TP53 and RPA3 Gene Variations Were Associated with Risk of Glioma in a Chinese Han Population

Abstract

Recent advances in human genetic studies have opened new avenues for the identification of susceptibility genes for many complex genetic disorders, especially in the field of rare cancers such as glioma. Glioma is one of the least understood human tumors and the etiology for glioma is barely known. Hundreds of single-nucleotide polymorphisms (SNPs) are found to be related to the risk of glioma in previous studies. This study is committed to investigate the role of heredity in this disorder. To examine and validate how common variants contribute to glioma susceptibility in the Han Chinese population, we evaluated 12 tagging SNPs in a case–control study in the Chinese Han population from Xi'an city of China (301 cases and 302 controls). Overall, two protective alleles and one risk allele for glioma were found by genetic model analyses. In dominant model, the allele “T” of rs6947203 in the RPA3 gene acts as a protective allele [odds ratio (OR), 0.59; 95% confidence interval (CI), 0.22–0.90; p=0.014]. In recessive model, the allele “C” of rs1042522 in the TP53 gene acts as a risk allele (OR, 1.65; 95% CI, 1.05–2.59; p=0.0314). In additive model, the allele “G” of rs4140805 in the RPA3 gene (OR, 0.73; 95% CI, 0.53–0.99; p=0.0437) and the allele “T” of rs6947203 in the RPA3 gene (OR, 0.62; 95% CI, 0.42–0.92; p=0.0177) both act as protective alleles. We also observed a haplotype of “CC” in the TP53 gene with an increased risk of 34% of developing glioma (p=0.0306). Our results, combined with previous studies, ascertain the potential role of the TP53 gene to glioma onset.

Introduction

Glioma is the most common brain tumor in humans and has a very poor prognosis with only 3% of glioblastoma patients surviving 5 years after diagnosis.¹ Over the past years, several relevant biomarkers with a potential clinical interest have been identified in gliomas using various techniques, such as karyotype analysis, microsatellite analysis, fluorescent in situ hybridization, and array-based chromosome comparative genomic hybridization technology.² Current evidence suggests that the heredity risk is due to the coinheritance of multiple low-risk variants.³ Genomic alterations described in gliomas, such as EGFR, CDKN2A, TP53, RB1, and PTEN, have been found in other common epithelial tumors.^4,5 Several genes affecting cell cycle and DNA repair functions have been proposed to play a role in glioma pathogenesis and progression.^6,7 Recent studies have emphasized on identifying the relationship between glioma susceptibility and single-nucleotide polymorphism (SNP).⁸

In this study, we investigate and validate the potential associations between glioma risk and 12 tagging SNPs (tSNPs) that were previously reported to be associated with glioma susceptibility. Our data shed new light on the association between common SNPs and glioma susceptibility in the Chinese population.

Materials and Methods

Study participants

We recruited a total of recently diagnosed and histologically confirmed 350 glioma patients from September 2009 to November 2011 into an ongoing molecular epidemiological study at the Department of Neurosurgery of the Tangdu Hospital affiliated with the Fourth Military Medical University in Xi'an city, China. All glioma cases were previously healthy. Age, gender, or disease stage restrictions were ignored for case recruitment. In our study, we excluded 49 cases that had incomplete clinical information. Finally, 301 glioma cases were successfully genotyped.

A random sample of unrelated healthy individuals from July 2011 to October 2011 were recruited from the medical center at Tangdu Hospital for genetic association research of human complex diseases, such as liver, renal cancer, and glioma. The recruitment and exclusion standard were used. Generally, all the subjects were healthy and had no chronic diseases involving vital organs (lung, heart, kidney, brain, and liver), and thus we reduced the known environmental and therapeutic factors that influence the variation of human complex diseases. We recruited a total of 302 healthy controls in this study. All the participants were restricted to Han Chinese who lived in Xi'an city and its surrounding areas.

Clinical data and demographic

A standardized questionnaire including information such as region, age, gender, alcohol use, ethnicity, education status, and family history of cancer were used to collect necessary data through in-person interview. For patients, we consulted treating physicians or medical chart review to collect related information. We also tested the α-fetoprotein and plasma carcinoembryonic antigen to ensure the quality of controls.

SNP selection and genotyping

Twelve tSNPs with minor allele frequency (MAF) >5% in HapMap Asian population in nine genes (CCNH, CHAF1A, ERCC1, PARP1, RPA3, TP53, XRCC3, XRCC4, and XRCC6) previously reported to be associated with^3,9

–13 glioma risk were successfully genotyped. Genomic DNA from whole blood was extracted using phenol–chloroform method and their concentration was measured by spectrometry (DU530 UV/VIS spectrophotometer; Beckman Instruments, Fullerton, CA). We used Sequenom MassARRAY Assay Design 3.0 Software to design Multiplexed SNP MassEXTEND assay.¹⁴ We performed Sequenom MassARRAY RS1000 to genotype SNPs using the standard protocol recommended by the manufacturer. Sequenom Typer 4.0 Software was used to perform data management and analyses.^14,15

Statistical analysis

Microsoft Excel and SPSS 16.0 statistical package (SPSS, Chicago, IL) were used for data management. All p-values in this study were two-sided and p=0.05 was considered the cutoff value of statistical significance. An exact test was used to test each tSNP frequency's departure from Hardy–Weinberg equilibrium in control subjects. χ² test was used to calculate the genotype frequencies of cases and controls.¹⁶ We tested odds ratios (ORs) and 95% confidence intervals (CIs) using unconditional logistic regression analysis with adjustment for age and gender.¹⁷ The possibility of gender differences as a source of population substructure was evaluated by a genotype test for each tSNP in men and women, and the number of significant results at the 5% level was compared with the number expected by χ² test.¹⁶ We did not detect ethnic specificity because all participants' ethnicity was limited to Han Chinese.

We used three models (allele model, dominant model, and recessive model) to perform the associations of certain tSNP contributed to the risk of glioma. ORs and 95% CIs were calculated using unconditional logistic regression analysis with adjustment for age and gender.¹⁷

We also used the Haploview software package (version 4.2) to estimate the pairwise linkage disequilibrium (LD) between markers and partition haplotype blocks.¹⁸

Finally, we used the SHEsis software platform (www.nhgg.org/analysis/) for analyses of LD, haplotype construction, and genetic association at polymorphism loci.¹⁹

Results

We included a total of 603 participants, with 301 patients (157 men, 144 women; median age at diagnosis 41.5 years) and 302 controls (155 men, 145 women; median age 42.3 years) for genetic association analyses. Basic characteristics of the cases such as gender, age, and pathology are listed in Table 1. The primers of the 12 selected tSNPs are shown in Table 2, which were designed by Sequenom MassARRAY Assay Design 3.0 Software.¹⁴ Twelve tSNPs were successfully genotyped in all cases and controls. Average call rate of tSNPs was 99.57% (range from 98.51% to 100%) (listed in Table 3). The relationships between the tSNPs and glioma risk were listed in Table 4. We found no association between tSNPs and glioma onset at 5% level.

Table 1.

Basic Characteristics of the Cases

	Men		Women		Total subjects
Group of pathology typing	No.	Mean age and range (years)	No.	Mean age and range (years)	No.	Mean age and range (years)
Astrocytoma	99	42 (5–77)	74	40 (2–74)	173	41 (2–77)
Ependymoma	9	38 (6–60)	11	17 (1–71)	20	20 (1–60)
Glioblastoma	22	40.5 (17–69)	20	54.5 (10–64)	42	46 (10–69)
Oligodendroglioma	4	49 (38–69)	5	25 (17–43)	9	41 (17–69)
Others	23	43 (22–81)	34	48 (10–69)	57	46.5 (10–81)

Table 2.

Primers Used for This Study

SNP ID		1st-PCR primer sequences	2nd-PCR primer sequences	UEP sequences
rs2266690	C/T	ACGTTGGATGAGCTCAGCAGAATGACATCG	ACGTTGGATGAAGAAGTATGAACCACCCAG	acggATGAACCACCCAGATCTGAAGAAG
rs105038	C/T	ACGTTGGATGGGAAGCTCAAGGGAACAATC	ACGTTGGATGCTGTAGGTTACAGAGCTGTT	ccGAGCTGTTTTCTCAGTTCTG
rs3212986	G/T	ACGTTGGATGTTTAGTTCCTCAGTTTCCCG	ACGTTGGATGCACAGGCCGGGACAAGAAG	GGACAAGAAGCGGAAG
rs1136410	G/T	ACGTTGGATGGCTTTCTTTTGCTCCTCCAG	ACGTTGGATGTGAGCAGACTGTAGGCCAC	tCAGGTTGTCAAGCATTTCC
rs4140805	G/T	ACGTTGGATGTGAGGGAAGGCCTATATGTG	ACGTTGGATGTTTGGGGATCACTAATGATG	GGGATCACTAATGATGTATGTT
rs6947203	T/C	ACGTTGGATGCTAAAACACCTATGAAACAG	ACGTTGGATGTGTACCCATAAGATACCTG	ttcCCTGTTTTAAGAAAATAACTTACA
rs2160138	C/T	ACGTTGGATGCTAGGGCATCTAAAACTACC	ACGTTGGATGAGAGGAGGAGGAGATAAGAG	AGATAAGAGCAATCCAGC
rs1042522	C/G	ACGTTGGATGGGTGTAGGAGCTGCTGGTG	ACGTTGGATGAGACCCAGGTCCAGATGAAG	cccacGCCAGAGGCTGCTCCCC
rs8079544	T/C	ACGTTGGATGACACCATGCCAGTGTCTGAG	ACGTTGGATGAGAGAGAATGTGAAGCAGCC	CCTGCTCCACAGGAAGCC
rs861530	A/G	ACGTTGGATGTTTGACACAGCCTCCTCCAC	ACGTTGGATGAGGCCAAGAGCGCAGCCTG	GCGCAGCCTGCACCGC
rs1056503	G/T	ACGTTGGATGTTCTAGGCCTGATTCTTCAC	ACGTTGGATGTGTTTTCAGCTGAGATGTGC	cagaTGAGATGTGCTCCTTTTT
rs6519265	A/G	ACGTTGGATGCATTCCTTTTCATGGAATGT	ACGTTGGATGCACCACCTCTATCTTATTCC	tttttTGGCTAAACAGTGGC

SNP, single-nucleotide polymorphism; PCR, polymorphism chain reaction; UEP, unextended mini-sequencing primer.

Table 3.

Basic Information of Candidate Single-Nucleotide Polymorphisms in This Study

SNP No.	Location	Gene	Position (genome build 36.3)	HWE p-Value	p-Value	Call rate (%)
rs2266690		CCNH	86731030	0.55	0.7623	99.17
rs105038	19p13.3	CHAF1A	4365710	0.86	0.8818	99.50
rs3212986	19q13.32	ERCC1	50604576	0.12	0.2749	100.00
rs1136410	1q42.12	PARP1	224621925	0.55	0.5463	98.51
rs4140805	7p21.3	RPA3	7693626	0.37	0.1122	99.50
rs6947203	7p21.3	RPA3	7703573	0.99	0.4968	99.17
rs2160138	7p21.3	RPA3	7722322	0.35	0.0780	100.00
rs1042522	17p13.1	TP53	7520197	0.42	0.0580	100.00
rs8079544	17p13.1	TP53	7520777	0.69	0.8034	100.00
rs861530	14q32.33	XRCC3	103243876	0.98	0.3531	99.67
rs1056503	5q14.2	XRCC4	82684733	0.98	0.2586	99.67
rs6519265		XRCC6		0.39	0.3538	99.67

HWE, Hardy–Weinberg equilibrium.

Table 4.

Association Between Single-Nucleotide Polymorphism Genotypes and Risk of Glioma

		No. (frequency)		Logistic regression
SNP ID	Genotype	Case	Control	OR (95% CI)	p-Value
rs1042522	CC	63 (20.9)	45 (14.9)	1.52 (0.95–2.45)	0.0831
rs1042522	CG	146 (48.5)	157 (52)	1.01 (0.7–1.45)	0.9536
rs1042522	GG	92 (30.6)	100 (33.1)	1 (referent)
rs105038	CC	56 (18.7)	54 (18)	0.97 (0.6–1.54)	0.8818
rs105038	CT	143 (47.7)	152 (50.7)	0.88 (0.61–1.26)	0.4718
rs105038	TT	101 (33.7)	94 (31.3)	1 (referent)
rs1056503	TT	35 (11.7)	25 (8.3)	1.47 (0.84–2.57)	0.1793
rs1056503	TG	117 (39)	121 (40.2)	1.01 (0.72–1.42)	0.942
rs1056503	GG	148 (49.3)	155 (51.5)	1 (referent)
rs1136410	CC	50 (16.8)	52 (17.6)	0.86 (0.53–1.4)	0.5537
rs1136410	CT	149 (50)	155 (52.4)	0.86 (0.6–1.24)	0.4318
rs1136410	TT	99 (33.2)	89 (30.1)	1 (referent)
rs2160138	CC	9 (3)	18 (6)	0.46 (0.2–1.06)	0.062
rs2160138	CT	86 (28.6)	93 (30.8)	0.86 (0.6–1.22)	0.393
rs2160138	TT	206 (68.4)	191 (63.2)	1 (referent)
rs2266690	CC	4 (1.3)	4 (1.3)	1.05 (0.26–4.25)	0.7741
rs2266690	CT	60 (20)	46 (15.4)	1.37 (0.9–2.09)	0.1435
rs2266690	TT	236 (78.7)	248 (83.2)	1 (referent)
rs3212986	TT	30 (10)	21 (7)	1.36 (0.74–2.49)	0.3188
rs3212986	TG	127 (42.2)	144 (47.7)	0.84 (0.6–1.17)	0.3032
rs3212986	GG	144 (47.8)	137 (45.4)	1 (referent)
rs4140805	GG	9 (3)	17 (5.6)	0.49 (0.21–1.13)	0.0881
rs4140805	GT	82 (27.4)	91 (30.2)	0.84 (0.59–1.19)	0.3255
rs4140805	TT	208 (69.6)	193 (64.1)	1 (referent)
rs6519265	AA	1 (0.3)	4 (1.3)	0.26 (0.03–2.3)	0.3878
rs6519265	AG	48 (16.1)	43 (14.2)	1.14 (0.73–1.78)	0.5691
rs6519265	GG	250 (83.6)	255 (84.4)	1 (referent)
rs6947203	TT	3 (1)	5 (1.7)	0.55 (0.13–2.33)	0.6421
rs6947203	TC	53 (17.7)	69 (23.2)	0.71 (0.47–1.05)	0.0872
rs6947203	CC	244 (81.3)	224 (75.2)	1 (referent)
rs8079544	TT	1 (0.3)	1 (0.3)	0.96 (0.06–15.46)	0.4958
rs8079544	TC	38 (12.6)	49 (16.2)	0.75 (0.47–1.18)	0.2082
rs8079544	CC	262 (87)	252 (83.4)	1 (referent)
rs861530	GG	63 (21)	54 (17.9)	1.32 (0.83–2.1)	0.2345
rs861530	GA	147 (49)	145 (48.2)	1.15 (0.8–1.65)	0.4553
rs861530	AA	90 (30)	102 (33.9)	1 (referent)

OR, odds ratio; CI, confidence interval.

We further analyzed the association of tSNPs and glioma risk using logistic test including dominant model, recessive model, and additive model. The MA of each tSNP was assumed to be a risk factor and their frequencies (MAF) were listed in Table 5. We found two protective alleles and one risk allele, the allele “T” of rs6947203 in the gene RPA3 acts as a protective allele in dominant (OR, 0.59; 95% CI, 0.22–0.90; p=0.014) and additive (OR, 0.62; 95% CI, 0.42–0.92; p=0.0177) model. The allele “G” of rs4140805 in the gene RPA3 (OR, 0.73; 95% CI, 0.53–0.99; p=0.0437) was found to play a protective role in additive model. The allele “C” of rs1042522 in the gene TP53 (OR, 1.65; 95% CI, 1.05–2.59; p=0.0314) acts as a risk allele in recessive model.

Table 5.

Association Between Single-Nucleotide Polymorphisms and Risk of Glioma Based on Logistic Test and Their Heterozygote and Homozygote Odds Ratios, per Allele Odds Ratios and Confidence Intervals

				Dominant model				Recessive model				Additive model
SNP No.	Minor allele	Case MAF	Control MAF	OR	95%	CI	p-Value	OR	95%	CI	p-Value	OR	95%	CI	P-Value
rs2266690	C	0.45	0.41	1.18	0.23	1.83	0.4735	1.23	0.27	5.53	0.7882	1.16	0.78	1.72	0.4728
rs105038	C	0.43	0.43	0.87	0.19	1.26	0.4585	1.07	0.69	1.67	0.7672	0.96	0.75	1.23	0.7381
rs3212986	T	0.31	0.28	0.84	0.17	1.18	0.3214	1.64	0.87	3.08	0.1240	0.98	0.75	1.29	0.9098
rs1136410	C	0.42	0.44	0.82	0.19	1.19	0.2918	0.97	0.62	1.53	0.9094	0.90	0.70	1.16	0.4361
rs4140805	G	0.17	0.21	0.71	0.19	1.03	0.0695	0.53	0.22	1.28	0.1570	0.73	0.53	0.99	0.0437^*
rs6947203	T	0.11	0.09	0.59	0.22	0.90	0.014^*	0.65	0.14	3.01	0.5844	0.62	0.42	0.92	0.0177^*
rs2160138	C	0.31	0.31	0.74	0.19	1.06	0.1028	0.48	0.20	1.17	0.1053	0.74	0.54	1.00	0.0521
rs1042522	C	0.17	0.21	1.13	0.19	1.63	0.5086	1.65	1.05	2.59	0.0314^*	1.23	0.96	1.58	0.1009
rs8079544	T	0.08	0.08	0.68	0.25	1.11	0.1183	0.72	0.03	15.31	0.8321	0.69	0.43	1.11	0.1226
rs861530	G	0.10	0.13	1.16	0.19	1.68	0.4181	1.18	0.76	1.81	0.4657	1.13	0.88	1.44	0.3447
rs1056503	T	0.07	0.08	1.11	0.17	1.56	0.5638	1.39	0.78	2.46	0.2657	1.13	0.87	1.46	0.3482
rs6519265	A	0.46	0.42	1.07	0.24	1.70	0.7832	0.34	0.04	3.16	0.3451	1.00	0.65	1.54	0.9838

means the p-values ≤0.05 and have statistical significance.

MAF, minor allele frequency.

The relationship between the TP53 haplotypes and glioma risk are listed in Table 6. Haplotype “CC” in the TP53 gene was found to be associated with risk of glioma (OR, 1.3384; 95% CI, 1.0273–1.7437; Fisher's p=0.0307; Pearson's p=0.0306).

Table 6.

Haploview Results for the Blocks on Glioma Risk

Gene	Haplotype	Frequency (case)	Frequency (ctrl)	χ²	Fisher's p-value	Pearson's p-value	Odds ratio	[95% CI]
TP53	C C ^*	0.3938	0.3268	4.6743	0.0307	0.0306	1.3384	[1.0273,1.7437]
	C T	0.0619	0.0768	0.8097	0.3683	0.3682	0.7941	[0.4803,1.3132]
Δ	G C	0.5442	0.5965	2.6629	0.1028	0.1027	0.8079	[0.6253,1.0440]

p value≤0.05 indicates statistical significance.

Discussion

In the current study, we successfully genotyped 12 tSNPs in nine genes (CCNH, CHAF1A, ERCC1, PARP1, RPA3, TP53, XRCC3, XRCC4, and XRCC6) in the Han Chinese population and indentified 2 protective tSNPs in the RPA3 gene (allele “T” of rs6947203 and allele “G” of rs4140805) and 1 risk tSNP in the TP53 gene (allele “C” of rs1042522) to be associated with glioma. We also observed a haplotype of “CC” of the TP53 gene with an increased risk of 34% of developing glioma.

TP53 is located in 17p13.1, encodes tumor protein TP53, which responds to diverse cellular stresses to regulate target genes that induce cell cycle arrest, apoptosis, senescence, DNA repair, or changes in metabolism. TP53 protein is expressed at low level in normal cells and at a high level in a variety of transformed cell lines, where it is believed to contribute to transformation and malignancy. TP53 is a DNA-binding protein containing transcription activation, DNA-binding, and oligomerization domains. It is postulated to bind to a TP53-binding site and activate expression of downstream genes that inhibit growth and/or invasion, and thus function as a tumor suppressor. Mutants of TP53 that frequently occur in a number of different human cancers fail to bind the consensus DNA binding site, and hence cause the loss of tumor suppressor activity. Alterations of this gene occur not only as somatic mutations in human malignancies, but also as germline mutations in some cancer-prone families with Li–Fraumeni syndrome. Multiple TP53 variants due to alternative promoters and multiple alternative splicing have been found. These variants encode distinct isoforms, which can regulate TP53 transcriptional activity. Our result demonstrates that mutation in TP53 plays a risk role in glioma onset; this is supported by the role for TP53 inactivation in astrocytoma initiation and is consistent with the high frequency of TP53 mutations in Grade II human astrocytomas.^20,21 Another study in Indians showed higher Arg/Arg (rs1042522 G/C) genotype in gliomas compared with normal population (38% vs. 13%).²² Previous study in New York also found that for Ex4+119 C>G SNP (rs1042522), women with the heterozygous genotype (G/C) had a 32% increase in breast cancer risk.²³ Ours and the previous study analysis of SNP suggest that TP53 genotypes, haplotypes, and locus–locus interactions are associated with glioma risk. These findings indicate TP53 gene plays an important role in glioma onset.

As to gene RPA3, which is located in 7p22, two protective tSNPs (allele “T” of rs6947203 and allele “G” of rs4140805) have been found for this gene in our study; however, we have not found any previous study that showed the relationship between these two tSNPs and glioma onset, in fact, study based on this gene is rare. The protein encoded by this gene has 8 exons and 121 amino acids, it is a single-stranded DNA-binding protein that functions in many aspects of DNA metabolism and has a central role in DNA replication, playing an essential function in both initiation and elongation.^24

–28

Our study is the first to find SNPs in the RPA3 gene are associated to glioma onset. Further study should be concentrated on the exact mechanism of how tSNPs affect glioma onset and whether it has population differentiation.

Some limitations cannot be ignored in our study. First, population admixture, which may cause type-I error (false positive) for association study. In our study, all the samples we used were from the same hospital to avoid two or more definite selection bias and they are neutralized because they did not differ in geographical distributions or genotype frequencies. The race of all participants was limited to Han Chinese who lived in Xi'an city or nearby, and so substantial population admixture can be ignored in our study.

Second, sample size (301 glioma patients and 302 control subjects) was not relatively large enough for association studies. We could not find any gender-specific significant tSNPs because our sample size was not large enough and could not be subgrouped by gender (data not shown). We selected SNPs whose frequencies were higher than 5% in HapMap Asian populations to affirm the statistical power. We also planned a haplotype-based study to investigate whether certain haplotypes were associated with glioma onset.

In conclusion, our study provides new evidences for the relationship between tSNPs and glioma onset, which may shed light on the etiology of glioma and the tSNPs we found can be applied in clinical diagnosis and therapy.

Footnotes

Acknowledgments

The authors are grateful to all the patients and individuals for their participation. We would also like to thank the clinicians and other hospital staff who contributed to the blood sample and data collection for this study.

Disclosure Statement

This work is supported by the National 863 High-Technology Research and Development Program (No. 2012AA02A519).

References

Schwartzbaum

, Ahlbom

, Lonn

et al. An international case-control study of interleukin-4Ralpha, interleukin-13, and cyclooxygenase-2 polymorphisms and glioblastoma risk. Cancer Epidemiol Biomarkers Prev, 2007; 16:2448.

Idbaih

, Criniere

, Ligon

et al. Array-based genomics in glioma research. Brain Pathol, 2010; 20:28.

Liu

, Shete

, Hosking

et al. Genetic advances in glioma: Susceptibility genes and networks. Curr Opin Genet Dev, 2010; 20:239.

Hartmann

, Mueller

, von Deimling

. Pathology and molecular genetics of oligodendroglial tumors. J Mol Med (Berl), 2004; 82:638.

Reifenberger

, Collins

. Pathology and molecular genetics of astrocytic gliomas. J Mol Med (Berl), 2004; 82:656.

Bhatia

, Malik

, Fernandez-L

, Kenney

. p27Kip1, a double-edged sword in Shh-mediated medulloblastoma: Tumor accelerator and suppressor. Cell Cycle, 2010; 9:4307.

Zhou

, Ma

, Shi

et al. Polymorphisms of DNA repair gene XRCC1 and risk of glioma: A case-control study in Southern China. Asian Pac J Cancer Prev, 2011; 12:2547.

, Jin

, Zhang

et al. Polymorphisms of TREH, IL4R and CCDC26 genes associated with risk of glioma. Cancer Epidemiol, 2012; 36:283–287.

Rajaraman

, Wang

, Rothman

et al. Polymorphisms in apoptosis and cell cycle control genes and risk of brain tumors in adults. Cancer Epidemiol Biomarkers Prev, 2007; 16:1655.

10.

Bethke

, Webb

, Murray

et al. Comprehensive analysis of the role of DNA repair gene polymorphisms on risk of glioma. Hum Mol Genet, 2007; 17:800.

11.

Zhou

, Liu

, Zhang

et al. XRCC3 haplotypes and risk of gliomas in a Chinese population: A hospital-based case-control study. Int J Cancer, 2009; 124:2948.

12.

Liu

, Zhou

, Zhang

et al. Polymorphisms of LIG4 and XRCC4 involved in the NHEJ pathway interact to modify risk of glioma. Hum Mutat, 2008; 29:381.

13.

Liu

, Zhang

, Zhou

et al. Tagging SNPs in non-homologous end-joining pathway genes and risk of glioma. Carcinogenesis, 2007; 28:1906.

14.

Gabriel

, Ziaugra

, Tabbaa

. SNP genotyping using the Sequenom MassARRAY iPLEX platform. Curr Protoc Hum Genet, 2009Chapter 2:Unit 2, 12.

15.

Thomas

, Baker

, Debiasi

et al. High-throughput oncogene mutation profiling in human cancer. Nat Genet, 2007; 39:347.

16.

Adamec

. [Example of the use of the nonparametric test. Test ×2 for comparison of 2 independent examples] Cesk Zdrav, 1964; 12:613.

17.

Bland

, Altman

. Statistics notes. The odds ratio. BMJ, 2000; 320:1468.

18.

Barrett

, Fry

, Maller

, Daly

. Haploview: Analysis and visualization of LD and haplotype maps. Bioinformatics, 2005; 21:263.

19.

Shi

, He

. SHEsis, a powerful software platform for analyses of linkage disequilibrium, haplotype construction, and genetic association at polymorphism loci. Cell Res, 2005; 15:97.

20.

Louis

, Ohgaki

, Wiestler

et al. The 2007 WHO classification of tumours of the central nervous system. Acta Neuropathol, 2007; 114:97.

21.

Wen

, Kesari

. Malignant gliomas in adults. N Engl J Med, 2008; 359:492.

22.

Jha

, Pathak

, Chosdol

et al. TP53 polymorphisms in gliomas from Indian patients: Study of codon 72 genotype, rs1642785, rs1800370 and 16 base pair insertion in intron-3. Exp Mol Pathol, 2011; 90:167.

23.

Gaudet

, Gammon

, Bensen

et al. Genetic variation of TP53, polycyclic aromatic hydrocarbon-related exposures, and breast cancer risk among women on Long Island, New York. Breast Cancer Res Treat, 2008; 108:93.

24.

Wold

. Replication protein A: A heterotrimeric, single-stranded DNA-binding protein required for eukaryotic DNA metabolism. Annu Rev Biochem, 1997; 66:61.

25.

Iftode

, Daniely

, Borowiec

. Replication protein A (RPA): The eukaryotic SSB. Crit Rev Biochem Mol Biol, 1999; 34:141.

26.

Fanning

, Klimovich

, Nager

. A dynamic model for replication protein A (RPA) function in DNA processing pathways. Nucleic Acids Res, 2006; 34:4126.

27.

Wold

, Kelly

. Purification and characterization of replication protein A, a cellular protein required for in vitro replication of simian virus 40 DNA. Proc Natl Acad Sci U S A, 1988; 85:2523.

28.

Walther

, Bjerke

, Wold

. A novel assay for examining the molecular reactions at the eukaryotic replication fork: Activities of replication protein A required during elongation. Nucleic Acids Res, 1999; 27:656.