Aeromonas hydrophila XS-91-4-1 was isolated from diseased Hypophthalmichthys molitrix from Prof. Aihua Li at the Institute of Hydrobiology, Chinese Academy of Sciences. Escherichia coli (E. coli) ATCC 25922 was purchased from ATCC and stored in our laboratory. Polydatin (CAS No. 27208-8006) with a purity of 98% was a commercial product of Sichuan Weikeqi biological technology Co., LTD, while enrofloxacin (CAS No. 93106-60-6) with a purity of 99% was purchased from the National Institutes for Food and Drug Control (Beijing, China). For in vitro studies, polydatin was prepared by dimethyl sulfoxide (DMSO, Sigma-Aldrich, St. Louis, USA) at 40.96 mg/ml. For the in vivo study, polydatin was dissolved in 0.5% carboxymethylcellulose sodium at 20 mg/ml.

Minimal inhibitory concentrations (MICs) of A. hydrophila XS-91-4-1 to enrofloxacin and polydatin were determined by the broth serial-dilution method in 96-well plates according to the guidance of CLSI M45-A3 (19). In brief, enrofloxacin and polydatin were diluted by two-fold in volumes of 100 μl in each well by Mueller-Hinton Broth (MHB, Hopebiol, Qingdao, China) at concentrations ranging from 64 to 0.125 μg/ml and 512 to 1 μg/ml, respectively. Bacterial strains stored at −70°C were thawed and sub-cultured for Luria-Bertani (LB, Hopebiol, Qingdao, China) medium at 28°C for 16–18 h. Bacterial cells were harvested by centrifugation (8,000 g, 2 min, 4°C) and washed three times with sterile PBS. Then, the concentration of bacterial cells was adjusted to about 5 × 10⁵ CFU/ml using the MHB medium by McFarLand standards. Bacterial suspension at a volume of 100 μl was added to each well and further incubated at 35°C for 16–20 h. DMSO was added into drug free well and was served as a positive control. E. coli ATCC 25922 was used as a quality control strain. The lowest concentration without visible bacterial growth was defined as the MICs.

Aeromonas hydrophila XS-91-4-1 was cultured in the LB medium for 16–18 h, then 1 ml of bacterial culture was inoculated to 100 ml fresh LB medium and further incubated at 28°C until the optical density at 600 nm (OD_600nm) reached 0.3. Bacterial cultures were divided into 5 glass flasks at a volume of 10 ml and then indicated concentrations of polydatin were added to each flask. The mixtures were further cultured for 5 h and the growth of bacteria was monitored by determining the absorption at 600 nm every 30 min. A drug-free group was co-cultured with DMSO to determine the impact of DMSO on bacterial growth.

An overnight bacterial culture (1 ml) was added to 100 ml of fresh LB medium and further cultured to OD_600nm of 0.3 at 28°C. Bacterial suspension was divided into five 50-ml glass flasks at a volume of 10 ml, then polydatin at concentrations ranging from 0 to 8 μg/ml was added to each flask and further incubated by shaking up to reach an OD_600nm of 1.5. Bacterial supernatants were collected by centrifugation and were used for determining aerolysin-induced hemolysis. Supernatants were treated with trypsin at room temperature for 10 min to activate aerolysin after being sterilized by 0.22 μm filters. Activated supernatants were used for the hemolytic activity assay. Briefly, the hemolytic reaction system was brought up in 1.5 ml tubes by 875 μl reaction buffer (25 mM Tris, 150 mM NaCl, pH = 7.2), 100 μl bacterial supernatant, and 25 μl freshly washed sheep red blood cells. The systems were mixed gently and further incubated at 37°C for 20 min. The mixtures were centrifuged at 10,000 g for 2 min to remove unlyzed erythrocytes. Erythrocytes treated with 0.1% Triton X-100 was served as a positive control (100% hemolytic activity). The hemolytic activities of bacterial supernatants treated with indicated concentrations of polydatin were determined by measuring the absorptions at 543 nm.

A primary anti-aerolysin polyclonal antibody was prepared by New Zealand white rabbits in accordance with the Animal Welfare and Research Ethics Committee of Yangtze River Fisheries Research Institute. Aerolysin was overexpressed and purified as per our previously reported method (20). Rabbits were hypodermically injected with aerolysin (500 μg) plus complete Freund's adjuvant, then boosted two times with aerolysin plus incomplete Freund's adjuvant at 2-weeks intervals. Serum collected from the carotid artery was used for the western blot assay.

Samples were prepared as described in the hemolysis assay. Before sampling, total proteins in supernatants were determined using a Pierce^TM BCA protein assay kit (Thermo Fisher Scientific, MA, USA). Samples were loaded on a sodium dodecyl sulfate (SDS)-polyacrylamide (12%) gel after being mixed with Laemmli sampling buffer and boiled. Proteins in the gel were transferred to a 0.22 μm PVDF membrane by a semi-dry transfer cell. Then, the membrane was incubated with the primary anti-aerolysin polyclonal antibody at a concentration of 1:1,000 for 2 h after being blocked by 5% non-fat milk in TBST for 1 h. The membrane was washed with TBST for 10 min, three times, and further incubated with a HRP-conjugated secondary goat anti-rabbit antiserum at a concentration of 1:40,000 for 1 h. After washing, proteins in the membrane were detected by ECL Western blotting detection regents. Pictures were photographed by a ChemiDoc XRS+ imager.

Pretreated A. hydrophila XS-91-4-1 cultures as mentioned earlier in the hemolysis assay were used for RNA extraction. Bacterial cells were acquired by centrifugation for 10 min at 4°C when OD600 nm of the cultures reached 1.5. A MolPure bacterial RNA kit was used to isolate total RNA from bacterial cells. cDNA was synthesized after determining the concentration of total RNA. Then a real-time PCR assay was conducted to determine the effect of polydatin on the transcription of aerolysin encoding gene aerA. As an internal housekeeping control gene, 16s rRNA was employed. Primer pairs for detecting aerA gene were 5'- TCTACCACCACCTCCCTGTC-3'(forward) and 5- GACGAAGGTGTGGTTCCAGT-3'(reverse), while 16s rRNA were 5'- TAATACCGCATACGCCCTAC-3'(forward) and 5- ACCGTGTCTCAGTTCCAGTG-3'(reverse) (21). CT values were used to calculate the relative expression levels of aerA by 2^−ΔΔT method as described in a previous study (22).

A549 alveolar carcinoma cells were cultured in the Dulbecco's Modified Eagle Medium (DMEM, Thermo Fisher Scientific, MA, USA) plus 10% fetal bovine serum with 5% CO₂ at 37°C. Cells were seeded into a 96-well plate at a density of 1.5 × 10⁵ cells per well and further cultured overnight at 37°C. To determine the protective effect of polydatin against aerolysin-induced cell injury, lactate dehydrogenase (LDH) release and live/dead assays were performed. Cells were incubated with bacterial supernatants treated with the indicated concentrations of polydatin for 1 h at 37°C. Cell supernatants were obtained and 120 μl of which were moved to a new plate, then 60 μl of LDH working solution was added to each well and further incubated at room temperature for 30 min. The percentage of LDH release was determined by measuring the OD_490nm values using a multi-mode microplate reader. For live/dead assay, cells treated previously were washed and were incubated with Calcein AM and Propidium iodide at concentrations of 2 and 4.5 μM, respectively. After a 30 min staining, cells were imaged by a fluorescence microscope.

Animal studies were carried out according to the guidance of the Institutional Animal Welfare and Research Ethics Committee. Channel catfish weighing 200 ± 10 g were separated into three groups, containing 30 fish in each group, and were maintained in 100 L glass tanks for 7 days prior infection. Bacterial cells of A. hydrophila XS-91-4-1 were collected when cultured to mid-logarithmic phase in LB medium. After washing, bacterial cells were re-suspended to a density of about 1.5 × 10⁸ CFU/ml in sterile PBS. Then, a fish infection model was established by intraperitoneal injection of bacterial suspension at a volume of 200 μl, while fish in the negative group were injected with the same volume of sterile PBS. A gavage needle was used to administer polydatin to fish in the polydatin treating group at a dosage of 20 mg/kg 6 h post-infection and then 12-h intervals for 3 days. Fish in positive and negative groups were given 0.5% carboxymethylcellulose as control. Deaths were monitored every 24 h for 8 days and mortality was analyzed.

The normality of hemolytic activity, gene expression, and cell viability data was analyzed by one sample K-S test and explore test, then the statistical significance was determined by Student's t-test. Kaplan–Meier estimates and log-rank tests were used to analyze the survival rate of animal studies by GraphPad Prism v5.04.

The MICs of enrofloxacin and polydatin against A. hydrophila XS-91-4-1 were 4 μg/ml and 512 μg/mL, indicating that polydatin had little inhibitory effect against A. hydrophila. Moreover, growth curves at 5 h were determined to assess the effect of polydatin on bacterial growth. As shown in Figure 1B, no impact was observed when A. hydrophila co-cultured with polydatin at concentrations ranging from 1 to 8 μg/ml. The addition of DSMO to polydatin-free groups both in MIC determination and growth curves assay showed that DMSO had no role on bacterial growth.

A hemolytic activity assay was performed to determine the impact of polydatin on hemolysis of bacterial supernatants co-cultured with indicated concentrations of polydatin. As shown in Figure 1C, hemolysis was reduced at a dose-dependent manner. The hemolytic activity decreased to 22.46 ± 3.42% when co-cultured with 8 μg/ml polydatin compared with 81.24 ± 13.08% of drug-free group. The hemolysis of bacterial supernatants with polydatin of 2 μg/ml was remarkably reduced (p = 0.024) compared with the polydatin-free group. Moreover, the western blot assay was performed to analyze the inhibitory effect of polydatin on the production of aerolysin in bacterial supernatants. As shown in Figure 1D, the levels of aerolysin in supernatants were decreased at a dose-dependent manner.

Real-time PCR (RT-PCR) was conducted to determine the effect of polydatin treatment on aerA transcription. As shown in Figure 2, the transcription of aerA was decreased following the addition of polydatin. The aerA gene was downregulated 1.99, 4.33, 6.68, and 9.51-fold with polydatin at concentrations ranging from 1 to 8 μg/ml compared with polydatin-free group. When A. hydrophila XS-91-4-1 co-cultured with polydatin at 2 μg/ml, the transcription of aerA was significantly (p = 0.018) downregulated. The findings demonstrated that polydatin inhibited the hemolytic activity and aerolysin production by suppressing the transcription of aerA gene.

It has been proven that A549 cells can be used as a cell line by assessing the cytotoxicity induced by aerolysin (23). Therefore, the protective effect of polydatin against aerolysin mediated cell injury was determined using A549 cells by live/dead staining and LDH release assays. Cells without any treatment were live and were stained with green fluorescence (Figure 3A), while the red of cells treated with polydatin-free bacterial supernatant (Figure 3B). Cells co-cultured with bacterial supernatant plus 8 μg/ml polydatin showed a visible decrease in dead cells (Figure 3C). Moreover, the results of the LDH release assay showed LDH release was decreased with increasing concentrations of polydatin (Figure 3D). When polydatin reached to 2 μg/ml, LDH release of A549 cells was reduced to 50.89 ± 5.42% compared with 82.03 ± 5.73%, which was statistical significance (p = 0.002, Figure 3D). Taken together, these findings demonstrated that treatment with polydatin could provide a significant protection to A549 cells from aerolysin induced cell injury.

In vitro results demonstrated that polydatin could significantly reduce aerolysin-mediated cell injury. Moreover, a previous study found that A. hydrophila deleting aerA showed a significant decrease in pathogenesis (24). Therefore, it is reasonable to believe that polydatin might have a therapeutic effect against A. hydrophila infection. As shown in Figure 4, deaths were observed on the first day post-infection, the survival rate of fish in the positive control group in 8 days' course was 0%, while 56.67% of fish in the 20 mg/kg polydatin treated group, which showed statistical significance (p = 0.0,003). No death was observed in the negative control group during the experimental period indicating that A. hydrophila infection resulted in deaths in the positive control and polydatin treated groups. These results suggested that the inhibition of aerolysin by polydatin could provide a significant protection to channel catfish challenged with A. hydrophila.