Rheb Activation Of Mtor

Posted on
  1. ‡ and
  1. Program in Signal Transduction Research, The Burnham Institute, La Jolla, California 92037
  1. Rheb can bind directly to the mTOR kinase domain, and association with inactive nucleotide-deficient Rheb mutants traps mTOR in a catalytically inactive state. Nevertheless, Rheb-GTP targets other than mTOR, such as FKBP38 (FK506-binding protein 38) and/or PLD1 (phospholipase D1), may also contribute to mTOR activation.
  2. Indeed, we found that Rheb is required for EGF-dependent mTOR activation in spinal cord astrocytes, whereas the Ras–MAP kinase pathway does not appear to be involved. Moreover, astrocyte growth and EGF-dependent chemoattraction were inhibited by the mTOR-selective drug rapamycin.
  3. Rheb activation of the nutrient and energy‐sensitive TOR pathway leads to the direct phosphorylation of two known downstream translational control targets by mTOR, the 40S ribosomal S6 kinase 1 (S6K1) and the eukaryotic translation initiation factor 4E (eIF4E)‐ binding protein 1 (4E‐BP1).
  4. MTOR Signalling, Nutrients and Disease 223 Activation of mTORC1 in two steps: Rheb-GTP activation of catalytic function and increased binding of substrates to raptor1 Joseph Avruch*2, Xiaomeng Long*, Yenshou Lin*, Sara Ortiz-Vega*, Joseph Rapley*, Angela Papageorgiou*.
  5. Rheb Activates mTORC1 In Vitro. The complex is inactive, however, the addition of recombinant Rheb causes activation of mTORC1 as detected by the phosphorylation of substrate protein 4E-BP1 using antibody against phospho-4E-BP1 ( Figure 3.2 ). The activation of mTORC1 by Rheb is dependent on the binding of GTP.

Rheb/mTOR Activation and Regulation in Cancer: Novel Treatment Strategies beyond Rapamycin Author(s): Justin T. Babcock, Lawrence A. Department of Biochemistry and Molecular Biology, Indiana University School of Medicine, 635 Barnhill Drive MS4075, Indianapolis, IN, USA.

  1. ↵‡ Supported by a Kirschstein-NRSA Fellowship F32 CA099354 from the NCI, National Institutes of Health. To whom correspondence should be addressed: Program in Signal Transduction Research, The Burnham Institute, 10901 N. Torrey Pines Road, La Jolla, CA 92037. Tel.: 858-646-3100 (ext. 3943); Fax: 858-713-6274; E-mail: gchiang{at}burnham.org.

Abstract

The mammalian target of rapamycin (mTOR) coordinates cell growth with the growth factor and nutrient/energy status of the cell. The phosphatidylinositol 3-kinase-AKT pathway is centrally involved in the transmission of mitogenic signals to mTOR. Previous studies have shown that mTOR is a direct substrate for the AKT kinase and identified Ser-2448 as the AKT target site in mTOR. In this study, we demonstrate that rapamycin, a specific inhibitor of mTOR function, blocks serum-stimulated Ser-2448 phosphorylation and that this drug effect is not explained by the inhibition of AKT. Furthermore, the phosphorylation of Ser-2448 was dependent on mTOR kinase activity, suggesting that mTOR itself or a protein kinase downstream from mTOR was responsible for the modification of Ser-2448. Here we show that p70S6 kinase phosphorylates mTOR at Ser-2448 in vitro and that ectopic expression of rapamycin-resistant p70S6 kinase restores Ser-2448 phosphorylation in rapamycin-treated cells. In addition, we show that cellular amino acid status, which modulates p70S6 kinase (S6K1) activity via the TSC/Rheb pathway, regulates Ser-2448 phosphorylation. Finally, small interfering RNA-mediated depletion of p70S6 kinase reduces Ser-2448 phosphorylation in cells. Taken together, these results suggest that p70S6 kinase is a major effector of mTOR phosphorylation at Ser-2448 in response to both mitogen- and nutrient-derived stimuli.

The mammalian target of rapamycin (mTOR)1 is a member of the phosphatidylinositol 3 (PI-3)-kinase-related kinase family (PIKK), which includes ATM, ATR, hSMG-1, and DNA-PK (1–3). These large Ser/Thr protein kinases play essential roles in cellular responses to growth factors and stress. In particular, mTOR plays a critical role in coordinating cell growth with growth factor inputs as well as cellular nutrient and energy status.

The bacterially derived macrolide ester, rapamycin, is clinically approved as an immunosuppressant and shows promising anti-tumor activity. Rapamycin, when complexed with FKBP-12 (FK506-binding protein, 12 kDa), binds specifically to mTOR at a conserved stretch of ∼100 amino acids termed the FKBP-12·rapamycin binding domain (4). Mutation of a critical serine residue (Ser-2035) in the FKBP-12·rapamycin binding domain to a more bulky amino acid, such as isoleucine, abrogates FKBP-12·rapamycin binding to mTOR, and generates a rapamycin-resistant form of mTOR (4, 5). The mechanism by which rapamycin inhibits mTOR function remains poorly understood.

Recent studies have provided significant insights into the growth factor and nutrient signaling pathway(s) upstream of mTOR. Stimulation of many growth factor receptors leads to activation of the PI-3 kinase-AKT pathway, and it appears that this pathway is centrally involved in the coupling of mitogenic stimuli to mTOR. Moreover, loss of the tumor suppressor PTEN provides a powerful signal for tumor progression and simultaneously confers increased sensitivity to the anti-proliferative effect of rapamycin (6, 7). Nutrient/energy status also regulates mTOR signaling. Although it has been suggested that mTOR directly senses ATP levels in the cell (8), a more plausible mechanism of mTOR regulation involves the ATP-regulated LKB1/AMPK pathway (9, 10). Signaling through both the PI-3 kinase/AKT and LKB1/AMPK pathways as well as nutrient cues all converge at the level of the TSC1/2 complex, which genetic and biochemical evidence establish as a negative regulator of mTOR (11–14). TSC2 is a direct target of both AKT and AMPK. Phosphorylation of TSC2 by AKT results in the inactivation of the TSC1/2 complex (15–17), whereas phosphorylation of TSC2 by AMPK results in enhanced activity (9).

Mtor

In addition to the regulation of mTOR by the TSC/Rheb pathway, our laboratory and others have provided evidence that mTOR may be a direct substrate for AKT (18, 19). The proposed AKT phosphorylation site (Ser-2448) in mTOR lies within a C-terminal regulatory region, which, when deleted, results in elevated mTOR activity in vitro and in cells (18, 20).

In this study, we demonstrate that mTOR phosphorylation at Ser-2448 is blocked by rapamycin, and this effect is independent of the AKT activation status, which suggests that Ser-2448 phosphorylation is catalyzed by a protein kinase other than AKT. We show that Ser-2448 is directly phosphorylated by p70S6 kinase in vitro and that expression of a rapamycin-resistant p70S6 kinase in cells maintains Ser-2448 phosphorylation in the presence of rapamycin. Furthermore, changes in amino acid availability, which affects p70S6 kinase activity, also modulates Ser-2448 phosphorylation. Finally, Ser-2448 phosphorylation is reduced in cells by siRNA-mediated depletion of p70S6 kinase. Taken together, these data suggest that p70S6 kinase is the major protein kinase responsible for Ser-2448 phosphorylation in mammalian cells.

EXPERIMENTAL PROCEDURES

Antibodies and Reagents—Rapamycin was obtained from the NCI, National Institutes of Health (Bethesda, MD). Wortmannin, the anti (α)-FLAG M2 antibody, α-FLAG M2-agarose, and the FLAG peptide were obtained from Sigma. The monoclonal α-AU1 and α-HA (12CA5) antibodies were purchased from Covance (Berkeley, CA). α-AKT and α-phospho-AKT (Ser-473) antibodies were purchased from Cell Signaling Technology (Beverly, MA), and α-p70S6 kinase antibody was purchased from Santa Cruz Biotechnology (Santa Cruz, CA). The α-mTOR and α-phospho-Ser-2448 mTOR (α-mTORp2) antibodies have been described previously (18, 21).

Cell Culture, DNA Constructs, siRNA, and Transfections—HEK 293 cells, MCF-7 cells, and HeLa cells were maintained in high glucose Dulbecco's modified essential medium (DMEM) (Irvine Scientific, Santa Ana, CA) supplemented with 10% FCS (Hyclone, Logan, UT). HEK 293 cells stably expressing AU1 epitope-tagged, rapamycin-resistant (SI) mTOR have been described (22). For experiments involving serum starvation, cells were washed twice with phosphate-buffered saline and then serum-starved for 16–18 h in DMEM supplemented with 0.1% FCS. For serum stimulation, cells were incubated for 15 min with 10% FCS prior to cell harvest. For amino acid deprivation, cells were washed twice with phosphate-buffered saline and incubated for 50 min in high glucose DMEM minus amino acids. Cells were restimulated for 10 min with serum-free, high glucose DMEM prior to cell harvest. Where indicated, cells were pretreated for 30 min with rapamycin or wortmannin before stimulation with amino acids or serum.

The wild type (WT), SI phosphorylation site mutants (Ser-2448 → Ala and Ser-2448 → Glu), and rapamycin-resistant, kinase-inactive mTOR cDNAs cloned into the expression vector pcDNA3 (Invitrogen) have been described (18, 22). The HA-tagged wild-type p70S6 kinase and the FLAG-tagged p70S6 kinase were provided by Dr. Naohiro Terada (University of Florida, Gainesville, FL). The HA-tagged, kinase-inactive (K100R) p70S6K mutant in the expression vector pRK7 and the HA-tagged, rapamycin-resistant ΔN/C p70S6K in the expression vector pBJ were provided by Dr. John Blenis (Harvard Medical School, Boston, MA).

Mtor Activation Training

The luciferase siRNA control and the p70S6K siRNA SMARTpool were purchased from Dharmacon (Lafayette, CO). siRNAs were transfected into HeLa cells with Oligofectamine (Invitrogen) according to the manufacturer's protocol. Cells were harvested 48 h posttransfection.

HEK 293 cells were transfected with FuGENE 6 (Roche Applied Science) according to the manufacturer's protocol. Cells were used in experiments at 36–48 h posttransfection.

Immunoprecipitations, in Vitro Kinase Reactions, and Immunoblotting—For mTOR immunoprecipitations, cells were suspended in lysis buffer (50 mm Tris-HCl, pH 7.4, 100 mm NaCl, 50 mm β-glycerophosphate, 10% glycerol (w/v), 1% Triton X-100, 1 mm EDTA, 10 μg/ml aprotinin, 1 μg/ml pepstatin A, 10 μg/ml leupeptin, 2 mm phenylmethylsulfonyl fluoride, 20 μm microcystin-LR, 25 mm NaF). Cleared extracts were immunoprecipitated with rabbit α-mTOR antibody or α-AU1 antibody. Immune complexes were collected on protein A-Sepharose (Sigma) or α-mouse IgG-agarose (Sigma), washed three times with lysis buffer, and resuspended in SDS-PAGE sample buffer.

For p70S6 kinase reactions, transfected HEK 293 cells were lysed as described above. Soluble proteins were immunoprecipitated with monoclonal α-HA antibody. Immune complexes were collected on α-mouse IgG-agarose, washed three times in lysis buffer, followed by one wash in p70S6 kinase buffer (50 mm Tris-Cl, pH 7.4, 10 mm NaCl, 10 mm MgCl2, 10% glycerol, 1 mm dithiothreitol). Immunoprecipitates were resuspended in 40 μl of p70S6 kinase buffer, and kinase reactions were initiated by addition of 1 μg of GST-mTOR RD fragment (amino acids 2405–2517) and 100 μm ATP. Reactions were incubated for 20 min at 30 °C and terminated with 2× SDS-PAGE sample buffer.

For phosphorylation of full-length mTOR by soluble p70S6 kinase, HEK 293 cells transfected with a FLAG-tagged p70S6 kinase expression vector were lysed as described above. Soluble proteins were immunoprecipitated with α-FLAG-agarose, and bound p70S6 kinase was eluted with p70S6 kinase buffer containing 0.15 mg/ml FLAG peptide. Anti-mTOR immunoprecipitates prepared from serum-starved HEK 293 cells were resuspended in 30 μl of p70S6 kinase buffer, and kinase reactions were initiated by the addition of 10 μl of soluble Flag-p70S6 kinase and 100 μm ATP. Reactions were incubated for 20 min at 30 °C and terminated with 2× SDS-PAGE sample buffer.

Samples were resolved by SDS-PAGE and transferred to polyvinylidene difluoride membranes (PerkinElmer Life Sciences). Western blotting was carried out essentially as described (21). Bound antibodies were detected with horseradish peroxidase-conjugated protein A (Amersham Biosciences) for polyclonal antibodies or horseradish peroxidase-conjugated goat anti-mouse antibody (Amersham Biosciences) for monoclonal antibodies followed by enhanced chemiluminescence using Western Lightning reagent (PerkinElmer Life Sciences).

RESULTS

Rapamycin Blocks Serum-stimulated Ser-2448 Phosphorylation—Previous reports demonstrated that mTOR was phosphorylated in vitro by AKT at a serine residue (Ser-2448) located in a C-terminal “repressor domain” of mTOR (18, 19). In later studies, we unexpectedly observed that phosphorylation at the Ser-2448 site was consistently suppressed in cells treated with rapamycin. As short term exposure to rapamycin does not inhibit AKT, these results suggested that Ser-2448 phosphorylation was mediated by a rapamycin-sensitive protein kinase that shared an overlapping substrate preference with AKT. To test this hypothesis, we cultured MCF-7 cells overnight in low serum (0.1% FCS)-containing medium. The cells were restimulated with 10% FCS in the absence or presence of 100 nm rapamycin or 100 nm wortmannin (Fig. 1A). As previously shown, serum stimulation increased Ser-2448 phosphorylation, with a concomitant increase in AKT phosphorylation at Ser-473. Treatment of the cells with 100 nm wortmannin, a concentration that inhibits PI-3 kinase, but not mTOR or other PI-3 kinase-related kinase family members (23), blocked the serum-dependent phosphorylation of mTOR, as well as the modification of AKT, consistent with the requirement for PI-3 kinase in AKT activation (24). As noted previously, Ser-2448 phosphorylation was also strongly suppressed by rapamycin; however, in contrast to wortmannin, rapamycin had no effect on the phosphorylation of AKT at Ser-473. Thus, changes in the phosphorylation of mTOR at Ser-2448 were not tightly correlated with the activation state of AKT in these cells.

We next asked whether binding of FKBP-12·rapamycin to mTOR was required for the observed decrease in Ser-2448 phosphorylation. To test this, we used HEK 293 cells that stably express a mTOR polypeptide bearing a Ser-2035 → Ile mutation in the FKBP-12·rapamycin binding domain (SI mTOR). The SI mTOR protein exhibits markedly reduced binding affinity for FKBP-12·rapamycin and confers rapamycin resistance when transfected into mammalian cells (5). We repeated the experiments described above, this time with serum-starved HEK 293 SI mTOR cells (Fig. 1B). In these cells, serum stimulation provoked a clear increase in the phosphorylation of the ectopically expressed SI mTOR protein at Ser-2448. The increase in Ser-2448 phosphorylation was inhibited by wortmannin; however, rapamycin had little or no effect on the phosphorylation of the SI mTOR protein. These results indicate that, as is the case for endogenous wild-type mTOR (see Fig. 1A), phosphorylation of the Ser-2448 site in SI mTOR remains dependent on PI 3-kinase. However, in contrast to the endogenous protein, phosphorylation of SI mTOR at Ser-2448 is resistant to rapamycin. This result confirms that the inhibitory effect of rapamycin on Ser-2448 phosphorylation is dependent on the inhibition of mTOR-dependent signaling by this drug.

mTOR Kinase Activity Is Required for Ser-2448 Phosphorylation—To test whether mTOR kinase activity was required for Ser-2448 phosphorylation, we transiently transfected HEK 293 cells with AU1 epitope-tagged, WT, SI mTOR, or a rapamycin-resistant, kinase-inactive mTOR double mutant. After 48 h, the transfected cells were treated with 100 nm rapamycin, and the phosphorylation of Ser-2448 was comparatively examined in the various AU1-tagged mTOR proteins (Fig. 1C). Rapamycin treatment strongly reduced the phosphorylation of both the WT and rapamycin-resistant, kinase-inactive mTOR polypeptides, whereas phosphorylation of the rapamycin-resistant SI mTOR was unaffected by the drug. Like SI mTOR, the rapamycin-resistant, kinase-inactive mTOR double mutant does not bind FKBP-12·rapamycin; however, this kinase-inactive mTOR mutant cannot confer rapamycin resistance when transfected into mammalian cells (18). Thus, the phosphorylation of the Ser-2448 site depends on the expression of catalytically active mTOR. Because Ser-2448 is not a mTOR autophosphorylation site (18), this finding raises the possibility that a downstream target of mTOR directly or indirectly mediates the phosphorylation of Ser-2448 in intact cells.

View larger version:
Fig. 1.

Inhibition of Ser-2448 phosphorylation by rapamycin and wortmannin.A, MCF-7 cells were starved in DMEM + 0.1% FCS for 16 h. Cells were incubated for 30 min with 100 nm rapamycin (Rap) or 100 nm wortmannin (Wm) as indicated prior to stimulation for 15 min with 10% FCS. Anti-mTOR immunoprecipitates were immunoblotted with phospho-Ser-2448-specific mTOR antibodies (α-mTORp2), and the same membrane was reprobed with α-mTOR antibodies. Whole cell extracts from the same samples were immunoblotted with α-phospho-AKT (Ser-473) antibodies followed by α-AKT antibodies. B, HEK 293 cells stably expressing AU1 epitope-tagged SI mTOR were serum-starved for 16 h in DMEM + 0.1% FBS. Cells were stimulated with 10% FBS in the presence of the indicated drug and were harvested after 15 min. Anti-AU1 immunoprecipitates were immunoblotted with α-mTORp2, and the same membrane was stripped and reprobed with α-mTOR antibodies. C, HEK 293 cells were transfected with AU1 epitope-tagged WT, SI, or rapamycin-resistant, kinase-inactive (SIDA) mTOR cDNAs. At 48 h posttransfection, cells were treated with 100 nm rapamycin for 30 min where indicated and harvested. Anti-AU1 immunoprecipitates were immunoblotted with α-mTORp2, and the same membrane was stripped and reprobed with α-mTOR antibodies. The data shown are representative of three independent experiments.

Phosphorylation of Ser-2448 by p70S6 Kinase in Vitro—Of the known mTOR target proteins, p70S6 kinase represents a prime candidate for the putative mTOR-regulated Ser-2448 kinase. To test whether p70S6 kinase phosphorylates mTOR at Ser-2448 in vitro, we transfected HEK 293 cells with an expression plasmid encoding a HA-tagged version of wild-type p70S6 kinase (WT) or a catalytically inactive p70S6 kinase mutant bearing a Lys to Arg substitution in the catalytic domain (KR) (25). Immune complex kinase reactions were performed with α-HA immunoprecipitates as a source of enzyme and a GST fusion protein containing mTOR amino acids 2405–2517 (GST-mTOR RD) as substrate. Phosphate incorporation into the substrate was visualized with the α-phospho-Ser-2448 antibody, α-mTORp2 (Fig. 2A). Immunoprecipitates containing the recombinant WT p70S6 kinase, but not the kinase-inactive KR mutant, mediated substantial phosphorylation of GST-mTOR RD at the Ser-2448 site. As a control for the specificity of the α-mTORp2 antibody, we repeated the immune complex kinase assays with a GST-mTOR RD fusion protein containing a Ser-2448 → Ala substitution as the substrate. Incubation of this substrate with WT p70S6 kinase-containing immunoprecipitates failed to increase the immunoreactivity of the mutated mTOR RD fragment with the α-mTORp2 antibody (data not shown).

View larger version:
Fig. 2.

Phosphorylation of Ser-2448 by p70S6 kinase.A, HEK 293 cells were transfected with empty vector, HA-tagged wild type (WT) or kinase-inactive (KR) p70S6 kinase. At 48 h posttransfection, cells were harvested, and HA-tagged p70S6 kinase was immunoprecipitated with α-HA antibodies. Top panel, immune complex kinase assay performed with α-HA immunoprecipitates and GST-mTOR RD (amino acids 2405–2517) as substrate. Substrate phosphorylation was visualized by immunoblotting with α-mTORp2. Bottom panel, the same membrane was immunoblotted with α-HA antibodies. Simcity societies patch windows 10. B, mTOR was immunoprecipitated from serum-starved HEK 293 cells and used as the substrate for in vitro kinase assays with soluble FLAG-tagged p70S6K. Substrate phosphorylation was visualized by immunoblotting with α-mTORp2, and the membrane was stripped and reprobed with α-mTOR antibodies. FLAG-tagged p70S6K was visualized by immunoblotting with anti-FLAG antibodies. C, HEK 293 cells were co-transfected with AU1-tagged wild-type mTOR and either wild-type (WT) or rapamycin-resistant (ΔN/C) p70S6 kinase. At 48 h posttransfection, cells were treated for 30 min with 100 nm rapamycin where indicated and harvested. α-AU1 immunoprecipitates were immunoblotted with α-mTORp2, and the same membrane was stripped and reprobed with α-AU1 antibodies. The data shown are representative of three independent experiments.

We also examined whether p70S6 kinase phosphorylated full-length mTOR in vitro. To generate this substrate, we serum-starved HEK 293 cells to reduce Ser-2448 phosphorylation to basal levels and isolated mTOR by immunoprecipitation. We then assayed the ability of a soluble FLAG-tagged p70S6 kinase to phosphorylate the mTOR immunoprecipitates. As described above, we visualized phosphate incorporation into Ser-2448 with the α-mTORp2 antibody (Fig. 2B). Anti-mTOR immunoprecipitates incubated with ATP alone displayed a basal level of Ser-2448 phosphorylation. The background phosphorylation of mTOR at Ser-2448 was observed when the kinase reactions were performed in the absence of Mg2+-ATP, indicating that the basal Ser-2448 phosphorylation had occurred prior to isolation of mTOR from the cells (data not shown). In agreement with our previous result with our GST-RD fusion protein, Ser-2448 phosphorylation of full-length mTOR was substantially increased upon addition of ATP and Flag-p70S6 kinase. Taken together, these results indicate p70S6 kinase shares with AKT (18, 19) the ability to phosphorylate mTOR at Ser-2448, at least under in vitro assay conditions.

Rapamycin-resistant p70S6 Kinase Restores Ser-2448 Phosphorylation—If p70S6 kinase serves as an important Ser-2448 kinase in intact cells, then phosphorylation of mTOR at this residue should display resistance to rapamycin in cells transfected with a drug-resistant p70S6 kinase mutant. To test this hypothesis, we co-expressed AU1-epitope-tagged WT mTOR with either wild-type p70S6 kinase (WT) or a N- and C-terminally truncated p70S6 kinase mutant (ΔN/C) whose catalytic activity is insensitive to rapamycin (25). As expected, phosphorylation of mTOR at Ser-2448 remained sensitive to rapamycin in cells transfected with WT p70S6 kinase (Fig. 2C). In contrast, rapamycin failed to suppress Ser-2448 phosphorylation in the ΔN/C p70S6 kinase-expressing cells. These results strongly suggest that p70S6 kinase is a major effector of Ser-2448 phosphorylation in serum-stimulated cells.

Ser-2448 Phosphorylation Is Stimulated by Amino Acids— The activity of p70S6 kinase is regulated by extracellular amino acids, as well as polypeptide growth factors (26, 27). If p70S6 kinase is a major contributor to the phosphorylation of mTOR at Ser-2448 in intact cells, then we would expect that readdition of amino acids to starved cells would lead to concomitant increases in p70S6 kinase activity and Ser-2448 phosphorylation. To address this prediction, we deprived HEK 293 cells of amino acids and then restimulated the cells with amino acids in the absence or presence of rapamycin (Fig. 3). Amino acid starvation decreased Ser-2448 phosphorylation to a level comparable with that observed in rapamycin-treated cells cultured under amino-acid-replete conditions. Readdition of amino acids to these cells provoked a strong increase in Ser-2448 phosphorylation, which was paralleled by the appearance of electrophoretically retarded forms of p70S6 kinase, a well established marker of kinase activation. Rapamycin exposure blocked the phosphorylation of p70S6 kinase and mTOR induced by cellular stimulation with amino acids, indicating that both responses were dependent on mTOR signaling functions. Importantly, neither amino acid stimulation nor rapamycin treatment altered the activation state of AKT, as monitored by changes in the phosphorylation of this protein kinase at Ser-473. Thus, these results argue that p70S6 kinase, rather than AKT, plays the dominant role in the regulation of mTOR phosphorylation at Ser-2448 by amino acids.

View larger version:
Fig. 3.

Amino acid stimulation increases Ser-2448 phosphorylation. HEK 293 cells were grown overnight in DMEM + 10% FCS. Cells were cultured in serum-free DMEM (+) or were starved for 50 min in serum-free DMEM minus amino acids (–). The indicated samples were stimulated for 10 min with serum-free DMEM plus amino acids (–/+). Cells were treated for 30 min with 100 nm rapamycin prior to harvest. Anti-mTOR immunoprecipitates were immunoblotted with α-mTORp2, and the same membrane was reprobed with α-mTOR antibodies. Whole cell extracts from the same samples were probed with α-p70S6 kinase antibodies, α-phospho-AKT (Ser-473) antibodies, and α-AKT antibodies. The data shown are representative of three independent experiments.

Depletion of p70S6 Kinase Expression in Cells Reduces Ser-2448 Phosphorylation—To confirm whether p70S6 kinase was responsible for Ser-2448 phosphorylation in cells, we used siRNA to down-regulate p70S6 kinase expression (Fig. 4). Transfection of HeLa cells with p70S6 kinase-specific siRNA reduced p70S6 kinase expression with a corresponding decrease in Ser-2448 phosphorylation, whereas control siRNA-transfected cells were unaffected. In both control and p70S6 kinase siRNA-treated cells, AKT phosphorylation was unaffected. These data provide compelling evidence that p70S6 kinase is responsible for Ser-2448 phosphorylation in cells.

DISCUSSION

Phosphorylation of mTOR at Ser-2448 has become a popular biomarker for the activation state of the PI-3 kinase pathway as well as the activation status of mTOR (28–31). Based on the knowledge that AKT phosphorylates the Ser-2448 site, the operative assumption is that changes in Ser-2448 phosphorylation reflect alterations in PI-3 kinase activity induced by endogenous factors or by therapeutic agents. The present findings indicate that the regulation of Ser-2448 phosphorylation is more complex than originally proposed and that this phosphorylation event is not obligatorily linked to AKT. We have confirmed earlier findings that Ser-2448 phosphorylation is wortmannin-sensitive (18, 19). Although the wortmannin results are consistent with the earlier conclusion that AKT is an effector of Ser-2448 phosphorylation, an equally plausible conclusion is that Ser-2448 phosphorylation is sensitive to wortmannin because another PI-3 kinase-activated kinase, PDK-1 (32), is required for phosphorylation of the critical Thr-229 residue in the p70S6 kinase activation loop (33, 34). Thus, our results suggest that, at least in the cell types examined in this study, Ser-2448 phosphorylation primarily reflects a feedback signal to mTOR from its downstream target, p70S6 kinase. As mTOR activity is essential for the activation of p70S6 kinase, we conclude that Ser-2448 phosphorylation is a reasonable indicator of the level of mTOR signaling in cells or tissues.

View larger version:
Fig. 4.

siRNA-mediated knock-down of p70S6K reduces Ser-2448 phosphorylation. HeLa cells were transfected with either a control luciferase siRNA or a p70S6K-specific siRNA SMARTpool. Cells were harvested at 48 h posttransfection. Anti-mTOR immunoprecipitates were immunoblotted with α-mTORp2, and the same membrane was reprobed with α-mTOR antibodies. Whole cell extracts from the same samples were probed with α-p70S6 kinase, α-phospho-AKT (Ser-473), α-AKT, and α-phospholipase Cγ (PLCγ) antibodies. The data shown are representative of three independent experiments.

Several lines of evidence support the notion that p70S6 kinase is the dominant kinase responsible for mTOR phosphorylation at Ser-2448 in cells. Previous studies demonstrated that Ser-2448 phosphorylation was stimulated by amino acids, which regulate p70S6 kinase activity, but not AKT (35, 36) Furthermore, perturbation of the TSC/Rheb pathway, which regulates p70S6 kinase, but not AKT, influences mTOR Ser-2448 phosphorylation. For example, overexpression of TSC1 and TSC2 decreased Ser-2448 phosphorylation and conversely, RNA interference-mediated depletion of TSC2 increased Ser-2448 phosphorylation (15). Overexpression of Rheb stimulates phosphorylation of p70S6 kinase and results in elevated Ser-2448 phosphorylation (12). Treatment of cells with 2-deoxyglucose, which mimics glucose deprivation by inhibiting hexokinase, activates the TSC pathway via AMPK, decreases p70S6 kinase activity, and results in a concomitant decrease in mTOR Ser-2448 phosphorylation (Ref. 9 and data not shown). Furthermore, the depletion of AKT-1 or AKT-2 isoforms by retrovirus-mediated RNA-interference did not affect mTOR Ser-2448 phosphorylation in macrophage colony-stimulating factor-stimulated bone marrow macrophages (37). It should be noted that other examples of feedback regulation by p70S6 kinase exist. In Drosophila, it has been recently described that phosphorylation of the IRS proteins by p70S6 kinase leads to their degradation (38). Similarly, hyperactivation of the mammalian p70S6 kinase pathway via loss of TSC1 or TSC2 results in the inactivation and degradation of IRS proteins, and leads to insulin resistance (39).

Our results help to explain the earlier finding that a mTOR Ser-2448 → Ala mutant retains the ability to signal to p70S6 kinase (18). With AKT as the proposed effector of Ser-2448 phosphorylation, it was difficult to explain why loss of the Ser-2448 site failed to disrupt signaling through the AKT → mTOR → p70S6 kinase pathway. The current results indicate that Ser-2448 phosphorylation is a consequence, rather than a cause of p70S6 kinase activation. Nonetheless, the functional significance of Ser-2448 phosphorylation with regard to mTOR signaling remains elusive. When assayed in vitro, we did not observe any apparent differences in mTOR kinase activity with either mTOR Ser-2448 → Ala or Ser-2448 → Glu (a “phosphomimetic” substitution) mutants when compared with wild-type mTOR (data not shown). Furthermore, ectopic expression of either mTOR Ser-2448 → Ala or Ser-2448 → Glu mutants that also contained the rapamycin-resistant SI mutation fully supported mTOR-dependent cell size and proliferative activity in rapamycin-treated HEK 293 cells (data not shown). However, we note several limitations of this experimental approach. First, rapamycin is now known to inhibit only a subset of mTOR-dependent signaling functions (22, 40, 41). A second, related caveat is that ectopic expression of any catalytically active, rapamycin-resistant mTOR mutant in the cells confers rapamycin resistance to p70S6 kinase and reconstitutes the phosphorylation of Ser-2448 on the endogenous mTOR polypeptide (data not shown). A definitive evaluation of the functional consequences of Ser-2448 mutations will require that the mTOR Ser-2448 → Ala and Ser-2448 → Glu alleles be expressed in cells that completely lack endogenous mTOR.

It is now apparent that the mTOR regulation through phosphorylation of the `repressor domain' is more complex than originally thought. In addition to Ser-2448 phosphorylation catalyzed by p70S6 kinase and AKT, Thr-2446 is phosphorylated by AMPK (42). Phosphorylation of Thr-2446 and Ser-2448 appears to be mutually exclusive and the significance of AMPK phosphorylation within this region of mTOR remains to be determined (42).

In summary, we conclude that mTOR phosphorylation at Ser-2448 is not controlled directly by AKT but rather occurs in a feedback fashion catalyzed by the downstream target of mTOR, p70S6 kinase. The presence of two juxtaposed, inter-regulated phosphorylation sites in the mTOR “repressor domain” (18, 42) strongly hints that these modifications are functionally significant, although their roles in mTOR signaling remain elusive. Further studies are required to determine whether feedback phosphorylation of mTOR by p70S6 kinase amplifies or dampens mTOR signaling in nutrient- or growth factor-stimulated cells.

Rheb Activation Of Mtor

Acknowledgments

Activation Of Mtor Survival Pathway

We thank Dr. John Blenis and Dr. Naohiro Terada for supplying p70S6 kinase reagents. We also thank members of the Abraham laboratory for reagents and helpful discussions.

Footnotes

  • ↵1 The abbreviations used are: mTOR, mammalian target of rapamycin; PI, phosphatidylinositol; FKBP, FK506-binding protein; siRNA, small interfering RNA; HA, hemagglutinin; HEK, human embryonic kidney; DMEM, Dulbecco's modified essential medium; FCS, fetal calf serum; WT, wild type; SI, rapamycin-resistant; GST, glutathione S-transferase; AMPK, AMP-activated protein kinase; TSC, tuberous sclerosis complex.

  • ↵* This work was supported in part by Grants CA76193 and CA52995 (to R. T. A.) from the NCI, National Institutes of Health. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

  • Received February 15, 2005.
  • Revision received April 26, 2005.
  • The American Society for Biochemistry and Molecular Biology, Inc.

References

  1. ↵
  2. ↵
  3. ↵
  4. ↵
  5. ↵
  6. ↵
  7. ↵
  8. ↵
  9. ↵
  10. ↵
  11. ↵
  12. ↵
  13. ↵
  14. ↵
  15. ↵
  16. ↵
  17. ↵
  18. ↵
  19. ↵
  20. ↵
  21. ↵
  22. ↵
  23. ↵
  24. ↵
  25. ↵
  26. ↵
  27. ↵
  28. ↵
  29. ↵
  30. ↵
  31. ↵
  32. ↵
  33. ↵
  34. ↵
  35. ↵
  36. ↵
  37. ↵