Unlike mammalian cells, malarial parasites lack the enzymes to salvage preformed pyrimidines. For this reason, a combination of a thymidylate synthase inhibitor and the nucleoside thymidine should provide selective antimalarial activity even in the absence of any known active site differences between malarial and mammalian thymidylate synthases. To test this hypothesis, we evaluated the in vitro antimalarial activity of ICI D1694, a quinazoline antifolate that inhibits thymidylate synthase in mammalian cells. ICI D1694 inhibited the in vitro proliferation of Plasmodium falciparum with a 50% inhibitory concentration of 20 microM. As predicted, this antimalarial activity was not affected by the presence of 10 microM thymidine in the culture medium. In contrast, five different mammalian cells, several of which were susceptible to nanomolar levels of ICI D1694 in the absence of thymidine, were rescued by thymidine. At doses of 100 microM ICI D1694 and 10 microM thymidine, the proliferation of parasites was completely inhibited, but the proliferation of all mammalian cells remained unaffected. A test of susceptibility patterns among five different isolates of P. falciparum revealed that strains resistant to pyrimethamine, cycloguanil, or chloroquine had susceptibilities to ICI D1694 essentially the same as those of wild-type parasites. These findings are consistent with the hypothesis that, intracellularly, ICI D1694 inhibits P. falciparum thymidylate synthase. Overall, it is clear that even with an inhibitor of malarial thymidylate synthase that is not particularly effective in itself, one can obtain selective inhibition of parasites if the antimalarial agent is used in combination with thymidine. More effective inhibitors of malarial thymidylate synthase will undoubtedly lead to selective chemotherapy in vivo.