Association of Arg194Trp, Arg280His and Arg399Gln Polymorphisms in X-ray Repair Cross-Complementing Group 1 Gene and Risk of Differentiated Thyroid Carcinoma in Iran

Abstract

Background: X-ray repair cross-complementing group 1 (XRCC1) gene is a DNA repair gene and its non-synonymous single nucleotide polymorphisms (SNP) may influence DNA repair capacity which has been considered as a modifying risk factor for cancer development. Methods: A case-control study was conducted to investigate impact of three frequently studied polymorphisms (Arg194Trp, Arg280His and Arg399Gln) on developing differentiated thyroid carcinoma (DTC). Results: Increased risks for DTC were shown in homozygous (odds ratio [OR]: 3.66, 95% confidence interval [CI]: 0.38-35.60) and in dominant trait (OR: 1.22, 95% CI: 1.64-2.32) of Arg194Trp genotype. Also, for Arg280His genotype, an increased risk for DTC was shown in dominant trait (OR: 1.42, 95% confidence interval [CI]: 0.76-2.68), while a mildly reduction of risk for DTC (OR: 0.77, 95% [CI]: 0.50-1.17) was estimated in dominant Gln genotype of Arg399Gln. Considering combinatory effects of Arg194Trp and Arg280His genotypes on DTC, the calculated OR and 95% CI for being heterozygous for one of Arg194Trp or Arg280His genotypes were 1.57 and 0.90-2.74, respectively. Conclusion: Genotyping of codons 194, 280 and 399 in XRCC1 gene may use in risk assessment of DTC.

Introduction

Exogenous and endogenous mutagens damage DNA, which may result in uncontrolled cell proliferation or programmed cell death. To protect the integrity of genome, some DNA repair systems have been developed in variety of organisms from bacteria to mammals. A few pathways of DNA repair operate on specific types of damaged DNA. Among them, base excision repair (BER) operates on nil lesions [1] and more than 20 gene products are related to BER [2]. The X-ray repair cross-complementation group 1 (XRCC1) protein plays an important role in BER [3]. The XRCC1 is a multi-domain protein, and is required for the efficient repair of single strand breaks and damaged bases in DNA [3, 4]. This protein is essential for mammalian viability because XRCC1 deficiency in mice results in embryonic lethality. As DNA damage may result in somatic mutations and subsequent stimulation of DNA repair processes, it can be considered that polymorphisms of DNA repair genes may alter cell repair function capacity and lead to carcinogenesis [5]. The XRCC1 gene is located on chromosome 19q13.2 and 17 exons of this gene encode a protein with 633 amino acids [6]. A number of single nucleotide polymorphisms (SNP) have been reported in the Ensembl database. Among them, Arg194Trp (rs1799782), Arg280His (rs25489) and Arg399Gln (rs25487) are highly studied and caused non-conservative changes [7]. Some studies have shown a significant association between these variants and risk of different types of human cancer [8, 9]. These findings suggest that non-conservative changes may result in DNA repair capacity, although there are inconsistent reports in favor of cancer types and different populations [10, 11]. Rationally, it seems non-conservative that Arg194Trp, Arg280His and Arg399Gln substitutions may attribute to cancer susceptibility.

Thyroid carcinomas are the most frequent endocrine malignancies which among these thyroid carcinomas, more than 90 percent are differentiated thyroid carcinomas (DTC). Pathologically, DTC include papillary, follicular, and Hürthle cell carcinoma [12]. To date, exposure to ionizing radiation is the only known risk factor for thyroid cancer [13]. However, there are evidences that some gene variants including DNA repair genes influence on DTC susceptibility. XRCC1 is one of the candidate genes which its variant relationship with thyroid cancer has not been extensively studied [14].

In the present case-control study, the genotype frequency distributions of three common XRCC1 SNP (Arg194Trp, Arg280His and Arg399Gln) are compared in DTC patients and thyroid cancer-free controls using usual statistical methods.

MATERIALS AND METHODS

Cases and controls. The cases were thyroid cancer patients with DTC pathologic diagnosis after surgical excision. The controls were individuals who were thyroid cancer free. The sample volumes are indicated in Table 1. Both cases and control groups were admitted to Research Institute for Nuclear Medicine of Shariati Hospital during the period of September 2008-2009 and signed a consent form for genetic tests. Individuals with cervical lymph node involvement, any other cancer history, heavy smokers, radiation exposure history and alcohol users were excluded from the research. Controls were frequency matched to cases based on age and sex.

Table 1.

Arg194Trp (A), Arg280His (B) and Arg399Gln (C): demographic and genotype frequencies of case and control subjects.

(A)
Variable					DTC (n = 157)										Controls (n = 187)
					n			%		P a					n			%
Sex										0.596
Male					34			21.7							45			24.1
Female					123			78.3							142			75.9
Age										0.137
<= 50					131			83.4							144			77.0
> 50					26			16.6							43			23.0
Arg194Trp										0.477
Arg/Arg					136			86.6							166			88.8
Arg/Trp					18			11.5							20			10.7
Trp/Trp					3			01.9		0.357					1			00.5
Trp allele frequency								07.6										05.9
(B)
Variable				DTC (n = 170)										Controls (n = 193)
				n			%			P a				n			%
Sex										0.826
Male				38			22.4							45			23.3
Female				132			77.6							148			76.7
Age										0.173
<= 50				141			82.9							149			77.2
> 50				29			17.1							44			22.8
Arg280His										0.411
Arg/Arg				146			85.88							173			89.64
Arg/His				23			13.53							18			09.33
His/His				1			0.59			0.366				2			01.04
His allele frequency							30.3										033.4
(C)
Variable				DTC (n = 155)										Controls (n = 190)
				n			%		P a					n			%
Sex									0.700
Male				34			22							45			23.7
Female				121			78							145			76.3
Age									0.131
<= 50				130			83.9							147			77.4
> 50				25			16.1							43			22.6
Arg280His									0.401
Arg/Arg				78			50.3							83			43.7
Arg/Gln				60			38.7							87			45.8
Gln/Gln				17			11							20			10.5
Gln allele frequency							30.3		0.385								033.4

^a P<0.05 is significant

Genotype analysis. Peripheral blood was collected in tubes containing EDTA and genomic DNA was extracted from leukocytes by previously described method in literature. XRCC1 codons 194 and 399 were determined using PCR-RFLP as described previously [11, 15]. Briefly, the primers used for XRCC1 codon 194 were GCCAGGGCCCCTCCTTCAA and TACC CTCAGACCCACGAG; and for codon 399 were TCCTCCACCTTGTGCTTTCT and AGTAGTCTGC TGGCTCTGGG. The PCR reactions for two polymorphisms were the same. Each 25 μl reaction contained 10 pmol of each primer, 2.0 mM MgCl₂, 200 mM each dNTP, 1 unit of Taq DNA polymerase (CinnaGen Smart Taq, Iran) and 50-250 ng of genomic DNA. The PCR conditions for codon 194 were 95^oC

Fig. 1.

High resolution melting of XRCC1 gene in 25 samples. RR-, RH- and HH-type curves are shown as RR, RH and HH, respectively. Small box shows difference graphs.

for 2 min; 35× (95^oC for 30 s, 57^oC for 30 s and 72^oC for 45 s); 72^oC for 7 min and for codon 399 were 95^oC for 2 min; 35× (95^oC for 30 s, 61^oC for 30 s and 72^oC for 45 s); 72^oC for 7 min . PCR products of codon 194 (485 bp) and codon 399 (517 bp) were digested by PvuII and BcnI restriction enzymes (Fermentas, Switzerland), respectively. Using submarine electro-phoresis, the DNA fragments were separated in 2% agarose gel and visualized by ethidium bromide. The expected fragments for codon 194 were 485 bp (194Arg) and 396 plus 89 bp (194Trp) as well as for codon 399 were 384 and 133 bp (399Arg) and 517 bp (399Gln).

For Arg280His genotyping, we developed a high resolution melting (HRM) technique and confirmed the 280His alleles by RLFP. A 125-bp fragment of XRCC1 gene containing codon 280 was proliferated using CCC CAG TGG TGC TAA CCT AAT and CTT CTC CTC GGG GTT TGC C primers. The PCR-HRM reactions carried out in 10-μl reactions using Type-it HRM PCR Kit (Qiagen, Germany) contained 10 pmol of each primers and 20-50 ng genomic DNA. The thermocycling program was 95^oC for 5 min; 45× (95^oC for 30 s, 60^oC for 30 s and 72^oC for 10 s); 72^oC for 4 min, followed by Pre-HRM heteroduplex enrichment (95^oC for 10 s; 65^oC for 5 min) and HRM protocol was ramp from 75^oC to 95^oC, rising by 0.05^oC each step and waiting for 2 s in each step. As it was the first time that PCR-HRM used for genotyping of Arg280His change in XRCC1 gene, for confirmation, all samples showed 280His allele were subjected to RsaI restriction enzyme treatment.

Statistical analysis. The relationship between the XRCC1 codons 194, 280 and 399 genotypes and DTC risk was analyzed by two-sided Chi-squared tests. In each codon we calculated the relative risk of the homozygous patients (Trp/Trp for codon 194, His/His for codon 280 and Gln/Gln for codon 399), heterozygous patients (Arg/Trp for codon 194, Arg/His for codon 280 and Arg/Gln for codon 399) compared with the wild type Arg/Arg genotypes (all three codons) for amino acid substitutions. Moreover, the recessive and dominant effects of the mentioned substitutions were estimated (codon 194: Trp/Trp versus Arg/Trp + Arg/Arg and Trp/Trp + Arg/Trp versus Arg/Arg, respectively; codon 280: His/His versus Arg/His + Arg/Arg and His/His + Arg/His versus Arg/Arg, respectively; codon 399: Gln/Gln versus Arg/Gln + Arg/Arg and Gln/Gln + Arg/Gln versus Arg/Arg, respectively). In addition, relative risks of having combinatory heterozygous form of three codons were calculated. The relative risk was expressed as odds ratio (OR) with 95% CI using logistic regression analyses.

Based on observed genotypes, the haplotype analysis was performed using PHASE program [16] for reconstruction of haplotypes. The frequencies of haplotypes were compared by chi-squared tests and the relative risks were estimated.

Results

Sex distributions of case and control subjects in addition to detailed genotype analysis results for the 194Trp, 280His and 399Gln are demonstrated in Table 1.

Genotyping of 194Trp and 399Gln was done by PCR-RFLP method (Fig. 1), while PCR-HRM and PCR-RFLP was used for 280His codon (Fig. 2).

Fig. 2.

RFLP of samples treated by RsaI restriction enzyme. Lane 1, untreated; lanes 2, 3, 5, 6, 9 and 10, 280RH; lanes 4, 7, 8, 11, 12 and 13, 280RR; M, 100 bp DNA ladder. Gel: agarose 3.5%.

According to the Hardy-Weinberg equilibrium model, the frequencies of observed genotypes of the controls did not deviate significantly from those expected (194Trp: χ² = 0.217, df = 1, P>0.05; 280His: χ² = 3.382, df = 1, P>0.05; 399Gln: χ² = 0.159, df = 1, P>0.05).

In Arg194Trp, Arg280His and Arg399Gln genotypes, the differences in genotype frequency distributions between DTC cases and controls were not significant (P = 0.477, P = 0.411 and P = 0.401, respectively).

Calculated OR are presented in Table 2. For the Arg194Trp change, a 3.6-fold increase in risk for DTC was estimated in homozygous Trp genotype. Also, when considering Trp as recessive trait, the risk was in concordance with it (3.62 fold). By separating Arg194Trp genotypes into groups with and without the Trp allele, the OR was calculated as 1.32 in favor of having DTC. In case of Arg280His change, the dominant His genotype was associated with increased

Table 2.

Crude and adjusted odd ratio (OR) and confidence interval (CI) of case and control subjects for Arg194Trp, Arg280His and Arg399Gln alleles.

Variable	Crude OR (95% CI), P value a	Sex-adjusted OR (95% CI), P value a	Age-adjusted OR (95% CI), P value a
Arg194Trp
Trp/Trp vs Arg/Arg	3.66 (0.38-35.60), 0.226	3.52 (0.37-34.46), 0.437	3.34 (0.34-32.54), 0.137
Trp/Trp + Arg/Trp vs Arg/Argb	1.22 (0.64-2.33), 0.545	1.21 (0.64-2.31), 0.732	1.21 (0.64-2.33), 0.276
Trp/Trp vs (Arg/Arg + Arg/Trp)c	3.62 (0.37-35.19), 0.229	3.53 (0.36-34.28), 0.416	3.35 (0.34-32.58), 0.131
Arg280His
His/His vs Arg/Arg	0.59 (0.05-6.60), 0.12	0.58 (0.05-6.43), 0.82	0.55 (0.05-6.18), 0.44
His/His + Arg/His vs Arg/Argb	1.42 (0.76-2.68),0.275	1.42 (0.75-2.68), 0.547	1.42 (0.76-2.69), 0.214
His/His vs (Arg/Arg + Arg/His)c	0.57 (0.05-6.29), 0.634	0.56 (0.05-6.22), 0.87	0.53 (0.05-5.85), 0.34
Arg399Gln
Gln/Gln vs Arg/Arg	0.90 (0.44-1.85), 0.784	0.90 (0.44-1.85), 0.951	0.89 (0.43-1.83), 0.270
(Gln/Gln + Arg/Gln) vs Arg/Argb	0.77 (0.50-1.17), 0.219	0.76 (0.50-1.17), 0.419	0.77 (0.50-1.18), 0.154
Gln/Gln vs (Arg/Arg + Arg/Gln)c	1.05 (0.53-2.08), 0.895	1.05 (0.53-2.08), 0.920	1.03 (0.52-2.09), 0.314

^a P<0.05 is significant, ^bdominant T allele , ^crecessive T allele

DTC (OR: 1.42, 95% [CI] 0.76-2.68). Finally, in Arg399Gln change, the dominant Gln genotype was borderline associated with reduction of DTC. Sex and age adjusted OR showed the same trend as the crude OR for all alleles.

Association between carrying different combinatory forms of heterozygous alleles and DTC were estimated and stated as OR. Considering combination of Arg194Trp and Arg280His heterozygous genotypes, the association between "being heterozygous in only one site" was borderline significant (OR 1.57, 95% [CI] 0.90-2.74, P = 0.111). The other data were not significant and not shown.

The retrieved haplotype frequencies and OR are shown in Table 3. The OR were calculated by comparison of each haplotype with the most frequent one (CGG) in controls. No significant risk association between XRCC1 haplotypes and DTC cases was found.

Table 3.

XRCC1 haplotype frequency and risk association.

Haplotype a	Frequency (%)			DTC
	DTC	Control		OR (95% CI)	P value
CGG	43.6	38.4		1.0 (Ref)
CAG	19.9	22.3		0.79 (0.37-1.65)	0.526
CAA	15.5	20.7		0.66 (0.30-1.45)	0.308
CGA	13.1	12.4		0.93 (0.38-2.26)	0.874
TGG	5.6	04.7		1.05 (0.29-3.85)	0.942
TGA + TAG + TAA	2.3	01.5		1.35 (0.17-11.05)	0.779

^a Haplotypes follow the SNP: XRCC1 Arg194Trp, Arg280His and Arg399Gln

Discussion

For the first time in Iran, the role of XRCC1 gene polymorphisms in relation with DTC was examined in a case-control study.

The impact of changes in DNA repair genes including XRCC1 gene has been studying in growing studies to elucidate possible susceptibility to various types of cancers [17, 18]. There are significant heterogeneities among studies due to cancer type, population and site of SNP [19-21]. There are only few studies which assessed XRCC1 gene polymorphism relations with thyroid carcinomas. In a case-control study [22], a significant increased risk of DTC for the Arg194Trp homozygous polymorphic genotype was calculated ([OR: 10.4, 95% CI: (1.0-105.5), while for Arg194Trp heterozygous polymorphic genotype, the OR was estimated as: 1.4, 95% [CI] (0.9-2.1). These results are in concordance with those of our survey

(Table 4). In the same research, the calculated OR for Arg399Gln homozygous and heterozygous were 0.5, 95% [CI] (0.3-0.8) and 0.8, 95% [CI] (0.6-1.0), respectively, demonstrating a decreased risk of DTC for Arg399Gln genotypes. Again, the data is in the same direction with our results. Increased risk for DTC is also shown with Arg194Trp [23] and Arg399Gln genotypes in another case-control study [24]. For Arg280His genotype, we found a trend toward having DTC in dominant trait (OR: 1.42, 95% [CI] (0.76-2.68) which is in opposite side of the results from both mentioned studies.

Although the haplotype analysis showed no significant results, the combined relative genetic risk of dominant trait of SNP showed some trends toward having DTC: OR _Arg194Trp X OR _Arg280His X OR _Arg399Gln = 1.22 X 1.42 X 0.77 = 1.33.

All mentioned SNP studied here were missense mutations and we expected impaired function of protein. Although we found an increased risk effect of Arg194Trp and Arg280His alleles for developing DTC, a minor protective effect of Arg399Gln genotype was found for this type of carcinoma. The latter result is in contrast with our hypothesis.

The results are not significant, s are not conclusive, probably because of small sample size or other unknown phenomena in developing thyroid carcinomas. Clearly, larger population samples from different populations and following meta-analysis assays will give better estimation of above SNP' association with DTC.