Leucine induced dephosphorylation of Sestrin2 promotes mTORC1 activation

Abstract

The studies described herein were designed to explore the role of Sestrin2 in mediating the selective action of leucine to activate mTORC1. The results demonstrate that Sestrin2 is a phosphoprotein and that its phosphorylation state is responsive to the availability of leucine, but not other essential amino acids. Moreover, leucine availability-induced alterations in Sestrin2 phosphorylation correlated temporally and dose dependently with the activation state of mTORC1, there being a reciprocal relationship between the degree of phosphorylation of Sestrin2 and the extent of repression of mTORC1. With leucine deprivation, Sestrin2 became more highly phosphorylated and interacted more strongly with proteins of the GATOR2 complex. Notably, in cells lacking the protein kinase ULK1, the activation state of mTORC1 was elevated in leucine-deficient medium, such that the effect of re-addition of the amino acid was blunted. In contrast, overexpression of ULK1 led to hyperphosphorylation of Sestrin2 and enhanced its interaction with GATOR2. Neither rapamycin nor Torin2 had any effect on Sestrin2 phosphorylation, suggesting that leucine deprivation-induced repression of mTORC1 was not responsible for the action of ULK1 on Sestrin2. Mass spectrometry analysis of Sestrin2 revealed three phosphorylation sites that are conserved across mammalian species. Mutation of the three sites to phospho-mimetic amino acids in exogenously expressed Sestrin2 promoted its interaction with GATOR2 and dramatically repressed mTORC1 even in the presence of leucine. Overall, the results support a model in which leucine selectively promotes dephosphorylation of Sestrin2, causing it to dissociate from and thereby activate GATOR2, leading to activation of mTORC1.

Keywords: Amino acids; Mechanistic target of rapamycin; Rag GTPases; Signaling; ULK1.

Figure 1

Leucine alters the electrophoretic migration of Sestrin2. (A) HEK293 cells were incubated in complete medium (CM) or medium lacking leucine (-L) for 2 h prior to harvest and samples were resolved on a BioRad Criterion gel prior to Western blot analysis for p70S6K1 (phospho-Thr389 (p-p70) and total (p70)), Sestrin2 (Sesn2), and GAPDH. (B and C) HEK293 cells were incubated in complete medium, medium lacking leucine for 2 h prior to stimulation with 760 μM leucine (LAB) for 30 min prior to harvest and cell extracts were fractionated on 7.5% polyacrylamide gels with 0.19% bisacrylamide. (D) HeLa cells were incubated as in panels B and C. All samples were analyzed on the same blot, but not in contiguous lanes; a white line separates non-contiguous lanes. * indicates significantly different than CM, p <0.05. Results represent the mean ± SEM of 3 independent experiments with 2 replicates performed in each experiment.

Figure 2

The altered migration of Sestrin2 is specific to changes in leucine content. (A and B) HEK293 cells were incubated in complete medium (CM) or medium lacking leucine, arginine, glutamine, glycine, and histidine (−5AA) for 2 h prior to stimulation with either 760 μM leucine (+L), 400 μM arginine (+R), 4 mM glutamine (+Q), or 200 μM histidine (+H) alone or in combination for 30 min prior to harvest. Samples were subjected to Western blot analysis for (A) total p70S6K1 (p70), p70S6K1 phosphorylated on Thr389 (p-p70), and GAPDH, or (B) Sestrin2. Letters above the bars indicate significant differences. Significance was set at p < 0.05 for all analysis. Results represent the mean ± SEM of 3 independent experiments with 1-2 replicates performed in each experiment.

Figure 3

Time course and dose response analysis of leucine-induced changes in Sestrin2 electrophoretic migration. (A) HEK293 cells were incubated in medium lacking leucine (−L) for the indicated time prior to harvest. (B) HEK293 cells were incubated in complete medium (CM) or medium lacking leucine for 2 h prior to stimulation with 760 μM leucine (LAB) for the indicated time. (C and D) HEK293 cells were incubated in CM or medium with the indicated concentration of leucine for 2 h prior to harvest. Sestrin2 (Sesn2) and GAPDH protein content and phosphorylation of p70S6K1 on Thr389 (p-p70) were determined by Western blot analysis. * indicates significantly different than 0 μM leucine. Significance was set at p < 0.05 for all analyses. Results represent the mean ± SEM of 3 independent experiments with 2 replicates performed in each experiment.

Figure 4

Leucine-induced changes in Sestrin2 electrophoretic mobility are due to phosphorylation. (A) HEK293 cells were transiently transfected with a control plasmid or one encoding FLAG-tagged Sestrin2 18 h prior to incubation in phosphate-free medium containing (CM) or lacking leucine (−L) for 1 h. [³²P]orthophosphate was added to the medium and cells were harvested 30 min later. FLAG-tagged Sestrin2 was immunoprecipitated, subjected to SDS-PAGE, and the gel was dried and ³²P_i incorporation was imaged on a Typhoon Imager. Incorporation of ³²P_i was from a single experiment. An aliquot of the immunoprecipitate was also subjected to Western blot analysis for FLAG content. (B) HEK293 cells were incubated in medium containing or lacking leucine for 2 h prior to harvest. A portion of the cell extracts were incubated with lambda protein phosphatase prior to Western blot analysis for Sestrin2 (Sesn2) and 4E-BP1 as indicated in the figure.

Figure 5

Identification of Sestrin2 phosphorylation sites by mass spectrometry. Sestrin2 was isolated from cells incubated in the presence or absence of leucine as described under “Methods” and sent to the Taplin Mass Spectrometry Facility for analysis. Three sets of samples were independently analyzed. Ser249 was identified as being phosphorylated in cells incubated in medium containing leucine and Thr232 and Ser279 were identified as being phosphorylated in cells deprived of leucine. (A) Alignment of Sestrin2 phosphorylation sites across species. Hs, Homo sapiens; Pt, Pan troglodytes; Bt, Bos Taurus; Mm, Mus musculus; Rn, Rattus norvegicus; Dr, Danio rerio. (B) Multisequence alignment of human Sestrin2 phosphorylation sites with Sestrin1 and Sestrin3.

Figure 6

Exogenous expression of a Sestrin2 phospho-mimetic variant results in repressed p70S6K1 phosphorylation and interaction with Mios in a leucine-independent manner. HEK293 cells were transiently transfected with either an empty vector (EV), a plasmid expressing wild-type Sestrin2 (WT Sesn2), or a plasmid expressing a triple phospho-mimetic Sestrin2 variant (Sesn2EDE). (A) Cells were incubated in medium lacking serum and leucine for 2 h (-SL) prior to stimulation with 760 μM leucine for 30 min (LAB). Phosphorylation of p70S6K1 on Thr389 (p-p70), FLAG-tagged Sestrin2 (FLAG-Sesn2), and GAPDH were evaluated by Western blot analysis. * indicates significantly different than LAB, # indicates significantly different than EV. Results represent the mean ± SEM of 5 independent experiments with 2-3 replicates performed in each experiment. (B) Cells were incubated in complete medium (CM) or deprived of serum and leucine for 2 h (-SL). FLAG-tagged Sestrin2 was immunoprecipitated from the supernatant fraction of cell lysates. Co-immunoprecipitation of Mios with FLAG-tagged Sestrin2 was assessed by Western blot analysis. Letters above the bars indicate significant differences. Results represent the mean ± SEM of 3 independent experiments with 2 replicates performed in each experiment.

Figure 7

ULK1 promotes Sestrin2 phosphorylation and interaction with GATOR2. (A) Wild type (ULK1^+/+) and ULK1 knockout (ULK1^−/−) MEF were incubated in medium lacking serum and leucine for 2 h (-SL) prior to stimulation with 760 μM leucine for 30 min (LAB). Phosphorylation of p70S6K1 on Thr389 (p-p70), total p70S6K1 (p70), ULK1, Sestrin2 (Sesn2), and GAPDH were evaluated by Western blot analysis. Results are representative of 2 independent experiments with 2-3 replicates performed in each experiment. Letters above the bars indicate significant differences. (B) HEK293 cells were transfected with the indicated plasmids, and the next day cell lysates were subjected to Western blot analysis for FLAG-Sestrin2 (FLAG-Sesn2) and HA-ULK1. Results are representative of 3 independent experiments with 2 replicates performed in each experiment. (C) HEK293 cells were transfected with an empty vector (EV) and/or plasmids expressing HA-p70S6K1 (HA-p70) and HA-ULK1. The next day cell lysates were immunoprecipitated using anti-HA beads and immunoprecipitates were subjected to Western blot analysis for HA and p70S6K1 phosphorylated on Thr389 (p-p70). Results represent 1 experiment with 2 replicates. (D) HEK293 cells were transfected with the plasmids indicated in the figure, and the next day FLAG-Sestrin2 was immunoprecipitated from the supernatant fraction of cell lysates. Co-immunoprecipitation of Mios with FLAG-tagged Sestrin2 was assessed by Western blot analysis. Results are representative of 2 independent experiments with 2 replicates performed in each experiment.

Figure 8

Working model for the mechanism whereby leucine availability acts to repress mTORC1 activity via phosphorylation of the Sestrin2 linker domain at Thr232, Ser249, and Ser279. CTD, C-terminal domain; NTD, N-terminal domain; linker, unstructured region between the C-terminal and N-terminal domains.