Research ArticleCell Biology

AMPK directly activates mTORC2 to promote cell survival during acute energetic stress

See allHide authors and affiliations

Science Signaling  11 Jun 2019:
Vol. 12, Issue 585, eaav3249
DOI: 10.1126/scisignal.aav3249
  • Fig. 1 AMPK associates with and phosphorylates mTOR within mTORC2.

    (A) Wild-type (WT) and AMPKα1/α2 DKO MEFs were serum-starved (20 hours), pretreated without or with torin1 (T; 100 nM; 30 min), and treated without (−) or with (+) AICAR (2.5 mM; 2 hours). Whole-cell lysates were immunoblotted (IB) as indicated. Graph represents quantification of the mean ratio ± SD of AktS473 phosphorylation over total Akt levels. Each bar represents the mean of five independent experiments performed in duplicate ± SD; thus, n = 10. ***P < 0.001, analysis of variance (ANOVA), WT compared to DKO. (B) MEFs were serum-starved (20 hours), pretreated with torin1 (100 nM) or BYL719 (10 μM; 30 min), and treated without (−) or with (+) AICAR (2.5 mM; 2 hours). Whole-cell lysates were immunoblotted as indicated. Blots are representative of three independent experiments. (C) rictor−/− MEFs stably expressing vector control (V) or HA-rictor were serum-starved (20 hours), pretreated without or with torin1 (100 nM; 30 min), and treated without (−) or with (+) AICAR (2.5 mM; 1 hour). Whole-cell lysates were immunoblotted as indicated. Blots are representative of three independent experiments. (D) Myc-mTOR WT and Myc-mTOR S1261A were immunoprecipitated (IP) with anti-Myc antibodies from transfected human embryonic kidney (HEK) 293 cells and incubated without (−) or with (+) recombinant active AMPKα1/β1/γ1 (α1) or AMPKα2/β1/γ1 (α2) (100 ng; 30 min at 30°C). IVK reactions and input were analyzed as indicated. Blots are representative of three independent experiments. (E) Rictor was immunoprecipitated and incubated with recombinant active AMPKα1/β1/γ1 or AMPKα2/β1/γ1 in vitro as described in (D). Blots are representative of five independent experiments. (F) mTOR Ser1261 peptide sequence compared to the AMPK consensus phosphorylation motif and the raptor Ser792 peptide sequence. (G) HEK293 cells were cotransfected with Flag-rictor, HA-AMPKα1, or HA-AMPKα2. Anti-Flag immunoprecipitates (IP) and whole-cell lysates (WCL) were immunoblotted as indicated. Blots are representative of six independent experiments. (H) WT and AMPKα1/α2 DKO MEFs were serum-starved (20 hours), pretreated with torin1 (100 nM; 30 min), and treated without (−) or with (+) AICAR as in (A). Rictor or raptor was immunoprecipitated, and immunoprecipitates and whole-cell lysates were immunoblotted as indicated. Blots are representative of four independent experiments.

  • Fig. 2 AMPK promotes mTORC2 signaling in response to energetic stress induced by diverse agents in cultured cells.

    (A) WT and AMPKα1/α2 DKO MEFs were serum-starved (20 hours) and refed with Dulbecco’s modified Eagle’s medium (DMEM) without (−) and with (+) glucose withdrawal (Glc W/D) for 8 hours to induce glucose deprivation without or with torin1 (100 nM). Graph represents quantification of the mean ratio ± SD of AktS473 phosphorylation over total Akt levels. n = 3 samples from three independent experiments. *P < 0.05 by unpaired t test. (B) WT and AMPKα1/α2 DKO MEFs were serum-starved (20 hours), pretreated without or with torin1 (100 nM; 30 min), and treated without (−) or with (+) phenformin (2 mM; 90 min). Graph represents quantification of the mean ratio ± SD of AktS473 phosphorylation over total Akt levels. n = 6 samples from three independent experiments. ***P < 0.001 by unpaired t test. (C) WT and AMPKα1/α2 DKO MEFs were treated as in (B) except without (−) or with (+) rotenone (2.5 μg/ml; 60 min). Graph represents quantification of the mean ratio ± SD of AktS473 phosphorylation over total Akt levels. n = 4 samples from four independent experiments. **P < 0.01 by unpaired t test. (D) MEFs were serum-starved (20 hours), pretreated without or with torin1 as above, and treated without (−) or with (+) A769662 (100 μM) for a time course (0 to 30 min). Blots are representative of three independent experiments. (E) HEK293 cells were transiently transfected with scrambled (Scr), AMPKβ1 (siAMPKβ1), or pan-AMPKα1/α2 (siAMPKα) siRNAs (96 hours), serum-starved (20 hours), pretreated without or with torin1 (100 nM; 30 min), and then stimulated without (−) or with (+) A769662 (100 μM; 10 min). Blots are representative of three independent experiments.

  • Fig. 3 AMPK promotes mTORC2 signaling independently of mTORC1-mediated negative feedback.

    (A) MEFs were serum-starved (20 hours), pretreated without or with torin1 (100 nM; 30 min), and treated without (−) or with (+) AICAR (2.5 mM), rapamycin (20 ng/ml), or insulin (100 nM) for the indicated times (0 to 240 min). Whole-cell lysates were immunoblotted as indicated. Blots are representative of three independent experiments. (B) Left: MEFs were serum-starved (20 hours), pretreated without (−) or with (+) rapamycin (20 ng/ml) or torin1 (100 nM; 30 min), and treated without (−) or with (+) AICAR (2.5 mM; 2 hours), phenformin (2 mM; 1 hour), glucose withdrawal (8 hours), or rotenone (2.5 μg/ml; 1 hour). Right: HEK293 cells were serum-starved and treated as in (A) except without (−) or with (+) A769662 (100 μM; 2 hours). Blots are representative of three independent experiments. (C) HEK293 cells were transfected with scrambled (Scr) or raptor (siRaptor) siRNAs (96 hours), serum-starved (20 hours), pretreated with torin1 (100 nM; 30 min), and treated without (−) or with (+) A769662 (100 μM; 10 min). Blots are representative of three independent experiments. (D) MEFs were serum-starved and torin1-treated as in (A) except without (−) or with (+) AICAR (2.5 mM; 2 hours), insulin (100 nM; 2 hours), or both. Blots are representative of three independent experiments.

  • Fig. 4 AMPK promotes mTORC2 signaling in cultured primary hepatocytes and in liver in vivo.

    (A) Primary hepatocytes expressing (AAV-GFP) or lacking AMPKα1 and α2 (AAV-Cre) were isolated from male AMPKα1/α2 floxed mice injected with AAVs for 14 days. The cells were then placed in vitro, serum-starved (20 hours), and treated without (−) or with (+) metformin (2 mM; 2 hours). Whole-cell lysates were immunoblotted as indicated. Blots are representative of two independent experiments. (B) Primary hepatocytes expressing (AAV-GFP) or lacking AMPKα1 and α2 (AAV-Cre) were serum-starved and treated without (−) or with (+) AICAR (2.5 mM; 2 hours). Blots are representative of two independent experiments. (C) Primary hepatocytes expressing (AAV-GFP) or lacking (AAV-Cre) AMPKα1 and α2 were serum-starved and treated with metformin (2 mM), insulin (100 nM), or both (2 hours). SE, short exposure; LE, long exposure. Blots are representative of two independent experiments. (D) Male AMPKα1/α2 floxed mice administered AAV-GFP viruses (control) or AAV-Cre viruses to excise AMPKα1/α2 (14 days) were fasted overnight (O/N) and injected intraperitoneally with saline or metformin (250 mg/kg; 1 hour). Liver lysates were immunoblotted as indicated. Blot is representative of one independent experiment using 3 to 5 individual mice per treatment group (14 mice total).

  • Fig. 5 AMPK directly increases mTORC2 catalytic activity.

    (A) AMPKα1/α2 DKO MEFs were serum-starved (20 hours), pretreated with torin1 (100 nM; 30 min), and treated without (−) or with (+) AICAR (2.5 mM; 2 hours) or insulin (INS; 100 nM; 30 min). Immunoprecipitations with anti-rictor antibodies or beads only (b/o) were incubated with recombinant His-Akt1 substrate (100 ng; 30 min) at 30°C. IVK reactions, immunoprecipitates, and whole-cell lysates were immunoblotted as indicated. Graph represents the mean ratio ± SD of AktS473 phosphorylation over total Akt levels. n = 3 samples from three independent experiments. *P < 0.05 by unpaired t test. (B) AMPKα1/α2 DKO MEFs were serum-starved (20 hours) and stimulated without (−) (lanes 1 to 7) or with (+) (lanes 8 and 9) insulin (100 nM; 30 min). Immunoprecipitations with anti-rictor antibodies (+) or beads only (−) were preincubated without or with compound C (CC; 0.5 mM) or torin1 (5 μM; 30 min) on ice, incubated without (−) or with (+) recombinant GST-AMPKα1β1γ1 (100 ng; 30 min) at 30°C (stage 1), washed twice, and incubated without (−) or with (+) His-Akt1 (100 ng; 30 min) at 30°C (stage 2). IVK reactions were immunoblotted as indicated. Graph represents the mean ratio ± SD of AktS473 phosphorylation over total Akt levels. n = 4 samples from four independent experiments. **P < 0.01 by unpaired t test. (C) Rictor (+) or beads-only (−) immunoprecipitates were incubated without (−) or with (+) recombinant active GST-AMPKα1/β1/γ1 (100 ng) and without (−) or with (+) compound C (25 μM; 30 min) at 30°C. Blots are representative of four independent experiments.

  • Fig. 6 AMPK-mediated mTORC2 signaling does not require mTOR Ser1261phosphorylation.

    (A) Two littermate-matched pairs of WT (mTOR+/+) and mTOR S1261A (mTORA/A) primary MEFs were serum-starved (20 hours) and treated without (−) or with (+) AICAR (2.5 mM; 2 hours). Whole-cell lysates were immunoblotted as indicated. Blots are representative of four independent experiments. (B) Primary MEFs from mTOR+/+ or mTORA/A littermates were serum-starved (20 hours) and refed with DMEM containing (−) or lacking (+) glucose (8 hours) to induce glucose deprivation (Glc W/D; glucose withdrawal) without or with torin1 (100 nM). Blots are representative of three independent experiments. (C) Primary hepatocytes from mTOR+/+ or mTORA/A littermates were serum-starved and treated without (−) or with (+) metformin (2 mM; 2 hours). Blots are representative of four independent experiments.

  • Fig. 7 AMPK, mTORC2, and Akt promote cell survival during acute energetic stress.

    (A and B) WT and AMPKα1/α2 DKO MEFs were serum-starved (20 hours), pretreated without (−) or with (+) torin1 (100 nM; 30 min), and treated without (−) or with (+) AICAR (2.5 mM; 5 hours) (A) or subjected to glucose withdrawal (Glc W/D) (+) or not (−). (B) Whole-cell lysates were immunoblotted as indicated. Blots are representative of three independent experiments. (C) AMPKα1/α2 DKO MEFs stably expressing vector control (V) or HA-AMPKα1 were serum-starved and treated with AICAR as in (A). Blots are representative of three independent experiments. (D and E) rictor−/− MEFs stably expressing vector control or HA-rictor were serum-starved and treated with AICAR as in (A) or subjected to glucose withdrawal (24 hours). Blots are representative of three independent experiments. (F) WT and AMPKα1/α2 DKO MEFs were treated with AICAR (5 hours) as in (A). ReadyProbes cell viability reagents were added to the culture medium, and live cells were imaged using blue [4′,6-diamidino-2-phenylindole (DAPI)] and green [fluorescein isothiocyanate (FITC)/GFP] filters on an inverted epifluorescence microscope. Cell death (%) was calculated using the ratio of green, nonviable cells over total blue cells (includes green cells). Graph represents the mean ± SD of n = 4 experiments, in which n = 300 cells were counted per treatment in each experiment. Statistical significance was tested by ANOVA followed by pairwise Tukey’s post hoc tests (**P < 0.001). (G) rictor−/− MEFs rescued with HA-rictor or vector control (Vector) were treated with AICAR (5 hours) as in (A). ReadyProbes cell viability reagents were added to the culture medium, and cells were imaged as in (F). Cell death (%) was calculated as in (F). Graph represents the mean ± SD of n = 3 experiments, in which n = 250 to 300 cells were counted per treatment in each experiment. Statistical significance was tested by ANOVA (**P < 0.001) followed by pairwise Tukey’s post hoc tests. (H) AMPK promotes cell survival during acute energetic stress by both mTORC2-dependent and mTORC2-independent pathways. (I) AMPK phosphorylates mTOR and/or partner proteins directly to increase mTORC2 catalytic activity, which functions to promote cell survival during acute energetic stress through Akt.

Supplementary Materials

  • stke.sciencemag.org/cgi/content/full/12/585/eaav3249/DC1

    Fig. S1. AMPK associates with and phosphorylates mTOR within mTORC2.

    Fig. S2. AMPK promotes mTORC2 signaling in response to energetic stress induced by diverse agents in cultured cells.

    Fig. S3. Time course experiments examining the effects of glucose withdrawal, phenformin, and rotenone on AMPK activation, mTORC2 signaling, and mTOR phosphorylation.

    Fig. S4. AMPK promotes mTORC2 signaling in liver in vivo and skeletal muscle ex vivo.

    Fig. S5. TOPO cloning, sequencing, and genotyping of mTORA/A mice bearing mTOR Ser1261 knock-in alleles using CRISPR-Cas9–mediated genome editing.

    Fig. S6. AMPK and Akt promote cell survival during acute energetic stress.

  • This PDF file includes:

    • Fig. S1. AMPK associates with and phosphorylates mTOR within mTORC2.
    • Fig. S2. AMPK promotes mTORC2 signaling in response to energetic stress induced by diverse agents in cultured cells.
    • Fig. S3. Time course experiments examining the effects of glucose withdrawal, phenformin, and rotenone on AMPK activation, mTORC2 signaling, and mTOR phosphorylation.
    • Fig. S4. AMPK promotes mTORC2 signaling in liver in vivo and skeletal muscle ex vivo.
    • Fig. S5. TOPO cloning, sequencing, and genotyping of mTORA/A mice bearing mTOR Ser1261 knock-in alleles using CRISPR-Cas9–mediated genome editing.
    • Fig. S6. AMPK and Akt promote cell survival during acute energetic stress.

    [Download PDF]

Stay Connected to Science Signaling

Navigate This Article