5-Deaza-7-desmethylene analogues of 5,10-methylene-5,6,7,8-tetrahydrofolic acid and related compounds: Synthesis and in vitro biological activity

1-4-(tert-Butyloxycarbonyl)phenyl]-3-pyrrolidinone and 1-3-(tert-butyloxycarbonyl)phenyl]-4-piperidinone were condensed with ethyl cyanoacetate or malononitrile to form ylidene derivatives, which were then subjected sequentially to (i) catalytic or chemical reduction, (ii) condensation with guanidine, and (iii) gentle tri-fluoroacetic acid treatment to obtain 3-(2,4-diamino-6(5H)-oxopyrimidin-5-yl)-1-(4-carboxyphenyl)pyrrolidine ( 27 ), 4-(2,4-diamino-6(5H)-oxopyrimidin-5-yl)-1-(carboxyphenyl)piperidine ( 35 ), and 3-(2,4,6-triaminopyrimidin-5-yl)-1-(carboxyphenyl)pyrrolidine ( 40 ). Condensation of 27, 35 , and 40 with diethyl or di-tert-butyl L-glutamate followed by removal of the ester groups yielded N-4-3-(2,4-diamino-6(5H)-oxopyrimidin-5-yl)pyr-rolidino]benzoyl]-L-glutamic acid ( 13 ), N-4-4-(2,4-diamino-6-(5H)-oxopyrimidin-5-yl)piperidino]benzoyl]-L-glutamic acid ( 14 ), and N-4-3-(2,4,6-triaminopyrimidin-5-yl)pyrrolidino]benzoyl]-L-glutamic acid ( 15 ). Compounds 13 and 14 may be viewed as 5-deaza-7-desmethylene analogues of 5,10-methylene-5,6,7,8-tetrahydrofolic and 5,10-ethylene-5,6,7,8-tetrahydrofolic acid, respectively. Compounds 13 and 15 were good substrates for mouse liver folylpolyglutamate synthetase, with K_m values of 20 and 18 μM and a relative first-order rate constant V_max/K_m of 2.2 (aminopterin = 1.0). In contrast, 14 was a very poor substrate, with a K_m of 490 μM and a relative V_max/K_m of 0.052. As expected from its structure, 15 was a dihydrofolate reductase inhibitor. However its potency was unexceptional (IC₅₀ = 1.2 μM). Compounds 13 and 14 were inactive at concentrations of up to 100 μM, and likewise showed no activity against thymidylate synthase or glycinamide ribotide formyltransferase, two other key enzymes of folate-mediated one-carbon metabolism. Compound 15 was moderately active as an inhibitor of the growth of cultured tumor cells (SCC25 human squamous cell carcinoma), with an IC₅₀ of 0.37 μM (72 hour exposure). By comparison the IC₅₀ of aminopterin was 0.0069 μM. Thus, even though 15 is a good folylpolyglutamate synthetase substrate, the deep-seated skeletal changes embodied in this structure are unfavorable for DHFR binding and may also be unfavorable for transport into cells.