Structural basis for the inhibition of the Bacillus subtilis c-di-AMP cyclase CdaA by the phosphoglucomutase GlmM

Cyclic-di-adenosine monophosphate (c-di-AMP) is an important nucleotide signalling molecule, which plays a key role in osmotic regulation in bacteria. Cellular c-di-AMP levels are tightly regulated, as both high and low levels have a negative impact on bacterial growth. Here, we investigated how the activity of the main Bacillus subtilis c-di-AMP cyclase CdaA is regulated by the phosphoglucomutase GlmM. c-di-AMP is produced from two molecules of ATP by proteins containing a deadenylate cyclase (DAC) domain. CdaA is a membrane-linked cyclase with an N-terminal transmembrane domain followed by the cytoplasmic DAC domain. Here we show, using the soluble catalytic B. subtilis CdaACD domain and purified full-length GlmM or the GlmMF369 variant lacking the C-terminal flexible domain 4, that the cyclase and phosphoglucomutase form a stable complex in vitro and that GlmM is a potent cyclase inhibitor. We determined the crystal structure of the individual B. subtilis CdaACD and GlmM proteins, both of which form dimers in the structures, and of the CdaACD:GlmMF369 complex. In the complex structure, a CdaACD dimer is bound to a GlmMF369 dimer in such a manner that GlmM blocks the oligomerization of CdaACD and formation of active head-to-head cyclase oligomers, thus providing molecular details on how GlmM acts as cyclase inhibitor. The function of a key amino acid residue in CdaACD in complex formation was confirmed by mutagenesis analysis. As the amino acids at the CdaACD:GlmM interphase are conserved, we propose that the observed inhibition mechanism of CdaA by GlmM is conserved among Firmicutes.


49
Nucleotide signalling molecules play important roles in helping bacteria to rapidly adapt to 50 changing environmental conditions (1,2). One such signalling nucleotide, cyclic-di-adenosine 51 monophosphate (c-di-AMP), which was discovered a little more than a decade ago (3), plays an 52 important function in the osmotic regulation of bacteria by controlling potassium and osmolyte uptake 53 (4-8). c-di-AMP also plays an important function in regulating cell size, either directly or indirectly 54 through its function in osmotic regulation, cell-wall integrity and susceptibility to beta-lactam 55 antibiotics, which target the synthesis of the peptidoglycan cell wall (9-12).

56
The function of c-di-AMP has been most extensively studied in a range of Firmicutes bacteria 57 including the Gram-positive model organism Bacillus subtilis and Gram-positive bacterial pathogens 58 such as Staphylococcus aureus, Listeria monocytogenes and several Streptococcus species (9,10,13-17).

59
From these studies, it has become apparent that the cellular level of c-di-AMP needs to be tightly 60 regulated as both an excess and a lack of c-di-AMP can negatively impact bacterial growth, physiology 61 and virulence (17,18). To achieve the optimal level, a dynamic equilibrium must exist between the 62 synthesis of c-di-AMP via diadenylate cyclases and its degradation into 5′-phosphadenylyl-adenosine 63 (pApA) or two molecules of AMP by phosphodiesterases (18)(19)(20). As part of the current study, we GlmM and with GlmMF369 that eluted as a single, higher-mobility species compared to the individual 154 proteins (Fig. 1). The peak fractions of each complex were further analysed by SDS-PAGE, confirming 155 the presence of both proteins (Fig. 1, inserts). We also determined the binding affinity between GlmM identity of the full-length proteins is 65%) (21), and S. aureus DacACD (PDB 6GYW; sequence identity 182 of the full-length proteins is 53%) (28) structures, all lacking the N-terminal transmembrane helices, 183 gave r.m.s.ds of 0.79 and 0.75, respectively, highlighting the overall structural similarities of these 184 enzymes (Fig. 3B). The B. subtilis CdaACD structure was solved as a dimer in the asymmetric unit with 185 hydrogen-bonding interactions observed at the interaction interface ( Fig 3C). Interactions were observed 186 between the side chains of amino acid residues Asn166, Thr172 and Leu174 (site 1) and residues 187 Leu150, Lys153 and Met155 (site 2) ( Fig 3C). Similar hydrogen-bonding interactions were also 188 identified in the S. aureus DacACD and L. monocytogenes ∆100CdaA structures with amino acid residues 189 in site 1 being absolutely conserved (28,30) (Fig. S1). Analysis of the interface with PDBePISA (36) 190 indicated a buried surface of 1400 Å 2 , which is similar to the value of 1460 Å 2 previously reported for 191 the S. aureus DacACD protein, indicative of a stable dimer formation. In this dimer confirmation, the 192 active sites face opposite directions and hence cannot be engaged in a catalytically active head-to-head 193 conformation (Fig. 3A). Taken together, these data indicate that the conformationally-inactive 5 dimerization interface is conserved among different CdaA homologs in Gram-positive bacteria and that 195 the enzyme needs to form higher oligomers for catalysis.

196
The His-tagged B. subtilis GlmM protein was crystallised and the structure solved by molecular 197 replacement using the B. anthracis GlmM structure (PDB 3PDK; (37)) as the search model (Table 1 198 and Fig. 3D). The B. subtilis GlmM protein displayed a four-domain architecture typical for 199 phosphoglucosamine mutase proteins (28,37) (Fig. 3D). Domains 1-3 are comprised of α-β mixed cores 200 linked via a flexible loop to domain 4, which displays a 3-stranded β sheet fold surrounded by two α-201 helices (Fig. 3D). While one GlmM molecule was present in the asymmetric unit, the typical "M-202 shaped" GlmM dimer arrangement was observed in the crystal cell packing (Fig. S2). Interactions were 203 formed between domains 1, leading to the formation of a large groove at the top of the dimer molecule, 204 mostly formed by domain 2 and the active site of each monomer subunit facing the opposite direction.

205
Two different structures were solved for the B. subtilis GlmM protein at 2.9 and 3.0 Å resolutions with 206 a superposition r.m.s.d. score of 0.29 (Table 1 and Fig. S2). One of the crystal structures was obtained 207 with a divalent cation bound to the catalytic serine residue, which during catalysis is thought to be 208 converted to a phosphoserine residue and the metal ion playing an important role during catalysis ( Fig.   209 S2). The exact type of metal ion could not be deduced due to the limitation of the structural resolution.

210
However, we speculate that it is a magnesium ion, as magnesium was present in the crystallisation 211 conditions and this metal ion is usually also bound in fully active enzymes. Furthermore, when a 212 magnesium ion was modelled into the structure and analyzed using the program CheckMyMetal (38), a 213 better fit was observed as compared to zinc or calcium ions, which could also fill the density. In the 214 second structure, a phosphate molecule (PO4) was bound to Arg419, located within a loop region in    (1) and 5 hydrogen bond and 3 ionic bond interactions with CdaACD(2) (Table S1 and Fig.  6 S5). On the other hand, only an average 220.3 Å 2 surface area is occluded in GlmMF369(2) (shown in 250 light pink in Fig. 4). Based on the PDBePISA analysis, GlmMF369(2) only formed two hydrogen bond 251 interactions with the CdaACD(2) monomer but no interaction with CdaACD(1) (Table S1 and Fig. S5). A 252 more detailed analysis of the interface showed that several interactions are made between two α-helices 253 from domain 2 of GlmMF369(1), α1 and α2, with the CdaACD(1) and CdaACD(2) monomers, respectively 254 ( Fig 4C and 4D). The main interactions in the complex were formed between three residues, D151, 255 E154, and D194 of domain 2 in GlmMF369(1) and residue R126 in each of the CdaACD monomers. More 256 specifically, ionic bonds were formed between residue D195 in GlmMF369 (1) and residue R126 in 257 CdaACD(2). In addition, salt bridges were formed between residues D151 and E154 in GlmMF369(1) and 258 residue R126 but this time from CdaACD(1) (Fig 4C and 4D, Table S2 and Fig. S5). The data suggest 259 that residue R126 in CdaACD is potentially one of the most critical residues for complex formation, as it 260 contributes to a number of ionic as well as hydrogen-bond interactions and even though the complex is 261 asymmetric, it contributes to interactions in both CdaACD monomers.

264
To gain insight how GlmM inhibits the activity of the c-di-AMP cyclase, we inspected the location of  CdaACD:GlmMF369 complex, were passed over an analytical size-exclusion column, followed by 291 continuous automated SAXS data collection throughout the run (Fig. 6, Fig S6 and   is responsible for this extra density. As control, a SAXS experiment was also performed using the 307 CdaACD:GlmMF369 complex sample for which the X-ray structure was obtained. The dimensions of the 308 CdaACD:GlmMF369 complex were Vc: 656.8, Rg: 37.51 Å and dmax: 117.5 Å and the molecular weight 309 was calculated to be 97.5 kDa, which is consistent with the theoretical molecular weight of 120 kDa for 310 a complex made of two CdaACD and two GlmMF369 molecules. Similarly, a good fit of the 311 CdaACD:GlmMF369 dimer complex structure was obtained when fitted into the reconstructed SAXS 312 envelope data (Fig 6D). These data suggest that the full-length GlmM protein likely forms a dimer-of-313 dimer complex with the c-di-AMP cyclase CdaACD and might assume a similar arrangement as observed 314 for the CdaACD:GlmMF369 complex.

316
Arginine 126 in B. subtilis CdaACD is essential for complex formation

317
The complex structure highlighted key interactions between residues D194 and residues D151/E154 in

334
CdaACD-R126A variant was active, although the activity was reduced as compared to wild-type CdaACD 335 (Fig. 7B). Importantly and in contrast to wild-type CdaACD, the enzyme activity of this variant was no 336 longer inhibited by the addition of GlmM (Fig. 7B). These data show that residue Arg126 in B. subtilis 337 CdaACD plays a critical role for complex formation and that GlmM can only inhibit the activity of the 338 c-di-AMP cyclase after the formation of a stable complex.

341
In this study, we show that the B. subtilis GlmM and CdaACD cyclase domain form a stable 342 dimer-of-dimer complex. GlmM acts through this protein-protein interaction as a potent inhibitor of the 343 c-di-AMP cyclase without requiring any additional factors. Based on the atomic-resolution complex 344 structure data, we suggest that GlmM inhibits the activity of CdaACD by preventing the formation of 345 active head-to-head cyclase oligomers.

346
For CdaA to produce c-di-AMP, two monomers need to be arranged in an active head-to-head 347 conformation. As part of this study, we determined the structure of the B. subtilis CdaACD protein and

348
show that it has the typical DAC domain fold. While the protein was also found as a dimer in the 349 structure, the dimer was in an inactive conformation, with the two active sites facing in opposite 350 directions. The interface creating the inactive dimer conformation is conserved among CdaA proteins.

351
The L. monocytogenes and S. aureus homologs, for which structures are available, were found in the 352 same inactive dimer conformation even though the proteins crystallized under different conditions and 353 were found in different space groups (21,28,30). This makes it less likely that a crystallographic previous work on the S. aureus homolog indicated that the inactive dimer conformation is very stable, 360 and in order for the protein to produce an active enzyme, the protein needs to form higher-level 361 oligomers (28). Given the similarity in the interaction interface, this is likely also the case for the B.

362
subtilis CdaACD enzyme and we would suggest that the B. subtilis CdaACD dimer observed in the 363 structure is unlikely to rearrange into an active dimer conformation.

364
We also solved the structure of the B. subtilis GlmM enzyme. The protein assumed the typical 365 4-domain architecture previously reported for GlmM enzymes (28,37) and the "M shape" in the dimer  CdaA and GlmM has now been reported for these proteins in several Firmicutes bacteria, and hence the 382 amino acids required for the interaction might be conserved. Indeed, a ConSurf (39) analysis using 250 383 CdaA protein sequences, showed that the residue corresponding to R126 in B. subtilis CdaA is conserved 384 between the different homologs (Fig. S8). Likewise, all the three negatively charged residues, D151, 385 E154 and D195 in GlmM, which mediate the primary electrostatic interactions with R126 of CdaACD, 386 are highly conserved (Fig. S8). In previous work, we have shown, that the S. aureus DacACD (the CdaACD

416
The level of c-di-AMP is regulated by a fine balance between the activities of the cyclase, which 417 synthesizes c-di-AMP, and the phosphodiesterases, which break it down. Interestingly, these two classes 418 of enzyme appear to be regulated very differently; whereas the activity of several phosphodiesterases 419 has been shown to be regulated by small molecules, cyclase activity appears to be regulated through 420 protein-protein interaction. For example, the stringent response alarmone (p)ppGpp has been shown to 421 inhibit the activity both GdpP and PgpH enzymes (11,20). Furthermore, binding of heme to the Per-

531
Protein crystallisation, data processing and analysis 532 For crystallisation, the histidine tag was removed from the purified B. subtilis CdaACD protein. This was 533 done by incubating 10 mg purified protein with 20 U thrombin overnight at 4˚C with agitation. The 534 following day, the tag less CdaACD was purified by size-exclusion chromatography as described above.

535
The CdaACD protein was crystallized at a concentration of 4 mg/ml in 0.1M sodium cacodylate pH 6.5,

564
To assess the activity of the B. subtilis CdaACD enzymes, 20 μl enzyme reactions were set up in 100 mM 565 NaCl, 40 mM HEPES pH 7 buffer containing 10 mM MnCl2 (or 10 mM MgCl2 or 10 mM CoCl2 for 566 metal dependent assays), 100 mM ATP spiked with a-P 32 -labelled ATP (Perkin Elmer; using 0.4 μl of 567 a 3.3 μM, 250 μCi solution per 20 μl reaction) and 5 μM CdaACD enzyme. The mixture was incubated 568 at 37˚C for 4 hours, followed by heat inactivation at 95˚C for 5 minutes. After centrifugation for 10 569 minutes at 21,000 x g, 1 μl of the mixture was deposited onto a polyethylenimine-modified cellulose 570 TLC plate (Millipore) and nucleotides separated by running the plate for 20 minutes using a 3.52 M 571 (NH4)2SO4 and 1.5 M KH2PO4 buffer system mixed at a 1:1.5 v/v ratio. Radioactive signals for ATP and 572 the c-di-AMP reaction product were detected using a Typhoon FLA-700 phosphor imager. The bands 573 were quantified using the ImageQuant program and the obtained values used to calculate the percent 574 conversion of ATP to c-di-AMP. For the time course experiment, a 100 µl reaction mixture was prepared 575 as described above and incubated at 37˚C. Ten µl aliquots were removed at the indicated time points 576 and the enzyme reactions stopped by incubation the removed aliquots at 95˚C for 5 minutes. To assess 577 the activity of CdaACD in the presence of GlmM or GlmMF369, the full length GlmM protein or C-578 terminally truncated GlmM variant were added to the reaction mixture at a 1:2 (CdaACD : GlmM or 12 CdaACD : GlmMF369) molar ratio and the reactions incubated at 37˚C for 4 hours, stopped and analysed 580 as described above. The enzyme activity assays were performed in triplicates with two independently 581 purified protein preparations.

584
For the SAXS analysis, purified CdaACD, GlmM, CdaACD:GlmM complex and CdaACD:GlmMF369 585 complex protein samples where purified by size exclusion chromatography as described above and 586 subsequently concentrated to 5 mg/ml for CdaACD, 24 mg/ml each for GlmM and the CdaACD:GlmM 587 complex and 10 mg/ml for the CdaACD:GlmMF369 complex. Next, 50 µl protein samples were loaded on 588 a high pressure Shodex column (KW403: range 10 kDa to 700 kDa) fitted to an Agilent 1200 HPLC 589 system at the B21 beamline at the Diamond Light Source (Didcot, UK). The size-exclusion column was 590 equilibrated with 30 mM Tris pH 7.5, 150 mM NaCl buffer prior to loading the protein sample and the 591 data were collected continuously throughout the protein elution. The analysis of the datasets was done 592 via ScÅtter (51) using the scattering frames corresponding to the elution peaks. The ab-initio analysis 593 of the SAXS data to reconstruct a low-resolution shape of the model was done using DAMAVER 594 (DAMMIF) program (52) which performs 13 ab-initio runs to generate models from each run that were 595 averaged to determine the most persistent three-dimensional shape of the protein.

638
Tables 806 Table 1: Crystallographic data and refinement statistics