Mechanisms of the synergistic lung tumorigenic effect of arsenic and benzo(a)pyrene combined- exposure

Abstract

Humans are often exposed to mixtures of environmental pollutants especially environmental chemical carcinogens, representing a significant environmental health issue. However, our understanding on the carcinogenic effects and mechanisms of environmental carcinogen mixture exposures is limited and mostly relies on the findings from studying individual chemical carcinogens. Both arsenic and benzo(a)pyrene (BaP) are among the most common environmental carcinogens causing lung cancer and other types of cancer in humans. Millions of people are exposed to arsenic via consuming arsenic-contaminated drinking water and even more people are exposed to BaP via cigarette smoking and consuming BaP-contaminated food. Thus arsenic and BaP combined-exposure in humans is common. Previous epidemiology studies indicated that arsenic-exposed people who were cigarette smokers had significantly higher lung cancer risk than those who were non-smokers. Since BaP is one of the major carcinogens in cigarette smoke, it has been speculated that arsenic and BaP combined-exposure may play important roles in the increased lung cancer risk observed in arsenic-exposed cigarette smokers. In this review, we summarize important findings and inconsistencies about the co-carcinogenic effects and underlying mechanisms of arsenic and BaP combined-exposure and propose new areas for future studies. A clear understanding on the mechanism of co-carcinogenic effects of arsenic and BaP combined exposure may identify novel targets to more efficiently treat and prevent lung cancer resulting from arsenic and BaP combined-exposure.

1. Introduction

Current practices in determining whether an environmental pollutant is carcinogenic or in assessing the carcinogenic potency of a chemical pollutant remain largely rely on the findings from studying the individual pollutant although humans are often exposed to a mixture of environmental pollutants. This represents a significant knowledge gap in determining the carcinogenicity of environmental pollutants and the underlying mechanisms. Indeed, studies showed that toxic effects including the carcinogenic effect of exposure to a mixture of environmental pollutants or combined exposure to different pollutants could be significantly different from that of exposure to individual pollutants alone [1-5]. However, the underlying mechanisms of how pollutant mixture-exposure or combined-exposure act significantly different from the single chemical exposure alone have been poorly understood.

Arsenic occurs naturally in rocks, soil, and water and is one of the most common environmental and occupational pollutants. While occupational arsenic exposure is mostly through inhalation, general population arsenic exposure happens mainly through consuming arsenic-contaminated drinking water [6]. Epidemiological and laboratory studies showed that chronic arsenic exposure via inhalation or through contaminated-drinking water increases the risk of skin, lung, bladder, liver, and prostate cancer [7-14]. However, the mechanism of arsenic carcinogenesis has not been clearly defined. It has been proposed that arsenic may promote carcinogenesis by inhibiting DNA repair [15,16], inducing oxidative stress [17,18], dysregulating epigenetics [19-21], altering cell signaling [22,23] or acting as a co-carcinogen [24,25].

Benzo(a)pyrene (BaP) is a member of the polycyclic aromatic hydrocarbon (PAH) family and BaP is produced usually when organic matters are incompletely burned. For example, cigarette smoke and well-done barbecued-meat contain high levels of BaP and are common sources of general population BaP exposure [26,27]. Although BaP is recognized as one of the most common environmental pollutants and carcinogens causing lung cancer [28], the mechanism of BaP carcinogenesis remains elusive. It has been proposed that BaP metabolic activation via cytochrome P450 (CYP) CYP1A1 and 1B1 and subsequent DNA adduct formation by its key metabolite benzo[a]pyrene-7,8-diol-9,10-epoxide (BPDE) plays an important role in BaP carcinogenesis [29,30].

Millions of people are exposed to arsenic through contaminated-drinking water. High levels of BaP are found in cigarette smoke and cooked-meat. It is envisioned that arsenic and BaP co-exposure could be common in humans. Moreover, epidemiology studies showed that arsenic-exposed people who were cigarette smokers had a significantly higher lung cancer risk than those who were non-smokers, revealing that arsenic exposure and cigarette smoking act synergistically with more than an additive effect on the risk of lung cancer [11,14,31]. Since BaP is one of the major carcinogens in cigarette smoke, it was speculated that the synergistic effect of arsenic exposure and cigarette smoking on lung cancer risk may be due to the combined effect of arsenic and BaP co-exposure. Together, these lay solid foundation for studying the combined-lung carcinogenic effect of arsenic and BaP co-exposure. The aim of this review is to summarize important findings from studies on the combined-lung tumorigenic effects and the underlying mechanisms of arsenic and BaP co-exposure and propose new directions for future studies.

2. The synergistic lung tumorigenic effects of arsenic and BaP combined-exposure

Although arsenic and BaP combined-exposure is likely common in humans and may increase human lung cancer risk, very few animal studies have been done to investigate the combined-lung carcinogenic effect of arsenic and BaP c-exposure.

2.1. Arsenic and BaP combined-exposure via respiration

The first co-carcinogenicity animal study on arsenic and BaP co-exposure was published in 1977 by Ishinishi N et al., who aimed to investigate the carcinogenic or co-carcinogenic properties of arsenic-containing copper ore and flue dust collected from a metal refinery and arsenic trioxide (As₂O₃) with BaP [32]. To mimic occupational arsenic exposure, the arsenic containing-flue dust (2 mg), arsenic trioxide (0.2 mg) and or BaP (0.4 mg) suspended in saline were given to 10-week old male Wistar-King albino rats via intratracheal instillation once a week for 15 weeks. At the end of exposure, all surviving rats were observed during their entire life span and allowed to die spontaneously. It was found that the incidence rate of lung squamous cell carcinoma in rats exposed to arsenic trioxide plus BaP was higher (3 out of 7 rats, 42.9%) than that in rats given to arsenic trioxide (0 out of 8 rats, 0%) or BaP (1 out of 7 rats, 14.3%) alone. Although the numbers of rats in each group were too small to yield significant differences, this study provided the first preliminary evidence suggesting that arsenic plus BaP co-exposure via inhalation may synergize in inducing lung cancer.

Pershagen et al. further investigated the carcinogenicity of arsenic trioxide and its co-carcinogenicity with BaP using male Syrian golden hamsters [33]. Similarly, arsenic trioxide (3 mg/kg body weight) and/or BaP (6 mg/kg body weight) suspended in saline solution containing 2 mM sulfuric acid were given to hamsters via intratracheal instillation once a week for 15 weeks. At the end of exposure, all surviving hamsters were observed for additional 125 weeks. It was found that carcinomas of the larynx, trachea, bronchus, or lung were found in 3, 17, and 25 of the 47, 40, and 54 animals examined in the arsenic, BaP, and arsenic plus BaP groups, respectively. No malignant lung tumors were observed in 53 animals in the control group. No significant difference in the occurrence of carcinomas between the BaP exposure alone and arsenic plus BaP co-exposure group was observed. However, if adenomatous tumors were counted and included (i.e., adenomas plus adenocarcinomas and adenosquamous carcinomas), the total lung tumor incidence in the arsenic plus BaP co-exposure group was significantly higher than that in BaP exposure alone group (p = 0.02). The findings from this study showed additional evidence supporting that arsenic and BaP co-exposure via inhalation may synergize in inducing lung tumorigenesis, which provided a potential mechanistic explanation for the synergism of lung cancer between occupational arsenic exposure and smoking in smelter workers.

2.2. Arsenic and BaP combined-exposure via oral ingestion

The arsenic used in above two arsenic plus BaP inhalation exposure animal studies was arsenic trioxide (As₂O₃), which is the most common arsenic compound in arsenic occupational exposure via inhalation in smelters. However, the most common route of general population arsenic exposure is through consuming arsenic-contaminated drinking water and one of the common arsenic compounds in arsenic-contaminated drinking water is sodium arsenite (NaAsO₂). Moreover, one common route of BaP exposure in general population non-smokers is via consuming BaP-contaminated food. Indeed, Hattemer-Frey and Travis reported that the food chain is the dominant pathway of human BaP exposure, accounting for about 97% of the total daily intake of BaP [34]. However, the co-carcinogenic effect of arsenic and BaP combined-exposure via oral ingestion is unknown.

To address this important question, we recently performed arsenic and BaP combined-exposure study using A/J mice [2]. Following Dr. Michael P Waalkes Group’s arsenic drinking water gestational and “whole-life” exposure protocols that significantly increase tumor incidences in lungs and other organs in offspring mice [35,36], our arsenic exposure in drinking water (sodium arsenite, 20 ppm) in pregnant A/J mice started from gestation day 18 [2]. After birth, the dams in arsenic exposure group continuously received the same arsenic water throughout lactation. At weaning (3 weeks of age), male offspring from maternal mice received arsenic water were randomly divided into two groups: arsenic exposure alone group and arsenic plus BaP combined-exposure group with 10 mice in each group. These two groups of mice continuously received the same arsenic water for 34 weeks. Our mouse BaP exposure followed Dr. Stephen S. Hecht Group’s BaP oral exposure protocol, a well-established and widely-used BaP oral exposure protocol for mouse lung tumorigenesis study [37]. Briefly, male offspring in BaP exposure alone group and arsenic plus BaP combined-exposure group were given BaP via oral gavage (3 μmol of BaP in 0.1 ml of corn oil) once a week starting from 3 weeks of age for 8 weeks. Proper control groups were included and all mice were sacrificed 34 weeks after the first BaP exposure following the protocol from the previous BaP oral exposure lung tumorigenesis study [37].

It was found that arsenic exposure alone did not cause any lung tumors in mice, while all BaP exposure alone mice developed lung tumor (adenomas only) [2]. No lung tumors were observed in control group mice. In contrast, mice exposed to arsenic plus BaP developed significantly more (tumor multiplicity) and larger (tumor burden) lung tumors. Moreover, three out of ten mice in arsenic plus BaP combined-exposure group also developed lung adenocarcinomas, whereas no adenocarcinoma was observed in BaP exposure alone mice. Together, our study revealed that arsenic and BaP combined-exposure via oral ingestion synergizes in inducing lung tumorigenesis, indicating that arsenic exposure through drinking water with BaP co-exposure via oral ingestion could significantly increase lung cancer risk in general population. Since BaP is one of major carcinogens in cigarette smoke, our findings also provide important mechanistic insight for early epidemiology observations showing that arsenic-exposed (via drinking water) people who were cigarette smokers had a significantly higher lung cancer risk than those who were non-smokers: it could be due to the synergistic lung tumorigenic effect of arsenic and BaP combined co-exposure.

3. Mechanisms of the synergistic lung tumorigenic effect of arsenic and BaP combined-exposure

Although animal studies provided substantial evidence supporting the synergistic lung tumorigenic effect of arsenic and BaP combined-exposure, the underlying mechanism has been poorly understood. A good number of in vitro and in vivo studies were performed to explore the mechanism of actions of arsenic and BaP combined-exposure, but controversial and inconsistent findings were reported in several proposed-mechanisms.

3.1. Genotoxic mechanisms

One of the proposed-mechanisms for arsenic carcinogenesis is its inhibitory effect on DNA repair to enhance DNA damage [15,16]. In addition, it is generally accepted that the carcinogenicity of BaP largely depends on its metabolic activation mainly by CYP1A1 and 1B1 and subsequent DNA adduct formation (BPDE-DNA adduct) by its key metabolite BPDE [29,30]. The mechanistic studies on the enhanced-carcinogenicity of arsenic and BaP combined-exposure have thus largely focused on the effects of arsenic on BaP-induced genotoxicity. However, the reported results were inconsistent and sometimes controversial.

Hartmann and Speit reported that arsenic (NaAsO₂, 80-100 μM) 2 h post-treatment significantly increases the persistence of 2 h BaP (10-50 μM) treatment-induced DNA lesions determined by single cell gel electrophoresis test (comet assay) in human white blood cells and lung fibroblasts [38]. This provided the first evidence suggesting that arsenic may inhibit the repair of DNA damage induced by BaP. This finding was supported by another study showing that co-injection of arsenic (NaAsO₂, 10 mg/kg) and BaP (800 μg) significantly extended the persistence of total DNA adduct levels induced by BaP in rat mammary glands [39]. Dr. Alvaro Puga’s group reported that arsenic (NaAsO₂, 10 μM, 30 min pretreatment) combined-exposure potentiates BaP (0.1, 0.5, 1.0 μM, 24 h treatment) genotoxicity as evidenced by significantly increasing BPDE-DNA adduct levels in mouse hepatoma Hepa-1 cells [40]. In contrast, Dr. Puga’s group found that arsenic treatment did not change BPDE-DNA adduct removal kinetics in BaP-treated cells, suggesting that arsenic treatment did not have a significant effect on the repair of DNA damage induced by BaP [40]. The lack of effect of arsenic on the repair of BPDE-DNA adduct was also reported in another study by transfecting BPDE-modified plasmid DNA into human lung cells, although the same arsenic (NaAsO₂, 3-50 μM, 72 h) treatment did enhance BPDE (3.75 μM)-induced mutagenesis in the same cells [41]. Moreover, Dr. Puga’s group also found that arsenic treatment did not alter BaP-inducible CYP1A1 enzymatic activity, suggesting that arsenic treatment may not have a significant effect on BaP metabolic activation [40]. This is in agreement with another study showing that arsenic (NaAsO₂) pretreatment did not affect BaP-mediated induction of CYP1A1 in human lung adenocarcinoma CL3 cells [42]. These are inconsistent with other studies showing that arsenic exposure alone is capable of increasing CYP1A1 expression level and activity in human cells and mouse lung tissues [43-45]. The mechanisms responsible for these inconsistent observations are unknown. Even more controversially, some studies showed that arsenic (NaAsO₂, 1-10 μM, or AS₂O₃, 1-5 μM) and BaP (2.5-25 μM) co-exposure (12-24 h) significantly decreased BaP-induced CYP1A1, CYP1A2 and CYP1B1 levels or activities in human Hepatocytes, liver cancer HepG2 and Hep3B cells and breast cancer T-47D cells [46-49]. Similarly, it was found that arsenic (NaAsO₂, 0.4 and 8 μM) and BaP (0.4 μM) co-exposure for 7 days also significantly decreased CYP1A enzyme activity (approximately 3-fold) in zebrafish when compared to BaP alone exposure [50].

However, one study from Dr. Scott W. Burchie’s group reported that environmentally relevant concentrations of arsenic (NaAsO₂, 5 nM) and the BaP metabolite (BaP-diol, 100 nM) co-exposure (18 h) significantly increased mRNA levels of CYP1A1 and CYP1B1 in primary thymus cells isolated from 8-10 week old C57BL/6J male mice [51]. Moreover, an increased genotoxicity determined using the comet assay was observed in the primary thymus cells co-treated with 5 nM NaAsO₂ +100 nM BaP-diol and 50 nM NaAsO₂ + 100 nM BPDE for 18 h. The authors also tested the hypothesis that the increase in DNA damage is due to the inhibition of Poly(ADP-ribose) polymerase (PARP) (an initiator of base excision repair of DNA damage) by arsenic and BaP metabolites co-exposure leading to decreased DNA repair in the co-exposed cells. Indeed, the arsenic and BaP metabolites co-exposure did significantly reduce the PARP activity in the thymus cells. Furthermore, co-treatment of a PARP inhibitor with BaP-diol caused an increase in DNA damage in thymus cells compared to BaP-diol treatment alone [51]. These findings suggest that the suppression of PARP activity and induction of CYP1A1/CYP1B1 may act together to increase genotoxicity resulting from arsenic and BaP metabolites co-exposure in thymus cells. This is in agreement with a recent study showing that arsenic 24 h post-treatment (NaAsO₂, 2 μM) significantly increased BPDE (1 μM) treatment (0.5 h)-caused DNA damage in a squamous carcinoma cell line SCC-7 cells determined by the comet assay [52]. Mechanistically, the authors proposed that BPDE treatment decreased arsenic methylation leading to cellular As⁺³ accumulation, which in turn inhibited nucleotide excision repair (NER) of the BPDE-DNA increasing DNA damage [52].

The effect of arsenic and BaP co-exposure on BaP genotoxicity has also been explored in animal studies. Five-week old female C57BL/6 mice were exposed to arsenic via drinking water (NaAsO₂, 2.1 ppm) for 13 days and/or BaP (200 nmol BaP/25 ml acetone) via skin topical application once per day during last 4 days of arsenic exposure [53]. It was found that arsenic co-exposure increased BaP-DNA adduct levels in both lung and skin tissues; and the increase in lung tissues was statistically significant (P = 0.038) [53]. However, Fischer et al. reported in another mouse study that BaP-DNA adduct levels were not higher in mice exposed to arsenic plus BaP compared to mice exposed to arsenic or BaP alone [54]. In Fischer’s study, FVB/N mice carrying the G11 PLAP transgene were crossed to C57B1/6 mice. Six-week old male hybrid mice were exposed to arsenic via drinking water (NaAsO₂, 10 ppm) for 10 weeks. Two weeks after starting arsenic exposure, half of the arsenic-exposed mice were also exposed to BaP once per day for 8 weeks by skin painting with 50 μl of acetone containing 100 nmol BaP [54]. While this arsenic and BaP combined-exposure did increase the skin mutagenicity as evidenced by producing more PLAP⁺ cells in mouse skin, the combined exposure group exhibited no more BaP-DNA adducts in skin tissues than the BaP exposure alone group. Moreover, the BaP-DNA adduct levels from the individual mice within each group did not correlate with the mutation data (number of PLAP⁺ cells). Whether this arsenic and BaP combined-exposure increased BaP-DNA adduct levels in mouse lung tissues was not determined. This finding is inconsistent with that from Evan’s animal study showing that arsenic co-exposure increased BaP-DNA adduct levels in both lung and skin tissues [53]. However, it is in agreement with our recent study showing that arsenic (NaAsO₂, 1 μM) and BaP (2.5 μM) co-exposure in immortalized human bronchial epithelial cells (BEAS-2B) for 30 weeks did not increase BPDE-DNA adduct levels although the co-exposure significantly increased cell malignant transformation and turnorigenicity [2].

The genotoxicity of arsenic and BaP combined-exposure was also investigated by assessing the chromosome damages in bone-marrow cells using the micronucleus test (MN) in mice. Six to seven-week old male C57BL/6J/Han mice were pre-exposed to arsenic (NaAsO₂, 50 ppm) in drinking water for 7 days and then exposed to a single dose of BaP (200 mg/kg body weight) by intra-peritoneal injection [55]. Control, arsenic or BaP exposure alone, and arsenic plus BaP combined exposure mice were sacrificed before or at 12, 24, 48 or 72 h after BaP exposure. In arsenic and combined-exposure mice, significantly higher levels of MN were detected in bone-marrow cells examined at 12, 24 and 48 h after BaP administration, compared with animals exposed to arsenic or BaP alone [55]. In contrast, Peng et al. reported that arsenic (NaAsO₂, 0.25-2.0 μM) and BaP (20 μM) co-exposure (24 h) did not significantly increase MN frequencies compared to BaP alone exposure in cultured human liver cancer HepG2 cells [56]. However, a recent study from the same group using the same cell line showed that arsenic (NaAsO₂, 10 μM) and BaP (2 μM) co-exposure (48-50 h) significantly increased MN formation in HepG2 cells determined using a flow cytometry-based MN test [57].

3.2. Epigenetic mechanisms

Epigenetics refers to heritable changes in the pattern of gene expression that are not caused by alterations in DNA sequences, but are mediated by DNA and RNA methylation, histone posttranslational modifications, and non-coding RNAs such as microRNAs (miRNAs) and long non-coding RNAs (lncRNAs). Since Drs. Andrew P. Feinberg and Bert Vogelstein reported the first epigenetic dysregulation evidence in human cancer tissues in 1983 [58], numerous studies have demonstrated that epigenetic dysregulations play important roles in cancer initiation and progression [59-61]. Indeed, some studies showed that epigenetic alterations are capable of largely, if not completely, replacing genomic instability to cause cancer development in the absence of extensive mutations. For example, McKenna ES et al. reported that epigenetic dysregulation due to disruption of a chromatin remodeling complex (SWI/SNF) causes cancer without genomic instability [62,63]. Moreover, Mack SC et al. found that epigenomic alterations define lethal CIMP-positive ependymomas of infancy, which exhibits a CpG island methylator phenotype but has extremely low mutation rate and zero significant recurrent somatic single nucleotide variants [64]. Collectively, these findings suggest that cancer may also be initiated by non-mutational mechanisms such as epigenetic mechanisms.

Given the controversial reports regarding the genotoxic effects of arsenic and BaP combined-exposure, it is essential to explore the other mechanism such as the epigenetic mechanism for the synergistic lung tumorigenic effect of arsenic and BaP combined-exposure. Interestingly, Vicente-Dueñas C. et al. recently proposed an epigenetic priming model for cancer initiation, emphasizing that “contribution of oncogenes to cancer development is mediated mainly through epigenetic priming of cancer-initiating cells” [65]. This new cancer initiation hypothesis argues that “stem or progenitor cells have a specific epigenetic program that renders them more susceptible to genetic alterations”; and “specific developmental stages with their associated epigenetic states may be more susceptible to oncogenes and progression to cancer” [65]. This new cancer initiation proposal provided additional rational to study the epigenetic mechanism of the co-carcinogenicity of arsenic and BaP combined-exposure.

Although either arsenic or BaP alone exposure has been shown to cause various epigenetic changes [19-21,66], very little is known about the epigenetic effects caused by arsenic and BaP combined-exposure. Our recent study provided the first evidence demonstrating that arsenic and BaP co-exposure synergizes in dysregulating epigenetics [2]. It was found that human bronchial epithelial cells (BEAS-2B) exposed to arsenic (NaAsO₂, 1 μM) plus BaP (2.5 μM) for 30 weeks did not have a higher level of BPDE-DNA adduct compared to BEAS-2B cells exposed to BaP alone for 30 weeks. In contrast, arsenic and BaP co-exposed cells exhibited a significantly higher level of histone H3 lysine 9 (H3K9) dimethylation (H3K9me2) compared to cells exposed to arsenic or BaP alone. Mechanistically, arsenic and BaP co-exposure synergizes in increasing H3K9me2 level by up-regulating the expression of a histone methyltransferase SUV39H1 via synergistically activating aryl hydrocarbon receptor (AhR) [2]. Moreover, the levels of SUV39H1 and H3K9me2 in mouse lung tumors resulting from arsenic and BaP combined-exposure were also significantly higher than that in mouse lung tumors caused by BaP exposure alone [2].

The significance of SUV39H1 up-regulation in the co-carcinogenicity of arsenic and BaP combined-exposure was assessed in our cell transformation experiments. First, knockdown of SUV39H1 in arsenic plus BaP co-exposure-transformed cells significantly reduced their H3K9me2 levels and the numbers of colonies formed in soft agar [2]. As the number of colonies formed in soft agar is widely used in cell transformation studies to assess the extent of cell transformation by a carcinogen exposure or an oncogene expression [67-70], these findings suggest that higher levels of SUV39H1 and H3K9me2 are essential in maintaining the transformed-phenotype of arsenic and BaP co-exposure-transformed cells. Second, whether up-regulation of SUV39H1 plays a role in cell transformation by arsenic and BaP co-exposure was also investigated. Using the same control shRNA and SUV39H1 shRNA knockdown vectors, we also generated shRNA control and SUV39H1 expression stable knock parental BEAS-2B cells. We re-performed cell transformation experiment by exposing control shRNA and SUV39H1 knockdown BEAS-2B cells to arsenic (NaAsO₂, 1 μM) plus BaP (2.5 μM) for 40 weeks. Strikingly, SUV39H1 knockdown greatly reduced the capability of arsenic plus BaP co-exposure to induce cell transformation as evidenced by decreasing soft agar colony formation by 46.5% compared to co-exposed-control shRNA cells. This finding suggests that SUV39H1 up-regulation contributes significantly and causally to arsenic plus BaP co-exposure-induced cell transformation. Since SUV39H1 up-regulation was also observed in arsenic and BaP combined exposure-caused mouse lung tumor tissues, these findings imply that SUV39H1 up-regulation and subsequent epigenetic dysregulations play important roles in arsenic and BaP co-exposure-induced synergistic lung tumorigenesis.

3.3. Activations of oncogenic signaling pathways and induction of cancer stem cell (CSC)-like property

The synergism of arsenic and BaP co-exposure or combined-exposure in increasing cell malignant transformation and tumorigenicity has also been investigated in cell culture models, which offered additional mechanistic insight for understanding the mechanism of synergistic lung tumorigenic effect of arsenic and BaP combined-exposure.

Using a rat lung epithelial cell (LEC) model, He et al. found that combined-BaP (100 nM, 24 h) and arsenic (NaAsO₂, 1.5 μM, 12 weeks) exposure synergized in inducing rat LEC cell transformation as evidenced by the fact that BaP plus arsenic-exposed cells formed significantly more colonies in soft agar than the cells exposed to BaP or arsenic alone [71]. In a follow-up study from the same group, Lau and Chiu reported that the LEC cells transformed by BaP plus arsenic combined-exposure had significantly higher levels of phosphor-Erk1/2 than that in LEC cells exposed to BaP or arsenic alone [72]. Further mechanistic study from this group revealed that BaP plus arsenic combined-exposure-transformed LEC cells displayed upregulations of glycolysis and oxidative phosphorylation for energy production to support their increased proliferation [73]. In a separate study, Pang et al. found that arsenic enhanced BaP-induced human bronchial epithelial (HBE) cell transformation by inducing the expression of HIF-2α to inhibit the ATM/Chk-2 pathway reducing the repair of BaP exposure-caused DNA damage [74].

Our recent studies revealed that the PI3K/Akt and Erk1/2 pathway activations due to reduced-SOCS3 expression and increased-integrin α4 (ITGA4) expression play important roles in arsenic and BaP co-exposure-induced cell transformation and tumorigenesis [2,75,76]. We found that arsenic plus BaP co-exposure synergized in activating AhR leading to the up-regulation of SUV39H1 expression, which in turn down-regulated SOCS3 expression by increasing H3K9me2 enrichment at SOCS3 promoter region [2]. SOCS3 down-regulation led to enhanced Akt and Erk/12 activation in arsenic and BaP co-exposure-transformed cells and the co-exposure-induced mouse lung tumor tissues [2]. Similar to the effects of stably knocking down SUV39H1 expression, stably overexpressing SOCS3 in arsenic plus BaP co-exposure-transformed cells significantly reduced their capability of forming soft agar colonies [2]. Moreover, stably overexpressing SOCS3 in parental BEAS-2B cells also greatly impaired the capability of arsenic plus BaP co-exposure to induce cell transformation [2]. Furthermore, inhibition of either Akt or Erk1/2 activity alone in arsenic plus BaP co-exposure-transformed cells significantly reduced their soft agar colony numbers and inhibiting both Akt and Erk1/2 simultaneously further reduced their soft agar colony numbers [2]. Together, these finds suggest that enhanced activation of Akt and Erk1/2 pathways may play important roles in promoting arsenic plus BaP combined-exposure-transformed cell and-induced tumor growth.

Cancer is now considered as a stem cell disease and it is proposed that cancer is initiated by cancer stem cells (CSCs) [77,78]. CSCs refer to a small population of cancer cells possessing characteristics associated with normal stem cells, especially the capability of self-renewal and generation of different types of cells found in a tumor. Although the CSC theory of cancer proposes that cancers are originated from CSCs, it remains largely unknown where CSCs or CSC-like cells come from. One potential source is that CSCs may come from adult tissue stem cells that are malignantly transformed through genetic mechanism or epigenetic reprograming [19]. Indeed, accumulating evidence has shown that chronic metal carcinogen exposure is capable of producing CSC-like cells [67,68,79-84].

Our recent study provided the first evidence showing that arsenic and BaP co-exposure synergized in producing CSC-like cells as evidenced by: (i) Arsenic and BaP co-exposure-transformed cells contained significantly more ALDH positive cells and were capable of forming significantly more tumorigenic pulmonary spheres compared to cells transformed by exposure to either arsenic or BaP alone [2]. Cellular ALDH activity analysis by the ALDEFLUOR assay and the suspension culture sphere formation assay are widely used assays to assess the presence of stem cells or CSCs or CSC-like cells [85]. (ii) Arsenic and BaP co-exposure-transformed cells displayed a significantly stronger tumor-initiating capability as determined by using the nude mouse xenograft tumorigenesis model [2]. (iii) Arsenic and BaP combined-exposure significantly increased mouse lung tumor multiplicity (i.e. produced more tumors in mouse lungs) [2]. Mechanistically, arsenic and BaP co-exposure-caused up-regulation of ITGA4 and SUV39H1 expression and down-regulation of SOCS3 expression play important roles in producing CSC-like cells by activating the PI3K/Akt pathway [2,75,76]. More specifically, enhanced activation of the PI3K/Akt pathway in arsenic and BaP co-exposure-transformed cells increased the deubiquitinase USP7 expression level, which stabilized the anti-apoptotic protein MCL-1 enabling apoptosis resistance and CSC-like property [76]. In addition, the synergistic activation of the PI3K/Akt pathway in arsenic and BaP co-exposure-transformed cells also destabilized and down-regulated SUFU protein resulting in a high activation of the Hedgehog/GLI-1pathway to enhance their CSC-like property [75]. Importantly, stably knocking down SUV39H1 expression in arsenic and BaP co-exposure-transformed cells significantly reduced their capability to form tumorigenic pulmonary spheres [2]. Moreover, stably knocking down SUV39H1 expression in parental BEAS-2B cells significantly reduced the capability of arsenic and BaP co-exposure to generate CSC-like cells as evidenced by a 40% reduction of tumorigenic sphere formation. Since CSCs or CSC-like cells are considered as cancer initiating cells, these findings strongly suggest that epigenetic dysregulations resulting from SUV39H1 up-regulation could play important roles in the enhanced CSC-like property and synergistic lung tumorigenesis induced by arsenic and BaP co-exposure.

4. Summary and perspectives

Accumulating strong evidence indicates that arsenic and BaP combined-exposure synergizes in inducing CSC-like property, cell malignant transformation and lung tumorigenesis. The demonstration of the synergistic lung tumorigenic effect of arsenic and BaP combined-exposure provided mechanistic insight for epidemiology findings showing that arsenic exposure and cigarette smoking act synergistically to increase human lung cancer risk. However, the mechanisms of how arsenic and BaP combined-exposure enhances lung tumorigenesis have been poorly understood. While some studies showed that arsenic and BaP combined-exposure synergizes in inducing genotoxicity in cultured cells and animal tissues, there are also studies reporting lack of enhanced genotoxicity in arsenic and BaP co-exposed cells and animals. Moreover, the reported mechanisms for the synergism in genotoxicity resulting from arsenic plus BaP combined-exposure are inconsistent and sometimes controversial. The inconsistencies may come from differences in cell types investigated, concentrations of arsenic and BaP used, treatment protocols (time durations and orders of the treatment) of arsenic and BaP combined-exposure, and animal species/strains used, etc. In addition to the enhanced genotoxicity, emerging evidence suggests that dysregulations in epigenetics, abnormal activations of oncogenic signaling pathways and induction of CSCS-like property and immunosuppression may also play critical roles in the synergistic lung tumorigenic effect of arsenic and BaP combined-exposure (Figure 1).

Figure 1.

A schematic description for the proposed mechanisms of the synergistic lung tumorigenic effect of arsenic and benzo(a)pyrene (BaP) combined- exposure. The solid black lines refer to the reported mechanisms discussed in this manuscript. The yellow description and yellow dot lines refer to the speculated mechanisms.

Given the facts that both arsenic and BaP are among the most common environmental carcinogens and that co-exposure to arsenic plus BaP is common in certain occupational workers and general populations, it is essential to clearly define the mechanism responsible for the synergistic lung tumorigenic effect of arsenic and BaP combined-exposure. A better understanding on the underlying mechanism may identify molecular targets to more efficiently treat and prevent lung tumors resulting from arsenic and BaP combined-exposure. Further studies are needed at least in following five areas. (i) Using appropriate cells, tissues and animal models to study the lung tumorigenic effect and the underlying mechanism of arsenic and BaP combined-exposure. Ideally, primary or immortalized lung epithelial cells and lung tumor-induction sensitive mouse strains are used. In this case, the identified-mechanisms may be better linked to the increased or decreased lung tumorigenic effect resulting from arsenic and BaP combined-exposure. In addition, epitranscriptomic dysregulations are emerging as new mechanisms of environmental carcinogenesis [86]. Further mechanistic studies are needed to explore whether arsenic and BaP combined-exposure causes epitranscriptomic dysregulation and its role in their co-carcinogenicity. (ii) Using human exposure relevant routes, durations and levels of arsenic and BaP to study the lung tumorigenic effect and the underlying mechanism of arsenic and BaP combined-exposure. Arsenic and BaP simultaneous co-exposure or sequential exposure (arsenic exposure first followed by BaP exposure or BaP exposure first followed by arsenic exposure) are considered as arsenic and BaP combined-exposure. (iii) Using both genders of animals to study the lung tumorigenic effect and the underlying mechanism of arsenic and BaP combined-exposure. Previous studies only investigated the lung tumorigenic effect of arsenic and BaP combined-exposure in male animals. It remains to be determined whether female animals respond similarly or differently to arsenic and BaP combined-exposure. (iv) Using appropriate animal models to determine the role of immunosuppression in the enhanced lung tumorigenic effect of arsenic and BaP combined-exposure. Immunosuppression is an important mechanism for cancer initiation and progression. While both arsenic and BaP display immunosuppressive effects, very few studies have been done to explore the co-immunosuppressive effect and its role in arsenic and BaP combined-exposure-induced lung tumorigenesis. (v) Using appropriate animal models to develop mechanism-based strategies to more efficiently diagnose, treat and prevent lung cancer caused by arsenic and BaP combined exposure. With increasingly understanding on the mechanisms of enhanced lung tumorigenic effect of arsenic and BaP combined-exposure, future studies also need to focus on identifying unique biomarkers, testing new therapeutic and preventive strategies to reduce the lung cancer risk in humans exposed to both arsenic and BaP.