In this review, the role of dietary antioxidants in the prevention of hepatocarcinogenesis is examined. Both human and animal models are discussed. Vitamin C, vitamin E, and selenium are antioxidants that are essential in the human diet. A number of non-essential chemicals also contain antioxidant activity and are consumed in the human diet, mainly as plants or as supplements, including beta-carotene, ellagic acid, curcumin, lycopene, coenzyme Q(10), epigallocatechin gallate, N-acetyl cysteine, and resveratrol. Although some human and animal studies show protection against carcinogenesis with the consumption of higher amounts of antioxidants, many studies show no effect or an enhancement of carcinogenesis. Because of the conflicting results from these studies, it is difficult to make dietary recommendations as to whether consuming higher amounts of specific antioxidants will decrease the risk of developing hepatocellular carcinoma.

PCBs are organic pollutants that persist and bioaccumulate in the environment. These chemicals induce and promote liver tumors in rodents. Previous studies have shown that they increase oxidative stress in the liver, including lipid peroxidation, oxidative DNA damage, and NF-κB activation. The objective of these studies was to determine if the promoting activities of PCBs could be inhibited by dietary antioxidants (vitamin E, selenium, or phytochemicals) or by knocking out the p50 subunit of NF-κB. In the antioxidant studies, female rats were first injected with DEN (150 mg/kg) and then administered 4 biweekly i.p. injections (300 μmol/kg/injection) of PCB-77, PCB-153, or vehicle; the number and volume of placental glutathione S-transferase (PGST)-positive foci were then quantified. Vitamin E did not influence the promoting activities of PCBs. Increasing dietary selenium above the recommended intake increased the number of foci induced but decreased their volume. Most of the phytochemicals examined (N-acetyl cysteine, β-carotene, resveratrol, EGCG) had no significant effect on the promoting activity of PCB-77. Ellagic acid increased and lycopene decreased the number of foci; ellagic acid, CoQ 10 , and curcumin decreased the volume of foci. In the NF-κB knockout study, male mice were first injected with DEN (90 mg/kg); controls not receiving DEN were also studied. Both p50 −/− and wild-type mice were then injected biweekly 20 times with PCB-153 (300 (μmol/kg). In DEN-treated and DEN + PCB-treated mice, the incidence of tumors was lower in the p50 −/− mice than in wild-type mice. In mice receiving PCB-153, the tumor incidence and tumor volume were higher. The volume of tumors that were positive for glutamine synthetase was increased in mice administered PCB-153. This study shows that the promotion of hepatocarcinogenesis by PCBs is largely unaffected by dietary antioxidants but is diminished when NF-κB activation is impaired by the absence of the p50 subunit.

In this study, the effect of dietary vitamin E on the hepatic tumor-promoting activity of PCB-77 and PCB-153 in female Sprague-Dawley rats (175-200 g) was investigated. One week after diethylnitrosamine injection, rats were fed purified diets containing 10, 50, or 250 mg/kg vitamin E in the form of alpha-tocopheryl acetate. Starting 1 wk later, we injected rats i.p. with vehicle (corn oil) or PCB-77 or PCB-153 (300 mumol/kg) every 14 d for 4 injections. All rats were killed 10 d after the last PCB injection. The number and volume of placental glutathione S-transferase (PGST)-positive foci were increased by PCB-77 but not by PCB-153. Vitamin E did not affect the induction of PGST-positive foci. PCB-77, but not PCB-153, increased hepatic NF-kappaB activity. In conclusion, dietary vitamin E supplementation does not protect against the induction of altered hepatic focal lesions by PCBs.

Polychlorinated biphenyls (PCBs) are persistent organic pollutants that have promoting activity in the liver. PCBs induce oxidative stress, which may influence carcinogenesis. Epidemiological studies strongly suggest an inverse relationship between dietary selenium (Se) and cancer. Despite evidence linking Se deficiency to hepatocellular carcinoma and liver necrosis, the underlying mechanisms for Se cancer protection in the liver remain to be determined. We examined the effect of dietary Se on the tumor promoting activities of two PCBs congeners, 3,3', 4,4'-tetrachlorobiphenyl (PCB-77) and 2,2', 4,4', 5,5'-hexachlorobiphenyl (PCB-153) using a 2-stage carcinogenesis model. An AIN-93 torula yeast-based purified diet containing 0.02 (deficient), 0.2 (adequate), or 2.0 mg (supplemental) selenium/kg diet was fed to Sprague-Dawley female rats starting ten days after administering a single dose of diethylnitrosamine (150 mg/kg). After being fed the selenium diets for 3 weeks, rats received four i.p. injections of either PCB-77 or PCB-153 (150 micromol/kg) administered every 14 days. The number of placental glutathione S-transferase (PGST)-positive foci per cm(3) and per liver among the PCB-77-treated rats was increased as the Se dietary level increased. Unlike PCB-77, rats receiving PCB-153 did not show the same Se dose-response effect; nevertheless, Se supplementation did not confer protection against foci development. However, the 2.0 ppm Se diet reduced the mean focal volume, indicating a possible protective effect by inhibiting progression of preneoplastic lesions into larger foci. Cell proliferation was not inhibited by Se in the liver of the PCB-treated groups. Se did not prevent the PCB-77-induced decrease of hepatic Se and associated reduction in glutathione peroxidase (GPx) activity. In contrast, thioredoxin reductase (TrxR) activity was not affected by the PCBs treatment or by Se supplementation. These findings indicate that Se does not inhibit the number of PGST-positive foci induced during promotion by PCBs, but that the size of the lesions may be inhibited. The effects of Se on altered hepatic foci do not correlate with its effects on GPx and TrxR.