RNA and Gene Expression

Transcriptional gene silencing and its role in disease

Efficient DNA damage repair is a fundamental process for every living organism. The accumulation of DNA damage affects cellular viability and leads to a variety of diseases, particularly cancer. Therefore, understanding of the molecular mechanisms necessary for DNA damage repair are of great importance. DNA double-strand breaks (DSBs) are highly toxic lesions, which cause genomic instability. DSBs activate non-homologous end joining (NHEJ) and homologous recombination (HR), two major DNA repair pathways that are orchestrated by a myriad of well-described enzymatic reactions in eukaryotic cells. However, the relevance of RNA metabolic enzymes for the DNA damage response (DDR) is poorly understood. We are interested in RNA dependent DNA damage response, a new concept, which is currently perceived as not well understood. We aim to provide new convincing data using state of art facilities at the Sir William Dunn School of Pathology and dissect the molecular mechanisms of DNA damage response, leading to efficient DNA repair. In particular, we study the interplay between RNAi, transcription and Cohesin recruitment upon DNA damage. All these factors possess canonical primary functions in gene silencing, gene expression and chromosome segregation. However, within physiological concept of DNA damage they might act as the key players in DNA damage response exhibiting their non-canonical functions in processing of long endogenous RNA, antisense transcription and mediation of DNA damage repair.

Transcription facilitates sister chromatid cohesion on chromosomal arms

Cohesin is a multi-subunit protein complex essential for sister chromatid cohesion, gene expression, recombination and DNA damage repair. Although structurally well studied, the underlying determinant of cohesion establishment on chromosomal arms remains enigmatic. We discovered two populations of functionally distinct cohesin on chromosomal arms. Using single-locus specific DNA-FISH analysis in vivo, we show that topologically bound cohesive cohesin coexists with its loader Mis4-Ssl3. In contrast, cohesin independent of its loader is unable to maintain stable cohesion in fission yeast. Cohesive sites overlap highly expressed genes and transcription inhibition reduces chromatin association of cohesion proteins. Reciprocally, heat shock induction leads to de novo recruitment of cohesive cohesin. Finally, we propose that transcription facilitates cohesin loading to chromatin also in human cells. Overall, our study suggests that transcription is the key determinant of cohesive sites on chromosomal arms in eukaryotes. The main experiments are performed by Dr. Shweta Bhardwaj and the bioinformatic analysis by Dr. Margarita Schlackow.

Dicer facilitates nuclear specific regulation of alternative poly(A) site selection

Alternative Cleavage and Polyadenylation (APA) plays a crucial role in the regulation of gene expression across eukaryotes. Although APA is extensively studied, its regulation within cellular compartments and its physiological impact remains largely enigmatic. We employed a rigorous subcellular fractionation approach to compare APA profiles of cytoplasmic and nuclear RNA fractions from human cell lines. This unique approach allowed us to extract APA isoforms that are subjected to regulation and provided us with a platform to interrogate the molecular regulatory pathways that shape the APA profiles in the subcellular locations. We found that APA isoforms with shorter 3’UTRs tend to be overrepresented in the cytoplasm and they appear to be cell type specific events. We also show that nuclear retention is partly a result of incomplete splicing and contributes to the observed cytoplasmic bias of transcripts with short 3’UTRs. We further show that endoribonuclease III, Dicer, contributes to the establishment of subcellular APA profiles not only by expected cytoplasmic miRNA mediated destabilisation of APA mRNA isoforms, but also by nuclear regulation. Experiments and analysis are performed by Dr. Kaspar Burger in collaboration with Prof Andre Furger.

Transcriptional regulation of the ERBB2 amplicon in breast cancer cells

Over-expression of human epidermal growth factor receptor type 2 (ERBB2) occurs in almost 30% of invasive breast carcinomas. Breast cancer cells, which over-express the ERBB2, result in hyper proliferation through the RAS-MAPK pathway and inhibit cell death through the mTOR pathway. Thus, strategies to target HER2 play a significant role in treatment of breast cancer. We are analysing how is the expression of the ERBB2 amplicon regulated on transcriptional level. High levels of de novo cohesin are recruited to ERBB2 amplicon in invasive breast cancer cells. We investigate whether and how cohesin might participate of ERBB2 transcriptional regulation.

The role of human nuclear Dicer in DNA damage response

The DNA damage response (DDR) is crucial for the maintenance of genome stability. A number of mechanisms exist to recognize and repair DNA lesions. The homologous recombination and non-homologous end-joining (NHEJ) pathways repair double strand breaks (DSBs). Recently, a new class of small regulatory RNA has been discovered in higher eukaryotes: site-specific DNA-damage RNA (DDRNA). DDRNA originate from both strands of DSBs and function at sequences that are in close proximity to these lesions. Drosha and Dicer processing of dsRNA precursors may mediate the maturation of DDRNA. The activation of major DNA repair factors, such as ATM is diminished upon Drosha or Dicer deletion, which underscores the connection between DDRNA, nuclear RNAi factors and DSB repair. We are currently investigating how Dicer and Drosha function in DNA damage response. Experiments and analysis are performed by Dr. Kaspar Burger.

Human nuclear Dicer is processing misfolded tRNA

Dicer is a type III endoribonuclease and a core part of the RNA interference (RNAi) machinery. RNAi involves processing of double-stranded RNA into small precursors that can have diverse effects but typically involve a repressive effect on a target nucleic acid. We have shown previously (White et al., NSMB, 2014) that Dicer also acts in nucleus to process endogenous double stranded RNA. Transfer RNAs are 75 nt RNAs that constitute 15% of total cellular RNA. There are 500 tRNA genes (tDNAs) in the human genome, of which only 20% are unique. After initial transcription as precursors by RNA polymerase III (pol III), 50 and 30 leader sequences are removed from pre-tRNAs by RNase P and RNAse Z, respectively. A large number of nucleobase modifications are known to occur in tRNAs some of which are thought to aid in the folding into the functional cloverleaf structure. The present study builds on Dicer association with chromatin in human cells, focusing on Dicer occupancy of transfer RNA (tRNA) genes. Our bioinformatics analysis shows that Dicer binds to actively transcribed tRNA, most likely through dsRNA. Knockdown of Dicer does not perturb transcription of selected tDNAs, although a cryptic alternative tRNA structure is observed upon knockdown of Dicer, suggesting a role for Dicer in tRNA quality control.

Classification and functional annotation of endogenous-siRNAs and other small RNAs

Over the course of the last 10 years the role of small RNAs (sRNA) in the regulation of eukaryotic cell biology and gene expression has become well established. However, while microRNAs (miRNAs) are now recognised as a relatively ubiquitous method by which cells control RNA and protein levels, it has recently become clear that endogenous siRNAs (endo-siRNAs) may also play a significant role in the regulation of expression in mammals. In order to better understand this class of molecule and their relationship to other sRNAs it is imperative that tools be developed. Such tools would allow the community to distinguish endo-siRNAs from amongst other sRNA classes in Next generation sequencing data (NGS). The identification of more widespread endo-siRNA expression has coincided with evidence that mammalian endo-siRNAs may play a key role in orchestrating histone modifications to silence transcription and affect alternative splicing. Tools that can predict this process will help guide research by the wider scientific community and significantly improve our knowledge of the regulatory effect of these sRNAs and their contribution to disease states. Experiments and analysis are performed by Dr. Matthew Davis in collaboration with Dr. Anton Enright.