Food Hygiene and Safety Science (Shokuhin Eiseigaku Zasshi) Vol. 58(2017) No. 4

Microbial concentration in samples of a food product lot has been generally assumed to follow the log-normal distribution in food sampling, but this distribution cannot accommodate the concentration of zero. In the present study, first, a probabilistic study with the most probable number (MPN) technique was done for a target microbe present at a low (or zero) concentration in food products. Namely, based on the number of target pathogen-positive samples in the total samples of a product found by a qualitative, microbiological examination, the concentration of the pathogen in the product was estimated by means of the MPN technique. The effects of the sample size and the total sample number of a product were then examined. Second, operating characteristic (OC) curves for the concentration of a target microbe in a product lot were generated on the assumption that the concentration of a target microbe could be expressed with the Poisson distribution. OC curves for Salmonella and Cronobacter sakazakii in powdered formulae for infants and young children were successfully generated. The present study suggested that the MPN technique and the Poisson distribution would be useful for qualitative microbiological test data analysis for a target microbe whose concentration in a lot is expected to be low.

A sensitive and reliable method for the simultaneous determination of four nitroimidazoles (ipronidazole (IPZ), dimetridazole (DMZ), metronidazole (MNZ) and ronidazole (RNZ)) and three metabolites (IPZ-OH, MNZ-OH and 2-hydroxymethyl-1-methyl-5-nitroimidazole (HMMNI)) in livestock and fishery products was developed. The analytes were extracted from samples with acetone containing acetic acid. The crude extracts were defatted by liquid–liquid partition using acetonitrile and n-hexane followed by solid-phase extraction using a cartridge column packed with divinylbenzene–N-vinylpyrolidone copolymer bearing sulfo groups. The analytes in the eluate from the cartridge column were extracted with ethyl acetate after addition of ammonium sulfate. The solvent was removed from the extract, and the residue was dissolved in 0.1 vol% formic acid. The HPLC separation was performed on a C18 column with a gradient formed from water containing 0.1 vol% formic acid and acetonitrile containing 0.1 vol% formic acid. For detection of the analytes, tandem mass spectrometry with positive ion electrospray ionization was used. The recovery tests were performed on 10 livestock and fishery products. The truenesse ranged from 74.6 to 111.1%, with repeatability of 0.5–8.3 RSD% for the entire procedure. The limit of quantification was 0.0001 mg/kg for IPZ, IPZ-OH, MNZ and MNZ-OH, and 0.0002 mg/mg for DMZ, RNZ and HMMNI.

Cleanup using two types of agents (S-NH₂ and S-Si) developed by the authors was investigated with the aim of removing interfering substances such as catechin and caffeine to enable analysis of pesticide residues in tea. S-NH₂ and S-Si removed approximately 100% of catechin and caffeine in 6 species of tea. Recoveries of 61 pesticides in tea were tested at the level of 0.1 μg/g, and 44 pesticides showed recovery within the range from 70 to 120%, with RSD of less than 10%. With cleanup using S-NH₂ and S-Si, pesticide residues in tea could be analyzed within two hours.

An LC-MS/MS method for the determination of diniconazole in agricultural products, livestock and marine products was developed. Diniconazole in agricultural products was extracted with acetone. The extract was concentrated and partitioned with n-hexane and 10% sodium chloride solution. Agricultural products such as grains and beans were defatted using n-hexane–acetonitrile. Livestock and marine products were extracted with a mixture of acetone and n-hexane, and the organic layer was evaporated to dryness. The residue was defatted using n-hexane–acetonitrile. Cleanup was carried out using a Florisil cartridge column and a graphitized carbon cartridge column for these samples. The LC separation was carried out on an Inertsil ODS-3 column with a linear gradient of 0.1% formic acid and acetonitrile containing 0.1% formic acid. MS was carried out in the positive ion electrospray ionization mode. The calibration curve was linear between 0.00125 to 0.00750 mg/L. Average recoveries (n=5) of diniconazole from 16 kinds of agricultural products, livestock and marine products fortified at the MRLs (0.01 ppm) were 88.3–108%, and the relative standard deviations were 0.5–5.1%. The limits of quantitation were 0.01 mg/kg.

During 2015–2016, we examined norovirus (NoV) RNA in swab specimens collected for investigation of suspected food poisoning outbreaks in Tokyo by real-time RT-PCR. Of 1,726 swab samples, 65 (3.8%) were NoV-positive and all positive swab samples were derived from NoV-positive outbreaks. Swab specimens were positive in 41 of 181 (22.7%) NoV outbreaks, while no positive swabs were detected in NoV-negative outbreaks. PCR fragments amplified from 32 swabs were sequenced, and all of them displayed complete homology with sequences from clinical and food samples. Though the results of swabs may be useful for determining the causative agent and infection route in some outbreaks, there was no case in which the results of swabs alone could elucidate the cause of food poisoning. Swabs may be useful in food poisoning investigations, if the results are interpreted in conjunction with epidemiological findings and clinical data. Swab samples are often collected several days after an outbreak, and the influence of disinfection should be taken into consideration. In NoV outbreaks, 55 out of 640 (8.6%) restroom swab specimens were NoV-positive whereas six of 618 (1.0%) were positive among kitchen swab specimens. In the restroom, the toilet bowl (43.6%) showed the highest positive rate and next was the toilet seat (14.5%). Additionally, NoV was detected at various sites in the restroom, including doorknob and floor. Since NoV-positive swab specimens may suggest that sanitation management is not performed properly in the facility, swab results may be utilized as a basis for hygiene guidance.

New automatic pretreatment equipment (FASVED; Food Automatic Analytical Systems for Veterinary Drugs) was developed. FASVED consists of ten main units: reagent dispenser, homogenizer, transfer hand, lid opening/closing device, centrifugal separator, pipette, shaker, column purification device, centrifugal evaporator and cooling device, and it is capable of freely combining operations by these units. A validation study was performed on two methods for determination of 178 veterinary drugs in livestock products, swine muscle, egg and shrimp, according to the method validation guideline of the Ministry of Health, Labour and Welfare of Japan. The numbers of analytes that satisfied the criteria of the guideline were 148 in swine muscle, 160 in egg and 151 in shrimp.