The diguanylate cyclase and phosphodiesterase locally regulate the virulence factors in Vibrio vulnificus
- Open Access
- 01.12.2025
- Research
Erschienen in:
Abstract
Introduction
Cyclic di-nucleotides serve as vital secondary messengers in bacteria, regulating various processes related to environmental information. They play essential roles in signaling pathways by conveying information about the environment that triggers appropriate cellular responses [1]. For instance, cyclic di-nucleotides like cyclic diguanylate (c-di-GMP) and cyclic diadenylate (c-di-AMP) are key regulators of biofilm formation, virulence, and stress responses [2]. These molecules allow bacteria to adapt to environment, thereby ensuring their survival and enabling them to thrive under diverse conditions. Specifically, c-di-GMP is crucial for signal transduction across many bacterial species, controlling the transition between motility and biofilm production [3‐5]. Diguanylate cyclase (DGC) contains a conserved GGDEF domain that contributes to c-di-GMP synthesis, modulating the intracellular accumulation of c-di-GMP, while it can be degraded by diguanylate phosphodiesterase (PDE) [1, 2].
The Vibrio genus is a Gram-negative, halophilic group of bacteria commonly found in warm coastal waters, capable of causing severe gastroenteritis when consuming raw seafood [6]. Pathogenic Vibrio species include V. cholerae, V. parahaemolyticus, and V. vulnificus. The pathogenic characteristics are related to quorum sensing, a process by which bacteria express their virulence factors through signaling molecules [7]. In V. cholerae, mutations in the regulatory genes hns, which encodes the histone-like nucleoid structuring protein (H-NS), and vieA, which encodes the cyclic diguanylate phosphodiesterase (PDE), are associated with a hypervirulent phenotype [8, 9]. The vieA gene is part of an operon that encodes the VieSAB signal transduction system, which regulates nearly 10% of the genome in the classical biotype of V. cholerae but is inactive primarily in the EI Tor biotype [10‐13]. In EI Tor biotype Vibrios, gene expression of vieSAB is negatively regulated by quorum sensing regulator HapR and stationary phase regulator LeuO [14]. The global regulator VieA contains a phospho-receiver domain, a helix-turn-helix motif, and an EAL domain that functions as a phosphodiesterase (PDE) [15]. VieS regulates gene expression by influencing the c-di-GMP pool through its interaction with VieA, while VieB associates with VieS to inhibit the phosphorylation of VieA [16, 17].
Anzeige
Although infections are relatively rare, V. vulnificus is responsible for the highest number of fatalities attributed to Vibrio species [14]. It causes severe gastroenteritis from consuming raw seafood, wound infections, and necrotizing fasciitis. The mortality rates associated with sepsis caused by V. vulnificus can reach as high as 50%, while those linked to wound infections are 17% [14]. Additionally, it presents the most significant economic burden per case among all food-related diseases in the United States [18]. In V. vulnificus, intracellular c-di-GMP regulates biofilm formation, rugose colony formation, and motility [19‐21]. The virulence-associated diguanylate cyclase, VdcR, possesses transcriptional regulatory activity to induce c-di-GMP phosphodiesterase expression, which leads to low accumulated c-di-GMP and suppresses pathogenicity in V. vulnificus [22]. The MARTX toxin in Vibrio groups is regulated by intracellular c-di-GMP, responding to extracellular calcium ions [9]. The cytolysin and hemolysin toxins are regulated by quorum-sensing regulators and cAMP response proteins. The overexpression of VdcR suppressed metalloprotease and hemolytic activity in V. vulnificus, revealing the indirect regulation between c-di-GMP and cytolysin/hemolysin toxins [22]. Nevertheless, the regulation of c-di-GMP-driven virulence in V. vulnificus remains poorly understood. This study discovered a diguanylate cyclase VV2380 that interacts with the globular transcriptional regulator VieA (VVA0648) to modulate virulence in V. vulnificus.
Materials and methods
Bacteria strains and cell lines used in this study, and the incubation conditions
The V. vulnificus YJ016 was obtained from the Bioresource Collection and Research Center (BCRC number 12B0001), which is regularly incubated in Luria-Bertani medium (BD Difco LB broth, Lennox) at 37 °C. The host for gene construction was E. coli DH5α, and the host for protein overexpression was E. coli BL21(DE3) (YB Biotech, ECOS-21). The E. coli and the subsequently constructed strains were incubated in the LB medium with appropriate antibiotics at 37 °C, shaking at 250 rpm. The human cell lines SW480, HCT-116, and HaCaT were obtained from ATCC and were incubated in RPMI or DMEM medium (Gibco) with 10% FBS (Cytiva) and a 1x penicillin/streptomycin/amphotericin B antibiotic solution at 37 °C in a 5% CO2 incubator. All the bacterial strains used in this study are listed in Table S1.
Sequence analysis and protein tertiary structure modeling
The sequence of VV2380 was downloaded from the Uniprot database (https://www.uniprot.org/) with accession Q7MIY3. The secondary structure of VV2380 was predicted by using JPred4 (https://www.compbio.dundee.ac.uk/jpred/) [23]. The tertiary structure of VV2380 was built by AlphaFold Protein Structure Database (https://alphafold.com/) [24, 25], and the PDB file was processed by PyMol (https://pymol.org/2/). To focus on the N-terminal (VV238028 − 141) or the C-terminal (VV2380155 − 327) structure, the tertiary structures were built by SWISS-MODEL using PDB 3PYP or PDB 4H54 as templates, respectively. PyMol processed the predicted tertiary structures to display the specific residues.
Gene construction and protein overexpression
This study used established methods to create variant strains [22]. The 500 bp regions of VV2380 were amplified using primers U500_F_NdeI_VV2380, U500_R_non_VV2380, AN_D500_F, and D500_R_XhoI_VV2380. These PCR products were templates for amplifying U500-D500 with U500_F_NdeI_VV2380 and D500_R_XhoI_VV2380 primers, yielding a 1000 bp product. It was incorporated into the suicide plasmid pKT132 to create the VV2380 gene knockout strain (ΔVV2380) through homologous recombination. Constructs were transformed into E. coli S17 as the DNA donor, while V. vulnificus served as the DNA recipient for conjugation. The donor and recipient bacteria were combined and incubated on a sterilized 0.22 μm filter membrane (Sartorius, 18407-47-N) on LB agar at 37 °C for 24 h. Following antibiotic selection, positively conjugated clones were exposed to a 10% sucrose LB medium to eliminate the vector component. The resulting VV2380 deficient strain (ΔVV2380) was verified through PCR. The coding gene VV2380 was amplified with the primers in Table S2 and incorporated into a pVSV105 vector with a 6-His-tag at the C-terminal. This plasmid was introduced into V. vulnificus WT or strain ΔVV2380 via conjugation, termed the VV2380 overexpression strain (oeVV2380) and the functional complement strain (cVV2380). In strains oeVV2380 and cVV2380, the expression of VV2380 was induced by Isopropyl β-d-1-thiogalactopyranoside (IPTG, BioFroxx).
Anzeige
All primers used in this study are listed in Table S2. The coding gene of VV2380 or VieA was cloned into the pET21a and pET28a plasmids, respectively, using the NdeI and NotI restriction sites or the NdeI and XhoI restriction sites. The resulting constructs, pET21a-vv2380 and pET28a-vieA, were transformed into E. coli BL21 (DE3) (YB Biotech, ECOS-21). The pET21a-vv2380 containing E. coli BL21 (DE3) was cultured in LB medium supplemented with 100 µg/mL of ampicillin (LB/Amp) at 37 °C with shaking at 250 rpm overnight. The pET28a-vieA containing E. coli BL21 (DE3) was cultured in LB medium supplemented with 50 µg/mL of kanamycin (LB/Km) at 37 °C with shaking at 250 rpm overnight. The overnight cultures were diluted 1:100 into fresh LB/Amp or LB/Km medium for a 4-hour incubation until the optical density reached 0.5 to 0.6 at 600 nm. The proteins VV2380 and VieA were induced with 0.25 mM and 1.00 mM of IPTG at 16 °C with shaking at 250 rpm for 16 h, respectively. The bacterial cells were harvested by centrifugation, and the pellet was stored at −80 °C.
Protein purification by Ni-NTA column
The bacteria pellet of E. coli BL21 (DE3) incorporated with pET21a-VV2380 was resuspended in buffer A (10 mM Tris-HCl, pH 8.0, 100 mM NaCl) containing a cOmplete proteinase inhibitor cocktail (Roche, 11697498001). Similarly, the bacteria pellet of E. coli BL21 (DE3) incorporated with pET28a-vieA was resuspended in buffer B (10 mM Tris-HCl, pH 7.5, 100 mM NaCl). The sample mixtures were treated with 0.1 mg/mL lysozyme at room temperature for 30 min. Subsequently, the bacterial cells were lysed using an ultrasonic homogenizer (QSONICA, q125) with 30 min of sonication in 10-second ON/10-second OFF cycles in an ice-water bath. Cell debris was removed through centrifugation at 12,000 rpm and 4 °C for 30 min to obtain the total crude extracts (CE). The crude extracts were incubated with Ni SepharoseTM 6 Fast Flow resin (GE Healthcare, 17–5318-02) at 4 °C for affinity binding with the His-tag fusion protein VV2380 or VieA. Non-binding proteins were discarded after one hour of incubation, while nonspecific binding proteins were washed out using buffer A containing low concentrations of imidazole (20 mM and 40 mM, respectively). The protein VV2380 or VieA was eluted with buffer A containing 50 mM to 150 mM imidazole. Finally, the eluted VV2380 or VieA was concentrated using 10 KDa Amicon Ultra-15 Centrifugal Filter Units (Merck) and then quantified using a bicinchoninic acid (BCA) assay.
VieA phosphodiesterase activity assays
N-methylanthraniloyl (MANT) is a fluorescent moiety commonly used to label nucleotides for biochemical studies. MANT-labeled c-di-GMP was used for phosphodiesterase activity assays. The reaction mixture included 1.0 µM of purified VieA and 0.4 µM MANT-c-di-GMP (Biolog Life Science Institute) in a 10 mM Tris-HCl buffer (pH 7.5) with 0.5 mM DTT, bringing the total volume to 100 µL. The released fluorescence was measured at two-minute intervals over thirty minutes using a fluorescence microplate reader (excitation at 355 nm; emission at 448 nm). Each reaction was assessed in five technical replicates. Buffers containing only protein and buffers with MANT-c-di-GMP were considered negative controls in this assay.
Pull-down assays and co-immunoprecipitation
The reaction mixture of pull-down assays included 5.0 µg of VV2380 in a 10 mM Tris-HCl buffer at pH 8.0, along with 150 mM NaCl and 5.0 mM MgCl2. It was incubated at 37 °C for 30 min, followed by the addition of 250 µM GTP, and continuing the incubation for an additional hour. This procedure yielded the active form of VV2380, whose product, c-di-GMP, was subsequently combined with 5.0 µg of VieA for another hour of incubation. Afterward, the reaction mixtures were incubated with Anti-Flag Magnetic Beads (MCE, HY-K0207) at room temperature for one hour on a Rota-Mixer (Genepure, GDS120). Once incubation was completed, the non-binding supernatant was removed, and the beads were washed with buffer containing 0.1% Nonidet P-40 (BIONOVAS, NP-40) five times, each wash lasting ten minutes. The resulting beads were then combined with 100 µL of buffer and heated at 80 °C for ten minutes to denature the proteins, followed by 12.5% polyacrylamide SDS-PAGE and Western analysis using an anti-flag antibody (Abclonal, AE005) and anti-6x His antibody (Abclonal, AE028).
The reaction mixtures for the co-immunoprecipitation (Co-IP) assay contained 5.0 µg of VV2380 with a His-tag fusion and 2.0 µg of VieA in 10 mM Tris-HCl buffer at pH 7.5, along with 150 mM NaCl and 0.1% NP-40. The mixture was incubated with Ni-NTA beads at 4 °C for 4 h on a Rota-Mixer. The non-binding supernatant was removed, and the beads were washed five times with buffer containing 50 mM imidazole, with each wash lasting five minutes. The beads were then eluted with 250 mM imidazole for subsequent Western analysis using an anti-flag antibody and an anti-6x His antibody.
GTP hydrolysis activity assays
The GTP hydrolysis activities were assessed using the malachite green phosphate assay kit (Sigma-Aldrich, MAK307). When VV2380 hydrolyzed GTP, the malachite green reagent identified the free phosphate released. Reaction mixtures containing 2.5 µM of VV2380, preincubated with 5.0 mM of MgCl2 at 37 °C for 30 min, were combined with 250 µM of GTP in buffer A and incubated at 37 °C for an additional hour. The reaction was diluted five-fold with DI water, and 200 µL of the mixture was terminated by adding 20 µL of working reagent (Sigma-Aldrich, MAK307). The resulting mixtures were incubated at room temperature for 30 min for color development. The absorbance of each sample was measured at 620 nm on a plate reader, and the released phosphate was quantified using the equation from the standard curve. All data points were averaged from triplicate experiments, with the standard deviation shown in the figure. A significant difference between the tested conditions was defined as a p-value less than 0.05 in the T-TEST.
C-di-GMP determination
The overnight cultures of the V. vulnificus variant strains in this study were harvested to prepare their total extracts using an ultrasonic homogenizer (QSONICA, q125). The supernatant was collected as total extracts for further c-di-GMP determination. The reaction product of diguanylate cyclase VV2380 and its derivative mutants, after one hour of incubation, was gathered for c-di-GMP determination using a cyclic di-GMP ELISA kit (Cayman Chemical, Item No. 501780), following standard procedures. The colorized results were measured at 450 nm using a 96-well microplate reader to align with the c-di-GMP quantification standard curve for quantification.
Colony morphology, motility, and biofilm formation assays
The overnight cultures of V. vulnificus variant strains in this study were subcultured into fresh LB medium until the optical density at 600 nm reached 0.5 to 0.6. Then, 2.0 mM of IPTG was added to induce VV2380 expression in oeVV2380 and cVV2380 strains at 30 °C for four hours of incubation. The resulting bacterial cultures were diluted to 0.1 of the optical density at 600 nm. Subsequently, two microliters of the cultures were spotted onto an LB agar plate containing 70 µg/mL of Congo Red. The colonies were incubated at 30 °C for 70 h, and their morphology was observed using a stereo microscope (Zoomkop, SM-63T).
Anzeige
The bacterial motility was assessed by placing the previously mentioned four-hour cultured samples onto a 0.3% agar LB plate for an additional 20 to 24 h of incubation. At least three independent experiments were conducted to calculate the colony area to determine the mean ± SD value.
Environmental calcium ions are known to be a critical factor in regulating biofilm formation in V. vulnificus [4]. This study measured biofilm production from V. vulnificus variant strains under LB medium with 10 mM calcium chloride. Five milliliters of the previously described four-hour cultured samples were placed into 20 mL glass vials and incubated for 24 h at 37 °C with shaking at 150 rpm. After incubation, the bacterial cultures were discarded, and the vials were rinsed three times with PBS buffer to remove unattached bacterial cells. The adhering biofilm was stained for 30 min with 0.23% crystal violet solution (Sigma-Aldrich, HT90132). The staining solution was then removed, and the vials were rinsed three times with PBS buffer before being allowed to air-dry for 30 min. Next, ten milliliters of 95% ethanol were added to each vial to dissolve the stained crystal violet, and its optical density at 545 nm was measured to quantify biofilm production. A minimum of three independent replicates was used to calculate the mean ± SD value. Statistical analysis, using ANOVA and T-TEST, indicated no significant differences among the tested strains in this study.
Protease activity assays
This study assessed the cytolysin activity of V. vulnificus variant strains using azocasein (Sigma-Aldrich, A2765) as a substrate. The azocasein was hydrolyzed by extracellular proteases, releasing azo-dye that could be detected at 450 nm. Six hundred microliters of each overnight culture’s supernatant were combined with 600 µL of a 5 mg/mL azocasein solution, and this mixture was incubated at 37 °C for 22 h. At each time point, 200 µl of the mixture were transferred to a new Eppendorf tube and combined with 350 µL of 5% Trichloroacetic acid solution (Sigma-Aldrich, T0699) to terminate the hydrolysis reaction, followed by incubation at room temperature for five minutes. One hundred microliters of the resulting supernatant were then mixed with 100 µL of 0.5 M NaOH to induce colorization, allowing detection at 450 nm. Each experiment included four technical replicates, and the data were obtained from at least three independent experiments. The reaction mixture without the bacterial culture supernatant served as a negative control. The readings of the WT group at each time point were set to 100% for calculating relative protease activity.
RNA Preparation and the determination of gene expression
To assess the gene expression of target genes in V. vulnificus variants, overnight cultures were transferred to fresh LB medium until the optical density at 600 nm reached 0.5 to 0.6. Next, 2.0 mM of IPTG was added to induce expression in oeVV2380 and cVV2380 strains, followed by four hours of incubation at 37 °C. The bacterial cultures were then harvested via centrifugation and lysed using the NC RNA extraction reagent (EBL, MRE-N3200). The RNA isolation was conducted according to the manufacturer’s instructions for the NC reagent. To eliminate gDNA contamination, RNA samples were treated with gDNA wiper (Vazyme Biotech). cDNA synthesis for each sample was performed with the HiScript III RT Super Mix for quantitative PCR (qPCR), using the 2x Universal SYBR Green Fast qPCR Mix (ABclonal) on a BIO-RAD CFX96 Touch Real-Time PCR Detection System. Bacterial 16 s rRNA gene expression served as the control, with data averaged from at least three biological replicates. All reverse transcription primer pairs can be found in Table S2. The 2^- ΔΔCt value was determined using the ΔCt of the target gene compared to WT, indicating the relative gene expression level.
Anzeige
The cytotoxicity to host cells
To evaluate the cytotoxicity of V. vulnificus variants on host cells, this study followed the established four-hour IPTG induction cultures for this assay. To simulate foodborne infection, we utilized colorectal epithelial cells SW480 and HCT-116, along with HaCaT skin cells, to represent the wound infection caused by V. vulnificus. We seeded 10^5 cells per well of each cell line into a 24-well tissue culture plate (BIOFIL) and incubated for one day. Afterward, the medium was discarded, and each well was washed three times with PBS buffer. Then, 10^6 cells per well of the V. vulnificus variants were introduced and co-incubated at 37 °C for three hours. Following incubation, the surviving cells still adhering to the bottom of each well were stained with 0.23% crystal violet solution for 30 min. After this, the solution was disposed of, and each well was washed three times with PBS. The stained cells were then dissolved in a 1% SDS solution while shaking at 70 rpm at room temperature for 30 min. The viable cells were quantified by measuring the optical density at 540 nm. The condition without bacteria served as a positive control, set at 100%, while the condition treated with RIPA solution served as a negative control, set at 0%, to calculate relative cell viability. Independent triplicate experimental data were used to compute the mean ± SD value.
Results
Phosphodiesterase VVA0648 from V. vulnificus is homologous to VieA from V. cholerae
The phosphodiesterase VieA drives a globular transcriptional regulator to modulate V. cholerae virulence. The phosphodiesterase VVA0648 is homologous to VieA in V. vulnificus, which possesses an N-terminal response regulator domain, an EAL domain, and a C-terminal helix-turn-helix structure (Fig. S1). The purified heterologously expressed VieA (VVA0648) was used in MANT-c-di-GMP degrading assays to validate its phosphodiesterase activity. The results indicated that VieA has phosphodiesterase activity capable of digesting 70% of c-di-GMP within 30 min (Fig. 1a). The response regulator VieA may interact with several diguanylate cyclases in V. vulnificus, according to STRING analysis, and VV2380 is one of them that may interact with VieA (Fig. S2). To investigate the protein-protein interaction between VV2380 and VieA, we fused a His-tag to the C-terminal of VV2380 and a flag-tag-TEV cut site-His-tag to the C-terminal of vieA for the pull-down assay. The fused His-tag on VieA will be used for protein purification using a Ni-NTA column; TEV digestion will then remove it. The pull-down assay revealed that there is no detectable nonspecific binding of VV2380 on anti-flag beads when it is incubated with VV2380 only (Fig. 1b, right). The anti-flag signals of VV2380 only or VieA only showed the non-specific anti-flag signals from antigen on anti-flag beads and VieA samples (Fig. 1b, left). The anti-His signals of VV2380 were pulled down by VieA, eluting from anti-flag beads (Fig. 1b, right). This protein-protein interaction between VV2380 and VieA was confirmed through Co-IP assays, where VV2380 was used as bait to bind with VieA and then captured on Ni-NTA beads (Fig. 1c). No anti-His signal was detected on the loaded VieA, indicating the absence of non-specific binding to the Ni-NTA resin (Fig. 1c, right). Moreover, the anti-Flag signal of VieA is due to its interaction with VV2380 (Fig. 1c, left). This result confirms the protein-protein interaction between VV2380 and VieA.
Fig. 1
C-di-GMP phosphodiesterase VieA activity assay and its protein-protein interaction with VV2380. (a). C-di-GMP phosphodiesterase activity of VieA determined by utilizing MANT-c-di-GMP as substrate. The data indicated a mean ± SD of three or more repeated data points. (b). Pull-down assays of VieA and VV2380. (c). Co-IP assays of VieA and VV2380
In silico analysis of diguanylate cyclase VV2380
Diguanylate cyclase VV2380 has 328 amino acids with a 36920.79 Da theoretical molecular weight. The GGDEF domain (K190-H328) of VV2380 was predicted by ScanProsite (https://prosite.expasy.org/scanprosite/), while the secondary structure of VV2380 was predicted by Jpred4 (Fig. 2a). According to the 3D structural prediction by AlphaFold, the modeling structure of VV2380 showed an α5-loop-α6 structure connected between the N-terminal PAS domain and the C-terminal GGDEF domain (Fig. 2b). Our previous genome-wide phosphoproteomics research of V. vulnificus YJ016 revealed that VV2380 was a reliable phosphorylated protein with specific phosphorylated sites (p-sites) Ser11 and Ser14 located at alpha-helix 1 (α1), and the p-sites Asp207, Thr208, and Tyr209 located at the loop structure between beta-sheet 7 (β7) and α7, while the p-site Asp215 is located at α7 (Fig. 2a). To focus on the C-terminal GGDEF domain, the tertiary structure of residue 155–327 of VV2380 (VV2380155 − 327) was built by SWISS-MODEL with E. coli’s DgcZ (PDB: 4H54) [26] as template. The dimer form of the annotated VV2380155 − 327 structure showed the conserved magnesium ion-binding residues D198 and D241, and the annotated active sites K203, N206, D207, H211, D215, R237, G239, G240, and E242 which participated in substrate GTP binding (Fig. 2c). The p-site D207 is predicted as one of the residues involved in active sites which means the phosphoryl-regulation may participate in the regulation of nucleotide binding (Fig. 2c).
Anzeige
Fig. 2
In silico analysis of diguanylate cyclase VV2380. (a). Schematic of VV2380 secondary structure. The phosphorylated sites D207, T208, Y209, and D215 were marked in red. (b). The tertiary structure of VV2380 was built using AlphaFold. The N-terminal domain and C-terminal domain were linked by the α5-loop-α6 structure. The phosphorylated sites were marked in yellow, and the GTP binding sites were marked in green. (c). SWISS-MODEL built the tertiary structure of the VV2380 C-terminal domain in dimer form with PDB 4H54 as a template to display the active site in detail. The dimer form of VV2380 C-terminal domains is shown in orange and cyan, respectively. The blue spheres are the magnesium ions incorporated in the modeling structure. The detailed active site was zoomed in on the right-hand side to display the GTP analog (GTP’) and its binding residues shown in spheres. The metal ion-binding sites D198 and D241 were shown in yellow sticks
Enzymatic characteristics of VV2380
The coding gene of VV2380 was constructed for protein overexpression in E. coli, and the protein was purified using an Ni-NTA affinity column. The optimal catalytic conditions were 1.0 µM of VV2380 pretreated with 5.0 mM magnesium ions at 37 °C for 30 min in 10 mM Tris pH 8.0 buffer (Fig. S3). Reactions were initialized by adding 250 µM of GTP as substrate and incubated at 37 °C for one hour. The released phosphate was quantified to calculate the specific activity of VV2380. Diguanylate cyclase is considered a metal-ion-dependent enzyme to produce c-di-GMP, and VV2380 has the best GTP hydrolysis activity when it is pretreated with 5.0 mM of Mg2+ and Cu2+ (Fig. S3).
To investigate its effects, the conserved residue within the GGDEF domain, D241, was mutated to Ala through site-directed mutagenesis. Additionally, the phosphorylation sites in D207 and D215 were mutated respectively to Ala to mimic the non-phosphorylated state, thereby eliminating the negative charge effects. Furthermore, the same sites were replaced with Glu to preserve the negative charge effects without the phosphorylated state. The kinetic analysis indicates that the phospho-mimic VV2380(D207E) mutant conserved specific activity with WT VV2380, while the nonphospho-mimic VV2380(D207A) mutant exhibited significantly lower GTP hydrolysis activity than WT, with lower maximal reaction velocity (0.69 ± 0.1190 µM/min) than WT (1.01 ± 0.0516 µM/min) (Fig. 3a; Table 1). This reveals that the negative charge effects on D207 are important for its GTP hydrolysis activity, and that this may be modulated by protein phosphorylation. Interestingly, the other mutants, including VV2380(D241A), VV2380(D215A), and VV2380(D215E), demonstrated significantly higher GTP hydrolysis activity than WT (Fig. 3a). This indicates that the VV2380(D215A), VV2380(D215E), and VV2380(D241A) mutants still maintain high GTP hydrolysis activity and that the negative charge at residue D215 did not affect GTP hydrolysis.
Table 1
Kinetic parameters of VV2380 WT and its mutants in GTP hydrolysis activity
kinetic parameters | WT | D207E | D207A | D215E | D215A | D241A |
|---|---|---|---|---|---|---|
Vmax (µM/min) | 1.01 ± 0.0516 | 1.13 ± 0.0650 | 0.69 ± 0.1190 | 1.35 ± 0.0830 | 1.17 ± 0.0380 | 1.27 ± 0.0410 |
Km (µM) | 198.25 ± 22.70 | 256.53 ± 30.09 | 256.05 ± 90.25 | 21.93 ± 7.46 | 32.20 ± 4.83 | 110.95 ± 9.87 |
Anzeige
To investigate the substrate specificity of VV2380 for hydrolyzing nucleoside triphosphates, ATP, GTP, CTP, UTP, and TTP were used as substrates in hydrolysis activity assays (Fig. 3b). The results revealed that, except for GTP, VV2380 WT can also hydrolyze ATP, CTP, UTP, and TTP. The findings indicate that VV2380 WT exhibits low substrate specificity for hydrolyzing nucleoside triphosphates, favoring ATP over GTP, while showing no preference for UTP (Fig. 3b). Figure 3c illustrates the kinetic analysis of VV2380 WT with various nucleoside triphosphates, and Table 2 summarizes the kinetic parameters.
Table 2
Kinetic parameters of VV2380 WT in nucleoside triphosphate hydrolysis activity
Kinetic parameters | GTP | ATP | CTP | UTP | TTP |
|---|---|---|---|---|---|
Vmax (µM/min) | 1.01 ± 0.0516 | 0.88 ± 0.0194 | 0.64 ± 0.0140 | 0.99 ± 0.0694 | ND |
Km (µM) | 198.25 ± 22.70 | 125.01 ± 7.32 | 68.29 ± 4.96 | 412.00 ± 50.47 | ND |
The c-di-GMP assay kit contained a highly selective c-di-GMP sensor to determine the amount of c-di-GMP, but not c-AMP-GMP. The products from a 1-hour GTP hydrolysis reaction of VV2380 WT and its mutants were collected for c-di-GMP quantification. Results indicated that VV2380 WT can synthesize 1000 ng/mL of c-di-GMP (Fig. 3d). The VV2380(D207E) mutant exhibited much greater c-di-GMP production compared to WT, while VV2380(D207A) produced c-di-GMP at levels similar to WT (Fig. 3d). This supports the hypothesis that VV2380’s c-di-GMP production is controlled by protein phosphorylation at residue D207. We observed that the GTP hydrolysis activity of VV2380(D215A), VV2380(D215E), and VV2380(D241A) mutated VV2380 was noted, but their c-di-GMP production was considerably lower. Notably, when D215 was mutated, it nearly completely lost the capacity to generate c-di-GMP, even though GTP hydrolysis activity remained (Fig. 3d).
Fig. 3
Enzymatic properties of WT and mutated VV2380 in hydrolyzing nucleotide triphosphates. (a). Comparison of GTP hydrolysis activity in WT and mutated VV2380, which corresponds to the Michaelis-Menten plot. (b). Substrate specificity test for WT VV2380 in nucleotide triphosphate hydrolysis. (c). Kinetic assessment of WT VV2380 in the hydrolysis of various nucleotide triphosphates. (d). Measurement of c-di-GMP production by WT and mutated VV2380 after GTP hydrolysis. The data indicated an average of three or more repeated data points. In panel b, we calculated the significant difference using a t-test, with ** signifying p < 0.01. In panel d, we performed a t-test to determine the significant difference between WT and mutated VV2380. The results were marked with * for p < 0.05, ** for p < 0.01, **** for p < 0.001, and n.s. for a non-significant difference
Anzeige
Intracellularly accumulated c-di-GMP regulates V. vulnificus’ pathogenicity
It has been well documented that intracellular c-di-GMP plays a pivotal role in regulating biofilm formation, rugose colony formation, and motility in V. vulnificus [3, 11, 12]. Based on in vitro enzymatic analysis, VV2380 is a diguanylate cyclase responsible for c-di-GMP production (Fig. 3d). The VV2380-derived V. vulnificus variant strains were constructed to investigate the effects of VV2380. The results indicated that when VV2380 is overexpressed in V. vulnificus (oeVV2380), its intracellular c-di-GMP levels reached 6337.7 ng/mL after six hours of 2.0 mM IPTG induction (Fig. 4a). In contrast, the VV2380-knockout strain (ΔVV2380) produced 1376.1 ng/mL of c-di-GMP, while the functionally complemented strain (cVV2380) produced 5529.7 ng/mL of c-di-GMP (Fig. 4a). Significantly, the c-di-GMP levels in both oeVV2380 and cVV2380 returned to 1906.1 ng/mL and 1503.8 ng/mL, aligning with the wild-type levels (1509.4 ng/mL) after 20 h of IPTG induction (Fig. 4a). Short-term induction cultures of each strain were used to study the effects of intracellular c-di-GMP produced by VV2380 in V. vulnificus. The examination of colony morphology showed consistent results, suggesting that elevated c-di-GMP levels in oeVV2380 and cVV2380 led to rugose colony morphology (Fig. 4b). The motility assays showed that an increase in c-di-GMP in oeVV2380 resulted in a significant decrease in motility. Conversely, cVV2380 maintained stable motility levels compared to WT (Fig. 4c). To investigate how VV2380-VieA’s modulation of intracellular c-di-GMP affects biofilm production, the VV2380-derived V. vulnificus variant strains were grown in plastic 96-well plates or glass sample vials with shaking for 24 h. Generally, biofilms attached to glass are found in greater quantities than those on plastic in this assay. The analysis of biofilm production indicated no significant differences when VV2380 was overexpressed, implying that the accumulation of c-di-GMP from VV2380 did not have a noticeable effect on biofilm production (Fig. 4d).
Fig. 4
VV2380-derived V. vulnificus variant strains’ phenotypic analysis. (a). c-di-GMP accumulation, (b). colony morphology, c. motility assay, and d. biofilm production. In panels a and c, we determined significant differences using a multiple t-test, with results indicated by * for p < 0.05, ** for p < 0.01, and **** for p < 0.001. Panel d shows no significant differences among strains. The scale bar in panel b represents 0.22 cm
V. vulnificus produces various secreted virulence factors essential for intragastric infections, such as hemolysin VvhA, metalloprotease VvpE, and MARTX. These bacterial toxins play a crucial role in the virulence of Vibrio species, causing cytotoxic effects on host cells. To evaluate the effects of c-di-GMP modulation from VV2380 on V. vulnificus pathogenesis, we assessed the metalloproteinase activities of the constructions in this study. The azocasein-digesting activity indicated that the metalloprotease activity of strain oeVV2380 was significantly reduced (Fig. 5a). The metalloprotease activity in the cVV2380 strain, which does not decrease as expected shown in Fig. 5a, may be due to its gene expression level (vvpE) not being significantly suppressed, as indicated by a p-value greater than 0.05 from the multiple t-test between WT and cVV2380. This suggests that a higher intracellular accumulation of c-di-GMP results in a decrease in V. vulnificus’ metalloprotease activity. The expression levels of vvpE and rtxA1 in various V. vulnificus strains showed that vvpE and rtxA1 expression were notably reduced in the oeVV2380 and cVV2380 strain after 4 h of IPTG-induced expression (Fig. 5b). To validate the down-regulation of V. vulnificus’ virulence when VV2380 was overexpressed, the human skin spontaneously immortalized cell line HaCaT and colorectal epithelial cell lines HCT-15 and SW480 were employed to confirm the cytotoxicity of V. vulnificus variant strains in mimicking wound infection and food-borne infection, respectively. Results showed that both oeVV2380 and cVV2380 have lower cytotoxicity, resulting in higher cell viability (Fig. 5c). The above data conclude that the intracellular c-di-GMP in V. vulnificus downregulates its pathogenicity.
Fig. 5
Pathogenesis assays and virulence-related gene expression in V. vulnificus variant strains. (a). metalloprotease activity, (b). relative gene expression level of vvpE and rtxA1, c. cytotoxicity assays. Each reaction was performed at least three times, and the data are shown as mean ± SD. We determined significant differences using a multiple t-test, with results indicated by * for p < 0.05, *** for p < 0.005, and **** for p < 0.001, while n.s. indicated a non-significant difference with p > 0.05
Discussion
Diguanylate cyclase VV2380 is an enzyme that depends on metal ions to catalyze the hydrolysis of GTP into c-di-GMP and pyrophosphate. Furthermore, VV2380 can also hydrolyze GTP, ATP, CTP, UTP, and TTP (Fig. 3b and c), indicating a low substrate specificity for VV2380. The sequence analysis of VV2380 revealed only a conserved GGDEF domain, recognized for its role in c-di-GMP production, and showed no additional annotated motifs. This suggests that while VV2380 can hydrolyze other nucleotides, it cannot produce any cyclic dinucleotide other than c-di-GMP. In comparison to c-di-GMP, c-di-AMP is another pivotal cyclic dinucleotide in bacteria, playing a vital role in the regulation of numerous physiological processes, including cell wall homeostasis, potassium uptake, DNA repair, biofilm formation, cell volume, and central metabolism [27‐29]. Cyclic di-AMP is synthesized by diadenylate cyclase (DAC). Several studies indicate that the conserved DAC domain is a structural feature of diadenylate cyclase. This domain generally exhibits a Rossmann-like fold, a prevalent structural motif observed in nucleotide-binding proteins. According to the domain architectures, the diadenylate cyclases can be classified into 22 families [30]. To our knowledge, diadenylate cyclase from V. splendidus is the only annotated DAC identified in the Vibrio genus. There is no known DAC in V. vulnificus, although it may have an enzyme with the same function. This study revealed a diguanylate cyclase VV2380 with low substrate specificity that can hydrolyze GTP and ATP; however, it lacked the conserved DAC domain or the structural characteristics indicating its ability to produce c-di-AMP.
The intracellular accumulation of c-di-GMP in V. vulnificus variant strains in this study exhibited dynamic modulation that depended on the incubation duration. The c-di-GMP level in the VV2380-deficient strain (ΔVV2380) reached 4354.7 ng/mL, which did not revert to the wild-type level (1509.4 ng/mL) after 20 h of incubation (Fig. 4a). This observation led us to hypothesize that locally accumulated c-di-GMP might trigger the activity of diguanylate phosphodiesterase VieA, resulting in the degradation of c-di-GMP after prolonged incubation. The c-di-GMP levels produced in the overnight culture of ΔVV2380 seem inadequate to activate VieA’s function, although they were considerably higher than those in WT, oeVV2380, and cVV2380 (Fig. 4a). Systematic studies using the model bacterium E. coli K-12 have demonstrated that most of its 12 DGCs and 13 PDEs are not only expressed simultaneously but also exhibit enzymatic activity [31‐34]. Furthermore, deleting individual DGC or PDE genes results in distinct phenotypes across various model organisms, despite no changes in intracellular c-di-GMP levels in the mutants or under target activation conditions [33, 35, 36]. These observations raise questions about how a diffusible second messenger like c-di-GMP can selectively regulate different responses through multiple DGCs, PDEs, and their effectors, which coexist and act in parallel. Consequently, it is hypothesized that c-di-GMP signaling occurs not only at a global cellular level but also locally within multiprotein complexes [32, 37]. Recent research has identified and closely examined examples of local c-di-GMP signaling, revealing that cells can dynamically combine both ambient and localized signaling mechanisms [38]. Furthermore, studies indicate that spatial organization contributes to enhancing the versatility and specificity of c-di-GMP signaling, with subcellular clustering noted in specific bacterial species. These results imply that c-di-GMP is not always freely diffusive but can be concentrated in distinct cellular areas to regulate particular functions [39]. In this study, we attempt to show evidence that VV2380 may interact with the c-di-GMP phosphodiesterase VieA, whose activity might be triggered by locally accumulated c-di-GMP produced by VV2380. We proposed that the long-term accumulated c-di-GMP in wild type, oeVV2380, and cVV2380 is high enough to activate VieA’s phosphodiesterase activity, resulting in c-di-GMP levels returning to those of the wild type. However, the locally accumulated c-di-GMP in the ΔVV2380 strain may not be sufficient to trigger VieA’s phosphodiesterase activity, which could explain the results shown in Fig. 4a.
This study investigated the VV2380-mediated pathogenesis in V. vulnificus, showing that the VV2380-associated local c-di-GMP signaling network reduces the activity of cytolysis (VvpE) (Fig. 5a) by inhibiting their gene expression levels (Fig. 5b). This downregulation of cytolysis and MARTX production results in decreased cytotoxicity to human skin cells HaCaT and colon epithelial cells SW480 (Fig. 5c). Since VV2380 lacks a sequence characteristic to function as a transcriptional regulator, it implies that the expression of vvpE and rtxA1 might be controlled by VieA, a potential collaborator in the local c-di-GMP signaling network associated with VV2380. In V. cholerae, the VieSAB operon serves as a signal transduction system, regulating approximately 10% of the genome in the classical biotype [3, 4]. The global regulator VieA consists of a phospho-receiver domain, a helix-turn-helix motif, and an EAL domain that acts as a phosphodiesterase [15]. This study demonstrated a protein-protein interaction between VV2380 and VieA through pull-down assays (Fig. 1), indicating that diguanylate cyclase VV2380 works in conjunction with phosphodiesterase VieA to regulate the c-di-GMP signaling network associated with VV2380. To validate the importance of VieA in this signaling network, this study tried to knock out the vieA gene in the V. vulnificus YJ016 genome. Although multiple attempts were made, the generation of a successful vieA gene knockout strain was not achieved. It precluded further validation of this study using the gene knockout strain.
In conclusion, our findings demonstrated that diguanylate cyclase VV2380 and phosphodiesterase VieA contributed to one of the local c-di-GMP signaling networks to modulate V. vulnificus’ phenotypic characteristics, including motility and colony morphology. It also regulated the cytolysis activity and cytotoxicity to host cells. This study proposes a novel explanation for the presence of various diguanylate cyclases and phosphodiesterases in the V. vulnificus genome, which regulate a range of corresponding responses.
Acknowledgements
The cell lines SW480 and HCT-116 are generously supplied by Dr. Shih-Hsiung Wu from Academia Sinica, Taiwan. Additionally, the HaCaT cell line is kindly provided by Dr. You-Cheng Hseu at China Medical University, Taiwan. We thank the Research Center for Cancer Biology, China Medical University, Taiwan, for supporting the instrument. This work was financially supported by the “Cancer Biology and Precision Therapeutics Center, China Medical University” from the Featured Areas Research Center Program within the Higher Education Sprout Project framework by the Ministry of Education (MOE) in Taiwan.
Declarations
Competing interests
The authors declare no competing interests.
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
