the use of macroarrays for the identification of mdr

7/30/2019 The Use of Macroarrays for the Identification of MDR

1/7

The use of macroarrays for the identification of MDR

Mycobacterium tuberculosis

T.J. Brown a,*, L. Herrera-Leon b, R.M. Anthony c, F.A. Drobniewski a

a Health Protection Agency Mycobacterium Reference Unit, Kings College Hospital (Dulwich), East Dulwich Grove, London SE22 8QF, UKb

Laboratorio de Referencia de Micobacterias, Servicio de Bacteriologia, Centro Nacional de Microbiologia, Instituto de Salud Carlos III,

Madrid, Spainc Biomedical Research, KIT (Royal Tropical Institute), Meibergdreef 39, 1105 AZ, Amsterdam, The Netherlands

Received 7 June 2005; received in revised form 5 August 2005; accepted 15 August 2005Available online 8 September 2005

Abstract

The emergence of Mycobacterium tuberculosis (Mtb), resistant to both isoniazid (INH) and rifampicin (RIF) (MDR-TB), is an

increasing threat to tuberculosis control programs. Susceptibility testing of Mtb complex isolates by phenotypic methods requires a

minimum of 14 days from a primary specimen. This can be reduced significantly if molecular analysis is used. Low density

oligonucleotide arrays (macroarrays) have been used successfully for the detection of RIF resistance in Mtb. We describe the use of

macroarray technology to identify Mtb complex isolates resistant to INH and/or RIF. The macroarray MDR-Mtb screen has been

designed to detect mutations in the RIF resistance determining region (RRDR) of Mtb rpoB and loci in katG and mabA-inhA

associated with INH resistance. A panel of Mtb isolates containing 38 different RRDR genotypes, 4 different genotypes withincodon 315 of katG and 2 genotypes at mabA-inhA was used to validate the macroarray. The wild type (WT) genotype was

correctly identified at all three loci. Of the 37 mutant rpoB genotypes, 36 were correctly detected; the single mutant not detected

contained a 9 base insertion. All mutations within katG and mabA-inhA were correctly identified. We conclude that this low cost,

rapid system can usefully detect the mutations associated with the vast majority of MDR-Mtb.

D 2005 Elsevier B.V. All rights reserved.

Keywords: Macroarray; MDR; Mycobacterium tuberculosis

1. Introduction

The WHO estimates that up to a third of the worldspopulation is infected with Mycobacterium tuberculosis

and globally someone dies of tuberculosis (TB) every

1 5 s . (www.who.int/gtb/publications/globrep). Where

TB is diagnosed in a timely manner a highly effective

multiple drug treatment regimen can be used, the two

most important constituents being isoniazid (INH) and

rifampin (RIF). Multidrug-resistant (MDR) strains,which are defined as being resistant to at least INH

and RIF, are emerging and worryingly can retain their

virulence as demonstrated by reports of their involve-

ment in several institutional and community outbreaks

(Bifani et al., 1996; Davies, 2003; Narvskaya et al.,

2002). The development of resistance to these two

drugs reduces the efficacy of standard anti-TB treat-

ment, increasing the rate of treatment failure (Quy et al.,

2003) and by inference the risk of transmission. Iden-

tification of these strains allows initiation of modified

0167-7012/$ - see front matterD

2005 Elsevier B.V. All rights reserved.doi:10.1016/j.mimet.2005.08.002

* Corresponding author. Health Protection Agency Mycobacterium

Reference Unit, Clinical Research Centre, Barts and The London,

Queen Marys School of Medicine and Dentistry, 2 Newark Street,

London E1 2AT, UK. Tel.: +44 20 7377 5895.

E-mail address: [email protected] (T.J. Brown).

Journal of Microbiological Methods 65 (2006) 294 300

www.elsevier.com/locate/jmicmeth
http:///reader/full/www.who.int/gtb/publications/globrephttp:///reader/full/www.who.int/gtb/publications/globrephttp:///reader/full/www.who.int/gtb/publications/globrep


2/7

treatment regimens, which should improve both patient

outcome and public health by minimizing the transmis-

sion of drug-resistant strains. A variety of culture based

methods for the determination of mycobacterial drug

susceptibilities provide definitive results but take at

least 14 days from primary isolation to produce (Collinset al., 1997). This can be reduced to a matter of hours

using DNA analysis techniques, but only where specific

markers have been identified.

Resistance to RIF in M. tuberculosis has been shown

to be associated with amino acid changes within the h-

subunit of RNA polymerase, encoded by M. tubercu-

losis rpoB (Telenti et al., 1993). Studies from diverse

countries have shown that more than 95% of RIF-

resistant isolates are associated with mutations within

an 81-bp rifampin-resistance determining region

(RRDR) region of rpoB (Herrera et al., 2003; Yue etal., 2003; Cavusoglu et al., 2002; Mani et al., 2001;

Bartfai et al., 2001; Telenti et al., 1993). Although these

mutations are seen throughout this RRDR approximate-

ly 6070% are found within two codons, 531 and 526.

(Tracevska et al., 2002; Huang et al., 2002; Cavusoglu

et al., 2002).

Mutations at a number of loci have been associated

with INH resistance in M. tuberculosis, these include

katG, mabA-inhA, oxyR-ahpC (Zhang et al., 1992;

Banerjee et al., 1994; Sreevatsan et al., 1997). Over

80% of INH-resistant isolates have been reported as

harbouring at least one of the two mutationsAGCNACC at codon 315 in katG and a CNT substi-

tution at15 at the mabA-inhA locus (Musser et al.,

1996; Martilla et al., 1998; Ahmad et al., 2002; Kim et

al., 2003; Bakonyte et al., 2003; Silva et al., 2003).

A variety of techniques have been used to detect

these limited loci (Garca de Viedma, 2003). Arrays,

which have found various applications in microbiology

(Anthony et al., 2001), can be used to analyse multiple

loci in parallel and have been previously described for

the analysis of mutations within M. tuberculosis rpoB

associated with RIF resistance (De Beenhouwer et al.,1995; Cooksey et al., 1997; Watterson et al., 1998).

In this paper we describe and validate a multiplex

PCR followed by hybridisation to low density DNA

oligonucleotide array that had been designed to detect

mutations associated with INH and RIF resistance.

2. Materials and methods

2.1. Bacterial isolates

A panel of 40 M. tuberculosis isolates was assembled

in order to give a wide range of genotypes at the rpoB

RRDR, katG315 and mabA-inh15 loci. These isolates

were cultured on LowensteinJensen (LJ) and drug

susceptibility testing was performed using the resistance

ratio method on LJ media (Collins et al., 1997).

2.2. Preparation of DNA extracts

Bacterial cells from LJ medium were suspended in

100 Al purified water and an equal volume of chloro-

form was added. The tubes were heated at 80 8C for

20 min, placed in the freezer for 5 min and mixed

briefly using a vortex mixer. Immediately before use

as PCR template tubes were centrifuged for 3 min at

12000 g.

2.3. PCR

Biotinylated target PCR products were generated

in a 20 Al multiplex PCR. This contained 1ammo-

nium reaction buffer (Bioline Ltd., London, UK),

dNTP at 0.2 mM each (Amersham Biosciences, Chal-

font St Giles, UK), MgCl2 at 1.5 mM (Bioline),

primers KatGP5IO (CGCTGGAGCAGATGGGCTTGG)

and KatGP6BIO (GTCAGCTCCCACTCGTAGCCG) at

0.25 mM, primers INHAP3BIO (CAGCCACGTTA-

CGCTCGTGG), TOMIP2BIO (CGATCCCCCGGTT-

TCCTCCGG), rpoBP1BIO (GGTCGGCATGTCGCQ

GGATGG) and BrpoB1420R (GTAGTGCGACGGQ

GTGCACGTC) at 1 mM (ThermoHybaid, Ulm, Ger-many), 1 U Taq-polymerase (Bioline) and 1 Al of

DNA template. All primers were biotinylated. Thermal

cycling was performed on a Perkin Elmer 9700 ther-

mocycler using a programme consisting of an initial

melting phase of 5 min at 95 8C followed by 30 cycles

of a 30 s melt at 95 8C and a combined annealing and

extension of 60 s at 65 8C, then a final extension hold

for 5 min at 72 8C.

2.4. Construction of MDR macroarray

Eleven probes as shown in Table 1 were used to

produce a macroarray. The first probe targeted an M.

tuberculosis complex specific locus of M. tuberculosis

rpoB. The next four probes were designed to analyse loci

associated with INH resistance. Two probes, K315WTC

and tomiwt, were designed to detect the wild type (WT)

genotypes atkatG315 and atmabA-inhA15 whilst two

further probes, K315GC and tomimut1, were designed to

detect the most frequently seen genotype at each locus,

katG315AGCYACC and mabA-inhA15CYT, respec-

tively. The remaining 6 probes, MRURP3, MRURP6,

MRURP9, MRURP12, MRURP17 and MRU1371A,

T.J. Brown et al. / Journal of Microbiological Methods 65 (2006) 294300 295


3/7

formed a scanning array designed to detect the WT

genotype of the RRDR of M. tuberculosis rpoB. In

order to optimise probe performance within the arrayoligonucleotide probes were synthesised with 3V polyT

tails (Brown and Anthony, 2000). Oligonucleotide

probes (Invitrogen, Paisley, UK) were diluted to 20

AM in water containing 0.001% bromo phenol blue

and applied to nylon membrane (Magnagraph 0.22

AM, Osmonics, Minnetonka, USA) using a hand-held

arraying device (VP Scientific, San Diego, USA). In

addition to the probes a permanent ink spot was

applied to the membrane in order to orientate the

array and a spot of primer rpoBP1BIO at 2 AM as a

colour development control. Probes were UV-cross-

linked to the nylon membrane. The membranes werewashed in 0.5% 20SSC (Sigma, Poole, UK) then

dried, cut and placed in 2 ml polythene hybridisation

tube (Alpha Labs, Eastleigh, UK).

2.5. Hybridisation and colour detection

The biotin labeled PCR products were denatured by

adding an equal volume of denaturation solution (0.4

M NaOH, 0.02 M EDTA) and incubating at room

temperature for 15 min. A 20 Al aliquot of the dena-

tured PCR was added to tube containing an array and500 Al hybridisation solution (5SSPE; 0.5% SDS)

that was agitated in a hybridisation oven at 60 8C for

15 min. The strips were then washed in wash buffer

(0.4SSPE, 0.5% SDS) at 60 8C for 10 min in the

hybridisation oven. The arrays were agitated in rinse

buffer (0.1 M Tris, 0.1 M NaCl, pH 7.5) at room

temperature (RT) for 1 min. This rinse step was re-

peated using the rinse buffer containing 0.1% blocking

reagent (Roche, Lewes, UK). The arrays were now

agitated at RT for 15 min in the rinse buffer with

0.1% blocking reagent and 1/200 dilution of concen-

trated streptavidin-alkaline phosphatase conjugate

reagent (BioGenex, San Ramon, USA). The mem-

branes were then washed twice in wash solution and

once in substrate buffer (0.1 M Tris, 0.1 M NaCl at pH9.5) before being incubated at RT for 5 min in sub-

strate buffer containing 0.34 mg/ml NBT (USB, Cleve-

land, USA) and 0.17 mg/ml BCIP (USB). The

membranes were washed in water before being air-

dried and the hybridisation patterns noted.

2.6. Interpretation of the macroarray

Hybridisation to any of the probes directed towards

rpoB is indicative of a WT genotype at that locus,

conversely lack of hybridisation with a given rpoB

probe is indicative of a mutant genotype at that locus.Hybridisation with K315WTC is indicative of a

katG315 WT genotype whereas absence indicates a

mutant genotype at this or surrounding this locus.

Absence of hybridisation with K315WTC and hybridi-

sation with K315GC is indicative of the katG315

AGCNACC genotype. Likewise, hybridisation with

TOMIWT is indicative of a mabA-inhA15 WT geno-

type whereas absence indicates a mutant genotype at

this or surrounding this locus. Absence of hybridisation

with TOMIWT and hybridisation with TOMIMUT1

suggests the mabA-inhA15CYT

genotype.

2.7. DNA sequencing

Single primer pairs (see PCR section above) were

used to generate single rpoB, katG or inhA PCR

products using the method given above. These were

diluted 1/100 in purified water and sequenced using

CEQ Quick Start sequencing kits and a CEQ 8000

instrument (Beckman Coulter, High Wycombe, UK)

according to the manufacturers instructions. The PCR

products were sequenced in both directions using the

amplification primers given above.

Table 1

DNA probes used in this study

Probe Array position Sequence

MRUMtb P1 CACCAGCCAGCTGAGCCAATTCATTTTTTTTTT

K315WTC P2 CTCGATGCCGCTGGTGATCGCTTTTTTTTTT

KAT315GC P3 GCGATCACCACCGGCATCGAGTTTTTTTTTTTOMIWT P4 GGCGAGACGATAGGTTGTCGGTTTTTTTTTT

TOMIMUT1 P5 GGCGAGATGATAGGTTGTCGGTTTTTTTTTT

MRURP3 P6 GCCAGCTGAGCCAATTCATGGACTTTTTTTTTT

MRURP6 P7 GCCAATTCATGGACCAGAACAACCTTTTTTTTTTTTTTT

MRURP9 P8 TGGACCAGAACAACCCGCTGTCTTTTTTTTTT

MRURP12 P9 ACAACCCGCTGTCGGGGTTGACTTTTTTTTTT

MRURP17 P10 GGTTGACCCACAAGCGCCGACTTTTTTTTTT

MRUR1371A P11 CGACTGTCGGCACTGGGGCCCGGTTTTTTTTTT

T.J. Brown et al. / Journal of Microbiological Methods 65 (2006) 294300296


4/7

3. Results

The panel of 40 M. tuberculosis isolates contained

30 MDR isolates, 5 RIF-monoresistant isolates, 1 INH

monoresistant isolate and 4 isolates sensitive to RIF and

INH. Sequencing of the RRDR of rpoB of these iso-lates revealed 36 different genotypes in addition to the

WT. Analysis of the codons most commonly associated

with RIF resistance showed two different mutations at

codon 531, 6 different mutations at codon 526, and 4

different mutations at codon 516. Mutations in codons

509, 511, 513, 515, 522, 528, 529 and 533 were also

seen. Seven isolates contained two separate single base

substitutions, four isolates contained insertions and

three contained deletions. The katG315 and mabA-

inhA15 genotypes of 28 of the isolates were deter-

mined. Three genotypes in addition to the WT were

seen at katG315 and a C to T substitution at mabA-inhA15 was seen in addition to the WT. The genotypes

of these isolates are shown in Table 2.

The crude DNA extracts from each of the isolates in

the panel were analysed using the MDR array, the

design of which is shown in Fig. 1 as are representative

examples of the developed arrays. All isolates produced

Table 2

Summary of phenotype and genotype of the isolates used in this study

Isolate Susceptibility by

phenotype

katG315/mabA-inhA15

genotype

rpoB genotype Probes showing

no hybridisation

Susceptibility

by array

INH RIF INH RIF

01/07786 R R 1302CNG/S509R+1351CNT/H526Y P2 P5 P10 R R

236-02 R R AGCNACC/WT 1307TNC/L511P+1322ANG/D516G P2 P5 P8 P7 R R

98/05219 S R 1307TNC/L511P+1351CNG/H526D P3 P5 P6 P10 S R

2936-99 R R WT/WT 1312CNA/Q513K P3 P5 P6 P7 S R

Is20043 R R 1313ANC/Q513P P2 P5 P6 P7 R R

98/05844 R R AGCNAAC/WT 1313ANT/Q513L P2 P3 P5 P6 P7 R R

02/07435 R R 1314 CCAACT ins 513 P2 P5 P6 P7 R R

2651-96 R R AGCNACC/WT 1315 TTC ins 514 P2 P5 P6 P7 R R

Is14373 R R 13151323 Del TTCATGGAC 514516 P2 P5 P6 P7 R R

Is11195 R R WT/WT 13161318 Del TCA 514515 P3 P5 P6 P7 S R

Is14786 R R 1318ANG/M515V+1351CNA/H526N P3 P4 P6 P7 P10 R R

1763-97 R R AGCN

AAC/WT 1318Ins ATTCAT 515 P2 P3 P5 P6 P7 R R 98/07530 R R 1320GNA/M515I P2 P5 P7 P8 R R

2883-97 R R AGCNACC/WT 132126 Del GACCAG 516517 P2 P5 P6 P7 P8 R R

Is11125 R R AGCNACC/WT 13212GANTT/D516F P2 P4 P7 R R

1579-96 S R WT/WT 1321GNT/D516Y P3 P5 P7 S R

Is14027 R R 1322ANG/D516G P3 P4 P7 P10 R R

1004-01 R R AGCNACA/WT 1322ANT/D516V P5 P7 R R

1071-98 R R AGCNACC/WT 1322ANT/D516V+1351CNG/H526D P2 P5 P6 P7 P8 P10 R R

98/00699 R R 1334 AGAACAACC ins 520 P3 P4 R S

1992-00 R R WT/WT 133940TCNCA/S522Q P3 P5 P9 S R

1445-01 R R AGCNACC/WT 1340CNG/S522W P2 P5 P9 R R

395-98 R R AGCNACC/WT 1340CNT/S522L P2 P5 P9 R R

03/02007 R R WT/WT 13501CCNTT/T525T+1351CNT/H526Y P3 P5 P10 S R

2323-02 R R AGCNACC/WT 13512CANTG/H526C P2 P5 P10 R R

1828-00 R R WT/ inhA C15T 1351CNG/H526D P3 P4 P10 R R 2031-02 S R WT/WT 1351CNT/H526Y P3 P5 P10 S R

3381-97 R R AGCNACC/WT 1352 ANG/H526R P2 P5 P10 R R

740-97 R R AGCNACC/WT 1352ANC/H526P P2 P5 P10 R R

1810-96 R R AGCNAAC/WT 1352ANT/H526L P2 P3 P5 P10 R R

01/03682 S S 1359CNT/R528R P3 P5 P10 S R

02/06539 S R 1361GNC/R529P P3 P5 P10 S R

1255-98 R R WT/ inhA C15T 1363CNA/L530M+1367CNT/S531L P3 P4 P6 P7 P11 R R

01/11196 R R 1367CNG/S531W P2 P5 P11 R R

Is5 R R WT/WT 1367CNT/S531L P3 P5 P11 S R

02/03056 S R WT/WT 1373TNC/L533P P3 P5 P11 S R

03/06044 S S WT/WT WT P3 P5 S S

03/04307 S S WT/WT WT P3 P5 S S

03/05269 S S WT/WT WT P3 P5 S S

1182-01 R S AGCNACA/WT WT P5 R S



5/7

interpretable hybridisation patterns with the array and

39 of the 40 different mutations present in the panel

were correctly detected when present. The one isolate

that failed to give a mutant genotype using the array

contained a nine base insertion in the rpoB RRDR. All

other isolates were correctly identified as mutant or

wild type. A mutation was detected in 35 of the RIF-

resistant isolates. A mutation was also detected in a RIF

susceptible isolate that did indeed carry a synonymous

mutation. The array detected mutations at katG315 or

mabA-inhA15 in twenty seven out of the 31 INH-

resistant isolates, the remaining four were wild type atthese loci (Table 2).

Amino acid codons 516, 526 and 531 are the most

prevalent codons involved in rifampin resistance. These

three codons may be responsible for 80% of RIF-resis-

tant M. tuberculosis cases. All the isolates with muta-

tions in these positions were correctly identified. The

rpoB531, rpoB526 and rpoB516 mutant alleles

showed a negative hybridisation signal for the P17

(Fig. 1.B4), the P6 (Fig. 1.B1) and the P17 (Fig.

1.B7) probes, respectively.

Other less frequent mutations at the 511, 513, 515,522, 529 and 533 codons and 6 double single muta-

tions, 3 different deletions and two insertions were

correctly identified. Only the mutant with an unusual

insertion of 9 nucleotides (AGAACAACC) at codon

520 was not identified.

Four MDR-resistant isolates showed a wild type

pattern for INH resistance with positive hybridisation

signal for the katGwt and inhA probes: this is possible

as resistance to INH is caused by a variety of mutations

at several chromosomal loci of M. tuberculosis. Muta-

tions in the katG and the regulatory region of the

mabA-inhA operon have not been found in between

10% and 30% of the INH-resistant M. tuberculosis

isolates.

All the isolates with known sequences of the rele-

vant part of katG and the regulatory region of the

mabA-inhA operon were correctly identified. The

INH-resistant M. tuberculosis isolates with different

mutations in katG codon 315 were correctly identified.

The INH-resistant isolates with the S315T mutation had

a pattern with a negative hybridisation signal for the

katGwt and a positive hybridisation signal for the

katGmut probe (Fig. 1.B1B2). Others with different

mutations in katG (S315 ACA, S315 AAC or S315AGG) showed a pattern with a negative hybridisation

signal for both probes (Fig 1.B5). The INH-resistant M.

tuberculosis isolates with the inhAC15T mutation

showed a negative hybridisation signal for the inhAwt

probe and a positive hybridisation signal for the inhA-

mut probe (Fig. 1.B6).

4. Discussion

Detecting drug resistance in M. tuberculosis isolates

by determining genotype is an attractive alternative toconventional phenotypic susceptibility testing because

results can be generated within hours with minimal

manipulation of live organisms. Obviously this ap-

proach can only be used where genotypic markers

for drug resistance have been identified. This is the

case for MDRTB where a range of mutations in the

RRDR of rpoB is highly specific to RIF-resistant

isolates and point mutations, one in katG and one

associated with inhA are strongly associated with

INH-resistant isolates. By analysing these three loci

80% of the MDRTB isolates present in the panel could

be identified. Using macroarray analysis all these loci

A

P1 MtbC

P3 katG315ACC

P4 inhAWT

P5 inhAMUT-15T

P6

P7

P8

P9

P10

P2 katGWT

Ink spot

control

P11

B

7

6

5

4

3

2

1

Fig. 1. Schematic of the MDR macroarray (A) and patterns of the M. tuberculosis strains (B): pattern 1, strain with the katG315 AGC-ACC

mutation and rpoB526 mutant allele; pattern 2, strain with the katG315 AGC-ACC mutation and insertion in the rpoB531 gene (lns TTC at the 514

codon); pattern 3, strain wild type; pattern 4, strain with the rpoB531 mutant allele; pattern 5, strain with the katG315 AGC-ACA mutation; pattern

6, starin with the inhAC15T mutation in the regulatory region of the mabA-inhA operon; pattern 7; strain with the rpoB516 mutant allele.



6/7

can be analysed in parallel. We have described such a

macroarray based assay for the detection of MDRTB

above. The principle of the MDR array assay is that a

mutation should impede the hybridisation of the target

to the relevant WT probe or in the case of the katG315

or inhA loci permit the hybridisation to the cor-responding mutant probe. The implementation de-

scribed was capable of detecting 35/ 36 different

mutations in the RRDR of rpoB, 3/3 different muta-

tions at katG315 and 1/1 at inhA15. It has been

reported that the range of mutations seen within the

loci scrutinized by this assay may vary with geograph-

ical location (Nikolayvsky et al., 2004). The mutations

used to challenge the array were detected in isolates

seen in the UK and Spain. The efficiency of the array

when used to analyse isolates from other geographical

locations may vary although it is capable of detectingall the commonly reported mutations within rpoB,

inhA and katG.

When susceptibility is designated by genotype, there

are four sources of discrepancy with the more definitive

phenotypic testing. Firstly a resistant isolate may not

contain the marker or markers targeted. According to

the literature this is seen in b5% of RIF-resistant iso-

lates and between 10% and 30% of INH-resistant iso-

lates with the markers described in this study. This type

of discrepancy was seen in 4 INHresistant isolates in

the present study. These discrepancies could be reduced

if as further markers are identified they were includedin the assay. Secondly, a susceptible isolate may contain

a synonymous mutation which when detected would

lead to the isolate being designated resistant if the assay

detects but does not identify mutations. These muta-

tions are rarely described in the literature at the loci

used in this assay and discrepancies caused by them

could be reduced by identification of mutations seen

either by sequencing mutant loci or inclusion on the

macroarray of probes directed at all possible mutations.

One rpoB mutant such as this was seen in the present

study. Thirdly the assay may fail to correctly detectmutations that are actually present. This type of dis-

crepancy is minimised by careful selection of the

probes used in the array. Because the hybridisation

behaviour of a given probe and target combination is

difficult to predict it is essential to validate all probes

with potential targets. In the present study only one

mutation was not detected. This was a 9 base insertion

which had the effect of producing a 3 base mismatch at

the 5V end of the 22 base probe MRURP9 which

presumably did not destabilise the hybridisation duplex

sufficiently to prevent the detection of hybridisation.

Lastly, where a non-synonymous mutation occurs but

does not result in the drug-resistant phenotype. This

type of discrepancy was not seen in this study and is

likely to be seen rarely, the effect on the predictive

value of the array could be negated by the inclusion

of all possible genotypes on the array although high

density platforms may be more suitable for this ap-proach. The platform described in this paper is low

cost using unmodified clearly defined oligonucleotides

and no specialized equipment. This also makes it flex-

ible and so ideal for the development of expanded or

novel probe sets for oligonucleotide arrays addressing

local needs. Oligonucleotide arrays, as described here,

consist of a series of probes immobilized on a solid

support all of which must exhibit similar hybridisation

destabilisation properties when probed with mis-

matched target DNA under the same stringent wash

conditions. There is no method for reliably predictingthe hybridisation properties of a probe target pairing

and so probes within the array must be validated em-

pirically. Experimentation is minimized by choosing

probes of between 18 and 21 bp, with equal estimated

melting temperature, with potential mismatch pairs

greater than 2 bases from either end and with ten

thiamine bases added to the 3V end of the probe. If

possible complimentary runs of greater than two bases

and guaninethymine mismatches should be avoided. If

specificity is poor probes can be improved by moving

the potential mismatch along the probe (Anthony et al.,

2003). If hybridisation signal intensity is a problem, thiscan be increased by adding thymine bases to the 3V end

of the probe or reduced by removing thymine bases.

In summary, the MDR screen macroarray can

identify M. tuberculosis complex isolates resistant to

INH and/or RIF, the two most important drugs in the

treatment of tuberculosis although phenotypic testing

is required to definitively identify all MDR-TB iso-

lates. The assay is easy to perform and interpret, uses

multipurpose equipment available in a basic molecu-

lar biology laboratory and could be performed in

routine clinical microbiology laboratories, most use-fully in areas with a high prevalence of MDR M.

tuberculosis.

References

Ahmad, S., Fares, E., Araj, F., Chugh, T.D., Mustafa, A.S., 2002.

Prevalence of S315T mutation within the katG gene in isoniazid-

resistant clinical Mycobacterium tuberculosis isolates from Dubai

and Beirut. Int. J. Tuberc. Lung Dis. 6, 920926.

Anthony, R., Brown, T., French, G., 2001. DNA array technology and

diagnostic microbiology. Expert Rev. Mol. Med. 1, 89 97.

Anthony, R.M., Schuitema, A.R.L., Chan, A.B., Boender, P.J.,Klaster, P.R., Oskam, L., 2003. Effect of secondary structure



7/7

on single nucleotide polymorphism detection with a porous

microarray matrix; implications for probe selection. Biotechni-

ques 34, 10821089.

Bakonyte, D., Baranauskaite, A., Cicenaite, J., Sosnovskaya, A.,

Satakenas, P., 2003. Molecular characterization of isoniazid-resis-

tant Mycobacterium tuberculosis clinical isolates in Lithuania.

Antimicrob. Agents Chemother. 47, 2009 2011.Banerjee, A., Dubnau, E., Quemard, A., Balasubramanian, U., Um,

K.S., Wilson, T., Collins, D., Lisle Jr., G., Jacobs, W.R., 1994.

inhA, a gene encoding a target for isoniazid and ethionamide in

Mycobacterium tuberculosis. Science 263, 227 230.

Bartfai, Z., Somoskovi, A., Kodmon, C., Szabo, N., Puskas, E.,

Kosztolanyi, L., Farago, E., Mester, J., Parsons, L.M., Salfinger,

M., 2001. Molecular characterization of rifampin-resistant isolates

of Mycobacterium tuberculosis from Hungary by DNA sequenc-

ing and line probe assay. J. Clin. Microbiol. 39, 3736 3739.

Bifani, P.J., Plikaytis, B., Kapur, V., Stockbauer, K., Pan, X., Lutfey,

M.L., Moghazeh, S.L., Eisner, W., Daniel, T.M, Kaplan, M.H.,

Crawford, J.T., Musser, J.M., Kreiswirth, B.N., 1996. Origin and

interstate spread of New York City multidrug-resistant Mycobac-

terium tuberculosis clone family. JAMA 275, 452457.Brown, T., Anthony, R., 2000. The addition of low numbers of 3V

thymine bases can be used to improve the hybridization signal of

oligonucleotides for use within arrays on nylon supports.

J. Microbiol. Methods 42, 203 207.

Cavusoglu, C., Hilmioglu, S., Guneri, S., Bilgic, A., 2002. Charac-

terization of rpoB mutations in rifampin-resistant clinical isolates

of Mycobacterium tuberculosis from Turkey by DNA sequencing

and line probe assay. J. Clin. Microbiol. 40, 4435 4438.

Collins, C.H., Grange, J.M., Yates, M.D., 1997. Tuberculosis Bacte-

riology, Organization and Practice, 2nd ed. Butterworth-Heine-

mann, Oxford.

Cooksey, R.C., Morlock, G.P., Glickman, S., Crawford, J.T., 1997.

Evaluation of a line probe assay kit for characterization ofrpo

B

mutations in rifampin-resistant Mycobacterium tuberculosis iso-

lates from New York City. J. Clin. Microbiol. 35, 12811283.

Davies, P.D., 2003. The worldwide increase in tuberculosis: how

demographic changes, HIV, infection and increasing numbers in

poverty are increasing tuberculosis. Ann. Med. 35, 235 243.

De Beenhouwer, H., Liang, Z., Jannes, G., Mijs, W., Machtelinckx,

L., Rossau, R., Traore, H., Portaels, F., 1995. Rapid detection of

rifampin resistance in sputum and biopsy specimens from tuber-

culosis patients by PCR and line probe assay. Tuberc. Lung Dis.

76, 425430.

Garca de Viedma, D., 2003. Rapid detection of resistance in Myco-

bacterium tuberculosis: a review discussing molecular

approaches. Clin. Microbiol. Infect. 9, 349359.

Herrera, L., Jimenez, M.S., Valverde, A., Garcia-Aranda, M.A., Saez-Nieto, J.A., 2003. Molecular analysis of rifampin-resistant M.

tuberculosis isolated in Spain (19962001). Description of new

mutations in the rpoB gene and review of the literature. Int. J.

Antimicrob. Agents 21, 403408.

Huang, H., Jin, Q., Ma, Y., Chen, X., Zhuang, Y., 2002. Characte-

rization of rpoB mutations in rifampin-resistant Mycobacterium

tuberculosis isolated in China. Tuberculosis 82, 7983.

Kim, S., Park, Y., Kim, W., Lee, S., Ludgerus Chang, C., Kang, S.,

Kang, C., 2003. Molecular analysis of isoniazid resistance in

Mycobacterium tuberculosis isolates recovered from South

Korea. Diagn. Microbiol. Infect. Dis. 47, 497 502.

Mani, C., Selvakumar, N., Narayanan, S., Narayanan, P.R., 2001.

Mutations in the rpoB gene of multidrug-resistantMycobacterium

tuberculosis clinical isolates from India. J. Clin. Microbiol. 39,

29872990.

Martilla, H., Soini, H., Eerola, E., Vyshnevskaya, E., Vyshnevskiy, B.,

Otten, T., Vasilyef, A., Viljanen, M., 1998. A Ser315Thr substi-

tution in KatG is predominant in genetically heterogeneous mul-

tidrug-resistant Mycobacterium tuberculosis isolates originating

from the St. Petersburg area in Russia. Antimicrob. Agents Che-mother. 42, 24432445.

Musser, J.M., Kapur, D., Williams, D.L., Kreiswirth, B.N., Van

Soolingen, D., Van Embden, J.D., 1996. Characterization of the

catalase-peroxidase gene (katG) and inhA locus in isoniazid

resistant and susceptible strains of Mycobacterium tuberculosis

by automated DNA sequencing: restricted array of mutation as-

sociated with drug resistance. J. Infect. Dis. 173, 196 202.

Narvskaya, O., Otten, T., Limeschenco, E., Sapozhnikova, N.,

Graschenkova, N., Steklova, L., Nikonova, A., Filipenko,

M.L., Mokrousov, I., Vysnevskiy, B., 2002. Nosocomial

outbreak of multidrug-resistant tuberculosis caused by a

strain of Mycobacterium tuberculosis W-Beijing family in

St. Petersburg Russia. Eur. J. Clin. Microbiol. Infect. Dis.

21, 596 602.Nikolayvsky, V., Brown, T.J., Balabanova, Y., Ruddy, M.C., Fedorin,

F.A., Drobniewski, F.D., 2004. Detection of mutations associated

with isoniazid and rifampin resistance in Mycobacterium tuber-

culosis isolates from Samara region Russian Federation. J. Clin.

Microbiol. 42, 44984502.

Quy, H., Lan, N., Borgdorff, M., Grosset, J., Linh, P., Tung, L., van

Soolingen, D., Raviglione, M., Co, N., Broekmans, J., 2003. Drug

resistance among failure and relapse cases of tuberculosis: is the

standard re-treatment regimen adequate? Int. J. Tuberc. Lung Dis.

7, 631636.

Silva, M., Senna, S., Ribeira, M., Valim, A., Telles, M., Kritski, A.,

Morlock, G., Cooksey, R., Zaha, A., Rossetti, M., 2003. Muta-

tions inkat

G,inh

A andahp

C genes of Brazillian isoniazid restant

isolates of Mycobacterium tuberculosis. J. Clin. Microbiol. 41,

44714474.

Sreevatsan, S., Pan, X., Zhang, Y., Deretic, V., Musser, J.M., 1997.

Analysis of the oxyR-ahpC region in isoniazid-resistant and sus-

ceptible Mycobacterium tuberculosis complex organisms reco-

vered from diseased humans and animals in diverse localities.

Antimicrob. Agents Chemother. 41, 600606.

Telenti, A., Imboden, P., Marchesi, F., Lowri, D., Cole, S., Colston,

M.J., Matter, L., Schopfer, K., Bodmer, T., 1993. Detection of

rifampicin-resistance mutations in M. tuberculosis. Lancet 341,

647650.

Tracevska, T., Jansone, I., Broka, L., Marga, O., Baumanis, V., 2002.

Mutations in the rpoB and katG genes leading to drug resistance

in Mycobacterium tuberculosis in Latvia. J. Clin. Microbiol. 40,37893792.

Watterson, S.A., Wilson, S.M., Yates, M.D., Drobniewski, F.A., 1998.

Comparison of three molecular assays for rapid detection of

rifampicin resistance in Mycobacterium tuberculosis. J. Clin.

Microbiol. 36, 19691973.

Yue, J., Shi, W., Xie, J., Li, Y., Zeng, E., Wang, H., 2003. Mutations

in the rpoB gene of multidrug-resistant M. tuberculosis isolates

from China. J. Clin. Microbiol. 41, 22092212.

Zhang, Y., Heym, B., Allen, B., Young, D., Cole, S.T., 1992. The

catalase peroxidase gene and isoniazid resistance in Mycobacte-

rium tuberculosis. Nature 358, 501 593.


the use of macroarrays for the identification of mdr

Documents