The ‘Mycoplasma mycoides cluster’ comprises the ruminant pathogens Mycoplasma mycoides subsp. mycoides the causative agent of contagious bovine pleuropneumonia (CBPP), Mycoplasma capricolum subsp. capripneumoniae the agent of contagious caprine pleuropneumonia (CCPP), Mycoplasma capricolum subsp. capricolum, Mycoplasma leachii and Mycoplasma mycoides subsp. capri. CBPP and CCPP are major livestock diseases and impact the agricultural sector especially in developing countries through reduced food-supply and international trade restrictions. In addition, these diseases are a threat to disease-free countries. We used a multilocus sequence typing (MLST) approach to gain insights into the demographic history of and phylogenetic relationships among the members of the ‘M. mycoides cluster’. We collected partial sequences from seven housekeeping genes representing a total of 3,816 base pairs from 118 strains within this cluster, and five strains isolated from wild Caprinae. Strikingly, the origin of the ‘M. mycoides cluster’ dates to about 10,000 years ago, suggesting that the establishment and spread of the cluster coincided with livestock domestication. In addition, we show that hybridization and recombination may be important factors in the evolutionary history of the cluster.
Citation: Fischer A, Shapiro B, Muriuki C, Heller M, Schnee C, et al. (2012) The Origin of the ‘Mycoplasma mycoides Cluster’ Coincides with Domestication of Ruminants. PLoS ONE 7(4): e36150. doi:10.1371/journal.pone.0036150
Editor: Mitchell F. Balish, Miami University, United States of America
Received: February 1, 2012; Accepted: March 27, 2012; Published: April 27, 2012
Copyright: © 2012 Fischer et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: The work was funded by the German Federal Ministry for Economic Cooperation and Development (Project No: 09.7860.1-001.00, Contract No: 81121408, http://www.bmz.de), and by the Swedish International Development Cooperation (SIDA reference: 2007–2510, Contribution No. 75000506, http://www.sida.se). JJ and AF were supported by the German Federal Ministry for Economic Cooperation and Development. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Members of the genus Mycoplasma belong to the most important bacterial livestock pathogens worldwide. Of particular importance are Mycoplasma mycoides subsp. mycoides (Mmm) and Mycoplasma capricolum subsp. capripneumoniae (Mccp), two members of the ‘Mycoplasma mycoides cluster’ , which are responsible for contagious bovine pleuropneumonia (CBPP) and contagious caprine pleuropneumonia (CCPP), respectively. Both diseases cause significant losses in livestock, in particular in Africa and Asia, and are a threat to disease-free countries.
The causative agent of CBPP was cultivated and characterized for the first time by Nocard and Roux in 1898 . CBPP was first recorded in Europe and was introduced into Africa, North America, Australia, and New Zealand during the colonial time period in the 18th and 19th centuries via livestock movement . Today, CBPP is present in sub-Saharan Africa and suspected in Asia.
CCPP was first described in Algeria in 1873 . Its highly contagious nature was acknowledged after an outbreak in South Africa in 1881, which was traced back to the importation of infected goats from Turkey. CCPP is a significant disease of goats in Africa, the Middle East and Western Asia and causes mortalities of up to 80%.
Besides the causative agents of CCPP and CBPP, the ‘M. mycoides cluster’ encompasses additional pathogens, including the bovine pathogen M. leachii  and the small ruminant pathogens M. mycoides subsp. capri (Mmc) and M. capricolum subsp. capricolum (Mcc). Diseases caused by members of the cluster are characterized by clinical symptoms including pneumonia, mastitis, septicaemia, meningitis, wound infections, and arthritis.
Several studies have attempted to resolve the evolutionary relationships between the members of the ‘M. mycoides cluster’ , , , or to infer the evolutionary history of single members within the cluster , , . However, despite the recent publication of complete genome data from single isolates belonging to the different lineages , a comprehensive overview of the evolutionary history of the ‘M. mycoides cluster’ and genetic relationship between populations is still lacking. Here, we partially sequence seven housekeeping genes from all members of the ‘M. mycoides cluster’, spanning their geographic distribution and isolated over the last 100 years. We use these data to infer the recent demographic and evolutionary history of these lineages, to estimate the timing of the origin of each member of the cluster, and correlate the estimated demographic history of the pathogens with that of their hosts.
Genetic relationship between populations
According to our STRUCTURE analysis, the Mycoplasma strains investigated here fall into four distinct populations, three of which belong to the ‘M. mycoides cluster’. These three are Mmm, Mmc, and M. capricolum/M. leachii. The fourth population consist of five strains of an unassigned Mycoplasma species (M. sp.) that were isolated from wild Caprinae. M. leachii appears to be a hybrid between Mmm and M. capricolum, with all 11 individuals showing at least 30% ancestry from Mmm and the remaining from M. capricolum (Figure 1). Strain B144P, isolated from cattle and formerly assigned as Mycoplasma sp. serogroup L , shows additional evidence for hybridization. This strain clusters with M. leachii and shows 60% ancestry from M. capricolum, 30% from Mmm, and 10% from Mmc.
Figure 1. Population structure and phylogenetic relationship of the ‘M. mycoides cluster’ and five non-species assigned Mycoplasma strains.
The 4 populations revealed by the STRUCTURE analysis using the linkage model and 7 housekeeping sequences are displayed on the top of the figure and marked with different colours. The ancestral parts of each strain are displayed in vertical lines. Subpopulations of the M. capricolum/M. leachii population are displayed below.doi:10.1371/journal.pone.0036150.g001
In a first phylogenetic analysis including all 123 Mycoplasma strains, the five strains of unassigned M. sp. clustered together as an outgroup to the ‘M. mycoides cluster’. Nevertheless the use of distant outgroups can lead to a distortion of the phylogeny. We plotted transition and transversion rates versus genetic distances (Figure S1). As expected, the observed number of transitions is higher than that of transversions among strains of the ‘M. mycoides cluster’ (ingroup). However, the observed number of transversions is higher than that of transitions for pairwise comparisons between the outgroup and the ingroup species, which implies that substitution saturation has occurred during the divergence between the outgroup and the ingroup. We therefore excluded M. sp. from the phylogenetic analysis. Instead we assumed equal evolutionary rates across all branches and used midpoint rooting for drawing the estimated trees.
The midpoint rooted maximum likelihood phylogeny estimated using the concatenated data set agrees with previously reported data  in that the ‘M. mycoides cluster’ is divided into two subclusters, one comprising Mmc and Mmm, the second M. capricolum and M. leachii (Figure S2). In addition, this analysis shows strong statistical support for the monophyly of the three ‘M. mycoides cluster’ populations defined above. Here however, M. leachii clusters separately from M. capricolum with strong statistical support. The subspecies of M. capricolum, Mccp is monophyletic, while Mcc is found to be paraphyletic. Phylogenies estimated for each locus independently varied in topology (Figure S3), and many were incongruent with the phylogenetic tree estimated from the concatenated data set. The incongruences in phylogenetic trees may be explained by hybridization between populations of the cluster, as identified in the structure analysis.
We computed three different summary statistics (θw, Tajima's D and Fu's Fs) to characterize the patterns of genetic diversity within the four populations from the structure analysis and treating the hybrid M. leachii as a fifth, separate subpopulation (Table 1). In addition, we estimated recombination rates per bp in three (sub)populations, Mmc, M. capricolum and M. leachi (Table 2). We did not estimate recombination in either Mmm or in the population of unassigned Mycoplasma strains because of their very low genetic diversity.
Table 1. Summary statistics for seven genes in five (sub)populations.doi:10.1371/journal.pone.0036150.t001
Table 2. Population recombination rate estimates (ρ) for three (sub)populations.doi:10.1371/journal.pone.0036150.t002
Mmc shows the highest genetic diversity (θw = 0.014). In addition, the test for recombination shows that significant levels of recombination occur in two genes within Mmc (Table 2). Levels of genetic diversity in the four other populations are low and we find no evidence for recombination. However, the sample sizes for several of the populations are very small, and it is possible that additional data may influence these results. In particular, as our analyses have revealed M. leachii as a likely hybrid, we should not exclude the possibility that recombination may be more important than this simple analysis suggests .
For each locus, we assessed how well a null model of constant population size and random mating fits each population by estimating Tajima's D and Fu's Fs statistics. Negative values reflect an excess of rare alleles, which can be either due to a demographic scenario such as population expansion or to positive selection. Three loci have a significantly negative Tajima's D for M. leachii, with negative values (although not significant) for the other four genes. Mmc also deviates from the null model in that Fu's Fs values are significantly negative for five of the seven loci, with negative Tajima's D values for all loci, although significantly negative for only one locus. Since it is unlikely that all the housekeeping genes used in this study are under positive selective pressure, we favour population expansion as an explanation of this departure from the null model.
The maximum clade credibility tree resulting from the combined BEAST analysis of the 110 strains for which a year of isolation was available (excluding the five unassigned M. sp. strains) is shown in Figure 2. All strains fall into monophyletic clusters with strong statistical support. The 110 strains included in this analysis shared a common ancestor ca. 10,000 years ago, while Mmc, M. capricolum and M. leachii all share common ancestors between 2,300 and 4,500 years ago (Table S3). Mccp, the agent of CCPP, shared a common ancestor between only 56 and 490 years ago.
Figure 2. Maximum credibility tree resulting from the BEAST analysis of the concatenated sequence alignment, as described in the main text.
Bayesian posterior probabilities are provided at the major nodes in the tree as a measure of support for clustering of the distinct strains. The scale is given in years before present.doi:10.1371/journal.pone.0036150.g002
The present study is based on 118 strains belonging to the ‘M. mycoides cluster’ and represents the largest comparative study of the ‘M. mycoides cluster’ to date. Strikingly, the estimated origin of the ‘M. mycoides cluster’ appears to coincide with the onset of domestication of small and large ruminants about 10,000 years ago , , , . Domestication was associated with both the establishment of large ruminant populations and the herding of mixed species. Both of these factors may have contributed to creating environmental conditions favouring the spread and diversification of the ‘M. mycoides cluster’ as the organisms adapted to different hosts.
It has been shown previously that genetic exchange between different Mycoplasma species sharing the same host is possible within the genus Mycoplasma , . However, our study provides evidence for genetic exchange between and within Mycoplasma populations having different primary hosts. For example, M. leachii appears to be a hybrid of Mmm and M. capricolum. In addition, the bovine strain B144P, which had been identified as Mycoplasma serogroup L , actually belongs to the M. leachii hybrid subpopulation, but contains ancestry from all ‘M. mycoides cluster’ populations defined in this study (Mmc, Mmm and M. capricolum). Since species belonging to the ‘M. mycoides cluster’ are obligate parasites and require the host for survival, any horizontal gene transfer or recombination must have happened within the host. Recent reports have shown that species of the ‘M. mycoides cluster’ can survive in non-primary hosts. For example, the bovine pathogens Mmm and M. leachii (Table S1) have been isolated from goats , and caprine pathogens M. capricolum or M. mycoides subsp. capri have been isolated from cattle (Table S1) . Other studies reported the isolation of different ‘M. mycoides cluster’ members from single, diseased individuals , , supporting the idea that the host itself might act as a hybridization oven for new ‘M. mycoides cluster’ variants. Mixed herding and pastoralist practices have been common in the past and remain widespread in Africa. These provide the opportunity for hosts to become infected with pathogens from other hosts. The resulting co-infection may facilitate the exchange of genetic material between pathogens.
The five strains of unassigned Mycoplasma sp. that were isolated independently from various Carprinae hosts on various continents over several years represent a distinct population (Table S1), related but not belonging to the ‘M. mycoides cluster’. We hypothesize that these isolates represent a novel species of Mycoplasma closely related to the ‘M. mycoides cluster’ that evolved in wild Caprinae . However, thorough analyses of both biochemical and genomic traits are required to confirm this hypothesis. Of particular interest would be the comparison of diseases patterns caused by the different strains in their respective hosts. Our MLST typing scheme is a useful tool to characterize not only Mycoplasma isolates belonging to the ‘M. mycoides cluster’ but also to type closely related mycoplasmas.
Materials and Methods
Strains and samples
We collected data from 123 Mycoplasma strains, including 50 Mmm, 33 Mmc, 11 M. leachii, nine Mcc, 14 Mccp, five unassigned Mycoplasma strains isolated from wild Caprinae in different continents, and one strain of Mycoplasma previously typed as serogroup L strain. Sequences from three fully annotated mycoplasma genomes from GenBank (GM12, Accession number CP002027; ATCC27343, Accession number NC_007633; and PG1, Accession number NC_005364) were part of our data collection. The full data set comprised strains from Europe, America, Australia, Asia, and Africa that had been collected during the period spanning 1931–2005. Additional information about the strains comprising the full data set, including country and year of isolation, are provided in Table S1.
Target amplification and sequencing
We selected 7 housekeeping genes, which were used previously for MLST analysis in other bacteria , , , : adenylate kinase (adk), guanylate kinase (gmk), DNA gyrase subunit B (gyrB), dihydrolipoamide S-acetyltransferase (pdhC), glucose-6-phosphate isomerase (pgi), recombination protein (recA), and DNA-directed RNA polymerase beta chain (rpoB). About 500 bp of each housekeeping gene were sequenced representing a total of 3816 bp per strain. The genes used are randomly distributed in the 1 MB genome of the Mmm type strain PG1 (Figure S4). PCR was carried out in duplicate (50 µl reaction volume) using GoTaq® Green master mix (Promega, USA) polymerase according to manufacturers' instructions, and primers listed in Table 2. Five to 20 ng of genomic DNA were used as template per reaction, annealing temperatures are provided in Table S2. PCR products were purified using QIAquick PCR purification kit (QIAGEN, Germany) and sequenced by Macrogen Inc. (Seoul, Korea), Agowa GmbH (Berlin, Germany), or StarSeq GmbH (Mainz, Germany). Sequence traces were assembled using the Staden program package (http://staden.sourceforge.net/) and trimmed (Table S2) by eye. Sequences are available in GenBank (Accession numbers JQ673623–JQ674483).
To infer recent demographic trends among members of the cluster, we calculated common summary statistics for the entire data set and for each member of the cluster separately using DNAsp v5 . We estimated nucleotide diversity (θw)  and summaries of allele frequencies (Tajima's D, Fu's Fs) , , with significance obtained by comparing the observed values to 1000 simulated data sets.
In order to quantify the extent of homologous recombination within populations, we estimated population recombination rates using an extension of the composite–likelihood approach described before  and implemented in the program LDhat . This method uses a finite-site model of substitution, which is more appropriate for bacterial evolution than the infinite sites model. Indeed, in bacteria, the rate of substitution is sufficiently high that some sites may have experienced multiple mutations in the history of the sample. We estimated the recombination rate ρ = 2Ner (where r is the per base rate of initiation of recombination and Ne the effective population size). We tested for significance of recombination using the permutation test available in LDhat.
We estimated population structure using the linkage model in STRUCTURE v2.3.2 , . To perform this test, we converted MLST sequence data from Extended FASTA Format into the Structure Format using xmfa2struct (available from http://www.xavierdidelot.xtreemhost.com/clonalframe.htm). This Bayesian approach uses multilocus genotypic data to define a set of populations with distinct allele frequencies, and assigns individuals/strains probabilistically to defined populations without prior knowledge of sampling location or sampled host. This program identifies admixed individuals/strains and gives an estimate of percent ancestry from ancestral population for each individual/strain. We performed three replications of the test, in which we initially discarded 10,000 Markov Chain Monte Carlo (MCMC) iterations as burn-in and kept the subsequent samples from 20,000 MCMC iterations for analysis. We tested values of K between 1 and 7, where K is the number of inferred populations. The results of the three independent runs were averaged for each K value to determine the most likely model, i.e. the one with the highest likelihood. Results were plotted using Distruct .
Phylogenetic analysis and estimates of divergence times
We first tested whether M. sp. was a suitable outgroup for our phylogenetic analyses by estimating substitution saturation. We used the program DAMBE  to plot pairwise transition and transversion distances versus total genetic distance. We inferred the phylogenetic history of each housekeeping gene separately using the Bayesian approach implemented in MrBayes v3 . We used the program jmodeltest1.0  to select the best fitting model of nucleotide substitution, which, for each data set, was the Generalized Time Reversible model with invariant sites and gamma-distributed rate heterogeneity (GTR+I+G) .
For each gene and the concatenated data set, we used MrBayes to estimate four independent MCMC chains (one cold and three hot), each running for ten million iterations with samples drawn every 1000 iterations. We removed the first 10% of each run to allow for burn-in, and assessed convergence using the program Tracer v1.4.1 (http://tree.bio.ed.ac.uk/software/tracer/). The majority consensus trees were drawn using Figtree v1.3.1 (http://tree.bio.ed.ac.uk/software/figtree/).
We performed two additional analyses of the phylogenetic history of the concatenated data set. First, we estimated a phylogeny for all the sequences concatenated and all the strains. Initially, we used MrBayes and assumed the GTR+G+I model of nucleotide substitution, with separate evolutionary models assigned to each locus. We had difficulties reaching convergence of the Markov Chains in our analyses using a Bayesian framework. We therefore estimated a maximum likelihood phylogeny for all populations using PhyML 3.0 . To assess statistical support for the resulting phylogeny, we performed 1000 bootstrap replicates assuming GTR+G+I model of nucleotide substitution. Secondly, we performed a molecular clock analysis of 110 strains, representing all populations and all loci, but excluding strains belonging to the population of strains of an unassigned Mycoplasma species, as well as the eight strains for which no year of isolation was available. We used the flexible Bayesian phylogenetic analysis package BEAST v1.6.1 , which allowed us to estimate the time of the divergence between the populations within the ‘M. mycoides cluster’. Here again, the GTR+G+I model proved to be an over parameterization of the data, and convergence of these particular parameters could not be achieved. For subsequent analyses, we used the HKY+G+I model, with different transition/transversion ratios, gamma parameters, and proportion of invariant sites estimated for each of the seven loci, but all loci informing the same tree. We assumed a strict molecular clock, with the evolutionary rate estimated using the collection date of each of the isolated sequences, and an internal, normally distributed calibration in which the age of the Mmm lineage is estimated to lie within a 95% confidence interval spanning 150–250 years ago , , . Independent evolutionary rates were estimated for each of the seven loci. To account for potential structure among the different populations, we used the flexible Bayesian Skyline coalescent model . Four BEAST analyses were run for 100 million iterations each, with trees and parameter values drawn from the posterior sample every 10,000 iterations. Chains were evaluated for convergence using Tracer. The first 10% of samples from each run were discarded as burn-in, and the remainder combined. The maximum clade credibility tree was estimated from the combined posterior sample of trees using TreeAnnotator v1.6.1 (http://beast.bio.ed.ac.uk/TreeAnnotator).
Plot of transitions (blue crosses) and transversions (green triangles) versus genetic distance (Generalized Time Reversible model (GTR)) for seven concatenated sequences.
Mid-point rooted phylogenetic tree displaying the phylogentic relationship of the ‘M. mycoides cluster’. The colour code used in Figure 1 was used to display the strain designation to different populations. The bootstrap values are displayed.
50% majority consensus tree for each of the seven partial gene sequences as estimated with MrBayes under the GTR+G+I substitution model.
Genomic location of MLST target genes based on the PG1 genome.
Strains used in this study.
Target genes, primer sequences and trimming region used for MLST of the ‘M. mycoides cluster’.
Bayesian MCMC estimates of time to the most recent common ancestor in years before present.
We thank Lucia Manso-Silvan for providing DNA of the strain VNCT. This is ILRI publication number 201105.
Conceived and designed the experiments: JJ JF. Performed the experiments: JJ CM CS. Analyzed the data: AF BS JJ. Contributed reagents/materials/analysis tools: JF MH EB-R EV. Wrote the paper: AF BS JJ.
- 1. Cottew GS, Breard A, DaMassa AJ, Erno H, Leach RH, et al. (1987) Taxonomy of the Mycoplasma mycoides cluster. Isr J Med Sci 23: 632–635.
- 2. Hutyra F, Marek J, Manninger R (1938) Contagious Bovine Pleuropneumonia. In: Greig JR, Mohler JR, Eichhorn A, editors. London: Balliere, Tindal and Cox.
- 3. Fisher J (2006) The origins, spread and disappearance of contagious bovine pleuro-pneumonia in New Zealand. Aust Vet J 84: 439–444.
- 4. Thomas P (1873) Rapport médical sur le Bou Frida. In: Jourdan A, editor. Publication du gouvernement général civil de l'Algérie. Algiers.
- 5. Manso-Silvan L, Vilei EM, Sachse K, Djordjevic SP, Thiaucourt F, et al. (2009) Mycoplasma leachii sp. nov. as a new species designation for Mycoplasma sp. bovine group 7 of Leach, and reclassification of Mycoplasma mycoides subsp. mycoides LC as a serovar of Mycoplasma mycoides subsp. capri. Int J Syst Evol Microbiol 59: 1353–1358.
- 6. Kim KS, Ko KS, Chang MW, Hahn TW, Hong SK, et al. (2003) Use of rpoB sequences for phylogenetic study of Mycoplasma species. FEMS Microbiol Lett 226: 299–305.
- 7. Vilei EM, Korczak BM, Frey J (2006) Mycoplasma mycoides subsp. capri and Mycoplasma mycoides subsp. mycoides LC can be grouped into a single subspecies. Vet Res 37: 779–790.
- 8. Manso-Silvan L, Perrier X, Thiaucourt F (2007) Phylogeny of the Mycoplasma mycoides cluster based on analysis of five conserved protein-coding sequences and possible implications for the taxonomy of the group. Int J Syst Evol Microbiol 57: 2247–2258.
- 9. Nwankpa ND, Manso-Silvan L, Lorenzon S, Yaya A, Lombin LH, et al. (2010) Variable Number Tandem Repeat (VNTR) analysis reveals genetic diversity within Mycoplasma mycoides mycoides Small Colony isolates from Nigeria. Vet Microbiol 146(3–4): 354–355.
- 10. Yaya A, Manso-Silvan L, Blanchard A, Thiaucourt F (2008) Genotyping of Mycoplasma mycoides subsp. mycoides SC by multilocus sequence analysis allows molecular epidemiology of contagious bovine pleuropneumonia. Vet Res 39: 14.
- 11. Manso-Silvan L, Dupuy V, Chu Y, Thiaucourt F (2011) Multi-locus sequence analysis of Mycoplasma capricolum subsp. capripneumoniae for the molecular epidemiology of contagious caprine pleuropneumonia. Veterinary research 42: 86.
- 12. Thiaucourt F, Manso-Silvan L, Salah W, Barbe V, Vacherie B, et al. (2011) Mycoplasma mycoides, from “mycoides Small Colony” to “capri”. A microevolutionary perspective. BMC genomics 12: 114.
- 13. Stipkovits L, El-Ebeedy A (1977) Biochemical and serological studies of avian mycoplasmas. Zentralblatt fur Veterinarmedizin Reihe B Journal of veterinary medicine Series B 24: 218–230.
- 14. Falush D, Wirth T, Linz B, Pritchard JK, Stephens M, et al. (2003) Traces of human migrations in Helicobacter pylori populations. Science 299: 1582–1585.
- 15. Bruford MW, Bradley DG, Luikart G (2003) DNA markers reveal the complexity of livestock domestication. Nature reviews Genetics 4: 900–910.
- 16. Naderi S, Rezaei HR, Pompanon F, Blum MG, Negrini R, et al. (2008) The goat domestication process inferred from large-scale mitochondrial DNA analysis of wild and domestic individuals. Proceedings of the National Academy of Sciences of the United States of America 105: 17659–17664.
- 17. Gotherstrom A, Anderung C, Hellborg L, Elburg R, Smith C, et al. (2005) Cattle domestication in the Near East was followed by hybridization with aurochs bulls in Europe. Proceedings Biological sciences/The Royal Society 272: 2345–2350.
- 18. Zeder MA, Hesse B (2000) The initial domestication of goats (Capra hircus) in the Zagros mountains 10,000 years ago. Science 287: 2254–2257.
- 19. Sirand-Pugnet P, Lartigue C, Marenda M, Jacob D, Barre A, et al. (2007) Being pathogenic, plastic, and sexual while living with a nearly minimal bacterial genome. PLoS Genet 3: e75.
- 20. Thomas A, Linden A, Mainil J, Bischof DF, Frey J, et al. (2005) Mycoplasma bovis shares insertion sequences with Mycoplasma agalactiae and Mycoplasma mycoides subsp. mycoides SC: Evolutionary and developmental aspects. FEMS Microbiol Lett 245: 249–255.
- 21. Kusiluka LJ, Ojeniyi B, Friis NF, Kokotovic B, Ahrens P (2001) Molecular analysis of field strains of Mycoplasma capricolum subspecies capripneumoniae and Mycoplasma mycoides subspecies mycoides, small colony type isolated from goats in Tanzania. Vet Microbiol 82: 27–37.
- 22. Thiaucourt F, Lorenzon S, David A, Breard A (2000) Phylogeny of the Mycoplasma mycoides cluster as shown by sequencing of a putative membrane protein gene. Vet Microbiol 72: 251–268.
- 23. DaMassa AJ, Brooks DL, Adler HE (1983) Caprine mycoplasmosis: widespread infection in goats with Mycoplasma mycoides subsp mycoides (large-colony type). American journal of veterinary research 44: 322–325.
- 24. Nicolas MM, Stalis IH, Clippinger TL, Busch M, Nordhausen R, et al. (2005) Systemic disease in Vaal rhebok (Pelea capreolus) caused by mycoplasmas in the mycoides cluster. Journal of clinical microbiology 43: 1330–1340.
- 25. Stackebrandt E, Frederiksen W, Garrity GM, Grimont PA, Kampfer P, et al. (2002) Report of the ad hoc committee for the re-evaluation of the species definition in bacteriology. International journal of systematic and evolutionary microbiology 52: 1043–1047.
- 26. Wirth T, Falush D, Lan R, Colles F, Mensa P, et al. (2006) Sex and virulence in Escherichia coli: an evolutionary perspective. Molecular microbiology 60: 1136–1151.
- 27. Mayor D, Jores J, Korczak BM, Kuhnert P (2008) Multilocus sequence typing (MLST) of Mycoplasma hyopneumoniae: a diverse pathogen with limited clonality. Vet Microbiol 127: 63–72.
- 28. Maiden MC, Bygraves JA, Feil E, Morelli G, Russell JE, et al. (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.
- 29. Librado P, Rozas J (2009) DnaSP v5: a software for comprehensive analysis of DNA polymorphism data. Bioinformatics 25: 1451–1452.
- 30. Watterson GA (1975) On the number of segregating sites in genetical models without recombination. Theoretical population biology 7: 256–276.
- 31. Tajima F (1989) Statistical method for testing the neutral mutation hypothesis by DNA polymorphism. Genetics 123: 585–595.
- 32. Fu YX (1997) Statistical tests of neutrality of mutations against population growth, hitchhiking and background selection. Genetics 147: 915–925.
- 33. Hudson RR (2001) Two-locus sampling distributions and their application. Genetics 159: 1805–1817.
- 34. McVean G, Awadalla P, Fearnhead P (2002) A coalescent-based method for detecting and estimating recombination from gene sequences. Genetics 160: 1231–1241.
- 35. Pritchard JK, Stephens M, Donnelly P (2000) Inference of population structure using multilocus genotype data. Genetics 155: 945–959.
- 36. Falush D, Stephens M, Pritchard JK (2003) Inference of population structure using multilocus genotype data: linked loci and correlated allele frequencies. Genetics 164: 1567–1587.
- 37. Rosenberg NA (2004) DISTRUCT: a program for the graphical display of population structure. Molecular Ecology Notes 4: 137–138.
- 38. Xia X, Xie Z (2001) DAMBE: software package for data analysis in molecular biology and evolution. The Journal of heredity 92: 371–373.
- 39. Ronquist F, Huelsenbeck JP (2003) MrBayes 3: Bayesian phylogenetic inference under mixed models. Bioinformatics 19: 1572–1574.
- 40. Posada D (2008) jModelTest: phylogenetic model averaging. Mol Biol Evol 25: 1253–1256.
- 41. Salemi M, Vandamme A-M, Lemey P (2009) The phylogenetic handbook: a practical approach to phylogenetic analysis and hypothesis testing. Cambridge: Cambridge University Press.
- 42. Guindon S, Dufayard JF, Lefort V, Anisimova M, Hordijk W, et al. (2010) New algorithms and methods to estimate maximum-likelihood phylogenies: assessing the performance of PhyML 3.0. Systematic biology 59: 307–321.
- 43. Drummond AJ, Rambaut A (2007) BEAST: Bayesian evolutionary analysis by sampling trees. BMC evolutionary biology 7: 214.
- 44. Fisher J (2003) To kill or not to kill: the eradication of contagious bovine pleuro-pneumonia in western Europe. Med Hist 47: 314–331.
- 45. Drummond AJ, Rambaut A, Shapiro B, Pybus OG (2005) Bayesian coalescent inference of past population dynamics from molecular sequences. Molecular biology and evolution 22: 1185–1192.