A Multicenter Proposal for a Fast Tool To Screen Biosecure Chicken Flocks for the Foodborne Pathogen Campylobacter.

The present multicenter study aimed at assessing the performance of air sampling as a novel method for monitoring Campylobacter in biosecure poultry farms. We compared, using a harmonized procedure, the bacteriological isolation protocol (ISO 10272-1:2017) and a real-time PCR method used on air filter samples. Air samples and boot swabs were collected from 62 biosecure flocks from five European countries during the summer of 2019. For air filters, the frequency of PCR-positive findings was significantly higher (n = 36; 58%) than that obtained with the cultivation methods (P < 0.01; standardized residuals). The cultivation protocols (one with Bolton enrichment and one with Preston enrichment) were comparable to each other but returned fewer positive samples (0 to 8%). The association between type of sample and frequency of PCR-positive findings was statistically confirmed (P < 0.01; Fisher´s exact test), although no culture-positive air filters were detected using direct plating. For the boot swabs, the highest number of positive samples were detected after enrichment in Preston broth (n = 23; 37%), followed by direct plating after homogenization in Preston (n = 21; 34%) or Bolton broth (n = 20; 32%). It is noteworthy that the flocks in Norway, a country known to have low Campylobacter prevalence in biosecure chicken flocks, tested negative for Campylobacter by the new sensitive approach. In conclusion, air sampling combined with real-time PCR is proposed as a multipurpose, low-cost, and convenient screening method that can be up to four times faster and four times more sensitive than the current boot-swab testing scheme used for screening biosecure chicken production.IMPORTANCECampylobacter bacteria are the cause of the vast majority of registered cases of foodborne illness in the industrialized world. In fact, the bacteria caused 246,571 registered cases of foodborne illness in 2018, which equates to 70% of all registered cases in Europe that year. An important tool to prevent campylobacters from making people sick is good data on where in the food chain the bacterium is present. The present study reports a new test method that quadruples the likelihood of identifying campylobacter-positive chicken flocks. It is important to identify campylobacter-positive flocks before they arrive at the slaughterhouse, because negative flocks can be slaughtered first in order to avoid cross-contamination along the production line.

Campylobacter in chicken - Critical parameters for international, multicentre evaluation of air sampling and detection methods.

The present pilot study aimed at evaluating air sampling as a novel method for monitoring Campylobacter in poultry farms. We compared the bacteriological isolation of Campylobacter from boot swabs and air filter samples using ISO 10272-1:2017. A secondary aim was to evaluate the use of molecular methods, i.e. real time PCR, on the same sample set. Samples from 44 flocks from five European countries were collected, and included air samples, in parallel with boot swabs. Campylobacter spp. was isolated from seven of 44 boot swabs from three of five partners using the enrichment method. Two of these positive boot swab samples had corresponding positive air samples. Using enrichment, one positive air sample was negative in the corresponding boot swabs, but Campylobacter spp. was isolated from direct plating of the boot swab sample. One partner isolated Campylobacter spp. from six of 10 boot swabs using direct plating. Overall, 33 air filter samples were screened directly with PCR, returning 14 positive results. In conclusion, there was a lack of correspondence between results from analysis of boot swabs and air filters using ISO 10272-1:2017. In contrast, the combination of air filters and direct real-time PCR might be a way forward. Despite the use of the detailed ISO protocols, there were still sections that could be interpreted differently among laboratories. Air sampling may turn into a multi-purpose and low-cost sampling method that may be integrated into self-monitoring programs.

Surveillance of Foodborne Pathogens: Towards Diagnostic Metagenomics of Fecal Samples.

Diagnostic metagenomics is a rapidly evolving laboratory tool for culture-independent tracing of foodborne pathogens. The method has the potential to become a generic platform for detection of most pathogens and many sample types. Today, however, it is still at an early and experimental stage. Studies show that metagenomic methods, from sample storage and DNA extraction to library preparation and shotgun sequencing, have a great influence on data output. To construct protocols that extract the complete metagenome but with minimal bias is an ongoing challenge. Many different software strategies for data analysis are being developed, and several studies applying diagnostic metagenomics to human clinical samples have been published, detecting, and sometimes, typing bacterial infections. It is possible to obtain a draft genome of the pathogen and to develop methods that can theoretically be applied in real-time. Finally, diagnostic metagenomics can theoretically be better geared than conventional methods to detect co-infections. The present review focuses on the current state of test development, as well as practical implementation of diagnostic metagenomics to trace foodborne bacterial infections in fecal samples from animals and humans.

Gene-Based Pathogen Detection: Can We Use qPCR to Predict the Outcome of Diagnostic Metagenomics?

In microbial food safety, molecular methods such as quantitative PCR (qPCR) and next-generation sequencing (NGS) of bacterial isolates can potentially be replaced by diagnostic shotgun metagenomics. However, the methods for pre-analytical sample preparation are often optimized for qPCR, and do not necessarily perform equally well for qPCR and sequencing. The present study investigates, through screening of methods, whether qPCR can be used as an indicator for the optimization of sample preparation for NGS-based shotgun metagenomics with a diagnostic focus. This was used on human fecal samples spiked with 10³ or 106 colony-forming units (CFU)/g Campylobacter jejuni, as well as porcine fecal samples spiked with 10³ or 106 CFU/g Salmonella typhimurium. DNA was extracted from the samples using variations of two widely used kits. The following quality parameters were measured: DNA concentration, qPCR, DNA fragmentation during library preparation, amount of DNA available for sequencing, amount of sequencing data, distribution of data between samples in a batch, and data insert size; none showed any correlation with the target ratio of the spiking organism detected in sequencing data. Surprisingly, diagnostic metagenomics can have better detection sensitivity than qPCR for samples spiked with 10³ CFU/g C. jejuni. The study also showed that qPCR and sequencing results may be different due to inhibition in one of the methods. In conclusion, qPCR cannot uncritically be used as an indicator for the optimization of sample preparation for diagnostic metagenomics.

Towards diagnostic metagenomics of Campylobacter in fecal samples.

BACKGROUND: The development of diagnostic metagenomics is driven by the need for universal, culture-independent methods for detection and characterization of pathogens to substitute the time-consuming, organism-specific, and often culture-based laboratory procedures for epidemiological source-tracing. Some of the challenges in diagnostic metagenomics are, that it requires a great next-generation sequencing depth and unautomated data analysis. RESULTS: DNA from human fecal samples spiked with 7.75 × 101-7.75 × 107 colony forming unit (CFU)/ml Campylobacter jejuni and chicken fecal samples spiked with 1 × 102-1 × 106 CFU/g Campylobacter jejuni was sequenced and data analysis was done by the metagenomic tools Kraken and CLARK. More hits were obtained at higher spiking levels, however with no significant linear correlations (human samples p = 0.12, chicken samples p = 0.10). Therefore, no definite detection limit could be determined, but the lowest spiking levels found positive were 7.75 × 104 CFU/ml in human feces and 103 CFU/g in chicken feces. Eight human clinical fecal samples with estimated Campylobacter infection loads from 9.2 × 104-1.0 × 109 CFU/ml were analyzed using the same methods. It was possible to detect Campylobacter in all the clinical samples. CONCLUSIONS: Sensitivity in diagnostic metagenomics is improving and has reached a clinically relevant level. There are still challenges to overcome before real-time diagnostic metagenomics can replace quantitative polymerase chain reaction (qPCR) or culture-based surveillance and diagnostics, but it is a promising new technology.

Microbial food safety: Potential of DNA extraction methods for use in diagnostic metagenomics.

The efficiency of ten widely applied DNA extraction protocols was evaluated for suitability for diagnostic metagenomics. The protocols were selected based on a thorough literature study. Chicken fecal samples inoculated with about 1×10(3) and 1×10(6) CFU/g Campylobacter jejuni were used as a model. The evaluation was performed based on total DNA yield measured by fluorometry, and quality and quantity of C. jejuni DNA measured by real-time PCR. There was up to a 25-fold variance between the lowest (NucliSens miniMAG, BIOMÉRIEUX) and highest (PowerLyzer PowerSoil DNA Isolation Kit, MO BIO Laboratories) yielding protocols. The PowerLyzer PowerSoil DNA Isolation Kit performed significantly better than all other protocols tested. Selected protocols were modified, i.e., extended heating and homogenization, resulting in increased yields of total DNA. For QIAamp Fast DNA Stool Mini Kit (Qiagen) a 7-fold increase in total DNA was observed following the protocol for human DNA analysis and including a 5 min heating step at 70°C. For the PowerLyzer PowerSoil and the PowerFecal DNA Isolation Kit (MO BIO Laboratories) the total DNA fold increase was 1.6 to 1.8 when including an extra 10 min of bead-vortexing. There was no correlation between the yield of total DNA and the amount of PCR-amplifiable DNA from C. jejuni. The protocols resulting in the highest yield of total DNA did not show correspondingly increased levels of C. jejuni DNA as determined by PCR. In conclusion, substantial variation in the efficiency of the protocols to extract DNA was observed. The highest DNA yield was obtained with the PowerLyzer PowerSoil DNA Isolation Kit, whereas the FastDNA SPIN Kit for Feces (MP Biomedicals) resulted in the highest amount of PCR-amplifiable C. jejuni DNA.

Monitoring Campylobacter in the poultry production chain­from culture to genes and beyond.

Improved monitoring tools are important for the control of Campylobacter bacteria in poultry production. Standardized reference culture methods issued by national and international standardization organizations are time-consuming, cumbersome and not amenable to automation for screening of large numbers of samples. The ultimate goal for rapid monitoring of Campylobacter is to prevent contaminated meat from entering the food market. Currently, real-time PCR is fulfilling abovementioned criteria to a certain extent. Further development of real-time PCR, microarray PCR, miniaturized biosensors, chromatographic techniques and DNA sequencing can improve our monitoring capacity at a lower cost. Combined with innovative sampling and sample treatment, these techniques could become realistic options for on-farm and liquid-sample monitoring at slaughterhouses.

Enumeration of salmonellae in table eggs, pasteurized egg products, and egg-containing dishes by using quantitative real-time PCR.

Salmonellae are a major cause of food-borne outbreaks in Europe, with eggs and egg products being identified as major sources. Due to the low levels of salmonellae in eggs and egg products, direct quantification is difficult. In the present study, enrichment quantitative real-time PCR (qPCR) was employed for enumeration of salmonellae in different matrices: table eggs, pasteurized egg products, and egg-containing dishes. Salmonella enterica serovar Enteritidis and S. enterica serovar Tennessee were used to artificially contaminate these matrices. The results showed a linear regression between the numbers of salmonellae and the quantification cycle (Cq) values for all matrices used, with the exception of pasteurized egg white. Standard curves were constructed by using both stationary-phase cells and heat-stressed cells, with similar results. Finally, this method was used to evaluate the fate of salmonellae in two egg-containing dishes, long egg and tiramisu, at abused refrigeration temperatures, and results indicated the growth of bacteria over a 1-week period. In conclusion, enrichment qPCR was shown to be reliable for enumeration of salmonellae in different egg products.

Evaluation of direct 16S rDNA sequencing as a metagenomics-based approach to screening bacteria in bottled water.

Deliberate or accidental contamination of food, feed, and water supplies poses a threat to human health worldwide. A rapid and sensitive detection technique that could replace the current labor-intensive and time-consuming culture-based methods is highly desirable. In addition to species-specific assays, such as PCR, there is a need for generic methods to screen for unknown pathogenic microorganisms in samples. This work presents a metagenomics-based direct-sequencing approach for detecting unknown microorganisms, using Bacillus cereus (as a model organism for B. anthracis) in bottled water as an example. Total DNA extraction and 16S rDNA gene sequencing were used in combination with principle component analysis and multicurve resolution to study detection level and possibility for identification. Results showed a detection level of 10(5) to 10(6) CFU/L. Using this method, it was possible to separate 2 B. cereus strains by the principal component plot, despite the close sequence resemblance. A linear correlation between the artificial contamination level and the relative amount of the Bacillus artificial contaminant in the metagenome was observed, and a relative amount value above 0.5 confirmed the presence of Bacillus. The analysis also revealed that background flora in the bottled water varied between the different water types that were included in the study. This method has the potential to be adapted to other biological matrices and bacterial pathogens for fast screening of unknown bacterial threats in outbreak situations.

Design and data analysis of experimental trials to test vaccine candidates against zoonotic pathogens in animals: the case of a clinical trial against campylobacter in broilers.

The development of effective vaccines against zoonotic pathogens represents a priority in public health protection programs. The design of clinical trials and appropriate data analysis of the experiments results are crucial for the assessment of vaccine effectiveness. This manuscript reviews important issues related to the assessment of the effectiveness of vaccines designed to obtain a quantitative reduction of the pathogen in animals or animal products. An effective vaccine will reduce the risk of human infections and therefore the number of human cases. Important considerations will be illustrated using a vaccination trial of a new campylobacter vaccine candidate developed to reduce the numbers of campylobacter in chickens and consequently the numbers of human campylobacteriosis cases. The design of the author's vaccination trial was based on the use of isolators, a parallel group design and several rotations. The effect of clustering or design effect was considered in the sample size calculations. Chickens were randomly assigned to different isolators (treatments) and challenged with Campylobacter jejuni. Samples were obtained at different intervals and processed in the laboratory. C. jejuni counts were determined as colony-forming unit-per-gram of chicken cecum or fecal mass in order to assess vaccine effectiveness. A desired vaccine effect of 2 logs reduction on the numbers of C. jejuni recovered from vaccinated chickens was selected. Sample-size calculations, desired vaccine effect, biological and epidemiological aspects, experimental design and appropriate statistical analysis of data considering group or clustering effects will be the focus of this manuscript.

Detection of Salmonella in meat: comparative and interlaboratory validation of a noncomplex and cost-effective pre-PCR protocol.

Cost-effective and rapid monitoring of Salmonella in the meat production chain can contribute to food safety. The objective of this study was to validate an easy-to-use pre-PCR sample preparation method based on a simple boiling protocol for screening of Salmonella in meat and carcass swab samples using a real-time PCR method. The protocol included incubation in buffered peptone water, centrifugation of an aliquot, and a boiling procedure. The validation study included comparative and interlaboratory trials recommended by the Nordic Organization for Validation of Alternative Microbiological Methods (NordVal). The comparative trial was performed against a reference method (NMKL 187, 2007) and a PCR method previously approved by NordVal with a semiautomated magnetic bead-based DNA extraction step using 122 artificially contaminated samples. The LOD was found to be 3.0, 3.2, and 3.4 CFU/sample for the boiling, magnetic bead-based, and NMKL 187 methods, respectively. When comparing the boiling method with the magnetic beads, the relative accuracy (AC), relative sensitivity (SE), and relative specificity (SP) were 98, 102, and 98%, respectively (Cohen's kappa index 0.95). When comparing results obtained by the boiling to the culture-based method, the AC, SE, and SP were found to be 98, 102, and 98%, respectively (kappa index 0.93). In the interlaboratory trial including valid results from 11 laboratories, apart from two false-positive samples by the boiling method combined with PCR, no deviating results were obtained (SP, SE, and AC were 100, 95, and 97%, respectively). This test is under implementation by the Danish meat industry, and can be useful for screening of large number of samples in the meat production, especially for fast release of minced meat with a short shelf life.

Validation of an open-formula, diagnostic real-time PCR method for 20-h detection of Salmonella in animal feeds.

A comparative study of a 20-h, non-commercial, open-formula PCR method and the standard culture-based method NMKL 187, for detection of Salmonella, was performed according to the validation protocol from the Nordic organisation for validation of alternative microbiological methods (NordVal) on 81 artificially or naturally contaminated animal feed samples. The PCR method is based on culture enrichment in buffered peptone water for 16 ± 2 h followed by a magnetic beads based semi automated DNA extraction and real-time PCR analysis, including an internal amplification control. The limit of detection (LOD50) was found to be 7.19 and 7.24 CFU/sample for the PCR method and NMKL187, respectively. A very good correlation between results obtained by the two methods was found (Cohen's kappa=0.92). The relative accuracy, relative sensitivity and relative specificity were found to be 97.5%, 102.0% and 96.6%, respectively. This method is the fastest open PCR based analysis protocol for detection of Salmonella in feed samples. Implementing rapid methods such as the one validated in this study can speed up Salmonella testing of feed for food-producing animals.

Direct detection of single-nucleotide polymorphisms in bacterial DNA by SNPtrap.

A major challenge with single-nucleotide polymorphism (SNP) fingerprinting of bacteria and higher organisms is the combination of genome-wide screenings with the potential of multiplexing and accurate SNP detection. Single-nucleotide extension by the minisequencing principle represents a technology that both is highly accurate and enables multiplexing. A current bottleneck for direct genome analyses by minisequencing, however, is the sensitivity, since minisequencing relies on linear signal amplification. Here, we present SNPtrap, which is a novel approach that combines the specificity and possibility of multiplexing by minisequencing with the sensitivity obtained by logarithmic signal amplification by polymerase chain reaction (PCR). We show a SNPtrap proof of principle in a model system for two polymorphic SNP sites in the Salmonella tetrathionate reductase gene (ttrC).

Development and evaluation of novel one-step TaqMan realtime RT-PCR assays for the detection and direct genotyping of genogroup I and II noroviruses.

BACKGROUND: Current detection and genotyping methods of genogroup (G) I and II noroviruses (NoVs) consist of a 2-step approach including detection of viral RNA by TaqMan realtime RT-PCR (RT-qPCR) followed by conventional RT-PCR and sequencing of partial regions of ORF1 or ORF2. OBJECTIVE: To develop novel long-template one-step TaqMan assays (L-RT-qPCR) for the rapid detection and direct genotyping of GI and GII NoVs and to evaluate the sensitivity and specificity of the assays. STUDY DESIGN: GI and GII-specific broadly reactive L-RT-qPCR assays were developed by combining existing NoV primers and probes targeting the open reading frame (ORF)1-ORF2 junction as well as region C at the 5'-ORF2. The assays were validated using GI and GII RNA transcripts and a coded panel of 75 stool samples containing NoV strains representing 9 GI genotypes and 12 GII genotypes, as well as sapoviruses, astroviruses, polioviruses, and rotaviruses. L-RT-qPCR products were typed by sequencing. RESULTS: The novel GI and GII L-RT-qPCR assays detected and typed all but one of the NoV positive panel samples. As few as 5-500 RNA copies could be accurately typed by sequencing of amplicons. CONCLUSIONS: We developed novel one-step TaqMan RT-qPCR assays for the sensitive detection and direct genotyping of GI and GII NoVs from clinical and environmental matrices.

Collaborative validation of a rapid method for efficient virus concentration in bottled water.

Enteric viruses, including norovirus (NoV) and hepatitis A virus (HAV), have emerged as a major cause of waterborne outbreaks worldwide. Due to their low infectious doses and low concentrations in water samples, an efficient and rapid virus concentration method is required for routine control. Three newly developed methods, A, B and C, for virus concentration in bottled water were compared against the reference method D: (A) Convective Interaction Media (CIM) monolithic chromatography; filtration of viruses followed by (B) direct lysis of viruses on membrane; (C) concentration of viruses by ultracentrifugation; and (D) concentration of viruses by ultrafiltration, for each methods' (A, B and C) efficacy to recover 10-fold dilutions of HAV and feline calicivirus (FCV) spiked in bottles of 1.5L of mineral water. Within the tested characteristics, all the new methods showed better performance than method D. Methods A, B and C shared a limit of detection (LOD(50)) of nine 50%-tissue culture infectious dose (TCID(50)) of FCV/1.5L, but differed with regard to the LOD(50)'s of HAV with 45, 361 and 3607 TCID(50)/1.5L, respectively, and the percentage of recoveries of HAV/FCV with 34/6, 32/25 and 0.3/0.5, respectively. Method B resulted in significantly (p<0.0001) lower C(t)-values for both HAV and FCV relative to the reference method D than any of the other methods. The most efficient method (B) was evaluated through a collaborative trial by five laboratories for the detection of HAV, FCV and NoV genogroup I and II (GI and GII), which resulted in the corresponding average LOD(50)'s and percentage of recoveries: 211 TCID(50)/1.5L and 51% for HAV; 66 TCID(50)/1.5L and 34% for FCV; 9 reverse transcriptase-PCR Units (RT-PCR U)/1.5L and 61% for NoV GI and 286 RT-PCR U/1.5L and 35% for NoV GII. The results indicate that method B could be considered robust enough for routine control and useful for harmonized data generation.

A novel strategy to obtain quantitative data for modelling: combined enrichment and real-time PCR for enumeration of salmonellae from pig carcasses.

Salmonella is a major zoonotic pathogen which causes outbreaks and sporadic cases of gastroenteritis in humans worldwide. The primary sources for Salmonella are food-producing animals such as pigs and poultry. For risk assessment and hazard analysis and critical control point (HACCP) concepts, it is essential to produce large amounts of quantitative data, which is currently not achievable with the standard cultural based methods for enumeration of Salmonella. This study presents the development of a novel strategy to enumerate low numbers of Salmonella in cork borer samples taken from pig carcasses as a first concept and proof of principle for a new sensitive and rapid quantification method based on combined enrichment and real-time PCR. The novelty of the approach is in the short pre-enrichment step, where for most bacteria, growth is in the log phase. The method consists of an 8h pre-enrichment of the cork borer sample diluted 1:10 in non-selective buffered peptone water, followed by DNA extraction, and Salmonella detection and quantification by real-time PCR. The limit of quantification was 1.4 colony forming units (CFU)/20 cm(2) (approximately 10 g) of artificially contaminated sample with 95% confidence interval of ± 0.7 log CFU/sample. The precision was similar to the standard reference most probable number (MPN) method. A screening of 200 potentially naturally contaminated cork borer samples obtained over seven weeks in a slaughterhouse resulted in 25 Salmonella-positive samples. The analysis of salmonellae within these samples showed that the PCR method had a higher sensitivity for samples with a low contamination level (<6.7 CFU/sample), where 15 of the samples negative with the MPN method was detected with the PCR method and 5 were found to be negative by both methods. For the samples with a higher contamination level (6.7-310 CFU/sample) a good agreement between the results obtained with the PCR and MPN methods was obtained. The quantitative real-time PCR method can easily be applied to other food and environmental matrices by adaptation of the pre-enrichment time and media.

Culture-independent quantification of Salmonella enterica in carcass gauze swabs by flotation prior to real-time PCR.

To facilitate quantitative risk assessment in the meat production chain, there is a need for culture-independent quantification methods. The aim of this study was to evaluate the use of flotation, a non-destructive sample preparation method based on traditional buoyant density centrifugation, for culture-independent quantification of intact Salmonella in pig carcass gauze swabs (100 cm(2)) prior to quantitative PCR (qPCR). A novel approach was investigated, excluding the homogenization step prior to flotation, to improve the detection limit and speed up the quantification procedure. The buoyant density of two Salmonella strains in different growth conditions was determined to be 1.065-1.092 g/ml. Based on these data, an optimal discontinuous flotation with three different density layers, ~1.200, 1.102 and 1.055 g/ml, was designed for extracting intact Salmonella cells from pig carcass swabs. The method allowed accurate quantification from 4.4 × 10(2) to at least 2.2 × 10(7)CFU Salmonella per swab sample using qPCR (without preceding DNA extraction) or selective plating on xylose lysine deoxycholate agar. Samples with 50CFU could be detected occasionally but fell outside the linear range of the standard curve. The swab samples showed a broad biological diversity; for seven samples not inoculated with Salmonella, the microbial background flora (BGF) was determined to 5.0 ± 2.2 log CFU/ml sample withdrawn after flotation. It was determined that the proceeding PCR step was inhibited by BGF concentrations of ≥ 6.1 × 10(8)CFU/swab sample, but not by concentrations ≤ 6.1 × 10(6)CFU/swab sample. By using the gauze swabs directly in the flotation procedure, the homogenization step normally used for preparation of food-related samples could be excluded, which simplified the culture-independent quantification method considerably.

Towards biotracing in food chains.

Biotracing is tracing (backward)/tracking (forward) biological contamination in the food/feed chain. Advances in detection technologies, improvements in molecular marker identification, clearer understanding of pathogenicity markers, improved modelling methodologies and, more importantly, the integration of these disciplines will lead to better capability in full-chain tracing and tracking biological contaminations (biotracing). The advantages of improved biotraceability are faster intervention, limited recalls and more targeted remedial action. The project is not dealing with risk assessments but developing tools that can be used in "second-generation" risk assessments involving quantitative microbiology. This concept is the core activity of BIOTRACER, which is an Integrated Project (2007-2011) funded by the EU 6th Framework Programme. The research in biotracing is organised into five Research Areas, and 21 cross-disciplinary work packages that cover tracing and tracking of contamination in feed, meat and dairy chains, in addition to accidental and deliberate contamination of bottled water. The BIOTRACER Consortium consists of 46 partners, including Europe's largest food/feed industries, several SMEs, and relevant International Cooperation (INCO) countries. The Consortium includes experts in predictive microbiology, database developers, software companies, risk assessors, risk managers, system biologists, food and molecular microbiologists, legislative officers, standardization and validation members and food retailers. The outcomes will ensure a more reliable and rapid response to a microbial contamination event.

Microarray-based genotyping of Salmonella: inter-laboratory evaluation of reproducibility and standardization potential.

Bacterial food-borne infections in humans caused by Salmonella spp. are considered a crucial food safety issue. Therefore, it is important for the risk assessments of Salmonella to consider the genomic variation among different isolates in order to control pathogen-induced infections. Microarray technology is a promising diagnostic tool that provides genomic information on many genes simultaneously. However, standardization of DNA microarray analysis is needed before it can be used as a routine method for characterizing Salmonella isolates across borders and laboratories. A comparative study was designed in which the agreement of data from a DNA microarray assay used for typing Salmonella spp. between two different labs was assessed. The study was expected to reveal the possibility of obtaining the same results in different labs using different equipment in order to evaluate the reproducibility of the microarray technique as a first step towards standardization. The low-density array contains 281 57-60-mer oligonucleotide probes for detecting a wide range of specific genomic marker genes associated with antibiotic resistance, cell envelope structures, mobile genetic elements and pathogenicity. Several critical methodology parameters that differed between the two labs were identified. These related to printing facilities, choice of hybridization buffer, wash buffers used following the hybridization and choice of procedure for purifying genomic DNA. Critical parameters were randomized in a four-factorial experiment and statistical measures of inter-lab consistency and agreement were performed based on the kappa coefficient. A high level of agreement (kappa=0.7-1.0) in microarray results was obtained even when employing different printing and hybridization facilities, different procedures for purifying genomic DNA and different wash buffers. However, less agreement (Kappa=0.2-0.6) between microarray results were observed when using different hybridization buffers, indicating this parameter as being highly critical when transferring a standard microarray assay between laboratories. In conclusion, this study indicates that DNA microarray assays can be reproduced in at least two different facilities, which is a pre-requisite for the development of standard guidelines.

Validation of a 20-h real-time PCR method for screening of Salmonella in poultry faecal samples.

Efficient and rapid monitoring of Salmonella in the poultry production chain is necessary to assure safe food. The objective was to validate an open-formula real-time PCR method for screening of Salmonella in poultry faeces (sock samples). The method consists of incubation in buffered peptone water for 18+/-2 h, centrifugation of a 1-ml subsample, DNA extraction on the pellet and PCR. The total analysis time is 20 h. The validation study included comparative and collaborative trials, based on the recommendations from the Nordic organization for validation of alternative microbiological methods (NordVal). The comparative trial was performed against a reference method from the Nordic Committee on Food Analysis (NMKL187, 2007) using 132 artificially and naturally contaminated samples. The limit of detection (LOD50) was found to be 24 and 33 CFU/sample for the PCR and NMKL187 methods, respectively. The relative accuracy, relative sensitivity and relative specificity were all 100%, when including naturally contaminated samples and samples artificially contaminated with 10-100 CFU/sample. The collaborative trial included six laboratories and valid results were obtained from five of them. Apart from one of the samples that was artificially contaminated with 1-10 CFU/sample being a false negative with PCR for one of the laboratories, no false-positive or false-negative results were reported. This test supplies the growing demand for validated diagnostic PCR methods for screening of samples in the meat production chain to assure safe food.