Methylarsonous acid causes oxidative DNA damage in cells independent of the ability to biomethylate inorganic arsenic

Abstract

Inorganic arsenic (iAs) and its toxic methylated metabolite, methylarsonous acid (MMA(III)), both have carcinogenic potential. Prior study shows iAs-induced malignant transformation in both arsenic methylation-proficient (liver) and methylation-deficient (prostate) cells, but only methylation-proficient cells show oxidative DNA damage (ODD) during this transformation. To further define whether arsenic methylation is necessary for transformation or ODD induction, here we chronically exposed these same liver or prostate cell lines to MMA(III) (0.25-1.0 μM) and tested for acquired malignant phenotype. Various metrics of oncogenic transformation were periodically assessed along with ODD during chronic MMA(III) exposure. Methylation-deficient and methylation-proficient cells both acquired a cancer phenotype with MMA(III) exposure at about 20 weeks, based on increased matrix metalloproteinase secretion, colony formation, and invasion. In contrast, prior work showed iAs-induced transformation took longer in biomethylation-deficient cells (~30 weeks) than in biomethylation-proficient cells (~18 weeks). In the present study, MMA(III) caused similar peak ODD levels at similar concentrations and at similar exposure times (18-22 weeks) in both cell types. At the approximate peak of ODD production, both cell types showed similar alterations in arsenic and oxidative stress adaptation factors (i.e., ABCC1, ABCC2, GST-π, SOD-1). Thus, MMA(III) causes oncogenic transformation associated with ODD in methylation-deficient cells, indicating that further methylation is not required to induce ODD. Together, these results show that MMA(III) and iAs cause an acquired malignant phenotype in methylation-deficient cells, yet iAs does not induce ODD. This indicates iAs likely has both genotoxic and non-genotoxic mechanisms dictated by the target cell's ability to methylate arsenic.

Figure 1

Activity of secreted matrix metalloproteinase (MMP)-2 and -9 in (A) liver and (B) prostate cells chronically exposed to MMA^III. Both cell lines were exposed to 1 μM MMA^III for 30 weeks, and MMP-2 and -9 in conditioned media were measured by SDS-PAGE zymography. Time-dependent increases in MMP activity were seen in both cell lines starting at 15 weeks of MMA^III exposure. Data represent mean ± SEM (n = 3). Asterisk (*) indicates statistical significance compared with time-matched untreated controls, p < 0.05.

Figure 2

Colony formation in soft agar and invasion during chronic exposure to MMA^III in liver and prostate cells. (A) Liver cells started showing a time-dependent increase at 5 weeks and (B) prostate cells started showing a time-dependent increase in colony formation at 15 weeks of MMA^III exposure. (C, D) Both cell lines showed very similar time-dependent increases in invasion starting at 15 weeks of MMA^III exposure. Data are expressed as mean ± SEM (n = 3). Asterisk (*) indicates statistical significance compared with time-matched untreated controls, p < 0.05.

Figure 3

PTEN levels during chronic MMA^III exposure of liver and porstate cells. (A, B) Transcript levels for PTEN show a similar depletion starting at 10 weeks in both the (A) liver and (B) prostate cell lines. Data are expressed as mean ± SEM (n = at least 3). Asterisk (*) indicates statistical significance compared with time-matched untreated controls, p < 0.05.

Figure 4

Oxidative DNA damage (ODD) during chronic MMA^III exposure. Methylation-proficient liver cells (A) and methylation-deficient prostate cells (B) were exposed to various concentrations of MMA^III for 30 weeks, and ODD was assessed by the formation of DNA–nitrone adducts using the immuno-spin trapping method. Strikingly similar temporal patterns of ODD were seen between the methylation-proficient and methylation-deficient cells. Data are expressed as mean ± SEM (n = 9).

Figure 5

Direct comparison of ODD peaks (at around 20 weeks of exposure; from Figure 4) in cells exposed to different concentrations of MMA^III. Peak ODD levels at the highest MMA^III concentration (1 μM) were similar between the two cell lines but at the lower concentrations (0.25 μM and 0.5 μM) tended to be somewhat lower in the methylation-proficient liver cells compared with the methylation-deficient prostate cells. Data were taken from Figure 4.

Figure 6

Comparison of MMP activity and ODD levels induced by exposure to iAs and MMA^III. MMP activity and ODD levels in the methylation-deficient prostate cells (A, B) and methylation-proficient liver cells (C, D) exposed to iAs and MMA^III at the approximate point of transformation for each arsenical. Asterisk (*) indicates statistical significance compared with time-matched, arsenical-matched untreated controls, p < 0.05. Data for iAs taken from Kojima et al. 2009.

Figure 7

Levels of arsenical efflux–associated factors. Transcript levels of GST-π, ABCC1, and ABCC2 were measured in prostate (A) and liver (B) cells during chronic MMA^III (1 μM) exposure for 30 weeks. Data are expressed as mean ± SEM (n = 3). Asterisk (*) indicates statistical significance compared with time-matched untreated controls, p < 0.05.

Figure 8

Levels of oxidative stress-associated factors. Transcript levels of NRF2, SOD-1, SOD-2, and HO-1 were measured in prostate (A) and liver (B) cells during chronic MMA^III (1 μM) exposure for 30 weeks. Data are expressed as mean ± SEM (n = 3). Asterisk (*) indicates statistical significance compared with time-matched untreated controls, p < 0.05.