Variability of Escherichia coli and Enterobacteriaceae counts on pig carcasses: A systematic review

BackgroundIndicator bacteria are considered an interesting target for microbiological analysis in order to obtain information about the hygiene of processes and products. Quantitative analysis targeting indicator bacteria offers several advantages over pathogenic bacteria since indicator bacteria prevalence and counts are higher, and their distribution is more homogeneous, than pathogens. However, although microbiological criteria addressing indicator bacteria are already in place in industrialised countries such as the EU and US, information on factors affecting their counts on carcasses at slaughter is scattered.ObjectivesThe aim of this review was to provide valuable information about the variability of indicator bacteria along the pig slaughterline, and to suggest suitable means for the implementation of GMP (Good Manufacturing Practices) and process hygiene in pig slaughterhouses.ConclusionsAlong the slaughterline, scalding and chilling stages effectively reduce both Escherichia coli and Enterobacteriaceae counts on pig carcasses and have to be considered in the implementation of HACCP systems. Surface pasteurisation under experimental slaughter conditions was also effective; in contrast, washing did not reduce microbial contamination on pig carcasses. Evisceration was confirmed as a key contamination point, in particular regarding Enterobacteriaceae, but it has been suggested that bacterial load can be controlled by anal plugging if it is performed according to good manufacturing practices (GMP).     

Potential of Escherichia coli as a surrogate indicator for postchill broiler carcasses with high Campylobacter counts

Surrogating Campylobacter contamination level in broiler carcasses with other bacterial indicators, used to evaluate the hygienic status of the slaughterline operations, might be stimulation to the broiler meat industry to improve control of Campylobacter during slaughter. Theoretically, Escherichia coli might have some practical merits as a potential indicator for carcasses contaminated with Campylobacter. This study investigates the correlation between the counts of E. coli and Campylobacter in 231 postchill broiler carcasses. The impact of setting a process hygiene target based on E. coli counts on reducing the frequency of carcasses contaminated with Campylobacter at level of ≥3 log₁₀ CFU/g was also investigated. Almost half (48.9% (46/94)) of the carcasses with enumerable Campylobacter (≥1 log₁₀ CFU/g) had E. coli counts between 3 and 4 log₁₀ CFU/g. In addition, 54.8% (17/31) of the carcasses contaminated with Campylobacter of ≥3 log₁₀ CFU/g were correlated with E. coli count range of ≥3 & <4 log₁₀ CFU/g. A theoretical scenario assuming that hygiene and processing measures could allow achieving a target for E. coli that not exceeding 3 log₁₀ CFU/g showed a parallel impact on Campylobacter contamination in broiler carcasses. In such scenario, the overall number of Campylobacter-positive carcasses could be dropped from 40.6% to 12.5%; in addition, 80.6% (25/31) of the carcasses contaminated with Campylobacter of ≥3 log₁₀ CFU/g could be eliminated. Findings from this study reveal that a hygiene target based on E. coli count could be used as an indirect sanitary tool for reducing the level of Campylobacter contamination in postchill broiler carcasses.     

Microbiological contamination of sheep carcases in Finland by excision and swabbing sampling

The aims of this study were to provide data on the microbiological contamination of sheep carcases slaughtered in Finland, and to compare excision and swabbing methods for microbiological sampling of sheep carcases. The results were also compared to the requirements laid down in Regulation (EC) No 2073/2005. A number of 50 sheep carcases were sampled both by excision and by swabbing with gauze at four slaughterhouses. The samples were analysed for total viable counts (TVC's), Enterobacteriaceae, and Escherichia coli. The same carcases were also sampled for Salmonella spp. by swabbing. The mean levels of TVC's and Enterobacteriaceae, and the results for Salmonella spp. were in line with the requirements laid down in Regulation (EC) No 2073/2005. The mean levels of TVC's for samples by excision and by swabbing were 3.77 log CFU/cm² and 3.16 log CFU/cm², respectively. Enterobacteriaceae were recovered from 72% and 76% of the carcases, and E. coli from 48% and 61% by excision and swabbing, respectively. No carcases were found positive for Salmonella spp. The mean levels of TVC's, Enterobacteriaceae and E. coli were significantly higher for the samples by excision than for the samples by swabbing. When the relationship between the sampling methods were analysed, the results for excision samples and swabbing samples were related for the TVC's, and there was a significant correlation between the sampling methods for Enterobacteriaceae. The results suggest that swabbing by gauze can be used as an alternative sampling method to excision. In addition, a significant correlation was seen between the results for Enterobacteriaceae and E. coli for the samples collected by swabbing, suggesting that E. coli can be used as an indicator bacterium instead for Enterobaceteriaceae for sampling of sheep carcases by swabbing.     

A risk-based approach for evaluation of hygiene performance at pig slaughter

In Denmark, the pig slaughterhouses have a daily input of pigs infected and/or contaminated with Salmonella, and the slaughter hygiene has major influence on the level of Salmonella contamination on the meat leaving the slaughterhouse. However, the relationship between the effect of improved hygiene performance and the consequential reduction of human health risk has not been estimated so far. In this study, swab samples from 2702 pig carcasses were collected, originally for other purposes, from five large Danish slaughterhouses in a period from 2005 to 2007, covering all seasons of the year. The samples were analysed quantitatively for E. coli and semi-quantitatively for Salmonella. A positive association between the number of E. coli on carcasses and the prevalence of Salmonella positive carcasses was shown. For carcasses positive for Salmonella, a positive association was also shown between the number of E. coli and the number of Salmonella on the carcass. As no biological association has been reported between faecal shedding of E. coli and presence of Salmonella, the relationship was considered to be associated with the level of faecal contamination. The positive association between E. coli and Salmonella was used as basis for developing a quantitative risk assessment model for Salmonella, using the level E. coli as model input. The model output associated the hygiene performance with a relative risk estimate of human salmonellosis. The overall objective was to develop a decision support tool that can be used to support risk-based hygiene interventions in pig slaughterhouses.     

Modeling the inactivation of Escherichia coli O157:H7 and Uropathogenic E. coli in ground beef by high pressure processing and citral

Escherichia coli O157:H7 are a well-known intestinal foodborne pathogen which were responsible for numerous foodborne illness outbreaks. Uropathogenic E. coli (UPEC) are common contaminants in meat and poultry, and may cause urinary tract infections after colonizing the gastrointestinal tract followed by accidental transfer of contaminated feces from the anus to the urethra. High pressure processing (HPP) has been demonstrated an effective means in reducing pathogenic E. coli levels in meat and poultry. Citral, an antimicrobial, also demonstrated some inactivation effect on pathogenic E. coli in ground beef (ca. 0.5–1.0 log CFU/g reduction at 1% w/w without HPP). With HPP alone, to achieve 5 log CFU/g reduction required a 500 MPa level and 15 min (for both O157:H7 and UPEC). However, 1% of citral addition may lower the pressure requirement to 380 MPa and 15 min which could reduce the food quality damage. To effectively inactivate E. coli O157:H7 and UPEC in meat, high pressure processing (HPP) in combination with the antimicrobial citral was studied. Ground beef inoculated with E. coli O157:H7 or UPEC were treated at different HPP conditions (250–350 MPa; 10–20 min), and citral (0.75–1.25%, w/w) following a central composite experimental design. Quadratic linear regression equations were developed to describe and predict the reductions of E. coli O157:H7 (R² = 0.93, p < 0.001) and UPEC (R² = 0.92, p < 0.001). Dimensionless nonlinear models consisting of three impact factors were also developed and compared with the linear models. These models were experimentally validated. Citral enhanced the inactivation of pathogenic E. coli, increasing the effectiveness of the HPP process. The models may assist the food industry and regulatory agencies in risk assessment of E. coli O157:H7 and UPEC on ground meats.     

Development of a real-time PCR procedure for quantification of viable Escherichia coli in populations of E. coli exposed to lactic acid,and the acid tolerance of verotoxigenic E. coli (VTEC) from cattle hides

The aims of this study were to develop a real-time PCR procedure for determining the effects on Escherichia coli of treatment for decontaminating beef carcasses with lactic acid solution, and determining if there were differences in the acid tolerance of E. coli generally and verotoxigenic E. coli (VTEC). Suspensions of E. coli were incubated with 4% lactic acid at pH 3.6. The numbers of surviving E. coli at different incubation times were determined from plate counts and from quantification by real-time PCR of the uidA gene in DNA preparations. The numbers of viable E. coli progressively declined, by about 4 log units during incubation for 6 h. The mean cycle threshold (Ct) values for uidA in DNA from samples collected at different times and treated or not treated with propidium monoazide (PMA) before DNA extraction were similar. Treatment with 1% sodium deoxycholate (SD) before PMA treatment resulted in an increase of >6 Ct when the reduction in viable cell number was around 1 log. When E. coli incubated with 4% lactic acid solutions of pH 2.4, 2.8, 3.2 or 3.6 were resuscitated in half strength brain heart infusion (BHI) for 2 h before treatments with SD and PMA, the slope of the plot relating Ct values to the numbers of viable E. coli was 1.85 Ct log cfu⁻¹ with the correlation coefficient (R²) being 0.80. The findings indicate that while the membranes of E. coli inactivated by 4% lactic acid were largely impermeable to PMA, the membranes of both dead and injured cells were rendered permeable to PMA by treatment with 1% SD. Resuscitation in BHI restored the membrane barrier properties of the injured cells. Treatment with lactic acid resulted in increases in Ct values of 4.1, 3.7, 2.5 and 1.5 for the uidA, stx₁, stx₂ and eae genes, respectively; and the increases in Ct values for the latter two genes were significantly different (p < 0.05) from that for the uidA gene. This indicates that VTEC carrying stx₂ and/or eae were more acid resistant than other E. coli. Thus, caution should be exercised when using generic E. coli as an indicator for VTEC for assessment of the antimicrobial efficacy of organic acid decontaminating treatments at abattoirs.     

Time trends in the prevalence of Escherichia coli and enterococci in bivalves harvested in Norway during 2007–2012

The objectives of this study were to describe time trends in the prevalence of Escherichia coli and enterococci in cultured blue mussels (Mytilus edulis) harvested from 152 localities along the coast of Norway during the six-year period from 2007 to 2012. Based on the available data, possible co-occurrence of these two indicator organisms of faecal contamination was assessed. Several localities for bivalve cultivation in Norway showed single high counts of E. coli, without any previous history of E. coli detection. For other localities, the pattern of E. coli detection was recurring, however low values were found, with some sporadic findings of higher values. There was a weak positive correlation between the detection of enterococci and E. coli, and a weak positive correlation between counts of E. coli/enterococci and rainfall. Sampling intervals should take into account knowledge of the occurring variation for bacterial faecal indicators, local knowledge on possible exposure to faecal material from livestock or humans, rainfall seasons, topography of the location, as well as tidal and water current patterns.     

Antimicrobial action and effects on beef quality attributes of a gaseous ozone treatment at refrigeration temperatures

The effects of gaseous ozone treatment at refrigeration temperatures, on microbial counts (total aerobic mesophilic heterotrophic microorganisms and inoculated Escherichia coli) in culture media and in beef samples were analyzed. The influence of ozone on beef quality properties such as surface color and rancidity was measured.The effect of gaseous ozone (154 × 10^?6 kg m^?3) in culture media inoculated with E. coli after 3- or 24-h treatment at 0° and 4 °C caused a total inactivation of this microorganism.For beef samples treated with the same gaseous ozone concentration, the highest microbial inhibition was observed at 0 °C and after 24-h exposure, producing a decrease of 0.7 and 2.0 log₁₀ cycles in E. coli and total aerobic mesophilic heterotrophic microorganism counts respectively; however, both the surface color and lipid oxidation of these beef samples were unacceptable. Shorter exposure times (3 h) to the tested ozone concentration at both temperatures (0–4 °C), reduced 0.5 log₁₀ cycles the counts of total aerobic mesophilic heterotrophic microorganisms and 0.6–1.0 log₁₀ cycles the counts of E. coli, without changing the color or producing rancidity in beef.     

Efficacy of the combined application of oregano and rosemary essential oils for the control of Escherichia coli,Listeria monocytogenes and Salmonella Enteritidis in leafy vegetables

This study assessed the effect of the combined application of essential oils (EOs) from Origanum vulgare L. – oregano (OVEO) and Rosmarinus officinalis L. – rosemary (ROEO), alone or in combination at subinhibitory concentrations, against three pathogenic bacteria that are associated with fresh leafy vegetables: Listeria monocytogenes (L. monocytogenes), Escherichia coli (E. coli) and Salmonella enterica Serovar Enteritidis (S. Enteritidis). The inhibitory effects were evaluated by determining the minimum inhibitory concentration (MIC) and the fractional inhibitory concentration index (FICI) and assessing the viable cell counts in vegetable broth and artificially infected vegetables over time. Still, the effects of the EOs on native spoilage native flora were assessed. The MIC of OVEO was 0.6 μL/mL against the test strains either in single and mixed inoculum. The MIC of ROEO was 5 μL/mL against L. monocytogenes and E. coli and 10 μL/mL against S. Enteritidis in single inocula, whereas it was 10 μL/mL against the mixed inoculum. The FICI of the combined EOs was 0.5 against the mixed bacterial inoculum, which suggested a synergic interaction. The incorporation of OVEO and ROEO alone (MIC) or combined at different subinhibitory concentrations in vegetable broth resulted in a decrease in the viable cell counts of all test strains over 24 h. Similarly, the EOs alone or in the tested combinations reduced the viable cell counts of all test strains in experimentally infected fresh vegetables, besides to decrease the counts of spoiling native flora (mesophilic bacteria, enterobacteria and fungi). These findings reinforce the rational for the use of OVEO and ROEO in combination at subinhibitory concentrations to guarantee the safety and extend the shelf life of fresh vegetables.     

Good agricultural practices in a Brazilian produce plant

A Brazilian produce plant was evaluated according to good agricultural practices (GAPs) using a checklist. The agricultural water was sampled for total and faecal coliform and Escherichia coli counts. Equipment/handler sanitary facilities, the handling of agrochemicals and pesticides, records of the composting process, prior analysis of the irrigation water for the identification of heavy metals and pesticide residues, inadequate temperature control of the raw material storage room, qualified person to manage the operations and periodic training programmes on food manipulation, hygiene and microbiology, were the main critical points found during the inspection. Microbiological analyses of the irrigation water gave counts between 0.3 and 1.2 log MPN/ml for total and faecal coliforms, respectively and absence of E. coli. The results are discussed and the necessary corrections suggested in order to improve process safety.     

Assessment of microbiological quality of raw fruit juice vended in Dar es Salaam city,Tanzania

Fresh fruit juices are highly nutritious food for human but the hygiene involved during preparation, packaging and storage make fresh juices prone to microbial contamination. This study was conducted to assess bacterial quality and establish the risk factors for contamination of raw fruit juices vended in Dar es Salaam city, Tanzania. Ninety fruit juice vendors were assessed for possible factors of microbial contamination in fruit juices. One juice sample per vendor was collected for microbial analysis using standard laboratory protocols of International Standards Organisation (ISO), Tanzania Bureau of Standards and Codex specifications. The results showed that the total plate counts (TPC) ranged between 2.32 and 8.54 (Log cfu/ml). About 72.2% of juice samples had TPC above Codex recommended maximum levels (3.7–4.7 Log cfu/ml). The prevalence of Escherichia coli in the juices was 80% with a range between 0.0 and 5.0 (Log MPN/ml) suggesting of direct faecal contamination or contamination from the environment. All samples were negative for Salmonella species. Risk factors for high TPC and E. coli counts which were statistically significant (P < 0.05) included type of juice, extraction methods, vending sites, storage containers and sex of the vendors. Generally, 78.9% of preparation and vending premises were unhygienic and encouraged contamination of the juices. It is concluded that, the overall handling, preparation practices and bacterial quality of unpasteurized fruit juices vended in Dare es Salaam city are poor. The government should educate the vendors on food safety and hygiene as well as enforcing regular monitoring of the quality of street fruit juices.     

Effects of lactic acid and lauricidin on the survival of Listeria monocytogenes,Salmonella enteritidis and Escherichia coli O157:H7 in chicken breast stored at 4 °C

Lauricidin and lactic acid were evaluated for their effects on growth and survival of Listeria monocytogenes (L55), Salmonella enteritidis (S552) and Escherichia coli O157:H7 (E19) inoculated onto raw chicken breast. Fresh, raw chicken breasts were purchased immediately after slaughter and transported on ice to the laboratory within 20 min. Each chicken breast was decontaminated by briefly dipping in 70% ethanol and passed through a flame of a Bunsen burner and then allowed to cool. The decontaminated Chicken breast was dipped in TSB broth, at room temperature (25 °C) for 15 min, containing approximately log 9 CFU/ml of L. monocytogenes, S. enteritidis or E. coli O157:H7. Initial counts of L. monocytogenes, S. enteritidis or E. coli O157:H7 counts in chicken breast immediately after dipping in TSB broth were in the range of log 7–log 8 CFU/g. After inoculation, the chicken breasts were kept at room temperature for 20 min to allow attachment. Each inoculated chicken breast (25 °C) was dipped in 0 (control – sterile water), 0.5%, 1%, 1.5% or 2% of lauricidin (w/v) or lactic acid (v/v) for 10, 20 or 30 min and then individually placed in oxygen-permeable polyethylene bags. Breasts were subjected to microbiological analyses after treatment (day 0) and after storage for 2, 5, 7, 10 and 14 d at 4 °C. Initial counts of L. monocytogenes, S. enteritidis and E. coli O157:H7, in chicken breast treated with lauricidin decreased by 2.90, 1.31 and 2.27 log CFU/g, respectively. Lauricidin was more effective in reducing L. monocytogenes population than S. enteritidis and E. coli O157:H7 population. Dipping chicken breast in lauricidin for 30 min caused a significant reduction of L. monocytogenes, S. enteritidis and E. coli O157:H7 population compared to 10 and 20 min dipping. Initial L. monocytogenes, S. enteritidis and E. coli O157:H7 counts on chicken breast treated with lactic acid decreased by 1.97, 1.71 and 2.59 log CFU/g, respectively. Lactic acid caused a higher reduction in initial S. enteritidis and E. coli O157:H7 counts compared to lauricidin.     

Hygienic conditions and microbiological status of chilled Ready-To-Eat products served in Southern Spanish hospitals

Chilled Ready-To-Eat (RTE) foods are of concern in hospital foodservices because they can support microbial growth when subjecting to time/temperature abuses during processing and distribution together with poor handling practices. This study was conducted in five different hospitals (A–E) of Southern Spain during 2008–2009 to perform an evaluation of their sanitary conditions and microbiological quality of two RTE meals: lettuce salads and cooked ham. A checklist based on hygiene principles embedded in Food European legislation was developed and applied in each hospital. In parallel, microbiological analysis of food contact surfaces, air quality and time/temperature measurements along the distribution chain were carried out. RTE samples (n = 150) were examined for mesophilic aerobic bacteria (MAB), total coliforms, coagulase-positive Staphylococci (CPS), Escherichia coli, Listeria spp. and Salmonella spp. Differences were found between hospitals regarding handling practices and cleanliness of working surfaces. Cooked ham samples presented lower counts of MAB and total coliforms (<10³ and <10 cfu/g respectively) than lettuce salads (10⁴ to 10⁵ and 10 to 10⁴ cfu/g respectively), although concentration of CPS was higher in cooked ham samples reaching maximum levels close to 10³ cfu/g. Neither Listeria spp. nor Salmonella spp. were detected in any food sample. Prevalence of E. coli was low (3%). Surface counts and air quality presented high variability among the different hospitals evaluated. It was concluded that good manufacturing practices and HACCP principles should be followed together with special training of food handlers. This study can help risk managers to better define the control measures to be adopted in healthcare settings in order to prevent foodborne infections.     

Combination of bioprotective cultures with EDTA to reduce Escherichia coli O157:H7 in frozen ground-beef patties

The effectiveness of bacteriocin-producing Lactobacillus curvatus CRL705 and Lactococcus lactis CRL1109 in combination with Na₂EDTA on frozen ground-beef patties contaminated with Escherichia coli O157:H7, was investigated under temperature abuse conditions (5 °C during 9 days). The presence of the bioprotective cultures (ca. 10⁷ CFU/g) and chelator (48 mM) resulted in one log CFU/g reduction for E. coli strain, compared to the control on day 0. Similarly, a significant decline for indigenous coliforms in ground-beef patties was also observed in the presence of bacteriocinogenic strains and chelator. However, in the absence of Na₂EDTA, neither E. coli nor coliforms were inhibited by the bioprotective cultures, the pathogen reaching similar counts than control samples (5.22 and 3.60 log CFU/g, respectively) at 9 days. When the growth of bacteriocinogenic strains on patties was evaluated, they were able to increase their population producing bacteriocins after 48 h up to the end of incubation period while a near neutral pH in the presence of Na₂EDTA was detected. Non substantial effect on of ground-beef patties color was produced in the presence of bioprotective cultures, while a darker color developed in those added with the chelator. The simultaneous treatment with bioprotective cultures and Na₂EDTA may be of value for the control of E. coli O157:H7 in temperature abused ground-beef patties.     

Efficacy of pulsed electric fields for the inactivation of indicator microorganisms and foodborne pathogens in liquids and raw chicken

This study investigated the ability of pulsed electric fields (PEF) to inactivate a range of microorganisms in liquid media and on raw chicken meat. The susceptibilities of ten Campylobacter isolates (seven Campylobacter jejuni isolates and three Campylobacter coli isolates), Escherichia coli (ATCC 25922) and Salmonella Enteritidis (ATCC 13076) to PEF in liquid media were investigated. All Campylobacter isolates tested in liquid were susceptible to PEF treatment (65 kV/cm, 5 μs, 500 Hz) with reductions of between 4.33 and 7.22 log₁₀ CFU/mL observed for the longest treatment (30 s). Significant differences in susceptibility were observed between Campylobacter isolates subjected to equivalent PEF treatments ranging from 2.41 to 5.19 log₁₀ CFU/mL. Campylobacter isolates processed in liquid media were found to be more sensitive to PEF than E. coli and S. Enteritidis (P < 0.05). The application of PEF (3.75 and 15 kV/cm, 10 μs, 5 Hz) to inoculated samples of raw chicken did not result in any significant reductions in total viable counts, Enterobacteriaceae, C. jejuni, E. coli or S. Enteritidis. Therefore, under the conditions used in this study, PEF technology may not be suitable as a food safety intervention measure for the control of microbial contaminants on broilers during processing although it may have potential to reduce contamination of process water (e.g. scald or spin chill water).     

Microbiological evaluation of fresh-cut organic vegetables produced in Zambia

This study was undertaken to assess the microbiological quality of fresh-cut organic vegetables produced in Zambia. Fresh-cut organic mixed vegetables and green beans produced in Zambia were analysed for aerobic plate counts, coliforms, Enterobacteriaceae, Escherichia coli, Bacillus cereus, Clostridium perfringens, Listeria monocytogenes, Salmonella spp., Staphylococcus aureus, and yeast and mould counts. The study included 160 samples for most of the parameters. The vegetables were grown on farms meant primarily for the export market. The vegetables were treated/washed with 150 μg ml⁻¹ chlorine solution at the processing plant prior to sampling. The aerobic plate count ranged between 3 log₁₀ and 9.7 log₁₀ CFU/g, with the highest count recorded for green beans. The largest grouping (26.1%) of vegetable samples fell between 3 and 4 log₁₀ CFU/g. Coliform counts were between 1.0 log₁₀ and 7.7 log₁₀ CFU/g. The highest incidence level was 31.4% for total coliform counts between 3 log₁₀ and 4 log₁₀ CFU/g. E. coli was only detected on mixed vegetables in the range of 0.6 log₁₀ to 3 log₁₀ CFU/g, while Enterobacteriaceae counts ranged between 1.6 log₁₀ and 9.8 log₁₀ CFU/g with the highest counts being found on green beans. The highest incidence level was of 25.8% for counts within the same range as the aerobic plate counts. Yeast and mould counts showed the highest incidence level between 5 log₁₀ and 6 log₁₀ CFU/g with an overall range between 1.5 log₁₀ and 5.6 log₁₀ CFU/g. L. monocytogenes, Salmonella spp. and S. aureus were detected in 20%, 23.1% and 83.9% of samples, respectively . C. perfringens and B. cereus were not detected in any of the samples analysed.     

Persistence of Escherichia coli following irrigation of strawberry grown under four production systems: Field experiment

A two-year field experiment was conducted in order to evaluate the persistence of generic Escherichia coli in strawberry after irrigation with naturally E. coli-contaminated surface water. Sixteen experimental plots representing actual field conditions were set, including two methods of irrigation (overhead and subsurface drip) and two mulch types (straw and plastic). Two irrigations were performed each year with water having an E. coli content varying between 460 and 2242 CFU per 100 ml. Strawberries were harvested before irrigation and 1, 4 and 24 h following irrigation. E. coli counts could not be determined in any of the 256 strawberry samples. Enrichment procedure revealed more positive samples under straw mulch (6.4%) compared to plastic mulch (4.3%), but this difference was not statistically significant (P = 0.3991). Higher strawberry contamination was also observed in overhead irrigation treatments (8.6%) compared to drip irrigation (2.1%) (P = 0.0674). The risk to detect E. coli in overhead-irrigated strawberries was 4.5-fold higher than in strawberries under drip irrigation. Four hours following irrigation, the risk to detect E. coli in fruits was 4.0-fold lower than the risk observed 1 h after irrigation. Increasing the delay to 24 h led to a 7.4-fold lower risk. In actual conditions that may be encountered in strawberry productions, this study showed a limited persistence of E. coli in strawberries following irrigation.     

Ratio between carcass-and skin-microflora as an abattoir process hygiene indicator

In two abattoirs, each slaughtering both cattle and pigs, 100 cattle and 100 pigs were randomly selected and sampled. From each animal, two samples were taken: a) immediately after sticking of bovines or stunning of pigs, approximately 2000 cm² hide (cattle) or 1500 cm² skin (pigs) areas were sponge-swabbed; and b) at the end of slaughter line but before chilling, the same areas on corresponding dressed carcasses were sponge-swabbed. In each swab-sample (400 in total), total viable count (TVC) and Enterobacteriaceae count (EC), as well as Escherichia coli O157 (in cattle) or Salmonella (in pigs) occurrence, were determined and used to assess process hygiene in the abattoirs. The results indicated that simply fitting mean TVC and/or EC on final carcasses into an acceptable, marginal or unacceptable process hygiene category (according to current microbiological EU process hygiene criteria) did not enable characterisation of each process with respect to its ability to reduce the transfer of incoming microbial loads (i.e. on skins) onto dressed carcasses. On the other hand, determining the ratio between mean TVC and/or EC on final carcasses and those on corresponding skins enabled more precise assessment of the hygiene of each abattoir process, as well as more reliable differentiation between abattoirs. On the other hand, occurrence of E. coli O157 in cattle or Salmonella in pigs (skins and/or carcasses), being dependant on varying factors including those on-farm/pre-abattoir, did not appear to be very useful for characterisation of the process hygiene but is valuable for the purposes of consumer exposure assessment and pathogen reduction.     

Trends in the occurrence and characteristics of Campylobacter jejuni and Campylobacter coli isolates from poultry meat in Northern Poland

The aim of this study was to investigate the occurrence of Campylobacter spp. in poultry meat available in retail stores in the northern part of Poland during a five-year period (2009–2013). A total of 742 poultry meat samples were collected from butcher shops and supermarkets including the following types of samples: chicken breast filets (n = 133), turkey breast filets (n = 112), chicken wings (n = 135), chicken leg quarters (n = 128), chicken drumsticks (n = 115), and chicken giblets (n = 119).The results indicated that 41.6% of the samples were positive for Campylobacter spp., and Campylobacter jejuni was predominant in this study. The prevalence of Campylobacter spp. changed during the study period, decreasing from 60.2% in 2009 to 32% in 2013.The characterization of the isolates revealed a high prevalence of Campylobacter virulence genes. All Campylobacter spp. isolates from poultry meat contained the cadF gene, which is responsible for adherence. The flaA gene, which is involved in motility, was present in all C. jejuni and Campylobacter coli strains. The cdtB, which is associated with toxin production, was present in 93.3% of C. jejuni strains and 89.6% of C. coli strains. The iam gene, which is associated with the invasiveness of Campylobacter spp., was predominant in C. coli strains (95.6%) compared to C. jejuni strains (84.5%).Resistance to four antimicrobials was also examined. The prevalence of resistance among the obtained C. jejuni and C. coli isolates was as follows: ciprofloxacin (62.8% and 72.2%, respectively), tetracycline (42.3% and 42.6%, respectively), erythromycin (3% and 1.7%, respectively) and azithromycin (1%). Multidrug resistance was more frequent among C. jejuni isolates (29.8%) than among C. coli isolates (18.2%).In conclusion, the results of this study demonstrated the importance of poultry meat as a source of Campylobacter spp., especially macrolide-resistant strains. The trend of decreasing Campylobacter spp. occurrence in retail poultry meat in this region of Poland requires further investigation, and monitoring.     

Evaluation of the factors affecting the activity of sakacin C2 against E. coli in milk

Sakacin C2 is a novel, broad spectrum bacteriocin secreted by Lactobacillus sake C2 isolated from Chinese traditional fermented cabbage. Effects of milk fat, emulsifiers, preservatives and homogenization on the activity of sakacin C2 against Escherichia coli ATCC 25922 in milk were evaluated by determined the changes of viable cell counts of E. coli ATCC 25922 after inoculation (10⁵ CFU/ml) then storage at 4 °C. Milk fat in pasteurized and homogenized milk products (low fat milk and whole milk) decreased the activity of sakacin C2 against E. coli ATCC 25922. Addition of 5 μl/ml of Tween 80 decreased the effect of sakacin C2 against E. coli ATCC 25922, but 5 mg/ml of lecithin increased the effect against E. coli ATCC 25922 in pasteurized and homogenized whole milk. Nisin and ε-Polylysine significantly increased the effect of sakacin C2 against E. coli ATCC 25922 in pasteurized and homogenized whole milk. Homogenization interfered with the activity of sakacin C2 against E. coli ATCC 25922 in pasteurized skim milk and whole milk. This study might lay the groundwork for the application of sakacin C2 as bio-preservative in milk and dairy products.