Comparison of Agar Media for Detection and Quantification of Shiga Toxin-Producing Escherichia coli in Cattle Feces

Effect of potassium tellurite concentration in a chromogenic agar medium on isolation of tellurite-resistant "Top Seven" Shiga toxin-producing Escherichia coli from ground beef

The effect of potassium tellurite concentration in a chromogenic agar medium on the detection of tellurite-resistant "top seven" Shiga toxin-producing Escherichia coli (STEC) in beef was evaluated. Samples of ground beef were inoculated with tellurite-resistant STEC O26, O45, O103, O111, O121, O145, or O157 strains at geometric mean (±standard error of the mean) levels of 0, 49 (±1), 490 (±1), or 4900 (±1) CFU/10 g and enriched 1:10 (90 mL) in EC broth (40°C for 6 h). Following enrichment, aliquots of broth culture were treated by immunomagnetic separation with one of three pools of beads against STEC serogroups; pool I: O26, O45, and O121; pool II: O103, O111, and O145; and pool III: O157. After immunomagnetic separation, 50 μL of washed bead suspensions in buffered peptone water was spiral plated onto a modified Possé medium containing 0.5, 1.0, or 1.5 mg/L potassium tellurite, and incubated at 37°C for 18 h. Up to four isolated colonies were picked from each spiral plate based on expected colony phenotypes for STEC, and isolate identity was confirmed with an 11-plex PCR assay targeting the O serogroups and virulence genes. Overall, across all inoculum levels and strains, modified Possé media containing 0.5, 1.0, or 1.5 mg/L potassium tellurite each had a positive predictive value of 100%, and medium containing 0.5 mg/L potassium tellurite had numerically the highest sensitivity (100%) and negative predictive value (100%), which was significantly different from 1.5 mg/L (92.9% and 40.0%, respectively; P < 0.05). Similarly, there was an inverse relationship between potassium tellurite concentration and analytical specificity (number of colonies tested that were STEC-positive): 0.5 (1463 of 1482; 98.7%), 1.0 (1356 of 1411; 96.1%), and 1.5 mg/L (1187 of 1278; 92.9%; P < 0.05). These results suggest that 0.5 mg/L gives better performance than 1.0 or 1.5 mg/L of potassium tellurite in Possé medium for isolation of tellurite-resistant "top seven" STEC from ground beef.

A systematic review and meta-analysis of published literature on prevalence of non-O157 Shiga toxin-producing Escherichia coli serogroups (O26, O45, O103, O111, O121, and O145) and virulence genes in feces, hides, and carcasses of pre- and peri-harvest cattle worldwide

OBJECTIVE

The objective of this study was to summarize peer-reviewed literature on the prevalence and concentration of non-O157 STEC (O26, O45, O103, O111, O121, and O145) serogroups and virulence genes (stx and eae) in fecal, hide, and carcass samples in pre- and peri-harvest cattle worldwide, using a systematic review of the literature and meta-analyses.

DATA SYNTHESIS

Seventy articles were eligible for meta-analysis inclusion; data from 65 articles were subjected to random-effects meta-analysis models to yield fecal prevalence estimates. Meta-regression models were built to explore variables contributing to the between-study heterogeneity.

RESULTS

Worldwide pooled non-O157 serogroup, STEC, and EHEC fecal prevalence estimates (95% confidence interval) were 4.7% (3.4-6.3%), 0.7% (0.5-0.8%), and 1.0% (0.8-1.1%), respectively. Fecal prevalence estimates significantly differed by geographic region (P < 0.01) for each outcome classification. Meta-regression analyses identified region, cattle type, and specimen type as factors that contribute to heterogeneity for worldwide fecal prevalence estimates.

CONCLUSIONS

The prevalence of these global foodborne pathogens in the cattle reservoir is widespread and highly variable by region. The scarcity of prevalence and concentration data for hide and carcass matrices identifies a large data gap in the literature as these are the closest proxies for potential beef contamination at harvest.

Performance of Chromogenic Agar Media for Isolation of Shiga Toxin-Producing Escherichia coli from Ground Beef

ABSTRACT

The performance of three chromogenic agar media for detection of the "top seven" Shiga toxin-producing Escherichia coli (STEC) in beef was compared. Samples of retail ground beef were inoculated with STEC O26, O45, O103, O111, O121, O145, or O157 at geometric mean (±standard error of the mean) levels of 0, 48 (±1), 420 (±1), 4,100 (±1), or 45,000 (±1) CFU/10 g and enriched 1:10 (90 mL) in EC broth (40°C for 6 h). Following enrichment, aliquots of broth culture were treated by immunomagnetic separation with one of three pools of beads against the seven STEC serogroups: pool I, O26, O45, and O121; pool II, O103, O111, and O145; and pool III, O157. After immunomagnetic separation, 50 μL of washed bead suspensions in buffered peptone water were spiral plated onto modified Rainbow Agar O157 (mRBA), CHROMagar STEC (CS), or modified Possé differential medium (mPossé2) and incubated at 37°C for 18 h. Up to six isolated colonies were picked from each spiral plate based on expected colony phenotypes for STEC on the respective media, and isolate identity was confirmed with an 11-plex PCR assay targeting the O serogroups and virulence genes. Overall, mRBA had the highest sensitivity (99.2%), correctly detecting a significantly higher proportion of STEC serogroups than either CS (79.4%; P < 0.05) or mPossé2 (91.7%; P < 0.05). mRBA also had the highest negative predictive value (90.0%), correctly identifying a significantly higher proportion of true-negative samples compared with CS (25.7%; P < 0.05) and mPossé2 (46.2%; P < 0.05). However, mRBA also had the lowest analytical specificity of 83.2% (P < 0.05), yielding the lowest proportion of colonies tested that were STEC positive (3,548 of 4,263) compared with 97.7% (3,607 of 3,693) for mPossé2 and 98.0% (2,875 of 2,935) for CS. Reduced specificity results in more work and higher expense due to the increased number of colonies that must be tested. Further improvements in agar culture media for non-O157 STEC isolation are needed.

HIGHLIGHTS

Evaluation of chromogenic selective agar (CHROMagar STEC) for the direct detection of Shiga toxin-producing Escherichia coli from faecal specimens

Shiga toxin-producing Escherichia coli (STEC) are zoonotic pathogens that cause symptoms of severe gastrointestinal disease, including haemolytic uraemic syndrome (HUS), in humans. Currently in England, STEC serotypes other than O157:H7 are not cultured at the local hospital laboratories. The aim of this study was to evaluate the utility of CHROMagar STEC for the direct detection of STEC from faecal specimens in a diagnostic setting, compared to the current reference laboratory method using PCR targeting the Shiga-toxin gene (stx) to test multiple colonies cultured on MacConkey agar. Of the 292 consecutive faecal specimens submitted to the Gastrointestinal Bacterial Reference Unit that tested positive for stx by PCR, STEC could not be cultured on MacConkey agar or CHROMagar STEC from 87/292 (29.8 %). Of the 205 that were cultured, 106 (51.7 %) were detected on both MacConkey agar and CHROMagar STEC and 99 (48.3 %) were detected on MacConkey agar only. All 106 (100 %) isolates that grew on CHROMagar STEC had the ter gene cassette, known to be associated with resistance to tellurite, compared to 13/99 (13.1 %) that were not detected on CHROMagar STEC. CHROMagar STEC supported the growth of 36/40 (90 %) isolates harbouring stx2a or stx2d, the subtypes most frequently associated with progression to HUS. Of the 92 isolates harbouring eae, an important STEC virulence marker, 77 (83.7 %) grew on CHROMagar STEC. CHROMagar STEC is a useful selective media for the rapid, near-patient detection of STEC that have the potential to cause HUS.

Detection, Prevalence, and Pathogenicity of Non-O157 Shiga Toxin-Producing Escherichia coli from Cattle Hides and Carcasses

Cattle are a major reservoir for Shiga toxin-producing Escherichia coli (STEC) and harbor these bacteria in the intestinal tract. The prevalence, concentration, and STEC serogroup isolated in cattle varies between individuals. Hide removal at slaughter serves as a major point of carcass contamination and ultimately beef products. Certain STEC serogroups, such as O26, O45, O103, O111, O121, O145, and O157, containing the intestinal adherence factor intimin, pose a large economic burden to food producers because of testing and recalls. Human infection with STEC can cause illnesses ranging from diarrhea to hemorrhagic colitis and hemolytic uremic syndrome, and is commonly acquired through ingestion of contaminated foods, often beef products. Previously, most studies focused on O157 STEC, but there is growing recognition of the importance of non-O157 STEC serogroups. This review summarizes detection methods, prevalence, and methods for prediction of pathogenicity of non-O157 STEC from cattle hides and carcasses. A synthesis of procedures is outlined for general non-O157 STEC and targeted detection of specific STEC serogroups. Standardization of sample collection and processing procedures would allow for more robust comparisons among studies. Presence of non-O157 STEC isolated from cattle hides and carcasses and specific factors, such as point of sample collection and season, are summarized. Also, factors that might influence STEC survival on these surfaces, such as the microbial population on hides and microbial adherence genes, are raised as topics for future investigation. Finally, this review gives an overview on studies that have used genetic and cell-based methods to identify specific phenotypes of non-O157 STEC strains isolated from cattle to assess their risk to human health.

Variability in Characterizing Escherichia coli from Cattle Feces: A Cautionary Tale

Shiga toxin-producing Escherichia coli (STEC) are diverse bacteria, with seven serogroups (O26, O45, O103, O111, O121, O145, O157; "Top 7") of interest due to their predominance in human disease. Confirmation of STEC relies on a combination of culturing, immunological and molecular assays, but no single gold standard for identification exists. In this study, we compared analysis of STEC between three independent laboratories (LAB) using different methodologies. In LAB A, colonies of Top 7 were picked after serogroup-specific immunomagnetic separation of feces from western-Canadian slaughter cattle. A fraction of each colony was tested by PCR (stx1, stx2, eae, O group), and Top 7 isolates were saved as glycerol stocks (n = 689). In LAB B, a subsample of isolates (n = 171) were evaluated for stx1 and stx2 using different primer sets. For this, approximately half of the PCR were performed using original DNA template provided by LAB A and half using DNA extracted from sub-cultured isolates. All Top 7 isolates were sub-cultured by LAB A and shipped to LAB C for traditional serotyping (TS) to determine O and H groups, with PCR-confirmation of virulence genes using a third set of primers. By TS, 76% of O groups (525/689) matched PCR-determined O groups. Lowest proportions (p < 0.05) of O group matches between PCR and TS (62.6% and 69.8%) occurred for O26 and O45 serogroups, respectively. PCR-detection of stx differed most between LAB A and LAB C. Excluding isolates where O groups by PCR and TS did not match, detection of stx1 was most consistent (p < 0.01) for O111 and O157:H7/NM. In contrast, for O45 and O103, stx1 was detected in >65% of isolates by LAB A and <5% by LAB C. Stx2 was only detected by LAB C in isolates of serogroups O121, O145, and O157:H7/NM. LAB B also detected stx2 in O26 and O157:H12/H29, while LAB A detected stx2 in all serogroups. Excluding O111 and O157:H7/NM, marked changes in stx detection were observed between initial isolation and sub-cultures of the same isolate. While multiple explanations exist for discordant O-typing between PCR and TS and for differences in stx detection across labs, these data suggest that assays for STEC classification may require re-evaluation and/or standardization.

Tellurite resistance profiles and performance of different chromogenic agars for detection of non-O157 Shiga toxin-producing Escherichia coli

Shiga toxin-producing Escherichia coli (STEC) are globally important food-borne pathogens. The isolation of non-O157 STEC is a significant public health challenge due to the dramatic diversity of their phenotypes and genotypes. In the present study, 476 non-O157 STEC strains representing 95 different O-serogroups were used to evaluate tellurite resistance and the performance of 12 different chromogenic agars. Of 476 strains, only 108 (22.7%) strains showed the minimal inhibitory concentration (MIC) values for potassium tellurite being higher than 4μg/ml, and 96 (20.2%) strains harbored intact ter genes cluster. The presence of ter genes was significantly correlated with tellurite resistance. Six commercial chromogenic agars (TBX, MAC, SMAC, Rainbow® Agar O157, CHROMagar™ ECC, and Fluorocult O157) supported the growth of all strains. However, CT-SMAC, CHROMagar™ O157, and CHROMagar™ STEC agars exhibited 12.2%, 31.1%, and 38.0% of growth inhibition, respectively. Furthermore, 4.6%, 33.2%, and 45.0% of strains were inhibited on RBA-USDA, RBA-NT, and BCM O157 agar media. Variations in tellurite resistance and colony appearance might result in discrepant performance of non-O157 STEC recovery from different chromogenic agars. Using inclusive agars or less selective agar in combination with highly selective agar should be suggested to recover most non-O157 STEC strains, which would increase the probability of recovering STECs from complex background microflora.

Culture-Based Quantification with Molecular Characterization of Non-O157 and O157 Enterohemorrhagic Escherichia coli Isolates from Rectoanal Mucosal Swabs of Feedlot Cattle

Enterohemorrhagic Escherichia coli (EHEC) strains are foodborne pathogens carried in the intestinal tracts of ruminants and shed in the feces. High concentrations (≥10⁴ colony-forming units [CFU]/g) of EHEC in cattle feces are associated with contamination of hides, and subsequently, carcasses and beef. Several studies using agar media have quantified O157 but few have quantified non-O157 EHEC in samples from cattle. Thus, the objective of this study was to determine the concentration of O157 and non-O157 EHEC in cattle, and to characterize the associated EHEC isolates for their virulence potential. Two hundred feedlot steers were sampled by rectoanal mucosal swab (RAMS) every 35 days over four sampling periods, and a spiral plating method using modified Possé differential agar was used to quantify EHEC organisms in these samples. Bacterial colonies from agar plates were tested by multiplex PCR for Shiga toxin and intimin genes (stx and eae, respectively), and confirmed EHEC isolates (i.e., positive for both stx and eae) were serotyped and characterized for virulence genes using a microarray. Organisms detected in this study included O26, O101, O103, O109, O121, O145, O157, and O177 EHEC, with all except O121 quantifiable and measuring within a range from 9.0 × 10² to 3.0 × 10⁵ CFU/g of RAMS sample. Organisms of the same EHEC serogroup were not detected in quantifiable concentrations from a single animal more than once. EHEC organisms most commonly detected at quantifiable levels were O26, O157, and O177. Interestingly, O26 EHEC isolates tested negative for stx₁ but positive for stx_2a. High concentrations of EHEC were detected in 11 (5.5%) of the steers at least once over the sampling period. These results indicate that in addition to O157, non-O157 EHEC are transiently present in high concentrations in the rectoanal mucosal region of cattle.

A randomized controlled trial to evaluate the effects of dietary fibre from distillers grains on enterohemorrhagic Escherichia coli detection from the rectoanal mucosa and hides of feedlot steers

Feeding high levels (≥40% dry matter) of distillers grains may increase the risk for cattle to carry enterohemorrhagic Escherichia coli (EHEC) O157. The mechanism for the increased risk is not known nor whether non-O157 EHEC are similarly affected. Our objective was to test whether the fibre content or other components of modified distillers grains plus solubles (MDGS) affects the probability for cattle to carry EHEC serogroups of public health importance. A 2 × 2 plus 1 factorial treatment arrangement within a randomized block design was utilized. Within each of four blocks, 25 feedlot pens (n = 8 steers/pen) were assigned randomly to (i) corn-based control diet; (ii) 20% dry matter (DM) MDGS; (iii) 40% DM MDGS; (iv) corn bran added to corn-based diet to match fibre of 20% MDGS or (v) 40% MDGS. Rectoanal mucosa swabs (RAMS) were collected on day (d)0, d35, d70 and d105; hide swabs were collected on the last feeding day. Samples were tested for EHEC by a molecular screening assay. The effects of fibre source and fibre level on EHEC carriage were tested using multilevel logistic regression (generalized linear mixed models; α = 0.05). EHEC O45 RAMS detection was associated with fibre level, source and sampling day. EHEC O103 RAMS detection increased by feeding 40% MDGS but not the corresponding corn bran diet. Hide contamination by EHEC O45 or O103 was less likely in cattle fed MDGS compared to corn bran diets. EHEC O111 RAMS detection decreased by feeding 40% MDGS but not by feeding the corresponding corn bran diet. Detection of EHEC O157 or O145 was not associated with dietary factors. Feeding 40% MDGS increased the probability for carriage of some EHEC serogroups but decreased probability of others, which indicated that EHEC serogroups have different risk factors associated with feeding MDGS and little association with dietary fibre.

Spiral Plating Method To Quantify the Six Major Non-O157 Escherichia coli Serogroups in Cattle Feces

Cattle are a major reservoir of the six major Shiga toxin-producing non-O157 Escherichia coli (STEC) serogroups (O26, O45, O103, O111, O121, and O145) responsible for foodborne illnesses in humans. Besides prevalence in feces, the concentrations of STEC in cattle feces play a major role in their transmission dynamics. A subset of cattle, referred to as super shedders, shed E. coli O157 at high concentrations (≥4 log CFU/g of feces). It is not known whether a similar pattern of fecal shedding exists for non-O157. Our objectives were to initially validate the spiral plating method to quantify the six non-O157 E. coli serogroups with pure cultures and culture-spiked fecal samples and then determine the applicability of the method and compare it with multiplex quantitative PCR (mqPCR) assays for the quantification of the six non-O157 E. coli serogroups in cattle fecal samples collected from commercial feedlots. Quantification limits of the spiral plating method were 3 log, 3 to 4 log, and 3 to 5 log CFU/mL or CFU/g for individual cultures, pooled pure cultures, and cattle fecal samples spiked with pooled pure cultures, respectively. Of the 1,152 cattle fecal samples tested from eight commercial feedlots, 122 (10.6%) and 320 (27.8%) harbored concentrations ≥4 log CFU/g of one or more of the six serogroups of non-O157 by spiral plating and mqPCR methods, respectively. A majority of quantifiable samples, detected by either spiral plating (135 of 137, 98.5%) or mqPCR (239 of 320, 74.7%), were shedding only one serogroup. Only one of the quantifiable samples was positive for a serogroup carrying Shiga toxin (stx1) and intimin (eae) genes; 38 samples were positive for serogroups carrying the intimin gene. In conclusion, the spiral plating method can be used to quantify non-O157 serogroups in cattle feces, and our study identified a subset of cattle that was super shedders of non-O157 E. coli. The method has the advantage of quantifying non-O157 STEC, unlike mqPCR that quantifies serogroups only.

Feedlot- and Pen-Level Prevalence of Enterohemorrhagic Escherichia coli in Feces of Commercial Feedlot Cattle in Two Major U.S. Cattle Feeding Areas

The objective of this study was to determine feedlot- and pen-level fecal prevalence of seven enterohemorrhagic Escherichia coli (EHEC) belonging to serogroups (O26, O45, O103, O111, O121, O145, and O157, or EHEC-7) in feces of feedlot cattle in two feeding areas in the United States. Cattle pens from four commercial feedlots in each of the two major U.S. beef cattle areas were sampled. Up to 16 pen-floor fecal samples were collected from each of 4-6 pens per feedlot, monthly, for a total of three visits per feedlot, from June to August, 2014. Culture procedures including fecal enrichment in E. coli broth, immunomagnetic separation, and plating on selective media, followed by confirmation through polymerase chain reaction (PCR) testing, were conducted. Generalized linear mixed models were fitted to estimate feedlot-, pen-, and sample-level fecal prevalence of EHEC-7 and to evaluate associations between potential demographic and management risk factors with feedlot and within-pen prevalence of EHEC-7. All study feedlots and 31.0% of the study pens had at least one non-O157 EHEC-positive fecal sample, whereas 62.4% of pens tested positive for EHEC O157; sample-level prevalence estimates ranged from 0.0% for EHEC O121 to 18.7% for EHEC O157. Within-pen prevalence of EHEC O157 varied significantly by sampling month; similarly within-pen prevalence of non-O157 EHEC varied significantly by month and by the sex composition of the pen (heifer, steer, or mixed). Feedlot management factors, however, were not significantly associated with fecal prevalence of EHEC-7. Intraclass correlation coefficients for EHEC-7 models indicated that most of the variation occurred between pens, rather than within pens, or between feedlots. Hence, the potential combination of preharvest interventions and pen-level management strategies may have positive food safety impacts downstream along the beef chain.