Genotype diversity of Mycoplasma Hyopneumoniae in Chinese swine herds based on multilocus sequence typing

Background Between 2018 and 2020, 989 clinical specimens from pigs showing clinical signs of a variety of swine diseases in 27 provinces in China were sampled and submitted for further testing. Nested PCR targeting the 16S rRNA gene of Mycoplasma hyopneumoniae and subsequent sequencing were used to analyse these specimens. Mycoplasma hyopneumoniae-positive samples were assayed by multilocus sequence typing (MLST). The aim of the study was to reveal the distribution of M. hyopneumoniae and determine the genotypes of M. hyopneumoniae in pig herds in China based on MLST. Results Among these 989 samples, 199 samples were M. hyopneumoniae-positive. The M. hyopneumoniae positivity rate was 7.2% (35/494) in 2018, 18.4% (38/207) in 2019, and 43.8% (126/288) in 2020. In total, 47 samples were successfully assayed by MLST. Sixteen new M. hyopneumoniae sequence types from 9 provinces were recorded in the present study. Conclusions This is the first report on sample positivity rates and molecular typing results for M. hyopneumoniae in swine herds in China. MLST has revealed high genotype diversity among M. hyopneumoniae from different provinces of China.


Introduction
Enzootic pneumonia (EP) in pigs caused by Mycoplasma hyopneumoniae results in large economic losses worldwide. Isolation and cultivation of M. hyopneumoniae is complicated and time-consuming. Genotyping analysis of clinical materials rather than isolates of M. hyopneumoniae has been very useful in field investigations. Multilocus sequence typing (MLST) [1,2], a molecular tool used in epidemiological investigations, has proven useful in investigations of outbreaks of EP in pigs and can be performed in clinical materials [3][4][5]. The data obtained can be easily shared among laboratories through online databases. While most MLST schemes utilize seven genes, it has been previously demonstrated that the use of only three genes (adk, rpoB and tpiA) provides the same resolution as the seven-gene scheme [1]. Molecular epidemiological studies on M. hyopneumoniae infection of pigs in China are rare, thus the aim of the present study was to reveal the distribution of M. hyopneumoniae and determine the genotypes of M. hyopneumoniae in pig herds in China based on MLST.

Samples
Between 2018 and 2020, a total of 989 clinical specimens (homogenates derived from the lungs, spleen, liver, and lymph nodes) from 989 pigs with clinical symptoms of a variety of swine diseases in 215 pig herds in 27 provinces in China were sampled. Each homogenate sample was derived from one pig. Among them, 494 clinical specimens were sampled from 146 herds in 18 provinces in 2018 (Fig. 1B), 207 clinical specimens were sampled from 24 herds in 10 provinces in 2019 (Fig. 1C), and 288 clinical specimens were sampled from 45 herds in 14 provinces in 2020 (Fig. 1D), The sampled provinces from 2018 and 2020 are shown in Fig. 1A.

DNA extraction, PCR amplification and sequencing
M. hyopneumoniae DNA was isolated from the homogenates using a magnetic particle-based sample preparation kit (Xi'an Tianlong Science & Technology Co., Ltd., Xi'an Shaanxi Province, China) in accordance with the manufacturer's protocol. Nested-PCR [6] was used for the detection of M. hyopneumoniae. The first PCR amplified a 649-bp region of the 16S rRNA gene. After the second PCR, DNA segments measuring 352 bp in length were cloned into a pMD18-T vector (TaKaRa, Dalian Laoning Province, China) and sequenced by the BGI Group (Qingdao, China). Positive samples were confirmed by using the BLAST (http:// blast. ncbi. nlm. nih. gov) and subjected to MLST.

MLST
Nucleotide sequence data of the adk, ropB, and tpiA genes were obtained from M. hyopneumoniae-positive samples by Sanger sequencing. In brief, the three genes were amplified by primers and conditions described previously [1]. The PCR products were analysed by 1.5% agarose gel electrophoresis and purified with the Wizard ® SV Gel and PCR Clean-Up System (Promega Corporation, Madison, WI, USA). DNA segments were sequenced by the BGI Group (Qingdao, China). Sequence data were submitted to the PubMLST database (http:// pubml st. org/ mhyop neumo niae) for sequence type (ST) designation. The PubMLST database contains nomenclature-allele definitions that provide an identifier for every unique allele sequence and MLST profiles that index each unique combination of alleles with a ST. Sequences were aligned using the CLUSTAL W and cluster analysis was performed using the neighbour-joining method (Kimura two-parameter method) with Molecular Evolutionary Genetics Analysis (MEGA) software (version 6.0; https:// www. megas oftwa re. net/). Bootstrap values were based on 1000 replicates. Values of Simpson's index of

Results and discussion
The results showed that 199 of 989 samples were M. Among the 199 samples, only 21 samples were sent with accompanying clinical sign information. These clinical signs included respiratory signs (dyspnoea, hyperpnea, coughing, fever or high fever), reproductive signs (abortion, weak piglets, mummified foetuses, or dead foetuses), neurological signs (ataxia, circling, paresis, or tremors), and gastrointestinal signs (vomiting or diarrhea). We assumed that most infections with M. hyopneumoniae were complicated by coinfection with other pathogens.
From 2018 to 2020, samples from 59 herds distributed in 20 of 27 provinces were positive for M. hyopneumoniae. Positive samples were detected from pig herds distributed throughout different regions of China (Fig. 1). Samples positive for M. hyopneumoniae were distributed in 9 of 18 provinces in 2018, 8 of 10 provinces in 2019, and 11 of 14 provinces in 2020 (Fig. 1). Seventy-three positive samples of 307 total samples were distributed in Hebei, Henan, and Guangxi provinces from 2018 to 2020, and 13 positive samples of 40 total samples were distributed in Shaanxi province in 2019 and 2020. No positive samples were detected in Gansu, Ningxia, Anhui, Zhejiang, Hubei, Hainan, or Jilin provinces. In these provinces, few samples were collected and most of them were collected in 2018. More samples should be collected from herds in these provinces to confirm the M. hyopneumoniae infection situation.
Direct contact and aerosol transmission are the main routes of M. hyopneumoniae transmission. Long-distance airborne transmission of the pathogen can occur [7].
Treatment with antibiotics such as tetracycline, macrolides and lincosamides can reduce clinical symptoms and the bacteria load in an individual [8], but it does not clear the infection. In a previous study, M. hyopneumoniae persisted in alveolar macrophages, interstitial macophages and type I pneumocytes in the lungs after experimental infection [9]. Therefore, a history of infection with M. hyopneumoniae in a herd could present a risk for persistence of the pathogen and reinfection. In China, the first M. hyopneumoniae strain was isolated in 1973 [10]. M. hyopneumoniae has existed for at least nearly 50 years and has evolved highly diverse genotypes.
In this study, M. hyopneumoniae genotyping was performed for the first time in China. Three housekeeping genes (adk, rpoB and tpiA) were detected in 47 of 199 M. hyopneumoniae-positive samples distributed among 26 herds. MLST allowed the amplification and sequencing of some individual genes (1-2) among 7 targets. However, the successful sequencing of all 7 targets was not possible for the majority of the samples. It is attributed to the low amount of M. hyopneumoniae DNA or the presence of unknown PCR inhibitors in samples [4,5]. In the rest of the positive samples, only some of the three housekeeping genes were amplified. Alleles and STs from the examined herds are summarized in Table 1 (Table 1). Except for ST128 and ST149 (detected in herds in Henan and Guangxi provinces), the STs were unique to the herds in each province. Since all the samples represented new STs based on the PubMLST database, we hypothesize that they belonged to exclusive Chinese genotypes. However, the ST data we obtained are limited and should be further studied in the future.
Some STs (ST128, ST149) were present in different herds in different provinces. Among these provinces, Jiangsu and Guangxi were the farthest apart, at 1600 km. The results imply that free movement of pigs is associated with a high risk for the transmision of M. hyopneumoniae. Since the first ASF case in China was reported on 3 August 2018, the Chinese government has strengthened measures to control animal movement. The implementation of strict movement measures can also improve EP control.
Phylogenetic analysis showed that the 16 new STs were clustered into five clades with 37 STs from 15 countries: Ia, I b, II, III, and IV, as shown in Fig. 2. Ia and I b were the predominant clades prevalent in Chinese pig herds. Clade Ia included ST149, ST151, ST129, ST174 and STs from the USA, France, Cuba, Hungary, UK, Australia, Canada, and Thailand. Clade Ib contained all STs from China, which means M. hyopneumoniae has existed in China for a long time and formed individual cluster. Clade II included ST172, ST169 and STs from Hungary, Switzerland, Greece, and France. Subtype III included ST173, ST130 and STs from the Czech Republic and France. No Chinese STs were present in calde IV. ST151 and ST149, which were detected in samples from different herds in Guangxi Province were phylogenetically related to ST61 representing Chinese strain 168, but only one of 3 alleles was the same (Fig. 2). This reference strain was isolated in 1979 from an EP outbreak in China [11]. This may imply that the original ST was replaced by the new STs via evolution.
ST152 and ST147, which were detected in samples from different herds from the same province (Guangxi), were phylogenetically close, with a 2-bp difference in  (Fig. 2). ST129 was detected in samples from different herds in Shandong Province in 2018 and 2019 and formed a distinct cluster. ST174 detected in samples from Shaanxi Province, which is geographically isolated from the other provinces, also formed a distinct cluster. The STs of samples from these provinces are possibly prone to the formation of individual clusters.
ST146 and ST148 are phylogenetically close to ST128, which is the most common ST. ST175 and ST145 are phylogenetically close, differing in the tpiA gene by 4 bp. The three alleles of ST173 and ST130 are different but form a distinct cluster (Fig. 2). These STs are geographically unrelated but form a cluster, which implies the spread of M. hyopneumoniae among provinces.

Conclusions
Positive samples were detected in pig herds distributed throughout different regions of China. MLST revealed high genotype diversity among M. hyopneumoniae isolates from different provinces of China. All the samples represented new STs based on the PubMLST database, and we hypothesize that they belong to exclusive Chinese genotypes. ST128, which was present in 4 provinces in different regions of China, was the most common ST in this study. The STs from most provinces are possibly prone to formation of their own cluster via evolution. Phylogenetic analysis showed that the Chinese STs belonged to four clades. It is difficult to elucidate the situation of M. hyopneumoniae in China in the long-term because of the high genotype diversity of the pathogen. This is the first study to report sample positivity rates and molecular typing results of M. hyopneumoniae in swine herds in China. To further understand the epidemiology and possible sources of M. hyopneumoniae, a surveillace plan for the disease should be formulated and implemented, and more samples could be collected and genotyped.