Despite years of research, numerous aspects of gene regulatory mechanisms and their contributions to shaping phenotypic characteristics remain elusive. In this thesis, I gain insights into mechanisms of transcriptional regulation, focusing particularly on the binding activity of transcription factors (TFs). In the comprised projects and analyses, I use models of closely related mouse species, as in vivo systems that leverage a pool of molecular variation, in order to study the activity and roles of TF binding within wider functional contexts. The molecular variation is manifested as variation in the sequence context and/or in the binding activity of the TFs, and it may be naturally fixed by evolution among the different species, and/or induced variation among different conditions, for example as a result to carcinogen exposures. The reflection of the underlying molecular variation in the functional roles of TF binding is also tied to some extent to functional outputs and is examined as part of general functional contexts, such as higher-order genome organisation or the neoplastic transformation of a tumour cell in cancer development.

In the first part of the thesis, I explore the evolutionary dynamics of the CTCF (CCCTC-binding factor) binding among mouse species and its interplay with the establishment and maintenance of topologically associating domains (TADs). TADs make up a level of the 3D genome organisation that dictates formation of regulatory landscapes. Although it is known that CTCF plays an important role in TAD formation, evidence of TAD boundaries being robust to CTCF depletion interferences has confounded the understanding of its exact role. My study reveals the occurrence of dynamically evolving clusters of neighbouring CTCF sites at the boundaries of TADs, which contribute to the resilience and flexibility of the TAD structures. My findings also highlight the necessity of examining the binding of CTCF not only as individual binding sites, but also as an ensemble of potentially functional sites that can exert their actions synergistically, additively or interchangeably, and contribute to fostering phenotypic features.

In the second part of the thesis, I present a project I worked on as a member of a collaborative effort of the Liver Cancer Evolution (LCE) consortium. The project makes use of a model of chemically induced liver carcinogenesis in different mouse species, which offer a controlled biological system with natural genomic and epigenomic variation. Here, I characterise the expression profile of the cell-of-origin, the hepatocyte, of chemically induced liver tumours and the shift of its cell-type specific phenotype along neoplastic transformation, which is underlined by gene dysregulation. In addition, I further explore associations of gene dysregulation in the cell-of-origin with mutagenesis (induced genetic variation as a result of carcinogen exposure) in liver TF binding sites that potentially underlie the onset of hepatocyte dedifferentiation in tumour development. My results disentangle patterns of mutation accumulation among distinct categories of TF binding sites based on their functional characteristics, such as combinatorial binding or association with with cis-regulatory elements. Also, they reveal sets of hyper-mutated TF binding sites in liver tumours, which are functionally enriched for liver-specific biological processes, cellular stress response and abnormal liver phenotype, and they are associated with dysregulated genes with hepatocyte-specific functions. Finally, these findings also highlight the necessity of studying the activity of TFs bound both in a singleton manner and in concert and within their broader functional context.