70. Knocking Down the Circular RNA ciRS-E2 Blocks Cancer Cell Proliferation Demonstrating Circular RNAs as a New Therapeutic Target

Abstract

Circular RNAs (circRNA) are a newly discovered class of abundant and well conserved non-coding RNAs (Memczak et al. and Hansen et al. Nature 2013). They are comprised of RNA sequences in which the 3’ end of a downstream exon has been backspliced to the 5’ end of an upstream exon, causing the formation of a continuous RNA loop. Thousand of different circRNAs have been identified so far in different cell types, including cancer cells, and their expression levels often reaches or exceeds many coding RNAs in the cell. A small number of circRNAs have been shown to control the activity of specific microRNAs by acting as molecular sponges, however the functions of the vast majority of circRNAs are still not known. Because the sequence of the 3’-5’ exon junction of a circRNA is unique in the transcriptome, we hypothesized that targeting this region may be used to generate RNAi vectors that can knockdown circRNAs without impacting their host gene.Using a deep-sequencing approach that we call Capture-seq and a novel mapping algorithm, we detected an abundant circRNA expressed in highly proliferative cells, which we dub circular E2 (ciRS-E2). ciRS-E2 is comprised of exon 2 of a coding gene which has circularized as a result of the 3’ end of exon 2 backsplicing to the 5’ end of the exon. Using a PCR that was specific for the backspliced junction, we found ciRS-E2 to be highly expressed in a number of cancers, including leukemia, melanoma, and ovarian.We generated two different shRNAs to target the unique 3’-5’ exon junction of ciRS-E2 and designed them to have different seed regions so that potential off-target effects could be accounted for. The shRNAs were cloned into lentiviral vectors encoding GFP (LV.GFP), and introduced into acute myelocytic leukemia or ovarian adenocarcinoma cells. More than 95% of cells were transduced. For comparison, the cells were transduced with two different control LV.GFP vectors encoding scrambled sequences. Impressively, both ciRS-E2 shRNAs knocked down ciRS-E2 by more than 80%, and there was no change in the expression of ciRS-E2's host gene's mRNA or protein. Strikingly, within 7 days of transduction there was a major loss in cells transduced with either ciRS-E2 shRNA, which did not happen with control cells. By 10 days, there were less than 20% GFP+ cells, indicating that ciRS-E2 is important for cancer cell fitness. Further analysis revealed that this was due to impaired cell proliferation as a result of the cells arresting in G0/G1. Transcriptomics analysis by RNA-seq indicated that loss of ciRS-E2 resulted in a significant decrease (P 2-fold) in the expression of a number of key oncogenes including Myc and Max.This study is one of the first to assign an important cellular function to a circRNA, and the first to identify a circRNA that controls cancer cell fitness. Importantly, we show that RNAi vectors can be generated to efficiently and specifically knock down a circRNA without impacting their host gene. circRNAs represent a potential new therapeutic target, and our work suggests that knockdown of ciRS-E2 may be used to control cancer cell proliferation.