Abstract
Background
Mitochondrial manganese superoxide dismutase (MnSOD) converts superoxide anion into H2O2, which is neutralized sequentially by either catalase (CAT) or glutathione peroxidase 1 (Gpx 1) into water or converted into highly reactive hypochlorous acid by myeloperoxidase (MPO). We hypothesize that gene variants for these enzymes might be associated with the risk of breast cancer in non-smoking, non-alcohol-consuming women.
Methods
Genotypes of oxidative stress-related enzymes (MnSOD1183T>C, MPO-463G>A, GPx1Pro198Leu and CAT-262C>T) were analysed in 260 non-smoking and non-alcohol-consuming female patients with breast cancer and 224 habit-matched controls.
Results
Subjects with the MnSOD1183T>C C carrier or those with the GPx1Pro198Leu CT genotype had significantly decreased age-adjusted risks (odds ratio [OR]: 0.56 and 0.16 with 95% confidence intervals [95% CI]: 0.38–0.83 and 0.08–0.29, respectively) for breast cancer. Certain combined genotypes of the polymorphisms also significantly modulated the age-adjusted risk.
Conclusions
We conclude that oxidative stress-related enzyme genetic variants, especially GPx1Pro198Leu CT, modify the risk of breast cancer development in non-smoking and non-alcohol-consuming women. The role of unidentified environmental factors predisposing to breast cancer development through an oxidative stress mechanism merits further investigation.
Introduction
Breast cancer has become one of the most prevalent cancers in Taiwanese women as is also the case in the Western nations. Risk factors for breast cancer include both non-modifiable factors such as genetics and age, and modifiable factors such as lifestyle or the environment. 1,2 However, differences in breast cancer risk among individuals with similar carcinogen exposure may be due to genetic predisposition and dietary factors that have been shown to modulate or neutralize carcinogen-derived oxidative radicals. 3–5
Exposure to endogenous and exogenous oxidant sources generates reactive oxygen species (ROS), and the most common ROS in the cells is the superoxide anion (O2 ·−), which is produced during oxidative phosphorylation within the mitochondria. The endogenous antioxidant enzyme system, including superoxide dismutase (SOD), which can scavenge O2 ·− into H2O2, along with catalase (CAT) and glutathione peroxidase (GPx), which can convert H2O2 into H2O and O2, regulates the concentration of ROS generated by the mitochondrial respiratory chain. H2O2, if not neutralized, may contribute to further generation of ROS by a reaction catalysed by myeloperoxidase (MPO). An imbalance between the production and detoxification of ROS results in oxidative stress, which is known to play a major role in cancer development. 6,7 Therefore, a change in the concentration of O2 ·− and of H2O2 in the mitochondria regulates the molecular mechanisms of cell proliferation, apoptosis and cellular adhesion, and eventually initiates cell transformation towards carcinogenesis. 8 The products of the MnSOD (manganese superoxide dismutase), CAT, GPx1 and MPO genes are involved in the metabolism mentioned above, and variability in these enzymes may change the status of oxidative stress in an organism. Many researchers have shown that genetic polymorphisms can alter protein expression and/or function and have reported potential relationships between these single nucleotide polymorphisms (SNPs) and risk of development of breast cancer and clinical outcomes. 4,9–13 The aetiology of breast cancer is very complex and is associated with both genetic and environmental factors acting separately or in tandem. Previous research has shown that in female Taiwanese breast cancer patients, the prevalence of cigarette smoking and alcohol consumption (which might increase carcinogenicity through a free-radical-mediated mechanism) is lower than that in Western nations. Given that many environmental substances other than cigarettes and alcohol might induce cellular damage mediated by oxidative stress, we suggest that there would be merit in examining the association between the oxidative stress-related genetic polymorphisms described above and risk of non-smoking, non-alcohol-associated breast cancer. Furthermore, concentrations of cytotoxic ROS may depend on the balance between the activities of MnSOD, MPO, CAT and GPx, the activity of these enzymes modulated by functional polymorphisms which will in turn therefore affect the concentrations of ROS, and influence the ROS and/or antioxidant status. In the present study, we explore the relationship and interaction between SNPs of ROS defense genes (MnSOD1183T>C, GPx1Pro198Leu and CAT-262C>T) and a ROS-producing gene (MPO-463G>A) in a cohort of non-smoking, non-alcohol-consuming Taiwanese breast cancer patients.
Materials and methods
Subjects
The study was conducted at the Kaohsiung Medical University under a project approved by the Ethics Committee at Kaohsiung Medical University Hospital. Informed consent, demographic information including age, age of menopause and other related information were obtained from participants by questionnaire at the time of blood sampling. A total of 260 breast cancer patients and 224 controls were enrolled from Kaohsiung Medical University Hospital (Kaohsiung, Taiwan) in this study. Women who had never smoked or consumed alcohol and had a first diagnosis of histopathologically confirmed breast cancer, from whom blood samples were available, were selected as the cases. None of the cases had a strong family or early onset breast cancer (under 35 years old). Non-smoking and non-drinking women, without a present or previous history of breast cancer, were recruited as control subjects when they attended Kaohsiung Medical University Hospital for a routine annual general health check. Women with benign breast tumours, mastitis, benign calcification or other malignant disease such as lung or liver cancer, which might be related to polymorphisms of oxidative-related enzymes, were excluded from both case and control groups.
Genotyping for SNPs
Blood samples were drawn into sterile tubes containing sodium ethylenediaminetetraacetic acid, and genomic DNA was extracted from the buffy coat by salt extraction. The MnSOD1183C>T, GPx1Pro198Leu, MPO-463G>A and CAT-262C>T SNPs were determined by the method of the realtime polymerase chain reaction (PCR) with a DNA thermal cycler (MJ Research PTC-200 Peltier Thermal Cycler; Bio-Rad Co, Hercules, CA, USA) in a 96-well format. The realtime PCR assay consisted of two primers for PCR amplification of the sequence of study and two allele-specific probes. The SNPs, primer and probe were as follows. MnSOD1183C>T, primers: forward 5′ AGCCTGCGTAGACGGTCC 3′ and reverse 5′ TCGGGGAGGCTGTGCTTC 3′, allele-specific probes: forward 5′ 6-FAM-AGCCCAGATACCCCAAAACCGGAGCC-TAMRA 3′ and reverse 5′HEX-AGCCCAGATACCCCAAAGCCGGAGCC-TAMRA 3′; GPx1 Pro198Leu, primers: forward 5′ CCCCTACGCAGGTACAGC 3′ and reverse 5′ ACACCCTCATAGATGAAAACCC 3′, allele-specific probes: forward 5′ HEX-CGCGATCGTCTCAAGGGCCCAGCTGTGCCTGATCGCG (BHQ1) 3′ and reverse 5′ 6-FAM-CGCGATCGTCTCAAGGGCTCAGCTGTGCCTGATCGCG (BHQ1) 3′; MPO-463G>A, primers: forward 5′ GCTGGTAGTGCTAAATTCAAAGG 3′ and reverse 5′ TAGATACAGGGTTTCACCATGTTG 3′, allele-specific probes: forward 5′ CGCGATCAGTGATCCACCCGCCTCAGCCTCGATCGCG 3′ and reverse 5′ CGCGATCAGTGATCCACCTGCCTCAGCCTCGATCGCG 3′; CAT-262C>T, primers: forward 5′ GGCCTGAAGGATGCTGATAACC 3′ and reverse 5′ AGGGTGCGGAAAGGAAGGG 3′, allele-specific probes: forward 5′ CGGCTATCCCGGGCAC 3′ and reverse 5′ CGGCTATTCCGGGCAC 3′. The genotypes of the PCR products were confirmed by the DNA sequence analysis. In each experiment, DNA samples from the subjects, together with two or three previously sequenced DNA samples serving as a quality control material (one for each genotype) to validate genotyping procedures, were concomitantly amplified by the realtime PCR. No MnSOD Ile58Thr polymorphism was found in any participants in our study. Meanwhile, none of the 260 patients was homozygous for the polymorphism of the GPx1Pro198Leu T allele and CAT-262C>T T allele.
Statistical methods
Data were analysed by SPSS for Windows (Version 14; SPSS Inc, Chicago, IL, USA). However, the power analysis was conducted using Power Analysis and Sample Size (PASS) software (NCSS, Kaysville, UT, USA) which employs calculation according to Hsieh et al. 14 In brief, the Student's t-test was performed to assess continuous variables such as age; the chi-squared test for contingency tables was used to assess differences in binominal variables such as menopausal status across the case and control groups. The unconditional logistic regression model was performed to calculate the age-adjusted odds ratios (OR) with their associated 95% confidence intervals (95% CI) between genotypes with and without breast cancer. For each SNP, the odds ratio and 95% CI were calculated for each genotype and then for combinations of genotypes. The two-order interaction was assessed in the logistic regression model by including dummy variables for each category defined by the cross-classification of the interacting variables, except the reference category. In brief, four related dummy variables were created for two different SNPs in the interaction analysis encoded as 0000, 0100, 0010 and 0001. The last three variables were included as independent variables for the logistic regression model to test both the independent effects of SNPs and their combined effect. A two-tailed P value of < 0.05 was considered to be statistically significant.
Results
General clinical characteristics of the cases and controls
Clinical characteristics of the patients with breast cancer and the controls
P value < 0.05: significant
Independent effects of MnSOD1183T>C, MPO-463G>A, GPx1Pro198Leu and CAT-262C>T
Estimated odds ratios (OR) and 95% confidence interval (95% CI) of genetic polymorphisms in 260 patients with breast cancer and 224 controls
*ORs are adjusted by age
Two-order epistasis (gene–gene interaction) effects of SNPs
Genotype distributions and age-adjusted odds ratios (OR) of different combinations of MnSOD1183T>C and either of the GPx1Pro198Leu, MPO-463G>A or CAT-262C>T genes in 260 patients with breast cancer and 224 controls
*ORs are adjusted by age
Discussion
The ultimate aim of this study was to evaluate the risk of previous oxidative stress or damage on the development of breast cancer (in the absence of exposure to cigarettes or alcohol) by utilizing the polymorphisms of oxidative stress-related enzyme genes as proxies, which supposes that the polymorphisms of these enzymes might affect the capacity to remove excess free radicals. The polymorphisms of oxidative stress-related enzyme genes studied here were reported in many previous studies to be related to enzyme activities, 9–11 and therefore modify the capacity to eliminate free radicals. In our study we found that the frequencies of certain gene polymorphisms in enzymes responsible for scavenging free radicals (in particular SOD, which converts O2 ·− to H2O2) were significantly different between the breast cancer cases and the controls. This suggests that O2 ·− might play a role in breast cancer development in non-smoking and non-alcohol-consuming Taiwanese women. Numerous publications have suggested that free radicals are involved in the pathophysiology of breast cancer. 16,17 Our study found that the frequency of the MnSOD1183T>C TT genotype was higher in the cases than in the controls, which suggests the importance of the C allele in protecting against, and the T allele in predisposing to, breast cancer. This finding is consistent with previous studies 18,19 which reported that the C allele promotes the import of MnSOD into the mitochondria; it might therefore be associated with lower cancer risk. O2 ·−, generated from the inner mitochondrial membrane, is vectorially released into the mitochondrial matrix, and is acted upon by MnSOD in the matrix to generate H2O2. However, inhibition of MnSOD (as in the Val form) can cause accumulation of O2 ·− leading to damage of the mitochondrial membrane, and release of cytochrome c followed by apoptosis. 8 However, this finding conflicts with data from other studies, 9,10 which suggest that women with the MnSOD1183T>C C allele were at higher risk of breast cancer. The inconsistency of results might be explained by different genomic types, or by different lifestyles, such as different degrees of exposure to chemical carcinogens and/or different amounts of intake of antioxidants. Several studies reported a gene–environment interaction between this polymorphism and the concentrations of antioxidants, 3,9 but these concentrations were strongly influenced by cultural dietary habits, which possibly explains the inconsistent results observed in different studies. 8 In comparing our study to that of Ambrosone et al., sample sizes and allele frequencies in control groups were similar, but genotype distribution in cases differed as did the age at diagnosis. Finally, the inconsistencies in findings between these studies might also be due to different pathways in carcinogenesis for early and late onset disease, or due to linkage disequilibrium of the polymorphism to another cancer-prone gene in proximity to MnSOD. It should be noted that the case group in our study consisted of subjects without prior exposure to cigarettes or alcohol. This differs from previous studies conducted in Western countries in which smoking and alcohol consumption were more prevalent in women and may constitute important risk factors for the development of breast cancer. Based on the results of our study, we suggest that a factor other than smoking or alcohol might serve as a free-radical-generating environmental agent and increase the risk of carcinogenesis of breast cancer in Taiwan.
A previous study by Hu et al. 20 indicated that allelic variants of the GPx1 gene were contributing factors in breast cancer development. Another study found that carriers with the variant T-allele of the GPX1 Pro198Leu polymorphism had a slightly higher risk of breast cancer than homozygous wild-type individuals. 5 Nevertheless, this study pointed out that the frequency of the low activity genotype (Leu-GPx1Pro198Leu) was higher in the control group. The role of glutathione (GSH) remains a controversial issue in breast cancer development. The high intracellular reduced GSH concentrations are associated with the enhancement of cell proliferation and resistance to apoptosis in cancer cells, while a reduction in intracellular erythrocyte reduced GSH concentration might be due to the increased detoxification capacity and defense against oxidative stress. Moreover, insufficient intracellular selenium (Se) could impair the stability and translation of GPx1 mRNA. 21 Consequently, GPX1Pro198Leu polymorphisms might not serve as reliable indicators for changes in enzyme activity without information on blood or intracellular Se concentrations. Unfortunately, in our study and other studies, it has not been possible to measure blood or intracellular Se concentrations at the time of, or before, cancer diagnosis. Conversely, the results indicated that women with the low-activity MPO-463G>A and CAT-262C>T polymorphisms had a non-significant trend for reduced and increased risk of breast cancer, respectively. However, these negative results do not entirely exclude the potential role of polymorphisms of these genes in the aetiology of breast cancer, even though similar results were reported by previous studies. 22,23
In addition, although the results from our study indicated that SNPs of MnSOD1183T>C and GPx1Pro198Leu showed significant association with the risk of breast cancer, O2 ·−, if not scavenged by SOD, may react with nitric oxide radical (•NO) to form the more strongly oxidative peroxynitrite radical (ONOO−). A greater understanding of the functions of the MnSOD1183T>C and GPx1Pro198Leu polymorphisms and analyses in conjunction with other genetic polymorphisms of oxidative stress-related enzymes will help to elucidate fully the contributions of these genes to breast cancer. A study of the joint effect of the combined genetic polymorphisms of ROS defense enzymes including MnSOD, GPx1 and CAT as well as ROS-producing enzymes including MPO for the same individual as a second step might therefore provide further information on the role of oxidative stress in breast cancer risk. All the two-order gene–gene interactions were significantly associated with the risk of breast cancer development (Table 3), which suggests that the task of scavenging O2 ·− is undertaken by oxidative stress-related enzymes acting in concert. Furthermore, the gene–gene interaction analyses also indicate that MnSOD and GPx1 (and therefore O2 ·− scavenging) play an important role in breast cancer risk (Table 3), and these findings are consistent with the data presented in Table 2.
The limitations of this study should be noted: firstly, a possible co-modulator role of the GSH concentrations on the GPx1 polymorphism was not investigated, and the possible interaction with Se was not assessed, since neither GSH nor Se concentration data were available. However, the discrepancies regarding the role of GPx1 expression in relation to risk for breast cancer remain to be clarified. Secondly, the deviation from Hardy–Weinberg equilibrium suggests that this polymorphism might be under selective pressure. The genotype distributions of CAT-262C>T and GPx1Pro198Leu transitions were not in Hardy–Weinberg equilibrium among the cases and controls (data not shown). Nevertheless, the lack of the TT homozygote in our study might be the result of the small sample size or analysis of the population with stratification based on gender. Thirdly, we did not evaluate the interaction of genotypes and environmental factors such as diet, passive smoking and other possible lifestyle risk factors. A previous study suggested that the association between polymorphisms and breast cancer might be modified by the consumption of fruit and vegetables. 24 It is possible, therefore, that dietary intake might bias the estimation of risk associated with antioxidant or oxidant enzyme polymorphisms. However, in a case-control study design, it is extremely difficult to collect accurate retrospective dietary intake data going back two or more decades. Fourthly, although case-control studies have well documented weaknesses as compared with a longitudinal cohort studies, including measuring exposure (recall bias) and temporal relationships between cause and effect, these were not important factors in this study. However, the possibility of a selection bias in a small study, particularly when the control group was recruited from the same hospital rather than the community, might have distorted the strength of associations observed. A larger scale population-based case-control study is required to confirm our findings.
In conclusion, our results indicate that MnSOD and GPx1 might play significant roles in the development of breast cancer in Taiwanese women. We suggest that environmental risk factors other than cigarette smoking and alcohol consumption may be involved in breast cancer development. Larger studies are required to confirm these findings and to explore the relationship between specific environmental factors (such as diet and in particular passive smoking) with the polymorphisms of oxidative stress-related enzymes in breast cancer.
DECLARATIONS