Introduction

Escherichia coli O157:H7 is an enterohemorrhagic E. coli (EHEC) currently recognized as one of the most significant foodborne pathogens globally [1,2]. This microorganism is the etiological agent of a range of illnesses, from mild gastroenteritis to severe diseases such as hemolytic colitis (HC) and hemolytic uremic syndrome (HUS) [6]. E. coli O157:H7 has been responsible for numerous foodborne outbreaks [3], for example a recent multi-state outbreak occurred in the United States affecting 83 people, resulting in 27 hospitalizations, 2 cases of HUS, and 1 fatality [4]. The pathogenicity of E. coli O157:H7 is linked to several virulence factors, with the eae gene and Shiga toxins (Stx1 and Stx2) being of particular importance. The eae gene facilitates the attachment of the microorganism to host intestinal cells, while Stx1 and Stx2 disrupt protein synthesis, contributing to the severity of HC and HUS [5, 6]. The presence of these virulence factors is crucial for the development of diseases, with the eae gene playing an essential role in bacterial adhesion to the intestinal lining, while Stx1 and Stx2 are directly involved in the pathogenic process and aggressivity of symptoms [7]. Given its low infectious dose and concentration in foods, the detection of E. coli O157:H7 demands selective, sensitive, and accurate methods. Furthermore, in food production industries, the ability to obtain rapid results is a critical feature of detection methods, enabling quick decision-making and contributing to food safety. Currently, there are several methods available for detecting E. coli O157:H7 in food, but many of these are complex, costly, and time-consuming. The most commonly used protocols involve plate culture, polymerase chain reaction (PCR), and enzyme-linked immunosorbent assays (ELISA), all of which require an overnight enrichment step before pathogen detection. Traditional plate culture is particularly laborious, taking between 2 to 7 days to complete, because it involves multiple steps such as pre-enrichment, selective enrichment, plating on agar media, biochemical testing, and serological confirmation. While PCR offers faster results, it can produce false positives due to the detection of genes associated with virulence factors that may not be exclusive to E. coli O157:H7. Moreover, PCR methods are unable to differentiate between viable and non-viable cells. ELISA assays also typically require additional enrichment or purification steps, further prolonging analysis time. Depending on the method, both PCR and ELISA can take anywhere from 5 to 25 hours to provide results, with certified methods commonly used in large food companies taking approximately 24 hours to generate results [1, 8].

In recent years, the use of bacteriophages for pathogen detection has gained significant attention due to their rapid, sensitive, and cost-effective nature [9 –15]. Bacteriophages, viruses that specifically infect bacteria, offer several advantages when detecting foodborne pathogens. Notably, bacteriophages are safe for humans, as they do not infect human cells. They exhibit high specificity, targeting only certain bacterial genus, species or serotypes, which enhances detection accuracy. Upon infecting their host bacteria, phages replicate rapidly, producing large quantities of new phages, which may result in bacterial cell lyses. Importantly, phages can only replicate in viable bacterial cells, providing a reliable method for identifying live pathogens and reducing the likelihood of false positives [16,17]. The PhageDx E. coli O157 Assay is an innovative detection method that uses a recombinant bacteriophage expressing luciferase to detect E. coli O157:H7. Upon infecting the target bacterium, the phage expresses luciferase, making it possible the emission of light when exposed to a specific substrate. The emitted light can then be detected by a luminometer, providing a rapid indication of the presence or absence of E. coli O157:H7 in the food samples [18]. This method has a short enrichment step (5 hours), which aroused the interest of our research group in testing the assay with E. coli O157:H7 isolated in Brazil. Although E. coli O157:H7 was not previously considered a significant risk in Brazilian food, recent studies have increasingly reported its presence in food samples from the country [19 – 22].  This raises concerns about whether the pathogen has recently entered the country [23] or whether earlier detection methods lacked sufficient sensitivity to identify circulating strains of E. coli O157:H7. Epidemiological data show that illnesses caused by E. coli O157:H7 have been linked to the consumption of undercooked ground beef, raw vegetables, milk, and contaminated water [6]. Brazilian production of all these foods is very significant and important, especially beef, as Brazil is the world's largest exporter. The objective of this study is to evaluate the performance of PhageDx recombinant bacteriophages for the detection of E. coli O157:H7 in vitro and in various food matrices.

Materials and Methods

PhageDx E. coli O157:H7 Assay

The PhageDx E. coli O157:H7 assay (PhageDx), developed by Laboratory Corporation of America Holdings (Labcorp), was provided as a donation to the Federal University of Rio Grande do Sul (UFRGS) after a scientific agreement was established. The assays were shipped from Saint Paul, Minnesota, USA to Brazil in refrigerated containers (<7°C). The assay utilizes a genetically engineered bacteriophage capable of specifically infecting E. coli O157:H7 in food samples. The procedure begins by incubating 25g or 25mL of food samples in 225mL of Tryptic Soy Broth (TSB) medium for 5 hours. Following this incubation, bacteriophages are added to the samples. If E. coli O157:H7 is present, the bacteriophage infects the bacterial cells and expresses the luciferase enzyme. After adding the substrate, the enzyme produces a bioluminescent signal, which is detected using a GloMax® Navigator Luminometer (Promega Corp., Madison, WI, USA). The detailed procedure is described below in section 2.3. The assay reads above 150 relative light units (RLU) to indicate the presence of E. coli O157:H7 in the sample [24]. Each sample was measured in duplicate with a 3-minute delay and 1-second integration period. The threshold of 150 RLU and the measurement protocol were defined by the assay manufacturer and applied as the standard in this study.

Experimental Design

PhageDx was evaluated in vitro for its ability to detect E. coli O157:H7 strains isolated from various sources, including lettuce irrigation water, bovine carcasses, bovine hides, and positive control strains from different laboratories in Brazil. The strains used are detailed in Table 1. Each sample was measured in duplicate, and the average RLU of these readings was recorded as results. This procedure was repeated in two independent tests for each strain, resulting in four values per strain to ensure accuracy and reproducibility. To determine the detection limit of PhageDx, E. coli O157:H7 samples were inoculated at three different concentrations: low (1 CFU), medium (10 CFU), and high (100 CFU), separatelly, into 10 wells of a microtiter plate. Negative controls, containing only TSB, phages, and detection reagents, were also tested. Each well was measured twice with the luminometer, and the average reading was used as the final value per well. This process was conducted in three independent assays, yielding 30 data points per inoculum level (1, 10 and 100 CFU/well), ensuring a robust assessment of the assay’s sensitivity. The sensitivity of the PhageDx assay was further evaluated with artificially contaminated food samples, including ground beef, pasteurized whole milk, organic curly lettuce, and non-carbonated mineral water. For the low (1 CFU) and medium (10 CFU) inocula, 10 wells were tested per microplate, while for the high inoculum (100 CFU), 4 wells were tested. Each well was measured twice with the luminometer, and the average RLU was recorded. Three independent tests were performed for each food matrix, resulting in 30 measurements for the low and medium inocula, and 12 for the high inoculum level, totaling 288 measurements across the food matrices.

* LMCA - ICTA: Microbiology and Food Control Laboratory (ICTA/UFRGS);

**INCQS 00171/FIOCRUZ: Fundação Oswaldo Cruz;

 *** USP: Universidade de São Paulo;

Table 1: E. coli O157:H7 strains isolated from food and food production environments tested at concentration 10⁸ CFU/mL by the PhageDx E. coli O157:H7 Assay. Results are expressed as Relative Light Units (RLU).

PhageDx Ability to Detect E. coli O157:H7 Strains

Each E. coli O157:H7 strain was cultured overnight in 5 mL of TSB at 37°C. The cell concentration was standardized to 10⁸ CFU/mL using a microplate reader (Loccus LMR 96, Brazil). To confirm the bacterial concentration, samples were plated on Tryptic Soy Agar (TSA) and incubated for 24 hours at 37°C. The strains were then added to a 96-well plate, with 100 µL of each sample and 10 µL of phage solution. After 2 hours of incubation at 37°C, 65 µL of a solution containing 50 µL NanoGlo buffer, 1 µL NanoGlo substrate (Nano-Glo® Luciferase Assay System, Promega Corp., Madison, WI, USA), and 15 µL Renilla lysis buffer (Renilla Luciferase Assay System, Promega Corp., Madison, WI, USA) was added to each well. Luminance was measured using a luminometer. Experiments were performed in duplicate.

PhageDx In Vitro Detection Limit

To assess the detection limit, a cocktail of five E. coli O157:H7 strains (USP, CC, A1P2, FIO and A8P1) was prepared. After standardization to 10⁸ CFU/mL, the strains were mixed and serially diluted to concentrations of 10, 100, and 1000 CFU/mL. Each dilution (100 µL) was added to the wells of a 96-well plate, resulting in three inoculum levels: low (1 CFU/well), medium (10 CFU/well), and high (100 CFU/well). A total of 30 samples were evaluated at each inoculum level. The bacterial concentration was confirmed using the procedure described in section 2.3.

PhageDx Sensitivity for Detection of E. coli O157:H7 in Artificially Contaminated Foods

The bacterial cocktail used for contaminating food samples was prepared as described in section 2.4, with an additional centrifugation step (2253 g for 10 min at 4°C) to pellet the bacteria, followed by three washes with 0.1% (w/v) peptone water. The cocktail was adjusted to concentrations of 1, 10 and 100 CFU/mL using peptone water. Food samples (ground beef, pasteurized whole milk, non-carbonated mineral water, and organic curly lettuce) were purchased from a local supermarket, transported in a refrigerated cooler (<7°C), and confirmed to be negative for E. coli O157:H7 by pre-incubation in modified Tryptic Soy Broth (mTSB) and analysis on Sorbitol MacConkey agar. The food samples (25g or 25mL) were placed in sterile Whirl-Pak® filter bags, and 75 mL of mTSB was added. Each sample was inoculated with 1 mL of bacterial cocktail at concentrations of 1, 10 and 100 CFU/mL. After 5 hours of incubation at 41°C, bioluminescence was measured as described in section 2.3.

Comparison of PhageDx and Real-Time PCR

Samples of 100 cattle carcasses (325g each) were collected from Brazilian slaughterhouses and incubated in 975 mL of STEC Enrichment Broth (SEB) at 41.5°C for 10 to 20 hours. After incubation, the samples were analyzed for the presence of Shiga-toxin E. coli (STEC) using real-time PCR iQ-Check STEC VirX kit (Bio-Rad). Concurrently, 1 mL of SEB broth was sent to the Microbiology and Food Control Laboratory (LMCA -ICTA/UFRGS) for analysis using the PhageDx assay. Upon arrival, the samples were thawed, homogenized, and 150 µL of each sample was added to a 96-well plate. Phage solution (10 µL) was added to each well, followed by a 2-hour incubation at 37°C. The bioluminescence assay was performed as described in section 2.3. All experiments were conducted in duplicate.

Results

All E. coli O157:H7 strains tested (Table 1) were detected by the PhageDx. The strains generated RLU values between 10⁶ – 10⁷. The non-O157:H7 E. coli strain (ATCC 25922) used as a negative control had an average RLU of 1.07 X10², which was below the lower limit of the method (150 RLU). Table 2 shows the percentage of wells out of a total of 30 wells for each inoculum level (low, medium and high) that were positive after incubation for 2 hours following the addition of bacteriophage solution. As expected, the percentage of detection varied with the inoculum concentration, with higher percentages of positive samples observed as the inoculum level increased. The detection limit of the PhageDx in different food matrices is demonstrated in Table 3. In summary, the assay was able to detect E. coli O157:H7 in all food samples contaminated with 1, 10 and 100 CFU/25g or mL after 5 hours of enrichment. Similar to the in vitro sensitivity test results, higher numbers of RLU were observed, when the number of cells inoculated in the foods increased. The highest RLU values were observed when 100 CFU were inoculated in lettuce (8.20x10⁶) and mineral water (1.88x10⁶), followed by ground beef (9.15x10⁵) and pasteurized milk (2.60x10⁵). The lowest RLU values were obtained when 1 CFU/25g was inoculated in pasteurized milk (4.75 x10³) and ground beef (7.68 x10³).

Table 2: Evaluation of the detection limit of the PhageDx E. coli O157:H7 Assay for cocktail composed for E. coli O157:H7 strains. The percentage indicates the percent positive samples at each concentration.

Table 3: PhageDx E. coli O157:H7 Assay sensitivity for E. coli O157:H7 detection on food after 5 h of incubation.

The negative controls, i.e., samples that were not inoculated, had even lower RLU values. The average RLUs found in these samples were 2.69 x 10¹ (mineral water), 6.00 x 10¹ (lettuce), 8.56 x 10¹ (milk) and 8.68 x 10¹ (ground beef). All 100 bovine carcasses tested negative for the presence of E. coli O157:H7 when tested by iQ-Check STEC VirX and PhageDx. None of the methods detected false-positive samples and both methods correctly identified the positive controls.

Discussion

The advancement of rapid methods for the detection of E. coli O157:H7 is critical for ensuring food and water safety. Such methods must not only provide rapid results but also be easy to use, require minimal sample and reagent volumes, demand low operator skill, and incur low costs. These attributes are essential for the widespread adoption of detection methods across various settings, particularly in food safety testing [1]. Developing methods that fulfil these criteria requires a careful evaluation process, including specificity and sensitivity analyses against microorganisms found in foods from specific regions. Only after these evaluations can a method be used for food safety in a given country. The specificity of a test encompasses two main aspects: inclusivity and exclusivity. Inclusivity refers to the ability of a method to detect a broad range of isolates/strains within the target microorganism, while exclusivity ensures that non-target strains are not detected [25]. The PhageDx method also demonstrated robust inclusivity, correctly identifying all 103 strains of E. coli O157:H7 tested in the study by Erickson et al. [18]. The method also detected all the Brazilian strains of E. coli O157:H7 tested by us, regardless of their origin. Erickson et al. [18] also demonstrated that PhageDx assays have good exclusivity capabilities, which was confirmed by testing 37 non-O157 E. coli strains, 12 big-six Shiga toxin-producing E. coli (STEC) strains and 64 non-E. coli strains, all of which tested negative. Corroborating these findings, our study demonstrated no false positive result with the inclusion of the non-O157 E. coli strain ATCC 25922 used as negative control. Sensitivity is another important characteristic for pathogen detection methods, especially when dealing with low-level contamination, which is typical for E. coli O157:H7 present on foods. According to our results, PhageDx was able to detect Brazilian E. coli O157:H7 strains without prior enrichment in approximately 80% of wells contaminated with only 10 CFU, and achieving 100% detection at 100 CFU/well. High sensitivity capacity is very important to detection methods because it is assumed that some foodborne pathogens can cause illnesses even with a single viable cell [26]. To ensure sensitivity in real-world food testing scenarios, a 5-hour enrichment step was tested to increase the concentration of target bacterial cells [27]. This enrichment step not only enhances the sensitivity of the assay but also ensures accurate detection, as bacteriophages require live cells for replication.

Enrichment in selective or non-selective media is commonly required to detect low levels of E. coli O157:H7 in food. Although this enrichment step is time-consuming, it increases the sensitivity of the assay by promoting the growth of the target cells and, in some cases, inhibiting the growth of competing microbiota. In the present study, TSB was used for enrichment without the addition of antibiotics, a notable advantage in terms of cost and ease of use. Since the PhageDx method has demonstrated specificity for E. coli O157:H7, the absence of antibiotics did not compromise the assay's performance, simplifying the protocol and reducing material costs. PhageDx method was certified by Association of Official Analytical Chemists (AOAC), involving five main steps - enrichment, sample concentration, phage infection, substrate addition, and reading in a luminometer. The sample concentration step, although simple, was omitted on purpose in our study, and the enrichment time was standardized to 5 hours, as recommended by the method developer. Even without sample concentration the method remained highly sensitive with short time to results, a significant advantage over traditional methods, which often require 2-5 days for pathogen detection. The PhageDx method demonstrated adequate performance in detecting E. coli O157:H7 at very low concentrations, with reliable results in different food matrices, including ground beef, milk, lettuce, and mineral water. Even with food samples containing as low as 1 CFU per 25 g or 25 mL, the assay detected E. coli O157:H7 with consistent results across different food types, underscoring the robustness and sensitivity of the method. Additionally, the bioluminescent signal generated by the NanoLuc® enzyme allowed for easy interpretation of results, providing high light emission for positive samples and low background signals for negative controls, ensuring accurate detection and reducing the likelihood of false positives [28]. When compared to previous studies using bacteriophage-based detection systems, such as those carried out by Ripp et al. [29] and Zhang et al. [30], the PhageDx method demonstrated better performance. While other studies reported longer detection times and occasional false positives, the PhageDx method achieved faster results and demonstrated greater sensitivity, detecting as little as 1 CFU of E. coli O157:H7 in 25 g of ground beef within only 7.5 hours. This is a significant improvement, particularly in comparison to the 9-hour detection time reported by Zhang et al. [30] or other rapid detection methods used in big food companies which require approximately 18 to 24 hours to result.

Brazil, as the largest exporter of beef, faces considerable pressure to ensure the safety of its meat products, especially in light of mandatory Shiga-toxin-producing E. coli (STEC) testing for beef intended for export [31, 32]. According to the United States Department of Agriculture (USDA), Brazil's beef production was estimated at 9.75 million tons in 2022, reflecting a growth of approximately 4.6% comparing to the previous six-month period [2]. In terms of domestic consumption, beef consumption in Brazil was estimated to be 7.14 million tons at the beginning of 2022, while the Brazilian estimative for beef exports and imports for 2023 was approximately 2.68 million tons and 70,000 tons, respectively. Although the prevalence of STEC in Brazil remains low, there is a clear need for rapid and reliable testing methods for monitoring pathogens like E. coli O157:H7 and other STECs. The PhageDx assay performed well when tested with Brazilian ground beef samples. These results indicate that PhageDx could be useful in ongoing surveillance efforts.

While the current study did not find E. coli O157:H7 in any of the 100 beef carcass samples tested by the iQ Check and PhageDx assay, the similarity of the results of both methods can be considered an adequate result because no false positive was found and there were no divergent results between methods. Actually, these results highlight the need for efficient and accurate testing methods reinforcing that in Brazil, the prevalence of STEC has been reported to be low, but cases of contamination do occur [20, 21]. For instance, studies have detected E. coli O157:H7 in various Brazilian foods, including ground beef, milk, and vegetables [22, 33, 34]. The present study demonstrated that PhageDx is capable of detecting E. coli O157:H7 in various food products, indicating its utility as a rapid diagnostic tool for ensuring food safety. The PhageDx assay demonstrated to be able to detect low levels of E. coli O157:H7 in food and water samples in a short timeframe, which can be useful for both food safety testing and outbreak investigations. This last question is particularly important because there are already reports of foodborne outbreaks caused by E. coli O157 in Brazil.  For example, Santos et al. [35] published the first report of human gastroenteritis caused by E. coli O157:NM in Brazil. In this outbreak, a 13-year-old adolescent developed bloody diarrhea and severe abdominal pain and E. coli O157:NM was isolated from his feces and from a tomato and cheese salad prepared in the canteen of the school where the adolescent attended. More recently, E. coli O157:NM was identified as the cause of death of two children in southern Brazil, but the microorganism was not found in any food analyzed [36]. These outbreaks emphasize the need to improve food safety and rapid testing methods for foodborne pathogens are essential to this goal.

Conclusion

The recombinant bacteriophages used in the PhageDx assay demonstrated a capability to detect low concentrations of viable E. coli O157:H7 in various food matrices. The assay successfully identified 1 cfu/25g or mL of food within 7.5 hours, highlighting its potential as a rapid pathogen detection method.

Acknowledgments

The authors would like to thank the Coordination for the Improvement of Higher Education Personnel (CAPES) for scholarship granted to the post-graduate student Nathanyelle Soraya Martins de Aquino.

References

1. A Rani, VB Ravindran, A Surapaneni, et al. Trends in point-of-care diagnosis for Escherichia coli O157:H7 in food and water. Int J Food Microbiol 349 (2021): 109233.
2. GF Santos, FG Sousa, SL Beier, et al. Escherichia coli O157:H7 strains in bovine carcasses and the impact on the animal production chain. Food Microbiol 54 (2023): 2243-2251.
3. AA Tarekegn, BA Mitiku, YF Alemu. Escherichia coli O157:H7 beef carcass contamination and its antibiotic resistance in Awi Zone, Northwest Ethiopia. Food Sci Nutri 11 (2023): 6140-6150.
4. CDC (Center of Disease Control). Investigation Update: E. coli Outbreak, Onions Served at McDonald’s - October 30, 2024. (2024).
5. JY Lim, JW Yoon, CJ Hovde. A brief overview of Escherichia coli O157:H7 and its plasmid O157. J Microbiol Biotechnol 20 (2010): 1–10.
6. WHO (World Health Organization). Shiga toxin-producing Escherichia coli (STEC) and food: attribution, characterization, and monitoring. In Microbiological risks. Assesment series. (2018).
7. X Yang, H Sun, R Fan, et al. Genetic diversity of the intimin gene (eae) in non-O157 Shiga toxin-producing Escherichia coli strains in China. Sci Rep 10 (2020): 3275.
8. N Razmi, M Hasanzadeh, M Willander, et al. Recent progress on the electrochemical biosensing of Escherichia coli O157:H7: Material and methods overview. Biosensors 10 (2020): 1–18.
9. C Cuntín-Abal, B Jurado-Sánchez, A Escarpa. Playing with biological selectivity: Antimicrobial peptides and bacteriophages-based optical biosensors for pathogenic bacteria detection. Trends Analyt Chem 172 (2024): 117565.
10. Y Qi, S Li, Y Zhang, C You. Recent advances in viability detection of foodborne pathogens in milk and dairy products. Food Control 160 (2024): 110314.
11. SJ Shiue, MS Wu, JY Liao, et al. Bacteriophage-linked superparamagnetic nanoparticle for sensitive and simple detection of pathogenic Escherichia coli O157:H7 bacteria. Microchem 207 (2024): 111896.
12. NSM Aquino, SO Elias, EC Tondo. Evaluation of PhageDX Salmonella assay for Salmonella detection in hydroponic curly lettuce. Foods 10 (2021): 1795.
13. NSM Aquino, SO Elias, LVA Gomes, EC Tondo. Phage-based assay for the detection of Salmonella in brazilian poultry products. J Food Nutr Res 04 (2021): 249–258.
14. J Kim, M Kim, S Kim, et al. Sensitive detection of viable Escherichia coli O157:H7 from foods using a luciferase-reporter phage phiV10lux. Int J Food Microbiol 254 (2017): 11–17.
15. L Wu, X Hong, T Luan, et al. Multiplexed detection of bacterial pathogens based on a cocktail of dual-modified phages. Anal Chim Acta, 1166 (2021): 338596.
16. W Hussain, M Ullah, W Farooq, et al. Bacteriophage-based advanced bacterial detection: Concept, mechanisms, and applications. Biosens Bioelectron 177(2021): 112973.
17. M Schmelcher, MJ Loessner. Application of bacteriophages for detection of foodborne pathogens. Bacteriophage 4 (2014): e28137.
18. S Erickson, J Gil, B Hopkins, et al. A Novel Phage-Based Escherichia coli O157:H7 Detection Method for Ground Beef. International Association for Food Protection (2016): 81601.
19. VS Castro, EES Figueiredo, K Stanford, et al. Shiga-toxin producing Escherichia coli in Brazil: A systematic review. Microorganisms, 7 (2019): 137.
20. DCS Assis, TML Silva, Brito RF, et al. Shiga toxin-producing Escherichia coli (STEC) in bovine meat and meat products over the last 15 years in Brazil: A systematic review and meta-analysis. Meat Science 173 (2021): 108394.
21. A Lucatelli, DFM Monte, PP Alvares, et al. Virulent shiga toxin-producing Escherichia coli (STEC) O157:H7 ST11 isolated from ground beef in Brazil. Braz J Microbiol (2024).
22. PRK Lopes, RM Souza, GLPA Ramos, et al. Shiga toxin-producing Escherichia coli in the milk production chain: Evaluation of virulence genes and clonal diversity of O157:H7 strains. Int Dairy J 161 (2025): 106119.
23. MR Loiko, CMD Paula, ACJ Langone, et al. Genotypic and antimicrobial characterization of pathogenic bacteria at different stages of cattle slaughtering in southern Brazil. Meat Sci 116 (2016): 193-200.
24. S Erickson, J Gil, M Nguyen. PhageDx^TM coli O157:H7 Assay in ground beef level 3 modification to add 375 g beef trim and new method procedure for 25 g ground beef: AOAC performance tested MethodSM 081601. J AOAC Int 104 (2021): 1338-1343.
25. M Wiedmann, S Wang, L Post, et al. Assessment criteria and approaches for rapid detection methods to be used in the food industry. J Food Prot, 77 (2014): 670-690.
26. JWF Law, NSA Mutalib, KG Chan, et al. Rapid methods for the detection of foodborne bacterial pathogens: Principles, applications, advantages and limitations. Front Microbiol 5 (2014): 1-19.
27. G Amagliani, L Rotundo, E Carloni, et al. Detection of Shiga toxin-producing Escherichia coli (STEC) in ground beef and bean sprouts: Evaluation of culture enrichment conditions. Food Res Int 103 (2018):398–405.
28. MP Hall, J Unch, BF Binkowski, et al. Engineered luciferase reporter from a deep sea shrimp utilizing a novel imidazopyrazinone substrate. ACS Chem Biol 7 (2012): 1848-1857.
29. S Ripp, P Jegier, CM Johnson, et al. Bacteriophage-amplified bioluminescent sensing of Escherichia coli O157:H7. Anal Bioanal Chem 391 (2008): 507-514.
30. D Zhang, CP Coronel-Aguilera, PL Romero, et al. The Use of a Novel NanoLuc-Based Reporter Phage for the Detection of Escherichia coli O157:H7. Sci Rep 6 (2016): 6-13.
31. ABIEC (Associação Brasileira das Indústrias Exportadoras de Carnes).Beef Report - Perfil da Pecuária no Brasil. (2021).
32. MAPA (Ministério da Agricultura e Pecuária do Brasil). Anuário dos Programas de Controle de Alimentos de Origem Animal do DIPOA 6 (2020).
33. CMDD Paula, MR Loiko, EC Tondo. Escherichia coli O157: H7: local epidemiology and disease spectrum in Brazil. Clinical and biomedical research. Porto Alegre 2 (2014): 113-121.
34. LT Decol, LS Casarin, CT Hessel, et al. Microbial quality of irrigation water used in leafy green production in Southern Brazil and its relationship with produce safety. Food Microbiol 65 (2017): 105-113.
35. RFC Santos, JDS Nascimento, MP Geimba, et al. First Report of Human Gastroenteritis Caused by Escherichia coli O157:NM in Brazil. Foodborne Pathog Dis 14 (2017) 665-666.
36. FW Bartz, LB Teixeira, R Schroder, et al. First Fatal Cases due to Escherichia coli O157 and Campylobacter jejuni subsp. jejuni Outbreak Occurred in Southern Brazil. Foodborne Pathog Dis 19 (2022): 4.