Natural co-occurrence of fumonisins, deoxynivalenol, zearalenone and aflatoxins in field trial corn in Argentina.

Corn samples collected from the main production area in Argentina in 1995 were surveyed for the natural occurrence of Fusarium mycotoxins and aflatoxins. Fumonisins B1, B2 and B3 and zearalenone were found in all samples. A positive relationship was found between fumonisins B1, B2 and B3, B1 and B3, and B2 and B3. Deoxynivalenol and aflatoxins were not detected. Mycological survey has also revealed the predominance of Fusarium moniliforme. This is the first report on the simultaneous occurrence of fumonisins and zearalenone in corn from the main production area in Argentina.

Photocatalytic degradation of cyanobacterial microcystin toxins in water.

The microcystins are hepatotoxins produced by a number of cyanobacterial species (blue green algae) in fresh water systems. The increasing eutrophication of natural waters has led to an increase in the incidence of algal blooms and the consequent increased risk of microcystin contamination of water resources. The removal of microcystins LR, YR and YA from contaminated water was investigated using an experimental laboratory-scale photocatalytic 'falling film' reactor in which an oxygen purge, UV radiation and semiconductor titanium dioxide (TiO2) catalyst were used to oxidatively decompose the microcystin pollutants. Preliminary studies, using algal extracts spiked into distilled water, indicated that the microcystins were rapidly decomposed in this reactor. The decomposition followed first order reaction kinetics with half-lives of less than 5 min with the reactor operating in a closed-loop mode. Reaction rates were strongly dependent on the amount of TiO2 catalyst (O-5 g/l), but only marginally influenced by a change in gas purge from oxygen to compressed air. The use of lake water, rather than distilled water, showed that this process is feasible in natural waters, although increased levels of catalyst (up to 5 g/l) were required to achieve comparable decomposition rates.

Fumonsin B1, B2, and B3 content of commercial unprocessed maize imported into South Africa from Argentina and the USA during 1992.

The widespread occurrence of F. moniliforme and the toxic effects of its secondary metabolites, the fumonisins B1(FB1), B2(FB2) and B3(FB3), make it imperative that fumonisin contamination of maize, a major constituent of animal feed as well as the staple diet of many populations, be closely monitored to reduce the risk of fumonisin exposure. Equine leukoencephalomalacia and porcine pulmonary oedema have been associated with the intake of feed heavily contaminated with fumonisins. In addition, high levels of fumonisins in the maize-based staple diets of certain populations have been linked to a high incidence of oesophageal cancer in the Transkei region of South Africa and in Linxian and Cixian Counties, China. Bulk shipments of maize imported into South Africa from the USA and Argentina during 1992 were sampled at the port of entry to determine fumonisin levels. Of the 79 samples from two US shipments, all were positive for fumonisins, with FB1 constituting approximately 71% of the total fumonisins with an overall mean of 2.35 micrograms/gFB1. The maximum FB1 level observed was 3.9 micrograms/g. These levels contrast with those obtained from two Argentinian bulk shipments, which also were all positive for fumonisins, but had a mean FB1 level of 0.31 microgram/g and a maximum observed level of 0.7 microgram/g FB1 measured over 47 composite samples.

The influence of deck storage and initial processing on patulin levels in apple juice.

Patulin, a secondary metabolite produced by Penicillium expansum and some other fungal species, is a common contaminant of ripened apples used for the production of apple juice concentrates. The limited availability of suitable storage facilities may result in fruit being subjected to storage in the open ('deck storage') for extended periods of time, prior to processing. A study was conducted to determine the influence that deck storage and subsequent initial processing practices had on patulin levels in freshly pressed juice. Over the study period, triplicate samples were collected at four strategic processing points from individual consignments of Granny Smith apples deck-stored for 7, 15 and 33 days, respectively. Over the study period, mean patulin levels in non-processed fruit increased from 90 to 2445 ng/g, respectively, but decreased to between 75 and 695 ng/g, respectively, following a water wash step. Subsequent removal of rotten/damaged fruit decreased patulin levels further (to between 55 and 405 ng/g, respectively), although the numerical decreases between sampling points were not shown to be statistically significant (P > 0.05). However, patulin levels were significantly higher (P < 0.05) in the rejected rotten/damaged fruit (mean levels ranged from 1120 to 6235 ng/g, respectively). P. expansum was the major patulin-producing fungus isolated from the juice samples. The mycological analyses tended to support the chemical data, in that removal of the rotten/damaged fractions significantly reduced total fungal counts in the juice samples.

Physicochemical data for some selected Fusarium toxins.

Fusarium toxins are a major group of secondary metabolites, produced by several species, that may contaminate food cereals and animal feeds. We describe results of a study in which a number of physicochemical constants for 12 Important Fusarium mycotoxins (zearalenone, diacetoxyscirpenol, T-2 toxin, neosolaniol monoacetate, deoxynivalenol, nivalenol, fumonisin B1, fumonisin B2, moniliformin, fusarenon-X, HT-2 toxin, and beta-zearalenol) were determined. Nuclear magnetic resonance, mass spectrometric, UV spectral, molar absorption coefficients, fluorescence spectra, melting points, and specific rotation data are presented.

Worldwide survey of fumonisin contamination of corn and corn-based products.

As part of a comprehensive risk assessment study for fumonisins, reliable data on exposure of populations to these dietary toxins must be obtained. To assess the extent of worldwide exposure, the published literature on the contamination of food and feed supplies has been reviewed and supplemented with unpublished material from various international sources. Fumonisin contamination of corn and corn-based products occurs in many countries. Animal mycotoxicoses such as equine leukoencephalomalacia and porcine pulmonary edema are caused by heavily contaminated animal feeds. For example, as much as 330 micrograms/g fumonisin B1 (FB1) has been found in swine feed. Although commercially available refined corn products for human consumption are generally contaminated at levels below 1 microgram/g FB1, individual products in certain countries can reach far higher levels. Health risks associated with consumption of these products depend on the extent to which they are consumed in a varied diet. Home-grown corn in certain rural areas, where it also constitutes the staple diet, can be contaminated at > 100 micrograms/g. Consumption of corn contaminated at these high levels has been associated with a high incidence of esophageal cancer in these areas.

Liquid chromatographic determination of fumonisins B1, B2, and B3 in corn: AOAC-IUPAC Collaborative Study.

A liquid chromatographic (LC) method for simultaneous determination of fumonisins B1 (FB1), B2 (FB2), and B3 (FB3) in corn was subjected to a collaborative study involving 12 participants from 10 countries, in which the accuracy and reproducibility characteristics of the method were established. Mean analyte recoveries from corn ranged from 81.1 to 84.2% for FB1 (at a spiking range of 500 to 8000 ng/g), from 75.9 to 81.9% for FB2 (at a spiking range of 200 to 3200 ng/g), and from 75.8 to 86.8% for FB3 (at a spiking range of 100 to 1600 ng/g). The valid data were statistically evaluated after exclusion of outliers. Relative standard deviations for within-laboratory repeatability ranged from 5.8 to 13.2% for FB1, from 7.2 to 17.5% for FB2, and from 8.0 to 17.2% for FB3. Relative standard deviations for between-laboratory reproducibility varied from 13.9 to 22.2% for FB1, from 15.8 to 26.7% for FB2, and from 19.5 to 24.9% for FB3. HORRAT ratios, calculated for the individual toxin analogues, ranged from 0.75 to 1.73. The LC method for determination of fumonisins B1, B2, and B3 in corn (at concentrations of 800-12800 ng total fumonisins/g) has been adopted by AOAC INTERNATIONAL.

Duckling toxicity and the production of fumonisin and moniliformin by isolates in the A and E mating populations of Gibberella fujikuroi (Fusarium moniliforme).

Two biological species of Gibberella fujikuroi (A and F mating populations) share the Fusarium moniliforme anamorph. Twenty strains of each of these biological species were tested for the ability to produce fumonisins B1, B2, and B3 and moniliformin and for toxicity to 1-day-old ducklings. Most of the members of the A mating population (19 of 20 strains) produced more than 60 micrograms of total fumonisins per g, whereas only 3 of 20 members of the F mating population produced more than trace levels of these toxins and none produced more than 40 micrograms of total fumonisins per g. In addition, only 3 of 20 members of the A mating population produced more than 1 microgram of moniliformin per g (and none produced more than 175 micrograms/g), while all 20 strains of the F mating population produced more than 85 micrograms of this toxin per g and 1 strain produced 10,345 micrograms/g. The duckling toxicity profiles of the strains of the two mating populations were similar, however, and the level of either toxin by itself was not strongly correlated with duckling toxicity. On the basis of our data we think that it is likely that the members of both of these mating populations produce additional toxins that have yet to be chemically identified. These toxins may act singly or synergistically with other compounds to induce the observed duckling toxicity.

The reliability and significance of analytical data on the natural occurrence of fumonisins in food.

Several methods are presently used to identify and quantify fumonisins in foods and feeds. HPLC procedures on derivatized fumonisins with fluorescence detection are most commonly used. The validity and significance of reported fumonisin levels depend on several factors such as the specificity, detection limit, accuracy and reproducibility of the analytical method as well as on the sampling procedure used, and the integrity and purity of the analytical standards. The importance of these factors is discussed and the results of two international collaborative studies are presented on the determination of fumonisins in corn by a reversed-phase HPLC method on o-phthaldialdehyde (OPA) derivatized fumonisins using fluorescence detection.

Toxicokinetics of the mycotoxin fumonisin B2 in rats.

Fumonisin B2 (FB2), a secondary metabolite of the fungus Fusarium moniliforme, was administered at a dose of 7.5 mg/kg body weight to male BD IX rats by ip injection or by gavage. FB2 was rapidly absorbed from the peritoneum, its level in plasma reaching a maximum within 20 min after injection. It was rapidly eliminated from plasma with a half-life of 26 min. After 24 hr, FB2 could not be detected in plasma (< 20 ng/ml). Analysis of rat plasma for FB2 following a gavage dose failed to detect any toxin over a 6-hr period after dosing. The elimination of FB2 in the urine and faeces was determined over a 3-day period after dosing. After i.p. injection, the mean urinary excretion over this period was 1.2% and faecal elimination accounted for 84.1% of the dose. Similarly, after dosing by gavage, 0.2 and 82.0% of the dose was recovered in urine and faeces, respectively. FB2 appeared to be excreted unmetabolized.

Liquid chromatographic determination of the mycotoxin fumonisin B2 in physiological samples.

The fungus Fusarium moniliforme produces a group of mycotoxins, the fumonisins, of which the most abundant are fumonisins B1 (FB1) and B2 (FB2). Previously developed analytical methods for the determination of FB1 in physiological samples have been modified for the determination of FB2 by the use of less polar extraction solvents. Plasma and urine extracts were purified on strong anion-exchange solid-phase extraction cartridges and fecal extracts on reversed-phase (C18) cartridges. FB2 in purified extracts was determined by reversed-phase HPLC with fluorescence detection using performed o-phthaldialdehyde derivatives. These methods were reproducible (R.S.D. of less than 6%) with recoveries greater than 85%. In a short preliminary study, they have been applied to the determination of the fate of FB2 dosed to rats by gavage. Of the dose given to the animals, over 90% was recovered unmetabolised in the feces within 48 h.

Fate of a single dose of 14C-labelled fumonisin B1 in vervet monkeys.

The mycotoxin fumonisin B1 (FB1) was dosed as 14C-labelled FB1, to male vervet monkeys (Cercopithecus aethiops) both by intravenous (i.v.) injection (2 monkeys, dose 1.72 mg [86 kBq]/kg body weight) and by gavage (2 monkeys, dose 6.42 mg [321 kBq]/kg body weight). Excreta were collected over a 24-hr period, whereafter the monkeys were sacrificed and selected organs and contents of the gut collected to determine the distribution of the 14C-label. The bulk of the radioactivity recovered from tissue was found in the liver (mean of 1.92% in i.v.-dosed monkeys; 0.64% in gavage-dosed monkeys). Of the other organs analysed, the following mean amounts of radioactivity were recovered in organs of i.v.- and gavage-dosed monkeys, respectively: muscle, 0.62% and 0.14%; kidney, 0.37% and 0.03%; brain, 0.08% and 0.02%; lung, 0.07% and 0.03%; heart, 0.04% and 0.01%; spleen, 0.02% and < 0.01%; plasma, 0.66% and 0.12%; red blood cells, 0.11% and 0.01%; while a further 68.1% and 64.0% were recovered in excreta, bile, and the gut contents. Analysis of faeces and gut contents showed that radioactivity was due to FB1, its partially hydrolysed metabolites, and trace amounts of the fully hydrolysed aminopentol moiety. Analysis of bile showed an absence of hydrolysis products, indicating that hydrolysis occurred only in the gut, resulting in the removal of the tricarballylic acid moiety at the C14-position. Determination of FB1, levels in plasma following a gavage dose indicated that only limited amounts of FB1 were absorbed, as plasma levels peaked after 1-2 hr with levels below 210 ng/ml.

Ear-rot fungi and mycotoxins in South African corn of the 1989 crop exported to Taiwan.

A shipment of South African corn (1989) exported to Taiwan, was analyzed for various ear-rot fungi and Fusarium mycotoxins. Two sets of samples, one from the points of origin in South Africa prior to shipment, and the other from the end-point distributors in Taiwan, were studied. Surface-sterilized kernels were plated onto two different agar media and the fungal colonies identified. High Performance Liquid Chromatography was used to analyze mycotoxin levels. The predominant ear-rot fungi, in decreasing order of isolation frequency, were Fusarium subglutinans, F. moniliforme, Diploidia maydis and F. graminearum. Aspergillus flavus and A. parasiticus were not isolated from samples prior to export, but a small number of A. flavus isolates were found after shipment. The predominant mycotoxins were fumonisins B1 (0-865 ng/g) and B2 (0-250 ng/g). Low levels of moniliformin (< or = 390 ng/g) were detected in some samples before shipment. Zearalenone (25 ng/g), and nivalenol (120 ng/g) were detected in two out of 32 samples taken in Taiwan. The samples contained no detectable levels of either aflatoxins (> 0.5 ng/g) or deoxynivalenol (> 100 ng/g) before or after shipment.

Distribution and excretion of a single dose of the mycotoxin fumonisin B1 in a non-human primate.

Fumonisin B1 (FB1), a toxic and carcinogenic secondary metabolite of the fungus Fusarium moniliforme Sheldon, was administered either by i.v. injection or by gavage to vervet monkeys (Cercopithecus aethiops). FB1 dosed by i.v. injection to two female vervet monkeys was rapidly eliminated from plasma with a mean half-life during the elimination phase of 40 min. Analysis of urine and faeces over a 5 day period after dosing gave an average 47% recovery of the dose as FB1 and its hydrolysed analogues. Two female vervet monkeys were given a single gavage dose of 14C-labelled FB1. During the subsequent 3 day period, faecal excretion of radioactivity accounted for an average of 61% of the administered dose and urinary excretion 1.2%. Residual radioactivity was recovered in low levels from skeletal muscle (1%), liver (0.4%), brain (0.2%), kidney, heart, plasma, red blood cells and bile (each 0.1%), while the contents of the intestines accounted for a further 12% of the radioactive dose. In total, 76% of the administered radioactivity was recovered. Analysis of the faeces, intestinal contents and urine indicated that over 90% of the radioactivity in these samples was due to FB1 and its hydrolysis products.

Biliary excretion of the mycotoxin fumonisin B1 in rats.

The biliary excretion of the mycotoxin fumonisin B1 (FB1), produced by the fungus Fusarium moniliforme Sheldon, has been measured in male Wistar rats. After ip injection of a solution of FB1 (7.5 mg/kg body weight), 67% of the applied dose was recovered in bile over a 24-hr period, 88% of this recovery being excreted in the first 4 hr after dosing. In contrast to these results, a similar dose of FB1 given by gavage resulted in only 0.2% recovery of the toxin in bile over a 24-hr period. Hence, although these results show that biliary excretion is a major route of elimination of FB1 from the circulation, only small amounts of the toxin appeared to be absorbed from the gut in rats.

Determination of the mycotoxin fumonisin B1 and identification of its partially hydrolysed metabolites in the faeces of non-human primates.

A method has been developed for the determination of fumonisin B1 (FB1) in the faeces of non-human primates (vervet monkeys). The animals were dosed with 14C-labelled FB1, and the radioactive compounds in faeces were recovered by repeated extractions with 0.1 M ethylenediaminetetraacetic acid. The extracts were cleaned-up on a reversed-phase (C18) solid-phase extraction cartridge, and FB1 was determined by o-phthaldialdehyde derivatization and reversed-phase HPLC. The analytical method for the determination of FB1 in the faecal extracts was reproducible [2.6% relative standard deviation (RSD)] and accurate (recovery from spiked blank extracts of 93 +/- 2.9% RSD). Confirmation of the identification of FB1 in faeces was achieved using HPLC and thin-layer chromatography, which showed that the radioactivity extracted corresponded mainly to FB1 and a new metabolite with chromatographic properties similar to those of the mycotoxin. The new metabolite was identified by mass spectrometry and nuclear magnetic resonance spectroscopy to be an equilibrium mixture of the two structural isomers of partially hydrolysed FB1, which are formed by hydrolysis of one of the ester groups of the mycotoxin.

Fumonisin-contaminated maize: physical treatment for the partial decontamination of bulk shipments.

Ten maize samples, randomly selected from a bulk shipment imported into South Africa, were characterized by a wide distribution in particulate size. Following fractionation by sieving through a 3 mm screen, the 'kernels' (fractions > or = 3 mm) corresponding to between 80.0 and 95.3% of the samples by mass, were contaminated with total fumonisin levels of between 530 and 1890 ng/g. Conversely, those fractions termed 'fines' (< 3 mm) had significantly higher total fumonisin concentrations of between 12,340 and 27,460 ng/g, and accounted for between 4.7 and 20.0% of the samples by mass. The data indicated that removal of the 'fines' resulted in overall reductions in total fumonisin levels of between 26.2 and 69.4%. It is suggested that initial removal of 'fines' from bulk shipments of maize, prior to further processing, could be considered as a preliminary fumonisin-decontamination procedure.

Determination of fumonisins in corn: Evaluation of two purification procedures.

The performance of two solid phase extraction (SPE) purification procedures, used in the determination of fumonlsin B1 (FB1), B2 (FB2) and B3 (FB2) In corn, was evaluated using both thin-layer chromatography (TLC) and high performance liquid chromatography (HPLC). Fewer interferences were observed In extracts prepared using the strong anion exchange (SAX) media, in contrast to those purified on C18 media, where on occasions, visual discernment of the TLC bands was hampered by the presence of interfering compounds. Precipitate formation, resulting In the blocking of SPE cartridges was also encountered when using the C18 procedure. HPLC analyses of extracts prepared by both media indicated that they gave comparable fumonlsin recoveries from naturally contaminated corn samples. The results suggest that the C18 procedure, originally developed for the TLC analyses of FB1 in mixed feeds, may also be applied to the determination of FB2 and FB2. However, where TLC is used quantitatively for fumonlsin levels <1 µg/g, purification of sample extracts on SAX media is recommended.

Fate of a single dose of the 14C-labelled mycotoxin, fumonisin B1, in rats.

The fate of the mycotoxin, fumonisin B1, (FB1) dosed to rats by i.p. injection and by gavage was traced using 14C-labelled FB1. Twenty-four hours after i.p. injection, 66% of the radioactivity was recovered in faeces, 32% in urine, 1% in liver and trace amounts (less than 1%) in kidney and red blood cells. When dosed by gavage, all (101%) radioactivity was recovered in faeces and trace amounts were found in urine, liver, kidney and red blood cells. The bulk of the radioactivity recovered was unmetabolized FB1.