Activity of TSC2 is inhibited by AKT-mediated phosphorylation and membrane partitioning

Abstract

Loss of tuberin, the product of TSC2 gene, increases mammalian target of rapamycin (mTOR) signaling, promoting cell growth and tumor development. However, in cells expressing tuberin, it is not known how repression of mTOR signaling is relieved to activate this pathway in response to growth factors and how hamartin participates in this process. We show that hamartin colocalizes with hypophosphorylated tuberin at the membrane, where tuberin exerts its GTPase-activating protein (GAP) activity to repress Rheb signaling. In response to growth signals, tuberin is phosphorylated by AKT and translocates to the cytosol, relieving Rheb repression. Phosphorylation of tuberin at serines 939 and 981 does not alter its intrinsic GAP activity toward Rheb but partitions tuberin to the cytosol, where it is bound by 14-3-3 proteins. Thus, tuberin bound by 14-3-3 in response to AKT phosphorylation is sequestered away from its membrane-bound activation partner (hamartin) and its target GTPase (Rheb) to relieve the growth inhibitory effects of this tumor suppressor.

Figure 1.

Tuberin localization is determined by Ser/Thr phosphorylation. (A) Western analyses of subcellular fractions from the indicated cell lines using an anti-tuberin antibody. β1-integrin and lamin A/C proteins were analyzed as fractionation controls. Similar results were also observed with canine kidney cell lysates (not depicted). (B) Fractionated NIH3T3 lysates were treated with either CIAP serine/threonine phosphatase or YOP tyrosine phosphatase at 30°C for 1 h, and the electrophoretic mobility of tuberin was determined by Western analysis. (C) Swiss3T3 fibroblasts were serum starved (SV) for 24 h followed by serum stimulation (SM) with 20% FBS for the indicated times, and fractionated lysates were immunoblotted with the indicated antibodies. Phospho-AKT (S473) and phospho- AKT (T308) were blotted as controls for the effectiveness of the indicated treatments. Total lysates (top, M + C) were analyzed as controls for tuberin levels. Tuberin protein levels within the membrane and cytosolic fractions were determined as the percentage of total tuberin detected within both fractions. A representative blot from two independent biological replications is shown. Quantitation obtained from quintuplicate runs (n = 5) from each biological replicate is shown below. (asterisks) Both the reduction in the membrane fraction and increase in cytosolic fraction of tuberin that occurred in response to serum stimulation for 1 h were significantly different than starvation controls (t test, P < 0.02) and the 6 h time point (t test, P < 0.01). (D, top) Western analyses of tuberin from membrane and cytosolic fractions of serum-starved MCF7 cells (24 h) that were stimulated for 1 h with 30 ng/ml IGF-1 in the absence or presence of 100 nM of the PI3K inhibitor wortmannin. The vertical line indicates nonadjacent lanes in a single blot. (bottom) Tuberin protein levels within the membrane and cytosolic fractions were determined as the percentage of total tuberin detected within both fractions. A representative blot from three independent biological replications is shown. Quantitation is obtained from a total of seven independent Western blots from the three experiments. (asterisks) Both the reduction in the membrane fraction and increase in cytosolic fraction of tuberin that occurred in response to IGF-1 stimulation were significantly different from starvation controls (t test, P < 0.01) and the + wortmannin (t test, P < 0.01).

Figure 2.

AKT-mediated phosphorylation of tuberin leads to subcellular translocation. (A) NIH3T3 cells were treated with 100 ng/ml EGF in the absence or presence of 200 nM wortmannin for 1 h. The fractionated cell lysates were immunoprecipitated with an anti-tuberin antibody (N19) followed by Western blot analysis with an anti–phospho-(S/T) AKT substrate antibody and an anti-tuberin antibody (C20). (B) NIH3T3 cells were serum starved for 24 h followed by treatment with 30 ng/ml IGF-1 in the absence or presence of 100 nM wortmannin for 1 h. The fractionated cell lysates were immunoprecipitated with an anti-tuberin antibody (N19) followed by Western blot analysis with an anti–phospho-(S/T) AKT substrate antibody and an anti-tuberin antibody (C20). Phospho-AKT (S473) was blotted as a control for the effectiveness of the indicated treatments. The vertical line indicates nonadjacent lanes in a single blot. (C) Western analyses of tuberin, AKT, and phospho-AKT (S473) were performed using membrane and cytosolic fractions from wild-type MCF7 and MCF7 cells that were stably transfected with constitutively active AKT (Myr-AKT). β1-integrin and LDH proteins were used as fractionation controls.

Figure 3.

Identification of S939 and S981 as phosphorylation sites that determine tuberin localization. (A) Schematic of tuberin phosphorylation sites with Flag-TSC2 constructs and their corresponding mutations listed below. (B) Western analysis of HEK293 membrane and cytosolic fractions using an anti-Flag antibody after transfection with wild-type and mutant Flag-TSC2 constructs. (C) Wild-type and mutant Flag-TSC2 constructs were overexpressed in HEK293 cells, and the membrane and cytosol lysates were immunoblotted with an anti-Flag antibody. The vertical line indicates nonadjacent lanes in a single blot.

Figure 4.

14-3-3 proteins bind and sequester tuberin in the cytosol. (A) GST–14-3-3 proteins were used to affinity purify proteins from HEK293 cells in the presence of phosphorylated or nonphosphorylated S939 and S981 tuberin peptides. Affinity-purified complexes were immunoblotted to detect the amount of tuberin interacting with 14-3-3. The vertical line indicates nonadjacent lanes in a single blot. (B) HEK293 cells were transfected with Flag-TSC2-WT or Flag-TSC2-2A in the absence or presence of 200 nM wortmannin. Western analysis of exogenous tuberin was performed after using GST–14-3-3 proteins to affinity purify Flag-tuberin. (C) HEK293 cells were transfected with Flag-TSC2 in the presence or absence of GFP-R18 14-3-3 decoy expression construct, and fractionated lysates were used for Western analyses with the indicated antibodies. LDH was used as a fractionation control.

Figure 5.

Hamartin increases the amount of tuberin retained in the membrane. (A) HEK293 cells were transfected with wild-type and mutant Flag-TSC2 constructs in the presence or absence of a Myc-TSC1 construct. Cells were fractionated and immunoblotted with indicated antibodies. (B) HEK293 cells were transfected with wild-type TSC2 constructs along with increasing amounts of Flag-TSC1. Cells were then fractionated and immunoblotted with indicated antibodies. (C) HEK293 cells were transfected with Flag-TSC2 and Myc-TSC1 constructs. The fractionated cell lysates were immunoprecipitated with an anti-Flag antibody followed by Western analysis with anti-Flag (mouse) and anti-Myc (mouse). The same blot was then reprobed with anti-tuberin (rabbit) and anti-hamartin (rabbit) as indicated.

Figure 6.

Tuberin colocalization with Rheb is disrupted in response to growth factor stimulation. HeLa cells were transfected in serum-free media with Flag-TSC2 WT and Myc-Rheb along with an untagged TSC1 construct. After 12 h without serum, cells were treated with 30 ng/ml IGF-1 for 1 h. Localization of exogenous Flag-tuberin (green) and Myc-Rheb (red) was detected by double immunofluorescence labeling and confocal microscopy. Colocalization is detected in the merged image with the yellow signal (right).

Figure 7.

AKT-mediated phosphorylation of tuberin relieves repression of mTOR-S6K signaling. (A) HEK293 cells were transfected with HA-S6K, Myc-TSC1, and Flag-TSC2 constructs as indicated. Cells were serum starved for 24 h and treated with or without 30 ng/ml IGF-1 for 1 h. Cell lysates were immunoblotted with the indicated antibodies, and phopho-S6K levels were compared with the amount of HA-S6K protein levels. Phospho-AKT (S473) was blotted as a control for the effectiveness of IGF-1 treatment. (B) Western analyses of membrane and cytosolic fractions from HEK293 cells transfected with wild-type or mutant Flag-TSC2 constructs. After 9 h, serum was either removed or allowed to remain for 24 h. The protein levels of phosphorylated endogenous S6K were compared with total S6K protein levels. Phospho-AKT (S473) and phospho-AKT (T308) were blotted as controls for the effectiveness of the indicated treatments.

Figure 8.

AKT phosphorylation of tuberin promotes Rheb-induced S6K1 activation through increased Rheb-GTP loading. (A) HEK293 cells coexpressing HA-S6K1, Myc-Rheb, Flag-TSC1, and either Flag-TSC2-WT or Flag-TSC2-2A (S939A and S981A) were serum starved and, where indicated, the PI3K–AKT signaling pathway was stimulated with 100 nM insulin for 7.5 and 15 min. These cells were subjected to in vivo radiolabeling, and the level of guanine nucleotide bound to immunoprecipitated Myc-Rheb was quantified. A representative blot from three independent biological replications is shown. The mean of the percentage of total Myc-Rheb bound to GTP (active state) is shown in the bar figure. (B) In parallel, HEK293 cells treated as in A were subjected to S6K1 activity assays. HA-S6K1 used in the activity assay was analyzed with an anti-HA antibody. ³²P-incorporation into GST-S6 substrate was quantified using a phosphorimager, and the fold activation of S6K1 is graphed. Protein levels of Flag-TSC1, Flag-TSC2, and AKT, and level of AKT phosphorylation at S473, were determined using Western analyses.

Figure 9.

Tuberin's GAP activity for Rheb in vitro is not affected by AKT phosphorylation. HEK293 cells coexpressing Flag-TSC1 and Flag-TSC2-WT, Flag-TSC2-S939A, Flag-TSC2-S981A, or Flag-TSC2-2A were serum starved and then stimulated with 100 nM insulin for 30 min. TSC1 and -2 were immunoprecipitated from the lysates and subjected to RhebGAP activity assays for 15, 30, or 60 min. Immunocomplexes containing wild-type and mutant tuberin activated Rheb similarly, as indicated by quantitation of percentage of GDP. The vertical line indicates nonadjacent lanes resolved on a single thin layer chromatography plate. For quality control of the α-[³²P]GTP used in the assay, the guanine nucleotide eluted from the 60-min vector control was resolved on a separate thin layer chromatography plate and shows that Rheb is 98% bound to GTP in the absence of hamartin and tuberin. Each assay contained approximately equal amounts of TSC1 and -2, and AKT activation was confirmed with a phospho-T1462 antibody by Western blot analysis (bottom).

Figure 10.

Model for regulation of tuberin under conditions of mitogenic sufficiency. (A) Tuberin normally functions as a GAP for Rheb within an intracellular membrane compartment, inhibiting Rheb and mTOR signaling, which suppresses cell growth. (B) Upon stimulation by growth factors, AKT is activated, leading to phosphorylation and cytosolic sequestration of tuberin by 14-3-3 proteins. This cytosolic translocation relieves tuberin repression of the Rheb-mTOR signaling, stimulating cell growth.