Much progress has been made in elucidating the biochemical and molecular mechanisms that underlie aflatoxin carcinogenesis. In humans, biotransformation of AFB1 to the putative carcinogenic intermediate, AFB-8,9-exo-epoxide, occurs predominantly by cytochromes P450 1A2 and 3A4, with the relative importance of each dependent upon the relative magnitude of expression of the respective enzymes in liver. Genetic variability in the expression of these and other cytochromes P450 may result in substantial interindividual differences in susceptibility to the carcinogenic effects of aflatoxins. Detoxification of AFB-8,9-epoxide by a specific alpha class glutathione S-transferase is an important protective mechanism in mice, and it accounts for the resistance of this species to the carcinogenic effects of AFB. This particular form of GST is expressed constitutively only at low levels in rats, but it is inducible by antioxidants such as ethoxyquin, and it accounts for much of the chemoprotective effects of a variety of substances, including natural dietary components that putatively act via an 'antioxidant response element' (ARE). In humans, the constitutively expressed GSTs have very little activity toward AFB1-8,9-exo-epoxide, suggeting that - on a biochemical basis - humans should be quite sensitive to the genotoxic effects of aflatoxins. If a gene encoding a high aflatoxin-active form of GST is present in the human genome, but is not constitutively expressed, and is inducible by dietary antioxidants (as occurs in rats), then chemo- and/or dietary intervention measures aimed at inducing this enzyme could be highly effective. However, as it is possible that human CYP 1A2 may also be inducible by these same chemicals (because of the possible presence of an ARE in this gene), the ultimate consequence of dietary treatment with chemicals that induce biotransformation enzymes via an ARE is uncertain. The balance of the rate of activation (exo-epoxide production) to inactivation (GST conjugation plus other P450-mediated non-epoxide oxidations) may be a strong indicator of individual and species susceptibility to aflatoxin carcinogenesis, if the experimental conditions are reflective of true dietary exposures. There is strong evidence that AFB-8,9-exo-epoxide binds to G:C rich regions of DNA, forming an adduct at the N7-position of guanine. Substantial evidence demonstrates that AFB1-8,9-epoxide can induce activating mutations in the ras oncogene in experimental animals, primarily at codon 12. Information on the activation of other oncogenes by aflatoxin is limited, and, to date, few (if any) studies have demonstrated an activated c-ras oncogene in human liver tumors from aflatoxin endemic areas. In contrast, substantial evidence implicates aflatoxin-induced G:C mutations (both G → T transversions and G → A transitions) in the inactivation of the human p53 tumor suppressor gene. These mutations occur with an extraordinary frequency at codon 249 of p53 and may ultimately serve as a 'carcinogen-specific' marker of aflatoxin-induced liver cancers in humans. The presence of hepatitis B virus appears to act synergistically with aflatoxin in the development of liver cancer. Analysis of epidemiological data of p53 mutations in the presence and/or absence of hepatitis B virus surface antigens suggests that the production of codon 249 mutations in p53 by aflatoxin may be enhanced by hepatitis virus infection, providing a rational, mechanistic basis for the observed synergy between these two human cancer risk factors.
