Plasmids, Bacterial Strains, Growth, Media and Animals
All the bacterial strains and plasmids used in this study, and their relevant characteristics and sources are described in Table 1. R. anatipestifer RA-GD and LJW-2 strains are cultured in tryptic soybean broth (TSB, Oxoid) or tryptic soy agar at 37℃ (TSA, Oxoid) with 5% calf serum in 5% CO2. E. coli X7213 is grown at 37℃ in Luria-bertani (LB, Oxoid) broth or on LB agar with 2,6-diaminoheptanedioic acid (DPA) of 50 μg/ml. 8-day-old Cerasus pseudocerasus ducklings were purchased from Sichuan Mian Ying Duck Co. Ltd (Chengdou, China). Animals are kept in clean buildings and the temperature of the room is maintained at 28℃–30℃. In addition, the room should be ventilated twice a day. All animal studies complied with the guidelines of Lanzhou Veterinary Research Institute Animal Care and Use Committee.
Construction of RA-GD∆RecA
All primers needed in this work were shown in Table 2. RA-GD∆RecA mutant with deletion of RecA gene were constructed as described previously . Briefly, the loci of RecA gene in the RA-GD∆RecA mutant were replaced with erythromycin (ErmF) resistant cassette which was amplified with the primers ErmF-F/ErmF-R, using the genomic DNA of LJW-2 strain as template. The primers Up-F/Up-R and Dp-F/Dp-R were used to amplify the upstream (600 bp) and downstream (600 bp) homologous arm regions of RecA gene, respectively, using the genomic DNA of RA-GD as template. Then, pRE112 plasmid was digested with SacI and SphI restriction endonucleases to obtain vector fragment. Each segment of three PCR fragments (upstream homologous region, ErmF resistant cassette and downstream homologous region) and pRE112 plasmid fragment were integrated with Unique CloneTM Plus Muti One Step Cloning Kit (Novogene, Tianjin, China) to generate the recombinant plasmid pRE112::ErmF-600H. The X7213 strains carrying pRE112::ErmF-600H and R. anatipestifer RA-GD strain were co-cultured and pRE112::ErmF-600H plasmids were introduced into RA-GD strain by conjugation. RA-GD∆RecA mutants were selected with erythromycin of 2 µg/ml and confirmed with conserved primers RA-OmpA-F/RA-OmpA-R and the identifying primers RecA-F/RecA-R and ErmF-F/ErmF-R.
Construction of RA-GD△RecA pCPRA::RecA
Complementation experiments were performed as described previously . Briefly, the RecA gene was amplified with pCP-RecA-F/pCP-RecA-R primers. RecA and pCPRA plasmid digested with PstI and SphI endonucleases were ligated using Unique CloneTM One Step Cloning Kit (Novogene, Tianjin, China) to generate the complemented plasmid pCPRA::RecA. The X7213 strains carrying pCPRA::RecA were selected with ampicillin of 100 µg/ml. pCPRA::RecA plasmid from X7213 was introduced into RA-GD△RecA by conjugation. The complemented strains RA-GD△RecA pCPRA::RecA were selected with erythromycin of 2 µg/ml and cefoxitin of 1 µg/ml. Further, the strains were confirmed with conserved primers RA-OmpA-F/RA-OmpA-R and the identifying primers RecA-F/RecA-R.
Growth curve and Competition Experiments in Vitro
The growth curves for RA-GD, RA-GD△RecA and RA-GD△RecA pCPRA::RecA strains were monitored under non-competitive conditions, based on OD600 values at different time points. Namely, 108 colony-forming unit (CFU) of each strain was inoculated into 5 ml TSB containing 5% calf serum and cultured at 37℃ with 200 rpm. The concentrations of bacteria solution under the 600nm wave length were measured from the first hour until the twelveth hour with an interval of one hour. The growth curve for each strain was drawn according to the OD600 values.
In vitro competition experiments were performed for RA-GD, RA-GD△RecA and RA-GD△RecA pCPRA::RecA strains as the reference . The competition index (CI) was calculated as the ratio between RA-GD CFU and RA-GD△RecA CFU. Similarly, CI between RA-GD△RecA pCPRA::RecA and RA-GD△RecA was calculated.
Drug sensitivity test
According to the Clinical and Laboratory Standards Institute guidelines (CLSI, 2016), the minimal inhibitory concentrations (MICs) of antimicrobial agents for RA-GD, RA-△RecA and RA-GD△RecA pCPRA::RecA strains were determined with 2-fold serial broth microdilution method. The procedure was carried out according to the references [26, 27]. The concentrations of antimicrobial agents ranged from 512 to 0.25µg/ml. E. coli ATCC 25922 was used as control. All tests were performed in triplicates. The MIC was recognized as the minimum concentration of the antimicrobial agent that could inhibit visible growth of bacteria (CLSI, 2016).
Duckling toxicity test
This experiment was performed as our previous work . The virulence of RA-GD, RA-GD△RecA and RA-GD△RecA pCPRA::RecA strains was defined according to the median lethal dose (LD50) and percent survival of the ducklings [26, 28]. After this work, the survivals were narcotized by pectoral muscle injection of pentobarbital sodium in 20 mg/kg of body weight. Then, the ducklings with deep narcosis were euthanized by cervical dislocation.
Transcription and expression of RecA gene
The parent strain RA-GD, deletion strain RA-GD△RecA and complemented strain RA-GD△RecA pCPRA::RecA were subjected to UV treatment. When the fresh bacterial solution reached to OD600 value of 1.0, the above three strains were irradiated with a UV dose of 60 μW/cm2. We inflicted the UV damage factor on the strains at 0, 15, 30 and 45 min, respectively, thereby acquiring a series of treated samples.
The transcriptional level of RecA gene was tested with the primers RecA-qRT-F/ RecA-qRT-R by qRT-PCR assay. 16S gene amplified with the primer pairs of 16S-F/16S-R was used as an internal control. In addition, the transcriptional levels of downstream RIA_1470 genes of RA-GD and RA-GD△RecA strains were detected with RIA-1470-F/RIA-1470-R primers to exclude the possibility that erythromycin resistant cassette had a polar effect on the transcription of adjacent genes. The amplification procedure was the same as the reference . The relative gene expression levels were quantified according to the comparative 2−△△CT method .
The complete RecA gene was amplified with the primers pET-RecA-F/pET-RecA-R, using the genomic DNA of RA-GD strain as the templates. RecA gene was cloned into pET-32a vector with BamHI and XhoI to generate recombinant plasmid pET-32a-RecA, and this plasmid was introduced into E. coli BL21(DE3). RecA protein carrying Trx-tag and His-tag was expressed in BL21(DE3) with 1 mM/L IPTG induction. The RecA protein, purified with nickel-nitrilotriacetic acid (NTA) column (TransGen, Beijing) according to the maufacture's manual, was used as antigen to immunize rabbit and generate polyclonal antibody. The expression level of RecA protein of R. anatipestifer was detected by Western blotting using rabbit polyclonal antibody and HRP-labeled goat anti-rabbit secondary antibody (Sigama, USA). These tests could determine the effect of UV on the transcription and expression levels of RecA.
DNA damage test
The experiment was carried out to measure the genome integrity of cells to analyze the damage condition of DNA, according to the reference  with appropriate modification. Briefly, for each sample, 500 μl of cell suspension (100 cells/μl) were irradiated with a UV dose of 60 μW/cm2 for 45 min in a 24-well plate. Then, genomic DNA was extracted from each cell sample with Universal Genomic DNA Extraction Kit (TakaRa, China). 55 μl of 0.2 M NaOH was added to 50 μl of DNA solution, and a pH value of 12.5 was reached under these conditions. Then, 105 μl of DNA solution for each sample was transferred to the 96-well plate. Thereafter the temperature setting was shifted to 30°C and kept constant for 60 min in order to allow alkaline unwinding of the DNA. Prior to the addition of 70 μl of neutralization solution (14 mM β-mercaptoethanol; 1 M glucose) at a rate of 200 μl/s, the temperature was shifted to 22°C. For the positive control sample, an internal standard representing cells with 100% double-stranded DNA, 70 μl of neutralization solution was added prior to 55 μl of 0.2 M NaOH. The negative control without any double-stranded DNA only consisted of 50 μl of DNA elution buffer, 55 μl of 0.2 M NaOH and 70 μl of neutralization solution. After dispensing 78 μl diluted SybrGreen (1:8333 in H2O), respectively, the samples were mixed by pipetting a volume of 253 μl up and down once at a rate of 100 μl/s. At last, samples were analysed in a 96-well-plate fluorescence reader at 492 nm excitation/520 nm emission immediately after SybrGreen addition. The relative fluorescence signals for each sample was calculated as follows: fluorescence signals (%) =Sample-Negative/Sample-Positive. Statistical significance was evaluated by using Student's t-test. In addition, in order to compare the changes of parental strain, deletion strain and complementary strain in different UV durations, we calculated and compared the colonies of the above three strains at different time durations of UV for 0 min, 15 min, 30 min and 45 min.
Statistical analysis was carried out using GraphPad Prism version 6.0 Windows software. Student’s t-test was used to ascertain the significance of the between-group differences. A value of *P < 0.05 was considered significant difference, and **P < 0.01 and ***P < 0.001 were all considered highly significant.