Article N-acetyl transferase 2 polymorphism and advanced stages of endometriosis in South Indian women

Transcription

1 RBMOnline - Vol 9. No Reproductive BioMedicine Online; on web 7 September 2004 Article N-acetyl transferase 2 polymorphism and advanced stages of endometriosis in South Indian women Dr Sisnthy Shivaji Sisnthy Shivaji obtained his MSc (Zoology) from B.I.T.S., Pilani, India in 1973 and PhD (Zoology) from the University of Delhi in 1978, where he lectured from 1976 to He was research associate at the Indian Institute of Science, Bangalore from 1978 to 1980, before joining the Centre for Cellular and Molecular Biology, Hyderabad, in As a group leader at CCMB, he has been pursuing two distinct areas of research, namely reproductive biology and antarctic microbiology, with focus on protein kinases and phosphatases that regulate motility acquisition and capacitation of spermatozoa by specifically up-regulating the phosphorylation of proteins. The group has pioneered studies on the bacterial diversity of Antarctica, using these unique cold-loving micro-organisms to understand the molecular basis of cold adaptation. More recently his research on the fertility status of lions, tigers and leopards has paved the way for the conservation of mega cats in India. He has published more than 100 research papers and is a Fellow of the National Academy of Sciences, India and the Zoological Society of London. K Arvind Babu 1, K Lakshmi Rao 1, N G Pavankumar Reddy 1, M K Kanakavalli 2, Krina T Zondervan 3, Mamata Deenadayal 2, Amarpal Singh 1,, S Shivaji 1, 5, Stephen Kennedy 4 1 Centre for Cellular and Molecular Biology, Uppal Road, Hyderabad , India; 2 Infertility Institute and Research Centre, Hyderabad, India; 3 Wellcome Trust Centre for Human Genetics, University of Oxford, Oxford, UK; 4 Nuffield Department of Obstetrics and Gynaecology, University of Oxford, John Radcliffe Hospital, Oxford, UK 5 Correspondence: Tel: ; Fax: ; shivas@ccmb.res.in Posthumously Abstract Arylamine-N-acetyl transferase is a phase II detoxification enzyme encoded by the gene NAT2. Single nucleotide polymorphism (SNP) changes from the wild type NAT2 *4 allele result in allelic variants *5, *6 and *7. Homozygotes for the NAT2 *4 wild type are fast acetylators; heterozygotes with one wild-type allele and a variant NAT2 *5, *6 or *7 allele have reduced enzyme activity and individuals with two variant alleles are slow acetylators. Previous studies have implicated NAT2 as a susceptibility factor in endometriosis. This study investigated the NAT2 allele frequencies and genotype distributions in 252 unrelated women with endometriosis and 264 controls of South Indian origin. No differences were found between the frequencies of fast and slow acetylators in cases (34.9% and 65.1%) and controls (33.3% and 66.7%). Two NAT2 genotypes *7/*7 (1.2%) and *5/*6/*7 (1.6%) were detected in endometriosis cases only. Four new combinations, 6D ( mutation), 7C ( ), 7D ( ) and 7E ( ) were detected, which have not been reported earlier. Similar genotype and phenotype results were obtained in 33 affected sister-pairs. The case control data from this study suggest there is no association between endometriosis and NAT2 in South Indian women; however, two new variant genotypes and seven SNP combinations were also identified in cases only, which suggests that the gene may still have some as yet undetermined role in the disease. Keywords: alleles, endometriosis, genotypes, NAT2, polymorphism Introduction Endometriosis is a common gynaecological disorder characterized by the presence of endometrial-like tissue outside the uterus, which affects up to 10% of women in the reproductive years. Its aetiology and pathogenesis remain unclear but there is increasing evidence that endometriosis is inherited as a complex trait, in which multiple gene loci interact with each other and the environment to produce the disease phenotype (Kennedy, 1999; for review see Simpson and Bischoff, 2003). Many candidate genes have been studied as susceptibility factors, and only a few have resulted in associations replicated across studies (Zondervan et al., 2001). Conflicting evidence has been reported of association between endometriosis and polymorphisms/deletions in a variety of genes, including some involved in detoxification (Baranov et al., 1996; Baranova et al., 1997, 1999; Bischoff et al., 1998; Nakago et al., 2001) because exposure to dioxin, an environmental pollutant, has been implicated as a risk factor (Rier et al., 1993). Simpson and Bischoff (2003) suggested that the discrepant results may also be attributed to faulty experimental design and the existing DNA polymorphism due to ethnicity of the population studied. 533

2 534 One such candidate is N-acetyl transferase 2 (NAT2), which is involved in the biotransformation of aromatic amines and hydrazines and catalyses the transfer of an acetyl group from acetyl CoA to the nitrogen of the substrate. Polymorphisms at the NAT2 gene locus, 8p22, result in impaired enzyme activity. Homozygotes for the NAT2 *4 wild-type allele are fast acetylators; heterozygotes with a wild-type allele and a variant NAT2 *5, *6 or *7 allele have reduced activity and individuals with two variant alleles are slow acetylators. Slow acetylation has been reported as a risk factor for bladder cancer (Risch et al., 1980; Cartwright et al., 1982), while fast acetylation has been implicated as a risk factor for colon cancer (Ilett et al., 1987). In endometriosis, findings have varied considerably in different populations. Fast acetylation has been associated with endometriosis in the UK population (Nakago et al., 2001), whereas among North Americans an association has been reported for the slow acetylator phenotype (Bischoff et al., 1998). It is difficult to choose an appropriate control group in endometriosis. In many early studies, pre-menopausal women aged years and males (Nakago et al., 2001), umbilical cord blood from newborn infants (Morizane et al., 2004) and healthy fertile females without any clinical diagnosis (Kado et al., 2002; Arvanitis et al., 2003) were used. It is the opinion of the authors that the best controls should include adult females, drawn from the same population as the cases, who are laparoscopically negative for endometriosis since such controls are unlikely to differ with respect to factors that are known to influence the disease. At the present time, several potential candidate genes such as GALT (galactose-1- phosphate uridyl transferase), Ah receptor (aryl hydrocarbon receptor), CYP1A1 (cytochrome P450), GST (glutathione S- transferase), NAT2 (N-acetyl transferase), ICAM-1 (intracellular adhesion genes), MSH2, MSH6, MCH1 and PMS1 (DNA mismatch repair genes), TP53 and PTEN (tumour suppressor genes) and a few other genes (reviewed in Simpson and Bischoff, 2003) have been studied to establish associations, if any, between the observed genetic polymorphism of the genes and endometriosis. These results indicate conflicting associations probably reflecting DNA polymorphism indigenous to the ethnicity of the population (Simpson and Bischoff, 2003). Thus keeping this in view and the fact that NAT2 association with endometriosis varied in populations from UK (Nakago et al., 2001) and North America (Bischoff et al., 1998), the present study was undertaken to ascertain the status of NAT2 gene polymorphism and endometriosis in South Indian women. Materials and methods Subjects Two-hundred and fifty-two unrelated women of South Indian origin with moderate severe (III IV) endometriosis staged using the revised American Fertility Society (rafs) classification system (1985) were recruited at the Infertility Institute and Research Centre (IIRC), Hyderabad, India. The cases included 33 probands with known familial history of the disease. All women had a trans-vaginal ultrasound scan (TVS) at screening followed by a laparoscopy to confirm the diagnosis (rafs III = 85; IV = 167). Women with ovarian cysts, adenomyosis, adhesions, ovarian cancer and fibroids were excluded from the study. All the subjects were nonsmokers who consumed little caffeine, reflecting the habits of most women in the region. Their mean age was 29.5 ± 5.5 (range 18 to 40) years. All the patients complained of dysmenorrhoea (mild = 47%; moderate = 30%; severe = 23%), and 75% had dyspareunia. Most women (97.7%) were infertile (primary = 78%; secondary = 22%). Written informed consent was obtained from all participants. The Institutional Review Board of Centre for Cellular and Molecular Biology (CCMB), Hyderabad, approved the study. Controls Two-hundred and sixty-four women were recruited from the same clinic population as the cases. All the 264 controls were also non-smokers who consumed little caffeine. Their mean age was 27.9 ± 4.95 (range 18 to 40) years. The clinical symptoms in the controls were dysmenorrhoea (mild = 29%; moderate = 0%; severe = 0%), dyspareunia = 20% and infertility (primary = 22%; secondary = 0%); the remaining 29% showed none of these symptoms. These controls consisted of 217 women without evidence of endometriosis at TVS followed by laparoscopy, and 47 women without evidence of ovarian endometriosis at TVS who subsequently, for various reasons, did not have a laparoscopy. Including these 47 individuals as controls, without evidence of ovarian endometriomas on ultrasound, was justifiable given that the cases all had rafs Stage III IV disease. Genotyping All genotyping was performed at CCMB. The NAT2 polymorphisms and allelic status were determined by the polymerase chain reaction/restriction fragment length polymorphism (PCR-RFLP) methods described by Nakago et al. (2001) with minor modifications. Similar to the human leukocyte antigen (HLA) system, the allelic status of NAT2 is based on multi-locus rather than single-locus base changes (Vatsis et al., 1991). Status was assessed at four loci: for example, compared with the wild-type *4 allele, allele *5 signifies the presence of a 481 C/T (KpnI) base change, an 803 A/G (DdeI) base change, or both (see Table 1). Genomic DNA was extracted and purified from 1 ml of peripheral blood using the standard sequential lysis protocol, followed by phenolchloroform treatment and ethanol precipitation (Miller et. al., 1988). A 547 bp fragment of NAT2 gene was amplified using the primers 5 -GCTGGGTCTGGAAGCTCCTC-3 (forward) and 5 -TTGGGTGATACATACACAAGGG-3 (reverse) which were synthesized at CCMB. The 50 μl PCR reaction mixture contained 1 PCR buffer, 2.5 mmol/l MgCl 2, 2.5 IU Taq polymerase (AmpliTaq Gold, Roche, Indianapolis, IN, USA), 0.2 mmol each dntp (Roche, Mannheim, Germany) and 50 ng of template DNA. PCR amplification was performed on PTC- 200 PCR machines (MJ Research Inc., Waltham, MA, USA) using the following thermal profile: initial denaturation at 95 C for 5 min, followed by 32 cycles of denaturation at 94 C for 1 min, annealing at 60 C for 40 s, elongation at 72 C for 1 min and final elongation at 72 C for 7 min. H 2 O was used as PCR control instead of DNA in all sets of PCR reactions to check any possible contamination.

3 The amplified product (7 μl) was run on 1.5% agarose gel to test the amplification and the remaining product in aliquots of 10 μl was separately digested simultaneously using four restriction enzymes KpnI (481 C/T), TaqI (590 G/A), DdeI (803 A/G) and BamHI (857 G/A) all purchased from New England Biolabs (NEB), Inc., Beverly, MA, USA. PCR product digestion was completed after overnight incubation at 37 C (65 C for TaqI-digested product) in a reaction mixture containing 10 μl of the PCR product and 5 μl of enzyme mix containing 1 reaction buffer (NEB) and bovine serum albumin (NEB). The cleaved fragments were separated by electrophoresis on a 2.5% agarose gel for KpnI- and TaqIdigested products, 3% agarose gel for BamHI-digested products and 7% polyacrylamide gel (PAGE) for DdeIdigested products. The gels were stained with ethidium bromide and images of bands were captured using the Syngene Gel Documentation system (Syngene, Cambridge, UK) as shown in Figure 1. Statistical analysis The genotypic distribution among controls was tested for Hardy Weinberg equilibrium ( using a chi-squared or exact test (where expected genotype frequencies were low). Genotypic and allelic associations were assessed using Cochran Armitage tests for trend (exact tests were used when expected frequencies were low), and Pearson s chi-squared tests for heterogeneity. All association analyses were conducted using SAS version A P-value of less than 0.05 was considered significant. Results All 252 cases and 264 controls were successfully genotyped. All the single nucleotide polymorphisms (SNP) tested were in Hardy Weinberg equilibrium except for 481 C/T (P = 0.02) in cases (Table 1). One explanation could be the high frequencies of the mutant and wild-type forms of the 481 C/T and 803 A/G (Lys268Arg) polymorphisms, neither of which affect enzyme activity if present alone. Therefore, genotype variance of only 13.76% between two polymorphisms in this population is not just by chance. Since this low genotype variance is not seen with other SNP (590 G/A and 857 G/A), it clearly suggests that the 481 C/T polymorphism has an influence over 803 A/G or vice versa. In addition, a considerable difference (chi-squared = 8.7; P = 0.03) in the loss of heterogeneity (481 ht ht) was also observed in cases compared with controls, implying that the decrease in heterogeneity is one of the factors for the 481 C/T polymorphism not obeying the Hardy Weinberg equilibrium (Table 2). The 547-bp amplified product showed different fragments when digested with KpnI (547 bp, 433 bp and 114 bp), TaqI (392 bp, 222 bp, 172 bp and 155 bp), DdeI (345 bp, 137 bp, 114 bp and 65 bp) and BamHI (547 bp, 490 bp and 57 bp) (Figure 1). There were no significant differences in SNP frequencies between the cases and controls, with the exception of 857 G/A, which was weakly associated (Table 1). Haplotype analysis of the four SNP revealed a total of 28 combinations, of which 20 were common to both cases and controls, seven were specific to cases and one was specific to controls (Table 3). The frequency of combination No. 13 (Table 3) was considerably less in cases compared with controls. Four new combinations namely mutation, , and were found, which have not been reported previously (Brockton et al., 2000). Table 4 shows the frequencies of the NAT2 genotypes, the four alleles and the acetylator phenotypes in cases and controls. The frequency of the slow acetylator status was similar (~66%) to the figure of 74% previously reported in the general South Indian population (Anitha et al., 2003). A total of 11 genotypes were Table 1. NAT2 polymorphisms in women of South Indian origin. NAT2 Endometriosis Controls P-value Chi- Hardy Weinburg polymorphisms stage III a ; stage IV a ; Total; n = 264 (%) (trend) b squared equilibrium n = 85 (%) n = 167(%) n =252 (%) Cases d Controls 481 C/T mt 8 (9.4) 26 (15.6) 34 (13.5) 31 (11.7) KpnI ht 31 (36.5) 63 (37.7) 94 (37.3) 116 (43.9) wt 46 (54.1) 78 (46.7) 124 (49.2) 117 (44.3) e G/A mt 6 (7.1) 27 (16.2) 33 (13.1) 36 (13.6) TaqI ht 47 (55.3) 66 (39.5) 113 (44.8) 132 (50.0) wt 32 (37.7) 74 (44.3) 106 (42.1) 96 (36.3) A/G mt 10 (11.8) 25 (15.0) 35 (13.9) 37 (14.0) DdeI ht 35 (41.2) 69 (41.3) 104 (41.3) 122 (46.2) wt 40 (47.1) 73 (43.7) 113 (44.8) 105 (39.7) G/A mt 3 (3.5) 1 (0.6) 4 (1.6) 0 (0) BamHI ht 13 (15.3) 30 (18.0) 43 (17.1) 34 (12.9) wt 69 (81.2) 136 (81.4) 205 (81.3) 230 (87.1) 0.04 c a Classified according to Revised American Fertility Stage of endometriosis. b Comparing total cases combined with controls. c Fisher s exact test used. d Calculated using permutation exact test. e Significant difference due to influence of 481 C/T and 803 A/G polymorphisms genotype match. mt = homozygous mutant; ht = heterozygous mutant; wt = wild type. 535

4 Table 2. NAT2 481 C/T and 803 A/G genotype match in 252 cases and 264 controls. All women; Cases; Controls; n = 516 (%) n = 252 (%) n = 264 (%) 445 (86.24) 208 (82.53) 237 (89.77) 481 ht ht 80 (31.74) a 108 (40.90) a 481 mt mt 28 (11.11) 29 (10.98) 481 wt wt 100 (39.68) 100 (37.87) Genotype variance 46 (18.25) 27 (10.22) a Heterogeneity test, chi-squared = 8.7; P = mt = homozygous mutant; ht = heterozygous mutant; wt = wild type. detected, of which NAT2 *7/*7 (1.2%) and *5/*6/*7 (1.6%) were unique in that they were absent in the controls. No significant differences were observed between the frequencies of fast/slow acetylators in cases (34.9% and 65.1%) and control women (33.3% and 66.7%); similarly, there were no differences in allele frequencies. Thirty-three sisters of the 33 cases with a known family history were also genotyped; all these additional women had surgically confirmed stage II IV disease. The NAT2 genotypes and acetylator status of the 33 affected sister-pairs are shown in Table 5. Thirteen (39.4%) had identical NAT2 genotypes and 24 (72.7%) had the same acetylator status. These figures were similar to the figures of 43.4 and 74.2, respectively, expected Figure 1. Electrophoresis results of 547-bp amplified fragment of NAT2 when digested with A, KpnI (547 bp, 433 bp and 114 bp) which was run on 2% agarose gel; B, TaqI (392 bp, 222 bp, 172 bp and 155 bp) digested product run on 2.5% agarose gel; C, DdeI (345 bp, 137 bp, 114 bp and 65 bp) digested product run on 7% DNA-PAGE and D, BamHI (547 bp, 490 bp and 57 bp) digested product run on 3% agarose gel. 536

5 Table 3. Different combinations of four NAT2 SNP tested and genotypes observed in cases and control women. S no Genotype Acetylation Cases; Controls; C/T G/A A/G G/A status n = 252 (%) n = 264 (%) KpnI TaqI DdeI BamHI 1 wt wt wt wt *4/*4 Fast 9 (3.6) 8 (3.0) 2 ht ht *4/*5 Fast 28 (11.1) 35 (13.3) 3 ht *4/*5 Fast 4 (1.6) 1 (0.4) 4 ht *4/*5 Fast 2 (0.8) 3 (1.1) 5 ht *4/*6 Fast 39 (15.5) 37 (14.0) 6 ht *4/*7 Fast 6 (2.4) 4 (1.5) 7 ht mt *5/*5 Slow 5 (2.0) 4 (1.5) 8 mt mt ht *5/*5 Slow 2 (0.8) 0 (0) 9 mt mt *5/*5 Slow 26 (10.3) 29 (10.1) 10 mt *5/*5 Slow 4 (1.6) 2 (0.8) 11 mt *5/*5 Slow 1 (0.4) 1 (0.4) 12 mt mt *5/*5 Slow 0 (0) 1 (0.4) 13 a ht ht ht *5/*6 Slow 37 (14.7) 63 (23.9) 14 ht mt ht *5/*6 Slow 2 (0.8) 1 (0.4) 15 b ht ht *5/*6 Slow 5 (2.0) 3 (1.1) 16 mt ht *5/*6 Slow 2 (0.8) 0 (0) 17 ht ht *5/*6 Slow 10 (4.0) 7 (2.7) 18 mt ht *5/*6 Slow 5 (2.0) 3 (1.1) 19 ht mt *5/*6 Slow 1 (0.4) 2 (0.8) 20 ht ht ht *5/*7 Slow 11 (4.4) 9 (3.4) 21 ht ht *5/*7 Slow 1 (0.4) 0 (0) 22 ht ht *5/*7 Slow 2 (0.8) 0 (0) 23 b mt ht *6/*6 Slow 1 (0.4) 1 (0.4) 24 mt *6/*6 Slow 25 (9.9) 30 (11.4) 25 ht ht *6/*7 Slow 17 (6.7) 20 (7.6) 26 mt *7/*7 Slow 3 (1.2) 0 (0) 27 b ht ht ht mt? Slow 1 (0.4) 0 (0) 28 b ht ht ht? Slow 3 (1.2) 0 (0) a Significant increase in frequency of combination No. 13 in controls compared to cases. b Combinations not reported earlier. mt = homozygous mutant; ht = heterozygous mutant; wt = wild-type.? = nomenclature not known. 537

6 Table 4. Comparison of genotypes, acetylation status and allele frequencies between populations from India and the UK. Genotype Acetylation South Indian women UK population a status III b ;n= IV b ;n= Cases; n= Controls; Cases; Controls; Males; 85 (%) 167 (%) 252 (%) n = 264 (%) n = 54 (%) n = 24 (%) n = 99 (%) *4/*4 Fast 4 (4.7) 5 (3.0) 9 (3.6) 8 (3.0) 3 (5.5) 2 (8.3) 4 (4.0) *4/*5 Fast 9 (10.6) 25 (15.0) 34 (13.5) 39 (14.8) 9 (16.7) 5 (20.8) 18 (18.2) *4/*6 Fast 19 (22.3) 20 (12.0) 39 (15.5) 37 (14.0) 19 (35.2) 1 (4.2) 8 (8.1) *4/*7 Fast 2 (2.3) 4 (2.4) 6 (2.4) 4 (1.5) 0 (0) 0 (0) 2 (2.0) *5/*5 Slow 11 (12.9) 27 (16.2) 38 (15.1) 37 (14.0) 7 (13.0) 5 (20.8) 21 (21.2) *5/*6 Slow 24 (28.2) 38 (22.7) 62 (24.6) 79 (29.9) 13 (24.1) 6 (25.0) 31 (31.3) *5/*7 Slow 3 (3.5) 11 (6.6) 14 (5.6) 9 (3.4) 2 (3.7) 1 (4.2) 1 (1.0) *6/*6 Slow 5 (5.9) 21 (12.6) 26 (10.3) 31 (11.7) 1 (1.9) 4 (16.7) 13 (13.1) *6/*7 Slow 4 (4.7) 13 (7.8) 17 (6.8) 20 (7.6) 0 (0) 0 (0) 1 (1.0) *7/*7 Slow 2 (2.3) 1 (0.6) 3 (1.2) 0 (0) 0 (0) 0 (0) 0 (0) (*5/*6/*7)? Slow 2 (2.3) 2 (1.2) 4 (1.6) 0 (0) 0 (0) 0 (0) 0 (0) Acetylation Fast 34 (40.0) 54 (32.3) 88 (34.9) 88 (33.3) 31 (57.4) 8 (33.3) 32 (32.3) Slow 51 (60.0) 113 (67.7) 164 (65.1) 176 (66.7) 23 (42.6) 16 (66.6) 67 (67.7) Alleles tested Allele frequency * * * * a Nakago et al., b Classified according to revised American Fertility Society (1985) stage of endometriosis.? = Nomenclature not known (see Table 3: ). Table 5. NAT 2 genotypes and NAT 2 acetylation status between sisters with endometriosis. Sister pair rafs stage a No. of Similarity between sisters disease severity Proband Sister sister pairs NAT2 NAT2 genotypes acetylation Mild + moderate III II 7 2 (28.6) 5 (71.4) Mild + severe IV II 8 1 (12.5) 4 (50) Moderate III III 3 0 (0) 1 (33.3) Moderate + severe IV III 9 7 (77.8) 8 (88.9) Severe IV IV 6 3 (50) 6 (100) One mild 15 3 (20.0) b 9 (60.0) c Both moderate + severe (55.6) b 15 (83.3) c Total sister pairs (39.4) 24 (72.7) a Revised American Fertility Society stage of endometriosis (American Fertility Society, 1985). b Comparison between sister-pairs with one mildly affected and those with both moderate severely affected: P = c Comparison between sister-pairs with one mildly affected and those with both moderate severely affected: P =

7 following laws of Mendelian segregation for the distribution of genotype and acetylator status of the probands. Interestingly, when both sisters had rafs stage IV disease, there was 100% acetylator status and 50% genotype concordance, whereas Mendelian expectation was 71.9% and 42.3%, respectively. Discussion The case control data of this study show no association between endometriosis and polymorphisms in the NAT2 gene in South Indian women; dividing the cases into rafs Stage III (n = 85) and Stage IV (n = 167) had no effect upon the results. The frequencies of slow/fast NAT2 acetylators in the cases and controls were also similar. Therefore, these data do not support previously reported studies (Bischoff et al., 1998; Baranova et al., 1999; Nakago et al., 2001). The results did not vary with respect to the association status between NAT2 and endometriosis even when the 47 women without evidence of endometriosis at TVS and who subsequently did not have a laparoscopy were excluded from the control group (data not shown). To further investigate differences between these studies and the current one NAT2 genotypes and allele frequencies were compared between South Indian women and those in the UK population, in whom an association between NAT2 activity and the disease has been described (Nakago et al., 2001). The frequencies of fast and slow acetylators in the South Indian controls and in unaffected female and male controls in the UK population were similar (Table 4). However, the two populations differed in many ways: for example, the frequencies of NAT2 *4/*6, *4/*7 and *6/*7 genotypes were higher in South Indian women; the frequencies of *4/*5 and *5/*5 were higher in the UK control population, and the NAT2 *4/*7, *6/*7 and *7/*7 genotypes were completely absent in UK endometriosis cases and unaffected women. Thus the distributions of NAT2 genotypes and acetylator status are different in women of South India and UK origins, which underlines the importance of studying the candidate genes in different populations in this or any other disease. A total of 28 combinations of four NAT2 SNP were observed in endometriosis cases (27) and control women (21) in the South Indian population (Table 3). Of these 28 combinations, seven (25%) mutant combinations (all slow acetylator) were detected in cases alone and not in controls, implying that the number of slow acetylator genotype combinations is higher in cases than controls. Although, the frequency of a slow acetylator combination No. 13 appeared less in cases compared with controls, numbers were insufficient to test this individual association. No overall association was found between endometriosis and slow acetylator status. Further, based on sister-pair data the present results suggest that sister-pairs in identical stages of endometriosis did not have the same NAT2 genotype (Table 5) but when the proband and the affected sister both had stage IV disease, they showed 100% NAT2 acetylation status (Table 5). In conclusion, the results of this case-control study suggest that there is no significant association between endometriosis and acetylator status in South Indian women. However, the occurrence of seven rare SNP combinations in cases alone and the possibility of an increased similarity in genotypes and NAT2 acetylator status in sister-pairs with severe endometriosis, imply that NAT2 may have a role to play in a small subset of endometriosis cases. Acknowledgements We sincerely acknowledge the Wellcome Trust for financial assistance, Ms Jayanthi and Ms Bhagya for their help in patient recruitment and Dr Manjula Bhanoori and Mr VV Suryanarayana for their support and helpful suggestions. References American Fertility Society 1985 Revised American Fertility Society classification of endometriosis Fertility and Sterility 43, Anitha A, Banerjee M 2003 Arylamine N-acetyltransferase 2 polymorphism in the ethnic populations of South India. International Journal Molecular Medicine 11, Arvanitis DA, Koumantakis GE, Goumenou AG et al CYP1A1, CYP19, and GSTM1 polymorphisms increase the risk of endometriosis Fertility and Sterility 79 (suppl. 1), Baranov VS, Ivaschenko T, Bakay B et al Proportion of the GSTM1 0/0 genotype in some Slavic populations and its correlation with cystic fibrosis and some multifactorial diseases. Human Genetics 97, Baranova H, Bothorishvilli R, Canis M et al Glutathione S transferase M1 gene polymorphism and susceptibility to endometriosis in a French population. Molecular Human Reproduction 3, Baranova H, Canis M, Ivaschenko T et al Possible involvement of arylamine N-acetyltransferase 2, glutathione S- transferase M1 and T1 genes in the development of endometriosis. Molecular Human Reproduction 5, Bischoff FZ, Marquez-Do D, Kosugi Y et al Association of N- acetyltransferase 2 (NAT2) genetic polymorphism resulting in decreased capacity to detoxify aromatic amines in women with endometriosis. Journal of the Society for Gynecological Investigations 5, 111A. Brockton N, Little J, Sharp L, Cotton SC 2000 N-Acetyltransferase polymorphisms and colorectal cancer: A HuGE review. American Journal of Epidemiology 151, Cartwright RA, Glashan RW, Rogers HJ et al Role of N- acetyltransferase phenotypes in bladder carcinogenesis: a pharmacogenetic epidemiological approach to bladder cancer. Lancet ii, Ilett KF, David BM, Detchon P et al Acetylation phenotype in colorectal carcinoma. Cancer Research 47, Kado N, Kitawaki J, Obayashi H, et al Association of the CYP17 gene and CYP19 gene polymorphisms with risk of endometriosis in Japanese women. Human Reproduction 17, Kennedy S 1999 The genetics of endometriosis. European Journal of Obstetrics and Gynecology and Reproductive Biology 82, Miller SA, Dykes DD, Polesky HF 1988 A simple salting out procedure for extracting DNA from human nucleated cells. Nucleic Acids Research 16, Morizane M, Yoshida S, Nakago S et al No association of endometriosis with glutathione S-transferase M1 and T1 null mutations in a Japanese population. Journal of the Society for Gynecologic Investigation 11, Nakago S, Hadfield RM, Zondervan KT et al Association between endometriosis and N-acetyl transferase 2 polymorphisms in a UK population. Molecular Human Reproduction 7, Rier SE, Martin DC, Bowman RE et al Endometriosis in rhesus monkeys (Macaca mulatta) following chronic exposure to 2,3,7,8-tetrachlorodibenzo-p-dioxin. Fundamental and Applied 539

8 Toxicology 21, Risch A, Wallace DM, Bathers S et al Slow N-acetylation genotype is a susceptibility factor in occupational and smoking related bladder cancer. Human Molecular Genetics 4, Simpson JL, Bischoff F 2003 Heritability and candidate genes for endometriosis. Reproductive BioMedicine Online 7, Vatsis KP, Martell KJ, Weber WW 1991 Diverse point mutations in the human gene for polymorphic N-acetyltransferase. Proceedings of the National Academy of Sciences USA 88, Zondervan KT, Cardon LR, Kennedy SH 2001 The genetic basis of endometriosis. Current Opinion on Obstetrics and Gynecology. 13, Received 30 July 2004; refereed 4 August 2004; accepted 13 August