¹Laboratory for Cancer Medicine, Western Australian Institute for Medical Research, Royal Perth Hospital, Perth, Western Australia

²Department of Pharmacology, University of Western Australia, Nedlands, Western Australia

³Department of Pharmacology, University of Oxford, Oxford, UK

Correspondence to: Dr N J Butcher, Laboratory for Cancer Medicine, Level 6, Medical Research Foundation Building, Royal Perth Hospital, Perth, Western Australia 6000 Tel: +61 8 9224 0338 Fax: +61 8 9224 0322 E-mail: nbutcher@receptor.pharm.uwa.edu.au

The arylamine N-acetyltransferases (NATs) are involved in the metabolism of a variety of different compounds that we are exposed to on a daily basis. Many drugs and chemicals found in the environment, such as those in cigarette smoke, car exhaust fumes and in foodstuffs, can be either detoxified by NATs and eliminated from the body or bioactivated to metabolites that have the potential to cause toxicity and/or cancer. NATs have been implicated in some adverse drug reactions and as risk factors for several different types of cancers. As a result, the levels of NATs in the body have important consequences with regard to an individual's susceptibility to certain drug-induced toxicities and cancers. This review focuses on recent advances in the molecular genetics of the human NATs.

The Pharmacogenomics Journal (2002) 2, 30-42. DOI: 10.1038/sj/tpj/6500053

N-acetyltransferase; NAT; acetylation; arylamine; polymorphism

Introduction

The arylamine N-acetyltransferases (NATs) are found in nearly all species from bacteria to humans. They catalyse the acetyltransfer from acetylcoenzyme A to an aromatic amine, heterocyclic amine or hydrazine compound. In humans, acetylation is a major route of biotransformation for many arylamine and hydrazine drugs, as well as for a number of known carcinogens present in the diet, cigarette smoke and the environment.^1,2,3,4 The reaction pathway is catalysed by two cytoplasmic acetyltransferases (NAT; EC 2.3.1.5), N-acetyltransferase Type I (NAT1) and N-acetyltransferase Type II (NAT2). The genes encoding both proteins were first isolated in 1989 by Grant et al who showed that each consists of an intronless open reading frame of 870 base pairs.⁵ The two genes are 87% homologous and are located at 8p22,^6,7,8 a chromosomal region commonly deleted in human cancers.^{9,10,11,12,13} Ohsako and Deguchi isolated the transcript for each gene from a human liver cDNA library in 1990.¹⁴ Sequencing of NAT1 and NAT2 revealed a number of allelic variants that affect activity of both genes in vivo. This work provided a genetic understanding for the long known functional polymorphism in NAT2 activity¹⁵ and, more recently, in NAT1 activity.^16,17,18 Genetic variation modulates the acetylator status of individuals and therefore may impact upon their predisposition to toxicity and disease. This review focuses on the genomics of the arylamine N-acetyltransferases and the impact of genetic variation of the enzymes on drug response in humans.

THE ARYLAMINE N-ACETYLTRANSFERASE GENE FAMILY

To date, 22 NAT-like genes have been identified in 14 different prokaryotic and eukaryotic species, although it is likely that additional genes will be discovered as more genomes become sequenced. The genes invariably have an intronless coding sequence that encodes for a protein between 254 and 332 amino acids in length (Figure 1). Figure 1 shows a multiple sequence alignment for 22 NAT proteins. The highest conserved regions occur at the amino terminus, whereas the carboxyl terminus shows very little conservation between species. Consistent with the recently published crystal structure for the Salmonella typhimurium NAT,¹⁹ all NATs possess a conserved cysteine, histidine and aspartate that have been implicated to form a catalytic triad. Inhibitor studies²⁰ and site-directed mutagenesis studies²¹ have confirmed that the cysteine (Cys⁶⁸ in the human proteins) is crucial for NAT activity.

A phylogenetic tree for the NAT proteins is shown in Figure 2. It indicates the separate clustering of the prokaryotic and eukaryotic sequences, with the exception of A. mediterranei NAT which is distant from both groups. This enzyme is part of the rifamycin B synthesis pathway and catalyses the final amide bond formation reaction.²² The NAT1 and NAT2 sequences for rat, mouse and hamster cluster together suggesting that the two proteins are encoded by genes that were present before the divergence of the three rodent species. By contrast, the two human proteins are more closely related to each other, perhaps because the genes duplicated later in evolution. The general relationship of the amino acid sequences is similar to that recently published for a limited number of NAT nucleotide sequences.²³

STRUCTURAL CHARACTERISTICS OF THE NAT PROTEINS

The first crystal structure of an arylamine N-acetyltransferase was recently published.¹⁹ Although the NAT was of bacterial origin, it revealed a number of surprising features that provided novel structural and functional information about the enzyme.¹⁹ Specifically, a cysteine-histidine-aspartate catalytic triad was identified in the N-terminus of the protein. Based on structural analysis, the protein has been divided into three domains. The first consists of a helical bundle, located from amino acid 1 to approximately 90 (based on numbering in Figure 1), which forms one side of a cleft in which the cysteine involved in acetyl transfer resides. All NATs are highly homologous in this region. The second domain consists of residues from approximately 90 to 210 and is located on the other side of the cleft. It mostly consists of -sheet structures. The last domain at the carboxyl terminus is a combination of -sheets and -helices, and this region shows the greatest diversity between species. The structural features surrounding the triad are similar to the cysteine protease superfamily of proteins which includes the transglutaminases, cathepsins and caspases. While these proteins traditionally catalyse the hydrolysis of amide substrates, the NATs and the transglutaminases catalyse an acyltransfer that results in amide bond formation. This is also the case for the NAT homolog found in A. mediterranei.²² To date, the crystal structure of the human NATs has not been resolved although the homology with NAT from bacteria suggests similar features will be present.

N-ACETYLTRANSFERASE NOMENCLATURE

The first attempt to devise a consensus nomenclature for the NATs was published in 1995 and included genes from six species and a total of 39 alleles.¹⁵ Since then, many new alleles have been identified and the nomenclature for these was updated in 2000.²⁴ The Human Gene Nomenclature Committee has agreed that the symbol NAT be assigned to the arylamine N-acetyltransferase genes. Currently, NAT also is used for unrelated genes such as yeast protein N-terminal acetyltransferase, human noradrenalin transporter, eukaryotic translation initiation factor, translation repressor protein and death associated protein 5.

The classification of the eukaryotic genes into NAT1 and NAT2 subfamilies has been done largely on a historical basis of the research area and is a consensus nomenclature.²⁴ While this has been appropriate for most members of the genes for NAT, it has raised some confusion with the mouse, hamster and rat Nat2 genes, which encode for proteins with substrate specificity similar to human NAT1 (acetylate p-aminobenzoic acid;^25,26,27,28). At this stage, mouse is the only species to have three genes for NAT, pseudogenes excepted. Classification of different alleles into different clusters is based on the most significant nucleotide substitution present.²⁴

The prokaryotic genes are sufficiently dissimilar to the eukaryotic genes for NAT to preclude subclassification. Consequently, the prokaryotic genes are referred to as Nat only. An international committee has been established to oversee the nomenclature of the N-acetyltransferases. The committee is responsible for nomenclature updates and assignment of new alleles. This has been found to be essential to ensure consistence of allelic names in the literature. A Web site that provides information about the naming of existing and new alleles for all species can be found at http://www.louisville.edu/medschool/pharmacology/NAT.html.

REACTION MECHANISM AND SUBSTRATE SPECIFICITY

The arylamine N-acetyltransferases differ from the many other acetyl coenzyme A-dependent transferases present in cells because of their ping-pong bi bi reaction mechanism.^29,30 The reaction takes place in two separate steps. Initially, acetyl coenzyme A binds to the enzyme and the acetyl moiety is transferred from the cofactor to a cysteine (Cys⁶⁸ for the human isoforms) of the protein. Coenzyme A is then released. The second step involves the binding of substrate to the acetylated enzyme following which the acetyl moiety is transferred to the substrate. Finally, the acetylated product is released from the enzyme. The first step of the reaction can proceed in the absence of arylamine substrate.³¹ It may be that the enzyme exists in an acetylated state in the cell when no substrate is present, although this has not been shown to date, and the stability of the acetylated intermediate should be assessed.

There is no clear structural motif that determines substrate specificity for the different isoforms of NAT. In general, p-aminobenzoic acid (PABA), p-aminobenzoyl glutamate and p-aminosalicylic acid (PAS) are considered specific substrates for human NAT1 (or mouse NAT2). These substrates can be characterised by the presence of relatively small hydrophilic substitutions in the para position of the aromatic ring. By contrast, sulfamethazine, procainamide and dapsone are acetylated primarily by human NAT2. Some compounds such as 2-aminofluorene are excellent substrates for both human NAT1 and NAT2.³

The prokaryotic NATs acetylate PABA poorly or not at all and their range of substrates, which includes isoniazid, suggests that they are functionally similar to the human NAT2 isoform.³² However, the prokaryotic NATs are also capable of acetylating the human NAT1 specific substrate 5-aminosalicylic acid,³³ indicating that prokaryotic NATs should not be considered functionally equivalent to either NAT1 or NAT2 in humans. In some bacterial strains such as H. pylori, NAT activity has been reported although there is no evidence of a NAT-like gene in their genome. The very high K_m values reported for substrates like PABA and 2-aminofluorene indicate that the acetylation reaction is very inefficient and is probably catalysed by a protein unrelated to the NATs.³⁴

One prokaryotic NAT homolog deserves further attention. The Rif gene cluster in A. mediterranei that encodes the proteins necessary for the synthesis of rifamycin B includes the gene RifF for rifamycin amide synthase. The amino acid similarity of RifF to the other prokaryotic NATs is approximately 45% (Figure 1) and the enzyme possesses the conserved catalytic triad in its amine terminus. RifF catalyses an amide bond formation during the latter stages of rifamycin synthesis (Figure 3).²² Other NAT-like proteins may become evident as more genomes are sequenced and it is possible that the NAT protein has evolved in different species to undertake quite different cellular tasks.³²

GENE LOCALISATION, STRUCTURE AND EXPRESSION

Two NAT isoenzymes have been identified in humans, namely NAT1 and NAT2, which are the products of distinct genetic loci, designated NAT1 and NAT2, respectively.⁶ A related pseudogene, NATP1, has also been identified,⁶ which contains multiple frameshift and nonsense mutations. The two functional NAT genes share an 87% nucleotide identity, which translates to an 81% homology at the amino acid level. While the entire transcript of NAT1 is derived from a single exon, that of NAT2 is derived from the protein encoding exon together with a second noncoding exon of 100 bp located about 8 kb upstream of the translation start site.^6,35 Human NAT1 and NAT2, as well as NATP1, have been localised to the short arm of chromosome 8,^6,36 more specifically in region 8p22.⁸ The NAT loci are separated by only 170-360 kb and are in the orientation NAT1 NATP1 NAT2, with NAT1 being on the centromeric side of marker D8S261 and NAT2 coinciding with marker D8S21.³⁶ Both NAT1 and NAT2 genes display pronounced allelic variation, with 26 different human NAT1 and 29 different human NAT2 alleles identified to date.²⁴

A similar situation is present in the mouse where three functional genes, designated Nat1, Nat2 and Nat3, have been identified to encode for NAT isoenzymes.^37,38 Mouse NAT2 shows similar substrate specificity with human NAT1, while mouse NAT1 is capable of metabolising the human NAT2-specific substrate isoniazid.³⁹ To date, no specific substrate has been identified for NAT3, although the encoding locus appears to be functional.^38,40 The localisation and genomic organisation of the mouse Nat genes is similar to their human counterparts. The three genes are clustered together in a 130-kb genomic region on mouse chromosome 8, cytogenetic band B3.1-B3.3⁴¹ and within a genetic distance of about 31 cM from the centromere.⁴² This chromosomal region is syntenic with the region harbouring the genes for human NAT on chromosome 8p. Polymorphism has been detected only in the Nat2 gene of both A/J³⁷ and A/HeJ⁴³ inbred mice, in the form of a missense A T mutation at position 296 of the open reading frame, causing the slow acetylator phenotype. It is of interest that mouse Nat2 also possesses a short non-coding exon, located about 6 kb upstream of the intronless open reading frame.⁴¹ This raises the possibility that mouse Nat2 may be the genetic orthologue of the human NAT2 gene, although mouse NAT2 and human NAT1 proteins appear to be functionally equivalent.

Information about the localisation and genomic organisation of the Nat genes in other eukaryotic species has been limited. All genes have an intronless open reading frame and polymorphism in NAT activity has been described in strains of rabbits,⁴⁴ hamsters⁴⁵ and rats,⁴⁶ all of which possess two Nat genes. Upstream non-coding exons have also been described for the rabbit Nat genes.⁴⁷ Cats and other felids only have one gene for NAT,⁴⁸ while dogs lack NAT activity, due to absence of Nat genes in their genome.⁴⁹

HUMAN NATs

Human NAT2 Alleles

Since the human NAT2 locus was established as the site of the classical acetylation polymorphism,^50,51 the study of NAT2 allelic variation has been an area of intense investigation. To date, 29 different NAT2 alleles have been detected in human populations (Table 1; reviewed in ^15,24,52). Each of the variant alleles is comprised of between one and four nucleotide substitutions, of which 13 have been identified, located in the protein encoding region of the gene. Nine of these lead to a change in the encoded amino acid (C¹⁹⁰T, G¹⁹¹A, T³⁴¹C, A⁴³⁴C, G⁴⁹⁹A, G⁵⁹⁰A, A⁸⁰³G, A⁸⁴⁵C, and G⁸⁵⁷A), while the remaining four are silent (T¹¹¹C, C²⁸²T, C⁴⁸¹T, and C⁷⁵⁹T).

Several studies have been performed that show clear correlations between NAT2 genotype and phenotype.^53,54,55 Early genotyping studies screened for the presence of the C⁴⁸¹T (M1), the G⁵⁹⁰A (M2), the G⁸⁵⁷A (M3) and sometimes the G¹⁹¹A (M4) nucleotide changes, all of which were shown to cause a slow acetylation phenotype. Moreover, there was a gene-dosage effect. Individuals who were homozygous for NAT2 polymorphisms had a slow acetylator phenotype, individuals heterozygous for NAT2 polymorphisms had an intermediate acetylator phenotype, and individuals who lacked NAT2 polymorphisms had a rapid acetylator phenotype. It should be noted that the method of detection of the above polymorphisms only identifies a subset of the variant alleles found in human populations, and there is potential for the misclassification of genotype and deduced phenotypes (reviewed in ⁵⁶).

Initial studies in liver tissue suggested that the slow acetylator phenotype associated with the presence of certain nucleotide substitutions in the protein encoding region of the NAT2 gene was due to a marked decrease in NAT2 protein content, while NAT2 mRNA levels remained unchanged.⁵⁷ Several studies have since investigated the mechanism by which nucleotide substitutions in the NAT2 gene affect acetylation capacity by the use of recombinant expression systems.^{58,59,60,61,62} Hein and coworkers⁶³ performed a comprehensive study that assessed the acetylation capacity of 16 different NAT2 alleles in a bacterial expression system. Of the seven specific NAT2 substitutions that they examined, the T³⁴¹C, G⁵⁹⁰A, G⁸⁵⁷A, and G¹⁹¹A substitutions produced recombinant NAT2 allozymes with reduced acetylation capacities, while the C⁴⁸¹T, C²⁸²T, and A⁸⁰³G substitutions produced recombinant NAT2 allozymes with acetylation capacities similar to the wild-type NAT2 4 protein. As a result, NAT2 alleles that contain any of the specific substitutions that produced recombinant NAT2 allozymes with reduced acetylation capacities are associated with a slow acetylator phenotype, and include the NAT2 5, NAT2 6, NAT2 7, NAT2 14, and NAT2 17 clusters (see Table 1).

The molecular mechanisms responsible for the production of the slow acetylator phenotypes are not well understood at present. Some base changes appeared to cause a slow acetylation phenotype by producing an unstable protein. NAT2 allozymes encoded by alleles with base substitutions at positions 191, 590, or 857 were found to be significantly more unstable in bacterial expression systems than the wild-type protein.^52,61,64 However, in these studies the amount of immunodetectable NAT2 protein was not different upon expression of the variant and wild-type alleles. This is in contrast to the earlier observations by Grant and coworkers⁵⁷ who showed that liver NAT2 content was markedly reduced in slow acetylators, suggesting that the artificial environment of bacterial expression systems may not accurately reflect what occurs in mammalian cells with regard to protein degradation.

Recently, a study by Leff and coworkers⁶² characterised several different human NAT2 alleles in a yeast expression system. They found that three novel alleles, namely NAT2*5D (T³⁴¹C), NAT2*14G (G¹⁹¹A, C²⁸²T, and A⁸⁰³G), and NAT2*6D (C¹¹¹T, C²⁸²T, and G⁵⁹⁰A), expressed proteins that had N- and O-acetylation capacities similar to the expressed protein of the commonly occurring slow NAT2*5B allele, and significantly less than that of the wild-type NAT2*4 allele. The expression of NAT2 5B and NAT2 5D was found to be significantly lower than that of the wild-type protein, suggesting that the base substitution at position 341, which is common to the NAT2*5 cluster, is sufficient for reduction in NAT2 protein expression. This was not found to be the case for NAT2 6D and NAT2 14G, which were expressed at levels comparable to wild-type. By contrast, NAT2 6D and NAT2 14G were found to be significantly less stable than wild-type.

The frequency of the slow acetylator phenotype varies considerably among ethnic groups,⁶⁵ and this is due to the differing frequencies of the polymorphisms that correspond to the slow acetylator alleles. In Caucasian and African populations, the frequency of the slow acetylation phenotype varies between 40 and 70%, while that of Asian populations, such as Japanese, Chinese, Korean, and Thai, range from 10 to 30% (reviewed in ⁶⁶). Caucasian and African populations have high frequencies of NAT2*5 alleles (>28%) and low frequencies of NAT2*7 alleles (<5%), while Asian populations have low incidences of NAT2*5 alleles (<7%) and higher incidences of NAT2*7 alleles (>10%). Also, NAT2*14 alleles are almost absent from Caucasian and Asian populations (<1%), but are present in African populations at comparably higher frequencies (>8%).

Human NAT1 Alleles

Historically, NAT1 was thought to be genetically invariant or monomorphic in nature. However, wide inter-individual variability in NAT1 activity towards PABA or PAS^{51,67,68,69,70,71,72} was suggestive of a genetic polymorphism, but NAT1 activities were generally unimodally distributed. It wasn't until 1993 when Vatsis and Weber⁷³ first reported the existence of several allelic variations at the NAT1 locus that interest in the NAT1 gene was aroused, marking the beginning of a systematic survey of NAT1 genotypes. To date, 26 different NAT1 alleles have been detected in human populations (Table 2; reviewed in ^15,24,52), however, only a small number have been shown to alter phenotype in vivo. Hughes and coworkers¹⁷ used PAS as a probe drug to phenotype a population for NAT1 activity. By measuring urinary metabolite ratios, they were able to detect individuals with marked impairments of NAT1 function. However, there was only a moderate correlation between phenotypes determined by in vivo and in vitro methods, and the authors themselves suggest that less than 50% of the phenotypic variation observed in vivo was related to variation in NAT1 function. It appears that the measurement of NAT1 activity of blood cells is the most reliable method of phenotyping for NAT1. While little is known about the relative expression of NAT1 in various human tissues, studies in the rabbit model suggest that NAT1 activity is comparable in most tissues, including blood cells.¹⁷ Therefore, it is reasonable to assume that NAT1 activity of blood cells is a good surrogate of systemic activity in humans.

The first report of a correlation between NAT1 genotype and phenotype was by Bell and coworkers in 1995.⁷⁴ They showed that the NAT1*10 allele was associated with activity two-fold higher than that of the wild-type allele, NAT1*4, in bladder and colon tissue samples.⁷⁴ In the bladder, higher levels of DNA adducts were detected in NAT1*10 heterozygotes compared with NAT1*4 homozygotes.^75,76 The NAT1*10 allele also has been associated with a marginally elevated activity in liver samples⁴ and erythrocytes.⁷⁷ NAT1*10 has no mutations in the protein encoding region of the gene, but contains two nucleotide substitutions (T¹⁰⁸⁸A and C¹⁰⁹⁵A) in its 3'-untranslated region. The T¹⁰⁸⁸A base change alters the consensus polyadenylation signal (AATAAA AAAAAA) leading to the suggestion that increased activity may be due to enhanced mRNA stability.^74,78 However, several recent studies,^{16,17,18,79,80} do not support the idea that the NAT1*10 allele is associated with elevated NAT1 activity. As a result, the functional significance of this allele remains unclear at present.

A population study¹⁶ showed a distribution of NAT1 activity that was clearly bimodal in nature, with 8% of the individuals being slow acetylators. Moreover, the above study was one of the first to report a correlation between NAT1 genotype and phenotype involving the slow acetylator alleles NAT1*14 and NAT1*17. Individuals that were heterozygous for either polymorphism had approximately half the activity of individuals that lacked these base changes. Furthermore, Western blots for NAT1 showed that low activity was due to a parallel decrease in NAT1 protein content, indicating that slow acetylator status was a result of a decrease in the amount of a functionally normal enzyme rather that the presence of a protein with altered acetylation capacity. A later study also found a correlation between NAT1 phenotype and the low activity NAT1*14 allele,⁷⁹ with heterozygote carriers of the allele having about 50% of the activity of noncarriers. Bruhn and coworkers⁷⁹ also found that individuals who possessed a NAT1*11 allele had slightly lower activities compared with individuals who were homozygous for NAT1*4, NAT1*10, or NAT1*3, all of which had similar activities. Interestingly, the same study identified an individual who was homozygous for the NAT1*15 allele and who had no measurable NAT1 activity. The NAT1*15 allele contains a base substitution (C⁵⁵⁹T) in the protein encoding region of the NAT1 gene that introduces a stop codon, leading to the production of a truncated, inactive protein.¹⁷ Hughes and coworkers¹⁷ also identified an individual who possessed two low activity alleles, namely NAT1*14B/NAT1*15, and who subsequently had very low acetylation capacity. As with several NAT2 low activity alleles, there appears to be a gene-dosage effect for the low activity NAT1 alleles, with heterozygotes having about half the activity of NAT1*4 wild-type individuals, and homozygotes (or compound heterozygotes) for low activity alleles having little or no NAT1 activity. The exception is a NAT1*14A/NAT1*14B heterozygote identified by Payton and Sim,⁷⁷ whose activity was less than the NAT1*4 homozygotes but was still detectable. The frequency of slow acetylator alleles for NAT1 is low. The most common low activity allele, NAT1*14, has been identified in Caucasian populations ranging from 1.3 to 3.7%.^{16,17,18,79,81,82} Interestingly, a much higher frequency of the NAT1*14 allele (25%) was reported for a Lebanese population.⁸³ Since no homozygous individuals were identified in the above study, 50% of the Lebanese population had a slow acetylator genotype. This indicates that NAT1, like NAT2, shows considerable interethnic variability.

Some of the more common variant NAT1 alleles have been characterised in bacterial and/or mammalian expression systems. Recombinant expression of NAT1*14 in a bacterial system by Hughes and coworkers¹⁷ showed that the variant protein had a 15- to 20-fold decrease in affinity for the substrate PAS and showed a 4-fold decrease in V_max compared with recombinant NAT1 4 wild-type protein. The same study also showed that expression of NAT1*15 in E.coli produced a truncated protein that was totally inactive. Therefore, NAT1*14 (A and B) and NAT1*15 are low activity alleles, which is consistent with phenotyping studies using human blood cells.^16,79,81 Two other low activity alleles, NAT1*17 and NAT1*22, have been expressed in bacterial systems and both produced variant proteins that had no detectable NAT1 activity towards PAS.¹⁸ Also, in the same study, immunoreactive NAT1 17 and NAT1 22 protein levels were markedly decreased compared with wild-type NAT1 4 protein levels. NAT1*19 was classified as a nonfunctional allele because the base substitution (C⁹⁷T) introduces a premature stop codon.¹⁸

The effects of coding and 3'-noncoding polymorphisms in the NAT1*11 allele were characterised by de Leon and coworkers⁸⁰ recently. Using recombinant expression of NAT1*11 in both bacterial and mammalian systems, they showed that no major differences existed in catalytic or other properties of the NAT1 11 protein compared with wild-type NAT1 4 protein. This is in agreement with an earlier study,¹⁷ which showed that the activity of recombinant NAT1 4 and NAT1 11 were similar, but is in contrast to another study which reported a slightly reduced activity of blood cells from individuals who carried the NAT1*11 allele.⁷⁹ de Leon and coworkers,⁸⁰ using a mammalian expression system, have recently shown that the NAT1*16 phenotype is caused by polymorphism in the 3'-untranslated region that leads to a decrease in protein expression. NAT1*16 has a triple adenosine insertion on the 3' side of the polyadenylation signal (AATAAA) which significantly alters the secondary structure of the pre-mRNA, leading to a 2-fold reduction in the amount of NAT1 16 protein and activity, compared with NAT1 4 and NAT1 10.

Some NAT1 alleles also may produce proteins with activities that are higher than that of the wild-type protein NAT1 4. Recombinant expression of NAT1*21, NAT1*24, and NAT1*25 in bacterial systems produced allozymes with activities 2- to 3-fold higher than NAT1 4.¹⁸ However, the levels of immunoreactive protein expressed by NAT1*4, NAT1*21, NAT1*24, and NAT1*25 were equivalent.¹⁸ The functional significance of the other NAT1 variants remains unclear at present.

NAT AND DISEASE

The association between acetylator status and the risk of various diseases has been extensively reported, and reviewed in detail.^84,85,86,87 Altered risk with either the slow or rapid phenotype has been observed for bladder, colon and breast cancer, systemic lupus erythematosis, diabetes, Gilbert's disease, Parkinson's disease and Alzheimer's disease. These associations imply a role for environmental factors that are metabolised by the NATs, in particular NAT2, in each disorder. However, identifying those factors has remained elusive. Humans are exposed to many toxic NAT substrates including the food-derived heterocyclics present in the diet as well as arylamines such as 4-aminobiphenyl and -naphthylamine present in tobacco smoke.^88,89,90,91 Moreover, occupational exposure to arylamine carcinogens such as benzidine has also been reported.^92,93

Because of the role of acetylation in the metabolic activation and detoxification of arylamine and heterocyclic carcinogens, acetylator status and cancer risk has been widely investigated. Unlike the relatively rare but highly penetrant genes involved in familial cancers, those genes responsible for metabolic polymorphisms have low penetrance and cause only a moderate increase in cancer risk. Nevertheless, their widespread occurrence in the general population suggests they are a significant contributor to individual risk. In 1979, Lower et al⁹⁴ first demonstrated an association between the slow acetylator phenotype and bladder cancer. This work was followed by many independent studies. However, few diseases have consistently demonstrated a relationship between phenotype and risk. For example, several studies have implicated the rapid phenotype as an increased risk factor for colon cancer,^95,96,97 whereas others have been unable to confirm this finding.^98,99,100 Geographical differences, ethnicity, lack of study power, dietary differences and differences in other risk factors between study groups have been suggested as reasons for variable results from independent laboratories. Recent reports suggesting that NAT activity may be altered by environmental factors and substrate-dependent down-regulation¹⁰¹ may also explain why inconsistent associations have been seen.

When acetylator phenotype has been linked to carcinogen exposure, more consistent results have been reported. For example, the rapid phenotype has emerged as a strong risk factor for colorectal cancer in those individuals who have a higher exposure to the food-derived heterocyclic amines.^99,100,102 These observations provide strong circumstantial evidence that the heterocyclic amines have an important role in colorectal cancer, and extensive animal studies support this.^{103,104,105,106} They also illustrate the importance of establishing associations between genetic polymorphisms and disease risk. From such studies, it should be possible to pursue the causative agent(s) of the disease where no obvious candidate agent is evident.

Recently, the NAT2 acetylator phenotype has been linked to increased risk associated with neurodegenerative diseases such as Parkinson's disease and Alzheimer's disease.^107,108,109 For late-onset Alzheimer's disease, an odds ratio of 3.0 (95% CI 1.3-7.3) has been reported for the rapid phenotype in non-apoE epsilon carriers.¹⁰⁹ By contrast, the slow phenotype appears to increase risk of Parkinson's disease with an odds ratio of 3.58 (95% CI 1.96-6.56).¹⁰⁷ Although these results need to be confirmed with larger epidemiological studies, they point to environmental factors that are substrates for the NATs having a role in the onset of these diseases. Alternatively, the different alleles for NAT2 that produce the rapid or slow phenotype may co-segregate with unrelated genes that are the causative agent for the different neurodegenerative diseases.

NAT and Drug Response

The genetic polymorphism in N-acetyltransferase activity was first discovered in patients treated with isoniazid for tuberculosis.¹¹⁰ This drug is primarily excreted following acetylation catalysed by NAT2. Since then, many therapeutic agents have been shown to be polymorphically acetylated in humans. These include hydralazine, procainamide, sulphamethazine, endralazine, a number of sulphonamides, nitrazepam and dapsone. However, the incidence of failed or less effective clinical response as a consequence of acetylation polyphorphism is uncommon. This is because most drugs that are metabolised by the NATs have a wide therapeutic window or because acetylation is a minor metabolic pathway. An exception is hydralazine. Early studies showed that the antihypertensive activity of hydralazine was less in rapid acetylators and that a 40% higher dose was necessary for a similar therapeutic effect compared with slow acetylators.¹¹¹ This difference appeared to be due to a change in the bioavailability of the drug which decreased from 33% in slow acetylators to less than 10% in rapid acetylators.¹¹²

A more common consequence of the polymorphic acetylation of therapeutic agents is an increase in the frequency and severity of side-effects associated with either the rapid or slow phenotype (Table 3). These adverse effects often arise as the result of a shift in the metabolic pathways responsible for the activation and detoxification of the drug. This is best illustrated by the amine-containing sulphonamides, such as sulphamethoxazole, that undergo hydroxylation to a reactive N-hydroxy metabolite capable of covalently binding to macromolecules and giving rise to idiosyncratic adverse reactions.¹¹³ These drugs can also be acetylated by NAT2 to non-reactive N-acetyl metabolites. In slow acetylators, a higher proportion of the drug is N-hydroxylated and consequently, these individuals are at a greater risk of sulphonamide-induced toxicity.^114,115,116 However, as pointed out by Spielberg, the incidence of severe adverse side effects to sulphonamides is much less than the incidence of the slow acetylator phenotype suggesting that other factors predispose individuals to idiosyncratic adverse reactions.¹¹³

Risk of developing side effects, such as neurotoxicity or haemolytic anemia, to dapsone therapy is very similar to that described for the sulphonamides.¹¹⁷ The most severe incidence of toxicity occurred in individuals with a slow acetylator phenotype who are rapid hydroxylators, which is consistent with the role each pathway has in the activation and detoxification of the drug.¹¹⁸

While slow acetylators are at a greater risk of toxicity from sulphonamides and dapsone, other therapeutic agents exhibit increased incidence of adverse reactions in rapid acetylators. Amonafide is a novel arylamine that has previously been used in clinical trials for the treatment of various cancers. It undergoes N-acetylation to an active metabolite that contributes to systemic toxicity. Several studies have shown that myelosuppression is greater in rapid acetylators (white blood cell nadirs of 500 l^-1) compared to slow acetylators (white blood cell nadirs of 3400 l^-1) following a standard dose of 300 mg m^-2 daily for 5 days.^119,120,121. This has led to different recommended doses for the two groups.^120,122

The recent discovery and cloning of prokaryotic NATs has raised the possibility that bacterial metabolism of drugs and other xenobiotics can contribute to their therapeutic and toxicological efficacy in vivo.^33,123 Payton and associates showed that M. smegmatis transformed with the M. tuberculosis NAT gene has a 3-fold higher resistance to isoniazid due to an increase in acetylation of the drug.¹²³ These observations suggested that the level of NAT expression in target bacteria may be an important therapeutic modifier for antibiotics that are extensively acetylated. In addition, polymorphisms in the bacterial NAT genes could lead to different therapeutic responses.

Okumura et al¹²⁴ found that the acetylated metabolites of a range of arylamines such as p-aminobenzoic acid, 4-aminobiphenyl and 1-aminopyrene were excreted in the urine and feces of dogs that lack N-acetyltransferase activity.⁴⁹ They showed that the intestinal microflora were responsible for the formation of the acetyl derivatives. Similarly, the microflora in the intestine of rats contribute to the acetylation of 2-nitrofluorene and the formation of DNA adducts in liver, kidney, lung and heart following oral administration.¹²⁵ Taken together, these studies suggest that bacterial NAT has a role in the activation and detoxification of xenobiotics in the host organism and may play an important role in the metabolism of anti-inflammatory drugs, such as 5-aminosalicylic acid.³³

CONCLUDING REMARKS

Although considerable allelic variation exists for both NAT1 and NAT2, our understanding of the molecular mechanisms and functional significance of many of these alleles, particularly for NAT1, is still limited. The majority of functional studies to date have been performed in bacterial expression systems, and the results of such studies may not necessarily accurately reflect what occurs in vivo, due to differences in degradation/processing pathways between bacterial and mammalian systems. Discrepancies between the two systems have been reported with regard to NAT protein content and slow acetylator alleles, and characterisation using mammalian systems may provide a better understanding of the molecular mechanisms leading to different NAT phenotypes.

Much of the research in the area of NATs has involved identifying relationships between allele frequencies and disease, particularly different forms of cancers. Although several studies have reported associations between different NAT alleles and various cancers, other studies have failed to do so. While these inconsistencies may be due to several factors, such as differences in exposure to arylamine carcinogens, it may well be that genotype does not necessarily accurately reflect phenotype. For example, although the regulation of NAT1 expression certainly has a genetic component, this only accounts for part of the observed variability in NAT1 activities. We have shown that significant variation in NAT1 activity can be observed within a single phenotype and within the same individuals measured on different occasions, suggesting that a considerable part of the variation in activity is environmentally based. In support of this, we have recently provided evidence for substrate-dependent regulation of NAT1,¹⁰¹ and identified a minimum promoter sequence for the human NAT1 that consists of an AP-1-like motif flanked on either side by a TCATT sequence (manuscript in preparation). Transient transfection assays showed that both the AP-1 motif and the 3'-TCATT sequence were essential for basal promoter activity, while the 5'-TCATT sequence appeared to act as an attenuator of phorbol 12-myristate 13-acetate induction. Moreover, antibody supershift assays suggested that c-jun, Oct-1, and YY1 transcription factors form complexes with the NAT1 minimum promoter. Therefore, environmental factors that alter the expression of transcription factors, such as those mentioned above also may modulate the basal expression of NAT1 in vivo. There also is the possibility that other promoter sequences and binding motifs exist further upstream from the minimum basal promoter sequence that could modulate NAT1 activity under certain conditions.

DUALITY OF INTEREST

None declared.

ABBREVIATIONS

NAT1 arylamine N-acetyltransferase 1

NAT2 arylamine N-acetyltranferase 2

PAS p-aminosalicylic acid

PABA p-aminobenzoic acid

1 Hein DW, Doll MA, Rustan TD, Gray K, Feng Y, Ferguson RJ et al. Metabolic activation and deactivation of arylamine carcinogens by recombinant human NAT1 and polymorphic NAT2 acetyltransferases. Carcinogenesis 1993; 14: 1633-1638. MEDLINE

2 Kato R. Metabolic activation of mutagenic heterocyclic aromatic amines from protein pyrolysates. Crit Rev Toxicol 1986; 16: 307-348. MEDLINE

3 Minchin RF, Reeves PT, Teitel CH, McManus ME, Mojarrabi B, Ilett KF et al. N- and O-acetylation of aromatic and heterocyclic amine carcinogens by human monomorphic and polymorphic acetyltransferases expressed in COS-1 cells. Biochem Biophys Res Commun 1992; 185: 839-844. MEDLINE

4 Zenser TV, Lakshmi VM, Rustan TD, Doll MA, Deitz AC, Davis BB et al. Human N-acetylation of benzidine: role of NAT1 and NAT2. Cancer Res 1996; 56: 3941-3947. MEDLINE

5 Grant DM, Blum M, Demierre A, Meyer UA. Nucleotide sequence of an intronless gene for a human arylamine N-acetyltransferase related to polymorphic drug acetylation. Nucl Acids Res 1989; 17: 3978.

6 Blum M, Grant DM, McBride W, Heim M, Meyer UA. Human arylamine N-acetyltransferase genes: isolation, chromosomal localization, and functional expression. DNA & Cell Biol 1990; 9: 193-203.

7 Franke S, Klawitz I, Schnakenberg E, Rommel B, Van de Ven W, Bullerdiek J et al. Isolation and mapping of a cosmid clone containing the human NAT2 gene. Biochem Biophys Res Communi 1994; 199: 52-55.

8 Hickman D, Risch A, Buckle V, Spurr NK, Jeremiah SJ, McCarthy A et al. Chromosomal localization of human genes for arylamine N-acetyltransferase. Biochem J 1994; 297: 441-445. MEDLINE

9 Bova GS, Carter BS, Bussemakers MJ, Emi M, Fujiwara Y, Kyprianou N et al. Homozygous deletion and frequent allelic loss of chromosome 8p22 loci in human prostate cancer. Cancer Res 1993; 53: 3869-3873. MEDLINE

10 Maitra A, Tavassoli FA, Albores-Saavedra J, Behrens C, Wistuba II, Bryant D et al. Molecular abnormalities associated with secretory carcinomas of the breast. Hum Pathol 1999; 30: 1435-1440. MEDLINE

11 Lutchman M, Pack S, Kim AC, Azim A, Emmert-Buck M, van Huffel C et al. Loss of heterozygosity on 8p in prostate cancer implicates a role for dematin in tumor progression. Cancer Genet Cytogenet 1999; 115: 65-69. MEDLINE

12 Brown MR, Chuaqui R, Vocke CD, Berchuck A, Middleton LP, Emmert-Buck MR et al. Allelic loss on chromosome arm 8p: analysis of sporadic epithelial ovarian tumors. Gynecol Oncol 1999; 74: 98-102. Article MEDLINE

13 Wistuba II, Behrens C, Virmani AK, Milchgrub S, Syed S, Lam S et al. Allelic losses at chromosome 8p21-23 are early and frequent events in the pathogenesis of lung cancer. Cancer Res 1999; 59: 1973-1979. MEDLINE

14 Ohsako S, Deguchi T. Cloning and expression of cDNAs for polymorphic and monomorphic arylamine N-acetyltransferases from human liver. J Biol Chem 1990; 265: 4630-4634. MEDLINE

15 Vatsis KP, Weber WW, Bell DA, Dupret JM, Evans DA, Grant DM et al. Nomenclature for N-acetyltransferases. Pharmacogenetics 1995; 5: 1-17. MEDLINE

16 Butcher NJ, Ilett KF, Minchin RF. Functional polymorphism of the human arylamine N-acetyltransferase type 1 gene caused by C190T and G560A mutations. Pharmacogenetics 1998; 8: 67-72. MEDLINE

17 Hughes NC, Janezic SA, McQueen KL, Jewett MA, Castranio T, Bell DA et al. Identification and characterization of variant alleles of human acetyltransferase NAT1 with defective function using p-aminosalicylate as an in-vivo and in-vitro probe. Pharmacogenetics 1998; 8: 55-66. MEDLINE

18 Lin HJ, Probst-Hensch NM, Hughes NC, Sakamoto GT, Louie AD, Kau IH et al. Variants of N-acetyltransferase NAT1 and a case-control study of colorectal adenomas. Pharmacogenetics 1998; 8: 269-281. MEDLINE

19 Sinclair JC, Sandy J, Delgoda R, Sim E, Noble ME. Structure of arylamine N-acetyltransferase reveals a catalytic triad. Nat Struct Biol 2000; 7: 560-564. Article MEDLINE

20 Andres HH, Klem AJ, Schopfer LM, Harrison JK, Weber WW. On the active site of liver acetyl-CoA. Arylamine N-acetyltransferase from rapid acetylator rabbits (III/J). J Biol Chem 1988; 263: 7521-7527. MEDLINE

21 Dupret JM, Grant DM. Site-directed mutagenesis of recombinant human arylamine N-acetyltransferase expressed in Escherichia coli. Evidence for direct involvement of Cys68 in the catalytic mechanism of polymorphic human NAT2. J Biol Chem 1992; 267: 7381-7385. MEDLINE

22 Floss HG, Yu TW. Lessons from the rifamycin biosynthetic gene cluster. Curr Opin Chem Biol 1999; 3: 592-597. MEDLINE

23 Upton A, Johnson N, Sandy J, Sim E. Arylamine N-acetyltransferases of mice, men and microorganisms. Trends Pharmacol Sci 2001; 22: 140-146. MEDLINE

24 Hein DW, Grant DM, Sim E. Update on consensus arylamine N-acetyltransferase gene nomenclature. Pharmacogenetics 2000; 10: 291-292. MEDLINE

25 Hein DW, Doll MA, Rustan TD, Gray K, Ferguson RJ, Feng Y. Construction of Syrian hamster lines congenic at the polymorphic acetyltransferase locus (NAT2): acetylator genotype-dependent N- and O-acetylation of arylamine carcinogens. Toxicol Appl Pharmacol 1994; 124: 16-24. Article MEDLINE

26 de Leon JH, Martell KJ, Vatsis KP, Weber WW. Slow acetylation in mice is caused by a labile and catalytically impaired mutant N-acetyltransferase (NAT2 9). Drug Metab Dispos 1995; 23: 1354-1361. MEDLINE

27 Ware JA, Svensson CK. Longitudinal distribution of arylamine N-acetyltransferases in the intestine of the hamster, mouse, and rat. Evidence for multiplicity of N-acetyltransferases in the intestine. Biochem Pharmacol 1996; 52: 1613-1620. MEDLINE

28 Drobitch RK, Divakaruni P, Svensson CK. Effect of streptolysin-O-on rat hepatic acetyl coenzyme-A: arylamine N-acetyltransferase and cytochrome P-450 2B 1/2 activities ex vivo. Immunopharmacol Immunotoxicol 1998; 20: 159-171. MEDLINE

29 Yamamura E, Sayama M, Kakikawa M, Mori M, Taketo A, Kodaira K. Purification and biochemical properties of an N-hydroxyarylamine O-acetyltransferase from Escherichia coli. Biochim Biophys Acta 2000; 1475: 10-16. MEDLINE

30 Kattner E, Dubbels R, Schloot W. Acetylation of procainamide and isoniazid by a rat liver-N-acetyl-transferase. Eur J Drug Metab Pharmacokinet 1981; 6: 81-84. MEDLINE

31 Riddle B, Jencks WP. Acetyl-coenzyme A: arylamine N-acetyltransferase. Role of the acetyl-enzyme intermediate and the effects of substituents on the rate. J Biol Chem 1971; 246: 3250-3258. MEDLINE

32 Payton M, Mushtaq A, Yu TW, Wu LJ, Sinclair J, Sim E. Eubacterial arylamine N-acetyltransferases-identification and comparison of 18 members of the protein family with conserved active site cysteine, histidine and aspartate residues. Microbiology 2001; 147: 1137-1147. MEDLINE

33 Delomenie C, Fouix S, Longuemaux S, Brahimi N, Bizet C, Picard B et al. Identification and functional characterisation of arylamine N-acetyltransferase in eubacteria: evidence for highly selective acetylation of 5-aminosalicylic acid. J Bacteriol 2001; 183: 3417-3427. MEDLINE

34 Chung JG, Wang HH, Tsou MF, Hsieh SE, Lo HH, Yen YS et al. Evidence for arylamine N-acetyltransferase activity in the bacterium Helicobacter pylori. Toxicol Lett 1997; 91: 63-71. MEDLINE

35 Ebisawa T, Deguchi T. Structure and restriction fragment length polymorphism of genes for human liver arylamine N-acetyltransferases. Biochem Biophys Res Commun 1991; 177: 1252-1257. MEDLINE

36 Matas N, Thygesen P, Stacey M, Risch A, Sim E. Mapping AAC1, AAC2 and AACP, the genes for arylamine N-acetyltransferases, carcinogen metabolising enzymes on human chromosome 8p22, a region frequently deleted in tumours. Cytogenet Cell Genet 1997; 77: 290-295. MEDLINE

37 Martell KJ, Vatsis KP, Weber WW. Molecular genetic basis of rapid and slow acetylation in mice. Mol Pharmacol 1991; 40: 218-227. MEDLINE

38 Kelly SL, Sim E. Arylamine N-acetyltransferase in Balb/c mice: identification of a novel mouse isoenzyme by cloning and expression in vitro. Biochem J 1994; 302: 347-353. MEDLINE

39 Martell KJ, Levy GN, Weber WW. Cloned mouse N-acetyltransferases: enzymatic properties of expressed Nat-1 and Nat-2 gene products. Mol Pharmacol 1992; 42: 265-272. MEDLINE

40 Estrada-Rodgers L, Levy GN, Weber WW. Substrate selectivity of mouse N-acetyltransferases 1, 2, and 3 expressed in COS-1 cells. Drug Metab Dispos 1998; 26: 502-505. MEDLINE

41 Fakis G, Boukouvala S, Buckle V, Payton M, Denning C, Sim E. Chromosome mapping of the genes for murine arylamine N-acetyltransferases (NATs), enzymes involved in the metabolism of carcinogens: identification of a novel upstream noncoding exon for murine Nat2. Cytogenet Cell Genet 2000; 90: 134-138. MEDLINE

42 Mattano SS, Erickson RP, Nesbitt MN, Weber WW. Linkage of Nat and Es-1 in the mouse and development of strains congenic for N-acetyltransferase. J Hered 1988; 79: 430-433. MEDLINE

43 Fretland AJ, Doll MA, Gray K, Feng Y, Hein DW. Cloning, sequencing, and recombinant expression of NAT1, NAT2, and NAT3 derived from the C3H/HeJ (rapid) and A/HeJ (slow) acetylator inbred mouse: functional characterization of the activation and deactivation of aromatic amine carcinogens. Toxicol Appl Pharmacol 1997; 142: 360-366. Article MEDLINE

44 Blum M, Grant DM, Demierre A, Meyer UA. N-acetylation pharmacogenetics: a gene deletion causes absence of arylamine N-acetyltransferase in liver of slow acetylator rabbits. Proc Natl Acad Sci USA 1989; 86: 9554-9557. MEDLINE

45 Ferguson RJ, Doll MA, Rustan TD, Hein DW. Cloning, expression, and functional characterization of rapid and slow acetylator polymorphic N-acetyl-transferase encoding genes of the Syrian hamster. Pharmacogenetics 1996; 6: 55-66. MEDLINE

46 Hein DW, Rustan TD, Bucher KD, Martin WJ, Furman EJ. Acetylator phenotype-dependent and -independent expression of arylamine N-acetyltransferase isozymes in rapid and slow acetylator inbred rat liver. Drug Metab Dispos 1991; 19: 933-937. MEDLINE

47 Blum M, Grant DM, Demierre A, Meyer UA. Nucleotide sequence of a full-length cDNA for arylamine N-acetyltransferase from rabbit liver. Nucl Acids Res 1989; 17: 3589.

48 Trepanier LA, Cribb AE, Spielberg SP, Ray K. Deficiency of cytosolic arylamine N-acetylation in the domestic cat and wild felids caused by the presence of a single NAT1-like gene. Pharmacogenetics 1998; 8: 169-179. MEDLINE

49 Trepanier LA, Ray K, Winand NJ, Spielberg SP, Cribb AE. Cytosolic arylamine N-acetyltransferase (NAT) deficiency in the dog and other canids due to an absence of NAT genes. Biochem Pharmacol 1997; 54: 73-80. MEDLINE

50 Blum M, Demierre A, Grant DM, Heim M, Meyer UA. Molecular mechanism of slow acetylation of drugs and carcinogens in humans. Proc Natl Acad Sci USA 1991; 88: 5237-5241. MEDLINE

51 Grant DM, Blum M, Beer M, Meyer UA. Monomorphic and polymorphic human arylamine N-acetyltransferases: a comparison of liver isozymes and expressed products of two cloned genes. Mol Pharmacol 1991; 39: 184-191. MEDLINE

52 Grant DM, Hughes NC, Janezic SA, Goodfellow GH, Chen HJ, Gaedigk A et al. Human acetyltransferase polymorphisms. Mutat Res 1997; 376: 61-70. Article MEDLINE

53 Bell DA, Taylor JA, Butler MA, Stephens EA, Wiest J, Brubaker LH et al. Genotype/phenotype discordance for human arylamine N-acetyltransferase (NAT2) reveals a new slow-acetylator allele common in African-Americans. Carcinogenesis 1993; 14: 1689-1692. MEDLINE

54 Cascorbi I, Drakoulis N, Brockmoller J, Maurer A, Sperling K, Roots I. Arylamine N-acetyltransferase (NAT2) mutations and their allelic linkage in unrelated Caucasian individuals: correlation with phenotypic activity. Am J Hum Genet 1995; 57: 581-592. MEDLINE

55 Hickman D, Risch A, Camilleri JP, Sim E. Genotyping human polymorphic arylamine N-acetyltransferase: identification of new slow allotypic variants. Pharmacogenetics 1992; 2: 217-226. MEDLINE

56 Hein DW, Doll MA, Fretland AJ, Leff MA, Webb SJ, Xiao GH et al. Molecular genetics and epidemiology of the NAT1 and NAT2 acetylation polymorphisms. Cancer Epidemiol, Biomarkers & Prevention 2000; 9: 29-42.

57 Grant DM, Morike K, Eichelbaum M, Meyer UA. Acetylation pharmacogenetics. The slow acetylator phenotype is caused by decreased or absent arylamine N-acetyltransferase in human liver. J Clin Invest 1990; 85: 968-972. MEDLINE

58 Dupret JM, Goodfellow GH, Janezic SA, Grant DM. Structure-function studies of human arylamine N-acetyltransferases NAT1 and NAT2. Functional analysis of recombinant NAT1/NAT2 chimeras expressed in Escherichia coli. J Biol Chem 1994; 269: 26830-26835. MEDLINE

59 Hein DW, Rustan TD, Ferguson RJ, Doll MA, Gray K. Metabolic activation of aromatic and heterocyclic N-hydroxyarylamines by wild-type and mutant recombinant human NAT1 and NAT2 acetyltransferases. Arch Toxicol 1994; 68: 129-133. Article MEDLINE

60 Hickman D, Palamanda JR, Unadkat JD, Sim E. Enzyme kinetic properties of human recombinant arylamine N-acetyltransferase 2 allotypic variants expressed in Escherichia coli. Biochem Pharmacol 1995; 50: 697-703. MEDLINE

61 Delomenie C, Goodfellow GH, Krishnamoorthy R, Grant DM, Dupret JM. Study of the role of the highly conserved residues Arg9 and Arg64 in the catalytic function of human N-acetyltransferases NAT1 and NAT2 by site-directed mutagenesis. Biochem J 1997; 323: 207-215. MEDLINE

62 Leff MA, Fretland AJ, Doll MA, Hein DW. Novel human N-acetyltransferase 2 alleles that differ in mechanism for slow acetylator phenotype. J Biol Chem 1999; 274: 34519-34522. MEDLINE

63 Hein DW, Doll MA, Rustan TD, Ferguson RJ. Metabolic activation of N-hydroxyarylamines and N-hydroxyarylamides by 16 recombinant human NAT2 allozymes: effects of 7 specific NAT2 nucleic acid substitutions. Cancer Res 1995; 55: 3531-3536. MEDLINE

64 Hein DW, Ferguson RJ, Doll MA, Rustan TD, Gray K. Molecular genetics of human polymorphic N-acetyltransferase: enzymatic analysis of 15 recombinant wild-type, mutant, and chimeric NAT2 allozymes. Hum Mol Genet 1994; 3: 729-734. MEDLINE

65 Evans DA. N-acetyltransferase. Pharmacol Therapeut 1989; 42: 157-234.

66 Meyer UA, Zanger UM. Molecular mechanisms of genetic polymorphisms of drug metabolism. Annu Rev Pharmacol Toxicol 1997; 37: 269-296. MEDLINE

67 Glowinski IB, Radtke HE, Weber WW. Genetic variation in N-acetylation of carcinogenic arylamines by human and rabbit liver. Mol Pharmacol 1978; 14: 940-949. MEDLINE

68 McQueen CA, Weber WW. Characterization of human lymphocyte N-acetyltransferase and its relationship to the isoniazid acetylator polymorphism. Biochem Genet 1980; 18: 889-904. MEDLINE

69 Cribb AE, Grant DM, Miller MA, Spielberg SP. Expression of monomorphic arylamine N-acetyltransferase (NAT1) in human leukocytes. J Pharmacol Exp Therapeut 1991; 259: 1241-1246.

70 Kilbane AJ, Petroff T, Weber WW. Kinetics of acetyl CoA: arylamine N-acetyltransferase from rapid and slow acetylator human liver. Drug Metab Dispos 1991; 19: 503-507. MEDLINE

71 Ward A, Hickman D, Gordon JW, Sim E. Arylamine N-acetyltransferase in human red blood cells. Biochem Pharmacol 1992; 44: 1099-1104. MEDLINE

72 Weber WW, Vatsis KP. Individual variability in p-aminobenzoic acid N-acetylation by human N-acetyltransferase (NAT1) of peripheral blood. Pharmacogenetics 1993; 3: 209-212. MEDLINE

73 Vatsis KP, Weber WW. Structural heterogeneity of Caucasian N-acetyltransferase at the NAT1 gene locus. Arch Biochem Biophys 1993; 301: 71-76. Article MEDLINE

74 Bell DA, Badawi AF, Lang NP, Ilett KF, Kadlubar FF, Hirvonen A. Polymorphism in the N-acetyltransferase 1 (NAT1) polyadenylation signal: association of NAT1*10 allele with higher N-acetylation activity in bladder and colon tissue. Cancer Res 1995; 55: 5226-5229. MEDLINE

75 Badawi AF, Hirvonen A, Bell DA, Lang NP, Kadlubar FF. Role of aromatic amine acetyltransferases, NAT1 and NAT2, in carcinogen-DNA adduct formation in the human urinary bladder. Cancer Res 1995; 55: 5230-5237. MEDLINE

76 Kadlubar FF, Badawi AF. Genetic susceptibility and carcinogen-DNA adduct formation in human urinary bladder carcinogenesis. Toxicol Lett 1995; 82-83: 627-632.

77 Payton MA, Sim E. Genotyping human arylamine N-acetyltransferase type 1 (NAT1): the identification of two novel allelic variants. Biochem Pharmacol 1998; 55: 361-366. MEDLINE

78 Bell DA, Stephens EA, Castranio T, Umbach DM, Watson M, Deakin M et al. Polyadenylation polymorphism in the acetyltransferase 1 gene (NAT1) increases risk of colorectal cancer. Cancer Res 1995; 55: 3537-3542. MEDLINE

79 Bruhn C, Brockmoller J, Cascorbi I, Roots I, Borchert HH. Correlation between genotype and phenotype of the human arylamine N-acetyltransferase type 1 (NAT1). Biochem Pharmacol 1999; 58: 1759-1764. MEDLINE

80 de Leon JH, Vatsis KP, Weber WW. Characterization of naturally occurring and recombinant human N-acetyltransferase variants encoded by NAT1. Mol Pharmacol 2000; 58: 288-299. MEDLINE

81 Bouchardy C, Mitrunen K, Wikman H, Husgafvel-Pursiainen K, Dayer P, Benhamou S et al. N-acetyltransferase NAT1 and NAT2 genotypes and lung cancer risk. Pharmacogenetics 1998; 8: 291-298. MEDLINE

82 Lo-Guidice JM, Allorge D, Chevalier D, Debuysere H, Fazio F, Lafitte LJ et al. Molecular analysis of the N-acetyltransferase 1 gene (NAT1*) using polymerase chain reaction-restriction fragment-single strand conformation polymorphism assay. Pharmacogenetics 2000; 10: 293-300. MEDLINE

83 Dhaini HR, Levy GN. Arylamine N-acetyltransferase 1 (NAT1) genotypes in a Lebanese population. Pharmacogenetics 2000; 10: 79-83. Article MEDLINE

84 Clark DW. Genetically determined variability in acetylation and oxidation. Therapeutic implications. Drugs 1985; 29: 342-375. MEDLINE

85 Evans DA. Survey of the human acetylator polymorphism in spontaneous disorders. J Med Genet 1984; 21: 243-253. MEDLINE

86 Evans DA. Genetic Factors in Drug Therapy Cambridge University Press: Cambridge, 1993.

87 Weber WW. Pharmacogenetics Oxford University Press: New York, 1997.

88 Besarati Nia A, Van Straaten HW, Kleinjans JC, Van Schooten FJ. Immunoperoxidase detection of 4-aminobiphenyl- and polycyclic aromatic hydrocarbons-DNA adducts in induced sputum of smokers and non-smokers. Mutat Res 2000; 468: 125-135. MEDLINE

89 Nagao M, Wakabayashi K, Ushijima T, Toyota M, Totsuka Y, Sugimura T. Human exposure to carcinogenic heterocyclic amines and their mutational fingerprints in experimental animals. Environ Health Perspect 1996; 104: 497-501. MEDLINE

90 Smith CJ, Livingston SD, Doolittle DJ. An international literature survey of 'IARC Group I carcinogens' reported in mainstream cigarette smoke. Food & Chem Toxicol 1997; 35: 1107-1130.

91 Sram RJ, Binkova B. Molecular epidemiology studies on occupational and environmental exposure to mutagens and carcinogens, 1997-1999. Environ Health Perspect 2000; 108: 57-70. MEDLINE

92 Mao Y, Hu J, Ugnat AM, White K. Non-Hodgkin's lymphoma and occupational exposure to chemicals in Canada. Canadian Cancer Registries Epidemiology Research Group. Ann Oncol 2000; 11: 69-73. MEDLINE

93 Zhou Q, Talaska G, Jaeger M, Bhatnagar VK, Hayes RB, Zenzer TV et al. Benzidine-DNA adduct levels in human peripheral white blood cells significantly correlate with levels in exfoliated urothelial cells. Mutat Res 1997; 393: 199-205. MEDLINE

94 Lower GM Jr, Nilsson T, Nelson CE, Wolf H, Gamsky TE, Bryan GT. N-acetyltransferase phenotype and risk in urinary bladder cancer: approaches in molecular epidemiology. Preliminary results in Sweden and Denmark. Environ Health Perspect 1979; 29: 71-79. MEDLINE

95 Wohlleb JC, Hunter CF, Blass B, Kadlubar FF, Chu DZ, Lang NP. Aromatic amine acetyltransferase as a marker for colorectal cancer: environmental and demographic associations. Int J Cancer 1990; 46: 22-30. MEDLINE

96 Ilett KF, David BM, Detchon P, Castleden WM, Kwa R. Acetylation phenotype in colorectal carcinoma. Cancer Res 1987; 47: 1466-1469. MEDLINE

97 Lang NP, Chu DZ, Hunter CF, Kendall DC, Flammang TJ, Kadlubar FF. Role of aromatic amine acetyltransferase in human colorectal cancer. Arch Surg 1986; 121: 1259-1261. MEDLINE

98 Brockton N, Little J, Sharp L, Cotton SC. N-acetyltransferase polymor- phisms and colorectal cancer: a HuGE review. Am J Epidemiol 2000; 151: 846-861.

99 Chen J, Stampfer MJ, Hough HL, Garcia-Closas M, Willett WC, Hennekens CH et al. A prospective study of N-acetyltransferase genotype, red meat intake, and risk of colorectal cancer. Cancer Res 1998; 58: 3307-3311. MEDLINE

100 Welfare MR, Cooper J, Bassendine MF, Daly AK. Relationship between acetylator status, smoking, and diet and colorectal cancer risk in the north-east of England. Carcinogenesis 1997; 18: 1351-1354. Article MEDLINE

101 Butcher NJ, Ilett KF, Minchin RF. Substrate-dependent regulation of human arylamine N-acetyltransferase-1 in cultured cells. Mol Pharmacol 2000; 57: 468-473. MEDLINE

102 Lang NP, Butler MA, Massengill J, Lawson M, Stotts RC, Hauer-Jensen M et al. Rapid metabolic phenotypes for acetyltransferase and cytochrome P4501A2 and putative exposure to food-borne heterocyclic amines increase the risk for colorectal cancer or polyps. Cancer Epidemiol, Biomarkers & Prevention 1994; 3: 675-682.

103 Kaderlik KR, Minchin RF, Mulder GJ, Ilett KF, Daugaard-Jenson M, Teitel CH et al. Metabolic activation pathway for the formation of DNA adducts of the carcinogen 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP) in rat extrahepatic tissues. Carcinogenesis 1994; 15: 1703-1709. MEDLINE

104 Kadlubar F, Kaderlik RK, Mulder GJ, Lin D, Butler MA, Teitel CH et al. Metabolic activation and DNA adduct detection of PhIP in dogs, rats, and humans in relation to urinary bladder and colon carcinogenesis. Princess Takamatsu Symp 1995; 23: 207-213. MEDLINE

105 Kristiansen E, Meyer O, Thorup I. The ability of two cooked food mutagens to induce aberrant crypt foci in mice. Eur J Cancer Prev 1997; 6: 53-57. MEDLINE

106 Purewal M, Fretland AJ, Schut HA, Hein DW, Wargovich MJ. Association between acetylator genotype and 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP) DNA adduct formation in colon and prostate of inbred Fischer 344 and Wistar Kyoto rats. Cancer Lett 2000; 149: 53-60. MEDLINE

107 Bandmann O, Vaughan JR, Holmans P, Marsden CD, Wood NW. Detailed genotyping demonstrates association between the slow acetylator genotype for N-acetyltransferase 2 (NAT2) and familial Parkinson's disease. Mov Disord 2000; 15: 30-35. Article MEDLINE

108 Tan EK, Khajavi M, Thornby JI, Nagamitsu S, Jankovic J, Ashizawa T. Variability and validity of polymorphism association studies in Parkinson's disease. Neurology 2000; 55: 533-538. MEDLINE

109 Ogawa M. [Biochemical, molecular genetic and ecogenetic studies of polymorphic arylamine N-acetyltransferase (NAT2) in the brain]. Fukuoka Igaku Zasshi - Fukuoka Acta Medica 1999; 90: 118-131.

110 Bonicke R, Reif W. Enzymatische inaktivierung von isonicotinsaurehydrazid im menschlichen und tierischen organismus. Arch Exp Pathol Pharmacol 1953; 220: 321-333.

111 Jounela AJ, Pasanen M, Mattila MJ. Acetylator phenotype and the antihypertensive response to hydralazine. Acta Medica Scand 1975; 197: 303-306.

112 Shepherd AM, Ludden TM, McNay JL, Lin MS. Hydralazine kinetics after single and repeated oral doses. Clin Pharmacol Therapeut 1980; 28: 804-811.

113 Spielberg SP. N-acetyltransferases: pharmacogenetics and clinical consequences of polymorphic drug metabolism. J Pharmacokinet Biopharmaceut 1996; 24: 509-519.

114 Das KM, Dubin R. Clinical pharmacokinetics of sulphasalazine. Clin Pharmacokinet 1976; 1: 406-425. MEDLINE

115 Shear NH, Spielberg SP, Grant DM, Tang BK, Kalow W. Differences in metabolism of sulfonamides predisposing to idiosyncratic toxicity. Ann Intern Med 1986; 105: 179-184. MEDLINE

116 Cribb AE, Nuss CE, Alberts DW, Lamphere DB, Grant DM, Grossman SJ et al. Covalent binding of sulfamethoxazole reactive metabolites to human and rat liver subcellular fractions assessed by immunochemical detection. Chem Res Toxicol 1996; 9: 500-507. Article MEDLINE

117 May DG, Porter JA, Uetrecht JP, Wilkinson GR, Branch RA. The contribution of N-hydroxylation and acetylation to dapsone pharmacokinetics in normal subjects. Clin Pharmacol Therapeut 1990; 48: 619-627.

118 Bluhm RE, Adedoyin A, McCarver DG, Branch RA. Development of dapsone toxicity in patients with inflammatory dermatoses: activity of acetylation and hydroxylation of dapsone as risk factors. Clin Pharmacol Therapeut 1999; 65: 598-605.

119 Ratain MJ, Mick R, Berezin F, Janisch L, Schilsky RL, Williams SF et al. Paradoxical relationship between acetylator phenotype and amonafide toxicity. Clin Pharmacol Therapeut 1991; 50: 573-579.

120 Ratain MJ, Mick R, Berezin F, Janisch L, Schilsky RL, Vogelzang NJ et al. Phase I study of amonafide dosing based on acetylator phenotype. Cancer Res 1993; 53: 2304-2308. MEDLINE

121 Ratain MJ, Rosner G, Allen SL, Costanza M, Van Echo DA, Henderson IC et al. Population pharmacodynamic study of amonafide: a Cancer and Leukemia Group B study. J Clin Oncol 1995; 13: 741-747. MEDLINE

122 Ratain MJ, Mick R, Janisch L, Berezin F, Schilsky RL, Vogelzang NJ et al. Individualized dosing of amonafide based on a pharmacodynamic model incorporating acetylator phenotype and gender. Pharmacogenetics 1996; 6: 93-101. MEDLINE

123 Payton M, Auty R, Delgoda R, Everett M, Sim E. Cloning and characterization of arylamine N-acetyltransferase genes from Mycobacterium smegmatis and Mycobacterium tuberculosis: increased expression results in isoniazid resistance. J Bacteriol 1999; 181: 1343-1347. MEDLINE

124 Okumura F, Ueda O, Kitamura S, Tatsumi K. N-acetylation and N-formylation of carcinogenic arylamines and related compounds in dogs. Carcinogenesis 1995; 16: 71-76. MEDLINE

125 Moller L, Zeisig M, Midtvedt T, Gustafsson JA. Intestinal microflora enhances formation of DNA adducts following administration of 2-NF and 2-AAF [published erratum appears in Carcinogenesis 1994 Dec; 15(12): 2969]. Carcinogenesis 1994; 15: 857-861. MEDLINE

126 Guo R, Thormann W, Lauterberg B. Relationship between high incidence of adverse dapsone reactions and slow acetylate phenotype or low plasma/lymphocyte glutathione level. Chin Med J 1996; 109: 933-936. MEDLINE

127 Zielinska E, Niewiarowski W, Bodalski J, Rebowski G, Skretkowicz J, Mianowska K et al. Genotyping of the arylamine N-acetyltransferase polymorphism in the prediction of idiosyncratic reactions to trimethoprim-sulfamethoxazole in infants. Pharmacy World & Science 1998; 20: 123-130.

128 Cribb AE, Nakamura H, Grant DM, Miller MA, Spielberg SP. Role of polymorphic and monomorphic human arylamine N-acetyltransferases in determining sulfamethoxazole metabolism. Biochem Pharmacol 1993; 45: 1277-1282. MEDLINE

129 Lemke LE, McQueen CA. Acetylation and its role in the mutagenicity of the antihypertensive agent hydralazine. Drug Metab Disposition 1995; 23: 559-565.

130 Strandberg I, Boman G, Hassler L, Sjoqvist F. Acetylator phenotype in patients with hydralazine-induced lupoid syndrome. Acta Medica Scand 1976; 200: 367-371.

131 Walubo A, Aboo A. Phenytoin toxicity due to concomitant antituberculosis therapy. S Afr Med J 1995; 85: 1175-1176. MEDLINE

132 Ohno M, Yamaguchi I, Yamamoto I, Fukuda T, Yokota S, Maekura R et al. Slow N-acetyltransferase 2 genotype affects the incidence of isoniazid and rifampicin-induced hepatotoxicity. Int J Tuberculosis Lung Dis 2000; 4: 256-261.

133 Zielinska E, Niewiarowski W, Bodalski J. The arylamine N-acetyltransferase (NAT2) polymorphism and the risk of adverse reactions to co-trimoxazole in children. Eur J Clin Pharmacol 1998; 54: 779-785. MEDLINE

134 Kitas GD, Farr M, Waterhouse L, Bacon PA. Influence of acetylator status on sulphasalazine efficacy and toxicity in patients with rheumatoid arthritis. Scand J Rheumatol 1992; 21: 220-225. MEDLINE

135 Pullar T, Hunter JA, Capell HA. Effect of acetylator phenotype on efficacy and toxicity of sulphasalazine in rheumatoid arthritis. Ann Rheum Dis 1985; 44: 831-837. MEDLINE

136 Mongey AB, Sim E, Risch A, Hess E. Acetylation status is associated with serological changes but not clinically significant disease in patients receiving procainamide. J Rheumatol 1999; 26: 1721-1726. MEDLINE

137 Tan EM, Rubin RL. Autoallergic reactions induced by procainamide. J Allergy Clin Immunol 1984; 74: 631-634. MEDLINE

138 Reidenberg MM, Drayer DE. Procainamide, N-acetylprocainamide, antinuclear antibody and systemic lupus erythematosus. Angiology 1986; 37: 968-971. MEDLINE

139 Paykel ES, West PS, Rowan PR, Parker RR. Influence of acetylator phenotype on antidepressant effects of phenelzine. Br J Psychiatry 1982; 141: 243-248. MEDLINE

140 Johnstone EC. The relationship between acetylator status and inhibition of monoamine oxidase, excretion of free drug and antidepressant response in depressed patients on phenelzine. Psychopharmacologia 1976; 46: 289-294. MEDLINE

141 Kawakubo Y, Nakamori M, Schopf E, Ohkido M. Acetylator phenotype in patients with p-phenylenediamine allergy. Dermatology 1997; 195: 43-45. MEDLINE

Figure 1 ClustalX multiple sequence alignment of 22 NAT proteins from 14 different species. A conservation score for each column is shown by the histogram. The shaded columns represent the catalytic triad which is conserved in all NATs. The three domains, identified from the crystal structure of NAT from S. typhimurium, are shown as shaded bars below the histogram.

Figure 2 Phylogenetic tree for the NATs determined from the amino acid alignments shown in Figure 1. The dendrogram shows the relative distance between each sequence on the horizontal axis. All sequences were acquired from GenBank including the chicken NATX (accession number J03737) which, to date, has not been assigned a name.

Figure 3 RifF, an NAT-like protein from Amycolatopsis mediterranei, catalyses an amide bond formation during the latter stages of rifamycin synthesis.²² Although RifF is similar to the other prokaryotic NATs, it has evolved to function specifically in the synthesis of rifamycin B.

Table 1 Human NAT2 alleles (adapted from References ⁵² and ⁵⁶)

Table 2 Human NAT1 alleles (adapted from References ⁵² and ⁵⁶)

Table 3 Effect of acetylator status on drug response and toxicity