Gene Sequencing for Routine Verification of Pyrazinamide Resistance
in Mycobacterium tuberculosis: a Role for pncA but Not rpsA
David C. Alexander,a,bJennifer H. Ma,aJennifer L. Guthrie,aJoanne Blair,aPam Chedore,aand Frances B. Jamiesona,b
Public Health Ontario, Public Health Laboratories, Toronto, Ontario, Canada,aand University of Toronto, Department of Laboratory Medicine and Pathobiology, Toronto,
geted gene sequencing as a supplement to phenotypic PZA susceptibility testing of Mycobacterium tuberculosis. Routine se-
priate tuberculosis (TB) therapy and help prevent the emergence
and spread of drug-resistant Mycobacterium tuberculosis strains.
Pyrazinamide (PZA) is a front-line drug for the treatment of TB.
Administered during the 2-month, intensive phase of the stan-
dard short-course regimen, PZA is effective primarily against
slowly replicating bacilli and, thus, complements the activities of
isoniazid (INH) and rifampin (RIF), which are bactericidal for
rapidly replicating organisms.
PZA is a prodrug. Conversion to the active form, pyrazinoic
acid (POA), is mediated by the pyrazinamidase (PZase) encoded
by the pncA gene. It is well established that mutations in pncA can
mulation of POA (6). However, some PZA-resistant (PZAr)
strains have wild-type pncA (pncAWT) alleles. In such strains, re-
sistance has been proposed to result from altered PZA uptake,
9). Recently, it has been demonstrated that POA binding to the
30S ribosomal protein S1 inhibits the trans-translation activity
required for efficient protein synthesis (7). Mutations in rpsA,
which encodes the S1 protein, result in altered POA binding and
can mediate PZA resistance in pncAWTstrains.
is challenging (2). PZase activity and the intracellular accumula-
tion of POA increase with decreasing pH, but Mycobacterium tu-
Laboratory Standards Institute (CLSI) recommends the Bactec
460TB radiometric system with Bactec 460TB PZA test medium
(BD Diagnostics Systems, Sparks, MD) as the reference method
for phenotypic PZA susceptibility testing (1, 4). However, the
460TB system has been discontinued and the 460TB PZA test
medium is no longer being manufactured. Many laboratories, in-
cluding Public Health Ontario (PHO), have adopted the Bactec
MGIT 960 (BT960) platform for PZA testing. At our large public
health laboratory, the switch to BT960-based testing was accom-
strains that were PZArby the BT960 method but PZA-susceptible
(PZAs) according to the reference 460TB method (3). To ensure
accurate susceptibility results, confirmatory testing of potential
PZArisolates is necessary. A phenotypic strategy, involving a sec-
ond round of BT960-based testing, can be effective but does not
resolve all cases (3, 8), and repeat testing requires an additional 5
to 7 days to complete (1). In contrast, confirmatory testing using
have investigated the utility of targeted gene sequencing for rapid
verification of PZArresults.
To assess the utility of sequencing, archived DNA from 141
were from the PHO strain collection and were originally isolated
spoligotype, 12-locus mycobacterial interspersed repetitive-unit–
variable-number tandem-repeat (MIRU-VNTR) pattern, and
first-line drug susceptibility profiles of each isolate were deter-
the supplemental material). Review of the original PZA suscepti-
that 77 isolates were PZA sensitive (PZAs) and 64 were PZA resis-
tant (PZAr). For the current study, the pncA gene was amplified
using primers pnc1 (5=-GGCGTCATGGACCCTATATC-3=) and
pnc2 (5=-CAACAGTTCATCCCGGTTC-3=) (5). The 670-bp am-
plicon encompassed the complete pncA coding region and 80 bp
of upstream DNA. Individual PCRs were prepared with HotStar
cycler (G-Storm Ltd.; Somerset, United Kingdom) under the fol-
extension (72°C for 1 min), and a final extension at 72°C for 10
cycle sequencing (Life Technologies, CA) with the Applied Bio-
systems 3730xl DNA analyzer (Life Technologies). All PZAs
strains were wild type for pncA, although a previously recognized
variant allele (C195T, Ser65) was observed in a subset of East
African Indian lineage isolates (10). In contrast, pncA mutations,
including insertions, deletions, nonsynonymous changes, and
1; see also Table S1 in the supplemental material for additional
Received 5 March 2012 Returned for modification 16 April 2012
Accepted 7 August 2012
Published ahead of print 15 August 2012
Address correspondence to Frances B. Jamieson, firstname.lastname@example.org.
Supplemental material for this article may be found at http://jcm.asm.org/.
Copyright © 2012, American Society for Microbiology. All Rights Reserved.
jcm.asm.org Journal of Clinical Microbiologyp. 3726–3728 November 2012 Volume 50 Number 11
details). Eleven isolates were phenotypically PZArbut genotypi-
cally pncAWT. Of these discordant isolates, 5 were monoresistant
PZArand 6 were resistant to multiple first-line drugs. Eight of the
discordant isolates were successfully resurrected from frozen
stocks, and first-line susceptibility testing was repeated using the
BT960 method (3). In contrast to the original 460TB results,
BT960-based testing indicated that 5 of these strains were PZAs
and, thus, was concordant with pncA sequencing. A cause for the
discrepant BT960 and 460TB findings was not able to be deter-
mined, but the results indicate that the “gold standard” 460TB
method is not infallible. Of the three remaining discordant iso-
lates, one was PZA monoresistant and two were multidrug-resis-
from the resurrected strains, and upon resequencing, pncA muta-
tions were identified in both MDR-TB isolates. No mutations
were found in the final isolate, and this one monoresistant strain
remained phenotypically PZArand genotypically pncAWT.
To examine the potential contribution of rpsA mutations to
PZA resistance, 13 PZAsstrains and the 11 isolates initially con-
sidered discordant were analyzed (Table 2). To obtain full-length
rpsA sequences, five overlapping amplicons were generated by
PCR using the following primers: rspA-1F, 5=-ATGCCGAGTCC
CACCGTC-3=; rspA-1R, 5=-ACCCTTGACGACCTCGATGA-3=;
rspA-2F, 5=-AAACGCGCGCAGTACG-3=; rspA-2R, 5=-GTGAC
CTCGTCACCAACCTG-3=; rspA-3F, 5=-GACGGTCTGGTGCA
rspA-4F, 5=-ATGGCTTGAGGGATTCGAAAA-3=; rspA-4R, 5=-A
GCTTGAG-3=; and rspA-M5R, 5=-CTACTGGCCGACGACTGA
T-3=. PCR and sequencing conditions were identical to those
described for pncA except that a longer extension time (72°C
for 1.5 min) was used. A synonymous rpsA mutation (A636C,
indicate that this is not a lineage-specific mutation. One isolate
also harbored a second synonymous change (G960A, Leu320).
Nonsynonymous mutations were present in two strains. G1318A
(Ala440Thr)wasobservedinthe Mycobacteriumbovis BCGrefer-
ence strain and is conserved among publicly available genomes of
M. bovis and M. bovis BCG. PZA resistance in M. bovis is a recog-
nized phenomenon that is traditionally attributed to the pncA
A169G (His57Asp) mutation. The impact of the RpsA Ala440Thr
?Ala438, has been characterized and found to impair POA bind-
PZAr/pncAWTstrains bearing dual RpsA Thr5Ser and Asp123Ala
similar mutation, rpsA A364G (Lys122Glu), was present in one
the Thr5Ser, Asp123Ala, or Lys122Glu mutations impact POA
binding or simply represent regions of RpsA that tolerate amino
acid substitutions. Our phenotypic findings suggest that RpsA
may also be innocuous. However, sequence comparison indicates
that Thr5Ser is present in the RpsA ortholog of Mycobacterium
avium. Similarly, the RpsA ortholog of Mycobacterium canettii
strain CIPT 140010059 exhibits 2 changes, Thr5Ala and
Thr210Ala, relative to M. tuberculosis H37Rv. Strains of the M.
avium complex and M. canettii isolates are considered PZAreven
though they are genotypically pncAWTand exhibit PZase activity
(9, 11). The Thr5Ser/Thr5Ala variants may explain this phenom-
Although rpsA may contribute to PZA resistance, rpsA se-
TABLE 1 Phenotypic and genotypic characteristics of pyrazinamide-
resistant clinical isolatesa
No. of isolates with each drug resistance
Total no. of
isolatesPZA only Polyc
5 (1) 1 (0)3 (0)9 (1)
aValues in parentheses are the numbers of strains remaining after repeat testing of
discordant isolates. A total of 8 strains were excluded. Five strains were reclassified as
PZA sensitive, and three were not able to be retested.
bDrug susceptibility testing was performed on the 460TB (all 64 isolates) and BT960
(11 discordant isolates) platforms.
cStrains that are resistant to (at least) PZA and INH but not RIF.
dStrains that are resistant to (at least) PZA, INH, and RIF.
eXDR, extensively drug resistant.
fC418A and an 8-bp deletion were observed in all 3 isolates.
TABLE 2 Genotypic characterization of rpsA in clinical and reference isolatesa
No. of isolates with:
Total no. of
Synonymous changeNonsynonymous change
460TB resistant andb:
M. bovis BCG11
aThe susceptibility profiles of eight strains were retested.
bIncludes 11 discordant isolates initially considered PZA resistant and pncAWT.
cStrain is a double mutant (A636C and G960A) and counted only once.
pncA and rpsA Sequencing for PZA Testing
November 2012 Volume 50 Number 11jcm.asm.org 3727
no phenotypically informative mutations were identified. In con-
all but four PZArisolates and no PZAsstrains. This indicates that,
for Mycobacterium tuberculosis, DNA sequencing of pncA but not
rpsA is a robust tool for routine and rapid verification of PZA
We acknowledge the excellent work of the Public Health Ontario TB and
Mycobacteriology Laboratory (PHL-Toronto) staff responsible for the
initial isolation, cultivation, and susceptibility testing of the clinical iso-
lates used in the study.
1. Becton Dickinson and Company. 2002. Bactec MGIT 960 PZA kit for the
antimycobacterial susceptibility testing of Mycobacterium tuberculosis.
Becton Dickinson and Company, Sparks, MD.
2. Chang KC, Yew WW, Zhang Y. 2011. Pyrazinamide susceptibility testing
in Mycobacterium tuberculosis: a systematic review with meta-analyses.
Antimicrob. Agents Chemother. 55:4499–4505.
3. Chedore P, Bertucci L, Wolfe J, Sharma M, Jamieson F. 2010. Potential
for erroneous results indicating resistance when using the Bactec MGIT
960 system for testing susceptibility of Mycobacterium tuberculosis to pyr-
azinamide. J. Clin. Microbiol. 48:300–301.
4. CLSI. 2011. Susceptibility testing of mycobacteria, nocardiae, and other
aerobic actinomycetes; approved standard—second edition. CLSI docu-
ment M24-A2. Clinical and Laboratory Standards Institute, Wayne, PA.
5. Nguyen D, et al. 2003. Widespread pyrazinamide-resistant Mycobacte-
rium tuberculosis family in a low-incidence setting. J. Clin. Microbiol.
6. Scorpio A, Zhang Y. 1996. Mutations in pncA, a gene encoding pyrazi-
namidase/nicotinamidase, cause resistance to the antituberculous drug
pyrazinamide in tubercle bacillus. Nat. Med. 2:662–667.
7. Shi W, et al. 2011. Pyrazinamide inhibits trans-translation in Mycobac-
terium tuberculosis. Science 333:1630–1632.
8. Simons SO, et al. 2012. Validation of pncA gene sequencing in combina-
tion with the mycobacterial growth indicator tube method to test suscep-
tibility of Mycobacterium tuberculosis to pyrazinamide. J. Clin. Microbiol.
9. Somoskovi A, et al. 2007. Sequencing of the pncA gene in members of the
Mycobacterium tuberculosis complex has important diagnostic applica-
tions: identification of a species-specific pncA mutation in “Mycobacte-
rium canettii” and the reliable and rapid predictor of pyrazinamide resis-
tance. J. Clin. Microbiol. 45:595–599.
10. Stavrum R, Myneedu VP, Arora VK, Ahmed N, Grewal HM. 2009.
In-depth molecular characterization of Mycobacterium tuberculosis from
New Delhi—predominance of drug resistant isolates of the ‘modern’
(TbD1) type. PLoS One 4:e4540. doi:10.1371/journal.pone.0004540.
11. Sun Z, Scorpio A, Zhang Y. 1997. The pncA gene from naturally pyrazi-
fers pyrazinamide susceptibility to resistant M. tuberculosis complex or-
ganisms. Microbiology 143:3367–3373.
Alexander et al.
jcm.asm.org Journal of Clinical Microbiology