We tested the hypothesis that the mechanism of action of the antifolate drug trimethoprim is through accumulation of bacterial dihydrofolate resulting in depletion of tetrahydrofolate coenzymes required for purine and pyrimidine biosynthesis. The folate pool of a strain of Escherichia coli (NCIMB 8879) was prelabeled with the folate biosynthetic precursor [(3)H]-p-aminobenzoic acid before treatment with trimethoprim. Folates in untreated E. coli were present as tetrahydrofolate coenzymes. In trimethoprim-treated cells, however, a rapid transient accumulation of dihydrofolate occurred, followed by complete conversion of all forms of folate to cleaved catabolites (pteridines and para-aminobenzoylglutamate) and the stable nonreduced form of the vitamin, folic acid. Both para-aminobenzoylglutamate and folic acid were present in the cell in the form of polyglutamates. Removal of trimethoprim resulted in the reconversion of the accumulated folic acid to tetrahydrofolate cofactors for subsequent participation in the one-carbon cycle. Whereas irreversible catabolism is probably bactericidal, conversion to folic acid may constitute a bacteriostatic mechanism since, as we show, folic acid can be used by the bacteria and proliferation is resumed once trimethoprim is removed. Thus, the clinical effectiveness of this important drug may depend on the extent to which the processes of either catabolism or folic acid production occur in different bacteria or during different therapeutic regimes.