Investigating PAX6 and SOX2 dynamic interactions at the single molecule level in live cells
The abundance of transcription factor (TF) molecules in the nuclei of eukaryotic cells are in the range of thousands. However, the functional binding sites of most TFs lie in the range of hundreds. This suggests that there is a surplus of the number of molecules for many TFs, relative to their binding sites at any given time. Nevertheless, precise TF levels are instrumental for normal development and maintenance, with haploinsufficiency (namely lowering the dosage of a TF by half) being a hallmark of many TF-related human developmental disorders. Qualitative methods assessing TF binding such as chromatin immunoprecipitation, provide static information, from fixed cell populations and so fail to provide insight into TF dynamic behaviour. Live-cell imaging methodologies such as Fluorescence Correlation Spectroscopy (FCS) offer the ability to measure kinetics of binding to chromatin, protein-protein interactions, absolute concentrations of molecules and the underlying cell-to-cell variability. SOX2 and PAX6 TFs exhibit haploinsufficiency in humans. Heterozygous point mutations, deletions or insertions in these genes can lead to a plethora of abnormal ocular developmental disorders (e.g. coloboma, aniridia, microphthalmia, anopthalmia). SOX2 encodes a high-mobility group (HMG) domain-containing TF, essential for maintaining self-renewal of embryonic stem cells and is expressed in proliferating central nervous system (CNS) progenitors. PAX6 contains two DNA binding domains; a PAIRED domain (PD) and a homeodomain (HD). Both DNA binding domains present in PAX6 (PD and HD) can function either jointly, or separately, to regulate a plethora of genes implicated in the development and maintenance of the CNS, the eye and the pancreas. Despite existing genetic and phenotypic evidence, it remains unclear how PAX6 and SOX2 influence each other at the molecular level and how sensitive their stoichiometry is during ocular development. In this thesis I investigated the dynamic interplay between PAX6/SOX2 and chromatin in live cells, at the molecular level. I compared wild-type protein function with pathogenic missense variants using advanced fluorescence microscopy techniques and assessed how these mutations quantitatively and qualitatively affected molecular behaviour. My results showed that both SOX2 and PAX6 pathogenic missense mutants display differential subnuclear localisation, as well as altered protein-protein and protein-chromatin interactions, linking molecular diffusion to pathogenic phenotype in humans. More importantly, I identified a novel role of SOX2 in stabilising PAX6- chromatin complexes in live cells, providing further insight into the complex and dynamic relation of PAX6 and SOX2 in ocular tissue specification, maintenance and development.