A New Protocol to Detect Multiple Foodborne Pathogens with PCR Dipstick DNA Chromatography after a Six-Hour Enrichment Culture in a Broad-Range Food Pathogen Enrichment Broth

A quick foodborne pathogen screening method after six-hour enrichment culture with a broad-range food pathogen enrichment broth is described. Pathogenic factors of Salmonella enterica, Shigella spp., enteroinvasive Escherichia coli, and enterohemorrhagic E. coli are amplified with a cocktail primer and rapid polymerase chain reaction (PCR), which finishes amplification in 30 min. The PCR amplicon was differentiated with a dipstick DNA chromatography assay in 5–10 min. Starting from a four- to six-hour enrichment culture, this assay was finished within 45 min. Detection sensitivity of this protocol was less than 2.5 CFU/25 g for S. enterica and 3.3 CFU/25 g for enterohemorrhagic E. coli in spiked ground meat experiments.


Introduction
Infectious gastroenteritis is a leading cause of morbidity and mortality worldwide, particularly in developing countries [1]. Risk factors for infectious gastroenteritis include exposure to various contaminated food products [2]. Several methods to detect pathogens directly in food samples have been reported [3,4]; however, most food analysis requires a 25 g food sample. Ideally, pathogen detection in food should be at the single-cell level [5].
Several methods based on polymerase chain reaction (PCR) have been developed to detect a single-cell pathogen from enrichment culture [6][7][8]. Cocktail PCR, carried out in a single PCR tube for simultaneous detection of more than one bacterial target, has been investigated as a more cost-effective and time-saving method [9,10]. However, it is difficult for small food laboratories to use ethidium bromide-based agarose gel. On the other hand, real-time PCR assays employing various types of fluorescence systems allow multiple detection during PCR [11][12][13]. This is an excellent method, but it requires an expensive real-time thermal cycler and reagents. Thus, small laboratories cannot afford this real-time method.
Another aspect of food analysis is the analysis time. Fresh food products must arrive to the market quickly, but current culture-based protocols require several days to confirm that the products are pathogen-free. Confirmation that fresh food is safe before shipping is, therefore, desired, but difficult in practice.

BioMed Research International
To solve these problems, we developed a quick cocktail PCR and dipstick DNA chromatography to differentiate PCR amplicons for Salmonella enterica, Shigella spp., enteroinvasive Escherichia coli (EIEC), and enterohemorrhagic E. coli (EHEC) from a single enrichment culture broth. In our previous report, we described a food pathogen enrichment (FPE) broth that supports the growth of Campylobacter without adding lysed blood and carbon dioxide [14]. The method detected a few Campylobacter cells in 25 g of chicken within 24 hours and was better than the conventional Bolton-based enrichment culture.
In this report, we describe a new protocol to detect S. enterica, Shigella spp., EIEC, and EHEC from the FPE broth. DNA preparation from the FPE broth was simplified and the cocktail PCR was designed to finish within 30 min. The PCR amplicon was visually differentiated using dipstick DNA chromatography within 5-10 min.

Bacterial Strains.
The bacterial strains of S. enterica, Shigella spp., EIEC, and EHEC and other strains are listed in Table 1. All strains were supplied from the Gifu Type Culture Collection of the Microbial Genetic Resource Stock Center, Gifu University Graduate School of Medicine (Gifu, Japan), supported by the Ministry of Education, Culture, Sports, Science, and Technology of Japan. All strains were cultured on heart infusion agars (BD, Tokyo, Japan) at 37 ∘ C under an aerobic atmosphere overnight. A fresh culture was used for each experiment.

Determination of Optimal Enrichment Culture with FPE
Broth. FPE broth is designed to support the growth of Campylobacter species without blood and carbon dioxide [14]. This broth was used in the present study to simplify the total food analysis protocol because the FPE broth supported most food borne pathogens in our preliminary experiment. The growth of the foodborne pathogen in the FPE broth was compared with that in conventional selective enrichment broth ( Figure 1). In the spiked ground meat experiment, diluted fresh bacterial solution and 25 g of beef were mixed in 225 mL of FPE broth and incubated at 37 ∘ C.

DNA Extraction.
DNA was extracted from 1 mL of culture broth using a physical disruption method (MORA-EXTRACT, AMR, Gifu, Japan) according to the manufacturer's instructions, with a final DNA elution volume of 200-400 L.
DNA extraction from FPE broth was performed using a simplified protocol. One milliliter of 2 to 18 hours enrichment culture was collected in a 2 mL Eppendorf tube and centrifuged at 12,000 g. The supernatant was completely removed and 1 mL of T10E1 buffer was added and centrifuged under the same conditions. After the complete removal of the supernatant, 200 L of T10E1 buffer was added to the tube and mixed. The solution was transferred to a tube containing beads and physically disrupted for 1 min with a Disrupter Genie (Scientific Industry Inc., Bohemia, NY, USA). The tube was boiled at 100 ∘ C for 3 min. Five microliters of the solution were used for the cocktail PCR assay.

Cocktail PCR.
The cocktail PCR conditions used in the present study are described below. The primers used are described in Table 2. PCR amplification was performed in 10 L of reaction mixture containing 5 L of 2× premix Ex Taq (Takara Bio, Shiga, Japan), 2.5 L of primer mixture, 0.5 L of distilled water, and 2 L of DNA template. PCR was carried out using the QuickBath thermal cycler (ThermoGen Ltd., Nagano, Japan) under the following conditions: 95 ∘ C for 3 min, 40 cycles of 95 ∘ C for 10 s, and 65 ∘ C for 10 s. The PCR cycles finished within 30 min.
2.5. Dipstick DNA Chromatography. The 5 terminus of the cocktail PCR amplicon was labeled with biotin, and the 3 terminus was labeled with four different tags. Streptavidincoated blue latex, kindly provided from Fujikura (Saitama, Japan), bound the biotinylated 5 terminal amplicon and the tagged 3 terminus was bound on the antitag lines printed on the DNA strip ( Figure 2).
After PCR, 30 L of streptavidin-coated blue latex solution was added to the PCR tube, and then the DNA strip was inserted into the PCR tube. The hybridized PCR amplicon was visualized in 5 to 10 min as a blue line, which represented the bound streptavidin-coated blue latex and biotin-labeled 5 terminus of the PCR amplicon.

A Protocol to Detect Multiple Foodborne Pathogens after 6 Hours Enrichment Culture of Ground
Meat. For detection of foodborne pathogens with our protocol, 25 g of beef and 225 mL of FPE broth were homogenized in Stomacher bags (Eiken Chemical Co., Tokyo, Japan). The entire homogenate was transferred to a culture bottle and incubated with shaking at 37 ∘ C. After incubation, 1 mL of the supernatant was collected, and DNA was extracted using the physical disruption method described above. Subsequently, 5 L of the extracted DNA was analyzed by cocktail PCR primers ( Table 2) using premix Ex Taq (Takara Bio, Shiga, Japan) and the QuickBath thermal cycler. After 30 min, the PCR amplicon was analyzed by dipstick DNA chromatography. Thirty microliters of streptavidin-coated blue latex solution were added to each tube. Subsequently, the DNA strip was inserted into each tube. After 5-10 min at room temperature, the amplicon bound on the appropriate line on the dipstick surface (Figure 2) was visualized by the blue latex of the biotin-labeled 5 terminus of the PCR amplicon.

Sensitivity and Specificity of the Cocktail PCR Dipstick DNA Chromatography (CPDC) Assay.
To measure the sensitivity of the CPDC assay, purified chromosomal DNA of E. coli O157 GTC14510 and S. enterica serovar Enteritidis GTC03838 were prepared at six different concentrations (2 ng, 200 pg, 20 pg, 2 pg, 200 fg, and 20 fg) and assayed (Table 3). Another sensitivity assay starting from quantitatively diluted culture supernatants was also performed ( Table 4). The specificity of the CPDC assay was determined using the 176 strains listed in Table 5.

Evaluation of CPDC Assay with Spiked Ground Meat.
The CPDC assay after enrichment culture in FPE broth was compared with a commercial immunochromatography system (Wako Pure Chem. Industries, Ltd., Osaka, Japan). Ground beef collected from local supermarkets was immediately transported to our laboratory in an insulated cooler box at 4 ∘ C. However, the isolation frequency of the target Shigella spp., Salmonella enterica, and E. coli O157 : H7 was less than 0.1% by culture-based conventional methods in our preliminary experiments. We decided, therefore, to evaluate the CPDC assay with spiked ground meat experiments. Ground meat was collected from a supermarket and confirmed target to be pathogen free by conventional methods. One milliliter of a mixed culture containing Shigella dysenteriae, S. enterica subspecies enterica serovar Enteritidis, and E. coli O157:H7 at three different concentrations were mixed with 25 g of ground meat, and 225 mL of FPE broth was then added. The total volume was incubated at 37 ∘ C. Immediately, and 4 hours, 6 hours, 8 hours, and 18 hours, 1 mL of the enrichment was used for extraction and then used in the CPDC assay. Another aliquot was used for the commercial immunochromatography kit for S. enterica serovar Enteritidis and E. coli O157.

Results and Discussion
Conventional culture has been the "gold standard" method for the detection of enteric bacterial pathogens. The advantages of this method include identification, facilitation of outbreak management, and generation of an antimicrobial susceptibility profile [15]. However, this conventional method has many disadvantages. Many different enrichment broths and solid media are used to screen for all possible foodborne Salmonella enterica serovar Enteritidis GTC03838 pathogens, and time-consuming protocols are prepared to generate a result. FPE broth is designed to support Campylobacter without adding lysed blood and carbon dioxide. Campylobacter, however, is a slow-growing organism and needs 24 hours to reach 10 4 CFU/mL. Therefore, addition of selective antibiotics to the FPE broth was essential to suppress contaminating other bacteria for 24 hours enrichment culture. FPE broth could also support the growth of Listeria without adding blood, but the growth of Listeria is also slow, needing 24 hours to reach 10 4 CFU/mL (unpublished data). The growth of pathogens in conventional enrichment culture and FPE broths were measured (Figure 1). Approximately 1-10 bacteria were spiked in 225 mL of enrichment broth. S. enterica and E. coli reached 10 4 CFU/mL after 6-hours incubation in FPE broth, buffered peptone water broth, and mEC broth, as shown in Figure 1. V. parahaemolyticus reached 10 4 CFU/mL after 4-hours incubation in both FPE and alkaline peptone broths. Based on these data, we selected 6 hours incubation for the CPDC assay.
In the present study, cocktail PCR was capable of simultaneously determining the presence of Salmonella spp., Shigella spp., EIEC, and EHEC by targeting invA, ipaH, stx1, and stx2 genes ( Figure 2).
To evaluate the detection limit of the CPDC assay for each pathogen, 2 ng to 20 fg of DNA per reaction was prepared. The sensitivity and specificity of this assay are shown in Tables 3-5. The detection limit was 200 fg for each pathogen per CPDC assay ( Table 3). The presence of the products with expected sizes was also confirmed by agarose gel electrophoresis, and nonspecific products were not observed (data not shown). The specificity of this CPDC assay was evaluated using 157 target strains (45 strains of Shigella spp., 54 strains of EHEC, 17 strains of EIEC, and 41 strains of Salmonella spp.) and 19 nontarget strains shown in Table 5. The detection limit of Salmonella and Escherichia from FPE culture supernatant (Table 4) was 10 3 CFU/mL, while the commercial immunochromatography kit required 10 5 CFU/mL. No false positive lines appeared on the dipstick DNA chromatography for any of the nontarget strains.
Immunochromatography is a simple technology to detect antigen in culture supernatant. However, immunochromatography targeting EHEC serotypes is not useful for testing food, because many kinds of E. coli serotypes produce shiga toxins. Thus, it is practically difficult to cover all of the EHEC serotypes by immunochromatography. Our method targeted both shiga toxin 1 and shiga toxin 2 toxins and detected non-O157 shiga toxin-producing serotypes (O26, O111, and O121). Three serotype (O45, O103, and O145) strains were not used because they are not available from our collection.
Shiga toxin 1 and shiga toxin 2 PCR products were designed to be bound on line 1 of the dipstick (Figure 2). The two genes were not equally amplified. The signal of the shiga toxin 2 amplicon was always bigger than the signal of the shiga toxin 1 amplicon.
The invasion plasmid antigen H (ipaH) gene is often used to diagnosis dysentery [16], because ipaH is carried by all four Shigella species as well as EIEC. In our cocktail primer, therefore, we selected the ipaH-specific primer to detect both Shigella and EIEC. The ipaH amplicon was designed to react on the second line of the dipstick chromatography strip, as shown in Figure 2. The CPDC assay was found to be effective at detecting Shigella species and EIEC from 4 to 6 hours FPE broth.
The CPDC assay required 4 to 6 hours FPE broth for the detection of S. enterica serovar Enteritidis. The detection limit of chromatography for Salmonella was, however, 10 5 CFU/mL in culture supernatant. To reach this cell number in FPE broth, incubation for more than 6 hours was necessary ( Table 6). There is another disadvantage to using immunochromatography. The commercially available immunochromatography kits for Salmonella serovars are limited. The products only detect serovar Enteritidis. Thus, it is difficult to screen many different Salmonella serotypes simultaneously, such as serovar Typhimurium, serovar Choleraesuis, serovar Dublin, serovar Typhi, and others.
The CPDC assay for Salmonella was evaluated by spiked ground meat experiments with three different inoculation levels from 2.5 to 10 CFU/25 g ( Table 6). The CPDC assay detected target pathogens on the third line of a dipstick DNA strip from 4 to 6 hours culture with FPE broth. On the other hand, the commercial immunochromatography kit only detected antigens from 18 hours enrichment culture because the method requires 10 5 CFU/mL of organism (Tables 4 and  6).
Multiplex PCR to detect many Salmonella serovars has been reported [17,18]; however, we selected the invA gene for our assay because all Salmonella serovars carry this gene [19].
Internal amplification control (IAC) is designed to bind to line 4 of the dipstick chromatography strip to check for the presence of PCR inhibitor and false negatives [20]. A general guideline proposed for PCR testing of foodborne pathogens also requires the presence of IAC in the reaction mixture [21].
Systematic review of clinical implications, public health considerations, and the cost effectiveness of rapid diagnostic tests for detection and identification of bacterial pathogens in feces and food suggests that adoption of rapid test methods, especially for PCR, in combination with a routine culture is unlikely to be cost-effective [7,22]. However, as the cost of rapid technologies decreases, total replacement with rapid technologies may be feasible.
The clinical impact of the decreased turnaround time means that bacterial diarrhea is more promptly ruled out using the CPDC assay compared to using conventional culture in small laboratories. This reduces the expenditure of infection control resources and, in particular, in cases of sporadic diarrhea, helps to reduce the requirement for scarce isolation rooms. In addition, the earlier availability of results is helpful in community-based management of outbreaks.
For detection of Salmonella spp., Shigella spp., EIEC, and EHEC, the overall time to confirm a positive result by conventional culture plus immunochromatography is at least two working days. Generating a negative report requires 48 hours. In contrast, a report can be generated for the CPDC assay within one working day. One advantage of an early laboratory report is early judgment of contamination, which can prevent food poisoning and additional outbreaks. The method also contributes to the quick shipment of fresh food products to the markets.

Conclusion
Cocktail PCR targeting multiple foodborne pathogens and simple dipstick DNA chromatography to differentiate the PCR product was designed. The method was applied to detect pathogens in ground meat after 6 hours enrichment broth, which supports the growth of broad range foodborne pathogens. This single tube PCR and enrichment method help to simplify food analysis protocol. As a result, the method offers rapid report to food suppliers and helps the quick shipment of safety-confirmed food products to markets.