Insights into the carcinogenic mode of action of arsenic.

That arsenic can induce cancer in humans has been known since the late 17th century, yet how arsenic induces cancer has been the subject of numerous scientific publications. Various modes of action (MOA) have been proposed for arsenic's carcinogenicity. In this paper we review our previous studies on the ability of arsenicals to cause DNA damage, the relative inability of these arsenicals to induce point mutations, and the involvement of arsenicals in spindle disruption. We present new evidence that shows that reduced glutathione (GSH) can chemically reduce inactive pentavalent arsenicals to trivalent arsenicals which can disrupt tubulin polymerization, and show that reactive oxygen species (ROS) are most likely not involved in tubulin disruption. A hypothesis is also presented on how arsenic may induce stable chromosome aberrations (CAs) that can lead to cancer, thus supporting a role for genetic damage in the MOA for arsenic. We then propose promising areas of research that might give insight into the MOA of arsenic.

Oxidation and methylation status determine the effects of arsenic on the mitotic apparatus.

We investigated the spindle inhibitory properties of six arsenicals differing in their methylation or oxidation state. Human lymphoblasts were exposed for 6 h to either sodium arsenate (NaAs(V)), sodium arsenite (NaAs(III)), monomethylarsonic acid (MMA(V)), monomethylarsonous acid (MMA(III)), dimethylarsinic acid (DMA(V)), or dimethylarsinous acid (DMA(III)). After exposure slides were prepared, and the mitotic indices (MI) were assessed. We also exposed tubulin directly to each arsenical and spectrophotometrically measured its effect on polymerization. NaAs(V) caused a small but significant increase in MI. MMA(V) also caused only a slight increase in MI that just reached statistical significance. In contrast, DMA(V) caused a significant increase in MI, producing approximately 75% the MI of demecolcine and approximately 4 times the MI of the control. NaAs(III) had no significant effect on MI and was quite toxic. MMA(III) induced more than a twofold increase in MI compared to the control, which was about 40% that caused by demecolcine. On a micromolar basis, MMA(III) was the most potent of the arsenicals tested. DMA(III) gave inconsistent results. None of the pentavalent arsenicals had a substantial effect (either inhibition or enhancement) on GTP-induced polymerization of tubulin. In contrast, NaAs(III) inhibited polymerization at concentrations of 1 mM and above and MMA(III) and DMA(III) at 10 microM and above. Taken together, these results present a complex picture of how arsenicals may affect cells. These studies demonstrate that the metabolites of arsenic are active not only as chromosome breaking and DNA damaging agents but can also interfere with cell division via tubulin disruption.

Genotoxicity studies of three triazine herbicides: in vivo studies using the alkaline single cell gel (SCG) assay.

Triazine herbicides are prevalent contaminants of groundwater in the agricultural regions of the United States. The literature on the genotoxicity of triazines is rife with conflicting data, though the general tendency is for most studies to report negative results. In order to investigate further the genotoxicity of triazines, we exposed mice to triazines by intraperitoneal injection up to the maximum tolerated doses. About 24h later, blood was removed, and the leukocytes subjected to DNA damage analysis using the alkaline single cell gel electrophoresis assay (SCG), one of the most sensitive DNA damage assays available. Our results indicate that atrazine induced a small dose-related increase in DNA damage. Simazine did not induce any dose-related increase in DNA damage. Cyanazine induced a marginal increase in DNA damage with dose, but no individual dose was significantly increased compared to the control. These results indicate that these triazines, even at extremely high concentrations, have only marginal DNA-damaging activity in vivo in mouse leukocytes.

Cytogenetic studies of three triazine herbicides. II. In vivo micronucleus studies in mouse bone marrow.

Atrazine, simazine, and cyanazine are widely used preemergence and postemergence triazine herbicides that have made their way into the potable water supply of many agricultural communities. Although there are several contradictory genotoxicity studies in the literature, our previous in vitro studies with human lymphocytes showed that atrazine, simazine, and cyanazine did not induce sister chromatid exchanges (SCEs) or chromosome aberrations (CAs) up to the limits of solubility in aqueous medium using 0.5% dimethyl sulfoxide. To expand upon these results and to ensure that our in vitro findings could be replicated in an in vivo system, mice were treated with each triazine by two intraperitoneal injections, 24h apart. The animals were sacrificed and the bone marrow removed for micronucleus (MN) analysis, 24h after the last injection. Two to four independent trials were performed for MN analysis in polychromatic erythrocytes, and in some trials the spleen was removed, cultured, and analyzed for SCEs and CAs. None of the triazines investigated induced MN in the bone marrow, even at doses that caused significant bone marrow suppression and/or death. These results indicate that atrazine, simazine, and cyanazine are not genotoxic as measured by the bone marrow MN assay in mice following high dose exposures.

Cytogenetic studies of three triazine herbicides. I. In vitro studies.

Atrazine, simazine, and cyanazine are widely used pre-emergence and post-emergence triazine herbicides that have made their way into the potable water supply of many agricultural communities. Because of this and the prevalence of contradictory cytogenetic studies in the literature on atrazine, simazine, and cyanazine, a series of in vitro experiments was performed to investigate the ability of these three triazines to induce sister chromatid exchanges (SCEs) and chromosome aberrations (CAs) in human lymphocyte cultures. Our results showed that all three triazines failed to produce any significant increases in SCEs or CAs up to the limits of solubility [using 0.5% dimethyl sulfoxide (DMSO)]. Our results are discussed in light of contradictory results in the literature.

Cell cycle specificity of cytogenetic damage induced by 3,4-epoxy-1- butene.

3,4-epoxy-1-butene (EB), a primary metabolite of butadiene, is a direct-acting "S-dependent" genotoxicant that can induce sister chromatid exchanges (SCEs) and chromosome aberrations (CAs) in cycling cells in vitro. However, EB is almost inactive when splenic or peripheral blood lymphocytes are exposed at the G(0) stage of the cell cycle. To investigate whether repair of DNA lesions is responsible for the lack of cytogenetic responses seen after G(0) treatments, we used cytosine arabinoside (ara-C) to inhibit DNA polymerization during DNA repair. If enough repairable lesions are present, double-strand breaks should accumulate and form chromosome-type ("S-independent") deletions and exchanges. This is exactly what occurred. EB induced chromosome deletions and dicentrics at the first division following treatment, when the EB exposure was followed by ara-C. Without ara-C treatment, there was no induction of CAs. These experiments indicate that the relatively low levels of damage induced by EB in G(0) lymphocytes are removed by DNA repair prior to DNA synthesis and thus, before the production of SCEs or chromatid-type aberrations.

Comparison of cytogenetic effects of 3,4-epoxy-1-butene and 1,2:3, 4-diepoxybutane in mouse, rat and human lymphocytes following in vitro G0 exposures.

To understand better the species differences in carcinogenicity caused by 1,3-butadiene (BD), we exposed G0 lymphocytes (either splenic or peripheral blood) from rats, mice and humans to 3, 4-epoxy-1-butene (EB) (20 to 931 microM) or 1,2:3,4-diepoxybutane (DEB) (2.5 to 320 uM), two of the suspected active metabolites of BD. Short EB exposures induced little measurable cytogenetic damage in either rat, mouse, or human G0 lymphocytes as measured by either sister chromatid exchange (SCE) or chromosome aberration (CA) analyses. However, DEB was a potent inducer of both SCEs and CAs in G0 splenic and peripheral blood lymphocytes. A comparison of the responses among species showed that the rat and mouse were approximately equisensitive to the cytogenetic damaging effects of DEB, but the situation for the human subjects was more complex. The presence of the GSTT1-1 gene (expressed in the erythrocytes) reduced the relative sensitivity of the lymphocytes to the SCE-inducing effects of DEB. However, additional factors also appear to influence the genotoxic response of humans to DEB. This study is the first direct comparison of the genotoxicity of EB and DEB in the cells from all three species.

Cytogenetic effects of butadiene metabolites in rat and mouse splenocytes following in vitro exposures.

As a first step in investigating the genotoxic effects of the principal metabolites of 1,3-butadiene (BD) in both rats and mice, splenocytes (which have little mixed function oxidase activity) from each specimen were exposed to a series of concentrations of either 3,4-epoxy-1-butene (EB) (20 to 931 microM) or 1,2:3,4-diepoxybutane (DEB) (2.5 to 160 microM) for 1 h. The splenocytes were then washed, cultured, and stimulated to divide with concanavalin A, and metaphases were analyzed for the induction of sister chromatid exchanges (SCEs) and chromosome aberrations (CAs). In addition, cells from some experiments were taken after exposure but before culture, and subjected to the single cell gel (SCG) assay to measure DNA damage in the form of DNA strand breakage and/or alkaline-labile sites. Initial studies indicate that EB does not induce cytogenetic damage in either rat or mouse G0 splenocytes. However, DEB was an extremely potent SCE- and CA-inducer in both species with no species differences apparent. Neither DEB nor EB produced any statistically significant DNA-damaging effects as measured by the SCG assay.

Cytogenetic effects in mice of divinylbenzene-55 inhalation.

Male B6C3F1 mice (8 weeks of age) were exposed by inhalation to divinylbenzene-55 (DVB-55), at target concentrations of 0, 25, 50 and 75 ppm for 6 h per day for 3 days. Following exposure the animals were killed blood smears were prepared for micronucleus (MN) analysis, and the spleens were removed and cultured for sister chromatid exchange (SCE) and chromosome aberration (CA) analyses. DVB-55 induced a dose dependent increase in SCE with the two highest doses reaching statistical significance. Similarly, there was a statistically significant although less pronounced increase in the frequency of CAs in splenocytes and MN in polychromatic erythrocytes. There was no indication of toxicity as measured by cell cycle kinetics in the splenocytes or the percentage of polychromatic erythrocytes in the peripheral blood smears. Thus, DVB-55 appears to be a weak genotoxicant in vivo.