Arsenic Toxicity - Carcinogenicity

Carcinogenicity

It is still a matter of debate whether DNA repair inhibition or alterations in the status of DNA methylation are responsible for the carcinogenic potential of As. As vicinal sulfhydryl groups are frequently found in DNA-binding proteins, transcription factors and DNA-repair proteins, interaction of arsenic with these molecules appears to be likely. However, in vitro, most purified DNA repair enzymes are rather insensitive to arsenic, but in cell culture, As produces a dose-dependent decrease of DNA ligase activity. This might indicate that inhibition of DNA repair is an indirect effect due to changes in cellular redox levels or altered signal transduction and consequent gene expression. In spite of its carcinogenicity, the potential of arsenic to induce point mutations is weak. If administered with point mutagens it enhances the frequency of mutations in a synergistic way.

Its comoutagenic effects may be explained by interference with base and nucleotide excision repair, eventually through interaction with zinc finger structures. DMA showed to effectuate DNA single stand breaks resulting from inhibition of repair enzymes at levels of 5 to 100 mM in human epithelial type II cells.

+3 MMA and +3 DMA were also shown to be directly genotoxic by effectuating scissions in supercoiled ΦX174 DNA. Increased arsenic exposure is associated with an increased frequency of chromosomal aberrations, micronuclei and sister-chromatid exchanges. An explanation for chromosomal aberrations is the sensitivity of the protein tubulin and the mitotic spindle to arsenic. Histological observations confirm effects on cellular integrity, shape and locomotion.

+3 DMA is able to form reactive oxygen species (ROS) by reaction with molecular oxygen. Resulting metabolites are the dimethylarsenic radical and the dimethylarsenic peroxyl radical. Both +5 DMA and +3 DMA were shown to release iron from horse spleen as well as from human liver ferritin if ascorbic acid was administered simultaneously. Thus, formation of ROS can be promoted. Moreover, arsenic could cause oxidative stress by depleting the cell’s antioxidants, especially the ones containing thiol groups. The accumulation of ROS like the cited above and hydroxyl radicals, superoxide radicals and hydrogen peroxides causes aberrant gene expression at low concentrations and lesions of lipids, proteins and DNA in higher concentrations which eventually lead to cellular death. In a rat animal model, urine levels of 8-hydroxy-2’-desoxyguanosine (as a biomarker of ROS DNA damage) were measured after treatment with DMA. In comparison to control levels, they turned out to be significantly increased. This theory is further supported by a cross-sectional study which found elevated mean serum lipid peroxides (LPO) in the As exposed individuals which correlated with blood levels of inorganic arsenic and methylated metabolites and inversely correlated with nonprotein sulfhydryl (NPSH) levels in whole blood. Another study found an association of As levels in whole blood with the level of reactive oxidants in plasma and an inverse relationship with plasma antioxidants. A finding of the latter study indicates that methylation might in fact be a detoxification pathway with regard to oxidative stress: the results showed that the lower the As methylation capacity was, the lower the level of plasma antioxidant capacity. As reviewed by Kitchin (2001), the oxidative stress theory provides an explanation for the preferred tumor sites connected with arsenic exposure. Considering that a high partial pressure of oxygen is present in lungs and +3 DMA is excreted in gaseous state via the lungs this seems to be a plausible mechanism for special vulnerability. The fact that DMA is produced by methylation in the liver, excreted via the kidneys and latter on stored in the bladder accounts for the other tumor localizations.

Regarding DNA methylation, some studies suggest interaction of As with methyltransferases which leads to an inactivation of tumor suppressor genes through hypermethylation, others state that hypomethylation might occur due to a lack of SAM resulting in aberrant gene activation. An experiment by Zhong et al. (2001) with arsenite-exposed human lung A549, kidney UOK123, UOK109 and UOK121 cells isolated eight different DNA fragments by methylation-sensitive arbitrarily primed PCR. It turned out that six of the fragments were hyper- and two of them were hypomethylated. Higher levels of DNA methltransferase mRNA and enzyme activity were found.

Kitchin (2001) proposed a model of altered growth factors which lead to cell proliferation and thus to carcinogenesis. From observations, it is known that chronic low-dose arsenic poisoning can lead to increased tolerance to its acute toxicity. MRP1-overexpressing lung tumor GLC4/Sb30 cells poorly accumulate arsenite and arsenate. This is mediated through MRP-1 dependent efflux. The efflux requires GSH, but no As-GSH complex formation.

Although many mechanisms have been proposed, no definite model can be given for the mechanisms of chronic arsenic poisoning. The prevailing events of toxicity and carcinogenicity might be quite tissue-specific. Current consensus on the mode of carcinogenesis is that it acts primarily as a tumor promoter. Its co-carcinogenicity has been demonstrated in several models. However, the finding of several studies that chronically arsenic-exposed Andean populations (as most extremely exposed to UV-light) do not develop skin cancer with chronic arsenic exposure, is puzzling.