Description:

In many eukaryotic organisms, splicing and transcription are both physically and functionally coupled. Exciting recent work from the Neugebauer lab has documented for the first time the timing of splicing relative to transcription for a few hundred transcripts in budding yeast, with evidence that splicing can occur quickly – as soon as an intron emerges from the exit channel of RNA polymerase II. A current challenge is to extend such an analysis of co-transcriptional splicing on a genome-wide scale to humans, where the genome encodes three orders of magnitude more introns and where alternative splicing undergoes intricate regulation; indeed, the timing of splicing relative to transcription has been invoked in some cases to rationalize splice site choice in humans, but this notion remains to be tested rigorously. To overcome this challenge in humans, we have implemented a genome-wide approach to study co-transcriptional splicing from the perspective of nascent lariat intermediates by co-transcriptional lariat sequencing (CoLa-seq). 

Through observation of the timing of splicing for thousands of introns in human cells, we have found evidence that splicing can be as fast as in budding yeast. A dominant mode of splice site recognition in humans is thought to require not only the 5’ and 3’ splice sites of an intron but also the 5’ splice site of the downstream intron, a mechanism termed exon definition. Nevertheless, we find evidence that a majority of introns can undergo splicing before the downstream 5’ splice site is even transcribed. Introns have also been subdivided into those that require the 3’ splice site for lariat intermediate formation and those that do not. Consistent with the latter class, we have observed lariat intermediate formation on some introns before the 3’ splice site has even emerged from the exit channel of RNA polymerase II. Previous studies have documented out-of-order splicing, in which downstream introns splice before upstream introns. We now report the observation of concurrent splicing, in which adjacent introns splice at the same time, hinting at coordination of splicing neighboring introns. Lastly, we have unveiled the first direct evidence that the competition between alternative 3’ splice sites is enabled by slower splicing of the upstream 3’ splice site. Together, these findings establish that significant features of co-transcriptional splicing in humans can be monitored by assaying nascent lariat intermediates. 

Jonathan Staley, PhD

University of Chicago

Professor, Cell and Molecular Biology