Cell Biochem Biophys (2015) 71:977–982 DOI 10.1007/s12013-014-0296-7
ORIGINAL PAPER
Susceptibility and REF1 Gene Polymorphism Towards Colorectal Cancer Shibin Yang • Yuanhui Lai • Longbin Xiao Fanghai Han • Wenhui Wu • Shuo Long • Wenfeng Li • Yulong He
•
Published online: 26 October 2014 Ó Springer Science+Business Media New York 2014
Abstract Published data on the relation between REF1 polymorphism and colorectal cancer risk showed inconclusive results. The aim of this study was to derive a comprehensive estimation of the association. Data on association between REF1 polymorphism and colorectal cancer risk were summarized. The association was estimated by calculating an odds ratio (OR) with corresponding 95 % confidence interval (95 % CI) with the fixed effects model when P [ 0.1 (from heterogeneity test) or with the random effects model when P \ 0.1. No significant association was revealed in any genetic model assumed for the overall analysis (OR = 1.03, 95 % CI = 0.81–1.32 for Glu/Glu vs. Asp/Asp; OR = 1.05, 95 % CI = 0.96–1.15 for Glu/Glu ? Asp/Glu vs. Asp/ Asp; OR = 0.97, 95 % CI = 0.76–1.23 for Glu/Glu vs. Asp/Glu ? Asp/Asp; OR = 1.03, 95 % CI = 0.92–1.16 for Glu vs. Asp; OR = 1.09, 95 % CI = 0.93–1.27 for Asp/Glu vs. Asp/Asp). In Caucasian population, nor did we find a significant association. This research indicates that REF1 polymorphism is unlikely to be associated with colorectal cancer risk. Keywords
REF1 Polymorphism Colorectal cancer
S. Yang L. Xiao (&) F. Han W. Wu S. Long W. Li Y. He Department of Gastrointestinal and Pancreatic Surgery, The First Affiliated Hospital, Sun Yat-sen University, Guangzhou 510700, Guangdong, China e-mail: [email protected] Y. Lai Department of General Surgery I, Huangpu Division of the First Affiliated Hospital of Sun Yat-sen University, Guangzhou 510700, China
Introduction A great deal of evidence shows that activation of genetic predisposition to cancer is regulated through cooperation of high-risk single nucleotide polymorphisms (SNPs) in a panel of genes with low or medium penetrance [1]. As one of the leading causes of cancer morbidity and mortality [2], colorectal cancer is a common malignancy initiated by genetic variations, accounting for 35 % of total occurrence [3]. Genome-wide association studies (GWAS) have emerged as a powerful approach to discover SNPs relevant to cancers across human genome using high-density SNP chips. Several GWAS focused on examining effects of SNPs on candidate genes have identified a large number of susceptibility loci (1q41, 3q26.2, 12q13.13, and 20q13.33) for colorectal cancer [4, 5]. These data implicate that genetic susceptibility is important in the aetiology of colorectal cancer and that association study is an effective way to provide new insights to the knowledge of the current genetic basis. Genes in the DNA repair system, which is responsible for the maintenance of DNA integrity and thus triggers apoptosis and inhibits progression of carcinogenesis, have been recognized as possible candidates with a significant role in colorectal cancer susceptibility [6–8]. Genetic variation in DNA repair genes causes accumulation of DNA damage, accompanied by subsequent apoptotic dysfunction or uncontrolled cell growth. The base excision repair (BER) belongs to DNA repair system and plays a major role in the repair of both endogenous and exogenous DNA damage [9, 10]. REF1, a rate-limiting enzyme in the BER process, is involved in repairing genomic damage arising from exogenous agents and reactive oxygen species formed during normal metabolism [11]. REF1 gene in humans is located
123
978
at chromosome 14 q11.2–12 with four introns and five exons spanning 2.21 kb and encoding a 317-amino acid protein [12, 13]. SNPs of DNA repair pathways including REF1, respond to DNA injury through regulating individual repair capacity. Deficiencies in such repair can cause a range of human diseases [14]. REF1 polymorphism (rs1130409), with a T to G substitution, has been implicated in several reports that the 148Glu allele might promote mitotic delay after ionizing radiation exposure, modulate endonuclease and DNA-binding activity, decrease communicative capability with other BER proteins, and increase DNA damage induced by oxidation via weakening repairing action [15–18]. The genetic role of REF1 polymorphism in susceptibility to colorectal cancer has been investigated in several single studies [19–21]. However, there is no consistent evidence for this association. The relative small sample size might account for the inconsistency. In this study, therefore, we performed a meta-analysis of early and recently released articles to derive a comprehensive estimation of the association between REF1 polymorphism and colorectal cancer risk.
Cell Biochem Biophys (2015) 71:977–982
Fig. 1 Flow chart shows study selection process
Disputes on any item were resolved by discussion among all investigators involved in this meta-analysis. Statistical Analysis
Materials and Methods Study Search Strategy Using the databases of PubMed, Embase, and China National Knowledge Infrastructure (CNKI), we performed a systematic and comprehensive electronic search for case– control studies of REF1 polymorphism and colorectal cancer risk in humans. The search terms used were ‘‘REF1’’, ‘‘polymorphism’’, ‘‘polymorphisms’’, ‘‘genotypes’’, ‘‘variants’’, ‘‘Asp148Glu’’, ‘‘T1349G’’, ‘‘colon cancer’’, ‘‘rectal cancer’’, and ‘‘colorectal cancer’’. Retrieved meta-analyses and review articles were supplementarily searched to obtain additional original data. Inclusion Criteria In order to accurately examine the relation between REF1 polymorphism and colorectal cancer risk in humans, the studies must have a case–control design and publish sufficient data for calculating an odds ratio (OR) with corresponding 95 % confidence interval (95 % CI).
The strength of association between REF1 polymorphism and colorectal cancer risk was estimated by ORs with 95 % CIs assuming Glu/Glu vs. Asp/Asp, Glu/Glu ? Asp/Glu vs. Asp/Asp, Glu/Glu vs. Asp/Glu ? Asp/Asp, Glu vs. Asp, and Asp/Glu vs. Asp/Asp comparisons, for total populations as well as Caucasian subjects. v2-based Q-test was used to calculate heterogeneity across studies. Based on the Q-test results (P [ 0.1), OR of each study was summarized with the fixed effects model (the Mantel–Haenszel method); otherwise, the random effects model (the DerSimoniane and Laird method) was more appropriate [22, 23], by which wider CIs were estimated when there was inconsistency in the results of the constituent studies. Publication bias was checked by using funnel plots and Egger’s test [24]. Deviation from Hardy–Weinberg equilibrium was evaluated in the control subjects of each study. Sensitivity analyses were also conducted to assess the stability of the results. All statistical analyses were performed with the STATA version 12.0 software (Stata Corporation, College Station, TX). P \ 0.1 was deemed as the significant level.
Data Extraction
Results
Two independent investigators extracted data in duplicate and reached a consensus on all of the following items: first author, journal and year of publication, country of origin, ethnicity of subjects, source of controls, and genotyping information.
Study Selection and Characteristics
123
Fig. 1 presents a flow diagram of study identification for this meta-analysis. A total of 23 published articles were
Cell Biochem Biophys (2015) 71:977–982 Table 1 Study characteristics of published studies on the relation of REF1 polymorphism and colorectal cancer risk
&
P value for Hardy–Weinberg equilibrium in the control group
979
First author
Year
Country
Ethnicity of subjects
No. of cases
No. of controls
Source of controls
Hardy–Weinberg (P value&)
Moreno
2006
Spain
Caucasian
359
312
Hospital
Berndt
2007
US
Caucasian
692
710
Population
0.1
Pardini
2008
Czech
Caucasian
531
530
Hospital
0.7
Jelonek
2010
Poland
Caucasian
113
153
Healthy blood donors
0.1
Canbay
2011
Turkey
Caucasian
79
247
Healthy blood donors
\0.1
Li
2013
China
Asian
451
631
mixed
0.4
0.1
Table 2 Summary of odds ratios for the relation of REF1 polymorphism and colorectal cancer risk Stratifying factor and genotype
No. of studies*
Odds ratio
95 % confidence interval
PQ-test
Model
All Glu/Glu vs. Asp/Asp
6
1.03
0.81, 1.32
0.063
Random
Glu/Glu ? Asp/Glu vs. Asp/Asp
6
1.05
0.96, 1.15
0.196
Fixed
Glu/Glu vs. Asp/Glu ? Asp/Asp
6
0.97
0.76, 1.23
0.042
Random
Glu vs. Asp
6
1.03
0.92, 1.16
0.048
Random
Asp/Glu vs. Asp/Asp
6
1.09
0.93, 1.27
0.084
Random
Glu/Glu vs. Asp/Asp
5
1.00
0.73, 1.38
0.033
Random
Glu/Glu ? Asp/Glu vs. Asp/Asp
5
1.06
0.95, 1.17
0.121
Fixed
Glu/Glu vs. Asp/Glu ? Asp/Asp Glu vs. Asp
5 5
0.93 1.03
0.68, 1.27 0.89, 1.20
0.021 0.025
Random Random
Asp/Glu vs. Asp/Asp
5
1.11
0.91, 1.35
0.049
Random
Ethnicity Caucasian
* The number of studies pooled depended on the amount of information provided in the original paper
identified from databases and other sources, of which 12 full texts remained for further evaluation. Six studies were excluded, because two articles did not have the available data for REF1 polymorphism [25, 26], two provided incomplete genotype information [27, 28], and two were review articles [1, 29]. As a result, 6 case–control studies were selected for the meta-analysis [19–21, 30–32]. For the included studies, Table 1 lists their basic characteristics, including first author, year of publication, country, ethnicity of subjects, source of control, total of cases and controls, and P value of HWE. Overall, most studies used Caucasian subjects and only one used Asian population. Except the control group of the study done by Canbay et al. [32], all genotype distributions of controls were in agreement with HWE. Meta-Analysis Summary ORs for the relation of REF1 polymorphism and colorectal cancer risk are shown in Table 2. Notably, no significant association was revealed in any genetic model assumed for the overall analysis (OR = 1.03, 95 %
CI = 0.81–1.32 for Glu/Glu vs. Asp/Asp; OR = 1.05, 95 % CI = 0.96–1.15 for Glu/Glu ? Asp/Glu vs. Asp/ Asp; OR = 0.97, 95 % CI = 0.76–1.23 for Glu/Glu vs. Asp/Glu ? Asp/Asp; OR = 1.03, 95 % CI = 0.92–1.16 for Glu vs. Asp; OR = 1.09, 95 % CI = 0.93–1.27 for Asp/Glu vs. Asp/Asp). Similar to the findings indicated in the overall populations, stratified analysis in Caucasians suggested that the increased or decreased risk of colorectal cancer was not associated with REF1 polymorphism (Figs. 2, 3 and Table 2). Heterogeneity and Sensitivity Analyses There was substantial heterogeneity among the studies of the association between REF1 polymorphism and colorectal cancer risk (Table 2). Rather than the fixed effects model, the random effects model, which could generate wider CIs, was used for the OR estimations. After excluding the study, which was conducted by Jelonek et al. [31] and identified by sensitivity analysis, the significant heterogeneity disappeared without quantitive influence on overall results (data not shown).
123
980
Cell Biochem Biophys (2015) 71:977–982
Fig. 2 Forest plot for the risk of colorectal cancer associated with REF1 polymorphism under the Glu/Glu vs. Asp/Asp contrast model. Pooled odds ratio (OR) and 95 % confidence interval (95 % CI) have been appropriately calculated with the random effects model
Fig. 3 Forest plot for the risk of colorectal cancer associated with REF1 polymorphism under the Glu/Glu ? Asp/Glu vs. Asp/Asp contrast model. Pooled odds ratio (OR) and 95 % confidence interval (95 % CI) have been appropriately calculated with the fixed effects model
Publication Bias We performed funnel plots and Egger’s test to assess potential publication bias. The studies were symmetrically distributed in the funnel plots (P = 0.452). Egger’s test suggested that publication bias in this meta-analysis was not statistically significant (P = 0.227 for Glu/Glu ? Asp/ Glu vs. Asp/Asp) (Fig. 4).
Discussion The present meta-analysis, which was consisting of 2 225 cancer cases and 2 583 control casese, examined the
123
association between a potentially functional polymorphism, REF1 gene and colorectal cancer risk. The overall results from this meta-analysis did not indicate any association between any of the investigated genetic models of REF1 polymorphism and colorectal cancer risk, which was consistent with the result from the sensitivity analysis. Stratified analysis in Caucasian population showed compatible results. These findings, however, need to be treated with caution due to the small study samples. Colorectal cancer has a complex aetiology, which is under extensive investigation. Environmental factors such as food intake, tobacco and alcohol consumption, and genetic variations may have a consistent role in the onset of this cancer by stimulating DNA damage and promoting
Cell Biochem Biophys (2015) 71:977–982
Fig. 4 Funnel plot of publication bias analysis for the association between REF1 polymorphism and colorectal cancer risk (P = 0.227 for Glu/Glu ? Asp/Glu vs. Asp/Asp)
instability [33, 34]. Such damages can be protected by DNA repair systems [6]. Efficient DNA repair mechanisms comprise a critical component in maintaining the integrity of the genetic material and counterattacking cancer progression [35]. REF1 allows successful DNA repair by its normal 30 -hydroxyl nucleotide termini that impede further gap filling or religation. SNPs in DNA repair genes appear to be associated with reduced function of their encoded proteins, and consequently attenuate DNA repair capability and drive tumorigenesis [36]. REF1 polymorphism has been recently concerned in a variety of cancers with discoveries suggesting that it was not associated with breast cancer or bladder cancer among Asians or non-Asians [37, 38]. Colorectal cancer is another field with great interest in the effects of REF1 polymorphism. Two meta-analyses regarding the association this polymorphism and cancer risk demonstrated controversial conclusions. Gu et al. supported that REF1 polymorphism was a low-penetrance risk factor for cancer development, including colorectal cancer which was investigated in three publications with 1 629 cases and 1 599 controls [39]. However, this finding was challenged by a more recent study [40]. Many factors may lead to the different findings, but the number of the study subjects should be accounted for the larger part of the deviation. In addition, many epidemiologic studies presented a positive association between REF1 polymorphism and colorectal cancer risk [31, 32]. However, when combining all scattered sources into one dataset, our meta-analysis implicated a different observation, in which REF1 polymorphism could not be considered as an effect modifier of colorectal cancer risk, also not for the Caucasian population. Several possible factors contribute to the inconsistency: (a) the low statistical power leading to an exaggerated discovery; (b) insufficient study number to detect a moderate effect; (c) significant heterogeneity
981
across study populations; and (d) potential publication bias among the studies. Although a rigorous method was used in the evaluation of the relation between REF1 polymorphism and colorectal cancer risk, several limitations must be addressed. Firstly, substantial heterogeneity was detected across the studied populations. Fortunately, this did not introduce significant alternation to the overall results, suggesting the feasibility and practicality of this meta-analysis. Secondly, available published sources were selected and those about to be published or under investigation were beyond the inclusion, thus potential publication is inevitable. Thirdly, cancer is a multifactorial disease which initiation should largely attribute to the combined effects of genetic candidates interacting with environmental carcinogens. The exploration of the joint influence was not allowed due to the insufficient data. In summary, in spite of the limitations mentioned above, our meta-analysis suggested that REF1 polymorphism was not statistically related to the susceptibility to colorectal cancer. Stratified analysis did not reveal any significant association, either. To further validate the current findings, large and well-designed studies in different ethnic populations with gene–gene and gene–environment interaction analyzed are desperately needed.
Conflict of interest interest.
The authors have not declared any conflicts of
Bibliography 1. Hung, R. J., et al. (2005). Genetic polymorphisms in the base excision repair pathway and cancer risk: a huge review. American Journal of Epidemiology, 162(10), 925–942. 2. Ferlay, J., et al. (2010). Estimates of worldwide burden of cancer in 2008: globocan 2008. International Journal of Cancer, 127(12), 2893–2917. 3. Lichtenstein, P., et al. (2000). Environmental and heritable factors in the causation of cancer–analyses of cohorts of twins from sweden, denmark, and finland. New England Journal of Medicine, 343(2), 78–85. 4. Tenesa, A., & Dunlop, M. G. (2009). New insights into the aetiology of colorectal cancer from genome-wide association studies. Nature Reviews Genetics, 10(6), 353–358. 5. Houlston, R. S., et al. (2010). Meta-analysis of three genomewide association studies identifies susceptibility loci for colorectal cancer at 1q41, 3q26.2, 12q13.13 and 20q13.33. Nature Genetics, 42(11), 973–977. 6. Hoeijmakers, J. H. (2007). Genome maintenance mechanisms are critical for preventing cancer as well as other aging-associated diseases. Mechanisms of Ageing and Development, 128(7–8), 460–462. 7. Kaina, B. (2003). DNA damage-triggered apoptosis: critical role of DNA repair, double-strand breaks, cell proliferation and signaling. Biochemical Pharmacology, 66(8), 1547–1554. 8. de Jong, M. M., et al. (2002). Low-penetrance genes and their involvement in colorectal cancer susceptibility. Cancer Epidemiology Biomarkers and Prevention, 11(11), 1332–1352.
123
982 9. Wood, R. D., et al. (2001). Human DNA repair genes. Science, 291(5507), 1284–1289. 10. Goode, E. L., Ulrich, C. M., & Potter, J. D. (2002). Polymorphisms in DNA repair genes and associations with cancer risk. Cancer Epidemiology Biomarkers and Prevention, 11(12), 1513–1530. 11. Lu, A. L., et al. (2001). Repair of oxidative DNA damage: mechanisms and functions. Cell Biochemistry and Biophysics, 35(2), 141–170. 12. Robson, C. N., et al. (1992). Structure of the human DNA repair gene HAP1 and its localisation to chromosome 14q 11.2-12. Nucleic Acids Research, 20(17), 4417–4421. 13. Evans, A. R., Limp-Foster, M., & Kelley, M. R. (2000). Going APE over ref-1. Mutation Research, 461(2), 83–108. 14. Friedberg, E. C. (2003). DNA damage and repair. Nature, 421(6921), 436–440. 15. Xi, T., Jones, I. M., & Mohrenweiser, H. W. (2004). Many amino acid substitution variants identified in DNA repair genes during human population screenings are predicted to impact protein function. Genomics, 83(6), 970–979. 16. Hu, J. J., et al. (2001). Amino acid substitution variants of APE1 and XRCC1 genes associated with ionizing radiation sensitivity. Carcinogenesis, 22(6), 917–922. 17. Hadi, M. Z., et al. (2000). Functional characterization of Ape1 variants identified in the human population. Nucleic Acids Research, 28(20), 3871–3879. 18. Au, W. W., Salama, S. A., & Sierra-Torres, C. H. (2003). Functional characterization of polymorphisms in DNA repair genes using cytogenetic challenge assays. Environmental Health Perspectives, 111(15), 1843–1850. 19. Moreno, V., et al. (2006). Polymorphisms in genes of nucleotide and base excision repair: risk and prognosis of colorectal cancer. Clinical Cancer Research, 12(7 Pt 1), 2101–2108. 20. Pardini, B., et al. (2008). DNA repair genetic polymorphisms and risk of colorectal cancer in the czech republic. Mutation Research, 638(1–2), 146–153. 21. Li, Y., et al. (2013). Polymorphisms in genes of APE1, PARP1, and XRCC1: risk and prognosis of colorectal cancer in a northeast chinese population. Medical Oncology, 30(2), 505. 22. Mantel, N., & Haenszel, W. (1959). Statistical aspects of the analysis of data from retrospective studies of disease. Journal of the National Cancer Institute, 22(4), 719–748. 23. DerSimonian, R., & Laird, N. (1986). Meta-analysis in clinical trials. Controlled Clinical Trials, 7(3), 177–188. 24. Egger, M., et al. (1997). Bias in meta-analysis detected by a simple, graphical test. BMJ, 315(7109), 629–634. 25. Gao, Y., et al. (2011). DNA repair gene polymorphisms and tobacco smoking in the risk for colorectal adenomas. Carcinogenesis, 32(6), 882–887. 26. Obtulowicz, T., et al. (2010). Oxidative stress and 8-oxoguanine repair are enhanced in colon adenoma and carcinoma patients. Mutagenesis, 25(5), 463–471.
123
Cell Biochem Biophys (2015) 71:977–982 27. Kasahara, M., et al. (2008). Association of mutyh Gln324His and APEX1 Asp148Glu with colorectal cancer and smoking in a japanese population. Journal of Experimental Clinical Cancer Research, 27, 49. 28. Ye, Cici, H.Z., Zhou, Chunyan, 2010. APE1 D148E, PARP1 V762A and XRCC1 R399Q polymorphisms and genetic susceptibility to colorectal cancer. World Chinese Journal of Digestology, 18(12), 1275–1279. 29. Karahalil, B., Bohr, V. A., & Wilson, D. M, 3rd. (2012). Impact of DNA polymorphisms in key DNA base excision repair proteins on cancer risk. Human and Experimental Toxicology, 31(10), 981–1005. 30. Berndt, S. I., et al. (2007). Genetic variation in base excision repair genes and the prevalence of advanced colorectal adenoma. Cancer Research, 67(3), 1395–1404. 31. Jelonek, K., et al. (2010). Association between single-nucleotide polymorphisms of selected genes involved in the response to DNA damage and risk of colon, head and neck, and breast cancers in a polish population. Journal of Applied Genetics, 51(3), 343–352. 32. Canbay, E., et al. (2011). Association of APE1 and hOGG1 polymorphisms with colorectal cancer risk in a Turkish population. Current Medical Research and Opinion, 27(7), 1295–1302. 33. Giovannucci, E. (2001). An updated review of the epidemiological evidence that cigarette smoking increases risk of colorectal cancer. Cancer Epidemiology Biomarkers Prevention, 10(7), 725–731. 34. Terry, P., et al. (2001). Fruit, vegetables, dietary fiber, and risk of colorectal cancer. Journal of the National Cancer Institute, 93(7), 525–533. 35. Paz-Elizur, T., et al. (2008). DNA repair of oxidative DNA damage in human carcinogenesis: potential application for cancer risk assessment and prevention. Cancer Letters, 266(1), 60–72. 36. Spitz, M. R., et al. (2003). Genetic susceptibility to lung cancer: the role of DNA damage and repair. Cancer Epidemiology Biomarkers Prevention, 12(8), 689–698. 37. Liu, C., et al. (2013). APE1 Asp148Glu gene polymorphism and bladder cancer risk: a meta-analysis. Molecular Biology Reports, 40(1), 171–176. 38. Wu, B., et al. (2012). Lack of an association between two BER gene polymorphisms and breast cancer risk: a meta-analysis. PLoS ONE, 7(12), e50857. 39. Gu, D., et al. (2009). The DNA repair gene APE1 T1349G polymorphism and cancer risk: a meta-analysis of 27 case-control studies. Mutagenesis, 24(6), 507–512. 40. Zhou, B., et al. (2011). The association of APE1 -656T [ G and 1349 T [ G polymorphisms and cancer risk: a meta-analysis based on 37 case-control studies. BMC Cancer, 11, 521.
ORIGINAL PAPER
Susceptibility and REF1 Gene Polymorphism Towards Colorectal Cancer Shibin Yang • Yuanhui Lai • Longbin Xiao Fanghai Han • Wenhui Wu • Shuo Long • Wenfeng Li • Yulong He
•
Published online: 26 October 2014 Ó Springer Science+Business Media New York 2014
Abstract Published data on the relation between REF1 polymorphism and colorectal cancer risk showed inconclusive results. The aim of this study was to derive a comprehensive estimation of the association. Data on association between REF1 polymorphism and colorectal cancer risk were summarized. The association was estimated by calculating an odds ratio (OR) with corresponding 95 % confidence interval (95 % CI) with the fixed effects model when P [ 0.1 (from heterogeneity test) or with the random effects model when P \ 0.1. No significant association was revealed in any genetic model assumed for the overall analysis (OR = 1.03, 95 % CI = 0.81–1.32 for Glu/Glu vs. Asp/Asp; OR = 1.05, 95 % CI = 0.96–1.15 for Glu/Glu ? Asp/Glu vs. Asp/ Asp; OR = 0.97, 95 % CI = 0.76–1.23 for Glu/Glu vs. Asp/Glu ? Asp/Asp; OR = 1.03, 95 % CI = 0.92–1.16 for Glu vs. Asp; OR = 1.09, 95 % CI = 0.93–1.27 for Asp/Glu vs. Asp/Asp). In Caucasian population, nor did we find a significant association. This research indicates that REF1 polymorphism is unlikely to be associated with colorectal cancer risk. Keywords
REF1 Polymorphism Colorectal cancer
S. Yang L. Xiao (&) F. Han W. Wu S. Long W. Li Y. He Department of Gastrointestinal and Pancreatic Surgery, The First Affiliated Hospital, Sun Yat-sen University, Guangzhou 510700, Guangdong, China e-mail: [email protected] Y. Lai Department of General Surgery I, Huangpu Division of the First Affiliated Hospital of Sun Yat-sen University, Guangzhou 510700, China
Introduction A great deal of evidence shows that activation of genetic predisposition to cancer is regulated through cooperation of high-risk single nucleotide polymorphisms (SNPs) in a panel of genes with low or medium penetrance [1]. As one of the leading causes of cancer morbidity and mortality [2], colorectal cancer is a common malignancy initiated by genetic variations, accounting for 35 % of total occurrence [3]. Genome-wide association studies (GWAS) have emerged as a powerful approach to discover SNPs relevant to cancers across human genome using high-density SNP chips. Several GWAS focused on examining effects of SNPs on candidate genes have identified a large number of susceptibility loci (1q41, 3q26.2, 12q13.13, and 20q13.33) for colorectal cancer [4, 5]. These data implicate that genetic susceptibility is important in the aetiology of colorectal cancer and that association study is an effective way to provide new insights to the knowledge of the current genetic basis. Genes in the DNA repair system, which is responsible for the maintenance of DNA integrity and thus triggers apoptosis and inhibits progression of carcinogenesis, have been recognized as possible candidates with a significant role in colorectal cancer susceptibility [6–8]. Genetic variation in DNA repair genes causes accumulation of DNA damage, accompanied by subsequent apoptotic dysfunction or uncontrolled cell growth. The base excision repair (BER) belongs to DNA repair system and plays a major role in the repair of both endogenous and exogenous DNA damage [9, 10]. REF1, a rate-limiting enzyme in the BER process, is involved in repairing genomic damage arising from exogenous agents and reactive oxygen species formed during normal metabolism [11]. REF1 gene in humans is located
123
978
at chromosome 14 q11.2–12 with four introns and five exons spanning 2.21 kb and encoding a 317-amino acid protein [12, 13]. SNPs of DNA repair pathways including REF1, respond to DNA injury through regulating individual repair capacity. Deficiencies in such repair can cause a range of human diseases [14]. REF1 polymorphism (rs1130409), with a T to G substitution, has been implicated in several reports that the 148Glu allele might promote mitotic delay after ionizing radiation exposure, modulate endonuclease and DNA-binding activity, decrease communicative capability with other BER proteins, and increase DNA damage induced by oxidation via weakening repairing action [15–18]. The genetic role of REF1 polymorphism in susceptibility to colorectal cancer has been investigated in several single studies [19–21]. However, there is no consistent evidence for this association. The relative small sample size might account for the inconsistency. In this study, therefore, we performed a meta-analysis of early and recently released articles to derive a comprehensive estimation of the association between REF1 polymorphism and colorectal cancer risk.
Cell Biochem Biophys (2015) 71:977–982
Fig. 1 Flow chart shows study selection process
Disputes on any item were resolved by discussion among all investigators involved in this meta-analysis. Statistical Analysis
Materials and Methods Study Search Strategy Using the databases of PubMed, Embase, and China National Knowledge Infrastructure (CNKI), we performed a systematic and comprehensive electronic search for case– control studies of REF1 polymorphism and colorectal cancer risk in humans. The search terms used were ‘‘REF1’’, ‘‘polymorphism’’, ‘‘polymorphisms’’, ‘‘genotypes’’, ‘‘variants’’, ‘‘Asp148Glu’’, ‘‘T1349G’’, ‘‘colon cancer’’, ‘‘rectal cancer’’, and ‘‘colorectal cancer’’. Retrieved meta-analyses and review articles were supplementarily searched to obtain additional original data. Inclusion Criteria In order to accurately examine the relation between REF1 polymorphism and colorectal cancer risk in humans, the studies must have a case–control design and publish sufficient data for calculating an odds ratio (OR) with corresponding 95 % confidence interval (95 % CI).
The strength of association between REF1 polymorphism and colorectal cancer risk was estimated by ORs with 95 % CIs assuming Glu/Glu vs. Asp/Asp, Glu/Glu ? Asp/Glu vs. Asp/Asp, Glu/Glu vs. Asp/Glu ? Asp/Asp, Glu vs. Asp, and Asp/Glu vs. Asp/Asp comparisons, for total populations as well as Caucasian subjects. v2-based Q-test was used to calculate heterogeneity across studies. Based on the Q-test results (P [ 0.1), OR of each study was summarized with the fixed effects model (the Mantel–Haenszel method); otherwise, the random effects model (the DerSimoniane and Laird method) was more appropriate [22, 23], by which wider CIs were estimated when there was inconsistency in the results of the constituent studies. Publication bias was checked by using funnel plots and Egger’s test [24]. Deviation from Hardy–Weinberg equilibrium was evaluated in the control subjects of each study. Sensitivity analyses were also conducted to assess the stability of the results. All statistical analyses were performed with the STATA version 12.0 software (Stata Corporation, College Station, TX). P \ 0.1 was deemed as the significant level.
Data Extraction
Results
Two independent investigators extracted data in duplicate and reached a consensus on all of the following items: first author, journal and year of publication, country of origin, ethnicity of subjects, source of controls, and genotyping information.
Study Selection and Characteristics
123
Fig. 1 presents a flow diagram of study identification for this meta-analysis. A total of 23 published articles were
Cell Biochem Biophys (2015) 71:977–982 Table 1 Study characteristics of published studies on the relation of REF1 polymorphism and colorectal cancer risk
&
P value for Hardy–Weinberg equilibrium in the control group
979
First author
Year
Country
Ethnicity of subjects
No. of cases
No. of controls
Source of controls
Hardy–Weinberg (P value&)
Moreno
2006
Spain
Caucasian
359
312
Hospital
Berndt
2007
US
Caucasian
692
710
Population
0.1
Pardini
2008
Czech
Caucasian
531
530
Hospital
0.7
Jelonek
2010
Poland
Caucasian
113
153
Healthy blood donors
0.1
Canbay
2011
Turkey
Caucasian
79
247
Healthy blood donors
\0.1
Li
2013
China
Asian
451
631
mixed
0.4
0.1
Table 2 Summary of odds ratios for the relation of REF1 polymorphism and colorectal cancer risk Stratifying factor and genotype
No. of studies*
Odds ratio
95 % confidence interval
PQ-test
Model
All Glu/Glu vs. Asp/Asp
6
1.03
0.81, 1.32
0.063
Random
Glu/Glu ? Asp/Glu vs. Asp/Asp
6
1.05
0.96, 1.15
0.196
Fixed
Glu/Glu vs. Asp/Glu ? Asp/Asp
6
0.97
0.76, 1.23
0.042
Random
Glu vs. Asp
6
1.03
0.92, 1.16
0.048
Random
Asp/Glu vs. Asp/Asp
6
1.09
0.93, 1.27
0.084
Random
Glu/Glu vs. Asp/Asp
5
1.00
0.73, 1.38
0.033
Random
Glu/Glu ? Asp/Glu vs. Asp/Asp
5
1.06
0.95, 1.17
0.121
Fixed
Glu/Glu vs. Asp/Glu ? Asp/Asp Glu vs. Asp
5 5
0.93 1.03
0.68, 1.27 0.89, 1.20
0.021 0.025
Random Random
Asp/Glu vs. Asp/Asp
5
1.11
0.91, 1.35
0.049
Random
Ethnicity Caucasian
* The number of studies pooled depended on the amount of information provided in the original paper
identified from databases and other sources, of which 12 full texts remained for further evaluation. Six studies were excluded, because two articles did not have the available data for REF1 polymorphism [25, 26], two provided incomplete genotype information [27, 28], and two were review articles [1, 29]. As a result, 6 case–control studies were selected for the meta-analysis [19–21, 30–32]. For the included studies, Table 1 lists their basic characteristics, including first author, year of publication, country, ethnicity of subjects, source of control, total of cases and controls, and P value of HWE. Overall, most studies used Caucasian subjects and only one used Asian population. Except the control group of the study done by Canbay et al. [32], all genotype distributions of controls were in agreement with HWE. Meta-Analysis Summary ORs for the relation of REF1 polymorphism and colorectal cancer risk are shown in Table 2. Notably, no significant association was revealed in any genetic model assumed for the overall analysis (OR = 1.03, 95 %
CI = 0.81–1.32 for Glu/Glu vs. Asp/Asp; OR = 1.05, 95 % CI = 0.96–1.15 for Glu/Glu ? Asp/Glu vs. Asp/ Asp; OR = 0.97, 95 % CI = 0.76–1.23 for Glu/Glu vs. Asp/Glu ? Asp/Asp; OR = 1.03, 95 % CI = 0.92–1.16 for Glu vs. Asp; OR = 1.09, 95 % CI = 0.93–1.27 for Asp/Glu vs. Asp/Asp). Similar to the findings indicated in the overall populations, stratified analysis in Caucasians suggested that the increased or decreased risk of colorectal cancer was not associated with REF1 polymorphism (Figs. 2, 3 and Table 2). Heterogeneity and Sensitivity Analyses There was substantial heterogeneity among the studies of the association between REF1 polymorphism and colorectal cancer risk (Table 2). Rather than the fixed effects model, the random effects model, which could generate wider CIs, was used for the OR estimations. After excluding the study, which was conducted by Jelonek et al. [31] and identified by sensitivity analysis, the significant heterogeneity disappeared without quantitive influence on overall results (data not shown).
123
980
Cell Biochem Biophys (2015) 71:977–982
Fig. 2 Forest plot for the risk of colorectal cancer associated with REF1 polymorphism under the Glu/Glu vs. Asp/Asp contrast model. Pooled odds ratio (OR) and 95 % confidence interval (95 % CI) have been appropriately calculated with the random effects model
Fig. 3 Forest plot for the risk of colorectal cancer associated with REF1 polymorphism under the Glu/Glu ? Asp/Glu vs. Asp/Asp contrast model. Pooled odds ratio (OR) and 95 % confidence interval (95 % CI) have been appropriately calculated with the fixed effects model
Publication Bias We performed funnel plots and Egger’s test to assess potential publication bias. The studies were symmetrically distributed in the funnel plots (P = 0.452). Egger’s test suggested that publication bias in this meta-analysis was not statistically significant (P = 0.227 for Glu/Glu ? Asp/ Glu vs. Asp/Asp) (Fig. 4).
Discussion The present meta-analysis, which was consisting of 2 225 cancer cases and 2 583 control casese, examined the
123
association between a potentially functional polymorphism, REF1 gene and colorectal cancer risk. The overall results from this meta-analysis did not indicate any association between any of the investigated genetic models of REF1 polymorphism and colorectal cancer risk, which was consistent with the result from the sensitivity analysis. Stratified analysis in Caucasian population showed compatible results. These findings, however, need to be treated with caution due to the small study samples. Colorectal cancer has a complex aetiology, which is under extensive investigation. Environmental factors such as food intake, tobacco and alcohol consumption, and genetic variations may have a consistent role in the onset of this cancer by stimulating DNA damage and promoting
Cell Biochem Biophys (2015) 71:977–982
Fig. 4 Funnel plot of publication bias analysis for the association between REF1 polymorphism and colorectal cancer risk (P = 0.227 for Glu/Glu ? Asp/Glu vs. Asp/Asp)
instability [33, 34]. Such damages can be protected by DNA repair systems [6]. Efficient DNA repair mechanisms comprise a critical component in maintaining the integrity of the genetic material and counterattacking cancer progression [35]. REF1 allows successful DNA repair by its normal 30 -hydroxyl nucleotide termini that impede further gap filling or religation. SNPs in DNA repair genes appear to be associated with reduced function of their encoded proteins, and consequently attenuate DNA repair capability and drive tumorigenesis [36]. REF1 polymorphism has been recently concerned in a variety of cancers with discoveries suggesting that it was not associated with breast cancer or bladder cancer among Asians or non-Asians [37, 38]. Colorectal cancer is another field with great interest in the effects of REF1 polymorphism. Two meta-analyses regarding the association this polymorphism and cancer risk demonstrated controversial conclusions. Gu et al. supported that REF1 polymorphism was a low-penetrance risk factor for cancer development, including colorectal cancer which was investigated in three publications with 1 629 cases and 1 599 controls [39]. However, this finding was challenged by a more recent study [40]. Many factors may lead to the different findings, but the number of the study subjects should be accounted for the larger part of the deviation. In addition, many epidemiologic studies presented a positive association between REF1 polymorphism and colorectal cancer risk [31, 32]. However, when combining all scattered sources into one dataset, our meta-analysis implicated a different observation, in which REF1 polymorphism could not be considered as an effect modifier of colorectal cancer risk, also not for the Caucasian population. Several possible factors contribute to the inconsistency: (a) the low statistical power leading to an exaggerated discovery; (b) insufficient study number to detect a moderate effect; (c) significant heterogeneity
981
across study populations; and (d) potential publication bias among the studies. Although a rigorous method was used in the evaluation of the relation between REF1 polymorphism and colorectal cancer risk, several limitations must be addressed. Firstly, substantial heterogeneity was detected across the studied populations. Fortunately, this did not introduce significant alternation to the overall results, suggesting the feasibility and practicality of this meta-analysis. Secondly, available published sources were selected and those about to be published or under investigation were beyond the inclusion, thus potential publication is inevitable. Thirdly, cancer is a multifactorial disease which initiation should largely attribute to the combined effects of genetic candidates interacting with environmental carcinogens. The exploration of the joint influence was not allowed due to the insufficient data. In summary, in spite of the limitations mentioned above, our meta-analysis suggested that REF1 polymorphism was not statistically related to the susceptibility to colorectal cancer. Stratified analysis did not reveal any significant association, either. To further validate the current findings, large and well-designed studies in different ethnic populations with gene–gene and gene–environment interaction analyzed are desperately needed.
Conflict of interest interest.
The authors have not declared any conflicts of
Bibliography 1. Hung, R. J., et al. (2005). Genetic polymorphisms in the base excision repair pathway and cancer risk: a huge review. American Journal of Epidemiology, 162(10), 925–942. 2. Ferlay, J., et al. (2010). Estimates of worldwide burden of cancer in 2008: globocan 2008. International Journal of Cancer, 127(12), 2893–2917. 3. Lichtenstein, P., et al. (2000). Environmental and heritable factors in the causation of cancer–analyses of cohorts of twins from sweden, denmark, and finland. New England Journal of Medicine, 343(2), 78–85. 4. Tenesa, A., & Dunlop, M. G. (2009). New insights into the aetiology of colorectal cancer from genome-wide association studies. Nature Reviews Genetics, 10(6), 353–358. 5. Houlston, R. S., et al. (2010). Meta-analysis of three genomewide association studies identifies susceptibility loci for colorectal cancer at 1q41, 3q26.2, 12q13.13 and 20q13.33. Nature Genetics, 42(11), 973–977. 6. Hoeijmakers, J. H. (2007). Genome maintenance mechanisms are critical for preventing cancer as well as other aging-associated diseases. Mechanisms of Ageing and Development, 128(7–8), 460–462. 7. Kaina, B. (2003). DNA damage-triggered apoptosis: critical role of DNA repair, double-strand breaks, cell proliferation and signaling. Biochemical Pharmacology, 66(8), 1547–1554. 8. de Jong, M. M., et al. (2002). Low-penetrance genes and their involvement in colorectal cancer susceptibility. Cancer Epidemiology Biomarkers and Prevention, 11(11), 1332–1352.
123
982 9. Wood, R. D., et al. (2001). Human DNA repair genes. Science, 291(5507), 1284–1289. 10. Goode, E. L., Ulrich, C. M., & Potter, J. D. (2002). Polymorphisms in DNA repair genes and associations with cancer risk. Cancer Epidemiology Biomarkers and Prevention, 11(12), 1513–1530. 11. Lu, A. L., et al. (2001). Repair of oxidative DNA damage: mechanisms and functions. Cell Biochemistry and Biophysics, 35(2), 141–170. 12. Robson, C. N., et al. (1992). Structure of the human DNA repair gene HAP1 and its localisation to chromosome 14q 11.2-12. Nucleic Acids Research, 20(17), 4417–4421. 13. Evans, A. R., Limp-Foster, M., & Kelley, M. R. (2000). Going APE over ref-1. Mutation Research, 461(2), 83–108. 14. Friedberg, E. C. (2003). DNA damage and repair. Nature, 421(6921), 436–440. 15. Xi, T., Jones, I. M., & Mohrenweiser, H. W. (2004). Many amino acid substitution variants identified in DNA repair genes during human population screenings are predicted to impact protein function. Genomics, 83(6), 970–979. 16. Hu, J. J., et al. (2001). Amino acid substitution variants of APE1 and XRCC1 genes associated with ionizing radiation sensitivity. Carcinogenesis, 22(6), 917–922. 17. Hadi, M. Z., et al. (2000). Functional characterization of Ape1 variants identified in the human population. Nucleic Acids Research, 28(20), 3871–3879. 18. Au, W. W., Salama, S. A., & Sierra-Torres, C. H. (2003). Functional characterization of polymorphisms in DNA repair genes using cytogenetic challenge assays. Environmental Health Perspectives, 111(15), 1843–1850. 19. Moreno, V., et al. (2006). Polymorphisms in genes of nucleotide and base excision repair: risk and prognosis of colorectal cancer. Clinical Cancer Research, 12(7 Pt 1), 2101–2108. 20. Pardini, B., et al. (2008). DNA repair genetic polymorphisms and risk of colorectal cancer in the czech republic. Mutation Research, 638(1–2), 146–153. 21. Li, Y., et al. (2013). Polymorphisms in genes of APE1, PARP1, and XRCC1: risk and prognosis of colorectal cancer in a northeast chinese population. Medical Oncology, 30(2), 505. 22. Mantel, N., & Haenszel, W. (1959). Statistical aspects of the analysis of data from retrospective studies of disease. Journal of the National Cancer Institute, 22(4), 719–748. 23. DerSimonian, R., & Laird, N. (1986). Meta-analysis in clinical trials. Controlled Clinical Trials, 7(3), 177–188. 24. Egger, M., et al. (1997). Bias in meta-analysis detected by a simple, graphical test. BMJ, 315(7109), 629–634. 25. Gao, Y., et al. (2011). DNA repair gene polymorphisms and tobacco smoking in the risk for colorectal adenomas. Carcinogenesis, 32(6), 882–887. 26. Obtulowicz, T., et al. (2010). Oxidative stress and 8-oxoguanine repair are enhanced in colon adenoma and carcinoma patients. Mutagenesis, 25(5), 463–471.
123
Cell Biochem Biophys (2015) 71:977–982 27. Kasahara, M., et al. (2008). Association of mutyh Gln324His and APEX1 Asp148Glu with colorectal cancer and smoking in a japanese population. Journal of Experimental Clinical Cancer Research, 27, 49. 28. Ye, Cici, H.Z., Zhou, Chunyan, 2010. APE1 D148E, PARP1 V762A and XRCC1 R399Q polymorphisms and genetic susceptibility to colorectal cancer. World Chinese Journal of Digestology, 18(12), 1275–1279. 29. Karahalil, B., Bohr, V. A., & Wilson, D. M, 3rd. (2012). Impact of DNA polymorphisms in key DNA base excision repair proteins on cancer risk. Human and Experimental Toxicology, 31(10), 981–1005. 30. Berndt, S. I., et al. (2007). Genetic variation in base excision repair genes and the prevalence of advanced colorectal adenoma. Cancer Research, 67(3), 1395–1404. 31. Jelonek, K., et al. (2010). Association between single-nucleotide polymorphisms of selected genes involved in the response to DNA damage and risk of colon, head and neck, and breast cancers in a polish population. Journal of Applied Genetics, 51(3), 343–352. 32. Canbay, E., et al. (2011). Association of APE1 and hOGG1 polymorphisms with colorectal cancer risk in a Turkish population. Current Medical Research and Opinion, 27(7), 1295–1302. 33. Giovannucci, E. (2001). An updated review of the epidemiological evidence that cigarette smoking increases risk of colorectal cancer. Cancer Epidemiology Biomarkers Prevention, 10(7), 725–731. 34. Terry, P., et al. (2001). Fruit, vegetables, dietary fiber, and risk of colorectal cancer. Journal of the National Cancer Institute, 93(7), 525–533. 35. Paz-Elizur, T., et al. (2008). DNA repair of oxidative DNA damage in human carcinogenesis: potential application for cancer risk assessment and prevention. Cancer Letters, 266(1), 60–72. 36. Spitz, M. R., et al. (2003). Genetic susceptibility to lung cancer: the role of DNA damage and repair. Cancer Epidemiology Biomarkers Prevention, 12(8), 689–698. 37. Liu, C., et al. (2013). APE1 Asp148Glu gene polymorphism and bladder cancer risk: a meta-analysis. Molecular Biology Reports, 40(1), 171–176. 38. Wu, B., et al. (2012). Lack of an association between two BER gene polymorphisms and breast cancer risk: a meta-analysis. PLoS ONE, 7(12), e50857. 39. Gu, D., et al. (2009). The DNA repair gene APE1 T1349G polymorphism and cancer risk: a meta-analysis of 27 case-control studies. Mutagenesis, 24(6), 507–512. 40. Zhou, B., et al. (2011). The association of APE1 -656T [ G and 1349 T [ G polymorphisms and cancer risk: a meta-analysis based on 37 case-control studies. BMC Cancer, 11, 521.
