Landscape characterization of chimeric RNAs in colorectal cancer.

Abstract

Gene fusions and their fusion products have been recognized as ideal biomarkers and drug targets for cancer. However, few recurrent gene fusions were found in colorectal cancer (CRC), despite comprehensive studies. We believe that chimeric RNAs, in the absence of chromosomal rearrangement, may represent a new repertoire of biomarkers and/or therapeutic targets in CRC. In this study, we aim to identify such recurrent chimeric RNAs, and investigate their clinical implications. To do so, we performed extensive data mining for chimeric RNAs using The Cancer Genome Atlas CRC RNA-Seq datasets. Multiple filtering criteria were applied, and the landscape of chimeric RNAs at multiple levels, from various angles, was analyzed. Eleven frequent, cancer biased chimeric RNAs were validated. The expression of RRM2-C2orf48 correlates with poor clinical outcomes, while the expression of parental RRM2 and C2orf48 correlates with positive clinical outcomes. Mechanistically, it is a product of cis-splicing between adjacent genes. Silencing of RRM2-C2orf48 resulted in reduced cellular proliferation in colon cancer cells, whereas overexpressed chimera promoted cell proliferation. These findings suggest that frequent chimeric RNAs are present in CRCs, and that chimeric RNAs may have different expression profiles and functions from parental genes, thus representing a new repertoire of biomarkers and therapeutic targets.