1.2.1. Overall structure of GNATs
Although the homology in the primary sequence between GNATs is moderate (3%–23%), they are folded into a highly conserved three-dimensional structure. The core secondary elements of the GNAT proteins consist of six or seven β-strands and four α-helices, connected in order β0-β1-α1-α2-β2-β3-β4-α3-β5-α4-β6. Four conserved motifs, known as the N-acetyltransferase domain, are found in this core arranged in order C-D-A-B (Figure 1C) (Dyda et al., 2000; Hentchel & Escalante-Semerena, 2015; Tercero et al., 1992).
Motifs C and D play an essential role in protein stability, while motifs A and B contain the residues involved in acyl-CoA and acceptor substrate binding, respectively. (Dyda et al., 2000; Vetting et al., 2005). Motif A contains a six-residue segment (Q/R)-X-X-G-X-(G/A), known as P-loop (phosphate-binding loop), for substrate recognition and binding and glutamic or aspartic residue to deprotonate the target lysine (Liu et al., 2008; Lu et al., 2017; Vetting et al., 2005). Across the entire GNAT superfamily, there is structure divergence in the loop β1β2, the α-4 helix of motif B, and strand β6 at the C-terminal, which together form the binding site for the acceptor substrate. Specifically, the C-terminal region contains a loop of varying length and position, and the residues in the motif B are not well conserved. These structural variations allow the recognition of various substrates (Dyda et al., 2000; Salah et al., 2016). For example, in the mycothiol synthase (Rv0819) from Mycobacterium tuberculosis,  the β-strand 1 in the N-terminal domain is missing, the position of helix 2, and a long loop inserted between α3′ and β5′ (Figure 1D), while aminoglycoside 6´-N-acetyltransferase from Enterococcus faecium has an additional α-helix between the strands β1 and β2 (Vetting et al., 2003; Wybenga-Groot et al., 1999).
The binding of the enzyme to the substrate is through interactions with the pantetheine and pyrophosphate moieties of CoA (Clements et al., 1999; Dutnall et al., 1998; Hentchel & Escalante-Semerena, 2015; Lin et al., 1999; Rojas et al., 1999; Wang et al., 2008;). The pantetheine binding is based on hydrogen bonds with the main chain of β4 and β 5, and the pyrophosphate is coordinated mostly by the main chain nitrogen atoms of the conserved phosphate binding loop between β4 and α3 (Majorek et al., 2017).