Biochemical characterization of a diiron halogenase

Abstract

The installation of a C–X bond (where X = F, Cl, Br, or I) modifies the physiochemical properties of molecules, often granting them improved bioactivity. Thus, halogenation reactions are highly valued in synthetic chemistry, particularly in the agricultural and pharmaceutical industries. However, synthetic halogenation often encounters significant challenges in achieving chemo and regioselectivity. Halogenation is also a prominent transformation in biology, present in all domains of life; to date, there have been over 5000 halogenated metabolites identified in the scientific literature. Nature has evolved halogenases, enzymes that perform C–X bond formation with remarkable efficiency and selectivity. While multiple classes of halogenases have been characterized in the past few decades, there are still halogenated metabolites biosynthesized via unknown C–X bond formation strategies, leading to gaps in our understanding of this important class of enzymatic transformations. In this thesis, we describe the characterization of diiron halogenases, an emerging subfamily of halogenase enzymes capable of radical-based halogenation. We use sequence-based information to determine and identify conserved features of these halogenases. Using information we gathered from our bioinformatic studies, we then biochemically characterize a new member of this halogenase subfamily, providing experimental evidence for the presence of a diiron cofactor. Together, these approaches expand our knowledge of enzymatic C–X bond formation and the reactivities of metal cofactors in biology. In Chapter 1, I provide a review of the current state of knowledge regarding enzymatic C–X bond formation. This chapter explores the three halogenase families: electrophilic, nucleophilic, and radical halogenases. This section provides an overview of the C–X bond formation strategies utilized by each family along with their substrate scope, accepted mechanisms, and structural bases for halogenation. This section ends with a discussion of the discovery of the first putative dimetal halogenase, CylC, and previous efforts to further our understanding of this new halogenase subfamily. Chapter 2 describes bioinformatic efforts to gain an understanding of key features of the protein sequences of the dimetal halogenase subfamily. By analyzing multiple sequence alignments, we identify distinct sequence motifs in dimetal halogenases that differentiate them from diiron N-oxygenases, their closest structurally characterized homologs. Prediction of the structures of dimetal halogenases suggests a basis for metallocofactor and halide binding, something previously obscured in older homology models. These findings provide an improved understanding of this enzyme subfamily and hypotheses to guide in vitro biochemistry. Chapter 3 describes the in vitro characterization of a CylC homolog and dimetal halogenase, NocO from nocuolactylate biosynthesis. We describe the purification of NocO with its cognate acyl carrier protein (ACP), NocM, and describe our initial efforts to characterize this protein alongside CylC. This section also discusses the difficulties associated with working with this halogenase subfamily, especially on preparative scales. We then detail the development of a coupled enzyme assay to generate ACP-tethered intermediates for the reconstitution of NocO’s chlorination activity and an exploration into its ability to use other anions for C–H functionalization. This chapter establishes important methods and a platform to guide work for metallocofactor characterization. In Chapter 4, we leverage the information we obtained from bioinformatics and the in vitro reconstitution of NocO’s chlorination activity to further characterize the nature of NocO’s metallocofactor. We discuss our efforts to optimize a platform for the expression of several unstable active site variants, as well as new insights gained into the residues responsible for metal binding. Finally, we present the first direct spectroscopic evidence of a diiron cofactor in this halogenase subfamily, thus confirming NocO’s identity as a non-heme diiron halogenase and the first diiron enzyme known to perform halogenation. Collectively, this thesis expands our knowledge of enzymatic C–X bond formation through bioinformatic analyses of the diiron halogenase family and the characterization of the first confirmed diiron halogenase. These findings could be used as a basis to further mechanistic studies of these enzymes and to ultimately expand enzymatic toolboxes for C–H functionalization.