Inhibition of S6 kinase suppresses the apoptotic effect of eIF4E ablation by inducing TGF-β-dependent G1 cell cycle arrest

Abstract

The mammalian target of rapamycin complex 1 (mTORC1) is a critical regulator of cap-dependent translation through its direct activation of ribosomal protein p70 S6 kinase (S6 kinase) and indirect activation of eukaryotic initiation factor 4E (eIF4E). We recently reported that inhibition of eIF4E expression caused apoptosis in cancer cells in the absence of serum. This was indicated by treatment with the mTORC1 inhibitor rapamycin, which suppressed both S6 kinase and 4E-BP1 phosphorylation (dephosphorylated 4E-BP1 binds and inactivates eIF4E), or by knockdown of eIF4E. We report here that knockdown of eIF4E also causes apoptosis in the presence of serum. This was unexpected because rapamycin induces G1 cell cycle arrest in the presence of serum. Upon investigation, we have found that inactivated S6 kinase prevents the apoptotic effect observed by singular knockdown of eIF4E and results in G1 cell cycle arrest. This effect is dependent on TGF-β (transforming growth factor-β) signaling which contributes to G1 cell cycle arrest. Suppression of S6 kinase phosphorylation alone is insufficient to mediate cell cycle arrest, indicating that complete G1 cell cycle arrest is due to suppression of both S6 kinase and eIF4E. These data indicate that the cytostatic effect of rapamycin is suppression of both S6 kinase and eIF4E, while the cytotoxic effects are due suppression of eIF4E in the absence of S6 kinase-dependent activation of TGF-β signals. Our findings place an importance on the evaluating the activity/expression level of S6 kinase and eIF4E as readouts for rapamycin/rapalog efficacy.

Introduction

It is widely accepted that rapamycin suppresses phosphorylation of mTOR complex 1 (mTORC1) substrate S6 kinase at low nano-molar doses. Through an allosteric mechanism, rapamycin preferentially inhibits mammalian target of rapamcyin complex 1 (mTORC1) but also inhibits mTORC2 under certain conditions [1], [2]. We recently reported that high (micro-molar) doses of rapamycin also suppressed phosphorylation of another mTORC1 substrate, eukaryotic initiation factor 4E (eIF4E)-binding protein 1 (4E-BP1). The high dose induced apoptosis in cancer cells in the absence of serum. We attributed cell death to indirect inhibition of the cap-dependent translation initiator, eIF4E, because de-phosphorylated 4E-BP1 binds eIF4E [3], [4]. Supportively, eIF4E ablation also caused apoptosis. This finding correlated with previous reports that the key mTORC1 substrate for cell proliferation and survival is eIF4E [5], [6]. In contrast, rapamycin induced G1 cell cycle arrest in the presence of serum that protected cells from progressing into S-phase [7], wherein the lack of eIF4E activity would otherwise cause apoptosis [7], [8]. G1 cell cycle arrest was dependent on TGF-β signaling [7] which is suppressed by mTOR [9], [10], Low-dose rapamycin concentrations sufficient to suppress phosphorylation of S6 kinase (but not 4E-BP1) induced TGF-β signaling [10]. While low-dose rapamycin treatment impeded G1 cell cycle progression, high-dose treatment was required to completely block progression from mitosis to S-phase [8]. In sum, rapamycin induces G1 cell cycle arrest at doses that effectively suppress the phosphorylation of both S6 kinase and 4E-BP1.

In this report, we investigated the role of eIF4E ablation on cell cycle arrest in the presence of serum. As aforementioned, inhibiting S6 kinase (low-dose rapamycin) was insufficient to mediate complete G1 arrest. Here, our results indicate similarly, eIF4E ablation is insufficient to induce cell cycle arrest and instead results in apoptotic cell death. Surprisingly, suppression of S6 kinase effectively prevented the apoptotic effect of eIF4E ablation. Dual inhibition of both proteins causes a TGF-β-dependent G1 cell cycle arrest. The findings reveal complex effects of rapamycin on mTORC1 substrates with clinical implications for using translational activators S6 kinase and eIF4E as readouts for clinical efficacy.