Coauthors

Itagaki Y., Fujita T., Naoki H., Yasuhara T., Andriantsiferana M., Nakajima T.

Spider venom glands store various novel neurotoxic acylpolyamines which present potent and irreversible inhibition of the glutamatergic synapses. We have developed a new highly sensitive analytical method to detect and characterize structures of those neurotoxins stored in a single venom gland using micro bore column high performance liquid chromatography (HPLC)-continuous flow FRIT fast atom bombardment mass spectrometry (HPLC-FAB/MS) and tandem mass spectrometry (MS/MS) array detection system. The high-energy collision induced dissociation (CID) spectra of sodium cationized spider toxins produced very strong structurally informative product ions which afforded the location of nitrogen atoms and the connectivity of methylene units within the polyamines. This methodology permitted detection of 40 amino acid containing acylpolyamines in the venom of Nephilengys borbonica from which structures of only five compounds were previously known. Nat. Toxins 5:1–13, 1997. © 1997 Wiley-Liss, Inc.

Laycock M.V., Kralovec J., Richards R.

A method is described to sulfate PSP toxins at various positions in the molecule and to prepare 35S labelled compounds using H2(35)SO4 in the presence of dicyclohexylcarbodiimide (DCC). The 11-sulfates of saxitoxin and neosaxitoxin, known as gonyautoxins, are often the most abundant of the PSP toxins in algae and contaminated shellfish. Receptor site binding and antibody assays based on these analogues should, therefore, better reflect toxicity than those in which saxitoxin is used. Although the specific activity of 35S-gonyautoxins is lower than that of commercially available 3H-saxitoxin, the label is strongly bound and is not lost through proton exchange with water as occurs with tritiated saxitoxin. The labelling procedure is rapid, inexpensive and can be done on a small scale. Sulfate can be removed from the 11-position of GTX's in methanolic-HCl and from the 21-position by mild acid hydrolysis and H2(35)SO4 added in 5-10-fold excess. Addition or exchange occurs rapidly on mixing DCC in dimethylformamide with dry toxin and sulfate. Reaction conditions were optimized and reaction products identified by capillary electrophoresis, autoradiography and ionspray mass spectrometry. Together with methods for selective removal of sulfate, the sulfation reaction provides an additional way to prepare some of the naturally occurring derivatives of saxitoxin, many of which are sulfates.

Haque A., Hossain M., Lambein F., Bell E.A.

In a study of 500 patients suffering from neurolathyrism in Bangladesh it was found that 60 (all male) complained of bone pain and showed skeletal deformities suggestive of osteolathyrism. On X-ray examination a failure of fusion in both vertebral and iliac epiphyses was found in two patients. At the age of these patients (30 and 37 years) such failure was considered a clear evidence of osteolathyrism. All 60 patients were accustomed to eating the green parts of Lathyrus sativus, which contain 2-cyanoethyl-isoxazolin-5-one, a compound that chemically and metabolically can produce the osteolathyrogen beta-aminopropionitrile (BAPN), as well as foods made from the seeds of the same plant which contain the neurotoxin 3-N-oxalyl-2,3-diaminopropanoic acid (beta-ODAP).

Wickstrom M., Haschek W., Henningsen G., Miller L.A., Wyman J., Beasley V.

The cyanobactenal hepatotoxin, microcystin-LR (MCLR), is a potent protein phosphatase inhibitor that disrupts actin microfilament, cytokeratin intermediate filament, and microtubule networks in hepatocytes. To determine ultrastructural and biochemical changes that develop concurrently with microcystin-induced cytoskeletal disorganization, isolated rat livers were perfused with MCLR at 0.1 to 5.0 μg/ml for 5 to 40 min. Lactate dehydrogenase, alanine aminotransferase, and aspartate aminotransferase changed over time, but trends for toxin-treated and control livers did not differ. The earliest toxin-induced ultrastructural changes, observed in livers perfused at 0.1 μg/ml for 15–20 min or at 0.3 μg/ml for 5–10 min, were loss of hepatocyte microvilli in the space of Disse, widening of sinusoidal fenestrae, disruption of sinusoidal endothelium, dilation of bile canaliculi with loss of microvilli, and widening of hepatocyte intercellular spaces. Lesions progressed with increasing toxin concentrations and exposure times. In livers perfused with MCLR at 0.5 μg/ml for 10–20 min, hepatocytes had plasma membrane blebs and concentric whorls of rough endoplasmic reticulum, and there was marked disassociation of hepatocytes resulting in disrupted hepatic cords. At toxin concentrations of 2.0 or 5.0 μg/ml for 10–20 min, there was mild dilation of mitochondrial cristae, cytoplasmic vacuolization or invagination of plasma membranes, redistribution of organelles, and sometimes nuclear degenerative change. Some hepatocytes exhibited clusters of plasma membrane blebs radiating from round cytoplasmic structures, which may be composed primarily of condensed microfilaments.

Harada K., Oshikata M., Uchida H., Suzuki M., Kondo F., Sato K., Ueno Y., Yu S., Chen G., Chen G.

Cyanobacterial toxins, microcystins, have a potent tumor-promoting activity. We investigated the level of microcystins in drinking water collected from 1992 to 1994 in Haimen City, China, where people who drink pond ditch water usually incurred a high incidence rate of hepatocellular carcinoma compared with those who drink well water. High-performance liquid chromatography, liquid chromatography/mass spectrometry (LC/MS), and protein phosphatase inhibition assay (pp assay) were used to identify and quantify the microcystins. Microcystin LR and [D-Asp3]microcystin LR were detected in 2 of 50 samples at a concentration less than 100 ng/L by LC/MS in 1992. Although no microcystins were found by the chemical method in 1993, 6 of 7 samples except for 3 tap water samples showed an approximate amount of 100 ng/L by using the pp assay in 1994. The obtained results supported the epidemiological results reported by Yu.

Alfonso D., Johnson H.A., Colman-Saizarbitoria T., Presley C.P., McCabe G.P., McLaughlin J.L.

Roy D.N., Sabri M.I., Kayton R.J., Spencer P.S.

The legume Vicia sativa (common vetch) harbors the neurotoxic nonprotein amino acid beta-cyano-L-alanine (BCLA) and its gamma-glutamyl derivative. BCLA elicits hyperexcitability, convulsions, and rigidity in chicks and rats after oral or intraperitoneal administration, but the mechanism of its action is unknown. The effect of different concentrations of BCLA (0.075-10.0 mM) has been investigated in an organotypic tissue culture system. BCLA concentrations of 0.075 and 0.60 mM had no effect, even up to 6 hr. No changes were observed in cultures treated with 1 mM BCLA for 4 hr. BCLA (2.0-10.0 mM) induces concentration-dependent changes in the explants. The explants display neurona vacuolation, chromatin, clumping, and dense shrunken cells, a pathological response generally seen with excitotoxin. MK-801 (35 microM), which blocks the open ion channel associated with the N-methyl-D-aspartate (NMDA) class of glutamate receptors, attenuates the neurotoxic property of BCLA, while the non-NMDA antagonist, 6-cyano-7-nitroquinoxaline-2,3-dione (10-20 microM), provides no significant protection. Treatment of isolated mouse brain mitochondria with up to 5 mM BCLA had no inhibitory effect on the activity of NADH dehydrogenase (complex I) or cytochrome or oxidase (complex IV), a cyanide-sensitive enzyme. These results suggest that the neurotoxicity of BCLA (or derivative) is mediated directly or indirectly through NMDA receptors.

Piñeiro M., Dawson R., Costarrica M.L.

A pilot study for monitoring mycotoxin contamination in food and feeds was implemented by the Technological Laboratory of Uruguay (LATU) with the technical assistance of the Food and Agriculture Organization of the United Nations (FAO). The scope of the study was to determine the potential hazard posed by priority food-contaminant and feed-contaminant combinations. The choice of foods and contaminants to be monitored was based on the importance of the food in the total diet, the economic importance of the product and the potential health risk posed by the specific combination. The principal commodities selected were wheat, barley and rice. Also included were com, soy, dairy products, feeds, dried fruits and legumes, oil seeds, cocoa beans and organ meats. Mycotoxins analyzed (TLC/densitometry) were aflatoxins, zearalenone, ochratoxin A, deoxynivalenol (DON) and ergot alkaloids. The survey results (1993-95) showed differences in both incidence and levels of mycotoxin content for the principal commodities. Of all food/feed categories analyzed, feed had the highest values for all mycotoxins. Samples containing DON in levels above 1000 ng/g were found in all groups. Ochratoxin A was not detected in any of the samples. Rice and soy beans were the categories with lowest aflatoxin incidence. Uruguayan regulatory limits for all toxins analyzed were exceeded for wheat, barley and rice in less than 3, 9 and 7% of samples, respectively. The data on actual mycotoxin levels in different foods will help identify sources of contaminations and areas where control measures should be improved, enable better risk assessment by proper estimation of mycotoxin intake, assist in the establishment of tolerances and adequate guidelines, aid in the implementation of a national program and provide economic benefits by improving grain quality.

Castegnaro M., Garren L., Gaucher I., Wild C.P.

The fumonisins are inhibitors of de novo sphingolipid biosynthesis in vitro and in vivo and thus possibly interfere with the regulation of cell growth, differentiation, and neoplastic transformation. In addition, the ratio of free sphinganine (Sa) to free sphingosine (So) has been proposed as a marker of exposure for animals or humans consuming feed or food contaminated by these toxins. A method to analyze these sphingolipid bases has been proposed [Merrill et al., 1988: Anal Biochem 171:373-381; Riley et al., 1994a: JAOAC 77:533-540] but involves numerous steps and consequently is not ideally suited to the analysis of large numbers of samples, as is often required in epidemiological studies. A new method was therefore developed for the analysis of the Sa/So ratio in tissues as well as human and rat urine. Briefly, the method involves isolation of exfoliated cells from as little as 0.5 ml of rat urine or 2 ml of human urine followed by a rapid and efficient extraction of sphingolipid bases in ethyl acetate, an optimized derivatization step with o-phthaldialdehyde and a high-pressure liquid chromatography separation on a 250 mm x 4.6 mm. 5 microns Kromasil C18 column, with a 4-step phosphate buffer/methanol gradient. Fluorescence was monitored at 340 nm excitation, 455 nm emission, and retention times for So, Sa, and C-20 Sa were about 11, 14, and 22 min, respectively. The method was adapted to tissue analysis by partially digesting approximately 30 mg tissue with trypsin to permit isolation of a cell pellet before extraction of the sphingolipids as described above. The method was applied to the analysis of So and Sa in urines and tissues of fumonisin B1 (FB1) treated and untreated male BDIV rats. The Sa/So ratio in urine of untreated rats varied from 0.1 to 0.7, and for treated rats (between 1-5 mg FB1/kg body weight daily by gavage), the ratio was between 1.2-10. In kidney, the ratio was 0.1 in control rats and varied from 4 to 10.3 in treated rats. In human urine, measurements could easily be made in 2 ml of urine in females, but in males much larger volumes were required due to the low levels of sphingolipid bases.

Prakash A.S., Pereira T.N., Smith B.L., Shaw G., Seawright A.A.

Bracken fern (Pteridium spp.) causes cancer of the oesophagus and the urinary bladder in cattle and sheep. Ptaquiloside (PT) is believed to be the carcinogenic principle which alkylates DNA when activated to its unstable dienone form (APT) under alkaline conditions. In this report we present evidence for the presence of PT-DNA adducts in the ileum of bracken fem-fed calves using the 32P-postlabelling assay. H-ras mutations were also observed in the ileum using single strand conformation polymorphism (SSCP) technique. Mutations corresponding to adenine to pyrimidine transversions in the codon 61 of H-ras were identified by the cycle sequencing method. In vitro DNA alkylation studies showed that APT alkylated H-ras primarily at the adenines. In addition, the rate of depurination of alkylated adenine was sequence dependent. Investigation of DNA template activity using a plasmid DNA showed that DNA synthesis by T7 DNA polymerase was terminated by the presence of all alkylated bases but certain apurinic sites allowed the DNA synthesis to continue. These results suggest that initial alkylation of adenine by PT in codon 61 followed by depurination and error in DNA synthesis lead to activation of H-ras proto-oncogene.

Yoshino N., Takizawa M., Akiba H., Okumura H., Tashiro F., Honda M., Ueno Y.

Recently we have reported that T-2 toxin, a trichothecene mycotoxin produced by Fusarium species, is a potent inducer of apoptosis in the human promyelotic cell line HL-60. To clarify the signal transduction pathway of apoptosis primed by T-2 toxin, T-2 toxin-induced apoptosis was investigated in detail using confocal laser microscopy and flow cytometry. Apoptosis in HL-60 cells induced by T-2 toxin was dose dependent when the cells were treated with concentrations of 5-100 ng/ml for more than 2 hr. The apoptosis proceeds through various cell cycle stages of HL-60 cells. Prior to apoptosis, the intracellular calcium ion (Ca+2i) level was markedly elevated within 3-5 min after exposure to T-2 toxin and returned to normal level thereafter. A well-known chelator for Ca+2i, ethylene-N,N,N', N'-tetraacetic acid 4K acetoxymethyl ester (BAPTA-AM), a Ca+2-dependent endonuclease inhibitor ZnCl2, and calpain inhibitor 1 sharply blocked T-2 toxin-induced apoptosis. These results strongly suggest that the Ca+2 signal triggered by T-2 toxin is transduced by the activation of endonuclease and protease, and ultimately evokes apoptosis.

Latus-Ziętkiewicz D., Chełkowski J., Foremska E., Goliński P., Grabarkiewicz-Szczesna J., Kostecki M., Lew M., Perkowski J., Piasecki M., Wiewiórowska M., Szebiotko K.

F. moniliforme and other species of Liseola section, F. culmorum, F. dlamini, and F. nygamai, were examined for their ability to produce gibberellic acid (GA3), fumonisins, trichothecenes, zearalenone, moniliformin, and bikaverin (TLC method). Gibberellic acid was produced by F. moniliforme strains in liquid medium and on rice kernels with a maximum concentration level of 470 mg/dm3 and 1 g/kg, respectively. No strain isolated in Poland produced GA3. High-yielding gibberellic acid strains produced neither trichothecenes and fumonisins nor other tested compounds. Also the rest of strains of examined species did not produce trichothecenes and other mycotoxins except for fumonisins which were found in rice cultures of F. moniliforme, F. proliferatum, and F. subglutinans. Bikaverin was produced by F. moniliforme always together with fumonisins. Filtrates of liquid cultures of gibberellin producing strains were tested for their toxicity to brine shrimps larvae (Artemia salina). It was found that GA3 presence does not increase toxicity of these filtrates.

Rotter B.A., Oh Y.

Fumonisin B1 (FB1), a mycotoxin, is a common fungal contaminant of corn and corn products. This sphingolipid-like compound causes a variety of animal diseases and is a suspected human carcinogen. Cellular targets of FB1 include hepatocytes and renal and immune cells. The effects of FB1 on nitric oxide (NO) production induced by lipopolysaccharide (LPS) were investigated in the present study by using a murine macrophage cell line as a model system. Stimulation of NO production was observed for increasing concentrations of FB1 (1, 10, and 100 microM) and either 0.005 or 0.01 microgram/ml LPS. Although with an increasing dose of FB1 the total protein content decreased for the stimulated and the unstimulated cells, the NO production remained elevated for the stimulated cells. It can be hypothesized that the potentiation of the LPS-dependent NO production by FB1 treatment could be due to direct interaction between FB1 and NO synthases and LPS receptors or to a disrupted sphingolipid metabolism.

Khan S.A., Wickstrom M.L., Haschek W.M., Schaeffer D.J., Ghosh S., Beasley V.R.

Microcystin-LR (MCLR) is a cyanobacterial hepatotoxin that inhibits protein phosphatases 1 and 2A. To characterize cytoskeletal changes over time, hepatocytes were incubated with the toxin at 13.3 microM for 0, 2, 4, 6, 8, 16, 32, or 64 minutes. Changes in the hepatocytes were compared to those in cultured kidney cells and skin fibroblasts incubated with the toxin at 133 microM for 0, 2, 4, 8, 12, 16, or 24 hours. Cells were fixed and incubated with rhodamine-conjugated phalloidin, or primary antibodies against beta-tubulin and either vimentin or cytokeratin intermediate filaments (IFs), followed by fluorescein-conjugated secondary antibodies. The number of affected cells per 400 counted (NAC) with alterations in a specific cytoskeletal element were determined at each time point. In fibroblasts as well as kidney cells, changes occurred first in IFs, followed by microtubules (MTs), and later microfilaments (MFs). In some hepatocytes, IFs were affected first, but after 16 minutes, the NAC with altered MTs exceeded the NAC with alterations in other cytoskeletal elements. In both hepatocytes and non-hepatocytes, IFs and MTs condensed and collapsed around the nucleus. MFs similarly collapsed, but some of the actin radiated outward, producing a star-like appearance. The similarity of the cytoskeletal changes induced by MCLR in hepatocytes and non-hepatocytes suggests a common mechanism of action. Differences among cell types in sequential cytoskeletal alterations may be due to differences in phosphorylation of intracellular proteins.

Ruhland M., Engelhardt G., Schäfer W., Wallnöfer P.R.

The metabolism of the mycotoxin ochratoxin A in plant cells was investigated using cell suspension cultures of wheat and maize. A number of metabolites were detected by HPLC-chromatography with fluorescence detection. The main metabolites were ochratoxin alpha, ochratoxin A methyl ester, two isomers of hydroxyochratoxin A, and the glucosides and methyl esters of both hydroxyochratoxin A isomers. The compounds were isolated by TLC and preparative HPLC and identified by mass spectrometry and specific enzymic reactions.