SF3B3 / SF3B5 form a metazoan specific transcription module of the U2 snRNP that coordinates Pol II elongation in a splicing independent manner

Technical Summary: SF3B3/SF3B5 as a Metazoan-Specific Transcription Module of the U2 snRNP

Problem Statement
Co-transcriptional splicing is a conserved feature of eukaryotic gene expression, yet the functional nature of the interplay between the transcription machinery and the spliceosome remains difficult to establish. While it is well-documented that splicing factors assemble onto the RNA Polymerase II (Pol II) C-terminal domain (CTD) and that global inhibition of splicing (e.g., via Pladienolide B or U1/U2 snRNP inhibition) disrupts transcription, it remains unclear whether splicing factors directly regulate transcription or if observed phenotypes are indirect consequences of uncoupled co-transcriptional activities. Specifically, the extent to which structural components of the U2 snRNP complex influence Pol II processivity independent of their canonical splicing roles is unresolved.

Methodology
The authors employed a multi-faceted approach combining real-time endogenous gene expression monitoring with high-throughput CRISPR/Cas9 screening and orthogonal protein degradation strategies:

• Reporter System: An endogenous c-MYC locus in K562 cells was engineered to express mNeonGreen (MYC-mNG), serving as a real-time readout for transcriptional activity.
• CRISPR Screens: Two distinct screens were performed: (1) a "temporal dependency" screen targeting 1,145 chromatin-related factors to monitor MYC-mNG loss over time, and (2) a "coactivator de-sensitisation" screen designed to identify factors that, when lost, overcome transcriptional repression induced by inhibitors of CBP/P300 (A-485), BET bromodomains (I-BET151), or ABL1 (Nilotinib).
• Orthogonal Degradation: To validate findings and study acute effects, the authors generated endogenous dTAG knock-in cell lines for SF3B3, SF3B1, and SF3B5, allowing for rapid, inducible protein degradation.
• Transcriptomics and Genomics: The study utilized multiple sequencing modalities to dissect transcriptional dynamics:
  • TT-seq (Transient Transcriptome Sequencing): To quantify nascent RNA synthesis.
  • PRO-seq (Precision Run-On Sequencing): To map the precise location and density of engaged Pol II.
  • Nanopore Long-Read Sequencing: To assess global splicing efficiency and intron retention in nascent transcripts.
  • ChIP-seq: To map the chromatin occupancy of Pol II (total, pSer5, pSer2), TBP, and transcriptional kinases (CDK9, CDK12, CDK13) and complexes (Integrator, PAF1c).
• Proteomics: RNA Pol II immunoprecipitation followed by mass spectrometry (IP-MS) and whole-cell proteomics were used to define protein interaction networks and stability dependencies.
• Structure-Function Analysis: Ectopic expression of wild-type and truncated SF3B3 mutants (specifically targeting the C-terminal domain) was used to dissect domain requirements for function.

Key Contributions and Results

• SF3B3 as a Transcriptional Regulator: CRISPR screens identified SF3B3, a subunit of the U2 snRNP, as a common dependency required for the maintenance of transcriptional repression following co-activator inhibition. Loss of SF3B3 phenocopied the reactivation of MYC expression even in the presence of potent co-activator inhibitors.
• Splicing-Independent Mechanism: Contrary to the effects of SF3B1 degradation or Pladienolide B treatment (which cause global splicing defects and transcriptional arrest), acute degradation of SF3B3 resulted in minimal intron retention and preserved splicing efficiency, as confirmed by Nanopore sequencing and intron inclusion PCR. This demonstrates that SF3B3's role in transcription is largely independent of its splicing function.
• Defect in Pol II Processivity: PRO-seq and TT-seq data revealed that SF3B3 loss causes "promoter proximal escape," characterized by an accumulation of Pol II immediately downstream of the Transcription Start Site (TSS) and a subsequent tapering of Pol II density toward the 3' end of genes. This phenotype is most pronounced in short genes (leading to their upregulation) and results in the downregulation of long genes due to reduced processivity.
• Coordination of Kinases and Complexes: ChIP-seq and IP-MS data demonstrated that SF3B3 is required for the stable chromatin occupancy of transcriptional kinases (CDK9, CDK12, CDK13) and the recruitment of the Integrator and PAF1 complexes to Pol II. In the absence of SF3B3, the association of these elongation factors with Pol II is significantly reduced, disrupting the balance necessary for controlled pause release and processive elongation.
• The SF3B3/SF3B5 Submodule: The study identified a metazoan-specific interaction where SF3B3 physically associates with SF3B5 via a disordered C-terminal tail containing a conserved 18-amino acid sequence. This interaction is essential for the stability of SF3B5. Crucially, the loss of SF3B5 mirrors the transcriptional defects of SF3B3 loss (promoter escape, reduced processivity) without affecting U2 snRNP assembly or splicing, whereas loss of the primary U2 subunit SF3B1 disrupts the entire complex.
• Domain Specificity: Truncation of the SF3B3 C-terminal tail (specifically the terminal 18aa) phenocopied the loss of SF3B3 function, preventing the physical association with SF3B5 and disrupting the transcriptional module, despite the protein retaining its ability to incorporate into the U2 snRNP core.

Significance
The paper claims to uncover a critical, splicing-independent role for the U2 snRNP components SF3B3 and SF3B5 in modulating RNA Pol II activity. The authors propose that this metazoan-specific submodule has evolved to coordinate the assembly and function of transcriptional kinases and co-activator complexes (PAF1c, Integrator) at the chromatin template. This finding challenges the dogma that structural components of co-transcriptional complexes possess uniform functions; instead, it suggests that specific subunits within these complexes can evolve distinct regulatory roles, in this case, ensuring optimal Pol II licensing and processivity to coordinate the efficiency of transcription and splicing. The study highlights that the U2 snRNP is not merely a splicing machine but also a key regulator of the transcriptional elongation machinery in metazoans.