Mhp182 (P102) binds fibronectin and contributes to the recruitment of plasmin(ogen) to the Mycoplasma hyopneumoniae cell surface

Proteomic and molecular analyses of P102 and P102 cleavage fragments

P102 is 904 amino acids in length and contains a signal peptide sequence (Adams et al., 2005). Analysis of the P102 sequence using the TmPred algorithm (http://www.ch.embnet.org/software/TMPRED_form.html) identified a single putative transmembrane domain between amino acids 11–34 (score 2095). Bioinformatic analyses failed to identify any known binding motifs within P102, but the region from amino acid 7–359 of P102 displays a level of similarity with the LppT protein from Mycoplasma conjunctivae (identity 24%, positive 44%). LppT contains a RGD motif involved in host cell adherence (Zimmermann et al., 2010).

Immunoblot analysis of a M. hyopneumoniae strain 232 whole-cell lysate probed with antisera raised to purified rP102 identified three proteins with masses of ∼100 kDa (P102 pre-protein), ∼72 kDa (P72) and ∼42 kDa (P42) (Djordjevic et al., 2004). The N-terminal cleavage fragment (P72) has not been identified previously by mass spectrometry (MS) and the cellular location of these cleavage fragments has not been determined in M. hyopneumoniae. To elucidate the location of P72 and P42, Triton X-114-fractionated proteins of M. hyopneumoniae were resolved by SDS-PAGE followed by characterization using liquid chromatography (LC)-MS/MS. Tryptic peptides that specifically aligned to the 60 kDa N-terminal and 42 kDa C-terminal portion of P102 were generated from proteins that resolved with masses of ∼55–70 kDa and ∼40–55 kDa respectively (Fig. 1A and B). As expected, peptide matches partitioned either side of the amino acid 556 cleavage site identified by Djordjevic et al. (2004). These data indicate that the original mass of 72 kDa given to the larger of the P102 cleavage fragments was over estimated and we have reassigned this fragment as P60. This is the first time P60 has been definitively characterized by LC-MS/MS, while the identity of P42 was reconfirmed by LC-MS/MS in this study. The band at ∼102 kDa (P102 pre-protein) appears inconsistently in immunoblot experiments and we were unable to confirm the presence of P102 by performing LC-MS/MS on proteins with masses of 90–150 kDa excised from an SDS-PAGE gel.

Cell surface location of P102, P60 and P42

Immunogold labelling experiments of M. hyopneumoniae cells derived from broth culture showed proteins recognized by anti-P102 antisera were not cell-associated but were found clustered in an extracellular matrix (Djordjevic et al., 2004). In similar experiments conducted on respiratory tissues from M. hyopneumoniae-infected pigs, anti-P102 antisera labelled with gold particles were observed within and on the cell surface of mycoplasmas and on the surface of cilia, with gold particles often clustered at high concentration at each of these sites (Adams et al., 2005). These data indicate that P102 or its cleavage fragments P60 and P42 are present on the surface of M. hyopneumoniae, but it is not clear precisely which of these proteins are present on the cell surface. In this study, enzymatic cell surface shaving experiments were used to determine if P102, P60 and P42 reside on the cell surface of M. hyopneumoniae. Freshly cultured, washed cells were exposed (5 min) to trypsin (50 µg ml⁻¹) and the shaved products were harvested for further analyses. LC-MS/MS of tryptic digests of these products identified nine unique peptides spanning amino acids 60–806 in P102 (Fig. 2A). To provide further evidence that P102 and/or its cleavage fragments P60 and P42 are present on the surface of M. hyopneumoniae, immunoblot studies of trypsin-shaved M. hyopneumoniae cells were performed (Fig. 2B and C). P60 and P42 were found to be digested by concentrations of trypsin of 50 µg ml⁻¹ and above, while L7/L12, a control cytosolic protein (Burnett et al., 2006), remained intact with treatment of trypsin at 300 µg ml⁻¹. These findings suggest both P60 and P42 are located on the cell surface.

Frequently a C-terminal lysine residue contributes to plasminogen binding sites (Plow et al., 1995; Sanderson-Smith et al., 2007). Two lysine residues are located at the C-terminus of P102 and this prompted our investigation of the potential of P102 to bind plasminogen. Ligand blots of the Triton X-114 detergent phase of the M. hyopneumoniae proteome detected plasminogen-binding proteins which correspond in size to P102 and P42 (Fig. 2D). Recombinant polyhistidine tagged proteins of P102 (rP102) and the known cleavage products P60 and P42 (rP60 and rP42) were expressed and purified (Fig. 2E and F) to enable further investigation of the potential role of P102 as a plasminogen-binding protein.

P102 domains bind plasminogen increasing the susceptibility of plasmin(ogen) to activation by tPA

To determine if P102 and its cleavage products bind plasminogen we utilized surface plasmon resonance to characterize real-time interactions of these molecules with plasminogen. rP102 bound plasminogen in a dose-dependant manner at physiologically relevant concentrations with a K_D of 76 ± 19 nM; k_a = (1.0 ± 1.0) ×  10⁵ M⁻¹ s⁻¹ (Fig. 3A–E). Deletion of the tandem C-terminal lysine residues from rP102 (rP102ΔKK) did not abolish plasminogen binding, and although the K_D did not change significantly (84 ± 8 nM), the level of binding was noticeably reduced. rP60 does not possess a C-terminal lysine, but an affinity for plasminogen with low binding levels was still observed [K_D 109 ± 47 nM; k_a = (4.8 ± 6.3) ×  10⁵ M⁻¹ s⁻¹]. The C-terminal P42 fragment still contains the two C-terminal lysine residues of P102. rP42 bound plasminogen in a dose-dependant manner at physiologically relevant concentrations [K_D 37 ± 15 nM; k_a = (7.4 ± 6.7) ×  10⁴ M⁻¹ s⁻¹]. Deletion of the C-terminal lysine residues from rP42 (rP42ΔKK) abolished the majority of observed binding, although binding was still physiologically relevant [K_D 88 ± 7 nM; k_a = (2.0 ± 1.7) ×  10⁵ M⁻¹ s⁻¹]. Binding of plasmin(ogen) by lysine residues alters the conformation of plasminogen rendering it more susceptible to activation (Lahteenmaki et al., 2005). C-terminal lysine residues appear to contribute to P102 binding plasminogen. To determine if binding by P102 increases the susceptibility of plasminogen to activation we investigated the proteolytic activity of P102-bound plasmin(ogen). Plasminogen was mixed with a tissue-specific plasminogen activator (tPA) following pre-incubation with rP102 and plasmin activity was determined with a plasmin-specific chromogenic substrate. Porcine plasmin(ogen) was more readily activated in the presence of rP102 compared with the tPA control (Fig. 3F) suggesting that plasminogen undergoes a conformation change upon binding by P102.

Activation of plasminogen bound by M. hyopneumoniae

Although M. hyopneumoniae binds plasminogen to its cell surface (Seymour et al., 2010), the potential for activation of this bound plasminogen has not been previously investigated. Activation of the bound zymogen to the serine protease plasmin is required for M. hyopneumoniae to utilize plasmin(ogen) as a potential mechanism for proteolysis of fibrin matrices, evasion of the innate immune system or to enable degradation and invasion of tissue barriers (Lähteenmäki et al., 2001; Degen et al., 2007). We have shown that activation of plasminogen by tPA is accelerated when plasminogen is pre-incubated with rP102. To determine whether M. hyopneumoniae surface-bound plasminogen can be converted to plasmin, M. hyopneumoniae cells were incubated with plasminogen and tPA and proteolytic activity was measured using a plasmin-specific chromogenic substrate. Human and porcine plasminogen were cleaved to form plasmin by human tPA (Fig. 4A). Porcine plasminogen was not activated as readily as human plasminogen, but it has previously been shown that human plasminogen is more readily activated than porcine plasminogen irrespective of the origin of the activator (Nieuwenhuizen and Keyser, 1985; Flight et al., 2006). No substrate cleavage was detected when tPA was excluded from the assay, indicating that M. hyopneumoniae does not produce its own activator. Minimal activation was seen when the bacteria were incubated with tPA only, which is most likely the result of their plasminogen recruitment (porcine and equine) from serum, since this is an essential component of the Friis medium routinely used for the culture of M. hyopneumoniae. Western blot analysis of M. hyopneumoniae whole-cell lysates with anti-porcine plasminogen antibodies illustrated that M. hyopneumoniae cells are able to recruit plasmin from Friis medium (Fig. 4B and C).

To demonstrate that M. hyopneumoniae surface-bound plasmin(ogen) possesses biological activity, the ability of plasminogen-coated M. hyopneumoniae cells to degrade fibrinogen in the presence and absence of tPA was examined. Fibrinogen degradation was illustrated by immunoblots probed with anti-fibrinogen antibodies (Fig. 4D). Plasminogen-coated M. hyopneumoniae cells were able to degrade full-length fibrinogen in the presence of tPA within 6 h. No fibrinogen degradation was observed in the absence of an activator and very little degradation occurred when tPA was added to M. hyopneumoniae cells incubated without porcine plasminogen. These data demonstrate that M. hyopneumoniae-bound plasminogen is accessible to activators and can be converted into the active enzyme plasmin.

Plasmin(ogen) is present in bronchoalveolar lavage fluid and lung tissue of both uninfected and M. hyopneumoniae-infected pigs

Previous studies with mice indicate that plasminogen is distributed throughout various tissues, including the lungs (Zhang et al., 2002; Swaisgood et al., 2007). To determine whether plasminogen is present in the airways of porcine lungs, a dot blot of bronchoalveolar lavage (BAL) fluid collected from both uninfected and M. hyopneumoniae-infected pigs was performed (Fig. 5A). Plasminogen was detected in all pigs pre-challenge, and in unchallenged pigs at day 18, indicating that plasminogen is present in the airways of healthy pigs. Plasminogen was also detected in M. hyopneumoniae-infected pigs at 18 days post challenge. Plasminogen levels in challenged and unchallenged pigs did not appear to vary. We also compared plasmin levels in the BAL fluid of pre-challenge and M. hyopneumoniae-infected pigs. In the majority of pigs plasmin activity was significantly higher in the BAL fluid following challenge with M. hyopneumoniae (Fig. 5B). These findings provides further evidence that M. hyopneumoniae interacts with the plasmin(ogen) system during colonization.

Sections of lung from unchallenged and M. hyopneumoniae-challenged pigs were subjected to immunohistochemical staining to determine the distribution of plasminogen in the tissue. Sections were initially stained with a M. hyopneumoniae-specific antibody to confirm that challenged pigs were infected with M. hyopneumoniae and that unchallenged pigs were uninfected. Samples from pigs 18 days post challenge with M. hyopneumoniae stained positive for this organism, while all other samples were negative (data not shown). All samples, from both infected and uninfected animals, stained positive for plasminogen along the ciliary borders of the bronchioles (Fig. 5C–F). As expected, staining was also observed in the blood vessels perfusing the lung tissue (data not shown). The distribution of staining in the bronchioles suggests that plasminogen is present in the extracellular fluid lining the airways. Staining in uninfected samples did not appear substantially different from that observed in infected samples, suggesting that plasminogen is available at the site of colonization both before and during M. hyopneumoniae infection.

P102 domains bind fibronectin

Although the function of P102 is unknown, the close genetic linkage of P102 to P97 in a single operon implies that this protein is involved in adherence to the host. Recently it has been discovered that M. hyopneumoniae is a fibronectin-binding pathogen with surface located fibronectin binding adhesins Mhp271, Mhp107 and P116 (Mhp108) (Deutscher et al., 2010; Seymour et al., 2010; Seymour et al., 2011). Using surface plasmon resonance, real-time interactions between rP102 and fibronectin were observed (Fig. 6A–C). rP102 binds fibronectin in a dose-dependant and physiologically relevant manner [K_D 26 ± 2 nM; k_a = (2.7 ± 7.8) × 10⁴ M⁻¹ s⁻¹]. The recombinant P102 cleavage fragment rP42 also bound fibronectin in a dose-dependant manner with a K_D of 33 ± 2 nM [corresponding k_a of (5.2 ± 16.1) × 10⁴ M⁻¹ s⁻¹]. rP60 also appeared to bind fibronectin but the rate constants of specific binding were not able to be calculated.

P102 domains promote binding to porcine kidney epithelial-like (PK15) cells

The observation that antibodies specific to P102 were localized to porcine cilia in M. hyopneumoniae-infected lungs (Adams et al., 2005) prompted us to investigate if recombinant P102 proteins promote adherence to PK15 cells. Zielinski et al. (1990) developed an in vitro model for adherence of M. hyopneumoniae using PK15 cells, since then this model has been used to elucidate adherence mechanisms employed by M. hyopneumoniae (Burnett et al., 2006; Wilton et al., 2009; Seymour et al., 2010; Seymour et al., 2011). Using PK15 cells as an in vitro model we found that rP102 and rP60 were able to promote adherence of microspheres to PK15 cells (Fig. 7).

Bacterial strains and culture conditions

Mycoplasma hyopneumoniae strains 232, J and Hillcrest (isolated in NSW, Australia) were grown in Friis medium and cultured as described previously (Bereiter et al., 1990; Scarman et al., 1997). Escherichia coli TOP10 and BL21 star (DE3) (Invitrogen, USA) were grown at 37°C on Luria–Bertani (LB) agar plates or cultured in LB medium with shaking at 200 r.p.m., supplemented with 100 µg ml⁻¹ ampicillin where appropriate.

Plasmid constructs, protein expression and purification

PfuUltra™ High-Fidelity II DNA Polymerase (Stratagene, USA) was used in polymerase chain reactions. Primer sequences are listed in Table S1. M. hyopneumoniae 232 chromosomal DNA and pMHP182 (provided by Prof. F. Chris Minion, Iowa State University, USA) were used as the template for amplification of mhp182. pMHP182 contains a truncated mhp182 gene with in frame TGA codons converted to TGG, as TGA encodes tryptophan in M. hyopneumoniae. The overlap extension mutagenesis method was used to amplify mhp182 to alter a remaining TGA codon to TGG (Ho et al., 1989). Subsequent cloning of mhp182 used the pET100/D-TOPO^® cloning kit (Invitrogen) according to the manufacturer's instructions. Segments of the pET100:mhp182 construct were amplified and cloned using the pET151/D-TOPO^® kit, resulting in the recombinant polyhistidine fusion proteins used in this study (rP102, rP102ΔKK, rP60, rP42 and rP42ΔKK). Nucleotide sequencing of all plasmids ensured conformity of the sequences to mhp182. Plasmid constructs were transformed into E. coli BL21 star (DE3) for expression of recombinant proteins; when cell growth reached an OD₆₀₀ of 0.8, protein expression was induced by the addition of 1 mM IPTG. Purification of proteins under native conditions used Ni-NTA agarose in accordance with the manufacturer's instructions (Qiagen, USA). Purification of rP42 and rP42ΔKK was performed under denaturing conditions as attempts to extract and purify the proteins under native conditions were unsuccessful. Purified proteins were dialysed into phosphate-buffered saline (PBS; 10 mM sodium phosphate, 137 mM sodium chloride, 2 mM potassium phosphate, 2.7 mM potassium chloride, pH 7.4). Polyclonal rabbit antisera were generated against recombinant proteins in New Zealand white rabbits using previously described methods (Jenkins et al., 2006).

Proteomics

A 0.1 g pellet of M. hyopneumoniae strain J cells was resuspended in 1 ml of solubilization buffer [7 M urea, 2 M thiourea, 40 mM Tris, 1% (w/v) C7Bz0] and disrupted by sonication at 10 W for 30 s. Proteins were reduced and alkylated with 20 mM acrylamide monomers, 5 mM tributylphosphine for 90 min. Proteins were precipitated in five volumes of ice-cold acetone for 30 min; the pellet was air dried and resuspended in SDS-PAGE sample buffer [15 mM Tris pH 6.8, 0.25% (w/v) SDS, 0.25% (v/v) β-mercaptoethanol, 2.5% (v/v) glycerol and 0.0025% (w/v) bromophenol blue]. Proteins were separated by SDS-PAGE and gel lanes cut into 16 equal slices for analysis. Following in-gel trypsin digestion, identification of proteins was achieved using LC-MS/MS as described previously (Deutscher et al., 2010).

To investigate the subcellular location of P102, P60 and P42, intact M. hyopneumoniae strain J cells were incubated in the presence and absence of trypsin and the supernatants were further digested and analysed by 2D-LC-MS/MS to detect peptides from proteins enzymatically released from the cell surface as previously described (Deutscher et al., 2010). Trypsin treatment of intact M. hyopneumoniae strain 232 cells was also performed and analysed with immunoblots using the method described by Wilton et al. (1998). Immunoblots were probed with pooled polyclonal rabbit sera raised against rP102 or against a cytosolic protein, L7/L12 (Burnett et al., 2006) as an intracellular control.

To perform immunoblots to investigate the ability of M. hyopneumoniae proteins to bind plasminogen, a 0.1 g pellet of M. hyopneumoniae cells was resuspended in 1 ml of 1% Triton buffer (1% Triton X-114, 10 mM Tris pH 8.0, 150 mM sodium chloride, 1 mM EDTA). Proteins were extracted as previously described (Wise and Kim, 1987). The protein pellet from the Triton X-114 detergent phase was resuspended in 0.5 ml MSS buffer [5 M Urea, 65 mM DTT, 2% 3-[(3-cholamidopropyl)-dimethylammonio]-1-propanesulfonate (CHAPS), 0.8% ampholytes, 40 mM Tris, 2 M thiourea, 2% sulfobetaine]. The detergent phase sample was sonicated at 10 W for 15 s, centrifuged and the supernatant collected. SDS-PAGE sample buffer was added to an aliquot of detergent phase sample and boiled for 5 min. The samples were separated by SDS-PAGE and visualized by staining with either Flamingo Fluorescent Gel Stain (Bio-Rad, USA) or Coomassie blue G-250. Alternatively protein was transferred to PVDF using a semi-dry transfer method as described previously (Kyhse-Andersen, 1984). Membranes were blocked with 5% (w/v) skim milk powder in PBS with 0.1% Tween 20 (v/v) (PBST) at room temperature for 1 h. Detection of P102 proteins was with polyclonal rabbit sera raised against purified rP60 and against rP42 diluted 1:100 in PBST for 1 h and a peroxidase-conjugated anti-rabbit IgG antibody, diluted 1:1000 for 1 h. Between incubations, the membrane was washed in three changes of PBST. The membrane was developed with SIGMAFAST™ 3,3′-diaminobenzidine tablets (Sigma-Aldrich, USA) as per manufacturer's instructions.

For detection of plasminogen-binding proteins, following blocking, the membrane was probed with purified biotinylated porcine plasminogen diluted 1:250 in PBST for 1 h. Porcine plasminogen was purified from porcine plasma as previously described (Andronicos et al., 1997; Sanderson-Smith et al., 2006), followed by gel filtration and ultrafiltration (Seymour et al., 2010). Spectrozyme PL (American Diagnostica, USA) was used to test the purity and activity of the purified plasminogen, while biotinylation was performed as described (Seymour et al., 2010). Membranes were then probed with ExtrAvidin^® peroxidase conjugate (Sigma-Aldrich) diluted 1:20 000 in PBST for 1 h. Between incubations, the membrane was washed in three changes of PBST. The membrane was developed with Amersham™ ECL Advance™ Western blotting detection kit (GE Healthcare, UK) and detected with a Chemi-doc XRS imaging system (Bio-Rad) for improved sensitivity of detection.

Surface plasmon resonance analyses

The ability of P102 protein fragments to bind plasminogen and fibronectin were investigated on a Biacore T100 instrument (Biacore AB, Sweden). The preparation of analyte proteins, ligand immobilization and kinetics assays have been described previously (Seymour et al., 2010). Plasminogen and fibronectin were immobilized as ligands for surface plasmon resonance assays; binding of recombinant P102 proteins was performed at 20°C with a concentration range of 0–200 nM. The 1:1 Langmuir binding model with BiaEvaluation software 3.1 (Biacore AB) was used to analyse plasminogen association kinetics. Fibronectin association kinetics were analysed with the sum of two exponentials for a heterogeneous surface (Deutscher et al., 2010).

Collection of porcine lung tissue and BAL fluid

Pigs at 9–10 weeks of age, sourced from a high health status herd, were anaesthetized and subjected to BAL using 4 ml of sterile PBS. The fluid was collected and stored at −80°C until required. At this time a number of pigs were inoculated intratracheally with 10 ml broth cultures of M. hyopneumoniae strain Hillcrest, while the remaining two pigs were not inoculated and served as uninfected controls. At 2.5 weeks post infection, all pigs were anaesthetized and subjected to BAL as described above. At 3 weeks post infection the pigs were euthanized and necropsies performed. Samples of lung tissue including any obvious lesions typical of M. hyopneumoniae infection, were collected into neutral-buffered formalin and fixed for 24 h.

Activation of plasmin(ogen) and plasmin activity in BAL fluid

Porcine plasminogen and rP102 in 50 mM Tris pH 7.4 were mixed in various molar ratios from 1:0.5 to 1:10 to investigate the effect of P102 on the activation of plasmin(ogen). Samples were incubated at 37°C for 1 h. Two hundred units of tPA (Calbiochem, Germany) in 50 mM Tris (pH 7.4) were added and the plate was incubated at 37°C for 15 min with shaking prior to the addition of 20 µl of 2.5 mM Spectrozyme PL. The absorbance at 405 nm was measured using a Multiskan Ascent plate reader (Thermo Labsystems, USA). Absorbance reads were taken every 15 min for up to 2 h 30 min. Between reads the plate was incubated at 37°C with shaking. Wells containing rP102 only or no plasminogen were employed as a negative control, while wells containing plasminogen and tPA only were used as a positive control. Absorbance measured in wells immediately after the addition of Spectrozyme PL was subtracted from the absorbance measured during subsequent readings.

To investigate the activation of M. hyopneumoniae surface-bound plasminogen, freshly cultured M. hyopneumoniae strain 232 cells were adhered to a Linbro/Titertek microtitre plate (MP Biomedicals, USA) and washed according to Seymour et al. (2010). Each well was blocked with 2% (w/v) BSA in PBS for 1 h. Five micrograms human (Sigma-Aldrich, USA) or porcine plasminogen in 1% (w/v) BSA in PBS was added to each well and incubated at 37°C for 1 h with shaking. Unbound plasminogen was removed by washing the plate three times with PBS containing 0.5% (v/v) Tween 20 and once with 50 mM Tris (pH 7.4). tPA was added and incubated as described for the activation of plasminogen in the presence of P102. Absorbance readings were taken every hour for the first 5 h, then again from 20 h until the absorbance for wells containing human plasminogen reached saturation. The average of the maximum absorbance for human plasminogen (time > 5 h) was set to 100% plasmin activity. Absorbance measured in wells without cells was subtracted from the absorbance of wells with cells. Wells containing cells incubated without plasminogen were employed as a negative control.

Bronchoalveolar lavage fluid from pre-challenge and M. hyopneumoniae-challenged pigs was tested for plasmin activity. Pre-warmed BAL fluid was incubated with 20 µl of 2.5 mM Spectrozyme PL at 37°C. Absorbance readings at 405 nm were taken over 22 h. Wells containing 1 µg of porcine plasminogen were employed as a negative control, while wells containing 1 µg of plasminogen and 200 units of tPA only were used as a positive control. Data analysis was performed using GraphPad Prism version 5.04.

M. hyopneumoniae cells recruit plasminogen from serum in Friis media

To assess whether M. hyopneumoniae cells recruit plasmin(ogen) from the serum in Friis media, a Western blot was performed using anti-porcine plasminogen antibodies (AbD Serotec, USA). Non-tPA-activated and tPA-activated porcine plasminogen were included as references. Porcine plasminogen (10 µg) was incubated in the presence or absence of 250 units of tPA in PBS for 23.5 h at 37°C. SDS-PAGE sample buffer was added to each sample and to PBS washed cultured M. hyopneumoniae strain 232 cells. All samples were boiled for 5 min prior to loading onto an SDS-PAGE gel. Proteins were either stained with GelCode Blue stain reagent (Pierce Biotechnology, USA) or transferred onto PVDF membrane. The membrane was blocked with 5% (w/v) non-fat dry milk in Tris-buffered saline (TBS) (10 mM Tris, 150 mM sodium chloride, pH 7.4) at room temperature for 1 h. Detection of plasmin(ogen) was with anti-porcine plasminogen antibodies diluted 1:2000 in TBS for 1.5 h and a peroxidase-conjugated sheep anti-rabbit IgG antibody (Chemicon, Australia) diluted 1:1000 in TBS for 1 h. Between incubations the membrane was washed with TBS containing 0.05% (v/v) Tween 20 twice for 5 min, followed by one 5 min wash in TBS. After equilibration with 100 mM Tris (pH 7.6) the membrane was developed with DAB substrate (0.05% 3,3′-diaminobenzidene, 0.03% hydrogen peroxide in 100 mM Tris, pH 7.6).

M. hyopneumoniae surface-bound plasmin(ogen) degradation of fibrinogen

Mycoplasma hyopneumoniae strain 232 cells were grown to log phase, washed and resuspended to an OD₆₀₀ of 0.9. M. hyopneumoniae cells (50 µl) were pre-incubated with 10 µg of porcine plasminogen in PBS for 30 min at 37°C. The cells were then washed three times with PBS prior to incubation with 5 µg of plasminogen-depleted human fibrinogen (Calbiochem) for 30 min at 37°C. Degradation of fibrinogen occurred in the presence of 225 units of tPA at 37°C. An aliquot of the sample was taken, the bacteria pelleted by centrifugation and the supernatant solubilized by boiling in SDS-PAGE sample buffer for 5 min. The samples were separated by SDS-PAGE, followed by transfer of the proteins onto PVDF membrane. The membrane was blocked with 5% (w/v) non-fat dry milk in TBS at room temperature for 1 h. Detection of fibrinogen peptides was with goat anti-human fibrinogen antiserum (Sigma-Aldrich) diluted 1:10 000 in TBS for 1.5 h and a peroxidase-conjugated anti-goat IgG antibody (Sigma-Aldrich) diluted 1:7500 in TBS for 1 h. Between incubations the membrane was washed with TBS containing 0.05% (v/v) Tween 20 twice for 5 min, followed by one 5 min wash in TBS. After equilibration with 100 mM Tris (pH 7.6) the membrane was developed with DAB substrate.

Anti-plasminogen dot blot

Bronchoalveolar lavage fluid collected from both unchallenged and M. hyopneumoniae-challenged pigs at time zero and 2.5 weeks post infection were thawed and diluted 1:20, 1:100 and 1:500 in PBS. Samples were then spotted in 100 µl volumes onto a pre-wetted PVDF membrane using a dot blot manifold. The membrane was blocked for 1 h in 2% BSA in PBS and then reacted with an anti-porcine plasminogen antibody diluted 1:500 in PBS. The membrane was washed 3 × 5 min in PBST and then incubated for 1 h in a 1:1500 dilution of horseradish peroxidase-conjugated sheep anti-rabbit antibody. The membrane was washed 3 × 5 min with PBST, equilibrated in 100 mM Tris (pH 7.6) and detected with DAB substrate.

Immunohistochemistry

Formalin-fixed tissues were placed into histology cassettes and embedded in paraffin. Sections of 4 µm thickness were cut from the paraffin blocks and mounted onto Superfrost slides. Sections were processed for immunohistochemistry by dewaxing the sections 2 × 3 min in xylene, rehydrating in an ethanol series and blocking with 1.8% peroxide in methanol for 10 min. Antigens were unmasked using 0.2% trypsin-0.1% CaCl₂ at 37°C for 30 min. Immunohistochemistry was then performed using the Vectastain ABC system (Vector Labs, USA) according to the manufacturer's instructions. To stain the sections for plasminogen or M. hyopneumoniae, slides were treated with a 1:500 dilution of anti-porcine plasminogen, or a 1:1000 dilution of M. hyopneumoniae-specific antiserum (Jenkins et al., 2006) respectively. Negative control slides were treated with PBS instead of primary antibody. Bound antibodies were detected with a chromagen solution containing 3-amino-9-ethylcarbazole and 0.015% hydrogen peroxide in dimethylformamide. Slides were then counterstained with haematoxylin and mounted with Faramount aqueous mounting medium (Dako, Denmark).

PK15 adherence assay

Protein-coated microspheres were used to investigate the ability of P102 proteins to promote adherence to PK15 cells. The assay was performed as described by Seymour et al. (2010) and Burnett et al. (2006). Briefly, fluorescent microspheres were coated with recombinant P102 proteins and incubated with PK15 monolayers for 1, 4 or 7 h. Monolayers were washed to remove unbound cells before fixing. Fixed PK15 cells were stained with a 1:200 dilution of Vybrant® Dil cell-labelling solution (Invitrogen) and imaged using a Nikon A1 Confocal Microscope. BSA coated microspheres were employed as a negative control and F1_P116 coated microspheres as a positive control. F1_P116 is a M. hyopneumoniae adhesin (Seymour et al., 2010).