Histone lysine methyltransferases (HKMTs) are a family of epigenetic enzymes responsible for catalyzing the methylation of lysines in histones and other proteins. During the past decade, the understanding of HKMTs has significantly advanced from structural and biochemical perspectives, enabling development of selective inhibitors. However, while structural evidence shows that HKMTs are conformationally dynamic in the context of protein methylation and small molecule binding, only limited knowledge is available about their full conformational ensembles. We combined experiment and computation to explore the conformational dynamics of SETD8, which is responsible for the methylation of H4K20 and p53, and implicated in a number of cancers, such as neuroblastoma. A series of X-ray structures of SETD8 in previously unseen conformations were solved with various ligands. Molecular dynamics simulations were conducted to collect ~10 ms of data in millions of conformations and analyzed with a Markov state model to reveal the free energy landscapes, hidden states, and pathways between different conformations. I will present a structural analysis of the SETD8 conformational ensemble, focusing on functionally important microsecond-timescale rearrangements of its sub-domains. I will highlight the usefulness of hierarchical kinetic state models for the description of molecular dynamics and their experimental validation. We also studied how cancer-related and designed mutations perturb the conformational ensemble, in terms of reweighting wild type conformations and causing structural rearrangements. The plasticity of known binding pockets and the openings of putative allosteric pockets were also analyzed, with the goal to design HKMT chemical probes to capture unique conformations of individual enzymes. This approach was extended to groups of HKMTs (such as NSD1/2/3, EZH2) at different evolutionary distances to comparatively model the dynamics of the protein family and their functional impact. The models will be analyzed to understand the free energy profiles along transitions between active and inactive conformations, the allosteric regulation of those activation profiles by protein-protein binding partners (e.g. of EZH2 in the PRC2 complex), and the role of flexible activation loops in the process.

Markov models of functional dynamics of histone methyltransferases by millisecond-timescale molecular simulation and chemical probing

2019 Fall National Meeting

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.