[Development for the Simultaneous Analytical Method of Furan and Alkyl Furans in Processed Foods].

A simultaneous analytical method was developed for the determination of alkyl furans (Furan, 2-methylfuran, 3-methylfuran and 2,5-dimethylfuran) in processed foods by headspace-GC-MS. Single-laboratory validation data of furan, 2-methylfuran, 3-methylfuran and 2,5-dimethylfuran showed good precision and accuracy. The mean recoveries ranged from 92 to 116%, the intermediate precision (RSDi) ranged from 0.9 to 12.9%. The level of LOQ ranged from 0.5 to 1.2 µg/kg (coffee), from 3.5 to 4.1 µg/kg (soy sauce), from 0.4 to 1.3 µg/kg (other foods: clear apple juice, infant formula and baby food), respectively. This method has the sensitivity to detect low levels of furan and alkyl furans contaminated in various foods and is thus applicable to surveillance for risk management in food safety.

[Verification and Improvement of an Official Method for Determining Chlorophyll Decomposition Products Including Pheophorbide].

The official method described in Direction Notification Kanshoku No. 99 （May 8, 1981） for the determination of chlorophyll decomposition products including pheophorbide was verified and improved in order to overcome several problems. Firstly, extraction with a mortar required improvement because of the difficulty of maintaining equal power for a long time. Secondly, the saturated sodium sulfate reagent caused a red-shift of the absorption maximum wavelength from the measured wavelength given in the official method; consequently, the absorption was decreased and a new absorption peak was detected around 729 nm. As a result, chlorophyll decomposition products including pheophorbide were underestimated. Lastly, it was impossible to make up the volume of the diethyl ether layer accurately to 20 mL before measuring the absorption. These points were improved in the modified method, and a validation test was performed. The mean recovery was 82.7％ and the within-laboratory reproducibility was 5.8％.

Four-year surveillance for ochratoxin a and fumonisins in retail foods in Japan.

Between 2004 and 2007 we examined foods from Japanese retail shops for contamination with ochratoxin A (OTA) and fumonisins B(1), B(2), and B(3). A total of 1,358 samples of 27 different products were examined for OTA, and 831 samples of 16 different products were examined for fumonisins. The limits of quantification ranged from 0.01 to 0.5 microg/kg for OTA and 2 to 10 microg/kg for the fumonisins. OTA was detected in amounts higher than limits of quantification in wheat flour, pasta, oatmeal, rye, buckwheat flour and dried buckwheat noodles, raisins, wine, beer, coffee beans and coffee products, chocolate, cocoa, and coriander. OTA was found in more than 90% of the samples of instant coffee and cocoa, and the highest concentration of OTA, 12.5 microg/kg, was detected in raisins. The concentration of OTA in oatmeal, rye, raisins, wine, and roasted coffee beans varied remarkably from year to year. Fumonisins were detected in frozen and canned corn, popcorn grain, corn grits, cornflakes, corn soups, corn snacks, beer, soybeans, millet, and asparagus. The highest concentrations of fumonisins B(1), B(2), and B(3) were detected in corn grits (1,670, 597, and 281 microg/kg, respectively). All of the samples of corn grits were contaminated with fumonisins, and more than 80% of the samples of popcorn grain and corn snacks contained fumonisins. OTA and fumonisins were detected in several food products in Japan; however, although Japan has not set regulatory levels for these mycotoxins, their concentrations were relatively low.

[Development of an HPLC-UV Method for Free Asparagine in Grains].

A novel analytical method was developed for the determination of free asparagine (Asn), which is a precursor of acrylamide, in grains. Asn was extracted from a sample with 5% (w/v) aqueous trichloroacetic acid solution, and the crude extract was cleaned up using a reversed-phase solid-phase cartridge. The cleaned extract was derivatized with dansyl chloride and analyzed by HPLC-UV. HPLC separation was performed by gradient elution on a ODS column using 0.01 mol/L ammonium acetate and acetonitrile mixture as the mobile phase. The calibration curve was linear in the range of 0.5-100 µg/mL. The mean recoveries from potato starch, non-glutinous rice flour and whole wheat flour ranged from 97.7 to 102.6%, the repeatability (RSDs) ranged from 0.8 to 2.0%, and the within-laboratory reproducibility (RSDwr) ranged from 1.4 to 6.2%. Limits of quantitation (LOQs) were 13 mg/kg for potato starch and 4 mg/kg for non-glutinous rice flour. The quantitative values obtained for about 15 kinds of grains using this developed method were approximately equal to those obtained with an automatic amino acid analyzer.

Occurrence of four Fusarium mycotoxins, deoxynivalenol, zearalenone, T-2 toxin, and HT-2 toxin, in wheat, barley, and Japanese retail food.

A survey of the contamination of wheat, barley, and Japanese retail food by four Fusarium mycotoxins, deoxynivalenol (DON), zearalenone (ZEN), T-2 toxin (T-2), and HT-2 toxin (HT-2), was performed between 2010 and 2012. A method for the simultaneous determination of the four mycotoxins by liquid chromatography-tandem mass spectrometry was validated by a small-scale interlaboratory study using two spiked wheat samples (DON was spiked at 20 and 100 µg/kg and ZEN, T-2, and HT-2 at 6 and 20 µg/kg in the respective samples). The recovery of the four mycotoxins ranged from 77.3 to 107.2%. A total of 557 samples of 10 different commodities were analyzed over 3 years by this validated method. Both T-2 and HT-2 were detected in wheat, wheat flour, barley, Job's tears products, beer, corn grits, azuki beans, soybeans, and rice with mixed grains. Only T-2 toxin was detected in sesame seeds. The highest concentrations of T-2 toxin (48.4 µg/kg) and HT-2 toxin (85.0 µg/kg) were present in azuki beans and wheat, respectively. DON was frequently detected in wheat, wheat flour, beer, and corn grits. The contamination level of wheat was below the provisional standard in Japan (1,100 µg/kg). The maximum contamination level of DON was present in a sample of a Job's tears product (1,093 µg/kg). ZEN was frequently detected in Job's tears products, corn grits, azuki beans, rice with mixed grains, and sesame seeds. A sample of a Job's tears product presented the highest ZEN contamination (153 µg/kg). These results indicate that continuous monitoring by multiple laboratories is effective and necessary due to the percentage of positive samples detected.

Survey of 7 trichothecenes in corn-derived feed and feed ingredients.

Corn is a major ingredient of mixed feed, and it has been reported that corn may be contaminated with deoxynivalenol (DON). There is also a possibility of contamination with other trichothecenes. Recently, corn-derived products such as Distiller's Dried Grains with Solubles (DDGS), corn gluten feed and corn gluten meal have been introduced and used for mixed feed. However the actual occurrence of trichothecenes has not been sufficiently investigated. So, in this study, we analyzed DON and 6 other trichothecenes, i.e., 3-acetyl-deoxynivalenol (3AcDON), 15-acetyl-deoxynivalenol (15AcDON), T-2 toxin, HT-2 toxin, nivalenol, and fusarenon-X, in DDGS, corn gluten feed, corn gluten meal, and mixed feed containing corn-derived ingredients. The major trichothecenes identified in the samples tested were DON, 3AcDON and 15AcDON. In particular, DON, 3AcDON and 15AcDON were detected in most DDGS and corn gluten feed samples. Most samples of mixed feed contained DON and 15AcDON, but only one mixed feed sample contained 3AcDON. In contrast, corn gluten meal was contaminated with lower levels of these compounds than the other samples tested. Among 36 corn gluten meal samples, DON was detected in 24 and 15AcDON was detected in 20 samples. 3AcDON was not detected in any of the corn gluten meal samples.

Investigation of the suitability of immunochemical-based test kits for quantitative analysis of deoxynivalenol in corn-derived feed and feed ingredients.

We examined whether immunochemical-based test kits designed for quantitative analysis of deoxynivalenol (DON) screening in grain crops are applicable to corn processing by-products. Commercially available test kits (two types of immunochromatographic kits and three types of ELISA kits) were used to assay three types of corn processing by-products and mixed feed. The results obtained with some kits were significantly different from those of LC-MS analysis. Since the differences might be caused by insufficient extraction of DON from samples, the extraction time of all kits was set to be 20 minutes, based on a study of the dependence of the amount of DON extracted on the shaking time. Moreover, the extract of corn processing by-products was acidic, resulting in inhibition of the antigen-antibody reaction, so neutralization and centrifugation processes were introduced to prevent denaturation of antibody. After these modifications, the recovery for all kits in assays of corn gluten meal was within the range of 80-120%, and all kits showed acceptable accuracy. The relative standard deviation (RSD) of repeatability tests for all kits was less than 11.3% for analyses of both corn processing by-products and mixed feeds, indicating good precision. The above results showed that the kits studied were applicable to the quantitative assay of DON in corn processing by-products and mixed feed after modifications as described in this paper.

[Analysis of heterocyclic amines in food by liquid chromatography-tandem mass spectrometry].

A analytical method for simultaneous determination of 10 heterocyclic amines (HCAs) applicable to prepared foods on the market was studied. HCAs were extracted with acidic methanol, and then purified on a diatomaceous column and an ion-exchange column prior to LC-MS/MS. The method was validated within laboratory using three groups among the total diet samples (oils and fats, fish and shellfish, meat and eggs). The method showed good precision and trueness (as recovery) in duplicate analyses over 5 days, though there were some unsatisfactory results. Limits of detection (LOD) and quantification (LOQ) of the method were estimated from the deviation of the analytical results in samples spiked at a level of near 1 ng/g. In addition,13 groups of total diet samples, 27 items of retail food ready to eat and a few foods cooked in the laboratory were analyzed using this validated method. The results showed that the method is applicable to the foods tested in this study and provided information on the content of HCAs in some foods in Japan.

Interlaboratory study of LC-UV and LC-MS methods for the simultaneous determination of deoxynivalenol and nivalenol in wheat.

To evaluate LC methods with UV or MS detection for simultaneous analysis of deoxynivalenol (DON) and nivalenol (NIV) in wheat, an interlaboratory study was conducted in 11 laboratories. DON and NIV were purified using a multifunctional column, and their concentrations were determined using LC-UV or LC-MS(/MS). No internal standards were used. Three fortified wheat samples (0.1, 0.5 and 1 mg/kg), one naturally contaminated wheat sample, and one blank wheat sample were used. The recoveries ranged from 90% to 110% for DON and from 76% to 83% for NIV. For DON, the relative standard deviations for repeatability (RSDr) ranged from 1.1% to 7.6%. The relative standard deviations for reproducibility (RSDr) ranged from 7.2% to 25.2%. For NIV, the RSDr ranged from 2.0% to 10.7%, and the RSDr ranged from 7.0% to 31.4%. Regardless of sample and detector, the HorRat values for DON and NIV ranged from 0.4 to 1.4. Both LC-UV and LC-MS(/MS) methods were considered to be suitable for application as an official method.

[A sensitive analytical method for six aflatoxins in rainbow trout muscle and liver].

A highly sensitive analysis method for six aflatoxins (aflatoxin B1, B2, G1, G2, M1 and aflatoxicol) in rainbow trout muscle and liver was developed. Aflatoxins (AFs) were extracted with acetonitrile-water (9 : 1), purified on an immunoaffinity column, and subjected to HPLC with fluorescence detection after post-column photochemical derivatization. The recoveries of AFs at 0.05 µg/kg spiking levels were 71.4-82.4% in muscle and 80.1-93.0% in liver, and the repeatability relative standard deviations (RSDr) were 0.87-4.6% in muscle and 2.0-6.2% in liver. Limits of quantitation (LOQs) and limits of detection(LODs)of AFs were estimated to be 0.004-0.029 µg/kg, and 0.002-0.012 µg/kg, respectively.

Uptake by dietary exposure and elimination of aflatoxins in muscle and liver of rainbow trout (Oncorhynchus mykiss).

Uptake and elimination of aflatoxins (AFs) by rainbow trout ( Oncorhynchus mykiss ) during a long-term (21 days) dietary exposure were studied to assess contamination by AFs in aquaculture fish fed AF-containing feed. The uptake factor (UF) of aflatoxin B(1) (AFB(1)) in muscle ranged from 0.40 × 10(-3) to 1.30 × 10(-3). AFB(1) concentrations in liver were 165-342 times higher than in muscle. AFs from feed were more highly accumulated in liver than in muscle. Aflatoxicol (AFL) and aflatoxin M(1) (AFM(1)) were detected in muscle and liver and also in the rearing water. AFL concentrations were higher than AFM(1) by 2 orders of magnitude in muscle, and AFL was a major metabolite of AFB(1). The elimination rate constants (α) of AFB(1) and AFL in muscle (1.83 and 2.02 day(-1), respectively) and liver (1.38 and 2.41 day(-1), respectively) were very large. The elimination half-life (t(1/2)) of AFB(1) was 0.38 days (9.12 h) in muscle and 0.50 days (12.00 h) in liver. The elimination half-life of AFL in muscle and liver was 0.34 day (8.16 h) and 0.29 day (6.96 h), respectively. These data show that AFs are eliminated rapidly and are not biomagnified in fish. Thus, AFB(1) concentration in muscle of fish fed AFB(1)-containing feed (ca. 500 µg/kg) decreased to below the detection limit (20 ng/kg) of the most sensitive analytical method at 1.54 days (36.96 h) after the change to uncontaminated feed.