Biokinetics and subchronic toxic effects of oral arsenite, arsenate, monomethylarsonic acid, and dimethylarsinic acid in v-Ha-ras transgenic (Tg.AC) mice.

Previous research demonstrated that 12-O-tetradecanoylphorbol-13-acetate (TPA) treatment increased the number of skin papillomas in v-Ha-ras transgenic (Tg.AC) mice that had received sodium arsenite [(As(III)] in drinking water, indicating that this model is useful for studying the toxic effects of arsenic in vivo. Because the liver is a known target of arsenic, we examined the pathophysiologic and molecular effects of inorganic and organic arsenical exposure on Tg.AC mouse liver in this study. Tg.AC mice were provided drinking water containing As(III), sodium arsenate [As(V)j, monornethylarsonic acid [(MMA(V)], and 1,000 ppm dimethylarsinic acid [DMA(V)] at dosages of 150, 200, 1,500, or 1,000 ppm as arsenic, respectively, for 17 weeks. Control mice received unaltered water. Four weeks after initiation of arsenic treatment, TPA at a dose of 1.25 [micro]g/200 [micro]L acetone was applied twice a week for 2 weeks to the shaved dorsal skin of all mice, including the controls not receiving arsenic. In some cases arsenic exposure reduced body weight gain and caused mortality (including moribundity). Arsenical exposure resulted in a dose-dependent accumulation of arsenic in the liver that was unexpectedly independent of chemical species and produced hepatic global DNA hypomethylation, cDNA microarray and reverse transcriptase-polymerase chain reaction analysis revealed that all arsenicals altered the expression of numerous genes associated with toxicity and cancer. However, organic arsenicals [MMA(V) and DMA(V)] induced a pattern of gene expression dissimilar to that of inorganic arsenicals. In summary, subchronic exposure of Tg.AC mice to inorganic or organic arsenicals resulted in toxic manifestations, hepatic arsenic accumulation, global DNA hypomethylation, and numerous gene expression changes. These effects may play a role in arsenic-induced hepatotoxicity and carcinogenesis and may be of particular toxicologic relevance. Key words: arsenicals (arsenic forms), gene expression, mouse liver, subchronic toxicity, toxicokinetics.

**********

Arsenic is an important environmental toxicant and carcinogen [International Agency for Research on Cancer (IARC) 1987; National Research Council (NRC) 1999]. Chronic exposure to arsenic via drinking water is a major health concern throughout the world (Gebel 2000; NRC 1999). The carcinogenic effects of environmental arsenic exposure in human populations are well documented (IARC 1987; NRC 1999), and exposure can lead to tumors in and toxicity of the skin, lung, urinary bladder, liver, and other sites.

The adverse effects of arsenic are dependent, in part, on its chemical form and metabolism (Aposhian 1997; Vahter 2002). Humans are exposed primarily to trivalent [arsenite, As(III)] and pentavalent ]arsenate, As(V)] inorganic arsenicals present in the environment, as well as to organic arsenic [e.g., dimethylarsinic acid, DMA(V)] in some situations (Kenyon and Hughes 2001; Shen et al. 2003b). In manamals, As(V) is first reduced to As(III), whereas As(III), produced by this reduction or from direct ingestion, is methylated primarily to pentavalent organic arsenicals including monomethylarsonic acid [MMA(V)] and DMA(V)]. MMA and DMA are the predominant metabolites of inorganic arsenic (Vahter 2002), although DMA may be further methylated to trimethylarsine oxide (TMAO) (Hughes 2002; Yoshida et al. 1998). The forms of arsenic to which humans are exposed, either directly or via metabolism, further complicate the elucidation of their toxic and carcinogenic mechanisms of action. Previously, inorganic arsenicals were thought to be more acutely toxic than organic species, as the methylation of inorganic arsenic was proposed to be a detoxification process. However, recent studies indicate that trivalent organic arsenicals [e.g., MMA(III) and DMA(III)] that are metabolic products of inorganic arsenic can be more toxic than the parent compound (Petrick et al. 2001; Styblo et al. 2000). Furthermore, DMA can act as a tumor promoter at various sites and as a complete carcinogen for the urinary bladder in rats (Salim et al. 2003; Wei et al. 2002; Yamamoto et al. 1997). MMA produces preneoplastic changes in liver and urinary bladder but does not produce overt neoplasia (Shen et al. 2003a), whereas TMAO can induce hepatocellular adenomas (Shen et al. 2003b). Therefore, it is important to compare and evaluate the toxicity of As(III), As(V), MMA(V), and DMA(V) under similar experimental conditions.

Recent studies demonstrated that arsenic acts as a co-promoter with 12-O-tetradecanoylphorbol-13-acetate (TPA) because together they enhance skin tumor development in transgenic (Tg.AC) mice, which overexpress the v-Ha-ras oncogene (Germolec et al. 1997, 1998; Trouba et al. 2003). Because hepatic metabolism in Tg.AC mice is not compromised by over-expression of the v-Ha-ras oncogene (Sanders et al. 2001), we hypothesized that organic and inorganic arsenicals produce similar yet distinct changes in Tg.AC liver gene expression that may be predictive of hepatotoxicity. The latter is important because the liver is an important target organ of arsenic toxicity in animals (Waalkes et al. 2000b) and humans (Lu et al. 2001). The liver is also a major target organ of arsenic carcinogenicity after in utero exposure in mice (Waalkes et al. 2003, 2004b) and in humans exposed to environmental arsenic (Centeno et al. 2002; Zhou et al. 2002). To address the above hypothesis, we examined the effects of subchronic inorganic and organic arsenical exposure on the Tg.AC mouse liver. Our results indicate that in Tg.AC mice, a) hepatic arsenic [e.g., As(III), As(V), MMA(V), and DMA(V))] accumulation, based on biokinetic analyses, was dose dependent; b) global DNA hypomethylation occurred after exposure to As(III) As(V), MMA(V), and DMA(V); c) pathological changes were present in the liver after exposure to As(III), MMA(V), and DMA(V); and d) arsenic-induced gene expression changes, determined using cDNA microarray and real-time reverse transcriptase-polymerase chain reaction (RT-PCR) analysis, occurred in the liver of animals treated with As(III), As(V), MMA(V), and DMA(V).

Materials and Methods

Chemicals

As(III) and As(V) were purchased from Aldrich Chemical Co. (Milwaukee, WI) and Fluka Chemical Corp. (Milwaukee, WI), respectively. MMA(V) was obtained from AccuStandard, Inc. (New Haven, CT). DMA(V) and TPA were purchased from Sigma Chemical Co. (St. Louis, MO). Customer-designed cDNA microarrays (600 genes) were purchased from BD Biosciences Clontech, Inc. (Palo Alto, CA). [[alpha]-[sup.32]P]-deoxyadenosine 5'-triphosphate was purchased from PerkinElmer, Inc. (Boston, MA), and [sup.3]H-labeled S-adenosyhnethionine ([[sup.3]H]-SAM) was from Amersham (Arlington Heights, IL).

Animal Treatment

All animals were handled and treated in compliance with the Guide for the Care and Use of Laboratory Animals (NRC 1996). Female, homozygous Tg.AC mice containing the fetal zeta-globin promoter fused to the v-Ha-ras structural gene (with mutations at codons 12 and 59) and linked to a simian virus 40 polyadenylylation/ splice sequence were obtained from Taconic Farms (Germantown, NY) (Leder et al. 1990). Mice were maintained in an animal facility at a temperature of 20-22[degrees]C, a relative humidity of 50%, and a 12-hr light/dark cycle. Mice were randomly assigned to five groups (n = 15 in each group) and were provided unaltered drinking water (control) and drinking water containing As(III) (150 ppm as arsenic), As(V) (200 ppm as arsenic), MMA(V) (1,500 ppm as arsenic), and DMA(V) (1,000 ppm as arsenic), respectively, for 17 weeks. The doses of arsenicals used were based on our previous studies (Germolec et al. 1997, 1998). Multiple doses of each arsenical were originally used to examine papilloma development. However, to detect gene expression changes in the liver that may be related to arsenic hepatotoxicity and hepatocellular carcinogenesis, animals treated with the maximal dose of each arsenical were selected for analysis.

Four weeks after initiation of arsenic treatment, TPA at a dose of 1.25 [micro]g/200 [micro]L acetone was applied twice weekly for 2 weeks to the shaved dorsal skin of all mice, including the mice not receiving arsenic (control). At 17 weeks the mice were sacrificed by C[O.sub.2] asphyxiation and necropsied. Liver tissue was excised and stored at -70[degrees]C until analysis or fixed for histology, as described below.

During the exposure to arsenic, mortality, moribundity, clinical symptoms, body weight, and water intake of the mice were monitored. All mice, including those found deceased or sacrificed as moribund, underwent complete necropsy.

Pathological Examination

Liver samples were fixed with neutral-buffered formalin, processed by standard procedures, embedded in paraffin, sectioned, and stained with hematoxylin and eosin for light microscopy examination. All pathological assessments were performed in a blind fashion.

Hepatic Arsenic Levels

A portion of the frozen liver (120-150 mg) was digested in nitric acid. Total arsenic, which would include inorganic and organic forms, was determined using graphic furnace atomic absorption spectrometry (Perkin-Elmer AAnalystl00; PerkinElmer, Inc., Norwalk, CT). Results were expressed as micrograms arsenic per gram wet weight liver, as reported in our recent publications (Liu et al. 2001a; Xie et al. 2004).

Global DNA Methylation Assay

Genomic DNA was extracted from liver tissue and purified using DNeasy Kits (Qiagen, Valencia, CA). Global DNA methylation status was assessed by methyl acceptance assay (Chen et al. 2004). Briefly, DNA (1 [micro]g) was incubated at 37[degrees]C for 2 hr in a 30-[micro]L mixture containing 1.25 [micro]M (3 [micro]Ci) [[sup.3]H]-SAM, 4 units CpG methylase (M. Sss I) (New England Biolabs, Inc., Beverly, MA), 10 mM DDT, Tris-EDTA buffer (100 mM Tris, 10 mM EDTA, pH 8.0), and 100 mM NaCl. The reaction was terminated on ice and transferred onto a Whatman DE81 filter (Whatman International Ltd., Maidstone, U.K.). The filter was washed on a vacuum filtration apparatus with 2 mL 0.5 M phosphate buffer (pH 7.0) 5 times, followed by a wash with 2 mL 70% ethanol and 2 mL absolute ethanol. After the filter was dried, the bound radioactivity was measured by scintillation (Beckman LS 6500 Scintillation Counter; Beckman Coulter, Inc., Fullerton, CA).

cDNA Microarray Analysis

Microarray analysis was performed as previously described (Xie et al. 2004). Briefly, total RNA was extracted from liver tissues with Trizol reagent and purified with RNeasy columns (Qiagen). Five micrograms pooled RNA (n = 5) was converted to [[alpha]-[sup.32]P]]-dATP-labeled cDNA probe with Atlas specific cDNA synthesis primers (BD Biosciences Clontech Inc.). The probe was purified with a NucleoSpin column (BD Biosciences Clontech), denatured at 100[degrees]C for 2-3 min, and hybridized to the membrane in triplicate with Expresshyb buffer (BD Biosciences Clontech) at 68[degrees]C overnight. The membranes were washed at 68[degrees]C four times (30 min each) in 2 x sodium chloride/sodium citrate (SSC)/1% sodium dodecyl sulfate (SDS), twice in 0.1 x SSC/0.5% SDS, and exposed to a phosphoimage screen. Images were acquired by Phosphorlmager Scanner (Model Storm 860; Molecular Dynamics, Sunnyvale, CA) and analyzed densito-metrically using AtlasImage software (version 2.01; Clontech).

Real-time RT-PCR Analysis

Total RNA was reverse transcribed with MMLV reverse transcriptase and oligodT primers (PerkinElmer Inc.). The PCR primers were designed with Primer Express software and the SYBR Green DNA PCR kit (Applied Biosystems, Foster City, CA) was used for real-time RT-PCR analysis. Differences in gene expression between groups were calculated using cycle time (Ct) values, which were normalized against [beta]-actin and expressed as relative increases/ decreases, setting control as 1.0. Assuming that the Ct value is reflective of the initial template amount (copy number) and that there is 100% efficiency, a difference of one cycle is equivalent to a 2-fold difference in initial copy number (Walker 2001).

Statistics

Data are expressed as mean [+ or -] SEM or as incidence (for mortality). For comparisons of gene expression between two groups, the Student t test was used. For comparisons among three or more groups, data were analyzed using a one-way analysis of variance, followed by Duncan's multiple range test. The p-value was calculated by Fisher's exact test for incidence data. The level of significance was set at p < 0.05 in all cases. Two-dimensional hierarchical cluster analysis of microarray data (hybrid intensity ratios to control values) was performed. The results from clustered analysis were examined by interactive graphical analysis using TreeView software (http://rana. lbl.gov/EisenSo ftware.htm).

Results

Clinical Symptoms

During the 17 weeks of arsenical exposure, several arsenic-treated mice were found deceased or were euthanized because of moribundity. Exposure to As(III) (150 ppm) and DMA(V) (1,000 ppm) resulted in 20% mortality, and exposure to MMA(V) (1,500 ppm) resulted in 40% mortality (Table 1). MMA(V) exposure alone produced significant toxicity when compared with control. In general, the body weight in arsenic-treated groups was lower than that in control groups. At the end of arsenic exposure (17 weeks), body weight was decreased by approximately 15, 8, 10, and 8% in mice treated with As(III) (150 ppm), As(V) (200 ppm), MMA(V) (1,500 ppm), or DMA(V) (1,000 ppm), respectively (Figure 1). Our findings suggest that exposure of Tg.AC mice to these arsenicals produced mild to moderate [for the MMA(V) group] toxicity.

[FIGURE 1 OMITTED]

Pathology

The treatment of As(III) plus TPA did not induce liver tumor formation in Tg.AC mice treated with arsenicals for 17 weeks (unpublished data). However, morphologic changes including inflammation, foci of apoptosis and necrosis, and hepatocellular degeneration were observed in arsenic-treated mice (Figure 2). Foci of apoptosis and necrosis were observed in animals treated with As(III) (150 ppm); however, no apparent histologic alterations were present in animals that received As(V) (200 ppm). MMA(V) (1,500 ppm) produced inflammatory cell infiltration, degeneration, and swelling; DMA(V) (1,000 ppm) produced foci of inflammation and hepatocellular degeneration. These findings indicate that subchronic arsenical exposure produces pathological alterations in the liver.

[FIGURE 2 OMITTED]

Hepatic Arsenic Content

Although not detectable in livers of controls, arsenic was found in the livers of all treatment groups (Figure 3). Particularly high levels of arsenic were present in the livers of the MMA-treated group (Figure 3A). When hepatic arsenic content was plotted against arsenical dose, a strong linear correlation was observed (r = 0.98) (Figure 3B), suggesting that subchronic arsenic exposure results in arsenic accumulation in the liver that is dose dependent.

[FIGURE 3 OMITTED]

Global DNA Methylation Status

Global DNA methylation was assessed by methyl acceptance assay (Figure 4). This assay uses a bacterial DNA methyltransferase that indiscriminately methylates all unmethyiated cytosines using [[sup.3]H]-SAM. Thus, higher [[sup.3]H]-SAM incorporation corresponds to a lower degree of methylation (i.e., hypomethylation) of cellular DNA. The amount of unmethylated DNA from all the arsenical-treated groups was significantly higher (p < 0.05) than control, indicating that DNA hypomethylation occurs in the Tg.AC mouse liver after subchronic exposure to arsenic, regardless of the chemical form. When this is correlated with actual arsenic dose, As(III) is the most potent hypomethylating agent; MMA(V) is the least.

[FIGURE 4 OMITTED]

Genomic Analysis by cDNA Microarray

Among the 600 genes examined via microarray analysis, 70 displayed increased or decreased expression after subchronic arsenic exposure. The hybrid intensity (ratio to control value) for these 70 genes was calculated for comparison then subjected to cluster analysis to compare alterations in gene expression patterns related to the type of arsenical exposure. TreeView revealed both similar and dissimilar changes in gene expression patterns among the four arsenicals (Figure 5).

[FIGURE 5 OMITTED]

The most significant arsenic-induced changes in gene expression are listed in Table 2. Genes associated with glutathione S-transferase (GST) function/metabolism, stress, apoptosis, cell proliferation, and early neoplasia are thought to be related to arsenic toxicity (Liu et al. 2004; Trouba et al. 2002, 2003; Xie et al. 2004) and thus are included for comparison. For example, all arsenicals produced increases in GSTs (alpha, mu, pi, and theta) and fibroblast growth factor 2, a gene related to cell proliferation. A significant increase in the expression of insulin-like growth factor binding protein 1 (IGFBP-1) also was found in MMA-treated mice. In general, all of the arsenicals produced similar effects (i.e., increase/decrease) on gene expression; however, the degree of change was different in some cases.

Real-Time RT-PCR Analysis

Real-time RT-PCR analysis was performed for selected genes in each cluster. Figure 6 shows data for some of the genes of interest. GST-[pi], early growth response protein 1 (FGR-1), heme oxygenase 1 (HO-1), c-myc, and [alpha]-fetoprotein gene expression was enhanced after arsenical exposure. Generally, real-time RT-PCR analysis confirmed our microarray results.

[FIGURE 6 OMITTED]

Discussion

This study demonstrated that subcbronic exposure of transgenic (Tg.AC) mice to both inorganic and organic arsenicals through drinking water produced various effects on the liver, a major target organ of arsenic toxicity and carcinogenesis (Centeno et al. 2002; NRC 1999; Waalkes et al. 2003, 2004b). Arsenic-induced toxicity was evidenced by an increase in moribundity and death, a depression in body weight, hepatic pathological changes, and significant changes in gene expression.

An original goal of our research was to examine the effects of inorganic and organic arsenic on TPA-promoted skin papilloma development in Tg.AC mice. Although TPA was administered to all mice (including controls that received no arsenic), the effects of this skin tumor promoter were not deemed critical to our analyses of liver pathology, DNA methylation, and gene expression. Interestingly, topical application of TPA in some experimental models has systemic effects; we recently found that it promoted liver tumors initiated by transplacental arsenic exposure in female mice (Waalkes et al. 2004b). In this study epidermal TPA treatment resulted in no mortality and did not affect hepatic pathology, indicating that the biological end points/ changes measured are most likely dependent on arsenical treatment alone.

Because the liver is a major target organ of arsenic toxicity and carcinogenesis (Waalkes et al. 2000b, 2003, 2004b), we examined gene expression as well as pathological changes in the livers of Tg.AC mice to further explore the usefulness of this system as an in vivo model of arsenic carcinogenesis and toxicity. To detect gene expression changes that may be related to arsenic toxicity, animals treated with the maximal dose of each arsenical were selected for analysis. Generally, 150 ppm As(III) produced more toxicity and more dramatic changes in gene expression than 200 ppm As(V). Organic arsenicals at doses [1,500 and 1,000 ppm as arsenic for MMA(V) and DMA(V), respectively] 5-to 10-fold higher produced toxic effects comparable to those produced by As(III). Although rats are tolerant to 200 ppm MMA(V) in drinking water for 104 weeks (Shen et al. 2003a), the mice in our study did not tolerate MMA(V) at 1,500 ppm, as 40% mortality (i.e., moribundity and death) occurred in these mice over the 17-week exposure period. The dose of DMA(V) in this study was also higher than the doses (50 and 200 ppm) used to induce urinary bladder tumors in rats (Wei et al. 2002) and also exceeded the maximum tolerated dose, as it produced 20% mortality.

In our study, promoted and nonpromoted, arsenic-treated Tg.AC mice did not display direct evidence of liver tumor formation. However, preneoplastic lesions (e.g., cell proliferation) occur in the liver after chronic oral arsenic exposures in several strains of mice (Chen et al. 2004; Shen et al. 2003a; Waalkes et al. 2000b) and were also observed in the liver of Tg.AC mice exposed to arsenic in this study. Exposure to arsenic in the drinking water resulted in a dose-dependent accumulation of arsenic in the liver that was independent of chemical form. The highest hepatic content, which was observed in the high-dose (1,500 ppm) MMA(V) group, might contribute to the high degree of mortality (40%) in this group. The hepatic arsenic contents in the Tg.AC mice receiving 150 ppm As(iii) and 200 ppm As(V) in this study were 1.2 and 2.0 [micro]g/g tissue, respectively. This was less than the arsenic content in the skin (8.3 [micro]g/g tissue) and much less than that in the hair (170.2 [micro]g/g tissue) of Tg.AC mice exposed to 200 ppm As(III) in the drinking water for 14 weeks in our previous study (Germolec et al. 1998), indicating that arsenic accumulation in the liver is lower than that in the hair or skin. This may be because liver is the major target organ for arsenic metabolism, and arsenic elimination generally occurs through the bile (Gregus et al. 2000) or urine.

DNA hypomethylation occurs after chronic arsenic exposure in cells (Zhao et al. 1997) and also in intact animals (Chen et al. 2004; Okoji et al. 2002). In the present study, all arsenicals produced significant DNA hypomethylation in the liver, regardless of dose. Although the doses of MMA(V) (1,500 ppm) and DMA(V) (1,000 ppm) used in our study were much higher than those of As(III) (150 ppm) and As(V) (200 ppm), MMA(V) and DMA(V) induced less hypomethylation of hepatic DNA than As(III) and As(V). This suggests that inorganic arsenicals are more potent stimulators of DNA hypomethylation compared with MMA(V) and DMA(V). It should be noted that global DNA hypomethylation could co-exist with regional or individual gene hypermethylation, as arsenic-induced p53 hypermethylation has been reported (Mass and Wang 1997). In our recent study, we proposed that arsenic-induced hypomethylation of the estrogen receptor-[alpha] gene plays an important role in hepatocellular proliferation (Chen et al. 2004; Waalkes et al. 2004a). Efforts are currently being undertaken to examine the methylation status of individual genes after arsenic exposure.

DNA hypomethylation is an important mechanism involved in aberrant gene expression and carcinogenesis (Baylin et al. 1998; Goodman and Watson 2002). In particular, it is thought that aberrant DNA methylation is central to the development of liver cancers (Goodman and Watson 2002) and is an epigenetic mechanism that underlines the aberrant expression of genes involved in mouse liver carcinogenesis (Counts et al. 1997). In the present study, As(III), As(V), MMA(V), and DMA(V) produced variable gene expression changes, accounting for approximately 10% of genes on the array. We focused primarily on a few categories, for example, glutathione (GSH)-, apoptosis-, and cell proliferation-related genes, and genes important for tumor development, as previous studies have shown these to be related to aberrant cell growth and neoplasia.

Glutathione systems play important roles in arsenic toxicity and carcinogenesis (NRC 1999; Trouba et al. 2002, 2003; Xie et al. 2004). In the present study, the expression of GST-[micro], GST-[pi], GST-[alpha], and GST-[tau] was increased by all arsenicals, although to a variable extent. GSTs are a group of enzymes catalyzing the conjugation and oxidation of GSH with arsenic (Xie et al. 2004). An increase in GST expression/activity (particularly GST-[pi]) has been reported to play an important role in cellular efflux of arsenic--GSH conjugates and to be a mechanism of arsenic tolerance (Brambila et al. 2002; Liu et al. 2001a; Wang et al. 1996). Increases in GST-[pi] positive foci have been proposed to be a hepatic preneoplastic biomarker in chronic arsenic-exposed populations (Nishikawa et al. 2002; Shen et al. 2003a). Changes in GST activity in humans also are associated with altered arsenic metabolism (Chiou et al. 1997; Marnell et al. 2003), and GST polymorphisms are thought to be a susceptibility factor for arsenic toxicity in humans (Marnell et al. 2003). Together, these data indicate that increases in GST gene expression and/or function are consistent events associated with arsenic carcinogenicity and toxicity.

Oxidative stress is proposed to play an important role in arsenic toxicity and carcinogenesis (Kitchin and Abroad 2003; Liu et al. 2001b; NRC 1999; Trouba et al. 2002; Xie et al. 2004). In addition to GSTs, other biomarkers for arsenic-induced oxidative stress such as HO-1 (Del Razo et al. 2001; Liu et al. 2001b), EGR-1 (Liu et al. 2004; Simeonova et al. 2000), DT- diaphorase (Pi et al. 2003), and cytochrome P450 3A25 (Liu et al. 2001b) were all increased in Tg.AC mice after exposure to arsenicals. Evidence is accumulating regarding the ability of arsenicals to produce reactive oxygen species and free radicals as measured using electron spin response. This includes inorganic As(III) and As(V) (Barchowsky et al. 1999) and organic MMA(III), DMA(III) (Nesnow et al. 2002), and DMA(V) (Yamanaka et al. 1990). Our data (e.g., gene expression changes and pathology) lend further evidence for the presence of oxidative stress during subchronic exposure to arsenicals.

Arsenic induces apoptosis involved in its mechanism of acute toxicity (NRC 1999). However, after chronic arsenic exposure and the induction of malignant transformation, the development of apoptosis resistance occurs (Brambila et al. 2002; Qu et al. 2002) and is associated with the downregulation of apoptosis-related genes (Brambila et al. 2002; Chen et al. 2001a). In the present study, arsenical exposure resulted in downregulation of apoptosis-associated genes such as FasL, tumor necrosis factor receptor-associated factor 3, Bad, and granzyme A, and also increased the expression of cell proliferation--related genes including c-myc, proliferating cell nuclear antigen, and fibroblast growth factor 2. These data are interesting in light of evidence that apoptosis tolerance and cell proliferation are important mechanisms involved in chemical carcinogenesis (Waalkes et al. 2000a), including arsenic. Apoptosis tolerance also is accompanied by cell proliferation, as seen in arsenic-transformed cells (Brambila et al. 2002; Chen et al. 2001b; Qu et al. 2002) in chronic arsenic-exposed animals (Chen et al. 2004; Xie et al. 2004), and in liver tumor and nontumor tissues from mice exposed to arsenic in utero (Liu et al. 2004; Waalkes et al. 2003). Thus, the depression of apoptosis genes and the overexpression of cell proliferation genes could be important in arsenic toxicity and carcinogenesis.

Liver is a major target of arsenic carcinogenesis in transplacentally exposed animal models (Waalkes et al., 2003) and in arsenic-exposed humans (Centeno et al. 2002; Chen et al. 1997; Zhou et al. 2002). The expression of [alpha]-fetoprotein (AFP), a biomarker for hepatocellular carcinogenesis, was increased in transplacental arsenic-induced hepatocellular carcinoma (HCC) and tumor-surrounding tissues (Liu et al. 2004). In the present study, all the arsenicals tested increased AFP expression up to 3-fold in MMA(V)-treated mice. The enhanced expression of AFP lends further support that preneoplastic alterations occur after subchronic arsenic exposure. Other notable alterations in gene expression were the overexpression of IGFBP-1 and suppression of insulin-like growth factor 1 (IGF-1). Chronic exposure to nongenotoxic chemicals such as oxazepam and Wyeth-14,643 increased the expression of IGFBP-1 in a time-dependent manner (Iida et al. 2003), and overexpression of IGFBP-I was also seen in transplacental arsenic-induced HCC and tumor-surrounding tissues (Liu et al. 2004). Dysregulation of the IGF axis has been implicated in liver tumor formation and progression (Scharf et al. 2001). Thus, subchronic exposure to arsenicals can produce aberrant gene expression related to hepatocarcinogenesis, some of which were confirmed in the present study.

In summary, this study demonstrated that subchronic exposure to As(III), As(V), MMA(V), and DMA(V) in the drinking water resulted in variable toxic effects, accumulation of arsenic in the liver, hepatic global DNA hypomethylation, and alterations in gene expression in Tg.AC mice. These findings indicate that liver is a target organ of subchronic arsenical exposure in this model and support the idea that altered DNA methylation and its effects on gene expression may contribute in an epigenetic manner to arsenic carcinogenesis.

REFERENCES

Aposhian HV. 1997. Enzymatic methylation of arsenic species and other new approaches to arsenic toxicity. Annu Rev Pharmacol Texicol 37:397-419.

Barchowsky A, Klei LR, Dudek EJ, Swartz HM, James PE. 1999. Stimulation of reactive oxygen, but not reactive nitrogen species, in vascular endothelial cells exposed to low levels of arsenite. Free Radic Biol Med 27:1405-1412.

Baylin SB, Herman JG, Graft JR, Vertino PM, Issa JP. 1998. Alterations in DNA methylation: a fundamental aspect of neoplasia. Adv Cancer Res 72:141-196.

Brambila EM, Achanzar WE, Qu W, Webber MM, Waalkes MP. 2902. Chronic arsenic-exposed human prostate epithelial cells exhibit stable arsenic tolerance: mechanistic implications of altered cellular glutathione and glutathione S-transferase. Toxicol Appl Pharmacol 183:99-107.

Centeno JA, Mullick FG, Martinez L, Page NP, Gibb H, Longfellow D, et al. 2002. Pathology related to chronic arsenic exposure. Environ Health Perspect 110(suppl 5):883-886.

Chen CJ, Yu MW, Liaw YF. 1997. Epidemiological characteristics and risk factors of hepatocellular carcinoma. J Gastroenterol Hepatol 12:S294-308.

Chen H, Li S, Liu J, Diwan BA, Barrett JC, Waalkes MP. 2004. Chronic inorganic arsenic exposure induces hepatic global and individual gene hypomethylation: implications for arsenic hepatocarcinogenesis. Carcinogenesis 25:1-8.

Chen H, Liu J, Merrick BA, Waalkes MP. 2001a. Genetic events associated with arsenic-induced malignant transformation: applications of cDNA microarray technology. Mol Carcinog 30:79-87.

Chen H, Liu J, Zhao CQ, Diwan BA, Merrick BA, Waalkes MP. 2001b. Association of c-myc over-expression and hyperproliferation with arsenite-induced malignant transformation. Toxicol Appl Pharmacol 175:260-268.

Chiou HY, Hsueh YM, Hsieh LL, Hsu LI, Hsu YH, Hsieh FI, et al. 1997. Arsenic methylation capacity, body retention, and null genotypes of glutathione S-transferase M1 and T1 among current arsenic-exposed residents in Taiwan. Mutat Res 386: 197-207.

Counts JL, McClain RM, Goodman JI. 1997. Comparison of effect of tumor promoter treatments on DNA methylation status and gene expression in B6C3F1 and C57BL/6 mouse liver and in B6C3F1 mouse liver tumors. Mol Carcinog 18:97-106.

Del Razo LM, Quintanilla-Vega B, Brambila-Colombres E, Calderon-Aranda ES, Manno M, Albores A. 2001. Stress proteins induced by arsenic. Toxicol Appl Pharmacol 177:132-148.

Gebel T. 2000. Confounding variables in the environmental toxicology of arsenic. Toxicology 144:155-162.

Germolec DR, Spalding J, Boorman GA, Wilmer JL, Yoshida T, Simeonova PP, et al. 1997. Arsenic can mediate skin neoplasia by chronic stimulation of keratinocyte-derived growth factors. Mutat Res 386:209-218.

Germolec DR, Spalding J, Yu HS, Chen GS, Simeonova PP, Humble MC, et al. 1998. Arsenic enhancement of skin neoplasia by chronic stimulation of growth factors. Am J Pathol 153:1775-1785.

Goodman JI, Watson RE. 2002. Altered DNA methylation: a secondary mechanism involved in carcinogenesis. Annu Rev Pharmacol Toxicol 42:501-525.

Gregus Z, Gyurasics A, Csanaky I. 2000. Biliary and urinary excretion of inorganic arsenic: monomethylarsonous acid as a major biliary metabolite in rats. Toxicol Sci 56:18-25.

Hughes MF. 2002. Arsenic toxicity and potential mechanisms of action. Toxicol Lett 133:1-16.

IARC. 1987. Arsenic and arsenic compounds (Group 1). IARC Monogr Eval Carcinog Risks Hum Suppl 7:100-103

Iida M, Anna CH, Harris J, Bruno M, Wetmore B, Dubin JR, et al. 2003. Changes in global gene and protein expression during early mouse liver carcinogenesis induced by non-genotoxic model carcinogens oxazepam and Wyeth-14,643. Carcinogenesis 24:757-770.

Kenyon EM, Hughes MF. 2001. A concise review of the toxicity and carcinogenicity of dimethylarsinic acid. Toxicology 160:227-236.

Kitchin KT, Ahmad S. 2003. Oxidative stress as a possible mode of action for arsenic carcinogenesis. Toxicol Lett 137:3-13.

Leder A, Kuo A, Cardiff RD, Sinn E, Leder P. 1990. v-Ha-ras transgene abrogates the initiation step in mouse skin tumorigenesis: effects of phorbol esters and retinoic acid. Proc Natl Acad Sci USA 87:9178-9182.

Liu J, Chen H, Miller DS, Saavedra JE, Keefer LK, Johnson DR, et al. 2001a. Overexpression of glutathione S-transferase II and multidrug resistance transport proteins is associated with acquired tolerance to inorganic arsenic. Mol Pharmacol 60:302-309.

Liu J, Kadiiska MB, Liu Y, Lu T, Qu W, Waalkes MP. 2001b. Stress-related gene expression in mice treated with inorganic arsenicals. Toxicol Sci 61:314-320.

Liu J, Xie Y, Ward JM, Diwan BA, Waalkes MP. 2004. Toxicogenomic analysis of aberrant gene expression in liver tumors and non-tumorous livers of adult mice exposed in uteroto inorganic arsenic. Toxicol Sci 77:249-257.

Lu T, Liu J, LeCluyse EL, Zhou YS, Cheng ML, Waalkes MP. 2001. Application of cDNA microarray to the study of arsenic-induced liver diseases in the population of Guizhou, China. Toxicol Sci 59:185-192.

Marnell LL, Garcia-Vargas GG, Chowdhury UK, Zakharyan RA, Walsh B, Avram MD, et al. 2003. Polymorphisms in the human monomethylarsonic acid (MMA(V)) reductase/hGST01 gene and changes in urinary arsenic profiles. Chem Res Toxicol 16:1507-1513.

Mass MJ, Wang L. 1997. Arsenic alters cytosine methylation patterns of the promoter of the tumor suppressor gene p53 in human lung cells:a model for a mechanism of carcinogenesis. Mutat Res 386:263-277.

Nesnow S, Roop BC, Lambert G, Kadiiska M, Mason RP, Cullen WR, et al. 2002. DNA damage induced by methylated trivalent arsenicals is mediated by reactive oxygen species. Chem Res Toxicol 15:1627-1634.

Nishikawa, T, Wanibuchi H, Ogawa M, Kinoshita A, Morimura K, Hiroi T, et al. 2002. Promoting effects of monomethylarsonic acid, dimethylarsinic acid and trimethylarsine oxide on induction of rat liver preneoplastic glutathione S-transferase placental form positive fool: a possible reactive oxygen species mechanism. Int J Cancer 100:136-139.

NRC. 1999. Arsenic in the drinking water. In: National Research Council Report. Washington, DC:National Academy Press, 1-310.

NRC. 1996. Guide for the Care and Use of Laboratory Animals. 7th ed. Institute of Laboratory Animal Research. Washington, DC:National Academy Press.

Okoji RS, Yu RC, Maronpot RR, Froines JR. 2002. Sodium arsenite administration via drinking water increases genome-wide and Ha-ras DNA hypomethylation in methyl-deficient C57BL/6J mice. Carcinogenesis 23:777-785.

Petrick JS, Jagadish B, Mash EA, Aposhian HV. 2001. Monomethylarsonous acid (MMA(III)) and arsenite: LD(50) in hamsters and in vitro inhibition of pyruvate dehydrogenase. Chem Res Toxicol 14:651-656.

Pi J, Qu W, Reece JM, Kumagai Y, Waalkes MP. 2003. Transcription factor Nrf2 activation by inorganic arsenic in cultured keratinocytes: involvement of hydrogen peroxide. Exp Cell Res 290:234-245.

Qu W, Bortner CD, Sakurai T, Hobson M J, Waalkes MP. 2002. Acquisition of apoptotic resistance in arsenic-induced malignant transformation: role of the JNK signal transduction pathway. Carcinogenesis 23:151-159.

Salim El, Wanibuchi H, Morimura K, Wei M, Mitsuhashi M, Yoshida K, et al. 2003. Carcinogenicity of dimethylarsinic acid in p53 heterozygous knockout and wild-type 057BL/6J mice. Carcinogenesis 24:335-342.

Sanders JM, Burka LT, Chanas B, Matthews HB. 2001. Comparative xenobiotic metabolism between Tg.AC and p53+/- genetically altered mice and their respective wild types. Toxicol Sci 61:54-61.

Scharf JB, Dombrowski F, Ramadori G. 2001. The IGF axis and hepatocarcinogenesis. Mol Pathol 54:138-144.

Shen J, Wanibuchi H, Salim EI, Wei M, Doi K, Yoshida K, et al. 2003a. Induction of glutathione S-transferase placental form positive foci in liver and epithelial hyperplasia in urinary bladder, but no tumor development in male Fischer 344 rats treated with monomethylarsonic acid for 104 weeks. Toxicol Appl Pharmacol 193:335-345.

Shen J, Wanibuchi H, Salim El, Wei M, Kinoshita A, Yoshida K, et al 2003b. Liver tumorigenicity of trimethylarsine oxide in male Fischer 344 rats--association with oxidative DNA damage and enhanced cell proliferation. Carcinogenesis 24:1827-1835.

Simeonova PP, Wang S, Toriuma W, Kommineni V, Matheson J, Unimye N, et al. 2000. Arsenic mediates cell proliferation and gene expression in the bladder epithelium: association with activating protein-1 transactivation. Cancer Res 60:3445-3453.

Styblo M, Del Raze LM, Vega L, Germolec DR, LeCluyse EL, Hamilton GA, et al. 2000. Comparative toxicity of trivalent and pentavalent inorganic and methylated arsenicals in rat and human cells. Arch Toxicol 74:289-299.

Trouba KJ, Hamadeh HK, Amin RP, Germolec DR. 2002. Oxidative stress and its role in skin disease. Antioxid Redox Signal 4:665-673.

Trouba KJ, Nyska A, Styblo M, Dunson D, Lomnitski L, Grossman S, et al. 2003. Effect of antioxidants on the papilloma response and liver glutathione modulation mediated by arsenic in Tg.AC transgenic mice. In: Arsenic Exposure and Health Effects V (Chappell WR, Abernathy CO, Calderon RL, eds). Amsterdam: Elsevier Science Ltd., 281-291.

Vahter M. 2002. Mechanisms of arsenic biotransformation. Toxicology 181-182:211-217.

Waalkes MP, Fox DA, States JC, Patierno SR, McCabe MJ Jr. 2000a. Metals and disorders of cell accumulation: modulation of apoptosis and cell proliferation. Toxicol Sci 56:255-261.

Waalkes MP, Keefer LK, Diwan BA. 2000b. Induction of proliferative lesions of the uterus, testes, and liver in Swiss mice given repeated injections of sodium arsenate: possible estrogenic mode of action. Toxicol Appl Pharmacol 166:24-35.

Waalkes MP, Liu J, Chen H, Xie Y, Achanzar WE, Zhou YS, et al. 2004a. Estrogen signaling in livers of male mice with hepatocellular carcinoma induced by exposure to arsenic in utero. J Natl Cancer Inst 96:466-474.

Waalkes MP, Ward JM, Diwan BA. 2004b. Induction of tumors of the liver, lung, ovary and adrenal in adult mice after brief maternal gestational exposure to inorganic arsenic: promotional effects of postnatal phorbol ester exposure on hepatic and pulmonary, but not dermal cancers. Carcinogenesis 25:133-141.

Waalkes MP, Ward JM, Liu J, Diwan BA. 2003. Transplacental carcioogenicity of inorganic arsenic in the drinking water: induction of hepatic, ovarian, pulmonary, and adrenal tumors in mice. Toxicol Appl Pharmacol 186:7-17.

Walker NJ. 2001. Real-time and quantitative PCR: applications to mechanism-based toxicology. J Biochem Mol Toxicol 15:121-127.

Wang Z, Dey S, Rosen BP, Rossman TG. 1996. Efflux-mediated resistance to arsenicals in arsenic-resistant and -hypersensitive Chinese hamster cells. Toxicol Appl Pharmacol 137:112-119.

Wei M, Wanibuchi H, Morimura K, Iwai S, Yoshida K, Endo G, et al. 2002. Carcinogenicity of dimethylarsinic acid in male F344 rats and genetic alterations in induced urinary bladder tumors. Carcinogenesis 23:1387-1397.

Xie Y, Liu J, Liu Y, Klaassen CD, Waalkes MP. 2004. Toxicokinetic and genomic analysis of chronic arsenic exposure in multidrug-resistance mdrla/lb(-/-) double knockout mice. Mol Cell Biochem 255:11-18.

Yamamoto S, Wanibuchi H, Hori T, Yano Y, Matsui-Yuasa I, Otani S, et al. 1997. Possible carcinogenic potential of dimethylarsinic acid as assessed in rat in vivo models: a review. Mutat Res 386:353-361.

Yamanaka K, Hoshino M, Okamoto M, Sawamura R, Hasegawa A, Okada S. 1990. Induction of DNA damage by dimethylarsine, a metabolite of inorganic arsenics, is for the major part likely due to its peroxyl radical. Biochem Biophys Res Commun 168:58-64.

Yoshida K, Inoue Y, Kuroda K, Chen H, Wanibuchi 14, Fukushima S, et al. 1998. Urinary excretion of arsenic metabolites after long-term oral administration of various arsenic compounds to rats. J Toxicol Environ Health A 54:179-192.

Zhao CQ, Young MR, Diwan BA, Coogan TP, Waalkes MP. 1997. Association of arsenic-induced malignant transformation with DNA hypomethylation and aberrant gene expression. Proc Natl Acad Sci USA 94:10907-10912.

Zhou YS, Du H, Cheng ML, Liu J, Zhang XJ. 2002. The investigation of death from diseases caused by coal-burning type of arsenic poisoning [in Chinese]. Chin J Endemiol 21:484-486.

Yaxiong Xie, (1) Kevin J. Trouba, (2) Jie Liu, (1) Michael P. Waalkes, (1) and Dori R. Germolec (2)

(1) Inorganic Carcinogenesis Section, Laboratory of Comparative Carcinogenesis, National Cancer Institute at the National Institute of Environmental Health Sciences and (2) Environmental Immunology, National Institute of Environmental Health Sciences, Research Triangle Park, North Carolina, USA

Address correspondence to D.R. Germolec, Environmental Immunology, National Institute of Environmental Health Sciences, P.O. Box 12233, Mail Drop C1-03, Research Triangle Park, NC 27709 USA. Telephone: [phone]. Fax: [phone]. E-mail: [e-mail]

The authors thank J. Pi, L. Benbrahim-Tallaa, and L. Keefer for their critical reviews and help during the preparation of" this manuscript.

The authors declare they have no competing financial interests.

Received 2 April 2004; accepted 17 June 2004.

Table 1. Mortality due to arsenical exposure. (a)

Arsenical                 No.         Found dead

Control                    15              0
AS(111) (150 ppm)          15              1
As(V) (200 ppm)            15              0
MMA(V) (1,500 plum)        15              3
DMA(V) (1,000 ppm)         15              2

                       Sacrificed
Arsenical               moribund     Total (%) (b)

Control                    0             0 (0)
AS(111) (150 ppm)          2             3 (20)
As(V) (200 ppm)            0             0 (0)
MMA(V) (1,500 plum)        3             6 (40) *
DMA(V) (1,000 ppm)         1             3 (20)

(a) Arsenicals (ppm as arsenic) were administered for 17 weeks in
drinking water; animal health was monitored twice daily.

(b) Total dead or euthanized before end of the experiment.

* Statistically significant (p < 0.05) compared with control.

Table 2. Effect of arsenicals on selected gene expression. (a)

                                                         Hybrid
                                        Accession       intensity
Protein/gene                              no. (b)       of control

GST gene
  GST-alpha                               J03958          5,306
  GST-mu                                  U24428          3,824
  GST-pi                                  D30687         25,368
  GST-theta-1                             X98055          3,892

Stress-related genes
  HO-1                                    M33203         13,480
  EGR-1                                   M20157          6,356
  DT diaphorase                           U12961          2,926
  Cytochrome P450 IIIA25 (CVP3A25)        Y11995         19,041

Genes related to apoptosis
and cell proliferation
  FasL                                    U06948          2,006
  TNF receptor-associated factor 3        U21050          3,452
  Bad                                     L37296          3,002
  Granzyme A                              M13226          2,802
  Proliferating cell nuclear antigen      X53068          2,768
  Fibroblast growth factor 2              M30644          1,582
  c-myc proto-oncogene                    X00195          1,981

Tumor-related genes
  Alpha fetoprotein                       V00743          1,631
  Insulin-like growth factor binding      X81579          7,605
    protein 1
  Insulin-like growth factor 1           AF056187         1,216

                                         Intensity relative to
                                            that of control

                                         As (III)         As (V)
Protein/gene                            (150 ppm)       (200 ppm)

GST gene
  GST-alpha                               2.38 *           2.25 *
  GST-mu                                  1.52             1.83 *
  GST-pi                                  1.40 *           1.25
  GST-theta-1                             1.70 *           1.42

Stress-related genes
  HO-1                                    1.19             1.31 *
  EGR-1                                   0.88             1.59 *
  DT diaphorase                           1.81 *           1.79 *
  Cytochrome P450 IIIA25 (CVP3A25)        1.22             1.21

Genes related to apoptosis
and cell proliferation
  FasL                                    0.60             0.65
  TNF receptor-associated factor 3        0.81             0.82
  Bad                                     0.70 *           0.67 *
  Granzyme A                              0.58 *           0.53 *
  Proliferating cell nuclear antigen      1.33 *           1.21
  Fibroblast growth factor 2              2.00 *           2.19 *
  c-myc proto-oncogene                    1.18             1.11

Tumor-related genes
  Alpha fetoprotein                       1.72 *           1.63
  Insulin-like growth factor binding      0.94             1.20
    protein 1
  Insulin-like growth factor 1            0.65             0.95

                                          Intensity relative to
                                            that of control

                                          MMA (V)        DMA (V)
Protein/gene                           (1,500 ppm)     (1,000 ppm)

GST gene
  GST-alpha                               1.23 *           1.17
  GST-mu                                  1.43             1.82 *
  GST-pi                                  0.96             1.40
  GST-theta-1                             1.03             1.12

Stress-related genes
  HO-1                                    1.31 *           1.26 *
  EGR-1                                   1.76 *           1.61 *
  DT diaphorase                           1.28             1.79 *
  Cytochrome P450 IIIA25 (CVP3A25)        1.48 *           1.47 *

Genes related to apoptosis
and cell proliferation
  FasL                                    0.63             0.33 *
  TNF receptor-associated factor 3        0.58 *           0.49 *
  Bad                                     0.81             0.92
  Granzyme A                              0.42 *           0.90
  Proliferating cell nuclear antigen      0.99             0.96
  Fibroblast growth factor 2              2.16 *           1.80 *
  c-myc proto-oncogene                    0.93             1.71

Tumor-related genes
  Alpha fetoprotein                       0.88             1.28
  Insulin-like growth factor binding      2.93 *           1.54 *
    protein 1
  Insulin-like growth factor 1            0.23             0.47

(a) Data are based on the average value of arrays run in triplicate.

(b) Accession numbers are from GenBank
(http://www.ncbi.nlm.nih.gov/entrez/query.fci?db=nucleotide).

* Original hybrid intensity is significantly different from that of
control (p < 0.05).