Association of DNA repair genes polymorphisms with childhood acute lymphoblastic leukemia: a high-resolution melting analysis

Acute lymphoblastic leukemia (ALL) is one of the most common cancers in children for which the exact pathogenesis is not yet known. Single-nucleotide variants (SNVs) in different DNA repair genes are reported to be associated with ALL risk. This study aimed to determine the association between XRCC1 (rs1799782) and NBN (rs1805794, rs709816) SNVs and childhood ALL risk in a sample of the Iranian population. Fifty children with ALL and 50 age- and sex-matched healthy children were included in this case–control study. Genotyping of the mentioned SNVs was done by high-resolution melting (HRM) analysis. The prevalence of all three SNVs in XRCC1 and NBN genes did not differ between the patient and control groups, and these polymorphisms were not associated with childhood ALL risk (P > 0.05). HRM was a practical method for the detection of SNVs in XRCC1 and NBN genes. We found no significant association between XRCC1 (rs1799782) and NBN (rs1805794, rs709816) SNVs and childhood ALL risk.


Introduction
Globally, acute lymphoblastic leukemia (ALL) is the first most common form of childhood cancer and constitutes about 25-30% of all malignancies in those aged 2-5 years [1]. ALL is a heterogeneous disorder that arises from a malignant transformation and proliferation of lymphoid progenitors in bone marrow, peripheral blood, lymph nodes, and thymus [2]. It has been hypothesized that the initiation of ALL occurs in early infancy [3]. Overall survival (OS) in children is estimated close to 90% but patients with relapsed/refractory ALL present a therapeutic challenge [1,4]. ALL progresses rapidly and it is suggested that a combination of genetic susceptibility and environmental factors contribute to disease development. However, the exact roles of both genetic and environmental factors as risk factors for ALL remain poorly understood and they require further study [5][6][7]. One of the major guilty of ALL pathogenesis is DNA damage. Humans are routinely adventured to endogenous and exogenous damaging factors [8]. Environmental exposures such as ultraviolet light, ionizing radiation, and chemical agents cause exogenous DNA damage. In addition, DNA may be damaged by endogenous factors such as oxidative stress, alkylation, hydrolysis (deamination, depurination, and depyrimidination), and mismatch of Open Access BMC Research Notes *Correspondence: ahmad.fatemi2@gmail.com DNA bases [8,9]. To overcome these lesions and maintain genome integrity, cells possess several DNA repair systems that remove many DNA damages [10]. These mechanisms are double-strand break repair, direct repair (MGMT), base excision repair (BER), nucleotide excision repair, and mismatch repair pathway [11]. In several studies have been reported that single nucleotide variants (SNVs) in genes encoding DNA repair proteins could be associated with ALL risk [12].
The BER pathway is primarily responsible for removing DNA lesions such as oxidized bases, alkylated bases, and abasic (AP) sites [13]. This pathway is a multistep process that requires the activation of several proteins and the deficiencies and changes in the efficiency of these proteins have been associated with diminished DNA repair capacity and individual susceptibility to cancer [14]. The x-ray repair cross-complementing group 1 (XRCC1) is one of the BER components. XRCC1 gene is located on the long arm of chromosome 19 (19q13.2) and consists of 17 exons encoding a 633 amino acid protein which plays a crucial role as a scaffold and multifunctional protein in BER [9]. XRCC1 has a domain for biologically significant interactions with other BER proteins, including poly (ADP-ribose) polymerase 1 (PARP-1), the gap-filling DNA polymerase β (POLβ), the DNA 30-phosphatase (PNKP), and DNA ligase 3a (LIG3a) [15]. SNVs are DNA base variants with a frequency of more than 1% in the human population [16]. SNV in the coding sequences of XRCC1 was shown to potentially associate with modulated DNA repair capacity, making it a risk factor to occurrence of various cancers, especially hematopoietic diseases [17]. Three common SNVs have been identified in the XRCC1 gene in humans (Arg194Trp, Arg280His, and Arg399Gln) [18,19]. One of them is rs1799782 (at position 26304 on exon 6, base G to A, amino acid Arg to Trp) in the codon 194 which may alter XRCC1 function [20].
Another DNA repair protein is Nibrin, encoded by the NBN gene located on chromosome 8q21. This protein is associated with both homologous recombination (HR) and non-homologous end-joining (NHEJ) DNA doublestrand breaks (DSBs) repair pathways [21][22][23]. The protein is a member of the NBS1/hMre11/RAD50 (N/M/R) complex that plays a key role in DNA DSBs repair [24][25][26]. Genetic variants in NBN may modulate the function of the protein and so the DNA repair capacity. In the past decade, 675 SNVs were reported for the NBN gene to be associated with multiple cancers [27]. Two common NBN SNVs, based on minor allele frequency of > 5%, are rs1805794 in the codon 185 (exon 8, base C to G, amino acid Glu to Gln) and rs709816 in the codon 399 (exon 8, base A to G, Asp to Asp) which may alter NBN function [28].
According to this fact that the prevalence of genetic polymorphisms is racially dependent, so evaluation of these variations in different populations provides valuable information about leukemia susceptibility and disease prognosis. To enhance our knowledge about these variations, we designed a case-control study to investigate the association between XRCC1 (rs1799782) and NBN (rs1805794, rs709816) genes polymorphisms and ALL risk in the Iranian population (Kerman province).

Methods Participants
This case-control study included 50 children diagnosed with ALL (based on current 2008 World Health Organization criteria) and 50 healthy children with no previous history of malignancy. All subjects in both case and control groups were matched for age and sex.

DNA extraction
Genomic DNA was extracted from 1.5 mL EDTA-anticoagulated whole blood samples using the modified saltingout method [29]. Extracted DNA samples were stored at − 20 °C until genotyping.

SNV genotyping by high-resolution melting (HRM) analysis
HRM analysis was performed using the Rotor-Gene ® Q instrument (QIAGEN's real-time PCR cycler). According to the manufacturer's protocol of 5 × HOT FIREPol ® EvaGreen ® HRM Mix (Solis BioDyne, Estonia), Components of an HRM reaction included: 1 × HRM Master Mix, 100 nM forward primer (10 pmol/μL), 100 nM reverse primer (10 pmol/μL), 50 ng/μL DNA template and H2O PCR grade. The final volume of the reaction mixture was 20 μL. Then, DNA was amplified under these cycling conditions: Initial activation at 95 °C for 12 min, followed by 40 cycles of amplification consisting of denaturation at 95 °C for 15 s, annealing at 67 °C for 20 s, and extension at 72 °C for 20 s. Finally, an HRM step was performed with an increase in temperature from 65 to 90 °C at a ramp rate of 0.1 °C per second. All data were analyzed using Rotor-Gene Q Series Software. Primer sequences and general information for each SNV are shown in Additional file 1: Table S1.

DNA sequencing
Because of obtaining different groups of melt curves and to ensure the accuracy of genotyping by HRM assay, three samples from each group were sent to BIONEER Corporation (South Korea) for direct DNA sequencing.

Statistical analysis
Pearson Chi-square test was used to compare the frequency of genotypes and alleles between case and control groups. To investigate the expected frequency of control genotypes, Hardy-Weinberg equilibrium (HWE) was applied. To determine the association of each genotype and risk of ALL, odds ratios (ORs) and 95% confidence intervals (CIs) were calculated using multivariate logistic regression. Statistical analyses were carried out with SPSS software version 23.0. A P value less than 0.05 was considered statistically significant.

Baseline characteristics of patients and controls
Our study included 50 children with ALL (age 5.5 ± 5.4 years) and 50 healthy children (age 6.2 ± 2.8 years). Demographic and hematologic characteristics of participants are presented in Table 1. All subjects in both case and control groups were matched for age (P = 0.41) and sex (P = 0.83). Hematologic parameters including red blood cell count, hemoglobin, hematocrit and platelet count were significantly associated with childhood ALL (P < 0.0001).

HRM and sequencing results
We used HRM assay to detect XRCC1 (rs1799782 G > A) and NBN (rs1805794 C > G, rs709816 A > G) gene polymorphisms. After analyzing, it was found that the melting curves show different patterns, so several samples from each group were selected for sequencing. Supplementary figures (Additional file 2: Fig. S1, Additional file 3: Fig. S2, Additional file 4: Fig. S3) show the results of sequencing for each SNV as well as the genotyping results obtained by the HRM technique presented in three models (melting curve, normalized graph, and difference graph).
In addition, the interaction analysis of the NBN SNVs (rs709816 by rs1805794) did not showed any statistically significant association with childhood ALL (OR = 3.06, 95% CI 0.31-29.94, P = 0.33). Multivariate regression analysis results, expressing the impact of each variable independently, are presented in Table 3. Accordingly, the studied SNVs were not associated with childhood ALL (P > 0.05).

Discussion
ALL is a complex disease influenced by both genetic and environmental factors [30]. One of the important factors in the development of this leukemia is genetic variations including SNVs as the most common type of variations [30]. In the present study, we investigated the influence of polymorphisms within XRCC1 (rs1799782 G > A) and NBN (rs1805794 C > G, rs709816 A > G) genes on susceptibility to childhood ALL in a sample of the Iranian population by HRM analysis as a powerful, cost-effective and practicable method [31,32] for detecting SNVs and discriminating the wild type and mutant genotypes with melting temperature shifts more than 0.4 °C [33]. rs1799782 G > A is located in the coding sequence of XRCC1 and associated with modulated DNA repair capacity [34]. According to the present results, there was no significant association between this SNV and ALL risk in our population. Similar to our finding, Joseph et al. showed that this polymorphism did not affect the susceptibility to ALL in Indian children [35]. Besides, no significant differences were observed among Turkish patients with childhood ALL and the controls concerning this polymorphism. In this study, however, Batar et al. reported that the combined XRCC1 AG/AA variant genotypes were associated with increased risk for ALL in females (OR = 5.47; 95% CI 1.49-20.10; p = 0.008) [34]. A meta-analysis of 38 case-control studies indicated that XRCC1 Arg194Trp may be a biomarker of cancer susceptibility [36]. Other SNVs examined in the present study were rs1805794 C > G and rs709816 A > G, localized to the coding sequences of the NBN gene, which may modulate the protein function and so the DNA DSB repair capacity [37]. In our study, no significant association was observed between the NBN polymorphisms and susceptibility to childhood ALL. Similar to our results, Mosor et al. genotyped six polymorphisms of the NBN gene in the Poland population and suggested that one specific variant of the gene may be associated with childhood acute leukemia but no significant association was detected for the remaining five NBN polymorphisms including rs1805794 and rs709816 [38]. Consistently, in the other study on 460 pediatric ALL cases and 552 healthy controls in the European population (Germany), Smolkova et al. also did not show any association between the NBN polymorphisms and ALL risk [39]. These results show that our population, at least in part, is closely related to the European population. In contrast, in the study of the Chinese population conducted on 175 patients with ALL and 350 controls, Jiang et al. observed a significant difference in frequencies of rs1805794 C/G between cases and controls (P < 0.0001) and reported for the first time that CC genotype carriers were associated with a 3.4-fold elevated risk of ALL, compared with the non-carries [27]. Also, a meta-analysis of rs1805794 including 3065 subjects revealed the association of CC genotype with approximately 1.70-fold increased risk of acute leukemia. In contrast, there was no association between rs709816 and risk of acute leukemia in the metaanalysis of including 1,485 samples [37].
In conclusion, we cannot recognize any association between the XRCC1 and NBN polymorphisms and susceptibility to childhood ALL in a sample of the Iranian population.

Limitations
• The discrepancy of our results with other studies may be due to the small sample size and so we suggest that similar studies be conducted on the larger sample size and different populations.