Tobacco smoke produces oxidative and alkylative DNA damage that necessitates repair by base excision repair coordinated by X-ray cross-complementing gene 1 (XRCC1). We investigated whether polymorphisms in XRCC1 alter DNA repair capacity and modify breast cancer risk associated with smoking. To show the functionality of the 280His variant, we evaluated single-strand break (SSB) repair capacity of isogenic Chinese hamster ovary cells expressing human forms of XRCC1 after exposure to hydrogen peroxide (H2O2), methyl methanesulfonate (MMS), or camptothecin by monitoring NAD(P)H. We used data from the Carolina Breast Cancer Study (CBCS), a population-based, case-control study that included 2,077 cases (786 African Americans and 1,281 Whites) and 1,818 controls (681 African Americans and 1,137 Whites), to examine associations among XRCC1 codon 194, 280, and 399 genotypes, breast cancer, and smoking. Odds ratios and 95% confidence intervals (95% CI) were calculated by unconditional logistic regression. Only cells expressing the 280His protein accumulated SSB, indicated by NAD(P)H depletion, from both H2O2 and MMS exposures. In the CBCS, positive associations were observed between breast cancer and smoking dose for participants with XRCC1 codon 194 Arg/Arg (Ptrend = 0.046), 399 Arg/Arg (Ptrend = 0.012), and 280 His/His or His/Arg (Ptrend = 0.047) genotypes. The 280His allele was in strong linkage disequilibrium with 194Arg (Lewontin's D′ = 1.0) and 399Arg (D′ = 1.0). These data suggest that less common, functional polymorphisms may lie within common haplotypes and drive gene-environment interactions. (Cancer Res 2006; 66(5): 2860-8)

Polymorphisms in genes responsible for maintaining genomic integrity seem to be potential modifiers of disease risk (1, 2). Consequently, several laboratory (1) and epidemiologic (reviewed in ref. 3) investigations have attempted to show a link between polymorphic DNA repair genes and a variety of malignancies. With breast cancer being the most frequently diagnosed malignancy in women, there is enormous interest in showing whether chemical exposures, genetics, or a combination of both are among the risk factors for this disease (47). The role of cigarette smoking and breast cancer risk is controversial, with some epidemiologic studies showing positive associations, whereas others showed inverse associations or no association (8). A recent literature review concluded that breast cancer risk may be increased by smoking of long duration and by exposure to passive smoking also called environmental tobacco smoke (ETS; ref. 9). Additionally, functional and observational approaches have focused on interactions between polymorphisms of DNA repair genes and smoking (4, 6, 10) as an example of gene-environment interactions involved in the etiology of breast cancer.

X-ray cross-complementing gene 1 (XRCC1) acts as a scaffolding protein for the base excision repair (BER) and single-strand break repair (SSBR; refs. 11, 12). These overlapping pathways participate in the constitutive response to endogenous mutagens and exogenous exposures, including tobacco smoke. Specifically, XRCC1-mediated pathways repair damage to DNA bases, from oxidation or covalent binding of nonbulky electrophiles, and to the deoxyribose phosphate backbone. Quick resolution of this genetic damage is imperative because repair intermediates, such as abasic sites and SSB, are generally more genotoxic and cytotoxic than the initial lesion (13). Three common polymorphisms within the XRCC1 gene have been identified at codon 194, 280, and 399 (Arg194Trp, Arg280His, and Arg399Gln; ref. 14). These nonconservative amino acid changes may alter XRCC1 function. This change in protein biochemistry leads to the supposition that variant alleles may diminish repair kinetics, thereby influencing susceptibility to adverse health effects, including cancer (15).

Laboratory experiments and epidemiologic studies have failed to reach a consensus regarding the functional effects of XRCC1 polymorphisms (reviewed in ref. 16). Some laboratory investigations of XRCC1 codon 399 Gln functionality in human cells suggested that this polymorphism is associated with increased levels of DNA damage after exposure to various mutagens (1719). Other reports offered conflicting evidence, suggesting that the 399Gln polymorphism has no adverse effect on DNA repair (2022). The 194Trp variant protein does not seem to negatively alter the DNA repair capacity of human cells (18, 20). Functional studies using lymphocytes suggested that the 280His polymorphism diminishes genomic stability (20, 21).

In the present study, we further characterized and confirmed the ability of isogenic mammalian cells transfected with human XRCC1 cDNA to amend SSB caused by genotoxic stress. We directly assessed the functionality of the 280His and 399Gln variant proteins through their expression within EM9 cells, a theoretical XRCC1 knockout model (reviewed in ref. 23), and comparison with repair-proficient cells. The choice of chemicals for exposure, hydrogen peroxide (H2O2) and methyl methanesulfonate (MMS), qualitatively mimics some of the genotoxic events resulting from tobacco smoke exposure (i.e., DNA oxidation and purine alkylation by N-nitrosamines). As a result, we could infer how BER and SSBR capacity in humans would be affected by XRCC1 variants after exposure to tobacco smoke. Additionally, exposure to the topoisomerase I inhibitor camptothecin allowed for the novel functional evaluation of XRCC1 variants within tyrosyl DNA phoshodiesterase 1 (TDP1)–mediated pathways (24, 25). We then applied our observations to a population-based, case-control study to evaluate the hypothesis that the XRCC1 280His allele increases the risk of breast cancer from exposure to tobacco smoke. We found that combining the use of transgenic cells and a novel screening assay for DNA repair capacity with a traditional epidemiologic approach has proven to be an effective union for providing an increased understanding of gene-environment interactions.

Cell line preparation and cell culture. Preparation of EM9 cells expressing the human wild-type (EM9-WT), 280His (EM9-280His), or 399Gln (EM9-399Gln) variant proteins or an empty pCMV vector (EM9-V) and culture conditions were described previously (26, 27). After reaching 90% to 100% confluency, cells were harvested by trypsin (Sigma, St. Louis, MO) for subculturing or chemical exposure.

Chemicals. Unless noted, all chemicals used for cell exposures and the NAD(P)H assay were purchased from Sigma. MMS was obtained from Aldrich (Milwaukee, WI). Dosing and control solutions of chemicals were prepared with 1× PBS (pH 7.4; Invitrogen, Grand Island, NY).

Chemical exposures and NAD(P)H assay. Exposed cells were analyzed for an imbalance of SSBR by noninvasively monitoring intracellular NAD(P)H levels using a colorimetric assay (27) with modification. Briefly, before chemical exposures, cells were seeded onto a 96-well plate (5 × 103/50 μL/well) in DMEM/F-12 (Invitrogen) supplemented with 10% fetal bovine serum (FBS) and 1% penicillin and streptomycin (Invitrogen) for an overnight incubation. For continuous exposures (i.e., MMS and camptothecin), each well has been adjusted to a volume of 110 μL with complete medium, dye, and test chemical. For H2O2 exposures, cells were exposed to H2O2 for 30 minutes at 37°C after replenishment with 50 μL serumless DMEM/F-12. To quench the oxidation reactions, DMEM/F-12 containing 20% FBS, catalase (18 units/mL), and dye were added to each well to give a final volume of 110 μL.

NAD(P)H levels were then monitored as described previously (27). Statistical evaluation of functional assay data was preformed using SAS version 9.1 (SAS Institute, Cary, NC). Due to the approximate negative exponential decay with increasing dose, NAD(P)H values were log transformed for multiple linear regression. For each chemical exposure, two-sided t tests were done comparing the regression coefficients for the wild-type response and the regression coefficient of each of the other cell lines to determine statistical significance with an α level of 0.05.

Carolina Breast Cancer Study. The Carolina Breast Cancer Study (CBCS) is a population-based, case-control study of invasive and in situ breast cancer conducted in 24 counties of central and eastern North Carolina (28). Incident cases were identified using a Rapid Case Ascertainment System in cooperation with the North Carolina Central Cancer Registry. Controls were selected from Division of Motor Vehicles (women <65 years old) and U.S. Health Care Financing Administration lists (women ≥65 years old). In-person interviews were conducted to obtain blood samples and information on potential breast cancer risk factors (28, 29).

Cases of invasive breast cancer were enrolled in two phases (phase 1: 1993-1996, phase 2: 1996-2001) with oversampling of African American and younger women (30). Controls were frequency matched to cases based on age and race (±5 years) using randomized recruitment (31). Cases of in situ breast cancer were enrolled between 1996 and 2001 and included women with ductal carcinoma in situ (DCIS) and DCIS with microinvasion to a depth of 2 mm. All cases of in situ breast cancer were eligible, with no oversampling according to age or race. Controls were frequency matched to in situ cases based on age (±5 years) and race. Race was classified according to self-report. Less than 2% of participants reported Native American or other race and were classified as White.

A total of 1,803 cases (787 African Americans and 1,016 Whites) and 1,564 controls (718 African Americans and 846 Whites) were enrolled in the invasive study, and a total of 508 cases (107 African Americans and 401 Whites) and 458 controls (70 African Americans and 388 Whites) were enrolled in the in situ study. Contact and cooperation rates for the CBCS and characteristics of cases and controls have been published previously (30). Response rates for blood draws and obtaining DNA were 90% for cases and 90% for controls. DNA samples were available for a total of 2,077 cases (786 African Americans and 1,281 Whites) and 1,818 controls (681 African Americans and 1,137 Whites). Odds ratios (OR) for breast cancer risk factors did not differ significantly between persons who gave DNA and those who did not (data not shown). XRCC1 codon 194 and 399 results for part of phase 1 of the CBCS were published previously (6). The present results combine genotypes from the entire CBCS (phases 1 and 2 and in situ studies). Results did not differ for African Americans and Whites or for invasive and in situ disease, so results are combined to increase precision.

Genotype analysis. DNA was extracted from peripheral blood lymphocytes by standard methods using an automated ABI-DNA extractor (Nuclei Acid Purification System, Applied Biosystems, Foster City, CA) in the University of North Carolina Specialized Program of Research Excellence (SPORE) Tissue Procurement Facility. Genotyping was conducted using the ABI 7700 Sequence Detection System or “Taqman” assay (Applied Biosystems). The following loci were genotyped: XRCC1 codon 194 (rs 1799782), 280 (rs 25489), and 399 (rs 25487). Primer and probe sequences as well as annealing temperatures for each genotyping assay are listed in Supplementary Table. Probes were labeled on the 5′ end with either FAM or VIC (Applied Biosystems). Probes were labeled on the 3′ end with the quencher dye 6-carboxy-N, N, N′, N′-tetramethylrhodamine.

PCR reactions were done in 15 μL reaction volumes. Reactions contained 0.7× Universal Master Mix (Applied Biosystems), 200 nmol/L of each allele-specific probe, 900 nmol/L of each primer, and 15 ng genomic DNA. After reactions, tubes were set up and amplification was done using a Perkin-Elmer GenAmp 9700 thermocycler (Perkin-Elmer, Wellesley, MA). Reaction tubes were placed into the thermocycler after the temperature reached 50°C. PCRs were carried out using the following conditions: 50°C for 2 minutes (AmpErase UNG Activation), 95°C for 10 minutes (AmpliTaq Gold Activation), and 40 cycles of 92°C for 15 seconds (denature) and the temperature listed in Supplementary Table for 1 minute (anneal/extend). Samples that failed to amplify were repeated. Those samples that failed to amplify on the second run were scored as missing. Missing genotypes for each loci were as follows: XRCC1 codon 194 (22 cases and 1 control), 280 (41 cases and 8 controls), and 399 (72 cases and 20 controls). A 10% random sample of genotypes were repeated for each locus, and results were identical to the initial analysis. For each genotyping assay, DNA samples from the Coriell Tissue Repository (Coriell Institute for Medical Research, Camden, NJ) that have been sequenced previously at the National Cancer Institute (NCI; http://www.nci.snp500.gov) were used as positive controls.

Statistical methods. Departures from Hardy-Weinberg equilibrium were evaluated by calculating expected genotype frequencies among controls based on observed allele frequencies and comparing the expected frequencies to observed genotype frequencies using χ2 tests. Differences between allele or genotype frequencies in cases and controls were estimated using χ2 tests or Fisher's exact tests when expected counts were <5. Tests for statistical significance were two-sided with an α level of 0.05. SAS Genetics version 8.2 (SAS Institute) was used to estimate XRCC1 codon 194 + 280 + 399 haplotype frequencies and to compare haplotype frequencies in cases and controls. Haplotype estimates from SAS Genetics are based on the EM algorithm (32). Lewontin's D′ value, an estimate of the extent of linkage disequilibrium, was calculated using SAS Genetics for pair-wise comparisons of XRCC1 codon 194 and 280 and codon 280 and 399.

Unconditional logistic regression was used to calculate ORs for breast cancer and 95% confidence intervals (95% CI). PROC GENMOD in SAS version 8.2 (SAS Institute) was used to incorporate offsets derived from sampling probabilities used to identify eligible participants (31) and to adjust for race (African American and White) and age (as an 11-level ordinal variable that reflected 5-year age categories).

Analysis of smoking effects used a common reference group of women who were not exposed to active or passive smoking. Ever-active smokers were defined as women who smoked at least 100 cigarettes in their lifetime. Exposure to passive smoking was defined as living with a smoker after age 18 (ETS after 18). Women who smoked on the reference date (date of diagnosis for cases or date of selection for controls) were classified as current smokers, whereas those women who no longer smoked on the reference date were designated former smokers. Women were asked about the amount of cigarettes smoked (packs per day) and the duration of smoking (the total number of years the participant smoked regularly). Information on duration of smoking was obtained by asking participants, “Keeping in mind that you may have stopped and started several times, overall how many years have you smoked regularly?” Dose of smoking was obtained by asking, “On average, how many cigarettes did you smoke per day?” ORs for smoking dose and duration were calculated separately for current smokers and former smokers, and these groups were combined in the present analysis because positive associations were observed in both groups. Information regarding dose and duration of smoking was missing for three White cases. Multivariable logistic regression was used to adjust for potential confounding factors. Confounding was evaluated by determining whether adding a variable to a model resulted in a change in the β coefficient of at least 10% for the exposure of interest. The following confounding variables were identified for the association of smoking and breast cancer: age at menarche (<12, ≥12 years), a composite term for age at first full-term pregnancy and parity (nulliparous, parity = 1 and age at first full-term pregnancy <26, parity = 1 and age at first full-term pregnancy ≥26, parity ≥2 and age at first full-term pregnancy <26, parity ≥2 and age at first full-term pregnancy ≥26), family history of breast cancer (yes/no for first-degree relative), and alcohol consumption (never/ever). ORs for XRCC1 genotypes and breast cancer were unchanged after adjusting for smoking and the other covariates listed and thus are presented adjusted for offsets (sampling probabilities), age and race only. Participants with missing values for any of the variables in a regression model were omitted from the analysis.

Stratified analyses were used to investigate modification of ORs for smoking and breast cancer by XRCC1 genotype. ORs for smoking were calculated according to each XRCC1 genotype separately. In addition, we wished to estimate effects for XRCC1 codon 194 and 399 separately, while ignoring codon 280 genotype, to compare our results with previous epidemiologic studies of XRCC1. Tests for trend for smoking dose and duration were conducted by calculating Ps for the β coefficient in logistic regression models with smoking dose or duration coded as an ordinal variable. Results were similar for African American and White participants; therefore, only combined results are shown. The term “Any genotype” refers to one or more copies of the less common allele (e.g., XRCC1 codon 194). “Any Trp” refers to “Arg/Trp or Trp/Trp genotype.”

Interactions between XRCC1 genotypes and smoking on a multiplicative scale were evaluated using likelihood ratio tests (LRT). An α value of 0.20 was used for statistical significance to account for the lower power of the test (33).

Interactions on an additive scale were assessed by estimating independent and joint effects for XRCC1 genotypes and smoking using a single reference of never-smokers and low-risk XRCC1 genotype. Departures from additive effects were assessed using interaction contrast ratios (ICR). ICRs greater than zero imply greater than additive effects or synergy (34).

Data from our functional evaluation of the 280His polymorphism support the hypothesis that some polymorphisms in DNA repair genes can alter the efficiency of repair pathways. To determine the phenotypic response to oxidative stress, the transfected EM9 cell lines were exposed to H2O2 (Fig. 1A). At the concentration range tested, the EM9-WT cells showed no depletion in NAD(P)H. However, the EM9-V cells exhibited a dose-dependent decrease in NAD(P)H to levels less than 40% of controls at the highest dose. Likewise, the EM9-280His cells showed a similar response with a 50% decrease in NAD(P)H after 4 hours of recovery. EM9-399Gln cells showed a slight reduction in NAD(P)H levels (83% of control level) at the highest dose.

Figure 1.

Graphical representation of NAD(P)H data as an indirect indicator of SSB accumulation. EM9 cells expressing human forms of XRCC1, including wild-type (EM9-WT; ◊), 399Gln (EM9-399Gln; ▪), or 280His (EM9-280His; ○) polymorphisms or an empty vector (EM9-V; ▴), were exposed (A) for 30 minutes to H2O2, (B) continuously to MMS, or (C) continuously to camptothecin. NAD(P)H levels were monitored in real-time for 4 hours during (MMS and camptothecin) or after (H2O2) exposure. NAD(P)H data for exposed wells were percentage relative to NAD(P)H levels (100%) in corresponding control wells of the same cell line dosed with PBS. Chemical exposures were conducted in triplicate and repeated on different days. Points, mean; bars, SD. *, P < 0.05; **, P < 0.01, significant difference from wild-type line.

Figure 1.

Graphical representation of NAD(P)H data as an indirect indicator of SSB accumulation. EM9 cells expressing human forms of XRCC1, including wild-type (EM9-WT; ◊), 399Gln (EM9-399Gln; ▪), or 280His (EM9-280His; ○) polymorphisms or an empty vector (EM9-V; ▴), were exposed (A) for 30 minutes to H2O2, (B) continuously to MMS, or (C) continuously to camptothecin. NAD(P)H levels were monitored in real-time for 4 hours during (MMS and camptothecin) or after (H2O2) exposure. NAD(P)H data for exposed wells were percentage relative to NAD(P)H levels (100%) in corresponding control wells of the same cell line dosed with PBS. Chemical exposures were conducted in triplicate and repeated on different days. Points, mean; bars, SD. *, P < 0.05; **, P < 0.01, significant difference from wild-type line.

Close modal

During MMS exposures (Fig. 1B), we observed clear differences in terms of SSBR proficiency between the EM9 cell lines. After 4 hours of continuous exposure to MMS, EM9-280His cells showed greater NAD(P)H depletion than EM9-WT cells at a concentration as low as 62.5 μmol/L. These data strongly suggest that the efficient removal of alkylated bases and other repair intermediates may be hindered by the expression of the XRCC1 280His genotype. EM9-399Gln cells seemed to have a similar depletion of intracellular NAD(P)H as EM9-WT cells in the cellular response to alkylative stress. EM9-V cells showed a massive reduction in NAD(P)H, with only 30% of control levels at the highest dose.

To determine the influence of XRCC1 polymorphisms on interactions with proteins involved in a TDP1-mediated pathway, we exposed the transfected cell lines to the topoisomerase I inhibitor camptothecin (Fig. 1C). After 4 hours of continuous exposure, the EM9-WT, EM9-399Gln, and EM9-280His cell lines showed <10% decreases in NAD(P)H relative to controls, indicating no influences of XRCC1 genotypes on this repair pathway. The repair-deficient EM9-V cells showed a 25% decrease in NAD(P)H at the highest dose level.

Because data indicated that the XRCC1 280His variant was a functionally detrimental polymorphism, we evaluated XRCC1 genotype and smoking history data from the CBCS. Genotype frequencies, allelic frequencies, and ORs for breast cancer for XRCC1 codon 194, 280, and 399 genotypes are presented in Table 1. Allele and genotype frequencies were similar in African Americans and Whites and between cases and controls within each racial group with respect to XRCC1 codon 194 and 280. The frequency of the codon 399 Gln variant was greater in White controls (q = 0.35) than African American controls (q = 0.14). Genotypes for each XRCC1 locus were observed to be in Hardy-Weinberg equilibrium among African American cases, African American controls, White cases, and White controls (data not shown). For each locus, comparisons between the Arg/Arg genotype and the variant genotypes did not yield any statistically significant increases in ORs for breast cancer. Haplotype frequencies for XRCC1 in African American and Whites are presented in Table 2. Haplotype frequencies did not differ between cases and controls for either racial group. The 194Arg + 280Arg + 399Arg haplotype was the most common in both African Americans and Whites. The 280His allele was in strong linkage disequilibrium with 194Arg (D′ = 1.0) and 399Arg (D′ = 1.0) in both racial groups. Results for non–African Americans were not affected by exclusion of the 2% of participants who were non-White.

Table 1.

XRCC1 genotype frequencies, allele frequencies, and ORs for breast cancer from the CBCS

LocusAfrican Americans
Whites
CasesControlsOR* (95% CI)CasesControlsOR* (95% CI)
XRCC1 codon 194       
    Arg/Arg 671 (86.7) 593 (87.0) Reference 1,126 (87.9) 987 (87.0) Reference 
    Arg/Trp 101 (13.0) 86 (12.6) 1.0 (0.7-1.3) 148 (11.5) 141 (12.4) 0.9 (0.7-1.2) 
    Trp/Trp 2 (0.3) 3 (0.4) 0.7 (0.1-4.4) 7 (0.6) 7 (0.6) 0.9 (0.3-2.8) 
    Any Trp 103 89 1.0 (0.7-1.3) 155 148 0.9 (0.7-1.2) 
    Fisher's exact test  P = 0.83   P = 0.81  
XRCC1 codon 194§       
    Arg 0.93 0.93  0.94 0.93  
    Trp 0.07 0.07  0.06 0.07  
    Fisher's exact test  P = 0.97   P = 0.48  
XRCC1 codon 280       
    Arg/Arg 710 (92.8) 642 (94.3) Reference 1,146 (90.2) 1,030 (91.2) Reference 
    Arg/His 54 (7.1) 38 (5.5) 1.3 (0.8-2.0) 125 (9.8) 97 (8.6) 1.2 (0.9-1.6) 
    His/His 1 (0.1) 1 (0.2) 1.1 (0.1-18.0) 2 (0.2) ND 
    Any His 55 39 1.3 (0.8-1.9) 125 99  
    Fisher's exact test  P = 0.55   P = 0.17 1.2 (0.9-1.6) 
XRCC1 codon 280§       
    Arg 0.96 0.97  0.95 0.96  
    His 0.04 0.03  0.05 0.04  
    Fisher's exact test  P = 0.28   P = 0.47  
XRCC1 codon 399       
    Arg/Arg 536 (70.4) 493 (72.9) Reference 504 (40.5) 480 (42.8) Reference 
    Arg/Gln 203 (26.7) 172 (25.5) 1.1 (0.9-1.5) 581 (46.7) 494 (44.0) 1.1 (0.9-1.3) 
    Gln/Gln 22 (2.9) 11 (1.6) 1.8 (0.8-3.8) 159 (12.8) 148 (13.2) 1.0 (0.8-1.3) 
    Any Gln 225 183 1.2 (0.8-1.5) 740 642 1.1 (0.9-1.3) 
    χ2 test  P = 0.22   P = 0.42  
XRCC1 codon 399§       
    Arg 0.84 0.86  0.64 0.65  
    Gln 0.16 0.14  0.36 0.35  
    χ2 test  P = 0.16   P = 0.51  
LocusAfrican Americans
Whites
CasesControlsOR* (95% CI)CasesControlsOR* (95% CI)
XRCC1 codon 194       
    Arg/Arg 671 (86.7) 593 (87.0) Reference 1,126 (87.9) 987 (87.0) Reference 
    Arg/Trp 101 (13.0) 86 (12.6) 1.0 (0.7-1.3) 148 (11.5) 141 (12.4) 0.9 (0.7-1.2) 
    Trp/Trp 2 (0.3) 3 (0.4) 0.7 (0.1-4.4) 7 (0.6) 7 (0.6) 0.9 (0.3-2.8) 
    Any Trp 103 89 1.0 (0.7-1.3) 155 148 0.9 (0.7-1.2) 
    Fisher's exact test  P = 0.83   P = 0.81  
XRCC1 codon 194§       
    Arg 0.93 0.93  0.94 0.93  
    Trp 0.07 0.07  0.06 0.07  
    Fisher's exact test  P = 0.97   P = 0.48  
XRCC1 codon 280       
    Arg/Arg 710 (92.8) 642 (94.3) Reference 1,146 (90.2) 1,030 (91.2) Reference 
    Arg/His 54 (7.1) 38 (5.5) 1.3 (0.8-2.0) 125 (9.8) 97 (8.6) 1.2 (0.9-1.6) 
    His/His 1 (0.1) 1 (0.2) 1.1 (0.1-18.0) 2 (0.2) ND 
    Any His 55 39 1.3 (0.8-1.9) 125 99  
    Fisher's exact test  P = 0.55   P = 0.17 1.2 (0.9-1.6) 
XRCC1 codon 280§       
    Arg 0.96 0.97  0.95 0.96  
    His 0.04 0.03  0.05 0.04  
    Fisher's exact test  P = 0.28   P = 0.47  
XRCC1 codon 399       
    Arg/Arg 536 (70.4) 493 (72.9) Reference 504 (40.5) 480 (42.8) Reference 
    Arg/Gln 203 (26.7) 172 (25.5) 1.1 (0.9-1.5) 581 (46.7) 494 (44.0) 1.1 (0.9-1.3) 
    Gln/Gln 22 (2.9) 11 (1.6) 1.8 (0.8-3.8) 159 (12.8) 148 (13.2) 1.0 (0.8-1.3) 
    Any Gln 225 183 1.2 (0.8-1.5) 740 642 1.1 (0.9-1.3) 
    χ2 test  P = 0.22   P = 0.42  
XRCC1 codon 399§       
    Arg 0.84 0.86  0.64 0.65  
    Gln 0.16 0.14  0.36 0.35  
    χ2 test  P = 0.16   P = 0.51  

Abbreviation: ND, not determined.

*

Adjusted for offsets and age.

Genotype frequencies, n (%).

Comparing cases and controls.

§

Allele frequencies (95% CI).

Unstable estimate.

Table 2.

XRCC1 haplotype frequencies in African Americans and Whites

Codon 194Codon 280Codon 399CasesControlsχ2 test*
African Americans      
Arg (C) Arg (G) Arg (A) 0.73 0.76 P = 0.13 
Arg Arg Gln (G) 0.16 0.14 P = 0.15 
Arg His (A) Arg 0.04 0.03 P = 0.28 
Arg His Gln <0.001 <0.001 ND 
Trp (T) Arg Arg 0.07 0.07 P = 0.85 
Trp Arg Gln <0.001 <0.001 ND 
Trp His Arg <0.001 <0.001 ND 
Trp His Gln <0.001 <0.001 ND 
Whites      
Arg (C) Arg (G) Arg (A) 0.53 0.54 0.48 
Arg Arg Gln (G) 0.36 0.35 0.56 
Arg His (A) Arg 0.05 0.05 0.50 
Arg His Gln <0.0001 <0.0001 ND 
Trp (T) Arg Arg 0.07 0.07 0.78 
Trp Arg Gln <0.0001 <0.0001 ND 
Trp His Arg <0.0001 <0.0001 ND 
Trp His Gln <0.0001 <0.0001 ND 
Codon 194Codon 280Codon 399CasesControlsχ2 test*
African Americans      
Arg (C) Arg (G) Arg (A) 0.73 0.76 P = 0.13 
Arg Arg Gln (G) 0.16 0.14 P = 0.15 
Arg His (A) Arg 0.04 0.03 P = 0.28 
Arg His Gln <0.001 <0.001 ND 
Trp (T) Arg Arg 0.07 0.07 P = 0.85 
Trp Arg Gln <0.001 <0.001 ND 
Trp His Arg <0.001 <0.001 ND 
Trp His Gln <0.001 <0.001 ND 
Whites      
Arg (C) Arg (G) Arg (A) 0.53 0.54 0.48 
Arg Arg Gln (G) 0.36 0.35 0.56 
Arg His (A) Arg 0.05 0.05 0.50 
Arg His Gln <0.0001 <0.0001 ND 
Trp (T) Arg Arg 0.07 0.07 0.78 
Trp Arg Gln <0.0001 <0.0001 ND 
Trp His Arg <0.0001 <0.0001 ND 
Trp His Gln <0.0001 <0.0001 ND 

NOTE: African Americans and Whites (cases and controls combined): D′ = 1.0 for Arg194 + His280 and D′ = 1.0 for His280 + Arg399.

*

Comparing cases and controls.

Nucleotide in parentheses.

ORs for breast cancer and smoking stratified by XRCC1 codon 194 genotypes are presented in Table 3. Results are presented combining African Americans and Whites and are adjusted for race. Former smokers with the Arg/Arg genotype showed an increased risk for breast cancer (OR, 1.4; 95% CI, 1.1-1.7). Participants with the Arg/Arg genotype showed statistically significant trends for increasing breast cancer risk with increased smoking dose (P = 0.046) and duration (P = 0.017). LRTs were significant for the interaction of smoking duration with codon 194 genotype (Table 3). ICRs for codon 194 Arg/Arg genotype and smoking status, dose, and duration were 0.35, 0.39, and 0.39, respectively. No associations with smoking were observed for participants with 194 Any Trp genotype.

Table 3.

ORs for smoking and breast cancer according to XRCC1 codon 194 genotypes from the CBCS

Smoking statusXRCC1 codon 194
Arg/Arg
Any Trp
Cases/controlsOR (95% CI)*Cases/controlsOR (95% CI)*
Unexposed to active and passive smoking 345/310 Reference 55/45 Reference 
Passive smoking (ETS >18) 598/546 1.1 (0.9-1.3) 96/76 1.2 (0.7-2.0) 
Former active 559/424 1.4 (1.1-1.7) 72/66 1.0 (0.5-1.7) 
Current active 292/300 1.0 (0.8-1.2) 35/50 0.6 (0.3-1.1) 
LRT: P = 0.23     
Dose of active smoking (packs per day)     
    1/2 or less 273/248 1.1 (0.9-1.4) 31/42 0.8 (0.4-1.5) 
    1/2 to 1 334/257 1.2 (1.0-1.6) 44/35 0.9 (0.5-1.7) 
    1 or more 239/214 1.3 (1.0-1.6) 31/39 0.8 (0.4-1.5) 
    Trend test  P = 0.046  P = 0.24 
LRT: P = 0.27     
Duration of active smoking (y)     
    ≤10 222/199 1.1 (0.8-1.4) 28/41 0.6 (0.3-1.2) 
    11-20 204/192 1.1 (0.8-1.4) 29/24 1.1 (0.5-2.2) 
    >20 419/330 1.4 (1.1-1.7) 49/50 0.9 (0.5-1.7) 
    Trend test  P = 0.017  P = 0.53 
LRT: P = 0.16     
Smoking statusXRCC1 codon 194
Arg/Arg
Any Trp
Cases/controlsOR (95% CI)*Cases/controlsOR (95% CI)*
Unexposed to active and passive smoking 345/310 Reference 55/45 Reference 
Passive smoking (ETS >18) 598/546 1.1 (0.9-1.3) 96/76 1.2 (0.7-2.0) 
Former active 559/424 1.4 (1.1-1.7) 72/66 1.0 (0.5-1.7) 
Current active 292/300 1.0 (0.8-1.2) 35/50 0.6 (0.3-1.1) 
LRT: P = 0.23     
Dose of active smoking (packs per day)     
    1/2 or less 273/248 1.1 (0.9-1.4) 31/42 0.8 (0.4-1.5) 
    1/2 to 1 334/257 1.2 (1.0-1.6) 44/35 0.9 (0.5-1.7) 
    1 or more 239/214 1.3 (1.0-1.6) 31/39 0.8 (0.4-1.5) 
    Trend test  P = 0.046  P = 0.24 
LRT: P = 0.27     
Duration of active smoking (y)     
    ≤10 222/199 1.1 (0.8-1.4) 28/41 0.6 (0.3-1.2) 
    11-20 204/192 1.1 (0.8-1.4) 29/24 1.1 (0.5-2.2) 
    >20 419/330 1.4 (1.1-1.7) 49/50 0.9 (0.5-1.7) 
    Trend test  P = 0.017  P = 0.53 
LRT: P = 0.16     

NOTE: Data are combined for African Americans and Whites.

*

Adjusted for offsets, age, race, age at menarche, age at first full-term pregnancy/parity composite, family history, and alcohol.

Reference group.

ORs for smoking and breast cancer stratified by XRCC1 codon 280 genotype are presented in Table 4. A statistically significant positive association between passive and former smoking and breast cancer was observed for participants with the 280 Any His genotype. Although the test for trend was statistically significant only for smoking dose (P = 0.047), ORs were elevated for all levels of smoking dose and duration as well as current and former smoking and passive smoking. LRTs were significant for the interaction of smoking status, dose, and duration with XRCC1 codon 280 genotype (Table 4). ICRs for XRCC1 codon 280 Any His and smoking status, dose, and duration were 0.87, 0.65, and 0.80, respectively.

Table 4.

ORs for smoking and breast cancer according to XRCC1 codon 280 genotypes from the CBCS

Smoking statusXRCC1 codon 280
Arg/Arg
Any His
Cases/controlsOR (95% CI)*Cases/controlsOR (95% CI)*
Unexposed to active and passive smoking 371/315 Reference 28/39 Reference 
Passive smoking (ETS >18) 621/583 1.0 (0.8-1.2) 67/37 2.8 (1.4-5.7) 
Former active 563/450 1.2 (1.0-1.5) 58/38 3.0 (1.4-6.2) 
Current active 299/324 0.9 (0.7-1.1) 26/24 2.0 (0.9-4.6) 
LRT: P = 0.01     
Dose of active smoking (packs per day)     
    1/2 or less 271/268 1.0 (0.8-1.2) 28/21 2.5 (1.1-5.8) 
    1/2 to 1 333/269 1.1 (0.8-1.4) 38/23 2.9 (1.3-6.7) 
    1 or more 252/233 1.1 (0.9-1.4) 18/17 2.7 (1.0-6.8) 
    Trend test  P = 0.34  P = 0.047 
LRT: P = 0.02     
Duration of active smoking (y)     
    ≤10 223/223 0.9 (0.7-1.1) 23/16 2.7 (1.1-6.7) 
    11-20 209/199 1.0 (0.7-1.3) 19/16 2.2 (0.9-5.5) 
    >20 424/348 1.2 (1.0-1.6) 41/30 2.7 (1.2-6.1) 
    Trend test  P = 0.11  P = 0.08 
LRT: P = 0.03     
Smoking statusXRCC1 codon 280
Arg/Arg
Any His
Cases/controlsOR (95% CI)*Cases/controlsOR (95% CI)*
Unexposed to active and passive smoking 371/315 Reference 28/39 Reference 
Passive smoking (ETS >18) 621/583 1.0 (0.8-1.2) 67/37 2.8 (1.4-5.7) 
Former active 563/450 1.2 (1.0-1.5) 58/38 3.0 (1.4-6.2) 
Current active 299/324 0.9 (0.7-1.1) 26/24 2.0 (0.9-4.6) 
LRT: P = 0.01     
Dose of active smoking (packs per day)     
    1/2 or less 271/268 1.0 (0.8-1.2) 28/21 2.5 (1.1-5.8) 
    1/2 to 1 333/269 1.1 (0.8-1.4) 38/23 2.9 (1.3-6.7) 
    1 or more 252/233 1.1 (0.9-1.4) 18/17 2.7 (1.0-6.8) 
    Trend test  P = 0.34  P = 0.047 
LRT: P = 0.02     
Duration of active smoking (y)     
    ≤10 223/223 0.9 (0.7-1.1) 23/16 2.7 (1.1-6.7) 
    11-20 209/199 1.0 (0.7-1.3) 19/16 2.2 (0.9-5.5) 
    >20 424/348 1.2 (1.0-1.6) 41/30 2.7 (1.2-6.1) 
    Trend test  P = 0.11  P = 0.08 
LRT: P = 0.03     

NOTE: Data are combined for African Americans and Whites.

*

Adjusted for offsets, age, race, age at menarche, age at first full-term pregnancy/parity composite, family history, and alcohol.

Reference group.

Results for smoking and XRCC1 codon 399 genotype are presented in Table 5. ORs for former smoking and trend tests for smoking dose (P = 0.012) and duration (P = 0.001) were statistically significant among participants with codon 399 Arg/Arg genotype. LRTs were significant for the interaction of smoking duration with XRCC1 codon 399 genotype (Table 5). ICRs for XRCC1 codon 399 Arg/Arg genotype and smoking status, dose, and duration were 0.17, 0.32, and 0.39, respectively. There was no association among participants with codon 399 Any Gln genotype.

Table 5.

ORs for smoking and breast cancer according to XRCC1 codon 399 genotypes from the CBCS

Smoking statusXRCC1 codon 399
Arg/Arg
Any Gln
Cases/controlsOR (95% CI)*Cases/controlsOR (95% CI)*
Unexposed to active and passive smoking 186/192 Reference 200/158 Reference 
Passive smoking (ETS >18) 373/353 1.2 (0.9-1.5) 304/264 1.2 (0.9-1.3) 
Former active 309/254 1.5 (1.1-2.0) 313/230 1.2 (0.9-1.6) 
Current active 172/174 1.1 (0.8-1.6) 145/173 0.7 (0.5-1.0) 
LRT: P = 0.34     
Dose of active smoking (packs per day)     
    1/2 or less 158/165 1.2 (0.8-1.6) 139/120 1.0 (0.7-1.4) 
    1/2 to 1 197/145 1.5 (1.1-2.1) 175/144 1.0 (0.7-1.4) 
    1 or more 123/116 1.4 (1.0-2.1) 141/136 1.0 (0.7-1.4) 
    Trend test  P = 0.012  P = 0.73 
LRT: P = 0.26     
Duration of active smoking (y)     
    ≤10 119/136 1.0 (0.9-1.6) 128/100 1.1 (0.7-1.5) 
    11-20 123/110 1.4 (1.0-2.0) 104/103 0.8 (0.5-1.2) 
    >20 236/181 1.7 (1.2-2.3) 222/197 1.0 (0.7-1.4) 
    Trend test  P = 0.001  P = 0.79 
LRT: P = 0.05     
Smoking statusXRCC1 codon 399
Arg/Arg
Any Gln
Cases/controlsOR (95% CI)*Cases/controlsOR (95% CI)*
Unexposed to active and passive smoking 186/192 Reference 200/158 Reference 
Passive smoking (ETS >18) 373/353 1.2 (0.9-1.5) 304/264 1.2 (0.9-1.3) 
Former active 309/254 1.5 (1.1-2.0) 313/230 1.2 (0.9-1.6) 
Current active 172/174 1.1 (0.8-1.6) 145/173 0.7 (0.5-1.0) 
LRT: P = 0.34     
Dose of active smoking (packs per day)     
    1/2 or less 158/165 1.2 (0.8-1.6) 139/120 1.0 (0.7-1.4) 
    1/2 to 1 197/145 1.5 (1.1-2.1) 175/144 1.0 (0.7-1.4) 
    1 or more 123/116 1.4 (1.0-2.1) 141/136 1.0 (0.7-1.4) 
    Trend test  P = 0.012  P = 0.73 
LRT: P = 0.26     
Duration of active smoking (y)     
    ≤10 119/136 1.0 (0.9-1.6) 128/100 1.1 (0.7-1.5) 
    11-20 123/110 1.4 (1.0-2.0) 104/103 0.8 (0.5-1.2) 
    >20 236/181 1.7 (1.2-2.3) 222/197 1.0 (0.7-1.4) 
    Trend test  P = 0.001  P = 0.79 
LRT: P = 0.05     

NOTE: Data are combined for African Americans and Whites.

*

Adjusted for offsets, age, race, age at menarche, age at first full-term pregnancy/parity composite, family history, and alcohol.

Reference group.

In our functional evaluation of XRCC1 polymorphisms, we determined that relative to the wild-type protein the 280His variant decreased the DNA repair capacity of mammalian cells exposed to chemical stresses, such as oxidation and alkylation, associated with tobacco smoke. This observation that the 280His variant is functionally relevant guided the subsequent evaluation of data from the CBCS for a potential gene-environment interaction between XRCC1 genotypes and exposure to tobacco smoke. Several epidemiologic studies have implicated the XRCC1 399Arg allele in the etiology of bladder, head and neck, and lung cancer (3537). However, our functional assay showed that the XRCC1 399Arg allele seems to be functionally competent like the 194Arg allele as shown in previous studies (20, 26, 38), suggesting that another genetic modifier may be the causative factor that increases breast cancer risk. The less common 280His variant appears only within the 194Arg + 399Arg haplotype of XRCC1 (Table 2). Because ORs for smoking and breast cancer were stronger for XRCC1 codon 280 His-containing genotypes compared with codon 194 Arg/Arg or codon 399 Arg/Arg, our results suggest that codon 280 His is the relevant functional polymorphism in XRCC1 with respect to smoking and breast cancer. Other complex phenotypes have been shown to be influenced by less common and rare alleles within common haplotypes (39). These phenomena show the importance of investigating less common alleles that lie within common haplotypes in human populations.

Here, we showed that the XRCC1 280His variant attenuated the DNA repair capacity of transgenic cells after exposure to oxidative stress. Additionally, our functional evaluation substantiated a previous observation (26) that the 280His variant hinders the efficient repair of DNA damage from alkylative stress. These observations were evident from greater NAD(P)H depletions caused by poly(ADP-ribose) polymerase-1 overactivation in response to the accumulation of SSB. Relative to the EM9-WT and EM9-399Gln cells, NAD(P)H depletions in EM9-280His cells were greater after exposure to H2O2 or MMS (Fig. 1A and B), suggesting an inability to efficiently amend DNA damage. Because NAD(P)H depletion in EM9-399Gln cells was similar to that in EM9-WT cells, it seems that the 399Gln variant protein does not negatively affect XRCC1-mediated repair. When exposed to the topoisomerase I inhibitor camptothecin, all cell lines, excluding the repair-deficient EM9-V line, had levels of NAD(P)H near 100% of control levels (Fig. 1C). These data suggest that the functionality of XRCC1 polymorphisms is relevant only to the removal of damaged bases or frank SSB and not abortive topoisomerase I activity.

Prior functional studies in human and rodent cell models support our observations regarding the XRCC1 399Gln and 280His variants. Human cells with the 399Gln allele were not sensitive to bleomycin-induced DNA damage compared with lymphocytes with the codon 399 Arg/Arg genotype (20). Expression of the 399Gln variant protein in an EM9 background restored DNA repair capacity and cell survival to a level similar to that of EM9-WT cells after exposure to MMS (38). Lymphocytes from individuals carrying the 280His allele showed increased genetic damage from bleomycin exposure relative to 280 Arg/Arg homozygotes (20). The 280His polymorphism was also associated with increased chromosomal aberrations in lymphocytes (21). Although not assessed in this study, prior investigations of the Arg194Trp variant protein in human cells have shown that this protein does not alter DNA repair capacity from bleomycin exposure (20). Additionally, after MMS exposure, EM9 cell lines expressing the 194Trp variant protein (EM9-194Trp) as well as EM9-WT and EM9-399Gln cells responded similar to repair-proficient AA8 cells in terms of survival (26).

A decrease in DNA repair capacity precipitated by the 280His variant seems to be biologically plausible. The 280 codon of the XRCC1 polypeptide lies within the AP endonuclease (APE)–binding domain (11). The nonsynonymous Arg280His polymorphism causes the replacement of arginine with histidine, which changes the amino acid sequence of XRCC1. This change in protein biochemistry could potentially alter XRCC1 structure and its ability to interact with APE. The 280His protein only seems to have a negative effect during BER or SSBR induced by either base damage or DNA oxidation, processes that both involve APE. During the repair of SSB formed by camptothecin exposure, a process independent of APE activity, EM9-280His cells show a phenotypic response similar to that of EM9-WT cells (Fig. 1C). Additionally, when expressed in EM9 cells, the 280His variant protein failed to localize to DNA damage foci with the same efficiency as the wild-type protein (26).

The association of XRCC1 genotypes and breast cancer has been examined in 13 epidemiologic studies (5, 4051) in addition to a previous report from phase 1 of the CBCS (6). Only the CBCS included significant numbers of African Americans. For codon 194, positive associations were observed for Trp-containing genotypes in four studies (40, 46, 48, 49), an inverse association in one study (47), and no association in five studies (5, 42, 44, 50, 51). Increased risk for codon 280 His-containing genotypes was observed in one study (5) and no association in three studies (40, 43, 46). For codon 399, positive associations were observed for Gln-containing genotypes in three studies (40, 44, 46) and no association in nine studies (5, 4143, 45, 4851). A meta-analysis by Hung et al. (16) combined results from 10 breast cancer studies (5, 6, 4451). Summary ORs were close to the null for codon 194 and 399 genotypes and breast cancer (16). Three epidemiologic studies of breast cancer analyzed XRCC1 haplotypes (5, 47, 52), and results were consistent with the presence of the less common codon 280 His allele solely on the codon 194 Arg + codon 399 Arg chromosomal background.

Four breast cancer studies examined interactions between XRCC1 genotypes and smoking (4143, 47). Han et al. (47) reported a trend of increasing breast cancer risk with increasing duration of smoking among study participants with the codon 194 Arg/Arg genotype but not among codon 194 Trp carriers, consistent with the results of our study. For codon 399, Metsola et al. (43) and Shen et al. (42) reported interactions between Gln-containing genotypes and smoking, but no interactions were observed by Figueiredo et al. (41) and Han et al. (47). Metsola et al. (43) reported a stronger association for codon 280 His-containing genotypes and breast cancer among heavy smokers. A meta-analysis by Hung et al. (16) of tobacco-related cancers (lung, upper aerodigestive tract, bladder, stomach, liver, pancreas, and myeloid leukemia) found a protective effect for codon 194 Trp-containing genotypes among ever-smokers. Codon 399 Gln-containing genotypes were associated with increased risk of tobacco-related cancers among light smokers but a decreased risk among heavy smokers (16). These results are compatible with our observation of a stronger association between breast cancer and increased duration and dose of smoking among study participants with codon 194 Arg/Arg and codon 399 Arg/Arg genotypes. Hung et al. (16) reported a summary OR close to the null for codon 280 His-containing genotypes and tobacco-related cancers, but the data were too sparse to stratify on smoking history. For results of additional epidemiologic studies of XRCC1, see Hung et al. (16) and Goode et al. (3).

Evaluating gene-environment interactions using a transgenic cell system as a screen for functional polymorphisms has advantages over human cell-based functional assays. The use of isogenic EM9 cells expressing human XRCC1 protein allowed for direct functional characterization of variant proteins without confounding by other genetic modifiers. Additionally, we found this Chinese hamster ovary model to be preferable over genetically matched lymphocyte cell lines from cases and controls carrying the 280His allele, because human lines exhibit different rates of growth (data not shown), a potential source of confounding, and a concern for assay variability. The use of a sensitive, real-time NAD(P)H assay to assess BER/SSBR capacity afforded us the flexibility to reproducibly test several exposure scenarios in a relatively short amount of time. The stable transfection of plasmids harboring human cDNA of other polymorphic genes into isogenic knockout cells would extend the applicability of this approach. Our combined study design provides a robust examination of the biological significance for XRCC1 polymorphisms. The precise functional evaluation of XRCC1 polymorphisms through a laboratory study lends biological plausibility to the findings from an epidemiologic study of breast cancer susceptibility. The strategy could prove useful for clarifying the biological significance of other genetic polymorphisms in DNA repair genes, particularly those with low allelic frequencies.

In summary, we further characterized the functionality of the XRCC1 280His polymorphism and used these observations to clarify the relationship between this allele, breast cancer, and smoking. The XRCC1 codon 280 His allele is in linkage disequilibrium with the more common variants for two other XRCC1 polymorphisms at codon 194 and 399. Functional and epidemiologic data suggest that the XRCC1 codon 280 His allele may be more important than codon 194 or 399 alleles with respect to smoking and breast cancer. Haplotype analyses, particularly using anonymous tagSNPs, may prove useful for identifying genetic heterogeneity when functional alleles are unknown. However, identification of functionally relevant alleles within defined haplotypes, as presented here, will also contribute important information for understanding gene-environment and gene-gene interactions.

Note: Supplementary data for this article are available at Cancer Research Online (http://cancerres.aacrjournals.org/).

Grant support: SPORE in Breast Cancer grant NIH/NCI P50-CA58223, Lineberger Comprehensive Cancer Center grant NIH/NCI P30-CA16086, Center for Environmental Health and Susceptibility grant NIEHS P30-ES10126, and Superfund Basic Research Program grant NIEHS P42-ES05948; “Open Research Center” Project for Private Universities: Matching Fund Subsidy from Ministry of Education, Culture, Sports, Science and Technology, Japan, 2004-2008 (Y. Kubota); and EPA STAR Graduate Fellowship and NCI training grant T32-CA72319 (B.F. Pachkowski).

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

We thank Allison Eaton, Kendra Worley, Jon Player, Allan Rene de Cotret (University of North Carolina High-Throughput Genotyping Core Laboratory), and Daynise Skeen (University of North Carolina SPORE Tissue Procurement Facility) for technical assistance and Dr. Michael Symons for statistical consultation with the functional assay.

1
Berwick M, Vineis P. Markers of DNA repair and susceptibility to cancer in humans: an epidemiologic review.
J Natl Cancer Inst
2000
;
92
:
874
–97.
2
Vodicka P, Kumar R, Stetina R, et al. Genetic polymorphisms in DNA repair genes and possible links with DNA repair rate, chromosomal aberrations and single-strand breaks in DNA.
Carcinogenesis
2004
;
25
:
757
–63.
3
Goode EL, Ulrich CM, Potter JD. Polymorphisms in DNA repair genes and associations with cancer risk.
Cancer Epidemiol Biomarkers Prev
2002
;
11
:
1513
–30.
4
Kennedy DO, Agrawal M, Shen J, et al. DNA repair capacity of lymphoblastoid cell lines from sisters discordant for breast cancer.
J Natl Cancer Inst
2005
;
97
:
127
–32.
5
Moullan N, Cox DG, Angele S, Romestaing P, Gerard J-P, Hall J. Polymorphisms in the DNA repair gene XRCC1, breast cancer risk, and response to radiotherapy.
Cancer Epidemiol Biomarkers Prev
2003
;
12
:
1168
–74.
6
Duell EJ, Millikan RC, Pittman GS, et al. Polymorphisms in the DNA repair gene XRCC1 and breast cancer.
Cancer Epidemiol Biomarkers Prev
2001
;
10
:
217
–22.
7
Kadouri L, Kote-Jarai Z, Hubert A, et al. A single-nucleotide polymorphism in the RAD51 gene modifies breast cancer risk in BRCA2 carriers, but not in BRCA1 carriers or noncarriers.
Br J Cancer
2004
;
90
:
2000
–5.
8
Palmer JR, Rosenberg L. Cigarette smoking and the risk of breast cancer.
Epidemiol Rev
1993
;
15
:
145
–56.
9
Terry PD, Rohan TE. Cigarette smoking and the risk of breast cancer in women: a review of the literature.
Cancer Epidemiol Biomarkers Prev
2002
;
11
:
953
–71.
10
Shi Q, Wang L-E, Bondy ML, Brewster A, Singletary SE, Wei Q. Reduced DNA repair of benzo[a]pyrene diol epoxide-induced adducts and common XPD polymorphisms in breast cancer patients.
Carcinogenesis
2004
;
25
:
1695
–700.
11
Caldecott KW. XRCC1 and DNA strand break repair.
DNA Repair
2003
;
2
:
955
–69.
12
Caldecott KW. Protein-protein interactions during mammalian DNA single-strand break repair.
Biochem Soc Trans
2003
;
31
:
247
–51.
13
Sobol RW, Kartalou M, Almeida KH, et al. Base excision repair intermediates induce p53-independent cytotoxic and genotoxic responses.
J Biol Chem
2003
;
278
:
39951
–9.
14
Shen MR, Jone IM, Mohrenweiser H. Nonconservative amino acid substitution variants exist at polymorphic frequency in DNA repair genes in healthy humans.
Cancer Res
1998
;
58
:
604
–8.
15
Rebbeck TR, Ambrosone CB, Bell DA, et al. SNPs, haplotypes, and cancer: applications in molecular epidemiology.
Cancer Epidemiol Biomarkers Prev
2004
;
13
:
681
–7.
16
Hung R, Hall J, Brennan P, Boffetta P. Genetic polymorphisms in the base excision repair pathway and cancer risk: a HuGE review.
Am J Epidemiol
2005
;
162
:
925
–42.
17
Au WW, Salama SA, Sierra-Torres CH. Functional characterization of polymorphisms in DNA repair genes using cytogenetic challenge assay.
Environ Health Perspect
2003
;
111
:
1843
–50.
18
Wang Y, Spitz MR, Zhu Y, Dong Q, Shete S, Wu X. From genotype to phenotype: correlating XRCC1 polymorphisms with mutagen sensitivity.
DNA Repair (Amst)
2003
;
2
:
901
–8.
19
Lunn RM, Langlois RG, Hsieh LL, Thompson CL, Bell DA. XRCC1 polymorphisms: effects on aflatoxin B1-DNA adducts and glycophorin A variant frequency.
Cancer Res
1999
;
59
:
2557
–61.
20
Tuimala J, Szekely G, Gundy S, Hirvonen A, Norppa H. Genetic polymorphisms of DNA repair and xenobiotic-metabolizing enzymes: role in mutagen sensitivity.
Carcinogenesis
2002
;
23
:
1003
–8.
21
Kiuru A, Lindholm C, Heilimo I, et al. Influence of DNA repair gene polymorphisms on the yield of chromosomal aberrations.
Environ Mol Mutagen
2005
;
46
:
198
–205.
22
Pastorelli R, Cerri A, Mezzetti M, Consonni E, Airoldi L. Effect of DNA repair gene polymorphisms on BPDE-DNA adducts in human lymphocytes.
Int J Cancer
2002
;
100
:
9
–13.
23
Thompson LH, West MG. XRCC1 keeps DNA from getting stranded.
Mutat Res
2000
;
459
:
1
–18.
24
Barrows LR, Holden JA, Anderson M, D'Arpa P. The CHO XRCC1 mutant, EM9, deficient in DNA ligase III activity, exhibits hypersensitivity to camptothecin independent of DNA replication.
Mutat Res
1998
;
408
:
103
–10.
25
Park S-Y, Lam W, Cheng Y-C. X-ray repair cross-complementing gene I protein plays an important role in camptothecin resistance.
Cancer Res
2002
;
62
:
459
–65.
26
Takanami T, Nakamura J, Kubota Y, Horiuchi S. The Arg280His polymorphism in X-ray repair cross-complementing gene 1 impairs DNA repair ability.
Mutat Res
2005
;
582
:
135
–45.
27
Nakamura J, Asakura S, Hester SD, de Murcia G, Caldecott KW, Swenberg JA. Quantitation of intracellular NAD(P)H can monitor an imbalance of DNA single strand break repair in base excision repair deficient cells in real time.
Nucleic Acids Res
2003
;
31
:
e104
.
28
Newman B, Moorman PG, Millikan R, et al. The Carolina Breast Cancer Study: integrating population-based epidemiology and molecular biology.
Breast Cancer Res Treat
1995
;
35
:
51
–60.
29
Millikan RC, Pittman GS, Newman B, et al. Cigarette smoking, N-acetyltransferases 1 and 2, and breast cancer risk.
Cancer Epidemiol Biomarkers Prev
1998
;
7
:
371
–8.
30
Millikan R, Eaton A, Worley K, et al. HER2 codon 655 polymorphism and risk of breast cancer in African Americans and Whites.
Breast Cancer Res Treat
2003
;
79
:
355
–64.
31
Weinberg CR, Sandler DP. Randomized recruitment in case-control studies.
Am J Epidemiol
1991
;
134
:
421
–32.
32
Zhao J, Curtis D, Sham P. Model-free analysis and permutation tests for allelic associations.
Hum Hered
2000
;
50
:
133
–9.
33
Selvin S. A note on the power to detect interaction effects. In: Kesley J, Marmot M, Stolley P, Vessey M, editors. Statistical analysis of epidemiologic data. 2nd ed. Oxford: Oxford University Press; 1996. p. 213–4.
34
Rothman KJ, Greenland S. Modern epidemiology. 2nd ed. Philadelphia: Lippincott-Raven; 1998.
35
David-Beabes GL, London SJ. Genetic polymorphisms of XRCC1 and lung cancer risk among African-Americans and Caucasian.
Lung Cancer
2001
;
34
:
333
–9.
36
Stern MC, Umbach DM, van Gils CH, Lunn RM, Taylor JA. DNA repair gene XRCC1 polymorphisms, smoking, and bladder cancer risk.
Cancer Epidemiol Biomarkers Prev
2001
;
10
:
125
–31.
37
Olshan AF, Watson MA, Weissler MC, Bell DA. XRCC1 polymorphisms and head and neck cancer.
Cancer Lett
2002
;
178
:
181
–6.
38
Taylor RM, Thistlethwaite A, Caldecott KW. Central role for the XRCC1 BRCT I domain in mammalian DNA single-strand break repair.
Mol Cell Biol
2002
;
22
:
2556
–63.
39
Cohen JC, Kiss RS, Pertsemlidis A, Marcel YL, McPherson R, Hobbs HH. Multiple rare alleles contribute to low plasma levels of HDL cholesterol.
Science
2004
;
305
:
869
–72.
40
Sigurdson A, Hauptmann M, Chatterjee N, et al. Kin-cohort estimates for familial breast cancer risk in relation to variants in DNA base excision repair, BRCA1 interacting, and growth factor genes.
BMC Cancer
2004
;
4
:
9
.
41
Figueiredo J, Knight J, Briollais L, Andrulis L, Ozcelik H. Polymorphisms XRCC1-399Q and XRCC3-241M and the risk of breast cancer at the Ontario site of the Breast Cancer Family Registry.
Cancer Epidemiol Biomarkers Prev
2004
;
13
:
583
–91.
42
Shen J, Gammon M, Terry M, et al. Polymorphisms in XRCC1 modify the association between polcyclic aromatic hydrocarbon-DNA adducts, cigarette smoking, dietary antioxidants, and breast cancer risk.
Cancer Epidemiol Biomarkers Prev
2005
;
14
:
336
–42.
43
Metsola K, Kataja V, Sillanpaa P, et al. XRCC1 and XPD genetic polymorphisms, smoking and breast cancer risk in a Finnish case-control study.
Breast Cancer Res
2005
;
7
:
R987
–97.
44
Kim S-U, Park S, Yoo K-Y, et al. XRCC1 genetic polymorphism and breast cancer risk.
Pharmacogenetics
2002
;
12
:
335
–8.
45
Shu X-O, Cai Q, Gao Y-T, Wen W, Jin F, Zheng W. A population-based case-control study of the Arg399Gln polymorphism in DNA repair gene XRCC1 and risk of breast cancer.
Cancer Epidemiol Biomarkers Prev
2003
;
12
:
1462
–7.
46
Chacko P, Rajan B, Joseph T, Mathew BS, Pillai M. Polymorphisms in DNA repair gene XRCC1 and increased genetic susceptibility to breast cancer.
Breast Cancer Res Treat
2005
;
89
:
15
–21.
47
Han J, Hankinson S, DeVivo I, et al. A prospective study of XRCC1 haplotypes and their interaction with plasma carotenoids on breast cancer risk.
Cancer Res
2003
;
63
:
8536
–41.
48
Smith T, Miller S, Lohman K, et al. Polymorphisms of XRCC1 and XRCC3 genes and susceptibility to breast cancer.
Cancer Lett
2003
;
190
:
183
–90.
49
Smith T, Levine E, Perrier N, et al. DNA-repair genetic polymorphisms and breast cancer risk.
Cancer Epidemiol Biomarkers Prev
2003
;
12
:
1200
–4.
50
Forsti A, Angelini S, Festa F, et al. Single nucleotide polymorphisms in breast cancer.
Oncol Rep
2004
;
11
:
917
–22.
51
Deligezer U, Dalay N. Association of the XRCC1 gene polymorphisms with cancer risk in Turkish breast cancer patients.
Exp Mol Med
2004
;
36
:
572
–5.
52
Chang-Claude J, Popanda O, Tan X-Y, et al. Association between polymorphisms in the DNA repair genes, XRCC1, APE1, and XPD and acute side effects of radiotherapy in breast cancer patients.
Clin Cancer Res
2005
;
11
:
4802
–9.

Supplementary data