Abstract
Background: Cigarette smoke contains carcinogenic aromatic and heterocyclic amines that are metabolized by N-acetyltransferase (NAT). These carcinogens also produce reactive oxygen species that are metabolized by manganese superoxide dismutase (MnSOD). The association between prostate cancer (PCA) and the polymorphism of MnSOD and NAT, and cigarette smoking was investigated. Patients and Methods: DNA samples from 187 PCA patients and 175 age-matched controls were genotyped for MnSOD, NAT1 and NAT2 by PCR restriction fragment length polymorphism analysis and DNA sequencing. Results: MnSOD AA genotype, as compared to MnSOD VV and VA, was associated with PCA (odds ratio, 1.65; 95% confidence interval, 1.03-2.66. There was no association of PCA with NAT or smoking. Results of exploratory analyses of the data suggest that the association of PCA and MnSOD exists only in the subpopulation of rapid NAT1 genotypes and smokers. Conclusion: The present study demonstrates the association of PCA and MnSOD. Oxidative stress and cigarette smoking may play an important role in the carcinogenesis of the prostate in those who have MnSOD AA and rapid NAT1 genotypes.
- Prostate cancer
- prostate-specific antigen
- manganese superoxide dismutase
- N-acetyltransferase
- smoking
- polymorphism
Abbreviations: MnSOD, manganese superoxide dismutase; NAT, N-acetyltransferase; PCA, prostate cancer; PSA, prostate-specific antigen; ROS, reactive oxygen species.
Prostate cancer (PCA) is the most prevalent non-skin malignancy in the US male and it is estimated that 1 in 6 males probably will develop invasive PCA in his lifetime (1). The etiology of PCA is multifactorial and it is suspected that environmental factors contribute to the majority of the cancers, including familial PCA (2). Chronic inflammation has been associated with the development of malignancy in several organs such as the esophagus, stomach, colon, liver and urinary bladder (3). Epidemiological data have correlated PCA with prostatitis and sexually transmitted diseases (4). Inflammation invariably produces reactive oxygen species (ROS), which can damage both mitochondrial and nuclear DNA (5, 6). Oxidative damage to nuclear DNA may result in the arrest or induction of transcription, induction of signal transduction pathways, replication errors or genomic instability, all of which may result in carcinogenesis (7, 8).
Approximately 2% of oxygen in the body is converted to superoxide radicals and its reactive metabolites in the mitochondria (9). Manganese superoxide dismutase (MnSOD) is present in the mitochondria and plays an important role in protection from ROS-mediated DNA damage. MnSOD converts superoxide radical to oxygen and hydrogen peroxide (10). The human MnSOD gene has a polymorphism at codon 16, which encodes for either alanine (A) or valine (V) (11). This polymorphism has been reported to be a risk factor for several malignancies. The A allele has been reported to be a risk factor for breast cancer (12, 13). It was further observed that breast cancer risk was elevated in premenopausal women who had the AA genotype with low consumption of dietary antioxidants (13). The risk was also increased in women with the AA genotype who smoked (14). A cohort study of Finnish male smokers suggested that men with the AA genotype had a 1.70-fold greater risk for PCA as compared to those with other MnSOD genotypes (15). However, this finding was not confirmed (16). Furthermore, it was reported that the AA genotype was a risk factor only for individuals with aggressive PCA who had low serum antioxidant levels, but not for the entire PCA cohort (17). Therefore, further investigation is needed to elucidate this association.
Cigarette smoking is a well-known risk factor for a variety of cancers, but there is no convincing evidence that it is associated with PCA (18). N-Acetyltransferases (NAT) are involved in the metabolic activation of carcinogenic aromatic and heterocyclic amines present in cigarette smoke (19). NAT1 and NAT2 genotypes are polymorphic in humans and they are generally described as rapid or slow acetylators (20, 21). Polymorphism of NAT as a risk factor for PCA has been under investigation (22). Cigarette smoke also produces ROS (23). In the present study, we investigated the association of the MnSOD genotype and prostate cancer risk as a function of NAT genotype and smoking status.
Patients and Methods
Study participants. The study was approved by the Institutional Review Board, SUNY Upstate Medical University. Patients were recruited between 1998 and 2001 from the Urology Clinic of University Hospital of SUNY Upstate Medical University, Syracuse, NY. The clinic serves the population of the rural central New York and the Syracuse area. Patients were either recently diagnosed with PCA or had PCA and were under follow-up or treatment. Age-matched controls were recruited from healthy males seeking annual urological examinations and from prostate cancer-screening programs. Individuals with normal digital rectal examination and prostate-specific antigen (PSA) <4 ng/ml were considered for inclusion as controls. Participants were questioned as to how many cigarettes and for how long they had smoked. Those who had smoked more than 100 cigarettes in their lifetime were categorized as smokers (24).
Genotyping. DNA was extracted from buccal swabs using QIAamp DNA Blood Mini Kits (Qiagen, Valencia, CA, USA) according to the supplier's instructions. MnSOD was PCR amplified with a primer pair (5′-GTGCTTTCTCGTCTTCAGCA-3′ and 5-AACGCCTCCTGGTACTTCTCC-3′) for 30 thermocycles: 94°C for 30 s, 55°C for 30 s, 72°C for 30 s. PCR products were digested with BsaWI (60°C, 1 h; New England BioLabs, Beverly, MA, USA). Digested products were visualized on a 2.5% agarose gel stained with ethidium bromide. Fragment patterns specific for three MnSOD genotypes were VV (GTT; 176 bp, 37 bp), VA (GTT/GCT; 213 bp, 176 bp, 37 bp) and AA (GCT; 213 bp).
NAT1 was PCR amplified from codon 124 through the polyadenylation region using a primer pair (5′-TGGGTTTGG ACGCTCATA-3′ and 5′-TCACCAATTTCCAAGATAAC-3′) with the following thermal program: 94°C for 1 min, 60°C for 2 min, 72°C for 1 min for 40 cycles. The amplicon was sequenced using primers (5′-GTGGCAGCCTCTGGAGT-3′ and 5′-GACTCT GAGTGAGGAAGAAATA-3′) with a Perkin Elmer ABI 310 automatic DNA sequencer. Genotypes were assigned according to the DNA sequences for the known NAT1 alleles reported by Deitz et al. (25). Eight NAT1 acetylator alleles (NAT1*3, NAT1*4, NAT1*10, NAT1*11, NAT1*14, NAT1*15, NAT1*16, and NAT1*18) and 14 genotypes were identified. Genes which contained at least one *10 allele were categorized as rapid NAT1 (NAT1R) genotypes and others were categorized as slow NAT1 (NAT1S) genotypes.
The entire 873 base-coding region of NAT2 was PCR amplified using a primer pair (5′-GATCATGGACATTGAAGCATATT-3′ and 5′-ATACATACACAAGGGTTTATTTTGT-3′) with the same thermal program. NAT2 was sequenced using nested primers (N109F and N440F; 5′-TTTGAACCTTAACATGCAT-3′ and 5′-CTTGCATTTTCTGCTTGACA-3′). Nested primer N109F gave the NAT2 sequence starting from base +109 and extending well past base +440. Likewise, nested primer N440F gave the sequence from the base +440 of NAT2 and continued into the polyadenylation region. There were 12 discernable individual NAT2 alleles (NAT2*4, NAT2*4A, NAT2*5A, NAT2*5B, NAT2*5C, NAT2*6A, NAT2*6B, NAT2*7B, NAT2*12A, NAT2*12B, NAT2*13 and NAT2*14B) found in this study. Analysis of the two sequenced portions of NAT2 genomic DNA showed it was not possible to differentiate between allele combinations *4/*5B and *5A/*12A, *4/*6A and *6B/*13, *4/*7B and *7A/*13, *4/*12B and *12A/*13, *5A/*12B and *5B/*13, and *6A/*12A and *6B/*12B. However they all are rapid genotypes as alleles *4, *12A, *12B and *13 give rapid NAT2 (NAT2R) phenotypes.
The laboratory personnel were unaware of the case-control status. To assess genotyping reproducibility, 25% of the samples were randomly selected for repeated determination and all genotypes were as initially determined.
Statistical analysis. Unconditional logistic regression models were used for statistical analyses using the software package JMP version 3.2.1, SAS Institute, Inc. (Cary, NC, USA). The 95% confidence interval (95% CI) was determined at the level of α=0.05 for type I error. The Pearson's chi square test or Fisher exact test was applied when the number in each sub-population was too small for logistic regression analysis after the population was stratified according to NAT genotype and smoking status. Likelihood ratio statistics were used when applicable. The analysis was adjusted for age, race and smoking status if possible; race and smoking status were treated as category variables and age as a numerical variable. Pearson's chi square test was applied to the Hardy-Weinberg equilibrium.
Results
This study consisted of 175 controls and 187 PCA patients whose genomic DNA was successfully genotyped for MnSOD. There were 145 Caucasians, 27 African-Americans and 3 other races in the control group and 162, 22 and 3, respectively, in the cancer group. The age was 62.3±9.9 and 63.3±8.0 years (mean±S.D.) for the control and cancer groups, respectively (p=0.842). There were 103 (61%) and 97 (52%) smokers in the control and cancer groups, respectively (p=0.182).
The genotype frequencies of MnSOD for VV, VA and AA were 0.229, 0.549 and 0.223 in the control group, and 0.219, 0.460 and 0.321 in the PCA group, respectively (Table I). The distribution of the genotypes in both the control group (p=0.438) or the cancer group (p=0.627) were consistent with the Hardy-Weinberg equilibrium. The observed numbers of these genotypes in the cancer group and the expected numbers of these genotypes derived from the observed frequencies in the control group were significantly different (p<0.01). As shown in Table I, the PCA group had a greater frequency of MnSOD AA genotype than did the control group.
Of the samples that were successfully genotyped for MnSOD, 170 control samples were successfully genotyped for both NAT1 and NAT2, and 179 and 180 PCA samples were successfully genotyped for NAT1 and NAT2, respectively. The results of logistic regression analysis suggested that neither NAT1 nor NAT2 was associated with PCA (Table I). Additionally, smoking did not appear to be a risk factor for PCA (Table I).
The distribution of MnSOD genotypes was stratified according to NAT1 genotype. There was an association between MnSOD AA and PCA in the rapid NAT1 group (odds ratio, OR=2.6; Table II). In contrast to the rapid NAT1, MnSOD AA was not associated with PCA in the slow NAT1 group (Table II). When stratified according to NAT2 genotype, the association between MnSOD and PCA was not significant (Table II).
The distribution of MnSOD was stratified according to smoking status and it appeared that the MnSOD AA genotype was a risk factor for PCA in smokers but not in nonsmokers (Table III). The OR was greatly increased in smokers who also had a rapid NAT1 genotype (Table IV). It appeared that the association between MnSOD AA and PCA was present in the subpopulation with NAT1R genotype who were also smokers. However, there was no interaction between MnSOD, NAT1 and smoking.
Discussion
The association between MnSOD and PCA has been investigated. Woodson et al. (15) suggested the MnSOD AA genotype was associated with PCA in smokers, especially in patients having a high-grade tumor. Li et al. (17) suggested that the MnSOD AA genotype was not associated with all PCA, but was associated with PCA in those patients who had low serum levels of antioxidants, such as selenium, lycopene and α-tocopherol. Their results suggest that those who with the MnSOD AA genotype are at risk for PCA under a condition of high ROS or low blood level of antioxidants. However, Choi et al. (16) suggested that there was no association between the MnSOD AA genotype and PCA, and smoking status did not modify associations. In the present study, we found an association between PCA and the MnSOD AA genotype, particularly in men who were smokers and/or had rapid NAT1 genotypes.
Although the function of the MnSOD polymorphism is not well understood, Sutton et al. (26) showed that the A-containing MnSOD is transported more efficiently through the mitochondrial membrane. Therefore, individuals with the AA genotype may have higher MnSOD activity as compared to those with other genotypes. In mitochondria, superoxide radical is converted by MnSOD into oxygen and hydrogen peroxide, which is further detoxified into water by glutathione peroxidase. The rate of hydrogen peroxide decomposition is proportional to both the glutathione level and the activity of glutathione peroxidase. Under physiological conditions, the ability to decompose hydrogen peroxide is thousands of times greater than is the generation of superoxide radical (27). However, at a high concentration of peroxide, the step of NADP reduction becomes rate-limiting, and the overall reaction rate of the detoxification of peroxide decreases by 100 times (27). Thus, high activity of MnSOD may lead to metabolic imbalance and induce toxicity if the rate of hydrogen peroxide decomposition decreases. Furthermore, cigarette smoke can produce superoxide radical (23), which is involved in the redox-cycling of Fenton reagents such as ferric/ferrous ions (28, 29). The regeneration of oxidized Fenton reagents provides continuous conversion of hydrogen peroxide to hydroxyl radical that can modify DNA.
It has been suspected that heterocyclic amines that are present in cooked meat and metabolized by NAT may contribute to the carcinogenesis of prostate (19, 30, 31). Therefore, several attempts have been made to correlate NAT with PCA. It has been reported that NAT1R is a risk factor for PCA (22, 30), but the results from another study do not support this conclusion (32). The NAT2S genotype has also been reported to be a risk factor, particularly for high grade PCA (33), but other studies have not shown the NAT2 genotype to be a risk factor for PCA (32, 34). We found that those individuals with NAT1S and NAT2S had a slightly increased risk for PCA but this was not statistically significant.
Despite numerous studies, cigarette smoking has not been conclusively shown to be a risk factor for PCA (18). We did not find an association between smoking and PCA either. However, we did find an association between MnSOD and PCA in smokers but not in nonsmokers. Furthermore, in the NAT1R/smoker group, those who with the MnSOD AA genotype had a 4-fold increased risk of PCA than those with other MnSOD genotypes (Table IV). That smoking increased the risk for PCA in individuals with NAT1R genotypes is consistent with the potential for acetyltransferase-mediated transformations of tobacco-derived carcinogenic aromatic and heterocyclic amines to metabolites capable of modifying DNA in the prostate (19). This is novel and important information, which may explain how cigarette smoking-derived carcinogens in urine may act directly by causing genetic damage and indirectly by causing inflammation that results in oxidative stress. Therefore, individuals with genetic variations which result in less efficient detoxification of carcinogens and handling of superoxide radicals would be at the greatest risk for developing PCA.
Our study has several limitations. Controls did not undergo biopsies to exclude PCA. This is an important aspect of verification bias that is inherent in most case-control studies, because even a negative prostate biopsy cannot definitively rule out occult PCA. When autopsied prostates were evaluated for cancer based on PSA <4 ng/ml, negative biopsy or both criteria, the occult cancer rates were 22%, 15% and 12%, respectively (35). Moreover, there was an association between PCA and the polymorphism of MnSOD in the autopsy cases where the control group was verified to be free of occult PCA (35). These data suggest that the contamination of occult cancer in the control group can result in false-negative results (36). The level of statistical significance of the present study may be diminished by the inclusion of as many as 22% occult cancer cases in the control group.
Our subpopulation of men who were smokers with NAT1R and MnSOD AA genotypes was a stratification of our larger study population, which was the basis of our original sample size calculation. Nevertheless, these observations provide important insight into the possible mechanism of PCA development which may be tested in larger cohorts.
In conclusion, our results suggest that the MnSOD AA genotype is a risk factor for PCA, particularly in smokers and in those who have rapid NAT1 genotypes. The risk is greatly increased in those who have both factors. MnSOD is involved in the metabolism of ROS, therefore, conditions that elicit inflammation and ROS may contribute to the carcinogenic process. Our data may help elucidate why studies focusing on the individual role of smoking, NAT and MnSOD may be inconclusive in their association with PCA, while our study investigating a combination of these risk factors was able to identify a subpopulation of individuals at an increased risk.
Acknowledgements
This study was supported in part by grants from the National Cancer Institute (CA097751; G. P. Haas) and the National Institute on Aging (AG021389; G. P. Haas).
- Received September 3, 2008.
- Revision received December 1, 2008.
- Accepted December 2, 2008.
- Copyright © 2009 International Institute of Anticancer Research (Dr. John G. Delinassios), All rights reserved