Introduction

It is well established that DNA repair plays a crucial role in maintaining normal function and genetic material stability of mammalian [1]. After DNA damage caused by endogenous or exogenous factors, it may arise carcinogenesis or apoptosis [2]. Polymorphism in DNA repair genes is one of the mechanisms that may lead to variation in DNA repair capacity [3]. It has been suggested that polymorphisms in genes involved in the non-homologous end joining (NHEJ) pathway, influence DNA repair capacity [4]. Ku, a heterodimeric DNA-binding complex, is directly involved in repair of DNA double-strand breaks as a member of the NHEJ pathway. NHEJ is thought to proceed through three key steps: recognition of the breaks, DNA processing to remove non-ligatable ends or other forms of damage at the termini, and finally ligation of the DNA ends. Recognition of the double-strand breaks is carried out by the Ku heterodimer, which is also required for recruitment of DNA-PKcs, the XRCC4/DNA ligase IV complex and XLF, reflecting its essential role in NHEJ [5]. The Ku complex binds to the ends of double-stranded DNA in a cell cycle-dependent manner, being associated with chromosomes of interphase cells, followed by complete dissociation from the condensing chromosomes in early prophase [6]. Ku consists of two subunits, Ku70 and Ku80, which are encoded by the XRCC6 (X-ray repair cross-complementing 6; OMIM: 152690) and XRCC5 (X-ray repair cross-complementing 5; OMIM: 194364) genes, respectively [7].

Several polymorphisms in the XRCC5 have been reported. One of them is a variable number of tandem repeats of a 21 bp (VNTR, rs. 6147172) polymorphism in the promoter region of XRCC5. This polymorphism has four alleles: 3R, 2R, 1R and 0R [8]. Number copies of cis elements in the promoter region of XRCC5 regulate its expression [9]. It has been shown that the VNTR polymorphism of XRCC5 is associated with the risk of bladder cancer [10]. Single-nucleotide polymorphism of T-991C (rs. 5751129) in the promoter region of XRCC6 has been reported. This polymorphism is associated with the risk of gastric, oral and lung cancer and also hepatocellular and renal cell carcinomas [11–15].

To date, there is no study investigating the association between the above-mentioned polymorphisms and susceptibility to breast cancer. The aim of the present study is to investigate the association between XRCC5 and XRCC6 polymorphisms and susceptibility to breast cancer.

Materials and methods

For the current study, 407 female patients were recruited from the chemotherapy department of Nemazi hospital in Shiraz, south of Iran. Eligibility criteria for cases were patients with pathologically confirmed primary adenocarcinoma of the breast and mental competence to give written informed consent. During the same time period, 395 cancer-free female controls, recruited among randomly selected blood donors, frequency matched to the cases by age. Participants with any previous history of cancer or diagnosed psychiatric diseases were excluded from the control group. The mean age (SD; Min–Max) of the patients and controls was 45.3 (10.7; 22–80) and 43.9 (8.8; 24–72) years, respectively. Considering the high heterogeneity of Iranian populations [16, 17], the participants (patients and controls) were selected from Persian Muslims (Caucasians) living in Fars Province (southern Iran). Ethical approval for the current study was obtained from Shiraz University institutional review board.

Genomic DNA was extracted from EDTA-treated blood samples. Genotyping for the XRCC5 VNTR polymorphism was carried out using high-resolution melting analysis (HRMA), following a pre-amplification step by Rotor-Gene 6000 instrument (Corbett Life Science). The reaction employed SYBR Premix Ex Taq II PCR master mix (Takara Bio Inc.) and previously described primers, 5′-AGG CGG CTC AAA CAC CAC AC-3′ (forward), and 5′-CAA GCG GCA GAT AGC GGA AAG-3′ (reverse). The PCR was performed in a total volume of 20 μl containing 100 ng genomic DNA, 10 μl SYBR Premix Ex Taq II, 0.8 μl (0.4 μM) of each forward and reverse primers, and the total reaction volume was brought to 20 μl with dH2O. Real-time PCR conditions were as follows: one cycle at 95 °C for 1 min followed by 40 cycles of denaturation at 95 °C for 10 s, annealing at 67 °C for 20 s and extension at 72 °C for 20 s. Melting curve data were collected from 85 to 95 °C at a ramping rate of 0.1 °C per second. Melting curve analysis was performed using SYBR green I channel, employing Rotor-Gene 6000 software (version 1.7) [18]. The T-991C polymorphism of XRCC6 was determined using the PCR–RFLP method with the primers as described previously [19]. The primers used were forward 5′-AAC TCA TGG ACC CAC GGT TGT GA-3′ and backward 5′-CAA CTT AAA TAC AGG AAT GTC TTG-3′. The cycling condition for the Ku70 promoter C-991T polymorphism was set as follows: one cycle at 94 °C for 8 min, 30 cycles at 94 °C for 30 s, 55 °C for 30 s, and 72 °C for 30 s, and one final cycle of extension at 72 °C for 10 min. The resultant 301 bp PCR product was mixed with 2 U of DpnII. The restriction site was located at −991 with a C/T polymorphism, and the C form PCR products could be further digested, while the T form could not. Two fragments measuring 101 and 200 bp were present if the product was digestible (C). The reaction was incubated for 2 h at 37 °C. Then, 10 μl of product was loaded into a 3 % agarose gel containing ethidium bromide for electrophoresis. The polymorphism was categorized as either (1) a C/C homozygote (digested), (2) T/T homozygote (undigested), or (3) C/T heterozygote.

A chi-square test was performed for the polymorphisms to determine if the control sample demonstrated Hardy–Weinberg equilibrium. The risk of breast cancer associated with the XRCC5 and XRCC6 polymorphisms was estimated. The relative associations between the genotypes and breast cancer risk were assessed by calculating crude odds ratios (ORs) and 95 % confidence intervals (CIs). To determine the effect(s) of XRCC5 and XRCC6 polymorphisms on age at diagnosis of breast cancer, the Kaplan–Meier survival analysis and Cox proportional hazards regression model were used. In analysis, breast cancer was defined as event, and age at diagnosis was included in the analysis as time period to event. Data analysis was performed using SPSS software version 11.5. All statistical tests are two-sided.

Results and discussion

Table 1 shows the genotype distribution of the studied polymorphisms in breast cancer cases and healthy controls. Control subjects were at Hardy–Weinberg equilibrium for both polymorphisms (For VNTR XRCC5 polymorphism: χ2 = 3.065, df = 6, P = 0.800; For T-991C XRCC6 polymorphism: χ2 = 1.767, df = 1, P = 0.183).

Table 1 Association between polymorphisms at promoter regions of XRCC5 and XRCC6 and risk of breast cancer
Full size table

For the VNTR XRCC5 polymorphism, in comparison with the 1R/1R genotype, the 0R/0R genotype significantly increased the risk of breast cancer (OR 9.55, 95 % CI 1.19–76.64, P = 0.034). The 1R/3R genotype compared with 1R/1R genotype decreased the risk of breast cancer (Fisher’s exact test P = 0.015). There was no significant association between T-991C polymorphism of XRCC6 and risk of breast cancer.

To determine the effect(s) of XRCC5 and XRCC6 polymorphisms on age at diagnosis of breast cancer, the Kaplan–Meier survival analysis and Cox proportional hazards regression model were used. Mean age at diagnosis of breast cancer for 0, 1, 2, 3, and >4 repeat in XRCC5 were 39.2, 41.9, 44.3, 45.8, and 47.3 years, respectively. The Kaplan–Meier survival analysis revealed that number of repeat was associated with age at diagnosis of breast cancer (log rank statistic = 13.90, df = 4, P = 0.008; Fig. 1). This means that more number of VNTR is associated with higher age at diagnosis of breast cancer. However, the XRCC6 polymorphism was not associated with age at diagnosis of breast cancer (log rank statistic = 0.584, df = 2, P = 0.747).

Fig. 1
figure 1

Association between number of repeats of XRCC5 VNTR and age at diagnosis of breast cancer

Full size image

Using Cox proportional hazards regression model, after adjustment for smoking habit, marital status, and the XRCC6 genotypes, there were significant associations between the number of repeats of XRCC5 and age at diagnosis of breast cancer. The 4 repeats (Hazard ratio, HR 0.229, 95 % CI 0.09–0.56, P = 0.002), 3 repeats (HR 0.216, 95 % CI 0.09–0.49, P < 0.001), 2 repeats (HR 0.231, 95 % CI 0.10–0.52, P < 0.001), and 1 repeat (HR 0.205, 95 % CI 0.08–0.47, P < 0.001) versus to the 0 repeat showed higher age at diagnosis of breast cancer.

There is strong evidence that some transcription factors can bind VNTR sequences [20]. The VNTR polymorphism of XRCC5 can alter the number of cis elements. The alleles with more tandem repeats include more Sp1 binding sites and could increase the affinity of Sp1 to the promoter of XRCC5 [10]. Our previous study indicated that the increase in the overall number of tandem repeats in the promoter region of XRCC5 down-regulates the gene expression [5].

The present study showed that 0R/0R genotype verses 1R/1R genotype is associated with increased risk of breast cancer. Also, we found that the 1R/3R vs 1R/1R genotypes decreased the risk of breast cancer (Table 1). It should be noted that based on the present study, the 2R/2R and 1R/2R genotypes did not alter the risk of breast cancer compared with the 1R/1R genotype (Table 1). It is self-evident that when 0R/0R and 1R/3R genotypes increased and decreased the risk of breast cancer compared with the 1R/1R genotype, respectively, it means that intermediate genotypes (such as 2R/2R and 1R/2R) may not alter the breast cancer risk. This finding is consistent with a report, indicating that fewer repeats of XRCC5 VNTR polymorphism is associated with increased risk of bladder cancer [10]. It should be mentioned that the over-expression of XRCC5 protein in several types of cancers have been reported previously [21–23]. Taken together, it might be suggested that in fewer tandem repeats, over-expression of XRCC5 leads to excess DNA repair, which interfere with normal apoptosis, thus increasing the likelihood for the development of breast cancer.

The strength of the results is tempered by small sample size, and this finding needs further replication in a larger sample. Considering the fact that ethnicity may influence the observed associations in multifactorial disease [24, 25], replication of this study in other countries is recommended.