(a) Overall fold and dimeric structure of BA2930. The part of the protein containing the cofactor (CoA) binding site is shown in green, while the part thought to be involved in putative ligand binding is in red.
(b) The molecular surface of the BA2930 dimer, with the electrostatic charge distribution as calculated by APBS⁴¹ and rendered by PyMOL.⁴²

Comparison of the BA2930 (dark gray) and Bacillus subtilis YokD (light gray) structures. Coenzyme A (purple) was co-crystallized with YokD, and CoA (blue) with BA2930. The largest structural variations between the two structures are colored blue in BA2930 and purple in YokD.

Surface representation and schematics of binding of BA2930 with ligands bound (citric acid (in the apo-form; (a), (aa)) or CoA ((b), (bb))), or docked (AcCoA and neomycin ((c), (cc))). The distances shown are calculated between non-hydrogen atoms. Atoms from the residue main chain are labeled next the box containing residue name and atoms from side chain are separated from this box by a solid line.

Different properties of residues in the Ba2390 structures plotted as a function of residue number. (a) A plot of the atom contacts between molecules in crystal lattice: red - residues forming close contacts (from 2.0 to 3.5 Å), green - residues forming hydrogen bonds, pale blue – residues bridged by water molecule, grey - residues forming other contacts (from 3.5 to 4.5 Å). The top graph represents contacts in the structure of the apo-form of BA2930, and the bottom in the CoA-bound structure. Contacts and distances were calculated using the CONTACT program from the CCP4 suite. (b) Plot of the RMSD and distance profiles comparing the apo and CoA-bound structures. Blue lines - RMSD values between corresponding residues. Red lines - distances between corresponding Cα atoms. The pale, thin lines represent pairwise chain comparisons between structures. Averaged results using a window of 4 comparisons are plotted with bold lines. (c) Cα distances between superposed A and B chains from one structure. The structure of the apo-form of BA2930 is represented in orange and the CoA-bound structure in green. (d) Percent deviation of the Cα temperature factors from the average value. All calculations were performed with the BioShell package.⁴³ Superposition of two BA2930 subunits yields Cα RMSD values of 0.4 Å for both apo- and CoA-bound forms of BA2930. Superposition of each chain of the apo-form on the corresponding chain of the CoA-bound form yields Cα RMSD values ranging from 0.3 to 0.4 Å. These differences can be attributed to mentioned before putative substrate binding lobe. We also observe a similar degree of flexibility in the crystallized mutants (data not shown).

Multiple sequence alignment (MSA) for selected members of the Antibiotic_NAT family. An MSA of the 56 Antibiotic_NAT (PFAM2522) family seed sequences was prepared with PCMA⁴⁴ followed by manual adjustments. The presented MSA shows a subset of the seed sequences, which represents the variability of the family alignment. Sequences are labeled with the name of the organism. The NCBI gene identification (gi) numbers of each sequence are: Vibrio shilonii – 149187267; Stigmatella aurantiaca - 115375112; Pseudomonas aeruginosa - 150956; Streptomyces fradiae – 62857296; Escheriachia coli - 134047089; Listeria monocytogenes - 167454283; Lactococcus lactis - 116326639; Clostridium butiricum - 182417486; Bacillus coagulans -124521550; Streptomyces coelicolor - 21220413; Salinispora arenicola – 159038228; Bacillus anthracis – 30257522; and Bacillus subtilis – 122921181. Amino acid residues responsible for binding of CoA/AcCoA by BA2930 are marked with red dots. Positions conserved among all sequences are highlighted in yellow and positions conserved in 80% of sequences are colored orange.