Fructose-1,6-bisphosphate aldolase encoded by a core gene of Mycoplasma hyopneumoniae contributes to host cell adhesion

Abstract

Mycoplasma hyopneumoniae is an important respiratory pathogen that causes great economic losses to the pig industry worldwide. Although some putative virulence factors have been reported, pathogenesis remains poorly understood. Herein, we evaluated the relative abundance of proteins in virulent 168 (F107) and attenuated 168L (F380) M. hyopneumoniae strains to identify virulence-associated factors by two-dimensional electrophoresis (2-DE). Seven proteins were found to be ≥ 1.5-fold more abundant in 168, and protein-protein interaction network analysis revealed that all seven interact with putative virulence factors. Unexpectedly, six of these virulence-associated proteins are encoded by core rather than accessory genomic elements. The most differentially abundant of the seven, fructose-1,6-bisphosphate aldolase (FBA), was successfully cloned, expressed and purified. Flow cytometry demonstrated the surface localisation of FBA, recombinant FBA (rFBA) mediated adhesion to swine tracheal epithelial cells (STEC), and anti-rFBA sera decreased adherence to STEC. Surface plasmon resonance showed that rFBA bound to fibronectin with a moderately strong K_D of 469 nM. The results demonstrate that core gene expression contributes to adhesion and virulence in M. hyopneumoniae, and FBA moonlights as an important adhesin, mediating binding to host cells via fibronectin.

Figure 1

Evaluation of virulence in Mycoplasma hyopneumoniae strains 168 and 168L. After treatment, piglets challenged with strain 168 showed symptoms of coughing, but no clinical signs of pneumonia were observed in the 168L or control groups. All pigs were alive during the entire experimental period. Lung lesion scores were subjected to statistical analysis after slaughter. The main lung lesion of the strain 168 challenged group is the pulmonary consolidation in the cranial and middle lobes of the lungs, and the lesions have obvious boundaries with the non-lesioned areas. Significant differences were observed between groups challenged with M. hyopneumoniae strain 168 and 168L (p < 0.05).

Figure 2

Identification of differentially abundant proteins by two-dimensional electrophoresis (2-DE). A Bacterial proteins from M. hyopneumoniae strain 168 cultured in KM2 medium. B Bacterial proteins from M. hyopneumoniae strain 168L cultured in KM2 medium. Yellow arrows on gel images indicate the seven protein spots listed in Table 1 increased in abundance by ≥ 1.5-fold in M. hyopneumoniae strain 168.

Figure 3

Interaction networks of the identified differentially abundant proteins and known putative virulence factors. Protein–protein interactions of differentially abundant proteins with a confidence score ≥ 0.4 are shown. Green nodes represent known putative virulence factors collected from published literature, and orange nodes represent differentially abundant proteins in M. hyopneumoniae strain 168 compared with strain 168L. Grey lines represent interactions between two nodes.

Figure 4

Predicted size of the pan-genome of M. hyopneumoniae. A Comparative overview of the pan-genome and core genome of M. hyopneumoniae. The M. hyopneumoniae pan-genome is shown in green (1152 genes), while the core genome is shown in orange (481 genes). Each plot point represents the mean value for gene clusters in the respective number of genomes, and curves represent power law fitting of the data. B Flower-plot schematic diagram of all nine M. hyopneumoniae strains for which genome data are available, showing the core genome size (flower centre) and the number of unique genes for each strain (flower petals). Numbers below the name of each strain indicate the total number of genes.

Figure 5

Western blot analysis of comparative proteomics data. A The left lane was loaded with bacterial proteins from M. hyopneumoniae strain 168. The right lane was loaded with bacterial proteins from strain 168L. The differentially abundant protein fructose-1,6-bisphosphate aldolase (FBA; 44 kDa) was analysed using the corresponding antibodies. Protein bands were visualised using Electro-Chemi-Luminescence (ECL) substrate. A Ponceau-S stained membrane was used as the loading control. B Image J software was used to calculate the optical density of the corresponding bands in the blots. The optical density of the corresponding bands was normalized to the total proteins of Ponceau-S staining of the same membrane. The level of abundance of FBA in M. hyopneumoniae strain 168L is expressed as the percentage of that in M. hyopneumoniae strain 168. The asterisk above the charts stands for statistically significant differences.

Figure 6

Detection of FBA on the surface of M. hyopneumoniae by flow cytometry. Negative control, M. hyopneumoniae strain 168 and 168L treated with preimmune serum; M. hyopneumoniae strain 168 and 168L: bacteria treated with anti-rFBA serum. The level of mean fluorescence intensity (MFI) of M. hyopneumoniae incubated with anti-rFBA sera is expressed as the percentage of the corresponding strain incubated with preimmune sera. The asterisks above the charts stand for statistically significant differences.

Figure 7

Role of rFBA in adhesion of M. hyopneumoniae to swine tracheal epithelial cells (STEC). Blue indicates STEC nuclei. Orange indicates A rFBA adhering to STEC membranes, and B BSA (negative control) adhering to STEC membranes. The left panel means protein adhered to the STEC membranes. The protein was labelled with tetraethyl rhodamine isothiocyanate (TRITC). The middle panel means cell nuclei of STEC stained with 6-diamidino-2-phenylindole (DAPI). The right panel means the merge of the left and middle panel. The white line indicates the scale.

Figure 8

Inhibition of M. hyopneumoniae adhesion to STEC. Adhesion rate = (number of bacteria recovered from cells incubated with anti-rFBA sera/number of bacteria recovered in the group incubated with preimmune sera) × 100. Data are expressed as mean ± SD of at least three experiments with samples performed in triplicate. The asterisks above the charts stand for statistically significant differences.

Figure 9

rFBA and fibronectin interaction analysis by Far-WB and Surface plasmon resonance (SPR) analysis. A Far-WB analysis of rFBA with fibronectin. The first lane: PVDF membrane with transferred rFBA protein incubated with anti- rFBA antibody as a positive control; the second lane: PVDF membrane with transferred rFBA protein incubated with fibronectin and anti-fibronectin antibody; the third lane: PVDF membrane with transferred BSA (negative control) incubated with fibronectin and anti-fibronectin antibody. Protein bands were visualized using ECL substrate. B Sensorgrams depict the binding of immobilised fibronectin to rFBA. Increasing concentrations of rFBA (5, 10, 25, 50 and 100 μg/mL) were injected at a flow rate of 30 μL/min for 180 s over immobilised fibronectin. The arrow indicates the end of the injection period, at which point dissociation of rFBA from fibronectin can be observed. RU resonance units.