Seasonal incidence and molecular characterization of Salmonella from dairy cows, calves, and farm environment.

The occurrence patterns and molecular characteristics of Salmonella are important for surveillance and control of the pathogens. Objectives of this study were to determine month-to-month variation and seasonal effects on the occurrence of Salmonella in dairy animals and environments and to characterize selected Salmonella isolates. A total of 7680 animal and environmental samples, collected monthly from a dairy farm, were analyzed for the presence of Salmonella during a 12-month study. Major sources of Salmonella on the dairy farm (% positive) were milking parlor air (62%) and bird droppings (63%) during winter; feeds (50-58%), water (53-67%), calf bedding (63%), soils (60-63%), milking parlor air (60%), and bird droppings (50%) in spring; all animal and environmental samples (40-92%) except milking parlor air (25%) and bulk tank milk (29%) in summer; and feeds (60-71%), cow beddings (59%), cow soils (50%), air (46-71%), and insects (63%) during fall. Salmonella ribotyping indicated that most serovars came from different sources but some might have originated from a common source and transmitted from site to site on the farm. These data provide some important information on key animal and environmental sampling sites needed to initiate on-farm management programs for control of this important foodborne pathogen.

Occurrence of Escherichia coli O157:H7 in diverse farm environments.

In the United States, foodborne outbreaks of Escherichia coli O157:H7 illness have often been linked to the consumption of contaminated, undercooked ground beef. However, the occurrence of E. coli O157:H7 has also been reported in other farm animals. The objective of this study was to evaluate the occurrence of E. coli O157:H7 on diverse farm types and from a variety of farm samples. Rectal swabs (n=1686) and environmental samples (n=576) were collected from 16 farms in five states over 24 months and analyzed for the presence of E. coli O157:H7. Overall, E. coli O157:H7 was found in 3.6% of beef cattle, 3.4% of dairy cattle, 0.9% of chicken, 7.5% of turkey, and 8.9% of swine samples. The pathogen was isolated sporadically from each of the environmental sample types. Of particular concern was the isolation of E. coli O157:H7 from fresh feed samples, indicating a potential vector for transmission. The data from this study indicate a high occurrence of E. coli O157:H7 on swine and turkey farms. This unexpected result suggests that more research on the occurrence of E. coli O157:H7 on these types of farms is required in order to better understand potential reservoirs of pathogenic E. coli.

Nannocystis exedens: a potential biocompetitive agent against Aspergillus flavus and Aspergillus parasiticus.

This study examined the potential for controlling toxigenic Aspergillus flavus and Aspergillus parasiticus by biological means using a myxobacterium commonly found in soil. The ability of Nannocystis exedens to antagonize A. flavus ATCC 16875, A. flavus ATCC 26946, and A. parasiticus NRRL 3145 was discovered. Cultures of aflatoxigenic fungi were grown on 0.3% Trypticase peptone yeast extract agar for 14 days at 28 degrees C. When N. exedens was grown in close proximity with an aflatoxigenic mold, zones of inhibition (10 to 20 mm) developed between the bacterium and mold colony. A flattening of the mold colony on the sides nearest N. exedens and general stunting of growth of the mold colony were also observed. When N. exedens was added to the center of the cross-streak of a mold colony, lysis of the colony by the bacterium was observed after 24 h. Microscopic observations revealed that N. exedens grew on spores, germinating spores, hyphae, and sclerotia of the molds. These results indicate that N. exedens may be a potential biocontrol agent against A. flavus and A. parasiticus.

Preliminary evidence that degradation of aflatoxin B1 by Flavobacterium aurantiacum is enzymatic.

The ability of crude protein extracts from Flavobacterium aurantiacum to degrade aflatoxin B1 (AB1) in aqueous solution was evaluated. Crude protein extracts (800 microg of total protein per ml) degraded 74.5% of AB1 in solution. An average of 94.5% of AB1 was recovered after incubation with heat-treated crude protein extracts (800 microg of total protein per ml). DNase I-treated crude protein extracts degraded 80.5% of AB1 in solution, suggesting that removal of aflatoxin by F. aurantiacum is not due to nonspecific binding with the bacterium's genomic DNA. Proteinase K-treated crude protein extracts degraded 34.5% of AB1, providing evidence that degradation of aflatoxin is linked to a protein that is possibly an enzyme. Solution pH affected the amount of AB1 degraded by crude protein extracts after 24 h. Maximum degradation was observed at pH 7 (pH levels tested: 5, 6, 7, and 8), with some AB1 degradation occurring at pH levels as low as 5 and as high as 8. Acidic pH levels were more detrimental to the ability of crude protein extracts to degrade AB1 than was basic pH. The results of this work indicate that the degradation of AB1 by F. aurantiacum may be enzymatic.

Antimicrobial activity of essential oils from plants against selected pathogenic and saprophytic microorganisms.

The beneficial health effects of extracts from many types of plants that are used as seasoning agents in foods and beverages have been claimed for centuries. The purpose of this study was to examine the effectiveness of selected herb and spice essential oils for control of growth and survival of microorganisms. Inhibition of growth was tested by the paper disc agar diffusion method. Antibiotic susceptibility discs were used as control. Minimum lethal concentration (MLC) was determined by the tube dilution method. Essential oils from anise, angelica, basil, carrot, celery, cardamom, coriander, dill weed, fennel, oregano, parsley, and rosemary were evaluated. Inhibition ranged from complete with oregano to no inhibition with carrot oil for each of the test strains that included: Listeria monocytogenes, Staphylococcus aureus, Escherichia coli O:157:H7, Yersinia enterocolitica, Pseudomonas aeruginosa, Lactobacillus plantarum, Aspergillus niger, Geotrichum, and Rhodotorula. Oregano essential oil showed the greatest inhibition (zone, > or = 70 to 80 mm) (MLC, approximately 8 ppm). Coriander and basil were also highly inhibitory (MLC, approximately 25 to 50 ppm) to E. coli O:157:H7 and to the other bacteria and fungi tested. Anise oil was not particularly inhibitory to bacteria (inhibition zone, approximately 25 mm); however, anise oil was highly inhibitory to molds. Because some of the herbal and spice essential oils are highly inhibitory to selected pathogenic and spoilage microorganisms, they may provide alternatives and supplements to conventional antimicrobial additives in foods.

Detection of Salmonella enterica somatic groups C1 and E1 by PCR-enzyme-linked immunosorbent assay.

Objectives of this study were to develop a PCR-based enzyme-linked immunosorbent assay (PCR-ELISA) for identification of Salmonella enterica somatic groups C1 and E1 and to evaluate this procedure along with a PCR-ELISA procedure for S. enterica somatic groups B, C2, and D in a masked study. Primers were selected from the rfb gene cluster, which is responsible for biosynthesis of O antigens of Salmonella lipopolysaccharide. Previously serogrouped Salmonella isolates (n = 169) were used to determine the sensitivity and specificity of the PCR-ELISA procedure. DNA from all isolates was amplified using the PCR procedure for selected somatic groups and amplified products were visualized on agarose gels, as well as subjected to the ELISA procedure. The PCR-ELISA technique correctly identified 97% of somatic group C1 and 87% of somatic group E1. The sensitivity of this procedure to correctly identify S. enterica somatic group C1 was 97% and 88% for somatic group E1. The specificity was 98% for both somatic groups C1 and E1. The PCR-ELISA techniques correctly identified 93% of Salmonella isolates belonging to somatic groups B, C1, C2, D, and E1. The overall sensitivity of this procedure to correctly identify S. enterica somatic groups was 96% and the specificity was 98%. Ninety-one percent of somatic group D, 92% of somatic group B, and 97% of somatic group C2 were identified correctly with this procedure. Results of this study indicate that the PCR-ELISA procedure is a rapid and accurate method for serogrouping Salmonella isolates. Utilization of the PCR-ELISA procedure for Salmonella serogrouping would aide in identification, surveillance, prevention, and control of Salmonella.

Osmotic dehydration of apple slices using a sucrose/CaCl2 combination to control spoilage caused by Botrytis cinerea, Colletotrichum acutatum, and Penicillium expansum.

The efficacy of sucrose combined with CaCl2 during osmotic dehydration (OD) was tested for the control of Botrytis cinerea, Colletotrichum acutatum, and Penicillium expansum growth on lightly processed apple slices. The objective of this work was to determine whether the addition of CaCl2 in the osmotic solutions would limit the proliferation of fungal decay organisms. Slices were submitted to OD for 1 h at 25 degrees C in solutions containing 5 to 65% sucrose. Calcium chloride was added to a similar set of sucrose solutions at 0 to 8%. Control slices were made of untreated slices, and slices were processed in water. The mass ratio of the slices did not vary when fruit pieces were processed in solutions containing 5 to 65% sucrose. These slices showed a high susceptibility to spoilage compared to the control slices not submitted to OD: a significant twofold and 60% increase in decay area caused by B. cinerea and P. expansum, respectively, was observed when slices were processed in 50% sucrose/0% CaCl2; C. acutatum showed a significant 50% increase in decay area when slices were processed in 20% sucrose/0% CaCl2. Calcium uptake was significantly increased when slices were processed in CaCl2 solutions, and the highest Ca content was observed when processed in 8% CaCl2, reaching 40 times that of the control slices processed in water. Calcium-treated slices were less susceptible to spoilage by all three pathogens, and the most effective combination in reducing apple slice spoilage was 20 to 30% sucrose combined with 2% CaCl2.

Molecular characterization of Salmonella spp. isolated from bulk tank milk and cull dairy cow fecal samples.

The consumption of meat from cull dairy cows and of raw milk has been associated with foodborne salmonellosis. This survey was conducted to establish the prevalence of Salmonella in cull dairy cow fecal samples and bulk tank milk and to determine the proportion of Salmonella-positive dairy farms (n = 30) in east Tennessee. Food and Drug Administration bacteriological analytical protocols were generally used for Salmonella isolation. Primary enrichment was performed with lactose broth, and secondary enrichment was conducted with tetrathionate broth. Eosin methylene blue, hektoen enteric, xylose lysine desoxycholate, bismuth sulfite, and brilliant green (BG) were used as isolation agars. BG agars supplemented with individual antibiotics and/or sulfur compounds were also evaluated. Six of 268 (2.24%) bulk tank milk samples and 9 of 415 (2.17%) fecal samples from 7 of 30 (25.3%) dairy farms were Salmonella-positive. Most isolates (11 of 15) were obtained between September and December. Salmonella isolates were further characterized using polyvalent somatic O Salmonella antiserum, o-nitrophenyl-beta-D-galactopyranoside (ONPG), and Analytical Profile Index (API) 20E strips for Enterobacteriaceae. Serological evaluation of presumptive positive Salmonella isolates resulted in substantial numbers of false positives (41.2%). ONPG and API 20E tests enabled further biochemical distinction of the majority of Salmonella spp. from Salmonella Arizonae and closely related members of Enterobacteriaceae like Citrobacter youngae. Pulsed-field gel electrophoresis of SpeI-digested Salmonella DNA was used to subtype isolates. The isolates grouped into four clusters. The baseline information generated in this survey is being used to develop preharvest pathogen reduction programs on selected farms.

Evaluation of methods for recovery of Salmonella from dairy cattle, poultry, and swine farms.

Current official methods for detection and isolation of Salmonella are mostly designed for foods. The objective of this study was to determine optimal methods for detection and isolation of Salmonella from animal and environmental samples of dairy, poultry, and swine farms. Preenrichment in lactose broth versus direct enrichment (no preenrichment) prior to selective enrichment in Rappaport-Vassiliadis, selenite cystine, and tetrathionate incubated at 35 and 42 degrees C and in four differential/selective plating media (brilliant green, bismuth sulfite, Hektoen enteric, and xylose-lysine-tergitol 4 agar base) were evaluated for their ability to recover Salmonella from artificially contaminated samples. The effects of pH adjustments to samples on Salmonella recovery were determined. A pH adjustment of the enrichment broth to 6.8 +/- 0.2 after addition of samples significantly improved recovery of Salmonella. The most effective medium combinations for isolation of Salmonella from farm samples depended on the type of samples. Generalizations of protocols for recovery of Salmonella from farm samples might result in poor recovery, increased recovery time, and increased sample processing costs.

Prevalence and molecular characterization of Escherichia coli O157:H7 in bulk tank milk and fecal samples from cull cows: a 12-month survey of dairy farms in east Tennessee.

A study on the prevalence of Escherichia coli O157:H7 was conducted on 30 dairy farms in east Tennessee between May 2000 and April 2001. This pathogen was isolated from 8 of 30 (26.7%) dairy farms at various sampling times. A total of 415 fecal samples from cull dairy cows and 268 bulk tank milk samples were analyzed. Overall, 10 of 683 (1.46%) samples (2 of 268 [0.75%] milk samples and 8 of 415 [1.93%] fecal samples) tested positive for E. coli O157:H7. Food and Drug Administration Bacteriological Analytical Manual protocols were used for the conventional isolation and confirmation of E. coli O157:H7. Samples were shake cultured (150 rpm) at 42 degrees C for 24 h in tryptic soy broth containing 2 mg of novobiocin per liter. White colonies isolated on cefixime-tellurite sorbitol MacConkey agar plates were evaluated for fluorescence on sorbitol MacConkey agar supplemented with 0.025 g of methylumbelliferyl-beta-D-glucuronide per liter. Nonfluorescing white colonies were biochemically typed and serologically confirmed. Multiplex polymerase chain reaction profiles of E. coli O157:H7 isolates indicated the presence of common virulence factors (Shiga toxin, enterohemolysin, and intimin) of Shiga toxin-producing E. coli, suggesting the potential human pathogenicity of bacterial isolates. Pulsed-field gel electrophoresis profiles of SpeI and XbaI restriction enzyme-digested genomic DNA were used to establish relatedness among bacterial isolates. Data from this study indicate that both cull dairy cows and bulk tank milk pose a potential hazard with regard to human foodborne illness. It is therefore imperative to develop on-farm and preharvest pathogen reduction programs to control the carriage of E. coli O157:H7 pathogens.

Genetic characterization of a diverse Escherichia coli O157:H7 population from a variety of farm environments.

Many of the current studies on the genetic diversity of Escherichia coli O157:H7 have focused on pathogenic clinical, veterinary, or food isolates. These studies did not explore the diversity of the larger population in the farm environment. Research on selected farm isolates address this wider diversity but have typically been limited to a specific geographic locale or farm type, thus giving limited insight into the greater diversity across geographic regions and varied environments. The objective of this study was to evaluate a diverse population of E. coli O157:H7 collected from a variety of locations and farm environments. Eighty-eight isolates were collected from four farm types (swine, dairy, beef, and poultry) across the southeastern and western United States. Eighteen farms were sampled every 3 months over a period of 24 months. Isolates were analyzed by ribotyping and pulsed field gel electrophoresis (PFGE). Real-time PCR was used to determine the presence or absence of key pathogenic genes (stx1, stx2, and eae). The data indicate a significant amount of genetic diversity, however, ribotype analysis revealed meaningful clusters within the larger population. These groupings were consistent with PFGE analysis. Most of these isolates were clustered by location (i.e. from the same state or region) or farm type. Of the isolates in these clusters, most did not contain pathogenic genes. Of notable interest is a single group in which the majority of isolates, collected from four of the five states sampled, contained at least one stx gene and the eae gene suggesting the existence of a specific pathogenic cluster. These data suggest that, while there is notable diversity within the broader E. coli O157:H7 population, pathogenic isolates may be limited to a subset of strains within the population.

Inhibition of aflatoxin production by selected insecticides.

The insecticide naled completed inhibition production of aflatoxins B1, B2, G1, and G2 by and growth of Aspergillus parasiticus at a 100-ppm (100 microgram/ml) concentration. The insecticides dichlorvos, Landrin, pyrethrum, Sevin, malathion, and Diazinon significantly (P = 0.05) inhibited production of aflatoxins at a 100-ppm concentration. However, at a concentration of 10 ppm, significant inhibition in production of aflatoxins was found only with naled, dichlorvos, Sevin, Landrin, and pyrethrum. Dichlorvos, Landrin, Sevin, and naled inhibited growth of A. parasiticus by 28.9 , 18.9, 15.7, and 100%, respectively, at 100 ppm. Stimulation of growth was observed when diazinon was added to cultures. Aflatoxin B1 was most resistant to inhibition by insecticides, followed by G1, G2, and B2, respectively.

Inhibition by antimicrobial food additives of ochratoxin A production by Aspergillus sulphureus and Penicillium viridicatum.

The effects of antimicrobial food additives on growth and ochratoxin A production by Aspergillus sulphureus NRRL 4077 and Penicillium viridicatum NRRL 3711 were investigated. At pH 4.5, growth and toxin production by both A. sulphureus and P. viridicatum were completely inhibited by 0.02% potassium sorbate, 0.067% methyl paraben, 0.0667% methyl paraben, and 0.2% sodium propionate. At pH 5.5, 0.134% potassium sorbate and 0.067% methyl paraben completely inhibited growth and ochratoxin A production by both fungi. Sodium bisulfite at 0.1%, the maximum level tested, was found to inhibit growth of A. sulphureus and P. viridicatum by 45 and 89%, respectively. Toxin production was inhibited by 97 and 99%, respectively. Sodium propionate (0.64%) at pH 5.5 inhibited growth of A. sulphureus and P. viridicatum by 76 and 90%, respectively. Toxin production was inhibited by greater than 99% for each fungus. Antimicrobial agents were ranked as to effectiveness by comparing the level required for complete inhibition of ochratoxin A production to the highest antimicrobial agent level normally used in food. At pH 4.5, the most effective inhibitor of growth and toxin production was potassium sorbate, followed by sodium propionate, methyl paraben, and sodium bisulfite, respectively, for both fungi. However, at pH 5.5, the most effective antimicrobial agents for inhibiting ochratoxin production were methyl paraben and potassium sorbate, followed by sodium propionate. Sodium bisulfite was not highly inhibitory to these toxigenic fungi at the higher pH value tested.

Fermentation of Lactose in Direct-Acid-Set Cottage Cheese Whey.

Kluyveromyces fragilis was more suitable than Candida pseudotropicalis or Kluyveromyces lactis for production of ethanol from whey. Direct-acid-set cottage cheese whey and the supernatant fluid resulting from heat treatment of the whey at 95 C for 20 min showed similar rates of fermentation when inoculated with K. fragilis . Inoculation rates of 10, 12 and 14 ml of active K. fragilis culture per 100 ml of media were not significantly different in rate of ethanol production. Samples incubated with K. fragilis at 35, 37, 40 and 42 C showed more rapid reduction in specific gravity than samples incubated at room temperature or 30 C. Lactose conversion in whey was 83% complete and in whey supernatant fluid, 77%.

Microbial Spoilage of Mexican-Style Sauces.

The objective of this study was to determine causes for spoilage of Mexican-style sauces prepared "in-house" by restaurants, and to suggest improvements in handling to eliminate this problem. A microbial profile was determined for spoiled enchilada and hot and mild taco sauces. There was no indication of a potential health problem associated with spoiled sauces since Salmonella was absent and Clostridium perfringens and Staphylococcus aureus were present only in low numbers. Spices used in preparing sauces had plate counts ranging from log 4.1 to log 7.7 bacteria per gram. Spoiled sauces had bacterial counts up to log 6.6 per gram. The enchilada and hot sauces which contained the greatest amount of spices had higher bacterial numbers and spoiled more rapidly than the mild taco sauce. Use of ethylene oxide-treated spices, prompt refrigeration of sauces and thorough sanitation reduced counts by 4 log cycles and eliminated spoilage problems.

Microbial profiles of country-cured hams aged in stockinettes, barrier bags, and paraffin wax.

No significant differences were found in surface microflora of county-cured hams covered with stockinettes, barrier bags, or a coating of paraffin wax during aging, except for a reduction in mold growth on waxed hams. The incidence of Clostridium spp. was lost in all treatments. Micrococcus spp. and Streptococcus spp. were the most common contaminants, but caused no apparent spoilage problem in any treatment.

Growth of Bacteria in Soy-Extended Ground Beef Stored at Three Temperatures.

The objective of this study was to determine the influence of five separate levels of textured soy protein (TSP) on growth of psychrotrophs, mesophiles, coliforms, Staphylococcus aureus , and fecal streptococci in soy-extended ground beef stored at -16°, 0° and 6°C. Highly significant increases in psychrotroph and mesophile counts accompanied increased levels of soy at 0° and 6°C, but not at -16°C. Soy-extended beef samples containing 20 and 40% TSP spoiled one day faster at 6°C and four days sooner at 0°C than non-extended ground beef. No significant differences in coliform, fecal streptococci or S. aureus counts could be attributed to increasing levels of TSP in extended ground beef at -16°, 0° or 6°C. Protein content did not vary significantly with TSP concentration; however, fat decreased as soy level increased. Moisture and carbohydrate-ash content increased significantly as soy level increased, as did pH which reached a maximum of 6.5.

Modification of the Processing Method for Home-Preservation of Tomato Juice.

A modification (low water level bath, LWL) of the recommended water bath (high water level bath, HWL) procedure was used to process tomato juice in quart jars. The LWL bath contained one-fifth the amount of water recommended for the HWL bath. Use of the HWL bath required 59 min and 1838 watt-hours of electricity to heat the bath and process hot packed (92°C) juice for 15 min. In comparison, 34 min and 1065 watt-hours of electricity were required when the LWL bath was used. Samples of juice were inoculated with log 3.0 Bacillus coagulans per ml, processed in each of the two baths, and stored up to 12 weeks at 27°C. Aerobic mesophiles were found only in juice processed in the HWL bath and stored 4 weeks and in juice processed in the LWL bath and stored 0 weeks. The aerobic mesophile count (log10) of juice processed in the HWL bath and stored 4 weeks was a mean log 1.4 per ml. Similar juice processed in the LWL bath had a mean log 1.3 aerobic mesophiles per ml. Juice processed in both water baths and stored for 8 and 12 weeks exhibited mesophilic counts of <1 log per ml. None of the inoculated, processed samples had a mean count greater than 1 log per ml of juice for aerobic, acid forming mesophiles; aerobic thermophiles; anaerobic mesophiles and thermophiles; and mold. Using temperature values and microbiological measurements, one may conclude that the LWL bath was as effective as the HWL bath for processing tomato juice while allowing for a substantial saving of time and electricity.