JOURNAL OF CLINICAL MICROBIOLOGY, June 1993, p. 1615-16180095-1137/93/061615-04$02.00/0Copyright X) 1993, American Society for Microbiology

Characterization of Mycobacterium tuberculosis Strains fromVietnamese Patients by Southern Blot HybridizationLILLY K. W. YUEN,* BRUCE C. ROSS, KATHY M. JACKSON, AND BRIAN DWYER

Clinical Pathology Laboratory, Fairfield Infectious Diseases Hospital, Fairfield, Victoria 3078, Australia

Received 12 November 1992/Accepted 2 March 1993

A total of 41 Mycobacterium tuberculosis strains from patients of Vietnamese origin were analyzed bySouthern blot hybridization with two different probes, IS6110 (Otal, I., et al., J. Clin. Microbiol. 29:1252-1254, 1991; Ross, B. C., et al., J. Clin. Microbiol. 30:942-946, 1992; Thierry, D., et al., J. Clin. Microbiol.28:2668-2673, 1990; van Soolingen, D., et al., J. Clin. Microbiol. 29:2578-2586, 1991) and pTBN12 (Ross,B. C., et al., J. Clin. Microbiol. 30:942-946, 1992). The restriction fragment patterns of nine of these strainswere virtually identical when the pTBN12 probe was used; five strains had a single copy of IS6110, and fourstrains failed to hybridize with the IS6110 probe. This relatively high frequency of strains with no or one copyof IS6110 suggests that the usefulness of IS6110 for epidemiological study may be limited in certain populations.

Despite the decline in the incidence of tuberculosis inhighly industrialized nations over the past 50 years, thedisease remains a major health problem in the world today.In Australia alone, approximately 600 new cases are stilldiagnosed annually (2).As part of efforts to further reduce the incidence of

tuberculosis, epidemiological investigations play an impor-tant role in the monitoring of control for this disease. Theavailability of a rapid, simple, and discriminatory typingmethod for Mycobacterium tuberculosis can be of greatbenefit to such investigations. Studies have shown thatrestriction fragment length polymorphism analysis of M.tuberculosis can detect genotypic variations between strainsby using repetitive DNA sequences as probes in Southernhybridization analyses (7, 10, 14).A DNA fragment containing a repetitive DNA sequence

has been cloned in plasmid pTBN12 from M. tuberculosis(10). This repeated sequence is highly repetitive in nature,and it has been estimated that at least 30 copies are presentin the genome. They are not confined to the M. tuberculosiscomplex but are present in a number of other mycobacterialspecies (10). By using pTBN12 as a probe, strains fromwithin known clusters of infection demonstrate identicalbanding patterns whilst nonrelated strains are different.Two insertion-sequence elements related to the enterobac-

terial IS3 family, IS6110 and IS986, have also been molecu-larly cloned from the genome of M. tuberculosis (7, 10, 12,14). Sequence comparisons have shown that IS6110 andIS986 differ from each other by only three nucleotides (7, 10,14). Although they are probably mobile genetic elementswith the potential to translocate within the chromosome,they are stable enough to provide identical restriction frag-ments for mycobacterial strains from within small clusters ofinfection (6, 7, 14). Previous studies have reported that allM. tuberculosis strains have at least one copy and usually 6to 15 copies of IS6110 present in their genome (7, 14).

Recently, we have found a group of M. tuberculosisstrains from two states in Australia with virtually identicalpTBN12 restriction fragment patterns which appear to haveeither one or no copies of IS6110. Most of these strainsappeared to be isolated from patients of Vietnamese origin.

* Corresponding author.

In this study, we investigate this finding by analyzing a

number of strains from patients of Vietnamese origin incomparison to randomly selected isolates from patients ofnon-Vietnamese origin.A total of 79 clinical isolates of M. tuberculosis were

selected for this study from the following reference labora-tories in Australia: Mycobacterium Reference Laboratory,Fairfield Hospital, Victoria; State Health Laboratory, Bris-bane, Queensland; Westmead Centre, Westmead, NewSouth Wales; and State Health Laboratory Services, Ned-lands, Western Australia. All isolates were identified as M.tuberculosis by standard procedures in the various referencelaboratories on the basis of positive nitrate reduction, niacinproduction, pyrazinamidase production, and/or resistance to5 jig of thiophene-2-carboxylic acid hydrazide per ml. Therewere no known epidemiological relationships between thesepatients except for a father and daughter. Of the 79 isolates,41 were from patients with Vietnamese origin, and as con-

trols, 38 isolates were randomly selected from patients not ofVietnamese origin. These control strains were isolated inVictoria, Queensland, and Western Australia. All M. tuber-culosis isolates received from interstate were retested in ourlaboratory and confirmed as M. tuberculosis.

Mycobacterial strains were grown and DNA was ex-tracted as previously described (10). The concentration ofDNA was estimated by agarose gel electrophoresis stainedwith ethidium bromide before restriction endonuclease di-gestion and Southern blot hybridization as described previ-ously (10). RNA and DNA probes used were digoxigeninlabelled with a nonradioactive digoxigenin nucleic acid la-belling and detection system supplied by Boehringer GmbH(Mannheim, Germany). Chromosomal DNA digested withAluI was hybridized with an RNA probe prepared fromplasmid pTBN12 (10). DNA digested with PvuII was hybrid-ized with a DNA probe prepared from a 123-bp polymerasechain reaction (PCR) product of the IS6110 element (10).

Slot blot analyses were performed as previously described(9). The same amount of DNA for each sample was appliedto duplicate nylon filters in the respective slots, baked at120°C, and hybridized with either a digoxigenin labelled65-kDa antigen gene probe (13) or an IS6110 probe.DNA used for PCR was extracted by sonication as de-

scribed previously (1). Briefly, a colony of M. tuberculosiswas removed with a bacterial loop and suspended in 200 ,ul

1615

Vol. 31, No. 6

on August 21, 2020 by guest

http://jcm.asm

.org/D

ownloaded from

J. CLIN. MICROBIOL.

23.1 kb-9.4kb -6.6 kb -

9.4 kb -

4.4 kb-

40.

6-6 kb -

4.4 kb-

2.3 kb-

2.0 kb -

2.0 kb -

1.7 kb- -S

1.0 kb -a

_I -

0.6

FIG. 1. Southern blot analysis of M. tuberculosis strains afterdigestion of chromosomal DNA withAlul and hybridization with anRNA probe directed against the cloned repeated element pTBN12.Lane 1, H37R, (ATCC 27294); lanes 2 to 10, M. tuberculosis strainsisolated from Vietnamese patients. The arrow in lane 8 indicates aband difference. The locations of molecular weight standards aremarked to the left.

of distilled water before centrifugation at 12,000 x g for 5min and resuspension in 50 ,ul of distilled water. The tubewas floated in a water bath containing the sonicator probe(Measuring and Scientific Equipment, Crawley, England)and sonicated for 10 min at 150 W. After sonication, sampleswere boiled for 10 min and centrifuged for 20 s at 12,000 x g,and the supernatant was collected. DNA samples wereamplified in a buffer supplied by the Taq DNA polymerasemanufacturer (Promega, Madison, Wis.) in a 20-,ul reactionmixture containing approximately 100 ng of DNA, 1 ,uMprimers, 1 U of Taq DNA polymerase, 100 puM dATP, 100,uM dCTP, 100 puM dGTP, and 100 puM dTTP (10). Theprimers TB1 and TB2 were designed to amplify a 123-bpfragment of IS6110 (4). The reactions were performed in anautomated thermal cycler (Hybaid) with a program consist-ing of an initial denaturation at 95°C for 3 min and then 35cycles of a two-step PCR of annealing and extension for 1min at 65°C and denaturation for 1 min at 94°C. The PCRproducts were then analyzed by electrophoresis through a10% polyacrylamide gel and stained with a silver staining kit(Bio-Rad Laboratories, Richmond, Calif.) according to themanufacturer's instructions.

Sequencing of PCR product was performed by the directincorporation protocol with aftnol DNA sequencing system(Promega) according to the manufacturer's instructions andwith the PCR primer TB1.The analysis of nine M. tuberculosis isolates from patients

of Vietnamese origin by Southern blot hybridization with theprobe pTBN12 is shown in Fig. 1. The DNA profilesproduced from the patient isolates (lanes 2 to 10) are

virtually identical. The only apparent variation was that of asingle band difference in one strain marked with an arrow inlane 8. The profile produced by the type strain H37Rv (ATCC

0.6kb -0.5kb -

1 2 3 4 5 6 7 8 9 10

FIG. 2. Southern blot analysis of M. tuberculosis strains afterdigestion of chromosomal DNA with PvuII and hybridization with aDNA probe directed against a 123-bp fragment of the IS6110element. Lane 1, H37R, (ATCC 27294); lanes 2 to 10, M. tubercu-losis strains from Vietnamese patients. The locations of molecularweight standards are to the left.

27294) in lane 1 is clearly different from the patient isolates.The similarity between these patient strains suggests anepidemiological relationship.To verify their relatedness, these same nine isolates were

digested with PvuII and hybridized with the IS6110 probe(Fig. 2). Since the 123-bp PCR product of the IS6110 elementused in this DNA probe lacks an internal PvuII site, eachhybridization band represents one copy of IS6110. TheH37Rv strain (lane 1) produced multiple copies of IS6110. Incontrast, no bands were visible in four of the Vietnameseisolates (lanes 2 to 5), and the remaining five isolates (lanes6 to 10) demonstrated one band, implying that these strainshave one copy of IS6110. Of the isolates with a single copyof IS6110, four produced a 1.3-kb fragment (lanes 5, 6, 7, and9) and one produced a 6.0-kb fragment (lane 8). The isolatewith the different size fragment is the same strain with a banddifference with the pTBN12 probe (Fig. 1, lane 8).To confirm the Southern blot result, slot blot analysis was

performed with DNA probes directed against the mycobac-terial common 65-kDa antigen gene and IS6110 (Fig. 3). Allmycobacterial strains studied were positive with the 65-kDaantigen probe (Fig. 3A), indicating that sufficient DNA wasapplied to the membrane to yield a positive result. The fourisolates which appeared to lack IS6110 by Southern blothybridization also failed to hybridize to IS6110 by slot blothybridization (Fig. 3B). Samples with a single copy andmultiple copies were positive with this probe, while speciesoutside the M. tuberculosis complex were all negative.PCR was performed as a final analysis on the four isolates

which lack IS6110 (Fig. 4). The four isolates tested failed toproduce a 123-bp fragment (lanes 3 and 4). Two of theseisolates produced a 280-bp fragment (lanes 5 and 6) whichwas sequenced to confirm that it was not related to IS6110(results not shown).

1616 NOTES

23.1 kb -

,;jillll

6 7 8 9 10

on August 21, 2020 by guest

http://jcm.asm

.org/D

ownloaded from

NOTES 1617

A B

9 1

2

3

-- 10 2

_ .. 11 3

4 12 4

5 13 5

6 14 6 -

7 15 7

8 8is; At\>

FIG. 3. Slot blot analyses of mycobacterial strasomal DNA hybridized with a 65-kDa antigen ge:IS6110 probe (B). For each sample, the same am(applied to the respective slot on the two filters. S;tuberculosis strains from Vietnamese patients; s(ATCC 27294); sample 11, clinical isolate of Mycsample 12, clinical isolate of M. bovis BCG; saisolate of Mycobacterium gastri; sample 14, Myvescens (NCTC 10271); sample 15, Mycobacteriun10268).

To determine if this group of nine isolate,tative of a generally low copy number of IS61Vietnamese origin, a further 32 strains weSouthern hybridization and compared withrandomly selected isolates of non-Vietnamestically, M. tuberculosis isolates from Vietnhave an average of 8 + 7 copies of IS6110.isolates from non-Vietnamese patients have E+ 4 copies of IS6110. This suggests thatcopies of IS6110 found in isolates from t

900 bp-

501 bp-

320 bp-

190 bp-

123 bp -

67 bp -

population tend to be fewer on average and have a widervariation in IS6110 copy number distribution than isolatesfrom the non-Vietnamese population.The results of this study demonstrate that IS6110 is not

9 present in the genome of all strains of M. tuberculosis. Thefour isolates without this element and the five strains which

10 have one copy of IS6110 appear to be genetically related onthe basis of their pTBN12 Southern blot profiles (Fig. 1). Incontrast, the other 31 Vietnamese strains with multiplecopies of IS6110 had variable Southern blot patterns with the

..12 pTBN12 probe (results not shown).

13 It is possible that the similarity of the pTBN12 Southernblot patterns is related to the copy number of IS6110,

14 although an epidemiological relationship between these pa-tients cannot be ruled out. However, it is reasonable to

15 suggest that the movement of IS6110 around the genome ofM. tuberculosis contributes to the genetic heterogeneityobserved with the pTBN12 Southern blots. Therefore, thelower the copy number of IS6110 in aM. tuberculosis strain,the less genetic diversity there may be over time. Thus, a

ains with chromo- strain may persist in a particular geographical location suchne probe (A) and as in Vietnam for a long time without significant alteration toount of DNA was its Southern blot profile because of the low copy number ofample 10,Hto I, IS6110. This supposition is supported by finding one strain;bacterum bovis; with a single copy of IS6110 which appears to be in a locationLmple 13, clinical different from those of the other single-copy strains (Fig. 2,ycobacterium fla- lane 8). This strain also has a single-band difference byn kansasii (NCTC Southern blot with a pTBN12 probe (Fig. 1, lane 8). This

result implies that in certain regions of the world, Southernblot hybridization may not be useful as an epidemiologicaltool.

s was represen- Published data have reported that IS986, which is virtually'10 in patients of identical to IS6110, can remain stable after 2 months ofre analyzed by passages in guinea pigs (6) as well as in vitro passaging ina group of 38 liquid medium every week for 6 months (14). Conversely, in

;e origin. Statis- our laboratory we have found strains with differences in oneiamese patients or two bands with IS6110 which appear identical withIn comparison, pTBN12 for multiple isolates from the same patient oran average of 12 isolates from a known cluster of infection (3, 8). For thisthe number of reason, our laboratory employs the probe pTBN12 forthe Vietnamese primary screening of isolates for epidemiologic relatedness

and the IS6110 probe for confirmation and/or subtyping ofstrains.Our results also have implications for the application of

PCR in laboratory diagnosis. PCR has the potential toprovide a more rapid, sensitive, and specific diagnostic assayfor M. tuberculosis in clinical specimens (1, 4). This studydemonstrates that not all M. tuberculosis strains can bedetected with primers directed towards IS6110. This canlead to false-negative results which, according to our data,would occur more frequently in the group of Vietnamesepatients reported in this study. Potentially this would presentproblems in the application of PCR diagnosis and Southernblot hybridization for epidemiology in areas where suchstrains have a high prevalence.

1 2 3 4 5 6

FIG. 4. PCR for the amplification of the 123-bp segment ofIS6110 with primers TB1 and TB2. DNA bands were visualized bysilver staining. Lane 1, M. tuberculosis H37Rk; lane 2, negativecontrol; lanes 3 to 6, M. tuberculosis strains negative for IS6110 bySouthern and slot blot analyses. The locations of molecular weightstandards are marked on the left.

We express our thanks to Aina Sievers of the MycobacteriumReference Laboratory, Fairfield Hospital, Victoria; David Dawsonof State Health Laboratory, Brisbane, Queensland; William Chewof Westmead Centre, Westmead, New South Wales; and Beth Gowand Frank Haverkort of the State Health Laboratory Services,Nedlands, Western Australia, for providing us with the M. tubercu-losis strains.

This work was funded by grant 920503 from the National Healthand Medical Research Council, Australia.

VOL. 31, 1993

1

on August 21, 2020 by guest

http://jcm.asm

.org/D

ownloaded from

J. CLIN. MICROBIOL.

REFERENCES1. Buck, G. E., L. C. O'Hara, and J. T. Summersgill. 1992. Rapid,

simple method for treating clinical specimens containing Myco-bacterium tuberculosis to remove DNA for polymerase chainreaction. J. Clin. Microbiol. 30:1331-1334.

2. Dawson, D., P. Anargyros, Z. Blacklock, W. Chew, H. Dagnia,B. Gow, K. Jackson, and A. Sievers. 1991. Tuberculosis inAustralia: an analysis of cases identified in reference laborato-ries in 1986-88. Pathogenesis 23:130-134.

3. Dwyer, B., K. Jackson, K. Raios, A. Sievers, E. Wilshire, and B.Ross. 1993. DNA restriction fragment analysis to define anextended cluster of tuberculosis in homeless men and theirassociates. J. Infect. Dis. 167:490-494.

4. Eisenach, K. D., M. D. Cave, J. H. Bates, and J. T. Crawford.1990. Polymerase chain reaction amplification of a repetitiveDNA sequence specific for Mycobacterium tuberculosis. J.Infect. Dis. 161:977-981.

5. Eisenach, K. D., M. D. Sifford, M. D. Cave, J. H. Bates, andJ. T. Crawford. 1991. Detection of Mycobacterium tuberculosisin sputum samples using a polymerase chain reaction. Am. Rev.Respir. Dis. 144:1160-1163.

6. Hermans, P. W. M., D. van Soolingen, J. W. Dale, A. R. J.Schuitema, R. A. McAdam, D. Catty, and J. D. A. van Embden.1990. Insertion element IS986 from Mycobacterium tuberculo-sis: a useful tool for diagnosis and epidemiology of tuberculosis.J. Clin. Microbiol. 28:2051-2058.

7. Otal, I., C. Martin, V. Vincent-Levy-Frebault, D. Thierry, andB. Gicquel. 1991. Restriction fragment length polymorphismanalysis using IS6110 as an epidemiological marker in tubercu-

losis. J. Clin. Microbiol. 29:1252-1254.8. Ross, B. C., and B. Dwyer. 1993. Rapid, simple method for

typing isolates of Mycobacterium tuberculosis by using thepolymerase chain reaction. J. Clin. Microbiol. 31:329-334.

9. Ross, B. C., K. Jackson, M. Yang, A. Sievers, and B. Dwyer.1992. Identification of a genetically distinct sub-species ofMycobacterium kansasii. J. Clin. Microbiol. 30:2930-2933.

10. Ross, B. C., K. Raios, K. Jackson, and B. Dwyer. 1992. Molec-ular cloning of a highly repeated DNA element from Mycobac-terium tuberculosis and its use as an epidemiological tool. J.Clin. Microbiol. 30:942-946.

11. Ross, B. C., K. Raios, K. Jackson, A. Sievers, and B. Dwyer.1991. Differentiation of Mycobacterium tuberculosis strains byuse of a nonradioactive southern blot hybridization method. J.Infect. Dis. 163:904-907.

12. Thierry, D., A. Brisson-Noel, V. Vincent-Levy-Frebault, S.Nguyen, J. Guesdon, and B. Gicquel. 1990. Characterization of aMycobacterium tuberculosis insertion sequence, IS6110, and itsapplication in diagnosis. J. Clin. Microbiol. 28:2668-2673.

13. Thole, J. E. R., and R. van der Zee. 1990. The 65kD antigen:molecular studies on a ubiquitous antigen, p. 37-67. In J.McFadden (ed.), Molecular biology of the mycobacteria-1990.Surrey University Press, London.

14. van Soolingen, D., P. W. M. Hermans, P. E. W. de Haas, D. R.Soll, and J. D. A. van Embden. 1991. Occurrence and stability ofinsertion sequences in Mycobacterium tuberculosis complexstrains: evaluation of an insertion sequence-dependent DNApolymorphism as a tool in the epidemiology of tuberculosis. J.Clin. Microbiol. 29:2578-2586.

1618 NOTES

on August 21, 2020 by guest

http://jcm.asm

.org/D

ownloaded from