Single Nucleotide Polymorphisms (SNPs) of RAD51-G172T and XRCC2-41657C/T Homologous Recombination Repair Genes and the Risk of Triple- Negative Breast Cancer in Polish Women

Double strand DNA breaks are the most dangerous DNA damage which, if non-repaired or misrepaired, may result in genomic instability, cancer transformation or cell death. RAD51 and XRCC2 encode proteins that are important for the repair of double-strand DNA breaks by homologous recombination. Therefore, genetic variability in these genes may contribute to the occurrence and progression of triple-negative breast cancer. The polymorphisms of the XRCC2 gene -41657C/T (rs718282) and of the RAD51 gene, −172G/T (rs1801321), were investigated by PCR-RFLP in 70 patients with triple-negative breast cancer and 70 age- and sex matched non-cancer controls. The obtained results demonstrated a significant positive association between the RAD51 T/T genotype and TNBC, with an adjusted odds ratio (OR) of 4.94 (p = 0.001). The homozygous T/T genotype was found in 60 % of TNBC cases and in 14 % of the used controls. Variant 172 T allele of RAD51 increased cancer risk (OR = 2.81 (1.72–4.58), p < .0001). No significant associations were observed between -41657C/T genotype of XRCC2 and the incidence of TNBC. There were no significant differences between the distribution of XRCC2 -41657C/T genotypes in the subgroups assigned to histological grades. The obtained results indicate that the polymorphism of RAD51, but not of XRCC2 gene, may be positively associated with the incidence of triple-negative breast carcinoma in the population of Polish women.


Introduction
Triple-negative breast cancer (TNBC) is a subtype of breast cancer defined by the lack of expression of estrogen, progesterone and HER-2 (human epidermal receptor-2) receptors. TNBC refers to about 15-20 % of all breast cancer cases. It is characterised by worse clinical outcome, poor prognosis and absence of prognostic indicators [1][2][3][4][5].
Mutations in DNA of double-strand breaks (DSB) repair genes are involved in the pathogenesis of tumours, however, it is still unclear whether any defects in this pathway may play any role in the aetiology of triple-negative breast cancer.
DSB in DNA may be rectified either by homologous recombination (HR) or non-homologous end joining (NHEJ) [6,7]. RAD51 is involved in the homologous recombination and repair of double-strand breaks in DNA and DNA cross-links and is responsible for the maintenance of chromosome stability [8].
RAD51 gene has been mapped to chromosome 15q14-15 and is polymorphic in nature. The involvement of RAD51 in the DNA repair determines its potential role in maintaining genetic stability, which is disturbed in cancer. Therefore the problem of genetic variability of the RAD51 gene in cancer is worthy studying. A substitution G/T at position 172 in 5′ untranslated region of the RAD51 gene (the 172G/T 5′UTR polymorphism) has been reported as a possible factor to affect the RAD51 mRNA stability [8,9]. 172G/T polymorphism of RAD51 is located in the regulatory element of the RAD51 promoter and is suggested to be associated with messenger RNA expression [8,9].
XRCC2 gene, located in 7q36.1, is an essential part of the homologous recombination repair pathway and a functional candidate for involvement in cancer progression [17].
Unfortunately, it is difficult to find in the literature reports directly binding -41657C/T SNP in DNA repair gene XRCC2 with breast cancer occurrence.
Because little is known on the association between RAD51-172G/T and XRCC2-41657C/T polymorphisms and TNBC, it was studied whether the polymorphisms in the two genes, being involved in the homologous recombination of doublestrand breaks, could have been linked with the risk for triplenegative breast cancer in that Polish population.

Patients with Triple-Negative Breast Cancer
A total of 70 patients with histologically-proven diagnosis of TNBC were included in the reported study. The distributions of sociodemographic characteristics, lifestyle risk factors and clinical characteristics of the patients are shown in Tables 1 and 2, respectively. Paraffin embedded tumour tissue specimens were obtained from women with triple-negative breast carcinoma, treated at the Department of Oncology, Institute of Polish Mothers Memorial Hospital, between 2000-2013. The age of the patients ranged in from 36 to 68 years (the mean age 46.2 ± 10.12). The median follow-up of patients at the time of analysis was 38 months (the range: 2-70 months). The average tumor size was 20 mm (the range 17-32 mm). All the diagnosed tumours were graded by criteria of the Scarf-Bloom-Richardson. DNA from non-cancerous breast tissue (n=70) served as control (mean age 45.41± Breast tissue samples (cancerous and non-cancerous) were routinely fixed in phormaldehyde, embedded in paraffin, cut into thin slices and stained with haematoxylin/eozin for pathological examination. DNA for analysis was obtained from archival pathological paraffin-embedded tumour specimens and from samples of non-cancerous breast tissue, which were deparaffinised in xylene and rehydratated in ethanol and distilled water. In order to ensure that the chosen histological material was representative for cancerous and non-cancerous tissue, each tissue sample, qualified for DNA extraction, was initially checked by a pathologist. DNA was extracted from the material, using a commercially available QIAmp DNA purification kit (Qiagen GmbH, Hilden, Germany), according to the manufacturer's instruction.

Statistical Analysis
For each polymorphism, departure of the genotype distribution from that expected from Hardy-Weinberg equilibrium was assessed using the standard χ 2 -test. Genotype frequencies in the study cases and the controls were compared by χ 2 -test. Genotype specific risks were estimated as odds ratios (ORs) with associated 95 % intervals (CIs) by unconditional logistic regression. P-values<0.05 were considered significant. All the statistical analyses were performed, using the STAT ISTICA 6.0 software (Statsoft, Tulsa, Oklahoma, USA). Table 3 shows genotype distribution of RAD51 polymorphism, illustrating the difference between the patients with TNBC and the controls. The Table clearly indicates significant differences (p<0.05) between the two investigated groups. A weak association was observed between the occurrence of triple-negative breast carcinoma and the presence of G/T and T/T genotypes. A stronger association was observed for T/T homozygotes than for G/T heterozygotes. Variant 172 T allele of RAD51 increased cancer risk. The frequencies of G/G, G/T and T/T genotypes, observed in the study patients, differed significantly (p<0.05) from the distribution values, as expected from the Hardy-Weinberg equilibrium.  No statistically significant differences were observed between the control group and the patients with triple-negative breast cancer, regarding either allele or genotype frequencies of XRCC2 -41657C/T gene polymorphism (see Table 4).

Results
Scarf-Bloom-Richardson grading was related to RAD51 and XRCC2 polymorphisms. Histological grade was evaluated in all the cases (n=70), where 20 cases were of grade I, 45 of grade II and 5 of grade III. Grade II and III were grouped together for the purposes of statistical analysis (see Table 5). No differences were observed in those groups between RAD51 and XRCC2 genotype distributions, while a lack of correlation was demonstrated between genotypes of the identified polymorphisms and TNBC invasiveness.
We did not find any association of the XRCC2 and RAD51 polymorphisms in patients group with cancer progression assessed by breast cancer with (N+) and without (N-) lymph node metastases (p>0.05). Our data did not demonstrate any statistically significant correlation between RAD51 and XRCC2 polymorphisms and the risk factors for breast cancer, such as BMI (body mass index), smoking status, alcohol consumption, menarche, menopause status, reproductive histories, exogenous hormone use and medical history (data not shown).

Discussion
Breast cancer classification is in constant evolution, as advances in molecular pathology as well as immunohistochemical staining allow researchers to define the molecular heterogenity of different disease subtypes and to guide the selection of appropriate treatment. The triple negative phenotype, defined as the lack of estrogen receptor, progesterone receptor and HER-2 expression, represents approximately 15-20 % of breast cancer cases and has a worse clinical outcome and prognosis than other breast cancer subtypes [1][2][3][4][5].
As it has already been mentioned above, little is known on the association between the homologous recombination repair RAD51 and XRCC2 SNPs and triple-negative breast cancer. Therefore, it was attempted to investigate the role of genetic variation in homologous recombination repair genes and the risk of TNBC.
As mentioned in Introduction mutations in DNA doublestrand breaks repair genes are involved in the pathogenesis of tumours. Defects in this pathway may play a role in development and progression of breast cancer [24,25].
No significant associations were observed between the 172G/T and breast cancer in Korean women. The genotype distributions of RAD51 SNP did not differ significantly from Hardy-Weinberg equilibrium (P>0.05) and the frequencies of minor alleles were 0.13 for RAD51 135C and 0.05 for RAD51 172 T. When the two loci in the same gene were evaluated for  linkage disequilibrium, two loci in RAD51 gene (135G/C and 171G/T) were in negative linkage disequilibrium [15]. Shin et al. examined the role of SNPs in RAD51 repair genes and the risk of breast cancer. In the reported study, the 135G/C but not 172G/T polymorphism, of RAD51 gene, was associated with breast carcinoma occurrence [16].
The role of -41657C/T variation of XRCC2 in breast cancer is still unknown. Some reports documented that the XRCC2 -41657C/T genotype was related to increased esophageal squamous cell carcinoma (ESCC), gastric cardia adenocarcinoma (GCA) and in smoking-drinking-related laryngeal cancer risk [22,29].
It appears from literature review that RAD51 and XRCC2 polymorphisms have been investigated in various types of malignancy. A large number of studies confirm a significant role of the influence of specific genetic variants on repair phenotype and cancer risk [29][30][31][32][33][34].
Unfortunately, it is difficult to find reports in the available literature, which would directly bind RAD51 and XRCC2 SNP in DNA repair gene by HR with clinical-pathological features of breast tumour. Only in single studies the researchers suggest that the homologous recombination repair gene polymorphism may play some role in the development of breast carcinoma [13].
Two single nucleotide polymorphisms of XRCC2-41657C/ T, and RAD51-172G/T were investigated in the reported study, as well as their association with human triple-negative breast cancer.
In literature, neither XRCC2-41657C/T nor RAD51-172G/ T polymorphism is correlated with risk of breast cancer [15,16]. What are important, recent reports introduce the role of RAD51 135G/C polymorphism in the development of triplenegative breast cancer [27].
In the recent studies, 135G/C polymorphism of RAD51 may be associated with an elevated tumour risk in the Polish populations, regarding TNBC [28], while there are still no data, which would be illustrating the significance of RAD51 polymorphism for triple-negative breast cancer development in other populations.
The obtained results confirm the important role of RAD51 G172T polymorphism for TNBC occurrence in Poland. In the reported study, RAD51 T/T genotype increased the risk of breast cancer in the Polish population. There was a 4.94-fold increased risk of breast carcinoma for RAD51-T/T genotype carriers, compared with subjects with RAD51-G/G, G/T genotype. The 172 T allele increased the risk of TNBC. RAD51 G172T polymorphism was not related to cancer grade. It is possible that the presence of the 172 T allele is in some disequilibrium with another, so far unknown, mutation, located outside the coding region in RAD51 gene, which may be of importance for RAD51 concentration in plasma.
No significant associations were observed between -41657C/T genotype of XRCC2 and the incidence of TNBC in the Polish women.
In conclusion, the obtained data confirm the important role of RAD51 172G/T -but not of -41657C/T of XRCC2 gene polymorphism in triple-negative breast cancer aetiology.
Finally, we suggest that RAD51 172G/T may be used as a prognosing factor of precancerous lesions, predicting TNBC in the Polish population. Further studies are required, regarding the role of these genes in the aetiology of triple-negative breast cancer.

Conflict of Interests
The authors declare that there are no conflicts of interests.
Open Access This article is distributed under the terms of the Creative Commons Attribution License which permits any use, distribution, and reproduction in any medium, provided the original author(s) and the source are credited.