Development and evaluation of an improved diagnostic PCR for Mycoplasma synoviae using primers located in the haemagglutinin encoding gene vlhA and its value for strain typing.

Abstract

Using published primers, detection of Mycoplasma synoviae and strain identification using the vlhA gene sequence was attempted. However, of 21 M. synoviae strains examined, three could not be amplified, so a new reverse primer was designed with a target in the conserved region of the vlhA gene. This allowed all 21 M. synoviae strains, a further nine strains and also material from 11 swab samples from M. synoviae-positive birds, to produce a PCR product, suggesting that the method could also be suitable for clinical specimens. The protocol was then tested on the type strains of M. synoviae and the other 22 recognised avian Mycoplasma species, with amplification of M. synoviae only. Further testing demonstrated that this PCR was equally or more sensitive than other PCR tests used to detect M. synoviae. Subsequent DNA sequence analysis of the PCR product based on percent similarity and evolutionary relationship appeared to be a useful tool for strain differentiation.

Full text

Page 1

Accepted Manuscript
Title: Development, evaluation of an improved diagnostic
PCR for Mycoplasma synoviae using primers located in the
haemagglutinin encoding gene vlhA and its value for strain
typing
Authors: P.P. Hammond, A.S. Ram´ ırez, C.J. Morrow, J.M.
Bradbury
PII:
DOI:
Reference:
S0378-1135(08)00486-0
doi:10.1016/j.vetmic.2008.10.011
VETMIC 4239
To appear in:
VETMIC
Received date:
Revised date:
Accepted date:
11-8-2008
6-10-2008
13-10-2008
Pleasecitethisarticleas:Hammond,P.P.,Ram´ ırez,A.S.,Morrow,C.J.,Bradbury,J.M.,
Development, evaluation of an improved diagnostic PCR for Mycoplasma synoviae
usingprimerslocatedinthehaemagglutininencodinggenevlhAanditsvalueforstrain
typing, Veterinary Microbiology (2008), doi:10.1016/j.vetmic.2008.10.011
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
Themanuscriptwillundergocopyediting,typesetting,andreviewoftheresultingproof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.
peer-00532523, version 1 - 4 Nov 2010
Author manuscript, published in "Veterinary Microbiology 136, 1-2 (2009) 61"
DOI : 10.1016/j.vetmic.2008.10.011

Page 2

Page 1 of 23
Accepted Manuscript
* Corresponding author at: Department of Veterinary Pathology, University of Liverpool
Development and evaluation of an improved diagnostic PCR for
1
Mycoplasma synoviae using primers located in the haemagglutinin encoding
2
gene vlhA and its value for strain typing
3
4
P.P. Hammond1, A.S. Ramírez2, ,C.J. Morrow3 and J.M. Bradbury2*
5
6
1Royal Veterinary College, Hawkshead House, Hawkhead Lane, North Mymms, Hatfield, 7
Herts. AL9 7TA. UK; present address: Crowshall Veterinary Services8
1 Crows Hall Lane, Attleborough, Norfolk NR17 1AD9
2Department of Veterinary Pathology, University of Liverpool Leahurst Campus, Neston, 10
CH64 7TE, UK; present address: Unit of Epidemiology and Preventive Medicine, Veterinary 11
Faculty, University of Las Palmas de G.C. Trasmontaña s/n, Arucas. 35416. Las Palmas, 12
Spain13
3Aviagen Limited, Newbridge, Midlothian, EH28 8SZ, Scotland, UK; present address: 14
Bioproperties Pty Ltd, 36 Charter St., Ringwood, Victoria 3134 Australia 15
16
17
18
19
20
21
Leahurst Campus, Neston, CH64 7TE, UK22
E.mail: jmb41@liverpool.ac.uk (J.M> Bradbury)
Tel: + 44 (0)151 794 6115, Fax: + 44 23
(0)151 794 6005 24
25
Manuscript
peer-00532523, version 1 - 4 Nov 2010

Page 3

Page 2 of 23
Accepted Manuscript
Abstract1
Using published primers, detection of Mycoplasma synoviae and strain identification 2
using the vlhA gene sequence was attempted. However, of 21 M. synoviae strains examined, 3
three could not be amplified, so a new reverse primer was designed with a target in the 4
conserved region of the vlhA gene. This allowed all 21 M. synoviae strains, a further nine 5
strains and also material from 11 swab samples from M. synoviae-positive birds, to produce a 6
PCR product, suggesting that the method could also be suitable for clinical specimens. The 7
protocol was then tested on the type strains of M. synoviae and the other 22 recognised avian 8
Mycoplasma species, with amplification of M. synoviae only. Further testing demonstrated 9
that this PCR was equally or more sensitive than other PCR tests used to detect M. synoviae. 10
Subsequent DNA sequence analysis of the PCR product based on percent similarity and 11
evolutionary relationship appeared to be a useful tool for strain differentiation. 12
13
Keywords: Mollicutes, Mycoplasma, Mycoplasma synoviae, diagnosis, polymerase chain 14
reaction, vlhA.15
16
peer-00532523, version 1 - 4 Nov 2010

Page 4

Page 3 of 23
Accepted Manuscript
properties suggested it as a suitable target for molecular strain typing (Benčina et al., 2001).
1. Introduction1
Mycoplasma synoviae is a chicken and turkey pathogen causing economic loss in 2
intensive production. Reliable, rapid diagnosis is needed to prevent dissemination of infection 3
and has traditionally been achieved by serological screening for antibodies or by culture of the 4
mycoplasma. Serological screening is still in widespread use but may not detect subclinical5
M. synoviae infections, and monitoring programmes that depend solely on detecting 6
seroconversion may be inadequate (Ewing et al., 1998; Kleven et al., 2001). Culture can be 7
costly and time consuming, and can also be inconclusive (Ewing et al., 1998) so PCR-based 8
tests are now routinely used for detecting pathogenic avian mycoplasmas. Some PCRs are9
based on the 16S rRNA gene (Garcia et al, 1996; Lauerman et al., 1993), some are in-house 10
tests (e.g. Lauerman, 1998), and others are commercially-produced kits. 11
M. synoviae PCRs have also been based on the vlhA haemagglutinin (HA) gene 12
(Benčina et al., 2001; Hong et al., 2004). Ben Abdelmoumen Mardassi et al. (2005) used a 13
duplex PCR targeting the HA genes to differentiate M. synoviae and M. gallisepticum while14
Jeffery et al. (2007) typed M. synoviae strains by single-strand conformation polymorphism or 15
high resolution melting curve analysis of the conserved 5’ end of the vlhA gene. The gene 16
product is an abundant immunodominant surface lipoprotein with a conserved and variable 17
region (Noormohammadi et al., 2000). The N-terminal (conserved) region exists as a single 18
copy whose sequence varies among strains. The gene includes tandem repeats that encode19
proline rich repeats (PRR) and also a region that is highly polymorphic (RIII) (Fig. 1). These 20
21
Our aim was to test the PCR assay of Benčina et al. (2001) on M. synoviae strains 22
from different hosts, locations and years. Since some of our strains were not amplified an23
alternative reverse primer, located in the conserved region of the vlhA gene (Fig. 1) was used. 24
peer-00532523, version 1 - 4 Nov 2010

Page 5

Page 4 of 23
Accepted Manuscript
positive chickens were also included in these tests. The five swabs from year 2003 had proved
The specificity and sensitivity of this PCR and its ability to detect M. synoviae in field isolates1
and swabs were established and sequence analysis was evaluated for strain typing.2
3
2. Material and methods4
5
2.1. Organisms and culture conditions6
7
The type strains of avian Mycoplasma species used were as follows: M. anatis 1340T, M. 8
anseris 1219T, M. buteonis BbT2gT, M. cloacale 383T, M. columbinasale 694T, M. 9
columbinum MMP/1T, M. columborale MMP/4T, M. corogypsi Bv1T, M. falconis H/T1T, M. 10
gallinaceum DDT, M. gallinarum PG16T, M. gallisepticum PG31T, M. gallopavonis WR1T, M. 11
glycophilum 486T, M. gypis B1/T1T, M. imitans 4229T, M. iners PG30T, M. iowae 695T, M. 12
lipofaciens R171T, M. meleagridis 17529T, M. pullorum CKKT, M. sturni UCMFT and M. 13
synoviae WVU1853T. The details of the M. synoviae field strains and swabs are given in 14
Table 1.15
The mycoplasmas were cloned once by filtration and their identity was confirmed by 16
the indirect fluorescent antibody test (Rosendal and Black, 1972). Then they were cultivated 17
in mycoplasma broth (Bradbury, 1977) at 37 ºC for between one and five days, in a CO2 rich 18
atmosphere. 19
Eleven swabs collected earlier from the trachea or choanal cleft of M. synoviae-20
21
to be IDEXX PCR positive but culture negative. 22
23
2.2. Sample preparation24
25
peer-00532523, version 1 - 4 Nov 2010

Page 6

Page 5 of 23
Accepted Manuscript
The specificity and sensitivity studies were carried out using the vlhAF-vlhAR2 PCR
DNA extraction was done using Chelex resin for broth cultures and heat treatment for 1
swabs as described elsewhere (Ramírez et al., 2006).2
3
2.3. Testing the primer pairs4
5
The primers used in the reaction are shown in Table 2. PCR amplification using vlhAF 6
and vlhAR1 primers was performed on 21 isolates. The reaction mixture contained 0.5 μM of 7
each primer (vlhAF and vlhAR1); 0.2mM dNTPs; 2.5 mM MgCl2; 1X reaction buffer8
(Abgene, AB-0192); 0.5 U Taq DNA polymerase (Abgene) made up to 49.5 μl with water9
(Sigma W4502). 10
DNA template (1 μl) was added per tube and Sigma water (1 μl) was added to another 11
tube to act as a negative control. Mineral oil (50 μl) was layered on the surface of each12
mixture. The thermal cycle, using Perkin Elmer PCR machine, was 94 ºC for 5 min; 36 cycles 13
of 94ºC for 1 min, 52ºC for 1 min and 72ºC for 1 min, then a final extension step of 72ºC for 2 14
min.15
Following the failure of these primers to amplify three of these strains, the combination 16
of vlhAF and vlhAR2 was tested on the same 21 strains and also on nine others and on the 11 17
swabs indicated in Table 1 following the same PCR protocol as before.18
19
2.4. Assay specificity and sensitivity20
21
22
assay. Specificity was tested with the type strains of all the 23 avian Mycoplasma species 23
listed above. Sensitivity was determined by analysing serial 10-fold dilutions to 10-9 of broth 24
cultures from the M. synoviae type strain (WVU1853T) and from a field strain (B155/02). 25
Viable counts were conducted for each culture and the numbers of colony-forming units 26
peer-00532523, version 1 - 4 Nov 2010

Page 7

Page 6 of 23
Accepted Manuscript
amino acid sequences. Nucleotide and amino acid sequences of all 41 M. synoviae strains (30
(CFU) per 100μl were calculated (Bradbury, 1977). The DNA was measured using a1
fluorometer (DyNA QuantTM 200, Hoefer, Pharmacia Biotech Inc.), following the 2
manufacturer’s instructions. PCR was carried out on each dilution.3
4
2.5. Comparison of the vlhAF-vlhAR2 PCR with three other diagnostic M. synoviae PCRs 5
6
7To compare the results of our PCR assay with other described methods, three other M.
synoviae PCR detection protocols were conducted with the type strain (WVU1853T) and a 8
field strain (B155/02). These were the Lauerman method (Lauerman, 1998), a modified 9
Lauerman procedure (Ramírez et. al., 2006) in which DNA extraction incorporated diethyl 10
pyrocarbonate (DEP) and a commercial kit (FlockChek® Mycoplasma synoviae DNA probe 11
kit, IDEXX Laboratories Inc., Westbrook, Maine, USA). 12
13
2.6. Sequencing14
15
Sequencing was conducted by Imperial College, London, and sequences were 16
determined for both strands of DNA. The resulting chromatograms were examined in 17
Chromas (version 1.45; School of Health Science, Griffith University, Australia). The 18
forward sequence and reverse complemented sequences were compared to produce a 19
consensus sequence using Generunner (version 3.05; Hastings Software, Inc.). Bioedit 7.0.0 20
(Hall, 1999) program was used for aligning all the sequences manually, and for deducing the 21
22
from cultures and 11 from swabs) were trimmed from nucleotide 49 to 421 corresponding to 23
the numbering of Benčina et al. (2001) for strain WVU1853T. They were assigned in the 24
Genbank with accession numbers AM998371, AJ580981 to AJ580983, AJ580985, 25
AJ58098187 to AJ580989, AJ58091 and FM164342 to FM164373.26
peer-00532523, version 1 - 4 Nov 2010

Page 8

Page 7 of 23
Accepted Manuscript
the vlhA gene PRR-coding region and the subtypes according to the sequence of the RIII
1
2.7. Percent DNA similarity2
3
The percent similarity of all the M. synoviae vlhA gene sequences was determined using 4
the program Bioedit version 7.0.0 (Hall, 1999). The strains were then grouped into those 5
showing 100% similarity and a representative of each group was selected for subsequent 6
phylogenetic analysis. The percent similarity between the representatives was calculated 7
including and omitting insertions/deletions.8
9
2.8. Analysis of sequences10
11
Dendrograms were constructed from alignments of the representative strains by the 12
neighbour-joining method (Saitou and Nei, 1987) with 1000 bootstrap replicate analyses 13
(Felsenstein, 1985), using the Molecular Evolutionary Genetic Analysis (MEGA) software 14
(Center for Evolutionary Functional Genomics, Tempe, AZ, USA; 15
http://www.megasoftware.net) for sequence alignments (Tamura et al., 2007). The 16
evolutionary distances were computed, with the same program, using the Maximum 17
Composite Likelihood method (Tamura et al., 2004).18
The method described by Benčina et al. (2001) was used initially to classify the M. 19
synoviae strains into types and subtypes. The types were assigned according to the length of 20
21
region. However after further examination of the percent similarity and of the relationships 22
revealed by the dendrogram a modified grouping was derived. 23
24
3. Results25
peer-00532523, version 1 - 4 Nov 2010

Page 9

Page 8 of 23
Accepted Manuscript
The next most sensitive was the modified Lauerman method which detected M. synoviae in
1
3.1. Testing the primer pairs2
3
4
Using the vlhAF-vlhAR1 primer combination three M. synoviae isolates (B27/00,
B142/02 and B154/02) failed to produce the expected fragment of around 621 bp but with 5
primer set vlhAF-vlhAR2 a PCR product was obtained with all 41 samples tested. This 6
included the previous three failed strains and the DNA extracted from the 11 swabs. The 7
amplicon sizes varied between 315 and 372 bp (Fig. 2).8
9
3.2. Specificity and sensitivity of the vlhAF-vlhAR2 M. synoviae PCR and its comparison with 10
three other diagnostic M. synoviae PCRs 11
12
13The vlhAF-vlhAR2 PCR was specific, producing an amplicon only with M. synoviae
template and with no amplified PCR products from the other 22 avian Mycoplasma species 14
tested (data not shown). 15
The PCR detected DNA template from strains WVU1853T and B115/02 up to the 16
dilution 10-5, which corresponded to an amount of DNA close to 1 pg for both strains (data 17
not shown). No colonies grew at this dilution. For the type strain the number of colony- 18
forming units per 100 µl was only 1 at a dilution of 10-3, while for the field strain eight 19
colonies were countable at a dilution of 10-4. The PCR was more sensitive than the other three 20
methods as none of them allowed detection of DNA from either strain at a dilution of 10-5. 21
22
both strains to dilutions of 10-4. The commercial kit detected M. synoviae DNA to a dilution 23
of 10-3 in both strains while the unmodified Lauerman protocol detected the type strain only at 24
10-1 dilution and the field strain at 10-2. 25
26
peer-00532523, version 1 - 4 Nov 2010

Page 10

Page 9 of 23
Accepted Manuscript
dendrogram was derived with the representative strain of each group (Fig. 4) that revealed the
3.3. Sequence analysis and strain typing1
2
Table 1 shows the sizes of amplicons and PRRs in nucleotides and amino acids 3
respectively, while Fig. 3 shows the point mutations and insertions/deletions in the nucleotide 4
alignments of the group representatives. The corresponding amino acid alignments are shown 5
in Supplementary Fig. 1.6
Strain typing was initially based on the size of the PRR region and point mutations of 7
the RIII region (Benčina et al., 2001). Using the length of the PRR the strains were assigned8
to Benčina’s types A to E. Most strains were assigned to type C (Table 1). There were none 9
belonging to type B but a further type F was added to accommodate two strains whose PRR 10
length was 108 nt (36 amino acids). After this, the strains were subdivided numerically based 11
on the RIII region and most of those in type C fell into subtype C3. However, while 21 of the 12
23 strains of this subtype had 100% similarity, the two remaining (identical) strains had only 13
99.6% similarity with the others. Percent similarity between the representatives of each group 14
with and without insertions/deletions is presented as Supplementary Table 1.15
Grouping based on 100% similarity placed the strains into 12 groups (Table 1). It was 16
apparent some groups had identical vlhA gene sequences except for a single 17
insertion/deletion, as for example strains B11/85 and J26/85 in Supplementary Fig. 2. The 18
same occurred between strains B142/02 and B95/04/K261, between strains B91/96/798 and 19
B133/99/05, and between strains B48/05/1 and B38/96/1704. Based on that information a 20
21
relationship between these groups.22
When groups 2 to 12 were compared with group 1 which contained the M. synoviae23
type strain WVU1853T (Table 1 and Fig. 3) all the other strains had a deletion at nt 112 24
corresponding to nt 148 of Benčina et al. (2001). Of these, 32 strains (the members of groups 25
peer-00532523, version 1 - 4 Nov 2010

Page 11

Page 10 of 23
Accepted Manuscript
Benčina et al. (2001) did not amplify three of our field isolates. All emanated from Hungary
2 to 7) had a deletion of 18 nt/6 aa and seven (the members of groups 8 to 12) a deletion of 57 1
nt/19 aa, however only five strains (the members of groups 2, 3 and 8) had an insertion 2
between nt 64 and 65 (nt 111 and 112 of Benčina et al. 2001).3
UK pheasant M. synoviae strains comprised a unique group (group 2 in Table 1). 4
These two strains plus a strain isolated from an Italian turkey (B38/96/1704) and two strains 5
from a UK chicken company (B48/05 and B57/05), contain the same insertion (5’ 6
ACTCCAACACCT 3’) in the PRR region (Fig. 3). Strains B133/99/5 and B133/99/12 which 7
were from the same Hungarian farm were homologous except that B133/99/5 exhibited a 578
nt deletion. Strains J15/ 85 and J26/85 from the mid 1980s exhibited a 57 nt deletion and were 9
identical except for one base change from G to A at nt 244 in J15/85. B7/86 and B11/85 10
isolated in the UK around the same time are identical to J26/85 except there was only an 18 11
bp deletion in the first two compared to a 57 bp deletion.12
13
Discussion 14
15
PCR assays for detection of M. synoviae have been reported earlier (Garcia et al, 1996; 16
Lauerman et al., 1993 and 1998; Benčina et al., 2001; Hong et al., 2004, Ben Abdelmoumen 17
Mardassi et al, 2005; Ramírez et al., 2006; Jeffery et al., 2007). Four of them target the 18
haemagglutinin-encoding vlhA gene (Benčina et al., 2001; Hong et al., 2004; Ben 19
Abdelmoumen Mardassi et al, 2005; Jeffery et al., 2007). In our hands the primers used by 20
21
and, after amplification with the new reverse primer, could be placed in Benčina subtype C3
22
and our groups 5 and 6. This lack of amplification can be explained by the fact that the 23
reverse primer of these authors, VlhAR1, targets the variable region of the vlhA gene (Fig. 1). 24
Hong et al. (2004) described a PCR with the PRR region as the target, excluding the RIII 25
peer-00532523, version 1 - 4 Nov 2010

Page 12

Page 11 of 23
Accepted Manuscript
When the four M. synoviae diagnostic PCRs were compared, the vlhAF-vlhAR2 PCR
region. Our primer vlhAR2 was designed to target the 3’ end of the conserved region, 1
including the RIII region. With the new reverse primer the DNA from all 21 isolates plus nine 2
more was amplified. 3
The results showed good specificity. This was demonstrated by no reactions with any of 4
22 other avian Mycoplasma species, including nine that fall within the M. synoviae cluster of 5
the 16S rDNA-based hominis phylogenetic group (Pettersson et al., 2000), and when used6
successfully on DNA extracted from 11 swabs, no nonspecific bands appeared. Thus our PCR 7
assay appears also to have sufficient sensitivity and specificity for clinical use. 8
The sensitivity was as good, if not better than several other PCR tests. The method 9
detected M. synoviae DNA of two strains when no colonies could be cultivated, while Ben 10
Abdelmoumen Mardassi et al. (2005) needed at least 90 CFUs for detection of M. synoviae11
with their duplex PCR assay based in the haemagglutinin genes. Hong et al. (2004) could 12
detect 4.7 X 102colour-changing units/ml and Lauerman et al. (1993) found a sensitivity of 13
100 CFUs for their M. synoviae PCR based on the 16S rRNA gene. The amount of DNA that 14
our PCR can detect is similar to the 1pg detected by a Mycoplasma synoviae PCR included in 15
a multiplex PCR described by Wang et al. (1997) and the PCR described by Ramírez et al. 16
(2006). These results agree with Razin’s observation that “sensitivity of PCR assays is usually 17
in the order of magnitude of a few femtograms of mycoplasmal DNA. When translated into 18
numbers of organisms, 1fg of mycoplasmal DNA is approximately equivalent to the genomic 19
DNA of a single mycoplasma cell, based on a genome size of about 1000 kb” (Razin, 2002). 20
21
described in this study performed best, detecting a lower amount of DNA than the other 22
methods. This could be due to the use of the resin Chelex in the DNA extraction, because23
similar results were seen by Ramírez et al. (2006) when the same method was used for 24
extracting DNA for a PCR based on the 16S-23S rRNA intergenic spacer region. 25
peer-00532523, version 1 - 4 Nov 2010

Page 13

Page 12 of 23
Accepted Manuscript
Hong et al. (2004) combined PCR and sequencing of the PRR repeat region of the vlhA gene
PCR and partial vlhA gene sequencing may also be useful in strain identification. 1
Benčina et al., (2001) compared the 5′ end haemagglutinin (vlhA) gene sequences of 30 M. 2
synoviae strains from chickens and turkeys and demonstrated 11 different types of vlhA3
sequences, while Hong et al. (2004) found 14 different groups among 43 strains from the 4
same species. The majority of strains in both studies came from the USA and Slovenia. 5
Jeffery et al. (2007) examined 35 different isolates from Australia and the USA (from 6
chickens and one turkey) and revealed 10 distinct profiles by single-strand length 7
polymorphism. These corresponded to the 10 genotypes derived by high-resolution melting-8
curve analysis. The 41 strains in our study were mainly from the UK and Hungary and 9
comprised chicken and turkey strains plus two from pheasants. They were assigned to 12 10
groups and showed considerable overlap with the groupings of Benčina et al. (2001) although 11
four strains, including the two pheasant strains did not fit into any of the Benčina types. 12
The vlhA gene has considerable divergence at the 3′ end of the gene due to 13
recombination with pseudogenes (presumably as a mechanism to evade the host immune 14
response), but relative conservation at the 5′ end (Noormohammadi et al., 2000). The 5′ vlhA 15
region is present in the M. synoviae genome as a single copy and does not change its sequence 16
in clonal populations of M. synoviae (Noormohammadi et al., 2000). This observation is 17
crucial to strain identification as downstream of this region the sequence can change even in 18
clonal populations of M. synoviae and therefore it cannot be considered a conserved sequence 19
that characterises individual strains (Noormohammadi et al, 2000). The test evaluated by 20
21
to detect and type M. synoviae strains without the need for isolation. However, it excluded the 22
RIII region (positions 343/400), which Benčina et al. (2001) found very useful for subtyping 23
M. synoviae strains. The PRR region is useful mainly due to insertions/deletions and the RIII 24
region is useful because of polymorphisms and our findings have confirmed this, as have 25
peer-00532523, version 1 - 4 Nov 2010

Page 14

Page 13 of 23
Accepted Manuscript
M. synoviae strains and the tree showed which species were more closely related. For
those of Jeffrey et al. (2007) who, by chance, designed a reverse primer almost identical to 1
ours.2
Nucleotide and amino acid sequence analyses of M. synoviae strains support the 3
observation that they are useful for strain identification and as a preliminary typing method. 4
Benčina et al. (2001) suggested that the polymorphisms in 5’ end of the vlhA gene might be 5
very useful for epidemiological analysis of M. synoviae isolates and it is of interest that the 6
two strains from UK pheasants comprised a unique group, (named group 2 in Table 1). 7
Insertions/deletions were observed in all the strains examined and may be related to 8
pathogenicity (Benčina et al., 2001). Although isolates B133/99/5 and B133/99/12 were from 9
the same farm, B133/99/5 exhibited a 57 nt deletion. These could be the same strain, with 10
strain B133/99/5 having undergone a deletion during passage on the farm but it is also 11
possible to have more than one strain on a farm.12
Isolates from the mid 1980s demonstrated greater similarity with the type strain 13
WVU1853T than the present isolates, apart from strain B95/04/K4412 whose sequence was 14
exactly the same as the M. synoviae type strain. These two strains from Germany were from 15
the same sender but no further information about their origin was available although we 16
assumed both to be field strains. Strains included in group 5 (Table 1) appeared in isolates 17
from Hungary, Holland and the UK and confirmed earlier observations by Hazel (PhD, 2002)18
using RAPD analysis.19
DNA sequence analysis was used to determine the phylogenetic relationships among the 20
21
example using the typing system of Benčina et al. (2001) it appears that strains J26/85 and 22
J15/85 are the same type but different to B11/85 whereas the phylogenetic tree reveals that 23
J26/85 is more closely related to B11/85 than to J15/85. This is reinforced by the DNA 24
homologies (Supplementary table 1) because comparison of J26/85 with B11/85 with the 25
peer-00532523, version 1 - 4 Nov 2010

Page 15

Page 14 of 23
Accepted Manuscript
Molecular basis of the length variation in the N-terminal part of Mycoplasma synoviae
insertion/deletion the similarity is 88.9% but without the insertion/deletion is 100%. In 1
contrast, when comparing J26/85 with J15/85 the homologies are 99.6% with or without the 2
insertion/deletion.3
This study confirms that DNA extracted from cultures as well as directly from swabs 4
can be successfully amplified with our primers and used in the diagnosis and strain typing of 5
M. synoviae. It also confirms the potential value of strain typing for epidemiological purposes 6
and suggests that the addition of phylogenetic studies is essential to understand the true 7
relationships between strains.8
9
Acknowledgements 10
The authors wish to thank to Aviagen Ltd for sponsoring PPH and Christine Yavari and 11
Cynthia Dare for their helpful assistance in the laboratory. 12
13
References14
15
Ben Abdelmoumen Mardassi, B., Ben Mohamed, R., Gueriri, I., Boughattas, S., Mlik, B., 16
2005. Duplex PCR to differentiate between Mycoplasma synoviae and Mycoplasma 17
gallisepticum on the basis of conserved species-specific sequences of their hemagglutinin 18
genes. J. Clin. Microbiol. 43, 948-958. 19
Benčina, D., Drobnič-Valič, M., Horvat, S., Narat, M., Kleven, S.H., Dovč, P., 2001. 20
21
hemagglutinin. FEMS Microbiol. Lett. 203, 115-123. 22
Bradbury, J.M., 1977. Rapid biochemical tests for characterisation of the Mycoplasmatales. J. 23
Clin. Microbiol. 5, 531-534.24
peer-00532523, version 1 - 4 Nov 2010