Liquid Chromatographic Determination of Ochratoxin A in Barley, Wheat Bran, and Rye by the AOAC/IUPAC/NMKL Method: NMKL Collaborative Study1

In an inter laboratory study of the AOAC/lnternational Union of Pure and Applied Chemistry (IUPAC)/ Nordic Committee on Food Analysis (NMKL) liquid chromatographic method for ochratoxin A (OTA), 3 spiked samples each of wheat bran and rye and 3 naturally contaminated barley samples with expected OTA contents between 2 and 9 μg/kg were analyzed by 12 European laboratories. Mean recoveries ranged from 64 to 72%. Within-laboratory standard deviations were 21,17, and 12% for wheat bran, rye, and barley, respectively. Corresponding between-laboratory standard deviations were 23-28,20-28, and 18-31%. Horwitz ratio values ranged from 0.50 to 0.77. The method was accepted as an official NMKL method for determining OTA at ≥2 μg/kg in cereals and cereal products.

The possible involvement of 3-methoxy-2(5H)-furanone in the etiology of Narthecium asiaticum maxim. associated nephrotoxicity in cattle

Two samples of Narthecium asiaticum Maxim leaves collected in Japan were found to contain 103 microg(-1) and 160 microg g(-1) dry matter of 3-methoxy-2(5H)-furanone respectively. 3-Methoxy-2(5h)-furanone was suggested to be the toxic principle of N. asiaticum causing nephrotoxicity in cattle in Japan. Two other furanones, which are thought to be non-toxic, were also isolated from the two samples. These were 4-methoxy-2(5H)-furanone and 5-hydroxy-4-methoxy-2(5H)-furanone.

The occurrence of culmorin and hydroxy-culmorins in cereals

Forty-five samples from 1988-1995 of naturally contaminated grain, barley, wheat and oats, three samples of mixed feed, and 16 samples of grain artificially inoculated with Fusarium culmorum during the flowering stage were analysed for deoxynivalenol (DON), 3-acetyldeoxynivalenol (3-acetyl-DON), culmorin and hydroxy-culmorins. These compounds are secondary metabolites produced by the fungal species F. culmorum and F. graminearum. Acetonitrile-water extract of the samples was purified on a Mycosep #225 column, derivetized using penta-fluoropropionic anhydride (PFPA) and analysed by gas chromatography-mass spectrometry (GC-MS). The amount of each of culmorin, 5-, 12-, 14 and 15-hydroxy-culmorin and one unknown hydroxy-culmorin were determined relative to the amount of DON plus 3-acetyl DON for each sample. The ratio between the total amount of culmorin compounds and the DON compounds ranged from 0.14 to 1.07 in the samples. This study shows that there is a strong correlation between the amount of DON present in the grain and the amount of culmorin and hydroxy-culmorins present. The ratio of each of the culmorin compounds relative to the amount of DON compounds were in the same range in the grain artificially inoculated by F. culmorum as found in an earlier study for F. culmorum strains cultivated on rice, while the hydroxy-culmorin profile in the naturally contaminated grain was more similar to what was found for the F. graminearum cultures in the same study. These results indicate that F. graminearum may be a relatively important source for DON in grain also in relatively cold areas.

Production of culmorin compounds and other secondary metabolites by Fusarium culmorum and F. graminearum strains isolated from Norwegian cereals

Twenty-three Fusarium culmorum and 21 F. graminearum isolates were studied for their ability to produce mycotoxins and other secondary metabolites. The strains were cultivated on rice, and the extracts analysed by gas chromatography mass spectrometry (GC-MS) after derivatization with pentafluoropropionic (PFP) reagent. Two F. culmorum strains formed nivalenol and its acetylated derivatives (chemotype II), while all F. graminearum and the other F. culmorum isolates produced deoxynivalenol (DON) via 3-acetyldeoxynivalenol (3-acetyl-DON) (chemotype IA). 15-hydroxy-culmorin, followed by 5-hydroxy-culmorin were the main other metabolites produced F. culmorum, while 5-, 12- and an unidentified hydroxy-culmorin, suggested to be 14-hydroxy-culmorin, were the main metabolites of F. graminearum. The hydroxy-culmorin profile was found to be significantly different for the two Fusarium species. Minor amounts of about ten other hydroxy-culmorins, four hydroxy-culmorones and 3,13-dihydroxy- epiapotrichothecene were also detected in most cultures. Traces of sambucinol seemed to be present in some of the isolates, but were not detected in any significant amounts. The precursors in the biosynthetic sequence to 3-acetyldeoxynivalenol, 7,8-dihydroxycalonectrin and 15-deacetyl-7,8-dihydroxycalonectrin, were detected in most cultures. We also report the assignment of both the 1H and 13C NMR data of 15-deacetyl-7,8-dihydroxycalonectrin, which has only been reported incorrectly before.

Placental transfer of the estrogenic mycotoxin zearalenone in rats

In order to study the possible placental transfer of the Fusarium mycotoxin zearalenone (ZON), Sprague Dawley rats were treated with a single dose (0.74 mg/kg b.w.) of ZON i.v. on day 12 or day 18 of pregnancy, or intragastrically (i.g.) on day 18 of pregnancy. Samples of placenta, foetus, and maternal liver and spleen were collected for chemical analyses 0.3 h after treatment on day 12, and 0.3, 4, and 24 h after treatment on day 18. Three rats were used for each pregnancy day, administration route, and exposure time. The concentrations of ZON and its metabolites alpha- and beta-zearalenol (-ZOL) were determined quantitatively by high-performance liquid chromatography (HPLC) after incubation with beta-glucuronidase and purification on immunoaffinity columns. Tissue distribution was studied by means of whole body autoradiography at 4 and 24 h after treatment with tritiated ZON (750 microCi/kg b.w; 7.4 mg/kg b.w.) on day 18 of pregnancy. ZON and alpha-ZOL were transferred into the foetus on both gestational days. However, a delay in distribution into the foetus, relative to the maternal tissue, was observed. Beta-ZOL was below the detection limit in the foetus. No specific site of foetal accumulation of ZON or its metabolites was apparent. In the maternal tissues, the highest levels of ZON and of alpha- and beta-ZOL were found in the liver.

The occurrence of HT-2 toxin and other trichothecenes in Norwegian cereals

A total of 449 grain samples, 102 barley, 169 wheat and 178 oat samples were collected from different regions of Norway from 1996-1998 crops, mainly from grain loads and silos. The samples were analysed for type A and B trichothecenes, the largest groups of mycotoxins produced by the Fusarium species, by gas chromatography with mass spectrometric detection (GC-MS). Factors affecting the presence of the different trichothecenes are discussed. Deoxynivalenol (DON) and HT-2 toxin were the trichothecenes most frequently detected, followed by T-2 toxin, nivalenol, and scirpentriol, scirpentriol being detected only in seven samples (> 20 micrograms/kg). Oats were the grain species most heavily contaminated with an incidence (% > 20 micrograms/kg) and mean concentration of positive samples of 70% (115 micrograms/kg) for HT-2 toxin, 30% (60 micrograms/kg) for T-2 toxin, 57% (104 micrograms/kg) for DON, and 10% (56 micrograms/kg) for nivalenol. The corresponding values for barley were 22% (73 micrograms/kg), 5% (85 micrograms/kg), 17% (155 micrograms/kg) and 6% (30 micrograms/kg), and for wheat 1.2% (20 micrograms/kg), 0.6% (20 micrograms/kg), 14% (53 micrograms/kg) and 0% for HT-2, T-2, DON and nivalenol, respectively. Norwegian oats were found to contain HT-2 and T-2 toxin in concentrations that might be at threat to human health for high consumers of oats. The amount of DON was significantly lower than in the crop from previous years.

Production of T-2 toxin by a Fusarium resembling Fusarium poae

A Fusarium species with a micro morphology similar to F. poae and a metabolite profile resembling that of F. sporotrichioides has been identified. Like typical F. poae, the microconidia have a globose to pyriform shape, but the powdery appearance, especially on Czapek-Dox Iprodione Dichloran agar (CZID), less aerial mycelium and the lack of fruity odour on Potato Sucrose Agar (PSA) make it different from F. poae. The lack of macroconidia, polyphialides and chlamydospores differentiates it from F. sporotrichioides. All 18 isolates investigated, 15 Norwegian, two Austrian and one Dutch, produced T-2 toxin (25-400 micrograms/g) on PSA or Yeast Extract Sucrose agar (YES). In addition, neosolaniol, iso-neosolaniol, HT-2 toxin, 4- and 15-acetyl T-2 tetraol, T-2 triol and T-2 tetraol and 4,15-diacetoxyscirpenol were formed in variable amounts. Neither nivalenol, 4- or 15-acetylnivalenol or 4,15-diacetylnivalenol were detected in any of the cultures, while these toxins were produced at least in small amounts by all the 12 typical F. poae isolates studied. The question of whether this Fusarium should be classified as F. poae or F. sporotrichioides or a separate taxon should be addressed.

Isolation and characterization of 3-methoxy-2(5H)-furanone as the principal nephrotoxin from Narthecium ossifragum (L.) Huds

The principal substance in Narthecium ossifragum (L.) Huds, responsible for the nephrotoxic effects on cattle, moose, goats and other ruminants has been isolated and identified by X-ray crystallography as 3-methoxy-2(5H)-furanone. The Fourier-transform infra-red, 1H and 13C nuclear magnetic resonance, and mass spectra are also given. The concentration in four different batches of plant material varied from 113 to 344 microg g(-1) (wet weight). Extracts of N. ossifragum and fractions derived from them, including purified 3-methoxy-2(5H)-furanone, were each dosed intraruminally, to young goats. 3-Methoxy-2(5H)-furanone of 99.9% purity (15 mg kg(-1) live weight) caused increased concentration of creatinine in serum within 2-3 days, typical of kidney damage caused by N. ossifragum, while toxic effect was obtained down to 4 mg kg(-1) live weight with less purified material (> or = 95%). Toxic effect was also obtained with synthesized 3-methoxy-2(5H)-furanone (30 mg kg(-1) live weight). The isomer 4-methoxy-2(5H)-furanone, detected in some of the batches of the plant material, was not toxic when dosed at 60 mg kg(-1) live weight.

Mycotoxin production and cytotoxicity of Fusarium strains isolated from Norwegian cereals

Thirty-four isolates of the eight most common Fusarium species isolated from Norwegian cereals; F. avenaceum, F. culmorum, F. equiseti, F. graminearum, F. poae, F. sporotrichioides, F. torulosum and F. tricinctum were studied for their cytotoxicity and ability to produce mycotoxins. The strains were cultivated on rice, and analysed for trichothecenes (all species), zearalenone (all species), fusarochromanone (F. equiseti), wortmannin (F. torulosum), moniliformin and enniatins (F. avenaceum, F. tricinctum and F. torulosum). The cytotoxicity of the extracts were examined with an (in vitro) MTT-cell culture assay. All F. graminearum and five of seven F. culmorum isolates belonged to chemotype IA, producing deoxynivalenol and 3-acetyl-deoxynivalenol, while the two other F. culmorum strains were nivalenol producers (chemotype II). The F. equiseti isolates and one of the F. poae isolates produced both type A and B trichothecenes, and relatively large quantities of fusarochromanone were detected in the F. equiseti cultures. All Fusarium species studied showed significant cytotoxicity, but with a large variation between species, and also within each species. F. sporotrichioides and F. equiseti showed the highest average cytotoxicity.

Ammoniated forage poisoning: concentrations of alkylimidazoles in ammoniated forage and in milk, plasma and urine in sheep and cow

4-methylimidazolc (4-MeI) has until now been the only identified toxic compound in ammoniated forage considered to be of possible etiological significance in ammoniated forage poisoning. However, several authors have concluded that 4-MeI alone cannot explain the toxicity observed (Morgan & Edwards 1986, Nielsen et al 1986, Motoi et at. 1997). On this background, we have examined samples of ammoniated forage and of milk, plasma, and urine collected from ewes and lambs during a previous experimental poisoning study (Sivertsen et al. 1993). We have also studied similar samples from a Norwegian Red Cattle dairy cow fed ammoniated hay for the first 5 days after calving. We have, in addition to the previously known ingredients 4-MeI and 2-methylimida-zole (2-MeI), identified 5 new compounds in these samples, all of which were found to be di-alkylsubstituted imidazoles: 1,2-dimcthylimi-dazole (1,2-diMeI), 1,4-dimethyl imidazole (1,4-diMeI), 1,5-dimethyl imidazole (1,5-diMeI), 2,4-dimethylimidazole (2,4-diMeI) and 2-ethyl, 4-methylimidazole (2Et-4MeI) (Müller et al. 1998). In this paper we present quantitative measurements of these di- and monoalky 1 imidazoles in a selection of the collected samples.

Effects of diets with graded levels of naturally deoxynivalenol-contaminated oats on immune response in growing pigs

A trial was conducted to evaluate the effect of including different levels of deoxynivalenol (DON)-contaminated oats in the complete diets of growing pigs on immune response and performance. The diets contained 0.6, 1.8 and 4.7 mg DON/kg, and both restricted and ad libitum feeding were used. Performance was recorded as weight gain, feed intake, efficiency of feed utilization and carcass quality. Immune response parameters recorded included primary and secondary antibody titres after injections of five different antigens: Human serum albumin (HSA), sheep red blood cells (SRBC), paratuberculosis vaccine (MPT), tetanus toxoid (TT) and diphteria toxoid (DT). A johnin test was also performed. Lymphocyte stimulation response was measured with three different mitogens (PWM, ConA and PHA). A significant, DON dose-dependent reduction in secondary antibody response to tetanus toxoid was observed. A slightly higher mitogen response after PHA stimulation in lymphocytes from the medium and high DON groups compared to the low DON group after 9 weeks was considered inconclusive. No other indication of dose-dependent immune response inhibition or stimulation was found. Significantly reduced feed intake with increased levels of DON was observed in groups fed restricted rations according to weight, but not in animals fed ad libitum.

Nephrotoxicity in goats caused by dosing with a water extract from the stems of Narthecium ossifragum plants

Seven goats were given a single dose of an aqueous extract derived from 30 g (wet weight) of Narthecium ossifragum per kg liveweight. Their serum creatinine and urea concentrations increased to day 5 but then fell to normal by day 10. Serum magnesium increased to day 4 and decreased to normal by day 9. Their serum calcium concentration was lower than normal on days 4, 5 and 6. Histopathological examination of the kidneys of goats killed or found dead 2, 4, 6, 8, 11 or 16 days after dosing revealed tubular epithelial cell degeneration and necrosis. Regeneration of the tubular epithelium and signs of interstitial fibroplast proliferation and fibrosis could be seen in animals killed on days 8, 11, 16 and 42. No signs of liver damage were observed in 3 goats dosed with the insoluble plant material from 40 g (wet weight) Narthecium ossifragum per kg liveweight. The total dose was divided into three doses, which were given intraruminally within 7 h. The activities of aspartate aminotransferase, gamma-glutamyltransferase and glutamate dehydrogenase remained within the normal range in all 10 goats after dosing.

Further studies on the presence, qualities and effects of the toxic principles from Narthecium ossifragum plants

One calf was dosed during one day with an aqueous extract from 3.0 kg (wet weight) of Narthecium ossifragum and another was dosed on the same day with the insoluble plant residue. The concentrations of serum creatinine and magnesium increased only in the calf dosed with the aqueous extract, while the activity of glutamate dehydrogenase increased only in the serum of the calf dosed with the plant residue, so differentiating the nephrotoxic and hepatotoxic principles as water-soluble and water-insoluble compounds, respectively. One calf was dosed with 30 g (wet weight) N. ossifragum flower stems per kg live weight during one day and another was dosed with 30 g (wet weight) N. ossifragum leaves per kg live weight on the same day. The serum creatinine and urea concentrations and also the activities of glutamate dehydrogenase, aspartate aminotransferase and gamma-glutamyltransferase in the serum increased in the calf dosed with the flower stems, whereas there was only a slight temporary increase in the creatinine concentration in serum from the calf dosed with the leaves. However, histopathological examination of the kidneys of the calf dosed with the flower stems revealed severe tubular necrosis and degeneration. It therefore appears that both the toxic principles are present in the flower stems of N. ossifragum rather than in its leaves. The serum creatinine concentration was significantly increased in a non-ruminating calf dosed with an aqueous extract from 32 g (wet weight) N. ossifragum per kg liveweight during one day, showing the intrinsic nephrotoxicity of the plant.

Determination of aluminium in liver from reindeer, moose and sheep by electrothermal atomic absorption spectrometry

Aluminium in liver from reindeer, moose and sheep from the northeast part of Norway was determined by graphite furnace atomic absorption spectrometry following digestion of the samples with nitric acid. The concentration of aluminium in the liver was markedly higher for reindeer than for moose and sheep; the median values obtained were 0.56 microgram g-1 Al (wet wt.) for 101 reindeer, 0.06 microgram g-1 Al for 72 moose and 0.09 microgram g-1 Al for 40 sheep. The detection limit of the method was 0.01 microgram g-1 Al. The NIST SRM 1577a Bovine Liver was also analyzed.

The effects of naturally deoxynivalenol-contaminated oats on the clinical condition, blood parameters, performance and carcass composition of growing pigs

A feeding trial with naturally deoxynivalenol (DON)-contaminated oats included in feed mixtures at graded levels was conducted in growing pigs. The DON concentrations were 0, 0.7, 1.7, and 3.5 mg/kg of complete feed mixture given ad libitum to different groups. The data recorded were feed consumption, body weight gain, slaughter weight, biochemical and haematological data including serum immunoglobulin A, clinical condition and post-mortem pathology including histopathology. Significantly decreasing body weight gain throughout the experimental period, decreased slaughter weight and reduced feed utilization efficiency were observed for the group fed a diet containing 3.5 mg/kg of DON. At the same DON concentration, there were increased liver weights and decreased concentrations of serum protein and albumin, and a temporary fall in packed blood cell volume, serum calcium and serum phosphorus. For the groups fed diets containing 1.7 and 3.5 mg/kg of DON, a statistically significant, dose-related decrease in daily feed consumption was observed. No other effects on haematological, biochemical or immunological parameters were recorded. The carcass quality was not affected in any group. It was concluded that significant effects in growing pigs may be observed at a dietary DON concentration of 1.7 mg/kg, originating from naturally contaminated oats included in a diet that was otherwise adequate and contained only minor traces of other mycotoxins.

Further arguments against 4-methylimidazole as causal factor in ammoniated forage toxicosis: experimental seed-hay poisoning in young lambs

4-methylimidazole (4-MeI) is up to now the main tremorogenic poison known to be present in ammoniated roughages. The concentration of 4-MeI is widely used as an indicator of the potential toxicity of individual batches of ammoniated straw and hay.

Ochratoxin A in plasma of Norwegian swine determined by an HPLC column-switching method

A method well suited for screening large numbers of plasma samples for ochratoxin A is presented. Proteins were precipitated with methanol and the supernatant diluted with 0.01 M phosphoric acid before 1 ml extract was injected into a high-performance liquid chromatograph (HPLC). The extract was further cleaned up and pre-concentrated on a polystyrene-divinylbenzene precolumn. After column-switching, the sample was chromatographed on a C18 analytical column, and ochratoxin A was detected with a fluorescence spectrophotometer, either directly or after postcolumn pH shift. The detection limit was 0.10 ng ochratoxin A/ml plasma. The method was used to determine the ochratoxin A concentration in 216 samples of swine plasma. They were collected from different herds in June 1991 from ten slaughterhouses, located in different parts of Norway. Eighty-two percent of the samples contained > or = 0.10 ng ochratoxin A/ml plasma while 0.9% contained > or = 5.0 ng/ml.

Automation of a clean-up procedure for determination of trichothecenes in cereals using a charcoal-alumina column

Automation of the clean-up procedure for trichothecenes on a charcoal-alumina column is described. Standard high-performance liquid chromatographic equipment was used for the clean-up step. An acetonitrile-water (84 + 16, v/v) extract of the sample was cleaned up on a column packed with charcoal-alumina-Celite, which was washed with acetonitrile between each sample. The eluates were collected directly in reaction vials and evaporated to dryness. The residual water was removed azeotropically with benzene. The sample was derivatized with 1-(trimethylsilyl)imidazole and analysed by capillary gas chromatography with electron-capture detection.