IQGAP1 binds AMPK and is required for maximum AMPK activation

Abstract

AMP-activated protein kinase (AMPK) is a fundamental component of a protein kinase cascade that is an energy sensor. AMPK maintains energy homeostasis in the cell by promoting catabolic and inhibiting anabolic pathways. Activation of AMPK requires phosphorylation by the liver kinase B1 or by the Ca²⁺/calmodulin-dependent protein kinase 2 (CaMKK2). The scaffold protein IQGAP1 regulates intracellular signaling pathways, such as the mitogen-activated protein kinase and AKT signaling cascades. Recent work implicates the participation of IQGAP1 in metabolic function, but the molecular mechanisms underlying these effects are poorly understood. Here, using several approaches including binding analysis with fusion proteins, siRNA-mediated gene silencing, RT-PCR, and knockout mice, we investigated whether IQGAP1 modulates AMPK signaling. In vitro analysis reveals that IQGAP1 binds directly to the α1 subunit of AMPK. In addition, we observed a direct interaction between IQGAP1 and CaMKK2, which is mediated by the IQ domain of IQGAP1. Both CaMKK2 and AMPK associate with IQGAP1 in cells. The ability of metformin and increased intracellular free Ca²⁺ concentrations to activate AMPK is reduced in cells lacking IQGAP1. Importantly, Ca²⁺-stimulated AMPK phosphorylation was rescued by re-expression of IQGAP1 in IQGAP1-null cell lines. Comparison of the fasting response in wild-type and IQGAP1-null mice revealed that transcriptional regulation of the gluconeogenesis genes PCK1 and G6PC and the fatty acid synthesis genes FASN and ACC1 is impaired in IQGAP1-null mice. Our data disclose a previously unidentified functional interaction between IQGAP1 and AMPK and suggest that IQGAP1 modulates AMPK signaling.

Keywords: AMP-activated kinase (AMPK); IQGAP1; calcium; calmodulin (CaM); homeostasis; metabolic regulation; metformin; protein–protein interaction; scaffold protein; signaling.

Figure 1

AMPKα1 and CaMKK2 co-immunoprecipitate with IQGAP1.A, HEK-293 cells were transfected with GFP-AMPKα1. 48 h after transfection, cells were lysed and endogenous IQGAP1 was immunoprecipitated (IP) with anti-IQGAP1 antibody (IQ1). Nonimmune rabbit serum (NIRS) was used as control. Samples were analyzed by SDS-PAGE and Western blotting using anti-GFP and anti-IQGAP1 antibodies. 1% of the lysate used for IP was resolved in parallel (Input). B, HEK-293 cells were transfected with GFP or GFP-AMPKα1. 72 h after transfection, cells were lysed and GFP-tagged proteins were immunoprecipitated with GFP-Trap Agarose. Samples were analyzed by SDS-PAGE and Western blotting using anti-GFP and anti-IQGAP1 antibodies. 1% of the lysate used for IP was resolved in parallel (Input). C, HEK-293 cells were transfected with GFP-CaMKK2. Immunoprecipitation of endogenous IQGAP1 and Western blotting were performed as described for panel (A). D, HEK-293 cells were transfected with GFP or GFP-CaMKK2. Immunoprecipitation of GFP proteins and Western blotting were performed as in panel (B). The positions of migration of molecular weight markers are indicated on the left. All data are representative of at least three independent experiments.

Figure 2

AMPKα1 and CaMKK2 bind directly to IQGAP1 via its IQ domain.A, schematic representation of IQGAP1 constructs. The identified protein interaction motifs (CHD, calponin homology domain; WW, two tryptophan-containing domain; IQ, IQ domain; GRD, GAP-related domain; RGCT, RasGAP C_terminus) and amino acid residues of each construct are indicated. These constructs correspond to full-length IQGAP1 (F, amino acids 2–1657), the N-half (N, 2–863), the IQ domain (IQ, 717–916), and the C-half (C, 864–1657) of IQGAP1. IQGAP1Δ746-860 has amino acids 746 to 860 deleted. B, a Coomassie-stained gel of the GST proteins (GST-AMPKα1, GST-CaMKK2, or GST alone) used for binding assays. Data are representative of two independent experiments. C, fragments of IQGAP1 (F, N, IQ or C) were expressed and labeled with [³⁵S]methionine using the T_NT system. The IQGAP1 (IQ1) fragments were incubated with purified recombinant GST-AMPKα1, GST-CaMKK2, or GST alone. Complexes were pulled down (PD) with glutathione-Sepharose beads and analyzed by SDS-PAGE and autoradiography. 1% of the T_NT products were analyzed in parallel (Input, right panel). The positions of migration of molecular weight markers are indicated on the left. Data are representative of at least two independent experiments. D, T_NT products of full-length IQGAP1(F) or IQGAP1Δ746-860 (Δ746-860) were labeled with biotinylated-lysine and were incubated with purified GST alone, GST-AMPKα1, or GST-CaMKK2. Complexes were pulled down and analyzed by SDS-PAGE. The gel was cut at ∼120 kDa. The upper portion of the gel (containing IQGAP1) was processed by Western blotting using IRDye-conjugated streptavidin (Strept.). The lower portion of the gel was stained with Coomassie blue. 1% of the T_NT products were analyzed in parallel (Input). Data are representative of five independent experiments.

Figure 3

IQGAP1 influences AMPK activation by Ca²⁺.A, wild-type (WT) and IQGAP1-null (−/−) MEF cells were treated with DMSO (−) or 10 μM A23187 (+) for 15 min. Cells were harvested and equal amounts of protein lysate were analyzed by Western blotting. Two separate blots (i and ii) were performed from the same lysates to probe for multiple proteins. Blot (i) was probed using anti-pAMPK and antitubulin (loading control) antibodies. Blot (ii) was probed with anti-IQGAP1 (IQ1), anti-AMPK, and antitubulin antibodies. The positions of migration of molecular weight markers are indicated on the left. B, the pAMPK and tubulin bands from the same blots were quantified with Image Studio 2.0 and pAMPK was corrected for the amount of tubulin in the corresponding sample. Data are expressed as mean ± SD (n = 8) with vehicle-treated wild-type cells set as 1. C, IQGAP1-null MEF cells were transfected with empty vector or Myc-IQGAP1 (Myc-IQ1). 48 h after transfection, cells were treated with DMSO (−) or 10 μM A23187 (+) for 15 min. Cells were harvested and equal amounts of protein lysate were analyzed by Western blotting using anti-pAMPK, anti-IQGAP1, and antitubulin antibodies. D, the pAMPK/tubulin ratios were calculated as described in panel B. Data are expressed as mean ± SD (n = 8) with vehicle-treated cells transfected with empty vector set as 1. All statistical analyses were performed using one-way ANOVA, with Tukey’s post-hoc test. ∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.001.

Figure 4

Knockdown of IQGAP1 alters AMPK activation in human cell lines.A, HepG2 cells were transfected with control or IQGAP1 (IQ1)-targeted siRNA. 72 h after transfection, DMSO (−) or 10 μM A23187 (+) was added for 15 min. Cells were harvested and equal amounts of protein lysate were analyzed by Western blotting. Two separate blots (i and ii) were performed to probe for multiple proteins. Blot (i) was probed with anti-IQGAP1, anti-pAMPK, and antitubulin (loading control) antibodies. Blot (ii) was probed with anti-AMPK and antitubulin antibodies. B, the pAMPK and tubulin bands from the same blots were quantified with Image Studio 2.0 and corrected for the amount of tubulin in the corresponding sample. Data are expressed as mean ± SD (n = 12) with vehicle-treated cells transfected with control siRNA set as 1. C, HeLa cells were transfected with control or IQGAP1-targeted siRNA. 72 h after transfection, cells were treated with DMSO (−) or 10 μM A23187 (+) for 15 min. Samples were processed as described for panel A. D, the pAMPK/tubulin ratios were calculated as described in panel B. Data are expressed as mean ± SD (n = 11) with vehicle-treated cells transfected with control siRNA set as 1. E, HeLa cells were transfected with control or CaMKK2-targeted siRNA. 72 h after transfection, DMSO (−) or 10 μM A23187 (+) was added for 15 min. Samples were processed as described for panel A. F, pAMPK/tubulin ratios were calculated as described for panel B. Data are expressed as mean ± SD (n = 5) with vehicle-treated cells transfected with control siRNA set as 1. G, equal amounts of protein lysate from the HeLa cells transfected as described in panel C were analyzed by Western blotting. Blots were probed with anti-IQGAP1 (IQ1), anti-pCaMK1, and antitubulin antibodies. H, the pCaMK1 and tubulin bands from the same blots were quantified and corrected for the total amount of tubulin in the sample. Data are expressed as mean ± SD (n = 4) with vehicle-treated cells transfected with control siRNA set as 1. All statistical analyses were performed using one-way ANOVA, with Tukey’s post-hoc test. ∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.001; ns, not significant.

Figure 5

IQGAP1Δ746-860 fails to rescue AMPK activation by Ca²⁺in cells with reduced endogenous IQGAP1.A, to knockdown endogenous IQGAP1, HeLa cells were transfected with IQGAP1-targeted siRNA (siIQGAP1-8). Scrambled siRNA was used as the control. 24 h later the cells were transfected with the following plasmids: empty vector or Myc-tagged constructs of wild-type IQGAP1 (WT) or IQGAP1Δ746-860 (Δ746-860). 48 h after transfection, cells were treated with DMSO (−) or 10 μM A23187 (+) for 15 min. Cells were harvested and equal amounts of protein lysate were analyzed by Western blotting. Two separate blots (i and ii) were performed to probe for multiple proteins. Blot (i) was probed with anti-IQGAP1, anti-pAMPK, and anti-GAPDH (loading control) antibodies. Blot (ii) was probed with anti-Myc, anti-AMPK, and anti-GAPDH antibodies. B, the pAMPK and GAPDH bands from the same blots were quantified by densitometry using Image Studio 2.0 and pAMPK was corrected for the amount of GAPDH in the corresponding sample. Data are expressed as means (n = 2) with vehicle-treated cells in each condition set as 1.

Figure 6

IQGAP1 influences AMPK activation by metformin.A, wild-type (WT) and IQGAP1-null (−/−) MEF cells were treated without (−) or with (+) 5 mM metformin for 4 h. Cells were harvested and equal amounts of protein lysate were analyzed by Western blotting. Two separate blots (i and ii) were performed from the same lysates to probe for multiple proteins. Blot (i) was probed using anti-pAMPK and antitubulin (loading control) antibodies. Blot (ii) was probed with anti-IQGAP1 (IQ1), anti-AMPK, and antitubulin antibodies. The positions of migration of molecular weight markers are indicated on the left. B, the pAMPK and tubulin bands from the same blots were quantified with Image Studio 2.0 and pAMPK was corrected for the amount of tubulin in the corresponding sample. Data are expressed as mean ± SD (n = 7) with untreated wild-type cells set as 1. Statistical analysis was performed using one-way ANOVA, with Tukey’s post-hoc test. ∗p < 0.05; ∗∗p < 0.01.

Figure 7

Expression of gluconeogenesis and fatty acid synthesis genes in IQGAP1-null mice. Wild-type (IQ1^+/+) or IQGAP1-null (IQ1^−/−) mice were fed or fasted for 16 h. Then, mice were euthanized with carbon dioxide and tissues were extracted. Quantitative RT-PCR analysis was performed to measure mRNA expression of PCK1 and G6PC in the liver (panels A and B) and FASN and ACC1 in the liver (panelsC and D) and epididymal fat pads (panels E and F). Expression of genes was normalized to the housekeeping gene 18S RNA. Data are expressed as mean ± SD (n = 2–4 mice per group, each assay was performed in triplicate) with fed wild-type mice set as 1. Statistical analysis was performed using unpaired t-test with Welch’s correction. ∗p < 0.05; ∗∗p < 0.01.