Meccanismi di riconoscimento del nucleosoma da parte di demetilasi istoniche

This thesis will present my research on LSD1 and LSD2, the only histone demethylases of the flavin class. They share the same substrate and reaction mechanism. Yet, they are recruited by distinct transcriptional macromolecular complexes with opposite effects on the chromatin state. In Chapter I, I will introduce the reader to the world of epigenetics, with a particular emphasis on nucleosome recognition by nuclear players. In the last six years, nucleosome-chromatin remodeler complex structures, obtained by X-ray crystallography and single particle cryo-EM provided deep insights into the structural basis of nucleosome recognition. However, structural characterization of interactions between flexible or dynamic regions has remained challenging. Combination of multiple and hybrid structural and biochemical approaches allowed characterization of the mechanism employed by LSD1 for nucleosome recognition, which necessitates DNA binding by the protein partner CoREST1. In Chapter II, I will present my research on the characterization of the relevant biomedical aspects of LSD1/CoREST. In particular, I analyzed the biochemical and structural effects of three pathological LSD1 single-residue mutations. It turned out that enzyme catalysis is only partially affected, and the structure not at tall, whereas diminished binding of the H3 tail and of other protein factors through the substrate-binding pocket might be the cause of such severe neurological and physiological symptoms. Here I attached the two published papers to which I contributed dissecting the effect of the pathological mutations on the enzyme folding and on histone H3 binding and catalysis, as well as on the recruitment of non-substrate regulatory proteins and transcription factors such as SNAIL1 and p53. The technical know-how I acquired during the characterization of LSD1/CoREST, greatly helped the structural and mechanistic investigation of its homolog LSD2 with its partner NPAC, illustrated in Chapter III. Yet, the LSD2/NPAC project turned out to be a more challenging and exciting team effort than ever expected. Particular technical and methodological choices greatly helped us to complete a solid description of LSD2/nucleosome complex formation mechanism. On this basis, I decided to describe in detail the rationale behind particular methodological choices, as well as the implementation and optimization of the protocols. Given that most of the data have been published (Marabelli et al., 2019), I will extensively present only those experiments I personally participated to. I am also including few unpublished results giving further insights on the biological function of the LSD2/NPAC system, whose chapter have been marked by an asterisk (*). The mechanism for nucleosome recognition by LSD2 is different than that of the previously characterized LSD1. LSD2/NPAC does not contact the core of the nucleosome, which is recruited by a tail-only mechanism. The short NPAC-linker module is extremely efficient in regulating the substrate histone tail processing, whereas other domains of NPAC also affect the avidity and processivity of the demethylase system. In accordance with literature, this machinery seems to be perfectly tailored to support the work of RNA-Polymerase II. The effort on studying the dehydrogenase domain of NPAC, led us to a new line of research, regarding the changing role of this enzyme during evolution: it appeared indeed that a single-point mutation might have shifted NPAC from being a cytosolic enzyme to a nuclear auxiliary subunit. This study has now been published and it is also reported in Chapter III, given its deep connection with the LSD2/NPAC story. The key role of the demethylase catalytic domain and the associated flavin cofactor, an aspect highly conserved in both enzymes, gives us a stimulating example of how a cell can employ the same tool for different purposes.