Comparison of Detection Methods for Escherichia coli O157 in Beef Livers and Carcasses

Abstract

Beef organ meat, such as liver, and beef are major food sources contaminated with Escherichia coli O157. This study investigated the detection method of E. coli O157 in beef liver and carcass. In an experiment with beef liver inoculated with E. coli O157, the direct plating method, plating after the immunomagnetic separation (IMS) method, and Shiga toxin (Stx)–producing E. coli detection and E. coli O157 detection loop-mediated isothermal amplification (LAMP) assays were compared for the detection of Stx-producing E. coli O157. Fifty percent and 45% of samples were positive by Stx-producing E. coli detection LAMP assay and E. coli O157 detection LAMP assay, respectively. Thirty-five percent and 10% of samples were positive by the IMS method and direct plating method, respectively. In an examination of beef swab samples, contamination frequencies with E. coli O157 were analyzed by LAMP assays and the IMS method. E. coli O157 was detected in 12 of 230 samples (5.2%). There was no sample positive for E. coli O157 isolation but negative for LAMP assays for Stx gene and O157 antigen gene. Four samples (1.7%) were positive by both LAMP assays but negative by the IMS method. The result that there was no sample positive for the O157 antigen gene, but not the Stx gene, indicated that the IMS method failed to detect E. coli O157. Twenty-nine samples (12.6%) were positive for the Stx gene but not the O157 antigen gene. The results indicated that screening of Stx gene and O157 antigen gene by LAMP assays is effective in saving time and effort to isolate E. coli O157 by the IMS method because the LAMP assay is more sensitive. This suggested that samples positive for Stx gene and O157 antigen gene should be examined by the IMS method to isolate E. coli O157.

Introduction

M ajor food sources associated with foodborne infection by Escherichia coli O157 are beef products, such as ground beef (Hussein, 2007; King et al., 2009; Rhoades et al., 2009), and vegetables (Friesema et al., 2008; Grant et al., 2008; Söderström et al., 2008). Other foods have been also examined in various countries. In The Netherlands, steak tartare hamburger has been associated with outbreaks of E. coli O157 infection (Doorduyn et al., 2006). In Japan, consumption of raw beef, including organ meats, and beef barbecue is well known to be related to outbreaks (Anonymous, 2007). Beef is the most prominent food to be considered in E. coli O157 infection control in many countries. To reduce E. coli O157 infections, reduction of the pathogen on farms is clearly important; however, infection control on farms is extensive because of the cost or other difficulties. Slaughterhouses and meat-processing plants are a key to reduce beef contamination with E. coli O157 because the workers are able to prevent contamination by the pathogen from feces and intestinal contents. In addition, E. coli O157 on the surface of beef carcasses is removed by disinfectants or cutting the surface if pathogen examination is positive. It is recommended that examination for E. coli O157 by culture methods and/or molecular methods is generally performed in slaughterhouses and meat-processing plants.

The main methods of E. coli O157 detection are culture methods, including enrichment, direct plating, and the immunomagnetic separation (IMS) method; however, these methods are time consuming and labor intensive. Isolating pathogens is essential to confirm their antigenicity and analyze their characteristics. Molecular methods have been developed to detect pathogens under various situations (López et al., 2003; McKillip and Drake, 2004). The advantages of molecular methods include time- and labor-saving techniques and their specificity and sensitivity, although they are not useful to isolate pathogens. The usage of molecular methods to screen for pathogens, followed by culture methods to isolate pathogens, is effective. Loop-mediated isothermal amplification (LAMP) is a novel nucleic acid amplification method that relies on autocycling strand displacement DNA synthesis performed by using the Bst DNA polymerase large fragment (Notomi et al. 2000; Nagamine et al. 2002). LAMP is different from polymerase chain reaction (PCR) in that four or six primers perform the amplification of the target gene, the amplification uses a single temperature step at 60°C–65°C for about 60 min, and the amplification products have many types of structures in large amounts. Thus, LAMP is more specific, rapid, and simple to perform than PCR. Further, gel electrophoresis is not needed, because the LAMP method synthesizes a large amount of DNA and the products can be detected by simple turbidity or fluorescence. Thus, expensive equipment is not necessary to give a high level of precision, equivalent or greater, when compared with the other PCR techniques. LAMP assay for Shiga toxin (Stx)–producing E. coli was developed and used to screen for the Stx gene (Hara-Kudo et al., 2008). The enrichment culture positive for the Stx gene is concentrated E. coli O26 and O157 cells by IMS methods. After inoculation of the concentrated sample onto selective medium and incubation, colonies are confirmed as E. coli O26 and O157. Screening is effective in saving time and labor.

In this study, liver artificially contaminated with E. coli O157 and beef carcasses were tested by culture methods and LAMP assays targeting Stx gene and O157 antigen gene to assess the effectiveness of the molecular methods. The investigation of beef carcasses elucidated the contamination frequency of Stx-producing E. coli and/or E. coli O157.

Materials and Methods

Experiment 1

Liver was purchased from a retail shop in Saitama prefecture, Japan, and stored in a refrigerator until use for experiments. A part of liver was tested negative for Stx-producing E. coli and E. coli O157 by the LAMP assays described below. Liver was sliced into 25-g samples before inoculation.

Four strains of E. coli O157:H7, strain no. EC8 and SEC212 from patients, E502 from pickles, and 96-14 from beef, were used for inoculation into liver samples. The strains were positive for agglutination with the E. coli O157 antiserum (Denka Seiken) and for Stx gene and O157 antigen gene by the LAMP assays described below. Each strain was incubated in tryptic soy broth (Difco; BD) at 37°C for 18 h. The culture was diluted in phosphate-buffered saline to 10⁻⁸. The dilutions were used as inocula for liver samples. To determine inoculation size, each 0.1 mL of 10⁻⁸ dilutions was plated onto 10 plates of tryptic soy agar (Difco). After incubation at 37°C for 24 h, the colonies were counted.

For artificial contamination with E. coli O157, inocula (1 mL) of each strain were added to five liver samples. After storage at 10°C for 1 h, the samples were incubated with 225 mL modified EC broth (mEC) with novobiocin (NmEC; Eiken Chemical) at 42°C for 18 h, respectively. Culture (20 μL) was used for direct plating method. One milliliter of culture was used for the E. coli O157 IMS method. Magnetic beads coated with O157 antibody (Denka Seiken) were finally suspended in 0.1 mL phosphate-buffered saline with 0.05% Tween-20. A portion (20 μL) was streaked onto sorbitol MacConkey agar medium (Oxoid) supplemented with 0.05 mg/L cefixime and 2.5 mg/L tellurite (Oxoid) (CT-SMAC). These agar plates were incubated at 37°C for 18–24 h. The colonies suspected as E. coli O157 were confirmed positive by agglutination with E. coli O157 antiserum (Denka Seiken) and Stx gene detection by the LAMP assay described below.

Experiment 2

In three slaughterhouses in Saitama prefecture, swab samples of 100-cm² surface of brisket and rump from 230 beef carcasses were collected from June 2003 to August 2004. The samples were immediately tested after collecting. Swab samples of beef carcass naturally contaminated with Stx-producing E. coli (STEC) were incubated with 20 mL NmEC at 42°C for 18 h. Culture (20 μL) was used for direct plating. One milliliter of culture was used for the E. coli O157 IMS method described above. A portion (20 μL) was streaked onto CT-SMAC, BCMO157 agar medium (Eiken Chemical), and CHROMagar O157 medium (CHROMagar). The colonies suspected as E. coli O157 were confirmed positive by agglutination with E. coli O157 antiserum (Denka Seiken) and Stx gene detection by the LAMP assay described below. Fifty colonies on agar media from samples that were positive for Stx gene but negative for E. coli O157 isolation were tested for Stx gene by the LAMP assay described below. The serotypes of Stx-positive isolates were determined using O and H antisera against E. coli (Denka Seiken) and Stx-production by a reversal passive latex agglutination kit (VET-RPLA; Denka Seiken).

LAMP assays

The LAMP reaction was performed using LAMP assay kits targeting Stx gene (verotoxin-producing E. coli detection kit; Eiken Chemical) (Hara-Kudo et al., 2007) and O157 antigen gene wzy and wbdO (Wang and Reeves, 1998) (E. coli O157 detection kit; Eiken Chemical). Each 50 μL culture of samples was added to 50 μL extraction solution (50 mM NaOH, pH 12.5) to extract DNA before heating at 95°C for 5 min. After brief flash centrifugation, the 50 μL samples were added to 8 μL of 1 M Tris-HCl (pH 7.0) and centrifuged (Tomy) at 2000 g, and then the supernatant was transferred to a new microtube and used as the template DNA solution for the LAMP assay. The LAMP reaction mixture contained the primers for Stx gene or O157 antigen gene detection (20 μL), Bst DNA polymerase (1 μL), and template DNA solution (5 μL). The reaction components were mixed in a tube, incubated at 65°C for 60 min using a Loopamp LA-200 (Teramecs), and then heated to 80°C for 2 min to terminate the reaction. LAMP amplification was detected as turbidity at 650 nm absorbance using a Loopamp LA-200 in real time. In addition, turbidity produced by the magnesium pyrophosphate byproduct was noted visually.

Statistics

Data from experiment 1 were analyzed by chi-square test for significant differences among the methods.

Results and Discussion

STEC infections are annually reported in 3000–4600 cases, and serogroup O157 occupy 65%–75% of STEC in Japan (Anonymous, 2009). A major food source associated with food-borne infection by E. coli O157 is beef, including organ meat (Itoh, 2006). It is suspected that hemolytic uremic syndrome in younger children is caused by consumption of raw or undercooked beef. Organ meat, in particular, such as liver, is consumed quickly because the meat easily loses freshness. For these reasons, rapid and effective methods using molecular techniques are needed to detect E. coli O157 in beef and beef organ meat. In this study, LAMP assays, a molecular method that quickly gives results, were compared with culture methods, which require more than 4 days.

In liver samples of experiment 1, the inoculation level of each strain of E. coli O157 into liver samples was 1–4 cfu/25 g (Table 1). No sample was positive for the isolation of E. coli O157 by culture methods, or the detection of Stx gene and O157 antigen gene by LAMP assay in the samples inoculated with strain no. 96-14. The reasons include the inoculation level of the strain was very low. E. coli O157 was not detected in samples inoculated with strains EC8 and SEC212 by the direct plating method. Overall, LAMP assays targeting Stx gene and O157 antigen gene successfully detected the target genes in 50% (10/20) and 45% (9/20) of samples, respectively; however, the direct plating method recovered E. coli O157 in 10% (2/20) of samples. The recovery rate was improved by the IMS method to 35% (7/20). In all the samples inoculated with E. coli O157 strains, LAMP assays for Stx gene and O157 antigen gene were superior to direct plating method (p < 0.01 and p < 0.05, respectively). All positive samples by the direct plating and IMS methods were positive for Stx gene and O157 antigen gene by LAMP assays. Commercial raw organ meat, including liver, was examined for E. coli O157 by plating after the E. coli O157 IMS method in 2000–2004 (Kitase and Ishii, 2005), and E. coli O157 was detected in 8.3% (2/24) of liver samples. The prevalence might increase by using molecular methods such as the LAMP assays used in this study.

Table 1.

Detection of Shiga Toxin–Producing Escherichia coli O157 in Liver Inoculated with the Pathogen by Culture Methods and Loop-Mediated Isothermal Amplification Assays

		Escherichia coli O157 isolation by culture methods		Detection by LAMP assays
Strain no.	No. of inoculated cells (cfu/25 g of liver)	Direct plating	Plating after IMS ^a	Shiga-toxin gene	O157 antigen gene
EC8	4	0/5^b	2/5	3/5	3/5
E502	3	2/5	4/5	5/5	5/5
SEC212	3	0/5	1/5	2/5	1/5
96-14	1	0/5	0/5	0/5	0/5
Total		2/20 (10%)^c	7/20 (35%)	10/20 (50%)	9/20 (45%)
Control^d	0	0/5	0/5	0/5	0/5

IMS method for O157 antigen.

The number of positive samples/the number of samples tested.

There were significant differences between direct plating and Shiga-toxin gene LAMP assay (p < 0.01), and direct plating and O157 antigen gene LAMP assay (p < 0.05).

Uninoculated samples were tested as control.

LAMP, loop-mediated isothermal amplification; IMS, immunomagnetic separation.

In swab samples from beef carcass of experiment 2, 12 swab samples (5.2%) were positive for the isolation of E. coli O157 by the IMS method (Table 2). The samples were also positive for Stx gene and O157 antigen gene by LAMP assays. The prevalence and levels of E. coli O157 in beef carcasses have been reported previously. In the United Kingdom, 3.2% (8/250) of samples were positive for E. coli O157 in a typical Irish beef abattoir (McEvoy et al., 2003). In another study, also in Ireland, E. coli O157 was recovered from 3% (4/132) of carcass samples at a beef slaughterhouse (Carney et al., 2006). In the United States, 1.8% (6/330) of postprocessing carcasses from four beef-processing plants were positive for E. coli O157 (Elder et al., 2000). Barkocy-Gallagher et al. (2003) also reported a prevalence of E. coli O157 of 1.2% (15/1232) and the level was <3 cfu/100 cm². Recently, 2.6% (39/1503) of carcasses in a slaughterhouse were reported to be positive for E. coli O157 (Fox et al., 2008). The prevalence (5%) in this study was similar to these data. The removal of the hide and intestinal tract in slaughterhouses is regarded as the major cause of carcass contamination with E. coli O157 (Elder et al., 2000; Barkocy-Gallagher et al., 2003; McEvoy et al., 2003). The prevalence of E. coli O157 in carcasses correlated to the fecal and hide prevalence (Elder et al., 2000). Reducing carcass contamination from hide and feces might prevent the consumption of meat contaminated with the pathogen. Examination of E. coli O157 in beef meat-processing plants might also remove meat contaminated with the pathogen.

Table 2.

Detection of Shiga Toxin–Producing Escherichia coli in Swab Samples of Beef Carcass from Slaughterhouses by Plating Method with Immunomagnetic Separation and Loop-Mediated Isothermal Amplification Assay

	LAMP assays
	Shiga toxin gene			O157 antigen gene
Plating method with IMS targeting O157	Positive	Negative	Total	Positive	Negative	Total
Positive	12^a	0	12	12	0	12
Negative	33^b	185	218	4^c	214	218
Total	45	185	230	16	214	230

Number of samples.

Shiga toxin–producing E. coli O159:H19 (four different samples), OUT:H21 (one sample), OUT:H19 (one sample), OUT:H7 (one sample) were isolated.

All four samples were positive for Shiga toxin gene by the LAMP assay.

In beef carcass swab samples of this study, Stx gene and O157 antigen gene were detected in 45 and 16 samples, respectively. Four samples (1.7%) positive for the O157 antigen gene and negative for the isolation of E. coli O157 by IMS method were positive for the Stx gene. It is possible that the IMS method failed to isolate E. coli O157 from the samples. In 33 samples positive for the Stx gene and negative for E. coli O157 by the IMS method, 29 samples (12.9%) were negative for the O157 antigen gene, indicating that the samples were contaminated with Stx-positive E. coli serotypes other than O157. In fact, Stx-positive E. coli O159:H19, OUT:H21, OUT:H19, and OUT:H7 were isolated from four, one, one, and one sample, respectively.

Meat and organ meat, such as the liver and intestinal tract, are consumed both raw and cooked in Japan. People, including children, become infected with E. coli O157 in barbecue restaurants serving raw beef and beef organ meats that are consumed raw or grilled by customers themselves. Cooking equipments used, such as chopsticks and tongs for raw meat, contaminate meat and so this method of consuming beef has a risk of infection. Although E. coli O157-free beef and the consumption of cooked meat are good solutions to prevent infection, they are difficult to achieve in practice. Examination of beef production and monitoring of food by rapid and sensitive methods such as the LAMP assay could help to abolish food contaminated with E. coli O157.

The present study indicates that LAMP assays targeting Stx gene and O157 antigen gene are more sensitive than direct plating and IMS methods for culture. The LAMP assays are effective for the rapid and sensitive detection of E. coli O157 in beef liver and carcasses. Rapid and sensitive examination of E. coli O157 in slaughterhouses and meat-processing plants would reduce E. coli O157 infections.

Footnotes

Acknowledgment

This research was supported by Health Science Research Grants from the Ministry of Health, Labor, and Welfare in Japan.

Disclosure Statement

No competing financial interests exist.

References

Anonymous. Enterohemorrhagic Escherichia coli infection in Japan as of April 2007. Infect Agents Surveill Rep, 2007; 28:131–132.

Anonymous. Enterohemorrhagic Escherichia coli infection in Japan as of April 2009. Infect Agents Surveill Rep, 2009; 30:119–120.

Barkocy-Gallagher

, Arthur

, Rivera-Betancourt

, Nou

, Shackelford

, Wheeler

, Koohmaraie

. Seasonal prevalence of Shiga toxin-producing Escherichia coli, including O157:H7 and non-O157 serotypes, and Salmonella in commercial beef processing plants. J Food Prot, 2003; 66:1978–1986.

Carney

, O'Brien

, Sheridan

, McDowell

, Blair

, Duffy

. Prevalence and level of Escherichia coli O157 on beef trimmings, carcasses and boned head meat at a beef slaughter plant. Food Microbiol, 2006; 23:52–59.

Doorduyn

, de Jager

, van der Zwaluw

, Friesema

, Heuvelink

, de Boer

, Wannet

, van Duynhoven

. Shiga toxin-producing Escherichia coli (STEC) O157 outbreak, The Netherlands, September–October 2005. Eurosurveillance, 2006; 11:182–185.

Elder

, Keen

, Siragusa

, Barkocy-Gallagher

, Koohmaraie

, Laegreid

. Correlation of enterohemorrhagic Escherichia coli O157 prevalence in feces, hides, and carcasses of beef cattle during processing. Proc Natl Acad Sci USA, 2000; 97:2999–3003.

Fox

, Renter

, Sanderson

, Nutsch

, Shi

, Nagaraja

. Associations between the presence and magnitude of Escherichia coli O157 in feces at harvest and contamination of preintervention beef carcasses. J Food Prot, 2008; 71:1761–1767.

Friesema

, Sigmundsdottir

, van der Zwaluw

, Heuvelink

, Schimmer

, de Jager

, Rump

, Briem

, Hardardottir

, Atladottir

, Gudmundsdottir

, van Pelt

. An international outbreak of Shiga toxin-producing Escherichia coli O157 infection due to lettuce, September-October 2007. Eurosurveillance, 2008; 13:pii:19065.

Grant

, Wendelboe

, Wendel

, Jepson

, Torres

, Smelser

, Rolfs

. Spinach-associated Escherichia coli O157:H7 outbreak, Utah and New Mexico, 2006. Emerg Infect Dis, 2008; 14:1633–1636.

10.

Hara-Kudo

, Konishi

, Otsuka

, Hiramatsu

, Tanaka

, Tsuchiya

, Konuma

, Takatori

. Detection of verotoxigenic Escherichia coli O157 and O26 in food by plating methods and LAMP method: a collaborative study. Int J Food Microbiol, 2008; 122:156–161.

11.

Hara-Kudo

, Nemoto

, Ohtsuka

, Segawa

, Takatori

, Kojima

, Ikedo

. Sensitive and rapid detection of Vero toxin-producing Escherichia coli using Loop-mediated isothermal amplification. J Med Microbiol, 2007; 56:398–406.

12.

Hussein

. Prevalence and pathogenicity of Shiga toxin-producing Escherichia coli in beef cattle and their products. J Anim Sci, 2007; 85,13 Suppl:E63–E72.

13.

Itoh

. Current topics of enterohemorrhagic Escherichia coli infection in Japan. Food Sanit Res, 2006; 56:9–16. (In Japanese.)

14.

King

, Mailles

, Mariani-Kurkdjian

, Vernozy-Rozand

, Montet

, Grimont

, Pihier

, Devalk

, Perret

, Bingen

, Espié

, Vaillant

. Community-wide outbreak of Escherichia coli O157:H7 associated with consumption of frozen beef burgers. Epidemiol Infect, 2009; 137:889–896.

15.

Kitase

, Ishii

. Distribution of enterohemorrhagic Escherichia coli O157 in internal organs of cattle on the market. Annu Rep Osaka City Inst Public Health Environ Sci, 2005; 67:15–19. (In Japanese.)

16.

López

, Bertolini

, Olmos

, Caruso

, Gorris

, Llop

, Penyalver

, Cambra

. Innovative tools for detection of plant pathogenic viruses and bacteria. Int Microbiol, 2003; 6:233–243.

17.

McEvoy

, Doherty

, Sheridan

, Thomson-Carter

, Garvey

, McGuire

, Blair

, McDowell

. The prevalence and spread of Escherichia coli O157:H7 at a commercial beef abattoir. J Appl Microbiol, 2003; 95:256–266.

18.

McKillip

, Drake

. Real-time nucleic acid-based detection methods for pathogenic bacteria in food. J Food Prot, 2004; 67:823–832.

19.

Nagamine

, Kuzuhara

, Notomi

. Isolation of single-stranded DNA from loop-mediated isothermal amplification products. Biochem Biophys Res Commun, 2002; 290:1195–1198.

20.

Notomi

, Okayama

, Masubuchi

, Yonekawa

, Watanabe

, Amino

, Hase

. Loop-mediated isothermal amplification of DNA. Nucleic Acids Res, 2000; 28:E63.

21.

Rhoades

, Duffy

, Koutsoumanis

. Prevalence and concentration of verocytotoxigenic Escherichia coli, Salmonella enterica and Listeria monocytogenes in the beef production chain: a review. Food Microbiol, 2009; 26:357–376.

22.

Söderström

, Osterberg

, Lindqvist

, Jönsson

, Lindberg

, Blide Ulander

, Welinder-Olsson

, Löfdahl

, Kaijser

, De Jong

, Kühlmann-Berenzon

, Boqvist

, Eriksson

, Szanto

, Andersson

, Allestam

, Hedenström

, Ledet Muller

, Andersson

. A large Escherichia coli O157 outbreak in Sweden associated with locally produced lettuce. Foodborne Pathog Dis, 2008; 5:339–349.

23.

Wang

, Reeves

. Organization of Escherichia coli O157 O antigen gene cluster and identification of its specific genes. Infect Immun, 1998; 66:3545–3551.