Absence of IQGAP1 Protein Leads to Insulin Resistance

Abstract

Insulin binds to the insulin receptor (IR) and induces tyrosine phosphorylation of the receptor and insulin receptor substrate-1 (IRS-1), leading to activation of the PKB/Akt and MAPK/ERK pathways. IQGAP1 is a scaffold protein that interacts with multiple binding partners and integrates diverse signaling cascades. Here we show that IQGAP1 associates with both IR and IRS-1 and influences insulin action. In vitro analysis with pure proteins revealed that the IQ region of IQGAP1 binds directly to the intracellular domain of IR. Similarly, the phosphotyrosine-binding domain of IRS-1 mediates a direct interaction with the C-terminal tail of IQGAP1. Consistent with these observations, both IR and IRS-1 co-immunoprecipitated with IQGAP1 from cells. Investigation of the functional effects of the interactions revealed that in the absence of IQGAP1, insulin-stimulated phosphorylation of Akt and ERK, as well as the association of phosphatidylinositol 3-kinase with IRS-1, were significantly decreased. Importantly, loss of IQGAP1 results in impaired insulin signaling and glucose homeostasis in vivo Collectively, these data reveal that IQGAP1 is a scaffold for IR and IRS-1 and implicate IQGAP1 as a participant in insulin signaling.

Keywords: Akt PKB; IQGAP1; insulin; insulin receptor; insulin receptor substrate 1 (IRS-1); insulin resistance; mitogen-activated protein kinase (MAPK); phosphatidylinositide 3-kinase (PI 3-kinase).

FIGURE 1.

IR and IQGAP1 interact in cells and in vitro. A, GST pull-down of IR from CHO/IR cell lysates using GST-IQGAP1 or GST alone analyzed by immunoblotting (top panels) and Coomassie Blue staining (bottom panels). B, immunoprecipitation (IP) with anti-IQGAP1 antibodies or NIRS from CHO/IR/IRS-1 cells transfected with Myc-IQGAP1. C, IP with anti-IQGAP1 antibodies or NIRS of endogenous IQGAP1 from C2C12 cells. D, schematic representation of IR-β and IR_cyt. TM/JM, transmembrane and juxtamembrane domains; TK, tyrosine kinase; C-term, C-terminal tail. E, GST pull-down of purified IQGAP1 using GST-IR_cyt or GST. Samples were analyzed by immunoblotting (top) and Coomassie Blue staining (bottom). Shown is GST pull-down of His₆-IR-β (F) or His₆-IR_cyt (G) using GST-IQGAP1 or GST alone. The middle portion of the gel was processed by Western blotting (top), whereas the top and bottom portions of the gel were stained with Coomassie Blue (middle and bottom). H, binding of biotin-labeled IQGAP1 proteins expressed using T_NT to GST-IR_cyt and GST alone. All data are representative of at least three independent experiments.

FIGURE 2.

IRS-1 binds to IQGAP1 in cell lysates. A, GST pull-down of endogenous IRS-1 from 3T3-L1 preadipocyte lysates using GST-IQGAP1 or GST alone. B, IP with anti-IQGAP1 polyclonal antibody from CHO/IR/IRS-1 cells transfected with Myc-IQGAP1. NIRS was the negative control. C, IP with anti-IQGAP1 antibodies or NIRS of endogenous IQGAP1 from C2C12 cells. D, schematic representation of IRS-1 constructs. Binding to IQGAP1 is depicted as follows: +, binding detected; −, no binding. E, GST pull-down of endogenous IQGAP1 from HEK-293H cells using GST-tagged IRS-N (N), IRS-C (C) or GST alone. Samples were resolved by SDS-PAGE. The top panel was probed for IQGAP1, and the bottom panel was stained with Coomassie Blue. F, GST pull-down of [³⁵S]methionine-labeled IRS-1 proteins using full-length GST-IQGAP1 (IQF) and GST alone. G, quantification of binding. Data are means ± S.E. (error bars) with binding of IRS-1 peptide 162–267 to IQGAP1 set as 1. The data are representative of three independent experiments.

FIGURE 3.

IRS-1 binds to the C-terminal region of IQGAP1. A, schematic representation of IQGAP1 constructs. Binding to IRS-1 is depicted as follows: +, binding detected; −, no binding. B, GST pull-down from HEK-293H cells transiently transfected with GFP-tagged full-length IQGAP1 (FL), IQGAP1-N (N), or IQGAP1-C (C) using GST-IRS-1(2–300) or GST alone. Blots were probed with anti-GFP antibody. C and D, GST pull-down (PD) of [³⁵S]methionine-labeled IQGAP1 fragments using GST-IRS-1(2–300) or GST alone. Input is shown in the bottom panel. Gels were dried and processed by autoradiography. The data are representative of at least two independent experiments.

FIGURE 4.

IRS-1, IR, and IQGAP1 associate. A, GST-IRS-1 or GST was preincubated with purified His₆-IR_cyt or IQGAP1. After removing unbound proteins, IQGAP1 or IR_cyt was added where indicated. IQGAP1 and IR_cyt binding were assessed by Western blotting. B, quantification of binding. Data are means ± S.E. (error bars) with binding to IQGAP1 or IR_cyt incubated with GST-IRS-1 alone set as 1 (n = 4). C, serum-starved CHO/IR/IRS1 cells, transfected with GFP-IQGAP1, were stimulated with (+) or without (−) 100 nm insulin for 10 min. Samples were immunoprecipitated with anti-IQGAP1 or rabbit IgG antibodies. Western blots were probed for IRS1, phospho-IR (pIR) Tyr-972, and IR, and then blots were stripped and reprobed with anti-IQGAP1 antibody, which detects both GFP-IQGAP1 and endogenous IQGAP1. D, quantification of binding corrected for the amount of IQGAP1 immunoprecipitated. Data are expressed as means ± S.E. with binding minus insulin set as 1 (n = 4).

FIGURE 5.

Effect of IQGAP1 on IR and IRS-1. Serum-starved control (WT) and IQGAP1-null (−/−) MEFs were incubated with 100 nm insulin for the indicated times. A, lysates were immunoprecipitated with anti-IR-β or IgG (control) antibodies, and blots were probed with anti-phosphotyrosine (pTyr) and anti-IR-β antibodies. B, quantification with Image Studio (LI-COR Biosciences). The amount of phosphotyrosine was corrected for IR-β immunoprecipitated from the same sample. Data are expressed as the means ± S.E. (error bars) with control cells at 2 min set as 1 (n = 3). C, IP of endogenous IRS-1. Blots were probed with anti-IRS-1 and anti-phosphotyrosine antibodies. D, the amount of phosphotyrosine signal was quantified as described for B and corrected for the amount of IRS-1 immunoprecipitated from the same sample. Data are means ± S.E. with control cells at 15 min set as 1 (n = 8). E, CHO/IR/IRS-1 cells were transfected with siRNA against IQGAP1 (siIQ) or nonspecific control siRNA sequences (Ctrl). Cells were treated with (+) or without (−) 100 nm insulin for 1 min, and IRS-1 was immunoprecipitated. Western blotting of lysates and immunoprecipitates was performed, and blots were probed for the proteins indicated. F, the amount of co-immunoprecipitated IR was quantified and corrected for IRS-1 immunoprecipitated from the same sample. Data are expressed as means ± S.E. with cells transfected with control siRNA and treated with vehicle set as 1 (n = 5). G, MEFs were incubated with (+) or without (−) 100 nm insulin for 5 min, and IRS-1 was immunoprecipitated. Western blotting of lysates and immunoprecipitates was performed, and blots were probed for the proteins indicated. H, the amount of co-immunoprecipitated p85α was quantified and corrected for the amount of IRS-1 immunoprecipitated from the same sample. Data are expressed as means ± S.E. with wild type MEFs incubated with insulin set as 1 (n = 5). *, p < 0.05; **, p < 0.01.

FIGURE 6.

Effect of IQGAP1 on Akt and ERK. A, MEFs were serum-starved and incubated with insulin for 0, 5, 15, or 30 min. After lysis, samples were processed by Western blotting. A, blots were probed for phospho-Akt Ser-473 (pAkt), Akt, and β-tubulin (loading control). B, after quantification, phospho-Akt was corrected for total Akt present in the same sample. Data are means ± S.E. (error bars) with control cells at 0 min set as 1 (n = 6). C, blots were probed with anti-pERK1/2 Thr-202/Tyr-204, ERK1/2, and β-tubulin antibodies. D, after quantification, phospho-ERK (pERK) was corrected for total ERK in the same sample. Data are means ± S.E. with control cells at 0 min set as 1 (n = 5). E, control and IQGAP1^−/− MEFs were transfected with empty vector (V) or Myc-tagged IQGAP1 full-length (FL) or N-half (N) plasmids. Cells were serum-starved and incubated with (+) or without (−) insulin for 10 min and processed by Western blotting with the antibodies indicated. F, quantification of phospho-Akt, corrected for Akt. Data are means ± S.E. with insulin-treated IQGAP1^−/− cells transfected with empty vector set as 1 (n = 6). G, control and IQGAP1^−/− MEFs were transfected with empty vector (V) or GFP-tagged C-half of IQGAP1 (C) plasmids. Cells were serum-starved and incubated with (+) or without (−) insulin for 10 min and processed by Western blotting with the antibodies indicated. H, quantification of phospho-Akt, corrected for Akt. Data are means ± S.E. with insulin-treated IQGAP1^−/− cells transfected with empty vector set as 1 (n = 4). *, p < 0.05; **, p < 0.01; n.s., not significant.

FIGURE 7.

Insulin-stimulated Akt phosphorylation is impaired in IQGAP1-null mice. Control (WT) and IQGAP1-null (−/−) mice fed a normal diet were fasted for 4 h. 15 min after intraperitoneal saline or insulin injection, mice were euthanized, and tissues were frozen and homogenized. A, IR was immunoprecipitated from lysates, and Western blots were probed with antibodies to phosphotyrosine and IR-β. B, quantification of tyrosine phosphorylation in A corrected for the amount of IR immunoprecipitated from the same sample. Data are means ± S.E. (error bars) with WT mice treated with insulin set to 1 (n = 3). To compare Akt activation, Western blotting for pAkt (Ser-473) and total Akt was performed using liver (n = 6) (C), quadriceps muscle (n = 6) (E), and epididymal fat (n = 4) (G) isolated from the mice. D, F, and H, pAkt levels were quantified and corrected for total Akt in the same sample. Data, normalized to insulin-treated control mice set as 1, are means ± S.E. *, p < 0.05; **, p < 0.01; n.s., not significant.

FIGURE 8.

Glucose regulation is impaired in IQGAP1-null mice. Glucose tolerance tests were performed using control (filled circles) and IQGAP1-null (empty circles) mice fed with either a regular diet (n = 20) (A) or a high fat diet (n = 15) (B). Data are means ± S.E. (error bars). C, insulin tolerance tests were performed in control (filled circles) and IQGAP1-null (empty circles) mice fed a regular diet. Data are means ± S.E. (n = 6). D, serum insulin was measured in mice fed or fasted for 16 h. Data are means ± S.E. (fed, 7 control and 4 IQGAP1^−/−; fasted, 5 control and 7 IQGAP1^−/−). p values are for control versus IQGAP1-null mice. *, p < 0.05; **, p < 0.01.