Improved genotyping of N-acetyltransferase 2: role of the ultra-slow acetylators.



N-acetyltransferase 2 polymorphisms are of high relevance in clinical toxicology (Lück et al., 2009[29]; Bing et al., 2011[4]; Costa et al., 2012[9]). The slow acetylator genotype of NAT2 has been demonstrated to be associated with an increased risk of anti-tuberculosis drug-induced liver damage (Cai et al., 2012[7]; Lv et al., 2012[30]; An et al., 2012[2]; Ben Mahmoud et al., 2012[3]; Bose et al., 2011[6]). Moreover, many urinary bladder carcinogens are substrates of NAT2 (Golka et al., 1996[16]; 2002[15]; Vineis et al., 2001[44]; Hung et al., 2004[25]; Moore et al., 2011[31]). Large meta-analyses have clearly shown an association between slow acetylation genotypes and increased risk of bladder cancer (Garcia-Closas et al., 2005[14]; 2011[13]; Sanderson et al., 2007[34]; Agúndez et al., 2008[1]; Hein, 2002[20], 2006[21], 2009[22]; Hein and Doll, 2012[23][24]). However, at the level of individual studies the results remain controversial. Of 46 studies included into one of the recent meta-analysis 35 did not reach statistical significance (Moore et al., 2011[31]).

To clarify the situation a recent study has been performed to identify the role of 'extreme' genotypes (Selinski et al., 2013[36]). This study is based on a population of 344 individuals that have been phenotyped by the caffeine test (Blaszkewicz, 2004[5]; Hakooz, 2009[18]; Jetter et al., 2009[26]). This test quantitatively determines the activity of NAT2 in vivo. A subgroup with an `ultra-slow´ in vivo metabolism of caffeine was identified.

Interestingly, these individuals with the ultra-slow NAT2 phenotype carried several slow acetylator alleles and could be identified as *6A/*6A, *6A/*7B and *7B/*7B genotypes. This combination of slow alleles, the 'ultra-slow genotype' was further tested in 1,712 bladder cancer cases and 2,020 controls. Remarkably, individuals with the 'ultra-slow' genotype showed an increased odds ratio for bladder cancer risk (OR=1.31, P=0.012) whereas the slow acetylators in general were not significantly associated with cancer risk.

Currently, a huge number of studies is performed to understand the association between genetic variations and phenotype (Daly, 2013[10]; Stewart and Marchan, 2012[41]; Partosch et al., 2013[32]; Sobin et al., 2011[40]; Tumer et al., 2012[42]; Zeller et al., 2012[45]; Escobar-García et al., 2012[11]). A special focus are drug metabolizing enzymes and their role in carcinogenesis (Chen et al., 2012[8]; Hanioka et al., 2011[19]; Santovito et al., 2011[35]; Fujihara et al., 2011[12]; Lankisch et al., 2008[27]; Ulusoy et al., 2007[43]). Genome-wide association studies have identified to which degree genetic variants influence bladder cancer risk (Golka et al., 2011[17]; Selinski et al., 2011[37], 2012[38][39]; Safarinejad et al., 2011[33]; Lehmann et al., 2010[28]). However, most of these approaches considered only the genotype in relation to disease. The present study (Selinski et al., 2013[36]) demonstrates the importance of understanding the association of haplotypes with enzyme activity and the relevance of extreme phenotypes.