Classical antifolate inhibitors of thymidylate synthase (TS) often require the reduced folate uptake system in order to exert their antitumor effects. In addition, these analogues are polyglutamylated via the enzyme folylpoly-gamma-glutamate synthetase (FPGS), which prevents analogue efflux from the cell and usually increases their inhibitory potency against TS. Impaired function of the reduced folate uptake system and that of FPGS are potential sources of resistance to such antifolates. We designed and synthesized a classical 6-5 ring-fused analogue N-[4-[(2-amino-6-methyl-3,4-dihydro-4-oxo-7H-pyrrolo[2,3- d]pyrimidin-5-yl)thio]-benzoyl]-L-glutamic acid (5) and a nonclassical 6-5 ring-fused analogue 2-amino-6-methyl-5-(pyridin-4-ylthio)-3,4-dihydro-4-oxo-7H-pyrrolo [2,3- d]pyrimidine (6) as TS inhibitors and antitumor agents. The syntheses of analogues 5 and 6 were achieved via the oxidative addition of the sodium salt of ethyl 4-mercaptobenzoate or 4-mercaptopyridine to 2-(pivaloylamino)-6-methyl-3,4-dihydro-4-oxo-7H-pyrrolo[2,3-d]pyri midine (17) in the presence of iodine. For the synthesis of 5 the ester obtained from the reaction was deprotected and coupled with diethyl L-glutamate followed by saponification. Compound 5 was a potent inhibitor of human and bacterial TS with IC50 values of 42 and 21 nM, respectively. Compound 6 was 10-fold less potent than 5 against human TS but more than 4700-fold less potent than 5 against Lactobacillus casei TS. The classical analogue 5 was neither a substrate nor an inhibitor of human FPGS derived from CCRF-CEM cells. Compound 5 was cytotoxic to CCRF-CEM and FaDu tumor cell lines as well as to an FPGS-deficient subline of CCRF-CEM. Thymidine protection studies established that TS was the primary target of 5.