Characterization of Mycobacterium caprae Isolates from Europe by Mycobacterial Interspersed Repetitive Unit Genotyping
http://www.100md.com
微生物临床杂志 2005年第10期
Department of Hygiene, Microbiology, and Social Medicine
Institute of Legal Medicine, Innsbruck Medical University, Innsbruck, Austria
Bavarian Health and Food Safety Authority, Oberschleissheim, Germany
Istituto Zooprofilattico Sperimentale della Lombardia e dell'Emilia Romagna, Brescia, Italy
Forschungszentrum Borstel, National Reference Center for Mycobacteria, Borstel, Germany
Agence Franaise pour la Securite Sanitaire des Aliments, Maisons Alfort, France
Departamento de Sanidad Animal, Universidad Complutense, Madrid, Spain
Central Veterinary Institute, Budapest, Hungary
Croatian Veterinary Institute, Zagreb, Croatia
Veterinary Faculty, University of Ljubljana, Slovenia
German National Veterinary Reference Laboratory for Tuberculosis, Friedrich-Loeffler-Institut, Jena, Germany
Veterinary Research Institute, Brno, Czech Republic
ABSTRACT
Mycobacterium caprae, a recently defined member of the Mycobacterium tuberculosis complex, causes tuberculosis among animals and, to a limited extent, in humans in several European countries. To characterize M. caprae in comparison with other Mycobacterium tuberculosis complex members and to evaluate genotyping methods for this species, we analyzed 232 M. caprae isolates by mycobacterial interspersed repetitive unit (MIRU) genotyping and by spoligotyping. The isolates originated from 128 distinct epidemiological settings in 10 countries, spanning a period of 25 years. We found 78 different MIRU patterns (53 unique types and 25 clusters with group sizes from 2 to 9) but only 17 spoligotypes, giving Hunter-Gaston discriminatory indices of 0.941 (MIRU typing) and 0.665 (spoligotyping). For a subset of 103 M. caprae isolates derived from outbreaks or endemic foci, MIRU genotyping and IS6110 restriction fragment length polymorphism were compared and shown to provide similar results. MIRU loci 4, 26, and 31 were most discriminant in M. caprae, followed by loci 10 and 16, a combination which is different than those reported to discriminate M. bovis best. M. caprae MIRU patterns together with published data were used for phylogenetic inference analysis employing the neighbor-joining method. M. caprae isolates were grouped together, closely related to the branches of classical M. bovis, M. pinnipedii, M. microti, and ancestral M. tuberculosis, but apart from modern M. tuberculosis. The analysis did not reflect geographic patterns indicative of origin or spread of M. caprae. Altogether, our data confirm M. caprae as a distinct phylogenetic lineage within the Mycobacterium tuberculosis complex.
INTRODUCTION
Several members of the Mycobacterium tuberculosis complex (MTC) that encompasses the causative agents of tuberculosis (TB) can be distinguished. Recently, M. pinnipedii (11) and M. caprae (1) have been added to M. tuberculosis, M. africanum, M. bovis, M. bovis BCG, M. microti, and M. canettii. M. caprae, with former names M. tuberculosis subspecies caprae (2) and M. bovis subspecies caprae (30), has originally been described as preferring goats to cattle as host within the same epidemiological settings in Spain (4, 19). M. caprae has been found to affect predominantly cattle among several other host species in Austria (34), France (20), Germany (15, 16), Hungary (16), Italy (M. B. Boniotti, L. Alborali, E. Tisato, and M. L. Pacciarini, Abstr. 25th Annu. Meet. Eur. Soc. Mycobacteriol., abstr. 61, 2004), Slovenia (16), and the Czech Republic (33). In addition, wild living species such as red deer (34) or wild boar (16, 27) infected with M. caprae may constitute a reservoir for resurgent TB in domestic animals. Human infections with M. caprae appear to be rare on a worldwide or a Europe-wide scale nowadays, due to the eradication campaigns and preventive measures taken against transmission of bovine TB in the last century. They primarily manifest in older individuals (31). However, in central European regions where M. caprae is the major cause of TB in cattle, it is also the predominant agent of "bovine TB" in humans (25, 34). Interestingly, genomic deletion analyses using large panels of MTC isolates (6, 29) suggested that both new MTC members have phylogenetically preceded M. bovis.
Genotyping techniques developed for MTC members have extended our understanding of the natural history of TB and have become an essential tool in TB epidemiology (47). This applies likewise to analyses of TB transmission among livestock (32) or between livestock and reservoir species such as badgers, possums, or deer (7). IS6110 restriction fragment length polymorphism (RFLP) has been the standard technique for differentiation of human isolates with at least five IS6110 copies (46) and can also discriminate M. bovis isolates, although such isolates most often possess only a single IS6110 element (9, 40, 48). However, PCR-based methods have challenged this costly method, as they promise a faster throughput at equal quality, e.g., methods aimed at tandemly arranged repetitive sequences in the genomes of MTC members (13). These variable-number tandem repeats (VNTRs) are found at multiple loci, and some loci show substantial variation in the number of repeat units.
Different VNTR typing methods have been established. Initially, Frothingham and Meeker (18) used six exact tandem repeat (ETR) loci. Mazars et al. (28) described the analysis of 12 loci of mycobacterial interspersed repetitive units (MIRUs) by a single-PCR method which was later developed into a multiplex PCR with automated sequencer analysis (45). MIRU genotyping has proven highly discriminant for worldwide MTC isolate collections (43, 45) and analyses of human TB outbreaks (21, 26, 28), for population-derived samples of isolates (5, 44) and for population-based studies (12). MIRU patterns show sufficient stability and appear to evolve at a slower pace than IS6110 patterns, as shown for serial isolates from patients (28, 38), and in some instances they discriminated IS6110-defined clusters (isolates with >6 bands) (12). MIRU typing has been studied for the discrimination of M. bovis isolates but has been found to be less discriminant than strategies aimed at other VNTRs (36). Skuce et al. (41) and Roring et al. (35) have described such VNTRs (termed QUBs) as particularly useful for M. bovis outbreaks and as more discriminant than ETR or MIRU typing with a standardized M. bovis panel (36).
M. caprae isolates have not been studied with VNTR analyses so far, and other techniques have their drawbacks with this MTC member. Spoligotyping, although instrumental in identifying M. caprae isolates (3), is little help in discriminating them: in central European countries, more than 50% of the isolates show the same spoligotype, lacking spacers 1, 3 to 16, 28, and 39 to 43 (16). IS6110 RFLP is highly discriminant, as M. caprae isolates generally have two to eight IS6110 copies (16), but is costly and time-consuming.
This study is the first to assess the potential of MIRU genotyping to differentiate M. caprae isolates. We have analyzed 232 M. caprae isolates from across Europe, representative of nearly all countries that have ever reported M. caprae isolates (Fig. 1), by MIRU typing and spoligotyping. We have analyzed the potential of the method to (i) discriminate between outbreaks among livestock, (ii) differentiate serial isolates obtained from natural TB foci over years, (iii) compare human isolates with each other and with animal isolates, and (iv) show the position of M. caprae among the MTC members on an evolutionary scale. The data obtained suggest that MIRU is a valuable tool in all of these aspects.
MATERIALS AND METHODS
Bacterial isolates and DNA extraction. MTC isolates from TB patients (48 isolates) or from wild or domestic animals (camel, 2 isolates; deer, 7; cattle, 156; goat, 11; wild boar, 8) were determined to be M. caprae by their specific spoligotype and by allele-specific PCR for the typical M. caprae nucleotide sequence polymorphisms in oxyR and pncA (17). The isolates originated from 10 European countries (Austria, 59 isolates; Croatia, 11; Czech Republic, 4; France, 16; Germany, 83; Hungary, 17; Italy, 27; Slovenia, 1; Spain, 11; Ukraine, 3), including several M. caprae isolates reported in the literature (16, 20, 34). A detailed list of all 232 MIRU-typed M. caprae isolates is given in the supplemental material, and the approximate geographic origins of the isolates are shown in Fig. 1. Genomic DNA for PCR was obtained by standard isolation techniques (49) or by boiling a loop full of bacteria in 200 μl of 1x Tris-EDTA buffer for 15 min and using the supernatant.
MIRU genotyping. MIRU typing was performed for 205 isolates in Innsbruck, Austria, and for the remaining 28 samples in Borstel, Germany. MIRU typing in Innsbruck was performed by 12 single PCRs as described previously (28) with the primer sequences and final MgCl2 concentrations as described previously (45) and the following modification: instead of a HotStarTaq DNA polymerase kit (QIAGEN, Hilden Germany), we used Dynazyme II DNA polymerase together with the appropriate 10x PCR buffer (Finnzymes, Espoo, Finland) and 10% dimethyl sulfoxide (molecular biology grade; Sigma, St. Louis, MO) in a 25-μl PCR. This protocol had been shown to yield identical MIRU results on a panel of RFLP-typed isolates and reference strains (data not shown). To minimize mislabeling and confounding errors, single strips of PCR tubes prefilled with 12 PCR mixes to amplify each of the 12 MIRU loci were used for each isolate. The PCR products were separated by electrophoresis on a 2% agarose (Seakem LE; Cambrex Biosciences, Rockland, Maine) gel in 1x Tris-acetate-EDTA containing 1 μg of ethidium bromide/ml in a Sub-Cell model 192 apparatus (Bio-Rad, Hercules, Calif.) and a 25- by 25-cm gel with three rows of 51 slots (slot width, 0.75 mm). Multichannel devices fitting both PCR strips and slot rows were used for loading the gel. A size marker (DNA molecular weight marker XIV; Roche, Mannheim, Germany) was loaded to every fifth lane. The fragments for each isolate were separated for 3 h at 150 V and photographed together. The MIRU copy number corresponding to the respective band size was calculated according to information provided by Philip Supply (www.ibl.fr/mirus/mirus.html) and entered into a spreadsheet file. At Borstel, MIRU codes were determined by T.K. and S.N. by multiplex-PCR MIRU typing according to a previously described method (45).
Spoligotyping and IS6110 RFLP analysis. Spoligotyping was carried out as described previously (23) by using membranes (Isogen Bioscience BV, Maarsen, The Netherlands). The octal codes for spoligotypes were determined as proposed previously (14). IS6110 RFLP was performed on selected series of isolates as described previously (49); for others, the results were retrieved from the literature (15, 16). The patterns were analyzed with Gelcompar version 4.2 (Applied Math, Sint-Martens-Latem, Belgium) by the unweighted-pair group method using arithmetic averages and the Dice coefficient for similarity with 1% band position tolerance.
MIRU-VNTR phylogenetic inference and evolutionary analysis of spoligotypes. For phylogenetic inference using the MIRU-VNTR data, we applied the neighbor-joining method (37) by the computer program PAUP 4.0b10 (phylogenetic analysis using parsimony and other methods; D. Swofford, Sinauer Associates, Sunderland, Mass.). MIRU-VNTR data were treated as ordered character states. To bring the 76 different MIRU patterns for M. caprae into a phylogenetic context with those from other MTC members, we combined the data from this study with that from 90 individuals belonging to M. tuberculosis (n = 70), M. africanum (n = 2), M. bovis (n = 12), M. pinnipedii (n = 3; originally described as M. bovis [27]), M. microti (n = 2), and M. canettii (n = 1) (45), retrieved from http://www.ibl.fr/mirus/mirus.html. Identical haplotypes were identified and represented in the analyses by one sequence only. This led to a final data set comprising 167 different types, including M. canettii as an out-group taxon.
Statistical analyses. The Hunter-Gaston discriminatory index (HGDI) was calculated as described previously (22). Allelic diversity (h) of MIRU alleles was assessed according to Selander et al. (39).
RESULTS
MIRU-VNTR and spoligotyping analyses. A complete MIRU code was obtained for 229 of the 232 M. caprae samples. For three isolates, the copy number of certain loci could not be established as no amplification product was obtained (for two isolates at locus 16 and for the third isolate at loci 4, 16, and 23). This failure was probably caused by a limiting amount or quality of target DNA. As the three isolates were nevertheless unequivocally different from all other isolates, they were also included in the analysis. Identical isolates from one definite epidemiological setting (same farm within the same year) or from one person are counted as one representative isolate (RI), making a total of 128 RIs for this study. The cluster analyses by spoligotyping and MIRU genotyping, respectively, are summarized in Table 1 (fingerprint patterns and epidemiological data are given in the supplemental material).
Spoligotyping revealed 17 different patterns, and 71 of 128 RIs (55%) belonged to the "classical" spoligotype of M. caprae, S1. A further 54 RIs were grouped into spoligotypes S2 to S14 (group sizes, 2 to 14 RIs), and only three isolates had unique spoligotypes. In contrast, MIRU genotyping showed 25 clusters (group sizes, 2 to 9 RIs) and 53 unique types, making 78 different patterns altogether (Table 1). In particular, the prevailing S1 spoligotype isolates were split up into 15 MIRU clusters and 27 unique patterns. The HGDI for MIRU typing was 0.941 for the 128 RIs (0.896 for all 232 isolates), whereas spoligotyping had HGDI values of 0.673 (n = 128) and 0.480 (n = 232), respectively.
Five isolates within MIRU clusters showed spoligotypes that differed from the one typically associated with the respective MIRU cluster (Table 2). In these cases, spoligotyping was more discriminant than MIRU typing, as the spoligotypes showed additional deletions of a single spacer or of a block of contingent direct repeat (DR) spacers. The M5/S4 type combination of isolate 21 appears as an intermediate between the very closely related and actually observed combinations M5/S1 and M9/S4.
Of a total of 25 MIRU clusters, 15 comprised only animal isolates, 3 had only human isolates, and 7 had both. Thirteen of the 15 animal clusters were found within the same country, often in the same or bordering regions. The two exceptions are cluster M1, which is found in parts of the northern and southern Alps (separated by 300 kilometers), and cluster M5, found at very distant locations in France, Germany, the Czech Republic, and Ukraine. In contrast, only 2 of 14 shared spoligotypes comprised only animal isolates, but 9 of 14 were mixed of human and animal isolates. The S1 type was found distributed in all countries except Spain, with a higher proportion among isolates from Eastern European countries. In contrast, S4 and S7, the other types shared between several countries, were both restricted to Germany (human isolates only), France, Italy, and Spain.
Performance of MIRU genotyping versus that of IS6110 RFLP. Since MIRU typing is being considered as the high-throughput alternative for IS6110 RFLP typing for M. tuberculosis, we compared the discrimination of both methods for a subset of M. caprae isolates with available RFLP results. Three larger outbreaks or series of outbreaks among cattle (i.e., affecting one or two farms in the same or neighboring communities), with a total of 80 parallel outbreak isolates, were analyzed. The respective outbreak strains are represented by RIs 2, 10, and 36 (see the supplemental material). Two outbreaks showed microheterogeneity (Table 3, outbreaks A and B). In the first outbreak (RI 10), 2 of 11 parallel isolates differed by one IS6110 band each, whereas all were MIRU identical (Table 3, outbreak A). In the second outbreak (RIs 35, 36, and 40), all 16 parallel isolates were RFLP identical, but one had a single MIRU copy number change (Table 3, outbreak B). Finally, in a border-crossing series of four outbreaks in the northern Alps with 53 parallel isolates (RI 2 of cluster M1), all isolates had an identical MIRU code, spoligotype, and RFLP pattern (Table 3, outbreak C). Interestingly, this MIRU type was also retrieved from distant sites in the southern Alps, but RFLP data are not available yet. In summary, among the 80 parallel isolates, two RFLP patterns differed from the respective outbreak RFLP pattern, and one MIRU type differed from the respective outbreak MIRU type.
Outbreaks among cattle or TB cases among wild boars that occurred separately over years, but in the same region within a range of 50 to 200 kilometers, showed greater heterogeneity among the isolates. With 23 isolates tested, the extent of copy number changes in MIRU typing agreed well with the number of band changes in RFLP typing in each setting (Table 3, outbreaks D to F).
M. caprae isolates from humans. Forty-seven of the 232 M. caprae isolates were derived from 46 human TB cases, the large majority being reactivation cases in aged individuals (data not shown; see reference 25). Altogether, 38 different MIRU patterns and 14 different spoligotypes were observed (see the supplemental material). Twenty-eight isolates (61%) were unique by MIRU typing but only two had unique spoligotypes. MIRU typing thus discriminated 13 isolates of spoligotype S1. Nine human isolates clustered with isolates from cattle or free-living animals. Six of these seven clusters are unlikely to represent a direct epidemiological connection, as animals and humans were from different countries. In cluster M23, the two human isolates differed by a unique spoligotype from the cattle isolate showing spoligotype S1. In one instance, the isolates from a farmer and his cow shared the same type (corresponding to cluster M25).
Nine further human isolates clustered with other human isolates only. Human-to-human transmission could not be ruled out or might even be suspected from the incomplete patient data. In the only case of childhood infection (isolate 56), a source (either animal or human) was not evident from this study. Cluster M11 comprised three individuals from the same region, aged 66, 70, and 80 years. However, their spoligotypes diverged by single DVR deletions. The other clusters did not show spoligotype diversity.
Allelic diversity in MIRU-VNTR loci. The distribution of MIRU allele numbers in our sample of 128 RIs is shown in Table 4. Two copies were found invariably at locus 39. On the other hand, a few isolates with more than two copies at locus 20 were detected, which has not been reported for other MTC isolates so far. Ten out of 122 isolates have copy numbers other than two at locus 24, with two having one copy. In 3 of all 232 isolates (1.3% [see the supplemental material]), two bands corresponding to different copy numbers were reproducibly found for one locus, indicating the emergence of new subclones. For 128 M. caprae isolates, allelic diversity (h) was highest for MIRU loci 26, 31, and 4, followed by loci 10 and 16. The other loci had low or no discriminatory potential.
MIRU-VNTR phylogenetic inference. As M. caprae isolates have not yet been included in any larger MIRU genotyping analysis of MTC isolates, we investigated how this method would group M. caprae within the MTC. Seventy-seven different MIRU patterns were observed for M. caprae in this study. Seventy-six (i.e., all except RI 126, which had three MIRU loci undetermined) were analyzed together with MIRU data for 90 well-studied MTC isolates published by Supply et al. (45). The resulting neighbor-joining tree for the 166 specimens, including M. canettii as the out-group individual, is shown in Fig. 2, and the positions of M. caprae isolates are indicated. All except one M. caprae isolate are separated from modern M. tuberculosis and found in one large group which has distinct branches for ancestral M. tuberculosis, classical M. bovis, M. pinnipedii, and M. microti. The only two M. africanum isolates are found between M. pinnipedii and M. caprae and between modern M. tuberculosis and M. caprae, respectively. Three M. bovis isolates (Fig. 2), i.e., one from Oryx gazella (isolate SAU 24) and two from humans (isolates NL 6 and NL 85), respectively, are positioned within M. caprae. The isolates of the dominant S1 spoligotype are evenly distributed among M. caprae, with the exception of the branch between RIs 26 and 19 that encompasses all S7 and most S4 type isolates.
DISCUSSION
Molecular epidemiology and epizootiology of TB have contributed significantly to a better understanding of the dynamics of transmission in human populations or between wild and/or domestic animals. In this study, we investigated the discriminatory potential of MIRU-VNTR genotyping for a comprehensive panel of isolates of M. caprae, a pathogen that contributes significantly to animal TB in continental Europe and to the evanescent number of "bovine TB" cases among humans in these countries. MIRU typing, as performed by electrophoresis of single-locus PCRs as readout (28), was found fully reproducible. Ninety-three percent of the isolates were fully typeable at the first run. The method also proved highly discriminant for this panel of M. caprae isolates, giving HGDI values of 0.941 and 0.896 for the representative set (n = 128) and the set including all outbreak isolates (n = 232), respectively. For MIRU genotyping, Sola et al. have reported HGDI values of 0.988 for a worldwide collection of 119 MTC isolates (43), and Sun et al. obtained values of 0.975 for 291 consecutive human M. tuberculosis isolates from Singapore (44).
Roring et al. calculated allelic diversity (h) for a defined set of 42 different M. bovis isolates and obtained a value of 0.74 for MIRU genotyping (36), in contrast to the h value of 0.933 for this M. caprae set (n = 128). These two panels cannot be compared directly, though, as the former was obtained selectively from Irish cattle and badgers, whereas the latter comprised human and animal isolates from across Europe. However, the most discriminating MIRU alleles were also found to differ between the two panels: MIRU 24 was moderately discriminating among the M. bovis set (36), as among M. tuberculosis (43, 45) isolates, but only poorly discriminating among M. caprae isolates. A similar finding was reported for a collection of 123 M. bovis isolates obtained from Belgian cattle over a period of 8 years (C. Allix, K. Walravens, V. Vandenpoorte, P. Supply, J. Godfroid, and M. Fauville-Dufaux, Abstr. 25th Annu. Meet. Eur. Soc. Mycobacteriol., abstr. 135, 2004). On the other hand, MIRU loci 10, 16, and 31 proved moderately discriminant among M. caprae samples but poorly resolving among the Irish M. bovis panel. Compared to large MTC panels analyzed by MIRU typing (and not including M. caprae) (43, 45), M. caprae isolates were found to be significantly less diverse at locus 40 and notably monomorphic at locus 39. Thus, an optimally reduced set of MIRUs for resolving M. caprae would consist of MIRUs 4, 10, 16, 26, and 31, which is different from other reduced MIRU sets postulated for M. bovis (36). Nevertheless, it will be interesting to test M. caprae within MIRU clusters with additional VNTRs, such as QUBs (41), reported to have a superior resolution in a well-defined M. bovis panel (36).
Direct comparison to IS6110 RFLP is regarded as a challenge for MTC subtyping methods (3, 48), provided that the isolates do not harbor more than three copies of IS6110. M. caprae isolates generally have more than one IS6110 copy and average around five (range, two to eight) copies (4, 16, 19, 34). Testing a subset of 103 M. caprae isolates from outbreaks or endemic foci, we found the same discrimination between isolates from endemic sources by MIRU typing and by IS6110 RFLP and very little variation among outbreak isolates (Table 3). These data suggest that MIRU typing is an alternative that could substitute for RFLP typing. This experience supports the conclusions from an early evaluation of a two-step approach adopted in the United States for genotyping M. tuberculosis from humans, i.e., a combination of automated spoligotyping and MIRU typing that reduces the need for RFLP typing (12).
The geographical distribution of MIRU types among the 128 epidemiological settings suggests that MIRU clusters are also specific enough on a large scale. Of 15 clusters that comprise animal isolates only, 13 were found in a single country and thus provided clues for local and regional transmission or persistence of TB. In contrast, the locations found with mixed clusters (i.e., comprising animal and human isolates) do not suggest epidemiological connections, although the observed transmission between cow and farmer (RIs 43 and 85 [see the al material]) is an exception. This may be explained by the fact that human study isolates originated predominantly from elderly patients and thus have been acquired several decades ago, whereas most animal isolates represent very recent infections. This hypothesis is supported by the high proportion of MIRU unique isolates from humans. In addition, a reduction of M. caprae strain diversity is very likely to have occurred after the eradication measures against "bovine tuberculosis" in the past.
A glimpse of ongoing strain differentiation could be caught in 3 of the 232 isolates, which showed two alleles for a single given MIRU locus. One would expect to "catch" the evolution of a subclone in rare instances, and this is actually an advantage of all VNTR-based methods. On the other hand, a potential problem associated with VNTR typing is homoplasia, i.e., that copy number changes in previously different isolates could lead to identical profiles, e.g., MIRU types. The instances in this study where different spoligotypes were found with identical MIRU patterns are best explained by additional deletion events in the DR gene locus (Table 2). Altogether, this data set does not infer that homoplasia is a major obstacle for the interpretation of MIRU genotyping results.
The underlying concept of MTC phylogeny has been demonstrated to be reflected by MIRU pattern analysis (45), except for M. caprae. These data were reanalyzed after addition of a comprehensive M. caprae sample (Fig. 2). M. caprae isolates (all but one) were distributed in one large group together with the distinct branches for ancestral M. tuberculosis, classical M. bovis, M. pinnipedii, and M. microti, respectively. All are set off from modern M. tuberculosis. Three "M. bovis" isolates (SAU 24, NL 6, and NL 85) were placed among M. caprae isolates at different sites. Those three have been shown to differ from any of the above MTC members of the M. bovis lineage by deletion analysis and single nucleotide polymorphisms (6). In accordance with these findings, 118 out of 128 M. caprae isolates had two copies at locus 24, a locus with a very slow evolution rate (45), as is the case for most isolates in the M. bovis lineage and for "ancestral" M. tuberculosis.
M. caprae has recently been defined as a separate member of the MTC (1). This development accommodates the findings from recent genomic deletion analyses (6, 29) that have likewise set apart "classical" M. bovis from M. pinnipedii or M. microti, i.e., pathogens that cause TB primarily in selected animal species. However, M. caprae causes TB among the same animal species as "classical" M. bovis does. In very practical terms, this implies that the lawful notification process should encompass M. caprae infections as TB and will have to be adapted wherever M. caprae is prevalent.
One interesting question remains. MIRU typing of this representative panel does yield a geographical pattern that would clearly suggest routes of spread of M. caprae in the past. On the one hand, the dominant S1 type is increasingly prevalent from the west part to the east part of Europe. As MTC strains are thought lose DR spacers and not to regain them, the S1 type can also be hypothesized to be ancestral to S4 and S7. On the other hand, a separate analysis of the human versus animal isolates in the same country shows a significantly higher proportion of the S1 type among cattle isolates (86%) than among human isolates (56%). This could reflect recent clonal expansions after an already achieved reduction of M. caprae prevalence by eradication measures. An analysis of 11,500 (classical) M. bovis isolates from the United Kingdom and Ireland by spoligotyping, followed by ETR-VNTR subtyping in part, has suggested clonal expansions leading to the current population of M. bovis in Great Britain (42). Historical animal trading across and from Europe has influenced the different repertoires of M. bovis spoligotypes, e.g., the ones in Great Britain and its trading-partner countries in contrast to those in continental European countries (8, 10, 16, 20, 41). M. caprae has been found almost exclusively in continental Europe and as far east as the eastern Ukraine. One human isolate has been reported from an inhabitant of the European part of Istanbul, Turkey (D. Satana, M. Uzun, Z. Erturan, Y. Yegenoglu, Abstr. 26th Annu. Meet. Eur. Soc. Mycobacteriol., abstr. 11, 2005), but information about the prevalence in Middle Eastern, Asian, or African countries is missing. The close relation of European M. caprae to an isolate from a gazelle found in Africa and Arabia is therefore very interesting and stimulating to further define its origin.
ACKNOWLEDGMENTS
We are grateful to Carolin Lechleitner, Dieter Koefler, and Daniela Loda for excellent technical assistance and to Walter Glawischnig and Maria Beatrice Boniotti for helpful discussions.
This work was supported by the European Commission (QLK2-CT-2000-00630), the Czech Ministry of Agriculture (MZE 0002716201), the Italian Ministry of Health (PRC2001008), the Spanish Ministry of Agriculture, Fisheries, and Food. A.S. is the recipient of a scholarship from the German Academic Exchange Service.
Supplemental material for this article may be found at http://jcm.asm.org/.
These authors contributed equally to this work.
REFERENCES
Aranaz, A., D. Cousins, A. Mateos, and L. Dominguez. 2003. Elevation of Mycobacterium tuberculosis subsp. caprae Aranaz et al. 1999 to species rank as Mycobacterium caprae comb. nov., sp. nov. Int. J. Syst. Evol. Microbiol. 53:1785-1789.
Aranaz, A., E. Liebana, E. Gomez Mampaso, J. Galan, D. Cousins, A. Ortega, J. Blazquez, F. Baquero, A. Mateos, G. Suarez, and L. Dominguez. 1999. Mycobacterium tuberculosis subsp. caprae subsp. nov.: a taxonomic study of a new member of the Mycobacterium tuberculosis complex isolated from goats in Spain. Int. J. Syst. Bacteriol. 49:1263-1273.
Aranaz, A., E. Liebana, A. Mateos, L. Dominguez, and D. Cousins. 1998. Restriction fragment length polymorphism and spacer oligonucleotide typing: a comparative analysis of fingerprinting strategies for Mycobacterium bovis. Vet. Microbiol. 61:311-324.
Aranaz, A., E. Liebana, A. Mateos, L. Dominguez, D. Vidal, M. Domingo, O. Gonzalez, E. F. Rodriguez Ferri, A. E. Bunschoten, J. D. van Embden, and D. Cousins. 1996. Spacer oligonucleotide typing of Mycobacterium bovis strains from cattle and other animals: a tool for studying epidemiology of tuberculosis. J. Clin. Microbiol. 34:2734-2740.
Banu, S., S. V. Gordon, S. Palmer, M. R. Islam, S. Ahmed, K. M. Alam, S. T. Cole, and R. Brosch. 2004. Genotypic analysis of Mycobacterium tuberculosis in Bangladesh and prevalence of the Beijing strain. J. Clin. Microbiol. 42:674-682.
Brosch, R., S. V. Gordon, M. Marmiesse, P. Brodin, C. Buchrieser, K. Eiglmeier, T. Garnier, C. Gutierrez, G. Hewinson, K. Kremer, L. M. Parsons, A. S. Pym, S. Samper, D. van Soolingen, and S. T. Cole. 2002. A new evolutionary scenario for the Mycobacterium tuberculosis complex. Proc. Natl. Acad. Sci. USA 99:3684-3689.
Collins, C. H. 2000. The bovine tubercle bacillus. Br. J. Biomed. Sci. 57:234-240.
Costello, E., D. O'Grady, O. Flynn, R. O'Brien, M. Rogers, F. Quigley, J. Egan, and J. Griffin. 1999. Study of restriction fragment length polymorphism analysis and spoligotyping for epidemiological investigation of Mycobacterium bovis infection. J. Clin. Microbiol. 37:3217-3222.
Cousins, D., R. Skuce, R. Kazwala, J. D. A. van Embden, et al. 1998. Towards a standardized approach to DNA fingerprinting of Mycobacterium bovis. Int. J. Tuberc. Lung Dis. 2:471-478.
Cousins, D., S. Williams, E. Liebana, A. Aranaz, A. Bunschoten, J. van Embden, and T. Ellis. 1998. Evaluation of four DNA typing techniques in epidemiological investigations of bovine tuberculosis. J. Clin. Microbiol. 36:168-178.
Cousins, D. V., R. Bastida, A. Cataldi, V. Quse, S. Redrobe, S. Dow, P. Duignan, A. Murray, C. Dupont, N. Ahmed, D. M. Collins, W. R. Butler, D. Dawson, D. Rodriguez, J. Loureiro, M. I. Romano, A. Alito, M. Zumarraga, and A. Bernardelli. 2003. Tuberculosis in seals caused by a novel member of the Mycobacterium tuberculosis complex: Mycobacterium pinnipedii sp. nov. Int. J. Syst. Evol. Microbiol. 53:1305-1314.
Cowan, L. S., L. Diem, T. Monson, P. Wand, D. Temporado, T. V. Oemig, and J. T. Crawford. 2005. Evaluation of a two-step approach for large-scale, prospective genotyping of Mycobacterium tuberculosis isolates in the United States. J. Clin. Microbiol. 43:688-695.
Crawford, J. T., C. R. Braden, B. A. Schable, and I. M. Onorato. 2002. National Tuberculosis Genotyping and Surveillance Network: design and methods. Emerg. Infect. Dis. 8:1192-1196.
Dale, J. W., D. Brittain, A. A. Cataldi, D. Cousins, J. T. Crawford, J. Driscoll, H. Heersma, T. Lillebaek, T. Quitugua, N. Rastogi, R. A. Skuce, C. Sola, D. van Soolingen, and V. Vincent. 2001. Spacer oligonucleotide typing of bacteria of the Mycobacterium tuberculosis complex: recommendations for standardised nomenclature. Int. J. Tuberc. Lung Dis. 5:216-219.
Erler, W., D. Kahlau, G. Martin, L. Naumann, D. Schimmel, and A. Weber. 2003. The epizootiology of tuberculosis of cattle in the Federal Republic of Germany. Berl. Muench. Tieraerztl. Wochenschr. 116:288-292. (In German.)
Erler, W., G. Martin, K. Sachse, L. Naumann, D. Kahlau, J. Beer, M. Bartos, G. Nagy, Z. Cvetnic, M. Zolnir-Dovc, and I. Pavlik. 2004. Molecular fingerprinting of Mycobacterium bovis subsp. caprae isolates from central Europe. J. Clin. Microbiol. 42:2234-2238.
Espinosa de los Monteros, L. E., J. C. Galán, M. Gutierrez, S. Samper, J. F. García Marín, C. Martín, L. Domínguez, L. de Rafael, F. Baquero, E. Gomez-Mampaso, and J. Blázquez. 1998. Allele-specific PCR method based on pncA and oxyR sequences for distinguishing Mycobacterium bovis from Mycobacterium tuberculosis: intraspecific M. bovis pncA sequence polymorphism. J. Clin. Microbiol. 36:239-242.
Frothingham, R., and O. C. Meeker. 1998. Genetic diversity in the Mycobacterium tuberculosis complex based on variable numbers of tandem DNA repeats. Microbiology 144:1189-1196.
Gutierrez, M., S. Samper, J. A. Gavigan, J. F. Garcia Marin, and C. Martin. 1995. Differentiation by molecular typing of Mycobacterium bovis strains causing tuberculosis in cattle and goats. J. Clin. Microbiol. 33:2953-2956.
Haddad, N., A. Ostyn, C. Karoui, M. Masselot, M. F. Thorel, S. L. Hughes, J. Inwald, R. G. Hewinson, and B. Durand. 2001. Spoligotype diversity of Mycobacterium bovis strains isolated in France from 1979 to 2000. J. Clin. Microbiol. 39:3623-3632.
Hawkey, P. M., E. G. Smith, J. T. Evans, P. Monk, G. Bryan, H. H. Mohamed, M. Bardhan, and R. N. Pugh. 2003. Mycobacterial interspersed repetitive unit typing of Mycobacterium tuberculosis compared to IS6110-based restriction fragment length polymorphism analysis for investigation of apparently clustered cases of tuberculosis. J. Clin. Microbiol. 41:3514-3520.
Hunter, P. R., and M. A. Gaston. 1988. Numerical index of the discriminatory ability of typing systems: an application of Simpson's index of diversity. J. Clin. Microbiol. 26:2465-2466.
Kamerbeek, J., L. Schouls, A. Kolk, M. van Agterveld, D. van Soolingen, S. Kuijper, A. Bunschoten, H. Molhuizen, R. Shaw, M. Goyal, and J. van Embden. 1997. Simultaneous detection and strain differentiation of Mycobacterium tuberculosis for diagnosis and epidemiology. J. Clin. Microbiol. 35:907-914.
Kremer, K., D. van Soolingen, R. Frothingham, W. H. Haas, P. W. M. Hermans, C. Martín, P. Palittapongarnpim, B. B. Plikaytis, L. W. Riley, M. A. Yakrus, J. M. Musser, and J. D. A. van Embden. 1999. Comparison of methods based on different molecular epidemiological markers for typing of Mycobacterium tuberculosis complex stains: interlaboratory study of discriminatory power and reproducibility. J. Clin. Microbiol. 37:2607-2618.
Kubica, T., S. Rüsch-Gerdes, and S. Niemann. 2003. Mycobacterium bovis subsp. caprae caused one-third of human M. bovis-associated tuberculosis cases reported in Germany between 1999 and 2001. J. Clin. Microbiol. 41:3070-3077.
Kwara, A., R. Schiro, L. S. Cowan, N. E. Hyslop, M. F. Wiser, H. S. Roahen, P. Kissinger, L. Diem, and J. T. Crawford. 2003. Evaluation of the epidemiologic utility of secondary typing methods for differentiation of Mycobacterium tuberculosis isolates. J. Clin. Microbiol. 41:2683-2685.
Machackova, M., L. Matlova, J. Lamka, J. Smolik, L. Melicharek, Hanzlikova.M., J. Docekal, Z. Cvetnic, G. Nagy, M. Lipiec, M. Ocepek, and I. Pavlik. 2004. Wild boar (Sus scrofa) as a possible vector of mycobacterial infections: review of literature and critical analysis of data from Central Europe between 1983 to 2001. Vet. Med. Czech. 48:51-65.
Mazars, E., S. Lesjean, A.-L. Banuls, M. Gilbert, V. Vincent, B. Gicquel, M. Tibayrenc, C. Locht, and P. Supply. 2001. High-resolution minisatellite-based typing as a portable approach to global analysis of Mycobacterium tuberculosis molecular epidemiology. Proc. Natl. Acad. Sci. USA 98:1901-1906.
Mostowy, S., D. Cousins, J. Brinkman, A. Aranaz, and M. A. Behr. 2002. Genomic deletions suggest a phylogeny for the Mycobacterium tuberculosis complex. J. Infect. Dis. 186:74-80.
Niemann, S., E. Richter, and S. Rusch-Gerdes. 2002. Biochemical and genetic evidence for the transfer of Mycobacterium tuberculosis subsp. caprae Aranaz et al. 1999 to the species Mycobacterium bovis Karlson and Lessel 1970 (approved lists 1980) as Mycobacterium bovis subsp. caprae comb. nov. Int. J. Syst. Evol. Microbiol. 52:433-436.
Pavlik, I., W. Y. Ayele, M. Havelkova, M. Svejnochova, V. Katalinic-Jankovic, and M. Zolnir-Dovc. 2004. Mycobacterium bovis infection in human population in four Central European countries during 1990-1999. Vet. Med. Czech. 48:90-98.
Pavlik, I., F. Bures, P. Janovsky, P. Pecinka, M. Bartos, L. Dvorksa, L. Matlova, K. Kremer, and D. van Soolingen. 2002. The last outbreak of bovine tuberculosis in cattle in the Czech Republic in 1995 was caused by Mycobacterium bovis supecies caprae. Vet. Med. Czech. 47:251-263.
Pavlik, I., L. Dvorksa, M. Bartos, I. Parmova, I. Melicharek, A. Jesenska, M. Havelkova, M. Slosarek, I. Putova, G. Martin, W. Erler, K. Kremer, and D. van Soolingen. 2002. Molecular epidemiology of bovine tuberculosis in the Czech Republic and Slovakia in the period 1965-2001 studied by spoligotyping. Vet. Med. Czech. 47:181-194.
Prodinger, W. M., A. Eigentler, F. Allerberger, M. Schnbauer, and W. Glawischnig. 2002. Infection of red deer, cattle, and humans with Mycobacterium bovis subsp. caprae in western Austria. J. Clin. Microbiol. 40:2270-2272.
Roring, S., A. Scott, D. Brittain, I. Walker, G. Hewinson, S. Neill, and R. A. Skuce. 2002. Development of variable-number tandem repeat typing of Mycobacterium bovis: comparison of results with those obtained by using existing exact tandem repeats and spoligotyping. J. Clin. Microbiol. 40:2126-2133.
Roring, S., A. N. Scott, H. R. Glyn, S. D. Neill, and R. A. Skuce. 2004. Evaluation of variable number tandem repeat (VNTR) loci in molecular typing of Mycobacterium bovis isolates from Ireland. Vet. Microbiol. 101:65-73.
Saitou, N., and M. Nei. 1987. The neighbor-joining method: a new method for reconstructing phylogenetic trees. Mol. Biol. Evol. 4:406-425.
Savine, E., R. M. Warren, G. D. Van Der Spuy, N. Beyers, P. D. van Helden, C. Locht, and P. Supply. 2002. Stability of variable-number tandem repeats of mycobacterial interspersed repetitive units from 12 loci in serial isolates of Mycobacterium tuberculosis. J. Clin. Microbiol. 40:4561-4566.
Selander, R. K., D. A. Caugant, H. Ochman, J. M. Musser, M. N. Gilmour, and T. S. Whittam. 1986. Methods of multilocus enzyme electrophoresis for bacterial population genetics and systematics. Appl. Environ. Microbiol. 51:873-884.
Skuce, R. A., D. Brittain, M. S. Hughes, L. A. Beck, and S. D. Neill. 1994. Genomic fingerprinting of Mycobacterium bovis from cattle by restriction fragment length polymorphism analysis. J. Clin. Microbiol. 32:2387-2392.
Skuce, R. A., T. P. McCorry, J. F. McCarroll, S. M. Roring, A. N. Scott, D. Brittain, S. L. Hughes, R. G. Hewinson, and S. D. Neill. 2002. Discrimination of Mycobacterium tuberculosis complex bacteria using novel VNTR-PCR targets. Microbiology 148:519-528.
Smith, N. H., J. Dale, J. Inwald, S. Palmer, S. V. Gordon, R. G. Hewinson, and J. M. Smith. 2003. The population structure of Mycobacterium bovis in Great Britain: clonal expansion. Proc. Natl. Acad. Sci. USA 100:15271-15275.
Sola, C., I. Filliol, E. Legrand, S. Lesjean, C. Locht, P. Supply, and N. Rastogi. 2003. Genotyping of the Mycobacterium tuberculosis complex using MIRUs: association with VNTR and spoligotyping for molecular epidemiology and evolutionary genetics. Infect. Genet. Evol. 3:125-133.
Sun, Y. J., R. Bellamy, A. S. Lee, S. T. Ng, S. Ravindran, S. Y. Wong, C. Locht, P. Supply, and N. I. Paton. 2004. Use of mycobacterial interspersed repetitive unit-variable-number tandem repeat typing to examine genetic diversity of Mycobacterium tuberculosis in Singapore. J. Clin. Microbiol. 42:1986-1993.
Supply, P., S. Lesjean, E. Savine, K. Kremer, D. van Soolingen, and C. Locht. 2001. Automated high-throughput genotyping for study of global epidemiology of Mycobacterium tuberculosis based on mycobacterial interspersed repetitive units. J. Clin. Microbiol. 39:3563-3571.
van Embden, J. D., M. D. Cave, J. T. Crawford, J. W. Dale, K. D. Eisenach, B. Gicquel, P. Hermans, C. Martin, R. McAdam, T. M. Shinnick, and P. M. Small. 1993. Strain identification of Mycobacterium tuberculosis by DNA fingerprinting: recommendations for a standardized methodology. J. Clin. Microbiol. 31:406-409.
van Soolingen, D. 2001. Molecular epidemiology of tuberculosis and other mycobacterial infections: main methodologies and achievements. J. Intern. Med. 249:1-26.
van Soolingen, D., P. E. de Haas, J. Haagsma, T. Eger, P. W. Hermans, V. Ritacco, A. Alito, and J. D. van Embden. 1994. Use of various genetic markers in differentiation of Mycobacterium bovis strains from animals and humans and for studying epidemiology of bovine tuberculosis. J. Clin. Microbiol. 32:2425-2433.
van Soolingen, D., P. E. de Haas, P. W. Hermans, and J. D. van Embden. 1994. DNA fingerprinting of Mycobacterium tuberculosis. Methods Enzymol. 235:196-205.(Wolfgang M. Prodinger, An)
Institute of Legal Medicine, Innsbruck Medical University, Innsbruck, Austria
Bavarian Health and Food Safety Authority, Oberschleissheim, Germany
Istituto Zooprofilattico Sperimentale della Lombardia e dell'Emilia Romagna, Brescia, Italy
Forschungszentrum Borstel, National Reference Center for Mycobacteria, Borstel, Germany
Agence Franaise pour la Securite Sanitaire des Aliments, Maisons Alfort, France
Departamento de Sanidad Animal, Universidad Complutense, Madrid, Spain
Central Veterinary Institute, Budapest, Hungary
Croatian Veterinary Institute, Zagreb, Croatia
Veterinary Faculty, University of Ljubljana, Slovenia
German National Veterinary Reference Laboratory for Tuberculosis, Friedrich-Loeffler-Institut, Jena, Germany
Veterinary Research Institute, Brno, Czech Republic
ABSTRACT
Mycobacterium caprae, a recently defined member of the Mycobacterium tuberculosis complex, causes tuberculosis among animals and, to a limited extent, in humans in several European countries. To characterize M. caprae in comparison with other Mycobacterium tuberculosis complex members and to evaluate genotyping methods for this species, we analyzed 232 M. caprae isolates by mycobacterial interspersed repetitive unit (MIRU) genotyping and by spoligotyping. The isolates originated from 128 distinct epidemiological settings in 10 countries, spanning a period of 25 years. We found 78 different MIRU patterns (53 unique types and 25 clusters with group sizes from 2 to 9) but only 17 spoligotypes, giving Hunter-Gaston discriminatory indices of 0.941 (MIRU typing) and 0.665 (spoligotyping). For a subset of 103 M. caprae isolates derived from outbreaks or endemic foci, MIRU genotyping and IS6110 restriction fragment length polymorphism were compared and shown to provide similar results. MIRU loci 4, 26, and 31 were most discriminant in M. caprae, followed by loci 10 and 16, a combination which is different than those reported to discriminate M. bovis best. M. caprae MIRU patterns together with published data were used for phylogenetic inference analysis employing the neighbor-joining method. M. caprae isolates were grouped together, closely related to the branches of classical M. bovis, M. pinnipedii, M. microti, and ancestral M. tuberculosis, but apart from modern M. tuberculosis. The analysis did not reflect geographic patterns indicative of origin or spread of M. caprae. Altogether, our data confirm M. caprae as a distinct phylogenetic lineage within the Mycobacterium tuberculosis complex.
INTRODUCTION
Several members of the Mycobacterium tuberculosis complex (MTC) that encompasses the causative agents of tuberculosis (TB) can be distinguished. Recently, M. pinnipedii (11) and M. caprae (1) have been added to M. tuberculosis, M. africanum, M. bovis, M. bovis BCG, M. microti, and M. canettii. M. caprae, with former names M. tuberculosis subspecies caprae (2) and M. bovis subspecies caprae (30), has originally been described as preferring goats to cattle as host within the same epidemiological settings in Spain (4, 19). M. caprae has been found to affect predominantly cattle among several other host species in Austria (34), France (20), Germany (15, 16), Hungary (16), Italy (M. B. Boniotti, L. Alborali, E. Tisato, and M. L. Pacciarini, Abstr. 25th Annu. Meet. Eur. Soc. Mycobacteriol., abstr. 61, 2004), Slovenia (16), and the Czech Republic (33). In addition, wild living species such as red deer (34) or wild boar (16, 27) infected with M. caprae may constitute a reservoir for resurgent TB in domestic animals. Human infections with M. caprae appear to be rare on a worldwide or a Europe-wide scale nowadays, due to the eradication campaigns and preventive measures taken against transmission of bovine TB in the last century. They primarily manifest in older individuals (31). However, in central European regions where M. caprae is the major cause of TB in cattle, it is also the predominant agent of "bovine TB" in humans (25, 34). Interestingly, genomic deletion analyses using large panels of MTC isolates (6, 29) suggested that both new MTC members have phylogenetically preceded M. bovis.
Genotyping techniques developed for MTC members have extended our understanding of the natural history of TB and have become an essential tool in TB epidemiology (47). This applies likewise to analyses of TB transmission among livestock (32) or between livestock and reservoir species such as badgers, possums, or deer (7). IS6110 restriction fragment length polymorphism (RFLP) has been the standard technique for differentiation of human isolates with at least five IS6110 copies (46) and can also discriminate M. bovis isolates, although such isolates most often possess only a single IS6110 element (9, 40, 48). However, PCR-based methods have challenged this costly method, as they promise a faster throughput at equal quality, e.g., methods aimed at tandemly arranged repetitive sequences in the genomes of MTC members (13). These variable-number tandem repeats (VNTRs) are found at multiple loci, and some loci show substantial variation in the number of repeat units.
Different VNTR typing methods have been established. Initially, Frothingham and Meeker (18) used six exact tandem repeat (ETR) loci. Mazars et al. (28) described the analysis of 12 loci of mycobacterial interspersed repetitive units (MIRUs) by a single-PCR method which was later developed into a multiplex PCR with automated sequencer analysis (45). MIRU genotyping has proven highly discriminant for worldwide MTC isolate collections (43, 45) and analyses of human TB outbreaks (21, 26, 28), for population-derived samples of isolates (5, 44) and for population-based studies (12). MIRU patterns show sufficient stability and appear to evolve at a slower pace than IS6110 patterns, as shown for serial isolates from patients (28, 38), and in some instances they discriminated IS6110-defined clusters (isolates with >6 bands) (12). MIRU typing has been studied for the discrimination of M. bovis isolates but has been found to be less discriminant than strategies aimed at other VNTRs (36). Skuce et al. (41) and Roring et al. (35) have described such VNTRs (termed QUBs) as particularly useful for M. bovis outbreaks and as more discriminant than ETR or MIRU typing with a standardized M. bovis panel (36).
M. caprae isolates have not been studied with VNTR analyses so far, and other techniques have their drawbacks with this MTC member. Spoligotyping, although instrumental in identifying M. caprae isolates (3), is little help in discriminating them: in central European countries, more than 50% of the isolates show the same spoligotype, lacking spacers 1, 3 to 16, 28, and 39 to 43 (16). IS6110 RFLP is highly discriminant, as M. caprae isolates generally have two to eight IS6110 copies (16), but is costly and time-consuming.
This study is the first to assess the potential of MIRU genotyping to differentiate M. caprae isolates. We have analyzed 232 M. caprae isolates from across Europe, representative of nearly all countries that have ever reported M. caprae isolates (Fig. 1), by MIRU typing and spoligotyping. We have analyzed the potential of the method to (i) discriminate between outbreaks among livestock, (ii) differentiate serial isolates obtained from natural TB foci over years, (iii) compare human isolates with each other and with animal isolates, and (iv) show the position of M. caprae among the MTC members on an evolutionary scale. The data obtained suggest that MIRU is a valuable tool in all of these aspects.
MATERIALS AND METHODS
Bacterial isolates and DNA extraction. MTC isolates from TB patients (48 isolates) or from wild or domestic animals (camel, 2 isolates; deer, 7; cattle, 156; goat, 11; wild boar, 8) were determined to be M. caprae by their specific spoligotype and by allele-specific PCR for the typical M. caprae nucleotide sequence polymorphisms in oxyR and pncA (17). The isolates originated from 10 European countries (Austria, 59 isolates; Croatia, 11; Czech Republic, 4; France, 16; Germany, 83; Hungary, 17; Italy, 27; Slovenia, 1; Spain, 11; Ukraine, 3), including several M. caprae isolates reported in the literature (16, 20, 34). A detailed list of all 232 MIRU-typed M. caprae isolates is given in the supplemental material, and the approximate geographic origins of the isolates are shown in Fig. 1. Genomic DNA for PCR was obtained by standard isolation techniques (49) or by boiling a loop full of bacteria in 200 μl of 1x Tris-EDTA buffer for 15 min and using the supernatant.
MIRU genotyping. MIRU typing was performed for 205 isolates in Innsbruck, Austria, and for the remaining 28 samples in Borstel, Germany. MIRU typing in Innsbruck was performed by 12 single PCRs as described previously (28) with the primer sequences and final MgCl2 concentrations as described previously (45) and the following modification: instead of a HotStarTaq DNA polymerase kit (QIAGEN, Hilden Germany), we used Dynazyme II DNA polymerase together with the appropriate 10x PCR buffer (Finnzymes, Espoo, Finland) and 10% dimethyl sulfoxide (molecular biology grade; Sigma, St. Louis, MO) in a 25-μl PCR. This protocol had been shown to yield identical MIRU results on a panel of RFLP-typed isolates and reference strains (data not shown). To minimize mislabeling and confounding errors, single strips of PCR tubes prefilled with 12 PCR mixes to amplify each of the 12 MIRU loci were used for each isolate. The PCR products were separated by electrophoresis on a 2% agarose (Seakem LE; Cambrex Biosciences, Rockland, Maine) gel in 1x Tris-acetate-EDTA containing 1 μg of ethidium bromide/ml in a Sub-Cell model 192 apparatus (Bio-Rad, Hercules, Calif.) and a 25- by 25-cm gel with three rows of 51 slots (slot width, 0.75 mm). Multichannel devices fitting both PCR strips and slot rows were used for loading the gel. A size marker (DNA molecular weight marker XIV; Roche, Mannheim, Germany) was loaded to every fifth lane. The fragments for each isolate were separated for 3 h at 150 V and photographed together. The MIRU copy number corresponding to the respective band size was calculated according to information provided by Philip Supply (www.ibl.fr/mirus/mirus.html) and entered into a spreadsheet file. At Borstel, MIRU codes were determined by T.K. and S.N. by multiplex-PCR MIRU typing according to a previously described method (45).
Spoligotyping and IS6110 RFLP analysis. Spoligotyping was carried out as described previously (23) by using membranes (Isogen Bioscience BV, Maarsen, The Netherlands). The octal codes for spoligotypes were determined as proposed previously (14). IS6110 RFLP was performed on selected series of isolates as described previously (49); for others, the results were retrieved from the literature (15, 16). The patterns were analyzed with Gelcompar version 4.2 (Applied Math, Sint-Martens-Latem, Belgium) by the unweighted-pair group method using arithmetic averages and the Dice coefficient for similarity with 1% band position tolerance.
MIRU-VNTR phylogenetic inference and evolutionary analysis of spoligotypes. For phylogenetic inference using the MIRU-VNTR data, we applied the neighbor-joining method (37) by the computer program PAUP 4.0b10 (phylogenetic analysis using parsimony and other methods; D. Swofford, Sinauer Associates, Sunderland, Mass.). MIRU-VNTR data were treated as ordered character states. To bring the 76 different MIRU patterns for M. caprae into a phylogenetic context with those from other MTC members, we combined the data from this study with that from 90 individuals belonging to M. tuberculosis (n = 70), M. africanum (n = 2), M. bovis (n = 12), M. pinnipedii (n = 3; originally described as M. bovis [27]), M. microti (n = 2), and M. canettii (n = 1) (45), retrieved from http://www.ibl.fr/mirus/mirus.html. Identical haplotypes were identified and represented in the analyses by one sequence only. This led to a final data set comprising 167 different types, including M. canettii as an out-group taxon.
Statistical analyses. The Hunter-Gaston discriminatory index (HGDI) was calculated as described previously (22). Allelic diversity (h) of MIRU alleles was assessed according to Selander et al. (39).
RESULTS
MIRU-VNTR and spoligotyping analyses. A complete MIRU code was obtained for 229 of the 232 M. caprae samples. For three isolates, the copy number of certain loci could not be established as no amplification product was obtained (for two isolates at locus 16 and for the third isolate at loci 4, 16, and 23). This failure was probably caused by a limiting amount or quality of target DNA. As the three isolates were nevertheless unequivocally different from all other isolates, they were also included in the analysis. Identical isolates from one definite epidemiological setting (same farm within the same year) or from one person are counted as one representative isolate (RI), making a total of 128 RIs for this study. The cluster analyses by spoligotyping and MIRU genotyping, respectively, are summarized in Table 1 (fingerprint patterns and epidemiological data are given in the supplemental material).
Spoligotyping revealed 17 different patterns, and 71 of 128 RIs (55%) belonged to the "classical" spoligotype of M. caprae, S1. A further 54 RIs were grouped into spoligotypes S2 to S14 (group sizes, 2 to 14 RIs), and only three isolates had unique spoligotypes. In contrast, MIRU genotyping showed 25 clusters (group sizes, 2 to 9 RIs) and 53 unique types, making 78 different patterns altogether (Table 1). In particular, the prevailing S1 spoligotype isolates were split up into 15 MIRU clusters and 27 unique patterns. The HGDI for MIRU typing was 0.941 for the 128 RIs (0.896 for all 232 isolates), whereas spoligotyping had HGDI values of 0.673 (n = 128) and 0.480 (n = 232), respectively.
Five isolates within MIRU clusters showed spoligotypes that differed from the one typically associated with the respective MIRU cluster (Table 2). In these cases, spoligotyping was more discriminant than MIRU typing, as the spoligotypes showed additional deletions of a single spacer or of a block of contingent direct repeat (DR) spacers. The M5/S4 type combination of isolate 21 appears as an intermediate between the very closely related and actually observed combinations M5/S1 and M9/S4.
Of a total of 25 MIRU clusters, 15 comprised only animal isolates, 3 had only human isolates, and 7 had both. Thirteen of the 15 animal clusters were found within the same country, often in the same or bordering regions. The two exceptions are cluster M1, which is found in parts of the northern and southern Alps (separated by 300 kilometers), and cluster M5, found at very distant locations in France, Germany, the Czech Republic, and Ukraine. In contrast, only 2 of 14 shared spoligotypes comprised only animal isolates, but 9 of 14 were mixed of human and animal isolates. The S1 type was found distributed in all countries except Spain, with a higher proportion among isolates from Eastern European countries. In contrast, S4 and S7, the other types shared between several countries, were both restricted to Germany (human isolates only), France, Italy, and Spain.
Performance of MIRU genotyping versus that of IS6110 RFLP. Since MIRU typing is being considered as the high-throughput alternative for IS6110 RFLP typing for M. tuberculosis, we compared the discrimination of both methods for a subset of M. caprae isolates with available RFLP results. Three larger outbreaks or series of outbreaks among cattle (i.e., affecting one or two farms in the same or neighboring communities), with a total of 80 parallel outbreak isolates, were analyzed. The respective outbreak strains are represented by RIs 2, 10, and 36 (see the supplemental material). Two outbreaks showed microheterogeneity (Table 3, outbreaks A and B). In the first outbreak (RI 10), 2 of 11 parallel isolates differed by one IS6110 band each, whereas all were MIRU identical (Table 3, outbreak A). In the second outbreak (RIs 35, 36, and 40), all 16 parallel isolates were RFLP identical, but one had a single MIRU copy number change (Table 3, outbreak B). Finally, in a border-crossing series of four outbreaks in the northern Alps with 53 parallel isolates (RI 2 of cluster M1), all isolates had an identical MIRU code, spoligotype, and RFLP pattern (Table 3, outbreak C). Interestingly, this MIRU type was also retrieved from distant sites in the southern Alps, but RFLP data are not available yet. In summary, among the 80 parallel isolates, two RFLP patterns differed from the respective outbreak RFLP pattern, and one MIRU type differed from the respective outbreak MIRU type.
Outbreaks among cattle or TB cases among wild boars that occurred separately over years, but in the same region within a range of 50 to 200 kilometers, showed greater heterogeneity among the isolates. With 23 isolates tested, the extent of copy number changes in MIRU typing agreed well with the number of band changes in RFLP typing in each setting (Table 3, outbreaks D to F).
M. caprae isolates from humans. Forty-seven of the 232 M. caprae isolates were derived from 46 human TB cases, the large majority being reactivation cases in aged individuals (data not shown; see reference 25). Altogether, 38 different MIRU patterns and 14 different spoligotypes were observed (see the supplemental material). Twenty-eight isolates (61%) were unique by MIRU typing but only two had unique spoligotypes. MIRU typing thus discriminated 13 isolates of spoligotype S1. Nine human isolates clustered with isolates from cattle or free-living animals. Six of these seven clusters are unlikely to represent a direct epidemiological connection, as animals and humans were from different countries. In cluster M23, the two human isolates differed by a unique spoligotype from the cattle isolate showing spoligotype S1. In one instance, the isolates from a farmer and his cow shared the same type (corresponding to cluster M25).
Nine further human isolates clustered with other human isolates only. Human-to-human transmission could not be ruled out or might even be suspected from the incomplete patient data. In the only case of childhood infection (isolate 56), a source (either animal or human) was not evident from this study. Cluster M11 comprised three individuals from the same region, aged 66, 70, and 80 years. However, their spoligotypes diverged by single DVR deletions. The other clusters did not show spoligotype diversity.
Allelic diversity in MIRU-VNTR loci. The distribution of MIRU allele numbers in our sample of 128 RIs is shown in Table 4. Two copies were found invariably at locus 39. On the other hand, a few isolates with more than two copies at locus 20 were detected, which has not been reported for other MTC isolates so far. Ten out of 122 isolates have copy numbers other than two at locus 24, with two having one copy. In 3 of all 232 isolates (1.3% [see the supplemental material]), two bands corresponding to different copy numbers were reproducibly found for one locus, indicating the emergence of new subclones. For 128 M. caprae isolates, allelic diversity (h) was highest for MIRU loci 26, 31, and 4, followed by loci 10 and 16. The other loci had low or no discriminatory potential.
MIRU-VNTR phylogenetic inference. As M. caprae isolates have not yet been included in any larger MIRU genotyping analysis of MTC isolates, we investigated how this method would group M. caprae within the MTC. Seventy-seven different MIRU patterns were observed for M. caprae in this study. Seventy-six (i.e., all except RI 126, which had three MIRU loci undetermined) were analyzed together with MIRU data for 90 well-studied MTC isolates published by Supply et al. (45). The resulting neighbor-joining tree for the 166 specimens, including M. canettii as the out-group individual, is shown in Fig. 2, and the positions of M. caprae isolates are indicated. All except one M. caprae isolate are separated from modern M. tuberculosis and found in one large group which has distinct branches for ancestral M. tuberculosis, classical M. bovis, M. pinnipedii, and M. microti. The only two M. africanum isolates are found between M. pinnipedii and M. caprae and between modern M. tuberculosis and M. caprae, respectively. Three M. bovis isolates (Fig. 2), i.e., one from Oryx gazella (isolate SAU 24) and two from humans (isolates NL 6 and NL 85), respectively, are positioned within M. caprae. The isolates of the dominant S1 spoligotype are evenly distributed among M. caprae, with the exception of the branch between RIs 26 and 19 that encompasses all S7 and most S4 type isolates.
DISCUSSION
Molecular epidemiology and epizootiology of TB have contributed significantly to a better understanding of the dynamics of transmission in human populations or between wild and/or domestic animals. In this study, we investigated the discriminatory potential of MIRU-VNTR genotyping for a comprehensive panel of isolates of M. caprae, a pathogen that contributes significantly to animal TB in continental Europe and to the evanescent number of "bovine TB" cases among humans in these countries. MIRU typing, as performed by electrophoresis of single-locus PCRs as readout (28), was found fully reproducible. Ninety-three percent of the isolates were fully typeable at the first run. The method also proved highly discriminant for this panel of M. caprae isolates, giving HGDI values of 0.941 and 0.896 for the representative set (n = 128) and the set including all outbreak isolates (n = 232), respectively. For MIRU genotyping, Sola et al. have reported HGDI values of 0.988 for a worldwide collection of 119 MTC isolates (43), and Sun et al. obtained values of 0.975 for 291 consecutive human M. tuberculosis isolates from Singapore (44).
Roring et al. calculated allelic diversity (h) for a defined set of 42 different M. bovis isolates and obtained a value of 0.74 for MIRU genotyping (36), in contrast to the h value of 0.933 for this M. caprae set (n = 128). These two panels cannot be compared directly, though, as the former was obtained selectively from Irish cattle and badgers, whereas the latter comprised human and animal isolates from across Europe. However, the most discriminating MIRU alleles were also found to differ between the two panels: MIRU 24 was moderately discriminating among the M. bovis set (36), as among M. tuberculosis (43, 45) isolates, but only poorly discriminating among M. caprae isolates. A similar finding was reported for a collection of 123 M. bovis isolates obtained from Belgian cattle over a period of 8 years (C. Allix, K. Walravens, V. Vandenpoorte, P. Supply, J. Godfroid, and M. Fauville-Dufaux, Abstr. 25th Annu. Meet. Eur. Soc. Mycobacteriol., abstr. 135, 2004). On the other hand, MIRU loci 10, 16, and 31 proved moderately discriminant among M. caprae samples but poorly resolving among the Irish M. bovis panel. Compared to large MTC panels analyzed by MIRU typing (and not including M. caprae) (43, 45), M. caprae isolates were found to be significantly less diverse at locus 40 and notably monomorphic at locus 39. Thus, an optimally reduced set of MIRUs for resolving M. caprae would consist of MIRUs 4, 10, 16, 26, and 31, which is different from other reduced MIRU sets postulated for M. bovis (36). Nevertheless, it will be interesting to test M. caprae within MIRU clusters with additional VNTRs, such as QUBs (41), reported to have a superior resolution in a well-defined M. bovis panel (36).
Direct comparison to IS6110 RFLP is regarded as a challenge for MTC subtyping methods (3, 48), provided that the isolates do not harbor more than three copies of IS6110. M. caprae isolates generally have more than one IS6110 copy and average around five (range, two to eight) copies (4, 16, 19, 34). Testing a subset of 103 M. caprae isolates from outbreaks or endemic foci, we found the same discrimination between isolates from endemic sources by MIRU typing and by IS6110 RFLP and very little variation among outbreak isolates (Table 3). These data suggest that MIRU typing is an alternative that could substitute for RFLP typing. This experience supports the conclusions from an early evaluation of a two-step approach adopted in the United States for genotyping M. tuberculosis from humans, i.e., a combination of automated spoligotyping and MIRU typing that reduces the need for RFLP typing (12).
The geographical distribution of MIRU types among the 128 epidemiological settings suggests that MIRU clusters are also specific enough on a large scale. Of 15 clusters that comprise animal isolates only, 13 were found in a single country and thus provided clues for local and regional transmission or persistence of TB. In contrast, the locations found with mixed clusters (i.e., comprising animal and human isolates) do not suggest epidemiological connections, although the observed transmission between cow and farmer (RIs 43 and 85 [see the al material]) is an exception. This may be explained by the fact that human study isolates originated predominantly from elderly patients and thus have been acquired several decades ago, whereas most animal isolates represent very recent infections. This hypothesis is supported by the high proportion of MIRU unique isolates from humans. In addition, a reduction of M. caprae strain diversity is very likely to have occurred after the eradication measures against "bovine tuberculosis" in the past.
A glimpse of ongoing strain differentiation could be caught in 3 of the 232 isolates, which showed two alleles for a single given MIRU locus. One would expect to "catch" the evolution of a subclone in rare instances, and this is actually an advantage of all VNTR-based methods. On the other hand, a potential problem associated with VNTR typing is homoplasia, i.e., that copy number changes in previously different isolates could lead to identical profiles, e.g., MIRU types. The instances in this study where different spoligotypes were found with identical MIRU patterns are best explained by additional deletion events in the DR gene locus (Table 2). Altogether, this data set does not infer that homoplasia is a major obstacle for the interpretation of MIRU genotyping results.
The underlying concept of MTC phylogeny has been demonstrated to be reflected by MIRU pattern analysis (45), except for M. caprae. These data were reanalyzed after addition of a comprehensive M. caprae sample (Fig. 2). M. caprae isolates (all but one) were distributed in one large group together with the distinct branches for ancestral M. tuberculosis, classical M. bovis, M. pinnipedii, and M. microti, respectively. All are set off from modern M. tuberculosis. Three "M. bovis" isolates (SAU 24, NL 6, and NL 85) were placed among M. caprae isolates at different sites. Those three have been shown to differ from any of the above MTC members of the M. bovis lineage by deletion analysis and single nucleotide polymorphisms (6). In accordance with these findings, 118 out of 128 M. caprae isolates had two copies at locus 24, a locus with a very slow evolution rate (45), as is the case for most isolates in the M. bovis lineage and for "ancestral" M. tuberculosis.
M. caprae has recently been defined as a separate member of the MTC (1). This development accommodates the findings from recent genomic deletion analyses (6, 29) that have likewise set apart "classical" M. bovis from M. pinnipedii or M. microti, i.e., pathogens that cause TB primarily in selected animal species. However, M. caprae causes TB among the same animal species as "classical" M. bovis does. In very practical terms, this implies that the lawful notification process should encompass M. caprae infections as TB and will have to be adapted wherever M. caprae is prevalent.
One interesting question remains. MIRU typing of this representative panel does yield a geographical pattern that would clearly suggest routes of spread of M. caprae in the past. On the one hand, the dominant S1 type is increasingly prevalent from the west part to the east part of Europe. As MTC strains are thought lose DR spacers and not to regain them, the S1 type can also be hypothesized to be ancestral to S4 and S7. On the other hand, a separate analysis of the human versus animal isolates in the same country shows a significantly higher proportion of the S1 type among cattle isolates (86%) than among human isolates (56%). This could reflect recent clonal expansions after an already achieved reduction of M. caprae prevalence by eradication measures. An analysis of 11,500 (classical) M. bovis isolates from the United Kingdom and Ireland by spoligotyping, followed by ETR-VNTR subtyping in part, has suggested clonal expansions leading to the current population of M. bovis in Great Britain (42). Historical animal trading across and from Europe has influenced the different repertoires of M. bovis spoligotypes, e.g., the ones in Great Britain and its trading-partner countries in contrast to those in continental European countries (8, 10, 16, 20, 41). M. caprae has been found almost exclusively in continental Europe and as far east as the eastern Ukraine. One human isolate has been reported from an inhabitant of the European part of Istanbul, Turkey (D. Satana, M. Uzun, Z. Erturan, Y. Yegenoglu, Abstr. 26th Annu. Meet. Eur. Soc. Mycobacteriol., abstr. 11, 2005), but information about the prevalence in Middle Eastern, Asian, or African countries is missing. The close relation of European M. caprae to an isolate from a gazelle found in Africa and Arabia is therefore very interesting and stimulating to further define its origin.
ACKNOWLEDGMENTS
We are grateful to Carolin Lechleitner, Dieter Koefler, and Daniela Loda for excellent technical assistance and to Walter Glawischnig and Maria Beatrice Boniotti for helpful discussions.
This work was supported by the European Commission (QLK2-CT-2000-00630), the Czech Ministry of Agriculture (MZE 0002716201), the Italian Ministry of Health (PRC2001008), the Spanish Ministry of Agriculture, Fisheries, and Food. A.S. is the recipient of a scholarship from the German Academic Exchange Service.
Supplemental material for this article may be found at http://jcm.asm.org/.
These authors contributed equally to this work.
REFERENCES
Aranaz, A., D. Cousins, A. Mateos, and L. Dominguez. 2003. Elevation of Mycobacterium tuberculosis subsp. caprae Aranaz et al. 1999 to species rank as Mycobacterium caprae comb. nov., sp. nov. Int. J. Syst. Evol. Microbiol. 53:1785-1789.
Aranaz, A., E. Liebana, E. Gomez Mampaso, J. Galan, D. Cousins, A. Ortega, J. Blazquez, F. Baquero, A. Mateos, G. Suarez, and L. Dominguez. 1999. Mycobacterium tuberculosis subsp. caprae subsp. nov.: a taxonomic study of a new member of the Mycobacterium tuberculosis complex isolated from goats in Spain. Int. J. Syst. Bacteriol. 49:1263-1273.
Aranaz, A., E. Liebana, A. Mateos, L. Dominguez, and D. Cousins. 1998. Restriction fragment length polymorphism and spacer oligonucleotide typing: a comparative analysis of fingerprinting strategies for Mycobacterium bovis. Vet. Microbiol. 61:311-324.
Aranaz, A., E. Liebana, A. Mateos, L. Dominguez, D. Vidal, M. Domingo, O. Gonzalez, E. F. Rodriguez Ferri, A. E. Bunschoten, J. D. van Embden, and D. Cousins. 1996. Spacer oligonucleotide typing of Mycobacterium bovis strains from cattle and other animals: a tool for studying epidemiology of tuberculosis. J. Clin. Microbiol. 34:2734-2740.
Banu, S., S. V. Gordon, S. Palmer, M. R. Islam, S. Ahmed, K. M. Alam, S. T. Cole, and R. Brosch. 2004. Genotypic analysis of Mycobacterium tuberculosis in Bangladesh and prevalence of the Beijing strain. J. Clin. Microbiol. 42:674-682.
Brosch, R., S. V. Gordon, M. Marmiesse, P. Brodin, C. Buchrieser, K. Eiglmeier, T. Garnier, C. Gutierrez, G. Hewinson, K. Kremer, L. M. Parsons, A. S. Pym, S. Samper, D. van Soolingen, and S. T. Cole. 2002. A new evolutionary scenario for the Mycobacterium tuberculosis complex. Proc. Natl. Acad. Sci. USA 99:3684-3689.
Collins, C. H. 2000. The bovine tubercle bacillus. Br. J. Biomed. Sci. 57:234-240.
Costello, E., D. O'Grady, O. Flynn, R. O'Brien, M. Rogers, F. Quigley, J. Egan, and J. Griffin. 1999. Study of restriction fragment length polymorphism analysis and spoligotyping for epidemiological investigation of Mycobacterium bovis infection. J. Clin. Microbiol. 37:3217-3222.
Cousins, D., R. Skuce, R. Kazwala, J. D. A. van Embden, et al. 1998. Towards a standardized approach to DNA fingerprinting of Mycobacterium bovis. Int. J. Tuberc. Lung Dis. 2:471-478.
Cousins, D., S. Williams, E. Liebana, A. Aranaz, A. Bunschoten, J. van Embden, and T. Ellis. 1998. Evaluation of four DNA typing techniques in epidemiological investigations of bovine tuberculosis. J. Clin. Microbiol. 36:168-178.
Cousins, D. V., R. Bastida, A. Cataldi, V. Quse, S. Redrobe, S. Dow, P. Duignan, A. Murray, C. Dupont, N. Ahmed, D. M. Collins, W. R. Butler, D. Dawson, D. Rodriguez, J. Loureiro, M. I. Romano, A. Alito, M. Zumarraga, and A. Bernardelli. 2003. Tuberculosis in seals caused by a novel member of the Mycobacterium tuberculosis complex: Mycobacterium pinnipedii sp. nov. Int. J. Syst. Evol. Microbiol. 53:1305-1314.
Cowan, L. S., L. Diem, T. Monson, P. Wand, D. Temporado, T. V. Oemig, and J. T. Crawford. 2005. Evaluation of a two-step approach for large-scale, prospective genotyping of Mycobacterium tuberculosis isolates in the United States. J. Clin. Microbiol. 43:688-695.
Crawford, J. T., C. R. Braden, B. A. Schable, and I. M. Onorato. 2002. National Tuberculosis Genotyping and Surveillance Network: design and methods. Emerg. Infect. Dis. 8:1192-1196.
Dale, J. W., D. Brittain, A. A. Cataldi, D. Cousins, J. T. Crawford, J. Driscoll, H. Heersma, T. Lillebaek, T. Quitugua, N. Rastogi, R. A. Skuce, C. Sola, D. van Soolingen, and V. Vincent. 2001. Spacer oligonucleotide typing of bacteria of the Mycobacterium tuberculosis complex: recommendations for standardised nomenclature. Int. J. Tuberc. Lung Dis. 5:216-219.
Erler, W., D. Kahlau, G. Martin, L. Naumann, D. Schimmel, and A. Weber. 2003. The epizootiology of tuberculosis of cattle in the Federal Republic of Germany. Berl. Muench. Tieraerztl. Wochenschr. 116:288-292. (In German.)
Erler, W., G. Martin, K. Sachse, L. Naumann, D. Kahlau, J. Beer, M. Bartos, G. Nagy, Z. Cvetnic, M. Zolnir-Dovc, and I. Pavlik. 2004. Molecular fingerprinting of Mycobacterium bovis subsp. caprae isolates from central Europe. J. Clin. Microbiol. 42:2234-2238.
Espinosa de los Monteros, L. E., J. C. Galán, M. Gutierrez, S. Samper, J. F. García Marín, C. Martín, L. Domínguez, L. de Rafael, F. Baquero, E. Gomez-Mampaso, and J. Blázquez. 1998. Allele-specific PCR method based on pncA and oxyR sequences for distinguishing Mycobacterium bovis from Mycobacterium tuberculosis: intraspecific M. bovis pncA sequence polymorphism. J. Clin. Microbiol. 36:239-242.
Frothingham, R., and O. C. Meeker. 1998. Genetic diversity in the Mycobacterium tuberculosis complex based on variable numbers of tandem DNA repeats. Microbiology 144:1189-1196.
Gutierrez, M., S. Samper, J. A. Gavigan, J. F. Garcia Marin, and C. Martin. 1995. Differentiation by molecular typing of Mycobacterium bovis strains causing tuberculosis in cattle and goats. J. Clin. Microbiol. 33:2953-2956.
Haddad, N., A. Ostyn, C. Karoui, M. Masselot, M. F. Thorel, S. L. Hughes, J. Inwald, R. G. Hewinson, and B. Durand. 2001. Spoligotype diversity of Mycobacterium bovis strains isolated in France from 1979 to 2000. J. Clin. Microbiol. 39:3623-3632.
Hawkey, P. M., E. G. Smith, J. T. Evans, P. Monk, G. Bryan, H. H. Mohamed, M. Bardhan, and R. N. Pugh. 2003. Mycobacterial interspersed repetitive unit typing of Mycobacterium tuberculosis compared to IS6110-based restriction fragment length polymorphism analysis for investigation of apparently clustered cases of tuberculosis. J. Clin. Microbiol. 41:3514-3520.
Hunter, P. R., and M. A. Gaston. 1988. Numerical index of the discriminatory ability of typing systems: an application of Simpson's index of diversity. J. Clin. Microbiol. 26:2465-2466.
Kamerbeek, J., L. Schouls, A. Kolk, M. van Agterveld, D. van Soolingen, S. Kuijper, A. Bunschoten, H. Molhuizen, R. Shaw, M. Goyal, and J. van Embden. 1997. Simultaneous detection and strain differentiation of Mycobacterium tuberculosis for diagnosis and epidemiology. J. Clin. Microbiol. 35:907-914.
Kremer, K., D. van Soolingen, R. Frothingham, W. H. Haas, P. W. M. Hermans, C. Martín, P. Palittapongarnpim, B. B. Plikaytis, L. W. Riley, M. A. Yakrus, J. M. Musser, and J. D. A. van Embden. 1999. Comparison of methods based on different molecular epidemiological markers for typing of Mycobacterium tuberculosis complex stains: interlaboratory study of discriminatory power and reproducibility. J. Clin. Microbiol. 37:2607-2618.
Kubica, T., S. Rüsch-Gerdes, and S. Niemann. 2003. Mycobacterium bovis subsp. caprae caused one-third of human M. bovis-associated tuberculosis cases reported in Germany between 1999 and 2001. J. Clin. Microbiol. 41:3070-3077.
Kwara, A., R. Schiro, L. S. Cowan, N. E. Hyslop, M. F. Wiser, H. S. Roahen, P. Kissinger, L. Diem, and J. T. Crawford. 2003. Evaluation of the epidemiologic utility of secondary typing methods for differentiation of Mycobacterium tuberculosis isolates. J. Clin. Microbiol. 41:2683-2685.
Machackova, M., L. Matlova, J. Lamka, J. Smolik, L. Melicharek, Hanzlikova.M., J. Docekal, Z. Cvetnic, G. Nagy, M. Lipiec, M. Ocepek, and I. Pavlik. 2004. Wild boar (Sus scrofa) as a possible vector of mycobacterial infections: review of literature and critical analysis of data from Central Europe between 1983 to 2001. Vet. Med. Czech. 48:51-65.
Mazars, E., S. Lesjean, A.-L. Banuls, M. Gilbert, V. Vincent, B. Gicquel, M. Tibayrenc, C. Locht, and P. Supply. 2001. High-resolution minisatellite-based typing as a portable approach to global analysis of Mycobacterium tuberculosis molecular epidemiology. Proc. Natl. Acad. Sci. USA 98:1901-1906.
Mostowy, S., D. Cousins, J. Brinkman, A. Aranaz, and M. A. Behr. 2002. Genomic deletions suggest a phylogeny for the Mycobacterium tuberculosis complex. J. Infect. Dis. 186:74-80.
Niemann, S., E. Richter, and S. Rusch-Gerdes. 2002. Biochemical and genetic evidence for the transfer of Mycobacterium tuberculosis subsp. caprae Aranaz et al. 1999 to the species Mycobacterium bovis Karlson and Lessel 1970 (approved lists 1980) as Mycobacterium bovis subsp. caprae comb. nov. Int. J. Syst. Evol. Microbiol. 52:433-436.
Pavlik, I., W. Y. Ayele, M. Havelkova, M. Svejnochova, V. Katalinic-Jankovic, and M. Zolnir-Dovc. 2004. Mycobacterium bovis infection in human population in four Central European countries during 1990-1999. Vet. Med. Czech. 48:90-98.
Pavlik, I., F. Bures, P. Janovsky, P. Pecinka, M. Bartos, L. Dvorksa, L. Matlova, K. Kremer, and D. van Soolingen. 2002. The last outbreak of bovine tuberculosis in cattle in the Czech Republic in 1995 was caused by Mycobacterium bovis supecies caprae. Vet. Med. Czech. 47:251-263.
Pavlik, I., L. Dvorksa, M. Bartos, I. Parmova, I. Melicharek, A. Jesenska, M. Havelkova, M. Slosarek, I. Putova, G. Martin, W. Erler, K. Kremer, and D. van Soolingen. 2002. Molecular epidemiology of bovine tuberculosis in the Czech Republic and Slovakia in the period 1965-2001 studied by spoligotyping. Vet. Med. Czech. 47:181-194.
Prodinger, W. M., A. Eigentler, F. Allerberger, M. Schnbauer, and W. Glawischnig. 2002. Infection of red deer, cattle, and humans with Mycobacterium bovis subsp. caprae in western Austria. J. Clin. Microbiol. 40:2270-2272.
Roring, S., A. Scott, D. Brittain, I. Walker, G. Hewinson, S. Neill, and R. A. Skuce. 2002. Development of variable-number tandem repeat typing of Mycobacterium bovis: comparison of results with those obtained by using existing exact tandem repeats and spoligotyping. J. Clin. Microbiol. 40:2126-2133.
Roring, S., A. N. Scott, H. R. Glyn, S. D. Neill, and R. A. Skuce. 2004. Evaluation of variable number tandem repeat (VNTR) loci in molecular typing of Mycobacterium bovis isolates from Ireland. Vet. Microbiol. 101:65-73.
Saitou, N., and M. Nei. 1987. The neighbor-joining method: a new method for reconstructing phylogenetic trees. Mol. Biol. Evol. 4:406-425.
Savine, E., R. M. Warren, G. D. Van Der Spuy, N. Beyers, P. D. van Helden, C. Locht, and P. Supply. 2002. Stability of variable-number tandem repeats of mycobacterial interspersed repetitive units from 12 loci in serial isolates of Mycobacterium tuberculosis. J. Clin. Microbiol. 40:4561-4566.
Selander, R. K., D. A. Caugant, H. Ochman, J. M. Musser, M. N. Gilmour, and T. S. Whittam. 1986. Methods of multilocus enzyme electrophoresis for bacterial population genetics and systematics. Appl. Environ. Microbiol. 51:873-884.
Skuce, R. A., D. Brittain, M. S. Hughes, L. A. Beck, and S. D. Neill. 1994. Genomic fingerprinting of Mycobacterium bovis from cattle by restriction fragment length polymorphism analysis. J. Clin. Microbiol. 32:2387-2392.
Skuce, R. A., T. P. McCorry, J. F. McCarroll, S. M. Roring, A. N. Scott, D. Brittain, S. L. Hughes, R. G. Hewinson, and S. D. Neill. 2002. Discrimination of Mycobacterium tuberculosis complex bacteria using novel VNTR-PCR targets. Microbiology 148:519-528.
Smith, N. H., J. Dale, J. Inwald, S. Palmer, S. V. Gordon, R. G. Hewinson, and J. M. Smith. 2003. The population structure of Mycobacterium bovis in Great Britain: clonal expansion. Proc. Natl. Acad. Sci. USA 100:15271-15275.
Sola, C., I. Filliol, E. Legrand, S. Lesjean, C. Locht, P. Supply, and N. Rastogi. 2003. Genotyping of the Mycobacterium tuberculosis complex using MIRUs: association with VNTR and spoligotyping for molecular epidemiology and evolutionary genetics. Infect. Genet. Evol. 3:125-133.
Sun, Y. J., R. Bellamy, A. S. Lee, S. T. Ng, S. Ravindran, S. Y. Wong, C. Locht, P. Supply, and N. I. Paton. 2004. Use of mycobacterial interspersed repetitive unit-variable-number tandem repeat typing to examine genetic diversity of Mycobacterium tuberculosis in Singapore. J. Clin. Microbiol. 42:1986-1993.
Supply, P., S. Lesjean, E. Savine, K. Kremer, D. van Soolingen, and C. Locht. 2001. Automated high-throughput genotyping for study of global epidemiology of Mycobacterium tuberculosis based on mycobacterial interspersed repetitive units. J. Clin. Microbiol. 39:3563-3571.
van Embden, J. D., M. D. Cave, J. T. Crawford, J. W. Dale, K. D. Eisenach, B. Gicquel, P. Hermans, C. Martin, R. McAdam, T. M. Shinnick, and P. M. Small. 1993. Strain identification of Mycobacterium tuberculosis by DNA fingerprinting: recommendations for a standardized methodology. J. Clin. Microbiol. 31:406-409.
van Soolingen, D. 2001. Molecular epidemiology of tuberculosis and other mycobacterial infections: main methodologies and achievements. J. Intern. Med. 249:1-26.
van Soolingen, D., P. E. de Haas, J. Haagsma, T. Eger, P. W. Hermans, V. Ritacco, A. Alito, and J. D. van Embden. 1994. Use of various genetic markers in differentiation of Mycobacterium bovis strains from animals and humans and for studying epidemiology of bovine tuberculosis. J. Clin. Microbiol. 32:2425-2433.
van Soolingen, D., P. E. de Haas, P. W. Hermans, and J. D. van Embden. 1994. DNA fingerprinting of Mycobacterium tuberculosis. Methods Enzymol. 235:196-205.(Wolfgang M. Prodinger, An)