Ronald J. Lorentzen

Carcinogenicity and mechanism of action of fumonisin B1: a mycotoxin produced by Fusarium moniliforme (= F. verticillioides).

Fumonisins are fungal metabolites and suspected human carcinogens. They inhibit ceramide synthase in vitro, enhance tumor necrosis factor alpha (TNFalpha) production, and cause apoptosis. Fumonisin B1 (FB1) was fed to rats and mice for 2 years or, in separate studies, given to rats or mice for up to 4 weeks. Kidney tubule adenomas and carcinomas were found in male rats fed > or = 50 ppm, whereas liver adenomas and carcinomas were found in female mice fed > or = 50 ppm for 2 years. In the short-term studies, increases in tissue concentration of the ceramide synthase substrate sphinganine (Sa) and the Sa to sphingosine (So) ratio were correlated with apoptosis. Further, hepatotoxicity was ameliorated in mice lacking either the TNFR1 or the TNFR2 TNFalpha receptors. Thus, FB1 was carcinogenic to rodents and thefindings support the hypothesis that disrupted sphingolipid metabolism and TNFalpha play important roles in its mode of action.

Benzo[a]pyrene dione‐benzo[a]pyrene diol oxidation‐reduction couples; involvement in DNA damage, cellular toxicity, and carcinogenesis

Three isomeric quinone metabolites of the environmental carcinogen benzo[a]pyrene undergo reversible, univalent oxidation-reduction cycles involving the corresponding benzo[a]pyrene diols and intermediate semiquinone radicals. Under anaerobic conditions, benzo[a]pyrene 1,6-dione, benzo[a]pyrene 3,6-dione, and benzo[a]pyrene 6,12-dione are readily reduced by mild biological agents such as NADH and glutathione. The benzo[a]pyrene diols, in turn, are very rapidly autooxidized to diones when exposed to air. Substantial amounts of hydrogen peroxide are produced during these autooxidations. The benzo[a]pyrene diol/benzo[a]pyrene dione interconversions proceed by one-electron steps; the corresponding semiquinone radicals were detected as intermediates when the reactions were carried out at high pH. Benzo[a]pyrene diones are electron-acceptor substrates for NADH dehydrogenase. Catalytic amounts of these metabolites, together with this respiratory enzyme, function as cyclic oxidation-reduction couples to link NADH and molecular oxygen in the continuous production of hydrogen peroxide. Benzo[a]pyrene diones induce strand scissions when incubated with T7 DNA. The damage is modified by conditions that indicate that reduced oxygen species propagate the reactions responsible for strand scission. Benzo[a]pyrene diones are cytotoxic at low concentrations to cultured hamster cells. The cytotoxic effect can be substantially reduced by depletion of oxygen from the growth medium and the atmosphere in which the cells are incubated. The results support the hypothesis that the biological activity of benzo[a]pyrene diones is due to the regenerative oxidation-reduction cycles involving quinone and hydroquinone forms; activated oxygen species and semiquinone radicals formed during these cycles are most likely responsible for the observed cytotoxic action. The role of activated oxygen species in carcinogenesis is discussed.

Comparing Human and Animal Cancer Risk Models Using Multistage Theory: Exponential vs. Relative vs. Additive Risk

Comparing human and animal cancer risk assessments often results in inconsistencies due to different extrapolation methodologies. Descriptive relative risk models using cumulative dose are typically used in human epidemiological studies, whereas multistage additive risk models using dose rate are typically used in high dose animal studies to extrapolate to human exposures. Biologically motivated nonclonal and clonal multistage risk models are reviewed for their unifying role in consistently comparing human and animal risks and for suggesting alternate cumulative dose epidemiological risk models.

Carcinogenic Risk of Benzene Derived from Animal and Human Data

Benzene is widely considered to be an animal and human carcinogen but there are wide-ranging estimates of its potency. Risk assessments of benzene and human leukemia have primarily focused on exposure by inhalation over an occupational lifetime. The Food and Drug Administration (FDA) regulates benzene as a contaminant in food additives and thus is interested in the risk from lifetime exposure by the oral route. The risk from lifetime ingestion of benzene is assessed using the NTP animal gavage studies and the National Institute of Occupational Safety and Health (NIOSH) epidemiology studies, and the results are compared. The unit risks derived from the human epidemiology data for leukemia are similar to those derived from the rodent data when the animal risks are summed over all organ sites in each sex grouping. The collective unit risk values calculated from the animal data were.038 and.039 per mg/kg body weight/day exposure for male and female mice, respectively. The unit risks calculated from the human epidemiology data were.043 and.024 per mg/kg body weight/day exposure using the relative and absolute risk models, respectively.

Consideration of Cell Proliferation and Adduct Formation as Parameters Affecting Risk Outcome

For the most part, the quantitative aspects of risk assessment for carcinogens currently involve using standard animal bioassay data and making assumptions about dose-response such as linearity at low-dosage levels and no threshold for the effect. These procedures are not considered to predict actual risk but to reflect an upper bound on risk. The true, but indeterminate, risk could be as high as this upper bound value and as low as zero. As we learn more about mechanisms of carcinogenesis, we anticipate that more biological information will lead to refinement of predicted dose-response curves and, hence, predicted risk. Two types of biological information currently being explored for their value in this regard are cell proliferation and covalent adduct formation with cellular macromolecules. Both of these types of data have the potential to allow reliable characterization of the dose-response curve to lower doses than is possible with bioassay data alone. This may allow in certain cases for the predictions of the upper bound risk to be closer to the actuarial risk at the low exposure levels usually encountered. Much work and careful interpretation need to be done before these approaches become routinely viable. However, the time has come to take seriously the potential that this kind of biological information can have on quantitative risk assessment.