Introduction

The type VI secretion system (T6SS) is a widespread, bacteriophage-like complex in Gram-negative bacteria that delivers effector proteins into target cells in a contact-dependent manner (Hespanhol et al. 2023). The system was initially described as a virulence factor of Pseudomonas aeruginosa in cystic fibrosis (CF) patients because Hcp1, secreted by T6SS, and its specific antibodies were detected in these patients (Mougous et al. 2006). It was found that hcp genes are consistently embedded within the T6SS gene cluster and are adjacent to vgrG genes, which encode a crucial structural protein of the T6SS (Mougous et al. 2006; Pukatzki et al. 2006; Ma et al. 2012). In Vibrio cholerae, deletion of hcp blocks the secretion of VgrG, and the deletion of vgrG similarly affects the secretion of Hcp (Pukatzki et al. 2006). Hcp proteins are required for the mature biofilm formation of P. aeruginosa, but they do not reduce biofilm formation (Gallique et al. 2017b). T6SS also functions as a cell-to-cell signal that promotes interbacterial communication depending on cell density (Gallique et al. 2017a). Furthermore, T6SS is utilized as an antibacterial weapon to kill rival bacteria in polymicrobial environments (Coulthurst 2013). T6SS-dependent bacterial killing against rival bacteria is considered a mechanism for pathogens to successfully colonize the host gut (Sana et al. 2016). Additionally, T6SS is also required for metal ion uptake and resistance to stress and host immunity (Wang et al. 2015). In summary, T6SS contributes to multiple bacterial behaviors, including colonization, pathogenesis, interbacterial interactions, biofilm formation, stress resistance, and ion uptake.

V. parahaemolyticus is a leading cause of seafood-associated gastroenteritis and can form stable biofilms on surfaces, which are surface-associated bacterial communities that serves as a strategy for bacteria to resist unfavorable environments (Sharan et al. 2022). The biofilm matrix, which forms a living environment for the cells, is mainly composed of exopolysaccharides (EPS), proteins, extracellular DNA (eDNA), and lipids (Flemming and Wingender 2010). The levels of proteins and eDNA in the biofilm matrix are strongly and positively correlated with biofilm formation in V. parahaemolyticus (Li et al. 2020). EPS and the genes associated with their synthesis also influence colony morphology and biofilm formation of V. parahaemolyticus (Liu et al. 2022; Wu et al. 2022). Furthermore, biofilm formation is strictly regulated by various factors, including quorum sensing and cyclic di-GMP (c-di-GMP) (Lu et al. 2018; Park and Sauer 2022). Biofilms formed by V. parahaemolyticus pose a serious threat to public health and can cause significant economic losses in the food industry (Sharan et al. 2022).

V. parahaemolyticus possesses two distinct T6SS gene loci located on chromosomes 1 and 2, termed T6SS1 (VP1386-1420) and T6SS2 (VPA1024-1046), respectively (Makino et al. 2003). T6SS1 can be detected in clinical isolates and all environmental strains causing acute hepatopancreatic necrosis disease, and it predominantly contributes to cell adhesion and antibacterial activity of V. parahaemolyticus (Yu et al. 2012; Salomon et al. 2013; Li et al. 2017). By contrast, T6SS2 is found in all isolates from both environmental and clinical samples and mainly contributes to cell adhesion activity (Yu et al. 2012; Salomon et al. 2013). However, further research is needed to determine whether these two T6SSs have other physiological roles in V. parahaemolyticus, as studies have shown that T6SS functions are diverse in other bacterial species (Wang et al. 2015; Gallique et al. 2017a, b).

The T6SS2 gene cluster includes 22 genes organized into three putative operons: VPA1027-1025, VPA1043-1028, and VPA1044-1046 (Ma et al. 2012; Zhang et al. 2017). Among these, the icmF2 gene (VPA1039) encodes the intracellular multiplication factor (IcmF), which acts as an inner-membrane protein of T6SS2 and is essential for the structure and function of T6SS2 in V. parahaemolyticus (Yu et al. 2012). Deletion of the icmF2 gene reduces the adhesion activity of V. parahaemolyticus to HeLa monolayers (Yu et al. 2012). In this study, we investigated the role of the icmF2 gene in biofilm formation of V. parahaemolyticus, using a combination of phenotypic and molecular approaches.

Materials and methods

Bacterial strains and growth conditions

V. parahaemolyticus RIMD2210633 was used as the wild type (WT) strain in this work. For construction of icmF2 mutant (ΔicmF2), the entire coding region of icmF2 was deleted from WT using the suicide plasmid pDS132, which was performed as previously described (Zhang et al. 2012). To complement the ΔicmF2 strain (Sun et al. 2014), the coding region of icmF2 was amplified using polymerase chain reaction (PCR) and then cloned into the pBAD33 vector. This vector contains an arabinose-induced PBAD promoter and a chloramphenicol resistance gene. The recombinant plasmid was introduced into the ΔicmF2 strain, yielding the complementation strain C-ΔicmF2. Control strains were created by transforming non-recombinant pBAD33 vector into both the WT and ΔicmF2 strains, respectively.

For the cultivation of V. parahaemolyticus (Li et al. 2021), an overnight bacterial culture in Luria-Bertani (LB) broth containing 10 g/L tryptone (OXOID, UK), 5 g/L yeast extract (OXOID, UK), and 10 g/L NaCl (Solarbio, China; LB-1.0% NaCl) was diluted 50-fold into 5 ml of LB-1.0% NaCl and incubated at 37 °C until it reached to an OD600 value of 1.4, which served as the bacterial seed. The bacterial seed was further diluted 1000-fold into 5 ml of marine (M) broth (2.0% NaCl; BD Biosciences, USA), LB-0.5% NaCl, LB-1.0% NaCl, or LB-3.0% NaCl, in a 24-well cell culture plate (Corning Inc., Corning, USA). The plate was incubated at 30 °C with shaking at 150 rpm, and bacterial cells were harvested at specific time points. The final concentration of chloramphenicol was 5 µg/ml, and that of L-arabinose was 0.1% (w/v).

Crystal violet staining assay

Crystal violet (CV) staining assay was performed as previously described (Zhang et al. 2023d). Briefly, the bacterial seed was diluted 1,000-fold into 2 ml of M broth (or LB broth with different concentrations of NaCl) in a 24-well cell culture plate, and incubated at 30ºC with shaking at 150 rpm for specific durations. The OD600 value, indicative of planktonic cells in the medium, was measured using a spectrophotometer. The surface-attached cells were then washed with deionized water and stained with 0.1% CV. The bound CV was dissolved with 20% acetic acid, after which the OD570 was measured. Relative biofilm formation was expressed as the ratio OD570/OD600.

Colony morphology assay

The colony morphology assay was performed similarly to that previously described (Zhang et al. 2023d). Briefly, the bacterial seed was diluted 50-fold into 5 ml of M broth and then statically incubated at 30ºC for 24 h, followed by thorough mixing. Two microliters of the culture were spotted onto LB plates with varying concentrations of NaCl and then incubated statically at 30 °C for 48 h.

Measurement of growth curves

The bacterial seed was diluted 1, 000-fold into 10 ml of LB-0.5% NaCl, LB-1.0% NaCl, or LB-3.0% NaCl, mixed thoroughly, and then divided into a 96 well cell culture plate with 200 µl in each well. There were 8 replicates for each strain under each condition. Growth curves were created using a microbial growth curve analyzer MGC-200 (Ningbo Scientz Biotechnology Co. Ltd., China) at 37 °C, by detecting OD600 values at a 30 min interval (Li et al. 2024).

Motility assays

Motility assays were performed as previously described (Lu et al. 2019; Zhang et al. 2023c). Briefly, 2 µl of bacterial seed were inoculated into a semi-solid LB plate containing 0.5%, 1.0%, or 3.0% NaCl and 0.5% (w/v) Difco Noble agar (BD Biosciences, USA) for swimming motility. For warming motility, the seed was spotted onto a solid LB plate containing 0.5%, 1.0%, or 3.0% NaCl and 1.5% (w/v) Difco Noble agar. The diameters of swimming and swarming circles were measured after incubation at 37ºC for various durations.

Determination of eDNA and proteins in the biofilm matrix

The relative content of eDNA and proteins in the biofilm matrix were measured similarly to that previously described (Zhu et al. 2023). Briefly, equal amounts of surface-attached biofilms formed by the WT and ΔicmF2 strains were washed twice with pre-cold phosphate-buffered solution (PBS), and then resuspended in 2 ml pre-cold 0.01 M KCl solution, followed by disruption with sonication. After centrifugation, the concentrations of total proteins and eDNA in the supernatant were measured using a Pierce BCA Protein Assay kit (ThermoFisher Scientific, USA) and a M5 Pic5 Green dsDNA Assay Kit (Mei5 Biotechnology, China), respectively.

Measurement of c-di-GMP level

Intracellular c-di-GMP concentrations in V. parahaemolyticus cells were measured similarly to those previously described (Zhang et al. 2023d). Briefly, the bacterial seed was diluted 1, 000-fold into 5 ml of LB-0.5% NaCl and then incubated at 37 °C with shaking at 200 rpm. Bacterial cells were harvested at the mid-log growth phase (OD600 ≈ 1.4), washed twice with ice-cold PBS, and then resuspended in 1 ml of ice-cold PBS, followed by sonication for 5 min. After centrifugation, the intracellular c-di-GMP concentration and total proteins in the supernatant were measured using a c-di-GMP enzyme-linked immunosorbent assay (ELISA) kit (Mskbio, China) and a Pierce BCA Protein Assay kit, respectively. The intracellular c-di-GMP level was calculated as pmol/g of protein.

RNA isolation and quantitative real-time PCR (qPCR)

Bacterial cells of V. parahaemolyticus strains were harvested during the mid-log growth phase, at which the OD600 value was approximately 1.4. Subsequently, total RNA was extracted using TRIzol Reagent (Invitrogen, USA). cDNA was generated from 1 µg of total RNA using a FastKing First Strand cDNA Synthesis Kit (Tiangen Biotech, China). qPCR was performed on a LightCycler 480 (Roche, Switzerland) with a SYBR Green master mixture. Relative gene expression was determined using the 2−ΔΔCt method, with 16 S rRNA serving as the internal control (Gao et al. 2011). All primers used in this study are listed in Table 1.

Table 1 Oligonucleotide primers used in this study
Full size table

Statistical methods

The CV staining assay, colony morphology assay, motility assays and measurement of growth curves were performed at least twice, with the presentation of similar results. Measurement of c-di-GMP level, extracellular proteins, and eDNA, as well as qPCR, were performed at least three times. The results were expressed as the mean ± standard deviation (SD). Student’s t-tests were used to calculate the statistical significance, with a p value less than 0.05 considered significant.

Results

IcmF2 promoted biofilm formation byV. parahaemolyticusin low salt conditions

Previously, our lab routinely used M broth to culture V. parahaemolyticus at 30ºC with shaking at 150 rpm to investigate its biofilm formation ability (Zhang et al. 2021, 2023d, f). In this study, the CV staining assay was employed to determine whether IcmF2 has a regulatory effect on the biofilm formation of V. parahaemolyticus under this growth condition. As shown in Fig. 1a, the ΔicmF2 strain produced similar CV staining results to the WT strain under all the tested time points, suggesting that IcmF2 did not have a regulatory effect on the biofilm formation of V. parahaemolyticus in marine-like growth conditions. V. parahaemolyticus is also capable of forming biofilms in nutrient-rich media such as Heart Infusion broth (Li et al. 2021). Therefore, V. parahaemolyticus strains were grown in nutrient-rich LB broth containing different concentrations of NaCl to quantify biofilms and compare colony morphology variations. As shown in Fig. 1b, the ΔicmF2 strain produced significantly less normalized CV staining compared to the WT strain under conditions of LB-0.5% NaCl and LB-1.0% NaCl (P < 0.05), while no significant differences in CV staining were observed between the ΔicmF2 and WT strains under the growth condition of LB-1.0% NaCl (P > 0.05). As further determined by the colony morphology assays (Fig. 1c), the ΔicmF2 strain exhibited less winkled colonies relative to the WT strain on LB plates containing 0.5% or 1.0% NaCl, while both the ΔicmF2 and WT strains exhibited smooth phenotypes on plates containing 3.0% NaCl. In addition, the biofilm formation capacities of both ΔicmF2 and WT under the growth condition of 0.5% NaCl were much stronger than those under the conditions of 1.0% and 3.0% NaCl (Fig. 1b and c). Moreover, there were no differences in growth rates between the ΔicmF2 and WT strains under the conditions of LB-0.5% NaCl, LB-1.0% NaCl, and LB-3.0% NaCl, suggesting that the deletion of icmF2 did not affect the growth of V. parahaemolyticus (Fig. 1d). In summary, these results suggested that IcmF2 promotes biofilm formation in V. parahaemolyticus under conditions with low salinities, and V. parahaemolyticus also exhibits a stronger biofilm formation capacity under the low salt condition.

Fig. 1
figure 1

Regulation of IcmF2 on biofilm formation by V. parahaemolyticus. The asterisks represent a p value less than 0.05, while ‘ns’ indicates a p value higher than 0.05. The biofilm formation capacities of the ΔicmF2 and WT strains were assessed using crystal violet staining (a and b) and colony morphology (c). The images of the colony morphology assay are representative of at least three independent experiments. Growth curves of ΔicmF2 and WT were created using a microbial growth curve analyzer MGC-200 by detecting OD600 values at a 30 min interval (d)

Full size image

IcmF2 did not regulate the swimming and swarming motility of

The diameters of swimming and swarming circles of the ΔicmF2 and WT strains were measured and compared to assess whether IcmF2 affects the flagella-propelled motilities of V. parahaemolyticus in LB medium with varying concentrations of NaCl. As shown in Fig. 2, there were no significant differences in both swimming and swarming motility between ΔicmF2 and WT under all tested growth conditions and all time points, suggesting that IcmF2 did not affect the swimming and swarming motility of V. parahaemolyticus. However, it appears that the higher the salinity of the medium, the stronger the motor capacity of V. parahaemolyticus, which might be due to the promotional effect of salinity on bacterial growth.

Fig. 2
figure 2

Regulation of IcmF2 on swimming and swarming motility of V. parahaemolyticus. The swimming (a) and swarming (b) motility of ΔicmF2 and WT were evaluated by detecting the diameters of swimming and swarming circles in HI plates containing different concentrations of NaCl, respectively. The results were presented as the mean ± standard deviation (SD). The ‘ns’ means a p value higher than 0.05

Full size image

Incubation time affected IcmF2-dependent biofilm formation

V. parahaemolyticus had a stronger ability to form biofilms at a salinity of 0.5% NaCl than the other tested salinities, and thus the growth condition of 0.5% NaCl was chose to perform the following experiments. CV staining assays were performed to detect whether incubation time affects IcmF2-dependent biofilm formation. As shown in Fig. 3, the highest biofilms produced by ΔicmF2 and WT were observed at an incubation time of 12 h, but a significant difference in biofilm formation capacity between the two strains were only observed at an incubation time of 24 h (P < 0.01). Therefore, incubation time was able to influence the regulation of biofilm formation by IcmF2 in V. parahaemolyticus.

Fig. 3
figure 3

Incubation time affected IcmF2-dependent biofilm formation. The CV staining assay was employed to detect whether incubation time affects IcmF2-dependent biofilm formation under the low salt growth condition. The double asterisks indicate statistically significant difference with a p value less than 0.01, whereas ‘ns’ indicates no statistically significant difference (p > 0.05)

Full size image

IcmF2 affected the contents of biofilm matrix components and c-di-GMP production

The main components of the biofilm matrix, which are essential for biofilm stability, include EPS, eDNA, extracellular proteins, and lipids (Flemming and Wingender 2010). The data presented here showed that the ΔicmF2 strain produced significantly less eDNA and proteins in the biofilm matrix than the WT strain (Fig. 4a and b). In addition, the ΔicmF2 strain also produced much less c-di-GMP than the WT strain (Fig. 4c). These results indicated that IcmF2 was capable of promoting the production of the main components of biofilm matrix and c-di-GMP in V. parahaemolyticus.

Fig. 4
figure 4

Effect of IcmF2 on the production of eDNA, extracellular proteins and intracellular c-di-GMP.The effect of IcmF2 on the main components of biofilm matrix was assessed by detecting the relative contents of eDNA (a) and extracellular proteins (b). Intracellular c-di-GMP levels were measured using a c-di-GMP Enzyme-linked Immunosorbent Assay (ELISA) Kit (c). The double asterisks represent a p value less than 0.01, whereas the single asterisk indicates a p value less than 0.05

Full size image

IcmF2 regulated the transcription of biofilm-associated genes

In this study, cpsA, the first gene of the cpsA-K operon responsible for EPS production (Liu et al. 2022), mfpA, which encodes the membrane fusion protein A contributing to biofilm formation (Enos-Berlage et al. 2005), and known c-di-GMP metabolism enzymic genes including scrJ (Kimbrough and McCarter 2021), scrL (Kimbrough and McCarter 2021), vopY (Wu et al. 2023), tpdA (Martínez-Méndez et al. 2021), gefA (Zhong et al. 2022), scrABC (Boles and McCarter 2002) and scrG (Kim and McCarter 2007), were selected as target genes for qPCR to investigate whether IcmF2 has regulatory effects on their transcription. As shown in Fig. 5, the mRNA levels of cpsA, mfpA, scrJ, scrL, vopY, tpdA, gefA, and scrG were significantly lower in the ΔicmF2/pBAD33 strain compared to the WT/pBAD33 and C-ΔicmF2 strains. In contrast, the mRNA level of scrA was significantly higher in the ΔicmF2/pBAD33 strain. Furthermore, the expression of these genes was restored in the C-ΔicmF2 strain when compared to the WT/pBAD33 strain. These results suggested that IcmF2 inhibits the transcription of scrABC but promotes the transcription of the other tested target genes.

Fig. 5
figure 5

Regulation of biofilm-related genes by IcmF2. Bacterial cells were harvested at an OD600 value of 1.4. The mRNA levels of each target gene were compared between WT/pBAD33, ΔicmF2/pBAD33, and C-ΔicmF2. The double (**) and single (*) asterisks denote a p-value less than 0.01 and 0.05, respectively, whereas the ‘ns’ indicates no statistically significant difference (p > 0.05)

Full size image

Discussion

The function of T6SS requires at least 13 conserved proteins, including membrane-associated proteins such as TssL, TssM, and TssJ, and the components of bacteriophage-like tail such as Hcp, VgrG, and TssB (Silverman et al. 2012). TssL, TssM, and TssJ form channels across the cell envelope that bind to the peptidoglycan layer through TssL (Silverman et al. 2012). Hcp forms a bacteriophage tail-like tubular structure, while VgrG resembles the tailspike-like proteins on the tip of the Hcp tube (Silverman et al. 2012). The Walker A motif of TssM is required for the secretion of Hcp and thus the function of T6SS (Ma et al. 2009; Silverman et al. 2012). In the T6SSs of some bacterial species, TssM is also named as IcmF, which is the homolog of TssM. Deletion of icmF from the genome of an avian pathogenic E. coli strain decreased adherence to and invasion of epithelial cells, intramacrophage survival, and biofilm formation capacity on abiotic surfaces (de Pace et al. 2011). In V. cholerae, deletion of icmF led to reduced motility but increased adherence to human intestinal epithelial cells and conjugation frequency (Das et al. 2002). V. parahaemolyticus possesses two kinds of T6SSs, namely T6SS1 and T6SS2, respectively (Makino et al. 2003). The icmF1 or hcp1 mutant of T6SS1 exhibited significantly reduced adhesion to Caco-2 and HeLa cells, whereas the icmF2 or hcp2 mutant of T6SS2 showed decreased adhesion only to HeLa cells (Yu et al. 2012). In addition, VgrG2 of V. parahaemolyticus T6SS2 was also able to induce autophagy in macrophages (Yu et al. 2015). However, the other roles that T6SS1 and T6SS2 may possess in V. parahaemolyticus remain obscure.

The data presented here demonstrated that IcmF2 promotes biofilm formation by V. parahaemolyticus in low salt conditions but does not in marine-like conditions (Fig. 1). However, the positive regulatory effect of IcmF2 on biofilm formation is limited, as the downregulation of the ΔicmF2 strain results in less than a two-fold reduction in biofilm formation (Fig. 1b and c). Therefore, it is challenging for us to construct a complement strain through plasmid supplementation to verify the regulatory relationships. This is particularly true for the pBAD33 plasmid, which we frequently employed in previous studies (Sun et al. 2014; Chen et al. 2023; Zhang et al. 2023d, e). The use of this plasmid necessitates the addition of L-arabinose and chloramphenicol, both of which significantly inhibit the biofilm formation of V. parahaemolyticus (Zhang et al. 2023a, b). Consequently, these additives would prevent us from observing the regulatory impact of IcmF2 on biofilm formation (data not shown). However, the data of qPCR assays confirmed that the icmF2 mutation is non-polar (Fig. 4). Therefore, the conclusion that IcmF2 promotes biofilm formation is deemed reliable. Previously, T6SS2 has been demonstrated to be most active at cold and warm temperatures in low salt conditions, as the secretion of Hcp2 was only detected in low salt media at 23ºC and 30ºC (Salomon et al. 2013). It is logical that T6SS2 promotes biofilm formation only under low salt growth conditions. T6SS2 likely contributes, at least in part, to the robust biofilm formation capacity of V. parahaemolyticus under low salt growth conditions (Li et al. 2021). V. parahaemolyticus optimally grows at salinities ranging from 2 to 4%, but the human gut is a low salt environment relative to marine ecosystems (Whitaker et al. 2010). In addition, V. cholerae almost immediately forms biofilms after adhering to intestinal cells (Sengupta et al. 2016). T6SS2 also can be expressed at the temperature of 37ºC (Salomon et al. 2013). Further research is needed to determine whether V. parahaemolyticus can form biofilms in the intestine and whether T6SS2 plays a role in this process.

The data also showed that the influence of IcmF2 on biofilm formation was influenced by the incubation time, with significant differences in biofilm formation capacity between ΔicmF2 and WT observed only at 24 h (Fig. 3). In addition, IcmF2 affected the production of eDNA, extracellular proteins and c-di-GMP (Fig. 5). eDNA and extracellular proteins, which are strongly positively correlated with the biofilm formation of V. parahaemolyticus, constitute the main components of the biofilm matrix (Flemming and Wingender 2010; Li et al. 2020). However, the specific role of IcmF2 in the secretion of extracellular DNA and proteins remains unclear. c-di-GMP is a second messenger that widely used by bacteria to control behaviors, including biofilm formation and motility (Homma and Kojima 2022). An elevated level of c-di-GMP in bacterial cells promotes biofilm formation but inhibits motility (Homma and Kojima 2022). Flagella-propelled swimming and swarming motility are required for the mature biofilm formation by V. parahaemolyticus (Yildiz and Visick 2009). However, the disruption of T6SS2 function did not affect the swimming and swarming motility of V. parahaemolyticus (Fig. 2), suggesting that T6SS2-dependent biofilm formation is not mediated by its influence on flagellar genes.

The cpsA-K and scvA-O operons are responsible for EPS production, but it is the cpsA-K operon that affects the colony morphology of V. parahaemolyticus on agar plates (Liu et al. 2022). Furthermore, the mfpABC operon, which encodes membrane fusion proteins, plays a positive role in biofilm formation by V. parahaemolyticus (Enos-Berlage et al. 2005). Deletion of icmF2 resulted in a more than 2-fold downregulation of the transcription levels of cpsA and mfpA (Fig. 5), indicating that IcmF2 promotes biofilm formation partly by positively affecting the transcription of these two gene loci. The synthesis of c-di-GMP is catalyzed by the GGDEF domain of diguanylate cyclase (DGC), whereas the degradation of this molecule is catalyzed by the HD-GYP or EAL domain of phosphodiesterase (PDE) (Homma and Kojima 2022). In V. parahaemolyticus, ScrJ (Kimbrough and McCarter 2021), ScrL (Kimbrough and McCarter 2021), and GefA (Zhong et al. 2022) act as active DGCs, while VopY (Wu et al. 2023), TpdA (Martínez-Méndez et al. 2021), ScrC (encoded by scrABC) (Boles and McCarter 2002), and ScrG (Kim and McCarter 2007) are active PDEs. IcmF2 had positive effects on the transcription of scrJ, scrL, vopY, tpdA, gefA, and scrG (Fig. 5). Additionally, IcmF2 had a slight inhibitory effect on the transcription of scrABC (Fig. 5). Therefore, IcmF2-dependent c-di-GMP production in V. parahaemolyticus may be mediated by the influence of IcmF2 on the expression of these c-di-GMP metabolism-associated genes. However, V. parahaemolyticus harbors more than 60 putative enzymic genes that are likely involved in c-di-GMP metabolism (Römling et al. 2013). Thus, further studies should be conducted to elucidate the mechanisms of IcmF2 on c-di-GMP metabolism in V. parahaemolyticus.

In conclusion, this study reported that IcmF2 positively affects biofilm formation by V. parahaemolyticus under low salt conditions, and this effect is also influenced by the incubation time. IcmF2 was able to promote the production of eDNA, extracellular proteins, and c-di-GMP. In addition, the transcription of cpsA, mfpA, and a group of genes associated with c-di-GMP metabolism is shown to be influenced by IcmF2. Therefore, the promotion of biofilm formation by IcmF2 is likely mediated by its influence on the production of eDNA, extracellular proteins, and c-di-GMP, as well as the transcription of cpsA, mfpA, and a group of genes associated with c-di-GMP metabolism. This work confirms new physiological roles for IcmF2, which promotes biofilm formation and c-di-GMP production in V. parahaemolyticus.