|Home | About | Journals | Submit | Contact Us | Français|
The mammalian target of rapamycin complex 1 (mTORC1) functions as an environmental sensor to promote critical cellular processes such as protein synthesis, cell growth, and cell proliferation in response to growth factors and nutrients. While diverse stimuli regulate mTORC1 signaling, the direct molecular mechanisms by which mTORC1 senses and responds to these signals remain poorly defined. Here we investigated the role of mTOR phosphorylation in mTORC1 function. By employing mass spectrometry and phospho-specific antibodies, we demonstrated novel phosphorylation on S2159 and T2164 within the mTOR kinase domain. Mutational analysis of these phosphorylation sites indicates that dual S2159/T2164 phosphorylation cooperatively promotes mTORC1 signaling to S6K1 and 4EBP1. Mechanistically, S2159/T2164 phosphorylation modulates the mTOR-raptor and raptor-PRAS40 interactions and augments mTORC1-associated mTOR S2481 autophosphorylation. Moreover, mTOR S2159/T2164 phosphorylation promotes cell growth and cell cycle progression. We propose a model whereby mTOR kinase domain phosphorylation modulates the interaction of mTOR with regulatory partner proteins and augments intrinsic mTORC1 kinase activity to promote biochemical signaling, cell growth, and cell cycle progression.
Aberrant signaling by mTOR, the mammalian target of rapamycin, contributes to the pathogenesis of myriad human diseases (e.g., cancer, benign tumor syndromes, type II diabetes, and obesity) and pathophysiologic conditions (e.g., cardiac hypertrophy and coronary artery stent restenosis). Cellular mTOR regulation remains incompletely defined, however (13, 24, 31). mTOR senses and integrates signals from diverse environmental cues such as growth factors and hormones (i.e., insulin, insulin-like growth factor [IGF], and epidermal growth factor [EGF]), nutrients (i.e., amino acids and glucose), and cellular stresses (15, 22, 34, 53, 72). mTOR interacts with different partner proteins to form at least two functionally distinct signaling complexes, mTOR complex 1 (mTORC1) and mTOR complex 2 (mTORC2) (2, 4). Acute rapamycin treatment inhibits the intrinsic catalytic activity and signaling capacity of mTORC1, which contains mTOR, mLST8 (lethal with sec13 protein 8)/GβL (G-protein, β-subunit-like protein), raptor, PRAS40 (proline-rich Akt substrate of 40 kDa), and deptor (DEP domain protein that interacts with mTOR) (25, 27, 38, 39, 43, 52, 57, 62, 67). Acute rapamycin treatment fails to inhibit mTORC2, which contains shared and distinct partners (2, 4, 22).
At the cellular level, mTORC1 promotes cellular anabolic processes, including ribosome biogenesis, protein and lipid synthesis, cell growth (increase in cell mass and size), and cell cycle progression, which drives cell proliferation (17, 22, 42, 45). During growth factor and nutrient sufficiency, mTORC1 phosphorylates the translational regulators p70 ribosomal S6 kinase 1 (S6K1) and eukaryotic initiation factor 4E (eIF4E) binding protein 1 (4EBP1) to coordinately upregulate protein synthesis (35, 45). mTORC1-mediated phosphorylation of S6K1 aids the assembly of the eukaryotic initiation factor 3 (eIF3) translation initiation complex, while phosphorylation of the translational repressor 4EBP1 induces its release from eIF4E, allowing eIF4E to initiate cap-dependent translation (28, 45). Both S6K1 and 4EBP1 contain a TOR signaling (TOS) motif that mediates an essential interaction with the scaffolding protein raptor to facilitate the recruitment of substrates to the mTOR kinase (10, 49, 59, 60). mTORC1 also inhibits autophagy, a catabolic process, by phosphorylating and inactivating the autophagic proteins unc-51-like kinase 1/2 (ULK1/2) and the autophagy-specific gene 13 (ATG13) product (37).
An intensive research effort has focused on identifying the biochemical pathways and molecular mechanisms that link environmental cues to mTORC1 regulation. The mTORC1-inhibitory tuberous sclerosis complex (TSC), a heterodimer composed of Tsc1 (hamartin) and Tsc2 (tuberin) proteins, functions as a nexus of convergent signals that regulate mTORC1 (30, 41). Inactivation of either Tsc1 or Tsc2 leads to strong and constitutive mTORC1 signaling, which causes benign tumors to develop in diverse organ systems (30, 41). Tsc2 contains a GTPase-activating protein (GAP) domain that acts on Rheb (Ras homologue enriched in brain), a small GTP binding protein that activates mTORC1 through an incompletely defined mechanism, possibly involving enhanced substrate recruitment (3, 23, 58, 65). The current model suggests that insulin/phosphatidylinositol 3-kinase (PI3K) signaling promotes Akt-mediated phosphorylation of Tsc2, which suppresses the inhibitory effect of Tsc1/2 on mTORC1, thus activating Rheb (30, 32, 46, 64). Growth factor-mediated activation of mTORC1 absolutely requires sufficient levels of amino acids. A current model proposes that upon amino acid addition after factor deprivation, mTORC1 rapidly translocates from an ill-defined subcellular compartment to lysosomal membranes that contain Rheb in a manner dependent on the Rag GTPases (40, 55, 56).
Attention has focused more recently on the role of mTORC1 component phosphorylation in mTORC1 regulation. Insulin/PI3K signaling leads to Akt- and mTOR-mediated phosphorylation of PRAS40, which relieves the inhibitory effect of PRAS40 on mTORC1 (20, 50, 57, 67, 69). Insulin/PI3K signaling also increases mTOR S1261 and mTOR-mediated raptor S863 phosphorylation, events that promote mTORC1 function (1, 21, 71). In addition to phosphorylating PRAS40 and raptor, activated mTOR also phosphorylates deptor, leading to its degradation and thus relieving its mTOR-inhibitory action (52). Via a parallel pathway, Ras activation leads to mitogen-activated protein kinase (MAPK)- and p90 ribosomal protein S6 kinase (RSK)-mediated phosphorylation of Tsc2 (44, 54, 63) and raptor (5, 6), events that promote mTORC1 signaling. In response to energy deprivation, AMP-activated protein kinase (AMPK) phosphorylates both Tsc2 and raptor to suppress mTORC1 function (26, 33). Thus, diverse upstream signals converge on Tsc2, PRAS40, and raptor to positively and negatively modulate mTORC1 function.
To elucidate the molecular mechanisms underlying mTORC1 regulation, we have investigated the phosphorylation of mTOR itself. Here we identify S2159 and T2164 as novel mTOR phosphorylation sites that lie at the beginning of the mTOR kinase domain. Collectively, our data demonstrate that mTOR kinase domain phosphorylation modulates the interactions of mTOR with raptor and PRAS40 and leads to increased intrinsic mTORC1 kinase activity, which promotes biochemical signaling, cell growth, and cell cycle progression.
Reagents were obtained from the following sources. Protein A-Sepharose CL4B, protein G-Sepharose Fast Flow, and 7-methyl-GTP–Sepharose 4E were from GE Healthcare; 3-[(3-cholamidopropyl)-dimethylammonio]-1-propanesulfonate (CHAPS) was from Pierce; Immobilon-P polyvinylidene difluoride (PVDF) membrane (0.45 μm) was from Millipore; autoradiography film (HyBlot CL) was from Denville Scientific; reagents for enhanced chemiluminescence (ECL) were from Millipore (Immobilon Western chemiluminescent horseradish peroxidase [HRP] substrate); and all chemicals were from either Fisher Chemicals or Sigma.
AU1, Myc (9E10), and HA.11 antibodies were from Covance. Flag-M2 antibody was from Sigma. HRP-conjugated donkey anti-rabbit and sheep anti-mouse secondary antibodies were from GE Healthcare. P-S6K1-T389 (rabbit monoclonal 108D2) (no. 9234), S6 (no. 2217), P-mTOR-S2481 (no. 2974), P-4EBP1-S65 (no. 9451), 4EBP1 (no. 9452), eIF4E (no. 9742), and glutathione S-transferase (GST) (no. 2622) antibodies were from Cell Signaling Technology. PRAS40 antibodies (05-998) were from Upstate.
Affinity-purified antipeptide antibodies to mTOR (amino acids 221 to 237; rat), P-mTOR-S1261 (amino acids 1256 to 1266; rat), S6K1 (C-terminal amino acids 485 to 502 of the 70-kDa isoform; rat), and P-S6 (amino acids 232 to 249) were generated as described previously (1). Phospho-specific antibodies against mTOR peptides phosphorylated at S2159 (catalog no. ABS79) or T2164 (catalog no. ABS88) were generated by EMD Millipore (serum-derived antibodies were used).
The pcDNA3/AU1-mTOR wild-type (WT), rapamycin-resistant (RR) (S2035I), kinase-dead (KD) (D2338A), and RR/KD (S2035I, D2338A) plasmids were generously shared by R. Abraham (Wyeth, Pearl River, NY); the pRK5/Myc-raptor plasmid was shared by D. Sabatini (MIT, Boston, MA); pRK5/Myc-mTOR, pRK5/Myc-mTOR-KD, and pRK5/HA-raptor were obtained from D. Sabatini via Addgene (no. 1861, 8482, and 8513, respectively); the pRK7/HA-S6K1, pKH3/HA-mLST8/GβL, and pRK7/Flag-Rheb plasmids were from J. Blenis (Harvard Medical School, Boston, MA); and pACTAG2/3HA-4EBP1 was from N. Sonenberg (McGill University, Montreal, Canada).
HEK293 cells (20 15-cm plates per condition; ~15 × 106 cells/15-cm plate) were untransfected or transiently transfected with AU1-tagged mTOR (25 μg) and cultured in Dulbecco's modified Eagle's medium (DMEM) with 10% fetal bovine serum (FBS). AU1-mTOR was immunoprecipitated overnight, resolved by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), stained with Coomassie blue R-250, excised from the gel, digested with trypsin, and analyzed by tandem mass spectrometry (MS/MS). Immunoisolated AU1-mTOR was prepared for analysis by MS/MS as described previously (1).
Site-directed mutagenesis was performed using QuikChange II XL (Stratagene), and mutated plasmids were fully sequenced. The following mutations were introduced in wild-type pcDNA3/AU1-mTOR: S2159D/T2164E (DE) and S2159A/T2164A (AA). The following mutations were introduced in the rapamycin-resistant (RR) backbone (S2035I) of plasmid pcDNA3/AU1-mTOR to generate the following plasmids: RR-S2159D/T2164E (RR/DE), RR-S2159A/T2164A (RR/AA), RR-S2159D, RR-S2159A, RR-T2164E, and RR-T2164A. A second rapamycin-resistant mutation, here designated RR-2 (S2035W), was introduced in pcDNA3/AU1-mTOR carrying WT or mutant mTOR alleles. The following mutations were introduced in pRK5/Myc-mTOR (wild-type backbone): S2159A, T2164A, S2159A/T2164A (AA), and S2159D/T2164E (DE). The following oligonucleotide primers were used to create point mutations in the rat mTOR cDNA (accession no. L37085) (capital letters indicate mismatches, and the three underlined nucleotides represent the codon mutated): for S2035I (RR), primer 1, 5′-ggcctagaagaggccATtcgcttgtactttggg-3′, and primer 2, 5′-cccaaagtacaagcgaATggcctcttctaggcc-3′; for S2035W (RR-2), primer 1, 5′-ggcctagaagaggcctGGcgcttgtactt-3′, and primer 2, 5′-aagtacaagcgCCaggcctcttctaggcc-3′; for S2159A, primer 1, 5′-ccatagccccgGctttgcaagtcatc-3′, and primer 2, 5′-gatgacttgcaaagCcggggctatgg-3′; for S2159D, primer 1, 5′-ccatagccccgGAtttgcaagtcatc-3′, and primer 2, 5′-gatgacttgcaaaTCcggggctatgg-3′; for T2164A, primer 1, 5′-gcaagtcatcGcatccaagcagaggcc-3′, and primer 2, 5′-ggcctctgcttggatgCgatgacttgc-3′; and for T2164E, primer 1, 5′-gcaagtcatcGAatccaagcagaggcc-3′, and primer 2, 5′-ggcctctgcttggatTCgatgacttgc-3′.
HEK293 cells were cultured in DMEM that contained high glucose (4.5 g/liter), glutamine (584 mg/liter), and sodium pyruvate (110 mg/liter) concentrations (Gibco/Invitrogen), supplemented with 10% fetal bovine serum (FBS) (HyClone), and were incubated at 37°C in a humidified atmosphere containing 5% CO2. Cells were serum deprived via incubation in DMEM supplemented with 20 mM HEPES (pH 7.2) for ~20 h. Insulin (100 nM) (Invitrogen) was added to serum-deprived cells and left for 30 min. Where indicated, serum-deprived cells were pretreated with rapamycin (20 ng/ml) (Calbiochem) for 30 min prior to the addition of insulin. For drug treatments under steady-state conditions (cycling in DMEM-FBS), cells were incubated in rapamycin (20 ng/ml) for 2 h, staurosporine (1 μM) (Calbiochem) for 3 h, or Torin1 (50 nM) (a kind gift from S. Sabatini) for 3 h. For amino acid deprivation, HEK293 cells were incubated in Dulbecco's phosphate-buffered saline (PBS) containing d-glucose (1 g/liter), sodium pyruvate (36 mg/liter), and 10% dialyzed FBS for 60 min. Unless indicated otherwise in the figure legends, HEK293 cells on 60-mm plates were transfected according to the manufacturer's directions using TransIT-LT1 (Mirus) and a total of 5 μg of DNA per plate. The specific amounts of experimental plasmid transfected are stated in the figure legends. Cells were lysed at ~24 to 48 h posttransfection.
Unless indicated otherwise, cells were washed twice with ice-cold PBS and scraped into ice-cold lysis buffer A containing NP-40 (0.5%) and Brij 35 (0.1%), as described previously (1). For Fig. 4, cells were lysed in buffer A containing CHAPS (0.3%). For Fig. 1 and in the mTOR-raptor coimmunoprecipitation experiment (see Fig. 5A), cells were lysed in buffer B (40 mM HEPES [pH 7.4], 120 mM NaCl, 1 mM EDTA, 10 mM pyrophosphate, 10 mM β-glycerophosphate, 50 mM NaF, 1 mM Na3VO4, 10 μg/ml leupeptin, 5 μg/ml pepstatin A, 40 μg/ml phenylmethylsulfonyl fluoride [PMSF]) containing CHAPS (0.3%), as originally described (38) (to maintain the mTOR-raptor interaction and to observe its prominent destabilization upon insulin stimulation). Lysates were spun at 13,200 rpm for 5 min at 4°C, and the postnuclear supernatants were collected. The Bradford assay was used to normalize protein levels for immunoprecipitation and immunoblot analysis. For immunoprecipitation, whole-cell lysate (WCL) was incubated with antibodies for ~2 h or overnight at 4°C, incubated with protein G-Sepharose beads for 1 h, washed three times in lysis buffer, and resuspended in 1× sample buffer, as described previously (1). For cap pulldown assays, WCL was incubated with m7GTP-Sepharose beads for 2 h, washed three times in lysis buffer, and resuspended in 1× sample buffer. Samples were resolved by SDS-PAGE and transferred to PVDF membranes by using Towbin transfer buffer, as described previously (1). Immunoblotting was performed by blocking PVDF membranes in Tris-buffered saline containing Tween 20 (TBST) and 3% nonfat milk and incubating the membranes in TBST with 2% bovine serum albumin (BSA) containing primary antibodies or secondary HRP-conjugated antibodies, as described previously (1). Blots were developed by ECL and detected with autoradiographic film or digitally with a Chemi-Doc-It HR 410 system (UVP).
The Invitrogen Flp-In system was used to generate stable HEK293 cell lines that express various AU1-mTOR alleles. This system utilizes a genetically engineered HEK293 cell line (HEK293/FRT) that possesses an FLP recombination target (FRT) integrated into a transcriptionally active region of the genome; thus, after drug selection, each clone expresses AU1-mTOR at equivalent levels, thus allowing one to simply pool colonies for analysis. After subcloning of AU1-mTOR alleles into pcDNA5/FRT, HEK293/FRT cells (maintained in DMEM–10% FBS plus Zeocin [100 μg/ml]) were cotransfected with pcDNA5/FRT/AU1-mTOR (0.2 μg) and pOG44 (2 μg) (recombinase). At 3 days posttransfection, stable integrants were selected in DMEM–10% FBS containing hygromycin (200 μg/ml). (Note that the Invitrogen HEK293/FRT cell line is not as insulin sensitive as the HEK293 cell line employed throughout this work [originally from the lab of J. Blenis].)
Flp-In HEK293 cells that stably express vector control or AU1-mTOR alleles were cultured in DMEM-FBS-hygromycin (100 μg/ml) in the absence or presence of rapamycin (20 ng/ml) and/or hydroxyurea (0.5 mM) for 96 h. For cell size and DNA content analysis, cells were harvested with PBS-EDTA (0.1%), fixed in 80% ethanol, and stained with propidium iodide (PI)-RNase A solution, as described previously (1). For cell size analysis, the mean forward scatter height (FSC-H) of ~3,000 single, unclumped G1-phase cells was determined by gating on PI fluorescence using a BD Biosciences FACSCalibur with CellQuest software. For cell cycle analysis, the PI fluorescence of ~10,000 single, unclumped cells was analyzed. Markers for each cell cycle phase (G0/G1, S, and G2/M) were set to quantitate percent G0/G1, S, and G2/M using CellQuest software.
High-titer lentiviruses encoding scrambled or TOR_2 short hairpin RNAs (shRNAs) were generated by the University of Michigan Vector Core. These LKO-based plasmids were originally from the D. Sabatini lab but were obtained from Addgene (catalogue no. 1856 and 1864, respectively). HEK293 cells were infected with lentiviral supernatant in the presence of protamine sulfate (6 μg/ml) for 24 h and then selected in puromycin (2.5 μg/ml) for 5 days.
For some figures, irrelevant lanes were removed from a scanned autoradiograph and flanking lanes juxtaposed using Adobe Photoshop. The presence of a thin, vertical black line indicates such a modification.
To elucidate the biochemical mechanisms that directly control mTORC1 signaling, we have investigated a role for site-specific mTOR phosphorylation in regulation of mTORC1 function. We employed liquid chromatography-tandem mass spectrometry (LC-MS/MS) and identified two novel in vivo phosphorylation sites (P sites) on AU1-tagged mTOR immunoprecipitated from transiently transfected HEK293 cells cultured under steady-state conditions (DMEM-FBS). MS/MS analysis identified serine 1261 (S1261) as an unambiguous site of mTOR phosphorylation, as reported previously (1). Such analysis also revealed a doubly phosphorylated peptide with a potential match for mTOR phosphorylation on serine 2159 (S2159) and threonine 2164 (T2164). Unfortunately, we were unable to unambiguously confirm this potential match, possibly due to low stoichiometry of phosphorylation on these sites (spectra not shown).
To gain further evidence for mTOR phosphorylation on S2159 and T2164, we generated rabbit polyclonal antibodies against mTOR peptides phosphorylated on either S2159 or T2164. To test the site specificity of these antibodies and confirm phosphorylation in intact cells, we generated phosphorylation site-defective Ala substitution mutants with mutations at S2159 and T2164. We transiently transfected HEK293 cells with a vector control or wild-type (WT), S2159A, T2164A, or double S2159A/T2164A (AA) Myc-mTOR. Myc-mTOR was then immunoprecipitated with Myc antibodies and immunoblotted with P-S2159 and P-T2164 antibodies. P-S2159 antibodies detected WT-mTOR but not S2159A- or AA-mTOR. P-T2164 antibodies, on the other hand, detected WT-mTOR but not T2164A- or AA-mTOR, thus confirming the site specificity of each antibody (Fig. 1A). Additionally, these data show that mTOR S2159 and T2164 phosphorylation events do not modulate each other, as P-S2159 antibodies detect phosphorylation on the T2164A-mTOR mutant and vice versa. To facilitate our analysis of mTOR phosphorylation, we employed the Invitrogen Flp-In system to generate stable HEK293 cell lines that express vector control or various AU1-mTOR alleles (WT, AA, or kinase dead [KD]) at equivalent levels. As shown in Fig. 1B, both P-S2159 and P-T2164 antibodies detected WT- but not S2159A/T2164A (AA)-AU1-mTOR, consistent with the transient-transfection analysis shown in Fig. 1A. To test whether S2159 and T2164 represents sites of autophosphorylation, similar to S2481, we analyzed phosphorylation on WT- versus KD-mTOR as well as phosphorylation on WT-mTOR isolated from cells treated with the ATP-competitive mTOR catalytic inhibitor Torin1. As expected, S2481 phosphorylation on KD-mTOR was abrogated, and Torin1 strongly reduced P-S2481 on WT-mTOR; mTOR S2159 and T2164 phosphorylation was normal on KD-mTOR and unaffected by Torin1, however, thus indicating that neither S2159 nor T2164 represents a site of mTOR phosphorylation (Fig. 1B). Phosphorylation on mTOR S2159 but not T2164 was reduced by the wide-spectrum kinase inhibitor staurosporine, indicating that two different kinases mediate these phosphorylation events (Fig. 1B). Additionally, in a subset of experiments, stable expression of AA-mTOR reduced phosphorylation of S6K1-T389 and the ribosomal protein S6 (rpS6), an S6K1 substrate, suggesting a dominant negative effect on mTORC1 function (Fig. 1B). Importantly, we confirmed S2159 and T2164 phosphorylation on endogenous mTOR immunoprecipitated from HEK293 cells (Fig. 1C). As seen with exogenously expressed mTOR, P-S2159 on endogenous mTOR was sensitive to staurosporine whereas P-T2164 was not (Fig. 1C). Lastly, in our HEK293 cell system, well-described mTORC1-regulating signals (e.g., serum, insulin, amino acids, glucose, and Rheb overexpression) did not modulate phosphorylation on mTOR S2159 or T2164 (data not shown), suggesting either that these phosphorylation events are constitutive or that we have uncovered a novel mTOR regulatory paradigm. S2159 and T2164 lie at the beginning of the C-terminal mTOR kinase domain and show conservation down to vertebrates, with T2164 showing conservation down to flies, worms, plants, and yeasts (Fig. 1D).
To investigate a role for mTOR S2159/T2164 phosphorylation in regulation of mTORC1 signaling, we compared the abilities of wild-type (WT) and mutant mTOR alleles to mediate the phosphorylation of S6K1 and 4EBP1, two well-characterized mTORC1 substrates. We introduced phospho-mimetic (S2159D/T2164E) and phosphorylation site-defective (S2159A/T2164A) substitutions at both S2159 and T2164 into the rapamycin-resistant (RR) mTOR backbone (S2035I) (7), which allows the signaling capacity of mTORC1 containing exogenously expressed mutant mTOR alleles to be studied in the absence of endogenous mTORC1 action upon pretreatment of cells with rapamycin to chemically knock out mTORC1 signaling. Point mutation of S2035 within the FKBP12/rapamycin binding (FRB) domain confers rapamycin resistance by abrogating the interaction of mTOR with the FKBP12/rapamycin complex (7). Upon transfection, we compared the abilities of AU1-mTOR WT and mutant alleles in the RR backbone to mediate the phosphorylation of cotransfected hemagglutinin (HA)-S6K1 on T389 in the presence of rapamycin.
Rapamycin completely inhibited the insulin-stimulated phosphorylation of HA-S6K1 in cells that coexpressed WT-mTOR, and expression of RR-mTOR, but not RR/kinase-dead (RR/KD)-mTOR, conferred rescue of HA-S6K1 phosphorylation (Fig. 2A), as expected. Expression of RR-S2159D/T2164E (RR/DE)-mTOR mediated stronger phosphorylation of HA-S6K1 than expression of RR- or RR-S2159A/T2164A (RR/AA)-mTOR in both the presence and absence of serum growth factors; expression of RR/AA-mTOR mediated weaker phosphorylation of HA-S6K1 than RR-mTOR. Similar results were observed under steady-state conditions in which HEK293 cells cycled asynchronously in DMEM-FBS (Fig. 2B). We also generated an alternate rapamycin-resistant allele of mTOR (S2035W; here designated RR-2) that signals to S6K1 in the presence of rapamycin more strongly than the originally described allele, S2035I (47). Similarly, phospho-mimetic DE substitutions in the RR-2 background promoted mTORC1 signaling to S6K1 more strongly than phospho-defective AA substitutions, and phospho-defective AA substitutions mediated weaker mTORC1 signaling relative to RR-mTOR (Fig. 2C). These data indicate that mTOR S2159/T2164 phosphorylation in the kinase domain promotes mTORC1-mediated signaling to S6K1.
As an alternative approach to transfecting RR-mTOR alleles in conjunction with rapamycin treatment, we performed an mTOR knockdown/rescue experiment whereby lentivirally delivered shRNA reduced the endogenous expression of human mTOR in HEK293 cells but did not target exogenously expressed rat Myc-mTOR. After knockdown, Myc-mTOR WT and mutant alleles in a wild-type FRB domain backbone (not rapamycin resistant) were transfected, and mTORC1 signaling capacity was assayed. As expected, WT-mTOR rescued the phosphorylation of rpS6, whereas KD-mTOR did not (Fig. 2D). Consistent with our data obtained by utilizing RR-mTOR alleles (Fig. 2A, B, and C), the DE-mTOR mutant mediated stronger phosphorylation of rpS6 than WT-mTOR, while the AA-mTOR mutant mediated weaker phosphorylation of rpS6 than DE-mTOR. These data further confirm that phosphorylation of mTOR S2159/T2164 enhances mTORC1 signaling toward S6K1 and rpS6.
mTORC1 signaling uniquely requires sufficient levels of amino acids, even during growth factor abundance. We thus tested whether mTOR S2159/T2164 phosphorylation modulates the mTORC1 response to acute amino acid stimulation. HEK293 cells were cotransfected with HA-S6K1 and various RR-AU1-mTOR alleles, amino acid deprived for 60 min in the presence of rapamycin, and then stimulated with amino acid-containing medium containing rapamycin for 30 min. Similar to results under insulin-stimulated conditions, RR/DE-mTOR signaled to HA-S6K1 more strongly than RR and RR/AA-mTOR in response to acute amino acid stimulation (Fig. 2E). Consistently, RR/AA-mTOR signaled to S6K1 in a defective manner relative to RR- and RR/DE-mTOR. Unlike under growth factor starvation conditions, however, the RR/DE mutant failed to increase basal HA-S6K1 phosphorylation in the absence of amino acids, even upon prolonged exposure of the Western blot signal (data not shown). These data indicate that mTOR S2159/T2164 phosphorylation enhances mTORC1 signaling only under conditions of amino acid abundance.
We next tested the abilities of transfected RR-, phospho-defective RR/AA-, and phospho-mimetic RR/DE-mTOR alleles to mediate the phosphorylation of S6K1 in response to various concentrations of insulin. Consistent with a gain-of-function phenotype, RR/DE-mTOR reached maximal signaling capacity at a submaximal insulin dose (5 nM) whereas RR-mTOR exhibited the expected submaximal signaling at this dose (Fig. 3A). Importantly, RR/AA-mTOR exhibited defective signaling at 100 nM insulin relative to RR-mTOR. We also tested the responses of the RR-, RR/AA-, and RR/DE-mTOR alleles to insulin treatment for 0 to 240 min. RR/DE-mTOR signaled to S6K1 more strongly than RR- and RR/AA-mTOR at early time points (Fig. 3B). Additionally, while mTORC1 signaling to S6K1 decreased at 240 min after insulin stimulation, the phosphorylation of S6K1 in cells expressing RR/DE-mTOR was maintained at higher levels at this time point. Consistently, RR/AA-mTOR signaled in a defective manner at 30 min of insulin stimulation relative to RR-mTOR. These data indicate that constitutive phosphorylation on mTOR S2159/T2164 accelerates the time course for insulin-stimulated phosphorylation of S6K1 and additionally renders TORC1 more resistant to downregulation at later time points.
We next examined the relative contributions of the individual S2159D and T2164E substitutions to the gain-of-function phenotype conferred by the double phospho-mimetic S2159D/T2164E mTOR mutant. As shown in Fig. 3C, neither RR/S2159D nor RR/T2164E enhanced the insulin-stimulated phosphorylation of HA-S6K1 as strongly as the RR/DE double mutant (Fig. 3C). Additionally, we tested the contributions of the individual S2159A and T2164A substitutions to the defective signaling conferred by the double S2159A/T2164A mTOR allele. As shown in Fig. 3D, single RR/S2159A- and RR/T2164A-mTOR mutants mediated phosphorylation of HA-S6K1 more strongly than the double RR/AA-mTOR mutant (Fig. 3D). Overall, these data indicate that phosphorylation events on S2159 and T2164 cooperate to regulate mTORC1 signaling to S6K1.
Recent work indicates that while rapamycin partially inhibits mTORC1-mediated phosphorylation of 4EBP1, novel ATP-competitive mTOR kinase inhibitors completely block 4EBP1 phosphorylation, indicating that not all mTORC1 substrates are equally sensitive to rapamycin (11, 12, 16, 66). Thus, rather than employing our RR-mTOR assay in conjunction with rapamycin treatment to study the role of mTOR S2159/T2164 phosphorylation in mTORC1 signaling to 4EBP1, we utilized AU1-mTOR alleles in a wild-type backbone in the absence of rapamycin treatment. We reasoned that such an approach would be feasible if expression of DE-mTOR was sufficient to augment 4EBP1 phosphorylation over that mediated by endogenous mTORC1 or if AA-mTOR was to dominantly inhibit mTORC1 function. We thus cotransfected HEK293 cells with various AU1-mTOR alleles together with the mTOR partner Myc-raptor and HA-4EBP1, followed by serum deprivation and acute insulin stimulation. To more sensitively measure 4EBP1 phosphorylation, we utilized a cap pulldown assay. In this assay, cellular eIF4E binds m7GTP, a chemical moiety that mimics the cap structure found at the 5′ ends of mRNAs, which is coupled to Sepharose beads. In the absence of insulin, hypophosphorylated 4EBP1 binds strongly to eIF4E and thus binds to beads; insulin stimulation, and thus 4EBP1 phosphorylation, induces the dissociation of 4EBP1 from eIF4E. As expected, insulin stimulation of WT-mTOR-transfected cells reduced 4EBP1-eIF4E association and increased 4EBP1 S65 phosphorylation (Fig. 4). In cells transfected with the active DE-mTOR allele, less 4EBP1 associated with eIF4E in the absence of serum growth factors than did so in WT- and AA-mTOR-transfected cells, indicating greater 4EBP1 phosphorylation. In cells transfected with AA-mTOR, however, more 4EBP1 associated with eIF4E in the absence or presence of insulin than did so in WT- and DE-mTOR-transfected cells, indicating reduced 4EBP1 phosphorylation. Additionally, AA-mTOR-expressing cells mediated weaker phosphorylation of 4EBP1S65 than RR- or RR/DE-mTOR expressing cells. Notably, in the absence of Myc-raptor cotransfection with AU1-mTOR, we do not observe the phenotypes described above (data not shown). Myc-raptor coexpression likely facilitates recruitment of 4EBP1 to mTOR, thus increasing the efficiency of HA-4EBP1 phosphorylation. These data suggest that mTOR S2159/T2164 phosphorylation mimics insulin stimulation to promote the phosphorylation of 4EBP1 and that overexpression of AA-mTOR functions dominantly to suppress the phosphorylation of 4EBP1 mediated by endogenous mTORC1.
Nutrients (e.g., amino acids and glucose) weaken the mTOR-raptor interaction, as measured by coimmunoprecipitation (38, 39), suggesting a conformational change in mTORC1 structure that correlates with active mTORC1. Similarly, we observe that insulin stimulation also weakens the mTOR-raptor interaction (1) (Fig. 5A). Insulin stimulation weakens the inhibitory raptor-PRAS40 interaction via both Akt- and mTORC1-mediated phosphorylation of PRAS40 (20, 50, 57, 67, 69, 70). We thus tested whether mTOR S2159/T2164 phosphorylation modulates the mTOR-raptor, mTOR-GβL (also known as mLST8), or raptor-PRAS40 interactions. To examine the mTOR-raptor and mTOR-GβL interactions, HEK293 cells were cotransfected with various wild-type backbone Myc-mTOR alleles together with HA-raptor and HA-GβL and lysed in a buffer that enables ready observation of the insulin-induced destabilization of mTOR and raptor. After immunoprecipitation of Myc-mTOR, we found that insulin destabilized the interaction of wild-type Myc-mTOR with HA-raptor, as expected (Fig. 5A). Strikingly, the interaction of Myc-DE-mTOR with HA-raptor was weak in the absence of serum growth factors, thus mimicking insulin stimulation; the interaction of Myc-AA-mTOR with HA-raptor was not appreciably strengthened (Fig. 5A). Importantly, the interaction of HA-GβL with Myc-mTOR-DE or -AA was unaltered, indicating that these amino acid substitutions do not adversely affect overall mTOR structure (Fig. 5A). These data additionally indicate that mutation of the kinase domain at S2159/T2164 does not alter the kinase domain structure, as GβL, which binds the mTOR kinase domain, interacts normally (39).
To examine the raptor-PRAS40 interaction, we cotransfected HEK293 cells with various AU1-mTOR alleles together with Myc-raptor and examined the levels of endogenous PRAS40 in Myc-raptor immunoprecipitates. As shown in Fig. 5B, insulin stimulation weakened the raptor-PRAS40 interaction, as expected (57, 67, 70). In cells expressing DE-mTOR, less PRAS40 associated with Myc-raptor than in cells expressing WT- and AA-mTOR under growth factor-deprived conditions. In cells expressing AA-mTOR, however, the raptor-PRAS40 interaction was stronger under both conditions. Our finding that phospho-mimetic substitutions at mTOR S2159/T2164 weaken while phospho-defective substitutions strengthen the inhibitory raptor-PRAS40 interaction provides an additional mechanism to explain the enhanced signaling capacity of DE-mTOR and decreased signaling capacity of AA-mTOR toward S6K1 and 4EBP1. By modulating inhibitory interactions of mTOR with mTORC1 components, mTOR S2159/T2164 phosphorylation shifts mTORC1 toward a more active conformation.
We have shown previously that insulin stimulation or Rheb overexpression promotes mTOR S2481 autophosphorylation within mTORC1, which monitors mTORC1 intrinsic catalytic activity in intact cells (62). Additionally, mTOR S1261 phosphorylation is required, at least in part, for the ability of Rheb to promote mTORC1-associated mTOR S2481 autophosphorylation (1). To investigate a role for mTOR S2159/T2164 phosphorylation in control of mTORC1 intrinsic kinase activity, we cotransfected HEK293 cells with various Myc-mTOR alleles (WT backbone) and HA-raptor in the absence or presence of Flag-Rheb. HA-raptor was immunoprecipitated to detect the autophosphorylation of mTORC1-associated mTOR S2481. As shown in Fig. 5C, mTORC1 that contained AA-mTOR displayed reduced S2481 autophosphorylation relative to mTORC1 that contained WT-mTOR, similar to S1261A-mTOR. Additionally, this experiment confirmed that mTOR S2159/T2164 phosphorylation does not affect mTOR S1261 phosphorylation (Fig. 5C), consistent with Fig. 1B. Taken together, these data reveal that mTOR S2159/T2164 phosphorylation as well as S1261 phosphorylation contributes to the activation of intrinsic mTORC1 catalytic activity.
To facilitate analysis of mTOR S2159/T2164 phosphorylation in mTORC1-mediated cell growth and cell cycle progression, critical cellular functions of mTORC1, we employed stable HEK293 cells generated via the Flp-In system that express AU1-mTOR-WT, -RR, -RR/DE, -RR/AA, or -RR/KD. Upon chemical knockout of endogenous mTORC1 signaling with rapamycin, we found that RR/DE-mTOR mediated stronger mTORC1 signaling than RR- and RR/AA-mTOR under both basal and insulin-stimulated conditions, as assayed by the phosphorylation of endogenous S6K1, rpS6, and 4EBP1 (Fig. 6A), similar to results from transient-transfection experiments. Additionally, RR/AA-mTOR signaled to S6K1, rpS6, and 4EBP1 in a defective manner relative to RR- and RR/DE-mTOR (Fig. 6A).
We next used these stable cell lines to investigate a role for mTOR S2159/T2164 phosphorylation in control of cell growth and cell cycle progression. To assay cell growth, we cultured these stable cell lines in complete medium in the absence or presence of rapamycin for 4 days and used flow cytometry to measure the relative cell size of G1-phase cells by employing the parameter mean forward scatter height (FSC-H). As expected, rapamycin treatment of WT-mTOR-expressing cells decreased the mean FSC-H, while RR-mTOR, but not RR/KD-mTOR, rescued the rapamycin-mediated decrease in cell size, in agreement with earlier work (1, 19) (Fig. 6B). Consistent with the gain-of-function phenotype in our biochemical assays, RR/DE-mTOR-expressing cells exhibited a modest but statistically significant increase in cell size (Fig. 6B). It is important to note that when these stable cell lines were subjected to culture conditions identical to those employed for cell size analysis for Fig. 6B and were analyzed for biochemical signaling, RR/DE-mTOR mediated stronger phosphorylation of rpS6 than RR-mTOR, while RR/AA-mTOR mediated weaker phosphorylation of rpS6 than RR-mTOR, consistent with earlier results (Fig. 6C). Unexpectedly, RR/AA-mTOR-expressing cells also displayed an increased cell size (Fig. 6B).
To understand how both RR/DE- and RR/AA-mTOR increase cell size, we determined how expression of AA- and DE-mTOR affects cell cycle progression relative to RR-mTOR, as perturbation of cell cycle progression can result in indirect effects on cell size. We reasoned that if RR/AA-mTOR-expressing cells were to spend an increased time in G1 phase but continued to grow, even at a reduced rate, they would display an increased cell size. To test this idea, stable cells were serum deprived for 24 h to induce G1 phase accumulation and then stimulated with serum containing medium (DMEM-FBS) for 24 h to promote cell cycle progression in the absence or presence of rapamycin. The percentage of cells in G1 phase was then determined on a flow cytometer by measuring DNA content after propidium iodide staining. As expected, WT-mTOR-expressing cells stimulated with serum in the presence of rapamycin displayed increased G1-phase content relative to cells stimulated in the absence of rapamycin, and expression of RR-mTOR completely rescued this G1-phase delay (Fig. 6D). Expression of RR/AA-mTOR, however, led to impaired rescue, while expression of RR/DE-mTOR conferred rescue (Fig. 6D). Thus, RR/AA-mTOR-expressing cells progress through G1 phase more slowly, thus having more time to grow, while RR/DE-mTOR-expressing cells progress through G1 phase normally with an augmented rate of cell growth; in the end, both situations lead to increased cell size.
To confirm the idea that cells expressing RR/AA-mTOR display a larger cell size due to delayed cell cycle progression, we repeated our cell size analysis with cells experiencing a cell cycle block. First, we treated WT-mTOR-expressing cells in the absence or presence of the DNA synthesis inhibitor 2-hydroxyurea (2-HU) for 4 days. 2-HU strongly reduced cell proliferation (data not shown) and perturbed cell cycle kinetics, causing significant accumulation in S phase (Fig. 7A). 2-HU also increased cell size as expected, and importantly, this increased cell size occurred in an mTORC1-dependent manner, as rapamycin blocked the 2-HU-induced cell size increase (Fig. 7B). We next analyzed cell size in the stable lines expressing AU1-mTOR-WT, -RR, -RR/DE, -RR/AA, and -RR/KD after 4 days of culture with rapamycin and 2-HU. We found that when cell cycle progression was eliminated as a complicating factor by using 2-HU, mTORC1 that contained mTOR-RR/DE promoted cell growth to a larger cell size while mTOR-RR/AA did not (Fig. 7C). Importantly, 2-HU caused similar S-phase accumulation in the mTOR-RR, -RR/DE, and -RR/AA cell lines (Fig. 7D). Thus, our original conclusion based on the cell size data in Fig. 6B was correct: AA-mTOR-expressing cells grow to a larger cell size than control cells under cycling conditions due to a primary defect in cell cycle progression, providing them more time to increase in mass and size prior to cell division.
Taken together, these data indicate that mTOR S2159/T2164 phosphorylation promotes both cell growth and cell cycle progression: constitutive phosphorylation on these sites promotes cell growth without appreciably accelerating cell cycle progression, while a lack of phosphorylation blunts cell cycle progression without appreciably reducing cell growth; both situations ultimately increase cell size. Moreover, they suggest the intriguing notion that mTOR S2159/T2164 phosphorylation differentially promotes cell growth and cell cycle progression in a manner dependent on relative stoichiometry of phosphorylation.
To elucidate molecular mechanisms underlying mTORC1 regulation, we have investigated the role of site-specific mTOR phosphorylation in mTORC1 function. Among the mTOR phosphorylation sites characterized in the literature thus far (i.e., S2448, S2481, T2446, and S1261), only mTOR S1261 phosphorylation has been reported to regulate mTORC1 signaling (1, 8, 9, 29, 48, 51, 61). Here we identified and characterized two novel mTOR phosphorylation sites, S2159 and T2164, that localize to the N terminus of the mTOR kinase domain. Mutational analysis demonstrates that mTOR S2159/T2164 phosphorylation promotes mTORC1-mediated signaling to S6K1 and 4EBP1, similar to mTOR S1261 phosphorylation (1). Mechanistically, mTOR S2159/T2164 phosphorylation modulates the mTOR-raptor interaction and weakens the inhibitory raptor-PRAS40 interaction. These conformational changes lead to increased mTORC1 intrinsic kinase activity, as monitored by mTOR S2481 autophosphorylation. Moving downstream, mTOR S2159/T2164 phosphorylation promotes both cell growth and cell cycle progression (Fig. 8). The rather weak signaling phenotypes conferred by phospho-mimetic and phosphorylation site-defective substitutions at mTOR S2159/T2164 suggest that these phosphorylation events modulate mTORC1 signaling rather than effect major regulatory function.
Tandem mass spectrometry suggested but did not unambiguously confirm dual mTOR phosphorylation on S2159/T2164 in intact cells. The generation of phospho-specific antibodies for P-S2159 and P-T2164 enabled us to confirm site-specific mTOR phosphorylation on S2159 and T2164 in intact cells. Using these antibodies, our data indicate that neither S2159 nor T2164 represents a site of mTOR autophosphorylation; moreover, two different kinases mediate mTOR S2159 versus T2164 phosphorylation: A staurosporine-sensitive kinase mediates P-S2159, while a staurosporine-insensitive kinase mediates P-T2164. In our HEK293 cell system, canonical mTORC1-regulating signals (e.g., insulin, amino acids, and glucose) did not modulate either phosphorylation event (data not shown), suggesting either that these events are constitutive or that we have uncovered a novel mTOR regulatory paradigm in which a currently unknown cellular signal regulates mTOR S2159 and T2164 phosphorylation. The rather weak affinity of the P-S2159 and P-T2164 antibodies for mTOR isolated from intact cells, coupled with our weak mass spectrometry data, suggest low stoichiometry of phosphorylation under steady-state conditions. That phospho-mimetic DE-mTOR mediates stronger substrate phosphorylation than wild-type mTOR supports this hypothesis. Low stoichiometry of phosphorylation may be explained by several potential reasons. As suggested above, an unknown cellular signal may regulate phosphorylation of these sites. Alternatively, mTOR S2159/T2164 phosphorylation may occur primarily in a specific subcellular compartment, on a fraction of total mTOR.
Insulin and nutrients (e.g., amino acids and glucose) weaken the mTOR-raptor interaction via an unknown mechanism(s), which correlates with active mTORC1 signaling (1, 21, 38, 39). Raptor possesses dual functions in mTORC1 regulation, as it both suppresses and promotes mTORC1 signaling via a strong, inhibitory interaction and a weaker, required interaction, respectively (38). Although Kim et al. reported that insulin stimulation fails to destabilize the mTOR-raptor interaction (38), our results indicate otherwise. This discrepancy may result from the fact that Kim et al. employed an HEK293T cell line that likely possesses poor insulin responsiveness due to high basal signaling, whereas we employed an HEK293 cell line that possesses strong insulin responsiveness with low basal signaling. Indeed, our experience with an HEK293T cell line shows that these cells respond poorly to insulin. We find that mTOR S2159/T2164 phosphorylation weakens the strong mTOR-raptor interaction found in serum-deprived cells, as phospho-mimetic DE-mTOR bypasses the growth factor requirement. Phospho-defective AA-mTOR does not dominantly stabilize the mTOR-raptor interaction upon insulin stimulation, however, suggesting that mTOR S2159/T2164 phosphorylation is not absolutely required for insulin-induced destabilization. Collectively, these data suggest that mTOR S2159/T2164 phosphorylation releases an inhibitory interaction of raptor with mTOR.
mTOR S2159/T2164 phosphorylation additionally modulates the interaction of raptor with PRAS40, as expression of DE-mTOR weakens the raptor-PRAS40 interaction in the absence of serum growth factors while expression of AA-mTOR strengthens the raptor-PRAS40 interaction in both the absence and presence of insulin. Thus, similar to Akt- and mTOR-mediated phosphorylation of PRAS40 (50, 57, 67, 69), phosphorylation on mTOR itself weakens the raptor-PRAS40 interaction to promote mTORC1 signaling. As PRAS40 contains a TOS motif and may thus inhibit mTORC1 signaling by functioning as a competitive substrate (20, 50, 70), mTOR S2159/T2164 phosphorylation may facilitate S6K1 or 4EBP1 docking and/or positioning by suppressing competitive PRAS40 action. mTOR S2159/T2164 phosphorylation also augments intrinsic mTORC1 catalytic activity, as expression of AA-mTOR reduces mTORC1-associated mTOR S2481 autophosphorylation that occurs upon Rheb overexpression. Collectively, our work reveals several mechanisms that cooperate to increase mTORC1 signaling capacity upon mTOR S2159/T2164 phosphorylation: mTOR kinase domain phosphorylation weakens inhibitory mTORC1 component interactions and promotes mTORC1 intrinsic catalytic activity, which leads to increased S6K1 and 4EBP1 phosphorylation.
Eukaryotic cells maintain a constant size over successive generations via the coordinated action of cell growth and cell cycle progression (17). Failure to couple these processes alters cell size homeostasis and can negatively affect development, tissue organization, and organismal physiology. For example, in S. cerevisiae, inactivation of the cyclin-dependent kinase Cdc28 causes cells to arrest in G1 phase at an abnormally large cell size because cell growth continues in the face of cell cycle arrest (36). Additionally, in mammals, overexpression of Cdk inhibitors induces cells to accumulate in G1 phase at a larger-than-normal cell size (19). These data indicate that while cell growth and cell cycle progression are generally coupled, they represent distinct processes (17, 19). mTORC1 functions as a critical controller of both cell growth and cell cycle progression via its positive control of anabolic metabolism (14, 17–19, 68). Coregulation of both cell growth and cell cycle progression by mTORC1 may thus represent a mechanism by which these processes are effectively coupled. We demonstrate here that mTOR S2159/T2164 phosphorylation promotes both mTORC1-mediated cell growth and G1-phase cell cycle progression. By expressing phospho-mimetic RR/DE-mTOR in the presence of rapamycin (to inhibit endogenous mTORC1 signaling), we find that constitutive mTOR S2159/T2164 phosphorylation is sufficient to promote cell growth to an increased cell size but insufficient to accelerate G1-phase progression. By expressing phospho-defective RR/AA-mTOR in the presence of rapamycin, we find that a complete lack of mTOR S2159/T2164 phosphorylation impairs G1-phase progression but does not strongly impair cell growth. Unexpectedly, we noted that cells expressing RR/AA-mTOR display an increased cell size relative to those expressing wild-type RR-mTOR. Our finding that mTORC1 signaling mediated by RR/AA-mTOR results in slower G1-phase progression resolves this apparent paradox: as RR/AA-mTOR-expressing cells progress more slowly through G1 phase, they have more time to accumulate mass than wild-type cells and therefore display an increased cell size due to an indirect effect. Importantly, we experimentally confirmed this notion by removing the effect of cell cycle progression on cell size. When cell cycle progression was blocked using the drug 2-hydroxyurea, RR/AA-mTOR-expressing cells no longer displayed an increased cell size relative to RR-mTOR-expressing cells, whereas RR/DE-mTOR-expressing cells still displayed an increase in cell size under these conditions. While a lack of mTOR S2159/T2164 phosphorylation did not produce a measurable defect in cell growth, it is reasonable to speculate that cells expressing RR/AA-mTOR indeed grow at a reduced rate relative to cells expressing wild-type mTOR.
Recent data suggest that mTORC1-mediated phosphorylation of 4EBP1 preferentially promotes cell cycle progression over cell growth, while mTORC1-mediated phosphorylation of S6K1 preferentially promotes cell growth over cell cycle progression (14). Thus, not only do cell growth and cell cycle progression represent separable processes, but their regulation may occur via distinct mTORC1-controlled biochemical pathways. Interestingly, we find that under insulin-stimulated conditions, RR/AA-mTOR dominantly inhibits mTORC1 signaling to 4EBP1 while RR-mTOR and RR/DE-mTOR signal to 4EBP1 similarly; these biochemical data may explain why RR/AA-mTOR dominantly impairs G1-phase progression while RR/DE-mTOR does not accelerate G1-phase progression. Additionally, under insulin-stimulated conditions, RR/DE-mTOR augments mTORC1-mediated S6K1 but not 4EBP1 phosphorylation; these data may explain why RR/DE-mTOR augments cell growth but does not accelerate G1-phase progression. Thus, it appears as though stoichiometry of phosphorylation on mTOR S2159/T2164 controls the strength of signaling along either the mTORC1/S6K1 or mTORC1/4EBP1 axes, which may explain our cell size phenotypes.
To fully understand the role of mTOR S2159/T2164 phosphorylation in mTORC1 regulation, it will be important to identify the mTOR S2159 and T2164 kinases and to understand their regulation in response to environmental cues. Additionally, it will be important to determine whether these phosphorylation events also occur on mTOR as part of mTORC2 and whether they control mTORC2 signaling. Emerging data indicate that complex multisite phosphorylation on Tsc2, raptor, and now mTOR underlies regulation of mTORC1 signaling in response to diverse environmental cues. A challenge for the future will be to identify the complete set of regulatory phosphorylation sites and understand how they cooperate to regulate mTORC1 signaling and its control of cellular and organismal physiology.
We thank all members of the Fingar laboratory for critical reading of the manuscript. Thanks go to John Blenis, Nahum Sonenberg, Robert Abraham, and David Sabatini for generously sharing reagents.
This work was supported by the NIH (R01-DK078135), the American Diabetes Association, and the Michigan Diabetes Research and Training Center (MDRTC) (D.C.F.) and by the American Heart Association (B.E.). This work utilized the Cell and Molecular Biology Core(s) of the MDRTC funded by NIH5P60 DK20572 from NIDDK at the University of Michigan.
Published ahead of print on 16 May 2011.