hOGG1 Ser326Cys polymorphism and risk of childhood acute lymphoblastic leukemia in a Chinese population
To whom correspondence should be addressed. E-mail: email@example.com
Oxidative DNA damage caused by reactive oxygen species can produce 8-oxoguanine (8-oxoG) in DNA, which is misread and leads to G:C→T:A transversions. This can be carcinogenic. Repair of 8-oxoG by the base excision repair pathway involves the activity of human 8-oxoG DNA glycosylase 1 (hOGG1). Accumulating evidence suggests that the hOGG1 Ser326Cys polymorphism affects the activity of hOGG1 and might serve as a genetic marker for susceptibility to several cancers. To determine whether this polymorphism is associated with risk of childhood acute lymphoblastic leukemia (ALL) in Chinese children, we genotyped the hOGG1 Ser326Cys polymorphism (rs1052133) in a case–control study including 415 cases and 511 controls. We found that there was a significant difference in the genotype distributions of the hOGG1 Ser326Cys polymorphism between cases and controls (P = 0.046), and the combined genotypes Ser/Ser and Ser/Cys were associated with a statistically significantly decreased risk of ALL (adjusted odds ratio [OR] = 0.66, 95% confidence interval [CI] = 0.49–0.88, P = 0.005). Furthermore, we found a decreased risk for high risk ALL (adjusted OR = 0.60, 95% CI = 0.40–0.88, P = 0.005), low risk ALL (adjusted OR = 0.68, 95% CI = 0.47–0.99, P = 0.042), and B-phenotype ALL (adjusted OR = 0.63, 95% CI = 0.46–0.86, P = 0.003) among children with the Ser/Ser and Ser/Cys genotypes. Our results suggest that the hOGG1 Ser326Cys polymorphism is associated with susceptibility to childhood ALL in a Chinese population. (Cancer Sci 2011; 102: 1123–1127)
Acute lymphoblastic leukemia (ALL) is a clonal disease of a lymphoblast and is the most frequent malignancy in children younger than 15 years, accounting for approximately 25% of all pediatric malignancies.(1) Studies of developed countries report that cancer incidence among children has risen steadily since the 1950s, with 38 new cases per million occurring every year.(2) Like other cancers, childhood ALL is generally believed to be caused by a combination of genetic susceptibility factors and environmental exposures.(3)
DNA damage to hematopoietic precursor cells is essential for the development of leukemia,(4,5) which occurs at an estimated rate of 10 000 lesions per cell per day.(6) Correspondingly, several DNA repair pathways have evolved to maintain the genetic integrity of cells and prevent carcinogenesis, including base excision repair (BER),(7–9) mismatch repair, and nucleotide excision repair.(7)
The DNA lesion 8-oxoguanine (8-oxoG), produced by reactive oxygen species, is one of the most common forms of oxidative damage to DNA, and leads to G:C→T:A transversions, causing carcinogenesis.(10,11) Human 8-oxoG DNA glycosylase 1 (hOGG1) is a key enzyme in the BER pathway and repairs 8-oxoG.(7,12) Many functional and epidemiological studies have suggested that the Ser326Cys polymorphism in exon 7 of the hOGG1 gene might affect the activity of the enzyme,(13,14) and therefore may serve as a genetic marker for susceptibility to various cancers.(15–18) Additionally, a recent study of a Polish population indicated that the hOGG1 Cys/Cys genotype increases the risk of childhood ALL.(19) There have been no studies of the hOGG1 Ser326Cys polymorphism and risk of childhood ALL in a Chinese population.
To determine whether the hOGG1 Ser326Cys polymorphism is associated with susceptibility to childhood ALL in a Chinese population, we genotyped 415 childhood ALL patients and 511 cancer-free controls and compared genotype frequencies among the two groups.
Materials and Methods
Study subjects. The study included 415 childhood ALL patients and 511 cancer-free controls from the Affiliated Nanjing Children’s Hospital of Nanjing Medical University (Nanjing, China) or the Affiliated Children’s Hospital of Soochow University (Suzhou, China), between January 2007 and April 2010. All subjects (age range, 1–18 years) were genetically unrelated Han Chinese. The 415 ALL cases were newly diagnosed by morphology, histochemistry, immunology, cytogenetics, and molecular biology. Risk level and immunophenotype of these cases were determined according to the Suggestion of Diagnosis and Treatment of ALL in Childhood in China.(20) Briefly, high risk ALL was assigned if: the patient was younger than 12 months; peripheral blood white blood cell count was ≥100 × 109/L; t(9;22)/BCR-ABL or t(4;11)/MLL-AF4 fusion gene was present; response to early treatment was poor; or standard treatment failed. Medium risk ALL was assigned if: the patient was older than 10 years; peripheral white blood cell count was ≥50 × 109/L; the patient had central nervous system leukemia and/or testis leukemia; the patient was diagnosed as T-ALL; or if an infaust cytogenetic characteristic was present such as hypodiploidy with chromosome number <45, fusion gene t(4;11)/MLL-AF4 or other gene rearrangement of MLL, or fusion gene t(9;22)/BCR-ABL. Low risk ALL was assigned if none of the above risk factors were present.
Patients with other hematological disorders, previous cancer, radiotherapy, chemotherapy, or were deceased were excluded.(21) The 511 controls from the same geographic area had no malignant neoplastic or hematological diseases and were frequency matched by age (±5 years) and gender to the cases. We obtained demographic information from a face-to-face questionnaire after receiving informed consent from the parents of eligible subjects. Smoking status was recorded as negative if neither of the parents smoked during or after the pregnancy; drinking status was recorded as negative if neither of the parents consumed alcohol during or after the pregnancy; house paint exposure was negative if the child’s house was not painted during pregnancy or ever after birth. The research protocol was approved by the Medical Ethics Committee of Nanjing Children’s Hospital affiliated to Nanjing Medical University.
Genotyping. Genomic DNA was isolated from peripheral blood samples collected from all cases and controls by the phenol/chloroform method. The hOGG1 Ser326Cys polymorphism was detected using the TaqMan SNP Genotyping assay, and the 384-well ABI 7900HT Real Time PCR System (Applied Biosystems, Foster City, CA, USA). The primer sequences were 5′-CCTCCTACAGGTGCTGTTCAGTG-3′ and 5′-ACCCTTTCTGCGCTTTGCT-3′, and the probes were 5′-FAM-CCAATGCCGCCATG-MGB-3′ and 5′-HEX-CGCCAATCCCGCCA-MGB-3′, which were devised and manufactured by Nanjing Steed Biotechnology (Nanjing, China). The final volume for each reaction was 5 μL, consisting of 1.25 μL ddH2O, 2.5 μL TaqMan Master Mix, 0.25 μL F-prime, 0.25 μL R-prime, 0.125 μL FAM-probe, 0.125 μL HEX-probe, 0.125 μL ROX, and 5 ng DNA (Nanjing Steed Biotechnology). The PCR profile consisted of an initial denaturation step at 95°C for 10 min and 40 cycles of 95°C for 15 s and 60°C for 1 min. The fluorescence level was measured with an ABI 7900HT Real Time PCR System. Allele frequencies were determined by abi sds software.
This assay was carried out on each sample by two persons, independently, in a blind fashion. Approximately 10% of the samples were randomly selected for repeated genotyping for confirmation and the results were 100% concordant.
Statistical analysis. The χ2-test was applied to compare the differences in demographic characteristics, risk factors, and the hOGG1 genotypes between the cases and controls. A goodness-of-fit χ2-test was used to examine the Hardy–Weinberg equilibrium of the controls’ genotype distributions. The crude and adjusted odds ratios (OR) and their 95% confidence intervals (CI) were obtained by univariate and multivariate logistic regression analyses to evaluate the association between the hOGG1 genotypes and risk of ALL. Those adjusted factors included age, gender, parental smoking status, parental drinking status, and house painting status. Furthermore, we carried out stratification analysis according to the subsection of age, gender, parental smoking status, parental drinking status, house painting status, immunophenotype, and treatment branch. All of the tests were two-sided and a P-value <0.05 was considered statistically significant. All statistical analysis was carried out using sas software (version 9.1.3; SAS Institute, Cary, NC, USA).
Characteristics of the study subjects. The characteristics of 415 ALL cases and 511 cancer-free controls are presented in Table 1. Overall, there was no significant difference between the cases and controls in the distribution of age, gender or parental smoking status. We did observe that a statistically significantly greater number of parents were positive for drinking status among cases (40.5%) than among the controls (15.7%, P < 0.001). Additionally, there were more families who had their houses painted among cases (34.2%) than among controls (11.5%, P < 0.001). Immunophenotyping revealed that the proportion of patients with T-phenotype ALL was 13.3%, B-phenotype ALL was 86.0%, and all others was 0.7%. Furthermore, 206 patients (49.6%) were in the low risk category, 48 patients (11.6%) were medium risk, and the remaining 161 patients (38.8%) were high risk.
Table 1. Distribution of selected variables between acute lymphoblastic leukemia (ALL) cases and controls
| <6||207 (49.9)||251 (49.1)||0.818|
| ≥6||208 (50.1)||260 (50.9)||–|
| Male||252 (60.7)||322 (63.0)||0.475|
| Female||163 (39.3)||189 (37.0)||–|
|Parental smoking status|
| Negative||149 (35.9)||215 (42.1)||0.056|
| Positive||266 (64.1)||296 (57.9)||–|
|Parental drinking status|
| Negative||247 (59.5)||431 (84.3)||<0.001|
| Positive||168 (40.5)||80 (15.7)||–|
|House painting status|
| Negative||273 (65.8)||452 (88.5)||<0.001|
| Positive||142 (34.2)||59 (11.5)||–|
| T-ALL||55 (13.3)||–||–|
| B-ALL||357 (86.0)||–||–|
| Other‡||3 (0.7)||–||–|
| Low risk||206 (49.6)||–||–|
| Medium risk||48 (11.6)||–||–|
| High risk||161 (49.6)||–||–|
Association between hOGG1 Ser326Cys polymorphism and ALL. The observed genotypes and allele frequencies of hOGG1 among cases and controls and their associations with risk of ALL are shown in Table 2. The hOGG1 Ser326Cys genotypes among the control subjects were in agreement with the Hardy–Weinberg equilibrium (P = 0.414). We observed that there was a significant difference in the genotype distributions of the hOGG1 Ser326Cys polymorphism between the cases and controls (P = 0.046). Among the control subjects, the frequency of the hOGG1 Cys allele was 0.596, which is similar to previous findings in a Chinese study of renal cancer.(22) The results of multiple logistic regression analysis indicated that the Ser326Cys polymorphism was not associated with a significant increased risk of ALL (for Ser/Cys heterozygotes, adjusted OR = 0.72, 95% CI = 0.47–1.09, P = 0.118, and for Cys/Cys homozygotes OR = 1.20, 95% CI = 0.78–1.84, P = 0.394). We subsequently combined Ser/Cys and Cys/Cys genotypes to construct a dominant genetic model and found that there was no association with a risk of ALL. However, when we used the 326Cys/Cys homozygous genotype as a reference to construct a recessive model, the combined group of individuals with genotypes Ser/Ser and Ser/Cys were associated with a statistically significantly lower risk of ALL (adjusted OR = 0.66, 95% CI = 0.49–0.88, P = 0.005) (Table 2).
Table 2. Logistic regression analysis of association between the hOGG1 Ser326Cys polymorphism and risk of acute lymphoblastic leukemia
|Ser/Ser||69 (16.6)||79 (15.5)||0.216||0.046||1.00||–|
|Ser/Cys||174 (41.9)||255 (49.9)||–||–||0.72 (0.47–1.09)||0.118|
|Cys/Cys||172 (41.5)||177 (34.6)||–||–||1.20 (0.78–1.84)||0.394|
|Ser/Cys + Cys/Cys||346 (83.4)||432 (84.5)||–||–||0.90 (0.61–1.33)||0.612|
|Cys/Cys||172 (41.5)||177 (34.6)||–||–||1.00||–|
|Ser/Ser + Ser/Cys||243 (58.5)||334 (65.4)||–||–||0.66 (0.49–0.88)||0.005|
In further analysis, as shown in Table 3, we observed an association between the combined Ser/Ser and Ser/Cys genotypes and decreased risk of high risk ALL (adjusted OR = 0.60, 95% CI = 0.40–0.88, P = 0.005), and low risk ALL (adjusted OR = 0.68, 95% CI = 0.47–0.99, P = 0.042), and B-phenotype ALL cases (adjusted OR = 0.63, 95% CI = 0.46–0.86, P = 0.003). No significant associations were observed between the genotypes and risk of medium risk ALL or T-ALL.
Table 3. Association between hOGG1 Ser326Cys genotypes and treatment branch or immunophenotype of acute lymphoblastic leukemia (ALL)
|Controls||177 (34.6)||–||334 (65.4)||–||–|
|Cases||172 (41.5)||–||243 (58.5)||–||–|
|T-ALL||20 (36.3)||1.00||35 (63.7)||0.90 (0.48–1.68)||0.586|
|B-ALL||150 (42.0)||1.00||207 (58.0)||0.65 (0.48–0.88)||0.003|
|Other‡||2 (66.7)||1.00||1 (33.3)||0.26 (0.02–3.08)||0.285|
|High risk ALL||71 (44.1)||1.00||90 (55.9)||0.60 (0.40–0.88)||0.005|
|Medium risk ALL||20 (41.7)||1.00||28 (58.4)||0.72 (0.38–1.39)||0.329|
|Low risk ALL||81 (39.3)||1.00||125 (60.7)||0.68 (0.47–0.99)||0.042|
In this case–control study, we found that the Ser326Cys polymorphism of hOGG1 was associated with childhood ALL in a Chinese population. Specifically, we found that the group of individuals with combined genotypes Ser/Ser and Ser/Cys had a statistically significantly decreased risk of high risk ALL, low risk ALL, and B-phenotype ALL among those with childhood ALL. The epidemiologic data revealed that there were more alcohol drinkers among the parents of children with ALL compared to controls, and that more families of children with ALL lived in houses that had been painted compared to controls.
Several studies have suggested that maternal alcohol consumption during pregnancy and exposure to house paint influences childhood leukemia,(23–25) and our findings are consistent with prior results that these environmental exposures are associated with an increased risk of ALL. We found no significant difference in numbers of smoking parents between cases and controls in our study, although other studies have found that parental smoking at home was associated with ALL.(26,27) This discrepancy may be due to some other unidentified risk factors that affected the results of our present study.(3,28) To the best of our knowledge, this is the first study to show an association between the hOGG1 Ser326Cys polymorphism and childhood ALL risk in a Chinese population.
8-Oxoguanine in DNA, one of the most common forms of oxidative damage caused by reactive oxygen species, can be mutagenic or carcinogenic(10,11) and is repaired by the BER pathway. In the first step of BER, the hOGG1 enzyme, encoded by the hOGG1 gene on chromosome 3p26, recognizes the 8-oxoG lesions and catalyzes the cleavage of the glycosylic bond between the modified base and the sugar moiety, leaving an abasic apurinic/apyrimidinic site in DNA. The base excision is filled through the action of DNA polymerase β and DNA ligase I and III.(7) Functional studies have shown that the glycosylase activity of the Cys326 variant of the hOGG1 enzyme is more sensitive to inactivation by oxidizing agents compared to the Ser326 hOGG1 enzyme, and that cells carrying Cys/Cys alleles could more readily accumulate mutations under conditions of oxidative stress.(13,29,30)
A recent study of Polish children supports these enzyme functional studies, reporting that the hOGG1 Cys/Cys genotype increases the risk of childhood ALL whereas the Ser/Ser genotype strongly reduces the risk of this cancer.(19) Other studies report that the Ser326Cys polymorphism might be correlated with an increased risk of carcinoma and poorer cancer outcomes.(15–18) Specifically, one study found that the hOGG1 326Cys/Cys genotype contributes to the risk of lung adenocarcinoma and that this genotype may serve as an independent marker for cancer recurrence.(16) Nam et al.(31) found that the hOGG1 Ser326Cys polymorphism was significantly associated with prostate cancer, and Dhillon et al.(32) reported that prostate cancer risk was significantly increased in individuals homozygous for the Cys326 variant. Several other studies have reported associations between the hOGG1 Ser326Cys polymorphism and risk for different types of cancers, including breast cancer,(33) larynx cancer,(34) and esophageal cancer.(35) However, Ferguson et al.(36) showed that there was no statistically significant association between this polymorphism and risk of esophageal adenocarcinoma, Barrett’s esophagus, or reflux esophagitis. Possible explanations for this inconsistency could be differences in genetic background and small sample size.
A recent genome-wide association study did not show an association between the hOGG1 Ser326Cys polymorphism and childhood ALL,(37,38) however, this research studied fewer children from a different ethnic group. Our study found that the hOGG1 Ser/Ser and Ser/Cys combined genotypes were associated with a statistically significantly decreased risk of childhood ALL compared to individuals with the Cys/Cys genotype in a limited group of Chinese children. One possible explanation for this is that the Cys326 allele may cause reduced glycosylase hOGG1 activity and DNA repair activity.(13,14) Additional studies with larger sample sizes are necessary to validate our results.
Our subgroup analysis also suggested that those individuals carrying the Ser/Ser and Ser/Cys genotypes had a higher risk for B-phenotype ALL than for T-phenotype ALL, suggesting that B-ALL and T-ALL may have a different etiology involving different genetic and epigenetic defects.(39,40) Our results also suggested that individuals who carry the Ser/Ser and Ser/Cys genotypes had a decreased risk for high risk ALL and for low risk ALL. In other words, the Ser/Ser and Ser/Cys genotypes may be protective against childhood ALL. These results should be confirmed in large-scale studies, however, before being used to make predictions of disease incidence and severity.
One limitation of this study is that it was hospital-based, which may introduce a selection bias of subjects with a particular genotype. However, we applied a rigorous epidemiological design in recruitment of control subjects, and matched them by age and sex, in an attempt to minimize potential confounding factors. Another limitation could be that we only explored the environmental risk factors of parental smoking, parental drinking, and house painting. Several additional chemical, physical, and biological exposures, as well as maternal reproductive history,(3) have been suggested as risk factors for childhood ALL. Therefore, there may be other confounding risk factors that should be considered in future studies. Larger studies on diverse ethnic groups will help determine genetic factors influencing childhood ALL.
In conclusion, our data suggest that the hOGG1 Ser326Cys polymorphism might be a genetic susceptibility marker for childhood ALL among Chinese children, especially in the high risk ALL, low risk ALL, and B-phenotype ALL groups. To verify our findings, further functional evaluations in vivo and larger studies with more detailed environmental exposure data including diverse ethnic populations are warranted.
This research was partly supported by the National Natural Science Foundation of China (81070436) and the Nanjing Special Foundation for Medical Science (ZKM05039).
The authors have no conflicts of interest to report.