The second messenger c-di-GMP [bis-(3′-5′)-cyclic dimeric guanosine monophosphate] plays a key role in bacterial growth, survival and pathogenesis, and thus its intracellular homeostasis should be finely maintained. Mycobacterium smegmatis encodes a GAF (mammalian cGMP-regulated phosphodiesterases, Anabaenaadenylyl cyclases and Escherichia coli transcription activator FhlA) domain containing bifunctional enzyme DcpA (diguanylate cyclase and phosphodiesterase A) that catalyzes the synthesis and hydrolysis of c-di-GMP. Here, we found that M. smegmatis DcpA catalyzes the hydrolysis of c-di-GMP at a higher velocity, compared with synthetic activity, resulting in a sum reaction from the ultimate substrate GTP to the final product pGpG [5′-phosphoguanylyl-(3′-5′)-guanosine]. Fusion with the N-terminal GAF domain enables the GGDEF (Gly-Gly-Asp-Glu-Phe) domain of DcpA to dimerize and accordingly gain synthetic activity. Screening of putative metabolites revealed that GDP is the ligand of the GAF domain. Binding of GDP to the GAF domain down-regulates synthetic activity, but up-regulates hydrolytic activity, which, in consequence, might enable a timely response to the transient accumulation of c-di-GMP at the stationary phase or under stresses. Combined with the crystal structure of the EAL (Glu-Ala-Leu) domain and the small-angle X-ray scattering data, we propose a putative regulatory model of the GAF domain finely tuned by the intracellular GTP/GDP ratio. These findings help us to better understand the concerted control of the synthesis and hydrolysis of c-di-GMP in M. smegmatis in various microenvironments.

Introduction

Bis-(3′,5′)-cyclic dimeric guanosine monophosphate (c-di-GMP) is an important second messenger produced by a variety of bacteria, which was first identified as an allosteric activator of cellulose synthase in Gluconacetobacter xylinus 30 years ago [1]. As a ubiquitous intracellular signaling molecule, c-di-GMP plays key roles in many cellular processes to help bacteria adapt to diverse environments [2]. For instance, it stimulates the synthesis of various adhesins and exopolysaccharide matrix substances in biofilms [3], or reduces cell motility by down-regulating flagellar expression or assembly [4]. Moreover, c-di-GMP regulates the expression of some bacterial virulence genes, as demonstrated in Vibrio cholerae [5,6]. Upon activating the stimulator of interferon genes (STING), c-di-GMP induces the secretion of type I interferon in response to microbial invasion [7]. In addition, the c-di-GMP signaling network usually cross-talks with other global regulatory pathways, such as the phosphorylation networks [8], quorum-sensing pathways [9] and other second messenger (e.g. cyclic GMP, cyclic AMP and guanosine tetraphosphate) involved pathways [1012]. Altogether, c-di-GMP is essential for the growth, survival and pathogenesis of bacteria.

In fact, the intracellular homeostasis of c-di-GMP is finely tuned by a couple of enzymes with opposing activities, namely diguanylate cyclases (DGCs) and phosphodiesterases (PDEs) [13]. One molecule of c-di-GMP is synthesized by DGCs with two molecules of GTP, or degraded into one molecule of 5′-phosphoguanylyl-(3′-5′)-guanosine (pGpG) or two molecules of GMP by PDEs. DGCs typically harbor a GGDEF (Gly-Gly-Asp-Glu-Phe) domain, named after the conserved GG(D/E)EF active site [14], which is critical for enzymatic activity [15]. PDEs containing an EAL (Glu-Ala-Leu) domain can hydrolyze c-di-GMP into the linear pGpG [16], which in some cases could be further cleaved into GMP by oligoribonucleases [17], whereas the HD-GYP-containing PDEs produce the ultimate products of two molecules of GMP [18]. Likewise, the EAL domain or HD-GYP domain is also named after the conserved motif at the active site.

The GGDEF domain and the EAL (or HD-GYP) domain could be encoded by two individual genes or a single fusion gene [19]. Interestingly, a fusion protein with both GGDEF and EAL domains usually harbors one or several regulatory domains at the N-terminus (Figure 1a), which are supposed to modulate the enzymatic activities in response to the internal or external stresses [2]. The regulatory domains fused with the isolated GGDEF domain or the EAL domain have been extensively investigated [20,21]; however, the regulatory mechanisms that co-ordinate the combined GGDEF and EAL domains in a single protein remain largely unknown. Among the six representatives in Figure 1a, Legionella pneumophila Lpl0329 possesses two regulatory domains, namely a receiver (REC) domain and a Per-Arnt-Sim (PAS) domain. Phosphorylation of the REC domain of Lpl0329 results in a decrease of the synthetic activity toward c-di-GMP, but no effect to the hydrolytic activity [22]. In addition, the PAS and PAS-associated C-terminal motif (PAC) regulatory domains of Streptomyces coelicolor RmdA were shown to specifically bind to hemin, the regulatory mechanism of which remains unclear [23].

DcpA possesses dual activities.

Figure 1.
DcpA possesses dual activities.

(a) Domain organization of several proteins from Gram-positive and Gram-negative bacteria that contain tandem GGDEF and EAL domains. The boundaries of each domain of DcpA were defined according to the multi-sequence alignment. The NCBI accession numbers of all the proteins are as follows: Mycobacterium smegmatis DcpA, WP_003893571.1; Mycobacterium tuberculosis Rv1354c, NP_215870.1; Streptomyces coelicolor RmdA, NP_625226.1; Shewanella woodyi Swoo_2750, ACA87026.1; Rhodobacter sphaeroides BphG1, YP_345384.1; Legionella pneumophila Lpl0329, CAH14560.1. REC, receiver; PAS, Per-Arnt-Sim; PAC, PAS-associated C-terminal motif; GAF, cGMP-regulated PDEs, adenylyl cyclases and FhlA. (b) Multiple-sequence alignment of the catalytic residues of DcpA and homologs. The alignment was performed with the programs Multalin and Espript. The corresponding secondary structural elements of DcpA are displayed above the sequences and the catalytic residues are marked with red triangles. All sequences were downloaded from the NCBI database (www.ncbi.nlm.nih.gov) with the following accession numbers: Mycobacterium smegmatis, WP_003893571.1; Mycobacterium tuberculosis, NP_215870.1; Rhodococcus fascians, WP_052067107.1; Thalassobacillus devorans, WP_084315756.1; Pseudorhodoferax sp. Leaf274, WP_056901087.1; Bacillus cereus, AIE81798.1; beta proteobacterium, KPF59561.1; Oxalobacteraceae bacterium, WP_020703251.1; Desulfovibrio magneticus, WP_024825048.1. (c) HPLC spectra to show the dual activities of DcpA at different time courses. (d) HPLC spectra of mutants DcpAE259Q and DcpAD506N&D507N toward GTP and c-di-GMP, respectively.

Figure 1.
DcpA possesses dual activities.

(a) Domain organization of several proteins from Gram-positive and Gram-negative bacteria that contain tandem GGDEF and EAL domains. The boundaries of each domain of DcpA were defined according to the multi-sequence alignment. The NCBI accession numbers of all the proteins are as follows: Mycobacterium smegmatis DcpA, WP_003893571.1; Mycobacterium tuberculosis Rv1354c, NP_215870.1; Streptomyces coelicolor RmdA, NP_625226.1; Shewanella woodyi Swoo_2750, ACA87026.1; Rhodobacter sphaeroides BphG1, YP_345384.1; Legionella pneumophila Lpl0329, CAH14560.1. REC, receiver; PAS, Per-Arnt-Sim; PAC, PAS-associated C-terminal motif; GAF, cGMP-regulated PDEs, adenylyl cyclases and FhlA. (b) Multiple-sequence alignment of the catalytic residues of DcpA and homologs. The alignment was performed with the programs Multalin and Espript. The corresponding secondary structural elements of DcpA are displayed above the sequences and the catalytic residues are marked with red triangles. All sequences were downloaded from the NCBI database (www.ncbi.nlm.nih.gov) with the following accession numbers: Mycobacterium smegmatis, WP_003893571.1; Mycobacterium tuberculosis, NP_215870.1; Rhodococcus fascians, WP_052067107.1; Thalassobacillus devorans, WP_084315756.1; Pseudorhodoferax sp. Leaf274, WP_056901087.1; Bacillus cereus, AIE81798.1; beta proteobacterium, KPF59561.1; Oxalobacteraceae bacterium, WP_020703251.1; Desulfovibrio magneticus, WP_024825048.1. (c) HPLC spectra to show the dual activities of DcpA at different time courses. (d) HPLC spectra of mutants DcpAE259Q and DcpAD506N&D507N toward GTP and c-di-GMP, respectively.

The intracellular human pathogen Mycobacterium tuberculosis is a causative agent of contagious tuberculosis and leads to millions of deaths annually around the world. Rv1354c is composed of the GGDEF and EAL domains in tandem and an N-terminal GAF domain (Figure 1a), which was named after mammalian cGMP-regulated PDEs, Anabaena adenylyl cyclases and E. coli transcription activator FhlA [24]. Rv1354c was proposed to be responsible for the synthesis and hydrolysis of c-di-GMP in M. tuberculosis [25]. Owing to the ease of growth and non-pathogenicity, Mycobacterium smegmatis usually serves as a model organism to study M. tuberculosis [26]. The ortholog of Rv1354c, M. smegmatis DcpA (diguanylate cyclase and phosphodiesterase A), shares a sequence identity of 63% with the counterpart domain in M. tuberculosis. Deletion of dcpA weakens long-term survival under nutrient starvation, whereas the phenotype can be complemented by Rv1354c [27]. However, the precise regulatory role of the N-terminal GAF domain of DcpA in co-ordinating the synthesis and hydrolysis of c-di-GMP remains unknown to date.

Here, we found that M. smegmatis DcpA indeed possesses dual activities to catalyze both the synthesis and hydrolysis of c-di-GMP using GTP as the ultimate substrate; however, the hydrolytic activity is higher than the synthetic activity, resulting in a sum reaction of producing only the final product pGpG. Notably, the dimerization mediated by the N-terminal GAF domain is indispensable for the synthetic activity of the GGDEF domain, whereas the full-length DcpA possesses a much higher hydrolytic activity compared with the isolated EAL domain. After screening the putative metabolites, we identified that GDP is the ligand of the GAF domain. Binding of GDP to the GAF domain decreases synthetic activity, but increases hydrolytic activity, which in consequence leads to the accelerated consumption and limited production of c-di-GMP, reminiscent of the decrease of the GTP/GDP ratio that enables the timely elimination of the accumulated c-di-GMP at the stationary phase or under stresses. In addition, we solved the crystal structure of the EAL domain, each subunit of which folds into a classic triosephosphate isomerase (TIM) barrel structure. Combining with the small-angle X-ray scattering (SAXS) data, we propose a putative model in which GDP binding to the GAF domain triggers conformational changes in the overall dimeric structure, which in consequence co-ordinates the dual activities of DcpA. These findings help us to better understand the fine control of the synthesis and hydrolysis of c-di-GMP in M. smegmatis and might provide hints for studying its homolog in M. tuberculosis.

Experimental procedures

Cloning, expression and purification of DcpA and mutants

The coding region of DcpA (MSMEG_2196) was amplified from the genomic DNA of M. smegmatis str. mc2 155 using the primers in Supplementary Table S1, and was cloned into a pET28a-derived vector with an N-terminal 6×His-tag. The construct was overexpressed in E. coli BL21 (DE3) strain (Novagen) at 16°C for 20 h after induction by 0.2 mM isopropyl β-d-1-thiogalactopyranoside at an A600 nm of 0.8. Cells were harvested and resuspended in the lysis buffer (20 mM HEPES (pH 8.0), 100 mM NaCl and 5% glycerol), and disrupted by sonication. After centrifugation, the supernatant was loaded onto a Ni-NTA column (Qiagen) equilibrated with the binding buffer (20 mM HEPES (pH 8.0), 100 mM NaCl and 5% glycerol). The target protein was eluted with 400 mM imidazole and further purified by gel filtration (Superdex 200, GE Healthcare) in the binding buffer plus 2 mM MgCl2. The fractions containing the target protein were collected and concentrated to 10 mg/ml for crystallization. For enzymatic activity assays, the protein samples were pooled at the highest peak fractions without concentration and stored at −80°C with 30% glycerol. The purity of protein was assessed by 12% (w/v) gel electrophoresis.

The selenomethionine (SeMet)-substituted DcpA was overexpressed in E. coli B834 (DE3) (Novagen). The transformed cells were grown at 37°C in SeMet medium (M9 medium supplemented with 50 mg/l SeMet and other essential amino acids at 50 mg/l) to an A600 nm of 0.8. The following steps in protein expression and purification were the same as those for the native protein.

For the enzymatic activities, we constructed several truncated versions of DcpA, namely DcpAGAF (Met1-Asp180), DcpAGGDEF (Asp180-Ser339), DcpAEAL (Asn350-Met615) and DcpAGAF–GGDEF (Met1-Asp339). They were overexpressed and purified in the same manner as the full-length DcpA, except that DcpAEAL was overexpressed in BL21 pKY206 (DE3) (Novagen). Mutants DcpAE259Q and DcpAD506N&D507N were obtained using the MutExpress Fast Mutagenesis Kit with the plasmid encoding the wild-type DcpA as the template. They were expressed, purified and stored as described above. The sequences of the primers for cloning the mutants of DcpA are listed in Supplementary Table S1.

Analytical gel filtration chromatography was performed to estimate the molecular mass of DcpA and mutants in the absence or presence of GDP in solution by a Superdex 200 Increase 10/300 column (GE Healthcare). Standard molecular markers used for calibration are as follows: ribonuclease A (13.7 kDa), ovalbumin (43.0 kDa), conalbumin (75.0 kDa), aldolase (158.0 kDa), ferritin (440.0 kDa) and thyroglobulin (669.0 kDa).

Crystallization, data collection and processing

Crystals of the SeMet-substituted proteins were grown at 16°C using the hanging-drop vapor-diffusion method, with a drop of 1 µl protein solution mixed with an equal volume of the reservoir solution (8% polyethylene glycol 6000, 0.1 M MgCl2 and 0.1 M MES (pH 6.0)). After optimization, crystals were transferred to cryoprotectant (reservoir solution supplemented with 30% glycerol) and flash-cooled with liquid nitrogen. The X-ray diffraction data were collected at 100 K in a liquid nitrogen stream using beamline BL19U1 with a PILATUS 6M detector at the Shanghai Synchrotron Radiation Facility (SSRF). All diffraction data were integrated and scaled with the program HKL2000 [28].

Structure determination and refinement

The structure of MorA (PDB ID: 4RNH) was used as the search model to determine the structure of the EAL domain by molecular replacement using the Molrep program [29] in the CCP4i program suite [30]. The model was further refined by the maximum likelihood method implemented in REFMAC5 [31] as part of the CCP4i program suite and rebuilt interactively with the program COOT [32]. Subsequent refinements were performed by REFMAC5 and COOT. The final model was evaluated with the web service MolProbity [33]. The crystallographic parameters are listed in Table 1. The interface area was calculated by PDBePISA [34]. All structure figures were prepared with PyMOL (https://pymol.org).

Table 1
Crystal parameters, data collection and structure refinement
 Se-DcpAEAL 
Data collection 
 Space group I422 
 Unit cell parameters 
  a, b, c (Å) 129.1, 129.1, 165.5 
  α, β, γ (°) 90, 90, 90 
 Resolution range (Å) 101.8–3.0 (3.2–3.0)1 
 Unique reflections 14 428 (2235) 
 Completeness (%) 99.4 (97.4) 
 11.1 (2.1) 
Rmerge2 (%) 18.7 (89) 
 Average redundancy 8.3 (8.7) 
Structure refinement 
 Resolution range (Å) 101.8–3.0 
 No. of reflections 13 688 
R-factor3/R-free4 (%) 25.6/31.0 
 No. of protein atoms 3370 
 No. of water atoms 29 
 Mean B factors (Å269.6 
 RMSD5 
  Bond lengths (Å) 0.0047 
  Bond angles (°) 0.9585 
 Ramachandran plot6 (% of residues)  
  Most favored 96.8 
  Additional allowed 3.2 
  Outliers 
 PDB entry 5YRP 
 Se-DcpAEAL 
Data collection 
 Space group I422 
 Unit cell parameters 
  a, b, c (Å) 129.1, 129.1, 165.5 
  α, β, γ (°) 90, 90, 90 
 Resolution range (Å) 101.8–3.0 (3.2–3.0)1 
 Unique reflections 14 428 (2235) 
 Completeness (%) 99.4 (97.4) 
 11.1 (2.1) 
Rmerge2 (%) 18.7 (89) 
 Average redundancy 8.3 (8.7) 
Structure refinement 
 Resolution range (Å) 101.8–3.0 
 No. of reflections 13 688 
R-factor3/R-free4 (%) 25.6/31.0 
 No. of protein atoms 3370 
 No. of water atoms 29 
 Mean B factors (Å269.6 
 RMSD5 
  Bond lengths (Å) 0.0047 
  Bond angles (°) 0.9585 
 Ramachandran plot6 (% of residues)  
  Most favored 96.8 
  Additional allowed 3.2 
  Outliers 
 PDB entry 5YRP 
1

The values in parentheses refer to statistics in the highest bin.

2

, where Ii(hkl) is the intensity of an observation and is the mean value for its unique reflection. Summations are over all reflections.

3

R-factor = ∑h|Fo(h) − Fc(h)|/∑hFo(h), where Fo and Fc are the observed and calculated structure factor amplitudes, respectively.

4

R-free was calculated with 5% of the data excluded from the refinement.

5

Root-mean-square deviation from ideal values.

6

Categories were defined by Molprobity.

Surface plasmon resonance assays

All surface plasmon resonance (SPR) experiments were performed at 25°C using a Biacore 3000 instrument (GE Healthcare) in phosphate-buffered saline (PBS), pH 7.4. The GAF domain was covalently immobilized on the carboxymethyl-dextran surface of the CM5 chip. Putative metabolites or different concentrations of GDP in the running buffer were serially injected over the sensor chips with the immobilized protein and the blank for 2 min at a flow rate of 30 µl/min. Afterwards, the sensor chip surface was regenerated after washing with PBS. The equilibrium responses were fitted with the linear fit and steady-state affinity mode using the BIA evaluation software. The graphs were plotted using the Origin 8.1 software.

Enzymatic activity assays

The standard samples were bought from Sigma–Aldrich and Biolog. The activities of DcpA and its mutants were assayed by high-performance liquid chromatography (HPLC) and performed in the buffer containing 100 mM HEPES (pH 7.5), 100 mM NaCl and 10 mM MgCl2. A 20 µl reaction mixture containing 0.5 mM GTP and 5 µM full-length DcpA was incubated at 37°C for different time courses. Then, the reactions were terminated by heating at 95°C for 10 min. After centrifugation, 10 µl of the supernatant was applied to the HPLC system (Agilent 1200 Series, U.S.A.). The buffer containing 20 mM KH2PO4/K2HPO4 (pH 7.0) and 5% methanol was used for pre-equilibration of the column (Zorbax 300 SB-C18 column, 4.6 × 250 mm, Agilent, U.S.A.) and separation of the components at a flow rate of 1 ml/min. The column temperature was kept at 25°C. The eluted fractions were detected and quantitated by the absorbance at 254 nm. The details of the enzymatic reactions concerning DcpAGGDEF, DcpAGAF–GGDEF, DcpAEAL, DcpAE259Q and DcpAD506N&D507N are shown in corresponding figure legends. The detection and quantitation of each reaction were the same as for DcpA. All assays were performed in three independent experiments to calculate the means and standard deviations.

Circular dichroism spectrometry

DcpA was purified in a PBS (pH 7.4), using the protocol used to purify proteins for crystallization. DcpA in the absence of GDP was diluted in the same buffer to 0.6 mg/ml, and its secondary-structure compositions were analyzed using a Jasco J-810 circular dichroism (CD) spectropolarimeter. Spectra were recorded from 190 to 280 nm with an interval of 1 nm at room temperature. The final spectrum represents the average data of three consecutive scans. The secondary structures of DcpA in the presence of GDP were analyzed in the same manner.

SAXS experiments

SAXS was used to investigate the overall conformation of DcpA in the absence or presence of GDP. SAXS experiments were performed at beamline BL19U2 of the National Center for Protein Science Shanghai (NCPSS) at SSRF [35]. Scattered X-ray intensities were collected using a Pilatus 1 M detector (DECTRIS Ltd). The measurements were carried out with 20 × 1 s exposures and scattering profiles for 20 frames were compared to evade the possible sample radiation damage. To reduce the radiation damage, a flow cell made of a cylindrical quartz capillary with a diameter of 1.5 mm and a wall of 10 µm was used during the measurements. The scattering was recorded in the range of the momentum transfer 0.01 < q < 0.45 Å−1, where q = (4πsinθ)/λ, 2θ is the scattering angle and λ = 1.033 Å is the X-ray wavelength. To exclude concentration dependence and a possible influence of the interaction of the protein macromolecules in solution, 1.0, 2.5 and 5.0 mg/ml protein concentrations were prepared and measured. The 2D scattering images were converted into 1D SAXS curves through azimuthal averaging after solid-angle correction and then normalized with the intensity of the transmitted X-ray beam, using the software package BioXTAS RAW [36]. The scattering curves were primarily processed using the program PRIMUS [37] in the ATSAS software package [38]. After subtraction from the buffer (20 mM HEPES (pH 7.5), 100 mM NaCl, 5% glycerol and 2 mM MgCl2, with or without GDP) scattering, the data were averaged and extrapolated to zero concentration. A linear Guinier plot was used to evaluate the radius of gyration (Rg) in the low q range (q × Rg < 1.3). The maximal diameter of the scattering objects Dmax and the pair distance distribution function P(r) were calculated by the program GNOM [39]. The ab initio reconstruction of each sample was performed using the program DAMMIN [40] with 20 runs for each experimental group, and DAMAVER [41] was used to analyze the normalized spatial discrepancy between 20 models. The filtered SAXS model by DAMFILT [41] was shown in PyMOL. Rigid-body modeling structures of the multi-domain protein DcpA and DcpA-GDP complex against the scattering curve were calculated with the program CORAL [38]. The theoretical scattering curves from the calculated rigid-body modeling structure of apo DcpA and liganded DcpA were calculated with the program CRYSOL [42] and compared with relative scattering profile with a χ2 value to indicate the discrepancy. The crystal structures and the SAXS models were matched using SUPCOMB [43]. The SAXS parameters were listed in Supplementary Table S2.

Results

DcpA catalyzes the synthesis and hydrolysis of c-di-GMP in the presence of GTP

The 615-residue DcpA from M. smegmatis contains three distinct domains: an N-terminal GAF domain, followed by a GGDEF domain and a C-terminal EAL domain (Figure 1a). The boundaries of each domain were defined according to the secondary-structure-based multiple-sequence alignment, and the catalytic residues of the GGDEF/EAL domain were deduced from the homologs (Figure 1b). Notably, the pattern of domain organization and putative catalytic residues are highly conserved in most mycobacterial species [27], suggesting that these fusion proteins should all possess both synthetic and hydrolytic activities. Nevertheless, most previously identified fusion proteins with both GGDEF and EAL domains only exhibit either synthetic or hydrolytic activity due to the mutation at one active site [19]. To verify if DcpA has dual activities, we applied the purified DcpA to the activity assays. As shown in Figure 1c, both c-di-GMP and pGpG could be detected at a proper time point, indicating that DcpA can use GTP to synthesize c-di-GMP, which is, in turn, hydrolyzed into pGpG, in agreement with the previous report [27]. However, only pGpG is accumulated in the reaction mixture with time, and c-di-GMP is maintained at a residual level, indicating that the reaction velocity of hydrolysis is much higher than that of synthesis (Figure 1c).

The fourth residue (Glu) at the conserved GG(D/E)EF active site was proposed to co-ordinate the metal ion that drives the synthetic activity [20], whereas the conserved DDFGTGYSS motif of the EAL domain is essential for the hydrolytic activity beyond the EAL motif [44]. Indeed, a single mutation E259Q (termed DcpAE259Q) at the GG(D/E)EF active site or double mutation D506N&D507N (termed DcpAD506N&D507N) at the DDFGTGYSS motif of DcpA completely abolished the corresponding activity, further proving that these residues are catalytic residues (Figure 1d).

The GAF domain is essential for the activities of DcpA

In many cases, the fused regulatory domain could induce the dimerization of DGC, which is indispensable for the synthetic activity [45]. Analytical gel filtration indicated that the isolated GAF domain (termed DcpAGAF) or GGDEF domain (termed DcpAGGDEF) is monomeric in solution, whereas the full-length DcpA and the EAL domain-deleted version (termed DcpAGAF–GGDEF) are dominantly dimeric (Figure 2a). In addition, the results of multiple-angle light-scattering (MALS) assays further supported that DcpAGAF and DcpAGGDEF are monomeric in solution, whereas DcpAGAF–GGDEF is most probably a dimer (Supplementary Figure S1). The results suggested that the GAF domain is required for the dimerization of the GGDEF domain. Moreover, activity assays indicated that DcpAGAF–GGDEF, but not DcpAGGDEF, possesses synthetic activity (Figure 2b). Therefore, it suggested that the GAF-mediated dimerization is indispensable for the synthetic activity of DcpA.

The GAF domain is essential for the activities of DcpA.

Figure 2.
The GAF domain is essential for the activities of DcpA.

(a) Analytical gel filtration profiles of the full-length and truncated DcpA. The molecular mass of each sample was calculated according to the standard curve. (b) HPLC spectra of the synthetic activity of DcpAGGDEF and DcpAGAF–GGDEF. DcpAGGDEF (20 µM) was incubated with 0.2 mM GTP at 37°C for 12 h, whereas the reaction time for DcpAGAF–GGDEF was 1 h. (c) HPLC spectra of the hydrolytic activity of DcpAEAL and DcpA. The reaction lasted for 2 h at 37°C in the presence of 10 µM enzyme and 0.2 mM c-di-GMP. The concentrations of substrates were chosen to ensure the reaction is in the linear interval.

Figure 2.
The GAF domain is essential for the activities of DcpA.

(a) Analytical gel filtration profiles of the full-length and truncated DcpA. The molecular mass of each sample was calculated according to the standard curve. (b) HPLC spectra of the synthetic activity of DcpAGGDEF and DcpAGAF–GGDEF. DcpAGGDEF (20 µM) was incubated with 0.2 mM GTP at 37°C for 12 h, whereas the reaction time for DcpAGAF–GGDEF was 1 h. (c) HPLC spectra of the hydrolytic activity of DcpAEAL and DcpA. The reaction lasted for 2 h at 37°C in the presence of 10 µM enzyme and 0.2 mM c-di-GMP. The concentrations of substrates were chosen to ensure the reaction is in the linear interval.

Furthermore, it showed that the hydrolytic activity of the isolated EAL domain (termed DcpAEAL) is much lower than that of the full-length DcpA (Figure 2c), suggesting that fusion domains contribute to the higher hydrolytic activity of DcpA. A previous study reported that the tandem GGDEF and EAL domains of DcpA, after deleting the GAF domain, showed no hydrolytic activity in vitro [27]. This indicated that the GAF domain could augment the hydrolytic activity of the EAL domain. Notably, the role of the GAF domain in the dual activities of DcpA is similar to that of the previously reported regulatory domain of Klebsiella pneumoniae BlrP1 [2].

GDP is the ligand of the GAF domain

The GAF domains, which are distributed widely from Archaea to mammals, represent one of the largest families of small-molecule-binding regulatory domains [46]. Though the GAF domains from different species display a similar three-dimensional structure, the binding ligands differ a lot, making it difficult to identify the ligand of the GAF domain only based on sequence alignment and/or structural comparison. It is known that most of the reported ligands are closely related to the corresponding signaling pathways regulated by the GAF domains. The up-regulation of the dcpA gene upon starvation or depletion of carbon source [47] strongly indicated that the ligand of the GAF domain is most probably involved in energy metabolism. Following these hints and taking the previously identified ligands into account, 26 putative metabolites were applied to the binding affinity screening using SPR. As displayed in Figure 3a, DcpAGAF only binds to GDP, but neither the previously reported ligands valine, isoleucine and pyruvate, nor the substrate GTP and analog GMP or cAMP. Furthermore, we determined the binding affinity of DcpAGAF toward GDP and revealed an equilibrium dissociation constant (Kd) of 2.25 mM (Figure 3b), which is quite similar to that with the full-length DcpA (Kd of 2.21 mM) (Figure 3c). However, the individual GGDEF or EAL domain does not interact with GDP.

GDP is the ligand of the GAF domain and could regulate the activities of DcpA.

Figure 3.
GDP is the ligand of the GAF domain and could regulate the activities of DcpA.

(a) SPR sensor diagram for the putative ligands against the immobilized DcpAGAF. The concentration of each small molecule is 1 mM. The steady-state equilibrium curves of (b) DcpAGAF and (c) the full-length DcpA against GDP. The binding affinity of DcpAGAF or full-length DcpA toward GDP was calculated from the data at the concentration varying from 0.2 to 3.2 mM using the linear fit and steady-state affinity mode. The relative (d) synthetic and (e) hydrolytic activity of DcpA in the presence of various concentrations of GDP. For the synthetic activity, 5 µM DcpAD506N&D507N was incubated with 0.2 mM GTP for 45 min at 37°C. For the hydrolytic activity, 5 µM DcpAE259Q was incubated with 0.5 mM c-di-GMP for 30 min at 37°C. (f) The relative synthetic activity of DcpA at different GTP/GDP ratios. Various concentrations of GDP were applied to the activity assays, with the concentration of GTP fixed at 0.5 mM. The reactions were started by adding 5 µM enzyme and terminated after 45 min. The experimental conditions for two mutants were chosen to make the reaction locate in the linear interval. The assays were performed by HPLC, and the relative activity was defined based on the peak areas of product. The error bars represent SD from three independent assays with three different protein preparations. Two-tailed Student's t-test is used for the comparison of statistical significance. The P-value of <0.001 is indicated with ***.

Figure 3.
GDP is the ligand of the GAF domain and could regulate the activities of DcpA.

(a) SPR sensor diagram for the putative ligands against the immobilized DcpAGAF. The concentration of each small molecule is 1 mM. The steady-state equilibrium curves of (b) DcpAGAF and (c) the full-length DcpA against GDP. The binding affinity of DcpAGAF or full-length DcpA toward GDP was calculated from the data at the concentration varying from 0.2 to 3.2 mM using the linear fit and steady-state affinity mode. The relative (d) synthetic and (e) hydrolytic activity of DcpA in the presence of various concentrations of GDP. For the synthetic activity, 5 µM DcpAD506N&D507N was incubated with 0.2 mM GTP for 45 min at 37°C. For the hydrolytic activity, 5 µM DcpAE259Q was incubated with 0.5 mM c-di-GMP for 30 min at 37°C. (f) The relative synthetic activity of DcpA at different GTP/GDP ratios. Various concentrations of GDP were applied to the activity assays, with the concentration of GTP fixed at 0.5 mM. The reactions were started by adding 5 µM enzyme and terminated after 45 min. The experimental conditions for two mutants were chosen to make the reaction locate in the linear interval. The assays were performed by HPLC, and the relative activity was defined based on the peak areas of product. The error bars represent SD from three independent assays with three different protein preparations. Two-tailed Student's t-test is used for the comparison of statistical significance. The P-value of <0.001 is indicated with ***.

GDP regulates the activities of DcpA

To investigate the putative effect of GDP on the synthetic and hydrolytic activities, we compared the corresponding activities of DcpA in the presence and absence of GDP. The double-mutant DcpAD506N&D507N that only possesses the synthetic activity and the single-mutant DcpAE259Q that just retains the hydrolytic activity (Figure 1d) were used as the positive controls for the comparisons. Upon the addition of GDP to 0.05 and 0.2 mM, the synthetic activity of DcpAD506N&D507N decreased to two-third and one-quarter, respectively (Figure 3d). Once the GDP concentration reached 1 mM, no synthetic activity could be detected (Figure 3d). In contrast, the addition of GDP sharply increased the hydrolytic activity of DcpAE259Q to 4–30-fold, in a GDP concentration-dependent manner (Figure 3e). These results revealed that binding of GDP to the GAF domain could down-regulate the synthetic activity, but up-regulate the hydrolytic activity of DcpA, which in consequence might lead to the consumption of intracellular c-di-GMP in M. smegmatis at a much faster rate in response to some given environmental stimuli.

It is known that the intracellular GTP/GDP ratio is strongly correlated to the growth phase and supply of environmental nutrients, and thus involved in bacterial growth and survival at starvation [48]. Previous reports showed that during the exponential growth phase, the intracellular concentration of c-di-GMP is maintained at a basal level [49], which is correlated to the high GTP/GDP ratio at this phase. In contrast, at the stationary phase or upon starvation, the intracellular c-di-GMP is accumulated and reaches a relatively high level [27,49], which in turn slows down the growth rate and initiates stress responses, such as biofilm formation, to help bacterial survival and adaption [27,50]. However, the accumulated c-di-GMP should be timely degraded to keep its intracellular homeostasis. On one hand, although the bifunctional enzyme DcpA consumes c-di-GMP at a higher velocity compared with its production in the normal condition, a decreasing intracellular GTP/GDP ratio will trigger an accelerated degradation of c-di-GMP (Figures 1c and 3e). On the other hand, the synthetic activity of DcpA is down-regulated along with the decrease in the intracellular GTP/GDP ratio (Figure 3f), leading to the very limited production of c-di-GMP. These dual mechanisms will guarantee the timely down-regulation of intracellular c-di-GMP to the basal level. It is also consistent with the previous finding of increased promoter activity of dcpA upon starvation [27].

Overall structure of the EAL domain

To obtain structural insights, we applied the full-length DcpA to crystallization. However, the refined structure from the optimized crystal only contains the EAL domain (DcpAEAL, Ala364-Pro606), which is probably due to the degradation of the other two domains during crystallization. Each asymmetric unit contains two molecules of the EAL domain, which form a homodimer with a total buried interface area of 895 Å2. The interface is stabilized by helices α4 and α6 in addition to the loop β5–α4 of each subunit (Figure 4a), sharing a dimerization pattern similar to other EAL-containing proteins, such as Thiobacillus denitrificans TBD1265, K. pneumoniae BlrP1 and Pseudomonas aeruginosa MorA [5153].

Structural analyses of DcpAEAL.

Figure 4.
Structural analyses of DcpAEAL.

(a) The overall structure of DcpAEAL. The structure is shown in cartoon representation. The two subunits are colored in cyan and red, respectively. The secondary structural elements (helices α4, α6 and strand β5) at the dimeric interface are labeled. The dash lines indicate the residues missing in the electron density map. The Mg2+ ions are shown as spheres in the same color with the corresponding subunit of DcpAEAL, whereas the water molecule (Wat) is displayed as gray spheres. (b) Structural superposition of DcpAEAL (cyan) on that of T. denitrificans TBD1265 (wheat; PDB ID: 2R6O). The water molecule of TBD1265 is shown as light blue spheres. The inset shows a close-up view of the active site. The metal-coordinated and catalytic residues of two structures are shown as sticks; however, only those of DcpAEAL are labeled. (c) Structural superposition of DcpAEAL (cyan) on the c-di-GMP-complexed EAL domain of TBD1265 (pink; PDB ID: 3N3T). c-di-GMP is shown as green sticks. (d) A schematic comparison of the DDFGTGYSS motifs. The helix α5 and loop β5–α5 of monomeric MorA (PDB ID: 4RNJ)/dimeric MorA (PDB ID: 4RNI), and helix α4 and loop β5–α4 of dimeric DcpAEAL in different lengths are shown.

Figure 4.
Structural analyses of DcpAEAL.

(a) The overall structure of DcpAEAL. The structure is shown in cartoon representation. The two subunits are colored in cyan and red, respectively. The secondary structural elements (helices α4, α6 and strand β5) at the dimeric interface are labeled. The dash lines indicate the residues missing in the electron density map. The Mg2+ ions are shown as spheres in the same color with the corresponding subunit of DcpAEAL, whereas the water molecule (Wat) is displayed as gray spheres. (b) Structural superposition of DcpAEAL (cyan) on that of T. denitrificans TBD1265 (wheat; PDB ID: 2R6O). The water molecule of TBD1265 is shown as light blue spheres. The inset shows a close-up view of the active site. The metal-coordinated and catalytic residues of two structures are shown as sticks; however, only those of DcpAEAL are labeled. (c) Structural superposition of DcpAEAL (cyan) on the c-di-GMP-complexed EAL domain of TBD1265 (pink; PDB ID: 3N3T). c-di-GMP is shown as green sticks. (d) A schematic comparison of the DDFGTGYSS motifs. The helix α5 and loop β5–α5 of monomeric MorA (PDB ID: 4RNJ)/dimeric MorA (PDB ID: 4RNI), and helix α4 and loop β5–α4 of dimeric DcpAEAL in different lengths are shown.

Each subunit displays a classic TIM barrel structure composed of eight β-strands surrounded by eight α-helices, leaving 19 residues (Glu390-Ser408) of one subunit and 24 residues (Arg388-Leu411) of the other subunit missing in the electron density map (Figure 4a). Compared with the known structures of EAL domains from other species, they appear to have a very similar tertiary structure (Supplementary Figure S2). The DALI search [54] revealed that DcpAEAL shares a highest structural homology with that of T. denitrificans TBD1265 (PDB ID: 2R6O, Z-score 30.5, sequence identity 38%), with a root-mean-square deviation of 2.86 Å over 395 Cα atoms. Superposition of the two structures revealed that DcpAEAL should also possess a two metal-coordinated active site, though only one Mg2+ ion could be defined in our structure (Figure 4b). This Mg2+ ion is positioned in a place similar to the primary one of TBD1265 and co-ordinated by the side chains of Glu384, Asn444, Glu476 and Asp506 (Figure 4b). Despite the co-ordinated residues Asp507 and Asp529 at the secondary metal-binding site, or the so-called low-affinity site, being conserved, the putative Mg2+ ion is missing in our structure (Figure 4b). Notably, Asp506 and Asp507 are part of the conserved DDFGTGYSS motif, which constitutes the loop β5–α4 at the dimeric interface. Besides, the conserved Glu563 is thought to be a general base catalyst that facilitates the water-mediated nucleophilic attack toward c-di-GMP (Figure 4b).

To date, 22 c-di-GMP-complexed structures of the EAL domain from nine species have been reported. Superimposing our structure with the c-di-GMP-complexed EAL domain of T. denitrificans TBD1265 (PDB ID: 3N3T) showed that c-di-GMP is supposed to bind to the C-terminal segment of the TIM barrel in DcpAEAL (Figure 4c), a similar site of known TIM barrel enzymes [55]. A previous report showed that site-directed mutagenesis preventing dimerization did not alter the substrate-binding ability of the EAL domain, but resulted in drastically reduced hydrolytic activity, indicating that dimerization of the EAL domain is indispensable for the activity [56]. Moreover, structural studies of MorA showed that dimerization leads to a shorter helix α5 of only six residues (corresponding to helix α4 in our structure) accompanying with an elongated loop β5–α5 of nine residues (corresponding to the loop β5–α4 in our structure) at the dimeric interface, eventually allowing the two aspartate residues to enter the active site for co-ordinating the catalytic metal ions [52]. A structure-based comparison showed that DcpAEAL is inclined to be an activated dimer (Figure 4d).

Overall conformational change of DcpA induced by GDP using SAXS

To investigate how the GAF domain regulates the dual activities upon GDP binding, we applied SAXS to compare the overall three-dimensional structures of DcpA in the presence and absence of GDP. The Guinier regions for both apo and liganded DcpA experimental groups are linear (Figure 5a), indicating that the measured system is monodispersed and homogeneous in solution. In addition, the experimental molecular mass obtained in SAXS indicated that DcpA adopts a dimeric structure either in the presence or absence of GDP, which is consistent with the result of analytical gel filtration (Figure 5b), indicating that there is no contribution of aggregates to the scatter. Based on the SAXS data, the calculated maximum particle dimension (Dmax) for apo DcpA is 169.5 Å, whereas that of the liganded is 200.5 Å (Figure 5c), revealing that DcpA becomes somewhat extended in the presence of GDP. Notably, the CD spectra showed that apo and liganded DcpA exhibit a similar secondary-structure composition (Figure 5d). These results suggested that GDP binding to the GAF domain might induce the rearrangements of the overall conformation of dimeric DcpA.

Conformational changes between apo and liganded DcpA.

Figure 5.
Conformational changes between apo and liganded DcpA.

(a) SAXS intensity profiles of apo (red line) and liganded DcpA (gray line). The inset shows the Guinier plots of apo and liganded DcpA. (b) Analytical gel filtration profiles of DcpA in the absence and presence of GDP. The molecular mass of each sample was calculated according to the standard curve. (c) Electron pair distribution function P(r) profiles of the apo (red line) and liganded DcpA (gray line). The Dmax of apo DcpA is 169.5 Å, and significantly increased to 200.5 Å upon the addition of GDP, showing a structural extension. (d) CD spectra of DcpA in the absence and presence of GDP.

Figure 5.
Conformational changes between apo and liganded DcpA.

(a) SAXS intensity profiles of apo (red line) and liganded DcpA (gray line). The inset shows the Guinier plots of apo and liganded DcpA. (b) Analytical gel filtration profiles of DcpA in the absence and presence of GDP. The molecular mass of each sample was calculated according to the standard curve. (c) Electron pair distribution function P(r) profiles of the apo (red line) and liganded DcpA (gray line). The Dmax of apo DcpA is 169.5 Å, and significantly increased to 200.5 Å upon the addition of GDP, showing a structural extension. (d) CD spectra of DcpA in the absence and presence of GDP.

Discussion

The GAF domains are widespread in all kingdoms of life through evolution for over 2 billion years [57]. They usually coexist with other catalytic domains, which are involved in a variety of cellular signaling pathways. It is well known that the GAF domains function through binding to small ligands, even though only a few of them have been identified. The hydrolytic activity of mammalian PDE5 is significantly increased upon binding of cGMP to the GAF domain, which might trigger the rearrangements of the overall conformation of the full-length protein [58]. In addition, the winged helix-turn-helix domain of Bacillus subtilis CodY splays when the N-terminal GAF domain binds to isoleucine, which in turn increases the DNA-binding affinity of CodY [59].

Bioinformatic analyses revealed that M. smegmatis encoded two enzymes MSMEG_2196 (DcpA) and MSMEG_2774 (MSDGC-2), which probably play a role in c-di-GMP synthesis/hydrolysis. We showed here DcpA that contains the tandem GGDEF and EAL domains indeed possesses both synthetic and hydrolytic activities (Figure 1c). However, it was reported that MSDGC-2 that contains only a putative GGDEF domain loses the synthetic activity due to the mutations of the catalytic residues [27]. Thus, DcpA becomes the only enzyme responsible for the synthesis and hydrolysis of c-di-GMP in M. smegmatis. Our results showed that the GAF domain is indispensable for the synthetic activity and could increase the hydrolytic activity of DcpA (Figure 2b,c). Moreover, screening of putative metabolites revealed that GDP is the ligand of the GAF domain (Figure 3a). The binding of GDP to the GAF domain down-regulates synthetic activity, but up-regulates hydrolytic activity (Figure 3d,e), thus in sum contributing to the accelerated assimilation of c-di-GMP upon environmental stresses. Then, using SAXS, we found that GDP binding to the GAF domain might induce the overall conformational change of dimeric DcpA (Figure 5).

Furthermore, we fitted the structures of the GGDEF–EAL and GAF domains into the ab initio molecular envelopes of apo and liganded DcpA, respectively. The envelopes were reconstructed from SAXS data by dummy bead modeling, and the one having the lowest normalized spatial discrepancy was picked as the most representative model. The structure of the fused GGDEF and EAL domains was gained by homology modeling based on the EAL domain of DcpA itself, combined with P. aeruginosa MorA (PDB ID: 4RNH), whereas the structure of the GAF domain was deduced from a cyanobacterial protein (PDB ID: 3W2Z). The positions of the input domains of apo and liganded DcpA were searched and optimized for 10 rounds based on the SAXS envelopes, respectively. In the model of apo DcpA (χ2 value 3.037), the GAF domain of one subunit interacts with the GGDEF domain of the other subunit, making DcpA a compact dimer (Figure 6a), in accordance with the results of analytical gel filtration (Figure 2a). However, in the presence of GDP (χ2 value 2.536), DcpA forms an extended dimer with an interface between the two EAL domains, leaving the two GAF domains kicked outwards (Figure 6b).

Superposition of the crystal structures against the SAXS models.

Figure 6.
Superposition of the crystal structures against the SAXS models.

The averaged and filtered ab initio SAXS envelopes of DcpA in the (a) absence and (b) presence of GDP were superimposed with the rigid-body modeling structures composed of GAF and GGDEF–EAL domains. The dimeric profile of DcpA is shown in green and yellow for two subunits, respectively.

Figure 6.
Superposition of the crystal structures against the SAXS models.

The averaged and filtered ab initio SAXS envelopes of DcpA in the (a) absence and (b) presence of GDP were superimposed with the rigid-body modeling structures composed of GAF and GGDEF–EAL domains. The dimeric profile of DcpA is shown in green and yellow for two subunits, respectively.

All together, these findings not only help us better understand the fine control of synthesis and hydrolysis of c-di-GMP in M. smegmatis, but also provide hints for exploration of the homolog Rv1354c in M. tuberculosis. However, more structural information and biochemical investigations are needed to clearly elucidate the fine catalytic mechanism of these dual-functional enzymes.

Abbreviations

     
  • c-di-GMP

    bis-(3′-5′)-cyclic dimeric guanosine monophosphate

  •  
  • DcpA

    diguanylate cyclase and phosphodiesterase A

  •  
  • DGC

    diguanylate cyclase

  •  
  • EAL

    Glu-Ala-Leu

  •  
  • GAF

    mammalian cGMP-regulated PDEs, Anabaenaadenylyl cyclases and Escherichia coli transcription activator FhlA

  •  
  • GGDEF

    Gly-Gly-Asp-Glu-Phe

  •  
  • HPLC

    high-performance liquid chromatography

  •  
  • Kd

    dissociation constant

  •  
  • PAS

    Per-Arnt-Sim

  •  
  • PBS

    phosphate-buffered saline

  •  
  • PDE

    phosphodiesterase

  •  
  • pGpG

    5′-phosphoguanylyl-(3′-5′)-guanosine

  •  
  • REC

    receiver

  •  
  • SAXS

    small-angle X-ray scattering

  •  
  • SeMet

    selenomethionine

  •  
  • SPR

    surface plasmon resonance

  •  
  • SSRF

    Shanghai Synchrotron Radiation Facility

  •  
  • TIM

    triosephosphate isomerase

Author Contribution

H.-J.C., Q.L., N.L., Y.C. and C.-Z.Z. designed the study, analyzed the data and wrote the paper. H.-J.C. performed the enzymatic experiments. H.-J.C., Y.L. and Y.-L.J. performed crystal screening and optimization, data collection and structural determination. H.-J.C. and Q.L. did the SAXS experiments and N.L. processed the SAXS data.

Funding

This work is supported by the National Natural Science Foundation of China (grant no. U1332114), the Ministry of Science and Technology of China (grant no. 2016YFA0400900) and the Youth Innovation Promotion Association of Chinese Academy of Science (grant no. 2017319).

Acknowledgments

We appreciate the assistance of the staff at the Shanghai Synchrotron Radiation Facility (SSRF) and the Core Facility Center for Life Sciences at the University of Science and Technology of China. We are grateful to all of the developers of the CCP4 suite, PyMOL and ESPript.

Competing Interests

The Authors declare that there are no competing interests associated with the manuscript.

References

References
1
Ross
,
P.
,
Weinhouse
,
H.
,
Aloni
,
Y.
,
Michaeli
,
D.
,
Weinberger-Ohana
,
P.
,
Mayer
,
R.
et al. 
(
1987
)
Regulation of cellulose synthesis in Acetobacter xylinum by cyclic diguanylic acid
.
Nature
325
,
279
281
2
Schirmer
,
T.
and
Jenal
,
U.
(
2009
)
Structural and mechanistic determinants of c-di-GMP signalling
.
Nat. Rev. Microbiol.
7
,
724
735
3
Römling
,
U.
and
Amikam
,
D.
(
2006
)
Cyclic di-GMP as a second messenger
.
Curr. Opin. Microbiol.
9
,
218
228
4
Wolfe
,
A.J.
and
Visick
,
K.L.
(
2008
)
Get the message out: cyclic-di-GMP regulates multiple levels of flagellum-based motility
.
J. Bacteriol.
190
,
463
475
5
Tamayo
,
R.
,
Pratt
,
J.T.
and
Camilli
,
A.
(
2007
)
Roles of cyclic diguanylate in the regulation of bacterial pathogenesis
.
Annu. Rev. Microbiol.
61
,
131
148
6
Tischler
,
A.D.
and
Camilli
,
A.
(
2005
)
Cyclic diguanylate regulates Vibrio cholerae virulence gene expression
.
Infect. Immun.
73
,
5873
5882
7
McWhirter
,
S.M.
,
Barbalat
,
R.
,
Monroe
,
K.M.
,
Fontana
,
M.F.
,
Hyodo
,
M.
,
Joncker
,
N.T.
et al. 
(
2009
)
A host type I interferon response is induced by cytosolic sensing of the bacterial second messenger cyclic-di-GMP
.
J. Exp. Med.
206
,
1899
1911
8
Lori
,
C.
,
Ozaki
,
S.
,
Steiner
,
S.
,
Böhm
,
R.
,
Abel
,
S.
,
Dubey
,
B.N.
et al. 
(
2015
)
Cyclic di-GMP acts as a cell cycle oscillator to drive chromosome replication
.
Nature
523
,
236
239
9
Srivastava
,
D.
and
Waters
,
C.M.
(
2012
)
A tangled web: regulatory connections between quorum sensing and cyclic di-GMP
.
J. Bacteriol.
194
,
4485
4493
10
An
,
S.Q.
,
Chin
,
K.H.
,
Febrer
,
M.
,
McCarthy
,
Y.
,
Yang
,
J.G.
,
Liu
,
C.L.
et al. 
(
2013
)
A cyclic GMP-dependent signalling pathway regulates bacterial phytopathogenesis
.
EMBO J.
32
,
2430
2438
11
Almblad
,
H.
,
Harrison
,
J.J.
,
Rybtke
,
M.
,
Groizeleau
,
J.
,
Givskov
,
M.
,
Parsek
,
M.R.
et al. 
(
2015
)
The cyclic AMP-Vfr signaling pathway in Pseudomonas aeruginosa is inhibited by cyclic di-GMP
.
J. Bacteriol.
197
,
2190
2200
12
Gupta
,
K.R.
,
Kasetty
,
S.
and
Chatterji
,
D.
(
2015
)
Novel functions of (p)ppGpp and cyclic di-GMP in mycobacterial physiology revealed by phenotype microarray analysis of wild-type and isogenic strains of Mycobacterium smegmatis
.
Appl. Environ. Microbiol.
81
,
2571
2578
13
Tal
,
R.
,
Wong
,
H.C.
,
Calhoon
,
R.
,
Gelfand
,
D.
,
Fear
,
A.L.
,
Volman
,
G.
et al. 
(
1998
)
Three cdg operons control cellular turnover of cyclic di-GMP in Acetobacter xylinum: genetic organization and occurrence of conserved domains in isoenzymes
.
J. Bacteriol.
180
,
4416
4425
PMID:
[PubMed]
14
Ausmees
,
N.
,
Mayer
,
R.
,
Weinhouse
,
H.
,
Volman
,
G.
,
Amikam
,
D.
,
Benziman
,
M.
et al. 
(
2001
)
Genetic data indicate that proteins containing the GGDEF domain possess diguanylate cyclase activity
.
FEMS Microbiol. Lett.
204
,
163
167
15
Malone
,
J.G.
,
Williams
,
R.
,
Christen
,
M.
,
Jenal
,
U.
,
Spiers
,
A.J.
and
Rainey
,
P.B.
(
2007
)
The structure-function relationship of WspR, a Pseudomonas fluorescens response regulator with a GGDEF output domain
.
Microbiology
153
(
Pt 4
),
980
994
16
Schmidt
,
A.J.
,
Ryjenkov
,
D.A.
and
Gomelsky
,
M.
(
2005
)
The ubiquitous protein domain EAL is a cyclic diguanylate-specific phosphodiesterase: enzymatically active and inactive EAL domains
.
J. Bacteriol.
187
,
4774
4781
17
Cohen
,
D.
,
Mechold
,
U.
,
Nevenzal
,
H.
,
Yarmiyhu
,
Y.
,
Randall
,
T.E.
,
Bay
,
D.C.
et al. 
(
2015
)
Oligoribonuclease is a central feature of cyclic diguanylate signaling in Pseudomonas aeruginosa
.
Proc. Natl Acad. Sci. U.S.A.
112
,
11359
11364
18
Ryan
,
R.P.
,
Fouhy
,
Y.
,
Lucey
,
J.F.
,
Crossman
,
L.C.
,
Spiro
,
S.
,
He
,
Y.W.
et al. 
(
2006
)
Cell-cell signaling in Xanthomonas campestris involves an HD-GYP domain protein that functions in cyclic di-GMP turnover
.
Proc. Natl Acad. Sci. U.S.A.
103
,
6712
6717
19
Hengge
,
R.
(
2009
)
Principles of c-di-GMP signalling in bacteria
.
Nat. Rev. Microbiol.
7
,
263
273
20
Chan
,
C.
,
Paul
,
R.
,
Samoray
,
D.
,
Amiot
,
N.C.
,
Giese
,
B.
,
Jenal
,
U.
et al. 
(
2004
)
Structural basis of activity and allosteric control of diguanylate cyclase
.
Proc. Natl Acad. Sci. U.S.A.
101
,
17084
17089
21
Hasegawa
,
K.
,
Masuda
,
S.
and
Ono
,
T.A.
(
2006
)
Light induced structural changes of a full-length protein and its BLUF domain in YcgF (Blrp), a blue-light sensing protein that uses FAD (BLUF)
.
Biochemistry
45
,
3785
3793
22
Levet-Paulo
,
M.
,
Lazzaroni
,
J.C.
,
Gilbert
,
C.
,
Atlan
,
D.
,
Doublet
,
P.
and
Vianney
,
A.
(
2011
)
The atypical two-component sensor kinase Lpl0330 from Legionella pneumophila controls the bifunctional diguanylate cyclase-phosphodiesterase Lpl0329 to modulate bis-(3′-5′)-cyclic dimeric GMP synthesis
.
J. Biol. Chem.
286
,
31136
31144
23
Hull
,
T.D.
,
Ryu
,
M.H.
,
Sullivan
,
M.J.
,
Johnson
,
R.C.
,
Klena
,
N.T.
,
Geiger
,
R.M.
et al. 
(
2012
)
Cyclic di-GMP phosphodiesterases RmdA and RmdB are involved in regulating colony morphology and development in Streptomyces coelicolor
.
J. Bacteriol.
194
,
4642
4651
24
Aravind
,
L.
and
Ponting
,
C.P.
(
1997
)
The GAF domain: an evolutionary link between diverse phototransducing proteins
.
Trends Biochem. Sci.
22
,
458
459
25
Cui
,
T.
,
Zhang
,
L.
,
Wang
,
X.
and
He
,
Z.G.
(
2009
)
Uncovering new signaling proteins and potential drug targets through the interactome analysis of Mycobacterium tuberculosis
.
BMC Genomics
10
,
118
26
Snapper
,
S.B.
,
Melton
,
R.E.
,
Mustafa
,
S.
,
Kieser
,
T.
and
Jacobs
, Jr,
W.R.
(
1990
)
Isolation and characterization of efficient plasmid transformation mutants of Mycobacterium smegmatis
.
Mol. Microbiol.
4
,
1911
1919
27
Bharati
,
B.K.
,
Sharma
,
I.M.
,
Kasetty
,
S.
,
Kumar
,
M.
,
Mukherjee
,
R.
and
Chatterji
,
D.
(
2012
)
A full-length bifunctional protein involved in c-di-GMP turnover is required for long-term survival under nutrient starvation in Mycobacterium smegmatis
.
Microbiology
158
(
Pt 6
),
1415
1427
28
Otwinowski
,
Z.
and
Minor
,
W.
(
1997
)
Processing of X-ray diffraction data collected in oscillation mode
.
Methods Enzymol.
276
,
307
326
29
Vagin
,
A.
and
Teplyakov
,
A.
(
1997
)
MOLREP: an automated program for molecular replacement
.
J. Appl. Crystallogr.
30
,
1022
1025
30
Winn
,
M.D.
,
Ballard
,
C.C.
,
Cowtan
,
K.D.
,
Dodson
,
E.J.
,
Emsley
,
P.
,
Evans
,
P.R.
et al. 
(
2011
)
Overview of the CCP4 suite and current developments
.
Acta Crystallogr. D Biol. Crystallogr.
67
(
Pt 4
),
235
242
31
Murshudov
,
G.N.
,
Skubák
,
P.
,
Lebedev
,
A.A.
,
Pannu
,
N.S.
,
Steiner
,
R.A.
,
Nicholls
,
R.A.
et al. 
(
2011
)
REFMAC5 for the refinement of macromolecular crystal structures
.
Acta Crystallogr. D Biol. Crystallogr.
67
,
355
367
32
Emsley
,
P.
and
Cowtan
,
K.
(
2004
)
Coot: model-building tools for molecular graphics
.
Acta Crystallogr. D Biol. Crystallogr.
60
,
2126
2132
33
Chen
,
V.B.
,
Arendall
,
W.B.
,
Headd
,
J.J.
,
Keedy
,
D.A.
,
Immormino
,
R.M.
,
Kapral
,
G.J.
et al. 
(
2010
)
Molprobity: all-atom structure validation for macromolecular crystallography
.
Acta Crystallogr. D Biol. Crystallogr.
66
,
12
21
34
Krissinel
,
E.
and
Henrick
,
K.
(
2007
)
Inference of macromolecular assemblies from crystalline state
.
J. Mol. Biol.
372
,
774
797
35
Li
,
N.
,
Li
,
X.
,
Wang
,
Y.
,
Liu
,
G.
,
Zhou
,
P.
,
Wu
,
H.
et al. 
(
2016
)
The new NCPSS BL19U2 beamline at the SSRF for small-angle X-ray scattering from biological macromolecules in solution
.
J. Appl. Crystallogr.
49
,
1428
1432
36
Nielsen
,
S.
,
Toft
,
K.N.
,
Snakenborg
,
D.
,
Jeppesen
,
M.G.
,
Jacobsen
,
J.
,
Vestergaard
,
B.
et al. 
(
2009
)
BioXTAS RAW, a software program for high-throughput automated small-angle X-ray scattering data reduction and preliminary analysis
.
J. Appl. Crystallogr.
42
,
959
964
37
Konarev
,
P.V.
,
Volkov
,
V.V.
,
Sokolova
,
A.V.
,
Koch
,
M.H.
and
Svergun
,
D.I.
(
2003
)
PRIMUS: a windows PC-based system for small-angle scattering data analysis
.
J. Appl. Crystallogr.
36
,
1277
1282
38
Petoukhov
,
M.V.
,
Franke
,
D.
,
Shkumatov
,
A.V.
,
Tria
,
G.
,
Kikhney
,
A.G.
,
Gajda
,
M.
et al. 
(
2012
)
New developments in the ATSAS program package for small-angle scattering data analysis
.
J. Appl. Crystallogr.
45
,
342
350
39
Svergun
,
D.
(
1992
)
Determination of the regularization parameter in indirect-transform methods using perceptual criteria
.
J. Appl. Crystallogr.
25
,
495
503
40
Fischer
,
H.
,
Oliveira Neto
,
M.
,
Napolitano
,
H.
,
Polikarpov
,
I.
and
Craievich
,
A.
(
2010
)
Determination of the molecular weight of proteins in solution from a single small-angle X-ray scattering measurement on a relative scale
.
J. Appl. Crystallogr.
43
,
101
109
41
Volkov
,
V.V.
and
Svergun
,
D.I.
(
2003
)
Uniqueness of ab initio shape determination in small-angle scattering
.
J. Appl. Crystallogr.
36
,
860
864
42
Svergun
,
D.
,
Barberato
,
C.
and
Koch
,
M.H.J.
(
1995
)
CRYSOL—A program to evaluate X-ray solution scattering of biological macromolecules from atomic coordinates
.
J. Appl. Crystallogr.
28
,
768
773
43
Kozin
,
M.B.
and
Svergun
,
D.I.
(
2001
)
Automated matching of high- and low-resolution structural models
.
J. Appl. Crystallogr.
34
,
33
41
44
Rao
,
F.
,
Yang
,
Y.
,
Qi
,
Y.
and
Liang
,
Z.X.
(
2008
)
Catalytic mechanism of cyclic di-GMP-specific phosphodiesterase: a study of the EAL domain-containing RocR from Pseudomonas aeruginosa
.
J. Bacteriol.
190
,
3622
3631
45
Schirmer
,
T.
(
2016
)
C-di-GMP synthesis: structural aspects of evolution, catalysis and regulation
.
J. Mol. Biol.
428
,
3683
3701
46
Zoraghi
,
R.
,
Corbin
,
J.D.
and
Francis
,
S.H.
(
2004
)
Properties and functions of GAF domains in cyclic nucleotide phosphodiesterases and other proteins
.
Mol. Pharmacol.
65
,
267
278
47
Bharati
,
B.K.
,
Swetha
,
R.
and
Chatterji
,
D.
(
2013
)
Identification and characterization of starvation induced msdgc-1 promoter involved in the c-di-GMP turnover
.
Gene
528
,
99
108
48
Bittner
,
A.N.
,
Kriel
,
A.
and
Wang
,
J.D.
(
2014
)
Lowering GTP level increases survival of amino acid starvation but slows growth rate for Bacillus subtilis cells lacking (p)ppGpp
.
J. Bacteriol.
196
,
2067
2076
49
Spangler
,
C.
,
Böhm
,
A.
,
Jenal
,
U.
,
Seifert
,
R.
and
Kaever
,
V.
(
2010
)
A liquid chromatography-coupled tandem mass spectrometry method for quantitation of cyclic di-guanosine monophosphate
.
J. Microbiol. Methods
81
,
226
231
50
Armitage
,
J.P.
and
Berry
,
R.M.
(
2010
)
Time for bacteria to slow down
.
Cell
141
,
24
26
51
Tchigvintsev
,
A.
,
Xu
,
X.
,
Singer
,
A.
,
Chang
,
C.
,
Brown
,
G.
,
Proudfoot
,
M.
et al. 
(
2010
)
Structural insight into the mechanism of c-di-GMP hydrolysis by EAL domain phosphodiesterases
.
J. Mol. Biol.
402
,
524
538
52
Phippen
,
C.W.
,
Mikolajek
,
H.
,
Schlaefli
,
H.G.
,
Keevil
,
C.W.
,
Webb
,
J.S.
and
Tews
,
I.
(
2014
)
Formation and dimerization of the phosphodiesterase active site of the Pseudomonas aeruginosa MorA, a bi-functional c-di-GMP regulator
.
FEBS Lett.
588
,
4631
4636
53
Barends
,
T.R.
,
Hartmann
,
E.
,
Griese
,
J.J.
,
Beitlich
,
T.
,
Kirienko
,
N.V.
,
Ryjenkov
,
D.A.
et al. 
(
2009
)
Structure and mechanism of a bacterial light-regulated cyclic nucleotide phosphodiesterase
.
Nature
459
,
1015
1018
54
Holm
,
L.
and
Rosenström
,
P.
(
2010
)
Dali server: conservation mapping in 3D
.
Nucleic Acids Res.
38
(
Suppl 2
),
W545
W549
55
Nagano
,
N.
,
Orengo
,
C.A.
and
Thornton
,
J.M.
(
2002
)
One fold with many functions: the evolutionary relationships between TIM barrel families based on their sequences, structures and functions
.
J. Mol. Biol.
321
,
741
765
56
Sundriyal
,
A.
,
Massa
,
C.
,
Samoray
,
D.
,
Zehender
,
F.
,
Sharpe
,
T.
,
Jenal
,
U.
et al. 
(
2014
)
Inherent regulation of EAL domain-catalyzed hydrolysis of second messenger cyclic di-GMP
.
J. Biol. Chem.
289
,
6978
6990
57
Martinez
,
S.E.
,
Beavo
,
J.A.
and
Hol
,
W.G.
(
2002
)
GAF domains: two-billion-year-old molecular switches that bind cyclic nucleotides
.
Mol. Interv.
2
,
317
58
Heikaus
,
C.C.
,
Stout
,
J.R.
,
Sekharan
,
M.R.
,
Eakin
,
C.M.
,
Rajagopal
,
P.
,
Brzovic
,
P.S.
et al. 
(
2008
)
Solution structure of the cGMP binding GAF domain from phosphodiesterase 5 insight into nucleotide specificity, dimerization, and cGMP-dependent conformational change
.
J. Biol. Chem.
283
,
22749
22759
59
Levdikov
,
V.M.
,
Blagova
,
E.
,
Young
,
V.L.
,
Belitsky
,
B.R.
,
Lebedev
,
A.
,
Sonenshein
,
A.L.
et al. 
(
2017
)
Structure of the branched-chain amino acid and GTP-sensing global regulator, CodY, from Bacillus subtilis
.
J. Biol. Chem.
292
,
2714
2728