Recombinant tandem epitope vaccination provides cross protection against Actinobacillus pleuropneumoniae challenge in mice

Actinobacillus pleuropneumoniae (A. pleuropneumoniae/APP) is the pathogen that causes porcine contagious pleuropneumonia. Actinobacillus pleuropneumoniae is divided into 18 serovars, and the cross protection efficacy of epitopes is debatable, which has resulted in the slow development of a vaccine. Consequently, epitope-based vaccines conferring Actinobacillus pleuropneumoniae cross protection have rarely been reported. In this study, B cell epitopes in the head domain of trimeric autotransporter adhesin were predicted, and 6 epitopes were selected. Then, the predicted epitopes (Ba1, Bb5, C1, PH1 and PH2) were connected by linkers to construct a recombinant tandem antigen (rta) gene. The RTA protein encoded by the recombinant rta gene was expressed, and finally the ICR mice were immunized with the RTA protein with or without inactivated Actinobacillus pleuropneumoniae (serovars 1 and 5b) and challenged with Actinobacillus pleuropneumoniae to evaluate the protective effect of the epitope-based vaccine and combined vaccine. The mice in the RTA-immunized group and RTA plus inactivated Actinobacillus pleuropneumoniae vaccine group had a significant improvement in clinical symptoms and a higher level of antibody in the serum than those in the control group. The RTA immune group had a 40% survival rate after Actinobacillus pleuropneumoniae infection, whereas the combination of RTA and inactivated Actinobacillus pleuropneumoniae produced very strong cross immune protection in mice, at least 50% (RTA IB1 + C5) and at most 100% (RTA IB5 + C1), whereas no cross immunoprotection was found in the solo Actinobacillus pleuropneumoniae immune group. Overall, the combination of the RTA protein and inactivated bacteria significantly enhanced the cross protection effects. This implies that RTA protein in combination with a suitable inactivated Actinobacillus pleuropneumoniae strain could be a candidate vaccine for porcine contagious pleuropneumonia.


Introduction
Porcine contagious pleuropneumonia caused by A. pleuropneumoniae is a serious respiratory infectious disease in pigs. It often causes acute outbreaks and is characterized by a high mortality rate, which causes very large economic losses in the global pig industry (Jeffrey et al. 2012). There are 18 serovars of A. pleuropneumoniae (Bosse et al. 2018;Liu et al. 2018), and the lack of cross immunoprotection among the major serovars has slowed the development of a vaccine. At present, research on an A. pleuropneumoniae vaccine has focused on the pilus protein (Sadilkova et al. 2012), lipopolysaccharide , outer membrane protein (Xie et al. 2016), adhesion (Zhou et al. 2013) and A. pleuropneumoniae toxin (Apx) exotoxin (Wu et al. 2018), which are involved in colonization, maintenance, lesion induction and host defense escape (Chiers et al. 2010). Apx toxins can be divided into four types, Apx I-IV; Apx IV exists in all serovars of A. pleuropneumoniae strains (Schaller et al. 1999) and is necessary for virulence (Liu et al. 2009). However, due to the complicated structure, the large size of the protein and limited effective epitopes, the subunit structure is not the best option for vaccine development.
Despite the existence of first-generation vaccines against A. pleuropneumoniae, the inactivated vaccine still has limitations. Due to the large number of A. pleuropneumoniae serovars, heterogeneous serovars cannot provide effective immune protection (Bosse et al. 2002). In the process of vaccine production, certain external factors may also reduce the immunogenicity of the vaccine. Different virulence factors are present in different serotypes, for example, APXI, is present in serovars 1, 5, 9, 10, 11, resulting in different immunization efficacies of inactivated whole-bacterium vaccines (Ramjeet et al. 2008;Seah et al. 2002). Therefore, commercially available inactivated vaccines do not provide complete protection against all serovars. Epitope-based vaccines are a new type of vaccine that has been developed in recent years. The characteristics that make these vaccines good candidates for their development are as follows: (i) Good safety and stability are conducive to convenient transportation and storage. (ii) Epitope-based vaccines can be produced in large quantities by genetic engineering. (iii) Specific antibodies produced against a small epitope can bind to the whole antigen with high specificity. (iv) Epitope-based vaccines can effectively stimulate the production of an effective immune response and avoid the side effects associated with targeting the nonepitope parts of the antigen, for example, hydrophobic amino acid sequences (Yin et al. 2016). However, there are few reports on A. pleuropneumoniae-focused epitope vaccine research, which has great potential for development and application.
We previously found that trimeric autotransporter adhesin (TAA) plays an important role in bacterial adhesion, aggregation and biofilm formation, and the head domain of TAA (ADH) is the main functional area (Qin et al. 2017;Wang et al. 2015;Xiao et al. 2012). We also found that Propionibacterium acnes (P. acnes) could prevent A. pleuropneumoniae infection (Lei et al. 2008). Further study showed that the PH1 epitope of Apx IV and the PH2 epitope of a high-affinity periplasmic zinc transporter protein had high homology with a single-stranded DNA-binding protein of P. acnes, which had a good immunoprotection effect (Lei et al. 2008). In addition, the cross epitopes Ba1, Bb5, and C1 of P. acnes and A. pleuropneumoniae screened by phage display technology could effectively prevent A. pleuropneumoniae infection . However, it is still unclear whether the combination of these epitopes has a good immune protection effect.
In this study, the epitopes of ADH were predicted by DNASTAR software and BepiPred 1.0 software, and the protective effect of the epitope-based vaccine was detected. These predicted epitopes were connected to the P. acnes antigen epitopes Ba1, Bb5, and C1 and the A. pleuropneumoniae epitopes PH1 and PH2 by the GSG linker. The recombinant rta gene combined with the P. acnes and A. pleuropneumoniae epitopes was synthesized, and the RTA protein was expressed by a prokaryotic expression system, purified through nickel column-affinity chromatography and identified by western blotting. Mice were immunized with the RTA protein and challenged with A. pleuropneumoniae. Then, the clinical symptoms, weight changes, lung index, number of T cells in the lung, levels of cytokines and survival rates were detected to evaluate the immune protective effect of the RTA protein.

Strains and growth conditions
Escherichia coli (E. coli) DH5α competent cells and BL21 (DE3) competent cells (Takara, Japan) were used for cloning and expressing RTA proteins, respectively and were grown at 37 °C in LB medium containing 50 μg/mL kanamycin.
The APP serotype 1 reference strain Shope 4074 (APP 1, CVCC259) and APP serotype 5 reference strain L20 (APP 5b L20, CVCC263) were donated by the Shanghai Entry-Exit Inspection and Quarantine Bureau. They were used for the challenge experiment and cultured in brain heart infusion (BHI, USA) medium supplemented with 10 µg/mL NAD (Sigma-Aldrich, USA) at 37 °C for 6 h with shaking at 180 rpm.

Generation of the target rta gene
TAAs (GenBank: ABN73547.1) were analyzed by daTAA software (https ://toolk it.tuebi ngen.mpg.de/dataa ), and the adhesin gene sequences were identified. Then, the DNASTAR and BepiPred 1.0 software programs were used to predict B cell epitopes of the adhesin gene; the epitopes containing α-helixes and β-folds were excluded as much as possible, and the epitopes containing random coils were selected as the dominant B cell epitopes.
These predicted epitopes were connected in series with Ba1, Bb5, C1, PH1 and PH2 by a linker (GSG). Moreover, the restriction enzyme sites EcoR I and Sal I, stop codon TAA and protective bases were added. The target rta gene (GenBank: MK603027) was synthesized by BGI-Tech after being converted into E. coli-preferred codons.

Construction of the recombinant vector
Bacteria containing the rta gene were cultured overnight to extract the PUC57 plasmid, which was digested by EcoR I and Sal I. Furthermore, the pET28a vector was treated similarly. The rta gene and the pET28a vector were recovered via 1% agarose gel electrophoresis and ligated overnight by T4 DNA ligase at 16 °C. The products were transferred into E. coli DH5α. After culture expansion, the plasmids were extracted and identified by double enzyme digestion. The identified plasmids were transferred into the E. coli BL21 (DE3) strain (ice bath 30 min, heat shock 45 s, ice bath 5 min) to express the RTA protein.

Expression and purification of the RTA protein
Cultures (200 mL) were grown to the exponential phase (OD600 nm = 0.6-0.8) with shaking at 37 °C in LB medium supplemented with 50 μg/mL kanamycin. Bacterial suspensions were induced with a final concentration of 1 mM IPTG and grown at 37 °C. Then, the bacteria were centrifuged at 5000× g for 8 min at 4 °C, washed three times with phosphate-buffered saline (PBS, pH 8.0), and disrupted by sonication (5 s of ultrasound with intervals of 5 s for a total of 30 min) in a mixture of ice and water.
The supernatants were harvested by centrifugation at 10000×g for 10 min at 4 °C and loaded onto a Ni-NTA agarose column (Qiagen, USA). Washing was performed with equilibration buffer, and the RTA protein was eluted with equilibration buffer supplemented with 250 mM imidazole. To remove the imidazole from the samples, the RTA protein was concentrated by ultrafiltration (Millipore, USA) and quantified by a bicinchoninic acid (BCA) protein assay kit (Thermo Fisher Scientific).

Identification by western blot analysis
The RTA protein was separated by 12% SDS-polyacrylamide gel electrophoresis (SDS-PAGE) and transferred to a PVDF membrane. Subsequently, anti-APP5b hyperimmune serum from pigs (stored in our laboratory, 1:5000 dilution) was added to the membrane and incubated for 60 min at 37 °C. Then, the peroxidase (HRP)-conjugated rabbit antibody against swine IgG was diluted (1:5000) and incubated with the membrane. Finally, TMB substrate was added, and the membrane was analyzed with a Tanon 5200 Multi (China).

Immunization and challenge in mice
For the immunization and challenge experiments, 166 female ICR mice (22-32 g) were purchased from the Animal Experiment Center of Jilin University. The specific laboratory environment was as follows: constant temperature (25 ℃), relative humidity at 40-70%, noise ≤ 60 dB, and artificial lighting (12 h cycle of day and night). All animal feeding and experiments were conducted according to the National Guidelines for Experimental Animal Welfare (Ministry of Science and Technology of China, 2006), and our experimental practices and standards were approved by the Animal Welfare and Research Committee at Jilin University.
Five kinds of vaccines (RTA, RTA + inactivated APP1, RTA + inactivated APP5b, inactivated APP1, and inactivated APP5b) were prepared and used in the study. RTA protein, RTA protein and inactivated APP1 (1 × 10 8 CFU; APP1, inactivated by adding formalin to the medium at a final volume of 0.25%), RTA protein and inactivated APP5b (1 × 10 8 CFU; APP5b, inactivated by adding formalin to the medium at a final volume of 0.25%), inactivated APP1 (1 × 10 8 CFU) and inactivated APP5b (1 × 10 8 CFU) were mixed with aluminum-gel brine adjuvant (V/V 4:1), respectively. All mice were injected subcutaneously (s.c.) with 0.2 mL of the appropriate vaccine at 0, 14 and 28 d, and nonimmune mice were administered PBS mixed with aluminum-gel brine adjuvant (V/V 4:1) at 0, 14 and 28 d. The details of the groups and treatments are shown in Tables 1 and 2. Challenge with APP was performed on the 35th day. A single clone of APP1 or APP5b grown on BHI medium was selected and added to 5 mL of BHI liquid medium for growth for 6 h at 37 °C with shaking at 180 rpm. The bacterial solution was centrifuged at 10,000×g for 5 min and resuspended in the original volume with PBS. The bacteria were washed three times with PBS, concentrated and enumerated. Each mouse was anesthetized and then intranasally challenged with 30 μL of APP1 (6 × 10 8 CFU) or 30 μL of APP5b (4 × 10 8 CFU) (Table 1). Seventy-two hours after infection, the mice used for sample collection were anesthetized with ketamine hydrochloride (35 mg/ kg, intramuscularly (i.m.)) and hydroxylamine hydrochloride (5 mg/kg, intramuscularly (i.m.)), then sacrificed and sampled (Table 1). Mice used for the survival experiment were observed until 72 h after infection, and the survivors were executed by cervical dislocation ( Table 2).

Detection of the antibody level in mouse serum by ELISA
Blood from mice in all groups was separately collected from the tail vein to isolate serum at 0, 14, 28 and 35 d. APP1 and APP5b bacterial suspensions were disrupted by sonication (ultrasound for 5 s at 5 s intervals) until the solution became clear and bright. The protein concentration was determined by the BCA method, and samples were then added to 96-well plates as coating antigens (1 μg per well). The antibody level in the mouse serum in each group was detected by indirect ELISA according to the reference (Yang et al. 2014). When the ratio of the OD value of the positive control serum and the negative control serum was ≥ 2.1, the test results were recorded at 450 nm.

Observation of clinical symptoms and weight changes in mice
Each mouse was scored according to its clinical symptoms, including appetite, mental state, activity, fur condition, eye secretions, and excreta, at 0, 6, 12, 24, 48 and 72 h after infection. Each clinical symptom of each mouse was scored from 0 to 3. A score of 0 indicated normal, 1 indicated mild symptoms, 2 indicated moderate symptoms, and 3 indicated severe symptoms. The average score of the six clinical symptoms of each mouse was calculated. Finally, the average score of each group was calculated at 0, 6, 12, 24, 48 and 72 h after infection.
The weight of each mouse was measured and recorded at 0, 6, 12, 24, 48 and 72 h after infection. The change in the weight of the mice over time was analyzed.

Lung index
Mice were sacrificed at 72 h after infection, autopsied, and the lung weight was measured. The lung index of

Detection of the number of T cells in the lungs by flow cytometry
The lungs of the mice were ground into a homogenate, which was filtered through a 200 mesh screen, and then the red blood cells were lysed with red blood cell lysis buffer. Finally, a single-cell suspension was prepared, and according to the instructions of the lymphocyte isolation kit (Tianjin HaoYang Biological Products Company, China), the lymphocytes were separated from the singlecell suspension and enumerated. Then, 1 × 10 6 lymphocytes were added to a 1.5 mL sterile tube and centrifuged for 10 min at 1250×g. The supernatant was removed, and 500 μL of PBS was added to resuspend the cells. Two microliters of FITC-labeled anti-CD3 antibody (1:250 dilutions) was added under dark conditions, and the mixture was gently mixed and incubated for 30 min at 4 °C under dark conditions. The samples were washed twice with 1 mL of PBS, resuspended in 500 μL of PBS and analyzed as soon as possible by flow cytometry.

Detection of cytokines
All mice were sacrificed at 72 h after challenge, bronchoalveolar lavage fluid (BALF) and serum were collected, and the levels of cytokines, including IL-1β, IL-6, IL-10, IL-17, IL-18, IL-21, TNF-α, IFN-γ, G-CSF, GM-CSF, Fas-Ligand, and MIP1β/CCL4, in the BALF and serum from each group were detected by indirect ELISA. Detection was performed according to the manufacturer's instructions (Shanghai, China). The purpose of this experiment was to observe the effect of the RTA protein on cytokine regulation. Samples from groups 1, 2, 3, and 4 were tested.

Survival rate examination
To confirm the protective effect of RTA protein in infected mice, we examined the survival rate of mice infected with APP in which vaccination and challenge were performed as described above and then recorded the survival rates for the analysis of the cross protection ability of RTA protein and inactivated strains at 72 h after infection. The group designations and treatments are detailed in Table 2.

Statistical analysis
One-way ANOVA and Student's t test were used to compare the differences between different groups. The survival rates of the mice challenged with A. pleuropneumoniae were analyzed by the Gehan-Breslow-Wilcoxon test. P-values < 0.05 were considered statistically significant. All statistical analyses were performed using GraphPad Prism 6.0 software.

Design of the rta gene and prokaryotic expression and purification of the RTA protein
To identify the effective epitope of ADH, TAA was analyzed by daTAA software, which showed that the Ylhead-binding sequence was densely distributed in the 124-612 aa region of ADH (Fig. 1a). Furthermore, the B cell epitope in the 124-612 aa region was predicted by DNASTAR (Additional file 1: Fig. S1A, S1B) and BepiPred 1.0 software. Epitopes containing α-helix and β-fold structures are difficult to produce and were therefore were excluded. Six epitopes with high antigenicity were selected, named ADH1-6, and connected in series with Ba1, C1, Bb5, PH1 and PH2 through a flexible linker (GSG) (Fig. 1b). The optimized nucleic acid sequences were synthesized by BGI-Tech.
The rta gene and pET28a vector were ligated and transferred into the E. coli DH5α strain. The plasmids were extracted and identified by double enzyme digestion. The size of the rta gene and pET28a (462 bp and 5356 bp) were predicted, and gel electrophoresis showed that the fragments were within the expected size range (Fig. 2a). The 12% SDS-PAGE results showed that the RTA protein was mainly present in the supernatant rather than in the precipitate (Fig. 2b). To obtain a large quantity of RTA protein, the expression conditions for RTA were optimized. The concentration of RTA protein reached a maximal value after 6-8 h at 37 °C and was unaffected by the inducer (IPTG) concentration (Additional file 2: Fig.  S2A-S2C). The RTA protein was expressed under optimal conditions, the cells were lysed by sonication, and the protein was purified by Ni-NTA affinity chromatography. The protein had a predicted weight of 20.6 kDa, and SDS-PAGE showed that the expressed protein was at the expected weight; thus, the relatively pure target protein was obtained (Fig. 2c). The reactivity of the RTA protein was verified by western blotting. The results showed that the RTA protein was specific enough to be recognized by anti-APP5b hyperimmune serum (Fig. 2d), indicating that the RTA protein had good reactivity.

A high level of cross-serotype antibody is produced by combined immunization
The serum antibody level before challenge was detected by indirect ELISA using APP1 (Fig. 3a) or APP5b (Fig. 3b) as the coating antigens and reading the OD value at 450 nm. The results showed that the level of antibody gradually increased even though the serum from each group was diluted 10 times and whole vaccinated groups showed higher levels of specific antibody than the PBS group (P < 0.05 or P < 0.01), almost reaching a peak at the 28 d post immunization time point. To our surprise, compared with the groups immunized with inactive APP alone, all mixed-antigen-immunized groups (RTA + IB1 or IB5) showed dramatic increases in the levels of APP1or APP5b-specific antibodies after 28 d (P < 0.01), but there was no difference between the groups immunized with RTA alone and APP alone (Fig. 3a, b). In addition, compared with the mice in the PBS group, the mice in the inactivated APP1 strain-immunized group produced higher levels of specific antibodies against both APP1 and APP5b; similarly, the mice immunized with the inactivated APP5b also produced higher levels of antibodies against both APP5b and APP1, but the cross reactivity of the serum was not equal to that against the same serotype, P < 0.05 at 28 d and P < 0.01 at 35 d (Fig. 3c1, c2). These results show that mice in the RTA immunization group produced many specific cross-reactive antibodies against APP1 or APP5b, and the mixture of RTA with inactivated APP produced extremely strong specific antibodies and distinctly moderately reactive cross-reactive antibodies.

RTA protein vaccination alleviated the clinical symptoms of mice infected with APP1 or APP5b
The clinical symptoms of the challenged mice in each group were observed. The clinical symptoms of the mice in the RTA immunization group were more significantly alleviated than those of the mice in the PBS control group at 6, 12, 24, and 48 h after infection with APP1 or APP5b (P < 0.01, Fig. 4a1), hinting that RTA protein provides cross protection in vivo. In particular, compared with the PBS control groups, the inactive APP1 group and inactive APP5b group showed a dramatic decrease in clinical symptom scores only when challenged with APP1 and APP5b, respectively (P < 0.01, Fig. 4a2), but no differences were noted in the IB1-C5 and IB5-C1 groups, suggesting that inactivated APP cannot induce clear cross protection. Additionally, there was no significant difference between the RTAC1, RTAC5 and IB1-C1, IB5-C5 groups.
The weight of each mouse was measured at 0, 6, 12, 24, 48 and 72 h after challenge, and the weight change in the mice was observed. The results showed that the mice in all of the groups rapidly lost weight within 24 h after the challenge, and then the trend abated. Notably, the RTA immunization groups had significantly greater body weights at 48 h and 72 h (P < 0.01) than the PBS control groups; in particular, the RTAC5 group displayed the least weight change (less than 1 g), which was less than that in the PBSC5 group (approximately 2.5 g) at 72 h (Fig. 4b1). Slight weight changes were observed in the IB1-C1 and IB5-C5 groups compared with the PBS control groups at 48 h and 72 h (P < 0.01, Fig. 4b2).
Moreover, infection with A. pleuropneumoniae causes pulmonary edema in animals, which can be evaluated through the detection of the pulmonary indexes in mice. There was no significant difference in the lung indexes between the groups. However, compared with that in the PBSC5 group, the degree of pulmonary edema in the RTAC5 group showed a downward trend (Fig. 4c).
Furthermore, host T lymphocyte populations have been reported previously during APP infection; therefore, to evaluate the immune status of the mice at 72 h after challenge, the number of T lymphocytes in the lungs of the mice was detected by flow cytometry. The results showed that there was no significant difference in the number of T lymphocytes in the lungs between the RTAC5 group and the PBSC5 group (Fig. 4d).
These results indicate that the RTA protein plays a protective role by alleviating the symptoms of challenged mice, and the effect does not depend on the T cell-mediated immune response. Moreover, the increased antibody level suggests that the protective effect of the RTA protein likely depends on the B cell-mediated immune response.

RTA protein vaccination reduced the level of cytokines produced by mice after APP infection
Because proinflammatory cytokines usually play an important role in pneumonia caused by microbes, the levels of cytokines in the BALF and serum of each group were detected by indirect ELISA at 72 h post challenge in samples from sacrificed mice.
The results showed that although there was no significant difference in the cytokine levels, the average level showed a robust change trend distinguishing the vaccinated and PBS control groups. Based on the average level, the IL-17, IL-21, G-CSF and CCL4 levels in the BALF and the Fas-Ligand, IL-21, G-CSF, and CCL4 levels in the blood of the RTA group were generally lower than those of the PBS group, whereas the IL-6, IL-18, and IFN-γ levels in the BALF and TNF-α in the blood of the RTA group were higher than those of the PBS group. There was no consistent change trend in The serum antibody level of the mice in all groups was detected by indirect ELISA. Each group had 6 mice. a APP1 was used as the coating antigen. b APP5b was used as the coating antigen. c1 At day 28 of immunization, the antibody level of the mice in all groups was detected by indirect ELISA. c2 At day 35 of immunization, the antibody level of the mice in all groups was detected by indirect ELISA the levels of other cytokines (Fig. 5a-l). These results indicate that the RTA protein response could decrease the level of inflammatory cytokines and improve the level of anti-inflammatory cytokines at 72 h after APP challenge, when the levels of multiple inflammatory cytokines were extensively increased in the lung, thus playing an immunoprotective role in the pathogenesis of APP. Furthermore, the levels of certain cytokines, such as TNF-α and IFN-γ, were distinctly changed in lungs and peripheral blood (Fig. 5i, k) These results suggest that cytokines that constitute a complex network in animals that differs in the lungs and blood, and the inflammatory response is the result of the interaction of many anti-inflammatory and proinflammatory cytokines.

Vaccination with the mixed RTA protein and inactive strain conferred strong cross protection to mice
To evaluate the effect of the RTA protein on the survival rate of mice, APP1 was used to challenge the mice immunized with the RTA protein, and the survival rates of the mice were recorded. The results showed that compared with that in the PBS control mice, the time to death in the mice treated with the RTA protein was extended within the first 12 h, and the protection rate in the RTAtreated mice was only 40% (Additional file 3: Fig. S3).

Fig. 4
Detection of the clinical symptoms, weight changes, lung indexes, and number of T cells in the lungs of the mice from different groups. Each group had 6 mice. a1, a2 The clinical symptoms of each mouse were observed and scored at 0, 6, 12, 24, 48 and 72 h after challenge. b1, b2 The weight of each mouse was recorded at 0, 6, 12, 24, 48, and 72 h after challenge, and the weight changes of the mice were calculated (weight changes = the initial weight-the real-time weight); * represents the RTAC1 group or the IB5-C1 group or the IB1-C1 group compared with the PBSC1 group and was statistically significant; # represents the RTAC5 group or the IB1-C5 group or the IB5-C5 group compared with the PBSC5 group and was statistically significant. c The body weight and the lung weight of the mice were recorded at 72 h after challenge, and the lung indexes of the mice were calculated. d The mice in the RTAC5 and PBSC5 groups were challenged with APP5b. The number of T cells in the lungs was detected by flow cytometry at 72 h after challenge. * or #, P < 0.05; ** or ##, P < 0.01 This implies that the RTA protein has a limited protective effect on mice infected with a lethal dose of APP1, but the effect was not sufficient to maintain a high survival rate. RTA immunization can significantly reduce the clinical symptoms of mice infected with serotype 1 and serotype 5, and its effect resembles that of the inactivated strain against homoserotypes (Fig. 4). Therefore, we mixed RTA with inactivated APP1 and APP5b bacterial proteins, immunized mice for 35 d to be challenged with APP1 or APP5b, and recording the clinical symptoms and survival rates (Tables 1 and 2).
The clinical symptoms of the combined immunization groups were generally mild. The clinical symptoms of the RTA IB1 + C5 group were significantly less severe than those of the PBSC5 control group (P < 0.01) at 6 h after challenge with APP5b, and the clinical symptom scores of the RTA IB5 + C1 group were significantly lower than those of the PBSC1 control group (P < 0.01) at 6 h and 12 h after APP1 infection (Fig. 6a1, a2). In the combined immunization group, body weight loss was significantly reduced after infection. Compared with those in the PBSC1 or PBSC5 group, mice in the RTA protein plus APP1 vaccination group lost very little body weight at 6 h regardless of APP1 or APP5b infection (P < 0.01, Fig. 6b1): furthermore, the immunization with RTA plus APP5b significantly reduced the body weight loss in mice at 12 h, 24 h, 48 h and 72 h (P < 0.01, Fig. 6b2).
In addition, the degree of pulmonary edema in all coadministration groups showed a downward trend, but there was no significant difference (Fig. 6c).
The survival experiments showed that the protection rate of mice immunized with RTA and inactive APP1 was 100% after challenge with APP1 (RTA IB1 + C1 group) and 50% against APP5b (RTA IB1 + C5 group); the protection rate of mice immunized with RTA and inactivated APP5b was 66% (RTA IB5 + C5 group) against APP5b but 100% against APP1 (RTA IB5 + C1), and no cross immunoprotection was found in the sole APP immune group (Fig. 6d, Additional file 3: Fig. S3). This result shows that compared with the individual antigens, the combination of RTA and inactivated APP can produce very strong immune protection, at least 50% (RTA IB1 + C5) and at most 100% (RTA IB5 + C1). Overall, the combination of the RTA protein and inactivated bacteria significantly enhances the cross protection effects, especially the combination of RTA and APP5b, which was the best antigen for serotype cross protection. The clinical symptoms of each mouse were observed and scored at 0, 6, 12, 24, 48 and 72 h after challenge. b1, b2 The weight of each mouse was recorded at 0, 6, 12, 24, 48, and 72 h after challenge, and the weight changes of the mice were calculated (weight changes = the initial weight-the real-time weight); * represents the RTA IB1 + C1 group or the RTA IB5 + C1 group compared with the PBSC1 group and was statistically significant; # represents the RTA IB1 + C5 group or the RTA IB5 + C5 group compared with the PBSC5 group and was statistically significant. * or #, P < 0.05; ** or ##, P < 0.01. c The body weight and the lung weight of the mice were recorded at 72 h after challenge, and the lung indexes of the mice were calculated. d. The survival rates of the mice immunized with PBS, RTA protein combined with inactivated APP1 or APP5b at 0, 14, and 28 d and challenged with APP1 and APP5b separately at 35 d. The mice were observed for 3 d to record the survival rate

Discussion
Porcine contagious pleuropneumonia is a serious threat to the modern pig industry, but vaccine development for this disease is slow due to the lack of cross protection induced by inactivated vaccines, and many studies have focused on the subunit structure. However, due to the complex subunit structure, there are few effective antigenic epitopes exposed on the surface of the protein that could cause a cross immune-protection response. Therefore, the epitopes of ADH were predicted by software and connected in series with the Ba1, C1, Bb5, PH1, and PH2 epitopes found by our group through a flexible linker (GSG) to form an epitope antigen. Then, the RTA protein was expressed, purified, and used to immunize mice that were subsequently challenged with A. pleuropneumoniae. Our results showed that the RTA protein could alleviate the clinical symptoms, regulate the levels of antibodies and cytokines, such as TNF-α and IFN-γ, and had a limited protective effect against APP infection; however, RTA proteins plus inactivated APP produced dramatic cross protection. This implies that RTA with a suitable inactivated APP strain could be a candidate vaccine for porcine contagious pleuropneumonia.
All the epitopes were short peptides with low molecular weights, which made their immunogenicity extremely low. To enhance the immunogenicity of these epitopes, they were connected head to tail with a GSG linker. The addition of a flexible linker was of great importance for increasing the immunogenicity of the polypeptide chain. Furthermore, the linker could reduce intermolecular interactions and maintain the protein structure. Yano et al. used two lysine residues as linkers to connect T cell epitopes and B cell epitopes to solve the problem of MHC restriction (Yano et al. 2005). RTA was successfully constructed using a GSG linker and induced a better immune response than other antigens, but we should perform further studies on the linker to improve its antigenicity.
In addition, we found that ADH-immunized mice had a 40% survival rate when infected with a lethal dose of APP5b (Xiao et al. 2012), whereas immunization with Ba1 and Bb5 produced 80% and 90% survival rates, respectively, when in mice infected with APP1 and APP5b . P. acnes Ssb, which has a high level of homology with PH1 and PH2, produced 80% and 60% survival rates in mice infected with APP1 and APP5b, respectively . In this study, the RTA protein produced a 40% protection rate in APP-infected mice. This result may be due to the high proportion of ADH epitopes in the RTA protein, whose protective effect was not as good as that of other polypeptide epitopes. The reason may also be that the selected linker is not optimal and causes changes to the RTA protein spatial structure due to intermolecular forces, thus blocking some of the key epitopes in the middle of the protein. In addition, the epitopes in the RTA protein are linked together, which limits the activity of each epitope and prevents specific immune responses to each epitope, thus reducing the protective effect of the RTA protein. This problem can be effectively overcome by immunizing mice with the combination of the RTA protein and inactivated bacteria, especially coimmunization of RTA with inactivated APP5b, which produced a 100% survival rate in mice challenged with APP1. This finding implies that choosing a suitable APP strain to mix with RTA may be a better way to improve candidate APP vaccines in the future.
There are many reports in which the current mouse model was used to evaluate novel A. pleuropneumoniae vaccines. For example, Shao et al. evaluated the immunological effect of recombinant Apx toxins and the recombinant outer membrane protein (OMP) (Shao et al. 2010). Park et al. used a murine model and showed that the M cell-targeting ligand-conjugated ApxIIA toxin fragment induces protective immunity against A. pleuropneumoniae infection (Park et al. 2015). In addition, we found that the clinical symptoms of mice infected by APP are similar to those of infected piglets when we researched the APP virulence factor ADH. This result indicates that the results obtained from the mouse model played an important role in guiding APP infection research in pigs ). Therefore, a mouse model can provide an important reference and information for researching the pathogenesis mechanisms of APP and developing vaccines against APP.
At present, most attention is paid to the ApxIV recombinant subunit vaccine, which also has cross protection, but other antigens, such as pili, could be helpful for producing immune protection (Shao et al. 2010). Current commercial vaccines cannot provide complete protection because the fixed antigen that induce complete immune protection have not been identified (Rycroft and Garside 2000). Some scholars have found that the recombinant ApxIV protein combined with inactivated APP1, known as a multicomponent vaccine, has a good protective effect against both homologous and heterologous A. pleuropneumoniae (Wu et al. 2018). In this study, the immunization with the combined vaccine addressed the defects of inactivated A. pleuropneumoniae and RTA protein alone, reducing the clinical symptoms and improving the survival rate of A. pleuropneumoniae-infected mice. This approach could improve the immune effect and provide a theoretical basis for developing a comprehensive and safe pleural pneumonia vaccine.