UC Berkeley

Nature has evolved myriad enzymes that perform selective C-H activation, a powerful method of building molecular complexity. These include radical Fe(II)/α-ketoglutarate (Fe(II)/αKG)-dependent halogenases which install the synthetically useful halogen functional group into a variety of substrate classes. Here, we focus on building an understanding of the mechanistic basis of selectivity in a family of amino acid-modifying halogenases and using that understanding to engineer non-native activity. First, we solved a set of crystal structures to explore the role of a C-terminal lid domain in substrate binding and specificity. We used our structural insights to successfully engineer the site- and substrate-selectivity of these enzymes. Next, we used X-ray crystallography, spectroscopy, and DFT calculations to build a revised mechanistic model of halogenase chemoselectivity. We identified the formation of a tyrosyl radical in a closely related radical hydroxylase and studied the competing pathways of substrate and protein oxidation. Lastly, we developed a Fe(II)/αKG-dependent peptide hydroxylase into an azidase and built a workflow to generate a protein halogenase. These insights represent a significant contribution to our understanding of enzymatic C-H activation and halogenation.