Vibrio parahaemolyticus RIMD2210633 (kindly provided by Professor Jinquan Li, College of Food Science and Technology, Huazhong Agricultural University, China) and derivatives were cultured in Tryptic Soy Broth (TSB) supplemented with 3% NaCl (w/v) or on Chromogenic Vibrio Agar, purchased from the Guangdong Huankai Co., Ltd (Guangzhou, China) at 37°C. Escherichia coli strain SM17-λ-pir (kindly provided by Professor Hongyu Ou, School of Life Sciences and Biotechnology, Shanghai Jiao Tong University, China) was used for plasmid maintenance and conjugation, whereas E. coli strain BL21 (DE3) was used for protein expression. Both E. coli strains were routinely cultured in Lysogeny broth (LB) or on LB agar plates.

Antibiotics were used at the following concentrations: E. coli, ampicillin at 100 μg/mL, chloramphenicol at 34 μg/mL, kanamycin at 50 μg/mL; V. parahaemolyticus, chloramphenicol at 5 μg/mL. The media were supplemented with 0.5 mM IPTG to induce plasmid gene expression.

All primers used for strain and plasmid construction in this study are shown in Supplementary Table 1. The Δvp0610 null mutations and C-Δvp0610 complementary strains were made using long-flanking homology PCR, with 34 μg/mL chloramphenicol added for selection. For IPTG-inducible expression in E. coli BL21 (DE3), the vp0610 coding sequence was amplified using an in-frame C-terminal 6× His tag from V. parahaemolyticus RIMD2210633 genomic DNA. Next, the PCR fragment was inserted into the multiple cloning site (MCS) of pET-B2M, which was derived from pET28a. Selection was conducted using 50 μg/mL kanamycin. The details of all strains and plasmids are listed in Table 1.

A mutant V. parahaemolyticus strain was constructed by deleting the vp0610 gene, as described previously with some minor modifications (Zhang et al., 2012). Briefly, competent SM17 λ-pir cells were transfected with the recombinant plasmid pDS132::vp0610 and conjugated with wild-type (WT) strains. The exconjugants were selected on CHROM agar Vibrio media supplemented with 5 μg/mL chloramphenicol and 100 μg/mL ampicillin. Colonies with characteristics indicating a double homologous recombination event [resistance to 10% (w/v) sucrose, chloramphenicol sensitivity were isolated and target gene mutations were confirmed using PCR with specific primers (Supplementary Table 1)].

pDS132, containing the complete vp0610 sequence with homology arm fragments, was propagated in E. coli SM17-λ-pir and conjugated to the corresponding vp0610 mutant to generate a complementary strain using a similar process to those used for mutant construction. The presence of vp0610 was verified using PCR and sequencing.

Biofilms of the parental strain and isogenic vp0610 mutant were produced as described previously, with minor modifications (O’Toole and Kolter, 1998). Briefly, each well of a 96-well polystyrene microtiter plate was inoculated with 200 μL of each culture diluted to OD₆₀₀ = 0.2 with TSB and then incubated at 28°C without shaking for 24, 48, or 72 h. After planktonic cells had been removed, biofilms attached to the wall were washed with phosphate-buffered saline (PBS, pH 7.4), and stained with 230 μL of 0.1% (w/v) crystal violet (CV) solution shielded from light for 15 min. After floating color had been removed, biofilms were quantified by eluting CV with 230 μL acetic acid (33 %) and measuring absorbance at 590 nm (OD₅₉₀) using a microplate spectrophotometer (EPOCH2, Bio Tek, Winooski, Vermont, United States). Biofilm formation ability was determined by calculating the biofilm formation index (BFI; Eq. 1), as follows:

where AB represents the OD₅₉₀ of CV-stained microorganisms, CW represents the OD₅₉₀ of stained blank wells containing only TSB, GB represents the OD₆₀₀ of the strain, and GW represents the OD₆₀₀ of the blank well.

To assess growth kinetics, the parental, mutant, and complementary strains were inoculated into a 96-well microplate (OD₆₀₀ = 0.05) with a flat bottom and cover. Bacterial growth was monitored by measuring the optical density at 600 nm of each well every 30 min for 24 h at 30°C using a microplate spectrophotometer (EPOCH2) and data were recorded using Gen 5 (EPOCH2, Bio Tek, Winooski, Vermont, United States).

Two types of motility occur depending on the environment: swimming in liquid media relies on the polar flagellum, whereas swarming on the surface of solid media depends upon lateral flagella (McCarter, 2004). We assessed the effect of mutations on swimming and swarming by examining bacterial motility on LB medium containing 0.2 and 1.5% (w/v) agar, respectively. Briefly, plates were spot inoculated with 2 μL of cell cultures (OD₆₀₀ = 0.6) and incubated at 37°C for 8 or 48 h.

Extracellular polysaccharides can be categorized into two main types: water-soluble EPS and water-insoluble EPS, both of which were prepared as described previously with minor modifications (Harimawan and Ting, 2016; Felz et al., 2019). Briefly, 2 mL of biofilm was collected from bacterial cultures grown for 24 h in 100 mL TSB and centrifuged at 5000 × g and 4°C for 20 min. The supernatant was used to measure water-soluble EPS, while the precipitate was re-suspended in 2 mL aqueous solution containing Buffer 1 (0.85% NaCl and 0.22% formaldehyde) at 80°C for 30 min to extract water-insoluble EPS by centrifugation at 4°C and 15000 × g for 30 min. Both soluble and insoluble EPS resuspensions (2 mL) were mixed with phenol (2 mL) and sulfuric acid (10 mL), placed in a boiling water bath for 15 min, and absorbance measured at 490 nm (OD₄₉₀).

Wild-type, Δvp0610, and C-Δvp0610 strains were streaked on Chromogenic Vibrio Agar with 5 μg/mL chloramphenicol and sub-cultured in TSB for 24 h. Total RNA was extracted using RNAiso plus and treated using a PrimeScript^TM RT reagent kit with gDNA Eraser (Takara, Liaoning, China) according to the manufacturer’s instructions. qPCR was performed using a 2-Step RT-qPCR System (LightCycler^®96, Roche, Basel, Switzerland) with TB Green^® Premix Ex Taq^TM (Tli RNaseH Plus). The primer sequences used for RT-qPCR are listed in Supplementary Table 1. Relative gene expression was calculated using the 2^–ΔΔt method, with 16S rDNA as a reference gene.

Biofilms were cultured in 24-well polystyrene microtiter plates for 24 h and then 1 mL bacterial culture (OD₆₀₀ = 0.20) was added to a cell culture coverslip. After cultivation, the biofilms on the coverslips were stained using a LIVE/DEAD^® BacLight Bacterial Viability Kit containing SYTO^® 9 green-fluorescent nucleic acid and propidium iodide red-fluorescent nucleic acid stain for 15 min in the dark and visualized using CSLM (LSM710, Zeiss, Jena, Germany). Biofilm images were processed using Zeiss Zen software (Zeiss).

c-di-GMP determination was performed as previously described with minor modifications (Meng et al., 2019). V. parahaemolyticus strains were grown for 12 and 24 h in TSB (3% NaCl) and centrifuged (2 mL) at 10000 × g for 10 min at 4°C. The cell pellet was washed three times and resuspended in 2 mL ice-cold PBS and sonicated in ice water for 15 min. After centrifugation, the supernatant containing the extracted c-di-GMP was collected and the pellet was resuspended in 2 mL of ice-cold PBS. This extraction step was repeated twice and intracellular c-di-GMP levels were determined using a c-di-GMP ELISA kit, which was purchased from Shanghai Ruifan Biotechnology Co., Ltd (Shanghai, China). Whole-cell protein concentration was determined using a BCA Protein Assay Kit, purchased from Beijing ComWin Biotech Co., Ltd (Beijing, China).

Escherichia coli BL21 (DE3) transformed with a plasmid expressing VP0610 [pVP0610 (6× His)] was grown in LB with kanamycin (50 μg/mL) to an OD₆₀₀ of 0.6. VP0610 expression was induced using 0.5 mM IPTG for 24 h at 20°C. The bacterial cultures were then centrifuged at 5000 × g for 10 min at 4°C, resuspended in binding buffer (20 mM sodium phosphate, 500 mM sodium chloride, 30 mM imidazole, pH 7.4), and treated with a high-pressure cell breaker. This process was repeated until the suspension became translucent and then the bacterial preparations were then filtered, immobilized using Ni Sepharose High Performance (GE Healthcare, Pittsburgh, Commonwealth of Pennsylvania, United States), and washed with 30% elution buffer. WT protein was loaded onto HisTrap Sepharose (GE Healthcare), eluted using different elution buffer concentrations, and analyzed using SDS-PAGE and silver staining. Next Western Blot was used to confirm the enrichment of VP0610 further. Protein bands were identified using mass spectrometry. Preliminary data were processed and converted into MGF format and used to retrieve data from the UniProt and NCBI databases. Data were analyzed using STRING (Damian et al., 2018).

Biological data for RT-qPCR experiments were 2^–ΔΔt transformed before statistically analyzed. All assays were analyzed using one-way ANOVA with Dunnett’s multiple comparisons test. The growth curves, EPS production, RT-qPCR, and c-di-GMP experiments were analyzed using GraphPad Prism 6. Different letters for the biofilm assays, EPS production, c-di-GMP production and RT-qPCR indicate statistically significant differences in the individual strains for the various assays tested. Asterisks indicate P-value < 0.0001 (^****); P-value < 0.001 (^∗∗∗); P-value < 0.01 (^∗∗); P-value < 0.05 (^∗); and non-significant (ns).

Till now, vp0610 is annotated as a hypothetical protein and we couldn’t find any research about VP0610 or its homologous protein in Vibrio spp. We applied Blast to search homologous protein and homology analysis revealed that VP0610 belongs to a YfgM family protein (Figure 1). However, some evidence suggest that YfgM serves as negative regulator of the transcriptional regulatory protein (RcsB)-dependent stress response pathway in E. coli (Westphal et al., 2012). RcsB is a transcriptional regulatory protein, controlling the expression of lux operon, thus affecting quorum sensing and biofilm formation (Gambino and Cappitelli, 2016), and the speculation is consistent with the comparative proteomic analysis (data not published). Furthermore, vp0610 is probably related to biofilm formation.

To elucidate the role of vp0610 in V. parahaemolyticus biofilm formation, we monitored this process at 24, 48, and 72 h using CV staining. vp0610 deletion significantly increased biofilm formation at 24 and 48 h, but increased biofilm formation by three-fold compared to the WT at 24 h. Interestingly, biofilm formation was faster in the WT than in the mutant after 24 h, meaning that this difference gradually decreased and disappeared completely by 72 h (Figure 2A). Moreover, quantitative analysis revealed that vp0610 exerted a clear inhibitory effect during the early stages of biofilm formation. To exclude the bacterial growth influence, we monitored OD₆₀₀ for 24 h and found that there was no significant difference between the WT and the mutant (Figure 2A). The results showed that vp0610 delayed the biofilms formation at early stage, and then the difference gradually disappeared toward the end.

Next, direct microscopic visualization with a CLSM and SEM was further used to confirm the effect of vp0610 on biofilm formation. The CLSM images showed that the mutant formed much thicker and more compact biofilms than the WT (Figure 2B), while the SEM images suggested that vp0610 deletion dramatically increased bacterial surface density, consistent with the CV staining results (Figure 2C). Together, these findings suggest that Δvp0610 affects biofilm formation by altering cell adhesion and aggregation.

A previous study discovered that flagellar motility benefits initial colonization, adhesion, thereby promoting biofilm formation in both pathogenic and non-pathogenic Vibrio species (Khan et al., 2020). Therefore, we investigated whether vp0610 affected V. parahaemolyticus biofilm formation by regulating its motility in semi-solid and on surfaces using swimming and swarming experiments, respectively. The WT strains grown in LB agar (0.2%) covered a much smaller diameter than the Δvp0610 strains (Figure 3A), indicating that vp0610 negatively affects swimming; however, all strains displayed similar swarming abilities (Figure 3B), suggesting that vp0610 does not affect swarming. Polar flagella mediate swimming motility. Our results suggested that vp0610 had an inhibitory effect on the motility of polar flagella. Deleting vp0610 resulted in the relieving of inhibition, causing the polar flagella to move quickly, and increasing the chances of cell contact with surfaces.

Extracellular polysaccharides can help the cells in a biofilm to adhere to the surface of the medium and form spatial structures, thereby protecting the cells in the biofilm (Limoli et al., 2015). To determine whether vp0610 was required for EPS synthesis in V. parahaemolyticus, we measured the production of both water-soluble and -insoluble EPS in the WT and mutant strains. Interestingly, the mutant strain produced significantly less water-soluble EPS than the WT strain but produced considerably more water-insoluble EPS (Figures 3C,D). Thus, vp0610 appears to positively regulate the production of soluble EPS and negatively regulate insoluble EPS production. The increase in water-insoluble EPS of the mutant enhances its adhesion and further facilitates biofilm formation, while the decrease of water-soluble EPS may result from the high density of bacteria in biofilm, which is consistent with the result of the biofilm measurement.

c-di-GMP can regulate the reversible transition of bacteria from a planktonic state into a biofilm state, with high c-di-GMP concentrations promoting biofilm formation and low concentrations promoting biofilm dispersion (Valentini and Filloux, 2016). Therefore, we examined whether vp0610 affected c-di-GMP production by measuring the c-di-GMP levels at 12 and 24 h. Although the loss of vp0610 did not alter c-di-GMP production at 12 h, a significant decrease was observed at 24 h (Figures 4A,B). Meanwhile, an obvious increase of in c-di-GMP production from 12 h to 24 h of WT was observed, but there was no change in the mutant strain. So, vp0610 makes no difference at the early stage, but causes a significant decrease later, partly explaining why the biofilm of the mutant strain appeared to be slowly formed after 24 h.

Having demonstrated that VP0610 participates in biofilm formation in V. parahaemolyticus, we aimed to identify new interaction partners of VP0610 by performing pull-down assays using hexahistidine-tagged VP0610 (His-VP0610) (Figure 5). Repeated experiments produced many protein bands that were mapped following in-gel tryptic digestion, revealing that 180 proteins interacted with His-VP0610 (Supplementary Table 2). Gene Ontology (GO) enrichment analysis indicated that these proteins could be divided into three groups: biological process, cellular component, and molecular function (Figure 6), while Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis revealed that the majority of the proteins participated in five main significant enrichment pathways, including RNA degradation, metabolic pathways, riboflavin metabolism, secondary metabolite biosynthesis, antibiotic biosynthesis, and lysine biosynthesis (Supplementary Table 3). Specifically, RNA binding protein (Hfq), adenylate cyclase (CyaA), and PTS-related protein components (VP0710 and VP0793) were identified as key proteins involved in biofilm formation via the quorum sensing and cAMP/CRP pathways (Supplementary Figure 1). Therefore, we will focus on cAMP/CRP pathways and quorum sensing to explore the effect of vp0610 on biofilm formation.

Based on the pull-down assay results, we subjected representative genes from the cAMP/CRP and quorum sensing pathways to RT-qPCR assays to determine how vp0610 affected these genes. Importantly, the mutant strain displayed significantly higher levels of vp0793, vp0710, opaR, flaE, and tdh transcripts than the WT strain; however, hfq, cyaA and aphA expression were not significantly higher and cdgC and mshA expression were significantly lower (Figure 7). Our results showed vp0610 could not only act on the cAMP/CRP pathways to affect biofilm formation, but also regulated flagellum-related genes through unknown pathways.