ldentification of an allosteric regulatory mechanism in the human histone demethylase KDM5A
- Author(s): Ortiz-Torres, Idelisse
- Advisor(s): Galonic Fujimori, Danica
- et al.
Histone lysine methylation plays an essential role in a variety of cellular processes including transcriptional regulation and cell differentiation. The retinoblastoma binding protein KDM5A is a histone H3 lysine 4 (H3K4) demethylase, capable of removing the trimethyl mark on K4 all the way to zero methylation. As trimethylation of H3K4 (H3K4me3) is often associated with promoters of actively transcribed genes, removal of H3K4me3 by KDM5A is believed to promote transcriptional repression of its target genes. Mis-regulation of KDM5A has been shown to contribute to pathogenesis of lung and gastric cancer, as well as to the resistance to receptor tyrosine kinase inhibitors in non-small cell lung cancer. Yet, the mechanisms by which KDM5A is recruited to its target loci and how it functions on chromatin are largely unknown.
In addition to its catalytic jumonji C (JmjC) eraser domain, KDM5A contains three PHD reader domains that are hypothesized to regulate the biological function of KDM5A. It is unknown, however, if and how these reader domains play a functional role in the demethylation reaction of KDM5A. In this thesis, using an array of biochemical and biophysical techniques, we show that the PHD1 domain specifically recognizes an unmethylated histone H3 tail, which interestingly is the final product of KDM5A. We also show that occupancy of PHD1 by its ligand allosterically stimulates the demethylase activity of the demethylase on histone tail peptides and on in vitro reconstituted nucleosome substrates.
Together, these observations reveal an unprecedented allosteric regulation of a histone demethylase by one of its auxiliary non-catalytic domain and, challenges the prevailing model that PHD domains solely act to recruit chromatin-modifying enzymes to their target sites. Moreover, our findings suggest a model by which demethylation could spread on chromatin through a positive feedback-based mechanism. Furthermore, this mode of regulation provides new alternative mechanisms for the development of cancer therapies to specifically target RBP2 activity when mis-regulated in lung and gastric cancer. Here, we will also present data toward our initial efforts for the development of a small molecule allosteric modulator of KDM5A as well as our initial attempts to determine structural features of the enzyme by electron microscopy EM.