Characterization of GefA, a GGEEF Domain-Containing Protein That Modulates Vibrio parahaemolyticus Motility, Biofilm Formation, and Virulence

ABSTRACT

Vibrio parahaemolyticus is a significant foodborne pathogen that causes economic and public health problems worldwide and has a high capacity to adapt to diverse environments and hosts. The second messenger cyclic diguanylate monophosphate (c-di-GMP) allows bacteria to shift from a planktonic form to a communal multicellular lifestyle and plays an important role in bacterial survival and transmission. Here, we characterized single-domain c-di-GMP synthetases in V. parahaemolyticus and identified a novel GGEEF domain-containing protein designated GefA that modulates bacterial swarming motility, biofilm formation, and virulence. GefA inhibits swarming motility by regulating the expression of lateral flagella, while it enhances biofilm formation by controlling exopolysaccharide biosynthesis. Under high-c-di-GMP conditions caused by scrABC knockout, we found that GefA is bifunctional, as it has no effect on swarming motility, but retains the ability to regulate biofilm formation. Subsequent studies suggested that GefA regulates the expression of type III secretion system 1 (T3SS1), which is an important virulence factor in V. parahaemolyticus. Here, we also revealed that the flagella participate in the infection of V. parahaemolyticus. We found that both the T3SS1 and flagella contribute to the GefA-mediated virulence of V. parahaemolyticus in the zebrafish model. Our results expand the knowledge of the V. parahaemolyticus c-di-GMP synthetases and their roles in social behaviors and pathogenicity.

IMPORTANCE The c-di-GMP metabolic enzymes constitute one of the largest clusters of potential orthologues in Vibrio parahaemolyticus. However, the specific roles that these individual c-di-GMP metabolic enzymes play are largely unknown. Here, we identified a GGEEF domain-containing protein designated GefA that regulates bacterial behaviors and virulence. We also demonstrated that flagella participate in the infection of this bacterium, through which GefA regulates bacterial virulence. To our knowledge, the roles that c-di-GMP and flagella play in V. parahaemolyticus virulence have never been revealed. Our findings contribute to a better understanding of the function of c-di-GMP and its synthetases in V. parahaemolyticus.

INTRODUCTION

To survive in complex external environments, bacteria have developed multiple strategies to synchronize their behaviors at the community level, which enable bacterial populations to solve problems that single bacterial cells cannot. The second messenger cyclic diguanylate monophosphate (c-di-GMP) plays a central role in regulating a wide spectrum of processes and phenotypic behaviors, such as cell morphology, antimicrobial tolerance, motility, biofilm formation, and virulence (1, 2). The intracellular concentration of c-di-GMP is controlled by the diguanylate cyclases (DGCs) and phosphodiesterases (PDEs) (2, 3). DGCs contain a GGD(/E)EF amino acid motif and produce c-di-GMP, while PDEs contain the EAL or HD-GYP domains and degrade c-di-GMP (1–3). The c-di-GMP metabolic enzymes constitute one of the largest clusters of potential orthologues in bacteria, and the cellular c-di-GMP homeostasis is dynamic and highly regulated (2, 4).

Vibrio parahaemolyticus is a Gram-negative, foodborne pathogen that thrives in warm climates within marine or estuarine environments (5, 6). It commonly causes acute gastroenteritis, which is associated with the consumption of raw or undercooked seafood (5). In rare cases, Vibrio parahaemolyticus septicemia has been reported (6). To adapt to fluctuating environmental conditions, V. parahaemolyticus must transition from a planktonic form to a communal multicellular lifestyle, which is directed by c-di-GMP and the surface colonization regulatory (Scr) program (7, 8). The Scr module is composed of ScrABC, and mutating scrABC increases cellular c-di-GMP (8). High levels of the second messenger c-di-GMP favor biofilm formation but inhibit swarming motility (8–10). Biofilm formation in V. parahaemolyticus requires exopolysaccharides (EPSs) to form an intercellular matrix (9, 11). The EPS synthesis cluster includes epsA to epsJ and is regulated by the c-di-GMP receptors, including CpsQ and CpsS (10, 12, 13). Swarming motility relies on the production of lateral flagella, which is regulated negatively by an increase in c-di-GMP levels (7, 14, 15). Although many c-di-GMP metabolic enzymes have been identified in V. parahaemolyticus genomes, their specific functions remain unexplored.

Although many virulence factors have been shown to contribute to the pathogenicity of V. parahaemolyticus, the pathogenesis remains incompletely understood, especially the mechanisms whereby the bacteria modulate gene expression according to external signals during infection (16–19). c-di-GMP has been shown to regulate virulence gene expression and thus impact bacterial infection or transmission (20). In the late-stage infection of V. cholerae, the expression of several DGC and PDE genes is induced (21). Moreover, the VieSAB three-component system can regulate c-di-GMP concentration by the PDE VieA and further modulate bacterial virulence (22–24). Therefore, c-di-GMP is an important external molecule to regulate bacterial infection. However, the role that c-di-GMP plays in V. parahaemolyticus virulence remains unclear.

As the GGD(/E)EF and EAL domains often reside in the same protein and many GGD(/E)EF or EAL domain-containing proteins have a regulatory domain, such as GAF and PAS, these hybrid proteins are multifunctional and their activity is frequently modulated by various mechanisms (2, 3, 25). To rapidly screen the c-di-GMP metabolic enzymes that contribute to the maintenance of c-di-GMP levels in V. parahaemolyticus, we carried out systematic research of the DGCs that contain only the GGD(/E)EF domain in the present study. We identified a novel GGEEF domain-containing protein designated GefA, which can synchronize bacterial behaviors and regulate V. parahaemolyticus virulence. These findings contribute to a better understanding of the specific roles that the individual c-di-GMP metabolic enzymes play in the regulation of c-di-GMP homeostasis and the behaviors of V. parahaemolyticus.

RESULTS

Identification of the single-domain DGCs that affect the swarming motility of V. parahaemolyticus.

Forty-four GGD(/E)EF family proteins were predicted in the V. parahaemolyticus genomes (https://www.ncbi.nlm.nih.gov/Complete_Genomes/c-di-GMP.html) (2, 4), and a further conserved domain analysis of these proteins revealed that only 11 proteins are single-domain DGCs (see Table S1 in the supplemental material). VP0699 and VP2076 are encoded on chromosome 1, while VPA0059, VPA0068, VPA0184, VPA0202, VPA0360, VPA0739, VPA0925, VPA1457, and VPA1478 are encoded on chromosome 2 (Fig. 1A). A further amino acid motif analysis confirmed that a conserved GGD(/E)EF domain exists in these DGCs except VPA0059, which contains a AGDEF motif (Fig. 1B and Fig. S1 in the supplemental material).

FIG 1.

Bioinformatics analysis of the single-domain DGCs in V. parahaemolyticus. (A) Distribution of the single-domain DGCs in the genomes of V. parahaemolyticus. The circular diagram depicts the location of 11 single-domain DGCs on the V. parahaemolyticus chromosome 1 and 2. Maps were established using the software CGView Server^BETA (http://cgview.ca/). (B) Protein sequence alignment of the single-domain DGCs. This area is part of Fig. S1. The red dotted line highlights the GGD(/E)EF motif.

To analyze the specific role that these single-domain DGCs play in the regulation of bacterial behaviors, we constructed gene deletion mutants of all these DGCs except for vp0699, vpa0739, and vpa0925. As lower levels of c-di-GMP favor the swarming motility of V. parahaemolyticus, we then compared the abilities of these mutants and the wild-type (WT) strain to swarm over solid agar. Δvpa0202 and Δvpa1478 strains exhibited a significantly increased swarming motility compared with that of the WT strain, while the swarming motility of the other mutant strains was not different compared with that of the WT strain (Fig. 2). These results suggested that both VPA0202 and VPA1478 were functional in V. parahaemolyticus and could regulate bacterial motility.

FIG 2.

Swarming motility of the WT and DGC gene mutant strains. The single swarming colony from the WT, Δvp2076, Δvpa0059, Δvpa0068, Δvpa0184, Δvpa0202, Δvpa0360, Δvpa1457, and Δvpa1478 strains. The migration diameter of the swarming colonies was measured by using ImageJ software. The migration of at least 3 biological replicates of each strain of interest was analyzed. The unpaired two-tailed Student’s t test was used for statistical analysis (**, P < 0.01).

Both VPA0202 and VPA1478 inhibit swarming through the activity of diguanylate cyclase.

To confirm the regulation of the swarming motility by VPA0202 and VPA1478, we overproduced these proteins in the corresponding deletion mutant and WT strains from plasmid pBAD24 containing an arabinose inducible promoter and analyzed the effect on swarming motility. We found that the overexpression of either VPA0202 or VPA1478 had a noticeable effect on the swarming motility. The strains with the pBAD24-VPA0202 or pBAD24-VPA1478 had a significant decrease in swarming motility compared with that of the corresponding strain with the empty plasmid pBAD24 (Fig. 3A and B).

FIG 3.

Both VPA0202 and VPA1478 inhibit swarming motility and rely on the GGEEF domain. (A) The swarming colony and the migration diameter of WT::pBAD24, WT::pBAD24-VPA0202, Δvpa0202::pBAD24, and Δvpa0202::pBAD24-VPA0202 strains. (B) The swarming colony and the migration diameter of WT::pBAD24, WT::pBAD24-VPA1478, Δvpa1478::pBAD24, and Δvpa1478::pBAD24-VPA1478 strains. (C) The swarming colony and the migration diameter of WT::pBAD24-VPA0202, WT::pBAD24-VPA0202^GGAAA, Δvpa0202::pBAD24-VPA0202, and Δvpa0202::pBAD24-VPA0202^GGAAA strains. (D) The swarming colony and the migration diameter of WT::pBAD24-VPA1478, WT::pBAD24-VPA1478^GGAAA, Δvpa1478::pBAD24-VPA1478, and Δvpa1478::pBAD24-VPA1478^GGAAA strains. (E) The swarming colony and the migration diameter of WT::pBAD24, WT::pBAD24-VPA0202, Δvpa1478::pBAD24, and Δvpa1478::pBAD24-VPA0202 strains. (F) The swarming colony and the migration diameter of WT::pBAD24, Δvpa0202::pBAD24, Δvpa0202::pBAD24-VPA1478, and Δvpa0202::pBAD24-VPA1478^GGAAA strains. The migration diameter of the swarming colonies was measured by using ImageJ software. The migration of at least 3 biological replicates of each strain of interest was analyzed. The unpaired two-tailed Student’s t test was used for statistical analysis (*, P < 0.05; **, P < 0.01; ***, P < 0.001).

Bioinformatics analysis showed that a conserved GGEEF domain exists in the C terminus of VPA0202 and VPA1478 (see Fig. S2 in the supplemental material). To evaluate whether the GGEEF motif contributes to the function of VPA0202 and VPA1478, we constructed the pBAD24-VPA0202^GGAAA and pBAD24-VPA1478^GGAAA plasmids in which the GGEEF amino acids were changed to GGAAA. We observed that the strain with the GGAAA variant plasmid had a significant increase in swarming motility compared with that of the strain with the WT plasmid (Fig. 3C and D). In addition, we replenished vpa0202 to the gene native locus of the Δvpa0202 strain and created the CΔvpa0202^WT and CΔvpa0202^GGAAA strains, and the results of the swarming motility of these strains were consistent with that of Fig. 3C (see Fig. S3 in the supplemental material). Therefore, the GGAAA variant motif disturbed the function of VPA0202 and VPA1478. These results suggested that both VPA0202 and VPA1478 carry an active DGC and the GGEEF domain is essential for their mediated swarming motility.

To investigate the relationship between VPA0202 and VPA1478 in modulating cellular c-di-GMP levels, the Δvpa0202 and Δvpa1478 strains were introduced with pBAD24-VPA1478 and pBAD24-VPA0202, respectively, and analyzed by swarming motility assays. We found that the swarming ability was significantly decreased in the Δvpa1478 strain when the VPA0202 protein was overproduced, and the VPA1478 protein exhibited the same function in the Δvpa0202 strain (Fig. 3E and F). These results indicated that VPA0202 and VPA1478 contribute equally to mediating swarming motility in V. parahaemolyticus. We then designated VPA0202 as GGEEF family protein A (GefA) and performed a series of assays to assess its specific roles and the underlying mechanisms in V. parahaemolyticus.

GefA inhibits swarming motility by regulating the expression of lateral flagella.

V. parahaemolyticus has two different flagellar morphologies, including polar flagellum and lateral flagellum (15). To investigate if the polar or the lateral flagellum is affected by GefA to regulate the swarming motility, we first generated the polar flagellum mutant strain ΔflgE and the lateral flagellum mutant strain ΔlfgB. We found that deletion of flgE, which encodes the hook protein of polar flagellum, caused a significant defect in V. parahaemolyticus swarming motility and swimming decreased compared with the WT strain (Fig. 4A and B). In contrast, by knocking out lfgB, which encodes the lateral flagellar basal body protein, the swarming motility significantly decreased in this mutant while the swimming motility was intact (Fig. 4A and B). The motility phenotypes of ΔflgE and ΔlfgB indicate that the polar flagellum is critically employed for swimming motility, yet facilitates swarming motility to some extent, and the lateral flagellum is employed primarily in swarming motility.

FIG 4.

GefA inhibits swarming motility and regulates the expression of lateral flagella. (A) The swarming colony and the migration diameter of WT, ΔgefA, ΔlfgB, ΔlfgBΔgefA, ΔflgE, and ΔflgEΔgefA strains. (B) The swimming colony and the migration diameter of WT, ΔgefA, ΔlfgB, ΔlfgBΔgefA, ΔflgE, and ΔflgEΔgefA strains. The migration diameter of the colonies was measured by using ImageJ software. The migration of at least 3 biological replicates of each strain of interest was analyzed. (C) The expression level of lfgB was assessed by measuring luminescence in P_lfgB-lux transcriptional fusion strains. (D) The expression level of fliM was assessed by measuring luminescence in P_fliM-lux transcriptional fusion strains. Luminescence expression was calculated as the luminescence per unit of OD₆₀₀. All data were presented as means ± SD deviation of triplicate samples from three independent experiments. The unpaired two-tailed Student’s t test was used for statistical analysis (nonsignificant [ns], P > 0.05; **, P < 0.01; ***, P < 0.001).

To assess GefA activity in the ΔflgE or ΔlfgB background, we generated the double gene mutant strains of ΔflgEΔgefA and ΔlfgBΔgefA. We found that GefA could still inhibit swarming motility in the ΔflgE mutant which has the intact lateral flagellum but a dysfunctional polar flagellum; however, GefA could no longer modulate V. parahaemolyticus motility without LfgB (Fig. 4A). These results indicated that GefA regulates swarming motility depending on the lateral flagellum.

To further investigate the mechanism by which GefA modulates the function of the lateral flagellum, we then analyzed the expression of the lateral flagellum in the WT or ΔgefA strains by using a bioluminescence reporter, pBBR-lux, which contained the lateral flagellar gene promoters (P_lfgB and P_fliM). The promoter activities of P_lfgB-lux and P_fliM-lux were both increased significantly in the ΔgefA and CΔgefA^GGAAA strains compared with those in the WT and CΔgefA^WT strains (Fig. 4C and D). In addition, the expression of the polar flagellum was measured by using pBBR-lux containing the polar flagellar gene promoters (P_flaD and P_flaK), and the results showed that the activities of the promoters were both identical in the WT and gefA relevant mutant strains (see Fig. S4 in the supplemental material). Thus, GefA inhibits swarming motility by regulating the expression of lateral flagella in V. parahaemolyticus.

GefA modulates biofilm formation and EPS biosynthesis in V. parahaemolyticus.

Biofilms are complex communities of microorganisms, which contribute to bacterial environmental survival and transmission (26). The EPS is a key chemical component in the biofilm formation of V. parahaemolyticus, and the expression of EPS is mediated by the c-di-GMP level (1, 10, 11). To analyze if the absence of gefA could alter the biofilm formation of V. parahaemolyticus, the WT, ΔgefA, CΔgefA^WT, and CΔgefA^GGAAA strains were cultivated for 36 h and assayed using crystal violet staining. We found that ΔgefA and CΔgefA^GGAAA strains exhibited a significant biofilm decrease compared with the WT and CΔgefA^WT strains (Fig. 5A). This result indicated that GefA could modulate biofilm formation in V. parahaemolyticus.

FIG 5.

GefA modulates biofilm formation and EPS expression. (A) Biofilm formation of WT, ΔgefA, CΔgefA^WT, and CΔgefA^GGAAA strains. Biofilm formation was measured by using crystal violet staining and quantified by measuring OD₅₉₅ at 36 h. (B) Schematic diagram showing the genetic organization of the EPS synthesis cluster in V. parahaemolyticus. The arrows indicate the direction of transcription. (C) The expression level of EPS was assessed by measuring luminescence in P_eps-lux transcriptional fusion strains. (D) The expression level of cpsQ was assessed by measuring luminescence in P_cpsQ-lux transcriptional fusion strains. All data were presented as means ± SD deviation of triplicate samples from three independent experiments. The unpaired two-tailed Student’s t test was used for statistical analysis (*, P < 0.05; **, P < 0.01; ***, P < 0.001).

The EPS synthesis cluster includes mainly epsA to epsJ in V. parahaemolyticus (Fig. 5B) (11–13). We then evaluated if the expression of EPS could be affected by GefA. The promoter region of epsA to epsJ was ligated with bioluminescence reporter pBBR-lux, which was further introduced into WT, ΔgefA, CΔgefA^WT, and CΔgefA^GGAAA strains. The promoter activity of P_eps-lux was significantly decreased in ΔgefA and CΔgefA^GGAAA strains compared with that in the WT and CΔgefA^WT strains (Fig. 5C).

Previous studies have demonstrated that CpsQ can bind the second messenger c-di-GMP and further mediate the bacterial EPS production (10, 14). We wanted to evaluate if the expression level of cpsQ could also be affected by GefA. Therefore, the promoter region of cpsQ was ligated with bioluminescence reporter pBBR-lux and then introduced into the WT, ΔgefA, CΔgefA^WT, and CΔgefA^GGAAA strains. We found that the promoter activity of P_cpsQ-lux was significantly decreased in ΔgefA and CΔgefA^GGAAA strains compared with that in the WT and CΔgefA^WT strains (Fig. 5D), which was consistent with that of P_eps-lux. These results suggested that GefA may modulate biofilm formation by controlling EPS biosynthesis in V. parahaemolyticus.

GefA plays different roles in swarming and biofilm formation in high-c-di-GMP strain.

Previous studies have indicated that the deletion of scrABC increases biofilm formation but inhibits swarming motility, which is due to the increased cellular c-di-GMP levels (8, 27). As GefA contains a GGEEF motif and belongs to a DGC, we then evaluated the relationship between GefA and ScrABC in regulating swarming and biofilm formation in V. parahaemolyticus. Both ΔgefAΔscrABC and ΔscrABC strains almost lost their swarming ability compared with the WT strain (Fig. 6A). By testing the expression of the lateral flagellum in the scrABC deleted strains, we found that the promoter activity of P_lfgB-lux was decreased significantly in both ΔgefAΔscrABC and ΔscrABC strains compared with that in WT and ΔgefA strains (Fig. 6B). These results indicated that GefA does not affect swarming under the high-c-di-GMP conditions caused by scrABC knockout in V. parahaemolyticus.

FIG 6.

GefA plays different roles in swarming and biofilm formation under the scrABC knockout background. (A) The swarming colony and the migration diameter of WT, ΔgefA, ΔscrABC, and ΔgefAΔscrABC strains. The migration diameter of the swarming colonies was measured by using ImageJ software. (B) The expression level of lfgB was assessed by measuring luminescence in P_lfgB-lux transcriptional fusion strains. (C) Biofilm formation of WT, ΔgefA, ΔscrABC, and ΔgefAΔscrABC strains. Biofilm formation was measured by using crystal violet staining and quantified by measuring OD₅₉₅ at 36 h. (D) The expression level of EPS was assessed by measuring luminescence in P_eps-lux transcriptional fusion strains. (E) The expression level of cpsQ was assessed by measuring luminescence in P_cpsQ-lux transcriptional fusion strains. All data were presented as means ± SD deviation of triplicate samples from three independent experiments. The unpaired two-tailed Student’s t test was used for statistical analysis (ns, P > 0.05; *, P < 0.05; **, P < 0.01; ***, P < 0.001).

We then tested whether GefA regulates biofilm formation in an scrABC knockout background, and we found that without ScrABC, GefA could still enhance biofilm formation in V. parahaemolyticus (Fig. 6C). The promoter activities of P_eps-lux and P_cpsQ-lux were both increased significantly in ΔscrABC strains compared with that in the ΔgefAΔscrABC strain (Fig. 6D and E). These results indicated that for biofilm formation in V. parahaemolyticus which is enhanced by c-di-GMP, GefA plays an important role in this biological process regardless of ScrABC function.

GefA regulates the virulence of V. parahaemolyticus via type III secretion system 1 (T3SS1) and flagella.

Previous studies have shown that c-di-GMP can regulate bacterial virulence, but this information remains unclear in V. parahaemolyticus (20, 28, 29). We used a zebrafish model to evaluate if GefA regulates the virulence of V. parahaemolyticus (30). We found that the zebrafish infected by the ΔgefA strain showed an obvious lower survival rate (0%) than zebrafish infected by the WT strain (20%) and the complemented strain CΔgefA^WT (30%) (Fig. 7A and Fig. S5 in the supplemental material). In addition, the zebrafish in the group challenged by phosphate-buffered saline (PBS) or the heated WT strain had a 100% survival rate (Fig. 7A and C). These results suggested that GefA regulates the virulence of V. parahaemolyticus in the zebrafish model.

FIG 7.

GefA regulates the virulence of V. parahaemolyticus via T3SS1 and flagella. (A, C, and D) Survival curves of zebrafish infected with WT, WT-heat, ΔgefA, CΔgefA^WT, Δt3ss1, ΔgefAΔt3ss1, ΔflgEΔlfgB, and ΔgefAΔflgEΔlfgB strains. Groups containing 15 zebrafish each were challenged with the indicated strains at a dose of 2.0 × 10⁷ CFU/fish and monitored until 72 h postinfection. The negative-control group was challenged with an equal volume of sterile PBS. (B) The mRNA levels of T3SS1 and T3SS2 in WT and ΔgefA strains as determined by qRT-PCR. The results of qRT-PCR are expressed as means ± SD from three independent experiments. The qRT-PCR data were analyzed using the unpaired two-tailed Student’s t test, and survival data were analyzed using the log-rank (Mantel-Cox) test. Asterisks indicate significant differences (ns, P > 0.05; *, P < 0.05; **, P < 0.01).

Previous studies indicated that the two type III secretion systems (T3SS1 and T3SS2) play important roles in the pathogenicity of V. parahaemolyticus (18, 31) and the c-di-GMP signaling pathway is involved in the regulation of T3SS expression (29, 32). We then evaluated if GefA affects the expression level of T3SS1 and T3SS2 by reverse transcription- quantitative PCR (qRT-PCR) analysis, and the results showed that the mRNA levels of T3SS1 were significantly higher in the ΔgefA strain than those in WT strain, whereas the mRNA levels of T3SS2 were identical in these strains (Fig. 7B). A further zebrafish survival assay showed that T3SS1 involved the GefA-mediated virulence of V. parahaemolyticus, as the zebrafish infected by the ΔgefAΔt3ss1 strain exhibited a higher survival rate (30%) than the zebrafish infected by the ΔgefA strain (0%) (Fig. 7C).

The flagellum is considered an important virulence factor in different pathogenic bacteria (33). However, this information remains unknown in V. parahaemolyticus. As GefA inhibits swarming motility, we wanted to determine whether flagella regulated by GefA participate in the virulence of V. parahaemolyticus. We found that zebrafish infected by the ΔflgEΔlfgB strain showed a higher survival rate (60%) than the zebrafish infected by the WT strain (20%), and the zebrafish infected by the ΔgefAΔlfgBΔflgE strain exhibited an obvious higher survival rate (50%) than the zebrafish infected by the ΔgefA strain (0%) (Fig. 7D). Therefore, flagella are also involved in the pathogenesis of ΔgefA strain infection. To better understand the effects of T3SS1 and flagella on the GefA-mediated infection of V. parahaemolyticus, we determined the bacterial load in infected zebrafish. We found that the bacterial loads from zebrafish challenged with the ΔgefAΔt3ss1 and ΔgefAΔlfgBΔflgE strains were both significantly reduced compared with those with the ΔgefA strain (see Fig. S6 in the supplemental material). These results indicated that both T3SS1 and flagella are important for in vivo bacterial proliferation and contribute to the GefA-mediated virulence of V. parahaemolyticus.

DISCUSSION

The ability of Vibrio species to initiate the colonization of host surfaces has been attributed to flagellar motility. Many different flagellar morphologies exist on the bacterial surface, such as polar flagellum and lateral flagellum (34). Previous studies showed that bacteria rely on the polar flagellum for swimming in liquid environments but employ the lateral flagellum for swarming on solid surfaces (15, 35). Here, we observed that in V. parahaemolyticus, the polar flagellum is employed critically for swimming motility and yet facilitates swarming motility to some extent, and the lateral flagellum is employed mainly in swarming motility. As flagellar motility is one of the high energy-utilizing cellular processes in bacteria, the genes responsible for the flagellar synthesis are tightly regulated (36). It has been shown that c-di-GMP plays a crucial role in the regulation of swarming motility in V. parahaemolyticus (10, 24). A high concentration of c-di-GMP in V. parahaemolyticus suppresses swarming motility, while a low concentration of c-di-GMP does the opposite. In this study, we found that GefA regulates swarming motility only when the strains contain an intact lateral flagellum regardless of whether the polar flagellum is functional or not. In addition, we observed that the gefA deletion induced lateral flagella formation and cell elongation of V. parahaemolyticus by transmission electron microscopy (see Fig. S7 in the supplemental material). Previous studies indicated that surface recognition, including second signals, affects swarmer cell differentiation, which has an elongated cell body and produces more lateral flagella (37). As c-di-GMP participates in the surface recognition of V. parahaemolyticus, it makes sense that the intracellular concentration of c-di-GMP could modulate the transformation between the swimmer and swarmer cell. In this study, we found that GefA inhibits swarming motility by repressing the expression of lateral flagellar genes, and overexpressing GefA caused a significant swarming defect which depends on the GGEEF motif of GefA. This evidence indicates that GefA acts as a diguanylate cyclase whose product could increase the concentration of the c-di-GMP pool of V. parahaemolyticus.

After colonization of a surface, bacteria begin to form biofilms, which can protect the encased microbial cells from environmental stressors (9). Previous studies have shown that c-di-GMP tends to promote biofilm formation in Vibrio species (3, 9, 15, 20). Here, we observed that the ΔgefA strain exhibited a significant biofilm decrease compared with the WT and CΔgefA^WT strains. The EPSs are the most prevalent component of biofilms. In V. cholerae, the transcription of EPS genes and the regulators vpsR are regulated by c-di-GMP (38). We evaluated the transcription of EPS genes in the ΔgefA strain, and our data suggest that GefA controls EPS biosynthesis in V. parahaemolyticus. In addition, CpsQ is the positive transcriptional regulator of EPS and has c-di-GMP binding ability in V. parahaemolyticus. Previous studies indicated that CpsQ contributes to EPS expression when c-di-GMP levels are high, and thus, it is the primary driver of biofilm formation (10). Here, we found that GefA also regulates CpsQ expression. Therefore, GefA may modulate biofilm formation by controlling EPS biosynthesis in V. parahaemolyticus, and the expression of EPS may be regulated directly by CpsQ and c-di-GMP.

The second messenger c-di-GMP was found to be ubiquitous across the bacterial kingdom (3). It is synthesized by DGCs containing a GGD(/E)EF domain and degraded by PDEs carrying either an EAL or HD-GYP domain (1, 3). In V. parahaemolyticus, there are 28 GGDEF-containing proteins, 13 EAL-containing proteins, 16 proteins with both GGDEF and EAL domains, and 5 HD-GYP-containing proteins (2, 4). Among these enzymes, the Scr proteins direct bacterial surface colonization by controlling the c-di-GMP homeostasis in V. parahaemolyticus. ScrC and ScrG are GGDEF-EAL proteins, whereas ScrJ and ScrL are GGEEF proteins. ScrC acts to synthesize c-di-GMP by itself, whereas it has c-di-GMP-degrading activity in the presence of ScrA and ScrB (8, 27). ScrG feeds into the Scr circuit to impact c-di-GMP and plays a similar role as ScrC (39). ScrJ and ScrL were identified recently as potent DGCs that participate in the Scr surface colonization regulatory network (40). As for the other the c-di-GMP metabolic enzymes in V. parahaemolyticus, their functions are largely unknown. Here, we identified 11 DGCs that contain only the GGD(/E)EF domain and found that 2 novel DGCs of VPA0202 and VPA1478 contribute to the swarming motility of V. parahaemolyticus. As lower levels of c-di-GMP favor the swarming motility, these DGCs may contribute to the modulation of the global c-di-GMP pool. We designated VPA0202 as GefA and further explored the relationship between GefA and Scr in the regulation of bacterial behaviors. Our results showed that GefA plays different roles in swarming and biofilm formation under a scrABC knockout background. As the deletion of the scrABC increases the cellular c-di-GMP levels, it is possible that the activity of GefA is disturbed under high-c-di-GMP conditions. Therefore, the activity of the gefA promoter in the ΔgefA and ΔscrABC strains was evaluated by using the bioluminescence reporter pBBR-lux. However, the results showed that the expression of GefA was not affected by the cellular c-di-GMP (see Fig. S8 in the supplemental material). The underlying mechanism requires further exploration.

Although many virulence factors have been characterized in V. parahaemolyticus, such as thermostable direct hemolysin (TDH), TDH-related hemolysin (TRH), T3SS1, and T3SS2 (41–43), the pathogenesis of this bacterium remains incompletely understood. Previous studies indicated that c-di-GMP can regulate virulence factor expression and thus contributes to bacterial pathogenicity (20, 29). However, it has not been elucidated in V. parahaemolyticus. In this study, we found that both flagella and T3SS1 are involved in the virulence of V. parahaemolyticus in a zebrafish model (Fig. 7). We also found that the deletion of the gefA gene increased the virulence of V. parahaemolyticus in the same model. To our knowledge, the relationship between c-di-GMP and V. parahaemolyticus virulence has never been revealed before. We then explored the mechanisms by which GefA contributes to virulence by evaluating whether GefA regulates T3SS expression. We found that GefA affects the mRNA levels of T3SS1 but not T3SS2, and a further zebrafish infection assay showed that T3SS1 is involved in the GefA-mediated virulence of V. parahaemolyticus. The previous study has indicated that ExsA (VP1699) could positively regulate the transcription of T3SS1 genes in V. parahaemolyticus by interacting directly with the promoter sequences of T3SS1 genes (44). Here, we found that the expression of exsA is regulated negatively by GefA and ExsA is involved in the GefA-mediated virulence of V. parahaemolyticus (see Fig. S9 in the supplemental material). However, by testing the promoter activity of T3SS1 regulated by GefA, we found that the promoter activity of these genes clusters was all identical in ΔgefA and WT strains (see Fig. S10 in the supplemental material). This result suggests that GefA may regulate the mRNA levels of T3SS1 through a posttranscriptional way.

Flagella play diverse roles in bacterial pathogenesis, such as reaching the host/target site, colonization or invasion, maintenance at the infection site, and dispersal to new hosts (33, 45). In the present study, we demonstrated that flagella participate in the virulence of V. parahaemolyticus. In the zebrafish animal model, we observed that the zebrafish infected by the ΔgefAΔlfgBΔflgE strain exhibited an obviously higher survival rate (50%) than zebrafish infected by the ΔgefAΔt3ss1 strain (30%). This finding suggests that that GefA regulation of bacterial virulence relies primarily on the activity of flagella. Bacteria have evolved chemoreceptor systems in conjunction with their flagella, and thus the enhanced flagella enable bacteria to sense the external environment and move in favorable directions (33). After the pathogens arrive at the optimal host site, flagella often continue to play roles in the interactions between the pathogens and their hosts during the establishment and progression of an infection (33). However, the molecular mechanisms whereby the GefA-mediated flagellar formation contributes to the pathogenicity of V. parahaemolyticus remain unclear.

In summary, we carried out systematic research of the DGCs that contain only the GGD(/E)EF domain and identified a novel GGEEF domain-containing protein designated GefA that can synchronize bacterial behaviors and regulate V. parahaemolyticus virulence. These findings will contribute to a better understanding of the function of these DGCs in V. parahaemolyticus. However, the underlying mechanisms whereby GefA functions under high-c-di-GMP conditions or regulates virulence remain unknown and require further exploration.

MATERIALS AND METHODS

Bacterial strains, culture conditions, and plasmids.

The bacterial strains and plasmids used in this work are listed and described in Table 1. V. parahaemolyticus was grown at 37°C in Luria-Bertani (LB) broth with 3% NaCl modified LB (MLB) containing appropriate antibiotics. Escherichia coli strains DH5α and CC118λpir were grown in LB broth with 1% NaCl at 37°C. When required, the culture medium was supplemented with 50 μg/mL streptomycin, 5 μg/mL chloromycetin, or 10% sucrose.

TABLE 1.

Bacterial strains and plasmids used in this study

Strains or plasmids	Description	Source
Bacterial strains
RIMD 2210633	V. parahaemolyticus, WT	Our laboratory
Δvpa0202 (ΔgefA)	Deletion mutant of vpa0202 with WT background	This study
Δvpa0360	Deletion mutant of vpa0360 with WT background	This study
Δvp2076	Deletion mutant of vp2076 with WT background	This study
Δvpa0059	Deletion mutant of vpa0059 with WT background	This study
Δvpa1457	Deletion mutant of vpa1457 with WT background	This study
Δvpa0068	Deletion mutant of vpa0068 with WT background	This study
Δvpa0184	Deletion mutant of vpa0184 with WT background	This study
Δvpa1478	Deletion mutant of vpa1478 with WT background	This study
Δvpa0264 (ΔlfgB)	Deletion mutant of vpa0264 with WT background	This study
Δvp0778 (ΔflgE)	Deletion mutant of vp0778 with WT background	This study
ΔscrABC	Deletion mutant of scrABC with WT background	This study
Δt3ss1	Deletion mutant of vp1696 with WT background	This study
ΔexsA (Δvp1699)	Deletion mutant of vp1699 with WT background	This study
CΔgefAWT (CΔvpa0202WT)	Complement of gefA with ΔgefA background	This study
CΔgefAGGAAA (CΔvpa0202GGAAA)	The GGEEF motif was changed to GGAAA with WT background	This study
ΔflgEΔgefA	Deletion mutant of flgE and gefA with WT background	This study
ΔlfgBΔgefA	Deletion mutant of lfgB and gefA with WT background	This study
ΔgefAΔscrABC	Deletion mutant of gefA and scrABC with WT background	This study
ΔgefAΔt3ss1	Deletion mutant of gefA and vp1696 with WT background	This study
ΔflgEΔlfgB	Deletion mutant of flgE and lfgB with WT background	This study
ΔgefAΔflgEΔlfgB	Deletion mutant of gefA, flgE, and lfgB with WT background	This study
ΔgefAΔexsA	Deletion mutant of gefA and vp1699 with WT background	This study
WT::pBAD24	WT with the plasmid pBAD24	This study
WT::pBAD24-VPA0202	WT with the plasmid pBAD24-VPA0202	This study
WT::pBAD24-VPA1478	WT with the plasmid pBAD24-VPA1478	This study
WT::pBAD24-VPA0202GGAAA	WT with the plasmid pBAD24-VPA0202GGAAA	This study
WT::pBAD24-VPA1478GGAAA	WT with the plasmid pBAD24-VPA1478GGAAA	This study
Δvpa0202::pBAD24	Δvpa0202 with the plasmid pBAD24	This study
Δvpa0202::pBAD24-VPA0202	Δvpa0202 with the plasmid pBAD24-VPA0202	This study
Δvpa0202::pBAD24-VPA0202GGAAA	Δvpa0202 with the plasmid pBAD24-VPA0202GGAAA	This study
Δvpa0202::pBAD24-VPA1478	Δvpa0202 with the plasmid pBAD24-VPA1478	This study
Δvpa0202::pBAD24-VPA1478GGAAA	Δvpa0202 with the plasmid pBAD24-VPA1478GGAAA	This study
Δvpa1478::pBAD24	Δvpa1478 with the plasmid pBAD24	This study
Δvpa1478::pBAD24-VPA1478	Δvpa1478 with the plasmid pBAD24-VPA1478	This study
Δvpa1478::pBAD24-VPA1478GGAAA	Δvpa1478 with the plasmid pBAD24-VPA1478GGAAA	This study
Δvpa1478::pBAD24-VPA0202	Δvpa1478 with the plasmid pBAD24-VPA0202	This study
E. coli DH5α	Cloning host for maintaining the recombinant plasmids	49
E. coli CC118λpir	Mobilization of plasmids into V. parahaemolyticus	50
Plasmids
pDS132	Suicide vector for V. parahaemolyticus mutagenesis	51
pDS132-gefA	Derived from pDS132 used to knock out gefA	This study
pDS132-vpa0360	Derived from pDS132 used to knock out vpa0360	This study
pDS132-vp2076	Derived from pDS132 used to knock out vp2076	This study
pDS132-vpa0059	Derived from pDS132 used to knock out vpa0059	This study
pDS132-vpa1457	Derived from pDS132 used to knock out vpa1457	This study
pDS132-vpa0068	Derived from pDS132 used to knock out vpa0068	This study
pDS132-vpa0184	Derived from pDS132 used to knock out vpa0184	This study
pDS132-vpa1478	Derived from pDS132 used to knock out vpa1478	This study
pDS132-lfgB	Derived from pDS132 used to knock out lfgB	This study
pDS132-flgE	Derived from pDS132 used to knock out flgE	This study
pDS132-scrABC	Derived from pDS132 used to knock out scrABC	This study
pDS132-vp1696	Derived from pDS132 used to knock out vp1696	This study
pDS132-vp1699	Derived from pDS132 used to knock out vp1699	This study
pDS132-gefAWT	Derived from pDS132 for the complement of gefA with ΔgefA background	This study
pDS132-gefAGGAAA	Derived from pDS132 used to make the GGAAA variant	This study
pBAD24	Arabinose-induced expression plasmid	52
pBAD24-VPA0202	Derived from pBAD24 used to overexpress VPA0202 protein	This study
pBAD24-VPA0202GGAAA	Derived from pBAD24-VPA0202 used to overexpress VPA0202GGAAA variant protein	This study
pBAD24-VPA1478	Derived from pBAD24 used to overexpress VPA1478 protein	This study
pBAD24-VPA1478GGAAA	Derived from pBAD24-VPA1478 used to overexpress VPA1478GGAAA variant protein	This study
pBBR-lux	The bioluminescence reporter containing luxCDABE operon of Vibrio harveyi	53
PlfgB-lux	pBBR-lux containing the promoter of lfgB	This study
PfliM-lux	pBBR-lux containing the promoter of fliM	This study
PflaD-lux	pBBR-lux containing the promoter of flaD	This study
PflaK-lux	pBBR-lux containing the promoter of flaK	This study
Peps-lux	pBBR-lux containing the promoter of epsA-J	This study
PcpsQ-lux	pBBR-lux containing the promoter of cpsQ	This study
PgefA-lux	pBBR-lux containing the promoter of gefA	This study
Pvp1667-lux	pBBR-lux containing the promoter of vp1667	This study
Pvp1668-lux	pBBR-lux containing the promoter of vp1668	This study
Pvp1676-lux	pBBR-lux containing the promoter of vp1676	This study
Pvp1682-lux	pBBR-lux containing the promoter of vp1682	This study
PvtrA-lux	pBBR-lux containing the promoter of vtrA	This study
PvtrB-lux	pBBR-lux containing the promoter of vtrB	This study

Bioinformatics analysis.

The complete collection of c-di-GMP metabolic enzymes were sourced from the website https://www.ncbi.nlm.nih.gov/Complete_Genomes/c-di-GMP.html, which shows the numbers of c-di-GMP-related domains in representative complete genomes from all bacterial and archaeal species (2, 4). The protein conserved domain was further analyzed by the Conserved Domain Database (CDD; available from the NCBI website). The BProm program (SoftBerry) was used to predict the promoter region.

Recombinant DNA techniques.

Standard procedures for DNA amplification and DNA assembly were used, and the primers used for PCR amplification are listed in Table 2. All restriction and DNA-modifying enzymes were purchased from APExBIO or New England BioLabs and performed according to the supplier instruction.

TABLE 2.

Primers used in this study

Primer	Primer sequence (5′–3′)
vpa0202 deletion	P1: GCTCTAGAGTAAAATTCACATTGTAAAATTTTTG
	P2: ACGTAGACAAAGCTTCTTTTGGTGTGGCAG
	P3: CCAAAAGAAGCTTTGTCTACGTAGGTATAC
	P4: TGTAGCATGCTTGGCAGTTGAGCGCCTTTTGGC
vpa0360 deletion	P1: GCTCTAGATTCACCATGGGGTTTCCGAA
	P2: TCTGTCGTCATTCTGCCAAACATCGAAGAC
	P3: GATGTTTGGCAGAATGACGACAGATTGATGTT
	P4: CGAGCTCCCTTTGCATCATAACGGATG
vp2076 deletion	P1: GCGTCTAGAGCAGAGCTTGGCGCGCGCGG
	P2: GATAGAGCGTTAGATATTGAAGGGATGAAGCAC
	P3: CCCTTCAATATCTAACGCTCTATCGATACGCGA
	P4: GCGGAGCTCTAGGTGAGCATTTTTTTGAG
vpa0059 deletion	P1: GCTCTAGATAAGAATGGGCTGCAAATTT
	P2: GACGAGAAAACATGCTGGTGAGCTGCGCAA
	P3: CACCAGCATGTTTTCTCGTCGAGCGGATAT
	P4: GCTCTAGACAGGAATATCGCGCCAAACA
vpa1457 deletion	P1: CGAGCTCAAGGCAATACGCTATCCGCC
	P2: CGCCATTTTTGAAAATTTCGACCACTTGTATCA
	P3: GGTCGAAATTTTCAAAAATGGCGATCGCGT
	P4: GCTCTAGAGCAATGGATGGCGAAGACATCT
vpa0068 deletion	P1: GCTCTAGACCAAGACAAAGAGCTCTACA
	P2: GCTCGCAAATGTTGAGTCCTCAGACGTCGA
	P3: TGAGGACTCAACATTTGCGAGCAATAAAGA
	P4: TGTAGCATGCAAATTAGGTGGCTTTATTGT
vpa0184 deletion	P1: GCTCTAGACGTCATTATTTATGGTGAAAC
	P2: ATCTATGCTATTACCCGAGCCGACGAAGCA
	P3: CGGCTCGGGTAATAGCATAGATACAAGAAA
	P4: TGTAGCATGCTGCCTTGATTTACCCAGCAGGT
vpa1478 deletion	P1: AAAACTGCAGTCAAAGACCATGAAGACAGCA
	P2: TAATACCGATGGATTCACGGCTGGTGGACGCAA
	P3: CACCAGCCGTGAATCCATCGGTATTACACTCCA
	P4: AAAACTGCAGAGCGGTTGGTTCGCTTGAACT
lfgB deletion	P1: GCTCTAGATGAACATTAAATTTTATTTTTGTACTCGAT
	P2: AAACGTCAGGCTAACGCGAAAATTGAGTGC
	P3: AATTTTCGCGTTAGCCTGACGTTTCTCAAC
	P4: CGAGCTCGTGTATTTGCCCGGCTTTAAGCCAAGCTCT
flgE deletion	P1: GCTCTAGAGTAGAAAGTGATAAGCCGCTTAGTGCGGAG
	P2: CAGGATATTTTGTTGTAGGGACAAACCGCTTAAAGA
	P3: TCTTTAAGCGGTTTGTCCCTACAACAAAATATCCTGCAG
	P4: CGAGCTCTTGAAGGTCAATCAGCGCGGTCATTTCACC
scrABC deletion	P1: CGAGCTCCCCCCGTTTTACCCATGAAGCCAACTGCGT
	P2: ACAAGCTCGACAACGCCATGTGGATTATCT
	P3: CATGGCGTTGTCGAGCTTGTGCTACACCTG
	P4: GCTCTAGAGATCGCCAAGATGGTTGCCATCGGTCCATG
vp1696 deletion	P1: GAATTCCTGCAGCCCGGGGGATCCGTGTGGTTTCGATGTCGTCTAATTC
	P2: GTTTACAAAGCGCGCCGAACAGTTCGCATGTTC
	P3: CTGTTCGGCGCGCTTTGTAAACGTGCAGTACTG
	P4: CTAAAGGGAACAAAAGCTGGAGCTCGCTGATCCTTTGTTGCTACTTGG
vp1699 deletion	P1: GCTCTAGATACTTGTGACTGCTGCTCGTTCAGTG
	P2: CTCATATTGTTGTCTCTTACTTTACTCAAAGCTATCG
	P3: TAAAGTAAGAGACAACAATATGAGATTGGTCTGAGTG
	P4: GCGGAGCTCTCAAGCTGGGAAAGGAGAAATGAAAGTG
pBAD24-VPA0202	P1: GCTCTAGATTAGATAGGCATCACTCGGTTTCG
	P2: GGGGTACCAAATGACTGACGAGTTTAAGAAAT
pBAD24-VPA0202GGAAA	P1: GCTCTAGATTAGATAGGCATCACTCGGTTTCG
	P2: GCTTCGGCGGCGCGGCGGCGGCGTTAATTGTG
	P3: CGCCGCCGCCGCGCCGCCGAAGCGGTATGC
	P4: GGGGTACCAAATGACTGACGAGTTTAAGAAAT
pBAD24-VPA1478	P1: GGGGTACCAATGGAACTACTGCTTAGCA
	P2: GCTCTAGATTAAAGTTCGGTACAGACTT
pBAD24-VPA1478GGAAA	P1: GGGGTACCAATGGAACTACTGCTTAGCA
	P2: ACAAGTGCTGCTGCACCACCAAATCGCACG
	P3: CGTGCGATTTGGTGGTGCAGCAGCACTTGTAATTCTTG
	P4: GCTCTAGATTAAAGTTCGGTACAGACTT
gefA complement	P1: GCTCTAGAGTAAAATTCACATTGTAAAATTTTTG
	P2: TGTAGCATGCTTGGCAGTTGAGCGCCTTTTGGC
GGAAA variant	P1: GCTCTAGAGTAAAATTCACATTGTAAAATTTTTG
	P2: GCTTCGGCGGCGCGGCGGCGGCGTTAATTGTG
	P3: CGCCGCCGCCGCGCCGCCGAAGCGGTATGC
	P4: GCTCTAGATTGGCAGTTGAGCGCCTTTT
PlfgB-lux	P1: CGAGCTCGTTGGACTTGGCTACGCGGTCTACTTC
	P2: CGGGATCCGCATCTTTCCTTACAGTCGGCTCGAAT
PfliM-lux	P1: CGAGCTCTTAATTACCTATTCTGAAATACTTAAATTG
	P2: CGGGATCCTTATATTTTACCGTGCGTTAACAAACTAAG
PflaD-lux	P1: CGAGCTCTTTGGCCCTCTCGGCTCAATTTGTAGAAAC
	P2: CGGGATCCGGTGATTTCTCCAATTGATTTTCCGATGTA
PflaK-lux	P1: CGGGATCCAAGTAAGAATGATTGCCTTTATTTTAGGTG
	P2: GACTAGTTAATCAAAACAATGCTTTAGGATCGACTTA
PgefA-lux	P1: CGGGATCCTTCTGCGATGTTACTAGCAAAATG
	P2: CGAGCTCGCTAAATGTACTGATTTGTTTTACG
Peps-lux	P1: GCGGAGCTCCTTTTCCTCATCCCTGCTTA
	P2: GCGGGATCCGACCTAGTTTCCCTTCTAGCA
PcpsQ-lux	P1: GCGGAGCTCAGATCACTCTATAGAGATGG
	P2: GCGGGATCCACTTTCATTAACTTAAGATTCTTAATT
Pvp1668-lux	P1: GCGGAGCTCATCCATTCCGGACCGATGCAT
	P2: GCGGGATCCTGTAAAAAATATGCGCAATGA
Pvp1667-lux	P1: GCGGAGCTCGGATCAAACGCGTTTGTTTG
	P2: GCGGGATCCTCACTTGCACTCCTTTCGTA
Pvp1676-lux	P1: GCGGAGCTCCCAGGATGAGAAACTACAAT
	P2: GCGGGATCCTTGTTAAGTATAACTTAACAGTCA
Pvp1682-lux	P1: GCGGAGCTCTCGATAGCCACAAACTTGGT
	P2: GCGGGATCCAATCTTTCCTCTTGTTTCCG
PvtrA-lux	P1: CGGGATCCTTACCGATCTTGTGAGCCTAGACTAATC
	P2: CGAGCTCATAACTCTTTTGATATGTATATTTCCTTTT
PvtrB-lux	P1: CGGGATCCTAGTAAGTAGTTTCTTAACATCGCTGA
	P2: CGAGCTCGCAATTATTTAAGGTCGATGACCAATATGCCGTCGC
vp1686-qRT-PCR	P1: GCCGTGATAACCAATCACACCAG
	P2: CAAATGTAGAACGCGATTACCGTGGG
vp1687-qRT-PCR	P1: GATTGACCAAGACACGGCAATCACTG
	P2: TAAACGGCTGGTGAGCCTTGAGTG
vp1696-qRT-PCR	P1: AGCAAGACTGCCGTTGTATC
	P2: CACACCGAGCTGACGTAAAT
vp1699-qRT-PCR	P1: CACAGCGATCTTCTCATGAC
	P2: GACTGTGCAGCAATAGAATCAGC
vpa1314 qRT-PCR	P1: TTGCCTTTGAGCTTCCATCT
	P2: TTGACCGGTGCATTGGTATTA
vpa1321-qRT-PCR	P1: GTACGCCTCTTGGACAGTTT
	P2: CCATTTACACCTCTTTGACCATTAG
vpa1332-qRT-PCR	P1: CTTGTGTGAGCAACCAGGAA
	P2: TGGGCTCTGATGTTACGAATTG
vpa1336-qRT-PCR	P1: GAGATACGGTAGACGGTGAATTT
	P2: AATGTCTCTGCTCTCGACTTTAC
16S rRNA-qRT-PCR	P1: AAGCGTGGGGAGCAAACAG
	P2: CGAAGGCACCAATCCATCTC

The plasmid pDS132 was used to delete the target gene from the bacterial genomes as described previously (13). Briefly, the upstream and downstream homology arms of the target gene were PCR amplified and assembled into the suicide plasmid pDS132 using specific primers with restriction enzyme sites. The recombinant plasmid was introduced into V. parahaemolyticus or E. coli by conjugation. The plasmid contains a chloromycetin resistance cassette and a sacB counter selectable marker, which could exchange genetic fragments twice with the genomes of V. parahaemolyticus by intermolecular recombination. A putative deletion mutant was identified using PCR and verified by sequencing. The CΔgefA^WT strain was constructed based on the ΔgefA strain by replenishing gefA to the gene native locus, while the replenished gefA sequence has several synonymous mutations after the start codon. The CΔgefA^GGAAA strain was constructed based on the ΔgefA strain by replenishing the gefA^GGAAA sequence to the gene native locus. Additionally, the plasmid pBAD24 was used to overexpress the proteins and corresponding mutants as described previously (46).

The pBBR-lux reporter plasmid was used to identify the activity of the target gene promoter. The promoter-proximal region of each target gene was predicted by the BProm program (SoftBerry) and amplified using specific primers. The promoter sequence was subcloned into the plasmid pBBR-lux, which was further introduced into WT and mutant strains by conjugation.

Swarming motility assays.

Swarming plates were prepared with 2.5% brain heart infusion (BHI) broth (Bacto), 1.2% agar (Bacto), 1.5% NaCl, and 50 μg/mL streptomycin. As for the strains containing the plasmid pBAD24, chloromycetin and arabinose were added to a final concentration of 5 μg/mL and 0.02% in the plates, respectively. The swarm plates were left to dry for 3 h at room temperature, and all bubbles were removed carefully. The single colony of the strain of interest from the MLB plate was picked by a sterile toothpick. Each swarming plate was inoculated with the colony by the toothpick. The inoculated swarming plates were incubated at 37°C for approximately 12 h. The swarming areas were photographed, and the diameter of the digital images were measured by using ImageJ software. Each sample procedure was repeated at least three times.

Swimming motility assays.

Swimming motility plates were prepared with 2.5% BHI, 0.25% agar, 1.5% NaCl, and 50 μg/mL streptomycin. The swim plates were left to dry for 3 h at room temperature, and all bubbles were removed carefully. A single colony of each strain of interest from the MLB plate was inoculated onto the center of the swimming plate using a sterile toothpick. The plates were incubated at 37°C for 4 h, and the migration diameter of the swimming colonies was measured by using ImageJ software. Each sample procedure was repeated at least three times.

Biofilm quantification.

The overnight V. parahaemolyticus cultures were subcultured into fresh MLB and grown to the logarithmic phase with shaking at 37°C for 6 h. The strains were diluted 100 times with MLB, and 2 mL of cultures was added to three replicates of 5-mL borosilicate culture tubes and incubated statically at 37°C for 36 h. Subsequently, the free-floating bacteria and liquid medium were decanted, and the tubes containing biofilm-associated cells were washed carefully with phosphate-buffered saline (PBS). The tubes then were stained with 0.04% crystal violet for 20 min and washed with PBS again. After the tubes were air-dried for 1 h, the crystal violet was solubilized with 2 mL 30% dimethyl sulfoxide (DMSO). The solubilized biofilm optical density at 595 nm (OD₅₉₅) was measured for three samples of each strain using the microplate reader (Biotek Instruments Inc.). Each sample procedure was repeated at least three times.

Bioluminescence reporter assay.

The overnight cultures of the strains harboring a recombinant bioluminescence plasmid (except P_lfgB-lux and P_fliM-lux) were subcultured at a dilution of 1:100 in fresh MLB, which were incubated statically at 37°C for 11 h. As for the strains containing P_lfgB-lux or P_fliM-lux, these strains were subcultured on swarming plates. Briefly, the swarming plate was inoculated with 10 μL of the strain and incubated at 37°C for 8 h, and then the strain was collected from the plates. The value of luminescence and OD₆₀₀ were measured using a Bio-Tek Synergy HT spectrophotometer. The OD₆₀₀ and luminescence of a clean MLB sample were used as a blank. Luminescence expression was calculated as the luminescence per unit of OD₆₀₀. Each sample procedure was repeated at least three times.

Ethics statement and zebrafish infection assay.

The zebrafish experiments were approved by the Institutional Animal Care and Use Committee of Science Technology Department of Zhejiang Province (permit number SYXK-2018-0010) in accordance with the Regulations for the Administration of Affairs Concerning Experimental Animals.

Adult zebrafish (5 months old) were acclimated at room temperature for 1 week before infection treatment. To investigate the survival curves, zebrafish were anesthetized with 1% ethyl 3-aminobenzoate methane sulfonate (Sigma) first and then injected intraperitoneally with 20 μL of bacteria. A total of 15 zebrafish were challenged with the strains of interest at a dose of 2.0 × 10⁷ CFU/fish, while another 15 zebrafish were injected with an equal volume of sterile PBS as a negative control. Survival rates were monitored for 72 h after challenge. For the bacterial burden assay, the zebrafish (8 zebrafish for each group) were challenged with the strains of interest at a dose of 1.0 × 10⁶ CFU/fish and euthanized at 8 h postinfection. The full intestines were harvested and homogenized in PBS. The homogenized samples were serially diluted and plated onto MLB plates with 50 μg/mL streptomycin to enumerate CFU after overnight incubation.

qRT-PCR analysis.

The overnight V. parahaemolyticus strains were subcultured into fresh MLB and grown to the logarithmic phase with shaking at 37°C for 6 h. The strains were then harvested by centrifugation and washed three times with PBS. Total RNA was isolated using TRIzol reagent (Vazyme, China) according to the manufacturer’s instructions, and cDNA was synthesized using the cDNA synthesis kit (Vazyme). qRT-PCR analysis was performed in a 20-μL reaction by using ChamQ yniversal SYBR qPCR master mix (Vazyme). A housekeeping gene, 16S rRNA, was used as the internal reference (47). Relative mRNA levels were quantified using the threshold cycle (2^−ΔΔCT) method (48). All primers used are listed in Table 2, and each sample procedure was repeated at least three times.

Transmission electron microscopy.

The flagella were observed by using transmission electron microscopy. The swarming plate was inoculated with 10 μL of the strain and incubated at 37°C for 8 h, and then the strain was collected from the plates. After being suspended in 50 μL monoethanolamine buffer, the samples were applied to carbon-coated copper grids and stood for 2 min at room temperature. Excess liquid was removed by filter paper, and the bacteria were stained with 10 μL of 0.5% phosphotungstic acid. The dried grids were examined using an H-7650 system (Hitachi) according to the manufacturer’s instructions. This experiment was performed independently three times.

Data analysis.

All data shown are representative of results from at least three independent experiments. The measurements of swarm diameter were prepared using ImageJ software. Statistical analyses were conducted using the unpaired two-tailed Student’s t test or the log-rank (Mantel-Cox) test with the GraphPad software package. For all tests, a P value of <0.05 was considered statistically significant.