Pulsed field, PCR ribotyping and multiplex PCR analysis of Yersinia enterocolitica strains isolated from meat food in San Luis Argentina

The characterization of phenotypic and genotypic virulence markers of Yersinia enterocolitica strains belonging to biotypes (B) 1A, 2 and 3, mostly isolated from food in San Luis, Argentina, and the assessment of their genotypic diversity using PFGE and PCR ribotyping, were performed in our laboratory for the first time. Thirty five Y. enterocolitica strains, two reference strains and 33 strains isolated in our laboratory were studied. The presence of virF, ail, ystA, and myfA genes was investigated by multiplex PCR. The pathogenic potential of B1A strains, the most predominant biotype of Y. enterocolitica strains isolated from meat in our region, was investigated by simple PCR. Four B1A strains were positive for ystB gene. Four Y. enterocolitica 2/O:9 (bio/serotype) and two 3/O:5 strains isolated in our laboratory showed virulence-related results in the phenotypic tests and multiplex PCR. A good correlation between the expression of virulence markers and their corresponding genotypes was observed for most strains. Sixteen genomic types (GT) and 9 different intergenic spacer region (SR) groups were generated by PFGE and PCR ribotyping, respectively. In both cases the Y. enterocolitica 2/O:9 strains were separately clustered from 1A and 3/O:5 strains. Meat foods might be vehicles of transmission of pathogenic Y. enterocolitica strains in our region.     

Detection and prevalence of pathogenic Yersinia enterocolitica in refrigerated and frozen dairy products by duplex PCR and dot hybridization targeting the virF and ail genes

Pathogenic Yersinia enterocolitica is involved in yersiniosis through expression of chromosome-borne or plasmid-borne virulence factors. Yersinia enterocolitica is a cold-tolerant pathogen frequently isolated from refrigerated or frozen foods. However, little attention has been focused on the prevalence of pathogenic Y. enterocolitica in refrigerated or frozen dairy samples in China. In this study, we developed a new duplex PCR targeting the plasmid-borne virF gene and chromosome-borne ail gene for detection of pathogenic Y. enterocolitica isolates. We established a detection limit for the duplex PCR of 6.5 × 10² cfu/mL in artificially contaminated dairy samples. In addition, the duplex PCR could detect directly 4.5 to 5.7 cfu of Y. enterocolitica in 5 mL of brain heart infusion broth after 6 h of enrichment at 28°C. A newly developed dot hybridization assay further confirmed specificity of the duplex PCR for detection of virulent Y. enterocolitica. Furthermore, 13 Y. enterocolitica and 5 pathogenic strains, from 88 commercial frozen or refrigerated dairy products, were detected successfully by the China National Standard method (GB/T4789.8-2008) and the duplex PCR, respectively. Finally, biotypes and serotypes of pathogenic Y. enterocolitica strains were further characterized. The duplex PCR developed here is reliable for large-scale screening, routine monitoring, and risk assessment of pathogenic Y. enterocolitica in refrigerated or frozen dairy products.     

Genotypic and phenotypic virulence characteristics and antimicrobial resistance of Yersinia spp. isolated from meat and milk products

A total of 300 food samples including 180 milk and 120 meat products have been examined for the presence of Yersinia spp. using the ISO 10273 and the cold enrichment method. The overall prevalence of Yersinia spp. was 84 (28%). Yersinia enterocolitica was isolated from 18 (6%) of the 300 samples. The other Yersinia species were detected in the samples Yersinia rohdei 15 (5%), Yersinia intermedia 14 (4.7%), Yersinia pseudotuberculosis 12 (4%), Yersinia ruckeri 12 (4%), Yersinia mollaretii 5 (1.7%), Yersinia bercovieri 4 (1.3%), and atypical Yersinia spp. 4 (1.3%). The conventionally identified Y. enterocolitica strains were also confirmed by the 16S rRNA gene sequencing. All Y. enterocolitica strains biotyped as 1A had negative results in the phenotypic virulence tests. The 84 Yersinia strains were also examined genotypically for the presence of virulence genes. None of the Y. enterocolitica and other Yersinia strains contained the ail, ystA, yadA, and virF except only 1 Y. intermedia and 2 Y. enterocolitica strains that were found to be positive for ystB. Antimicrobial resistance of 84 Yersinia to 16 antimicrobial agents was determined by the disk diffusion method. All strains were sensitive to tobramycine and imipenem while resistant to clindamycin. Although 84.5% of the strains were resistant to at least 3 or more antimicrobial agents, 64.3% of them were resistant to 4 or more antimicrobial agents.     

Pathogenic Yersinia enterocolitica 2/O:9 and Yersinia pseudotuberculosis 1/O:1 strains isolated from human and non-human sources in the Plateau State of Nigeria

Foodborne yersiniosis, caused by enteropathogenic Yersinia, especially Yersinia enterocolitica, is an important cause of diarrhea in developed countries, especially in temperate zones. Since studies concerning the presence of enteropathogenic Yersinia in humans and foods are rare in developing countries and tropical areas, human and non-human samples were studied in Plateau state of Nigeria to obtain information on the epidemiology of Y. enterocolitica and Yersinia pseudotuberculosis. Surprisingly, ail-positive Y. enterocolitica and inv-positive Y. pseudotuberculosis were isolated in Plateau state of Nigeria from several samples of human and non-human origin. Bioserotype 1/O:1 was the only Y. pseudotuberculosis type found. Y. enterocolitica belonging to bioserotype 2/O:9 was the dominating type found in most samples. Bioserotype 4/O:3 was isolated only from one pig and one sheep. Using PFGE, 5 genotypes were obtained among 45 Y. enterocolitica 2/O:9 strains with NotI, ApaI and XhoI enzymes and 3 among 20 Y. pseudotuberculosis 1/O:1 strains with NotI and SpeI enzymes. All human Y. pseudotuberculosis 1/O:1 strains were indistinguishable from pig, sheep or food strains. The dominating genotype of Y. enterocolitica 2/O:9 strains among humans was also found among strains isolated from pig, fermented cow milk and traditional intestine pepper soap samples.     

Detection of Pathogenic Yersinia enterocolitica in Meat using Real-Time PCR

Yersinia enterocolitica is an important zoonosis, which can cause disease in humans and animals. The culture methods available for detection of Y. enterocolitica in food samples are time-consuming and seldom successful. Using DNA-based methods, like PCR, this pathogen can be detected more rapidly and with greater sensitivity. The aim of this study was to establish a rapid and accurate real-time PCR method to detect pathogenic Y. enterocolitica in pork. The chromosomal ail−gene, which is only present in pathogenic Y. enterocolitica strains, was used as DNA target for the 5' nuclease PCR assay. The probe was labelled at the 5' end with the fluorescent reporter dye (FAM) and at the 3' end with the quencher dye (TAMRA). A 2-step protocol with 45 cycles was used in a multicolour real-time PCR detection system. A Ct value over 40 indicated a negative result. The DNA extraction procedure for the natural samples was rapid and simply. This qualitative real-time PCR method was shown to be specific and sensitive. Detection rate of ail-positive Y. enterocolitica in 200 pig tonsils was 88 % and 35 % with PCR and culture methods, respectively. When 100 raw pork samples were studied, 7 were positive with PCR and all were culture negative.     

Evaluation of the ISO 10273:2003 method for the isolation of human pathogenic Yersinia enterocolitica from pig carcasses and minced meat

Pig carcass swabs (n = 254) and minced meat samples (n = 82) were examined for pathogenic Yersinia enterocolitica using different routinely used enrichment protocols. All samples were obtained in the context of the official Yersinia monitoring program in Belgium. In total, 28 carcasses (11.0%) were contaminated with Y. enterocolitica bioserotype 4/O:3 and one (0.4%) with bioserotype 2/O:9. Four minced meat samples (4.9%) tested positive for Y. enterocolitica bioserotype 4/O:3. Using the ISO 10273:2003 method, eight out of the 29 Yersinia-positive carcasses (27.6%) and none of the contaminated minced meat samples (0.0%) were detected. Reducing the enrichment time in PSB from 5 to 2 days increased the number of positive samples. Overall, enrichment in PSB at 25 °C recovered more positive carcasses and minced meat samples than selective enrichment and cold enrichment. As the exclusive use of the ISO 10273:2003 method results in a strong underestimation of Y. enterocolitica positive carcasses and minced meats, efforts are needed to optimize the current version of the ISO method. In addition, isolation of pathogenic Y. enterocolitica requires experience and the use of a stereomicroscope to avoid false negative results.     

Yersinia enterocoliticaand related species isolated from animals slaughtered for human consumption

Yersinia enterocoliticaand related species recovery frequency was studied in recently slaughtered animals in San Luis, Argentina. A total of 493 samples from bovine rectal contents (100), young goat (100), chicken (50) and pig (103) caecal contents, and chicken skins (140) were analyzed. Forty-sixY. enterocoliticastrains, threeYersinia intermediastrains and oneYersinia frederikseniistrain were obtained from 7.5% of the samples. A highestY. enterocoliticarecovery was from the caecal contents from pigs (26.2% positive samples). Most of the strains belonged to biovar 1, serovar 0:5. In porcines, the isolated biovars were different from those recovered from these animals in Europe. The pathogenicity was assessed by virulence tests with positive autoagglutination and calcium dependence at 37°C for threeY. enterocoliticastrains from porcines and negative results for the remaining ones.     

Prevalence of Yersinia enterocolitica in milk and dairy products and the effects of storage temperatures on survival and virulence gene expression

Yersinia enterocolitica was isolated from raw milk and dairy products from 10% of examined samples. The highest isolation rate was 22%, from raw milk, followed by 12%, 4% and 2% from fermented milk (Rayeb), pasteurised milk and ripened salted cheese, respectively. The virulence-associated genes ail and yst were detected in 30% and 10% of the isolates, respectively, while these genes were present simultaneously in 10% of the isolates. All the isolates showed susceptibility to gentamicin, ciprofloxacin and chloramphenicol, while only two of the isolates exhibited multidrug resistance. Storage of inoculated pasteurised milk at refrigeration (4 °C), freezing (−20 °C) and room (25 °C) temperatures revealed significant differences in Y. enterocolitica counts and relative expression of the two virulence genes. The isolation of potentially pathogenic Y. enterocolitica isolates from retail dairy products indicates risk to consumers; screening of prevalence, pathogenicity potential and antibiotic resistance is essential to implement control measures.     

Detection of Pathogenic Yersinia enterocolitica in Meat using Real-Time PCR

Yersinia enterocolitica is an important zoonosis, which can cause disease in humans and animals. The culture methods available for detection of Y. enterocolitica in food samples are time-consuming and seldom successful. Using DNA-based methods, like PCR, this pathogen can be detected more rapidly and with greater sensitivity. The aim of this study was to establish a rapid and accurate real-time PCR method to detect pathogenic Y. enterocolitica in pork. The chromosomal ail−gene, which is only present in pathogenic Y. enterocolitica strains, was used as DNA target for the 5' nuclease PCR assay. The probe was labelled at the 5' end with the fluorescent reporter dye (FAM) and at the 3' end with the quencher dye (TAMRA). A 2-step protocol with 45 cycles was used in a multicolour real-time PCR detection system. A Ct value over 40 indicated a negative result. The DNA extraction procedure for the natural samples was rapid and simply. This qualitative real-time PCR method was shown to be specific and sensitive. Detection rate of ail-positive Y. enterocolitica in 200 pig tonsils was 88 % and 35 % with PCR and culture methods, respectively. When 100 raw pork samples were studied, 7 were positive with PCR and all were culture negative. Received: January 30, 2007; Received in revised form: February 27, 2007; Accepted: February 28, 2007.     

A real-time PCR for Detection of Pathogenic Yersinia enterocolitica in food combined with an Universal Internal Amplification Control System

A real-time PCR system with an internal amplification control was developed for detection of pathogenic Yersinia) enterocolitica in food samples. The chromosomally encoded ail gene was chosen as PCR target. Sequences of plasmid pUC19 served as target for the internal amplification control. The method was validated in combination with sample enrichment in PSB and TSB broth using different food matrices spiked with Y. enterocolitica and naturally contaminated slaughterhouse samples. The results of the real-time PCR with internal control were verified by the cultural method according to EN ISO 10273:2003. The sensitivity of the real-time PCR with internal control is about 5 genome copies per reaction. Artificial contamination of food samples resulted in a detection level of 5 cfu per 25 g Y. enterocolitica in food samples. 100% of porcine tonsils and about 22% meat from pig heads were contaminated. The screening of samples by PCR prior to cultural analysis allows focusing on positive samples in routine analysis. This could result in a higher detection rate by cultural analysis.     

Occurrence of Bacillus cereus and Yersinia enterocolitica in South African retail meats

Ground beef, chicken broilers and processed meats (Vienna sausages, ham, salami) obtained from supermarkets in the Pretoria area (South Africa) were tested for the presence of Bacillus cereus and Yersinia enterocolitica. B. cereus counts were carried out on Bacillus cereus -selective agar and confirmed by the rapid confirmation procedure. Y. enerocolitica was isolated on Y. enterocolitica -selective agar. Total aerobic counts were performed on all the samples. Out of 51 samples, B. cereus was recovered from one broiler, three salami and five Vienna sausage samples, but not from gound beef or ham samples. The levels of detection ranged from log₁₀1·0 to log₁₀3·1 g⁻¹sample. Y. enterocolitica occurred in eight ground beef, eight broiler, one Vienna sausage and two ham samples. Our results emphasize the potential hazard of the presence of B. cereus and Y. enterolitica in meat and poultry products in South Africa. This necessitates the implementation of methods to prevent the entry or proliferation of these pathogens in meat products.     

Yersinia enterocolitica in slaughter pig tonsils: Enumeration and detection by enrichment versus direct plating culture

Tonsil samples from 139 slaughter pigs were examined for the presence of pathogenic Yersinia enterocolitica by enrichment procedures based on the standard method ISO 10273:2003. In addition, samples were tested by direct plating method to evaluate its efficiency compared to the enrichment culture methods and to quantify the level of contamination in porcine tonsils. In total, 52 samples (37.4%) were positive for pathogenic Y. enterocolitica, all belonging to bioserotype 4/O:3. Fifty out of the 52 positive samples (96.2%) were detected by direct plating. Enumeration showed an average concentration of 4.5 log₁₀ CFU g⁻¹ and 4.4 log₁₀ CFU g⁻¹ tonsil on Salmonella-Shigella-desoxycholate-calcium chloride (SSDC) and cefsulodin-irgasan-novobiocin (CIN) agar plates, respectively. The enrichment procedures recommended by the ISO 10273:2003 method were not optimal for the isolation of pathogenic Y. enterocolitica from pig tonsils: two days enrichment in irgasan-ticarcillin-potassium chlorate (ITC) broth resulted in an isolation rate of 84.6%, while 5 days enrichment in peptone-sorbitol-bile (PSB) broth recovered only 59.6% of positive samples. Reducing the enrichment time in PSB from 5 to 2 days resulted in a significantly higher recovery rate (94.2%) and might serve as an appropriate enrichment protocol for the isolation of pathogenic Y. enterocolitica from pig tonsils. Compared to enrichment culture methods, results based on direct plating can be obtained in a shorter time course and provide quantitative data that might be needed for further risk assessment studies.     

Growth of Salmonella enteritidis and Salmonella typhimurium in the presence of quorum sensing signalling compounds produced by spoilage and pathogenic bacteria

The effect of acylated homoserine lactones (AHLs) and autoinducer-2 (AI-2) signalling compounds present in the cell-free culture supernatants (CFS), of Pseudomonas aeruginosa, Yersinia enterocolitica-like GTE 112, Serratia proteamaculans 00612, Y. enterocolitica CITY650 and Y. enterocolitica CITY844, on the growth of two Salmonella Enteritidis and two S. Typhimurium strains was assessed though monitoring of changes in conductance of the medium. Detection times (T_det), area and slope of conductance curves were recorded. Except for P. aeruginosa 108928, which was not found to produce AI-2, all other strains produced both AHLs and AI-2. Thereafter, aliquots (20% in the final volume) of these CFS were transferred into NZ Amine broth inoculated with ca. 10³ CFU/ml of stationary phase cultures of each Salmonella strain. While the CFS of P. aeruginosa induced a shorter detection time, i.e. acceleration of the metabolic activity, the CFS of the other microorganisms increased the detection time of Salmonella strains compared to control samples (i.e. without CFS). Results indicate that the growth of Salmonella may be affected by the presence of Quorum sensing (QS) signalling compounds and/or other novel signals existing in CFS, produced by other bacterial species and confirm the complexity of bacterial communication.     

Control of Yersinia enterocolitica in raw pork and pork products by γ-irradiation

γ-Radiation response of Y. enterocolitica 5692 and 152 was studied at 0°C and at −40°C in phosphate buffer (pH 7.00) as well as in 10% raw meat/salami homogenate. The strains investigated did not differ in their response and were found to be sensitive to γ-radiation but exhibited a tailing phenomenon in the survival curve. The D₁₀ in homogenate was 0.25 kGy at 0°C. This response was not affected at −40°C. Storage studies of packs, inoculated artificially with heavy inoculum of Y. enterocolitica (10⁶ cfu/g) showed that while samples of salami and cooked ham could be decontaminated at doses of 4 and 3 kGy respectively; cells could not be eliminated from raw pork meat even at the higher dose of 6 kGy. The role of different treatments given prior to irradiation for revival of Y. enterocolitica after irradiation storage was also studied. The dose of 1 kGy at −40°C was efficient in eradicating low numbers (<10³) of naturally occuring Y. enterocolitica from raw pork meat without any revival during storage at refrigeration temperature.     

A real-time PCR for Detection of Pathogenic Yersinia enterocolitica in food combined with an Universal Internal Amplification Control System

Summary:  A real-time PCR system with an internal amplification control was developed for detection of pathogenic Yersinia) enterocolitica in food samples. The chromosomally encoded ail gene was chosen as PCR target. Sequences of plasmid pUC19 served as target for the internal amplification control. The method was validated in combination with sample enrichment in PSB and TSB broth using different food matrices spiked with Y. enterocolitica and naturally contaminated slaughterhouse samples. The results of the real-time PCR with internal control were verified by the cultural method according to EN ISO 10273:2003. The sensitivity of the real-time PCR with internal control is about 5 genome copies per reaction. Artificial contamination of food samples resulted in a detection level of 5 cfu per 25 g Y. enterocolitica in food samples. 100% of porcine tonsils and about 22% meat from pig heads were contaminated. The screening of samples by PCR prior to cultural analysis allows focusing on positive samples in routine analysis. This could result in a higher detection rate by cultural analysis.

---

Zusammenfassung:  Für den Nachweis von pathogenen Yersinia enterocolitica wurde ein real-time PCR System mit interner Amplifikationskontrolle entwickelt. Das Nachweissystem für pathogene Y. enterocolitica basiert auf dem chromosomal kodierten ail-Gen. Als Zielsequenz der internen Amplifikationskontrolle dient eine Sequenz aus dem Plasmid pUC19. Zur Validierung der Methode wurden sowohl natürlich kontaminierte Proben aus einem Schlachthof als auch künstlich kontaminierte Proben verschiedener Lebensmittelmatrices verwendet. Die Anreicherung der Proben vor der molekularbiologischen Untersuchung erfolgte parallel in Tryptikase Soja-Bouillon (TSB) und in Pepton-Sorbit-Gallensalz-Bouillon (PSB). Die Ergebnisse der molekularbiologischen Untersuchungen wurden anschlie?end kulturell in Anlehnung an das Standardverfahren nach EN ISO 10273:2003 verifiziert. Die real-time PCR mit interner Amplifikationskontrolle weist eine Sensitivit?t von 5 Genomkopien pro Reaktionsansatz auf. Die Nachweisgrenze des Verfahrens, bestimmt anhand künstlich kontaminierter Proben, betr?gt etwa 5 KbE Y. enterocolitica pro 25 g Lebensmittel. Von den natürlich kontaminierten Proben aus einem Schlachthof waren die Tonsillen vom Schwein zu 100% mit pathogenen Y. enterocolitica kontaminiert, Schweinefleischabschnitte aus dem Kopfbereich wiesen einen Kontaminationsgrad von 22% auf. Ein Screening von Proben durch PCR erlaubt in der Routineanalytik die Fokussierung der kulturellen Analyse auf positive Proben. Dies k?nnte zu einer h?heren Nachweisrate durch das kulturelle Verfahren führen.

---

Received: March 5. 2008; accepted: March 14. 2008     

Loss of crystal violet binding activity in stationary phase Yersinia enterocolitica following gamma irradiation

Ionizing radiation can eliminate virulent Yersinia enterocolitica from meat. It is possible, however unlikely, that a small number of Y. enterocolitica could survive the pasteurization process. The virulence ofY. enterocolitica is dependent upon the presence of a 70 kb plasmid. The effect of low-dose ionizing radiation on the plasmid-associated virulence trait of crystal violet binding was investigated. Y. enterocolitica strains which carried the virulence plasmid were suspended in Butterfield's Phosphate Buffer or raw ground pork and irradiated to a dose of 1·0 or 1·25 kGy, respectively. Loss of crystal violet binding increased 10-fold following exposure to ionizing radiation, regardless of the suspending medium. Polymerase chain reaction analysis of Y. enterocolitica isolates that did not bind crystal violet following irradiation indicated that loss of the virulence plasmid, as opposed to mutation of a single plasmid-encoded gene, was the major mechanism for elimination of the crystal violet binding trait.     

Enumeration of Yersinia enterocolitica O:3 from the porcine oral cavity,and its occurrence on cut surfaces of pig carcasses and the environment in a slaughterhouse

The number of Yersinia enterocolitica serogroup O:3/biovar 4 recovered from five porcine tongues, ranged from 0–186/cm², the corresponding range for the tonsils was, 0–1720/cm². This bacterium was isolated from the oral cavity of 25 (83.3%) of 30 freshly eviscerated slaughter pigs. The isolation rates obtained from the different cut surfaces were: cranial incision, 46.7%; abdominal incision, 43.3%; and circumanal incision, 26.7%. Seven pigs harboured the bacteria in the oral cavity only, and one pig yielded Y. enterocolitica serogroup O:3/biovar 4 only from the abdominal incision. In eighteen pigs the bacteria was obtained from the oral cavity and at least one of the cut surfaces, four of these pigs yielding the bacteria from all sites sampled. Y. enterocolitica O:3/biovar 4 was isolated from samples taken from the following sites in the slaughterhouse: the floor in the bleeding area, the floor in the eviscerating area, the viscera table, and the floor next in line after the meat inspection area. The organism was not found on the floor in the meat inspection area. It is concluded that modern slaughtering technology and the routines followed by meat inspection personnel have probably contributed to the relatively high rates of this bacteria on carcass cut surfaces detected in this study. The results of this investigation thus emphasize the need to consider the introduction of changes in slaughtering technology and meat inspection practices presently being employed in Norwegian slaughterhouses.     

Low dose gamma irradiation of raw meat. I. Bacteriological and sensory quality effects in artificially contaminated samples

Filet américain, consisting of raw, ground meat (beef) mixed with mayonnaise sauce, and the corresponding raw beef (without sauce) were experimentally contaminated with different strains of Salmonella. Yersinia enterocolitica and Campylobacter jejuni and irradiated with doses up to 1.5 kGy.Radiation resistance (D value) was determined immediately after irradiation. For the evaluation of the irradiaion effect on Y. enterocolitica in relation to storage, a number of samples were stored for a few days at 3°C and the microbiologically examined.D values of S. typhimurium, S. anatum, S. panama and S. stanley (strains isolated from retail filet américain) irradiated in filet américain were found to be 0.37, 0.45, 0.41 and 0.61 kGy and in raw beef 0.55, 0.67, 0.66 and 0.78 kGy, respectively. D values of Y. enterocolitica serotypes 0:3, 0:5, 27 and 0:9 irradiated in filet américain were 0.043, 0.065, and 0.080 kGy and in raw beef, 0.10, 0.16 and 0.21 kGy, respectively. D values of C. jejuni (three strains) irradiated in filet américain were 0.11, 0.08 and 0.09 and in beef 0.15, 0.14 and 0.16 kGy, respectively.It is concluded that doses as low as 1 kGy are effective in reducing Salmonella by approximately 1.6–2.7 log cycles in filet américan and 1.3–1.8 log cycles in ground meat. Numbers of Y. enterocolitica and C. jejuni are always reduced by more than 4 log cycles with this dose.On sensory evaluation 38% of samples of filet américain irradiated directly with 1 kGy were not acceptable for a taste panel; preference was for the unirradiated sample in 82% of the cases. However, when beef was irradiated with the same dose prior to the addition of mayonnaise sauce no significant taste differences could be observed between nonirradiated and irradiated samples.     

Genotypic and phenotypic characteristics of Yersinia spp. isolates from food and man

Yersinia strains collected by the Unit of Enteric Pathogens of the Istituto Superiore di Sanità were analysed for bioserogroup, virulence-related characteristics, DNA plasmid profile and biotype. The isolates were from clinical and food samples, 26 were Y. enterocolitica, four Y. intermedia and four Y. frederiksenii. Seventeen were isolated in Northern Italy, 11 in Southern Italy, five in Central Italy and one from a sample of mussels imported from Greece. Virulence-related characteristics were expressed only by four Y. enterocolitica, bioserogroup 4:O3 associated with the presence of a plasmid weighing 76·8 Kbp. The DNA digested by Spe I and Xba I and analysed by PFGE generated restriction patterns of 12–23 bands, ranging between 400 and 20 Kbp. The similarities of the DNA fingerprintings ofY. enterocolitica strains generated using Spe I and Xba I and calculated by Dice's index were not closely related to bioserogroup.     

Technological characterization of Geotrichum candidum strains isolated from a traditional Spanish goats' milk cheese

Forty-one strains of Geotrichum candidum isolated from Armada cheese, Sobado variety, were screened for their enzymatic activities, including proteolytic and lipolytic activities and aminopeptidase activity. The highest extracellular proteolytic activity was detected for 8 strains with values ranging between 2.086 and 4.685 mM Gly L⁻¹ of milk. Extracellular lipolytic activity was high for all but one of the G. candidum strains, with values ranging between 67 and 131 μmol oleic mL⁻¹. Cell-bound lipase activity showed values which were considerably lower than those for extracellular activity, ranging between 32.50 and 42.50 μmol oleic mL⁻¹ and falling below 20 μmol oleic mL⁻¹ in 28 strains. Aminopeptidase activity was not very high in the cell-free extract (CFE) of any of the strains, showing the highest values toward the substrate Lys-p-nitroanilide, with levels of activity ranging between 6.12 and 10.05 UE mg⁻¹ protein in 8 strains. In accordance with the results obtained, four G. candidum strains were selected as co-cultures in order to study their role in the ripening of a semi-hard goat’s milk cheese.