Protein Kinase C θ Inhibits Insulin Signaling by Phosphorylating IRS1 at Ser1101
2004; Elsevier BV; Volume: 279; Issue: 44 Linguagem: Inglês
10.1074/jbc.c400186200
ISSN1083-351X
AutoresYu Li, Timothy Soos, Xinghai Li, Jiong Wu, Matthew DeGennaro, Xiao‐Jian Sun, Dan R. Littman, Morris J. Birnbaum, Roberto D. Polakiewicz,
Tópico(s)Plant Gene Expression Analysis
ResumoObesity and stress inhibit insulin action by activating protein kinases that enhance serine phosphorylation of IRS1 and have been thus associated to insulin resistance and the development of type II diabetes. The protein kinase C θ (PKCθ) is activated by free-fatty acids, and its activity is higher in muscle from obese diabetic patients. However, a molecular link between PKCθ and insulin resistance has not been defined yet. Here we show that PKCθ phosphorylates IRS1 at serine 1101 blocking IRS1 tyrosine phosphorylation and downstream activation of the Akt pathway. Mutation of Ser1101 to alanine makes IRS1 insensitive to the effect of PKCθ and restores insulin signaling in culture cells. These results provide a novel mechanism linking the activation of PKCθ to the inhibition of insulin signaling. Obesity and stress inhibit insulin action by activating protein kinases that enhance serine phosphorylation of IRS1 and have been thus associated to insulin resistance and the development of type II diabetes. The protein kinase C θ (PKCθ) is activated by free-fatty acids, and its activity is higher in muscle from obese diabetic patients. However, a molecular link between PKCθ and insulin resistance has not been defined yet. Here we show that PKCθ phosphorylates IRS1 at serine 1101 blocking IRS1 tyrosine phosphorylation and downstream activation of the Akt pathway. Mutation of Ser1101 to alanine makes IRS1 insensitive to the effect of PKCθ and restores insulin signaling in culture cells. These results provide a novel mechanism linking the activation of PKCθ to the inhibition of insulin signaling. Insulin regulates blood glucose uptake and metabolism by binding to the insulin receptor (IR) 1The abbreviations used are: IR, insulin receptor; IRS, insulin receptor substrate; PI3K, phosphatidylinositol 3-kinase; PKC, protein kinase C; IKK, IκB kinase; ERK, extracellular signal-regulated kinase; JNK, c-Jun NH2-terminal kinase; TPA, 12-O-tetradecanoylphorbol-13-acetate; TNF, tumor necrosis factor; CHO, Chinese hamster ovary; MARCKS, myristoylated alanine-rich C kinase substrate; YFP, yellow fluorescent protein; GFP, green fluorescent protein; HA, hemagglutinin; GST, glutathione S-transferase. and activating its intrinsic tyrosine kinase activity. The IR phosphorylates downstream targets such as the insulin receptor substrate (IRS) proteins, which mediate most of the biological actions of insulin (1Zick Y. Trends Cell Biol. 2001; 11: 437-441Abstract Full Text PDF PubMed Scopus (186) Google Scholar). IRS1 is the best-studied IR kinase substrate. Upon tyrosine phosphorylation, IRS1 interacts with Src homology domain-2-containing proteins such as the p85 subunit of phosphatidylinositol 3-kinase (PI3K), Grb2, SHP2, Nck, and others (1Zick Y. Trends Cell Biol. 2001; 11: 437-441Abstract Full Text PDF PubMed Scopus (186) Google Scholar). Activation of PI3K leads to stimulation of Akt, which contributes to increased glucose uptake, glycogen synthesis, and protein synthesis (2Saltiel A.R. Pessin J.E. Trends Cell Biol. 2002; 12: 65-71Abstract Full Text Full Text PDF PubMed Scopus (503) Google Scholar, 3White M.F. Science. 2003; 302: 1710-1711Crossref PubMed Scopus (558) Google Scholar). Prolonged activation of the insulin receptor, free fatty acids, and cellular stress can result in serine phosphorylation and reduced tyrosine phosphorylation of IRS1, thereby attenuating insulin signaling (1Zick Y. Trends Cell Biol. 2001; 11: 437-441Abstract Full Text PDF PubMed Scopus (186) Google Scholar, 3White M.F. Science. 2003; 302: 1710-1711Crossref PubMed Scopus (558) Google Scholar). This mechanism is thought to contribute to insulin resistance and the development of type II diabetes (2Saltiel A.R. Pessin J.E. Trends Cell Biol. 2002; 12: 65-71Abstract Full Text Full Text PDF PubMed Scopus (503) Google Scholar, 3White M.F. Science. 2003; 302: 1710-1711Crossref PubMed Scopus (558) Google Scholar). IRS1 contains many potential serine phosphorylation sites and several protein kinases, such as JNK (4Aguirre V. Werner E.D. Giraud J. Lee Y.H. Shoelson S.E. White M.F. J. Biol. Chem. 2002; 277: 1531-1537Abstract Full Text Full Text PDF PubMed Scopus (769) Google Scholar, 5Lee Y.H. Giraud J. Davis R.J. White M.F. J. Biol. Chem. 2003; 278: 2896-2902Abstract Full Text Full Text PDF PubMed Scopus (340) Google Scholar), IKKβ (6Gao Z. Hwang D. Bataille F. Lefevre M. York D. Quon M.J. Ye J. J. Biol. Chem. 2002; 277: 48115-48121Abstract Full Text Full Text PDF PubMed Scopus (599) Google Scholar), ERK (7De Fea K. Roth R.A. J. Biol. Chem. 1997; 272: 31400-31406Abstract Full Text Full Text PDF PubMed Scopus (267) Google Scholar), Akt (8Paz K. Liu Y.F. Shorer H. Hemi R. LeRoith D. Quan M. Kanety H. Seger R. Zick Y. J. Biol. Chem. 1999; 274: 28816-28822Abstract Full Text Full Text PDF PubMed Scopus (168) Google Scholar), mTOR/FRAP (9Gual P. Gonzalez T. Gremeaux T. Barres R. March Le -Brustel Y. Tanti J.F. J. Biol. Chem. 2003; 278: 26550-26557Abstract Full Text Full Text PDF PubMed Scopus (66) Google Scholar), PKCζ (10Liu Y.F. Paz K. Herschkovitz A. Alt A. Tennenbaum T. Sampson S.R. Ohba M. Kuroki T. LeRoith D. Zick Y. J. Biol. Chem. 2001; 276: 14459-14465Abstract Full Text Full Text PDF PubMed Scopus (164) Google Scholar), and PKCδ (11Greene M.W. Morrice N. Garofalo R.S. Roth R.A. Biochem. J. 2004; 378: 105-116Crossref PubMed Scopus (75) Google Scholar), have been shown to phosphorylate IRS1. However, it has been difficult to determine which serine residues are the most relevant for different physiological or pathological stimuli. Extensive analysis suggests that phosphorylation of mouse IRS1 at Ser307 (Ser312 in human IRS1) by JNK is important for inflammatory stimuli-mediated inhibition of IRS1 function and in mouse models of obesity-driven type II diabetes (4Aguirre V. Werner E.D. Giraud J. Lee Y.H. Shoelson S.E. White M.F. J. Biol. Chem. 2002; 277: 1531-1537Abstract Full Text Full Text PDF PubMed Scopus (769) Google Scholar, 5Lee Y.H. Giraud J. Davis R.J. White M.F. J. Biol. Chem. 2003; 278: 2896-2902Abstract Full Text Full Text PDF PubMed Scopus (340) Google Scholar, 12Hirosumi J. Tuncman G. Chang L. Gorgun C.Z. Uysal K.T. Maeda K. Karin M. Hotamisligil G.S. Nature. 2002; 420: 333-336Crossref PubMed Scopus (2636) Google Scholar, 13Aguirre V. Uchida T. Yenush L. Davis R. White M.F. J. Biol. Chem. 2000; 275: 9047-9054Abstract Full Text Full Text PDF PubMed Scopus (1172) Google Scholar). Chronic elevation of plasma free fatty acid levels has been linked to insulin resistance in skeletal muscle and correlates with obesity and type II diabetes (14Dresner A. Laurent D. Marcucci M. Griffin M.E. Dufour S. Cline G.W. Slezak L.A. Andersen D.K. Hundal R.S. Rothman D.L. Petersen K.F. Shulman G.I. J. Clin. Invest. 1999; 103: 253-259Crossref PubMed Scopus (1023) Google Scholar, 15Roden M. Price T.B. Perseghin G. Petersen K.F. Rothman D.L. Cline G.W. Shulman G.I. J. Clin. Invest. 1996; 97: 2859-2865Crossref PubMed Scopus (1241) Google Scholar). High plasma free fatty acid concentrations have also been associated with increased expression and activation of PKCθ, the most abundant PKC isoform in skeletal muscle (16Qu X. Seale J.P. Donnelly R. J. Endocrinol. 1999; 162: 207-214Crossref PubMed Scopus (94) Google Scholar, 17Itani S.I. Pories W.J. Macdonald K.G. Dohm G.L. Metabolism. 2001; 50: 553-557Abstract Full Text PDF PubMed Scopus (78) Google Scholar, 18Itani S.I. Zhou Q. Pories W.J. Macdonald K.G. Dohm G.L. Diabetes. 2000; 49: 1353-1358Crossref PubMed Scopus (184) Google Scholar, 19Griffin M.E. Marcucci M.J. Cline G.W. Bell K. Barucci N. Lee D. Goodyear L.J. Kraegen E.W. White M.F. Shulman G.I. Diabetes. 1999; 48: 1270-1274Crossref PubMed Scopus (979) Google Scholar). This novel PKC isoform can be activated by different intracellular fatty acid metabolites such as fatty acyl-CoA and diacylglycerol, as well as by phorbol esters (TPA) (20Baier G. Baier-Bitterlich G. Meller N. Coggeshall K.M. Giampa L. Telford D. Isakov N. Altman A. Eur. J. Biochem. 1994; 225: 195-203Crossref PubMed Scopus (78) Google Scholar, 21Yu C. Chen Y. Cline G.W. Zhang D. Zong H. Wang Y. Bergeron R. Kim J.K. Cushman S.W. Cooney G.J. Atcheson B. White M.F. Kraegen E.W. Shulman G.I. J. Biol. Chem. 2002; 277: 50230-50236Abstract Full Text Full Text PDF PubMed Scopus (1196) Google Scholar). Since IRS1 is the major mediator of the insulin response in the muscle (22Kido Y. Burks D.J. Withers D. Bruning J.C. Kahn C.R. White M.F. Accili D. J. Clin. Invest. 2000; 105: 199-205Crossref PubMed Scopus (423) Google Scholar), it has been hypothesized that that PKCθ activation may down-regulate insulin signaling by affecting serine phosphorylation of IRS1 (11Greene M.W. Morrice N. Garofalo R.S. Roth R.A. Biochem. J. 2004; 378: 105-116Crossref PubMed Scopus (75) Google Scholar, 21Yu C. Chen Y. Cline G.W. Zhang D. Zong H. Wang Y. Bergeron R. Kim J.K. Cushman S.W. Cooney G.J. Atcheson B. White M.F. Kraegen E.W. Shulman G.I. J. Biol. Chem. 2002; 277: 50230-50236Abstract Full Text Full Text PDF PubMed Scopus (1196) Google Scholar). Fatty acids can induce IRS1 phosphorylation at Ser307 in muscle, with a concomitant reduction in its tyrosine phosphorylation and IRS1-associated PI3K (21Yu C. Chen Y. Cline G.W. Zhang D. Zong H. Wang Y. Bergeron R. Kim J.K. Cushman S.W. Cooney G.J. Atcheson B. White M.F. Kraegen E.W. Shulman G.I. J. Biol. Chem. 2002; 277: 50230-50236Abstract Full Text Full Text PDF PubMed Scopus (1196) Google Scholar). However, PKCθ is not likely to directly phosphorylate IRS1 at Ser307 because it is not a proline-directed kinase. One possibility is that, as described in T cells (23Altman A. Isakov N. Baier G. Immunol. Today. 2000; 21: 567-573Abstract Full Text Full Text PDF PubMed Scopus (135) Google Scholar), PKCθ activation in the muscle stimulates JNK and IKKβ, two reported Ser307 kinases. This hypothesis would be consistent with studies showing that fat-induced insulin resistance is blocked in JNK1 and IKKβ knock-out mice (12Hirosumi J. Tuncman G. Chang L. Gorgun C.Z. Uysal K.T. Maeda K. Karin M. Hotamisligil G.S. Nature. 2002; 420: 333-336Crossref PubMed Scopus (2636) Google Scholar, 24Yuan M. Konstantopoulos N. Lee J. Hansen L. Li Z.W. Karin M. Shoelson S.E. Science. 2001; 293: 1673-1677Crossref PubMed Scopus (1630) Google Scholar). Another possibility, explored in this report, is that PKCθ may directly phosphorylate and inhibit IRS1 function. Here we demonstrate that PKCθ inhibits insulin signaling by phosphorylating IRS1 at a serine 1101. Ser1101 phosphorylation in muscle cells is induced by insulin, phorbol esters, free fatty acids, and TNFα and contributes to reduced tyrosine phosphorylation of IRS1 and signaling to the PI3K/Akt cascade. Furthermore, we show that mutation of Ser1101 to alanine makes IRS1 refractory to the inhibitory effect of PKCθ on insulin signaling. Reagents—Insulin and arachidonic and oleic acids were obtained from Sigma. Wortmannin and bisindolylmaleimide I were purchased from Calbiochem. All antibodies used in this study, except for the PKCθ antibody (BD Biosciences-Transduction Laboratories), were from Cell Signaling Technology. The phosphospecific IRS1(S1101) antibody was raised by immunizing rabbits with a synthetic peptide containing phosphorylated Ser1101 (see supplementary Table I) and following methods described before (25Polakiewicz R.D. Schieferl S.M. Gingras A.C. Sonenberg N. Comb M.J. J. Biol. Chem. 1998; 273: 23534-23541Abstract Full Text Full Text PDF PubMed Scopus (165) Google Scholar). PKCθ kinase was purchased from Upstate Biotechnology, and other PKC enzymes were obtained from PanVera (now Invitrogen). Cell Culture and Differentiation—Murine C2C12 myocytes and 3T3 L1 fibroblasts were maintained in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum. Differentiation was induced according to published protocols (26Moyers J.S. Bilan P.J. Reynet C. Kahn C.R. J. Biol. Chem. 1996; 271: 23111-23116Abstract Full Text Full Text PDF PubMed Scopus (112) Google Scholar). After differentiation, cells were incubated in serum-free medium overnight prior ligand stimulation and preincubated with inhibitors for 30 min as indicated in the figure legend to Fig. 1. Cells were then lysed, and extracts were prepared for immunoprecipitation and/or immunoblotting as described below. Chinese hamster ovary (CHO)-IR/IRS1 and wild-type CHO cells were cultured in Dulbecco's modified Eagle's medium containing 10% fetal bovine serum. Transfections were performed using Polyfect (Qiagen). Animals and Treatment—C57BL/6J mice were purchased from the Jackson Laboratory. Mice including Akt2 (27Cho H. Mu J. Kim J.K. Thorvaldsen J.L. Chu Q. Crenshaw III, E.B. Kaestner K.H. Bartolomei M.S. Shulman G.I. Birnbaum M.J. Science. 2001; 292: 1728-1731Crossref PubMed Scopus (1547) Google Scholar) and PKCθ (28Sun Z. Arendt C.W. Ellmeier W. Schaeffer E.M. Sunshine M.J. Gandhi L. Annes J. Petrzilka D. Kupfer A. Schwartzberg P.L. Littman D.R. Nature. 2000; 404: 402-407Crossref PubMed Scopus (790) Google Scholar) knock-out were maintained on Teklad Mouse Diet number 7002 and housed in colony cages in a 12-h light/12-h dark cycle. Male mice of age 6–8 weeks were fasted for 12 h and injected intraperitoneally with insulin at 1 unit/kg body weight in phosphate-buffered saline or with phosphate-buffered saline alone (total volume of 100–200 μl) for 20 min before sacrifice. The muscle gastrocnemii were then isolated and snap-frozen in liquid nitrogen and stored in –80 °C before use. Immunoprecipitation and Immunoblotting—Muscle tissue was cut into small pieces on ice, resuspended in cold lysis buffer (50 mm Hepes, pH 7.5, 150 mm NaCl, 10% glycerol, 1% Triton X-100, 5 mm EDTA, 1× complete protease inhibitor mixture (Roche Applied Science), 5 mm Na3VO4, 20mm NaF, 20 mm sodium pyrophosphate, and 200 nm okadaic acid) and Polytron-homogenized. The samples were spun at 14,000 rpm for 15 min at 4 °C. 2 mg of protein from the supernatant was diluted in lysis buffer to 500 μl and immunoprecipitated with 5 μg of rabbit anti-IRS-1 antibody for 2 h at 4 °C. Immunoprecipitated IRS1 complexes from muscle and cultured cell extracts were used for immunoprecipitation and/or immunoblotting as described (29Li Y. Eitan S. Wu J. Evans C.J. Kieffer B. Sun X. Polakiewicz R.D. Mol. Cell. Biol. 2003; 23: 6255-6266Crossref PubMed Scopus (51) Google Scholar). Protein Kinase Assays—PKC and Akt kinases (50 milliunits per assay, normalized for equivalent specific activity) and 200 μm ATP were incubated with 10 μg of purified GST-IRS1-C (900–1235) or paramyosin-MARCKS (amino acids 149–160 of MARKCS) fusion proteins at 37 °C for 30 min in 50 μl (final volume) of kinase buffer (25 mm Tris, pH 7.5, 5 mm β-glycerolphosphate, 2 mm dithiothreitol, 0.1 mm Na3VO4, and 10 mm MgCl2). Lipid mix (0.5 mg/ml phosphatidylserine and 0.05 mg/ml diglycerides) was added in the PKC kinase assays. The reaction was terminated with 25 μl of 3× SDS sample buffer (187.5 mm Tris-HCl, pH6.8, 6% w/v SDS, 30% glycerol, 150 mm dithiothreitol, 0.03% w/v bromphenol blue), and samples were immunoblotted as described (29Li Y. Eitan S. Wu J. Evans C.J. Kieffer B. Sun X. Polakiewicz R.D. Mol. Cell. Biol. 2003; 23: 6255-6266Crossref PubMed Scopus (51) Google Scholar). The paramyosin-MARCKS fusion protein was prepared by incubating the MARKCS peptide with bacterially expressed paramyosin and isolated using the IMPACT-CN system (New England Biolabs). The GSK3β-paramyosin fusion protein was from Cell Signaling Technology. Expression Constructs—The wild-type IRS1-HA cDNA construct was kindly provided by Michael J. Quon. A point mutaion of Ser1101 to alanine was generated using site-directed mutagenesis by Bio S&T Inc. (Montreal, Canada). Wild-type murine PKCθ was cloned into pEYFP-C1 (BD Biosciences) to create PKCθ-YFP fusion protein. Site-directed mutagenesis generated the dominant negative PKCθ construct by changing lysine 409 to arginine in the ATP binding domain. Constitutively active PKCθ was created by changing alanine 148 to glutamate in the pseudo-substrate domain. All point mutations were confirmed by sequencing. The PKCθ mutations have been characterized functionally before (30Baier-Bitterlich G. Uberall F. Bauer B. Fresser F. Wachter H. Grunicke H. Utermann G. Altman A. Baier G. Mol. Cell. Biol. 1996; 16: 1842-1850Crossref PubMed Google Scholar). The critical role of the PI3K/Akt cascade in insulin signaling prompted us to search for potential serine phosphorylation sites in IRS1 embedded within a putative Akt phosphorylation motif. We searched for such motifs within the IRS1 sequence using the Scansite program (scansite.mit.edu) (31Yaffe M.B. Leparc G.G. Lai J. Obata T. Volinia S. Cantley L.C. Nat. Biotechnol. 2001; 19: 348-353Crossref PubMed Scopus (464) Google Scholar). Scansite predicted with high stringency at least three potential Akt sites in human IRS1, Ser307, Ser330, and Ser1101 (Ser302, Ser325, and Ser1095 in mouse IRS1, respectively). Earlier studies have proposed that Ser307 and Ser330 were involved in positive regulation of human IRS1 (8Paz K. Liu Y.F. Shorer H. Hemi R. LeRoith D. Quan M. Kanety H. Seger R. Zick Y. J. Biol. Chem. 1999; 274: 28816-28822Abstract Full Text Full Text PDF PubMed Scopus (168) Google Scholar). Recently, the phosphorylation of Ser307 (Ser302 in mouse/rat IRS1) has been confirmed (32Giraud J. Leshan R.L. Lee Y.H. White M.F. J. Biol. Chem. 2003; 279: 3447-3454Abstract Full Text Full Text PDF PubMed Scopus (82) Google Scholar). We focused on Ser1101 located at the COOH-terminal domain of human IRS1, which is highly conserved in mouse IRS1 as well as in human and mouse IRS-2 (supplemental Table I). To determine whether Ser1101 is phosphorylated in vivo, we raised an anti-IRS1-Ser1101 phosphopeptide antibody (see “Experimental Procedures”). This antibody was tested using extracts from C2C12 muscle cells (Fig. 1A, right panel) and 3T3L1 adipocytes (Fig. 1A, left panel) stimulated with insulin. Extracts were prepared, and IRS1 was immunoprecipitated and immunoblotted with the phospho-Ser1101 antibody. Ser1101 became phosphorylated upon insulin stimulation within 15 min. Because of the high homology around the Ser1101 site, the anti-phospho-Ser1101 antibody also detected Ser1149 in human IRS2 immunoprecipitated from CHO-IR-IRS2 cells stimulated with insulin (data not shown). To verify the specificity of the phospho-Ser1101 antibody, CHO cells overexpressing IR and IRS1 were transfected with a wild-type HA-tagged IRS1 construct or a HA-tagged-IRS1 with Ser1101 mutated to alanine (S1101A). Insulin stimulated Ser1101 phosphorylation of wild-type IRS1 but not of the S1101A IRS1 protein (Fig. 1B). These results indicate that the anti-phospho-Ser1101 antibody was specific for IRS1 phosphorylated at Ser1101. We then asked whether the PI3K/Akt pathway regulates Ser1101 phosphorylation. Differentiated C2C12 myocytes were stimulated with insulin for 15 min after 30-min preincubation with inhibitors of PI3K (wortmannin), PKC (bisindoleimide I), and MEK1/2 (U0126). Ser1101 phosphorylation was not inhibited by wortmannin (Fig. 1C), or by the MEK inhibitor, U0126, suggesting that the PI3K/Akt and the ERK pathways do not regulate this site. Instead, Ser1101 phosphorylation was blocked by the PKC inhibitor, bisindoleimide I (Fig. 1C). Consistently, the phorbol ester (TPA), which activates classical and novel PKCs, induced the phosphorylation at Ser1101 without activating Akt (Fig. 1D). Ser1101 phosphorylation in differentiated C2C12 cells was also stimulated by free fatty acids (arachidonic acid and oleic acid) and TNFα without inducing Akt phosphorylation (Fig. 1D). Due to the sequence homology around Ser1101 and Ser1149 the signal detected in Fig. 1, C and D, could also derive from IRS2. Altogether these results suggest that Akt is not involved in Ser1101 phosphorylation. Proline-directed kinases known to phosphorylate IRS1 are unlikely to phosphorylate Ser1101 because it is not followed by a proline. Instead, a PKC isoform or downstream basophilic kinase responsive to fatty acids and TNFα is likely to phosphorylate Ser1101. We thus tested several PKC isoforms for their ability to phosphorylate Ser1101in vitro. A GST-IRS1 fusion protein containing the COOH terminus of IRS1, including Ser1101, was incubated with purified, active PKC isoforms α, β1, β2, γ, η, ζ, ϵ, δ, or θ (Fig. 2A). Upon completion, the reactions were immunoblotted with the anti-phospho-IRS1 Ser1101 antibody. PKCθ potently phosphorylated Ser1101in vitro (Fig. 2A). PKCδ and -γ also phosphorylated Ser1101, though less effectively, whereas all the other PKC isoforms did not. All PKC isoforms used were equally active phosphorylating a peptide containing the PKC site Ser152/156 of the MARCKS protein (33Herget T. Oehrlein S.A. Pappin D.J. Rozengurt E. Parker P.J. Eur. J. Biochem. 1995; 233: 448-457Crossref PubMed Scopus (79) Google Scholar), chemically fused to the paramyosin protein. PKCθ also phosphorylated full-length IRS1 pulled down from unstimulated CHO/IR/IRS1 cells at Ser1101 (data not shown). The Akt kinase did not phosphorylate full-length or the GST-COOH-terminal IRS1 proteins at Ser1101, although it was active on a known substrate, a GSK3β peptide including Ser9 (data not shown). Based on these results PKCθ, but not Akt, is a good candidate to be an IRS1 Ser1101 kinase. We then investigated the role of PKCθ in the phosphorylation of Ser1101 within cells. CHO-IR/IRS1 cells were co-transfected with HA-tagged IRS1 alone or together with a GFP-tagged dominant negative mutant of PKCθ. Insulin stimulated the phosphorylation of immunoprecipitated HA-tagged IRS1 at Ser1101, which was reduced to background levels in the presence of co-expressed dominant negative PKCθ (Fig. 2B, upper panel). Endogenous and transfected PKCθ was detected with a PKCθ antibody (Fig. 2B, bottom panel). Differentiated C2C12 also expressed detectable levels of endogenous PKCθ (data not shown). Phosphorylation of Ser1101 was also detected in IRS1 immunoprecipitated from muscle extracts of wild-type and Akt2 knock-out mice injected with insulin. However, no phospho-Ser1101 immunoreactivity was observed in muscle extracts of PKCθ knock-out mice (Fig. 2C). These results suggest that Ser1101 phosphorylation is a physiological event mediated by PKCθ in vivo and further confirmed that Akt2, the Akt isoform involved in glucose metabolism in liver and muscle (27Cho H. Mu J. Kim J.K. Thorvaldsen J.L. Chu Q. Crenshaw III, E.B. Kaestner K.H. Bartolomei M.S. Shulman G.I. Birnbaum M.J. Science. 2001; 292: 1728-1731Crossref PubMed Scopus (1547) Google Scholar), has no role in Ser1101 phosphorylation. Furthermore, overexpression of constitutively active GFP-tagged PKCθ in CHO-IR/IRS1 cells induced the phosphorylation of co-transfected wild-type IRS1 at Ser1101 in an insulin-independent fashion (Fig. 3A). Along with Ser1101 phosphorylation, expression of active PKCθ attenuated the insulin-induced tyrosine phosphorylation of cotransfected wild-type IRS1 but not of S1101A IRS1 (Fig. 3A). To further examine the role of Ser1101 as a mediator of PKCθ inhibitory signal, we used wild-type CHO cells, which express very low levels of endogenous IRS1. Consistent with its effect on IRS1 tyrosine phosphorylation, constitutively active PKCθ reduced insulin-stimulated Akt phosphorylation (Fig. 3B). This effect of active PKCθ on insulin signaling was blocked by overexpression of S1101A-IRS1 but not wild-type IRS1 (Fig. 3B). These results demonstrate that Ser1101 phosphorylation is necessary for the inhibitory effect of PKCθ on insulin signaling. What is the physiological role of insulin-stimulated phosphorylation of IRS1 on Ser1101? Our results suggest that Ser1101 phosphorylation could be an inhibitory feedback signal, as shown for phosphorylation of other serines in IRS1 (4Aguirre V. Werner E.D. Giraud J. Lee Y.H. Shoelson S.E. White M.F. J. Biol. Chem. 2002; 277: 1531-1537Abstract Full Text Full Text PDF PubMed Scopus (769) Google Scholar). Indeed, skeletal muscle of PKCθ knock-out mice displayed a hypersensitive insulin response measured by elevated tyrosine phosphorylation of IRS1 (Fig. 3C). This indicates that PKCθ plays a significant role down-regulating the insulin stimulus. Under pathological conditions that keep PKCθ activity chronically elevated, for instance high free fatty acid levels, preexisting Ser1101 would strongly attenuate insulin signaling (17Itani S.I. Pories W.J. Macdonald K.G. Dohm G.L. Metabolism. 2001; 50: 553-557Abstract Full Text PDF PubMed Scopus (78) Google Scholar). Reduced tyrosine phosphorylation of IRS1 caused by overexpression of catalytically active PKCθ probably mimics this scenario (Fig. 3A). In contrast, in a recent study overexpression of dominant negative PKCθ in skeletal muscle of transgenic mice was shown to cause obesity associated with insulin resistance (34Serra C. Federici M. Buongiorno A. Senni M.I. Morelli S. Segratella E. Pascuccio M. Tiveron C. Mattei E. Tatangelo L. Lauro R. Molinaro M. Giaccari A. Bouche M. J. Cell. Physiol. 2003; 196: 89-97Crossref PubMed Scopus (56) Google Scholar). We cannot explain this discrepancy, but we speculate it could be due to nonspecific or compensatory mechanisms related to overexpression of the dominant negative kinase. Our results, however, are in agreement with earlier and recent studies suggesting that PKCθ mediates inhibitory effects of free fatty acids on insulin signaling (21Yu C. Chen Y. Cline G.W. Zhang D. Zong H. Wang Y. Bergeron R. Kim J.K. Cushman S.W. Cooney G.J. Atcheson B. White M.F. Kraegen E.W. Shulman G.I. J. Biol. Chem. 2002; 277: 50230-50236Abstract Full Text Full Text PDF PubMed Scopus (1196) Google Scholar, 36Kim J.K. Filmore J.J. Sunshine M.J. Albrecht B. Higashimori T. Kim D-W. Liu Z-X. Soos T.J. Cline G. O'Brien W.R. Littman D.R. Shulman G.I. J. Clin. Invest. 2004; 114: 823-827Crossref PubMed Scopus (419) Google Scholar). It is possible that PKCθ also regulates IRS1 via Ser1101-independent mechanisms. PKCθ could directly phosphorylate other sites in IRS1 (11Greene M.W. Morrice N. Garofalo R.S. Roth R.A. Biochem. J. 2004; 378: 105-116Crossref PubMed Scopus (75) Google Scholar), or it could regulate other inhibitory kinases such as JNK and/or IKK complexes (23Altman A. Isakov N. Baier G. Immunol. Today. 2000; 21: 567-573Abstract Full Text Full Text PDF PubMed Scopus (135) Google Scholar). Whether and how these kinases work coordinately to inhibit IRS1 under physiological and pathological conditions remain to be studied. Other important questions remain open: how does insulin activate PKCθ in skeletal muscle, and whether phosphorylation of Ser1101 and other sites in IRS1 is coordinated? Our results suggest that PKCθ is a Ser1101 kinase in muscle cells, but we cannot completely rule out other kinases acting downstream to PKCθ. It is also likely that other kinases may phosphorylate Ser1101 in cells where PKCθ is not abundantly expressed. Such kinases would predictably prefer basic amino acid-containing motifs like the one around Ser1101. For example, PKCδ is highly homologous to PKCθ and can phosphorylate Ser1101in vitro (Fig. 2A). Moreover, free fatty acids infusion activates PKCδ together with inhibition of insulin signaling in the liver (35Boden G. Cheung P. Stein T.P. Kresge K. Mozzoli M. Am. J. Physiol. 2002; 283: E12-E19Crossref PubMed Scopus (185) Google Scholar). Similarly to PKCθ (Fig. 3), transfection of a catalytically active PKCδ construct blocks IRS1 tyrosine phosphorylation (11Greene M.W. Morrice N. Garofalo R.S. Roth R.A. Biochem. J. 2004; 378: 105-116Crossref PubMed Scopus (75) Google Scholar). The same report identified multiple sites in IRS1 phosphorylated by PKCδ and PKCθ in vitro, but in contrast to our results (Fig. 2A), Ser1101 was not among them. Overall, this study establishes a direct and novel mechanistic connection between the activation of PKCθ and inhibition of insulin signaling through phosphorylation of Ser1101, a novel regulatory site in IRS1. We thank Michael Comb for encouragement and helpful discussions. We also thank Michael Quon for the IRS1-HA construct and Ming Xu and Nicole Kelesoglu for the paramyosin protein and technical advise. Download .pdf (.01 MB) Help with pdf files
Referência(s)