Abstracts / Toxicology Letters 196S (2010) S37–S351 S331

ane:ethyl acetate:acetone (1:5:4, 1:4:5, 1:3:6), respectively. Thenthe combined solutions was evaporated under nitrogen and theresidue was dissolved in 1 ml of buffer and analyzed by HPLC usingcolumn (250 mm × 4.6 mm ID, 5 �m C18 stationary phase) with aUV detector at 276 nm, mobile phase of acetonitrile–water (1:10,v/v) and a flow rate of 1 ml/min. The results showed that all investi-gated apple juice samples were contaminated with patulin at levelsranging from 6.56 to 106.01 ppb. The mean concentration of pat-ulin in samples was 26.926 ppb. 13.3% of the apple juice sampleshad patulin above the maximum tolerable limit (50 ppb) for applejuice in Iran.

doi:10.1016/j.toxlet.2010.03.1046

P309-013Safety assessment of almond skins

Y. Song, X. Jia, X. Zhang, N. Li

Institute of Nutrition and Food Safety, Chinese Center for DiseaseControl and Prevention, China

Almond skins are rich in fibre and a wide variety of flavonoid, andhave been suggested to have some potential benefits. To study thegenotoxic potential of almond skins, a bacterial reverse mutationassay, two in vivo animal tests including micronucleus test andmammalian bone marrow chromosome aberration test were con-ducted. To investigate its subchronic toxicity, a 90-feeding studywas conducted in rats. In the genotoxic tests, almond skins exertedno mutagenic activity in various bacterial strains of Salmonellatyphimurium either in the absence or presence of metabolic acti-vation at all doses. Various doses of almond skins did not affectthe proportions of immature to total erythrocytes, the numberof micronuclei in the immature erythrocytes, and the number ofstructural and numerical chromosomal aberrations of Swiss albinomice. The results demonstrate that almond skins are not geno-toxic under the conditions of the in vitro bacterial reverse mutationassay and two in vivo tests-micronucleus test and mammalianbone marrow chromosome aberration test. In the 90-feeding study,Sprague–Dawley rats were randomly divided into four groups(20 rats/sex/group) and received a diet containing 0%, 2.5%, 5.0%and 10% (w/w) almond skins. Daily clinical observations and weeklymeasurement of body weights and food consumption were con-ducted. Blood samples were obtained for the measurement ofhematology, coagulation and clinical chemistry parameters. Urinesamples were collected for urinalysis. Histological examinationwas performed on all tissues in the control and high-dose groups.No significant changes of mortality, body weight, hematology,coagulation, urinalysis parameters, macroscopic or microscopicexaminations were observed. Differences between treated andcontrol groups in weight gain, food consumption, clinical chem-istry, and organ weight were not considered treatment-related. Theno-observed-adverse-effect-level (NOAEL) for almond skins wasconsidered to be 10% (w/w) for both genders (females, 9.7 g/kg bodyweight/day; males, 8.2 g/kg body weight/day).

doi:10.1016/j.toxlet.2010.03.1047

P309-014Effects of selenium supplementation on lipid peroxidationinduced by thermally oxidized sunflower oil

A. Rajaeefard, N. Tabatabaei, J. Jamalian, R. Ramezani, A.A. Owji,N. Karbalaie

Shiraz School of Health and Nutrition, Iran

The present study compared the effects of four isocaloric dietson rat serum malondialdehyde (MDA) concentrations, liver glu-tathione peroxidase (GPx) activity and serum and liver seleniumcontents.

Material and methods: Male Sprague Dawley rats were orallyadministered 4 different isocaloric diets for 43 days as follow:(1) fresh sunflower oil not supplemented with selenium (Fresh),(2) oxidized sunflower oil not supplemented with selenium (oxi-dized), (3) fresh sunflower oil supplemented with 1 ppm seleniumas sodium selenite (Fresh + Se), (4) oxidized sunflower oil supple-mented with 1 ppm selenium as sodium selenite (Oxidized + Se).The oxidized oil used was prepared by heating fresh sunflower oilat 180 ◦C for 48 h. Results: Serum and liver selenium contents andliver GPx activity were significantly higher in the selenium sup-plemented groups compared to the non-selenium supplementedgroups, but these parameters did not differ significantly betweenthe oxidized oil fed groups and the fresh oil fed groups. SerumMDA concentrations increased significantly in the Oxidized groupcompared to the fresh group. This suggests that the ingestion ofoxidized oil resulted in, in vivo lipid peroxidation. Serum MDA con-centrations remained significantly higher even in comparison of theOxidized + Se group with the Oxidized group.

Conclusions: Our results emphasize that the consumption ofoxidized oil increases in vivo lipid peroxidation and thus can bedeleterious to health. However, we did not observe a significantbeneficial effect of selenium supplementation upon the ingestionof thermally oxidized oil on lipid peroxidation.

doi:10.1016/j.toxlet.2010.03.1048

P309-015Transcellular transport of domoic acid in Caco-2 cellmonolayers

T. Endo 1, O. Kimura 1, Y. Kotaki 2, N. Hamaue 1, K. Haraguchi 3

1 Health Sciences University of Hokkaido, Japan, 2 Kitasato University,Japan, 3 Daiich College of Pharmaceutical Sciences, Japan

Domoic acid is a tricarboxylic amino acid which contains glu-tamic acid moiety, and causes deficit in short-term memory bythe binding at the glutamate receptor as an agonist of glutamicacid. We investigated the intestinal absorption of domoic acidusing a human colon adenocarcinoma cell line Caco-2. Caco-2 cellmonolayers cultured on the permeable membranes were incubatedwith 100 �M domoic acid from either the apical or basolateralside, and the transcellular transport of domoic acid were mea-sured. The transcellular transport of domoic acid from the apicalto basolateral side was about two times higher than that in theopposite direction. The transcellular transport of domoic acidfrom the apical side was temperature-dependent, but it was pH-and sodium-independent. Coincubation of domoic acid with 4,4′-diisothiocyanatostilbene-2,2′-disulfonic acid (DIDS), which is aninhibitor of anion exchangers, significantly decreased the apical-to-basolateral transport of domoic acid by 48%. Coincubation withprobenecid (a non-specific inhibitor of anion transport) and mono-

S332 Abstracts / Toxicology Letters 196S (2010) S37–S351

carboxylic acids such as benzoic acid and acetic acid significantlydecreased the apical-to-basolateral transport by 35, 29 and 32%,respectively. In contrast, coincubation with p-aminohippuric acid(a typical substrate for organic anion transporters), tetraethy-lammonium (a typical substrate for organic cation transporters),succinic acid (a dicarboxylic acid), citric acid (a tricarboxylic acid)and glutamic acid decreased slightly or did not decrease the apical-to-basolateral transport of domoic acid. These results suggestedthat the apical-to-basolateral transport of domoic acid across theCaco-2 cell monolayers is mediated by the DIDS-sensitive aniontransporters.

doi:10.1016/j.toxlet.2010.03.1049

P309-016Application of the threshold of toxicological concern-conceptin safety assessment of chemically complex food matrices

M. Rennen, S. Koster, C. Krul, L. Krul, G. Houben

TNO Quality of Life, The Netherlands

Toxicological safety assessment is usually performed as a sequen-tial process (hazard identification–hazard assessment–exposureassessment–risk assessment), which is often time consuming,expensive and uses many animals. To obtain a more efficient pro-cess new concepts would be needed that integrate knowledgeand information from different disciplines (i.e. chemical analysis,toxicological research, exposure and risk assessment) right fromthe start throughout the whole research process to fine-tune therequired level of detail. The Threshold of Toxicological Concern(TTC) is potentially such a new concept.

The TTC concept was developed to assess substances in foodwith known identity but lacking toxicological information. Appli-cation of TTC to chemically complex food matrices (CCFM) suchas migrants from packaging (especially the non intentionallyadded substances), natural fragrances and flavourings (e.g. smokeflavours) and processed foods (e.g. presence of acrylamide andfurans) is limited due to a lack of chemical identity.

We drafted a framework to enable application of TTC to suchCCFM. The approach is based on exclusion of specific groups ofcompounds following the Kroes et al. TTC decision tree (Food andChemical Toxicology 42 (2004) 65–83) and modifications as pro-posed by Munro et al. (Toxicology Letters 180 (2008) 151–156). Weconcluded that the highest feasible toxicological threshold whichcan be applied for unknowns after exclusion of specific groupsof hazardous chemicals is 540 �g/person/day. To determine theamount of substances above a certain threshold a conversion ofresponse in a chromatogram (visualized as ‘forest of peaks’ inchromatograph) into concentrations and subsequently into intakeis needed. Those substances which appear above the applicablethreshold should be identified, characterized and assessed for theirtoxicological safety.

The safety assessment framework for CCFM and required ana-lytical and test strategy innovations (such as genotoxicity screeningmethods) that are under development will be presented.

doi:10.1016/j.toxlet.2010.03.1050

P309-017Determination of furan levels in selected foods in Brazil

A. Arisseto, E. Vicente, M. Ueno, M.C. Toledo

Institute of Food Technology, USA

Detection of high amounts of furan in thermally treated foods byAmerican researchers in 2004 raised for the first time a concernon the potential risks of this contaminant to human health. Furanhas been found to be carcinogenic in experimental animals andis classified as a possible human carcinogen by the IARC. Sev-eral countries, especially from Europe and North America, havereported the occurrence of furan in commonly consumed foods,including canned and jarred foods, cereal products and coffee, inlevels up to 6500 �g/kg. The objective of this study was to con-duct a preliminary survey in Brazil on furan levels in foods. Forthat, 43 samples (36 of canned/jarred foods, 3 of roasted cof-fee, 3 of soy sauce and 1 of bread) were analyzed by using anin-house validated method based on gas chromatography cou-pled to mass spectrometry preceded by headspace solid phasemicroextraction (HS-SPME-GC/MS). The limits of detection (LOD)and quantitation (LOQ) were 0.7 and 2.4 �g/kg, respectively. Thelevels of furan in samples varied from not detected to 5021 �g/kg.Furan was found in quantifiable amounts in 24 samples and thehighest concentrations were observed in roasted coffee powder(1946–5021 �g/kg) and soy sauce (13–50 �g/kg). The high levelsfound in coffee are probably due to the roasting process where thetemperatures exceed most other food-processing procedures. Theresults obtained in the present study are comparable with thosereported in other countries. These data are only a first overview ofthe distribution of furan in foods in Brazil, and should be refinedin the future by examining a larger number of samples and foodgroups.

FAPESP (Proc. 2008/50095-0) and CNPq (Proc. 474267/2008-3and Proc. 578381/2008-7).

doi:10.1016/j.toxlet.2010.03.1051

P309-018Safety assessment of the Cordyceps Militaris Fruitbody

J. Huang

Centers for Disease Control and Prevention of Guangdong Province,China

Cordyceps Sinensis, a valued traditional Chinese medicine, is anendangered species. Cordyceps Militaris is similar to it in thecomponents, Cordyceps Militaris Fruitbody is artificial culturedusing Paecilomyces militaris extracted from Cordyceps Militaris.For being rich of cordycepin, Cordyceps Militaris attracted specialinterests. Cordyceps Militaris was reported to possess wide rangeof bio-functions, such as enhancing body’s immunity, anti-aging,anti-fatigue, anti-cancer, anti-virus and hypoglycemic effects, andcordycepin was also conformed as one of anticancer drugs belong-ing to antitumor antibiotics family by National Cancer Institute ofthe USA. Cordyceps Militaris Fruitbody, as a new health food, areused by more and more people in the East Asia. However, littleis known about the safety of Cordyceps Militaris Fruitbody. Theaims of the present study are to assess the genotoxicity and sub-chronic toxicity of Cordyceps Militaris Fruitbody. The genotoxicityof Cordyceps Militaris Fruitbody were evaluated by micronucleustest of mice, in vitro mammalian cell chromosome aberration testand Ames test. Compared to the solvent control, no obvious effects