Infection, Genetics and Evolution 10 (2010) 459–466

Single nucleotide polymorphisms in cell wall biosynthesis-associated genesand phylogeny of Mycobacterium tuberculosis lineages§

Pei-Chun Chuang a,b, Yi-Ming A. Chen c, Huang-Yau Chen a, Ruwen Jou a,*a Reference Laboratory of Mycobacteriology, Research and Diagnostic Center, Centers for Disease Control, 161 Kun-Yang Street, Nan-Kang, Taipei 115, Taiwan, ROCb Institute of Public Health, National Yang-Ming University, 155, Sec 2, Linong Street, Taipei 112, Taiwan, ROCc AIDS Prevention and Research Center and Institute of Microbiology and Immunology, National Yang-Ming University, 155, Sec 2, Linong Street, Taipei 112, Taiwan, ROC

A R T I C L E I N F O

Article history:

Received 5 November 2009

Received in revised form 1 March 2010

Accepted 2 March 2010

Available online 16 March 2010

Keywords:

Mycobacterium tuberculosis

Lineages

Single nucleotide polymorphisms (SNPs)

Cell wall biosynthesis-associated genes

Antigen 85 complex

Mannosyltransferase

A B S T R A C T

To investigate specific single nucleotide polymorphisms (SNPs) of different lineages of Mycobacterium

tuberculosis, cell wall biosynthesis-associated genes encoding antigen 85 complex (fbpA, fbpB, and fbpC)

and mannosyltransferase (pimB) were analyzed. Genetically diversified and predominant M. tuberculosis

and Mycobacterium bovis genotypes identified in Taiwan (26 Beijing and 44 non-Beijing) were included

in the study. Sequence analyses revealed that nine novel SNPs were found within main lineages of M.

tuberculosis complex, including East-African-Indian (EAI), Beijing, Central-Asian (CAS), Bovis, and one

lineage containing Latin American and Mediterranean (LAM), Haarlem and T. Specifically, a SNP in pimB

codon 270 was identified in EAI, fbpA codon 156 in ancestral Beijing, fbpB codon 238 in modern Beijing,

fbpA codon 4 and fbpC codon 158 in CAS, fbpA codon 311 in M. bovis and an additional SNP in fbpB codon

140 in M. bovis-BCG, and pimB codon 107 in other spoligotypes lineages including an additional SNP in

fbpC codon 103 in LAM. In addition, we proved that isolates with spoligotype shared type (ST) no. 523

(carrying all 43 spacers), designated as unknown lineage in an international spoligotyping database

(SpolDB4), belong to an early ancestral Beijing sublineage. Accordingly, a phylogenetic tree was

constructed using those SNPs, and an evolutionary hypothesis for lineages of M. tuberculosis was

proposed. These novel lineage-specific SNPs will be informative genetic markers in molecular

epidemiological and evolutionary studies of M. tuberculosis.

� 2010 Elsevier B.V. All rights reserved.

Contents lists available at ScienceDirect

Infection, Genetics and Evolution

journal homepage: www.elsev ier .com/ locate /meegid

1. Introduction

Mycobacterium tuberculosis with relatively low genetic varia-tions is considered as a monomorphic pathogen as compared toother pathogenic bacteria (Achtman, 2008). Strains of M. tubercu-

losis seen today appear to have a relatively most recent commonancestor (MRCA) (Smith et al., 2009). Molecular and bioinformatictechniques have been applied to study genetic characteristics of M.

tuberculosis. Before the whole genome of M. tuberculosis wassequenced in 1998, Sreevatsan et al. (1997) analyzed 26 structuralgenes and revealed that M. tuberculosis could be grouped into threeprinciple genetic groups (PGGs), PGG1, PGG2, and PGG3 usingsingle nucleotide polymorphisms (SNPs) of two drug resistanceassociated genes, katG codon 463 and gyrA codon 95. PGG1 is

§ Nucleotide sequence data reported in this paper are available in the GenBank,

EMBL and DDBJ databases under the accession numbers GQ150313 to GQ150315

for fbpA gene, GQ150316 to GQ150317 for fbpB gene, GQ150318 to GQ150320 for

fbpC gene, and GQ150321 to GQ150324 for pimB gene.

* Corresponding author. Tel.: +886 2 26531370; fax: +886 2 26531387.

E-mail addresses: peichun@cdc.gov.tw (P.-C. Chuang), arthur@ym.edu.tw

(Y.A. Chen), hopeseed427@yahoo.com.tw (H.-Y. Chen), rwj@cdc.gov.tw (R. Jou).

1567-1348/$ – see front matter � 2010 Elsevier B.V. All rights reserved.

doi:10.1016/j.meegid.2010.03.003

considered to be evolutionarily more ancestral (Sreevatsan et al.,1997). However, classification of PGGs was derived from a highlybiased strain collection causing grouping of diversified M.

tuberculosis complex lineages into an ‘‘artificial’’ group, PGG1(Comas et al., 2009). Strains possessing region of M. tuberculosis

specific deletion 1 (TbD1) were considered as ‘‘ancient’’ while‘‘modern’’ ones had the deletion of TbD1 region (Brosch et al.,2002). Furthermore, SNPs and large sequence polymorphisms(LSPs) identified by comparative genomics in M. tuberculosis andMycobacterium bovis were applied for phylogeny analyses(Gutacker et al., 2002, 2006; Baker et al., 2004; Tsolaki et al.,2005; Filliol et al., 2006; Gagneux et al., 2006). Lineages deduced inseveral studies were concordant (Gagneux and Small, 2007).Analyses using genetic markers of SNPs, mycobacterial inter-spersed repetitive units (MIRU) and spacer oligonucleotide typing(spoligotyping) have addressed the sequential evolutionary eventsof M. tuberculosis. Therefore, evolutionary history of M. tuberculosis

was proposed based on the characteristics of genotypes andgenetic markers (Arnold, 2007). Hershberg et al. (2008) proposed aphylogeny of M. tuberculosis complex by sequencing 89 genes(approximately seven megabase pairs) of 108 representativestrains. Based on more than 300 SNPs, M. tuberculosis complex

P.-C. Chuang et al. / Infection, Genetics and Evolution 10 (2010) 459–466460

can be phylogenetic grouped into six lineages (Lineage 1–6) and oneanimal-adapted lineage. This SNP-based phylogeny was concordantwith that deduced from LSPs as well as presence and absence ofTbD1. Besides, spoligotype groupings also correlated with SNP-based lineages. For example, East-African-Indian (EAI) was corre-sponding to Lineage 1, Central-Asian (CAS) to Lineage 3, and M.

africanum AFR1 and AFR2 to Lineage 6 and 5, respectively. However,spoligotype families can be just one of the sublineages within aLineage such as Beijing family is part of Lineage 2, and severalspoligotype families (Haarlem, Latin American and Mediterranean(LAM), X, Cameroon, and Uganda etc.) were sublineages withinLineage 4 (Comas et al., 2009). Nevertheless, compared tospoligotyping and MIRU, DNA sequencing resulted phylogeny treeshowed no homoplasy (Comas et al., 2009).

According to the international spoligotyping database(SpolDB4), the phylogeography of EAI, Beijing, and CAS lineagesshowed they were more prevalent (over 60%) in Asia regionsincluding Middle East, Central Asia, and Far-East Asia (Brudey et al.,2006). In India, these three ancestral lineages were predominantand indicated that India is a relatively ancient endemic focus oftuberculosis (Brosch et al., 2002; Gutierrez et al., 2006). In Taiwan,Beijing lineage was the most prevalent and pathogenic genotypefollowed by Haarlem and EAI lineages, while CAS lineage wasrarely found (Jou et al., 2005; unpublished data). Previous studieshave revealed specific SNPs in isolates possessing particularspoligotypes. SNPs in DNA repair genes, mutT2, mutT4, and ogt,were identified in Beijing genotypes suggesting that sequentialacquisition of the SNPs might elucidate evolution of Beijingsublineages (Rad et al., 2003). Whereas SNPs in DNA repair genes,including ogt gene codon 15, ung gene codon 167, and the non-synonymous SNP in mgtC codon 182, have been recognized asHaarlem genotype-specific SNPs (Alix et al., 2006; Olano et al.,2007). Therefore, those SNPs were suggested as specific geneticmarkers for distinct spoligotypes of M. tuberculosis.

The pathogenicity of M. tuberculosis was suggested to be derivedfrom the multifactorial and sequential accumulation of mutations infunctional genes. Virulence-associated genes in M. tuberculosis mayplay important roles in host interaction and immune response.Those genes included those involved in cell secretion proteins, cellwall components, enzymes associated with cellular metabolism,and transcriptional regulators (Smith, 2003). The cell wall of M.

tuberculosis contains numerous insoluble and non-covalentlyassociated glycolipids and proteins that are presumed to be requiredfor pathogenic viability during infection (Glickman and Jacobs,2001). The major cell wall components include mycolylarabinogalactan–peptidoglycan complex and interspersed phos-phatidylinositol mannosides (PIMs), lipomannan (LM), and lipoar-abinomannan (Dover et al., 2007). Interestingly, after screeningsequences of several known virulence genes involved in biosynth-esis of cell wall components, we identified SNPs in genes of theantigen 85 complex (fbpA, fbpB, fbpC), and pimB. The antigen 85complex, a family of fibronectin-binding proteins, has mycolyl-transferase activity and contributes to cell wall biosynthesis.Immunogenicity and protective efficiency of antigen 85 complexproteins were also illustrated (Wiker and Harboe, 1992; Horwitzet al., 1995). According to the whole genome sequence of M.

tuberculosis, the locations of fbpA, fbpB, and fbpC of antigen 85complex were separated, and the genetics of their promoter regionswas diverse (Cole et al., 1998). Their mycolyltransferase activitiesand the proportion of secreted proteins were distinct (Harth et al.,1996; Puech et al., 2002). Copenhaver et al. (2004) showed that themutant strain lacking fbpA was attenuated, and the virulence couldnot to be compensated by fbpB or fbpC. Whereas lipoarabinomannansynthesized from PIMs and LM has been implicated in the down-regulation of cell-mediated immunity (Moreno et al., 1988; Chanet al., 1991). Several mannosyltransferases and glycosyltransferases

that participate in the biosynthesis of lipoarabinomannan have beenidentified (Alderwick et al., 2007). The PimB, encoded by pimB, withknown mannosyltransferase activity catalyzes the formation ofPIM2 in the early stage of biosynthesis (Schaeffer et al., 1999).Inactivation of pimB gene in M. tuberculosis resulted in decreasedlipoarabinomannan and LM content, altered macrophage phenotypeand increased macrophage death rate (Torrelles et al., 2009).Therefore, we investigated associations between SNPs of fbpA, fbpB,fbpC, and pimB and different lineages of M. tuberculosis, and theimplication for putative phylogeny.

2. Materials and methods

2.1. Study population

Predominant genotypes of M. tuberculosis were identifiedfrom the genomic database, comprising 5949 isolates referring to649 spoligotypes, established by the Taiwan Centers forDisease Control. Nationwide survey indicated that the majority(80.0%) of the spoligotype lineages were Beijing, Haarlem andHaarlem-like, EAI, T, and LAM (unpublished data). In addition,CAS and Bovis, were included in this study. M. africanum isolatewas not yet identified in Taiwan. To ensure isolates with abreadth of genetic diversities, 70 representative genotypes,including the lineages/sublineages described above, wereselected based on the diversities of the IS6110-restrictionfragment length polymorphism (RFLP) patterns and spoligotypesfor SNP analysis.

2.2. Genotype analysis

Spoligotyping using a commercially available kit (IsogenBioscience BV, Maarssen, The Netherlands) was describedpreviously (Kamerbeek et al., 1997). Only isolates that hybridizedto all of the last nine spacer oligonucleotides (spacers 35–43) weredefined as the characteristic Beijing genotype, whereas isolatesthat hybridized to only some of the last nine spacers were definedas the Beijing-like genotype. Beijing family genotypes includeboth Beijing and Beijing-like genotypes (Sebban et al., 2002;Kremer et al., 2004). Non-Beijing family isolates were designatedas such compared to those included in the SpolDB4 database(Brudey et al., 2006). IS6110-RFLP was performed following thestandardized protocol (van Embden et al., 1993). Spoligotypes andRFLP patterns were scanned and analyzed with Bionumerics1

software (version 4.601; Applied Maths, Kortrijk, Belgium). Thecluster and clade analysis was carried out using the unweightedpair group method with the arithmetic averages algorithm(UPGMA) and the Dice coefficient, according to the manufac-turer’s instructions. A clade was defined as a group of RFLPpatterns with at least 85% similarity.

2.3. Sequence analysis

SNPs in the fbpA, fbpB, fbpC, and pimB genes were analyzed bypolymerase chain reaction (PCR) amplification and sequenced withthe respective oligonucleotide primers designed by us (Table 1).Primers were designed using free FastPCR software (Kalendar et al.,2009). Forward and reverse primers for PCR and sequencing werelocated in the 50 and 30 untranslated regions of the genes. The fulllength (100%) of genes coding regions were covered by sequencinganalysis. The PCR reactions for the four genes were performed asfollows: an initial denaturing step of 95 8C for 10 min, 25 cycles of95 8C for 1 min, 69 8C (for fbpA), 64 8C (for fbpB), 56 8C (for fbpC), or62 8C (for pimB) for 1 min, and 72 8C for 2 min, and a final extensionstep of 72 8C for 7 min. The PCR products were sequenced by an ABI3730 automated sequencer (Applied Biosystems, USA) using

Table 1Primers used in PCR reactions and sequencing.

Gene Locus tag PCR product (bp) Forward PCR primer/internal sequencing primer Reverse PCR primer/internal sequencing primer

Name Sequence Name Sequence

pimB Rv0557 1335 Forward 50-TGAAAATTCGGCGTCTGCAA-30 Reverse 50-CAACACGGTATTGGTCAGGTC-30

pimB-F545 50-GGGTGGACGTGCAACGTTTCG-30 pimB-R675 50-CGTGAGCCGGTCGACATGCTT-30

fbpA Rv3804c 1214 Forward 50-CCCAGCGCCTGCAGTCT-30 Reverse 50-GCCCGTCAGGTAGCTAACCA-30

fbpA-F353 50-AGTCAAGCTTCTACTCCGACTGG-30 fbpA-R620 50-ATCAGGGTGGGACCCATCGCCT-30

fbpB Rv1886c 1078 Forward 50-ACACGGTATGTTCTGGCC-30 Reverse 50-GTATATCTCACGTGGACG-30

fbpB-F513 50-GGCAATGATCTTGGCCGCCTACC-30 fbpB-R672 50-ACTCGAGGGACCCCACATGTC-30

fbpC Rv0129c 1122 Forward 50-CGTCCTCGGTGTGTTTCACTC-30 Reverse 50-CCGATGCTGGCTTGCTGG-30

fbpC-F540 50-CGCGTACTACCCGCAGCAGTT-30 fbpC-R728 50-CGGGGAATCTGAACCATTGGGTCG-30

P.-C. Chuang et al. / Infection, Genetics and Evolution 10 (2010) 459–466 461

standardized conditions. Data were analyzed using SequencingAnalysis 5.2.0 software (Applied Biosystems, USA), and the genomeof H37Rv was used as a reference (GenBank accession no.NC_000962) (Cole et al., 1998) for comparison.

2.4. NTF region analysis

The NTF region is genetic marker for Beijing strains.Unidirectional evolution of NTF region in Beijing family ofM. tuberculosis was addressed by Mokrousov et al. (2005).The ancestral (atypical) and modern (typical) Beijing sublineageswere defined by the absence or presence of IS6110 insertion(s) inthe NTF region, respectively (Mokrousov et al., 2002, 2005).NTF1 (50-CCAGATATCGGGTGTGTCGAC-30) and INS2 (50-GCGT-AGGCGTCGGTGACAAA-30) primers were used (Jiao et al., 2007).The PCR reaction was performed as follows: an initial denaturingstep of 95 8C for 10 min, 35 cycles of 95 8C for 1 min, 64 8C for1 min, and 72 8C for 2 min, and a final extension step of 72 8C for

Table 2Genetic characteristics of 70 Mycobacterium tuberculosis and Mycobacterium bovis isola

Spoligotype

lineage

Shared

type no.

Octal value No. of RFLP

pattern

fbpAa

Bovis 684 666773677777600 1 Gly GGC 311 GG

Bovis_BCG 482 676773777777600 1 Gly GGC 311 GG

Beijing 1 000000000003771 8 Arg AGG 156 AT

Beijing 1 000000000003771 13 wt

Beijing 1 000000000003771 1 wt

Beijing 250 000000000000371 1 wt

Beijing 1 000000000003771 3 wt

CAS 26 703777740003771 4 Val GTT 4 GTC V

CAS 357 703777740000771 1 Val GTT 4 GTC V

EAI 19 677777477413771 3 wt

EAI 19 677777477413771 1 wt

unknown 523 777777777777771 4 wt

Haarlem 50 777777777720771 4 wt

Haarlem 50 777777777720771 1 wt

Haarlem 316 777777770020731 1 wt

Haarlem 742 777777770020771 4 wt

Haarlem-like TW6b 747777770020771 1 wt

Haarlem-like TW7b 757777770020771 3 wt

Haarlem-like TW12b 777377750020771 1 wt

Haarlem-like TW13b 777777770000371 1 wt

LAM 33 776177607760771 3 wt

LAM 64 777777607560771 1 wt

T 53 777777777760771 1 wt

Undesignated TW1b 577767777760731 4 wt

Undesignated Undesignated 777777760160731 1 wt

H37Rv 451 777777477760771 1 wt

T 52 777777777760731 1 wt

T 280 770000777760771 1 wt

a wt: wild type (identical to H37Rv strain, GenBank accession no. NC_000962).b TW1, TW6, and TW7: designated by Chuang et al. (2008).

7 min. A 967-bp PCR product for the IS6110 insertion in the NTFregion was designated as indicative of modern Beijing sub-lineages.

2.5. Phylogenetic analysis

The concatenated SNPs of the fbpA, fbpB, fbpC, and pimB genesresulted in a short nucleotide sequence, which was analyzedusing Phylip 3.67 software with the maximum likelihood (ML)method. This method tends to outperform alternative methodssuch as parsimony or distance methods when very shortsequences are analyzed. Labels on branching points of thephylogenetic tree indicated groups separated significantlyaccording to ML-calculated P-values. SNPs of M. tuberculosis

strains H37Rv (GenBank accession no. NC_000962) (Cole et al.,1998) and F11 (GenBank accession no. NC_009565), andM. bovis strains AF2122/97 (GenBank accession no.NC_002945) (Garnier et al., 2003) and BCG Pasteur 1173P2

tes.

fbpBa fbpCa pimBa

G Gly wt wt Gly GGT 107 GGC Gly

G Gly Phe TTC 140 CTC Leu wt Gly GGT 107 GGC Gly

G Met wt wt Gly GGT 107 GGC Gly

Pro CCC 238 CCA Pro wt Gly GGT 107 GGC Gly

Pro CCC 238 CCA Pro Thr ACC 143 GCC Ala Gly GGT 107 GGC Gly

Pro CCC 238 CCA Pro wt Gly GGT 107 GGC Gly

wt wt Gly GGT 107 GGC Gly

al wt Gly GGC 158 AGC Ser Gly GGT 107 GGC Gly

al wt Gly GGC 158 AGC Ser Gly GGT 107 GGC Gly

wt wt Gly GGT 107 GGC Gly,

Ala GCC 270 GCT Ala

wt Gly GGG 27 AGG Arg Gly GGT 107 GGC Gly,

Ala GCC 270 GCT Ala

wt wt Gly GGT 107 GGC Gly

wt wt wt

wt wt Arg CGC 152 CCC Pro

wt wt wt

wt wt wt

wt wt wt

wt wt wt

wt wt wt

wt wt wt

wt Glu GAG 103 GAA Glu wt

wt Glu GAG 103 GAA Glu wt

wt wt wt

wt wt wt

wt wt wt

wt wt wt

wt wt wt

wt wt wt

P.-C. Chuang et al. / Infection, Genetics and Evolution 10 (2010) 459–466462

(GenBank accession no. NC_008769) (Brosch et al., 2007)obtained from the NCBI database were analyzed together with70 study isolates.

3. Results

3.1. Genotypes of the study isolates

Of the 70 genotypes, 37.1% (26/70) and 62.9% (44/70) wereBeijing and non-Beijing lineages, respectively. Compared withSpolDB4, 44 non-Beijing lineages included two Bovis, five CAS, fourEAI, ten Haarlem, six Haarlem-like (as defined in our previousstudy) (Chuang et al., 2008), one H37Rv, four LAM, three T, fourshared type (ST) no. 523 (unknown lineage), and five undesignatedgenotypes (Table 2). Based on a similarity analysis of the RFLPpatterns, eight, thirteen, and two Beijing genotypes were groupedinto clades I, II, and III, respectively. Furthermore, the results of theNTF region analysis indicated that clade I and III genotypescomprised the ancestral Beijing family and clade II the modern one.Nevertheless, three ancestral Beijing genotypes were not in anyclade (Fig. 1).

3.2. SNPs in the fbpA, fbpB, and fbpC genes

Of the 26 Beijing family genotypes analyzed, eight clade I(ancestral Beijing) sublineages displayed an unique non-synon-ymous SNP in fbpA codon 156 (Arg AGG to ATG Met), while another13 clade II (modern Beijing) sublineages and two other ancestralBeijing sublineages with less than 85% similarity displayed asynonymous SNP in fbpB codon 238 (Pro CCC to CCA Pro) (Table 2and Fig. 1). These two unique SNPs were not observed in otherlineages (Table 2). In addition, no mutations were found in eitherthe fbpA or fbpB genes in the three ancestral Beijing genotypes,including two in clade III (Fig. 1).

In addition, all five CAS genotypes harbored SNPs in fbpA codon4 (Val GTT to GTC Val) and fbpC codon 158 (Gly GGC to AGC Ser).

Fig. 1. Dendrogram of RFLP genotypes of 26 Beijing sublineages of Mycobacterium tube

sublineages.

Furthermore, both M. bovis (ST no. 684) and M. bovis-BCG Tokyo172 (ST no. 482) genotypes carried a SNP in codon 311 (Gly GGC311 GGG Gly) of the fbpA gene, while only M. bovis-BCG Tokyo 172had an additional SNP in codon 140 (Phe TTC to CTC Leu) of the fbpB

gene (Table 2). Our findings are concordant with publishedsequences of M. bovis strain AF2122/97 and M. bovis-BCG strainPasteur 1173P2.

The SNP in fbpC codon 103 (Glu GAG to GAA Glu) was identifiedin all four LAM lineage genotypes. This result is concordant with aprevious study indicating that this SNP could be a lineage-specificgenetic marker for LAM isolates (Gibson et al., 2008). In addition, anon-synonymous SNP in fbpC gene codon 27 (Gly GGG 27 AGG Arg)was identified in one EAI (ST no. 19) lineage genotype, and codon143 (Thr ACC 143 GCC Ala) in one Beijing (ST no. 1) lineagegenotype (Table 2).

3.3. SNPs in the pimB gene

In our study population, compared with sequences of H37Rvgenes, 41 genotypes, including Beijing, CAS, EAI, ST no. 523, andBovis, carried a silent polymorphism in pimB codon 107 (Gly GGT toGGC Gly). We also found that all four EAI genotypes harbored aunique concurrent SNP in pimB codon 270 (Ala GCC to GCT Ala). Anon-synonymous SNP in pimB gene codon 152 (Arg CGC to CCC Pro)was identified in one Haarlem (shared type no. 50) lineage (Table 2).

3.4. Phylogenetic analysis

A total of 12 SNPs identified in fbpA, fbpB, fbpC and pimB geneswere concatenated, resulting in a short nucleotide sequence foreach genotype analyzed. The phylogenetic tree generated showedthat the Beijing sublineage carrying fbpA or fbpB SNPs, EAI, CAS,LAM lineages of M. tuberculosis, as well as M. bovis, were groupedseparately (P < 0.01). In addition, the M. bovis BCG lineage could bedistinguished from M. bovis based on only one SNP (fbpB codon140) difference (Fig. 2).

rculosis. (*) NTF analyses were performed to identify ancestral (A) or modern (M)

Fig. 2. Phylogenetic tree analysis of 68 Mycobacterium tuberculosis, 2 Mycobacterium bovis isolates, and 4 reference strains (H37Rv, F11, AF2122/97, and Pasteur 1173P2) using

12 SNPs by the maximum likelihood method (unrooted phylogenetic tree). Beijing_fbpA156 and Beijing_fbpB238 represented Beijing sublineages harboring SNPs in fbpA codon

156 and fbpB codon 238, respectively. *P < 0.01.

P.-C. Chuang et al. / Infection, Genetics and Evolution 10 (2010) 459–466 463

4. Discussion

Nine novel SNPs in cell wall biosynthesis-associated genes,fbpA, fbpB, fbpC, and pimB were found in specific lineages of M.

tuberculosis and M. bovis in this study. These novel SNPs werephylogenetic informative but however not recognized in previousstudies (Gutacker et al., 2002; Filliol et al., 2006). According to ourSNPs findings (Fig. 2) and the deletion of TbD1 region, we maypresume that the M. tuberculosis descended from a most recentcommon ancestor (MRCA) and then evolved into each lineage andsublineage with specific SNPs (Fig. 3). Our results are comparableto previous studies (Baker et al., 2004; Filliol et al., 2006; Gagneuxand Small, 2007) as well as congruent to that of Comas et al. (2009)using more than 300 SNPs.

In our study, SNPs were identified by comparing with respectivegenes of H37Rv reference strain. We found lineages of M.

tuberculosis including EAI, Beijing, CAS, and M. bovis harboredGGC in pimB gene codon 107; while other spoligotype lineagescorresponding to Lineage 4 and H37Rv had GGT (Table 2). We alsochecked GGC in the pimB gene codon 107 in M. canettii (http://www.sanger.ac.uk/sequencing/Mycobacterium/canetti/). There-fore, we concluded that the SNP in pimB codon 107 GGC to GGTwas a robust marker for Lineage 4. Furthermore, the commonancestor of EAI lineage possesses the pimB gene codon 270 SNP.Since EAI lineage, having TbD1 region, was considered as ancient,EAI lineage can be separated from other Lineages (Lineages 2 and3). For the Beijing lineage, the early ancestral Beijing sublineageshad no SNPs in the fbpA and fbpB genes, and then descended intoancestral sublineages harboring the SNP in fbpA gene codon 156;whereas Beijing sublineages carrying a SNP in fbpB gene codon 238evolved into modern sublineages. The CAS lineage might haveevolved with SNPs in fbpA gene codon 4 and fbpC gene codon 158.

Fig. 3. Evolutionary hypothesis lineages/sublineages of the Mycobacterium tuberculosis based on the SNPs in the fbpA, fbpB, fbpC, and pimBgenes and the deletion of TbD1

region. MRCA stands for most recent common ancestor of M. tuberculosis complex. Lineage 1–4 and animal-adapted lineage are designated in Comas et al. (2009) study.

P.-C. Chuang et al. / Infection, Genetics and Evolution 10 (2010) 459–466464

In addition, M. bovis acquired one SNP in fbpA gene codon 311,whereas M. bovis-BCG subsequently harbored another SNP in fbpB

gene codon 140. We provided the first evidence from Asia andconfirmed the SNP in fbpC codon 103 (Glu GAG to GAA Glu) as amarker for LAM proposed by Gibson et al. (2008) (Fig. 3).

To identify six main lineages (Lineages 1–6) and one animal-adapted lineage, Comas et al. (2009) extracted 93 phylogeneticallyinformative SNPs, many of them were redundant, from more than300 original SNPs. EAI lineage (Lineage 1) can be subdivided intotwo distinct groups, the Philippines and the Rim of Indian Oceansublineages (Hershberg et al., 2008; Comas et al., 2009). The SNP ofpimB codon 270 (GCC to GCT) was found in sequences of those twoEAI sublineages of Philippines (isolates T17, T46, T92) and the Rimof Indian Ocean (isolate EAS054) (http://www.broadinstitute.org/).Besides, we matched sequences of two Beijing isolates (in Lineage2), 02_1987 (from South Korea) and T85 (from China) (http://www.broadinstitute.org/). No SNP of fbpA codon 156 and fbpB

codon 238 was found in 02_1987 isolate while only SNP of fbpB

codon 238 was found in T85. We may suggest the South KoreaBeijing strain 02_1987 should be early ancestral Beijing whileChina strain T85 modern Beijing. This was further justified by thefact that ancestral and modern Beijing was prevalent in SouthKorea and China, respectively (personal communication). Inaddition, our findings of fbpA codon 4 (GTT to GTC) and fbpC

codon 158 (GGC to AGC) in CAS lineage is consistent with SNPsidentified as informative markers in Lineage 3 (Fig. 3).

In this study, ancestral and modern Beijing sublineages can bedistinguished based on SNPs of fbpA and fbpB of antigen 85complex genes. We found that two ancestral Beijing genotypesharbored a SNP in fbpB gene codon 238, suggesting that they mightbe transient sublineages positioned between the ancestral andmodern sublineages. The SNP in fbpB gene codon 238 might haveevolved before the IS6110 insertion in the NTF region. Since twoancestral Beijing genotypes in clade III had no mutations in eitherthe fbpA or the fbpB gene, they might be earlier ancestral than thosein clade I. This result was similar to that of the study in mutatorgenes which indicated that ancestral Beijing isolates did not haveSNP in mutT4, mutT2, or ogt. Sublineages of the modern Beijingfamily diverged after acquired mutations in mutT4 codon 48, ogt

codon 37, or mutT2 codon 58 (Rad et al., 2003). Additionally, Rindiet al. (2009) studied the evolutionary pathway of Beijing lineage ofM. tuberculosis based on SNPs in mutT2 and mutT4 were compatibleto that based on LSPs (RD105, 181, 150, and 142). Musser et al.

(2000) reported SNP in the fbpB codon 238 in one of the six Beijingisolates analyzed, however, the SNP was not considered as agenetic marker. In addition, the ST no. 523 isolates with no SNPs inthe fbpA, fbpB, or fbpC genes can be the early ancestral Beijingsublineage in the clade III (Fig. 3). Flores et al. (2007) reported thattwo ST no. 523 strains (94_M4241A and 93_2461) harbored adeletion in RD105, a marker for Beijing family, and therefore mightbelong to a Beijing lineage of M. tuberculosis. Hershberg et al.(2008) showed that 94_M4241A was included in the East Asialineage, and was an early offshoot of the ancestral Beijingsublineages. Therefore, in this study, we contributed an additionalevidence to prove ST no. 523 is part of the Beijing family. Indeed,certain spoligotypes (designated as unknown lineages in spolDB4)were still not well-defined genetically. Those unclassified spoli-gotypes may be reclassified upon additional SNPs and LSPs.

In this study, the novel and specific SNPs in M. tuberculosis weresynonymous, except those in fbpA codon 156 and fbpC codon 158.Crystal structure of antigen 85 complex proteins revealed that theactive site of the catalytic triad were residues Ser 124 (amino acid169, 166, and 170 in FbpA, FbpB, and FbpC, respectively), His 260(amino acid 305, 302, and 306 in FbpA, FbpB, and FbpC,respectively), and Glu 228 (amino acid 273, 270, and 274 in FbpA,FbpB, and FbpC, respectively) (Takayama et al., 2005). These twonon-synonymous SNPs identified in fbpA codon 156 and fbpC

codon 158 in the Beijing and CAS lineages, respectively, might notbe involved in the active site or in the fibronectin binding region(Naito et al., 1998). However, studies have also shown that theantigen 85 complex plays a role in the survival of tuberculousbacillus and its ability to escape the human immune response(Armitige et al., 2000; Mariani et al., 2000). The residues involvedin catalysis or fibronectin binding were structurally moreconserved in three proteins of the antigen 85 complex, while ahigh variability in surface residues was found for the remainingregions of the protein surface. Variations in the antigen 85 complexmight be important in the survival of M. tuberculosis (Ronning et al.,2004). Therefore, specific SNPs found in particular M. tuberculosis

lineages might result from environmental or host selectionpressures, leading to clonal expansion in subsequent populations.The impact of SNPs associated with cell wall synthesis on the clonalexpansion of M. tuberculosis, as well as in its escape from theimmune response, requires further investigation.

In conclusion, we found that lineage-specific SNPs couldunambiguously elucidate differentiated lineages of M. tuberculosis

P.-C. Chuang et al. / Infection, Genetics and Evolution 10 (2010) 459–466 465

complex and might have implications for the evolution of M.

tuberculosis. Consequently, using only nine novel and informativeSNPs in the fbpA, fbpB, fbpC, and pimB genes as specific geneticmarkers could facilitate the molecular epidemiological studies andrapid diagnosis of lineages/sublineages of M. tuberculosis and M.

bovis.

Acknowledgments

This study was supported by grants DOH97-DC-2502 andDOH96-DC-2006 from the Centers for Disease Control, Departmentof Health, Taiwan.

The authors wish to thank nine Taiwan-CDC-contractedMycobacteriology laboratories for providing Mycobacterium tuber-

culosis isolates, as well as Hsin-Fu Liu, Chen-Che Chiu, Kan-Ju Ho,and Su-Ying Chang for their technical support.

References

Achtman, M., 2008. Evolution, population structure, and phylogeography of geneti-cally monomorphic bacterial pathogens. Annu. Rev. Microbiol. 62, 53–70.

Alderwick, L.J., Birch, H.L., Mishra, A.K., Eggeling, L., Besra, G.S., 2007. Structure,function and biosynthesis of the Mycobacterium tuberculosis cell wall: arabi-nogalactan and lipoarabinomannan assembly with a view to discovering newdrug targets. Biochem. Soc. Trans. 35, 1325–1328.

Alix, E., Godreuil, S., Blanc-Potard, A.B., 2006. Identification of a Haarlem genotype-specific single nucleotide polymorphism in the mgtC virulence gene of Myco-bacterium tuberculosis. J. Clin. Microbiol. 44, 2093–2098.

Armitige, L.Y., Jagannath, C., Wanger, A.R., Norris, S.J., 2000. Disruption of the GenesEncoding Antigen 85A and Antigen 85B of Mycobacterium tuberculosis H37Rv:Effect on Growth in Culture and in Macrophages. Infect. Immun. 68, 767–778.

Arnold, C., 2007. Molecular evolution of Mycobacterium tuberculosis. Clin. Microbiol.Infect. 13, 120–128.

Baker, L., Brown, T., Maiden, M.C., Drobniewski, F., 2004. Silent nucleotide poly-morphisms and a phylogeny for Mycobacterium tuberculosis. Emerg. Infect. Dis.10, 1568–1577.

Brosch, R., Gordon, S.V., Marmiesse, M., Brodin, P., Buchrieser, C., Eiglmeier, K.,Garnier, T., Gutierrez, C., Hewinson, G., Kremer, K., Parsons, L.M., Pym, A.S.,Samper, S., van Soolingen, D., Cole, S.T., 2002. A new evolutionary scenario forthe Mycobacterium tuberculosis complex. Proc. Natl. Acad. Sci. U.S.A. 99, 3684–3689.

Brosch, R., Gordon, S.V., Garnier, T., Eiglmeier, K., Frigui, W., Valenti, P., Dos Santos,S., Duthoy, S.p., Lacroix, C., Garcia-Pelayo, C., Inwald, J.K., Golby, P., Garcia, J.N.e.,Hewinson, R.G., Behr, M.A., Quail, M.A., Churcher, C., Barrell, B.G., Parkhill, J.,Cole, S.T., 2007. Genome plasticity of BCG and impact on vaccine efficacy. Proc.Natl. Acad. Sci. U.S.A. 104, 5596–5601.

Brudey, K., Driscoll, J., Rigouts, L., Prodinger, W., Gori, A., Al-Hajoj, S., Allix, C.,Aristimuno, L., Arora, J., Baumanis, V., Binder, L., Cafrune, P., Cataldi, A., Cheong,S., Diel, R., Ellermeier, C., Evans, J., Fauville-Dufaux, M., Ferdinand, S., de Viedma,D., Garzelli, C., Gazzola, L., Gomes, H., Guttierez, M.C., Hawkey, P., van Helden, P.,Kadival, G., Kreiswirth, B., Kremer, K., Kubin, M., Kulkarni, S., Liens, B., Lillebaek,T., Ly, H., Martin, C., Martin, C., Mokrousov, I., Narvskaia, O., Ngeow, Y.,Naumann, L., Niemann, S., Parwati, I., Rahim, Z., Rasolofo-Razanamparany, V.,Rasolonavalona, T., Rossetti, M.L., Rusch-Gerdes, S., Sajduda, A., Samper, S.,Shemyakin, I., Singh, U., Somoskovi, A., Skuce, R., van Soolingen, D., Streicher, E.,Suffys, P., Tortoli, E., Tracevska, T., Vincent, V., Victor, T., Warren, R., Yap, S.,Zaman, K., Portaels, F., Rastogi, N., Sola, C., 2006. Mycobacterium tuberculosiscomplex genetic diversity: mining the fourth international spoligotyping data-base (SpolDB4) for classification, population genetics and epidemiology. BMCMicrobiol. 6, 23.

Chan, J., Fan, X.D., Hunter, S.W., Brennan, P.J., Bloom, B.R., 1991. Lipoarabinoman-nan, a possible virulence factor involved in persistence of Mycobacteriumtuberculosis within macrophages. Infect. Immun. 59, 1755–1761.

Chuang, P.C., Liu, H., Sola, C., Chen, Y.M., Jou, R., 2008. Spoligotypes of Mycobacteriumtuberculosis isolates of a high tuberculosis burden aboriginal township inTaiwan. Infect. Genet. Evol. 8, 553–557.

Cole, S.T., Brosch, R., Parkhill, J., Garnier, T., Churcher, C., Harris, D., Gordon, S.V.,Eiglmeier, K., Gas, S., Barry, C.E., Tekaia, F., Badcock, K., Basham, D., Brown, D.,Chillingworth, T., Connor, R., Davies, R., Devlin, K., Feltwell, T., Gentles, S.,Hamlin, N., Holroyd, S., Hornsby, T., Jagels, K., Krogh, A., McLean, J., Moule,S., Murphy, L., Oliver, K., Osborne, J., Quail, M.A., Rajandream, M.A., Rogers, J.,Rutter, S., Seeger, K., Skelton, J., Squares, R., Squares, S., Sulston, J.E., Taylor, K.,Whitehead, S., Barrell, B.G., 1998. Deciphering the biology of Mycobacteriumtuberculosis from the complete genome sequence. Nature 393, 537–544.

Comas, I., Homolka, S., Niemann, S., Gagneux, S., 2009. Genotyping of geneticallymonomorphic bacteria: DNA sequencing in Mycobacterium tuberculosis high-lights the limitations of current methodologies. PLos ONE 4, e7815.

Copenhaver, R.H., Sepulveda, E., Armitige, L.Y., Actor, J.K., Wanger, A., Norris, S.J.,Hunter, R.L., Jagannath, C., 2004. A mutant of Mycobacterium tuberculosis H37Rv

that lacks expression of antigen 85A is attenuated in mice but retains vaccino-genic potential. Infect. Immun. 72, 7084–7095.

Dover, L.G., Alderwick, L.J., Brown, A.K., Futterer, K., Besra, G.S., 2007. Regulation ofcell wall synthesis and growth. Curr. Mol. Med. 7, 247–276.

Filliol, I., Motiwala, A.S., Cavatore, M., Qi, W., Hazbon, M.H., Bobadilla del Valle, M.,Fyfe, J., Garcia-Garcia, L., Rastogi, N., Sola, C., Zozio, T., Guerrero, M.I., Leon, C.I.,Crabtree, J., Angiuoli, S., Eisenach, K.D., Durmaz, R., Joloba, M.L., Rendon, A.,Sifuentes-Osornio, J., Ponce de Leon, A., Cave, M.D., Fleischmann, R., Whittam,T.S., Alland, D., 2006. Global phylogeny of Mycobacterium tuberculosis based onsingle nucleotide polymorphism (SNP) analysis: insights into tuberculosisevolution, phylogenetic accuracy of other DNA fingerprinting systems, andrecommendations for a minimal standard SNP set. J. Bacteriol. 188, 759–772.

Flores, L., Van, T., Narayanan, S., DeRiemer, K., Kato-Maeda, M., Gagneux, S., 2007.Large sequence polymorphisms classify Mycobacterium tuberculosis strains withancestral spoligotyping patterns. J. Clin. Microbiol. 45, 3393–3395.

Gagneux, S., DeRiemer, K., Van, T., Kato-Maeda, M., de Jong, B.C., Narayanan, S.,Nicol, M., Niemann, S., Kremer, K., Gutierrez, M.C., Hilty, M., Hopewell, P.C.,Small, P.M., 2006. Variable host-pathogen compatibility in Mycobacteriumtuberculosis. Proc. Natl. Acad. Sci. U.S.A. 103, 2869–2873.

Gagneux, S., Small, P.M., 2007. Global phylogeography of Mycobacterium tubercu-losis and implications for tuberculosis product development. Lancet Infect. Dis.7, 328–337.

Garnier, T., Eiglmeier, K., Camus, J.C., Medina, N., Mansoor, H., Pryor, M., Duthoy, S.,Grondin, S., Lacroix, C., Monsempe, C., Simon, S., Harris, B., Atkin, R., Doggett, J.,Mayes, R., Keating, L., Wheeler, P.R., Parkhill, J., Barrell, B.G., Cole, S.T., Gordon,S.V., Hewinson, R.G., 2003. The complete genome sequence of Mycobacteriumbovis. Proc. Natl. Acad. Sci. U.S.A. 100, 7877–7882.

Gibson, A.L., Huard, R.C., Gey van Pittius, N.C., Lazzarini, L.C.O., Driscoll, J., Kurepina,N., Zozio, T., Sola, C., Spindola, S.M., Kritski, A.L., Fitzgerald, D., Kremer, K.,Mardassi, H., Chitale, P., Brinkworth, J., Garcia de Viedma, D., Gicquel, B., Pape,J.W., van Soolingen, D., Kreiswirth, B.N., Warren, R.M., van Helden, P.D., Rastogi,N., Suffys, P.N., Lapa e Silva, J., Ho, J.L., 2008. Application of sensitive and specificmolecular methods to uncover global dissemination of the major RDRio sub-lineage of the Latin American-Mediterranean Mycobacterium tuberculosis spo-ligotype family. J. Clin. Microbiol. 46, 1259–1267.

Glickman, M.S., Jacobs, J., 2001. Microbial pathogenesis of Mycobacterium tubercu-losis: dawn of a discipline. Cell 104, 477–485.

Gutacker, M.M., Mathema, B., Soini, H., Shashkina, E., Kreiswirth, B.N., Graviss, E.A.,Musser, J.M., 2006. Single-nucleotide polymorphism-based population geneticanalysis of Mycobacterium tuberculosis strains from 4 geographic sites. J. Infect.Dis. 193, 121–128.

Gutacker, M.M., Smoot, J.C., Lux Migliaccio, C.A., Ricklefs, S.M., Hua, S., Cousins, D.V.,Graviss, E.A., Shashkina, E., Kreiswirth, B.N., Musser, J.M., 2002. Genome-wideanalysis of synonymous single nucleotide polymorphisms in Mycobacteriumtuberculosis complex organisms: resolution of genetic relationships amongclosely related microbial strains. Genetics 162, 1533–1543.

Gutierrez, M.C., Ahmed, N., Willery, E., Narayanan, S., Hasnain, S.E., Chauhan, D.S.,Katoch, V.M., Vincet, V., Locht, C., Supply, P., 2006. Predominance of ancestrallineages of Mycobacterium tuberculosis in India. Emerg. Infect. Dis. 12, 1367–1374.

Harth, G., Lee, B.Y., Wang, J., Clemens, D.L., Horwitz, M.A., 1996. Novel insights intothe genetics, biochemistry, and immunocytochemistry of the 30-kilodaltonmajor extracellular protein of Mycobacterium tuberculosis. Infect. Immun. 64,3038–3047.

Hershberg, R., Lipatov, M., Small, P.M., Sheffer, H., Niemann, S., Homolka, S., Roach,J.C., Kremer, K., Petrov, D.A., Feldman, M.W., Gagneux, S., 2008. High functionaldiversity in Mycobacterium tuberculosis deiven by genetic drift and humandemography. PLoS Biol. 6, e311.

Horwitz, M.A., Lee, B.W., Dillon, B.J., Harth, G., 1995. Protective immunity againsttuberculosis induced by vaccination with major extracellular proteins of Myco-bacterium tuberculosis. Proc. Natl. Acad. Sci. U.S.A. 92, 1530–1534.

Jiao, W., Mokrousov, I., Sun, G., Li, M., Liu, J., Narvskaya, O., Shen, A., 2007. Molecularcharacteristics of rifampin and isoniazid resistant Mycobacterium tuberculosisstrains from Beijing, China. Clin. Med. J. (Engl.) 120, 814–819.

Jou, R., Chiang, C.Y., Huang, W.L., 2005. Distribution of the Beijing family genotypesof Mycobacterium tuberculosis in Taiwan. J. Clin. Microbiol. 43, 95–100.

Kalendar, R., Lee, D., Schulman, A.H., 2009. FastPCR software for PCR primer andprobe design and repeat search. http://www.biocenter.helsinki.fi/bi/programs/fastpcr.htm [online].

Kamerbeek, J., Schouls, L., Kolk, A., van Agterveld, M., van Soolingen, D., Kuijper, S.,Bunschoten, A., Molhuizen, H., Shaw, R., Goyal, M., van Embden, J., 1997.Simultaneous detection and strain differentiation of Mycobacterium tuberculosisfor diagnosis and epidemiology. J. Clin. Microbiol. 35, 907–914.

Kremer, K., Glynn, J.R., Lillebaek, T., Niemann, S., Kurepina, N.E., Kreiswirth, B.N.,Bifani, P.J., van Soolingen, D., 2004. Definition of the Beijing/W lineage ofMycobacterium tuberculosis on the basis of genetic markers. J. Clin. Microbiol.42, 4040–4049.

Mariani, F., Cappelli, G., Riccardi, G., Colizzi, V., 2000. Mycobacterium tuberculosisH37Rv comparative gene-expression analysis in synthetic medium and humanmacrophage. Gene 253, 281–291.

Mokrousov, I., Ly, H.M., Otten, T., Lan, N.N., Vyshnevskyi, B., Hoffner, S., Narvskaya,O., 2005. Origin and primary dispersal of the Mycobacterium tuberculosis Beijinggenotype: clues from human phylogeography. Genome Res. 15, 1357–1364.

Mokrousov, I., Narvskaya, O., Otten, T., Vyazovaya, A., Limeschenko, E., Steklova, L.,Vyshnevskyi, B., 2002. Phylogenetic reconstruction within Mycobacterium

P.-C. Chuang et al. / Infection, Genetics and Evolution 10 (2010) 459–466466

tuberculosis Beijing genotype in northwestern Russia. Res. Microbiol. 153, 629–637.

Moreno, C., Mehlert, A., Lamb, J., 1988. The inhibitory effects of mycobacteriallipoarabinomannan and polysaccharides upon polyclonal and monoclonalhuman T cell proliferation. Clin. Exp. Immunol. 74, 206–210.

Musser, J.M., Amin, A., Ramaswamy, S., 2000. Negligible genetic diversity ofMycobacterium tuberculosis host immune system protein targets: evidence oflimited selective pressure. Genetics 155, 7–16.

Naito, M., Ohara, N., Matsumoto, S., Yamada, T., 1998. The Novel fibronectin-bindingmotif and key residues of mycobacteria. J. Biol. Chem. 273, 2905–2909.

Olano, J., Lopez, B., Reyes, A., del Pilar Lemos, M.a., Correa, N., Del Portillo, P., Barrera,L., Robledo, J., Ritacco, V., Mercedes Zambrano, M.a., 2007. Mutations in DNArepair genes are associated with the Haarlem lineage of Mycobacterium tuber-culosis independently of their antibiotic resistance. Tuberculosis 87, 502–508.

Puech, V., Guilhot, C., Perez, E., Tropis, M., Armitige, L.Y., Gicquel, B., Daffe, M., 2002.Evidence for a partial redundancy of the fibronectin-binding proteins for thetransfer of mycoloyl residues onto the cell wall arabinogalactan termini ofMycobacterium tuberculosis. Mol. Microbiol. 44, 1109–1122.

Rad, M.E., Bifani, P., Martin, C., Kremer, K., Samper, S., Rauzier, J., Kreiswirth, B.,Blazquez, J., Jouan, M., van Soolingen, D., Gicquel, B., 2003. Mutations in putativemutator genes of Mycobacterium tuberculosis strains of the W-Beijing family.Emerg. Infect. Dis. 9, 838–845.

Rindi, L., Lari, N., Cuccu, B., Garzelli, C., 2009. Evolutionary pathway of the Beijinglineage of Mycobacterium tuberculosis based on genomic deletions and mutTgenes polymorphisms. Infect. Genet. Evol. 9, 48–53.

Ronning, D.R., Vissa, V., Besra, G.S., Belisle, J.T., Sacchettini, J.C., 2004. Mycobacteriumtuberculosis Antigen 85A and 85C Structures Confirm Binding Orientation andConserved Substrate Specificity. J. Biol. Chem. 279, 36771–36777.

Schaeffer, M.L., Khoo, K.H., Besra, G.S., Chatterjee, D., Brennan, P.J., Belisle, J.T.,Inamine, J.M., 1999. The pimB gene of Mycobacterium tuberculosis encodes a

mannosyltransferase involved in lipoarabinomannan biosynthesis. J. Biol.Chem. 274, 31625–31631.

Sebban, M., Mokrousov, I., Rastogi, N., Sola, C., 2002. A data-mining approach tospacer oligonucleotide typing of Mycobacterium tuberculosis. Bioinformatics 18,235–243.

Smith, I., 2003. Mycobacterium tuberculosis pathogenesis and molecular determi-nants of virulence. Clin. Microbiol. Rev. 16, 463–496.

Smith, N.H., Hewinson, R.G., Kremer, K., Brosch, R., Gordon, S.V., 2009. Myths andmisconceptions: the origin and evolution of Mycobacterium tuberculosis. Nat.Rev. Microbiol. 7, 537–544.

Sreevatsan, S., Pan, X., Stockbauer, K.E., Connell, N.D., Kreiswirth, B.N., Whittam, T.S.,Musser, J.M., 1997. Restricted structural gene polymorphism in the Mycobac-terium tuberculosis complex indicates evolutionarily recent global dissemina-tion. Proc. Natl. Acad. Sci. U. S. A 94, 9869–9874.

Takayama, K., Wang, C., Besra, G.S., 2005. Pathway to synthesis and processing ofmycolic acids in Mycobacterium tuberculosis. Clin. Microbiol. Rev. 18, 81–101.

Torrelles, J.B., DesJardin, L.E., MacNeil, J., Kaufman, T.M., Kutzbach, B., Knaup, R.,McCarthy, T.R., Gurcha, S.S., Besra, G.S., Clegg, S., Schlesinger, L.S., 2009. Inacti-vation of Mycobacterium tuberculosis mannosyltransferase PimB reduces thecell wall lipoarabinomannan and lipomannan content and increases the rate ofbacterial-induced human macrophage cell death. Glycobiology 19, 743–755.

Tsolaki, A.G., Gagneux, S., Pym, A.S., Goguet de la Salmoniere, Y.-O., Kreiswirth, B.N.,van Soolingen, D., Small, P.M., 2005. Genomic deletions classify the Beijing/Wstrains as a distinct genetic lineage of Mycobacterium tuberculosis. J. Clin.Microbiol. 43, 3185–3191.

van Embden, J.D., Cave, M.D., Crawford, J.T., Dale, J.W., Eisenach, K.D., Gicquel, B.,Hermans, P., Martin, C., McAdam, R., Shinnick, T.M., Small, P.M., 1993. Strainidentification of Mycobacterium tuberculosis by DNA fingerprinting: recommen-dations for a standardized methodology. J. Clin. Microbiol. 31, 406–409.

Wiker, H.G., Harboe, M., 1992. The antigen 85 complex: a major secretion product ofMycobacterium tuberculosis. Microbiol. Rev. 56, 648–661.