- Email: [email protected]
Development of a single multi-locus sequence typing scheme for Taylorella equigenitalis and Taylorella asinigenitalis
Veterinary Microbiology 167 (2013) 609–618
Contents lists available at ScienceDirect
Veterinary Microbiology journal homepage: www.elsevier.com/locate/vetmic
Development of a single multi-locus sequence typing scheme for Taylorella equigenitalis and Taylorella asinigenitalis Fabien Duquesne a,*, Laurent He´bert a, Marie-France Breuil a, Motoo Matsuda b, Claire Laugier a, Sandrine Petry a a
ANSES, Dozule´ Laboratory for Equine Diseases, Bacteriology and Parasitology Unit, 14430 Dozule´, France Laboratory of Molecular Biology, Graduate School of Environmental Health Sciences, Azabu University, Fuchinobe 1-17-71, Chuo-ku, Sagamihara 252-5201, Japan
b
A R T I C L E I N F O
A B S T R A C T
Article history: Received 10 April 2013 Received in revised form 12 September 2013 Accepted 13 September 2013
We describe here the development of a multilocus sequence typing (MLST) scheme for Taylorella equigenitalis, the causative agent of contagious equine metritis (CEM), and Taylorella asinigenitalis, a nonpathogenic bacterium. MLST was performed on a set of 163 strains collected in several countries over 35 years (1977–2012). The MLST data were analyzed using START2, MEGA 5.05 and eBURST, and can be accessed at http:// pubmlst.org/taylorella/. Our results revealed a clonal population with 39 sequence types (ST) and no common ST between the two Taylorella species. The eBURST analysis grouped the 27 T. equigenitalis STs into four clonal complexes (CC1–4) and five unlinked STs. The 12 T. asinigenitalis STs were grouped into three clonal complexes (CC5–7) and five unlinked STs, among which CC1 (68.1% of the 113 T. equigenitalis) and CC5 (58.0% of the 50 T. asinigenitalis) were dominants. The CC1, still in circulation in France, contains isolates from the first CEM outbreaks that simultaneously emerged in several countries in the late 1970s. The emergence in different countries (e.g. France, Japan, and United Arab Emirates) of STs without any genetic relationship to CC1 suggests the existence of a natural worldwide reservoir that remains to be identified. T. asinigenitalis appears to behave same way since the American, Swedish and French isolates have unrelated STs. This first Taylorella sp. MLST is a powerful tool for further epidemiological investigations and population biology studies of the Taylorella genus. ß 2013 Elsevier B.V. All rights reserved.
Keywords: Taylorella genus MLST Contagious equine metritis
1. Introduction Taylorella equigenitalis and Taylorella asinigenitalis, small Gram-negative microaerophilic coccobacilli, are the only two species of the Taylorella genus classified in the Alcaligenaceae family (Jang et al., 2001; Sugimoto et al., 1983). T. equigenitalis is the causative agent of contagious equine metritis (CEM), a sexually-transmitted infection of horses first reported in 1977 (Crowhurst, 1977; Timoney
* Corresponding author. Tel.: +33 2 31 79 22 76; fax: +33 2 31 39 21 37. E-mail address: [email protected] (F. Duquesne). 0378-1135/$ – see front matter ß 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.vetmic.2013.09.016
et al., 1977). CEM is currently present in many parts of the world and various horse breeds (Timoney, 2011), and is notifiable to the World Organisation for Animal Health (OIE, 2012). Whereas stallions are typically asymptomatic carriers (Timoney, 2011), 30–40% of infected mares show abundant mucopurulent vaginal discharge and variable degrees of vaginitis, endometritis and cervicitis, which leads to temporary infertility (Wakeley et al., 2006). A proportion of mares may become long-time symptomless carriers and, together with infected stallions, may spread infection during mating (Luddy and Kutzler, 2010). T. asinigenitalis is currently considered a nonpathogenic bacterium, despite clinical signs of metritis and cervicitis
610
F. Duquesne et al. / Veterinary Microbiology 167 (2013) 609–618
in mares following experimental intra-uterine infusion with field isolates of T. asinigenitalis (Katz et al., 2000). T. asinigenitalis was first reported on late 1997–early 1998 in the states of California and Kentucky (Katz et al., 2000). It is characterized by slight differences from T. equigenitalis regarding colony morphology, growth rates and immunofluorescence characteristics. Due to their high degree of relatedness, it remains difficult to differentiate both Taylorella species only using the conventional methods, which must be associated with species-specific PCR. This difficulty is illustrated by the fact that T. asinigenitalis isolates have been mis-identified as T. equigenitalis (Breuil et al., 2011). Every CEM outbreak (the last major one occurred in the United States (US) in 2008–2010, (Erdman et al., 2011; Timoney, 2011)) highlights the insidious threat of this disease to the horse breeding industry, due to its infectiousness, short-term infertility in mares, and the existence of a carrier state independent of the sex of the horse. Pulsed-field gel electrophoresis (PFGE) is currently the main tool used to genotype CEM isolates for epidemiological analysis, the most recent example being the PFGE genotyping of T. equigenitalis isolated in the US from 1978 to 2010 (Aalsburg and Erdman, 2011). Several other molecular typing tools have also been used including field inversion gel electrophoresis (Bleumink-Pluym et al., 1990), chromosomal DNA fingerprinting (Thoresen et al., 1995) and crossed-field gel electrophoresis (Miyazawa et al., 1995). However, these molecular epidemiological tools are poorly portable because of their index of variation and it is difficult to compare results among laboratories (Maiden et al., 1998). Accordingly, these tools are not well suited for the global epidemiological studies of CEM outbreaks. The international trade in Equidae leads to the need to identify and track the global spread of T. equigenitalis and T. asinigenitalis, and thus investigate the genetic relationship between isolates from different outbreaks and both bacterial species. In this regard, multilocus sequence typing (MLST) seems to be more appropriate to the current need for large-scale epidemiological studies. MLST is based on DNA sequence comparison of internal fragments of housekeeping genes chosen because their products play vital function and are present in all isolates of a given species, and mutations within them are assumed to be neutral (Maiden et al., 1998; Selander et al., 1986). The accumulation of nucleotide changes in housekeeping genes is a relatively slow process and their allelic profile is sufficiently stable over time for the method to be ideal for global epidemiology (Enright and Spratt, 1999). Because MLST is based on nucleotide sequence, it is highly discriminatory and provides unambiguous results that are directly comparable among laboratories via the Internet (Enright et al., 2000). The recent characterization and comparison of complete genome sequences of T. equigenitalis MCE9 (He´bert et al., 2011) and T. asinigenitalis MCE3 (He´bert et al., 2012) now allows the development of such a tool. We describe here the development of an MLST scheme for T. equigenitalis and T. asinigenitalis based on the nucleotide sequences of seven housekeeping genes.
2. Materials and methods 2.1. Bacterial strains As listed in Table 1, a total of 113 T. equigenitalis and 50 T. asinigenitalis strains were used in the present study, including reference strains CIP79.7T, CIP79.44 and CIP 107673T from the Collection Institut Pasteur (CIP; France). This set of isolates is both temporally and geographically diverse with strains collected in six countries (France, US, UK, Japan, Australia, Sweden) over a 35-year period of time (1977–2012). As French reference laboratory for CEM, we have an exhaustive strain collection of the most CEM cases identified in France since 1992. On the basis of known epidemiological data (date, host source and geographic location), we selected 97 T. equigenitalis and 48 T. asinigenitalis to ensure the best representation of French Taylorella population. Among them, 43 strains were isolated from 16 animals that remained CEM-positive after several samplings within a single year or over a period of several years, and 33 strains were isolated from five French farms with at least two animals that remained CEM-positive. To extrapolate our results globally, we completed our French selection by strains from different countries (US, UK, Japan, Australia, and Sweden). All isolates had been previously identified using routine phenotypic tests and confirmed by PCR (Breuil et al., 2011; Duquesne et al., 2007). Full details of all of the strains used in this study are available on the MLST database (http://pubmlst.org/taylorella/). Each strain was inoculated on a ready-to-use heated blood agar enriched with a polyvitamin supplement (AES Chemunex, Combourg, France). Plates were incubated at 37 8C in 7% (v/v) CO2 in air for 48 h (T. equigenitalis) or 72 h (T. asinigenitalis). 2.2. Choice of loci for MLST Initially, 22 potential housekeeping genes were identified by comparing the T. equigenitalis MCE9 (He´bert et al., 2011) and T. asinigenitalis MCE3 (He´bert et al., 2012) genome sequences using the Artemis Comparison Tool (Berriman and Rutherford, 2003; Rutherford et al., 2000) available at http://www.sanger.ac.uk/resources/software/artemis/. On the base of a pilot MLST study performed on 20 Taylorella sp. strains, the set was reduced to seven gene fragments (Table 2) by eliminating fragments that could not be readily amplified from all tested isolates as well as those with close chromosomal locations. The chromosomal locations of these seven housekeeping genes suggest that it is unlikely for any of them to be coinherited in the same recombination event since the minimum distance between two loci is 95 kbp except for the txn gene, which is located 7 kbp away from the gltA gene. 2.3. Amplification and nucleotide sequence determination Genomic DNA of each strain was extracted using the NucleoSpin Tissue kit according to the manufacturer’s protocol (Macherey-Nagel). Amplification/sequencing
F. Duquesne et al. / Veterinary Microbiology 167 (2013) 609–618
611
Table 1 Characteristics of Taylorella isolates and MLST data. Strain (other reference name) 113 T. equigenitalis strains MA01 (KY-188) MA02 MA04 MA05 MA06 MA07 MA14 MCE496 (ATCC35865, CIP79.7T) MCE16 MCE17 MCE26 MCE39 MCE41 MCE42 MCE43 MCE48 MCE51 MCE54 MCE55 MCE60 MCE63 MCE64 MCE68 MCE70 MCE75 MCE80 MCE91 MCE104 MCE109 MCE110 MCE8, MCE9a MCE58 MCE65 MCE81 MCE103 MCE102 MCE71 MCE79 MCE101 MCE49 MCE62 MCE495 (NCTC 11225, CIP79.44) MA03 MCE543 MCE552 MCE553 MCE555 MCE44 MCE77 MCE78 MCE83 MCE107 MCE108 MCE542 MCE27 MCE28 MCE33 MCE38 MCE40 MCE56 MCE61 MCE82 MCE106 MCE494 MCE525 MCE526 MCE528 MCE532
eBURST group
ST
Allelic profilec
Host sourced
Countrye
Year
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 6 7 8 12 13 14 19 20 30 30 30 32 33 33 33 33 33 33 33 34 34 34 34 34 34 34 34 34 34 34 34 34 34
1-1-1-1-1-1-1 1-1-1-1-1-1-1 1-1-1-1-1-1-1 1-1-1-1-1-1-1 1-1-1-1-1-1-1 1-1-1-1-1-1-1 1-1-1-1-1-1-1 1-1-1-1-1-1-1 1-1-1-1-1-1-1 1-1-1-1-1-1-1 1-1-1-1-1-1-1 1-1-1-1-1-1-1 1-1-1-1-1-1-1 1-1-1-1-1-1-1 1-1-1-1-1-1-1 1-1-1-1-1-1-1 1-1-1-1-1-1-1 1-1-1-1-1-1-1 1-1-1-1-1-1-1 1-1-1-1-1-1-1 1-1-1-1-1-1-1 1-1-1-1-1-1-1 1-1-1-1-1-1-1 1-1-1-1-1-1-1 1-1-1-1-1-1-1 1-1-1-1-1-1-1 1-1-1-1-1-1-1 1-1-1-1-1-1-1 1-1-1-1-1-1-1 1-1-1-1-1-1-1 1-1-1-1-2-1-1 1-1-1-1-2-1-1 1-1-1-1-2-1-1 1-1-1-1-2-1-1 1-1-1-1-2-1-1 1-1-1-1-2-4-1 1-1-1-1-3-1-4 1-1-1-1-4-1-1 1-1-6-1-2-1-1 1-1-7-1-1-1-1 1-1-8-1-1-1-4 1-4-1-1-1-1-1 2-1-1-1-1-1-1 1-1-1-1-1-12-5 1-1-1-1-1-12-5 1-1-1-1-1-12-5 1-1-1-1-1-13-5 1-1-1-1-1-1-4 1-1-1-1-1-1-4 1-1-1-1-1-1-4 1-1-1-1-1-1-4 1-1-1-1-1-1-4 1-1-1-1-1-1-4 1-1-1-1-1-1-4 1-1-1-1-1-1-5 1-1-1-1-1-1-5 1-1-1-1-1-1-5 1-1-1-1-1-1-5 1-1-1-1-1-1-5 1-1-1-1-1-1-5 1-1-1-1-1-1-5 1-1-1-1-1-1-5 1-1-1-1-1-1-5 1-1-1-1-1-1-5 1-1-1-1-1-1-5 1-1-1-1-1-1-5 1-1-1-1-1-1-5 1-1-1-1-1-1-5
Female horse Female horse Female horse Female horse Female horse Female horse Unknown Horse Female horse (1) Female horse (1) Horse Female horse (2) Horse Horse Horse (3) Horse (3) Female horse (2) Horse Horse Horse Horse Horse Horse Horse Horse Horse Horse Unknown Unknown Unknown Male horse Horse Horse Horse Horse Horse Horse Horse Unknown Horse Horse Female horse Female horse Male horse (6) Female horse (4) Female horse (4) Female horse (4) Horse Horse Horse Horse Unknown Unknown Male horse (5) Horse Horse Horse Male horse Female horse Female horse (7) Female horse (7) Female horse Male horse Male horse (8) Male horse (8) Male horse (9) Male horse Male horse
US Australia Australia Australia Australia Australia Unknown UK France France France France France France France France France France France France France France France France France France France France France France France France France France France France France France France France France US Australia France France France France France France France France France France France (1) France France France France France France France France France France (2) France (2) France (2) France (2) France (2)
1978 1978 1981 1981 1981 1979 Unknown 1977 2005 2005 2004 2005 2002 2002 2001 2001 2005 2001 2001 2001 1999 1999 1999 1999 1999 1999 2001 2000 1998 1998 2005 2001 1999 1999 2000 2000 1999 1998 2000 2001 2000 1978 1978 2010 2011 2011 2011 2001 1998 1998 1999 1998 1998 2010 2004 2004 2004 2005 2002 2001 2001 1999 2005 2007 2009 2009 2009 2009
612
F. Duquesne et al. / Veterinary Microbiology 167 (2013) 609–618
Table 1 (Continued ) Strain (other reference name)
eBURST group
ST
Allelic profilec
Host sourced
Countrye
Year
MCE533 MCE534 MCE535 MCE544, MCE545a MCE554 MCE69 MCE314 MCE536 MCE1 MCE15 MCE486, MCE487, MCE488a MCE492 MCE493 MCE502 MCE504 MCE505 MCE529 MCE557 MCE558 MCE559, MCE560a MCE90 MCE98 MCE57 MCE99 MCE25 MCE66 MCE67 MCE88 MCE89 MCE96 MCE546 MA09 (EQ59) MA10 (EQ70) MA11 MA12 MA13 MCE84 MCE95 MA08 (SS28) MCE556
1 1 1 1 1 1 1 2 2 2 2 2 2 2 2 2 2 2 2 2 3 3 4 4
34 34 34 34 34 35 39 15 16 16 16 16 16 16 16 16 16 16 16 16 10 11 17 18 4 4 4 4 4 4 4 3 3 3 3 3 5 5 9 31
1-1-1-1-1-1-5 1-1-1-1-1-1-5 1-1-1-1-1-1-5 1-1-1-1-1-1-5 1-1-1-1-1-1-5 1-1-1-1-1-1-6 11-1-1-1-1-1-5 1-2-4-1-2-2-4 1-2-4-1-2-3-4 1-2-4-1-2-3-4 1-2-4-1-2-3-4 1-2-4-1-2-3-4 1-2-4-1-2-3-4 1-2-4-1-2-3-4 1-2-4-1-2-3-4 1-2-4-1-2-3-4 1-2-4-1-2-3-4 1-2-4-1-2-3-4 1-2-4-1-2-3-4 1-2-4-1-2-3-4 1-1-5-3-1-1-1 1-1-5-3-5-1-1 1-3-6-1-2-2-4 1-3-9-1-2-2-4 1-2-5-3-2-2-3 1-2-5-3-2-2-3 1-2-5-3-2-2-3 1-2-5-3-2-2-3 1-2-5-3-2-2-3 1-2-5-3-2-2-3 1-2-5-3-2-2-3 1-1-3-2-2-2-3 1-1-3-2-2-2-3 1-1-3-2-2-2-3 1-1-3-2-2-2-3 1-1-3-2-2-2-3 1-2-1-1-2-2-3 1-2-1-1-2-2-3 1-1-2-1-1-2-2 1-11-2-1-14-3-3
Male horse Male horse (9) Male horse Male horse (6) Female horse (4) Male donkey Male horse Female horse with signs of metritis Male horse Male horse Male horse Male horse (10) Male horse Female horse with signs of infertility Male horse (10) Male horse (5) Male horse Male horse (5) Male horse (11) Male horse (11) Horse Unknown Horse Horse Horse Horse Horse Female horse Horse Horse Horse Female horse Female horse Female horse Female horse Female horse Horse Unknown Female horse Horse
France France France France France France France France France France France France France France France France France France France France France France France France France France France France France France France Japan Japan Japan Japan Japan France France Japan UAE
2009 2009 2009 2010 2011 1999 2005 2009 2004 2005 2007 2007 2007 2008 2008 2008 2009 2012 2012 2012 2001 2001 2001 2001 2004 1999 1999 2001 2001 2001 2010 1980 1980 1996 1996 1996 2001 2001 1980 2009
5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 6 6 6 6 6
22 22 22 22 22 22 22 22 22 22 22 22 22 22 22 22 22 22 22 22 22 22 26 37 21 21 21 21 21
4-9-10-4-11-5-9 4-9-10-4-11-5-9 4-9-10-4-11-5-9 4-9-10-4-11-5-9 4-9-10-4-11-5-9 4-9-10-4-11-5-9 4-9-10-4-11-5-9 4-9-10-4-11-5-9 4-9-10-4-11-5-9 4-9-10-4-11-5-9 4-9-10-4-11-5-9 4-9-10-4-11-5-9 4-9-10-4-11-5-9 4-9-10-4-11-5-9 4-9-10-4-11-5-9 4-9-10-4-11-5-9 4-9-10-4-11-5-9 4-9-10-4-11-5-9 4-9-10-4-11-5-9 4-9-10-4-11-5-9 4-9-10-4-11-5-9 4-9-10-4-11-5-9 4-9-10-8-11-5-9 4-9-10-4-11-5-12 3-7-12-6-10-6-7 3-7-12-6-10-6-7 3-7-12-6-10-6-7 3-7-12-6-10-6-7 3-7-12-6-10-6-7
Male donkey Male donkey Male donkey Male donkey Male donkey Male donkey Male donkey Male donkey Male donkey Male donkey Male donkey (12) Male donkey (12) Male donkey (12) Male donkey (12) Male horse Male donkey (13) Male donkey Male donkey (14) Male donkey (13) Male donkey (13) Male donkey (13) Male donkey (13) Male donkey (14) Male horse Male donkey (15) Male donkey (15) Male donkey Female horse Male donkey
France France France France France France France France France France France France France France France France France France France France France France France France France France France France France
50 T. asinigenitalis strains MCE4 MCE10 MCE11 MCE12 MCE13 MCE14 MCE21 MCE22 MCE23 MCE24 MCE85 MCE92 MCE93 MCE97 MCE235 MCE265 MCE475 MCE476 MCE477 MCE514, MCE515a MCE516, MCE517, MCE518a MCE519, MCE520, MCE521a MCE266 MCE59 MCE2b MCE6, MCE7a MCE47 MCE86 MCE473
(2) (2)
(2)
(1) (1) (1) (1) (1)
(3) (3) (3) (3)
(4) (3) (4) (4) (4) (4) (4) (4) (5) (5) (5) (5)
2004 2005 2005 2005 2005 2005 2004 2004 2004 2004 2001 2001 2001 2001 1995 2005 2006 2006 2006 2008 2008 2008 2005 2001 2004 2005 2001 2001 2006
F. Duquesne et al. / Veterinary Microbiology 167 (2013) 609–618
613
Table 1 (Continued ) Strain (other reference name)
eBURST group
ST
Allelic profilec
Host sourced
Countrye
Year
MCE474 MCE3b MCE94 MCE479 MCE480 MCE46 MCE73 MCE76 MCE124 MCE549 MCE550, MCE551a MCE20 MCE497 (UCD-1, ATCC700933, CIP107673T) MCE513 (Bd3751/05)
6 6 7 7 7
21 36 25 38 38 23 23 23 23 24 24 27 28 29
3-7-12-6-10-6-7 3-7-12-6-10-6-8 10-8-16-7-12-11-13 10-8-16-7-12-11-14 10-8-16-7-12-11-14 6-6-11-4-6-8-11 6-6-11-4-6-8-11 6-6-11-4-6-8-11 6-6-11-4-6-8-11 9-9-13-11-9-8-17 9-9-13-11-9-8-17 5-5-14-5-13-7-10 7-10-15-9-7-9-15 8-10-17-10-8-10-16
Male Male Male Male Male Male Male Male Male Male Male Male Male Male
France France (5) France France (5) France (5) France France France France France France France US Sweden
2006 2004 2005 2006 2006 2001 1999 1999 1998 2011 2011 2004 1997 2004
donkey donkey (15) donkey (15) horse donkey (15) donkey donkey donkey donkey horse (16) horse (16) donkey donkey horse
a
Strains isolated during the same test from different sampling sites. MCE2 and MCE3 strains were isolated during the same test from different sampling sites. Allelic profile of gltA-gyrB-fh-shmt-tyrB-adk-txn loci, respectively. d The numbers in parentheses are arbitrarily attributed to animals remaining CEM-positive after being sampled several times in a single year or over a period of several years. e The numbers in parentheses are arbitrarily attributed to farms with at least two animals that remained CEM-positive; US, United States; UK, United Kingdom; UAE, United Arab Emirates. b c
primer pairs were designed from the MCE9 and MCE3 genomes using DNASTAR Lasergene1 software (version 5.51) to amplify 455–682 bp internal gene fragments (Table 2). Each 50-ml PCR was carried out in a Bio-Rad Laboratories MyCycler Thermal Cycler1 using 2 ml genomic DNA extraction, 10 pM [each] forward and reverse primers, 25 ml PCR master mix containing 1.5 mM MgCl2, 200 mM [each] deoxynucleotide triphosphates and 1.25 U Thermoprime Taq DNA polymerase (Thermo Scientific, France), and PCR-grade water. All primer pairs were designed to ensure they had similar
annealing temperatures. Amplification conditions were as follows: initial denaturation at 95 8C for 5 min and 35 cycles of denaturation at 94 8C for 30 s, primer annealing at 55 8C for 1 min and extension at 72 8C for 1 min. Amplification products were sequenced by Beckman Coulter Genomics (Stansted, UK). The forward and reverse sequences of a given locus were edited, aligned and trimmed to the desired length (MLST fragment, Table 2) using the http://multalin.toulouse.inra.fr/multalin/ website and MEGA v5.05 software (http://www.megasoft ware.net).
Table 2 Gene fragments and primer details for Taylorella sp. MLST. Locusa Gene ID accessionb PCR and sequencing primers (50 !30 )c Amplicon size (bp) Locus size (bp) MLST fragment start (50 )/stop (30 ) T. equigenitalis
T. asinigenitalis
gltA
11186031
GCTCAGACAGGCATGTTTACTTA (F) GTCCCCATAGGCAAGAAATAC (R)
682
623
ATGTTTAC/TCAGGTAC ATGTTTAC/TCAGGTAC
gyrB
10114465
GTTGCTATGCAATGGACTGA (F) GCACGTGATTTTTCTACATT (R)
614
518
AATAATAT/CAAGCAAT AATAACAT/CAAGCAAT
fh
10114863
AGAGGCTTGCGATGAGG (F) AGTCAGGAGCCCACTCTAC (R)
659
574
TGCGTTTA/TCGTTAGA TGCGAATG/TCTCTAGA
shmt
10114226
GACATTGCCCACTATTCAGG (F) CTCTTGTAGTCATAGCTGGTGTTC (R)
511
435
TATTCAGG/TACCTAAT TATTCAGG/TTCCAAAC
tyrB
10115298
CCTATTTTGGGGCTTTCTGACT (F) GGTCTACGCCAGTAGGATT (R)
532
462
TTGGTGTT/TAATCCTA TAGGTGTT/CAATCCTA
adk
10115205
ATGCGTCTAATTCTTCTTG (F) AACGTCTCCGACACCATT (R)
615
538
TAAATACA/TAAATGGT TAAATACA/TCAATGGT
txn
10115427
ACGGGGGACCGCATAAAGCC (F) AGCGTTTCTGTACCCGTGCGA (R)
455
371
ATTTATGG/ATTCGCAC ATTTATGG/ATTCGCAC
a gltA, citrate synthase; gyrB, gyrase subunit B; fh, putative fumarate hydratase; shmt, serine hydroxymethyltransferase; tyrB, tyrosine aminotransferase; adk, adenylate kinase; txn, thioredoxine. b From T. equigenitalis MCE9 genome (He´bert et al., 2011). c F, forward; R, reverse.
F. Duquesne et al. / Veterinary Microbiology 167 (2013) 609–618
614
2.4. Data analysis For each locus, all the sequences were compared and an allelic number was assigned to each unique allele for a given locus. The allelic profile of each strain was generated by combining the allelic numbers of each of the seven loci in the following order: gltA, gyrB, fh, shmt, tyrB, adk, and txn. A novel sequence type (ST) designation was given to each unique allelic profile; strains with the same allelic profile belonged to the same ST. The number of polymorphic nucleotide sites was calculated using MEGA v5.05 software. For each locus, the degree of selective pressure was assessed by calculating the dN/dS ratio of nonsynonymous substitutions (dN) to synonymous substitutions (dS) using START v2 software (http://pubmlst.org/software/analysis/start2). A dN/dS ratio higher than 1 implies selection for amino acid changes (Yang and Bielawski, 2000). Multilocus linkage disequilibrium was assessed using the index of association (IA) using START v2 software. In case of linkage disequilibrium because of frequent recombination events, the expected value of IA is zero. Clonal populations are identified by an IA that differs significantly from zero (Smith et al., 1993). The seven allelic sequences for each strain were concatenated (3521 bp) to generate a neighbor-joining phylogenetic tree, with 1000 bootstrap replicates, using the Jukes Cantor algorithm in the MEGA v5.05 software. Using the eBURST v3 program, the strains were grouped into clonal complexes according to the MLST definition: a cluster of strains sharing at least six of the seven alleles sequenced (www.mlst.net) (Jolley and Maiden, 2010). STs were classified as single-locus variants (SLVs), double-locus variants (DLVs) or individual unlinked STs. 3. Results 3.1. Variation of the seven MLST loci The sequence of the seven chosen loci (gltA, gyrB, fh, shmt, tyrB, adk, and txn) was determined for the 113 T. equigenitalis and 50 T. asinigenitalis strains isolated (Table 1), and allelic profiles were assigned. As shown in Table 2, the mean average allele length for each locus was 503 pb and ranged from 371 pb (txn) to 623 pb (gltA). All alleles for a given locus were of equal length and, to aid further analysis, were in the correct reading
frame. Among the 163 investigated isolates, 3 (gltA and shmt) to 9 alleles (fh) were present at each locus for T. equigenitalis and 6 (gyrB) to 11 alleles (txn) were present at each locus for T. asinigenitalis (mean average = 5.4 and 8.0, respectively). The proportion of variable nucleotide sites at each locus varied from 0.5% (gltA and shmt) to 6.1% (tyrB), and from 1.3% (gyrB) to 6.7% (txn) for T. equigenitalis and T. asinigenitalis, respectively (mean average = 1.9% and 4.4%) (Table 3; Fig. S1). Supplementary material related to this article can be found, in the online version, at http://dx.doi.org/10.1016/ j.vetmic.2013.09.016. The dN/dS ratios were calculated to determine the degree of selective pressure. For each locus, the dN/dS ratios were less than 1 (Table 3), indicating that no strong positive selective pressure was present at any selected loci, validating their suitability for inclusion into the Taylorella sp. MLST scheme. IA was calculated to estimate the degree of association and recombination between alleles at different loci, based on the allelic profile data. The IA computed using the all 163 isolates was 3.64 (p < 0.001). This indicates significant linkage disequilibrium and suggests a clonal population structure in the Taylorella genus and limited recombinational events, even between the glta and txn loci that are separated by only 7 kb. The IA remained significantly different from zero (2.21) when only one representative of each ST was included in the computation, which rules out any bias due to taxonomic sampling (p < 0.001). The calculated IA values for T. equigenitalis and T. asinigenitalis populations were 1.9845 (p < 0.001) and 1.6428 (p < 0.001), respectively. 3.2. Relatedness of Taylorella sp. strains A total of 27 and 12 STs were respectively assigned to the 113 T. equigenitalis and 50 T. asinigenitalis strains used in this study (Table 1). The most common STs were identified 30 (ST-1), 20 (ST-34) and 15 (ST-16) times among the 113 T. equigenitalis strains, and 27 (ST-22) and 7 (ST-21) times among the 50 T. asinigenitalis strains. No STs or alleles were common between T. equigenitalis and T. asinigenitalis. The neighbor-joining tree of concatenated gltA, gyrB, fh, shmt, tyrB, adk, and txn sequences (3521 bp) showed that both species are divided into two distinct groups and that the genetic diversity is more important for T. asinigenitalis than for T. equigenitalis (Fig. S2).
Table 3 Analysis of the seven loci in the Taylorella sp. population of 163 strains. Locus
gltA gyrB fh shmt tyrB adk txn a
MLST fragment size (bp)
623 518 574 435 462 538 371
T. equigenitalis
T. asinigenitalis
No. of alleles
Polymorphic sites, no (%)
dN/dSa
No. of alleles
Polymorphic sites, no (%)
dN/dS
3 5 9 3 6 6 6
3 4 13 2 28 5 8
0.1495 0.2369 0.1313 0.0000 0.7317 0.1723 0.6658
8 6 8 8 8 7 11
27 7 26 19 19 30 25
0.0226 0.0000 0.0610 0.0377 0.0266 0.0631 0.2334
(0.5) (0.8) (2.3) (0.5) (6.1) (0.9) (2.2)
dN/dS: ratio of nonsynonymous (dN) to synonymous (dS) substitutions.
(4.3) (1.3) (4.5) (4.4) (4.1) (5.6) (6.7)
F. Duquesne et al. / Veterinary Microbiology 167 (2013) 609–618
615
Fig. 1. Population snapshot of the Taylorella genus. Clusters of related STs (corresponding to clonal complexes) and individual unlinked STs within the study strains are displayed as a single eBURST diagram by setting the group definition to zero of seven shared alleles. STs are represented as circles; the size of a circle is proportional to the number of strains within a given ST. Primary founders and subgroup founders (cofounders) are respectively shown in white and gray circles. Single-locus variant links are represented as black lines. Each clonal complex was defined by a number in boldface type (CC1–CC7). Distances between STs, indicated by connecting lines, are arbitrary.
Supplementary material related to this article can be found, in the online version, at http://dx.doi.org/10.1016/ j.vetmic.2013.09.016. The likely patterns of chromosomal evolutionary descent and identity of the probable founding genotypes were inferred from an eBURST analysis (Feil et al., 2004) of the MLST data where the number of resamplings made for bootstrapping was 1000 and only SLVs were considered. Using these parameters seven clonal complexes were identified among the 39 STs (Table 1; Fig. 1). In details, the 27 T. equigenitalis STs were grouped into four clonal complexes (CC1–CC4) and five individual unlinked STs, and the 12 T. asinigenitalis STs into three clonal complexes (CC5–CC7) and five individual unlinked STs. The largest T. equigenitalis clonal complex (CC1) contained 16 STs (68.1% of the T. equigenitalis tested) with the predicted founder ST1 and a cofounded ST-34. Each remaining T. equigenitalis clonal complex contained 2 STs with 1.8% (CC3 and CC4) to 14.2% (CC2) of the tested T. equigenitalis. The largest T. asinigenitalis clonal complex (CC5) contained 3 STs (58.0% of the tested T. asinigenitalis), and the ST-22 could be considered as the predicted founder. Each of the remaining T. asinigenitalis clonal complexes contained 2 STs with 6.0% (CC7) to 16.0% (CC6) of the tested T. asinigenitalis. The individual unlinked STs represented 14.2% of T. equigenitalis strains and 20.0% of T. asinigenitalis strains.
When DLVs were considered, the eBURST analysis showed that (i) CC1 and CC3 are linked since ST-10 has two DLVs with ST-1 and ST-13, and (ii) CC2, CC4, ST-4 and ST-5 are linked since ST-5 has two DLVs with ST-4 and ST-15, and ST-15 has three DLVs with ST-5, ST-17 and ST-18 (Fig. S3). Supplementary material related to this article can be found, in the online version, at http://dx.doi.org/10.1016/ j.vetmic.2013.09.016. 3.3. Relationship between clonal complex and epidemiologic data The data set presented in the study was not specifically designed to investigate the relationship between molecular diversity and host or geographical origin of the strains. However, the following points can be emphasized: (i) the ST-1 contains T. equigenitalis strains from Australia (1978 and 1981), France (1998– 2011), UK (1977) and US (1977 and 1978), (ii) the Japanese and Emirati T. equigenitalis strains are classified into unlinked STs (ST-3 and ST-9 for Japanese strains and ST-31 for the Emirati strain), (iii) the two non-French T. asinigenitalis strains are classified into unlinked STs (ST28 for the American strain and ST-29 for the Swedish strain).
616
F. Duquesne et al. / Veterinary Microbiology 167 (2013) 609–618
Among the five farms where several animals were identified as CEM-positive, one had strains with a common ST, two had strains with different STs of the same clonal complex (farms no. 2 and 4), and two had strains with different STs of different clonal complexes (farms no. 1 and 5). Among the 16 animals identified as CEM-positive from several samples taken during the same year or over several years, three carried strains with different STs from the same clonal complex (animals no. 4, 6 and 14) and two carried strains with different STs from two clonal complexes (animals no. 5 and 15). The only T. equigenitalis strain (MCE69) isolated from a donkey was classified in the CC1, like 68.1% of the T. equigenitalis strains, and one can note that the five T. asinigenitalis strains isolated from horses do not form a separate clonal complex among the 50 T. asinigenitalis strains. 4. Discussion MLST is a well-established molecular typing method that has been successfully used for the determination of the population structure of many bacterial species (Maiden, 2006; Maiden et al., 1998). In this study, we used the recent genome sequences of T. equigenitalis and T. asinigenitalis (He´bert et al., 2011, 2012) to develop an unambiguous and discriminatory Taylorella sp. MLST scheme based on the allelic variations of seven housekeeping genes common to both bacterial species. Internal fragments of the seven loci selected for the final MLST scheme (gltA, gyrB, fh, shmt, tyrB, adk, and txn) were amplified from 113 T. equigenitalis and 50 T. asinigenitalis strains. Despite a notable genetic divergence between T. equigenitalis and T. asinigenitalis (He´bert et al., 2011, 2012), the selection of the housekeeping genes was validated by a dN/dS ratio lower than 1 for all of them, indicating a very low contribution of environmental selection to the variation of these genes (Table 3). The data generated have been used to construct a publicly open and curated MLST website with the Taylorella sp. MLST scheme and data (http://pubmlst.org/taylorella/). In terms of data validation, the sequence chromatograms produced by automated sequencers can be sent by email to the Taylorella sp. MLST website curator to be checked before publication. Based on calculated IA values, it appears that the Taylorella population can be classified as clonal (Smith et al., 1993), indicating an absence or a low probability of intra- and inter-species recombination. This observation is consistent with the presence of a conserved synteny (He´bert et al., 2012), and the absence of common ST or allele identified between T. equigenitalis and T. asinigenitalis. Despite 2.3 times more strains of T. equigenitalis than T. asinigenitalis used in this study, the genetic diversity of T. equigenitalis appeared to be lower than that of T. asinigenitalis (Table 3). This observation is confirmed by a higher IA value for the T. equigenitalis population (1.9845, p < 0.001) than for the T. asinigenitalis population (1.6428, p < 0.001), and suggests that the T. equigenitalis population has a more ‘‘epidemic’’ structure than T. asinigenitalis (Smith et al., 1993).
The analysis of MLST results among the 163 strains tested from several countries over a 35-year period of time showed the presence of 39 STs, including 39% of unlinked STs (Fig. 1). The emergence of new STs over time suggests that Equidae carrying these strains were contaminated by an external source of Taylorella originating from a still unidentified natural ecological reservoir. Moreover, these emergences were observed in different countries all over the world (Table 1), suggesting that this potential reservoir is spread worldwide. The first CEM outbreaks, that emerged simultaneously in the late 1970s in several countries (at least UK, US and Australia according the present results), were reported in the literature as the cause of significant economic impact in the horse breeding industry because of a rapid worldwide spread and the presence of characteristic clinical signs of reproductive tract disease in mares (Timoney, 2011). The present results revealed that these first CEM outbreaks were associated with the T. equigenitalis founding clonal complex CC1 confirming an epidemiological relationship and suggesting that these strains are originated from a single initial bacterial clone of unknown origin. Interestingly, in France, T. equigenitalis strains associated with the CC1 are still being isolated from mares showing few or no clinical signs (Table 1). This observation suggests that the initially virulent strain lost its virulence traits in favor of persistence traits allowing it to be asymptomatically carried for years, as suggested for other pathogens and symbionts (Marchetti et al., 2010; Sachs et al., 2011). Moreover, with the exception of the CC1, all CCs and STs contain strains originated from a single country. This observation suggests a lower international dissemination of these strains compare to CC1 strains, and suggests that international prophylactic measures taken to prevent dissemination of CEM over the world after the first CEM outbreak have been generally effective. However, more data on international CEM strains remain necessary to evaluate this effectiveness. In this context, it remains important to maintain these prophylactic measures, to prevent the dissemination of new epidemic clones. For example, the CC2, circulating in France since the 2000s, should be carefully monitored to limit its spread to other parts of the world, since the two females carrying a T. equigenitalis strain associated with the CC2 developed signs of reproductive tract disease (strains MCE502 and MCE536; Table 1). Comparisons are frequently drawn between the levels of resolution provided by MLST and PFGE. Many studies showed similar levels of discrimination between these techniques (Godoy et al., 2003; Peacock et al., 2002) but MLST was found to be more discriminatory in other studies (Kotetishvili et al., 2003, 2002; Revazishvili et al., 2004). In previous studies, 130 Japanese T. equigenitalis strains, including MA08 to MA13 strains, isolated from 1980 to 1996 were examined by PFGE and appeared to have a common genotype, designed ‘‘genotype J’’, suggesting a common source for all of them (Matsuda et al., 1999; Miyazawa et al., 1995). Here, we show that the MA08 (SS28) strain isolated in 1980 does not have the same ST as the other Japanese strains (Table 1), suggesting actually
F. Duquesne et al. / Veterinary Microbiology 167 (2013) 609–618
two separate outbreaks in Japan during the early 1980s. Moreover, whereas the previous 16S rDNA analysis of T. asinigenitalis strains separated them into two profiles (Ba˚verud et al., 2006; Breuil et al., 2011), the present Taylorella sp. MLST allows refinement of the phylogenetic analysis of T. asinigenitalis into 12 STs. The T. asinigenitalis 16S rDNA group 1 (Breuil et al., 2011) is exclusively associated with the clonal complex 5 while the T. asinigenitalis 16S rDNA group 2 (Breuil et al., 2011) has a genetic variability associated with two clonal complexes (CC6 and CC7) and five unlinked STs (Fig. 1). Altogether, these results allow to conclude that the Taylorella sp. MLST scheme developed in this study is a powerful tool for molecular epidemiological studies of the Taylorella genus. The use of the MLST database by other researchers from all around the world will increase the representatively of these results, thus offering a new way to study the population biology of these bacteria. Acknowledgements This publication made use of the MLST website (http:// pubmlst.org/mlst/) developed by Keith Jolley and situated at the University of Oxford; the development of this site was funded by the Wellcome Trust (Jolley and Maiden, 2010). We are very grateful to the French veterinary laboratories approved for CEM diagnosis that contributed the Taylorella strains used in this study, to Viveca Ba˚verud (National Veterinary Institute, Sweden) for providing the T. asinigenitalis Bd3751/05 strain, and to Ulrich Wernery (Central Veterinary Research Laboratory, United Arab Emirates) for providing the Emirati T. equigenitalis strain. We also would like to thank Christine Delorme (INRA Jouyen-Josas, France) for her technical assistance as well as Dana Pottratz and Guillaume He´bert from SC Partners for their editorial assistance. This work was supported in part by the European Union in conjunction with the European Union Reference Laboratory for equine diseases. References Aalsburg, A.M., Erdman, M.M., 2011. Pulsed-field gel electrophoresis genotyping of Taylorella equigenitalis isolates collected in the United States from 1978 to 2010. J. Clin. Microbiol. 49, 829–833. Ba˚verud, V., Nystrom, C., Johansson, K.E., 2006. Isolation and identification of Taylorella asinigenitalis from the genital tract of a stallion, first case of a natural infection. Vet. Microbiol. 116, 294–300. Berriman, M., Rutherford, K., 2003. Viewing and annotating sequence data with Artemis. Brief. Bioinform. 4, 124–132. Bleumink-Pluym, N., ter Laak, E.A., van der Zeijst, B.A., 1990. Epidemiologic study of Taylorella equigenitalis strains by field inversion gel electrophoresis of genomic restriction endonuclease fragments. J. Clin. Microbiol. 28, 2012–2016. Breuil, M.F., Duquesne, F., Laugier, C., Petry, S., 2011. Phenotypic and 16S ribosomal RNA gene diversity of Taylorella asinigenitalis strains isolated between 1995 and 2008. Vet. Microbiol. 148, 260–266. Crowhurst, R.C., 1977. Genital infection in mares. Vet. Rec. 100, 476. Duquesne, F., Pronost, S., Laugier, C., Petry, S., 2007. Identification of Taylorella equigenitalis responsible for contagious equine metritis in equine genital swabs by direct polymerase chain reaction. Res. Vet. Sci. 82, 47–49. Enright, M.C., Day, N.P., Davies, C.E., Peacock, S.J., Spratt, B.G., 2000. Multilocus sequence typing for characterization of methicillin-resistant and methicillin-susceptible clones of Staphylococcus aureus. J. Clin. Microbiol. 38, 1008–1015.
617
Enright, M.C., Spratt, B.G., 1999. Multilocus sequence typing. Trends Microbiol. 7, 482–487. Erdman, M.M., Creekmore, L.H., Fox, P.E., Pelzel, A.M., Porter-Spalding, B.A., Aalsburg, A.M., Cox, L.K., Morningstar-Shaw, B.R., Crom, R.L., 2011. Diagnostic and epidemiologic analysis of the 2008–2010 investigation of a multi-year outbreak of contagious equine metritis in the United States. Prev. Vet. Med. 101, 219–228. Feil, E.J., Li, B.C., Aanensen, D.M., Hanage, W.P., Spratt, B.G., 2004. eBURST: inferring patterns of evolutionary descent among clusters of related bacterial genotypes from multilocus sequence typing data. J. Bacteriol. 186, 1518–1530. Godoy, D., Randle, G., Simpson, A.J., Aanensen, D.M., Pitt, T.L., Kinoshita, R., Spratt, B.G., 2003. Multilocus sequence typing and evolutionary relationships among the causative agents of melioidosis and glanders, Burkholderia pseudomallei and Burkholderia mallei. J. Clin. Microbiol. 41, 2068–2079. He´bert, L., Moumen, B., Duquesne, F., Breuil, M.F., Laugier, C., Batto, J.M., Renault, P., Petry, S., 2011. Genome sequence of Taylorella equigenitalis MCE9, the causative agent of contagious equine metritis. J. Bacteriol. 193, 1785. He´bert, L., Moumen, B., Pons, N., Duquesne, F., Breuil, M.F., Goux, D., Batto, J.M., Laugier, C., Renault, P., Petry, S., 2012. Genomic characterization of the Taylorella genus. PLoS One 7, e29953. Jang, S.S., Donahue, J.M., Arata, A.B., Goris, J., Hansen, L.M., Earley, D.L., Vandamme, P.A., Timoney, P.J., Hirsh, D.C., 2001. Taylorella asinigenitalis sp. nov. a bacterium isolated from the genital tract of male donkeys (Equus asinus). Int. J. Syst. Evol. Microbiol. 51, 971–976. Jolley, K.A., Maiden, M.C., 2010. BIGSdb: scalable analysis of bacterial genome variation at the population level. BMC Bioinform. 11, 595. Katz, J.B., Evans, L.E., Hutto, D.L., Schroeder-Tucker, L.C., Carew, A.M., Donahue, J.M., Hirsh, D.C., 2000. Clinical, bacteriologic, serologic, and pathologic features of infections with atypical Taylorella equigenitalis in mares. J. Am. Vet. Med. Assoc. 216, 1945–1948. Kotetishvili, M., Stine, O.C., Chen, Y., Kreger, A., Sulakvelidze, A., Sozhamannan, S., Morris Jr., J.G., 2003. Multilocus sequence typing has better discriminatory ability for typing Vibrio cholerae than does pulsed-field gel electrophoresis and provides a measure of phylogenetic relatedness. J. Clin. Microbiol. 41, 2191–2196. Kotetishvili, M., Stine, O.C., Kreger, A., Morris Jr., J.G., Sulakvelidze, A., 2002. Multilocus sequence typing for characterization of clinical and environmental Salmonella strains. J. Clin. Microbiol. 40, 1626–1635. Luddy, S., Kutzler, M.A., 2010. Contagious equine metritis Within the United States: a review of the 2008 outbreak. J. Equine Vet. Sci. 30, 393–400. Maiden, M.C., 2006. Multilocus sequence typing of bacteria. Annu. Rev. Microbiol. 60, 561–588. Maiden, M.C., Bygraves, J.A., Feil, E., Morelli, G., Russell, J.E., Urwin, R., Zhang, Q., Zhou, J., Zurth, K., Caugant, D.A., Feavers, I.M., Achtman, M., Spratt, B.G., 1998. Multilocus sequence typing: a portable approach to the identification of clones within populations of pathogenic microorganisms. Proc. Natl. Acad. Sci. U.S.A. 95, 3140–3145. Marchetti, M., Capela, D., Glew, M., Cruveiller, S., Chane-Woon-Ming, B., Gris, C., Timmers, T., Poinsot, V., Gilbert, L.B., Heeb, P., Medigue, C., Batut, J., Masson-Boivin, C., 2010. Experimental evolution of a plant pathogen into a legume symbiont. PLoS Biol. 8, e1000280. Matsuda, M., Miyazawa, T., Anzai, T., 1999. Detection of a common genotype among strains of Taylorella equigenitalis isolated from thoroughbred horses in Japan between 1994 and 1996. J. Basic Microbiol. 39, 127–130. Miyazawa, T., Matsuda, M., Isayama, Y., Samata, T., Ishida, Y., Ogawa, S., Takei, K., Honda, M., Kamada, M., 1995. Genotyping of isolates of Taylorella equigenitalis from thoroughbred brood mares in Japan. Vet. Res. Commun. 19, 265–271. OIE, 2012. Contagious Equine Metritis. Manual of Diagnostic Tests and Vaccines for Terrestrial Animals. Chap. 2.5.2. Peacock, S.J., de Silva, G.D., Justice, A., Cowland, A., Moore, C.E., Winearls, C.G., Day, N.P., 2002. Comparison of multilocus sequence typing and pulsed-field gel electrophoresis as tools for typing Staphylococcus aureus isolates in a microepidemiological setting. J. Clin. Microbiol. 40, 3764–3770. Revazishvili, T., Kotetishvili, M., Stine, O.C., Kreger, A.S., Morris Jr., J.G., Sulakvelidze, A., 2004. Comparative analysis of multilocus sequence typing and pulsed-field gel electrophoresis for characterizing Listeria monocytogenes strains isolated from environmental and clinical sources. J. Clin. Microbiol. 42, 276–285. Rutherford, K., Parkhill, J., Crook, J., Horsnell, T., Rice, P., Rajandream, M.A., Barrell, B., 2000. Artemis: sequence visualization and annotation. Bioinformatics 16, 944–945.
618
F. Duquesne et al. / Veterinary Microbiology 167 (2013) 609–618
Sachs, J.L., Skophammer, R.G., Regus, J.U., 2011. Evolutionary transitions in bacterial symbiosis. Proc. Natl. Acad. Sci. U.S.A. 108 (Suppl. 2) 10800–10807. Selander, R.K., Caugant, D.A., Ochman, H., Musser, J.M., Gilmour, M.N., Whittam, T.S., 1986. Methods of multilocus enzyme electrophoresis for bacterial population genetics and systematics. Appl. Environ. Microbiol. 51, 873–884. Smith, J.M., Smith, N.H., O’Rourke, M., Spratt, B.G., 1993. How clonal are bacteria? Proc. Natl. Acad. Sci. U.S.A. 90, 4384–4388. Sugimoto, C., Isayama, Y., Sakazaki, R., Kuramochi, S., 1983. Transfer of Haemophilus equigenitalis Taylor et al. 1978 to the genus Taylorella gen. nov. asTaylorella equigenitalis comb. nov. Curr. Microbiol. 9, 155–162. Thoresen, S.I., Jenkins, A., Ask, E., 1995. Genetic homogeneity of Taylorella equigenitalis from Norwegian trotting horses revealed by chromosomal DNA fingerprinting. J. Clin. Microbiol. 33, 233–234.
Timoney, P.J., 2011. Horse species symposium: contagious equine metritis: an insidious threat to the horse breeding industry in the United States. J. Anim. Sci. 89, 1552–1560. Timoney, P.J., Ward, J., Kelly, P., 1977. A contagious genital infection of mares. Vet. Rec. 101, 103. Wakeley, P.R., Errington, J., Hannon, S., Roest, H.I., Carson, T., Hunt, B., Sawyer, J., Heath, P., 2006. Development of a real time PCR for the detection of Taylorella equigenitalis directly from genital swabs and discrimination from Taylorella asinigenitalis. Vet. Microbiol. 118, 247–254. Yang, Z., Bielawski, J.P., 2000. Statistical methods for detecting molecular adaptation. Trends Ecol. Evol. 15, 496–503.
Contents lists available at ScienceDirect
Veterinary Microbiology journal homepage: www.elsevier.com/locate/vetmic
Development of a single multi-locus sequence typing scheme for Taylorella equigenitalis and Taylorella asinigenitalis Fabien Duquesne a,*, Laurent He´bert a, Marie-France Breuil a, Motoo Matsuda b, Claire Laugier a, Sandrine Petry a a
ANSES, Dozule´ Laboratory for Equine Diseases, Bacteriology and Parasitology Unit, 14430 Dozule´, France Laboratory of Molecular Biology, Graduate School of Environmental Health Sciences, Azabu University, Fuchinobe 1-17-71, Chuo-ku, Sagamihara 252-5201, Japan
b
A R T I C L E I N F O
A B S T R A C T
Article history: Received 10 April 2013 Received in revised form 12 September 2013 Accepted 13 September 2013
We describe here the development of a multilocus sequence typing (MLST) scheme for Taylorella equigenitalis, the causative agent of contagious equine metritis (CEM), and Taylorella asinigenitalis, a nonpathogenic bacterium. MLST was performed on a set of 163 strains collected in several countries over 35 years (1977–2012). The MLST data were analyzed using START2, MEGA 5.05 and eBURST, and can be accessed at http:// pubmlst.org/taylorella/. Our results revealed a clonal population with 39 sequence types (ST) and no common ST between the two Taylorella species. The eBURST analysis grouped the 27 T. equigenitalis STs into four clonal complexes (CC1–4) and five unlinked STs. The 12 T. asinigenitalis STs were grouped into three clonal complexes (CC5–7) and five unlinked STs, among which CC1 (68.1% of the 113 T. equigenitalis) and CC5 (58.0% of the 50 T. asinigenitalis) were dominants. The CC1, still in circulation in France, contains isolates from the first CEM outbreaks that simultaneously emerged in several countries in the late 1970s. The emergence in different countries (e.g. France, Japan, and United Arab Emirates) of STs without any genetic relationship to CC1 suggests the existence of a natural worldwide reservoir that remains to be identified. T. asinigenitalis appears to behave same way since the American, Swedish and French isolates have unrelated STs. This first Taylorella sp. MLST is a powerful tool for further epidemiological investigations and population biology studies of the Taylorella genus. ß 2013 Elsevier B.V. All rights reserved.
Keywords: Taylorella genus MLST Contagious equine metritis
1. Introduction Taylorella equigenitalis and Taylorella asinigenitalis, small Gram-negative microaerophilic coccobacilli, are the only two species of the Taylorella genus classified in the Alcaligenaceae family (Jang et al., 2001; Sugimoto et al., 1983). T. equigenitalis is the causative agent of contagious equine metritis (CEM), a sexually-transmitted infection of horses first reported in 1977 (Crowhurst, 1977; Timoney
* Corresponding author. Tel.: +33 2 31 79 22 76; fax: +33 2 31 39 21 37. E-mail address: [email protected] (F. Duquesne). 0378-1135/$ – see front matter ß 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.vetmic.2013.09.016
et al., 1977). CEM is currently present in many parts of the world and various horse breeds (Timoney, 2011), and is notifiable to the World Organisation for Animal Health (OIE, 2012). Whereas stallions are typically asymptomatic carriers (Timoney, 2011), 30–40% of infected mares show abundant mucopurulent vaginal discharge and variable degrees of vaginitis, endometritis and cervicitis, which leads to temporary infertility (Wakeley et al., 2006). A proportion of mares may become long-time symptomless carriers and, together with infected stallions, may spread infection during mating (Luddy and Kutzler, 2010). T. asinigenitalis is currently considered a nonpathogenic bacterium, despite clinical signs of metritis and cervicitis
610
F. Duquesne et al. / Veterinary Microbiology 167 (2013) 609–618
in mares following experimental intra-uterine infusion with field isolates of T. asinigenitalis (Katz et al., 2000). T. asinigenitalis was first reported on late 1997–early 1998 in the states of California and Kentucky (Katz et al., 2000). It is characterized by slight differences from T. equigenitalis regarding colony morphology, growth rates and immunofluorescence characteristics. Due to their high degree of relatedness, it remains difficult to differentiate both Taylorella species only using the conventional methods, which must be associated with species-specific PCR. This difficulty is illustrated by the fact that T. asinigenitalis isolates have been mis-identified as T. equigenitalis (Breuil et al., 2011). Every CEM outbreak (the last major one occurred in the United States (US) in 2008–2010, (Erdman et al., 2011; Timoney, 2011)) highlights the insidious threat of this disease to the horse breeding industry, due to its infectiousness, short-term infertility in mares, and the existence of a carrier state independent of the sex of the horse. Pulsed-field gel electrophoresis (PFGE) is currently the main tool used to genotype CEM isolates for epidemiological analysis, the most recent example being the PFGE genotyping of T. equigenitalis isolated in the US from 1978 to 2010 (Aalsburg and Erdman, 2011). Several other molecular typing tools have also been used including field inversion gel electrophoresis (Bleumink-Pluym et al., 1990), chromosomal DNA fingerprinting (Thoresen et al., 1995) and crossed-field gel electrophoresis (Miyazawa et al., 1995). However, these molecular epidemiological tools are poorly portable because of their index of variation and it is difficult to compare results among laboratories (Maiden et al., 1998). Accordingly, these tools are not well suited for the global epidemiological studies of CEM outbreaks. The international trade in Equidae leads to the need to identify and track the global spread of T. equigenitalis and T. asinigenitalis, and thus investigate the genetic relationship between isolates from different outbreaks and both bacterial species. In this regard, multilocus sequence typing (MLST) seems to be more appropriate to the current need for large-scale epidemiological studies. MLST is based on DNA sequence comparison of internal fragments of housekeeping genes chosen because their products play vital function and are present in all isolates of a given species, and mutations within them are assumed to be neutral (Maiden et al., 1998; Selander et al., 1986). The accumulation of nucleotide changes in housekeeping genes is a relatively slow process and their allelic profile is sufficiently stable over time for the method to be ideal for global epidemiology (Enright and Spratt, 1999). Because MLST is based on nucleotide sequence, it is highly discriminatory and provides unambiguous results that are directly comparable among laboratories via the Internet (Enright et al., 2000). The recent characterization and comparison of complete genome sequences of T. equigenitalis MCE9 (He´bert et al., 2011) and T. asinigenitalis MCE3 (He´bert et al., 2012) now allows the development of such a tool. We describe here the development of an MLST scheme for T. equigenitalis and T. asinigenitalis based on the nucleotide sequences of seven housekeeping genes.
2. Materials and methods 2.1. Bacterial strains As listed in Table 1, a total of 113 T. equigenitalis and 50 T. asinigenitalis strains were used in the present study, including reference strains CIP79.7T, CIP79.44 and CIP 107673T from the Collection Institut Pasteur (CIP; France). This set of isolates is both temporally and geographically diverse with strains collected in six countries (France, US, UK, Japan, Australia, Sweden) over a 35-year period of time (1977–2012). As French reference laboratory for CEM, we have an exhaustive strain collection of the most CEM cases identified in France since 1992. On the basis of known epidemiological data (date, host source and geographic location), we selected 97 T. equigenitalis and 48 T. asinigenitalis to ensure the best representation of French Taylorella population. Among them, 43 strains were isolated from 16 animals that remained CEM-positive after several samplings within a single year or over a period of several years, and 33 strains were isolated from five French farms with at least two animals that remained CEM-positive. To extrapolate our results globally, we completed our French selection by strains from different countries (US, UK, Japan, Australia, and Sweden). All isolates had been previously identified using routine phenotypic tests and confirmed by PCR (Breuil et al., 2011; Duquesne et al., 2007). Full details of all of the strains used in this study are available on the MLST database (http://pubmlst.org/taylorella/). Each strain was inoculated on a ready-to-use heated blood agar enriched with a polyvitamin supplement (AES Chemunex, Combourg, France). Plates were incubated at 37 8C in 7% (v/v) CO2 in air for 48 h (T. equigenitalis) or 72 h (T. asinigenitalis). 2.2. Choice of loci for MLST Initially, 22 potential housekeeping genes were identified by comparing the T. equigenitalis MCE9 (He´bert et al., 2011) and T. asinigenitalis MCE3 (He´bert et al., 2012) genome sequences using the Artemis Comparison Tool (Berriman and Rutherford, 2003; Rutherford et al., 2000) available at http://www.sanger.ac.uk/resources/software/artemis/. On the base of a pilot MLST study performed on 20 Taylorella sp. strains, the set was reduced to seven gene fragments (Table 2) by eliminating fragments that could not be readily amplified from all tested isolates as well as those with close chromosomal locations. The chromosomal locations of these seven housekeeping genes suggest that it is unlikely for any of them to be coinherited in the same recombination event since the minimum distance between two loci is 95 kbp except for the txn gene, which is located 7 kbp away from the gltA gene. 2.3. Amplification and nucleotide sequence determination Genomic DNA of each strain was extracted using the NucleoSpin Tissue kit according to the manufacturer’s protocol (Macherey-Nagel). Amplification/sequencing
F. Duquesne et al. / Veterinary Microbiology 167 (2013) 609–618
611
Table 1 Characteristics of Taylorella isolates and MLST data. Strain (other reference name) 113 T. equigenitalis strains MA01 (KY-188) MA02 MA04 MA05 MA06 MA07 MA14 MCE496 (ATCC35865, CIP79.7T) MCE16 MCE17 MCE26 MCE39 MCE41 MCE42 MCE43 MCE48 MCE51 MCE54 MCE55 MCE60 MCE63 MCE64 MCE68 MCE70 MCE75 MCE80 MCE91 MCE104 MCE109 MCE110 MCE8, MCE9a MCE58 MCE65 MCE81 MCE103 MCE102 MCE71 MCE79 MCE101 MCE49 MCE62 MCE495 (NCTC 11225, CIP79.44) MA03 MCE543 MCE552 MCE553 MCE555 MCE44 MCE77 MCE78 MCE83 MCE107 MCE108 MCE542 MCE27 MCE28 MCE33 MCE38 MCE40 MCE56 MCE61 MCE82 MCE106 MCE494 MCE525 MCE526 MCE528 MCE532
eBURST group
ST
Allelic profilec
Host sourced
Countrye
Year
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 6 7 8 12 13 14 19 20 30 30 30 32 33 33 33 33 33 33 33 34 34 34 34 34 34 34 34 34 34 34 34 34 34
1-1-1-1-1-1-1 1-1-1-1-1-1-1 1-1-1-1-1-1-1 1-1-1-1-1-1-1 1-1-1-1-1-1-1 1-1-1-1-1-1-1 1-1-1-1-1-1-1 1-1-1-1-1-1-1 1-1-1-1-1-1-1 1-1-1-1-1-1-1 1-1-1-1-1-1-1 1-1-1-1-1-1-1 1-1-1-1-1-1-1 1-1-1-1-1-1-1 1-1-1-1-1-1-1 1-1-1-1-1-1-1 1-1-1-1-1-1-1 1-1-1-1-1-1-1 1-1-1-1-1-1-1 1-1-1-1-1-1-1 1-1-1-1-1-1-1 1-1-1-1-1-1-1 1-1-1-1-1-1-1 1-1-1-1-1-1-1 1-1-1-1-1-1-1 1-1-1-1-1-1-1 1-1-1-1-1-1-1 1-1-1-1-1-1-1 1-1-1-1-1-1-1 1-1-1-1-1-1-1 1-1-1-1-2-1-1 1-1-1-1-2-1-1 1-1-1-1-2-1-1 1-1-1-1-2-1-1 1-1-1-1-2-1-1 1-1-1-1-2-4-1 1-1-1-1-3-1-4 1-1-1-1-4-1-1 1-1-6-1-2-1-1 1-1-7-1-1-1-1 1-1-8-1-1-1-4 1-4-1-1-1-1-1 2-1-1-1-1-1-1 1-1-1-1-1-12-5 1-1-1-1-1-12-5 1-1-1-1-1-12-5 1-1-1-1-1-13-5 1-1-1-1-1-1-4 1-1-1-1-1-1-4 1-1-1-1-1-1-4 1-1-1-1-1-1-4 1-1-1-1-1-1-4 1-1-1-1-1-1-4 1-1-1-1-1-1-4 1-1-1-1-1-1-5 1-1-1-1-1-1-5 1-1-1-1-1-1-5 1-1-1-1-1-1-5 1-1-1-1-1-1-5 1-1-1-1-1-1-5 1-1-1-1-1-1-5 1-1-1-1-1-1-5 1-1-1-1-1-1-5 1-1-1-1-1-1-5 1-1-1-1-1-1-5 1-1-1-1-1-1-5 1-1-1-1-1-1-5 1-1-1-1-1-1-5
Female horse Female horse Female horse Female horse Female horse Female horse Unknown Horse Female horse (1) Female horse (1) Horse Female horse (2) Horse Horse Horse (3) Horse (3) Female horse (2) Horse Horse Horse Horse Horse Horse Horse Horse Horse Horse Unknown Unknown Unknown Male horse Horse Horse Horse Horse Horse Horse Horse Unknown Horse Horse Female horse Female horse Male horse (6) Female horse (4) Female horse (4) Female horse (4) Horse Horse Horse Horse Unknown Unknown Male horse (5) Horse Horse Horse Male horse Female horse Female horse (7) Female horse (7) Female horse Male horse Male horse (8) Male horse (8) Male horse (9) Male horse Male horse
US Australia Australia Australia Australia Australia Unknown UK France France France France France France France France France France France France France France France France France France France France France France France France France France France France France France France France France US Australia France France France France France France France France France France France (1) France France France France France France France France France France (2) France (2) France (2) France (2) France (2)
1978 1978 1981 1981 1981 1979 Unknown 1977 2005 2005 2004 2005 2002 2002 2001 2001 2005 2001 2001 2001 1999 1999 1999 1999 1999 1999 2001 2000 1998 1998 2005 2001 1999 1999 2000 2000 1999 1998 2000 2001 2000 1978 1978 2010 2011 2011 2011 2001 1998 1998 1999 1998 1998 2010 2004 2004 2004 2005 2002 2001 2001 1999 2005 2007 2009 2009 2009 2009
612
F. Duquesne et al. / Veterinary Microbiology 167 (2013) 609–618
Table 1 (Continued ) Strain (other reference name)
eBURST group
ST
Allelic profilec
Host sourced
Countrye
Year
MCE533 MCE534 MCE535 MCE544, MCE545a MCE554 MCE69 MCE314 MCE536 MCE1 MCE15 MCE486, MCE487, MCE488a MCE492 MCE493 MCE502 MCE504 MCE505 MCE529 MCE557 MCE558 MCE559, MCE560a MCE90 MCE98 MCE57 MCE99 MCE25 MCE66 MCE67 MCE88 MCE89 MCE96 MCE546 MA09 (EQ59) MA10 (EQ70) MA11 MA12 MA13 MCE84 MCE95 MA08 (SS28) MCE556
1 1 1 1 1 1 1 2 2 2 2 2 2 2 2 2 2 2 2 2 3 3 4 4
34 34 34 34 34 35 39 15 16 16 16 16 16 16 16 16 16 16 16 16 10 11 17 18 4 4 4 4 4 4 4 3 3 3 3 3 5 5 9 31
1-1-1-1-1-1-5 1-1-1-1-1-1-5 1-1-1-1-1-1-5 1-1-1-1-1-1-5 1-1-1-1-1-1-5 1-1-1-1-1-1-6 11-1-1-1-1-1-5 1-2-4-1-2-2-4 1-2-4-1-2-3-4 1-2-4-1-2-3-4 1-2-4-1-2-3-4 1-2-4-1-2-3-4 1-2-4-1-2-3-4 1-2-4-1-2-3-4 1-2-4-1-2-3-4 1-2-4-1-2-3-4 1-2-4-1-2-3-4 1-2-4-1-2-3-4 1-2-4-1-2-3-4 1-2-4-1-2-3-4 1-1-5-3-1-1-1 1-1-5-3-5-1-1 1-3-6-1-2-2-4 1-3-9-1-2-2-4 1-2-5-3-2-2-3 1-2-5-3-2-2-3 1-2-5-3-2-2-3 1-2-5-3-2-2-3 1-2-5-3-2-2-3 1-2-5-3-2-2-3 1-2-5-3-2-2-3 1-1-3-2-2-2-3 1-1-3-2-2-2-3 1-1-3-2-2-2-3 1-1-3-2-2-2-3 1-1-3-2-2-2-3 1-2-1-1-2-2-3 1-2-1-1-2-2-3 1-1-2-1-1-2-2 1-11-2-1-14-3-3
Male horse Male horse (9) Male horse Male horse (6) Female horse (4) Male donkey Male horse Female horse with signs of metritis Male horse Male horse Male horse Male horse (10) Male horse Female horse with signs of infertility Male horse (10) Male horse (5) Male horse Male horse (5) Male horse (11) Male horse (11) Horse Unknown Horse Horse Horse Horse Horse Female horse Horse Horse Horse Female horse Female horse Female horse Female horse Female horse Horse Unknown Female horse Horse
France France France France France France France France France France France France France France France France France France France France France France France France France France France France France France France Japan Japan Japan Japan Japan France France Japan UAE
2009 2009 2009 2010 2011 1999 2005 2009 2004 2005 2007 2007 2007 2008 2008 2008 2009 2012 2012 2012 2001 2001 2001 2001 2004 1999 1999 2001 2001 2001 2010 1980 1980 1996 1996 1996 2001 2001 1980 2009
5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 6 6 6 6 6
22 22 22 22 22 22 22 22 22 22 22 22 22 22 22 22 22 22 22 22 22 22 26 37 21 21 21 21 21
4-9-10-4-11-5-9 4-9-10-4-11-5-9 4-9-10-4-11-5-9 4-9-10-4-11-5-9 4-9-10-4-11-5-9 4-9-10-4-11-5-9 4-9-10-4-11-5-9 4-9-10-4-11-5-9 4-9-10-4-11-5-9 4-9-10-4-11-5-9 4-9-10-4-11-5-9 4-9-10-4-11-5-9 4-9-10-4-11-5-9 4-9-10-4-11-5-9 4-9-10-4-11-5-9 4-9-10-4-11-5-9 4-9-10-4-11-5-9 4-9-10-4-11-5-9 4-9-10-4-11-5-9 4-9-10-4-11-5-9 4-9-10-4-11-5-9 4-9-10-4-11-5-9 4-9-10-8-11-5-9 4-9-10-4-11-5-12 3-7-12-6-10-6-7 3-7-12-6-10-6-7 3-7-12-6-10-6-7 3-7-12-6-10-6-7 3-7-12-6-10-6-7
Male donkey Male donkey Male donkey Male donkey Male donkey Male donkey Male donkey Male donkey Male donkey Male donkey Male donkey (12) Male donkey (12) Male donkey (12) Male donkey (12) Male horse Male donkey (13) Male donkey Male donkey (14) Male donkey (13) Male donkey (13) Male donkey (13) Male donkey (13) Male donkey (14) Male horse Male donkey (15) Male donkey (15) Male donkey Female horse Male donkey
France France France France France France France France France France France France France France France France France France France France France France France France France France France France France
50 T. asinigenitalis strains MCE4 MCE10 MCE11 MCE12 MCE13 MCE14 MCE21 MCE22 MCE23 MCE24 MCE85 MCE92 MCE93 MCE97 MCE235 MCE265 MCE475 MCE476 MCE477 MCE514, MCE515a MCE516, MCE517, MCE518a MCE519, MCE520, MCE521a MCE266 MCE59 MCE2b MCE6, MCE7a MCE47 MCE86 MCE473
(2) (2)
(2)
(1) (1) (1) (1) (1)
(3) (3) (3) (3)
(4) (3) (4) (4) (4) (4) (4) (4) (5) (5) (5) (5)
2004 2005 2005 2005 2005 2005 2004 2004 2004 2004 2001 2001 2001 2001 1995 2005 2006 2006 2006 2008 2008 2008 2005 2001 2004 2005 2001 2001 2006
F. Duquesne et al. / Veterinary Microbiology 167 (2013) 609–618
613
Table 1 (Continued ) Strain (other reference name)
eBURST group
ST
Allelic profilec
Host sourced
Countrye
Year
MCE474 MCE3b MCE94 MCE479 MCE480 MCE46 MCE73 MCE76 MCE124 MCE549 MCE550, MCE551a MCE20 MCE497 (UCD-1, ATCC700933, CIP107673T) MCE513 (Bd3751/05)
6 6 7 7 7
21 36 25 38 38 23 23 23 23 24 24 27 28 29
3-7-12-6-10-6-7 3-7-12-6-10-6-8 10-8-16-7-12-11-13 10-8-16-7-12-11-14 10-8-16-7-12-11-14 6-6-11-4-6-8-11 6-6-11-4-6-8-11 6-6-11-4-6-8-11 6-6-11-4-6-8-11 9-9-13-11-9-8-17 9-9-13-11-9-8-17 5-5-14-5-13-7-10 7-10-15-9-7-9-15 8-10-17-10-8-10-16
Male Male Male Male Male Male Male Male Male Male Male Male Male Male
France France (5) France France (5) France (5) France France France France France France France US Sweden
2006 2004 2005 2006 2006 2001 1999 1999 1998 2011 2011 2004 1997 2004
donkey donkey (15) donkey (15) horse donkey (15) donkey donkey donkey donkey horse (16) horse (16) donkey donkey horse
a
Strains isolated during the same test from different sampling sites. MCE2 and MCE3 strains were isolated during the same test from different sampling sites. Allelic profile of gltA-gyrB-fh-shmt-tyrB-adk-txn loci, respectively. d The numbers in parentheses are arbitrarily attributed to animals remaining CEM-positive after being sampled several times in a single year or over a period of several years. e The numbers in parentheses are arbitrarily attributed to farms with at least two animals that remained CEM-positive; US, United States; UK, United Kingdom; UAE, United Arab Emirates. b c
primer pairs were designed from the MCE9 and MCE3 genomes using DNASTAR Lasergene1 software (version 5.51) to amplify 455–682 bp internal gene fragments (Table 2). Each 50-ml PCR was carried out in a Bio-Rad Laboratories MyCycler Thermal Cycler1 using 2 ml genomic DNA extraction, 10 pM [each] forward and reverse primers, 25 ml PCR master mix containing 1.5 mM MgCl2, 200 mM [each] deoxynucleotide triphosphates and 1.25 U Thermoprime Taq DNA polymerase (Thermo Scientific, France), and PCR-grade water. All primer pairs were designed to ensure they had similar
annealing temperatures. Amplification conditions were as follows: initial denaturation at 95 8C for 5 min and 35 cycles of denaturation at 94 8C for 30 s, primer annealing at 55 8C for 1 min and extension at 72 8C for 1 min. Amplification products were sequenced by Beckman Coulter Genomics (Stansted, UK). The forward and reverse sequences of a given locus were edited, aligned and trimmed to the desired length (MLST fragment, Table 2) using the http://multalin.toulouse.inra.fr/multalin/ website and MEGA v5.05 software (http://www.megasoft ware.net).
Table 2 Gene fragments and primer details for Taylorella sp. MLST. Locusa Gene ID accessionb PCR and sequencing primers (50 !30 )c Amplicon size (bp) Locus size (bp) MLST fragment start (50 )/stop (30 ) T. equigenitalis
T. asinigenitalis
gltA
11186031
GCTCAGACAGGCATGTTTACTTA (F) GTCCCCATAGGCAAGAAATAC (R)
682
623
ATGTTTAC/TCAGGTAC ATGTTTAC/TCAGGTAC
gyrB
10114465
GTTGCTATGCAATGGACTGA (F) GCACGTGATTTTTCTACATT (R)
614
518
AATAATAT/CAAGCAAT AATAACAT/CAAGCAAT
fh
10114863
AGAGGCTTGCGATGAGG (F) AGTCAGGAGCCCACTCTAC (R)
659
574
TGCGTTTA/TCGTTAGA TGCGAATG/TCTCTAGA
shmt
10114226
GACATTGCCCACTATTCAGG (F) CTCTTGTAGTCATAGCTGGTGTTC (R)
511
435
TATTCAGG/TACCTAAT TATTCAGG/TTCCAAAC
tyrB
10115298
CCTATTTTGGGGCTTTCTGACT (F) GGTCTACGCCAGTAGGATT (R)
532
462
TTGGTGTT/TAATCCTA TAGGTGTT/CAATCCTA
adk
10115205
ATGCGTCTAATTCTTCTTG (F) AACGTCTCCGACACCATT (R)
615
538
TAAATACA/TAAATGGT TAAATACA/TCAATGGT
txn
10115427
ACGGGGGACCGCATAAAGCC (F) AGCGTTTCTGTACCCGTGCGA (R)
455
371
ATTTATGG/ATTCGCAC ATTTATGG/ATTCGCAC
a gltA, citrate synthase; gyrB, gyrase subunit B; fh, putative fumarate hydratase; shmt, serine hydroxymethyltransferase; tyrB, tyrosine aminotransferase; adk, adenylate kinase; txn, thioredoxine. b From T. equigenitalis MCE9 genome (He´bert et al., 2011). c F, forward; R, reverse.
F. Duquesne et al. / Veterinary Microbiology 167 (2013) 609–618
614
2.4. Data analysis For each locus, all the sequences were compared and an allelic number was assigned to each unique allele for a given locus. The allelic profile of each strain was generated by combining the allelic numbers of each of the seven loci in the following order: gltA, gyrB, fh, shmt, tyrB, adk, and txn. A novel sequence type (ST) designation was given to each unique allelic profile; strains with the same allelic profile belonged to the same ST. The number of polymorphic nucleotide sites was calculated using MEGA v5.05 software. For each locus, the degree of selective pressure was assessed by calculating the dN/dS ratio of nonsynonymous substitutions (dN) to synonymous substitutions (dS) using START v2 software (http://pubmlst.org/software/analysis/start2). A dN/dS ratio higher than 1 implies selection for amino acid changes (Yang and Bielawski, 2000). Multilocus linkage disequilibrium was assessed using the index of association (IA) using START v2 software. In case of linkage disequilibrium because of frequent recombination events, the expected value of IA is zero. Clonal populations are identified by an IA that differs significantly from zero (Smith et al., 1993). The seven allelic sequences for each strain were concatenated (3521 bp) to generate a neighbor-joining phylogenetic tree, with 1000 bootstrap replicates, using the Jukes Cantor algorithm in the MEGA v5.05 software. Using the eBURST v3 program, the strains were grouped into clonal complexes according to the MLST definition: a cluster of strains sharing at least six of the seven alleles sequenced (www.mlst.net) (Jolley and Maiden, 2010). STs were classified as single-locus variants (SLVs), double-locus variants (DLVs) or individual unlinked STs. 3. Results 3.1. Variation of the seven MLST loci The sequence of the seven chosen loci (gltA, gyrB, fh, shmt, tyrB, adk, and txn) was determined for the 113 T. equigenitalis and 50 T. asinigenitalis strains isolated (Table 1), and allelic profiles were assigned. As shown in Table 2, the mean average allele length for each locus was 503 pb and ranged from 371 pb (txn) to 623 pb (gltA). All alleles for a given locus were of equal length and, to aid further analysis, were in the correct reading
frame. Among the 163 investigated isolates, 3 (gltA and shmt) to 9 alleles (fh) were present at each locus for T. equigenitalis and 6 (gyrB) to 11 alleles (txn) were present at each locus for T. asinigenitalis (mean average = 5.4 and 8.0, respectively). The proportion of variable nucleotide sites at each locus varied from 0.5% (gltA and shmt) to 6.1% (tyrB), and from 1.3% (gyrB) to 6.7% (txn) for T. equigenitalis and T. asinigenitalis, respectively (mean average = 1.9% and 4.4%) (Table 3; Fig. S1). Supplementary material related to this article can be found, in the online version, at http://dx.doi.org/10.1016/ j.vetmic.2013.09.016. The dN/dS ratios were calculated to determine the degree of selective pressure. For each locus, the dN/dS ratios were less than 1 (Table 3), indicating that no strong positive selective pressure was present at any selected loci, validating their suitability for inclusion into the Taylorella sp. MLST scheme. IA was calculated to estimate the degree of association and recombination between alleles at different loci, based on the allelic profile data. The IA computed using the all 163 isolates was 3.64 (p < 0.001). This indicates significant linkage disequilibrium and suggests a clonal population structure in the Taylorella genus and limited recombinational events, even between the glta and txn loci that are separated by only 7 kb. The IA remained significantly different from zero (2.21) when only one representative of each ST was included in the computation, which rules out any bias due to taxonomic sampling (p < 0.001). The calculated IA values for T. equigenitalis and T. asinigenitalis populations were 1.9845 (p < 0.001) and 1.6428 (p < 0.001), respectively. 3.2. Relatedness of Taylorella sp. strains A total of 27 and 12 STs were respectively assigned to the 113 T. equigenitalis and 50 T. asinigenitalis strains used in this study (Table 1). The most common STs were identified 30 (ST-1), 20 (ST-34) and 15 (ST-16) times among the 113 T. equigenitalis strains, and 27 (ST-22) and 7 (ST-21) times among the 50 T. asinigenitalis strains. No STs or alleles were common between T. equigenitalis and T. asinigenitalis. The neighbor-joining tree of concatenated gltA, gyrB, fh, shmt, tyrB, adk, and txn sequences (3521 bp) showed that both species are divided into two distinct groups and that the genetic diversity is more important for T. asinigenitalis than for T. equigenitalis (Fig. S2).
Table 3 Analysis of the seven loci in the Taylorella sp. population of 163 strains. Locus
gltA gyrB fh shmt tyrB adk txn a
MLST fragment size (bp)
623 518 574 435 462 538 371
T. equigenitalis
T. asinigenitalis
No. of alleles
Polymorphic sites, no (%)
dN/dSa
No. of alleles
Polymorphic sites, no (%)
dN/dS
3 5 9 3 6 6 6
3 4 13 2 28 5 8
0.1495 0.2369 0.1313 0.0000 0.7317 0.1723 0.6658
8 6 8 8 8 7 11
27 7 26 19 19 30 25
0.0226 0.0000 0.0610 0.0377 0.0266 0.0631 0.2334
(0.5) (0.8) (2.3) (0.5) (6.1) (0.9) (2.2)
dN/dS: ratio of nonsynonymous (dN) to synonymous (dS) substitutions.
(4.3) (1.3) (4.5) (4.4) (4.1) (5.6) (6.7)
F. Duquesne et al. / Veterinary Microbiology 167 (2013) 609–618
615
Fig. 1. Population snapshot of the Taylorella genus. Clusters of related STs (corresponding to clonal complexes) and individual unlinked STs within the study strains are displayed as a single eBURST diagram by setting the group definition to zero of seven shared alleles. STs are represented as circles; the size of a circle is proportional to the number of strains within a given ST. Primary founders and subgroup founders (cofounders) are respectively shown in white and gray circles. Single-locus variant links are represented as black lines. Each clonal complex was defined by a number in boldface type (CC1–CC7). Distances between STs, indicated by connecting lines, are arbitrary.
Supplementary material related to this article can be found, in the online version, at http://dx.doi.org/10.1016/ j.vetmic.2013.09.016. The likely patterns of chromosomal evolutionary descent and identity of the probable founding genotypes were inferred from an eBURST analysis (Feil et al., 2004) of the MLST data where the number of resamplings made for bootstrapping was 1000 and only SLVs were considered. Using these parameters seven clonal complexes were identified among the 39 STs (Table 1; Fig. 1). In details, the 27 T. equigenitalis STs were grouped into four clonal complexes (CC1–CC4) and five individual unlinked STs, and the 12 T. asinigenitalis STs into three clonal complexes (CC5–CC7) and five individual unlinked STs. The largest T. equigenitalis clonal complex (CC1) contained 16 STs (68.1% of the T. equigenitalis tested) with the predicted founder ST1 and a cofounded ST-34. Each remaining T. equigenitalis clonal complex contained 2 STs with 1.8% (CC3 and CC4) to 14.2% (CC2) of the tested T. equigenitalis. The largest T. asinigenitalis clonal complex (CC5) contained 3 STs (58.0% of the tested T. asinigenitalis), and the ST-22 could be considered as the predicted founder. Each of the remaining T. asinigenitalis clonal complexes contained 2 STs with 6.0% (CC7) to 16.0% (CC6) of the tested T. asinigenitalis. The individual unlinked STs represented 14.2% of T. equigenitalis strains and 20.0% of T. asinigenitalis strains.
When DLVs were considered, the eBURST analysis showed that (i) CC1 and CC3 are linked since ST-10 has two DLVs with ST-1 and ST-13, and (ii) CC2, CC4, ST-4 and ST-5 are linked since ST-5 has two DLVs with ST-4 and ST-15, and ST-15 has three DLVs with ST-5, ST-17 and ST-18 (Fig. S3). Supplementary material related to this article can be found, in the online version, at http://dx.doi.org/10.1016/ j.vetmic.2013.09.016. 3.3. Relationship between clonal complex and epidemiologic data The data set presented in the study was not specifically designed to investigate the relationship between molecular diversity and host or geographical origin of the strains. However, the following points can be emphasized: (i) the ST-1 contains T. equigenitalis strains from Australia (1978 and 1981), France (1998– 2011), UK (1977) and US (1977 and 1978), (ii) the Japanese and Emirati T. equigenitalis strains are classified into unlinked STs (ST-3 and ST-9 for Japanese strains and ST-31 for the Emirati strain), (iii) the two non-French T. asinigenitalis strains are classified into unlinked STs (ST28 for the American strain and ST-29 for the Swedish strain).
616
F. Duquesne et al. / Veterinary Microbiology 167 (2013) 609–618
Among the five farms where several animals were identified as CEM-positive, one had strains with a common ST, two had strains with different STs of the same clonal complex (farms no. 2 and 4), and two had strains with different STs of different clonal complexes (farms no. 1 and 5). Among the 16 animals identified as CEM-positive from several samples taken during the same year or over several years, three carried strains with different STs from the same clonal complex (animals no. 4, 6 and 14) and two carried strains with different STs from two clonal complexes (animals no. 5 and 15). The only T. equigenitalis strain (MCE69) isolated from a donkey was classified in the CC1, like 68.1% of the T. equigenitalis strains, and one can note that the five T. asinigenitalis strains isolated from horses do not form a separate clonal complex among the 50 T. asinigenitalis strains. 4. Discussion MLST is a well-established molecular typing method that has been successfully used for the determination of the population structure of many bacterial species (Maiden, 2006; Maiden et al., 1998). In this study, we used the recent genome sequences of T. equigenitalis and T. asinigenitalis (He´bert et al., 2011, 2012) to develop an unambiguous and discriminatory Taylorella sp. MLST scheme based on the allelic variations of seven housekeeping genes common to both bacterial species. Internal fragments of the seven loci selected for the final MLST scheme (gltA, gyrB, fh, shmt, tyrB, adk, and txn) were amplified from 113 T. equigenitalis and 50 T. asinigenitalis strains. Despite a notable genetic divergence between T. equigenitalis and T. asinigenitalis (He´bert et al., 2011, 2012), the selection of the housekeeping genes was validated by a dN/dS ratio lower than 1 for all of them, indicating a very low contribution of environmental selection to the variation of these genes (Table 3). The data generated have been used to construct a publicly open and curated MLST website with the Taylorella sp. MLST scheme and data (http://pubmlst.org/taylorella/). In terms of data validation, the sequence chromatograms produced by automated sequencers can be sent by email to the Taylorella sp. MLST website curator to be checked before publication. Based on calculated IA values, it appears that the Taylorella population can be classified as clonal (Smith et al., 1993), indicating an absence or a low probability of intra- and inter-species recombination. This observation is consistent with the presence of a conserved synteny (He´bert et al., 2012), and the absence of common ST or allele identified between T. equigenitalis and T. asinigenitalis. Despite 2.3 times more strains of T. equigenitalis than T. asinigenitalis used in this study, the genetic diversity of T. equigenitalis appeared to be lower than that of T. asinigenitalis (Table 3). This observation is confirmed by a higher IA value for the T. equigenitalis population (1.9845, p < 0.001) than for the T. asinigenitalis population (1.6428, p < 0.001), and suggests that the T. equigenitalis population has a more ‘‘epidemic’’ structure than T. asinigenitalis (Smith et al., 1993).
The analysis of MLST results among the 163 strains tested from several countries over a 35-year period of time showed the presence of 39 STs, including 39% of unlinked STs (Fig. 1). The emergence of new STs over time suggests that Equidae carrying these strains were contaminated by an external source of Taylorella originating from a still unidentified natural ecological reservoir. Moreover, these emergences were observed in different countries all over the world (Table 1), suggesting that this potential reservoir is spread worldwide. The first CEM outbreaks, that emerged simultaneously in the late 1970s in several countries (at least UK, US and Australia according the present results), were reported in the literature as the cause of significant economic impact in the horse breeding industry because of a rapid worldwide spread and the presence of characteristic clinical signs of reproductive tract disease in mares (Timoney, 2011). The present results revealed that these first CEM outbreaks were associated with the T. equigenitalis founding clonal complex CC1 confirming an epidemiological relationship and suggesting that these strains are originated from a single initial bacterial clone of unknown origin. Interestingly, in France, T. equigenitalis strains associated with the CC1 are still being isolated from mares showing few or no clinical signs (Table 1). This observation suggests that the initially virulent strain lost its virulence traits in favor of persistence traits allowing it to be asymptomatically carried for years, as suggested for other pathogens and symbionts (Marchetti et al., 2010; Sachs et al., 2011). Moreover, with the exception of the CC1, all CCs and STs contain strains originated from a single country. This observation suggests a lower international dissemination of these strains compare to CC1 strains, and suggests that international prophylactic measures taken to prevent dissemination of CEM over the world after the first CEM outbreak have been generally effective. However, more data on international CEM strains remain necessary to evaluate this effectiveness. In this context, it remains important to maintain these prophylactic measures, to prevent the dissemination of new epidemic clones. For example, the CC2, circulating in France since the 2000s, should be carefully monitored to limit its spread to other parts of the world, since the two females carrying a T. equigenitalis strain associated with the CC2 developed signs of reproductive tract disease (strains MCE502 and MCE536; Table 1). Comparisons are frequently drawn between the levels of resolution provided by MLST and PFGE. Many studies showed similar levels of discrimination between these techniques (Godoy et al., 2003; Peacock et al., 2002) but MLST was found to be more discriminatory in other studies (Kotetishvili et al., 2003, 2002; Revazishvili et al., 2004). In previous studies, 130 Japanese T. equigenitalis strains, including MA08 to MA13 strains, isolated from 1980 to 1996 were examined by PFGE and appeared to have a common genotype, designed ‘‘genotype J’’, suggesting a common source for all of them (Matsuda et al., 1999; Miyazawa et al., 1995). Here, we show that the MA08 (SS28) strain isolated in 1980 does not have the same ST as the other Japanese strains (Table 1), suggesting actually
F. Duquesne et al. / Veterinary Microbiology 167 (2013) 609–618
two separate outbreaks in Japan during the early 1980s. Moreover, whereas the previous 16S rDNA analysis of T. asinigenitalis strains separated them into two profiles (Ba˚verud et al., 2006; Breuil et al., 2011), the present Taylorella sp. MLST allows refinement of the phylogenetic analysis of T. asinigenitalis into 12 STs. The T. asinigenitalis 16S rDNA group 1 (Breuil et al., 2011) is exclusively associated with the clonal complex 5 while the T. asinigenitalis 16S rDNA group 2 (Breuil et al., 2011) has a genetic variability associated with two clonal complexes (CC6 and CC7) and five unlinked STs (Fig. 1). Altogether, these results allow to conclude that the Taylorella sp. MLST scheme developed in this study is a powerful tool for molecular epidemiological studies of the Taylorella genus. The use of the MLST database by other researchers from all around the world will increase the representatively of these results, thus offering a new way to study the population biology of these bacteria. Acknowledgements This publication made use of the MLST website (http:// pubmlst.org/mlst/) developed by Keith Jolley and situated at the University of Oxford; the development of this site was funded by the Wellcome Trust (Jolley and Maiden, 2010). We are very grateful to the French veterinary laboratories approved for CEM diagnosis that contributed the Taylorella strains used in this study, to Viveca Ba˚verud (National Veterinary Institute, Sweden) for providing the T. asinigenitalis Bd3751/05 strain, and to Ulrich Wernery (Central Veterinary Research Laboratory, United Arab Emirates) for providing the Emirati T. equigenitalis strain. We also would like to thank Christine Delorme (INRA Jouyen-Josas, France) for her technical assistance as well as Dana Pottratz and Guillaume He´bert from SC Partners for their editorial assistance. This work was supported in part by the European Union in conjunction with the European Union Reference Laboratory for equine diseases. References Aalsburg, A.M., Erdman, M.M., 2011. Pulsed-field gel electrophoresis genotyping of Taylorella equigenitalis isolates collected in the United States from 1978 to 2010. J. Clin. Microbiol. 49, 829–833. Ba˚verud, V., Nystrom, C., Johansson, K.E., 2006. Isolation and identification of Taylorella asinigenitalis from the genital tract of a stallion, first case of a natural infection. Vet. Microbiol. 116, 294–300. Berriman, M., Rutherford, K., 2003. Viewing and annotating sequence data with Artemis. Brief. Bioinform. 4, 124–132. Bleumink-Pluym, N., ter Laak, E.A., van der Zeijst, B.A., 1990. Epidemiologic study of Taylorella equigenitalis strains by field inversion gel electrophoresis of genomic restriction endonuclease fragments. J. Clin. Microbiol. 28, 2012–2016. Breuil, M.F., Duquesne, F., Laugier, C., Petry, S., 2011. Phenotypic and 16S ribosomal RNA gene diversity of Taylorella asinigenitalis strains isolated between 1995 and 2008. Vet. Microbiol. 148, 260–266. Crowhurst, R.C., 1977. Genital infection in mares. Vet. Rec. 100, 476. Duquesne, F., Pronost, S., Laugier, C., Petry, S., 2007. Identification of Taylorella equigenitalis responsible for contagious equine metritis in equine genital swabs by direct polymerase chain reaction. Res. Vet. Sci. 82, 47–49. Enright, M.C., Day, N.P., Davies, C.E., Peacock, S.J., Spratt, B.G., 2000. Multilocus sequence typing for characterization of methicillin-resistant and methicillin-susceptible clones of Staphylococcus aureus. J. Clin. Microbiol. 38, 1008–1015.
617
Enright, M.C., Spratt, B.G., 1999. Multilocus sequence typing. Trends Microbiol. 7, 482–487. Erdman, M.M., Creekmore, L.H., Fox, P.E., Pelzel, A.M., Porter-Spalding, B.A., Aalsburg, A.M., Cox, L.K., Morningstar-Shaw, B.R., Crom, R.L., 2011. Diagnostic and epidemiologic analysis of the 2008–2010 investigation of a multi-year outbreak of contagious equine metritis in the United States. Prev. Vet. Med. 101, 219–228. Feil, E.J., Li, B.C., Aanensen, D.M., Hanage, W.P., Spratt, B.G., 2004. eBURST: inferring patterns of evolutionary descent among clusters of related bacterial genotypes from multilocus sequence typing data. J. Bacteriol. 186, 1518–1530. Godoy, D., Randle, G., Simpson, A.J., Aanensen, D.M., Pitt, T.L., Kinoshita, R., Spratt, B.G., 2003. Multilocus sequence typing and evolutionary relationships among the causative agents of melioidosis and glanders, Burkholderia pseudomallei and Burkholderia mallei. J. Clin. Microbiol. 41, 2068–2079. He´bert, L., Moumen, B., Duquesne, F., Breuil, M.F., Laugier, C., Batto, J.M., Renault, P., Petry, S., 2011. Genome sequence of Taylorella equigenitalis MCE9, the causative agent of contagious equine metritis. J. Bacteriol. 193, 1785. He´bert, L., Moumen, B., Pons, N., Duquesne, F., Breuil, M.F., Goux, D., Batto, J.M., Laugier, C., Renault, P., Petry, S., 2012. Genomic characterization of the Taylorella genus. PLoS One 7, e29953. Jang, S.S., Donahue, J.M., Arata, A.B., Goris, J., Hansen, L.M., Earley, D.L., Vandamme, P.A., Timoney, P.J., Hirsh, D.C., 2001. Taylorella asinigenitalis sp. nov. a bacterium isolated from the genital tract of male donkeys (Equus asinus). Int. J. Syst. Evol. Microbiol. 51, 971–976. Jolley, K.A., Maiden, M.C., 2010. BIGSdb: scalable analysis of bacterial genome variation at the population level. BMC Bioinform. 11, 595. Katz, J.B., Evans, L.E., Hutto, D.L., Schroeder-Tucker, L.C., Carew, A.M., Donahue, J.M., Hirsh, D.C., 2000. Clinical, bacteriologic, serologic, and pathologic features of infections with atypical Taylorella equigenitalis in mares. J. Am. Vet. Med. Assoc. 216, 1945–1948. Kotetishvili, M., Stine, O.C., Chen, Y., Kreger, A., Sulakvelidze, A., Sozhamannan, S., Morris Jr., J.G., 2003. Multilocus sequence typing has better discriminatory ability for typing Vibrio cholerae than does pulsed-field gel electrophoresis and provides a measure of phylogenetic relatedness. J. Clin. Microbiol. 41, 2191–2196. Kotetishvili, M., Stine, O.C., Kreger, A., Morris Jr., J.G., Sulakvelidze, A., 2002. Multilocus sequence typing for characterization of clinical and environmental Salmonella strains. J. Clin. Microbiol. 40, 1626–1635. Luddy, S., Kutzler, M.A., 2010. Contagious equine metritis Within the United States: a review of the 2008 outbreak. J. Equine Vet. Sci. 30, 393–400. Maiden, M.C., 2006. Multilocus sequence typing of bacteria. Annu. Rev. Microbiol. 60, 561–588. Maiden, M.C., Bygraves, J.A., Feil, E., Morelli, G., Russell, J.E., Urwin, R., Zhang, Q., Zhou, J., Zurth, K., Caugant, D.A., Feavers, I.M., Achtman, M., Spratt, B.G., 1998. Multilocus sequence typing: a portable approach to the identification of clones within populations of pathogenic microorganisms. Proc. Natl. Acad. Sci. U.S.A. 95, 3140–3145. Marchetti, M., Capela, D., Glew, M., Cruveiller, S., Chane-Woon-Ming, B., Gris, C., Timmers, T., Poinsot, V., Gilbert, L.B., Heeb, P., Medigue, C., Batut, J., Masson-Boivin, C., 2010. Experimental evolution of a plant pathogen into a legume symbiont. PLoS Biol. 8, e1000280. Matsuda, M., Miyazawa, T., Anzai, T., 1999. Detection of a common genotype among strains of Taylorella equigenitalis isolated from thoroughbred horses in Japan between 1994 and 1996. J. Basic Microbiol. 39, 127–130. Miyazawa, T., Matsuda, M., Isayama, Y., Samata, T., Ishida, Y., Ogawa, S., Takei, K., Honda, M., Kamada, M., 1995. Genotyping of isolates of Taylorella equigenitalis from thoroughbred brood mares in Japan. Vet. Res. Commun. 19, 265–271. OIE, 2012. Contagious Equine Metritis. Manual of Diagnostic Tests and Vaccines for Terrestrial Animals. Chap. 2.5.2. Peacock, S.J., de Silva, G.D., Justice, A., Cowland, A., Moore, C.E., Winearls, C.G., Day, N.P., 2002. Comparison of multilocus sequence typing and pulsed-field gel electrophoresis as tools for typing Staphylococcus aureus isolates in a microepidemiological setting. J. Clin. Microbiol. 40, 3764–3770. Revazishvili, T., Kotetishvili, M., Stine, O.C., Kreger, A.S., Morris Jr., J.G., Sulakvelidze, A., 2004. Comparative analysis of multilocus sequence typing and pulsed-field gel electrophoresis for characterizing Listeria monocytogenes strains isolated from environmental and clinical sources. J. Clin. Microbiol. 42, 276–285. Rutherford, K., Parkhill, J., Crook, J., Horsnell, T., Rice, P., Rajandream, M.A., Barrell, B., 2000. Artemis: sequence visualization and annotation. Bioinformatics 16, 944–945.
618
F. Duquesne et al. / Veterinary Microbiology 167 (2013) 609–618
Sachs, J.L., Skophammer, R.G., Regus, J.U., 2011. Evolutionary transitions in bacterial symbiosis. Proc. Natl. Acad. Sci. U.S.A. 108 (Suppl. 2) 10800–10807. Selander, R.K., Caugant, D.A., Ochman, H., Musser, J.M., Gilmour, M.N., Whittam, T.S., 1986. Methods of multilocus enzyme electrophoresis for bacterial population genetics and systematics. Appl. Environ. Microbiol. 51, 873–884. Smith, J.M., Smith, N.H., O’Rourke, M., Spratt, B.G., 1993. How clonal are bacteria? Proc. Natl. Acad. Sci. U.S.A. 90, 4384–4388. Sugimoto, C., Isayama, Y., Sakazaki, R., Kuramochi, S., 1983. Transfer of Haemophilus equigenitalis Taylor et al. 1978 to the genus Taylorella gen. nov. asTaylorella equigenitalis comb. nov. Curr. Microbiol. 9, 155–162. Thoresen, S.I., Jenkins, A., Ask, E., 1995. Genetic homogeneity of Taylorella equigenitalis from Norwegian trotting horses revealed by chromosomal DNA fingerprinting. J. Clin. Microbiol. 33, 233–234.
Timoney, P.J., 2011. Horse species symposium: contagious equine metritis: an insidious threat to the horse breeding industry in the United States. J. Anim. Sci. 89, 1552–1560. Timoney, P.J., Ward, J., Kelly, P., 1977. A contagious genital infection of mares. Vet. Rec. 101, 103. Wakeley, P.R., Errington, J., Hannon, S., Roest, H.I., Carson, T., Hunt, B., Sawyer, J., Heath, P., 2006. Development of a real time PCR for the detection of Taylorella equigenitalis directly from genital swabs and discrimination from Taylorella asinigenitalis. Vet. Microbiol. 118, 247–254. Yang, Z., Bielawski, J.P., 2000. Statistical methods for detecting molecular adaptation. Trends Ecol. Evol. 15, 496–503.