Serine Phosphorylation by mTORC1 Promotes IRS-1 Degradation through SCFβ-TRCP E3 Ubiquitin Ligase

Abstract

The insulin receptor substrate IRS-1 is a key substrate of insulin and insulin-like growth factor (IGF) receptor tyrosine kinases that mediates their metabolic and growth-promoting actions. Proteasomal degradation of IRS-1 is induced following activation of the downstream kinase mTOR complex 1 (mTORC1) to constitute a negative feedback loop. However, the underlying mechanism remains poorly understood. Here we report that Ser 422 of IRS-1 is phosphorylated by mTORC1 and required for IRS-1 degradation induced by prolonged IGF stimulation. Phosphorylation of Ser 422 then recruits the SCF^β-TRCP E3 ligase complex, which catalyzes IRS-1 ubiquitination. Phosphorylation-dependent IRS-1 degradation contributes to impaired growth and survival responses to IGF in cells lacking TSC2, a negative regulator of mTORC1. Inhibition of IRS-1 degradation promotes sustained Akt activation in IGF-stimulated cells. Our work clarifies the nature of the IRS-1-mTORC1 feedback loop and elucidates its role in temporal regulation of IGF signaling.

Keywords: Biochemical Mechanism; Biochemistry; Molecular Interaction; Molecular Physiology.

Graphical abstract

Figure 1

Activation of mTORC1 Is Necessary and Sufficient to Induce IRS-1 Degradation (A) Immunoblot (IB) analysis of whole-cell lysates derived from L6 cells that were serum starved, treated with the indicated inhibitors, and then collected at the indicated time periods following IGF-I stimulation. Asterisk indicates a non-specific band. (B) L6 cells stably expressing FLAG-IRS-1 (L6: FLAG-IRS-1) were serum starved and then collected at the indicated time periods following IGF-I stimulation. The cell lysates were subjected to immunoprecipitation (IP) with anti-FLAG antibody, and the bound proteins were eluted under denaturing conditions. The denatured fraction was then re-immunoprecipitated (ReIP) with the indicated antibody for ubiquitination assay as described in Transparent Methods. Samples were analyzed by immunoblotting with the indicated antibodies. (C) L6 cells lacking TSC2 (TSC2 KO) and wild-type (WT) cells were serum starved and then collected at the indicated time periods following IGF-I stimulation. The cell lysates were subjected to immunoblotting with the indicated antibodies. See also Figure S1D. (D) Immunoblot analysis of whole-cell lysates derived from L6 WT and TSC2 KO cells that were serum starved with the indicated inhibitors for 12 hr. The data shown are representative of three independent experiments. See also Figure S1.

Figure 2

Cullin1 and β-TRCP1/2 Are Required for IRS-1 Degradation Induced by Prolonged IGF Stimulation (A) HEK293T cells were transfected with Myc-IRS-1 together with the indicated E3 ligase components, and lysates were subjected to anti-FLAG immunoprecipitation (IP) and immunoblotting (IB) with the indicated antibodies. (B) L6 cells were transfected with siRNA targeting the indicated E3 ligase components, serum starved, and then collected at the indicated time periods following IGF-I stimulation. Asterisk indicates a non-specific band. The knockdown efficiency of Cullin7 and Fbxw8 was validated by quantitative RT-PCR shown in Figure S2A. (C) FLAG-IRS-1 was transiently expressed in HEK293 cells, and the protein complex of FLAG-IRS-1 was isolated with anti-FLAG immunoprecipitation and subjected to liquid chromatography-tandem mass spectrometry (LC-MS/MS) analysis. The recovered IRS-1-associated proteins and their reproducibility from four independent experiments are shown. (D) Schematic structure of human β-TRCP1 and β-TRCP2. The substrate recognition site R474 in β-TRCP1 and its homologous site R423 in β-TRCP2 are depicted in red. (E) HEK293T cells were transfected with Myc-IRS-1 together with the indicated F box proteins, and lysates were subjected to anti-FLAG immunoprecipitation and immunoblotting with the indicated antibodies. (F) HEK293T cells were transfected with Myc-IRS-1 together with FLAG-β-TRCP2 WT or R423A mutant, and lysates were subjected to anti-FLAG immunoprecipitation and immunoblotting with the indicated antibodies. (G and H) L6 cells were transfected with two siRNAs targeting β-TRCP1 and β-TRCP2, serum starved, and then collected at the indicated time periods following IGF-I stimulation. The cell lysates were subjected to immunoblotting with the indicated antibodies. The knockdown efficiency of β-TRCP2 was validated by quantitative RT-PCR shown in Figure S2D. Immunoblots of IRS-1 for (G) were quantified, and the graph is shown as mean ± SEM of three independent experiments (*p < 0.05) (H). (I) L6 cells stably expressing FLAG-IRS-1 (L6: FLAG-IRS-1) were transfected with two siRNAs targeting β-TRCP1 and β-TRCP2, serum starved, and then collected at the indicated time periods following IGF-I stimulation. The cell lysates were subjected to re-immunoprecipitation (ReIP) with anti-FLAG antibody under denatured conditions for ubiquitination assay as described in Transparent Methods. Samples were analyzed by immunoblotting with the indicated antibodies. The data shown are representative of at least three independent experiments. See also Figures S2 and S3.

Figure 3

IGF Stimulation Triggers the Interaction of IRS-1 with β-TRCP in an mTORC1-Dependent Manner (A) L6 TSC2 KO cells were transfected with siRNAs targeting β-TRCP1+2 or Cullin1 and serum starved with or without Torin1 for 12 hr. The collected cell lysates were subjected to immunoblotting (IB) with the indicated antibodies. Asterisk indicates a non-specific band. (B) L6 cells stably expressing FLAG-IRS-1 (L6: FLAG-IRS-1) were serum starved, treated with the indicated inhibitors, and then collected at the indicated time periods following IGF-I stimulation. The cell lysates were subjected to immunoprecipitation (IP) and immunoblotting with the indicated antibodies. The data shown are representative of three independent experiments. See also Figure S3.

Figure 4

Ser 422 in IRS-1 Is Required for the Interaction with β-TRCP and IGF-I-Induced Degradation of IRS-1 (A) Schematic illustration of IRS-1 mutants used in (B). The regions colored in orange include the Ala substitution of possible phosphorylated Ser/Thr residues that are authenticated in PhosphoSitePlus. See also Figure S4E. (B) Immunoblot (IB) analysis of whole-cell lysates derived from L6 cells stably expressing the indicated FLAG-IRS-1 mutants that were serum starved and then collected at the indicated time periods following IGF-I stimulation. (C) Schematic representation of the candidate β-TRCP recognition sites that are included in the IRS-1 region corresponding to amino acid residues 301–501. The Ser residues colored in red indicate the authenticated phosphosites that are listed in PhosphoSitePlus. (D) Immunoblot analysis of whole-cell lysates derived from L6 cells stably expressing the indicated FLAG-IRS-1 mutants that were serum starved and then collected at the indicated time periods following IGF-I stimulation. (E) Amino acid sequence alignment of Ser 422 and its surrounding sequence among IRS-1 proteins of vertebrates. Identical residues are highlighted in gray, a critical Ser for IRS-1 degradation is in red, and Gly, a conserved residue only in IRS-1 but not in IRS-2, is in black. (F) L6 cells stably expressing the indicated FLAG-IRS-1 mutants were serum starved and then collected at the indicated time periods following IGF-I stimulation. The cell lysates were subjected to immunoprecipitation (IP) and immunoblotting with the indicated antibodies. (G) Immunoblot analysis of whole-cell lysates derived from L6 cells stably expressing the indicated FLAG-IRS-1 mutants that were serum starved and then collected at the indicated time periods following IGF-I stimulation. (H) L6 cells stably expressing FLAG-IRS-1 WT, S422A, or empty vector were serum starved and then collected at the indicated time periods following IGF-I stimulation. The cell lysates were subjected to re-immunoprecipitation (ReIP) with anti-FLAG antibody under denatured conditions for ubiquitination assay as described in Transparent Methods. Samples were analyzed by immunoblotting with the indicated antibodies. The data shown are representative of three independent experiments. See also Figure S4.

Figure 5

mTORC1 Phosphorylates Ser 422 of IRS-1 in IGF-Stimulated Cells (A) p-IRS-1 (S422/S423) antibody preferentially recognizes S422/S423-phosphorylated IRS-1. L6 cells stably expressing the indicated FLAG-IRS-1 mutants were serum starved and then collected at the indicated time periods following IGF-I stimulation. FLAG-IRS-1 was immunoprecipitated (IP), and one set of the WT immunoprecipitates was treated with calf intestine alkaline phosphatase (CIP). The immunoprecipitated IRS-1 phosphorylation and protein levels were then detected by immunoblotting (IB) with anti-p-IRS-1 (S422/S423) antibody and anti-FLAG antibody, respectively. (B) L6 cells were serum starved, treated with the indicated inhibitors, and then collected at the indicated time periods following IGF-I stimulation. The endogenous IRS-1 proteins were immunoprecipitated, and one set of the immunoprecipitates was treated with CIP. The immunoprecipitated IRS-1 phosphorylation and S6K1/S6 phosphorylation were analyzed by immunoblotting with the indicated antibodies. (C and D) L6 cells stably expressing FLAG-IRS-1 (L6: FLAG-IRS-1) were serum starved and then collected at the indicated time periods following IGF-I stimulation. The cell lysates were subjected to immunoprecipitation and immunoblotting with the indicated antibodies (C). Immunoblots of Tyr, Ser 307, and Ser 422/Ser 423 phosphorylation of IRS-1, and p85 PI3K and β-TRCP1 binding to IRS-1 for (C) were quantified and their time-dependent changes are shown in the graph (D). (E) L6 cells were transfected with siRNA targeting Raptor or Rictor, serum starved, treated with or without Torin1, and then collected at the indicated time periods following IGF-I stimulation. The cell lysates were subjected to immunoprecipitation and immunoblotting with the indicated antibodies. (F) L6 cells stably expressing FLAG-IRS-1 were serum starved and then collected at the indicated time periods following IGF-I stimulation. The cell lysates were subjected to immunoprecipitation and immunoblotting with the indicated antibodies against mTOR complex components. See also Figures S5E and S5F. (G) In vitro phosphorylation of IRS-1 S422/S423 by mTOR. An active mutant FLAG-mTOR^SL1+IT and FLAG-GFP were expressed in HEK293T cells and were immunoprecipitated with anti-FLAG antibody. The substrates GST-IRS-1 fragment (amino acid residues 256–443), GST-4EBP1, and GST were expressed and purified from Escherichia coli. The kinase reaction was carried out as described in Transparent Methods. The phosphorylation levels of IRS-1 and 4EBP1 were analyzed by immunoblotting with anti-p-IRS-1 (S422/S423) antibody and p-4EBP1 (T37/46) antibody, respectively. The total amounts of recombinant proteins used were examined by Coomassie Brilliant Blue (CBB) staining. See also Figure S5G. The data shown are representative of three independent experiments except Figures 5D and 5E being from two independent experiments. See also Figure S5.

Figure 6

Expression of IRS-1 S422A Mutant Restores IGF-I Sensitivity in Cells Lacking TSC2 (A) Schematic illustration of wild-type (WT) IRS-1, IRS-1 S422A, and 4SA mutants. (B) Immunoblot (IB) analysis of whole-cell lysates derived from L6 TSC2 KO cells stably expressing TSC2, FLAG-IRS-1 WT, S422A, 4SA, or empty vector that were serum starved and then collected at 1 hr following Torin1 stimulation. (C and D) Immunoblot analysis of whole-cell lysates derived from L6 TSC2 KO cells stably expressing TSC2, FLAG-IRS-1 WT, S422A, 4SA, or empty vector that were serum starved and then collected at 5 min following IGF-I stimulation at the indicated concentration (C). Immunoblots of Akt and FoxO1/3a phosphorylation for (C) were quantified, and the graph is shown as mean ± SEM of four independent experiments (*p < 0.05, **p < 0.01, ***p < 0.001) (D). (E) Effects of expression of TSC2 or IRS-1 mutants on IGF-I-induced DNA synthesis in L6 TSC2 KO cells. L6 TSC2 KO cells stably expressing TSC2, FLAG-IRS-1 WT, S422A, 4SA, or empty vector were serum starved and then treated with IGF-I at the indicated concentration for 12 hr. [³H]Thymidine incorporation into DNA during 8–12 hr of IGF-I stimulation was measured. The graph is shown as mean ± SEM (n = 4; **p < 0.01). (F) L6 TSC2 KO cells stably expressing TSC2, FLAG-IRS-1 WT, S422A, 4SA, or empty vector were placed in serum-free medium with or without IGF-I at the indicated concentration for 6 hr. The collected cell lysates were subjected to immunoblotting with the indicated antibodies. See also Figure S6A. The data shown are representative of at least three independent experiments. See also Figure S6.

Figure 7

Inhibition of Ser 422 Phosphorylation-Dependent IRS-1 Degradation Promotes Sustained Akt Activation in IGF-Stimulated Cells (A and B) L6 cells were transfected with two siRNAs targeting β-TRCP1 and β-TRCP2, serum starved, and then collected at the indicated time periods following IGF-I stimulation. The cell lysates were subjected to immunoblotting (IB) with the indicated antibodies (A). Immunoblots of Akt phosphorylation for (A) were quantified, and the graph is shown as mean ± SEM of four independent experiments (*p < 0.05) (B). (C and D) Immunoblot analysis of whole-cell lysates derived from L6 cells stably expressing FLAG-IRS-1 WT, S422A, or empty vector that were serum starved and then collected at the indicated time periods following IGF-I stimulation (C). Immunoblots of Akt phosphorylation for (C) were quantified, and the graph is shown as mean ± SEM of three independent experiments (*p < 0.05, **p < 0.01) (D). (E) Model for temporal regulation of mTORC1-mediated IRS-1 degradation and the PI3K/Akt signaling following IGF stimulation. The figure depicts the interplay of mTORC1-induced Ser 422 phosphorylation of IRS-1, SCF^β-TRCP-mediated ubiquitination of IRS-1, and downregulation of the PI3K/Akt signaling as a function of time following the stimulation. See text for details.