Diagnostic Microbiology and Infectious Disease 77 (2013) 53–57

Contents lists available at SciVerse ScienceDirect

Diagnostic Microbiology and Infectious Disease

j ourna l homepage: www.e lsev ie r .com/ locate /d iagmicrob io

Combination of multiplex PCR with denaturing high-performance liquidchromatography for rapid detection of Mycobacterium genus and simultaneousidentification of the Mycobacterium tuberculosis complex

Ru Chen a,⁎, Xiao-Bo Gao b, Zhi-Hui Liu c, Xiao-Bing Shen d, Ai-Zhen Guo e, Yan-Yu Duan a, Zhi-Ling Liu a,Xiao-Wei Wu a, Dao-Zhong Zhu a

a Technical Center, Guangdong Entry-Exit Inspection and Quarantine Bureau, Guangzhou 510623, Chinab Department of Genetics, National Research Institute for Family Planning, Beijing 100081, Chinac Guangzhou Chest Hospital, Guangzhou 510095, Chinad National Institutes for Food and Drug Control, Beijing 100050, Chinae College of Animal Science, Huazhong Agriculture University, Wuhan 430070, China

a b s t r a c ta r t i c l e i n f o

⁎ Corresponding author. Tel.: +86-20-38290911; fax:E-mail address: ychenru@sina.com (R. Chen).

0732-8893/$ – see front matter © 2013 Elsevier Inc. Alhttp://dx.doi.org/10.1016/j.diagmicrobio.2013.06.003

Article history:Received 9 April 2013Received in revised form 22 May 2013Accepted 1 June 2013Available online 16 July 2013

Keywords:DHPLCMultiplex PCRMycobacteriumThe Mycobacterium tuberculosis complex

A new assay with the combination of multiplex polymerase chain reaction and denaturing high-performanceliquid chromatography analysis was developed for simultaneous detection of Mycobacterium genus andidentification of the Mycobacterium tuberculosis complex (MTC). Targeting at genus-specific 16S rRNAsequence ofMycobacterium and specific insertion elements IS6110 and IS1081 of MTC, the assaywas validatedwith 84 strains covering 23 mycobacteria species and 30 strains of non-mycobacteria species. No crossreactivity was observed. Clinical application was carried out on 198 specimens (155 human sputum and 43bovine tissue samples) and compared with culture. The multiplex assay detected all culture-positive (36 innumber) and 35.2% (57/162) culture-negative specimens. The molecular assay was fast that could becompleted within 1 h on purified DNA, with the limit of detection as 0.8–1.6 pg per reaction on DNA template.This work provided a useful laboratory tool for rapid identification of Mycobacterium and differentiation ofMTC and nontuberculous mycobacteria.

+86-20-38290422.

l rights reserved.

© 2013 Elsevier Inc. All rights reserved.

1. Introduction

Mycobacterium causes tuberculosis and other diseases of healthproblem. The genus Mycobacterium comprises members of theMycobacterium tuberculosis complex (MTC) and nontuberculousmycobacteria. MTC members as a group are pathogen of human andanimal tuberculosis. Nontuberculous mycobacteria (NTM) includemore than 100 mycobacterial species. NTM may spread from water,soil, and animal to human and cause mycobacterial infection in bothimmunocompromised and immunocompetent individuals (Glassroth,2008). According to the National Tuberculosis Epidemiology Survey inChina, the infection of NTM in human has increased from 4.9% in 1990to 11.1% in 2000 (Duanmu, 2002). Major animal NTM diseases includeParatuberculosis and avian tuberculosis that have brought heavyeconomic losses in animal husbandry. Infection of MTC and NTMmembers causes similar clinical symptom but requires differentantibiotic therapy. NTM infection may induce false “positive” reactionof tuberculin skin test on animal because mycobacteria species arehighly homologous in antigenic and genetic aspects. Rapid identifi-

cation of mycobacterium groups and species not only serves toprompt accurate diagnosis and treatment of mycobacterial diseasesbut also aids with appropriate border restriction measures oninternational animal trade.

Although culture and biochemical test is considered as the “gold-standard”method for identifying mycobacteria, the procedure is timeconsuming and usually takes several weeks or longer. Moleculardiagnostic methods allow for direct detection of mycobacteria DNAfrom specimens before culture. Polymerase chain reaction (PCR)–based methods have shorten the diagnostic time to within 1 day. PCRor real-time PCR methods for direct detection, group or speciesidentification of mycobacterium (Anilkumar et al., 2012; Issa et al.,2012; Reddington et al., 2011; Richardson et al., 2009), and multiplexassay or high-density oligonucleotide array for simultaneous identi-fication of different mycobacterial species have been developed (Chenet al., 2010; Park et al., 2005; Troesch et al., 1999). The currentlyavailable molecular diagnostic assays have advantages but withlimitations. Conventional PCR relied on gel-based analysis is laboriousand prone to cause contamination by exposure of DNA amplicon forelectrophoresis, due to the fact that it may produce and release nucleicacid aerosol when opening the tube to take DNA sample forelectrophoresis. Real-time PCR is fast and highly sensitive but limitedin multiplex detection, and the amplicon could not be further

Table 1Detection on mycobacterial strains by the multiplex PCR-DHPLC assay.

Species Source Results

IS6110 IS1081 16S rRNA

M. tuberculosis H37Rv ATCC 27294 + + +M. tuberculosis isolate strains(n = 41)

Human + + +

M. bovis ATCC 27291 + + +M. bovis BCG Vaccine + + +M. bovis isolate strains (n = 9) Cattle + + +Mycobacterium avium ATCC 15769 − − +M. avium ATCC 25291 − − +M. avium isolate strain Cattle − − +M. avium isolate strains (n = 2) Goat − − +M. avium subsp. paratuberculosis CVCC 68605 − − +M. avium subsp. paratuberculosis CVCC 68623 − − +M. avium subsp. paratuberculosis CVCC 68627 − − +M. avium subsp. paratuberculosis CVCC 68635 − − +M. avium subsp. paratuberculosis CVCC 68636 − − +M. avium subsp. paratuberculosis CVCC 68601 − − +M. avium subsp. paratuberculosis CVCC 68637 − − +Mycobacterium chelonae ATCC 14472 − − +Mycobacterium phlei ATCC 11758 − − +Mycobacterium kansasii ATCC 12478 − − +Mycobacterium intracellulare ATCC 13950 − − +Mycobacterium smegmatis ATCC 19420 − − +Mycobacterium gastri ATCC 15754 − − +Mycobacterium scrofulaceum ATCC 19981 − − +Mycobacterium fortuitum ATCC 6481 − − +Mycobacterium xenopi ATCC 19970 − − +Mycobacterium diernhoferi CGMCC 4.1179 − − +Mycobacterium nonchromogenium CMCC(B)95007 − − +Mycobacterium mamo CMCC(B)95010 − − +Mycobacterium farcinogenes CMCC(B)95012 − − +Mycobacterium marinum CMCC(B)95014 − − +Mycobacterium asiaticum CMCC(B)95016 − − +Mycobacterium gordonae CMCC(B)95018 − − +Mycobacterium szulgai CMCC(B)95019 − − +Mycobacterium chelonae subsp.abscessus

CMCC(B)95021 − − +

Mycobacterium flavescens CMCC(B)95030 − − +

CVCC = China Veterinary Culture Collection Center; CGMCC = China GeneralMicrobiology Culture Collection Center; + = positive; − = negative.

54 R. Chen et al. / Diagnostic Microbiology and Infectious Disease 77 (2013) 53–57

analyzed in assays based on TaqMan® fluorescent labeled probe. DNAmicroarray techniques consist of complicate steps of DNA amplifica-tion and hybridization and require expensive reagents and apparatus,which limit routine application of the techniques. Research ondeveloping new diagnostic methods to overcome limitations ofcurrent technology will improve the diagnostic system for mycobac-terial infection.

Denaturing high-performance liquid chromatography (DHPLC) fornucleic acid analysis is the relatively new technology that has beenused to detect PCR-amplified genes and genome mutations. DHPLCtechnique can precisely distinguish gene fragments with minor sizedifference and provides automatic separation and collection of largenumbers of gene amplicons. More recently, DHPLC has been applied inclinical and environmental microbiology, immunology, virology, andtoxicology (Troedsson et al., 2008a). For example, DHPLC was used inidentification and typing of parasites, virus, and bacteria strains(Goldenberg et al., 2005; Hurtle et al., 2003; Li et al., 2003; Troedssonet al., 2008a, 2008b). In the field of mycobacteria research, DHPLC hasbeen successfully applied in identification of drug-resistant MTCstrains (Evans et al., 2004, 2009; Shi et al., 2007; Wilson., 2011).

In this study, a novel method with combination of multiplex PCRand DHPLC analysis was developed for rapid detection of Mycobac-terium genus and simultaneous identification of the MTC. The assayreported here could support differential diagnosis of tuberculous andnontuberculous mycobacteria infection.

2. Materials and methods

2.1. Bacteria

A total of 84 mycobacterial strains covering 23 species used in thisstudy are shown in Table 1. This study also used reference strains of 30non-mycobacteria species, including species of Actinomycetales. Aslisted below, the 30 reference strains were either from the AmericanType Culture Collection Center (ATCC), the China Veterinary CultureCollection Center (CMCC), or the Agricultural Culture Collection ofChina (ACCC): including Streptomyces coralus Dietz (ATCC 23901),Streptomyces lateritius (ATCC 19776), Corynebacterium glutamicum(ACCC 04261), Nocardioides albus (ACCC 41020), Rhodococcus opacus(ACCC 41021), Listeria innocua (ATCC 33090), Listeria monocytogenes(ATCC 7644), Salmonella enteritidis (ACCC 01996), Salmonella typhi-murium (CMCC 50115), Enterobacter sakazakii (ATCC 29544), Proteusmirabilis (CMCC 49003), Listeria welshimeri (ATCC 35897), Listeriagrayi (ATCC 25401), Streptococcus epidermidis (CMCC 26096), Enter-obacter aerogenes (ATCC 13048), Shigella sonnei (CMCC 51334),Streptococcus hemdyticus (CMCC 32210), Pseudomonas aeruginosa(CMCC AS1.0212), Escherichia coli (ACCC 10503), Escherichia coli(ATCC 25922), enterotoxigenic E. coli (ATCC 35401), EnteropathogenicE. coli (ATCC 43887), enteroinvasive E. coli (ATCC 43893), Vibrioalginolyticus (ATCC 17749), Proteus vulgaris (CMCC AS1.1527),Staphylococcus aureus (ATCC 29213), Vibrio parahaemolyticus (ATCC17802), Yersinia enterocolitica (ATCC 96100), Campylobacter jejuni(ATCC 33560), and Bacillus cereus (ACCC 03315).

2.2. Clinical specimens and treatment

Human sputum specimens were collected in the outpatientdepartment for pneumonia in Guangzhou Chest Hospital in southChina. Bovine lung tissue specimens collected in north China werefrom suspected infected herds positive for tuberculin skin tests.Treatment on specimens and culture followed standard methods. Inbrief, sputum and tissue specimens were subjected to microbiologicaldecontamination by processed with 4% NaOH and then neutralized,followed by centrifugation. Aliquots of latter specimens were culturein BACTEC MGIT 960 System (Becton Dickinson, Franklin Lakes, NJ,USA). The remaining portion was used for DNA extraction. The culture

tubes were kept and observed, and the negative report requiresobservation for at least 6 weeks. The culture with bacteria growthwassubjected to checking the presence of acid-fast bacilli by Ziehl-Neelsen technique. Strains were identified by para-nitrobenzoic acidand other conventional biochemical methods.

2.3. DNA extraction

Extraction and purification of mycobacterium DNA followed themethod established in our previous study (Chen et al, 2010). Thealiquot from each decontaminate specimen was suspended in 1 mL of0.01 mol/L, pH 7.6 phosphate-buffered saline, followed by vortex andmicrocentrifugation at 15000×g for 10 min. Fifty microliters of DNAextraction buffer (100mmol/L Tris–HCl of pH 8.0, 0.01% Triton X-100)was added to the pellet; the suspension was then boiled for 10 minfollowed by a brief centrifugation and further treatment with equalvolume of chloroform. The mixture was microcentrifuged, and thesupernatant was directly used as DNA template for amplification.

2.4. Primers

The primer set for genus Mycobacterium targeted at the highlyconserved regions of the genus-specific 16S ribosomal RNA gene (16SrRNA). The assay used 2 primer sets, respectively, targeting at theinsertion element IS6110 and IS1081 for identification of the MTC.Primers were designed by using the Primer PREMIER 5.0 software

Table 2Primers used in the multiplex PCR-DHPLC assay.

Organism Target gene GenBank accession Primer sequence (5’→3’) Amplicon size (bp)

MTC IS6110 Y14048 Forward: CCAACAAGAAGGCGTACTCReverse: TTGATCGTCTCGGCTAGTG

165

MTC IS1081 X84741 Forward: CGATGAGCGGTCCAATCReverse: GACGCGGCCTGCCT

379

Mycobacterium 16S rRNA FJ468345 Forward: TAACTGTGAGCGTGCGReverse: GGCACGGATCCCAA

240

55R. Chen et al. / Diagnostic Microbiology and Infectious Disease 77 (2013) 53–57

(PREMIER Biosoft International, Palo Alto, CA, USA). Sequence andamplicon information of the 3 primer sets were showed in Table 2.

2.5. Multiplex PCR-DHPLC assay

Multiplex PCR was performed by using the 3 primer sets and thefast and high-fidelity KOD DNA polymerase (TOYOBO, Osaka, Japan).The optimized 50-μL reaction mixtures contain 1 × PCR buffer, 0.2mmol/L of each dNTP, 0.2 μmol/L of each primer, 1.0 mmol/L MgSO4, 1unit KOD DNA polymerase, and purified DNA template. Amplificationswereperformed in a thermocycler (ABIVeriti Thermal Cycler, Foster, CA,USA) with the optimized cycling condition: 94 °C, 2 min (1 cycle); 35cycles of 94 °C, 15 s; 62 °C, 10 s; and 68 °C, 10 s; 72 °C, 1 min (1 cycle).

DHPLC analyses were conducted by using the Transgenomic Wave4500 systemequippedwith aDNASep Cartridge (Transgenomic, catalogno. DNA-99-3510) and UV detector. The products of the multiplex PCRwere loaded directly onto the DHPLC instrument without purification.The DHPLC running was performed under the application type of Ds-Multiple-Fragments under non-denaturing mode, with a columntemperature of 50 °C and a linear acetonitrile elution gradient formedby using WAVE buffer A, consisting of 0.1 mol/L triethylammonium(Transgenomic, StOmaha,NE,USA), andWAVEbufferB consistingof 0.1mol/L triethylammonium in 25% (vol/vol) acetonitrile (HPLC/Spectrograde; Tedia, Fairfield, OH, USA). The running method included settingnumber of segments as 3 with base pair length ranging from 100 bp to450 bp and setting flow rate as 0.9 mL/min. A total DHPLC sample runtime was 10.4 min, including cleaning and equilibration. The optimalrunning conditions are listed in Table 3. The DHPLC data were obtainedand analyzed using Navigator software version 1.6.2 (Transgenomic).

The result of a test was considered as positive forMycobacterium ifthe test showed positive for 16S rRNA. When the test also showedpositive for either IS6110 or IS1081, the result was regarded aspositive for MTC. The result was considered as positive for NTMwhenthe test only showed positive for 16S rRNA.

2.6. Determination of sensitivity

The limit of detection (LOD) on DNA template of the multiplexPCR-DHPLC assay was evaluated by using cloned plasmid DNA. Thetarget sequence of mycobacteria 16S rRNA andMTC insertion elementIS6110 and IS1081was, respectively, cloned into pEASY-Blunt CloningVector (Beijing Transgen Biotech., Beijing, China). Cloned plasmid

Table 3Optimized condition for detection of mycobacteria using the Transgenomic Wave4500 system.

Gradient Time (min) % Buffer A % Buffer B

— 0.0 55.0 45.0100 bp (3.5 min) 0.5 50.2 49.8265 bp (6.9 min) 4.6 40.4 59.6430 bp (9.8 min) 8.8 36.9 63.1Start Clean 8.9 0.0 0.0Stop Clean 9.4 0.0 0.0Start Equilitrate 9.5 55.0 45.0Stop Equilibrate 10.4 55.0 45.0

DNA was extracted, the concentration at A260 was measured by usinga spectrophotometer (NanoDrop, Wilmington, DE, USA), and thegenome copies of the templates were calculated based on theconcentration and molecular weight of the plasmid. Serial 10-folddiluted cloned plasmid DNA samples were tested in duplicates by themultiplex assay, and the concentrations and genome copies of the endpoint DNA samples were calculated.

3. Results

3.1. Specificity of the multiplex PCR-DHPLC assay

Based on online BLAST search (http://blast.ncbi.nlm.nih.gov/Blast.cgi) on 10,000 nucleic acid sequences, the sequence of the primer setfor 16S rRNA aligned with sequences from more than 100mycobacteria species. The BLAST analysis also revealed that sequenceof the primer sets for IS6110 and IS1081 exclusively align with MTCstrains sequences.

The specificity of the multiplex PCR-DHPLC assay was validated bydetection on 84 strains covering 23 mycobacteria species listed inTable 1. The detection on MTC strains, including 1 reference and 41isolated M. tuberculosis strains, 1 reference and 9 isolated strains ofMycobacterium bovis, and 1 strain of M. bovis bacilli calmette guerin(BCG) vaccine, all showed positive of the 3 specific amplicons byDHPLC analysis, including the 165 bp amplicon of IS6110, the 379 bpamplicon of IS1081, and the 240 bp amplicon of the 16S rRNA. Thedetection on the 21 species of NTM strains listed in Table 1 onlydemonstrated the 240 bp amplicon of the 16S rRNA. Detection on the30 non-mycobacteria species revealed no cross-reaction. Fig. 1 andFig. 2, respectively, demonstrated the DHPLC analysis of ampliconsfrom MTC and NTM strains after multiplex PCR.

3.2. Sensitivity of the assay

The LOD of the multiplex PCR-DHPLC was evaluated by testing onserial diluted cloned DNA templates as described in above. The LOD

Fig. 1. DHPLC peak profiles obtained from M. tuberculosis strains (n = 3) and M. bovisstrains (n = 3). Horizontal axis: elution time (min); vertical axis: fluorescence (mV)negative: negative strains.

;

Fig. 2. DHPLC peak profiles obtained from NTM strains. Horizontal axis: elution time(min); vertical axis: fluorescence (mV); negative: negative strains.

56 R. Chen et al. / Diagnostic Microbiology and Infectious Disease 77 (2013) 53–57

for targeted gene sequence IS6110, IS1081, and 16S rRNA by the assaywas 0.8, 1.0, and 1.6 pg per reaction, respectively, equivalent to 190,220, and 370 genomic copies per reaction, respectively.

3.3. Detection on clinical specimens and comparison with culture

The clinical usefulness of the multiplex PCR-DHPLC assay wasevaluated by detection on human sputum and bovine tissuespecimens and comparison to culture.

As showed in Table 4, the multiplex PCR-DHPLC detected 31.0%(48/155) positive for MTC and 4.5% (7/155) positive for NTM inhuman sputum specimens, compared to culture with 9.0% (14/155)positive for MTC and 2.6% (4/155) positive for NTM. The molecularassay detected all sputum samples positive by culture, and the 2methods were in agreement on identifying MTC and NTM.

The detection on bovine specimens revealed that the multiplexPCR-DHPLC also detected more positive than culture. Totally, 43bovine lung specimens were collected from cattle positive bytuberculin tests, the multiplex PCR-DHPLC detected 38 (88.4%)samples positive as MTC, compared to culture with 18 (41.9%)samples determined as positive. Samples determined as positive byculture were all confirmed positive by the multiplex PCR-DHPLC. NoNTM positive was detected by both methods in bovine specimens.

In summary, for 198 human and bovine specimens tested in thisstudy, the multiplex PCR-DHPLC assay detected all culture-positive(36 in number) and 57 in 162 (35.2%) culture-negative specimens.

4. Discussion

Currently, mycobacteria research using DHPLC technology mainlyfocused on differentiation and identification of drug-resistant MTCstrains, which were based on detection of gene mutation underdenaturing mode (Evans et al., 2009; Shi et al., 2007; Wilson., 2011).Combination of PCR and non-DHPLC for typing of M. tuberculosisstrains was first reported in 2004 (Evans et al, 2004). According to our

Table 4Detection of mycobacteria on sputum specimens (n = 155) by the multiplex PCR-DHPLC and comparison with culture.

DHPLC Culture

MTC positivea NTM positiveb Negative

MTC positivea 14 0 34NTM positiveb 0 4 3Negative 0 0 100Total 14 4 137

a Number of samples determined as positive of MTC.b Number of samples determined as positive of NTM.

current knowledge, application of multiplex PCR combined withDHPLC for identification of mycobacterium genus and simultaneousdifferentiation of tuberculous and nontuberculous mycobacteria hasnot been reported. The assay reported here was verified by variousmycobacterial species and species of non-mycobacterial microorgan-ism. The clinical usefulness of the assay was demonstrated byidentifying mycobacteria directly in human and cattle specimens.

This assay used genus-specific primer set targeted at the 16S rRNAgene of mycobacteria. The 16S rRNA gene is highly conserved andcontains genus- and species-specific sequence variations useful formicroorganism identification purpose. Molecular methods targeted atthe 16S rRNA gene of mycobacteria have successfully identified MTCand several nontuberculous species (Foongladda et al., 2009; Kox etal., 1997; Richardson et al., 2009). This study provided genus-specificresolution. Currently, the primer set has been verified by testing on 23species of mycobacterial reference strains; though online BLASTanalysis has confirmed that the primer set covered more than 100mycobacteria species, further verification by testing should proceedand requires collection of more mycobacteria species.

Identification of MTC by the assay was achieved by targeting at 2specific insertion gene elements. The 2 insertion elements, IS6110and IS1081, were widely recognized as marker gene sequence ofMTC strains. M. tuberculosis strains commonly carried high copynumbers of IS6110 but only single or few copies of IS1081, while thegenome of bovine tuberculosis strains usually contains more copiesof IS1081 than IS6110 (Cai et al., 1999; Dziadek et al., 2001). Isolatestrains of Mycobacterium tuberculosis with no or low copy numbersof IS6110 have been found (Warren et al., 2004; Yuen et al., 1993).Therefore, we considered using both the insertion gene elements astargets would enhance clinical detection of MTC in both human andanimal specimens.

The assay demonstrated good performance on 84 strains covering23 mycobacteria species and 30 strains and species of non-mycobacteria, with no cross-reaction between MTC and nontubercu-lous strains or between mycobacteria and non-mycobacteria micro-organisms. According to the clinical application on 155 humansputum and 43 cattle tissue specimens, the molecular assay detectedall specimens positive by culture (36 in number) and detected 35.2%positive from culture-negative specimens. To further analyzed theextra positives detected by the DHPLC assay, DNA samples fromspecimens negative by culture but positive asMTC by the DHPLC assaywere subjected to more sensitive real-time PCR detection, and allthese DNA samples, with 34 human sputum and 20 cattle tissues innumber, were detected as MTC positive by standard real-time PCRmethods targeted at IS6110 or IS1081 gene element (Chen et al,2011). Due to the lack of additional method specific for NTM, the 3extra positive NTM samples were not further tested. Although it couldnot be concluded that the extra positives were true positive, theadditional detection as positive of the samples by another molecularmethod should further prove the accuracy of the DHPLC assay.Molecular assay targets at nucleic acid without the necessity for livepathogen and fresh specimens; this advantage may contribute to thehigher detection of positive than culture in this study.

DHPLC technique has been proved to be a reliable tool for high-throughput and cost-effective nucleic acid analysis by studiesworldwide. It provides rapid, fully automatic, and accurate separationand detection of PCR amplified gene fragments and avoids laboriousand time-consuming gel-electrophoresis handling. The DHPLC systemintegrating with fragment collector allows collection of genefragments ready for downstream applications such as cloning andsequencing. The DHPLC-based assay developed in this study demon-strated increased sensitivity, accuracy in sizing of gene fragments, andreproducibility over gel-based conventional PCR methods. Accordingto our current study, although the multiplex PCR DHPLC assay wasless sensitive than real-time PCR assay in single target detection (datanot showed), it was with the advantage of multiplex detection,

57R. Chen et al. / Diagnostic Microbiology and Infectious Disease 77 (2013) 53–57

convenience in handling large number of samples, and capability ofcollecting amplicon for further verification and application. The assayenabled rapid detection, which took less than 1 h on purified DNAincluding 10 min for DHPLC running of each sample, and the wholeprocedure including sample treatment and DNA extraction could becompleted within 1 day.

The practical value of the assay described here includes usefulnessin combined application with other technology in clinical diagnosis.For example, the assay could be used as screening method beforeculture to reduce laborious culture work and detection time; it couldalso be used after culture to rapidly identified MTC and NTM strains.

In summary, in addition to rapid identification of Mycobacteriumand differentiation of MTC and NTM strain from positive cultures, theassay presented here enables fast and reliable detection on clinicalspecimens. It should be helpful for rapid screening of tuberculosis andother mycobacterial infection in human and animal.

Acknowledgment

This work was supported by a research project (Code: 2010IK020)from the General Administration of Quality Supervision, Inspectionand Quarantine of the People's Republic of China.

We thank Prof Bao-Wen Chen and Mrs Ru-Su Xu for providingbacterial samples.

References

Anilkumar AK, Madhavilatha GK, Paul LK, Radhakrishnan I, Kumar RA, Mundayoor S.Standardization and evaluation of a tetraplex polymerase chain reaction to detectand differentiate Mycobacterium tuberculosis complex and nontuberculous Myco-bacteria—a retrospective study on pulmonary TB patients. Diagn Microbiol InfectDis 2012;72(3):239–47.

Cai H, Li JL, Wu J, Zhang M, Chen Y. Species identification of Mycobacterium tuberculosiswith insertion elements. Acta Microbiol Sini 1999;39(2):120–6.

Chen R, Bi Y, Yang G, Liu Z, Liu Z, Zeng B, Tong T. Development of a fluorescentmicrosphere-based multiplex assay for simultaneous rapid detection of Mycobac-terium tuberculosis complex and differentiation of M. tuberculosis and M. bovis inclinical samples. Diagn Mol Pathol 2010;19(3):172–9.

Chen R, Liu ZY, Yang GH, Guo AZ, Zeng BJ, Han HQ, et al. National standard of the People'sRepublic of China: real time PCR method for the detection of tuberculosis pathogenicorganisms GB/T 27639–2011. 1st ed. Beijing: China Standard Press; 2011.

Duanmu HJ. National Technic Steering Group of the Epidemiological Sampling Surveyfor Tuberculosis. Report on fourth national epidemiological sampling survey oftuberculosis. Chin J Tuberc Respir Dis 2002;25(1):3–7.

Dziadek J, Sajduda A, Borun TM. Specificity of insertion sequence-based PCR assays forMycobacterium tuberculosis complex. Int J Tuberc Lung Dis 2001;5(6):569–74.

Evans JT, Hawkey PM, Smith EG, Boese KA, Warren RE, Hong G. Automated high-throughput mycobacterial interspesed repetitive unit typing of Mycobacterium

tuberculosis strains by a combination of PCR and nondenaturing high-performanceliquid chromatography. J Clin Microbiol 2004;42(9):4175–80.

Evans JT, Parveen A, Smith GE, Xu L, Chan EW, Chan RC, Hawkey PM. Application ofdenaturingHPLC to rapidly identify rifampicin-resistantMycobacterium tuberculosis inlow- and high-prevalence areas. J Antimicrob Chemother 2009;63(2):295–301.

Foongladda S, Pholwat S, Eampokalap B, Kiratisin P, Sutthent R. Multi-probe real-timePCR identification of common Mycobacterium species in blood culture broth. J MolDiagn 2009;11(1):42–8.

Glassroth J. Pulmonary disease due to nontuberculous mycobacteria. Chest 2008;133:243–51.

Goldenberg O, Herrmann S, Adam T, Marjoram G, Hong G, Göbel UB, Graf B. Use ofdenaturing high-performance liquid chromatography for rapid detection andidentification of seven Candida species. J Clin Microbiol 2005;43(12):5912–5.

Hurtle W, Bode E, Kaplan RS, Garrison J, Kearney B, Shoemaker D, et al. Use ofdenaturing high-performance liquid chromatography to identify Bacillus anthracisby analysis of the 16S-23S rRNA interspacer region and gyrA gene. J Clin Microbiol2003;41(10):4758–66.

Issa R, Mohd Hassan NA, Abdul H, Hashim SH, Seradja VH, Abdul Sani A. Detection anddiscrimination of Mycobacterium tuberculosis complex. Diagn Microbiol Infect Dis2012;72(1):62–7.

Kox LF, Jansen HM, Kuijper S, Kolk AH. Multiplex PCR assay for immediate identificationof the infecting species in patients with mycobacterial disease. J Clin Microbiol1997;35(6):1492–8.

Li J, Gerhard DS, Zhang Z, Huettner PC, Wright J, Nguyen L, et al. Denaturing high-performance liquid chromatography for detecting and typing genital humanpapillomavirus. J Clin Microbiol 2003;41(12):5563–71.

Park H, Jang H, Song E, Chang CL, Lee M, Jeong S, et al. Detection and genotyping ofMycobacterium species from clinical isolates and specimens by oligonucleotidearray. J Clin Microbiol 2005;43(4):1782–8.

Reddington K, O’Grady J, Dorai-Raj S, Maher M, van Soolingen D, Barry T. Novelmultiplex real-time PCR diagnostic assay for identification and differentiation ofMycobacterium tuberculosis, Mycobacterium canettii, and Mycobacterium tuberculo-sis complex strains. J Clin Microbiol 2011;49(2):651–7.

Richardson ET, Samson D, Banaei N. Rapid identification of Mycobacterium tuberculosisand nontuberculous mycobacteria by multiplex, real-time PCR. J Clin Microbiol2009;47(5):1497–502.

Shi R, Zhang J, Li C, Kazumi Y, Sugawara I. Detection of streptomycin resistance inMycobacterium tuberculosis clinical isolates from China as determined by denaturingHPLC analysis and DNA sequencing. Microbes Infect 2007;9(14–15):1538–44.

Troesch A, Nguyen H, Miyada CG, et al. Mycobacterium species identification andrifampin resistance testing with high-density DNA probe arrays. J Clin Microbiol1999;37:49–55.

Troedsson C, Lee RF, Stokes V, Walters TL, Simonelli P, Frischer ME. Development of adenaturing high-performance liquid chromatography method for detection ofprotist parasites of metazoans. Appl Environ Microbiol 2008a;74(14):4336–45.

Troedsson C, Lee RF, Walters T, Stokes V, Brinkley K, Naegele V, Frischer ME. Detectionand discovery of crustacean parasites in blue crabs (Callinectes sapidus) by using18S rRNA gene-targeted denaturing high-performance liquid chromatography.Appl Environ Microbiol 2008b;74(14):4346–52.

Warren RM, Victor TC, Streicher EM, Richardson M, van der Spuy GD, Johnson R, et al.Clonal expansion of a globally disseminated lineage of Mycobacterium tuberculosiswith low IS6110 copy numbers. J Clin Microbiol 2004;42(12):5774–82.

Wilson ML. Recent advances in the laboratory detection of Mycobacterium tuberculosiscomplex and drug resistance. Clin Infect Dis 2011;52(11):1350–5.

Yuen LKW, Ross BC, Jackson KM, et al. Characterization of Mycobacterium tuberculosisstrains from Vietnamese patients by southern blot hybridization. J Clin Microbiol1993;31(6):1615–8.