Association of MUTYH Gln324His and APEX1 Asp148Glu with colorectal cancer and smoking in a Japanese population

Background Genetic polymorphisms of DNA repair enzymes may lead to genetic instability and colorectal cancer carcinogenesis. Our objective was to measure the interactions between polymorphisms of repair genes and tobacco smoking in colorectal cancer. Methods The case-control study involved sixty-eight colorectal cancer patients and 121 non-cancer controls divided into non-smokers and smokers according to pack-years of smoking. The genetic polymorphisms of DNA repair enzymes,OGG1 Ser326Cys, MUTYH Gln324His, APEX1 Asp148Glu and XRCC1 Arg399Gln, were examined using PCR-RFLP. Results The MUTYH Gln324His showed strong significant associations with a risk of colorectal cancer (crude odds ratio [OR] 3.30, 95% confidence interval [95%CI] 1.44–7.60, p = 0.005; adjusted OR3.53, 95%CI 1.44–8.70, p = 0.006). The ORs for the APEX1 Asp148Glu were statistically significant (crude OR 2.69, 95%CI 1.45–4.99, p = 0.002; adjusted OR 2.33, 95%CI 1.21–4.48, p = 0.011). The ORs for the MUTYH Gln324His and the APEX1 Asp148Glu were statistically significant for colon cancer (adjusted OR 3.95, 95%CI 1.28–12.20, p = 0.017 for MUTYH Gln324His ; adjusted OR 3.04, 95%CI 1.38–6.71, p = 0.006 for APEX1 Asp148Glu). The joint effect of tobacco exposure and the MUTYH Gln324His showed a significant association with colorectal cancer risk in non-smokers (adjusted OR 4.08, 95%CI 1.22–13.58, p = 0.022) and the APEX1 Asp148Glu was significantly increased in smokers (adjusted OR 5.02, 95%CI 1.80–13.99, p = 0.002). However, the distributions of OGG1 Ser326Cys and XRCC1 Arg399Gln were not associated with a colorectal cancer risk. Conclusion Our findings suggest that the MUTYH Gln324His and the APEX1 Asp148Glu constitutes an increased risk of colorectal cancer, especially colon cancer. The MUTYH Gln324His is strongly associated with colorectal cancer susceptibility in never smoking history, whereas the APEX1 Asp148Glu genotype constitutes an increased risk of colorectal cancer when accompanied by smoking exposure.


Introduction
Colorectal cancer is a major cause of death and is influenced by genetic characteristics and environmental factors. Humans are exposed daily to a large variety of toxic and carcinogenic compounds due to habits such as tobacco smoking. Tobacco smoking produces major classes of carcinogenic compounds: polycyclic aromatic hydrocarbons (PAHs), aromatic amines, and heterocyclic amines (HCA). Several of these compounds can produce bulky DNA adducts [1]. The colorectal mucosa is exposed to these compounds through either the alimentary tract or the circulatory system. DNA adducts were detected in the colonic mucosa of smokers than in nonsmokers [2]. A previous study found that heavy smokers have a 2-3-fold elevated risk of colorectal adenoma [3]. Our previous data showed that genetic polymorphisms of NAT2 and CYP1A2 in metabolic processes contributed to colorectal cancer risk depending on smoking status in Japanese population [4]. Therefore, tobacco smoking might be a potential risk factor for colorectal cancer.
DNA repair genes are increasingly being studied for cancer risk because of their critical role in maintaining genome integrity. The base excision repair (BER) pathway, one of four major DNA repair pathways, has a principal role in the repair of mutations caused by oxidized or reduced bases [5]. Therefore, polymorphisms of DNA repair genes may increase the risk of colorectal cancer. In addition, smoking-induced oxidative DNA base modifications and single-strand breaks are repaired by the BER pathway. In the current study, we focused on genes encoding four key proteins in the BER pathway: OGG1 (8-oxoguanine DNA glycosylase), MUTYH/MYH (Mut Y homolog), APEX1/ APE1 (Apurinic/apyrimidinic endonuclease-1), and XRCC1 (X-ray cross-complementing group 1).OGG1 is a DNA glycosylase that removes 8-oxo-7, 8-dihydro-2'deoxyguanosine (8-oxo-G), which is the most stable form of a highly mutagenic oxidative DNA adduct that pairs with cytosine [6]. MUTYH is another DNA glycosylase that removes adenine paired with 8-oxo-G or 1, 2-dihydro-2-oxoadenine (2-OH-A) paired with guanine [7]. The 2-OH-A level is increased by exposure to reactive oxygen species [8]. APEX1 removes abasic sites formed in DNA cleavage by OGG1 and MUTYH and recruits DNA polymerase β and DNA ligase III [9]. X-ray cross-complementing group 1 (XRCC1) is a multidomain protein that interacts with poly-ADP-ribose polymerase, DNA ligase III and DNA polymerase β, and repairs DNA single-strand breaks by generating a single nucleotide repair patch (short-patch BER) [10].
We conducted a hospital-based case-control study to elucidate the DNA repair gene polymorphisms, OGG1 Ser326Cys (rs1052133), MUTYH Gln324His (rs3219489),APEX1 Asp148Glu (rs1130409) and XRCC1 Arg399Gln (rs25487). OGG1 Ser326Cys, APEX1 Asp148Glu and XRCC1 Arg399Gln have also been linked to a risk of colorectal cancer [11][12][13]. Germ-line variants, Tyr165Cys and Gly382Asp, of the MUTYH gene have been associated with colorectal adenomas in Caucasians, not in Asians [14][15][16]. Recent studies reported that MUTYH Gln324His mutation was the most frequent mutation in Japanese patients with adenomatous popyposis, and the gene polymorphisms was associated with the risk of proximal colon cancer in the Japanese population [17,18]. To our knowledge, few previous studies have examined the effect of these polymorphisms on the association between smoking and colorectal cancer [19,20]. These polymorphisms were analyzed to evaluate genetic susceptibility to colorectal cancer and the possible modification effect on the relationship between smoking and colorectal cancer risk.  [4]. The study design was approved by the Ethics Review Committee on Genetic and Genomic Research, Kobe University Graduate School of Medicine. Informed consent was obtained and detailed data on smoking were collected in a personal interview. The amount of smoke exposure was calculated in pack-years: the product of the number of years an individual had smoked and the average number of cigarettes smoked per day, converted into a standard pack of 20 cigarettes. Samples were coded after blood and data collection.

Genotyping
Genomic DNA used in the study was isolated for a previous study [4]. For MUTYH Gln324His, PCR was performed in a 25-μL reaction mixture containing 50 ng DNA, 250 mmol/l of each primer, 250 μmol/l of dNTPs, 1× PCR buffer, and 0.75 U of Ex Taq DNA polymerase (Takara, Shiga, Japan) using a programmable thermocycler PC-701 (Astec K.K, Fukuoka, Japan). The primer sequences were 5'-TGC CGA TTC CCT CCA TTC TCT CTT G-3' and 5'-TCT TGG CTT GAG TAG GGT TCG GG-3'. PCR conditions were 3 min at 94°C; 38 cycles of 1 min at 94°C, 1 min at 64°C, and 1 min at 72°C; followed by final extension for 10 min at 72°C. After amplification, the PCR products were digested with HpyCH4 III (New England Biolabs, Beverly, MA) for 4 h at 37°C in a final volume of 23 μl.
Digested fragments were separated by electrophoresis on 12% polyacrylamide gel and visualized by ethidium bromide staining using a 20-bp DNA Ladder (Takara, Shiga, Japan) as a size marker. All experiments included positive and negative controls for each polymorphism. In PCR-RFLP genotyping of MUTYH, complete digestion of PCR products produced 292-bp fragments for the Gln allele, and 239-bp and 53-bp fragments for the His allele (the 53bp fragment was too small to resolve accurately)( Figure  1). The genotypes of OGG1 Ser326Cys [21], APEX1 Asp148Glu [22], and XRCC1 Arg399Gln [23] were determined by PCR-RFLP analysis.

Statistical analysis
Differences in demographic variables, smoking and grouped genotypic frequencies between the patients and controls were evaluated using a Chi-square test. All reported P-values are two-sided with p < 0.05 considered to be significant. Hardy-Weinberg equilibrium was tested using a Chi-square test to compare the observed genotype frequencies with the expected frequencies among the control subjects. The association between variant genotypes and risk of colorectal cancer was estimated by computing the odds ratios (OR) and the 95% confidence interval (95%CI) from unconditional logistic regression analysis, adjusted for age, gender, and smoking habit. Statistical analysis was performed with SPSS software (version 14.0 for Windows; SPSS Japan Inc., Tokyo, Japan). The subjects were divided into two groups according to pack-years of smoking: non-smokers (pack-years = 0), and smokers (pack-years > 0). The OR for each variant genotype was calculated for each subgroup.
Genotyping and allele frequencies of OGG1 Ser326Cys,MUTYH Gln324His,APEX1 Asp148Glu, and XRCC1 Arg399Gln adjusted for gender, age and smoking habit are shown in

Discussion
The association between the risk of colorectal cancer and polymorphisms of four DNA repair genes in the BER pathway was investigated in a small case-control study. No significant relationship was apparent between OGG1 Ser326Cys and colorectal cancer risk. Previous reports have suggested that OGG1 Ser326Cys is associated with colorectal cancer in Caucasians [11,24], but not among Koreans [25]. Our findings in a Japanese population are consistent with the results from the Korean population study.
Interestingly, we found that the MUTYH Gln324His genotype has a strong significant association with colorectal cancer risk, especially colon cancer. Tao et al. [18] reported MUTYH Gln324His in Japanese was statistically significantly associated with increased risk of proximal colon, but not distal colon or rectal cancer. Therefore, their results are consistent with our study. Moreover, a recent study found that the activity of MUTYH Gln324His is 34% less active than that of wild type [26]. 8-oxo-G is generated by direct oxidation of DNA by a hydroxyl radical, whereas 2-OH-A is exclusively generated by oxidation of dATP in the nucleotide pool [6,7]. The 2-OH-A level is increased in human cancerous tissues compared to normal tissues [27]. Thus, for colorectal cancer, it is also possible that the enzyme of MUTYH Gln324His may have partially impaired in repair of 2-OH-A opposite guanine, compared to repair of adenine opposite 8-oxo-G, because of the difference in the origin of each oxidized base. We also found that MUTYH Gln324His was statistically associated with increased risk in never smokers. These results suggest that the MUTYH 324His variation may be associated with a risk of colorectal cancer due to an increased mutation frequency, containing environmental factors except smoking.
We indicated that the APEX1 Asp148Glu genotype has a specifically association with colon cancer risk. A previous study reported that this genotype was especially an increased risk of colon cancer risk [28]. We also found a statistically significant association between the APEX1 Asp148Glu genotype and colorectal cancer risk in combination with smoking exposure. Ito et al. [29] reported that the gene-environment interaction between current smoking and APEX1 148 Glu/Glu genotype was statistically significant for lung cancer risk. However, a previous study didn't found about the effect of smoking habit on association between the APEX1 Asp148Glu genotype and colorectal cancer risk [28]. This polymorphism is located within the endonuclease domain of the protein [30], but it does not reduce endonuclease activity [31]. The 148 Glu allele has also been associated with increased mitotic delay after exposure to ionizing radiation [22]. Our results indicate that the APEX1 variation may play an important role in colorectal cancer risk, containing a reduced ability to communicate with the other BER proteins. In contrast, for XRCC1 Arg399Gln variants, we found no relationship with colorectal cancer. The XRCC1 399Gln allele has been linked with a reduced risk of colorectal adenomas [12,13], and XRCC1 has also been associated with improved progress in patients who underwent chemotherapy, but not in those who received surgery alone [11]. The smoking has an effect on colon adenoma risks among carriers of XRCC1 codon 399 Arg alleles [19,20]. However, we were unable to detect these relationships in our cases.
Our data may be biased by the relatively small number as a hospital-based case-control study, because we have several limitations. Therefore, we would require further verification as predictive biomarkers in a larger study population and need to clarify the gene-environment interaction between smoking and these genotypes.
In conclusion, MUTYH Gln324His and APEX1 Asp148Glu polymorphisms are important risk factors for colorectal cancer, especially colon cancer, in the Japanese population. In particular, the MUTYH Gln324His genotype is associated with colorectal cancer susceptibility in never