ELSEVIER FEMS Microbiology Letters 144 (1996) 103-108

Downloaded from https://academic.oup.com/femsle/article/144/1/103/527475 by guest on 16 September 2023

Rifampicin resistance and mutation of the rpoB gene in
Mycobacterium tuberculosis
Hatsumi Taniguchi al*, Hironori Aramaki c, Yoshihiko Nikaido b,
Yasuo Mizuguchi d, Masahiro Nakamura e, Toshihiko Koga e, Shin-ichi Yoshida a
il Department of Microbiology, University of Occupational and Environmental Health, I-1 Iseigaoka, Yahatanishi-ku,
Kitakyushu, Fukuoka 807, Japan
’ Second Department of Internal Medicine, University of Occupational and Environmental Health, I-l Iseigaoka.
Yahatanishi-ku, Kitakyushu, Fukuoka 807, Japan
’ Daiichi College of Pharmaceutical Sciences. 22-l Tamagawa-cho. Minami-ku, Fukuoka 815, Japan
’ Public Health Laboratory of Chiba Prefecture, Nitona, Chuo-ku, Chiba 260. Japan
e Koga Hospital Medical Research Institute, 120 Tenjin-cho, Kurume 830. Japan

Received 26 May 1996; accepted 26 August 1996

Abstract

Using 39 clinical isolates of Mycobacterium tuberculosis strains with a broad range of susceptibility to rifampicin, we
examined the relationship between the degree of resistance to rifampicin and mutational sites of the rpoB gene. All rifampicin-
resistant strains had missense mutations. Twenty strains (95%) had a mutation in the cluster I region, which has also been
reported in Escherichia coli [Jin and Gross (1988) J. Mol. Biol. 202, 45-581, and the remaining one strain had a mutation at
codon 381 [Ala + Val] in the N-terminal region, which has not been reported in E. coli. Among 18 rifampicin-susceptible
strains, two had a mutation in the cluster I region and the other three strains had a mutation in the cluster III region. The
mutations at codons 513 (So/,), 526 (33%) or 531 (43%) in the cluster I region led to high level resistance to rifampicin (50 ug
ml-‘SMIC). The mutations at the other sites, in the cluster III region (codons 679 or 687) and even in the cluster I region
(codon 514, 521, or 533), showed low level (MIC = 12.5 ug ml-‘) or no (MIC < 0.39 ug ml-l) resistance to rifampicin. These
results suggest that mutations in the rpoB gene are, mostly, but not necessarily, associated with rifampicin resistance of
M. tuberculosis, and the sites of mutations on the rpoB gene will affect the level of resistance to rifampicin.

Keywords: Mycobacterium tuberculosis; Rifampicin resistance; rpoB gene; Mutational site; Level of resistance

1. Introduction tin is believed to interfere with transcription by bind-

Rifampicin has been proved to be an effective anti-
tuberculosis agent against susceptible Mycobacterium
tuberculosis [2,3]. The action mechanism of rifampi-
* Corresponding author. Tel.: +81 (93) 691-7242;
Fax: +81 (93) 6024799; E-mail: hatsumi@med.uoeh-u.ac.jp
ing the drug to the p subunit of RNA polymerase
(rpoB gene) at a region formed by the appropriate
complexing of the different RNA polymerase sub-
units [4]. Mutated sites at the rpoB gene in rifampi-
tin-resistant Escherichia coli strains were identified
by Jin and Gross [l], and Lisitsyn et al. [S]. The sites
were classified into three clusters (namely regions I,

037%1097/96/ $12.00 Copyright 0 1996 Federation of European Microbiological Societies. published by Elsevier Science B.V.
PIISO378-1097(96)00346-l
104 H. Taniguchi LJIul. IFEMS
II, and III) [l] and one site at a 146 amino acid
residue upstream of the cluster I region [5]. Further-
more, it has also been reported that a level of resist-
ance to rifampicin depended on the mutational sites
in the rpoB gene in E. cd, mutations in the cluster I
region of the rpoB gene leading to especially high
levels of resistance to rifampicin.
Recently, several researchers have investigated nu-
cleotide sequences of the cluster I and II regions of
the rpoB gene of rifampicin-resistant A4. tuberculosis
strains. They reported that although most rifampi-
tin-resistant A4. tuberculosis clinical isolates exhibited
mutations with a single nucleotide change or inframe
deletions and insertions in the cluster I region, some
3-10% of the rifampicin-resistant strains did not
show any mutations within this region [6-l 11. So
we wanted to know whether those strains had muta-
tions in other regions of the rpoB gene than the
cluster I region or in other genes than the rpoB
gene. Also no information is available on the rela-
tionship between the level of resistance to rifampicin
and mutational sites of the rpoB gene in A4. tubercu-
losis.
So, we also determined the nucleotide sequence of
the N-terminal region, designated N, and the cluster
III region other than the cluster I and II regions of
the rpoB gene for M. tuberculosis clinical isolates
having a broad range of susceptibility to rifampicin.
And we summarized the relationship between the
mutational sites and the degree of resistance to rif-
ampicin.
2. Materials and methods
2.1. Bacterial strains
Thirty-nine clinical isolates of Mycobacterium tu-
berculosis (including multiple drug-resistant strains)
were received from Koga hospital, Nougata central
hospital, and Fukuoka-Higashi hospital in Fukuoka,
Japan.
2.2. Drug susceptibility testing
Rifampicin susceptibility tests were carried out by
two methods. One method is a proportion method,
using 1% Ogawa’s egg medium (Nissui Seiyaku Co.
Microbiology Letters 144 (1996) 103-108
Ltd. Tokyo, Japan) containing 0, 10 and 50 pg ml-’
rifampicin (Table 1). According to the instructions of
the suppliers, a resistant level is shown by a percent
degree comparing between the colony numbers
grown on 1% Ogawa’s egg medium containing 10
and 50 pg ml-’ of rifampicin and that on 1% Oga-
wa’s egg medium not containing drug (control medi-
Downloaded from https://academic.oup.com/femsle/article/144/1/103/527475 by guest on 16 September 2023
um). This method was used at the time of isolation
in each hospital. The other method is an absolute
concentration-microtiter method. Cells grown in
Middlebrook 7H9 broth (Difco Lab., Detroit, MI,
USA) were adjusted to the turbidity of McFarland
No. 1 standard, and 50 pl of the cell suspensions
were inoculated into 50 ~1 Middlebrook 7H9 broth
with serial twofold dilution of rifampicin (0.39-200
pg ml-‘). The microtiter plates were incubated at
37°C for 4 weeks. The minimum inhibitory concen-
tration (MIC) was defined as the lowest concentra-
tion of rifampicin which inhibits a growth of M.
tuberculosis. As shown in Table 1, M. tuberculosis
isolates were divided into resistant and sensitive
strains according to MIC.
2.3. DNA preparution for PCR
To extract DNA, all clinical isolates were cultured
on 1% Ogawa’s egg medium at 37°C for 2 weeks.
Cells were suspended in 0.5 ml. of 50 mM EDTA
pH 8.0 and after centrifugation (15 OOOxg, 10 min,
4”C), resuspended into 0.5 ml of 1% Triton X-1Oc
50 mM EDTA at 30°C overnight. Cells were har-
vested by centrifugation (15 000 Xg, 10 min, 4”C),
and suspended in 0.5 ml 1% SDS, 10 mg ml-’ lyso-
zyme-36 U ml-’ proteinase K in 50 mM EDTA, and
were further incubated overnight at 37°C. The prep-
aration was treated twice with phenol<hloroform
(1: 1). The DNA was precipitated with ethanol, and
solubilized in 20 pl of TE buffer (10 mM Tris, 1 mM
EDTA, pH 8.0).
2.4. Amplification by PCR
The DNA template (l-20 ng) was amplified in a
100 yl reaction mixture containing dATP, dTTP,
dCTP and dGTP, all at 200 yM, each primer at
1.O PM, 2.5 U of Taq DNA polymerase (Perkin-El-
mer Cetus, Norwalk, CT), 10 mM Tris-HCl (pH
8.3), 50 mM KC1 and 1.5 mM MgCIP. The amplifi-
 H. Taniguchi et al. I FEMS Microbiology Letters 144 (1996) 103-108 105

cation was performed for 30 cycles (1 min at 94”C, Among the 21 resistant strains, four showed a low
1 min at 55”C, 2 min at 72°C) by a model 480 ther- level of resistance to rifampicin (MIC = 12.5 ug ml-i).
mocycler (Perkin-Elmer Cetus). As shown in Fig. 1, In E. co/i, mutations of rifampicin-resistant strains
cluster I region was amplified using primers TR9 (5’- were located in three distinct clusters (cluster I, II,
TCGCCGCGATCAAGGAGT-3’) and TR8 (5’- and III regions) [l], and in the N-terminal region
TGCACGTCGCGGACCTCCA-3’), TRl (5’-TAC- (codon 146) [5] of the rpoB gene (Fig. 1). Most of
GGTCGGCGAGCTGATCC-3’) and TR2b (5’- the mutations leading to high level resistance were

Downloaded from https://academic.oup.com/femsle/article/144/1/103/527475 by guest on 16 September 2023

TACGGCGTTTCGATGAACC-3’) [lo]. The latter located in the cluster I region in E. coli [l]. We there-
combination could also amplify the cluster II region. fore determined the nucleotide sequence of the clus-
Furthermore, the 1050 bp or 750 bp rpoB DNA ter I region of the rpoB gene of all isolates. If no
fragments containing the N region or the cluster mutations were detected in the cluster I region, the
III region were amplified using primers RpoN (5’- nucleotide sequences of the N, cluster II, and cluster
GGGCACGTTCATCATCAACG-3’) and TR2b III regions were determined.
(5’-TACGGCGTTTCGATGAAC-3’), TRl (5’-TA- The results are shown in Table 1. All of the 21
CGGTCGGCGAGCTGATCC-3’) and RpoC (5’- rifampicin-resistant strains had a missense mutation
CCCGGTGCCCACCAGCGGGGC-3’). in the rpoB gene. No mutants with insertions or de-
letions were found. Among them, 20 strains (95%)
2.5. Sequencing of the PCR product exhibited mutations in the cluster I region, but strain
No. 59 (5%) did not. This proportion of mutant
The PCR products were cloned into TA cloning strains to non-mutant strains in the cluster I region
vector (InVitrogen, Funakoshi Co. Ltd., Tokyo, Ja- is the same as previously reported [611]. However,
pan) or M13mpl8 single strand phage DNA vector. strain No. 59 had a missense mutation at codon 381
The nucleotide sequence of the PCR-generated rpoB in the N-terminal region in which alanine changed to
gene was analyzed with a DSQlOOO automated DNA valine. This mutation did not coincide with that of
sequencer (Shimazu Co. Ltd., Tokyo, Japan) or ABI the E. coli mutant at codon 146. However, we as-
373A DNA sequencer (Perkin Elmer Japan Co. Ltd., sume that the mutation might be associated with
Tokyo, Japan). Recombinant clones were sequenced rifampicin resistance, because this strain did not
with a Takara Taq cycle sequencing core kit (Takara have any mutations in the cluster I region, and the
Co. Ltd., Kyoto, Japan) and a Dye primer cycle rifampicin-sensitive strains so far tested showed no
sequencing kit (Perkin Elmer Japan Co. Ltd.) using mutation at this codon. Further investigation is nec-
a fluorescein-labeled universal primer (5’- essary to determine the relationship between muta-
TGTAAAACGACGGCCAGT-3’) ‘and an Ml 3 re- tions at this codon and drug resistance. However,
verse primer (5’-CAGGAAACAGCTATGACC-3’). these results suggest that strains without any muta-
tion in the cluster I region as previously reported [6-
1 l] might have a mutation in another region of the
3. Results and discussion rpoB gene.
Nineteen of the 21 strains had one point mutation
Since evaluation of the resistance level by the pro- while two strains (Nos. 72 and 20) exhibited a com-
portional method using 1% Ogawa’s egg medium bination of two point mutations (codons 516, 526
may not be suitable for accurate determination, we and codons 514, 516), suggesting that they were
retested the rifampicin susceptibilities of M. tubercu- two-step mutations. Interestingly, one of them
losis strains by determination of the MIC (see Sec- (strain No. 20) showed a low MIC (discussed below).
tion 2.2). The results determined by two methods are No mutations were found in the cluster II and clus-
shown in Table 1. Strains could be classified into ter III regions in the rifampicin-resistant strains ex-
resistant and sensitive ones according to the MIC amined.
values. Of 39 strains, 21 were identified as resistant In 13 (72%) out of 18 rifampicin-sensitive strains,
to rifampicin (12.5 ug ml-‘SMIC), and the remain- no mutations were found in any of the regions (N,
ing 18 strains as sensitive (MIC<0.39 pg ml-‘). cluster I, II, and III regions), while five strains ex-
106 H. Taniguchi et al. iFEMS Microbiology Letters 144 (1996) 103-108

Table 1
Sequence analysis of the rpoB gene of M. tuberculws clinical isolates

Strain no. Number MIC” ‘%,growth on” Mutational position’ Amino acid Nucleotide
of isolates tug ml-‘) change change
+RflO 50 N I II III

Resistant strains (n = 21 i
59 1 200 100 100 381’1 -1’ _ Ala + Val GCG + GTG
58 1 > 200 100 100 513 Gln + Leu CAA -+ CTA

Downloaded from https://academic.oup.com/femsle/article/144/1/103/527475 by guest on 16 September 2023

12 1 > 200 100 100 516 Asp + Ala GAC + GCC
526 His -+ Asp CAC + GAC
23 1 200 80 8.2 526 His + Asp CAC + GAC
7 I > 200 80 60 526 His*Tyr CAC + TAC
5, 73, 74 3 > 200 100 100 526 His + Tyr CAC --) TAC
8 1 > 200 100 100 526 His + Pro CAC + CCC
1, 3, 4, 21, 15, 76 6 > 200 100 100 531 Ser + Leu TCG + TTG
6 1 > 200 100 20 531 Ser -+ Leu TCG + TTG
26 1 100 100 100 531 Ser --) Leu TCG + ‘PI-G
2 I 50 100 100 531 Ser + Leu TCG -+ TTG
22 1 12.5 100 30 533 Leu + Pro CTG + CCG
64 I 12.5 100 100 533 Leu + Pro CTG --) CCG
20 I 12.5 100 100 514 Phe + Leu TTC + TTG
516 Asp --) Val GAC + GTC
Sensitive strains (n = 181
24 I CO.39 50 0 521 Leu + Pro CTG + CCG
69 1 CO.39 so 0 533 Leu + Pro CTG + CCG
60 1 CO.39 25 0 679 Ala + Ser GCC + TCC
66 1 CO.39 0 0 619 Ala + Ser GCC + TCC
70 1 CO.39 25 0 687 Arg --f Pro CGC + CCC
25 1 CO.39 3 0 _
27 1 < 0.39 0.7 0 _
71 1 CO.39 20 0 _
65, 67 2 ~0.39 25 0 _
50 I CO.39 40 0 _
52 1 CO.39 50 0
62 1 10.39 60 0 _
53 1 CO.39 80 0
55, 56 2 CO.39 80 50
61 1 < 0.39 100 0
57 I < 0.39 100 80 _.

“By the absolute concentration method using Middlebrook 7H9 broth

bBy the proportion method; % of colony numbers grown on 1% Ogawa’s egg medium containing 10 or 50 ug ml-’ rifampicin to that on the
same medium not containing drug.
“Mutational position on the N, cluster I, 11, and III regions of the rpoB gene, previously reported in E. coli [1,5].
dNumbers correspond to E. coli RNA polymerase amino acid position [I].
K’ indicates that mutations were not found as a result of nucleotide sequencing.

hibited point mutations. Two of them had a muta-
tion in the cluster I region (codon 521 or 533) and
three strains showed a mutation in the cluster III
region (codon 679 or 687). This suggests that muta-
tions in the cluster III region are not associated with
rifampicin resistance or confer only a very low level
of resistance in M. tuberculosis (MICc 0.39 ug
ml-‘). Also in E. coli, cluster III mutants showed a
low level of resistance [l]. Because no strain having a
mutation in the cluster II region was present, we do
not know whether a mutation at cluster II is in-
volved in rifampicin resistance in M. tuberculosis.
It has been reported that in E. co& the sites of
mutation in the cluster I region will affect the level
of resistance to rifampicin [l]. However, the relation-
ship has not been reported between the degree of
 H. Taniguchi et al. IFEMS Microbiology Letters 144 (1996) 103-108 107

rpoB ( /3 subunit)
Amino acid 1 1342

Mutations sites reported

inRifrE. wli a

primerC RgN lL &9 T;r8 l&J R&

Downloaded from https://academic.oup.com/femsle/article/144/1/103/527475 by guest on 16 September 2023

Region d N I II III
Rif no./mutants no. e O/O l/l 22/24 o/o 013
Fig. 1. A diagram of regions analyzed by indicated primers, mutational sites in the rpoB gene, and a relationship with rifampicin resist-
ance of M. tuberculosis. Line 1: Numbers indicate codon numbers corresponding to the fi subunit of E. coli RNA polymerase amino acid
position [I]. Line 2: Numbers in parentheses are mutational sites found in this study. Line 3: Primers TRl, TRZb, TR8, and TR9 were
reported previously by Telenti et al. [IO]. RpoN and RpoC were designated in this study. Line 4: Nucleotide sequence of these regions
were analyzed. Cluster I, II, III regions are according to Jin and Gross [l]. The N region was designated in this study. Line 5: Number
of strains expressingd rifampicin resistance/number of strains harboring a mutation in each region.

resistance to rifampicin and the mutational sites in don 381, 513, 514, 516, 526, 531, or 533 led to re-
IV. tuberculosis. So we summarized this relationship sistance to rifampicin. Among them, a mutation at
(Fig. 2). Our results showed that a mutation at co- codon 513, 526 or 531 led to high MICs (50 pg

N Region 1
El
CCC CCG ACC AAAGAGTCA EC: CAG ICC CTC TTG GAA AAC TTC TTC TTC AAG GAG
375 -Pro Pro Thr tys Glu Ser Ala Gin Thr teu Lou Glu Ass leu Phe Phe tyr Glu-392

1
I Region a
I I
a b
1 I bl

CTG ICC CAA TTC ATG GAC CAG AAC AAC CCC CTG TCG CCC TTG ACl: CAC AAG CGC CGA CTG TCG CCC CTG
Sll- leu Ser Glo Phe Met Asp Gin Asn Asn Pro teu Ser Gly Leu Tbr His lys Arg Arg teu Ser Ala teu-533
CTG AGC CAA TTC ATG GAC CAG AAC AAC CCC CTG TCG GGG TTG ACC CAC AAG CGC CGA CTG TCG CCC CTG

III Region
GCC AAC CGT CCC CTC ATG GCG GCA AAC ATG CAG CGC CAC GCG
616 -Ala Ass Arg Ala teu Met Gly Ala Asn Met Gin Ar CIn Ala- 689
2 1
K!l r El

Fig. 2. Mutational sites of M. tuberculosis rpoB gene. The mutational positions of rifampicin-resistant M. tuberculosis harboring 12.5 pg
ml-‘ SMICs or MICs < 0.39 pg ml-’ are illustrated above or below the wild-type sequence. The numbers on the left side above the boxes
are those of strains which harbored 50 pg ml-‘SMICs, and the numbers on the right side above the boxes are those of strains which har-
bored MIC= 12.5 kg ml-‘. ‘a’ and ‘b’ indicate a combination of two mutations harbored in one strain, corresponded to strains 20 and
72 in Table 1. We used the rpoB codon numbering system used by Telenti et al. [lo] for comparison between our data and those gener-
ated by them. The codon numbers are designated on the basis of alignment of the translated E. coli rpoB sequence, and are not the actual
M. tuberculosis rpoB codons.
108 H. Taniguchi et ul. I FEMS Microbiology Letters 144 (1996) 103-108
mll’<MIC). Especially a mutation at codon 526 or
531 occurred most frequently (33% and 43%). How-
ever, even in the cluster I region, a mutation at co-
don 514, 521, or 533 led to low MICs (MIC= 12.5
ug ml-‘) or had no influence (MIC < 0.39 pg ml-‘).
These phenomena strongly suggest a relationship be-
tween the degree of resistance to rifampicin and the
mutational sites. However, confusing phenomena
were also observed. The mutation at codon 516 or
533 of the cluster I region exhibited different levels of
resistance. In the case of a mutation at codon 516
(strains No. 20 and 72), both of them had another
mutation in combination. Strain No. 72 changed as-
pargic acid to alanine at codon 516 and also had
another mutation at codon 526, and was high level
resistant to rifampicin (200 ug ml-‘IMIC). Strain
No. 20 changed aspargic acid to valine at codon 516,
and had another mutation at codon 514, and was
low level resistant to rifampicin (MIC = 12.5 p.g
ml-‘). Two explanations are possible: either there
is a difference of amino acid change at codon 516,
or there is a difference of other mutational sites har-
bored by the same strain. Although we could not
exclude the former, the latter might be reasonable,
because a single mutation at codon 526 led to a high
MIC, as observed in another six strains. On the
other hand, in the case of a mutation at codon
533, strains No. 22, 64, and 69 exhibited the same
mutation which changed leucine to proline. How-
ever, two of them (strains 22 and 64) exhibited a
low level of resistance (MIC= 12.5 pg ml-‘) and
strain 69 was sensitive to rifampicin (MIC < 0.39
pg ml-l). One possibility is that some other muta-
tion(s) occurred outside the sequenced part of the
rpoB gene or in a gene affecting the permeability of
rifampicin influenced the level of resistance in these
mutants. Nucleotide sequencing of the entire rpoB
gene may be necessary to answer this question.
Comparing the degree of resistance to rifampicin
determined by the absolute concentration method
using Middlebrook 7H9 broth, and the proportion
method using 1% Ogawa’s egg medium, and the mu-
tational sites determined by nucleotide sequencing,
high level resistance to rifampicin (50 ltg
mlF’~MIC) determined by the absolute concentra-
tion method coincided accurately with a mutation at
codons 513, 526, and 531. As the absolute concen-
tration method requires at least 1.5-2 months for
identification, the determination of a mutational
site might be useful and helpful for a rapid and ac-
curate diagnosis.
References
111Jin, D.J. and Gross, C.A. (1988) Mapping and sequencing of
mutations in the Escherichia coli rpoB gene that lead to rifam-
Downloaded from https://academic.oup.com/femsle/article/144/1/103/527475 by guest on 16 September 2023
picin resistance. J. Mol. Biol. 202, 45558.
PI Mitchison, D.A. and Nunn, A.J. (1986) Influence of initial
drug resistance on the response to short-course chemotherapy
of pulmonary tuberculosis. Am. Rev. Respir. Dis. 133, 423-
430.
131Vail-Spinosa, A., Lester, W.. Moulding, T., Davidson, P.T.
and McClatchy, J.K. (1970) Rifampin in the treatment of
drug-resistant Mycobacterium tuberculosis infections. New
Engl. J. Med. 283, 616621.
141Johnston, D.E. and McClure, W.R. (1975) Abortive initiation
of in vitro RNA synthesis on bacteriophage lambda DNA
polymerase. In: RNA Polymerase (R. Losick and M. Cham-
berlain, Eds.), pp. 413428. Cold Spring Harbor Laboratory.
Cold Spring Harbor, N.Y.
PI Lisitsyn, N.A., Sverdlov, E.D., Moiseyeva, E.P., Danilev-
skaya, O.N. and Nikiforov, V.G. (1984) Mutation to rifampi-
tin resistance at the beginning of the RNA polymerase B
subunit gene in Escherichia coli. Mol. Gen. Genet. 196, 173%
114.
PI Kapur, V., Li, L.L., Iordanescu, S., Hamrick, M.R., Wanger,
A., Kreiswirth, B.N. and Musser, J.M. (1994) Characteriza-
tion by automated DNA sequencing of mutations in the gene
(rpoB) encoding the RNA polymerase S subunit in rifampin-
resistant Mycobacterium tuberculosis strains from New York
City and Texas. J. Clin. Microbial. 32, 1095-1098.
[71Miller L.P., Crawford, J.T. and Shinnick, T.M. (1994) The
rpoB gene of Mycobacterium tuberculosis. Antimicrob. Agents
Chemother. 38, 805-811.
PI Ohno, H., Koga, H., Kohno, S., Higashiyama, Y., Miyazaki.
Y., Ogawa, K., Yanagihara, K., Yamamoto, Y., Noda, T.,
Miyamoto, T., Hashimoto, A., Tomono, K., Kaku, M. and
Hara, K. (1994) Evaluation for rapid detection of rifampicin
resistant Mycobacterium tuberculosis by polymerase chain re-
action-single strand conformation polymorphism. Kekkaku
69, 713- 778.
[9] Suzuki, Y., Katsukawa, C., Inoue, K., Yin, Y., Tasaka, H.,
Ueba, N. and Makino, M. (1995) Mutations in rpoB gene of
rifampicin resistant clinical isolates of Mycobacterium tubercu-
losis in Japan. Kansenshougaku-zasshi 69, 413419.
[lo] Telenti, A., Imboden, P., Marchesi, F., Lowrie, D., Cole, S.,
Colston, M.J., Matter, L., Schopfer, K. and Bodmer, T.
(1993) Detection of rifampicin-resistance mutations in Myco-
bacterium tuberculosis. Lancet 341, 647-650.
[II] Williams, D.L., Waguespack, C., Eisenach, K., Crawford,
J.T., Portaels, F., Salfinger, M., Nolan, C.M., Abe, C.,
Groh, VS. and Gillis, T.P. (1994) Characterization of rifam-
picin resistance in pathogenic mycobacteria. Antimicrob.
Agents Chemother. 38, 238&2386.