|Home | About | Journals | Submit | Contact Us | Français|
Nitric oxide (NO) induces cytotoxicity and angiogenesis, and may play a role in prostate carcinogenesis, potentially modulated by environmental exposures. We evaluated the association of prostate cancer with genetic polymorphisms in two genes related to intracellular NO: NOS2A [inducible nitric oxide synthase (NOS); −2892T>C, Ex16 + 14C>T (S608L), IVS16 + 88T>G and IVS20 + 524G>A] and NOS3 [endothelial NOS; IVS1 − 762C>T, Ex7 − 43C>T (D258D), IVS7 − 26A>G, Ex8 − 63G>T (E298D) and IVS15 − 62G>T]. Prostate cancer cases (n=1320) from the screening arm of the Prostate, Lung, Colorectal and Ovarian Cancer Screening Trial were frequency matched to controls (n=1842), by age, race, time since initial screening and year of blood draw. An antioxidant score [range 3–12; low (3–7) versus high (8–12)] was created by summing the quartile levels of vitamin E, β-carotene and lycopene, which were coded from 1 to 4, respectively. The global tests for all eight single-nucleotide polymorphisms (SNPs) (excluding NOS2A −2892T>C, with low minor allele frequency) were statistically significant for prostate cancer (P=0.005), especially for aggressive cancer (stage III–IV or Gleason score≥7) (P=0.01). The NOS2A IVS16+88 GT/TT was associated with increased prostate caner risk (odds ratio=1.24, 95% confidence interval=1.00–1.54), whereas the IVS20+524 AG/GG was associated with decreased risk (0.77, 0.66–0.90). The NOS3 IVS7 − 26GG was associated with increased prostate caner risk (1.33, 1.07–1.64). All these SNPs showed significant associations with aggressive cancer and not for non-aggressive cancer. In the evaluation of effect modification, the effect of the NOS2A IVS16+88 GT/TT on aggressive cancer was stronger among subjects with higher antioxidant intake (1.61, 1.18–2.19; Pinteraction=0.01). Our results suggest that NOS gene polymorphisms are genetic susceptibility factors for aggressive prostate cancer.
Age, race and family history of prostate cancer are well-established risk factors for prostate cancer (1); recently, several genetic risk factors for this disease have been identified (2–5), spurring renewed interest in the investigation of common genetic variants as prostate cancer risk factors.
Nitric oxide (NO) is a multifaceted compound that may inhibit carcinogenesis through cell death or promote carcinogenesis through angiogenesis (6,7). Nitric oxide synthase (NOS) is a family of enzymes that generate NO from L-arginine and have been classified as calcium-dependent endothelial NOS (NOS3) and neuronal NOS (NOS1) and calcium-independent inducible NOS (NOS2). The expression of inducible NOS (gene: NOS2A) is increased in prostate carcinoma (8,9) and endothelial NOS (gene: NOS3) is reported to protect prostate cancer cells from apoptosis (10). Genetic polymorphisms in these genes may play an important role in prostate cancer development, as indicated by Medeiros et al. (11) who reported prostate cancer risk associations in a small case–control study (125 cases) with an NOS3 27 bp repeat polymorphism in intron 4.
To better evaluate the potential role of selected genetic polymorphisms of NOS2A and NOS3 in the development of prostate cancer, we conducted a large nested case–control study with >1000 cases in the Prostate, Lung, Colorectal and Ovarian Cancer Screening Trial.
Details of study setting and subjects have been described (12). In brief, men randomized to the screening arm of the Prostate, Lung, Colorectal and Ovarian Trial were eligible for this nested case–control study, if they had at least one valid screening for prostate cancer (prostate-specific antigen and/or digital rectum exams), before 1 October 2001 (the censor date for this analysis), completed the baseline risk factor questionnaire, provided a blood sample and signed the informed consent for etiologic studies of cancer (n=26975). All men were followed from their initial valid prostate cancer screen (prostate-specific antigen and/or digital rectum exams), to first occurrence of prostate cancer, loss-to-follow-up, death or 1 October 2001, whichever came first.
The eligible group included 1320 prostate cancer cases (1213 non-Hispanic Caucasians and 107 African-Americans). For comparison, we selected 1842 controls (1433 non-Hispanic Caucasians and 409 African-Americans) using risk-set sampling, frequency matched to cases by age (55–59, 60–64, 65–69 and 70–74), ethnicity (case–control ratio of 1.2:1 for Caucasians and 4:1 for African-Americans), time since initial screening (1 year time windows) and year of blood draw (1 year calendar periods).
At enrollment, all participants were asked to complete a questionnaire including age, ethnicity, education, occupation, current and past smoking behavior, history of cancer and other diseases, use of selected drugs, recent history of screening exams and prostate-related health factors. Usual dietary intake over the 12 months before enrollment was assessed with a 137 item food frequency questionnaire, which included an additional 14 questions about intake of vitamin and mineral supplements (The food frequency questionnaire is available at http://www.parplco.org and access can be requested by contacting parplcohelpdesk/at/westat.com). Daily dietary nutrient intake was calculated by multiplying the daily frequency of each consumed food item by the nutrient value of the sex-specific portion size, using the nutrient database from the US Department of Agriculture (13). Total vitamin and mineral intake was calculated by adding dietary and supplemental intake.
An antioxidant score (range 3–12, low to high) was created for each man by summing the quartile levels of vitamin E, β-carotene and lycopene, which were coded from 1 to 4, respectively. Based on this score, we categorized all men into two groups with antioxidant scores of 3–7 (lower antioxidant intake) and 8 to 12 (higher antioxidant intake).
Genotype analysis was performed at the National Cancer Institute Core Genotyping Facility (http://cgf.nci.nih.gov). All TaqMan® assays (Applied Biosystems, Foster City, CA) were optimized on the ABI 7900 HT detection system with 100% concordance with sequence analysis of 102 individuals listed on the SNP500Cancer website (ref. 14; http://snp500cancer.nci.nih.gov).
We selected nine single-nucleotide polymorphisms (SNPs) in NOS2A (GenBank ID: NM_000625, 17q11.2–q12) and NOS3 (GeneBank ID: NM_000603, 7q36) genes for analysis: NOS2A −2892T>C (rs9282799), Ex16 + 14C>T (S608L) (rs2297518), IVS16 + 88T>G (rs9282801), IVS20 + 524G>A (rs944722), NOS3 IVS1 − 762C>T (rs2853796), Ex7 − 43C>T (D258D) (rs1549758), IVS7 − 26A>G (rs1007311), Ex8 − 63G>T (E298D) (rs1799983) and IVS15 − 62G>T (rs2853796). These SNPs were selected based on minor allele frequencies >5%, laboratory evidence of function or prior association with human disease (11,15,16). Replicate samples from 45 study subjects (assayed three to seven times) were interspersed throughout the batches for all genotyping assays. The concordance rates for replicate samples were 99.6–100% for all assays.
All statistical procedures were conducted using SAS version 9.1.3 (SAS Institute, Cary, NC), unless otherwise indicated. We used conditional logistic regression models to estimate odds ratios (ORs) and 95% confidence intervals (95% CIs) for prostate cancer risk, conditioned on the matching factors (age, time since initial screening and year of blood draw).
Genotype data were analyzed with the homozygote of the common allele as the reference group. For each SNP, potential dominant effects were evaluated by combining homozygote and heterozygote variant carriers for comparison with the reference group. Trend tests were conducted by assigning the ordinal values 1, 2 and 3 to homozygous wild-type, heterozygous and homozygous variant genotypes, respectively, and by using these scores as a continuous variable in logistic regression model. We note that some trends were statistically significant even though clear dose–response per allele does not seem to exist. In those cases, the statistical significance for trend may be driven by the significant dominant or recessive effect. The NOS2A −2892T>C polymorphism was excluded from individual SNP tests as well as haplotype analysis described below, due to low variant allele frequency in both ethnic groups (T: 0% in Caucasian controls and 3.4% in African-American controls).
To assess the global significance of the association between polymorphisms in NOS genes and prostate cancer risk, we used the likelihood ratio χ2 statistic, comparing the logistic regression model that included all eight SNPs (excluding NOS2A −2892T>C, with low minor allele frequency) as main effects (i.e. variant allele-containing genotypes versus homozygotes of the common allele) against the null model that included none of the SNPs.
To explore differences by disease severity, cases were divided into aggressive (stage III–IV or Gleason score≥7) and non-aggressive (stage I–II and Gleason score < 7) and results presented separately (Table II). For African-Americans, only combined results are presented, as the samples size for the African-American subjects was too small for meaningful subgroup analyses. To explore potential effect modification by antioxidant intake (higher versus lower) or smoking status (ever versus never), we carried out stratified analyses and evaluated multiplicative interactions, by creating product terms. Statistical significance of multiplicative interactions was evaluated by comparison of the log-likelihood statistics between the main effect model and the joint effect model with product terms.
Haplotype analyses were conducted using the haplo.stats package (http://mayoresearch.mayo.edu/mayo/research/schaid_lab/software.cfm) in the R program version 2.2.1 (http://www.r-project.org), which uses an expectation–maximization algorithm to estimate haplotypes from genotype data (17). Haplotypes were estimated excluding subjects missing all the genotype data. The generalized linear model implemented in haplo.stats was used to estimate the effect of individual haplotypes by fitting an additive model, adjusting for age, time to diagnosis and year of blood draw. The overall difference in haplotype frequencies between cases and controls was assessed by race, using a global score test, adjusting for age, time to diagnosis and year of blood draw.
Characteristics of study subjects were reported in our previous study (12). All SNPs did not divert from Hardy–Weinberg equilibrium among controls. Global tests of all eight SNPs (excluding NOS2A −2892T>C) were statistically significant for prostate cancer (P=0.005), especially for aggressive cancer (stage III–IV or Gleason score≥7) (P=0.002). When stratified by race, the global test results were significant for both Caucasians (P=0.04), especially for aggressive cancer (P=0.01), and for African-Americans (P=0.02). Separate evaluation by genes showed that the test was significant for NOS2A only (P=0.0004 for all subjects).
For individual genotypes, the NOS2A IVS16 + 88 GT/TT genotype was associated with increased prostate caner risk among both Caucasians (OR=1.17, 95% CI=1.00–1.37) and African-Americans (OR=1.74, 95% CI=1.11–2.73), whereas the NOS2A IVS20 + 524 AG/GG genotype was associated with decreased risk among both ethnic groups (OR=0.82, 95% CI=0.69–0.96 and OR=0.50, 95% CI=0.31–0.81, respectively) (Table I); these significant associations were observed for only aggressive prostate cancer (stage III–IV or Gleason score≥7) (OR=1.26, 95% CI=1.03–1.55 and OR=0.71, 95% CI=0.58–0.87, respectively) (Table II). The NOS2A haplotype distribution was significantly different between aggressive cases and controls among Caucasians after adjustment for age, time since initial screening and year of blood draw (Pomnibus=0.03) (Table III).
Similar to the results of two SNPs in NOS2A, the NOS3 IVS7 − 26GG genotype was associated with increased prostate caner risk among both Caucasians (OR=1.28, 95% CI=1.02–1.61) and moderately associated with the increased risk among African-Americans (OR=1.85, 95% CI=0.93–3.68) (Table I). The effect of NOS3 IVS7 − 26GG was also observed for aggressive cancer (OR=1.44, 95% CI=1.07–1.95) and not for non-aggressvie cancer (Table II); however, no significant differences were observed for the other four SNPs in NOS3 or the NOS3 haplotypes between cases and controls (data not shown).
Stratified analysis by antioxidant vitamin intake showed that the NOS2A IVS16 + 88 GT/TT was significantly associated with aggressive prostate cancer among high vitamin intakers (OR=1.61, 95% CI=1.18–2.19) and not among low vitamin intakers (Pinteraction=0.01) (Table IV). No significant interactive effect was observed in stratified analysis by smoking status.
Our results indicate that NOS2A and NOS3 genetic polymorphisms are associated with prostate cancer risk in both Caucasians and African-Americans, and the associations were observed especially for aggressive cancer. Consistent with our results, Meideros et al. (11) reported the association of the IVS4 27 bp repeat polymorphism of NOS3 gene with high histological grade (Gleason≥7) of prostate cancer and the shedding of circulating tumor cells in the blood (18). These findings are supported by evidence that NO promotes the growth rate, vascular density and invasiveness of tumors (19) and has a role in maintaining tumor blood supply in patients with lung and prostate cancer (20). A recent study suggested that intratumoral inducible NOS activity favors development of prostate cancer cells that are able to proliferate androgen receptor independently, thereby promoting prostate tumor progression (21). Moreover, the association between genetic polymorphisms in NOS3 and invasive cancer has also been reported for ovary (16) and breast cancer (22).
We also show in an exploration of interrelationships of NOS with potential environmental factors that risk association for NOS2A IVS16 + 88T>G is influenced by antioxidant intake. This potential effect modification could be interpreted in terms of the role of NO in angiogenesis (19,20) and antioxidant function (23,24). Endogenously generated NO may play a role as a reactive oxygen species (ROS) scavenger protecting growing tumor-spheroids from ROS-induced apoptosis (10,24). Lower ROS levels were observed in invasive prostate cancer cells (PC3) compared with non-invasive ones (LNCaP) (25). Thus, the effect of NOS2A IVS16 + 88T>G on aggressive prostate cancer among those with higher antioxidant intake level might be due to decreased ROS-induced apoptosis or increased angiogenesis in prostate cancer cells.
About 72% of the subjects in our study were included in a recent genome-wide association study [Cancer Genetic Markers of Susceptibility (CGEMS)], including the characterization of >500000 SNPs. Among the nine SNPs evaluated in this study, only one [NOS2A Ex16 + 14C>T (S608L)] was evaluated in CGEMS and, consistent with our result, the NOS2A Ex16 + 14 T allele showed no significant association with prostate cancer (http://cgems.cancer.gov). Further, the NOS2A IVS20 + 524A>G in our study is in strong linkage disequilibrium (r2=0.96) with another CGEMS SNP (rs2274894) (http://www.hapmap.org); although the association between rs2274894 and aggressive prostate cancer had only marginal statistical significance (P=0.10) in CGEMS, the direction of the effect was consistent with our study. The other two SNPs with significant results in our study (NOS2A IVS16 + 88 GT/TT and NOS3 IVS7 − 26GG) could not be completely evaluated for linkage disequilibrium with the SNPs of CGEMS because they were not included in HapMap.
Our study was limited in the number of SNPs that were genotyped, and thus our analysis, including haplotype analysis for NOS3 gene, was not comprehensive and warrants cautious interpretation. Also, none of the three SNPs that showed significant association with prostate cancer in our study has known functional significance. Data for African-Americans were limited because of small sample size and conclusions for this group are limited. On the other hand, as strengths, our study was of a nested case–control study design and had a large number of Caucasian subjects that enabled us to do subgroup analysis by tumor aggressiveness.
In conclusion, our results suggest that polymorphisms of NOS genes are genetic susceptibility factors for prostate cancer, especially for aggressive disease. Potential interactive effects of NOS gene polymorphisms with antioxidant vitamin intake warrant further studies.
Intramural Research Program of the Division of Cancer Epidemiology and Genetics and contracts from the Division of Cancer Prevention, National Cancer Institute, National Institutes of Health, Department of Health and Human Services.
The authors thank Drs. Christine Berg and Philip Prorok, Division of Cancer Prevention, National Cancer Institute; the Screening Center investigators and staff of the Prostate, Lung, Colorectal and Ovarian Cancer Screening Trial; Mr. Tom Riley and staff, Information Management Services and Ms. Barbara O'Brien and staff, Westat. Most importantly, we acknowledge the study participants for their contributions to making this study possible.