For this study, we wanted to focus on the cooperation between genetic aberrations to drive disease progression in early-stage prostate cancer, which has rarely been addressed in the current literature. To this end, we decided to use a sensitised in vivo transposon mutagenesis screen of a genetically modified mouse model to uncover different tumour-promoting genes.
The PI3K-AKT pathway is frequently mutated in prostate cancer, with the Pten gene often involved. Given that the homozygous prostate-specific deletion of Pten in mice leads to prostatic intraepithelial neoplasia and then invasive carcinoma within weeks, we opted to use mice with a heterozygous mutation in PTEN to give us better scope to determine what was responsible for tumour promotion.
From our piggyBac transposon screen, we identified Bzw2 and Eif5a2 as cooperating with the heterozygous Pten mutation to promote prostate neoplasia. Both genes have been shown to have a role in protein translation. Using shRNA in prostate organoids, we validated Bzw2, demonstrating that it can activate the PI3K-AKT pathway when its gene expression is reduced. Interestingly, we found that Bzw2 action was context-specific and that lower expression levels can promote tumour formation in some circumstances – in contrast with what the existing literature has shown.
Our analysis of patient data found that EIF5A2 is frequently amplified and associated with higher gene expression in tumours, including prostate cancer. Our validation studies also showed that Eif5a2 activated the PI3K-AKT pathway, increasing the expression of AKT in the prostate and promoting lesion formation. Finally, an analysis of human prostate cancer samples from two independent cohorts revealed a correlation between increased Eif5a2 expression and upregulation of a PI3K-AKT pathway gene signature. Importantly, organoids with high levels of EIF5A2 were also found to be sensitive to AKT inhibitors.
Overall, our work suggests that the expression of Pten within prostate cells is under multiple forms of control and that it may be possible to exploit this during tumour formation. It also highlights the variety of ways in which the PI3K pathway can be activated, confirming the need to do much more research in this area. We conclude our paper by proposing that AKT inhibitors be more widely used in the clinic, particularly to treat patients with high expression levels of Eif5a2.
Written by:
- Amanda Swain, Group Leader, Division of Cancer Biology, Institute of Cancer Research, London, UK
- Jeffrey C. Francis, Senior Scientific Officer, Division of Cancer Biology, Institute of Cancer Research, London, UK