Loop-mediated isothermal amplification (LAMP) assay was effective in detecting Salmonella enterica in naturally contaminated liquid egg samples. Salmonella was detected in 110 samples taken from four eggbreaking plants. The egg samples were pre-enriched in buffered peptone water (BPW) at 37°C for 20 h. The selective enrichment was done in Rappaport-Vassiliadis or tetrathionate broth and plated onto xylose lysine deoxycholate agar and brilliant green agar, modified. In addition, the PCR assay was used to detect Salmonella after pre-enrichment in BPW at 37°C for 20 h. The culture method and PCR assay were compared to the LAMP assay, which was also performed after pre-enrichment in BPW. PCR failed to detect Salmonella in 10% of 110 samples, whereas the culture method and LAMP assay successfully identified Salmonella in all samples. However, the LAMP assay was found to be much more rapid than the culture method and as sensitive in detecting Salmonella from liquid eggs. In all of the egg-breaking plants studied, Salmonella was isolated on most tested days. The positive samples showed that more than 75% of the Salmonella strains had identical genetic patterns when analyzed by pulsed-field gel electrophoresis. This suggests that the same Salmonella strains having survived long periods of time in the plants were contaminating the production line. The LAMP assay is rapid, specific, and sensitive for Salmonella detection in liquid eggs and is able to monitor Salmonella contamination in egg-handling plants more reliably.

A loop-mediated isothermal amplification (LAMP) assay was developed to detect Vero toxin (VT)-producing Escherichia coli rapidly (within 60 min). The 24 strains of VT-producing E. coli were successfully amplified, but 6 strains of non-VT-producing E. coli and 46 bacterial species other than E. coli were not. The sensitivity of the LAMP assay was found to be >0.7 c.f.u. per test using serogroups O157, O26 and O111 of VT-producing E. coli; this sensitivity is greater than that obtained by PCR assay. Furthermore, the LAMP assay was examined for its ability to detect VT-producing E. coli in food because of the difficulty of detection in food samples. The recovery of VT-producing E. coli by LAMP assay from beef and radish sprouts inoculated with the pathogen was high, similar to that obtained using culture methods with direct plating and/or plating after immunomagnetic separation. Although PCR assay was unable to recover VT-producing E. coli from half of the radish samples, LAMP assay was successful in most samples. In addition, VT-producing E. coli was successfully detected in cultures of the beef samples by LAMP assay, but not by the culture method. The LAMP products in naturally contaminated beef samples were analysed to confirm the specific amplification of the VT-encoding gene, and were found to show a specific ladder band pattern on agarose gel after electrophoresis. Additionally the sequences of the LAMP products coincided well with the expected sequences of the VT-encoding gene. These results indicate that the proposed LAMP assay is a rapid, specific and sensitive method of detecting the VT-producing E. coli.

Intake of a small dose of foodborne pathogens can cause infection. In this study, an estimation of the infectious dose of the pathogens was obtained by conducting microbiological risk assessments. The contamination levels of foodborne pathogens were analysed in 17 outbreaks of Salmonella, Escherichia coli O157, enterotoxigenic E. coli, Vibrio parahaemolyticus, and Campylobacter jejuni occurring in Japan between 2004 and 2006. The infectious dose was estimated in 14 of the 17 outbreaks utilizing existing data. In three outbreaks of Salmonella infection in which the infection rate was 89-100%, the dose of the ingested pathogens was estimated to be 259,000-14,000,000,000 c.f.u. In other outbreaks of Salmonella infection, the infection rate and dose of the ingested pathogens were 10-66·4% and 81-1560 c.f.u. or most probable number (MPN), respectively. The ingested Salmonella dose is likely to be related to the infection rate; however, storage conditions should be taken into account when making this determination. In an outbreak of E. coli O157 infection, the infection rate and ingestion dose were 100% and 2 to <9 c.f.u., respectively, while in an outbreak of enterotoxigenic E. coli infection, they were 93% and 25-1000 c.f.u., respectively. Finally, in an outbreak of C. jejuni infection, the infection rate and ingestion dose were 37·5% and 360 MPN, respectively. These results will be particularly valuable for risk assessment.

No abstract

No abstract