The rabbits used in this study were purchased from KangDa Biotechnology Co., Ltd. (Qingdao, China). All animal experimental procedures were evaluated and approved by the Institutional Animal Care and Use Committee (IACUC) of the Qingdao Haihua Bio-pharmaceutical Technology Co., Ltd. (Haihua).

The optimized gene fragment sequence (in the supplementary document) in our study, three fragments including the p30 gene (GenBank No. M96354.1), which encodes amino acids 46 to 251, the p54 gene (GenBank No. MZ812353.1), which encodes amino acids 18 to 203, and the p72 gene (GenBank No. MN886930.1), which encodes amino acids 213 to 358, were used as the target gene fragments, intending that specific antigens play a vital role in vaccines to improve high antigenicity and hydrophilicity of antigens, and the base sequences were obtained from NCBI, as well as the LTB gene (GenBank Nos. AF359362.1) and the His-tag gene sequences, which were inserted into downstream of the target genes, and all of the genes were cloned into the pMG36e vector (Lactobacillus expression plasmid vector, derived from our laboratory). Six recombinant plasmids, including pMG36e-p30/p54/p72-His and pMG36e-p30/p54/p72-LTB-His, were synthesized by Zhongmei Taihe Biotechnology Co., Ltd. (Beijing, China) and sequenced by Paisenor Gene Biotechnology Co., Ltd. (Qingdao, China).

For the construction of the recombinant Lactococcus lactis MG1363, the MG1363 competent cells were prepared as follow. The MG1363 stored in our laboratory at –80°C was scribed on a GM17 plate and incubated at 30°C for 36 h, Then, single colony was inoculated in 2 ml of GM17 liquid medium and incubated at 30°C for 24 h, and then inoculate into 30 ml SGM17G hypertonic medium according to 5% inoculation ratio and cultured until OD₆₀₀ = 0.8. Before centrifuging at 5,000 g for 10 min, the bacterial solution was incubated in an ice bath for 10 min at 4°C, and the supernatant was discarded. The precipitate was resuspended by 10 ml pre-chilled Wash Buffer with 10% glycerin containing 0.5 M sucrose by centrifugation at 5,000 g for 10 min at 4°C, discard the supernatant, and repeat twice. Ultimately, the cells were resuspended as competent cells by 400 μl pre-cooled washing solution.

The six recombinant plasmids were electroporated into MG1363 competent cells using an electrotransformer (2,200 V, 200 Ω, 2 msec). The recombinant plasmids pMG36e-p30/p54/p72-His and pMG36e-p30/p54/p72-LTB-His were identified by PCR (forward primer: 5’-AATATCGTAGCGCCGGGGTA-3′; reverse primer: 5’-GCCTCCTCATCCTCTTCATC-3′) and sequencing by Paisenor Gene Biotechnology Co., Ltd. (Qingdao, China). The positive recombinant L. lactis were named MG1363/pMG36e-p30-His, MG1363/pMG36e-p54-His, MG1363/pMG36e-p72-His, MG1363/pMG36e-p30-LTB-His, MG1363/pMG36e-p54-LTB-His, MG1363/pMG36e-p72-LTB-His, and the negative control was named MG1363/pMG36e.

The recombinant bacteria were cultured in 2 ml GM17 liquid medium containing Erm (5 μg/ml) at 30°C for 18 h, subsequently transferred to 30 ml GM17 liquid medium with 2% inoculum and incubated at 30°C for 14 h until OD₆₀₀ = 0.8, centrifuged for 10 min at 12,000 rpm at 4°C, and resuspended in pre-chilled PBS. After blending, samples were processed by ultrasonic fragmentation and analyzed by the same quantity in the precipitations of each sample was isolated by sodium dodecyl sulfate 15% polyacrylamide gel electrophoresis (SDS-PAGE). Ultimately, Western blot analysis followed by an anti-His tag monoclonal antibody as the primary antibody and HRP-labeled goat anti-mouse IgG as the secondary antibody (Beijing Zhongshan Jinqiao Biotechnology Co., Ltd.).

A total of 40 female New Zealand rabbits were randomly divided into four groups, with 10 rabbits per group (n = 10), namely, a PBS group, a MG1363/pMG36e group, a group without LTB adjuvant vaccine, and a group with LTB adjuvant vaccine. In group 1, each rabbit received 3 ml of PBS (pH = 7.4); in group 2, each rabbit was given the same dose concentration of 1.0 × 10⁸ CFU/ml MG1363/pMG36e; in group 3, each rabbit was immunized with the same dose and concentration of 1 ml each of recombinant L. lactis MG1363/pMG36e-p30 + p54 + p72-His, as well as in group 4, 1 ml of each of the recombinant L. lactis MG1363/pMG36e-p30 + p54 + p72-LTB-His immunized rabbits, respectively. Rabbits were immunized on days 0 and 17. Two rounds of immunization were administered at 14-day intervals, each lasting 3 days. The experimental period was 34 days, and the details of the vaccination are described in Table 1.

The samples of the sera, the small intestine for the jejunal segment, and the spleens were collected by five rabbits per group randomly selected on 17th and 34th days post-vaccination (dpv; Figure 1). The blood supernatant was stored at-80°C to access the humoral immune response after centrifuging at 3,000 rpm for 15 min at ambient temperature. Using the same centrifugation methods, treated supernatants from a 5 cm jejunal segment were placed in 5 ml PBS (pH = 7.4) and incubated at 4°C for 2 h to assess mucosal immune response. The collected spleen was rinsed with PBS (pH = 7.4), put in a stainless steel mesh filter with a pore size of 0.075 mm, and ground with a syringe wick until it became white flocculent. The filter was then rinsed and filled with PBS, and the cell suspension was collected in a sterile centrifuge tube and centrifuged for 10 min at 1,500 rpm. The cells were rinsed once with RPMI-1640 culture media and again with PBS. The cells were then suspended in RPMI-1640 culture medium, which contains 10% fetal bovine serum and 100 U m/L of penicillin and streptomycin. Typan blue was then used to demonstrate the viability of the cells, and the cell survival ratio was >90%. After adjusting the cell density, the cells were inoculated at a density of 2 × 10⁶ cells m/L and incubated at 37°C in a 5% CO₂ incubator, and the samples were stored at –80°C for evaluating the spleen cell survival rate.

The levels of IgG in serum and sIgA in small intestinal contents were determined using the rabbit-derived ASFV p30-IgG, p54-IgG, and p72-IgG enzyme-linked immunosorbent assay (ELISA) kits and the rabbit-derived sIgA ELISA kit (Shanghai Yongwin Biotechnology Co., Ltd.), according to the manufacturer’s instructions. Briefly, the negative control well contained 50 μl, the positive control well contained 50 μl, and the sample wells contained 10 μl of the sample to be tested and 40 μl of diluent, which were added with 100 μl HRP-labeled detection antigen and incubated at 37°C for 1 h. Then washed five times with wash buffer. Subsequently, each well added 50 μl each of substrate A and substrate B which were incubated for 15 min at 37°C, avoiding light. Within 15 min, add 50 μl of termination solution to each well, and after that, the absorbance was read at 450 nm using an enzyme-labeling measuring instrument (Thermo Fisher Scientific, United States). Ultimately, the IgG or sIgA antibody levels of our samples were calculated based on the standard curve of the kits.

The levels of IL-4 and IFN-γ in serum were performed using ELISA kits according to the manufacturer’s recommendation (Shanghai Landon Biotechnology Co., Ltd.). The method was similar to that previously mentioned in 2.7. The concentrations of IL-4 and IFN-γ in serum were obtained from the standard curve of each ELISA plate.

The spleen cells from immunized rabbits were prepared for the lymphocyte survival rate assay using the CCK-8 Cell Viability Assay Kit (Shanghai Langdon Biotechnology Co., Ltd.) according to the manufacturer’s instructions. In brief, each well with 100 μl of splenocyte suspension was incubated in a 5% CO₂ incubator (SANYO, Japan) at 37°C for 24 h. Then, the plate was added 100 μl of complete medium, incubating for 48 h in a 37°C, 5% CO₂ incubator. After that, 10 μl of CCK-8 solution was added to each well and incubated for 3 h in a 37°C, 5% CO₂ incubator. Subsequently, the absorbance at 450 nm was measured by an enzyme-labeling measuring instrument (Thermo Fisher Scientific, United States). Finally, Cell viability (%) = (OD₄₅₀ of sample cells/OD₄₅₀ of control cells) × 100%.

The experimental data were analyzed by GraphPad Prism software (San Diego, CA, USA), and the significance of differences between groups was tested by One-way ANOVA.

As shown in Figures 2A–F, six new functional recombinant plasmids were successfully constructed. According to Figure 2G, the empty vector and six recombinant L. lactis were identified by PCR, and the target bands of recombinant L. lactis MG1363/pMG36e, MG1363/pMG36e-p30-His, MG1363/pMG36e-p54-His, MG1363/pMG36e-p72-His, MG1363/pMG36e-p30-LTB-His, MG1363/pMG36e-p54-LTB-His, MG1363/pMG36e-p72-LTB-His were 442 bp, 1,111 bp, 1,081 bp, 928 bp, 1,573 bp, 1,543 bp, and 1,390 bp, respectively, which were consistent with the expected target fragment sizes.

The expressions of the p30-His protein, p54-His protein, p72-His protein, p30-LTB-His fusion protein, p54-LTB-His fusion protein, and p72-LTB-His fusion protein were successfully detected by Western blot (with His-tag as the detection antigen) in Figures 3A–D. The p30-His and p54-His molecular weights were 24.5 kD and 23.4 kD, respectively (Figure 3A). The p30-LTB-His molecular weight was 41.4 kD (Figure 3B). The p54-LTB-His molecular weight was 40.3 kD (Figure 3C). The p72-His and p72-LTB-His molecular weights were 17.8 kD and 34.7 kD, respectively (Figure 3D). The results indicated that the proteins of interest were consistent with the expected size, proving that the recombinant L. lactis successfully expressed the foreign protein.

On 17 and 34 dpv, specific antibodies to p30, p54, and p72 were detected in the serum of rabbits. Tables 2–4 demonstrated that there were 4 positive samples in groups 3 and 4 on 17 dpv, with an 80% positive sample rate. On 34 dpv, however, there were 5 positive samples in groups 3 and 4, with a 100% positive rate for the samples. Additionally, on 17 and 34 dpv, neither group 1 nor group 2 had a positive rate. Although this test is qualitative, its methodology allows it to be semi-quantitative, and the size of the OD₄₅₀ value somewhat predicts the level of antibodies. The differences between groups 3 and 4 and groups 1 and 2 are highlighted in the data, suggesting that the recombinant bacteria can induce a humoral immune response.

The results were judged as follows: critical value = negative control hole OD₄₅₀ value +0.15, sample OD value < critical value is judged as negative; sample OD value > critical value is judged as positive.

The OD₄₅₀ of the negative control well of p30 IgG was 0.096, and its critical value OD₄₅₀ = 0.246.

The OD₄₅₀ of the negative control well of p54 IgG was 0.066, and its critical value OD₄₅₀ = 0.216.

The OD₄₅₀ of the negative control well of p72 IgG was 0.082, and its critical value OD₄₅₀ = 0.232.

The sIgA levels in the intestinal mucosa were measured by ELISA on 17 and 34 dpv. According to the findings in Figure 4, the results showed that the sIgA of groups 3 and 4 were significantly higher than that of groups 1 (p < 0.01) and 2 (p < 0.01), respectively. Group 4 differed slightly from group 3 (p < 0.05), with an increasing trend from days 17 to 34. These data suggested that oral administration of recombinant bacteria could stimulate the mucosal immune system in rabbits.

The levels of cytokines IL-4 and IFN-γ in serum were measured using ELISA on 17 and 34 dpv (Figures 5A,B). As shown in Figure 5A, the serum IL-4 concentrations of group 3 and group 4 on 17 and 34 dpv were significantly different than those of group 1 (p < 0.01), respectively. There was no significant difference (p > 0.05) in the level of IL-4 concentrations observed between groups 2 and 3 on 17 dpv, but they were significantly higher (p < 0.05) on 34 dpv. However, group 4 performed slightly better (p < 0.05) on 17 dpv and significantly better (p < 0.01) on 34 dpv than group 2. Although there was no difference (p > 0.05), there was an increasing trend between groups 3 and 4. These results demonstrated that vaccination with LTB adjuvant can produce significantly higher levels of Th2-associated cytokine IL-4 in rabbits.

The level of cytokine IFN-γ in serum was determined via ELISA on 17 and 34 dpv. The results, as shown in Figure 5B, indicated that on 17 dpv, there was a significant difference between groups 1 and 3 (p < 0.01), while no significant differences were found between group 1 and group 4 (p > 0.05), and compared to group 2, group 3 and group 4 also showed no difference (p > 0.05). Unexpectedly, the level of IFN-γ in group 3 was significantly higher than in group 4 (p < 0.05). Following the strength of immunization, the data on 34 dpv suggested that groups 3 and 4 were monumentally higher than groups 1 (p < 0.01) and 2 (p < 0.01). There was a significant difference between groups 3 and 4 (p < 0.05). The results showed that oral administration without the LTB adjuvant resulted in significantly higher levels of the Th1-associated cytokine IFN-γ.

The cell viability of splenocytes after the experiment was detected by the CCK-8 kit on 17 and 34 dpv. Only groups 2 and 4 (p < 0.05) showed significance in Figure 6 on 17 dpv. On 34 dpv, groups 3 and 4 were monumentally higher than groups 1 (p < 0.01) and 2 (p < 0.01), respectively, and there was a statistical difference between groups 3 and 4 (p < 0.01). These data demonstrated that immunization of rabbits with recombinant bacteria increased the percentage of splenocytes, indicating the activation of T cells in the rabbit model.