Carrier- and receptor-mediated transport of folate antagonists targeting folate-dependent enzymes: correlates of molecular-structure and biological activity.

Abstract

The transport properties and growth-inhibitory potential of 37 classic and novel antifolate compounds have been tested in vitro against human and murine cell lines expressing different levels of the reduced folate carrier (RFC), the membrane-associated folate binding protein (mFBP), or both. The intracellular targets of these drugs were dihydrofolate reductase (DHFR), glycinamide ribonucleotide transformylase (GARTF), folylpolyglutamate synthetase (FPGS), and thymidylate synthase (TS). Parameters that were investigated included the affinity of both folate-transport systems for the antifolate drugs, their growth-inhibitory potential as a function of cellular RFC/mFBP expression, and the protective effect of either FA or leucovorin against growth inhibition. Methotrexate, aminopterin, N10-propargyl-5,8-dideazafolic acid (CB3717), ZD1694, 5,8-dideazaisofolic acid (IAHQ), 5,10-dideazatetrahydrofolic acid (DDATHF), and 5-deazafolic acid (efficient substrate for FPGS) were used as the basic structures in the present study, from which modifications were introduced in the pteridine/quinazoline ring, the C9-N10 bridge, the benzoyl ring, and the glutamate side chain. It was observed that RFC exhibited an efficient substrate affinity for all analogues except CB3717, 2-NH2-ZD1694, and glutamate side-chain-modified FPGS inhibitors. Substitutions at the 2-position (e.g., 2-CH3) improved the RFC substrate affinity for methotrexate and aminopterin. Other good substrates included PT523 (N alpha-(4-amino-4-deoxypteroyl)-N delta-hemiphthaloyl-L-ornithine), 10-ethyl-10-deazaaminopterin, and DDATHF. With respect to mFBP, modifications at the N-3 and 4-oxo positions resulted in a substantial loss of binding affinity. Modifications at other sites of the molecule were well tolerated. Growth-inhibition studies identified a series of drugs that were preferentially transported via RFC (2,4-diamino structures) or mFBP (CB3717, 2-NH-ZD1694, or 5,8-dideazaisofolic acid), whereas other drugs were efficiently transported via both transport pathways (e.g., DDATHF, ZD1694, BW1843U89, or LY231514). Given the fact that for an increasing number of normal and neoplastic cells and tissue, different expression levels of RFC and mFBP are being recognized, this folate antagonist structure-activity relationship can be of value for predicting drug sensitivity and resistance of tumor cells or drug-related toxicity to normal cells and for the rational design and development of novel antifolates.