Influence of body condition and forage type on prevalence of Escherichia coli O157:H7 and Salmonella in grazing beef cows

Authors


Michael L. Looper, USDA-ARS, Dale Bumpers Small Farms Research Center, Booneville, AR, USA. E-mail: mike.looper@ars.usda.gov

Abstract

Aim:  To determine the influence of body condition (BC) and forage type on the prevalence of faecal shedding of Escherichia coli O157:H7 and Salmonella from beef cows.

Methods and Results:  Thin or moderately conditioned cows (= 115) were randomly assigned to graze either common bermudagrass (= 3 pastures) or toxic endophyte-infected tall fescue (= 3 pastures) for 62 days. Faecal samples were collected on day 0, 30 and 62. Overall percentage of faecal samples positive for E. coli O157:H7 was 2·6% and 2·0% for Salmonella. Percentage of cows positive for both E. coli O157:H7 and Salmonella on at least one occasion was 6·1%. BC, forage type or the interaction did not influence the prevalence of E. coli O157:H7 or Salmonella in the faeces of cows.

Conclusions:  BC at initiation of the grazing period or loss of BC in moderate conditioned cows during the grazing period did not influence faecal shedding of E. coli O157:H7 or Salmonella. Consumption of either forage type did not influence faecal shedding of either E. coli O157:H7 or Salmonella in beef cows of thin or moderate BC.

Significance and Impact of the Study:  Change in BC that typically occurs during the normal production cycle in grazing cows did not influence faecal shedding of pathogenic bacteria regardless of forage type.

Introduction

Inadequate nutrition can lead to poor performance and reduced reproductive efficiency of cattle (Richards et al. 1986; Randel 1990). A commonly used nine-point body condition (BC) scoring system is a management tool that describes the amount of adipose tissue of cattle (Wagner et al. 1988). Generally, cows in moderate BC lose condition during the postpartum and breeding period, while cows in thin BC typically gain condition (Flores et al. 2007). Fitzgerald et al. (2003) reported that common production stressors influenced faecal shedding of Escherichia coli O157:H7 and Salmonella in dairy cattle. Influence of BC and/or changes in BC on the prevalence of pathogenic bacteria in the faeces of cows is nonexistent in the scientific literature.

Changes in diet may influence faecal shedding of pathogenic bacteria from ruminants (Callaway et al. 2003; Looper et al. 2006). Cattle grazing toxic endophyte-infected (EI) tall fescue, a cool-season forage found throughout the Southeastern United States are exposed to ergot alkaloids that cause increased body temperature during summer months, reduced reproductive performance and growth rate and decreased milk production (Hoveland et al. 1983; Paterson et al. 1995). Further, dry matter intake is usually reduced in ruminants consuming toxic EI tall fescue (Paterson et al. 1995; Looper et al. 2006). Ruminants fed at below-maintenance requirements generally have reduced ruminal volatile fatty acid concentrations and increased pH in the rumen, which can result in increased prevalence of E. coli O157:H7 (Brownlie and Grau 1967; Rasmussen et al. 1993) and Salmonella in faeces (Brownlie and Grau 1967). Although faecal shedding patterns of E. coli O157:H7 and Salmonella from cows grazing toxic EI tall fescue have not been conclusive (Looper et al. 2003, 2006), the combined interaction of BC and toxic EI tall fescue on the prevalence of pathogenic bacteria in the faeces of cows has not been investigated. A better understanding of common, on-farm production factors that influence pathogenic bacteria is needed. Therefore, an experiment was conducted to determine the effects of BC and forage type on the prevalence of faecal shedding of E. coli O157:H7 and Salmonella from grazing beef cows.

Materials and methods

Animal model and collection of faecal samples

All animal procedures used in this study were approved by the Committee for Animal Welfare at the USDA-ARS, Dale Bumpers Small Farms Research Center in Booneville, AR. Multiparous Brahman-influenced (1/4 to 3/8 Bos indicus) cows were managed to achieve a thin (= 50) or moderate (= 65) body condition score (BCS = 4·5 ± 0·1 and 6·6 ± 0·1 for thin and moderate BCs, respectively, where the scale ranged from 1 = emaciated to 9 = obese; Wagner et al. 1988). Cows within each BC were randomly assigned to graze (one cow-calf per 0·8 ha) either common bermudagrass [CB; Cynodon dactylon (L.) Pers; = 3 pastures; = 59 cows] or toxic EI tall fescue [Lolium arundinaceum (Schreb.) Darbysh. = Schedonorus arundinaceum (Schreb.) Dumort; cultivar ‘Kentucky 31’; = 3 pastures; = 56 cows] for 62 days. Cows suckled their calves throughout the study. Body weight (BW) and BCS were recorded, and faecal samples collected on day 0 (9 May), 30 (11 June) and 62 (10 July) of grazing. To minimize animal handling and stress and to optimize labour resources, faecal samples were collected when BWs of cows were recorded. All cows were treated with an anthelmintic (Ivermectin, 1 ml 50 kg−1 BW; Durvet, Inc., Blue Springs, MO, USA) for control of internal parasites at the initiation of the experiment. Approx. 50 g of faecal material was obtained via the rectum using a separate sterile veterinary palpation sleeve for each animal. Faecal samples were packed on ice on the day of collection and shipped overnight to the USDA-ARS, Food and Feed Safety Research Unit, College Station, TX. Prevalence of E. coli O157:H7 and Salmonella in faecal samples was determined within 24 h of collection.

Bacterial culture and isolation

A manual immunomagnetic separation technique was used to isolate E. coli O157:H7 (Edrington et al. 2009). Faeces (10 g) were enriched in 90 ml of tryptic soy broth (TSB) with phosphate (30 g TSB, 2·31 g KH2PO4, 12·54 g K2HPO4 per litre, final pH 7·2) and incubated at 25°C for 2 h, 42°C for 6 h and then overnight at 25°C. Following enrichment, 20 μl of anti-E. coli O157:H7 antibody-labelled paramagnetic beads (Neogen Corp., Lansing, MI, USA) were added to 1 ml of the above enrichment for 15 min at room temperature, while shaking. The paramagnetic beads were removed from the mixture, washed and resuspended in phosphate-buffered saline with 0·05% Tween 20. Fifty microlitres of the resulting suspension was plated on Chrom-O157 (DRG International, Mountain Side, NJ, USA) agar plates containing novobiocin (5 mg l−1) and potassium tellurite (2·5 mg l−1) and incubated at 37°C for 18–20 h. Colonies exhibiting E. coli O157:H7 colony morphology (flat, mauve colonies lacking a distinct centre) were confirmed as E. coli O157:H7 using the Reveal® microbial screening test (Neogen Corp.).

Salmonella was cultured by enriching approx. 10 g of faecal material in 90 ml of TSB for 6 h at 37°C. The enrichment (200 μl) was added to 5 ml of Rappaport-Vassiliadis R10 broth (Difco Laboratories, Detroit, MI) and incubated for 24 h at 42°C before plating on brilliant green agar supplemented with novobiocin (25 μg ml−1). Following incubation (24 h, 37°C), colonies exhibiting Salmonella morphology were confirmed biochemically using lysine agar and triple sugar iron agar tests. Salmonella-positive samples were confirmed by slide agglutination with Salmonella antiserum (Difco Laboratories, Detroit, MI, USA). Treatment groups were unknown to the technician performing bacterial culture and isolation.

Statistical analyses

Chi-square analysis, using the FREQ procedure of sas (SAS Inst., Inc., Cary, NC, USA), was used to determine the influence of pasture, BC, and forage type on the frequency of positive E. coli O157:H7 and Salmonella samples at each collection date. Pasture did not influence the prevalence of E. coli O157:H7 (= 0·63) or Salmonella (= 0·80); therefore, data from replicate pastures were pooled for analysis. Change in BC during the study was analysed by anova using the MIXED procedure of sas, with a model that included BC, forage type, and the interaction. Cows were categorized as gaining, maintaining, or losing BC during the 62-day study. Chi-square analysis, using the FREQ procedure of sas, was used to determine the effect of BC change on the frequency of positive E. coli O157:H7 and Salmonella samples. POWER procedure in sas was used to calculate a best estimate of the retrospective power of this study using the appropriate applications for tests of binomial proportions.

Results

Overall, the percentage of samples positive for E. coli O157:H7 was 2·6 (9 of 344 samples). The percentage of cows positive for E. coli O157:H7 on at least one occasion was 6·1 (7 of 115 cows). Overall percentage of samples positive for Salmonella was 2·0 (7 of 344 samples). Similar to the prevalence of E. coli O157:H7, the percentage of cows positive for Salmonella on at least one occasion was 6·1 (7 of 115 cows).

The BC did not influence the prevalence of E. coli O157:H7 (= 0·45) or Salmonella (= 0·99) in the faeces of cows. Average prevalence of E. coli O157:H7 was 3·4% and 2·1% for thin and moderate BC cows respectively (Table 1). Salmonella prevalence averaged 2·0% for thin cows and 2·1% for moderate cows. Cumulative total percentages of thin and moderate BC cows positive for both pathogens are tabulated in Table 1.

Table 1.   Percentage of faecal samples positive for Escherichia coli O157:H7 and Salmonella from cows in thin (4·5 ± 0·1) or moderate (6·6 ± 0·1) body condition (BC) grazing either bermudagrass or toxic endophyte-infected tall fescue pastures for 62 days
ItemBCForage type
ThinModerateBermudagrassFescue
  1. > 0·10 for main effects (BC and forage type) and the interaction (forage type × BC).

No. of cows50655956
E. coli, %
 Day 04·01·53·41·8
 Day 302·01·53·40·0
 Day 624·13·13·43·6
Cumulative total, %10·16·110·25·4
Salmonella, %
 Day 04·03·13·43·6
 Day 300·00·00·00·0
 Day 622·03·11·73·6
Cumulative total, %6·06·25·17·2

Forage type did not influence the prevalence of E. coli O157:H7 (= 0·35) or Salmonella (= 0·66). For E. coli O157:H7, prevalence averaged 3·4% for cows grazing CB and 1·8% for cows grazing toxic EI fescue (Table 1). Percentage of cows grazing CB and shedding Salmonella averaged 1·7% compared to 2·4% of cows grazing toxic EI fescue and shedding Salmonella. Cumulative total percentages for both pathogens of cows grazing CB or EI fescue forages are listed in Table 1.

Interaction of BC and forage type did not alter prevalence (> 0·10) of E. coli O157:H7 or Salmonella in the faeces of grazing beef cows (data not shown). Change in BC from day 0 to 62 was affected by a forage type × BC interaction (< 0·04). Moderate BC cows grazing either toxic EI fescue or CB lost 0·7 BC units, while thin BC cows grazing toxic EI fescue lost 0·2 BC units during the study; thin cows grazing CB gained 0·2 BC units. Change in BC was not associated with changes in faecal shedding of E. coli O157:H7 (= 0·31) or Salmonella (= 0·57).

Discussion

Prevalence of E. coli O157:H7 and Salmonella detected in the current study are comparable to previous reports from our laboratory (Looper et al. 2003, 2006). Shedding of pathogenic bacteria in the faeces of grazing cattle is highly sporadic, with some animals testing positive for specific pathogenic bacteria at one sample collection and negative at a subsequent collection (Callaway et al. 2004; Looper et al. 2006). Two cows were positive for E. coli O157:H7 on more than one collection date. However, Salmonella was not detected in the same cow more than once in the current study. No cow was positive for both E. coli O157:H7 and Salmonella on same collection dates.

BC of cows did not influence prevalence of E. coli O157:H7 or Salmonella in faeces during the study period. Cows in moderate BC, regardless of forage type grazed, usually lose BC compared with thin BC cows during the postpartum period (Flores et al. 2007). In the current study, cows in moderate BC lost BC regardless of the forage type compared with cows in thin BC grazing CB. However, this loss in BC did not alter shedding of pathogenic bacteria from cows in the current study. Faecal shedding of E. coli from experimentally inoculated calves was increased when calves were fasted before (Cray et al. 1998) but not after inoculation with E. coli (Cray et al. 1998; Harmon et al. 1999). Changes in BC occur gradually (i.e. 62 days in the current study) and would not necessarily include a complete nutrient restriction that occurs during fasting and may explain differences between studies. Common production stressors such as lactation status, parity, and number of days in milk influenced faecal shedding of pathogenic bacteria from dairy cattle (Fitzgerald et al. 2003). To our knowledge, these preliminary data are the first to investigate the relationship of BC and their changes with the prevalence of faecal shedding of pathogenic bacteria from grazing cows. Loss of condition in cows in moderate BC that typically occurs during the postpartum and breeding period did not appear to alter the prevalence of faecal shedding of pathogenic bacteria.

Type of forage grazed by beef cows did not alter the shedding of either E. coli O157:H7 or Salmonella. Cattle grazing toxic EI tall fescue generally have a range of symptoms including increased body temperature and reduced dry matter intake that results in decreased performance (Hoveland et al. 1983; Paterson et al. 1995). Effects of toxic EI fescue on faecal shedding of E. coli O157:H7 in naturally infected cattle have not been consistent. Prevalence of faecal shedding of E. coli O157:H7 was either decreased (Looper et al. 2003) or not altered (Looper et al. 2006) in cattle grazing toxic EI fescue. Short-term feeding of toxic EI fescue seed diets reduced dry matter intake and statistically increased the prevalence of faecal shedding of E. coli O157:H7 in experimentally inoculated ewes compared with experimentally inoculated ewes consuming a nontoxic fescue seed diet (Looper et al. 2007). Difference in results between studies might be due to the duration of exposure to different forages (Looper et al. 2006), experimental animal model (naturally infected vs experimentally inoculated), seasonal effects (Edrington et al. 2006), percentage of animals shedding pathogenic bacteria at the initiation of the study, and/or other variables. In the current study, forage type, specifically toxic EI tall fescue, did not influence faecal shedding of E. coli O157:H7 or Salmonella.

The interaction of BC and forage type did not alter the prevalence of E. coli O157:H7 or Salmonella in the faeces of grazing beef cows. A POWER analysis was conducted on these data to determine whether more animals per treatment were needed. It indicated that a total population of 244–327 cows (0·80 power) would be needed to detect significant differences. The best estimate of POWER, currently, is somewhere between 0·33 and 0·63 based on the prevalence of faecal shedding of E. coli O157:H7 and Salmonella with a sample population of 115 cows.

In conclusion, BC of cows at the initiation of the study or change in BC over 62 days did not influence the prevalence of faecal shedding of E. coli O157:H7 or Salmonella. Further, consumption of toxic EI fescue did not influence faecal shedding of either E. coli O157:H7 or Salmonella from beef cows in thin or moderate BC. It appears from these data that BC changes that typically occur in grazing cattle do not alter faecal shedding of E. coli O157:H7 or Salmonella. Increased number of cows and frequency of faecal collection and/or use of ruminants inoculated with pathogenic bacteria should be the basis of future studies investigating the effects of BC and/or forage diet on faecal shedding of E. coli O157:H7 and Salmonella.

Acknowledgements

Product names are necessary to report factually on available data; however, the USDA does not guarantee or warrant the standard of the product, and the use of the name by the USDA implies no approval of the product to the exclusion of others that also may be suitable. We gratefully acknowledge S. Tabler, B. Woolley and S. Bradshaw, USDA-ARS, for technical assistance and faecal collection; K. Yeater, USDA-ARS, for assistance with statistical analysis.

Ancillary