Epigenetic Mechanisms of Chemical Carcinogenesis: Commentary

Anthony B. DeAngelo, Ph.D.*

National Health and Environmental Effects Research Laboratory, U.S. Environmental Protection Agency, Research Triangle Park, NC 27711

Tel: (919) 541-2568

Fax: (919) 541-0694

E-mail: deangelo.anthony@epa.gov

Klaunig, Kamendulis, and Yong have provided the reader with a breathtaking vista of cellular mechanisms by which non-genotoxic chemicals can bring about cancer in laboratory animals. They catalog an impressive amount of experimental data for each of the five epigenetic mechanisms considered in their article. What appears to be lacking is the sense that no one mechanism can account for the carcinogenicity of a particular non-genotoxic carcinogen and consideration of the interrelationships between mechanisms. In fairness, the balance between some of the cellular processes was mentioned but only in passing. However, the reader would benefit from an in-depth discussion of the interrelationships and dependency of mechanisms postulated to underlie the carcinogenicity of specific chemicals in order to gain an appreciation for the complexity of the cellular processes affected by non-genotoxic chemicals.

It has long been appreciated that liver cancer can not be predicted by altered hepatocyte proliferation¹ and that the disturbance of the mitotic-apoptotic balance² may be more fundamental to the induction of hepatocellular cancer by non-genotoxic chemicals. Dichloroacetic acid is a case in particular and illustrative of the complexity of the carcinogenic process. Dichloroacetic acid (DCA) was found to induce hepatocellular cancer in male rats and mice^3-4. A considerable body of data indicates that DCA acts through non-genotoxic mechanisms⁵. DCA did not enhance the proliferation of hepatocyes outside of pre-neoplastic and neoplastic lesions at any time point observed throughout the two year exposures. DCA actually inhibited hepatocyte proliferation early in the carcinogenic process and when added directly to hepatocytes in primary culture^6-8. The dose-time depression of hepatocyte proliferation was accompanied by a suppression of spontaneous apoptosis⁹.

DCA has been shown to induce peroxisome proliferation at doses higher than those required to enhance hepatocellular cancer⁴ and presumably through PPARa. Since there is evidence to link PPARa with the suppression of apoptosis¹⁰ and since DCA was reported to activate a reporter gene spliced into a PPARa" binding response element¹¹ it is not unreasonable to consider DCA acting through this receptor at doses that do not induce frank peroxisome proliferation. With respect to other members of the steroid receptor family, DCA inhibited the binding activity and subcellular localization of the glucocorticoid receptor in mouse liver¹² and inhibited the secretion of steroid hormones by the adrenal gland (unpublished observations). The glucocorticoid hormones are known to exert an effect on both cell proliferation and cell death.

Although cell proliferation and cell death were inhibited in mouse hepatocytes during short-term exposures to DCA^6-9, chronic exposures to the chemical resulted in an enhanced cell proliferation in large altered foci of cells (AFC) and adenomas relative to either non-involved hepatocytes or carcinomas. The enhanced proliferation seen in the large AFC and in adenomas was likely mediated by the enhanced expression of c-jun and could be correlated with a depressed rate of apoptosis and decreased cell size, resulting in the development of a carcinoma phenotype within these pre-neoplastic lesions^13-15. These alterations are consistent with the conclusion that the mechanism(s) of DCA hepatocarcinogenicity partially involves progressive cellular adaptation to a heptotoxic insult and selection of cytotoxicant-resistant cells.

For this one chemical a variety of mechanisms leading to perturbations in the control of cell proliferation and cell death within hepatocytes and within the various proliferative lesions leading to hepatocellular cancer appear to be in play. These would include direct disruption of the cell cycle, receptor mediated processes, altered gene expression, and possibly endocrine disruption. Not mentioned, but supported by experimental data, are DCA effects on cell-cell communication, the induction of oxidative damage and effects on DNA repair systems.

The authors have provided a valuable overview of non-genotoxic mechanisms of carcinogenesis which should provide interested investigators, and especially those new to the field, a foot-hold in a rather complex, confusing and sometimes contradictory arena of research. It can serve as a guide-post to the lines of research required for understanding chemically induced cancer and to alert the investigator to the dangers of putting too much significance on any one mechanism. Unstated, but clearly evident is the proposition that mechanistic research must be conducted and related to the conditions of the cancer bioassay, and that ultimately the subject in question (rodent, fish, human) has the final say.

REFERENCES