3. Cellular mechanisms of arsenic toxicity

The levels of ROS play a key role in normal cell signaling, and its alteration can result in aberrant expression of genes that are activated by redox mechanisms. Notably, genes associated with redox mechanisms include those regulating cellular proliferation, differentiation and apoptosis. The consequences of ROS production can further lead to DNA damage which typically involves the conversion of 2-deoxyguanine to 8-hydroxyl-2<sup>0</sup> -deoxyguanosine (8-OHdG),

Figure 2. Biogeochemical cycle of arsenic.

which is considered as a marker indicating oxidative stress of DNA. Arsenic was capable of inducing specific DNA lesions consistent with oxidative damage like 8-OHdG generation. Moreover, 8-OHdG has also been detected in the skin of patients with arsenic-related Bowen's disease and in the liver of rats exposed to dimethylarsinic acid (DMAV). These results indicate that ROS generation is a major pathway for arsenic-mediated genotoxicity in mammalian cells [14, 15].

results in the inclusion of arsenic in the environment [10]. The earth's atmosphere also has a significant presence of arsenic species owing to wind erosion processes, sea spray, hot springs, volcanic emissions, forest fires and volatilization (in cold climates). Human activities like pharmaceutical manufacturing, glassmaking industry, wood processing, chemical weapons, burning of arsenic-rich fossil fuels and electronics industry also contribute to the addition of arsenic compounds into the environment [11]. Industrial by-products and wastes, ore smelting, mineral mining and well drilling can also mobilize and intensify arsenic into the

Microbial metabolisms like arsenate reduction, arsenite oxidation and methylation processes are also a determining factor of the occurrence of the various arsenic oxidation states in the environment. Reduction of arsenate to arsenite by arsenate reductase enzymes is a common feature in the microbial world, while incidences of oxidation of arsenite to arsenate have also been reported in contaminated environments. These reactions also contribute to the protective and/or energy metabolisms of the bacteria from various arsenic-induced stress conditions (Figure 2) [12, 13].

The levels of ROS play a key role in normal cell signaling, and its alteration can result in aberrant expression of genes that are activated by redox mechanisms. Notably, genes associated with redox mechanisms include those regulating cellular proliferation, differentiation and apoptosis. The consequences of ROS production can further lead to DNA damage which typi-


environment.

60 Arsenic - Analytical and Toxicological Studies

3. Cellular mechanisms of arsenic toxicity

Figure 2. Biogeochemical cycle of arsenic.

cally involves the conversion of 2-deoxyguanine to 8-hydroxyl-2<sup>0</sup>

Glutathione and other aminothiols such as cysteine and cysteamine comprise the non-protein sulfhydryls (NPSHs) in a cell and have significant free radical scavenging abilities. Therefore, depletion of intracellular glutathione levels is known to have an effect on arsenic mutagenesis. Studies have shown that pretreatment of cells with an inhibitor of glutathione biosynthesis (buthionine sulfoximine) reduces NPSH levels in the cell, resulting in enhancement of both the cytotoxicity and mutagenicity of arsenic. In contrast, glutathione and cysteine pretreatments are capable of protecting mammalian cells against the toxic effects of arsenite [16].

In a similar way, various antioxidants also have a significant effect on arsenic-induced genotoxicity. The balance between the rate of generation of free radicals and the rate of their removal by various antioxidant enzymes dictates the deleterious effect of oxidative stress. Enzymes like superoxide dismutase (SOD) and catalase are capable of partially suppressing both the toxicity and the mutagenic potential of sodium arsenite. These enzymes catalyze the dismutation of superoxide anions and prevent the formation of hydroxyl radicals by removal of hydrogen peroxide, respectively. Therefore, catalase and SOD are capable of reducing the mutagenic potential of arsenic. This is also consistent with other reports which reveal the ability of sodium arsenite to induce heme oxygenase, an oxidative stress protein, and peroxidase in various human cell lines. Moreover, the arsenite-induced occurrence of sister chromatid exchanges is reduced by SOD in cultured human lymphocytes [16].

In mammalian liver, the methylation of arsenic to MMA and DMA occurs at a high level by an incompletely characterized methyltransferase (Figure 3) using S-adenosylmethionine (SAM) as a methyl donor. SAM is a global methyl donor, required for DNA methylations, and its depletion can lead to hypomethylation of DNA resulting in alteration of gene expression like c-Myc, c-Met, cyclin D1 and induction of carcinogenesis [17, 18].

DNA methylation is an epigenetic modification that plays an important role in controlling the expression of various genes. Methylation generally occurs at cytosine residues located in symmetrical CpG nucleotide sequences, and its alteration, both in the global and regional levels, has been associated with oncogenesis. Methylation of CpG islands in the promoter region suppresses gene expression, as 5-methylcytosine interferes with the binding of transcription factors or other DNA-binding proteins causing reduced transcription. On the other hand, promoter hypomethylation causes overexpression of associated genes. Therefore, aberrant DNA methylation could be an underlying epigenetic mechanism causing altered gene expression that contributes towards the formation of cancers. This has been studied well in hepatocytes where chronic arsenic exposure induces hepatic DNA hypomethylation, which can potentially lead to aberrant gene expression and oncogenic growth in the liver, therefore suggesting a plausible mechanism of hepatocarcinogenesis (major cellular effects of arsenic are summarized in Figure 4) [18].

Figure 3. A homology model for arsenite methyltransferase from humans (AS3MT\_HUMAN) showing arsenic bound to Cys residues. PDB ID: 5EVJ with 42% sequence identity spanning residues 38–327 was used to build the model. The coordinates were downloaded from https://swissmodel.expasy.org/repository/uniprot//Q9HBK9 and refined to introduce the arsenic atom.

transformation. Altogether, cyclin D1 overexpression was seen upon arsenic exposure in multiple in vitro and in vivo model systems of arsenic carcinogenesis, which includes skin and bladder cancers in rodents. Thus, under conditions of arsenic-induced carcinogenesis,

Mechanisms of Arsenic-Induced Toxicity with Special Emphasis on Arsenic-Binding Proteins

http://dx.doi.org/10.5772/intechopen.74758

63

In mouse lung tissue, reduced expression of proteins associated with cellular migration was observed when exposed to low dose of arsenic. On lung tissue of mice fed low-dose arsenic, changes in extracellular matrix (ECM) protein expression and a large increase in matrix metalloproteinase (MMP)-9 expression were revealed [21]. MMPs are responsible for ECM degradation among other proteolyses. MMP-9 is the most prominently studied MMP in the lung and has been associated with a variety of lung diseases [22]. An increase in the ratio of MMP-9 to tissue inhibitor of matrix metalloproteinase (TIMP)-1 was observed under low-level arsenic exposure [23]. This imbalance between MMP-9 and TIMP-1 can cause changes in epithelial wound response, thereby contributing to the progression of airway remodeling. Altered wound response is partly due to increased secretion and activity, upon increasing concentration of arsenic. Therefore, arsenic ingestion may alter wound response and, specifically, MMP-9/TIMP-1 ratios in the lung. To conclude, arsenic is capable of causing or exacerbating lung diseases by directly affecting

Studies have revealed that both c-Jun NH2-terminal kinases (JNKs) and extracellular signalregulated protein kinases (Erks) are activated by arsenite, with their activation varying temporally and depending on the dosage. Various results also indicate that Erk activation but not JNK activation is required for arsenite-induced cell transformation. Expression of the dominantnegative mutant JNK1 blocked induction of apoptosis by arsenite or arsenate compared with

signaling pathways involved in cell migration and remodeling of the airway [24].

overexpression of cyclin D1 is observed consistently [18–20].

Figure 4. Cellular effects of arsenic toxicity.

Estrogens are considered to be liver carcinogens in rodents and are suspected to cause carcinogenesis in humans [19]. Evidence suggests that they cause hepatocellular proliferation and aberrant mitogenesis through ER-mediated mechanisms in addition to the likelihood that they confer epigenetic modifications. Hypomethylation of estrogen receptor-α (ER-α) promoter region caused by arsenic exposure and ER-α overexpression have been found to trigger associated formation of proliferative lesions and hepatocellular carcinogenesis. Therefore, chronic arsenic exposure causes overexpression of ER-α creating hypersensitivity of hepatic cells to endogenous steroids. As evidenced by microarray analysis, various cell cycleregulating genes like cyclin D1, cyclin D2 and cyclin D3 were overexpressed on exposure to arsenic. Liver cells that acquired malignant properties upon arsenic treatment also showed cyclin D1 overexpression. In addition, this overexpression had a direct effect on the observed malignant transformation, as selective cyclin D1 overexpression in the liver was sufficient enough to initiate hepatocellular carcinogenesis. Cyclin D1 can, therefore, be considered as a hepatic oncogene. Cyclin D1 is also known to be upregulated transcriptionally by various growth factors which potentially include estrogens. In estrogen-responsive tissues like the liver and uterus, proliferative lesions and co-overexpression of ER-α and cyclin D1 after chronic arsenic exposure are reported. Cyclin D1 activation by arsenic may be a secondary effect to ERα overexpression as cyclin D1 is potentially an ER-α-linked gene. Therefore, we can expect that aberrant expression of cyclin D1 along with that of other oncogenes leads to carcinogenic Mechanisms of Arsenic-Induced Toxicity with Special Emphasis on Arsenic-Binding Proteins http://dx.doi.org/10.5772/intechopen.74758 63

Figure 4. Cellular effects of arsenic toxicity.

Estrogens are considered to be liver carcinogens in rodents and are suspected to cause carcinogenesis in humans [19]. Evidence suggests that they cause hepatocellular proliferation and aberrant mitogenesis through ER-mediated mechanisms in addition to the likelihood that they confer epigenetic modifications. Hypomethylation of estrogen receptor-α (ER-α) promoter region caused by arsenic exposure and ER-α overexpression have been found to trigger associated formation of proliferative lesions and hepatocellular carcinogenesis. Therefore, chronic arsenic exposure causes overexpression of ER-α creating hypersensitivity of hepatic cells to endogenous steroids. As evidenced by microarray analysis, various cell cycleregulating genes like cyclin D1, cyclin D2 and cyclin D3 were overexpressed on exposure to arsenic. Liver cells that acquired malignant properties upon arsenic treatment also showed cyclin D1 overexpression. In addition, this overexpression had a direct effect on the observed malignant transformation, as selective cyclin D1 overexpression in the liver was sufficient enough to initiate hepatocellular carcinogenesis. Cyclin D1 can, therefore, be considered as a hepatic oncogene. Cyclin D1 is also known to be upregulated transcriptionally by various growth factors which potentially include estrogens. In estrogen-responsive tissues like the liver and uterus, proliferative lesions and co-overexpression of ER-α and cyclin D1 after chronic arsenic exposure are reported. Cyclin D1 activation by arsenic may be a secondary effect to ERα overexpression as cyclin D1 is potentially an ER-α-linked gene. Therefore, we can expect that aberrant expression of cyclin D1 along with that of other oncogenes leads to carcinogenic

Figure 3. A homology model for arsenite methyltransferase from humans (AS3MT\_HUMAN) showing arsenic bound to Cys residues. PDB ID: 5EVJ with 42% sequence identity spanning residues 38–327 was used to build the model. The coordinates were downloaded from https://swissmodel.expasy.org/repository/uniprot//Q9HBK9 and refined to introduce

the arsenic atom.

62 Arsenic - Analytical and Toxicological Studies

transformation. Altogether, cyclin D1 overexpression was seen upon arsenic exposure in multiple in vitro and in vivo model systems of arsenic carcinogenesis, which includes skin and bladder cancers in rodents. Thus, under conditions of arsenic-induced carcinogenesis, overexpression of cyclin D1 is observed consistently [18–20].

In mouse lung tissue, reduced expression of proteins associated with cellular migration was observed when exposed to low dose of arsenic. On lung tissue of mice fed low-dose arsenic, changes in extracellular matrix (ECM) protein expression and a large increase in matrix metalloproteinase (MMP)-9 expression were revealed [21]. MMPs are responsible for ECM degradation among other proteolyses. MMP-9 is the most prominently studied MMP in the lung and has been associated with a variety of lung diseases [22]. An increase in the ratio of MMP-9 to tissue inhibitor of matrix metalloproteinase (TIMP)-1 was observed under low-level arsenic exposure [23]. This imbalance between MMP-9 and TIMP-1 can cause changes in epithelial wound response, thereby contributing to the progression of airway remodeling. Altered wound response is partly due to increased secretion and activity, upon increasing concentration of arsenic. Therefore, arsenic ingestion may alter wound response and, specifically, MMP-9/TIMP-1 ratios in the lung. To conclude, arsenic is capable of causing or exacerbating lung diseases by directly affecting signaling pathways involved in cell migration and remodeling of the airway [24].

Studies have revealed that both c-Jun NH2-terminal kinases (JNKs) and extracellular signalregulated protein kinases (Erks) are activated by arsenite, with their activation varying temporally and depending on the dosage. Various results also indicate that Erk activation but not JNK activation is required for arsenite-induced cell transformation. Expression of the dominantnegative mutant JNK1 blocked induction of apoptosis by arsenite or arsenate compared with vector-transfected JB6 cells, indicating the role of activation of JNKs in arsenic-induced apoptosis. Studies have found that both arsenite and arsenate can cause transactivation of activation protein-1 (AP-1). Since increased activation of AP-1 by arsenite could be inhibited by either treating cells with MAP kinase Erk kinase (MEK)1 inhibitor or overexpression of dominantnegative protein kinase C α (PKCα), this induction appears to occur through activation of mitogen-activated protein (MAP) kinases and PKC. Moreover, in AP-1-luciferase reporter transgenic mice, transactivation of AP-1 was caused by both arsenite and arsenate. Recent data also indicates that PKC, upstream from the MAP kinases, may be involved in mediating arseniteinduced signal transduction. Activation of PKC requires it to be translocated from the cytosol to the membrane, and this phenomenon is observed within 15 minutes when cells are treated with arsenite. Moreover arsenite-induced AP-1 activity, phosphorylation of Erks, JNKs and p38 kinase were blocked once PKC activation was inhibited. These results suggest that PKC plays a critical role in arsenite-induced activation of MAP kinases [25, 26].

Nuclear factor kappa B (NF-κB) is a rapidly induced stress-responsive transcription factor that may play an important role in arsenic-induced signal transduction, cell transformation and apoptosis [27]. Reports suggest that in arsenic-induced oxidative stress, H2O2 and superoxide are the predominant reactive species in endothelia cells and may be the mediators for the activation of the NF-κB pathway. It was also shown that arsenic could induce activation of NF-κB in different cell culture models. Expression of a dominant-negative inhibitory kappa-Βα blocked arsenic-induced activation of NF-κB and apoptosis [26].

4.1. Binding sites of arsenic in proteins

introduce the arsenic atom.

donors [35–37].

Cysteine and histidine residues are thought to be the most frequent targets of metals like zinc, copper and iron resulting in such metals binding to peptides and proteins [32, 33]. However, the binding of arsenic to histidine is not well understood and is yet to be established. There is no change in nuclear magnetic resonance (NMR) spectra once arsenic was added to a buffered solution of histidine signifying the absence of interaction between histidine and arsenic [34]. Many studies have also used site-directed mutagenesis to replace cysteine residues with serine residues on the reason that interaction between arsenic and serine is very weak. Arsenic is however known to bind to zinc finger protein in C3H1 motif and not in C2H2 motif, releases zinc, and thus decreases the capacity of the protein to bind to DNA. Selenocysteine—a cysteine analogue—also has the ability to bind to arsenic species. This amino acid is present in selenoproteins and has a lower pKa which increases nucleophilicity. The amino acid residues in the vicinity of cysteine (or selenocysteine) act as proton

Figure 5. A homology model for arsenical resistance operon repressor protein from E. coli (ARSR\_ECOLI) showing arsenic bound to Cys residues. PDB ID: 1SMT with 40% sequence identity spanning residues 8–90 was used to build the model. The coordinates were downloaded from https://swissmodel.expasy.org/repository/uniprot//P37309 and refined to

Mechanisms of Arsenic-Induced Toxicity with Special Emphasis on Arsenic-Binding Proteins

http://dx.doi.org/10.5772/intechopen.74758

65

Studies with some enzymes reveal that serine residues can be potential targets of arsenic species, thereby inhibiting their function. It was found that in serine hydrolases and the arsenic moieties interacting with hydroxyl containing serine, pentavalent forms of arsenic rather than trivalent forms were prevalent. The complex between the serine residue and the pentavalent

arsenic consists of a tripartite oxyanion hole in the proximity of the active site [38].
