Nutrients suppress phosphatidylinositol 3-kinase/Akt signaling via raptor-dependent mTOR-mediated insulin receptor substrate 1 phosphorylation

Abstract

Nutritional excess and/or obesity represent well-known predisposition factors for the development of non-insulin-dependent diabetes mellitus (NIDDM). However, molecular links between obesity and NIDDM are only beginning to emerge. Here, we demonstrate that nutrients suppress phosphatidylinositol 3 (PI3)-kinase/Akt signaling via Raptor-dependent mTOR (mammalian target of rapamycin)-mediated phosphorylation of insulin receptor substrate 1 (IRS-1). Raptor directly binds to and serves as a scaffold for mTOR-mediated phosphorylation of IRS-1 on Ser636/639. These serines lie close to the Y(632)MPM motif that is implicated in the binding of p85alpha/p110alpha PI3-kinase to IRS-1 upon insulin stimulation. Phosphomimicking mutations of these serines block insulin-stimulated activation of IRS-1-associated PI3-kinase. Knockdown of Raptor as well as activators of the LKB1/AMPK pathway, such as the widely used antidiabetic compound metformin, suppress IRS-1 Ser636/639 phosphorylation and reverse mTOR-mediated inhibition on PI3-kinase/Akt signaling. Thus, diabetes-related hyperglycemia hyperactivates the mTOR pathway and may lead to insulin resistance due to suppression of IRS-1-dependent PI3-kinase/Akt signaling.

FIG. 1.

Rapamycin inhibits phosphorylation of IRS-1 on Ser636/639 and enhances PI3-kinase/Akt signaling. (A) L6 myotubes were serum starved for 6 h, pretreated with rapamycin (Rap) for 30 min, and then treated with insulin for another 30 min as described in Materials and Methods, and endogenous IRS-1 was immunoprecipitated (IP). The presence of PI3-kinase in the immunoprecipitated material was determined by Western blotting (top right panel) and its activity was measured by TLC (left panel). The bottom right panel shows normalized IRS-1-associated PI3-kinase activity (mean values ± SEM) of the results from three independent experiments. (B and C) Total cell lysates treated as described for panel A were analyzed by Western blotting. (D) Normalized mean values ± SEM of results from three independent experiments shown in panel C; asterisks indicate that P values of <0.01. +, present; −, absent.

FIG. 2.

Phosphorylation of IRS-1 on Ser 636/639 but not on Ser616 uncouples insulin signaling. (A) 293HEK cells transiently transfected with specified IRS-1 mutants (4 μg of each cDNA per 10-cm plate) were serum starved for 6 h followed by 5 min of insulin stimulation and immunoprecipitation (IP) of IRS-1. The presence of PI3-kinase in the immunoprecipitated material was determined by Western blotting, and its activity was measured by TLC. The top panel shows normalized PI3-kinase activity (mean values ± SEM) of three independent experiments. The bottom panel shows Western blot analysis of total cell lysates. (B) Same as described for panel A with different IRS-1 mutants. Note that in panels A and B, phosphospecific antibodies do not recognize mutated serine residues in IRS-1, verifying that mutations have been introduced correctly. (C) L6 myotubes were homogenized and fractionated as described in Materials and Methods into cytosol and high-speed pellet. Each fraction was analyzed by Western blotting. The same amount of protein (30 μg) was loaded on each lane; asterisks indicate P values of < 0.01. +, present; −, absent.

FIG. 3.

Activation of mTOR/S6K1 by Rheb suppresses insulin signaling. (A) 293HEK cells were transfected with different amounts of Rheb cDNA for 36 h, serum starved for 6 h, and then stimulated with insulin for 30 min. Total cell lysates were analyzed by Western blotting (WB). (B) 293HEK cells were transfected as indicated on the figure. Cells were serum starved for 6 h, and total cell lysates were analyzed by Western blotting. (C) C2C12 fibroblasts stably transfected with Rheb or empty vector were grown in DMEM supplemented with 20% FBS and treated with cycloheximide (40 μg/ml) in the presence or absence of rapamycin (50 nM) for different periods of time. Levels of endogenous IRS-1 in cell lysates were determined by Western blotting. The graph shows the IRS-1 protein level (as percentage of control) normalized to actin of three independent experiments (mean values ± SEM). +, present; −, absent.

FIG. 4.

Nutrients stimulate phosphorylation of IRS-1 on Ser636/639 and inhibit insulin signaling. (A) L6 myotubes were starved in glucose-free KRP for 6 h, pretreated with glucose for 30 min, and stimulated with insulin for another 30 min as indicated. (B) L6 myotubes were starved in glucose-free KRP for 6 h and stimulated with the indicated concentrations of glucose for 30 min. (C) 293HEK cells were transfected with the indicated constructs and were starved in glucose-free KRP for 6 h following stimulation with glucose for 30 min. Total cell lysates were analyzed by Western blotting. (D) 3T3-L1 adipocytes and L6 myotubes were starved in glucose- and amino acid-free KRP for 6 h, treated with rapamycin as indicated for 30 min, and stimulated with leucine and insulin for another 30 min. Total cell lysates were analyzed by Western blotting. (E) 293HEK cells were transfected with HA-Rheb and were starved in glucose-free KRP or in serum-free DMEM following insulin stimulation for 30 min as indicated on the figure. HA-Rheb-bound GDP was measured as described in Materials and Methods. Results of two independent experiments are presented. The first four lanes of the TLC panel show the calibration curve. (F) Normalized results (mean values ± SEM) of three independent experiments shown in panel E. The asterisk indicates a P value of < 0.01. +, present; −, absent.

FIG. 5.

LKB1, metformin, and 2-deoxyglucose suppress phosphorylation of IRS-1 on Ser636/639 and enhance Akt phosphorylation. (A to C) 293HEK cells were transfected with indicated constructs. Cells were either serum starved for 6 h (A and B) or grown in 20% FBS (C). Total cell lysates were analyzed by Western blotting. (D) HepG2 cells were serum starved for 6 h, pretreated for 60 min with indicated doses of metformin, and stimulated with insulin for another 30 min. Total cell lysates were analyzed by Western blotting. (E) Insulin-induced Akt Thr308 phosphorylation in HepG2 cells in the absence and in the presence of 5 mM metformin was normalized by actin. The panel shows mean values ± SEM of the results from three independent experiments. The asterisk indicates a P value of <0.03. A.U., arbitrary units. (F) 3T3-L1 adipocytes were pretreated with increasing concentrations of 2-deoxyglucose (2-DG) for 30 min and then stimulated with insulin for another 30 min. Total cell lysates were analyzed by Western blotting. +, present; −, absent.

FIG. 6.

TNF-α does not stimulate the mTOR/S6K1 pathway and IRS-1 Ser636/639 phosphorylation. L6 myotubes and 3T3-L1 adipocytes were serum starved for 6 h and stimulated with the indicated doses of either TNF-α or insulin for another 30 min. Total cell lysates were analyzed by Western blotting. +, present; −, absent.

FIG. 7.

Raptor is essential for Rheb-, insulin-, and nutrient-dependent phosphorylation of IRS-1 on Ser636/639 and attenuation of PI3-kinase/Akt signaling. (A) 293HEK cells were transfected with Rheb, siRNA against Raptor, or control siRNA as indicated on the figure. Cells that were serum starved for 6 h were stimulated or not with insulin, and total cell lysates were analyzed by Western blotting (WB). (B) 293HEK cells were transfected with Myc-IRS-1 and the indicated siRNAs. Cells were incubated in serum-free media for 6 h, and total cell lysates were analyzed by Western blotting. (C) 293HEK cells were transfected with indicated amounts of siRNA against Raptor or with control siRNA. Cells were grown in 20% FBS, and total cell lysates were analyzed by Western blotting. (D) The domain structure of wild-type Raptor and mutants (not to scale). (E) 293HEK cells were transfected with the indicated constructs (2 μg each per 10-cm plate). HA-tagged Raptor mutants were immunoprecipitated (IP) with anti-HA antibody; immunoprecipitates (left) and total cell lysates (right) were analyzed by Western blotting. (F and G) 293HEK cells were transfected with indicated constructs. In panel F, cells were serum starved for 6 h, and in panel G, cells were grown in 20% FBS. +, present; −, absent; IgG, immunoglobulin G.

FIG. 8.

Raptor/mTOR/S6K1 complex phosphorylates in vitro IRS-1 on Ser636/639. (A) 293HEK cells were transfected with indicated cDNAs. Cells were harvested in 0.3% CHAPS, and Raptor/mTOR complexes were immunoprecipitated (IP) with anti-Myc antibody. In vitro phosphorylation with recombinant IRS-1 as a substrate was performed as described in Materials and Methods. Phosphorylation products were analyzed by Western blotting with phosphospecific antibody (top left panel) or by autoradiography (top right panel). Asterisks indicate degradation products of recombinant IRS-1. (B) 293HEK cells were transfected as indicated and incubated in serum-free medium (SFM) or in 20% FBS as shown on the figure. Total cell lysates were analyzed by Western blotting (WB). (C) NIH 3T3 cells were transfected as indicated, incubated in serum-free media for 6 h, pretreated with rapamycin (Rap) for 30 min, and stimulated with insulin for another 30 min. Total cell lysates were analyzed by Western blotting. +, present; −, absent; IgG, immunoglobulin G.

FIG. 9.

Diabetic milieu (hyperglycemia and hyperinsulinemia) down-regulates PI3-kinase/Akt signaling via Raptor-dependent mTOR/S6K1 mediated phosphorylation of IRS-1 on Ser636/639 (not to scale).