Ulrike Lyhs

Microbiological quality and shelf-life of vacuum-packaged 'gravad' rainbow trout stored at 3 and 8 °C

Microbiological and sensory changes of vacuum-packaged ‘gravad’ rainbow trout slices were studied during storage at 3 and 8 °C. At the time of spoilage, after 27 and 20 days of storage at 3 and 8 °C, respectively, both mesophilic viable counts (MVC) and psychrotrophic viable counts (PVC) reached 106–107cfu/g at 3 °C and 107–108 cfu/g at 8 °C. H2S-producing bacteria constituted a high proportion of the PVCs and lactic acid bacteria (LAB) counts were lower than the other determined bacterial counts. Sensory scores decreased with increasing MVC and PVC. The judges considered samples unfit for human consumption at MVC and PVC levels exceeding 106 and 107 cfu/g for samples stored at 3 and 8 °C, respectively. At respective levels of 107 and 108 cfu/g, most of the samples were deemed unfit. The main reasons for sensory rejection at both storage temperatures were the lack of the typical product odour or an ammonia off-odour and colour change to dark violet. The shelf-lives of the rainbow trout slices based on microbiological and sensory analyses were 20 days and 18 days at 3 and 8 °C, respectively.

Characterisation of lactic acid bacteria from spoiled, vacuum-packaged, cold-smoked rainbow trout using ribotyping

A total of 405 lactic acid bacteria (LAB) isolated from spoiled, vacuum-packaged, salted, sodium nitrite- or potassium nitrate-treated, cold-smoked rainbow trout stored at 4 degrees C or 8 degrees C were characterised and identified using a molecular method. The isolates were initially classified according to their restriction endonuclease profiles using HindIII and EcoRI restriction endonucleases and further characterised by rRNA gene restriction patterns (ribotypes). Numerical analysis of these ribopatterns was performed together with 19 reference LAB strain patterns in order to identify the isolates to species level. The strains were divided with HindIII and EcoRI ribopatterns into ten and nine clusters at the similarity level of 65% and 50%, respectively. The Leuconostoc-clusters and the Lb. sakei/Lb. curvatus-clusters formed the two main groups. Only one isolate was identified as Lactobacillus plantarum and no Carnobacterium strains were discovered. For both enzymes, the 35 isolates possessing six individual ribotypes and forming five clusters could not be identified further with the reference strains used. The relative proportion of Leuconostoc mesenteroides subsp. mesenteroides was higher in all samples stored at 4 degrees C. Most of the Leuconostoc citreum were found in the samples stored at 8 degrees C, and particularly in the nitrite-treated samples.

Leuconostoc gelidum and Leuconostoc gasicomitatum strains dominated the lactic acid bacterium population associated with strong slime formation in an acetic-acid herring preserve.

Spoilage characterised by strong slime and gas formation affected some manufacture lots of an acetic-acid Baltic herring (Culpea haerengus membras) preserve after few weeks of storage at 0-6 degrees C. The product consisted of herring filets in acetic acid marinade containing sugar, salt, allspice and carrot slices. Microbiological analyses of the spoiled product showed high lactic acid bacterium (LAB) levels ranging from 4.5x10(8) to 2.4x10(9) CFU/g. Yeasts were not detected in any of the herring samples. Since LAB contaminants are seldom associated with fresh fish, LAB populations associated with marinade ingredients (carrots, allspice) were also analyzed. The highest LAB levels exceeding 10(7) CFU/g were detected in equilibrium modified atmosphere packaged baby carrots whereas the levels detected in the allspice samples did not exceed 4.3x10(5). A total of 176 randomly selected LAB isolates originating from herring, carrot and allspice samples were further identified to species level using a 16 and 23S rRNA gene RFLP (ribotyping) database. Leuconostoc gelidum and Leuconostoc gasicomitatum strains dominated both in the spoiled herring and carrot samples. These species are heterofermentative-producing CO(2) from glucose and they also produce dextran from sucrose. Inoculation of some commercial-herring products with spoilage-associated L. gelidum and L. gasicomitatum strains verified that these strains have the capability of producing slime and gas in herring preserves although slime formation was not as strong as in the original samples. Since L. gelidum and L. gasicomitatum strains were commonly detected in carrots, carrot slices used for the fish marinade were considered to be the probable source of these specific spoilage organisms.

The spoilage flora of vacuum-packaged, sodium nitrite or potassium nitrate treated, cold-smoked rainbow trout stored at 4°C or 8°C

The spoilage flora of vacuum-packaged, salted, cold-smoked rainbow trout fillets, with or without the addition of nitrate or nitrite, stored at 4 degrees C and 8 degrees C, was studied. Of 620 isolates, lactic acid bacteria were the major fraction (76%), predominating in all samples of spoiled product. However, the phenotypical tests used were insufficient to identify the lactic acid bacteria to the species level. Gram-positive, catalase-positive cocci, gram-negative, oxidase-negative rods and gram-negative, oxidase-positive rods were found in 6%, 16% and 2% of the samples, respectively. Of 39 gram-positive, catalase-positive cocci, 29 were identified as staphylococci and 10 as micrococci. Eighty-five isolates were found to belong to the family Enterobacteriaceae, with 45 of those being Serratia plymuthica. Eleven isolates from the nitrate treated samples stored at 8 degrees C were identified as Pseudomonas aeruginosa. The occurrence of P. aeruginosa and staphylococci in the nitrate-containing samples, stored at 8 degrees C, may cause problems with respect to the safety of the product. The types of lactic acid and other bacteria in the spoilage flora were generally reduced by the addition of nitrate or nitrite to fillets.

Use of ultraviolet irradiation to reduce Campylobacter jejuni on broiler meat

The effects of UV irradiation at a wavelength of 254 nm on the survival of Campylobacter jejuni on the surfaces of broiler meat, skin, and carcasses were studied. On broiler carcasses, the effects of UV were also studied in combination with activated oxygen. The surfaces were inoculated with varying counts of C. jejuni and treated with UV irradiation using doses ranging between 9.4 and 32.9 mW/s per square centimeter. The log reductions in C. jejuni counts were determined by dilution plating. The effects of both treatments on the sensory quality of broiler meat, including visual appearance, odor, and fatty acid composition, were also evaluated. On broiler meat, the maximum reduction achieved was 0.7 log and on broiler skin 0.8 log. On broiler carcasses, the maximum reduction using UV irradiation was 0.4 log, and using UV in combination with activated oxygen 0.4 log. No significant differences were found in the sensory quality between the samples and the controls. The use of UV irradiation alone or in combination with activated oxygen cannot be recommended as a primary decontamination method for C. jejuni on broiler carcasses. The use of these methods in combination with other decontamination techniques, and processing with proper processing plant sanitation and hygiene, might be more effective in reducing the C. jejuni counts on broiler carcass surfaces than the use of these methods only.

PCR assay for the detection of Campylobacter in marinated and non-marinated poultry products

During a period of 9 months, 194 marinated and non-marinated poultry products were collected from retail shops in a defined area in Western Finland and tested for Campylobacter spp. using a conventional enrichment culture and Polymerase Chain Reaction (PCR) method. For marinated poultry products, the study involved modification of a commercial DNA isolation method. Using either a conventional culture or PCR method, a total of 25 (12.9%) of all investigated samples were Campylobacter positive. In marinated poultry products, Campylobacter was detected at a prevalence of 21.1% and 9.5% in turkey and chicken products, respectively. In August, there was a peak with 28.9% positive Campylobacter samples. Campylobacter inoculation tests were carried out to test the detection limit of both methods. The PCR method used is faster than microbiological analyses. However, enrichment of the samples is necessary due to the low occurrence of Campylobacter spp. in retail Finnish poultry products.

Lactobacillus sakei/curvatus is the prevailing lactic acid bacterium group in spoiled maatjes herring.

A total of 164 lactic acid bacteria (LAB) isolated from spoiled maatjes herring stored in air and under modified atmosphere at 4 or 10 degrees C were characterised and identified using an rRNA gene restriction pattern (ribotype) database. The isolates were initially grouped according to their HindIII restriction endonuclease profiles and further identified to species level using numerical analysis. Lactobacillus sakei, Lactobacillus curvatus and strains of the L. curvatus spp./Lactobacillus fuchuensis group were the main species detected. Of all the isolates, six were identified as Lactococcus spp.

Characterisation of ropy slime-producing Lactobacillus sakei using repetitive element sequence -based PCR

Abstract Eighteen previously characterised Lactobacillus sakei strains exhibiting varying slime production capabilities in vacuum-packaged meat products were analysed using repetitive element sequence-based PCR (rep-PCR). The single primers BOXA1R and RW3A and the primer pair REP1R-Dt and REP2R-Dt were evaluated for their applicability in L. sakei genotyping. The five different patterns produced by RW3A were the least useful, with the discriminatory power equal to ribotyping. BOXA1R and REP–primer pair both produced six different banding patterns and the combination of these results yielded seven different rep-types. Rep-PCR was concluded to have approximately the same discriminatory power as randomly amplified polymorphic DNA (RAPD) analysis, but was inferior to pulsed-field gel electrophoresis (PFGE). However, if the results of rep-PCR and RAPD were combined, the discrimination was comparable to PFGE, with the exception that within Ribogroup I non-slime-producing strains were indistinguishable from weak slime producers. It was concluded that the combination of the two PCR-based typing techniques, rep-PCR and RAPD, would be a valuable tool in large scale contamination studies at meat processing plants, since results can be obtained rapidly with fewer isolates needing further analysis by PFGE.

Extraintestinal pathogenic Escherichia coli in poultry meat products on the Finnish retail market

BackgroundExtraintestinal pathogenic Escherichia coli bacteria (ExPEC) exist as commensals in the human intestines and can infect extraintestinal sites and cause septicemia. The transfer of ExPEC from poultry to humans and the role of poultry meat as a source of ExPEC in human disease have been discussed previously. The aim of the present study was to provide insight into the properties of ExPEC in poultry meat products on the Finnish retail market with special attention to their prevalence, virulence and phylogenetic profiles. Furthermore, the isolates were screened for possible ESBL producers and their resistance to nalidixic acid and ciprofloxacin was tested.MethodsThe presence of ExPEC in 219 marinated and non-marinated raw poultry meat products from retail shops has been analyzed. One E. coli strain per product was analyzed further for phylogenetic groups and possession of ten virulence genes associated with ExPEC bacteria (kpsMT K1, ibeA, astA, iss, irp2, papC, iucD, tsh, vat and cva/cv) using PCR methods. The E. coli strains were also screened phenotypically for the production of extended-spectrum β-lactamase (ESBL) and the susceptibility of 48 potential ExPEC isolates for nalidixic acid and ciprofloxacin was tested.ResultsE. coli was isolated from 207 (94.5%) of 219 poultry meat products. The most common phylogenetic groups were D (50.7%), A (37.7%), and B2 (7.7%). Based on virulence factor gene PCR, 23.2% of the strains were classified as ExPEC. Two ExPEC strains (1%) belonged to [O1] B2 svg+ (specific for virulent subgroup) group, which has been implicated in multiple forms of ExPEC disease. None of the ExPEC strains was resistant to ciprofloxacin or cephalosporins. One isolate (2.1%) showed resistance to nalidixic acid.ConclusionsPotential ExPEC bacteria were found in 22% of marinated and non-marinated poultry meat products on the Finnish retail market and 0.9% were contaminated with E. coli [O1] B2 svg+ group. Marinades did not have an effect on the survival of ExPEC as strains from marinated and non-marinated meat products were equally often classified as ExPEC. Poultry meat products on the Finnish retail market may have zoonotic potential.

The temporal, PFGE and resistance pattern associations suggest that poultry products are only a minor source of human infections in Western Finland

In order to compare human and retail poultry meat thermophilic Campylobacter isolates originating in a regional area in Western Finland, minimum inhibitory concentration (MICs) for six antimicrobials (96 isolates) and pulsed-field gel electrophoresis (PFGE) (102 isolates) were analysed. Campylobacter spp. were detected in 10.5% out of 305 fresh poultry products studied; 29 (90.5%) isolates were identified as Campylobacter jejuni. Among the 70 human isolates, 66 (94.3%) isolates were identified as C. jejuni. Only one C. jejuni domestic poultry isolate showed resistance (ampicillin), whereas domestic human C. jejuni isolates were more commonly resistant to ciprofloxacin, nalidixic acid, ampicillin and tetracycline. The resistance in foreign human isolates was significantly more common than among domestic isolates. PFGE analysis with KpnI restriction enzyme resulted in 59 different PFGE types among the poultry and human isolates. Three types were detected first in poultry meat and thereafter during the following month in domestic human samples, whereas the other conjoint types were detected only after many months. This study suggests that poultry products play only a minor role in human campylobacteriosis in the study area and that the resistance found in domestic human isolates is not likely related to retail poultry meat products.