Exp Toxic Pathol 2002; 53: 481–487 URBAN & FISCHER http://www.urbanfischer.de/journals/exptoxpath 1

Department of General and clinical pathology and 2Department of Surgery, Faculty of Veterinary Medicine, Thracian University, Stara Zagora, Bulgaria 3 Central Science Laboratory, Ministry of Agriculture, Fisheries and Food, Sand Hutton, York, UK 4 Department of Biochemistry, Imperial College of Science, Technology and Medicine, London, UK

Experimental one year ochratoxin A toxicosis in pigs S. D. STOEV1, M. PASKALEV2, S. MACDONALD3, and P. G. MANTLE4 With 5 figures and 1 table Received: August 20, 2001; Revised: October 12, 2001; Accepted: October 19, 2001 Address for correspondence: S. D. STOEV, Department of General and clinical pathology, Faculty of Veterinary Medicine, Thracian University, Students campus, 6000 Stara Zagora, Bulgaria; Fax: ++35942 45101; E-mail: [email protected] Key words: Mycotoxins; Mycotoxic porcine nephropathy; Ochratoxicosis; Ochratoxins; Ochratoxin A; Pathology, Serum analysis. Abbreviations: MPN – mycotoxic porcine nephropathy, OTA – ochratoxin A.

Summary

Introduction

Mild mycotoxic nephropathy was induced in 6 pigs by a diet containing ochratoxin A at 800 ppb, several times higher than that naturally encountered in some feed for pig production in Bulgaria. The nephropathy was expressed only as slightly hypertrophied kidneys with a faintly mottled surface, discernible at the end of the experiment to a skilled observer but probably not recognisable in routine slaughterhouse processing. Histological examination showed two types of changes: degenerative – affecting epithelial cells in some proximal tubules of pigs after 6 months, and proliferative changes in the interstitium which predominated after 1 year of exposure to ochratoxin A. Telangiectasis and lymph stasis were rarely seen. The renal lesions were similar to those described for classical mycotoxic porcine nephropathy formerly encountered in Denmark, but they were rather different from the porcine nephropathy which occurs spontaneously in Bulgaria. Measurement of ochratoxin A in serum provided analytical values complementary to feed intake and with similar concentration values. It also showed both accumulation with time, from 3 months to 6 months (approximately 1 ppm), and a 2-fold range of values within a group eating from a common feed source, as in commercial pig production. Mild symptomatology in this long, single-mycotoxin experiment serves to lessen somewhat the current perception of the direct renal toxicity of ochratoxin A alone, though a role in multi-toxin contexts is unquestioned.

Mycotoxic porcine nephropathy (MPN) has been comprehensively reviewed (KROGH 1976) and the toxic effect of ochratoxin A (OTA) was considered to be responsible to this nephropathy (KROGH et al. 1974; SZCZECH et al. 1973). Mycotoxins and especially OTA have also been suggested as a possible aetiological agent in Balkan endemic nephropathy in humans (KROGH 1972; STOEV 1998), which is closely associated with a high frequency of carcinoma in the renal pelvis, ureter and urinary bladder (CHERNOZEMSKY et al. 1977; CUKURANOVIC et al. 1991; NIKOLOV and CHERNOZEMSKY 1990; RADOVANOVIC et al. 1991). Similarly, neoplasia (fibroma, adenoma and fibroadenoma) in kidneys were recorded in the Bulgarian cases of MPN (STOEV et al. 1998a), but they have not been reported elsewhere. These neoplastic changes could be due to the carcinogenicity of OTA, as has been previously reported in mice (KANISAWA and SUZUKI 1978; BENDELE et al. 1985) and rats (BOORMAN et al. 1989) or to some other mycotoxins encountered in the same feeds. While agreeing that the most important toxicological target of OTA in the pig is the kidney, the principal descriptions of the pathology of MPN vary considerably with respect to some other details, and according to the dosing regime and the duration of OTA exposure. Secondary bacterial enteric disease, as a result of OTA-in0940-2993/02/53/06-481 $ 15.00/0

481

duced immunosuppression, may also influence significantly the complex clinicomorphological picture of MPN (STOEV et al. 2000b). Our recent studies on naturally occurring porcine nephropathy in Bulgaria have identified it as a notable disease, which can only partly be attributed to OTA in the feed (STOEV et al. 1998a). This nephropathy was characterized by vascular lesions, renal haemorrhages, enlargement of the renal lymph nodes and furrowed surface of kidneys, in contrast to the classic MPN as described in Denmark (KROGH 1976). The incidence of MPN in Bulgaria is higher, possibly by one to two orders of magnitude, than it was in the Danish bacon industry in the 1960s and 70s (STOEV et al. 1998a). However, the average concentrations of OTA in some feeds for pigs in Bulgaria (100–200 ppb) were substantially lower than the 1–2 ppm required to reproduce in Denmark the classical MPN of a severity similar to that observed in Bulgaria (STOEV et al. 1998a, b, c). It seems, therefore, that Bulgarian MPN may have a multitoxic aetiology, because it cannot be explained by the concentration of OTA alone. Experiments with diets contaminated artificially with both OTA and penicillic acid (STOEV et al. 2001) have indicated that mixtures of these mycotoxins given for up to 3 months can cause MPN reminiscent of that seen naturally in Bulgaria. Notably, this was achieved with OTA contamination close to some higher natural values. The present experiment was designed to study the effect of OTA alone, at 4–8 times the average natural contamination value, given continuously over one year. This has been much longer than has been attempted before, and resources were necessarily focused on providing the large amount of artificially contaminated feed to large animals which was particularly demanding during the latter months. Data collection in the experiment was therefore restricted to monitoring animal health and growth, and renal pathology at 6 and 12 months. The OTA content of serum was also measured at 2 stages by the most sensitive, specific and reproducible methodology to establish the range of values amongst animals feeding on a standard diet from the same source.

Materials and methods Ochratoxin A: Aspergillus ochraceus (isolate D2306, as used by TAPIA and SEAWRIGHT 1984, and STOEV et al. 2000b) was grown on sterilised shredded wheat (40 g) in 500 ml conical flasks, moistened by a 40% (v/w) addition of sterile water and incubated on a rotary shaker at 27 °C for 2 weeks (HARRIS and MANTLE 2001). The brown granular product, which bore no obvious sign of fungal growth or sporulation, was sterilised at 80 °C for 1 hour (yield 2 kg) and stored at –20 °C. A sample was analysed by HPLC with diode array detection for OTA and found to contain ~2 mg/g. This detection method is appropriate for analysis of material of high OTA concentration and for monitoring for other mycotoxins. No other mycotoxins (eg penicillic acid) were present, although there was typically a small ad482

Exp Toxic Pathol 53 (2002) 6

ditional proportion of the biologically-inactive deschloroanalogue ochratoxin B. The ochratoxin-rich moulded shredded wheat was intimately homogenised into pig ration (diluted at least 2000-fold) to give the required concentration of OTA in diet. Experimental design: Twelve young pigs (Landrace × Bulgarian white from a commercial stock source in Bulgaria, 6 male and 6 female) were purchased at about eight weeks of age and approximately equal body weight (12–14 kg) and then were castrated. After that, pigs were arranged in two groups of six (3 male and 3 female) and fed ad libitum either a good quality commercial pig starter ration (controls) or the same ration artificially contaminated with the fermented OTA-rich shredded wheat to give 800 ppb OTA. Fresh drinking water was available ad libitum. Before commencing the OTA-contaminated diet the tomography of both kidneys was imaged and measured by ultrasonography (Dynamic Imaging Ltd., Livingstone, Scotland) and found to be within normal parameters. At 6 months, body fat precluded ultrasonography as a feasible way of assessing abnormal kidney dimensions. To protect against developing secondary bacterial enteric disease, to which pigs given OTA may be susceptible (STOEV et al. 2000b), pigs were given Dor-novTM (Dorvet Ltd, Nes-Ziona, Israel) at a prophylactic dose in drinking water for the first experimental week, and were then similarly given RidzolTM (Merck, Sharp & Dohme, Holland) up to day 14. The same 2-week prophylactic treatment was repeated after 1 month of the experimental period. Pigs were weighed again after 6 and 12 months. At 6 months, one male and one female from the experimental and control groups were exposed to unilateral nephrectomy of the left kidneys which were weighed immediately after excission. The neuroleptic Acepromazini maleas (CombistressTM – KELA Laboratoria B-2320, Hoogstraten, Belgium) was employed for premedication of general anaesthesia and Thiopental-sodium (ThiopentalTM – BIOCHEMIE GmbH, Vienna, Austria) was used for achieving the combined narcosis (VLAKHOV et al. 1977). At the end of the 12 month experimental period, the other 4 pigs from each of the experimental and control groups were slaughtered and examined generally, but particularly for morphological changes in their left kidneys which were also weighed at that time. Kidney tissue for histological examination was fixed in 10% neutral buffered formalin. Fixed tissues were embedded in paraffin wax, sectioned at 6 µm and stained with haematoxylin-eosin and PAS. The non-parametric Mann-Whitney-U-test and Student’s t-test were used to estimate the significance of the differences between the mean values of the body and the kidney weights in pigs of control and experimental groups. Measurement of OTA concentration in serum: Samples (1ml) were mixed with 10 ml 0.5 M phosphoric acid in 2M NaCl and extracted exhaustively twice with an equal volume of chloroform. Combined extract was evaporated and the residue dissolved in 20 ml 5% acetonitrile in PBS for processing through an immunoaffinity column (Ochraprep, Rhone Diagnostics) according to the Manufacturer’s protocol, yielding the OTA in 4 ml of eluate. 0.4 ml of the eluate was analysed by reversed phase HPLC through a Spherisorb ODS 2 (5 µm) Excel column (2.5 cm

× 4.6 mm i.d.) in acetonitrile:water:acetic acid (99:99:2, v/v/v) at 1 ml/min. Post-column 1.1 M ammonium hydroxide solution was added at 0.3 ml/min to enhance OTA fluorescence which was detected at 440 nm (390 nm excitation wavelength). Two serum samples were also spiked with OTA (1000 ng/ml) for 30 min before extraction to determine analytical recovery (79%) so as to express OTA concentrations in serum allowing for protein binding and inefficiencies in the analytical process.

Results Clinical signs were not seen in any of the experimental pigs. A slight reduction in weight gain, significant only at 1 year, was only observed in the OTA-treated group (table 1). An increase of absolute and relative kidney weight measured at 6 months became just statistically significant at the end of the 12 month experiment. The only macroscopic lesions in the pigs fed on OTAcontaminated diet for 6 and 12 months were a few small grey-white foci on the kidney surface. The major renal histopathological changes were mainly in the epithelium of proximal tubules: different degrees of karyo- and plasmolysis, karyopyknosis or karyomegaly (fig. 1). Degenerative changes of moderate to marked cloudy swelling, granular (fig. 2) or vacuolar (fig. 3) degeneration were the main changes in the tubular epithelial cells. Some proximal tubules were occluded by the swollen epithelial cells, whereas others, especially in pigs slaughtered at the

end of 1 year experimental period, were atrophied and surrounded by connective tissue (figs. 4 and 5). Some dilated tubules were lined by squamous epithelium or contained proteinaceous and necrotic debris in the lumen. Cellular casts were only rarely observed in convoluted or collecting tubules. Telangiectasis and minute lymphatic cysts containing serous fluid and lined by endothelial cells were sometimes encountered in the cortex. Capillary endothelial proliferation occurred occasionally. In the interstitium of some renal cortical regions, there was limited proliferation of connective tissue and focal infiltration of mononuclear inflammatory cells (fig. 2). Focal interstitial fibrosis, glomerular sclerosis (fig. 4) and thickening of tubular basement membranes were only seen in 2 kidneys of pigs slaughtered after 1 year OTA exposure. There were no large retention cysts or pronounced interstitial fibrosis in any of the kidneys. There were no histopathological changes in kidneys of any of the control pigs. OTA analysis of the serum samples taken at the end of the 3rd and 6th month revealed the toxin in all control pigs (table 1) and showed the presence of OTA in the commercial pig starter ration. Serum concentration of OTA in experimental pigs was of the order of 1 ppm, but individually they varied by as much as a factor of 2 at 3 months, even though the group ate from the same source. Individuals selected randomly for unilateral nephrectomy reached higher values (c. 1.5 ppm) at 6 months. Perception of correlations with weight or sex is difficult in

Table 1. Concentration of OTA in serum and values of body weight (b.w.) and absolute (k. a. w.) or relative (k. r. w.) weight of left kidney in pigs exposed to 0.8 ppm OTA in feed for 3, 6 or 12 months and in control pigs. Groups

experimental

Sex

3rd month serum OTA (ng/ml)

6th month serum OTA (ng/ml)

male male male female female female

1048 1342 493 621 709 900 852 ± 126

male male male female female female

40 68 53 100 154 30 74 ± 18

mean ± SEM control

mean ± SEM

6th month

12th month

b. w. (kg)

k. a. w. (g)

k. r. w. (g/kg b.w.)

b. w. (kg)

k. a. w. (g)

k. r. w (g/kg b.w.)

1542 # # # # 1623 1582 ± 40

56 65 67 72 74 74 68.0 ± 2.8

141 # # # # 153 147.0 ± 6.0

2.52 # # # # 2.07 2.29 ± 0.22

155 160 170 195 205 190 179.2 ± 8.3*

# 266 289 293 261 # 277.2 ± 8.0*

# 1.66 1.70 1.50 1.27 # 1.53 ± 0.09*

37 # # # 126 # 81 ± 44

95 70 74 94 70 76 79.8 ± 4.7

112 # # # 110 # 111.0 ± 1.0

1.18 # # # 1.57 # 1.37 ± 0.19

230 190 195 220 210 210 209.2 ± 6.1

# 208 226 238 # 221 223.2 ± 6.2

# 1.09 1.16 1.08 # 1.05 1.09 ± 0.02

± SEM (standard error of the mean) * – significant difference compared with controls (p