Comparative subchronic toxicity studies of nixtamalized and water-extracted Fusarium moniliforme culture material

~

Pergamon

Food and Chemical Toxicology 34 (1996) 623-632

Comparative Subchronic Toxicity Studies of Nixtamalized and Water-extracted Fusarium rnoniliforme Culture Material K. A. VOSSt, C. W. BACON, F. I. M E R E D I T H and W. P. N O R R E D Toxicology & Mycotoxin Research Unit, United States Department of Agriculture, Agricultural Research Service, South Atlantic Area, Richard B. Russell Agricultural Research Center, PO Box 5677, Athens, GA 30604-5677, USA

(Accepted 6 February 1996) Abstract--Fumonisins are mycotoxins produced by Fusarium moniliforme, F. proliferatum and other Fusariurn species, which are commonly found on corn, cause a variety of species-specific toxicoses, and have been linked to human oesophageal cancer in areas of southern Africa and China where corn is a dietary staple. The effect of nixtamalization, the process by which masa flour is produced by alkaline hydrolysis of corn, on the organ-specific toxicity of F. moniliforme culture material containing fumonisin B~ (FBI) was studied and the effectiveness of nixtamalization and water extraction for detoxifying culture material was compared. Male rats (n = 10/group) were fed diets containing 5% culture material equivalent weights of nixtamalized culture material (NX diet) providing 58 ppm hydrolysed FB1 but no FB1, water-extracted culture material (WE diet) providing 8 ppm FB1, or untreated culture material (CM diet) providing 71 pF,m FB1 for 4 wk. An additional control group was fed a diet containing sound seed corn. Serum chemical and histopathological findings confirmed that the nixtamalized culture material was hepatotoxic and nephrotoxic. Hepatopathy was found in all rats fed the NX or CM diets. The lesions were qualitatiwdy similar in these two groups, but were noticeably less severe in rats fed the NX diet. In contrast, only one rat fed the WE diet exhibited mild hepatopathy. Mild-to-moderate nephropathy resembling that induced by FB 1 was found in all rats fed the NX, WE or CM diet. Thus, the organ-specific effects of nixtamalized culture material, containing no detectable FB1, were similar to those of the FBl-containing diet prepared from untreated culture material. Furthermore, nixtamalization was not as effective as water extraction as a detoxification method. Copyright © 1996 Published by Elsevier Science Ltd.

INTRODUCTION

Fusarium moniliforme and F. proliferatum are fungi which are f o u n d worldwide. They c o m m o n l y c o n t a m i n a t e corn ( M u r p h y et al., 1993; Nelson et al., 1993; Sanchis et al., 1994; S y d e n h a m et al., 1993; U e n o et al., 1993; Zoller et al., 1994) a n d also produce a class of mycotoxins, the fumonisins (Fig. 1), which have been s h o w n to cause the fatal syndromes o f horse,; k n o w n as equine leukoencephalomalacia (Kellerman et al., 1990; M a r a s a s et al., 1988) a n d o f swine k n o w n as porcine p u l m o n a r y oedema (Colvin et al., 1993; H a r r i s o n et al., 1990). A l t h o u g h the effects of fumonisins on h u m a n health are stilt speculative, there is considerable evidence associating F. moniliforme and, by inference, fumonisins with the high rates of tAuthor for correspondence. ALT = alanine aminotransferase; A P = alkaline phosphatase; AST = asparate aminotransferase; BUN = blood urea nitrogen; FB1 = fumonisin B~; FB2 = fumonisin B2; GGT = 7-glutamyl transpeptidase; HFBI = hydrolysed fumonisin B~; H F B 2 = hydrolysed fumonisin B2; LDH = lactate dehydrogenase; NOEL = no-observed-effect level.

Abbreviations:

oesophageal cancer f o u n d in areas o f s o u t h e r n Africa a n d C h i n a (Chu a n d Li, 1994; Marasas, 1993) where corn is a dietary staple. In rats, b o t h culture material o f F. moniliforme ( G e l d e r b l o m et al., 1988; Voss et al., 1990) a n d purified fumonisin B~ (FB1), the most c o m m o n fumonisin found on corn ( G e l d e r b l o m et al., 1988; Voss et al., 1993 a n d 1995), are hepatotoxic a n d nephrotoxic. F. moniliforme a n d FB 1 ( G e l d e r b l o m et al., 1988) have c a n c e r - p r o m o t i n g activity, F. moniliforme culture material may or m a y not be hepatocarcinogenic (Jaskiewicz et al., 1987), and, in one study ( G e l d e r b l o m et al., 1991), purified FB1 induced liver cancer when fed to rats. F u m o n i s i n s have been f o u n d in commercially produced corn meal as well as in processed corn-based food products from a r o u n d the world, albeit at lower levels (Bullerman a n d Tsai, 1994; Pittet et al., 1992; Sanchis et al., 1994; S y d e n h a m et al., 1991 a n d 1992; Thiel et al., 1992; U e n o et al., 1993; Yoshizawa et al., 1994; Zoller et al., 1994). C o r n tortillas m a d e from masa flour are a p o p u l a r dietary staple in Mexico a n d adjacent areas. S y d e n h a m et al. (1991) did not find appreciable fumonisin c o n c e n t r a t i o n s in m a s a flour samples from

0278-6915/96/$15.00 + 0.00 Copyright © 1996 Published by Elsevier Science Ltd. All rights reserved. Printed in Great Britain PII S0278-6915(96)00024-5

624

K.A. Voss et al.

Peru or the United States. Masa is made by soaking corn in heated calcium hydroxide, a process called nixtamalization, and base hydrolysis such as nixtamalization removes the carballylic acid moieties from the molecule, converting fumonisin homologues into their respective hydrolysed forms (Hendrich et al., 1993; Hopmans and Murphy, 1993) (Fig. 1). In their study, Sydenham et al. (1991) did not report whether the nixtamalized masa contained hydrolysed fumonisins; however, others have found hydrolysed fumonisins in foods prepared from masa (Hopmans and Murphy, 1993). This is of interest because results of some investigations suggest that hydrolysed fumonisins exert the same toxicological effects in vitro (Gelderblom et al., 1993; Merrill et al., 1993) and the same liver cancer promoting activity in vivo (Hendrich et al., 1993) as the parent compounds. However, others (Gelderblom et al., 1993) have observed that, in contrast to FB1 and fumonisin B2 (FB2), hydrolysed fumonisins B~ (HFB1) and B2 (HFB2) neither adversely affected body weight nor induced formation of y-glutamyl transpeptidase (GGT)-positive foci in the liver when fed (1000 ppm in the diets) to rats and suggested that HFBI and HFB2 were not absorbed from the gastrointestinal tract. Given the worldwide use of corn and the regional popularity of masa-based tortillas, the health effects of fumonisin-contaminated corn after nixtamalization should be determined. Furthermore, because nixtamalization chemically alters fumonisins, the usefulness of this process as well as other treatments as potential detoxification strategies should be investigated. The authors have previously reported (Voss et al., 1990) that extracting fumonisincontaminated culture material with water significantly reduced its hepatotoxicity. The purpose of the

present study was twofold: to determine the subchronic, organ-specific effects of nixtamalized culture material in vivo and to compare the effectiveness of nixtamalization and water extraction for detoxifying F. moniliforme culture material using hepatotoxicity and nephrotoxicity induced by fumonisin-containing culture material as endpoints. MATERIALS AND METHODS

Culture material preparation and processing

Corn culture material ofF. moniliforme strain MRC 826 (W.F.O. Marasas, MRC, Tygerburg, Republic of South Africa) was prepared, freeze-dried and ground to a powder as previously described (Voss et al., 1990). After mixing, approximately one-third (800 g) of the culture material was nixtamalized according to the specifications of Hendrich et al. (1993). Briefly, two 400-g aliquots of the freeze-dried, powdered culture material were heated (80-100°C) for 1 hr in 1.2% aqueous calcium hydroxide (250 g culture material/ litre) and the mixture steeped overnight. The nixtamalized culture material was then washed (three times) with 1600 ml deionized water. After the final wash, the nixtamalized product was vacuum filtered and air-dried. The dried nixtamalized culture material was then reground and remixed. Another 800 g culture material (two 400-g aliquots) was washed (three times) with deionized water (250 g culture material/litre), vacuum filtered, air-dried, reground and remixed in the same manner. The dried culture material and nixtamalized culture material were stored frozen. Chemical analyses

FB1 and HFB1 concentrations of the formulated diets were calculated values based on analysis of the

COOH

I COCH2CHCH2COOH O

OH

21CH30

I

OH

23CH 3 OH

NH 2

COCHE~HCHECOOH COOH Fumonisin

B 1

OH

OH

21CH30H23CH3

OH

OH

NI"I2

Hyrdrolysed fumonisin B l Fig. I. Chemical structure of fumonisins and hydrolysed fumonisins.

Toxicity of nixtamalized corn culture material unprocessed, nixtaraalized and water-extracted culture material for FBI and HFBI. FBI was determined by HPLC using previously published methods (Voss et al., 1995). For H F B I , the nixtamalized culture material (1 g) was extracted overnight with 2 N I¢~OH (20 ml) at 70°C. The K O H extract was adjusted to pH 4 with concentrated HC1 and then partitioned with butanol (four times 10 ml). The butanol fractions were combined and the solvent removed under a stream of nitrogen. The concentrated extract was dissolved in 2 ml methanol for orthophthalate anhydride (Pierce, Rockford, IL, USA) derivatization and HPLC analysis, which were accomplished in the same manner as for FBI. Quantitation was accomplished by comparison of the peak area to that of an HFB1 standard. The standard was prepared by base hydrolysis of authentic FB1 and compared by HPLC analyses to two separate reference standard~; of HFBI (R. D. Plattner, National Center for Agricultural Utilization Research, Agricultural Research Service, USDA, Peoria, IL, USA and R. T. Riley, Toxicology and Mycotoxin Research Unit, Agricultural Research Service, USDA, Athens, GA, USA) which had been verified by HPLC and gas chromatography/mass spectrometry (Plattner and Branham, 1994).

Diets All diets contained 90% (w/w) basal feed (RMH 3000 certified rode~at diet, Agway, Waverly, NY, USA, containing < 0 . 5 p p m FB1 and < 2 p p b aflatoxin). Culture material (5%, w/w) (CM diet), nixtamalized culture material (5% culture material equivalent weights) (NX diet), or water-extracted culture material (5% culture material equivalent weights) (WE diet) and ground sound seed corn were mixed with the basal feed using a Patterson Kelley blender with a stirring intensifier bar. The control diet (SC diet) was mixed using sound seed corn (10%, w/w) and basal feed only. Diets were prepared and dispensed fresh wee]dy.

Animals and animal husbandry Male Sprague-Dawley rats (Harlan Industries, Indianapolis, IN, USA), 3 wk of age on receipt, were individually housed in stainless-steel, wire-mesh cages placed in an environmentally controlled (20-24°C) room having a 12-hr light/dark cycle. Food and fresh tap water were available ad lib. Acclimatized (1 wk) animals were assilgned to groups by random stratification according to body weight.

35o

_

30°

"~

250

625

• SC • WE • NX

*

0

150

, 0

1

I 2

I 3

I 4

Wk Fig. 2. Body weights of male rats fed SC (control) (Q), WE (&), NX ( 0 ) or CM (11) diet for 4wk. Each point represents the mean of 10 observations/group; error bars denote standard deviations. *t:~, Groups significantly different from each other at wk 1-4 (P < 0.05). Data were analysed by ANOVA/Duncan's Multiple Range Test, Student's t-test or Kruskal-Wallis/Distribution Free Multiple Comparisons, as given in the scheme of Gad and Weil (1982). aminotransferase (ALT), aspartate aminotransferase (AST), alkaline phosphatase (AP), G G T and lactate dehydrogenase (LDH) activities, as well as serum total cholesterol, triglyceride, total bilirubin, blood urea nitrogen and creatinine concentrations (Voss et al., 1995). After 4 wk the rats were killed, and examined by autopsy. The adrenal glands, brain, heart, kidneys, liver, lungs and testes were excised and weighed. These organs or representative portions thereof were fixed in 10% Carson's buffered formalin, processed and microscopically examined. Data were statistically analysed using appropriate parametric and non-parametric methods according to the scheme of Gad and Weil (1982). RESULTS

Analytical chemistry The unprocessed culture material contained 1420 ppm (n = 4; range 1240-1590 ppm) FB1. After processing, the water-extracted culture material (total air-dried weight --- 480 g) contained 271 ppm (n = 4; range 243-296 ppm). No detectable FBI remained in the nixtamalized culture material (total air-dried weight = 608 g). However, 1528 ppm (n = 3; range 1512-1545ppm) of HFB1 was found in the nixtamalized culture material. Thus, the calculated concentrations of FBI in the CM and the WE diets were approximately 71 and 8 ppm, respectively, while the calculated concentration of HFB1 in the NX diet was 58 ppm.

Experimental

Body weight and food consumption

Groups of 10 rats were fed the CM, NX, WE or SC diet for 4 wk. The animals were observed daily for toxicity and weighed weekly. Food consumption was measured weekly, l/~lood specimens were collected after 2 and 4 wk for measurement of serum alanine

Gross body weight (g) (Fig. 2) and cumulative weight gain of all groups differed significantly at wk 1-4 in the order: SC d i e t > W E d i e t > N X diet > CM diet. Weekly weight gain of rats fed the WE diet was reduced compared with that of the

626

K. A. Voss et al.

,oo

diet was decreased to wk 2. Thereafter, prorated food consumption of these groups was comparable with that of the controls (SC diet). Prorated weekly food consumption of rats fed the CM diet was significantly decreased during wk 1-3, but was significantly greater than groups fed the SC, WE or NX diet during wk 4.

Diet SC D WE

~ 9o~

NX

a -b a

vo L "~ o

E~] CM

70

Serum chemistry

8 "- - 60 50 40

3 4 Wk Fig. 3. Weekly food consumption of male rats fed SC (control), WE, NX or CM diet for 4wk. Each point represents the mean of 10 observations/group; error bars denote standard deviations, a, b, c. Groups not sharing letters differ significantly(P < 0.05). Data were analysed by ANOVA/Duncan's Multiple Range Test, Student's t-test or Kruskal-Wallis/Distribution Free Multiple Comparisons, as given in the scheme of Gad and Weil (1982). 1

2

control group (SC diet) only during wk 1. In contrast, weekly weight gain of groups fed the CM or the NX diet was similar to that of the controls only during wk 4. Weekly food consumption (g/rat) of rats fed the NX or the WE diet was less than the controls, while food consumption of rats given the CM diet was less than all other groups. However, when prorated for body weight (Fig. 3), weekly food consumption (g/100 g body weight) of rats fed the NX diet was decreased during wk 1 and that of rats fed the WE

350

350

Diet [-q SC WE mNX k~! CM

300 250

Serum chemistry findings were essentially the same after both wk 2 and 4. There were no significant differences in LDH and there were no significant differences between the control and WE-diet groups for any other serum chemistry indicators of hepatotoxicity (Fig. 4). However, ALT, AST and AP of the group fed the NX diet were significantly elevated compared with that of the controls and WE-diet groups, while cholesterol and triglycerides of the NX-diet group were increased compared with that of the controls only. All indicators of hepatotoxicity (except LDH) were increased in the group fed the CM diet compared with the control or WE-diet groups. LDH, cholesterol and bilirubin excepted, indicators of hepatotoxicity were also increased in the CM-diet group compared with those in rats fed the NX diet. Serum blood urea nitrogen (BUN) concentrations of the CM, NX and WE-diet groups were slightly, but significantly, increased compared with those of the controls. Average BUN of these groups ranged (mean ___SD) from 19.3 + 1.72 mg/dl (NX diet) to 21.0 + 3.09 mg/dl (CM diet). Serum creatinine concentration of the WE-diet group (0.53 + 0.05 mg/ dl) was also significantly increased, while serum creatinine concentrations of the NX (0.65 +__0.06 mg/

300

t

250 200

200

tt

150 100

50

ALT

oi

150 100

AST AP GGT CHOL TRGD BI L (x0.5) (x0.1) (xl00) (xlO0) Fig. 4. Selected serum chemical findings in male rats fed SC (control), WE, NX or CM diet for 4 wk. Each value represents the mean of l0 observations/group; error bars denote standard deviations. *t~ Groups not sharing the same symbol differ significantly ( P < 0.05). CHOL =cholesterol; TRGD = triglb,cerides; BIL = bilirubin. Data were analysed by ANOVA/Duncan's Multiple Range Test, Student's t-test or Kruskal-Wallis/Distribution Free Multiple Comparisons, as given in the scheme of Gad and Weil (1982).

Toxicity of nixtamalized corn culture material 120

I'-1 SC m

WE 1

NX ~

CM

110 100 90 •

80

"t

70 60 50 G

%BWT%BRN Li~er

G

%BWT %BRN Kidney

Fig. 5. Liver and kidney weights of male rats fed SC (control), WE, NX or CM diet for 4 wk. Each value represents the mean of 10 observations/group; error bars denote standard deviations. *t:~ Groups not sharing the same symbol differ significantly (P < 0.05). G = weight in grams; %BWT = per cent body weight = (organ weight (g)/body weight (g))x 100; % B R N = p e r cent brain weight = (organ weight (g)/brain weight (g) x 100. Data were analysed by ANOVA/Duncan's Multiple Range Test, Student's t-test or Kruskal-Wallis/Distribution Free Multiple Comparisons, as given in the scheme of Gad and Weil (1982).

dl) and CM-diet (0.64 + 0.08 mg/dl) groups were significantly greater than both the control and WE-diet groups. Control B U N and creatinine concentrations were 17.2 _ 1.58 and 0.47 _ 0.05 mg/ dl, respectively.

Organ weights Relative (% body weight) kidney and liver weights of the NX-diet group and relative kidney weight of the WE-diet group were decreased compared with those of the controls (Fig. 5). Relative kidney weight, but not relative liver weight, of rats fed the C M diet was also decreased. When expressed as organ-tobrain weight ratio, kidney weights of all groups differed significantly; SC diet > WE diet > N X diet > C M diet, while liver weights of groups fed the N X or C M diet were less than the W E diet and control groups. Relative adrenal weight and adrenalto-brain weight ratios of animals fed the C M diet were increased, ave:raging 150 and 111% of the control values, respectively. N o differences in adrenal weight were found a m o n g the control (SC diet), W E and NX-diet groups. Other minor statistically significant organ weight differences among groups were found (data not shown), but were unaccompanied by any apparent histopathological abnormalities and, therefore, were considered secondary to overall body weight effects of the C M and N X diets.

Histopathology Histopathological effects related to dietary treatment were limited to the liver, kidneys and adrenals. N o liver lesions were found in the controls (SC diet) (Table 1) or in eight of 10 animals fed the W E diet.

627

One animal from the WE-diet group exhibited increased (diffuse hepatocellular) cytoplasmic vacuolation but, because necrosis, apoptosis, or increased mitotic index were not observed, the finding was not attributed to fumonisin exposure. The remaining rat in the WE-diet group showed minimal signs o f FBl-induced hepatopathy, as indicated by the presence of single-cell necrosis (putatively apoptotic bodies) (Howard et al., 1995) and mitotic figures (Plate 1). Single-cell necrosis, mitosis and apparent 'collapse' of the centrilobular parenchyma were found in the livers of all rats fed the N X diet. However, the overall parenchymal cytoarchitecture of this group remained intact and was closer in appearance to the unaffected W E diet and control groups than the CM-diet group. In the latter group, lesions were found in all animals and were obviously more advanced than in the NX-diet group. Diffusely scattered single-cell necrosis, increased mitotic figures, more extensive parenchymal collapse, heterogeneity of both the size and the nucleus-to-cytoplasm ratio of hepatocytes, as well as sporadic fibrosis with centrilobular 'bridging' were readily observed in rats fed the C M diet. Nephropathy was present in all animals fed the CM, N X or W E diet and, based on subjective scoring (Table 1), no significant differences in the extent or severity of these lesions were found. Lesions were limited to the outer medulla (sometimes referred to a s the cortico-medullary junction) and consisted of single-cell necrosis (putatively apoptosis) (Howard et al., 1995), sloughing of eosinophilic epithelial cells with pyknotic nuclei into the tubular lumina, cytoplasmic basophilia, decreased epithelial height with an apparent increase in nucleus-to-cytoplasm ratio, and occasional mitotic figures (Plate 2). Cytoplasmic vacuolation of the zona fasciculata of the adrenal cortex was found in all rats fed the C M diet, six of 10 rats fed the N X diet and seven o f 10 Table I. Incidence and severity of microscopic liver and kidney effects Incidence and severity of lesions in animals (n = 10/group) according to diet Organ SC diet CM diet NX diet WE diet Liver Normal (0) 10~ 0b 0b 9~ Hepatopathy Minimal (1) 0 0 8 1 Mild (2) 0 0 2 0 Moderate (3) 0 10 0 0 Mean score 0" 3.0~ 1.2b 0.1" Kidney Normal (0) 10" 0b 0~ 0b Nephropathy Mild (1) 0 0 2 4 Intermediate (2) 0 5 4 6 Moderate (3) 0 5 4 0 Mean score 0" 2.5b 1.9b 1.6b ".bValues in rows with different letters are significantly different (P < 0.05). Data were analysed by Kruskal-Wallis/Distribution Free Multiple Comparisons, (mean score) or Fishers' Exact Probability Test (incidence) as given in the scheme of Gad and Weil (1982).

628

K.A. Voss et al.

animals fed the WE diet. Lesions in the CM-diet group were most (subjectively) advanced. In contrast, mild cytoplasmic vacuolation was noted only in one control animal. Adrenal lesions were considered to be stress related and may have been secondary to hepatopathy.

DISCUSSION Nixtamalization, which is essentially a base hydrolysis, converts fumonisins to their hydrolysed forms (Hendrich et al., 1993; Hopmans and Murphy, 1993). Following nixtamalization, no detectable FB1 remained in the culture material. However, in contrast to previous investigations in which the 'fumonisin equivalents' found in corn (Sydenham et al., 1995) or F. proliferatum culture material (Hendrich et al., 1993) after nixtamalization were decreased significantly, the 'fumonisin equivalents' recovered from the nixtamalized culture material as HFB1 increased. As a result, 71 ppm FB1 were present in the CM diet and 103ppm 'FB1 equivalents' (58 ppm HFB1) were present in the NX diet. The factors contributing to this are unknown. Differences in fumonisin-matrix interactions related to the chemical composition of the F. moniliforme culture material, F. proliferatum culture material (Hendrich et al., 1993) and corn (Sydenham et al., 1995) may influence the amount of HFB1 which partitions into the liquid phase during nixtamalization. Secondly, unlike in previous investigations, the nixtamalized culture material was base hydrolysed (with 2 N KOH) during the first step of the analytical procedure. It is possible that tightly bound HFB1 (or FB1), which was otherwise not readily extractable, was released by this step. Additional research is needed to clarify this point, specifically to determine whether differences in matrix composition (culture materials v. corn), temperature during nixtamalization (80-100°C v. ambient), or analytical procedures (presence or absence of KOH treatment) affect the extractability of HFBI. In any event, nixtamalization did not detoxify the culture material. Weight gain of the NX-diet group was decreased compared with that in the control and WE-diet groups. The NX diet was unequivocally hepatotoxic, although less so than the CM diet, and was also nephrotoxic. Despite the absence of detectable FB1 in the culture material following nixtamalization, both liver and kidney lesions found in the NX-diet group were qualitatively indistinguishable from those found in rats fed fumonisin-containing culture material in this (CM diet) and other studies (Voss et al., 1990 and 1992), corn naturally contaminated with F. moniliforme and fumonisins (Voss et al., 1989), or purified FB1 (Gelderblom et al., 1991; Voss et al., 1993 and 1995). Similarly, Hendrich et al. (1993) found that nixtamalized F.

proliferatum culture material containing 8-11 ppm

H F B I , but no detectable FB1, had the same (liver) cancer-promoting activity in diethylnitrosaminetreated rats as did unprocessed culture material containing about 45 ppm FB1, but no detectable HFB1. Therefore, nixtamalized corn contains one or more toxic constituents, putatively hydrolysed fumonisins such as HFB 1, which induce fumonisin-like liver and kidney effects. Therefore, even in the absence of measurable fumonisins, nixtamalized corn may remain toxic. Gelderblom et al. (1993) have shown that HFB1 and hydrolysed FB2 (HFB2) were more toxic than the parent fumonisins when LDH release from cultured primary hepatocytes was the indicator of toxicity. Furthermore, fumonisins specifically disrupt cellular sphingolipid metabolism causing, among other things, increased levels of the sphingoid base sphinganine and an increased sphinganine/sphingosine ratio. This has been demonstrated in vivo in the liver and kidney of rats fed purified FBI for 4 wk (Riley et al., 1994), and in primary hepatocytes in vitro (Merrill et al., 1993). In preliminary studies (data not shown) increased sphinganine/sphingosine ratios were also noted in precision-cut rat liver slices treated with HFB1 in vitro. Given this evidence and the similarities in liver and kidney lesions found in rats fed fumonisins or hydrolysed fumonisins, the effects of hydrolysed fumonisins and nixtamalized culture materials on sphingolipids in vivo should be investigated. Other evidence suggests that the in vivo effects of nixtamalized culture materials may be due to previously unidentified compounds. Gelderblom et al. (1993) reported that, unlike the parent fumonisins, hydrolysed homologues had no effect on body weight and outward appearance, and had no initiation activity, for example induction of macroscopic nodules and/or GGT-positive foci in the liver when fed to rats. They proposed that, although inherently more cytotoxic than their parent fumonisins, HFB1 and HFB2 were apparently not toxic in vivo because they may not have been absorbed from the gastrointestinal tract. However, fumonisins may exert serum chemical, sphingolipid, organ weight and/or histopathological effects at doses that do not cause significant weight loss or other outward signs of toxicity (Riley et al., 1994; Voss et al., 1993 and 1995). Furthermore, measurement of initiation potential may not adequately assess either toxicity or bioavailability because toxigenicity per se and initiation may not occur by the same molecular mechanism. Clearly, FBI and FB2 possess only weak initiation potential even when dietary levels are quite high (Gelderblom et al., 1993 and 1994), while their cancer-promoting activity (Gelderblom et al., 1988) and organ-specific toxicities (Gelderblom et al., 1988 and 1991; Voss et al., 1993 and 1995), which occur at much lower exposure levels, have been well documented. In any event, it remains to be proven

Toxicity of nixtamalized corn culture material

Plate 1. Photomicrograph of liver from a male rat fed the NX diet containing 58 ppm HFBI for 4 wk. Degenerate hepatocytes (curved arrow) were widely scattered throughout the otherwise normal-appearing parenchyma (open arrow), but were more numerous around central veins (closed arrow) (haematoxylin and eosin, x 475). Plate 2. Photomicrograph of kidney (outer medulla) from a male rat fed the NX diet containing 58 ppm HFBI for 4 wk. Note the apoptotic epithelial cells having (short arrow) a decreased cytoplasm-to-nucleus ratio, and sloughing of necrotic epithelial cells into the tubular lumina (long arrow). Cytoplasm was basophilic and le'.~ions were generalized throughout the outer medulla with focal extension into the cortex or lower medulla (haematoxylin and eosin, x 475).

629

Toxicity of nixtamalized corn culture material whether the hydrolysed fumonisins are in fact responsible for the fumonisin-like hepato- and nephro-toxicities o f the nixtamalized culture material. The authors previously found that fumonisins are water-soluble and that water extraction decreases both the fumonisin content and hepatotoxic potential of F. moniliforme culture material (Voss et al., 1990). In this study, water extraction was more effective than nixtamalization in reducing toxicity as demonstrated by the relatively higher weight gains and the absence o f hepatoto~:icity in the WE-diet group. While the FB1 concentration of the W E diet (8 ppm) was well below the no-observed-effect level ( N O E L ) o f 50 ppm for hepatol:oxic effects, which was found in male rats fed FBI for 4 wk (Voss et al., 1993), it was probably sufficient to cause nephrotoxicity. Renal lesions have been shown to occur in fumonisin-fed rats, particularly males, at dietary levels well below the N O E L for hepatotoxicity (Voss et al., 1993 and 1995); reported N O E L s for nephrotoxicity in male rats were less than 15 ppm following 4 wk o f exposure (Voss et al., 1993) and 3 ppm following 13 wk of exposure (Voss et al., 1995). In summary, nixtamalization of fumonisin-containing culture materials failed to eliminate their organ-specific toxicity, as illustrated by the present results, and their cancer-promoting activity, as reported by Hendrich et al. (1993), and was less effective than extraction with water for reducing their hepatotoxicity. Given the popularity of nixtamalized corn products and the fumonisin-like effects of the nixtamalized culture materials, additional toxicological assessments of hydrolysed fumonisins and nixtamalized corn are clearly warranted.

Acknowledgements--The technical expertise of R. Bennett, N. Brice, G. Huff, P. Malcom, M. Nelms, P. Stancel and K. Tate are appreciated. The consultations of R. Plattner and R. T. Riley are gratefully acknowledged.

RFA~'ERENCES

Bullerman L. B. and Tsai W.-J. J. (1994) Incidence and levels of Fusarium monilforme, Fusarium proliferatum and fumonisins in corn ard corn-based foods and feeds. Journal of Food Protec~ion 57, 541-546. Chu F. S. and Li G. Y. (1994) Simultaneous occurrence of fumonisin B1 and other mycotoxins in moldy corn collected from the People's Republic of China in regions with high incidences of esophageal cancer. Applied and Environmental Microbiology 60, 847-852. Colvin B. M., Cooley A. J. and Beaver R. W. (1993) Fumonisin toxicosis in swine: clinical and pathologic findings. Journal of Veterinary Diagnostic Investigations 5, 232-241. Gad S. C. and Well C. S. (1982) Statistics for toxicologists. In Principles and Methods of Toxicology. Edited by A. W. Hayes. pp. 273-320. Raven Press, New York. Gelderblom W. C. A., Cawood M. E., Snyman S. D. and Marasas W. F. O. (1994) Fumonisin BI dosimetry in relation to cancer initiation in rat liver. Carcinogenesis 15, 209-214.

631

Gelderblom W. C. A., Cawood M. E., Snyman S. D., Vleggaar R. and Marasas W. F. O. (1993) Structure-activity relationships of fumonisins in short-term carcinogenesis and cytotoxicity assays. Food and Chemical Toxicology 31, 407-414. Gelderblom W. C. A., Jaskiewicz K., Marasas W. F. O., Thiel P. G., Horak R. M., VIcggaar R. and Kriek N. P. J. (1988) Fumonisins---novel mycotoxins with cancer promoting activity produced by Fusarium moniliforme. Applied and Environmental Microbiology 54, 1806-1811. Gelderblom W. C. A., Kriek N. P. J., Marasas W. F. O. and Thiel P. G. (1991) Toxicity and carcinogenicity of the Fusarium moniliforme metabolite, fumonisin Bl in rats. Carcinogenesis 12, 1247-1251. 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 B1, a toxic metabolite of Fusarium moniliforme. Journal of Veterinary Diagnostic Investigations 2, 217-221. Hendrich S., Miller K. A., Wilson T. M. and Murphy P. A. (1993) Toxicity of Fusarium proliferatum-fermented nixtamalized corn-based diets fed to rats: effect of nutritional status. Journal of Agricultural and Food Chemistry 41, 1649-1654. Hopmans E. C. and Murphy P. A. (1993) Detection of fumonisins BI, B2 and B3 and hydrolyzed fumonisin BI in corn-containing foods. Journal of Agricultural and Food Chemistry 41, 1655-1658. Howard P. C., Thurman J. D., Lorentzen R. J., Voss K. A,. Bucci T. J. and Dooley K. L. (1995) The induction of apoptosis in the liver and kidneys of male and female F344 rats fed diets for 28 days containing the mycotoxin fumonisin B1. (Abstract). Proceedings of the American Association for Cancer Research 36, 785. Jaskiewicz K., van Rensburg S. J., Marasas W. F. O. and Gelderblom W. C. A. (1987) Carcinogenicity of Fusarium moniliforme culture material in rats. Journal of the National Cancer Institute 78, 321-325. Kellerman T. S., Marasas W. F. O., Thiel P. G., Gelderblom W. C. A., Cawood M. and Coetzer J. A. W. (1990) Leukoencephalomalacia in two horses induced by oral dosing of fumonisin BI. Onderstepoort Journal of Veterinary Research 57, 269275. Marasas W. F. O. (1993) Occurrence of Fusarium moniliforme and fumonisins in maize in relation to human health. South African Medical Journal 83, 383-384. Marasas W. F. O., Kellerman T. S., Gelderblom W. C. A., Coetzer J. A. W., Thiel P. G. and van der Lugt J. J. (1988) Leukoencephalomalacia in a horse induced by fumonisin B1 isolated from Fusarium moniliforme. Onderstepoort Journal of Veterinary Research 55, 197-203. Merrill A. H., Jr, Wang E., Gilchrist D. G. and Riley R. T. (1993) Fumonisins and other inhibitors of de novo sphingolipid biosynthesis. Advanced Lipid Research 26, 215-234. Murphy P. A., Rice L. G. and Ross P. F. (1993) Fumonisin B1, B2, and B3 content of Iowa, Wisconsin, and Illinois corn and corn screenings. Journal of Agricultural and Food Chemistry 41, 263-266. Nelson P. E., Desjardins A. E. and Plattner R. D. (1993) Fumonisins, mycotoxins produced by Fusarium species: biology, chemistry and significance. Annual Review of Phytopathology 31, 233-252. Pittet A., Parisod V. and Schellenberg M. (1992) Occurrence of fumonisins B1 and B2 in corn-based products from the Swiss market. Journal of Agricultural and Food Chemistry 40, 1352-1354. Plattner R. D. and Branham B. E. (1994) Labeled fumonisins: production and use of fumonisin Bl containing stable isotopes. Journal of AOA C International 77, 525-532. Riley R. T., Hinton D. M., Chamberlain W. J., Bacon C. W., Wang E., Merrill A. H. Jr and Voss K. A. (1994)

632

K . A . Voss et al.

Dietary fumonisin BI induces disruption of sphingolipid metabolism in Sprague-Dawley rats: a new mechanism of nephrotoxicity. Journal of Nutrition 124, 594-4503. Sanchis V., Abadias M., Oncins L., Sala N., Vinas I. and Canela R. (1994) Occurrence of fumonisins BI and B2 in corn-based products from the Spanish market. Applied and Environmental Microbiology 60, 2147-2148. Sydenham E. W., Shephard G. S. and Thiel P. G. (1992) Liquid chromatographic determination of fumonisins B 1, B2, and B3 in foods and feeds. Journal of AOAC International 75, 313-318. Sydenham E. W., Shephard G. S., Thiel P. G., Marasas W. F. O., Rheeder J. P., Peralta Sanhueza C. E., Gonzalez H. H. L. and Resnik S. L. (1993) Fumonisins in Argentinian field-trial corn. Journal of Agricultural and Food Chemistry 41, 891-895. Sydenham E. W., Shephard G. S., Thiel P. G., Marasas W. F. O. and Stockenstrom S. (1991) Fumonisin contamination of commercial corn-based human foodstuffs. Journal of Agricultural and Food Chemistry 39, 2014--2018. Sydenham E. W., Stockenstrom S., Thiel P. G., Shephard G. S., Koch K. R. and Marasas W. F. O. (1995) Potential of alkaline hydrolysis for the removal of fumonisins from contaminated corn. Journal of Agricultural and Food Chemistry 43, 1198-1201. Thiel P. G., Marasas W. F. O., Sydenham E. W., Shephard G. S. and Gelderblom W. C. A. (1992) The implications of naturally occurring levels of fumonisins in corn for human and animal health. Mycopathologia 117, 3-10. Ueno Y., Aoyama S., Sugiura Y., Wang D-S., Lee U-S., Hirooka E. Y., Hara S., Karki T., Chen G. and Yu S-Z. (1993) A limited survey of fumonisins in corn and

corn-based products in Asian countries. Mycotoxin Research 9, 27-34. Voss K. A., Chamberlain W. J., Bacon C. W., Herbert R. A., Walters D. B. and Norred W. P. (1995) Subchronic feeding study of the mycotoxin fumonisin BI in B6C3FI mice and Fischer 344 rats. Fundamental and Applied Toxieology 24, 102-110. 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 B1. Natural Toxins 1, 222 228. Voss K. A., Norred W. P. and Bacon C. W. (1992) Subchronic toxicological investigations of F. moniliformecontaminated corn, culture material and ammoniated culture material. Mycopathologia 117, 97-104. 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 leukoencephalomalacia. Food and Chemical Toxieology 27, 89-96. Voss K. A.. Plattner R. D., Bacon C. W. and Norred W. P. (1990) Comparative studies of hepatotoxicity and fumonisin B1 and B2 content of water and chloroform/ methanol extracts of Fusarium moniliforme strain MRC 826 culture material. Mycopathologia 112, 81-92. Yoshizawa T., Yamashita A. and Luo Y. (1994) Fumonisin occurrence in corn from high- and low-risk areas for human esophageal cancer in China. Applied and Environmental Microbiology 60, 1626-1629. Zoller O., Sager F. and Zimmerli B. (1994) Vorkommen von fumonisinen in lebensmitteln. Mittelungen aus dem Gebiete der Lebensmitteluntersuchung und Hygiene 85, 81 99.