Molecular Discrimination of Mycobacterium bovis in São Paulo,Brazil

Abstract

Mycobacterium bovis, a member of the Mycobacterium tuberculosis complex, is the most common agent of cattle tuberculosis, a zoonosis that causes losses in meat and milk production in several countries. In order to support epidemiological studies aimed at controlling the disease, several methods for molecular discrimination of M. bovis isolates have recently been developed. The most frequently used are spacer oligonucleotide typing (spoligotyping), mycobacterial interspersed repetitive units (MIRU), and exact tandem repeat (ETR), but they all have different discriminatory power. In the present study, allelic diversity was calculated for each MIRU and ETR locus, and the Hunter-Gaston discriminatory index (HGI) was calculated for spoligotyping, 10 MIRUs, and 3 ETRs, in 116 isolates of M. bovis obtained from cattle. The analysis of allelic diversity indicated that MIRUs 16, 26, and 27, and ETRs A, B, and C, showed the greatest diversity between the assayed loci. The HGIs for each of the techniques were: spoligotyping=0.738381; MIRU=0.829835; and ETR=0.825337. The associations of the methods' improved discriminatory power were: spoligotyping+MIRU=0.930585; spoligotyping+ETR=0.931034; and MIRU+ETR=0.953373. The greatest discriminatory power was obtained when the three techniques were associated (HGI=0.98051). Considering the analyses of the present study, spoligotyping should be the first method to be used because it differentiates M. bovis from the other members of the Mycobacterium tuberculosis complex. As the associations of MIRU and ETR with spoligotyping resulted in nearly identical HGIs, ETR seems to be the best choice after spoligotyping, because it is faster and more economical than MIRU. Finally, MIRU should be the last method used. In spite of this finding, the choice of the method used should be based on the discriminatory power necessary for the objective at hand.

Introduction

A lthough animal tuberculosis is widely known for affecting cattle and other ruminants, it may affect other animal species as well as humans. The main causative agent of bovine tuberculosis is Mycobacterium bovis, a member of the Mycobacterium tuberculosis complex (Smith et al. 2006). Infection caused by this agent has considerable economic importance in cattle breeding, and it is highly prevalent in animals in some developing countries, causing losses to the milk and meat production chains. In addition, animal tuberculosis has zoonotic potential, and is therefore also a public health concern (Cosivi et al. 1998; Renwick et al. 2007).

High-quality epidemiological data increases the efficiency of bovine tuberculosis control programs. With the new tools available in molecular epidemiology, there is an increasing possibility of finding answers for issues such as the importance of transmission between individual animals and its risk factors, and the role of wild animals as reservoirs (Hilty et al. 2005). The association of molecular methods, mainly those based on polymerase chain reaction (PCR), such as polymerase chain reaction-PCR restriction analysis (PCR-PRA; Telenti et al. 1993), and spacer oligonucleotide typing (spoligotyping; Kamerbeek et al. 1997), mycobacterial interspersed repetitive units-variable numbers of tandem repeats (MIRU-VNTR; Supply et al. 1997, 2000), and exact tandem repeat (ETR; Goyal et al. 1994; Frothingham et al. 1998), have greatly aided in the study of the molecular epidemiology of the M. tuberculosis complex. These methods aid the identification and discrimination of species, and are important tools in epidemiological studies.

A clearer understanding of M. bovis epidemiology in cattle may contribute to better control or to the eradication of disease. The ability to identify and differentiate species, and to track the origin of the infection, may make these tasks easier. In order to do that, it is necessary to have a precise correlation between typing and epidemiological data (Romano et al. 1996).

The intensification of dairy production combined with cattle movement has contributed to M. bovis transmission, especially in the absence of appropriate control measures (Gilbert et al. 2005). Cattle trade between neighboring and partner countries led to the spread of M. bovis, and the dominance of clonal complexes in wide areas (Müller et al. 2008). A recent study of M. bovis clonal complex phylogeography, a method that analyzes the geographical location of molecular types, assessed the existence of four large M. bovis clonal complexes: African 1, African 2, European 1, and European 2 (Müller et al. 2009; Berg et al. 2011; Smith et al. 2011; Rodriguez-Campos et al. 2012). European 1 clonal complex is spread worldwide, and has already been identified in the British Isles, in former British colonies, and in the Americas, as well as in Kazakhstan and Korea, suggesting that existing cattle breeds such as Hereford, created in the U.K. in the 18th century, may have been good vehicles for the worldwide distribution of this closely related group of strains, and that human trade of cattle all over the world led to this worldwide spread of the bovine pathogen over the last 200 years (Smith 2012), corroborating Webb's 1936 hypothesis.

The objectives of the present study were to assess how different spoligotypes of M. bovis isolated from cattle in São Paulo, Brazil, behave in MIRU and ETR, and to calculate the genetic diversity for each MIRU and ETR locus, as well as the discriminatory power of these techniques, either alone or in combination.

Materials and Methods

A total of 116 M. bovis isolates were analyzed. These isolates came from foci of bovine tuberculosis in the state of São Paulo, Brazil, and were initially submitted to PCR-PRA, and subsequently to spoligotyping, MIRU, and ETR. With the results obtained by the three techniques, allelic diversity was calculated for each locus (h) in order to determine the most polymorphic loci, as recommended by Selander and associates (1986). The results were also analyzed by means of the Hunter-Gaston discriminatory index (HGI), as recommended by Hunter and Gaston (1988).

Results

All isolates were submitted to PCR-PRA, and classified in the Mycobacterium genus. Using the profile produced by BstEII and HaeIII enzyme restriction, these isolates were classified as belonging to the M. tuberculosis complex (Institut Pasteur 2000).

Spoligotyping of the 116 samples showed SB0295 (35.34%), followed by SB0121 (32.76%), as the most common spoligotypes.

Analysis of allelic diversity of each locus (h) showed that the most polymorphic loci in MIRU and ETR were MIRUs 16, 26, and 27, and ETRs A, B, and C, followed by MIRUs 23, 31, 39, and 40, and lastly, MIRUs 2, 4, and 10, which did not show polymorphism (Table 1).

Table 1.

Number of Occurrences of MIRUs and ETRs, and Allelic Diversity of Each Locus in São Paulo, Brazil

	Allele number
Locus	0	1	2	3	4	5	6	7	8	9	Allelic diversity (h)
MIRU 2	-	-	116	-	-	-	-	-	-	-	−0.0087
MIRU 4	-	-	-	116	-	-	-	-	-	-	−0.0087
MIRU 10	-	-	116	-	-	-	-	-	-	-	−0.0087
MIRU 16	-	-	11	90	11	4	-	-	-	-	0.373463
MIRU 23	-	-	1	-	115	-	-	-	-	-	0.008546
MIRU 26	-	2	6	12	12	78	4	2	-	-	0.517841
MIRU 27	-	21	2	88	5	-	-	-	-	-	0.384258
MIRU 31	-	-	6	105	5	-	-	-	-	-	0.168966
MIRU 39	-	1	115	-	-	-	-	-	-	-	0.008546
MIRU 40	-	2	113	1	-	-	-	-	-	-	0.042429
ETR A	-	-	2	2	9	13	78	12	-	-	0.513793
ETR B	-	-	-	-	-	8	5	86	16	1	0.41964
ETR C	-	-	-	-	-	27	-	88	1	-	0.364768

MIRU, mycobacterial interspersed repetitive units; ETR, exact tandem repeat.

The HGIs calculated for each technique were: spoligotyping=0.738381; ETR=0.825337; and MIRU=0.829835. When the techniques were associated, there was a significant increase in discriminatory power: spoligotyping+MIRU=0.930585; spoligotyping+ETR=0.931034; MIRU+ETR=0.953373. The greatest discriminatory power was observed when the three techniques were combined (HGI=0.98051; Table 2).

Table 2.

Discriminatory Power of Spoligotyping, MIRU, and ETR, Used Either Alone or in Combination in São Paulo, Brazil

Method	HGI
Spoligotyping	0.73838
MIRU	0.82983
ETR	0.82533
Spoligotyping+MIRU	0.93058
Spoligotyping+ETR	0.93103
MIRU+ETR	0.95337
Spoligotyping+MIRU+ETR	0.98051

MIRU, mycobacterial interspersed repetitive units; ETR, exact tandem repeat; HGI, Hunter-Gaston discriminatory index.

Discussion

Isolates were chosen as a function of spoligotyping results. The five most frequent spoligotypes (SB0121, SB0295, SB0120, SB0140, and SB0881) obtained in cattle slaughtered in the state of São Paulo in the study by Rodriguez and colleagues (2004), were selected.

The occurrence of the different spoligotypes identified in this study in several countries of the world is summarized in Table 3.

Table 3.

List of Different Spoligotypes and Their Occurrence in Different Countries

Spoligotype	Countries
SB0295	Majority in Spain, Mexico, Portugal, and Brazil (Aranaz et al. 1996; Cobos-Marín et al. 2005; Duarte et al. 2008; Parreiras et al. 2012)
	Identified in Brazil, Belgium, France, and Algiers (Zumárraga et al. 1999; Rodriguez et al. 2004; Haddad et al. 2001; Sahraoui et al. 2009)
SB0121	Identified in the Netherlands (Haddad et al. 2001), and in Brazil (Zanini et al. 2001; Rodriguez et al. 2004; Parreiras et al. 2012)
SB0120	Majority in France, with 26% of the strains (Haddad et al. 2001)
	Identified in Belgium, China, Denmark, Iran, Japan, Portugal, Russia, South Africa, Spain, Sri Lanka, the Netherlands, Algiers, and Brazil (Aranaz et al. 1996; Kremer et al. 1999; Van Embden et al. 2000; Sahraoui et al. 2009; Parreiras et al. 2012)
SB0140	Majority in Australia, Argentina, Ireland, and the United Kingdom (Cousins et al. 1998; Zumarraga et al. 1999; Costello et al. 1999; Haddad et al. 2001)
	Identified in Paraguay, Uruguay, Brazil, and Mexico (Aranaz et al. 1996; Kremer et al. 1999; Van Embden et al. 2000; Haddad et al. 2001; Sahraoui et al. 2009)
SB0881	Identified in France and Brazil (Haddad et al. 2001; Rodriguez et al. 2004; Parreiras et al. 2012)

Spoligotyping, the most common molecular epidemiological technique applied to strains of the M. tuberculosis complex, can be used as a preliminary screen to identify members of this complex. M. bovis isolates discriminated by this technique may be compared by researchers in internet databases, making it easier to compare results between laboratories (Kamerbeek et al. 1997).

Analysis of allelic diversity showed that the most polymorphic loci in MIRU and ETR were MIRUs 16, 26, and 27; and ETRs A, B, and C; followed by MIRUs 23, 31, 39, and 40 (Table 1). Hilty and associates (2005), studying M. bovis isolates in the Republic of Chad (Central Africa), observed similar results, with MIRUs 26 and 27, and ETRs A, B, and C, as the most polymorphic loci. Results of the present study also classified these loci as highly polymorphic. In a study carried out in Xinjiang, China, with 135 samples of M. bovis, Sun and colleagues (2011) identified the greatest allelic diversity in nine loci, among them ETRs A, B, D, and E.

A study carried out by Parreiras and co-workers (2012) in Brazil calculated the allelic diversity of 12 MIRU loci. The results showed a moderate polymorphism for MIRU 16 and a high index for MIRU 26. This latter finding is in agreement with the results of the present study.

The analysis of the MIRU and ETR loci showed that these two techniques had similar HGIs (0.829835 and 0.825337, respectively; Table 2). Although the isolates have been chosen based on spoligotyping results, HGI was also calculated for this method, and was equal to 0.738381. Similar results were reported by Kremer and associates (2005), who analyzed the discriminatory power of nine typing methods based on PCR in 100 samples of Mycobacterium from 38 countries. Discrimination levels of the methods tested as determined by HGI were: 0.998 (Queens University of Belfast, QUB-VNTR); 0.995 (MIRU-VNTR); 0.994 (FLiP); 0.987 (ligation-mediated PCR and LM-PCR), and 0.982 (spoligotyping). The difference of the HGI results obtained by Kremer and colleagues (2005) and the ones reported here was due to the inclusion criteria used for the isolates in the present study. In a study carried out by Sun and associates (2004b) in 291 samples collected in Singapore and analyzed for M. tuberculosis genetic diversity using MIRU, the HGI was 0.975, indicating that this technique has high discriminatory power.

In the present study, when spoligotyping was associated with the other two techniques, there was a significant increase in discriminatory power: HGI=0.930585 for spoligotyping+MIRU, and HGI=0.931034 for spoligotyping+ETR (Table 3). In order to evaluate the performance of the combination of MIRU-VNTR and spoligotyping in the detection of tuberculosis transmission chains, Oelemann and co-workers (2007) studied 154 samples from the M. tuberculosis complex in Germany. These authors defined a set of 24 MIRU-VNTR loci, including 15 discriminatory loci. These sets of loci were used to improve the discriminatory power of MIRU-VNTR, especially when associated with spoligotyping, and the results were compared with those obtained by RFLP-IS6110. They concluded that the sets of 15 or 24 MIRU-VNTR loci, combined with spoligotyping, represent the first-choice PCR method for the study of tuberculosis transmission, with operational parameters (sensitivity and specificity) comparable to those of the gold standard (RFLP-IS6110).

Similar results were reported by Parreiras and associates (2012) in Brazil; the discriminatory power of spoligotyping was 0.85, and of MIRU was 0.86. The combination of the two techniques increased the HGI to 0.94.

Table 2 also shows that the HGI for the combination of MIRU and ETR was 0.953373. Sun and colleagues (2004a) reported that the combination of these same techniques produced an HGI of 0.994 for human isolates of M. tuberculosis in Singapore, greater than the result obtained for RFLP-IS6110 (HGI=0.931), or spoligotyping (HGI=0.923). Chin and Jou (2005), studying 502 isolates of M. tuberculosis in Taiwan, also observed that the combination of MIRU and ETR significantly improved the HGI.

Table 2 shows that the greatest discriminatory power was obtained when the three techniques were combined (spoligotyping+MIRU+ETR=0.98051). In 2003, Sola and co-workers evaluated the discriminatory power of these three methods in the genotyping of the M. tuberculosis complex, either alone or in combination, using 103 samples isolated from different geographic origins. In the individual evaluation, the results showed that MIRU typing was more discriminatory (HGI=0.988) than the other two methods (spoligotyping HGI of 0.965; MIRU-VNTR HGI of 0.959), supporting its use in combination with spoligotyping in the genotyping of a large number of samples. The study also showed that MIRU-VNTR may be the best alternative to RFLP. In spite of the fact that they did not report an HGI for the combination, Sola and associates (2003) reported that the results obtained with the association of the three methods were excellent.

Spoligotyping has been demonstrated to be a fast and cost-effective method for first-line typing (Haddad et al. 2004). Smith and colleagues (2006) inferred that a combination of spoligotyping and VNTR typing resulted in the simplest and most cost-effective method for routine molecular typing and epidemiological tracing of bovine tuberculosis in Great Britain.

It should be emphasized that typing by MIRU and ETR have similar costs to perform. However, ETR has a lower final cost, because it has greater discriminatory power with the analysis of fewer loci.

Conclusions

Although the association of the three methods increased the discriminatory power (HGDI=0.98051), the choice of the method used depends on the discriminatory power necessary for the objective at hand. Considering the analyses of the present study, the initial method should be spoligotyping, because it differentiates M. bovis from the other members of the M. tuberculosis complex. As the associations of MIRU and ETR with spoligotyping resulted in nearly identical HGIs, ETR seems to be the best choice after spoligotyping, because it uses fewer loci than MIRU. Finally, MIRU should be the last method to be used.

In the samples isolated in São Paulo, Brazil, MIRUs 16, 26, and 27, and ETRs A, B, and C showed the greatest allelic diversity (i.e., the greatest polymorphism between the assayed loci).

Footnotes

Acknowledgments

The authors would like to thank Fundacao de Amparo A Pesquisa do Esatdo de São Paulo (FAPESP) and Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) for their financial support.

Author Disclosure Statement

No competing financial interests exist.

References

Aranaz

, Liébana

, Mateos

et al. Spacer oligonucleotide typing of Mycobacterium bovis strains from cattle and other animals: a tool for studying epidemiology of tuberculosis. J Clin Microbiol, 1996; 34:2734–2740.

Berg

, Garcia-Pelayo

, Müller

et al. African 2, a clonal complex of Mycobacterium bovis epidemiologically important in East Africa. J Bacteriol, 2011; 193:670–678.

Chin

P-J

, Jou

. A modified automated high-throughput mycobacterial interspersed repetitive unit method for genotyping Mycobacterium tuberculosis. Diag Microbiol Infect Dis, 2005; 53:325–327.

Cobos-Marín

, Montes-Vargas

, Zumarraga

et al. Spoligotype analysis of Mycobacterium bovis isolates from Northern México. Can J Microbiol, 2005; 51:996–1000.

Cosivi

, Grange

, Daborn

et al. Zoonotic tuberculosis due to Mycobacterium bovis in developing countries. Emerg Infect Dis, 1998; 4:59–70.

Costello

, O'Grady

, Flynn

et al. Study of restriction fragment length polymorphism analysis and spoligotyping for epidemiological investigation of Mycobacterium bovis infection. J Clin Microbiol, 1999; 37:3217–3222.

Cousins

, Williams

, Liébana

et al. Evaluation of four DNA typing techniques in epidemiological investigations of bovine tuberculosis. J Clin Microbiol, 1998; 36:168–178.

Duarte

, Domingos

, Amado

et al. Spoligotype diversity of Mycobacterium bovis and Mycobacterium caprae animal isolates. Vet Microbiol, 2008; 130:415–421.

Frothingham

, Meeker-O'Connell

. Genetic diversity in the Mycobacterium tuberculosis complex based on variable numbers of tandem DNA repeats. Microbe, 1998; 144:1189–1196.

10.

Gilbert

, Mitchell

, Bourn

et al. Cattle movements and bovine tuberculosis in Great Britain Periodical. Nature, 2005; 435:491–496.

11.

Goyal

, Young

, Zhang

et al. PCR amplification of variable sequence upstream of katG gene to subdivide strains of Mycobacterium tuberculosis complex. J Clin Microbiol, 1994; 32:3070–3071.

12.

Haddad

, Masselot

, Durand

. Molecular differentiation of Mycobacterium bovis isolates. Review of main techniques and applications. Res Vet Sc, 2004; 76:1–18.

13.

Haddad

, Ostyn

, Karoui

et al. Spoligotype diversity of Mycobacterium bovis strains isolated in France from 1979 to 2000. J Clin Microbiol, 2001; 39:3623–3632.

14.

Hilty

, Diguimbaye

, Schelling

et al. Evaluation of the discriminatory power of variable number tandem repeat (VNTR) typing of Mycobacterium bovis strains. Vet Microbiol, 2005; 109:217–222.

15.

Hunter

, Gaston

. Numerical index of the discriminatory ability of typing systems: an application of Simpson's index of diversity. J Clin Microbiol, 1988; 26:2465–2466.

16.

Institut Pasteur. Identification of Mycobacteria. 2000. http://app.chuv.ch/prasite/index.html

17.

Kamerbeek

, Schouls

, Van Agterveld

et al. Simultaneous detection and strain differentiation of Mycobacterium tuberculosis for diagnosis and epidemiology. J Clin Microbiol, 1997; 35:907–914.

18.

Kremer

, Arnold

, Cataldi

et al. Discriminatory power and reproducibility of novel DNA typing methods for Mycobacterium tuberculosis complex strains. J Clin Microbiol, 2005; 43:5628–5638.

19.

Kremer

, Van Sooligen

, Frothingham

et al. Comparison of methods based on different molecular epidemiological markers for typing of Mycobacterium tuberculosis strains: interlaboratory study of discriminatory power and reproducibility. J Clin Microbiol, 1999; 37:2607–2618.

20.

Müller

, Hilty

, Berg

et al. African 1, an epidemiologically important clonal complex of Mycobacterium bovis dominant in Mali, Nigeria, Cameroon, and Chad. J Bacteriol, 2009; 191:1951–1960.

21.

Müller

, Steiner

, Bonfoh

et al. Molecular characterization of Mycobacterium bovis isolated from cattle slaughtered at the Bamako abattoir in Mali. BMC Vet Res, 2008; 4:26.

22.

Oelemann

, Diel

, Vatin

et al. Assessment of an optimized Mycobacterial Interspersed Repetitive-Unit-Variable-Number Tandem-Repeat typing system combined with Spoligotyping for population-based molecular epidemiology studies of tuberculosis. J Clin Microbiol, 2007; 45:691–697.

23.

Parreiras

, Andrade

, Nascimento

et al. Spoligotyping and variable number tandem repeat analysis of Mycobacterium bovis isolates from cattle in Brazil. Mem Inst Oswaldo Cruz, Rio de Janeiro, 2012; 107:64–73.

24.

Renwick

, White

, Bengis

. Bovine tuberculosis in southern African wildlife: a multi-species host–pathogen system. Epidemiol Infect, 2007; 135:529–540.

25.

Rodriguez-Campos

, Schürch

, Dale

et al. European 2 - A clonal complex of Mycobacterium bovis dominant in the Iberian Peninsula. Infect Genet Evol, 2012; 12:866–872.

26.

Rodriguez

CAR

, Zumárraga

, Oliveira

EMD

et al. Caracterização molecular de isolados de Mycobacterium bovis do Estado de São Paulo Brasil, utilizando a técnica de Spoligotyping. Arq Inst Biol, 2004; 71:277–282.

27.

Romano

, Alito

, Fisanotti

, Bigi

, Kantor

, Cicuta

, Cataldi

. Comparison of different genetic markers for molecular epidemiology of bovine tuberculosis. Vet Microbiol., 1996 May 50:59–71.

28.

Sahraoui

, Müller

, Guetarni

et al. Molecular characterization of Mycobacterium bovis strains isolated from cattle slaughtered at two abattoirs in Algeria. BMC Vet Res, 2009; 5:4.

29.

Selander

, Caugant

, Ochman

et al. Methods of multilocus enzyme electrophoresis for bacterial population genetics and systematics. Appl Environ Microbiol, 1986; 51:873–884.

30.

Smith

, Berg

, Dale

et al. European 1: A globally important clonal complex of Mycobacterium bovis. Infect Genet Evol, 2011; 11:1340–1351.

31.

Smith

, Gordon

, de La Rua-Domenech

et al. Bottlenecks and broomsticks: the molecular evolution of Mycobacterium bovis. Nat Rev Microbiol, 2006; 4:670–681.

32.

Smith

. The global distribution and phylogeography of Mycobacterium bovis clonal complexes. Infect Genet Evol, 2012; 12:857–865.

33.

Sola

, Filliol

, Legrand

et al. Genotyping of the Mycobacterium tuberculosis complex using MIRUs: association with and spoligotyping for molecular epidemiology and evolutionary genetics. Infect Genet Evol, 2003; 3:125–133.

34.

Supply

, Magdalena

, Himpens

et al. Identification of novel intergenic repetitive units in a mycobacterial two-component system operon. Mol Microbiol, 1997; 26:991–1003.

35.

Supply

, Mazars

, Lesjean

et al. Variable human minisatellite-like regions in the Mycobacterium tuberculosis genome. Mol Microbiol, 2000; 36:762–771.

36.

Sun

Y-J

, Bellamy

, Lee

ASG

et al. Use of Mycobacterial Interspersed Repetitive Unit–Variable-Number Tandem Repeat typing to examine genetic diversity of Mycobacterium tuberculosis in Singapore. J Clin Microbiol, 2004b. 42:1986–1993.

37.

Sun

Y-J

, Lee

ASG

, Sze Ta Ng

et al. Characterization of ancestral Mycobacterium tuberculosis by multiple genetic markers and proposal of genotyping strategy. J Clin Microbiol, 2004a. 42:5058–5064.

38.

Sun

, Cao

, Tian

et al. Evaluation of Spoligotyping and MIRU-VNTR for Mycobacterium bovis in Xinjiang, China. Res Vet Sc, 2011, 10.1016/j.rvsc.2011.03.012

39.

Telenti

, Marchesi

, Balz

et al. Rapid identification of Mycobacteria to the species level by polymerase chain reaction and restriction enzyme analysis. J Clin Microbiol, 1993; 31:175–178.

40.

Van Embden

, Van Gorkom

, Kremer

et al. Genetic variation and evolutionary origin of the direct repeat locus of Mycobacterium tuberculosis complex bacteria. J Bacteriol, 2000; 182:2393–2401.

41.

Zanini

, Moreira

, Lopes

MTP

et al. Mycobacterium bovis: polymerase chain reaction identification in bovine lymph node biopsies and genotyping in isolates from Southeast Brazil by spoligotyping and restriction fragment length polymorphism. Mem Inst Oswaldo Cruz, 2001; 96:809–813.

42.

Zumarraga

, Martín

, Samper

et al. Usefulness of spoligotyping in molecular epidemiology of Mycobacterium bovis–related infections in South America. J Clin Microbiol, 1999; 37:296–303.