we evaluated the risk association between six common SNPs of the XRCC1 gene and ... Conclusions: The XRCC1 194Trp variant allele may be associated with ...
THYROID Volume 19, Number 2, 2009 ª Mary Ann Liebert, Inc. DOI: 10.1089=thy.2008.0153
Association of XRCC1 Polymorphisms and Risk of Differentiated Thyroid Carcinoma: A Case–Control Analysis Tang Ho,1,2 Guojun Li,1,3 Jiachun Lu,3 Chong Zhao,1 Qingyi Wei,3 and Erich M. Sturgis1,3
Background: Numerous single-nucleotide polymorphisms (SNPs) of the DNA repair gene XRCC1 have been described. These SNPs have been increasingly studied in the epidemiology of various cancer types. In this study we evaluated the risk association between six common SNPs of the XRCC1 gene and differentiated thyroid carcinoma (DTC). Methods: We conducted a case–control study of 251 subjects with DTC, 145 subjects with benign thyroid disease, and 503 cancer-free controls. Polymerase chain reaction–restriction fragment length polymorphism assays were performed for genotyping. Multivariate logistic regression analysis was performed for risk estimation. Expectation-maximization algorithm and bayesian methods were used to estimate haplotype frequencies. Results: Multivariate analysis demonstrated an increased risk of DTC for the Arg194Trp heterozygous polymorphic (CT) genotype (odds ratio [OR]: 1.4, 95% confidence interval [CI]: 0.9–2.1). Multivariate analysis demonstrated a decreased risk of DTC for the Arg399Gln homozygous polymorphic (AA) genotype (OR: 0.5, 95% CI: 0.3–0.8) and the polymorphic (A) allele (OR: 0.7, 95% CI: 0.5–1.0). Compared to the most commonly observed haplotype (CGTCGA), multiple haplotypes were associated with a significantly increased risk of DTC, with the CGTTGG haplotype demonstrating the strongest association (OR: 5.0, 95% CI: 1.9–13.2). Conclusions: The XRCC1 194Trp variant allele may be associated with increased risk of DTC, while the XRCC1 399Gln variant allele may be associated with decreased risk of DTC. The utility of XRCC1 haplotypes in predicting DTC risk deserves further investigation with direct haplotype measurement.
Introduction
D
NA damage occurs through various pathways, including exposure to endogenously produced reactive oxygen species and exogenous carcinogens. If unrepaired, genetic damage can lead to cell death or uncontrolled proliferation and contribute to the carcinogenesis of various malignancies. Among the different DNA repair mechanisms is base excision repair, which guards primarily against singlestrand DNA damage resulting from cellular metabolism. The X-ray repair cross-complementing group 1 (XRCC1) gene participates in base excision repair by encoding a protein that aids in the repair of single-strand DNA breaks in conjunction with a number of other proteins, including poly (ADP-ribose) polymerase, DNA ligase III, and DNA polymerase beta (1). It has also been shown that XRCC1 is required for the efficient repair of single-strand DNA damage (2). Furthermore, experimental evidence has suggested that cells lacking the
XRCC1 gene product are hypersensitive to exogenous insults such as ionizing radiation and alkylating agents (3). Numerous single-nucleotide polymorphisms (SNPs) of the XRCC1 gene have been described, and they have been increasingly studied in the epidemiology of various cancer types (4,5). However, the functional significance of these SNPs remains largely unknown. Thyroid carcinoma is the most frequently encountered endocrine malignancy in the United States; nearly 30,000 new cases were expected in 2006 (6). The majority of thyroid carcinomas are differentiated thyroid carcinomas (DTCs), a category that includes the pathologic subtypes of papillary, follicular, and Hu¨rthle cell carcinoma (7). Limited and often controversial risk factors for DTC have been reported, involving dietary habits, ethnicity, and exposure to various potentially carcinogenic agents; however, exposure to ionizing radiation at a young age remains the only clear risk factor for DTC to date (8–10).
1
Department of Head and Neck Surgery, The University of Texas M. D. Anderson Cancer Center, Houston, Texas. Bobby R. Alford Department of Otolaryngology–Head and Neck Surgery, Baylor College of Medicine, Houston, Texas. 3 Department of Epidemiology, The University of Texas M. D. Anderson Cancer Center, Houston, Texas. 2
129
130
HO ET AL.
In this molecular epidemiological case–control study, we compared the genotype frequency distributions of six common (minor allele frequencies 5%) XRCC1 SNPs in patients with DTC and cancer-free controls. Subjects with benign thyroid gland lesions (benign thyroid disease [BTD]) were included as an intermediate-risk comparison group. Three of the SNPs we studied are located in the XRCC1 promoter region (C310T, 539del542, and T1915C), and three nonsynonymous SNPs are located within the XRCC1 coding region (Arg399Gln, Arg194Trp, and Arg280His). We estimated the risk of DTC associated with these six SNPs using a multivariate regression analysis model. We also estimated haplotype frequencies and associated DTC risk using statistical methods.
classified as smokers, and those who had quit smoking more than 1 year prior to enrollment in the study were classified as former smokers. Subjects who used alcohol at least once a week for more than 1 year were classified as drinkers, and those who had quit such alcohol use more than 1 year prior to enrollment were classified as former drinkers. Previous radiation exposure was defined as previous whole-body or head-and-neck-specific irradiation. After institutional review board–approved informed consent was obtained, each participating subject donated 20 mL of blood for cell culture and DNA extraction.
Materials and Methods
From each blood sample, a leukocyte cell pellet was obtained from the buffy coat by centrifugation of 1 mL of whole blood. The pellet was used for genomic DNA extraction with the QIAGEN DNA Blood Mini Kit (QIAGEN Inc., Valencia, CA) according to the manufacturer’s instructions. A polymerase chain reaction–restriction fragment length polymorphism (PCR-RFLP) assay was used to amplify the XRCC1 regions containing the polymorphisms of interest. The XRCC1 SNPs and restriction enzyme used for each SNP are listed in Table 1. The fragments were amplified separately but under the same conditions in a 10-mL reaction mixture containing 15 ng of genomic DNA, 2.0 pmol of each primer, 0.1 mL of 10 mM dNTP mixture, 0.3 mL of 50 mM MgCl2, 1.0 mL of 10PCR buffer (50 mM KCl, 10 mM Tris HCl [pH 9.0 at 258C], 0.1% Triton X-100), 0.25 U of Taq DNA polymerase (Denville Scientific Inc., Metuchen, NJ), and 7.15 mL of double-distilled H2O. The PCR-RFLP profile consisted of an initial denaturing step of 958C for 5 minutes, followed by 35 cycles at 958C for 30 seconds, 598C for 30 seconds, 728C for 30 seconds, and a final elongation step of 728C for 5 minutes. The PCR products were verified on a 2% NuSieve agarose gel (Cambrex, Inc., Rockland, ME), and then stored in 108C incubators. All of the restriction enzymes used were from New England Biolabs (Ipswich, MA). The restriction enzyme HhaI was used to distinguish the XRCC1 C310T promoter SNP, in which the loss of a HhaI restriction site occurs with the polymorphic T allele. The restriction enzyme MspI was used to distinguish the XRCC1 539del542 promoter SNP, in which the loss of a MspI restriction site occurs with the deletion polymorphism. The restriction enzyme BsrBI was used to distinguish the XRCC1 T1915C promoter SNP, in which the loss of a BsrBI restriction site occurs with the polymorphic C allele. The restriction enzyme PvuII
Study subjects We conducted a tertiary cancer center–based case–control study, which was approved by our institutional review board. Case subjects were recruited between November 1999 and December 2005 from patients who presented to the Head and Neck Surgery Clinic at The University of Texas M. D. Anderson Cancer Center for evaluation of thyroid gland masses and who subsequently underwent surgical excision. Final diagnosis was determined by histopathologic examination of the resected specimen. Patients with anaplastic thyroid carcinoma, medullary thyroid carcinoma, lymphoma of the thyroid, and mucoepidermoid carcinoma of the thyroid were excluded from the study. Patients with BTD on final histopathologic review were included as an intermediate-risk comparison group. Control subjects were cancer-free visitors to our institution who had participated as control subjects in a molecular epidemiological study of head and neck squamous cell carcinoma ongoing during approximately the same time period, from November 1996 to March 2005. Any control subject with a past history of cancer (except nonmelanoma skin carcinoma) was excluded from the study. The final genotype analysis included a total of 251 patients with DTC, 145 patients with BTD, and 503 cancer-free control subjects. All subjects provided demographic data, information about their exposure to carcinogens, and information about their family’s cancer history via a self-administered questionnaire. Ethnicity was categorized as non-Hispanic white or other. A positive family history of cancer was defined as reportedly having any first-degree relative with a history of cancer except for nonmelanoma skin carcinoma. Subjects who had smoked more than 100 cigarettes in their lifetimes were
Genotyping of XRCC1 SNPs
Table 1. XRCC1 Single Nucleotide Polymorphisms (GenBank Accession Number AF512504) Region
Genomic position
Nucleotide change
Reference SNP ID
Amino acid change
Reported allele frequency
Restriction enzymea
Promoter Promoter Promoter Exon 6 Exon 9 Exon 10
310 539..542 1915 24049 25211 25897
C?T G deletion T?C C?T G?A G?A
rs2682585 N=A rs3213245 rs1799782 rs25489 rs25487
N=A N=A N=A Arg194Trp Arg280His Arg399Gln
0.13 0.22 0.22 0.12 0.10 0.23
HhaI MspI BsrBI PvuII RsaI NciI
a Restriction fragment length polymorphism fragment length in base pairs: C310T (CC: 111, 24; TT: 135; CT: 135, 111, 24); 539del542 (GG: 153, 85; OO: 228; GO: 228, 153, 85); T1915C (TT: 151, 92; CC: 243; TC: 243, 151, 92); Arg194Trp (CC: 485; TT: 396, 89; CT: 485, 396, 89); Arg280His (GG: 120, 57; AA: 57; GA: 177, 120, 57); Arg399Gln (GG: 461, 287, 132; AA: 593, 287; GA: 593, 461, 287, 132). SNP, single nucleotide polymorphism.
XRCC1 AND RISK OF DIFFERENTIATED THYROID CARCINOMAS was used to distinguish the XRCC1 C24049T SNP in exon 6, in which the gain of a PvuII restriction site occurs with the polymorphic T allele. The restriction enzyme RsaI was used to distinguish the XRCC1 G25211A SNP in exon 9, in which the loss of a RsaI restriction site occurs with the polymorphic A allele. The restriction enzyme NciI was used to distinguish the XRCC1 G25897A SNP in exon 10, in which the loss of a NciI restriction site occurs with the polymorphic A allele. The PCR products (4 mL) were digested using 4.0 U of the restriction enzyme with 1 mL of the 10buffer supplied with the restriction enzyme in a 10-mL reaction mixture incubated at 378C overnight. Digested product was then separated on a 2% NuSieve agarose gel and photographed using the Digital Imaging System (Model IS-1000; Alpha Innotech Co., San Leandro, CA). Statistical analysis Two-sided chi-squared tests were used to calculate the difference in frequency between cases and controls in age, sex, ethnicity, family history of cancer, tobacco use, alcohol use, and radiation exposure. Univariate analysis using twosided chi-squared tests was done to compare the genotype and allele frequencies between case and control subjects. The observed genotype frequencies were compared with those calculated from Hardy–Weinberg equilibrium theory ( p2 þ 2pq þ q2 ¼ 1, p ¼ frequency of the variant allele, q ¼ 1 p). The odds ratios (ORs) and their 95% confidence in-
131
tervals (CIs) were calculated by logistic regression analyses with or without adjustment for age, sex, ethnicity, family history of thyroid cancer, tobacco use, alcohol use, and radiation exposure. For the haplotype analysis, we used both the HAPLOTYPE procedure in SAS=Genetics software (Version 9.0; SAS Institute Inc., Cary, NC) and the PHASE program to reconstruct the haplotypes for each subject on the basis of the observed genotypes in the studied subjects. The procedure HAPLOTYPE applies the expectation-maximization algorithm to generate maximum likelihood estimates of haplotype frequencies (11). The PHASE program uses a bayesian Markov chain–Monte Carlo algorithm (12,13). The haplotypes resulting from SAS=Genetics and PHASE were compared and found to be consistent. Haplotype frequency comparisons between cases and controls (frequency 5%) were performed using chisquared tests. Associations between the reconstructed haplotypes and cancer risks were evaluated using logistic regression models in SAS=Genetics software. ORs representing the risk per copy of each carried haplotype were adjusted for age, sex, ethnicity, tobacco use, alcohol use, radiation exposure, and family history of cancer. All statistical tests were two-sided, and p < 0.05 was considered statistically significant. Results Age, sex, ethnicity, family history of cancer, tobacco use, alcohol use, and radiation exposure distributions for case and
Table 2. Demographic and Exposure Characteristics of Case and Control Subjects DTC (n ¼ 251) Variable Age, years 50 >50 Sex Male Female Ethnicity Non-Hispanic whites Other Family history of cancerb Yes No Tobacco usec Current Former Never Alcohol used Current Former Never Radiatione exposure No Yes a
n
%
161 90
64.1 35.9
86 165
34.3 65.7
174 77
69.3 30.7
123 128
49.0 51.0
22 59 170
8.8 23.5 67.7
81 25 145
32.3 10.0 57.8
246 5
98.0 2.0
BTD (n ¼ 145)
p value
a
n
%
73 72
50.3 49.7
31 114
21.4 78.6
112 33
77.2 22.8
76 69
52.4 47.6
29 28 88
20.0 19.3 60.7
53 18 74
36.6 12.4 51.0
139 6
95.9 4.1
0.930
Controls (n ¼ 503) a
p value
n
%
321 182
63.8 36.2
240 263
47.7 52.3
393 110
78.1 21.9
225 278
44.7 55.3
100 114 289
19.9 22.7 57.5
173 73 257
34.4 14.5 51.1
487 16
96.8 3.2
0.003
