Molecular Typing of Indian Mycoplasma synoviae Isolates

Mycoplasma synoviae is an economically significant pathogen of poultry reported worldwide. It is mostly associated with subclinical respiratory infection in poultry. Co-infection with respiratory bacteria and viruses causes air sacculitis (Kleven et al., 1972). Respiratory form of infection is highly common as compared to other forms and its severity is affected by concurrent infection with other respiratory pathogens (Dijkman et al., 2014). Systemic infection results in infectious synovitis in poultry and turkey. Another condition caused by this pathogen is eggshell apex abnormality.
       
The prevalence of M. synoviae in the Indian poultry industry is well documented (Baksi et al., 2016; Sumitha and Sukumar, 2017; Rajkumar et al., 2018). Previous study shows a higher prevalence of M. synoviae than M. gallisepticum in poultry flocks in India (Rajkumar et al., 2018). Maintenance of Mycoplasma free flock is highly difficult in multi-aged farms and mortality with lesions of airsacculitis are common during adverse weather conditions like summer and winter. Antibiotic treatment is generally used for control of disease due to Mycoplasma in India. Despite the wide prevalence of this pathogen in India, not much has been done for the genetic data analysis and molecular typing of the Indian isolates. 
       
The immune dominant variable lipoprotein haemagglutinin encoded by the vlhA gene which shows antigenic variation which helps in binding with host cell receptors and colonisation (Razin et al., 1998; Bercic et al., 2008). The vlhA gene has a conserved and variable region. The conserved 5’ end of the vlhA gene (nucleotides 1-410) encoding for major surface protein B (MSPB), occurs in a single chromosomal copy, where as the remaining coding sequence occurs as multiple copies (Noormohammadi et al., 2000). Antigenic variation of vlhA protein occurs due to recombination of 3¢ end of the expressed vlhA gene with one of the multiple pseudogenes. The conserved 5' end region is a highly polymorphic region encoding the Proline Rich Repeats (PRR) in the N-terminal part of MSPB protein. Isolates varies in the length of the PRR. Studies have suggested length of the PRR is associated with invasiveness (Bencina et al., 2001).

Isolation and identification of M. synoviae
 
Tracheal/choanal cleft/oviduct/Hock joint swabs were collected from carcasses showing respiratory lesions or arthritis during routine postmortem examination at an organized poultry farm at Hyderabad, Telangana State. Also, tracheal/choanal cleft swabs from live birds were collected from poultry flocks from 7 Indian states as per the guidelines of the Institutional Animal Ethics Committee (IAEC) during the period from March 2013 to February 2014. None of the flocks were vaccinated against M. synoviae. The activities were carried out at Avian Health Lab, ICAR-Directorate of Poultry Research, Hyderabad.  M. synoviae  was isolated in Frey’s medium as described by Kleven (2003) and DNA was isolated using QIAamp, DNA mini kit (QIAGEN, Germany)  and were screened for the presence of M. synoviae using MSF-MSR primer pair (International Office of Epizootics and Committee2008). A total of 817 swab samples were screened by PCR and from positive samples, 19 field isolates (details are given in Table 1) from 5 states and from carcasses showing respiratory lesions, salpingitis and arthritis were selected for gene targeted sequence analysis.
 

 
PCR amplification, sequencing, phylogenetic analysis and typing
 
The vlhA gene was amplified by PCR as per El Gazzar et al., (2012) and was sequenced. The sequences of the single copy conserved region at the 5’ end of vlhA gene from all the isolates and reference strains were trimmed from nucleotide 24 to 376 of the sequence of reference strain WVU1853 (Genbank accession number KC832824) and was used for phylogenetic analysis. The details of the reference strains used are given in the Table 2. The nucleotide sequence editing was accomplished using DNASTAR (Madison, WI). Evolutionary analyses were conducted in MEGA X (Kumar et al., 2018). Molecular Phylogenetic tree was constructed using the Maximum Likelihood method based on the Tamura-Nei model (Tamura and Nei, 1993). The bootstrap consensus tree inferred from 1000 replicates was taken to represent the evolutionary history of the taxa analyzed. Isolates were typed and subtyped based on nucleotide and corresponding amino acid insertion/deletion in the proline-rich repeat (PRR) region and the nucleotide polymorphisms of the RIII region of the 52  region of the vlhA (Bencina et al., 2001; Hammond et al., 2009; Limpavithayakul et al., 2016).
 

Partial sequence analysis of vlhA genes of field isolates showed the presence of variations and the grouping based on 100% similarity placed the isolates into 7 sequence types (Table 1). Nucleotide sequence similarity between these sequence types varied from 92.8% to 98.3%. One reason for the occurrence of variation can be that Mycoplasma species have a high percentage of mutations when compared to other bacteria (Woese et al., 1985; Heldtander et al., 2001). Phylogenetic tree was constructed using nucleotide sequences of 19 field isolates and 12 reference strains of M. synoviae. Phylogenetic analysis showed clustering of field isolates in 4 groups (Fig 1).  The cluster I included sequence types 1 and 2 (Hyderabad) and one reference strain SP 267 from Spain. The cluster II consisted of sequence types 3 and 4 from Kolkata and Bengaluru respectively. Cluster III included one novel field isolate MGS 482, which had only 95.2-98.3% similarity with other 18 field isolates, but had 99.1% similarity with MS H vaccine strain. The cluster IV had sequence types5 (MGS 996 from Hyderabad, 1336 and 1342 from Namakkal) and 6 (MGS 13B from Bengaluru). They had 100% identity with that of reference strain ULB 925KF (Slovenia). By performing phylogenetic analysis of the vlhA gene, we observed that most of the field isolates clustered independently from foreign strains except for sequence type 5 which was 100% identical to ULB 925KF. Hundred percent sequence similarity with foreign strains suggest that they may have originated from a common global strain. Identical types 5 and 6 were obtained from poultry farms of Telangana, Tamil Nadu and Karnataka, three important states for commercial poultry production in India. The culling and replacement of poultry flocks and transport of birds for meat purpose may be an important reason for the prevalence of the same strain in the southern states as suggested by Buim et al., (2010).
 

       
There was apparent deletion of large DNA segments, so that the sequence types in same cluster were identical except for the region of deletion. For example, in cluster II, sequence type 3 and 4 are 100% identical, but type 4 has a deletion of 60 nucleotides and in cluster III as compared to type 3, sequence type 5 and 6 are 100% identical, but type 6 has a deletion of 48 nucleotides. It could be due to that both originated from a single strain and one had undergone deletions due to passage in field or both types entirely two different strains. Similar deletions were described by Hammond et al., (2009) in isolates from a single farm. These insertions or deletions in vlhA gene observed in the isolates may be also be related to pathogenicity (Bencina et al., 2001).
       
Insertions or deletions in the PRR and nucleotide polymorhysms in RIII region are highly useful for epidemiological analysis of M. synoviae isolates (Bencina et al., 2001; Limpavithayakul et al., 2016). In our study, the 19 field isolates were classified into 3 genotypes C (n=8), E (n=6) and L (n=5) based on the length of the PRR (Table 1). Based on the point mutations in the RIII region, the subtypes identified were C1 (n=3) and C3 (n=1) and other subtypes were novel. Alignment of the amino acid sequence of the PRR is given in Fig 2. All the types encoded a conserved signal peptide sequence (IAPAVIAIS). The first eight AAs of their predicted MSPBs had an identical sequence (CGDQTPAP). Isolates from Hyderabad belonged to types L, C1 and E. Isolates from Bengaluru, Kolkata and Namakkal belonged to types E, C and C1 respectively. Types E, L and C had deletions of 19, 3 and 6 amino acids respectively as compared to the reference sequence of strain WVU 1853. Bencina et al., (2001) described that the highly invasive strain K1968 is of type B, having a long PRR and the length of PRR can be associated with invasiveness of the isolate. In the present study most of the isolates were from birds with only respiratory lesions or symptoms. Synovitis cases were very rare and the isolate from synovitis (MGS996) was found to be of subtype C1. Sequence analysis showed that this isolate was 100% identical to strain ULB925 (AF314228) which was also isolated from chicken joint (Bencina et al., 2001). Multiple types (C1, E and L) were found in one particular farm and all other farms were affected by only single type.
 

The frequency of M. synoviae infection has been found increasing worldwide as compared to Mycoplasma gallisepticum, the most important avian Mycoplasma (Landman, 2014). The differentiation of prevalent strains is necessary for tracing the spread of the pathogen. This study provides the first molecular typing and subtyping of Indian M. synoviae isolates. The genetic analysis showed that most of the farms had single sequence type, but one particular sequence type was found in three major states for commercial poultry production, which could be due to interstate transport of birds or chicks. Also there was 100% identity between few Indian isolates and a foreign strain. The isolates from different clinical conditions (salpingitis, arthritis and air sacculitis) did not show any variation in the sequences. Variation was also evident from the length of the vlhA gene region encoding PRR. The prevalence of M. synoviae in Indian poultry flocks along with the results of the present study insists the urgent need for implementation of effective prophylactic measures, such as application of proper biosecurity measures and vaccination to control the M. synoviae infection in Indian poultry flocks.

The authors are grateful to Director, ICAR-Directorate of Poultry research (DPR) for providing all facilities necessary to conduct this research. Authors are also thankful to staff in the Avian Health Lab, DPR for their help in carrying out this research work.

There is no conflict of interest among authors and there is no financial or personal relationship between authors and other organizations which that might inappropriately influence or bias their work.