Mycoplasma hyopneumoniae (M. hyopneumoniae) ES-2 strain is a China field isolate from the lung with gross lesions of Enshi Black pig. And the lung was from the slaughterhouse. M. hyopneumoniae strain ES-2 was isolated by using Friis liquid medium modified Friis liquid medium and modified Friis solid medium as previously described (19).

To improve the isolation rate of M. hyopneumoniae, the lung is required to be harvested from 3–4-month-old, chronic coughing, and skinny pigs with obvious lesions of M. hyopneumoniae pneumonia in the apical and cardiac lobes of the lung. The lungs harvested from the diseased pig were immediately put into the clean bench. A total of 9 lung tissue samples from the different sites of the same lung were placed in a centrifuge tube with sterile phosphate-buffered saline (PBS), and then cut the lung tissue in the centrifuge tube into small pieces with scissors and grind it with a tissue grinder. Subsequently, the supernatants were collected by centrifugation at 5,000 rpm for 10 min, and bacteria in the supernatants were removed by 0.45 μm filtration. The supernatants were transferred into a fresh Friis medium. The mixtures were incubated in 5% CO₂ at 37°C for 6 days. On day 6 of culture, M. hyopneumoniae in Friis medium was identified by PCR amplifying the highly conserved 16S rRNA region from M. hyopneumoniae (19). The samples identified correctly by PCR were continued to subculture. If PCR identification of the culture at passage 4 was still correct, the positive samples were placed on the modified Friis solid medium. After culture for 9 days, a single colony on the solid medium was cultured in the modified Friis liquid medium. After three times of purification on the modified Friis solid medium, the purified colony was identified by PCR and sequencing.

Minimal inhibitory concentrations (MICs) of kanamycin for planktonic cultures of M. hyopneumoniae ES-2 were determined using a broth microdilution method previously described (35).

For the purpose of the visualization of M. hyopneumoniae ES-2 strain, scanning electron microscopy (SEM) and transmission electron microscopy (TEM) were performed as previously described with some modifications (36). The insoluble impurities in the fresh-modified Friis medium were removed by centrifugation at 10,000 rpm for 20 min and then centrifuged fresh-modified Friis medium was used to culture the ES-2 strain. Two days after culture, 100 ml of cultured ES-2 was centrifuged at 12,000 rpm for 15 min at 4°C, and the pellets were washed three times with PBS. The pellets were fixed with 2.5% glutaraldehyde in PBS at 4°C for 2 h. And then samples were sent to Servicebio for further processing. ES-2 was visualized using an HT7700 TEM and a SU8010 SEM. The adobe photoshop CC was used to determine the diameter of ES-2.

More than 20 continuous passages of M. hyopneumoniae ES-2 were performed. And the colonies of the ES-2 strain were cultured to confirm that there are no other microorganisms except M. hyopneumoniae ES-2 strain in the medium. And then ES-2 was inoculated into fresh-modified Friis medium. The mixture was cultured in 5% CO₂ at 37°C for 6 days. Meanwhile, the growth of ES-2 was examined every 12 h, between 0 and 144 h with the metabolic activity recorded by the color-changing units (CCU) which was indicated by the phenol red as previously described (37).

To observe the growth characteristic of ES-2 strain on a solid culture medium, the modified Friis solid medium was prepared as previously described (19). ES-2 at the exponential growth phase with five-fold serial dilutions in fresh-modified Friis medium was spotted (200 ml) onto the modified Friis solid medium. After ES-2 was cultured in 5% CO₂ at 37°C for 9 days, ES-2 strain colonies were observed by Inverted Microscope (Olympus, Japan).

Three-week-old piglets were purchased from a commercial herd. Before M. hyopneumoniae inoculation, blood samples were collected from each piglet. Anti-M. hyopneumoniae IgG antibody in sera, PCV-2 and PRRSV in blood were separately detected by the test kit (IDEXXCo.) and PCR was used to confirm that no piglets were exposed to M. hyopneumoniae, PCV-2, or PRRSV. On the 28th day, 4-week-old pigs were weighed and randomly allocated to two different groups. Different groups with 5 pigs in each group were housed in different rooms to avoid the spread of M. hyopneumoniae between groups. All pigs were inoculated and then euthanized as described previously (38) with minor modifications. Experimental group was inoculated intratracheally with 7 ml inoculum containing 7×10⁷ CCU of M. hyopneumoniae ES-2 strain. The control group was inoculated with 7 ml modified Friis liquid medium without ES-2.

On day 38 after infection, all pigs were euthanized and necropsied as previously described (38). For histopathological analysis, the parts of the lungs were fixed in 4% paraformaldehyde. And then, the fixed lungs were processed, stained with H&E, and visualized under a light microscope. To observe the ciliated epithelial cells and microvilli of the tracheal surface of the infected lungs, the lung bronchi were gently washed three times with 0.1 M PBS, and then were immediately put into fixative (G1102, Servicebio) at 4°C for 2–4 h for TEM. The remaining lungs were stored at −20°C, which were used to identify M. hyopneumoniae by PCR and isolate the ES-2 strain.

After recovery of the frozen M. hyopneumoniae ES-2 strain, the liquid culture with ES-2 strain was placed on a modified Friis solid medium. After culture for 9 days, the colony of M. hyopneumoniae ES-2 strain was cultured in 150 ml modified Friis liquid medium at 37°C. And then the M. hyopneumoniae ES-2 strain was harvested by centrifugation at 14,000 g for 15 min. Afterward, the DNA of the ES-2 strain was extracted by using a TIANamp Genomic DNA Kit (Q5502, Tiangen, China). The complete genome of the M. hyopneumoniae ES-2 strain was sequenced using the PacBio method. The raw reads (approximately 906,783,884 bp polymerase read bases) were obtained by the third-generation sequencing. Polymerase read quality was 0.876. The reads were assembled de novo using the HGAP2 protocol on the SMRT Portal v2.3.0 (39). A circular chromosome was obtained with 751× average base coverage. The annotation and analysis of the sequences of the genome were performed using Glimmer3 v3.02 (40) with the default parameters.

The complete sequences of a total of 162 mycoplasma strains were collected from the genome assembly database. A phylogenetic tree including 59 human isolates, 35 bovine isolates, 17 porcine isolates, 18 chicken isolates, 13 sheep isolates, and 20 other isolates (unknown host) was generated based on CDS existence information in 162 mycoplasma strains using amino acid multiple sequence alignment in Roary v3.11.2 (http://sanger-pathogens.github.io/Roary/) with default parameters (41). To analyze the genes which are present in pathogenic M. hyopneumoniae strains and absent in nonpathogenic M. hyopneumoniae strains, BlastP with a similarity cutoff (0.75, 0.80, 0.85, 0.90, 0.95, and 0.99) (42) was performed to distinguish homologous protein-coding regions from nonhomologous ones. These diagrams were developed by VENNY.

After the first generation was cultured for a week, 8 out of 9 samples were observed to be positive by PCR (Figure 1A). The 8 positive samples were separately transferred into a fresh Friis liquid medium (1:4). After one week's culture, 5 out of 8 samples were found to be positive by PCR (Figure 1B). These 5 positive samples were separately transferred into fresh liquid Friis medium (1:4). After one week's culture, 3 out of 5 samples were detected to be positive by PCR (Figure 1C). In this continuous passage culture, 2 samples of the fourth generation remained positive by PCR (Figure 1D). Subsequently, the serial passage culture of these 2 positive samples was performed in the modified Friis liquid medium. After 20 passage cultures, M. hyopneumoniae was still identified by PCR (Figure 1E).

To get pure M. hyopneumoniae strain, the liquid cultures were placed on the modified Friis solid medium. And then a single colony was cultured in the modified Friis liquid medium. After three times of purification, sequencing results showed that the microorganism purified in liquid medium is M. hyopneumoniae strain (GenBank accession number GCA_004768725.1). Ultimately, a field M. hyopneumoniae strain was successfully isolated and purified from the lung of Enshi black pigs suffering from typical swine mycoplasma pneumonia. And we named this newly isolated strain M. hyopneumoniae strain ES-2. Meanwhile, this study found that the minimum inhibitory concentration of M. hyopneumoniae ES-2 strain to kanamycin was 2 μg/ml.

To study the colony morphology of ES-2 strain on solid media, after 9 days of culture, the colonies from the M. hyopneumoniae ES-2 strain on solid medium were observed through microscopic in this study. The colonies were found to be very small and granular with irregular margins with their central parts burrowed into the medium, which exhibited a typical “fried egg” appearance (Figures 2A,B). To further observe the morphology of cells, M. hyopneumoniae ES-2 strain was collected by centrifugation. ES-2 strain cells were found to be spherical or elliptic and they differed in sizes under SEM (Figures 2C,D). Meanwhile, TEM results exhibited that the morphology of most cells was spherical and a very small number of them was irregular in morphology (Figures 2E,F). The diameters of ES-2 strain cells ranged from 0.2 to 1.6 μm and the diameters of approximately 80% of cells were between 0.4 and 1.0 μm under the electron microscope (Figure 2G).

Color-changing units (CCU) of liquid medium were commonly used to determine the viability of M. hyopneumoniae. After M. hyopneumoniae ES-2 strain culture was transferred into a fresh medium, the results showed that the number of ES-2 strain cells was approximate 10⁷ CCU/ml on day 0 (Figure 3A). From day 0 to day 2, the number of ES-2 strain cells was constantly increasing. However, the pH of the color shift continued to decline (pH 7.55–6.37) during the whole incubation process. And the viability in cultures reached a maximum (10¹⁰ CCU/ml) approximately on day 2 postincubation with the corresponding pH of 7.24. After 2 days of culture, the viability in cultures continuously declined. On day 8, only 10 CCU/ml of ES-2 strains viability was detected with the corresponding color shift pH being 6.37. On day 7, color shift pH was minimum (6.37) and stopped declining (Figures 3A,B).

Some studies reported that M. hyopneumoniae could infect pigs of various ages resulting in typical lesions of EP (43). In this study, the 28-day-old pigs infected with M. hyopneumoniae ES-2 strain for 38 days showed typical pneumonia in the accessory lobes, apical lobes, cardiac lobes, and the cranial portion of the caudal lobes of the lungs (Figure 4F). However, typical lesions of EP were not discovered in the lungs of pigs inoculated with fresh medium without ES-2 strain in the control group (Figures 4A–D). In the pathogenesis of M. hyopneumoniae infections, it is crucial for pathogenic M. hyopneumoniae to damage the ciliated epithelial cells of the trachea, bronchi, and bronchioles (1, 2). SEM results showed that the cilia of the bronchi in the lungs of the pigs infected with the ES-2 strain were tangled, clumped, split, and lost, compared to those of the healthy lungs (Figures 4G–I). In addition, the histological examination results of the lungs of pigs infected with M. hyopneumoniae ES-2 strain showed that much less air was present in the infected lung due to the massive infiltration of inflammatory and lymphocytes cells (Figure 4J), compared to that in the healthy lung (Figure 4E). Therefore, it could be concluded that the M. hyopneumoniae ES-2 strain isolate is a pathogenic strain.

The genomes of 9 M. hyopneumoniae strains were available from the National Center for Biotechnology Information (NCBI) database. Of these 9 genomes of M. hyopneumoniae strains, the assembly level of 6 genomes (J, 7422, 7448, 168, 168-L, 232, and KM014) was complete. The other 2 genomes (11 and TB1) were incomplete. The genome size of M. hyopneumoniae strains ranged from 0.8 to 1.0 Mb (Supplementary Table S1). In this study, the complete genome of M. hyopneumoniae ES-2 strain (GenBank accession number GCA_004768725.1), which was sequenced using the PacBio method, contained a 956,514 bp (GC content 28.36%) single circular chromosome. The total number of genes predicted was 733, and the average gene length was 1,164 bp (Supplementary Table S1). Subsequently, 25 virulence factors of mycoplasma reported (http://www.mgc.ac.cn/cgi-bin/VFs/compvfs.cgi?Genus=Mycoplasma) were collected from the virulence factor database to investigate the distribution of virulence-related factors among 9 M. hyopneumoniae strains. By aligning the proteins (identity ≥ 80% and coverage ≥ 80%) from each strain against the reported 25 virulence factors, a total of 11 virulence factors were identified in these genomes. In total, 10 out of 11 virulence factors were found to be present in each of 9 M. hyopneumoniae strains, and some virulence factors had multiple copies in these genomes. The cytoadherence protein (P200) (44) of Mycoplasma pneumoniae which functions in cytoadherence and gliding motility was found to be present only in the genome of the ES-2 strain and absent in other strains (Supplementary Table S1). So putative protein P200 has existed in M. hyopneumoniae ES-2.

In addition, the above experiments show that M. hyopneumoniae strain ES-2 is pathogenic for pigs. Among the M. hyopneumoniae strains which have been reported, M. hyopneumoniae strains 7422, 7448, 168, 168-L, and 232 were also reported to be pathogenic for pigs. Only M. hyopneumoniae strain J was nonpathogenic. The pathogenicity of the remaining 3 strains (KM014, TB1, and 11) was unknown.

To understand the phylogenetic relationship between different mycoplasma species, all 162 complete genomes of different species of mycoplasma strains were collected from the NCBI database. However, no core gene was found in the 162 mycoplasma strains. Therefore, the phylogenetic tree was first constructed based on CDS existence information in 162 mycoplasma strains. The phylogenetic tree showed that 15 porcine mycoplasmas (8 M. hyopneumoniae, 1 M. flocculare, and 6 M. hyorhinis), 2 M. dispar, 1 M. bovoculi, 3 other mycoplasmas form a group; 53 human mycoplasmas form another group; 10 sheep mycoplasmas and 14 bovine mycoplasmas form the third group; 12 chicken mycoplasmas, 2 human mycoplasmas, 2 porcine mycoplasmas, 1 bovine mycoplasma, and 7 other mycoplasmas form the fourth group; and 17 bovine mycoplasmas, 6 chicken mycoplasmas, 53 human mycoplasmas, and 13 other mycoplasmas form the fifth group (Figure 5).

To detect the genes which are present in pathogenic M. hyopneumoniae strains but absent in nonpathogenic M. hyopneumoniae strains, each gene of 4 pathogenic M. hyopneumoniae strains ES-2, 168, 7,448, and 7,422 and each gene of 1 nonpathogenic M. hyopneumoniae strain J were aligned based on the different levels of identity (0.75, 0.8, 0.85, 0.9, 0.95, and 0.99) at the amino acid level. Venn diagram of the results showed that the number of common genes shared by 4 pathogenic M. hyopneumoniae strains and 1 nonpathogenic M. hyopneumoniae strain J were, respectively, 686 (identity = 0.75), 687 (identity = 0.80), 683 (identity = 0.85), 678 (identity = 0.90), 666 (identity = 0.95), and 521 (identity = 0.99). Excluding the common genes shared by pathogenic and nonpathogenic strains, 12 (identity = 0.75), 10 (identity = 0.80), 12 (identity = 0.85), 13 (identity = 0.90), 12 (identity = 0.95), and 25 (identity = 0.99) genes were found to be shared by all 4 pathogenic M. hyopneumoniae strains at the amino acid level, respectively (Figure 6). Further analysis of the genes which are present in pathogenic M. hyopneumoniae strains but absent in nonpathogenic M. hyopneumoniae strains indicated that 34 genes specific to pathogenic M. hyopneumoniae strains were detected at the amino acid level, of which 18 were hypothetical proteins, and another 16 were annotated in NCBI database (Table 1). By BlastN, 2 out of 34 genes specific to pathogenic M. hyopneumoniae strains at the amino acid level were absent in the nonpathogenic M. hyopneumoniae J strain at the nucleotide level (Table 2). ES-2_00484 gene was only identified in pathogenic M. hyopneumoniae strains 168, 168-L, 7422, and 7448. A partial sequence of the ES-2_00744 gene was only identified in pathogenic M. hyopneumoniae strains 232, 168-L, 7,422, and 7,448. Although the pathogenicity of M. hyopneumoniae KM014 is unclear, both genes were identified in the strain.