Pharmacokinetics, metabolism, and carcinogenicity of arsenic.

The carcinogenicity of arsenic in humans has been unambiguously demonstrated in a variety of epidemiological studies encompassing geographically diverse study populations and multiple exposure scenarios. Despite the abundance of human data, our knowledge of the mechanism(s) responsible for the carcinogenic effects of arsenic remains incomplete. A deeper understanding of these mechanisms is highly dependent on the development of appropriate experimental models, both in vitro and in vivo, for future mechanistic investigations. Suitable in vitro models would facilitate further investigation of the critical chemical species (arsenate/arsenite/MMA/DMA) involved in the carcinogenic process, as well as the evaluation of the generation and role of ROS. Mechanisms underlying the clastogenic effects of arsenic, its role in modulating DNA methylation, and the phenomenon of inducible tolerance could all be more completely investigated using in vitro models. The mechanisms involved in arsenic's inhibition of ubiquitin-mediated proteolysis demand further attention, particularly with respect to its effects on cell proliferation and DNA repair. Exploration of the mechanisms responsible for the protective or anticarcinogenic effects of arsenic could also enhance our understanding of the cellular and molecular interactions that influence its carcinogenicity. In addition, appropriate in vivo models must be developed that consider the action of arsenic as a promoter and/or progressor. In vivo models that allow further investigation of the comutagenic effects of arsenic are also especially necessary. Such models may employ initiation-promotion-progression bioassays or transgenic animals. Both in vitro and in vivo models have the potential to greatly enhance our current understanding of the cellular and molecular interactions of arsenic and its metabolites in target tissues. However, refinement of our knowledge of the mechanistic aspects of arsenic carcinogenicity is not alone sufficient; an understanding of the pharmacokinetics and target tissue doses of the critical chemical species is essential. Additionally, a more thorough characterization of species differences in the tissue kinetics of arsenic and its methylated metabolites would facilitate the development of more accurate and relevant PBPK models. Improved models could be used to further investigate the existence of a methylation threshold for arsenic and its relevance to arsenic carcinogenicity in humans. The significance of alterations in relative tissue concentrations of SAM and SAH deserves further attention, particularly with respect to their role in modulating methyltransferases involved in arsenic metabolism and DNA methylation. The importance of genetic polymorphisms and nutrition in influencing methyltransferase activities must not be overlooked. In vivo models are necessary to evaluate these factors; transgenic or knockout models would be particularly useful in the investigation of methylation polymorphisms. Further evaluation of methylation polymorphisms in human populations is also warranted. Other in vivo models incorporating dietary manipulation could provide valuable insight into the role of nutrition in the carcinogenicity of arsenic. With more complete knowledge of the pharmacokinetics of arsenic metabolism and the mechanisms associated with its carcinogenic effects, development of more reliable risk assessment strategies are possible. Integration of data, both pharmacokinetic and mechanistic in nature, will lead to more accurate descriptions of the interactions that occur between the active chemical species and cellular constituents which lead to the development of cancer. This knowledge, in turn, will facilitate the development of more accurate and reliable risk assessment strategies for arsenic.