A screening method was developed for the isolation of Escherichia coli O157:H7 from raw ground beef. Suspensions at a 1:10 dilution of beef were made in a modified EC broth with novobiocin (mEC+n; EC broth with 1.12 g/L instead of 1.5 g/L Bile salts #3 and novobiocin at 20 mg/L). The samples were macerated in a Stomacher for 2 min and either shaken at 37°C (100 RPM) for 6 h, or incubated static at 35°C for 24 h. Appropriate dilutions of the cultures were then spread plated on 150×15 mm plates of MacConkey sorbitol agar (MSA). The MSA plates were incubated at 42°C overnight. A set of two plates consisting of a deep (40 ml/plate) phenol red sorbitol agar plate with 4-methylumbelliferyl ß-D-glucuronide (PRS-MUG), and a Levine EMB agar plate with added agar for a final concentration of 3%, were gridded into 12 numbered sections. Sorbitol negative colonies were picked from the MSA plates, spread on the appropriate section of the EMB, and stabbed into the corresponding section on the PRS-MUG plate. Those cultures that were sorbitol negative and MUG negative on PRS-MUG and were typically E. coli on EMB were confirmed biochemically and serologically. By this procedure O157:H7 was isolated from 5 of 10 meat samples inoculated at 0.6 organisms/g, and 10 of 10 samples at the 5/g level using the 6 h shaken method. With the 24 h static incubation method, O157:H7 was isolated from 8 of 10 samples at the 0.6/g level and 10 of 10 at the 5/g level. Thirteen strains of O157:H7 inoculated at levels between 0.4 and 0.6/g were tested and 9 of the 13 were isolated with the 6 h method, and 13 of the 13 with the 24 h method. The method is reliable and simple enough to be used in large screening programs.

Five enrichment broths and five selective and differentia] plating media were tested for efficiency of isolation of Aeromonas spp. from chicken, beef and pork. An overnight incubation of sample in Trypticase soy broth containing 10 μg of ampicillin/ml which was spread on starch ampicillin agar or on MacConkey mannitol ampicillin agar, gave the best results. A small survey was conducted on 10 samples each of chicken thigh-meat, ground beef, and pork sausage or ground unseasoned pork purchased from local food stores. Aeromonads were found in all of the samples in numbers ranging from 4.44 × 10−2−>4.44× 103/g except for two of the pork products from which the organisms could not be isolated. Fifty-eight isolates from this survey were tested for hemolysin production and cytotoxin production; 36 isolates were tested for production of cholera-like toxin. Cytotoxin, as detected by mouse adrenyl Y1 cells and Chinese hamster ovary cells, was produced by 92.8% of the Aeromonas hydrophila isolates, by 84.6% of the Aeromonas sobria isolates and by 17.6% of the Aeromonas caviae isolates. Hemolysin production paralleled cytotoxin production in A. hydrophila and A. caviae. Of the A. sobria isolates, 69.2% were hemolysin producers. None of the isolates tested produced cholera-like toxin. It is not known whether the presence of cytotoxin- and hemolysin-producing Aeromonas species in retail meat and poultry has any public health significance, since to date there have been no reported outbreaks of Aeromonas-caused gastroenteritis traced to meat or poultry.

Escherichia coli 0157 specific antibody, coated on magnetic beads, was used to concentrate and remove the E. coli 0157:H7 from mixed cultures and meat samples. The problem of nontarget organism carryover was addressed by adding Protamine to the culture-bead sample, washing the beads three times in saline, and changing the test tubes with each wash. These modifications reduced the nontarget colony counts obtained from uninoculated meat samples. This procedure enabled consistent recovery of E. coli 0157:H7 from inoculated meat samples. The percentage of E. coli 0157:H7 cells captured, compared to the total number of cells captured, ranged from 48 to 100%. Two strains of E. coli 0157, H7 and :non-H7, appeared to compete with one another and thus reduce or prevent isolation.

A screening method was devised incorporating a commercially available reactive disc blot ELISA for Escherichia coli 0157 antigen, into a cultural screening program for the isolation of E. coli 0157:H7 from meat and poultry products. The method includes the inoculation of a raw or cooked meat sample into an enrichment broth, incubation with shaking at 37°C for 6 to 8 h, followed by inoculation of 3M Petrifilm™ E. coli Count plates with dilutions of the enrichment culture. The Petrifilm plates were incubated at 42°C for 18 h and tested for the presence of the 0157 antigen. The enrichment cultures were reincubated static at 35°C after the initial shaken incubation. Isolation was attempted from the positive Petrifilm plates by both a direct picking and streaking method and by the 3M Prompt™ isolation method. Isolation also was attempted from the 24-h enrichment cultures by spread plating serial dilutions on 150 × 15 mm MacConkey sorbitol agar (MSA) and MSA with 5-bromo-4-chloro-3-indoxyl-β-D-glucuronic acid cyclohexylammonium salt (BCIG). This fast and efficient screening procedure identifies negative and presumptive positive samples in 26–28 h. Isolation and confirmation of the presumptive positive isolates require an additional 3 to 4 d.

The addition of 5-bromo-4-chloro-3-indoxyl-β-D-glucuronide (BCIG) at the 0.1 g/L level, to MacConkey sorbitol agar (MSA) plates aided in the isolation of Escherichia coli 0157:H7 from raw ground beef samples by differentiating β-glucuronidase positive from β-glucuronidase negative colonies. E. coli 0157:H7 colonies, being sorbitol negative, β-glucuronidase negative, remained white, while sorbitol negative, β-glucuronidase positive colonies turned green to blue. Addition of BCIG to the MSA agar reduced the number of false suspect colonies picked from the primary plating medium by 36% when compared to MSA. E. coli 0157:H7 was isolated from 11 out of 12 inoculated meat samples (0.7 E. coli 0157:H7/g) using MSA-BCIG as compared to 8 out of 12 samples using MSA without BCIG.