Natural products have long been appreciated for their potent biological activities, and their unique scaffolds and pharmacophores have served as inspiration for many pharmaceuticals. As the structural complexity of these compounds can make their production and diversification challenging for chemical synthesis, the study and engineering of natural product biosynthetic pathways can facilitate more efficient and sustainable production of natural products and their analogues. Here, we present our findings from the biosynthetic studies of several medicinally active natural products produced by Streptomyces, which reveal some unusual pathways and enzymes. In doing this work, we also demonstrate how heterologous expression can be used to expediently determine the necessity and function of individual biosynthetic enzymes, gain insight into biosynthetic mechanisms, and produce new natural product analogues.

The compounds of focus here include the cholinesterase inhibitor physostigmine, the peptidyl epoxyketone proteasome inhibitor eponemycin, and the family of antimycin-type depsipeptides. The in vivo and in vitro characterization of physostigmine biosynthesis revealed an unexpected pathway involving an acetylation-deacetylation-dependent reaction cascade and a unique indole methyltransferase. Heterologous expression-based investigations on the biosynthesis of eponemycin then revealed an unprecedented flavin-dependent enzyme to be necessary and sufficient for the formation of the terminal epoxyketone pharmacophore in addition to indicating the involvement of a decarboxylation step. Further efforts to study the biosynthesis of antimycins likewise resulted in the in vivo reconstitution of the 3-formamidosalicylate pharmacophore, and studies on the related neoantimycins led to advances in the biosynthetic understanding and heterologous expression of this ring-expanded class of antimycin-type depsipeptides. Collectively, these biosynthetic insights not only enrich our enzymatic knowledge but can also expand our ability to perform combinatorial biosynthesis.

The alkyne functionality has attracted much interest due to its diverse chemical and biological applications. However, the underdeveloped alkyne biosynthetic toolbox severely limits the applications of this functionality in drug discovery and chemical biology. De novo biosynthesis is a recently developing and one of the most promising strategies for installing the terminal alkyne functionality into various molecular scaffolds, including diverse bioactive natural products, through the in situ generation and incorporation of the terminal alkyne functionality. We recently elucidated an acyl carrier protein (ACP)-dependent alkyne biosynthetic pathway comprised of JamABC from Moorea producens JHB that produces a terminal alkyne-tagged polyketide using hexanoic acid as a starting substrate for terminal alkyne generation. However, little is known about ACP interactions with the other alkyne biosynthetic enzymes in the JamABC pathway, including an acyl-ACP ligase (JamA) and a membrane-bound bi-functional desaturase/acetylenase (JamB). In this study, we demonstrated that JamB has a more stringent interaction with ACPs than JamA and subsequently pursued two different strategies to pinpoint key protein-protein interactions between JamB and several non-cognate ACPs. We first employed rational engineering of a non-cognate ACP (TtuC from Teredinibacter turnerae T7901) through site-directed mutagenesis, which significantly improved its compatibility with JamB and suggested a possible electrostatic interaction at the ACP-JamB interface. We also utilized an error-prone PCR and screening protocol towards a second non-cognate ACP (PeACP from Pseudomonas entomophila L48) to identify two hot spots on that ACP that are important for interacting with JamB and yielded mutants which were better recognized by JamB than the wild-type ACP. This study thus not only provides insights into the ACP interactions in alkyne biosynthesis, but it also potentially aids in future combinatorial biosynthesis of alkyne-tagged metabolites for chemical and biological applications.

The production of ABE (acetone, butanol, ethanol) by members of the anaerobic bacterial genus Clostridium is a bioprocess capable of converting various renewable substrates into high-value solvents and fuels. While ABE fermentation by clostridia has been performed on an industrial scale since the early 20th Century, several significant technical challenges have prevented this process from being economically competitive with the conventional production of solvents and fuels from petroleum. Here, we discuss the role which signaling small molecules (SSMs) play in controlling industrial traits of solvent-producing species of Clostridium. This includes our discovery and characterization of a family of polyketides from the model ABE-producer Clostridium acetobutylicum ATCC 824, which we demonstrate are important in influencing solvent production and controlling morphological differentiation in this organism. Additionally, we describe the development of genetic tools and techniques for genome editing of another industrial solvent-producing species, Clostridium saccharoperbutylacetonicum strain N1-4 (C. s. N1-4). Applying these methods, we present preliminary results which suggest that SSMs may play critical roles in controlling ABE production in C. s. N1-4. Finally, we present our discovery of the first known non-ribosomal peptide isolated from an anaerobic organism. Through this work, we hope to highlight the significant role of secondary metabolites in clostridial fermentations (as well as anaerobes in general), and provide a relatively unexplored strategy for improving this promising renewable bioprocess.

The terminal alkyne functionality is widely used in organic synthesis, pharmaceutical science, and materials. It is also a useful moiety in the azide-alkyne [3+2] cycloaddition reaction (often known as “click chemistry”), which has recently emerged to be one of the most powerful tools in drug discovery and chemical biology. In particular, by tagging medicinally active natural products with alkynes and coupling them with subsequent click chemistry, researchers have enriched, visualized, and studied the mode of action of natural products. Despite the importance of the alkyne functionality, its applications in natural product research are limited by the underexplored alkyne biosynthetic tools and tagging strategies.

In this study, we first elucidated an unprecedented carrier protein–dependent terminal alkyne biosynthetic pathway in microbes that featured an acyl-acyl carrier protein (ACP) synthetase (JamA), a membrane-bound bifunctional desaturase/acetylenase (JamB), and an ACP (JamC). We further demonstrated that this enzymatic machinery can be exploited for the in situ generation and incorporation of terminal alkynes into two natural product scaffolds (a polyketide pyrone, and antimycin, respectively) in Escherichia coli. However, JamB showed stringent substrate specificities and weak catalytic activities, which restricted its broader applications in synthetic and chemical biology. We addressed these limitations in two ways. In silico analysis suggested that this carrier protein–dependent terminal alkyne biosynthetic mechanism could be widespread in bacteria. We screened additional gene cassettes that are homologous to jamABC and discovered a new terminal alkyne biosynthetic pathway comprised of TtuABC from Teredinibacter turnerae T7901. Different from JamABC, TtuABC displayed altered substrate specificities and improved catalytic activities. Using TtuABC, we further in situ generated and incorporated the alkyne functionality into another medicinally important natural product, a peptidyl epoxyketone. In a parallel effort, directed evolution was adopted to create a JamB variant with improved catalytic efficiency in E. coli. To quickly assess the protein engineering outcomes, we developed a new platform for quantifying extracellular alkyne-tagged metabolites through a fluorogenic click reaction. Overall, this study not only provides us with alkyne biosynthetic tools that are useful in synthetic and chemical biology, but it can also find applications in which enzymatic generation of terminal alkynes and in situ click chemistry is required or preferred.

Humans have exploited small molecule natural products have exploited throughout history. Recently, advances in chemistry and biology have led to an explosion in drug development, revolutionizing the treatment of infectious diseases, cancer, and other ailments. However, current techniques have already been exploited nearly to their limits. The work described in this dissertation aims to expand the pharmacological potential of small molecule natural products by developing a range of techniques that exploit functional chemical tags that are chemically bonded to the small molecules. These tags enable natural products to be studied in new ways using techniques such as Raman microscopy, computational analysis of LCMS data, and optical lock-in detection of FRET. By fundamentally altering small molecule natural products, we have raised the possibility to discover and study drug-like molecules that would be inaccessible using traditional techniques.