Skip to main content
PMC home page
ACS AuthorChoice logoLink to ACS AuthorChoice
. 2011 Feb 4;24(2):165–167. doi: 10.1021/tx1004419

Epigenetic Changes in Individuals with Arsenicosis

Lisa Smeester †,#, Julia E Rager †,#, Kathryn A Bailey †, Xiaojun Guan §, Nikia Smith †, Gonzalo García-Vargas ∥, Luz-Maria Del Razo ⊥, Zuzana Drobná ‡, Hemant Kelkar §, Miroslav Stýblo ‡, Rebecca C Fry †,*
PMCID: PMC3042796  PMID: 21291286

Abstract

graphic file with name tx-2010-004419_0001.jpg

Inorganic arsenic (iAs) is an environmental toxicant currently poisoning millions of people worldwide, and chronically exposed individuals are susceptible to arsenicosis or arsenic poisoning. Using a state-of-the-art technique to map the methylomes of our study subjects, we identified a large interactome of hypermethylated genes that are enriched for their involvement in arsenic-associated diseases, such as cancer, heart disease, and diabetes. Notably, we have uncovered an arsenic-induced tumor suppressorome, a complex of 17 tumor suppressors known to be silenced in human cancers. This finding represents a pivotal clue in unraveling a possible epigenetic mode of arsenic-induced disease.

Inorganic arsenic (iAs) is an environmental toxicant currently poisoning tens of millions of people worldwide. Individuals chronically exposed to iAs are susceptible to arsenicosis or chronic arsenic poisoning. Heavily suffering areas such as West Bengal and Bangladesh saw a rise in incidents of arsenicosis when government officials and international aid agencies, in the hopes of mitigating waterborne diseases, introduced tube wells fed from arsenic-contaminated aquifers.(1) Other regions, such as Mexico, are affected by both naturally occurring arsenic as well as anthropogenic sources such as smelters and ore mining operations. Chronic exposure to iAs is associated with the development of various diseases including heart disease, diabetes, and cancer, and exposed individuals often present with hallmark skin lesions.(2) Premalignant skin lesions may indicate increased risk for arsenic-related cancer.(3) While the precise mode of action in arsenic-induced disease is unknown, one of the proposed mechanisms is altered gene regulation via epigenetic modes of action such as DNA methylation.(4) Supporting this is the finding that early life exposure can result in long-term health consequences,5,6 suggesting that there are heritable changes to the genome.

Previous studies highlight the association of arsenicosis with altered gene expression patterns in humans displaying the hallmark skin lesions.(7) Moreover, gene-specific analyses suggest the role of altered DNA methylation at target sites such as tumor protein p53 (p53), cyclin-dependent kinase inhibitor 2A (CDKN2A/p16), and Ras association (RalGDS/AF-6) domain family member 1 (RASSF1A).8,9 However, it remains to be shown whether multiple genes and pathways are affected by epigenetic processes in individuals with signs of arsenicosis. Therefore, we set out to identify differentially methylated genomic regions associated with arsenicosis in humans from Zimapan, Hidalgo State, Mexico who were exposed to varying levels of iAs via their drinking water as assessed by urinary arsenic (Supporting Information, Table 1).

To our knowledge, this is the first study to examine genome-wide site-specific DNA methylation alterations due to arsenic-induced toxicity in a population with ongoing exposure. Peripheral blood lymphocyte DNA of 16 of the individuals, half with established elevated levels of iAs exposure and showing signs of arsenicosis (skin lesions), was analyzed using a methylated CpG island recovery (MIRA)-chip assay. In addition to the difference in skin lesion status, the two groups showed different levels of iAs exposure as assessed in urine (p < 0.001). CpG-methylated DNA was isolated using a methyl binding domain protein complex and hybridized to Affymetrix Human Promoter arrays, assessing over 4.6 million sites tiled through human promoter regions. Our analysis enabled a comprehensive examination of DNA methylation levels within CpG islands for over 14,000 genes.

We employed a comparative analytical approach to identify differentially methylated CpG islands and found 183 genes with differential patterns, of which 182 were hypermethylated in individuals with signs of arsenicosis (Figure 1). Specifically, the identified genes showed a statistically significant (false discovery rate (FDR) q-value <0.05) difference in average DNA methylation for each CpG island. This assay allowed for interrogation of one the three genes previously identified as being hypermethylated in individuals with signs of arsenicosis, e.g., p16. While the fold change of p16 did not meet the statistical threshold for this study, here we also observed increased promoter methylation in individuals with signs of arsenicosis (fold change of 1.24).

Figure 1.

Figure 1

Open in a new tab

Arsenicosis-associated patterns of DNA methylation. The heat map illustrates the average DNA methylation levels in promoter regions of 183 genes. Data are z-score normalized for each gene. Red represents a relative increase in CpG island methylation level, and blue represents a relative decrease in methylation level.

Using a systems level approach, the 183 genes were analyzed for known molecular interactions, and a large interactome of hypermethylated genes was identified (Figure 2A). These are enriched for their involvement in cancer-associated pathways mediated by genes such as p53 (Figure 2B). Interestingly, we found that many of the proteins encoded by genes with differentially methylated CpG islands are known players in arsenic-associated disease, such as heart disease, diabetes, and cancer (Supporting Information, Table 5).

Figure 2.

Figure 2

Open in a new tab

Epigenetically modified iAs-induced networks. (A) Large interacting network of hypermethylated genes. (B) Tumor protein p53 (tp53)-associated network. (C) The iAs-induced tumor suppressorome. P-values are shown in the top right corners of each network. Networks are displayed with symbols representing products of hypermethylated genes (red symbols) or the proteins associated with these genes (clear symbols).

Notably, we have also identified an arsenic-methylated tumor suppressorome (Figure 2C), a pivotal clue in unravelling a possible epigenetic mode of arsenic-induced disease. The tumor suppressorome is a complex of 17 known or putative tumor suppressors silenced in human cancers. It comprises the following hypermethylated genes: C11orf70 (chromosome 11 open reading frame 70), CENPE (centromere protein E, 312 kDa), EEF1E1 (eukaryotic translation elongation factor 1 epsilon 1, also known as p18), ENDOG (endonuclease G), FOXF1 (forkhead box F1), HOXB5 (homeobox B5), HOXB9 (homeobox B9), hsa-mir-126 (human microRNA 126), MMP15 (matrix metallopeptidase 15 (membrane inserted)), MSX1 (msh homeobox 1, also known as HOX7), POLD4 (polymerase (DNA-directed), delta-4, also known as p12), PRDM2 (PR domain containing 2, with ZNF domain, also known as RIZ), RNF20 (ring finger protein 20), SMARCD2 (SWI/SNF related, matrix associated, actin dependent regulator of chromatin, subfamily d, member 2), SUFU (suppressor of fused homologue (Drosophila)), TBR1 (T-box, brain, 1), and TSC22D3 (TSC22 domain family, member 3).

Among the members of the tumor suppressorome, of particular interest are those with known associations to arsenic-induced diseases such as cancer of the bladder, kidney, lung, liver, and prostate, as well as cardiovascular disease and diabetes mellitus (Figure 2C, Supporting Information, Table 5). Interestingly, the expression levels of specific members of the tumor suppressorome have previously been shown to be altered via iAs exposure. For example, iAs exposure in vitro results in the downregulation of both MSX1(10) and CENPE.(11) In this study, we find that the CpG islands within the promoter regions of the identified genes are hypermethylated in individuals with skin lesions. As mentioned, DNA hypermethylation of the promoter regions of three genes has been reported in arsenic-induced disease.8,9 Notably, the results from our study vastly increase this list of gene targets. Examination of these gene targets would be the next step in understanding how epigenetic changes regulate gene expression and, subsequently, cause dysregulation leading to disease.

In this study we have analyzed epigenetic changes in the peripheral blood lymphocyte DNA from iAs exposed and diseased individuals and do not directly measure alterations in target organs. Recent studies support the utilization of lymphocyte DNA to detect genomic and epigenetic biomarkers of organ-specific disease.12,13 In future research, it will be possible to compare the epigenetic alterations of the tumor suppressorome from tissue samples of arsenic-exposed individuals.

In conclusion, these results demonstrate that a large number of genes are epigenetically modified in the lymphocyte DNA of individuals exposed to iAs with related arsenicosis. It is likely that the pathways we have identified here are influenced at the transcriptional level resulting in the repression of their activity in exposed individuals. Our findings demonstrate the significant effects of iAs on the epigenome. The identified methylation sites and differential DNA methylation patterns may serve as biomarkers of adverse health effects associated with iAs exposure. Through the identification of differential patterns of methylation, we hope to detail arsenic effects in humans in order to understand arsenic-induced disease and to identify potential methods for disease prevention.

Acknowledgments

We thank P.A. Wade, A. Dhasarathy, and especially A.Y. Lai for their invaluable help.

Glossary

Abbreviations

iAs

inorganic arsenic

C11orf70

chromosome 11 open reading frame 70

CDKN2A/p16

cyclin-dependent kinase inhibitor 2A

CENPE

centromere protein E, 312 kDa

EEF1E1

eukaryotic translation elongation factor 1 epsilon 1, also known as p18

ENDOG

endonuclease G

FOXF1

forkhead box F1

FDR

false discovery rate

HOXB5

homeobox B5

HOXB9

homeobox B9

hsa-mir-126

human microRNA 126

MIRA-Chip

methylated CpG Island Recovery-Chip assay

MMP15

matrix metallopeptidase 15 (membrane inserted)

MSX1

msh homeobox 1, also known as HOX7

p53

tumor protein p53

POLD4

polymerase (DNA-directed), delta-4, also known as p12

PRDM2

PR domain containing 2, with ZNF domain, also known as RIZ

RASSF1A

ras association (RalGDS/AF-6) domain family member 1

RNF20

ring finger protein 20

SMARCD2

SWI/SNF related, matrix associated, actin dependent regulator of chromatin, subfamily d, member 2

SUFU

suppressor of fused homologue (Drosophila)

TBR1

T-box, brain, 1

TSC22D3

TSC22 domain family, member 3.

Supporting Information Available

Experimental methods and detailed gene lists. This material is available free of charge via the Internet at http://pubs.acs.org.

Microarray data have been submitted to NCBI's Gene Expression Omnibus (GEO) repository (http://www.ncbi.nlm.nih.gov/geo/) and are available under accession number GSE26073.

This research was supported in part by grants from the National Institute of Environmental Health Sciences (P30ES010126 and T32 ES7018) to R.C.F., US EPA STAR Grant No. 832735 to M.S., and by a grant from Pfizer to R.C.F.

Author Contributions

These individuals contributed equally to this work.

Funding Statement

National Institutes of Health, United States

Supplementary Material

tx1004419_si_001.pdf (489.3KB, pdf)

References

  1. Bhattacharjee Y. (2007) Toxicology: a sluggish response to humanityʼs biggest mass poisoning. Science 315, 1659–1661. [DOI] [PubMed] [Google Scholar]
  2. Yoshida T.; Yamauchi H.; Fan Sun G. (2004) Chronic health effects in people exposed to arsenic via the drinking water: dose-response relationships in review. Toxicol. Appl. Pharmacol. 198, 243–252. [DOI] [PubMed] [Google Scholar]
  3. NRC (2001) Arsenic in Drinking Water, 2001 Update, National Academy Press, Washington, DC. [Google Scholar]
  4. Ren X.; McHale C. M.; Skibola C. F.; Smith A. H.; Smith M. T.; Zhang L. (2011) An emerging role for epigenetic dysregulation in arsenic toxicity and carcinogenesis. Environ. Health Perspect. 119, 11–19. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Waalkes M. P.; Liu J. (2008) Early-life arsenic exposure: methylation capacity and beyond. Environ. Health Perspect. 116, 3A104. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Yuan Y.; Marshall G.; Ferreccio C.; Steinmaus C.; Liaw J.; Bates M.; Smith A. H. (2010) Kidney cancer mortality: fifty-year latency patterns related to arsenic exposure. Epidemiology 21, 103–108. [DOI] [PubMed] [Google Scholar]
  7. Argos M.; Kibriya M. G.; Parvez F.; Jasmine F.; Rakibuz-Zaman M.; Ahsan H. (2006) Gene expression profiles in peripheral lymphocytes by arsenic exposure and skin lesion status in a Bangladeshi population. Cancer Epidemiol. Biomarkers Prev. 15, 1367–1375. [DOI] [PubMed] [Google Scholar]
  8. Chan M. W. Y.; Chan L. W.; Tang N. L. S.; Lo K. W.; Tong J. H. M.; Chan A. W. H.; Cheung H. Y.; Wong W. S.; Chan P. S. F.; Lai F. M. M.; To K. F. (2003) Frequent hypermethylation of promoter region of RASSF1A in tumor tissues and voided urine of urinary bladder cancer patients. Int. J. Cancer 104, 611–616. [DOI] [PubMed] [Google Scholar]
  9. Chanda S.; Dasgupta U. B.; GuhaMazumder D.; Gupta M.; Chaudhuri U.; Lahiri S.; Das S.; Ghosh N.; Chatterjee D. (2006) DNA hypermethylation of promoter of gene p53 and p16 in arsenic-exposed people with and without malignancy. Toxicol. Sci. 89, 431–437. [DOI] [PubMed] [Google Scholar]
  10. Flora S. J. S.; Mehta A. (2009) Monoisoamyl dimercaptosuccinic acid abrogates arsenic-induced developmental toxicity in human embryonic stem cell-derived embryoid bodies: comparison with in vivo studies. Biochem. Pharmacol. 78, 1340–1349. [DOI] [PubMed] [Google Scholar]
  11. Zheng X. H.; Watts G. S.; Vaught S.; Gandolfi A. J. (2003) Low-level arsenite induced gene expression in HEK293 cells. Toxicology 187, 39–48. [DOI] [PubMed] [Google Scholar]
  12. Anglim P. P.; Alonzo T. A.; Laird-Offringa I. A. (2008) DNA methylation-based biomarkers for early detection of non-small cell lung cancer: an update. Mol. Cancer 7, 81. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Sinnaeve P. R.; Donahue M. P.; Grass P.; Seo D.; Vonderscher J.; Chibout S. D.; Kraus W. E.; Sketch M. Jr.; Nelson C.; Ginsburg G. S.; Goldschmidt-Clermont P. J.; Granger C. B. (2009) Gene expression patterns in peripheral blood correlate with the extent of coronary artery disease. PLoS One 4, e7037. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

tx1004419_si_001.pdf (489.3KB, pdf)

Articles from Chemical Research in Toxicology are provided here courtesy of American Chemical Society

RESOURCES