Mycobacteriology

Direct detection and identication of acid-fast bacteria from

smear-positive broth cultures using a pyrosequencing method

Jian R. Bao , Richard B. Clark, Ronald N. Master, Arelis E. Piscitelli, Praveena R. Tummala, Lynn L. Eklund,
Beatriz G. Poselero, Jackie Wright
Quest Diagnostics Nichols Institute, Chantilly, VA 20151
a b s t r a c t a r t i c l e i n f o
Article history:
Received 2 November 2013
Received in revised form 28 February 2014
Accepted 4 March 2014
Available online 15 March 2014
Keywords:
Acid-fast bacteria
Identication
Broth culture
Pyrosequencing
Broth culture is a standard method for detection of acid-fast bacteria (AFB) (e.g., Mycobacteriumand Nocardia)
from patient specimens. Direct nucleic acid-based identication from smear-positive broths expedites the
infectious disease diagnosis. We developed and evaluated the performance of a pyrogram-based technique
(direct-broth-pyrosequencing [DBP]) to identify AFB directly from smear-positive broths. One hundred
thirteen AFB-positive broths frompatient specimens were tested. Bacterial DNA was amplied by polymerase
chain reaction and sequenced using the PyroMark ID system. The DBP method correctly identied the AFB
species/group in 109 (97%) of the 113 broths, including 15 Mycobacterium species and 4 Nocardia species. Three
broths that yieldedindeterminateresults were foundtobeAFB-AFBmixedbroths andrequiredpuriedcolonies on
solid media for denite identication. The 4th broth was repeatedly identied by sequencing to be Mycobacterium
intracellulare, even though the organism was not isolated and the AccuProbe was negative. This method did not
identify the AFB organisms from broths containing 2 AFB organisms, but did not produce false identication. No
cross-reactionwas observedwhen AFB-positive broths were spiked withnon-AFB microorganisms, indicating that
the DBP method was specic to AFB. The DBP method gives rapid (within 8 h), accurate AFB identication directly
from broth cultures and provides another useful AFB identication tool in a clinical laboratory.
2014 Elsevier Inc. All rights reserved.
1. Introduction
Acid-fast bacteria (AFB) (e.g. Mycobacteriumand Nocardia species)
are usually signicant microorganisms in patient specimens. Their
rapid detection and identication are critical for optimal patient care.
Liquid (broth) culture systems recommended by the World Health
Organization (Nyendak et al., 2009), together with simultaneously
culturing on solid media, have become a standard method for AFB
detection in most clinical laboratories. Automated broth culture
systems expedite AFB-positive detection. The subsequent AFB
organism identication from AFB-positive broth cultures may still
be delayed since most available AFB identication methods require
developed colonies on solid media. The gold standard for AFB species
identication is largely relied on DNA sequences of specic target
genes, usually the 16S ribosomal RNA gene, in combination with
certain phenotypic characteristics in some cases. For most clinical
laboratories, the feasibility of AFB identication using sequences is
likely low. Further studies are need to establish the gold standard
sequencing method for identication directly from broth cultures.
Direct AFB identication from AFB-positive broth has been available
by using molecular or immunological techniques (Brent et al., 2011;
Ichiyama et al., 1997; Lu et al., 2011; Miller et al., 2002; Moon et al.,
2012; Pfyffer et al., 1994; Reischl et al., 1994; Saidet al., 2011; Shenet al.,
2011; Soini and Musser, 2001). The available methods are typically
targeting a single species (e.g., Mycobacterium tuberculosis complex
[MTBC]) or a limited number of AFB species (for example, AccuProbe
method with only 4 Mycobacterium species) and thus leave other AFB
organisms unidentied. The AccuProbe test, a direct DNA-probe
method (Gen-Probe, San Diego, CA, USA), is a widely-used method to
detect and identify AFB organisms directly frombroth cultures (Miller
et al., 1994; OSullivan et al., 2002; Telenti et al., 1994). The AccuProbe
method identies 4 groups of Mycobacterium species (MTBC,
Mycobacterium gordonae, Mycobacterium kansasii, and Mycobacterium
avium complex), but not other AFB organisms. The method is relative
labor intensive and expensive as it requires 1 probe for 1 targeted
organism. Newly emerging mass spectrometry methods, suchas matrix-
assisted laser desorption ionization/time of ight (MALDI-TOF), can
be a good option for rapid AFB identication (Balada-Llasat et al.,
2013), but lack sufcient data in patient cultures. Certain methods,
such as high performance liquid chromatography (HPLC) (Butler and
Kilburn, 1998), are not practical for most clinical laboratories as they
require signicant expertise. A rapid detection and identication
method that can simultaneously target various acid-fast organisms in
the genera Mycobacterium and Nocardia and other related AFB
organisms could be valuable in speeding diagnosis and thus potentially
improving patient care.
Diagnostic Microbiology and Infectious Disease 79 (2014) 228232
Conict of interest: The authors declare that they have no conict of interest.
Corresponding author. Tel.: +1-703-802-6900x65217; fax: +1-703-802-7017.
E-mail address: jian.r.bao@questdiagnostics.com (J.R. Bao).
http://dx.doi.org/10.1016/j.diagmicrobio.2014.03.003
0732-8893/ 2014 Elsevier Inc. All rights reserved.
Contents lists available at ScienceDirect
Diagnostic Microbiology and Infectious Disease
j our nal homepage: www. el sevi er . com/ l ocat e/ di agmi cr obi o
We developed a rapid method using polymerase chain reaction
(PCR) coupled with pyrosequencing (direct-broth-pyrosequencing
[DBP]) to detect and identify various AFB directly from broth cultures.
This study evaluated the feasibility of DBP in a single-shift clinical
laboratory setting.
2. Materials and methods
2.1. Culture media, AFB-positive specimens, and standard AFB
identication
Three types of broth media from different vendors were used for
AFB culture in this study: Microbial Growth Indicator Tubes (MGIT)
(BD Company, Towson, MD, USA), VersaTREK
TM
Myco (TREK
Diagnostic Systems, Cleveland, OH, USA), and BacT/Alert
TM
MP
(bioMerieux, Durham, NC, USA). The latter 2 broth types and some
MGIT broth cultures were processed and incubated by outside
laboratories and were submitted directly to the Nichols Institute for
AFB identication. Only broth cultures (no-agar cultures) were used
in this study. MGIT was the sole broth mediumused in this laboratory
for AFB cultures.
For the routine AFB culture at the Nichols Institute, patient
specimens of various sources were processed according to the
manufacturers instructions (Becton Dickinson) and microbiological
manual (Pfyffer and Palicova, 2011). Each processed specimen was
inoculated simultaneously into both MGIT broth and Lowenstein-
Jensen (L-J) slant (Becton Dickinson). The inoculated broth tubes were
incubated in a BACTEC 960 system(Becton Dickinson) at 3538 C, and
the inoculated L-J slants were incubated at 3537 C with 57% CO
2
.
Broth tubes identied as positive by the BACTEC 960 system were
subjected to a Ziehl-Neelsen staining procedure (Atlas and Snyder,
2011) to determine the presence of AFB (smear positive). The broth
cultures for this study were randomly selected fromthose AFB-positive
culture pools and negative cultures were not used. AFB identication
using the DBP method was assessed either immediately after the smear
procedure or held at roomtemperature (2025 C) for 35 days (a few
held up to 6 weeks). The DBP identication results were compared to
those obtained from established standard laboratory AFB identication
methods. The standard AFB identication methods used in this
laboratory included AccuProbe (Gen-Probe) and a pyrosequencing/
phenotypic method validated previously (Bao et al., 2010). The
AccuProbe method was used to identify the 4 Mycobacterium species
or groups (MTBC, M. gordonae, M. kansasii, M. avium complex) from
broth cultures according to the manufacturers instructions. The
pyrosequencing/phenotypic method was used to identify AFB from
colonies on L-J or Middlebrook 7H10 medium plates (Becton
Dickinson). Additional cultures on solid medium were obtained by
sub-culturing the positive broths onto new L-J slants, and the slants
were incubated at 3537 C with 57% CO
2
. Mixed AFB-AFB broth
cultures were subcultured to Middlebrook plates to obtain pure
isolated colonies for identication.
The organisms used in spiking experiments (AFB and non-AFB)
were type strains obtained fromthe American Type Culture Collection
(ATCC, Manassas, VA, USA) unless otherwise specied. AFB cultures
were 2-week-old cultures on L-J slants (Becton Dickinson) incubated
at 3537 C with 57% CO
2
. Non-AFB and yeast were cultured
overnight (24 h) at 3537 C on 5% sheep blood trypticase soy agar
plates (bacteria) or on potato dextrose agar (BD Company) plates
(yeast). The inoculums were prepared by suspending the organisms
in sterile MGIT broth and adjusting the organism cell density to 0.5
McFarland units, followed by serial dilutions.
2.2. DNA extraction from broth cultures
Bacterial cells were harvested by centrifugation of 0.31.0 mL of
the broth culture (Table 1) at 10,000 g for 5 min in a clean 2-mL
micro-centrifuge tube at room temperature. The bacterial cell pellet
was washed twice with 1000 L of sterile distilled water using the
suspension-centrifuging-suspension alternation procedure. The pel-
leted cells were re-suspended in 300 L DNA extract buffer (Bao et al.,
2010) and incubated in a heating block at 100 5 C for 10 min. The
DNA-containing supernatant was obtained by centrifugation of the
suspension at 10,000 g for 5 min at room temperature and used
immediately for PCR or held at 20 C for later testing.
2.3. DNA amplication, pyrosequencing, and data interpretation for AFB
identication
The DNA amplication, amplicon visualization, and pyrosequen-
cing procedures have been described previously (Bao et al., 2010). In
brief, 5 L of the DNA extract was used for each 50-L nal volume
PCR. The PCR was performed on an ATC 401 thermocycler (Nycktech,
San Diego, CA, USA) or a Mastercycler Pro thermocycler (Eppendorf,
Germany) for 40 cycles using a pair of PCR primers (forward: 5-
TGGCGAACGGGTGAGTAACA, reverse: 5-biotin-GCTACCCGTCGTCGC-
CTTG; synthesized by Invitrogen, Carlsbad, CA, USA). The amplicons
were visualized using the E-Gel system (Invitrogen). The PCR
amplicon was bonded to sepharose beads and denatured in 0.5N
NaOH buffer according to the manufacturers instructions (Qiagen,
Valencia, CA, USA). Pyrosequencing was performed on a PyroMark ID
system (Qiagen) using a sequencing primer (5-AAACTGGGTCTAA-
TACCG; synthesized by Invitrogen) (Tuohy et al., 2005) with 60 (15
4) nucleotide-dispensing cycles.
The AFB identication of this method is to use short signature
sequences (typically 3050 base pairs) to differentiate genus and
species. Resulting sequences containing a minimum of 30 nucleotide
bases were aligned to an in-house-developed database and a second
database developed by Michigan State Universitys RDP (http://rdp.
cme.msu.edu). The species-calling criteria were referring to those that
have been used in other related studies (Bao et al., 2010; Innings et al.,
2005; Tuohy et al., 2005). Sequence homology of 95% or greater to a
specic organism was considered acceptable for acid-fast organism
identication from a broth culture. All identications using this DBP
method were compared to those obtained using 1 or more of the
standard laboratory identication methods described above.
2.4. Impact of non-AFB microorganisms on AFB detection and
identication
Cross-reactivity with nonacid-fast organisms was tested by
spiking AFB-positive broths with inoculated non-AFB organisms.
AFB-positive MGIT broths were obtained by inoculating with either M.
chelonae (ATCC 35752) or M. fortuitum (ATCC 6841) at 10
5
bacterial
cells each. The inoculated broths were incubated in a BACTEC 960
system (Becton Dickinson) at 3538 C. Inoculated MGIT broths that
exhibited positive signals (in about 23 days) on the BACTEC system
were spiked with 1 of 6 nonacid-fast organisms (5 bacteria and 1
yeast) (Table 2). Each MGIT tube was spiked with 500 L of a 0.5-
McFarland bacterial cell suspension prepared from MGIT broth as
described above, and the nal non-AFB bacterial concentration in
tested MGIT tube was at approximately 5 10
6
cells/mL. The controls
Table 1
Broth volume used for DNA extraction based on visual bacterial density observation.
Scale Visual description for the broth culture Volume sampled for
DNA extraction (mL)
1 Culture is clear, no signicant sign of bacterial
growth in the broth.
1.0
2 Culture is clear, but visible sign of bacterial growth
on the bottom portion of the broth tube.
0.5
3 Culture somewhat cloudy with sign of heavy
bacterial growth.
0.3
229 J.R. Bao et al. / Diagnostic Microbiology and Infectious Disease 79 (2014) 228232
for the experiments consisted of either uninoculated MGIT broth
(blank) or broths spiked with both the above AFB organisms or with 1
nonacid-fast organism alone. Each spiked combination was per-
formed in triplicate. The DBP AFB identication procedure was
performed as described above.
2.5. Parallel detection and identication using both the DBP and
AccuProbe methods
AFB smearpositive MGIT broth cultures were blindly tested for direct
AFB identication fromthe broths, using both the DBP and the AccuProbe
method at the same time for each batch. Two batches of randomly
selected AFB-smear positive MGIT broths (37 totals) were tested on 2
different days by different technologists. The results reported here were
only from the single runs, and repeat data were not counted in this
experiment. Only 2 of the AccuProbe probes (for MTBC and MAC) were
used in this test. The DBP method was performed as described above.
3. Results
3.1. AFB detection and identication directly from AFB-positive
broth cultures
Among the 113 AFB smearpositive broth cultures tested in this
study, 71 were from sputum specimens (63%), 19 from bronchial
washing or aspirates (17%), 10 from tissues (9%), and 13 from
miscellaneous or unknown (11%). Of the positive cultures, 109 (97%)
were correctly identied to the species or species complex level using
the DBP method (Table 3). The identied organisms belonged to 15
Mycobacterium species or complexes, 4 Nocardia species or com-
plexes, and 1 Streptomyces species. The identication results were
conrmed with 1 or more validated standard laboratory identication
methods as described in the Materials and Methods. In comparison
with the standard identication methods from developed colonies,
the DBP identication time saved for other AFB organisms ranged
from 6 to more than 40 days (data not shown).
Three (3%) MGIT broth cultures were identied as indeterminate
initially by the DBP method (Table 4), since either the sequence
homology was below 95% for denite identication, or the second
organism was not isolated on solid media. In any case, no false
identication to an unrelated organism was produced by the DBP
method among the mixed broths. Two indeterminate broth cultures
were later found to contain 2 acid-fast organisms each; thus,
approximately 2% of the broths tested contained AFB-AFB mixed
cultures. The acid-fast organisms fromeach mixed broth were isolated
on solid media and identied using standard laboratory identication
methods (Table 4). The third indeterminate broth culture was
identied as Mycobacterium lentiavum by the DBP method but was
positive for M. gordonae by the AccuProbe method. The colonies
recovered on L-J medium slants had morphology typical of M.
lentiavum and were identied as M. lentiavum in 3 separate tests
using the standard laboratory identication method as described
above. M. gordonae was not isolated from the broth on either L-J
mediumslants or Middlebrook mediumplates. The 4th indeterminate
smear-positive broth culture was identied as M. intracellulare using
the DBP method. This culture was negative by AccuProbe, and the
organism was not recovered after repeated subcultures on L-J slants.
The DBP method exhibited no AFB identication discrepancy
among the 3 different types of broth used in this study. Sixteen
different AFB species or groups were identied from 93 MGIT broth
cultures, 7 different AFB organisms from a total of 7 VersaTREK
TM
Myco (TREK) cultures, and 6 AFB organisms from a total of 12 Bact/
Alert
TM
MP (bioMerieux) bottles. All the last 2 types of positive broth
cultures were referred from outside laboratories.
Table 2
Identication of acid-fast organisms by DBP in broth cultures spiked with nonacid-fast organisms.
Inoculated organism(s) Spiking organism for each treatment Identication by DBP (match %)
M. chelonae (ATCC 35752) or M. fortuitum (ATCC 6841) Bacillus cereus (ATCC 10867) M. chelonae (100) or M. fortuitum (100)
Escherichia coli (ATCC 25922)
Staphylococcus aureus (ATCC 25923)
Enterococcus faecalis (ATCC 29212)
Pseudomonas aeruginosa (ATCC 27853)
Candida albicans (ATCC 14055)
(Blank control)
M. chelonae (ATCC 35752) + M. fortuitum (ATCC 6841) None M. fortuitum (94)
None Bacillus cereus (ATCC 10867) No ID
Escherichia coli (ATCC 25922)
Staphylococcus aureus (ATCC 25923)
Enterococcus faecalis (ATCC 29212)
Pseudomonas aeruginosa (ATCC 27853)
Candida albicans (ATCC 14055)
(Blank control)
Table 3
AFB organisms identied directly from smear-positive broths using the DBP method.
Organism ID by DBP No. SH, %
a
Mycobacterium spp. (Total: 104)
MTBC 17 100
M. avium 9 100
M. chelonae-abscessus group 12 100
M. fortuitum 7 100
M. gilvum 2 100
M. gordonae 11 100
M. intracellulare 18 100
M. kansasii 6 100
M. lentiavum 2 100
M. magaritense 1 100
M. marinum 3 100
M. mucogenicum group 5 100
M. nebraskense/seoulense group 1 100
M. asiaticum 1 100
M. xenopi 9 100
Nocardia spp. (Total: 4)
N. farcinica 1 100
N. nova 1 100
N. nova complex 1 100
N. wallacei 1 100
SH = sequence homology.
a
Some data were obtained from repeated assays. The species of M. kansasii was
determined by using a sequencing primer 2 (Qiagen) in pyrosequencing (Bao et al.,
2010). The databases used here are described in Materials and Methods section.
230 J.R. Bao et al. / Diagnostic Microbiology and Infectious Disease 79 (2014) 228232
3.2. Impact of non-AFB organisms on AFB detection and identication by
the DBP method
The two pre-inoculated acid-fast organisms (M. chelonae and M.
fortuitum) were correctly identied from the MGIT broths that had
been heavily spiked with nonacid-fast organisms (Table 2). No
misidentication of AFB organisms was observed from AFB broths
spiked with non-AFB organisms or from broths containing non-AFB
organisms alone. The presence of nonacid-fast organisms in the AFB-
positive broth cultures had no impact on the correct identication by
this method in this study. The MGIT broths that had been inoculated
with both M. chelonae and M. fortuitum produced a result of 94%
sequence homology to M. fortuitum, below the 95% homology
threshold for correct identication. The method did not misidentify
the mixed organisms as unrelated organisms.
3.3. Parallel detection and identication using both the DBP and
AccuProbe methods
Among the 37 positive broth cultures tested for AFB identication
using both the DBP and AccuProbe methods, the DBP method
correctly detected and identied AFB from 35 (95%) to the species
or species complex level (Table 5). The DBP method was able to
differentiate the M. avium/M. intracellulare complex (MAC) into
individual species of either M. avium or M. intracellulare. The method
failed to identify acid-fast organisms in 2 smear-positive broth
cultures in the rst run. One culture was later conrmed to contain
tuberculosis (TB) complex and the other to contain M. avium; both
were identied correctly in a repeat run. The AccuProbe method using
2 probes (MTBC and MAC) detected and correctly identied acid-fast
organisms from 20 broth cultures (55%).
Of the 37 broth cultures, 28 contained AFB organisms of either
MTBC or MAC that were detectable by both the AccuProbe (2 probes)
and the DBP methods. The other 9 broth cultures grew other AFB
organisms that were detectable by this DBP method, but not by the
AccuProbe method. Among the AFB organisms from the 28 broth
cultures detectable by both methods, the AccuProbe method correctly
identied acid-fast organisms from 20 of the 28 broth cultures (71%)
and failed to identify organisms from 8 broth cultures (2 MTBC and 6
MAC). The DBP method correctly identied acid-fast organisms from
26 of the 28 broth cultures (93%) and failed to identify organisms from
2 broth cultures.
4. Discussion
AFB are very important microorganisms for patients as well as
public health. The DBP method developed in this study provides
another valuable tool to detect and identify acid-fast organisms
directly from positive broth cultures. The whole procedure can be
nished within a single 8-h shift. It can detect and identify a wide
variety of AFB organisms to species/groups of Mycobacterium and
Nocardia, as well as of Rhodococcus, Tsukamurella, and Gordonia
(data shown here). In comparison, the currently available methods
for direct identication from broths are typically to target either a
single species or complex (MTBC) (Ismail et al., 2009; Miller et al.,
2002; Moon et al., 2012; Said et al., 2011) or certain dened AFB
species (AccuProbe test, Miller et al., 1994; OSullivan et al., 2002;
Telenti et al., 1994). This direct AFB identication method
described here expedites the identication speed to a single day
from positive broth cultures. Developed colony on a solid medium
becomes not a prerequisite as do many other available AFB
identication methods.
Nonacid-fast microorganismcontamination in AFB broth cultures
accounts for 2% to more than 10% of cultures and can interfere with
the correct identication of targeted organisms (Corneld et al., 1997;
Zheng et al., 2001). In this study, AFB-positive tubes were spiked with
high concentrations of non-AFB or yeast (approximately 10
6
non-
AFB bacterial cells/mL). Both of the targeted acid-fast organisms (M.
chelonae and M. fortuitum) in the spiked broths were correctly
identied using the DBP method (Table 2). Under actual culturing
conditions, non-AFB concentration in the broths could be relatively
low as the culturing MGIT broth contains a mixture of 5 antibiotics to
inhibit non-AFB growth (Package Insert of the product; BD). These
results indicate that the DBP method is specic to acid-fast organisms.
Table 5
Single parallel run to detect and identify acid-fast organisms directly from AFB smearpositive MGIT broths using both DBP and AccuProbe methods.
a
AFB organism detected Total
MGIT,
number
a
Identied correctly Final
ID
c
Number missed
AccuProbe
b
DBP AccuProbe (%) DBP (%)
TB complex 6 4 5 6 2 (33) 1 (17)
M. intracellulare 13 16 (MAC) 12 13 6 (27) 0
M. avium 9 9 9 1 (5)
M. gordonae 4 NA 4 4 0
M. chelonae-abscessus group 2 NA 2 2 0
M. fortuitum 1 NA 1 1 0
M. kansasii 2 NA 2 2 0
Total 37 20 35 37 8 2 (5)
a
The results were based on a single run for either method. The correct identication results were obtained on repeated assays.
b
Two types of probes were used in this study for either TB complex or MAC.
c
Final ID is based on the standard laboratory identication methods as described in Materials and Methods.
Table 4
Other or undetermined bacterial organisms identied directly from broth cultures using the DBP method.
ID by DBP (SH, %) No ID by other methods Final identication
Streptomyces sp. (100) 1 Streptomyces sp. by phenotypic characteristics and AFB staining Streptomyces sp.
M. intracellulare (93) 1 M. gordonae by AccuProbe.
Two acid-fast organisms were obtained on solid media.
M. intracellulare and M. gordonae
Tsukamurella sp. (93) 1 Two acid-fast organisms were isolated on solid media. M. fortuitum and Gordonia sp.
M. lentiavum (100) 1 M. gordonae by AccuProbe.
Only M. lentiavum was isolated on solid media.
M. gordonae and M. lentiavum
M. intracellulare (100) 1 Negative by sub-culture and AccuProbe
a
M. intracellulare
a
The culture was not available for repeat testing.
231 J.R. Bao et al. / Diagnostic Microbiology and Infectious Disease 79 (2014) 228232
The presence of non-AFB organisms did not impair AFB identication
with this method, though the number of non-AFB organisms tested in
this study was small. The interference of contaminations from non-
AFB organisms to AFB identication was not observed in these
experiments. The such specicity to AFB organisms of this method
was also evidenced by the fact that all 109 identied patient broths
did not exhibit misidentication due to possible presence of non-AFB
organisms in the broth cultures.
AFB identications were indeterminate in three positive broths.
The identication results from all 3 of these broths had sequence
homology belowthe 95% threshold for correct identication of an AFB
organism. The DBP method did not incorrectly identify other
unrelated organisms in any of the 3 broths. These results indicate
that the DBP method did not misidentify the organisms in AFB-AFB
mixed broths. Similar results were also observed in spiked experi-
ments where 2 different acid-fast organisms were growing in a single
broth (Tables 24). For broths containing 2 or more acid-fast
organisms, isolation of each organism is required, or an alternative
identication method must be used. In this sense, the DBP method
should be a robust method to detect and identify acid-fast organisms
directly from broth cultures. Identications with sequence homology
below 95% should be interpreted with caution, as this might indicate
the presence of more than 1 acid-fast organism in the same broth.
In 1 unresolved case, a broth was identied as containing
M. lentiavum by DBP but as M. gordonae by the AccuProbe method.
M. lentiavumwas repeatedlyidentiedfromcolonies, andnoM. gordonae
was recovered on solid media fromthis broth. A similar scenario was also
observed in other reported procedures (OSullivan et al., 2002). This
couldbe causedbya lowcopynumber (overgrowthby therst organism)
or dormancy of the second organism in the culture.
The detection sensitivity of the DBP method was not specically
tested. This method was designed to detect and identify the AFB
organisms from AFB smear positive broth cultures. These results
described here indicate that the detection limit of the DBP method
was sufcient to accurately identify AFB organisms in AFB-positive
broths dened by the BACTEC system and conrmed by AFB smear.
In summary, the DBP method is a rapid, sensitive, and accurate
method to detect and identify acid-fast organisms directly from positive
broth cultures. It can identify MTBC and other acid-fast organisms,
including Nocardia species. The procedure should prove to be a valuable
tool in AFB identication for the clinical laboratory and patient care.
Acknowledgments
We would like to thank the personnel in our Microbiology
Department who provided technical assistance and Dr Kileen Shier
for manuscript review. We wouldlike toespecially thank Mr Jeff Radcliff
for his critical review and recommendations for the manuscript.
References
Atlas RM, Snyder JW. Reagents, stains, and media: Bacteriology. In: Versalovic J,
Jorgensen JH, et al, editors. Manual of clinical microbiology. 10th ed. Washington
DC: ASM Press; 2011. p. 272303.
Balada-Llasat JM, Kamboj K, Pancholi P. Identication of mycobacteria from solid and
liquid media by matrix-assisted laser desorption/ionization time-of-ight mass
spectrometry in the clinical laboratory. J Clin Microbiol 2013;51:28759.
Bao JR, Master RN, Schwab DA, Clark RB. Identication of acid-fast bacilli using
pyrosequencing analysis. Diagn Microbiol Infect Dis 2010;67:2348.
Brent AJ, Mugo D, Musyimi R, Mutiso A, Morpeth S, Levin M, et al. Performance of the
MGIT TBc identication test and meta-analysis of MPT64 assays for identication of
the Mycobacterium tuberculosis complex in liquid culture. J Clin Microbiol 2011;49:
43436.
Butler WR, Kilburn JO. Identication of major slowly growing pathogenic Mycobacteria
and Mycobacterium gordonae by high-performance liquid chromatography of their
mycolic acids. J Clin Microbiol 1998;26:503.
Corneld DB, Beavis KG, Greene JA, Bojak M, Bondi J. Mycobacterial growth and
bacterial contamination in the Mycobacteria Growth Indicator Tube and BACTEC
460 culture systems. J Clin Microbiol 1997;35:206871.
Ichiyama S, Iinuma Y, Yamori S, Hasegawa Y, Shimokata K, Nakashima N. Mycobac-
terium growth indicator tube testing in conjunction with the AccuProbe or the
AMPLICOR-PCR assay for detecting and identifying Mycobacteria from sputum
samples. J Clin Microbiol 1997;35:20225.
Innings A, Krabbe M, Ullberg M, Herrmann B. Identication of 43 Streptococcus species
by pyrosequencing analysis of the rnpB gene. J Clin Microbiol 2005;43:598391.
Ismail NA, Baba K, Pombo D, Hoosen AA. Use of an immunochromatographic kit for the
rapid detection of Mycobacterium tuberculosis frombroth cultures. Int J Tuberc Lung
Dis 2009;13:13.
Lu PL, Yang YC, Huang SC, Jenh YS, Lin YC, Huang HH, et al. Evaluation of the Bactec
MGIT 960 system in combination with the MGIT TBc identication test for
detection of Mycobacterium tuberculosis complex in respiratory specimens. J Clin
Microbiol 2011;49:22902.
Miller N, Cleary T, Kraus G, Young AK, Spruill G, Hnatyszyn HJ. Rapid and specic
detection of Mycobacterium tuberculosis from acid-fast bacillus smear-positive
respiratory specimens and BacT/ALERT MP culture bottles by using uorogenic
probes and real-time PCR. J Clin Microbiol 2002;40:41437.
Miller N, Hernandez SG, Cleary TJ. Evaluation of Gen-Probe amplied Mycobacterium
tuberculosis direct test and PCR for direct detection of Mycobacteriumtuberculosis in
clinical specimens. J Clin Microbiol 1994;32:3937.
Moon HW, Hur M, Kim H, Kim JY, Park CM, Yun YM. Isothermal target and probe
amplication for Mycobacteriumtuberculosis identication frombroth cultures. Int J
Tuberc Lung Dis 2012;16:51620.
Nyendak MR, Lewinsohn DA, Lowinsohn M. New diagnostic methods for tuberculosis.
Curr Opin Infect Dis 2009;22:17482.
OSullivan CE, Miller DR, Schneider PS, Roberts GD. Evaluation of Gen-Probe
amplied Mycobacterium tuberculosis direct test by using respiratory and
nonrespiratory specimens in a tertiary care center laboratory. J Clin Microbiol
2002;40:17237.
Pfyffer GE, Kissling P, Wirth R, Weber R. Direct detection of Mycobacterium tuberculosis
complex in respiratory specimens by a target-amplied test system. J Clin
Microbiol 1994;32:91823.
Pfyffer GE, Palicova F. Mycobacterium: General characteristics, laboratory detection, and
staining procedures. In: Versalovic J, Jorgensen JH, et al, editors. Manual of Clinical
Microbiology. 10th edn. Washington DC: ASM Press; 2011. p. 472502.
Reischl U, Pulz M, Ehret W, Wolf H. PCR-based detection of mycobacteria in sputum
samples using a simple and reliable DNA extraction protocol. Biotechniques
1994;17:8445.
Said HM, Ismail N, Osman A, Velsman C, Hoosen AA. Evaluation of TBcID
immunochromatographic assay for rapid identication of Mycobacterium tubercu-
losis complex from broth cultures. J Clin Microbiol 2011;49:193942.
Shen GH, Chiou CS, Hu ST, Wu KM, Chen JH. Rapid identication of the Mycobacterium
tuberculosis complex by combining the ESAT-6/CFP-10 immunochromatographic
assay and smear morphology. J Clin Microbiol 2011;49:9027.
Soini H, Musser JM. Molecular Diagnosis of Mycobactera. Clin Chemistry 2001;47:
80914.
Telenti M, de Quiros JF, Alvarez M, Santos Rionda MJ, Mendoza MC. The diagnostic
usefulness of a DNA probe for Mycobacterium tuberculosis complex (Gen-Probe) in
Bactec cultures versus other diagnostic methods. Infection 1994;22:1823.
Tuohy MJ, Hall GS, Sholtis M, Procop GW. Pyrosequencing as a tool for the identication
of common isolates of Mycobacterium sp. Diagn MicrobiolInfect Disease 2005;51:
24550.
Zheng X, Pang M, Engler HD, Tanaka S, Reppun T. Rapid detection of Mycobacterium
tuberculosis in contaminated BACTEC 12B broth cultures by testing with amplied
Mycobacterium tuberculosis direct test. J Clin Microbiol 2001;39:371820.
232 J.R. Bao et al. / Diagnostic Microbiology and Infectious Disease 79 (2014) 228232