An 18-month carcinogenicity study was conducted in male weanling F344 rats (28/group) to examine the effects of the simultaneous feeding of selected concentrations of ethionine and 0.05% phenobarbital in a normal chow diet. The effects of a 1-6-week feeding of phenobarbital and ethionine on the hepatic levels of the related metabolites S-adenosylmethionine, S-adenosylhomocysteine and S-adenosylethionine were also examined. Ethionine at 0.3% or 0.1% induced hepatocellular carcinoma (HCCa) at incidences of 90% (19/21) and 89% (24/27), respectively. Adding phenobarbital to the 0.1% ethionine diet reduced the incidence of HCCa to 36% (10/28) and reduced the number of liver tumor-associated deaths occurring prior to terminal sacrifice from 10/27 to 1/28. No hepatic tumors were observed in rats fed 0, 0.003, 0.01, or 0.03% ethionine. Phenobarbital alone or combined with 0.03% ethionine produced no hepatic tumors. Dietary ethionine at 0.1% reduced the intracellular hepatic level of S-adenosylmethionine to <50% of that seen in control rats. Phenobarbital alone had little effect on either S-adenosylmethionine or S-adenosylhomocysteine levels. The combination of phenobarbital and 0.1% ethionine led to increases in the hepatic levels of S-adenosylmethionine of 40-60% after 3 and 6 weeks of feeding, compared to those seen in rats receiving 0.1% ethionine alone. Ethionine feeding resulted in high levels of S-adenosylethionine in the livers. Combining phenobarbital with ethionine in the diet led to 30-50% reductions in hepatic S-adenosylethionine content. The results indicate that phenobarbital inhibits hepatocarcinogenesis by ethionine, that ethionine may cause HCCa via methyl group insufficiency, and that at levels of < or =0.03% ethionine did not show evidence of tumorigenicity.