Purified Fumonisin B1 Decreases Cardiovascular

56, 240 –249 (2000) Copyright © 2000 by the Society of Toxicology

TOXICOLOGICAL SCIENCES

Purified Fumonisin B 1 Decreases Cardiovascular Function but Does Not Alter Pulmonary Capillary Permeability in Swine Geoffrey W. Smith,* ,† ,1 Peter D. Constable,* ,† Robert M. Eppley,‡ Mike E. Tumbleson,§ Laura A. Gumprecht,† ,2 and Wanda M. Haschek-Hock† Departments of *Veterinary Clinical Medicine, †Veterinary Pathobiology, and §Veterinary Biosciences, College of Veterinary Medicine, University of Illinois, Urbana, Illinois 61802; and ‡Center for Food Safety and Applied Nutrition, Food and Drug Administration, Laurel, Maryland 20708 Received January 27, 2000; accepted April 7, 2000

Fumonisins are mycotoxins produced by Fusarium verticillioides, which induce acute pulmonary edema in swine. We previously reported that ingestion of fumonisin-containing culture material decreases cardiovascular function in swine (1996,a,b; Fundam. Appl. Toxicol. 31, 169 –172; 33, 140 –148; 1999, Am. J Vet. Res. 60, 1291–1300). The main purpose of this study was to confirm that fumonisin B 1 was responsible for the observed cardiovascular changes. Treated pigs (n ⴝ 6) were given daily intravenous injections of purified fumonisin B 1 at 1 mg/kg for 4 days, while controls (n ⴝ 6) were injected with equal volumes of saline. On day 5, pigs were anesthetized with butorphanol-chloralose and instrumented for hemodynamic studies. Terminally, bronchoalveolar lavage was performed on each pig to determine the relative permeability index of the pulmonary endothelium. Fumonisin B 1-treated pigs had marked decreases in the maximal rate of change of left ventricular pressure (dP/dt max), mean aortic pressure, cardiac output, and arterial pO 2, accompanied by increases in mean pulmonary artery pressure, oxygen extraction ratio, and blood hemoglobin concentration. Plasma and left ventricular sphingosine and sphinganine concentrations were markedly increased in treated pigs at day 5; however, there was no difference in the relative permeability index between groups. Serum cholesterol concentrations and activities of hepatic-derived enzymes were increased, and hepatocyte apoptosis and mitoses were present in the livers of fumonisin-treated pigs. In the lungs of treated pigs, there was proteinaceous edema and membranous accumulations in capillary endothelial cells. These results indicate that cardiovascular function is altered by fumonisin B 1, and that fumonisin-induced pulmonary edema is caused by left-sided heart failure and not by altered endothelial permeability. Because of the potential for contamination of human foodstuffs by fumonisins, the cardiovascular toxicity of these compounds must be taken into consideration. Key Words: fumonisin; cardiovascular; heart failure; pulmonary edema; swine.

1 To whom correspondence should be addressed at the Department of Veterinary Clinical Medicine, University of Illinois, 1008 West Hazelwood Drive, Urbana, IL 61802. Fax: (217) 244-7421. E-mail: [email protected]. 2 Present address: Department of Safety Assessment, Merck Research Laboratories, West Point, PA 19486.

Fumonisins are a group of mycotoxins primarily produced by the fungus Fusarium verticillioides (formerly F. moniliforme), one of the most prevalent fungi associated with corn intended for human and animal consumption (Marasas et al., 1984). Since their initial isolation in the late 1980⬘s, these toxic fungal metabolites have been implicated in naturally occurring outbreaks of porcine pulmonary edema (Harrison et al., 1990; Osweiler et al., 1992) and equine leukoencephalomalacia (Wilson et al., 1990), and have also been linked epidemiologically with human esophageal cancer (Rheeder et al., 1992). We previously reported that ingestion of fumonisin-containing culture material decreases cardiac contractility, heart rate, cardiac output, mean arterial pressure, arterial and mixed venous blood oxygen tensions, and systemic oxygen delivery. In addition, it increases mean pulmonary artery pressure, pulmonary artery wedge pressure, oxygen consumption, and oxygen extraction ratio in swine (Constable et al., 1999; Smith et al., 1996a,b; Smith et al., 1999). Our findings suggested that fumonisininduced pulmonary edema in swine was due to acute left-sided heart failure. However, in vitro (Ramasamy et al., 1995) and ultrastructural studies (Gumprecht et al., 1998) indicated that fumonisins also induce pulmonary endothelial damage, thereby increasing pulmonary capillary permeability and potentially resulting in pulmonary edema. Although pulmonary edema in swine has been experimentally reproduced using purified fumonisin B 1 (Harrison et al., 1990; Haschek et al., 1992), all of our previous studies examining the cardiovascular effects of fumonisin in swine (Constable et al., 1999; Smith et al., 1996a,b; Smth et al., 1999) have used culture material prepared from strains of Fusarium verticillioides. In addition to fumonisin B 1, this fungus is known to produce other toxins such as fumonisin B 2, moniliformin, fusarin A, C, and F, and fusaric acid (Bacon and Nelson, 1994; Savard and Miller, 1992), beauvericin (Gupta et al., 1991) and fusaproliferin (Ritieni et al., 1995). Moniliformin is cardiotoxic in poultry (Nagaraj et al., 1996); however, none of the culture material samples that we have used contained detectable levels of this toxin (detection limit ⫽ 0.5 ppm). Fusaric acid (5-butylpicolinic acid) also has profound

240

PURIFIED FUMONISIN B 1 DECREASES CARDIOVASCULAR FUNCTION IN SWINE

effects on the cardiovascular system, as a potent inhibitor of dopamine ␤-hydroxylase in vivo and in vitro (Nagatsu et al., 1970). Intraperitoneal injection of 20 mg/kg fusaric acid in rabbits, rats, cats, and dogs produced profound and persistent decreases in mean arterial pressure (Hidaka et al., 1969), and oral administration of fusaric acid to humans decreased mean arterial pressure because of a decrease in endogenous norepinephrine concentrations (Nagatsu et al., 1970). The systemic hypotensive response to fusaric acid is similar to that observed in pigs fed culture material containing fumonisins (Smith et al., 1996b; Smith et al., 1999). In fact, the hypotensive effects of fusaric acid were so profound that it was once clinically investigated as an antihypertensive drug in humans (Terasawa and Kameyama, 1971). We recently analyzed samples of the fumonisin-containing culture material used in our previous studies and found them to contain fusaric acid at concentrations of 5.7 ppm (Smith et al., 1996a) and 1.7 ppm (Smith et al., 1996b). These fusaric acid concentrations are considered very low (Bacon et al., 1996; Smith and Graca-Sousadias, 1993). However, it was important to verify that the cardiovascular depression observed in our previous studies was actually due to fumonisin B 1 and not fusaric acid or another unidentified mycotoxin. Therefore the aims of this study were to confirm that the cardiovascular changes we observed following ingestion of fumonisin-containing culture material were due to fumonisin B 1 and not another toxin or toxins, and to determine whether fumonisin B 1 increased pulmonary capillary permeability in swine.

MATERIALS AND METHODS Animals and toxin isolation. Our institutional committee on the care and use of laboratory animals approved this study. Twelve healthy, castrated male crossbred pigs weighing 25 ⫾ 3 kg (range 22 to 29 kg) were housed individually in stalls beginning 5 days prior to instrumentation. Pigs had free access to water and were fed an 18% protein complete grower diet that was free of fumonisin B 1, fumonisin B 2 (detection limit ⫽ 0.5 ppm for both fumonisin B 1 and B 2 by HPLC), aflatoxin (detection limit ⫽ 3 ppb by HPLC), zearalenone (detection limit ⫽ 1 ppm by HPLC), ochratoxin (detection limit ⫽ 50 ppb by HPLC), vomitoxin (detection limit ⫽ 1 ppm by gas chromatography), and T-2 toxin (detection limit ⫽ 1 ppm by thin-layer chromatography) on analysis at the University of Illinois Laboratory of Veterinary Diagnostic Medicine. The HPLC technique for fumonisin determination has been previously described (Bennett and Richard, 1994). Fumonisin B 1 was isolated from Fusarium proliferatum grown on whole corn. This fungus is not as prevalent as F. verticillioides, but has been shown to produce fumonisin B 1. The toxin was extracted from the culture medium with methanol/water (7:3 by volume) and purified by preparatory liquid column chromotography. It was purified using a series of preparative columns, alternating between C-18 and cyano phases. Water was removed from the purified fumonisin B 1 by freeze-drying. The purity of the final product (as the free acid) was determined to be greater than 95% by analytical liquid chromatography, nuclear magnetic resonance, and mass spectral analysis. The major impurities were determined to be the partial hydrolysis, methyl ester, and amide derivatives of fumonisin B 1. For this study, quantities of toxin were weighed and dissolved in phosphate-buffered saline to yield a fumonisin B 1 concentration of 3 mg/ml (pH 6.8).

241

Experimental protocol. At the end of the 5-day acclimation period, pigs were held off feed and water overnight. The following day, general anesthesia was induced with an intramuscular injection of xylazine (2 mg/kg) and ketamine hydrochloride (10 mg/kg) followed by mask induction with 3 to 5% halothane in 100% O 2. Pigs were orotracheally intubated, placed in dorsal recumbency on a water-circulating heating blanket, maintained at 1.5% halothane in 100% O 2 in a semi-closed circuit, and allowed to spontaneously ventilate. The right jugular vein was identified by surgical cutdown using aseptic technique and a 16-gauge polyethylene catheter (Arrow International, Reading, PA) was inserted. The catheter was secured around the neck and pigs were returned to their stalls and allowed to recover from anesthesia for 24 h. Following anesthetic recovery, a blood sample was drawn for sphingolipid analyses and pigs were randomly assigned to 2 groups. Treated pigs (n ⫽ 6) were given daily intravenous injections of 1 mg fumonisin B 1 per kg of body weight for 4 days, while control pigs (n ⫽ 6) were given an equivalent volume of saline. Pharmacokinetic studies of fumonisin B 1 indicate that the bioavailability of orally administered fumonisin is 3 to 6% in swine, and that intravenously administered fumonisin is rapidly cleared and does not bind in appreciable amounts to plasma proteins (5% bound) (Prelusky et al., 1994). Assuming a bioavailability of 5%, a daily intravenous dose of 1-mg fumonisin B 1/kg is equivalent to a daily oral dose of 20 mg/kg. This fumonisin dose causes pulmonary edema and death within 5 days of initial exposure in our laboratory (Smith et al., 1999). Every 24 h, the resting respiratory rate was determined and pigs were placed sternally in a sling. Pigs were given either fumonisin B 1 or saline intravenously and catheters were filled with heparinized saline (100 IU/ml) to prevent thrombosis. On day 5, pigs were again anesthetized using the same anesthetic protocol as above, but were ventilated with supplemental oxygen (4 L/min) on a volume respirator (Harvard Apparatus Co. Inc, Dover, MS) at a tidal volume of 10 ml/kg and respiratory frequency of 20 breaths/minute. Blood samples were drawn for serum biochemical and sphingolipid analyses, and arterial and mixed venous blood-gas samples were collected at the beginning of the anesthesia, after pigs had been ventilated with room air instead of 100% oxygen for 5 min. Halothane was discontinued and anesthesia subsequently maintained by IV ␣-chloralose administration (50 mg/kg initially, then 15 mg/kg –1/h –1, IV; Sigma Chemical Co, St. Louis, MO) and butorphanol (0.5 mg/kg, IM, every 3 h). This anesthetic protocol produces minimal myocardial depression in swine (Constable et al., 2000) and has been recommended when prolonged anesthesia is required for studies of cardiovascular physiology (Lang et al., 1992). Arterial pH, pCO 2, and pO 2 were measured periodically during instrumentation and maintained within normal limits (pH, 7.40 to 7.50; pCO 2, 35 to 45 mm Hg; pO 2 ⬎ 100 mm Hg) by adjusting tidal volume and ventilatory frequency. Ventilator settings were not altered once instrumentation had been completed (approximately 2 h). Lactated Ringer’s solution (4 ml/kg –1/h –1) was administered intravenously for the duration of the study. The left jugular vein, left carotid artery, and left femoral vein were identified by surgical cutdowns and intravenous heparin (Elkins-Sinn, Inc, Cherry Hill, NJ) was administered (100 IU/kg initially, then 50 IU/kg, every 90 min). A polyethylene catheter (3-mm outside diameter) was placed in the femoral artery for measurement of mean arterial blood pressure and to obtain arterial blood for analysis. A 7-F Swan-Ganz thermodilution catheter (Baxter Healthcare Corp, Irvine, CA) was advanced through the left jugular vein so the distal port was in the pulmonary artery and the proximal port was in the cranial vena cava or right atrium. A 5-F catheter equipped with a tip micromanometer (Millar Instruments, Houston, TX) was advanced along the left carotid artery and positioned in the left ventricle for recording of left ventricular pressure. Correct catheter positions were determined by evaluation of pressure tracings, which were monitored continuously on a multi-channel strip-chart recorder (Gilson Medical Electronics, Middleton, WI), and fluid pressures were referenced to the center of the thorax. After instrumentation had been completed, pigs were monitored for 15 min to ensure hemodynamic stability. Cardiovascular measurements were then obtained. Measurements. Cardiac output was measured by the thermodilution technique with the aid of a CO computer (American Edwards Laboratories, Inc.,

242

SMITH ET AL.

Irvine, CA). Three milliliters of 5% dextrose (0°C) were injected rapidly into the proximal port of the Swan-Ganz catheter for CO measurement, and the mean value of 2 to 3 determinations was used as the experimental value for each period. Lead (II) electrocardiogram, left-ventricular pressure, mean arterial pressure (MAP), mean pulmonary artery pressure (MPAP), pulmonary artery wedge pressure (PAWP), and mean central venous pressure (CVP), were recorded on the strip-chart recorder. The electrocardiogram and left ventricular pressure signals were digitized at 500 Hz by a 12-bit microcomputer system, and data stored on the hard disk for subsequent analysis. Data to be digitized were recorded at end-expiration with the ventilator turned off. Serum obtained from blood taken on day 5 was analyzed for creatinine, urea nitrogen, total protein, albumin, calcium, phosphorus, sodium, potassium, chloride, glucose, cholesterol, and total bilirubin concentrations along with creatine kinase, alkaline phosphatase (ALP), gamma glutamyl transferase (GGT), aspartate aminotransferase, and sorbitol dehydrogenase (SDH) activities (Hitachi 704 Automated Chemistry Analyzer, Boehringer Mannheim Diagnostics, Indianapolis, IN). Blood pH, pO 2, and pCO 2 were measured in arterial and mixed venous blood-gas samples (Ciba-Corning 288 blood-gas system, Medfield, MA), and were corrected for blood temperature. Bicarbonate (HCO 3) and base excess values were calculated. Hemoglobin concentration was also measured using the same blood-gas analyzer. Systemic oxygen delivery, oxygen consumption, oxygen extraction ratio, the alveolar-arterial oxygen gradient, and the physiologic shunt to total blood flow ratio were calculated. Systemic O 2 delivery was calculated as the product of arterial O 2 content and cardiac output and normalized to body weight. Total blood-O 2 content was calculated to be 1.39 ml of O 2/g of hemoglobin plus dissolved O 2 equal to 0.3 volume %/100 mm of Hg. Mass specific O 2 consumption (V O2) was calculated by the difference between arterial (Ca O2) and mixed venous oxygen content (Cv O2), multiplied by the CO: V O2 ⫽ CO ⫻ (Ca O2 – Cv O2)/body weight. Systemic O 2 extraction ratio was calculated as the ratio of the arteriovenous O 2 content difference to the arterial O 2 content and expressed as a percentage. Room air alveolar-arterial O 2 gradient [PAO 2 – PaO 2] was calculated by use of the alveolar gas equation: PA O2 ⫽ PI O2 – (Pa CO2/R) ⫹ F, where PI O2 was the inspired partial pressure of oxygen calculated from the barometric pressure and PA O2 was the calculated alveolar O 2 tension. The respiratory exchange ratio (R) was assumed to equal 0.8, and F (a small correction factor) was ignored (West, 1987). The physiologic shunt to total blood flow ratio (Q sQ t) was calculated by use of the shunt equation: Q sQ t ⫽ (Ci O2 – Ca O2)/ (Ci O2 – Cv O2), where CiO 2 is the oxygen content of ideal end-pulmonary capillary blood, which is commonly considered to be equal to the alveolar oxygen concentration (West, 1987). Analysis. Left ventricular pressures data were analyzed off-line using a personal computer, Conduct-PC software (Leycom, The Netherlands), and custom-designed software. Left ventricular end-diastolic pressure (LVEDP) was defined as the ventricular pressure at the start of the R wave, and represented the mean value for 10 consecutive beats. The maximal rate of change of left ventricular pressure (dP/dt max) was determined using a 3-point Lagrangian interpolation on digitized pressure data smoothed with a 3-point moving average. Systemic vascular resistance (SVR) was calculated as SVR ⫽ (MAP – CVP)/CO, and pulmonary vascular resistance was calculated as PVR ⫽ (MPAP – LVEDP)/CO. Relative permeability index. At the completion of the study, a blood sample was collected and pigs were euthanized with an overdose of sodium pentobarbital (60 mg/kg, IV). The thorax was opened and the heart and lungs removed for determination of the relative permeability index (RPI) (Olson et al., 1985). The left lung was tied off and carefully excised at the hilus, so that no blood leaked retrograde into the bronchi. A 14-F Foley catheter was inserted into the bronchus of the left diaphragmatic lobe and sealed by inflating the balloon. Bronchoalveolar lavage was then performed with 3 20-ml aliquots of phosphate-buffered saline. Approximately 50% of the infused phosphatebuffered saline was recovered from each lavage and the 3 samples were pooled for analysis. An aliquot (2-ml) of fluid was used to determine the total number of polymorphonuclear cells and for differential determination of all types present. The remaining lavage fluid was centrifuged at 2,000 rpm for 20 min.

The supernatant was decanted and protein concentrations determined by automated methods (Hitachi 704 automatic analyzer, Hitachi, Tokyo, Japan). The RPI was calculated as RPI ⫽ (bronchoalveolar lavage protein concentration)/ (serum total protein concentration immediately before euthanasia). Sphingolipid analysis. Venous blood samples were collected in EDTA tubes on days 1 and 5 of the study, and samples of the left ventricle were obtained after euthanasia. Tissue and plasma samples were stored at –20°C and thawed immediately before determining free-sphinganine and sphingosine concentrations as described previously (Smith et al., 1999). Pathology and lung wet weight to dry weight ratio. Selected tissues (lung, liver, heart, and kidney) were immersion-fixed in 10% neutral buffered formalin for histopathology, and the right cranial lung lobe was fixed by intrabronchial instillation of formalin. Sections of lung and heart were also immersed in Karnovsky’s fixative for ultrastructural evaluation. Formalin-fixed tissues were processed routinely, and 3 to 5 ␮m paraffin-embedded sections were stained with hematoxylin and eosin. The number of hepatocytes undergoing apoptosis and mitoses was determined by light microscopy at 100⫻ in 5 nonsequential fields in each of 2 sections per pig. A section of the right diaphragmatic lung lobe was weighed, dried at 110°C for 48 h, and re-weighed for determination of wet and dry weights, and the lung wet-weight to dryweight ratio was calculated. Statistical analysis. Data are presented as mean ⫾ SD. A p value of ⬍0.05 was regarded as significant. One-way analysis of variance was used to compare baseline cardiovascular, blood gas, serum biochemical data, and lung wetweight to dry-weight ratios between control and treated pigs. Variables with non-normal distributions or unequal variances were log transformed or ranked before analysis of variance was performed. The frequency of hepatocyte apoptosis and mitosis was compared between control and treated pigs using the Mann-Whitney test. Two-way analysis of variance (group, time) with repeated measures on one factor (time) was used for comparison of fumonisin and control group data from the pharmacological studies. Appropriate post-tests were conducted whenever the F test was significant. Multiple pairwise comparisons were conducted between groups and within each series of drug administration, using the Bonferonni inequality in order to keep the experiment-wise error rate at p ⬍ 0.05 for each group of comparisons. Within-group comparisons were to the baseline value for each series of drug administration. Between-group comparisons for each variable were made at each drug administration dose rate.

RESULTS

Pigs dosed with purified fumonisin B 1 appeared clinically normal during the study. Feed consumption, weight gain, and respiratory rate were not different between groups. Cardiovascular effects. Fumonisin B 1-treated pigs had decreased left ventricular dP/dt max, cardiac output, and mean aortic pressure along with an increase in mean pulmonary artery pressure, as compared to control pigs (Table 1, Fig. 1). Heart rate tended (p ⫽ 0.050) to be lower in fumonisin-treated pigs. Mean central venous pressure, left ventricular end-diastolic pressure, systemic vascular resistance, pulmonary vascular resistance, and stroke volume were not different between groups. Blood gas analyses and oxygen transport. Although differences were not seen in arterial pH, pCO 2, or pO 2 when pigs were breathing 100% O 2 during anesthesia, fumonisin-treated pigs had decreased arterial and mixed venous pO 2 tensions when breathing room air (Table 2). Base excess values and HCO 3 concentrations were similar between groups. Fumoni-

243


TABLE 1 Effects of Intravenous Administration of Purified Fumonisin B 1 at 1.0 mg per kg of Body Weight for 4 Days on Selected Cardiovascular Parameters in Swine

Heart rate (beats/min) Cardiac output (L/min) Stroke volume (mL) Mean aortic pressure (mm Hg) Mean pulmonary artery pressure (mm Hg) Mean central venous pressure (mm Hg) Left ventricular dP/dt max (mm Hg/s) Left ventricular end-diastolic pressure (mm Hg) Systemic vascular resistance (dyn sec/cm 5) Pulmonary vascular resistance (dyn sec/cm 5)

Control

Fumonisin

p value

114 ⫾ 20 4.16 ⫾ 0.81 37 ⫾ 6 120 ⫾ 18 13 ⫾ 2 3⫾1 1742 ⫾ 178 9.2 ⫾ 1.6 2323 ⫾ 580 70 ⫾ 26

95 ⫾ 7 2.75 ⫾ 0.49* 29 ⫾ 7 86 ⫾ 10* 17 ⫾ 1* 2⫾1 1381 ⫾ 96* 10.0 ⫾ 2.2 2537 ⫾ 681 195 ⫾ 158

0.050 0.003 NS 0.003 0.019 NS 0.001 NS NS NS

Note. Values are mean ⫾ SD; NS ⫽ not significant. *p ⬍ 0.05, compared with control group.

sin-treated pigs also had an increased arterial hemoglobin concentration; however, systemic oxygen delivery was not different between treated and control animals (Table 2). The oxygen extraction ratio was increased and oxygen consumption tended to be increased (p ⫽ 0.055) in treated pigs; however, pulmonary arterial blood temperature, alveolar-arterial O 2 gradient, and physiologic shunt fraction (Q sQ t) were unchanged between groups. Sphingolipid analyses. Plasma sphinganine and sphingosine concentrations were not different between groups before the study began; however, both were markedly increased in fumonisin-treated pigs on day 5 (Fig. 2). Sphinganine and sphingosine concentrations were also increased in the left ventricle of treated pigs (Fig. 3). Vascular permeability. There was no difference between the amount of BAL fluid recovered from control (33 ⫾ 8 ml) and treated pigs (34 ⫾ 4 ml). Total protein concentrations and polymorphonuclear cell counts from bronchoalveolar lavage samples along with the calculated relative permeability index of the alveolar-capillary membrane were not different between groups (Fig. 4). Clinical pathology. Increases in serum liver-associated enzyme activities were moderate in pigs dosed with fumonisin B 1. Fumonisin-treated pigs had increased cholesterol concentrations, and alkaline phosphatase, gamma-glutamyl transferase, aspartate aminotransferase, and sorbitol dehydrogenase activities (Table 3). Serum creatinine and total protein concentrations were also increased in treated animals; however, serum urea nitrogen, calcium, phosphorus, sodium, potassium, chloride, and glucose concentrations were not different between groups. Pathology and lung wet-weight to dry-weight ratios. Pulmonary edema was not observed at necropsy. The lung wetto-dry weight ratio was higher (p ⫽ 0.002) in fumonisintreated pigs (6.1 ⫾ 0.3) than in control pigs (5.3 ⫾ 0.1), suggesting that mild pulmonary edema was present in the

treated group (Gumprecht et al., 1998). Histologic lesions were observed in lung and liver. In the lungs, edema was characterized by widening of connective tissues around airways and vessels, widening of interlobular septa, and dilation of lymphatics. Edema was scored as minimal, mild, or moderate. In control pigs, there was minimal non-proteinaceous edema, presumably due to the recumbency of the pigs during the cardiovascular studies. In 5 of 6 fumonisin-treated pigs there was mild to moderate proteinaceous interstitial edema, in one case with the edema extending into alveoli and airways. In the liver, there was scattered hepatocyte apoptotic cell death and mitosis. Apoptosis was not observed in control pigs, whereas it occurred frequently (p ⫽ 0.0028) in fumonisintreated pigs (median 5, range 1 to 16). Mitotic hepatocytes were observed more frequently (p ⫽ 0.018) in fumonisintreated pigs (median 2, range 0 to 4) as compared to controls (median 0, range 0 to 1). In the kidneys, there was mild dilation of tubules in treated pigs. The hearts from both control and fumonisin-treated pigs had multifocal areas of contraction with hyalinization and vacuolation of myocytes, as well as multifocal areas of mild disorganization and hypercellularity. These changes were presumably related to cardiovascular instrumentation or were artifacts. In previous studies using non-instrumented pigs, myocardial lesions were not observed (Haschek et al., 1992; Motelin et al., 1994). On ultrastuctural evaluation, there were no differences between the hearts of fumonisin-treated and control pigs. In the lungs of treated pigs, accumulations of membranous material were present in the pulmonary capillary endothelial cells. DISCUSSION

The major finding of this study is that intravenous administration of purified fumonisin B 1 decreases cardiovascular function in swine in an identical manner to that observed when swine ingested fumonisin-containing culture material (Consta-

244

SMITH ET AL.

FIG. 1. Representative data from a control pig and a fumonisin-treated pig showing maximal rate of change of left ventricular pressure (dP/dt), and lead (II) ECG.

ble et al., 1999; Smith et al., 1996a,b; Smith et al., 1999). This finding conclusively demonstrates that fumonisin, and not another toxin, was responsible for the cardiovascular depression in swine fed fumonisin-containing culture material. In addition, inhibition of cardiac function occurs without alterations in alveolar-capillary membrane permeability. Coupled with our recent findings that fumonisin ingestion terminally increases pulmonary artery wedge pressure and decreases cardiac contractility and mean aortic pressure (Constable et al., 1999; Smith et al., 1999), we conclude that fumonisin-induced pulmonary edema in swine is due to acute left-sided heart failure. The mechanism of fumonisin toxicity is widely believed to be related to altered sphingolipid biosynthesis. Fumonisins inhibit sphingosine-N-acyltransferase and sphinganine-N-acyltransferase (Wang et al., 1991), enzymes that are key compo-

nents in the pathway for de novo sphingolipid biosynthesis. Fumonisin-induced enzyme inhibition results in increased concentrations of free sphinganine and sphingosine in the plasma and tissues of swine, as well as depletion of complex sphingolipids (this study, Gumprecht et al., 1998; Riley et al., 1993; Smith et al., 1999). Although the physiologic role of sphinganine is unknown, sphingosine is an important intracellular second messenger (Hannun and Bell, 1989) that inhibits L-type calcium channels in myocardial cells, thereby decreasing sarcoplasmic reticulum Ca 2⫹-induced Ca 2⫹-release and cardiac contractility (Dettbarn et al., 1994; McDonough et al., 1994; Sabbadini et al., 1992; Webster et al., 1994). Cardiovascular toxicity could therefore result from fumonisin-induced increases in plasma and tissue sphingosine concentrations. Sphingosine inhibits excitation-contraction coupling in rat ven-

245


TABLE 2 Effects of Administration of Purified Fumonisin B 1 at 1.0 mg per kg of Body Weight for 4 Days on Blood Gas, Oxygen Delivery, and Oxygen Consumption Values in Dorsally Recumbent Swine Ventilated with Room Air or 100% Oxygen Control

Fumonisin

p value

64 ⫾ 7 33 ⫾ 2 10.2 ⫾ 0.8 29 ⫾ 5 21 ⫾ 6 19.7 ⫾ 3.1 4.9 ⫾ 1.2 25 ⫾ 4

53 ⫾ 9* 24 ⫾ 3* 12.9 ⫾ 1.9* 38 ⫾ 12 26 ⫾ 11 17.8 ⫾ 4.1 6.4 ⫾ 1.2 38 ⫾ 10*

0.036 0.002 0.008 0.10 NS NS 0.055 0.016

7.44 ⫾ 0.01 199 ⫾ 65 47 ⫾ 5 33.1 ⫾ 2.2 7.2 ⫾ 1.8

7.44 ⫾ 0.01 151 ⫾ 35 45 ⫾ 5 30.9 ⫾ 2.4 6.2 ⫾ 1.5

NS NS NS NS NS

Room air Arterial pO 2 (mm Hg) Mixed venous pO 2 (mm Hg) Hemoglobin concentration (g/dL) (p[A-a]O 2) (mm Hg) Physiologic shunt to blood flow ratio (%) Systemic oxygen delivery (ml O 2 min –1 kg –1) Oxygen consumption (ml O 2 min –1 kg –1) Oxygen extraction ratio (%) 100% Oxygen Arterial Arterial Arterial Arterial Arterial

pH pO 2 (100% O 2) (mm Hg) pCO 2 (mm Hg) HCO 3 (mEq/L) BE (mEq/L)


tricular myocytes (IC 50, 0.5–1.0 ␮M) by reducing the amount of entering trigger Ca 2⫹ for Ca 2⫹-induced Ca 2⫹ release (L-type Ca 2⫹ channel blockade) and raising the ryanodine-receptor threshold for Ca 2⫹ release, although the effects of sphingosine on the L-type calcium channels occurred at lower concentrations than the ryanodine receptor (McDonough et al., 1994). In studies conducted on skinned myocardial cells and isolated junctional sarcoplasmic reticulum, sphingosine concentrations ⱖ1 ␮M inhibited Ca 2⫹ release from the sarcoplasmic reticulum ryanodine receptor in a dose-dependent manner when release

FIG. 2. Effects of intravenous administration of purified fumonisin B 1 at 1.0 mg per kg of body weight for 4 days on plasma sphingosine and sphinganine concentrations in swine. Data presented as mean ⫾ SD and represent plasma concentrations before fumonisin treatment began (Baseline) and at the conclusion of the study (euthanasia). Asterisks indicate significant difference from controls.

was induced by caffeine (IC 50, 1 ␮M), doxorubicin (IC 50, 2 ␮M), or calcium (Dettbarn et al., 1994; Sabbadini et al., 1992). Fumonisin-treated pigs in this study had mean sphingosine concentrations of 0.5 ␮M/L in plasma and 7 ␮M/kg in the left ventricle (Figs. 3 and 4) that are physiologically relevant, as these concentrations block L-type Ca 2⫹ channels in vitro. Sphingosine-mediated L-type Ca 2⫹ channel blockade has been observed in canine ventricular cells (Dettbarn et al., 1994), rat ventricular cells (McDonough et al., 1994), rabbit ventricular cells (Webster et al., 1994), and cat ventricular cells (Oral et al., 1997), while fumonisin B 1 itself inhibits calcium

FIG. 3. Effects of intravenous administration of purified fumonisin B 1 at 1.0 mg per kg of body weight for 4 days on sphingosine and sphinganine concentrations in the left ventricle of swine. Data presented as ␮M of sphinganine and sphingosine per kilogram of wet tissue (mean ⫾ SD). Asterisks indicate significant difference from controls.

246

SMITH ET AL.

FIG. 4. Effects of intravenous administration of purified fumonisin B 1 at 1.0 mg per kg of body weight for 4 days on alveolar-capillary membrane permeability of swine. Data are presented as mean ⫾ SD and include the protein concentration of bronchoalveolar lavage (BAL) samples (mg/dl), the calculated relative permeability index (RPI), and the concentration of polymorphonuclear cells (PMNs) in the bronchoalveolar lavage fluid (⫻10 cells/ ␮l) from fumonisin-treated and control pigs.

channels in frog atrial tissue (Sauviat et al., 1991). The mechanism of this L-type Ca 2⫹ channel blocking does not appear to involve effects on bulk surface charge or a voltage-dependent block. Instead, sphingosine appears to alter channel gating by increasing the closed time probability of the Ca 2⫹ channel, with the blocking action being related to the polar head groups or the primary amine of sphingosine (Yasui and Palade, 1996). This blocking of sphingosine is reversible (Titievsky et al., 1998; Yasui and Palade, 1996) and therefore should not alter the responsiveness of cardiac tissue to calcium or G-proteinmediated ligands such as isoproterenol, as seen in this study. Since the onset of hemodynamic changes are temporally associated with increases in plasma sphingosine (Smith et al., 1999), and pulmonary edema is seen only when pigs ingest diets containing high fumonisin concentrations for more than 3 days, we believe that sphingosine causes a dose-dependent L-type Ca 2⫹-channel blockade,which ultimately results in acute left ventricular failure and subsequent pulmonary edema. The two most common mechanisms of pulmonary edema in mammals are increased capillary hydrostatic pressure (i.e., left-sided heart failure) and increased capillary permeability (West, 1987). Previously, Ramasamy and colleagues reported that fumonisin B 1 doubled the rate of albumin transfer across porcine endothelial cell monolayers in vitro and speculated that disruption of endothelial barrier function was responsible for the pathogenesis of porcine pulmonary edema and equine leukoencephalomalacia (Ramasamy et al., 1995). Although increased pulmonary capillary permeability has been stated to be the cause of pulmonary edema in swine ingesting fumonisin

(Fazekas et al., 1998), results of in vivo studies indicate that increased capillary permeability does not occur with fumonisin toxicosis. Evans blue dye clearance is an accurate and reproducible method for assessing alterations in vascular permeability (Patterson et al., 1992), and we previously found no difference between clearance rates of Evans blue dye in fumonisintreated and control pigs (Smith et al., 1999). In this study, we used bronchoalveolar lavage to calculate the alveolar-capillary RPI. Increases in bronchoalveolar lavage protein concentrations correlate with alveolar-capillary membrane injury as damage to the endothelial membrane permits intravascular proteins to move into alveolar spaces (Holter et al., 1986). Although techniques using tracer dyes or radio-labeled particles are considered to have greater sensitivity, BAL fluid protein concentrations have been widely used to assess changes in pulmonary endothelial permeability (Byrne et al., 1990; Olson et al., 1985). In this study we found no difference between control and treated animals. For comparison, studies done in swine using endotoxin as a model of pulmonary vascular damage have demonstrated 40-fold increases in the RPI (Olson et al., 1985) and 5-fold increases in bronchoalveolar lavage protein concentrations (Byrne et al., 1990). Studies performed in vitro have demonstrated that fumonisins induce direct cytotoxicity in numerous cell lines, including endothelial cell monolayers (Ramasamy et al., 1995), hepatocytes (van der Westhuizen et al., 1998), keratinocytes and esophageal epithelial cells, (Tolleson et al., 1996), splenocytes (Wu et al., 1995a), colon cells (Schmelz et al., 1998), macrophages (Chatterjee et al., 1995), lymphocytes (Dombrink-

TABLE 3 Effects of Intravenous Administration of Purified Fumonisin B 1 at 1.0 mg per kg of Body Weight for 4 Days on Serum Biochemical Parameters in Swine

Creatinine (mg/dL) Urea nitrogen (mg/dL) Total protein (g/dL) Albumin (g/dL) Calcium (mg/dL) Phosphorus (mg/dL) Sodium (mEq/L) Potassium (mEq/L) Chloride (mEq/L) Glucose (mg/dL) Cholesterol (mg/dL) Total bilirubin (mg/dL) Creatine kinase (IU/L) Alkaline phosphatase (IU/L) Gamma-glutamyl transferase (IU/L) Aspartate aminotransferase (IU/L) Sorbitol dehydrogenase (IU/L)

Control

Fumonisin

p value

0.9 ⫾ 0.1 12 ⫾ 3 5.8 ⫾ 0.2 3.1 ⫾ 0.2 9.8 ⫾ 0.5 9.9 ⫾ 0.9 139 ⫾ 3 4.3 ⫾ 0.2 98 ⫾ 3 105 ⫾ 15 91 ⫾ 11 0.10 ⫾ 0.01 830 ⫾ 280 179 ⫾ 29 37 ⫾ 9 36 ⫾ 5 2⫾1

1.6 ⫾ 0.2* 13 ⫾ 2 6.8 ⫾ 0.3* 3.4 ⫾ 0.2 9.6 ⫾ 0.4 9.5 ⫾ 1.1 140 ⫾ 5 4.2 ⫾ 0.5 99 ⫾ 3 105 ⫾ 13 166 ⫾ 21* 0.25 ⫾ 0.19 807 ⫾ 320 318 ⫾ 144* 64 ⫾ 22* 66 ⫾ 29* 5 ⫾ 1*

0.001 0.09 0.003 0.08 NS NS NS NS NS NS 0.001 NS NS 0.043 0.019 0.033 0.002



Kurtzman et al., 1994), chondrocytes (Wu et al., 1995b) and kidney cells (Yoo et al., 1996). Many of these studies used fumonisin doses (25 to 100 ␮M) well above the physiologic range. Based on the pharmokinetics of the toxin in swine, fumonisin is poorly absorbed after oral administration (5% bioavailability) and has a volume of distribution (V d) of 2.4 L/kg (Prelusky et al., 1994). Assuming a molecular weight of 722 for fumonisin B 1, the maximum theoretical plasma concentration is 0.6 ␮M following administration of 1.0 mg/kg fumonisin B 1 IV, or 20 mg/kg orally which causes pulmonary edema. We must therefore use caution when drawing conclusions about the potential in vivo toxicity of fumonisin B 1 from in vitro studies that use supraphysiologic concentrations of fumonisin. Fumonisin-treated pigs had a decrease in arterial oxygen tension when compared to control pigs in this study. Although the arterial O 2 tensions for both groups were lower than the expected normal range of 90 to 100 mm Hg (Table 2), these samples were collected after pigs had been under anesthesia (without supplemental oxygen) and in dorsal recumbency for 10 to 15 min. In this study, the systemic arterial hypoxemia of fumonisin-treated pigs was abolished by administering supplemental oxygen, which is in agreement with a previous study (Smith et al., 1996b). This indicated that an anatomic shunt was not associated with the hypoxemia (West, 1987), and that diffusion impairment was the primary cause. Ultrastructural studies have demonstrated that fumonisin treatment results in a thickening of pulmonary endothelial cells due to an accumulation of membranous material within the cytoplasm (this study, Gumprecht et al., 1998). The results from this study indicate that fumonisin B 1 is responsible for induction of this alteration. Although this alteration could potentially create oxygen diffusion abnormalities leading to systemic arterial hypoxemia, we attribute most of the diffusion impairment to the interstitial edema observed histologically, which resulted from progressive left-sided heart failure. The 31% increase in oxygen consumption in fumonisintreated pigs confirms our previous findings in pigs fed fumonisin-containing culture material where we observed increases in oxygen consumption of 38% (Smith et al., 1999) and 49% (Smith et al., 1996b). This increase in systemic oxygen consumption is potentially important since fumonisin decreases the average daily gain of pigs (Motelin et al., 1994; Smith et al., 1996b), even at concentrations as low as 1 ppm (Rotter et al., 1996). Since feed consumption was generally not affected in these studies, it is likely the increase in systemic oxygen consumption was responsible for the decreased growth rates of fumonisin-treated pigs. The cellular mechanism responsible for the increased oxygen consumption is not clear; however, we speculate that sphingosine-mediated inhibition of L-type Ca 2⫹ channels, or sphingosine-1-phosphate activated Ca 2⫹ release from the endoplasmic reticulum may be involved. Increased leakage of calcium from intracellular stores such as the endoplasmic reticulum will lead to increased oxygen consumption

247

and decreased mechanical efficiency of the heart, as the calcium that leaks out has to be pumped back in to the endoplasmic reticulum by membrane bound Ca 2⫹-ATPase (Takasago et al., 1993). Our recent research has indicated that fumonisintreated pigs have a significant reduction in left ventricular mechanical efficiency (Constable et al., 2000). Most of the current regulatory debate surrounding fumonisin B 1 centers on its potential carcinogenicity. Male BD-1X rats fed 50 ppm purified fumonisin B 1 developed hepatic regenerative nodules with a progression to hepatic cirrhosis and hepatocellular carcinoma after 20 months (Gelderblom et al., 1991). A more recent 2-year study conducted by the National Toxicology Program showed a dose-dependent increase in renal tubular carcinomas in male Fischer 344 rats fed 50 and 150 ppm purified fumonisin B 1, and a dose-dependent increase in hepatocellular tumors in female B6C3F 1 mice fed fumonisin B 1 at 50 and 80 ppm (Howard, 1999). While the potential carcinogenicity of the toxin is important, fumonisin B 1 can induce lethal pulmonary edema in swine at concentrations of less than 80 ppm (Motelin et al., 1994), cause leukoencephalomalacia in horses at concentrations as low as 5 to 8 ppm (Ross et al., 1991; Wilson et al., 1990, 1992), and increase serum cholesterol levels in swine at ⱖ1 ppm (Rotter et al., 1996). Fumonisin-induced hepatotoxicity characterized by hepatocyte apoptosis and proliferation, and elevated serum cholesterol concentrations (as seen in this study) occur in all mammals studied experimentally. Fumonisins have also been shown to induce right ventricular hypertrophy and medial hypertrophy of the small pulmonary arteries when fed to pigs for 7 months in diets containing 150 to 170 ppm fumonisin B 1 in culture material (Casteel et al., 1994). While more studies are needed to examine the potential long-term effects of fumonisin on the cardiovascular system, we believe the findings in this study raise important regulatory issues that must be considered. In summary, the results of this study indicate that purified fumonisin B 1 decreases cardiovascular function in swine without altering pulmonary capillary permeability. These findings indicate that fumonisin-induced pulmonary edema in swine is due to acute left-sided heart failure, which is most likely induced by sphingosine-mediated L-type Ca 2⫹-channel blockade. Because swine are considered to be the best animal models for studying human cardiovascular disease (McKenzie, 1996), future research is needed to examine the potential for fumonisin cardiotoxicity in other species, including humans.

ACKNOWLEDGMENTS We thank Dr. Helen Parker and Catherine Simutis for technical assistance, and Dr. James Porter for conducting the fusaric acid assays. The study was supported by an American Heart Association Grant-in-Aid (97-GB-01). Dr. Smith is supported by an American Heart Association Fellowship Award (9804717X).

248

SMITH ET AL.

REFERENCES Bacon, C. W., and Nelson, P. E. (1994). Fumonisin production in corn by toxigenic strains of Fusarium moniliforme and Fusarium proliferatum. J. Food Prot. 57, 514 –521. Bacon, C. W., Porter, J. K., Norred, W. P., and Leslie, J. F. (1996). Production of fusaric acid by Fusarium species. Appl. Environ. Microbiol. 62, 4039 – 4043. Bennett, G. A., and Richard, J. L. (1994) Liquid chromatographic method for analysis of the naphthalene dicarboxaldehyde derivative of fumonisins. J. AOAC Int. 77, 501–506. Byrne, K., Cooper, K. R., Carey, P. D., Berlin, A., Sielaff, T. D., Blocher, C. R., Jenkins, J. K., Fisher, B. J., Hirsch, J. I., Tatum, J. L., Fowler, A. A., and Sugerman, J. (1990). Pulmonary compliance: Early assessment of evolving lung injury after onset of sepsis. J. Appl. Physiol. 69, 2290 –2295. Casteel, S. W., Turk, J. R., and Rottinghaus, G. E. (1994). Chronic effects of dietary fumonisin on the heart and pulmonary vasculature of swine. Fundam. Appl. Toxicol. 23, 518 –524.

Howard, P. C. (1999). Toxicology and carcinogenesis studies of fumonisin B 1 in F344/N rats and B6C3F 1 mice. National Toxicology Technical Report #496 (NIH Publication No. 99 –3955). Lang, R. M., Marcus, R. H., Neumann, A., Janzen, D., Hansen, D., Fujii, A. M., and Borow, K. M. (1992). A time-course study of the effects of pentobarbital, fentanyl, and morphine chloralose on myocardial mechanics. J. Appl. Physiol. 73, 143–150. Marasas, W. F. O., and Nelson, P. E. (1984). Toxigenic Fusarium Species. Pennsylvania State University Press, University Park, PA. McDonough, P. M., Yasui, K., Betto, R., Salviati, G., Glembotski, C. C., Palade, P. T., and Sabbadini, R. A. (1994). Control of cardiac Ca 2⫹ levels. Inhibitory actions of sphingosine on Ca 2⫹ transients and L-type Ca 2⫹ channel conductance. Circ. Res. 75, 981–989. McKenzie, J. E. (1996). The swine as a model in cardiovascular research. In Advances in Swine in Biomedical Research, (M. E. Tumbleson and L. B. Schook, Eds.), Vol. 1, pp. 7–17. Plenum Press, New York.

Chatterjee, D., Mukherjee, S. K., and Dey, A. (1995). Nuclear disintegration in chicken peritoneal macrophages exposed to fumonisin B 1 from Indian maize. Lett. Appl. Microbiol. 20, 184 –185.

Motelin, G. K., Haschek, W. M., Ness, D. K., Hall, W. F., Harlin, K. S., Schaeffer, D. J., and Beasley, V. R. (1994). Temporal and dose-response features in swine fed corn screenings contaminated with fumonisin mycotoxins. Mycopathologia 126, 27– 40.

Constable, P. D., Smith, G. W., Rottinghaus, G. E., and Haschek W. M. (2000). Ingestion of fumonisin B 1-containing culture material decreases cardiac contractility and mechanical efficiency in swine. Toxicol. Appl. Pharmacol. 162, 151–160.

Nagaraj, R. Y., Wu, W., Will, J. A., and Vesonder, R. F. (1996). Acute cardiotoxicity of moniliformin in broiler chickens as measured by electrocardiography. Avian Dis. 40, 223–227.

Dettbarn, C. A., Betto, R., Salviati, G., Palade, E., Jenkins, G. M., and Sabbadini, R. A. (1994). Modulation of cardiac sarcoplasmic reticulum ryanodine receptor by sphingosine. J. Mol. Cell. Cardiol. 26, 229 –242. Dombrink-Kurtzman, M. A., Bennett, G. A., and Richard, J. L. (1994). An optimized MTT bioassay for determination of cytotoxicity of fumonisins in turkey lymphocytes. J. AOAC Int. 77, 512–516. Fazekas, B., Bajmocy, E., Glavits, R., Fenyvesi, A., and Tanyi, J. (1998). Fumonisin B 1 contamination of maize and experimental acute fumonisin toxicosis in pigs. Zentralbl Veterinarmed 45, 171–181. Gelderblom, W. C., Kriek, N. P., Marasas, W. F., and Thiel, P. G. (1991). Toxicity and carcinogenicity of the Fusarium moniliforme metabolite, fumonisin B 1, in rats. Carcinogenesis 12, 1247–1251. Gumprecht, L. A., Beasley, V. R., Weigel, R. M., Parker, H. M., Tumbleson, M. E., Bacon, C. W., Meredith, F. I., and Haschek, W. M. (1998). Development of fumonisin-induced hepatotoxicity and pulmonary edema in orally dosed swine: Morphological and biochemical alterations. Toxicol. Pathol. 26, 777–788. Gupta, S., Krasnoff, S. B., Underwood, N. L., Renwick, J. A., and Roberts, D. W. (1991). Isolation of beauvericin as an insect toxin from Fusarium semitectum and Fusarium moniliforme var. subglutinans. Mycopathologia 115, 185–189. Hannun, Y. A., and Bell, R. M. (1989). Functions of sphingolipid and sphingolipid breakdown products in cellular regulation. Science 243, 500 –507. Harrison, L. R., Colvin, B. M., Greene, J. T., Newman, L. E., and Cole, J. R., Jr. (1990). Pulmonary edema and hydrothorax in swine produced by fumonisin B 1, a toxic metabolite of Fusarium moniliforme. J. Vet. Diagn. Invest. 2, 217–221. Haschek, W. M., Motelin, G., Ness, D. K., Harlin, K. S., Hall, W. F., Vesonder, R. F., Peterson, R. E., and Beasley, V. R. (1992). Characterization of fumonisin toxicity in orally and intravenously dosed swine. Mycopathologia 117, 83–96.

Nagatsu, T., Hidaka, H., Kuzuya, H., Takeya, K., and Umezawa, H. (1970). Inhibition of dopamine beta-hydroxylase by fusaric acid (5-butylpicolinic acid) in vitro and in vivo. Biochem. Pharmacol. 19, 35– 44. Olson, N. C., Brown, T. T., Jr., and Anderson, D. L. (1985). Dexamethasone and indomethacin modify endotoxin-induced respiratory failure in pigs. J. Appl. Physiol. 58, 274 –284. Oral, H., Dorn, G. W., II, and Mann, D. L. (1997). Sphingosine mediates the immediate negative inotropic effects of tumor necrosis factor-␣ in the adult mammalian cardiac myocyte. J. Biol. Chem. 272, 4836 – 4842. Osweiler, G. D., Ross, P. F., Wilson, T. M., Nelson, P. E., Witte, S. T., Carson, T. L., Rice, L. G., and Nelson, H. A. (1992). Characterization of an epizootic of pulmonary edema in swine associated with fumonisin in corn screenings. J. Vet. Diagn. Invest. 4, 53–59. Patterson, C. E., Rhoades, R. A., and Garcia, J. G. (1992). Evans blue dye as a marker of albumin clearance in cultured endothelial monolayer and isolated lung. J. Appl. Physiol. 72, 865– 873. Prelusky, D. B., Trenholm, H. L., and Savard, M. E. (1994). Pharmacokinetic fate of 14C-labelled fumonisin B 1 in swine. Nat. Toxins 2, 73– 80. Ramasamy, S., Wang, E., Hennig, B., and Merrill, A. H., Jr. (1995). Fumonisin B 1 alters sphingolipid metabolism and disrupts the barrier function of endothelial cells in culture. Toxicol. Appl. Pharmacol. 133, 343–348. Rheeder, J. P., Marasas, W. F. O., Thiel, P. G., Sydenham, E. W., Shepherd, G. S., and van Schalwyk, D. J. (1992). Fusarium moniliforme and fumonisins in corn in relation to human esophageal cancer in Transkei. Phytopathology 82, 353–357. Riley, R. T., An, N. H., Showker, J. L., Yoo, H. S., Norred, W. P., Chamberlain, W. J., Wang, E., Merrill, A. H., Jr., Motelin, G., Beasley, V. R., and Haschek, W. M. (1993). Alteration of tissue and serum sphinganine to sphingosine ratio: An early biomarker of exposure to fumonisin-containing feeds in pigs. Toxicol. Appl. Pharmacol. 118, 105–112.

Hidaka, H., Nagatsu, T., Takeya, K., Takeuchi, T., and Suda, H. (1969). Fusaric acid, a hypotensive agent produced by fungi. J. Antibiot. 22, 228 – 230.

Ritieni, A., Fogliano, V., Randazzo, G., Scarallo, A., Logrieco, A., Moretti, A., Mannina, L., and Bottalico, A. (1995). Isolation and characterization of fusaproliferin, a new toxic metabolite from Fusarium proliferatum. Nat. Toxins 3, 17–20.

Holter, J. F., Weiland, J. E., Pacht, E. R., Gadek, J. E., and Davis, W. B. (1986). Protein permeability in the adult respiratory distress syndrome. Loss of size selectivity of the alveolar epithelium. J. Clin. Invest. 78, 1513–1522.

Ross, P. F., Rice, L. G., Reagor, J. C., Osweiler, G. D., Wilson, T. M., Owens, D. L., Nelson, H. A., Plattner, R. D., Harlin, K. A., Richard, J. L., Colvin, B. M., and Banton, M. I. (1991). Fumonisin B 1 concentrations

PURIFIED FUMONISIN B 1 DECREASES CARDIOVASCULAR FUNCTION IN SWINE in feeds from 45 confirmed equine leukoencephalomalacia cases. J. Vet. Diagn. Invest. 3, 238 –241. Rotter, B. A., Thompson, B. K., Prelusky, D. B., Trenholm, H. L., Stewart, B., Miller, J. D., and Savard, M. E. (1996). Response of growing swine to dietary exposure to pure fumonisin B 1 during an eight-week period: Growth and clinical parameters. Nat. Toxins 4, 42–50. Sabbadini, R. A., Betto, R., Teresi, A., Fachechi-Cassano, G., and Salviati, G. (1992). The effects of sphingosine on sarcoplasmic reticulum membrane calcium release. J. Biol. Chem. 267, 15475–15484. Sauviat, M. P., Laurent, D., Kohler, F., and Pellegrin, F. (1991). Fumonisin, a toxin from the fungus Fusarium moniliforme sheld, blocks both the calcium current and the mechanical activity in frog atrial muscle. Toxicon 29, 1025–1031. Savard, M. E., and Miller, J. D. (1992). Characterization of fusarin F, a new fusarin from Fusarium moniliforme. J. Nat. Prod. 55, 64 –70. Schmelz, E. M., Dombrink-Kurtzman, M. A., Roberts, P. C., Kozutsumi, T., Kawasaki, T., and Merrill, A. H., Jr. (1998). Induction of apoptosis by fumonisin B 1 in HT29 cells is mediated by the accumulation of endogenous free sphingoid bases. Toxicol. Appl. Pharmacol. 148, 252–260. Smith, G. W., Constable, P. D, Bacon, C. W., Meredith, F. I., and Haschek, W. M. (1996a). Cardiovascular effects of fumonisins in swine. Fundam. Appl. Toxicol. 31, 169 –172. Smith, G. W., Constable, P. D., and Haschek, W. M. (1996b). Cardiovascular responses to short-term fumonisin exposure in swine. Fundam. Appl. Toxicol. 33, 140 –148. Smith, G. W., Constable, P. D., Tumbleson, M. E., Rottinghaus, G. E., and Haschek, W. M. (1999). Sequence of cardiovascular changes leading to pulmonary edema in swine fed fumonisin-containing culture material. Am. J. Vet. Res. 60, 1292–1300. Smith, T. K., and Graca-Sousadias, M. (1993). Fusaric acid content of swine feedstuffs. J. Agric. Food Chem. 41, 2296 –2298. Takasago, T., Goto, Y., Kawaguchi, O., Hata, K., Saeki, A., Nishioka, T., and Suga, H. (1993). Ryanodine wastes oxygen consumption for Ca 2⫹ handling in the dog heart. A new pathological heart model. J. Clin. Invest. 92, 823– 830. Terasawa, F., and Kameyama, M. (1971). The clinical trial of a new hypotensive agent, “fusaric acid (5-butylpicolinic acid)”: The preliminary report. Jpn. Circ. J. 35, 339 –357. Titievsky, A., Titievskaya, I., Pasternack, M., Kaila, K., and Tornquist, K. (1998). Sphingosine inhibits voltage-operated calcium channels in GH4C1 cells. J. Biol. Chem. 273, 242–247.

249

Tolleson, W.H., Melchior, W. B., Jr., Morris, S. M., McGarrity, L. J., Domon, O. E., Muskhelishvili, L., James, S. J., and Howard, P. C. (1996). Apoptotic and anti-proliferative effects of fumonisin B 1 in human keratinocytes, fibroblasts, esophageal epithelial cells and hepatoma cells. Carcinogenesis 17, 239 –249. van der Westhuizen, L., Shephard, G. S., Snyman, S. D., Abel, S., Swanevelder, S., and Gelderblom, W. C. (1998). Inhibition of sphingolipid biosynthesis in rat primary hepatocyte cultures by fumonisin B 1 and other structurally related compounds. Food Chem. Toxicol. 36, 497–503. Wang, E., Norred, W. P., Bacon, C. W., Riley, R. T., and Merrill, A. H., Jr. (1991). Inhibition of sphingosine biosynthesis by fumonisins. Implications for diseases associated with Fusarium moniliforme. J. Biol. Chem. 266, 14486 –14490. Webster, R., Sabbadini, R. A., Dettbarn, C. A., and Paolini, P. (1994). Sphingosine effects on the contractile behavior of skinned cardiac myocytes. J. Mol. Cell. Cardiol. 26, 1273–1290. West, J. B. (1987). Pulmonary Pathophysiology-The Essentials, 3rd ed. Williams & Wilkins, Baltimore. Wilson, T. M., Ross, P. F., Owens, D. L., Rice, L. G., Green, S. A., Jenkins, S. J., and Nelson, H. A. (1992). Experimental reproduction of ELEM. A study to determine the minimum toxic dose in ponies. Mycopathologia 117, 115–120. Wilson, T. M., Ross, P. F., Rice, L. G., Osweiler, G. D., Nelson, H. A., Owens, D. L., Plattner, R. D., Reggiardo, C., Noon, T. H., and Pickrell, J. W. (1990). Fumonisin B 1 levels associated with an epizootic of equine leukoencephalomalacia. J. Vet. Diagn. Invest. 2, 213–216. Wu, W., Liu, T., and Vesonder, R. F. (1995a). Comparative cytotoxicity of fumonisin B 1 and moniliformin in chicken primary cell cultures. Mycopathologia 132, 111–116. Wu, W., Li, G., Liu, T., and Vesonder, R. F. (1995b). The effect of fumonisin B 1 on isolated chondrocytes and on bone formation. Poult. Sci. 74, 1431– 1436. Yasui, K., and Palade, P. (1996). Sphingolipid actions in sodium and calcium currents of rat ventricular myocytes. Am. J. Physiol. 270, C645–C649. Yoo, H. S., Norred, W. P., Showker, J., and Riley, R. T. (1996). Elevated sphingoid bases and complex sphingolipid depletion as contributing factors in fumonisin-induced cytotoxicity. Toxicol. Appl. Pharmacol. 138, 211–218.