The Effects of Temperature on the Growth and Heat Resistance of Cronobacter spp.

　The growth and survival of Cronobacter isolates were examined under incubation at different temperatures, and their thermal death behavior was investigated at high temperature conditions of above 50℃. Seventy three strains isolated from fresh vegetables, dried foods and soil were tested: 28 of Cronobacter sakazakii, 5 of C. dublienensis, 27 of C. malonaticus and 13 of C. turicensis. All Cronobacter strains grew and multiplied predominantly at 35 and 44℃ until 16 hours of incubation, but showed poor growth at 15℃, and no growth at 5℃. At 48℃, the bacteria grew slightly during 6 to 8 h-incubation but decreased or were inactivated after 16 h-incubation. The heat resistance of Cronobacter spp. was measured under the conditions of 50, 55, 60, 65 and 70℃. Cronobacter strains survived almost without decrement for 30 min at 50℃, but decreased suddenly and perished completely within 10 to 20 min at 55℃ and within 2 - 5 min at above 60℃. Some food materials should be stored below 5℃ until the preparation, and dried food including powdered infant milk formula should be utilized immediately after reconstitution and preparation.

Occurrence of Cronobacter spp. in Dried Foods, Fresh Vegetables and Soil.

　The present study surveyed the occurrence of Cronobacter spp. in dried foods including milk powder, spices and herbs and others, and fresh vegetables commercially available in markets, and ground soil materials for the agriculture. Cronobacter spp. were isolated from 15% of 33 spice and herb samples and 3% of 36 taste foods, and these were C. turicensis, C. malonaticus, C. sakazakii and C. dubliensis. Cronobacter spp. from fresh vegetables were detected in 12% of field vegetables and 13% of hydroponic vegetables. C. turicensis was prevalent in field vegetables, and C. malonaticus was in hydroponic ones. And, Cronobacter spp. in shredded vegetables were detected from 44% of 9 samples, and these were C. dubliensis, C. turicensis and C. sakazakii. Also, Cronobacter spp. in soil from rice field, vegetable field and sandpits were predominantly C. sakazakii and C. malonaticus.

Identification of Cereulide-Producing Bacillus cereus by Nucleic Acid Chromatography and Reverse Transcription Real-Time PCR.

RNA extracts were analyzed with a nucleic acid sequence-based amplification (NASBA) - nucleic acid chromatography and a reverse transcription-quantitative PCR assay (RT-qPCR) based on the TaqMan probe for identification of cereulide-producing Bacillus cereus. All 100 emetic B. cereus strains were found to give positive results, but 50 diarrheal B. cereus strains and other bacterial species showed negative results in the NASBA-chromatography. That is, the assay could selectively identify the emetic strains among B. cereus strains. Also, the B. cereus contents of more than 10(7) cfu/ml were required for the identification of the cereulide-producing strains in this assay. In qRT-PCR assays, all 100 emetic type strains of B. cereus produced 10(2) - 10(4) copy numbers per ng of the RNA preparation, and the strains produced 10(4) copies including ones which had the high vacuolation activities of HEp-2 cells.

Evaluation of the immunochromatographic device for the detection of verotoxins in cultures of food materials.

The immunochromatographic assay, which targets Shiga toxin 1/verotoxin 1 (VT1) and/or Shiga toxin 2/verotoxin 2 (VT2) independently with same test device, was used for easily, rapidly and specifically detecting verotoxin-producing Escherichia coli among E. coli strains from food and fecal materials. All 10 strains of VT 1 and/or VT 2- producing E. coli among E. coli isolates from various sources showed a positive reaction to VT1- or VT2- antibodies, but other gram-negative and positive bacterial species had a negative reaction. Bacterial counts of 10(8) cfu/ml in enrichment broth and food suspension were required for the detection of VT-producing E. coli. The IC assay described here could detect easily and specifically the verotoxin-producing E. coli within 20 min by pure culture.

Detection of emetic Bacillus cereus by real-time PCR in foods.

The simplex real-time PCR assays based on the TaqMan probe and SYBR green I， which target cereulide synthetase genes (ces genes) were used for rapidly, reliably and sensitively identifying the emetic strains from among Bacillus cereus strains isolated from different sources. Only the emetic strains showed positive reactions to the real-time PCR assays, but all examined strains of diarrheal B. cereus, other Bacillus species and other gram positive and negative bacteria gave negative results. The final identification of emetic B. cereus was possible within 1 to 1.5 h in both simplex real-time PCR procedures. The detection limit of emetic strains in food was suggested to be 10(4)-10(5) cfu/g. Both simplex real-time PCR assays were found to be rapid, sensitive and reliable diagnostic tools that complement the different cellular, immunological and chemical detection methods for cereulide- producing B. cereus, even if a relatively expensive device is required for the assays.

Evaluation of immunochromatography for the rapid and specific identification of Listeria monocytogenes from food.

To rapidly, simply and specifically detect and identify Listeria monocytogenes from food samples, an immunochromatographic assay, based on gold-labeled monoclonal antibodies directed against an antigen common to all serovars of L. monocytogenes, was used. All strains of L. monocytogenes serovars showed a positive reaction to the assay, but all other gram positive and negative bacteria did not. The detection limit of the assay was in the order of 10(6) cfu/ml in fluid medium. The assay could simply and rapidly identify L. monocytogenes within 30 min by a pure culture without special instruments. Even if the selective enrichment cultivation was employed for the isolation and growth of bacteria from food materials, the application of the assay system could detect and identify L. monocytogenes precisely in various food materials within 2 to 3 days.

LC-MS analysis of the emetic toxin, cereulide, produced by Bacillus cereus.

A quantitative and chemical assay of cereulide produced in the cultures by some strains of Bacillus cereus was performed on a HPLC and a ESI electrospray ion trap mass analyzer, using the synthetic cereulide as a standard. All 20 strains of emetic B. cereus were found to produce 27 - 740 ng/ml of cereulide by the LC-MS analysis. In contrast, none of the 10 diarrheal strains produced it. 10(2) cfu/ml of the cereulide producible strain with a 210 ng/ml yield was inoculated into the 10% suspensions of 14 food products, and was incubated at 32°C for 24h. The B. cereus counts in the cultures grew in the order of 10(8) to 10(9) cfu/ml, although the bacteria could not grow in fruits, and the yields of cereulide ranged from 5.18µg in curry to 0.03µg/g of raw material and/or powder material, except for fruits. These culture supernatants were also tested for the biological activity in the HEp-2 cell culture assay. Consequently, a certain correlation was shown between the yields of cereulide and the HEp-2 vacuolation activities. In addition, the supernatants were administered i.p. to 5 Suncus marinus test animals. The emetic dose was calculated to be approximately 16µg/kg.

Rapid identification of emetic Bacillus cereus by immunochromatography.

The immunochromatographic assay, which targets a marker protein co-expressed during the synthesis of cereulide by an emetic Bacillus cereus strain, was used for easily, rapidly and specifically identifying the emetic strains among B. cereus strains from various materials associated with food poisonings. All 50 of the emetic strains showed a positive reaction to the assay, but all 50 diarrheal strains had a negative reaction. The bacterial counts of 108 cfu/ml in enrichment broth and 109 cfu/ml in food-containing enrichment were required for the identification of emetic B. cereus. The present assay could identify easily and specifically the emetic type of B. cereus within 30 min by a pure culture without special techniques and instruments.

Evaluation of an enzyme-linked fluorescent assay for the detection of Listeria monocytogenes from food.

The VIDAS Listeria monocytogenes II (LMO2) method, which is an automated enzyme-linked fluorescent immunoassay (ELFA), was used for rapidly, specifically and sensitively detecting L. monocytogenes in food samples. All 31 L. monocytogenes strains examined gave positive results. The other bacterial species except Salmonella showed completely negative results, but it was suspected that Salmonella spp. had given a false-positive reaction to the assay. As the detectable limit of the assay was 106 cfu/ml in a food suspension, food samples were required to be enriched in order to increase the number of the bacteria to the detectable limit. However, the ELFA was reconfirmed to be applied satisfactorily as a rapid and precise method for the detection of L. monocytogenes in various food samples, even if the culture had to be enriched for 12 h prior to the assay.

The rapid detection of Salmonella from food samples by loop-mediated isothermal amplification (LAMP).

Loop-mediated isothermal amplification (LAMP) assay was applied to the detection of Salmonella in food and human materials. It was possible for the assay to detect Salmonella within 60 min. All of 54 serovars of Salmonella tested were amplified, but all bacteria tested other than Salmonella were not. The LAMP assay could detect 10(2) cfu/ml levels of Salmonella. The specificity was similar to that of a PCR assay, but the sensitivity of LAMP was considered to be greater. Thus, the LAMP assay was confirmed to be a rapid, specific and sensitive detection method for Salmonella.

Susceptibility of biofilm Escherichia coli, Salmonella Enteritidis and Staphylococcus aureus to detergents and sanitizers.

This study was conducted to investigate the susceptibility of the biofilm cells of Escherichia coli O157, Salmonella Enteritidis, and Staphylococcus aureus to some cleaning detergents and sanitizers. No weakly acidic, neutral, and weakly alkaline detergent could remove the biofilm bacteria from stainless steel chips at commonly used concentrations recommended by manufacturers. Among sanitizers, sodium hypochlorite did not completely inactivate any biofilm bacteria at active chlorine concentrations of 25 to 200 microg/ml. Benzalkonium chloride, alkyldiaminoethyl glycine hydrochloride, chlorhexidine digluconate, and polyhexamethylenebiganide inactivated the great majority of E. coli and S. Enteritidis at commonly used concentrations, but did not inactivate S. aureus effectively enough. The biofilm S. aureus population was shown to be more tolerant than the E. coli and/or S. Enteritidis populations to the sanitizers.

Detection of Listeria monocytogenes from food samples by PCR after IMS-plating.

For rapid, accurate and sensitive detection of Listeria monocytogenes in food samples, colonies developed on the selective agar (Oxford agar) after immunomagnetic separation (IMS) were subjected to polymerase chain reaction (PCR) assay with the prf A1-2 primer pair. The proposed assay system was shown experimentally to be capable of specifically detecting the bacteria from food samples contaminated at more than 10(2) cfu/g. However, the enrichment culture after a short period of 16 h with the appropriate selective broth was needed before IMS-plating, because the bacterial contents in most actual food were as low as less than 10(2) cfu/g. However, even if the enrichment cultivation was employed before IMS, L. monocytogenes was detected within 3 days.

A PCR assay based on a sequence-characterized amplified region marker for detection of emetic Bacillus cereus.

A PCR assay for the detection of Bacillus cereus strains able to produce an emetic toxin (cereulide) was developed in this study based on a sequence-characterized amplified region (SCAR) derived from a random amplified polymorphic DNA (RAPD) fragment. One of the RAPD fragments generated was selected, cloned, and sequenced. A set of PCR primers was newly designed from the SCAR obtained (the sequence of the cloned RAPD fragment) and used in this assay. To determine the specificity of the assay, 30 different B. cereus strains, 8 other Bacillus strains (of six species), and 16 other non-Bacillus strains (from 16 genera) were tested. Results were positive for every emetic B. cereus strain and for only one nonemetic B. cereus strain. For all other bacterial strains, results were negative. Bacterial DNA for PCR was prepared by a simple procedure using Chelex 100 resin from the bacterial colony on the agar plate or from culture after growth in brain heart infusion medium. This PCR assay enabled us to detect the bacteria of emetic B. cereus grown on agar plates but not the bacteria of nonemetic B. cereus. To test this PCR assay for the monitoring of the emetic bacteria, 10 to 70 CFU of B. cereus DSM 4312 (emetic) per g of food was inoculated into several foods as an indicator, followed by a 7-h enrichment culture step. Because this PCR assay based on the SCAR derived from the RAPD fragment was able to detect bacterial cells, this assay should be useful for rapid and specific detection of emetic B. cereus.

Detection of seg, seh, and sei genes in Staphylococcus aureus isolates and determination of the enterotoxin productivities of S. aureus isolates Harboring seg, seh, or sei genes.

To investigate the distribution of staphylococcal enterotoxin (SE) A to I (SEA to SEI) genes (sea to sei) in Staphylococcus aureus, 146 isolates obtained in Japan from humans involved in and samples from food poisoning outbreaks, healthy humans, cows with mastitis, and bovine raw milk were analyzed by multiplex PCR. One hundred thirteen (77.4%) S. aureus isolates were found to be positive for one or more se genes. The se genotype was classified into 14 genotypes. seg and sei coexisted in the same S. aureus strain. The newly developed sandwich enzyme-linked immunosorbent assay showed that most seh-harboring S. aureus isolates were able to produce a significant amount of SEH. However, most of the S. aureus isolates harboring seg and about 60% of the isolates harboring sei did not produce a detectable level of SEG or SEI, while reverse transcription-PCR analysis proved that the mRNAs of SEG and SEI were transcribed in S. aureus strains harboring seg and sei genes. These results suggest the importance of quantitative assessment of SEG and SEI production in foods in order to clarify the relationship between these new SEs and food poisoning.