The effects of progressing and nonprogressing Mycobacterium avium ssp. paratuberculosis infection on milk production in dairy cows.

Longitudinal data from 3 commercial dairy herds in the northeast United States, collected from 2004 to 2011, were analyzed to determine the effect of Mycobacterium avium ssp. paratuberculosis (MAP) infection status and progression path on milk production. Disease status, as indicated by MAP test results, was determined through quarterly ELISA serum testing, biannual fecal culture, and culture of tissues and feces at slaughter. Milk production data were collected from the Dairy Herd Information Association. Animals with positive MAP test results were categorized, based on test results over the full course of the study, as high path (at least one high-positive culture) or low path (at least one positive culture or ELISA). The cumulative numbers of positive ELISA and culture results were recorded. The effects of both MAP infection path, status, and number of positive tests on milk production were analyzed using a mixed linear model with an autocorrelation random effect structure. Low- and high-path animals produced more milk before their first positive test than always-negative animals, especially high-path animals. Although mean production decreased after a first positive test, low-path animals were shown to recover some productivity. High-path animals continued to exhibit a decrease in milk production, especially after their first high-positive fecal culture. These results show that not all animals that test positive for MAP will have long-term production losses. Milk production decreased significantly with each additional positive test. Ultimately, production loss appeared to be a function of MAP infection progression.

Molecular detection of the index case of a subclinical Salmonella Kentucky epidemic on a dairy farm.

Salmonella enterica commonly colonizes the intestinal tract of cattle and is a leading cause of foodborne illness. A previously described investigation into the prevalence of S. enterica on a dairy farm revealed an 8-year-long asymptomatic S. enterica epidemic caused by serotypes Cerro and Kentucky in the lactating herd. To investigate the source of the S. Kentucky strains, the genomes of two S. Kentucky isolates were sequenced; one collected prior to the epidemic (2004) and one collected during the epidemic (2010). Comparative genomic analysis demonstrated significant polymorphisms between the two strains. PCR primers targeting unique and strain-specific regions were developed, and screening of the archived isolates identified the index case of the asymptomatic S. Kentucky epidemic as a heifer that was raised off-site and transported onto the study farm in 2005. Analysis of isolates collected from all heifers brought onto the farm demonstrated frequent re-introduction of clones of the epidemic strain suggesting transmission of pathogens between farms might occur repeatedly.

Antimicrobial resistance of Salmonella enterica isolates from bulk tank milk and milk filters in the United States.

Salmonella isolates were recovered from bulk tank milk as part of the National Animal Health Monitoring System (NAHMS) Dairy 2002 and 2007 surveys. In-line milk filters were also tested in the 2007 survey. The objective of this study was to determine the prevalence of antimicrobial resistance among Salmonella enterica isolates from bulk milk and milk filters in the NAHMS Dairy 2002 and 2007 surveys and to further characterize resistant isolates. Susceptibilities to 15 antibiotics were determined for 176 Salmonella isolates of 26 serotypes using an automated antimicrobial susceptibility system. Resistant isolates were screened by PCR for the presence of the extended-spectrum ß-lactamase (bla(CMY)) gene and class I integrons and further characterized by pulsed-field gel electrophoresis. Thirty isolates (17.0%) representing six S. enterica serotypes exhibited resistance to at least one antimicrobial agent (serotypes Newport [14 of 14 isolates exhibited resistance], Dublin [7 of 7], Typhimurium [3 of 5], Kentucky [4 of 22], Anatum [1 of 13], and Infantis [1 of 2]). Twenty isolates (11.4%), including all 14 Newport, 3 Dublin, 2 Typhimurium, and 1 Infantis isolate, displayed the typical multidrug-resistant, bla(CMY)-positive (MDR-AmpC) phenotype which included resistance to ampicillin, chloramphenicol, streptomycin, sulfonamide, and tetracycline, plus resistance to amoxicillin-clavulanic acid and extended-spectrum cephalosporins. Five of the MDR-AmpC isolates carried class I integrons (2.8%). Two-enzyme (XbaI and BlnI) pulsed-field gel electrophoresis discerned clades within serotypes and, together with the resistance profiles, identified strains that appeared to have persisted temporally and geographically. These results suggest that there is a low but appreciable risk of infection with MDR Salmonella from consumption of nonpasteurized milk and dairy products.

Environmental contamination with Mycobacterium avium subsp. paratuberculosis in endemically infected dairy herds.

Environmental contamination with Mycobacterium avium subsp. paratuberculosis (MAP) is thought to be one of the primary sources of infection for dairy cattle. The exact link between fecal shedding of MAP by individual cows and environmental contamination levels at the herd level was explored with a cross-sectional analysis of longitudinally collected samples on 3 dairy farms. Composite samples from multiple environmental sites in 3 commercial dairy herds in the Northeast US were cultured quarterly for MAP, providing 1131 samples (133 (11.8%) were culture-positive), and all adult animals in the herds were tested biannually by fecal culture (FC), for 6 years. Of the environmental sites sampled, manure storage areas and shared alleyways were most likely to be culture-positive. Environmental sample results were compared to FC results from either the concurrent or previous sampling date at both the herd and the pen level. At the herd level, a 1 log unit increase in average fecal shedding increased the odds of a positive non-pen environmental sample by a factor of 6 and increased the average amount of MAP in non-pen samples by 2.9 cfu/g. At the pen level, a 1 log unit increase in average fecal shedding in the pen increased the odds of a positive environment by a factor of 2.4 and the average amount of MAP was increased by 3.5 cfu/g. We were not able to model the relationship between non-pen environmental sample status and the distance between shedding animals and the sample's location, and neighboring pens did not significantly affect the results of the pen-level analysis. The amount of MAP in pen-level samples and the probability of a pen testing positive for MAP were both positively but non-significantly correlated with the number of animals in the pen shedding >30 cfu/g of MAP. At least 6 environmental samples met the criteria for the U.S. Voluntary Bovine Johne's Disease Control Program on 47 of the 72 sampling dates; of these, 19 of the 47 FC-positive sampling dates were positive by the 6-sample environmental testing method, resulting in a herd sensitivity of 0.40 (95% CI: 0.26-0.54). None of the 3 FC-negative sampling dates produced positive environmental samples. Although environmental sampling can be used as a tool in understanding the level of MAP infection in a herd or pen, it did not appear to be a sensitive diagnostic method for herd positivity in these low prevalence herds, and its use may require caution.

Identification of loci associated with tolerance to Johne's disease in Holstein cattle.

Johne's disease, caused by Mycobacterium avium subspecies paratuberculosis (Map), is a fatal disease in cattle. The objective of this study was to identify loci associated with tolerance in cows infected with Map. Tolerance was defined as a cow's fitness at a given level of Map infection intensity. Fitness was measured by Map faecal cultures, and Map infection intensity was measured by culturing four gut tissues. The quantitative phenotype of tolerance was defined by numerical indexes of cultures of peak (peak tolerance, PT) and average (average tolerance, AT) faecal and tissue Map from 245 Holstein cows. The categorical phenotype was defined as: ≥ 100 cfu Map tissue infection, and faecal shedding ≥ 75 cfu (intolerant) or <10 cfu (tolerant cows). In 94 cows, Map was identified in ≥ 1 tissue, including 44 cows with ≥ 100 Map tissue cfu and 36 with ≥ 1 faecal cfu. A genome-wide association analysis was performed after filtering, leaving genotypes for 45,789 SNPs in 90 animals for the quantitative phenotype and 16 cases and 25 controls for the categorical analysis of tolerance. rs41748405:A>C (BTA15) was associated with PT (P = 1.12 × 10(-7)) and AT (P = 2.17 × 10(-6)). Associations were identified with PT and adjacent SNPs ss61512613:A>G and ss61530518:A>G (BTA6) (P < 3.0 × 10(-5)), and with AT for ss61469568:A>G (BTA 2) (P = 3.3 × 10(-5)) and ss86284768:A>G (BTA1) (P = 3.31 × 10(-5)). For the categorical phenotype, an association was found with ss8632653:A>G (BTA6) (P < 5.0 × 10(-5)). This is the first study to identify loci associated with tolerance to Johne's disease.

Effect of Johne's disease status on reproduction and culling in dairy cattle.

Among the costs attributed to Mycobacterium avium ssp. paratuberculosis (MAP) infection in dairy cattle, the effects on reproduction and culling are the least documented. To estimate the cost of MAP infections and Johne's disease in a dairy herd, the rates of calving and culling were calculated for cows in each stage of MAP infection relative to uninfected cows. Data from 6 commercial dairy herds, consisting of 2,818 cows with 2,754 calvings and 1,483 cullings, were used for analysis. Every cow in each study herd was tested regularly for MAP, and herds were followed for between 4 and 7 yr. An ordinal categorical variable for Johne's disease status [test-negative, low-positive (low-shedding or ELISA-positive only), or high-shedding] was defined as a time-dependent variable for all cows with at least 1 positive test result or 2 negative test results. A Cox regression model, stratified on herd and controlling for the time-dependent infection variable, was used to analyze time to culling. Nonshedding animals were significantly less likely to be culled in comparison with animals in the low-shedding or ELISA-positive category, and high-shedding animals had nonsignificantly higher culling rates than low-shedding or ELISA-positive animals. Time to calving was analyzed using a proportional rates model, an analog to the Andersen-Gill regression model suitable for recurrent event data, stratifying on herd and weighted to adjust for the dependent censoring caused by the culling effects described above. High-shedding animals had lower calving rates in comparison with low-shedding or ELISA-positive animals, which tended to have higher calving rates than test-negative animals.

Biofilm in milking equipment on a dairy farm as a potential source of bulk tank milk contamination with Listeria monocytogenes.

The objective of this study was to assess the presence of a Listeria monocytogenes-containing biofilm in milking equipment as a potential source of bulk tank milk contamination on a dairy farm where milk contamination had been previously documented. Samples were collected from milking equipment and milking parlor premises on 4 occasions and analyzed for the presence of L. monocytogenes. Pulsed-field gel electrophoresis (PFGE) typing was conducted on L. monocytogenes isolates from the milking equipment, parlor and storage room floors, bulk tank milk, and in-line milk filters. Pieces from milk meters and rubber liners were obtained to visually assess the presence of a biofilm using scanning electron microscopy. A total of 6 (15%), 4 (25%), and 1 (6%) samples were culture-positive for L. monocytogenes in the first, second, and third sample collection, respectively. Two samples were L. monocytogenes hly PCR-positive but were culture-negative in the fourth sample collection. Combined AscI and ApaI restriction analysis yielded 6 PFGE types for 15 L. monocytogenes isolates obtained from milking equipment, parlor, bulk tank milk, and milk filters. A predominant and persistent PFGE type (PFGE type T) was observed among these L. monocytogenes isolates (9/15 isolates). Scanning electron microscopy of samples from the bottom cover of 2 milk meters showed the presence of individual and clusters of bacteria, mainly associated with surface scratches. The presence of a bacterial biofilm was observed on the bottom covers of the 2 milk meters. Prevention of the establishment of biofilms in milking equipment is a crucial step in fulfilling the requirement of safe, high-quality milk.

Effect of somatic cell count in goat milk on yield, sensory quality, and fatty acid profile of semisoft cheese.

This study investigated the effect of somatic cell count (SCC) in goat milk on yield, free fatty acid (FFA) profile, and sensory quality of semisoft cheese. Sixty Alpine goats without evidence of clinical mastitis were assigned to 3 groups with milk SCC level of <500,000 (low), 500,000 to 1,000,000 (medium), and 1,000,000 to 1,500,000 (high) cells/mL. Thirty kilograms of goat milk with mean SCC levels of 410,000 (low), 770,000 (medium), and 1,250,000 (high) cells/mL was obtained for the manufacture of semisoft cheese for 2 consecutive weeks in 3 lactation stages. The composition of milk was analyzed and cheese yield was recorded on d 1. Cheese samples on d 1, 60, and 120 were analyzed for total sensory scores, flavor, and body and texture by a panel of 3 expert judges and were also analyzed for FFA. Results indicated that milk composition did not change when milk SCC varied from 214,000 to 1,450,000 cells/mL. Milk with higher SCC had a lower standard plate count, whereas coliform count and psychrotrophic bacteria count were not affected. However, milk components (fat, protein, lactose, casein, and total solids) among the 3 groups were similar. As a result, no significant differences in the yield of semisoft goat cheeses were detected. However, total sensory scores and body and texture scores for cheeses made from the high SCC milk were lower than those for cheeses made from the low and medium SCC milks. The difference in milk SCC levels also resulted in diverse changes in cheese texture (hardness, springiness, and so on) and FFA profiles. Individual and total FFA increased significantly during ripening, regardless the SCC levels. It is concluded that SCC in goat milk did not affect the yield of semisoft cheese but did result in inferior sensory quality of aged cheeses.

A whole genome association analysis identifies loci associated with Mycobacterium avium subsp. paratuberculosis infection status in US holstein cattle.

The purpose of this study was to identify loci associated with Mycobacterium avium subspecies paratuberculosis (Map) infection status in US Holsteins using the Illumina BovineSNP50 BeadChip whole genome single nucleotide polymorphism (SNP) assay. Two hundred forty-five cows from dairies in New York, Pennsylvania and Vermont enrolled in longitudinal herd studies between January 1999 and November 2007 were assessed for the presence of Map in both faecal and tissue samples. An animal was considered tissue infected if any sample contained at least one colony forming unit of Map per gram of tissue (CFU/g) and the same definition was employed for faecal samples. Each animal was genotyped with the Illumina BovineSNP50 BeadChip and after quality assurance filtering, 218 animals and 45 683 SNPs remained. We sought to identify loci associated with four different case/control classifications: presence of Map in the tissue, presence of Map in faeces, presence of Map in both tissue and faeces and presence of Map in tissue but not faeces. A case-control genome wide association study was conducted to test the four different classifications of Map infection status (cases) when compared with a Map-negative control group (control). Regions on chromosomes 1, 5, 7, 8, 16, 21 and 23 were identified with moderate significance (P < 5 x 10(-5)). Two regions, one on chromosome 3 (near EDN2) and another on chromosome 9 (no positional gene candidates), were identified with a high level of association to the presence of Map in tissue and both tissue and faeces respectively (P < 5 x 10(-7), genome-wide Bonferonni P < 0.05).

A longitudinal study on the impact of Johne's disease status on milk production in individual cows.

Longitudinal data from 3 commercial dairy herds in the northeast United States were collected from 2004 to 2007. Johne's disease status, as indicated by Mycobacterium avium ssp. paratuberculosis infection levels, was determined through quarterly ELISA serum testing, biannual fecal culture, and culture of tissues at slaughter. Milk production data were collected from the Dairy Herd Improvement Association. The effect of Johne's disease status on milk production was analyzed using a mixed linear model with an autocorrelation random effect structure. Infected animals produced more milk than uninfected cows before they began shedding M. avium ssp. paratuberculosis. Cows infected with M. avium ssp. paratuberculosis had monthly decreases of 0.05 to 1 kg in daily milk production relative to uninfected animals, with greater decreases in progressive disease categories. Animals with fecal culture results of >30 cfu/g produced approximately 4 kg less milk per day compared with uninfected cows. These results will be valuable in calculating the economic effect of Johne's disease.

Dynamics of endemic infectious diseases of animal and human importance on three dairy herds in the northeastern United States.

Endemic infectious diseases in dairy cattle are of significant concern to the industry as well as for public health because of their potential impact on animal and human health, milk and meat production, food safety, and economics. We sought to provide insight into the dynamics of important endemic infectious diseases in 3 northeastern US dairy herds. Fecal samples from individual cows and various environmental samples from these farms were tested for the presence of major zoonotic pathogens (i.e., Salmonella, Campylobacter, and Listeria) as well as commensal bacteria Escherichia coli and enterococci. Additionally, the presence of Mycobacterium avium ssp. paratuberculosis was tested in fecal and serum samples from individual cows. Test results and health and reproductive records were maintained in a database, and fecal, plasma, DNA, and tissue samples were kept in a biobank. All bacteria of interest were detected on these farms and their presence was variable both within and between farms. The prevalence of Listeria spp. and L. monocytogenes in individual fecal samples within farm A ranged from 0 to 68.2% and 0 to 25.5%, respectively, over a period of 3 yr. Within farm B, continuous fecal shedding of Salmonella spp. was observed with a prevalence ranging from 8 to 88%; Salmonella Cerro was the predominant serotype. Farm C appeared less contaminated with Salmonella and Listeria, although in the summer of 2005, 50 and 19.2% of fecal samples were positive for Listeria and L. monocytogenes, respectively. The high prevalence of E. coli (89 to 100%), Enterococcus (75 to 100%), and Campylobacter (0 to 81%) in feces suggested they were ubiquitous throughout the farm environment. Fecal culture and ELISA results indicated a low prevalence of Mycobacterium avium ssp. paratuberculosis infection in these farms (0 to 13.6% and 0 to 4.9% for culture-positive and ELISA-positive, respectively), although the occasional presence of high shedders was observed. Results have major implications for food safety and epidemiology by providing a better understanding of infectious disease dynamics on dairy farms. Comprehensive understanding of these infections may lead to better farm management practices and pathogen reduction programs to control and reduce the on-farm contamination of these pathogens and to prevent their further entry into the food-chain.

Risk factors associated with the presence of viable Listeria monocytogenes in bulk tank milk from US dairies.

The objective of the study was to screen a large number of herd management practices and herd characteristics from US dairies to identify herd-level risk factors associated with the presence of Listeria monocytogenes in bulk tank milk (BTM). A total of 71 variables was univariately evaluated for their association with the presence of L. monocytogenes in BTM. Results from the univariate analysis indicated that using automatic take offs and having an open herd management increased the risk of BTM contamination with L. monocytogenes, while storing manure in outside pens not accessible to cattle decreased the risk. These variables, however, were not sustained in the multivariable model, which indicated that the presence of L. monocytogenes in BTM was significantly associated with region of the country (farms in the southeast and northeast were six and four times more likely respectively, to have BTM contamination than farms in the west) and number of milking cows (herds with >500 milking cows were five times more likely to have BTM contamination than herds with <100 milking cows). In conclusion, our results suggest that risk factors associated with BTM contamination are different depending on the geographical region and herd size of the operation.

Environmental sampling to predict fecal prevalence of Salmonella in an intensively monitored dairy herd.

Although dairy cattle are known reservoirs for salmonellae, cattle that are shedding this organism are often asymptomatic and difficult to identify. A dairy herd that was experiencing a sustained, subclinical outbreak of Salmonella enterica subsp. enterica Cerro was monitored for 2 years. Fecal samples from the lactating cows were collected every 6 to 8 weeks and tested for the presence of Salmonella. Fecal prevalence of Salmonella fluctuated throughout the observation period and ranged from 8 to 88%. Manure composites and water trough samples were collected along with the fecal samples, and bulk milk and milk filters were cultured for the presence of Salmonella on a weekly basis. Over 90% of the manure composites--representing high-animal-traffic areas-were positive at each sampling. Salmonella was detected in 11% of milk samples and in 66% of the milk filters. Results of weekly bulk milk quality testing (i.e., bulk tank somatic cell score, standard plate count, preliminary incubation count) were typically well within acceptable ranges. Milk quality variables had low correlations with herd Salmonella fecal prevalence. When observed over time, sampling period average prevalence of Salmonella in milk filters closely paralleled fecal prevalence of Salmonella in the herd. Based on results of this study, milk filters appear to be an effective method for monitoring shedding prevalence at the herd level. In-line filter testing is also a more sensitive measure of Salmonella, and perhaps other pathogens, in raw milk than testing the milk alone.

The importance of culling in Johne's disease control.

Johne's disease is caused by Mycobacterium avium subsp. paratuberculosis (MAP) infection and results in economic losses in the dairy industry. To control MAP transmission in herds, test-based culling has been recommended and immediate culling of high shedding animals is typically implemented. In this study, we quantified the effects of MAP control in US dairy herds, using the basic reproduction ratio R(0). The effectiveness of culling strategies was evaluated for good and poor herd management (low- and high-transmission rates, respectively) by a phase diagram approach. To establish a quantitative relationship between culling rates and test properties, we defined the average detection times for low and high shedding animals. The effects of various culling strategies and test characteristics, such as test sensitivity, test turnaround time, and testing interval, were analyzed. To understand the overall effect of model parameters on R(0), we performed global uncertainty and sensitivity analyses. We also evaluated the effectiveness of culling only high shedding animals by comparing three test methods (fecal culture, fecal polymerase chain reaction, PCR, and enzyme-linked immunosorbent assay, ELISA). Our study shows that, in the case of good herd management, culling of only high shedding animals may be effective in controlling MAP transmission. However, in the case of poor management, in addition to immediate culling of high shedding animals, culling of low shedding animals (based on the fecal culture test) will be necessary. Culling of low shedding animals may be delayed 6-12 months, however, if a shorter testing interval is applied. This study suggests that if farmers prefer culling only high shedding animals, faster MAP detection tests (such as the fecal PCR and ELISA) of higher sensitivity should be applied with high testing frequency, particularly on farms with poor management. Culling of infectious animals with a longer testing interval is generally not effective to control MAP.

A mathematical model of the dynamics of Salmonella Cerro infection in a US dairy herd.

We developed a mathematical model of the transmission dynamics of salmonella to describe an outbreak of S. Cerro infection that occurred in a Pennsylvania dairy herd. The data were collected as part of a cooperative research project between the Regional Dairy Quality Management Alliance and the Agricultural Research Service. After the initial detection of a high prevalence of S. Cerro infection in the herd, a frequent and intensive sampling was conducted and the outbreak was followed for 1 year. The data showed a persistent presence of S. Cerro with a high prevalence of infection in the herd. The dynamics of host and pathogen were modelled using a set of nonlinear differential equations. A more realistically distributed (gamma-distributed) infectious period using multiple stages of infection was considered. The basic reproduction number was calculated and relevance to the intervention strategies is discussed.

Longitudinal study of a clonal, subclinical outbreak of Salmonella enterica subsp. enterica serovar Cerro in a U.S. dairy herd.

Salmonellae are a major group of foodborne pathogens known to affect both humans and animals. Dairy cattle are a known reservoir of these bacteria and human Salmonella infections have been associated with the consumption of improperly processed or contaminated dairy products. Many of the over 2500 known serotypes of Salmonella are known to infect cattle, resulting in asymptomatic to fatal salmonellosis. This study describes the course of a Salmonella outbreak and subsequent endemic infection on a dairy farm in Pennsylvania. The outbreak was initially detected when a few cows with clinical symptoms and one fatality were found to be infected with Salmonella enterica subsp. enterica serovar Typhimurium var. Copenhagan. Based upon sampling of the farm environment, Salmonella Typhimurium var. Copenhagan was succeeded within 3 months by Salmonella enterica subsp. enterica serovar Kentucky. Salmonella enterica subsp. enterica serovar Cerro ultimately supplanted Typhimurium var. Copenhagan and Kentucky in individual animals and environmental samples and persisted in the herd at high prevalence for almost 2 years. Since there were no obvious clinical consequences of the Salmonella Cerro infection, these data suggest that some serotypes of S. enterica subsp. enterica can behave as commensal organisms in dairy cattle and illustrate the difficulties of controlling Salmonella in milk production systems. The consistent finding of Salmonella in the environment reinforces the potential for human exposure to this pathogen and the need to understand the dynamics and ecology of Salmonella in dairy production settings.

Survival of Escherichia coli in cowpats in pasture and in laboratory conditions.

AIMS: To compare survival of Escherichia coli and faecal coliforms (FC) in bovine faeces deposited in a pasture or incubated in a controlled laboratory environment at temperatures within the same range. METHODS AND RESULTS: Faecal samples from three cow herds were deposited as shaded and nonshaded cowpats in a field and incubated in a laboratory for one month at 21.1, 26.7 and 32.2 degrees C. Both FC and E. coli concentrations increased as much as 1.5 orders of magnitude both in the field and in the laboratory during the 1st week and subsequently decreased. In shaded cowpats, the die-off of E. coli and FC was significantly slower, and the proportion of E. coli in FC was significantly larger as compared with nonshaded cowpats. The die-off was faster in the field than in the laboratory at similar temperatures. CONCLUSIONS: FC and E. coli die-off rates were substantially lower in laboratory conditions than in the field within the same range of temperatures. SIGNIFICANCE AND IMPACT OF THE STUDY: This study underscores the importance of field data on survival of manure-borne FC and E. coli, and indicates that laboratory die-off rates have to be corrected to be used for field condition simulations.

Incidence of Escherichia coli O157:H7 and E. coli virulence factors in US bulk tank milk as determined by polymerase chain reaction.

Samples of bulk tank milk from dairies across the United States, taken as part of the National Animal Health Monitoring Dairy 2002 survey, were analyzed for the presence of several genes encoding virulence factors associated with enterohemorrhagic forms of Escherichia coli (EHEC) using real-time and conventional PCR assays. Samples from 859 farms in 21 states were collected and enriched in EC medium at 42.5 degrees C to amplify any E. coli present, and DNA was isolated from the resulting biomass. The eaeA gene encoding intimin, a virulence factor associated with enteropathogenic forms of E. coli and EHEC, was detected in 199 (23%) of the samples. Thirty-six samples (4.2%) were positive for eaeA, the gamma allele of the translocated intimin receptor (gamma-tir), found in EHEC strains of O157:H7, and one or both shiga-like toxin genes (stx1 and stx2), a combination that may be indicative of the presence of O157:H7 EHEC. Testing these 36 samples with a commercially available real-time PCR kit for detection of O157:H7 indicated that 5 samples could be contaminated with O157:H7. A multiplex PCR to detect the presence of fliC, rfbE, and hlyA genes found in O157:H7 reduced to 2 (0.2% of all samples) the number of samples likely to be contaminated with this organism. A strain of O157:H7 (eaeA+, gamma-tir+, stx2+, rfbE+, fliC+, hlyA+) was subsequently isolated from one sample. Thirty-four eaeA-positive samples did not contain detectable gamma-tir but did contain one or both of the stx genes suggesting the presence of EHEC strains other than O157:H7. These results indicate a low incidence of O157:H7 in bulk tank milk but suggest that a risk from other enteropathogenic and EHEC forms of E. coli may exist and that PCR targeting virulence factors associated with highly pathogenic forms of E. coli may be an effective means of detecting potential dangers in raw milk.

Comparison of release and transport of manure-borne Escherichia coli and enterococci under grass buffer conditions.

AIM: To test the hypothesis that Escherichia coli and enterococci bacteria have similar release rates and transport characteristics after being released from land-applied manure. METHODS AND RESULTS: Turfgrass soil sod was placed into 200 cm long boxes that had the top two 25 cm sections separated to monitor the release and infiltration of bacteria, which affected bacteria transport in the rest of the box. Dairy manure with added KBr was broadcast on the top two sections. Boxes with either live or dead grass stand were placed under a rainfall simulator for 90 min. Runoff and infiltration samples were collected and analysed for Br, E. coli, enterococci and turbidity. Significant differences in release kinetics of E. coli and enterococci were found. A change from first-order release kinetics to zero-order kinetics after 1 h of rainfall simulation was observed. CONCLUSION: Differences in release rates but not in the subsequent transport were observed for E. coli and enterococci. SIGNIFICANCE AND IMPACT OF THE STUDY: Because both E. coli and enterococci are currently used as indicator organisms for manure-borne pathogens, the differences in their release rates may affect the efficiency of using these organisms as indicators.

Subtyping Listeria monocytogenes from bulk tank milk using automated repetitive element-based PCR.

Sixty-one Listeria monocytogenes strains from raw milk were analyzed with an automated repetitive element-based PCR (rep-PCR) system to examine the utility of this system for serotype grouping and to determine whether specific regional relationships could be identified. Results of the similarity analysis revealed two primary clusters of L. monocytogenes isolates. Cluster 2 exclusively contained serogroup 1/2a isolates; however, two 1/2a isolates were also found in cluster 1. Isolates of serogroups 1/2b, 4b, 3b, and 4c were also in cluster 1. Clusters 1 and 2 were separated at a relative similarity of 86%. Listeria species other than L. monocytogenes (L. ivanovii, L. seeligeri, L. welshimeri, L. grayi, and L. innocua) had similarity scores of less than 80% in pairwise comparisons with the L. monocytogenes isolates. Thus, this method may be useful for species identification once an isolate is characterized as Listeria. When rep-PCR fingerprints of the L. monocytogenes 1/2a isolates were compared, there was no apparent regional grouping. However, discrimination between isolates suggests that the rep-PCR assay might be useful for tracking L. monocytogenes 1/2a and for tracking isolates across regions or within smaller ecological niches. The automated rep-PCR method could not discriminate between serotypes 1/2b and 4b but may be useful for discriminating between 1/2a and other serotypes and for tracking isolates within serotype 1/2a.