NBS1 rs2735383 polymorphism is associated with an increased risk of laryngeal carcinoma

In this study, 342 patients with histopathologically diagnosed primary laryngeal carcinoma were recruited between 2014 and 2016 in the Affiliated Tumor Hospital of Guangxi Medical University. They had a response rate of 95% among all cancer patients diagnosed in the hospital. Clinical and pathological information on all laryngeal carcinoma diagnoses were confirmed by manual review of the pathology reports and endoscopic findings of Otorhinolaryngology Department. Of the 342 cases, 37 were poorly differentiated squamous cell carcinoma (SCC), 63 were moderately differentiated SCC, 76 were well-differentiated SCC, and 166 remain unknown. Totally, 345 cancer-free controls who have matched age (±5 years) and sex with the cases were recruited from the same hospital. These control individuals had a response rate of 84%. We only recruited people whose ethnicity is Han or Zhuang.

After having given a written informed consent, all individuals were interviewed according to a structured questionnaire in order to collect personal information including age, sex, smoking status, drinking status and so on. The participants who have smoked less than 100 cigarettes in their lifetime were defined as never-smokers; otherwise were defined as ever-smokers [23]. Similarly, the participants who have consumed alcohol at least once a week for more than one year were defined as alcohol ever-drinkers and the remaining as alcohol never-drinkers [24]. Each study participant was asked to donate a one-time blood sample of 5 ml for later examination. This study was approved by the Medical Ethics Committee of Guangxi Medical University (GXMU-20140307-4).

Based on the previous studies [25‐27], we selected and genotyped two NBS1 SNPs (rs1805794G > C in exon 5 and rs2735383G > C in 3’-UTR). According to the dbSNP database, the current study defined the C allele of both SNPs in the antisense strand (i.e., the corresponding allele is G in the sense strand in the database) as minor allele [20, 22]. We used the polymerase chain reaction-restriction fragment length polymorphism (PCR-RFLP) method in Zheng’s study [20] to verity the association between NBS1 polymorphism and laryngeal carcinoma risk.

The primer pairs used to amplify the DNA fragment containing the rs1805794G > C polymorphism were 5’-ACCTT TCAAT TTGTG GAGGC-3′ (forward) and 5’-AGTCG GTCTT TGGTC ACTGC-3′ (reverse), to produce a fragment of 289 bp. The primer pairs for rs2735383G > C were 5’-TGCAG TGTTC TACAC CTTGC TT-3′ (forward) and 5′-AGGTG ACATC TGCAC CACTG-3′ (reverse), to produce a fragment of 156 bp. PCR was performed in a 25 μl reaction system, containing 5 mM MgCl_2, 0.1 mM deoxynucleotide triphosphates, 3.0 U of Taq polymerase (MBI Fermentas, Vilnus, Lithuania), and the manufacturer’s buffer. After an initial melting step at 94 °C for 5 min, the PCR procedure consists of 35 cycles including denature at 94 °C for 45 s, anneal at 62.9 °C for rs1805794 or 61.0 °C for rs2375383 for 45 s, and extension at 72 °C for 45 s, followed by a final extension step at 72 °C for 7 min. The amplified fragment containing rs1805794 polymorphism was cut by Hinf1 (Takara BioTech, Dalian, China) at 37 °C for at least 3 h. The major G allele produced a single 289 bp band, while the minor C allele produced 30- and 259-bp bands. The digestion products were separated and visualized in 3% agarose gel electrophoresis. Meanwhile, the amplified fragment containing rs2735383 polymorphism was cut by CviQI (New England Biolabs, Beijing, China) at 25 °C for at least 3 h. The minor G allele produced a single 156-bp band, while the major C allele produced two separate bands of 57-bp and 99-bp.

Total RNA from 32 laryngeal carcinoma tissues were extracted using TriPure Reagent (Roche Applied Science) and reverse transcribed to complementary DNA using cDNA synthesis kit ThermoScrept™ RT-PCR System (Invitrogen). The relative mRNA expression levels of NBS1 covering rs2735383 site and β-actin (as an internal reference) were measured by the ABI Prism 7500 Sequence Detection Systems (Applied Biosystems) using the SYBR-Green method, with the same baseline and threshold set for each plate to generate threshold cycle (Ct) values for both genes in each sample. The primers used for detection of NBS1 were as follows: 5′-TTGGT TGCAT GCTCT TCTTG-3′ (forward) and 5’-GGCTG CTTCT TGGAC TCAAC-3′ (reverse); and for β-actin 5’-GGCGG CACCA CCATG TACCC T-3′ (forward) and 5′-AGGGG CCGGA CTCGT CATAC T-3′ (reverse). The 2 ^-△△Ct method was used to measure the level of NBS1 gene’s expression.

We used Chi-square (χ²) test to analyze the differences of selected characteristics (age, sex, ethnicity, smoking status and drinking status) between cases and controls. Hardy-Weinberg equilibrium (HWE) was tested using a goodness-of-fit χ² test by comparing the expected genotype frequencies with that of the controls. The associations between the two polymorphisms of NBS1 and risk of laryngeal carcinoma were respectively calculated by odds ratios (ORs) with 95% confidence intervals (CIs) using a unconditional logistic regression model, after adjusting age, sex, smoking status and drinking status. Student’s t-test was used to evaluate the differences in NBS1 mRNA expression between different groups. All tests were analyzed using the Stata 13.0 software (Stata Corp LP, College Station, Texas, United States). The statistical power was calculated by using the PS Software (http://​biostat.​mc.​vanderbilt.​edu/​twiki/​bin/​view/​Main/​PowerSampleSize). P < 0.05 was considered statistically significant.

In order to determine whether the rs1805794 and rs2735383 polymorphisms of NBS1 gene are associated with laryngeal carcinoma, we examined the genotypic distribution of these two polymorphisms of NBS1 in both cases and controls. In this study, 342 laryngeal carcinoma cases and 345 controls were examined. The basic characteristics of laryngeal carcinoma cases and controls were shown in Table 1. There were no significant differences in distributions of sex, age, and ethnicity between cases and controls (P > 0.05 for all), while the smoking status and drinking status were significantly different between cases and controls (P = 0.024 and P = 0.000 respectively).

The genotype distributions of rs1805794 and rs2735383, and their respective associations with laryngeal carcinoma risk are summarized in Table 2. The observed genotype frequencies of these two SNPs among the control subjects were all in agreement with the Hardy-Weinberg Equilibrium (P = 0.081 for rs1805794 and P = 0.963 for rs2735383). As shown in Table 2, the respective frequencies of rs2735383 GG, GC and CC genotypes in the cases were 26.608%, 47.076%, 26.316%, differing significantly from 33.043%, 48.986%, 17.971% in the controls (P = 0.019). The logistical model showed that rs2735383CC variant genotype conferred a significantly increased risk of laryngeal carcinoma in comparison with the common genotype of GG (adjusted OR = 1.884, 95% CI =1.215–2.921), but the GC genotype fails (adjusted OR = 1.236, 95% CI =0.861–1.775). When combining these two genotypes, the rs2735383C variant genotypes (GC + CC) contributed to a 1.410-fold increased risk of laryngeal carcinoma in comparison to that of the GG genotype (adjusted OR = 1.410, 95% CI =1.004–1.980). The C allele of rs2735383 in a dose-dependent model was associated with an increased cancer risk (P_trend = 0.005).

In contrast, no significant association was observable between rs1805794G > C polymorphism and the risk of laryngeal carcinoma (P = 0.917, adjusted OR = 1.074, 95% CI = 0.759–1.518 for GC; adjusted OR = 1.100, 95% CI = 0.678–1.787 for CC; adjusted OR = 1.079, 95% CI = 0.774–1.505 for GC + CC). Combined analyses were conducted to evaluate the cumulative effect of rs1805794 and rs2735383 (Additional file 1). In comparison with the homozygous genotypes, the combined variant genotypes were not significantly associated with increased laryngeal carcinoma risk. Thus, our study demonstrated that NBS1 SNPs rs2735383G > C might be associated with the elevated risk of laryngeal carcinoma. We then focused on SNP rs2735383 in the following analyses.

Next, we tried to stratify the subjects according to different factors and then determined the effects of SNP rs2735383 on laryngeal carcinoma risk. As showed in Table 3, the rs2735383C variant genotypes (GC + CC) were significantly associated with an increased risk of laryngeal carcinoma in the male subgroup (adjusted OR = 1.505, 95%CI =1.066–2.126) and the ever-smoker subgroup (adjusted OR = 1.602, 95%CI = 1.023–2.510). But we failed to observe any significant interactions between NBS1 polymorphisms and these two confounding factors (all P_interaction > 0.05), possibly due to the limited sample size in each subgroup.

As shown in Fig. 1(a), among 32 laryngeal carcinoma tissues examined, the tissues of CC genotype have significantly lower mRNA expression (0.12 ± 0.06 for 7 cases with CC) of NBS1 than those with GC or GG genotype (0.23 ± 0.12 for 16 cases with GC and 0.29 ± 0.13 for 9 cases with GG; ANOVA test: P = 0.02). Furthermore, as seen in Fig. 1(b), the dichotomized analysis showed that the mRNA levels of NBS1 from those tissues with combined GC and CC genotypes (0.18 ± 0.09 for 23 cases with GC + CC) were significantly lower than that of GG genotype (0.29 ± 0.13 for 9 cases with GG; Student’s t-test: P = 0.003).