TOXICOLOGY AND APPLIED PHARMACOLOGY 147.63-70 (1997) 7878 ARTICLE NO. T0978272

Mycotoxin-Induced Elevation of Free Sphingoid Bases in Precision-Cut Rat Liver Slices: Specificity of the Response and Structure-Activity Relationships1

William P. Norred.*·~ Ronald D. Plattner.t Mary Ann Dombrink-Kurtzman.t Filmore 1. Meredith.* and Ronald T. Riley* *Toxicalogy and M.\·c(}/oxin Research Vnit. Richard B. Russell Agricullural Research Center. ARSIVSDA. P.O. Box 5677. Athens. Georgia 30604·5677; and tMycolOxin Research Vnil. National Cenrer for AgriculTUral Vtili:ation Research. ARSIVSDA. 18/5 N. Vnil'ersity Sneel. Peoria. lllinois 61604

Received February 24. 1997: accepted July 24. 1997

Many are known to cause animal diseases. can Effects of Analogs and Structurally Unrelated Myco­ also be phytotoxic, and are suspected causative agents of on Elevation of Free Sphingoid Bases in Precision-Cut Rat some human illnesses, including cancer (Rheeder et al.. Liver Slices:)~p-ecificity of the Response and Structure Activity 1992). , produced by several Fusarium species. Relationships. Norred, W. P., Plattner, R. D., Meredith, F. I. and Riley, R. T. (1997). Taxieal. Appl. Pharmaeal147, 63-70. are frequent contaminants of com and corn-based feed and food (Nelson et al., 1991; Ross et al.. 1991: Sydenham et Fumonisin BI (FBI) is the predominant member of a family of toxic metabolites produced by several species of Fusarium and is al.. 1991). Fumonisin B I (FBI)' the most prevalent of several commonly found on com. FBI is a potent competitive inhibitor of naturally occurring analogs, has been shown to cause the ceramide synthase. which catalyzes the conversion of sphinganine fatal livestock diseases, equine leucoencephalomalacia and sphingosine to ceramide. The resultant accumulation of free (Marasas et al.. 1988), and porcine pulmonary edema (Har­ sphingoid bases and the disruption of sphingolipid metabolism is rison et al.. 1990). FB I has been studied in purified form in believed to be the mechanism of of the fumonisins. The laboratory rats and mice and shown to cause hepato- and objectives of this study were to determine the relative potency of nephrotoxicity (Voss et aI., 1993, 1994) and, in one study, analogs of FBI to inhibit ceramide synthase and to determine hepatocellular carcinoma (Gelderblom et aI., 1991). Several whether the inhibition is specific to mycotoxins with fumonisin­ reviews on the occurrence, toxicity, and chemistry offumon­ like structures. Fumonisins B • B • B.I , B , C , and TA (a I 1 4 4 isins are available (Norred, 1993: Powell and Plattner. 1995: structurally similar produced by the tomato pathogen. Dutton. 1996). High rates of human esophageal cancer in Alternaria altemata f. sp. lyeapersiel) were approximately equipo­ region~ of China. South Africa. and Italy where corn is con­ tent inhibitors. Hydrolyzed fumonisins BI • B1 • and B.,. which lack the tricarballylic side chains, were only 30-409c as potent as the sumed as a staple are correlated with the presence of fumoni­ parent toxins. N-acetylated FBI (FA I) did not block ceramide syn­ sins or fumonisin-producing fungi in corn and corn-based thase. suggesting that FA I is nontoxic. Inhibition of ceramide syn­ food (Chu and Li. 1994; Sydenham et al., 1990; Logrieco thase by fumonisin analogs did not appear to be related to the ('t al.. 1995). Only recently discovered (Gelderblom et al., Iipophilicity of the compounds, as determined by computer estima­ 1988). the fumonisins are an important emerging class of tion of log P values. The ability of relatively high (10 and 100 fJ M) mycolOxins that are ofgreat concem to corn producers, com­ doses of other mycotoxins that bear no structural similarity to modity groups. and industries and to regulatory agencies fumonisins, including B , cyclopiazonic acid. beauveri­ 1 responsible for food safety and grain quality. cin, T-2 toxin. ,luteoskyrin. verrucarin A, scirpen­ triol, and , to block ceramide synthase was also deter­ Fumonisin B I was the first agent discovered that inhibits mined. All of the toxins tested were negative in the bioassay with ceramide synthase (sphinganine (Sa) and sphingosine (So) the exception offumonisins. indicating that disruption of sphingo­ N-acyltransferase). which catalyzes the conversion of Sa (or lipid metabolism is a specific cytotoxic response. So) to <.:eramide (Wang et aI., 1991). The fumonisins are structurally similar to these sphingoid bases, differing mainly by the presen<.:e of tricarballylic acid side chains (Fig. I), Toxic fungal metabolites. or mycotoxins. are diverse in and it was for this reason that fumonisins B and B~ were structure. biological activity' and environmental prevalen<.:e. I tested for their ability to alter the conversion of serine into sphingolipids in an in vitro assay (Wang et al., 1991). Other I Presented in pan at the 35th Annual Meeting of the Soclet~ of Tmucol· ogy. Anaheim. CA. agents thai interfere with sphingolipid metabolism have e To whom correspondence should be addressed. sin<.:c heen discovered. including compounds such as AAL

0041-008XJ97 64 NORRED ET AL.

OH sphingolipid metabolism are alteration of signal transduction pathways through repression ofprotein kinase C and stimula­ H tion of cyclic AMP response elements (Huang et al.. 1995) .. Sphinganine and inhibition of protein serine/threonine phosphatases (Fu­ kuda et al.. 1996). Thus. whereas the principal subcellular effect of fumonisins appears to be simple, i.e.. inhibition of ceramide synthase, the ramifications of this action for ad­ verse biological consequences may be complex and far reaching. Disruption of sphingolipid metabolism could therefore lead to long-term toxicity and. possibly. carcinoge­ nicity. but not necessarily acute toxicity. Indeed. fumonisins are not particularly acutely toxic either in vivo or in \'itro R1 R2 R3 R4 RS (Norred and Voss, 1994). Several reviews discussing the FB1 H OH OH TCA TCA role of sphingolipids and the disruption of their metabolism in the toxicity and carcinogenicity of fumonisins are avail­ FB2 H OH H TCA TCA able (Merrill et al.. 1995a,b. 1996a: Riley et al.. 1996). FB3 HH OH TCA TCA A number of structural analogs of FBI have now been H H H TCA TCA FB4 discovered (Powell and Plattner. 1995). and it is known that FA1 COCH3 OH OH TCA TCA several of these, including fumonisins BI, B2 , B3 and hy­ HFB1 H OH OH H H drolyzed FB I (Fig. 1). share the ability to inhibit ceramide HFB2 - -- - H OH H HH synthase (Wang et al.. 1991; Merrill et al.. 1993). The pur­ HFB3 H H OH HH pose of the present study was twofold. First, we wanted to determine the relative capability of structural analogs of fumonisin B I to induce elevated levels of free sphinganine when incubated with precision-cut rat liver slices, a bioassay system which we previously reported to be a sensitive indica­ tor of FBI activity (Norred et al.. 1996). Second, we wanted to determine whether the inhibition of ceramide synthase is specific to fumonisins or fumonisin-like compounds or R1 R2 R3 R4 RS whether elevated sphinganine levels could be a generalized IFC4 H H TCA TCA CH2CH3 response of cells to toxic insult by other, structurally unre­ ITA TOXIN OH OH TCA HH lated toxins.

FIG. I. Structures of sphinganine and fumonisin analogs tested. MATERIALS AND METHODS

Rats and li~'er slices. Tissues were obtained from healthy male. un­ toxin (TA toxin: Merrill et al.. 1993) and australifungin fasted Sprague-Dawley rats (200-300 g; Harlan Industries. Indianapolis, (Mandala et al.. 1995). Normally. ceramide is converted to IN) maintained in stainless-steel cages mounted in racks equipped with complex sphingolipids. including sphingomyelin and glyco­ automatic drinking water and excreta flushing in temperature- and humidity­ sphingolipids (Fig. 2). The inhibition of ceramide synthase controlled animal rooms. Feed (RMH 3000 certified. Agway. Waverly. NY) caused by FBI disrupts sphingolipid metabolism. causing was analyzed as previously described to ensure fumonisin content was below 0.5 ppm (Chamberlain eT of.. 1993). Individual rats were killed during elevated levels of free intracellular Sa and So. increased the morning hours (6:30-7:30 AM) by inhalation of carbon dioxide. and amounts of sphingoid base degradation products such as the liver was quickly removed and placed in ice-cold Sack's buffer (5.5 sphinganine-l-phosphate. depletion of complex sphingoli­ mM KH~PO •• 14.3 mM NaHCO,. lOA mM K~HPO.·3H~O, 6 mM KHCO,. pids. and increased biosynthesis of phosphatidylethanolam­ 206 mM mannitol. and 3.7 mM MgCI~' 6H~O). Cores of liver (8 mm diame­ ine (Smith and Merrill. 1995: Yoo et al.. 1996). Other effects ter) were cut with minimum pressure using a stainless-steel borer filled on a specially designed drill press (Alabama Research and Development. of FB I which have been noted and which likely are end Mumford. All. The cores were placed in ice-cold Sack's buffer and trans­ results of disrupted sphingolipid metabolism, include inhibi­ ferred one at a time to a Krumdieck tissue slicer (Alabama Research and tion of cell proliferation and increased cell death (Yoo et Development). Slices (250 pm) were cut according to the manufacturer's al.. 1996: Wang et al., 1996: Tolleson et al.. 1996). reduction instructions. The slices were transferred to Teflon/titanium mesh rollers of the barrier function of endothelial cells thus allowing (Vitron. Inc.. Tucson. AZ: two slices/roller). and then the rollers were placed in 20-ml glass scintillation vials and incubated in Waymouth's me­ "leakage" of albumin across membranes (Ramasamy et al.. dium (Gibco. Grand Island. NY) with various doses of test agents dissolved 1995). and stimulation of DNA synthesis (Schroeder et al.. in 10 pi of solvent (distilled water or dimethyl sulfoxide) for' 48 hr as 1994). Other effects which could be related to disruption of previously described (Norred eT of.. 1996). except that the final volume of FUMONISIN SAR IN LIVER SLICES 65

SERINE

+ 3-KETOSPHINGANINE SPHINGANINE PA.LMITOYL-CoA

FUMONISIN •••••..-

SPHINGOMYELIN ~

CERAMIDE ...,....----- DIHYDROCERAMIDE ~. GLYCOSPHINGOUPIDS t. j FUMONISIN •••••..-f

SPHINGOSINE

SPHINGOSINE-1-PO.

FIG. 2. Pathway of sphingolipid biosynthesis and catabolism.

medium in the incubation vials was 1.7 ml instead of 2.0 ml. Control slices phosphate-buffered saline (PBS, Gibco, Grand Island, NY). and transferred were treated with solvent only. For cytotoxicity assays, positive control to a well of an 8-well tissue culture Multiplate (LUX, Nunc, Inc., Naperville, slices were dosed with 10 JlI of Triton X-IOO surfactant (Sigma Chemical IL) containing I ml of fresh PBS. Fifty microliters of MIT solution (5 mg! Co.. St. Louis. MO). ml of PBS) was added to each well, and the plates were incubated at 37· for 2 hr. We determined experimentally that under these conditions, MIT Toxins and reagents. Fumonisins B,. B2 , B). B., A,. C•• HFB,. HFB 2 • and HFB) were isolated from cultures of Fusarium monilifonne or F. prolif­ would penetrate slices up to 500 f.Lm thick within 2 hr. After incubation, erarum grown on either com or rice or in liquid medium and purified by PBS was aspirated from the wells. leaving the slices behind, and I ml of previously described methods (Plattner et al.. 1992; Branham and Plattner, isopropanol was added to each well. The plates were placed on a gyrorotary 1993; Meredith et al.. 1996a.b; Poling and Plattner. 1996). TA toxin was shaker for 30 min to extract the formazan crystals from the slices. Intensity a generous gift of Dr. W. F. O. Marasas (Medical Research Council, Tyger­ of the purple color was measured spectrophotometrically at 570 nm against berg. South Africa). All the fumonisin analogs and the TA toxin were an isopropanol blank. Repeated isopropanol extraction of the slices resulted judged to be >95% pure. as determined by high-performance liquid chro­ in only negligible (less that 5%) increase in recovery. matography-mass spectrometry (HPLC-MS). For each of the analogs, there Protein determination, After extraction of the liver slices used for the was no evidence of the presence of any other fumonisin analog. also deter­ MIT assay with isopropanol, the slices were solubilized in 0.2 N NaOH mined by HPLC-MS. Aflatoxin B,. cyclopiazonic acid, beauvericin, T-2 and protein content was determined (Lowry et al.. 1951). toxin, sterigmatocystin. verrucarin A.luteoskyrin. scirpentriol. and zearalen­ n-Octanol:water partition coefficient. The 10glO of the octanol:water one were purchased (Sigma Chemical Co.). o-erythro-Dihydrosphingosine panition coefficient (log P) of fumonisin B, was experimentally deter­ (sphinganine) (a gift of A. H. Merrill. Jr., Emory University) was complexed mined. Equal volumes (0.75 mJ) of n-octanol and a solution ofFB, in water in water with an equimolar concentration of fatty acid-free bovine serum (0.122 mg!ml) were placed in a 2-ml plastic tube and shaken overnight at albumin (Sigma Chemical Co.). Other chemicals were reagent grade or room temperature. The liquid phases were separated by centrifugation better and were commercially available. (8000g for 10 min), and the octanol layer was collected by pipette. An Sphingoid base analysis. Content of free sphinganine and sphingosine aliquot (0.1 ml) of the octanol layer was dried under nitrogen, and the in the slices was determined by a high-performance liquid chromatography residue was dissolved in acetonitrile:water (1:1). Fumonisin B, was quanti­ procedure as previously described (Norred et al.. 1996). tated by high-performance liquid chromatography as described by Ross et Cytotoxicity assay. The ability of liver slices to conven MIT (3-[4.5­ al. (1991). Estimates of log P for FB Io FAIo FB., and HFB, were deter­ dimethylthiazol-2-ylj-2.5-diphenYlietrazoliumbromide) to purple formazan mined using a computer modeling program (Program ClogP, BioByte crystals by mitochondrial dehydrogenase was determined by a modification Corp.. Claremont, CAl. of the procedure described by Mosmann (1983). After the incubation period, Statistical analyses. Three replicate experiments were conducted for one of the two slices from each vial was removed, rinsed with Dulbecco's each toxin and dose tested, and the results were analyzed by analysis of 66 NORRED ET AL.

TABLE 1 Effect of Incubation of Rat Liver Slices with Fumonisin Analogues on Free Spbingoid Bases"

Concentration of Toxin (pM)

0.05 0.5 5.0 0.05 0.5 5.0

Toxin Sphinganine Content (nmollg tissue::: SD) Sphinganine:sphingosine + SD

FB, 0.7::: 0.3 6.4 ::: 3.5* 6.9::: 3.0* 0.18 ::: 0.03* 1.25 ::: 0.36* 1.36::: 0.17* FB, 1.0 ::: 1.1 3.1 ::: 1.6* 5.8 ::: 3.4* 0.20::: 0.13 0.62::: 0.06* 1.17 ::: 0.10* FB, 0.7 ::: 0.4 3.5 ::: 1.7* 5.5 ::: 3.4* 0.19::: 0.04* 0.79::: 0.14* 1.09 ::: 0.19* FB. 2.1 ::: 1.0* 5.2::: 4.0* 7.3 ::: 4.7* 0.40 :!: 0.10* 0.91 ::: 0.46* 1.44 ::: 0.23* FA, 0.5::: 0.1 0.5 ::: 0.2 1.0 :!: 0.6 0.09 :!: om 0.11 ::: 0.06 0.15 ::: 0.07 FC 1.4 ::: 0.6* 3.8 :!: 3.1 * 5.0::: 3.1* 0.36:!: 0.16* 0.76 ::: 0.09* J.l4 ::: 0.05* HFB, 0.5 ::: 0.2 1.5::: 0.7* 3.2 :!: 2.0* 0.12 :!: 0.03 0.34:!: 0.04* 0.57 ::: 0.16* HFB, 0.6 ::: 0.4 1.2 :!: 0.9* 2.4 ::: 1.3* 0.12 :!: 0.06 0.26 :!: 0.11* 0.47 :!: 0.12* HFB, 0.5 ::: 0.3 0.6 :!: 0.4 1.9 ::: 1.3* 0.11 :!: 0.03 0.15 :!: 0.06 0.34::: 0.06* TA TOXIN 2.1 ::: 1.1 * 6.5::: 2.8* 8.9::: 3.6* 0.43 :!: 0.05* 1.19 :!: 0.11 * 1.63::: 0.09*

"Slices were incubated for 20 hr as described under Materials and Methods. Values are means:!: SD of three experiments. For control slices (DMSO or water treated). sphinganine content of slices was 0.5 ::: 0.3 nmollg tissue and ratio of sphinganine to sphingosine was 0.11 :!: 0.05. * Significantly different (p < 0.05) from control slices.

variance procedures using a commercially available computer program (Sig­ significantly elevated Sa:So (Fig. 3). A number of the myco­ maS tal. Jandel Scientific. San Rafael. CAl. An experiment was defined as toxins used were cytotoxic at these doses. as measured by the mean of triplicate determinations with liver slices obtained from one the ability of the test compounds to reduce the conversion rat. Thus for each parameter measured. the n for statistical analysis was 3. Availability of some of the analogs precluded more experimental replica­ of MIT to the purple dye. formazan (Table 2). The cytotoxic tions. When significant differences were detected. Dunnett's method was compounds included aflatoxin B I• T-2 toxin. sterigmatocys­ used to identify values that were significantly different from control values. tin. verrucarin A, and scirpentriol. Cyclopiazonic acid. with a significance level of p < 0.05. beauvericin. Sa. FBI. luteoskyrin. and zearalenone were not toxic by the MIT conversion assay at the dose levels used. RESULTS n-Octanol:water partition coefficients. The experimen­ Effectsoffumonisinanalogsonfreesphingoidbases. Con­ tally determined log P for fumonisin BI• -1.84. compared centration of free So was not altered by exposure of slices favorably with the computer generated predicted value of to any of the agents tested in this study (data not shown). -1.534. The calculated log P values for the other analogs Significant elevations in the content of free Sa and the Sa:So were FA) = 0.627; FB4 = 2.880; HFB, = 0.610. ratio were produced in liver slices after incubation of the slices for 20 hr with 0.5 or 5.0 pM concentrations of all the DISCUSSION analogs tested. except fumonisin AI (Table 1). Incubation with the low dose (0.05 pM) of FB 4 • FC4 • and TA toxin It has been proposed that the fumonisins exert their toxic elevated Sa. and Sa:So was elevated by the low dose of FB) • effects through the disruption of sphingolipid metabolism FB 3 • FB 4 • Fe. and TA toxin. Hydrolyzed fumonisins B I • because of their inhibition of ceramide synthase (sphinga­ B2 • and B3 were less potent inducers of Sa. The elevations nine and sphingosine N-acyltransferases) (Riley et al.. 1996; in Sa:So ratio produced by 0.5 and 5.0 pM doses of HFB 1 • Merrill et al.. 1995b). The result of fumonisin exposure in HFB 2 • and HFBJ were 30 to 40% of the elevations produced all animals. , and eukaryotic cells tested thus far is the by equimolar concentrations of the parent fumonisins. FA) accumulation of free sphingoid bases. especially Sa. and the had no significant effect on Sa content or the Sa:So ratio at depletion of complex sphingolipids. such as sphingomyelin any of the concentrations tested. __ and glycosphingolipids. Toxicity and effects on sphingoid bases ofselected myco­ Since free sphinganine is normally present in cells in very toxins. The ability of several structurally unrelated toxins low (picomoles/gram tissue) concentrations (Merrill et al.. and sphinganine to elevate. sphingoid bases was tested at 1986; Riley et al.. 1993) and since fumonisin causes much two dose levels. 10 and 100 pM. by incubation with liver larger accumulation of Sa than So. the elevation of the ratio, slices for 20 hr. Of the compounds tested. only fumonisin Sa:So. has been suggested as a biomarker for fumonisin B) and sphinganine (D-eT)'thro-dihydrosphingosine). itself. exposure and contamination (Riley et al.. 1993, 1994), and FUMONlSIN SAR IN LIVER SLICES 67

10 ,------,

.Qco Dose of Toxin: c: '~~'''*~4 10 I.l M OJ 'wc: o - 100I.lM OJ c: ~ (/)

0.01

FIG. 3. Effect of incubation of rat liver slices for 20 hr with selected mycotoxins or sphinganine (o-erythro-dihydrosphinganine) on the ratio of the free sphingoid bases. Bars with asterisks indicate ratio values are significantly different (p < 0.05) from that of DMSO-treated controls. a United States patent has been approved for such use (Mer­ dieting relative toxicity of the analogs. Furthermore, since rill et aI., 1996c). The results of the present study support the bioassay may be a useful method for detecting fumoni­ the use ofthe ratio as a more sensitive indicator of fumonisin sins in corn or corn-based products or for evaluating poten­ exposure, since the low dose (0.05 f-LM) of several of the tial detoxification methods, it was important to establish that analogs tested significantly elevated the Sa:So ratio, but not the elevation in sphingoid bases is due to ability of a com­ the concentration of free Sa. Furthermore, the experimental pound to inhibit ceramide synthase and that it is not a non­ variability as measured by standard deviation was less for specific response to compounds that produce toxicity by Sa:So values than for sphinganine values (Table 1), an obser­ other mechanisms. Therefore. other mycotoxins bearing no vation in agreement with that reported by Riley et al. (1994). structural similarity to FB, were tested at relatively high As previously reported for hepatocytes (Wang et al., 1991). doses (10 and 100 f-LM) for their ability to block ceramide levels of free sphingosine in liver slices were not signifi­ synthase. cantly elevated by fumonisin treatment. Thus the elevation The results of this study demonstrate that a number of in Sa:So ratio was the parameter we used to evaluate the structural analogs of have similar abilities to ability of the various mycotoxins we tested for ability to elevate Sa and the Sa:So ratio. The results also suggest that disrupt sphingolipid biosynthesis. base hydrolysis to remove the tricarballylic side chains re­ We have developed a bioassay system based on the use duces the biological activity and acetylation of the amino of precision-cut tissue slices to determine the ability of fu­ group eliminates the activity. This is consistent with the monisin B, or extracts of fumonisin-containing corn to ele­ hypothesis that fumonisin interacts with both the binding vate Sa:So. The method was found to be relatively quick site for sphinganine and the site for the fatty acyl-CoAs of and highly sensitive (Norred et aI., 1996). In the present ceramide synthase (Merrill et aI., I966b). Incubation of liver studies, we used the bioass.ay to evaluate the relative ability slices with sphinganine (D-erythro-dihydroceramide) itself of analogs of fumonisin B, to inhibit ceramide synthase in also resulted in elevated levels of free sphinganine in the order to provide information about structure-activity rela­ cells, probably as a result of uptake of the sphinganine from tionships of fumonisins and thus provide a basis for pre- the surrounding medium. The most potent inhibitor of cera- 68 NORRED ET AL.

TABLE 2 analogs relative to each other. could be related to different Cytotoxic Response of Rat Liver Slices Incubated degrees of membrane permeability and thereby differences with Various Toxins for 20 hr in accessibility of the toxins to ceramide synthase sites within the liver cells. Lipid solubility per se does not appear MTT metabolism" to correlate with the ability of the various fumonisins to Concentration (A~71/mg inhibit ceramide synthase. The log P value of FB I which Toxin (pM) protein:: SD) we determined experimentally compared favorably with the computer-generated value and reflected the high degree of Aflatoxin BI 10 0.032 ::: 0.010* water solubility of this zwitterionic toxin. Hydrolyzed FB" Aflatoxin BI 100 0.057 :: 0.023* Cyclopiazonic acid 10 0.317 :: 0.261 which had a calculated log P of 0.61. indicating approxi­ Cyclopiazonic acid 100 0.242 :: 0.175 mately 100-fold greater lipophilicity than FB, , was actually Beauvericin 10 0.280 :: 0.312 less than half as effective as FB, in inhibition of ceramide Beauvericin 100 0.112:: 0.104 formation. Fumonisin B4 • which has two less hydroxyl D-eryrhro-Dihydrosphingl)sine 10 0.274 :: 0.225 groups that FBI and a calculated log P of 2.880. had about 100 0.202 ::!: 0.139 the same ceramide synthase inhibition activity as FB,. Fumonisin B I 10 0.468 ::!: 0.348 Fumonisin B,< 100 0.299 ::!: 0.170 Since fumonisins are competitive inhibitors of the syn­ T-2 toxin 10 0.035 ::!: 0.022* thase enzyme (Merrill et al.. 1996b), the analogs may have T-2 toxin 100 0.078 ::!: 0.060* different dissociation constants. That the acetylated FA! was Sterigmatocystin 10 0.048 :: 0.023* unable to block ceramide synthase at all indicates that the Sterigmatocystin 100 0.024 ::!: 0.009* Luteoskyrin 10 0.223 :: 0.163 amino group is essential for the biological activity offumoni­ Luteoskyrin 100 0.290 ::!: 0.081 sins, as suggested by Merrill et al. (1995b). Gelderblom et Verrucarin A 10 0.254 :: 0.239 al. (1993) found that fumonisins B, and B~ were cytotoxic Verrucarin A 100 0.051 :: 0.054* (at high doses of 1-2 mM) to rat liver hepatocytes and also Scirpentriol 10 0.034 ::: 0.009· Scirpentriol 100 0.052 :: 0.039* produced positive responses in a short-term carcinogenesis Zearalenone 10 0.476 :: 0.373 bioassay in rats. The acetylated forms, FA! and FA~. were Zearalenone 100 0.4 17 :: 0.283 neither as cytotoxic nor were they positive in the carcinogen­ DMSO 10 pi 0.398 ::!: 0.317 esis bioassay, thus lending further support to the conclusion Triton X-IOO 10 ill 0.017 ::!: 0.009* that the amino group is necessary for toxicity. It should be " Values are means::: SD of three replicate experiments. noted here that in our study, initial bioassay of FA! produced • Significantly different (p < 0.05) from DMSO-treated control slices. a positive response; i.e., Sa content of slices and Sa:So were elevated. Mass spectrometric analysis indicated that the sam­ ple contained a small amount of FB, and possibly other mide synthase was the Altemaria altemata toxin. TA toxin fumonisins (Poling and Plattner, unpublished observation). (Table 1). TA toxin has a number of structural differences Additional cleanup procedures were required to remove from that of FB,. most notably a shorter hydrocarbon chain these contaminants (Poling and Plattner, 1996), after which backbone than fumonisin. positioning of the amino group at the material was negative in the bioassay. Thus, caution CI instead of C2. and the absence of one of the TCA groups should be taken if future toxicity studies are undertaken with (Fig. 3). Under the conditions of our study. we were not FA 1 to ensure that all contaminating fumonisins are removed. able to differentiate between the potency of fumonisins B,. Further studies are needed to more fully define those parame­ ters that determine structure-activity relationships of this B> B,. or C4 to elevate the Sa:So ratio. It is possible that by replicating the experiments many more times. statistical family of mycotoxins. differences in the potencies of these analogs would be mani­ Nixtamalization is the process by which corn is con­ fested. However. the important consideration is that these verted to masa, used in the preparation of tortillas and other fumonisins all cause very large increases in free Sa relative foods. The process involves soaking corn in heated base to that found in control slices and therefore would be ex­ (calcium hydroxide) solution and has been shown to con­ pected to be equipotent in toxicity assuming they are ab­ vert fumonisins in the corn to TeA-hydrolyzed by-products sorbed and distributed in similar manners. Indeed, Cawood (Hendrich et al.. 1993). Two studies have reported that et al. (1994) found that the lowest levels of FB I and FB~ rats fed nixtamalized corn that was cultured with either F. that bound to hepatocytes and elicited cytotoxicity were sim­ moniliforme or F. proliferatum had less severe signs of ilar for the two analogs. anc;i .therefore the effective dose toxicity, including nephrosis and hepatosis, than did rats levels were not different. fed the same cultured corn which was not nixtamalized The reduced ability of hydrolyzed fumonisins to block (Hendrich et aI., 1993; Voss et al.. 1996). The results of ceramide synthase. and the different potencies of the other our study with hydrolyzed fumonisins suggest a reduction, FUMONISIN SAR IN LIVER SLICES 69 but not elimination of the biological activity of fumonisins. Cawood. M. E.. Gelderblom, W. C. A.. Alberts. J. F.. and Snyman. S. D. and thus are consistent with the in vivo investigations that (1994). Interaction of '·C-Iabelled fumonisin B, mycotoxins with primary rat hepatocytes. Food Chem. Toxicol. 32, 627-632. have been reported. Of interest in this regard is the recent Chamberlain. W. J.. Voss. K. A.. and Norred. W. P. (1993). Analysis of finding in our laboratory that a positive correlation existed commercial laboratory rat rations for fumonisin B,. a mycotoxin pro­ between the degree of elevation of tissue Sa and Sa:So and duced on com by Fusarium moniliforme. Conremp. Topics Lab. Anim. severity of toxic effects (as measured by reduced body Sci. 32, 26-28. weight gains, elevated serum chemistry values, kidney to Chu. F. S.. and Li. G. Y. (1994). Simultaneous occurrence offurnonisin B, brain or liver to brain weight ratios. and extent of micro­ and other mycotoxins in moldy com collected from the People's Republic scopic lesions in liver and kidney) observed in rats fed of China in regions with high incidences of esophageal cancer. Appl. Em·iron. Microbiol. 60,847-852. diets containing F. moniliforme culture material (CM). nix­ Dutton. M. F. (1996). Fumonisins. mycotoxins of increasing importance: tamalized CM. water extracted CM, or fumonisin-free seed Their nature and their effects. Pharmacol. Ther. 70, 137-161. corn (K. A. Voss. personal communication). Fukuda. H.. Shima. H., Vesonder, R. F.. Tokuda, H.. Nishino. H.. Katoh. Mycotoxins that are structurally dissimilar to fumonisins S.. Tamura, S., Sugimura, T., and Nagao. M. (1996). Inhibition of protein did not inhibit ceramide synthase, even at doses that caused serine threonine phosphatases by fumonisin B,. a mycotoxin. Biochem. overt toxicity (Fig. 2. Table 2). This finding provides evi­ Biophys. Res. Cammun. 220, 160-165. dence that the disruption of sphingolipid metabolism by inhi­ Gelderblom. W. C. A.. Cawood. M. E.. Snyman. S. D.. Vleggaar. R.. and bition of ceramide synthase is specific to fumonisins, and Marasas. W. F. O. (1993). Structure-activity relationships of fumonisins in short-term carcinogenesis and cytotoxicity assays. Faod Chem. Tax· structurally related compounds such as TA toxin. However, icol. 31,407-414. at least one fungal metabolite with little structural resem­ Gelderblom. W. C. A.. Jaskiewicz, K., Marasas. W. F. 0 .. Thiel. P. G.. blance to fumonisins. australifungin. has been reported to be Horak. R. M.. Vleggaar. R.. and Kriek. N. P. J. (1988). Fumonisins­ an inhibitor...{Mandala et al.. 1995), and it is conceivable novel mycotoxins with cancer-promoting activity produced by Fusarium that there are other. as yet undiscovered, agents that disrupt maniliforme. Appl. Environ. Microbiol. 54, 1806-1811. sphingolipid metabolism by this mechanism. We are cur­ Gelderblom. W. C. A.. Kriek. N. P. 1.. Marasas. W. F. 0 .. and Thiel. P. G. rently using the liver slice bioassay system to screen a variety (1991 ). Toxicity and carcinogenicity of the Fusarium monilifomle metab­ olite. fumonisin B,. in rats. Carcinogenesis 12,1247-1251. of fungal cultures for ceramide synthase inhibition activity. Harrison. L. R.. Colvin. B. M., Greene. 1. T.. Newman. L. E.. and Cole, In summary, the results of this investigation suggest that J. R. (1990). Pulmonary edema and hydrothorax in swine produced by the naturally occurring fumonisins (B" B1 , B), B4 , C4 ,) and fumonisin B,. a toxic metabolite of Fusarium moniliforme. J. Vet. Diagn. TA toxin are approximately equipotent in their ability to Im'est. 2, 2 I7-221. disrupt sphingolipid metabolism and would be expected to Hendrich. S.. Miller. K. A., Wilson, T. M.. and Murphy. P. A. (1993). Tox­ have similar degrees of toxicity if they also share similar icity of Fusarium proliferatum-fermented nixtamalized com-based diets toxicokinetic properties. Removal of the tricarballylic acid fed to rats: Effect of nutritional status. J. Agric. Food Chem. 41, 1649­ groups. as occurs by processing procedures such as nixtamal­ 1654. ization. reduces but does not eliminate the biological activity, Huang. c.. Dickman. M.. Henderson. G.. and Jones. C. (1995). Repression of protein kinase C and stimulation of cyclic AMP response elements by and apparently, as shown in other studies. the toxicity as fumonisin. a fungal encoded toxin which is a carcinogen. Cancer Res. well. The results also demonstrate that the primary amino 55,1655-1659. group of fumonisin-like compounds is necessary for cera­ Logrieco. A.. Moretti. A.. Ritieni. A.. Bottalico. A.. and Corda. P. (1995). mide synthase inhibition, since acetylation of FB) to FA) Occurrence and toxigenicity of Fusarium proliferatum from preharvest completely inhibited the response. Finally, the results dem­ maize ear ro!. and associated mycotoxins. in Italy. Dis. 79,727­ onstrate that the action of fumonisins and similar compounds 731. on ceramide synthase is specific and is not a generalized Lowry. O.H.. Rosebrough. N.1.. Farr. A.L.. and Randall. R.J. (1951). Protein measurement with the Folin phenol reagent. J. BioI. Chem. 193, response of cells to cytotoxic compounds such as aflatoxin. 263-275. and other mycotoxins. Mandala. S. M.. Thornton. R. A.. Fromer. B. R.. Curotto. 1. E., Rozdilsky. W.. Kurtz. M. B.. Giacobbe, R. A.. Bills. G. F.. Cabello. M. A.. Martin. ACKNOWLEDGMENTS I.. Pelaez, F.. and Harris. G. H. (1995). The discovery of australifungin. a novel inhibitor of sphinganine N-acyltransferase from Sporormiella The authors thank Pamela J. Malcom for excellent technical assistance austalis. J. Antibiotics 48, 349-356. and Steve M. Poling for assistance in purification of fumonisin analogs. D­ Marasas. W. F. 0., Kellerman. T. S.• Ge1derblorn. W. C. A., Coetzer, er\'lhro-dihydrosphingosine and synthetic- ClO-sphinganine were generous J. A. W.. Thiel. P. G., and van der Lugt. J. J. (1988). Leukoencephaloma­ gifts of Dr. A. H. Merrill. Jr.. Emory University. lacia in a horse induced by fumonisin B, isolated from Fusarium monili­ forme. Onderstepoort J. Vet. Res. 55, 197-203. REFERENCES Meredith. F. I.. Bacon, C. W., Norred. W. P.. and Plattner. R. D. (I 996a). Isolation and purification of fumonisin B, and B2 from rice culture. In Branham. B. E., and Plattner, R. D. (1993). Isolation and characterization Fumonisins in Food (L. Jackson, J. W. DeVries, and L. B. Bullerman. of a new fumonisin from liquid cultures of Fusarium moniliforme. J. Eds.), pp. 113-122. Plenum, New York. NY. Nat. Products 56, 1630-1633. Meredith. F. r.. Bacon. C. W.. Plattner. R. D.. and Norred. W. P. (1996b). 70 NORRED ET AL.

Preparative LC isolation and purification of fumonisin B, from rice cul­ Riley, R. T.. An, N.-H.. Showker. J. L., Yoo. H.-S.. Norred. W. P.. Cham· ture. J. Agric. Food Chem. 44,195-198. berlain. W. J.• Wang, E.. Merrill. A. H.• Jr.. Motelin. G.• Beasley. V. R.. Merrill. A. H.. Jr.. Liona. D. C. and Riley, R. T. (I996a). Fumonisins: and Haschek, W. M. (1993). Alteration of tissue and serum sphinganine Fungal toxins that shed light on sphingolipid function. Trends Cell BioI. to sphingosine ratio: An early biomarker in pigs ofexposure to fumonisin­ 6, 218-223. containing feeds. Toxicol. Appl. Pharmacol. 118, 105-112. Merrill. A. H.. Jr.. Schmelz. E.-M.. Wang, E.. Schroeder, J.1.. Dillehay, Riley, R. T., Wang. E.• and Merrill. A. H.. Jr. (1994). Liquid chromato­ D. L.. and Riley. R. T. (I 995a). Role of dietary sphingolipids and inhibi­ graphic determination of sphinganine and sphingosine: Use of the free tors of sphingolipid metabolism in cancer and other diseases. J. Nutr. sphinganine-to-sphingosine ratio as a biomarker for consumption of fu­ 125(Suppl.). 1677S-1682S. monisins. J. Assoc. Off. Anal. Chern. Int. 77, 533-540. Merrill. A. H.. Jr., Sereni. A. M.. Stevens, V. L.. Hannun, Y. A.. Bell, R. M., Riley, R. T., Wang, E., Schroeder, J. 1.. Smith, E. R., Plattner, R. D.. Abbas. and Kinkade. J. M.. Jr. (1986). Inhibition of phorbol ester-dependent H., Yoo, H.-S., and Merrill, A. H., Jr. (1996). Evidence for disruption differentiation of human promyelocytic leukemic (HL-60) cells by sphin­ of sphingolipid metabolism as a contributing factor in the toxicity and ganine and other long-chain bases. J. Bioi. Chern. 261, 12610-12615. carcinogenicity of fumonisins. Nat. Toxins 4, 3-15. Merrill. A. H., Jr.. Wang, E.. Gilchrist, D. G., and Riley, R. T. (1993). Ross, P. F.• Rice, L. G., Plattner, R. D.. Osweiler. G. D., Wilson. T. M.. Fumonisins and other inhibitors of de novo sphingolipid biosynthesis. Owens, D. L., Nelson. H. A.. and Richard. J. L. (1991). Concentrations Adv. Lipid Res. 26,215-231. of fumonisin B1 in feeds associated with animal health problems. Myco· Merrill. A. H.. Jr.. Wang. E.. Schroeder. J. 1.. Smith, E. R., Yoo, H.-S.. and pathologia 114, 129-135. Riley. R. T. (1995b). Disruption of sphingolipid metabolism in the toxic­ Schroeder, J. J., Crane, H. M., Xia, J., Liona, D. C, and Merrill. A. H., Jr. ity and carcinogenicity of fumonisins. In Molecular Approaches to Food (1994). Disruption of sphingolipid metabolism and stimulation of DNA

Safety: Issues Involving Toxic Microorganisms (M. Eklund, J. L. Richard, synthesis by fumonisin B 1 : A molecular mechanism for carcinogenesis and K. Mise. Eds.). pp. 429-443. Alaken. Fon Collins, CO. associated with Fusarium moniliforme. J. BioI. Chem. 269,3475-3481. Merrill, A. H.. Jr., Wang. E., Vales, T. R., Smith, E. R.• Schroeder. J. 1.. Smith, E. R., and Merrill, A. H., Jr. (1995). Differential roles of de novo Menaldino, D. S.. Alexander. C.. Crane. H. M., Xia, 1.. Liona. D. C. sphingolipid biosynthesis and turnover in the "burst" offree sphingosine Meredith. F.I".and Riley, R. T. (1996b). Fumonisin toxicity and sphin­ and sphinganine, and their I-phosphates and N-acyl-derivatives, that oc­ golipid biosynthesis. In Fumonisins in Food (L. Jackson. J. DeVries, and curs upon changing the medium of cells in culture. J. Bioi. Chem. 270, L. Bullerman. Eds.), pp. 297-306. Plenum. New York. NY. 18749-18758. Merrill, A. H.• Jr., Wang. E. W.. and Riley, R. T. (1996c). Method of alter­ Sydenham. E. W., Shephard, G. S., Thiel, P. G., Marasas, W. F. 0., and ing sphingolipid metabolism and detecting fumonisin ingestion and con· Stockenstrom, S. (1991). Fumonisin contamination of commercial com­ tamination. Off. Ga~etIe U.S. Pat. Trademark Off. 1186, 1979- I980. based human foodstuffs. J. Agric. Food Chem. 39, 2014-2018. Mosmann, T. (1983). Rapid colorimetric assay for cellular growth and Sydenham, E. W., Thiel, P. G., Marasas, W. F. 0., Shephard, G. S., Van survival: Application to proliferation and cytotoxicity assays. J. Immunol. Schalkwyk, D. J., and Koch, K. R. (1990). Natural occurrence of some Methods 65, 55-63. Fusarium mycotoxins in corn from low and high esophageal cancer prev­ Nelson, P. E.. Plattner. R. D., Shackelford, D. D.• and Desjardins, A. E. alence areas of the Transkei, South Africa. J. Agric. Food Chem. 38, (1991). Production of fumonisins by Fusarium moniliforme strains from 1900-1903. various substrates and geographic areas. Appl. Environ. Microbiol. 57, Tolleson, W. H., Dooley, K. L.. Sheldon, W. G.• Thurman, J. D.• Bucci. 2410-2412. T. J., and Howard, P. C. (1996). The mycotoxin fumonisin induces Norred, W. P. (1993). Fumonisins-mycotoxins produced by Fusarium apoptosis in cultured human cells and in livers and kidneys of rats. Adv. moniliforme. J. Toxicol. Environ. Health 38, 309-328. £rp. Med. BioI. 392,237-250. Norred. W. P.. Riley. R. T., Meredith. F. 1.. Bacon. C. W., and Voss, K. A. Voss. K. A.. Bacon, C W., Meredith. F. 1., and Norred. W. P. (1996). Com­ (1996). Time- and dose-response effect of the mycotoxin. fumonisin parative subchronic toxicity studies of nixtamalized and water-extracted B,. on sphingoid base elevations in precision-cut rat liver and kidney Fusarium moniliforme culture material. Food Chem. Toxicol. 34, 623­ slices. Toxicol. In Vitro 10, 349-358. 632. Norred, W. P.. and Voss, K. A. (1994). Toxicity and role of fumonisins in Voss, K. A.• Chamberlain. W. J.• Bacon. C W., and Norred. W. P. (1993). animal diseases and human esophageal cancer. J. Food Protect. 57,522­ A preliminary investigation on renal and hepatic toxicity in rats fed 527. purified fumonisin B,. Nat. Toxins 1,222-228. Planner, R. D.. Weisleder, D.. Shackelford, D. D., Peterson, R.. and Powell. Voss. K. A., Chamberlain. W. J.• Bacon, C W.• Herbert, R. A., Walters. R. G. (1992). A new fumonisin from solid cultures of Fusarium monili­ D. B.. and Norred, W. P. (1994). Subchronic feeding study of the myco­ forme. Mycopathologia 117,23-28. toxin fumonisin B, in B6C3FI mice and Fischer 344 rats. Fundam. Appl. Poling. S. M.. and Plattner. R. D. (19%). Rapid purification of fumonisins Toxicol. 24, 102- 110. B, and B. with solid phase extraction columns. J. Agric. Food Chem. Wang. E.. Norred. W. P.. Bacon. C W.• Riley. R. T., and Merrill. A. H.• 44, 2792-2796. Jr. (1991 ). Inhibition of sphingolipid biosynthesis by fumonisins: implica­ Powell, R. G.. and Plattner. R. D. (1995). Fumonisins. In Alkaloids: Chemi­ tions for diseases associated with Fusarium moniliforme. J. Bioi. Chem. cal and Biological Perspectives (S. W. Pelletier. Ed.), pp. 247-278. 266, 14486-14490. Pergamon. Tarrytown. NY. Wang, H., Jones. C. Ciacci-Zanella. J.• Holt, T.• Gilchrist, D. G.. and Ramasamy. S.. Wang, E.. Hennig, B., and Merrill, A. H.. Jr. (1995). Fu­ Dickman. M. B. (1996). Fumonisins and Alternaria altemata lycopersici monisin B, alters sphingolipid metabolism and disrupts the bamer function toxins: Sphinganine analog mycotoxins induce apoptosis in monkey kid­ of endothelial cells in culture. Toxicol. Appl. Pharmacol. 133, 343-348. ney cells. Proc. Nat. Acad. Sci. USA 93, 3461-3465. Rheeder. J. P.. Marasas, W. F. 0 .. Thiel, P. G.. Sydenham, E. W.. Shephard, Yoo, H.-S.. Norred, W. P., Showker, 1.. and Riley. R. T. (1996). Elevated G. S.. and Van Schalkwyk, D. J. (1992). Fusarium moniliforme and fu­ sphingoid bases and complex sphingolipid depletion as contributing fac­ monisins in com in relation to human esophageal cancer in Transkei. tors in fumonisin-induced cytotoxicity. Toxicol. Appl. Pharmacol. 138, Phytopathology 82, 353-357. 211-218.

Supplied by U.S. Dept of Agriculture National Center for Agricultural Utilization Research, Peoria, IlHnois