Lysine demethylases are a newly discovered group of enzymes that have rapidly expanded over evolutionary time by the acquisition of multiple functional domains, in addition to the unifying catalytic JmjC domain. There are thirty members of the JmjC-domain family in humans. A proportion of lysine demethylases catalyse the removal of methyl modifications from lysine residues of histones and non-histone proteins. The discovery of mutations in histone demethylase genes, in a number of human syndromes, stresses the functional importance of these enzymes in development and disease. Therefore, the phenotypic dissection of animal models of histone lysine demethylases will provide invaluable insights into the molecular mechanisms that underlie human disease. In mammals, the Kdm3 family of histone demethylases includes Kdm3a, Kdm3b, Jmjd1c and Hairless. However, in zebrafish, there are two kdm3 genes, one of which encodes a protein similar to both the mammalian Kdm3a and Kdm3b. Morpholino knock-down of the kdm3 gene in zebrafish faithfully recapitulates classical ciliary phenotypes, although the underlying causalities are still unclear. In recent years, Kdm3a function has been extensively dissected through the use of mouse models and cell culture studies, focusing on the nuclear histone demethylation function. Kdm3a gene-trap and knock-out mouse models present with obesity, infertility, sex reversal and predisposal to diabetes, reminiscent of a human ciliopathy syndrome. No mouse models for Kdm3b have been characterised yet. In this study, I hypothesized that the murine Kdm3a and Kdm3b histone demethylases have diverged biological roles and that the zebrafish kdm3 fulfils the functions of both. The aims of my thesis were: 1) to compare the evolutionary conservation of the zebrafish kdm3 and murine Kdm3b in function and check their spatial expression, 2) to dissect the phenotype of Kdm3b gene-trapped mice and 3) to characterise an alternative murine Kdm3a isoform. Protein sequence comparison studies show that the zebrafish kdm3 protein is closer in sequence to the mammalian Kdm3b. Both the zebrafish kdm3 and murine Kdm3b are di-methyl lysine 9 (H3K9me2) demethylases, however, they have diverged spatial expression during embryogenesis. In agreement with the phenotype of kdm3 morphants, over-expression of the zebrafish kdm3 reduces ciliation efficiency when transfected into animal cells. Notably, the phenotype analysis of Kdm3b gene-trapped mice does not resemble classical ciliary phenotypes, as one would expect from the zebrafish data. Homozygous Kdm3b gene-trapped mice are postnatally growth retarded, with plausible defects in thymus organ development. Interestingly, an alternative murine Kdm3a isoform (Kdm3a-i2) shows both nuclear and cytoplasmic localisation. Over-expression studies revealed that Kdm3ai2 retains its histone demethylation function, and a proportion of the over-expressed construct localises to the centrosome. In addition, over-expression of Kdm3a-i2 reduces ciliation efficiency. Overall, the data from my studies suggests that: 1) the zebrafish kdm3 is more similar in sequence to the murine Kdm3b than Kdm3a, is a histone demethylase and has a distinct spatial expression during embryogenesis. However, the phenotype of kdm3 zebrafish morphants is more closely related to the Kdm3a-than Kdm3bdeficient mice, 2) the murine Kdm3a and Kdm3b have distinct biological roles, as evidenced by the mouse models, 3) the Kdm3a-i2 isoform shares the same nuclear demethylation function as the full length Kdm3a and has a plausible centrosomal function.

MLL/KMT2A amplifications and translocations are prevalent in infant, adult and therapy-induced leukemia. However, the molecular contributor(s) to these alterations are unclear. Here we demonstrate that histone H3 lysine 9 mono- and di-methylation (H3K9me1/2) balance at the MLL/KMT2A locus regulates these amplifications and rearrangements. This balance is controlled by the cross-talk between lysine demethylase KDM3B and methyltransferase G9a/EHMT2. KDM3B depletion increases H3K9me1/2 levels and reduces CTCF occupancy at the MLL/KMT2A locus, and in turn, promotes amplification and rearrangements. Depleting CTCF is also sufficient to generate these focal alterations. Furthermore, the chemotherapy Doxorubicin (Dox), which associates with therapy-induced leukemia and promotes MLL/KMT2A amplifications and rearrangements, suppresses KDM3B and CTCF protein levels. KDM3B and CTCF overexpression rescues Dox-induced MLL/KMT2A alterations. G9a inhibition in human cells or mice also suppresses MLL/KMT2A events accompanying Dox treatment. Therefore, MLL/KMT2A amplifications and rearrangements are controlled by epigenetic regulators that are tractable drug targets, which has clinical implications. The data presented in this thesis were published in Cell in 2023. / Biomedical Sciences