Open AccessRegioisomerization of Antimalarial Drug WR99210 Explains the Inactivity of a Commercial Stock
Author(s) -
Thomas P. Remcho,
Sravanthi Devi Guggilapu,
Phillip Cruz,
Glenn Nardone,
Gavin D. Heffernan,
Robert O’Connor,
Carole A. Bewley,
Thomas E. Wellems,
Kristin D. Lane
Publication year - 2020
Publication title -
antimicrobial agents and chemotherapy
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 2.07
H-Index - 259
eISSN - 1070-6283
pISSN - 0066-4804
DOI - 10.1128/aac.01385-20
Subject(s) - dihydrofolate reductase , plasmodium falciparum , active site , stereochemistry , enzyme , chemistry , structural isomer , nuclear magnetic resonance spectroscopy , biochemistry , biology , combinatorial chemistry , malaria , immunology
WR99210, a former antimalarial drug candidate now widely used for the selection of Plasmodium transfectants, selectively targets the parasite's dihydrofolate reductase thymidine synthase bifunctional enzyme (DHFR-TS) but not human DHFR, which is not fused with TS. Accordingly, WR99210 and plasmids expressing the human dhfr gene have become valued tools for the genetic modification of parasites in the laboratory. Concerns over the ineffectiveness of WR99210 from some sources encouraged us to investigate the biological and chemical differences of supplies from two different companies (compounds 1 and 2). Compound 1 proved effective at low nanomolar concentrations against Plasmodium falciparum parasites, whereas compound 2 was ineffective, even at micromolar concentrations. Intact and fragmented mass spectra indicated identical molecular formulae of the unprotonated (free base) structures of compounds 1 and 2; however, the compounds displayed differences by thin-layer chromatography, reverse-phase high-performance liquid chromatography, and UV-visible spectroscopy, indicating important isomeric differences. Structural evaluations by 1 H, 13 C, and 15 N nuclear magnetic resonance spectroscopy confirmed compound 1 as WR99210 and compound 2 as a dihydrotriazine regioisomer. Induced fit computational docking models showed that compound 1 binds tightly and specifically in the P. falciparum DHFR active site, whereas compound 2 fits poorly to the active site in loose and varied orientations. Stocks and concentrates of WR99210 should be monitored for the presence of regioisomer 2, particularly when they are not supplied as the hydrochloride salt or are exposed to basic conditions that may promote rearrangement. Absorption spectroscopy can serve for assays of the unrearranged and rearranged triazines.