Chemoenzymatic Synthesis of Azaphilone Natural Products

Abstract

Over the course of evolution, Nature has crafted innumerous enzymatic machines, capable of great feats of chemistry, across nearly all walks of life. These biocatalysts can construct complex molecular scaffolds, demonstrated by the abundance of intricate natural products isolated from producing organisms. These compounds have inspired synthetic chemists for decades, but it is only recently that technology has advanced to the point where the use of enzymes for organic synthesis has become feasible. Not only do biocatalysts offer advantages over their small molecule counterparts in terms of ease of use, mild conditions, greater selectivity, and faster reaction times, but also in the sense that many have evolved to operate synergistically with one another. This inherent coaction opens the possibility of performing enzymatic and/or chemoenzymatic cascades to construct complicated molecules in a one-pot fashion, chaining together multiple transformations and streamlining chemical syntheses. Chapter 1 summarizes the history of biocatalysis in the context of organic synthesis, the current state of this field, and future prospects for this methodology in the synthesis of complex molecules. The remainder of this thesis describes the use of enzymes from biosynthetic pathways to construct members of a class of fungal natural products called azaphilones. This family of compounds contains hundreds of scaffolds with unique structural features that impart a wide array of biological properties. Though azaphilones represent a potentially untapped pharmacophore, their construction has proven challenging through conventional chemical syntheses. In Chapter 2, I describe our rationale behind the motivation for using biocatalysis to construct azaphilones, the initial challenges associated with this goal, and our search for enzymatic catalysts to mediate the necessary dearomatization reaction towards constructing the azaphilone core. Using a sequence similarity network (SSN), we identified several previously uncharacterized flavin-dependent monooxygenase (FDMO) enzymes capable of mediating the desired reaction in a model system. Further analysis of these FDMOs also provided evidence for structural features of these enzymes that control the facial selectivity of the dearomatization reaction. Having identified and characterized useful FDMO enzymes, we then constructed several azaphilone natural products through chemoenzymatic syntheses in Chapter 3. Two FDMO homologs, AzaH and AfoD, provided access to both enantiomers of an azaphilone bicycle, upon which the natural products trichoflectin and lunatoic acid A were built. Obtaining both enantiomers of trichoflectin through this methodology provided evidence for the structural revision of this compound, uncovering an error in the azaphilone literature on the absolute configuration assignment of these compounds. AzaH also provided access to another natural product, deflectin-1a, allowing for its structural revision, as well. Finally, in Chapter 4, I describe the development of a dual-enzyme platform to construct azaphilone tricycle analogs. Several acyltransferases were identified using an SSN. Ultimately, the acyltransferase (AT) MrPigD (PigD) was investigated for its ability to acylate the C7 hydroxyl group of azaphilone bicycles. Our studies indicate that this enzyme operates synergistically with AzaH in a one-pot reaction, providing acylated azaphilones directly from five different orcinaldehyde precursors. This platform also provided access to the natural product, rubropunctatin, through cyclization of the acylated intermediate to afford a linear tricycle. Following analysis of the compound by UPLC, quantities of the natural product sufficient for NMR were obtained through a preparative scale biocatalytic reaction. This method was also applied to the synthesis of two other tricycle analogs, whose 1H NMR spectra suggest they are also linear tricycles.