Skip Navigation


Carcinogenesis Advance Access originally published online on March 20, 2007
Carcinogenesis 2007 28(7):1520-1525; doi:10.1093/carcin/bgm063
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
28/7/1520    most recent
bgm063v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (2)
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Breton, C. V.
Right arrow Articles by Christiani, D. C.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Breton, C. V.
Right arrow Articles by Christiani, D. C.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© The Author 2007. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oxfordjournals.org

Susceptibility to arsenic-induced skin lesions from polymorphisms in base excision repair genes

Carrie V. Breton1,*, Wei Zhou2, Molly L. Kile2, E.A. Houseman3, Quazi Quamruzzaman4, Mahmuder Rahman4, Golam Mahiuddin4 and David C. Christiani1,2

1 Department of Epidemiology, Harvard School of Public Health, 677 Huntington Avenue, Boston, MA 02115, USA
2 Department of Environmental Health, Harvard School of Public Health, 665 Huntington Avenue, Boston, MA 02115, USA
3 Department of Work Environment, University of Massachusetts Lowell, Kitson Hall, 202E, One University Avenue Lowell, MA 01854, USA
4 Dhaka Community Hospital, 190/1 Baro Moghbazar, Wireless Railgate, 1217, Dhaka, Bangladesh

* To whom correspondence should be addressed. Tel: +617 416 9897; Fax: +617 432 3441; Email: cbreton{at}hsph.harvard.edu


   Abstract
Top
 Abstract
Introduction
 Materials and methods
 Results
 Discussion
 References
 
Genetic polymorphisms in the base excision DNA repair pathway may influence individual susceptibility to arsenic and the development of arsenic-induced skin lesions. Data from a case–control study of 792 cases and 792 matched controls conducted in Bangladesh from 2001 to 2003 were analyzed using conditional logistic regression to assess the associations between four common base excision repair (BER) genetic polymorphisms X-ray repair cross-complementing group 1 (XRCC1) Arg399Gln, XRCC1 Arg194Trp, human 8-oxoguanine DNA glycosylase (hOGG1) Ser326Cys and apurinic/apyrimidinic endonuclease (APE1) Asp148Glu and arsenic-induced skin lesions including melanosis and keratosis. Adjusted for toenail arsenic, body mass index, education, smoking and betel nut use, individuals with the APE1 148Glu/Glu polymorphism had a 2-fold increased odds of skin lesions compared with individuals with the 148Asp/Asp genotype (1.93; 95% confidence interval 1.15, 3.19). Gene–environment interactions between toenail arsenic and XRCC1 Arg194Trp and APE1 Asp148Glu were observed. Within the lowest arsenic tertile, APE1 148Glu/Glu had 2.5 times the odds ratio compared with wild-type, whereas within the highest tertile of arsenic the odds ratios for skin lesions did not differ. In contrast, at low arsenic levels, the odds ratios for skin lesions did not differ much by XRCC1 Arg194Trp genotype. However, at the highest tertile of arsenic, the XRCC1 194Arg/Arg polymorphism conferred a 3-fold larger odds ratio for skin lesions compared with XRCC1 194Trp/Trp. Individuals may have different odds for developing skin lesions based in part on their genetic profile for BER and their arsenic exposure history. Future research on arsenic-induced skin lesions should consider the impact of genetic variation to individual susceptibility to arsenic toxicity.

Abbreviations: APE1, apurinic/apyrimidinic endonuclease; BER, base excision repair; BMI, body mass index; CIs, confidence intervals; hOGG1, human 8-oxoguanine DNA glycosylase; HWE, Hardy–Weinberg equilibrium; POL ß, polymerase ß; RORint, ratio of odds ratios for the interaction; RRint, interaction risk ratio; XRCC1, X-ray repair cross-complementing group 1


   Introduction
Top
 Abstract
 Introduction
Materials and methods
 Results
 Discussion
 References
 
Chronic arsenic exposure has been linked to a wide variety of diseases including cancers, peripheral vascular disease, chronic cough, bronchitis, cardiac disease and peripheral neuropathy (1,2). In Bangladesh, where arsenic-contaminated drinking water is endemic in many regions, the earliest observed physical manifestations of arsenic toxicity are skin lesions including melanosis and keratosis, which can later lead to skin cancer such as Bowen's disease, basal cell and squamous cell carcinomas (3,4,5,6). However, not everyone exposed to arsenic develops skin lesions, suggesting other factors play a role in individual susceptibility to arsenic toxicity.

Arsenic can generate DNA damage that needs to be repaired (7) and change cellular capacity for DNA repair (8,9,10,11). If the development of skin lesions is dependent in part on the accumulation of unrepaired DNA damage resulting in proliferation of aberrant skin cells that subsequently progress into visible skin lesions, polymorphisms in DNA repair pathways could influence susceptibility to arsenic. Not only could polymorphisms in DNA repair pathway genes alter DNA repair function in general but also they may do so in a manner conditional on arsenic exposure.

Base excision repair (BER) and nucleotide excision repair are two main DNA repair pathways through which the body repairs spontaneous and endogenously produced DNA damage (12,13,14). Nucleotide excision repair plays a role in repairing non-bulky lesions and a polymorphism in ERCC2 was recently shown to confer increased risk for development of hyperkeratosis among an arsenic-exposed population in West Bengal (15). However, BER is the most important pathway for the removal of oxidative lesions (14), which are hypothesized to play a key role in arsenic toxicity (7,16,17), and thus was the focus for this investigation.

Many proteins are involved in BER, but three of the most well-studied are X-ray repair cross-complementing group 1 (XRCC1), apurinic/apyrimidinic endonuclease (APE1) and human 8-oxoguanine DNA glycosylase (hOGG1). XRCC1, a scaffolding protein, can interact with multiple enzymatic components at each stage of the repair process, including DNA polymerase ß (POL ß), hOGG1 and APE (18). Additionally, XRCC1 repairs single strand breaks resulting either from the BER process itself or damage to deoxyribose. APE1 is the essential protein that excises apurinic/apyrimidinic sites generated when glycosylases initiate repair of a damaged base (18). APE1 also helps recruit POL ß and facilitate further steps in the repair process. hOGG1 is a protein specific for the repair of oxidative damage, primarily the DNA adduct 8-Hydroxy-2'-deoxyguanosine (19).

Polymorphisms in XRCC1, APE1 and hOGG1 have been shown to reduce the capacity to repair oxidative damage (20), which could result in an increased risk for accumulating DNA damage and developing diseases such as skin lesions or skin cancer. Recent human studies indicate possible associations between XRCC1 and hOGG1 polymorphisms and various cancers (13,21,22,23), although only XRCC1 has been associated specifically with basal cell and squamous cell carcinomas of the skin (24,25). APE1 has not yet been studied extensively in relation to human disease risk (26).

Limited research suggests that certain BER genetic polymorphisms influence skin cancer (24,27,28). Because DNA damage, particularly that caused by reactive oxygen species, has been shown to play a role in skin carcinogenesis (29), we investigated the association of four common polymorphisms in BER genes on the presence of arsenic-induced skin lesions in a large case–control study conducted in an arsenic-endemic region of Bangladesh. Specifically, we evaluated the combination of XRCC1 Arg399Gln (rs25487), XRCC1 Arg194Trp (rs1799782), hOGG1 Ser326Cys (rs1052133) and APE1 Asp148Glu (rs3136820) polymorphisms, including gene–environment interactions with arsenic exposure.


   Materials and methods
Top
 Abstract
 Introduction
 Materials and methods
Results
 Discussion
 References
 
Participant selection
Nine hundred case–control pairs were recruited from 23 villages within the Pabna district of Bangladesh that are serviced by the Pabna Community Clinic, a primary care satellite clinic of Dhaka Community Hospital. District residents were made aware of the study and asked to participate through a series of community meetings from 2001 to 2003. Two physicians, trained by a dermatologist in characterizing arsenic related skin lesions, screened volunteers in their homes to identify eligible cases and controls. Individuals were considered eligible cases if they resided within the Pabna Community Clinic catchment zone, were at least 16 years of age and were diagnosed with one or more types of skin lesions: diffuse/spotted keratosis, diffuse/spotted melanosis, hyperkeratosis, leukomelanosis or squamous cell carcinoma. A subset of lesions, including all suspected squamous cell carcinomas, were histologically confirmed. Controls were individuals selected from the same communities and did not have any visible skin lesions. One control was selected per case and matched on gender, age (within 3 years) and area of residence. In most cases, the control was selected from the same village as the case, but when a matching control in the same village could not be found a neighboring village was chosen. At the time of recruitment, all participants underwent a clinic or in-home visit, completed a behavioral and demographic questionnaire administered by a trained interviewer and provided a blood, toenail and water sample.

The study was designed to investigate effect modifiers of the association between arsenic exposure and skin lesions, not main effects of arsenic. Previous studies have described in detail the main effects of arsenic exposure from drinking water and the risk of skin lesions,(30,31,32) but few have investigated factors that could influence susceptibility to arsenic toxicity. Approximately 80% of controls were selected from suspected ‘low-exposure’ arsenic (<50 µg/l) communities and 20% of the controls were from suspected ‘high-exposure’ (>50 µg/l) areas. This distribution was chosen to reflect the reported background distribution of wells in Pabna (33) and to ensure heterogeneity of exposure, which enables better investigation of effect modification (34). Enrolled individuals determined to have arsenic exposure >50 µg/l, the Bangladeshi health standard, were notified and counseled on the health hazards by the staff of the Arsenic Mitigation Program, a governmental program.

Institutional Review Boards at the Harvard School of Public Health and Dhaka Community Hospital approved the protocol for this study. Informed consent was obtained prior to participation.

Data collection and analysis
Toenail clippings were collected and prepared as described by Chen et al. (35). Total arsenic was measured using inductively coupled plasma mass spectrometry (ICP-MS Model 6100 DRC, PerkinElmer, Norwalk, CT) and each sample was subjected to five replicate analyses. Instrument performance and the digestion process were validated using standard reference material water (NIST 1643d and NIST 1643e Trace Elements in Water; National Institute of Standards and Technology, Gaithersburg, MD) and certified human hair reference material (CRM Hair; Shanghai Institute of Nuclear Research, Academia Sinica, China) as described previously (36).

Toenails provide an easily measurable and valid biomarker of arsenic exposure. Several studies have demonstrated a strong association between drinking water arsenic exposure and arsenic concentration in toenails. In a previous study in this population in Bangladesh, we found a non-linear association between drinking water arsenic and toenail arsenic. A 1.6-fold increase in toenail arsenic concentration was observed for every 10-fold increase in drinking water arsenic above 2 µg As/l. No association was present when drinking water arsenic concentrations were below 2 µg As/l (37). Similar results were also observed from a study in New Hampshire, in which toenail arsenic concentration increased 2-fold for every 10-fold increase in drinking water arsenic concentration when drinking water concentrations were above 1 µg As/l. (38).

DNA from 1800 cases and controls was available for genotyping. DNA was extracted from whole blood using the Puregene DNA Isolation kit (Gentra Systems, Minneapolis, MN). XRCC1 Arg399Gln, XRCC1 Arg194Trp, hOGG1 Ser326Cys and APE1 Asp148Glu polymorphisms were detected by the Taqman method using the ABI Prism 7900HT Sequence Detection System (Applied Biosystems, Foster City, CA). Primers, probes and reaction conditions are available upon request. Five percent of the samples were randomly selected and subjected to repeat analysis to verify genotyping procedures. Two researchers independently reviewed all genotyping results; the final concordance rates were 96% for XRCC1 Arg194Trp, APE1 Asp148Glu and hOGG1 Ser326Cys and 98% for XRCC1 Arg399Gln. Discordant pairs (n = 25) were excluded from all analyses.

Statistical analysis
Physical and sociodemographic characteristics of the cases and controls were compared using the chi-square test for categorical data and t-test for comparison of means. Toenail arsenic concentrations were not normally distributed; thus, median arsenic values were compared using the Wilcoxon rank sum test. Allele frequencies, genotype frequencies and linkage disequilibrium between the two XRCC1 polymorphisms were determined using SAS Proc Allele. Conformity of genes to Hardy–Weinberg equilibrium (HWE) was tested using an exact test (SAS Proc Allele).

Analyses were conducted using conditional logistic regression. Odds ratios, the ratio of odds ratios for the interaction (RORint) and 95% confidence intervals (CIs) were computed for the parameters of interest. The functional form of toenail arsenic concentrations was explored using penalized splines and found to be linear on the log scale (39,40). Therefore, values were subsequently transformed to their common logarithms. The four polymorphisms in the BER pathway were first analyzed for their associations with arsenic-induced skin lesions in individual models (main effects models 1a–e) and then in a single model that adjusted for the following potential confounders: the four polymorphisms, body mass index (BMI), education, ever smoked biri or cigarettes and ever chewed betel nuts (main effects model 2). Self-reported measures of skin reactivity including childhood tan level and childhood skin reaction to 2 hours of sun exposure, as well as number of hours exposed to the sun while working, were also evaluated for potential confounding.

Subset analyses of main effects by the following types of skin lesions were performed: melanosis, leukomelanosis and keratosis. Keratosis was defined as any diffuse or spotted lesion characterized by hard and roughened skin observed on the palm or dorsum of the hands or the sole or plantar of the foot. Melanosis was defined as any diffuse or spotted lesion characterized by darkening of color on the face, oral cavity, neck, upper and lower limbs, chest or back. Hyperkeratosis was defined as extensively thickened keratosis easily visible from a distance. Leukomelanosis was defined as depigmentation characterized by black and white spots present anywhere on the body. There were too few cases of hyperkeratosis or skin cancer to analyze separately. The four polymorphisms were also analyzed along with their gene–arsenic interactions to discern the most important predictors of skin lesions within the pathway. For interaction terms, genes were categorized into three levels (wild-type, heterozygote and homozygous variant) and arsenic was treated as a continuous variable. Because the frequency of the XRCC1 194Trp/Trp genotype was low (1.4%), analyses were also run in which the heterozygote and homozygote variant genotypes were combined.

In addition, a joint effects model was constructed in which the logarithms of toenail arsenic concentrations were categorized into tertiles. These tertiles corresponded to toenail arsenic ranges of 0.11–1.29 µg/g, 1.29–3.98 µg/g and 3.98–97.3 µg/g.

A case-only analytic approach was used to confirm the presence of gene–arsenic interactions (41,42). In the case-only analysis, genotypes were collapsed such that the homozygous variant and heterozygous genotype were combined into one category and the wild-type genotype was used as the reference group. Unconditional logistic regression was used to calculate the case-only interaction risk ratio (RRint) and corresponding 95% CI for each gene by regressing genotype on toenail arsenic treated as a continuous variable. The RRint was calculated as the exponent of the beta coefficient for the arsenic variable and was compared with the RORint calculated from the beta coefficient of the interaction term in the case–control analysis. The statistical programs SAS version 9.1.2 (SAS Institute, Cary, NC) and R version 2.0.1 (R Foundation for Statistical Computing, Vienna, Austria) were used for the analyses.


   Results
Top
 Abstract
 Introduction
 Materials and methods
 Results
Discussion
 References
 
The participation rate for the cases and controls was 98.7%. The primary reasons for non-participation were fear of stigmatization, fear of giving biological samples and disbelief that the water was harmful. Of the 1800 participating cases and controls, 1701 (95%) were genotyped successfully for all four BER genotypes. Complete information on toenail arsenic, BMI, education, smoking and betel nut chewing was available for 1684 of these participants, one of whom was excluded due to an implausible computed BMI of 60. This resulted in 99 case–control pairs in which either case or control information was missing. Thus, 792 complete matched pairs (1584 individuals) were available for analysis.

Cases were more likely than controls to have a lower education level and were less likely to smoke cigarettes (Table I). As expected, median toenail arsenic values were higher in cases than in controls. No meaningful differences between cases and controls were found with respect to BMI, chewing betel nuts or smoking biri.


View this table:
[in this window]
[in a new window]

  		Table I. Physical and social characteristics of 1584 cases and controls in Pabna, Bangladesh

 
The XRCC1 and hOGG1 polymorphisms were consistent with HWE in the case and control populations. The two XRCC1 polymorphisms were in linkage disequilibrium (D = 0.44, P < 0.0001). The APE1 polymorphism departed from HWE among the controls (P = 0.02). The overall allelic frequency for APE1 148Glu did not vary between cases and controls, though it was much lower than frequencies reported in other populations (43,44,45). However, controls were more likely than cases to have the heterozygote genotype and less likely to have the homozygous variant Glu/Glu or the wild-type Asp/Asp genotype.

Crude (models 1a–e) and adjusted (model 2) odds ratios showing the effect of toenail arsenic and each genotype on skin lesions are displayed in Table II. After adjusting for covariates and other genes, a 10-fold increase in toenail arsenic concentration was associated with a 7.35-fold increase in odds of skin lesions (95% CI 5.17, 10.45). Individuals with the APE1 148Glu homozygous variant had a nearly 2-fold increased odds of skin lesions compared with those homozygous for the 148Asp allele (odds ratio 1.93; 95% CI 1.15, 3.19). No associations were observed for the XRCC1 and hOGG1 polymorphisms. When the XRCC1 194Trp/Trp and Trp/Arg genotypes were combined into a single categorical variable, similar results were observed. Furthermore, within model 2, subgroup analyses were performed to determine whether the effect of the APE1 148Glu homozygous variant was specific to a type of skin lesion (e.g. keratosis, melanosis, etc). Although statistical power was lacking, a comparison of point estimates suggested that the effect of the APE1 148Glu homozygous variant was greatest for individuals with melanosis (OR 1.75; 95% CI 0.75, 4.11), followed by those with leukomelanosis (OR 1.28; 95% CI 0.57, 2.85) and keratosis (OR 0.87; 95% CI 0.20, 3.79). There were too few individuals with hyperkeratosis and squamous cell carcinoma to analyze separately.


View this table:
[in this window]
[in a new window]

  		Table II. Results from conditional logistic regression models that evaluated the main effects of BER genotypes on the presence of skin lesions (N = 1584)

 
Several gene–environment interactions were also evaluated. Table III illustrates the interaction between the genotypes, XRCC1 Arg194Trp and APE1 Asp148Glu, with toenail arsenic. A 10-fold increase in toenail arsenic concentration translated to an odds ratio for skin lesions of 2.49 (95% CI 0.34, 17.89) for the XRCC1 194Trp homozygous variant, 4.24 (95% CI 2.02, 8.93) for the heterozygote and 9.49 (95% CI 5.01, 17.99) for the wild-type genotype (Table III). A model in which the homozygous variant and heterozygote genotypes were combined yielded similar results (data not shown). When toenail arsenic was categorized into tertiles and evaluated jointly with genotype (Table IV), the odds ratio for skin lesions was highest among individuals with the XRCC1 wild-type genotype and the highest tertile of toenail arsenic concentration. Within the highest tertile, the variant genotypes conferred a much smaller odds ratio compared with the wild-type genotype. These results suggest that the presence of skin lesions was decreased in individuals possessing at least one variant allele as arsenic concentrations increased.


View this table:
[in this window]
[in a new window]

  		Table III. Results from a single conditional logistic regression model evaluating the interaction between genetic polymorphisms and toenail arsenic concentration

 


View this table:
[in this window]
[in a new window]

  		Table IV. Odds ratiosa and 95% CIs for the joint effects of having APE1 Asp148Glu or XRCC1 Arg194Trp polymorphisms and tertiles of toenail arsenic values (N = 1584)

 
A marginal interaction was also observed between APE1 and toenail arsenic concentration. The odds ratio for skin lesions for a 10-fold change in toenail arsenic was 7.08 (95% CI 2.18, 23.01) for the APE1 148Glu homozygous variant, 5.62 (95% CI 2.95, 10.69) for the heterozygote and 9.49 (95% CI 5.01, 17.99) for the wild-type genotype. However, since there remained a highly significant negative main effect for the APE1 148Glu/Glu variant in the interaction model, the results are best interpreted by evaluating the overall effect of jointly having the variant APE1 genotype and various levels of toenail arsenic (Table IV). For APE1, the most notable differences in odds ratios were observed at low levels of arsenic, where individuals with the APE1 homozygous variant had a greater odds ratio compared with the heterozygote or wild-type genotypes. The largest odds ratios for skin lesions uniformly occurred in the highest tertile of toenail arsenic; however, these odds ratios did not differ much by genotype.

Lastly, the case-only analyses confirmed the presence of gene–arsenic interactions observed in the case–control analyses. The RRint estimates suggested strong gene–arsenic interactions for XRCC1 194Trp/Trp + Arg/Trp (RRint 0.61; 95% CI 0.44, 0.85) and APE1 148GluGlu + Asp/Glu (RRint 0.57; 95% CI 0.43, 0.74) genotypes compared with wild-type. These were similar to the case–control results for XRCC1 194Trp/Trp + Arg/Trp (RORint 0.42; 95% CI 0.23, 0.78) and APE1 148GluGlu + Asp/Glu (RORint 0.62; 95% CI 0.37, 1.03).


   Discussion
Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
References
 
This study provides epidemiological evidence that the BER genes APE1 and XRCC1 affected the odds of skin lesions for individuals living in an arsenic-endemic environment. While odds of skin lesions increased with toenail arsenic concentrations for all individuals regardless of genotype, the XRCC1 194Trp allele offered some protection. When arsenic concentrations were low, the odds of skin lesions were generally low, and few differences were observed between genotypes. However, as arsenic concentrations increased, individuals with at least one XRCC1 194Trp variant allele had lower odds of skin lesions compared with individuals with the wild-type genotype.

One possible explanation for this observation is that XRCC1 is known to interact with several enzymes in the BER pathway and, since the Arg194Trp polymorphism resides in the linker region separating the POL ß domain from the poly (ADP-ribose) polymerase (PARP)-interacting domain, the polymorphism could affect XRCC1’s ability to bind to either POL ß or poly (ADP-ribose) polymerase (12). If this resulted in decreased DNA repair ability, it could lead to increased accumulation of unrepaired DNA damage that enhances apoptosis, thereby reducing the probability of the cell cycle replicating mutated DNA (24). Since arsenic exposure is genotoxic, individuals with the variant allele and high arsenic levels may have damaged skin cells that apoptose more readily, making the variant protective. However, experimental evidence in support of functional changes of XRCC1 Arg194Trp is limited (45,46,47,48).

The association could also be explained if XRCC1 was differentially down-regulated as a result of arsenic exposure. While this mechanism is currently unknown, arsenic has been shown to down-regulate expression of several DNA repair genes in a time- and dose-dependent manner (28). If arsenic down-regulates transcription of the XRCC1 wild-type genotype more effectively than the XRCC1 heterozygote or homozygous variant, then the variant genotype would be protective because DNA repair would be inhibited to a greater extent in individuals with the wild-type genotype particularly at high arsenic levels.

Despite the lack of knowledge regarding XRCC1 Arg194Trp function, human evidence of a protective effect of the 194Trp allele is growing. Several reviews and a recent meta-analysis suggest that the 194Trp allele is protective against cancers (13,21,26), although the only study to evaluate an association between the Arg194Trp polymorphism and skin cancer found no effect (49). However, two other studies identified positive associations between the XRCC1 Arg399Gln polymorphism and skin cancer (24,25). In the present study, neither did we observe any association between XRCC1 Arg399Gln and skin lesions nor did we find evidence for a gene–environment interaction with arsenic exposure.

At low to moderate levels of toenail arsenic, an increased odds ratio for skin lesions was observed among individuals with the APE1 148Glu variant compared with wild-type. This could be a result of a general defect in BER capacity because individuals with the APE1 variant allele may have a compromised ability to repair DNA, resulting in genetic instability and increased frequency of mutation even at very low levels of arsenic. As arsenic levels increase, arsenic-induced DNA damage is also likely to increase and may overwhelm the BER pathway such that it is less important which APE1 genotype is present because neither is capable of repairing the accumulating damage. This would explain the observed APE1–arsenic interaction in which genotype seems to play a more important role at low arsenic levels than at high arsenic levels.

This observation is supported by in vitro and in vivo animal experiments. Functional studies on the APE1 polymorphism show that the 148Glu allele may alter endonuclease and DNA-binding activity and reduce ability to communicate with other BER proteins (50). Bacteria deficient in APE1 activity exhibit elevated sensitivity to oxidizing and alkylating agents and elevated spontaneous mutation frequencies (51). The loss of a single APE1 allele in mice increased their predisposition to UV radiation-induced skin cancer (51). In addition, RNA-interference experiments of APE1 function demonstrate that excess damage could lead to the accumulation of unrepaired oxidative lesions, abasic damage and POL ß-mediated cross-link formation (52). Thus, cells may need to maintain high levels of APE1 to handle the routine load of DNA damage generated by metabolic processes. With regard to individuals chronically exposed to arsenic, reduced expression or function of APE1 could result in more DNA damage that goes unrepaired, leading to abberant skin cells and the development of skin lesions.

Despite evidence from animal models that suggests mutations in the hOGG1 gene increase susceptibility to skin carcinogenesis (27), no evidence supports an association with skin cancer in humans (23). Similarly, we observed no association between the hOGG1 326Cys allele and increased odds of skin lesions. Two other mechanisms exist that repair the oxidized base 8-oxo-7,8-dihydroguanine (12), which could compensate for any decrease in hOGG1 function, making it impossible to observe effects of the hOGG1 polymorphism.

While the principal analyses reported in this study resulted from conditional logistic regression models using data from a case–control population, we also conducted a case-only analysis to evaluate the gene–environment interactions with greater statistical power (41,42,53). These results support the conclusions drawn from the traditional case–control analyses. For both genes, the direction of the estimates was consistent, though the magnitudes varied. For APE1, the case-only CI was also much narrower, providing stronger indirect evidence for a gene–environment interaction.

Several limitations of the present study warrant consideration. This analysis only examined one or two common polymorphisms in three of the many genes involved in the BER pathway. Thus, other polymorphisms or other BER genes may actually be responsible for the observed effects. Furthermore, BER is just one method for DNA repair and it is possible that genes in other DNA repair mechanisms such as nucleotide excision repair may also affect the occurrence of skin lesions. Nor can we rule out the possibility of confounding by sun exposure since UV exposure generates DNA damage and has been implicated as a cocarcinogen with arsenic on the development of skin lesions (6,54). We attempted to control for measures of sun exposure by evaluating three variables in our data set: childhood tan level, childhood skin reaction to two hours of sun exposure and number of hours exposed to the sun while working. While these variables provide only a crude measure of skin type and sun exposure, including them in our model did not appreciably change our results.

In the case–control analysis, APE1 Asp148Glu departed from HWE among the controls. However, in the case-only analysis, which met HWE, the APE1 interaction with arsenic was observed to be significant and supported the conclusions drawn from the case–control analysis. Although the case-only design relies on the assumption of independence between genotype and environmental exposure, we believe this assumption is robust since it is unlikely that individuals would know their BER genotypes or modify their exposure based upon this knowledge.

The outcomes evaluated in this study were a heterogeneous pool of skin lesions. Consequently, our results may be obscured or misleading if the true effect of arsenic and BER polymorphisms is only for one specific type of lesion. Subset analyses conducted to address this issue suggested that the main genetic effects were similar for individuals with melanosis or leukomelanosis but may differ for keratosis. However, these analyses lacked power in general and could neither be conducted for individuals with hyperkeratosis nor gene–environment interactions be evaluated. Lastly, the odds ratios and CIs reported in the current study were not adjusted for multiple comparisons and should be validated in future research.

In conclusion, arsenic-induced skin lesions are an increasingly common health burden in Bangladesh and yet the etiology of their origin is not well understood. Individuals may have inherently different odds for developing skin lesions based in part on their genetic profile for BER and their arsenic exposure history. We observed that two genes in the BER pathway, APE1 and XRCC1, were associated with arsenic-induced skin lesions. Whereas only 6% of the population was homozygous for the high-risk APE1 variant allele, 78% of the population possessed the high-risk XRCC1 wild-type 194Trp/Trp genotype. Therefore, future research on arsenic-induced skin lesions should consider the impact of genetically susceptible subpopulations.


   Acknowledgments
 
The authors thank our colleagues, technicians, laboratory and administrative staff at Dhaka Community Hospital and the Pabna Community Clinic in Bangladesh. We also acknowledge the academic assistance of Tom Smith and Paul Catalano, and the technical expertise of Janna Frelich, Lia Shimada, Ian James, Li Su, Ema Rodrigues and Meredith Jones. This publication was made possible by National Institutes of Health grants T32 ES07069, ES011622, ES05947 and ES00002. Its contents are solely the responsibility of the authors and do not necessarily represent the official views of the National Institutes of Health.

Conflict of Interest Statement: None declared.


   References
Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 

  1. Bates MN, et al. Arsenic ingestion and internal cancers: a review. Am. J. Epidemiol. (1992) 135:462–476.[Free Full Text]
  2. Szymanska-Chabowska A, et al. Some aspects of arsenic toxicity and carcinogenicity in living organism with special regard to its influence on cardiovascular system, blood and bone marrow. Int. J. Occup. Med. Environ. Health (2002) 15:101–116.[Medline]
  3. Centeno JA, et al. Pathology related to chronic arsenic exposure. Environ. Health Perspect. (2002) 110((suppl. 5)):883–886.[ISI][Medline]
  4. Khan MM, et al. Magnitude of arsenic toxicity in tube-well drinking water in Bangladesh and its adverse effects on human health including cancer: evidence from a review of the literature. Asian Pac. J. Cancer Prev. (2003) 4:7–14.[Medline]
  5. Yoshida T, et al. Chronic health effects in people exposed to arsenic via the drinking water: dose-response relationships in review. Toxicol. Appl. Pharmacol. (2004) 198:243–252.[CrossRef][ISI][Medline]
  6. Yu HS, et al. Arsenic carcinogenesis in the skin. J. Biomed. Sci. (2006) 13:657–666.[CrossRef][ISI][Medline]
  7. Huang C, et al. Molecular mechanisms of arsenic carcinogenesis. Mol. Cell Biochem. (2004) 255:57–66.[CrossRef][ISI][Medline]
  8. Hartwig A, et al. Interaction of arsenic(III) with nucleotide excision repair in UV-irradiated human fibroblasts. Carcinogenesis (1997) 18:399–405.[Abstract/Free Full Text]
  9. Schwerdtle T, et al. Arsenite and its biomethylated metabolites interfere with the formation and repair of stable BPDE-induced DNA adducts in human cells and impair XPAzf and Fpg. DNA Repair (Amst.) (2003) 2:1449–1463.[CrossRef][Medline]
  10. Witkiewicz-Kucharczyk A, et al. Damage of zinc fingers in DNA repair proteins, a novel molecular mechanism in carcinogenesis. Toxicol. Lett. (2006) 162:29–42.[CrossRef][ISI][Medline]
  11. Snow ET, et al. Arsenic, mode of action at biologically plausible low doses: what are the implications for low dose cancer risk? Toxicol. Appl. Pharmacol. (2005) 207:557–564.[CrossRef][Medline]
  12. Frosina G. Commentary: DNA base excision repair defects in human pathologies. Free Radic. Res. (2004) 38:1037–1054.[CrossRef][ISI][Medline]
  13. Hu Z, et al. XRCC1 polymorphisms and cancer risk: a meta-analysis of 38 case-control studies. Cancer Epidemiol. Biomarkers Prev. (2005) 14:1810–1818.[Abstract/Free Full Text]
  14. Cooke MS, et al. DNA repair is responsible for the presence of oxidatively damaged DNA lesions in urine. Mutat. Res. (2005) 574:58–66.[ISI][Medline]
  15. Banerjee M, et al. Polymorphism in the ERCC2 codon 751 is associated with arsenic-induced premalignant hyperkeratosis and significant chromosome aberrations. Carcinogenesis (2006) doi:10.1093/carcin/bgl181.
  16. Ding W, et al. Inorganic arsenic compounds cause oxidative damage to DNA and protein by inducing ROS and RNS generation in human keratinocytes. Mol. Cell Biochem. (2005) 279:105–112.[CrossRef][ISI][Medline]
  17. Shi H, et al. Oxidative mechanism of arsenic toxicity and carcinogenesis. Mol. Cell Biochem. (2004) 255:67–78.[CrossRef][ISI][Medline]
  18. Caldecott KW. XRCC1 and DNA strand break repair. DNA Repair (Amst.) (2003) 2:955–969.[CrossRef][Medline]
  19. Shibutani S, et al. Insertion of specific bases during DNA synthesis past the oxidation-damaged base 8-oxodG. Nature (1991) 349:431–434.[CrossRef][Medline]
  20. Vodicka P, et al. Association of DNA repair polymorphisms with DNA repair functional outcomes in healthy human subjects. Carcinogenesis (2007) 28:657–664.[Abstract/Free Full Text]
  21. Goode EL, et al. Polymorphisms in DNA repair genes and associations with cancer risk. Cancer Epidemiol. Biomarkers Prev. (2002) 11:1513–1530.[Abstract/Free Full Text]
  22. Weiss JM, et al. Polymorphic variation in hOGG1 and risk of cancer: a review of the functional and epidemiologic literature. Mol. Carcinog. (2005) 42:127–141.[CrossRef][ISI][Medline]
  23. Vogel U, et al. No association between OGG1 Ser326Cys and risk of basal cell carcinoma. Cancer Epidemiol. Biomarkers Prev. (2004) 13:1680–1681.[Free Full Text]
  24. Nelson HH, et al. The XRCC1 Arg399Gln polymorphism, sunburn, and non-melanoma skin cancer: evidence of gene-environment interaction. Cancer Res. (2002) 62:152–155.[Abstract/Free Full Text]
  25. Han J, et al. Genetic variation in XRCC1, sun exposure, and risk of skin cancer. Br. J. Cancer (2004) 91:1604–1609.[ISI][Medline]
  26. Hung RJ, et al. Genetic polymorphisms in the base excision repair pathway and cancer risk: a HuGE review. Am. J. Epidemiol. (2005) 162:925–942.[Abstract/Free Full Text]
  27. Kunisada M, et al. 8-Oxoguanine formation induced by chronic UVB exposure makes Ogg1 knockout mice susceptible to skin carcinogenesis. Cancer Res. (2005) 65:6006–6010.[Abstract/Free Full Text]
  28. Hamadeh HK, et al. Coordination of altered DNA repair and damage pathways in arsenite-exposed keratinocytes. Toxicol. Sci. (2002) 69:306–316.[Abstract/Free Full Text]
  29. Nishigori C, et al. Role of reactive oxygen species in skin carcinogenesis. Antioxid. Redox. Signal. (2004) 6:561–570.[CrossRef][ISI][Medline]
  30. Ahsan H, et al. Associations between drinking water and urinary arsenic levels and skin lesions in Bangladesh. J. Occup. Environ. Med. (2000) 42:1195–1201.[CrossRef][ISI][Medline]
  31. Rahman M, et al. Prevalence of arsenic exposure and skin lesions. A population based survey in Matlab, Bangladesh. J. Epidemiol. Community Health (2006) 60:242–248.[Abstract/Free Full Text]
  32. Fewtrell L, et al. An estimation of the global burden of disease due to skin lesions caused by arsenic in drinking water. J. Water Health (2005) 3:101–107.[Medline]
  33. British Geological Survey. Arsenic contamination of the groundwater. In: British Geologic Survey and Bangladesh Department of Public Health and Engineering—Kinniburgh DG, Smedley PL, eds. (2001) Vol. II. Keyworth, UK.
  34. Greenland S. Basic problems in interaction assessment. Environ. Health Perspect. (1993) 101((suppl. 4)):59–66.
  35. Chen KL, et al. Determination of total arsenic concentrations in nails by inductively coupled plasma mass spectrometry. Biol. Trace Elem. Res. (1999) 67:109–125.[ISI][Medline]
  36. Breton CV, et al. Gender-specific protective effect of hemoglobin on arsenic-induced skin lesions. Cancer Epidemiol. Biomarkers Prev. (2006) 15:902–907.[Abstract/Free Full Text]
  37. Kile M, et al. Toenail arsenic concentrations, GSTT1 gene polymorphisms, and arsenic exposure from drinking water. Cancer Epidemiol. Biomarkers Prev. (2005) 14:2419–2426.[Abstract/Free Full Text]
  38. Karagas MR, et al. Measurement of low levels of arsenic exposure: a comparison of water and toenail concentrations. Am. J. Epidemiol. (2000) 152:84–90.[Abstract/Free Full Text]
  39. Ruppert D, et al. Semiparametric Regression (2003) Cambridge, UK: Cambridge University Press.
  40. Hastie TJ, et al. Generalized Additive Models (1990) Chapman and Hall, New York.
  41. Piegorsch WW, et al. Non-hierarchical logistic models and case-only designs for assessing susceptibility in population-based case-control studies. Stat. Med. (1994) 13:153–162.[ISI][Medline]
  42. Khoury MJ, et al. Nontraditional epidemiologic approaches in the analysis of gene-environment interaction: case-control studies with no controls! Am. J. Epidemiol. (1996) 144:207–213.[Abstract/Free Full Text]
  43. Ito H, et al. Gene-environment interactions between the smoking habit and polymorphisms in the DNA repair genes, APE1 Asp148Glu and XRCC1 Arg399Gln, in Japanese lung cancer risk. Carcinogenesis (2004) 25:1395–1401.[Abstract/Free Full Text]
  44. Popanda O, et al. Specific combinations of DNA repair gene variants and increased risk for non-small cell lung cancer. Carcinogenesis (2004) 25:2433–2441.[Abstract/Free Full Text]
  45. Hu JJ, et al. Amino acid substitution variants of APE1 and XRCC1 genes associated with ionizing radiation sensitivity. Carcinogenesis (2001) 22:917–922.[Abstract/Free Full Text]
  46. Wang Y, et al. From genotype to phenotype: correlating XRCC1 polymorphisms with mutagen sensitivity. DNA Repair (Amst.) (2003) 2:901–908.[CrossRef][Medline]
  47. Zienolddiny S, et al. Polymorphisms of DNA repair genes and risk of non-small cell lung cancer. Carcinogenesis (2006) 27:560–567.[Abstract/Free Full Text]
  48. Lunn RM, et al. XRCC1 polymorphisms: effects on aflatoxin B1-DNA adducts and glycophorin A variant frequency. Cancer Res. (1999) 59:2557–2561.[Abstract/Free Full Text]
  49. Winsey SL, et al. A variant within the DNA repair gene XRCC3 is associated with the development of melanoma skin cancer. Cancer Res. (2000) 60:5612–5616.[Abstract/Free Full Text]
  50. Hadi MZ, et al. Functional characterization of Ape1 variants identified in the human population. Nucleic Acids Res. (2000) 28:3871–3879.[Abstract/Free Full Text]
  51. Wilson DM 3rd, et al. The major human abasic endonuclease: formation, consequences and repair of abasic lesions in DNA. Mutat. Res. (2001) 485:283–307.[ISI][Medline]
  52. Demple B, et al. Molecular and biological roles of Ape1 protein in mammalian base excision repair. DNA Repair (Amst.) (2005) 4:442–449.
  53. Botto LD, et al. Commentary: facing the challenge of gene-environment interaction: the two-by-four table and beyond. Am. J. Epidemiol. (2001) 153:1016–1020.[Abstract/Free Full Text]
  54. Rossman TG, et al. Arsenite is a cocarcinogen with solar ultraviolet radiation for mouse skin: an animal model for arsenic carcinogenesis. Toxicol. Appl. Pharmacol. (2001) 176:64–71.[CrossRef][ISI][Medline]
Received November 28, 2006; revised January 23, 2007; accepted March 15, 2007.


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?



This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
28/7/1520    most recent
bgm063v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (2)
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Breton, C. V.
Right arrow Articles by Christiani, D. C.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Breton, C. V.
Right arrow Articles by Christiani, D. C.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?