The I. galbana strain (NMBjih021-2) used in this study was obtained from the Marine Biotechnology Laboratory of Ningbo University, China. The microalgae were cultured in NMB3 medium (Peng et al., 2020), which contained KNO₃ (100 mg/L), KH₂PO₄ (10 mg/L), MnSO₄·H₂O (2.5 mg/L), FeSO₄·7H₂O (2.5 mg/L), EDTA-Na₂ (10 mg/L), vitamin B₁ (6 mg/L), and vitamin B₁₂ (0.05 mg/L). The microalgae were cultivated at 25°C with a 12 h/12 h light/dark cycle in a light-controlled incubator (GXZ-280B, China), and the light intensity was 100 μmol photon m⁻² s⁻¹. Axenic I. galbana was maintained as described by Cao et al. (2019).

Seven bacterial strains, i.e., Alteromonas sp., Oceanicaulis sp., Dinoroseobacter sp., Pseudoalteromonas flavipulchra, Alteromonas macleodii, Marinobacter sp., and Bacillus jeotgali, were isolated from the cultures of I. galbana in the exponential growth phase ( Supplementary Table S1 , Supplementary Figure S1 ). Vibrio vulnificus, V. parahaemolyticus and V. cholerae were isolated from the water samples collected at the Dalai Experimental Base in the Luoyuan County of Fujian, China ( Supplementary Table S1 , Supplementary Figure S2 ). All bacteria were freshly plated on 2216E agar plates in a biochemical incubator (SPX-50, China) and grown in 2216E medium at 28°C with shaking at 220 rpm in a thermostatic shaking incubator (THZ-103B, China).

The cultures of the seven bacterial strains (OD₆₀₀ = 0.4–0.6) were centrifuged at 8,000 rpm for 7 min and washed twice with sterile NMB3 medium, and the collected bacteria were added to the axenic culture of I. galbana in the exponential phase, whose cell density was approximately 1 × 10⁶ cells/mL, to give a co-culture with a bacteria/algae ratio of 1:1. The mixture was cultivated at 25°C for 14 days under fluorescent light (4000 lux) with a 12 h/12 h light/dark cycle, and the algal growth was assessed every two days by cell counting (Qin et al., 1999). All experiments were carried out in triplicate. Axenic I. galbana culture was used as the control.

The seven bacterial strains were tested for their antagonistic activities against V. vulnificus, V. parahaemolyticus, and V. cholerae using the disc diffusion method (Bauer et al., 1966). The isolates of the seven bacterial strains were prepared in a liquid 2216E medium, and sterilized filter paper discs were soaked in the culture medium of the bacterial strains for 1 h. The discs were then spread on plates containing solidified 2216E medium that was pre-inoculated with fresh Vibrio spp. cultures (100 µL, about 24 h old). The growth inhibition zones (GIZ) around each disc were measured after incubation at 28°C for 24–48 h. Cephalosporin and 2216E liquid medium were used as the positive and negative controls, respectively. To further verify the inhibitory effect of the bacteria on Vibrio growth, the co-cultures of Vibrio spp. with the bacterial strains were incubated in 2216E liquid medium at 28°C for 24 h. The cultures from each group were sampled and spread evenly on thiosulfate citrate bile salts sucrose agar plates, and the colonies of Vibrio spp. were counted.

The culture of P. flavipulchra was centrifuged (TGL-18M, China) at 4°C and 8000 rpm for 10 min, and the supernatant was passed through a sterilized 0.22 µm Millipore filter. The obtained solution of the P. flavipulchra ECP was stored at 4°C until use. To measure the inhibitory effect of the ECP on Vibrio spp., a 1:1 v/v mixture of the ECP solution and the culture of Vibrio spp. was prepared. The mixture was incubated at 28°C with shaking at 220 rpm, and OD₆₀₀ was measured every 3 h using a spectrophotometer (TU-180, China). The control group used the mixture of 2216E liquid medium and the culture of Vibrio spp. To evaluate the effect of ECP concentration on inhibition effect, the ECP solution (1–6 mL) was added into the culture of Vibrio spp. (10 mL). The mixtures were incubated at 28°C with shaking at 220 rpm, and OD₆₀₀ was measured using a spectrophotometer every 3 h up to 24 h. The control group used the Vibrio spp. culture alone with the addition of ECP.

The following treatments were applied separately to the P. flavipulchra ECP: (1) heating at 100°C in a water bath for 30 min, (2) adjustment of the solution pH 12 using 3 M NaOH, (3) incubation with proteinase K (1 mg/mL) at 37°C for 2 h. After adjusting to pH 7 (if needed), the treated ECP was then added to Vibrio spp. cultures to assess the antibacterial activity.

Proteins were isolated from the P. flavipulchra ECP through ammonium sulfate precipitation using a solution of (NH₄)₂SO₄ at 70%, 80%, or 90% saturation. The mixture was maintained at 4°C for 12 h, then centrifuged at 4°C and 8,000 rpm for 10 min. The collected precipitates were resuspended in phosphate-buffered saline (PBS) and dialyzed using a dialysis membrane. The antibacterial activity of the dialysate was tested by adding it to the culture of Vibrio spp. The mixture was incubated at 28°C with shaking at 220 rpm for 24 h, and OD₆₀₀ was measured using a spectrophotometer. The precipitates obtained using 90% ammonium sulfate were subsequently added into the culture of Vibrio spp. at varying concentrations, and the antibacterial activity was measured.

Statistical analysis was carried out using SPSS 22.0 (SPSS Inc., USA). All experiments were run in three replicates, and the mean value and standard deviation were calculated accordingly. Differences were considered statistically significant when P < 0.05.

All seven bacterial strains effectively promoted the growth of I. galbana. Oceanicaulis sp. and Marinobacter sp. significantly promoted the growth of I. galbana starting from day 2, whereas B. jeotgali, Alteromonas sp., P. flavipulchra, and A. macleodii promoted the microalgae growth starting from day 4. Dinoroseobacter sp. enhanced the growth of I. galbana starting from day 8 ( Figure 1 ). The growth of axenic I. galbana started to plateau after day 6, but the growth of I. galbana co-cultured with bacteria either did not slow down over the 14-day experimental period or had an inflection point at a much later time.

All seven bacterial strains were further tested for their antagonistic activities against V. vulnificus, V. parahaemolyticus, and V. cholerae using the disc diffusion method. Only the discs soaked with P. flavipulchra or cephalosporin produced clear growth inhibition zones ( Figure 2 ), suggesting that P. flavipulchra was the only bacterial strain that suppressed the growth of V. cholerae, V. vulnificus, and V. parahaemolyticus. To further verify the inhibitory effect of the bacteria on Vibrio growth, the co-cultures of Vibrio spp. with different bacterial strains were grown in 2216E liquid medium. The cell densities of V. cholerae, V. vulnificus, and V. parahaemolyticus were significantly reduced compared to the control upon co-culturing with P. flavipulchra ( Figure 3 ). The results corroborated the findings from the disc diffusion screening and confirmed that P. flavipulchra could inhibit V. cholerae, V. vulnificus, and V. parahaemolyticus.

To further analyze whether antibacterial activity of P. flavipulchra is due to the secretion of extracellular antibacterial substances, the P. flavipulchra ECP was added to Vibrio spp. cultures. The growth of V. vulnificus, V. cholerae, and V. parahaemolyticus was significantly inhibited by the addition of the P. flavipulchra ECP ( Figures 4A–C ), and the inhibitory effect increased with rising ECP concentrations. The inhibition rate reached 35.2%, 18.1%, and 30.7% for V. vulnificus, V. cholerae, and V. parahaemolyticus, respectively, when 6 mL ECP was added into 10 mL Vibrio culture ( Figures 4D–F ).

Next, V. parahaemolyticus was selected as a representative substrate to analyze the effects of heat, alkali, and proteinase K treatments on the antibacterial activity of the P. flavipulchra ECP ( Figure 5 ). The P. flavipulchra ECP completely lost its antibacterial activity against V. parahaemolyticus after it was heated, exposed to alkali, or treated with proteinase K. Thus, the antibacterial substance in the ECP of P. flavipulchra likely had proteinaceous characteristics.

To further determine whether the antibacterial substances in P. flavipulchra ECP are proteins, the proteins in the P. flavipulchra ECP were isolated by precipitation using ammonium sulfate. The proteins precipitated using 90% ammonium sulfate had the highest antibacterial activity and were used for further assessments ( Figure 6A ). Further testing of these protein fractions revealed a concentration-dependent increase in antibacterial activity against on V. parahaemolyticus, demonstrating that higher protein concentrations enhanced the antibacterial effect ( Figure 6B ).

To date, many MGPB have been identified. They demonstrate important regulatory effects on the growth and metabolism of microalgae. For instance, co-culturing Pseudomonas putida and Chlorella vulgaris significantly increases the cell density of C. vulgaris (Shen et al., 2017), and Rhizobium strain 1011 enhances the chlorophyll levels and lipid content in microalgae by supplying vitamin B₁₂ (Do Nascimento et al., 2013). Nevertheless, the impact of MGPB on the aquaculture animals feeding on microalgae is underexplored, particularly concerning their antagonistic effects against Vibrio species. This work thus aimed to screen MGPB capable of Vibrio antagonism and analyze the antibacterial substances from the MGPB.

Vibrio species are widely recognized as the major causes of mass mortality in marine aquaculture (Wang et al., 2022). Although antibiotics have long been employed to prevent Vibriosis, the excessive application of antibiotics has led to environmental degradation and the emergence of antibiotic-resistant bacteria (Rico et al., 2017). This situation has heightened the need for alternative approaches. Probiotics have emerged as a promising option for disease prevention in aquaculture (Dawood and Koshio, 2016; Lieke et al., 2020; Wang et al., 2008; Doan et al., 2020). Recent studies have demonstrated that probiotics can improve growth performance (Li et al., 2024), enhance immunological responses (Costa et al., 2024), and inhibit pathogenic microorganisms (Wang et al., 2020a). However, most studies on probiotics focus on the isolation and screening of bacterial strains from healthy aquaculture animals, and limited attention is given to the antagonistic effects of MGPB against pathogens such as Vibrio species. The present study revealed that P. flavipulchra simultaneously promoted the growth of I. galbana and inhibited V. vulnificus, V. cholerae, and V. parahaemolyticus, suggesting that it might be a promising probiotic for shellfish aquaculture. Nevertheless, the application of P. flavipulchra as a probiotic in aquaculture faces several challenges. For instance, variations in environmental conditions, such as temperature and salinity, can influence the activity and survival of the bacterium, and further study is needed to evaluate if P. flavipulchra can maintain its anti-Vibrio activity despite these fluctuations. In addition, it may not be straightforward to deliver P. flavipulchra in a way that ensures high survival and activity, and methods like freeze-drying or incorporation into the shellfish feed may reduce bacterial viability. Therefore, the application of P. flavipulchra as a probiotic in aquaculture still requires additional investigation.

Several mechanisms have been proposed for the inhibition of various Vibrio species by probiotics, including competition for adhesion sites, competition for nutrients, immune system stimulation, disruption of quorum sensing (QS), and production of inhibitory substances (Alcaide et al., 2005; Chauhan and Singh, 2019). Most commonly, the probiotics synthesize and secrete antibiotics, antibacterial agents, lysozymes, proteases, bacteriocins, butyric acid, small molecules, or organic acids that have antibacterial properties (Kesarcodi-Watson et al., 2008; Wang et al., 2019). For example, Weissella cibaria KY10 completely inhibits the growth of V. parahaemolyticus T11.1 through the secretion of various antibacterial substances, primarily organic acids (Kanjan et al., 2022). Bacillus subtilis produces amicoumacin A to inhibit the growth of V. parahaemolyticus and other Vibrio species, thereby reducing the risk of vibriosis outbreaks (Wang et al., 2020b; Chen et al., 2024). Lactobacillus plantarum W2 inhibits seven pathogenic bacteria, including V. parahaemolyticus, and organic acids are believed to be active antibacterial agents (Wei et al., 2022). In this study, the ECP isolated from P. flavipulchra exhibited antibacterial effects against V. parahaemolyticus, but its potency was lost after treatment with heat, alkali, or proteinase K. Analogously, Fontoura et al. reported that the purified antibacterial substance from Pseudomonas sp. strain 4B is partially inactivated by proteinase K or trichloroacetic acid and suggested that a protein moiety is involved in the antibacterial activity (Fontoura et al., 2008). Thus, the antibacterial substances in the ECP of P. flavipulchra are likely proteinaceous.

Many studies have demonstrated that bacteria can produce proteinaceous substances to inhibit Vibrio species. For instance, Lactobacillus sp. M31 produces a novel iturin, known as iturin V, which exhibits antibacterial activity against Vibrio species (Singh et al., 2021). Pseudoalteromonas sluteoviolacea produces a 100 kDa protein that possesses L-amino acid oxidase activity and exerts antibacterial activity against Vibrio species such as V. parahaemolyticus (Gómez et al., 2008). Pseudoalteromonas sp. strain X153 generates an 87 kDa antimicrobial protein that can protect bivalve larvae against Vibrio (Longeon et al., 2004). In this study, the proteins precipitated from the ECP of P. flavipulchra using 90% ammonium sulphate exhibited strong antibacterial activity and inhibited the growth of V. parahaemolyticus in a concentration-dependent manner. However, further investigation is needed to identify and fully characterize the exact protein responsible for the antibacterial property.

Pseudoalteromonas is garnering increasing attention in aquaculture due to its ability to produce a myriad of antibacterial compounds, including chitinase, pigments, and antibiotics (Bosi et al., 2017; Richards et al., 2017). In addition, P. flavipulchra has been noted to produce vesicle-like structures, which may be involved in some novel antibacterial mechanism (Wang et al., 2021). Vibrio spp. typically use QS to regulate virulence, biofilm formation, bioluminescence, sporulation, swarming motility, host colonization, and other population behaviors (Milton, 2006). Pseudoalteromonas sp. strain DL3 exhibits not only broad-spectrum antibacterial activity but also the ability to quench QS signal molecules (Zhao et al., 2023). Xu et al. reported that healthy cuttlefish harbored both the pathogenic V. alginolyticus and the antagonistic Pseudoalteromonas piscicida and attributed the inhibition of V. alginolyticus to chemotaxis rather than the production of antibacterial substances (Xu et al., 2024). While the ECP of P. flavipulchra contained proteinaceous material that was antimicrobial, further research is necessary to find out whether P. flavipulchra also utilizes other mechanisms to inhibit Vibrio species.