Previous work in our laboratory has shown that dietary megadoses of nicotinamide, used in the prevention of diabetes, cause increases in hepatic poly(ADP-ribose). Poly(ADP-ribose) is synthesized from NAD+ by a nuclear enzyme, poly(ADP-ribose)polymerase, which is activated by DNA strand breaks. The nicotinamide-induced increase in poly(ADP-ribose) could result from an increase in substrate, NAD+, or the induction of strand breaks in DNA. Strand breaks may result from the depletion of single carbon groups, through the excretion of methylated derivatives of nicotinamide. To differentiate between these mechanisms, a 3 x 3 factorial experiment was conducted in which rats were fed diets containing various supplements of choline bitartrate (0, 2, 20 g/kg diet) and nicotinamide (0, 1, 2 g/kg diet). At the conclusion of treatments, blood NAD+ and liver lipid, NAD+ and poly(ADP-ribose) levels were determined. Choline deficiency caused the characteristic accumulation of fat in the liver at all levels of nicotinamide. In choline deficient rats, nicotinamide supplements further increased liver lipid concentration. Blood and liver NAD+ concentrations were increased by nicotinamide supplementation, irrespective of choline status. In contrast, liver poly(ADP-ribose) levels were increased by nicotinamide supplementation only in choline deficient rats. These results show that nicotinamide-induced increases in poly(ADP-ribose) levels appear to be dependent on decreased methyl donor status and suggest that adequate choline status is important for preventing some deleterious effects of nicotinamide treatment.