2851 Journal of Food Protection, Vol. 69, No. 12, 2006, Pages 2851–2855 Copyright 䊚, International Association for Food Protection

Survival of Escherichia coli O157:H7 in Bovine Feces over Time under Various Temperature Conditions A. ECHEVERRY,1 G. H. LONERAGAN,2 1Department

AND

M. M. BRASHEARS1*

of Animal and Food Sciences, Texas Tech University, Box 42141, Lubbock, Texas 79409; and 2Division of Agriculture, West Texas A&M University, WTAMU Box 60998, Canyon, Texas 79016, USA MS 06-167: Received 21 March 2006/Accepted 8 July 2006

ABSTRACT Although Escherichia coli O157:H7 prevalence estimates in cattle have increased over time due to improvements in detection methods, fecal sample transport conditions from farm to microbiological laboratories for further analysis may be a factor for prevalence underestimation. The objective of this study was to compare and determine the survival characteristics of E. coli O157:H7 in bovine feces under various storage conditions that could be encountered during transport. Fecal pats were inoculated with a four-strain cocktail of antibiotic-resistant E. coli O157:H7 to contain ⬃1 ⫻ 105 CFU/g. Inoculated and control samples were taken after 0, 24, 48, 120, and 168 h at each storage temperature and examined for presence and numbers of E. coli O157:H7. Each sample was subdivided and placed at each of the four following temperatures: 37, 23, 4.4⬚C, and in a plastic cooler with refrigerant packs (0, 4, 4, 21, and 23⬚C at five sampling times, respectively) to simulate transportation conditions. A statistically significant decrease in the population of the pathogen was observed after 48 h in samples held at 37⬚C (P ⬍ 0.01) and after 168 h at 4.4⬚C (P ⫽ 0.02). At 37⬚C, E. coli O157:H7 was not detected after 48 h, either by direct plating (P ⬍ 0.01) or by immunomagnetic separation. Overall, the results of this study showed that E. coli O157:H7 survived without significant detriment in bovine fecal material inside the cooler for up to 168 h. These results indicate that shipment and storage under these conditions before microbiological analysis would be acceptable and should not affect pathogen detection.

Escherichia coli O157:H7 is considered a serious threat to public health in developed countries. In the United States alone, E. coli O157:H7 is the single greatest cause of hemorrhagic colitis and hemolytic uremic syndrome in children (1, 27). According to the Centers for Disease Control and Prevention, nearly 74,000 cases and 61 deaths annually are attributable to this pathogen (7). E. coli O157:H7 is widely prevalent in beef feedlot cattle (24). In recent studies, detection of E. coli O157:H7 in environmental samples has been possible because of the use of sensitive methods that allow detection of the pathogen even when present in very small numbers. Studies have shown that up to 38.6% of individual animals (8), 72% of lots (11), and 100% of feedlots (24) have at least one sample that tested positive for the pathogen, suggesting that E. coli O157 is more common in groups of animals and feedyards than previously reported. E. coli O157:H7 can enter the meat supply during slaughter from the hides, dust in the environment, or by direct contact with feces or digesta from the intestinal tract during processing (3, 6, 22). Decreasing foodborne pathogens in the animal, on the hides and in the preharvest environment should decrease the occurrence of the pathogen in the meat supply and protect consumers from foodborne illness. Although several preharvest intervention methods have been investigated (5, 16, 26, 30), the methods associated * Author for correspondence. Tel: 806-742-2469; Fax: 806-742-2427; E-mail: [email protected].

with these studies have varied and it is not possible to accurately compare the efficacy of the interventions in relationship to one another. Some differences among studies include the type of sample used, timing of sample collection, effects of other pathogens and background flora, effects of transportation conditions on animals’ shedding, and the methods surrounding the isolation, identification, and quantification of the pathogen (2, 3, 9, 12, 14, 22, 23, 25, 29). Although E. coli O157:H7 prevalence estimates in cattle have increased over time due to improvements in diagnostic and detection methodologies (4, 9, 22), transport conditions of the fecal samples to the microbiological laboratories for their further analysis may also affect survival and, therefore, prevalence. Prevalence underestimation of the pathogen may be occurring as a direct consequence of sample handling and storage conditions. The objectives of this study were to determine the survival characteristics of E. coli O157:H7 in bovine feces under various storage conditions that could be encountered during transport. We simulated and focused on the transportation and handling conditions that fecal samples undergo during shipping in order to study the effect that different temperatures and length of storage have on the pathogen’s population. MATERIALS AND METHODS Study design. This study had two treatments (inoculated and control) with four storage conditions analyzed over time making it a 2 ⫻ 4 factorial experimental design. One fresh fecal pat was

2852

ECHEVERRY ET AL.

used for each of the treatments with three replicates conducted for both treatments. Fecal pats were collected on separate occasions and all fecal samples were initially tested to determine whether they were positive for E. coli O157. Randomization was performed when assigning to the fecal material either to the control or to the inoculated treatment and immediately after mixing and subdividing the sample for storage at each of the four temperatures. Sample collection. Fresh fecal samples collected from the ground were aseptically obtained from cattle housed at the Texas Tech University Burnett Center facilities in New Deal. At each replication, two fresh fecal samples, each weighing approximately 1,000 g, were obtained directly from the pen using sterile, disposable latex gloves and placed in sterile plastic bags. After collection, bags were stored in a plastic cooler with ice packs, transported directly to the Texas Tech University Food Microbiology Laboratory, and examined within 30 to 60 min after collection. Preparation of pathogens. A cocktail of four different streptomycin-resistant (1,000 ␮g/ml) E. coli O157:H7 strains (920, 922, 944, 966) from the Texas Tech University Food Microbiology Laboratory was used. Antibiotic-resistant strains were used to facilitate recovery in the presence of background flora. All E. coli O157:H7 strains were originally isolated from cattle. Stock cultures maintained in tryptic soy broth (TSB; Difco, Becton Dickinson, Sparks, Md.) at ⫺80⬚C were grown separately and subcultured twice in TSB containing streptomycin (1,000 ␮g/ml) for 24 h at 37⬚C. Equal volumes were then combined to obtain the fourstrain cocktail for inoculation. Inoculated sample preparation. Immediately after arrival, 10 g of fecal material from each initial sample were tested for naturally present E. coli O157 using an immunomagnetic separation (IMS) method (described below). While performing this microbiological analysis, fecal material remained under refrigeration (5⬚C) for a maximum of 3 days. If negative E. coli O157 results were obtained, 99 ml of buffered peptone water (BPW; Difco, Becton Dickinson) containing 11 ml of the previously described four-strain cocktail was added to 1,000 g of feces and mixed manually for 2 min to evenly distribute the inoculum. After mixing, the inoculated sample was divided into four 250-g subsamples using sterile spatulas and placed in sterile Whirl-Pak bags for each of the following storage conditions: (i) 37⬚C, (ii) room temperature at 23⬚C, (iii) a cooler, and (iv) refrigeration at 4.4⬚C. Control samples were prepared simultaneously as inoculated samples by adding 110 ml of BPW to 1,000 g of feces and mixed manually with special attention given to prevent cross-contamination. Noninoculated manure was also divided in four subsamples, randomly assigned to the same storage conditions, and placed into sterile Whirl-Pak bags as described previously. Microbial analysis for detection of E. coli O157:H7. The IMS protocol used in this study for the isolation and identification of E. coli O157:H7 has been described previously (5, 11). This method was used to detect E. coli O157 in noninoculated as well as inoculated and control samples after initial negative results. Briefly, immediately after subdividing the fecal samples, 10 g of feces for each treatment and control were enriched in 90 ml of GN broth (Difco, Becton Dickinson) supplemented with 10 ␮g/ liter vancomycin (Sigma-Aldrich, St. Louis, Mo.), 8 ␮g/liter cefsulodin (Sigma-Aldrich), and 50 ␮g/liter cefixime (Dynal, Lake Success, N.Y.) (VCC-GNBroth) and incubated at 37⬚C for 6 h. After preenrichment, 1 ml of each subsample, control and inoculated treatment, was analyzed using IMS with 20 ␮l of Dynabeads anti–E. coli O157 (Dynal Biotech, Oslo, Norway) for 30

J. Food Prot., Vol. 69, No. 12

min at room temperature. The sample was washed three times in phosphate-buffered saline with Tween 20 (PBST; Sigma-Aldrich) while subjected to a magnetic field via a Dynal magnetic concentrator in order to separate and capture the bacteria attached to the paramagnetic beads (4, 8). The beads were then resuspended in 0.05 ml of the same buffer, vortexed, and spread onto sorbitol MacConkey agar plates (SMAC; Difco, Becton Dickinson) supplemented with cefixime (0.05 ␮l/liter) and potassium tellurite (50 ␮l/liter) (CT-SMAC) and incubated at 37⬚C for 18 to 24 h in an aerobic chamber (8, 18, 21). After incubation, two typical colorless sorbitol-negative colonies per plate were picked and streaked for isolation on CTSMAC and placed overnight at 37⬚C. A single sorbitol-nonfermenting colony was selected and subcultured onto MacConkey (MAC; Difco, Becton Dickinson) and Fluorocult E. coli O157:H7 agars (Fluorocult; EM Science, Gibbstown, N.J.) and incubated overnight at 37⬚C (17). After incubation, pink lactose-positive, 4methylumbelliferyl-␤-D-glucuronide (MUG) negative colonies (not fluorescent under UV light) were transferred from the MAC plates into TSB (Sigma-Aldrich), triple sugar iron agar tubes (TSI; Difco, Becton Dickinson), and MacConkey broth (MACb; Difco, Becton Dickinson) for further testing and incubated overnight at 37⬚C. Colonies that were indole positive on TSB, dextrose, lactose and/or sucrose positive with gas production on TSI, and Gramnegative lactose fermenting on MACb were considered as presumptive positive E. coli O157:H7. For confirmation, presumptive colonies from the MAC plates stored at 4.4⬚C for less than 24 h were subjected to a commercial latex agglutination test specific for the O157:H7 antigen (Remel, Lenexa, Kans.) (5, 19). Final confirmation of presumptive positive samples was made by polymerase chain reaction using the BAX (Dupont Qualicon, Wilmington, Del.) system. Quantification of E. coli O157:H7 by direct plating. Immediately after subdividing both inoculated and control samples, 11 g of feces were weighed into stomacher filter bags containing 99 ml of BPW and homogenized in a Stomacher 400 Laboratory blender (Seward Medical, London, UK) for 2 min. E. coli O157: H7 counts were determined by duplicate plating of appropriate dilutions using an autoplater (Autoplate 4000, Spiral Biotech, Norwood, Mass.). Thereafter, 50 ␮l of the suspension was spread onto plates of EC medium containing MUG (ECMUG; Difco) supplemented with streptomycin (1,000 ␮g/liter) and incubated at 37⬚C for 18 to 24 h. Following incubation, all plates were counted using a Q-count (Spiral Biotech), and then examined under UV light to detect MUG-negative colonies. Up to 10 MUG-negative colonies were selected per plate, streaked for isolation on MAC agar plates, and incubated overnight at 37⬚C. After incubation pink, lactose-positive colonies from each plate were subjected to a latex agglutination test for final confirmation (18, 19). Both selective media and the use of antibiotic resistant E. coli O157:H7 were necessary to enumerate the pathogen in the presence of background flora. Sampling times and temperature conditions. As detailed previously, fecal samples were tested for the presence of naturally occurring E. coli O157 before inoculation. Qualitative (IMS) and quantitative (direct plating) analyses for E. coli O157:H7 were conducted for each subsample in every replication at times 0 (immediately after subdividing samples), 24, 48, 120, and 168 h. Inoculated and control subsamples were stored separately under the following conditions: incubator (37⬚C), room temperature (23⬚C), refrigeration (4.4⬚C), and inside a insulated cooler (model 9482, Rubbermaid; 21.75 by 13.625 by 15.5 in.). The samples inside the plastic cooler were covered with 10 refrigerant packs

J. Food Prot., Vol. 69, No. 12

SURVIVAL OF E. COLI O157:H7 IN BOVINE FECES OVER TIME

(7.5 by 6.5 by 1 in. each) (Utek, Polyfoam Packers Corporation, Wheeling, Ill.) that were not replaced during the course of the study. Average temperatures inside the cooler were 0, 4, 4, 21, and 23⬚C at times 0, 24, 48, 120, and 168 h, respectively. Statistical analysis. All experiments were performed three times for each temperature condition. After the counts were obtained for each subsample, corrections were performed (i) by taking into account the exact weight of each fecal sample before addition of BPW, and (ii) the percentage of colonies that were by the agglutination test before transforming the results into log data. To analyze statistically significant differences in the survival of E. coli O157:H7 among the different treatments (the effect of each temperature on bacterial growth), the data were analyzed using the Proc Mixed Procedure of the SAS system (SAS System for Windows, release 8.2, SAS Institute, Cary, N.C.). The dependent variable of interest was the bacterial population at a given time (log CFU). The initial model included treatment and time and its interaction as independent variables. Replicate and its interactions were included as random variables and first-order covariance matrices were used to account for the repeated-measures nature of the study. Because significant differences were expected between controls and inoculated samples and the primary objective was to evaluate various storage conditions on positive samples, the control samples were excluded from a separate analysis. This model was similar to the one described above. If a significant interaction was detected, means within days were separated using multiple comparisons of means techniques.

RESULTS AND DISCUSSION Studies on the survival characteristics of E. coli O157: H7 in bovine feces and manure are important because during epidemiological surveys or studies analyzing the effectiveness of preharvest interventions that are used to reduce the pathogen, accurate results are required in order to assess the status of the animals, pens, or feedlots before and after the investigation. However, pathogen survival can be affected by shipping and handling, and therefore, the outcome of the investigations can also be influenced by these results. Various other conditions such as temperature, time, pH, solid content, aeration, bacterial concentration, and moisture can reportedly influence pathogen survival in fecal material (16, 22, 28). A few studies report the survival characteristics of E. coli O157:H7 in bovine feces. In one of the earliest studies, pathogen survival rates were affected by the temperature at which the samples were held, the inoculum level at the beginning of the experiment, and perhaps the moisture content of the fecal sample (28). In that study, inoculated samples were stored under three different environmental conditions. Although no growth was observed at 5⬚C, the pathogen was able to survive for 70 days. Although samples were not held this long in our experiments because transportation and holding of samples for analysis would not be expected to take longer than a week, our results are similar in that the pathogen survived for the course of the study at 4.4⬚C. Other studies have revealed similar long-term survival rates for E. coli O157:H7 in bovine feces when held under similar conditions. Fukushima et al. (13) detected the pathogen in samples collected just after excretion from 4 to 8

2853

TABLE 1. Overall least square mean values for the population of antibiotic-resistant E. coli O157:H7 in inoculated bovine feces held under different temperature conditionsa Mean values (log CFU/g) at sampling time: Treatments

Coolerb 4.4⬚C 23⬚C 37⬚C

0h

4.91 5.06 5.04 5.01

Ac A A A

24 h

5.00 5.03 4.99 4.71

A A A A

48 h

5.06 5.06 5.04 3.44

A A A B

120 h

168 h

5.05 A 4.81 A 4.99 A ND C

5.02 A 4.69 A 4.87 A ND C

a

ND, not detectable. E. coli O157:H7 colonies were not recovered by plating or IMS. b Cooler: The average temperatures inside the cooler were 0, 4, 4, 21, and 23⬚C at times 0, 24, 48, 120, and 168 h, respectively. c Rows with different letters differ (P ⬍ 0.05).

weeks at 25⬚C and for 16 to 18 weeks at 15⬚C. Kudva et al. (15) reported a 2-log reduction in populations after 48 h at 4⬚C with the population stabilized thereafter. A similar trend was observed in our samples; however, the initial reduction was not as large as that previously reported, probably due the higher inoculum (6.6 ⫻ 109 CFU/g) in that study (15). Kudva et al. (15) also reported no viable bacteria after approximately 5 days when samples were held at 37⬚C. In our experiment, subsamples were analyzed to simultaneously detect the presence of E. coli O157:H7 with IMS and to quantify the population of the pathogen under different storage conditions. For each treatment, least square mean estimates of average E. coli O157:H7 fecal concentration (log CFU per gram) are presented in Table 1. The average initial E. coli O157:H7 inoculum in bovine feces was 5.01 log CFU/g, and ranged from 4.91 to 5.06 log CFU/g with significant differences in initial populations. When the uninoculated samples were removed from the model, a significant interaction between treatment and time was detected (P ⬍ 0.05). No treatment effect was detected at 0 h (P ⫽ 0.88) or 24 h (P ⫽ 0.37); however, a measurable treatment effect was observed at 48 h (P ⬍ 0.01), 120 h (P ⬍ 0.01), and 168 h (P ⬍ 0.01). For all comparisons at 48, 120, and 168 h, E. coli O157:H7 counts were greater at 4.4, 23⬚C, and in the cooler than those obtained in the incubator (P ⬍ 0.01). Across time points, no time effect was detected for the cooler (P ⫽ 0.70), 4.4⬚C (P ⫽ 0.14), or 23⬚C (P ⫽ 0.43). A time effect was detected for those samples stored at 37⬚C (P ⬍ 0.01). Concentrations at 48, 120, and 168 h were lower in samples held at 37⬚C than those at 0 and 24 h (P ⬍ 0.01 for all comparisons). In the cooler, E. coli O157:H7 survived during the entire study with very few differences observed. In addition to the quantitative data, the pathogen was detected in samples held in the cooler using IMS from 0 to 168 h. The numbers of E. coli O157:H7 enumerated from 0 to 168 h varied during the study, and although a maximum increase was observable at 48 h, this result was not significant (P ⫽ 0.73) As a way to simulate transportation conditions that samples could encounter during shipping, ice packs inside

2854

ECHEVERRY ET AL.

the cooler were not replaced in order to measure the internal temperature change inside the container. Temperatures were 0, 4, 4, 21, and 23⬚C at 0, 24, 48, 120, and 168 h, respectively. This slow, but continuous increase in the temperature inside the container may have allowed the pathogen to increase in numbers during the study but are likely not practical or affecting enumeration studies. At 4.4⬚C, a slight decrease in the population of E. coli O157:H7 was observed by the end of the study, but this decrease was not statistically significant (P ⫽ 0.14). At 23⬚C, there was no observable E. coli O157:H7 growth in the samples. At room temperature, a trend toward a decrease in population was observed; however, it was not statistically significant at any sampling time. The pathogen was detected by IMS in the control samples from 0 to 120 h; however, none of the control samples was positive at 168 h. Although IMS results were positive for the controls after preenrichment, none of the samples was positive using enumeration techniques. Therefore, we concluded that the number of naturally present E. coli O157 would not have impacted the enumeration results and were excluded from the analysis. At 37⬚C, the population of E. coli O157:H7 decreased from 5.05 log CFU/g at 0 h to 4.72 log CFU/g (P ⬍ 0.0001) at 24 h. After 48 h, the population decreased dramatically by 1.57 log CFU/g (P ⬍ 0.01). After 120 and 168 h, the pathogen was not recovered by direct plating (P ⬍ 0.01) or IMS following preenrichment. Some control samples at 37⬚C were positive using IMS, at 0 h on replications 1 and 2, and at 24 h on replications 1 and 3; however, none of the incubated control samples was positive after 120 h. One of the most surprising results was that obtained from the control fecal samples. Despite being analyzed by IMS for the pathogen before addition of BPW, once the experiment started some control samples yielded positive IMS results, indicating that E. coli O157 was naturally present in the samples at the time of collection, but perhaps at undetectable levels during initial IMS screening. Small numbers of the pathogen would not be present in all samples, thus indicating that larger sample sizes are needed to adequately determine the natural presence in bovine feces. These pathogens also could have been injured when the initial samples were collected and recovered during storage; however, because the samples were collected promptly after the animal defecated, this would appear likely. Although IMS detected the pathogen, the enumeration methods were insufficiently sensitive and would not have impacted our final enumeration results. Another possibility is that the pathogen was not evenly distributed in the fecal pat as some authors have suggested (10, 20), causing us to believe its absence during initial IMS detection in the control samples but being able to be recovered from the entire fecal sample after addition of BPW and homogenization. The sensitivity of IMS allows injured bacterial cells to recover and increase in number if present in the sample, allowing higher rates of recovery than when other methods such as direct plating are used and accounts for detection even when no counts

J. Food Prot., Vol. 69, No. 12

were obtained from the same subsamples when examined quantitatively (8, 12). Overall, our results show that the pathogen was recovered by direct plating and with the use of IMS for up to 168 h (7 days) when all of the temperatures were equal or below 23⬚C; but no colonies were recovered after 120 h when bovine fecal samples where incubated at 37⬚C, maybe because the pathogen is entering the decline phase of the growth curve. Cooler conditions allowed recovery of the pathogen for the duration of the study, and from 24 to 168 h the population even increased as the temperature inside the container increased during holding, indicating that shipping and handling of fecal samples under these conditions before processing are suitable. These findings show that survival and detection of E. coli O157:H7 in feces is not affected by shipping and refrigeration of the samples if processed within 7 days. REFERENCES 1.

2.

3.

4.

5.

6.

7.

8.

9.

10.

11.

12.

Banatvala, N., P. M. Griffin, K. D. Greeene, T. J. Barret, W. F. Bibb, J. H. Green, and J. G. Wells. 2001. The United States national prospective hemolytic uremic syndrome study: microbiologic, serologic, clinical, and epidemiologic findings. J. Infect. Dis. 183:1063–1073. Barham, A. R., B. L. Barham, A. K. Johnson, D. M. Allen, J. R. Blanton, Jr., and M. F. Miller. 2002. Effects of the transportation of beef cattle from the feedyard to the packing plant on prevalence levels of Escherichia coli O157 and Salmonella spp. J. Food Prot. 65:280–283. Barkocy-Gallagher, G. A., T. M. Arthur, M. Rivera-Betancourt, X. Nou, S. D. Shackelford, T. L. Wheeler, and M. Koohmaraie. 2003. Seasonal prevalence of shiga toxin-producing Escherichia coli O157: H7 and non-157 serotypes, and Salmonella in commercial beef processing plants. J. Food Prot. 66:1978–1986. Barkocy-Gallagher, G. A., E. D. Berry, M. Rivera-Betancourt, T. M. Arthur, X. Nou, and M. Koohmaraie. 2002. Development of methods for the recovery of Escherichia coli O157:H7 and Salmonella from beef carcass sponge samples and bovine fecal and hide samples. J. Food Prot. 65:1527–1534. Brashears, M. M., M. L. Galyean, G. H. Loneragan, J. E. Mann, and K. Killinger-Mann. 2003. Prevalence of Escherichia coli O157:H7 and performance by beef feedlot cattle given Lactobacillus directfed microbials. J. Food Prot. 66:748–754. Byrne, C. M., D. J. Bolton, J. J. Sheridan, D. A. McDowell, and I. S. Blair. 2000. The effects of preslaughter washing on the reduction of Escherichia coli O157:H7 transfer from cattle hides to carcasses during slaughter. Lett. Appl. Microbiol. 30:142–145. Centers for Disease Control and Prevention. 2003. Escherichia coli O157:H7: general information. Last reviewed 3 September 2003. Available at: http://www.cdc.gov/ncidod/dbmd/diseaseinfo/escherichiacoli㛮g. htm. Accessed 21 October 2003. Chapman, P. A., D. J. Wright, and C. A. Siddons. 1994. A comparison of immunomagnetic separation and direct culture for the isolation of verocytotoxin-producing Escherichia coli O157 from bovine faeces. J. Med. Microbiol. 40:424–427 De Boer, E., and A. E. Heuvelink. 2000. Methods for the detection and isolation of shiga toxin-producing Escherichia coli. J. Appl. Microbiol. Symp. Supp. 2000 88:133–143. Echeverry, A., M. M. Brashears, G. H. Loneragan, and B. A. Wagner. 2005. Effect of intensity of fecal pat sampling on estimates of Escherichia coli O157. Am. J. Vet. Res. 66:2023–2027. Elder, R. O., J. E. Keen, G. R. Siragusa, G. A. Barkocy-Gallagher, M. Koohmaraie, and W. W. Laegreid. 2000. Correlation of enterohemorrhagic Escherichia coli O157 prevalence in feces, hides, and carcasses of beef cattle during processing. Proc. Natl. Acad. Sci. USA 97:2999–3003. Foster, G., G. F. Hopkins, G. J. Gunn, H. E. Ternent, F. ThompsonCarter, H. I. Knight, D. J. L. Graham, V. Edge, and B. A. Synge.

J. Food Prot., Vol. 69, No. 12

13.

14.

15.

16.

17. 18.

19.

20.

21.

22.

SURVIVAL OF E. COLI O157:H7 IN BOVINE FECES OVER TIME

2003. A comparison of two pre-enrichment media prior to immunomagnetic separation for the isolation of E. coli O157 from bovine faeces. J. Appl. Microbiol. 95:155–159. Fukushima, H., K. Hoshina, and M. Gomyoda. 1999. Long-term survival of shiga toxin-producing Escherichia coli O26, O111, and O157 in bovine feces. Appl. Environ. Microbiol. 65:5177–5181. Johnson, J. L., C. L. Brooke, and S. J. Fritschel. 1998. Comparison of the BAX for screening/E. coli O157:H7 method with conventional methods for detection of extremely low levels of Escherichia coli O157:H7 in ground beef. Appl. Environ. Microbiol. 64:4390–4395. Kudva, I. T., K. Blanch, and C. J. Hovde. 1998. Analysis of Escherichia coli O157:H7 survival in ovine or bovine manure and manure slurry. Appl. Environ. Microbiol. 64:3166–3174. Kudva, I. T., S. Jelacic, P. I. Tarr, P. Youderian, and C. J. Hovde. 1999. Biocontrol of E. coli O157 with O157-specific bacteriophages. Appl. Environ. Microbiol. 65:3767–3773. Manafi, M. 2000. New developments in chromogenic and fluorogenic culture media. Int. J. Food Microbiol. 60:205–218. March, S. B., and S. Ratnam. 1986. Sorbitol-MacConkey medium for detection of Escherichia coli O157:H7 associated with hemorrhagic colitis. J. Clin. Microbiol. 23:869–872. March, S. B., and S. Ratnam. 1989. Latex agglutination test for detection of Escherichia coli serotype O157. J. Clin. Microbiol. 7: 1675–1677. Naylor, S. W., J. C. Low, T. E. Besser, A. Mahajan, G. J. Gunn, M. C. Pearce, I. J. McKendrick, D. G. E. Smith, and D. L. Gally. 2003. Lymphoid follicle-dense mucosa at the terminal rectum is the principal site of colonization of enterohaemorrhagic Escherichia coli O157:H7 in the bovine host. Infect. Immun. 71:1505–1512. Parham, N., J. Spencer, D. Taylor, H. Ternent, G. Innocent, D. Mellor, M. Roberts, and A. Williams. 2003. An adapted immunomagnetic cell separation method for use in quantification of Escherichia coli O157:H7 from bovine faeces. J. Microbiol. Methods 53:1–9. Ransom, J. R., K. E. Belk, R. T. Bacon, J. N. Sofos, J. A. Scanga,

23.

24.

25.

26.

27. 28.

29.

30.

2855

and G. C. Smith. 2002. Comparison of sampling methods for microbiological testing of beef animal rectal/colonal feces, hides, and carcasses. J. Food Prot. 65:621–626. Sanderson, M. W., J. M. Gay, D. D. Hancock, C. C. Gay, L. K. Fox, and T. E. Besser. 1995. Sensitivity of bacteriologic culture for detection of Escherichia coli O157:H7 in bovine feces. J. Clin. Microbiol. 33:2616–2619. Smith, D., M. Blackford, S. Younts, R. Moxley, J. Gray, L. Hungerford, T. Milton, and T. Klopfenstein. 2001. Ecological relationships between the prevalence of cattle shedding Escherichia coli O157:H7 and characteristics of the cattle or conditions of the feedlot pen. J. Food Prot. 64:1899–1903. Stephan, R., S. Ragettli, and F. Untermann. 2000. Occurrence of verotoxin-producing Escherichia coli (VTEC) in fecal swabs from slaughter cattle and sheep: an observation from a meat hygiene view. Schweiz. Arch. Tierheilkd. 142:110–114. Tkalcic, S., T. Zhao, B. G. Harmon, M. P. Doyle, C. A. Brown, and P. Zhao. 2003. Fecal shedding of enterohemorrhagic Escherichia coli in weaned calves following treatment with probiotic Escherichia coli. J. Food Prot. 66:1184–1189. Varade, W. S. 2000. Hemolytic uremic syndrome: reducing the risks. Contemp. Pediatr. 17:54–64. Wang, G., T. Shao, and M. P. Doyle. 1996. Fate of enterohemorrhagic Escherichia coli O157:H7 in bovine feces. Appl. Environ. Microbiol. 62:2567–2570. Ware, L. M., M. L. Kain, J. N. Sofos, K. E. Belk, and G. C. Smith. 1999. Comparison of sponging and excising as sampling procedures for microbiological analysis of fresh beef-carcass tissue. J. Food Prot. 62:1255–1259. Younts-Dahl, S. M., M. L. Galyean, G. H. Loneragan, N. A. Elam, and M. M. Brashears. 2004. Dietary supplementation with Lactobacillus- and Propionibacterium-based direct-fed microbials and prevalence of Escherichia coli O157 in beef feedlot cattle and on hides at harvest. J. Food Prot. 67:889–893.