Appalachian mountaintop mining particulate matter induces neoplastic transformation of human bronchial epithelial cells and promotes tumor formation.

Epidemiological studies suggest that living near mountaintop coal mining (MTM) activities is one of the contributing factors for high lung cancer incidence. The purpose of this study was to investigate the long-term carcinogenic potential of MTM particulate matter (PMMTM) exposure on human bronchial epithelial cells. Our results show that chronic exposure (3 months) to noncytotoxic, physiological relevant concentration (1 μg/mL) of PMMTM, but not control particle PMCON, induced neoplastic transformation, accelerated cell proliferation, and enhanced cell migration of the exposed lung cells. Xenograft transplantation of the PMMTM-exposed cells in mice caused no apparent tumor formation, but promoted tumor growth of human lung carcinoma H460 cells, suggesting the tumor-promoting effect of PMMTM. Chronic exposure to the main inorganic chemical constituent of PMMTM, molybdenum but not silica, similarly induced cell transformation and tumor promotion, suggesting the contribution of molybdenum, at least in part, in the PMMTM effects. These results provide new evidence for the carcinogenic potential of PMMTM and support further risk assessment and implementation of exposure control for PMMTM.

Introduction

Lung cancer is the leading cause of cancer-related death, and, after smoking, environmental and occupational exposure is a major cause.[1,2] The Appalachian Mountains stretch across 13 states of the United States from southern New York to northern Mississippi. Health disparities, most notably cancer incidence and mortality rate, are higher in the Appalachian region compared to the rest of the country.[3,4] Previous epidemiology studies demonstrated elevated lung cancer mortality in coal-mining areas of Appalachia,[5,6] suggesting that environmental contaminants from coal-mining activities may contribute to the increased lung cancer risk.

Mountaintop removal mining (MTM) is a major form of surface coal mining in Appalachia, especially in West Virginia and Kentucky.[7] In southern West Virginia, almost 40 million tons of coals were extracted by MTM in 2012.[8] Particulate matter (PM) is generated from these active MTM sites by blasting and combustion from heavy equipment and may represent a potential toxicant that is elevated in ambient air.[9] The lungs are the primary target organ for these airborne MTM-derived PM (PMMTM) exposures.[10] To date, there have been no experimental reports on the potential carcinogenic effect of PMMTM, either in vitro or in vivo. Because carcinogenesis is a multistep process commonly associated with long-term exposure to carcinogens,[11,12] we studied the chronic effects of PMMTM exposure on human bronchial epithelial cells, one of the major cellular targets of lung carcinogenesis. Such information is necessary to provide a scientific basis for the epidemiological finding on increased lung cancer mortality in the coal-mining areas of Appalachia.

In the present study, we chronically exposed human bronchial epithelial BEAS-2B cells to noncytotoxic, physiologically relevant concentration of PMMTM or control PM (PMCON) over a 3-month period in culture. The exposed lung cells were then evaluated for their neoplastic transformation, proliferative, and migratory properties in vitro and tumorigenicity in vivo. We also studied the effect of inorganic chemical constituents of PMMTM by similarly exposing bronchial epithelial cells to silica (Si) and molybdenum (Mo), the main inorganic chemical constituents of PMMTM. Our data indicate the cell-transforming and tumor-promoting effects of PMMTM; thus supporting the prudent adoption of prevention strategies and implementation of exposure control for PMMTM. The described chronic exposure model could further be used for mechanistic studies and risk assessment of PMMTM, which may not be feasible in vivo.

Materials and Methods

A more detailed description of Materials and Methods used in this study is available as Supporting Information at http://pubs.acs.org/.

Cell Culture

Human bronchial epithelial BEAS-2B and nonsmall cell lung cancer H460 cells were obtained from American Type Culture Collection (ATCC; Manassas, VA) and were cultured as described previously.[13]

Collection of MTM and Control Particulate Matters

Air samples were taken at two rural residential sites located within 1 mile of an active MTM site in Edwight, WV, U.S.A. For control, air was similarly sampled from selected rural areas in Green Bank, WV, which does not have coal mining.[14] PMMTM and PMCON were collected on PTFE fiber-backed filters with a pore size of 5 μm (Whatman, Springfield Mill, U.K.) for 2–4 weeks. The filters were extracted according to the method previously described (see Supporting Information Table S1 for PM mass).[15] It is worth noting that this method of PM collection could not preserve the volatile organic compounds. Scanning electron microscope-energy-dispersive X-ray spectroscopy (SEM-EDX), which was limited to the analysis of inorganic compounds, was further used to perform PM compositional analysis (RTI International, Research Triangle Park, NC). In comparison with PMCON, Si and Mo were found to be the main inorganic chemical elements in PMMTM with the % weight-average of 48.15 ± 26.91% and 28.90 ± 4.16% , respectively, for Si and Mo versus 23.75 ± 15.07% and 0.00 ± 0.00% of the elements in PMCON (see Table S2 for analysis of organic elements).

Cytotoxicity Assay

Cell viability was determined by MTT assay, as described previously.[16] All particles were suspended in phosphate-buffered saline (PBS) containing 5% bovine serum albumin (BSA) and were lightly sonicated prior to use to disperse the particles. The absorbance ratio of MTT formazan product of treated and nontreated cells was calculated and presented as relative cell viability.

Chronic Particle Exposure

Subconfluent cultures of bronchial epithelial BEAS-2B cells were continuously exposed to noncytotoxic concentration (1 μg/mL) of PMMTM or PMCON in 6-well plates for 3 months and were passaged biweekly. PMMTM- and PMCON-exposed BEAS-2B cells were designated as B-PMMTM and B-PMCON cells. Parallel culture grown with the same background level of dispersant provided a passage-matched control (B-NTX cells). To study the effect of PMMTM inorganic chemical elements, cells were similarly exposed to a noncytotoxic concentration (1 μg/mL) of Si or Mo for 3 months (designated as B-Si and B-Mo cells). All cells were cultured in complete medium (without treatment) for at least 10 passages prior to experiments to rule out any reversible effects.

Dosage Calculation and Human Extrapolation

PM exposure dose of 1 μg/mL in the 6-well plates (growth area ∼10 cm2) at the total volume of 1 mL corresponds to the surface area dose of 0.1 μg/cm2. Based on the reported rat lung surface area of 5000 cm2,[17] this exposure dose is equivalent to a bolus exposure of PM at 0.5 mg in the rats, which was previously shown to induce pathological changes.[18] Assuming the pulmonary surface area in humans of 100 m2, the human burden is equal to 100 mg/lung. Considering a respiratory deposition of ∼40%[14] and an adult inhalation rate of ∼16 m3/day,[19] the experimental dose could be reached within 8.5 years of human inhalation exposure at 5 μg/m3 (average total PM mass concentration in Edwight and Green Bank, WV).[14]

Soft Agar Colony Formation Assay

The chronically exposed cells at 3 × 104 cells per 24-well plate were mixed with culture medium containing 0.5% agar. The resulting cell suspensions were immediately plated onto dishes coated with 0.5% agar in culture medium. After 2 weeks, colonies larger than 50 μm in diameter were scored as positive for growth.[20]

Cell Counting

The exposed cells (3 × 104 cells) were seeded in 24-well plates and cultured in complete medium. The cells were stained with 0.4% trypan blue (Invitrogen) (to indicate dead cells), and healthy cell number was scored using a Countess automated cell counter (Invitrogen) at 2 and 5 days.

Proliferative Index

The chronically exposed cells (2 × 106 cells) were labeled with CellVue Claret Far Red Fluorescent Cell Linker (Sigma) according to the manufacturer’s protocol. After 4 days of culture, proliferative index was determined on the basis of far red fluorescence intensity using FSC Express 4 Flow Cytometry software (De Novo Software, Los Angeles, CA).

Cell Cycle Analysis

The chronically exposed cells were serum starved for 12 h and incubated in the complete medium for 8 h. The cells were then stained with 20 μg/mL PI and the percentage of cells in different phases of cell cycle was determined by FSC Express 4 software.

Migration Assay

Cell migration was determined by wound healing assay as previously described.[16] Briefly, a monolayer of chronic exposed cells was cultured in 24-well plate, and a wound space was created with a 1 mm width tip. The cell monolayers were incubated in complete medium and allowed to migrate for 24 h.

Xenograft Mouse Model

Animal care and experimental procedure described in this study were in accordance with the Guidelines for Animal Experiments at West Virginia University (IACUC no. 12-0502). Immunodeficient NOD/SCID gamma mice, strain NOD.Cg-Prkdcscid Il2rgtm1Wjl/SzJ (NSG; Jackson Laboratory, Bar Harbor, ME), were maintained under pathogen-free conditions within the institutional animal facility.[21,22] Mice were injected subcutaneously (SC) with 3 × 105 luciferase (Luc2; Capital Biosciences, Rockville, MD)-labeled lung cancer H460 cells and 6 × 105 PM-exposed cells (1:2 ratio) suspended in 100 μL of ExtraCel hydrogel (Advanced BioMatrix, San Diego, CA). Tumor growth of luciferase-labeled cells was monitored weekly using IVIS bioimaging (PerkinElmer, Waltham, MA). At the end of experiments, mice were euthanized, and SC tumors were dissected and weighted. Metastasis of tumor cells to neighbor tissues was analyzed by IVIS imaging after removal of primary SC tumors.

Statistical Analysis

The data represent means ± SD from three or more independent experiments as indicated. Statistical analysis was performed by Student’s t test at a significance level of P < 0.05.

Results

Effects of PMMTM on Cytotoxicity

The purpose of this study was to establish an experimental human lung cell model for PMMTM carcinogenesis studies that would allow further molecular and cellular mechanistic studies underlying cancer-like phenotypes. We first characterized the acute cytotoxic effect of PMs to determine their noncytotoxic concentrations for subsequent long-term exposure studies. Human bronchial epithelial cells were exposed to various concentrations (0.1–10 μg/mL) of PMCON and PMMTM for 48 h, and cell viability was determined by MTT assay. The results showed that none of the PMCON and PMMTM treatments caused a significant effect on cell viability as compared to nontreated (NTX) control (Figure 1A). We similarly tested the dose effect of inorganic chemical elements of PMMTM (Si and Mo) on cell viability. The results similarly showed the noncytotoxic effect of Si and Mo at the treatment doses of 0.1–10 μg/mL (Figure 1B). As we observed a slight increase in cell viability (proliferation), albeit not significant, at the high dose of PMMTM (10 μg/mL), we used a lower dose (1 μg/mL) in our subsequent chronic exposure studies.

Figure 1

Effect of acute PMMTM exposure on cytotoxicity of human bronchial epithelial cells. (A) Subconfluent monolayers of BEAS-2B cells were left untreated (nontreatment, NTX) or treated with various concentrations (0.1–10 μg/mL) of PMMTM and PMCON for 48 h and analyzed for cell viability using MTT assay. (B) Cells were treated with Si or Mo at the same concentration range and analyzed for cell viability after 48 h by MTT assay. Data are mean ± SD (n = 4).

Chronic PMMTM Exposure Induces Neoplastic Transformation

To mimic the long-term carcinogenic process, cells were chronically exposed to a noncytotoxic concentration of PMs at 1 μg/mL (0.1 μg/cm2 surface area dose) and passaged biweekly. This surface area dose mimics the in vivo dose in rodents of 0.5 mg or approximately 8.5 years of human inhalation exposure as described in Materials and Methods. In this study, the cells were exposed to PMMTM, PMCON, or left untreated for 3 months (Figure 2A), after which they were grown in complete medium (without treatment) for at least 10 passages and examined for anchorage-independent growth by soft-agar colony formation assay, which is one of the most stringent indicators of neoplastic transformation.[23] To determine the key inorganic constituents of PMMTM that may contribute to its pathological effect, cells were similarly exposed to Si and Mo, and analyzed for cell transformation. Figure 2B,C show that as compared to particle-control B-PMCON cells, the B-PMMTM and B-Mo cells formed larger and greater numbers of colony, whereas the B-Si cells exhibited a similar colony forming activity. These results indicate the neoplastic transformation of B-PMMTM and B-Mo cells.

Figure 2

Chronic exposure to PMMTM induces neoplastic transformation of human bronchial epithelial cells. (A) Schematic representation of chronic exposure model. BEAS-2B cells were continuously exposed to noncytotoxic concentration (1 μg/mL) of PMCON, PMMTM, Si, and Mo for 3 months and designated as B-PMCON, B-PMMTM, B-Si, and B-Mo cells, respectively. BEAS-2B cells maintained in culture without particle exposure (B-NTX) served as passage control cells. (B, C) Cells were seeded on 0.5% agar plates, and after 2 weeks they were visualized under a phase contrast microscope. (C) Quantification of large colonies (>50 μm in diameter). Data are mean ± SD (n = 4). *P < 0.05 (power >95%) vs passage control B-NTX cells. **P < 0.05 (power >80%) vs B-PMCON cells.

Chronic PMMTM Exposure Alters Cell Growth Characteristic

Excessive cell growth is one of the carcinogenic properties of malignant cells.[24,25] To determine whether chronic PM exposure affects cell growth characteristic, the exposed cells were analyzed for cell proliferation by direct cell counting and dye-based assays. Figure 3A shows that the B-PMMTM and B-Mo cells exhibited a significantly higher proliferation rate than the B-PMCON cells, which grew at a similar rate as the B-NTX and B-Si cells. To confirm this result which was based on direct cell counting assay, the cells were stained with membrane dye CellVue Claret, and their proliferative index was determined by flow cytometry. This dye-based assay measures cell proliferation based on the principle of dye dilution upon cell division. Consistent with the direct cell counting result, the dye-based assay indicated a higher proliferative index of B-PMMTM and B-Mo cells compared to B-PMCON and B-Si cells (Figure 3B). Analysis of cellular fluorescence intensity further indicated the division of parental cells with the observed seventh generation of daughter cells only in the B-PMMTM and B-Mo cells (Figure 3C), thus substantiating the above finding.

Figure 3

Chronic PMMTM exposure accelerates proliferation of human bronchial epithelial cells. (A) B-PMCON, B-PMMTM, B-Si, B-Mo, and B-NTX cells were plated in 24-well plates at the density of 3 × 104 cells in growth medium. After 2 and 5 days, the cells were counted using an automated cell counter. Data are mean ± SD (n = 3). *P < 0.05 (power >95%) vs passage control B-NTX cells. (B) Cells were labeled with membrane dye CellVue Claret. After 4 days of culture, cellular fluorescence intensity was determined by flow cytometry, and proliferative index was calculated. (C) Representative flow cytometric histograms from three independent experiments showing brightly stained parental cells and weakly stained daughter cells.

To delineate the mechanism of PMMTM-induced cell proliferation, we investigated the cell cycle progression of synchronized B-PMMTM, B-PMCON, B-Si, and B-Mo cells by flow cytometry using propidium iodide (PI) DNA staining assay. As depicted in Figure 4A, a higher percentage of B-PMMTM cells compared to B-PMCON cells entered the S phase (∼80% vs 50%) and reached the G2/M transition phase (∼5% vs 1%), whereas the B-Mo cells had a significant portion in the G2/M phase (∼10%). These results indicated the promotion of S phase entry by chronic PMMTM exposure and the transition to G2/M phase by chronic Mo exposure.

Figure 4

Chronic PMMTM exposure alters cell cycle of human bronchial epithelial cells. (A) B-NTX, B-PMCON, B-PMMTM, B-Si, and B-Mo cells were serum starved for 12 h to synchronize their cell cycle. They were then cultured in a complete medium for 8 h and analyzed for their cell cycle by flow cytometry. Representative histograms from three independent experiments were shown. (B) Plots are percentages of total cells (50 000 events) in each phase of the cell cycle (G1, S, and G2/M).

Chronic PMMTM Exposure Promotes Cell Migration

The aggressive behavior of PMMTM-exposed cells was examined by assessing their migratory activity, which is a key determinant of tumor invasion and progression.[26,27] Cell migration was determined by scratch or wound healing assay. At 24 h after the scratch, B-PMMTM and B-Mo cells showed a significantly higher motility rate toward the wound compared to B-NTX, B-PMCON and B-Si cells, as judged by their greater wound closure (Figure 5A,B). These results indicate the induction of aggressive cell behavior by chronic exposure to PMMTM and Mo.

Figure 5

Chronic PMMTM exposure enhances migration of human bronchial epithelial cells. (A) Confluent monolayers of B-NTX, B-PMCON, B-PMMTM, B-Si, and B-Mo cells were wounded, and the cells were allowed to migrate for 24 h. Wound space was visualized under a phase contrast microscope and analyzed by comparing the change in wound space as a percentage of wound closure. Data are mean ± SD (n = 3). *P < 0.05 (power >80%) vs passage control B-NTX cells. **P < 0.05 (power >80%) vs B-PMCON cells. (B) Representative micrographs from three independent experiments were shown.

Tumorigenicity Assessment of PMMTM Cells in Mice

Carcinogenesis is a multistep sequential process consisting of three major stages, namely, initiation, promotion, and progression.[12,28,29] Certain carcinogens can act in one or all of these stages, which results in neoplastic transformation and tumor development.[30] Having demonstrated the neoplastic transformation of B-PMMTM cells, we next assessed their tumorigenic potential in vivo. The B-PMMTM cells and their control B-PMCON cells as well as B-NTX, B-Mo, and B-Si cells (1 × 106) were injected into NSG mice subcutaneously and tumor formation was determined over time. No tumor formation was observed with any of the above treatments including those injected with the neoplastic B-PMMTM and B-Mo cells (Figure 6A), indicating their inherent nontumorigenicity. To test whether these cells might possess tumor-promoting activity, we coinjected the B-PMMTM, B-PMCON, B-Si, or B-Mo cells (6 × 105) with tumorigenic human lung cancer H460 cells (3 × 105), which have been modified to express luciferase to aid quantitation of tumor formation in mice by bioluminescence imaging (Figure 6B). Tumor luminescence signals were quantified over time and normalized to their initial signal at the time of inoculation (day 1). At 1 week postinjection, tumor luminescence was higher in mice bearing the H460 cells with B-PMMTM, B-Mo, or B-Si cells as compared to the mice bearing the H460 with B-PMCON cells (Figure 6C). At 2 weeks postinjection, the tumor luminescence intensity was high only in the mice injected with H460 cells and B-PMMTM or B-Mo cells, but not B-PMCON or B-Si cells (Figure 6D). These results indicate the tumor-promoting activity of B-PMMTM and B-Mo cells.

Figure 6

Chronic PMMTM exposed cells promote tumor formation of human nonsmall cell lung cancer H460 cells in mice. (A) Growth kinetics of H460 or transformed B-PMCON, B-PMMTM, B-Si, and B-Mo cells (1 × 106 cells) when SC injected into the NSG mice alone. E indicates the end of experiment. (B) Transformed cells at the dose of 6 × 105 cells were coinjected with luciferase-labeled H460 cells at the dose of 3 × 105 cells (2:1 ratio) into the left and right flanks of NSG mice. Tumor formation was monitored weekly by IVIS bioluminescence imaging. Representative IVIS images of mice at day 1 and week 2 are shown. (C, D) Normalization of tumor bioluminescence signals at 1 (C) and 2 (D) weeks postinjection to their initial signal at day 1. Data are mean ± SD (n = 4). *P < 0.05 (power >60%) vs H460 and B-PMCON coinjection.

At the end of the experiments (week 3), SC tumors were dissected and their bioluminescence were determined and compared between groups (Figure 7A). Figure 7B shows a stronger bioluminescence signal in tumors from H460 and B-PMMTM or H460 and B-Mo cells, compared to those from H460 and B-PMCON or H460 and B-Si cells. Analysis of tumor weight of the samples further supported the tumor-promoting role of B-PMMTM and B-Mo cells (Figure 7C). Interestingly, we observed notable tumor bioluminescence signals in mice bearing H460 and B-PMMTM, B-Si or B-Mo cells after the dissection of SC tumors (Figure 7A,D), suggesting metastasis of tumor cells to neighboring tissues and strengthening the important role of chronic PMMTM and Mo exposure in tumor promotion.

Figure 7

Analysis of ex vivo tumors of human lung cancer H460 cells. At 3 weeks postinjection, SC tumors were dissected from mice bearing H460 and B-PMCON, B-PMMTM, B-Si, or B-Mo cells. (A) Representative bioluminescence images of mice and SC tumors. (B) Quantitative analysis of bioluminescence signals from SC tumors. Data are mean ± SD (n = 4). *P < 0.05 (power >60%) vs H460 and B-PMCON coinjection. (C) The weight of dissected SC tumors. (D) Quantitative analysis of bioluminescence signals from the back of mice.

Discussion

A growing body of evidence links living in proximity to MTM activities to greater risk of serious health consequences, including significantly higher reports of cancer.[31] The MTM operation uses explosive and excavation equipment to remove vegetation, rock, and dirt from mountaintops to expose coal seams, and thus, it consists of active areas of blasting, crushing, and grinding.[7] MTM activities result in the production of atmospheric PM (PMMTM) that might be associated with human health effects. Currently, the direct relationship between chronic pulmonary exposure to PMMTM and lung cancer risk has not been investigated. In this study, we reported a combined in vitro–in vivo model for PMMTM lung carcinogenesis studies using chronically exposed human bronchial epithelial BEAS-2B cells and a mouse xenograft model. Bronchial epithelial cells were chosen in this study as they are one of the key targets for lung carcinogenesis. Airway epithelium lines the body’s first physiological barrier to inhaled PM and the particles in range of 1–5 μm generally deposited in the tracheobronchial region of the airways—such deposition appears to be a close correlation with the incidence of primary cancer sites.[32]

Anchorage-independent growth has been well correlated with the tumorigenicity and invasiveness of several cancer cell types.[24] Colony formation under soft agar assay, the gold standard test to evaluate the ability of cells to undergo anchorage-independent growth, is therefore the most stringent indicator for neoplastic transformation. We showed that chronic PMMTM-exposed B-PMMTM cells induced larger number and size of colonies as compared to chronic PMCON-exposed B-PMCON cells and passage-matched control B-NTX cells (Figure 2). It has previously been reported that BEAS-2B cells might undergo squamous differentiation in the presence of serum.[33,34] However, because B-PMCON and B-NTX cells showed no phenotypic changes or neoplastic behavior under the culture condition, it can be concluded that the B-PMMTM cells, not B-PMCON and B-NTX cells, have undergone differentiation and show altered phenotype due to continued exposure. In order to delineate the chemical effects of PMMTM inorganic elements, bronchial epithelial cells were similarly exposed to Si and Mo and neoplastic transformation was observed in the Mo-exposed B-Mo cells, but not Si-exposed B-Si cells. Given that the magnitude of Mo effect was similar to the PMMTM, despite its higher concentration than those presented in PMMTM, it is likely that (i) some other elements could be involved in the PMMTM effect and that (ii) such an effect arose from the synergistic effect of more than one component (e.g., Si and Mo).

Various carcinogenic properties representing the hallmarks of malignant cells were further assessed in this study. The transformed B-PMMTM and B-Mo cells demonstrated excessive cell growth and altered cell cycle (Figures 3 and 4). To our knowledge, this is the first demonstration of the induction of cell proliferation by chronic low-dose PM exposure, although the inhibition of cell proliferation[35,36] and induction of cytotoxicity[37,38] by acute high-dose PMs from urban and industrial areas have previously been demonstrated. Interestingly, B-PMMTM and B-Mo cells promoted cell cycle at different phases, possibly due to (i) the low content of Mo present in the PMMTM that might not be sufficient to drive the cells to G2/M phase, and (ii) the PMMTM proliferative effect was probably the result of components other than Mo or a synergistic effect of more than one component. Cell motility was also shown to increase significantly in the B-PMMTM and B-Mo cells as compared to B-NTX cells (Figure 5), thus indicating their aggressive behaviors, which could be important in tumor progression.

The potential role of PMMTM in lung carcinogenesis was further evaluated in vivo using a mouse xenograph model. Our results demonstrated that B-PMMTM cells, although not directly tumor-inducing in mice, promoted tumor formation and metastasis of human lung cancer H460 cells (Figures 6 and 7). The limitation of our in vivo study is the relatively small number of animals per group, and given the high individual biological variation, we could not obtain statistical power >80%. Despite the low statistical power, we were still able to achieve statistical significance (P < 0.05). These results however are in good agreement with previous reports showing the hypermethylation of tumor suppressor p16 by PMs from urban areas, which could lead to cancer development[39] and the induction of lung carcinoma by Mo inhalation exposure in mice.[40]

Taken together, the present study demonstrated that chronic exposure to PMMTM induced neoplastic transformation of human bronchial epithelial cells with cancer-like properties. Although the data did not indicate tumor initiation by PMMTM, the lung tumor promotion and progression by PMMTM are a health concern as a cancer promoter. Our finding strengthens previous epidemiological studies linking MTM to increased incidence of lung cancer[5,6,41] and supports prudent adoption of prevention strategies and exposure control for PMMTM. As more than 60 000 cancer cases has been estimated to correlate with MTM activities in West Virginia,[31] this finding on the cancer-promoting effect of PMMTM and related epidemiological data are crucial to raise public health awareness to reduce cancer risk. Our study also suggested that Mo could be one of the key inorganic elements responsible for the cancer-promoting effect of PMMTM, although we could not rule out the involvement of organic elements as well as the synergistic effect of more than one element in the process. Finally, the chronic exposure model described in this study may be useful in further mechanistic studies and risk assessment of PMMTM pathogenicity.