Acetylation, Slow

A number sign (#) is used with this entry because the slow and fast acetylation phenotypes are due to polymorphisms in the gene encoding N-acetyltransferase-2 (NAT2; 612182).

Clinical Features

The antituberculosis agent isoniazid (INH) is rendered therapeutically inactive by acetylation. Most or perhaps all populations of the world are polymorphic for 'rapid inactivation' versus 'slow inactivation.' As shown by the study of Evans et al. (1960), the 'slow inactivator' person is homozygous for a slow inactivator allele; the 'rapid inactivator' person may be either homozygous or heterozygous for a rapid inactivator allele. The method described by Sunahara et al. (1961) permitted separation of the homozygotes and heterozygotes, i.e., 3 genotypes in all. The rapid versus slow acetylation of sulfadiazine in rabbits (Frymoyer and Jacox, 1963) is similar. The polymorphism in acetylation extends to the acetylation of sulfamethazine, which can be used as a test (Parker, 1969). Isoniazid, hydralazine, and some sulfa drugs are acetylated by a common mechanism (Evans and White, 1964). Administration of INH with phenytoin (Dilantin) results in high, even toxic levels of the anticonvulsant (Kutt et al., 1970), and the effects of the drug interaction are greater in slow acetylators. Hydralazine (Apresoline) is acetylated through the INH-type mechanism, as is procainamide (Pronestyl). INH hepatotoxicity is more frequent in slow inactivators (Timbrell et al., 1977).

McLaren et al. (1977) found a significantly higher proportion of fast acetylators in a group of diabetics without neuropathy than in those with neuropathy (see 603933) or in the normal population. Using dapsone (anti-leprosy drug) acetylation rate, Vansant et al. (1978) found a normal distribution of phenotypes in idiopathic systemic lupus erythematosus (SLE; see 152700). Drug-induced SLE has been thought to be more frequent in slow acetylators and several workers have reported the same for spontaneous SLE. Sonnhag et al. (1979) could find no relation between acetylator phenotype and proneness to develop SLE-like syndrome. Reidenberg et al. (1980) found an excess of slow acetylator phenotype in SLE. On the other hand, Baer et al. (1986) could find no association between acetylator phenotype and SLE and from a review of the literature concluded that most workers have had similar results.

Harmer et al. (1986) found no correlation between the acetylator and the sparteine hydroxylation phenotypes. This is perhaps not surprising inasmuch as the N-acetyltransferase enzyme is cytosolic in liver and jejunal mucosa, whereas the polymorphic enzyme that governs hydroxylation of sparteine, debrisoquine, and other drugs is a liver P450 (124030). In addition to isoniazid, Harmer et al. (1986) listed the following drugs as being acetylated by the polymorphic enzyme: sulfadimidine, hydralazine, dapsone, procaine amide, sulfapyridine, a reduced metabolite of nitrazepam, and a metabolite of caffeine. Nhachi (1988) found bimodality for percent urinary sulfamethazine acetylated in a 6-hour sample in a Zimbabwe population. The allele for slow acetylation was estimated to have a frequency of 0.72.

Roberts-Thomson et al. (1996) found that the fast acetylator phenotype was associated with odds ratios of 1.1 and 1.8 for adenoma and colorectal cancer, respectively. The highest risk occurred in the youngest tertile (less than 64 years) of cases. There was no difference between the sexes. The risk of adenoma or cancer increased with increasing intake of meat in fast but not in slow acetylators. The findings suggested that acetylator status modulates the risk of colorectal neoplasia associated with meat intake.

Schnakenberg et al. (1998) analyzed the NAT2 gene of blood and tumor DNA from 60 patients with primary bladder cancer and DNA of blood samples from 154 healthy individuals. They found that 70% of patients with bladder cancer were slow acetylators, while genotyping of controls resulted in 61% with slow acetylation. In addition, dividing bladder cancer patients in males and females, the genotype NAT2*5B/NAT2*6A occurred with much higher frequencies in males. Furthermore, investigating bladder cancer tissues, they detected loss of heterozygosity in slow and rapid acetylator genotypes. In 11 of 60 tumor samples (18.3%), they observed allelic loss at the NAT2 locus, while in control DNA of blood from the same patients both alleles were still detectable.

Biochemical Features

In cultured rabbit hepatocytes, McQueen et al. (1982) found a relationship between acetylator phenotype and DNA damage by chemicals that undergo N-acetylation. DNA repair, an index of DNA damage, was produced by hydralazine in hepatocytes from slow acetylator rabbits but not in those from rapid acetylators. In contrast, hepatocytes from rapid acetylators were more sensitive to toxicity from the carcinogen 2-aminofluorene and displayed greater amounts of DNA repair. The amount of DNA repair measured with each chemical was dose dependent. Thus, McQueen et al. (1982) concluded that the acetylation polymorphism may be a factor in susceptibility to toxicity and perhaps carcinogenicity of these chemicals.

The polymorphic enzyme responsible for the acetylator phenotype is arylamine N-acetyltransferase (EC 2.3.1.5). In the New Zealand white rabbit, which is a widely used animal model for the human acetylation polymorphism, Blum et al. (1989) showed that the defective arylamine N-acetylation is caused by a gene deletion. Reviews were provided by Weber (1987) and Evans (1989). Grant et al. (1990) concluded that the slow acetylator phenotype is the result of decreased or absent arylamine N-acetyltransferase in the liver. In a study of the acetylator phenotype of 26 surgical patients, they used caffeine as the probe drug and the measurement of the 5-acetyl-amino-6-formylamino-3-methyluracil to 1-methylxanthine molar ratio in urine. Liver wedge biopsies taken from these patients were used for a measurement of N-acetyltransferase activity with the substrate sulfamethazine and for quantitation of immunoreactive N-acetyltransferase protein. The ratio of caffeine metabolites in urine and in vitro sulfamethazine acetylation showed a correlation of 0.98. Furthermore, slow acetylation was associated with a decrease in the quantity of immunodetectable N-acetyltransferase protein. They isolated 2 kinetically distinct enzyme activities, designated NAT1 (108345) and NAT2, from low- and high-activity livers. Low acetylation was related to decreases in the liver content of both of these immunologically related proteins. Grant et al. (1990) concluded that parallel decrease in the 2 enzymes is involved. Unlike the rabbit, in which gene deletion is responsible for slow acetylation, Southern analyses of genomic DNA from slow and rapid acetylator humans suggested that in man the deletion mechanism is either unlikely or uncommon.

Mapping

The slow and rapid acetylation phenotypes are caused by polymorphisms in the NAT2 gene, which maps to chromosome 8p23.1-p21.3 (Hickman et al., 1994).

Molecular Genetics

The highly homologous human genes for N-acetyltransferase, NAT1 and NAT2, appear to code for the genetically invariant and variant NAT proteins, respectively. NAT1, which is responsible for N-acetylation of certain arylamine drugs, displays no genetic variation, whereas the rapid or slow acetylation of therapeutic and carcinogenic agents is due to variation at the NAT2 locus. Vatsis et al. (1991) generated 1.9-kb genomic EcoRI fragments by PCR with liver and leukocyte DNA from 7 subjects phenotyped as homozygous or heterozygous acetylators. Direct sequencing demonstrated multiple point mutations in a coding region of 2 distinct NAT2 variants. One of these, derived from leukocytes of a slow acetylator, showed a G-to-A transition at nucleotide 590 leading to replacement of arginine-197 by glutamine; the mutated guanine was part of a CpG dinucleotide and a TaqI site (612182.0001). A second NAT2 variant originated from liver with low N-acetylation activity. It was characterized by 3 nucleotide transitions giving rise to a silent mutation and 2 amino acid substitutions: ile114-to-thr (612182.0002) and lys268-to-arg (612182.0003). The results show conclusively that the genetically variant NAT is encoded by NAT2.

Mashimo et al. (1992) described a method for determining the polymorphic N-acetyltransferase phenotype from Southern blot analysis of genomic DNA from leukocytes. Abe et al. (1993) described a rapid and simple method for genotyping the N-acetyl transferase polymorphism using a PCR-based RFLP. They determined 10 different genotypes reliably.

Lin et al. (1994) found that 4 mutations--191A, 481T, 590A (612182.0001), and 857A (612182.0004)--accounted for nearly all slow acetylator alleles among blacks, whites, Asian Indians, Hispanics, Koreans, Japanese, Hong Kong Chinese, Taiwanese, Filipinos, and Samoans. The ethnic distribution supported an interpretation that the acetylation polymorphism existed before Paleolithic splitting of human populations from Africa.

Cascorbi et al. (1995) identified 7 different alleles of the NAT2 gene coding for the slow acetylation phenotype. They found a slow acetylation genotype in 58.9% of the 844 unrelated German subjects studied. In vivo acetylation capacity of homozygous wildtype subjects was significantly higher than in heterozygous genotypes. All mutant alleles showed low in vivo acetylation capacities, including the newly defined alleles. Moreover, distinct slow genotypes differed significantly among each other, as reflected in lower acetylation capacity of some alleles as compared with others.

Lee et al. (1998) studied the frequencies of NAT2 alleles and genotypes in 216 colorectal cancer patients and 187 controls among Chinese in Singapore. Their results confirmed the findings of Roots et al. (1989) and were contrary to earlier reports that the rapid N-acetylation phenotype was associated with an increased risk of colorectal cancer in Caucasians (Lang et al., 1986; Ilett et al., 1987). The frequency of rapid acetylator genotypes in patients with right-sided cancer, however, was significantly greater as compared to that of controls (odds ratio = 1.899). On the other hand, the frequency of the NAT2*7A allele, commonly associated with the slow acetylator status, was increased among colorectal cancer patients. The NAT2*7A genotypes, however, were more frequent among patients with cancer on the left side (odds ratio = 2.872) and along the sigmoid/rectal region (odds ratio = 2.642).

Population Genetics

To investigate the role of population history and natural selection in shaping variation in the closely clustered NAT1 and NAT2 genes on chromosome 8p, Patin et al. (2006) characterized genetic diversity through the resequencing and genotyping of NAT1, NAT2, and the pseudogene NATP in 13 different populations with distinct ethnic backgrounds and demographic pasts. They defined a detailed map of linkage disequilibrium of the NAT region as well as performed a number of sequence-based neutrality tests and the long-range haplotype (LRH) test. The data showed distinctive patterns of variability for the 2 genes: the reduced diversity observed at NAT1 was consistent with the action of purifying selection, whereas NAT2 functional variation contributed to high levels of diversity. The LRH test identified a particular NAT2 haplotype (NAT2*5B; 612182.0002) under recent positive selection in western/central Eurasians. This haplotype harbors the mutation 341T-C and encodes the 'slowest-acetylator' NAT2 enzyme, suggesting a general selective advantage for the slow acetylator phenotype. The NAT2*5B haplotype, which seemed to have conferred a selective advantage during the previous 6,500 years, exhibits today the strongest association with susceptibility to bladder cancer and adverse drug reactions. The patterns observed for NAT2 illustrate how geographically and temporally fluctuating xenobiotic environments may have influenced not only our genome diversity but also our present day susceptibility to disease. The diversity patterns observed in the NAT region illustrate the current vision of the human genome as a 'mosaic of segments,' each with its own individual evolutionary history (Paabo, 2003).

Magalon et al. (2008) genotyped 138 unrelated individuals from 6 populations in central Asia, including long-term sedentary agriculturalists (2 Tajik populations) and recent sedentary agriculturalists (Kazakhs and Uzbeks), whose diets are less dominated by meat, and nomad pastoralists (2 Kirghiz populations), whose diet is dominated by meat. The Tajiks and Kazakhs exhibited the highest frequency of the slow acetylator haplotype NAT2*5B (612182.0002), ranging from 22 to 26%. The NAT2*6A haplotype (612182.0001) was present at high frequencies in the Tajik populations (39 to 45%), whereas it was at the lowest frequency in Kazakhs (13%). The Kazakhs exhibited the highest frequency (23%) of NAT2*7B (612182.0004), which is mainly restricted to East Eurasian populations. The 'fast' haplotype NAT2*4, defined as the ancestral state of the gene and the reference NAT2 haplotype, was found at high frequency in the Kirghizs (46 to 48%) and in the Uzbeks (42%), but at nearly 2-fold lower frequencies in the Tajiks (23 to 26%), and intermediate frequency in the Kazakhs (38%). Overall, the Tajiks exhibited significantly higher proportions of slow acetylators (55 to 63%; p less than 0.05) as compared to the Uzbeks, Kirghizs, and Kazakhs, who presented proportions of slow acetylators ranging from 26 to 35%. Magalon et al. (2008) suggested that being a slow acetylator confers an advantage in long-term agriculturalist populations in central Asia, and indicated that selective natural environmental pressures can affect the evolution of genetic diversity.

Animal Model

The rapid versus slow acetylation of sulfadiazine in rabbits (Frymoyer and Jacox, 1963) is similar to the human isoniazid inactivation phenotype. In cultured rabbit hepatocytes, McQueen et al. (1982) found a relationship between acetylator phenotype and DNA damage by chemicals that undergo N-acetylation. DNA repair, an index of DNA damage, was produced by hydralazine in hepatocytes from slow acetylator rabbits but not in those from rapid acetylators. In contrast, hepatocytes from rapid acetylators were more sensitive to toxicity from the carcinogen 2-aminofluorene and displayed greater amounts of DNA repair. The amount of DNA repair measured with each chemical was dose dependent. The polymorphic enzyme responsible for the 'acetylator phenotype' is arylamine N-acetyltransferase (EC 2.3.1.5). In the New Zealand white rabbit, which is a widely used animal model for the human acetylation polymorphism, Blum et al. (1989) showed that the defective arylamine N-acetylation is caused by a gene deletion.

A mouse model for the human acetylation polymorphism was developed by Glowinski and Weber (1982), Mattano and Weber (1987), and Tannen and Weber (1980). The model involves the A/J (slow acetylator) and C57BL/6J (rapid acetylator) inbred strains. Mattano et al. (1988) demonstrated linkage between the Nat and the esterase (Es-1) genes, located on mouse chromosome 8. A recombination frequency of about 12% was observed between the 2 loci.

History

The finding in the late 1950s that an impairment of a phase I reaction, hydrolysis of the muscle relaxant succinylcholine by butyrylcholinesterase (177400), was inherited served as an early stimulus for the development of pharmacogenetics (Kalow, 1962). At almost the same time, Evans et al. (1960) observed that a common genetic variation in a phase II pathway of drug metabolism, N-acetylation, could result in striking differences in the half-life and plasma concentrations of drugs metabolized by N-acetyltransferase. Weinshilboum (2003) reviewed the subject of inheritance and drug response beginning from these 2 historic examples. Evans and McLeod (2003) discussed pharmacogenomics more broadly, from the standpoints of drug disposition, drug targets, and side effects.