Bioinformatics analysis of multi-omics data elucidates U2 snRNP function in transcription

Abstract

Transcription by RNA polymerase II (Pol II) is an important step in cell function and regulation. Pol II transcription has been shown to be coupled to pre-mRNA splicing, but the underlying mechanisms remain poorly understood. Co-transcriptional splicing requires the assembly of a functional spliceosome on nascent pre-mRNA, but whether and how this influences Pol II transcription remains unclear. To investigate this, we used a human erythroleukemic cell line and performed transient transcriptome sequencing (TT-seq) and mammalian native elongating transcript sequencing (mNET-seq) upon fast inhibition of U2 snRNP function. We further studied how the positive transcription elongation factor b (P-TEFb) recruitment is related to the Pol II pause duration, using chromatin immunoprecipitation and sequencing (ChIP-seq) of the PTEF-b kinase cyclin T1 (CycT1) upon U2 snRNP inhibition. I performed a bioinformatics analysis of the different datasets generated for this study and two additional published datasets. I also conducted a multiomics analysis combining TT-seq and mNET-seq data to calculate and quantify transcription kinetic parameters such as Pol II productive initiation frequency, pause duration and elongation velocity. Here we show that inhibition of pre-mRNA branch site recognition by the spliceosome component U2 snRNP leads to a widespread and strong decrease in new RNA synthesis from human genes. We further show that inhibition of U2 snRNP function increases the duration of Pol II pausing in the promoter-proximal region, impairs recruitment of the pause release factor P-TEFb, and reduces Pol II elongation velocity at the beginning of genes. Our results indicate that efficient release of paused Pol II into active transcription elongation requires the formation of functional spliceosomes and that eukaryotic mRNA biogenesis relies on positive feedback from the splicing machinery to the transcription machinery. We further show that the fast U2 snRNP inhibition affects the expression of genes related to RNA synthesis and it is not related to stress response genes. Our new multi-omics approach for the calculation of Pol II elongation velocity can be applied to further study the impact on Pol II kinetics regarding different splicing and transcription factors. This is of great importance to unravel the mechanisms behind Pol II transcription and splicing and understand how the disruption of this regulation leads to several pathological cell phenotypes and diseases.