Abstract
The estrogen metabolite, 4-hydroxy-estrogen, has been shown to play a role in malignant transformation of male kidneys. To counteract the effects of this catechol-estrogen, the catechol-O-methyltransferase (COMT) enzyme is capable of neutralizing the genotoxic effects of this compound. A polymorphic variant of COMT has been shown to have a reduced enzyme activity, and thus, we hypothesize that single nucleotide polymorphisms of the COMT gene can be a risk factor for renal cell cancer (RCC). To determine this hypothesis, a study of a Japanese male population was used and the genetic distributions of COMT polymorphisms at codons 62 (C→T), 72 (G→T), and 158 (G→A) were analyzed in 157 normal healthy subjects and 123 sporadic RCC (clear cell type) samples by using a sequence-specific PCR technique. These experiments show that the variant genotype (P = 0.025) and allele (P = 0.011) at codon 62 is a risk factor for RCC. The odds ratio and 95% confidence interval for cancer were 3.16 and 1.29 to 7.73, respectively, for the T/T genotype as compared with wild-type. No associations for renal cancer were found at either codons 72 or 158 in this Japanese male population. However, codons 62 and 158 were observed to be in linkage disequilibrium, and haplotype analysis shows the combined forms of T-A, T-G, and C-A to be associated with RCC as compared with C-G (P < 0.001). When evaluating the risk of COMT polymorphisms with grade of cancer, no associations were observed for any of the genotypes. This study is the first to report COMT polymorphism to be associated with RCC. These results are important in understanding the role of COMT polymorphisms in the pathogenesis of RCC. (Cancer Epidemiol Biomarkers Prev 2007;16(1):92–7)
Introduction
The global annual incidence rate of kidney cancer is estimated to be >200,000 new cases with a death rate of >100,000 persons (1). In the U.S., the estimated incidence rate of this cancer is >38,000 cases with a death rate of >12,000 per year (2). When compared with all adult cancers, malignancies of the kidney account for nearly 3% (2). Interestingly, gender differences occur because men in the U.S. account for over two-thirds of both the incidence (>24,000 cases) and death (>8,000 cases) rates (2). Among the various forms of kidney cancers that develop in adults, renal cell carcinoma (RCC) is the most common, accounting for ∼85% of cases (3). Furthermore, RCC consists of several types, of which the clear cell form comprises the majority (3). Despite the incidence and death rates, the genetic basis of this cancer is not fully understood.
A factor that has been shown to cause cancer is estrogen. Although the classical role of estrogen in cells is to bind to its receptors and produce their biological effects, when given exogenously, estrogens have been shown to induce or promote tumor formation at various locations in animals (4-6). In the Syrian male hamster, treatment with estrogen leads to the induction of kidney tumors (7), and additionally, estradiol-17β has been shown to cause multiple bilateral renal tumors with a 100% incidence rate (7, 8). Interestingly, the prevention of these estrogen-induced kidney tumors can be achieved by the concomitant administration of estrogen antagonists (9). Thus, these data show renal carcinogenesis to be an estrogen-related process.
Much of the tumorigenic effects of estrogens are believed to be through its normal metabolic products and includes 4-hydroxy-estrogen (8, 10). This catechol-estrogen is produced via hydroxylation of the parent estrogen by the cytochrome P450 (CYP) 1B1 enzyme (11), which has been shown to be expressed in human kidney cells (12). Indeed, when 4-hydroxy-estradiol was injected into animals, tumors at locations such as the uterus (13) and kidneys (8, 10) were formed. 4-Hydroxy-estrogen undergoes further metabolic reactions to form other reactive compounds such as estradiol-3,4-quinone and semiquinones (14), and mechanistically, treatment of animals with these metabolites has been shown to cause DNA single-strand breakage (15), mutation (16), and depurination (16). Thus, these observations show that 4-hydroxy-estrogen and its associated metabolites play a causative role in the process of carcinogenesis.
To counter the effects of these naturally produced but carcinogenic catechol-estrogens and its metabolic products, enzymes are present in cells that can detoxify these compounds. A key enzyme that can react with and nullify the detrimental effects of 4-hydroxy-estrogen is catechol-O-methyltransferase (COMT). This enzyme attaches a methyl group from the coenzyme S-adenosyl-l-methionine to the hydroxyl group at either the 2-, 3-, or 4- positions of the catechol ring (17), and thus, forms a methoxy compound (18). Consequentially, the formation of highly reactive estradiol-quinones and semiquinones that are damaging to DNA (19) or toxic to cells (20) are prevented. Additionally, these methoxyestrogens are biologically inactive because they have little or no affinity for estrogen receptors, as well as no estrogenic effects on target tissues (21). Thus, the COMT enzyme is shown to play a protective role in the body.
Polymorphisms and mutations in genes have been shown to cause changes in the function of the resultant mutated enzymes. In the case of CYP2A6 and CYP2D6, the altered base causes a reduction in enzyme function (22, 23). On the contrary, the variant form of the protein causes the enhanced activity of CYP1A1 and CYP1B1 enzymes (24, 25). Polymorphisms of various genes have also been associated with numerous cancers (26, 27). Thus, polymorphisms of COMT may be a risk factor for RCC.
There are many polymorphisms that have been identified in the COMT gene. Several have been characterized in the coding region and include codons 62 (C→T), 72 (G→T), and 158 (G→A; Fig. 1A; refs. 28, 29). Variants at two of these codons lead to altered amino acids which are codon 72 (Ala→Ser) and 158 (Val→Met). Studies show that the Met variant at codon 158 of COMT results in an enzyme activity that is reduced by up to four times compared with the wild-type (28), and recently, the codon 72 variant (Ser) was shown to have a reduced activity as well (30). The third polymorphism at codon 62 is silent (His). Although the base change on this codon does not lead to an amino acid change, studies show that synonymous mutations could lead to structural changes in mRNA (31-33), and as a result, causes altered protein expression (32) as well as disease (33). Furthermore, polymorphisms of COMT have been found to have associations with neurologic/psychological diseases (34, 35) and have also been investigated in various cancers such as breast (36), bladder (37), liver (38), endometrial (39), and prostate (40) cancers.
A. Schematic representation of the COMT gene structure. Three polymorphic sites located at codons 62 (C→T), 72 (G→T), and 158 (G→A). The polymorphism at codons 62 and 72 are located in exon 3 and codon 158 is in exon 4. Corresponding amino acid changes due to the polymorphism are listed with codon 62 being silent (adapted from Tanaka et al.; ref. 40). B. Representative gels displaying polymorphic genotypes of COMT for codons 62, 72, and 158. Top gels (WT), wild-type of each polymorphic locus; bottom gels (VT), variant type of each polymorphic locus. Top band within gels, the first PCR product; lower bands within gels, the SSP product. Lane 1, 100 bp ladder marker; lane 2, homozygous wild-type; lane 3, heterozygous type; lane 4, homozygous variant genotype.
A. Schematic representation of the COMT gene structure. Three polymorphic sites located at codons 62 (C→T), 72 (G→T), and 158 (G→A). The polymorphism at codons 62 and 72 are located in exon 3 and codon 158 is in exon 4. Corresponding amino acid changes due to the polymorphism are listed with codon 62 being silent (adapted from Tanaka et al.; ref. 40). B. Representative gels displaying polymorphic genotypes of COMT for codons 62, 72, and 158. Top gels (WT), wild-type of each polymorphic locus; bottom gels (VT), variant type of each polymorphic locus. Top band within gels, the first PCR product; lower bands within gels, the SSP product. Lane 1, 100 bp ladder marker; lane 2, homozygous wild-type; lane 3, heterozygous type; lane 4, homozygous variant genotype.
Thus, COMT polymorphisms have been studied in various diseases. However, to date, there has been no study investigating the role of polymorphism of this gene in RCC. Because polymorphisms are inherited, interindividual differences in disease risk may occur and include renal cancer. Therefore, in the present study, the risk of COMT polymorphisms with RCC was determined.
Materials and Methods
Subjects
A total of 123 male patients with primary RCC were recruited from the hospital at Shimane Medical University (Izumo, Japan) and Toho University Hospital (Tokyo, Japan). All specimens were of the clear cell carcinoma type. RCC specimens were collected by radical nephrectomy during the period from 1995 to 2003. The average age of cancer patients was 61.0 ± 1.3 years. Samples were pathologically characterized in terms of their grade (General Rules for Clinical and Pathological Studies on Renal Cancer by the Japanese Urological Association and the Japanese Pathological Society) and numbers in each category were as follows: grade 1, 44 patients; grade 2, 64 patients; and grade 3, 13 patients; with two patients of unknown status. To assess genotyping, control blood samples from 157 male Japanese volunteers were also obtained. These samples were collected during the same period as the radical nephrectomy operations. Volunteers were closely matched with patients with RCC and had a mean age of 61.7 ± 1.3 years. To ascertain that volunteers were healthy and free of cancer, they all underwent various tests that included physical exams, questionnaires about their health and history, chest X-rays, blood and urine tests for various tumor markers, abdominal ultrasound, gastric endoscopy, and colon enema. Based on doctor examinations and past history, all were confirmed to be free of cancer. Thus, all patients with RCC and healthy volunteers were of the same race (Japanese), sex (male), and age-matched, and were selected at random during the sample collection period. Informed consent from all subjects was obtained and the study was approved by the Shimane Medical University and Toho University.
DNA Extraction
A DNA extraction kit (Qiagen, Valencia, CA) was used to extract DNA from all RCC and control samples. A spectrophotometer at 260 and 280 nm wavelengths was used to determine the quantity and quality of DNA.
Analyses of COMT Polymorphisms
To analyze COMT polymorphisms in samples, a two-step PCR procedure was designed. The PCR conditions for the three polymorphic sites studied have been previously described (40). In brief, for both the first and sequence-specific PCR (SSP), 10 ng of DNA was amplified in a 20 μL reaction containing 1.5 mmol/L of MgCl2, 0.8 mmol/L of deoxynucleotide triphosphate mix, PCR buffer, and 0.5 units of Red-Taq polymerase (Sigma-Aldrich, St. Louis, MO), along with primer sets as shown in Table 1.
Summary of first PCR and SSP primer sets along with annealing temperatures for COMT polymorphisms in exons 3 and 4
Exon 3 . | First PCR primer . | Sequence . | Annealing . | |||||
---|---|---|---|---|---|---|---|---|
E3-S | gctggaacgagttcatcctg | 55°C | ||||||
E3B-AS | ctgctcgcagtaggtgtcaa | |||||||
Codon | Allele | SSP Primer | Sequence | Annealing | ||||
62 | C | 62-C-S with E3B-AS | agcagcgcatcctgaaccac | 65°C | ||||
T | 62-T-S with E3B-AS | agcagcgcatcctgaaccat | ||||||
72 | G | 72-G-AS with E3-S | gcctccagcacgctctgtgc | 65°C | ||||
T | 72-T-AS with E3-S | gcctccagcacgctctgtga | ||||||
Exon 4 | First PCR primer | Sequence | Annealing | |||||
E4B-S | catcaccatcgagatcaacc | 55°C | ||||||
E4-AS | ccctttttccaggtctgaca | |||||||
Codon | Allele | SSP Primer | Sequence | Annealing | ||||
158 | G | 158-G-AS with E4B-S | gcatgcacaccttgtccttcac | 65°C | ||||
A | 158-A-AS with E4B-S | gcatgcacaccttgtccttcat |
Exon 3 . | First PCR primer . | Sequence . | Annealing . | |||||
---|---|---|---|---|---|---|---|---|
E3-S | gctggaacgagttcatcctg | 55°C | ||||||
E3B-AS | ctgctcgcagtaggtgtcaa | |||||||
Codon | Allele | SSP Primer | Sequence | Annealing | ||||
62 | C | 62-C-S with E3B-AS | agcagcgcatcctgaaccac | 65°C | ||||
T | 62-T-S with E3B-AS | agcagcgcatcctgaaccat | ||||||
72 | G | 72-G-AS with E3-S | gcctccagcacgctctgtgc | 65°C | ||||
T | 72-T-AS with E3-S | gcctccagcacgctctgtga | ||||||
Exon 4 | First PCR primer | Sequence | Annealing | |||||
E4B-S | catcaccatcgagatcaacc | 55°C | ||||||
E4-AS | ccctttttccaggtctgaca | |||||||
Codon | Allele | SSP Primer | Sequence | Annealing | ||||
158 | G | 158-G-AS with E4B-S | gcatgcacaccttgtccttcac | 65°C | ||||
A | 158-A-AS with E4B-S | gcatgcacaccttgtccttcat |
Abbreviation: S, sense; AS, antisense.
Gel Electrophoresis
Products of SSP were separated electrophoretically on 3% agarose gels using 200 V at ambient temperature. After staining with ethidium bromide, products were then visualized on gels under UV light.
DNA Sequencing
Products of the first PCR were subjected to direct DNA sequencing to confirm genotyping. From gels, DNA was purified using a PCR purification kit (Qiagen). The ABI 377 Sequencer and Dye Terminator Cycle sequencing kit (Applied Biosystems Inc., Foster City, CA) was then used to analyze the sequence of purified DNA products by using first PCR primers. DNA sequence was confirmed on at least three representative samples for each of the polymorphic types.
Statistical Analyses
The frequencies of each of the COMT polymorphic genotypes and allele types for both healthy controls and RCC along with grades of cancer were determined and tabulated. To determine differences in the genotypic and allelic distributions between groups, χ2 analysis was used. For each polymorphic site, odds ratios along with 95% confidence intervals were also determined for each genotype as compared with wild-type. Linkage disequilibrium between polymorphic sites were measured in healthy control samples and haplotype frequency differences between RCC and controls were calculated using the SNPAlyze version 5.1 software (Dynacom, Tokyo, Japan).
Results
The structure of the COMT gene is shown in Fig. 1A, with polymorphic sites located in exons 3 and 4. Representative gels displaying the polymorphic genotypes at codons 62 (C→T), 72 (G→T), and 158 (G→A) are shown in Fig. 1B. For each locus, the upper gels show wild-type and lower gels show the variant type. Within gels, lanes 1 to 4 show the 100 bp DNA ladder marker, homozygous wild-type, heterozygous, and homozygous variant genotypes, respectively. The upper band within gels is the first PCR product and the lower band shows the SSP products. SSP products were confirmed by DNA sequencing (data not shown).
Table 2A and B show the genotypic and allelic frequencies, respectively, for codons 62, 72, and 158 of the COMT gene in men that are healthy and in men with RCC. The frequencies of all polymorphic sites in healthy controls are in agreement with the Hardy-Weinberg equilibrium. Interestingly, in this Japanese population, the variant (T/T) at codon 62 is significantly higher in patients with RCC compared with controls because more than twice the number of patients were observed to have the variant (12.2% versus 5.1% for RCC versus controls, respectively; χ2, P = 0.025). The odds ratios for T/T compared with wild-type C/C was 3.16 with a 95% confidence interval of 1.29 to 7.73. Likewise, the variant allele T was also observed to be significantly associated with RCC as 35% of cancer patients had the variant compared with 25% of healthy controls (χ2, P = 0.011). No differences were found between patients with RCC and controls at codon 72. Interestingly, the polymorphism at this codon is relatively rare, with only 1% of total samples (3 of 280) displaying the T/T variant genotype. The T allele frequency was just 9.2% and 7.3% for controls and patients with RCC, respectively. Also, no association for RCC was observed for the codon 158 polymorphic site, with the variant A/A genotype being observed in 8.1% of patients with RCC compared with 7.0% of controls.
The genotypic (A) and allelic (B) frequencies for codons 62, 72, and 158 of the COMT gene in men with and without RCC
(A) . | . | . | . | |||
---|---|---|---|---|---|---|
Genotype . | Controls (n = 157) . | RCC (n = 123) . | Odds ratio (95% confidence intervals) . | |||
Codon 62 | ||||||
C/C | 86 (54.8%) | 51 (41.5%) | 1.00 (reference) | |||
C/T | 63 (40.1%) | 57 (46.3%) | 1.53 (0.93-2.50) | |||
T/T | 8 (5.1%) | 15 (12.2%) | 3.16 (1.29-7.73) | |||
P = 0.025 | ||||||
Codon 72 | ||||||
G/G | 130 (82.8%) | 106 (86.2%) | 1.00 (reference) | |||
G/T | 25 (15.9%) | 16 (13.0%) | 0.78 (0.40-1.53) | |||
T/T | 2 (1.3%) | 1 (0.8%) | 0.61 (0.06-6.03) | |||
Codon 158 | ||||||
G/G | 85 (54.1%) | 59 (48.0%) | 1.00 (reference) | |||
G/A | 61 (38.8%) | 54 (43.9%) | 1.28 (0.78-2.09) | |||
A/A | 11 (7.0%) | 10 (8.1%) | 1.31 (0.52-3.30) | |||
(B) | ||||||
Allele | Controls (n = 314) | RCC (n = 246) | ||||
Codon 62 | ||||||
C | 235 (74.8%) | 159 (64.6%) | ||||
T | 79 (25.2%) | 87 (35.4%) | ||||
P = 0.011 | ||||||
Codon 72 | ||||||
G | 285 (90.8%) | 228 (92.7%) | ||||
T | 29 (9.2%) | 18 (7.3%) | ||||
Codon 158 | ||||||
G | 231 (73.6%) | 172 (69.9%) | ||||
A | 83 (26.4%) | 74 (30.1%) |
(A) . | . | . | . | |||
---|---|---|---|---|---|---|
Genotype . | Controls (n = 157) . | RCC (n = 123) . | Odds ratio (95% confidence intervals) . | |||
Codon 62 | ||||||
C/C | 86 (54.8%) | 51 (41.5%) | 1.00 (reference) | |||
C/T | 63 (40.1%) | 57 (46.3%) | 1.53 (0.93-2.50) | |||
T/T | 8 (5.1%) | 15 (12.2%) | 3.16 (1.29-7.73) | |||
P = 0.025 | ||||||
Codon 72 | ||||||
G/G | 130 (82.8%) | 106 (86.2%) | 1.00 (reference) | |||
G/T | 25 (15.9%) | 16 (13.0%) | 0.78 (0.40-1.53) | |||
T/T | 2 (1.3%) | 1 (0.8%) | 0.61 (0.06-6.03) | |||
Codon 158 | ||||||
G/G | 85 (54.1%) | 59 (48.0%) | 1.00 (reference) | |||
G/A | 61 (38.8%) | 54 (43.9%) | 1.28 (0.78-2.09) | |||
A/A | 11 (7.0%) | 10 (8.1%) | 1.31 (0.52-3.30) | |||
(B) | ||||||
Allele | Controls (n = 314) | RCC (n = 246) | ||||
Codon 62 | ||||||
C | 235 (74.8%) | 159 (64.6%) | ||||
T | 79 (25.2%) | 87 (35.4%) | ||||
P = 0.011 | ||||||
Codon 72 | ||||||
G | 285 (90.8%) | 228 (92.7%) | ||||
T | 29 (9.2%) | 18 (7.3%) | ||||
Codon 158 | ||||||
G | 231 (73.6%) | 172 (69.9%) | ||||
A | 83 (26.4%) | 74 (30.1%) |
NOTE: P values reflect χ2 test.
Linkage between the COMT polymorphic codons was calculated and Table 3 shows D values among healthy controls. Interestingly, codons 62 and 158 were observed to be in linkage disequilibrium with a D value of 0.1619. No linkage was observed between codon 72 with either of the two codons.
Linkage disequilibrium among three polymorphisms of COMT in normal healthy samples
Codon . | 62 . | 72 . | 158 . |
---|---|---|---|
62 | — | −0.0232 | 0.1619 |
72 | −0.0232 | — | −0.0244 |
158 | 0.1619 | −0.0244 | — |
Codon . | 62 . | 72 . | 158 . |
---|---|---|---|
62 | — | −0.0232 | 0.1619 |
72 | −0.0232 | — | −0.0244 |
158 | 0.1619 | −0.0244 | — |
NOTE: D values shown.
Haplotype frequencies between codons 62 and 158 were calculated and results are shown in Table 4. The 62C-158G haplotype is expressed in 71.2% in healthy subjects compared with 56.6% of patients with RCC. Interestingly, when compared with this C-G haplotype, the other types (T-A, T-G, and C-A) were observed to be significantly associated with RCC (P < 0.001).
Haplotype frequencies of codons 62 and 158 of COMT between healthy controls and patients with RCC
Haplotype . | Control (n = 314) . | RCC (n = 246) . |
---|---|---|
62C-158G | 224 (71.2%) | 139 (56.6%) |
62T-158A | 72 (22.8%) | 54 (22.0%) |
62T-158G | 7 (2.3%) | 33 (13.4%) |
62C-158A | 11 (3.6%) | 20 (8.1%) |
P < 0.001 |
Haplotype . | Control (n = 314) . | RCC (n = 246) . |
---|---|---|
62C-158G | 224 (71.2%) | 139 (56.6%) |
62T-158A | 72 (22.8%) | 54 (22.0%) |
62T-158G | 7 (2.3%) | 33 (13.4%) |
62C-158A | 11 (3.6%) | 20 (8.1%) |
P < 0.001 |
NOTE: P value is based on 62C-158G compared with other haplotypes combined.
RCC samples were classified in terms of pathologic grade. The frequency of the various genotypes and alleles at codons 62, 72, and 158 of COMT with respect to grade of cancer is shown in Table 5A and B, respectively, with two samples of unknown status. Because the number of patients with grade 3 cancer was small, they were combined with grade 2 cancers. No differences in genotypic and allelic frequencies were observed between grades 1 and ≥2 for all COMT polymorphic sites. The results of these experiments suggest that the COMT polymorphism may be a risk factor for RCC.
The genotypic (A) and allelic (B) frequencies of COMT at codons 62, 72, and 158 with pathologic grade (1 versus ≥2) in men with RCC
(A) . | . | . | . | |||
---|---|---|---|---|---|---|
Genotype . | Grade 1 (n = 44) . | Grade ≥2 (n = 77) . | Total (n = 121) . | |||
Codon 62 | ||||||
C/C | 18 (40.9%) | 32 (41.6%) | 50 (41.3%) | |||
C/T | 20 (45.4%) | 36 (46.8%) | 56 (46.3%) | |||
T/T | 6 (13.6%) | 9 (11.7%) | 15 (12.4%) | |||
Codon 72 | ||||||
G/G | 38 (86.4%) | 66 (85.7%) | 104 (86.0%) | |||
G/T | 6 (13.6%) | 10 (13.0%) | 16 (13.2%) | |||
T/T | 0 (0.0%) | 1 (1.3%) | 1 (0.8%) | |||
Codon 158 | ||||||
G/G | 23 (52.3%) | 35 (45.4%) | 58 (47.9%) | |||
G/A | 17 (38.6%) | 36 (46.8%) | 53 (43.8%) | |||
A/A | 4 (9.1%) | 6 (7.8%) | 10 (8.3%) | |||
(B) | ||||||
Allele | Grade 1 (n = 88) | Grade ≥2 (n = 154) | Total (n = 242) | |||
Codon 62 | ||||||
C | 56 (63.6%) | 100 (64.9%) | 156 (64.5%) | |||
T | 32 (36.4%) | 54 (35.1%) | 86 (35.5%) | |||
Codon 72 | ||||||
G | 82 (93.2%) | 142 (92.2) | 224 (92.6%) | |||
T | 6 (6.8%) | 12 (7.8%) | 18 (7.4%) | |||
Codon 158 | ||||||
G | 63 (71.6%) | 106 (68.8%) | 169 (69.8%) | |||
A | 25 (28.4%) | 48 (31.2%) | 73 (30.2%) |
(A) . | . | . | . | |||
---|---|---|---|---|---|---|
Genotype . | Grade 1 (n = 44) . | Grade ≥2 (n = 77) . | Total (n = 121) . | |||
Codon 62 | ||||||
C/C | 18 (40.9%) | 32 (41.6%) | 50 (41.3%) | |||
C/T | 20 (45.4%) | 36 (46.8%) | 56 (46.3%) | |||
T/T | 6 (13.6%) | 9 (11.7%) | 15 (12.4%) | |||
Codon 72 | ||||||
G/G | 38 (86.4%) | 66 (85.7%) | 104 (86.0%) | |||
G/T | 6 (13.6%) | 10 (13.0%) | 16 (13.2%) | |||
T/T | 0 (0.0%) | 1 (1.3%) | 1 (0.8%) | |||
Codon 158 | ||||||
G/G | 23 (52.3%) | 35 (45.4%) | 58 (47.9%) | |||
G/A | 17 (38.6%) | 36 (46.8%) | 53 (43.8%) | |||
A/A | 4 (9.1%) | 6 (7.8%) | 10 (8.3%) | |||
(B) | ||||||
Allele | Grade 1 (n = 88) | Grade ≥2 (n = 154) | Total (n = 242) | |||
Codon 62 | ||||||
C | 56 (63.6%) | 100 (64.9%) | 156 (64.5%) | |||
T | 32 (36.4%) | 54 (35.1%) | 86 (35.5%) | |||
Codon 72 | ||||||
G | 82 (93.2%) | 142 (92.2) | 224 (92.6%) | |||
T | 6 (6.8%) | 12 (7.8%) | 18 (7.4%) | |||
Codon 158 | ||||||
G | 63 (71.6%) | 106 (68.8%) | 169 (69.8%) | |||
A | 25 (28.4%) | 48 (31.2%) | 73 (30.2%) |
NOTE: Samples with unknown status (n = 2).
Discussion
Estrogens in the cells of the body undergo metabolic activation to carcinogenic forms via CYP1B1 (11). One such structure formed via CYP1B1 is 4-hydroxy-estrogen and this compound has been shown to be tumorigenic in males (8, 10). This catechol-estrogen undergoes further metabolic conversions to form semiquinones and quinones such as estradiol-3,4-quinone, which in turn, are also known to react with DNA (15, 16). Thus, in order to prevent mutations and damage to the cells of the body, detoxification of 4-hydroxy-estrogen is a key important step and the COMT enzyme is responsible for this defense.
Two types of COMT proteins occur in humans which are (a) a soluble form (S-COMT) containing 221 amino acids with a mass of 24.4 kDa, and (b) a membrane-bound form (MB-COMT) that contains 271 amino acids with a mass of 30.0 kDa (41). These two forms of COMT are identical and derived from the same gene located on chromosome 22, band q11.2, with the extra amino acids in MB-COMT responsible for anchoring the enzyme into the membrane (41, 42). The COMT gene contains six exons in which the coding region starts from exon 3, and on this exon, there are two distinct ATG-start codons for promoters that code for S-COMT and MB-COMT (43).
Crucial in the protection of cells of the body toward the mutagenic capability of catechol-estrogens is the proper enzymatic function of COMT, which attaches a methyl group to the catechol compound, making it inert. However, polymorphisms in genes have been shown to alter enzymatic activity, and in certain cases, the variant results in a reduction in activity (22). Polymorphic variants may thus play a mechanistic role by which COMT activity is reduced and result in the accumulation of mutagenic catechol compounds. There are several polymorphisms in the COMT gene and three of these in the coding region studied (codons 62, 72, and 158) are located in exons 3 and 4 (Fig. 1A). Codon 62 polymorphism results in no amino acid change (His) and is thus silent (29). The other two codons, however, result in amino acid substitutions: Ala→Ser for codon 72 and Val→Met for codon 158 (29).
The present study evaluated COMT polymorphisms and their risk for RCC (clear cell type) in a Japanese male population. The variant T allele in codon 62 was observed in 25.2% of controls and is consistent with a study on another sampling of a Japanese population that shows the T allele at a frequency of 28.6% (44). The genotypic frequencies of the present study were also similar with a second Asian population (Korean; ref. 45). Interestingly, there was a significant difference in the genotypic and allelic distribution at codon 62 of COMT between RCC and healthy controls with the T/T genotype having a 3.2-fold risk in patients with cancer. The heterozygous C/T was also higher with a 1.5-fold risk for RCC. Thus, the codon 62 variant, which remains silent as His62 seems to be of importance in the pathogenesis of RCC. Studies of this polymorphic site and its risk for cancer are few, but in support for a potential effect by this codon site, a previous study by Tanaka et al. (40) also observed the variant T/T genotype at codon 62 to be significantly associated with prostate cancer in a Japanese population.
The codon 72 variant T/T genotype is rare, being observed in only 1% of total samples with an allele frequency of just 8.4%. This low frequency of the variant was also observed among Koreans (45), and in a separate Japanese population, Yoshimura et al. (44) found a similar allele frequency of 10.3%. However, no association for RCC was observed at codon 72. As is the case with codon 62, studies involving codon 72 and cancer risk are even rarer but were attributed to this site being nonpolymorphic among Caucasians and African-Americans (46). In another study involving cancer at this site, no association was observed between codon 72 and prostate cancer (40). Thus, although a recent study shows that the variant Ser72 causes a reduction in enzyme activity (30), the polymorphism at this codon site (Ala→Ser) does not seem to be responsible for the development of these urological cancers in the Japanese male population.
Of particular interest is the polymorphism on codon 158 which has been widely studied and leads to an amino acid change, primarily due to the Met variant displaying a much weaker enzymatic activity (28) as well as enhanced thermolability (28, 47) as compared with the wild-type. The results of the present study, however, show no correlation between codon 158 polymorphism and RCC in this Japanese male population. The variant allele is expressed in approximately a quarter of healthy subjects and is consistent with other Asian populations (36, 45). In contrast to the limited population studies on codons 62 and 72, the polymorphism at the 158 codon has been extensively studied in a variety of cancers, although none involve RCC. Nonetheless, as was the case in this study, the codon 158 variant was also observed to be nonassociated with other types of cancers such as ovarian (48), bladder (37), endometrial (39), and liver cancers (38), and was inconclusive in breast cancer (reviewed in ref. 36).
RCC samples were classified in terms of pathologic grade and their correlation with COMT genotypes and alleles were analyzed. The classifications evaluated were based on grade 1 versus grade 2 or greater because the n size was small for grade 3 cancers. No association for genotype as well as allele frequency was observed according to grade of cancer at all polymorphic sites studied. Thus, COMT polymorphisms do not seem to play a role in the pathologic grade of RCC. This lack of association with grade of cancer for COMT polymorphisms was also observed with prostate cancer (40).
Thus, the polymorphism on codon 62 seems to be important for RCC. This codon was observed to be in linkage disequilibrium with codon 158 despite the absence of an association between the latter polymorphic site and RCC. Analyses show the codon 62-158, C-G haplotype to be significantly lower in RCC cases compared with healthy controls, and interestingly, the T-G haplotype accounted for much of this difference being several-fold higher in cases. The codon 62 T variant may thus be a factor in the etiology of RCC. Although the mechanism by which the T/T genotype may play a role in the process of carcinogenesis is not fully understood. The polymorphism is silent as the amino acid remains unchanged as His. However, gene expression could be affected as a result of a structural change in the RNA due to the variant, and may lead to an alteration in processing or efficiency of translation (49). Several studies have shown evidence that synonymous mutations can affect mRNA secondary structure and result in abnormalities (31-33). In a study by Duan et al. (32), the silent 957 C→T polymorphism of the human dopamine receptor D2 (DRD2) gene has been shown to alter mRNA folding, reduce mRNA stability and translation, as well as alter dopamine-induced DRD2 expression. In another study, Capon et al. (33) compared the stability of mRNA transcribed from haplotypes from the psoriasis-associated corneodesmosin (CDSN) gene. They observed that the risk haplotype resulted in a 2-fold increase in mRNA stability compared with the neutral haplotype and was associated with the disease. They show that a single synonymous polymorphism (971 C→T) of the CDSN gene accounts for the increase in RNA stability. Interestingly, these two studies (32, 33) involved a C→T polymorphic base change, and further experimentation is necessary to determine if codon 62 C→T could also result in the alteration of COMT mRNA stability.
Aside from its role in the body regulating various biological substrates such as dopamine and norepinephrine (50), the COMT enzyme is also important in the elimination of toxic or carcinogenic compounds that include catechol-estrogens which have been shown to induce tumors in animals (8, 10, 13), DNA single-strand breakage (15), mutation (16), and depurination (16). Proper RNA processing and its translation to the COMT enzyme at proper levels is therefore essential to provide a defense against these genotoxic compounds formed in the cells of the body. Thus, a polymorphic variant, with the potential to affect mRNA stability and/or alter processing of its enzyme, may lead to RCC.
This study is the first to report the polymorphisms at three locations in the COMT gene (codons 62, 72, and 158) and its association with RCC. Interestingly, the polymorphism at codon 62 was observed to be a risk factor for RCC. Because polymorphisms are inherited, the codon 62 polymorphism may thus be responsible for interindividual differences in RCC risk associated with catechol-estrogens. In conclusion, COMT polymorphisms may be important in understanding the pathogenesis of RCC.
Grant support: NIH T32 award no. DK07790.
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.