A tail of order from chaos : the impact of RNA polymerase II CTD length on liquid-liquid phase separation, genome architecture, and transcriptional dynamics

The majority of eukaryotic genes are transcribed by RNA polymerase II (Pol II), a multiprotein complex whose largest subunit has a flexible, tail-like C-terminal domain (CTD). The CTD is intrinsically disordered, adopting a range of rapidly shifting conformations instead of a fixed 3D structure. This facilitates multivalent interactions that promote a process called liquid-liquid phase separation (LLPS), where proteins and other molecules cluster together to form condensates. These condensates can act as membraneless organelles, which play an increasingly recognised role in coordinating spatiotemporal cellular organisation and regulating processes such as transcription. This thesis investigates cells expressing mutant Pol II variants with different CTD lengths, to explore how its intrinsic disorder drives the highly ordered 3D shape of the genome and the tightly regulated dynamics of gene expression. Single-cell RNA sequencing revealed transcriptional changes associated with these mutants, including the upregulation of genes involved in LLPS. Longer CTDs were shown to enhance LLPS, driving the expression of genes such as CRYAB and CTSL, which encode proteins that suppress aggregation of LLPS-prone proteins. These findings suggest that cells adaptively upregulate protective mechanisms to counteract the detrimental effects of excessive LLPS. This was supported by observations of how mutant Pol II expression levels affect transcription. Genes with expression levels strongly correlated with the mutant transcript included MALAT1, HNRNPU, and DDX5, all of which interact with the Pol II protein directly or indirectly. Their upregulation likely reflects feedback mechanisms compensating for LLPS-induced sequestration. LLPS-related genes were significantly enriched in the sets of genes whose expression was correlated with the exogenous tags, and therefore the mutant RPB1. There was also a significant depletion of LLPS-related genes in those genes whose expression is negatively correlated with Dendra2 or HaloTag, compared to genes with no significant correlation. These observations are consistent with the hypothesis that higher mutant Pol II expression leads to an increase in global LLPS which triggers increased expression of LLPSrelated genes. To further investigate the LLPS of Pol II in vivo, an improved version of the lightinducible CasDrop system was developed. Modifications included replacing suboptimal promoter sequences with more stable alternatives, introducing new fluorophores to facilitate simultaneous imaging of mutant Pol II, and optimising experimental conditions for droplet formation. Fluorescent microscopy confirmed that Pol II could reversibly condense into CasDrop droplets, enabling dynamic studies of its LLPS. The impact of LLPS on chromatin architecture was examined using chromatin interaction analysis by paired-end tag sequencing, which revealed that enhanced LLPS correlated with increased interchromosomal interactions, suggesting a role for phase separation in reorganising genome topology by bringing together distant regulatory elements. Microscopy revealed that the observed negative colocalisation of CasDrop condensates and DNA supports prior findings that LLPS reshapes the spatial organisation of chromatin. Together, these findings contribute to our understanding of the intricate relationship between LLPS and its regulatory effects on the spatial organisation of the genome and the dynamics of gene expression.