Computational and experimental analyses of mammalian arylamine N-acetyltransferase structure and function.

Arylamine N-acetyltransferases (NATs) play an important role in the metabolism of arylamine and hydrazine drugs and many arylamine procarcinogens. The two human N-acetyltransferases, NAT1 and NAT2, are widely distributed in human tissues and are highly polymorphic. Although many xenobiotic procarcinogens and drugs are known mammalian NAT substrates, it is unclear what physiological roles these enzymes might play, what endogenous substrates they primarily act upon, or the mechanisms underlying the functional effects of specific NAT gene coding region single-nucleotide polymorphisms. Analyses of mammalian NAT protein structures can greatly help to answer these questions. Homology modeling techniques can be used to approximate mammalian NAT structures using known bacterial NAT crystal structures as templates. In comparison to the bacterial template NATs used for homology modeling, mammalian NATs have a 17-residue insert of unknown structure and function. Homology modeling analyses yielded two different alignments (Modeler 8v1 or 3DCoffee algorithms) that placed this insert in two likely alternative locations. Secondary structure prediction techniques and experimental analyses of a series of human NAT2 mutants with artificial deletions/replacements of the insert region distinguished one of these alternatives as the most likely insert location and provided a better understanding of its structure and function. This study demonstrates both the utility and limitations of computational structural modeling with proteins that differ as much as the mammalian and bacterial NATs.