Food Control 30 (2013) 76e85

Contents lists available at SciVerse ScienceDirect

Food Control
journal homepage: www.elsevier.com/locate/foodcont

A new multiplex real-time PCR developed method for Salmonella spp. and Listeria
monocytogenes detection in food and environmental samples
Alejandro Garrido, María-José Chapela, Belén Román, Paula Fajardo, Jorge Lago, Juan M. Vieites,
Ana G. Cabado*
Microbiology and Toxins Area, ANFACO-CECOPESCA, Campus Univ. 16, 36310 Vigo PO, Spain

a r t i c l e i n f o a b s t r a c t

Article history: Despite efforts done by industries some well known foodborne pathogens, like Salmonella spp. and
Received 10 January 2012 Listeria monocytogenes, continue to be a challenge to public health institutions and a threat for
Received in revised form consumers. The aim of this study was to develop a complete, rapid and reliable multiplex real-time PCR
7 June 2012
(qPCR) method for the simultaneous detection of these two bacteria in food and environmental samples,
Accepted 20 June 2012
including a novel single enrichment broth (TA10) for both bacteria. TA10 broth was modified (pH and
buffer concentration) to enhance simultaneous growth of both pathogens in the presence of high
Keywords:
numbers of competitors bacteria. Also two different DNA-extraction protocols were compared. qPCR
TA10
Salmonella spp.
efficiency above 90% was obtained, covering 5 orders of magnitude. Complete method achieved low limit
Listeria monocytogenes of detection (5 cfu/25 g), and all quality parameters of the method returned values over 90%. Complete
Multiplex real-time PCR qPCR method was applied to 95 natural samples covering a wide variety of food types proving that the
Food safety qPCR method described, including the use of one single enrichment broth, modified TA10, was suitable
for the simultaneous and reliable screening of Salmonella spp. and L. monocytogenes in food and envi-
ronmental samples.
Ó 2012 Elsevier Ltd. All rights reserved.

1. Introduction present in very low numbers against a background of indigenous

microflora, rendering the recovery of target organisms difficult
Foodborne diseases are a widespread and growing public health (Kawasaki, Fratamico, Kamisaki-Horikoshi, et al., 2010).
concern affecting both developed and developing countries, In the early 1990s the polymerase chain reaction (PCR) was used
microbiologically contaminated food and water, being the major for simple identification of pure bacterial cultures or colonies on
causes of diarrheal diseases. Although their global incidence is agar plates. Since then, the development of sample preparations
difficult to estimate, authors generally agree that the percentage of suitable for PCR detection of bacteria in food or pre-enrichment
the population suffering from foodborne diseases each year could media has expanded enormously (Rijpens & Herman, 2002) and
be up to 30% in industrialized countries and figures could be even several PCR validation studies have reported that the PCR method is
worse in developing countries (Germini, Masola, Carnevali, & one of the most promising among the rapid microbiological
Marchelli, 2009). In order to minimize the risk of infection for methods for the detection and identification of bacteria in a wide
consumers, microbiological control of the food chain is being variety of samples (Myint, Johnson, Tablante, & Heckert, 2006) as
increasingly applied. Thus, the availability of reliable, rapid, and previously demonstrated (Chapela et al., 2010; Garrido et al., 2012).
internationally accepted test systems for determination of the PCR-based techniques such as nested-PCR, reverse transcription-
presence or absence of foodborne pathogens has become increas- PCR, PCR-based fingerprinting, quantitative PCR, and alternative
ingly important for the agricultural and food industry, as well as for amplification techniques such as nucleic acid sequence-based
legislative regulation of food safety (Malorny, Hoorfar, Bunge, & amplification (NASBA), among others, were introduced over the
Helmuth, 2003) as stated in regulation 2073/05 and the amend- years (Rijpens & Herman, 2002). Within them, real-time PCR
ment 1441/07 ((EC), 2005, 2007). Bacterial pathogens are often possess sensitivity equal to culture methods but it is rapid and
allow testing to be completed in 48 h (Navas et al., 2006). However
reliability of PCR-based detection methods partly depends on the
* Corresponding author. Tel.: þ34 986 469 303x344; fax: þ34 986 469 269. target bacterial cell number, i.e., the copy numbers of target
E-mail address: [email protected] (A.G. Cabado). molecules present in food samples. Detection methods for

0956-7135/$ e see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.foodcont.2012.06.029
 A. Garrido et al. / Food Control 30 (2013) 76e85 77

Salmonella spp. and Listeria monocytogenes pathogens in food are To achieve the required limit of detection of Salmonella spp.
based on enrichment using selective media to increase the (absence in 25 g of sample), and the no tolerance of
concentration of viable bacteria which can be followed by PCR for L. monocytogenes in ready-to-eat food in most countries, assays
identification. There are two major obstacles for simple and rapid have been coupled to pre-enrichment and DNA purification steps
detection of Salmonella spp. and L. monocytogenes pathogens in (Badosa et al., 2009). Nevertheless, not much attention was paid to
food: (a) the lack of one single medium for the simultaneous the development of novel broths for simultaneous recovery of both
enrichment of Salmonella spp. and L. monocytogenes; and (b) the bacteria as can be assessed by the scarce studies published during
presence of inhibitors of the PCR in foods such as beef or chicken, last years, underlining among others the following (Kawasaki et al.,
synthetic media and DNA preparation solutions (Bhaduri & Cottrell, 2005; Kim & Bhunia, 2008; Kobayashi et al., 2009). Universal pre-
2001; Lantz, Hahnhagerdal, & Radstrom, 1994; Lantz et al., 1998). enrichment broth (UP) was developed for the simultaneous
Though the sensitivities of many of the modern detection detection of Salmonella and Listeria in foods, although in that time
methods have improved significantly, an enrichment step is still many foods used to be analyzed for these two pathogens, but not
needed to overcome the problems of low pathogen numbers and to simultaneously. This medium allowed sublethally injured bacteria
limit the risk of detecting dead cells (Badosa, Chico, Pla, Pares, & to resuscitate and proliferate to high levels so that selective
Montesinos, 2009; Kim & Bhunia, 2008; Lantz et al., 1994; secondary enrichment media can be used (Hammack, Amaguana,
Mahmuda, Kawasaki, & Kawamoto, 2007) as their sensitivity is Johnson, & Andrews, 2003). Later studies proved that No. 17 broth
limited (Kimura et al., 1999). Concerning PCR, the reaction can showed a higher recovery than UP broth for injured Salmonella due
dramatically decrease if the sample contains PCR inhibitors (Lantz to the higher amount of nutritious compounds (Kamisaki-
et al., 1994). The enrichment step is required not only to increase Horikoshi et al., 2011).
the target-pathogen concentration in a sample but also to resus- The aim of the present work was the development of a complete
citate physiologically stressed or injured cells. Selective enrichment protocol for the detection of Salmonella spp. and L. monocytogenes,
is also necessary to suppress the natural background microorgan- including one single enrichment broth, a DNA-extraction method
isms as well as to improve detection efficiency and to avoid false- and a multiplex Real-Time PCR (qPCR) detection system able of
negative results. However, the drawbacks of some of the selective achieving a low limit of detection, even in the presence of high
enrichment broths are that selective agents can be inhibitory or can numbers of competitors. The medium selected was evaluated for its
delay growth of healthy microorganisms or the recovery of stressed ability to achieve high bacterial density of S. enterica and
target pathogens, thus affecting pathogens detection that have the L. monocytogenes when cultured either individually or simulta-
potential to recover and grow when the food is consumed neously with other common competitors. DNA-extraction method
(Kawasaki et al., 2009; Kim & Bhunia, 2008). In addition, DNA- chosen was compared with a common boiling protocol and,
extraction solutions condition the effectiveness of the PCR by furthermore, correct PCR reaction was controlled with an internal
reducing assay sensitivity due to loss of target DNA. The array of amplification control. Finally, this modified complete protocol was
methods developed for preparing PCR samples determines the applied to natural samples to evaluate presence of Salmonella spp.
need for techniques that are tailored to each type of food (Bhaduri & and L. monocytogenes.
Cottrell, 2001).
Several studies have reported the use of multiplex PCR systems 2. Materials and methods
to detect two or more pathogens in one assay (Mahmuda et al.,
2007). The multipathogen detection approach is attractive and 2.1. Bacterial strains and culture media
economically favorable since it can reduce the total space
requirement for handling a large number of samples, as well as the Spanish Culture Collection (CECT) strains used as reference
bench space, supplies, reagents, and labor needed, thus reducing strains for evaluation of modified TA10 broth were: S. enterica CECT
the overall cost of testing per pathogen (Elizaquivel & Aznar, 2008; 4594 (serovar Typhimurium, serotype 4, 5, 12: i: 1, 2) and
Ruiz-Rueda, Soler, & Calvo, 2010; Singh, Batish, & Grover, 2011). L. monocytogenes CECT 935. Bacteria were stored frozen at 20  C
Furthermore, multiplex detection is a rational approach since many until use. Additionally two strains of Salmonella spp. and one of
foods, such as milk, diary products, meat, poultry, fruits and L. monocytogenes were isolated from meat samples in a local food
vegetables, are common carriers of more than one pathogen like, control laboratory, and were used for spiking experiments. All other
Salmonella enterica, Escherichia coli O157:H7, and L. monocytogenes organisms used in this study are summarized in Table 1.
(Kim & Bhunia, 2008). Fresh cultures of all strains used in the present work were ob-
As can be observed from the high number of singleplex PCR tained by inoculating 10 mL tubes of Tryptic Soy Broth (TSB, Bio-
methods published for the detection of Salmonella spp. and Mérieux S.A., France). All strains were incubated at 37  C overnight,
L. monocytogenes (Calvo, Martinez-Planells, Pardos-Bosch, & Garcia- except for Bacillus subtilis which was incubated at 31  C.
Gil, 2008; Cheng et al., 2008; Hyeon et al., 2010; Kawasaki, Kimura, In order to know the reference values of each of both pathogenic
& Fujii, 2001; Lofstrom, Hansen, & Hoorfar, 2010; Malorny, Huehn, bacteria under study they were grown as described above, ten-fold
Dieckmann, Kramer, & Helmuth, 2009; Myint et al., 2006; Navas serially diluted in Buffered Peptone Water (BPW, Biokar diagnostics
et al., 2006; Poltronieri, de Blasi, & D’Urso, 2009; Rantsiou, Ales- S.A., France) and plated on Tryptic Soy Agar (TSA, Biokar diagnostics
sandria, Urso, Dolci, & Cocolin, 2008; Rodriguez-Lazaro, Jofre, S.A., France) for S. enterica and on Tryptic Soy Yeast Extract Agar
Aymerich, Hugas, & Pla, 2004; Rossmanith, Krassnig, Wagner, & (TSYEA, Biolife Italiana S.r.l., Italy) for L. monocytogenes. Plates were
Hein, 2006; Vanegas, Vasquez, Martinez, & Rueda, 2009) and also incubated 18 h at 37  C. Reference values of competitors were
multiplex PCR methods (Badosa et al., 2009; Elizaquivel, Gabaldon, plated on TSA, except for Vibrio cholera and Vibrio parahaemolyticus
& Aznar, 2010; Germini et al., 2009; Jofre et al., 2005; Kawasaki, which were plated on Nutrient Agar (NA, Biokar diagnostics S.A.,
Fratamico, Horikoshi, et al., 2010; Kim & Bhunia, 2008; Mahmuda France) supplemented with 10 g/L of sodium chloride.
et al., 2007; Omiccioli, Amagliani, Brandi, & Magnani, 2009; Ruiz- TA10 broth, commercial name of No. 17 broth (Kamisaki-
Rueda et al., 2010; Singh et al., 2011; Suo, He, Tu, & Shi, 2010; Horikoshi et al., 2011; Kawasaki et al., 2005), was chosen for
Zhang et al., 2009) among others, it is clear that these pathogens qPCR sample enrichment. The medium was modified as suggested
continue to be a health hazard and a serious problem for the food by Omiccioli et al. (2009), who stated that dextrose was not
industry which needs fast and reliable methods. necessary for successful recovery of Salmonella spp. and
78 A. Garrido et al. / Food Control 30 (2013) 76e85

Table 1 absorbance to evaluate capacity of selected bacterial species to

Strains used for productivity evaluation and sample inoculation. grow in the differently buffered versions of the broth (1). Second,
Bacteria Strain RT-PCR Co-culture individual growth of each bacterium without competitors, was
invA/hlyA reference evaluated by pure culture in three differently buffered broths and
value (cfu/mL) compared to a reference medium BPW (2). Third, simultaneous
Salmonella entericaa CECT 4594 þ/ 10e102 growth of Salmonella and L. monocytogenes, in all three differently
Salmonella sppb S1 þ/ e
buffered broths was evaluated with different competitors and, in
Salmonella sppb S2 þ/ e
Listeria monocytogenesa CECT 935 /þ 10e102 parallel, by adding selective agents to inhibit competitors, at
L. monocytogenesb Lm1 /þ e different times (3). Three different buffering salt concentrations
Escherichia colia CECT 434 / 102e103 were tested, A) the buffer specified in the ISO method (ISO, 1996)
Bacillus subtilisa CECT 435 / 20 for Half-Fraser broth (ISO Buffer: Na2HPO4 12.0 g/L; KH2PO4 1.35 g/
Staphylococcus aureusa CECT 240 / 3  107
Enterococcus faecalisa CECT 481 / 4  103
L), B) 150 mM (Na2HPO4 29.0 g/L; KH2PO4 5.0 g/L) and C) 100 mM
Vibrio parahaemolyticusa CECT 511 / 2  107 (Na2HPO4 19.3 g/L; KH2PO4 3.4 g/L), see Table 2. All three buffers
Vibrio choleraea CECT 514 (O1) / 108 tested rendered a final pH value of 7.2  0.2.
a
Strains used for evaluation of productivity. Inoculation of the different broths in the three steps of the
b
Natural strains isolated from meat in a local food laboratory. invA: Salmonella evaluation was done following the procedure described below. To
invasion gene, hlyA: hemolysin gene. CECT: Spanish type culture collection. obtain the desired bacterial concentration, from an overnight
culture in TSB of S. enterica and L. monocytogenes, turbidity was
L. monocytogenes. An additional modification performed in this adjusted to 2 McFarland with TSB in a Densimat (BioMérieux S.A.,
study, was to change the amount of buffering salts used to increase France) turbidimeter and ten-fold serially diluted in BPW.
the final pH of the broth to 7.2  0.2 (final medium composition: Initial turbidity of 2 McFarland corresponds to a theoretical
tryptose 10.0 g/L, beef extract 5.0 g/L, yeast extract 5.0 g/L, sodium concentration of 6  108 cfu/mL. Dilutions were done to reach
chloride 5.0 g/L, disodium phosphate 19.3 g/L, monopotassium theoretical inocula of 6 cfu (viable counts in the laboratory showed
phosphate 3.4 g/L). growth between 10 and 100 cfu/mL) for the inoculation of mTA10
Evaluation of simultaneous growth of Salmonella spp., and broth. Dilutions were plated on TSA and TSYEA for S. enterica and
L. monocytogenes, was determined by plating ten-fold serial dilutions L. monocytogenes respectively to get a viable bacteria reference
in Xylose Lysine Deoxycholate (XLD) agar and COMPASS Listeria agar value. Plates were incubated at 37  C overnight.
(Biokar diagnostics S.A., France) for viable counts of S. enterica and
L. monocytogenes respectively. Plates were incubated at 37  C for 24 h. 2.2.1. Growth kinetics
The growth kinetics study was done with two purposes, one, to
2.2. Evaluation of modified TA10 broth (mTA10) verify the suitability of the buffers selected for the growth of the
bacteria of interest, and second, to check if any of them was clearly
In order to evaluate the selected and modified broth, three steps better than the others in final absorbance, generation time or
were done, see Fig. 1. First, bacterial growth was monitored by duration of the lag phase. Buffering salts concentrations selected

Fig. 1. Schematic representation of the steps followed for the buffer evaluation of mTA10 broth. 1) Growth kinetics measured in microtiter plates in a SPECTRAmax instrument. A
and B are growth curves generated for S. enterica and L. monocytogenes respectively. 2) Viable counts after growth of S. enterica and L. monocytogenes in the different media tested
and comparison against reference broth. 3) Simultaneous growth of S. enterica and L. monocytogenes with other competitors under three conditions, NS: no supplement, 1 h:
supplementation after 1 h of incubation, 5 h: supplementation after 5 h of incubation.
 A. Garrido et al. / Food Control 30 (2013) 76e85 79

Table 2 was transferred to a new tube and centrifuged at 13,000 rpm for
Individual bacterial growth evaluation. 5 min, supernatant was discarded and the resulting pellet was
BPW ISO buffer 150 mM 100 mM resuspended in 1 mL PBS with Tween 20 (Rossmanith et al., 2006),
broth broth broth and centrifuged again under the same conditions, supernatant was
S. enterica 9.19  0.09ab 8.96  0.08a 9.13  0.11ab 9.32  0.21b discarded and bacterial cell pellet was processed according to the
L. monocytogenes 8.88  0.04a 9.02  0.14a 8.84  0.32a 9.21  0.02a respective method.
aeb
Different lower cases in the same row indicate statistically significant differences
(p < 0.05). Results expressed in Log cfu/mL. Data analyzed with one-way ANOVA 2.3.1. Boiling DNA-extraction method
Tukey-b.
Bacterial cell pellet was resuspended in 200 mL of PBS with
Tween 20 and boiled for 15 min. After boiling, suspension was
centrifuged at 13,000 rpm for 5 min at 4  C.
were tested for the growth of S. enteric and L. monocytogenes, by
spiking 200 mL of mTA10 medium prepared with the different
2.3.2. Lysis-GuSCN DNA-extraction method (Kawasaki et al., 2005)
buffers, with 2 mL rendering a viable count of 10e100 cfu, as rec-
Bacterial cells were resuspended in 200 mL of the enzyme
ommended in ISO 11133-2:2003 for productivity inocula (ISO,
solution containing 1 mg/mL achromopeptidase and 1 mg/mL
2003, p. 27), of the corresponding bacterium in a Falcon 96 well
lysozyme in TE 1 buffer. After incubation at 37  C with constant
plate and measured using SpectraMax M5 microplate reader (Bio-
agitation at 1000 rpm for 1 h, in a Thermomixer 5436 dry bath
Nova Científica S.L., Spain). Growth curves were constructed by
(Eppendorf AG, Germany) the solution was mixed with 300 mL of
measuring bacterial growth of S. enterica and L. monocytogenes at
4 M GuSCN containing 2% (wt/vol) Tween 20. A portion (400 mL) of
600 nm every 15 min for 24 h at 35  C.
the supernatant was transferred to a new tube containing 400 mL of
100% isopropanol. After mixing, the mixture was centrifuged for
2.2.2. Pure culture of S. enterica and L. monocytogenes in mTA10
10 min at 13,000 rpm, and the resulting DNA pellet was rinsed with
broth
75% isopropanol. The pellet was then dissolved in 160 mL of sterile
The growth of S. enterica and L. monocytogenes on mTA10 was
milli-Q water by heating with constant agitation at 14,000 rpm, at
evaluated by inoculating three series of tubes containing 10 mL of
70  C for 3 min. Prior to use, the template DNA solution was
mTA10 and BPW, with 10e100 cfu of the corresponding microor-
centrifuged for 5 min at 13,000 rpm to further remove water-
ganism. All tubes were incubated at 35  C for 24 h.
insoluble impurities.
After incubation ten-fold serial dilutions were done in BPW and
After extraction with both methods, DNA was quantified and
plated on TSA and TSYEA for S. enterica and L. monocytogenes
purity assessed, using a NanoDrop 1000 spectrophotometer
respectively. Plates were incubated at 37  C for 24 h.
(Thermo Fischer Scientific, Inc., USA) software ND-1000 v3.7.1.
2.2.3. Simultaneous growth of S. enterica and L. monocytogenes in
2.4. Genes, primers and probes used for qPCR method
mTA10 broth
For the evaluation of simultaneous bacterial growth 100 mL of
Primers and probe described by Cheng et al. (2008) targeting
medium were inoculated with 10e100 cfu of each target bacterium.
invA gene (Cheng et al., 2008) for the detection of Salmonella spp.
Competitors were also added, see Table 1. Inoculated medium was
tagged with FAM fluorescence dye, and those described by
incubated at 35  C 24 h.
Omiccioli et al. (2009) targeting hlyA for detection of
As previous authors have reported the advantages of using
L. monocytogenes tagged with Cy3 fluorescence dye were used. The
selective media for the simultaneous enrichment of Salmonella spp.
Internal Amplification Control (IAC) developed by Calvo et al.
and L. monocytogenes (Kim & Bhunia, 2008), in parallel the same
(2008) tagged with Texas Red fluorescence dye was also used.
broths were supplemented with selective agents (2 mg of Nalidixic
These Primers and probes were provided by IDT (Integrated DNA
acid, 2.5 mg of Acriflavine hydrochloride, and 100 mg of ferric
Technologies, Inc., USA).
ammonium citrate) added to the broth after 1 and 5 h of incubation
Specificity of primers and probes for selected genes was
(Wu, 2008), time left for recovery of stressed cells; then all media
extensively tested in original works, where Salmonella spp. primers
were incubated to fulfill the 24 h.
and probe were tested against 328 various Salmonella serovars and
After incubation ten-fold serial dilutions were done in BPW and
56 non-Salmonella strains (Cheng et al., 2008). Regarding
plated in XLD and COMPASS Listeria agar (Biokar diagnostics S.A.,
L. monocytogenes a total of 90 target and non-target bacterial strains
France) for both sets of experiments, none-supplement and selec-
were used to test the specificity (Omiccioli et al., 2009).
tive broth, with two “recovery times”. These plates were incubated
24 h at 37  C. After incubation, viable counts of presumptive
2.5. Multiplex qPCR detection method for Salmonella spp. and
colonies of S. enterica and L. monocytogenes were obtained. In
L. monocytogenes
addition to plate counts, qPCR was done after enrichment in every
variant of the broth (different buffers, with or without selective
The qPCR reaction was carried out in a final volume of 50 mL with
agents).
the following components: 30 mL of SsoFastÔ Probes Supermix
(Bio-Rad Laboratories, Inc., USA), 1000 nM primers and 250 nM
2.3. DNA-extraction methods comparison probe were used for Salmonella spp., 900 nM primers and 200 nM
probe were used for L. monocytogenes; and for Internal Amplifica-
Two different DNA-extraction methods were tested for their tion Control (IAC) 200 nM primers, 45 nM probe and 9  102 copies
application in the qPCR method described: conventional boiling of IAC DNA were added per reaction.
method, and the lysis-Guanidine isothiocyanate method (lysis- 5 mL of template DNA was added per reaction tube. Stratagene
GuSCN) (Kawasaki et al., 2005) with slight modifications. Mx3005p thermocycler (Agilent Technologies, Inc., USA) was used
For the comparison of both methods, 1 mL of pure cultures of with the following thermal profile: 2 min at 95  C for the activation
S. enterica and L. monocytogenes, grown overnight in 10 mL of TSB, of the polymerase (Hot Start), followed by 40 cycles, each cycle
were taken. Aliquots were centrifuged at 2000 rpm for 2 min in consisted on a denaturation step of 10 s at 95  C, and annealing-
a Biofuge fresco (Heraeus Instruments, England), the supernatant extension step at 63  C for 30 s.
80 A. Garrido et al. / Food Control 30 (2013) 76e85

Table 3 The qPCR method consisted of the enrichment of 25 g of sample

DNA-extraction protocols. in 225 mL mTA10 broth at 35  C for 24  2 h. After primary
Method DNA concentration Abs 260/280 Abs 260/230 enrichment, 1 mL was transferred to a tube containing 10 mL of
(ng/mL) new mTA10 broth and further incubated for a minimum of 8 h at
Boiling Salmonella 159.40  56.56 2.17  0.03 1.26  0.13a 35  C. After incubation, 1 mL was taken for DNA extraction as
Lysis-GuSCN 195.89  32.73 2.17  0.01 0.39  0.08a described in Section 2.3.2 and 5 mL were used as template.
Salmonella
Real-Time PCR results were gathered from both, pre-enrichment
Boiling L. 46.28  7.21b 2.71  0.02b 1.50  0.10b
monocytogenes broth and re-seeded tube, and statistically compared to evaluate
Lysis-GuSCN L. 141.03  42.05b 2.17  0.04b 0.27  0.08b the effect of the secondary enrichment.
monocytogenes

Data analyzed with ManneWhitney U-test.

a
Statistical differences for Salmonella. 2.8. Estimation of relative sensitivity, relative specificity, relative
b
Statistical differences for L. monocytogenes. accuracy, positive and negative predictive value and index kappa of
concordance of the qPCR method
2.6. qPCR efficiency
With the data from the LOD and additional spiked and blind
For calculation of the efficiency, overnight pure cultures of samples (with known result as were previously analyzed), see
S. enterica and L. monocytogenes in 10 mL of TSB, incubated at 37  C Table 5, relative sensitivity (SE), relative specificity (SP), relative
were used. DNA of each bacterium was extracted as described in accuracy (AC), positive and negative predictive values (PPV and
Section 2.3.2, and measured with the NanoDrop 1000 spectro- NPV) of the method were calculated. Evaluation of these parame-
photometer. For singleplex qPCR efficiency calculation, DNA from ters was done by comparing the qPCR method with the expected
S. enterica and L. monocytogenes were ten-fold serially diluted in results of the spiked samples or the blind, previously analyzed
sterile milli-Q water. In the same way for multiplex qPCR efficiency samples.
calculation, DNA from both bacteria were mixed in equal volumes, Each sample with positive (PA) and negative (NA) Accordance
again ten-fold serially diluted and analyzed. For both sets of were defined as samples presenting the same result, positive or
experiments 5 mL of each dilution was used as template for qPCR. negative, for the qPCR method and the expected result for spiked
qPCR efficiency was determined in duplicate for individual samples. Negative deviations (ND) are the number of samples ex-
singleplex detection of each pathogen. When calculating the effi- pected positive with a negative result, and positive deviations (PD),
ciency for simultaneous multiplex qPCR, three replicates were are the number of samples expected negative with a positive result.
done. SE was defined as the percentage of positive samples giving
The Mx3005pro software automatically calculates the standard a correct positive signal (SE ¼ PA/(PA þ ND)  100).
curve for each run based on the Cycle threshold (Ct) for each SP was defined as the percentage of negative samples giving
standard. The formula from which the amplification efficiency was a correct negative signal (SP ¼ NA/(PD þ NA)  100).
calculated is e ¼ 101/s  1 (Kawasaki, Fratamico, Horikoshi, et al., AC is defined as the degree of correspondence between the
2010), where s is the slope of the standard curve. response obtained by the expected result and the method on
identical samples (AC ¼ [(PA þ NA)/N]  100; where N ¼ number of
2.7. Evaluation of the limit of detection (LOD) in food samples by analyzed samples).
qPCR PPV and NPV are measures of the performance of the method by
giving the probability of a sample being really positive or negative
Evaluation of the LOD was done as previously described by when the method shows a positive or negative result PPV ¼ [(PA/
Garrido et al. (2012). Briefly, ten samples were inoculated with low PA) þ PD]  100; NPV ¼ [(NA/NA þ ND)  100].
concentration of both pathogens and 90% of positive results must Finally the index kappa of concordance shows the degree of
be achieved. concordance between the method and the expected result
Samples used for LOD are specified in Table 5 along with other k ¼ 2  (PA  NA  ND  PD)/[(PA þ PD)  (PD þ NA) þ
spiked and blind analyzed samples. The procedure was done twice (PA þ ND)  (ND þ NA)] (Anderson et al., 2011; Tomas, Rodrigo,
and spiking procedure was done following the same procedure Hernandez, & Ferrus, 2009).
described below.
Both target bacteria were grown overnight at 37  C in 10 mL TSB.
Turbidity to 0.5e2 McFarland was adjusted with sterile TSB. Ten- 2.9. Natural samples
fold serial dilutions were done to get a final concentration
ranging 1.5 to 6 cfu/mL for the inoculation of samples and TSA A survey was carried out with samples from local suppliers.
plates, for S. enterica, and TSYEA, for L. monocytogenes, to get Samples were transported to the laboratory, either frozen or
a reference value of viable bacteria after 18  2 h incubation at refrigerated, and were kept under the same conditions until
37  C. analysis.

Table 4
RT-PCR efficiency for S. enterica and L. monocytogenes pure DNA and DNA mixture.

RT-PCR DNA mixturea RT-PCR pure DNAb

S. enterica L. monocytogenes S. enterica L. monocytogenes

Amplification efficiency (%) 96.2  0.7 99.5  8.0 95.8  4.2 109.4  2.9
r2 0.999  0.000 0.998  0.002 0.997  0.000 0.999  0.000
Standard curve slope 3.415  0.018 3.344  0.187 3.430  0.110 3.117  0.058
a
All values are averages of three replicates.
b
All values are averages of two replicates.
 A. Garrido et al. / Food Control 30 (2013) 76e85 81

Table 5 (Na2HPO4 19.3 g/L; KH2PO4 3.4 g/L), each bacterium was grown in
Spiked and blind samples for evaluation of the method. 200 mL of medium inoculated with 10e100 cfu of the corresponding
Food type N RT-PCR Salmonella spp. L. monocytogenes bacterium and monitored by measuring absorbance at 600 nm
samples result
PA PD NA ND PA PD NA ND
every 15 min for 24 h at 35  C.
invA/hlyA S. enterica and L. monocytogenes grew correctly in all buffers
Bivalves 12 10/10 10 0 1 1 10 0 2 0 tested for mTA10 broth at 35  C, as it can be seen in Fig. 1 “A” and
Cephalopods 5 1/0 1 0 4 0 0 0 5 0
“B” but none of them showed differences respect to the others,
Meat 21 7/3 7 0 13 1b 6 0 14 1b
Vegetables 5 2/1 2 0 3 0 1 0 4 0 regarding maximum absorbance, maximum rate of growth or lag
Ready-to-eat 8 1/1 1 0 7 0 1 0 7 0 phase.
a
Boiled mussels 11 9/9 9 0 1 1 9 0 1 1
a
Samples used for LOD calculation. PA: positive agreement, PD: positive devia- 3.1.2. Pure culture of S. enterica and L. monocytogenes in mTA10
tion, NA: negative agreement, ND: negative deviation. broth
b
Inoculation level below the LOD. invA: Salmonella invasion gene, hlyA: hemo- Growth of S. enterica and L. monocytogenes in mTA10 broth was
lysin gene.
evaluated by inoculating 10 mL of medium with the different
buffering salts, with approximately 10 cfu of each bacterium. Ten
A total of 95 samples were analyzed which included fish, milliliters BPW tubes were inoculated in parallel, as a reference
vegetables, bivalves, cephalopods, crustaceans, ready-to-eat foods, medium. After incubation at 35  C for 24 h, ten-fold serial dilutions
byproducts, and environmental samples, see Table 6 for details. were done in BPW and plated in TSA and TSYEA for S. enterica and
L. monocytogenes respectively. Plates were incubated at 37  C for
2.10. Statistical analysis 24 h.
Statistical analysis did not show significant differences
Two different statistical analyses were applied. Comparison of (p < 0.05) in the viable counts of S. enterica in mTA10 with 100 mM
target bacteria growth in pure culture and in co-culture with or 150 mM buffers, and the reference broth (BPW). However, plate
competitors plus supplement was performed with one-way ANOVA counts increased significantly from 100 mM respect to the ISO
Tukey-b. For comparison of the Ct values obtained in co-cultures of buffer, see Table 2. Regarding L. monocytogenes although no
S. enterica and L. monocytogenes, Ct values from whole food matrix statistical significant differences were observed, there was a growth
versus the re-seeded tube qPCR, and DNA-extraction parameters increase in 100 mM broth similar to S. enterica proliferation.
(DNA concentration, and purity ratios 260/280 and 260/230),
ManneWhitney U-test was used. 3.1.3. Simultaneous growth of S. enterica and L. monocytogenes in
These analysis were performed using SPSS 15.0 software (SPSS mTA10 broth
Inc., Chicago, IL, USA). Evaluation of the co-culture was done by inoculating 100 mL of
the corresponding medium with a theoretical inoculum of 10 cfu of
each microorganism, in presence of competitors. After incubation
3. Results
ten-fold serial dilutions were done and plated in XLD and COMPASS
Listeria agar.
3.1. Evaluation of mTA10 broth
With the ISO buffer, for S. enterica highest colony counts were
obtained supplementing the medium with selective agents after
The evaluation of mTA10 broth was done in three consecutive
1 h, being the second highest result the medium without any
steps: 1) checking growth kinetics, 2) comparing growth against
supplement; regarding L. monocytogenes these results inversed,
a reference medium (BPW), and 3) checking co-culture growth of
highest values for the medium without supplement and second,
the target bacteria with competitors and with/without selective
supplemented after 1 h. For both bacteria, supplementation after
agents, as illustrates Fig. 1.
5 h showed the lowest results.
When the 150 mM buffer was used for S. enterica, lowest
3.1.1. Growth kinetics
productivity was obtained supplementing the broth with antibi-
To evaluate the suitability of the medium with the different
otics after 1 h. Higher viable counts were obtained when supple-
buffers tested, A) ISO Buffer (Na2HPO4 12.0 g/L; KH2PO4 1.35 g/L), B)
mented after 5 h or without supplementation. Regarding
150 mM (Na2HPO4 29.0 g/L; KH2PO4 5.0 g/L) and C) 100 mM
L. monocytogenes no statistical differences were observed.
Finally when the 100 mM buffer was tested, no statistical
Table 6 differences were observed in the data obtained for S. enterica
Natural samples analyzed. however, for L. monocytogenes significantly higher results were
Food type N RT-PCR results Food samples obtained with the medium without supplement.
samples Salmonella spp./ To summarize, even in presence of high number of competitors,
L. best results were obtained for S. enterica and L. monocytogenes
monocytogenes
without selective agents. Among these three buffers, statistical
Bivalves 14 0/2 Boiled musselsa,c analysis did not show any differences in the growth of
Fish 53 0/5 Pangassius,c patagonian
toothfish,c kingklipc and
L. monocytogenes, but it did for S. enterica in which significantly
tunaa,c higher data were collected for the ISO and the 100 mM buffer.
Cephalopods 9 0/0 Results obtained indicated that 150 mM buffer was the worst of
Crustaceans 1 0/0 the three buffers analyzed, so it was discarded. The Ct values ob-
Vegetables 1 0/1 Pre-fried onionc
tained for the ISO buffer and the 100 mM were compared with the
Ready-to-eat 9 0/1 Hake roe
Environmental 7 0/0 ManneWhitney U-test. Statistical differences (p < 0.05) were
Food byproducts 1 1/0 Fishmealb observed for L. monocytogenes but not for S. enterica, see lower part
a
Corresponding pathogen found in two samples.
of Fig. 2. The 100 mM buffer was selected considering that highly
b
Detected Salmonella spp. buffered broths are recommended in the literature when no
c
Detected L. monocytogenes. selective agents are added (Kawasaki et al., 2005; Omiccioli et al.,
82 A. Garrido et al. / Food Control 30 (2013) 76e85

Fig. 2. Simultaneous growth of S. enterica and L. monocytogenes in mTA10 broth with ISO buffer, 150 mM and 100 mM buffers. Upper part corresponds to viable counts obtained
without supplement (NS) and supplementation after 1 or 5 h post-incubation (1 h and 5 h respectively). aecDifferent lower cases in the same buffer group indicate statistically
significant differences (p < 0.05). Lower part corresponds to Ct (cycle threshold) values obtained for those viable counts shown in upper part. *Indicates statistically significant
differences (p < 0.05).

2009) and that in this study statistically lower Ct values were ob- (b ¼ 3.415  0.018) and 99.5% (b ¼ 3.344  0.187) respectively,
tained for L. monocytogenes using this type of broths, the 100 mM with an r2 of 0.999 and 0.998. When qPCR efficiency was deter-
buffer was selected to be added to the enrichment broth. mined using only pure DNA from each bacterium, values obtained
were 95.8% (b ¼ 3.430  0.110) and 109.4% (b ¼ 3.117  0.058)
3.2. DNA-extraction methods comparison respectively, with r2 of 0.997 and 0.999 respectively, see Table 4.

A simple boiling protocol and lysis-GuSCN method (Kawasaki 3.4. Evaluation of the limit of detection (LOD) for the multiplex
et al., 2005) were compared. qPCR method in inoculated food samples
Data obtained for S. enterica only showed significant differences
(p < 0.05) for 260/230 ratio, being higher for the boiling method. The evaluation of the LOD was done by spiking 10 samples
Regarding L. monocytogenes the boiling method also gave statisti- following the procedure described below. These experiments were
cally higher values for 260/230 ratio, but showed lower concen- done twice.
tration of DNA obtained and a lower 260/280 ratio, see Table 3. Samples were processed as described in Section 2.7. For the LOD
Finally the lysis-GuSCN method was selected. to be established, 90% of positive results must be achieved.
The first group of samples spiked for the LOD gave 9 out of 10
3.3. Singleplex and multiplex qPCR efficiency positive samples with an LOD of 4 cfu in 25 g, for Salmonella and 10
out of 10 for L. monocytogenes with 8 cfu in 25 g. As the objective of
Evaluation of singleplex efficiency was done by using pure DNA the study was to achieve a low LOD, it was re-evaluated with new
of S. enterica or L. monocytogenes. For multiplex qPCR efficiency, inoculations of low pathogens concentrations. Finally 9 out of the
DNA of each bacterium was mixed (ratio 1:1). Either for singleplex 10 samples were obtained simultaneously positive for S. enterica
or multiplex qPCR efficiency DNA was ten-fold serially diluted and and L. monocytogenes with an LOD of 5 cfu in 25 g of sample, for
5 mL of each dilution were used as template. Standard curve was each bacterium.
calculated by plotting the amount of DNA in ng, versus the Ct value Statistical analysis of the Ct data obtained indicated that after
obtained. 8 h of secondary enrichment in this broth, values obtained are
Efficiency for simultaneous detection of S. enterica and significantly lower than those reported from the whole food matrix
L. monocytogenes yielded an average efficiency of 96.2% sample.
 A. Garrido et al. / Food Control 30 (2013) 76e85 83

3.5. Estimation of relative sensitivity, relative specificity, relative On the other hand, different authors have shown that a two-step,
accuracy, positive and negative predictive value and index kappa of primary and secondary, enrichment is necessary to obtain
concordance of the qPCR method a higher number of positive results by PCR with L. monocytogenes
(Navas et al., 2006).
With all spiked and blind (natural samples previously analyzed) In addition to the presence of inhibitors in food and growth
samples analyzed in the present study LOD, relative sensitivity (SE), media, DNA extraction traditionally includes extraction of the sample
relative specificity (SP), relative accuracy (AC), positive and negative in phenolechloroform followed by precipitation of the DNA using
predictive values (PPV and NPV) and the index kappa of concordance cold ethanol. A disadvantage of this method is that a major fraction of
(k) were calculated following the formula shown in Section 2.8. the DNA is usually lost during the procedure and it is difficult to
All parameter analyzed for the evaluation of the quality of the process several samples simultaneously in a short time. Aqueous
method, SE, SP, AC, PPV and NPV showed values higher than 90%, two-phase systems have previously been used as a rapid and simple
see Table 7. Finally the k values calculated were 0.92 and 0.93. Data sample preparation method to separate PCR inhibitors in soft cheese
are summarized in Tables 5 and 7. from L. monocytogenes as well as to separate PCR inhibitors in faeces
from Helicobacter pylori. A disadvantage of this method is that the
3.6. Natural samples PCR sensitivity may be negatively affected if the target bacteria
partition to the interface of the aqueous two-phase system or to the
A total of 95 natural non spiked samples which covered a wide phase containing the PCR inhibitors. Buoyant density centrifugation
range of food and environmental samples were analyzed following has been used to remove PCR inhibitors when detecting pathogenic
the method developed. bacteria in blue cheese and in beef. A reduction in sensitivity of the
Salmonella spp. was detected in one sample (fishmeal) and 9 PCR may be attributed to attachment of the bacteria to the food
resulted positive for L. monocytogenes, including seafood and sample, especially minced meat. These attachments may result in
vegetables, see Table 6. a loss of detectable bacteria since only non-attached bacteria are
separated by buoyant density centrifugation (Lantz et al., 1998).
4. Discussion When a negative result is detected it is important to know whether
PCR failure occurred of if it was a real negative PCR result. A negative
Traditional microbiological methods for the detection of path- PCR result does not necessarily indicate that no template DNA was
ogenic bacteria involve several steps of enrichment, plate isolation present in the sample. Inhibitory substances present in the sample
and biochemical identification, taking up to 7e10 days in the case of may influence the outcome of the PCR by lowering or completely
Salmonella spp. and L. monocytogenes (Jofre et al., 2005). However preventing the amplification (Rip & Gouws, 2009).
PCR-based techniques have the potential to allow for rapid and Previous studies have classically used two broths for the
sensitive detection of foodborne pathogens. Since PCR can target enrichment of Salmonella spp. and L. monocytogenes (Buffered
unique genetic sequences such as virulence genes of microorgan- Peptone Water and Half-Fraser broth, respectively (Badosa et al.,
isms, it also has the advantage of potentially being an extremely 2009; Chua & Bhagwat, 2009; Jofre et al., 2005; Ruiz-Rueda et al.,
specific assay (Fratamico & Strobaugh, 1998). But, even these rapid 2010)) or have tried the application or development of single broth
detection methods still require an enrichment procedure (Hyeon for both bacteria, either general (Tryptic Soy Broth (Germini et al.,
et al., 2010; Kobayashi et al., 2009). 2009), Nutrient Broth (Zhang et al., 2009), Universal Pre-
Bacterial pathogens may coexist, at different concentrations, in enrichment Broth (Bhagwat, 2003), SEB (Kobayashi et al., 2009),
the same food sample, but they usually occur at low levels. Their No. 17 (Kawasaki et al., 2005)) or selective (SEL (Kim & Bhunia,
detection is usually preceded by an enrichment step to increase cell 2008)). One of the main factors affecting bacterial recovery from
numbers to the detection level. For the simultaneous detection of food is the drop in pH associated with bacterial growth. This is the
more than one pathogen, differences in growth requirements and reason why many authors have suggested the use of highly buffered
growth rates must be considered (Alarcon, Garcia-Canas, Cifuentes, broths and poor in carbohydrates (Kawasaki et al., 2005; Omiccioli
Gonzalez, & Aznar, 2004). A second major issue added to the et al., 2009). A factor not considered in most previous studies is that
selection of the appropriate enrichment medium is the presence of most media intended for the recovery of L. monocytogenes and
certain substances that inhibit the PCR reaction. These compounds Salmonella spp. present slightly basic pH values (BPW, TSB, Fraser
can contaminate the DNA templates extracted from food samples or Broth, Buffered Listeria Enrichment Broth, among others). The
environmental samples such as air, soil, and water, and may in turn modified medium described in the present work was designed
generate false-negative results (Hyeon et al., 2010). following this concept, with a final pH value 7.2  0.2. In addition,
Many authors have demonstrated the advantages of using a simple DNA-extraction method without phenol and chloroform
multiplex PCR methods (Elizaquivel & Aznar, 2008; Ruiz-Rueda ensures not only the high sensitivity of the multiplex PCR assay but
et al., 2010; Singh et al., 2011). In the multiplex PCR assay for the also safety in handling for practical use (Kawasaki, Fratamico,
detection of different bacterial strains, the ability to detect small Kamisaki-Horikoshi, et al., 2010).
numbers of cells in foods was considered strongly dependent of the Results obtained from the kinetics with the different buffering
DNA-extraction procedure used (Kawasaki et al., 2005). Detection strengths show that all broths studied in this work would be suit-
of L. monocytogenes with PCR is less sensitive than the corre- able for the enrichment of Salmonella, but the medium with
sponding methods for Gram-negative pathogens, which can be 100 mM achieved the highest growth for L. monocytogenes.
related to low levels of Listeria in the primary enrichment cultures. These results were verified by enriching pure cultures of each
microorganism and comparing them with the growth obtained in
Table 7 BPW. According to plate counts, growth of Salmonella in broth with
Method evaluation.
100 mM buffer was significantly higher than in other buffers and
SE SP AC PPV NPV k BPW; regarding L. monocytogenes there was no statistical difference
S. enterica 91 100 95 1 91 0.92 in plate counts.
L. monocytogenes 93 100 97 1 94 0.93 Finally, all three buffers were compared for the simultaneous
SE: relative sensitivity, SP: relative specificity, AC: relative accuracy, PPV: positive growth of both target bacteria, but with the addition of other
predictive value, NPV: negative predictive value, k: index kappa of concordance. competitors and selective agents. Data from all analysis showed, in
84 A. Garrido et al. / Food Control 30 (2013) 76e85

general, a better performance of the different broths without the Limit of detection was established in 5 cfu/25 g for each path-
selective supplement, thus these were compared to check which one ogen, with a 90% of positive results in spiked samples. Several
is the best and it was found that there were no significant differences spiked samples proved that detection could be as low as 3 cfu/25 g,
for the growth of L. monocytogenes. In the case of Salmonella plate but this inoculum concentration was not further evaluated as ND
counts were higher for the ISO and the 100 mM buffer. were observed for both pathogens, see Table 5. Inclusion of the
Previous authors (Kim & Bhunia, 2008; Mahmuda et al., 2007) secondary enrichment step did not enhance the number of positive
have reported problems on the recovery of L. monocytogenes when samples for LOD but did significantly decreased the Ct values ob-
high numbers of competitors were present in the samples. In the tained, when compared to that of whole matrix. This effect may be
present study, with the results gathered from the different steps in due to extending sample incubation in new medium and
the evaluation of the most suitable buffer for the broth, the 100 mM decreasing the amount of food particles. Reduction of food particles
was chosen as its buffering capacity is higher than the ISO buffer. has been previously recommended for enhanced DNA extraction
Also high viable counts were obtained for both pathogens, thus (Kawasaki et al., 2009; Malorny et al., 2009). The method described
supporting simultaneous growth of Salmonella spp. and in the present work proved to overcome troubles related to LOD
L. monocytogenes even in the presence of a high concentration of expressed by previous authors (Mahmuda et al., 2007).
competitors with different requirements. These results were When the method was applied to natural non spiked samples
confirmed by plate counts and qPCR, see Fig. 2. detection of both pathogens was possible in different types of
Evaluation of DNA-extraction protocols did not show statistical samples (fish, bivalves, vegetables and fishmeal), thus proving its
differences when applied for S. enterica but, when data for applicability to samples from different origins.
L. monocytogenes were analyzed, significant differences were ob-
tained for all three parameters evaluated. Results are consistent
5. Conclusions
with data reported by Kawasaki et al. (2005) who showed that even
there were no differences in the sensitivity achieved by boiling of
In the present study the modification of TA10 broth was
Lysis-GuSCN method when applied either on Salmonella Enteritidis
completely evaluated for the simultaneous growth of S. enterica and
or E. coli O157:H7 (both Gram-negative bacteria) but when used
L. monocytogenes in the presence of high numbers of different
with L. monocytogenes (Gram-positive) only the sensitivity of the
competitors. Previous authors reported difficulties recovering,
Lysis-GuSCN method was comparable to that obtained for Gram-
specially L. monocytogenes, when it was overnumbered by
negative bacteria.
competitors (Kim & Bhunia, 2008; Mahmuda et al., 2007).
In the present paper the purity of the DNA obtained was also
The evaluation of the lysis-GuSCN DNA-extraction method
evaluated. Concerning Salmonella no differences were observed for
showed that for Gram-positive bacteria a statistically higher
260/280 ratio, but the boiling method showed statistically higher
amount of DNA could be obtained. The drawback noted was the
values of 260/230 ratio, closer to the optimum of pure DNA
lack of DNA purity, specially observed by the low 260/230 ratio.
(z2.0e2.2). In the case of L. monocytogenes the boiling method had
The qPCR method showed high efficiency that coupled with the
statistically higher values of 260/280 thus, further from ideal pure
enrichment medium and the DNA-extraction protocol selected, was
DNA (z1.8); also significant differences were observed for 260/230
able to detect 3 cfu of Salmonella and L. monocytogenes in 25 g of
were the boiling method exhibited better results, closer to ideally
samples, even though the LOD was established at 5 cfu in 25 g for
pure DNA. The slightly high values obtained with the lysis-GuSCN
both bacterial pathogens.
method for 260/280 ratio may be due to traces of GuSCN in the
Inclusion of an IAC to discard possible PCR reaction inhibition
final extract as this compound absorbs at 260 nm. The low values
and avoid false-negative results, makes the method described in
measured for 260/230 may be related to the presence of organic
the present study suitable for application in routine food and
compounds or chaotropic salts (GuSCN) in the purified DNA. If this
environmental samples control.
method was intended for other purposes where additional DNA
To summarize, the use of mTA10 broth with 100 mM buffer with
purity was required, the extract obtained may be coupled with any
the secondary enrichment step, coupled with the lysis-GuSCN
commercial purification kit.
DNA-extraction method can be used with the qPCR detection
Even though the DNA purity parameters measured were out of
system described in the present study for the simultaneous
the ideally pure DNA values, the overall method performed perfectly
screening of Salmonella spp. and L. monocytogenes from food and
well as it could be extrapolated from the Ct values observed for the
environmental samples.
IAC from all samples, were no remarkable alterations were detected,
which may have indicated presence of inhibitory substances not
eliminated during the DNA-extraction protocol. Furthermore, all Acknowledgments
quality parameters calculated for the method (SE, SP, AC, PPV and
NPV) were above 90%, even more, the k value for both, Salmonella This work is financially supported by the Secretary General for
and L. monocytogenes, were above 0.9, keeping in mind that values of the Sea of the Spanish Ministry of Agricultural, Land and Marine
0.81e1 indicate a “very good concordance” (DG, 1991) between the Resources (MARM), by order ARM/1193/2009. Authors also thank
method and the expected result. Victoria Docampo and Angeles Marcote for technical assistance,
The PCR characteristics can be defined from a standard curve and Lema & Bandin Laboratories for the strains isolated from meat.
based on ten-fold serial dilutions of the DNA or cDNA (reference
material), within the dynamic range of the method. The slope of the References
linear regression line, ideally 3.3219, results in a real-time PCR
efficiency of 1 indicating that the number of target molecules Alarcon, B., Garcia-Canas, V., Cifuentes, A., Gonzalez, R., & Aznar, R. (2004). Simul-
taneous and sensitive detection of three foodborne pathogens by multiplex
exactly doubles in one PCR cycle. Slopes between 3.1 and 3.6,
PCR, capillary gel electrophoresis, and laser-induced fluorescence. Journal of
with efficiency percentage between 90 and 110 are generally Agricultural and Food Chemistry, 52(23), 7180e7186.
acceptable (Raymaekers, Smets, Maes, & Cartuyvels, 2009). All Anderson, A., Pietsch, K., Zucker, R., Mayr, A., Muller-Hohe, E., Messelhausser, U.,
values calculated for efficiency either singleplex or multiplex et al. (2011). Validation of a Duplex real-time PCR for the detection of Salmo-
nella spp. in different food products. Food Analytical Methods, 4(3), 259e267.
resulted in values above 90% and below 110% and the slope values Badosa, E., Chico, N., Pla, M., Pares, D., & Montesinos, E. (2009). Evaluation of ISO
between 3.1 and 3.4, see Table 3. enrichment real-time PCR methods with internal amplification control for
 A. Garrido et al. / Food Control 30 (2013) 76e85 85

detection of Listeria monocytogenes and Salmonella enterica in fresh fruit and Kim, H. C., & Bhunia, A. K. (2008). SEL, a selective enrichment broth for simulta-
vegetables. Letters in Applied Microbiology, 49(1), 105e111. neous growth of Salmonella enterica, Escherichia coli O157:H7, and Listeria
Bhaduri, S., & Cottrell, B. (2001). Sample preparation methods for PCR detection of monocytogenes. Applied and Environmental Microbiology, 74(15), 4853e4866.
Escherichia coli O157:H7, Salmonella typhimurium, and Listeria monocytogenes Kimura, B., Kawasaki, S., Fujii, T., Kusunoki, J., Itoh, T., & Flood, S. J. A. (1999).
on beef chuck shoulder using a single enrichment medium. Molecular and Evaluation of TaqMan PCR assay for detecting Salmonella in raw meat and
Cellular Probes, 15(5), 267e274. shrimp. Journal of Food Protection, 62(4), 329e335.
Bhagwat, A. A. (2003). Simultaneous detection of Escherichia coli O157:H7, Listeria Kobayashi, H., Kubota, J., Fujihara, K., Honjoh, K., Ito, M., Fujiki, N., et al. (2009).
monocytogenes and Salmonella strains by real-time PCR. International Journal of Simultaneous enrichment of Salmonella spp, Escherichia coli O157:H7, Vibrio
Food Microbiology, 84(2), 217e224. parahaemolyticus, Staphylococcus aureus, Bacillus cereus, and Listeria mono-
Calvo, L., Martinez-Planells, A., Pardos-Bosch, J., & Garcia-Gil, L. J. (2008). A new cytogenes by single broth and screening of the pathogens by multiplex real-time
real-time PCR assay for the specific detection of Salmonella spp. Targeting the PCR. Food Science and Technology Research, 15(4), 427e438.
bipA gene. Food Analytical Methods, 1(4), 236e242. Lantz, P. G., Hahnhagerdal, B., & Radstrom, P. (1994). Sample preparation methods in
Chapela, M. J., Fajardo, P., Garrido, A., Cabado, A. G., Ferreira, M., Lago, J., et al. (2010). PCR-based detection of food pathogens. Trends in Food Science & Technology,
Comparison between a TaqMan polymerase chain reaction assay and a culture 5(12), 384e389.
method for ctx-positive Vibrio cholerae detection. Journal of Agricultural and Lantz, P. G., Knutsson, R., Blixt, Y., Abu Al-Soud, W., Borch, E., & Radstrom, P. (1998).
Food Chemistry, 58(7), 4051e4055. Detection of pathogenic Yersinia enterocolitica in enrichment media and pork by
Cheng, C. M., Lin, W., Van, K. T., Phan, L., Tran, N. N., & Farmer, D. (2008). Rapid a multiplex PCR: a study of sample preparation and PCR-inhibitory compo-
detection of Salmonella in foods using real-time PCR. Journal of Food Protection, nents. International Journal of Food Microbiology, 45(2), 93e105.
71(12), 2436e2441. Lofstrom, C., Hansen, F., & Hoorfar, J. (2010). Validation of a 20-h real-time PCR
Chua, T., & Bhagwat, A. A. (2009). A rapid and simple DNA extraction procedure to method for screening of Salmonella in poultry faecal samples. Veterinary
detect Salmonella spp. and Listeria monocytogenes from fresh produce using Microbiology, 144(3e4), 511e514.
real-time PCR. Food Analytical Methods, 2(2), 96e101. Mahmuda, Y., Kawasaki, S., & Kawamoto, S. (2007). Evaluation of multiplex PCR
DG, A. (1991). Practical statistics for medical research. New York. system for simultaneous detection of Escherichia coli O157:H7, Listeria mono-
(EC), C. R. (2005). Microbiological criteria for foodstuffs. Official Journal of the cytogenes and Salmonella enteritidis in shrimp samples. Bangladesh Journal of
European Union, 2073/2005. Microbiology, 24(1), 42e46.
(EC), C. R. (2007). Amending Regulation (EC) No 2073/2005 on microbiological Malorny, B., Hoorfar, J., Bunge, C., & Helmuth, R. (2003). Multicenter validation of
criteria for foodstuffs. Official Journal of the European Union, 1441/2007. the analytical accuracy of Salmonella PCR: towards an international standard.
Elizaquivel, P., & Aznar, R. (2008). A multiplex RTi-PCR reaction for simultaneous Applied and Environmental Microbiology, 69(1), 290e296.
detection of Escherichia coli O157:H7, Salmonella spp. and Staphylococcus Malorny, B., Huehn, S., Dieckmann, R., Kramer, N., & Helmuth, R. (2009). Polymerase
aureus on fresh, minimally processed vegetables. Food Microbiology, 25(5), chain reaction for the rapid detection and serovar identification of Salmonella in
705e713. food and feeding stuff. Food Analytical Methods, 2(2), 81e95.
Elizaquivel, P., Gabaldon, J. A., & Aznar, R. (2010). Quantification of Salmonella spp, Myint, M. S., Johnson, Y. J., Tablante, N. L., & Heckert, R. A. (2006). The effect of pre-
Listeria monocytogenes and Escherichia coli O157 H7 in non-spiked food prod- enrichment protocol on the sensitivity and specificity of PCR for detection of
ucts and evaluation of real-time PCR as a diagnostic tool in routine food anal- naturally contaminated Salmonella in raw poultry compared to conventional
ysis. Food Control, 22(2), 158e164. culture. Food Microbiology, 23(6), 599e604.
Fratamico, P. M., & Strobaugh, T. P. (1998). Simultaneous detection of Salmonella spp Navas, J., Ortiz, S., Lopez, P., Jantzen, M. M., Lopez, V., & Martinez-Suarez, J. V. (2006).
and Escherichia coli O157:H7 by multiplex PCR. Journal of Industrial Microbiology Evaluation of effects of primary and secondary enrichment for the detection of
& Biotechnology, 21(3), 92e98. Listeria monocytogenes by real-time PCR in retail ground chicken meat. Food-
Garrido, A., Chapela, M. J., Ferreira, M., Atanassova, M., Fajardo, P., Lago, J., et al. borne Pathogens and Disease, 3(4), 347e354.
(2012). Development of a multiplex real-time PCR method for pathogenic Vibrio Omiccioli, E., Amagliani, G., Brandi, G., & Magnani, M. (2009). A new platform for
parahaemolyticus detection (tdhþ and trhþ). Food Control, 24(1e2), 128e135. real-time PCR detection of Salmonella spp., Listeria monocytogenes and Escher-
Germini, A., Masola, A., Carnevali, P., & Marchelli, R. (2009). Simultaneous detection ichia coli O157 in milk. Food Microbiology, 26(6), 615e622.
of Escherichia coli O175:H7, Salmonella spp., and Listeria monocytogenes by Poltronieri, P., de Blasi, M. D., & D’Urso, O. F. (2009). Detection of Listeria mono-
multiplex PCR. Food Control, 20(8), 733e738. cytogenes through real-time PCR and biosensor methods. Plant Soil and Envi-
Hammack, T. S., Amaguana, R. M., Johnson, M. L., & Andrews, W. H. (2003). Effec- ronment, 55(9), 363e369.
tiveness of universal pre-enrichment broth for recovery of Salmonella from Rantsiou, K., Alessandria, V., Urso, R., Dolci, P., & Cocolin, L. (2008). Detection,
selected dairy foods. Journal of AOAC International, 86(4), 714e718. quantification and vitality of Listeria monocytogenes in food as determined by
Hyeon, J. Y., Hwang, I. G., Kwak, H. S., Park, C., Choi, I. S., & Seo, K. H. (2010). Eval- quantitative PCR. International Journal of Food Microbiology, 127(3), 321, (Vol.
uation of PCR inhibitory effect of enrichment broths and comparison of DNA 121, pg 99, 2008).
extraction methods for detection of Salmonella Enteritidis using real-time PCR Raymaekers, M., Smets, R., Maes, B., & Cartuyvels, R. (2009). Checklist for optimi-
assay. Journal of Veterinary Science, 11(2), 143e149. zation and validation of real-time PCR assays. Journal of Clinical Laboratory
ISO. (1996). Microbiology of food and animal feeding stuffs e Horizontal method for Analysis, 23(3), 145e151.
the detection and enumeration of Listeria monocytogenes e Part 1: Detection Rijpens, N., & Herman, L. (2002). Molecular methods for identification and detection
method, Vol. 11290-1. of bacterial food pathogens. Journal of AOAC International, 85(2), 984e995.
ISO. (2003). Microbiology of food and animal feeding stuffs e Guidelines on prepa- Rip, D., & Gouws, P. A. (2009). Development of an internal amplification control
ration and production of culture media e Part 2: Practical guidelines on perfor- using multiplex PCR for the detection of Listeria monocytogenes in food prod-
mance testing of culture media, Vol. 11133-2. ucts. Food Analytical Methods, 2(3), 190e196.
Jofre, A., Martin, B., Garriga, M., Hugas, M., Pla, M., Rodriguez-Lazaro, D., et al. Rodriguez-Lazaro, D., Jofre, A., Aymerich, T., Hugas, M., & Pla, M. (2004). Rapid
(2005). Simultaneous detection of Listeria monocytogenes and Salmonella by quantitative detection of Listeria monocytogenes in meat products by real-time
multiplex PCR in cooked ham. Food Microbiology, 22(1), 109e115. PCR. Applied and Environmental Microbiology, 70(10), 6299e6301.
Kamisaki-Horikoshi, N., Okada, Y., Takeshita, K., Sameshima, T., Kawasaki, S., Rossmanith, P., Krassnig, M., Wagner, M., & Hein, I. (2006). Detection of Listeria
Kawamoto, S., et al. (2011). Evaluation of TA10 broth for recovery of heat- and monocytogenes in food using a combined enrichment/real-time PCR method
freeze-injured Salmonella from beef. Journal of AOAC International, 94(3), targeting the prfA gene. Research in Microbiology, 157(8), 763e771.
857e862. Ruiz-Rueda, O., Soler, M., & Calvo, L. (2010). Multiplex real-time PCR for the
Kawasaki, S., Fratamico, P. M., Horikoshi, N., Okada, Y., Takeshita, K., Sameshima, T., simultaneous detection of Salmonella spp. and Listeria monocytogenes in food
et al. (2009). Evaluation of a multiplex PCR system for simultaneous detection samples. Food Analytical Methods, .
of Salmonella spp., Listeria monocytogenes, and Escherichia coli O157:H7 in foods Singh, J., Batish, V. K., & Grover, S. (2011). Simultaneous detection of Listeria mon-
and in food subjected to freezing. Foodborne Pathogens and Disease, 6(1), 81e89. ocytogenes and Salmonella spp. in dairy products using real time PCR-melt curve
Kawasaki, S., Fratamico, P. M., Horikoshi, N., Okada, Y., Takeshita, K., Sameshima, T., analysis. Journal of Food Science and Technology, .
et al. (2010). Multiplex real-time polymerase chain reaction assay for simulta- Suo, B., He, Y. P., Tu, S. I., & Shi, X. M. (2010). A multiplex real-time polymerase chain
neous detection and quantification of Salmonella species, Listeria mono- reaction for simultaneous detection of Salmonella spp., Escherichia coli O157,
cytogenes, and Escherichia coli O157:H7 in ground pork samples. Foodborne and Listeria monocytogenes in meat products. Foodborne Pathogens and Disease,
Pathogens and Disease, 7(5), 549e554. 7(6), 619e628.
Kawasaki, S., Fratamico, P. M., Kamisaki-Horikoshi, N., Okada, Y., Takeshita, K., Tomas, D., Rodrigo, A., Hernandez, M., & Ferrus, M. A. (2009). Validation of real-time
Sameshima, T., et al. (2010). Development of the multiplex PCR detection kit for PCR and enzyme-linked fluorescent assay-based methods for detection of
Salmonella spp., Listeria monocytogenes, and Escherichia coli O157:H7. JARQ e Salmonella spp. in chicken feces samples. Food Analytical Methods, 2(3), 180e189.
Japan Agricultural Research Quarterly, 45(1), 77e81. Vanegas, M. C., Vasquez, E., Martinez, A. J., & Rueda, A. M. (2009). Detection of
Kawasaki, S., Horikoshi, N., Okada, Y., Takeshita, K., Sameshima, T., & Kawamoto, S. Listeria monocytogenes in raw whole milk for human consumption in Colombia
(2005). Multiplex PCR for simultaneous detection of Salmonella spp., Listeria by real-time PCR. Food Control, 20(4), 430e432.
monocytogenes, and Escherichia coli O157:H7 in meat samples. Journal of Food Wu, V. C. H. (2008). A review of microbial injury and recovery methods in food. Food
Protection, 68(3), 551e556. Microbiology, 25(8), 1001, (Vol. 25, pg 735, 2008).
Kawasaki, S., Kimura, B., & Fujii, T. (2001). Comparison of TaqMan (TM) Salmonella Zhang, D., Zhang, H., Yang, L., Guo, J., Li, X., & Feng, Y. (2009). Simultaneous
amplification/detection kit with standard culture procedure for detection of detection of Listeria monocytogenes, Staphylococcus aureus, Salmonella enterica
Salmonella in meat samples. Journal of the Food Hygienic Society of Japan, 42(1), and Escherichia coli O157:H7 in food samples using multiplex PCR method.
33e39. Journal of Food Safety, 29(3), 348e363.