Combination of Single Nucleotide Polymorphism and Variable ...

Combination of Single Nucleotide Polymorphism and VariableNumber Tandem Repeats for Genotyping a Homogenous Population of Mycobacterium tuberculosis Beijing Strains in China Tao Luo,a Chongguang Yang,a Sebastien Gagneux,b Brigitte Gicquel,c Jian Mei,d and Qian Gaoa Key Laboratory of Medical Molecular Virology, Institutes of Biomedical Sciences and Institute of Medical Microbiology, Fudan University, Shanghai, Chinaa; Swiss Tropical and Public Health Institute, Basel, Switzerlandb; Mycobacterial Genetics Unit, Institut Pasteur, Paris, Francec; and Departments of TB Control, Shanghai Municipal Centers for Disease Control and Prevention, Shanghai, Chinad

The standard 15- and 24-locus variable-number tandem repeat (VNTR) genotyping methods have demonstrated adequate discriminatory power and a small homoplasy effect for tracing tuberculosis (TB) transmission and predicting Mycobacterium tuberculosis lineages in European and North American countries. However, its validity for the definition of transmission in homogenous M. tuberculosis populations in settings with high TB burdens has been questioned. Here, we genotyped a populationbased collection of 191 Beijing strains based on standard 15-locus VNTR (VNTR-15) and 8 single nucleotide polymorphisms (SNPs) in Shanghai, China. Limited discriminatory power and high rates of VNTR homoplasy were observed in the homogenous population of evolutionarily “modern” Beijing strains. Additional typing of three hypervariable loci (VNTR3820, VNTR4120, and VNTR3232) was performed for VNTR-15-based clusters. High variations of hypervariable alleles were observed in clusters with inconsistent SNP sublineages. We concluded that SNPs and hypervariable VNTR loci are helpful to enhance the discriminatory power and decrease the VNTR homoplasy effect for defining clusters. We recommend the combination of standard VNTR-15 and SNPs as first-line typing methods and the hypervariable loci for second-line typing of clustered strains for molecular epidemiology studies of homogenous M. tuberculosis populations.

S

tudy of the molecular epidemiology of tuberculosis (TB) has enhanced our understanding of this ancient disease. Three main molecular genotyping methods, i.e., IS6110 restriction fragment length polymorphism (RFLP), spoligotyping, and variablenumber tandem repeat (VNTR) analysis, have been successfully applied to investigate TB outbreaks, detect recent transmission, identify expansion of successful bacterial clones, and distinguish between endogenous reactivation and exogenous reinfection (29). Until now, several developed countries in Europe and North America have established databases with genotyping data from clinical isolates for tracking particular isolates of M. tuberculosis in local communities and detecting international transmission (9, 10, 41, 43). Because of the many advantages of VNTR genotyping compared to IS6110 RFLP (41), the newly proposed 15- and 24locus VNTR genotyping schemes have been introduced to study the molecular epidemiology in homogenous M. tuberculosis populations, where TB is caused mainly by transmission of a large number of genetically closely related strains (5, 18, 23, 25, 35). However, the ability of these methods to track transmission is limited in those settings, as many clustered strains are found to lack epidemiological links, which limits the identification of transmission chains (18, 23, 25, 35). This lack of epidemiological links can be due to limitations during contact-tracing investigations (29). On the other hand, the low discrimination power and the problem of homoplasy in molecular markers may lead to false classification of clustered strains which are distantly related and may involve historically remote (as opposed to ongoing) transmission chains (8, 18, 25, 35). China has one of the highest TB burdens in the world (48). In most Chinese regions, TB is caused mainly by Beijing strains (17, 25, 28, 44, 49, 50). The Beijing strains includes several divergent sublineages which can be defined based on large sequence poly-

0095-1137/12/$12.00

Journal of Clinical Microbiology

p. 633– 639

morphisms (LSPs) or single nucleotide polymorphisms (SNPs) (14, 20, 42, 46). However, the most frequent Beijing strains in China and elsewhere, with the exception of Japan and Korea, belong to a genetically homogenous population originally characterized by one or two IS6110 insertions in the noise transfer function (NTF) region of the Mycobacterium tuberculosis genome (4, 13, 19, 32). These strains have been referred to as the “modern” (or typical) Beijing branch (34), and Beijing strains without an IS6110 insertion in NTF region have been referred to as the “ancient” (or atypical) branch (34). The modern Beijing branch comprises a large group of genetically closely related strains which show high similarity in terms of IS6110 RFLP and VNTR profiles (19, 25, 33, 34, 44). As a consequence, the discriminatory power of the standard 15- and 24-locus VNTR genotyping schemes was repeatedly demonstrated to be low for epidemiology studies in settings dominated by modern Beijing strains (18, 25, 35). Modern Beijing strains are also characterized by SNPs in the mutT2 and ogt genes, and their population structure has been further explored by using more recently evolved LSPs (RD142 and RD150) and SNPs (11, 31, 42). However, the frequencies of these polymorphisms vary in different settings (13, 19, 24, 31). Compared to VNTRs, SNPs and LSPs have the advantage of

Received 20 August 2011 Returned for modification 28 September 2011 Accepted 21 December 2011 Published ahead of print 28 December 2011 Address correspondence to Qian Gao, [email protected]. Supplemental material for this article may be found at http://jcm.asm.org/. Copyright © 2012, American Society for Microbiology. All Rights Reserved. doi:10.1128/JCM.05539-11

jcm.asm.org

633

Luo et al.

TABLE 1 Distribution of 191 Beijing strains in each sublineage as defined by polymorphism of eight SNPs SNPa Beijing recR mutT4 recX mutT2 uvrD1 adhE2 ligD ogt No. (%) sublineage (codon 44) (codon 48) (codon 59) (codon 58) (codon 462) (codon 124) (codon 580) (codon 37) of strains Subgroup HGI (95% CI) Bmyc2 Bmyc4 Bmyc6 Bmyc25 Bmyc26 Bmyc10 Bmyc13 Bmyc210 a

W M M M M M M M

W W M M M M M M

W W W W M M M M

W W W W W M M M

W W W W W W M W

W W W W W W W M

M W W W W W W W

W W W M W W W W

9 (4.7) 7 (3.7) 2 (1.0) 31 (16.2) 19 (9.9) 73 (38.2) 20 (10.5) 30 (15.7)

Ancient Ancient Ancient Ancient Ancient Modern Modern Modern

0.944 (0.871–1) 0.952 (0.855–1) 0.981 (0.957–1) 0.959 (0.915–1) 0.975 (0.953–0.998) 0.937 (0.863–1) 0.968 (0.940–0.995)

W, wild type; M, mutant.

assigning a particular strain to certain lineage unambiguously. However, SNP and LSP typing is less informative in molecular epidemiology studies because of low discriminatory power (5). Here we combined SNPs and VNTRs to genotype Beijing strains in Shanghai. We showed that SNPs in modern Beijing strains can help to correct the clusters which were misclassified due to the limited discriminatory power and/or homoplasy of VNTRs. MATERIALS AND METHODS Study population and isolates. In Shanghai, all M. tuberculosis isolates from new pulmonary tuberculosis patients who were sputum smear positive and/or culture positive for M. tuberculosis were sent to the Shanghai Municipal Centers for Disease Control and Prevention (CDC) for mycobacterial identification and drug susceptibility testing. From January 2007 to June 2010, a total of 284 isolates were collected from the initial samples from pulmonary tuberculosis patients in the Chongming district. Two hundred sixty-seven isolates (94.0%; 267/284) were successfully identified as M. tuberculosis complex, among which 224 (83.9%; 224/267) were further identified as Beijing family strains by deletion-targeted multiplex PCR (DTM-PCR) in this study or previously (6, 26). One hundred ninetyone Beijing strains (85.3%; 191/224) that had enough DNA for VNTR and SNP genotyping were included in this study. Genotyping and computer-assisted analysis. The standard VNTR-15 genotyping was applied to all 191 Beijing strains (41). A further genotyping of three hypervariable VNTR loci (VNTR3820, VNTR3232, and VNTR4120) was performed for VNTR-15 defined clustered strains. Loci of VNTR-15 were amplified individually using the primers and conditions described previously (49). The three hypervariable loci were amplified in 1⫻ GC buffer I (Takara Biotech Co. Ltd., Dalian, China) with an extension time of 1.5 min at 72°C. The sizes of amplicons were analyzed on 1.0 or 1.5% agarose gels for 1.5 to 2 h at 150 V with 50-bp ladder and 100-bp high-ladder size standards (CoWin Biotech Co. Ltd., Beijing, China). Hypervariable alleles from clustered strains were electrophoresed in lanes close together on the same gel to minimize the possibility of false-positive discriminations caused by potential errors of misscoring within each cluster. For each gel, the corresponding VNTR allele of H37Rv was used for quality control. For SNP genotyping, eight SNPs in 3R genes were previously found in Beijing strains from Shanghai (31). Excluding the SNP in ogt codon 12, which is redundant with the SNP in mutT2 codon 58, the other seven SNPs were included for genotyping in this study. One extra SNP at position 2532616 (according to the genome of H37Rv) which was constantly found in modern Beijing strains was also included (13, 19, 24). In total, eight SNPs (Table 1) were genotyped with a novel real-time PCR melting curve assay which was developed previously by us for detecting drug-resistant mutations (27). According to previous studies, modern Beijing strains all carry a mutation in mutT2 codon 58 (19, 24). Thus, we defined strains with mutant mutT2 as modern Beijing and strains with wild-type mutT2 as ancient in this study. The Hunter-Gaston index (HGI)

634

jcm.asm.org

and 95% confidence interval (CI) of theVNTR-15 was calculated as described previously (16, 21) and was used to evaluate the population diversity within each sublineage. BioNumerics 5.0 was used to construct minimal spanning trees (MSTs) for all Beijing strains and for strains of each sublineage based onVNTR-15 with the rules described previously (46).

RESULTS

Population structure analysis based on SNPs. The 191 M. tuberculosis Beijing strains were divided into 8 sublineages by genotyping of the 8 SNPs (Table 1). Comparing these results with those of our previous study (31), one additional sublineage was characterized based on the mutation in adhE2 codon 124 (position 2532616 in the H37Rv genome). This new sublineage was named Bmyc210, as the corresponding mutation was first identified in strain 210 (14). Based on the mutation in mutT2 codon 58, sublineages Bmyc10, Bmyc13, and Bmyc210 were classified as being in the modern Beijing group and sublineages Bmyc2, Bmyc6, Bmyc4, Bmyc25, and Bmyc26 were classified as being in the ancient group (Table 1). The three sublineages of the modern Beijing branch constituted the majority (64.4%; 123/191) of Beijing strains in our population, which is in agreement with our previous study (31). Cluster and diversity analysis based on VNTR-15. SNP genotyping has the advantage of classifying Beijing strains into phylogenetically robust sublineages but lacks the discriminatory power for epidemiological study and further diversity analysis. To study the potential transmission clusters and population diversity of these Beijing strains, VNTR-15 was applied to all 191 isolates. The VNTR genotyping divided the 191 Beijing strains into 110 unique types and 27 clusters, with a clustering rate of 28.3% and a discriminatory index of 0.990 (95% CI, 0.984 to 0.996). To further analyze the diversity within sublineages, we calculated the HGI for each sublineage separately (Table 1). Except for sublineage Bmyc6, where the HGI was not applicable due to the small sample size, the other sublineages all showed high HGIs which indicate high diversity within each sublineage. Considering that high diversity results from long clonal expansion, this may indicate early times of divergence of these sublineages. Thus, strains from different sublineages, including the three modern sublineages, are more likely to be historically distantly related to each other. Congruence between VNTR and SNP. According to previous studies, MSTs of VNTR-15 are phylogenetically informative for Beijing strains (45, 46). To study the relationship of VNTR genotypes and SNP-defined sublineages in our population, we constructed the MST for all 191 Beijing strains based onVNTR-15. Sublineage information was mapped onto the network (Fig. 1).

Journal of Clinical Microbiology

Genotyping of a Homogenous M. tuberculosis Population

FIG 1 Minimal spanning tree of the 191 Beijing strains based on VNTR-15. Each circle corresponds to a certain VNTR genotype. Circle size corresponds to the number of isolates with a particular genotype. Circles are colored according to the SNP-based sublineages (Table 1). Five clusters with mixed sublineages are indicated. The two distinct branches are indicated by a dashed curve.

The MST grouped 176 of the 191 Beijing strains to a single, albeit quite diverse, complex, with the remaining 15 isolates (13 genotypes) distributed to two small minor complexes or occupying unique positions. At a low resolution, the concordance between the MST and SNP classifications was quite good, as 120 of 121 strains from the left branch of the major complex belonged to the modern Beijing group and 54 of 55 strains from the right branch belonged to the ancient group (Fig. 1). However, the concordance was found to be relatively low at a higher resolution. The MST consistently clustered strains from different sublineages and failed to classify strains of each sublineage to a specific MST branch. Furthermore, five VNTR-15-based clusters were found to contain strains from more than one sublineage (Fig. 1). This high discordance between MST and SNP sublineages indicates a high level of homoplasy in VNTR-15 genotypes. Composite tree based on SNP and VNTR. Because of the high rate of homoplasy, the population modeling of 191 Beijing strains based on VNTR-15 was less meaningful than previous studies had shown for ancient Beijing strains (45, 46). In order to further study the evolution and diversity of each sublineage, we constructed a composite tree, in which the MST of each sublineage was combined with an SNP phylogeny drawn according to our previous study (31). After exclusion of sublineages Bmyc4 and

March 2012 Volume 50 Number 3

Bmyc6 due to small sample sizes, each MST of the other six sublineages was represented as a star-like network, which indicated the homogenous populations and more recent expansions in each sublineage. The populations of the five ancient sublineages might have been founded by distinct VNTR genotypes, and no VNTR genotype shared between them was found. In contrast, the three modern sublineages shared several VNTR genotypes, which also confirmed their close genetic relationships. The shared VNTR genotypes might be caused by the same founder genotype in different sublineages or by VNTR homoplasy. Sublineages Bmyc10 and Bmyc13 were more likely to originate from the same VNTR-15 genotype of cluster 1 (Fig. 2). Thus, VNTR-15 failed to differentiate these historically distantly related strains and falsely classified them as a cluster (Fig. 1). The potential parallel evolutions of this founder VNTR genotype might lead to another three genotypes shared between Bmyc10 and Bmyc13, which resulted in misclassified clusters 2, 4, and 5. The potential founder genotypes of Bmyc10 and Bmyc210 were different (VNTR genotypes of clusters 1 and 2). Thus, the genotype of cluster 3 shared between two sublineages might be the result of convergent evolution. Genotyping of hypervariable VNTR loci. Previous studies have shown that hypervariable loci are helpful in differentiating epidemically unrelated strains (23, 34, 35). In order to study the

jcm.asm.org 635

Luo et al.

FIG 2 A composite tree combining SNP phylogeny and MSTs of each sublineage. The genotypes of the five mixed clusters in Fig. 1 are labeled with different colors.

usefulness of these loci for further differentiation of strains of the 27 VNTR-15 based clusters, three hypervariable loci (VNTR3820, VNTR4120, and VNTR3232) were genotyped in these 81 isolates. Three strains showed two bands at locus VNTR4120 or VNTR3232 (Fig. 3; see the supplemental material), which may indicate the coexistence of a clonal variant according to a previous study (23). For the five clusters with mixed sublineages, most of the strains within each cluster showed distinct genotypes in the three hypervariable loci, and only 8 of these 30 strains (26.7%) were further grouped into four new clusters (Fig. 3; see the supplemental material). Three of the four clusters contained only strains with a unique sublineage; however, the other one still contained two strains from different sublineages (Bmyc210 and Bmyc13). For the other 22 VNTR-15-defined clusters with consistent sublineages, the alleles of three hypervariable loci were much more conserved within clusters, and 74.5% (38/51) of these strains were further classified into 18 clusters (see the supplemental material). DISCUSSION

The newly proposed 15- and 24-locus VNTR genotyping methods have been shown to be appropriate for studying the epidemiology of TB in developed countries with low TB incidences and where recent transmission of TB is largely prevented by efficient TB control programs (1, 3, 38, 41). However, these tools were found to be less reliable for tracking transmission in countries with high TB burdens, especially in settings where the TB epidemic was dominated by Beijing family strains (18, 23, 25, 35). Among the six main M. tuberculosis lineages, the Beijing family was thought be the most successful and has undergone a large population expansion in the few recent centuries (47). The recent worldwide expansions of Beijing strains were caused mainly by the modern Beijing sublineage, which led to the homogenous M. tuberculosis populations in East Asia, Eurasia, and South Africa (19, 33, 44). Two

636

jcm.asm.org

major inherent characteristics of VNTR-15 typing may lead to limitations for epidemiological studies in areas where M. tuberculosis populations are homogenous. First, the discriminatory power of VNTR-15 is relatively low in those populations. Although the concordance between VNTR genotyping and contact tracing data was higher than that with IS6110 RFLP typing in some settings (1, 38), previous studies from Japan, South Africa, and China found that many 15- or 24-locus VNTR-defined clusters could be further differentiated by IS6110 RFLP typing (18, 23, 25). In this study, we found that some historically distantly related strains from sublineages Bmyc10 and Bmyc13 were more likely to have kept their original founder VNTR-15 profile after their divergence, which highlights the limitations of the low discrimination power of VNTR-15 for epidemiology studies in such populations. The next limitation is the homoplasy of VNTRs. High levels of homoplasy at individual VNTR loci have been previously reported (8). However, the combination of multiple VNTR loci can partially compensate for the homoplasy of an individual locus, as 15- and 24locus VNTR profiles have been shown to be predictive of the M. tuberculosis lineages and ancient sublineages among the Beijing family (2, 5, 41, 45, 46). In this study, the VNTR-15-based MST successfully differentiated the modern and ancient branches of the Beijing family. However, VNTR-15 failed to further differentiate Beijing strains at the sublineage level. The success of using 15- and 24-locus VNTR to predict genetically distantly related M. tuberculosis lineages could be explained by the distinct VNTR profiles of populations of different lineages. In this case, VNTR homoplasy between lineages will only have small effects on MST-based population modeling to differentiate them (30, 47). However, at sublineage levels, especially for closely related sublineages whose VNTR profiles are highly similar, homoplasy at a single VNTR locus would possibly lead to a cluster of strains from different sublineages determined by MSTs. Our results indicated that VNTR homoplasy would also lead to a nonnegligible effect on

Journal of Clinical Microbiology

Genotyping of a Homogenous M. tuberculosis Population

FIG 3 VNTR-15-based dendrogram and 3-hypervariable-VNTR-locus profiles of the five clusters with mixed sublineages. The boxes in the sublineage section indicate the five clusters (numbered as in Fig. 1); boxes in hypervariable locus section indicate clustered isolates defined by both VNTR-15 and 3 hypervariable VNTR loci.

epidemiology studies, as it consistently caused misclassifications of historically distantly related strains as clusters. The addition of hypervariable loci was helpful for both enhancing discriminative power and decreasing the homoplasy rate of VNTR-15 in this study. Due to the clonal instability and large alleles, hypervariable loci were excluded for standard VNTR genotyping (41). However, a recent study indicated that these loci were not extraordinarily unstable compared to other loci and IS6110 RFLP patterns (23). Furthermore, several studies in Japan and countries of the former Soviet Union indicated the necessity to include these loci to increase the discriminatory power and differentiate epidemically unrelated Beijing strains which appeared to be clustered by standard 15- or 24-locus VNTR typing (23, 35, 36). In our study, the relatively conserved profiles of hypervariable alleles within VNTR-15- and SNP-based clusters may also prove their clonal stabilities during transmissions. In contrast, the high variations of hypervariable loci in five VNTR-15-based clusters

March 2012 Volume 50 Number 3

with mixed sublineages further indicated that there might be no epidemic relevance between most of these strains. Hypervariable loci were thought to be at high individual homoplasy levels (12). However, the inclusion of three hypervariable loci was helpful to decrease the homoplasy rate in the modern Beijing populations, as the five VNTR-15-based clusters with mixed sublineages could be further classified as unique strains or smaller clusters comprising single SNP sublineage. Similar results have been obtained in a more recent study, in which a higher mutation rate of a 24-locus VNTR data set was found to efficiently reduce the homoplasy level by modeling a homogenous M. tuberculosis population for 100 years (39). Although homoplasy could be largely reduced by inclusion of hypervariable loci, it still caused problems for epidemiology studies in homogenous populations, as one cluster defined by both VNTR-15 and three hypervariable loci still contained two distantly related strains from two sublineages in our study. Tracing transmissions of M. tuberculosis in countries with high

jcm.asm.org 637

Luo et al.

TB burdens is challenging. As shown in this and previous studies, the homogenous M. tuberculosis population in those settings usually led to a relatively low discrimination power of traditional genotyping methods and to VNTR homoplasy (18, 35). One way to address this challenge is whole-genome sequencing (WGS), which provides the highest discrimination power and where homoplasy is not an issue (7). In recent years, with the development of high-throughput next-generation sequencing technology, several studies have applied WGS to reconstruct transmission chains between VNTR-defined cluster cases (15, 22, 40), and some clustered strains defined by traditional molecular genotyping methods were identified as distantly related isolates (37). However, because of the relatively high cost and the complexity of data analysis, the application of this technology to study the molecular epidemiology of M.tuberculosis will be limited for some time, especially in developing countries. An alternative way to address this challenge is to combine genotyping of robust phylogenetic markers such as SNPs with traditional genotyping methods. SNPs and other robust polymorphism markers were traditionally used to differentiate genetically distantly related M. tuberculosis lineages which could also be generally classified by 15- or 24-locus VNTR (5, 8, 20). Thus, these phylogenetically robust markers were thought to be less important for epidemic studies. Here we showed, for the first time, that SNPs could help to correct the clusters that were misclassified due to the low discriminatory power and homoplasy of VNTR-15. We also showed the potential of hypervariable VNTR loci to further differentiate distantly related strains clustered by VNTR-15. Thus, we recommended the standard VNTR-15 and SNPs as first-line typing methods for large-scale typing and the hypervariable loci for second-line typing of clustered strains following use of the VNTR-15- and SNPbased scheme in countries with high TB burdens. Although very few SNPs are presently available (14, 31), with the tremendous increase of whole-genome sequencing data, we believe that an increasing number of informative markers will be found for studying the molecular epidemiology of TB. ACKNOWLEDGMENTS This work was supported by the Key Project of Chinese National Programs (2012ZX10003004-006), the International Cooperation Project of the Ministry of Science and Technology (2010DFA34440), and the Key Project of Science and Technology, Commission of Shanghai Municipality (10JC1413700 and 10411955000). Tao Luo is a trainee of the National Institutes of Health (grant D43 TW007887).

REFERENCES 1. Allix-Beguec C, Fauville-Dufaux M, Supply P. 2008. Three-year population-based evaluation of standardized mycobacterial interspersed repetitive-unit-variable-number tandem-repeat typing of Mycobacterium tuberculosis. J. Clin. Microbiol. 46:1398 –1406. 2. Allix-Beguec C, Harmsen D, Weniger T, Supply P, Niemann S. 2008. Evaluation and strategy for use of MIRU-VNTRplus, a multifunctional database for online analysis of genotyping data and phylogenetic identification of Mycobacterium tuberculosis complex isolates. J. Clin. Microbiol. 46:2692–2699. 3. Allix-Beguec C, Supply P, Wanlin M, Bifani P, Fauville-Dufaux M. 2008. Standardised PCR-based molecular epidemiology of tuberculosis. Eur. Respir. J. 31:1077–1084. 4. Bifani PJ, Mathema B, Kurepina NE, Kreiswirth BN. 2002. Global dissemination of the Mycobacterium tuberculosis W-Beijing family strains. Trends Microbiol. 10:45–52. 5. Cardoso Oelemann M, et al. 2011. The forest behind the tree: phylogenetic exploration of a dominant Mycobacterium tuberculosis strain lineage from a high tuberculosis burden country. PLoS One 6:e18256.

638

jcm.asm.org

6. Chen J, et al. 2007. Deletion-targeted multiplex PCR (DTM-PCR) for identification of Beijing/W genotypes of Mycobacterium tuberculosis. Tuberculosis (Edinb.) 87:446 – 449. 7. Comas I, et al. 2010. Human T cell epitopes of Mycobacterium tuberculosis are evolutionarily hyperconserved. Nat. Genet. 42:498 –503. 8. Comas I, Homolka S, Niemann S, Gagneux S. 2009. Genotyping of genetically monomorphic bacteria: DNA sequencing in Mycobacterium tuberculosis highlights the limitations of current methodologies. PLoS One 4:e7815. 9. Cowan LS, Crawford JT. 2002. Genotype analysis of Mycobacterium tuberculosis isolates from a sentinel surveillance population. Emerg. Infect. Dis. 8:1294 –1302. 10. Devaux I, Kremer K, Heersma H, Van Soolingen D. 2009. Clusters of multidrug-resistant Mycobacterium tuberculosis cases, Europe. Emerg. Infect. Dis. 15:1052–1060. 11. Ebrahimi-Rad M, et al. 2003. Mutations in putative mutator genes of Mycobacterium tuberculosis strains of the W-Beijing family. Emerg. Infect. Dis. 9:838 – 845. 12. Estoup A, Jarne P, Cornuet JM. 2002. Homoplasy and mutation model at microsatellite loci and their consequences for population genetics analysis. Mol. Ecol. 11:1591–1604. 13. Faksri K, et al. 2011. Genetic diversity of the Mycobacterium tuberculosis Beijing family based on IS6110, SNP, LSP and VNTR profiles from Thailand. Infect. Genet. Evol. 11:1142–1149. 14. Filliol I, et al. 2006. Global phylogeny of Mycobacterium tuberculosis based on single nucleotide polymorphism (SNP) analysis: insights into tuberculosis evolution, phylogenetic accuracy of other DNA fingerprinting systems, and recommendations for a minimal standard SNP set. J. Bacteriol. 188:759 –772. 15. Gardy JL, et al. 2011. Whole-genome sequencing and social-network analysis of a tuberculosis outbreak. N. Engl. J. Med. 364:730 –739. 16. Grundmann H, Hori S, Tanner G. 2001. Determining confidence intervals when measuring genetic diversity and the discriminatory abilities of typing methods for microorganisms. J. Clin. Microbiol. 39:4190 – 4192. 17. Han H, et al. 2007. Utility of mycobacterial interspersed repetitive unit typing for differentiating Mycobacterium tuberculosis isolates in Wuhan, China. J. Med. Microbiol. 56:1219 –1223. 18. Hanekom M, et al. 2008. Discordance between mycobacterial interspersed repetitive-unit-variable-number tandem-repeat typing and IS6110 restriction fragment length polymorphism genotyping for analysis of Mycobacterium tuberculosis Beijing strains in a setting of high incidence of tuberculosis. J. Clin. Microbiol. 46:3338 –3345. 19. Hanekom M, et al. 2007. A recently evolved sublineage of the Mycobacterium tuberculosis Beijing strain family is associated with an increased ability to spread and cause disease. J. Clin. Microbiol. 45:1483–1490. 20. Hershberg R, et al. 2008. High functional diversity in Mycobacterium tuberculosis driven by genetic drift and human demography. PLoS Biol. 6:e311. 21. Hunter PR, Gaston MA. 1988. Numerical index of the discriminatory ability of typing systems: an application of Simpson’s index of diversity. J. Clin. Microbiol. 26:2465–2466. 22. Ioerger TR, et al. 2009. Genome analysis of multi- and extensively-drugresistant tuberculosis from KwaZulu-Natal, South Africa. PLoS One 4:e7778. 23. Iwamoto T, et al. 2007. Hypervariable loci that enhance the discriminatory ability of newly proposed 15- and 24-loci variable-number tandem repeat typing method on Mycobacterium tuberculosis strains predominated by the Beijing family. FEMS Microbiol. Lett. 270:67–74. 24. Iwamoto T, Yoshida S, Suzuki K, Wada T. 2008. Population structure analysis of the Mycobacterium tuberculosis Beijing family indicates an association between certain sublineages and multidrug resistance. Antimicrob. Agents Chemother. 52:3805–3809. 25. Jiao WW, et al. 2008. Evaluation of new variable-number tandem-repeat systems for typing Mycobacterium tuberculosis with Beijing genotype isolates from Beijing, China. J. Clin. Microbiol. 46:1045–1049. 26. Li X, et al. 2011. Non-Beijing strains of Mycobacterium tuberculosis in China. J. Clin. Microbiol. 49:392–395. 27. Luo T, et al. 2011. A multiplex real-time PCR melting curve assay to detect drug resistant mutations of Mycobacterium tuberculosis. J. Clin. Microbiol. 49:3132–3138. 28. Ma X, et al. 2006. rpoB gene mutations and molecular characterization of rifampin-resistant Mycobacterium tuberculosis isolates from Shandong Province, China. J. Clin. Microbiol. 44:3409 –3412.

Journal of Clinical Microbiology

Genotyping of a Homogenous M. tuberculosis Population

29. Mathema B, Kurepina NE, Bifani PJ, Kreiswirth BN. 2006. Molecular epidemiology of tuberculosis: current insights. Clin. Microbiol. Rev. 19: 658 – 685. 30. McEvoy CR, van Helden PD, Warren RM, Gey van Pittius NC. 2009. Evidence for a rapid rate of molecular evolution at the hypervariable and immunogenic Mycobacterium tuberculosis PPE38 gene region. BMC Evol. Biol. 9:237. 31. Mestre O, et al. 2011. Phylogeny of Mycobacterium tuberculosis Beijing strains constructed from polymorphisms in genes involved in DNA replication, recombination and repair. PLoS One 6:e16020. 32. Mokrousov I, et al. 2005. Origin and primary dispersal of the Mycobacterium tuberculosis Beijing genotype: clues from human phylogeography. Genome Res. 15:1357–1364. 33. Mokrousov I, et al. 2004. Analysis of the allelic diversity of the mycobacterial interspersed repetitive units in Mycobacterium tuberculosis strains of the Beijing family: practical implications and evolutionary considerations. J. Clin. Microbiol. 42:2438 –2444. 34. Mokrousov I, et al. 2002. Phylogenetic reconstruction within Mycobacterium tuberculosis Beijing genotype in northwestern Russia. Res. Microbiol. 153:629 – 637. 35. Mokrousov I, et al. 2008. Mycobacterium tuberculosis Beijing genotype in Russia: in search of informative variable-number tandem-repeat loci. J. Clin. Microbiol. 46:3576 –3584. 36. Mokrousov I, et al. 2009. Penitentiary population of Mycobacterium tuberculosis in Kyrgyzstan: exceptionally high prevalence of the Beijing genotype and its Russia-specific subtype. Infect. Genet. Evol. 9:1400 – 1405. 37. Niemann S, et al. 2009. Genomic diversity among drug sensitive and multidrug resistant isolates of Mycobacterium tuberculosis with identical DNA fingerprints. PLoS One 4:e7407. 38. Oelemann MC, et al. 2007. 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. 45:691– 697.

March 2012 Volume 50 Number 3

39. Reyes JF, Chan CH, Tanaka MM. 30 May 2011. Impact of homoplasy on variable numbers of tandem repeats and spoligotypes in Mycobacterium tuberculosis. Infect. Genet. Evol. doi:10.1016/j.meegid.2011.05.018. 40. Schurch AC, et al. 2010. High-resolution typing by integration of genome sequencing data in a large tuberculosis cluster. J. Clin. Microbiol. 48:3403– 3406. 41. Supply P, et al. 2006. Proposal for standardization of optimized mycobacterial interspersed repetitive unit-variable-number tandem repeat typing of Mycobacterium tuberculosis. J. Clin. Microbiol. 44:4498 – 4510. 42. Tsolaki AG, et al. 2005. Genomic deletions classify the Beijing/W strains as a distinct genetic lineage of Mycobacterium tuberculosis. J. Clin. Microbiol. 43:3185–3191. 43. van Embden JD, et al. 1993. Strain identification of Mycobacterium tuberculosis by DNA fingerprinting: recommendations for a standardized methodology. J. Clin. Microbiol. 31:406 – 409. 44. van Soolingen D, et al. 1995. Predominance of a single genotype of Mycobacterium tuberculosis in countries of east Asia. J. Clin. Microbiol. 33:3234 –3238. 45. Wada T, Iwamoto T. 2009. Allelic diversity of variable number of tandem repeats provides phylogenetic clues regarding the Mycobacterium tuberculosis Beijing family. Infect. Genet. Evol. 9:921–926. 46. Wada T, Iwamoto T, Maeda S. 2009. Genetic diversity of the Mycobacterium tuberculosis Beijing family in East Asia revealed through refined population structure analysis. FEMS Microbiol. Lett. 291:35– 43. 47. Wirth T, et al. 2008. Origin, spread and demography of the Mycobacterium tuberculosis complex. PLoS Pathog. 4:e1000160. 48. World Health Organization. 2010. Global tuberculosis control 2010, p 7. WHO Press, Geneva, Switzerland. 49. Zhang L, et al. 2008. Highly polymorphic variable-number tandem repeats loci for differentiating Beijing genotype strains of Mycobacterium tuberculosis in Shanghai, China. FEMS Microbiol. Lett. 282:22–31. 50. Zhang M, et al. 2005. Detection of mutations associated with isoniazid resistance in Mycobacterium tuberculosis isolates from China. J. Clin. Microbiol. 43:5477–5482.

jcm.asm.org 639