Decarboxylations/carboxylations are chemically and biologically important reactions that involve the transfer of a carbon atom to an organic scaffold. In biochemistry, carboxylations are often coupled to other metabolic reactions as a way to drive them forward. Industrially, (de)carboxylation reactions are of considerable interest as they provide routes to many valuable compounds. Though thermodynamically favorable, decarboxylations are kinetically difficult because they involve the formation of a high-energy carbanion intermediate. Nature overcomes this barrier by using cofactors such as thiamine pyrophosphate (TPP), pyridoxal phosphate (PLP) and metal ions. In 2015, our lab co-discovered a new modified flavin mononucleotide (FMN) cofactor. Dubbed the prenylated flavin (prFMN), this cofactor is used by enzymes in the UbiD family of reversible decarboxylases.
Our lab is interested in studying the enzymes in this family and elucidating their mechanisms. Currently, we are studying two enzymes: ferulic acid decarboxylase (FDC) which converts phenylacrylic acid to styrene, and PhdA which decarboxylates phenazine-1-carboxylic acid (PCA). Both of these enzymes are involved in detoxification systems, but the prFMN cofactor is broadly utilized throughout the microbial world: for example, prFMN enzymes are involved in bacterial ubiquinone biosynthesis and in secondary metabolism.
Current Research Interests
Elucidating the mechanism and kinetics of ferulic acid decarboxylase
Investigating the reactivity of PhdA and probing its mechanism
Identifying other prFMN-dependent decarboxylases and studying how they utilize the prFMN cofactor
Phenazine-1-carboxylic acid decarboxylase (PhdA) is a member of the expanding class of prenylated-FMN-dependent (prFMN) decarboxylase enzymes. These enzymes have attracted interest for their ability to catalyze (de)carboxylation reactions on aromatic rings and conjugated double bonds. Here we describe a method to reconstitute PhdA with prFMN that produces an active and stable form of the holo-enzyme that does not require prereduction with dithionite for activity. We establish that oxidized phenazine-1-carboxylate (PCA) is the substrate for decarboxylation, with kcat = 2.6 s–1 and KM = 53 μM. PhdA also catalyzes the much slower exchange of solvent deuterium into the product, phenazine, with an apparent turnover number of 0.8 min–1. The enzyme was found to catalyze the decarboxylation of a broad range of polyaromatic carboxylic acids, including anthracene-1-carboxylic acid. Previously described prFMN-dependent aromatic (de)carboxylases have utilized electron-rich phenolic or heterocyclic molecules as substrates. PhdA extends the substrate range of prFMN-dependent (de)carboxylases to electron-poor and unfunctionalized aromatic systems, suggesting that it may prove a useful catalyst for the regioselective (de)carboxylation of otherwise unreactive aromatic molecules.
PAD1 produces prFMN from reduced FMN and the prenyl donor, dimethylallyl pyrophosphate
Recent Publication
P. M. Datar, E. N. G. Marsh (2021). "Decarboxylation of Aromatic Carboxylic Acids by the Prenylated-FMN-dependent Enzyme Phenazine-1-carboxylic Acid Decarboxylase" ACS Catal. 11(18): 11723-11732
Ferulic acid decarboxylase catalyzes the decarboxylation of various substituted phenylacrylic acids to their corresponding styrene derivatives and CO2 using the recently discovered cofactor prenylated FMN (prFMN). The mechanism involves an unusual 1,3-dipolar cycloaddition reaction between prFMN and the substrate to generate a cycloadduct capable of undergoing decarboxylation. Using native mass spectrometry, we show the enzyme forms a stable prFMN–styrene cycloadduct that accumulates on the enzyme during turnover. Pre-steady state kinetic analysis of the reaction using ultraviolet–visible stopped-flow spectroscopy reveals a complex pattern of kinetic behavior, best described by a half-of-sites model involving negative cooperativity between the two subunits of the dimeric enzyme. For the reactive site, the cycloadduct of prFMN with phenylacylic acid is formed with a kapp of 131 s–1. This intermediate converts to the prFMN–styrene cycloadduct with a kapp of 75 s–1. Cycloelimination of the prFMN–styrene cycloadduct to generate styrene and free enzyme appears to determine kcat for the overall reaction, which is 11.3 s–1.
PAD1 produces prFMN from reduced FMN and the prenyl donor, dimethylallyl pyrophosphate
Recent Publication
A.K. Kaneshiro, K.J. Koebke, C. Zhao, K.L. Ferguson, D.P. Ballou, B.A. Palfey, B.T. Ruotolo, E.N.G. Marsh (2021). "Kinetic Analysis of Transient Intermediates in the Mechanism of Prenyl-Flavin-Dependent Ferulic Acid Decarboxylase." Biochemistry 60(2): 125-134.
PAD1 produces prFMN from reduced FMN and the prenyl donor, dimethylallyl pyrophosphate
PAD1 is responsible for producing the prFMN cofactor in yeast. Its bacterial homologue, UbiX, had previously been found to produce prFMN from reduced FMN and the prenyl donor dimethylallyl phosphate (DMAP). To determine the kinetics of PAD1, our lab developed a coupled assay for PAD1 that utilized FDC to measure the production of prFMN. In contrast to the bacterial enzyme, we found that PAD1 was about 20-fold more active with dimethylallyl pyrophosphate (DMAPP) as the prenyl donor compared to DMAP. Our investigation also revealed that PAD1 is a surprisingly slow enzyme (kcat of ~12 h-1), allowing for analysis of pre-steady state kinetics. Both production of prFMN and consumption of FMN matched well to a first-order kinetic model with a rate constant of ~20 h-1, suggesting that a chemical step was rate-determining.
Recent Publication
N. Arunrattanamook and E.N.G. Marsh (2018). “Kinetic Characterization of Prenyl-Flavin Synthase from Saccharomyces cerevisiae”. Biochemistry. 57 (5), 696-700. DOI: 10.1021/acs.biochem.7b01131
Reaction of FDC with FNVB results in stable FNVB-prFMN cycloadduct
The proposed mechanism for FDC involves a 1,3-dipolar cycloaddition between the prFMN and the double bond of trans-cinnamic acid, but this chemistry has never been observed in enzymes before. We characterized the adduct that forms between prFMN and its substrate by using a mechanism based inhibitor designed to stabilize this adduct. 2-fluoro-2-nitro-vinylbenzene (FNVB) proved to be a potent inhibitor and the resulting FNVB-prFMN adduct was analyzed using UV-Vis spectroscopy and native tandem mass spectroscopy. The FNVB-prFMN adduct was confirmed to be a cycloadduct, providing the first direct evidence that FDC performs 1,3-dipolar cycloaddition chemistry and revealing that enzymes are capable of performing such chemistry.
Recent Publication
K.L. Ferguson, J.D. Eschweiler, B.T. Ruotolo, E.N.G. Marsh (2017). “Evidence for a 1,3-Dipolar Cyclo-addition Mechanism in the Decarboxylation of Phenylacrylic Acids Catalyzed by Ferulic Acid Decarboxylase”. J. Am. Chem. Soc. 139 (32), 10972-10975. DOI: 10.1021/jacs.7b05060
Analysis of FDC’s decarboxylation reaction suggests rate-determining step is cycloelimination step
Linear free energy relationships provide a powerful tool to probe reaction mechanisms, but are difficult to apply to enzyme reactions. Taking advantage of FDC’s tolerance for substituted aromatic substrates, we used Hammett analysis and kinetic isotope effects to diagnose the rate-limiting step in this complex reaction. Our analysis indicated that the rate-limiting step of FDC’s reaction was likely to be the cycloelimination step where styrene is produced, a notable contrast to most decarboxylation reactions in which the rate-determining step is decarboxylation.
Recent Publication
K.L. Ferguson, N. Arunrattanamook, E.N.G. Marsh (2016). “Mechanism of the Novel Prenylated Flavin-Containing Enzyme Ferulic Acid Decarboxylase Probed by Isotope Effects and Linear Free-Energy Relationships”. Biochemistry. 55 (20), 2857-2863. DOI: 10.1021/acs.biochem.6b00170
PAD1 synthesizes a modified FMN cofactor that is required for FDC decarboxylase activity
Initially, it was thought that the fungal enzymes ferulic acid decarboxylase (FDC) and phenylacrylic acid decarboxylase (PAD1) were isofunctional enzymes that performed the decarboxylation of phenylacrylic acids. However, we determined that only FDC possessed decarboxylase activity. We showed that PAD1 was mis-assigned as a decarboxylase and instead that it synthesized a modified form of FMN that functioned as the cofactor in the decarboxylation reactions catalyzed by FDC. The structure of this FMN-derived cofactor was shown by David Leys’ group to be prenylated-FMN.
Recent Publication
F. Lin, K.L. Ferguson, D.R. Boyer, X.N. Lin, E.N.G. Marsh (2015). “Isofunctional Enzymes PAD1 and UbiX Catalyze Formation of a Novel Cofactor Required by Ferulic Acid Decarboxylase and 4-Hydroxy-3-polyprenylbenzoic Acid Decarboxylase”. ACS Chem. Biol. 10 (4), 1137-1144. DOI: 10.1021/cb5008103