The AKT signalling pathway is heavily mutated in a variety of human malignancies. In fact, mutations of AKT pathway components and upstream regulators cover nearly the entire spectrum of human cancers, suggesting a broad requirement for AKT activation in tumourigenesis (). Although genetic alterations of AKT are relatively rare in human cancers, multiple mouse studies have demonstrated that the expression of constitutively active AKT isoforms is sufficient to drive tumourigenesis (Mende et al, 2001
; Majumder et al, 2003
; Tan et al, 2008
). Furthermore, AKT hyperactivity has been shown to be critically required for tumourigenesis caused by more frequently occurring genetic lesions upstream of AKT signalling, such as PTEN loss (Chen et al, 2006
). Despite a wealth of knowledge on genetic mutations leading to the oncogenic activation of AKT and a growing appreciation for the ability of AKT to coordinately regulate mRNA translation, the extent to which deregulated AKT translational control functions as an oncogenic driver remains largely undefined. Recent studies have, however, highlighted a critical requirement for enhanced translation initiation downstream of oncogenic AKT signalling. Strikingly, oncogenic AKT seems to enhance translation initiation largely through hyperactivation of the eIF4E translation initiation factor, which is a bona-fide oncogene.
Common mutations in the PI3K–AKT–mTOR signalling pathway
The oncogenic potential of eIF4E has been well described both in vitro
and in vivo
. Overexpression of eIF4E is sufficient to induce transformation of fibroblasts and primary epithelial cells in culture, and eIF4E overexpression in mice leads to increased cancer susceptibility in a range of tissues (Lazaris-Karatzas et al, 1990
; Avdulov et al, 2004
; Ruggero et al, 2004
). While these findings, along with evidence of eIF4E overexpression in human cancers (Flowers et al, 2009
; Graff et al, 2009
; Wang et al, 2009
), support the notion that eIF4E is oncogenic, a direct connection between eIF4E and translational deregulation downstream of oncogenic AKT signalling has only recently been described. Some of the first evidence for such a connection came from a study showing that pharmacological inhibition of oncogenic RAS and AKT in glioblastoma cells caused a rapid and profound change in mRNA translation that far outweighed transcriptional changes and was associated with loss of mTORC1-dependent phosphorylation of 4EBPs (Rajasekhar et al, 2003
). This study identified translational regulation of several mRNA targets important for cancer development, and suggested that altered translational control downstream of eIF4E hyperactivation may be required for AKT-driven cellular transformation.
Our group demonstrated in vivo
that hyperactivation of eIF4E is necessary for AKT-mediated tumourigenesis. Using a T-cell lymphoma model driven by overexpression of constitutively active AKT, we showed that enhanced protein synthesis through eIF4E hyperactivation was required for AKT-mediated tumourigenesis. We found that AKT overexpressing pretumour progenitor T cells possessed a distinct survival advantage, which was abrogated when eIF4E hyperactivity was restored to wild-type levels. Using a candidate gene approach, we found that this survival advantage was due in part to translational upregulation of the antiapoptotic Mcl-1. Importantly, we were also able to pharmacologically inhibit eIF4E hyperactivity downstream of oncogenic AKT, which resulted in significant inhibition of tumour growth (see below; Hsieh et al, 2010
). As such, we identified the 4EBP/eIF4E axis as a druggable target that regulates translation downstream of oncogenic AKT.
The requirement for eIF4E hyperactivity in AKT-driven tumours has been further substantiated by recent studies. For example, it was found that the efficacy of an AKT inhibitor in human cancer cell lines correlated with its ability to inhibit phosphorylation of 4EBPs and block cap-dependent translation. This study showed that in cell lines where AKT inhibition failed to block phosphorylation of 4EBPs, the MAPK signalling pathway was frequently activated. The authors further demonstrated that combined pharmacological inhibition of AKT and MAPK signalling was able to inhibit phosphorylation of 4EBPs and prevent the in vivo
growth of cell lines resistant to AKT inhibition alone. Importantly, the authors were able to attribute this combinatorial drug effect directly to the inhibition of eIF4E hyperactivity, as the overexpression of a non-phosphorylatable form of 4EBP1 was sufficient to block the growth of these cells in xenografts (She et al, 2010
In addition to 4EBP-dependent control, eIF4E activity is positively regulated through phosphorylation at serine 209 by the MAP kinase targets MNK1/2. Whole body expression of a knock-in mutant of eIF4E, which can no longer be phosphorylated at this residue, was found to decrease the incidence and grade of prostatic intraepithelial neoplasia in a mouse prostate cancer model driven by PTEN loss (Furic et al, 2010
). While this study supports a role for eIF4E hyperactivation downstream of oncogenic AKT signalling, it raises several questions: Do all tissues rely on phosphorylation of serine 209 for hyperactivation of eIF4E downstream of oncogenic AKT signalling? More broadly, what is the tissue-specific dependence of eIF4E hyperactivation, which could be achieved by different mechanisms, downstream of oncogenic AKT? Indeed, there is convincing genetic evidence that oncogenic eIF4E alone is sufficient to drive tumourigenesis in specific tissues. Transgenic mice that ubiquitously overexpress eIF4E show that distinct tissues, including the lungs, liver, and the lymphoid compartment, are more prone to oncogenic transformation (Ruggero et al, 2004
). As such, we can speculate that there may be tissue-specific requirements for the eIF4E oncogenic activity downstream of AKT hyperactivation in tumour development. Although many important questions remain to be addressed, the above studies show that eIF4E hyperactivation is not only critically required for AKT-driven tumours but it might also serve as a node on which multiple oncogenic signalling pathways converge, thus representing an attractive therapeutic target.