Investigating prokaryotic transcriptomes and the impact of crosstalk between noncoding RNA and messenger RNA interactions
Thesis DisciplinePlant Biotechnology
Degree GrantorUniversity of Canterbury
Degree NameDoctor of Philosophy
Prokaryotes have a complex noncoding RNA (ncRNA) based regulatory system, resembling that of eukaryotes. Recent transcriptomics studies also point out the abundance of highly expressed uncharacterized RNAs in archaea and bacteria. However, despite the recent advances indicating the prevalence of ncRNAs in prokaryotes, it is still unknown to what extent these uncharacterized transcripts are functional. Therefore, we have proposed a phylogeny informed approach to design new RNA sequencing (RNAseq) experiments, which increases the information harnessed from transcriptome data for ncRNA detection. Many regulatory ncRNAs engage in RNARNA interactions, where RNA molecules bind to form a duplex. Predictions of true targets for an RNA enables a successful functional characterization, these can be estimated by bioinformatics methods. However, the algorithms developed to date are imperfect and it is an open question as to which ones perform well and whether these can be improved upon. Towards this goal we performed a computational benchmark study to find reliable algorithms for RNARNA interaction prediction. We found that energy based methods, which include the accessibility of interaction regions, are currently the most accurate. Many ncRNAs, including housekeeping ncRNA genes, are highly expressed. The abundances of interacting RNA molecules enable RNARNA duplex formation. In chapter IV we explore the impact of high abundance RNAs on protein expression due to crosstalk RNARNA interactions between mRNAs and ncRNAs. With extensive RNARNA interaction predictions we reveal that RNA avoidance is an evolutionarily conserved phenomenon among prokaryotes, which means that core mRNAs have evolved to avoid crosstalk interactions with abundant ncRNAs. Our predictions also reveal that RNA avoidance may influence protein expression. To test this, we investigated the stability of interactions between mRNAs and core ncRNAs. These predictions show that the RNA avoidance influences the final protein abundances. In conclusion, the primary aims of this study are to investigate the prokaryotic transcriptome for novel ncRNA genes and examine the effects of crosstalk RNA interactions. We present a method to increase information gained from transcriptome in prokaryotes for ncRNA identification. We also present the most comprehensive benchmark of RNARNA interaction prediction algorithms to date. Lastly, we introduce and test a ‘RNA avoidance hypothesis’ that shows the influence of crosstalk RNA interactions on protein expression in bacteria.