Regulation of Mitogen-activated Protein Kinase Phosphatase-1 Induction by Insulin in Vascular Smooth Muscle Cells

In this study, we examined the regulation of mitogen-activated protein kinase phosphatase (MKP-1) expression by insulin in primary vascular smooth muscle cell cultures. Insulin caused a rapid time- and dose-dependent induction of MKP-1 mRNA and protein expression. Blockade of nitric-oxide synthase (NOS) withN G-monomethyl-l-arginine acetate, and cGMP with RpcGMP, completely inhibited MKP-1 expression. Insulin-mediated MKP-1 expression was preceded by inducible NOS (iNOS) induction and cGMP production. Blockade of phosphatidylinositol 3-kinase (PI3-kinase) signaling with wortmannin inhibited insulin-mediated iNOS protein induction, cGMP production, and MKP-1 expression. To evaluate potential interactions between NOS and the mitogen-activated protein kinase (MAPK) signaling pathways, we employed PD98059 and SB203580, two specific inhibitors of ERKs and p38 MAPK. These inhibitors abolished the effect of insulin on MKP-1 expression. Only PD98059 inhibited insulin-mediated iNOS protein induction. Vascular smooth muscle cells from spontaneous hypertensive rats exhibited a marked decrease in MKP-1 induction due to defects in insulin-induced iNOS expression because of reductions in PI3-kinase activity. Treatment with sodium nitroprusside and 8-bromo-cGMP restored MKP-1 mRNA expression to levels comparable with controls. We conclude that insulin-induced MKP-1 expression is mediated by PI3-kinase-initiated signals, leading to the induction of iNOS and elevated cGMP levels that stimulates MKP-1 expression.

regulating cellular events required for cell growth, differentiation, and cell homeostasis (1)(2)(3)(4). Three major subclasses of MAPKs have been identified recently and comprise the ERK, SAPK/JNK, and p38 HOG families (1)(2)(3)(4)5). Full activation of MAPKs requires phosphorylation on critical tyrosine and threonine residues. Several upstream dual-specificity kinases catalyzing this modification have been identified (6). Once activated, these kinases are responsible for the activation and phosphorylation of additional kinases as well as a battery of regulatory proteins, including transcription factors required for the expression of genes involved in cell growth and/or differentiation (6).
The activities of all the three members of the MAPK family are regulated by reversible phosphorylation of tyrosine and threonine residues, indicating that protein phosphatases play a critical role in controlling enzyme activity. Recent studies indicate that inactivation or attenuation of MAPK signaling is mediated by a class of dual specificity protein phosphatases (7)(8). These include MKP-1 (also known as CL100, Erp, and hVH-1, which is encoded by murine gene, 3ch134 (9), MKP-2, MKP-3, PAC-1, and B23 (10). MKP-1, the most ubiquitously expressed and best studied of these phosphatases, has dual catalytic activity toward phosphotyrosine-and phosphothreonine-containing proteins and is known to inactivate ERKs, JNK, and p38 HOG in vivo as well as in vitro (7,11). MKP-1 and the other family members are principally regulated at the transcriptional level as evidenced by very low to undetectable mRNA expression in quiescent cells and a rapid induction upon treatment of cells with growth factors, as well as agents that cause oxidative stress and heat shock (7,11). MKP-1 has been implicated in a feedback loop serving to inactivate MAPK after stimulation by mitogens as well as during the cellular response to stress (7,(11)(12). Despite its high level of induction following treatment with FBS and angiotensin II (13), a function for this phosphatase in response to insulin or IGF-1 has not been established in vascular smooth muscle cells (VSMCs).
Hypertension is frequently associated with insulin-resistant states such as diabetes and obesity (14 -15). However, mechanisms linking hypertension with insulin resistance are not clear. VSMCs are a major constituent of blood vessel walls responsible for the maintenance of vascular tone (16). Accelerated VSMC growth, hypertrophy, and abnormal vascular tone play a central role in the development of hypertension (17)(18). Although alterations in insulin action of the vasculature have been proposed to contribute to atherosclerosis and the regulation of vascular tone, little is known about the pathways of insulin signaling that control vascular tone and cell growth or the mechanism of their regulation in VSMCs.
Elegant studies by Baron and co-workers (19) have shown that insulin is a potent vasodilator, and this effect of insulin is mediated by nitric oxide (NO, see Ref. 19). NO is also known to influence the growth of VSMCs. However, the mechanism of NO action remains unclear.
We have recently shown that confluent primary cultures of VSMCs isolated from spontaneous hypertensive rats (SHR) exhibit increased responsiveness to insulin in terms of MAPK activation and DNA synthesis (20). It is not known whether the observed increase in MAPK activation in SHR is due to defective regulation of MKP-1 mRNA induction resulting from alterations in intracellular insulin signaling pathways that mediate MKP-1 expression or unstable MKP-1 mRNA.
In this study, in order to gain insight into potential mechanisms linking insulin signal transduction with hypertension, we examined the kinetics of MKP-1 induction by insulin and evaluated the contribution of PI3-kinase, nitric oxide, and cGMP signaling pathway(s) in insulin regulation of MKP-1 expression using primary cultures of VSMCs isolated from SHR and normotensive Wistar Kyoto rats (WKY). The results of this study indicate that insulin increases MKP-1 mRNA expression in VSMCs mainly via the PI3-kinase/NO-cGMP signaling pathway, and the observed defects in MKP-1 expression in SHR are due to defective PI3-kinase/NOS signaling leading to reductions in cGMP which may mediate MKP-1 gene expression.

EXPERIMENTAL PROCEDURES
Materials-Minimal essential medium-␣, fetal bovine serum (FBS), antibiotics, trypsin, L-glutamine, and freezing medium were obtained from Life Technologies, Inc., ␣-[ 32 P]dCTP (specific activity, 3000 Ci/ mmol), -[ 33 P]ATP (specific activity, 2000 Ci/mmol), and 125 I-protein A were purchased from NEN Life Science Products. Type 1 collagenase was from Worthington. Antibodies against MKP-1 and iNOS were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). Antibodies against IRS-1 were obtained from Upstate Biotechnology Inc. (Lake Placid, NY). Protein A/G-agarose was from Oncogene Science (Cambridge, MA). Wortmannin, Rp-8 CPT-cyclic guanosine monophosphate (RpcGMP), 8-bromo cGMP, N G -monomethyl-L-arginine acetate (L-NMMA), PD98059, and SB203580 were from Biomol (Plymouth Meeting, PA). SDS-polyacrylamide gel electrophoresis and Western blot analyses reagents were from Bio-Rad. Rat MKP-1 cDNA was a kind gift from Dr. Jyotirmoy Kusari (Tulane University, New Orleans). Phosphatidylinositol and all other chemicals and reagents were purchased from the Sigma. The cGMP assay kit was purchased from Amersham Pharmacia Biotech. The nitric oxide Griess reaction assay kit was from Boehringer Mannheim.
Culture of VSMCs-VSMCs in primary culture were obtained by enzymatic digestion of the aortic media of male normotensive Wistar Kyoto rats (WKY) and spontaneous hypertensive rats (SHR) of 200 -220 g body weight, as described in our recent publication (20). Subcultures of VSMCs from passages 3-5 were used in all the experiments. VSMCs prepared from the two strains of rats were not contaminated with fibroblasts or endothelial cells as evidenced by Ͼ99% positive immunostaining of smooth muscle ␣-actin with fluorescein isothiocyanate-conjugated ␣-actin antibody (Sigma). All experiments on MKP-1 induction, nitric oxide, and cGMP generation were performed on highly confluent cells (9 -11 days in culture) at identical passages. Prior to each experiment, cells were serum-starved for 24 -48 h in serum-free Dulbecco's modified Eagle's medium containing antibiotics.
Northern Blot Analysis of MKP-1 mRNA Expression-Confluent serum-starved VSMCs from WKY and SHR were treated with and without insulin (0 -1000 nM) or IGF-1 (10 nM) for 0 -60 min. In some experiments, VSMCs were pretreated with various inhibitors for 30 min followed by insulin for 10 -60 min. RNA was extracted with guanidinium isothiocyanate using a Qiagen RNAeasy kit as per the manufacturer's instructions and quantitated by measuring the absorbance at 260/280 nm. Equal amounts of RNA (10 g/lane) were separated on a 1.2% agarose-formaldehyde denaturing gel, transferred overnight to a nitrocellulose membrane, and hybridized with 32 P-labeled MKP-1 cDNA, and detected by autoradiography by standard protocols (21). The membrane was stripped by boiling at 100°C for 5 min in 1% SDS and reprobed with either ␤-actin or glyceraldehyde-3-phosphate dehydrogenase. The MKP-1 mRNA and ␤-actin or glyceraldehyde-3-phosphate dehydrogenase expressions were quantitated by densitometric analyses of the autoradiograms. The MKP-1 mRNA was normalized with respect to ␤-actin or glyceraldehyde-3-phosphate dehydrogenase.
Immunoblot Analysis of MKP-1 and iNOS Protein Expression-Immunodetection of MKP-1 and iNOS proteins in control and insulintreated VSMCs was performed by Western blot analyses as described in our recent publications (22)(23)(24). Briefly, 20 -50 g of cell lysate proteins were separated by SDS-polyacrylamide gel electrophoresis and transferred to polyvinylidene difluoride membranes (22). The membranes were probed with anti-MKP-1 antibody and anti-iNOS antibody according to the manufacturer's protocols. Visualization of the primary antibody was with horseradish peroxidase-conjugated secondary antibodies by enhanced chemiluminescence.
Measurement of Cellular cGMP Levels-cGMP was extracted from control and insulin-treated VSMCs by treatment with 75% aqueous ethanol. The ethanol extracts were sonicated on ice, centrifuged at 2000 ϫ g for 15 min, and the supernatants dried. The dried samples were reconstituted in cGMP assay buffer and acetylated according to the manufacturer's protocol followed by measurement of cGMP using a highly sensitive radioimmunoassay kit from Amersham Pharmacia Biotech. cGMP levels were expressed as picomoles/mg of protein.
Immunoprecipitation and Assay of IRS-1-associated PI3-Kinase Activity-Serum-starved VSMCs were stimulated with insulin (0 -1000 nM) for 1-30 min. The cells were quickly rinsed with ice-cold PBS containing 2 mM vanadate and dropped in liquid nitrogen. The frozen dishes were thawed and the cells lysed with buffer containing 20 mM HEPES, pH 7.5, 137 mM NaCl, 1 mM MgCl 2 , 1 mM CaCl 2 , 1 mM sodium orthovanadate, 10% glycerol, 1% Nonidet P-40, a mixture of protease and phosphatase inhibitors (22)(23)(24). Insoluble material was removed by centrifugation for 15 min at 12,000 rpm at 4°C. Immunoprecipitation of cell lysates normalized to 200 g of protein were performed overnight at 4°C with 4 g of anti-rabbit IRS-1 antibody directed against the plextrin homology domain (Upstate Biotechnology Inc.). The immunocomplexes were precipitated the next day by incubation with 50 l of protein G plus/protein A-agarose beads (50% v/v, Calbiochem). The immunoprecipitates were washed exhaustively with buffers, and PI3kinase activity was assayed in the immunoprecipitates as described previously (26). The reaction products were separated by thin layer chromatography on oxalate-treated Silica Gel 60 plates in a solvent of chloroform:methanol:water:ammonia (60:47:12.5:2). Cold phosphatidylinositol 3-phosphate was used as a standard and visualized by iodine vapor. The 33 P-labeled phosphatidylinositol 3-P was identified by autoradiography and quantitated by cut and count technique.
Protein Assay-Proteins in the cellular extracts and lysates were quantitated by bicinchoninic acid (25) or by the Bradford technique (27).
Statistics-The results are presented as means Ϯ S.E. of four to six independent experiments each performed in triplicate at different times. Unpaired Student's t test or analysis of variance was used to compare the mean values among WKY and SHR. A p value of Ͻ0.05 was considered statistically significant.

Differential Effect of Insulin on MKP-1 mRNA Induction in
VSMCs Isolated from WKY and SHR-To understand the mechanism of sustained MAPK activation in SHR, we examined the effect of insulin on MKP-1 induction. As seen in Fig.  1A, low levels of MKP-1 mRNA are expressed in VSMCs in the basal state in both WKY and SHR. Insulin as well as IGF-1 treatment for 30 min caused a 100% increase over basal levels in MKP-1 mRNA expression in WKY. In contrast, VSMCs from SHR exhibited a lower basal as well as a significant reduction in insulin/IGF-1-mediated MKP-1 mRNA expression (Fig. 1). Kinetic analyses of MKP-1 mRNA induction in VSMCs treated with insulin revealed a 2.5-fold increase after 10 min in WKY ( Fig. 2A). The stimulation was sustained for 30 min with a return toward basal levels in 60 min. The time course of MKP-1 mRNA induction by insulin is comparable with the previously reported kinetics with 10% FBS and angiotensin II (13). Although VSMCs from SHR exhibited an approximate 2-fold increase in MKP-1 mRNA induction over the basal levels after 10 min of insulin treatment, the stimulation was not sustained for longer periods. This may be due to instability of the mes-sage. Over all a 50 -70% reduction in insulin-stimulated MKP-1 mRNA expression was observed in SHR at all time points studied when compared with WKY ( Fig. 2A). The insulin effect on MKP-1 mRNA expression was concentration-dependent with a maximum response seen between 1 and 10 nM insulin and the effect sustained up to 100 nM insulin (Fig. 2B). Higher concentrations of insulin (above 100 nM) appear to inhibit MKP-1 mRNA expression. VSMCs isolated from SHR exhibit resistance to insulin at all concentrations tested. Western blot analyses of MKP-1 protein accumulation 1 h after insulin treatment revealed a dose-dependent increase in MKP-1 protein over the basal levels in WKY, whereas MKP-1 induction in SHR in response to insulin was markedly decreased (Fig. 2C). Treatment with 10 g/ml IGF-1 receptor antibody (␣-IR3) did not abolish the effects of insulin on MKP-1 induction (data not shown). This observation together with the fact that low concentrations of insulin induce MKP-1 expression suggest that insulin mediates MKP-1 expression through interaction with its own receptor rather than the IGF-1 receptor.
Analyses of Potential Signaling Components That Mediate Insulin-stimulated MKP-1 Expression-A number of recent studies suggest that activation of PI3-kinase is crucial for the metabolic effects of insulin and DNA synthesis in many cell types (28 -30). These observations together with the recent reports that the vasodilatory effects of insulin are mediated via NO production (19, 31-32) prompted us to examine the role of PI3-kinase and NO signaling pathway in insulin-induced MKP-1 expression. The signaling through these pathways was blocked by pretreatment of VSMCs with selective inhibitors of PI3-kinase, NOS, and cGMP, the downstream effector of NOS. Incubation of VSMCs with 1) 100 nM wortmannin, a selective PI3-kinase inhibitor, 2) 1 mM L-NMMA, a potent NOS inhibitor, and 3) 100 M RpcGMP, a cGMP antagonist, for 30 min prior to insulin exposure resulted in a complete loss of the stimulatory effect of insulin on MKP-1 mRNA expression (Fig. 3, compare  lanes 3-5 with lane 2). Furthermore, 100 nM sodium nitroprusside (SNP), the NO donor as well as 100 M 8-bromo-cGMP, a cGMP agonist, mimicked the effect of insulin on MKP-1 induction (Fig. 3, compare lanes 7 and 8 with lane 2). Treatment of VSMCs with 100 nM insulin together with SNP or 8-bromo-cGMP did not cause an additive effect on MKP-1 expression (Fig. 3, compare lanes 6 and 9 with lane 2). More importantly, the magnitude of MKP-1 induction in response to 100 nM SNP (the NO donor) and cGMP agonist (a downstream effector of NOS) was comparable between WKY and SHR (see Fig. 3, lanes  7 and 8). These results suggest that defects in NO generation because of defective NOS may be responsible for the observed impairment in insulin-mediated induction of MKP-1 in SHR.
As expected, inhibition of insulin receptor tyrosine kinase activation by pretreatment with erbstatin A or herbimycin completely blocked the subsequent effect of insulin on MKP-1 mRNA expression (data not shown). NOS inhibitor and cGMP antagonist did not block FBS-induced MKP-1 expression, suggesting that multiple signaling pathways may be involved in MKP-1 expression. Both the inhibitors also caused a modest decrease in basal expression of MKP-1 mRNA (data not shown).
VSMCs express inducible form of NOS (iNOS, see Ref. 33). To evaluate the role of the NO signaling pathway in insulinmediated MKP-1 induction and to identify potential defects in this pathway in SHR further, we measured the kinetics of iNOS protein induction and cGMP generation in insulin-stimulated VSMCs. As seen in Fig. 4A, insulin caused a rapid time-dependent increase in iNOS protein expression in WKY (2-3-fold increase over basal levels). Maximal increase in iNOS protein expression was observed after 5 min, and the effect was sustained for 30 min and returned to basal levels after 1 h. In contrast, VSMCs from SHR were resistant to insulin as evidenced by a lack of insulin-induced iNOS protein expression (Fig. 4A). Inhibition of PI3-kinase with wortmannin blocked the stimulatory effect of insulin on iNOS protein expression in WKY (Fig. 4B).
Insulin-mediated iNOS protein induction was accompanied by a time-dependent increase in cGMP levels. A Ͼ3-fold increase in cellular cGMP levels was observed after 10 min of exposure to insulin with a return to basal levels in 60 min (Fig.  5A). VSMCs from SHR exhibited a 40% reduction in basal cGMP levels and a marked impairment (Ͼ80%) in cGMP production in response to insulin when compared with WKY (Fig.  5, A and B). However, SNP-induced cGMP production was comparable between WKY and SHR (Fig. 5B). This confirms that the defect in SHR is at the level of NO production and not due to defective guanylylcyclase activity. Inhibition of PI3kinase with wortmannin and iNOS with L-NMMA completely prevented insulin-mediated cGMP production (Fig. 5B). Wortmannin and LNMMA did not block SNP mediated cGMP production (data not shown).
Effect of Insulin on PI3-Kinase Activation in SHR and WKY-Our results on the inhibitory effects of wortmannin on insulin-mediated MKP-1 and iNOS induction suggested that PI3-kinase signaling may mediate the effects of insulin on iNOS induction, cGMP production, and MKP-1 expression. To evaluate further whether insulin differentially activates PI3kinase in WKY and SHR, we measured PI3-kinase activity in IRS-1 immunoprecipitates. As seen in Fig. 6, A and B, insulin rapidly stimulates IRS-1-associated PI3-kinase activity in a dose-dependent manner in both cell types. In WKY, 10 -100 nM insulin caused a 2-3-fold increase in IRS-1 associated PI-3 kinase activity. A maximal stimulation (Ͼ6 fold over basal) of IRS-1-associated PI3-kinase was observed with 1000 nM insulin. When compared with WKY, SHR exhibited 50 -70% decrease in PI3-kinase activity at all insulin concentrations tested, although fold activation over basal value was comparable between the two cell types.
Analyses of Interaction between NOS and MAPK Signaling Pathways-A number of recent studies in different cell types suggest that MKP-1 induction is regulated by MAPK family members (7, 34 -36). Therefore, to evaluate the potential contribution of MAPK signaling pathway in insulin-mediated iNOS and MKP-1 induction, cells were pretreated with PD98059 (a specific MEK inhibitor to inhibit ERKs, Ref. 20) and SB203580, a specific inhibitor of p38 HOG MAPK, for 30 min followed by insulin exposure for 30 min. The induction of iNOS protein and MKP-1 mRNA was examined. As seen in Fig. 7, A were treated with insulin (1-100 nM) for 30 min, and proteins were allowed to accumulate for 1 h followed by extraction and analyses of MKP-1 proteins by SDS-polyacrylamide gel electrophoresis and immunoblotting with MKP-1 antibody. A representative autoradiogram is shown. Defective MKP-1 expression in SHR is accompanied by impaired iNOS protein induction. VSMCs were treated with 100 nM insulin for indicated times followed by extraction of protein in lysis buffer containing 1% Triton X-100. Equal amounts of proteins (50 g) were separated on 7.5% denaturing gel followed by Western blot analyses with a polyclonal iNOS antibody. A representative autoradiogram is shown. Similar results were obtained in four separate experiments. B, wortmannin blocks insulin-induced expression of iNOS in WKY. VSMCs were treated with 100 nM wortmannin for 30 min followed by insulin treatment for 30 min. Cell lysates were examined for iNOS as detailed above. A representative autoradiogram is shown. and B, PD98059 completely blocked insulin-mediated expression of iNOS and prevented the subsequent effect of insulin on MKP-1 expression. As reported in our recent publication (20), in addition to the observed inhibition of iNOS and MKP-1, PD98059 also blocked growth stimulatory effects of insulin in VSMCs by inhibiting insulin-mediated MAPK activation (20). In contrast, SB203580, a p38 HOG kinase inhibitor, did not affect insulin-induced iNOS but inhibited MKP-1 expression to below basal levels (Fig. 7B). These observations together with the fact that insulin does not affect p38 HOG MAPK activity suggest that ERKs may regulate NOS and MKP-1 induction, but the other MAPK family members may coordinate with ERKs in regulating MKP-1 using additional unidentified signaling pathways. DISCUSSION The results presented in this study clearly indicate that low concentrations of insulin rapidly induce MKP-1 mRNA and protein expression in VSMCs. The insulin-induced MKP-1 mRNA expression appears to be transient with a return toward basal levels within 60 min. The time course of MKP-1 induction by insulin parallels our recently reported kinetics of MAPK inactivation in VSMCs isolated from WKY (20). Thus, MKP-1 may act as an inhibitory feedback signal in attenuating MAPK signaling in VSMCs.
Several lines of evidence presented in this study suggest that MKP-1 induction by insulin in VSMCs is mediated via PI3kinase-initiated NOS-cGMP signaling pathway. First, inhibition of PI3-kinase with wortmannin blocked insulin-mediated iNOS, cGMP, as well as MKP-1 mRNA induction. Second, blockade of NOS activity with L-NMMA inhibited the effect of insulin on MKP-1 mRNA induction. Third, treatment with RpcGMP, a cGMP antagonist, prevented insulin-induced MKP-1 mRNA expression. Finally, treatment with cGMP agonist, 8-bromo-cGMP, as well as SNP, a NO donor, mimicked the effect of insulin on MKP-1. Moreover, the effects of SNP, 8-bromo-cGMP, and insulin were not additive. More importantly, the NOS-cGMP signaling pathway appears to be dependent upon MAPK signaling as inhibition of ERKs completely blocks iNOS induction by insulin. These results suggest a complex cross-talk between ERKs and NOS-signaling pathway.  5. A, insulin-mediated iNOS induction is accompanied by increased production of cAMP. VSMCs were incubated with insulin (100 nM) for the indicated times followed by extraction and assay of cGMP levels as detailed in the text. Results are the mean of two independent experiments each performed in duplicate. B, wortmannin (Wort) and L-NMMA block insulin (Ins)-mediated cGMP production. VSMCs were pretreated wortmannin and L-NMMA for 30 min followed by insulin for 10 min. Cellular contents of cGMP were measured by radioimmunoassay. Results are the mean Ϯ S.E. of four separate experiments performed in duplicate. * denotes p Ͻ 0.05 versus control; **, versus insulin.
FIG. 6. Insulin increases PI3-kinase activity associated with IRS-1; defective MKP-1 induction in SHR is accompanied by reductions in PI3-kinase activity. VSMCs were treated with various concentrations of insulin (0 -1000 nM) for 5 min. Equal amounts of precleared cell lysate proteins were immunoprecipitated with IRS-1 antibody at 4°C overnight. The immunoprecipitates were captured using 50 l (50% v/v) of protein G plus/protein A-agarose followed by the assay of PI3-kinase activity in the immunoprecipitates. The PI3kinase reaction products were extracted and separated by thin layer chromatography. Details are given under "Experimental Procedures." A, an autoradiogram from a representative experiment is shown. B, the radioactivity from TLC plate was cut out, eluted, and counted in the scintillation counter. Activity is expressed as femtomoles of [ 33 P] incorporated into phosphatidylinositol 3-phosphate (PIP)/mg protein/min. Results are the mean of two independent experiments.
In contrast to WKY, VSMCs from SHR exhibit a marked insulin resistance in terms of MKP-1 induction. The observed impairment in insulin-induced MKP-1 expression may be responsible for the sustained MAPK activation and increased cell growth observed by us in VSMCs isolated from SHR (see Ref. 20). To our knowledge, this is the first study to demonstrate insulin-mediated expression of MKP-1 mRNA in VSMCs and its abnormal induction in hypertension. Other studies have demonstrated induction of MKP-1 by FBS, AII, and plateletderived growth factor (13). Studies by Lai et al. (37) reported that balloon injury of rat carotid artery was accompanied by a marked decrease in MKP-1 mRNA expression, whereas p44 MAPK activity was increased. The results presented in this study add a new dimension to the observations that sustained MAPK activation seen in hypertension and atherosclerosis may be due to inherent reductions in MKP-1 mRNA expression resulting from defective regulation of MKP-1 gene expression in response to insulin and IGF-1. These observations led us to examine the precise intracellular insulin signal transduction pathway(s) that mediate MKP-1 expression.
Given the knowledge that insulin acts as a potent vasodilator and increases the production of NO in endothelial cells (38), we examined the contribution of the NO signaling pathway in insulin-mediated MKP-1 expression in VSMCs. In the initial studies, we observed that incubation of VSMCs with low concentrations of SNP, an exogenous NO donor, rapidly induced MKP-1 mRNA expression. The effects of SNP and insulin were not additive. Furthermore, inhibition of NOS signaling by treatment of VSMCs with L-NMMA, a potent NOS inhibitor, completely blocked insulin-induced MKP-1 expression. In addition, blockade of cGMP, a downstream effector of NOS signaling by treatment with RpcGMP, a cGMP antagonist, prevented insulin-induced MKP-1 expression. These preliminary results suggested that NOS signaling pathway may play a significant role in insulin-mediated MKP-1 expression. Further confirmation of the role of NOS/cGMP signaling in insulinmediated MKP-1 expression came from studies demonstrating a rapid induction of iNOS protein by insulin. The insulin effect on iNOS was accompanied by a rapid increase in cellular cGMP levels. Pretreatment of VSMCs with the NOS inhibitor, L-NMMA, completely blocked cGMP production and MKP-1 expression. Treatment of VSMCs with 8-bromo-cGMP, a cGMP agonist, also mimicked the effect of insulin on MKP-1 expression and bypassed the inhibitory effects of L-NMMA. The effects of cGMP and insulin on MKP-1 expression were not synergistic. These results clearly indicate that the NOS/cGMP signaling pathway may play a dominant role in insulin-mediated MKP-1 expression. In support of our observations, studies by Sugimoto et al. (39) indicate that atrial natriuretic peptide, a potent vasorelaxing factor, inhibits MAPK activation and cell proliferation by inducing MKP-1 via activation of guanylate cyclase.
The inducible form of NOS is the predominant isoform of the NOS family of proteins that is expressed in VSMCs (40), although recent studies suggest that VSMCs may express constitutive endothelial NOS as well (40). In the present study we observed that induction of iNOS protein was rapid upon treatment with insulin and the protein levels return to basal values in 1 h, suggesting a rapid turnover.
To understand further the mechanism whereby insulin regulates NOS/cGMP signaling in VSMCs, we examined the contribution of the PI3-kinase pathway in insulin-mediated iNOS induction. Insulin caused a rapid time-and dose-dependent increase in IRS-1-associated PI3-kinase activity. Treatment with wortmannin, a potent PI3-kinase inhibitor, completely blocked PI3-kinase activation by insulin and inhibited the induction of iNOS, cGMP production, and MKP-1 mRNA expression. Studies by Zheng and Quon (38) in human vascular endothelial cells suggested that PI3-kinase may participate in the vasodilatory effects of insulin by increasing the production of NO. In addition to MKP-1 induction, PI3-kinase activation is required for MAPK activation and cell growth as wortmannin blocked the stimulatory of insulin on MAPK activation and mitogenesis (20). Thus, it appears from the current study as well as from our previous results that insulin-mediated PI3kinase activation leads to activation of distinct signal transduction pathways that mediate MAPK activation and cell growth as well as MKP-1 induction to terminate MAPK signaling.
VSMCs from SHR exhibited marked reductions in PI3-kinase activity, iNOS induction, and cGMP production in response to insulin. However, the effects of cGMP agonist and SNP on MKP-1 induction were preserved in SHR. These observations suggest that the impaired induction of MKP-1 in SHR is due to defects upstream of NO, at the level of PI3-kinase/ NOS signaling.
Several studies using NIH3T3 fibroblasts and other cell types indicate that MKP-1 induction may be mediated by SAPK signaling and/or ERK signaling pathways (34 -35). Our preliminary studies with anisomycin, a potent stimulator of JNK/SAPK, indicate that MKP-1 mRNA expression can be induced by anisomycin in VSMCs with a resultant inhibition of mitogenesis. Moreover, inhibition of ERKs and p38 MAPK signaling with PD98059 and SB203580, respectively, blocked insulin-mediated induction of MKP-1 mRNA expression. The above findings suggested the existence of a complex, cell typespecific cross-talk between the ERKs/SAPK and p38 MAPK signaling cascades. Alternatively, MAPK signaling pathways may be interacting with the NOS pathway to cause MKP-1 induction. To test this possibility, VSMCs were exposed to PD98059 and SB203580, respectively, followed by insulin treatment and examined for iNOS induction and MKP-1 expression. Inhibition of ERKs with PD98059 completely blocked the effects of insulin on iNOS induction and MKP-1 mRNA expression. In contrast to PD98059, inhibition of p38 HOG kinase activity with SB203580 did not affect iNOS induction by insulin but prevented MKP-1 mRNA expression. Given that insulin increases only ERKs activity in VSMCs and does not affect p38 HOG kinase activity, it appears that ERK signaling pathways may cross-talk with iNOS signaling pathway to regulate MKP-1 induction in response to insulin.
Several recent studies suggest that p38 MAPK activity increases upon the withdrawal of serum, and the addition of growth factors as well as insulin inhibit p38 MAPK activity (41). The presence of detectable levels of phosphorylated p38 MAPK in VSMCs under basal conditions suggests that this enzyme may be needed to maintain adequate expression of MKP-1 mRNA in the basal state. Upon insulin treatment VSMCs may use the MAPK-mediated NOS signaling pathway to cause MKP-1 expression. Further studies with constitutively active as well as dominant negative mutants of ERK and p38 MAPK will help in understanding the exact contribution of ERKs and p38 in NOS activation and MKP-1 induction. It should be noted that the NOS signaling pathway does not directly control MAPK activation in VSMCs, as inhibition of NOS with L-NMMA did not prevent ERK activation but increased its activation status presumably due to inhibition of MKP-1 expression.
In VSMCs from SHR, it appears that defective MKP-1 induction by insulin may be due to reductions in PI3-kinase-generated signals leading to impaired induction of iNOS and reduced cGMP production. The above defects in MKP-1 induction could be corrected by direct treatment with SNP, a NO donor, as well as with cGMP agonist.
In summary, the results of the present study indicate that insulin induces MKP-1 expression in VSMCs by employing PI3-kinase-initiated signals leading to the induction/activation of iNOS resulting in cGMP production that may mediate MKP-1 gene expression. The observed impairment in MKP-1 induction in SHR is due to defective NOS signaling leading to sustained MAPK activation and accelerated cell growth. This study also highlights the possibility of potential cross-talk between MAPK and NOS signaling pathways.