RNA polymerase II (Pol II) regulation during early elongation has emerged as a regulatory hub in the gene expression of multicellular organisms. Prior research links the BRD4 protein to this control point, regulating the release of paused Pol II into productive elongation. However, the exact roles and mechanisms by which BRD4 influences this and potentially other post-initiation regulatory processes remain unknown. This study combines rapid BRD4 protein degradation and multi-omics approaches, including nascent elongating transcript sequencing (NET-seq), to uncover BRD4’s direct protein functions.
Applying NET-seq in comparative studies required experimental adaptations. First, analyses with spiked-in mouse cells proved essential for reliable normalization. Second, the study identified a disproportional enrichment of a chromatin-associated RNA class as NET-seq’s major limitation. Incorporating an additional enrichment step solved this problem and significantly increased Pol II coverage.
The resulting high-sensitivity NET-seq method confirmed BRD4’s proposed role in early elongation by revealing a global defect in Pol II pause release upon BRD4 degradation. Observations from proteomics and chromatin immunoprecipitation followed by sequencing (ChIP-seq) experiments suggest that the failed recruitment of Pol II-associated factors (PAF) causes an assembly defect of a competent elongation complex.
Interestingly, the elongation defect also affected transcribed enhancers. Pol II occupancy increased in a region proximal to the enhancer center, strikingly similar to the impaired Pol II pause release at genes. An integrated multi-omics analysis that included genome-wide 3D genome information revealed reduced interactions between these enhancers and other regulatory regions.
Another unexpected result was the widespread Pol II readthrough transcription quantified by the developed readthrough index, revealing an apparent transcriptional termination defect. The implementation of long-read nascent RNA-sequencing (nascONT-seq) combined with a 3’-RNA cleavage efficiency test detected impaired 3’-RNA processing. Notably, those 3’-RNA cleavage defects correlated with the observed termination defects. A potential explanation is the BRD4-dependent recruitment of general 3’-RNA processing factors to the 5’-control region. These observations start to establish regulatory links between 5’ and 3’ control that require further validation. Overall, the results indicate a general BRD4-dependent 5’ elongation control point required for 3’-RNA processing and termination.
