You are here

Jean Beggs

Co-workers:

Vahid Aslanzadeh, David Barrass, Jim Brodie, Susana De Lucas, Eve Hartswood, Bella Maudlin, Gonzalo Mendoza-Ochoa, Ema Sani, Edward Wallace
Beggs Lab website

Regulation of splicing and functional links between splicing and transcription

Transcription and RNA splicing are at the centre of gene expression in eukaryotes, controlling gene expression levels and removing introns from the primary transcripts. The mechanisms and machineries involved in both transcription and RNA splicing are highly conserved throughout eukaryotes, and the budding yeast Saccharomyces cerevisiae makes an excellent model system, permitting the application of genetic approaches in parallel with molecular studies. In addition to investigating the functions and molecular interactions of yeast splicing factors, we are interested in links between RNA splicing and other metabolic processes, especially transcription. Our approaches include: quantitative RT-PCR, ChIP-seq, RNA-seq, biochemical analyses of splicing and molecular genetics. 

To investigate links between transcription and splicing, we performed high resolution kinetic ChIP experiments to follow the recruitment of RNA polymerase II (Pol II) and splicing factors to inducible reporter genes. We found that, soon after the initiation of transcription, Pol II pauses transiently near the 3’ end of an intron in response to splicing. The carboxyterminal domain (CTD) of the paused Pol II large subunit is hyper-phosphorylated on serine 5 and on serine 2, suggesting regulation through phosphorylation (Alexander et al., 2010). We propose that this represents a novel splicing–dependent transcriptional checkpoint that may be associated with the quality control activities of splicing ATPases, such as Prp5 (Figure 1A). This hypothesis is supported by the finding that mutations (e.g. prp5-1) that block pre-spliceosome formation cause a transcription defect, with pSer5 Pol II accumulating on introns (Figure 1B). In the prp5-1 mutant, the U2-associated Cus2p is thought to remain in a defective transcription/splicing complex, but deletion of CUS2 suppresses the transcription defect. We propose that Cus2p is a potential checkpoint factor that signals the status of prespliceosome formation to the transcription machinery (Chathoth et al., 2014). In future work we will characterise these factors and investigate the mechanism by which splicing affects transcription and how coupling these processes benefits gene expression.

To facilitate studies of how other processes affect splicing, we have developed a procedure for labelling RNA in vivo with 4-thio-uracil for very short periods, followed by biotinylation of the thiolated RNA and its affinity purification. Reverse transcription then produces cDNA of the newly synthesised transcripts, which can be analysed by deep sequencing. In this way, short-lived precursor RNAs can be sequenced during a brief time course of labelling (Barrass et al., 2015). Comparing intron and exon reads for each transcript allows a comparison of the rate of splicing of different pre-mRNAs (Figure 2).
 

Selected publications:

Alexander, R., Innocente, S., Barrass, J.D. and Beggs, J.D. (2010). Splicing causes RNA polymerase pausing in yeast. Mol Cell 40, 582-593.

Chathoth, K., Barrass, J.D., Webb, S. and Beggs, J.D. (2014) A splicing-dependent transcriptional checkpoint associated with pre-spliceosome formation. Mol Cell 53, 779-790.

Barrass, J.D., Reid, J.E.A., Huang, Y., Hector, R.D., Sanguinetti, G., Beggs, J.D. and Granneman, S. (2015) Transcriptome-wide RNA processing kinetics revealed using extremely short 4tU labeling. Genome Biol 16:282.
 


 

1. Model for a splicing-coupled transcriptional checkpoint associated with pre-spliceosome formation in wild-type cells (A) or in a prp5-1 mutant (B) (Alexander et al., 2010; Chathoth et al., 2014).
 

2. RNA-seq analysis of splicing kinetics. RNA labelled in vivo for 1.5, 2.5 or 5.0 min with 4-thio-U was affinitypurified, reverse transcribed and sequenced along with steady state RNA. Shown are genome browser shots
of sequence reads for three intron-containing genes. With time, the intron reads relative to exon reads decline towards the steady state level. Faster splicing transcripts reach the steady state level faster than slower splicing transcripts. (Barrass et al., 2015)