A quantitative prevalence of Escherichia coli O157 in different food samples using real‐time qPCR method

Abstract Escherichia coli serogroup O157 is the main causative agent of several intestinal and extra‐intestinal foodborne diseases in humans through consumption of low‐dose contaminated foods such as milk, beef, and vegetables. To date, studies regarding the quantitative prevalence of E. coli O157 in foods are so limited. Therefore, this study aimed to evaluate the quantitative prevalence rate of E. coli serogroup O157 in raw milk (n = 144), vegetable salad (n = 174), and minced beef samples (n = 108) using the real‐time qPCR SYBR green melting curve method targeting the rfbA gene. First, we evaluated the method and found a sensitive and specific qPCR assay with 1 log of CFU/ml detection limit to detect E. coli O157 (Tm = 80.3 ± 0.1°C). About 2.77%, 10.18%, and 9.19% of raw milk, minced beef, and vegetable salad samples, respectively, were contaminated with E. coli O157. Minced beef and vegetable salad samples were significantly more contaminated than raw milk samples. Population average of E. coli O157 in raw milk, minced beef, and vegetable salad samples were 2.22 ± 0.57, 3.30 ± 0.40, and 1.65 ± 0.44 log CFU/ml or gr, respectively. Significantly higher levels of population of E. coli O157 were observed in minced beef samples. Minced beef can be regarded as the main food in the transmission of this foodborne pathogen. Routine quantitative rapid monitoring is strongly suggested to be carried out to prevent foodborne diseases caused by E. coli O157.

million foodborne cases, 320,000 hospitalizations, and 5000 deaths have been reported annually in the United States (Mead et al., 1999).
The World Health Organization declared that more than 1.5 million people died from the acute diarrheal diseases caused by different foodborne pathogens around the world (Pires et al., 2021). The most common foodborne bacterial pathogens are pathogenic Escherichia coli, Salmonella spp., Campylobacter spp., Staphylococcus aureus, Clostridium perfringens, Listeria monocytogenes, Shigella spp., and Cronobacter sakazakii (Gourama, 2020).
Escherichia coli is a rod-shaped, Gram-negative, facultative anaerobe, nonsporulating bacterium belonging to the Enterobacteriaceae family, which is part of the normal gastrointestinal microbiota of humans and animals. It is also commonly found in abundance in soil and water (Denamur et al., 2021). Different serotypes of E. coli have been identified and isolated from clinical, food, water, and environmental samples. Escherichia coli O157 has been known as the most common serotype of pathogenic E. coli strains identified as enterohemorrhagic and Shiga-toxin producing bacterial foodborne pathogens (Rani et al., 2021). In 1982, the first outbreaks caused by E. coli O157 were reported in Michigan and Oregon states, USA.
The organism was isolated from patients with abdominal cramp and severe bloody diarrhea after consumption of hamburgers in local restaurants (Lim et al., 2010;Pennington, 2010). Escherichia coli serotype O157: H7 is now recognized as the main causative agent of hemolytic uremic syndrome, hemorrhagic colitis, permanent organ failure, and thrombotic thrombocytopenic purpura in human of all ages (Rangel et al., 2005). Escherichia coli serogroup O157 accounts for more than 20% of all foodborne diseases around the world (Tack et al., 2021). More than 350 outbreaks caused by E. coli O157: H7 occurred between 1982 and 2006 in USA (Manning et al., 2008). More than 52% of these outbreaks linked with food consumption, 9% with consumption of contaminated water, and other ones were related with person to person and animal contacts. It is interesting to note that most of these outbreaks were associated with the consumption of contaminated unpasteurized milk, salads, and poorly cooked meat products. Consequently, it has been regarded as one of the most important foodborne bacterial pathogens (Heiman et al., 2015;Kim et al., 2020;Tack et al., 2021).
Hospitalization and fatality rates of E. coli O157: H7 are significantly higher than that of other E. coli serogroups and foodborne pathogens including Campylobacter and Salmonella. The infectious dose of this foodborne pathogen is 10-100 bacterial cells, leading to different intestinal or extra-intestinal diseases, which might progress into the stages that threaten human life (Rani et al., 2021;Singha et al., 2022). Several strategies have previously been investigated to control and prevent the spread of E. coli O157 as recommended by center for disease control (CDC) (Singha et al., 2022). Quantitative prevalence evaluation of different food samples has been reported as one of the most efficient strategies to control and prevent any outbreak caused by this pathogen (Laidlaw et al., 2019). Limited studies have been conducted on quantitative prevalence of E. coli O157 serogroup in different food samples in Iran or other parts of the world (Pakbin et al., 2020). Therefore, the aim of this study was to investigate the quantitative prevalence of E. coli O157 serogroup, by using real-time qPCR of different food samples collected from different areas of Tehran city, Iran. with ice packs to the central microbiology laboratory for total DNA extraction and further molecular analysis.
All strains were purchased from Pasteur Institute (Pasteur In.,).
Reference strains were grown in tryptic soy broth (TSB, Merck,) and cultured at 37°C for 24 h. Ten-fold dilution series of the reference strains (positive and negative controls) were prepared to design and develop the real-time PCR assay in this study. Serially diluted cultures were subjected to total DNA extraction.

| DNA extraction
Food samples preparation was performed according to the method previously described by Fukushima et al. (2007) and Liu et al. (2019) (Fukushima et al., 2007;Liu et al., 2019). Prepared food samples were subjected to total genome extraction by using the SinaClon commercial total DNA extraction kit (SinaClon Co.,) according to the manufacturer's instructions. The quality and quantity of the extracted total genomes were evaluated using the NanoDrop-1000 spectrophotometer (ThermoFisher,). Concentrations of the all extracted DNA were adjusted to 50 ng. μL −1 before real-time PCR assay. DNA templates were kept at −20°C until the further analysis.
Real-time PCR tubes contained 10 μl of the 2X real-time PCR master mix, 1 μl of each primer (20 μM), 1 μl of the DNA template (50 ng. μL −1 ), and sterilized nuclease free water up to a final reaction volume of 20 μl. The amplification procedure of real-time PCR was performed using the following conditions: initial denaturation step at 94°C for 5 min, followed by 40 cycles of denaturation at 94°C for 40 s, annealing at 60°C for 40 s, and extension at 72°C for 30 s, followed by a melting step with the increasing temperature from 70 to 95°C with the raising temperature rate of 0.2°C/s. The size of amplicons produced by rfbA primers was 242 bp. The analysis of amplification and melting curves was carried out using the Rotor-Gene 6000 software version 2.3.5 (Qiagen Corbett,).

| Statistical analysis
One-way analysis of variance (ANOVA) and Chi-square tests were used to evaluate the significant differences (p < .05) between the contamination rates and the average microbial populations, respectively, using the SPSS software version 23.0.1. All experimental and statistical measurements were performed in triplicate.

| RE SULTS
In this study, we have used qPCR to investigate the quantitative prevalence of E. coli serogroup O157 in different food samples. First of all, we developed and optimized the qPCR method to quantify the E. coli O157 counts. We detected the rfbA gene in E. coli O157 strain for identification of this foodborne pathogen through real-time PCR assay at the present study. Figure 1 represents the melting curves of PCR amplicons of decimally diluted pure broth culture of E. coli O157 DNA templates to determine the limit of detection of the method.
The average melting temperature (T m ) of rfbA amplicon generated in real-time PCR was 80.3 ± 0.1 °C. No non-O157 E. coli strains was identified by using the developed method. The limit of detection for rfbA primer set using DNA extracted from serially diluted pure broth culture of E. coli serogroup O157 was found to be 10 CFU/ml (1 log of CFU/ml) with the associated C t value. After plotting the standard curve, the amplification efficiency was calculated 97.8%. Standard curve of the real-time PCR for quantification of E. coli O157 count generated by plotting the C t value of reactions against the E. coli O157 log of the CFU/ml was demonstrated in Figure 2. After recovering E. coli serogroup O157, the negative and positive results of detection were confirmed by the developed real-time PCR method.
As shown in Figure 2, the detection limit of the reference strain also was 1 log. The regression value of the standard curve equation Figure 2 showed a significant relationship between the C t values and bacterial counts in the developed qPCR method. We found real-time qPCR method, targeting the rfbA gene, a sensitive and specific assay to investigate the quantitative prevalence of E. coli O157 in food samples.
The prevalence rate of E. coli O157 in different food samples including minced beef, vegetable salad, and raw milk was evaluated by using a sensitive and specific real-time qPCR method which was developed in this study. Figure 3 illustrates the prevalence of E. coli serogroup O157 in raw milk, minced beef, and vegetable salad samples. About 2.77% (4 out of 144), 10.18% (11 out of 108), and 9.19% (16 out of 174) of raw milk, minced beef, and vegetable salad samples, respectively, F I G U R E 1 Melting curves of real-time PCR using a 10-fold dilution series of E. coli serogroup O157 were contaminated with E. coli serogroup O157 in this study. In total, E. coli O157 was detected in 7.27% (31 out of 426) of food samples.
Escherichia coli O157 was also detected significantly (p < .05) in higher levels in minced beef and vegetable salad samples than that in raw milk samples. Populations of E. coli O157 serogroup in different E. coli O157-positive food samples are shown in Table 1. As it can be seen in Table 1, average population of E. coli O157 in raw milk, minced beef, and vegetable salad samples were observed 2.22 ± 0.57, 3.30 ± 0.40, and 1.65 ± 0.44 log CFU/ml or gr, respectively. The average population of E. coli O157 in minced beef samples was significantly (p < .05) higher than that observed in raw milk and vegetable salad samples. Also, the concentrations of E. coli O157 varied considerably among different food samples, which ranged from 1.5 to 2.9 log CFU/ml, from 2.7 to 4.0 log CFU/mg, and from 1.1 to 3.0 log CFU/mg in raw milk, minced beef, and vegetable salad samples, respectively.

| DISCUSS ION
Escherichia coli O157 is the most common serogroup member of pathogenic E. coli strains identified differently as Shiga-toxin producing, verotoxin producing, and enterohemorrhagic bacterial pathogen. Escherichia coli O157 is recognized as particularly as virulent foodborne bacterial pathogen which only requires a small number of bacterial cells to cause disease in the host (Pakbin et al., 2021). Also, low levels of contamination have been detected in the infected foods, which were consumed and caused to several food poisoning cases (Franz et al., 2019). This foodborne pathogen caused different intestinal and extra-intestinal illnesses such as acute bloody diarrhea and HUS in human (Kim et al., 2020).
Cattle have been recognized mainly as the most important reservoir of E. coli serogroup O157 causing zoonotic diseases through consumption of various foods contaminated with bovine feces (e.g., vegetables contaminated with animal-based agricultural fertilizers), undercooked meat, and unpasteurized dairy products in humans (Gutema et al., 2021). However, this bacterial pathogen has also been isolated from naturally acquired infections from the consumption of other animal species such as sheep, deer, goat, moose, chicken, turkey, among others. Thus, E. coli O157 has been considered as one of the main challenges and major concerns in public health and food safety in developing and developed countries around the world (Abreham et al., 2019). Molecular assays such as polymerase chain reaction (PCR)-based methods are considerably more sensitive. However, these methods are relatively expensive and require instrumentation. qPCR is a specific molecular PCR-based assay commonly used to quantify different bacterial strains (DNA) in food samples .
qPCR method provides faster, safer, cheaper, and more practical strategy than conventional methods for diagnosis and quantification of different foodborne pathogens in food sample, and in the recent decades, this method is strongly appreciated by clinicians and would aid their suitable treatment of their patients with foodborne diseases (Bustin, 2010;Kubista et al., 2006;Yang & Rothman, 2004). Also, the simplicity and high-throughput adaptability of the method would enable rapid diagnostics of spoiled food improving food safety over the entire food production and processing chain in developing countries. Moreover, in a clinical setting, it would enable a rapid treatment response that does safe life particularly when patients have to travel from remote areas to seek treatment and where on presentation toxic symptoms and systemic infection are often in an advanced state (Chen et al., 2021). However, the main limitation of this method is its inability to differentiate live from dead cell DNA . qPCR assays were used to detect and quantify E. coli serogroup O157 strains in different clinical, environmental, and food samples (Kim & Oh, 2021). In this study, we also used real-time qPCR SYBR green melting curves, targeting the rfbA gene, to determine the prevalence rate and quantify E. coli O157 in food samples. We found this method very specific and sensitive for the detection of E. coli O157 in food samples (detection limit of the developed method was  (Azinheiro et al., 2020). Kim et al. (2020) designed and optimized qPCR method to detect E. coli O157 and Salmonella typhimurium in cabbage and iceberg without any pre-enrichment step. They demonstrated that this method was able to detect 7 CFU of E. coli O157 in 25 g of iceberg and cabbage samples (Kim & Oh, 2021). Several other researchers also declared that qPCR SYBR green melting curve analysis method is sufficiently sensitive and specific with an excellent limit of detection to detect E. coli O157 in food samples in single and multiplex reactions (Barbau-Piednoir et al., 2018). (3 out of 30), and 3.7% (1 out of 27), respectively. They reported that 13.8% of all beef samples were contaminated with this foodborne pathogen, which was significantly higher than that we reported in this study (Sallam et al., 2013). Recently, Ahmad et al. (2021) (Furukawa et al., 2018). Cooking sufficient and safe and hygienic food handling and distribution procedures are strongly suggested to be used to prevent infection and gastroenteritis outbreaks caused by E. coli O157 via consumption of contaminated foods, especially minced meat products (Burrus et al., 2017;Rangel et al., 2005;Rani et al., 2021).

| CON CLUS ION
In conclusion, we showed that real-time qPCR SYBR green melting curve analysis is a sensitive and specific assay with a low detection limit for the detection and identification of E. coli serogroup O157 in food samples. Also, we found that minced beef and vegetable salad samples were significantly more contaminated with E. coil O157 than raw milk samples. The population average of E. coli serogroup O157 was observed in significantly higher levels in minced beef samples than that in raw milk and vegetable salad samples. Sufficient thermal treatments, hygienic practices, and routine quantitative rapid monitoring are strongly recommended to be implemented to prevent foodborne diseases and outbreaks caused by E. coli O157 through consumption of contaminated foods.

We appreciate our colleagues and technicians in Medical
Microbiology Research Center, Qazvin University of Medical Sciences, who assisted us in this research project.

CO N FLI C T O F I NTE R E S T
All the authors have declared no conflict of interest.

DATA AVA I L A B I L I T Y S TAT E M E N T
We confirm that all data and findings of this study are available within the article.