Research

Toolbox for Studying Protein Post-translational modifications: Lysine and Arginine Methylation

From a chemist’s point of view, it is remarkable that the placement of one methyl group in a particular position of an enormous protein-DNA complex is enough to control gene expression. But this is exactly the role of methylation of lysine residues in histone proteins that make up nucleosomes. It turns out that such a small change creates a new binding site for other proteins through that small perturbation of its properties, including its hydrogen bonding ability, desolvation cost, and van der Waals interactions. This phenomenon captured our imaginations and led us to try to understand it, develop sensors for it, and determine how to inhibit it, since dysregulation of lysine methylation is associated with many disease states, including a wide range of cancers.

Studying the mechanism of Kme3 binding

Counterintuitively, trimethyllysine (Kme3), which is cationic, binds in an aromatic pocket in Kme3 “reader proteins”, which would generally be described as hydrophobic. We have performed some of the seminal work determining the driving force for the recognition of trimethyllysine in an aromatic pocket, demonstrating that the interaction is driven by cation-pi interactions, in which the cationic trimethylammonium group interacts favorably with the electron-rich face of the aromatic rings. We continue those studies using more advanced strategies such as unnatural amino acid mutagenesis.

Mimicking proteins with small molecules

We have developed methods to mimic reader proteins using a biomimetic selection method called dynamic combinatorial chemistry (DCC) to identify synthetic receptors for methylated lysine and arginine. DCC utilizes La Chatelier’s principle to amplify the best binders from an equilibrating mixture of potential synthetic receptors. Thus, instead of designing synthetic receptors, we design building blocks containing functional groups that can form reversible covalent bonds. The receptors then self-assemble around the guest of interest, and are thus amplified. In effect, we let the guest pick out its own best host. When a new host is discovered, we determine how it works and then improve upon it by designing new building blocks with improved binding features. This has led to receptors that bind Kme3 with tighter affinity than the proteins we aim to mimic.