The A. pleuropneumoniae 4074 strain, serovar 1, is highly virulent and cultured on tryptic soy agar (TSA; Difco Laboratories, USA) or in tryptic soy broth (TSB, Difco Laboratories, USA) supplemented with 10% fetal bovine serum (FBS) (Gibco, USA) and 10 μg/mL NAD (Sigma-Aldrich, USA). The ΔompW and CΔompW strains were constructed in this study, and the culture conditions were the same as the A. pleuropneumoniae 4074 strain. The pEMOC2 plasmid was from the laboratory of Prof. gerald-f. Gerlach and the pJFF224-XN plasmid was stored in our laboratory. The pEMΔompW and pJFF-ompW plasmid were constructed in this study. The E. coli β2155 strain was cultured on Luria-Bertani (LB, Oxoid Ltd. UK) agar or in broth. The A. pleuropneumoniae and E. coli strains were cultured at 37°C. If necessary, 2 μg/mL or 50 μg/mL of chloramphenicol (Sigma-Aldrich, USA) was added to the culture medium.

The method for construction of ΔompW has been reported previously (25). Briefly, a 648 bp internal deletion of the ompW gene was amplified with primers ompWUF/ompWUR and ompWDF/ompWDR utilizing single-overlap extension polymerase chain reaction (SOE PCR). Then a 2073 bp PCR product (pEompWUF/pEompWUR) was cloned into the suicide plasmid vector pEMOC2 to obtain the plasmid pEMΔompW (26), which included the 648 bp deletion fragment in frame. The pEMΔompW plasmid in E. coli β2155 strain was transformed into the wild-type strain of A. pleuropneumoniae with a single-step transconjugation system (27, 28). The ΔompW mutant was obtained by two homologous recombination steps and was identified with ompWMF/ompWMR primers.

The ompW gene was amplified with ompWF/ompWR primers for construction of CΔompW. The ompW gene was then connected to the shuttle vector pJFF224-XN and was electrically transferred into the ΔompW mutant to construct the corresponding complemented strain. Voltage was 2.5 KV, capacitance 25 μFD, and pulse resistance 800 Ω using an electroporation apparatus (Bio-Rad, USA). The complemented strain was screened with 2 μg/mL chloramphenicol and was identified with PJompWR/ PJompWF primers. The bacterial strains, plasmids used in this study are listed in Table 1 and the primers used in this study are listed in Table 2.

The effect of the ompW gene deletion on A. pleuropneumoniae growth was assessed as described previously (29). Briefly, the A. pleuropneumoniae, the ΔompW, and CΔompW strains were cultured in TSB supplemented with 10% FBS and 10 μg/mL NAD at 37°C overnight. The bacterial were then transferred to fresh TSB supplemented with 10% FBS and 10 μg/mL NAD at a 1:100 (V/V) dilution for logarithmic growth. The initial OD600 value was adjusted to 0.2 and the OD600 value of the bacterial suspension determined 1 h to 8 h after culture initiation for kinetic analysis.

The A. pleuropneumoniae and the ΔompW biofilm formation were as described previously with minor modifications (30). Briefly, an overnight bacterial culture was transferred to fresh TSB supplemented with 10% FBS and 10 μg/mL NAD at a 1:100 (V/V) dilution. Then a 100 μL bacterial suspension was added to the wells of a 96-well microplate and incubated for 12 h, 24 h, 36 h, and 48 h at 37°C. The bacterial culture liquid was removed, each well washed with PBS, and 100 μL of Crystal Violet dye solution (0.1%) added for 30 min. The wells were washed with PBS to remove unbound crystal violet dye and dried for 5 h at 37°C. Then 100 μL of a glacial acetic acid solution (33%, v/v) was added to each well and the microplate was shaken for 1 h. The absorbance at 590 nm was measured.

The bacterial morphologies of the A. pleuropneumoniae and the ΔompW strains were examined by SEM and TEM as described previously with minor modifications (31, 32). Briefly, the strains were cultured to logarithmic and steady state phases of growth. The bacteria were collected, washed three times with PBS, and fixed with 4% glutaraldehyde. Specimens were dehydrated, dried, and sprayed with gold. Bacterial morphologies were observed with a HITACHI SU8010 scanning electron microscope (Hitachi, Japan). For TEM, the strains were fixed with 0.1 M cacodylate buffer containing 5% glutaraldehyde and 0.15% ruthenium red at 37°C for 4 h and then treated with 1 mg/mL polycationic ferritin. Thin sections were prepared with a Tecnai G² 20 TWIN transmission electron microscope (FEI, Portland, OR, USA).

Stress resistance tests were conducted using the A. pleuropneumoniae, the ΔompW, and CΔompW strains. These strains were grown in TSA supplemented with 10% FBS and 10 μg/mL NAD at 37°C overnight. The bacterial were then transferred to fresh TSB supplemented with 10% FBS and 10 μg/mL NAD at a 1:100 (V/V) dilution for logarithmic growth. At an OD600 value of approximately 0.6, cells from 1 ml of broth cultures were centrifuged at 5000 g for 5 min. For the oxidative tolerance test, the cells were resuspended in 1 ml of TSB supplemented with 10% FBS, 10 μg/mL NAD and 1 mM hydrogen peroxide for 10 min or 20 min, respectively. The control samples of each strain were resuspended in 1 ml of TSB supplemented with 10% FBS and 10 μg/mL NAD without any treatment. Then, the cultures from oxidative stress resistance test were serially diluted in TSB, and spread on TSB plates supplemented with 10% FBS and 10 μg/mL NAD for CFU counting. Oxidative stress resistance was calculated as ((oxidative stressed sample CFU mL⁻¹)/(control sample CFU mL⁻¹)) × 100. The experiments were carried out in triplicate (20).

The minimal inhibitory concentration (MIC) of the A. pleuropneumoniae, ΔompW, and CΔompW strains for 19 antibiotics (rifampicin, Ofloxacin, polymyxin B, vancomycin, acriflavine, cefalexin, lincomycin, gentamicin, kanamycin, neomycin, penicillin, amoxicillin, tetracycline, streptomycin, vibramycin, enrofloxacin, florfenicol, sulfamethoxazole, and ceftiofur sodium; Yuanye biology Co., Ltd, China) was determined as described previously with minor modifications (33). Briefly, 50 μL of a TSB culture supplemented with 10% FBS and 10 μg/mL NAD of each strain were added to 96-well plates. The appropriate concentration of antibiotics was diluted 12 times continuously, and then 50 μL of diluted antibiotics were added to the 96 well plate to make the final concentration of 0.125 to 256 μg/mL or 0.015625 to 32 μg/mL. Bacterial concentration was adjusted to a 0.5 Macbeth turbidity using 5 mL of sterile normal saline. Then 100 μL the bacterial suspension was added to 10 mL TSB supplemented with 10% FBS and 10 μg/mL NAD for measurement. The positive control (with bacteria and without drugs) and the negative control (without bacteria and drugs) were assessed by adding 50 μL of the tested bacterial solution to each antibiotic well. The tested bacterial solution was cultured at 37°C for 24 h. The highest concentration without bacterial growth was observed as the MIC of the antibiotic and recorded.

To explore the effect of the ompW gene deletion on the expression levels of chose genes, total RNA of bacteria was extracted by TRIzol reagent (Invitrogen, USA) (34). The strains cultured in TSB supplemented with 10% fetal bovine serumand 10 μg/mL NAD. The RNA was reverse-transcribed to cDNA with a PrimeScript^TM II 1st Strand cDNA kit (Takara, Dalian, China). Amplification of cDNA was carried out using the SYBR Premix Ex Taq kit (Takara, Dalian, China). The rpmA, rpmB, xylG, modA, rplT, rpsA, malK_1, phnS_2, rpsC, rplU, fhuB, prokka_00896, rplC genes were analyzed and the kdsB gene encoding 3-deoxy-manno-octulosonate cytidylyltransferase was used as the internal control (35). The primers used for RT-PCR are found in Table 2.

Experimental data are presented as means ± SD. The difference between two groups was analyzed by the two-tailed Student's t-test. P values of < 0.05 were considered significant, ^*p < 0.05 and ^**p < 0.01.

The ompW gene deletion mutant, ΔompW, was constructed by homologous recombination and verified by PCR (Figure 1A). The CΔompW strain was confirmed with PJompWF/PJompWR (Figure 1A).

The in vitro effect of the ompW gene deletion on 8 h growth was assessed for the A. pleuropneumoniae strain, the ΔompW mutant strain, and the CΔompW strain. Results showed the growth kinetics of the A. pleuropneumoniae, the ΔompW mutant strain, and the CΔompW strain to not differ significantly for this incubation period (Figure 1B).

We explored the effect of ompW gene deletion on biofilm formation of A. pleuropneumoniae and ΔompW using a quantitative assay. The results showed that deletion of ompW gene reduced the biofilm formation ability of A. pleuropneumoniae at 12 h to 48 h, but the difference is not significant (Figure 2).

SEM and TEM morphologies were assessed for the A. pleuropneumoniae strain, the ΔompW mutant strain, and the CΔompW strain during logarithmic and steady state growth. There were no significant morphological differences for any of the strains during the logarithmic growth phase as judged by SEM (Figure 3A). However, for steady growth, the cell surface of the ΔompW mutant became rough, while the surfaces of the A. pleuropneumoniae strain and the CΔompW strain were smooth by SEM (Figure 3A), which indicated absence of the ompW gene resulted in a rough surface for A. pleuropneumoniae during steady growth. These surface morphologies were confirmed by TEM (Figure 3B).

The A. pleuropneumoniae, ΔompW, and CΔompW strains were exposed to oxidative stress conditions. When the strains were treated with 1 mM hydrogen peroxide for 10 min, the ΔompW strain survival rate was 0.21%, which was much lower than that of the A. pleuropneumoniae strain (0.64%) and the CΔompW strain (0.58%; Figure 4). When the strains were treated with 1 mM hydrogen peroxide for 20 min, the ΔompW strain survival rate was 0.18%, which was lower than that of the A. pleuropneumoniae strain (0.46%) and the CΔompW strain (0.44%). These results suggest that OmpW has a role in the tolerance of A. pleuropneumoniae to oxidative stress.

The A. pleuropneumoniae, ΔompW, and CΔompW strains were assessed for bacterial susceptibility to; rifampicin, ofloxacin, polymyxin B, vancomycin, acriflavine, cefalexin, lincomycin, gentamicin, kanamycin, neomycin, penicillin, amoxicillin, tetracycline, streptomycin, vibramycin, enrofloxacin, florfenicol, sulfamethoxazole, and ceftiofur sodium. As judged by MIC, deletion of the ΔompW gene did not affect bacterial susceptibility to most assessed antibiotics (Table 3). However the deletion resulted in two times sensitive to polymyxin B, kanamycin, and penicillin (Table 3).

The effect of the A. pleuropneumoniae and ΔompW strains on ribosome synthesis related genes and ABC transporter genes at the mRNA level was determined by RT-PCR. Three genes, rpmA, rpmB, and rplT were significantly upregulated and three genes, rpsC, rplU, and rplC were significantly downregulated within the ribosome synthesis pathway (Figure 5). For the ABC transporter pathway, deletion of the ompW gene upregulated xylG, and modA, while phnS_2, PROKKA_00896, and fhuB were downregulated (Figure 5).