Regulation of the small GTPase Rheb by amino acids

Transcription

1 ORIGINAL ARTICLE Regulation of the small GTPase Rheb by amino acids M Roccio, JL Bos and FJT Zwartkruis (2006) 25, & 2006 Nature Publishing Group All rights reserved /06 $ Department of Physiological Chemistry and Centre for Biomedical Genetics, University Medical Center Utrecht, Universiteitsweg 100, 3584 CG, Utrecht, The Netherlands The mtor/s6k/4e-bp1 pathway integrates extracellular signals derived from growth factors, and intracellular signals, determined by the availability of nutrients like amino acids and glucose. Activation of this pathway requires inhibition of the tumor suppressor complex TSC1/2. TSC2 is a GTPase-activating protein for the small GTPase Ras homologue enriched in brain (Rheb), GTP loading of which activates mtor by a yet unidentified mechanism. The level at which this pathway senses the availability of amino acids is unknown but is suggested to be at the level of TSC2. Here, we show that amino-acid depletion completely blocks insulin- and TPAinduced Rheb activation. This indicates that amino-acid sensing occurs upstream of Rheb. Despite this, amino-acid depletion can still inhibit mtor/s6 kinase signaling in TSC2 / fibroblasts. Since under these conditions Rheb- GTP levels remain high, a second level of amino-acid sensing exists, affecting mtor activity in a Rhebindependent fashion. (2006) 25, doi: /sj.onc ; published online 19 September 2005 Keywords: Rheb; mtor; S6 kinase Introduction Cell growth in eucaryotic organisms is controlled by the combined action of extracellular signals like growth factors and cell intrinsic cues such as nutrient availability (for a review, see Shamji et al., 2003). Biochemical and genetic studies have revealed a central role for the serine/threonine kinase target of rapamycin, mtor, in this process. mtor controls protein synthesis and cell growth by phosphorylation of S6 kinase and 4EBP1 (Thomas and Hall, 1997; Fingar and Blenis, 2004). S6 kinase phosphorylates the ribosomal protein S6 as well as various other targets whose function remains unknown (Harada et al., 2001; Wang et al., 2001; Raught et al., 2004; Richardson et al., 2004). S6 phosphorylation was suggested to enhance the rate of Correspondence:Dr FJT Zwartkruis, Department of Physiological Chemistry and Centre for Biomedical Genetics, University Medical Center Utrecht, Universiteitsweg 100, 3584 CG Utrecht, The Netherlands. g.j.t.zwartkruis@med.uu.nl Received 19 April 2005; revised 18 July 2005; accepted 29 July 2005; published online 19 September 2005 translation of messenger RNAs containing 5 0 oligopyrimidine tracts (TOP mrnas) (Grammer et al., 1996). However, recent data show that it does not act alone in this pathway (Tang et al., 2001; Stolovich et al., 2002). Phosphorylation of 4E-BP1 results in its release from the inactive 4E-BP1/eIF4E complex, allowing eif4e to take part in CAP-site-dependent translation (Gingras et al., 1999). mtor is activated by various growth factors, including insulin. Indeed, the insulin-signaling pathway has emerged as a major regulator of cell growth (Pan et al., 2004). In Drosophila, many elements of this pathway affect cell size and for some of them a conserved function has been demonstrated in mice. In the insulin-signaling pathway phosphatidylinositol-3- kinase (PI-3 kinase) and protein kinase B (PKB) act upstream of mtor. PKB phosphorylates and thereby inhibits the tuberous sclerosis (TSC) tumor suppressor TSC2, which is found in a complex with TSC1 (Potter et al., 2002). The only currently identified catalytic domain in TSC2 is a GTPase-activating protein (GAP) domain at the C-terminus. Genetic screens in Drosophila recently identified Ras homologue enriched in brain (Rheb), a member of the Ras-like family of GTPases (Yamagata et al., 1994), as a component of the insulin/ mtor signaling pathway regulating cell growth (Saucedo et al., 2003; Stocker et al., 2003). Biochemical studies confirmed that Rheb is a substrate for TSC2 GAP activity (Garami et al., 2003; Inoki et al., 2003a; Tee et al., 2003b; Zhang et al., 2003) and is regulated by insulin in a PI-3 kinase-dependent manner (Garami et al., 2003). Increasing GTP levels of Rheb, either by inhibition of TSC2 (Inoki et al., 2002) or by overexpression of Rheb (Inoki et al., 2003a), leads to activation of mtor and S6 kinase. Conversely, overexpression of TSC2 inhibits mtor activity as well as Ser2448 phosphorylation (Inoki et al., 2002). Furthermore, decreasing Rheb activity by RNA interference blocks S6 kinase phosphorylation (Garami et al., 2003; Saucedo et al., 2003). Thus, Rheb is an essential intermediate in insulin-mediated mtor/s6k/4e-bp1 signaling. Amino-acid depletion is known to inhibit mtor activity, decrease basal S6 kinase activity and block its activation normally seen after growth factor stimulation (Hara et al., 1998; Beugnet et al., 2003). Also, 4E-BP1 phosphorylation is inhibited by amino-acid deprivation, suggesting a block at the level of mtor or further upstream (Hara et al., 1998; Iiboshi et al., 1999).

2 658 Importantly, activation of PKB and elements further upstream of the insulin-signaling pathway are not affected by amino-acid withdrawal (Hara et al., 1998). It has been suggested that sensing the availability of nutrients involves TSC2. For example, cells lacking TSC2 have increased S6 kinase activity, which is more resistant to amino-acid depletion (Gao et al., 2002; Saucedo et al., 2003) as compared to that of wild-type (WT) cells. Also, overexpression of Rheb renders S6 kinase activity less dependent on the presence of amino acids (Garami et al., 2003; Inoki et al., 2003a; Stocker et al., 2003; Tee et al., 2003b). Furthermore, overexpression of TSC1/2 reduces the amino-acid-induced phosphorylation of mtor targets (Inoki et al., 2002; Tee et al., 2002). We now show by measuring endogenous Rheb-GTP levels that amino-acid deprivation interferes upstream of Rheb to restrict mtor activity:gtp loading of Rheb is only mildly and transiently inhibited by amino-acid depletion, which increased upon stimulation of depleted cells with amino acids. More importantly, Rheb activation by insulin or TPA is blocked when cells are deprived of amino acids. This is most likely not caused by impaired guanine nucleotide exchange activity on Rheb, since in TSC2- deficient cells Rheb-GTP levels remain invariantly high when amino acids are removed from the culture medium. However, TSC2-deficient cells are still capable of downregulating mtor and S6 kinase activity upon amino-acid starvation. Given the fact that under these conditions endogenous Rheb remains fully activated, these results reveal the existence of an additional regulatory mechanism that restricts mtor activity independently of TSC2/Rheb. Results Regulation of S6 kinase by insulin is mediated by PI-3 kinase-dependent changes in the activity of TSC1/2, leading to increased Rheb-GTP levels and mtor activity. PKC, on the other hand, has been reported to activate S6 kinase in a PI-3 kinase-independent fashion. To test if both pathways were operative in mouse embryonic fibroblasts (MEFs), cells were treated with insulin or TPA and activation of S6 kinase was determined using phospho-specific S6 kinase antibodies. Both stimuli led to a clear increase of T389 and T421 phosphorylation (Figure 1a). In contrast, PKB activation as measured by phosphorylation at S473 was seen only following insulin treatment. To further demonstrate that TPA activates S6 kinase via a distinct pathway, cells were pretreated with the PI-3 kinase inhibitor wortmannin (wm) or the PKC inhibitor bisindolylmaleimide (BIM). As expected, wm completely blocked S6 kinase activation by insulin but not that by TPA. BIM, on the other hand, selectively prohibited TPA-induced S6 kinase activation. Using a distinct PKC inhibitor, Go 6976, identical results were obtained (data not shown). Apart from S6 kinase phosphorylation, also 4E-BP1 was found to be hyperphosphorylated by TPA, suggesting the involvement of mtor. Figure 1 (a) WT MEFs were either untreated or stimulated for 30 min with insulin (5 mg/ml) or TPA (0.1 mg/ml). Where indicated, cells were pretreated with wm (0.1 mm) or BIM (6 mm) for 30 min before stimulation with insulin and TPA. Phosphorylation of S6K, 4E-BP1, PKB and ERK1-2 at the indicated residues was determined using phospho-specific antibodies. (b) Graphic representation of Rheb-GTP content following stimulation with insulin (5 mg/ml) or TPA (0.1 mg/ml) for 30 min. Pretreatment with BIM or wm was for 30 min. Each bar represents the average of at least three independent experiments. Error bars indicate s.d. values of untreated cells which were set at 100%. Quantification was performed using the ImageQuant program using a PhosphorImager. In the lower panel, a representative example is shown. The percentage of GTP bound to Rheb for each sample is indicated above the PhosphorImager picture. GTP loading of Rheb is known to be pivotal for insulin-induced mtor activation. If the PKC and the PI-3 kinase pathways converge at the level of TSC1/2, Rheb-GTP levels are expected to increase following TPA treatment. Therefore, we measured Rheb-GTP levels in cells from parallel dishes treated in exactly the same manner, but labeled with [ 32 P]orthophosphate. As can be seen in Figure 1b, insulin treatment caused a

3 1.5-fold increase in Rheb-GTP, which is slightly lower than what we previously observed in A14-NIH-3T3 cells, which overexpress the insulin receptor (Garami et al., 2003). TPA elevated the fraction Rheb-GTP 1.4- fold. Pretreatment with BIM abolished this effect and actually decreased Rheb-GTP to levels below those seen under basal conditions. Wm did not affect GTP loading by TPA, but, as shown previously (Garami et al., 2003), blocked insulin-induced Rheb activation. Treatment of nonstimulated cells with BIM or wm lowered the basal levels of S6 kinase phosphorylation and also reduced Rheb-GTP to 75 and 86%, respectively, of that seen under basal conditions (Supplementary Figure S1). Together, these results show that growth factor signaling affects Rheb-GTP levels via distinct pathways. Apart from growth factors, amino acids also control the activity of the mtor/s6k/4e-bp1 pathway. In order to obtain more direct evidence for a role of TSC1/ 2/Rheb in amino-acid sensing, we examined the effect of withdrawal of amino acids on the level of GTP-bound Rheb and the effect of resupplementing amino acids to cells previously starved. We observed a moderate and transient inhibitory effect on Rheb activity upon depletion of amino acids for 60 min (Figure 3a, middle white bar) and for 30 and 90 min (data not shown). Interestingly, after h of amino-acid depletion, Rheb GTP levels had returned to their basal level (92% of that of control cells, n ¼ 4). Since after 2 1 2h of amino acid depletion, the activities of both mtor and S6 kinase were still drastically inhibited (Figures 2b and 3c), this result shows that downregulation of mtor by the lack of amino acids can take place independently of Rheb. When we measured the effect of resupplementing amino acids to cells that had been depleted for 150 min, we observed a clear induction of Rheb-GTP loading (1.3- fold, n ¼ 4) (Figure 2a). As expected, phosphorylation of T389 of S6 kinase and S235/236 of S6 was concomitantly increased (Figure 2b). Adding amino acid to cells that had been starved for shorter periods also consistently increased Rheb-GTP levels, although these effects were more variable (data not shown). Together with the observation that overexpression of TSC2 blocks aminoacid-induced mtor activity (Inoki et al., 2002; Tee et al., 2002), these results show that amino acids regulate GTP loading of Rheb. It is known that the absence of amino acids blocks mtor activity downstream of PKB, which is normally activated in amino-acid-starved cells (Hara et al., 1998). To determine the level at which amino-acid depletion interferes with insulin-induced mtor activity, we tested the effect of amino-acid withdrawal on insulin-induced Rheb activation. Whereas insulin stimulation of MEFs in the presence of amino acids increased Rheb-GTP levels 1.5-fold, the same treatment did not lead to Rheb activation when cells had been depleted of amino acids for 150 min (Figure 3a and b). When cells were depleted for 60 min, intermediate results were obtained:insulin treatment still gave rise to an increase in the GTP levels bound to Rheb, but this increase was clearly reduced compared to the induction observed in the presence of amino acids (Figure 3a, middle bars). The activity of the Figure 2 (a) Graphic representation of Rheb-GTP levels following 150 min of amino acid (aa) starvation (white bars), after which aa were readded for 30 min (gray bars; note that in the case of 0 min starvation, this results in a doubling of the aa concentration). Quantification was performed using the ImageQuant program using a PhosphorImager. Values of nonstarved cells were set at 100%. Each bar represents the average of four independent experiments. Error bars indicate s.d. (b) Phosphorylation of S6K (T389) and S6 (S ) and total S6K levels were determined in cells untreated, starved for aa for 150 min or starved for aa and subsequently resupplemented with aa for 30 min as described above. mtor/s6k/4e-bp1 pathway mirrored the effects seen on Rheb. Insulin-induced phosphorylation of T389 of S6 kinase and T37 of 4E-BP1 was almost completely abolished in cells that had been depleted of amino acids for 150 min (Figure 3c). In contrast, phosphorylation of PKB and its direct target FOXO3A was not affected by amino-acid depletion (Figure 3d). In addition, depletion of cells for 60 min only partially blocked S6 kinase and S6 phosphorylation (data not shown). Finally, we tried to block insulin-induced S6 kinase phosphorylation in MEFs by removing only leucine from the tissue culture medium. Although this treatment lowered the basal level of S6 kinase phosphorylation, it did not affect the insulin-induced S6 kinase phosphorylation. Also, GTP loading of Rheb was normal under these conditions (Figure 4). To further consolidate the observation that in aminoacid-starved cells, growth factor signaling toward mtor is blocked as a consequence of the inability to increase Rheb-GTP levels, similar experiments were performed using TPA as a stimulus. Indeed, TPAinduced Rheb activity was abrogated in cells that had been starved of amino acids for 150 min (Figure 5a). This correlated with the inhibition of S6K and S6 phosphorylation (Figure 5b). ERK activation by TPA was not affected by amino-acid depletion, as is evident from probing total cell lysates with a phospho-erk antibody (Figure 5b). Altogether, these results demonstrate that Rheb can no longer be activated by insulin or 659

4 660 Figure 3 (a) Graphic representation of Rheb-GTP content following stimulation for 30 min with insulin (5 mg/ml) in cells that were either starved for aa for 60 and 150 min or left in their presence (0 min). The graph shows the average of four independent experiments. Values are expressed as percentage vs the untreated sample (100%). Error bars indicate s.d. (b) A representative example of the experiments from (a) is shown. The percentage of GTP-bound Rheb for each sample is indicated above the PhosphorImager picture. (c) Phosphorylation of S6K, 4EBP1, and PKB in cells from parallel dishes used for the experiment in (b) as determined by Western blotting using phospho-specific antibodies. Total S6K and tubulin staining were performed to check equal loading. (d) Phosphorylation of FOXO3A as determined by Western blotting from lysates of an independent experiment performed as shown in (c). TPA in the absence of amino acids, thereby restricting mtor activation. The results described so far show that in the absence of amino acids, growth factor signaling to Rheb is impaired. Increases in S6 kinase/s6/4e-bp1 phosphorylation are always accompanied by an increase in Rheb- GTP levels and provide support for the idea that the activity of TSC1/2 depends on the nutritional status of a cell. However, an alternative possibility is that a GAP other than TSC2 is responsible for the observed block in growth factor-induced Rheb-GTP accumulation after amino-acid starvation, or that a GEF for Rheb is regulated by amino-acid availability. We therefore turned to TSC2 knockout cells. We previously showed that in TSC2 knockout cells, Rheb-GTP levels are increased to more than 70% under normal growth conditions, while the GTP/GDP ratio of Ras-bound nucleotides is similar to that in WT MEFs (Garami et al., 2003; Supplementary Figure S2A). This high level of Rheb-GTP is thought to be responsible for the constitutive mtor activity causing elevated levels of phosphorylated S6 kinase and 4E-BP1. Indeed, only direct inhibition of mtor by rapamycin completely negates S6 kinase phosphorylation, and this occurs without any change on the GTP content of Rheb (Figure 6a). We next tested the effect of amino-acid depletion in Rheb-GTP levels in TSC2 knockout cells, however, even prolonged amino-acid depletion did not cause any change in Rheb-GTP levels in TSC2 knockout cells (Figure 6b). In the course of these experiments, we noted that when TSC2 knockout cells are serum starved overnight and subsequently depleted of amino acids for 2 h, phosphorylation of T389 of S6 kinase becomes undetectable (Figure 6c). In addition, a clear reduction in phosphorylation 4E-BP1 was seen. Since TSC2 is not present in these cells and Rheb is still fully activated under these conditions another amino-acid sensor has to be operative in these cells, which blocks mtor activity. In summary, we conclude that restriction of mtor activation in amino-acid-deprived conditions, takes place upstream of Rheb, most likely at the level of TSC2, and that an additional control mechanism exists that can downregulate mtor activity independently of TSC1/2/Rheb. Discussion The integration of growth factor signaling and nutrient sensing is an important step in the control of cell proliferation and cell size, both of which are essential for

5 661 Figure 4 (a) WT MEFs were starved of leucine for 3 h or left in the presence of all essential aa, and then treated for 30 min with insulin or resupplemented with leucine. Phosphorylation of S6 kinase, S6 and PKB are shown. Tubulin is shown as a loading control. (b) Graphic representation of Rheb-GTP content following stimulation for 30 min with insulin (5 mg/ml) or leucine (100 mg/ l) in cells that were either starved for leucine for 180 min ( Leu) or left in their presence ( þ aa). The graph shows the average of two experiments performed in duplo. The adding back of leucine was performed once in duplo. Values are expressed as percentage vs the untreated sample (100%). Error bars indicate s.d. Figure 5 (a) Graphic representation of Rheb-GTP content following stimulation with TPA (0.1 mg/ml) in the presence or absence of aa. Cells were starved for aa for 150 min. Each bar represents the average of three independent experiments. The GTP levels of untreated cells were set at 100%. Error bars indicate s.d. (b) Effect of TPA stimulation on phosphorylation of S6 kinase, S6 and ERK in the presence or absence of aa as determined with phospho-specific antibodies in cells treated in the exact manner as those used in (a). Tubulin staining was performed to show equal loading. proper development and prevention of tumor formation. The recent discovery of Rheb as a downstream target of TSC1/2 has provided a direct read out for the activity of these tumor suppressor genes. It is important to note, however, that TSC1/2 and Rheb are present in a delicate balance in cells (Im et al., 2002), making the measurement of the GTP/GDP ratio of endogenous Rheb, the only reliable method to determine Rheb activity. Here, we demonstrate that apart from the previously described PI-3 kinase-induced Rheb activation, growth factors can also use a PKC-dependent route to increase Rheb-GTP levels. In agreement with this, Tee et al. (2003a) recently reported that TPA could induce phosphorylation of TSC2 on residues partially overlapping with those phosphorylated by PI-3 kinase. Serine 1798 has been shown to be a RSK-dependent phosphorylation site on TSC2, regulated by the Ras/ MAPK pathway (Roux et al., 2004). Moreover, overexpression of TSC1/2 blocked TPA-induced S6 kinase activation (Tee et al., 2003a). Together with the data presented here, it is clear that TSC1/2 and Rheb are common signaling intermediates in mtor activation by distinct growth factors. In Drosophila, PI-3K/PKB are not strictly required for S6 kinase activation (Radimerski et al., 2002). Perhaps this can be explained by alternative mechanisms to inactivate TSC1/2, as shown here for TPA. TSC2 has been put forward as a nutrient sensor, required for the downregulation of mtor following amino-acid and glucose deprivation. Evidence for such a role comes from the observation that cells lacking a functional TSC1/2 complex are more resistant to nutrient deprivation with respect to mtor inactivation than WT cells (Gao et al., 2002; Saucedo et al., 2003). Furthermore, overexpression of its downstream target Rheb has the same effect (Garami et al., 2003; Inoki et al., 2003a; Stocker et al., 2003; Tee et al., 2003b). Indeed, we find that Rheb is clearly under the control of amino acids. First, readdition of amino acids to cells, which have been deprived for 2 h, gives a significant increase in Rheb-GTP levels. Secondly, upon depletion of amino acids, Rheb is no longer GTP loaded following stimulation with insulin or TPA. The duration of amino-acid starvation appears to affect the response of Rheb to growth factor stimulation:a progressive unresponsiveness of Rheb to insulin is seen upon prolonged amino-acid depletion. From these studies, it then appears that an increase in the level of Rheb-GTP acts as a permissive signal for activation of mtor. This is in good agreement with the finding that activation of

6 662 Figure 6 (a) Determination of Rheb-GTP levels in TSC2 / cells following treatment with rapamycin for 1 h. The reduction in phosphorylated S6 kinase as determined by Western blotting is shown in the lower panel. (b) Determination of the Rheb-GTP content in TSC2 / cells following overnight starvation for serum and subsequent depletion of aa for the indicated times. The percentage of GTP is indicted above the PhosphorImager picture. A representative experiment of three is shown. (c) Phosphorylation of S6 kinase, S6, 4E-BP1 and mtor, as determined by Western blotting using phospho-specific antibodies in TSC / and TSC2 þ / þ MEFs, serum starved overnight and subsequently depleted of aa for the indicated times. Tubulin staining was performed to show equal loading. mtor/s6kinase/4e-bp1 signaling by insulin, TPA and amino acids can all be blocked by overexpression of TSC1/2. This view differs from our previous conclusion that Rheb was required and sufficient for mtor activity (Garami et al., 2003). The simplest explanation for this is that overexpressed Rheb by mass action drives mtor into an active signaling complex. The mechanism by which amino-acid depletion prevents GTP loading of Rheb upon growth factor stimulation remains unknown for the moment. We tried to test the involvement of TSC1/2 by comparing the effects of amino-acid depletion on the GTP loading of Rheb in WT and TSC2 knockout cells. Whereas in WT cells, a mild and transient decrease was found after 60 min (Figure 3a, middle white bar) and 30 and 90 min (unpublished observation), in TSC2 knockdown cells, Rheb-GTP levels remained invariantly high after aminoacid starvation. This result shows that amino-acidstarved cells do not activate a Rheb-GAP distinct from TSC1/2. However, it is formally possible that inactivation of a yet unknown GEF for Rheb accounts for the observed block in growth factor-induced Rheb activation. In TSC2 knockout cells, this effect would be obscured by the complete lack of Rheb-GAP activity. Indeed, it has recently been argued that the TSC1/2 complex is not involved in amino-acid sensing. This conclusion was based on the fact that in TSC2 knockout cells mtor activity was still sensitive to the lack of amino acids (Smith et al., 2005). Indeed, we also observed that in TSC2 knockout cells serum starved overnight withdrawal of amino acids strongly reduces both S6K and 4E-BP1 phosphorylation. In conclusion, amino acids affect the activity of mtor both in a Rheb-dependent and a Rhebindependent manner. One of the reasons for two levels of amino-acid control may be a tighter control. Indeed, compared to WT cells, in TSC2 / cells, more stringent amino-acid depletion is required to downregulate mtor/s6 kinase. Alternatively, Rheb might have functions other than the regulation of mtor and require its own independent regulation by amino acids. It should be noted that our study has been performed in MEFs and it may well be that not all cells have both sensing mechanisms. A similar mechanism of dual regulation of mtor takes place, for example, in the control of energy levels, involving the integration of signals from TSC1/2 with more direct effects on mtor. Under conditions where cells cannot maintain normal ATP levels, the concentration of AMP increases to activate 5 0 AMP-activated protein kinase. Subsequently, AMP-kinase activates TSC2 by phosphorylation of Thr1227 and Ser1345, leading to inhibition of mtor

7 (Inoki et al., 2003b). However, earlier studies indicated that limiting the amounts of ATP also directly restrict mtor activity (Dennis et al., 2001). The mechanism by which a lack of amino acids blocks insulin-induced Rheb activation remains to be investigated, but may involve phosphorylation of TSC2 at specific sites as shown for AMP-activated protein kinase. However, alternative mechanisms such as increased stabilization or loss of binding may equally well be involved in maintaining high TSC2 activity (Li et al., 2002). Our results do not exclude the possibility that the dual control mechanism originates from a single amino-acid sensor. A putative candidate could be GCN2, an eif2a kinase, known to be activated by amino-acid starvation (Dever et al., 1992) and glucose starvation (Yang et al., 2000). Such an aminoacid sensor could, for example, directly phosphorylate TSC2 as well as proteins in the TOR complex. Indeed, when this paper was under revision, Long et al. (2005) reported that a complex containing mtor and Rheb was dependent on the presence of amino acids. A major question now to be answered is how Rheb activates mtor. An interesting option is that Rheb regulates complex formation of mtor with Raptor or GbL, which function as scaffolding proteins to bring mtor and its substrates together (Kim et al., 2003). Identification of Rheb effectors will help to resolve this issue further. Materials and methods Reagents Reagents were obtained from the following sources:insulin, wm and TPA, were from Sigma Aldrich Co, BIM I and Go 6976 were from Calbiochem and rapamycin was from BioMOL. Anti-Rheb antibody (C-19), anti-p53 and antip70s6k were from Santa Cruz Biotechnology. Phosphospecific antibodies (phospho-thr389-s6k, Thr421-S6K, Ser235/236-S6, Ser240/244-S6, Thr37-4E-BP1, Ser473-PKB, Thr202/Tyr204 p42 p44 MAPK) were purchased from Cell Signaling and phospho-thr32-fkhrl1 and anti-tubulin were purchased from Upstate Biotechnology and Calbiochem, respectively. Cell culture and amino-acid depletion TSC2 þ / þ and TSC2 / cells were maintained in DMEM supplemented with 10% FCS. Cells were grown to 90% confluency. Where indicated, cells were serum starved overnight. In all cases parallel dishes were prepared and treated in the same manner for measuring the Rheb (GTP/GDP) content and for making of total cell lysates for Western blotting; the only difference was the replacement of [ 32 P]orthophosphate by regular phosphate in the latter. Cells were labeled for 3.5 h with [ 32 P]orthophosphate in phosphate-free DMEM (purchased from Bio Whitacker as phosphate- and amino-acid free DMEM containing 1 g/l glucose), resupplemented with all essential amino acid (Essential amino-acid mix 50 (Bio Whitacker)), all essential amino acids minus leucine (US Biological) or without amino acids. In samples where we studied the effect of amino-acid resupplementation, essential amino acids (or leucine, where indicated) were added back for 30 min at the concentration in which they are normally present in DMEM. Simultaneously, the medium of nondepleted cells was refreshed. Insulin and TPA were added at final concentrations of 5 and 0.1 mg/ml, respectively. wm (0.1 mm), BIM (6 mm) and Go 6976 (1 m) were preincubated for 30 min prior stimulation. Rapamycin was used at a final concentration of 50 nm for 1 h. Radioactive activation assay for small GTPases Measurement of GTP/GDP charged state of endogenous Rheb and Ras was performed as described earlier (Wolthuis et al., 1997). After in vivo labeling and treatment, cells were lysed in Ral-Tag buffer and endogenous Rheb and Ras were pulled down by the C19 antibody and Y13-259, respectively. 20 ml of antibody were used for each immunoprecipitation. The antibody was precoupled overnight to a 1:1 mix of ProtA and ProtG beads. Quantification was performed using a PhosphoImager (STORM820, Molecular Dynamics). Cell lysates and immunoblotting Cells from parallel dishes, treated in the same manner (so also kept in phosphate-free (and eventually amino-acid-free) DMEM) as the ones used for the radioactive GTP loading of Rheb were used for making total cell lysates. After treatment, cells were lysed in laemmli sample buffer and equal amounts of lysates were analysed by SDS PAGE and transferred to PVDF membrane (Perkin-Elmer). Western blots were blocked for 1 h at 41C in TBS/0.1% Tween 20 (TBST) containing 2% nonfat dried milk (Protifar, Nutricia) and 0.5% bovine serum albumin (Sigma). When using phosphorylationspecific antibodies, blots were blocked with 2% bovine serum albumin. Primary antibodies were diluted in TBST (1:5000). HRP-conjugated secondary antibodies were diluted 1: and visualized by chemiluminescence assay. Acknowledgements We thank G Thomas and members from his lab for continuous discussion and critical comments. This work was supported by the Dutch Cancer Society, Koningin Wilhelmina Fonds (MR) and the Center for Biomedical Genetics (FZ). 663 References Beugnet A, Tee AR, Taylor PM, Proud CG. (2003). Biochem J 372: Dennis PB, Jaeschke A, Saitoh M, Fowler B, Kozma SC, Thomas G. (2001). Science 294: Dever TE, Feng L, Wek RC, Cigan AM, Donahue TF, Hinnebusch AG. (1992). Cell 68: Fingar DC, Blenis J. (2004). 23: Gao X, Zhang Y, Arrazola P, Hino O, Kobayashi T, Yeung RS et al. (2002). Nat Cell Biol 4: Garami A, Zwartkruis FJ, Nobukuni T, Joaquin M, Roccio M, Stocker H et al. (2003). Mol Cell 11: Gingras AC, Gygi SP, Raught B, Polakiewicz RD, Abraham RT, Hoekstra MF et al. (1999). Genes Dev 13: Grammer TC, Cheatham L, Chou MM, Blenis J. (1996). Cancer Surv 27: Hara K, Yonezawa K, Weng QP, Kozlowski MT, Belham C, Avruch J. (1998). J Biol Chem 273:

8 664 Harada H, Andersen JS, Mann M, Terada N, Korsmeyer SJ. (2001). Proc Natl Acad Sci USA 98: Iiboshi Y, Papst PJ, Kawasome H, Hosoi H, Abraham RT, Houghton PJ et al. (1999). J Biol Chem 274: Im E, von Lintig FC, Chen J, Zhuang S, Qui W, Chowdhury S, Worley PF et al. (2002). 21: Inoki K, Li Y, Xu T, Guan KL. (2003a). Genes Dev 17: Inoki K, Li Y, Zhu T, Wu J, Guan KL. (2002). Nat Cell Biol 4: Inoki K, Zhu T, Guan KL. (2003b). Cell 115: Kim DH, Sarbassov dos D, Ali SM, Latek RR, Guntur KV, Erdjument-Bromage H et al. (2003). Mol Cell 11: Li Y, Inoki K, Yeung R, Guan KL. (2002). J Biol Chem 277: Long X, Ortiz-Vega S, Lin Y, Avruch J. (2005). J BIol Chem 280: Pan D, Dong J, Zhang Y, Gao X. (2004). Trends Cell Biol 14: Potter CJ, Pedraza LG, Xu T. (2002). Nat Cell Biol 4: Radimerski T, Montagne J, Rintelen F, Stocker H, van der Kaay J, Downes CP et al. (2002). Nat Cell Biol 4: Raught B, Peiretti F, Gingras AC, Livingstone M, Shahbazian D, Mayeur GL et al. (2004). EMBO J 23: Richardson CJ, Broenstrup M, Fingar DC, Julich K, Ballif BA, Gygi S et al. (2004). Curr Biol 14: Roux PP, Ballif BA, Anjum R, Gygi SP, Blenis J. (2004). Proc Natl Acad Sci USA 101: Saucedo LJ, Gao X, Chiarelli DA, Li L, Pan D, Edgar BA. (2003). Nat Cell Biol 5: Shamji AF, Nghiem P, Schreiber SL. (2003). Mol Cell 12: Smith EM, Finn SG, Tee AR, Browne GJ, Proud CG. (2005). J Biol Chem 280: Stocker H, Radimerski T, Schindelholz B, Wittwer F, Belawat P, Daram P et al. (2003). Nat Cell Biol 5: Stolovich M, Tang H, Hornstein E, Levy G, Cohen R, Bae SS et al. (2002). Mol Cell Biol 22: Tang H, Hornstein E, Stolovich M, Levy G, Livingstone M, Templeton D et al. (2001). Mol Cell Biol 21: Tee AR, Anjum R, Blenis J. (2003a). J Biol Chem 279: Tee AR, Fingar DC, Manning BD, Kwiatkowski DJ, Cantley LC, Blenis J. (2002). Proc Natl Acad Sci USA 99: Tee AR, Manning BD, Roux PP, Cantley LC, Blenis. (2003b). Curr Biol 13: Thomas G, Hall MN. (1997). Curr Opin Cell Biol 9: Wang X, Li W, Williams M, Terada N, Alessi DR, Proud CG. (2001). EMBO J 20: Wolthuis RM, de Ruiter ND, Cool RH, Bos JL. (1997). EMBO J 16: Yamagata K, Sanders LK, Kaufmann WE, Yee W, Barnes CA, Nathans D et al. (1994). J Biol Chem 269: Yang R, Wek SA, Wek RC. (2000). Mol Cell Biol 20: Zhang Y, Gao X, Saucedo LJ, Ru B, Edgar BA, Pan D. (2003). Nat Cell Biol 5: Supplementary Information accompanies the paper on the website (