| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Brief Communications |
1 Department of Pharmacology, Medical School, University of Extremadura, Badajoz, Spain;2 Institute for Occupational Physiology (IfADo), University of Dortmund, Dortmund, Germany;3 Fakultät Statistik, Technische Universität Dortmund, Dortmund, Germany;4 Department of Biochemistry and Molecular Biology, School of Biological Sciences, University of Extremadura, Badajoz, Spain;
aaddress correspondence to this author at: Department of Pharmacology, Medical School, University of Extremadura, Avda. de Elvas s/n, 06071-Badajoz, Spain. Fax +34 24 27 94 58; e-mail jagundez{at}unex.es.
Abstract
Background: Arylamine N-acetyltransferase 2 (CoASAc; NAT2, EC 2.3.1.5) is a drug-metabolizing enzyme that displays common polymorphisms leading to impaired drug metabolism and adverse drug effects. Determination of the N-acetyltransferase 2 (arylamine N-acetyltransferase) (NAT2) genotype in clinical practice is hampered by the occurrence of ambiguous haplotype combinations that may lead to patient misclassification. We determined the frequencies for ambiguous NAT2 haplotypes and diplotypes in a white population and investigated the use of PHASE v2.1.1, a statistical program for haplotype reconstruction, to clarify this ambiguity and classify individuals according to their acetylation status.
Methods: By means of allele-specific haplotype mapping and sequencing, we determined the haplotypes for 7 common single-nucleotide polymorphisms in the NAT2 gene (n = 2624 haplotypes). To test the performance of PHASE, actual genotypes were deconstructed and then reconstructed by haplotype prediction.
Results: We identified 21 NAT2 allelic variants, including a new variant allele that combines the single-nucleotide polymorphisms rs1801279, rs1799929, and rs1208. In contrast, the previously described variant alleles *5G, *5J, *6E, *7A, *11A, *11B, and *14B were not identified in the study population. Ambiguous haplotypes were observed in 98 alleles (3.7%), and ambiguous diplotypes were observed in 64 individuals (4.9%). Eleven individuals (0.8%) were misclassified by the use of haplotype prediction.
Conclusions: Ambiguous NAT2 genotyping data are common. Actual NAT2 genotypes cannot be fully determined by haplotype prediction techniques. This study provides real haplotype data that can be used as a guide to convert NAT2 haplotypes and diplotypes into actual genotypes in white individuals.
The arylamine N-acetyltransferase 2 (NAT21 ; E.C. 2.3.1.5) is a polymorphic enzyme involved in the metabolism of drugs and aromatic amines. About 50% of white individuals are classified as slow acetylators, and these individuals show impaired metabolism of many therapeutically useful arylamine and hydrazine drugs (1). In patients with tuberculosis who receive isoniazid, determination of the N-acetyltransferase 2 (arylamine N-acetyltransferase) (NAT2) genotype or phenotype has been proposed as a method for predicting adverse reactions as well as before the concomitant administration of drug combinations such as procainamide-phenytoin or doxycycline-rifampin (1)(2)(3). In addition, the NAT2 polymorphism modulates the risk for the development of bladder (4)(5) and liver cancer (6).
Diverse NAT2 genotyping methods have been developed, including amplification-restriction (7) and allele-specific methods (8), real-time PCR (9), and high-throughput microarrays (10). All of these genotyping methods simultaneously analyze 2 alleles (i.e., the paternal and the maternal allele). This characteristic is a major source of uncertainty in the case of NAT2 analysis, because 2 or more enzyme-inactivating mutations can be located in 1 allele, leaving the other allele intact, or the mutations can be distributed between the 2 alleles, leading to a lack of functional NAT2 alleles (1)(5). In fact, rather than indicating real genotypes, most genotyping method results indicate the sum of 2 haplotypes (diplotypes), and many such diplotypes can represent different genotypes with different functional consequences. This fact, together with the increasing importance of haplotypes in association studies (11), makes it necessary to determine specific haplotype combinations.
Four common allelic variants, NAT2*4, NAT2*5B, NAT2*6A, and NAT2*7B, account for most NAT2 gene variants in white individuals and lead to unambiguous genotypes (1)(12). Many other variants have been described, however, up to a total of 52, and these variants may lead to ambiguous genotypes (13). The frequencies for rare NAT2 allelic variants have not been investigated in detail because such an investigation requires a large population group, and haplotyping methods are expensive and time-consuming. Instead, diverse statistical approaches have been proposed to reconstruct NAT2 haplotypes (14). The estimated performance of these reconstruction programs for NAT2 genotyping, however, relies on inferred haplotype frequencies (14). To date no comparisons between actual haplotypes and reconstructed haplotypes have been performed in a study group large enough to include rare variant alleles in white individuals.
We analyzed the allele frequencies for common and rare NAT2 allele variants in a large white population. Because of their small sample sizes, previous haplotyping studies, as well as HapMap data (15), have been unable to detect rare alleles. With the sample size used for the present study (n = 2624 alleles), we were able to detect rare variant alleles that occur with frequencies below 0.0005. By comparing actual data with reconstructed data, we tested the PHASE haplotype reconstruction program to determine its usefulness for clarifying ambiguity in haplotype analysis.
All study participants were white Spanish individuals randomly selected from a group of medical students and university and hospital staff. All study participants gave informed consent, and the protocol was approved by the ethics committee of the Infanta Christina University Hospital (Badajoz, Spain).
Study samples were obtained by venipuncture; 10 mL peripheral blood was obtained and genomic DNA extracted as described elsewhere (7). Mutation analysis at the NAT2 gene locus was achieved by the use of PCR-based mutation-specific amplification. First, a DNA fragment spanning 1212 bp, which contained the whole coding region of the NAT2 gene, was amplified from genomic DNA from study samples by use of primers specifically designed to amplify the NAT2 gene, avoiding amplification of the highly homologous genes N-acetyltransferase 1 (arylamine N-acetyltransferase) (NAT1) and arylamide acetylase pseudogene (AACP; also known as NATP) (7). The amplified DNA fragments were then analyzed for the presence of the 7 most common single nucleotide polymorphisms (SNPs) at the coding region of the gene, namely G191A (rs1801279), C282T (rs1041983), T341C (rs1801280), C481T (rs1799929), G590A (rs1799930), A803G (rs1208), and G857A (rs1799931). Once the SNPs were identified, we performed separate SNP analyses for both alleles for each study participant to obtain a complete haplotype map for each allele in each individual. Details of the whole procedure are given elsewhere (16). The occurrences of rare haplotypes (those with frequencies below 0.01) were confirmed by sequencing as described elsewhere (17).
We performed haplotype reconstruction with the PHASE v2.1.1 program (18)(19) using the default model for recombination rate variation (20). We performed 7 independent runs with 1000 iterations, 500 burn-in iterations, and a thinning interval of 1. Of the 7 runs we selected as the best run the one that showed the maximum consistency of results across all runs. We also examined the consistency of the results obtained by applying the PHASE algorithm repeatedly with default and varying values for the number of iterations, the thinning interval, and the burn-in iterations.
The NAT2 diplotypes and the actual genotypes identified in the study group are shown in Table 1
. The distribution of NAT2 haplotypes in 2624 genes is shown in a Supplemental Data Table available with the on-line version of this manuscript at www.clinchem.org. Some variant alleles, such as *5B, *5F, *5H, *5I, *5L, and *5M (cluster NAT2*5B) or *6A, *6D, *6G, *6H, *6K, and *6L (cluster NAT2*6A), are shown grouped in the tables, because these variant alleles consist of the same combination of the 7 SNPs analyzed here and differ in other rare SNPs (13). Note that 3 of the 21 haplotypes identified, namely NAT2*4, the cluster NAT2*5B, and the cluster NAT2*6A account for more than 90% of NAT2 haplotypes in white individuals. We did not identify 7 rare variant alleles previously described (13), designated as *5G, *5J, *6E, *7A, *11A, *11B, and *14B.
|
|
One of the aims of the present study was to analyze the frequency of ambiguous haplotypes. One type of ambiguous haplotype, as in the case of NAT2*12A, *12B, *12C, *12D, and *13, contains mutations, but the variant alleles do not lead to a slow acetylation status. If one of these mutated-rapid haplotypes occurs in combination with a slow allele, the individuals may be misclassified as slow acetylators. According to our findings, these mutated-rapid haplotypes account for 75 of the 2624 genes analyzed (2.8%). Sixty individuals of the 1312 involved in the study (4.6%) carried such mutated-rapid alleles in combination with a slow allele.
A second type of ambiguous allele also exists. Because most genotyping procedures simultaneously analyze both alleles, paternal and maternal, for each individual, certain haplotypes containing rare SNP combinations can be ambiguous when they occur in heterozygosity with a rapid allele, a situation that may lead to misclassification. An example is the NAT2*5E allele which, in addition to SNPs at 341 typically occurring in NAT2*5 alleles, contains SNPs at 590 that typically occur in NAT2*6 alleles. Likewise, the variants NAT2*6C, *6F, *14C, *14D, and *14F may lead to misclassification when these occur in combination with a rapid allele. Of the 2624 genes studied, 23 (0.9%) fulfilled such criteria, and 4 individuals in the study group (0.3%) carried these ambiguous haplotypes in combination with a rapid allele. Taking together both types of ambiguous haplotypes, our data indicate that 98 alleles (3.7%) and 64 individuals (4.9%) may be misclassified by NAT2 genotyping techniques.
We identified 51 diplotypes, although the number of genotypes was higher (n = 56), because 5 diplotypes corresponded to 2 different genotypes. In most cases, genotype reconstruction fully corresponded to actual genotypes, indicating that PHASE is a suitable tool to clarify haplotype ambiguities. In some cases, however, such as the diplotypes 0100010, 1000010, 1011010, 1012020, 1021020, 1100100, and 1111110 noted in Table 1
, the most probable predicted genotypes did not correspond to the actual ones. Eleven individuals (0.8%) were not correctly classified by the most probable genotypes predicted by PHASE. For this reason, and because haplotype reconstruction programs may not be readily available in some clinical laboratories, we included in Table 1
all the diplotypes identified in the study and their corresponding genotypes. These data can be used as a reference for clinical laboratories, to convert diplotypes into genotypes and to estimate the frequency for the different NAT2 genotypes.
Our data show that ambiguous haplotypes and diplotypes are common in the population analyzed, and that actual NAT2 genotypes cannot be fully determined by haplotype prediction techniques. Furthermore, this study allows a validation of haplotype prediction techniques, i.e., to determine the correspondence of predicted and actual haplotypes. The ambiguity of the haplotype reconstruction with PHASE v2.1.1 was 1.2% in this study and 0.1% in a previous study (14). This difference is attributable to the variant alleles *14, which were uncommon in 2 previous studies (14)(21), and are the major source of discordance between predicted and actual haplotypes in the present study. The frequency for carriers of variant alleles *14 in the study population analyzed here was statistically significantly different (
2 test, P < 0.001) from that seen in previous studies (14).
As a general rule, haplotype prediction techniques seem to be adequate, but exceptions exist. Novel data presented in this study constitute a useful guide for converting diplotypes obtained by any routine method into genotype data without the need for a haplotype reconstruction program. These data may simplify and improve the interpretation of NAT2 gene analyses regardless the genotyping method used.
Acknowledgments
Grant/Funding Support: This work was supported in part by grants FIS 05/1056, 06/1252 and RETICS RD07/0064/0016 from Fondo de Investigación Sanitaria, Instituto de Salud Carlos III, Spain and FUNDESALUD, Mérida, Spain.
Financial Disclosures: None declared.
Acknowledgment: We are grateful to Professor James McCue for assistance in language editing.
Footnotes
1 Nonstandard abbreviations: NAT2, N-acetyltransferase 2; SNP, single-nucleotide polymorphism. 
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
