Whole genome sequencing and analysis of a goose-derived Mycoplasma gallisepticum in Guangdong Province, China

Partial air-sac tissues were obtained from five suspected Mycoplasma gallisepticum-infected geese on a meat goose farm located in Sanshui District, Foshan City, Guangdong Province in China.

Clinical symptoms in diseased geese were systematically observed, and post-mortem examinations were conducted on deceased geese. Detailed records of the findings were made.

Air-sac tissues from affected geese were collected for the isolation of MG. A small amount of air-sac tissue was added to modified Frey’s liquid medium (Qingdao Haibo Biotechnology Co., Qingdao, China), vortexed and then incubated at 37°C. When the medium colour changed from red to yellow, the sample was plated. Colonies exhibiting clear Mycoplasma characteristics were selected for cultivation and sub-culturing. The purified second-generation bacterial fluid was inoculated onto modified Frey’s solid medium and incubated at 37°C in a 5% CO₂ incubator for 7 d to observe colony morphology under a microscope (Nikon, Tokyo, Japan). To rule out the possibility of erroneously studying L-form bacteria, which have certain similarities to mycoplasma, the isolated pathogen was cultured and was observed conditionally on its displaying a stable and typical “fried-egg” appearance.

Twenty healthy 7-day-old Yangzhou geese, obtained from a goose hatching facility (Qingyuan Hongxing, Qingyuan, China), were used in the study. The geese were evenly distributed, with an equal representation of males and females. The criteria for selection included the absence of respiratory symptoms and negative results in a plate agglutination test and tracheal swab PCR. The geese were randomly divided into two groups of 10: one for inoculation and the other as a control group. A 0.5 mL sample of the third-generation bacterial fluid in the logarithmic growth phase was inoculated twice daily into the trachea of goslings in the inoculation group. Continuous monitoring of clinical symptoms was performed in the inoculation group and the control group. After 15 d, post-mortem examinations were conducted to observe pathological changes.

A 10 μL reaction system was prepared according to Table 1, and amplification was carried out using a PCR machine (Touch T960; JingGe Scientific Instrument, Hangzhou, China). The selection of primers was based on a thorough literature review. The intergenic spacer region primers (5'16S-IGSR-23S-3') (31), designed to differentiate between Mycoplasma and Mycoplasma gallisepticum, were synthesised by Sangon Biotech Co. (Shanghai, China)). The details of the primers are provided in Table 2. The identification of amplification products was accomplished through agarose gel electrophoresis.

The identified clinical isolates were purified again, and part of the samples were sent to the Chinese Center for Type Culture Collection (CCTCC, Wuhan, China). The isolates were named Mycoplasma gallisepticum MG-GD01/22 according to their species classification and assigned preservation number CCTCC M 20232686.

A minimal inhibitory concentration (MIC) test was performed utilising an agar dilution method with slight modifications incorporating blank controls and quality controls (10). Modified Frey’s solid medium with an appropriate drug concentration was prepared. The MIC of eight antibiotics were determined against the goose-origin MG-GD01/22, the antibiotics spanning four classes: enrofloxacin and danofloxacin as quinolones, spectinomycin as aminoglycosides, tilmicosin, tylosin, and acetylisovaleryltylosin tartrate as macrolides, and tiamulin and valnemulin as pleuromutilins. Bacterial fluid containing the isolate was diluted to 10⁶ colony-forming units (CFU)/mL, and 100 μL of the bacterial fluid was inoculated on the drug-containing solid plates, ensuring even distribution. The plates were then incubated at 37°C in a 5% CO₂ incubator for 7 d.

A 5-mL volume of MG-GD01/22 bacterial fluid was inoculated into 45 mL of modified Frey’s liquid medium and cultured at 37°C to the logarithmic phase. After centrifugation at 12,000 rpm and 4°C for 30 min, the culture medium was discarded. The white precipitate was washed with ultrapure water, and the liquid was discarded after centrifugation. The collected precipitate was resuspended in 120 μL of ultrapure water and vortexed evenly. Extraction of total DNA from the MG-GD01/22 strain was achieved with a Mag-Bind Bacterial DNA 96 Kit (Omega Bio-tek, Norcross, GA, USA) as stated in the manufacturer’s manual. The DNA concentration was determined by micro-UV spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA). The qualified DNA was stored at −20°C and then sent to Sangon Biotech (Shanghai, China) for third-generation whole-genome sequencing. The whole genome of MG-GD01/22 was sequenced using PacBio combined with Illumina technology, and sequence splicing was performed using Canu 1.3 (23), GapFiller v. 1.11 (7) and PrlnSeS-G 1.0.0 (29). The Prokka tool (32) was utilised to predict basic gene information from the MG-GD01/22 whole-genome sequencing results. Clustered, regularly interspaced short palindromic repeats (CRISPR) were predicted using the CRISPR Recognition Tool (6), gene islands were predicted using IslandPath-DIMOB (5), and prophage prediction analysis using PhiSpy (1). Gene function annotation was performed for the isolate. The bioinformatics software used in this study is listed in Table 3.

The 16S rRNA and mgc2 gene sequences of MG-GD01/22 were subjected to comparative analysis with those of various representative strains of MG and a single strain of MI archived in the GenBank database. The phylogenetic tree and the nucleotide sequence homology assessment of 16S rRNA and the evolutionary tree for the mgc2 gene were constructed by MEGA XI; FigTree and MegAlign Pro software (DNASTAR, Madison, WI, USA). A gene collinearity analysis between MG-GD01/22 and evolutionarily proximate MG representative strains was also performed using MUMmer3.1 software (25).

The affected geese manifested lethargy, unkempt feathers and decreased food consumption. Some animals presented symptoms including nasal discharge, head shaking, neck extension, open-mouth breathing and elevated body temperature. Auscultation revealed wet rales in the lungs.

Initial dissection revealed mucus accumulation in the trachea and bronchi, pinpoint bleeding and congestion in lung tissues. A minor fibrous exudate between the thoracic and abdominal membranes and the air sac membranes was also found. Subsequent observations indicated patchy deposits on the air sacs accompanied by a degree of air-sac thickening and turbidity. In advanced stages of the disease, the air sacs contained cheesy white or yellowish substances, while the lungs and bronchi were filled with purulent secretions. Localised congestion or parenchymal consolidation was evident in lung tissues (Fig. 1A and B).

Fig. 1.

Isolation and identification of goose-origin Mycoplasma gallisepticum. A – thickened air-sac wall of an affected goose: wall with white viscous liquid; B – blood in statis in the lung tissue of an affected goose, and large amounts of yellow and white cheese-like material in the air sacs; C – colonies of clinical MG from goose isolates (100×); D – the PCR identification result of goose-origin MG. M – DL1000 Marker; 1 – MG-GD01/22; 2 – MG S6; 3 – Mycoplasma synoviae strain MS wvu1853; 4 – Control

The inoculation of the purified bacterial fluid on modified Frey’s solid medium yielded mycoplasma colonies exhibiting characteristic of “fried-egg” morphology under microscopic examination (Fig. 1C). None of the bacteria identified as L-form, and the viable cell count surpassed 10⁸ CFU/mL.

A distinctive band aligned with the MG S6 standard strain manifested at around 800 bp (Fig. 1D). This and the colony morphology identification methods collectively confirmed the goose-origin nature of this clinical isolate. The strain was designated as MG-GD01/22.

Animals in the infected group exhibited clinical manifestations, including open-mouthed breathing, head shaking, nasal discharge, depression and decreased food intake over a 15-day period. Necropsies revealed the nasal cavity to be filled with yellow purulent secretions, the air sacs to be slightly thickened and the existence of point-shaped and patchy light-yellow cheesy exudates. Contrastingly, animals in the control group displayed neither evident clinical symptoms nor pathological alterations (Fig. 2).

Fig. 2.

Changes caused by Mycoplasma gallisepticum in geese. A – clean nasal cavity; B – normal, thin and transparent air sacs in a control-group goose; C – yellow, viscous secretion in the nasal cavity of an infected-group goose; D – slightly thickened air sacs withs cloudy, light-yellow cheese-like exudation in an infected-group goose

The findings suggested that MG-GD01/22 had resistance to quinolones (levofloxacin and enrofloxacin) and aminoglycosides (florfenicol) but susceptibility to macrolides (tilmicosin, tylosin and tiamulin) and pleuromutilins (thiamphenicol and valnemulin) (Table 4).

The concentration of total DNA extracted from MG-GD01/22 was measured at 147.5 ng/μL. The sample purity was evaluated by means of the OD 260 nm : OD 280 nm ratios as 1.91 and OD 260 nm : OD 230 nm ratios as 2.00, which met the prescribed standards for whole-genome sequencing. Following first-generation sequencing validation, the species origin was conclusively identified as Mycoplasma gallisepticum. Utilising the third-generation gene sequencing technology, the comprehensive genome length of MG-GD01/22 was determined to be 1,024,330 bp, featuring a guanine + cytosine content of 31.56%. A total of 1,666 coding genes were predicted, constituting 72.25% of the genome, complemented by 33 transfer RNAs and 6 ribosomal RNAs (Fig. 3A).

Fig. 3.

Basic information and bioinformatics prediction for Mycoplasma gallisepticum strain MG-GD01/22 isolated from inoculated Chinese geese. A – visualisation of the MG-GD01/22 genome; B – clustered, regularly interspaced short palindromic repeat fragment SeqLogo of MG-GD01/22. COG – clusters of orthologous groups; GC – guanine + cytosine; CDS – coding sequence; rRNA – ribosomal RNA; tRNA – transfer RNA

There were three genomic islands and three prophage structures were anticipated within MG-GD01/22. According to the CRISPR prediction, the repetitive sequence in MG-GD01/22 began at position 867,297 and concluded at position 869,120. The repetitive sequence spanned 36 bp, featuring an interspace of 30 bp between consecutive repetitive sequences. A total of 28 repetitive sequence structures were identified in this study (Fig. 3B).

The COG system was established by the National Center for Biotechnology Information (NCBI) to annotate genes in prokaryotes for the direct homologous relationships among them. It categorises genes into three primary functional divisions. The system’s evolutionary relationship clusters genes with homologous traits from diverse species into distinct groups, wherein genes within the same group exhibit like functionalities. This enables the prognostication and elucidation of the acquired gene sequences. In this study, the predicted protein sequences of genes were aligned with the COG database for annotation, and the annotated genes were subsequently classified into specific groups (Fig. 4). Successfully annotated genes were classified within the MG-GD01/22 genome. The largest gene cluster was predicted to be associated with translation, ribosomal structure and biogenesis and to comprise 174 genes involved in cellular processes and signal transduction. The next-largest category within information storage and processing genes was replication, recombination and repair, encompassing 115 genes. The most prevalent category predicted was the metabolic cluster, which included 49 genes specifically involved in energy production and conversion. Separately, another subset comprised 45 genes associated with carbohydrate transport and metabolism. Additionally, the study identified two other distinct groups of 36 genes each, one involved in nucleotide transport and metabolism, and the other in the transport and metabolism of inorganic ions. The synthesis, transport and metabolism of secondary metabolites had the fewest predicted genes with only one of each. There remained 22 genes with unknown functions (Fig. 4).

Fig. 4.

Functional analysis of the Mycoplasma gallisepticum strain MG-GD01/22 isolated from inoculated Chinese geese based on Cluster of Orthologous Groups (of proteins) (COG) database annotation results

Gene ontology serves as an internationally standardised gene function classification system, offering a dynamically updated standard vocabulary for a thorough depiction of the characteristics of genes and their products within an organism. It encompasses three principal categories: molecular function of genes, cellular location and participation in biological processes. In this study, genes were categorised based on the GO system, and the outcomes are delineated in Fig. 5. We conducted GO classification on the acquired genes and identified the predominant gene category associated with biological processes. Among these genes, 724 were linked to metabolic processes, 694 to cellular processes, 103 to localisation, 71 to response to stimulus, 51 to biological regulation processes and 17 to biological adhesion. The next most numerous category comprised molecular function genes, predominantly catalytic activity genes (682) and binding factor genes (664). There were 122 transporter protein activity genes and 82 structural molecular activity genes, while 10 genes were associated with nucleic acid binding transcription factor activity. The category with the lowest predicted gene count was that of cell components, encompassing 577 genes predicted to be involved in cell-related processes, 244 genes related to membranes and 124 genes associated with cellular organelles (Fig. 5).

Fig. 5.

Functional analysis of the Mycoplasma gallisepticum strain MG-GD01 isolated from inoculated Chinese geese based on the Gene Ontology (GO) database annotation results

Fig. 6.

Functional analysis of the Mycoplasma gallisepticum strain MG-GD01 isolated from inoculated Chinese geese based on Kyoto Encyclopedia of Genes and Genomes (KEGG) database annotation results

The Pathogen Host Interactions database (PHI-base), dedicated to cataloguing pathogen–host interactions, was used in this study. Annotation efforts revealed ten genes associated with loss of pathogenicity, seven genes increased the pathogen’s virulence, seven genes acted as effector, three genes with negligible impact on pathogenicity, and two genes reduced virulence (Table 5).

The Virulence Factor Database (VFDB) collates pathogenic factors in bacteria, chlamydiae and mycoplasmas. The setA gene is its core, symbolising experimentally verified virulence genes. The comparison of gene sequences of this study with the VFDB led to the annotation of 25 virulence genes in MG-GD01.

Predominantly, these genes belonged to the category of adenosine triphosphate synthase–binding cassette transporters (48%). Additional predictions encompassed vlhA (pMGA), pvpA, mgc2, gapA/crmA, groEL, tuf, relA encoding (p)ppGpp, msrA, msrB and various other virulence-related genes.

The Comprehensive Antibiotic Resistance Database (CARD) is a database of bacterial antibiotic resistance genes. A total of 22 antibiotic resistance genes in MG-GD01 were identified in this study. Notably, the highest number of genes conferred resistance to quinolone antibiotics (10 genes), a lower number were associated with aminocoumarin resistance (7 genes), and a single gene was noted to confer streptomycin resistance (Table 6).

The Carbohydrate-Active enzymes (CAZy) database encompasses six major enzyme categories: auxiliary oxidoreductases, glycoside hydrolases, glycosyltransferases, polysaccharide lyases and carbohydrate esterases. A search of the database for genes in the MG-GD01/22 strain found nine encoding proteins related to carbohydrate enzymes, including four proteins from the glycosyltransferase family 2 (carbohydrate structure domains) and two components of the pyruvate dehydrogenase complex dihydrolipoamide dehydrogenase (E3) (glycolipase). There were annotations for one gene each in polysaccharide lyases, auxiliary oxidoreductases and glycosyltransferases, with the highest proportion being proteins containing the found-in-various-architecture domain. No genes encoding proteins related to the glycoside hydrolase superfamily were annotated.

Thirty-two MG strains and one MI strain with high similarity to the 16S rRNA gene sequence (246,487 bp–247,991 bp; 1,505 bp) of MG-GD01/22 were selected by the NCBI basic local alignment search tool (BLAST) as reference strains for nucleotide sequence homology and phylogenetic tree analysis. The current isolate MG-GD01/22 differed minimally in nucleotide sequence from MG reference strains, demonstrating homology ranging from 99.4% to 99.9%. Its closest phylogenetic relationship was identified with MG S6 at 99.9%, whereas the homology with MI was comparatively lower at 95.2% (Fig. 7).

Fig. 7.

Nucleotide sequence homology analysis of the 16S ribosomal RNA of the Mycoplasma gallisepticum strain MG-GD01/22 isolated from inoculated Chinese geese

Recovery sequencing confirmed that the sequencing outcomes were entirely consistent with those of the 16S rRNA and mgc2 gene fragments initially isolated from the strain. The evolutionary tree topology of the current strain alongside reference strains indicated MG-GD01/22’s closest relationship being to the MG S6 strain and its remoteness from MI 4229 (Fig. 8).

Fig. 8.

Phylogenetic tree of 16S ribosomal RNA of the Mycoplasma gallisepticum strain MG-GD01/22 (indicated by the red star) isolated from inoculated Chinese geese

A meticulous comparative analysis was executed on the mgc2 gene sequence (155,060 bp–156,418 bp; 1,359 bp) of MG-GD01/22. Twenty-five MG strains exhibiting high similarity were chosen through NCBI BLAST to serve as reference strains for comprehensive comparison and systematic evolutionary tree construction. The resultant evolutionary tree topology indicated the distinctive nature of MG-GD01/22’s mgc2 gene typing in comparison to the 25 established strains, albeit with only minimal differences to 3 of them (Fig. 9).

Fig. 9.

Phylogenetic tree of mgc2 of MG-GD01/22 (indicated by the red star) isolated from inoculated Chinese geese

MUMmer is a suffix tree algorithm designed to find exact maximal unique matches (MUMs) of some minimum length between two input sequences. The match lists produced by MUMmer can be used alone to generate alignment dot plots, or can be passed on to the clustering algorithms for the identification of longer non-exact regions of conservation. The outcomes unveiled a substantial genomic homology between the clinical isolate and MG S6, although the collinearity was comparatively deficient. Noteworthy modifications were observed in the distribution of numerous genes, encompassing instances of gene loss and inversion (Fig. 10).

Fig. 10.

Collinearity analysis of the genomes of the Mycoplasma gallisepticum strain MG-GD01/22 (x-axis) isolated from inoculated Chinese geese and the MG S6 strain (y-axis). Forward maximal unique matches (MUMs) are plotted as red lines/dots while reverse MUMs are plotted as blue lines/dots. A line of dots with slope = 1 represents an undisturbed segment of conservation between the two sequences, while a line of slope = −1 represents an inverted segment of conservation between the two sequences