Epigenetic effects of cadmium in cancer: focus on melanoma.

Cadmium is a highly toxic heavy metal, which has a destroying impact on organs. Exposure to cadmium causes severe health problems to human beings due to its ubiquitous environmental presence and features of the pathologies associated with pro-longed exposure. Cadmium is a well-established carcinogen, although the underlying mechanisms have not been fully under-stood yet. Recently, there has been considerable interest in the impact of this environmental pollutant on the epigenome. Be-cause of the role of epigenetic alterations in regulating gene expression, there is a potential for the integration of cadmium-induced epigenetic alterations as critical elements in the cancer risk assessment process. Here, after a brief review of the ma-jor diseases related to cadmium exposure, we focus our interest on the carcinogenic potential of this heavy metal. Among the several proposed pathogenetic mechanisms, particular attention is given to epigenetic alterations, including changes in DNA methylation, histone modifications and non-coding RNA expression. We review evidence for a link between cadmium-induced epigenetic changes and cell transformation, with special emphasis on melanoma. DNA methylation, with reduced expression of key genes that regulate cell proliferation and apoptosis, has emerged as a possible cadmium-induced epigenetic mechanism in melanoma. A wider comprehension of mechanisms related to this common environmental contaminant would allow a better cancer risk evaluation.

INTRODUCTION

Heavy metals are the most common environmental pollutants all over the world. Heavy metal contamination provoked by natural and anthropogenic sources has become a global issue. The ubiquitous and indestructible nature of heavy metals turns them into one of the main causes of environmental pollution.

Environmental pollution has become a matter of grave importance in relation with industrial development, provoking pedosphere, hydrosphere and atmosphere deterioration. Unsuitable disposal practices related to industrial production are considered to be one of the main reasons for environmental pollution [1]. It has been demonstrated that industrial effluents contain high concentrations of metals, such as zinc, copper, iron, manganese, lead, nickel and cadmium [2]. Such effluents deposit heavy metals in soil and change their intake in the food chain, causing bioaccumulation. Although it iswell known that heavy metals can disturb fundamental cellular processes, e.g. by interfering with intracellular redox state regulation, their potential to provoke disease is still not fully understood [3, 4]. The same goes for their role in the onset and progression of tumors, despite the fact that environmental metal carcinogenesis is considered a major public health problem. In particular, while the association of heavy metal pollution with cancer has been known for some time, the mechanisms underlying malignant transformation remain to be clarified.

Cadmium is considered to be the most toxic metal present in industrial effluents. The pollutant is led into human beings by water, metal plating, batteries, fertilizers, mining, pigments, stabilizers, alloy material and sewage sludge [5]. Non-occupational exposure is mainly due to diet and smoking [6]. In fact, tobacco plants can absorb high quantities of cadmium from the environment and smoking contributes considerably to cadmium exposure. A cigarette contains about 2.0 ug of cadmium [7]. Cadmium levels in blood and body fluids are double in people who smoke relative to non-smokers [8]. More recently, cadmium exposure has been also associated with airborne particulate CdO and with cadmium-containing quantum dots in medical applications [8, 9]. Cadmium, as it has been said, is extremely toxic and tends to lay in tissues [10]. The average cadmium blood level is 0.315 μg/L, whereas the mean urine level of is 0.193 μg/g creatinine (0.185 μg/L) [11]. As soon as blood absorbs it, most of cadmium binds with proteins, such as albumin and metallothionein (MT). Liver is the first organ to be reached and there cadmium activates MT production. After having triggered hepatocyte necrosis and apoptosis, cadmium-metallothionein groups move into sinusoidal blood and enters the enterohepatic cycle through secretion into the biliary tract in the form of cadmium-glutathione conjugates. Once it has been enzymatically degraded to cadmium-cysteine complexes in the biliary tree, cadmium reaches the small intestine, from which it can be reabsorbed [12]. Kidney is the main organ for long-term cadmium accumulation [13]. Here the half-life for cadmium is approximately 10 years. A life-long intake can therefore lead to a cadmium accumulation in the kidney, thus provoking tubulus cell necrosis. The presence of high cadmium levels in the blood and urine indicates a recent and past exposure, respectively [14]. Cadmium is excreted via faeces and urine. (Fig. ) shows a scheme on the handling of cadmium in the human body. Several studies took into exam the role of cadmium in smokers’ chronic obstructive pulmonary diseases (COPD) [15]. Smokers, or former ones, show higher levels of cadmium relative to non smokers and the presence of cadmium is linked to lung diseases provoked by smoking [7], suggesting that cadmium plays a pivotal role in lung disease and in lung cancer [16, 17]. A massive exposure to cadmium, for example in the workplace, provokes respiratory diseases, such as rhinitis, olfactory epithelium destruction, anosmia and bronchitis [15, 18].

Foods which have been grown in areas contaminated by cadmium are generally the main source of cadmium exposure. People take up about 30-50 micrograms per day [19, 20], but they usually absorb orally only small quantities (1- 10%) [21].

A low intake of vitamin D, calcium and the presence of metals, such as zinc and copper, increase cadmium absorption. In the case of zinc and calcium, their molecular homology could be one of the mechanisms underlying higher cadmium absorption [22]. Foulkes showed that, in rat jejunum, cadmium uptake was depressed by high concentrations of polyvalent cations, including Pb, Ni, Cr3+, Sr, and Mg [23].

Cadmium intake is increased by diets rich in fibers and poor in iron [24]. People with low iron intakes have a higher cadmium absorption relative to people with well-balanced iron supplies [25]. For example, patients suffering from anaemia and iron lack, such as children and menstruating women, absorb higher levels of cadmium. Interestingly, low iron levels in blood stimulate a divalent cation transporter, which is employed for cadmium absorption in the gastro-intestinal tract [26].

CADMIUM AND DISEASES

In the world, there are areas showing high levels of cadmium in their soils and people who live there are exposed to cadmium absorption. For example, in the Jinzu and Kakehashi river basins in Japan, soils are contaminated and rice absorbs cadmium. Eating it for prolonged periods of time leads to serious kidney illnesses and provokes a bone disorder called “Itai-Itai”, which mainly affect women [27, 28]. Patients show symptoms resulting from low levels of bone mineralization, such as fractures, osteoporosis and pains in their bones. In a study on this topic, patients showed the latter symptoms after having eaten rice grown on fields irrigated with highly cadmium-polluted water. Fractures provoked by osteomalacia and skeleton decalcification were also noticed. These studies were criticised as mostly post-menopausal women were involved [29] who had a high percentage of osteoporosis [30]. In animals, continuous cadmium absorption causes high systolic blood pressure even in absence of renal disease. These effects have been linked to depressed levels of atrial natriuretic peptide, increased blood amounts of aldosterone, and retention of sodium and water [15]. Cadmium exposure in human beings might be also related to hypertension. There are several studies on this topic. The Cadmibel study found no relationship between cadmium and blood pressure [31]. This study was carried out on people that underwent low-level exposure to cadmium. The follow-up of the original Cadmibel cohort, the PheeCad study, confirmed this result [32]. However, other studies on the same topic [33, 34] have shown a role for cadmium in peripheral artery disease. In particular, a correlation was observed between the levels of cadmium exposure and the pathological effects exerted by smoke on peripheral arteries.

People who work with cadmium can have problems if proper industrial hygiene measures are not applied. Occupational exposure occurs predominantly via inhalation of dusts in metallurgic industries, where batteries are produced and in electroplating processes [5, 35]. Cadmium has a very long biological life, which makes it a cumulative toxin, so even a past exposure to such a metal can provoke damages after a long time [36]. Cadmium toxic properties have been known for over 150 years [37], but the issue regarding low-level exposure to it is still on.

Cadmium severe intoxication provokes injuries to the testes, liver and lungs [37, 38]. Chronic exposure provokes respiratory diseases, emphysema, end-stage renal failure, diabetic and renal complications, deregulated blood pressure, bone problems and immune suppression [35, 24, 40]. The main organ affected by cadmium exposure is the kidney. Cadmium inhalation and ingestion may lead to nephrotoxicity. Data from human studies suggest renal damage due to cadmium exposure can be seen after 10 years. In animals, massive cadmium exposure provokes renal toxicity, but few data are available in people.

Increasing cadmium exposure can be related to growing renal tubular dysfunction. Renal malfunction can be evidenced at levels of 2 μg/g creatinine by microscopic tubular proteinuria or using biomarkers such as β2-microglobulin and α1-microglobulin. At urinary cadmium levels of 4 μg/g creatinine, N-acetyl-B-glucosaminidase (NAG) is elevated in urine and signs of glomerular damage are observed. In the final stages of cadmium intoxication it is possible to evidence nephropathy, glycosuria, calcium and phosphate loss, and altered calcium metabolism with osteoporosis and osteomalacia [41, 42]. Some studies argue that microproteinuria due to cadmium exposure is not progressive and that cadmium nephropathy becomes progressive and irreversible when it reaches cadmium levels higher than 4µg/g creatinine or a β2- microglobinuria higher than 1,000 µg/g creatinine [43, 27]. Toxic effects on kidneys depend on cadmium doses [44]. Some researches state that renal tubular dysfunction related to cadmium is irreversible [45]. Cadmium nephropathy is one of the main causes of mortality in people who work in contaminated environments. The risk of nephropathy in people who work in cadmium-contaminated places was high when total airborne exposure was higher than 300 mg/m3, urine cadmium levels were higher than 10 µg/g creatinine, and renal cortex levels were higher than 200 ppm [41]. People show early signs of kidney damage with urine levels between 2-4 nmol/mmol creatinine, as shown by a considerable literature on this topic [27, 42, 43, 46-49]. These studies show that even small amounts of cadmium can provoke kidney diseases. The World Health Organization (WHO) has stated that 200 μg/g levels wet weight in kidney cause detrimental changes in 10% of the population [50]. As mentioned above, literature on the topic showed that the threshold for renal damage was urinary cadmium levels of 2-4 nmol/mmol creatinine [46]; however, the OSCAR study showed that people who exhibited a urine cadmium level of 1 nmol/mmol creatinine had a threefold risk of increased α-1 microglobulin [42]. However, it has not been determined yet if changes in kidney biomarkers, as seen in conditions of low-level cadmium exposure, have any correlation with the impairment of renal function [49]. At the moment, researchers are at work to define the amount of cadmium which provokes kidney diseases. Recent studies in polluted areas suggest that a dose of cadmium is 2.0 grams can be tolerated in men and women [51, 52]. In smokers there is a slight margin between critical renal concentration and body burdens [6]. Some researchers also suggests that children might be especially susceptible to cadmium [51]. Early signs of glomerular damage provoked by cadmium are increased excretion of albumin and transferrin. Glomerular damage is dose-dependent and can be irreversible [53]. Glomerular filtration rate (GFR) decreases slowly but progressively and it’s possible that cadmium accelerates the normal age-related decline in kidney function. Uremia is not common, but people with normal baseline GFR and serum creatinine, who work in environments contaminated by cadmium, show a decreased filtration reserve capacity [46]. Medium-level exposure to cadmium can cause decreased GFR and chronic renal failure, which manifest itself with aminoaciduria, glicosuria, hypercalciuria, hyperphosphaturia, polyuria, and reduced buffering capacity for acids [53]. Kidney stones are common in people exposed to cadmium, above all in those with occupational exposure. Stone formation, that is secondary to kidney damage, can be related to hypercalciuria and hyperphosphaturia. Other factors which contribute to stone formation are uric aciduria, reduced urinary citrate, and renal tubular acidosis. Cadmium also accumulates in bones where it might promote pathological changes. However, bone disease resulting from excessive cadmium exposure is believed to be mainly secondary to changes in calcium metabolism due to cadmium-induced renal damage [19]. Bone lesions occur in cases of severe cadmium poisoning and include pseudo-fractures and other effects of osteomalacia and osteoporosis. Spontaneous fractures can occur anywhere in the skeleton or in places where the main arteries cross the bone. In Sweden the OSCAR study, which we have mentioned before, has examined the relationship between cadmium exposure and risk for reduced density in bone [42, 54, 55]. Researchers found out a negative correlation between cadmium in urines and bone density. Skeletal effects appeared to be secondary to increased urinary calcium and phosphorus losses due to cadmium-induced effects on kidney [42]. These included decreased renal hydroxylation of vitamin D which provokes deficiency of its active form. Moreover, an inhibitory effect on calcium absorption from the gastrointestinal tract can be exerted by cadmium [56]. When animals are pregnant, cadmium crosses the placenta inducing malformations in the newborns. Amounts of cadmium exceeding 2.5 mg/kg cause severe placental damage and lead to fetal death. In the case of pregnant women, it has been reported that cadmium exposure in the working place does not induce malformations in newborns. However, in Japan, women with higher urinary cadmium levels have increased rates of preterm delivery than mothers with lower levels. Infant weight was also lower in the former group [57]. However, other studies did not show that cadmium causes pre-term labour [58]. At this time, the evidence of cadmium's effects on pregnancy is inconsistent and requires further investigation. Cadmium can seriously damage the reproductive system [59] as it can mimic estrogen function by virtue of its binding with high affinity to estrogen receptors (ER) [60], leading to gene activation [61]. Alternatively, estrogen receptors can exert their effects through other mechanisms, such as the activation of secondary messengers and enzymatic pathways [62]. Likely, the high incidence of tumors of the genital sphere in industrialized countries can be related to pollutants, such as cadmium, interfering with the hormonal network [63, 64]. In some, but not all, cell culture systems cadmium exerts either genomic or non-genomic effects [65-67]. The different hormonal effects exerted by cadmium have been attributed to the presence of different pools of ER-alpha with different reactivity to cadmium, as shown in (Fig. ) [68, 69]. A recent review evaluated the effects on neurodevelopmental disorders of pre- or post-natal exposure to arsenic, cadmium, and manganese in children up to 16 years of age. Very little evidence was found about effects of cadmium that seemed consistent with neurodevelopmental and/or behavioral disorders. [70]. No significant differences have been observed between children and adults and developmental or behavioral changes have not been generally reported in populations exposed to cadmium. Therefore further studies are needed to carefully examine these potential effects of cadmium exposure. A limited number of animal studies suggest that young animals may absorb more cadmium than adults and may therefore display increased bone loss and fragility [11]. Recent studies have indicated that cadmium, even at non-toxic concentrations, can induce inflammatory changes in the airways [71] and in the vascular bed [72], and that these effects might be mediated by its ability to induce CXCL1 and CXCL2 chemokines [73]. These chemokines have central role in promoting neutrophil infiltration, which, in turn, can further amplify the production of not only CXCL1 and CXCL2 but also and of the potent pro-inflammatory cytokine IL-1β [74, 75]. Further studies will be needed to ascertain whether cadmium can directly potentiate the activation of receptors of the innate immune system, including the Toll-like receptors [76-78] and Nod-like receptors [79], leading to the induction of proinflammatory cytokines. All adverse effects caused by cadmium are due to its low excretion rate (half-life as long as 15–20 years) [80], and its consequent accumulation in the organism. In (Fig. ) are reported the effects of cadmium on several organ systems.

CADMIUM AND CANCER

Studies have shown an increasing incidence of cancer in association with increasing contamination of the environment by cadmium. The effects of cadmium at the cellular level can result in a variety of metabolic derangements, alterations in the proliferative activity with reduced or increased cell growth, perturbations of apoptosis programs, modifications in the homeostasis of redox systems, cytotoxic as well as genotoxic responses. Along with all these activities, cadmium can also act as a carcinogen, making it extremely dangerous to human health. The mechanisms involved could include aberrant gene expressions, errors in DNA methylation, apoptosis blockage and differentiation disruption, to cite a few. These phenomena can result in cadmium-induced carcinogenic transformation in the absence of direct or indirect genetic damage, which is consistent with the metal’s poor ability to produce mutations at sub-lethal dosages. Cadmium can activate oncogenes or genes associated with cell proliferation, such as c-myc, c-jun or c-fos, both in vivo and in vitro [81] and it can also induce up-regulation of signal pathways resulting in increased mitogenesis, such as, for example, the AP-1 and MAP kinase pathways [82]. A direct action of cadmium on the expression of E-cadherin, a transmembrane Ca(II)-binding glycoprotein, has not been demonstrated yet, but the few available data indicate that cadmium may disrupt E- cadherin-dependent cell communication and promote cell division [83]. Such event may be one of the different mechanisms by which cadmium promotes tumour formation. Another mechanism may involve suppression of DNA repair that can allow damaged cells to enter the S phase of the cell cycle. Enhanced proliferation may by-pass apoptosis with a consequent accumulation of cells harbouring serious DNA impairment and development of pre-neoplastic or early neoplastic cells [84, 85]. Thus, cadmium-induced cell growth enhancement and reduced apoptosis may together cooperate in increasing cell population and promoting cancerogenesis.

Moreover, cadmium effects may synergize with those of the other pro-neoplastic agents, such as carcinogens present in cigarette smoke, with whom it is often associated. The understanding the mechanisms of cadmium resistance involved in the acute and chronic responses to low, non cytotoxic concentrations of this pollutant may be particularly useful, since they may provide a linkage between human health and environmental, occupational, and every-day life conditions. Although cadmium was associated with the development of prostate and testicular tumours in rats [86], no studies have definitively proved that cadmium can cause prostate cancer in humans [87]. But, nevertheless, the IARC (International Agency for Research on Cancer) classified cadmium as a group I carcinogen by virtue of its effect in inducing renal cancer [88, 89]. Some evidences indicate that cadmium may be tumorigenic only when it is absorbed through the respiratory tract [14]. The molecular mechanisms of cadmium transformation have not yet been well defined, although several signal transduction pathways have been implicated. Either increased proliferation rate or decreased apoptosis have been linked to cadmium-induced neoplasia. Alterations in DNA mismatch repair genes have also been described. Moreover, cadmium was shown to interfere with the transcriptional machinery by substituting for zinc in some transcriptional factors [90]. Epidemiological studies have provided further evidence that in addition to lung cancer, other cancers can be induced by cadmium [35, 91, 92]. Altogether, the German MAK Commission concluded that an increased relative risk of renal cancer has to be assumed [93] and also International Agency for Research on Cancer IARC stated a positive association with respect to renal and prostate cancer [94]. Finally, an association was also found between cadmium toxicity and breast and endometrial cancer [95, 63]. As mentioned above, cadmium appears not to be a direct carcinogen, as it weakly binds to DNA. However cadmium can induce perturbations of genome stability through several indirect mechanisms, such as the production of reactive oxygen species (ROS), the disturbance of cellular signalling processes, DNA repair, mitotic cycle regulation, induction of apoptosis as well as epigenetic alterations, which may act together and lead to carcinogenesis [96]. The biological effects of cadmium are strictly dose-dependent. Cadmium exposure induces mitogenic signaling at concentrations above 1μM by increasing the levels of kinases of the RAS pathway: RAF-1, MAPK, MEK1/2, ERK1/2, p38 MAPK and JNK, beyond its effects on the protoncogenes c-fos, c-myc, and c-jun [97]. At lower concentrations, it enhances cell proliferation through distinct mechanisms [98]. The effects of cadmium on apoptosis are complex and both pro-apoptotic and anti-apoptotic consequences can be observed under different conditions of exposure to the metal. Several groups of researchers showed that cadmium-induced apoptosis is mediated by up-regulation and activation of p53 [99, 100, 85]. However, the implication of p53 is not always found in all biological systems [101, 102], being oxidative stress often involved in the activation of apoptosis by cadmium. Moreover, interactions with the DNA damage response systems [36] and the induction of oxidative stress appear to be relevant in cadmium-induced genotoxicity. It is worth to note that the increase in ROS levels is not direct, as cadmium lacks redox activity and it is not able to direct Fenton type-reaction. Furthermore, several studies show that cadmium associates with proteins, such as metallothionein or albumin, rather than with DNA [103, 104] and this is probably why cadmium-induced genotoxicity is mostly due to indirect processes [35, 105]. It is well known that cellular redox homeostasis, which is maintained by an elaborate endogenous system of enzymatic and non-enzymatic anti-oxidant biological molecules [96], is heavily involved in the regulation of apoptosis. Decreased antioxidant activities induce apoptosis and overproduction of ROS leads to inhibition of apoptosis [106-112]. In any case, it has not been well established whether suppression of apoptosis contributes to carcinogenesis induced by cadmium, particularly in the lung [113, 114]. Another issue is the increased resistance to high doses of cadmium and the cross-tolerance to an otherwise lethal dose of oxygen observed in animals subjected to cadmium inhalation [115, 116]. Moreover, lung cells adapted to cadmium by repeated exposure, in vitro or in vivo, showed greater resistance to oxidation, mainly through modifications in cellular thiols, such as MT and GSH [117]. However, although exerting a protective effect, this response may give to the cells the opportunity to evade from apoptosis and to clonally expand. This would be particularly damaging in a stem cell like the type II pneumocyte that is the putative target of carcinogenesis caused by cadmium [118]. Recently, it has been shown that cadmium is able to induce apoptosis in naive alveolar epithelial cells [119], likely by a mechanism of oxidative stress [120]. Indeed, cadmium may also affect antioxidative proteins, such as catalase and glutathione reductase, which have a crucial role in ROS elimination and induce, thereby, a reduction in cellular antioxidants and release of ROS by mitochondria [121]. Cadmium is believed to impair not only the repair of DNA damage caused by oxidative stress [92, 122, 123] but also the repair of DNA double-strand breaks and DNA interstrand crosslinks.

It is noteworthy that cadmium has been shown to act as a tumour “progressor” in mouse models of sarcomas [124]. Therefore, when cadmium exposure occurs in the presence of other human carcinogens, such as those contained in tobacco smoking, or with UV, synergistic effects may be observed. For example, cadmium interfered with the removal of thymine dimers following UV irradiation [125]. Moreover smoking delays epithelial wound healing [126] and disrupts the antioxidant/oxidant balance [127, 128]. Several findings indicate that cadmium is a strong multi-route, multi-site, multi-species carcinogen in rodents. Cadmium exposure has been associated with tumors of the lung, testes, injection site, prostate, hematopoietic system, pancreas, adrenals, liver, kidney and pituitary and several of these sites concur with potential target sites of cadmium carcinogenesis in humans [8]. The in vivo mechanisms of action for cadmium carcinogenesis is still not clear, with the exception of rodent testes. Given that high doses of cadmium are required to induce testicular tumors in rodent, it is uncertain whether these malignancies are relevant to human cadmium exposure. On the other hand, there is evidence that multiple, target tissue-specific mechanisms of cadmium carcinogenesis may apply and no correlation was demonstrated between cadmium exposure and cytogenetic effects in humans [129]. Therefore, the lack of convincing evidence in humans reduces the likelihood that genotoxicity is the main mechanism of cadmium carcinogenesis.

EPIGENETIC EFFECTS OF CADMIUM

Over the years it has become increasingly evident that mutagenesis is not the only mechanism underlying the activities of carcinogens [130, 131]. Instead, epigenetic events appear to be involved in the starting phase of heavy metal carcinogenesis by “writing” and “erasing” epigenetic signals at the promoter regions of several cancerous genes.

Epigenetic alterations regulate key events in cellular homeostasis, including transcriptional and translational regulation of gene expression. The most well studied epigenetic alterations are DNA methylation, the covalent, post-translational modification of histones, and non-coding small RNAs (miRNAs) [132-135]. Epigenetic alterations can be induced by environmental stimuli, and much attention has been given to the role of the epigenome in human disease [136]: a role that arises primarily from the control that the epigenome exerts over the transcriptome and the proteome. The transcriptome, which is the total transcribed RNA of a cell at a given point in time, is regulated primarily through transcription and mRNA stability and degradation. DNA methylation, histone modifications, and miRNA all regulate the transcriptome through transcriptional processes. The proteome, the total proteins of the cell, is the functional mediator between the genome and the cell and is regulated primarily through post-transcriptional regulation of mRNA transcripts. Epigenetic regulation of the proteome occurs primarily through the action of miRNAs on mRNA transcript stability and translation. In (Fig. ), the main epigenetic effects induced by heavy metals are reported. The recent scientific interest aroused by cadmium for its effects on human carcinogenesis has been enhanced by its ability to interfere with gene expression despite its inability to alter gene structure. At the moment, cadmium is predominantly considered as an epigenetic carcinogen. Many studies converge on its ability to induce epi-mutations, mostly by modifying the DNA methylation pattern of target genes, although the definition of ways and mechanisms involved is still far from being achieved. The carcinogenic potential of cadmium through epigenetic changes includes silencing of DNA repair and tumor-suppressor genes.

DNA Methylation

Both DNA hypomethylation and hypermethylation were found according to the time of exposure to the metal, but while some researchers have shown that hypermethylation appears to be an early signature for cadmium-induced cancer [137], other studies have highlighted that the acute exposure to cadmium prevents DNA methyltransferase activity leading to global DNA hypomethylation, while extended exposure results in increased DNA methyltransferase activity and DNA hypermethylation. Likely, the discrepancies may be due to the different species examined, humans or animals, and to the different cell types used.

Recently, it was found that cadmium-induced transformation of rodent liver cells is associated with DNA hypomethylation, likely due to inhibition of DNA methyltransferases in the early stages of cadmium exposure [138]. Another work indicates that cadmium is an effective inhibitor of nuclear DNA methyltransferases in rats [139].

There is evidence also in humans that cadmium exposure can, under certain circumstances, alter DNA methylation, which may represent an additional mechanism for cadmium-induced carcinogenesis [140]. For example, life-long cadmium exposure in women, as evidenced by low-level urinary cadmium concentrations, was inversely associated with a marker of global DNA methylation, such as long interspersed nuclear element-1 (LINE-1) methylation, and with the expression of DNMT3B, which is involved in de novo CpG methylation. These data are of concern, since genome-wide methylation loss from LINE-1 promoter sites, which are normally heavily methylated at CpG sites, is considered as a frequent epigenetic change in malignancies and may play a role in carcinogenesis [141-143]. However, urinary cadmium was not associated with alterations in the expression of DNMT1, a factor that plays a role in maintaining CpG methylation patterns in the genome. The association between urinary cadmium and decreased LINE-1 methylation was stronger among carriers of the rare alleles of DNMT1 rs10854076 and rs2228611, indicating that cadmium effects can vary according to DNMT1 polymorphisms. A recent work links arsenite and cadmium exposure with agglomerative DNA methylation, confirming that this may be one avenue by which these chemicals drive carcinogenesis [144]. Agglomerative DNA methylation, in which many differentially methylated regions (DMRs) are concentrated within a certain region of a chromosome, can be part of long-range epigenetic silencing (LRES), which is implicated in cancer progression and associated with changes in histone modifications [145]. While the role of LRES in cancers continues to gain supporting evidence, which environmental factors trigger this kind of epigenetic change, if any, has been unclear. Agglomerative DNA hypermethylation has been observed in different kinds of human tumors in association with a reduction in plasticity-associated gene transcription of consecutive genes [145]. Moreover, agglomerative DNA hypermethylation was found in toxicant-transformed cell lines RWPE-1 and UROtsa (derived, respectively, from the prostate and ureter and used as models to study epigenetic dysfunction related to malignant transformation induced by arsenicum or cadmium), thus providing a link between in vitro chronic exposure to both arsenic and cadmium and hypermethylation of large chromosomal regions [144]. Over the course of cadmium-induced transformation, global DNA methylation and activity of DNMTs increased, along with DNMT1 and DNMT3a expression. Moreover, four key genes related to DNA repair, hMSH2, ERCC1, XRCC1, and hOGG1, were hypermethylated at the promoter and suppressed at the mRNA and protein levels in cadmium-transformed cells compared to controls. Tying these findings together, DNA demethylation treatment partially restored mRNA and protein expression of the DNA repair genes, brought down elevated DNMTs levels and suppressed the growth of tumor-derived cells and of cells passaged 35 times in the presence of cadmium chloride. Hypermethylation and repression of DNA repair genes, therefore, appear to be an early signature for cadmium-induced cancer and may also constitute part of the mechanism by which the toxicant induces tumorigenesis [137]. Another study has shown that following acute exposure (1 week) to cadmium, TRL 1215 rat liver cells exhibited decreased DNA methyltransferase activity and global DNA hypomethylation, whereas extended exposure (10 week) resulted in increased DNA methyltransferase activity and DNA hypermethylation [138]. A study using human prostate cells was in line with these results by showing that a 10-week exposure to cadmium led to malignant transformation in association with DNA hypermethylation at the global level, overexpression of DNMT3b and increased DNMT activity, promoter hypermethylation and decreased expression of the tumor suppressor genes RASSF1A and p16INK4A [146]. Recently, it has been shown that the level of genomic DNA methylation markedly increased in long-term cadmium-treated human B lymphoblats [147]. The methylation status of p16INK4A in the cadmium-exposed cells was significantly increased compared with control cells. Moreover the expression of p16 INK4A protein in the cadmium-treated cells was consistent with p16INK4A mRNA expression both in the presence and in the absence of 5-aza-dC treatment. These data provided evidence that DNA hypermethylation of p16INK4A gene contributes to cadmium-enhanced cell proliferation of human lymphocytes and spurred further investigations about the epigenetic mechanism for cadmium-induced carcinogenesis. However, another research group reported that cadmium-induced global DNA hypomethylation led to increased cell proliferation in the chronic myelogenous leukemia K562 cell line [148]. Recently, a causal association was demonstrated between caspase 8 promoter hypermethylation, pathological changes in the liver of rats and mice exposed to cadmium, and loss of the apoptotic pathway [149]. The functional loss of caspase 8, an apoptosis-related cysteine protease [150], was previously attributed to DNA hypermethylation in approximately 61% of human neuroblastoma specimens [151]. In addition, caspase 8 promoter was aberrantly methylated in all liver tissues from patients with hepatic cell carcinoma (HCC) while it was unmethylated in all the normal liver tissues. These data suggest that the hypermethylation-mediated inactivation of caspase 8 contributes to HCC by regulation of apoptotic cell death [152]. Moreover, it has been shown that the hypermethylation of caspase 8 promoter is a common feature of relapsed glioblastoma multiforme [153]. Although some studies have indicated that acute exposure to cadmium alone induces apoptotic cell death through different cell death pathways [154, 101], other in vitro studies showed that cadmium's inhibition of apoptotic cell death induced by other stressors, such as chromium, occurs through caspase-3 repression [155]. Therefore, it has been considered that the inhibition of apoptosis may be the main feature of cadmium carcinogenic mechanism. The relationships between altered activity of DNA methyltransferases and cadmium-induced cell proliferation needs to be better investigated.

Histone Tail Modification

As far as we know, very few studies have shown posttranslational modifications of N-terminal tails of histones as a result of cadmium poisoning. Acute exposure of non-transformed human urothelial cells (UROtsa) to both arsenicum and cadmium markedly increased the expression of enolase (ENO) 2, a useful biomarker for identifying tumoral neuroendocrine differentiation. It has been demonstrated that the expression of ENO2 mRNA was markedly elevated following treatment with the histone deacetlyase inhibitor MS-275 or in cadmium-and arsenicum-transformed UROtsa cells and tumour transplants [156]. A similar mechanism has been proposed for MT-3 gene expression, but in reverse. MT-3 gene is silent either in cadmium or arsenicum-transformed cell lines derived from UROtsa cells [157] through a mechanism that implies histone changes [158]. After treatment with MS-275, higher expression of MT-3 mRNA was found in the cadmium and arsenicum-transformed cell lines as compared with the parental UROtsa cells. Further studies are recommended to explore whether cadmium can alter gene expression by inducing histone changes at a global or gene-specific level.

Modulation of miRNA Expression

miRNAs are small, non-coding RNAs transcripts of ~21–23 nucleotides in length that regulate gene expression at the post-transcriptional level through complementary binding of 3′ or 5′ untranslated regions (3′ or 5′ UTR) of mRNA. miRNAs target the transcripts for degradation or to prevent translation, resulting in decreased target protein expression, though it has been reported that miRNAs may activate translation [159, 160]. miRNAs may also regulate transcription by directly binding to gene promoter sequences and inducing chromatin remodeling [161]. miRNA sequences are located in specifically associated promoters, or even embedded in host genes [160]. After transcription miRNAs are processed in the nucleus to yield hairpin structures known as pre-miRNA. Pre-miRNAs are then exported into the cytoplasm where they undergo further processing into mature miRNAs. Mature miRNAs involved in gene silencing are incorporated into RNA-induced silencing complexes (RISC) [162]. At present, over 17,000 different miRNAs sequences have been identified in approximately 140 species [163]. Aberrant expression of miRNAs has been associated with several human disease states, including cancer, cardiovascular disease, and genetic disorders [164]. A number of studies provide hints about the involvement of miRNAs in the toxic mechanism induced by cadmium. Cadmium has been shown to alter miRNA levels in vitro. In the human hepatocellular carcinoma cell line HepG2, a useful in vitro model for metal toxicity studies, miRNA expression was analyzed after cadmium exposure [165]. A large percentage of modified microRNAs belonged to the let-7 family, which is reported to have oncosuppressor functions, as evidenced by its ability to regulate DNA repair and cell division [166]. Another potential tumor suppressor that might be down-modulated by cadmium is mir-15b, an element that can target Bcl-2, thereby regulating apoptosis [167]. Several pathways connected to cancer were regulated at the transcriptional level, and miRNAs had a potential impact on the modulation of this regulation. At the low cadmium concentration (2 µM) only eleven genes belonging to the metallothionein family were regulated. At the higher concentration (10 µM) the pathway enrichment analysis for the 536 up-regulated genes showed a large number of pathways related to cancer, whereas the 424 down-regulated genes were enriched on pathways correlated to liver function. Focal adhesion (hsa04510) and the MAPK signaling pathway (hsa04010) were the two top pathways in the analysis of enrichment in KEGG pathways of the downregulated miRNAs targets. Focal adhesion is a fundamental mechanism regulating a number of crucial cell functions, such as gene expression, proliferation, differentiation, survival and motility. Metastasis and local invasion in cancer cells are facilitated by motility events requiring the regulation of focal adhesion and the deregulation of the actin cytoskeleton [168, 169]. Cell proliferation and migration are associated with increased signaling through the MAPK cascade, a highly conserved activation pathway [170]. In the same hepatoma cell line, very recently, Urani and his colleagues [171] showed that mir-372 and mir-138, both connected to carcinogenesis [172, 173], were upregulated after exposure to 10 µM cadmium. Cell cycle regulation can be affected by mir-372 that binds to the 3’UTR of p21Cip1/WAF-1a, a downstream protein of the p53 network [174]. Accordingly, mir-372 can promote cell-cycle progression and cell proliferation [173]. Oncosuppression was recently linked with increased mir-138 expression and, conversely, its downregulation was associated with the development of head and neck squamous cell carcinoma (HNSCC). However, although miR-138 downregulation is associated with HNSCC and other tumors, the exact contribution of miR-138 in oncogenesis and the mechanisms underlying this effect remain unclear [175].

In studying the role of airway pollutants in cystic fibrosis, Hassan et al. showed that cadmium reduced the expression of the cystic fibrosis transmembrane conductance regulator (CFTR) in human bronchial epithelial cells through upregulation of miR-101 and miR-144 [176]. Notably, miR-101 might promote inflammation by downregulating the expression of MAPK phosphatase-1 (MKP-1). Indeed, this dual specific phosphatase inactivates MAPKs, and thereby acts as an important negative regulator of innate immunity responses [177, 178]. Furthermore, miR-144 targets the hedgehog (Hh) signaling pathway that plays an important role in glioblastoma development [179], as well as the isocitrate dehydrogenase 2 (IDH2) pathway, that has been shown to contribute to atherosclerosis progression. Indeed, the identification of the latter pathway, which involves 7-ketocholesterol (7-KC) in addition to miR-144, has provided new insight into 7-KC function and microRNA biology in cardiovascular disease [180]. Moreover, miR-144 expression levels are inversely related with cell proliferation. For example, bladder cancer cell lines or cancer tissue samples display decreased miR-144 levels. In addition, blocking the expression of endogenous miR-144, promotes cell proliferation. Conversely, miR-144 overexpression results in blockade of cell proliferation, whereas miR-144 downregulation mitigates miR-144-mediated repression of zeste homolog 2 (EZH2) and subsequent cell proliferation [181].

Therefore, it can be speculated that cadmium is implicated in a lot of disparate disease states through alterations of miRNA levels and that such alterations may be useful in the risk characterization steps.

CADMIUM AND MELANOMA

Melanoma incidence has been on the rise over the last years. Among the risk factors identified so far for the development and the progression of the disease are some trace elements of nutritional and toxicological interest. Data obtained from trace element determination in toenails of 58 cases and 58 controls from the Modena province of Italy showed higher levels of copper and lower concentrations of iron in patients with melanoma of the skin, whereas no differences were observed for cadmium. Therefore, the authors excluded an involvement of cadmium exposure in this disease [182]. However, an involvement of cadmium in the pathogenesis of skin melanoma cannot be excluded, as little is known on cadmium absorption through the skin. In 1991, Wester et al. showed a higher adsorption of cadmium from contaminated soil and water into the skin rather than in plasma [183]. Importantly, rats (shaved skin) subjected to percutaneous administration of cadmium chloride solutions showed dermal hyperkeratosis and acanthosis with occasional ulceration, increased mitotic index of the skin cells and elevated cadmium levels in blood, liver and kidney [184]. Cadmium may be absorbed by the skin through its binding to sulfhydryl radicals of cysteine in epidermal keratins, or its complexing with metallothionein [185]. The analysis of mouse B16 melanoma cell lines with high and low constitutive MT expression revealed that cadmium resistance was associated with constitutive MT accumulation in the absence of heavy-metal induction [186]. Resistance to anticancer drugs and radiotherapy, and thereby a poor prognosis, have been associated with MT overexpression in a variety of cancers, according to a number of recent studies. For example, a tendency towards disease progression is associated with MT overexpression in primary melanoma. This feature is useful to identify thin melanomas (< 1.5 mm) carrying a high risk of progression, since MT expression is independent from Breslow thickness [187]. It has been known for many years that MT expression in malignant melanoma is significantly associated with progressive disease and might therefore be a useful prognostic indicator, as shown in paraffin-embedded tissues from 63 cases of malignant melanoma and 13 secondary deposits [188]. An association has been found between exposure to cadmium and the development of uveal melanoma [189]. Moreover, in uveal melanoma cultures, an increased resistance to cadmium-induced cytotoxicity characterizes the vasculogenic mimicry (VM)-forming tumor cells. VM patterns are present in a wide variety of malignant tumours and are associated with a poor prognosis, since they reflect a marked ability of tumour cells to form perfusion pathways [190-194]. C918 uveal melanoma cells have been studied in both traditional two-dimensional (2D) and extracellular matrix (ECM, Matrigel)-containing 3D cultures. While growing as a monolayer in 2D cultures, C918 cells formed complex patterns in 3D cultures, involving morphologically distinct cell subpopulations. For example, monolayers were formed both on Matrigel surfaces and on the bottom of the culture dish, whereas VM patterns were produced in other areas. At 4 to 5 days after the beginning of culture, 2D and 3D cultures were exposed to various concentrations of cadmium chloride, ranging from 20 to 1000 μM [195] and observed daily for toxicity. Under these conditions, cadmium-exposed cells showed dose-dependent toxicity, while control cultures remained viable. Interestingly, however, VM-forming tumour cells showed increased survival following cadmium exposure relative to other subpopulations in 3D cultures or cells cultured in 2D. Moreover, when exposure to cadmium was stopped after initiation of cisplatin treatment, residual viable cells in VM patterns could function as foci for new growth in 3D cultures [196]. These data suggest that VM-forming cells in uveal melanoma cultures are less susceptible to cisplatin or cadmium toxicity. Moreover, such increased resistance may represent a mechanism whereby VM-forming melanoma cells might contribute to a poor prognosis [196]. In addition, several years ago it was well documented that the mouse melanoma cell invasion of organ samples obtained from syngeneic mice who had been administered heavy metals, including cadmium, was enhanced through the induction of MTs. Therefore, it has been suggested that heavy metal-induced MTs serve as host-derived factors in malignant disease and are strictly associated with metastasis [197]. Thus from a review of the literature, a picture emerges of conflicting results on the involvement of cadmium in melanoma. Some studies could not find any role of the metal in the pathogenesis of this tumor, while increased resistance to cadmium was associated with some features of malignant melanomas. Indeed, more attention should perhaps be paid to the effects of non-toxic doses of cadmium on normal and transformed melanocytes. Indeed, this issue has never been studied in depth, despite cadmium was shown to be comutagenic with UV, the leading cause of the onset of melanoma. For example, as previously indicated, cadmium interferes with nucleotide excision repair by interfering with the removal of thymine dimers after UV irradiation [125]. However, to the best of our knowledge, the possible role of cadmium in inducing epigenetic modifications of key genes in the regulation of proliferation and apoptosis in transformed melanocytes has not been studied. As mentioned, investigations of this type could be of importance, in view of the recognized role of cadmium as a cancer promoter in combination with ultraviolet radiations. Moreover, patterns of epigenetic modifications could serve as markers of gene activity and expression [198-200]. To get initial insights into this issue, we evaluated epigenetic effects on genes involved in apoptosis and proliferation after acute cadmium exposure. In other transformed cell systems, inhibition of the expression of tumor suppressor genes responsible for the regulation of the cell cycle, such as p16INK4A, and of apoptosis, such as caspase 8, was demonstrated following exposure to cadmium. Similarly, we could document that short-term cadmium exposure had a critical impact on the behavior of melanoma cells through epigenetic modifications of p16INK4A and caspase 8, resulting in their altered expression and in aberrant cell proliferation and apoptosis. Indeed, after a short cadmium exposure we observed an increase in cell population in either cutaneous or uveal melanoma However, the mechanisms appeared to be quite different. In skin melanoma the extrinsic pathway of apoptosis was affected and caspase 8 hypermethylation and inactivation was indicated as the main pathogenic mechanism. In uveal melanoma, instead, p16INK4A aberrant methylation and silencing was called as the determinant cause [201]. It is worth of notice that in some cell lines the effects were more pronounced after 48 hours of exposure to cadmium than those observed after 72 hours of treatment, even when using low concentrations of the metal. The other genes investigated did not undergo any modification of their expression following exposure to cadmium regardless of whether they were expressed or not. Our results show that cadmium-treated uveal melanoma cell lines have >80% methylated clones (i.e., severe hypermethylation and complete methylation) in the p16 INK4A promoter versus >60% unmethylated clones (i.e., severe hypomethylation and complete unmethylation) in the same gene region of the corresponding untreated cells. Clones of cutaneous melanoma cell lines with <2 0% methylation of the caspase 8 promoter (i.e., completely unmethylated and mildly hypermethylated) were found in control untreated cells, whereas clones with >70% methylation (i.e., moderately–to-severely hypermethylated) occurred in the same cell lines exposed to cadmium. Integrative analysis of DNA methylation and RNA expression data revealed a positive association between DNA methylation and transcript expression of p16INK4A and caspase 8 either before or after treatment with cadmium. Inquiring about the role of DNMTs evidenced a prevalent role for DNMT3a and DNMT1 in the methylation of, respectively, caspase 8 and p16INK4A while DNMT3b seemed to have much less impact. These results are in agreement with those of others, highlighting that during malignant transformation cadmium is able to alter total DNMTs activities and to progressively increase global DNA methylation [137, 138, 142]. The methylation profile of the other genes that we have studied revealed generally hypomethylated or mildly hypomethylated clones with no significant difference between the untreated and cadmium-treated cells. The silencing of p16INK4A and caspase 8 by cadmium in melanoma cells is of particular importance in light of the negligible role played by constitutional epimutations of CDKN2A (p16INK4A and p14ARF) genes in melanoma families, whereby it was concluded that it is unlikely that they play a role in susceptibility to the disease in families without CDKN2A mutations [202]. These findings are coherent with those of others which aimed at identifying germline epimutations of genes that were found to be mutated in familial cutaneous malignant melanoma. No association was found between the risk of cutaneous malignant melanoma and peripheral blood mononuclear cell methylation levels of p16INK4A [203]. These data confirm the ones presented in this study, and highlight a crucial role of the environment in inducing epigenetic changes in a key gene in melanoma development, such as p16INK4A, in absence of inherited mutations. The relevance of the occurrence of p16INK4A promoter methylation in melanoma was well established by a very recent work which has shown that hypermethylation of the p16INK4A promoter was greater in the vertical growth-phase (60%) melanomas than in the radial (40%, P = 0.063) melanomas and in those displaying epidermal involvement (P = 0.046). Notably, p16INK4A methylation well correlated with increased melanoma thickness and marginally with increased Clark level. Low (1-30%) p16INK4A expression was detected in the majority (19 of 54 or 35%)) of melanoma cases [204]. And more, loss of CDKN2A expression was shown to frequently occur in primary invasive melanoma [205] and absent p16INK4A expression was associated with adverse prognostic markers of the cancer [206]. For these reasons, p16INK4A immunohistochemical expression is believed to be an independent prognostic biomarker of potential value in routine melanoma diagnostic practice and was found to be useful in confirming the benign nature of mildly and moderately atypical cutaneous blue nevi (CBN) compared to severely atypical CBN and melanomas [207]. Absence or downregulation of caspase 8 is a frequent event in cancer and is correlated with an unfavourable outcome [208, 209], since it may cause resistance to apoptosis [146, 210]. Several studies have demonstrated that expression of caspase 8 may be controlled by epigenetic modifications, which differ according to the type of cancer cells [211-213]. Hypermethylation of the promoter region has been recognized as the most recurrent mechanism causing functional loss of caspase 8 gene. However, although an association between aberrant DNA methylation of caspase 8 and carcinogenesis has been demonstrated in several studies [214, 215], the functional relevance of this epigenetic trait in melanomagenesis has not yet been fully understood [216]. Our findings, showing for the first time that caspase 8 expression is impaired by DNA hypermethylation following cadmium treatment in cutaneous melanoma cell lines, highlight the potential relevance of such a mechanism in the context of this tumor and extend earlier in vitro and in vivo observations linking exposure to cadmium with decreased expression of caspase 8 in tumours different from melanoma [217, 218]. The pivotal role of caspase 8 overexpression in reactivating the apoptosis program in cutaneous melanoma cells exposed to cadmium with hypermethylated caspase 8 promoter suggests that cadmium might act as a tumor promoter for melanoma development. Based on evidence that both the programmed re-expression of caspase 8 by demethylation and caspase 8 gene transfer can sensitize tumour cell lines for drug-induced apoptosis [216, 219, 220], it is possible to hypothesize that this pollutant may prevent drug-induced apoptosis and interfere with therapeutic strategies targeting cutaneous melanoma. Therefore, these findings may be of interest because they: i) highlight the crucial role of acute cadmium exposure in silencing key genes involved in the control of cell cycle and apoptosis in melanoma; ii) discriminate between cutaneous and uveal melanoma on the basis of the genes epigenetically modified by cadmium; iii) indicate cadmium pollution as one of the potential causes of inactivation of p16INK4A, a previously recognized biomarker of malignant melanoma; iv) show for the first time the significance of caspase 8 methylation and silencing in cutaneous melanoma and the role of cadmium in this effect. In conclusion, our data suggest that aberrant DNA methylation of p16INK4A and caspase 8 gene promoters is associated with single-phenotype defects in melanoma cells after acute exposure to low non cytotoxic doses of cadmium. Further studies are warranted to address the role of epigenetic mechanisms induced by environmental pollutants in worsening the melanoma progression.

CONCLUSION

Cadmium is a heavy metal, widely distributed in the environment, that affects human health in a multiple ways. In mammals it exerts toxic effects on several organs, including kidney, liver, stomach, respiratory tract, brain, and bones, although the underlying mechanisms are far from being elucidated.

Cadmium was classified as a human carcinogen by the International Agency for Research on Cancer, since many studies have linked exposure to cadmium with pulmonary, prostatic, and renal cancer in humans and some studies have associated cadmium exposure with human cancer of the liver, hematopoietic system, urinary bladder and stomach. This review emphasizes the importance for applying the study of cadmium carcinogenesis to melanomagenesis, as there is some indication that cadmium might be important in melanoma despite the limited number of studies on the subject.

Relying on available evidence, cadmium apparently has little direct genotoxic activity and acts apparently through various indirect mechanisms, including ROS generation, DNA repair inhibition, impairment of the cellular antioxidant defence system, interference with signal transduction, and possibly the disruption of cell–cell adhesion with the consequent initiation of cancer.

The evidence described in this review highlights the importance of considering epigenetics as a possible mechanism underlying the toxicity and cell-transforming ability of this metal by demonstrating that exposure to cadmium perturbs DNA methylation levels, global and gene-specific histone posttranslational modification markers, and miRNA expression levels. As most of the epigenetic changes are reversible, insights into this topic might provide novel therapeutic strategies based on molecules that modify the activities of epigenetic enzymes, such as DNMTs and HDACs, and/or affect miRNA expression. Thus, epigenome modification might be used to restore normal gene transcription levels. Results presented here indicate that the effects of cadmium on melanomagenesis may be more important than previously thought. Moreover, they highlight the role of epigenetics in melanoma progression following exposure to cadmium. It is speculated that in cutaneous melanoma cadmium affects the extrinsic apoptotic pathway, that is prevalent in this kind of tumor, through methylation and inactivation of caspase 8, while in uveal melanoma the metal alters the cell cycle through methylation and silencing of p16INK4A. It’s possible that in a near future well characterized epigenetic modifications will aid in the molecular diagnosis and treatment of a variety of cancers and other disorders induced by cadmium.