Chromogenic Agar Medium for Detection and Isolation of Escherichia coli Serogroups O26, O45, O103, O111, O121, and O145 from Fresh Beef and Cattle Feces3

Non-O157 Shiga toxin–producing Escherichia coli (STEC) strains are clinically important foodborne pathogens. Unlike E. coli O157:H7, these foodborne pathogens have no unique biochemical characteristics to readily distinguish them from other E. coli strains growing on plating media. In this study, a chromogenic agar medium was developed in order to differentiate among non-O157 STEC strains of serogroups O26, O45, O103, O111, O121, and O145 on a single agar medium. The ability of this chromogenic agar medium to select and distinguish among these pathogens is based on a combination of utilization of carbohydrates, b-galactosidase activity, and resistance to selective agents. The agar medium in combination with immunomagnetic separation was evaluated and successfully allowed for the detection and isolation of these six serogroups from artificially contaminated fresh beef. The agar medium in combination with immunomagnetic separation also allowed successful detection and isolation of naturally occurring non-O157 STEC strains present in cattle feces. Thirty-five strains of the top six non-O157 STEC serogroups were isolated from 1,897 fecal samples collected from 271 feedlot cattle. This chromogenic agar medium could help significantly in routine screening for the top six non-O157 STEC serogroups from beef cattle and other food. In the early 1980s, Escherichia coli O157:H7, a member of the enterohemorrhagic E. coli (EHEC) group, gained recognition as the causative agent for an outbreak of severe bloody diarrhea traced to consumption of improperly cooked hamburgers (28). EHEC bacteria excrete potent toxins called Shiga toxins and carry genes for virulence factors, such as intimin and hemolysin, which are responsible for attaching to and effacing host epithelial cells and for lysis of red blood cells, respectively. Although E. coli O157:H7 is currently the most widely recognized EHEC member, strains of more than 200 other non-O157 Shiga toxin–producing Escherichia coli (STEC) serotypes have been implicated in cases of human disease (5). NonO157 STEC strains are similar to E. coli O157:H7 in that these bacteria produce Shiga toxins, can be isolated from bovine feces, and are found as contaminants on the hides and carcasses of processed beef cattle (3). Many strains have been found with the same serotypes and virulence genotypes as those determined to cause human disease. According to the Centers for Disease Control and Prevention (CDC), the most common non-O157 STEC serogroups associated with human disease in the United States include O26, O103, O111, O121, O45, and O145 (5). The newly emerging enteroaggregative E. coli serotype O104:H4 that was identified as the causative agent in a large outbreak in Germany and other European countries was not included in the top six serogroups listed by the CDC because meat animals are not reservoirs of enteroaggregative E. coli, including the aggregative serotype O104:H4 (6, 30). It has been estimated that non-O157 STEC strains cause approximately 112,752 cases of foodborne illness in the United States annually (29). The true number of illnesses caused by non-O157 STEC may be underestimated, as only about 4% of clinical laboratories routinely screen for these pathogens (15). In September 2011, the Food Safety and Inspection Service, U.S. Department of Agriculture, published a notice in the Federal Register of their intent to regulate non-O157 STEC bacteria of serogroups O26, O45, O103, O111, O121, and O145 as adulterants in certain raw beef products, as had been done for E. coli O157:H7 several years earlier (31). The isolation of non-O157 STEC in stool and foodstuffs is laborious and time consuming due to the lack of differential and selective plating media for isolation. Over the years, varied strategies have been developed for the cultivation of pathogens. The ability to detect the * Author for correspondence. Tel: 402-762-4224; Fax: 402-762-4149; E-mail: [email protected]. { Mention of trade names or commercial products in this article is solely for the purpose of providing specific information and does not imply recommendation or endorsement by the U.S. Department of Agriculture. The USDA is an equal opportunity provider and employer. 192 Journal of Food Protection, Vol. 76, No. 2, 2013, Pages 192–199 doi:10.4315/0362-028X.JFP-12-182 presence of a specific enzyme using suitable substrates, in particular fluorogenic or chromogenic enzyme substrates, has led to the development of a great number of methods for the identification of microorganisms in primary isolation media. Several chromogenic agar media for the detection and isolation of E. coli O157:H7 are available commercially, such as Rainbow agar O157 (Biolog, Inc., Hayward, CA), BCM O157:H7 (Biosynth AG, Staad, Switzerland), Fluorocult E. coli O157:H7 (Merck, Darmstadt, Germany), R & F E. coli O157:H7 (R & F Laboratory, Downers Grove, IL), CHROMagar O157 (CHROMagar Microbiology, Paris, France), CHROMagar O26/O157 (CHROMagar Microbiology), and CHROMagar STEC (CHROMagar Microbiology), as well as a noncommercial chromogenic agar medium (26). However, there is no true chromogenic agar medium commercially available to differentiate strains of all top six non-O157 serogroups of STEC on a single plate due to their variation in biochemical characteristics. At the U.S. Meat Animal Research Center (USMARC), we recently developed a unique chromogenic agar medium for the detection and isolation of strains of serogroups O26, O45, O103, O111, O121, and O145. In the present study, the USMARC chromogenic agar medium was evaluated for its efficiency in detecting and isolating STEC from artificially contaminated fresh beef and naturally contaminated cattle feces. MATERIALS AND METHODS Bacterial strains and growth conditions. The bacterial cultures from the USMARC culture collection listed in Table 1, including gram-negative and gram-positive strains, were grown in nutrient broth (Difco, BD, Sparks, MD) for 16 to 18 h at 37uC. The strains of Enterobacteriaceae and Pseudomonas were streaked for isolation on violet red bile agar (Difco, BD) and tryptic soy agar (TSA; Difco, BD), respectively, while STEC strains were streaked on sorbitol MacConkey agar (Difco, BD). Listeria monocytogenes, Staphylococcus aureus, and lactic acid bacteria were streaked on modified Oxford, Baird-Parker, and De Man Rogosa Sharpe agar plates (Difco, BD), respectively. All the plates were incubated at 37uC for 16 to 18 h. A single colony from each plate was stabbed into nutrient soft agar (nutrient broth with 0.8% agar; Difco, BD) and served as a stock culture for the studies. Biochemical and enzymatic reactions. Selected STEC strains were analyzed for their carbohydrate metabolism and enzymatic reaction profiles using API 50 CHB/E and API ZYM test kits (bioMérieux, Hazelwood, MO). Reference strains, including O26:H11 strain 3392, O45:H2 strain 01E-1269, O103:H2 strain 2421, O111:NM strain 1665, O121:H19 strain 02E-2074, O145:NM strain G55578620, and O157:H7 strain ATCC 43895, were streaked on TSA (Difco, BD) and incubated at 37uC for 24 h. Colonies from TSA agar plates were grown in suspension, inoculated into test strips, and incubated, and the results interpreted according to the manufacturer’s recommendations. Medium formulation and preparation. The USMARC chromogenic agar medium was formulated based on the composition of MacConkey medium but without lactose and neutral red. The formulation of the selective differential medium was as follows: Bacto Peptone (BD, Franklin Lakes, NJ) at 17.0 g/liter, Proteose Peptone (BD) at 30 g/liter, sodium chloride (Sigma, St. Louis, MO) at 5.0 g/liter, crystal violet (Sigma) at 1.0 mg/liter, L-sorbose (Sigma) at 6.0 g/liter, D-raffinose (Sigma) at 6.0 g/liter, phenol red (Sigma) at 20 mg/liter, bromothymol blue (Sigma) at 1.5 mg/liter, and Bacto agar (BD) at 15 g/liter. All the ingredients were mixed well with 900 ml of distilled water, the pH adjusted to 7.4 ¡ 0.1, and the mixture autoclaved for 10 min at 115uC. The autoclaved medium was cooled to 50uC before adding 100 ml of a mixture of filter-sterilized (pH 7.1 ¡ 0.1) bile salts no. 3 (BD) at 3 g/liter, 5-bromo-4-chloro-3-indoxyl-b-Dgalactopyranoside (BCIG; Gold Biotechnology, St. Louis, MO) at 0.05 g/liter, isopropyl-b-D-thiogalactopyranoside (Sigma) at 0.05 g/ liter, novobiocin (Sigma) at 5 mg/liter, and potassium tellurite (Sigma) at 0.125 mg/liter. The concentration of potassium tellurite can be increased for purposes of more selectivity or greater inhibition of the background flora. The base agar medium, formulated with 0.125 mg/liter of potassium tellurite, was based on the good growth of serogroup O121 strains. Phenol red has a pH range from 6.8 to 8.4, changing from yellow (acid color) to red (alkali color), while bromothymol blue has a pH range from 6.0 to 7.6, changing from yellow (acid color) to blue (alkali color) (8). Growth of bacterial strains on differential medium. The differential medium described above was inoculated separately with STEC strains and non–E. coli control strains (Table 1). All strains were grown in nutrient broth for 16 to 18 h at 37uC. Each strain was streaked on USMARC chromogenic agar medium. The plates were incubated at 37uC for 24 h and then at room temperature for half an hour for full color development. Detection of STEC strains from inoculated fresh beef tissues. A cocktail mixture of STEC strains, including O26:H11 strains 3392 and 3891, O45:H2 strain 01E-1269, O45 strain WDG3, O103:H2 strain 2421, O111:NM strains 1665 and ECRC 3007:85, O121:H19 strain 02E-2074, an O121:H7 strain, and an O145:NM strain, was inoculated on fresh beef flanks. Each bacterial strain was grown in nutrient broth for 16 h at 37uC, the cell concentration was adjusted to approximately 1.5 | 10 CFU/ml using a spectrophotometer (Thermo Spectronic, Fisher Scientific, Pittsburgh, PA) at 600 nm, and the population level was confirmed using Enterobacteriaceae Count Plates (3M, St. Paul, MN) according to the manufacturer’s recommendations. Beef flanks obtained from a local cattle processing plant were aseptically cut into 96 pieces (3 replicates of 32 pieces) with an area of 25 cm (5 by 5 cm). The background bacterial population of fresh beef flanks was determined using Petrifilm aerobic count plates (3M) and was approximately 2 | 10 CFU/cm. A total of 96 pieces were cut and divided into three replicates. Each piece was randomly inoculated with a cocktail mixture to a total final concentration for all strains of approximately 3 | 10 CFU/cm and allowed to stand for 15 min at room temperature to allow bacterial cell attachment. The inoculated 5-cm beef flank square was placed in a stomacher bag and chilled for 48 h at 2 to 4uC to simulate chilling of carcasses before fabrication. Exposure to refrigeration temperature causes sublethal injury to microbial cells, which are susceptible to selective agents added to medium used for detection and/or enumeration of target organisms. Sublethal injury of bacteria can be reversed by self-repair during enrichment process in nonselective media. Therefore, an aliquot of 50 ml of tryptic soy broth (Difco, BD) was added to each chilled beef flank, and it was hand massaged and enriched at 25uC for 2 h and at 42uC for 6 h and held overnight at 4uC before immunomagnetic separation (IMS) as previously described (4). The next day, a 1-ml aliquot of the enriched samples was subjected to IMS using a mixture of Dynabeads EPEC/VTEC O26, O103, O111, and O145 (5 ml of each bead). The bacterial bead complexes were captured J. Food Prot., Vol. 76, No. 2 DIFFERENTIAL MEDIUM FOR NON-O157 E. COLI 193