Fumonisin B 1 Is Hepatotoxic and Nephrotoxic in Milk-Fed Calves

Transcription

1 TOXICOLOGICAL SCIENCES 60, (2001) Copyright 2001 by the Society of Toxicology Fumonisin B 1 Is Hepatotoxic and Nephrotoxic in Milk-Fed Calves Sheerin Mathur,* Peter D. Constable,*,1 Robert M. Eppley, Amy L. Waggoner, Mike E. Tumbleson, and Wanda M. Haschek *Departments of Veterinary Clinical Medicine, Veterinary Pathobiology, and Veterinary Biosciences, College of Veterinary Medicine, University of Illinois, Urbana, Illinois 61802; and U.S. Food and Drug Administration, Washington, DC Received August 31, 2000; accepted December 11, 2000 Fumonisins are a group of mycotoxins that alter sphingolipid biosynthesis and induce leukoencephalomalacia in horses and pulmonary edema in pigs. Experimental administration of fumonisin induces hepatotoxicity in all species, including cattle, as well as nephrotoxicity in rats, rabbits, and sheep. We investigated the hepatotoxicity and nephrotoxicity of fumonisin B 1 to calves. Ten milk-fed male Holstein calves aged 7 to 14 days were instrumented to obtain blood and urine. Treated calves (n 5) were administered fumonisin B 1 at 1 mg/kg, iv, daily and controls (n 5) 10 ml 0.9% NaCl, iv, daily until euthanized on day 7. Fumonisin B 1 - treated calves were lethargic and had decreased appetite from day 4 onward, serum biochemical evidence of severe liver and bile duct injury, and impaired hepatic function. Treated calves also had biochemical evidence of renal injury that functionally involved the proximal convoluted tubules. Sphinganine and sphingosine concentrations in liver, kidney, lung, heart, and skeletal muscle were increased in treated calves. Sphinganine, but not sphingosine, concentration was increased in brains of treated calves. In fumonisin B 1 -treated calves, hepatic lesions were characterized by disorganized hepatic cords, varying severity of hepatocyte apoptosis, hepatocyte proliferation, and proliferation of bile ductular cells. Renal lesions in treated calves consisted of vacuolar change, apoptosis, karyomegaly, and proliferation of proximal renal tubular cells, as well as dilation of proximal renal tubules, which contained cellular debris and protein. This is the first report of fumonisin B 1 -induced renal injury and organ sphingolipid alterations in cattle. Key Words: fumonisin; sphingosine; sphinganine; sphingolipid; hepatotoxicity; cholesterol; nephrotoxicity; proximal renal tubules. Fumonisins are a group of naturally occurring mycotoxins produced primarily by two fungi, Fusarium verticillioides and F. proliferatum, which frequently are found in corn. Fumonisins have been implicated in field cases of equine leukoencephalomalacia (Ross et al., 1993), and porcine pulmonary edema (Osweiler et al., 1992). Experimentally, fumonisins A portion of the results of this article were reported as an Abstract at the Society of Toxicology-2000 Annual Meeting in Philadelphia (abstract #1881). 1 To whom correspondence should be addressed at the University of Illinois at Urbana-Champaign, Department of Veterinary Clinical Medicine, 1008 W. Hazelwood Dr., Urbana, IL p-constable@uiuc.edu. cause liver damage in all species studied to date, including pigs (Haschek et al., 1992; Osweiler et al., 1992), horses (Ross et al., 1993), primates (Jaskiewicz et al., 1987), cattle (Osweiler et al., 1993), sheep (Edrington et al., 1995), rabbits (Gumprecht et al., 1995; Voss et al., 1989), and rats (Suzuki et al., 1995). Fumonisin also cause species-specific target-organ toxicity, such as in brain in horses (Ross et al., 1993), heart in pigs (Casteel et al., 1994; Constable et al., 2000; Smith et al., 1999), kidney in sheep (Edrington et al., 1995), rabbits (Gumprecht et al., 1995; Laborde et al., 1997), and rats (Voss et al., 1989; Voss et al., 1993; Suzuki et al., 1995) as well as esophagus in rats and pigs (Casteel et al., 1994; Lim et al., 1996). Epidemiologic data also has suggested an association between ingestion of corn contaminated with F. verticillioides and human esophageal cancer (Sydenham et al., 1991). While administration of fumonisin has been fatal in a number of species, the cause of death has only been determined in pigs, where fumonisin causes acute left-sided heart failure and pulmonary edema consistent with sphingosine-mediated L-type calcium channel blockade of the heart and systemic vasculature (Constable et al., 2000; Smith et al., 1999, 2000). The pathophysiology of fumonisin-induced lethality is unknown except in pigs, and the reason for species-specific target organ toxicity remains enigmatic. Fumonisin B 1 is the most commonly found and most toxic form of fumonisin (Bucci and Howard, 1996), and ruminants are considered to be more resistant to fumonisin toxicity than horses and pigs (Prelusky et al., 1995). Nevertheless, oral administration of fumonisin B 1 containing culture material has induced toxicity in cattle (Baker and Rottinghaus, 1999; Osweiler et al., 1993; Richard et al., 1996). Ingestion of fumonisin B 1 (105 ppm, daily for 31 days) increased serum aspartate aminotransferase (AST), gamma-glutamyl transferase (GGT), and lactate dehydrogenase (LDH) activities, and serum total bilirubin and cholesterol concentrations in beef calves, but ingestion of feed containing lower fumonisin concentrations ( 26 ppm, daily) had no effect (Osweiler et al., 1993). Ingestion of fumonisin B 1 (75 ppm, daily for 14 days) increased serum cholesterol concentration and decreased feed intake and milk production in Jersey cows (Richard et al., 1996). Fumonisin B 1 (94 ppm, daily for 253 days) increased serum AST and 385

2 386 MATHUR ET AL. GGT activities and induced mild histologic evidence of hepatocellular injury and biliary epithelial hyperplasia in Holstein steers (Baker and Rottinghaus, 1999). Therefore, diets containing 75 ppm fumonisin B 1 are hepatotoxic to cattle. Because little is known regarding the toxicity of fumonisin in adult cattle and calves, it was important to document the toxic effects of acute high-dose fumonisin exposure in this species. We therefore examined the effects of intravenously administered fumonisin B 1 and characterized the response using clinicopathological and histological techniques. We investigated the toxicity of fumonisin in cattle because corn is consumed widely by humans and animals, and it is important to clarify common toxicologic mechanisms of fumonisin mycotoxicosis. Milk-fed calves were used as the experimental model instead of adult cattle, because limited supplies of purified fumonisin B 1 were available. We administered fumonisin B 1 intravenously because of the low oral bioavailability in cattle (Prelusky et al., 1995) and cost of purified fumonisin B 1. Our experimental hypothesis was that intravenous administration of high doses of fumonisin B 1 would be hepatotoxic and nephrotoxic in milk-fed calves, as in sheep (Edrington et al., 1995). This is the first report documenting the toxic effects of purified fumonisin B 1 in cattle and the first report of fumonisin-induced renal injury in cattle. MATERIALS AND METHODS Animals. This study was approved by our institutional committee on the care and use of laboratory animals. Ten healthy male Holstein calves, colostrum-fed and aged between 7 and 14 days (average weight 43 7 kg; mean SD), were obtained from local sources. Calves were housed at an ambient temperature of approximately 21 C in individual calf pens, and were fed a high quality milk replacement (milk origin protein, crude protein 20%, crude fat 20%, crude fiber 0.15%) at 10% of their body weight/day, divided into 2 feedings at 12-h intervals for the duration of the study. Calves were fed milk replacement through an esophageal feeding tube if they did not suckle their milk replacement within 8 min. Calves had access to supplemental water at all times. Instrumentation. Calves were instrumented under halothane anesthesia to obtain cardiovascular measurements (Mathur et al., 2001). A urine collection device was fitted to each calf, consisting of an empty 1-L intravenous fluid bag (Abbott Laboratories, North Chicago, IL) with an IV drip set (Abbott Laboratories) glued to the prepuce of the calf by a commercial adhesive (Goop, Eclectic Products, Mount Prospect, IL) as described (Walker et al., 1998). Experimental protocol. Following full recovery from instrumentation (12 to 24 h), calves were assigned randomly to 2 groups of 5 calves each. Treated calves (n 5) were administered purified fumonisin B 1 at 1 mg/kg, iv, daily for 7 days. Fumonisin B 1 was purified ( 95% free-acid form) as described in Smith et al., 2000, dissolved in phosphate buffered 0.9% NaCl solution (ph 7.0), and the concentration adjusted to produce an administration volume of approximately 10 ml/day. Assuming an oral bioavailability of 5% (Prelusky et al., 1994) and a daily dry-matter intake of 2.5% body weight; this intravenous dose was equivalent to ingestion of feed containing 800 ppm fumonisin B 1. Control calves (n 5) were administered 10 ml of 0.9% NaCl solution at the same time. Blood for serum biochemical analysis was obtained at 8 A.M. each day, and urine voided in the preceding 12-h period was collected to permit assessment of renal function. Blood was allowed to clot at room temperature for 2 to 4 h, centrifuged, and the serum harvested and stored at 20 C for up to 1 month before being analyzed. Bovine serum constituents are stable when handled in this manner (West, 1991). Body weight was recorded every 24 h, immediately before feeding. At this time, each calf was examined and a clinical hydration score (0 to 3), fecal consistency score (0 to 3), and clinical depression score (0 to 3) determined (Walker et al., 1998). Determination of degree of clinical hydration score was as follows: 0 normal, eyes are bright and skin feels pliable; 1 mild dehydration, slight loss of skin elasticity, skin tent 3 s, eyes not recessed into orbit; 2 moderate hydration, skin tent 3 s and eyes slightly recessed into orbit; and 3 severe dehydration, skin tent 10 seconds, eyes markedly recessed into orbit. Degree of fecal consistency: 0 normal, manure is normal and well formed; 1 abnormal feces but not diarrhea, manure is pasty (softer than normal); 2 mild diarrhea (feces are semi liquid, but still have a solid component); and 3 liquid feces only. Degree of clinical depression: 0 normal, 1 mild depression, calf suckles but not vigorously; 2 moderate depression, calf able to stand with assistance and suckle; 3 severe depression, calf unable to stand or suckle. At the completion of the study (day 7), each calf was euthanized with an overdose of sodium pentobarbital (60 mg/kg, iv). A complete necropsy examination was performed and tissues saved for histopathologic examination. Heart, kidney cortex, liver, lung, and cerebral cortical tissue were saved for determination of organ-specific sphingosine and sphinganine concentrations and were stored at 20 C until analyzed. Hepatic function and injury assessment. Critical indicators of hepatic function and injury that were evaluated included serum total bilirubin, bile acid, cholesterol, and glucose concentrations, and serum AST, sorbitol dehydrogenase (SDH), GGT, and alkaline phosphatase (ALP) activities, which were measured using automated methods. Renal function and injury assessment. Serum and urine biochemical analyses were completed by automated methods (Hitachi 704 automatic analyzer, Hitachi, Tokyo, Japan). Urine was examined for the presence of blood, ketones, glucose, proteins, and ph using a commercial urine test strip (Labstix, Bayer, Elkhart, IN). Evidence for renal tubular damage was evaluated by calculating the urinary GGT activity to creatinine ratio (Sommardahl et al., 1997) and by microscopic examination of urine sediment for the presence of casts. Evidence for glomerular and tubular damage was evaluated by calculating the urinary protein to creatinine ratio, which provides a sensitive and reliable diagnostic method for detection and quantification of proteinuria (White et al., 1984). Renal tubular function was assessed by measuring urine specific gravity using refractometry, and by calculating fractional clearances of sodium, potassium, phosphorus, and GGT, and expressing fractional clearance as a percentage of substance (FCx), such that: FCx (Ux/Px) 100/(Ucr/Pcr) (Sommardahl et al., 1997). Fractional clearance of GGT was calculated because this index compares the extent of tubular damage to the amount of functioning kidney mass, rather than to muscle mass, which is the index used when the urinary GGT to creatinine ratio is calculated (Amodio et al., 1985). Serum biochemical analysis. Serum total protein, albumin, calcium, phosphorus, sodium, potassium, chloride, -OH butyrate, and non-esterified fattyacid concentrations, and serum creatine kinase activity were determined by automated methods (Hitachi 704 automatic analyzer, Hitachi, Tokyo, Japan). Anion gap was calculated as anion gap ([Na] [K]) ([Cl] [HCO 3 ]) (Constable et al., 1997). Organ sphingolipid analysis. Frozen tissues were thawed, homogenized in 0.05 M potassium phosphate buffer, and 50 mg (fumonisin-treated calves) or 200 mg (control calves) aliquots were obtained. An internal standard of the 20-carbon analog of sphinganine (sphinganine C20) was added to each sample. Homogenates were hydrolysed and extracted with a mixture of chloroform and 0.2 M KOH in methanol at 40 C for 2 h. Samples were washed (Yoo et al., 1996), dried through Na 2 SO 4 columns, evaporated to dryness under a stream of nitrogen, and derivatized with o-phthaldialdehyde (Riley et al., 1994). Histologic examination. The following tissues were fixed in 10% formalin, embedded in paraffin, sectioned at 5 m, stained with hematoxylin and eosin, and examined microscopically: heart, lung, liver, kidney, pancreas, spleen, esophageal mucosa, and cerebral cortex.

3 FUMONISIN HEPATOTOXICITY AND NEPHROTOXICITY 387 TABLE 1 Effect of Intravenous Administration of Fumonisin B 1 (Fumonisin, n 5) or Isotonic Saline (Control, n 5) on Physical and Serum Biochemical Parameters in Milk-Fed Calves Day of study Baseline Clinical depression score (0 3) Fumonisin * * * Control Number of calves suckling allotted volume of milk replacement within 8 min Fumonisin 4/5 4/5 2/5 1/5 NA Control 4/5 4/5 5/5 4/5 NA Calcium (mg/dl) Fumonisin Control Phosphorus (mg/dl) Fumonisin * * Control Anion gap (meq/l) Fumonisin * 23 3* 21 5* Control Non-esterified fatty acid ( M/l) Fumonisin Control Note. NA not applicable. *p 0.05 compared with baseline value. p 0.05 compared with control group at the same time. Statistical analysis. Data are presented as mean SD. A p value 0.05 was considered significant. Non-normally distributed variables were log transformed or ranked before statistical analysis was performed. Two-way analysis of variance (group, time) with repeated measures on one factor (time) was used for comparison of serum biochemical and renal function parameters. Appropriate Bonferroni-adjusted post-tests were conducted whenever the F value was significant. Within-group comparisons were to the baseline value. Between-group comparisons for each variable were made at each time interval. One way analysis of variance (group) was used for comparison of organ sphingolipid concentrations. A statistical software package (SAS Release 6.12, SAS Institute Inc., Cary, NC) was used for analysis. RESULTS Clinical signs. Calves treated with fumonisin B 1 were lethargic and anorectic from day 4 until euthanasia on day 7, as indicated by higher clinical depression scores and more calves failing to suckle the allotted volume of milk replacement solution within 8 minutes (Table 1). Body weight, clinical hydration score, fecal consistency score, blood temperature, and respiratory rate did not change in treated or control calves (data not shown). Hepatic function and injury assessment. Severe liver and bile duct injury was evident in fumonisin B 1 -treated calves by day 4, as indicated by increased serum AST, SDH, and ALP activities, and increased serum GGT activity relative to calves in the control group (Fig. 1). Hepatic function also was impaired in fumonisin B 1 -treated calves, as indicated by marked increases in serum bile acid concentration by day 2, and increased serum total bilirubin and cholesterol concentrations by day 4 (Fig. 2). Serum GGT activity declined over 7 days in control calves but not in treated calves. Serum glucose concentration decreased over the 7-day study period in both treated and control calves, but serum glucose concentration decreased at a faster rate in fumonisin-treated calves (Fig. 2). Renal function and injury assessment. Serum creatinine concentration was increased in fumonisin B 1 -treated calves by day 6 (Fig. 3), and serum urea nitrogen concentration and urine specific gravity were increased in the same calves by day 7 (Table 2). Urine creatinine, sodium, potassium, chloride, and phosphorus concentrations and ph did not change in treated and control calves (data not shown). Urinary fractional clearance of potassium, phosphorus, and GGT, and urine to serum creatinine ratio were increased in treated calves by day 6; whereas, urine fractional clearance of sodium was unchanged in treated calves, but was greater than the control group on day 6 (Fig. 3, Table 2). The indicators of renal injury or altered renal function were changed from baseline: on day 4, urine protein concentration, urine GGT to creatinine ratio, and urine fractional clearance of phosphorus; on day 6, urine protein to creatinine ratio and urine fractional clearance of potassium and GGT, serum creatinine concentration; and on day 7, urine

4 388 MATHUR ET AL. FIG. 1. Effect of intravenous administration of fumonisin B 1 (fumonisin, n 5) or isotonic saline (control, n 5) on serum biochemical indices of hepatic injury (AST and SDH) and bile ductile injury (ALP and GGT). *p 0.05 compared with baseline value. p 0.05 compared with control calves. specific gravity, urine GGT concentration, and serum urea nitrogen concentration. Neither treated nor control calves had urinary blood or ketones present. None of the calves had glycosuria except for one treated calf (100 mg/dl) on day 7. Proteinuria ( 30 mg/dl) was found in treated calves on day 4 (1/5), day 5 (2/5), day 6 (3/5) and day 7 (5/5); whereas, in control calves, proteinuria was detected only on day 5 in one calf. Occasional hyaline and coarse or fine granular casts were seen in two treated calves (from day 2 onward), and one control calf from day 4 onward. Triple phosphate crystals were found in treated calves (3/5) starting day 2; whereas, only one control calf had triple phosphate crystals in its urine. Rare to occasional renal epithelial cells were found in 2 treated calves from day 2 onward, but not in control calves. FIG. 2. Effect of intravenous administration of fumonisin B 1 (fumonisin, n 5) or isotonic saline (control, n 5) on serum biochemical indices of hepatic function. *p 0.05 compared with baseline value. p 0.05 compared with control calves.

5 FUMONISIN HEPATOTOXICITY AND NEPHROTOXICITY 389 FIG. 3. Effect of intravenous administration of fumonisin B 1 (fumonisin, n 5) or isotonic saline (control, n 5) on serum and urine biochemical indices of renal injury and function. *p 0.05 compared with baseline value. p 0.05 compared with control calves. Serum biochemical analysis. Serum total protein, albumin, sodium, potassium, chloride, and -OH butyrate concentrations did not change in treated and control calves, and physiologically important changes in serum creatine kinase were not present (data not shown). Serum non-esterified fatty-acid concentration was greater in treated calves on days 4 and 6, relative to control calves (Table 1). Serum calcium concentration was increased transiently in treated calves from days 2 to TABLE 2 Effect of Intravenous Administration of Fumonisin B 1 (Fumonisin, n 5) or Isotonic Saline (Control, n 5) on Renal Parameters in Milk-Fed Calves Day of study Baseline Serum urea nitrogen (mg/dl) Fumonisin * Control Urine specific gravity Fumonisin * Control Urine GGT (U/L) Fumonisin * Control Urine protein (mg/dl) Fumonisin * * * Control Urine fractional clearance of sodium (%) Fumonisin * Control Urine fractional clearance of potassium (%) Fumonisin * * Control *p 0.05 compared with baseline value. p 0.05 compared with control group at the same time.

6 390 MATHUR ET AL. FIG. 4. Effect of intravenous administration of fumonisin B 1 (fumonisin, n 5) or isotonic saline (control, n 5) on sphinganine concentration in selected organs. p 0.05 compared with control calves. FIG. 5. Effect of intravenous administration of fumonisin B 1 (fumonisin, n 5) or isotonic saline (control, n 5) on sphingosine concentration in selected organs. p 0.05 compared with control calves. 4; serum phosphorus concentration was decreased in treated calves on days 6 and 7, and tended to decrease in control calves over the same time interval (Table 1). Anion gap was increased in treated calves from day 4 onwards, reflecting metabolic acidosis in the absence of increased serum albumin and phosphorus concentrations (Table 1). Anion gap did not change in the control group. Sphingolipid alterations. Sphinganine concentrations in control calves were highest in the lung ( M/l) followed by brain ( M/l), liver ( M/l), heart ( M/l), kidney ( M/L), and skeletal muscle ( M/L; Fig. 4). Sphinganine concentrations were markedly increased in fumonisin-treated calves, relative to control calves, for lung, brain, liver, kidney, heart, and skeletal muscle, with the largest increases occurring in the liver and kidney. Sphingosine concentrations in control calves were highest in the lung ( M/L) followed by brain ( M/l), liver ( M/l), kidney ( M/l), heart ( M/l), and skeletal muscle ( M/l; Fig. 5). Sphingosine concentrations were markedly increased in fumonisin-treated calves, relative to control calves, for lung, kidney, liver, heart, and skeletal muscle, but not for brain. The largest increases in sphingosine concentrations occurred in liver and kidney. Sphinganine to sphingosine ratios were similar for most organs in control calves (Fig. 6). In fumonisin B 1 -treated calves, the largest sphinganine-to-sphingosine ratios were in liver and kidney, with the brain having the lowest ratio. Pathology. Thrombi were found in the right jugular vein of several treated and control calves, and were associated with placement of the Swan-Ganz catheter for hemodynamic measurement (Mathur et al., 2001). The liver appeared pale in all calves and multifocal atelectasis was present in lungs of some calves in both groups. Two fumonisin B 1 -treated calves appeared jaundiced. Treatment-related lesions were present in the livers and kidneys of all fumonisin-treated calves. Liver lesions (Fig. 7) consisted of disorganization of hepatic cords; scattered hepatocyte apoptosis, karyomegaly, and proliferation. In addition, there was proliferation of bile ductular epithelium with almost complete bridging across lobules in the 3 more-severely affected calves. Kidney lesions were consistent with tubular nephrosis (Fig. 8) with proximal convoluted tubules being most severely affected. Tubules were dilated and contained cellular debris, apoptotic cells, and occasional mitotic cells. Scattered tubules were lined by cuboidal basophilic cells, with occasional karyomegaly indicating regeneration. All treated and control calves had foci of myofiber hypereosinophilia and contraction bands in the heart, and one treated calf had chronic FIG. 6. Effect of intravenous administration of fumonisin B 1 (fumonisin, n 5) or isotonic saline (control, n 5) on sphinganine to sphingosine ratio in selected organs. p 0.05 compared with control calves.

7 FUMONISIN HEPATOTOXICITY AND NEPHROTOXICITY 391 FIG. 7. Liver from control and fumonisin B 1 -treated calves at day 7. In the control liver (A and B), normal architecture (cv central vein, b bile ductile) with well delineated hepatic cords is present. In the treated liver (C, D, and E), normal architecture is no longer present, hepatocytes are rounded, and there is bile ductular epithelial proliferation (C, arrows). In D, higher magnification of C, hepatocytes are undergoing mitosis (arrowheads), and in E, scattered apoptosis is present. H&E, 150 (A, C), and 300 (B, D, E). ischemic renal lesions; these were assumed to be related to catheterization. Incidental lesions in some treated and control calves consisted of mild focal pulmonary inflammation and mild non-suppurative periportal inflammation. DISCUSSION The major findings of this study were that fumonisin B 1 is hepatotoxic and nephrotoxic in milk-fed calves, and that fu-

8 392 MATHUR ET AL. FIG. 8. Kidney cortex from control and fumonisin B 1 -treated calves at day 7. In the control kidney (A and B), normal glomeruli and proximal tubules are present. In the treated kidney (C and D), there is tubular dilation with loss of proximal tubular epithelium and presence of cellular and proteinaceous casts. In D, higher magnification of C, apoptotic tubular epithelial cells (arrows) and cellular debris are present within dilated tubules that are lined by few epithelial cells. Other tubules are lined by basophilic epithelial cells with occasional karyomegaly (arrowhead). H&E, 150 (A, C), and 300 (B, D). monisin B 1 increased sphinganine and sphingosine concentrations to the greatest extent in liver and kidney. We believe this to be the first report of renal injury and organ sphingolipid alterations in cattle administered fumonisin B 1. Daily intravenous administration of 0.40 to 1.15 mg of fumonisin B 1 /kg body weight is fatal within 5 days in pigs (Harrison et al., 1990; Haschek et al., 1992) and daily iv administration of mg fumonisin B 1 /kg is fatal within 9 days in horses (Marasas et al., 1988). In contrast, we administered fumonisin B 1 intravenously at 1 mg/kg daily for 7 days and did not observe any fatalities, although calves appeared depressed and had decreased appetite by day 4 of the study. The results of this study and other cattle studies (Baker and Rottinghaus, 1999; Osweiler et al., 1993; Richard et al., 1996) indicate that cattle are more resistant to fumonisin toxicity than pigs and horses. Fumonisin causes a dose-dependent hepatotoxicity in all species studied to date, including cattle (Baker and Rottinghaus, 1999; Osweiler et al., 1993), sheep (Edrington et al., 1995), and goats (Gurung et al., 1998). In this study, we found that fumonisin was hepatotoxic in milk-fed calves, as indicated by alterations in liver function indices (increased serum bile acid, total bilirubin, and cholesterol concentrations), liver injury indices (increased serum AST and SDH activities), and bile-duct injury indices (increased serum ALP and GGT activities). Our finding that serum glucose concentration tended to decline at a faster rate in treated than control calves, coupled with the finding that serum non-esterified fatty acid concentration was greater in treated calves than control calves, suggested that fumonisin-treated calves were in a state of negative energy balance, even

9 FUMONISIN HEPATOTOXICITY AND NEPHROTOXICITY 393 though we failed to detect a change in body weight in either group. Serum bile acid concentration provides a sensitive and specific measure of hepatic injury and function in cattle (West, 1991), and concentrations increase whenever there is a decrease in hepatic reuptake of bile acids (Garry et al., 1994). Serum bile acid concentrations greater than 45 to 88 M/l, observed from day 2 onward in treated calves, indicate the presence of hepatic disease in adult dairy cattle (Craig et al., 1992; Pearson, 1992; West, 1991; ). Serum bile acid concentrations are variable in fed cattle (Craig et al., 1992; Garry et al., 1994; Pearson, 1992) and variability can be reduced by measuring at a fixed time after feeding (Cornelius, 1987), as in this study. Serum total bilirubin concentration is a sensitive and specific measure of hepatic function in the absence of intravascular hemolysis (Cornelius, 1987), and concentrations 0.9 mg/dl, observed from day 2 onward in treated calves, are considered abnormal in calves aged 1 to 2 weeks (Pearson, 1995). Increased serum cholesterol concentrations were observed within 2 days of commencing fumonisin administration. Similar increases in serum cholesterol concentration have been observed in all mammalian species administered fumonisin B 1 (with no increase occurring in turkey poults [Kubena et al., 1997]), including pigs at 1 ppm (Rotter et al., 1996), rats at 55 ppm (Voss et al., 1996), calves at 105 ppm (Osweiler et al., 1993), and lambs at 273 ppm (Edrington et al., 1995). The mechanism for fumonisin-induced hypercholesterolemia is currently unclear, although studies in hamster ovary cells indicate that hypercholesterolemia is not the result of sphinganine- or sphingosine-mediated inhibition of cholesterol esterification (Ridgway, 1995). Taken together, the changes in serum bile acid, total bilirubin, and cholesterol concentrations in treated calves indicate that fumonisin had an early and marked effect on hepatic function. Serum SDH activity provides a sensitive and specific indicator of hepatic injury in cattle (Bartholomew et al., 1987; West, 1991). The increase in serum SDH activity in fumonisintreated calves by day 4 indicated the presence of hepatocellular necrosis. Serum AST activity also increased in fumonisintreated calves, probably as a result of hepatocellular necrosis, although this test is not as specific as SDH. Serum GGT and ALP activities decreased in control calves over the 7 days of the study, which was attributed to natural decay of colostralderived GGT (Thompson and Pauli, 1981) and intestine-derived ALP associated with absorption of colostrum (Healy, 1975; Pauli, 1983; Thompson and Pauli, 1981). Serum ALP activity increased early and to a large extent in fumonisin B 1 treated calves; the magnitude of increase was greater than for serum GGT activity. Although we did not examine ALP isoenyzmes in this study, 3 ALP isoenzymes (bone, liver, intestinal) have been identified in normal cattle serum, with bone the predominant isoenzyme form (Healy, 1971). We suspect the proportionately greater increase in ALP, relative to GGT, partially reflected mobilization of bone phosphorus (Hidiroglou and Thompson, 1980) in response to increased urinary loss of phosphorus in treated calves, in conjunction with liver injury. Increased bone mobilization of calcium and phosphorus also would explain the transient increase in serum calcium concentration in fumonisin treated calves. Serum biochemical changes were consistent with fumonisin induced hepatocellular injury, as well as biliary injury or cholestasis. We believe this to be the first report of fumonisin induced renal injury in cattle. Fumonisin consistently produces dose dependent acute and subacute renal tubular nephrosis in sheep (Edrington et al., 1995), rabbits (Gumprecht et al., 1996), and rats (Voss et al., 1996), with tubular regeneration occurring at low doses (Voss et al., 1996). Renal toxicity has been reported inconsistently in pigs administered fumonisin, with some reporting renal tubular nephrosis (Colvin et al 1993; Harvey et al., 1995, 1996) and others reporting no change (Casteel et al., 1994; Gumprecht et al., 1998; Haschek et al., 1992). The mechanism of renal injury appears to involve disordered sphingolipid metabolism and induction of apoptosis (rat) and acute massive tubular epithelial cell death (sheep, rabbit) (Bucci and Howard, 1996; Gumprecht et al., 1995; Riley et al., 1994). Sphingosine is directly toxic to renal tubular cells, with both the rate and extent of cell killing being concentration-dependent (Iwata et al., 1995). Histopathological findings in fumonisin nephrotoxicosis are characterized by renal tubular necrosis (single or multiple necrotic tubular epithelial cells with pyknotic nuclei; Riley et al., 1994), as observed in this study. Most evidence for fumonisin-induced nephrotoxicity has come from histologic examination, (Voss et al., 1989, 1993, 1996); only a few researchers (Bondy et al., 1995; Edrington et al., 1995; Gumprecht et al., 1995; Smith et al., 1996) have examined and reported changes in renal function indices, such as serum creatinine and urea nitrogen concentrations. Increases in the latter two parameters, although present in sheep (Edrington et al., 1995), goats (Gurung et al., 1998), rabbits (Gumprecht et al., 1995), rats (Bondy et al., 1995) and pigs (Smith et al., 1996) administered fumonisin, do not specifically indicate the presence of nephrotoxicosis, as they can be caused by decreased renal blood flow and glomerular filtration rate (prerenal azotemia). These changes can result from decreased cardiac output and mean arterial pressure, which occur in pigs (Constable et al., 2000; Smith et al.; 1999), but not in milk-fed calves (Mathur et al., 2001). Moreover, serum creatinine and urea concentration also increase in response to decreased plasma volume, which we have observed in fumonisin treated pigs (Smith et al., 1999) and others have observed in rats (Bondy et al., 1995). Because plasma volume, cardiac output, and mean arterial pressure did not change in the milk-fed calves in this study (Mathur et al., 2001), the azotemia was due entirely to renal injury and decreased tubular function. We suspect the increased anion gap in treated calves was due to the presence of unmeasured strong anions, particularly anions associated with uremia (Constable et al., 1997). In rats, fumonisin B 1 altered several markers of nephrotox-

10 394 MATHUR ET AL. icity, including increased urine volume, decreased urine osmolality, proteinuria, enzymuria, and decreased ion transport (Suzuki et al., 1995), suggesting that fumonisin exerts a direct toxic effect on renal glomeruli and tubular cells. Our findings that fumonisin caused a rapid and large increase in urine GGT and protein concentrations, urine fractional clearance of GGT, urine GGT to creatinine ratio, and urine protein to creatinine ratio, and moderate increases in fractional clearance of potassium and phosphorus, are consistent with proximal tubular injury. In calves, kidney has the highest GGT content of any organ (Barakat and Ford, 1988). Almost all GGT activity in urine is derived from the brush border of proximal tubular cells, where GGT is involved in amino acid reabsorption by means of the gamma-glutamyl cycle (Amodio et al., 1985). Urine GGT concentrations and GGT to creatinine ratios increase to the greatest extent when renal tubular cells degenerate (Amodio et al., 1985), as in this study where renal tubular epithelial apoptosis was observed. Proteinuria was considered to be tubular in origin, as renal tubular injury results in decreased reabsorption of filtered low molecular weight proteins such as 2 -microglobulin and lysozyme, and the urine protein to creatinine ratio was 13, which was more indicative of tubular than glomerular proteinuria (Lulich and Osborne, 1990). Moreover, serum albumin concentration was unchanged in fumonisin treated calves; whereas, we would have expected this parameter to decrease in calves with proteinuria due to glomerular disease. Biochemical and histologic changes observed in this study were consistent with fumonisin induced proximal tubular damage. Sphinganine concentrations in the liver and kidney of untreated calves were similar to those in pigs (Gumprecht et al., 1998; Rotter et al., 1996), rats (Riley et al., 1994), rabbits (Gumprecht et al., 1995), and horses (Goel et al., 1996), but less than those reported in healthy goats (Gurung et al., 1998). Liver and kidney sphingosine concentrations of untreated calves were similar to those in horses (Goel et al., 1996) and goats (Gurung et al., 1998), but lower than those in pigs (Rotter et al., 1996; Gumprecht et al., 1998), rats (Riley et al., 1994), and rabbits (Gumprecht et al., 1995). The biological relevance of these differences is unknown. Liver sphinganine and sphingosine concentrations in fumonisin B 1 -treated calves were at least 10 times greater than those reported in pigs dying of pulmonary edema (Gumprecht et al., 1998; Riley et al., 1993). Kidney sphinganine and sphingosine concentrations in treated calves were at least 3 times greater than those in pigs dying of pulmonary edema (Gumprecht et al., 1998; Riley et al., 1993). The relatively greater increase in liver and kidney sphinganine and sphingosine concentrations in fumonisin-treated calves, compared to pigs dying from fumonisin-induced pulmonary edema, may be associated with the greater severity of hepatic and renal disease (this paper) and lower serum sphingolipid concentrations (Mathur et al., 2001) in fumonisin-treated calves. Two potential limitations of this study are the route of fumonisin administration and the use of neonatal calves. The intravenous route of administration does not simulate ingestion, in that the serum fumonisin concentration-time relationship is different, the first capillary bed traversed is pulmonary instead of hepatic, and ruminal microbial metabolism to potentially more toxic metabolites is prevented, although fumonisin appears to be metabolized minimally by rumen microflora (Gurung et al., 1999). In addition, neonatal calves may metabolize or excrete fumonisin differently to adult cattle. The main advantages of administering purified fumonisin B 1 intravenously is cost and the clinicopathologic effects of administration, which can be attributed directly to fumonisin B 1. In conclusion, fumonisin B 1 is hepatotoxic and nephrotoxic in milk-fed calves. The clinicopathologic manifestations of fumonisin toxicity in the calf liver and kidney were similar to those reported in other species. The high dose of fumonisin B 1 administered in this study did not induce leukoencephalomalacia or pulmonary edema. ACKNOWLEDGMENTS This article fulfilled part of the requirements for the MS degree for S.M. This study was supported by funding from the U.S. Department of Agriculture Cooperative Regional Project NC129: Fusarium mycotoxins in cereal grains. REFERENCES Amodio, P., Bazzeria, G., Malatesta, R., and Gatta, A. (1985). Reference ranges and methodological aspects in the urinary measuring of lysozyme, malate-dehydrogenase, -glutamyltransferase and -glucosidase. Enzyme 33, Baker, D. C., and Rottinghaus, G. E. (1999). Chronic experimental fumonisin intoxication of calves. J. Vet. Diagn. Invest. 11, Bartholomew, M. L., Willett, L. B., Liu, T. T., and Moorhead, P. D. (1987). Changes in hepatic function tests to induced toxicity in the bovine liver. J. Anim. Sci. 64, Bondy, G., Suzuki, C., Baker, M., Armstrong, S. F., Hierlihy, L., Rowsell, P., and Mueller, R. (1995). Toxicity of fumonisin B 1 administered intraperitoneally to male Sprague-Dawley rats. Fund. Chem. Toxic. 33, Bucci, T. J., and Howard, P. C. (1996). Effect of fumonisin mycotoxins in animals. J. Toxicol. 15, 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, Colvin, B. M., Cooley, A. J., and Beaver, R. W. (1993). Fumonisin toxicosis in swine: Clinical and pathological findings. J. Vet. Diagn. Invest. 5, 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. Pharm. 162, Constable, P. D, Streeter, R. N., Koenig, G. J., Perkins, N. R., Gohar, H. M., and Morin, D. E. (1997). Determinants and utility of the anion gap in predicting hyperlactatemia in cattle. J. Vet. Int. Med. 11, Cornelius, C. E. (1987). A review of new approaches to assessing hepatic function in animals. Vet. Res. Comm. 11, Craig, A. M., Pearson, E. G., and Rowe, K. (1992). Serum bile acid concen-

11 FUMONISIN HEPATOTOXICITY AND NEPHROTOXICITY 395 trations in clinically normal cattle: Comparison by type, age, and stage of lactation. Am. J. Vet. Res. 53, Edrington, T. S., Kamps-Holtzapple, C.A., Harvey, R. B., Kubena, L. F., Elissalde, M. H., and Rottinghaus, G. E. (1995). Acute hepatic and renal toxicity in lambs dosed with fumonisin-containing culture material. J. Anim. Sci. 73, Garry, F. B., Fettman, M. J., Curtis, R., and Smith, J. A. (1994). Serum bile acid concentration in dairy cattle with hepatic lipidosis. J. Vet. Int. Med. 8, Goel, S., Schumacher, J., Lenz, S. D., and Kemppainen, B. W. (1996). Effects of Fusarium moniliforme isolates on tissue and serum sphingolipid concentrations in horses. Vet. Human Toxicol. 38, Gumprecht, L. A., Beasley, A. R., Weigel, R. M., Parker, H. M., Tumbleson, M. E., Bacon, C. W., Meredith, F. L., and Haschek, W. M. (1998). Development of fumonisin-induced hepatotoxicity and pulmonary edema in orally dosed swine: Morphological and biochemical alterations. Environ. Toxicol. Pathol. 26, Gumprecht, L. A., Marcucci, A., Weigel, R. M., Vesonder, R. F., Riley, R. T., Shwoker, J. L., Beasley, V. R., and Haschek, W. M. (1995). Effects of intravenous fumonisin B 1 in rabbits: Nephrotoxicity and sphingolipid alterations. Nat. Toxins. 3, Gurung, N. K., Rankins, D. L., Jr, and Shelby, R. A. (1999). In vitro ruminal disappearance of fumonisin B1 and its effect on in vitro dry matter disappearance. Vet. Human. Toxicol. 41, Gurung, N. K., Rankins, D. L., Jr, Shelby, R. A., and Goel, S. (1998). Effects of fumonisin B 1 -contaminated feeds on weanling Angora goats. J. Anim. Sci.. 76, 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, Harvey, R. B., Edrington, T. S., Kubena, L. F., Elissalde, M. H., Casper, H. H., Rottinghaus, G. E., and Turk, J. R. (1996). Influence of aflatoxin and fumonisin B 1 -containing culture material, deoxynivalenol-contaminated wheat, or their combination on growing barrows. Am. J. Vet. Res. 57, Harvey, R. B., Edrington, T. S., Kubena, L. F., Elissalde, M. H., and Rottinghaus, G., E. (1995). Influence of aflatoxin and fumonisin B 1 -containing culture material on growing barrows. Am. J. Vet Res. 56, 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, Healy, P. J. (1971). Serum alkaline phosphatase activity in cattle. Clinica. Chemica. Acta 33, Healy, P. J. (1975). Isoenzymes of alkaline phosphatase in serum of newly born lambs. Res. Vet. Sci. 19, Hidiroglou, M., and Thompson, B. K. (1980). Serum alkaline phosphatase activity in beef cattle. Res. Vet. Sci. 11, Iwata, M., Herrington, J., and Zager, R. A. (1995). Sphingosine: A mediator of acute renal tubular injury and subsequent cytoresistence. Proc. Natl. Acad. Sci. 92, Jaskiewicz, K., Marasas, W. F., and Taljaard, J. J. (1987). Hepatitis in Vervet monkeys caused by Fusarium moniliforme. J. Comp. Path. 97, Kubena, L. F., Edrington, T. S., Harvey, R. B, Buckley, S. A., Phillips, T. D., Rottinghaus, G. E., and Caspers, H. H. (1997). Individual and combined effects of fumonisin B 1 present in Fusarium moniliforme culture material and T-2 toxin or deoxynivalelnol in broiler chicks. Poultry Sci. 76, Laborde, J. B., Terry, K. K., Howard, P. C., Chen, J. J., Collins, T. F. X., Shackelford, M. E., and Hansen, D. K. (1997). Lack of embryotoxicity of fumonisin B 1 in New Zealand white rabbits. Fundam. Appl. Toxicol. 40, Lim, C. W., Parker, H. M., Vesonder, R. F., and Haschek W. M. (1996). Intravenous fumonisin B 1 induces cell proliferation and apoptosis in the rat. Nat. Toxins 4, Lulich, J. P., and Osborne, C. A. (1990). Interpretation of urine proteincreatinine ratios in dogs with glomerular and nonglomerular disorders. Comp. Contin. Ed. Pract. Vet. 12, Marasas, W. F., Kellerman, T. S., Gelderblom, W. C., Coetzer, J. A., Thiel, P. G., and Vander Lugt, J. J. (1988). Leukoencephalomalacia in a horse induced by fumonisin B1 isolated from Fusarium moniliforme. Onderstepoort J. Vet. Res. 55, Mathur, S., Constable, P. D., Eppley, R. M., Tumbleson, M. E., Smith, G. W., Tranquilli, W. J., Morin, D. E., and Haschek, W. M. (2001). Fumonisin B 1 increases serum sphinganine concentration but does not alter serum sphingosine concentration or induce cardiovascular changes in milk-fed calves. Toxicol. Sci. 60, Osweiler, G. D., Kehrli, M. E., Stabel, J. R., Thurston, J. R., Ross, P. F., and Wilson, T. M. (1993). Effects of fumonisin-contaminated corn screenings on growth and health of feeder calves. J. Anim. Sci. 71, 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, Pauli, J. V. (1983). Colostral transfer of gamma-glutamyl transferase in lambs. N. Z. Vet. J. 31, Pearson, E. G. (1992). Variability of serum bile acid concentrations over time in dairy cattle, and effect of feed deprivation on the variability. Am. J. Vet. Res. 53, Pearson, E. G. (1995). Evaluation of liver function tests in neonatal calves. J. Am. Vet. Med. Assoc. 207, Prelusky, D. B., Savard, M. E., and Trenholm, H. L. (1995). Pilot study on the plasma pharmacokinetics of fumonisin B 1 in cows following a single dose by oral gavage or intravenous administration. Nat. Toxins 3, Prelusky, D. B., Trenholm, H. L., and Savard, M. E. (1994). Pharmacokinetic fate of 14C-Labelled fumonisin B1 in swine. Nat. Toxins 2, Richard, J. L., Meerdink, G., Maragos, C. M., Tumbleson, M., Bordson, G., Rice, L. G., and Ross, P. F. (1996). Absence of detectable fumonisins in the milk of cows fed Fusarium proliferatum (Matsushima) Nirenberg culture material. Mycopathologia 133, Ridgway, N. D. (1995). Inhibition of acyl-coa: Cholestrol acyltransferase in Chinese hamster ovary (CHO) cells by short-chain ceramide and dihydroceramide. Biochem. et Biophys. Acta 1256, Riley, R. T., An, N. H., Showker, J. L., Yoo, H. S., Norred, W, P., Chamberlain, W. J., Wang, E., Merrill, A. H., Jr., Motelm, G., Beasley, V. R., et al. (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, Riley, R. T., Hinton, D. M., Chamberlain, W. J., Bacon, C. W., Wang, E., Merrill, A. H., Jr, and Voss, K. A. (1994). Dietary fumonisin B 1 induces disruption of sphingolipid metabolism in Sprague-Dawley rats: A new mechanism of nephrotoxicity. Nutr. Pharm. Toxicol. 124, Ross, P. F., Ledet, A. E., Owens, D. L., Rice, L. G., Nelson, H. A., Osweiler, G. D., and Wilson, T. T. (1993). Experimental equine leukoencephalomalacia, toxic hepatosis, and encephalopathy caused by corn naturally contaminated with fumonisins. J. Vet. Diagn. Invest. 5, Rotter, B. A., Thompson, B. K., Prelusky, D. B., Trentholm, 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, Smith, G. W., Constable, P. D., Eppley, R. M., Tumbleson, M., Gumprecht,

12 396 MATHUR ET AL. L. A., and Haschek-Hock, W. M. (2000). Purified fumonisin B 1 decreases cardiovascular function but does not alter pulmonary capillary permeability in swine. Toxicol. Sci. 56, Smith, G. W., Constable, P. D., Smith, A. R., Bacon, C. W., Meredith, F. I., Wollenberg, G. K., and Haschek, W. M. (1996). Effects of fumonisincontaining culture material on pulmonary clearance in swine. Am. J. Vet. Res. 57, Smith, G. W., Constable, P. D., Tumbleson, M., Rottinghaus, G. E., and Haschek, W. M. (1999). Sequence of cardiovascular changes leading to pulmonary edema in swine fed culture material containing fumonisin. Am. J. Vet. Res. 60, Sommardahl, C., Olchowy, T., Provenza, M., and Saxton, A. M. (1997). Urinary diagnostic indices in calves. J. Am. Vet. Med. Assoc. 211, Suzuki, C. A., Hierlihy, L., Barker, M., Curran, I., Mueller, R., and Bondy, G. S. (1995). The effects of fumonisin B 1 on several markers of nephrotoxicity in rats. Toxicol. Appl. Pharmacol. 133, Sydenham, E. W., Shephard, G. S., Theil, P. G., Marasas, W. F. O., and Stockenstrom S. (1991). Fumonisin contamination of commercial cornbased human foodstuffs. J. Agric. Food Chem. 39, Thompson, J. C., and Pauli, J. V. (1981). Colostral transfer of glutamyl transpeptidase in calves. N. Z. Vet. J. 29, Voss, K. A., Chamberlain, W. J., Bacon, C. W., and Norred, W. P. (1993). A preliminary investigation on renal and hepatic toxicity in rats fed purified fumonisin B 1. Nat. Toxins 1, Voss, K. A., Norred, W. P., Plattner, R. D., and Bacon, C. W. (1989). Hepatotoxicity and renal toxicity in rats of corn samples associated with field cases of equine leukoencephalomalacia. Food Chem. Toxicol. 27, Voss, K. A., Riley, R. T., Bacon, C. W., Chamberlain, W. J., and Norred, W. P. (1996). Subchronic toxic effects of Fusarium moniliforme and fumonisin B 1 in rats and mice. Nat. Toxins 4, Walker, P. G., Constable, P. D., Morin, D. E., Drackley, J. K., Foreman, J. H., and Thurmon, J. C. (1998). A reliable, practical, and economical protocol for inducing diarrhea and severe dehydration in the neonatal calf. Can. J. Vet. Res. 62, West, H. J. (1991). Evaluation of total serum bile acid concentrations for the diagnosis of hepatobiliary disease in cattle. Res. Vet. Sci. 51, White, J. V. (1984). Use of protein-to-creatinine ratio in a single urine specimen for quantitative estimation of canine proteinuria. J. Am. Vet. Med. Assoc. 185, Yoo, H., Norred, W. P., Wang, E., Merrill, A. H., and Riley, R. (1996). Fumonisin inhibition of de novo sphingolipid biosynthesis and cytotoxicity in LLC-PK 1 cells. Toxicol. Appl. Pharmacol. 114, 9 15.