Cleaning and handling implements as potential reservoirs for bacterial contamination of some ready-to-eat foods in retail delicatessen environments.

This study assessed the association of bacteria with cleaning tools, such as floor mops (n = 25) and cleaning cloths (n = 39), and handling devices, such as disposable plastic gloves (n = 20), used during filled baguette and assorted salad preparation in four selected retail delicatessens in Johannesburg, South Africa. Samples of each cleaning or handling tool were prepared for aerobic (APC), coliform (CC), Escherichia coli (EC), Bacillus cereus (BCC), and Staphylococcus aureus (SAC) counts, as well as tested for the incidence of Listeria monocytogenes (LM) and Salmonella (SALM) by standard plating methods. Bacterial populations attached to the cleaning and handling tools were observed by scanning electron microscopy (SEM). Ten selected gram-positive isolates were further analyzed by 16S rRNA sequence analysis and compared with isolates from filled baguettes and assorted salads. The floor mops consistently yielded the highest APCs, CCs, and ECs (5.7, 4.1, and 3.0 log CFU/g, respectively), while gloves had the lowest corresponding counts (3.6, 2.0, and 1.0 log CFU/g, respectively). Low BCCs and SACs were recorded in this study (ca. 1.2 log CFU/g), while SALM and LM were each detected in five cleaning tool samples. SEM showed rods and cocci attached to handling and cleaning tools. Furthermore, results of 16S rRNA sequence analysis indicated that several gram-positive isolates were identified as S. aureus, Staphylococcus pasteuri, Staphylococcus sciuri, and Enterococcus faecalis. Genetically similar strains (100% similarity) were isolated from cleaning and handling tools and associated ready-to-eat (RTE) foods. Cleaning and handling tools may act as reservoirs of contamination for RTE foods during preparation in retail delicatessens in South Africa. The transfer of potential pathogens, such as S. aureus, to foods from cleaning and handling tools may hold food safety implications.

Carbon : nitrogen : phosphorus ratios influence biofilm formation by Enterobacter cloacae and Citrobacter freundii.

AIMS: To test the effects of C : N : P ratio modification of a well-known nutrient medium formulation, the Endo formulation on biofilm formation by Enterobacter cloacae Ecl and Citrobacter freundii Cf1 in both single-species and binary species biofilms. METHODS AND RESULTS: The C : N : P atom : atom ratio of a well-known nutrient medium formulation, the Endo formulation, that has been applied in fermentative biohydrogen studies, was modified to include two different C concentrations, one containing 17.65 g l(-1) and the other 8.84 g l(-1) sucrose, each containing four different C : N : P ratios, two at higher C : N : P ratios (334 : 84 : 16.8 and 334 : 84 : 3) and two at lower C : N : P ratios (334 : 28 : 5.6 and 334 : 28 : 1). Attached cells were enumerated after dislodging the biofilms that had formed on granular activated carbon (GAC). The modified medium containing 17.65 g l(-1) sucrose and having a C : N : P ratio of 334 : 28 : 5.6 resulted in significantly (P < 0.05) higher counts of attached cells for both single-species biofilms at 7.73 log(10) CFU g(-1) GAC and 9.3 log(10)CFU g(-1) GAC for Ent. cloacae Ecl and Cit. freundii Cf1, respectively, and binary species biofilms at 8.2 log(10) CFU g(-1) GAC and 6.34 log(10) CFU g(-1) GAC for Ent. cloacae Ecl and Cit. freundii Cf1, respectively. Scanning electron micrographs showed qualitative evidence that the 334 : 28 : 5.6 ratio encouraged more complex and extensive biofilm growth for both single-species and binary species biofilms. CONCLUSIONS: The differences in the attachment numbers between the different ratios were found not to be a result of the individual actions of the bacterial isolates involved but rather because of the effects of the various C : N : P ratios. The 334 : 28 : 5.6 ratio showed significantly (P < 0.05) higher counts of attached cells for both single-species and binary species biofilms. SIGNIFICANCE AND IMPACT OF THE STUDY: This study indicates that C : N : P ratios should be a key consideration with regard to maximizing biofilm formation in shake flask and fluidized bed bioreactor studies as well as understanding fundamental factors affecting biofilm growth in natural environments.

Bacterial biofilms within the clinical setting: what healthcare professionals should know.

Bacterial biofilm formation is the prevailing microbial lifestyle in natural and manmade environments and occurs on all surface types. Biofilm formation develops in several phases and is influenced by various parameters, both environmental and inherent to the attaching cell. Biofilms also serve as protective niches for particular pathogens when outside a host. Although it is accepted that biofilms are ubiquitous in nature, the significance of biofilms in clinical settings, especially with regard to their role in medical-related infections, is often underestimated. It has been found that several aspects of human pathogenesis within a clinical context are directly related to biofilm development. Various types of surfaces in clinical settings are prone to biofilm development and an increased risk of disease may be a direct consequence of their formation. This review describes the process of biofilm formation, highlights the importance of bacterial associations with surfaces in clinical settings and describes various methods for biofilm visualization and control.

Improving street food vending in South Africa: achievements and lessons learned.

Until the late 1990s there were limited scientific data on the microbiological quality and safety of street-vended foods in South Africa, while information was already available in other developing countries, including those within the African region. At that time street-vended foods were perceived as unsafe and street food vending in South Africa was regarded as a practice, which should be outlawed. The first comprehensively documented scientific research into the safety of street-vended foods in South Africa was carried out through university-based research. This research found that street food vendors in South Africa were capable of producing relatively safe foods, with low bacterial counts, although there was still a need for proper hygiene conditions and access to basic sanitary facilities. The Department of Health of South Africa, when coordinating an FAO Technical Cooperation Programme (TCP) project on Improving Street Foods in South Africa, drew similar conclusions. This article provides information of the efforts by universities and health authorities in South Africa towards improving the safety and promoting the sale of street-vended foods. It is shown that a successful transition from street food vending being perceived as a nuisance by health authorities can be made to these authorities promoting and improving street food vending instead.

Biofilm-spore response in Bacillus cereus and Bacillus subtilis during nutrient limitation.

This study aimed to trace the dynamics of biofilm formation by vegetative cells and endospores of Bacillus cereus DL5 and Bacillus subtilis 168. Counts of B. cereus DL5 and B. subtilis 168 vegetative cells and spores either attached to glass wool or, correspondingly, planktonic cells were determined by standard plate-counting methods. Results from this study highlighted the biofilm-forming potential of both spores and vegetative cells of two different Bacillus species. It was shown that once Bacillus spores had attached to a surface, the spores germinated under favorable (B. cereus DL5) and even unfavorable (B. subtilis 168) nutrient conditions, resulting in biofilms containing both spores and vegetative populations. Furthermore, it was suggested that vegetative B. cereus DL5 cells exhibited a low propensity for spore formation in attached and planktonic growth forms in nutrient-limited growth medium. By contrast, vegetative B. subtilis 168 cells readily formed spores in planktonic and attached microcosms when exposed to nutrient-limited growth conditions. Sporulation in attached Bacillus populations is an important practical consideration for many food industries, such as dairy processing, where bacilli are routinely isolated from populations attached to processing-equipment surfaces.

Spore formation in Bacillus subtilis biofilms.

Spore formation by a Bacillus strain (Bacillus subtilis SpoIVFB-GFP) engineered with a green fluorescent protein (GFP) fused to a polytopic membrane protein (SpoIVF) that fluoresces during sporulation was observed. Biofilms of B. subtilis SpoIVFB-GFP containing ca. 8 log CFU/ml vegetative cells and spores below the lower detection limit (i.e., <1 log CFU/ ml) were allowed to develop on glass wool (37 degrees C). These biofilms were subsequently exposed to nutrient limitation to stimulate spore formation, which was monitored for fluorescence by confocal scanning laser microscopy. Sporulation in corresponding planktonic cells was also monitored for comparative purposes. Planktonic B. subtilis SpoIVFB-GFP cells began fluorescing after 5 h, while B. subtilis SpoIVFB-GFP biofilm cells began fluorescing after 30 h. Results suggested that an existing biofilm of vegetative B. subtilis cells may be stimulated to form spores when exposed to conditions of nutrient limitation. From a practical point of view, it may be suggested that a window of time does exist before sporulation occurs in attached Bacillus biofilms highlighting the need for shorter operating runs between cleaning and sanitation of food-processing equipment surfaces.

The presence of Enterococcus, coliforms and E. coli in a commercial yeast manufacturing process.

This study evaluated a typical commercial yeast manufacturing process for bacterial contamination. Product line samples of a commercial yeast manufacturing process and the corresponding seed yeast manufacturing process were obtained upstream from the final compressed and dry yeast products. All samples were analysed before (non-PI) and after preliminary incubation (PI) at 37 degrees C for 24 h. The PI procedure was incorporated for amplification of bacterial counts below the lower detection limit. Enterococcus, coliform and Escherichia coli counts were quantified by standard pour-plate techniques using selective media. Presence at all stages and progressive increases in counts of Enterococcus, coliforms and E. coli during processing in the commercial manufacturing operation suggested that the primary source of contamination of both compressed and dry yeast with these bacteria was the seed yeast manufacturing process and that contamination was amplified throughout the commercial yeast manufacturing process. This was confirmed by surveys of the seed yeast manufacturing process which indicated that contamination of the seed yeast with Enterococcus, coliforms and E. coli occurred during scale up of seed yeast biomass destined as inoculum for the commercial fermentation.

In vitro growth response of bread-spoilage Bacillus strains to selected natural antimicrobials.

This study assessed the in vitro responses of Bacillus (B.) strains isolated from ropey bread to natural antimicrobials under optimum growth conditions. The responses of six Bacillus strains [B. subtilis (2), B. licheniformis (2) and B. pumilus (2)] to acetic acid (AA), lactic acid (LA), calcium lactate (CL) and a lactate-containing cocktail (LCC), singly and in combinations were determined and compared to calcium propionate (CP). Isolates were each inoculated into flasks containing Nutrient Broth (NB) and the respective antimicrobial treatments and pHs were left unadjusted. A duplicate set of flasks, also containing NB and the respective antimicrobials, but adjusted to pH's corresponding to those of baked brown bread containing the same antimicrobials was also inoculated. Growth curves were obtained spectrophotometrically and used to estimate lag times. The organic acids used in this study [0.1% (v/v) AA and 0.25% (v/v) LA] singly and in combination with each other and with CL, CP or LCC, completely inhibited the growth of all six Bacillus strains, but only at non-adjusted pHs. The efficacies of LA, AA and CL notably decreased when the pH of the test media containing the respective preservatives was adjusted to the corresponding in situ (bread) values. However, the natural antimicrobials were still as effective as CP in retarding growth of the six Bacillus strains at the in situ (bread) pH values.

Adaptation of a neutrophilic dairy-associated Bacillus cereus isolate to alkaline pH.

AIMS: This study identified and studied the response of five Bacillus strains, isolated from alkaline cleaning in place (CIP) solutions, to alkaline conditions. METHODS AND RESULTS: Isolates were identified as B. cereus by 16S rDNA sequencing. External and internal cell pH and buffering capacity data of a representative strain, Bacillus DL5, were compared to B. cereus ATCC 10702. Results indicated that a buffering system was induced when the pH of the growth medium increased to above pH 10, which was effective up to pH 12 and presumably cell wall associated. Volume measurements and confocal scanning laser microscope images of Bacillus DL5 cells showed that cells exhibited more pronounced stress symptoms when exposed to pH 10 than at pHs above 10. Long-term exposure of Bacillus DL5 to pH 10 or 10.5 indicated that cells grew in planktonic form and formed biofilms at both pHs. CONCLUSIONS: Bacillus DL5 was a neutrophile with a growth pH range similar to B. cereus ATCC 10702, but tolerated alkaline pH. This may be a general trait of the B. cereus species rather than a specific phenomenon of isolates from alkaline ecosystems. SIGNIFICANCE AND IMPACT OF THE STUDY: Other neutrophilic B. cereus isolates may exhibit similar responses to alkaline conditions as the isolates studied here. These results may have important implications for dairy manufacturers.

Differential efficacy of a chlorine dioxide-containing sanitizer against single species and binary biofilms of a dairy-associated Bacillus cereus and a Pseudomonas fluorescens isolate.

AIMS: Daily exposure to 100 p.p.m. chlorine dioxide of single species and binary biofilms of dairy-associated Bacillus cereus DL5 and Pseudomonas fluorescens M2, attached to stainless steel surfaces in a laboratory flow system, was studied. METHODS AND RESULTS: Surfaces were sampled daily before and after sanitizer treatment and cells and spores dislodged and enumerated by standard methods. Duplicate surfaces were prepared for confocal scanning laser microscopy (CSLM) and scanning electron microscopy. Higher counts of Ps. fluorescens M2 were obtained in single species biofilms, microcolonies stained green (viable) in CSLM images and were closely packed on attachment surfaces. By contrast, higher counts of B. cereus DL5 were obtained in binary biofilms, microcolonies stained green in CSLM images, but were more spread out. Lower spore counts were obtained for B. cereus DL5 in binary biofilms. The survival of Ps. fluorescens M2 cells after exposure to chlorine dioxide was apparently enhanced by the presence of B. cereus DL5 in binary biofilms. By contrast, B. cereus DL5 showed increased susceptibility to sanitizer treatment in the presence of Ps. fluorescens M2. CONCLUSIONS: Co-cultured bacteria in biofilms influence each other with respect to attachment capabilities and sanitizer resistance/susceptibility. SIGNIFICANCE AND IMPACT OF THE STUDY: Binary biofilms endemic in food-processing industries can survive sanitation regimes and may represent reservoirs of product contamination leading to subsequent spoilage and/or food safety risks.

Effect of selected natural antimicrobials on Baker's yeast activity.

AIMS: To evaluate the responses of Baker's yeast (Saccharomyces cerevisiae) activity to the natural antimicrobials acetic acid, calcium lactate, a lactate-containing cocktail and lactic acid compared to calcium propionate. METHODS AND RESULTS: A dough fermentometer test was used to measure Baker's yeast activity in the presence of these natural antimicrobials and calcium propionate. Yeast activity generally decreased as a function of increasing antimicrobial concentrations, but the lactate-containing cocktail showed no relationship between concentration and yeast activity reduction. At in situ concentrations, calcium propionate resulted in the highest yeast activity reduction, followed by calcium lactate, acetic acid, the lactate-containing cocktail and lactic acid in decreasing order. CONCLUSION: Based on yeast activity reduction, all natural antimicrobials tested showed potential as possible replacements for calcium propionate. SIGNIFICANCE AND IMPACT OF THE STUDY: This has practical implications since calcium propionate inhibits Baker's yeast activity and attracts negative consumer perceptions as a chemical bread preservative.

Genotypic analysis of Escherichia coli strains from poultry carcasses and their susceptibilities to antimicrobial agents.

Plasmid profiling and amplified fragment length polymorphism (AFLP) analysis were used to genotype 50 Escherichia coli strains from poultry carcasses. Thirty different plasmid profiles were evident, and clustering of the AFLP data showed that they were a distinctly heterogeneous group of strains. Susceptibility testing against five antimicrobial agents used in the South African poultry industry showed all strains to be susceptible to danofloxacin and colistin, while the majority (96%) were resistant to two tetracyclines.

Antimicrobial susceptibilities of isolates of Staphylococcus aureus, Listeria species and Salmonella serotypes associated with poultry processing.

The broth microdilution method was used to determine the activities of selected antimicrobial agents used in the South African poultry industry (danofloxacin, neomycin, chlortetracycline, oxytetracycline, tylosin and colistin) and vancomycin against bacterial isolates previously obtained from carcasses and selected equipment surfaces and environmental sources associated with poultry processing. The antimicrobial susceptibilities of 38 isolates of Staphylococcus (S.) aureus, 25 Listeria (L.) innocua, 18 L. monocytogenes, and 62 isolates belonging to six Salmonella (Salm.) serotypes (Salm. agona, Salm. blockley, Salm. enteritidis, Salm. isangi, Salm. reading and Salm. typhimurium) were determined. The most active antimicrobial agent against all the isolates tested was danofloxacin with minimum inhibitory concentrations (MICs) for 90% of the isolates (MIC90) not exceeding 0.25 and 2 microg/ml for gram-negative and gram-positive isolates, respectively. Conversely, high MICs were recorded for all the isolates tested against chlortetracycline and oxytetracycline (MIC90 range of 32 to > 512 microg/ml), except for the L. monocytogenes and Salm. enteritidis isolates (MIC range of < or = 0.5-4 microg/ml). Neomycin was found to be active against S. aureus, L. innocua, L. monocytogenes, Salm. enteritidis and Salm. isangi isolates, with MICs not exceeding 8 microg/ml. MIC ranges for tylosin and vancomycin, which were only tested against the gram-positive isolates, were from 1 to > 512 microg/ml and from 1 to 4 microg/ml, respectively. The MIC range for the remaining antimicrobial agent, colistin, which was only tested against the Salmonella isolates, was 0.5-16 microg/ml. The lack of MIC breakpoints for the antimicrobial agents used in the poultry industry did not allow for definite conclusions as to the level of resistant bacteria associated with poultry carcasses and the processing environment in this study.

Microbiological hazard identification and exposure assessment of street food vending in Johannesburg, South Africa.

One hundred and thirty-two samples of beef, chicken, salad and gravy were collected from two street vendors over eleven replicate surveys to assess microbiological safety and quality. For each food type samples were collected during preparation and holding. Dish water was also collected and food preparation surfaces swabbed during preparation and display. Standard methods were used to determine aerobic plate counts, Enterobacteriaceae counts, coliform counts and spore counts. Six hundred and seventy-five predominant colonies were isolated from aerobic plate counts of all samples and characterised. The incidence of selected foodborne bacterial pathogens and non-pathogenic E. coli 1 was also determined. In most cases mean bacterial counts of the raw materials were significantly higher (P < 0.05) than those of corresponding cooked foods. No significant differences (P > 0.05) in all count types were observed between food samples collected during cooking and those collected during holding. In addition, no significant differences (P > 0.05) in all count types were observed between prepared salads and their raw materials. Mean bacterial counts of water and swab samples collected from vendor 1 were lower than those of water and swab samples collected from vendor 2.The predominant populations isolated from the aerobic plate counts were Bacillus spp., Staphylococcus spp., Enterobacteriaceae and Alcaligenes spp. Bacillus cereus was detected in 17%, Clostridium perfringens in 1%, Staphylococcus aureus in 3% and Vibrio metchnikovii in 2% of the food samples. Campylobacter jejuni, Listeria monocytogenes, and Escherichia coli O157:H7 were not detected. Non-pathogenic E. coli 1 was detected in 13% of food samples, in 86 and 36% of dish water samples collected from vendors 1 and 2, respectively, and in 36% of surface swab samples from vendor 2.

Bacterial populations associated with bulk packaged beef supplemented with dietary vitamin E.

Five Simmentaler type calves were fed diets supplemented with 500 mg vitamin E per day and five fed control diets, Rump steaks from each carcass were PVC-overwrapped and bulk packaged in 100% CO2 or 20% CO2:80% O2. Bulk packs were stored up to 42 days at 4 degrees C and PVC-overwrapped samples subsequently displayed up to 7 days at 4 degrees C. After display the Aerobic Plate Count (APC) of steaks was determined and four colonies were randomly selected from the highest dilution APC plates showing growth. A total of 627 colonies were obtained. Gram-reaction, catalase, oxidase, morphology and motility of the isolates were determined. The gram-negative and gram-positive isolates were then identified using a dichotomous identification key. Enterobacteriaceae, Pseudomonas spp. and Acinetobacter spp. predominated on rump steaks from both feeding treatments and in packaging treatments. After 42 days bulk storage Enterobacteriaceae, Pseudomonas spp., lactic acid bacteria and Acinetobacter spp. predominated in 20% CO2:80% O2 and 100% CO2 bulk packaging. Enterobacteriaceae, Pseudomonas spp. and Acinetobacter spp. predominated on rump steaks, from both feeding and packaging treatments, during the aerobic display period of 7 days.

Cytotoxicity of alkaline-tolerant dairy-associated Bacillus spp.

The cytotoxicity of five Bacillus spp. isolated from alkaline cleaning solutions in South African dairies was evaluated against McCoy mouse cells using a 3-(4,5-dimethylthiazol-2-yl)-2, 5-diphenyltetrazolium bromide (MTT)-based assay, confocal scanning laser microscopy and scanning electron microscopy. According to the MTT-based assay, two of the Bacillus isolates (Bacillus licheniformis 5 and B. pumilus 122) were cytotoxic to McCoy cells and the cytotoxic components were heat labile. Confocal scanning laser microscopy combined with fluorescent staining using propidium iodide and fluorescein diacetate indicated that cytotoxic effects occurred within 3 h, appeared to be membrane active and resulted in cell necrosis. Scanning electron microscopy showed that McCoy cells exposed to the cytotoxic components exhibited morphological damage.

Physiology of dairy-associated Bacillus spp. over a wide pH range.

Bacillus species isolated from alkaline wash solutions used for cleaning in place in South African dairy factories have been suggested to contaminate product contact surfaces of dairy processing equipment and result in post-pasteurization spoilage of milk and milk products. Growth and attachment of such Bacillus isolates under alkaline and acidic conditions have not been previously described. Therefore, the in vitro growth temperature and pH ranges, attachment abilities and hydrophobicity, and enzyme production capabilities of four Bacillus isolates (tentatively identified as B. subtilis115, B. pumilus122, B. licheniformis137 and B. cereus144) previously isolated from the alkaline wash solutions in a South African dairy were examined. Growth pH ranges were determined in buffered Standard One-like Nutrient Broth and in unbuffered 1% Milk Medium at pH values ranging from 3 to 12. Growth and attachment to stainless steel surfaces and production of protease and lipase enzymes were determined in 1% Milk Medium at pH 4, 7 and 10. Colony hydrophobicity of each isolate by the Direction of Spreading Method (DOS) was also determined at pH 4, 7 and 10. In addition, Arrhenius plots were used to examine the growth temperature ranges of the isolates. All isolates grew at pH values ranging from 4.5 to 9.5 in buffered Standard One-like Nutrient Broth, and from pH 4 to 10 in 1% Milk Medium. All isolates also attached to stainless steel at pH 4, 7 and 10 in 1% Milk Medium. Generally the attachment of B. subtilis115, B. pumilus122 and B. lichenformis137 to stainless steel surfaces was enhanced at pH 4 and 10, compared to pH 7. By contrast, the best attachment of B. cereus144 cells to stainless steel surfaces was at pH 7. Planktonic and attached cells of all isolates produced proteolytic enzymes at pH 7 and 10, but not at pH 4. Similarly, planktonic and attached cells of B. subtilis115, B. pumilus122 and B. licheniformis137 produced lipolytic enzymes at pH 7 and 10, and weak lipolysis was observed at pH 4. The Bacillus cereus144 isolate showed no lipolytic activity at pH 10. All isolates exhibited low hydrophobic properties at all pH values even though attachment to stainless steel at the same pH values occurred. None of the isolates grew below 11 degrees C or above 56 degrees C, and optimum growth temperatures were in the high mesophilic range (36-44 degrees C).

Characterization and determination of origin of lactic acid bacteria from a sorghum-based fermented weaning food by analysis of soluble proteins and amplified fragment length polymorphism fingerprinting.

The group that includes the lactic acid bacteria is one of the most diverse groups of bacteria known, and these organisms have been characterized extensively by using different techniques. In this study, 180 lactic acid bacterial strains isolated from sorghum powder (44 strains) and from corresponding fermented (93 strains) and cooked fermented (43 strains) porridge samples that were prepared in 15 households were characterized by using biochemical and physiological methods, as well as by analyzing the electrophoretic profiles of total soluble proteins. A total of 58 of the 180 strains were Lactobacillus plantarum strains, 47 were Leuconostoc mesenteroides strains, 25 were Lactobacillus sake-Lactobacillus curvatus strains, 17 were Pediococcus pentosaceus strains, 13 were Pediococcus acidilactici strains, and 7 were Lactococcus lactis strains. L. plantarum and L. mesenteroides strains were the dominant strains during the fermentation process and were recovered from 87 and 73% of the households, respectively. The potential origins of these groups of lactic acid bacteria were assessed by amplified fragment length polymorphism fingerprint analysis.

Bacterial counts associated with poultry processing at different sampling times.

Aerobic plate counts, Enterobacteriaceae counts and Pseudomonas counts were performed on neck skin samples from six processing steps in a poultry abattoir at three different sampling times. Sampling time 1 was shortly after start-up of processing operations, time 2 after a tea break which was preceded by a cold water rinse-down of equipment surfaces, and time 3 before shut-down. No significant differences (P > 0.05) in microbial numbers of neck skin samples were observed between the three sampling times at the six sampling sites. At this particular processing plant, therefore, sampling at any time of the processing shift would thus not lead to significantly different bacterial counts of neck skins. The lowest aerobic plate counts, over all three sampling times, were obtained for neck skins sampled after spray washing, and the highest for neck skins sampled after packaging. This indicated the efficacy of the washing step in reducing microbial contamination but subsequent re-contamination of carcasses. Despite the Pseudomonas counts of neck skins being lower than the Enterobacteriaceae counts at the beginning of processing, packaging of carcasses resulted in Pseudomonas counts that were higher than the Enterobacteriaceae counts.

Combined effect of modified atmosphere bulk packaging, dietary vitamin E supplementation and microbiological contamination on colour stability of Musculus gluteus medius.

Ten Simmentaler type calves were fed diets supplemented with 500 mg vitamin E per day and ten fed control diets. After completion of a 100 day feeding period the cattle were slaughtered and rump steaks (M. gluteus medius) from each carcass PVC-overwrapped and subsequently bulk packaged in 100% CO(2) or 20% CO(2): 80% O(2). Steaks with low levels of bacteriological contamination were also prepared and packaged. Bulk packs were stored at 4°C for 0, 14, 28 and 42 days and the PVC-overwrapped samples subsequently displayed for 0, 4 and 7 days at 4°C. After display saturation, surface metmyoglobin and oxymyoglobin accumulation of the steaks were determined and acceptability of the steaks assessed by sensory evaluation using a trained panel. The dietary vitamin E supplemented steaks were more acceptable than the steaks from cattle not supplemented with vitamin E. Steaks prepared with low levels of bacteriological contamination, supplemented with dietary vitamin E, were more acceptable and discoloured less than all the other treatments. Beef rump steaks bulk packaged in 20% CO(2): 80% O(2) or 100% CO(2) were acceptable and colour stable for up to 14 days bulk storage at 4°C and a subsequent 2 days retail display at 4°C.