Transforming Growth Factor β Integrates Smad 3 to Mechanistic Target of Rapamycin Complexes to Arrest Deptor Abundance for Glomerular Mesangial Cell Hypertrophy*

Background: Transforming growth factor β (TGFβ) induces renal hypertrophy and fibrosis. Results: TGFβ-induced deptor down-regulation is necessary for prolonged activation of TORC1/2 and mesangial cell hypertrophy. Conclusion: TGFβ-stimulated Smad 3 contributes to deptor suppression and mammalian target of rapamycin activation. Significance: Sustained deptor expression may alleviate renal glomerular hypertrophy and fibrosis. In many renal diseases, transforming growth factor β (TGFβ)-stimulated canonical Smad 3 and noncanonical mechanistic target of rapamycin (mTOR) promote increased protein synthesis and mesangial cell hypertrophy. The cellular underpinnings involving these signaling molecules to regulate mesangial cell hypertrophy are not fully understood. Deptor has recently been identified as an mTOR interacting protein and functions as an endogenous inhibitor of the kinase activity for both TORC1 and TORC2. Prolonged incubation of mesangial cells with TGFβ reduced the levels of deptor concomitant with an increase in TORC1 and TORC2 activity. Sustained TGFβ activation was required to inhibit association of deptor with mTOR, whereas rapid activation had no effect. Using the mTOR inhibitor PP242, we found that TGFβ-induced both early and sustained activation of TORC1 and TORC2 was necessary for deptor suppression. PP242-induced reversal of deptor suppression by TGFβ was associated with a significant inhibition of TGFβ-stimulated protein synthesis and hypertrophy. Interestingly, expression of siRNA against Smad 3 or Smad 7, which blocks TGFβ receptor-specific Smad 3 signaling, prevented TGFβ-induced suppression of deptor abundance and TORC1/2 activities. Furthermore, overexpression of Smad 3 decreased deptor expression similar to TGFβ stimulation concomitant with increased TORC1 and TORC2 activities. Finally, knockdown of deptor reversed Smad 7-mediated inhibition of protein synthesis and mesangial cell hypertrophy induced by TGFβ. These data reveal the requirement of both early and late activation of mTOR for TGFβ-induced protein synthesis. Our results support that TGFβ-stimulated Smad 3 acts as a key node to instill a feedback loop between deptor down-regulation and TORC1/2 activation in driving mesangial cell hypertrophy.

In many renal diseases, transforming growth factor ␤ (TGF␤)-stimulated canonical Smad 3 and noncanonical mechanistic target of rapamycin (mTOR) promote increased protein synthesis and mesangial cell hypertrophy. The cellular underpinnings involving these signaling molecules to regulate mesangial cell hypertrophy are not fully understood. Deptor has recently been identified as an mTOR interacting protein and functions as an endogenous inhibitor of the kinase activity for both TORC1 and TORC2. Prolonged incubation of mesangial cells with TGF␤ reduced the levels of deptor concomitant with an increase in TORC1 and TORC2 activity. Sustained TGF␤ activation was required to inhibit association of deptor with mTOR, whereas rapid activation had no effect. Using the mTOR inhibitor PP242, we found that TGF␤-induced both early and sustained activation of TORC1 and TORC2 was necessary for deptor suppression. PP242-induced reversal of deptor suppression by TGF␤ was associated with a significant inhibition of TGF␤-stimulated protein synthesis and hypertrophy. Interestingly, expression of siRNA against Smad 3 or Smad 7, which blocks TGF␤ receptor-specific Smad 3 signaling, prevented TGF␤-induced suppression of deptor abundance and TORC1/2 activities. Furthermore, overexpression of Smad 3 decreased deptor expression similar to TGF␤ stimulation concomitant with increased TORC1 and TORC2 activities. Finally, knockdown of deptor reversed Smad 7-mediated inhibition of protein synthesis and mesangial cell hypertrophy induced by TGF␤. These data reveal the requirement of both early and late activation of mTOR for TGF␤-induced protein synthesis. Our results support that TGF␤-stimulated Smad 3 acts as a key node to instill a feedback loop between deptor down-regulation and TORC1/2 activation in driving mesangial cell hypertrophy.
Chronic renal diseases that eventuate in glomerulosclerosis including diabetic nephropathy are characterized by increased local production of transforming growth factor ␤ (TGF␤). Functional consequences of TGF␤ action consist of altered glomerular hemodynamics, whole kidney hypertrophy, and glomerular hypertrophy (1)(2)(3). TGF␤ mediates these effects by binding to its high affinity type II receptor to form a tetrameric complex with the type I receptor (4). In the absence of TGF␤, the type I receptor binds to the negative regulatory protein FKBP12. Binding of TGF␤ induces phosphorylation of TGF␤ receptor I by the type II receptor in the GS domain. This phosphorylation leads to activation of the type I receptor, releasing FKBP12 to recruit receptor-specific transcription factors Smad 2 and Smad 3 onto the receptor, which phosphorylates them at the C terminus (4,5). Phosphorylated Smads dissociate from the type I receptor and the SARA (Smad anchor for receptor activation), a Smad-recruiting protein to the plasma membrane (6). Subsequently, phosphorylated Smad 2 and Smad 3 bind to the co-Smad, Smad 4, and the heterodimer translocates to the nucleus to recruit coactivators or corepressors for target gene regulation (7)(8)(9).
Apart from the canonical signal transduction pathway described above, the TGF␤ receptor activates downstream signaling kinases including ERK1/2, JNK1/2, p38 kinase, and c-Src * This work was supported, in whole or in part, by National Institutes of Health Grant RO1 DK50190 and Veterans Affairs Merit Review grants (to G. G. C.). □ S This article contains supplemental Figs. S1-S9. 1 (8,10,11). We have previously reported that activation of PI 3-kinase signaling in response to TGF␤ in renal glomerular mesangial cells contributes to hypertrophy of these cells prior to expansion of matrix proteins that precedes fibrosis (12,13). We also showed involvement of mechanistic target of rapamycin (mammalian target of rapamycin; mTOR) 5 kinase in TGF␤induced mesangial cell hypertrophy (14). Although TOR exists as two independent kinases in yeast, mammalian TOR is coded by a single gene (15). Mice with homozygous deletion of mTOR die after implantation (16,17). mTOR null embryos show deficiency in macromolecular synthesis, including protein synthesis necessary for cellular hypertrophy (16 -18). The catalytic mTOR kinase forms two distinct complexes known as mTOR complex 1 (TORC1) and TORC2, which contain common, mLST8, and distinct proteins that confer functional specificity to these complexes (19,20). Raptor and PRAS40 are exclusive components of TORC1. Similarly, rictor (protor 1), protor 2, and mSin1 distinguish TORC2 from TORC1. Raptor is essential for TORC1 activation and serves as the docking site for the substrates, whereas PRAS40 acts as a negative regulator of TORC1 activity (15,21,22). Similarly, rictor and mSin1 along with mLST8 maintain the integrity of the TORC2 (15,23). Deficiency of any of these components abolishes TORC2 activity (23)(24)(25). But lack of mLST8 does not affect TORC1 activity (24).
Recently, deptor has been identified as a partner of both TORC1 and TORC2 and acts as a negative regulator of mTOR kinase activity (26). mTOR regulates protein synthesis necessary for diabetic renal hypertrophy including mesangial cell hypertrophy (27-30). TGF␤-induced mesangial cell hypertrophy is mediated by mTOR (14). Although TGF␤ rapidly activates mTOR in glomerular mesangial cells, the precise mechanism by which TGF␤ induces hypertrophy is not known. In the present study, we show that TGF␤ induces down-regulation of deptor and prolonged activation of mTOR in mesangial cells. Both early and sustained activation of mTOR kinase by TGF␤ is required for deptor down-regulation. Moreover, our results demonstrate a contribution of TGF␤-specific Smad 3 to the reduced deptor abundance necessary for mesangial cell protein synthesis and hypertrophy. These data provide evidence for the presence of a novel cross-talk between Smad 3 and mTOR to induce deptor down-regulation, resulting in mesangial cell hypertrophy.
Cell Culture and Adenovirus Infection-Normal human kidney glomerular mesangial cells were grown in DMEM in the presence of 10% fetal bovine serum as described previously (31-33). For the present experiments, the cells were used between passage 7 and 12. The cells were infected with adenovirus vectors at a multiplicity of infection of 50 for 24 h, essentially as described previously (9,14). An adenovirus expressing green fluorescence protein (Ad GFP) was used as control. Before TGF␤ (2 ng/ml) incubation, the cells were incubated in serum-free medium for 24 h to make them quiescent.
Cell Lysis, Immunoblotting, and Immunoprecipitation-Cells were washed with PBS and harvested in radioimmune precipitation assay buffer (20 mM Tris-HCl, pH 7.5, 5 mM EDTA, 150 mM NaCl, 1 mM Na 3 VO 4 , 1 mM PMSF, 0.1% protease inhibitor mixture, and 1% Nonidet P-40) and incubated at 4°C for 30 min. The crude extracts were centrifuged at 12,000 ϫ g for 30 min at 4°C. The protein concentration was determined in the cleared cell lysate. Equal amounts of protein were separated by SDS-polyacrylamide gel electrophoresis and transferred to membrane. Immunoblotting was performed using the indicated antibodies. The protein bands were developed with HRP-conjugated secondary antibody using ECL reagent as described previously (9,14,28,29,34). Equal amounts of cleared cell lysates were immunoprecipitated with the indicated antibodies as described (9,14,35). The immune beads were resuspended in sample buffer, proteins were separated by SDS-polyacrylamide gel electrophoresis and immunoblotted as described above.
RNA Isolation and RT-PCR-Total RNA was prepared from mesangial cells using TRI Reagent. Expression of deptor was determined by quantitative real-time RT-PCR. cDNAs were prepared by reverse transcription from 0.5 g of total RNA using TaqMan reverse transcription reagents (number N808-0234). The cDNA was amplified and quantified in 96-well plates using TaqMan deptor primers (Applied Biosystem) in a 7500 real time PCR machine (Applied Biosystem). The PCR condition was 95°C for 10 min followed by 45 cycles at 95°C for 15 s and 60°C for 1 min. The relative mRNA levels were normalized to the reference GAPDH in the same sample. Data analyses were done by the comparative C t method as described previously (36).
Measurement of Cellular Hypertrophy-At the end of the incubation, the cells were trypsinized and counted using a hemocytometer. Cells were then pelleted at 4000 ϫ g at 4°C, washed with PBS, and lysed in RIPA buffer as described above. The protein content in the total number of cells was determined. Hypertrophy was expressed as an increase in the ratio of cellular protein content to cell number as described previously (28, 29).
Statistics-The significance of the results was assessed by analysis of variance followed by Student-Newman-Keuls analysis as described previously (9,14,28,29,34,35). The mean Ϯ S.E. of the indicated measurements are shown. A p value of less than 0.05 was considered significant.

TGF␤ Induces Down-regulation of Deptor for Prolonged Activation of TORC1 and TORC2-We and others have recently
shown that TGF␤ rapidly increases mTOR kinase activity (14,37). One mechanism involves activation of Akt followed by phosphorylation and inactivation of the negative regulator TSC2, leading to activation of TORC1 (14). More recently, deptor was identified as a component of the mTOR kinase complex that negatively controls the activity of both TORC1 and TORC2 (26). We examined the effect of TGF␤ on the expression of deptor in mesangial cells. Short-term incubation of these cells with TGF␤ did not show any effect on abundance of deptor ( Fig. 1A and supplemental Fig. S1A). However, prolonged treatment with TGF␤ significantly decreased the levels of deptor (Fig. 1B). The reduction was evident at 6 h of TGF␤ treatment and sustained for 24 h (Fig. 1B). To determine the half-life of deptor, we incubated mesangial cells with TGF␤ in the presence of cycloheximide. As shown (supplemental Fig.  S1B), the apparent protein half-life of deptor was 18 h. TGF␤ acts through activation of its type I receptor serine/threonine kinase. SB431542, an inhibitor of TGF␤ receptor I, prevented the TGF␤-induced down-regulation of deptor ( Fig. 1C and supplemental Fig. S1C). The reduction in deptor protein level was associated with a significant decrease in its mRNA in response to TGF␤ (Fig. 1D). SB431542 reversed the TGF␤induced decrease in deptor mRNA (Fig. 1D). These results indicate TGF␤ receptor serine/threonine kinase activity is required for the decrease in deptor abundance in response to TGF␤.
As deptor is a suppressor of mTOR activity and because deptor was not decreased at early time points of TGF␤ stimulation, we tested the activation of this kinase. Phosphorylation of S6 kinase at Thr-389 was used as a surrogate for TORC1 activation. Incubation of mesangial cells with TGF␤ increased phosphorylation of S6 kinase within 2.5 min, which was sen- sitive to the TGF␤ receptor inhibitor SB431542 ( Fig. 2A and  supplemental Fig. S2, A and B). Similarly, prolonged incubation of mesangial cells with TGF␤ increased phosphorylation of S6 kinase ( Fig. 2B and supplemental Fig. S2C). SB431542 inhibited TGF␤-stimulated phosphorylation of S6 kinase at 24 h (supplemental Fig. S2D). Moreover, suppression of deptor using two FIGURE 2. TGF␤ increases TORC1 and TORC2 activities in mesangial cells. A, B, G, and H, lysates from quiescent mesangial cells treated with TGF␤ for the indicated times were used for immunoblotting with phospho-S6 kinase (Thr-389) for TORC1 activity (panels A and B) and phospho-Akt (Ser-473) for TORC2 activity (panels G and H). The same cell lysates were immunoblotted with phospho-Akt (Thr-308), S6 kinase, and Akt antibodies as indicated. C-F, I, and J, human mesangial cells were transfected with deptor sh1 (panels C, E, and I) or deptor sh2 (panels D, F, and J), two independent shRNA vectors targeting different regions of deptor mRNA. Vector expressing a scrambled RNA was used as control. The cell lysates were immunoblotted with phospho-S6 kinase, S6 kinase, deptor, actin antibodies (panels C and D); phospho-4EBP-1 (Thr-37/46), phospho-4EBP-1 (Ser-65), 4EBP-1, Deptor, and actin antibodies (panels E and F); phospho-Akt (Ser-473), phospho-Akt (Thr-308), deptor, Akt, and actin antibodies (panels I and J). Quantifications of these panels are shown in supplemental Fig. S2, B, C, E-J, M, and N. independent shRNAs increased phosphorylation of S6 kinase supporting the hypothesis that late activation of TORC1 is mediated by a decrease in deptor levels (Fig. 2, C and D, and supplemental Fig. S2, E and F). This activation of TORC1 was confirmed by observing increased phosphorylation of another substrate, 4EBP-1, as a result of deptor down-regulation (Fig. 2, E and F, and supplemental Fig. S2, G and H). Note that early activation of TORC1 ( Fig. 2A) does not coincide with deptor down-regulation (Fig. 1B). These results indicate that downregulation of deptor is not required for early activation of TORC1. Similar to the activation of TORC1, TGF␤ rapidly increased phosphorylation of Akt at the hydrophobic site, Ser-473, suggesting activation of TORC2 ( Fig. 2G and supplemental  Fig. S2I). Also, prolonged incubation of mesangial cells with TGF␤ increased TORC2 activity (Fig. 2H and supplemental Fig.  S2J). Both early and prolonged activation was inhibited by SB431542 (supplemental Fig. S2, K and L). shRNA-mediated inhibition of deptor expression also resulted in activation of TORC2 (Fig. 2, I and J, and supplemental Fig. S2, M and N). These results suggest the presence of an inverse correlation between deptor abundance and sustained TORC1/TORC2 activation in response to TGF␤. However, early activation of these kinases does not involve deptor.
mTOR Regulates TGF␤-induced Phosphorylation of Deptor-In previous studies deptor was shown to be phosphorylated at 13 Ser/Thr residues located between the C-terminal DEP domain and PDZ domain (26). Because both TORC1 and TORC2 were activated early in response to TGF␤ (Fig. 2, A and  G), we considered a role of both kinases in regulating phosphorylations of deptor. We used an ATP competitive inhibitor of mTOR, PP242, which blocks both kinase complexes (38). Mesangial cells were incubated with PP242 followed by TGF␤. Deptor immunoprecipitates were immunoblotted with antibody that recognizes serine-phosphorylated proteins. TGF␤ increased phosphorylation of deptor (Fig. 3A). PP242 significantly inhibited this phosphorylation ( Fig. 3A and supplemental Fig. S3A). To confirm this observation, we inhibited both TORC1 and TORC2 simultaneously by down-regulating raptor and rictor, using two independent shRNAs. Inhibition of TORC1 and TORC2 blocked TGF␤-induced serine phosphorylation of deptor (Fig. 3, B and C, and supplemental Fig. S3, B  andC). Moreover, we used a mutant of FLAG-tagged deptor in which all 13 phosphorylation sites were mutated to alanine (13X). This mutant and wild type deptor were transfected into mesangial cells followed by incubation with TGF␤. As shown in Fig. 3D, TGF␤ did not have any effect on mutant deptor, whereas wild deptor was significantly down-regulated (supplemental Fig. S3D).
Deptor interacts with mTOR in both TORC1 and TORC2 (26). Because deptor is an endogenous inhibitor of mTOR activity and we found very rapid activation of both TORC1 and TORC2 (Fig. 2, A and G), we assessed the complex formation between mTOR and deptor in the presence of TGF␤ at early time points. Lysates of mesangial cells incubated with TGF␤ for 5 min were immunoprecipitated with mTOR antibody. This immunopurified mTOR was immunoblotted with deptor antibody. As predicted, mTOR was complexed with deptor in the absence of TGF␤ (Fig. 4A). No change in complex formation between mTOR and deptor in response to TGF␤ was observed at this early time point (Fig. 4A and supplemental Fig. S4A). The reciprocal experiment showed identical results ( Fig. 4B and  supplemental Fig. S4B). Complex formation between deptor and mTOR was inhibited in mesangial cells incubated with TGF␤ for 24 h (Fig. 4, C and D, and supplemental Figs. S4, C and  D). This decreased association may be due to the reduced abundance of deptor at this time of TGF␤ stimulation (Fig. 1B).
Both mTOR Kinases Regulate TGF␤-induced Expression of Deptor-Our results above demonstrate a role of deptor downregulation in the prolonged activation of mTOR. Furthermore, we show that mTOR regulates phosphorylation of deptor. Previously it was shown that mTOR kinase activity is required for the decrease in deptor abundance (26). But the requirement of mTOR for TGF␤-induced suppression of deptor has not been investigated. Preincubation of mesangial cells with PP242, which inhibits both TORC1 and TORC2 at an early time point, reversed the TGF␤-induced down-regulation of deptor ( Fig. 5A and supplemental Fig. S5, A-C). Interestingly, PP242 also prevented the TGF␤-suppressed deptor levels when mesangial cells were treated with the drug after 15 min of incubation with TGF␤ (Fig. 5A, compare lane 2 with lane 4, and supplemental Fig. S5A). This reversal of deptor abundance by PP242 was associated with inhibition of TORC1 and TORC2 (Fig. 5, B and C, and supplemental Fig. S5, D and E). These results conclusively indicate that prolonged activation of both TORC1 and TORC2 is necessary for decreased deptor abundance in response to TGF␤.

TGF␤-stimulated Prolonged Inactivation of 4EBP-1 Regulates Mesangial Cell Protein Synthesis Necessary for
Hypertrophy-Phosphorylation of 4EBP-1 at Thr-37/46 and Ser-65 results in its inactivation (39). Recent reports including our own (20, 34) show involvement of both TORC1 and TORC2 in phosphorylation and inactivation of 4EBP-1. Therefore, we used PP242, which inhibits these kinase complexes (38). Mesangial cells were either treated with PP242 prior to incubation with TGF␤ or first incubated with TGF␤ for 15 min followed by treatment with PP242. In both cases cells were exposed to TGF␤ for 24 h. TGF␤ increased phosphorylation of 4EBP-1 at Thr-37/46 and Ser-65 (Fig. 6A). Both pre-and posttreatment of mesangial cells with PP242 inhibited TGF␤-induced phosphorylation of 4EBP-1 at these sites ( Fig. 6A and  supplemental Fig. S6). Inactivation of 4EBP-1 by TORC1-mediated phosphorylation at these specific serine/threonine sites is necessary for increased protein synthesis (39). TGF␤-induced protein synthesis was significantly inhibited by pre-as well as post-treatment with PP242 (Fig. 6B). Similarly, PP242 markedly blocked hypertrophy of mesangial cells in response to TGF␤ under both conditions (Fig. 6C). These results suggest that prolonged activation of both TORC1 and TORC2 by TGF␤ contributes to phosphorylation and inactivation of 4EBP-1, resulting in increased protein synthesis and hypertrophy of mesangial cells.
TGF␤-stimulated Smad 3 Regulates Abundance of Deptor-Activated TGF␤ receptor I initiates signal transduction by direct phosphorylation of Smad 3, which regulates TGF␤-induced biological activities (4, 40). Incubation of mesangial cells with TGF␤ increased phosphorylation of Smad 3 (supplemen-tal Fig. S7A). To examine the involvement of Smad 3 phosphorylation in deptor down-regulation, we used adenovirus-mediated expression of Smad 7, an endogenous inhibitor of TGF␤stimulated Smad 3 phosphorylation/signaling. Expression of Smad 7 in mesangial cells reversed the TGF␤-induced downregulation of deptor concomitant with inhibition of phosphorylation of Smad 3 ( Fig. 7A and supplemental Fig. S7, A and B). This reversal of TGF␤-induced deptor suppression by Smad 7 was associated with inhibition of TORC1 as well as TORC2 activities (Fig. 7, B and C, and supplemental Fig. S7, C and D). To directly test the role of Smad 3, we employed Smad 3 siRNA. A pool of three siRNAs directed against Smad 3 mRNA significantly reversed TGF␤-suppressed deptor levels ( Fig. 7D and  supplemental Fig. S7E). Expression of Smad siRNAs inhibited both TORC1 and TORC2 activities induced by TGF␤ (Fig. 7, E  and F, and supplemental Fig. S7, F and G). To further confirm the role of Smad 3, we directly expressed Smad 3. As expected, TGF␤ repressed the levels of deptor (Fig. 7G). Similarly, overexpression of Smad 3 decreased the expression of deptor (Fig.  7G, compare lane 1 with lane 3 and supplemental Fig. S7H). Furthermore, expression of Smad 3 in the presence of TGF␤ maintained deptor abundance analogous to the levels observed with TGF␤ or Smad 3 alone (Fig. 7G). Concomitantly, the decrease in deptor expression by Smad 3 was accompanied by increased TORC1 and TORC2 activities similar to those found with TGF␤ stimulation (Fig. 7, H and I, and supplemental Fig.  S7, I a n dJ). These results conclusively demonstrate that TGF␤stimulated Smad 3 signal transduction regulates deptor down-regulation.
TGF␤-stimulated Smad 3 Signaling Regulates Deptor-mediated Mesangial Cell Protein Synthesis and Hypertrophy-We have shown above that TGF␤-induced protein synthesis is sen-FIGURE 3. mTORC1/2 regulate TGF␤-induced phosphorylation of deptor. A, human mesangial cells were treated with 1 M PP242 for 1 h prior to incubation with 2 ng/ml of TGF␤ for 4 h. This incubation time with TGF␤ was chosen because beginning at 6 h a significant deptor down-regulation was detected (Fig.  1B). The cell lysates were immunoprecipitated (IP) with control antibody against actin or deptor antibody as indicated. The immunoprecipitates were immunoblotted with phosphoserine antibody. B and C, human mesangial cells were transfected with two shRNAs against raptor and rictor (sh1, panel B; sh2, panel C). Transfected cells were treated with TGF␤ as described in panel A. The cell lysates were immunoprecipitated with actin or deptor antibody followed by immunoblotting (IB) with phosphoserine antibody as described in panel A. D, phosphorylation of deptor is required for its down-regulation by TGF␤. Human mesangial cells were transfected with FLAG-tagged phospho-deficient mutant deptor 13X (lanes 1 and 2) or wild type deptor (lanes 3 and 4). Transfected cells were incubated with 2 ng/ml of TGF␤ for 24 h. The cell lysates were immunoblotted with FLAG and actin antibodies as indicated. Quantifications of these panels are shown in supplemental Fig. S3, A-D. sitive to mTOR inhibition (Fig. 6B). We also showed that deptor down-regulation activates both TORC1 and TORC2 (Fig. 2,  C-F, I, and J). We examined whether deptor regulates mesangial cell protein synthesis. Down-regulation of deptor using two independent shRNAs significantly increased protein synthesis similar to TGF␤ treatment (supplemental Fig. S8, A and B). Next, we tested if the integration between the TGF␤-induced canonical Smad 3 signaling and non-canonical mTOR activation by deptor down-regulation contributes to protein synthesis. Expression of Smad 7, which inhibits TGF␤-induced activation of Smad 3 (supplemental Fig. S7A), blocked TGF␤-stimulated protein synthesis (Fig. 8, A and B, and supplemental Fig. S9, A and B). Interestingly, inhibition of deptor expression by two independent shRNAs significantly reversed the inhibitory effect of Smad 7 on TGF␤-stimulated protein synthesis (Fig. 8, A and B, and supplemental Fig. S9, A and B). Similarly, expression of Smad 7 abrogated TGF␤-stimulated mesangial cell hypertrophy, which was significantly prevented by suppression of deptor expression (Fig. 8, C and D, and supplemental Fig. S9, C and D). These results demonstrate that TGF␤-induced Smad 3 signaling regulates deptor-mediated mesangial cell hypertrophy.

DISCUSSION
In the present study we identified deptor, an endogenous inhibitor of mTOR activity, as a pathological target of TGF␤ receptor serine/threonine kinase. We present evidence demonstrating Smad 3-dependent activation of mTOR kinase through down-regulation of its endogenous inhibitor deptor. Our results also show that this cross-talk between Smad 3 and deptor abundance regulates mesangial cell hypertrophy in response to TGF␤ (Fig. 9).
In physiologic kidney hypertrophy such as compensatory renal hypertrophy following uninephrectomy and in disease states such as diabetes, a role for TGF␤ has been recognized (41-43). For example, TGF␤1 null renal cells display reduced hypertrophy in response to high glucose (44). Furthermore, type 1 diabetic mice heterozygous for the type II TGF␤ receptor exhibit significantly decreased glomerular and mesangial cell hypertrophy, thus reinforcing the requirement of TGF␤-induced signal transduction in this process (45). We and others have shown a pivotal role of mTOR in pathologic renal hypertrophy in vitro and in vivo (14,30,42,46). Mechanistically, we established phosphorylation of the tumor suppressor protein tuberin by the TGF␤-activated PI 3-kinase/Akt signaling pathway resulting in activation of TORC1, which lead to mesangial cell hypertrophy (14).
By definition, hypertrophy results from an increase in protein synthesis in the absence of DNA synthesis. A role of TORC1 in protein synthesis is well established (39). It is executed mainly by TORC1 substrates 4EBP-1 along with S6 kinase. Phosphorylated and activated S6 kinase binds or phosphorylates multiple proteins acting at the rate-limiting steps in phases of mRNA translation initiation and elongation (20). Moreover, studies have postulated a role of TORC2 via the Akt hydrophobic motif phosphorylation in protein synthesis, which may involve phosphorylation of 4EBP-1 (13,34,47). Although mTOR activity in both TORC1 and TORC2 has been shown to be positively regulated by PI 3-kinase, recent studies have identified negative regulators of mTOR, such as FBXW7 and deptor (26,48,49).
Deptor was originally discovered as a protein that contains two tandem DEP (dishevelled, egl-10, and pleckstrin) domains in the N terminus and a PDZ (postsynaptic density 95, discs large, zonular occludens-1) domain in the C terminus (26). It binds directly through its PDZ domain to the FAT domain of mTOR and blocks the kinase activity of both TORC1 and TORC2 (26). Moreover, deptor is an extremely unstable protein, which undergoes rapid degradation via a proteasomal pathway, resulting in activation of both TORC1 and TORC2 (26, 50). In the present study, we demonstrate that activation of FIGURE 6. TORC1 and TORC2 regulate TGF␤-stimulated 4EBP-1 phosphorylation, protein synthesis, and hypertrophy in mesangial cells. Quiescent mesangial cells were incubated with PP242 and TGF␤ essentially as described in the legend of Fig. 5. Panel A, cell lysates were immunoblotted with phospho-4EBP-1 (Thr-37/46), phospho-4EBP-1 (Ser-65), and 4EBP-1 antibodies. Quantification of these data are shown in supplemental Fig. S6. B, TGF␤-incubated cells were labeled with [ 35 S]methionine. Protein synthesis was determined essentially as described (13,14,28,29). C, TGF␤-incubated cells were trypsinized and counted, and protein content was determined. The cell hypertrophy was expressed as an increase in ratio of total protein to cell number as described under "Experimental Procedures" (13,28). For panels B and C, mean Ϯ S.E. of 6 measurements are shown. *, p Ͻ 0.001 versus control; **, p Ͻ 0.001 versus TGF␤stimulated; #, p Ͻ 0.001 versus TGF␤ stimulated.
Although deptor contains phosphorylation sites for multiple kinases including ERK1/2, PKC, and GSK3␤, they do not regulate its stability (51). Recent mechanistic studies identified CK1␣, S6 kinase 1, and RSK1 as other kinases that contribute to the degradation of deptor (51)(52)(53). Furthermore, deptor contains multiple phosphorylation sites for mTOR (TORC1 and TORC2) in a serine-rich domain between the second DEP and C-terminal PDZ domains (26, 51). In fact, phosphorylation of deptor by mTOR (either TORC1 or TORC2 or both) primes its phosphorylation by other kinases, which confer its ability to bind to the F box protein ␤TRCP1 (51). In line with this observation, we found enhanced TGF␤-induced phosphorylation of deptor, which was sensitive to both TORC1 and TORC2 inhibition (Fig. 3). Furthermore, the TGF␤-induced decrease in deptor expression was dependent on both TORC1 and TORC2 activities (Fig. 5). In fact, our results demonstrate that both rapid and prolonged activation of both kinase complexes are necessary for reduction in deptor levels, resulting in phosphorylation/activation of S6 kinase, phosphorylation/inactivation of 4EBP-1, and activation of Akt as judged by the deptor suppression-sensitive phosphorylation at Ser-473 by TORC2 (Figs. 5 and 6A). Moreover, both early activation of TORC1 and TORC2, and late activation of mTOR, which is controlled by diminished deptor levels, are necessary for TGF␤-induced protein synthesis and hypertrophy of mesangial cells (Figs. 5 and 6).
Full activation of Akt requires phosphorylation at both Thr-308 and Ser-473 sites (20,47,54). It has been shown previously that TORC2-mediated phosphorylation may be necessary for phosphorylation of Akt at Thr-308 by PDK-1 (20, 24). In line FIGURE 7. TGF␤-stimulated Smad 3 signaling regulates deptor down-regulation and, TORC1 and TORC2 activation. Mesangial cells were infected with Ad Smad 7 (panels A-C) and Ad Smad 3 (panels G--I) in serum-free medium for 24 h prior to incubation with 2 ng/ml of TGF␤ for 24 h as described under "Experimental Procedures." As control, Ad GFP was used. D-F, mesangial cells were transfected with a pool of three siRNAs against Smad 3 or scramble RNA. Transfected cells were incubated with TGF␤ as described above. Cell lysates were immunoblotted with deptor and actin antibodies (panels A, D, and G); phospho-S6 kinase (Thr-389) and S6 kinase antibodies (panels B, E, and H); phospho-Akt (Ser-473), phospho-Akt (Thr-308), and Akt antibodies (panels C, F, and I). Expression of Smad 7 is shown by immunoblotting with FLAG (panels A-C). Expression of Smad 3 is shown by immunoblotting with Smad 3 (panels D-F) and anti-HA (panels G-I) antibodies, respectively. Quantification of these data are shown in supplemental Fig. S7, B-J.
with this, we find that TGF␤-induced deptor down-regulation resulted in an increase in TORC2 activity, which phosphorylates Akt at Ser-473, and is concomitant with an increase in its phosphorylation at the PDK1 site Thr-308 (Fig. 2, H and J). As inhibition of TORC1 results in increased PI 3-kinase-mediated phosphorylation of Akt at Thr-308, one would expect increased phosphorylation at this site by TGF␤ in the presence of PP242, which is an ATP-competitive inhibitor of mTOR (26,38,55,56). In contrast to this notion, we found inhibition of Akt phosphorylation at Thr-308 in TGF␤-treated mesangial cells in the presence of PP242 (Fig. 5C). These results suggest that TGF␤ forced inhibition of deptor expression and consequently TORC2 activation contributes to the full activation of Akt in mesangial cells.
The role of canonical TGF␤ receptor-specific Smad 3 in mediating renal pathology, including hypertrophy, has been reported (57)(58)(59). Our results demonstrate that TGF␤-induced Smad 3 activation is necessary for deptor down-regulation (Fig.  7, A, D, and G). To our knowledge, this is the first evidence for Smad 3 regulation of deptor. TGF␤ receptor-specific Smads including Smad 3 contain N-terminal MH-1 and C-terminal MH-2 domains separated by a proline-rich linker region (4). The MH-1 domain confers nuclear import of proteins and takes part in DNA binding, whereas the MH-2 domain binds to the type I receptor and forms a complex with Smad 4 (4, 60). Although Smad 3 acts as a transcriptional activator for many genes induced by TGF␤, it has also been shown to repress certain genes (4,61). Deptor has been shown to be regulated partly at the level of mRNA expression (26). We found that TGF␤ decreased the expression of deptor mRNA (Fig. 1D). In fact using a reporter plasmid, deptor promoter activity was significantly inhibited by TGF␤ (data not shown). Therefore, our FIGURE 8. TGF␤-stimulated Smad regulates mesangial cell protein synthesis and hypertrophy via deptor down-regulation. Human mesangial cells were transfected with two independent shRNAs (250 ng/well) targeting deptor (deptor sh1 and deptor sh2) along with Ad Smad 7 infection with 50 multiplicity of infection as described under "Experimental Procedures." The cells were incubated with TGF␤ for 24 h. A and B, the cells were labeled with [ 35 S]methionine and its incorporation was determined as described previously (13,28). C and D, cellular hypertrophy was determined as ratio of protein content to cell number as described in legend of Fig. 5C (13, 28). Mean Ϯ S.E. of triplicate measurements are shown. In panel A, *, p Ͻ 0.01 versus control; **, p Ͻ 0.05 versus TGF␤; @, p Ͻ 0.05 control; #, p Ͻ 0.01 versus TGF␤ ϩ Smad 7. In panel B, *, p Ͻ 0.01 versus control; **, p Ͻ 0.05 versus TGF␤; @, p Ͻ 0.01 control; #, p Ͻ 0.01 versus TGF␤ ϩ Smad 7. In panels C and D, *, p Ͻ 0.001 versus control; **, p Ͻ 0.001 versus TGF␤; @, p Ͻ 0.001 control; #, p Ͻ 0.001 versus TGF␤ ϩ Smad 7. Expression of Smad 7 and deptor is shown in supplemental Fig. S9. results demonstrating down-regulation of deptor in response to TGF␤ or by Smad 3 may result from transcriptional repression (Fig. 7, A, D, and G).
The MH-2 domain of Smad 3 is known to interact with the phospholipid-binding FYVE domain containing the protein SARA, which targets Smads to the endosomes where active TGF␤ signaling occurs (4, 6, 40). Recent evidence supports intracellular localization of mTOR at the endosomes (62)(63)(64)(65). Whether Smad 3 binds mTOR in TGF␤-stimulated cells is not known. However, our results provide strong evidence for the requirement of Smad 3 for activation of TORC1/2 and mesangial cell hypertrophy (Figs. 7, B, C, E, F, H, and I, and 8).
Abundance of deptor is significantly low in many cancers (26, 50). A key biological consequence of deptor suppression is regulation of cancer cell survival (26, 50). In fact, deptor inhibition protects tumor cells from apoptosis induced by serum withdrawal. This effect occurs in the absence of phosphorylation of Akt, suggesting a lack of involvement of this kinase in the biological activity elicited by the deptor-repressed state (26). In contrast to these results, in mesangial cells we found that reduced deptor levels in response to TGF␤ increased phosphorylation of Akt at both sites Ser-473 and Thr-308, leading to increased protein synthesis (Figs. 1, 2, 5, and 6). Thus, to induce mesangial cell hypertrophy, the contribution of deptor is to regulate Akt activation. In summary, the results presented here are consistent with a model in which TGF␤-stimulated Smad 3 induces deptor down-regulation, which contributes to sustained activation of TORC1 and TORC2 to maintain mesangial cell hypertrophy. Thus, deptor may be amenable for rational drug designing to increase its expression that may ameliorate renal disease such as diabetic nephropathy where mesangial cell hypertrophy plays a pathologic role.