Original Research Article

Transcription

1 IUBMB Life, 49: , 2000 Copyright c 2000 IUBMB /00 $ Original Research Article Regulation of Cardiac Jak-2 in Animal Models of Insulin Resistance Fernanda Alvarez Rojas, 1 Carla R. O. Carvalho, 1 Veronica Paez-Espinosa, 2 and Mario J. A. Saad 1 1 Departamento de Cl õ nica Médica, FCM 2 Departamento de Fisiologia e Biof õ sica, Instituto de Biologia, UNICAMP, Campinas, SP, Brasil Summary Insulin induces phosphorylation and activation of JAK2 tyrosine, as well as its association with STAT1 and SHP2 in insulinsensitive tissues of intact rats, thus demonstrating a new pathway in transduction of insulin signals. We investigated this pathway in hearts of rats in three situations of insulin resistance: 72 h of fasting, chronic treatment with dexamethasone, and acute treatment with epinephrine. The acute treatment with epinephrine showed no difference in insulin-induced JAK2 tyrosine phosphorylation or JAK2/STAT1 and JAK2/SHP2 association in comparison with the control. In fasted rats the JAK2 protein concentration decreased, accompanied by a decrease in the stoichiometry of the phosphorylation to 70%, an increase in association of JAK2/STAT1 to 160%, and a decrease in JAK2/SHP2 association to 85%. In the dexamethasone-treated group, the JAK2 protein concentrations increased but the stoichiometry of its phosphorylation decreased to 20%, whereas the JAK2/STAT1 and JAK2/SHP2 associations changed by 70% and 170%, respectively. In fasting and dexamethasone-treated rats, therefore, insulin-induced JAK2 tyrosine phosphorylation decreases, and the JAK2 protein expression is differentially regulated such that the insulin-induced JAK2 association with SHP2 and STAT1 shows opposite interactions with the kinase. IUBMB Life, 49: , 2000 Keywords Insulin; insulin action; insulin resistance; JAK2; SHP2; STAT-1. The insulin receptor is the principal mediator of insulin action in the cellular mitogenic and metabolic responses to this hormone. After binding insulin, the intrinsic tyrosine kinase activity of the transmembrane b subunit of the receptor increases, Received 2 November 1999; accepted 28 March Address correspondence to Mário J. A. Saad, MD, Departamento de Cl õ nica Medica, FCM-UNICAMP, Campinas, SP, Brasil, Fax: ; fcmadm@head.fcm.unicamp.br allowing it to phosphorylate itself as well as intracellular substrates (1). Numerous studies have shown that the insulin receptor has various substrates, including insulin receptor substrate 1 (IRS-1), 3 IRS-2, IRS-3, Shc, and possibly others (2 9). The Janus kinase (JAK) family of protein tyrosine kinases constitutes a novel signal transduction pathway activated in response to a wide variety of polypeptide ligands. The kinases of this family are generally involved in intracellular cross-talk between different signaling pathways. Currently, the JAK family of nonreceptor protein tyrosine kinases has four known members: JAK1 (10), JAK2 (11), JAK3 (12), and Tyk2 (13). After activation, JAKs phosphorylate STAT (signal transducer and activator of transduction) proteins, which then form homodimers or heterodimers before translocating to the nucleus, where they activate target genes by interaction with speci c DNA sequences (14 17). Recently, insulin-induced JAK2 activation was shown to be accompanied by association with STAT1 (Carvalho, Thironi, Velloso, and Saad, unpublished). Insulin also stimulated the association of the JAK2 with the phosphotyrosin e phosphatase 1D (also known as SHP2, PTP2C, or Syp). The SHP2 isoenzyme has been suggested to act as a bridge or an adapter between tyrosine phosphorylated IRS-1 and JAK2, and one possible function of constitutive SHP2 JAK2 binding is to translocate SHP2 to the membrane to facilitate association with IRS-1. SHP2 activity may also be responsible for the dephosphorylation of the kinase JAK2 in HIRc cells (18). However, the modulation of these new insulin signaling pathways were not investigated in insulin-sensitive tissues of intact animals. Prolonged fasting, an excess of glucocorticoids, or excess epinephrine is well established to induce insulin resistance in rats (19 21). These are relevant models in which insulin resistance is accompanied by different amounts of insulin. We have 3 Abbreviations: IRS, insulin receptor substrate; ITT, insulin tolerance test; JAK, Janus kinase; SDS-PAGE, sodium dodecyl sulfate polyacrylamide gel electrophoresis; STAT, signal transducer and activator of transduction. 501

2 502 ROJAS ET AL. previously demonstrated differential regulation of the early steps in insulin action in the liver and muscle of these animal models. Although heart muscle has been shown to be an excellent model for the study of insulin action (22), much less is currently known about cardiac insulin resistance, in contrast to the responses in liver and muscle. In the present study, we investigated the regulation of insulin-induced JAK2 tyrosine phosphorylation and its association with STAT1 and SHP2 in heart tissue of rats in three models of insulin resistance: 72 h of fasting, chronic dexamethasone treatment, and acute epinephrine treatment. MATERIALS AND METHODS Materials. The reagents and apparatus for sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and immunoblotting were obtained from Bio-Rad (Richmond, CA). Tris, phenylmethylsulfonyl uoride, aprotinin, silicone, and dithiothreitol were obtained from Sigma Chemical Co. (St. Louis, MO). Sodium amobarbital and human recombinant insulin (Humulin R) were purchased from Eli Lilly Co. (Indianapolis, IN). Protein A Sepharose 6MB was purchased from Pharmacia (Uppsala, Sweden). 125 I-labeled protein A was obtained from Amersham (Aylesbury, UK), and nitrocellulose (BA85; 0.2 l m pore size) was obtained from Schleicher and Schuell (Keene, NH). Male Wistar rats were from the State University of Campinas (UNICAMP) Central Animal Breeding Center. Monoclonal anti-phosphotyrosine antibodies were obtained from Upstate Biotechnology (Lake Placid, NY). Anti-JAK2, anti-shp2, and anti-stat1 antibodies were from Santa Cruz Biotechnology (Santa Cruz, CA). Animals. Male Wistar rats, 6 weeks old (mean SD body weight g), were fed standard rodent chow and water ad libitum. Food was withdrawn 12 to 14 h before the experiments for the epinephrine- or dexamethasone-treated rats and their respective controls. To examine the in uence of starvation, some rats were fasted for 72 h; their control rats continued to be fed. Chronic hypercortisolism was induced with dexamethasone administered for 5 days (1 mg/kg of body weight each day, intraperitoneally). In experiments with epinephrine the rats were injected with 25 l g of epinephrine per 100 g of body weight 5 min before the abdominal cavity was opened. All experiments were approved by the UNICAMP Ethics Committee. Methods. The rats were anesthetized with sodium amobarbital (15 mg/kg body weight, intraperitoneally) and samples were taken 10 to 15 min later, as soon as anesthesia was assured by the loss of foot and corneal re exes. The abdominal cavity was opened, and 1 ml of isotonic saline (0.9% NaCl), with or without 10 5 M, insulin was injected in the cava vein. The heart was removed 90 s after the infusion of insulin, minced coarsely, and homogenized immediately in 10 volumes of solubilization buffer A at 4 ± C, by using a Polytron PT 10/35 homogenizer tted with a PTA 20S blade (Brinkmann Instruments) operated at maximum speed for 30 s. Solubilization buffer A consisted of 1% Triton X-100, 50 mm Hepes (ph 7.4), 100 mm sodium pyrophosphate, 100 mm sodium uoride, 10 mm EDTA, 10 mm sodium vanadate, 2 mm phenylmethylsulfonyl uoride, and aprotinin, 0.1 mg/ml. Each heart was extracted in an identical fashion in all the models studied. The extracts were centrifuged at g in a Beckman 70.1 Ti rotor at 4 ± C for 20 min to remove insoluble material; the resulting supernatant was then immunoprecipitated with 13 l l of polyclonal anti-jak2 antibody. After reprecipitation with protein A Sepharose 6MB, the immune complexes were washed three times with 50 mm Tris (ph 7.4), 2 mm sodium vanadate, and 0.1% Triton X-100. Protein Analysis by Immunoblotting. After washing, the pellet was resuspended in Laemmli sample buffer (23) with 100 mm dithiothreitol and heated in a boiling water bath for 5 min. The samples were subjected to SDS-PAGE (6.5% Trisacrylamide) in a Bio-Rad miniature slab gel apparatus. Electrotransfer of proteins from the gel to nitrocellulose was performed for 90 min at 120 V (constant) in a Bio-Rad miniature transfer apparatus (Mini-Protean) as described by Towbin et al. (24). Nonspeci c protein binding to the nitrocellulose was reduced by preincubating the lter overnight at 4 ± C in blocking buffer (3% bovine serum albumin, 10 mm Tris, 150 mm NaCl, and 0.02% Tween 20). The prestained molecular mass standards used were myosin (206 kda), b -galactosidase (120 kda), bovine serum albumin (85 kda), and ovalbumin (47 kda). The nitrocellulose lter thus treated was next incubated for 4 h at 22 ± C with anti-phosphotyrosine antibody (0.5 l g/ml, diluted in blocking buffer) and then washed for 30 min in blocking buffer without serum albumin. The blots were then incubated with 125 I-labeled protein A (30 l Ci/l g) in 10 ml of blocking buffer for 1 h at 22 ± C and washed again. The 125 I-labeled protein A bound to the antibodies was detected by autoradiography with pre ashed Kodak XAR lm and using Cronex Lightning Plus intensifying screens at 70 ± C for 12 to 48 h. Band intensities were quanti ed by optical densitometry (Molecular Dynamics) of the developed autoradiogram. The effect of the treatments on the ability of insulin to stimulate glucose disposal was estimated by the intravenous insulin tolerance test (ITT). The ITT was determined in control rats and in treated rats after 14 h of food deprivation. The rats were anesthetized as previously described; tail blood samples were drawn before injection (0, or basal, sample) of 6 mg of insulin and at 4, 8, 12, and 16 min after hormone administration. Plasma glucose concentrations in the samples were determined by the glucose oxidase method, with use of a commercial kit (Labtest Diagnóstica). The rate constant for plasma glucose disappearance, as determined in the ITT (K itt ), was calculated by using the formula 0.693/t 1/2. The plasma glucose t 1/2 was calculated from the slope of the least-squares analysis of the plasma glucose concentrations during the linear decay phase (25) Statistical Analysis. The experiments were always performed by studying the physiological or pathological group of rats in parallel with a control group. Comparisons of the data from fed (0) versus fasted (72 h) rats, control versus epinephrine-treated

3 REGULATION OF CARDIAC JAK rats, and control versus dexamethasone-treated rats were analyzed by using unpaired Student s t test. The level of signi cance used was P < RESULTS Animal Characteristics Fasting for 72 h resulted in substantial reductions in plasma glucose and serum insulin in comparison with fed rats. To demonstrate that the fasting rats were insulin resistant, the animals were submitted to an ITT. The glucose disappearance rate (K itt ; %/min) during this test was less in fasted rats than in fed rats (fed , fasting ; P < 0.05). Dexamethasone induced a state of insulin resistance characterized by a twofold increase in plasma glucose and serum insulin and a signi cant decrease in the glucose disappearance rate (control , Dexa ; P < 0.05). Acute infusion of epinephrine did not markedly change the concentrations of plasma glucose and serum insulin, but it did induce a signi cant decrease in the glucose disappearance rate (control , epinephrine ; P < 0.05) Effect of Fasting on Insulin-Induced JAK2 Tyrosine Phosphorylation in Heart Tissues of Rats To investigate the tyrosine phosphorylation of JAK2 after stimulation by insulin in physiological and pathophysiological states of insulin resistance, we infused insulin into the vena cava of rats and then removed and homogenized the heart and immunoprecipitated the proteins with a polyclonal anti-jak2 antibody. The JAK2 immunoprecipitates were analyzed for tyrosyl phosphorylation by immunoblotting with a monoclonal anti-phosphotyrosine antibody. The nitrocellulose membranes were also stripped and reblotted with anti-shp2 and anti-stat1 antibodies, to assess the association of these proteins with JAK2. JAK2 protein in the hearts of fasted rats decreased to 68% 9% (P < 0.003) of that in fed rats, as determined by immunoprecipitation and immunoblotting with the anti-jak2 antibody (Fig. 1A). Before insulin stimulation, the extent of JAK2 tyrosine phosphorylation was higher in fed than in fasted animals. After insulin infusion, the extent of phosphorylation increased in both groups of animals but now was signi cantly higher in the fed rats (fasted 49.5% 19%, fed 100% 15%; P = 0.02) (Fig. 1B). Furthermore, when the data for fasted rats were expressed as a function of the amount of JAK2 protein, the stoichiometry of JAK2 phosphorylation decreased to 70% (P < 0.02) of that seen in fed animals (Fig.1B). Previous studies have suggested a relatively stable, highaf nity interaction between JAK2/STAT1, such that both proteins are coprecipitated by antibodies to either protein. In heart samples previously immunoprecipitated with anti-jak2 antibody and then immunoblotted with anti-stat1 antibody, a band corresponding to the latter protein was seen in both fed and fasted rats, but the amount of STAT1 associated with JAK2 was higher in heart tissues of fed rats than in 72-h fasted rats. After stimulation with insulin, the intensity of this band increased in both groups of rats, in agreement with the formation of a stable association between JAK2 and STAT1. Comparison of the bands corresponding to stimulation by insulin showed that the amount of STAT1 associated with JAK2 increased to 160% 16% (P < 0.05) in the hearts from fasted rats (Fig. 1C). When the same membranes were blotted with anti-shp2 antibody, JAK2 immunoprecipitated along with this tyrosine phosphatase in the basal state. This basal association was higher in fed than in fasted animals. Comparison of the JAK2/SHP2 association after insulin stimulation showed that in 72-h fasted rats, this association decreased to 85% 0.1% (P < 0.002) of that seen in fed rats (Fig. 1D). There was no change in concentrations of STAT1 and SHPTP2 proteins in heart tissues of fasted rats (data not shown). Effect of Dexamethasone on Insulin-Induced JAK2 Tyrosine Phosphorylation in Heart Tissues of Intact Rats. Fig. 2 shows the effect of the glucocorticoid dexamethasone on expression of JAK2 protein in rat heart. The amount of JAK2 protein increased signi cantly (463% 19%; P < 0.05) in heart tissues of rats treated with dexamethasone (Fig. 2A). Immunoblotting the homogenates with anti-phosphotyrosine antibody showed a nonsigni cant decrease (to 81% 31% of that in the control group) in the extent of insulin-induced JAK2 tyrosine phosphorylation in dexamethasone-treated rats (Fig. 2B). In contrast, correction of the data for the amount of protein present revealed a dramatic decrease in the stoichiometry of the JAK2 phosphorylation to 20% (P < 0.05) in the dexamethasone-treated rats. The insulin-induced JAK2/STAT1 association (Fig. 2C) also showed a reduction, to 70% 18 (P < 0.05) of controls, as well as a decrease in the stoichiometry (number of associated molecules per number of JAK2 protein molecules) to 15% ( P < 0.02) in heart samples from dexamethasone-treated rats (Fig. 2D). When heart extracts that had been previously immunoprecipitated with anti-jak2 antibody were blotted with anti-shp2 antibody (Fig. 2E), a signi cant increase (to 170% 9%; P < 0.05) in the insulin-induced JAK2/SHP2 association was seen in dexamethasone-treated rats, but there was a decrease in the stoichiometry (associated molecules per JAK2 protein molecule) to 36% (P < 0.02) of the control values in dexamethasone-treated animals (Fig. 2F). Heart tissues of dexamethasone-treated rats also showed an increase to 150% in SHP2 protein concentrations, but no change in STAT1. Effect of Epinephrine on Insulin-Induced JAK2 Tyrosine Phosphorylation in Heart Tissues of Rats There was no change in JAK2 protein expression after acute stimulation with epinephrine (Fig. 3A). Similarly, there was no signi cant difference in the insulin-induced JAK2 tyrosine phosphorylation between epinephrine-treated and control rats (control 100% 15%, epinephrine 78% 11%; Fig. 3B). In agreement with this, the stoichiometry of insulin-induced JAK2

4 504 ROJAS ET AL. Figure 1. Effect of fasting on extent of JAK2 (Janus kinase 2) tyrosine phosphorylation in cardiac tissue. Rats were anesthetized, and the abdominal wall was incised to expose viscera. Isotonic saline ( ) or 6 l g of insulin (+) was infused into the vena cava as a bolus injection; 90 s later the heart was excised, and the tissues were homogenized in extraction buffer A as described in Materials and Methods. The cardiac tissue from fed and fasted rats were immunoprecipitated with anti-jak2 antibody and then blotted with anti-jak2 antibody (A), anti-phosphotyrosine antibody (B), anti-stat1 antibody (C), or anti-shp2 antibody (D). The bar graphs represent the mean S.E.M. of the scanning densitometry of six experiments.

5 REGULATION OF CARDIAC JAK Figure 2. Insulin-stimulates JAK2 tyrosine phosphorylation in the cardiac tissue of dexamethasone-treated rats. Rats were anesthetized, and the abdominal wall was incised to expose viscera. Isotonic saline ( ) or 6 l g of insulin (+) was infused into the vena cava as a bolus injection; 90 s later the heart was excised and the tissues were homogenized in extraction buffer A as described in Materials and Methods. The cardiac tissue from dexamethasone-treated and control rats was immunoprecipitated with anti-jak2 and then was blotted with anti-jak2 antibody (A), anti-phosphotyrosine antibody (B), anti-stat1 antibody (C), with anti-stat1/protein antibody (D), anti-shp2 antibody (E), or anti-shp2/protein antibody (F). The bar graphs represent the mean S.E.M. of the scanning densitometry of six experiments.

6 506 ROJAS ET AL. Figure 3. Insulin stimulates JAK2 tyrosine phosphorylation in the cardiac tissue of epinephrine-treated rats. Rats were anesthetized, and the abdominal wall was incised to expose viscera. Isotonic saline ( ) or 6 l g of insulin ( +) was infused into the vena cava as a bolus injection; 90 s later the heart was excised, and the tissues were homogenized in extraction buffer A as described in Materials and Methods. The cardiac tissue from epinephrine-treated and control rats were immunoprecipitated with anti-jak2 and then were blotted with anti-jak2 antibody (A), anti-phosphotyrosine antibody (B), anti-stat1 antibody (C), or anti-shp2 antibody (D). The bar graphs represent the mean S.E.M. of the scanning densitometry of six experiments.

7 REGULATION OF CARDIAC JAK tyrosine phosphorylation did not change in the catecholaminetreated rats. Analysis of the insulin-induced JAK2/STAT1 association also showed no signi cant change in this association in epinephrinetreated rats (control 100% 1%, epinephrine 107% 5%; Fig. 3C). Similar nonsigni cant results were obtained for the insulin-induced JAK2/SHP2 association (control 100% 1%, epinephrine 93% 3%; Fig. 3D). DISCUSSION JAK2 belongs to a family of proteins that, in addition to an active kinase domain, also possess a second kinase-like domain but have no Src homology 2 (SH2), Src homology 3 (SH3), or membrane-spanning region (10 13). The receptors for erythropoietin (26), prolactin (27), growth hormone (28), and angiotensin II (29) bind to and activate JAK2. We have recently shown that after the infusion of insulin, the tyrosines of JAK2 are rapidly phosphorylated and the enzyme is activated in insulinsensitive tissues of the intact rat (30). JAK2 also is coimmunoprecipitated with the insulin receptor and with IRS-1, suggesting that this cytoplasmic kinase is involved in transduction of insulin signals (30). After activation by different hormones and cytokines, JAK2 phosphorylates STAT proteins, which in turn form homodimers or heterodimers before being translocated to the nucleus, where they activate target genes by interaction with speci c DNA sequences. We have demonstrated that insulin induces phosphorylation of the STAT1 tyrosines in a time- and dose-dependent fashion; moreover, the clear association between JAK2 and STAT1 after insulin infusion suggests that the latter is activated by the former (Carvalho et al., unpublished). The time course of insulininduced STAT1 phosphorylation and the binding of this protein to JAK2 are closely correlated. Furthermore, JAK2 has kinase activity towards STAT1 after exposure to insulin (Carvalho et al., unpublished). SHP2 has been suggested to act as a bridge or an adapter between tyrosine-phosphorylated IRS-1 and JAK2 after stimulation by insulin. One possible function of constitutive JAK2 SHP2 binding is the translocation of SHP2 to the membrane to allow easier association with IRS-1. SHP2 activity has also been implicated in the dephosphorylation of JAK2 in HIRc cells (18). Prolonged fasting inrats is characterized by insulin de ciency and insulin resistance, despite an increase in the number of insulin receptors (19, 20). Heart tissues of fasting rats have an increase in the insulin-induced phosphorylation of the insulin receptor that is paralleled by an increase in IRS-1 tyrosine phosphorylation (Silva et al., 79th annual meeting of The Endocrine Society, p. 582 of the program book). We have shown here that after a 72-h fast there is a decrease in JAK2 protein expression and a more marked decrease in insulin-induced JAK2 tyrosine phosphorylation. These results suggest a tissue-speci c regulation of amount of JAK2 tyrosine phosphorylation, given that in the liver of 72-h fasted rats the insulin-induced tyrosine phosphorylation of this kinase is increased (31). Before insulin infusion, JAK2 tyrosine phosphorylation is also higher in fed rats, probably as consequence of the higher insulin levels these animals present. The insulin-induced association of JAK2 with other proteins was differentially regulated in hearts from 72-h fasted rats, there being an increase in JAK2/STAT1 and a decrease in JAK2/SHP2 association. The reason for this divergent regulation in association, which was also observed in dexamethasone-treated rats, is at present unknown. The reductions in JAK2 protein expression and tyrosine phosphorylation in heart tissues from fasted rats may be of limited biological signi cance because the association of this kinase with STAT1, which seems to be an important mitogenic pathway, is preserved. JAK2 has been described as an alternative pathway for insulin-induced IRS-1 tyrosine phosphorylation (18). However, tyrosine phosphorylation of IRS-1 in hearts from 72-h fasted rats is increased in parallel with insulin-induced phosphorylation of insulin receptors, suggesting that the insulin receptor is the main modulator of IRS-1 tyrosine phosphoryla - tion in this situation (Silva et al., 79th annual meeting of The Endocrine Society, p. 582 of the program book). An excess of epinephrine has long been known to cause insulin resistance (21, 31 33). Catecholamines antagonize the action of insulin by stimulating gluconeogenesis, glycogenolysis, and lipolysis and by inhibiting peripheral use of glucose by way of a b -adrenergic mechanism that may also involve a decrease in cellular glucose transport (21). High intracellular concentrations of cyclic AMP seem to induce insulin resistance at both receptor and postreceptor levels (34). We have recently shown that epinephrine reduces insulin-induced phosphorylation of insulin receptors and IRS-1 tyrosines in liver, muscle, and heart (35). This decrease may be secondary to an increased activity of protein kinase A (PKA), which can phosphorylate insulin receptor and probably IRS-1 at serine, thereby reducing the kinase activity and autophosphorylation of the insulin receptors as well as the tyrosine phosphorylation of IRS-1. Epinephrine did not change the extent of insulin-induced tyrosine phosphorylation of JAK2 or the association of JAK2 with STAT1 and SHP2. The reason for this differential regulation of the insulin receptor/ IRS-1 and JAK2 in hearts from epinephrine-treated rats is dif - cult to explain, but it may indicate that PKA does not phosphorylate JAK2 at serine or that serine phosphorylation of JAK2 does not have a regulatory role. Alternatively, epinephrine may activate phosphotyrosin e phosphatases that can dephosphorylat e the insulin receptor and IRS-1 but not JAK2. An excess of glucocorticoids is a well-known situation of insulin resistance (36, 37). Hypercortisolemia is associated with increased glucose production by the liver, decreased transport and utilization of peripheral glucose, decreased protein synthesis, and increased protein degradation in muscle (38 40). In contrast to their catabolic effect on skeletal muscle, glucocorticoids are anabolic in the heart (41). The increased accumulation of total cardiac protein during glucocorticoid administration is

8 508 ROJAS ET AL. mediated entirely through increased rates of synthesis. In agreement with this, we showed here that a glucocorticoid excess markedly increased the expression of JAK2 protein in the rat heart. However, the stoichiometry of the insulin-induced phosphorylation of JAK2 tyrosine was reduced by 80% in rats exposed to excess glucocorticoid. Because JAK2 may have a role in insulin-induced phosphorylation of IRS-1 (21), the reduction in JAK2 tyrosine phosphorylation after dexamethasone treatment may contribute to the reduced IRS-1 tyrosine phosphorylation observed in rats (21), although this was not the case in fasted rats. Dexamethasone also differentially modulated the interactions of JAK2 with SHP2 and STAT1. The insulin-induced association of JAK2/SHP2 was increased, whereas that of JAK2/STAT1 was decreased in hearts from dexamethasone-treated rats. These results from fasted and dexamethasone-treated rats suggest that in situations of diminished insulin-induced phosphorylation of JAK2 tyrosine, JAK2/STAT1 and JAK2/SHP2 interact competitively, such that when one increases, the other decreases. The mechanisms responsible for this dissociation are currently unknown, but at least four possibilities should be considered. First, a differential coupling between JAK2 and STAT1 or SHP2, modulated by the reduced phosphorylation of speci c tyrosines on JAK2, may drive the interaction. Second, given the increase observed in concentrations of SHP2 protein in dexamethasonetreated rats, this change may have contributed to the increased interaction of JAK2/SHP2 in this speci c situation. Third, STAT1 and SHP2 may be competing for the same or similar binding sites on JAK2, because each of these proteins appears to have opposing interactions with JAK2. And nally the apparent competitive nature of JAK2/SHP2 may be attributed to a fourth possible mechanism involving the downstream regulation of STAT1- or SHP2-driven JAK2 inhibitors; a resulting inhibitory feedback mechanism, if present, might account for the observed opposing interactions of JAK2. The consequences of these regulations to other pathways of insulin signal transduction are currently unknown. Interestingly, however, one possible function of SHP2/JAK2 association is to translocate SHP2 to the membrane to facilitate its association with IRS-1 and to dephosphorylate IRS-1 (30); in such a case, an opposite modulation between JAK2/SHP2 association and IRS-1 tyrosine phosphorylation would be expected. We previously described that in heart tissues of intact animals, in conditions identical to those of this study, IRS-1 tyrosine phosphorylation was increased in 72-h fasted rats and decreased in dexamethasone-treated rats (Silva et al., 79th annual meeting of The Endocrine Society, p. 582 of the program book). Taking these data with the results of the present study (JAK2/SHP2 association decreased in 72-h fasted rats and increased in dexamethasone-treated rats), we suggest that JAK2/SHP2 interaction may have an important role in modulating IRS-1 tyrosine phosphorylation. In summary, in fasting and dexamethasone-treated rats (two insulin-resistant states), the insulin-induced JAK2 tyrosine phosphorylation is decreased, and the JAK2 protein expression is differentially regulated such that the insulin-induced JAK2 associations with SHP2 and STAT1 show opposite directions of response. REFERENCES 1. Wilden, P. A., Kahn, C. R., Siddle, K., and White, M. F. (1992) Insulin receptor kinase domain autophosphorylatio n regulates receptor enzymatic functions. J. Biol. Chem. 267, Sun, X. J., Rothenberg, P. L., Kahn, C. R., and White, M. F. (1991) Structure of the insulin receptor substrate IRS-1 de nes a unique signal transduction protein. Nature 352, Rothenberg, P. L., Lane, W. S., Karasik, A., Backer, J., White, M. F., and Kahn, C. R. (1991) Puri cation and partial sequence analysis of pp185, the major cellular substrate of the insulin receptor tyrosine kinase. J. Biol. Chem. 266, Tamemoto, H., Kadowaki, T., Tobe, K., Yagi, T., Sakura, H., Hayakawa, T., Terauchi, Y., Ueki, K., Satoh, S., Secchara, H., Yohioka, S., Horlkoshi, H., Furuta, Y., Ikawa, Y., Kasuga, M., Yazaki, Y., and Aizawa, S. (1994) Insulin resistance and growth retardation in mice lacking insulin receptor substrate 1. Nature 372, Araki, E., Lipes, M. A., Patti, M. A., Bruning, J. C., Haag, B., Johnsn, R. S., and Kahn, C. R. (1994) Alternative pathway of insulin signal in mice with target disruption of the IRS-1 gene. Nature 372, Sun, X. J., Wang, L. M., Patti, M. A., Bruning, J. C., Haag, B., Johson, R. S., and Kahn, C. R. (1995) Role of IRS-2 in insulin and cytokine signalling. Nature 337, Skolnik, E. Y., Lee, C. H., Batzer, A., Vicentini, L. M., Zhou, M., Daly, R., Myers, M. G., Backer, J. M., Ullrich, A., White, M. F., and Schlessinger, J. (1993) The SH2/SH3 domain-containin g protein Grb 2 interacts with tyrosine-phosphorylate d IRS-1 and Shc: implication for insulin control of ras signalling. EMBO J. 12, Okada, T., Kawano, Y., Sahihara, T., Hazeki, O., and Ui, M. (1994) Essential role of phosphatidylinisitol 3-kinase in insulin induced glucose transport and antilipolysis in rat adipocytes. Studies with selective inhibitor wortamannin. J. Biol. Chem. 269, Smith-Hall, J., Pons, S., Patti, M. E., Burks, D. J., Yenush, L., Sun, X. L., Kahn, C. R., and White, M. F. (1997)The 60 kda receptor substrate functions like an IRS protein (pp60irs3) in adipose cells. Biochemistry 36, Wilks, A. F., Harpur, A. G., Kirban, R. R., Rooph, S. J., Zuercker, G., and Ziemiecki, A. (1992) Two novel protein-tyrosine kinases, each with a second phosphotransferase-relate d catalytic domain, de ne a new class of protein kinase. Mol. Cell. Biol. 11, Harpur, A. G., Andres, A. C., Ziemecki, A., Aston, R. R., and Wilks, A. F. (1992) JAK2, a third member of JAK family of protein tyrosine kinase. Oncogene 7, Johnston, J. A., Kawamura, M., Kirken, R. A., Chen, Y. Q., Blake, T. B., Shibuya, K., Ortaldo, J. R., McVicar, D. W., and O Shea, J. J. (1994) Phosphorylation and activation of the JAK3 Janus kinase in response to interleukin-2. Nature 370, Firmbach-Kraft, I., Byers, M., Shows, T., Dalla-Favera, R., and Krolewski, J. J. (1994) Tyk2, prototype of a novel class of non-receptor tyrosine kinase genes. Oncogene 5, Ihle, J. N., Witthuhn, B. A., Quelle, F. W., Yamamoto, K., Thierfelder, W. E., Kreider, B., and Silvennoinen, O. (1994) Signaling by the cytokine receptor superfamily: JAKs and STATs. Trends Biochem. Sci. 19, Shuai, K., Horvath, C. M., Tsai Huang, L. H., Quereshi, S. A., Cowburn, D., and Darnell, J. E. (1994) Interferon activation of the transcription factor STAT1 involves dimerization through SH2 phosphotyrosyl peptide interactions. Cell 76, Tian, S.-S., Lamb, P., Seidel, H. M., Stein, R. B., and Rosen, J. (1994) Rapid activation of the STAT3 transcription factor by granulocyte colonystimulating factor. Blood 84,

9 REGULATION OF CARDIAC JAK De Koning, J. P., Dong, F., Smith, L., Shelen, A. M., Barge, R. M., Van Der Plas, D. C., Hoefsloot, L. H., Lowenberg, B., and Tonw, I. P. (1996) The membrane-distal cytoplasmatic region of a human granulocyte colonystimulating factor receptor is required for STAT3 but not STAT1 homodimer formation. Blood 87, Maegawa, H., Kashiwagi, A., Fujita, T., Ugi, S., Hasegawa, M., Obata, T., Nishiio, Y., Kojima, H., Hidaka, H., Kikkawa, R. (1996) SHPTP2 serves adapter protein linking between Janus kinase 2 and insulin receptor substrates. Biochem. Biophys. Res. Commun. 228, Penicaud, L., Kande, J., and Lemagnen, J. (1985) Insulin action during fasting and refeeding in rat determined by euglycemic clamp. Am. J. Physiol. 249, E514 E Defronzo, R. A., Hendler, R., and Simonson, D. (1982) Insulin resistance is a prominent feature of insulin-dependen t diabetes. Diabetes 31, Chiasson, J. L., Shikama, H., Chu, D. T. W., and Exton, J. H. (1981) Inhibitory effect of epinephrine on insulin-stimulated glucose uptake by rat skeletal muscle. J. Clin. Invest. 68, Eckel, J., Wirdeier, A., Herberg, L., and Reinauer, H. (1985). Insulin resistance in the heart: studies on isolated cardiocytes of genetically obese Zucker rats. Endocrinology 116, Laemmli, U. K. (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, Towbin, H., Staehlin, J., and Gordon, J. (1979) Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets. Procedure and some applications. Proc. Natl. Acad. Sci. U.S.A. 76, Bonora, E., Moghetti, P., Zancanaro, C., Cigolin, M., Queren, M., Cacciotori, V., Corgnati, A., and Muggeo, M. (1989) Estimates of in vivo insulin action in man: comparision of insulin tolerance test with euglycemia and hyperglycemic glucose clamp studies. J. Clin. Endocrinol. Metab. 68, Witthuhn, B. A., Quelle, F. W., Silvennoinen, O., Yi, T., Tang, B., Miura, O., and Ihle, J. N. (1993) JAK2 is associated with the erythropoietin receptor and is tyrosine phosphorylated and activated following stimulation with erythropoietin. Cell 74, Lebrun, J. J., Ali, S., Sofer, L., Ullrich, A., and Kelly, P. A. (1994) Prolactininduced proliferation of Nb2 cells involves tyrosine phosphorylatio n of the prolactin receptor and its associated tyrosine kinase JAK2. J. Biol. Chem. 269, Argentsinger, L. S., Campbell, G. S., Yang, X., Witthuhn, B. A., Silvennoinen, O., Ihle, J. N., and Carter-Su, C. (1993) Identi cation of JAK2 as a growth hormone receptor-associated tyrosine kinase. Cell 74, Marrero, M. B., Schieffer, B., Paxton, W. G., Heerdt, L., Berk, B. C., Delafontaine, P., and Berstein, K. E. (1995) Direct stimulation of JAK/STAT pathway by the angiotensin II AT1 receptor. Nature 375, Saad, M. J. A., Carvalho, C. R. O., Thirone, A. C. P., and Velloso, L. A. (1996) Insulin induces tyrosine phosphorylatio n of JAK2 in insulin-sensitive tissues of intact rat. J. Biol. Chem. 271, Kirsch, D., Kemmler, W., and Haring, H. U. (1983) Cyclic AMP modulates insulin binding and induces post-receptor insulin resistance of glucose transport in isolated rat adipocytes. Biochem. Biophys. Res. Commun. 197, Wallberg-Henriksson, H. (1987) Glucose transport into skeletal muscle. In- uence of contractile activity, insulin, catecholamines and diabetes mellitus. Acta Physiol. Scand. Suppl. 564, Bonen, A., Megeney, L. A., Mccarthy, S. C., Mcdermott, J. C., and Tan, M. H. (1992) Epinephrine administration stimulates GLUT4 translocation but reduces glucose transport in muscle. Biochem. Biophys. Res. Commun. 187, Kuroda, M., Honnor, R. C., Cushman, S. W., Londos, C., and Simpson, I. A. (1987) Regulation of insulin-stimulated glucose transport in the isolated rat adipocyte. CAMP-independent effects of lipolytic and antilipolytic agents. J. Biol. Chem. 262, Saad, M. J. A., Hartmann, L. G. C., Carvalho, D. S., Galoro, C. A. O., Brenelli, S. L., and Carvalho, C. R. O. (1995) Modulation of early steps in insulin action in the liver and muscle of epinephrine treated rats. Endocrine 3, Kahn, C. R., Gold ne, I. D., Neville, D. M., and Demeyts, P. (1978) Alteration in insulin binding induces changes in vivo in the levels of glucocorticoids and growth hormone. Endocrinology 103, Caro, J. F., and Amatruda, J. M. (1982) Glucocorticoid-induce d insulin resistance. J. Clin. Invest. 69, Amatruda, J. M., Livingston, J. N., and Lockwood, D. H. (1985) Cellular mechanisms in selected states of insulin resistance: human obesity, glucocorticoid excess and chronic renal failure. Diabetes Metab. Rev. 3, Knutson, V. O. (1986) The acute and chronic effect of glucocorticoids on insulin receptor and insulin responsiveness. Transient uctuations in intracellular receptor level parallel transient uctuations in responsiveness. J. Biol. Chem. 261, Moller, D. E., and Flier, J. S. N. (1991) Insulin resistance mechanisms, syndromes, and implications. N. Engl. J. Med. 325, Bensky, A. S., Kothadia, J. M., and Covitz, W. (1996) Cardiac effect of dexamethason e in very low birth weight infants. Pediatrics 97,