Introduction

Endometrial carcinoma (EC) is the most common tumor of the female genital tract in the Western world [1]. The great majority of EC cases are type I (estrogen-related), frequently showing microsatellite instability and mutations in the PTEN, PIK3CA, K-Ras and β-catenin genes. These mutations may reflect the genomic instability which is most common symptom of the cancer cells [2, 3]. This instability may be caused by a continuous exposure to genotoxic stress, including that evoked by estrogens, which can induce bulky DNA adducts and minor modifications to the DNA bases [4]. These lesions are removed by nucleotide and base excision repair (NER and BER), respectively. NER includes recognition of DNA damage by the RAD23B-XPC complex, followed by binding of the XPA/RPA dimer to the lesion. XPA is an important factor for accurate positioning of the ERCC1-ERCC4 (XPF) endonuclease. Two helicases ERCC3 (XPB) and ERCC2 (XPD) are responsible for unwinding the DNA helix, and the ERCC5 (XPG) and ERCC1-ERCC4 nucleases excise a single stranded DNA fragment containing the lesion. The remained gap is filled by DNA polymerase δ/ε and DNA ligase I using the intact strand as a template. The base excision repair (BER) pathway corrects most base modifications caused by reactive oxygen species (ROS). A damaged base is recognized by a specific glycosylase, which cleaves the bond between the base and sugar, creating an abasic site, which is cleaved by an endonuclease. Resulting gap is filled by polβ and the remaining nick is sealed by DNA ligase LIG1 or LIG3 complexed with XRCC1.

Because NER and BER are involved in removing a substantial number of DNA damages, which can contribute to the genome instability, it is reasonable to check whether variability in the genes coding for BER and NER products may be associated with EC. In the present work we searched for an association between EC and the variants of single nucleotide polymorphisms (SNPs) of the BER/NER genes: ERCC2, OGG1 and XRCC1. We studied 4 SNPs occurring in 3 BER and NER genes: p.Arg194Trp, p.Arg399Gln of the XRCC1 gene, p.Ser326Cys of the hOGG1 gene and p.Lys751Gln of the ERCC2 gene (rs1799782, rs25487, rs1052133 and rs13181 respectively). These polymorphisms have been correlated with various tumors, including lung, breast and skin cancers [5–13], but little is known about their association with EC.

Materials and Methods

Patients

Blood was obtained from 94 women (median age 48 years and median BMI 28) with EC treated in 2004–2006 at the Polish Mother’s Memorial Hospital (Lodz, Poland). All patients had histologically confirmed EC and agreed to complete a risk factor questionnaire. The characteristics of the subjects enrolled in this study are presented in Table 1. Control samples consisted of DNA extracted from blood cells from age-matched 114 cancer-free women. The study was approved by the Local Ethic Committee and each patient gave a written consent.

Table 1 Characteristics of the study population
Full size table

Genotype Determination

Genomic DNA was prepared using GeneMatrix Blood DNA purification Kit (EURx, Gdansk, Poland) according to the manufacturer instruction. Genotypes were determined by PCR-RFLP (polymerase chain reaction-restriction fragment length polymorphism). Genome regions that include studied polymorphisms were amplified by PCR using primers listed in Table 2. The PCR reaction (total volume 25 μl) was launched with a mixture containing 100 ng genomic DNA, 5 mM dNTPs, 5 pmol each primer and 1 U Taq DNA polymerase (Biotools, Madrid, Spain) which was added into PCR buffer containing 10 mM Tris–HCl, 1.5 μM MgCl2 and 50 mM KCl. PCR conditions were as follows: initial denaturation step at 95 °C for 5 min, 30 cycles at 95 °C for 30 s and 30 s at the 62 °C annealing temperature, and at 72 °C for 30 s. The final extension step was performed at 72 °C for 5 min. The PCR was carried out in a MJ Research, INC thermal cycler, model PTC-100 (Waltham, MA, USA). Following PCR, 20 ml aliquots were removed and subjected to restriction digestion with PvuII (for codon 194), BcnI (for codon 399), SatI (for codon 326) or PstI (for codon 751). All restriction enzymes were from Fermentas, Vilnius, Lithuania). The digested products were resolved on a 8 % acrylamide gel and stained with 0.5 μg/ml ethidium bromide. The cleavage of the XRCC1 fragment with PvuII produced bands of 292/174/21, 313/292/174/21 and 313/174 bp corresponding to the Arg/Arg, Arg/Trp and Trp/Trp genotypes, respectively. The BcnI restrictase having acted on the same fragment produced bands of 159/89, 248/159/89 and 248 bp corresponding to the 399 Arg/Arg, Arg/Gln and Gln/Gln genotypes, respectively. The SatI restriction enzyme yielded products of 200, 200/100/100 and 100 bp corresponding to the Ser/Ser, Ser/Cys and Cys/Cys genotypes of the OGG1 gene, respectively. The cleavage with PstI produced fragments of 161, 161/120/41 and 120/41 bp corresponding to the Lys/Lys, Lys/Gln and Gln/Gln genotypes of the ERCC2 gene, respectively.

Table 2 Primers used to analyze p.Arg194Trp, p.Arg399Gln polymorphisms of the XRCC1 gene, p.Ser326Cys polymorphism of the hOGG1 and p.Lys751Gln polymorphism of the ERCC2 gene
Full size table

Data Analysis

Logistic regression analysis was used to compute odds ratio (OR) and associated 95 % confidence interval (95 % CI) relating each of the SNPs as well as combinations of SNPs and another analysed factors presented in Table 1 to the risk of EC. Only matching variables and factors that altered the ORs by 10 % were included in the final multivariate models. Analyses were performed using STATISTICA 10 package (Statsoft, Tulsa, OK, USA).

Results

All distributions of genotypes did not differ significantly (p < 0.05) from those expected by the Hardy-Weinberg equilibrium. An association (OR 3.95; 95 % CI 1.88–8.31) was found between the Gln/Gln genotype of the p.Lys751Gln polymorphism of ERCC2 gene and EC occurrence (Table 3). There were no differences in the genotype distributions between cancer patients and controls for the remaining polymorphisms (Tables 4, 5 and 6). We also analyzed combined genotype of all polymorphism pairs. The Arg/Arg genotype of the XRCC1 gene increased the risk of EC for the carriers of the 751 Gln/Gln variant of the ERCC2 gene (Table 7). We also found that the Cys/Cys and Arg/Arg genotypes of the p.Ser326Cys polymorphism of the hOGG1 gene and the Arg/Gln genotype of the XRCC1 gene decreased EC risk (OR 0.50; 95 % CI 0.25–0.99) (Table 8). No difference between genotype distributions was found for others combined genotypes of the polymorphisms (data not shown). Adjustment for first degree relatives cancer history increased OR for the Gln/Gln genotype of the p.Lys751Gln polymorphism of ERCC2 gene from OR 3.95; 95 % CI 1.88–8.31 to OR 4.97; 95 % CI 1.98–12.48. Other remaining confounders, including postmenopausal hormone use and body mass index, did not modify the observed estimates of association.

Table 3 The allele and genotype frequency and odds ratio (OR) of p.Lys751Gln polymorphism of the ERCC2 gene in endometrial cancer
Full size table
Table 4 The allele and genotype frequency and odds ratio (OR) of the p.Ser326Cys polymorphism of the hOGG1 gene in endometrial cancer
Full size table
Table 5 The allele and genotype frequency and odds ratio (OR) of the p.Arg399Gln polymorphism of the XRCC1 gene in endometrial cancer
Full size table
Table 6 The allele and genotype frequency and odds ratio (OR) of the p.Arg194Trp polymorphism of the XRCC1 gene in endometrial cancer
Full size table
Table 7 The distribution of combined genotypes of the of the p.Arg194Trp polymorphism of the XRCC1 gene and p.Lys571Gln polymorphism of the ERCC2 gene in endometrial cancer
Full size table
Table 8 The distribution of combined genotypes of the p.Ser326Cys polymorphism of the hOGG1 gene and p.Arg399Gln polymorphism of the XRCC1 gene in endometrial cancer
Full size table

Discussion

In the present study we genotyped four common polymorphisms of the XRCC1, hOGG1 and ERCC2 DNA repair genes and tested the association between the distributions of their genotypes with EC. These polymorphisms have been shown to have functional significance and may be in part responsible, for the inter-individual difference in capacity of DNA repair in the general population and for low DNA repair efficacy in cancer patients [5–7, 14–17]. We obtained a significantly higher OR than for other analyzed polymorphisms, odds ratio for the Gln/Gln genotype of the p.Lys751Gln polymorphism of the ERCC2 gene than for genotypes of remaining polymorphisms. The protein encoded by the ERCC2 gene is involved in transcription-coupled NER and is an important member of the basal transcription factor TFIIH. Exchange of 751 Lys for Gln in the ERCC2 can lead to a conformational change in the encoded protein at the domain of the interaction between ERCC2 and its helicase activator, p44, inside the TFIIH complex [18]. The Gln/Gln variant of the ERCC2 gene has been associated with an increased risk of lung cancer [10, 11], and correlated with higher risk of skin, bladder and breast cancer [12, 19, 20]. Surprisingly, this polymorphism has been also linked with non-cancer diseases, such as cataract [21]. To date, none studies have addressed the association between alterations in this region of the ERCC2 gene and EC. Because a proper functioning of the ERCC2 gene is important for the genomic stability, its alternations may be associated with a higher cancer susceptibility.

Type I EC are estrogen-related. The mechanisms by which estrogens might cause the development of EC remain unclear. Estrogens have the unique chemical structure that distinguish them from other groups of hormones and their metabolism in eukaryotic cells include formation of a variety of intermediate forms and production of ROS. Estrogens undertake oxidative metabolism through hydroxylation pathway, but the major intermediates are 2-OH and 4-OH estrogens [22]. These chemicals are further oxidized to semiquinones and quinones, which may form bulky DNA adducts and may undergo redox cycling, producing ROS that may cause oxidative stress, lipid peroxidation, and DNA damage [23, 24]. Consequently, estrogen metabolism in human cells may play a role in tumor initiation via direct damage to the DNA by the formation of bulky DNA adducts and/or by producing ROS that cause oxidative DNA damage. These types of DNA damage are usually repaired by NER and BER.

In our study we analyzed the association between three polymorphisms of two genes of BER and EC. We did not find any association when we analyzed each polymorphism separately, but the analysis of combined genotypes showed that they might significantly increase the risk of EC. The results obtained suggest that polymorphisms of the XRCC1 and ERCC2 genes may modulate the risk and therefore play a role in the etiology of EC. The XRCC1 protein has no known catalytic activity but serves to orchestrate BER through its role as a central scaffolding protein for DNA ligase III, DNA polymerase β, and poly(ADP-ribose) polymerase (PARP) [25]. Arg/Trp variant of the p.Arg/Trp polymorphism of the XRCC1 gene occurs in proliferating cell nuclear antigen binding region, but few studies have examined the influence of the Trp/Trp genotype of this polymorphism on the function of the XRCC1 protein [26, 27]. This variant has been associated with a lower bleomycin and benzo(a)pyrene diol-epoxide sensitivity in vitro [16, 28]. These data suggest a protective role of the Trp/Trp genotype of the p.Arg/Trp polymorphism of the XRCC1 gene against the development of cancer and this function can be underlined by increasing the activity of BER. This is in agreement with our result suggesting a potential role of the Arg/Arg genotype of the p.Arg/Trp polymorphism of the XRCC1 gene with reduced BER capacity as compared with Trp/Trp genotype in EC.

We have also found that cancer history in first degree relatives increased endometrial cancer risk in the Gln/Gln variant of the p.Lys751Gln polymorphism of the ERCC2 gene. This result may suggest hereditary background of EC cancer and/or major contribution of the p.Lys751Gln polymorphism of the ERCC2 gene in cancer development but more studies performed on larger population is needed to draw a final conclusion.

In summary, our results suggest that the 751 Gln/Gln variant of the p.Lys751Gln polymorphism of the ERCC2 gene can be associated with the occurrence of EC. We have also showed that the Arg/Arg variant of the p. Arg194Trp polymorphism of the XRCC1 gene increased the risk of EC in individuals with the Gln/Gln variant of the ERCC2 gene. The data obtained suggest also that positive cancer history in first degree relatives in connection with Gln/Gln variant of the p.Lys751Gln polymorphism of the ERCC2 gene may be associated with EC.