We studied the effects of nitroso-chloramphenicol, chloramphenicol, amino-chloramphenicol, and thiamphenicol on the activity of mitochondrial DNA polymerase of rat liver. 3H-thymidine triphosphate incorporation into DNA was used to measure the DNA polymerase activity in the mitochondrial matrix fraction. This fraction was in the supernatant of sonicated mitochondria obtained by ultracentrifugation. Under standard experimental conditions, thymidine triphosphate incorporation was time dependent up to 10 minutes. This activity was enhanced by beta-mercaptoethanol and was blocked by the known polymerase inhibitors ethidium bromide and 2',3'-dideoxythymidine 5'-triphosphate. Chloramphenicol and its analogues, amino-chloramphenicol and thiamphenicol, did not have a significant effect on the polymerase activity, whereas nitroso-chloramphenicol was inhibitory. The degree of inhibition was dependent on the experimental conditions. Thus, in the absence of beta-mercaptoethanol, nitroso-chloramphenicol caused inhibition; however, in its presence, there was no significant inhibitory effect. Under similar conditions, the addition of dithiothreitol also provided partial protection. On the other hand, the inhibition by nitroso-chloramphenicol was significantly enhanced with its preincubation in the mitochondrial matrix fraction before the addition of nucleotides and DNA; thus after 40 minutes of preincubation, nitroso-chloramphenicol at a concentration of 200 mumol/L gave 53% inhibition, and produced total inhibition at 600 mumol/L. The addition of NADH or NADPH to the preincubation medium produced substantial protection against nitroso-chloramphenicol, whereas nicotinamide-adenine dinucleotide had no effect. These results suggest that mitochondrial DNA polymerase may be a target for nitroso-chloramphenicol action. The potentiation of that action by preincubation and the protection against it by NADH and NADPH suggest the involvement of intermediate metabolic steps for maximal inhibition.