Project Summary

Establishment of a Non-O157 Shiga Toxin-Producing E. coli (STEC) Baseline in Raw Ground Beef in the U.S.

Principle Investigator(s):
YenTe Liao, Guy H. Loneragan, Mark F. Miller, J Chance Brooks, Alejandro Echeverry, Kendra Nightingale, Mindy Brashears
Texas Tech University
Completion Date:
May 2012

Shiga toxin-producing Escherichia coli (STEC), which are capable of producing Shiga toxin 1 and/or 2, have been associated with foodborne outbreaks and severe human illness - bloody diarrhea, hemolytic uremic syndrome (HUS), and kidney failure, for example - since 1982, when E. coli O157:H7 was frequently isolated from undercooked hamburger beef patties. Although not all non-O157 STEC are pathogenic to humans, some non-O157 STEC strains have been related to sporadic cases of human illness and are able to cause disease as severe as E. coli O157:H7, such as bloody diarrhea, hemorrhagic colitis, HUS, or even death. STEC primarily resides in the gastrointestinal tract of ruminants, especially cattle, and will not cause any illness to their hosts. These strains can be easily disseminated to the environment or other surfaces through direct fecal contamination, which is also the primary food safety issue during the production of beef products. Centers for Disease Control and Prevention (CDC) estimated that six non-O157 STEC strains, including serotypes O26, O45, O103, O111, O121, and O145, accounted for over 70% of non-O157 STEC outbreaks in the United States. Due to the increasing number of non-O157 STEC-associated illness and capability of developing severe disease, USDA’s Food Safety and Inspection Service (FSIS) has declared the six most prevalent non-O157 STEC (O26, O45, O103, O111, O121, and O145) as adulterants in ground beef or other non-intact beef products. 

Due to non-phenotypic characteristics of non-O157 STEC, it has been a challenge to detect these bacteria with the fast and accurate methods; therefore, illness associated with non-O157 STEC has always been underestimated. Polymerase chain reaction (PCR)-based detection methods have been considered as valuable tools to trace outbreaks of bacterial pathogens in the food supplies. The BAX system is one of the commercial PCR-based systems that utilize the amplification of a specific target DNA sequence for the detection of pathogens in the food. 

To date, unlike E. coli O157:H7, the information regarding the prevalence of non-O157 STEC in commercial ground beef samples from the retailers is scarce. Therefore, there is a need to set up the baseline of non-O157 STEC in the commercial ground beef samples and to determine fast and reliable methods to detect those stains. 

The objective of this study was to identify the prevalence of the six pathogenic non-O157 STECs (serotypes O26, O45, O103, O111, O121, O145) in commercial ground beef products, and their association with different harvest parts from carcass, and lean/fat proportion.


Ground Beef Samples Collection

Ground beef samples (n = 1,000) were purchased from grocery stores in 24 states (NE, IA, TX, KS, WY, CO, WI, CA, OK, MS, OH, MI, IN, IL, MO, FL, AZ, NM, AL, SC, GA, NC, TN, and AR) in October 2011 to February 2012. Commercial ground beef products with various lean/fat portion (96/4, 93/7, 90/10, 85/15, 81/19, 80/20, or 73/27), different location of cut (chuck, round, or sirloin), and different packaging (overwrap, vacuum, or chub) were chosen. Samples were stored in the coolers during transportation to the Microbiology Lab at ESB in Texas Tech University. Ground beef samples were documented and stored in plastic containers at -20°C. 

Microbiological Analysis

Ground beef samples were thawed out in the walk-in cooler for 48 hours, and 25 grams of each sample was placed in a stomacher bag with addition of 225 mL of tryptic soy broth (TSB). After stomaching at 230 rpm for 1 min, ground beef samples were incubated at 41oC for 18±2 h. After enrichment, enriched ground beef broths were subjected to BAX® system to detect the six non-O157 STECs (O26, O45, O103, O111, O121, and O145). 

For BAX DNA isolation, 200 μL of BAX DNA lysis were mixed with 20 μL of sample. The cells were first lysed at 37°C for 20 min, followed by 95°C for 10 min. After cooling at 4°C for 5 min, DNA was loaded onto STEC stx/eae screening kits, and positive samples would be further subjected to BAX using STEC Panel 1 and STEC Panel 2 kits to determine the presence of serotypes. IMS were performed with Dynabeads using the BeadRetrieverTM instrument. All ground beef samples were subjected to IMS by mixing 1 mL of the enriched samples with 20 μL of anti-non-O157 beads, including O26, O103, O111, and O145. Beads were washed three times in PBS-Tween 20, and 50 μL of the bead-bacteria mixture of E. coli O157 beads were spread onto Rainbow® Agar plates. The plates were incubated at 37°C overnight. All morphologically different colonies from Rainbow® Agar plates were tested against serogroup-specific antisera by a slide agglutination test. The blue polystyrene latex particles used in the kit are coated with an antibody against the E. coli O26, O103, O111 or O145 somatic antigen. When these latex particles were mixed with fresh colonies of E. coli O26, O103, O111 or O145, the bacteria bound to the antibody, causing the latex particles to agglutinate (positive reaction).


Experiment 1: Prevalence rates of the non-O157 STEC in ground beef samples:  The results of this study indicated that the prevalence of non-O157 STEC was 0.9% in the ground beef samples (Table 1). Detection of only stx- or eae-genes was in 2.4% and 2.3% of samples, respectively. In addition, ground beef made of different part of meat from the carcass, such as round, chuck or sirloin, was not related to the detection of the non-O157 STEC; however, non-O157 STECs were frequently isolated from the 85/15 (lean/fat) portion of ground beef products (Table 2). In this study, a total of 12 STEC isolates were obtained in the commercial ground beef, and the most common O groups detected were O26 (n = 4), followed by O103 (n = 3), O45 (n = 2), O145 (n = 2), and O121 (n = 1); nevertheless, serotype O111 was not isolated from the ground beef in this study (Table 2).

Table 1. Detection of genes (stx and/or eae) and confirmation tests in the ground beef samples using the BAX screening and confirmation steps of the STEC suite*.




stx & eae

Confirmed STECa

% of samples





*Numbers of ground beef samples (n = 1000).

aInclude serotypes of O26, O45, O103, O111, O121, or O145.

Table 2. Shiga toxin-producing E. coli (STEC) isolated in the ground beef using IMS.



Type of meat













O26, O145
















O121, O45, O103