N -Acetyltransferase ( NAT2 ) Polymorphism and Breast Cancer Susceptibility: A Lack of Association in a Case-Control Study of Turkish Population

MATERIALS AND METHODS

Eighty-four women diagnosed with incident primary, histologically confirmed breast cancer by Ankara Numune Oncology and SSK Ankara Oncology Hospitals were then matched to 103 control women who had been living in Ankara region for at least 10 years based on age, time of blood donation, and menopausal status. The control group comprised of Gazi University staffs, students, and individuals from Turkish State Railways Hospital who were reported to be in good health, not taking any medications, and not having any relatives diagnosed breast or any type of cancer. All three hospitals are in Ankara, and serve patients predominantly from Central Anatolia. After written informed consent, whole blood samples were collected from cases and controls. At the time of blood donations, all participants completed a questionnaire, assisted by physician or assistant, that included information on age (menarche, first full-term pregnancy [FTP], menopause for postmenopausal women), menopausal status, height, weight, parity, number of pregnancies, use of oral contraceptives, hormone replacement therapy [HRT], bilateral oophorectomy, history of breast biopsy, family history of breast cancer, history of goitre, history of cigarette smoking and alcohol consumption, diet, and residence area.

Whole blood samples (5 to 10 ml) were withdrawn via venepuncture into tubes containing EDTA and samples were stored at −20°C until DNA extraction. Genomic DNA was extracted from whole blood using a sodium perchlorate/chloroform extraction as described by Daly et al. (1996).

The NAT2 genotyping was detected by a modification of the methods of Hickman et al. (1992) and Bell et al. (1993), which are designed to avoid the pitfalls reported for NAT2 genotyping (Cascorbi and Roots 1999). The sequences of the primers were NAT-Hu 14 (5′-GACATTGAAGCATATTTTGAAAG-3′) for the forward and NAT-Hu 16 (5′-GATGAAAGTATTTGA-TGTTTAGC-3°) for the reverse primer. A 60-μl polymerase chain reaction (PCR) was performed using approx 0.2 to 0.5 μg genomic DNA, 0.25 μM each primer, 2 mM of each dNTP, 5 Unit Taq DNA polymerase (Bioline) in the KCl-containing buffer supplied by the manufacturer (10 mM Tris-HCl, pH 8.3, 50 mM KCl, 1.5 mM MgCl₂). Amplification was carried out for 40 cycles by denaturing at 93°C for 1 minute, annealing at 58°C for 1.5 minutes, extending at 70°C for 2 minutes, and final extention at 70°C for 10 minutes for 1 cycle. A product of 1000 bp was obtained. Four 10-μl aliquots of the PCR products were digested with Kpn I, Dde I, Taq I, and BamH I restriction enzymes (New England Biolabs) to detect NAT2 ^* 5A, NAT2 ^* 5B, NAT2 ^* 5C, NAT2 ^* 6, and NAT2 ^* 7 alleles, respectively. These variants defined by the analyzed mutations are listed in Table 1. For each batch of 20 samples, controls known to be wild-type for alleles (NAT2 ^* 4/NAT2 ^* 4), heterozygotes for NAT2 ^* 5B, NAT2 ^* 6, and NAT2 ^* 7, and homozygotes for NAT2 ^* 5B, NAT2 ^* 6, NAT2 ^* 7 were also analyzed. For the Kpn I and BamH I digests, an internal control containing a single site for the relevant enzyme was added to each digest. These were CYP2D6 fragments (for Kpn I, bp 4093 to +125, 350 bp; for BamH I bp 2859 to 3482, 622 bp). The other restriction enzymes cut the NAT2 product at the several positions and internal controls for digestion are therefore not required. Samples digested with Dde I and Taq I were analyzed on 3% Nusieve agarose gels, samples digested with Kpn I on 2% agarose gels, and BamH I on 2% agarose gels with ethidium bromide. Samples that were heterozygous or homozygous mutant by Dde I analysis were further analysed by allele-specific PCR (ASPCR) for the T341C in order to (i) distinguish between the rapid acetylator NAT2 ^* 12 allele and the slow acetylator NAT2 ^* 5 allele; and (ii) determine if also heterozygous by Kpn I and negative for the other polymorphisms between an overall genotype of NAT2 ^* 4/NAT2 ^* 5B (rapid acetylator) and NAT2 ^*5A/NAT2 ^* 5C (slow acetylator). For the ASPCR assay, the primers were NAT-Hu 16 primer (as above) combined with either the wild-type–specific DH 1 (5′-TTC TCC TGC AGG TGA CCA T-3′) or the mutant-specific DH 2 (5′-TTC TCC TGC AGG TGA CCA C-3′). One microliter of the primary PCR product in a volume of 25 μl was reamplified under same conditions as for the primary PCR, except that PCR conditions were 35 cycles of 1 minute at 93°C, 1.5 minutes at 68°C, and 2 minutes 70°C. The products of amplification were analyzed on a 1% agarose gels with ethidium bromide. Three controls of known genotype for T341C were included in each batch of samples.

Departure from a Hardy-Weinberg equilibrium was tested among case and control groups using chi-square (χ ²) tests. The relationship between NAT2 genotypes and other known or suspected risk factors for breast cancer were studied by stratified analysis. The median value for cases and controls were used to dichotomize body mass index (BMI) (24 kg/m²), age (45 years), age at menarche (13 years), and age at first FTP (20 years). Variables of interest were (BMI) (≤24, >24 kg/m²), age at menarche (≤13, >13 years), and first FTP (nulliparous, ≤20, >20 years), smoking history (never, light [≤10 per day], heavy [>10 per day]), current alcohol use (never, ever), diet (vegetarian, fried, balance), postmenopausal use of estrogen (never, ever), use of oral contraceptives (never, ever), and history of goitre (yes, no). The results are also shown stratified by menopausal status (at the time of the diagnosis of the case patients). Women who reported natural menopause or had undergone bilateral oophorectomy were classified as postmenopausal. All of the others were classified as premenopausal. Four patients who had a family history of breast cancer were excluded from the study. Odds ratios (ORs), adjusted for age, age at menarche, age at first FTP, BMI, and smoking status were calculated, with 95% confidence interval (95% CI), by a binary logistic regression model. Analyses were done by using SPSS v.10.0 software.

RESULTS

Our study population included 84 breast cancer patients and 103 women controls matched by age, time of blood donation, and menopausal status. The mean (± SD) age for cases and controls were 48 years (±13, range 23–78) and 47 years (±12, range 20–78), respectively, yielding slightly fewer cases being premenopausal than controls (37 versus 47%). Case and control variables for the susceptibility of breast cancer are given in Table 2. The rapid acetylators of NAT2 allele frequencies in cases and controls were 50% and 43.7%, respectively, and slow acetylators of NAT2 allele frequencies were 50% and 56.3%, respectively. The genotype distributions of NAT2 in cases and controls are shown in Table 3. The frequency of rapid genotype in cases was slightly more common than controls, although there was no significant difference in the genotype frequency of the NAT2 rapid allele between two groups (the adjusted OR for age, age at menarche, age at first FTP, BMI, and smoking status is 0.78 [95% CI 0.44–1.38, p = .39]). The most common slow allele was the NAT2 ^* 5B allele in both cases (38.1%) and controls (38.9%). Among the NAT2 ^* 5B slow alleles in cases (25%), NAT2 ^* 5B/NAT2 ^* 6 was slightly higher than among controls (21.4%), whereas NAT2 ^* 5B/NAT2 ^* 5B slow allele in cases (9.5%) was slightly lower than among controls (14.6%). The second most frequent slow allele, NAT2 ^* 6/NAT2 ^* 6 was the same as among cases (9.5%) and controls (10.7%). Slightly less controls (43.7%) than cases (50%) had the wild-type allele NAT2 ^* 4. Some 8.3% of the cases and 6.8% of all controls were homozygous rapid allele carriers (NAT2 ^* 4/NAT2 ^* 4). Only four and one case subject were found to be NAT2 ^* 4/NAT2 ^* 5C and NAT2 ^* 12A/NAT2 ^* 12C, respectively, whereas no control subject was found. Only three controls were found to be NAT2 ^* 6/NAT2 ^* 7, whereas no case subject was found (Table 3).

Because different risk factors may contribute to breast cancer development, the association between NAT2 genotypes and breast cancer was further analyzed by stratifying for different variables and evaluated against control groups. It was suggested that the association of NAT2 genotype with breast cancer risk might depend on smoking. Although the smoking case and control group consisted of only 12 patients and 30 subjects, respectively, we found no significant associations with active or passive smoking, and ORs were <1.0. The association of genotype with enhanced susceptibility for breast cancer did not vary significantly with age, age at menarche, age at first FTP, ever use of oral contraceptives, history of alcohol use, diet, BMI, history of bilateral oophorectomy, history of goitre (not all data shown) (Table 4).

DISCUSSION

The NAT2 acetylation polymorphism is one of the most common polymorphisms known in human populations. More than 50% of Turkish population as well as Caucasians have NAT2 slow acetylator phenotype and genotype. Because NAT2 is responsible for the activation and deactivation of aromatic and heterocyclic amine carcinogens as well as clinically important aromatic and heterocyclic amine and hydrazine drugs, it is of great public health interest to understand the role of the acetylation polymorphism in cancer etiology (Weber and Hein 1984).

Studies investigating the relationship between NAT2 acetylator phenotype and genotype and breast cancer have yielded mixed results. NAT2 acetylator phenotype was not associated with breast cancer in three studies, although other studies have suggested that the rapid NAT2 acetylator phenotype is associated with breast cancer risk, or advanced disease at first presentation (Hein et al. 2000a). In Turkish population, only one phenotype study was carried out by our groups to examine the relation between NAT2 gene and breast cancer, although the number of patients was small, it was found that there was a higher percentage of rapid acetylator phenotypes with breast cancer (60.7%) than in control subjects (35.3%) (Şardaş et al. 1990). Therefore, the purpose of this study was to investigate whether NAT2 gene polymorphism is associated with breast cancer risk in Turkish women. To our knowledge, this is the first genetic study on the association of NAT2 genotypes with breast cancer in the Turkish population. We did not observe a positive relationship between NAT2 genotype and breast cancer, either overall or among subgroups of women defined by menopausal status, smoking, or other variables. None of the studies examining NAT2 genetic polymorphisms found it to be independently associated with breast cancer risk (Agundez et al. 1995; Ambrosone et al. 1996; Hunter et al. 1997; Millikan et al. 1998; Delfino et al. 2000a; Krajinovic et al. 2001). Dunning et al. (1999) examined the effect of common alleles of 18 different genes on breast cancer risk by combining results in meta-analysis. They concluded that for polymorphism in NAT2, the best estimate of risk either from individual studies or meta-analyses was sufficiently precise to exclude a relative risk of 1.5 or greater.

The association between NAT2 genotype and breast cancer has been investigated in relation to smoking and diet. The relationship between NAT2 genotype and breast cancer among smoking women has varied in studies. Elevated risk of breast cancer was reported among postmenopausal women who smoked and had the NAT2 slow acetylator genotype by Ambrosone et al. (1996) and Huang et al. (1999). In a recent study, Chang-Claude et al. (2002) suggest that active smoking variables, such as pack-years, duration of smoking, and time since cessation, showed significant dose-response relationships with breast cancer risk among slow acetylators but not rapid acetylators. On the contrary, Millikan et al. (1998) and Morabia et al. (2000) found that postmenopausal fast acetylators were at higher risk of breast cancer if they smoked, and also the associations of both passive and active smoking with breast cancer appear stronger in rapid than in slow NAT2 genotypes. The study done by Krajinovic et al. (2001) was consistent with the findings that the frequency of rapid acetylators was increased among 37 smokers. We found no association between breast cancer patients who are current/passive smokers and acetylator status due to their NAT2 genotypes, although the number of current smokers was small in both groups (14% versus 29%). Due to small sample size, it was difficult to say that our data were in agreement with the findings of Hunter et al. (1997) with current smokers of 55, and those of Delfino et al. (2000a) with current smokers of 54, suggesting that smoking was not a risk factor for breast cancer among NAT2 acetylators.

Consumption of well-done meat has been reported to be a risk factor in breast cancer. In our study, although we did not find any associations between breast cancer risk and the diet related to genotypes, we could not examine this factor in relation to cancer and NAT2 genotypes due to lack of individual information of red meat consumption. Four studies found that red meat consumption and NAT2 genotype were not associated with breast cancer risk. Ambrosone et al. (1998) reported that in general, consumption of meats, especially red meat as well as an index of concentrated sources of HA, was not associated with increased breast cancer risk for premenopausal or post-menopausal women, nor did risk vary by NAT2 genotypes. The findings were in agreement with the other study results (Gertig et al. 1999; Delfino et al. 2000b; Krajinovic et al. 2001). However, rapid/intermediate NAT2 genotypes were associated with a dose-dependent, nearly eightfold elevated, breast cancer risk in women who consistently consume very well-done meat in the study (Deitz et al. 2000). In a very recent study, Krajinovic et al. (2001) found that the frequency of NAT2 rapid acetylators increased 2.6-fold among smokers, however, this result did not modify the effect of well-done meat consumption on the risk of breast cancer in French-Canadians population.

Inherited alterations in the activity of any of enzymes involved in estrogen metabolism hold the potential to define differences in breast cancer risk associated with estrogens carcinogenesis. Because estrogen has been proposed to trigger breast cancer development via an initiating mechanism involving its metabolite catechol estrogen, which is formed by polymorphic CYP1B1 and inactivated by COMT (Thompson and Ambrosone 2000), we investigated in the same population in a parallel study. Our results shown that the CYP1B1 ^* 3 (L432V) allele was associated with a significantly increased susceptibility of breast cancer, with adjusted OR 2.32, (95% CI = 1.26–4.25, p = .007), whereas the COMT-L allele was not, OR 0.86 (95% CI = 0.46–1.60, p = .63) (Kocabaş et al. 2002). When NAT2 slow allele was combined with either COMT-HL and COMT-LL or CYP1B1 ^* 1/CYP1B1 ^* 3 and CYP1B1 ^* 3/CYP1B1 ^* 3 genotypes and all gene-gene interaction together, the risk for developing breast cancer was not significantly increased, OR 1.30 (95% CI = 0.71–2.37), OR 0.93 (95% CI = 0.52–1.65), and OR 1.28 (95% CI = 0.55–2.96), respectively.

Because the designation of the wild-type allele is dependent upon the ethnicity of the population studied, NAT2 ^* 4 has been designated as the wild-type. The slow genotype frequencies of NAT2 in controls (56.3%) are consistent with other phenotype (62%) and genotype (57.4%) studies on Turkish population, as well as Caucasians. The NAT2 slow allele frequency (56.3%), determined in the control group, was similar to previous population-based studies: African-American (55%) (Bell et al. 1993); Caucasian (58.9%) (Cascorbi et al. 1995); American (59.5%) (Gross et al. 1999); Swiss (56.5%) (Morabia et al. 2000); French-Canadian (61.1%) (Krajinovic et al. 2001); and German (60%) (Chang-Claude et al. 2002), but found to be somewhat significantly different from Taiwanese (21.1%) (Huang et al. 1999) and Chinese (24%) (Zhao et al. 2000). The G191A substitution common to the NAT2 ^* 14 gene cluster is present in African-Americans and native Africans, but it is virtually absent in Caucasians, Asians, and Turks. Because of its absence in Turkish population (Aynacioglu et al. 1997), we did not examine the NAT2 ^* 14 gene. The most prevalent genotypes in our control subjects were NAT2 ^* 5B/NAT2 ^* 6, NAT2 ^* 4/NAT2 ^* 6, NAT2 ^* 5B/NAT2 ^* 5B, NAT2 ^* 4/NAT2 ^* 5B, and NAT2 ^* 6/NAT2 ^* 6, which accounted for 21.4%, 18.4%, 14.6%, 13.6%, and 10.7% of the total genotypes, respectively. These genotype frequencies are nearly the same as those reported by Bell et al. (1993), Aynacioǧlu et al. (1997), and Gross et al. (1999).

We examined several of the most likely candidates for gene-environment interaction, but we failed to show a correlation between genotype and other factors. The lack of association between NAT2 genotype and breast cancer in the current study could be attributable to a number of factors. Because the number of study subjects is small, only moderate and large effects can be ruled out and, in stratified analyses, even moderate effects may have been undetected. Nevertheless, because the results in subgroup analyses were all based on small number of subjects, leading to risk estimates with wide CIs, these findings remain to be confirmed in future studies with even larger sample sizes. Furthermore, information on specific factors that might have influenced the association between the NAT2 genotype and breast cancer risk, such as detail of diet (red meat consumption, etc.), was unknown. Also, other enzymes (gene-gene interaction), including NAT1, several cytochrome P450 enzymes such as CYP1A2, and sulfotransferases, play an important role in the detoxifying mechanism of aromatic amines and the relationship of polymorphisms in these genes to breast cancer risk is still unclear. There is evidence for the existence of polymorphism in each of the genes encoding these enzymes, and it is possible that combinations of polymorphic enzymes may be better predictors of breast cancer risk than polymorphisms in one or two genes alone. Also, further investigations are needed to be carried out on genetic polymorphisms at carcinogen-metabolizing loci and analysis of gene-gene and gene-environment interactions in large series of patients.