Mechanism-based inhibition of human folylpolyglutamate synthetase: Design, synthesis, and biochemical characterization of a phosphapeptide mimic of the tetrahedral intermediate

Folylpolyglutamate synthetase (FPGS) catalyzes an ATP-dependent ligation reaction that results in the synthesis of poly(γ-glutamate) metabolites of folates and some antifolates. We have synthesized and characterized the prototype of a new class of mechanism-based FPGS inhibitor in which a phosphonate moiety mimics the tetrahedral intermediate formed during the ligation reaction. This phosphonate, 4-amino-4-deoxy-10-methyl-pteroyl-L- glutamyl-γ-[Ψ{P(O) (OH)-O}]glutarate (4-NH₂-10-CH₃-Pte-L-Glu-γ- [Ψ{P(O)(OH)-O}]glutarate), is not a substrate for human FPGS, but is a linear, competitive inhibitor (K(in) = 46 nM) with respect to methotrexate as the variable substrate. Inhibition is not time-dependent and preincubation of FPGS with this phosphonate does not increase the degree of inhibition, suggesting that it is not a slow, tight-binding inhibitor involving a time- dependent isomerization, EI → EI*. Substructures containing the phosphonate moiety but lacking the pterin are much less inhibitory to FPGS, indicating that a significant portion of the inhibitor binding energy is derived from the pterin moiety, a feature also observed in substrate binding. 4-NH₂-10- CH₃-Pte-L-Glu-γ-[Ψ{P(O)(OH)-O}]glutarate is also an analog of a proposed tetrahedral intermediate in the reaction catalyzed by γ-glutamyl hydrolase (γ-GH), another enzyme of importance in controlling folate homeostasis in cells. This intermediate would arise from direct attack of H₂O on the dipeptide, 4-NH₂-10-CH₃-Pte-L-Glu-γ-L-Glu. The fact that 4-NH₂-10-CH₃- Pte-L-Glu-γ-[Ψ{P(O)(OH)-O}]glutarate is not an inhibitor of γ-GH strongly suggests that hydrolysis of poly-γ-glutamates catalyzed by γ-GH does not involve the direct attack of water at the scissile amide bond. Methotrexate, its γ-glutamyl dipeptide metabolite, and 4-NH₂-10-CH₃-Pte- L-Glu-γ-[Ψ{P(O)(OH)-O}]glutarate are equipotent as inhibitors of human dihydrofolate reductase (the primary target of methotrexate), but the phosphonate does not significantly inhibit another important folate-dependent enzyme, thymidylate synthase. Thus, the phosphonate moiety in this analog represents an important new lead in the development of FPGS inhibitors.