and insulin were added in combination, ATP dramatically reduced the insulin-stimulated Akt activation (2-fold above basal). Thus these results are consistent with ATP relieving an insulin-induced Akt-dependent inhibitory effect on the ERK signaling pathway, leading to synergistic stimulation of CASMC proliferation.

P2Y receptor regulation of PAI-1 expression in vascular smooth muscle cells.

P2Y-type purine and pyrimidine nucleotide receptors play important roles in the regulation of vascular hemostasis. In this article, the regulation of plasminogen activator inhibitor-1 (PAI-1) expression in rat aortic smooth muscle cells (RASMCs) by adenine and uridine nucleotides was examined and compared. Northern analysis revealed that RASMCs express multiple P2Y receptor subtypes, including P2Y(1), P2Y(2), and P2Y(6). Treatment of RASMCs with UTP increased PAI-1 mRNA expression and extracellular PAI-1 protein levels by 21-fold (P<0.001) and 7-fold (P<0.001), respectively. The ED(50) for the effect of UTP on PAI-1 expression was approximately 1 micromol/L, and its maximal effect occurred at 3 hours. UDP stimulated a 5-fold increase (P<0.005) in PAI-1 expression. In contrast to these potent stimulatory effects of uridine nucleotides, ATP and 2-methylthioadenosine triphosphate (2-MeSATP) caused a small and transient increase in PAI-1 mRNA at 1 hour, followed by a rapid decrease to baseline levels. ADP produced only an inhibitory effect, reducing PAI-1 mRNA levels by 63% (P<0.05) at 3 hours. The relative nucleotide potency in stimulating PAI-1 expression is UTP>UDP>ATP=2-MeSATP, consistent with a predominant role of the P2Y(6) receptor. Further studies revealed that exposure of RASMCs to either ATP or ADP for 3 hours inhibited both UTP- and angiotensin II-stimulated PAI-1 expression by up to 90% (P<0.001). Thus, ATP induced a small and transient upregulation of PAI-1 that was followed by a strong inhibition of PAI-1 expression. These results show that extracellular adenine and uridine nucleotides exert potent and opposing effects on vascular PAI-1 expression.

APS, an adapter protein with a PH and SH2 domain, is a substrate for the insulin receptor kinase.

APS (adapter protein with a PH and SH2 domain) is the newest member of a family of tyrosine kinase adapter proteins including SH2-B and Lnk. We previously identified SH2-B as an insulin-receptor-binding protein and substrate [Kotani, Wilden and Pillay (1998) Biochem J. 335, 103-109]. Here we show that APS interacts with the insulin receptor kinase activation loop through its SH2 domain and insulin stimulates the tyrosine-phosphorylation of APS. Furthermore, the phosphorylation of activation-loop tyrosine residues 1158 and 1162 are required for this interaction.

ATP-stimulated smooth muscle cell proliferation requires independent ERK and PI3K signaling pathways.

Vascular smooth muscle cells respond to the purinergic agonist ATP by increasing intracellular calcium concentration and increasing the rate of cell proliferation. In many cells the extracellular signal-regulated kinase (ERK) cascade plays an important role in cellular proliferation. We have studied the effect of extracellular ATP on ERK activation and cell proliferation. ATP binding to a UTP-sensitive P2Y nucleotide receptor activates ERK1/ERK2 in a time- and dose-dependent manner in coronary artery smooth muscle cells (CASMC). ATP-induced activation of ERK1/ERK2 is dependent on the dual-specificity kinase mitogen-activated protein kinase/ERK kinase (i.e., MEK) but independent of phosphatidylinositol 3-kinase (PI3K) activity. We provide evidence that both ERK1/ERK2 and PI3K activities are required for CASMC proliferation. Thus ATP-stimulation of CASMC proliferation requires independent activation of both the ERK and PI3K signaling pathways.

SH2-Balpha is an insulin-receptor adapter protein and substrate that interacts with the activation loop of the insulin-receptor kinase.

We identified SH2-Balpha as an insulin-receptor-binding protein based on interaction screening in yeast hybrid systems and co-precipitation in cells. SH2-Balpha contains pleckstrin-homology ('PH') and Src homology 2 (SH2) domains and is closely related to APS (adapter protein with a PH domain and an SH2 domain) and lnk, adapter proteins first identified in lymphocytes. SH2-Balpha is ubiquitously expressed and is present in rat epididymal adipose tissue, liver and skeletal muscle, physiological sites of insulin action. On SDS/PAGE, SH2-Balpha migrates at a molecular mass of 98 kDa, although the predicted size of SH2-Balpha is 79.6 kDa. Insulin causes an electrophoretic mobility shift. SH2-Balpha can be immunoprecipitated using anti-(insulin receptor) antibody from insulin-stimulated cells. Anti-phosphotyrosine antibody or the growth factor receptor-binding protein 2 (Grb2) SH2 domain precipitate SH2-Balpha after insulin stimulation, suggesting that SH2-Balpha is tyrosine-phosphorylated and may be a substrate for the insulin receptor. The SH2-Balpha SH2 domain did not interact with insulin-receptor substrate (IRS) proteins or epidermal-growth-factor receptor. Mutation of the juxtamembrane and C-terminus of the insulin receptor did not abolish the interaction with the SH2 domain. This was further confirmed using a panel of activation-loop single point mutants where mutation of Tyr1158, Tyr1162 and Tyr1163 abolished interaction. Thus SH2-Balpha is a likely component in the insulin-signalling pathway and may function as an alternative signalling protein by interacting with the activation loop of the insulin-receptor cytoplasmic domain.

CSF-1 receptor/insulin receptor chimera permits CSF-1-dependent differentiation of 3T3-L1 preadipocytes.

A chimeric growth factor receptor (CSF1R/IR) was constructed by splicing cDNA sequences encoding the extracellular ligand binding domain of the human colony stimulating factor-1 (CSF-1) receptor to sequences encoding the transmembrane and cytoplasmic domains of the human insulin receptor. The addition of CSF-1 to cells transfected with the CSF1R/IR chimera cDNA stimulated the tyrosine phosphorylation of a protein that was immunoprecipitated by an antibody directed against the carboxyl terminus of the insulin receptor. Phosphopeptide maps of the 32P-labeled CSF1R/IR protein revealed the same pattern of phosphorylation observed in 32P-labeled insulin receptor beta subunits. CSF-1 stimulated the tyrosine phosphorylation of insulin receptor substrate-1 (IRS-1) and Shc in cells expressing the CSF1R/IR chimera. Lipid accumulation and the expression of a differentiation-specific marker demonstrated that 3T3-L1 preadipocytes undergo CSF-1-dependent differentiation when transfected with the CSF1R/IR chimera cDNA but not when transfected with the expression vector alone. A 12-amino acid deletion within the juxtamembrane region of the CSF1R/IR (CSF1R/IRDelta960) blocked CSF-1-stimulated phosphorylation of IRS-1 and Shc but did not inhibit CSF-1-mediated differentiation of 3T3-L1 preadipocytes. These observations indicate that adipocyte differentiation can be initiated by intracellular pathways that do not require tyrosine phosphorylation of IRS-1 or Shc.

Insulin receptor structural requirements for the formation of a ternary complex with IRS-1 and PI 3-kinase.

Insulin receptor substrate-1 (IRS-1) is a major substrate of the insulin receptor tyrosine kinase and is an intermediate in insulin signaling. Phosphotyrosyl-IRS-1 binds to other signaling proteins including phosphatidylinositol 3-kinase (PI 3-kinase). We examined the role of three insulin receptor tyrosine autophosphorylation domains in association of the receptor with IRS-1. Our data support the idea that tyrosine phosphorylation of the insulin receptor juxtamembrane domain is necessary for receptor association with IRS-1. We provide evidence that the kinase regulatory domain, which is part of a loop structure at the mouth of the catalytic cleft, when mutated to replace Tyr1146, Tyr1150, and Tyr1151 with phenylalanine can bind receptor substrates without tyrosine phosphorylation of residues in the receptor juxtamembrane region. In addition, our data show that the amount of PI 3-kinase directly associated with the insulin receptor C-terminus is low when compared to the PI 3-kinase associating with IRS-1. We also demonstrate that a substantial amount (approximately 25%) of the IRS-1 associated PI 3-kinase is associated with the insulin receptor in a ternary complex of insulin receptor/IRS-1/PI 3-kinase.

Combination of insulinomimetic agents H2O2 and vanadate enhances insulin receptor mediated tyrosine phosphorylation of IRS-1 leading to IRS-1 association with the phosphatidylinositol 3-kinase.

To analyze the mechanism of action of the insulinomimetic agents H2O2, vanadate, and pervanadate (H2O2 and vanadate), CHO cells or CHO cells that overexpress wild-type or mutant insulin receptor and/or the insulin receptor substrate (IRS-1) were used. H2O2 or vanadate treatment alone had little or no effect on tyrosine phosphorylation of cellular proteins; however, pervanadate treatment dramatically enhanced tyrosine phosphorylation of a number of proteins including the insulin receptor and IRS-1. However, the insulin receptor and IRS-1 coimmunoprecipitate from insulin-treated but not from pervanadate-treated cells. Pervanadate-induced tyrosine phosphorylation of the insulin receptor led to an increase in insulin receptor tyrosine kinase activity toward IRS-1 in vivo and IRS-1 peptides in vitro equal to that induced by insulin treatment. Pervanadate-enhanced phosphorylation of IRS-1 led to a fifteenfold increase in IRS-1-associated phosphatidylinositol (PtdIns) 3-kinase activity. However, insulin receptor-associated PtdIns 3-kinase activity from pervanadate-treated cells was not detectable, while insulin receptor-associated PtdIns 3-kinase activity from insulin-treated cells was 20% of the IRS-1-associated activity. Thus, pervanadate but not H2O2 or vanadate alone under these conditions mimics many of insulin actions, but pervanadate treatment does not induce insulin receptor/IRS-1 association.

Effect of phosphotyrosyl-IRS-1 level and insulin receptor tyrosine kinase activity on insulin-stimulated phosphatidylinositol 3, MAP, and S6 kinase activities.

The role of tyrosine phosphorylation of the insulin receptor substrate 1 (IRS-1) was studied utilizing parental CHO cells or CHO cells that overexpress IRS-1, the insulin receptor, or both IRS-1 and the insulin receptor. Insulin stimulation of these four cell lines led to progressive levels of IRS-1 tyrosine phosphorylation of one, two, four, and tenfold. Maximal insulin-stimulated IRS-1 associated PtdIns 3'-kinase activit in these cells was 1-, 1.5-, 3-, and 3-fold, while insulin sensitivity, as determined by ED50, was 1-, 2.5-, 10-, and 10-fold. Both sensitivity and maximal response paralleled the increased level of phosphotyrosyl-IRS-1; however, the increased level of phosphotyrosyl-IRS-1 seen in CHO/IR/IRS-1 cells did not further increase these responses. Likewise, maximal insulin-stimulated MAP kinase activity in these cell lines increased in parallel with IRS-1 tyrosine phosphorylation except in the CHO/IR/IRS-1 cell lines with activity levels of one-, five-, nine-, and ninefold. However, insulin sensitivity of the MAP and S6 kinases and maximal insulin-stimulated S6 kinase activity was not changed by a twofold increase in phosphotyrosyl-IRS-1, but an increase was observed with insulin-stimulated receptor autophosphorylation and kinase activity in CHO/IR cells which led to a tenfold increase in insulin receptor autophosphorylation and a fourfold increase in IRS-1 tyrosine phosphorylation. Thus, these three kinase activities may be differentially coupled to the activation of the insulin receptor kinase activity via IRS-1 and other possible cellular substrates.

Characterization of phorbol ester-stimulated serine phosphorylation of the human insulin receptor.

Phorbol 12-myristate 13-acetate (PMA)-stimulated phosphorylation of the human insulin receptor (IR) was characterized and compared in two cell types of different lineage: normal rat kidney epithelial (NRK) cells and Chinese hamster ovary (CHO) fibroblasts. PMA stimulation increased IR beta-subunit phosphorylation to 252 +/- 43 and 25- +/- 47% (+/- S.D.) of the unstimulated control in NRK and CHO cells respectively. Tryptic phosphopeptide analysis by Tricine/SDS/PAGE revealed significant differences in the PMA-stimulated phosphorylation of the IR in these two cell types. This phosphorylation of the IR was predominantly located in two tryptic phosphopeptides, and these phosphopeptides were absent in an IR mutant truncated by 43 C-terminal amino acids. The major PMA-stimulated tryptic phosphopeptide from in vivo-labelled CHO/IR was immunoprecipitated with an antibody against residues Ser1315 to Lys1329, and this precipitation was blocked with excess unlabelled peptide containing this sequence. Radiosequencing by manual Edman degradation revealed that this tryptic phosphopeptide was phosphorylated at Ser1315. This PMA-stimulated phosphorylation did not inhibit autophosphorylation of the IR in vivo. These results demonstrate that PMA-stimulated phosphorylation of the IR can exhibit significant differences when expressed in different cell types, and that Ser1315 is a major PMA-stimulated phosphorylation site on the human IR.

The level of insulin receptor tyrosine kinase activity modulates the activities of phosphatidylinositol 3-kinase, microtubule-associated protein, and S6 kinases.

The role of insulin receptor tyrosine kinase activity in stimulation of intracellular enzymes linked to insulin action [phosphatidylinositol 3-kinase (PtdIns 3-kinase), microtubule-associated protein (MAP) kinase, and S6 kinases] was studied in Chinese hamster ovary cells which overexpress wild type human insulin receptors, receptors with reduced kinase activity due to substitution of Phe for Tyr1146 (single-Phe), Tyr1150,1151 (double-Phe), and Tyr1146,1150,1151 (triple-Phe), or kinase-inactive receptors with a substitution of Ala for Lys1018 in the ATP binding site (A1018). We have previously shown that receptor autophosphorylation and kinase activity of these mutants were reduced by approximately 50, 65, 85, and 100%, respectively. Glycogen and DNA synthesis parallel the level of receptor autophosphorylation and kinase activity; however, receptor serine and threonine phosphorylation was independent of receptor tyrosine kinase activity and receptor internalization was completely dependent on maximal receptor kinase activity. Overexpression of the wild type insulin receptor increased both maximal insulin receptor substrate-1-associated and total insulin-stimulated PtdIns 3-kinase activity, as well as S6 and MAP kinase activities 2.0- to 3.6-fold. In addition there was a leftward shift of the dose-response curves for PtdIns 3-kinase and S6 kinases by approximately 10-fold. Expression of the single- and double-Phe mutant receptors also enhanced maximal PtdIns 3-kinase activity, but had no effect on insulin sensitivity, whereas expression of either the triple-Phe or kinase-inactive receptors did not enhance insulin stimulation or increase insulin sensitivity as compared to the control cells. When comparing the mutant and wild type receptors, differences in insulin sensitivity were least for insulin-stimulated MAP kinase and greatest for S6 kinase; with the latter there was greater than a 1000-fold difference in insulin sensitivity when cells that overexpress wild type vs. kinase-inactive insulin receptors were compared. Thus, the level of insulin receptor tyrosine autophosphorylation and kinase activity regulate both maximal activation and insulin sensitivity of these intracellular kinases in the insulin action pathway which may lead to glycogen and/or DNA synthesis. The differential sensitivity of these enzymes to changes in receptor activation suggests that they may be differently coupled to the receptor kinase.

Insulin stimulates serine and tyrosine phosphorylation in the juxtamembrane region of the insulin receptor.

Insulin-stimulated autophosphorylation of the cytoplasmic juxtamembrane region of the human insulin receptor was examined by Tricine/SDS-PAGE. Various mutant receptor molecules were used to identify two tryptic phosphopeptides associated with the juxtamembrane region which accounts for 15% of the autophosphorylation of partially purified insulin receptor. These phosphopeptides were immunoprecipitated with an antipeptide antibody against the juxtamembrane sequence and were phosphorylated exclusively on tyrosine. Substitution of both Tyr960 and Tyr953 with alanine eliminated insulin-stimulated phosphorylation of the juxtamembrane region without affecting tyrosine autophosphorylation in the C terminus or regulatory regions. Monosubstitution of Tyr960 with phenylalanine or alanine reduced phosphorylation in the juxtamembrane region by more than 50%, and manual Edman degradation indicated that Tyr960 was phosphorylated in wild-type receptor. In vivo, phosphorylation of the juxtamembrane region accounts for one-third of the insulin receptor phosphorylation and contains both phosphoserine and phosphotyrosine. Deletion of Tyr960 and 11 adjacent amino acids eliminated insulin-stimulated phosphorylation of the juxtamembrane region. Substitution of Tyr960 reduced this phosphorylation by more than 50%. The insulin receptor also undergoes serine phosphorylation outside of the juxtamembrane region which depends on the presence of Tyr1151. Together with our previous studies, this report suggests that phosphorylation of Tyr960 may play an important role in signal transduction by the insulin receptor.

Insulin receptor kinase domain autophosphorylation regulates receptor enzymatic function.

We have studied a series of insulin receptor molecules in which the 3 tyrosine residues which undergo autophosphorylation in the kinase domain of the beta-subunit (Tyr1158, Tyr1162, and Tyr1163) were replaced individually, in pairs, or all together with phenylalanine or serine by in vitro mutagenesis. A single-Phe replacement at each of these three positions reduced insulin-stimulated autophosphorylation of solubilized receptor by 45-60% of that observed with wild-type receptor. The double-Phe replacements showed a 60-70% reduction, and substitution of all 3 tyrosine residues with Phe or Ser reduced insulin-stimulated tyrosine autophosphorylation by greater than 80%. Phosphopeptide mapping each mutant revealed that all remaining tyrosine autophosphorylation sites were phosphorylated normally following insulin stimulation, and no new sites appeared. The single-Phe mutants showed insulin-stimulated kinase activity toward a synthetic peptide substrate of 50-75% when compared with wild-type receptor kinase activity. Insulin-stimulated kinase activity was further reduced in the double-Phe mutants and barely detectable in the triple-Phe mutants. In contrast to the wild-type receptor, all of the mutant receptor kinases showed a significant reduction in activation following in vitro insulin-stimulated autophosphorylation. When studied in intact Chinese hamster ovary cells, insulin-stimulated receptor autophosphorylation and tyrosine phosphorylation of the cellular substrate pp185 in the single-Phe and double-Phe mutants was progressively lower with increased tyrosine replacement and did not exceed the basal levels in the triple-Phe mutants. However, all the mutant receptors, including the triple-Phe mutant, retained the ability to undergo insulin-stimulated Ser and Thr phosphorylation. Thus, full activation of the insulin receptor tyrosine kinase is dependent on insulin-stimulated Tris phosphorylation of the kinase domain, and the level of autophosphorylation in the kinase domain provides a mechanism for modulating insulin receptor kinase activity following insulin stimulation. By contrast, insulin stimulation of receptor phosphorylation on Ser and Thr residues by cellular serine/threonine kinases can occur despite markedly reduced tyrosine autophosphorylation.

The role of insulin receptor kinase domain autophosphorylation in receptor-mediated activities. Analysis with insulin and anti-receptor antibodies.

The role of specific tyrosine autophosphorylation sites in the human insulin receptor kinase domain (Tyr1158, Tyr1162, and Tyr1163) was analyzed using in vitro mutagenesis to replace tyrosine residues individually or in combination. Each of the three single-Phe, the three possible double-Phe a triple-Phe and a triple-Ser mutant receptors, stably expressed in Chinese hamster ovary cells, were compared with the wild-type receptor in their ability to mediate stimulation of receptor kinase activity, glycogen synthesis, and DNA synthesis by insulin or the human-specific anti-receptor monoclonal antibody 83-14. At a concentration of 0.1 nM insulin which produced approximately half-maximal responses with wild-type receptor, DNA synthesis and glycogen synthesis mediated by the three single-Phe mutants ranged from 52 to 88% and from 32 to 79% of the wild-type receptor, respectively. The corresponding figures for the double-Phe mutants averaged 15 and 6%, whereas the triple-mutants were unresponsive in both assays. The level of biological function approximately paralleled the insulin-stimulated tyrosine kinase activity in the intact cell as estimated by tyrosine phosphorylation of the insulin receptor and its endogenous substrate pp 185/IRS-1. Interestingly, all mutants showed a marked decrease in insulin-stimulated receptor internalization. Anti-receptor antibody stimulated receptor kinase activity and mimicked insulin action in these cells. In general, the impairment of the metabolic response was greater and impairment of the growth response was less when antibody was the stimulus. These experiments show that the level and specific sites of autophosphorylation are critical determinants of receptor function. The data are consistent with a requirement for the receptor tyrosine kinase either as an obligatory step or a modulator, in both metabolic and growth responses, and demonstrate the important role of the level of insulin receptor kinase domain autophosphorylation in regulating insulin sensitivity.

Insulin stimulation of phosphatidylinositol 3-kinase activity maps to insulin receptor regions required for endogenous substrate phosphorylation.

We have studied the phosphatidylinositol 3-kinase (PtdIns 3-kinase) in insulin-stimulated Chinese hamster ovary (CHO) cells expressing normal (CHO/IR) and mutant human insulin receptors. Insulin stimulation of CHO/IR cells results in an increase in PtdIns 3-kinase activity associated with anti-phosphotyrosine (alpha PY) immunoprecipitates, which has been previously shown to correlate with the in vivo production of PtdIns(3,4)P2, and PtdIns(3,4,5)P3 (Ruderman, N., Kapeller, R., White, M.F., and Cantley, L.C. (1990) Proc. Natl. Acad. Sci. U.S.A. 87, 1411-1415). Stimulation was maximal within 1 min and showed a dose response identical to that of insulin receptor autophosphorylation. The PtdIns 3-kinase also associated with the insulin receptor in an insulin-stimulated manner, as approximately 50% of the total alpha PY-precipitable activity could be specifically immunoprecipitated with anti-insulin receptor antibody. Mutant insulin receptors displayed variable ability to stimulate the PtdIns 3-kinase, but in all cases the presence of PtdIns 3-kinase in alpha PY immunoprecipitates correlated closely with the tyrosyl phosphorylation of the endogenous substrate pp185. In CHO cells expressing a kinase-deficient mutant (IRA1018), there was no observable insulin stimulation of PtdIns 3-kinase activity in alpha PY immunoprecipitates and no tyrosyl phosphorylation of pp185. Substitution of Tyr1146 in the insulin receptor regulatory region with phenylalanine partially impaired receptor autophosphorylation, pp185 phosphorylation, and insulin-stimulated increases in alpha PY-precipitable PtdIns 3-kinase activity. In contrast, a deletion mutant lacking 12 amino acids from the juxtamembrane region (IR delta 960) displayed normal in vivo autophosphorylation but failed to stimulate the PtdIns 3-kinase or phosphorylate pp185. Finally, a mutant receptor from which the C-terminal 43 amino acids had been deleted (IR delta CT) exhibited normal insulin-stimulated autophosphorylation, pp185 phosphorylation, and stimulation of the PtdIns 3-kinase activity in alpha PY immunoprecipitates. These data suggest that the PtdIns 3-kinase is itself a substrate of the insulin receptor kinase or associates preferentially with a substrate. A comparison of the biological activities of the mutant receptors with their activation of the PtdIns 3-kinase furthermore suggests that the PtdIns 3-kinase may be linked to insulin's ability to regulate DNA synthesis and cell growth.

Structure of the insulin receptor substrate IRS-1 defines a unique signal transduction protein.

Since the discovery of insulin nearly 70 years ago, there has been no problem more fundamental to diabetes research than understanding how insulin works at the cellular level. Insulin binds to the alpha subunit of the insulin receptor which activates the tyrosine kinase in the beta subunit, but the molecular events linking the receptor kinase to insulin-sensitive enzymes and transport processes are unknown. Our discovery that insulin stimulates tyrosine phosphorylation of a protein of relative molecular mass between 165,000 and 185,000, collectively called pp185, showed that the insulin receptor kinase has specific cellular substrates. The pp185 is a minor cytoplasmic phosphoprotein found in most cells and tissues; its phosphorylation is decreased in cells expressing mutant receptors defective in signalling. We have now cloned IRS-1, which encodes a component of the pp185 band. IRS-1 contains over ten potential tyrosine phosphorylation sites, six of which are in Tyr-Met-X-Met motifs. During insulin stimulation, the IRS-1 protein undergoes tyrosine phosphorylation and binds phosphatidylinositol 3-kinase, suggesting that IRS-1 acts as a multisite 'docking' protein to bind signal-transducing molecules containing Src-homology 2 and Src-homology-3 domains. Thus IRS-1 may link the insulin receptor kinase and enzymes regulating cellular growth and metabolism.

The insulin receptor with phenylalanine replacing tyrosine-1146 provides evidence for separate signals regulating cellular metabolism and growth.

We have studied the function of a mutant insulin receptor (IR) molecule in which Tyr-1146, one of the first autophosphorylation sites in the beta subunit, was replaced with phenylalanine (IRF1146). Autophosphorylation of the partially purified IRF1146 was reduced 60-70% when compared to the wild-type IR but was still stimulated by insulin. The phosphotransferase activity of the dephospho form of both the IR and IRF1146 toward exogenous substrates was stimulated 3- to 4-fold by insulin. However, the wild-type IR was activated 12-fold by autophosphorylation, whereas the IRF1146 was activated only 2-fold. When the IRF1146 was expressed in Chinese hamster ovary (CHO) cells, insulin binding was normal, whereas autophosphorylation was reduced 80% when compared to cells expressing the wild-type IR. Endogenous substrates of the insulin receptor kinase were not detected during insulin stimulation of CHO cells expressing the IRF1146. Moreover, the IRF1146 did not internalize insulin rapidly or stimulate DNA synthesis in the presence of insulin. In contrast, both the IR and IRF1146 stimulated glycogen synthase equally in CHO cells. These data suggest that activation of the IR tyrosine kinase can be resolved into two components: the first is dependent on insulin binding and the second is dependent on the subsequent insulin-stimulated autophosphorylation cascade. Thus, at least two signal transduction pathways diverging from the IR are implicated in the mechanism of insulin action.

IGF-1-dependent subunit communication of the IGF-1 holoreceptor: interactions between alpha beta heterodimeric receptor halves.

Examination of 125I-IGF-1 affinity cross-linking and beta-subunit autophosphorylation has indicated that IGF-1 induces a covalent association of isolated alpha beta heterodimeric IGF-1 receptors into an alpha 2 beta 2 heterotetrameric state, in a similar manner to that observed for the insulin receptor [Morrison, B.D., Swanson, M.L., Sweet, L.J., & Pessin, J.E. (1988) J. Biol. Chem. 263, 7806-7813]. The formation of the alpha 2 beta 2 heterotetrameric IGF-1 receptor complex from the partially purified alpha beta heterodimers was time dependent with half-maximal formation in approximately 30 min at saturating IGF-1 concentrations. The IGF-1-dependent association of the partially purified alpha beta heterodimers into an alpha 2 beta 2 heterotetrameric state was specific for the IGF-1 receptors since IGF-1 was unable to stimulate the protein kinase activity of the purified alpha beta heterodimeric insulin receptor complex. Incubation of the alpha 2 beta 2 heterotetrameric IGF-1 holoreceptor with the specific sulfhydryl agent iodoacetamide (IAN) did not alter 125I-IGF-1 binding of IGF-1 stimulation of protein kinase activity. In addition, IAN did not affect the Mn/MgATP-dependent noncovalent association of IGF-1 receptor alpha beta heterodimers into an alpha 2 beta 2 heterotetrameric state. However, IAN treatment of the alpha beta heterodimeric IGF-1 receptors inhibited the IGF-1-dependent covalent formation of the disulfide-linked alpha 2 beta 2 heterotetrameric complex. These data indicate that IGF-1 induces the covalent association of isolated alpha beta heterodimeric IGF-1 receptor complexes into a disulfide-linked alpha 2 beta 2 heterotetrameric state whereas Mn/MgATP induces a noncovalent association.(ABSTRACT TRUNCATED AT 250 WORDS)

Wheat germ agglutinin stimulation of alpha beta heterodimeric insulin receptor beta-subunit autophosphorylation by noncovalent association into an alpha 2 beta 2 heterotetrameric state.

The purified human placental alpha 2 beta 2 heterotetrameric insulin-receptor complex was reduced and dissociated into functional alpha beta heterodimers by a combination of alkaline pH and dithiothreitol treatment. Wheat germ agglutinin (WGA) was observed to stimulate the beta-subunit autophosphorylation of both the alpha 2 beta 2 heterotetrameric and alpha beta heterodimeric complexes in the absence of insulin. However, WGA was without effect on the insulin stimulation of beta-subunit autophosphorylation in these insulin-receptor complexes. In contrast, monomeric WGA was unable to stimulate the basal exogenous substrate protein kinase activity in either the alpha 2 beta 2 heterotetrameric or alpha beta heterodimeric insulin receptor complexes. The stimulatory effect of WGA was biphasic, increasing the basal insulin receptor beta-subunit autophosphorylation at low concentrations (250 ng/ml to 2.5 micrograms/ml) and inhibiting at high concentrations (greater than 25 micrograms/ml). Similarly, equilibrium tracer insulin binding was not significantly altered by low concentrations of WGA (less than 1 microgram/ml) but was dramatically reduced at high WGA concentrations (greater than 2.5 micrograms/ml). In contrast to the insulin-induced covalent association of the alpha beta heterodimeric insulin receptors to form a disulfide-linked alpha 2 beta 2 heterotetrameric complex, nonreducing sodium dodecyl sulfate-polyacrylamide gel electrophoresis demonstrated that the WGA stimulation of beta-subunit autophosphorylation in the alpha beta heterodimer preparation occurred in the absence of covalent association. Nondenaturing Bio-Gel A-1.5m gel filtration chromatography and [125I]insulin affinity cross-linking demonstrated that WGA treatment of the alpha beta heterodimeric insulin receptors induced a noncovalent association into an alpha 2 beta 2 heterotetrameric state. These data support the hypothesis that the insulin receptor protein kinase activity, although dependent upon alpha beta heterodimeric subunit interactions, does not necessarily require covalent disulfide bond formation between the individual alpha beta heterodimeric species.

Relationship between insulin receptor subunit association and protein kinase activation: insulin-dependent covalent and Mn/MgATP-dependent noncovalent association of alpha beta heterodimeric insulin receptors into an alpha 2 beta 2 heterotetrameric state.

The purified human placenta alpha 2 beta 2 heterotetrameric insulin receptor was reduced and dissociated into a functional alpha beta heterodimeric complex by a combination of alkaline pH and dithiothreitol treatment. In the presence of Mn/MgATP, insulin binding to the isolated alpha beta heterodimeric insulin receptor was found to induce the formation of a covalent disulfide-linked alpha 2 beta 2 heterotetrameric complex. In the absence of insulin, a noncovalent association of the alpha beta heterodimeric insulin receptor complex into an alpha 2 beta 2 heterotetrameric state required the continuous presence of both a divalent metal ion (Mn or Mg) and an adenine nucleotide (ATP, ADP, or AMPPCP). Thus, Mn/MgATP binding and not insulin receptor autophosphorylation was responsible for the noncovalent association into the alpha 2 beta 2 heterotetrameric state. However, the divalent metal ions or NaATP separately was ineffective in inducing the noncovalent association between the alpha beta heterodimers. The specific sulfhydryl agent iodoacetamide (IAN) was observed to inhibit the insulin-dependent covalent association of the alpha beta heterodimers without affecting the Mn/MgATP-induced noncovalent association into the alpha 2 beta 2 heterotetrameric state. Insulin treatment of the isolated alpha beta heterodimeric complex in the presence of IAN demonstrated that the Mn/MgATP-induce noncovalent association into the alpha 2 beta 2 heterotetrameric state was sufficient for insulin stimulation of beta-subunit autophosphorylation and exogenous substrate protein kinase activity. These data indicate that although interaction between the individual insulin receptor alpha beta heterodimers is necessary for insulin stimulation of protein kinase activity it does not require covalent disulfide bond formation.