Amino acid substitution variants of APE1 and XRCC1 genes ...

4 downloads 52 Views 84KB Size Report
XRCC3 is a member of the Rad51 DNA repair gene family;. Exposure to ionizing radiation (IR) has been linked to cancers it functions in the HRR pathway for ...
Carcinogenesis vol.22 no.6 pp.917–922, 2001

Amino acid substitution variants of APE1 and XRCC1 genes associated with ionizing radiation sensitivity

Jennifer J.Hu1,2,5, Tasha R.Smith1, Mark Steven Miller1, Harvey W.Mohrenweiser3, Andrea Golden4 and L.Douglas Case2 1Department

of Cancer Biology and 2Department of Public Health Sciences, Wake Forest University School of Medicine, Winston-Salem, NC 27157, 3Biology and Biotechnology Research Program, Lawrence Livermore National Laboratory, Livermore, CA 94550 and 4Lombardi Cancer Center, Georgetown University Medical Center, Washington, DC 20007, USA 5To

whom correspondence should be addressed Email: [email protected]

Although several variants of DNA repair genes have been identified, their functional significance has not been determined. Using samples collected from 135 cancerfree women, this study evaluated whether amino acid substitution variants of DNA repair genes contribute to ionizing radiation (IR) susceptibility as measured by prolonged cell cycle G2 delay. PCR–restriction fragment length polymorphism (RFLP) assays were used to determine four genotypes: X-ray repair cross complementing group 1 (XRCC1, exon 6, C/T, 194 Arg/Trp and exon 10, G/A, 399 Arg/Gln), XRCC group 3 (XRCC3, exon 7, C/T, 241 Thr/Met) and apurinic/apyrimidinic endonuclease 1 (APE1, exon 5, T/G, 148 Asp/Glu). Fluorescence-activated cell sorter (FACS) analysis was used to measure cell cycle delay. APE1 (exon 5) genotype was significantly associated with mitotic delay (P ⍧ 0.01), with the Glu/Glu genotype having prolonged delay compared with the other two genotypes. The mitotic delay index (mean ⍨ SD) in women with the APE1 codon 148 Asp/Asp, Asp/Glu and Glu/Glu genotypes was 30.95 ⫾ 10.15 (n ⍧ 49), 30.65 ⍨ 10.4 (n ⍧ 60) and 39.56 ⫾ 13.12 (n ⍧ 21), respectively. There was a significant interaction between family history (FH) and APE1 (exon 5) genotype (P ⍧ 0.007) as well as FH and XRCC1 (exon 10) genotype (P ⍧ 0.005) in mitotic delay. Lastly, prolonged cell cycle delay was significantly associated with number of variant alleles when APE1 Asp148Glu and XRCC1 Arg399Gln genotypes were evaluated in a four-level model (χ2 for linear trend ⍧ 10.9; P ⍧ 0.001). These results suggest that amino acid substitution variants of XRCC1 and APE1 may contribute to IR hypersensitivity. Introduction Exposure to ionizing radiation (IR) has been linked to cancers of the thyroid, breast and lung as well as leukemia (1). Ionizing radiation exposure damages cellular DNA in many ways, Abbreviations: APE1, apurinic/apyrimidinic endonuclease; AT, ataxia telangiectasia; BER, base excision repair; CI, confidence interval; FACS, fluorescence-activated cell sorter; FH, family history; HRR, homologous recombination repair; IR, ionizing radiation; Lig III, ligase III; NER, nucleotide excision repair; OR, odds ratio; PARP, poly(ADP-ribose)polymerase; Pol β, polymerase β; RFLP, restriction fragment length polymorphism; SNPs, singlenucleotide polymorphisms; XRCC, X-ray repair cross complementing. © Oxford University Press

requiring the concerted action of a number of DNA repair enzymes to restore genomic integrity. The base excision repair (BER) and homologous recombination repair (HRR) pathways are particularly important (2–5). Since oxidative base damage and strand breaks induced by IR are repaired mainly by these two pathways, defective/deficient repair activities may contribute to IR sensitivity and elevated cancer risk. Hypersensitivity to IR and defective DNA repair were observed in breast cancer patients and healthy women with a positive family history (FH) compared with healthy women without a FH of breast cancer (6–14). Higher levels of radiation-induced G2 phase delay (i.e. mitotic delay) were present in lymphocytes from ataxia telangiectasia (AT) patients, breast cancer patients and women with a FH of breast cancer (15–17). Genetic variability in DNA repair may contribute to IR sensitivity and cancer susceptibility. Several studies have identified many single-nucleotide polymorphisms (SNPs) in genes involved in nucleotide excision repair (NER), BER, double-strand break repair/HRR and cell-cycle checkpoint (18–22). These DNA repair genetic variants could be classified as cancer susceptibility genes, particularly if SNPs have functional significance. In this study, we evaluated the association between IR sensitivity as measured by cell cycle G2 delay and four amino acid substitution variants of three DNA repair genes: XRCC1 (exon 6, codon 194 Arg/Trp and exon 10, codon 399 Arg/Gln), XRCC3 (exon 7, codon 241 Thr/Met) and APE1 (exon 5, codon 148 Asp/Glu). XRCC1 plays an important role in BER and participates as a scaffolding intermediate by interacting with ligase III (Lig III), DNA polymerase β (pol β) and poly(ADP-ribose)polymerase (PARP) in the C-terminal, N-terminal and central regions of XRCC1, respectively (23–25). XRCC1 mutant cells have increased sensitivity to IR, UV, hydrogen peroxide and mitomycin C (5). The XRCC1 variant allele, Arg194Trp (exon 6), results in a nonconservative substitution in a hydrophobic region of XRCC1, and the SNP of Arg399Gln occurs within the BRCA1 Cterminal domain known to interact with PARP (19). The XRCC1 194Arg and 399Gln alleles were associated with increased risk for oral cavity and pharyngeal cancers (26). The XRCC1 399Gln allele was associated with lung cancer risk, as well as higher levels of DNA adduct and sister chromatid exchange (27–29). XRCC3 is a member of the Rad51 DNA repair gene family; it functions in the HRR pathway for repairing double-strand breaks, which plays important roles in maintaining genome stability (30). XRCC3 mutant cells show moderate hypersensitivity to IR, UV and monofunctional alkylating agents but extreme sensitivity to DNA cross-linking drugs, such as mitomycin C or cisplatin (30–33). APE1 (HAP1/Ref-1; EC 4.2.99.18) is the rate-limiting enzyme in the BER pathway (34). It cleaves 5⬘ of DNA abasic sugar residues generated from exogenous factors, such as IR and environmental carcinogens, as well as endogenous agents from normal cellular 917

J.J.Hu et al.

metabolism (35). Two studies suggested that APE1 allows pol β and Lig III to enter DNA repair sites and assembles pol β onto AP sites (36,37). Three variants (L104R, E126D and R237A) of the APE1 gene show reduced endonuclease activity (38). In contrast to rare, highly penetrant alleles, low-penetrance susceptibility alleles may contribute to a substantial proportion of cancer cases, since some are very common in the general population (39,40). It is clearly important to evaluate genetic variants of DNA repair in human cancer risk, but first, their functional significance must be elucidated. At the present time, one of the most rapidly growing research areas on DNA repair has focused on determining the association between sequence variations with heritable phenotypes and cancer susceptibility. In this study, we investigated the association between prolonged cell cycle delay in response to IR and four amino acid substitution variants of three DNA repair genes.

respectively. The use of NlaIII for XRCC3 (exon 7) created an internal control band size of 140 bp. The digested products for XRCC3 codon 241 Thr/Thr, Thr/Met and Met/Met genotypes had band sizes of 315/140, 315/210/140/105 and 210/140/105 bp, respectively. The BfaI restricted products of APE1 codon 148 Asp/Asp, Asp/Glu and Glu/Glu genotypes had band sizes of 164, 164/ 144/20 and 144/20 bp, respectively.

Materials and methods

Statistical analysis All the laboratory assays were conducted in a blinded fashion and then unblinded for statistical analysis. ANOVA was used to compare the mitotic delay index among different groups of subjects. A log transformation of the mitotic delay data was used to stabilize variances and increase the normality of the residuals. Analysis of covariance on the log transformed data was used to assess the joint effect of FH and genotypes and their two-way interactions on mitotic delay. Logistic regression was used to evaluate the association between cell cycle G2 delay and genotypes. Odds ratio (OR) and 95% confidence interval (CI) were calculated from the logistic model estimates. All the statistical analyses were carried out using the Statistical Analysis System (SAS Institute, Cary, NC).

Study subjects Cancer-free women were recruited from the Comprehensive Breast Center and the Cancer Assessment and Risk Evaluation program at Georgetown University Medical Center from August 1995 to November 1996 as study controls for a breast cancer case–control study (17). Subjects received a detailed description of the study protocol and signed informed consent as approved by the Institutional Review Board of the Georgetown University Medical Center. Blood samples (20 ml) were collected from all consenting study subjects. Peripheral lymphocytes were isolated from whole blood within 2 h using neutrophil isolation medium (Cardinal Associates, Santa Fe, NM). The lymphocyte fraction was cryopreserved (5⫻106 cells/vial) in RPMI medium with 50% fetal bovine serum and 10% DMSO and stored at –140°C until assay. Questionnaires, medical records and pathology reports were reviewed to confirm cancer-free status. In the parent study, a 20 page questionnaire was used by the Biomarker Core of Lombardi Cancer Center to collect information on age, race, FH, menstrual history, parity and medication. PCR–RFLP genotyping assays Genomic DNA was isolated from peripheral lymphocytes using an ASAP Genomic DNA Isolation kit (Boehringer Mannheim, Indianapolis, IN). PCR– RFLP assays were used to determine DNA repair genotypes. Four primer pairs, synthesized by the DNA Synthesis Core Laboratory at Wake Forest University School of Medicine were used to individually amplify each genetic region. The primer pairs used were XRCC1 Arg194Trp, forward, 5⬘-GCCCCGTCCCAGGTA-3⬘ and reverse, 5⬘-AGCCCCAAGACCC TTTCACT-3⬘; XRCC1 Arg399Gln, forward, 5⬘-TCTCCCTTGGTCTCCAACCT3⬘ and reverse, 5⬘-AGTAGTCTGCTGGCTCTGG-3⬘; XRCC3 Thr241Met, forward, 5⬘-GGTCGAGTGACAGTCCAAAC-3⬘ and reverse, 5⬘-TGCAACGGCTGAGGGTCTT-3⬘ and APE1 Asp148Glu, forward, 5⬘-CTGTTTCATTTCTATAGGCTA-3⬘ and reverse, 5⬘-AGGAACTTGCGAAAGGCTTC-3⬘. About 50–200 ng genomic DNA in a total volume of 20 µl was amplified using a GeneAmp 9700 (Perkin Elmer, Foster City, CA). The reaction mixture consisted of GeneAmp PCR Gold buffer (150 mM Tris–HCl, pH 8.0, 500 mM KCl), 2.5 mM MgCl2, 0.2 mM each dNTPs, 0.2 µM each primer, 1 U AmpliTaq Gold polymerase (Perkin Elmer) and nuclease-free water (Promega, Madison, WI). Negative controls lacking DNA templates were set up with all PCR reactions. Each genotype assay was repeated at least two to three times to confirm study results. To avoid contamination, barrier tips were used, and all reaction tubes and tips were UV irradiated for at least 15 min prior to reaction mixture preparation. PCR conditions were 95°C for 2 min, followed by 40 cycles of 94°C for 15 s, 57°C for 45 s, 72°C for 45 s and a final elongation step at 72°C for 5 min. PCR products were digested with specific restriction enzymes obtained from New England Biolabs (Beverly, MA) that recognized and cut either the wild-type or variant sequence site. PCR products were digested at 37°C for 3 h for XRCC1 (exon 6) and at least 3 h to overnight for other genotypes. The digested products were resolved on 1.5% agarose gels and stained with 0.5 µg/ml ethidium bromide (Gibco BRL, Gaithersburg, MD). The PvuII restricted products of XRCC1 codon 194 Arg/Arg, Arg/Trp and Trp/Trp genotypes had band sizes of 490, 490/294/196 and 294/196 bp, respectively. The MspI restricted products of XRCC1 codon 399 Arg/Arg, Arg/Gln and Gln/Gln genotypes had band sizes of 269/133, 402/269/133 and 402 bp,

918

Cell cycle G2 delay assay The quick-thawed lymphocytes with ⬎95% viability were incubated in RPMI medium with 20 µg/ml phytohemagglutinin for 72 h. Using a 137Cs source, the cells were exposed to 3 Gy γ-irradiation, which gave the best discrimination between AT patients and normal controls (15). Cells with or without (as control) exposure to radiation were harvested 24 h later, and the percentages of cells in the G0/G1, G2/M and S phase of the cell cycle were determined by using a dual laser fluorescence activated cell sorter (FACStar Plus; Becton Dickinson, Mountain View, CA) as described previously (11,12). DNA staining was performed by adding 0.05 ml propidium iodide in PBS (250 µg/ml propidium iodide, 5 mg/ml RNase and 10 mg/ml Triton X-100) to the cell suspension. The mitotic delay index was calculated as (percentage of cells in G2/M with IR exposure – percentage of control cells in G2/M) / (percentage of control cells in S phase)⫻100%. The median of the total study population was used as the cut-off for prolonged cell cycle G2 delay (IR hypersensitive).

Results In the parent study, a total of 643 subjects were recruited in collaboration with the Biomarker Core at the Lombardi Cancer Center, Georgetown University Medical Center. All the study subjects gave blood. Thawed lymphocytes with ⬎ 80% viability were used for DNA repair functional assays. Data on mitotic delay were available for 464 subjects (including controls, benign breast diseases, atypical hyperplasia and invasive breast cancer). In this paper, we only present data obtained from disease-free controls. The cryopreserved lymphocyte samples were stored about 8 months (range 3–12 months) before cell cycle G2 delay assay. During this short storage period, the viability and cellular response to mitogen are not altered (L.Grossman, personal communication). During the development of the cell cycle G2 delay assay, repeated measurements on six subjects were carried out. The mean coefficient of variation was 21% (range 4–57%). In this study, we evaluated genotype and cell cycle G2 delay data from 135 disease-free women. Their ages ranged from 26 to 82 with a median of 53 years. Sixty-four percent of the women were 50 years or older. A 20 page questionnaire was used by the Biomarker Core to collect information on race, FH, menstrual history, parity and medication. Data on FH were available for 83 women; 38 (46%) of these had at least one first degree relative with breast cancer. Twenty-one (16%) only had mothers with breast cancer, nine (7%) only had sisters and eight (6%) had both. Twenty-six of the women in our sample (31%) had one first degree relative with breast cancer, 11 (13%) had two and one (1%) had three. Many study subjects did not return the questionnaire, probably because the questionnaire was extremely time consuming. Unfortunately, the Institutional Review Board guidelines did not allow us to contact study subjects for missing information. To address the

APE1 and XRCC1 genetic variants and IR sensitivity

Table I. Distribution of delay index by age and family history Characteristics Age ⬍40 40–49 50–59 60–69 艌70 Mother with breast cancer No Yes Missing Sister with breast cancer No Yes Missing No. affected relatives 0 1 2 Missing aP-value,

n

Mitotic delay index (mean ⫾ SD)

14 35 51 21 14

30.03 31.04 32.39 33.73 34.40

54 29 52

32.44 ⫾ 10.33 31.79 ⫾ 14.87 32.20 ⫾ 9.97

66 17 52

32.22 ⫾ 12.20 32.20 ⫾ 11.68 32.20 ⫾ 9.97

45 26 12 52

32.87 30.14 34.27 32.20

⫾ ⫾ ⫾ ⫾ ⫾

⫾ ⫾ ⫾ ⫾

P-valuea 0.57

8.57 10.57 10.67 13.09 14.74 0.45

0.81

0.35 11.12 13.09 13.33 9.97

ANOVA based on log-transformed data.

potential bias related to missing FH data, we first ran analyses to compare G2 delay and genotype data between the two groups of subjects, women with and without questionnaire information. There was no significant difference in the distribution of DNA repair genotypes and G2 delay marker between these two groups. Additionally, age was similar for those with and without FH data. Therefore, it is unlikely that missing questionnaire data will introduce serious systematic error or bias in this study. However, missing questionnaire data lowered the sample size and statistical power for those analyses including FH information. Eighty-five percent of the subjects carried the XRCC1 (exon 6) wild-type/wild-type (WW) genotype; the remaining 15% were wild-type/variant (WV). The distribution of XRCC1 (exon 10) is 44% WW, 46% WV and 10% VV. These percentages are 38% WW, 46% WV and 16% VV for APE1 (exon 5) genotype and 37% WW, 47% WV and 16% VV for XRCC3 (exon 7) genotype. Genotype distributions at each locus were consistent with Hardy–Weinberg equilibria. Analysis of variance on the log transformed values of delay index was used to assess the univariate effect of subject characteristics and amino acid substitution variants in DNA repair genes on cellular response to IR. Results are summarized in Tables I and II. In Table I, there was a slight but nonsignificant increase of mitotic delay with age (P ⫽ 0.57). The correlation between mitotic delay and age considered continuously was also non-significant (P ⫽ 0.39). The mean ⫾ SD of the mitotic delay index in women with 0, 1, and 2⫹ affected first-degree relatives was 32.87 ⫾ 11.12 (n ⫽ 45), 30.14 ⫾ 13.09 (n ⫽ 26) and 34.27 ⫾ 13.33 (n ⫽ 12), respectively. The mitotic delay did not differ significantly among these three groups of women (P ⫽ 0.35). Distributions of the mitotic delay by DNA repair genotypes are shown in Table II. Only APE1 (exon 5) genotype was significantly associated with mitotic delay index (P ⫽ 0.01), with the VV (Glu/Glu) genotype having prolonged delay compared with the other two genotypes. The mitotic delay index in women with the APE1 codon 148 Asp/Asp, Asp/Glu and Glu/Glu genotypes was 30.95 ⫾ 10.15 (n ⫽ 49),

30.65 ⫾ 10.4 (n ⫽ 60) and 39.56 ⫾ 13.12 (n ⫽ 21), respectively. The results in Table II also suggest that the XRCC1 399Gln allele may alter IR sensitivity to some extent; the mitotic delay index in women with the Arg/Arg, Arg/Gln and Gln/Gln genotypes was 30.17 ⫾ 10.22 (n ⫽ 59), 33.61 ⫾ 11.97 (n ⫽ 61) and 34.28 ⫾ 11.82 (n ⫽ 14), respectively. However, the difference was not statistically significant (P ⫽ 0.26). Multivariable regression models were then used to assess the joint effect of the variables shown in Tables I and II and their two-way interactions on mitotic delay index. Due to the large number of individuals with missing FH information, models involving FH which included only the 83 subjects with these data were initially assessed. Interestingly, it appears there is a significant interaction between FH and APE1 (exon 5) genotype (P ⫽ 0.007) and between FH and XRCC1 (exon 10) genotype (P ⫽ 0.005) in mitotic delay. Summary statistics for delay index broken down by FH for each genotype are shown in Table III. In the absence of FH, neither genotype is significantly associated with mitotic delay. For those with one involved relative, both APE1 (exon 5) and XRCC1 (exon 10) are significantly associated with mitotic index, with both WV and VV differing from WW. Data are too sparse to make conclusions regarding differences in genotypes for those with 2⫹ involved relatives. None of the gene–gene interactions were statistically significant, even the one between APE1 and XRCC1, although both play critical roles in BER and singlestrand break repair. We also analyzed these data to see which variables (or combination of variables) were predictive of prolonged mitotic delay (referred to as IR sensitivity), which was arbitrarily defined as a delay index greater than the median (which was 32 for all the subjects with FH data). Due to small numbers, the WV and VV categories for each genotype were combined as were those with one or more involved family members. Logistic regression was used to assess which factors were jointly predictive of IR sensitivity. As in the previous analysis, there was a significant interaction between XRCC1 (exon 10) and FH (P ⫽ 0.04). In the absence of a FH of breast cancer, the WV/VV category of XRCC1 (exon 10) was associated with only a 12% increase in the odds of IR sensitivity. However, in the presence of a FH, the WV/VV category of XRCC1 (exon 10) was associated with a 96% increase in the odds of IR sensitivity. APE1 was borderline significant in this analysis (P ⫽ 0.06), with the odds of IR sensitivity increasing by 183% for the WV/VV category compared with the WW category. Data in Table IV show the distribution of subjects with prolonged G2 delay for combined XRCC1 and APE1 genotypes. Even though this interaction was not significant in the logistic regression model just described, the results in Table IV suggest that prolonged cell cycle delay was significantly associated with number of variant alleles when APE1 Asp148Glu and XRCC1 Arg399Gln genotypes were evaluated in a four-level model (χ2 for linear trend ⫽ 10.9; P ⫽ 0.001). Odds ratio (OR) ⫽ 1.0 for those without variant allele (referent); OR ⫽ 0.81 (95% CI 0.27–2.45) for those with one variant allele; OR ⫽ 2.19 (95% CI 0.68–7.02) and OR ⫽ 6.67 (95% CI ⫽ 1.38–32.26) for those with at least three variant alleles. Future larger studies are warranted to further test this potential combined effect of XRCC1 and APE1 genotypes. 919

J.J.Hu et al.

Table II. Distribution of delay index by DNA repair genotypes Genes

Variant (base position) (amino acid position)

XRCC1 (exon 6) XRCC1 (exon 10) XRCC3 (exon 7) APE1 (exon 5)

C/T (26304) Arg/Trp (194) G/A (28152) Arg/Gln (399) C/T (18067) Thr/Met (241) T/G (2197) Asp/Glu (148)

Mitotic delay indexa WWb

WVb

% of WW

VVb

% of WW

32.20 ⫾ 11.02 (n ⫽ 115) 30.17 ⫾ 10.22 (n ⫽ 59) 32.68 ⫾ 10.67 (n ⫽ 50) 30.95 ⫾ 10.15 (n ⫽ 49)

32.25 ⫾ 12.72 (n ⫽ 20) 33.61 ⫾ 11.97 (n ⫽ 61) 31.69 ⫾ 11.08 (n ⫽ 63) 30.65 ⫾ 10.4 (n ⫽ 60)

100

N/A

N/A

0.92

111

34.28 ⫾ 11.82 (n ⫽ 14) 32.63 ⫾ 13.27 (n ⫽ 22) 39.56 ⫾ 13.12 (n ⫽ 21)

114

0.26

100

0.84

128

0.01

aData are presented as mean ⫾ SD (n). bWW, wild-type/wild-type; WV, wild-type/variant; cP-value,

P-valuec

97 99

VV, variant/variant.

ANOVA based on log-transformed data.

Table III. Distribution of delay index by XRCC1 and APE1 genotypes and family history Genes

No. affected relatives

XRCC1 (exon 10) APE1 (exon 5)

0 1 2 0 1 2

Mitotic delay indexa n

WWb

18 9 6 12 9 5

31.46 25.51 33.81 33.19 23.86 32.83

⫾ ⫾ ⫾ ⫾ ⫾ ⫾

7.49 11.03 19.68 11.42 8.75 3.39

n

WVb

22 14 6 22 11 6

34.32 30.74 34.73 31.49 30.69 29.23

⫾ ⫾ ⫾ ⫾ ⫾ ⫾

13.38 13.69* 1.71 10.69 13.46* 8.27

n

VVb

4 3 0 8 5 1

29.86 41.25 NA 37.97 36.44 71.73

⫾ 13.91 ⫾ 12.43* ⫾ 10.88 ⫾ 15.33*

aData are presented as mean ⫾ SD. bWW, wild-type/wild-type; WV, wild-type/variant;

VV, variant/variant. *P ⬍ 0.05 compared with WW genotype group; ANOVA contrast based on log-transformed data.

Table IV. Association between XRCC1/APE1 genotypes and prolonged cell cycle G2 delay Group

No. of variant Total No. subjects allelesa Prolonged Normal delay (%) delay (%)

OR (95% CI)b

1 2 3 4

0 1 2 3⫹

Referent 0.81 (0.27–2.45) 2.19 (0.68–7.02) 6.67 (1.38–32.26)

17 58 38 17

7 (41) 21 (36) 23 (61) 14 (82)

10 (59) 37 (64) 15 (39) 3 (18)

aTotal

number of variant alleles in XRCC1 (exon 10) and APE1 (exon 5) genotypes. bχ2 ⫽ 10.9, P ⫽ 0.001, test for linear trend.

Discussion This study investigated the relationship between four amino acid substitution variants in DNA repair genes and IR-induced cell cycle G2 delay in mitogen-stimulated peripheral lymphocytes. We demonstrate that the APE1 148 Glu allele and/or XRCC1 399Gln allele may interact with FH and contribute to a prolonged cell cycle G2 delay, which is associated with IR hypersensitivity and susceptibility to breast cancer (15–17). The study results provide preliminary evidence that amino acid substitution variants in DNA repair genes may contribute to IR sensitivity and have potential application as markers for cancer susceptibility. Since some of the amino acid substitution variants in DNA repair genes are very common, our findings have biological and public health significance. Exposure to IR has been implicated in the etiology of cancer 920

(1), and lesions induced by IR are repaired by BER and HRR (2–4,30). Hypersensitivity to IR could result from somatic mutations of repair genes, polymorphisms of BER genes (e.g. APE1 and XRCC1), and a reduction in protein expression secondary to other cellular dysfunction. Our data suggest that amino acid substitution variants in BER genes are associated with mitotic delay in response to IR. Since inherited hypersensitivity to IR and deficient repair of IR-induced damage may serve as markers for low-penetrant predisposition genes in breast cancer, it is reasonable to conclude that amino acid substitution variants in BER genes may contribute to hereditary IR sensitivity and cancer susceptibility. However, detailed structure–function relationships between amino acid substitution variants and enzyme activities remain to be elucidated. The multifunctional mammalian APE1 is responsible for the repair of AP sites in DNA. This enzyme also functions as a redox factor facilitating the DNA-binding capability of AP-1, and numerous other transcription factors (41). Abasic sites represent ubiquitous DNA lesions that arise spontaneously or are induced by DNA-damaging agents. They block DNA replication and are considered cytotoxic and mutagenic. The major mammalian APE1 plays a central role in BER in two distinct ways (42). First, APE1 initiates repair of AP sites in DNA produced either spontaneously or after removal of uracil and alkylated bases in DNA by monofunctional DNA glycosylases. Second, APE1 can act as a 3⬘-phosphoesterase to initiate repair of strand breaks, either directly induced by reactive oxygen species or indirectly through the AP lyase reaction of DNA damage-specific glycosylases. Furthermore, although APE1 plays an important role in the repair of DNA strand breaks, other genes, such as pol β, are rate-limiting

APE1 and XRCC1 genetic variants and IR sensitivity

factors for uracil repair (42); hence, it will be necessary to evaluate other BER genes as well. In the current study, we report an association between SNPs of two DNA repair genes and IR sensitivity. Since the SNPs presented in this paper are amino acid substitution variants, structural changes of APE1 and XRCC1 proteins may have functional significance. First, the APE1 148 Glu variant allele has a small but non-significant effect on endonuclease activity (94% of wild-type) and DNA binding activity (Kd, 20.3 ⫾ 3.4 versus 25.8 ⫾ 12.2 nM in wild-type) (38). Although the small differences in binding and lower endonuclease activity may be within experimental error, it is possible that the lower Kd of the variant implies a higher affinity between APE1 protein and damaged DNA after catalysis and as such turns over less effectively. In addition, isolated protein was studied, effects of APE1 interaction with other BER components not studied (38). These other proteins may also affect APE1 turnover. Therefore, future studies are needed to investigate whether APE1 148 Glu allele may alter the ability of the APE1 protein to communicate with other BER proteins and thereby influence BER efficiency. Second, the genetic variant of XRCC1 Arg399Gln occurs within the BRCA1 C-terminal domain, which interacts with PARP (19). Considering the important roles of BRCA1 and PARP in DNA repair, the XRCC1 variant may have functional significance. Future detailed analyses of protein structures and functions are needed to further evaluate the mechanisms involved. Using a cytogenetic assay, cellular radiosensitivity in family members of radiosensitive (breast cancer patients) and nonsensitive individuals was studied (12). The results from their segregation analysis suggested that radiosensitivity may be heritable, with a single major gene accounting for 82% of the variance among family members. The addition of a second, rarer gene to the model resulted in a better fit of the data. These results further suggest that cancer-predisposing genes are probably common, low-penetrance alleles, such as amino acid substitution variants in DNA repair genes. The results from our study support their proposed model that multiple, low-penetrance, common genes (e.g. APE1 and XRCC1) contribute to IR sensitivity. To the best of our knowledge, the relationship between APE1 Asp148Glu genotype and cancer susceptibility has not been investigated previously. In an ongoing study, we are evaluating whether this genotype is associated with breast cancer (43). The genetic variant of XRCC1 Arg399Gln has been evaluated in cancer of the head and neck, lung and breast (26,27,43). In this study, we demonstrate that the XRCC1 Arg399Gln genotype may influence cellular response to IR, particularly in women with positive FH. Our data are consistent with the previous findings that the XRCC1 399Gln allele was associated with higher levels of DNA adducts and sister chromatid exchange (28,29). Ionizing radiation induces many types of damage to DNA, requiring multiple repair pathways (e.g. BER and HRR) to restore genomic integrity. Most of the repair pathways are extremely complex, and many genes are involved in different repair pathways. With limited questionnaire data and sample size, this manuscript focuses on the genotype–phenotype relationship. It is conceivable that many other genetic and non-genetic factors (i.e. other SNPs, smoking, alcohol consumption, menopausal status, medication, diet and environmental exposures) may also influence DNA damage/repair, but

the study of those relationships is beyond the scope of this paper. At the present time, we are conducting a larger study to test these other factors. The results from this study suggest that amino acid substitution variants of XRCC1 and APE1 may alter the functions of these proteins, resulting in hypersensitivity to IR. Future larger studies are warranted to further test the functional significance of these genotypes. Acknowledgements We gratefully acknowledge the contributions of Drs Bruce J.Trock, Caryn Lerman, Marc E.Lippman, Susan Honig, Claudine Isaacs, Owen C.Blair, Satyajit Ray and Mr Linus Truong from Georgetown University Medical Center. We also thank Drs Lawrence Grossman and Ian Hickson for helpful discussions. The study was supported by NIH grant CA73629 and ACS grant no. RPG-97-115-01 (to J.J.H.). The cell cycle analysis was conducted by the Lombardi Cancer Center Flow Cytometry/Cell Sorting Shared Resource. Work performed at Lawrence Livermore National Laboratory was under the auspices of US DOE contract No.W-7405-ENG-48 to the University of California.

References 1. Ron,E. (1998) Ionizing radiation and cancer risk: evidence from epidemiology. Radiat. Res., 150, S30–S41. 2. Demple,B. and Harrison,L. (1994) Repair of oxidative damage to DNA: enzymology and biology. Annu. Rev. Biochem., 63, 915–948. 3. Wallace,S.S. (1994) DNA damages processed by base excision repair: biological consequences. Int. J. Radiat. Biol., 66, 579–589. 4. Lindahl,T. (2000) Suppression of spontaneous mutagenesis in human cells by DNA base excision–repair. Mutat. Res., 462, 129–135. 5. Thompson,L.H. and West,M.G. (2000) XRCC1 keeps DNA from getting stranded. Mutat. Res., 459, 1–18. 6. Helzlsouer,K.J., Harris,E.L., Parshad,R., Perry,H.R., Price,F.M. and Sanford,K.K. (1996) DNA repair proficiency: potential susceptibility factor for breast cancer. J. Natl Cancer Inst., 88, 754–755. 7. Parshad,R., Price,F.M., Bohr,V.A., Cowans,K.H., Zujewski,J.A. and Sanford,K.K. (1996) Deficient DNA repair capacity, a predisposing factor in breast cancer. Br. J. Cancer, 74, 1–5. 8. Hu,J.J., Dubin,N., Berwick,M., Miesner,J., Roses,D.F., Harris,M.N. and Roush,G.C. (1997) Poly(ADP-ribose) polymerase in human breast cancer: A case–control analysis. Pharmacogenetics, 7, 309–316. 9. Chakraborty,R., Little,M.P. and Sankaranarayanan,K. (1997) Cancer predisposition, radiosensitivity and the risk of radiation-induced cancers: III. Effects of incomplete penetrance and dose-dependent radiosensitivity on cancer risks in populations. Radiat. Res., 147, 309–320. 10. Patel,R.K., Trivedi,A.H., Arora,D.C., Bhatavdekar,J.M. and Patel,D.D. (1997) DNA repair proficiency in breast cancer patients and their firstdegree relatives. Int. J. Cancer, 73, 20–24. 11. Grossman,L., Matanoski,G., Farmer,E., Hedayati,M., Ray,S., Trock,B., Hanfelt,J., Roush,G., Berwick,M. and Hu,J.J. (1999) DNA repair as a susceptibility factor in chronic diseases in human populations. In Dizdaroglu,M. and Karakaya,A.E. (eds) Advances in DNA Damage and Repair. Kluwer Academic/Plenum Publishers, New York, pp. 149–167. 12. Roberts,S.A., Spreadborough,A.R., Bulman,B., Barber,J.B., Evans,D.G. and Scott,D. (1999) Heritability of cellular radiosensitivity: a marker of low-penetrance predisposition genes in breast cancer? Am. J. Hum. Genet., 65, 784–794. 13. Scott,D., Barber,J.B., Spreadborough,A.R., Burrill,W. and Roberts,S.A. (1999) Increased chromosomal radiosensitivity in breast cancer patients: a comparison of two assays. Int. J. Radiat. Biol., 75, 1–10. 14. Rajeswari,N., Ahuja,Y.R., Malini,U., Chandrashekar,S., Balakrishna,N., Rao,K.V. and Khar,A. (2000) Risk assessment in first degree female relatives of breast cancer patients using the alkaline Comet assay. Carcinogenesis, 21, 557–561. 15. Lavin,M.F., Le Poidevin,P. and Bates,P. (1992) Enhanced levels of radiation-induced G2 phase delay in ataxia telangiectasia heterozygotes. Cancer Genet. Cytogenet., 60, 183–187. 16. Lavin,M.F., Bennett,I., Ramsay,J., Gardiner,R.A., Seymour,G.J., Farrell,A. and Walsh,M. (1994) Identification of a potentially radiosensitive subgroup among patients with breast cancer. J. Natl Cancer Inst., 86, 1627–1634. 17. Hu,J.J., Legault,C., deLeon,A., Trock,B.J., Lerman,C. and Craven,T. (1999) Hypersensitivity to ionizing radiation and human breast cancer risk. Proc. Am. Assoc. Cancer Res., 40, 3757.

921

J.J.Hu et al. 18. Broughton,B.C., Steingrimsdottir,H. and Lehmann,A.R. (1996) Five polymorphisms in the coding sequence of the xeroderma pigmentosum group D gene. Mutat. Res., 362, 209–211. 19. Shen,M.R., Jones,I.M. and Mohrenweiser,H. (1998) Nonconservative amino acid substitution variants exist at polymorphic frequency in DNA repair genes in healthy humans. Cancer Res., 58, 604–608. 20. Bell,D.W., Wahrer,D.C., Kang,D.H., MacMahon,M.S., FitzGerald,M.G., Ishioka,C., Isselbacher,K.J., Krainer,M. and Haber,D.A. (1999) Common nonsense mutations in RAD52. Cancer Res., 59, 3883–3888. 21. Fan,F., Liu,C., Tavare,S. and Arnheim,N. (1999) Polymorphisms in the human DNA repair gene XPF. Mutat. Res., 406, 115–120. 22. Ford,B.N., Ruttan,C.C., Kyle,V.L., Brackley,M.E. and Glickman,B.W. (2000) Identification of single nucleotide polymorphisms in human DNA repair genes. Carcinogenesis, 21, 1977–1981. 23. Nash,R.A., Caldecott,K.W., Barnes,D.E. and Lindahl,T. (1997) XRCC1 protein interacts with one of two distinct forms of DNA ligase III. Biochemistry, 36, 5207–5211. 24. Kubota,Y., Nash,R., Klungland,A., Schar,P., Barnes,D. and Lindahl,T. (1996) Reconstitution of DNA base-excision repair with purified human proteins: interaction between DNA polymerase β and the XRCC1 protein. EMBO J., 15, 6662–6670. 25. Masson,M., Niedergang,C., Schreiber,V., Muller,S., Menissier-de Murcia,J. and de Murcia,G. (1998) XRCC1 is specifically associated with poly (ADP-ribose) polymerase and negatively regulates its activity following DNA damage. Mol. Cell. Biol., 18, 3563–3571. 26. Sturgis,E.M., Castillo,E.J., Li,L., Zheng,R., Eicher,A., Clayman,G.L., Strom,S.S., Spitz,M.R. and Wei,Q. (1999) Polymorphisms of DNA repair gene XRCC1 in squamous cell carcinoma of the head and neck. Carcinogenesis, 20, 2125–2129. 27. Divine,K.K., Gilliland,F.D., Crowell,R.E., Stidley,C.A., Bocklage,T.J., Cook,D.L. and Belinsky,S.A. (2001) The XRCC1 399 glutamine allele is a risk factor for adenocarcinoma of the lung. Mutat. Res., 461, 271–278. 28. Lunn,R., Langlois,R., Hsieh,L., Thompson,C. and Bell,D. (1999) XRCC1 polymorphisms: effects on aflatoxin B1–DNA adducts and glycophorin A variant frequency. Cancer Res., 59, 2557–2561. 29. Duell,E.J., Wiencke,J.K., Cheng,T.J., Varkonyi,A., Zuo,Z.F., Ashok,T.D.S., Mark,E.J., Wain,J.C., Christiani,D.C. and Kelsey,K.T. (2000) Polymorphisms in the DNA repair genes XRCC1 and ERCC2 and biomarkers of DNA damage in human blood mononuclear cells. Carcinogenesis, 21, 965–971. 30. Brenneman,M.A., Weiss,A.E., Nickoloff,J.A. and Chen,D.J. (2000) XRCC3 is required for efficient repair of chromosome breaks by homologous recombination. Mutat. Res., 459, 89–97.

922

31. Liu,N., Lamerdin,J.E., Tebbs,R.S. et al. (1998) XRCC2 and XRCC3, new human Rad51-family members, promote chromosome stability and protect against DNA cross-links and other damages. Mol. Cell, 1, 783–793. 32. Fuller,L.F. and Painter,R.B. (1998) A Chinese hamster ovary cell line hypersensitive to ionizing radiation and deficient in repair replication. Mutat. Res., 193, 109–121. 33. Cui,X., Brenneman,M., Meyne,J., Oshimura,M., Goodwin,E.H. and Chen,D.J. (1999) The XRCC2 and XRCC3 repair genes are required for chromosome stability in mammalian cells. Mutat. Res., 434, 75–88. 34. Ramana,C.V., Boldogh,I., Izumi,T. and Mitra,S. (1998) Activation of apurinic/apyrimidinic endonuclease in human cells by reactive oxygen species and its correlation with their adaptive response to genotoxicity of free radicals. Proc. Natl Acad. Sci. USA, 95, 5061–5066. 35. Barzilay,G. and Hickson,I. (1995) Structure and function of apurinic/ apyrimidinic endonucleases. BioEssays, 17, 713–719. 36. Bennett,R., Wilson III,D., Wong,D. and Demple,B. (1997) Interaction of human apurinic endonuclease and DNA polymerase β in the base excision repair pathway. Proc. Natl Acad. Sci. USA, 94, 7166–7169. 37. Kakolyris,S., Kaklamanis,L., Giatromanolaki,A. et al. (1998) Expression and subcellular localization of human AP endonuclease 1 (HAP1/Ref-1) protein: a basis for its role in human disease. Histopathology, 330, 561–569. 38. Hadi,M.Z., Coleman,M.A., Fidelis,K., Mohrenweiser,H.W. and Wilson III,D.M. (2000) Functional characterization of Ape1 variants identified in the human population. Nucleic Acids Res., 28, 3871–3879. 39. Teare,M.D., Wallace,S.A., Harris,M., Howell,A. and Birch,J.M. (1994) Cancer experience in the relatives of an unselected series of breast cancer patients. Br. J. Cancer, 70, 102–111. 40. Ford,D., Easton,D.F., Stratton,M. et al. (1998) Genetic heterogeneity and penetrance analysis of the BRCA1 and BRCA2 genes in breast cancer families. The Breast Cancer Linkage Consortium. Am. J. Hum. Genet., 62, 676–689. 41. Xu,Y., Moore,D.H., Broshears,J., Liu,L., Wilson,T.M. and Kelley,M.R. (1997) The apurinic/apyrimidinic endonuclease (APE/ref-1) DNA repair enzyme is elevated in premalignant and malignant cervical cancer. Anticancer Res., 17, 3713–3719. 42. Izumi,T., Hazra,T.K., Boldogh,I., Tomkinson,A.E., Park,M.S., Ikeda,S. and Mitra,S. (2000) Requirement for human AP endonuclease 1 for repair of 3⬘-blocking damage at DNA single-strand breaks induced by reactive oxygen species. Carcinogenesis, 21, 1329–1334. 43. Smith,T.R., Miller,M.S., Lohman,K., Mohrenweiser,H.W. and Hu,J.J. (2000) Polymorphisms of DNA repair genes in breast cancer. Proc. Am. Assoc. Cancer Res., 41, 817. Received August 25, 2000; revised March 2, 2001; accepted March 7, 2001