DNA Repair Gene Polymorphisms at XRCC1 (Arg194Trp,Arg280His, and Arg399Gln) in a Healthy Tunisian Population:

Interethnic Variation and Functional Prediction

Ghada Ben Salah,1 Imen Ayadi,2 Nourhene Fendri-Kriaa,1 Fakhri Kallabi,1 Emna Mkaouar-Rebai,1

Amine Fourati,3 Faiza Fakhfakh,1 Hammadi Ayadi,2 and Hassen Kamoun1

The genetic polymorphisms in DNA repair genes might affect the repair activities of the enzymes, predisposingindividuals to cancer risk. Due to these genetic variants, interethnic differences in DNA repair capacity wereobserved in various populations. Hence, our study aimed to determine the prevalence of three nonsynonymoussingle-nucleotide polymorphisms (SNPs) in an X-ray repair cross-complementation group 1 gene (XRCC1)(Arg194Trp, Arg280His, and Arg399Gln) in a healthy Tunisian population (TUN) and to compare that withHapMap (www.hapmap.org) populations. Also, we predicted their eventual functional effect based on theprotein conservation analysis by Sorting Intolerant From Tolerant (SIFT; http://sift.jcvi.org/www/SIFT_dbSNP.html) software. The genotypes of 154 healthy individuals were determined by the polymerasechain reaction–restriction fragment length polymorphism. Tunisians showed a relative relatedness with Cau-casians (European ancestry) for Arg194Trp and Arg399Gln that may be explained by the strategic geographiclocation of Tunisia in the Mediterranean, allowing exchanges with European countries. However, a characteristicpattern was observed in Arg280His polymorphism, which could be explained by the high inbreeding rate inTUN. The analysis of protein conservation showed that the three amino acid substitutions were predicted asdamaged. The results presented here provide the first report on XRCC1 polymorphisms about Tunisians andmay establish baseline database for our future clinical and genetic studies.

Introduction

DNA bearing indispensable inheritance informationmust remain stable in order to undertake its crucial

physiologic functions, but it is persistently vulnerable to en-dogenous and/or exogenous damage; thus, mutations canaccumulate, and carcinogenesis may occur. The DNA repairsystem plays a pivotal role in maintaining the functions ofnormal cells and genome integrity (Richard et al., 2001;Branzei and Foiani, 2008).

X-ray repair cross-complementation gene 1 (XRCC1), one ofthe DNA repair proteins, plays an important role in base ex-cision repair (BER). XRCC1 is a scaffold protein that interactswith many other components of BER as DNA polymerase b,APE1, hOGG1, poly-(ADP-ribose) polymerase, and DNAligase III in the NH2-terminal, central, and COOH-terminalregions, respectively, as summarized in Figure 1 (Vidal et al.,2001; Marsin et al., 2003; Audebert et al., 2004; Brem and Hall,

2005). Eight common single-nucleotide polymorphisms (SNPs)leading to an amino acid substitution have been described inXRCC1 (Ruttan and Glickman, 2002); three of which are locatedin or near important protein domains (Shen et al., 1998) (Fig. 1).

The C > T (rs1799782) polymorphism causing arginine totryptophane substitution at codon 194 of exon 6 (Arg194Trp)occurs near the N-terminal domain. The 194Trp variant allelehas been associated with a decrease in cancer risk (Goodeet al., 2002; Hung et al., 2005a, 2005b). This unexpected pro-tective effect has been confirmed by experimental studies(Wang et al., 2003; Takanami et al., 2005).

The G > A (rs25489) polymorphism causing arginine tohistidine substitution at codon 280 of exon 9 (Arg280His)occurs near the BRCT1 domain. It has been reported that the280His variant allele was associated with a slight increase incancer risk (Hu et al., 2005). However, an approximate 3.5-foldincrease in skin cancer risk has been, recently, reported inmeta-analysis study (Zhang et al., 2011).

1Laboratory of Human Molecular Genetics, Faculty of Medicine, University of Sfax, Sfax, Tunisia.2Centre of Biotechnology, University of Sfax, Sfax, Tunisia.3Research Unit, Chemical Tunisian Group, Sfax, Tunisia.

GENETIC TESTING AND MOLECULAR BIOMARKERSVolume 16, Number 10, 2012ª Mary Ann Liebert, Inc.Pp. 1218–1225DOI: 10.1089/gtmb.2012.0035

1218

The G > A (rs25487) polymorphism causing arginine toglutamine substitution at codon 399 of exon 10 (Arg399Gln)occurs at a conserved residue in the poly (ADP-ribose) poly-merase-binding domain of XRCC1. The Arg399Gln poly-morphism has been extensively investigated. A positiveassociation between the 399Gln variant allele and cancer riskhas been reported in several studies (Divine et al., 2001; Huanget al., 2011; Samulak et al., 2011; Wu et al., 2011; Zhou et al.,2011; Yin et al., 2012). However, no overall association hasbeen seen by others (Smith et al., 2003a, 2003b; Wang et al.,2009; Xue et al., 2011).

The variant allele frequencies of Arg194Trp, Arg280His,and Arg399Gln polymorphisms in the human population are0.131, 0.061, and 0.266, respectively, according to the NCBIdbSNP (www.ncbi.nlm.nih.gov/projects/SNP/) database.However, a difference in their prevalence has been reportedamong ethnic groups as well as within the same ethnic group,according to the HapMap(www.hapmap.org) project data-base. To the best of our knowledge, there is no study in thisregard in the Tunisian population (TUN) and, therefore, fur-ther investigation is merited to establish our genotype profile.

Hence, our present study aimed to determine the genotypeand the allele frequencies of XRCC1 polymorphismsArg194Trp, Arg280His, and Arg399Gln in the healthy TUN,and to compare them with those reported in the HapMap(www.hapmap.org) populations from other ethnic groups. Inaddition, we predicted their eventual functional effect basedon the protein conservation analysis by SIFT (http://sift.jcvi.org/www/SIFT_dbSNP.html) software, in order to dis-tinguish between neutral and deleterious amino acid substi-tutions for further clinical and genetic studies.

Materials and Methods

Studied population and DNA extraction

In the present study, 154 healthy unrelated individuals (56women and 98 men) from Tunisia were enrolled after ap-proval from the ethics committee of the institute. Their ageranged from 20 to 60 years, with a mean – SD of 37.5 – 11.Information about social habits and health problems of eachindividual were gathered through standardized question-naire. We excluded those who had a history of cancer. Blood

samples (5 mL) were collected after getting proper writteninformed consent, and processed for DNA extraction usingphenol-chloroform standard procedure (Lewin and Stewart-Haynes, 1992).

Analysis of XRCC1 polymorphisms

The analysis of XRCC1 polymorphisms was carried outwith polymerase chain reaction–restriction fragment lengthpolymorphism (PCR-RFLP) using primers F: 5¢-GCCCCGTCCCAGGTA-3¢ and R: 5¢-AGCCCCAAGCCCTTTCACT-3¢ for exon 6 (491 bp) (Ruth et al., 1999); F: 5¢-TGGGGCCTGGATTGCTGGGTCTG-3¢ and R: 5¢-CAGCACCACTACCACACCCTGGCA-3¢ for exon 9 (280 bp) (Tuimala et al., 2002); andF: 5¢-GCCCCTCAGATCACACCTAC-3¢ and R: 5¢-CATTGCCCAGCACAGGATAA-3¢ for exon 10 (214 bp) (Vodickaet al. 2004).

These fragments were amplified separately under the fol-lowing conditions: a reaction mixture (50mL) contained ap-proximately 50 ng of genomic DNA, 10 mM of each primer,2 mM MgCl2, 10 mM mix dNTP, 1 · PCR buffer, and 1.5 U TaqDNA polymerase. The mixtures were amplified with a Gen-eAmp PCR System 9700 Thermal cycler (Applied Biosys-tems). The PCR profile consisted of an initial melting step of95�C for 5 min, followed by 35 cycles with melting at 94�C for50 s, annealing at 62�C (exon 6) for 60 s, 71.5�C (exon 9) for50 s, and 61�C (exon 10) for 30 s, and elongation at 72�C for 50 swith a final elongation step of 72�C for 10 min. The PCRproducts were checked on a 2% agarose gel and photo-graphed using Gel doc and were then subjected to RFLPanalysis.

The restriction enzyme MspI (Fermentas) was used to dis-tinguish the codon 194 C/T polymorphism of exon 6, inwhich a loss of MspI restriction site occurs in the polymorphicallele. PCR-RFLP patterns resulted in three bands of 178, 292,and 21 bp in the homozygous wild type (Arg/Arg), whereasin the heterozygous mutant type (Arg/Trp), four bands at313, 292, 178, and 21 bp were produced. The same restrictionenzyme MspI was used to distinguish the codon 399 G/Apolymorphism at exon 10, in which a created MspI restrictionsite occurs in the polymorphic allele. The homozygous wildtype (Arg/Arg) has the PCR product band 214 bp, whereasthe homozygous mutant type (Gln/Gln) has two bands of 160

FIG. 1. Linear schematic ofhuman XRCC1 gene and proteinstructure. (a) The diagram showsthe structure of the gene with themost common and studied single-nucleotide polymorphisms (SNPs):R194W, R280H, and R399Q. Eachof them is detailed in the text. (b)Indicated are the four majorconserved domains of XRCC1: TheN-terminal domain (NTD), thenuclear localization signals (NLS),the BRCT1, and the BRCT2domains. (c) The domains ofXRCC1 that interact with specifiedprotein partners of BER as DNApolymerase b, APE1, hOGG1,poly-(ADP-ribose) polymerase(PARP), and Ligase III.

DNA REPAIR GENE POLYMORPHISMS AT XRCC1 1219

and 54 bp. The restriction enzyme RsaI (Fermentas) was usedto distinguish the codon 280 G/A polymorphism of exon 9,in which a created RsaI restriction site occurs in the poly-morphic allele. The homozygous wild type (Arg/Arg) has thePCR product band (280 bp), and the homozygous mutanttype (His/His) resulted in two fragments of 140 and 140 bp.

Digestion of the PCR products was carried out using 10 Uof MspI or RsaI and the 10 · buffer supplied with each re-striction enzyme at 37�C overnight. The digestion productswere separated on a 2.5% and 3.5% agarose gel and photo-graphed using Gel doc film (Bio-Rad).

Bioinformatics analysis

The genotype and the allele frequencies of XRCC1 poly-morphisms (Arg194Trp, Arg280His, and Arg399Gln) in theother ethnic groups were obtained from the HapMap(www.hapmap.org) database. The pairwise linkage disequi-librium (LD) (r2) was calculated using the Haploview software(version 4.2) and HapMap (www.hapmap.org) data. The pos-sible haplotypes in our population and their estimated fre-quencies were performed using PHASE software (version 2.1).

Evolutionary protein conservation was examined bySIFT (http://sift.jcvi.org/www/SIFT_dbSNP.html) software(version 1.03) developed by Ng and Henikoff (2001, 2002).SIFT (http://sift.jcvi.org/www/SIFT_dbSNP.html) predictswhether an amino acid substitution affects protein function.SIFT (http://sift.jcvi.org/www/SIFT_dbSNP.html) predic-tion is based on the degree of conservation of amino acidresidues in sequence alignments derived from closely relatedsequences, collected through PSI-BLAST. The median se-quence conservation ranges from 0 to 4.32; ideally the numberwould be between 2.75 and 3.5, the default median sequenceconservation in the range of 3.0. The amino acid substitution ispredicted damaging if the SIFT score is £ 0.05, and tolerated ifthe score is > 0.05 (The SIFT score ranges from 0 to 1.).

Statistical analysis

The allele frequency was calculated from the genotypefrequency. Hardy–Weinberg equilibrium was examined forall SNPs, using a Chi-square (w2) test with one degree offreedom. The 95% confidence interval (CI) of the genotypefrequencies was calculated based on a binomial distribution.The comparison of genotype frequency distributions as agender function was performed by w2 test. The differencebetween TUN and HapMap populations was examined by w2

test, using the allele frequencies, and a p-value < 0.05 wasconsidered statistically significant. These calculations wereconducted by the computer SPSS program (Version 17.0).

Results and Discussion

The discordance in genetic variation observed betweenethnicities highlights the necessity for researchers to establisha specific genotype profile for each population, in order toprioritize substitutions for further studies. In the present one,we studied the major three SNPs in XRCC1 gene asArg194Trp (C > T, rs1799782), Arg280His (G > A, rs25489),and Arg399Gln (G > A, rs25487).

The genotype and the allele frequencies of XRCC1 poly-morphisms in our population are summarized in Table 1. Thegenotype distributions showed no statistical difference as agender function ( p > 0.05; Table 1). The observed genotypefrequencies are consistent with Hardy–Weinberg expectations( p > 0.05; Table 1).

Genotype and allele frequencies of XRCC1(C > T) Arg194Trp

The observed CC, CT, and TT genotype frequencies were0.863, 0.137, and 0.000, respectively (Table 1). The HapMapdata for this SNP were only available for the ethnic groups,Caucasian Europeans (CEU), Asians (HCB and JPT), and Sub-

Table 1. Genotype Distributions and Allele Frequencies of XRCC1 Polymorphisms

Among Tunisian Population

Total Women Men Allele frequency

Genotype (n) Frequencya (n) Frequency (n) Frequency Wild-type allele Variant alleleHWEb

(p-value)

rs1799782 C > T (Arg194Trp)Total 154 56 98C/C 133 0.863 47 0.839 86 0.877 (C) (T) 0.45C/T 21 0.137 9 0.161 12 0.123 0.931 0.069T/T 0 0.000 0 0.000 0 0.000c

rs25489 G > A (Arg280His)Total 154 56 98G/G 99 0.643 36 0.643 63 0.643 (G) (A) 0.32G/A 47 0.305 14 0.250 33 0.337 0.795 0.205A/A 8 0.052 6 0.107 2 0.020c

rs25487 G > A (Arg399Gln)Total 154 56 98G/G 64 0.416 25 0.446 39 0.398 (G) (A) 0.98G/A 69 0.448 26 0.464 43 0.349 0.640 0.360A/A 21 0.136 5 0.090 16 0.103c

n, Number of subjects.aGenotype frequencies.bThere was no deviation from the expected Hardy–Weinberg frequencies ( p > 0.05).cp > 0.05, no difference as a function of gender.

1220 BEN SALAH ET AL.

Saharan Africans (YRI). The distribution of genotype fre-quencies with 95% of CI for Tunisians (TUN) and the otherethnic groups is shown graphically in Figure 2. The wild-type(C) and variant allele (T) frequencies for Tunisians (TUN) andthe other HapMap groups are shown in Table 2. The preva-lence of Arg194Trp polymorphism in the Tunisian population(TUN) is consistent with those reported in both Caucasian(CEU) and Sub-Saharan African (Negroid YRI) populations( p > 0.05; Table 2). However, a clear discordance was ob-served compared with those observed in Asian populations( JPT and HCB) ( p < 0.05; Table 2).

Genotype and allele frequencies of XRCC1(G > A) Arg280His

The observed GG, GA, and AA genotype frequencies were0.643, 0.305, and 0.052, respectively (Table 1). The distributionof genotype frequencies with 95% of CI for Tunisians (TUN)and the other ethnic groups is shown graphically in Figure 3.The wild-type (G) and variant allele (A) frequencies for Tu-nisians (TUN) and the other HapMap groups are shown inTable 2. The Tunisian population appears to have a signifi-cantly higher prevalence compared with all other ethnicgroups included in our study, such as Caucasians (CEU andTSI), Asians (HCB, CHD, GIH, and JPT), Africans Sub-Saharan (LWK, MKK, and YRI), American-Africans (ASW),and American-Mexicans (MEX) ( p < 0.05; Table 3).

This unexpected data could be explained by the high rate ofinbreeding in the Tunisian population being the most im-portant effect. It has been reported that the unions betweenTunisian relatives represent more than 32%, which may reach60% in rural areas (Riou et al., 1989; Ben Arab et al., 2004;Kerkeni et al., 2007). This endogamy creates groups in thepopulation where chances that two gametes carrying thesame genetic information meet each other are more elevatedthan in the general population. Therefore, endogamy has di-rect consequences on distribution, structure, heterogeneity,and homozygosity tendency in Tunisian population (BenMrad and Chalbi, 2004).

Genotype and allele frequencies of XRCC1(G > A) Arg399Gln

The observed GG, GA, and AA genotype frequencies were0.416, 0.448, and 0.136, respectively (Table 1). The distributionof genotype frequencies with 95% of CI for Tunisians (TUN)and the other ethnic groups is shown graphically in Figure 4.The wild-type (G) and variant allele (A) frequencies for Tu-nisians (TUN) and the other HapMap groups are shown inTable 2. Unlike 194 and 280 polymorphisms, the Arg399Glnpolymorphism showed a relatively high prevalence in Tuni-sian population and in all HapMap groups. Compared withother groups, a similarity was seen between Tunisian popu-lations and either Caucasian (CEU and TSI), Asian (GIH,

FIG. 2. Genotype frequencies of C > T (Arg194Trp) poly-morphism in Tunisian (TUN) and HapMap populations( JPT, HCB, CEU, and YRI) (www.hapmap.org). Circle: fre-quency of homozygous wild-type (CC); square: frequency ofheterozygous (CT); and triangle: frequency of homozygousmutant type (TT); bars are 95% confidence intervals.

Table 2. Allele Frequencies of XRCC1 Arg194Trp

in HapMap and Tunisian (TUN) Populations

Allele frequency

PopulationWild-type

alleleVariant

allele

Pairwisep-value between

TUN andother populations

C > T (Arg194Trp) C TCEU (n = 120) 0.908 0.092 0.53HCB (n = 90) 0.756 0.244 6 10 - 4

JPT (n = 90) 0.722 0.278 8 10 - 5

YRI (n = 120) 0.917 0.083 0.68TUN (n = 154) 0.931 0.069

n, Number of individuals.Population data source: (www.hapmap.org).CEU, Utah residents with Northern and Western European

ancestry from CEPH collection; HCB, Han Chinese in Beijing, China;JPT, Japanese in Tokyo, Japan; YRI, Yoruban in Ibadan, Nigeria.

p < 0.05, a significant difference between TUN and other popula-tions.

p > 0.05, no significant difference between TUN and other popu-lations.

FIG. 3. Genotype frequencies of G > A (Arg280His) poly-morphism in Tunisian (TUN) and HapMap populations(HCB, GIH, TSI, CHD, JPT, MEX, CEU, MKK, ASW, LWK,and YRI) (www.hapmap.org). Circle: frequency of homozy-gous wild-type (GG); square: frequency of heterozygous(GA); and triangle: frequency of homozygous mutant type(AA); bars are 95% confidence intervals.

DNA REPAIR GENE POLYMORPHISMS AT XRCC1 1221

HCB, CHD, and JPT), and American-Mexican (MEX) popu-lations ( p > 0.05; Table 2). However, a significant differencewas shown compared with American-Africans (ASW) andAfrican Sub-Saharans (LWK and MKK and YRI) ( p < 0.05;Table 2).

LD and haplotype analysis

Calculations of pairwise LD showed an equilibrium statusfor XRCC1 polymorphisms; r2 was 0, 0.01, and 0.01 for(rs25489/rs25487), (rs1799782/rs25489), and (rs1799782/

rs25487), respectively. Haplotype distributions are shown inTable 4. Four out of 8 possible haplotypes were observed inour population with frequencies greater than 5%. The mostfrequent haplotype was GGC (0.455 – 0.009) that is composedof three wild-type alleles (Table 4), followed by GAT haplo-type (0.279 – 0.008) that contained the variant alleles of G > A(rs25489) and C > T (rs1799782). To elucidate the influence ofXRCC1 haplotypes on cancer risk, Saadat evaluated whetherthe Arg194Trp and Arg399Gln haplotypes contribute tobreast cancer susceptibility among Caucasoid subjects (Saa-dat, 2010). Saadat reported a borderline increased risk ofbreast cancer associated with the Arg194-Gln399 haplotypeversus the Arg194-Arg399 haplotype. However, a higher as-sociation has been observed between head and neck cancerand XRCC1 haplotypes, especially with Arg194Trp-Arg399Arg haplotype (Kowalski et al., 2009). To the best of

FIG. 4. Genotype frequencies of G > A (Arg399Gln) poly-morphism in Tunisian (TUN) and HapMap populations(GIH, CEU, TSI, MEX, CHD, JPT, HCB, MKK, ASW, YRI,and LWK) (www.hapmap.org). Circle: frequency of homo-zygous wild-type (GG); square: frequency of heterozygous(GA); triangle: frequency of homozygous mutant type (AA);bars are 95% confidence intervals.

Table 3. Allele Frequencies of XRCC1 Arg280His and Arg399Gln in HapMap

and Tunisian (TUN) Populations

Allele frequency Pairwise p-valuebetween TUN

and otherpopulations

Allele frequency Pairwise p-valuebetween TUN

and otherpopulationsPopulation

Wild-typeallele

Variantallele Population

Wild-typeallele

Variantallele

G > A (Arg280His) G A G > A (Arg399Gln) G ACEU (n = 226) 0.956 0.044 7 · 10 - 4 CEU (n = 224) 0.634 0.366 0.92TSI (n = 204) 0.902 0.098 0.04 TSI (n = 204) 0.634 0.366 0.96HCB (n = 274) 0.901 0.099 0.043 HCB (n = 270) 0.748 0.252 0.09CHD (n = 218) 0.913 0.087 0.021 CHD (n = 218) 0.716 0.284 0.25GIH (n = 202) 0.901 0.099 0.043 GIH (n = 202) 0.604 0.396 0.59JPT (n = 226) 0.920 0.080 0.013 JPT (n = 226) 0.730 0.270 0.17LWK (n = 220) 0.986 0.014 2 · 10 - 5 LWK (n = 218) 0.899 0.101 10 - 5

MKK (n = 312) 0.962 0.038 3 · 10 - 4 MKK (n = 312) 0.814 0.186 5 · 10 - 3

YRI (n = 294) 0.986 0.014 2 · 10 - 5 YRI (n = 294) 0.895 0.105 2 · 10 - 5

ASW (n = 112) 0.973 0.027 10 - 4 ASW (n = 114) 0.825 0.175 3 · 10 - 3

MEX (n = 114) 0.921 0.079 0.012 MEX (n = 114) 0.693 0.307 0.42TUN (n = 154) 0.795 0.205 TUN (n = 154) 0.640 0.360

n, Number of individuals.Population data source: (www.hapmap.org).CEU, Utah residents with Northern and Western European ancestry from the CEPH collection; TSI, Tuscan in Italy; HCB, Han Chinese in

Beijing, China; CHD, Chinese in Metropolitan Denver, Colorado; GIH, Gujarati Indians in Houston, Texas; JPT, Japanese in Tokyo, Japan;LWK, Luhya in Webuye, Kenya; MKK, Maasai in Kinyawa, Kenya; YRI, Yoruban in Ibadan, Nigeria; MEX, Mexican ancestry in Los Angeles,California; ASW, African ancestry in Southwest United States.

p < 0.05, significant difference between TUN and other populations.p > 0.05, no significant difference between TUN and other populations.

Table 4. The Frequencies of the Possible

Haplotypes for XRCC1 Polymorphisms

in Tunisian Population (n = 154)

Index Haplotype Frequency Standard error

rs25487/rs25489/rs17997821 G G C 0.455 0.0092 G G T 0.043 0.0053 G A C 0.132 0.0084 G A T 0.005 0.0035 A G C 0.279 0.0086 A G T 0.016 0.0057 A A C 0.064 0.0078 A A T 0.002 0.002

The major haplotypes are in bold (frequency > 5%).

1222 BEN SALAH ET AL.

our knowledge, there is no study pertaining to the XRCC1haplotypes in our population; therefore, these data would beuseful in epidemiologic and experimental studies. However,largest population studies might be more suitable to assess theeffect of each haplotype.

Effect of Arg194Trp substitution on protein function

The arginine (R) to tryptophane (W) substitution at 194codon was predicted as damaging with a SIFT score of 0.00(Table 5). In our analysis, the SIFT prediction was based on 98homologous proteins in the alignments. The Arg194Trp sub-stitution occurs near the N-terminal domain of the XRCC1gene (Shen et al., 1998; Marintchev et al., 1999), within possiblebinding domains of several interactive protein partners, in-cluding APE1, polymerase b, hOGG1, and PARP, as de-scribed in Figure 1 (Ladiges, 2006). This area is rich in proline,serine, arginine, and lysine residues, and, therefore, the tran-sition from positively charged arginine to a hydrophobictryptophan may affect binding and DNA repair efficiency(Ladiges, 2006).

Effect of Arg280His substitution on protein function

The arginine (R) to histidine (H) substitution at 280 codonwas predicted as damaging with a SIFT score of 0.05 (Table 5).The Arg280His substitution occurs near the BRCT1 domain(breast cancer susceptibility protein-1 domain) of the XRCC1gene (Fig. 1) (Shen et al., 1998), and lies in between the DNApolymerase b and PARP binding areas (Thompson and West,2000). According to Caldecott (2003), the transition from apositively charged arginine to histidine within the conservedregion could potentially alter XRCC1 function and its abilityto interact with the apurinic/apyrimidinque endonuclease(APE)–binding domain. All these findings highlight the del-eterious effect of Arg280His polymorphism on XRCC1 func-tion. In this regard, the 280His variant allele has beensuggested to confer increased mutagen sensitivity (Tuimalaet al., 2002).

Effect of Arg399Gln substitution on protein function

The arginine (R) to glutamine (Q) substitution at 399 co-don was predicted as damaging with a SIFT score of 0.05(Table 5). This substitution occurs within the breast cancer

C-terminal BRCT domain at a conserved residue in the poly(ADP-ribose) polymerase PARP binding domain (Zhanget al., 1998; Ladiges, 2006). Taylor et al. (2002) showed thatalthough BRCT1 domain is critical for efficient single-strandbreak repair and cell survival, Arg399Gln polymorphismlocated within this domain did not appear to significantlyaffect XRCC1 function. On the contrary, by using moleculardynamics techniques, Monaco et al. (2007) predicted thestructure of wild-type and polymorphic form of BRCT1 do-main of XRCC1 demonstrating that the polymorphism inexon 10 changed the XRCC1’s secondary structure (Tayloret al., 2002; Monaco et al., 2007). Therefore, any change in theBRCT structure has the potential to lead to defects in thedetection of the DNA damage and hence the activation ofthe BER pathway. In fact, several reports suggested that 399polymorphism may result in deficient DNA repair and in-dividuals with Gln variant allele exhibited higher level ofDNA damage than those with homozygous wild-type (Arg/Arg) and heterozygous (Arg/Gln) genotypes (Lunn et al.,1999; Abdel-Rahman and El-Zein, 2000; Miller et al., 2001;Matullo et al., 2001).

In the overall, our findings revealed an interethnic vari-ability in prevalence of XRCC1 polymorphisms. Tunisia withits high levels of endogamy could provide a unique resourcefor dissecting complex disease etiology. Our findings lead usto suggest that the Tunisian population could be more variedin the capacity of DNA repair than other populations in theworld, and therefore the susceptibility to disease incidencesdue to environmental pollutants will vary; thus, extrapolatedresults from other populations cannot be applied on the Tu-nisian population. The present study establishes the preva-lence and the possible effects of XRCC1 polymorphisms. Thisdata will provide a basic database for the future clinical andgenetic studies pertaining to variability and the defect of DNArepair capacity.

Acknowledgments

The authors would like to thank the volunteers for theircooperation. We were also grateful to Pr. Jamil JAOUA,Founder and former head of the English Department at theSfax Faculty of Science, for proofreading this article. Thiswork was supported by the Ministry of Higher Education andScientific Research in Tunisia and Chemical Tunisian Group.

Table 5. SIFT Evolutionary Conservation Status of XRCC1 Polymorphisms

dbSNPs SubstitutionProtein

ID AaaSIFT

predictionSIFTscore

Medianinfo

Number ofsequence at position

rs1799782 R194W NP_006288 R TOLERATED 1.00 2.42 98W DAMAGING 0.00 2.42 98

rs25489 R280H NP_006288 R TOLERATED 1.00 2.35 98H DAMAGING 0.05 2.35 98

rs25487 R399Q NP_006288 R TOLERATED 1.00 2.35 98Q DAMAGING 0.05 2.35 98

aAmino acid.SIFT score: The amino acid substitution is predicted DAMAGING if the SIFT score £ 0.05, and TOLERATED if SIFT score > 0.05.Median Info: This is used to measure the diversity of the sequences used for prediction; it ranges from 0 to 4.32; ideally the number would

be between 2.75 and 3.5.The number of sequences at position: The number of homologue sequences that have an amino acid at the position prediction (SIFT

automatically chooses the sequences).

DNA REPAIR GENE POLYMORPHISMS AT XRCC1 1223

Author Disclosure Statement

The authors declare that they have no actual or potentialconflict of interest, including any financial, personal, or otherrelationships, with other people or organizations concerningthis work.

References

Abdel-Rahman SZ, El-Zein RA (2000) The 399Gln polymorphismin the DNA repair gene XRCC1 modulates the genotoxic re-sponse induced in human lymphocytes by the tobacco-specificnitrosamine NNK. Cancer Lett 159:63–71.

Audebert M, Salles B, Calsou P (2004) Involvement of poly(ADP-ribose) polymerase-1 and XRCC1/DNA ligase III in analternative route for DNA double-strand breaks rejoining.J Biol Chem 279:117–126.

Ben Arab S, Masmoudi S, Beltaief N, et al. (2004) Consanguinityand endogamy in Northern Tunisia and its impact on non-syndromic deafness. Genet Epidemiol 27:74–79.

Ben Mrad L, Chalbi N (2004) Is the matrimonial choice trans-mittable in Tunisia? Antropo 7:31–37.

Branzei D, Foiani M (2008) Regulation of DNA repair through-out the cell cycle. Nat Rev Mol Cell Biol 9:297–308.

Brem R, Hall J (2005) XRCC1 is required for DNA single-strand break repair in human cells. Nucleic Acids Res 8:2512–2520.

Caldecott KW. XRCC1 and DNA strand break repair. DNARepair (Amst) 2003;2:955–969.

Divine KK, Gilliland FD, Crowell RE, et al. (2001) The XRCC1399 glutamine allele is a risk factor for adenocarcinoma of thelung. Mutat Res 461:273–278.

Goode EL, Ulrich CM, Potter JD (2002) Polymorphisms in DNArepair genes and associations with cancer risk. Cancer Epi-demiol Biomarkers Prev 12:1513–1530.

Hu Z, Ma H, Chen F, et al. (2005) XRCC1 polymorphisms andcancer risk: a meta-analysis of 38 case-control studies. CancerEpidemiol Biomarkers Prev 7:1810–1818.

Huang GL, Guo HQ, Yu CY, et al. (2011) XRCC1 polymorphismsand risk of nasopharyngeal carcinoma: a meta-analysis. AsianPac J Cancer Prev 9:2329–2333.

Hung RJ, Brennan P, Canzian F, et al. (2005a) Large-scale investi-gation of base excision repair genetic polymorphisms and lungcancer risk in a multicenter study. J Natl Cancer Inst 8:567–576.

Hung RJ, Hall J, Brennan P, et al. (2005b) Genetic polymorphismsin the base excision repair pathway and cancer risk: a HuGEreview. Am J Epidemiol 10:925–942.

Kerkeni E, Monastiri K, Saket B, et al. (2007) Interplay of socio-economic factors, consanguinity, fertility, and offspring mor-tality in Monastir, Tunisia. Croat Med J 48:701–707.

Kowalski M, Przybylowska K, Rusin P, et al. (2009) Geneticpolymorphisms in DNA base excision repair gene XRCC1 andthe risk of squamous cell carcinoma of the head and neck.J Exp Clin Cancer Res 28:37.

Ladiges WC (2006) Mouse models of XRCC1 DNA repairpolymorphisms and cancer. Oncogene 25:1612–1619.

Lewin HA, Stewart-Haynes JA (1992) A simple method for DNAextraction from leukocytes for use in PCR. Biotechniques4:522–524.

Lunn RM, Langlois RG, Hsieh LL, et al. (1999) XRCC1 poly-morphisms: effects on aflatoxin B1-DNA adducts and glyco-phorin A variant frequency. Canc Res 59:2557–2561.

Marintchev A, Mullen MA, Maciejewski MW, et al. (1999) So-lution structure of the single-strand break repair proteinXRCC1 N-terminal domain. Nat Struct Biol 6:884–893.

Marsin S, Vidal AE, Sossou M, et al. (2003) Role of XRCC1 in thecoordination and stimulation of oxidative DNA damage re-pair initiated by the DNA glycosylase hOGG1. J Biol Chem45:44068–44074.

Matullo G, Palli D, Peluso M, et al. (2001) XRCC1, XRCC3, XPDgene polymorphisms, smoking and (32) P-DNA adducts in asample of healthy subjects. Carcinogenesis 22:1437–1445.

Miller MC 3rd, Mohrenweiser HW, Bell DA (2001) Geneticvariability in susceptibility and response to toxicants. ToxicolLett 120:269–280.

Monaco R, Rosal R, Dolan MA, et al. (2007) Conformational ef-fects of a common codon 399 polymorphism on the BRCT1domain of the XRCC1 protein. Protein J 8:541–546.

Ng PC, Henikoff S (2001) Predicting deleterious amino acidsubstitutions. Genome Res 11:863–874.

Ng PC, Henikoff S (2002) Accounting for human polymorphismspredicted to affect protein function. Genome Res 12:436–446.

Richard D, Wood L, Michael M, et al. (2001) Human DNA repairgenes. Science 291:1284–1289.

Riou S, El Younsi C, Chaabouni H (1989) Consanguinity in thepopulation of Northern Tunisia. Tunis Med 67:167–172.

Ruth ML, Ronald GL, Ling LH, et al. (1999) XRCC1 Polymorph-isms: effects on aflatoxin B1-DNA adducts and glycophorinA variant frequency. Cancer Res 59:2557–2561.

Ruttan CC, Glickman BW (2002) Coding variants in humandouble-strand break DNA repair genes. Mutat Res 509:175–200.

Saadat M (2010) Haplotype analysis of XRCC1 (at codons 194and 399) and susceptibility to breast cancer, a meta-analysis ofthe literatures. Breast Cancer Res Treat 3:785–791.

Samulak D, Romanowicz-Makowska H, Smolarz B, et al. (2011)Association between Arg399Gln polymorphism of X-ray repaircross-complementing 1 (XRCC1) gene and sporadic endome-trial cancer in the Polish population. Eur J Gynaecol Oncol5:491–500.

Shen MR, Jones IM, Mohrenweiser H (1998) Nonconservativeamino acid substitution variants exist at polymorphic fre-quency in DNA repair genes in DNA repair genes in healthyhumans. Cancer Res 58:604–608.

Smith TR, Levine EA, Perrier ND, et al. (2003a) DNA-repairgenetic polymorphisms and breast cancer risk. Cancer Epi-demiol Biomarkers Prev 12:1200–1204.

Smith TR, Miller MS, Lohman K, et al. (2003b) Polymorphisms ofXRCC1 and XRCC3 genes and susceptibility to breast cancer.Cancer Lett 190:183–190.

Takanami T, Nakamura J, Kubota Y, et al. (2005) The Arg280Hispolymorphism in X-ray repair cross-complementing gene 1impairs DNA repair ability. Mutat Res 582:135–145.

Taylor RM, Thistlethwaite A, Caldecott KW (2002) Central rolefor the XRCC1 BRCT I domain in mammalian DNA single-strand break repair. Mol Cell Biol 8:2556–2563.

Thompson LH, West MG (2000) XRCC1 keeps DNA from get-ting stranded. Mutat Res 459:1–18.

Tuimala J, Szekely G, Gundy S, et al. (2002) Genetic poly-morphisms of DNA repair and xenobiotic-metabolizingenzymes: role in mutagen sensitivity. Carcinogenesis 23:1003–1008.

Vidal AE, Boiteux S, Hickson ID, et al. (2001) XRCC1 coordinatesthe initial and late stages of DNA abasic site repair throughprotein-protein interactions. EMBO J 22:6530–6539.

Vodicka P, Kumar R, Stetina R, et al. (2004) Genetic polymor-phisms in DNA repair genes and possible links with DNArepair rates, chromosomal aberrations and single-strandbreaks in DNA. Carcinogenesis 25:757–763.

1224 BEN SALAH ET AL.

Wang Y, Spitz MR, Zhu Y, et al. (2003) From genotype to phe-notype: correlating XRCC1 polymorphisms with mutagensensitivity. DNA Repair 8:901–908.

Wang Y, Yang H, Li H, et al. (2009) Association between X-rayrepair cross complementing group 1 codon 399 and 194polymorphisms and lung cancer risk: a meta-analysis. CancerLett 2:134–140.

Wu K, Su D, Lin K, et al. (2011) XRCC1 Arg399Gln genepolymorphism and breast cancer risk: a meta-analysisbased on case-control studies. Asian Pac J Cancer Prev 9:2237–2243.

Xue H, Ni P, Lin B, et al. (2011) X-ray repair cross-complementinggroup 1 (XRCC1) genetic polymorphisms and gastric cancerrisk: a huge review and meta-analysis. Am J Epidemiol 173:363–375.

Yin G, Morita M, Ohnaka K, et al. (2012) Genetic polymorphismsof XRCC1, alcohol consumption, and the risk of colorectalcancer in Japan. J Epidemiol 1:64–71.

Zhang H, Li W, franklin MJ, et al. (2011) Polymorphisms in DNArepair gene XRCC1and skin cancer risk: a meta-analysis. An-ticancer Res 11:3945–3952.

Zhang X, Morera S, Bates PA, et al. (1998) Structure of an XRCC1BRCT domain: a new protein–protein interaction module.EMBO J 17:6404–6411.

Zhou LQ, Ma Z, Shi XF, et al. (2011) Polymorphisms of DNArepair gene XRCC1 and risk of glioma: a case-control study inSouthern China. Asian Pac J Cancer Prev 10:2547–2550.

Address correspondence to:Ghada Ben Salah, Ph.D.

Laboratory of Human Molecular GeneticsFaculty of Medicine

University of SfaxAv. Magida Boulila

Sfax 3029Tunisia

E-mail: ghada.bsalah@gmail.com

Hassen Kamoun, Pr. Agr.Laboratory of Human Molecular Genetics

Faculty of MedicineUniversity of Sfax

Av. Magida BoulilaSfax 3029

Tunisia

E-mail: hassen.kamoun@fmsf.rnu.tn

DNA REPAIR GENE POLYMORPHISMS AT XRCC1 1225