A Computational Analysis of Alternative Splicing across Mammalian Tissues Reveals Circadian and Ultradian Rhythms in Splicing Events

Abstract

Mounting evidence points to a role of the circadian clock in the temporal regulation of post-transcriptional processes in mammals, including alternative splicing (AS). In this study, we carried out a computational analysis of circadian and ultradian rhythms on the transcriptome level to characterise the landscape of rhythmic AS events in published datasets covering 76 tissues from mouse and olive baboon. Splicing-related genes with 24-h rhythmic expression patterns showed a bimodal distribution of peak phases across tissues and species, indicating that they might be controlled by the circadian clock. On the output level, we identified putative oscillating AS events in murine microarray data and pairs of differentially rhythmic splice isoforms of the same gene in baboon RNA-seq data that peaked at opposing times of the day and included oncogenes and tumour suppressors. We further explored these findings using a new circadian RNA-seq dataset of human colorectal cancer cell lines. Rhythmic isoform expression patterns differed between the primary tumour and the metastatic cell line and were associated with cancer-related biological processes, indicating a functional role of rhythmic AS that might be implicated in tumour progression. Our data shows that rhythmic AS events are widespread across mammalian tissues and might contribute to a temporal diversification of the proteome.