The 3<i>S</i> Enantiomer Drives Enolase Inhibitory Activity in SF2312 and Its Analogues

• Published in: Molecules
• Main Authors: Federica Pisaneschi, Yu-Hsi Lin, Paul G. Leonard, Nikunj Satani, Victoria C. Yan, Naima Hammoudi, Sudhir Raghavan, Todd M. Link, Dimitra K. Georgiou, Barbara Czako, Florian L. Muller
• Format: Article
• Language: English
• Published: MDPI AG 2019-07-01
• Subjects: glycolysis; enolase; chiral; phosphonate; hydroxamate; <i>E. coli</i>; enzyme inhibitor; enzyme structure; natural product; X-ray crystallography
• Online Access: https://www.mdpi.com/1420-3049/24/13/2510

We recently reported that SF2312 ((1,5-dihydroxy-2-oxopyrrolidin-3-yl)phosphonic acid), a phosphonate antibiotic with a previously unknown mode of action, is a potent inhibitor of the glycolytic enzyme, Enolase. SF2312 can only be synthesized as a racemic-diastereomeric mixture. However, co-crystal structures with Enolase 2 (ENO2) have consistently shown that only the (3<i>S</i>,5<i>S</i>)-enantiomer binds to the active site. The acidity of the alpha proton at C-3, which deprotonates under mildly alkaline conditions, results in racemization; thus while the separation of four enantiomeric intermediates was achieved via chiral High Performance Liquid Chromatography (HPLC) of the fully protected intermediate, deprotection inevitably nullified enantiopurity. To prevent epimerization of the C-3, we designed and synthesized MethylSF2312, ((1,5-dihydroxy-3-methyl-2-oxopyrrolidin-3-yl)phosphonic acid), which contains a fully-substituted C-3 alpha carbon. As a racemic-diastereomeric mixture, MethylSF2312 is equipotent to SF2312 in enzymatic and cellular systems against Enolase. Chiral HPLC separation of a protected MethylSF2312 precursor resulted in the efficient separation of the four enantiomers. After deprotection and inevitable re-equilibration of the anomeric C-5, (3<i>S</i>)-MethylSF2312 was up to 2000-fold more potent than (3<i>R</i>)-MethylSF2312 in an isolated enzymatic assay. This observation strongly correlates with biological activity in both human cancer cells and bacteria for the 3<i>S</i> enantiomer of SF2312. Novel X-ray structures of human ENO2 with chiral and racemic MethylSF2312 show that only (3<i>S,</i>5<i>S)</i>-enantiomer occupies the active site. Enolase inhibition is thus a direct result of binding by the (3<i>S,</i>5<i>S)</i>-enantiomer of MethylSF2312. Concurrent with these results for MethylSF2312, we contend that the (3<i>S</i>,5<i>S</i>)-SF2312 is the single active enantiomer of inhibitor SF2312.