Clindamycin, which has been reported to have no significant in vitro activity against Toxoplasma gondii, actually markedly inhibits the growth of this parasite in infected human fibroblasts. When measured 3 days after treatment, the concentration required to reduce parasite growth by 50%o is about 1 ng/ml. Some observers failed to note this inhibition because of its markedly delayed onset. At 6 ng/ml, clindamycin is parasiticidal, and the rate and extent of parasite killing increase with higher drug concentrations. With the aid of chemical mutagenesis, we isolated a parasite mutant that is approximately 100-fold more resistant to clindamycin than is the wild type. Lincomycin inhibits T. gondii at a higher 50% inhibitory concentration, about 100 ng/ml. The clindamycin-resistant mutant is partially cross-resistant to lincomycin.While Derouin et al. (3) observed that clindamycin concentrations as low as 0.5 ng/ml inhibited the growth of Toxoplasma gondii, at least three other laboratories have reported that the drug has no in vitro activity (5,6,8). These latter observations are in marked contrast both to the well-established efficacy of clindamycin phosphate against toxoplasmosis in infected mice (1, 6) and to the preliminary evidence for the successful use of clindamycin phosphate to treat toxoplasmic encephalitis in AIDS patients (2). We show here that clindamycin is actually a potent parasiticidal agent in cultured cells infected with T. gondii, with a 50% inhibitory concentration (IC50) that is among the lowest reported for any antitoxoplasma drug. This activity previously escaped detection by several laboratories because the drug has little effect on parasite multiplication until many cell divisions have occurred. MATERIALS AND METHODSThe host cells for all experiments were confluent cultures of human fibroblasts that were maintained as previously described (13), except that the medium for growth contained 10% calf serum and that for infection contained 1% fetal bovine serum. The parasites were our cloned line (13) of the RH strain or mutants derived from this clone. The parasites were maintained by serial subculturing in human fibroblasts. Infected cells were disrupted by forced extrusion through a 25-gauge needle to release intracellular parasites. The newly released parasites were assayed by a plaque procedure (13). In all final experiments, this procedure was carried out with 25-cm2 flasks and the plaques were counted without staining. For preliminary plaque assays with multiwell trays, the monolayer was first fixed with 0.3 N trichloroacetic acid and then stained with Coomassie blue. Parasites were cloned by the detection of a single plaque 6 days after infection of 0.15-cm2 wells with suitably diluted suspensions of extracellular T. gondii.

Mutants of Toxoplasma gondii resistant to drugs that appear to affect the mitochondrial bc1 complex were isolated with the aid of mutagenesis with ethylnitrosourea. Mutant DeqR-1 was > 1,000-fold more resistant to decoquinate than was the wild type but more sensitive to atovaquone (formerly called 566C80). Mutant AtoR-1 was 20-fold more resistant to atovaquone than was the wild type and was also partially cross-resistant to decoquinate. Both drugs rapidly inhibited oxygen uptake by freshly prepared extracellular parasites, and the mutants were resistant to this inhibition. Neither the addition of uracil to the medium nor the use of a mutant of T. gondii with a defect in pyrimidine salvage had a substantial effect on the in vitro anti-parasitic activity of these drugs, suggesting that de novo pyrimidine synthesis was not the major biochemical target of either drug.

No abstract