Final Report Summary - PRP8CHROMATIN (Molecular analysis of the mechanisms linking co-transcriptional splicing with chromatin) The expression of genes is tightly regulated in response to various cues (i.e. changes in the environmental conditions, exposure to various chemicals etc.). One of the major directions in molecular biology research is to understand the mechanisms that govern changes in gene expression. Transcription is the first step in gene expression, in which a gene is copied into ribonucleic acid (RNA). In higher organisms many genes on the deoxyribonucleic acid (DNA) are segmented with coding and non-coding sequences, named as exons and introns, respectively. In the transcribed RNA the introns are excised and exons are joined together in a process known as splicing and the final product undergoes 3'-end cleavage and polyadenylation. Recent studies have shown that the transcription in the opposite, antisense, direction is also widespread in the eukaryotic genomes. We are focused on understanding the mechanism of regulation of Flowering Locus C (FLC) gene expression, which encodes an important flowering regulator in a model plant Arabidopsis thaliana as well as in commercially important Brassicaceae species. FLC expression is regulated by long antisense non-coding RNAs collectively named COOLAIR, which are expressed from the FLC locus. There are two major forms of COOLAIR resulting from alternative polyadenylation - termed proximal and distal COOLAIR. The Arabidopsis RNA binding protein FCA strongly represses the expression of FLC. A forward mutagenesis screen had been designed to identify components required for FCA-induced repression of FLC expression. This has revealed that FCA function required the conserved 3' end processing factors CstF64 and CstF77 to promote 3'-end processing of the proximal COOLAIR. It also revealed the requirement of the histone lysine demethylase FLD for reduction of the coding, sense FLC transcription. Another component that was identified in the forward mutagenesis screen was characterised as a central component of the spliceosome, PRP8. The focus of my project was to characterise the interactions between the sense and antisense transcription of FLC and understand how PRP8 impacted this interaction. Complete inactivation of the PRP8 gene is lethal in Arabidopsis thaliana. Such mutations, characterised by sus-2 alleles lead to embryonic arrest and subsequent death of the seedlings. The prp8 mutation discovered in the FLC regulator forward mutagenesis screen was not lethal, just delayed in flowering time. Assuming that this hypomorphic (partial effect) mutation in PRP8 might possibly affect splicing efficiency, a comprehensive assessment of splicing efficiency of the sense and antisense introns of the FLC locus was performed. The hypomorphic prp8 mutation specifically reduced splicing efficiency of the short version of the antisense intron and reduced formation of the short 'proximal' antisense transcript. Thus, in the wild-type context PRP8 appears to promote splicing of the short antisense intron, which promotes use of the proximal antisense polyadenylation site and this represses sense FLC transcription. In order to assess the extent of COOLAIR regulation on FLC transcription we transformed Arabidopsis FLC mutant plants with a construct that expresses sense FLC, but was disrupted in COOLAIR expression. Interestingly, the transformants exhibited increased levels of the unspliced sense FLC and the prp8 mutation no longer affected FLC expression. These data indicates that antisense transcription was important for the PRP8 repression of FLC expression. The next step was to directly assess the effect of splicing and polyadenylation of the proximal COOLAIR on the expression of FLC. We focused on the intron of the proximal antisense transcript that was inefficiently spliced in the prp8 mutant. We prevented use of the first intron 3' splice site in this transcript by mutating the dinucleotide AG to AA. This construct was introduced into FLC mutant plants that did or did not carry the prp8 mutation. We found that the 3' splice site mutation was sufficient to significantly increase levels of FLC expression. Moreover, this mutation was associated with lower levels of proximal antisense polyadenylation and the PRP8 repression of FLC expression was overcome. In summary, we revealed the mechanism by which PRP8 represses sense FLC expression and showed it involved altered splicing and polyadenylation of an antisense transcript that in turn regulated sense strand transcription. Given that pervasive transcription has been detected in many genomes this work has the potential to inform regulation of transcription in many organisms. In addition to detailed analysis of the role of PRP8 on FLC regulation I have also initiated experiments to monitor the localisation and dynamics of different FLC transcripts in the cell. Transgenic Arabidopsis plants expressing unspliced and spliced tagged FLC RNAs were generated. The tags chosen were the MS2 and BoxB repeats system, but the latter gave RNA that would not function correctly. We are now visualising the RNAs with an MS2-GFP protein that binds to the FLC-MS2 transcripts. In addition to visualising the cellular localisation of the FLC transcripts we plan to use the tagged RNAs in proteomic studies to comprehensively analyse protein components that interact with FLC RNAs. Related documents Final Report - PRP8CHROMATIN (Molecular analysis of the mechanisms linking co-transcriptional splicing with chromatin)
