5.1 Bacterial strains, plasmids and culture conditions
The wild-type strain A. veronii TH0426 used in this study was initially isolated from a farmed yellow catfish in Zhejiang Province, China. All E. coli in this experiment were cultured at 37°C using Luria-Bertani (LB) solid or liquid medium. A. veronii was cultured at 28°C using LB solid or liquid medium and Aeromonas-selective solid Rimler-Shotts (RS) medium. The pEASY-Blunt Zero (pEASY), suicide plasmid pRE112 and broad-host expression plasmid pBBR1-MCS, were used for gene amplification, conjugation with the genome of A. veronii and gene expression, respectively. When required, the concentrations of antibiotics were 100 μg/mL ampicillin (Amp) for E. coli and A. veronii and 50 μg/mL chloramphenicol for E. coli. The bacterial strains and plasmids used in this study and their relevant characteristics are listed in Table 1.
5.2 Ethics statement
Adult AB/TU wild red zebrafish (3-6 months old), purchased from a commercial fish market in Changchun City, China, were approximately 0.15 g of body weight and approximately 2.8 cm of body length. The experimental animals were kept in a clean, pathogen-free (SPF) barrier environment. All experiments were performed in strict accordance with the regulations of the Animal Care and Use Committee of Jilin Agriculture University (JLAU08201409) and the National Institutes of Health Guide for the Care and Use of Laboratory Animals (NIH Publications No.8023). All remaining experimental animals were euthanized by bringing the concentration of clove oil in the water to 80 mg/L.
5.3 Sequence analysis
The nucleotide sequence of the A. veronii dotU gene was obtained from the A. veronii genome. The sequence of dotU was analysed using the BlAST program at the National Center for Biotechnology Information (NCBI) and the Expert Protein Analysis System. The DotU protein structure was analysed using the corresponding protein structure online prediction tools.
5.4 Construction of an A. veronii mutant and complemented strain
To construct the dotU deletion mutant strain of A. veronii, ∆dotU, 720 bp of the dotU gene were disrupted. In short, the first and second PCRs were performed with dotU-f/dotU-int-r and dotU-int-f/dotU-r to amplify the upstream and downstream fragments of the dotU gene, respectively. The two purified flanks were then ligated by fusion PCR, and inserted into a linear vector, pRE112, digested at the same restriction site, named pRE112-UD dotU. The resultant suicide vector was then transformed into the gene-engineering strain WM3064. After that, WM3064 and A. veronii were co-cultured to induce homologous recombination. The transconjugants that underwent the first homologous recombination were selected on LB agar plates supplemented with Amp and Cm. Subsequently, double crossover (DC) recombination was induced on plates containing Amp and 10% sucrose to spontaneously achieve suicide vector excision from the genome. All the primers for the construction of the corresponding strains are listed in Table 2.
A promoter was accurately screened to drive the expression of the dotU gene and carried by the dotU gene to be amplified using primers dotU-O-f/dotU-O-r. Then, the retrieved PCR product was ligated into the broad-host expression plasmid pBBR1-MCS of gram-negative bacteria to construct the expression plasmid pBBR-dotU. The plasmid was then introduced into the dotU mutant to construct the complemented strain. PCR and RT-PCR were used to detect whether the mutant and complemented strains were successfully constructed.
5.5 Growth characteristics, morphology and haemolysis activity
According to a previous method with some modifications, the growth curve of the bacterial strains was monitored by determining the OD600 value of each bacterial culture. Briefly, for colony counting, the three strains (wild-type A. veronii, mutant ∆dotU and complemented strain C-dotU) were cultured overnight for approximately 12 h. Then, the bacterial concentration was adjusted to the same starting point and the OD600 value was determined. Then the same density of the three bacterial suspensions was inoculated onto an LB agar plate and cultured for 48 h. The bacterial colony morphology was observed by Gram staining. The same volume of bacterial suspension was transferred to a new 50 mL conical flask containing liquid LB and cultured at 28°C with 180 rpm shaking, which was collected at intervals of 1 h. The OD600 value was determined, and the growth was recorded. The same volume of bacterial suspension was inoculated onto a sheep blood plate and cultured in a 28℃ incubator for 24 h to observe lytic activity. All experiments were performed at least 3 times.
5.6 Flagella formation and motility test
Flagellar staining of A. veronii, the deletion strain ∆dotU and the complemented strain C-dotU was performed using a flagellar staining kit (modified Ryu method, Solarbio Inc, Beijing). Then, the flagellar synthesis of the corresponding strains was observed by light microscopy, according to the kit instructions for specific test procedures. The swimming and swarming abilities were detected by LB containing 0.3% agar, 0.5% agar, and 5% glucose at 28°C for 24 h. Then the distances were measured later. Both experiments were repeated at least three times.
5.7 Biofilm assay
Based on the biofilm formation ability detection method used by Watnick PI et al[34], the 96-well plate method was used to detect the biofilm formation ability of the bacteria using the biofilm formation conditions of A. veronii with appropriate improvements. Briefly, overnight culture supernatant was removed, and the wells were gently rinsed 3 times with fresh phosphate-buffered saline (PBS). The attached cells were fixed with 99% methanol for 20 min, and after discarding, the wells were stained with 2% crystal violet dye solution for 10 min. After drying, the crystal violet was dissolved with 33% acetic acid and fully mixed. Ten replicates were set up per bacterium in each experiment to detect biofilm formation, and the assay was performed at least three times. A previous report[35] showed that A. veronii TH0426 was an efficient biofilm-producer and the evaluation standard of biofilm formation ability was performed as follows. i: when the OD575 value of the experimental group was less than or equal to that of the negative control group (OD575N), there was no biofilm formation ability (negative); ii: OD575N < OD575 ≤ 2OD575N, 2OD575N < OD575 ≤ 4OD575N and OD575 > 4OD575N were judged to be weaker, medium and stronger biofilm formation abilities, respectively; and iii: the OD575 value was exhibited by the mean (mean) ± standard deviation (SD) of three experiments.
5.8 Median lethal dose (LD50) in the zebrafish model
The LD50 values of all strains were determined in zebrafish as described previously to assess the pathogenicity of the three strains (A. veronii TH0426, ∆dotU and C-dotU). Bacterial culturing, colony counting and concentration adjustment were performed according to the above method. The bacterial liquid was diluted 10-fold with sterile PBS and set to 8 gradients. Then healthy zebrafish were divided into twenty-four groups with ten fish in each tank. Each of the fish was injected intraperitoneally with a 10 μL bacterial solution and the control group was treated with an equal volume of PBS. The mortalities were recorded over ten days after infection and moribund fish were removed and inspected visually for the status of the disease. Then LD50 values were calculated by Kou’s law[36].
5.9 Adhesion and invasion of EPC cells
The detection of adhesion and invasion ability which was improved by referring to the relevant method reported previously[37], was conducted to analyse the adhesion and invasion ability of the three strains (A. veronii, the mutant ∆dotU and the complemented strain C-dotU). In brief, EPC[38] cells were cultured in M199 medium containing 10% heat-inactivated foetal bovine serum and 1% double antibiotic (penicillin and streptomycin) at 25°C in an incubator with 5% CO2. EPC monolayers were grown for 24 h in 24-well tissue culture plates and infected by the three strains, which were washed and resuspended with M199 at a multiplicity of infection of 10:1. The same volume of sterile PBS was used as a control. After infecting cells for 1 h, the cells were washed several times with PBS and then lysed with 1% Trion X-100 at 25°C for 1 h. The cell lysate was mixed thoroughly and diluted properly, and then the number of bacteria was quantified by plate counting.
5.10 Cytotoxicity test
Based on related research[39], lactate dehydrogenase was used as a marker to evaluate the cytotoxicity of bacteria. Here, we evaluated the toxicity of the three strains to EPC cells according to the instructions of the cytotoxic kit. The infection status was observed to determine the optimal time of cytotoxicity detection. Finally, cytotoxicity was displayed by calculating the lactate dehydrogenase (LDH) release percentage for each strain.
5.11 Tolerance test
The stress resistance of the three strains was detected as described previously with some modifications[40, 41]. Briefly, an overnight culture was adjusted for concentration and then treated as follows. For the oxidative stress test: the washed cells were resuspended in 1 mL of sterile PBS containing 1 mM hydrogen peroxide at 28℃ with 180 rpm shaking for 1 h. For the pH tolerance test: the bacteria were suspended in 1 mL of sterile PBS at pH values of 3, 4, 5, 6, 8 and 10 and placed at 28℃ with 180 rpm shaking to culture for 1 h. For the heat stress tolerance test, the cells were resuspended in 1 mL of sterile PBS and incubated in 55℃ water for 1 h. For the osmotic pressure tolerance test, 1 mL of a bacterial solution in the logarithmic growth phase was serially diluted and then directly coated on solid LB plates containing 0.4 M NaCl. After that, the bacterium-coated plate was incubated overnight in a 28℃ incubator, and the colonies were counted. All tolerance tests contained a bacterial population resuspended in 1 mL of sterile PBS for 1 h at 28°C that was used as a blank control. Each tolerance test was repeated three times in parallel.
5.12 Statistical analysis
The data were analysed with one-way analysis of variance (ANOVA) followed by Duncan’s new multiple range test and Tukey’s test with SPSS 13.0 software. A significant difference was considered at *P<0.05 and **P<0.01.