Abstract
Rho is a hexameric helicase that hydrolyses ATP to translocate along nascent mRNA transcripts and terminate the transcription of specific bacterial genes. Usually, Rho has three conserved domains (N-terminal, RNA-binding, and ATPase). Some species, including mycobacteria, might have Rho with a functional insertion region. Transcription termination plays an essential role in the regulation of bacterial gene expression, and it is important to study how the structure of Rho varies among bacteria. However, the last broad analysis of Rho in bacteria was performed more than ten years ago.
This thesis provides a modern compilation and analysis of the currently known sequence diversity and phylogeny of Rho. Standard and developed Hidden Markov models for each domain were used in a high-quality genome dataset of 2,730 bacterial species. The Rho sequences were classified based on the presence and position of the additional regions (Types 1-4). The typical Escherichia coli like Rho (Type 1) was detected in almost half of the species analysed and atypical forms of Rho were found in most of the other species. The main domains were highly conserved, but I also described novel variations in the bicyclomycin binding pocket and a double RNA-binding domain in certain species. The Rho additional regions have biased amino acid compositions comprised four significant motifs. Furthermore, these regions are predicted to provide extra sites to RNA-binding and to be intrinsically disordered, undergo phase separation, or have prion-like behaviour.
Liquid-Liquid Phase Separation (LLPS) is an alternative mechanism of compartmentalisation in cells and is associated with protection against stress and metabolic regulation. LLPS systems are not well characterised in bacteria and there is only one study showing this ability of Rho. Furthermore, I showed experimentally that Mycolicibacterium smegmatis Type 2 Rho forms droplets both in vitro and in vivo by LLPS. The additional regions of slow and fast-growing Mycobacteria also form condensates, and this process is associated with RNA. Then, I demonstrated that the ability of M. smegmatis cells to tolerate acid stress is consistent with the additional regions of Rho that undergo LLPS. In contrast, the engineered Rho mutants lacking these additional regions were neither able to tolerate acid stress nor form droplets.
Overall, this study revealed the diversity of atypical Rho factors among bacteria. The Rho additional regions are likely to provide further cellular roles. One example that I showed was the acid tolerance of mycobacteria due to the ability of its atypical Rho to undergo LLPS.