This article is mentioned in:

Introduction

Esophageal cancer occurs frequently, with an incidence ranked 8th among all tumor types. This form of cancer is associated with a high degree of malignancy and a poor 5-year survival rate of <10% (1–2). The prevalence of esophageal cancer, in particular esophageal squamous cell carcinoma (ESCC), is extremely high in China, with ~250,000 new cases arise annually, accounting for 50% of all cases worldwide (3).

However, despite the frequent occurrence and high mortality of esophageal cancer, its pathogenesis remains unknown. Esophageal cancer is generally accepted to be a complex polygenic disease caused by genetic and environmental factors. Studies have demonstrated that defects in DNA damage repair lead to accumulated DNA damage, which confers a higher risk of cancer (4). The most important DNA damage repair mechanism in humans is nucleotide excision repair (NER) (5). NER is performed by a number of proteins, for deficiencies that cause human diseases, including Xeroderma pigmentosum and Cockayne syndrome. A key NER protein, Xeroderma pigmentosum group D (XPD), a subunit of transcription factor IIH, has ATP-dependent 5′→3′ helicase activity and is involved in gene transcription with the action of RNA polymerase II (6,7). Additionally, XPD interacts with p53 to co-regulate cell apoptosis (8). Thus, XPD is essential to DNA repair capacity and complete XPD gene deletion results in embryonic death (9).

XPD, also known as excision repair cross-complementing, maps to a 5.4-kb region of human chromosome 9. Eight single nucleotide polymorphisms (SNPs) have been found in the coding region thus far and among these, variations in codons 312 (G→A) and 751 (A→C) are the most common (10). Numerous studies have investigated potential associations between XPD polymorphisms and various types of cancer, including head and neck squamous, lung squamous and basal cell carcinomas and breast and gastric cancers (11–15). However, a limited number of studies have explored the relationship between XPD polymorphisms and ESCC and have yielded inconsistent results (16–19). Therefore, this case-control study assessed whether SNPs in codons 312 and 751 of XPD are associated with ESCC in a Chinese population.

Materials and methods

Study subjects

The present study was a population-based case-control study. Between January 2008 and December 2011, 400 patients with ESCC (case group) whose diagnoses were confirmed by pathological biopsy of tissues removed by endoscopic or surgical excision were recruited. Healthy individuals (control group) were confirmed by endoscopic census to be free of esophageal cancer and precancerous lesions over the same period in the same geographical area. Information on gender, age, smoking status and alcohol consumption of case and control group individuals was obtained. The 400 healthy individuals were gender- and age-matched with the case group. Smoking was defined as ≥1 cigarette/day continuously for >6 months. Alcohol consumption was defined as at least once/week continuously for >6 months. The study was approved by the First People’s Hospital of Yancheng City Ethics Committee (Jiangsu, China) and all subjects provided written informed consent.

Specimen collection

Peripheral venous blood samples were obtained from all participants in a fasting state and stored at −20°C for subsequent use.

Extraction of genomic DNA

Wizard genomic DNA extraction kits (Promega Corporation, Madison, WI, USA) were used to extract genomic DNA according to the manufacturer’s instructions.

Concentration and purity of genomic DNA

ND-1000 UV/VIS spectrophotometer (Nanodrop Corporation, Wilmington, DE, USA) was used to perform quantitative determination of the DNA samples. Template DNA concentration was adjusted to 25–50 ng/μl.

Polymerase chain reaction (PCR)

Genotypes at the 312 and 751 loci of XPD were determined using PCR-restriction fragment length polymorphism (RFLP). Primers were synthesized by Sangon Biological Engineering Technology Co., Ltd. (Shanghai, China). For codon 312, the primer sequences used were: upstream, 5′-CTGTTGGTGGGTGCCCGTATCTGTT GGTCT-3′; downstream, 5′-TAATATCGGGGCTCACCCT GCAGCACTTCCT-3′. PCR mix (20 μl) containing 100 ng DNA template, 2.0 μl 10X PCR buffer solution (with 15 mM MgCl2), 0.4 μl dNTPs (10 mM), 1.5 μl each primer (150 mM), 1 unit Taq DNA polymerase and 16.5 μl ddH2O. Reaction conditions were as follows: 94°C for 5 min; 35 cycles of 94°C for 45 sec, 69°C for 45 sec and 72°C for 45 sec; and 72°C for 7 min. PCR products were digested using the StyI restriction enzyme (New England Biolabs, Ipswich, MA, USA) at 37°C for 16 h. For codon 751, primer sequences used were: upstream, 5′-GCCCGCTCTGGATTATACG-3′ downstream, 5′-CTATC ATCTCCTGGCCCCC-3′. PCR mix (20 μl) containing 100 ng DNA template, 2.0 μl 10Χ PCR buffer solution (Mg2+-free), 1.2 μl Mg2+ solution (25 mM MgCl2), 0.5 μl dNTPs (10 mM), 1.5 μl each primer (150 mM) and 25 units Taq DNA polymerase. Reaction conditions were as follows: 94°C for 5 min; 35 cycles of 94°C for 45 sec, 63°C for 45 sec and 72°C for 45 sec; and 72°C for 7 min. PCR products were digested using the PstI restriction enzyme (New England Biolabs) at 37°C for 16 h. Codon products were visualized on 2% agarose gels using a gel imaging system. Restriction fragments were sent to Beijing Genomics Institute (Beijing, China) for direct gene sequencing.

Statistical analysis

SPSS 17.0 statistical software was used for statistical analysis. The two-sample t-test was used to compare differences between the two groups and the χ2 test was used to compare age, gender, smoking status, alcohol consumption and genotype and allele frequencies between the two groups. The relationship between gene polymorphism and ESCC was analyzed using logistic regression to obtain odds ratios (OR) and 95% confidence intervals (CI). Analyses were performed with two-sided tests. P<0.05 was considered to indicate a statistically significant difference.

Results

Population characteristics

Age, gender, smoking status and alcohol consumption of the 400 ESCC cases and 400 healthy controls included in this study are presented in Table I. Controls were age- and gender-matched to cases. Age ranges and gender distributions of the two groups were not identified as significantly different, although ESCC occurred twice as often in males. However, statistically significant increases in cigarette smoking and alcohol consumption were observed in the ESCC patients (44.0 and 34.3%, respectively) compared with the controls (33.5 and 26%, respectively; P<0.05).

Table I Comparison of basic characteristics between ESCC cases and healthy controls.

Table I

Comparison of basic characteristics between ESCC cases and healthy controls.

Variable	Controls (n=400)	Cases (n=400)	t/χ2a	P-value
Age, years (mean ± SD)	63.2±9.6	62.9±9.5	0.498	0.618
Age, years (%)
<60	137 (34.3)	150 (37.5)	0.918	0.338
≥60	263 (65.8)	250 (62.5)
Gender (%)
Male	272 (68.0)	257 (64.3)	1.256	0.262
Female	128 (32.0)	143 (35.8)
Smoking status (%)
Non-smoking	266 (66.5)	224 (56.0)
Smoking	134 (33.5)	176 (44.0)	9.290	0.002
Alcohol consumer (%)
No	296 (74.0)	263 (65.8)
Yes	104 (26.0)	137 (34.3)	6.467	0.011

{ label (or @symbol) needed for fn[@id='tfn1-mmr-07-02-0674'] } ESCC, esophageal squamous cell carcinoma.

a t value for Age, χ² value for other variables.

Comparison of genotype and allele frequencies for SNPs 312 and 751 between ESCC and healthy individuals

The expected PCR product size for the 312 locus was 751 bp. StyI restriction digestion generated various products based on the genotype: GG resulted in 2 fragments, 507 and 244 bp; GA resulted in 4 fragments, 507, 474, 244 and 33 bp; and AA resulted in 3 fragments, 474, 244 and 33 bp (Fig. 1). For the 751 locus, the expected PCR product size was 436 bp. PstI restriction digestion generated the following: 2 fragments for the AA genotype, 290 and 146 bp; 4 fragments for AC, 290, 227, 146 and 63 bp; and 3 fragments for CC, 227, 146 and 63 bp (Fig. 2). Sequencing of these restriction products was consistent with the expected sequence (Figs. 3 and 4).

Figure 1 Electrophoretogram of three genotypes at XPD codon 312. M, 100 bp DNA ladder. Lanes 1, GG; 2, GA and 3, AA genotypes. XPD, Xeroderma pigmentosum group D.

Figure 2 Electrophoretogram of three genotypes at XPD codon 751. M, 100 bp DNA ladder. Lanes 1, AA; 2, AC and 3, CC genotypes. XPD, Xeroderma pigmentosum group D.

Figure 3 Sequencing graphs of restriction fragments for XPD codon 312. (A) GG, (B) GA, and (C) AA genotypes. XPD, Xeroderma pigmentosum group D.

Figure 4 Sequencing graphs of restriction fragments for XPD codon 751. (A) AA, (B) AC, and (C) CC genotypes. XPD, Xeroderma pigmentosum group D.

The distributions of genotypes for XPD codons 312 and 751 were consistent with Hardy-Weinberg equilibrium for cases and controls (P>0.05). No statistically significant differences in genotype or allele frequencies for codon 312 were observed between cases and controls (Table II). However, statistically significant differences in genotype and allele frequencies for codon 751 were observed between cases and controls (P<0.05; Table III).

Table II Comparison of genotype and allele frequencies for XPD codon 312 between groups, no. (%).

Table II

Comparison of genotype and allele frequencies for XPD codon 312 between groups, no. (%).

		Genotype			Allele
Group	No.	GG	GA	AA	G	A
Controls	400	351 (87.8)	47 (11.8)	2 (0.5)	749 (93.6)	51 (6.4)
Cases	400	342 (85.5)	56 (14.0)	2 (0.5)	740 (92.5)	60 (7.5)
χ2			0.903		0.784	0.784
P-value			0.637		0.376	0.376

[i] XPD, Xeroderma pigmentosum group D.

Table III Comparison of genotype and allele frequencies for XPD codon 751 between groups, no. (%).

Table III

Comparison of genotype and allele frequencies for XPD codon 751 between groups, no. (%).

		Genotype			Allele
Group	No.	AA	AC	CC	A	C
Controls	400	321 (80.3)	73 (18.3)	6 (1.5)	715 (89.4)	85 (10.6)
Cases	400	283 (70.8)	105 (26.3)	12 (3.0)	671 (83.9)	129 (16.1)
χ2			10.144		10.444	10.444
P-value			0.006		0.001	0.001

[i] XPD, Xeroderma pigmentosum group D.

Relationship between genotype and ESCC

Results of the logistic regression analysis indicated that gender was a confounding factor for ESCC (P=0.039), while age, smoking status and alcohol consumption had no correlation with the occurrence of ESCC (P>0.05). Excluding the effect of gender resulted in no correlation between SNP in codon 312 and occurrence of ESCC (P>0.05). However, a correlation was observed between codon 751 genotype and ESCC occurrence. No significant increase in ESCC risk was observed for cases with the AC genotype compared with those with the AA genotype (P=0.137). ESCC risk in cases with the CC genotype was increased by 1.6-fold (OR, 1.600; 95% CI; 1.137–2.253; P=0.007; Table IV).

Table IV Association between XPD polymorphism and ESCC.

Table IV

Association between XPD polymorphism and ESCC.

Variable	Controls (n=400)	Cases (n=400)	OR (95% CI)	P-value
Age
<60	137 (34.3)	150 (37.5)	Reference
≥60	263 (65.8)	250 (62.5)	0.866 (0.644–1.164)	0.341
Gender
Male	272 (68.0)	257 (64.3)	Reference	0.039
Female	128 (32.0)	143 (35.8)	1.391 (1.016–1.905)
Smoking status
Non-smoking	266 (66.5)	224 (56.0)	Reference
Smoking	134 (33.5)	176 (44.0)	1.507 (0.985–2.305)	0.059
Alcohol consumer
Never consume alcohol	296 (74.0)	263 (65.8)	Reference
Consume alcohol	104 (26.0)	137 (34.3)	1.177 (0.748–1.850)	0.481
Codon
312 GG	351 (87.8)	342 (85.5)	Reference
312 GA	47 (11.8)	56 (14.0)	1.166 (0.762–1.783)	0.480
312 AA	2 (0.5)	2 (0.5)	1.013 (0.138–7.441)	0.990
751 AA	321 (80.3)	283 (70.8)	Reference
751 AC	73 (18.3)	105 (26.3)	2.149 (0.785–5.884)	0.137
751 CC	6 (1.5)	12 (3.0)	1.600 (1.137–2.253)	0.007

[i] ESCC, esophageal squamous cell carcinoma. OR, odds ratio; CI, confidence interval.

Discussion

Previous studies on the association between XPD polymorphisms and ESCC are controversial. However, the present study provides additional confirmation of a correlation between a SNP in XPD, specifically within codon 751 and ESCC. No significant correlation was observed between the SNP in codon 312 and occurrence of ESCC, consistent with results of previous studies on the relationship between polymorphisms of XPD codon 312 and risk of esophageal cancer. Specifically, Zhou et al(17) and Wu et al(19) reported no correlation between polymorphisms at this locus and occurrence of ESCC in Chinese populations. Similarly, Tse et al(20) found no such correlation in a predominantly Caucasian population. By contrast, we report that a polymorphism in codon 751 of XPD is significantly correlated with occurrence of ESCC and that individuals with the CC genotype (vs. AA) have a 1.6-fold higher risk of ESCC. Thus, the CC genotype confers increased susceptibility to ESCC, consistent with specific previous studies. In particular, researchers in the Taiwan Province of China reported that the XPD codon 751 SNP is correlated with risk of esophageal cancer (21); Ye et al(16) found a similar correlation in a Swedish population. However, an earlier study in northern China identified no correlation between XPD gene polymorphisms and ESCC (19), indicating that XPD polymorphisms may have different effects on genetic susceptibility to ESCC in various populations. A previous study indicated that, compared with the AA genotype, CC individuals have significantly reduced repair capacity of chromatid-type DNA aberrations (22). The structure of codon 751 is highly evolutionarily conserved, indicating a role in maintaining function of XPD. Therefore, mutation at this locus may affect the repair ability of XPD, as well as affecting its transcriptional activity and role in apoptosis, thereby promoting tumorigenesis.

In short, XPD codon 751 polymorphism is markedly associated with ESCC susceptibility and the CC genotype may be a risk factor for ESCC. Therefore, detection of XPD polymorphism at this locus may aid in prevention and early diagnosis of ESCC. Corresponding functional studies are required to reveal molecular mechanisms that promote ESCC in individuals carrying XPD variants.

References