The goal of the research in my laboratory is to understand the biosynthetic pathways plants, microbes, and other organisms use to produce a vast array of chemicals that allow them to survive and prosper in a multitude of challenging ecosystems (Figure 1). Our aims are to understand the chemical, structural, and evolutionary principles that underlie this extraordinary form of biological diversity, as well as to harness and alter these pathways to produce chemical “scaffolds” that can provide the starting points for the development of chemical tools for modulating proteins, cells, and organisms.
Why are these metabolic pathways useful and important for understanding biodiversity and evolution? Diverse molecular changes associated with specialized metabolism are often preserved genetically, functionally, and structurally as a result of the increased fitness of host organisms in diverse and challenging ecosystems. These specialized metabolic pathways and their “chemical output” present us with a rich evolutionary record of where biosynthetic pathways, natural products, and biosynthetic enzymes have been (vestigial biochemical traits), what adaptive significance these complex enzymatic systems hold in the present, and ultimately where these pathways may be heading in the future.
Type III Polyketide Synthases
All polyketide synthases (PKSs), like their fatty acid synthase (FAS) ancestors, possess a β-ketosynthase (KS) activity that catalyzes the sequential head-to-tail incorporation of two-carbon acetate units into a growing polyketide chain. Unlike iterative and modular type I and iterative type II PKSs, type III or chalcone synthase (CHS)-like PKSs, which are the focus of our research, maintain a simple three-dimensional structure, a multitude of diverse specificities, and a facile expression profile in heterologous hosts, making them amenable to in vitro examination and manipulation as well as to high-resolution structural analyses (Figure 2). (This work was supported by grants from the National Institutes of Health and the National Science Foundation.)
New Organizational Paradigms in Type III PKS Structure and Catalysis
Our examination of available DNA-sequencing contigs for Dictyostelium discoideum recently revealed two unprecedented type III PKS-containing genes in this slime mold. These two genes encompass not isolated type III PKS genes but CHS-like type III PKSs clearly linked to large type I FAS/PKS multidomain enzymes (Figure 3).
Our preliminary in vitro analyses using structural and biochemical approaches of the type III PKS domains of these fungal type I/III hybrid PKSs confirms our speculation, based on gene organization, that these domains iteratively elongate and cyclize a non-standard-length acyl starter produced by the attached type I PKS. The end product of one type III PKS domain produces a critical intermediate, 1-[2,4,6-trihydroxyphenyl]-hexan-1-one, used by the biosynthetic pathway for the small-molecule modulator of cellular differentiation in D. discoideum known as differentiation initiation factor 1 (DIF-1). The second type III PKS domain produces acyl pyrones when initiated with a variety of short-chain acyl-CoAs.
Sesquiterpene Synthases (Cyclases)
Terpenes comprise the most diverse family of natural products known. Terpene synthases (cyclases) catalyze electrophilic transformations of acyclic, achiral, isoprenoid substrates into multicyclic, multichiral products. Remarkably, cyclization of acyclic terpenoid substrates in a terpene cyclase-active site proceeds through several carbocationic intermediates, which readily undergo dramatic and stereochemically specific structural rearrangements.
We are using Nictotiana tabacum 5-epi-aristolochene synthase (TEAS) to investigate the structural basis for product specificity of sesquiterpene (15-carbon) cyclases (Figures 4 and 5). This was the first terpene cyclase structure elucidated by our laboratory, and the initial structures have led to the development of hypotheses concerning the mechanisms of carbocation-mediated catalysis.
Comparative analyses of TEAS and the evolutionarily related Hyoscyamus muticus premnaspirodiene synthase (HPS) yielded deeper insights into the mechanisms of sesquiterpene cyclases. TEAS and HPS are closely related cyclases (72 percent identical at the amino acid level) that share a common reaction mechanism but diverge at the penultimate step of catalysis where either a methyl migration or an alkyl shift occurs to produce 5-epi-aristolochene or premnaspirodiene, respectively (Figure 5). Using the TEAS structure as a guide, we successfully converted TEAS into a premnaspirodiene synthase (PS) through the sequential introduction of nine amino acid changes that line the periphery of the active site (TEAS M9). All of the changes form a group of second-tier residues that likely support the structure and dynamics of the active-site cavity but nonetheless play essential roles in directing product specificity.
We also developed structure-based combinatorial protein engineering (SCOPE) of sesquiterpene cyclase genes for the rapid, inexpensive, and complete synthesis of gene libraries to lay the genetic foundation for the exploration of the relationship between structure and function in enzymes of specialized metabolism. Separation of gene synthesis into discrete steps is a critical property of SCOPE that enables one to control recombination through pairing gene fragments and genes that give rise to designed and anticipated combinations of crossovers. (This work was supported by a grant from the National Institutes of Health.)
SCOPE-ing Out Evolutionary Landscapes
Our goal is to understand how the three-dimensional folds of sesquiterpene cyclases link a single FPP starting material to the diversity of products known or yet to be discovered in nature (Figure 6). Although the successful conversion of TEAS into an active PS through the introduction of nine sequential changes is an intellectually satisfying application of heuristic approaches toward enzyme engineering, the question remains—what is the catalytic potential of all possible permutations (29 = 512) at these nine positions? In other words, can we, to a first approximation, completely sample the chemical diversity inextricably linked to the evolutionary landscape separating TEAS (tobacco) from its phylogenetic relative HPS (henbane)? What combinations, if any, of these nine mutations is sufficient for this property change? Is the change gradual or does it occur only as the end point is reached? Apart from addressing fundamental evolutionary questions, this large collection of terpene cyclases will be an important source of diverse chiral cyclic terpene scaffolds. (This work was supported by a grant from the National Institutes of Health.)
Diversity-Oriented Biosynthesis/Chemical Synthesis-Chemoenzymatic Approaches
Since all biological targets of small molecules are chiral, a premium is placed on the use of chiral chemicals in the pharmaceutical industry. The sesquiterpene hydrocarbons produced either in vitro or in vivo using sesquiterpene cyclases possess clear advantages as starting points for the synthesis of new classes of chiral molecules due to their small size (204 daltons), multiple chiral centers (two or more), and now facile means of production on the multigram scale in a heterologous fermention system.
To exploit this class of chiral chemical scaffolds will require a strategy that combines the power of enzymes for “chiral resolution” with the power of chemistry for diversity-oriented synthesis. Since we now have gram quantities of these rare chiral molecules, we are exploring synthetic methods that can readily “functionalize” these conformationally restricted and chiral starting materials.
To demonstrate the utility of this combined use of biosynthetically derived natural products and chemical derivatization, we are exploring agonism and antagonism in the farnesoid-X receptor (FXR). We have carried out in silico docking studies of FXR with several sesquiterpene products successfully produced by fermentation at the gram level. These computational studies have identified premnaspirodiene as an initial target for further chemical elaboration. A single binding mode of premnaspirodiene to the FXR ligand-binding pocket orients specific carbons on the hydrocarbon scaffold in spatial locations that map to regions we have previously shown to regulate agonism and antagonism of FXR ligands (Figure 7).
Engineered Biosynthesis in Heterologous Hosts
Over the long term, we aim to combine the type of directed engineering studies described above for type III PKS and sesquiterpene cyclases with structure-based engineering of larger pathways for specialized metabolites. Two of our goals are (1) to be able to modify pathways such that natural products and the protein complexes that make them can be produced endogenously in amounts that are bioactive within the cell producing them and (2) to be able to regulate the genetic and molecular pathways that make these molecules with such precision that they can be moved with ease between animal, plant, and microbial cells. The ultimate goal is to induce in any cell type the ability to produce bioactive chemicals with temporal and spatial accuracy.
