31 Emerging Bacterial Foodborne Pathogens and Methods of Detection 719ples (Hoffman and Wedman 2001). In one outbreak
involving hot dogs in the United States, illness was
attributed to contamination levels 0.3 cfu/g of hot
dog meat (Donnelly 2001). DNA-based methods of
detection employ ways of amplifying the signal of a
few cells more quickly than waiting for cell replica-
tion. The following methods represent some of the
common ways in which nucleic acids have been
used as detection targets for foodborne pathogens,
and some comparison of the techniques is presented
in Table 31.3.
Polymerase Chain Reaction (PCR)
PCR is the basis of many nucleic acid–based detec-
tion systems. With this method, total DNA is ex-
tracted from the food sample. Next, two
oligonucleotide primers are selected that bind to a
pathogen-specific target gene at opposite ends of
opposing strands of DNA. A DNA polymerase is
added along with the four types of deoxynucleotides
(dNTPs) and appropriate buffers, and the mixture is
inserted into a thermocycler. This machine cycles
through a temperature regime that usually involves
94°C to denature the DNA template, an annealing
temperature of 50–65°C to allow the oligonucleotide
primers to bind to the gene, and 72°C to allow ex-
tension by the DNA polymerase. The thermocycler
runs through these temperatures a number of times to
amplify the gene fragment exponentially. The result-
ing fragment is stained with ethidium bromide and
visualized on an agarose gel under UV light. The
presence of a pathogen can be detected by simply
determining if a band representing the pathogen gene
of interest is present.
Performing PCR takes only a few hours, shorten-
ing the time required for pathogen detection dramat-
ically, and is conducive to automation and high
throughput processing using 96-well plates. Aznar
and Alarcón (2002) and Shearer et al. (2001) claim
that PCR-based detection of L. monocytogenesis
more sensitive than culture-based methods for de-
tecting the pathogen in contaminated food samples
because more samples turned up positive for the L.
monocytogenesusing the PCR method. The authors
attribute the increase in sensitivity of the PCR
method over culturing methods to the fact that the
former does not have an initial selection step. The
PCR method circumvents the problem that some
cells do not grow on the selective media. The au-
thors also claim that some of the false negatives re-
ported for culturing methods are due to methods that
use color changes to differentiate between species,
and that these methods are influenced by the subjec-
tivity of the observer (Aznar and Alarcón 2002,
Shearer et al. 2001).
Among the target genes for PCR detection of L.
monocytogenesare the hlyAgene (encoding LLO;
Blais et al. 1997, Hough et al. 2002, Hudson et al.
2001, Lehner et al. 1999, Lunge et al. 2002, Ryser et
al. 1996, Weidmann et al. 1997), the iapgene (en-
coding an invasion-associated protein; Cocolin et al.
2002, Schmid et al. 2003), inlB (encoding internalin
B; Ingianni et al. 2001, Jung et al. 2003, Lunge et al.
2002, Pangallo et al. 2001) and 16S rRNA (Call et
al. 2003, Schmid et al. 2003). Among these genes,
the most commonly used has been hlyA(Aznar and
Alarcón 2002).
Wan et al. (2003) compared the use of PCR to
detect L. monocytogenesin salmon with Interna-
tional Organization for Standardization culturing
method 11290-1, and found the two methods gave
comparable results in spiked samples if enrichment
culturing is used prior to PCR to lower the detection
limit for the L. monocytogenes. The only difference
is that the PCR method requires 58–60 hours to per-
form rather than 5 days (Wan et al. 2003).
Jung et al. (Jung et al. 2003) also used PCR to
detect L. monocytogenesusing the internalin AB
(inlAB) gene. PCR allowed the specific detection of
a number of serotypes of L. monocytogenes,while
no bands from other nonpathogenic species of Lis-
teriawere detected. The limit of detection of the
pathogen in pure cultures was 10^5 cells/mL culture.
When frankfurters were spiked with the pathogen,
the limit of detection was improved to 10 cells in a
25 g sample, provided the sample was first enriched
in a modified enrichment broth for at least 16 hours
(Jung et al. 2003). This method allowed detection of
the pathogen within 24 hours.
PCR methods are able to detect the presence of
the pathogen, but they are not able to quantify the
level of pathogen contamination. One way to ap-
proach this problem is the use of competitive PCR
(see Fig. 31.1). In this case, a competitor fragment
of DNA that matches the gene to be amplified is
introduced into the sample. In general, the competi-
tor fragment is synthesized as a deletion mutant that
can be distinguished from the pathogen gene by its
smaller size. In order to determine the level of