Abstract 1943 Strategies for new anti-folate development targeting the dihydroneopterin aldolase enzyme

The highly successful antibacterial treatments trimethoprim and sulfamethoxazole target enzymes within the folate biosynthetic pathway, yet no inhibitors for other enzymes within this pathway have come to market. Our objective is to explore the feasibility of new anti-folate approaches, and we have focused on the dihydroneopterin aldolase (DHNA) enzyme. An early structure-based screening campaign for DHNA inhibitors was carried out by Abbott labs and reported the enzyme could be readily targeted (Sanders, et al, 2004). However, DHNA inhibitors failed to halt growth of bacterial cells, and this was attributed to high intracellular levels of 7,8-dihydroneopterin, the substrate for DHNA. Subsequent studies identified a unique shunt within the folate pathway of M. tuberculosis that could be manipulated for feedback inhibition of DHNA within cells (Hajian, et al., 2019). Overall, it appears feasible to inhibit DHNA as a means of controlling bacterial cell growth, but such compounds must overcome the high intracellular concentrations of substrate. We used classical methodological approaches combining enzyme kinetics, crystallography, and in silico docking. Kinetic studies included screening for inhibitors based on structural similarity to the native substrate; however, this produced little useful chemical material for further development. We supplemented these with crystallographically determined structures of the DHNA co-crystallized and also soaked with different potential binding compounds. These structures all confirmed a canonical octameric oligomer with active sites spanning between two monomers. Mutagenesis of the active site lysine allowed us to capture the substrate within the pocket, and an energetic strain in the conformation around the scissle bond was evident. Additional mutagenesis followed by kinetic and structural determination gave insights into residues critical for product release. Docking studies explored the energetics of interactions of substrate and product, mimetics of each, and expanded to include other known folate inhibitors. While some compounds had stronger docked energies, they also highlighted a two-fold asymmetry within the overall octameric oligomer, such that every other active site adopted one of two conformations consistent with processive catalysis. Among notable features within the binding pocket of the active sites is a highly conserved cysteine, consistent with other aldolase families. We determined that this Cys is not catalytic; however, alkylation of this residue significantly reduced kinetic turnover and distorted the active site geometry. The culmination of these investigations is a proposal that a covalent inhibitor with a reactive group appended to a substrate-like pterin structure would impart specificity for folate pathway enzymes, including DHNA, and would overcome the previously identified limitation imposed by high intracellular substrate concentrations. Sanders, et al. (2004) J Med Chem 47(7):1709. Hajian, et al. (2019) Cell Chem Biol 26(6):781. Funding provided by an Institutional Development Awards from NIGMS, grant numbers P20GM103447 and P20GM103640, and internal funding from the OU Vice President for Research and Partnerships.