The polymorphic N-Acetyltransferase 2 (NAT2) is a well-characterised key enzyme in the phase II metabolism of aromatic amines and heterocyclic aromatic amines present in a wide range of xenobiotics such as pharmaceutical drugs and carcinogens from tobacco smoke, workplace, environment and food (Selinski et al. 2014). Patterns of polymorphisms—so-called haplotypes or NAT2 alleles—result in a reduced N-, O- and N,O-acetylation of aromatic and heterocyclic amines—in general rather due to a reduction in protein as due to structural changes in the enzyme (McDonagh et al. 2014; Hein et al. 2008). Persons with an impaired NAT2 capacity—so-called slow acetylators—have two slow copies of the gene, whereas the presence of one or two rapid or wild-type copies of the gene leads to an intermediate or rapid genotype—so-called rapid acetylators. The relationship between genotype and NAT2 activity has been investigated extensively in vitro using bacterial, yeast and eukaryotic cell systems (Selinski et al. 2014; Walraven et al. 2008; García-Closas et al. 2011). Furthermore, the caffeine test (using a further NAT2 substrate) allows the determination of the metabolic capacity of NAT2 in humans by the ratio of two caffeine metabolites in urine (Selinski et al. 2011).

Recent in vivo studies indicate remarkable differences in the subgroup of slow acetylators leading to a differentiation between “ultra-slow” and “all other slow” acetylators (Selinski et al. 2013b; Ruiz et al. 2012). Ultra-slow acetylators are persons with two copies of the frequent NAT*6A haplotype and/or perhaps other infrequent haplotypes from the NAT2*6 (sharing rs1799930, G590A with *6A) cluster or NAT2*7B (sharing rs1041983, C282A with *6A). Two independent studies in Europeans found a 35 (Ruiz et al. 2012) and 46 % (Selinski et al. 2013b), respectively, decreased NAT2 capacity in NAT2*6/*6 genotypes compared to slow *5/*5. From the functional perspective, the missense SNP rs1799930 (G590A) is more likely than the synonymous rs1041983 (C282A) to result in an even more reduced metabolic capacity than the “all other slow” NAT2*5 and further less frequent slow haplotypes (*11 and *14). However, NAT2*7B shows in vivo also a reduced metabolic capacity though data are sparse (frequency: 2–3 % in Europeans). New studies and re-evaluation of older studies indicate that this further reduced metabolic capacity of the ultra-slow acetylators may have a particular impact on a number of biological outcomes ranging from drug efficacy and side effects, bladder cancer risk to age-related impairments. Interestingly, in vitro studies show contrary results suggesting the NAT2*5 haplotype cluster (rs1801280, T341C) to result in clearly lower acetylation capacity than NAT2*6 and *7 haplotypes (Hein 2002).

NAT2 shows a remarkable interethnic variability (Garcia-Martin 2008; Selinski et al. 2013a) regarding the haplotype distribution and the proportion of slow acetylators. So, interpretation of NAT2 effects should consider these differences, e.g. absence of particular haplotypes in certain populations. These interethnic differences give rise to the question of evolutionary processes that might have been involved in the development and distribution of slow haplotypes. Slow haplotypes have a high prevalence (>50 %) in populations of European descent and particular African and South Asian populations, but are rather infrequent in East Asians (<25 %, Sabbagh et al. 2011). Change of subsistence from hunting and gathering to pastoralism and agriculture and the resulting change of ingested xenobiotics are hypothesised as selection factor in favour of slow haplotypes (Sabbagh et al. 2011; Luca et al. 2008), especially the ultra-slow haplotype NAT2*6A (Patillon et al. 2014). Possibly relevant NAT2 substrates in food are, for instance, heterocyclic amines present in well-done meat (Turesky and Le Marchand 2011). Interestingly, European and Northeast Asian populations have a similar frequency of ultra-slow NAT2*6A haplotypes but differ in the slow NAT2*5B, virtually absent in Northeast Asians but frequent in Europeans, and NAT2*7B that is much more frequent in East Asians than in Europeans (Sabbagh et al. 2011). Reasons for these large interethnic differences are still unclear but might be due to genetic drift or local selective pressures (Sabbagh et al. 2011).

NAT2 polymorphisms were initially discovered as risk factor for drug-induced liver injury (DILI) in tuberculosis patients receiving isoniazid therapy (Hughes et al. 1954). Studies on anti-tuberculosis drug-induced hepatotoxicity (ATDH) are mostly conducted in East Asian and South American populations where tuberculosis is a major health concern (Daly 2013; Du et al. 2013). A recent meta-analysis indicated that ultra-slow acetylators are at highest risk of ATDH (Selinski et al. 2014) though results have to be interpreted with care due to the absence of NAT2*5B in East Asians.

The impact of slow NAT2 status on bladder cancer risk in persons exposed to carcinogenic NAT2 substrates, e.g. aromatic amines from tobacco smoke, is well investigated since decades (Figueroa et al. 2014; Moore et al. 2011; Risch et al. 1995). In recent studies, the effect of slow NAT2 status is diminishing even in smokers, possibly due to reductions in the general environmental and occupational exposure (Ovsiannikov et al. 2012). However, ultra-slow acetylators seem to be still susceptible to bladder cancer when exposed (Selinski et al. 2013b). This can be seen, for instance, in a large case–control study (1692 cases, 1995 controls, Selinski et al. 2013b) where slow acetylators had no particular bladder cancer risk in general (OR 1.02, 95 % CI 0.89–1.17) or when smoking (current smokers OR 1.07, 95 % CI 0.81–1.41), whereas ultra-slow acetylators had still higher bladder cancer risks (OR 1.29, 95 % CI 1.03–1.62), especially when smoking (current smokers OR 1.51, 95 % CI 0.99–2.31).

Though an effect of NAT2 on ageing itself can be excluded (Agundez et al. 1997; Korrapati et al. 1997; Muiras et al. 1998), age-related impairments have been subject of a number of studies. Age-related cataract was reported to be more frequent in slow acetylators (Tamer et al. 2005; Meyer et al. 2003) though the sample size in both studies was quite small. Slow, especially ultra-slow, acetylators seem to be more susceptible to age-related hearing impairment than rapid acetylators (for review, see Uchida et al. 2014), although this effect could not be confirmed in all study groups with sufficient sample size (Dawes et al. 2015). To date, the mechanism how NAT2 is involved in age-related impairments is unclear. An effect of the decreased antioxidant capacity of the slow NAT2 genotype is hypothesised (Uchida et al. 2014).

In addition to age-related declines in basic sensory hearing functions, a recent study also suggests an association of ultra-slow NAT2 with higher cognitive auditory functions in elderly persons (Selinski et al. 2015). Using an auditory distraction paradigm and electrophysiological measures, it could be demonstrated that ultra-slow acetylators are more prone to distracting sounds than rapid and slow acetylators. Moreover, the analysis of event-related potentials indicated that ultra-slow acetylators show deficits specifically in the automatic detection of changes in the acoustic environment and in the re-focusing of attention after a distracting event. Future research will show whether NAT2 effects are confined to auditory functions, or can also be found for further cognitive functions, like visual perception and attention.

In essence, recent studies show that in parallel genotyping and phenotyping still lead to new insights in NAT2-mediated metabolism. Furthermore, it is of high importance to differentiate between the “all other slow” and “ultra-slow” acetylators as both groups may still differ in the strength of the induced biological effects. Finally, it has to be considered that acetylation—similar to methylation—is a common metabolic process that has been investigated only in a few fields such as adverse drug reactions and urinary bladder cancer risk. Recent studies indicate the impact of NAT2-mediated acetylation is broader.