| Backgrounds & ObjectivesColorectal cancer including colon and rectum cancer is one of the most common cancers. In 2002, the number of global incident cases was about 1 million accounting for 9.4% of all cancers which is only less than lung cancer (1.35 million) and breast cancer (1.15 million). In terms of prevalence, it is only second to breast cancer (4.4 million, 17.9%) with an estimated 2.8 million persons alive with colorectal cancer diagnosed within 5 years of diagnosis which accouts for 11.5% of all cancers in world. In China, colorectal cancer is the third prevalent cancer both in males and females with an increasing incidence, of which the estimates in 2005 were 15.0 and 9.7 per 100,000 for males and females, respectively.As for the genetic predisposition to colorectal cancer, high-penetrance genes such as APC and DNA mismatch repair genes may account for hereditary colorectal cancer which is less than 5% of all colorectal cancer, such as familial adenomatous polyposis (FAP) and hereditary nonpolyposis colorectal cancer (HNPCC) known as monogenic diseases. Low-penetrance cnadidate genes mainly include DNA repair genes, metabolic enzyme genes, cell cycle regulatory genes, immune regulatory genes, oncogenes and tumor suppressing genes that are more common in natural population. Low-penetrance genes have an important influence on genetic predisposition to sporadic colorectal cancer (known as polygenic disease), which have intricate networks with environmental factors in carcinogenesis. DNA repair genes are the most important candidate genes, as the DNA repair system plays an important role in maintaining genome integrity which can protect the genome from carcinogens-induced damage to some extent. The genetic polymorphisms in DNA repair genes have an effect on individual's genetic susceptibility to spontaneous or induced cancer with the altered DNA repair capacity. P53 is a crucial tumor suppressor taking effect on DNA repair, cell cycle arrest and apoptosis in some cases via its stress response way, which is essential for genomic stability, and plays an important role in preventing tumor formation. Inactivating alterations in P53 gene is presented in about 50% of all human cancers. MDM2, encoded by mouse double minute 2 homolog gene (MDM2), is a key negative regulator of P53 stress response way. As a part of an autoregulatory negative feedback loop with P53 protein, overexpression of MDM2 gene can result in excessiveinactivation of P53 by blocking its transcription and mediating its degradation with E3 ubiquitin ligase activity. Polymorphisms in P53 and MDM2 may impair the function of P53 stress response way.A population-based case-control study was conducted to describe the frequency distribution of polymorphims in DNA repair genes (OGG1 Ser326Cys, XRCC1 Arg194Trp, Arg280His, Arg399Gln, XPD Lys751Gln and XRCC3 Thr241 Met) and in P53 pathway genes (P53 Intron3 16bp duplication, Exon4 Arg72Pro, Intron6 G/A transition, MDM2 Dell518 and SNP309) in natural Chinese population, and to explore the association of colorectal cancer risk with the above single polymorphism, haplotypes of XRCC1, P53 and MDM2, gene-environment interaction between the above polymorphims and common environmental exposure factors, and gene-gene interaction within the same pathway or between the different pathways.Materials & MethodsIn this population-based case-control study, 206 primay colorectal cancer cases and 845 cancer-free healthy controls were enrolled in, and they are all Chinese Han people coming from a follow-up cohort which has been built since 1989 in Jiashan County, the county with the highest age-standardized colorectal cancer mortality in China. After informed consent, all subjects were interviewed with a questionnaire mainly including demographic characteristics, individual lifestyle and family history of cancers by professional trained interviewers with face-to-face method. To ensure the validity of data, a repeat interview by telephone with five percent was performed. Also with the subjects' permission, a total 5ml blood was collected after interview, which was separated into two sections of 2ml blood with sodium citrate anticoagulation and 3ml blood without anticoagulation to gain blood serum. All blood samples were stored at -60°C for long-term conservation. The study was approved by the Medical Ethical Committee of Zhejiang University School of Medicine.The genomic DNA for each subject was extracted from whole blood using improved salting out procedure. As for the genotyping of polymorphisms in DNA repair genes, polymerase chain reaction-restriction fragment length polymorphism (PCR-RFLP) technique was used, which also was used for the genotyping of P53 Exon4 Arg72Pro and Intron6 G/A transition. The direct electrophoresis after PCR amplification was used for the genotyping of P53 Intron3 16bp duplication and MDM2 Del1518. Finally, the PCR- primer introduced restriction analysis (PIRA) method was used for the MDM2 SNP309 genotyping. Ten percent of all samples were randomly selected for repeat analysis to ensure obtaining the correct genotype information. All genotyping work was finished under the condition without knowledge of the subjects' case/control status.Student's t-test was used to evaluate the difference in mean age between cases and controls. Pearson's x~2 test was used to compare the distribution difference in categorial variables (genotype distribution et al) between cases and controls. The Hardy-Weinberg equilibrium test in control group was finished by X~2 goodness-of-fit test. The associations of different factors (0GG1 variants et al) with colorectal cancer risk were estimated by calculating the odds ratios (ORs) and their 95% confidence intervals (95% C/s) with multivariate logistic regression adjusted by age and sex. EH Linkage Software1.2 was used to identify the linkage disequilibrium. PHASE2.0 was used to construct haplotypes and estimate their frequency distributions. At last, MDR Software 1.0.0 was used for analyzing the gene-gene interaction. All statistical analysis was finished in Excel2003, SPSS 13.0 for windows, EH Linkage Softwarel.2, PHASE2.0, MDR-Data Tool Software 0.4.3 and MDR Software 1.0.0. All tests of statistical significance (α=0.05) were two-sided.Results1. Demographic characteristics, lifestyle factors and colorectal cancerThe mean age of colorectal cancer cases was 65.25 (SD=9.47) years, which was statistically differentfrom that of controls (X±SD: 61.84±10.83 years, P<0.01). Based on the quartile distribution of controls, the significant difference between two groups was also observed in the age distribution (P<0.01). As compared with the low age group (<53 years), the higher (>61 years) and the highest age group (≥72 years) had significant risks of colorectal cancer, and the ORs (95% C/s) were 2.89 (1.79-4.68) and 2.37 (1.45-3.88) adjusted by sex, respectively. There was no statistically significant difference in distributions of sex, BMI and family history of cancer in first and second relatives. However, as adjusted by sex and age, family history of cancer in first and second relatives was significantly associated with colorectal cancer risk (OR=1.51, 95%CI=1.05-2.17). As for the lifestyle factors, the frequency of heavy smokers (>20 cigarettes per day) in cases was statistically higher than that in controls (24.27% Vs. 16.18%), and the adjusted OR (95% CI) was 1.19 (1.05-1.35).2. Genetic polymorphisms of DNA repair genes and risk of colorectal cancerThe polymorphic variants of OGG1 Ser326Cys, XRCC1 Arg194Trp, Arg280His, Arg399Gln, XPD Lys751Gln and XRCC3 Thr241 Met in controls were 54.91%, 34.25%, 11.68%, 26.19%, 8.16% and 4.41%, which in cases were 54.57%, 29.95%, 11.22%, 28.22%, 10.15% and 4.90%, respectively. No significant difference was observed between cases and controls in allels and genotypes distribution. Association analysis for the different sub-type colorectal cancer indicated that 194Trp variant decreased risks of rectum cancer, the comparative early-onset colorectal cancer (age at diagnosis <60 years) and male colorectal cancer significantly, and the adjusted ORs (95% C/s) for variantheterozygotes and homozygotes as referred to wild homozygotes were 0.66 (0.44-0.99), 0.59 (0.38-0.92) and 0.64 (0.41-0.99), respectively. EH Linkage Software1.2 showed that there existed LD among XRCC1 three locuses, Arg194Trp, Arg280His and Arg399Gln, both in case and control group (Cases: X~2=36.74, Controls: X~2=200.71, df=7, P<0.01). However, no significant difference of the XRCC1 haplotype distribution existed between two groups (X~2=13.92, df=7, P>0.05). The PHASE progrom predicted CGG, TGG, CAG and CGA were the four most common haplotypes among all eight hyplotypes. Adjusted by age and sex, TGG haplotype (only with 194Trp variant) was associated with a statistically significant decreased risk of colorectal cancer compared with CGG hyplotype (without mutant variants), and the OR was 0.75 (95%CI=0.57-0.99). X~2 trend analysis found that 399Gln variant was correlated with the increase of colorectal cancer risk corresponding to the increase of smoke exposure dose (cigarettes per day) (P<0.05). The interaction between XPD 751 Lys/Gln & Gln/Gln genotypes and alcohol drinking (ever or current Vs. never) had an increasing risk of colorectal cancer but didn't reach statistically significant level (OR=1.90, 95%CI=0.97-3.71, P=0.06). Gene-gene interaction revealed that individuals with four putative high-risk genotypes statistically increased colorectal cancer risk (OR=3.63, 95%CI=1.24-10.62, P<0.05).3. Genetic polymorphisms of P53 pathway genes and risk of colorectal cancerThe polymorphic variants of P53 Intron3 16bp duplication, Exon4 Arg72Pro, Intron6 G/A transition, MDM2 Del1518 and SNP309 in controls were 3.80%, 46.49%, 5.97%, 40.65% and 43.78%, which in cases were 2.48%, 47.26%, 4.46%, 32.50% and 40.59%, respectively. Significant difference was observed between cases and controls in MDM2 Del1518 allels and genotypes distribution (P<0.05). Multivariate logistic regression adjusted by age and sex indicated that MDM2 Del1518 mutant homozygotes statistically decreased colorectal cancer risk especially rectum cancer risk, and ORs (95%C/s) were 0.50 (0.31-0.81) for colorectal cancer and 0.40 (0.20-0.79) for rectum cancer. EH Linkage Softwarel.2 revealed that LD existed among P53 Intron3 16bp duplication, Exon4 Arg72Pro and Intron6 G/A transition three locuses in controls (X~2=120.89, df=7, P<0.001), and between MDM2 Del1518 and SNP309 two locuses both in cases and controls (Cases: X~2=29.46, Controls X~2=25.53, df=3, P<0.01). What's more, statistical differences of the P53 (X~2=14.28, df=7, P<0.05) and MDM2 (X~2=22.78, df=3, P<0.01) haplotypes distribution existed between case and control groups. Compared with "- T" haplotype (both with wild alleles), "+ G" haplotype (both with mutant alleles) made colorectal cancer risk decrease statistically (adjusted OR=0.25, 95%CI=0.14-0.47). Gene-environment interaction analysis found that the interaction of P53 Intron3 16bp duplication and age decreased colorectal cancer risk significantly (OR=0.29, 95%CI=0.09-0.98. On the other hand, the interactions of P53 Exon4 Arg72Pro and family history of cancer or smoke exposure dose, MDM2 SNP309 and smoke exposure dose could increase risk of colorectal cancer but didn't reach the statistical significant level. As compared with individuals having less than threeputative high-risk genotypes, individuals with three or more putative high-risk genotypes had a statistical increased risk of colorectal cancer. And the ORs (95%C/s) were 3.94 (1.18-13.17) for individuals with three high-risk genotypes, 4.63 (1.41-15.17) for individuals with four high-risk genotypes and 6.08 (1.81-20.36) for individuals with five high-risk genotypes, respectively. MDR analysis found that the interaction of P53 Intron3 16bp duplication and MDM2 Del 1518 made colorectal cancer risk decrease statistically. The wild combinative genotype (without variants) was statistically associated with increase of colorectal cancer risk as compared with other three combinative genotypes (OR=1.37, 95%CI= 1.00-1.89). In consideration of all polymorphisms, single XRCC1 Arg194Trp or P53 Intron6 G/A transition polymorphism was associated with a statistical decrease of colorectal cancer risk referred to the wild combinative genotype. However, the mutant combinative genotype also was associated with the decrease of colorectal cancer risk but didn't have statistical significance.ConclusionsThe main findings from this population-based case-control study carried out in Chinese Han population were as follows:1. Age (≥61 years), family history of cancer in first and second relatives and heavy smoking (≥20 cigarettes per day) may contribute to the colorectal cancer predisposition.2. XRCC1 Arg194Trp polymorphism may have effect on colorectal cancer especially on rectum cancer, the comparative early-onset colorectal cancer (age at diagnosis <60 years) and male colorectal cancer. The gene-environment interaction between XRCC1 Arg399Gln polymorphism and cigarette smoke exposure may also have effect on the colorectal carcinogenesis. The accumulation of multi-polymorphisms in DNA repair genes has a contribution to the genetic predisposition of colorectal cancer.3. MDM2 Del1518 polymorphism and its interaction with P53 Intron3 16bp duplication may have influence on colorectal cancer especially on rectum cancer. P53 Intron3 16bp duplication and may be able to counteract the increasing risk of colorectal cancer resulting from age. The accumulation of multi-polymorphisms in P53 pathway genes plays an important role in colorectal cancer susceptibility.
|
PDF Full Text Request