ERF-202: A Hidden Tumor Suppressor Isoform Linked to Prostate Cancer Progression

Most genes are represented by different mRNA isoforms which in turn may produce different proteins with similar or opposing functions. While this diversity is known for decades, relatively little is known about specific isoforms and their biology. Here, we interrogate a harmonized prostate cancer RNA sequencing compendium with 1140 clinical samples for thus far hidden, isoform-specific transcript regulations and their associations with disease progression. We identify a shorter isoform of the tumor suppressor gene ERF (ERF-202) which is drastically reduced in expression with disease progression. In contrast to the canonical isoform of ERF (ERF-201), ERF-202 has a truncated ETS DNA binding domain and localizes not to the nucleus but rather to mitochondria. Its function and association with cancer growth and progression remains elusive.

To this end, we have secured a tumor suppressive function of ERF-202 in a xenograft tumor model in vivo. The outlined research project will further interrogate the regulation, the function, and emerging therapeutic opportunities linked to this hidden and frequently downregulated tumor suppressor. More specifically, we aim to elucidate in Aim 1 why ERF-202 expression is silenced during disease progression by investigating genetic and epigenetic modification as well as transcription bindings sites at the second promoter controlling ERF-202 expression. In Aim 2, we will address the question on how ERF-202 operates at mitochondria by determining its molecular environment, its influence on oxidative phosphorylation, and more general influences on cell metabolism. In Aim 3, we will investigate if and how the loss of ERF-202 affects resistance to androgen deprivation in vivo and systematically search for synthetic lethal interactions related to ERF-202 loss which may be exploited therapeutically. For the latter, we will use a CRISPR-based loss-of-function library capable of eliminating functional redundancies in the human genome by co-deleting gene paralogues.

The project may establish a paradigm for a cancer-relevant gene whose protein isoforms (i.e., ERF-201 & -202) have unrelated functions in different cellular compartments. Emerging synthetic lethal interactions may open new therapeutic avenues for exploiting vulnerabilities related to ERF-202 loss-of-function. Finally, the project will provide with the Prostate Cancer Atlas (www.prostatecanceratlas.org), a user-friendly web resource to investigate isoform-specific associations in an unbiased manner for the larger research community (Open Science).