A role for splicing factore and R-loop formation in the pathophysiology of Fanconi anemia

Postdoc Assoc Pediatrics
15 York St, New Haven, CT 06510-3221

Objective: To determine the effects of R-loop formation on the FA pathway and to determine the effects of the FA pathway on transcription. Methods: siRNA transfections were used to reduce protein levels of splicing factors.
Survival assays were performed to determine cell sensitivity to mitomycin C (MMC). Chromatin extractions were performed to determine chromatin loading of proteins. Immunoprecipitations were used to determine protein-protein interactions. Results: Previous studies in our lab have shown that siRNA mediated knock down of the splicing factors ASF/SF2 and SC35 but not SRp55 result in decreased levels of FANCD2 ubiquitylation following MMC treatment and hypersensitivity to MMC. As depletion of the splicing factor ASF/SF2 has also been demonstrated to increase cellular genomic instability through the formation of increased levels of R-loop structures, we next wanted to determine what effects increased levels of RNaseH would have on activation of the FA pathway in cells with decreased ASF/SF2 expression. Interestingly, RNaseH overexpression was able to partially rescue the decreased levels of FANCD2 ubiquitylation following MMC treatment and hypersensitivity to MMC seen in cells with decreased protein levels of ASF/SF2, indicating that at least part of the FA-phenotype seen with ASF/SF2 knockdown is due to R-loop formation. As R-loop formation has been shown to occur co-transcriptionally, we next went on to examine overall transcription levels in mutant versus corrected cells by looking at epigenetic histone modifications universally accepted to signify active transcription. Remarkably we found increased levels of H2B-ub, H3K9ac, and H3K14ac in FA-A mutant cells compared to corrected cells. We went on to look at baseline levels of transcription of three housekeeping genes in mutant versus corrected cells by qPCR and again saw increased levels of transcription in the mutant cells with each gene. These differences in overall transcription led us to question whether transcription occurs normally in cells lacking an intact FA pathway. Interestingly, we discovered that proper degradation of the hyperphosphorylated, transcription competent form of RNAPII in response to DNA damage is dependent upon an intact FA pathway as cells mutant in FANCA or FANCD2 show delayed RNAPII degradation following MMC treatment. Accordingly, we also saw a decrease in the levels of RNAPII interacting with FANCD2 in chromatin after similar, short term MMC treatments. This was accompanied by FANCD2 interaction with RNAPII and BARD1 in a FANCD2 ubiquitylation-dependent manner. Conclusions: These results suggest that the FA pathway may play a part in regulating transcription, either through a connection to splicing factors or through direct interaction with the transcriptional machinery itself in addition to playing a role in the DNA damage response.