Cattle Diets and Experimental Design
All procedures in this study were approved by the Institutional Animal Care and Use Committee (IACUC protocol 00-002). Four ruminally fistulated and four nonfistulated , nonlactating Holstein cows (average 900 kg BW) were adapted to a high-grain diet (Table 1) in a stepwise fashion. Cows were fed according to NRC recommendations (NRC, 2000) and were allowed ad libitum access to water. Two ruminally fistulated and two nonfistulated animals were randomly assigned to each treatment (control or chlorate-treated). Cattle (n = 8) were housed in environmentally controlled facilities and were screened for the presence of E. coli capable of growth on antibiotic-supplemented agar prior to experimental inoculation with E. coli O157:H7 strains.
Bacterial Cultures
Escherichia coli O157:H7 strain 933 (ATCC 43895), strain 6058 (graciously provided by Dan Rice, Field Disease Isolation Unit, Washington State Univ., Pullman), and strain 86-24 (graciously provided by Francisco Diez-Gonzalez, Dept. of Food Sci. and Nutr., Univ. of Minnesota, St. Paul) were cultivated in anoxic Tryptic Soy Broth (TSB) medium at 37°C. Strains 933 and 6058 were isolated from human victims of hemorrhagic colitis outbreaks caused by ingestion of improperly cooked ground beef. Strain 933 was resistant to novobiocin and nalidixic acid (20 and 25 μg/mL, respectively), strain 6058 was resistant to rifampicin (25 μg/mL), and strain 86-24 was resistant to streptomycin (100 μg/mL).
Overnight cultures (1,000 mL each) of all three strains of E. coli O157:H7 were harvested by centrifugation (7,500 × g, 24°C, 10 min) and the cell pellets of all three strains were mixed and collectively resuspended in TSB medium (150 mL). Each cow was inoculated with E. coli O157:H7 (1 × 1010 cfu of strain 933, 2 × 1010 cfu of strain 6058, and 3 × 1010 CFU of strain 86- 24) through the ruminal fistula or via injection through the paralumbar fossa at −48 h (Figure 1). Ruminal fluid samples were collected via ruminal cannula (fistulated cows only) and were strained through sterilized nylon paint strainers (0.1 mm mesh size) and fecal samples were collected via rectal grab (all cows) every 12 h following inoculation. Fecal and ruminal populations of each O157:H7 strain, as well as generic E. coli, total coliforms, and total culturable anaerobes were enumerated as described below.
Chlorate Treatment
Feed and water were withdrawn from cattle for 24 h to stimulate water intake (Figure 1). Cattle were then given access to full feed and ad libitum access to drinking water supplemented with either 2.5 mM KNO3 and 100 mM NaCl (controls) or 2.5 mM KNO3 and 100 mM NaClO3 (chlorate-treated). Potassium nitrate was included in treatments in order to stimulate induction of nitrate reductase activity among the intestinal microflora. Cattle received control or chlorate-supplemented drinking water treatments for 24 h prior to harvest.
Hide Swabs and Gastrointestinal Sample Collection Cattle were humanely euthanatized and exsanguinated. Hide swabs were performed on a 15.2- × 15.2- cm area of the hock and ventral midline of the hanging carcass using sterile gauze pads saturated with sterile PBS (pH 7.0) and sterilized metal templates. Gauze pads were immediately placed in sterile bags for transport to the laboratory. Sterile PBS was added to the bags containing hide swipe pads to dilute material from the hide 10-fold, and bags and swipes were mechanically massaged (via Stomacher) for 1 min to thoroughly mix each sample. Fluid from this initial dilution was used for subsequent dilutions and enrichments of hide swipes.
Intestinal contents and epithelial tissues from the rumen, ileum, cecum, colon, and rectum were aseptically collected upon necropsy as grab samples. Samples were diluted as described below for quantitative enumerations. Sample aliquots and epithelial tissues were added to TSB for overnight qualitative enrichment of E. coli O157:H7 strains. Overnight enrichments were plated as described below. Samples of ruminal, cecal, and rectal contents were immediately transferred to sealed tubes containing anoxic reinforced clostridial agar (described below) for determination of total culturable anaerobic bacteria as described below.
Bacterial Enumeration
Ruminal, intestinal, and fecal contents were serially diluted (10-fold increments) in sterile PBS. Dilutions were plated on MacConkey’s agar (for total coliforms), MacConkey’s agar supplemented with novobiocin (20 μg/mL) and nalidixic acid (25 μg/mL), rifampicin (25 μg/mL), or streptomycin (100 μg/mL) (for inoculated E. coli O157:H7 strains 933, 6058, and 86-24, respectively), and M-Endo agar LES (for enumerating total E. coli) and incubated overnight at 37°C. Colonies that grew on agar plates after a 24-h incubation were directly counted (quantitative enumeration). To qualitatively confirm the presence of inoculated E. coli O157:H7 strains, intestinal contents and epithelial tissue samples were incubated overnight in TSB at 37°C and were streaked on all three antibiotic-supplemented MacConkey’s agars. Plates that contained colonies after a 24-h incubation were classified as positive for each E. coli O157:H7 strain.
Most probable number (MPN) estimates of total culturable anaerobic bacteria from ruminal, cecal, and rectal fluid were determined by serially diluting anaerobic samples harvested at necropsy and performing a triplicate three-tube MPN test using anoxic reinforced clostridial agar supplemented with 1.67 mM xylose, 0.73 mM cellobiose, and 40% filter-sterilized ruminal fluid. Tubes were incubated at 37°C, turbidity after 7 d was indicative of growth in each dilution, and MPN values were estimated using standardized calculations (AOAC, 1980). Intestinal contents were analyzed for pH and VFA concentrations as previously described (Corrier et al., 1990).
Reagents and Supplies
Unless otherwise noted, all media and agar were from Difco Laboratories, Detroit, MI. Reagents and antibiotics were obtained from Sigma Chemical Co., St. Louis, MO.
Statistics
Data were compared via Student’s t-test using GraphPad Prism version 3.00 for Windows (GraphPad Software, San Diego, CA).
Answers to all your Biology Questions
