ENVIRONMENTAL MICROBIOLOGY

crossm

Real-Time PCR Method for Detection of

Salmonella spp. in Environmental
Samples
Kuppuswamy N. Kasturi,a Tomas Drgonb
Northeast Regional Laboratory, U.S. Food and Drug Administration, Jamaica, New York, USAa; Office of
Policy and Risk Management, Office of Regulatory Affairs, U.S. Food and Drug Administration, Rockville,
Maryland, USAb

ABSTRACT The methods currently used for detecting Salmonella in environmental

samples require 2 days to produce results and have limited sensitivity. Here, we de- Received 17 March 2017 Accepted 2 May
2017
scribe the development and validation of a real-time PCR Salmonella screening
Accepted manuscript posted online 12 May
method that produces results in 18 to 24 h. Primers and probes specific to the gene 2017
invA, group D, and Salmonella enterica serovar Enteritidis organisms were designed Citation Kasturi KN, Drgon T. 2017. Real-time
and evaluated for inclusivity and exclusivity using a panel of 329 Salmonella isolates PCR method for detection of Salmonella spp. in
representing 126 serovars and 22 non-Salmonella organisms. The invA- and group environmental samples. Appl Environ
Microbiol 83:e00644-17. https://doi.org/10
D-specific sets identified all the isolates accurately. The PCR method had 100% inclu- .1128/AEM.00644-17.
sivity and detected 1 to 2 copies of Salmonella DNA per reaction. Primers specific for Editor Charles M. Dozois, INRS—Institut
Salmonella-differentiating fragment 1 (Sdf-1) in conjunction with the group D set Armand-Frappier
had 100% inclusivity for 32 S. Enteritidis isolates and 100% exclusivity for the 297 non- This is a work of the U.S. Government and is
not subject to copyright protection in the
Enteritidis Salmonella isolates. Single-laboratory validation performed on 1,741 envi- United States. Foreign copyrights may apply.
ronmental samples demonstrated that the PCR method detected 55% more posi- Address correspondence to Kuppuswamy N.
tives than the Vitek immunodiagnostic assay system (VIDAS) method. The PCR Kasturi, [email protected].
results correlated well with the culture results, and the method did not report any
false-negative results. The receiver operating characteristic (ROC) analysis docu-
mented excellent agreement between the results from the culture and PCR methods
(area under the curve, 0.90; 95% confidence interval of 0.76 to 1.0) confirming the
validity of the PCR method.
IMPORTANCE This validated PCR method detects 55% more positives for Salmonella
in half the time required for the reference method, VIDAS. The validated PCR
method will help to strengthen public health efforts through rapid screening of Sal-
monella spp. in environmental samples.
KEYWORDS S. Enteritidis, environmental samples, Salmonella, validation of PCR
method, real-time PCR

S almonella is one of the major causes of foodborne illnesses, with the U.S. Centers for
Disease Control and Prevention (CDC) estimating that ⬃1.2 million cases of salmo-
nellosis occur annually in the United States alone. Salmonella infections cause severe
illness in infants, the elderly, and immunocompromised individuals. Most infections in
humans result from eating contaminated food or drinking contaminated water. Sources
of Salmonella infections include foods of animal origin, dairy products, pet food, fresh
produce, and foods contaminated during processing. Moreover, Salmonella outbreaks
following the consumption of eggs contaminated with Salmonella enterica serovar
Enteritidis are relatively common (1).
The expeditious detection of Salmonella species in food and environmental samples
requires rapid, efficient, and validated methods. A number of conventional and real-
time PCR methods have been reported for the detection of Salmonella in naturally or
artificially contaminated food samples (2–9) and fecal samples (10). A few validation

July 2017 Volume 83 Issue 14 e00644-17 Applied and Environmental Microbiology aem.asm.org 1
Kasturi and Drgon Applied and Environmental Microbiology

studies have been reported for food samples that were spiked with Salmonella (8, 9, 11,
12, 42), but no validation studies have been reported for environmental samples. The
primer sets that have been used for the amplification of Salmonella DNA by different
methods differed in their detection limits, lacked disclosure of the internal amplification
control (IAC) DNA sequence used (8, 9, 11–14), or did not include a wide range of
epidemiologically important isolates. In addition, the matrices and levels of contami-
nating competing organisms present in each environmental sample appeared to be
unique, making the validation of methods for screening environmental samples a
challenge.
The presence of high concentrations of PCR inhibitors in samples (15, 16) and the
lack of adequate validation studies appear to be major factors underlying the lack of
use of rapid PCR methods for routine screening. Hence, most public health organiza-
tions and commercial food-processing firms use an immunological screening method,
the Vitek immunodiagnostic assay system (VIDAS), for detecting Salmonella in environ-
mental samples and report presumptive results after 48 h (17). Since the detection of
Salmonella by VIDAS requires thousands of target organisms in test samples (17), the
probability of missing potentially positive samples is relatively high. A recent study
reported that VIDAS failed to detect the presence of Salmonella in a number of spiked
samples, although the Salmonella sp. used to spike the samples was in fact isolated
from those samples (18). Notably, nonspiked VIDAS-negative samples are not usually
tested by the culture method. Furthermore, Salmonella isolates were not recovered
from a number of VIDAS-positive samples, because the VIDAS reagent appeared to
react with O-antigen epitopes of Enterobacter cloacae, Escherichia coli, etc., present in
the samples (19). Together, these data indicate the need for the development of
alternate rapid, sensitive, and reliable screening methods.
Several PCR methods, both conventional and real time, have been reported for the
identification of the O groups A, B, C1, C2, D, and E (20, 21, 41) and Salmonella enterica
serovars, such as Enteritidis, Typhimurium, Typhi, Paratyphi A, and Paratyphi B (20,
22–24). A multiplex bead-based suspension array method has also been reported for
the determination of common O groups of Salmonella isolates (25). All these methods
require Salmonella isolates as the starting material for such identification. At present,
there is no rapid molecular method for the direct screening of Salmonella or the S.
Enteritidis serovar in environmental samples.
This study reports the development of a real-time PCR method that detected 126
epidemiologically important serovars belonging to all subspecies of Salmonella en-
terica. An IAC was included in the PCR assays to detect the presence of PCR inhibitors
and reduce false-negative results (15, 16). The method reported herein was designed to
be selective and did not detect organisms that were cross-reactive in antibody-based
assays. The results obtained from screening 1,741 environmental samples were com-
pared with those obtained with the reference method, VIDAS (17, 26). All samples that
were positive based on either screening method were followed up using the culture
method, as described in the Bacteriological Analytical Manual (27). The results demon-
strate that the new PCR method described here is sensitive, specific, and reliable for
screening environmental samples in less than half the time required for the standard
VIDAS method. This method could be a valuable tool for the detection of S. Enteritidis
directly in environmental samples collected from poultry farms.

RESULTS
Limits of detection. The sensitivity of the PCR method was determined using
purified positive-control DNA from S. Enteritidis at concentrations ranging from 1 to 105
copies per reaction in quadruplicate reactions. The results from a representative assay
(Table 1) demonstrated that the invA primer-probe set detected the presence of a
single copy of Salmonella genomic DNA with a probability of 50%, whereas two or more
copies of the genome were detectable with a probability of 100%. The assay was
equally sensitive for detecting Salmonella group D and Salmonella Enteritidis targets.
The logarithm of the number of Salmonella DNA copies in a sample correlated ex-

July 2017 Volume 83 Issue 14 e00644-17 aem.asm.org 2

Evaluation of Real-Time PCR for Salmonella Species Applied and Environmental Microbiology

TABLE 1 Sensitivity of real-time PCRa

invA CT valueb
No. of DNA Mean ⴞ SD
copies Expt 1 Expt 2 Expt 3 Expt 4 invA CT value
105 22.48 23.29 23.14 23.16 23.02 ⫾ 0.37
104 26.63 26.74 26.77 26.90 26.76 ⫾ 0.11
103 29.74 29.79 29.56 29.50 29.65 ⫾ 0.14
102 33.31 33.06 32.35 32.99 32.93 ⫾ 0.41
10 36.29 35.45 36.11 36.46 35.83 ⫾ 0.44
2 37.73 36.53 36.27 36.33 36.72 ⫾ 0.69
1 38.26 Neg Neg 38.70 38.48 ⫾ 0.31
aThe lower limit of detection by the PCR method was determined by testing various concentrations of
Salmonella Enteritidis DNA, ranging from 1 to 105 copies per reaction. The assay was performed in
quadruplicate.
bThe C values were obtained from the 4 assays and represent the number of PCR cycles at which the
T
fluorescence generated crossed the minimum threshold set for the assay. Neg, negative (no CT value was
displayed).

tremely well with the invA threshold cycle (CT) value that represented the DNA copy
number in the PCR method (Pearson correlation r ⫽ ⫺0.99). The correlation was linear
over the entire range of DNA concentrations analyzed (from a single copy to 105
copies), suggesting that the PCR method can be used to quantify the number of
Salmonella cells present in samples.
Validation of the PCR method. The exclusivity of the primers and probes used in
the multiplex PCR was confirmed by testing 105 copies of DNA from each of the 22
non-Salmonella target organisms (Table 2). The inclusivity of the primers and probes
used was first confirmed by testing a panel of 12 control Salmonella organisms and
then extended to testing a random panel of 329 Salmonella isolates obtained from food
and environmental samples. The results demonstrated that the Salmonella-specific invA
primer-probe set identified all 329 Salmonella isolates that contained 126 serovars

TABLE 2 Exclusivity of PCR primers and probes was confirmed by testing 105 copies of
purified DNA from each of the 22 non-Salmonella target organisms in the multiplex real-
time PCR
Result by PCR targeta
Control organism invA prt Sdf-1 IAC
Escherichia coli (ATCC 8759) Neg Neg Neg Pos
E. coli (ATCC 43888) Neg Neg Neg Pos
E. coli (ATCC 43895) Neg Neg Neg Pos
E. coli (ATCC 6209) Neg Neg Neg Pos
E. coli (ATCC 6210) Neg Neg Neg Pos
Shigella flexneri (ATCC 29903) Neg Neg Neg Pos
Cronobacter sakazakii (ATCC 12868) Neg Neg Neg Pos
Enterobacter aerogenes (ATCC 13048) Neg Neg Neg Pos
Enterobacter cloacae (lab isolate) Neg Neg Neg Pos
Citrobacter freundii (lab isolate) Neg Neg Neg Pos
Klebsiella pneumoniae (ATCC 13883) Neg Neg Neg Pos
Proteus hauseri (ATCC 13315) Neg Neg Neg Pos
Pseudomonas aeruginosa (ATCC 9027) Neg Neg Neg Pos
Pantoea sp. (lab isolate) Neg Neg Neg Pos
Vibrio cholerae (ATCC 35971) Neg Neg Neg Pos
Yersinia enterocolitica (ATCC 23715) Neg Neg Neg Pos
Acinetobacter sp. (lab isolate) Neg Neg Neg Pos
Bacillus subtilis (ATCC 6633) Neg Neg Neg Pos
Staphylococcus aureus (ATCC 6538) Neg Neg Neg Pos
Rhodococcus equi (ATCC 6939) Neg Neg Neg Pos
Listeria monocytogenes (ATCC 19115) Neg Neg Neg Pos
Listeria innocua (ATCC 33090) Neg Neg Neg Pos
No-template control Neg Neg Neg Pos
S. Enteritidis (ATCC 13076) Pos Pos Pos Pos
aTheresults shown here were obtained after 40 cycles of PCR. Pos, positive (displayed a CT value in the PCR);
Neg, negative (displayed no CT value in the PCR).

July 2017 Volume 83 Issue 14 e00644-17 aem.asm.org 3

Kasturi and Drgon Applied and Environmental Microbiology

TABLE 3 Summary of Salmonella isolates tested for the inclusivity PCRa

No. of isolates No. of serovars
O group tested per O group
35 1 1
39 2 1
51 3 1
61 3 2
B 47 14
C1 61 21
C2 35 10
D1 54 10
D2 3 2
E1 40 8
E2 2 1
E4 9 2
F 4 3
G1 3 3
G2 4 4
H 4 4
I 14 8
J 1 1
K 2 1
L 3 2
M 8 6
N 2 2
O 1 1
P 5 2
Q 2 2
R 1 1
S 2 2
T 2 2
U 4 2
V 2 2
W 1 1
X 2 2
Y 1 1
Z 1 1

Total 329 126

aThepanel included 126 serovars from 34 O groups. Fifty-seven isolates were from the D1 and D2 O groups,
of which 32 were S. Enteritidis.

belonging to all subspecies of S. enterica (Table 3). Similarly, the group D-specific
primers and probe (prt) identified all 57 group D isolates in the panel (contained 12
distinct serovars of D1 and D2 groups) and did not react with the remaining 272 isolates
that were not group D. The tyv primer and probe set gave results similar to those for
the prt primer-probe set, thus establishing that either the prt or tyv primer-probe set
could be used for identifying group D serovars. Notably, both the prt- and tyv-specific
reagent sets recognized all serovars that contained the antigenic factor O-9, including
those isolates that were identified as the antigenic formula and listed under group D
by the WHO Collaborating Centre for Reference and Research on Salmonella (28). The
S. Enteritidis-specific Sdf-1 primer-probe set recognized all 32 S. Enteritidis serovars
included in the panel. However, the Sdf-1 primer-probe set reacted with one out of 297
non-Enteritidis Salmonella isolates, thus exhibiting an exclusivity of 99.7% for non-
Enteritidis Salmonella serovars.
PCR, VIDAS, and culture results of environmental samples. A total of 1,741
environmental samples collected from 14 different food manufacturing companies
were screened by PCR and VIDAS assays, and culture analyses were performed on all
samples that were positive with either the VIDAS or PCR. Of 1,741 samples tested, the
reference method, VIDAS, detected Salmonella in 12 samples, PCR detected Salmonella
in 11 samples, and the culture method found Salmonella in 9 samples (Table 4). None
of the environmental samples tested showed an inhibitory effect in PCR, as evidenced
by the PCR results of the IAC. Of the 12 VIDAS-positive samples, Salmonella was found

July 2017 Volume 83 Issue 14 e00644-17 aem.asm.org 4

Evaluation of Real-Time PCR for Salmonella Species Applied and Environmental Microbiology

TABLE 4 Comparison of results obtained from the PCR, VIDAS, and culture methodsa
PCR result VIDAS
Source (no. of samples) or CT value result Culture resultb
Cheese product plant (132) 22.49 Negative Positive
38.13 Negative Positive
37.79 Negative Negative
Cheese product company (140) 25.75 Negative Positive
26.31 Positive Positive
27.25 Positive Positive
28.84 Positive Positive
31.26 Positive Positive
34.08 Negative Positive
34.40 Negative Positive
Negative Positive Negative
Whey powder company (154) All negative 7 positives 7 of 7 negative
6 food product companies (686) 36.64 All negative 1 of 1 negative
3 dairy product companies (309) All negative All negative Not performed
2 food and cheese companies (320) All negative All negative Not performed

Total (1,741) 11 positive 12 positive 9 positive

aA total of 1,741 environmental samples from 14 food-processing facilities were screened by PCR and VIDAS
methods. Culture analyses were performed for a total of 19 samples that were positive according to the
PCR or VIDAS method. Salmonella isolates were obtained from 9 of 11 PCR-positive samples and 4 of 12
VIDAS-positive samples. VIDAS failed to detect Salmonella in 5 culture-positive samples.
bCulture analyses were performed for all samples that were positive with PCR or VIDAS screening.

by the culture method in only 4 samples. In contrast, Salmonella was isolated from 9 of
11 PCR-positive samples. Notably, the VIDAS method failed to detect the presence of
Salmonella in 5 samples that were positive by the culture method. Most public health
and regulatory organizations perform culture analyses on all presumptive positives
found by the screening method, and no culture analysis is generally performed on
samples that were reported as negative by the screening method. In summary, the PCR
method is sensitive and specific, detected 55% more culture-positive samples than the
VIDAS method, and reported no false negatives. There was an excellent correlation
between the culture method and the PCR method.
The ROC analysis of the agreement of the PCR method with the culture method
produced an AUC of 0.90, with a 95% confidence interval between 0.76 and 1.0 (Fig.
1A). The Cohen’s kappa of this agreement (29) was 0.79 ⫾ 0.13 (standard error of the
mean [SEM]), and the empirical probability, as described in Materials and Methods, of

FIG 1 (A) ROC analysis of the agreement between the PCR method and culture method (gold standard).
The AUC of 0.90 signifies excellent agreement between the two methods. (B) ROC analysis of the
agreement between the VIDAS method and culture method (gold standard). The AUC of 0.67 signifies
modest agreement between the two methods, whereas the confidence interval (CI) indicates high
signal-to-noise ratio.

July 2017 Volume 83 Issue 14 e00644-17 aem.asm.org 5

Kasturi and Drgon Applied and Environmental Microbiology

this level of agreement being reached by chance was 0.0013. These results establish
statistically robust (significant) agreement between the new PCR method and the
culture method, which has been used as the gold standard for this comparison (Fig. 1A).
Moreover, the same ROC analysis of the agreement between the VIDAS results and the
culture method results yielded an AUC of 0.67, with a 95% confidence interval between
0.46 and 0.89 (Fig. 1B). The Cohen’s kappa for the VIDAS versus culture agreement
could not be clearly established, owing to a very high likelihood that the agreement
was produced by chance. The empirical P value for this agreement being reached by
pure chance was 0.9, as described in Materials and Methods. Taken together, these
observations suggest that the new PCR method significantly outperforms the reference
VIDAS screening method for detecting Salmonella in environmental samples.

DISCUSSION
The PCR results for 329 Salmonella isolates and the 22 non-Salmonella environmen-
tal organisms confirmed the specificities of the invA, prt, tyv, and Sdf-1 primer-probe
sets used in the assay. The invA reagent detected all 6 subspecies of Salmonella enterica.
The majority of the Salmonella spp. isolated from the environmental samples belonged
to the subspecies enterica, and only a few isolates were from the other 5 subspecies of
S. enterica. The prt and tyv target-specific reagents detected all the 57 group D-positive
isolates (12 distinct serovars) that shared the antigenic factor O-9, whereas the other
272 isolates of 114 serovars from the 32 O groups were negative, thus confirming the
absence of the shared antigenic factor O-9 in those isolates. The Sdf-1-specific set
recognized all S. Enteritidis isolates, as expected. Surprisingly, this reagent also detected
a non-Enteritidis Salmonella isolate from a thyme sample. But, this isolate did not
contain the O-9-specific sequence that was common to all group D serovars and has
been identified as S. enterica serovar Montevideo of group C1. This result demonstrated
that the Sdf-1-specific reagent, if used as a stand-alone PCR reagent, possibly reacts
with non-Enteritidis Salmonella serovars that contained the Sdf-1 target sequence and
thus might not be a reliable reagent for the identification of S. Enteritidis.
Analysis of genomic sequences from S. Montevideo isolates indicated that S. Mon-
tevideo isolates exhibited sequence identity of 99% or higher, covering 98 to 100% of
the genomic sequence, whereas 98 to 99% sequence identity was shared with other
serovars of the C1 group that covered a region of 93 to 95%. The segment of the
genome that showed greater similarity decreased to 90 or 91% with isolates belonging
to group B. The data from the Sequence Read Archive (SRA) of the Sdf-1-positive S.
Montevideo isolate showed the presence of a stretch of 382 nucleotides (accession no.
SRR1220746) that covered the entire Sdf-1 region of 333 nucleotides of reference S.
Enteritidis with 98% identity (Fig. 2). Thus, it is not surprising that the Sdf-1 reagent
reacted positively with this isolate, even though the group D-specific PCR reagent did
not react. In addition to sequence similarity with Sdf-1, the isolate also exhibited
sequence similarity to Sdf-3, Sdf-4, Sdf-5, Sdf-6, and Sdf-8 sequences (data not shown).
The sequences (from the SRA) of several other S. Montevideo isolates examined did not
have any significant similarity with Sdf-1 or other Salmonella differential fragment
sequences.
The mechanism involved in the acquisition of the Sdf-1 sequence by non-Enteritidis
Salmonella serovars is not known. We are currently investigating the possibility that
genetic elements from S. Enteritidis were horizontally transferred to non-Enteritidis
Salmonella serovars via phage infection. Plasmid-mediated lateral transmission of
genetic elements, like the O-antigen biosynthesis genes, has been reported (30, 31).
Thus, an approach using the O group-specific reagent together with a serovar-specific
Sdf-1 PCR reagent would be more reliable for identifying the S. Enteritidis serovar than
existing methods that use only one of the targets common to most S. Enteritidis strains.
All S. Enteritidis isolates tested in this study were positive with the prt-, tyv-, and
Sdf-1-specific reagents, as expected. All other serovars of group D were positive with
both of the prt and tyv PCR reagents but not with the Sdf-1 reagent. Thus, the new
method is more reliable than current methods: the protE6-specific PCR reagent recog-

July 2017 Volume 83 Issue 14 e00644-17 aem.asm.org 6

Evaluation of Real-Time PCR for Salmonella Species Applied and Environmental Microbiology

FIG 2 Similarity between the nucleotide sequence of an S. Montevideo isolate and the Sdf-1 sequence. The Sdf-1
sequences (333 nucleotides) for S. Enteritidis strain OLF-00D989 87-1 (accession no. CP011942.1) and an S.
Montevideo isolate (accession no. SRR1220746) were aligned to show the similarity. The sequence of Sdf-1 of S.
Enteritidis is underlined. The Sdf-1 PCR primer and probe sequences used in the PCR are highlighted: the forward
primer sequence is in green, the probe sequence is in yellow, and the reverse primer sequence is in gray. Sbjct,
subject.

nized several phage types of S. Enteritidis but not all (9, 32), and the Sdf-1-specific
reagent (33, 34) would have identified the S. Montevideo isolate as S. Enteritidis.
The environmental swab samples were collected from different places and various
production environments that included surfaces associated with food preparation,
kitchen floor, enrobing rooms, and packing and storage areas. Although the VIDAS
method has been validated using a large panel of food matrices after artificially spiking
them with Salmonella, a similar validation has not been done for environmental
samples that contain varied matrices and a variety of competing organisms. The PCR
results of environmental samples demonstrated an excellent correlation with the
culture method. In addition to our observations, a recent study has shown the VIDAS
method could not detect the presence of Salmonella in certain produce samples that
were spiked with Salmonella (false-negative results), even though the Salmonella used
for spiking was isolated from those samples using the culture method (18). The PCR
method could be the alternative and reliable method for screening such samples. The
PCR method is highly sensitive, detecting the presence of as few as 40 Salmonella
cells/ml (preenrichment) with a probability of 50%, which increased to 100% for 80 or
more cells. On the other hand, the VIDAS method did not detect the presence of
Salmonella in 5 samples that were PCR positive, and Salmonella isolates were obtained
from the enrichments that were used for the VIDAS assay. The above-described results

July 2017 Volume 83 Issue 14 e00644-17 aem.asm.org 7

Kasturi and Drgon Applied and Environmental Microbiology

clearly demonstrated that secondary enrichments from the above-mentioned 5 VIDAS-

negative samples contained Salmonella cells, although fewer cells than were required
for detection by VIDAS. Alternatively, the Salmonella cells present in those enrichments
did not have a sufficient number of antigenic epitopes required for detection by the
VIDAS method; this possibility was unlikely because the same Salmonella serovar was
isolated from both VIDAS-positive and -negative samples. Based on the CT values
obtained from the PCR assays, it appears that there was no direct relationship between
the number of Salmonella cells present in the preenrichments and the negative VIDAS
results (18). These results suggest that the diversity of matrices and the composition
and levels of competing organisms present in different environmental samples could
have influenced the outcome of the VIDAS results. The fact that the PCR method
detected 55% more Salmonella positives than VIDAS confirmed the higher sensitivity of
the PCR method. In addition, PCR did not yield any false-negative results, unlike the
VIDAS method. The results from the PCR screening of 1,741 environmental samples
showed an excellent correlation with the culture method, thus confirming the reliability
and validity of the PCR method for screening environmental samples for detecting
Salmonella. In conclusion, our PCR-based approach detected 55% more positives in half
the time required for the VIDAS method, and thus the PCR method has great potential
to strengthen public health initiatives through expeditious detection of Salmonella spp.
and S. Enteritidis in environmental samples.

MATERIALS AND METHODS

Control organisms. A Salmonella control panel and a non-Salmonella (negative) control panel
containing common environmental organisms were used for confirming the inclusivity and exclusivity of
the PCR. The Salmonella panel included 12 reference organisms, i.e., Salmonella bongori (ATCC 43975); 6
serovars of S. enterica subsp. enterica: Paratyphi A, group A (ATCC 9281), Typhimurium, group B (ATCC
13311), Javiana, group D (ATCC 10721), Enteritidis, group D (ATCC 13076), Gaminara, group I (ATCC 8324),
and Cerro, group K (ATCC 10723), and 5 reference organisms from other subspecies of S. enterica, i.e., S.
enterica subsp. salamae (ATCC 15786), S. enterica subsp. arizonae IIIa 35:-:x,z15 (CFSAN014667), S. enterica
subsp. diarizonae, IIIb 47:i:z53:z57 (ATCC 12325), S. enterica subsp. houtenae, IV 43:z4,z23:- (CFSAN011677),
and S. enterica subsp. indica, VI 1,6,14,25:a:e,n,z15 (CFSAN015097). The DNA extracts from 12 control
organisms of a Salmonella panel were prepared from 1.0 ml of overnight culture using InstaGene matrix,
according to the manufacturer’s instructions (catalog no. 732-6030; Bio-Rad); briefly, 1.0 ml of each
bacterial suspension was pelleted by centrifugation at 14,000 ⫻ g for 2 min in an Eppendorf microcen-
trifuge, and each pellet was mixed thoroughly with 200 ␮l of InstaGene matrix by vigorous agitation at
high speed on a vortex mixer for 10 s and incubated for 15 to 30 min in an Eppendorf thermomixer at
56°C. The samples were removed from the thermomixer, agitated at high speed on a vortex mixer for a
few seconds, and placed in a heat block at 100°C for 10 min. The samples were removed from the heat
block, agitated on a vortex mixer for 10 s, and centrifuged at 14,000 ⫻ g for 2 min. The supernatants
containing the extracted DNA were diluted 1:1,000 and used in the PCR for confirming the inclusivity of
Salmonella-specific PCR primers and probes. The non-Salmonella control panel contained 22 non-
Salmonella organisms, including 18 ATCC reference organisms and 4 laboratory isolates (Table 2). Purified
DNA from the non-Salmonella control organisms and a positive-control Salmonella (serovar Enteritidis)
strain (ATCC 13076) was prepared from 1.0-ml overnight cultures of bacteria using the DNeasy blood and
tissue kit (catalog no. 69504; Qiagen, Inc., USA) and an automated QIAcube instrument. The concentra-
tion of purified DNA was determined with the Qubit double-stranded DNA (dsDNA) BR assay kit (catalog
no. Q32850), according to the protocol provided by the manufacturer. Copies of genomic DNA present
in the positive-control S. Enteritidis organism and all non-Salmonella environmental control organisms
were determined using the online URI Genomics & Sequencing Center copy number calculator for
double-stranded DNA. A total of 105 copies of the purified DNA from each non-Salmonella organism were
used for confirming the exclusivity of PCR primers and probes.
Salmonella isolates used for PCR method validation. For validation of the PCR method, a panel of
329 Salmonella isolates, confirmed by pulsed-field gel electrophoresis at the FDA Northeast Regional
Laboratory (NRL) and with conventional serology performed at the FDA Arkansas Regional Laboratory
(ARL), were randomly chosen from the NRL Salmonella culture bank and tested for the inclusivity of the
PCR assay. The panel consisted of 126 distinct serovars belonging to 34 O groups that were commonly
found in food or the environment. The CDC-recommended taxonomical nomenclature was used to
identify species, subspecies, and serovars in this report (35).
Isolates from frozen stocks were first plated on selective agar medium, and a single typical colony was
picked and grown in 4.0 ml of Luria-Bertani medium overnight at 37°C in a shaker incubator. DNA was
extracted from 1.0 ml of culture, according to the procedure described above for the preparation of DNA
from the control Salmonella panel, with 200 ␮l of InstaGene matrix. DNA extracts were stored at ⫺20°C
until use. DNA samples from isolates were diluted 1:1,000 prior to use in PCR.
Processing of environmental samples. A total of 1,741 environmental sponge swab samples
collected by the FDA during the surveillance inspections of 14 different food manufacturing companies

July 2017 Volume 83 Issue 14 e00644-17 aem.asm.org 8

Evaluation of Real-Time PCR for Salmonella Species Applied and Environmental Microbiology

(from 2015 to 2017) were used for this study. The swab samples were analyzed within 24 h of receipt.
To each of the swab/sponge samples, lactose broth or modified buffered peptone water was added in
a Whirl-Pak bag, and samples were incubated for 24 ⫾ 2.0 h at 35°C, according to the procedure
described in the Bacteriological Analytical Manual (27) and AOAC official methods 2009.03 (17). One-tenth
of a milliliter of primary enrichment from each sample was transferred into 10 ml of SX2 broth and
incubated at 42 ⫾ 1°C for an additional 24 ⫾ 2.0 h; aliquots from the secondary enrichments were used
for the VIDAS. One milliliter of preenrichment from each sample was used for the extraction of DNA for
the PCR screening.
VIDAS screening. One-half of a milliliter of secondary enrichment from each sample was transferred
to VIDAS strips, and VIDAS was performed according to the Easy SLM method (17).
PCR screening. DNA was extracted from 1.0 ml of preenrichment samples, with 200 ␮l of InstaGene
matrix, according to the manufacturer’s instructions described for the extraction of DNA from bacterial
cultures (catalog no. 732-6030; Bio-Rad). Five microliters of the extracted DNA, undiluted (representing
25.0 ␮l of the preenrichment sample), was used in the real-time PCR.
Primers and TaqMan probes. PCR primer pairs and TaqMan probes specific for the gene invA of
Salmonella spp. (36), the paratose synthase (prt) gene, and the tyvelose epimerase (tyv) gene of group
D and group A Salmonella, the Salmonella-differentiating fragment 1 (Sdf-1) sequence of S. Enteritidis,
and the IAC DNA were designed using the online primer design software of Integrated DNA Technol-
ogies, Inc., USA (http://www.idtdna.com/calc/analyzer). IDT synthesized a gene construct of 139 bp of IAC
DNA as G block DNA: ATGATTACGAATTCACTGATCTGACTTCACTCCTGTCTACTATACAGATATAACGCTGAC
TCGTCCATACAATAAGACACCTTGATGTAGTCTGATAGTGGTCAGTGACGAGTGTCATCTAGAGTCGAATTCACT
GCAA. The IAC served as the reagent activity control to monitor the presence of PCR inhibitors in
samples. The target sequence of invA was chosen from a conserved region shared by Salmonella bongori
and the subspecies of S. enterica (36). A cell surface polysaccharide present in O groups A, B, and D is a
polymer made of a repeating unit of four sugars, of which three form the backbone mannosyl-
rhamnosyl-galactose common to all three O groups, whereas the fourth sugar is a dideoxyhexose linked
to a mannose residue, which is specific to each O group (37). The gene rfbS is responsible for the
synthesis of CDP-paratose, the fourth sugar of the group A epitope (O-2), whereas the rfbE is responsible
for converting CDP-paratose to CDP-tyvelose of the group D epitope (O-9) (37). All group A members
contain rfbE and were derived from group D by a single frameshift mutation in the rfbE sequence (37).
Thus, both group A and group D serovars contain rfbS and rfbE sequences, whereas rfbE is nonfunctional
in group A. Thus, either the prt or tyv PCR primer-probe set could be used for the identification of groups
A and D. No group A serovar was found among the 1,700 Salmonella isolates recovered from food and
environmental samples in our laboratory during the past 18 years or among the 10,000 isolates that were
sequenced by the FDA from 2012 (http://www.ncbi.nlm.nih.gov/projects/pathogens). Essentially all
PCR-positive environmental and food samples are likely to be of the group D serovar. For each gene
target, four sets of primer-probe sequences were evaluated for secondary structures and dimer-forming
potential with other primers and probes using the Oligo Analyzer application software program of IDT.
Thus, a set of 12 sequences for each target was first tested against 36 primer sequences specific to other
three targets in a chessboard-type analysis. Then, four primer-probe sequence sets exhibiting the least
heterodimer-forming potential were selected for use in multiplex PCR and were synthesized by IDT
(Table 5). The stock solutions (100 ␮M) of primers and probes were prepared according to IDT’s
instructions. The optimal concentrations of primers and probes, annealing and elongation temperatures,
and the specificity of the reactions were initially tested for each individual primer-probe set and then
confirmed in the multiplex reaction. The real-time PCR assay described herein has been optimized for
detecting three Salmonella target genes, invA, prt, and tyv, as well as Sdf-1 in a single reaction.
Multiplex real-time PCR. The optimal PCR conditions were configured for the SmartCycler II
platform in 25-␮l reaction mixtures. The method is easily adaptable to the Fast 7500 (Applied Biosys-
tems). USB VeriQuest Fast Probe quantitative PCR (qPCR) master mix with no reference dye (product no.
75685; Affymetrix, Inc., USA) was used for the assays. Master mix reagents from GE, Eurogentec, and
Qiagen were also tested and gave comparable results. The optimal conditions for performing the
multiplex assay were first determined using DNA templates from a positive-control S. Enteritidis and 22
non-Salmonella target organisms. The concentrations of primers and probes used in PCR were as follows:
200 nM invA forward and reverse primers and 30 nM probe; 250 nM prt forward and reverse primers and
200 nM probe (or 300 nM tyv forward primer, 200 nM tyv reverse primer, and 200 nM probe); 250 nM
Sdf-1 forward primer, 300 nM reverse primer, and 40 nM probe; 200 nM IAC forward and reverse primers
and 100 nM probe. The PCR master mix was prepared from the reaction components that were thawed
and kept on ice. A known quantity of IAC DNA (100 to 200 copies) was added to provide a CT value of
32 to 34. The PCR master mix was supplemented with 0.10 U of VeriQuest Taq DNA polymerase. Master
mix (20 ␮l) was distributed to the required number of SmartCycler reaction tubes for each run, and the
caps were loosely closed. To the no-template control (NTC) tubes, 5.0 ␮l of molecular biology-grade
water was added, and the caps were closed tightly. To the sample-containing tubes, 5.0 ␮l of DNA extract
from samples was added and the caps closed tightly. Finally, 5.0 ␮l of positive control containing 103 or
104 DNA copies, giving a CT value of 29 to 26, was added to positive-control tubes. The tubes were briefly
centrifuged to bring the reaction mixtures to the bottom of the tubes. The tubes were then placed in
thermal cyclers and subjected to a two-step PCR protocol. The protocol consisted of a 180-s hold at 95°C,
followed by 40 cycles of 10 s at 94°C and 30 s at 60°C. The fluorescence generated in each reaction was
recorded at the extension step of each cycle. The PCR run took ⬃46 min.
Each lot of master mix we prepared was subjected to quality control to ensure that each reaction
mixture contained 100 to 200 copies of IAC DNA and that a single copy of Salmonella genomic DNA

July 2017 Volume 83 Issue 14 e00644-17 aem.asm.org 9

Kasturi and Drgon Applied and Environmental Microbiology

TABLE 5 List of primer and probe sequences used in the multiplex PCRa
No. of Position in
Name bases genome Sequence (5= to 3=)b
Primers
invA forward 20 2924483–2924506 AGCGTACTGGAAAGGGAAAG
invA reverse 24 2924598–2924579 ATACCGCCAATAAAGTTCACAAAG
prt forward 22 2175083–2175104 AGCTCCATAGAAATGCTCCAAT
prt reverse 22 2175213–2175192 GAACATCACTGCCACCAAATAC
tyv forward 25 2174357–2174376 ACTAAGTATATGCCTGATAGCTGTT
tyv reverse 20 2174454–2174462 GCCGTACTGCCTCAAGTAAA
Sdf-1 forward 26 1472650–1472675 CTTTCTCAGATTCAGGGAGTATATCA
Sdf-1 reverse 23 1472772–1472750 TGAACTACGTTCGTTCTTCTGGT
IAC forward 27 16–42 CTGATCTGACTTCACTCCTGTCTACTA
IAC reverse 24 116–93 GACACTCGTCACTGACCACTATCA

Probes
invA (⫹) 26 2924529–2924554 6FAM/CGTCACCTTTGATAAACTTCATCGCA–BHQ1
prt (⫺) 26 2175150–2175125 Cy3/CCGCCGCCATTATAGATAAAGTTTGT–BHQ2
tyv (⫺) 25 2174454–2174430 Cy3/TGCAGGTCAAGTGGCAATGACTACA–BHQ2
Sdf-1 (⫹) 24 1472690–1472713 TexRdXN/ATCAGCCTGTTGTCTGCTCACCAT–BHQ2
IAC (⫹) 28 54–81 Cy5/CGCTGACTCGTCCATACAATAAGACACC–BHQ2
aThe sequences, lengths, and locations of primers and probes are shown within the Salmonella enterica
serovar Enteritidis genome (strain SA20082034, accession no. CP007425.2). The sequence, length and
location of the IAC primers and probe are shown on the IAC gene. The lengths of the PCR products
generated by the primer pairs were as follows: invA, 115 nucleotides; IAC, 100 nucleotides; prt, 130
nucleotides; tyv, 130 nucleotides; and Sdf-1, 123 nucleotides.
b6FAM, 6-carboxyfluorescein; BHQ1, black hole quencher 1; TexRd, Texas Red.

could be detected. Duplicate reactions of positive and negative controls were included with each
run. An increase in the CT value of the IAC target in a sample compared with the CT of NTC indicated
the presence of PCR inhibitors in that sample. The IAC could yield a negative result if a sample
contained ⱖ104 copies of the positive target as a consequence of one or more reagents of the PCR
mixture becoming depleted sooner than the CT value of IAC reached the threshold set for the assay.
A sample that yielded a positive CT value for at least one of the targets was considered a valid test.
Samples showing inhibition of IAC in the absence of a positive target in the assay were required to
be reprocessed and retested.
Culture analysis. The secondary enrichments from all VIDAS- and PCR-positive samples were plated
onto xylose lysine deoxycholate (XLD), Hektoen enteric (HE), and bismuth sulfite (BS) agar plates and
incubated for 24 h at 35 ⫾ 2°C (27). Typical and atypical control Salmonella cultures (ATCC 8324 and
ATCC 29934) were included as positive controls. Colonies showing typical/atypical morphology were
inoculated on triple sugar iron/lysine iron agar (TSI/LIA) slants, and the cultures showing characteristic
reactions were identified biochemically using the Vitek 2 GN card and AOAC Official Methods of Analysis
method 2011.17 and confirmed with serological tests.
Statistical analysis. All statistical analyses were performed in R (38), except for the Cohen’s kappa
calculation, which was performed using an online calculator (http://vassarstats.net/kappa.html). The
correlation between the number of DNA copies (and thus the number of Salmonella cells present in each
sample) and the PCR signal (InvA CT) was established with the Pearson correlation coefficient (38).
The agreement of the results of the PCR and VIDAS methods with the microbial culture method (gold
standard) was assessed as follows. (i) Agreement between the PCR method and culture method and
between the VIDAS method and the culture method was assessed with receiver operating characteristic
(ROC) analysis. The level of agreement was enumerated as the area under the curve (AUC) (39). (ii) The
level of agreement was calculated using unweighted Cohen’s kappa (29). (iii) The probability of chance
agreement was enumerated with an empirical statistics model as follows. The number of agreements
(positive-positive or negative-negative) was calculated for each comparison (PCR versus culture, and
VIDAS versus culture). The number of agreements in each comparison represented the agreement score
for that comparison. The sets of results from the test method and the gold standard method were
randomized and randomly matched. The agreement score was calculated for this random match. This
procedure was repeated 10,000 times and produced a distribution of 10,000 chance agreement scores
for each comparison. The distribution of the agreement scores was compared to the real agreement
score for each comparison, and the probability of chance agreement was enumerated as the fraction of
trials that reached the agreement score that was equal or higher than the real agreement score from the
experiment.

ACKNOWLEDGMENTS
We thank Angela Lara, Marisa Reeves, Kazi Rafiquzzaman, Evelyn Ariza, Julia Mullins,
Marianna Rhatigan, Tony Sepla, Darren Gopaul, Khadija Auguste, Susan Joseph, and
Victoria Rosales for providing the preenrichments of environmental samples, Sandra

July 2017 Volume 83 Issue 14 e00644-17 aem.asm.org 10

Evaluation of Real-Time PCR for Salmonella Species Applied and Environmental Microbiology

Thompson and Marianna Rhatigan for providing Salmonella isolates from the Northeast
Regional Laboratory culture bank, and Allan Littell and Lawrence James for purchasing
supplies used in the study. We also wish to thank Azad Kaushik, Department of Cellular
and Molecular Biology, University of Guelph, Canada, for critical reading of the manu-
script. K.N.K. wishes to thank Yelena Karaseva and John Leazer for the encouragement
of this study.
The opinions expressed in this paper are those of the authors and not necessarily
those of the FDA.

REFERENCES
1. Braden CR. 2006. Salmonella enterica serotype Enteritidis and eggs: a 18. Welch LJ, Melvin CD, Smiley RD. 2016. Preliminary examination of spiked-
national epidemic in the United States. Clin Infect Dis 43:512–517. microgreens data for Vidas Salmonella AOAC Official Method 2004.03
https://doi.org/10.1086/505973. screening method compiled from data entered in the field accomplishment
2. Chen J, Zhang Paoli LGC, Shi C, Tu S, Shi X. 2010. A real-time PCR method computerized tracking system (FACTS). Lab Inf Bull 4611:1–9.
for the detection of Salmonella enterica from food using a target se- 19. Limpitakis NC, Genigeorgis C. 2001. Elimination of cross reactions during
quence identified by comparative genomic analysis. Int J Food Microbiol detection of Salmonella spp., with commercial ELISA kits, p 502–504. Pro-
137:168 –174. https://doi.org/10.1016/j.ijfoodmicro.2009.12.004. ceedings of the 4th International Symposium on the Epidemiology and
3. Cheng CM, Lin W, Van KT, Phan L, Tran NN, Farmer D. 2008. Rapid Control of Salmonella and other Food Borne Pathogens in Pork, 2 to 5
detection of Salmonella in foods using real-time PCR. J Food Prot September 2001, Leipzig, Germany. http://lib.dr.iastate.edu/safepork/2001/
71:2436 –2441. https://doi.org/10.4315/0362-028X-71.12.2436. allpapers/128/.
4. González-Escalona N, Hammack TS, Russell M, Jacobson AP, De Jesús AJ, 20. Levy H, Diallo S, Tennant SM, Livio S, Sow SO, Tapia M, Fields PI, Mikoleit
Brown EW, Lampel KA. 2009. Detection of live Salmonella sp. cells in M, Tamboura B, Kotloff KL, Lagos R, Nataro JP, Galen JE, Levine MM.
produce by a TaqMan-based quantitative reverse transcriptase real-time 2008. PCR method to identify Salmonella enterica serovars Typhi, Para-
PCR targeting invA mRNA. Appl Environ Microbiol 75:3714 –3720. https:// typhi A, and Paratyphi B among Salmonella isolates from the blood of
doi.org/10.1128/AEM.02686-08. patients with clinical enteric fever. J Clin Microbiol 46:1861–1866.
5. Hein I, Flekna G, Krassnig M, Wagner M. 2006. Real-time PCR for the https://doi.org/10.1128/JCM.00109-08.
detection of Salmonella spp. in food: an alternative approach to a 21. Lim BK, Thong KL. 2009. Application of PCR-based serogrouping of
conventional PCR system suggested by the FOOD-PCR project. J Micro- selected Salmonella serotypes in Malaysia. J Infect Dev Ctries 3:420 – 428.
biol Methods 66:538 –547. https://doi.org/10.1016/j.mimet.2006.02.008. 22. Alvarez J, Sota Vivanco MAB, Perales I, Cisterna R, Rementeria A, Garaizar J.
6. Josefsen MH, Krause M, Hansen F, Hoorfar J. 2007. Optimization of a 2004. Development of a multiplex PCR technique for detection and epide-
12-hour TaqMan PCR-based method for detection of Salmonella bacteria miological typing of Salmonella in human clinical samples. J Clin Microbiol
in meat. Appl Environ Microbiol 73:3040 –3048. https://doi.org/10.1128/ 42:1734–1738. https://doi.org/10.1128/JCM.42.4.1734-1738.2004.
AEM.02823-06. 23. Hadjinicolaou AV, Demetriou VL, Emmanuel MA, Kakoyiannis CK, Kostri-
7. Malorny B, Hoorfar J, Bunge C, Helmuth R. 2003. Multicenter validation kis LG. 2009. Molecular beacon-based real-time PCR detection of primary
of the analytical accuracy of Salmonella PCR: towards an international isolates of Salmonella Typhimurium and Salmonella Enteritidis in envi-
standard. Appl Environ Microbiol 69:290 –296. https://doi.org/10.1128/ ronmental and clinical samples. BMC Microbiol 9:1–14. https://doi.org/
AEM.69.1.290-296.2003. 10.1186/1471-2180-9-1.
8. Malorny B, Paccassoni E, Fach P, Bunge C, Martin A, Helmuth R. 2004. 24. Muñoz N, Diaz-Osorio M, Moreno J, Sanchez-Jimenez M, Cadona-Castro N.
Diagnostic real-time PCR for detection of Salmonella in food. Appl 2010. Development and evaluation of a multiplex real-time polymerase
Environ Microbiol 70:7046 –7052. https://doi.org/10.1128/AEM.70.12 chain reaction procedure to clinically type prevalent Salmonella enterica
.7046-7052.2004. serovars. J Mol Diagn 12:220 –225. https://doi.org/10.2353/jmoldx.2010
9. Malorny B, Bunge C, Helmuth R. 2007. A real-time PCR for the detection of .090036.
Salmonella Enteritidis in poultry meat and consumption eggs. J Microbiol 25. Fitzgerald C, Collins M, Duyne SV, Mikoleit M, Brown T, Fields P. 2007.
Methods 70:245–251. https://doi.org/10.1016/j.mimet.2007.04.013. Multiplex, bead-based suspension array for molecular determination of
10. Löfström C, Hansen F, Hoorfar J. 2010. Validation of a 20-h real-time PCR common Salmonella serogroups. J Clin Microbiol 45:3323–3334. https://
method for screening of Salmonella in poultry fecal samples. Vet Micro- doi.org/10.1128/JCM.00025-07.
biol 144:511–514. https://doi.org/10.1016/j.vetmic.2010.02.019. 26. McMahon WA, Shultz AM, Johnson RL. 2004. Evaluation of Vidas Salmo-
11. Anderson A, Pietsch K, Zucker R, Mayr A, Muller-Hoche E, Messelhausser nella (SLM) immunoassay method with Rappaport-Vassiliadis (RV) me-
U, Sing A, Busch U, Huber I. 2011. Validation of a duplex real-time PCR dium for detection of Salmonella in foods: collaborative study. J AOAC
for the detection of Salmonella spp. in different food products. Food Int 87:867– 883.
Anal Methods 4:259 –267. https://doi.org/10.1007/s12161-010-9142-8. 27. Andrews WH, Wang H, Jacobson A, Hammack TS. 2016. Bacteriological
12. Cheng CM, Van KT, Lin W, Ruby RM. 2009. Interlaboratory validation of analytical manual, chapter 5: Salmonella. U.S. Food and Drug Administra-
a real-time PCR 24-hour rapid method for detection of Salmonella in tion, Silver Spring, MD. https://www.fda.gov/Food/FoodScienceResearch/
Foods. J Food Prot 72:2436 –2441. LaboratoryMethods/ucm070149.htm.
13. Mozola M. 2006. Genetic-based methods for detection of Salmonella 28. Grimont PAD, Weill F-X. 2007. Antigenic formulae of the Salmonella
spp. in foods. J AOAC Int 89:517–529. serovars, 9th ed. WHO Collaborating Centre for Reference and Research
14. Wolffs PFG, Glencross K, Thibaudeau RR, Griffiths MW. 2006. Direct on Salmonella, Institut Pasteur, Paris, France. https://www.pasteur.fr/
quantitation and detection of salmonellae in biological samples without sites/default/files/veng_0.pdf.
enrichment, using two-step filtration and real-time PCR. Appl Environ 29. Cohen J. 1960. A coefficient of agreement for nominal scales. Educ
Microbiol 72:3896 –3900. Psychol Meas 20:37– 46. https://doi.org/10.1177/001316446002000104.
15. Abu Al Soud W, Radstrom P. 2000. Effect of amplification facilitators on 30. Keenleyside WJ, Whitfield CJ. 1995. Lateral transfer of rfb genes: a mobiliz-
diagnostic PCR in the presence of blood, feces, and meat. J Clin Micro- able ColE1-type plasmid carries the rfb O:54 (O:54 antigen biosynthesis)
biol 38:4463– 4470. gene cluster from Salmonella enterica serovar Borreze. J Bacteriol 177:
16. Schrader C, Schelke A, Ellerbroek L, Johne R. 2012. PCR inhibitors–occurrence, 5247–4233. https://doi.org/10.1128/jb.177.18.5247-5253.1995.
properties and removal. J Appl Microbiol 113:1014–1026. https://doi.org/10 31. Popoff MY, Le Minor L. 1985. Expression of antigenic factor O54 is
.1111/j.1365-2672.2012.05384.x. associated with the presence of a plasmid in Salmonella. Ann Inst
17. Johnson R, Mills J, Colon-Reveles J. 2009. Vidas Salmonella (SLM) assay Pasteur Microbiol (1985) 136B:169 –179.
method Easy SLM with ChromID Salmonella (SM2) agar. J AOAC Int 32. González-Escalona N, Zhang G, Brown EW. 2012. Salmonella–a diversi-
92:1861–1864. fied superbug, p 541–562. In Kumar Y (ed), Multiplex TaqMan real-time

July 2017 Volume 83 Issue 14 e00644-17 aem.asm.org 11

Kasturi and Drgon Applied and Environmental Microbiology

PCR (qPCR) assay targeting prot6E and invA genes for fast and accurate 38. R Core Team. 2016. A language and environment for statistical comput-
detection of Salmonella Enteritidis. InTech, Rijeka, Croatia. ing. R Foundation for Statistical Computing, Vienna, Austria. https://
33. Agron PG, Walker RL, Kinde H, Sawyer SJ, Hayes DC, Wollard J, Andersen www.R-project.org.
GL. 2001. Identification by subtractive hybridization of sequences spe- 39. Robin X, Turck N, Hainard A, Tiberti N, Lisacek F, Sanchez JC, Müller M.
cific for Salmonella enterica serovar Enteritidis. Appl Environ Microbiol 2011. pROC: an open-source package for R and S⫹ to analyze and
67:4984 – 4991. https://doi.org/10.1128/AEM.67.11.4984-4991.2001. compare ROC curves. BMC Bioinformatics 12:77. https://doi.org/10.1186/
34. Agron PG, Andersen GL, Walker RL. Oct 2008. Method for the detection 1471-2105-12-77.
of Salmonella enterica serovar Enteritidis. US patent 7442517 B2. 41. Lavalett L, Sanchez Munoz MMN, Moreno J, Cardona-Castro N. 2009.
35. Brenner FW, Villar RG, Angulo FJ, Tauxe R, Swaminathan B. 2000. Sal- Development and validation of a multiplex polymerase chain reaction
monella nomenclature. J Clin Microbiol 38:2465–2467. for molecular identification of Salmonella enterica serogroups B, C2, D
36. Boyd EF, Li J, Ochman H, Selander RK. 1997. Comparative genetics of the and E. Biomedica 29:244 –252. (In Spanish.) https://doi.org/10.7705/
inv-spa invasion gene complex of Salmonella enterica. J Bacteriol 179: biomedica.v29i2.26.
1985–1991. https://doi.org/10.1128/jb.179.6.1985-1991.1997. 42. Löfström C, Krause M, Josefsen MH, Hansen F, Hoorfar J. 2009. Validation
37. Verma N, Reeves P. 1989. Identification and sequence of rfbS and rfbE, which of a same-day real-time PCR method for screening of meat and carcass
determine antigenic specificity of group A and group D salmonellae. J swabs for Salmonella. BMC Microbiol 9:85. https://doi.org/10.1186/1471
Bacteriol 171:5694–5701. https://doi.org/10.1128/jb.171.10.5694-5701.1989. -2180-9-85.

July 2017 Volume 83 Issue 14 e00644-17 aem.asm.org 12