Mycoplasma hyorhinis strains HEF-16 and JS-15 were isolated from pigs showing typical clinical signs of M. hyorhinis infection and were able to reproduce the disease in challenge test (Supplementary Figure S1). M. hyorhinis strain HUB-1, kindly provided by Prof. Shaobo Xiao from Huazhong Agricultural University, China, was unable to induce any disease in a challenge test (Supplementary Figure S1). Pig kidney cell line PK-15 (ATCC, Manassas, VA, United States, CCL-33) and human airway epithelial cell line NCI-H292 (ATCC, Manassas, VA, United States, CRL-1848) were cultured in Dulbecco’s modified Eagle medium (DMEM) and RPMI 1640 (Thermofisher Scientific, Waltham, MA, United States), respectively, supplemented with 10% fetal bovine serum (FBS). The cells were grown at 37°C in humidified air with provision of 5% CO₂.

The gene encoding M. hyorhinis DnaK (NCBI accession number: CP002170) was synthesized by GenScript Biotech Corp, Nanjing, China. The sequence was optimized with E. coli-preferred codons, and two TGA codons (coding Trp in mycoplasmas) were mutated into TGG codons. The gene was inserted into the pET-32a (+) vector (GenScript, Nanjing, China) between EcoRI and HindIII restriction sites to generate recombinant plasmid pET-32a-rMhr-DnaK that express a Trx-fusion protein. The plasmid was transformed into the E. coli strain BL21 (DE3). The bacterial cells were cultured in Luria–Bertani (LB) medium at 37°C. When the OD_600nm values of the bacterial cultures reached approximately 0.8, isopropyl-beta-D-thiogalactopyranoside (IPTG) at a final concentration of 1 mM was added to induce protein expression and incubated at 18°C overnight. The bacterial cells were collected by centrifugation at 12,000 × g for 10 min and resuspended in Tris–HCl buffer (pH = 8.0) with 300 mM NaCl and 20 mM imidazole. Then, the cells were lysed by sonication at 4°C, and the cell pellets were removed by centrifugation at 10,000 × g for 30 min. Supernatants was incubated with Ni Sepharose 6 FF resin for 1 h at 4°C, and the target protein was eluted with 150 mM imidazole. Purified protein was confirmed by SDS-PAGE and named as rMhr-DnaK.

Polyclonal antibody against the recombinant protein rMhr-DnaK was obtained by immunizing 4-week-old New Zealand white rabbits. Purified rMhr-DnaK was emulsified with complete (the first immunization) and incomplete (the second and third immunization) Freund’s adjuvant (1:1, v/v). Rabbits were subcutaneously immunized with 1 mg of recombinant protein three times at 2-week intervals. Antisera were collected 2 weeks after the third immunization and antibody titer was detected by ELISA. The blood was collected from heart when the titer was higher than 1,00,000, and antisera was extracted and store at −20°C until use.

The ability of the prepared polyclonal antibody to recognize DnaK from M. hyorhinis and rMhr-DnaK was determined by Western blotting. The whole cell lysate of M. hyorhinis and the purified rMhr-DnaK protein were subjected to 12% SDS-PAGE and transferred to a polyvinylidene fluoride (PVDF) membrane. After blocking with 5% skim milk in Tris-buffered saline with Tween 20 (TBST) buffer for 1 h at 37°C, the membrane was incubated with the anti-DnaK serum (1:1,000 dilution) overnight at 4°C, followed by horseradish peroxidase (HRP)-conjugated goat anti-rabbit IgG (1:10,000 dilution; Boster, China) for 1 h at 37°C. Finally, filters were developed with Electro-Chemi-Luminescence (ECL) substrate using a ChemiDoc XRS + system (Bio-Rad, Hercules, CA, United States). Preimmune serum was used as a negative control instead of anti-DnaK serum.

The surface localization of DnaK in M. hyorhinis (strain HEF-16) was first investigated by colony blotting with DnaK-specific antisera under conditions that would not damage the cell membrane of M. hyorhinis (Supplementary Table S1). PVDF membranes were gently placed on mycoplasma colonies on the surface of agar plates. After 5 min, filters were removed, blocked for 1 h at 37°C with Tris-buffered saline (TBS) containing 5% skim milk, and incubated overnight at 4°C in TBS containing 5% skim milk and anti-DnaK serum (1:1,000 dilution). Filters were washed four times with TBS with an interval of 15 min and treated with HRP-conjugated goat anti-rabbit IgG (1:10,000 dilution; Boster, China) for 1 h at 37°C. Finally, filters were developed with ECL substrate using the ChemiDoc XRS + system. Preimmune serum was used as a negative control instead of anti-DnaK serum. Antiserum against GAPDH whose surface expression has been identified previously (Wang et al., 2021) was used as a positive control.

Subsequently, flow cytometry (FACS) analysis was used to confirm the surface localization of DnaK and analyze the difference between strains with different virulence. In brief, M. hyorhinis cultures of strains HEF-16, JS-15, and HUB-1 (10⁸ color change units (CCU)/mL) were centrifuged at 15,000 × g for 20 min at 4°C, respectively. After blocking with phosphate-buffered saline (PBS) containing 1% bovine serum albumin (BSA), M. hyorhinis cells in the precipitate were collected and incubated with anti-DnaK serum or anti-GAPDH serum at a 1:100 dilution in PBS for 1 h at 37°C. M. hyorhinis incubated with preimmune rabbit serum was used as negative control. After washing with PBS, M. hyorhinis cells were stained with fluorescein isothiocyanate (FITC)-conjugated goat anti-rabbit IgG at a 1:500 dilution (Boster, China) for 1 h at 37°C. The fluorescence intensity was detected using a flow cytometer (BD Accuri^® C6). The mean fluorescence intensity (MFI) of M. hyorhinis incubated with anti-DnaK serum or anti-GAPDH serum was expressed as the percentage of that of M. hyorhinis incubated with preimmune serum.

Mycoplasma hyorhinis cells (1 × 10⁷ CCU/mL) were washed three times with PBS and pre-incubated with anti-DnaK serum or preimmune serum (1:20 dilution) at 37°C for 30 min. Bacteria suspended in cell medium were added to 24-well cell plates containing confluent PK-15 or NCI-H292 cells and incubated at 37°C for 6 h. After washing with PBS to remove non-adherent mycoplasmas, the cells were digested with 0.25% trypsin at 37°C for 5 min, and the amount of mycoplasma were determined by quantitative real-time PCR (Fourour et al., 2018).

An indirect immunofluorescence assay was used to determine whether the rMhr-DnaK protein adhered to cell surface. The 1 × 10⁵/mL cells were propagated in a 96-well cell culture dish for 24 h. The original medium was removed, and 100 μL of rMhr-DnaK in medium (200 μg/mL) was added and incubated at 37°C for 2 h. Cells incubated with the recombinant protein purified by nickel column affinity chromatography from E. coli containing the empty pET-32a (+) vector were used as negative control. Cells incubated with the recombinant rMhr-GAPDH protein were used as positive control. The unbound proteins were washed with PBS. Then, the cells were fixed with cold ethanol for 30 min at 4°C and blocked with PBS containing 5% BSA for 1 h at 37°C. The adherence was assessed by rabbit anti-DnaK serum (1:500 dilution) or rabbit anti-GAPDH serum (1:500 dilution) as primary antibodies and FITC-labeled goat anti-rabbit IgG (1:500 dilution; Boster, China) as secondary antibody. Finally, cell nuclei were stained with DAPI, and the immunofluorescence was detected using a fluorescence microscope (Zeiss, Jena, Germany).

The cell membrane proteins of PK-15 and NCI-H292 cells were prepared by a commercial Membrane and Cytosol Protein Extraction Kit according to the manufacturers’ instructions (Tiangen Biotech, China). The ability of rMhr-DnaK to bind cell membrane proteins was quantitatively determined by a microtiter plate adhesion assay (MPAA) (Xiong et al., 2016). In brief, a 96-well ELISA plate was coated with 100 μL cell membrane proteins (10 μg/mL) overnight at 4°C. After blocking with 5% BSA, the plate was incubated for 2 h at 37°C with 100 μL of rMhr-DnaK solution at different concentrations (ranging from 1.56 to 100 μg/mL) or PBS. Unbound proteins were removed by washing with PBS Tween (PBST), and the adherence was evaluated by adding 100 μL of mouse anti-Trx-tag monoclonal antibody (1:1,000 dilution; Sangon, China) followed by 100 μL of HRP-conjugated goat anti-mouse IgG (1:10,000 dilution; Boster, China). After washing, the substrate containing 3,3′,5,5′-tetramethylbenzidine and H₂O₂-urea was added, and the plates were incubated at 37°C for 15 min. Then, 2 M H₂SO₄ was added to stop the reaction, and the optical density (OD) of the solution was measured at 450 nm. For the adherence inhibition assay, 50 μg/mL of rMhr-DnaK was mixed with anti-DnaK or preimmune serum at various dilutions (1:10, 1:25, and 1:50) and added into the microtiter plate.

ELISA plate was coated with 100 μL of 3 μg/mL fibronectin (Sigma-Aldrich, Burlington, MA, United States, 10838039001), collagen type IV (Sigma-Aldrich, Burlington, MA, United States, C7521), laminin (Sigma-Aldrich, Burlington, MA, United States, L2020), or vitronectin (Sigma-Aldrich, Burlington, MA, United States, 5051) overnight at 4°C. After blocking with 5% BSA, 100 μL of rMhr-DnaK or BSA solutions at different concentrations (ranging from 1.56 to 100 μg/mL) or PBS was added. After washing with PBST, the adherence was evaluated by adding 100 μL of mouse anti-Trx-tag monoclonal antibody (1:1,000 dilution; Sangon, China) followed by 100 μL of HRP-conjugated goat anti-mouse IgG (1:10,000 dilution; Boster, China). After washing, the substrate was added, and the absorbance was measured at 450 nm.

The ability of rMhr-DnaK to bind plasminogen was first detected. An ELISA plate was coated with 100 μL of 3 μg/mL plasminogen (Sigma-Aldrich, Burlington, MA, United States, SRP6518). After blocking, 100 μL of rMhr-DnaK or BSA solutions at different concentrations (ranging from 1.56 to 100 μg/mL) or PBS was added. The binding of rMhr-DnaK to plasminogen was determined using the method described in the section “Binding of rMhr-DnaK to Cell Membrane Proteins.”

To determine the function of lysine residues of DnaK in its interaction with plasminogen, the lysine analog ε-aminocaproic acid (ε-ACA, Sigma-Aldrich, Burlington, MA, United States) was added as a competitive inhibitor. ELISA plates were coated with 100 μL rMhr-DnaK solution (30 μg/mL) overnight at 4°C. After blocking, the plate was incubated for 2 h at 37°C with 100 μL of 5 μg/mL plasminogen in the presence or absence of increasing concentrations (100, 200, 400, 800, and 1,600 mM) of ε-ACA. After washing, bound plasminogen was detected by adding rabbit anti-plasminogen polyclonal antibody (1:2,000 dilution; Boster, China) followed by HRP-conjugated goat anti-rabbit IgG (1:10,000 dilution). After washing, the substrate was added, and the absorbance was measured at 450 nm.

The effect of rMhr-DnaK binding on the activation of plasminogen was assessed. An ELISA plate was coated with 100 μL of 10 μg/mL rMhr-DnaK overnight at 4°C. After blocking with 5% BSA, the plate was incubated for 2 h at 37°C with 100 μL of 5 μg/mL plasminogen. After washing with PBS, 100 μL of tissue-type plasminogen activator (tPA; 200 ng/mL, Sigma-Aldrich, Burlington, MA, United States) was added, and the plate was incubated for 2 h at 37°C. After washing, 100 μL of the plasmin-specific substrate D-valyl-leucyl-lysine-p-nitroanilide dihydrochloride (Sigma-Aldrich, Burlington, MA, United States) was added at a final concentration of 0.4 mM. Plates were incubated overnight at 37°C, and absorbance was detected at 405 nm. BSA-coated wells were used as the negative control. Wells without tPA were set to determine whether the recombinant protein activated the plasminogen by itself.

The activation was further determined under a solution state. rMhr-DnaK (20 μg/mL) and plasminogen (20 μg/mL) were mixed and incubated at 37°C for 1 h and then added into a flat-bottomed 96-well ELISA plate. Then, tPA was added to a concentration of 500 ng/mL. After 15 min incubation, 0.4 mM of substrate was added, and the plate were incubated at 37°C. OD_405nm was measured every 15 min for 120 min. Wells containing only plasminogen with tPA or rMhr-DnaK with plasminogen or rMhr-DnaK with tPA were used as controls.

Fresh M. hyorhinis cultures were inactivated with 0.01% formaldehyde at 37°C for 24 h and mixed with Tween-80 (4%, v/v). Vaccines were made by emulsifying the aqueous phase with the Marcol white mineral oil adjuvant at 10:25 (v/v). Six-week-old Bama miniature pigs originally developed from caesarian-derived colostrum-deprived (CDCD) pigs were obtained from a pig farm. Six pigs, confirmed to be free from colonization by M. hyorhinis by nested PCR on nasal swabs, were randomly separated into two groups. Animals in group 1 were intramuscularly inoculated with 2 mL of M. hyorhinis inactivated vaccine (10⁹ CCU/mL), and the inoculation was repeated 14 days later. Animals in group 2 were not immunized. Serum samples were collected every 14 days until 42 days after the first immunization and assayed for antibody against DnaK. In brief, ELISA plates were coated with the rMhr-DnaK protein (10 μg/mL) overnight at 4°C. After blocking with 5% BSA, each well was incubated for 30 min at 37°C with 100 μL of serum sample diluted in PBS containing 5% BSA at 1:100, followed by 100 μL HRP-conjugated goat anti-swine IgG (Bethyl Laboratories, Montgomery, United States) at a dilution of 1:10,000. After washing, the substrate was added, and the absorbance was measured at 450 nm.

Data are expressed as mean ± SD. Statistical analysis was performed using GraphPad Prism software (Version 8.2.1). Independent samples t-test was used to analyze the data difference of FACS between specific antiserum and preimmune serum, the data in antibody-mediated adherence inhibition assay, the data difference between rMhr-DnaK and BSA in ECM and plasminogen binding assay. Repeated measures ANOVA was used to analyze the dynamic changes of plasminogen activation and anti-DnaK antibody production. The statistical analysis on other data was performed using one-way analysis of variance (ANOVA) followed by Dunnett’s multiple comparisons test. P-values < 0.05 were considered statistically significant.

The Full-length DnaK gene was designed according to the amino acid sequence of DnaK of M. hyorhinis strain HUB-1 and optimized according to E. coli codon usage. The synthetic gene was inserted into the expression vector pET-32a (+) (Figure 1A) and transformed into E. coli. The correct sequence was verified by DNA sequencing, and the recombinant engineering strain was obtained and named as pET-32a-rMhr-DnaK. IPTG was added to the E. coli cultures to induce the expression of recombinant protein. The SDS-PAGE results showed a distinct band between 70 and 100 kDa (Figure 1B, Lane 2), which was consistent with the theoretical molecular weight (79.8 kDa) of the recombinant protein rMhr-DnaK. After cell lysis, the recombinant protein in the supernatant was purified by nickel column affinity chromatography (Figure 1B, Lane 3). Rabbit anti-DnaK polyclonal antibody was prepared by immunizing the purified rMhr-DnaK protein. The serum was collected after three immunizations, and the antibody titer was detected by ELISA to be 1:128,000. The antibody could specifically react to the DnaK from M. hyorhinis and purified rMhr-DnaK protein (Figure 1C).

Surface localization of DnaK in M. hyorhinis was investigated by detecting the binding of anti-DnaK antibody to bacterial surface under conditions that did not damage the cell membrane of M. hyorhinis. First, colony blotting was carried out. As a result, significant signals were demonstrated after incubation of the PVDF membranes with the anti-DnaK serum or anti-GAPDH serum but not after its incubation with preimmune serum (Figure 2A). Thereafter, the surface expression of DnaK was further examined by FACS analysis. As shown in Figure 2B, significantly higher MFI of M. hyorhinis incubated with anti-DnaK serum or anti-GAPDH serum were observed than that of M. hyorhinis incubated with the preimmune serum (P < 0.01). Three strains of M. hyorhinis with different virulence were compared for their surface expression of DnaK. As shown in Figure 3, the virulent strains HEF-16 and JS-15 showed higher MFI than the strain HUB-1 which lacked the ability to cause clinical disease (P < 0.01).

Antibody inhibition assay was performed to investigate the function of surface-exposed DnaK in cytoadhesion of M. hyorhinis. The number of M. hyorhinis attached to swine-origin PK-15 cells (Figure 4A) and human-origin NCI-H292 cells (Figure 4B) were greatly reduced after being incubated with anti-DnaK serum, compared with those of the M. hyorhinis incubated with preimmune serum (P < 0.01). It was indicated that surface-localized DnaK plays a role in the adherence of M. hyorhinis to host cells.

The role of DnaK as M. hyorhinis adhesin was further investigated by determining its ability to adhere to host cells. The rMhr-DnaK protein was added to the cultures of PK-15 and NCI-H292 cells, and the adhered recombinant protein was detected after 2 h incubation by indirect immunofluorescence method. The results are shown in Figure 5. Distinct green fluorescence signals were observed in the wells of both kinds of cells incubated with rMhr-DnaK or rMhr-GAPDH, which was distributed in the whole cell-covered area. No green fluorescence signal was observed in the intercellular space. The wells with control protein had no positive fluorescence signal. Furthermore, the binding of rMhr-DnaK to cell membrane proteins was quantitatively detected by MPAA. The results showed that rMhr-DnaK significantly bound to the membrane proteins of both the PK-15 and NCI-H292 cells (P < 0.01). With the increase in rMhr-DnaK concentration, the OD value gradually increased in a dose-dependent manner (Figures 6A,B). The binding was significantly inhibited by anti-DnaK serum, compared to the preimmune serum (Figures 6C,D).

The binding of rMhr-DnaK to extracellular matrix (ECM) was assessed by detecting the adherence of rMhr-DnaK to the 96-well ELISA plate coated with different ECM components. As shown in Figure 7, the recombinant protein significantly bound to all the four different ECM components, fibronectin, type IV collagen, laminin and vitronectin, in a dose-dependent manner (P < 0.01, compared with BSA).

The ability of rMhr-DnaK to bind plasminogen was revealed by the microtiter plate test (P < 0.01, Figure 8A). The binding was significantly down-regulated by adding a lysine analog, ε-ACA (P < 0.01, Figure 8B). With the increase of the concentration of ε-ACA, the amount of plasminogen bound to rMhr-DnaK decreased gradually. The ability of rMhr-DnaK to capture plasminogen from normal pig plasma was demonstrated in Supplementary Figure S2. Subsequently, we determined whether the rMhr-DnaK-bound plasminogen was activated by the host agonist to form plasmin. tPA was added to activate the rMhr-DnaK-bound plasminogen. Thereafter, a plasmin-specific chromogenic substrate was added to detect their hydrolytic activity. As shown in Figure 8C, an increased OD_405nm was detected in the wells coated with rMhr-DnaK with the addition of plasminogen and tPA (P < 0.01) but not in the BSA-coated control wells. No activation was observed in the rMhr-DnaK-coated wells without tPA or plasminogen. To further determine whether the binding of rMhr-DnaK proteins increases the susceptibility of plasminogen to be activated by tPA, the activation of plasminogen was measured in a solution with or without rMhr-DnaK protein. As shown in Figure 8D, plasminogen was activated more rapidly in the wells containing rMhr-DnaK, plasminogen, and tPA than in wells that contained only plasminogen and tPA (P < 0.01). No activation was observed in the absence of tPA or plasminogen.

To assess the ability of DnaK of M. hyorhinis to induce a specific immune reaction, the DnaK-specific antibody in the serum of pigs inoculated with inactivated M. hyorhinis vaccine was determined using ELISA. As shown in Figure 9, the production of serum IgG antibody against DnaK was detected at 14 days after the second immunization, and it increased thereafter (P < 0.05). No antibody response was detected in the control group.