Insulin-like growth factor-I-stimulated insulin receptor substrate-1 negatively regulates Src homology 2 domain-containing protein-tyrosine phosphatase substrate-1 function in vascular smooth muscle cells

Abstract

Vascular smooth muscle cells maintained in normal (5.6 mm) glucose respond to insulin-like growth factor-I (IGF-I) with increased protein synthesis but do not proliferate. In contrast, hyperglycemia alters responsiveness to IGF-I, resulting in increased SHPS-1 phosphorylation and assembly of a signaling complex that enhances MAPK and phosphatidylinositol 3-kinase pathways. Hyperglycemia also reduces the basal IRS-1 concentration and IGF-I-stimulated IRS-1-linked signaling. To determine if failure to down-regulate IRS-1 alters vascular smooth muscle cell (VSMC) responses to IGF-I, we overexpressed IRS-1 in VSMCs maintained in high glucose. These cultures showed reduced SHPS-1 phosphorylation, transfer of SHP-2 to SHPS-1, and impaired Shc and MAPK phosphorylation and cell proliferation in response to IGF-I. In vitro studies demonstrated that SHPS-1 was a substrate for type I IGF receptor (IGF-IR) and that IRS-1 competitively inhibited SHPS-1 phosphorylation. Exposure of VSMC cultures to a peptide that inhibited IRS-1/IGF-IR interaction showed that IRS-1 binding to IGF-IR impairs SHPS-1 phosphorylation in vivo. IRS-1 also sequestered SHP-2. Expression of an IRS-1 mutant (Y1179F/Y1229F) reduced IRS-1/SHP-2 association, and exposure of cells expressing the mutant to the inhibitory peptide enhanced SHPS-1 phosphorylation and SHP-2 transfer. This result was confirmed by expressing an IRS-1 mutant that had both impaired binding to IGF-IR and to SHP-2 IGF-I increased SHPS-1 phosphorylation, SHP-2 association with SHPS-1, Shc MAPK phosphorylation, and proliferation in cells expressing the mutant. We conclude that IRS-1 is an important factor for maintaining VSMCs in the non-proliferative state and that its down-regulation is a component of the VSMC response to hyperglycemic stress that results in an enhanced response to IGF-I.

FIGURE 1.

Hyperglycemia decreases total IRS-1 protein levels and IRS-1-mediated signaling. A, confluent VSMC cultures were maintained in medium containing 5 mm (NG) or 25 mm (HG) glucose prior to these analyses. They were changed to serum-free medium containing the same glucose concentration for 12, 24, or 48 h without (top) or with (second panel) IGF-I (50 ng/ml). The anti-β-actin antibody (loading control) is shown in the lower panel. After cell lysis, aliquots containing equal amounts of protein were separated directly by SDS-PAGE and immunoblotted (IB) with anti-IRS-1 antibody. The bar graph is representative of three independent experiments without IGF-I. Error bars, mean ± S.E. **, p < 0.01. B, the extent of IRS-1 tyrosine phosphorylation was determined by immunoprecipitating (IP) cell lysates with an anti-IRS-1 antibody and then immunoblotting with an anti-Tyr(P)⁹⁹ (PY99) antibody (first panel), and Grb-2 association with IRS-1 was determined by immunoprecipitating with an anti-Grb2 antibody and then immunoblotting with anti-IRS-1 antibody (third panel). The membranes were stripped and reprobed with either anti-IRS-1 or anti-Grb2 antibody to detect the total levels of IRS-1 (second panel) or Grb2 (fourth panel), respectively. The amount of immunoprecipitate analyzed was adjusted to correct for differences in total IRS-1 between the NG and HG cultures. The bar graph represents three independent experiments. Error bars, mean ± S.E. ***, p < 0.001 when the amount of tyrosine phosphorylation of IRS-1 or the amount of Grb2 associated with IRS-1 at 5 min in response to IGF-I is compared between NG and HG cultures. C, VSMCs expressing either LacZ or IRS-WT were serum-starved in DMEM-HG and analyzed for recombinant protein expression. Cell lysates were immunoblotted using an anti-HA antibody (top), anti-IRS-1 antibody (middle), and an anti-β-actin antibody loading control (bottom). D, confluent VSMCs expressing LacZ or IRS-WT cultured in HG were serum-deprived and stimulated without or with IGF-I (50 ng/ml) for 12 or 24 h. Equal amounts of protein were separated by SDS-PAGE and immunoblotted with anti-IRS-1 antibody (without IGF-I (top) or with IGF-I (middle)) or with anti-β-actin antibody (bottom). The bar graph is representative of three independent experiments. Error bars, mean ± S.E. **, p < 0.01.

FIGURE 2.

IRS-1 overexpression impairs IGF-I-mediated SHPS-1 phosphorylation and subsequent SHPS-1 complex assembly, leading to reduction in phosphorylation of MAPK and cellular proliferation. A, LacZ and IRS-WT cells were plated (3 × 10⁴ cells) in DMEM-HG with 2% FBS prior to exposure to IGF-I in DMEM-HG with 0.2% platelet-poor plasma. Forty-eight hours after the addition of IGF-I, the cell number was determined. **, p < 0.01 when the change in cell number in response to IGF-I in LacZ control cells is determined; *, p < 0.05 when the number of cells proliferating after exposure to IGF-I is compared between LacZ and IRS-WT cells. Error bars, mean ± S.E. B, LacZ and IRS-WT cells were grown in 6-well dishes in DMEM-HG containing 10% FBS. After wounding, they were allowed to migrate with or without IGF-I in medium containing 0.2% FBS for 48 h. The total number of cells migrating past the wound line in the predetermined areas was counted. The results shown are the mean ± S.E. of at least three independent experiments. **, p < 0.01 when the change in migrating cells in response to IGF-I in LacZ cells is compared with LacZ cells without IGF-I; *, p < 0.05 when the number of cells migrating in response to IGF-I is compared between LacZ and IRS-WT cells. C, confluent LacZ and IRS-WT cells were serum-starved for 16 h in DMEM-HG and then exposed to IGF-I for the times indicated. The extent of Shc and IRS-1 phosphorylation was determined by immunoprecipitating (IP) p52^shc (first panel) or IRS-1 (fourth panel) and then immunoblotting (IB) with an anti-phosphotyrosine (Tyr(P)⁹⁹ (PY99)) antibody. Similarly, the lysates were immunoprecipitated with anti-Grb2 antibody, and the bottom portion of the blot was used for immunoblotting for p52^shc (second panel), whereas the top portion of the same blot was immunoblotted for IRS-1 (fifth panel). An equal amount of protein from cell lysates from the same experiment was used to immunoblot for total Shc (third panel) and total IRS-1 (sixth panel). Similarly, an equal amount of protein from the same cell lysates was immunoblotted using anti-phospho-MAPK (seventh panel). The blots were stripped and reprobed using anti-β-actin antibody (bottom panel). The bar graphs are representative of at least three independent experiments and two independent transductions. Error bars, mean ± S.E. ***, p < 0.001 when the amount of Grb2 associated with Shc at 10 min in response to IGF-I is compared between LacZ and IRS-WT cells; **, p < 0.01 when the amount of phosphorylation of MAPK in response to IGF-I at 10 min is compared between LacZ and IRS-WT cells. D, confluent LacZ- and IRS-WT-expressing cells were serum-starved for 16 h in DMEM-HG and then exposed to IGF-I for 5 min, and the extent of SHPS-1 phosphorylation and SHP-2 association was determined by immunoprecipitating with an anti-SHPS-1 antibody and immunoblotting with Tyr(P)⁹⁹ (first panel) or with anti-SHP-2 antibody (second panel). The extent of IGF-IR tyrosine phosphorylation was assessed by immunoprecipitating IGF-IR and immuoblotting for phosphotyrosine (fourth panel). Lysates from the same experiments were immunoblotted for total SHPS-1 (third panel) or for total IGF-IR (fifth panel). The bar graph shows the relative increase in SHPS-1 phosphorylation for at least three independent experiments. Error bars, mean ± S.E. ***, p < 0.001 when increase in SHPS-1 phosphorylation in response to IGF-I is compared between LacZ- and IRS-WT-expressing cells.

FIGURE 3.

IGF-IR kinase phosphorylates SHPS-1 and IRS-1 binding to IGF-IR impairs its ability to phosphorylate SHPS-1 in cells and in a cell-free system. A, SHPS-1/CD generated by in vitro translation using wheat germ extract was incubated in vitro (denoted by the plus or minus sign) with or without activated IGF-IR kinase as described under “Materials and Methods.” The supernatant was analyzed by immunoblotting (IB) for phosphotyrosine (top) or SHPS-1 (bottom). The relative -fold increase in the phosphorylation of SHPS-1/CD is shown in the bar graph. Error bars, mean ± S.E. ***, p < 0.001. B, confluent LacZ and IGF-IR-WT cells were serum-starved for 16 h in DMEM-HG and exposed to IGF-I for 5 min. The extent of SHPS-1 phosphorylation was determined by immunoprecipitating (IP) SHPS-1 and immunoblotting with Tyr(P)⁹⁹ (first panel). Lysates from the same experiments were immunoblotted for total SHPS-1 (second panel). C, in similar experiments, additional samples that contained IGF-IR and SHPS-1/CD were exposed to the IGF-IR tyrosine kinase inhibitor, PQ401 15 μm (lane 3) or 50 μm (lane 5), and vehicle control (DMSO) was added to lanes 2 and 4. Half of the supernatant was analyzed by immunoblotting using Tyr(P)⁹⁹ (PY99) antibody representing IGF-IR phosphorylation (top) and SHPS-1/CD phosphorylation (bottom). Both bands were identified by electrophoretic mobility. The other half of the supernatant was analyzed by immunoblotting for total IGF-IR and total SHPS-1 (lower panels). The relative -fold increase in the SHPS-1/CD phosphorylation and suppression by the inhibitor is shown in the bar graph. Error bars, mean ± S.E. ***, p < 0.001. D, VSMCs cultured and maintained in DMEM-HG were serum-starved and then incubated with or without PQ401 (50 μm) for 1 h. IGF-I was added for 5 min. Cell lysates were immunoprecipitated with anti-IGF-IR antibody (first panel) or with anti-SHPS-1 antibody (second panel) and immunoblotted for Tyr(P)⁹⁹. Cell lysates were immunoblotted with anti-IGF-IR (third panel) or anti-SHPS-1 antibody (fourth panel). E, confluent IGF-IR-overexpressing cells were serum-starved for 16 h in DMEM-HG and then exposed to IGF-I for the times indicated. The cell lysates were immunoprecipitated with anti-IGF-IR antiserum. Some aliquots of reconstituted immunoprecipitate were incubated with SHPS-1/CD in in vitro phosphorylation assays (lanes 1–3) or with buffer alone (Con). After centrifugation, the resultant supernatants were immunoblotted using a phosphotyrosine antibody IGF-IR (top) or phospho-SHPS-1/CD (middle). Similar lysates were immunoprecipitated and immunoblotted with anti-IGF-IR antibody as a loading control (bottom). To ascertain the specificity of the bands, an antibody only control was also used. The bar graph represents the relative increase in SHPS-1/CD phosphorylation in response to IGF-I at 10 min compared with non-stimulated immunoprecipitates; ***, p < 0.001 based on at least three independent experiments. Error bars, mean ± S.E. F, in an in vitro assay, the SHPS-1/CD and activated IGF-IR were incubated with increasing concentrations of recombinant full-length IRS-1, 20 ng (lane 3) and 40 ng (lane 4), for 30 min. An aliquot of each supernatant was analyzed by immunoblotting with an anti-phosphotyrosine antibody that recognizes IGF-IR tyrosine phosphorylation (top) or SHPS-1/CD (middle). The third panel indicates the total SHPS-1/CD. The -fold increase in SHPS-1/CD phosphorylation in response to IGF-IR is shown in the bar graph. Error bars, mean ± S.E. ***, p < 0.001. The -fold decrease in SHPS-1 phosphorylation in the presence of 40 ng of IRS-1 (lane 4) is shown (bar graph) in comparison with no IRS-1 addition (lane 2). Error bars, mean ± S.E.; **, p < 0.01.

FIGURE 4.

Inhibition of IRS-1 binding to IGF-IR increases association of SHPS-1 with IGF-IR and increases SHPS-1 phosphorylation. A and B, confluent cultures expressing WT-IRS-1 were serum-starved overnight and then incubated with or without cell-permeable peptides, IGF-IR (peptide 299) (A) or IRS-1 (peptide 301) (B) at 20 μg/ml for 2 h. IGF-I was added for either 5 or 10 min. Cell lysates were immunoprecipitated (IP) with anti-IGF-IR antibody and then immunoblotted for IRS-1 (top panel) or for SHPS-1 (second panel). The blots were stripped and reprobed with anti-IGF-IR antibody (third panel). Cell lysates from the same experiments were immunoprecipitated with anti-SHPS-1 antibody and immunoblotted with Tyr(P)⁹⁹ antibody (fourth panel). The blots were stripped and reprobed with anti-SHPS-1 antibody (fifth panel). B, in addition, cell lysates were immunoprecipitated with anti-SHPS-1 antibody and immunoblotted for SHP-2 (sixth panel). Similar lysates were immunoprecipitated with anti-SHP-2 antibody and immunoblotted for IRS-1 (seventh panel) or SHP-2 (eighth panel). The top bar graph shows the relative increase in IGF-IR/SHPS-1 association for at least three independent experiments. Error bars, mean ± S.E. *, p < 0.05 when increase in IGF-IR/SHPS-1 association in response to IGF-I at 10 min is compared between IRS-WT cells with and without peptide 299. The middle bar graph shows the relative increase in SHPS-1 phosphorylation for at least three independent experiments. Error bars, mean ± S.E. ***, p < 0.001 when increase in SHPS-1 phosphorylation in response to IGF-I after 10 min is compared in IRS-WT-expressing cells with or without peptide 301. C, VSMCs cultured and maintained in DMEM-NG were serum-starved and then incubated with or without cell-permeable IGF-IR/IRS-1 peptide 301 at a concentration of 5 μg/ml for 2 h. IGF-I was added for 5 min with and without the peptide. Cell lysates from the above experiments were immunoprecipitated with anti-IRS-1 antibody and immunoblotted for either IGF-IR (upper panel) or Tyr(P)⁹⁹ (second panel). The blots were stripped and reprobed with anti-IRS-1 antibody (third panel). Cell lysates from the same experiments were immunoprecipitated with anti-SHPS-1 antibody and immunoblotted with Tyr(P)⁹⁹ antibody (fourth panel) or with anti-SHP-2 antibody (fifth panel). The blots were stripped and reprobed with anti-SHPS-1 antibody (sixth panel). The bar graph shows the relative increase in SHPS-1 phosphorylation for at least three independent experiments. Error bars, mean ± S.E. **, p < 0.01 when the increase in SHPS-1 phosphorylation in response to IGF-I is compared with or without peptide 301.

FIGURE 5.

IRS-1 sequesters SHP-2 and impairs its transfer to SHPS-1. A, VSMCs overexpressing IRS-WT and the IRS-FF mutant were serum-starved overnight and analyzed for recombinant protein expression. Cell lysates were immunoblotted (IB) for HA (top) and for β-actin (bottom). B, confluent cultures of IRS-WT and IRS-FF cells were serum-starved for 16 h and then exposed to IGF-I for the times indicated. Cell lysates were immunoprecipitated (IP) with anti-HA and immunoblotted for SHP-2 (top). The membranes were stripped and probed with anti-HA antibody (bottom). C, confluent cultures of control LacZ cells, IRS-WT cells, and IRS-FF cells were serum-starved overnight and stimulated with IGF-I for 5 or 10 min. The cell lysates were immunoprecipitated with anti-SHPS-1 and immunoblotted for Tyr(P)⁹⁹ (top panel) or SHP-2 (second panel). They were stripped and reprobed with anti-SHPS-1 antibody (third panel). Lysates from similar experiments were immunoprecipitated with anti-IGF-IR and immunoblotted for IRS-1 (fourth panel). The blots were stripped and reprobed for IGF-IR (bottom panel). D, confluent IRS-FF cell cultures were serum-starved for 14–16 h and incubated with or without cell-permeable IGF-IR/IRS-1 peptide 301 (20 μg/ml) for 2 h followed by IGF-I for either 5 or 10 min. Cell lysates were immunoprecipitated with anti-SHPS-1 and immunoblotted for IGF-IR (first panel), Tyr(P)⁹⁹ (second panel), SHP-2 (third panel), or SHPS-1 (fourth panel). Lysates were immunoprecipitated with Tyr(P)⁹⁹ (PY99) and immunoblotted with anti-p52^shc (fifth panel) or immunoblotted directly for total Shc (bottom panel). The bar graph shows the relative increase in Shc phosphorylation for at least three independent experiments. Error bars, mean ± S.E. **, p < 0.01 when increase in Shc phosphorylation in response to IGF-I is compared between IRS-FF mutant cells with or without peptide 301.

FIGURE 6.

Inhibition of IRS-1 binding to both IGF-IR and SHP-2 restores SHPS-1 phosphorylation and SHPS-1 complex assembly and enhances the cellular proliferation response to IGF-I. A, VSMCs expressing IRS-WT and the IRS-R3Q-FF mutant were serum-starved overnight and analyzed for recombinant protein expression. Cell lysates were immunoblotted (IB) for HA (top) or β-actin (bottom). B, confluent cultures of IRS-WT and IRS-R3Q-FF cells were serum-starved and then exposed to IGF-I for the indicated time periods. Cell lysates were immunoprecipitated (IP) with anti-HA and immunoblotted for IGF-IR (top panel), Tyr(P)⁹⁹ (PY99; second panel), SHP-2 (third panel), or total HA (fourth panel). Cell lysates from similar experiments were immunoprecipitated with Grb2 and immunoblotted for IRS-1 (fifth panel) or total Grb2 (sixth panel). The bar graphs are representative of at least three independent experiments. Error bars, mean ± S.E. ***, p < 0.001 when the amount of Grb2 associated with IRS-1 at 5 min in response to IGF-I is compared between IRS-WT and IRS-R3Q-FF cells. C, the cell lysates from the above experiments and similar experiments were immunoprecipitated with anti-SHPS-1 antibody and immunoblotted for IGF-IR (top panel), phosphotyrosine (second panel), SHP-2 (third panel), or total SHPS-1 (fourth panel). Similarly, lysates were immunoprecipitated with Tyr(P)⁹⁹ antibody and immunoblotted for p52^shc (fifth panel). Similar lysates were immunoprecipitated with anti-Grb2 and immunoblotted for Shc (sixth panel) or Grb2 (seventh panel). Ten micrograms of protein from the same cell lysates was immunoblotted directly for phospho-MAPK (eighth panel) or β-actin (bottom panel). The bar graphs are representative of at least three independent experiments. Error bars, mean ± S.E. **, p < 0.01 when the amount of Grb2 associated with Shc (middle graph) or amount of phosphorylation of MAPK (bottom graph) at 10 min in response to IGF-I is compared between IRS-WT and IRS-R3Q-FF cells. D, IRS-WT and IRS-R3Q-FF cells were plated (3 × 10⁴ cells) in DMEM-HG with 2% FBS prior to exposure to IGF-I in DMEM-HG with 0.2% platelet-poor plasma. Forty-eight hours after the addition of IGF-I, cell number was determined. **, p < 0.01 when the change in number of cells proliferating in response to IGF-I is compared between IRS-WT and IRS-R3Q-FF cells. Error bars, mean ± S.E.

FIGURE 7.

Proposed role of IRS-1 in regulating SHPS-1 phosphorylation under hyperglycemic conditions. A, in VSMCs under normal glucose conditions, in response to IGF-I, competitive binding of IRS-1 to IGF-IR impairs the access of SHPS-1 to IGF-I receptor kinase, thereby decreasing SHPS-1 phosphorylation. Further, IRS-1 sequesters SHP-2, leading to impaired assembly of the signaling complex on SHPS-1. B, in high glucose conditions, reduction in IRS-1 levels allows SHPS-1 improved access to the IGF-I receptor kinase, leading to enhanced phosphorylation of SHPS-1 and subsequent SHP-2 transfer. This facilitates formation of the SHP-2·Src·Shc·Grb-2 signaling complex, which augments phosphatidylinositol 3-kinase and MAPK pathway activation. C, overexpression of IRS-1 decreases SHPS-1 access to the IGF-IR kinase, SHPS-1 phosphorylation, and subsequent complex assembly. The IRS-1 double mutant with impaired binding to IGF-IR and SHP-2 allows increased SHPS-1 access to IGF-IR kinase, thus increasing SHPS-1 phosphorylation and SHP-2 transfer, thereby leading to increased phosphorylation of MAPK and cellular proliferation in response to IGF-I.