Characterization of the Mycobacterium bovis restriction fragment length polymorphism DNA probe pUCD and performance comparison with standard methods

In this study, the newly described Mycobacterium bovis restriction fragment length polymorphism (RFLP) typing probe pUCD was characterized by sequence analysis and the previously observed polymorphic banding pattern was reproduced with a combination of three oligonucleotide probes in a single, mixed hybridization. In addition, the ability of pUCD to distinguish between 299 M. bovis isolates from the Republic of Ireland was assessed in relation to established methods and a statistical function for objective comparison of RFLP probes was derived. It was found that typing with pUCD alone produced greater discrimination between M. bovis isolates than typing with the commonly used mycobacterial DNA probes IS6110, PGRS, and DR and also by the spoligotyping technique. pUCD and DR in combination produced the highest level of discrimination while maintaining a high level of concordance with known epidemiological data relating to the samples. The reduction of pUCD to the level of oligonucleotides should in future allow pUCD and DR to be included together in a mixed hybridization, thus producing a high level of M. bovis strain type discrimination from a single round of RFLP analysis.     

Identification of a novel DNA probe for strain typing Mycobacterium bovis by restriction fragment length polymorphism analysis

Bovine tuberculosis caused by Mycobacterium bovis remains a significant disease of farmed cattle in many countries despite ongoing tuberculosis eradication programs. Molecular typing methods such as restriction fragment length polymorphism (RFLP) analysis and spoligotyping have been used to identify related herd breakdowns in an attempt to identify more precisely the route of infection into cattle herds and to trace the transmission of bovine tuberculosis. A recent geographical survey of Irish M. bovis isolates demonstrated that a significant proportion of isolates ( approximately 20%) exhibit a common strain type, limiting the value of current strain typing methods as an epidemiological tool. We have identified and cloned a region of the M. bovis genome, pUCD, which generates a clear, highly polymorphic banding pattern when used as an RFLP probe on AluI restriction-digested M. bovis genomic DNA and which effectively subdivides this common strain type. When used to type 60 Irish M. bovis isolates, pUCD exhibited greater discriminatory power than the commonly used mycobacterial RFLP probes IS6110, PGRS, and DR and detected an equivalent number of strain types to a combination of these three probes. pUCD also detected significantly more strain types than the spoligotyping technique, while maintaining a high level of concordance between epidemiologically related and unrelated herd breakdowns. The polymorphic element within pUCD remains to be fully characterized, however the potential for this probe to greatly decrease the workload necessary to genotype M. bovis by RFLP analysis is compelling.     

Study of Restriction Fragment Length Polymorphism Analysis and Spoligotyping for Epidemiological Investigation of Mycobacterium bovis Infection

Restriction fragment length polymorphism (RFLP) analysis with probes derived from the insertion element IS6110, the direct repeat sequence, and the polymorphic GC-rich sequence (PGRS) and a PCR-based typing method called spacer oligonucleotide typing (spoligotyping) were used to strain type Mycobacterium bovis isolates from the Republic of Ireland. Results were assessed for 452 isolates which were obtained from 233 cattle, 173 badgers, 33 deer, 7 pigs, 5 sheep, and 1 goat. Eighty-five strains were identified by RFLP analysis, and 20 strains were identified by spoligotyping. Twenty percent of the isolates were the most prevalent RFLP type, while 52% of the isolates were the most prevalent spoligotype. Both the prevalent RFLP type and the prevalent spoligotype were identified in isolates from all animal species tested and had a wide geographic distribution. Isolates of some RFLP types and some spoligotypes were clustered in regions consisting of groups of adjoining counties. The PGRS probe gave better differentiation of strains than the IS6110 or DR probes. The majority of isolates from all species carried a single IS6110 copy. In four RFLP types IS6110 polymorphism was associated with deletion of fragments equivalent in size to one or two direct variable repeat sequences. The same range and geographic distribution of strains were found for the majority of isolates from cattle, badgers, and deer. This suggests that transmission of infection between these species is a factor in the epidemiology of M. bovis infection in Ireland.     

False molecular clusters due to nonrandom association of IS6110 with Mycobacterium tuberculosis

We sought to determine whether nonrandom association of IS6110 with Mycobacterium tuberculosis could result in false-positive clustering in unselected collections of isolates. We typed 196 strains of M. tuberculosis from an unselected community-based study in northern Tanzania by IS6110 and polymorphic GC-rich repetitive-sequence (PGRS) methodologies. The strains were analyzed by Gelcompar computer software. Analysis of 13 out of 25 groups showed that isolates with identical IS6110 and PGRS patterns were likely to be the same strain. Some IS6110 groups containing strains with identical PGRS patterns had similar IS6110 patterns that differed only by movement of the element. Isolates assigned to a single group (i.e., group 11) on the basis of sharing an identical IS6110 fingerprint pattern did not share identical PGRS fingerprint patterns. Six out of the nine bands in these isolates were in hot-spot locations, as previously defined. This indicates that nonrandom association may result in false-positive clustering in unselected community-based studies. Only strains with identical PGRS and IS6110 patterns are likely to be recently transmitted.     

Molecular epidemiology of Mycobacterium tuberculosis strains isolated from patients in a human immunodeficiency virus cohort in Switzerland.

From 1989 to 1995, 46 patients infected with the human immunodeficiency virus were diagnosed with tuberculosis at the University Hospital in Zurich. Using the IS6110 insertion sequence as a genetic marker, restriction fragment length polymorphism analyses were done for 52 Mycobacterium tuberculosis isolates. We have found a large degree of IS6110 polymorphism, ranging from 1 to 16 copies. For isolates from patients from whom multiple isolates had been available, the IS6110 pattern remained virtually stable over a period of up to 4 years, as well as during emerging drug resistance. In none of the cases was a reinfection of a patient with another strain detected. For isolates from 10 patients we detected identical patterns which could be associated with four clusters. In one of these, the strains exhibited a low IS6110 copy number (four bands), and the strains were further analyzed by hybridizing with (i) the polymorphic GC-rich repetitive sequence (PGRS) and (ii) the 36-bp direct-repeat (DR) cluster sequence. One of these isolates had a different pattern with the PGRS as well as with the DR sequence and could therefore be safely excluded from that cluster. These findings point to the importance of applying more than one genetic criterion in the molecular biological study of strain relatedness.     

Evaluation of Four DNA Typing Techniques in Epidemiological Investigations of Bovine Tuberculosis

DNA fingerprinting techniques were used to type 273 isolates of Mycobacterium bovis from Australia, Canada, the Republic of Ireland, and Iran. The results of restriction fragment length polymorphism (RFLP) analysis with DNA probes from IS6110, the direct repeat (DR), and the polymorphic GC-rich sequence (PGRS) were compared with those of a new PCR-based method called spacer oligonucleotide typing (spoligotyping) developed for the rapid typing of Mycobacterium tuberculosis (J. Kamerbeek et al., J. Clin. Microbiol. 35:907–914, 1997). Eighty-five percent of the isolates harbored a single copy of IS6110, and 81.5% of these carried IS6110 on the characteristic 1.9-kb restriction fragment. RFLP analysis with IS6110 identified 23 different types, RFLP analysis with the DR probe identified 35 types, RFLP analysis with the PGRS probe identified 77 types, and the spoligotyping method identified 35 types. By combining all results, 99 different strains could be identified. Isolate clusters were frequently associated within herds or were found between herds when epidemiological evidence confirmed animal movements. RFLP analysis with IS6110 was sufficiently sensitive for the typing of isolates with more than three copies of IS6110, but RFLP analysis with the PGRS probe was the most sensitive typing technique for strains with only a single copy of IS6110. Spoligotyping may have advantages for the rapid typing of M. bovis, but it needs to be made more sensitive.     

Secondary typing of Mycobacterium tuberculosis isolates with matching IS6110 fingerprints from different geographic regions of the United States

Fifty-nine isolates of Mycobacterium tuberculosis obtained from different states in the United States and representing 25 interstate clusters were investigated. These clusters were identified by computer-assisted analysis of DNA fingerprints submitted during 1996 and 1997 by different laboratories participating in the CDC National Genotyping and Surveillance Network. Isolates were fingerprinted with the IS6110 right-hand probe (IS6110-3'), the IS6110 left-hand probe (IS6110-5'), and the probe pTBN12, containing the polymorphic GC-rich sequence (PGRS). Spoligotyping based on the polymorphism in the 36-bp direct-repeat locus was also performed. As a control, 43 M. tuberculosis isolates in 17 clusters obtained from patients in Arkansas during the study period were analyzed. Of the 25 interstate clusters, 19 were confirmed as correctly clustered when all the isolates were analyzed on the same gel using the IS6110-3' probe. Of the 19 true IS6110-3' clusters, 10 (53%) were subdivided by one or more secondary typing methods. Clustering of the control group was virtually identical by all methods. Of the three different secondary typing methods, spoligotyping was the least discriminating. IS6110-5' fingerprinting was as discriminating as PGRS fingerprinting. The data indicate that the IS6110-5' probe not only is a useful secondary typing method but also probably would prove to be a more useful primary typing method for a genotyping network which involves isolates from different geographic regions.     

Evaluation of tuberculosis transmission in a community by 1 year of systematic typing of Mycobacterium tuberculosis clinical isolates.

Interhuman transmission of Mycobacterium tuberculosis was investigated by using molecular typing, including restriction fragment length polymorphism with probes IS6110, DR (direct repeat) and PGRS (polymorphic GC-rich sequence) and a PCR method using the inverted repeat sequences of IS6110 as primers. From 105 patients hospitalized for tuberculosis during a 1-year survey in three hospitals in Paris, France, 111 isolates were collected and analyzed. Eighty-eight patients were infected with genetically different isolates, demonstrating the clonal heterogeneity of M. tuberculosis in these patients originating from various geographical areas. Fourteen patients were infected by strains clustered with identical fingerprints. An epidemiological relatedness was demonstrated for isolates from only seven of these patients. Thus, the typing of isolates from all tuberculous patients in hospitals during 1 year allows the detection of transmission in the general community. This would improve the case findings, thereby further improving the detection of outbreaks.     

Molecular characterization of Mycobacterium tuberculosis H37Rv/Ra variants: distinguishing the mycobacterial laboratory strain

The Mycobacterium tuberculosis strains H37Rv and H37Ra are the most commonly used controls for M. tuberculosis identification in the clinical and research laboratory setting. To reduce the likelihood of misidentification and possible cross-contamination with this laboratory neotype, it is important to be able to distinguish H37 from clinical isolates. To provide a reference for identifying H37, we used multiple molecular techniques to characterize H37 strains, including 18 of the most frequently used variants available through the American Type Culture Collection. Isolates were genotyped using gene probes to IS6110 and IS1085. In addition, we performed polymorphic GC-rich sequence typing (PGRS), spoligotyping, determination of variable number of tandem repeats (VNTR), and PCR amplification of the mtp40, msx4, and mpp8 polymorphic regions. Southern hybridization with IS6110 provided the most discrimination, differentiating the 18 H37 isolates into 10 discrete patterns made up of 9 H37Rv variants and 1 H37Ra variant. PGRS, IS1085, mpp8, and spoligotyping were not able to distinguish any H37 variants, while VNTR and msx4 discriminated two. Only IS6110 and spoligotyping could distinguish the H37 strain from clinical isolates. In summary, spoligotyping and IS6110 provide a rapid and accurate way to identify H37 contamination, though IS6110 can, in addition, classify many of the H37 variants that would otherwise require phenotypic segregation.     

Differentiation by molecular typing of Mycobacterium bovis strains causing tuberculosis in cattle and goats.

Forty Mycobacterium bovis isolates from cattle and goats were analyzed by using different repetitive genetic markers. The 23 M. bovis strains from goats were found to carry six to eight copies of the insertion sequence IS6110. In contrast, most of the bovine isolates contained only a single copy of this element. The standardized IS6110 fingerprinting by restriction fragment length polymorphism (RFLP), described for Mycobacterium tuberculosis strains, allowed the differentiation of caprine strains. Although this method was not useful for typing bovine isolates, the repetitive elements pTBN12 and DR proved to be suitable for this purpose. A procedure using PCR which amplifies IS6110 in the outward direction was found to be as sensitive as RFLP for typing M. bovis strains from goats. The use of PCR and RFLP methods based on the IS6110 polymorphism would be useful for epidemiological studies of caprine tuberculosis. The results are consistent with different strains of M. bovis being implicated in bovine and caprine tuberculosis.     

Fast ligation-mediated PCR, a fast and reliable method for IS6110-based typing of Mycobacterium tuberculosis complex

IS6110 restriction fragment length polymorphism (RFLP) analysis is the most widely applied method for strain differentiation of Mycobacterium tuberculosis complex. We have previously described mixed-linker PCR, an IS6110-based PCR method that favorably compared with other typing methods for M. tuberculosis complex according to reproducibility and ability to differentiate between strains. Here we report the further development of this method, called fast ligation-mediated PCR (FLiP), which allows analysis of strains within one working day and starting from less than 1 ng of mycobacterial DNA or a crude cell lysate. Blinded analysis of a standard set of 131 M. tuberculosis complex and nontuberculous isolates showed the ability to differentiate 81 types among 90 M. tuberculosis complex isolates with 84 different IS6110 RFLP fingerprint patterns and detected 97% of the 31 duplicate samples. We suggest that FLiP can serve to rapidly detect chains of transmission prior to starting high-throughput analysis or standard IS6110 RFLP. It may as well serve as a secondary typing technique for other, non-IS6110-based methods.     

Evaluation of spoligotyping in a study of the transmission of Mycobacterium tuberculosis.

Spoligotyping (for spacer oligotyping) is an easy, economical, and rapid way of typing Mycobacterium tuberculosis complex strains with the DR spacer markers (J. Kamerbeek et al., J. Clin. Microbiol. 35:907-914, 1997; D. van Soolingen et al., 33:3234-3248, 1995). The stability of the markers was demonstrated by showing that all the Mycobacterium bovis BCG strains tested gave the same spoligotyping pattern. None of the 42 atypical mycobacterial strains tested gave a spoligotyping signal, indicating the specificity of the technique for M. tuberculosis complex. The utility of the spoligotyping method was demonstrated by analyzing 106 isolates of M. tuberculosis obtained over 1 year in three Paris hospitals. The results obtained by this technique were compared to those obtained by Torrea et al. (G. Torrea et al., J. Clin. Microbiol. 34:1043-1049, 1996) by IS6110-based restriction fragment length polymorphism (RFLP) analysis. Strains from patients with epidemiological relationships that were in the same IS6110-RFLP cluster were also in the same spoligotyping group. Spoligotyping was more discriminative than RFLP analysis for strains with one or two copies of IS6110. RFLP analysis did not discriminate between the nine strains with one or two IS6110 bands with no known epidemiological relation, whereas spoligotyping distinguished between eight different types. IS6I10-RFLP analysis split some of the spoligotyping clusters, particularly when the IS6110 copy number was high. Therefore, we propose a strategy for typing M. tuberculosis strains in which both markers are used.     

Differentiation of Mycobacterium bovis isolates from animals by DNA typing.

The insertion sequence IS6110 and the direct repeat (DR) specific to tuberculosis complex mycobacteria and the highly repeated DNA sequence, the polymorphic GC-rich repeat sequence (PGRS), were systematically used to identify restriction fragment length polymorphisms (RFLPs) within 210 isolates of Mycobacterium bovis. The isolates were primarily of bovine origin, but isolates from badgers, feral deer, sheep, humans, and a pig were included. The RFLP probes IS6110, DR, and PGRS individually identified 17, 18, and 18 different RFLP types, respectively, but in combination these probes identified a total of 39 different M. bovis RFLP types. The recommendations (J. D. A. van Embden, M. D. Cave, J. T. Crawford, J. W. Dale, K. D. Eisenach, B. Gicquel, P. W. M. Hermans, C. Martin, R. McAdam, T. M. Shinnick, and P. M. Small, J. Clin. Microbiol. 31:406-409, 1993) for a standardized RFLP analysis for M. tuberculosis were adapted to facilitate gel documentation, image analysis, and construction of a database of RFLP types. In the present study the same M. bovis RFLP types were evident in the various animal species included, indicating that the strains were not host restricted. Application of these techniques to defined field studies should help elucidate more accurately aspects of the epidemiology of bovine tuberculosis in different countries.     

Molecular typing of Mycobacterium bovis isolates from Cameroon

In order to gain a better understanding of the molecular epidemiology of Mycobacterium bovis isolates in Cameroon, 75 isolates of M. bovis collected in three provinces of northern Cameroon were studied by spoligotyping. For 65 of these isolates, typing was also carried out by pulsed-field gel electrophoresis (PFGE) with DraI, and 18 of the isolates were also typed by restriction fragment length polymorphism (RFLP) analysis with probe IS6110-RHS. Molecular typing of the isolates by these techniques revealed a high degree of homogeneity, with 10 spoligotypes for 75 isolates, four PFGE profiles for 65 isolates, and three RFLP types for 18 isolates. Some types were present in the three different provinces, while some were confined to one or two areas. These results suggest that geographical mapping of M. bovis strains could be helpful for the control of bovine tuberculosis at the regional level. An interesting feature of all the spoligotypes was the absence of spacer 30, suggesting a common origin for all of the Cameroon isolates tested; an evolutionary scenario for the isolates is discussed. In addition, a comparison of the three techniques showed that for M. bovis strain differentiation in Cameroon and in surrounding countries, spoligotyping would be a more discriminating and practical tool for molecular typing than the other two techniques used in this study.     

Molecular characteristics of strains of the cameroon family, the major group of Mycobacterium tuberculosis in a country with a high prevalence of tuberculosis

A preliminary investigation of the genetic biodiversity of Mycobacterium tuberculosis complex strains in Cameroon, a country with a high prevalence of tuberculosis, described a group of closely related M. tuberculosis strains (the Cameroon family) currently responsible for more than 40% of smear-positive pulmonary tuberculosis cases. Here, we used various molecular methods to study the genetic characteristics of this family of strains. Cameroon family M. tuberculosis strains (i) are part of the major genetic group 2 and lack the TbD1 region like other families of epidemic strains, (ii) lack spacers 23, 24, and 25 in their direct repeat (DR) region, (iii) have an identical number of repeats in 8 of 12 variable-number tandem repeats of mycobacterial interspersed repetitive unit (MIRU-VNTR) loci, (iv) have similar IS6110-restriction fragment length polymorphism (RFLP) multiband patterns (10 to 15 copies) with seven common IS6110 bands, (v) do not have an IS6110 element in their DR locus, and (vi) have four IS6110 elements in open reading frames (adenylate cyclase, phospholipase C, moeY, and ATP binding genes). Analysis by spoligotyping, MIRU-VNTR, and IS6110-RFLP typing methods revealed differences not observed in previous studies; polymorphism as assessed by MIRU-VNTR typing was lower than suggested by spoligotyping, and in rare cases, strains with identical IS6110-RFLP patterns had spoligotypes differing by as much as 15 spacers. Our findings confirm the recent expansion of this family in Cameroon and indicate that the interpretation of molecular typing results has to be adapted to the characteristics of the strain population within each setting. The knowledge of this particular genotype, with its large involvement in tuberculosis in Cameroon, allows greater refinement of tuberculosis transmission studies by interpreting data in the context of this geographic area.     

Usefulness of Spoligotyping in Molecular Epidemiology of Mycobacterium bovis-Related Infections in South America

Two hundred twenty-four Mycobacterium bovis isolates, mainly from South American countries, were typed by spoligotyping, and 41 different spoligotypes were identified. A total of 202 M. bovis isolates (90%) were grouped into 19 different clusters. The largest cluster contained 96 isolates (42.8%) on the basis of the most frequently observed spoligotype, spoligotype 34. Nineteen M. bovis isolates from humans in Argentina had spoligotypes and polymorphic GC-rich repetitive sequence (PGRS) types that represented the most common types found among isolates from cattle. All five isolates from Uruguay and three of the six isolates from Paraguay had spoligotypes that were also detected for isolates from Argentina. The spoligotypes of isolates from Brazil, Costa Rica, and Mexico and of some of the isolates from Paraguay could not be found in Argentina. A total of 154 M. bovis isolates were selected in order to compare the discriminative power of spoligotyping and restriction fragment length polymorphism (RFLP) analysis with direct repeat (DR) and PGRS probes. By spoligotyping, 31 different types were found, while AluI-digested DR probe-associated RFLP analysis identified 42 types, and RFLP analysis with the PGRS probe also detected 42 types; these were partly independent of the DR types. By combining the results obtained by spoligotyping and by RFLP analysis with the DR and PGRS probes, 88 different types were obtained. Although the differentiation of M. bovis by spoligotyping was less discriminatory than differentiation by RFLP analysis with the DR and PGRS probes, spoligotyping is easier to perform and its results are easier to interpret. Therefore, for the purpose of typing of M. bovis isolates, spoligotyping could be performed first and the isolates could be grouped into clusters and then analyzed by RFLP analysis with the DR and PGRS probes.     

Evaluation of Mycobacterium tuberculosis typing methods in a 4-year study in Schleswig-Holstein, Northern Germany

In order to evaluate the discriminatory power of different methods for genotyping of Mycobacterium tuberculosis complex (MTBC) isolates, we compared the performance of (i) IS6110 DNA fingerprint typing, (ii) spoligotyping, and (iii) 24-loci mycobacterial interspersed repetitive units-variable number of tandem repeats (MIRU-VNTR) typing in a long-term study on the epidemiology of tuberculosis (TB) in Schleswig-Holstein, the northernmost federal state of Germany. In total, we analyzed 277 MTBC isolates collected from patients between the years 2006 and 2010. The collection comprised a broad spectrum of 13 different genotypes, among which strains of the Haarlem genotype (31%) were most prominent, followed by strains belonging to the Delhi and Beijing lineages (7% and 6%, respectively). On the basis of IS6110 restriction fragment length polymorphism (RFLP) and spoligotyping analyses, 211 isolates had unique patterns (76%) and 66 isolates (24%) were in 20 clusters. MIRU-VNTR combined with spoligotyping analyses revealed 202 isolates with unique patterns (73%) and 75 isolates in 18 clusters (27%). Overall, there was 93.1% concordance between the typing results obtained; 198 strains were identified as unique, and 60 isolates were clustered by both typing combinations (including all 31 isolates with confirmed epidemiological links). Of the remaining 19 isolates with discrepant results, 15 were falsely clustered by MIRU-VNTR (six Beijing genotype strains) and four were clustered by IS6110 RFLP (low IS6110 copy number) only. In conclusion, in the study population investigated, a minority of isolates, especially of the Beijing genotype, clustered by standard 24-loci MIRU-VNTR and without an obvious epidemiological link may require second-line typing by IS6110 RFLP or hypervariable MIRU-VNTR loci.     

Use of a PCR method based on IS6110 polymorphism for typing Mycobacterium tuberculosis strains from BACTEC cultures.

Two PCR typing methods, based on polymorphism of the insertion sequence IS6110, were compared with Mycobacterium tuberculosis strains by using a single primer complementary to the inverted repeats of IS6110. Total M. tuberculosis DNA either was amplified directly (IS6110-PCR) or was amplified following digestion and ligation (IS6110-inverse-PCR). Both PCR techniques showed a similar degree of discrimination. Because of its simplicity, IS6110-PCR was chosen to confirm that a single M. tuberculosis strain was responsible for an outbreak of tuberculosis in a secondary school. IS6110-PCR was used to study the degree of differentiation in 85 clinical M. tuberculosis isolates from BACTEC 12B broth cultures. Results were consistent with those of the standardized IS6110 restriction fragment length polymorphism (RFLP) analysis method, showing identical PCR types for identical RFLPs, although the degree of discrimination was greater by RFLP analysis. The study concludes that due to its simplicity, IS6110-PCR is a good screening method when quick differentiation between M. tuberculosis strains is needed because BACTEC cultures may be used directly.     

A combination of two genetic markers is sufficient for restriction fragment length polymorphism typing of Mycobacterium tuberculosis complex in areas with a high incidence of tuberculosis

The incidence of tuberculosis (TB) in Madagascar is 150 cases per 100,000 people. Because of this endemicity, we studied the genetic diversity of Mycobacterium tuberculosis strains isolated in four big cities in 1994 to 1995 with the aim of monitoring TB transmission. Isolates from 316 cases of pulmonary TB (PTM(+)) were typed by Southern hybridization with genetic markers IS6110 and DR. Of the 316 PTM(+) strains, 66 (20.8%) had a single IS6110 band and were differentiated by the DR marker into 33 profiles. Using both markers, 37.7% (119) of the patients were clustered, a proportion similar to that in countries with a high prevalence of TB. There was no significant difference between clustered and nonclustered patients in age, sex, Mycobacterium bovis BCG status, and drug susceptibility of strains. Clustering was significantly greater in the capital, Antananarivo, than in the other cities, suggesting a higher rate of transmission. However, most of the patients in clusters were living in different areas, and, within a distance of 0.7 km, we did not find epidemiologically unrelated strains with the same restriction fragment length polymorphism profile. Despite an apparently low polymorphism, genetic markers such as IS6110 are potentially valuable for monitoring TB transmission. However, the high proportion of Malagasy isolates with a single IS6110 copy makes this marker alone unsuitable for typing. Additional markers such as DR are necessary for the differentiation of the isolates and for epidemiological surveys.