[Modification of HPLC conditions for the determinations of raw materials, intermediates and subsidiary colors in 5 kinds of food azo colors].

Modification of HPLC conditions for the determinations of raw materials, intermediates and subsidiary colors (organic impurities) in 5 kinds of food azo colors were studied in order to analyze them simply and rapidly. The organic impurities were determined by HPLC using L-column ODS and a gradient system (0.02 mol/L ammonium acetate solution for 10 min, and a gradient with a mixture of acetonitrile and water (7:3)). The organic impurities in 163 samples of azo colors subjected to inspection in fiscal year 1999 were determined under the proposed HPLC conditions. It was found that the contents of organic impurities in 163 samples were less than the limit of Japan's Specifications and Standards for Food Additives, 7th Edition.

[Preparation of HPLC test solutions for organic impurities in aluminum lakes of food red no. 40 (allura red AC) and food yellow no. 5 (sunset yellow FCF)].

The HPLC determination of organic impurities in Food Red No. 40 aluminum lakes (R-40Als) as directed by Japan's Specifications and Standards for Food Additives, 7th Ed. (JSFA-VII), has problems, such as reproducibility and low recovery. ICP analyses suggested that the problem was caused by the aluminum in the test solution. In the improved method for preparation of the test solution, aluminum was precipitated as a hydroxide gel by boiling with 1% aqueous ammonia. After centrifugation, the supernatant was used for the HPLC analysis of the organic impurities in the lakes. Recoveries of organic impurities were more than 85% from R-40Al spiked at the 0.1 and 1.0% levels of R-40. The proposed method was also adapted for Food Yellow No. 5 aluminum lakes.

Structural determination of the subsidiary colors in food blue No. 1 (brilliant blue FCF) aluminum lake detected by paper chromatography.

One of eight Food Blue No. 1 aluminum lakes (B-1Als) used in the official inspection of coal-tar colors in fiscal year 1999 had a violet sub-spot during paper chromatography and was rejected. To clarify the orgin of the sub-spot, the violet subsidiary color (Sub-V) was isolated from the sample. On the basis of NMR and MS analyses and ion chromatography, the structure of the subsidiary color was elucidated to be 2-[[4-[N-ethyl-N-(3- sulfophenylmethyl)amino]phenyl][4-hydroxyphenyl]methylio]benzenesulfonic acid. The relative content of Sub-V to that of m,m-B-1 in the rejected sample was determined to be 39.5% by HPLC. The relative contents in other submitted samples of B-1Al were in the range of 1.1-3.6%.

[Estimated production by the official inspection of tar colors (including aluminum lakes) in fiscal year 2000].

There were 176 official inspections of tar colors and their lakes in fiscal year 2000, and 175 samples were qualified. The quantity of tar colors that passed inspection in Japan in fiscal year 2000 reached 137.5 tons. Tar color production is estimated by month and by manufacturer. The food tar color produced in the largest quantity was Food Yellow No. 4, accounting for 43.4% during this period.

High-performance liquid chromatographic determination with photodiode array detection of ellagic acid in fresh and processed fruits.

A high-performance liquid chromatographic (HPLC) procedure based on an isocratic elution with photodiode array detection has been developed for a simple and rapid determination of ellagic acid (EA) in fresh and processed fruits. The homogenized sample was refluxed with methanol and then the extract was refined using a solid-phase cartridge before HPLC. We analyzed EA in 40 kinds of fresh fruits and 11 kinds of processed fruits by the developed method. EA was found in several berries, fueijoa, pineapple and pomegranate. This is the first occurrence of the detection of EA in bayberry, fueijoa and pineapple.

Determination of phenolic acids in fruit juices by isocratic column liquid chromatography.

A simple and rapid analytical method of five phenolic acids, gallic, chlorogenic, caffeic, ellagic and ferulic acid, which are naturally occurring bioactives, were determined in fruit juices by isocratic LC using photodiode array UV detection. The sample was pre-treated by solid-phase extraction (a combination of Sep-Pak Plus tC18 and Bond Elut PSA).

[Studies on procedure for sample preparation of ellagic acid in several kinds of foodstuffs].

A sample preparation for ellagic acid in several kinds of foodstuffs, which is included in the List of Existing Food Additives as natural antioxidants in Japan, functioning as an antioxidant, was studied. The solid samples were refluxed with methanol, and then the extract was refined using a solid-phase cartridge. The liquid samples were directly pretreated by solid-phase extraction. On the other hand, test solutions for cooking oils and fats were inapplicable to solid-phase extraction in this work, because the recovery tests from samples spiked with ellagic acid gave low recoveries. Consequently, they were prepared by modified frozen method using methanol or acetonitrile-2-propanol-ethanol (2:1:1). The recoveries from tested foodstuffs spiked with ellagic acid (50 micrograms/g) were 75.2-96.9%. The limits of quantification for ellagic acid were 0.05 microgram/g by the proposed method.

[Studies on rejected food blue no. 1 (Brilliant Blue FCF) aluminum lake].

One of the brilliant blue FCF aluminum lakes (B-1Als) in the official inspection of coal-tar dyes in fiscal year 1999 was rejected. The results of tests of the rejected sample were submitted to JSFA-VII except for a violet sub-spot by paper chromatography. Its visual spectrum had same lambda max 630 nm as B-1 standard and a very small shoulder at 580 nm in comparing to B-1 standard. HPLCs detected at 625 nm of subsidiary dyes and 580 nm were performed on the rejected sample of B-1Al. A sub-V in HPLCs of th sample which eluted at 37.6 min had its relative peak area of 7.9% at 580 nm and 0.24% at 625 nm, and its peak ratio (at 580 nm/at 625 nm) was extremely large. Consequently, it was presumed that the sub-V in those HPLCs corresponded to the violet sub-spot in paper chromatography.

[Estimated production by the official inspection of coal-tar dyes (including dye aluminum lakes) in fiscal year 1999].

There were 260 official inspections of coal-tar dyes and their lakes in fiscal year 1999, and 259 of the lots qualified, one of the 8 samples of Food Blue No. 1 aluminum lakes was rejected. The quantity of coal-tar dyes that passed inspection in Japan in fiscal year 1999 reached 151.9 tons. Coal-tar dye production is estimated by month and by manufacturer. The food coal-tar dye produced in the largest quantity was Food Yellow No. 4, accounting for 44.2% during this period.

Influence of jam processing on the radical scavenging activity and phenolic content in berries.

Six selected phenolic aglycons (caffeic and ellagic acids, kaempferol, quercetin, myricetin, and morin) in nine types of berries, and their changes as influenced by jam processing, have been evaluated using optimized HPLC with diode-array detection. The berry samples, fresh and after jam processing, were analyzed, and the total amounts of selected phenolics as aglycons were identified and determined by acid hydrolysis. Their contents in fresh and jam samples did not indicate appreciable changes; therefore, the influence of jam processing on these selected phenolics in berries was suggested to be small, and was mostly present in berries as several conjugated forms that were glycosylated, esterified, etc., in the samples. The total phenolic content of each sample was also determined by the Folin-Ciocalteu method. The three samples of each berry, namely fresh, jam, and acid hydrolysate of the berry, had similar total phenolic contents. On the other hand, the scavenging effect on the 1,1-diphenyl-2-picrylhydrazyl (DPPH) radical was measured, and acid hydrolysates showed stronger activity than that of the fresh and jam-processed samples for all of the berry types.

Oligomeric hydrolysable tannins from Tibouchina multiflora.

Two hydrolysable tannins, nobotanin O and nobotanin P, were isolated from the leaf extract of Tibouchina multiflora (Melastomataceae) and their dimeric and tetrameric structures elucidated on the basis of spectral data and chemical correlations with nobotanin B and K, respectively. Thirteen known hydrolysable tannins including nobotanins A, B, C and J, which are oligomers characteristic of the Melastomataceae, were also isolated.

Tannins and related polyphenols of euphorbiaceous plants. XI. Three new hydrolyzable tannins and a polyphenol glucoside from Euphorbia humifusa.

Three new hydrolyzable tannins, euphormisins M1, M2, and M3, were isolated from Euphorbia humifusa WILLD., and respectively characterized as 1,3,6-tri-O-galloyl-4-O- brevifolincarboxyl-beta-D-glucose (19), an oxidative metabolite (23) of geraniin, and 1,3,6-tri-O-galloyl-alpha-D-glucose (18), by spectroscopic and chemical methods. A new ellagic acid glucoside (16) and fifteen known tannins, including geraniin (8) and four dimers [euphorbins A (13), B (14), excoecarianin (15) and eumaculin A (12)], were also isolated.