Increased proliferation of atypical smooth muscle cells in pulmonary LAM and in kidney tumors in TS occurs due to mutational inactivation of tumor suppressor TSC2
and mTORC1 activation (54
). This study demonstrates that TSC2-dependent increased cell proliferation and survival also requires mTORC2 and its downstream effector RhoA GTPase. Thus, inhibition of mTORC1 and mTORC2 signaling with siRNAs for raptor and rictor, respectively, had comparable inhibitory effects on TSC2-null cell growth, demonstrating that both mTORC1 and mTORC2 are required for TSC2-null cell proliferation. Importantly, siRNA for rictor inhibited both increased P-Ser473 Akt and RhoA activity, demonstrating mTORC2-dependent regulation of Akt and RhoA GTPase in TSC2-null and LAMD cells. We also show that RhoA activity is necessary for TSC2-null cell survival and that inhibition of RhoA promotes apoptosis through downregulation of Bcl2 and upregulation of the proapoptotic proteins Bim, Bok, and Puma. In vivo
, the combination of the proapoptotic activity of simvastatin with the cytostatic effect of rapamycin abrogated xenographic TSC2-null tumor growth in nude mice, improved animal survival, and prevented tumor recurrence after treatment withdrawal.
While the critical role of mTORC1 signaling in TSC2-deficient cell and tumor growth makes it a logical therapeutic target for LAM and TS, recent preclinical and clinical studies have demonstrated that the mTORC1 inhibitor rapamycin and its analogs have a primarily cytostatic effect in TSC2-deficient cells and tumors and that termination of treatment results in disease reversal (3
). Thus, data from rapamycin (sirolimus) clinical trials for patients with pulmonary LAM and TS (3
) show marked regression in renal angiomyolipoma volume during rapamycin therapy (3
). However, approximately 1 year after treatment withdrawal, tumors reappeared and reached approximately 86% of their original volume (3
). Similarly, the rapamycin analog RAD001 attenuated renal tumor development in TSC2+/−
mice, and its withdrawal led to marked tumor regrowth (72
). Our data demonstrate that 50-day rapamycin treatment in doses that were 4 to 10 times lower than those used in previously published reports (1 mg/kg versus 4 to 10 mg/kg) (59
) had growth-inhibitory effects on TSC2-null tumors in nude mice and suppressed mTORC1 signaling without induction of apoptosis. Not surprisingly, rapamycin withdrawal resulted in tumor recurrence within 1 week after the last treatment, with tumor regrowth reaching 10% of the animal body weight by day 41. Thus, our findings further demonstrate that rapamycin as a single agent has limitations in inhibiting TSC2-null tumor growth and suggest that induction of apoptosis in TSC2-null tumors should have therapeutic benefits.
Lamb and colleagues showed that TSC1, which forms a tumor suppressor complex with TSC2, promotes stress fiber formation through activation of RhoA GTPase (56
). We demonstrate that TSC2 loss induces TSC1-dependent stress fiber formation due to activation of RhoA GTPase that is rapamycin insensitive (27
). Rapamycin-insensitive mTORC2 regulates the actin cytoskeleton and Rho GTPase activity (44
), suggesting that mTORC2 may play a role in modulating Rho activity and Rho-dependent stress fiber formation due to TSC2 loss. Here, we demonstrate that the mTOR and rictor, but not raptor, are required for stress fiber formation in TSC2-null cells and that rictor regulates the activity of RhoA GTPase. Activated Rho rescued rictor-induced stress fiber disassembly in TSC2-null cells, demonstrating that rictor acts upstream of RhoA in regulating its activity and stress fiber formation due to TSC2 loss.
Rho GTPases have prosurvival and antiproliferative effects in different diseases, including human cancers (47
). We show that RhoA GTPase is necessary for TSC2-null cell proliferation and survival, and either RhoA knockdown or specific inhibition of Rho activity significantly inhibits proliferation and induces apoptosis in TSC2-null and LAMD cells in an mTORC1-independent manner. Importantly, inhibition of mTORC2 signaling with siRNA for rictor markedly inhibited TSC2-null cell proliferation at levels comparable to those of siRNA for raptor, while it had little effect on mTORC1-dependent S6 phosphorylation. Importantly, constitutively active Rho rescued rictor siRNA-induced inhibition of proliferation, suggesting that mTORC2 modulates TSC2-null cell proliferation via Rho GTPase. Thus, our data show that TSC2 loss, in addition to activation of mTORC1, leads to mTORC2-dependent activation of RhoA GTPase, which acts as a prosurvival molecule and is required for the proliferation of TSC2-null cells. Thus, these data suggest that therapeutic targeting of RhoA may inhibit abnormal cell growth due to TSC2 loss and promote apoptosis in TSC2-null cells.
Extensive efforts have been made to develop selective Rho inhibitors; however, to date, none have been approved for clinical use (74
). Rho GTPases require prenylation (geranylgeranylation) for membrane binding and activation. Statins, which are HMG-CoA reductase inhibitors, suppress GGPP production, leading to nonselective inhibition of Rho GTPases. Statins, which are widely used clinically to reduce cholesterol levels (89
), also have an oncoprotective effect in humans (15
) and inhibit proliferation and promote apoptosis in different cancer cells (4
); a natural statin, simvastatin, induces apoptosis in numerous human cancer cell lines via suppression of RhoA GTPase activity (10
We show that simvastatin inhibited Rho GTPase activity, attenuated proliferation, induced cleavage of caspase 3, and promoted apoptosis in TSC2-null cells. Importantly, the constitutively active Rho prevented simvastatin-induced caspase 3 cleavage, suggesting that simvastatin inhibits TSC2-related cell proliferation and promotes apoptosis via inhibition of Rho GTPase. In TSC2-null and LAMD cells, the RhoA-dependent apoptotic response involves downregulation of antiapoptotic Bcl2 and upregulation of proapoptotic Bim, Bok, and Puma. Further, simvastatin attenuated TSC2-null subcutaneous tumor growth in nude mice and prolonged animal survival. Interestingly, simvastatin-dependent inhibition of tumor growth, which was observed after 12 days of simvastatin treatment, correlates with statin pharmacokinetic studies (62
). Importantly, simvastatin promoted apoptosis and attenuated proliferation while it had little effect on mTORC1 signaling, confirming that the effects of simvastatin are mTORC1 independent. Although a relatively high dose of simvastatin was used in this study, because mice metabolize simvastatin more rapidly than humans, lower doses may be effective in humans (48
Consistent with our in vivo
findings, simvastatin inhibited proliferation, induced apoptosis, and cooperated with rapamycin in the inhibition of TSC2-null ELT3 cell growth in vitro
. Importantly, similar data were obtained for primary LAMD cells dissociated from the nodules from the lungs of LAM patients (30
), suggesting that our findings may be applicable to LAM disease. The concentrations of simvastatin that inhibited growth and induced apoptosis in TSC2-null ELT3 and LAMD cells (0.3 to 1 μM) were comparable to the plasma levels of statins achieved in humans in clinical trials (2.32 ± 1.27 μM at peak concentration) (86
) and did not result in severe drug toxicity in the patients (91
). Because rapamycin elevates triglycerides and total cholesterol levels (69
), statins are routinely used in rapamycin-treated transplant patients to decrease the risk of cardiovascular disease and prolong survival (2
), and studies show no significant pharmacokinetic drug-drug interactions between rapamycin and statins (53
We demonstrate that simvastatin and rapamycin cooperated in inhibiting TSC2-null and LAMD cell proliferation in vitro, in reducing TSC2-null tumor size, and in improving tumor-bearing mouse survival in vivo. The combination of simvastatin and rapamycin induced apoptosis, inhibited mTORC1 signaling, and abrogated DNA synthesis in TSC2-null tumors. Importantly, in contrast to rapamycin-treated mice, which demonstrated marked tumor recurrence after rapamycin withdrawal, no tumor regrowth was detected in simvastatin- and simvastatin-rapamycin-treated mice during 9 months of observation following the last treatment. Taken together, these data demonstrate that the combination of cytostatic rapamycin with proapoptotic simvastatin improves survival, suppresses tumor growth, and prevents tumor regrowth upon treatment withdrawal.
Collectively, our data demonstrate that mTORC2-dependent RhoA GTPase activation is necessary for TSC2-null cell growth and survival. In vivo targeting of RhoA GTPase with simvastatin and mTORC1 with rapamycin abrogates TSC2-null tumor growth, induces apoptosis, increases tumor-bearing mouse survival, and prevents posttreatment tumor regrowth. Our findings have important implications, because they show RhoA GTPase is a potential therapeutic target for combinational therapy in diseases associated with TSC2 dysfunction.