Enumeration of clostridia in goat milk using an optimized membrane filtration technique.

A membrane filtration technique developed for counting butyric acid bacteria in cow milk was further developed for analysis of goat milk. Reduction of the sample volume, prolongation of incubation time after addition of proteolytic enzyme and detergent, and a novel step of ultrasonic treatment during incubation allowed filtration of goat milk even in the case of somatic cell counts (SCC) exceeding 10(6)/mL. However, filterability was impaired in milk from goats in late lactation. In total, spore counts were assessed in 329 farm bulk goat milk samples. Membrane filtration technique counts were lower than numbers revealed by the classic most probable number technique. Thus, method-specific thresholds for milk to evaluate the risk of late blowing have to be set. As expected, the spore counts of milk samples from suppliers not feeding silage were significantly lower than the spore counts of milk samples from suppliers using silage feeds. Not only were counts different, the clostridial spore population also varied significantly. By using 16S rRNA gene PCR and gene sequencing, 342 strains from 15 clostridial species were identified. The most common Clostridium species were Clostridium tyrobutyricum (40.4%), Clostridium sporogenes (38.3%), Clostridium bifermentans (7.6%), and Clostridium perfringens (5.3%). The 2 most frequently occurring species C. tyrobutyricum and C. sporogenes accounted for 84.7% of the isolates derived from samples of suppliers feeding silage (n=288). In contrast, in samples from suppliers without silage feeding (n=55), these species were detected in only 45.5% of the isolates.

Stress survival islet 1 (SSI-1) survey in Listeria monocytogenes reveals an insert common to listeria innocua in sequence type 121 L. monocytogenes strains.

Listeria monocytogenes strains (n = 117) were screened for the presence of stress survival islet 1 (SSI-1). SSI-1(+) strains (32.5%) belonged mainly to serotypes 1/2c, 3b, and 3c. All sequence type 121 (ST-121) strains included (n = 7) possessed homologues to Listeria innocua genes lin0464 and lin0465 instead of SSI-1.

Novel real-time PCR assay for simultaneous detection and differentiation of Clostridium chauvoei and Clostridium septicum in clostridial myonecrosis.

A real-time PCR assay based on the 16S rRNA gene sequence was designed for differentiation of blackleg-causing Clostridium chauvoei and Clostridium septicum, a phylogenetically closely related bacterium responsible for malignant edema. In order to exclude false-negative results, an internal amplification control was included in the assay. A set of three probes, one specific for C. chauvoei, one specific for C. septicum, and one specific for both species, permitted unequivocal detection of C. chauvoei in tests of 32 Clostridium sp. strains and 10 non-Clostridium strains. The assay proved to be sensitive, detecting one genome of C. chauvoei or C. septicum per PCR and 1.79 x 10(3) C. chauvoei cells/g artificially contaminated muscle tissue. In tests of 11 clinical specimens, the real-time PCR assay yielded the same results as an established conventional PCR method.

Evaluation of paramagnetic beads coated with recombinant Listeria phage endolysin-derived cell-wall-binding domain proteins for separation of Listeria monocytogenes from raw milk in combination with culture-based and real-time polymerase chain reaction-based quantification.

The aim of this study was to evaluate a fast and simple bead-based method using paramagnetic beads covered with recombinant Listeria phage endolysin-derived cell-wall-binding domain proteins specific for Listeria spp. for separation of the foodborne pathogen Listeria monocytogenes from artificially contaminated raw milk. The method was combined with subsequent detection and quantification by the traditional plate-count technique and real-time polymerase chain reaction (PCR). To account for differences in cell properties, recovery rates and the detection limit were determined using five different L. monocytogenes strains for preparation of a 10-fold dilution series in raw milk, spanning an 8-log scale. Two independent test series were performed for each strain, yielding mean recovery rates of 46.6% to 122.8% for detection with the plate-count method, and 64.7% to 95.1% for detection by real-time PCR. A high correlation was found between the number of L. monocytogenes added to the samples and the number of colony forming units recovered by plate count (0.980), as well as the number of bacterial cell equivalents obtained by real-time quantitative PCR (0.987). The detection limit of the combined cell-wall-binding domain proteins/real-time PCR approach ranged from 10(2) to 10(3) colony forming units per milliliter, which is close to the theoretical detection limit of the method.

Stress- and growth rate-related differences between plate count and real-time PCR data during growth of Listeria monocytogenes.

To assess the overestimation of bacterial cell counts in real-time PCR in relation to stress and growth phase, four different strains of L. monocytogenes were exposed to combinations of osmotic stress (0.5 to 8% [vol/vol] NaCl) and acid stress (pH 5 to 7) in a culture model at a growth temperature of 10 degrees C or were grown under optimal conditions. Growth curves obtained from real-time PCR, optical density, and viable count data were compared. As expected, optical density data revealed entirely different growth curves. Good to moderate growth conditions yielded good correlation of real-time PCR data and plate count data (r(2) = 0.96 and 0.99) with similar cell counts. When growth conditions became worse, the numbers of CFU decreased during the stationary phase, whereas real-time-PCR-derived bacterial cell equivalents differed in this regard; the correlation worsened (r(2) = 0.84). However, fitted growth curves revealed that maximum growth rates calculated from real-time PCR data were not significantly different from those derived from plate count data. The overestimation of bacterial cell counts by real-time PCR observed in the stationary phase under higher-stress conditions might be explained by the accumulation of viable but nonculturable bacteria or dead bacteria and extracellular DNA. Considering these results, real-time PCR data collected from naturally contaminated samples should be viewed with caution.

Studying the effect of single mismatches in primer and probe binding regions on amplification curves and quantification in real-time PCR.

Comparison of a broad range of characteristics of real-time PCR amplification curves yielded only slight alterations for low numbers of mismatches in the primer binding regions, resulting in a quantification error up to 63.12%. The effects were more pronounced for mismatches in the probe binding region and resulted in a quantification error up to 33%.

Temporal and spatial distribution of Cronobacter isolates in a milk powder processing plant determined by pulsed-field gel electrophoresis.

A milk powder processing line was sampled for the presence of Enterobacteriaceae and the opportunistic neonatal pathogen Cronobacter at six different sampling sites during an 11-month period. The highest number of Enterobacteriaceae-positive samples was recovered from the raw milk concentrate before pasteurization (78.2%) and from nonproduct samples of the processing line (86.5%), which included swabs from the drying tower and screw conveyers, swabs from the explosion chamber, waste water after the automated cleaning-in-place procedure, air filter cut-outs, and floor samples underneath the outlet of the packaging machine. The prepackaged and packaged final product was contaminated at a rate of 6.6% and 7.1%, respectively. The prevalence of Cronobacter in the prefinal product and the prepackaged and packaged final product was 14.3%, 3.8%, and 2.1%, respectively. Pulsed-field gel electrophoresis (PFGE) analysis of 133 Cronobacter isolates yielded 40 different PFGE profiles. Long-term persistence in the processing line of some of these PFGE profiles was observed. Comparison of the PFGE profiles recovered at different sampling sites revealed the supply air as a potential source for extrinsic Cronobacter contamination. In addition, recovery of the same PFGE profiles before and after CIP events followed by heat treatment indicated the inefficiency of these hygiene measures to completely eliminate Cronobacter from all areas of the processing line. This information provides an essential basis for developing control and prevention strategies concerning this opportunistic pathogen.

Monitoring hygiene on- and at-line is critical for controlling Listeria monocytogenes during produce processing.

The prevalence of Listeria monocytogenes in different types of produce and on processing plant environments was investigated over a 4-year period in a large produce processing plant in Poland. Prevalence of L. monocytogenes was 46% in frozen vegetables and 41.3% in swab samples taken from the plant environment. Survival studies using artificial inocula demonstrated that the number of Listeria in frozen produce stored for 100 days did not significantly decrease in relation to the initial contamination level. A subset of 129 L. monocytogenes isolates originating from produce and the plant environment were typed by pulsed-field gel electrophoresis. Seventy-six of these isolates were retyped by ribo- and serotyping. Thirteen pulsotypes and 18 ribotypes were distinguished. Persistent Listeria isolates were found even when cleansing and sanitization was applied on a daily basis. Nine (69.2%) of 13 pulsotypes were recovered during a period of more than 2 years. L. monocytogenes of the same pulsotype was isolated from broccoli sampled directly before and after blanching, thus suggesting that blanching at 92 to 95 degrees C for 4 to 8 min did not result in a Listeria-free product, most likely due to massive recontamination. This finding is of importance since blanching is the only critical control point in produce processing. Cross-contamination between the two lines was demonstrated through isolating L. monocytogenes strains indistinguishable by pulsed-field gel electrophoresis from contaminated gloves and floor surfaces.

Insufficient differentiation of live and dead Campylobacter jejuni and Listeria monocytogenes cells by ethidium monoazide (EMA) compromises EMA/real-time PCR.

Recently, ethidium monoazide (EMA) has been proposed as a means of reducing the real-time PCR signal originating from free DNA and dead bacterial cells by selectively entering damaged cells and blocking the DNA for PCR amplification via photoactivation. The present study investigated the effect of EMA on viable and dead bacterial cells using real-time PCR, plate count method and microscopy. The foodborne pathogens Campylobacter jejuni and Listeria monocytogenes were used as a Gram-negative and a Gram-positive model organism, respectively. EMA/real-time PCR analysis of heat-treated cultures of C. jejuni and L. monocytogenes containing 2.6x10(5) and 4x10(5) viable and 3x10(6) and 2x10(6) dead cells/ml, respectively, yielded 2x10(3) and 5.2x10(4) bacterial cell equivalents/ml after EMA treatment, thus underestimating the viable cell count in the samples. Similar results were obtained when analyzing late exponential phase cultures of C. jejuni and L. monocytogenes. Inhibition of growth by EMA was observed. It depended on the concentration of the bacterial cells present in the sample and the EMA concentration used (100-1 microg/ml). An EMA concentration at which dead cells would stain brightly and viable cells would not stain at all or would be very pale was not identified, as revealed by comparison with the results of a commercial live/dead stain. The results suggest that EMA influences not only dead but also viable cells of C. jejuni and L. monocytogenes. Thus EMA/real-time PCR is a poor indicator of cell viability.

Development of matrix lysis for concentration of gram positive bacteria from food and blood.

The development of a fast, reliable and inexpensive protocol for the concentration of bacteria from food by the removal of fat, carbohydrates and proteins that is compatible with downstream alternative DNA-based quantification methods is described. The protocol was used for dairy products, cooked and smoked fish and meat, carbohydrate-rich cooked products, ready-to-eat sauces, egg and blood. Lysis resulted in pellets of reasonable size for further processing. Starch, plant materials, fungi, tissues such as sinew, and chalaza could not be dissolved. Using L. monocytogenes, S. aureus and B. cereus as model organisms, microscopic analysis of the remaining bacterial pellets revealed that the recovered bacteria remained physically intact, albeit that the viability of the cells was compromised. Using real-time PCR, 7.3 CFU of L. monocytogenes were detected in artificially contaminated ultra-high temperature treated (UHT) milk and raw milk.

Ethidium monoazide and propidium monoazide for elimination of unspecific DNA background in quantitative universal real-time PCR.

Unspecific background DNA in quantitative universal real-time PCR utilizing a hydrolysis probe was completely suppressed by the addition of EMA or PMA to the PCR mix via cross-linking of the dyes to DNA during 650 W visible light exposure. The proposed procedure had no effect on the sensitivity of the real-time PCR reaction.

Inhibitor-free DNA for real-time PCR analysis of synovial fluid from horses, cattle and pigs.

The potential of five different commercial DNA isolation methods to remove real-time PCR inhibitors from the synovial fluid of horses, cattle and pigs was investigated. All kits with the exception of one included a silica column-based purification of the DNA. With the fifth kit, DNA purification is achieved by removing contaminating macromolecules by a desalting process. We used a recently developed method based on comparison of the real-time PCR signal of an artificial target incorporated into each PCR reaction in the presence of the isolated DNA from the sample, and in control samples containing water instead of isolated DNA. This was followed by statistical analysis of the data. Inhibition and subsequent reduction of the endpoint fluorescence in the real-time PCR reaction was encountered in many cases. Less frequently, the target copy number in the samples was underestimated. However, we found no experimental evidence of a negative influence of the reduced endpoint fluorescence signal on the detection limit of the real-time PCR assay. All kits tested were useful for analyzing pelleted synovial fluid from horses, cattle and pigs. When analyzing non-pelleted synovial fluid, three kits - two based on silica columns and one employing a desalting process - yielded inhibitor-free DNA for real-time PCR analysis.

Detection of Listeria monocytogenes in food using a combined enrichment/real-time PCR method targeting the prfA gene.

A combined enrichment/real-time PCR method for the detection of Listeria monocytogenes is presented. The method is based on a conventional PCR assay targeting the prfA gene, which has been validated and suggested as an international standard PCR method for identifying L. monocytogenes in food. This real-time PCR assay includes an internal amplification control. Inclusivity and exclusivity were 100% each when testing 100 L. monocytogenes isolates, 30 Listeria spp. isolates other than L. monocytogenes, and 29 non-Listeria isolates. The theoretical detection limit was one copy of the target gene per PCR reaction and the practical detection limit was about 5 copies per PCR. Using the combined enrichment/real-time PCR method, 7.5 CFU/25 ml of artificially contaminated raw milk, and 9, 1, and 1 CFU/15 g of artificially contaminated salmon, pâté, and green-veined cheese, respectively, were detected. When analyzing 76 naturally contaminated food samples of various types and comparing the results with the ISO 11290-1 standard method, the relative accuracy was 96%, the relative specificity 100%, and the relative sensitivity, 76.9%.

Real-time PCR for the detection of Salmonella spp. in food: An alternative approach to a conventional PCR system suggested by the FOOD-PCR project.

A real-time PCR assay using non-patented primers and a TaqMan probe for the detection and quantification of Salmonella spp. is presented. The assay is based on an internationally validated conventional PCR system, which was suggested as a standard method for the detection of Salmonella spp. in the FOOD-PCR project. The assay was sensitive and specific. Consistent detection of 9.5 genome equivalents per PCR reaction was achieved, whereas samples containing an average of 0.95 genome equivalents per reaction were inconsistently positive. The assay performed equally well as a commercially available real-time PCR assay and allowed sensitive detection of Salmonella spp. in artificially contaminated food. After enrichment for 16 h in buffered peptone water (BPW) or universal pre-enrichment broth (UPB) 2.5 CFU/25 g salmon and minced meat, and 5 CFU/25 g chicken meat and 25 ml raw milk were detected. Enrichment in BPW yielded higher numbers of CFU/ml than UPB for all matrices tested. However, the productivity of UPB was sufficient, as all samples were positive with both real-time PCR methods, including those containing less than 300 CFU/ml enrichment broth (enrichment of 5 CFU/25 ml raw milk in UPB).

Evidence of metabolically active but non-culturable Listeria monocytogenes in long-term growth at 10 °C.

Cultures of Listeria monocytogenes at low temperatures (10 °C) in a broth model revealed long-term survival at about 0.04% cell density in relation to the log phase. In contrast, direct viable counts and PMA real-time PCR data suggested that 50% and 1% of the population retain membrane integrity, respectively. To elucidate the observed difference, the metabolic activity of the bacterial population was investigated by multiparametric flow cytometry, including the assessment of membrane integrity, reductase activity, as well as forward and side scatter properties. These analyses were complemented by 16S rRNA real-time PCR. The majority of the cells retained their membrane integrity and reductase activity until day 29. On day 42, 48.00 ± 4.00% (L. monocytogenes strain 3251) and 68.67 ± 3.74% (L. monocytogenes strain 535) of the cells had intact membranes, whereas 57.23 ± 1.85% (strain 3251) and 74.97 ± 3.01% (strain 535) exhibited high reductase activity. On day 42, mean 16S rRNA copy numbers of 3.98 ± 1.37 (membrane integrity) and 3.86 ± 1.32 (reductase activity) remained per intact or active cell. Our data suggest the transition of L. monocytogenes into a state of metabolic dormancy during long-term culture at low temperature.

Quantification of Staphylococcus aureus in unpasteurised bovine and caprine milk by real-time PCR.

A sensitive and reproducible real-time PCR assay targeting the nuc gene of Staphylococcus aureus was applied for quantification of this microorganism in artificially and naturally contaminated raw milk samples. The S. aureus cell equivalents (SCEs) estimated by the real-time PCR method were two log scales higher than colony forming units (CFUs) estimated from a plate count method in artificially contaminated milk. The repeatability of the real-time PCR assay including the DNA isolation procedure was assessed by analysing the data derived from naturally contaminated samples. The relative standard deviation of the log-transformed data of four real-time PCR measurements including duplicate DNA isolations ranged between 11.3 and 1.0%. When analysing 80 bovine and 107 caprine naturally contaminated raw milk samples, the real-time PCR method yielded 19.3% more positive samples than the plate count method. With the exception of one sample, SCEs were always higher than CFUs. The difference between SCEs and CFUs was highly variable, and it was not possible to correlate real-time PCR-derived SCEs and CFUs. However, as each SCE detected by real-time PCR indicates a S. aureus cell, which is or has been present in the sample, this method offers the advantage of a retrospective analysis even of processed samples to aid food poisoning-related risk assessment.

Application of single-strand conformation polymorphism and denaturing gradient gel electrophoresis for fla sequence typing of Campylobacter jejuni.

Campylobacter jejuni is a frequent cause of bacterial gastroenteritis in humans all over the world. Several molecular typing methods are used to study the epidemiology of Campylobacter spp. infections. The aim of the present study was to investigate the application of single-strand conformation polymorphism (SSCP) and denaturing gradient gel electrophoresis (DGGE) analysis as rapid primary subtyping methods for C. jejuni. A variable fragment from the 3' end of the flaA to the 3' end of the intergenic region, separating the flaA and flaB genes, was subjected to SSCP and DGGE analysis. A total of 48 clinical C. jejuni isolates, 49 C. jejuni strains isolated from poultry, 2 strains isolated from ducks and 1 strain isolated from a pheasant were assigned to 24 distinct SSCP patterns. Sequence analysis of the respective DNA fragments revealed that every different fla sequence type could be distinguished by SSCP. DGGE proved to be equally discriminatory. Both methods can be applied as primary subtyping methods, because pulsed-field gel electrophoresis (PFGE) and amplified fragment length polymorphism (AFLP) analysis further differentiated isolates belonging to the same fla sequence types.

CmeR-dependent gene Cj0561c is induced more effectively by bile salts than the CmeABC efflux pump in both human and poultry Campylobacter jejuni strains.

The multidrug efflux pump CmeABC described in the food-borne pathogen Campylobacter jejuni was shown to be an important element of bile resistance and significant for successful colonization of chicken intestines. Recently, another gene (Cj0561c) strongly suppressed by the same repressor (CmeR) that regulates expression of CmeABC was identified in C. jejuni NCTC 11168. Initial data suggested that, similarly to cmeABC, Cj0561c could be induced by bile salts. In the present study, the occurrence of the Cj0561c gene and bile-salt-dependent induction was investigated in ten poultry and eight human C. jejuni strains. The Cj0561c gene was present in all strains. When cultured without addition of bile salts, the transcription level of that gene was about tenfold higher than that of cmeABC. Bile salts cholate and taurocholate induced transcription of cmeABC 1.66-fold and 2.71-fold and that of Cj0561c 3.71-fold and 10.99-fold, respectively. Thus Cj0561c was more effectively induced by bile salts than cmeABC and taurocholate was superior to cholate as an inducer of transcription. More efficient induction of both cmeABC and Cj0561c by taurocholate might be the reason for the higher minimum inhibitory concentrations (MICs) observed with taurocholate than with cholate (100 mg/ml vs. 10 mg/ml). There was no significant difference between poultry and human C. jejuni strains concerning the transcription levels of cmeABC and Cj0561c in cultures without bile salts and concerning bile-salt-induced changes in transcription of cmeABC and Cj0561c. Thus, higher MIC values observed for taurocholate in human strains than in poultry strains (200 mg/ml vs. 75 mg/ml) could not be explained by different responses of cmeABC and Cj0561c to bile salts. Therefore, they must be due to another mechanism.

Advanced sample preparation for the molecular quantification of Staphylococcus aureus in artificially and naturally contaminated milk.

Sample treatment is an essential element when using real-time PCR for quantification of pathogens directly on food samples. This study comparatively evaluated three different principles of sample treatment, i.e. immunomagnetic separation based on phage-derived cell wall binding molecules, matrix solubilization and flotation, in order to establish their suitability for quantifying low numbers of Staphylococcus aureus in milk. All three procedures succeeded to remove S. aureus from the milk matrix, either raw or pasteurized, and, as a result of the concentration of the target cells, minimized the effect of milk associated PCR inhibitors. Sample preparation based on immunomagnetic separation albeit of being user friendly, specific and rapid, failed to allow quantification of low and medium numbers (<10(4)CFU) of S. aureus. In a mastitic milk model cell wall binding domain (CBD)-based target cell extraction revealed results most closely matching those derived from culture-based quantification. Both matrix lysis and flotation allowed quantification of S. aureus at a level of 1-10 cells per ml. Both methods resulted in higher numbers of bacterial cell equivalents (bce) than plating could reveal. Since both methods harvest cells that have been subjected to either mechanical and chemical stresses before quantification, we concluded that the higher bce numbers resulted from a disaggregation of S. aureus clusters initially present in the inoculum. Conclusively, since likely each S. aureus cell of a toxigenic strain contributes to enterotoxin production, molecular quantification could provide an even more realistic impact assessment in outbreak investigations than plating does.

A novel real-time PCR assay for specific detection and quantification of Mycobacterium avium subsp. paratuberculosis in milk with the inherent possibility of differentiation between viable and dead cells.

BACKGROUND: Mycobacterium avium subsp. paratuberculosis (MAP) is the etiological agent of paratuberculosis (Johne's disease) in ruminants and is suggested to be one of the etiologic factors in Crohn's disease in humans. Contaminated milk might expose humans to that pathogen. The aim of the present study was to develop a novel real-time PCR assay providing the additional possibility to detect viable Mycobacterium avium subsp. paratuberculosis (MAP) based on the MAP-specific Mptb52.16 target. The design included an internal amplification control to identify false negative results. FINDINGS: Inclusivity and exclusivity tested on 10 MAP strains, 22 non-MAP mycobacteria, and 16 raw milk microflora strains achieved 100%. The detection limit in artificially contaminated raw milk was 2.42 × 101 MAP cells/ml milk. In a survey of naturally contaminated samples obtained from dairy herds with a known history of paratuberculosis, 47.8% pre-milk and 51.9% main milk samples tested positive. Real-time PCR-derived MAP-specific bacterial cell equivalents (bce) ranged from 1 × 100 to 5.1 × 102 bce/51 ml; the majority of samples had less than one bce per ml milk. Expression of the chosen target was detected in artificially contaminated raw milk as well as inoculated Dubos broth, thus confirming the real-time PCR assay's potential to detect viable MAP cells. CONCLUSIONS: Concentrating the DNA of a large sample volume in combination with the newly developed real-time PCR assay permitted quantification of low levels of MAP cells in raw milk and pasteurized milk. The selected target - Mptb52.16 - is promising with regard to the detection of viable MAP. Future studies integrating quantitative DNA- and RNA-based data might provide important information for risk assessment concerning the presence of MAP in raw milk and pasteurized milk.