Cyclic di-AMP (c-di-AMP) is a relatively new member of the family of bacterial cyclic dinucleotide second messengers. It has attracted significant attention in recent years because of the abundant roles it plays in a variety of Gram-positive bacteria. The structural features that allow diverse bacterial proteins to bind c-di-AMP are not fully understood. Here we report the biophysical and structural studies of c-di-AMP in complex with a bacterial cation-proton antiporter (CpaA) RCK (regulator of the conductance of K(+)) protein from Staphylococcus aureus (Sa). The crystal structure of the SaCpaA_RCK C-terminal domain (CTD) in complex with c-di-AMP was determined to a resolution of 1.81 Å. This structure revealed two well-liganded water molecules, each interacting with one of the adenine bases by a unique H2Olp-π interaction to stabilize the complex. Sequence blasting using the SaCpaA_RCK primary sequence against the bacterial genome database returned many CpaA analogues, and alignment of these sequences revealed that the active site residues are all well-conserved, indicating a universal c-di-AMP binding mode for CpaA_RCK. A proteoliposome activity assay using the full-length SaCpaA membrane protein indicated that c-di-AMP binding alters its antiporter activity by approximately 40%. A comparison of this structure to all other reported c-di-AMP-receptor complex structures revealed that c-di-AMP binds to receptors in either a "U-shape" or "V-shape" mode. The two adenine rings are stabilized in the inner interaction zone by a variety of CH-π, cation-π, backbone-π, or H2Olp-π interaction, but more commonly in the outer interaction zone by hydrophobic CH-π or π-π interaction. The structures determined to date provide an understanding of the mechanisms by which a single c-di-AMP can interact with a variety of receptor proteins, and how c-di-AMP binds receptor proteins in a special way different from that of c-di-GMP.

Since the publication of this article, the authors have noticed several errors that in any case do not affect the original conclusions presented. The authors apologize for any inconvenience caused.The gel presented in panel A of Figure 2 suggests a slightly incorrect size for the purified CYC domain of XC_250. Panel E in the same figure incorrectly describes D28 as D41 (see below). The correct figure and legend are shown below. Source data for this figure is now available on the online supplementary information page.The predicted critical amino acid sites of XC_0250 were reported incorrectly in the section 'The cyclase domain of XC_0250 is active in cyclic GMP synthesis', with D28 described as D41. The text should have read:'The two critical metal-ion binding aspartates are conserved (D28 and D71) as well as an alanine (A150) residue that occupies a substrate-specifying position. However, the transition state-stabilizing asparagine and arginine residues are substituted by leucine (L157) and alanine (A161). The importance of both conserved and altered residues (D28, D41, D71, L73, A150, L157, A161) for the enzymatic activity of this domain was examined by assessing the effects of alanine or serine substitutions'.And at a later point in the same section: 'Several of these residues (D28, D71, A150, L157) are conserved in the R. centenum guanylyl cyclase (Supplementary Fig S1).' Consistent with this, the alignment in Supplementary Fig S1 panel A incorrectly shows the position of D28. The correct figure and legend are available below.

FleQ is a master regulator that controls bacterial flagellar gene expression. It is a unique enhancer-binding protein or repressor protein comprising an N-terminal FleQ domain, an AAA(+)/ATPase σ54-interaction domain and a helix-turn-helix DNA-binding domain. FleN is a putative ATPase with a deviant Walker A motif that works together with FleQ by binding to the FleQ N-terminal domain to fully express pel, psl and cdr operons in the presence of c-di-GMP to enhance biofilm formation. Stenotrophomonas maltophilia is an emerging human pathogen that causes fatal infections in humans. In order to understand the interaction between the FleN and FleQ domains and its effect on S. maltophilia biofilm formation, determination of the FleQ-c-di-GMP and FleN-FleQ-c-di-GMP complex structures was embarked upon. Towards this goal, the FleQ N-terminal domain from S. maltophilia was first cloned and expressed in Escherichia coli. Native and SeMet-labelled FleQ domains were successfully crystallized and diffracted to resolutions of 2.08 and 2.58 Å, respectively.

No abstract

No abstract