Mycotoxins, selected: ochratoxins, trichothecenes, ergot (EHC 105, 1990)

INTERNATIONAL PROGRAMME ON CHEMICAL SAFETY

ENVIRONMENTAL HEALTH CRITERIA 105

SELECTED MYCOTOXINS:
OCHRATOXINS, TRICHOTHECENES, ERGOT

This report contains the collective views of an international group of
experts and does not necessarily represent the decisions or the stated
policy of the United Nations Environment Programme, the International
Labour Organisation, or the World Health Organization.

Published under the joint sponsorship of
the United Nations Environment Programme,
the International Labour Organisation,
and the World Health Organization

World Health Organization
Geneva, 1990

The International Programme on Chemical Safety (IPCS) is a
joint venture of the United Nations Environment Programme, the
International Labour Organisation, and the World Health
Organization. The main objective of the IPCS is to carry out and
disseminate evaluations of the effects of chemicals on human health
and the quality of the environment. Supporting activities include
the development of epidemiological, experimental laboratory, and
risk-assessment methods that could produce internationally
comparable results, and the development of manpower in the field of
toxicology. Other activities carried out by the IPCS include the
development of know-how for coping with chemical accidents,
coordination of laboratory testing and epidemiological studies, and
promotion of research on the mechanisms of the biological action of
chemicals.

WHO Library Cataloguing in Publication Data

Selected mycotoxins : ochratoxins, trichothecenes, ergot.

(Environmental health criteria ; 105)

1.Ochratoxins 2.Trichothecenes 3.Ergot alkaloids
I.Series

ISBN 92 4 157105 5 (NLM Classification: QW 630)
ISSN 0250-863X

The World Health Organization welcomes requests for permission
to reproduce or translate its publications, in part or in full.
Applications and enquiries should be addressed to the Office of
Publications, World Health Organization, Geneva, Switzerland, which
will be glad to provide the latest information on any changes made
to the text, plans for new editions, and reprints and translations
already available.

^(c) World Health Organization 1990

Publications of the World Health Organization enjoy copyright
protection in accordance with the provisions of Protocol 2 of the
Universal Copyright Convention. All rights reserved.

The designations employed and the presentation of the material
in this publication do not imply the expression of any opinion
whatsoever on the part of the Secretariat of the World Health
Organization concerning the legal status of any country, territory,
city or area or of its authorities, or concerning the delimitation
of its frontiers or boundaries.

The mention of specific companies or of certain manufacturers'
products does not imply that they are endorsed or recommended by the
World Health Organization in preference to others of a similar
nature that are not mentioned. Errors and omissions excepted, the
names of proprietary products are distinguished by initial capital
letters.

CONTENTS

ENVIRONMENTAL HEALTH CRITERIA FOR SELECTED MYCOTOXINS: OCHRATOXINS,
TRICHOTHECENES, AND ERGOT

INTRODUCTION

SUMMARY AND RECOMMENDATIONS FOR FURTHER RESEARCH

1. Ochratoxin A
1.1. Natural occurrence
1.2. Analytical methods
1.3. Metabolism
1.4. Effects on animals
1.5. Effects on man

2. Trichothecenes
2.1. Natural occurrence
2.2. Analytical methods
2.3. Metabolism
2.4. Effects on animals
2.5. Effects on man

3. Ergot
3.1. Natural occurrence
3.2. Analytical methods
3.3. Effects on animals
3.4. Effects on man

4. Recommendations for further research
4.1. General recommendations
4.2. Ochratoxin A
4.3. Trichothecenes
4.4. Ergot

I. OCHRATOXINS
I.1 Properties and analytical methods
I.1.1 Chemical properties
I.1.2 Methods for the analysis of foodstuffs and
biological samples
I.2 Sources and occurrence
I.2.1 Fungal formation
I.2.2 Occurrence in foodstuffs
I.2.2.1 Plant products
I.2.2.2 Residues in food of animal origin
I.3 Metabolism
I.3.1 Absorption
I.3.2 Tissue distribution
I.3.2.1 Animal studies
I.3.2.2 Studies on man
I.3.3 Metabolic transformation
I.3.4 Excretion
I.4 Effects on animals
I.4.1 Field observations
I.4.1.1 Pigs
I.4.1.2 Poultry

I.4.2 Experimental animal studies
I.4.2.1 Acute and chronic effects
I.4.2.2 Teratogenicity
I.4.2.3 Mutagenicity
I.4.2.4 Carcinogenicity
I.4.2.5 Biochemical effects and mode of action
I.5 Effects on man
I.5.1 Ochratoxin A, Balkan endemic nephropathy, and
tumours of the urinary system
I.6 Evaluation of the human health risks

II. TRICHOTHECENES

II.1 Properties and analytical methods
II.1.1 Physical and chemical properties
II.1.1.1 Physical properties
II.1.1.2 Chemical properties
II.1.2 Analytical methods for trichothecenes
II.1.2.1 Chemical methods
II.1.2.2 Immunological methods
II.1.2.3 Biological methods
II.2 Sources and occurrence
II.2.1 Taxonomic considerations
II.2.2 Ecology of trichothecene-producing fungi
II.2.3 Natural occurrence
II.2.3.1 Agricultural products
II.2.3.2 Trichothecenes in human foodstuffs
II.3 Metabolism
II.3.1 Absorption and tissue distribution
II.3.1.1 Animal studies
II.3.2 Metabolic transformation
II.3.3 Excretion
II.3.3.1 Animal studies
II.3.3.2 Excretion in eggs and milk
II.4 Effects on animals
II.4.1 Field observations
II.4.2 Effects on experimental animals
II.4.2.1 General toxic effects
II.4.2.2 Haematological and haemostatic changes
II.4.2.3 Disturbances of the central nervous
system
II.4.2.4 Dermal toxicity
II.4.2.5 Impairment of immune response
II.4.2.6 Carcinogenicity
II.4.2.7 Mutagenicity
II.4.2.8 Teratogenicity and reproductive effects
II.4.3 Biochemical effects and mode of action
II.4.3.1 Cytotoxicity
II.4.3.2 Inhibition of protein synthesis
II.4.3.3 Inhibition of nucleic acid synthesis
II.4.3.4 Alterations of cellular membranes
II.4.3.5 Other biochemical effects
II.4.4 Structure-activity relationships
II.4.5 Prevention and therapy of trichothecene toxicosis

II.5 Effects on man
II.5.1 Contemporary episodes of human disease
II.5.2 Historical Fusarium-related diseases
II.5.3 Skin irritation
II.5.4 Studies of haemostasis
II.5.5 Airborne trichothecene-related diseases
II.5.6 Toxicological information on man, obtained from
therapeutic uses
II.6 Evaluation of the human health risks

III. ERGOT

III.1 Properties and analytical methods
III.1.1 Chemical properties
III.1.2 Analytical methods for ergot and ergot
alkaloids
III.1.2.1 Ergot
III.1.2.2 Ergot alkaloids
III.2 Sources and occurrence
III.2.1 Fungal producers
III.2.2 Biosynthesis
III.2.3 Occurrence in foodstuffs
III.2.4 Fate of ergolines during food processing
III.3 Metabolism
III.4 Effects on animals
III.4.1 Field studies
III.4.2 Experimental animal studies
III.4.2.1 Cattle
III.4.2.2 Sheep
III.4.2.3 Poultry
III.4.2.4 Swine
III.4.2.5 Primates
III.5 Effects on man
III.5.1 Ergometrine-related outbreaks
III.5.2 Clavine-related outbreaks
III.6 Evaluation of the human health risks

REFERENCES

RESUME

RESUMEN

WHO TASK GROUP ON SELECTED MYCOTOXINS: OCHRATOXINS, TRICHOTHECENES,
AND ERGOT

Members

Professor W.W. Carlton, Department of Veterinary Pathobiology,
School of Veterinary Medicine, Purdue University, West Lafayette,
Indiana, USA

Mr T. Demeke, Health Service Department, Ministry of Health, Addis
Ababa, Ethiopia

Dr J. Gilbert, Ministry of Agriculture, Fisheries and Food,
Norwich, United Kingdom

Professor P. Krogh, Department of Microbiology, Royal Dental
College, Copenhagen, Denmark (Co-Rapporteur)

Dr M. Nakadate, Section of Information and Investigation, Division
of Information on Chemical Safety, National Institute of Hygienic
Sciences, Tokyo, Japan

Dr J. Parizek, Institute of Nuclear Biology and Radiochemistry,
Czechoslovak Academy of Sciences, Prague, Czechoslovakia

Dr A.E. Pohland, Division of Chemical Contaminants, Center for Food
Safety and Applied Nutrition, Food and Drug Administration, US
Department of Health and Human Services, Washington DC, USA

Professor H.D.Tandon, ex-President, National Academy of Medical
Sciences, New Delhi, India (Chairman)

Professor Y. Ueno, Department of Toxicology and Microbial
Chemistry, Faculty of Pharmaceutical Sciences, Science University
of Tokyo, Tokyo, Japan (Co-Rapporteur)

Observers

Dr K. Ohtsubo, Department of Clinical Pathology, Tokyo Metropolitan
Institute of Gerontology, Tokyo, Japan

Dr T. Yoshizawa, Department of Bioresource Sciences, Faculty of
Agriculture, Kagawa University, Kagawa, Japan

Secretariat

Dr M. Gilbert, International Programme on Chemical Safety, World
Health Organization, Geneva, Switzerland

Dr A. Prost, Division of Environmental Health, World Health
Organization, Geneva, Switzerland

NOTE TO READERS OF THE CRITERIA DOCUMENTS

Every effort has been made to present information in the
criteria documents as accurately as possible without unduly
delaying their publication. In the interest of all users of the
environmental health criteria documents, readers are kindly
requested to communicate any errors that may have occurred to the
Manager of the International Programme on Chemical Safety, World
Health Organization, Geneva, Switzerland, in order that they may be
included in corrigenda, which will appear in subsequent volumes.

ENVIRONMENTAL HEALTH CRITERIA FOR SELECTED MYCOTOXINS:
OCHRATOXINS, TRICHOTHECENES, AND ERGOT

A WHO Task Group on Environmental Health Criteria for Selected
Mycotoxins met in London on 14-18 November, 1988. Dr Malcolm
HUTTON opened the meeting on behalf of the Director of the
Monitoring and Assessment Research Centre (MARC), King's College,
London, which hosted the meeting on behalf of the three cooperating
organizations of the International Programme on Chemical Safety
(WHO/ILO/UNEP). The Task Group reviewed and revised the draft
criteria document and made an evaluation of the health risks of
exposure to selected mycotoxins.

The draft documents for Ochratoxins and Ergot were prepared by
Professor P. KROGH. Those for Trichothecenes were prepared by Dr
M. NAKADATE AND HIS COLLEAGUES in the National Institute of
Hygienic Sciences and by Professor Y. UENO AND HIS COLLEAGUES in
Tokyo. During the task group meeting, several members of the Group
agreed to undertake a substantial revision of the draft. Dr A.
PROST was responsible for the overall scientific content of the
document and Mrs M. O. HEAD of Oxford, England, for the editing.

The Secretariat wishes to acknowledge the contributions of: Dr
J. GILBERT (Chemistry and analytical methods for trichothecenes);
Dr A.E. POHLAND (Sources and natural occurrence of trichothecenes);
and Professor W.W. CARLTON (Animal studies and metabolism of
trichothecenes).

The Secretariat also wishes to thank Professor Y. UENO, Co-
rapporteur of the Task Group, for his significant contributions and
revisions of the draft document during the meeting. Professor H.
TANDON, Chairman of the Task Group, and Professor P. KROGH and
Professor Y. UENO, Co-rapporteurs, met with members of the
Secretariat in Tokyo, 25-30 July, 1989, to review the final
document before its release.

The efforts of all who helped in the preparation and
finalization of the document, especially those of Dr H. KURATA and
Dr M. ICHINOE (National Institute of Hygienic Sciences, Tokyo) and
Dr K. OHTSUBO (Tokyo Metropolitan Institute of Gerontology), are
gratefully acknowledged.

* * *

Partial financial support for the publication of this criteria
document was kindly provided by the United States Department of
Health and Human Services, through a contract from the National
Institute of Environmental Health Sciences, Research Triangle Park,
North Carolina, USA -- a WHO Collaborating Centre for Environmental
Health Effects.

INTRODUCTION

A decade has passed since the publication of Environmental
Health Criteria 11: Mycotoxins (WHO, 1979), but this field of
research has expanded rapidly, and recent data indicate that the
health effects of several of the mycotoxins dealt with in the above
publication should be updated. More than 200 mycotoxins are now
known to exist. They are present in the environment and, in some
cases, human exposure has been documented, mainly through food
contamination or occasionally through inhalation. However,
information on adverse human health effects is often lacking and,
in many cases, the association between exposure to selected
mycotoxins and the occurrence of health disorders remains
hypothetical.

Ochratoxin A has been found as a contaminant in foods with a
frequency in the range of 2-30% in all countries where attempts to
perform food analysis have been made. Field cases of ochratoxin A-
associated nephropathy in farm animals have been encountered in
many countries, underlining the nephrotoxic potential of this
compound. More recently, ochratoxin A has been detected in the
blood of 6-18% of the human population in some areas where Balkan
endemic nephropathy is prevalent. Ochratoxin A has also been found
in human blood samples outside the Balkan peninsula. In some
studies, more than 50% of the samples analysed have been
contaminated. A high incidence of tumours of the urinary system is
strongly correlated with the prevalence of Balkan endemic
nephropathy. For these reasons, the human health effects of
ochratoxin A have been re-evaluated.

A considerable increase in trichothecenes research has been
seen over the last decade. The potential for the production of a
number of trichothecenes among Fusarium species is well documented,
and the toxicology of a few trichothecenes and the natural
occurrence of some of these compounds in food is fairly well
established. Thus, food-borne exposure of human beings to some
trichothecenes, in particular deoxynivalenol (vomitoxin) and
nivalenol, is likely to occur. Several reports have recently
associated outbreaks of human disease with the presence of
trichothecenes in food. For this reason, the health effects of
trichothecenes have been re-evaluated.

Ergotism is by far the oldest known mycotoxicosis in man and
animals. Recent episodes of Claviceps purpurea-associated
intoxication in Ethiopia, as well as episodes of Claviceps
fusiformis-associated intoxications in areas of India indicate that
ergotism is still a disease of public health importance,
particularly in developing countries. The evaluation of ergot as a
food contaminant has therefore been included in the present updated
environmental health criteria dealing with selected mycotoxins.
However, the review of available documentation has concentrated on
studies dealing with the naturally occurring ergot alkaloids.
Derivatives produced by the pharmaceutical industry have been
deliberately excluded. No attempt has been made to review the
literature on the pharmacology of derivatives of lysergic acid. The

section on ergot in this publication aims at alerting the
scientific community about the present status of ergotism as a
disease of our time, and about the differences in clinical symptoms
that are observed between Asia and Africa in relation to the
chemical differences in responsible toxins.

Studies on the etiology of hepatocellular carcinoma, in
particular those indicating that there is a consistent and specific
causal association between hepatitis B virus and hepatocellular
carcinoma, as well as the existence of other etiological factors
that may cause hepatocellular carcinoma independently, have
attracted much attention since the publication of the previous EHC
on mycotoxins (WHO, 1979). A report by WHO (1983) recognized the
value of available methods for the direct assessment of individual
exposure to aflatoxins, and the methods and their use in field
studies were considered in a later report by IARC (1984). Since the
evaluation of the carcinogenic risks of aflatoxins for human beings
has recently been reviewed by IARC (1987a), the present update on
mycotoxins will not assess the issue and readers should refer to
the IARC evaluation.

SUMMARY AND RECOMMENDATIONS FOR FURTHER RESEARCH

1. Ochratoxin A

1.1 Natural occurrence

Ochratoxins are produced by several species of the fungal
genera Aspergillus and Penicillium. These fungi are ubiquitous and
the potential for the contamination of foodstuffs and animal feed
is widespread. Ochratoxin A, the major compound has been found in
a number of countries in Australasia, Europe, and North America.
Ochratoxin formation by Aspergillus species appears to be limited
to conditions of high humidity and temperature, whereas at least
some Penicillium species may produce ochratoxin at temperatures as
low as 5 °C.

The highest incidences of ochratoxin A contamination have been
found in cereals, and to a lesser extent in some beans (coffee,
soya, cocoa). Ochratoxin B occurs extremely rarely.

1.2 Analytical methods

Analytical techniques have been developed for the
identification and quantitative determination of ochratoxin levels
in the µg/kg range.

1.3 Metabolism

Residues of unchanged ochratoxin A have been found in the
blood, kidney, liver, and muscle of pigs in slaughter houses and in
the muscle of hens and chickens. However, residues of ochratoxin
A have not generally been found in ruminants. The in vitro
binding of ochratoxin A to serum albumin is particularly strong in
cattle, pigs, and man. Experimental studies on pigs and hens have
shown that higher levels of ochratoxin A occur in the kidneys.
Microsomal hydroxylation might represent a detoxification reaction
in pigs, rats, and man. In experimental studies, residues could
still be identified in pig kidneys, one month after the termination
of exposure.

1.4 Effects on animals

Field cases of ochratoxicosis in farm animals (pigs, poultry)
have been reported from several European countries, the primary
manifestation being chronic nephropathy. The lesions include
tubular atrophy, interstitial fibrosis and, at later stages,
hyalinized glomeruli. Ochratoxin A has also been found in pig
blood collected at Canadian slaughterhouses. It has produced
nephrotoxic effects in all species of single-stomach animals
studied so far, even at the lowest level tested (200 µg/kg feed in
rats and pigs).

Teratogenic effects were observed in mice exposed orally to 3
mg/kg body weight. Fetal resorption was observed in rats given
doses from 0.75 mg/kg body weight orally. Teratogenic effects,

which in the rat were enhanced by a diet low in protein, have also
been observed in hamsters.

There is no evidence of ochratoxin A activity in short-term
tests for mutagenicity (bacteria and yeasts). Rats exposed orally
showed single-strand breaks in DNA in renal and hepatic tissues.
Ochratoxin A induced renal cell neoplasms in male mice and in both
sexes of rats dosed orally. Hepatic cell neoplasms were reported
in only one mouse strain and not in the rat.

Ochratoxin A is an inhibitor of protein synthesis and tRNA
synthetase in microorganisms, hepatoma cells, and in renal mRNA in
the rat.

Ochratoxin A can inhibit macrophage migration. In mice, a dose
of 0.005 µg/kg body weight suppressed the immune response to sheep
erythrocytes; however, contradictory results have also been
obtained.

Ochratoxin A has been shown to be carcinogenic to the renal
tubular epithelium in male mice and in both sexes in rats.

1.5 Effects on man

Human exposure, as demonstrated by the occurrence of ochratoxin
A in food, blood, and in human milk, has been observed in various
countries in Europe. Available epidemiological information
indicates that Balkan nephropathy may be associated with the
consumption of foodstuffs contaminated by this toxin.

A highly significant relationship has been observed between
Balkan nephropathy and tumours of the urinary tract, particularly
with tumours of the renal pelvis and ureters. However, no data
have been published that establish a direct causal role of
ochratoxin A in the etiology of such tumours.

2. Trichothecenes

2.1 Natural occurrence

To date, 148 trichothecenes, characterized chemically as having
the same basic tetracyclic scirpenol ring system, are known. These
compounds are produced primarily by moulds belonging to the genus
Fusarium, though other genera, including Trichoderma,
Trichothecium, Myrothecium, and Stachybotrys, are also known to
produce metabolites now characterized as trichothecenes. Only a
few of the known trichothecenes have been found to contaminate food
or animal feed including: deoxynivalenol (DON), nivalenol (NIV),
diacetoxyscirpenol (DAS), and T-2 toxin and, less frequently,
certain derivatives (3-Ac-DON, 15-Ac-DON, fusarenon-X and HT-2
toxin). Of these, by far the most commonly encountered in food and
animal feed is DON, with lesser amounts of NIV usually found as co-
contaminants. Some macrocyclic trichothecenes, such as satratoxins
G and H, and the verrucarins, occur occasionally in animal feed
(straw, hay) but there are no reports of their presence in foods.

Surveys for the presence of trichothecenes have revealed the
world-wide occurrence of DON, primarily in cereals, such as wheat
and corn, at levels occasionally as high as 92 mg/kg, though
average levels are considerably lower and vary with commodity.
There are isolated reports of the occurrence of DON in barley,
mixed feeds, potatoes, etc. NIV, though not normally reported in
cereals in Canada or the USA, is commonly found in conjunction with
DON in Asian and European grains; the highest concentration
recorded to date for NIV is 37.9 mg/kg. T-2 toxin and DAS have
been reported infrequently and at much lower concentrations.

Processing and milling studies have shown little reduction in
DON levels from the cereal to the finished product. Similarly,
baking is not effective in destroying DON. In general,
commercially available human foodstuffs rarely contain detectable
levels of DON and NIV.

2.2 Analytical methods

Analytical methods based on TLC, GC, HPLC, and immunological
techniques are available for the determination of the four most
frequently encountered toxins (DON, T-2 toxin, DAS, NIV) with
detection limits below 1 µg/g. Several of these methods have been
tested collaboratively. In addition, research methods, such as
GC/MS and LC/MS, are available for confirmation of identity.

2.3 Metabolism

Metabolic studies have been carried out on animals, principally
with T-2 toxin, but a few with DON. These trichothecenes are
rapidly absorbed from the alimentary tract, but quantitative data
are not available. The toxins are distributed fairly evenly
without marked accumulation in any specific organ or tissue.
Trichothecenes are metabolically transformed to less toxic
metabolites by such reactions as hydrolysis, hydroxylation,
de-epoxidation, and glucuronidation. Trichothecenes, such as T-2
toxin and DON, are rapidly eliminated in the faeces and urine. For
example, almost 100% of an oral dose of T-2 toxin in cattle was
eliminated within hours of dosing; in chickens, about 80% had been
eliminated 48 h after dosing. In the rat, 25% of DON was
eliminated in the urine and 65% in the faeces, 96 h after dosing.
The results of transmission of T-2 toxin in the laying hen and
lactating cow showed that less than 1% of the administered dose of
this toxin and its metabolites was present in eggs and milk. Tissue
residues of oral T-2 toxin and metabolites in chicken meat were
below 2% of the dose, 24 h after dosing.

2.4 Effects on animals

Ingestion of animal feed of plant origin is the main route of
exposure to trichothecenes. T-2 toxin and DAS, which are the most
potent for laboratory animals of the trichothecenes commonly
reported as feed contaminants (T-2 toxin, DAS, NIV, and DON),
induce a similar toxic response. NIV is less potent in some
systems than the previous two compounds and DON is the least toxic

of the four (examples of potency include the oral LD₅₀s in the
mouse: T-2 toxin, 10.5 mg/kg body weight and DON, 46.0 mg/kg).

The more potent trichothecenes, such as T-2 toxin and DAS,
produce acute systemic effects when administered experimentally to
rodents, pigs, and cattle, via the oral, parenteral, or inhalation
(pig, mouse) route. Epithelionecrosis is a lesion produced by
contact exposure with potent trichothecenes, such as T-2 toxin and
DAS (dose of 0.2 µg per spot for T-2 toxin). Larger doses of other
trichothecenes (NIV, 10 µg per spot) are required to produce an
irritant effect. The cytotoxic trichothecenes, such as T-2 toxin,
produce necrosis of the intestinal crypt epithelium and of lymphoid
and haematopoietic tissues after oral, parenteral, or inhalation
exposure. Haematological and coagulopathic abnormalities follow
exposure to cytotoxic trichothecenes, such as T-2 toxin and DAS.
Severe toxicosis can result in pancytopenia. Suppression of cell-
mediated and humoral immunity has been demonstrated in studies with
T-2 toxin, DON, and DAS, and observations include effects such as
reduced concentrations of immunoglobulins and depressed phagocytic
activity of both macrophages and neutrophils. The results of
experimental animal studies have indicated that the immunodepressive
effect of such trichothecenes as T-2, DAS, and DON, results in
decreased resistance to secondary infection by bacteria
(Mycobacteria, Listeria monocytogenes), yeasts (Cryptococcus
neoformans), and viruses (Herpes simplex virus).

T-2 toxin has been reported to be teratogenic in the mouse,
when given by intraperitoneal injection (unusual route of
administration for teratogenic studies). DON was reported to be
teratogenic in mice after gastric intubation, but was not
teratogenic in rats when the toxin was provided in the feed. NIV
was not teratogenic in mice. T-2 toxin, DAS, and DON were not
mutagenic in an Ames-type assay. T-2 toxin had weak clastogenic
activity in some assays. There is no evidence from the published
long-term toxicity studies in animals to indicate that T-2 toxin,
fusarenon-X, and NIV are tumorigenic in animals. No long-term
studies of DON toxicity have been published.

Trichothecenes are toxic for actively dividing cells, such as
the intestinal crypt epithelium and the haematopoietic cells. The
cytotoxicity has been associated with either impairment of protein
synthesis by the binding of the compounds to the ribosomes of
eukaryotic cells, or the dysfunction of cellular membranes.
Inhibition of protein synthesis has been associated with the
induction of labile and regulatory proteins, such as IL-2 in
immunocytes. Transport of small molecules is impaired in cell
membranes by extremely low concentrations of trichothecenes.

2.5 Effects on man

Ingestion of contaminated foods of plant origin is the main
route of exposure to trichothecenes, but other routes have been
reported occasionally, such as accidental skin contact amongst
laboratory research workers, and airborne trichothecenes in dust.

Reported cases of illness associated with exposure to
trichothecenes are scarce and none has been established as being
due to trichothecenes. However, a causative role is suggested by
the two outbreaks referred to below.

One disease outbreak was reported from China and was associated
with the consumption of scabby wheat containing 1.0-40.0 mg DON/kg.
The disease was characterized by gastrointestinal symptoms. No
deaths occurred in human beings. Swine and chicks fed the leftover
cereals were also affected.

An analogous outbreak was reported from India and was
associated with consumption of baked bread made from contaminated
wheat. The disease was characterized by gastrointestinal symptoms
and throat irritation, which developed within 15 minutes to one
hour following ingestion of the bread. The following mycotoxins
were detected in samples of refined wheat flour used in the
preparation of the bread: DON (0.35-8.3 mg/kg), acetyldeoxynivalenol
(0.64-2.49 mg/kg), NIV (0.03-0.1 mg/kg) and T-2 toxin (0.5-0.8
mg/kg). However, there was no confirmation of the identity of the
detected trichothecenes. The concomitant occurrence of DON and NIV
with T-2 toxin is unusual.

Two diseases of historical interest, alimentary toxic aleukia
(ATA) in the USSR and scabby wheat toxicosis in Japan and Korea,
have been associated with the consumption of grain invaded by
Fusarium moulds. Some trichothecenes have since been identified
under laboratory conditions in fungal cultures of Fusarium moulds
isolated from grains involved in the incidents. Studies linking
ATA and scabby grain toxicosis to trichothecenes exposure could not
be made at the time that the disease occurred, because the toxins
were not known.

3. Ergot

3.1 Natural occurrence

Ergot is the name given to sclerotia of fungal species within
the genus Claviceps. Biologically active alkaloids contained in
the sclerotia cause the development of toxicoses when the sclerotia
are consumed by man and animals through contaminated food or animal
feed.

Ergot alkaloids produce two different patterns of diseases,
depending on the fungal organism involved (C. Purpurea, C.
fusiformis) and hence the alkaloids produced. Ergotism, induced by
ergotamine-ergocristine alkaloids produced by C. purpurea, is
characterized predominantly by gangrene of the extremities as well
as gastrointestinal symptoms. Intoxication induced by millet
contaminated with C. fusiformis is mainly characterized by
gastrointestinal symptoms, and is related to clavine alkaloids.
There are no signs or symptoms suggesting vaso-occlusion.

3.2 Analytical methods

Ergot alkaloids (ergolines) are derivatives of lysergic acid.
The individual alkaloids vary in the magnitude of their biological
activity. Determination of C. purpurea ergot alkaloids has been
carried out by HPLC with fluorescence detection. Concentrations of
0.2 µg ergoline per litre of human plasma can be measured.
Ergotamine and ergocristine can be determined very specifically by
radioimmunoassays at levels of 3.5 picomoles and 0.8 picomoles,
respectively.

3.3 Effects on animals

Ergolines, mainly ergotamine and ergotaminine, have been
associated with outbreaks of bovine abortion. Sheep, administered
ergotamine orally, rapidly became ill and intestinal inflammation
was observed. Orally exposed poultry, pigs, and primates
experienced slight effects. No data on the mutagenicity,
teratogenicity, and carcinogenicity of ergolines were available to
the task group.

3.4 Effects on man

Claviceps-infected grain is a source of human exposure to
ergolines. In most toxicological studies, identification of
specific alkaloids has not been undertaken. The published
information from only one survey of cereals and cereal products
indicates a total daily human intake of ergolines in Switzerland of
approximately 5.1 µg per person, the contents of certain
commodities being up to 140 µg/kg. Baking reduces the ergolines
present in contaminated flour by 25-100%.

An outbreak of ergotism in Ethiopia in 1978 resulted from
exposure to ergolines from C. purpurea sclerotia. The grain
contained up to 0.75% ergot; ergometrine was detected specifically.
Symptoms included dry gangrene with loss of one or more limb (29%
of cases), feeble or absent peripheral pulses (36%), and
desquamation of the skin. Gastrointestinal symptoms occurred in
only a few cases. Lower extremities were involved in 88% of
patients.

In India, several outbreaks have occurred since 1958 as a
result of ingesting pearl millet containing clavine-type ergot from
C. fusiformis. Symptoms included nausea, vomiting, and giddiness.
Pearl millet containing 15-26 mg ergoline/kg caused the toxic
symptoms.

Since autopsies were not performed in either of the episodes,
no information is on record of the pathological effects on human
viscera.

4. Recommendations for further research

4.1 General recommendations

A network of reference centres should be established to assist
Member States in confirming the identity of individual mycotoxins
found in human foods and tissues. These reference centres should
also provide mycotoxin reference samples, upon request, to
reinforce the intercomparability of analytical results obtained in
different parts of the world.

4.2 Ochratoxin A

(a) Extended retrospective, as well as focal prospective,
epidemiological studies on the association of ochratoxin A
with Balkan type endemic nephropathy and with urinary-tract
tumours should be conducted in the Balkan peninsula and the
Mediterranean region.

(b) Blood analysis for ochratoxin A should be performed on
patients with urinary-tract tumours, outside the Balkan
peninsula.

(c) The source of ochratoxin A exposure, as indicated by human
blood analysis, should be elucidated in countries outside the
Balkan peninsula.

(d) The mechanism of the sex differences in renal, neoplastic, and
non-neoplastic disease, caused by ochratoxin A in experimental
animals, should be elucidated.

(e) Extended surveys on the ochratoxin A contents of foods in
different parts of the world are required. Such surveys are
particularly important in regions of the world where high
incidence rates of urinary-tract tumours, renal tumours, or
nephropathy occur.

4.3 Trichothecenes

(a) Follow-up studies should be performed in the areas of India
and the People's Republic of China in which episodes of
trichothecenes intoxication in human beings have been
encountered recently. The unusual pattern of trichothecene
occurrence in these episodes should be further elucidated.

(b) The effects of long-term exposure of experimental animals to
DON, including the carcinogenic effects, should be studied.
Because the response to DON by different species varies
greatly, the test species must be chosen carefully.

(c) Secondary microbial infection in experimental animals
following trichothecenes exposure should be further
elucidated.

(d) The influence of environmental conditions, including the
presence of insecticides and other man-made chemicals, on the
fungal production of trichothecenes should be studied.

(e) The effects of food processing on trichothecenes should be
clarified.

(f) Agricultural plants, resistant to infection by trichothecene-
producing fungi, should be developed, using biotechnological
approaches.

(g) The possible synergistic effects in experimental animals of
combined exposure to trichothecenes, aflatoxins, ochratoxin A,
and other mycotoxins should be studied.

(h) Studies on the intake of trichothecenes by human beings should
be performed.

(i) Rapid and sensitive screening procedures for trichothecenes
should be developed, and surveys for trichothecenes in grain
and processed foods in temperate zones of the world should be
conducted.

4.4 Ergot

(a) Methods of analysis for agroclavines should be developed.

(b) Information should be made available to developing countries
on the use of pathological seed screening and milling
procedures for minimizing the problems of ergot.

(c) Epidemiological studies should be performed on the possible
effects of low levels of ergolines on the human population.

(d) Pharmacological and toxicological studies should be performed
using individual and combined ergolines on experimental
animals.

(e) The possible transmission of ergolines through the mother's
milk to the infant should be elucidated.

I. OCHRATOXINS

I.1 PROPERTIES AND ANALYTICAL METHODS

I.1.1 Chemical properties

The ochratoxins constitute a group of closely-related
derivatives of isocoumarin linked to L-phenylalanine (Fig. 1), and
classified according to biosynthetic origin as pentaketides within
the group of polyketides (Turner, 1971). The topic has been
reviewed by Scott (1977) and Steyn (1977, 1984).

The first compound discovered, ochratoxin A, was isolated from
a culture of Aspergillus ochraceus, hence the name (Van der Merwe
et al., 1965). The acids, including 4-hydroxy-ochratoxin A, the
methyl and ethyl esters, and the isocoumarin part of ochratoxin B
(ochratoxin) have all been isolated from fungal cultures under
experimental conditions. On acid hydrolysis, ochratoxin A yields
phenylalanine, and the isocoumarin part, ochratoxin alpha, a
cleavage product also found in the intestines, faeces, urine, and
liver of rodents experimentally fed an ochratoxin A-containing diet
(Galtier & Alvinerie, 1976). Ochratoxin A and, very rarely,
ochratoxin B are the only compounds found as natural contaminants
in plant material, and most of the information available concerns
ochratoxin A. It can be stored in ethanol in the refrigerator for
more than a year without loss (Chu & Butz, 1970); however, such
solutions should be protected from light, since decomposition
occurs on exposure to fluorescent light for several days (Neely &
West, 1972).

Ochratoxin A is a colourless, crystalline compound, obtained by
crystallization from benzene, with a melting point of about 90 °C,
and containing approximately one mole of benzene. After drying for
1 h at 60 °C, it has a melting point in the range of 168-173 °C.
It is soluble in polar organic solvents, slightly soluble in water,
and soluble in dilute aqueous bicarbonate. Physical data on
ochratoxin A, based on a collaborative study, have been published
by IUPAC (Pohland et al., 1982). Ochratoxin A is optically active:
[alpha]²¹_D: -46.8 °C (c = 2650 µmol/litre in chloroform). The

publication cited above includes information on the following
spectra of ochratoxin A: ultraviolet absorption spectrum; infrared
absorption spectrum; electron impact mass spectrum; nuclear
magnetic resonance spectrum.

I.1.2 Methods for the analysis of foodstuffs and biological
samples

Ochratoxin A in acidified commodities is readily soluble in
many organic solvents, and this characteristic has been used as the
principle of extraction in several methods.

A number of methods using thin-layer chromatography have been
published, one of the most commonly used being the official AOAC
method developed for barley (Nesheim et al., 1973). In this
method, ochratoxins A and B are extracted from ground samples with
chloroform, after acidification. The toxins are trapped in a
column containing diatomaceous earth impregnated with a basic
aqueous solution. After column clean-up, the toxins are eluted and
thin-layer chromatography is performed using long-wave UV
irradiation for visualization of the fluorescent ochratoxin spots
(limit of detection: 12 µg/kg). The method has been
collaboratively studied, revealing coefficients of variation
(between laboratories) in the range of 31-54% (Nesheim, 1973).
This method has been published as an IUPAC recommended procedure
(IUPAC, 1976). The sensitivity can be improved by exposing the
developed plate to ammonia fumes, resulting in a limit of detection
of a few µg/kg. A slightly modified method, developed for green
coffee (Levi, 1975), has a coefficient of variation in the range of
33-49%, based on a collaborative study.

The two methods have been combined into one procedure,
published as an IARC procedure (Nesheim, 1982). In an
international check sample survey on ochratoxin A in animal feed,
the AOAC procedure was used by 61% of the participants, and the
performance was slightly better than that of the other methods,
with a coefficient of variation of 69% compared to 79% for the
other methods combined (Freisen & Garren, 1983). Paulsch et al.
(1982) developed a procedure for the determination of ochratoxin A
in the kidneys of swine, using a liquid-liquid partitioning step
instead of column clean-up, followed by two-dimensional thin-layer
chromatography using an acidic and an alkaline developing solvent.
In addition, the procedure included a confirmatory test, based on
the formation of ochratoxin A methyl ester on the plate.

High-performance liquid chromatographic procedures have been
developed for the determination of ochratoxin A in food of plant
origin (Hunt et al., 1978; Josefsson & Moller, 1979), with limits
of detection in the range of 1-12 µg/kg, and with recoveries in the
55-92% range. A procedure for the determination of ochratoxin A
residues in renal tissue has been published, in which enzymic
digestion of the sample tissue is followed by dialysis and high
performance liquid chromatography (Hunt et al., 1979). The limit
of detection is about 1 µg/kg, and recoveries of 77-78% have been
observed. Rapid screening methods for ochratoxin A are available,

based on minicolumn chromatography, with limits of detection in
the 8-12 µg/kg range (Hald & Krogh, 1975; Holaday, 1976). By using
antisera to ochratoxin A, an enzyme-linked immunosorbent assay
(ELISA) has been constructed in which the toxin in barley can be
determined with only 0.5% cross-reaction for ochratoxin B, and with
a lower level of detection of 10 pg ochratoxin A/well of the ml
plate (Morgan et al., 1982). Immunological methodology has been
further improved by the use of monoclonal antibody (IgG) in a
radioimmunoassay with high specificity for ochratoxin A, which can
detect levels of this toxin as low as 0.2 µg/kg in swine kidney
tissue (Candlish et al., 1986; Rousseau et al., 1987). A method
has been described in which ochratoxin A is cleaved to ochratoxin
alpha and phenylalanine using the enzyme carboxypeptidase. The
quantification of ochratoxin A is based on the loss of fluorescence
intensity at 380 nm, the limit of detection being 4 µg/kg of
barley (Hult & Gatenbeck, 1976). The procedure has also been
applied to swine blood with a limit of detection of 2 µg/litre
(Hult et al., 1979).

The same procedure has been used in screening human blood for
the presence of ochratoxin A, with a limit of detection of 1-2
µg/litre serum for a 2-g sample. High-performance liquid
chromatography was used as confirmation (Hult et al., 1982). A
screening method involving flow injection has been developed by
which ochratoxin A concentrations of more than 10 µg/litre can be
determined, based on 50-µlitre samples of serum (Hult et al.,
1984). As the specificity of the method is limited, positive
results have to be confirmed by conventional methods requiring much
larger blood samples.

A method has been developed for the detection of ochratoxin
A (and aflatoxin B and citrinin) in human urine using hydrolysis
of urine, solid-phase extraction, and reversed-phase liquid
chromatography with fluorescence detection (Orti et al., 1986).
The detection limit for ochratoxin A was approximately 10 ng/ml in
samples of 10 ml urine.

I.2 SOURCES AND OCCURRENCE

I.2.1 Fungal formation

The ochratoxins were isolated in 1965 from a culture of
Aspergillus ochraceus, hence the name (Van der Merwe et al., 1965),
but subsequent investigations have revealed that a variety of
fungal organisms included in the genera Aspergillus and Penicillium
are able to produce ochratoxins (Table 1).

The effects of water activity (a_w) and temperature, the main
factors controlling mycotoxin formation, have been elucidated in
relation to growth and ochratoxin production for 3 fungal
organisms: A. ochraceus, P. cyclopium, and P. viridicatum
(Northolt et al., 1979). The minimum a_w values for ochratoxin
production ranged between 0.83 and 0.87, 0.87 and 0.90, and 0.83
and 0.86, respectively. At 24 °C, optimum a_w values for A.
ochraceus and for both P. cyclopium and P. viridicatum were 0.99
and 0.95-0.99, respectively (Fig. 2).

Table 1. Ochratoxin-producing fungi^a
-------------------------------------------------------------------
Penicillium link

Monoverticillata:
P. frequentans series: P. purpurrescens Sopp
Asymmetrica lanata:
P. commune series: P. commune Thom
Asymmetrica fasciculata:
P. viridicatum series: P. viridicatum Westling
P. palitans Westling
P. cyclopium series: P. cyclopium Westling
Biverticillata symmetrica:
P. purpurogenum series: P. varibile Sopp

Aspergillus Micheli

Aspergillus ochraceus group: A. sulphureus (Fres.)
Thom & Church
A. sclerotiorum Huber
A. alliaceus Thom & Church
A. melleus Yukawa
A. ochraceus Wilhelm
A. ostianus Wehmer
A. petrakii Vörös
-------------------------------------------------------------------
^a From: Krogh (1978).

At optimum a_w, the temperature range for ochratoxin production
by A. ochraceus was 12-37 °C, whereas that of P. cyclopium and P.
viridicatum was 4-31 °C. These laboratory data correspond with
those from field observations on ochratoxin contamination in crops.
Thus, the more frigophilic Penicillia, particularly P. viridicatum,
are the major ochratoxin producers in crops in colder climatic
zones, such as Scandinavia (Krogh, 1978; Rutqvist et al., 1978;
Häggblom, 1982), and in Canada (Scott et al., 1972).

In contrast, ochratoxin-producing fungal potential has been
found in 28-50% of A. ochraceus strains isolated from crops in
warmer climatic zones, such as Australia (Connole et al., 1981),
and Yugoslavia (Pepeljnjak & Cvetnic, 1981), and strains isolated
from coffee beans (Stack et al., 1982).

I.2.2 Occurrence in foodstuffs

I.2.2.1 Plant products

Ochratoxin A was first encountered as a natural contaminant in
maize (Shotwell et al., 1969), and subsequent surveys have
established that ochratoxin A is a contaminant of cereals and some
beans (coffee beans, soya beans, cocoa beans) in many areas of the
world (Table 2). Although the mean level of ochratoxin A in all
reported surveys up to 1979 was 1035 µg/kg (Fig. 3), 83% of the
samples contained less than 200 µg/kg (Krogh, 1980).

Table 2. Natural occurrence of ochratoxin A in foodstuffs and animal feed of plant origin
---------------------------------------------------------------------------------------------------------
Commodity Country Number of Percentage Range of Reference
samples contaminated ochratoxin A
analysed levels (µg/kg)
---------------------------------------------------------------------------------------------------------
FOOD
maize USA 293 1.0 83-166 Shottwell et al. (1971)
maize (1973) France 463 2.6 15-200 Galtier et al. (1977b)
maize (1974) France 461 1.3 20-200 Galtier et al. (1977b)
wheat (red winter) USA 291 1.0 5-115 Shottwell et al. (1976)
wheat (red spring) USA 286 2.8 5-115 Shottwell et al. (1976)
barley (malt) Denmark 50 6.0 9-189 Krogh (1978)
barley USA 127 14.2 10-40 Nesheim (1971)
coffee beans USA 267 7.1 20-360 Levi et al. (1974)
maize Yugoslavia^a 542 8.3^b 6-140 Pavlovic et al. (1979)
wheat Yugoslavia^a 130 8.5^b 14-135 Pavlovic et al. (1979)
wheat bread Yugoslavia^a 32 18.8^b Pavlovic et al. (1979)
barley Yugoslavia^a 64 12.5^b 14-27 Pavlovic et al. (1979)
barley Czechoslovakia 48 2.1 3800 Vesela et al. (1978)
bread United Kingdom^c 50 2.0 210 Osborne (1980)
flour United Kingdom^c 7 28.5 490-2900 Osborne (1980)
beans Sweden 71 8.5 10-442 Akerstrand &
peas Sweden 72 2.8 10 Josefsson (1979)
maize United Kingdom 29 37.9 50-500 Ministry of
cornflour United Kingdom 13 30.8 50-200 Agriculture (1980)
soya bean United Kingdom 25 36.0 50-500 Ministry of
soya flour United Kingdom 21 19.0 50-500 Agriculture (1980)
cocoa beans (raw) United Kingdom 56 17.9 100-500 Ministry of
(roasted) United Kingdom 19 15.8 100 Agriculture (1980)
grain German Democratic 49 4.1 18-22 Fritz et al.
Republic (1979)
grain (barley, Poland 296 6.8 20-470 Szebiotko et al. (1981)
wheat, rye)
grain (wheat, rye) Denmark 151 1.3 15-50 Pedersen &
bran (wheat) Denmark 57 10.5 5-20 Hansen (1981)
maize Bulgaria^a 22 27.3 25-35 Petkova-Bocharova &
maize Bulgaria 22 9.0 10-25 Castegnaro (1985)
beans Bulgaria^a 24 16.7 25-27 Petkova-Bocharova &
beans Bulgaria^a 28 7.1 25-50 Castegnaro (1985)
---------------------------------------------------------------------------------------------------------

Table 2 (contd.)
---------------------------------------------------------------------------------------------------------
Commodity Country Number of Percentage Range of Reference
samples contaminated ochratoxin A
analysed levels (µg/kg)
---------------------------------------------------------------------------------------------------------
FOOD (continued)
barley, wheat, Poland 150 5.3 50-200 Juszkiewicz &
oats, rye, maize Piskorska-
Pliszczynska (1976)
FEED
mixed feed Poland 203 4.9 10-50 Juszkiewicz &
Piskorska-
Pliszczynska (1977)
maize Yugoslavia 191 25.7 45-5125 Balzer et al. (1977)
barley, oats Sweden 84 8.3 16-409 Krogh et al. (1974b)
wheat, hay Canada 95 7.4 30-6000 Prior (1976)
wheat, oats, Canada^c 32 56.3 30-27 000 Scott et al. (1972)
barley, rye
barley, oats Denmark^c 33 57.6 28-27 500 Krogh et al. (1973)
mixed feed Canada 474 1.1 30-100 Prior (1981)
mixed feed Canada^c 51 7.8 48-5900 Abramson et al. (1983)
mixed feed Australia 25 4.0 70 000 Connole et al. (1981)
mouldy bread Italy^d 1 80 000 Visconti &
Bottalico (1983)
---------------------------------------------------------------------------------------------------------
^a From an area with endemic nephropathy.
^b Average values for a period of 2-5 years.
^c All samples suspected of containing mycotoxins.
^d This sample contained in addition 9600 µg ochratoxin B/kg.
Ochratoxin B has only been found in 3 samples and, thus, occurs
extremely rarely; the other ochratoxins have never been found in
plant products. Levels of ochratoxin A and the frequency of
contamination are generally higher in animal feed than in foodstuffs
(Table 2). Although the mean level of ochratoxin A in all reported
surveys up to 1979 was 1035 µg/kg (Fig. 3), 83% of the samples
contained less than 200 µg/kg (Krogh, 1980).

I.2.2.2 Residues in food of animal origin

Residues of ochratoxin A are not generally found in ruminants,
because ochratoxin A is cleaved in the forestomachs by protozoan and
bacterial enzymes (Galtier & Alvinerie, 1976; Hult et al., 1976;
Patterson et al., 1981). The non-toxic cleavage product, ochratoxin
alpha, has been found in the kidneys at levels below 10 µg/kg and in
the blood of calves fed a diet containing 300-500 µg ochratoxin A/kg
(Patterson et al., 1981). In half of the calves, the kidneys
contained low levels of ochratoxin A (up to 5 µg/kg), a feature that
may reflect that the calves were not yet functioning as ruminants.
When 2 milking cows were fed a ration containing 317-1125 µg
ochratoxin A/kg for 11 weeks, a residue of 5 µg ochratoxin A/kg was
found in the kidneys of one of the animals but not in any other
tissue or the milk. Ochratoxin alpha was not found in any tissue
(Shreeve et al., 1979). Residues of ochratoxin A have been detected
in a number of tissues in single-stomach food animals, such as pigs.
The carry-over of ochratoxin A from feed to animal tissues was
elucidated in a study in which groups of pigs were exposed for 3-4
months to dietary levels of ochratoxin A of 200, 1000, or 4000 µg/kg
(Krogh et al., 1974a). At termination (slaughter), the highest
levels of ochratoxin A residues were found in the kidneys (mean
levels 50 µg/kg at the 4000 µg/kg feed level), with lower levels
in the liver, muscle, and adipose tissue; other tissues, including
blood, were not analysed. There was a high correlation between the
feed level of ochratoxin A and the residue levels in the 4 tissues
investigated (Table 3).

Table 3. Correlation between feed level and tissue levels
(residues) of ochratoxin A in pigs^a
-----------------------------------------------------------
Tissue Regression equation ^r
-----------------------------------------------------------
kidney Y = 2.15 + 0.0123X 0.86
liver Y = 0.35 + 0.0095X 0.82
adipose Y = 2.51 + 0.0099X 0.78
-----------------------------------------------------------
^a Modified from Krogh et al. (1974a).
X = ochratoxin A in feed (µg/kg).
Y = ochratoxin A residue (µg/kg tissue).
r = correlation coefficient.
The regression is calculated on feed levels of ochratoxin A
in the range of 200-4000 µg/kg.

Ochratoxin A is present in the blood bound to serum-albumin
(Chu, 1971; Galtier, 1974a) and as free ochratoxin A; saturation (in
the rat) occurs at 70 mg ochratoxin A/litre plasma. The binding of
ochratoxin A to serum-albumin is particularly strong in cattle,
pigs, and man, based on in vitro studies (Table 4).

Table 4. In vitro binding of ochratoxin A to
serum-albumin in several species^a
-----------------------------------------------
Animal Number of Intrinsic
species binding Association
sites constant (M-1)
-----------------------------------------------
Cattle 0.58 94 600
Pig 0.56 71 100
Man 0.58 63 500
Horse 0.57 57 400
Chicken 0.51 52 700
Rat 0.68 40 100
Sheep 0.88 22 600
-----------------------------------------------
^a Adapted from: Galtier (1979).

The presence of ochratoxin A in the blood of pigs has been
elucidated in experimental animal studies as well as by surveys of
blood samples from pigs at farms (Hult et al., 1979, 1980). A
total of 1200 pig blood samples, collected from slaughterhouses
over various periods during 1986 in western Canada, was screened
for the presence of ochratoxin A using an HPLC procedure (Marqhardt
et al., 1988). It was shown that 3.6-4.2% of the blood samples
contained the toxin at concentrations higher than 20 ng/ml. It
appears that the concentration of ochratoxin A in the blood is
higher than that in any other tissue. In feeding studies, bacon
pigs were exposed for various periods to rations containing
ochratoxin A concentrations in the range of 58-1878 µg/kg; blood,
kidney, liver, muscle, and adipose tissues were analysed, at
slaughter, for residues of ochratoxin A (Mortensen et al., 1983).
The statistical association between residue levels in various
tissues is indicated by the regression analyses (Table 5).

Table 5. Regression between residues of ochratoxin A
in the serum and certain other tissues^a
------------------------------------------------------
Tissue Regression r s_b
equation
------------------------------------------------------
kidney Y = 0.0651X 0.89 0.002
muscle Y = 0.0346X 0.88 0.001
liver Y = 0.0259X 0.89 0.001
adipose Y = 0.0191X 0.84 0.001
------------------------------------------------------
^a Adapted from: Mortensen et al. (1983).
X = µg ochratoxin A/litre serum.
Y = µg ochratoxin A/kg in the other tissues.
r = correlation coefficient
s_b = Standard error of slope

Epidemiological studies on the basis of data from meat
inspection in Danish slaughterhouses revealed prevalence rates of
porcine nephropathy ranging from 10 to 80 cases per 100 000
slaughtered pigs (Krogh, 1976a). Surveys in a number of European

countries for residues of ochratoxin A in kidneys from cases of
porcine nephropathy revealed that 25-39% of the cases contained
ochratoxin A levels in the range 2-100 µg/kg (Table 6).

Table 6. Surveys for ochratoxin A residues in kidneys from
cases of porcine nephropathy, based on meat inspection data
-------------------------------------------------------------------
Country No. of porcine Percentage Range of Reference
nephropathy containing ochratoxin
kidneys residues of A residues
investigated ochratoxin (µg/kg wet
A weight)
-------------------------------------------------------------------
Denmark 60 35 2-68 Krogh (1977)

Germany, 104 21 0.1-1.8 Bauer et al.
Federal (1984)
Republic of

Hungary 122 39 2-100 Sandor et
al. (1982)
Poland 113 24 1-23 Golinski et
al. (1984)
Sweden 129 25 2-104 Rutqvist et
al. (1977)
Sweden 90 27 2-88 Josefsson
(1979)
-------------------------------------------------------------------

Ochratoxin A levels of up to 29 µg/kg were found in the muscle
of hens and chickens collected at one slaughterhouse (Elling et
al., 1975). The birds had been condemned because of nephropathy.
In another study, groups of hens were exposed for 1-2 years to
dietary levels of ochratoxin A of 0.3 or 1 mg/kg (Krogh et al.,
1976b). The kidneys contained the highest residues with a mean
value of 19 µg/kg tissue in the group fed 1 mg ochratoxin A/kg; the
liver and muscle contained lower levels of residues and no
ochratoxin A was found in the eggs. These results are in
accordance with those of subsequent experimental animal studies.
For example, in a study in which 4 groups of hens were fed diets
containing 0, 0.5, 1, or 4 mg ochratoxin A/kg, the kidneys
contained the highest level of ochratoxin A residues (31 µg/kg) in
the highest dose group; eggs were not analysed (Prior & Sisodia,
1978).

In another study, hens were fed 1 mg ochratoxin A/kg feed for 8
weeks; the kidneys contained 3-10 µg ochratoxin A/kg and the liver,
1.5-2.5 µg/kg. The eggs were not analysed (Reichmann et al.,
1982). When groups of hens were fed diets containing 2.5 or 10
mg/kg, the kidneys contained 1-6 µg ochratoxin A/kg, lower levels
being found in the plasma, muscle, and liver (Juszkiewicz et al.,
1982). Eggs from the high-dose group contained 0.7-1.3 µg
ochratoxin A/kg, but none was detected in eggs from the low-dose
group.

White Leghorn hens (54 birds), divided into four groups, were
fed diets for one month containing the following concentrations of
ochratoxin A (mg/kg): 0, 1.3, 2.6, and 5.2 (Bauer et al., 1988).
At termination, three tissues were analysed for residues. The
following ranges of dose-dependent residues were found: serum,
4.7-11.7 ng/ml, liver, 9.1-18.0 ng/g, and yolk, 1.6-4.0 ng/g; very
little ochratoxin A was found in the egg white, and none was
detected in the tissues of the control group.

I.3 METABOLISM

I.3.1 Absorption

In a study on rats exposed by gavage to a single dose of
ochratoxin A at 10 mg/kg body weight, Galtier (1974b) found the
highest tissue level of unchanged ochratoxin A in the stomach wall
during the first 4 h following administration. The small and large
intestine and caecum contained small amounts of unchanged
ochratoxin A, and it was concluded that ochratoxin A was absorbed
mainly in the stomach. Small amounts (1-3% of the total dose) were
detected in the caecum and the large intestine, as the isocoumarin
moiety (ochratoxin alpha), most likely as the result of the
hydrolysing action of the intestinal microflora (Galtier &
Alvinerie, 1976; Hult et al., 1976).

In a study on intestinal absorption using the same animal
species, Kumagai & Aibara (1982) came to the conclusion that the
site of maximal absorption of ochratoxin A was the proximal
jejunum, and that the portal vein was the primary route of
transport from the intestinal tract, though part of the
transportation took place through the lymphatics.

Using a highly specific antibody against ochratoxin A, which is
used in a peroxidase-antiperoxidase detection method, Lee et al.
(1984) studied absorption and tissue distribution in Swiss mice
over a 48-h period, after the administration of a single dose of 25
mg ochratoxin A/kg body weight. Ochratoxin A was found in large
amounts, indicated by staining, in epithelial cells as well as in
macrophages of the lamina propria in the duodenum. Smaller amounts
were found in jejunal epithelial cells and lamina propria, and much
smaller amounts in the epithelial cells in the esophagus and
stomach; no toxin was found in the ileum. These findings suggest
that absorption mainly takes place in the duodenum and the jejunum.

I.3.2 Tissue distribution

I.3.2.1 Animal studies

In slaughterhouse cases of mycotoxic porcine nephropathy
studied by Hald & Krogh (1972), residues of unchanged ochratoxin A
were found in all tissues investigated (kidney, liver, and muscle),
the highest level (up to 67 µg/kg) occurring in the kidney. In
experimental studies on pigs ingesting feed containing ochratoxin
A, residues of the toxin were found in all 4 tissues in the

decreasing order of kidney, liver, muscle, adipose tissue (Krogh et
al., 1974a). A subsequent study revealed that the concentration of
ochratoxin A residues in the blood of the pig was higher than those
in the other tissues mentioned above (Mortensen et al., 1983).
When rats were exposed orally to an ochratoxin A dose of 10 mg/kg
body weight, Galtier (1974b) recovered 0.3% of the administered
dose in the whole kidneys, 0.9% in the whole liver, and 0.6% in the
total muscle tissue, 96 h after exposure. Chang & Chu (1977),
using a single intraperitoneal injection of 1 mg ochratoxin A per
rat (labelled with ¹⁴C in phenylalanine), found that the kidney
contained twice as much unchanged ochratoxin A as the liver after
30 minutes, amounting to 4-5% of the total dose.

In the study by Lee et al. (1984), the largest amounts of
ochratoxin A found in the kidney (as indicated by staining
intensity) were in the epithelium of the proximal convoluted
tubules, and to a lesser extent in the distal convoluted tubules,
the descending loop of Henle, and glomeruli and Bowman's capsule.
Small amounts were found in hepatocytes as well as in the lumina of
bile ducts, but not in biliary epithelium, indicating biliary
excretion.

Using pig renal cortical slices, it was found that ochratoxin A
enters the proximal tubule cells by the common organic anion
transport system (Friis et al., 1988). Ochratoxin A inhibited p-
aminohippurate (PAH) and phenolsulfophthalein uptake in a dose-
dependent manner.

I.3.2.2 Studies on man

In a study in Yugoslavia near Slavonski Brod, where endemic
nephropathy is prevalent, 639 samples of serum from the inhabitants
of 2 villages were screened for the presence of ochratoxin A; 42
(6.6%) were positive for the toxin "with concentrations in the
range of 1-57 ng ochratoxin A/g" (Hult et al., 1982). Detection
was carried out using the enzymic spectrofluorometric procedure,
and positives were confirmed by the esterification of ochratoxin A
in serum and of ochratoxin alpha obtained from the enzymetic
hydrolisates; the esters were measured by high performance liquid
chromatography.

In a screening of serum samples in Poland using the same method
of analysis, 77 out of 1065 samples (7.2%) contained ochratoxin A,
with a mean concentration of 0.27 ng/ml and a maximum value of 40
ng/ml (Colinski, 1987). In the Federal Republic of Germany, 173
out of 306 serum samples (56.6%) contained ochratoxin A, as
measured by an HPLC procedure, with a mean of 0.6 ng/g and a range
of 0.1-14.4 ng/g (Bauer & Gareis, 1987). Three out of 46 kidneys
(6.6%) contained ochratoxin A, with a mean of 0.2 ng/g and a range
of 0.1-0.3 ng/g. In a study of blood plasma in Denmark, 46 out of
96 samples (47.9%) contained ochratoxin A, as measured by an HPLC
procedure, with a mean of 1.7 ng/g and a range of 0.1-9.2 ng/g
(Hald, 1989). In Bulgaria, where endemic nephropathy also occurs
in some areas, 45 out of 312 blood samples contained ochratoxin A
(14.4%), with a mean of approximately 14 ng/g (Petkova-Bocharova et
al., 1988).

I.3.3 Metabolic transformation

It has been shown from in vitro studies that ochratoxin A binds
to serum-albumin (Chu, 1971, 1974b); this binding has also been
observed in in vivo studies on rats (Galtier, 1974a; Chang & Chu,
1977). Ochratoxin alpha has been detected in the urine and faeces
of rats injected intraperitoneally with ochratoxin A (Nel &
Purchase, 1968; Chang & Chu, 1977), indicating the cleavage of
ochratoxin A to ochratoxin alpha and phenylalanine, under these
conditions.

Studies with ¹⁴C-labelled ochratoxin A indicated that some
other, not yet identified, metabolites are formed in the body.
Less than half of the radioactivity excreted in rat urine within 24
h of a single intraperitoneal injection of ¹⁴C-phenylalanine-
labelled ochratoxin A was identified as ochratoxin A (Chang & Chu,
1977).

In both albino rats and brown rats given ochratoxin A orally or
intraperitoneally, 1-1.5% of the dose was excreted as (4R)-4-
hydroxyochratoxin A and 25-27% as ochratoxin alpha in the urine
(Storen et al., 1982). In in vitro studies using liver microsomes
from the pig, rat, and man, both (4R)-and (4S)-4-hydroxy-ochratoxin
A were produced in a hydroxylation process involving cytochrome
P-450 (Stormer & Pedersen, 1980; Stormer et al., 1981).
4-Hydroxyochratoxin A is non-toxic for rats in amounts up to 40 mg/kg
body weight (Hutchison et al., 1971); thus, it has been concluded
that the microsomal hydroxylation most likely represents a
detoxification reaction.

I.3.4 Excretion

The results of studies in which ¹⁴C-labelled ochratoxin A was
injected intraperitoneally in rats demonstrated that the toxin was
excreted primarily in the urine (Chang & Chu, 1977), though faecal
elimination also occurred to some extent (Galtier, 1974b; Chang &
Chu, 1977). In a study in which ¹⁴C-labelled ochratoxin A was
given orally to rats as a single dose (15 mg/kg body weight), the
cumulative elimination after 120 h was 11% ochratoxin A and 23%
ochratoxin alpha in the faeces; 11% ochratoxin A and 12% ochratoxin
alpha in the urine; and 33% ochratoxin A in the bile (Suzuki et
al., 1977). Absorption was influenced by enteritis, which was
caused by the high ochratoxin A doses given (75% of the LD₅₀
value). Ochratoxin A injected intravenously as a single dose (4.1
mg/kg) in albumin-deficient and normal rats was excreted in the
bile and urine 20-70 times faster in the albumin-deficient rats
than in normal rats, indicating that the binding of ochratoxin A to
blood albumin delays the excretion of the compound through the
liver and kidney (Kumagai, 1985).

In rats given unlabelled ochratoxin A orally, Storen et al.
(1982) found 6% ochratoxin A, 1.5% (4R)-4-hydroxyochratoxin A, and
25-27% ochratoxin alpha in the urine; 12% ochratoxin A and 9%
ochratoxin alpha were found in the faeces.

The excretion of ochratoxin A in the milk was studied in
rabbits intravenously injected with 1-4 mg/kg body weight, as
single dose (Galtier et al., 1977a). At the highest dose injected,
the milk contained 1 mg ochratoxin A/litre; ochratoxin alpha and
4-hydroxy-ochratoxin A were not detected. Goats were given a
single dose of tritium-labelled ochratoxin A (0.5 mg/kg) and the
cumulative excretion (in terms of radioactivity) after 7 days
amounted to 53% in the faeces, 38% in the urine, 6% in the milk,
and 2% in the serum (Nip & Chu, 1979). Only a small fraction of
the radioactivity in milk was ochratoxin A, amounting to 0.026% of
the total ochratoxin A given.

In the Federal Republic of Germany, a study of human milk
obtained from women in two hospitals (patient category not stated)
revealed that 4 out of 36 samples (11.1%) contained ochratoxin A,
with a mean value of 0.024 ng/ml and a range of 0.017-0.030 ng/ml
(Bauer & Gareis, 1987; Gareis et al., 1988).

In mice given ¹⁴C-labelled ochratoxin A intravenously at
various stages of pregnancy, the toxin was shown to cross the
placental barrier on day 9 of pregnancy, at which time it is most
effective in producing fetal malformations (Appelgren & Arora,
1983). The highest toxin concentration was found in the bile,
which contained 5 times as much as the blood.

Ochratoxin A has been detected in the urine of bacon pigs
suffering from nephropathy (Krogh, unpublished information
communicated to the Task Group).

In a study on the disappearance rates for various tissues,
female bacon pigs were fed ochratoxin A at a level of 1 mg/kg feed
for one month and then kept on a toxin-free diet for another month,
during which animals were sacrificed at regular intervals (Krogh et
al., 1976a). Ochratoxin A disappeared exponentially (Table 7) from
the 4 tissues investigated (kidney, liver, muscle, and adipose
tissue) with residual life values (RL₅₀)^a in the range of 3.5-4.5
days; the toxin could still be detected in the kidneys one month
after termination of exposure.

Table 7. The rate of disappearance of ochratoxin A residues
from pig tissues after termination of a one-month exposure to
ochratoxin A at 1 mg/kg feed^a
--------------------------------------------------------------
Tissue Ochratoxin A (µg/kg tissue) at time t (days)
after termination of exposure
---------------------------------------------------------------
kidney 28.22 exp (-0.1522 t)
liver 19.49 exp (-0.1598 t)
muscle 12.94 exp (-0.2096 t)
adipose 4.62 exp (-0.0565 t)
---------------------------------------------------------------
^a From: Krogh et al. (1976a).

-------------------------------------------------------------------
^a RL₅₀ = half residual life, calculated from the exponential
equations shown in Table 7.

No data are available on ochratoxin levels in human urine or
faeces.

When the level in the serum is known, the ochratoxin A residues
in the four other tissues can be calculated (Table 5).

I.4 EFFECTS ON ANIMALS

I.4.1 Field observations

I.4.1.1 Pigs

The effects of ochratoxins on animals have been reviewed by
Krogh (1976a, 1978). Cases of mycotoxic porcine nephropathy have
been regularly encountered in studies in Denmark since the disease
was first discovered 50 years ago (Larsen, 1928). The disease is
endemic in all areas of the country, though unevenly distributed.
Prevalence rates in 1971 varied from 0.6 to 65.9 cases per 10 000
pigs, and epidemics encountered in 1963 and 1971 were associated
with a high moisture content in the grain caused by unusual
climatic conditions (Krogh, 1976b). On the basis of these studies,
Krogh (1978) concluded that ochratoxin A is the substance most
frequently associated with porcine nephropathy, though other
factors, such as citrinin, may also be involved.

A survey on porcine nephropathy was conducted at six
slaughterhouses in Sweden during the spring months of 1978
(Josefsson, 1979). A prevalence rate of 4.4 cases per 10 000 pigs
was encountered corresponding to the endemic level of prevalence
rates in Denmark; 26.7% of nephropathic kidneys contained residues
of ochratoxin A. In Hungary, an epidemiological study on porcine
nephropathy and the association with ochratoxin A was conducted in
1980-81 covering 4 areas in the country (Sandor et al., 1982). A
prevalence rate of 2.0 cases per 10 000 pigs was measured,
comparable to endemic prevalence rates in Scandinavia; 39% of
nephropathic kidneys contained residues of ochratoxin A. The
morphological changes in the kidneys in cases of mycotoxic porcine
nephropathy were characterized by degeneration of the proximal
tubules followed by atrophy of the tubular epithelium, interstitial
fibrosis in the renal cortex, and hyalinization of some glomeruli
(Elling & Moller, 1973).

In Poland, surveys for porcine nephropathy in 1983 and 1984
revealed prevalence rates of 4.7-5.7 cases per 10 000 pigs, with 5-
55% of the nephropathic kidneys containing detectable residues of
ochratoxin A, apparently depending on the season of the year
(Golinski et al., 1984, 1985). Porcine nephropathy has been
encountered in the Federal Republic of Germany (Bauer et al., 1984)
and in Belgium (Rousseau & van Peteghem, 1989), with respectively,
21% and 18% of the affected kidneys containing residues of
ochratoxin A, in the range of 0.1-12 ng/g. In Canada, 1200 samples
of pig blood, collected at a slaughterhouse, were screened for
ochratoxin A using HPLC (Marquardt et al., 1988). Levels exceeding
10 ng/ml (11.3%) were found in 136 samples; detection was confirmed
by derivative formation and spectrometry. No kidney examination

was conducted, but the ochratoxin A concentrations detected in the
blood suggest that nephropathy might have been present in some of
the pigs.

I.4.1.2 Poultry

In a preliminary study in Denmark on chickens condemned by meat
inspectors because of renal lesions, 4 out of 14 birds (29%) were
found to have nephropathy associated with the ingestion of
ochratoxin A, as revealed by the presence of residues of ochratoxin
A in tissues (Elling et al., 1975). The renal lesions were
characterized by degeneration of proximal and distal tubules of
both reptilian and mammalian nephrons and interstitial fibrosis.

I.4.2 Experimental animal studies

I.4.2.1 Acute and chronic effects

The acute and chronic effects of ochratoxins on experimental
animals have been reviewed by Chu (1974a), Harwig (1974), and Krogh
(1976a). Different species vary in their susceptibility to acute
poisoning by ochratoxin A with LD₅₀ values ranging from 3.4 to 30.3
mg/kg (Table 8). When ochratoxin A was administered orally to rats
and guinea-pigs, the female was more sensitive than the male. In
rats, the kidney is the target organ, but necrosis of periportal
cells in the liver has also been noted during studies on acute
effects (Purchase & Theron, 1968).

Table 8. Acute toxicity of ochratoxin A
-------------------------------------------------------------------
Animal LD₅₀ (mg/kg Route of Reference
body weight) administration
-------------------------------------------------------------------
mouse 22 intraperitoneal Sansing et
(female) al. (1976)

rat (male) 30.3 oral Galtier et
al. (1974)

rat 21.4 oral Galtier et
(female) al. (1974)

rat (male) 28 oral Kanizawa et
al. (1977)

rat (male) 12.6 intraperitoneal Galtier et
al. (1974)

rat 14.3 intraperitoneal Galtier et
(female) al. (1974)

guinea-pig 9.1 oral Thacker &
(male) Carlton
(1977)
-------------------------------------------------------------------

Table 8 (contd.)
-------------------------------------------------------------------
Animal LD₅₀ (mg/kg Route of Reference
body weight) administration
-------------------------------------------------------------------
guinea-pig 8.1 oral Thacker &
(female) Carlton
(1977)

white 3.4 oral Prior et
leghorn al. (1976)

turkey 5.9 oral Prior et
al. (1976)

Japanese 16.5 oral Prior et
quail al. (1976)

rainbow 4.7 intraperitoneal Doster et
trout al. (1972)

beagle dog 9 oral^a Szczech et
(male) (total dose) al. (1973a)

pig 6 oral^b Szczech et
(female) (total dose) al. (1973b)
-------------------------------------------------------------------
^a All 3 dogs, dosed daily with 3 mg/kg body weight, died within
3 days.
^b Both pigs receiving 2 mg/kg daily were moribund and killed
within 3 days, and both pigs receiving 1 mg/kg daily were
moribund and killed within 6 days.

The lesions observed in field cases of mycotoxic nephropathy
have been reproduced by feeding diets containing levels of
ochratoxin A identical to those encountered in the naturally
contaminated products. In a study by Krogh et al. (1974a), 39 pigs
fed rations containing ochratoxin A, at levels ranging from 200 to
4000 µg/kg, developed nephropathy after 4 months at all levels of
exposure. Changes in renal function were characterized by
impairment of tubular function, indicated particularly by a
decrease in Tm_PAH/C_in,^a and reduced ability to produce concentrated
urine. These functional changes corresponded well with the changes
in renal structure observed at all exposure levels, including
atrophy of the proximal tubules and interstitial cortical fibrosis.
Sclerosed glomeruli were also observed in the group receiving the
highest dose of ochratoxin A of 4000 µg/kg feed. Changes were not
seen in any other organs or tissues.

-------------------------------------------------------------------
^a Tm_PAH = transport maximum for para-aminohippuric acid.
C_in = clearance of insulin.

Kidney damage, identical to naturally occurring porcine
nephropathy, was produced in another study by feeding pigs (9
animals) with crystalline ochratoxin A in amounts corresponding to
a feed level of 1 mg/kg for 3 months. Significant renal tubular
functional impairment as measured by a decrease in Tm_PAH C_in was
detected after only 5 weeks of ochratoxin exposure (Krogh et al.,
1976b). The study was continued for a 2-year period during which
the renal impairment aggravated slightly without reaching a state
of terminal renal failure (Krogh et al., 1979).

In 2 pigs and 9 rats dosed orally with ochratoxin A (400 and
250 µg/kg body weight, respectively) for 5 days, ochratoxin A was
detected in the epithelial cells of the proximal convoluted tubules
of the nephron of all animals. The method of detection was
immunofluorescence microscopy using an antibody against ochratoxin
A that had been formed in rabbits after injection of an albumin-
ochratoxin A conjugate (Elling, 1977).

Groups of 80 F 344/N rats of each sex were administered 0, 21,
70, or 210 µg ochratoxin A/kg in corn oil by gavage, 5 days per
week for 103 weeks (Boorman, 1988). The administration of
ochratoxin A to male and female rats caused a spectrum of
degenerative and proliferative changes in the kidney. The
predominant non-neoplastic lesion in treated rats was degeneration
of the renal tubular epithelium in the inner cortex and the outer
stripe of the outer medulla (nephropathy).

In four pigs given 0.8 mg ochratoxin A/kg body weight orally
for 5 consecutive days, the activity of catalase and CN-insensitive
palmitoyl-CoA-dependent NAD (nicotinamide adenine di-nucleotide) in
renal homogenates decreased, but that in hepatic homogenates did
not, suggesting peroxisomal changes. This was confirmed by
ultrastructural observations of peroxisomes in the proximal
tubules in kidneys of ochratoxin A-treated animals (Elling et
al., 1985).

When 11 SPF pigs and 23 beagle dogs were given high oral doses
of ochratoxin A corresponding to feed levels of more than 5-10
mg/kg (levels rarely found in nature), pathological effects, mainly
necrosis, were observed in the liver, intestine, spleen, lymphoid
tissue, leukocytes, and kidney (Szczech et al., 1973a,b,c). Three
groups of Wistar rats, each consisting of 15 animals, were exposed
to feed levels of ochratoxin A ranging from 0.2 to 5 mg/kg for 3
months. Renal damage in the form of tubular degeneration was
observed at all dose levels (Munro et al., 1974). A decrease in
urinary osmolality, glycosuria, and proteinuria were observed in
an unstated number of Sprague-Dawley and Wistar rats administered
daily doses intraperitoneally of 0.75-2 mg ochratoxin A/kg body
weight (Berndt & Hayes, 1979).

In a study of coagulation factors, rats were given daily oral
doses of 4 mg/kg body weight over 4-10 days, resulting in decreases
in plasma-fibrinogen levels, at levels of factors II, VII, and X,
and in the platelet and megakaryocyte counts (Galtier et al.,
1979).

In 4 calves fed rations containing 0.1-2 mg ochratoxin A/kg
body weight for 30 days, the only signs observed were polyuria,
increased levels of glutamic oxalacetic transaminase (GOT) in
serum, and mild enteritis (Pier et al., 1976); there was mild
tubular degeneration in the kidneys. Cows given ochratoxin A
orally for 4 days at doses ranging from 0.2 to 1.66 mg/kg body
weight remained clinically normal (Ribelin et al., 1978). A cow
given a single dose of 13.3 mg/kg body weight (corresponding to 865
mg/kg feed) developed diarrhoea, anorexia, and cessation of milk
production, one day after.

Avian nephropathy, similar to that in spontaneously occurring
cases, developed in Leghorn hens (27 birds per group) exposed to
dietary levels of 0.3 or 1 mg ochratoxin A/kg for one year (Krogh
et al., 1976c). The renal changes included degeneration of the
tubular epithelium, mainly confined to the proximal and distal
tubules of both reptilian and mammalian nephrons; impairment of
glomerular and tubular function was also observed. "Acute
nephrosis" and "visceral gout" were observed in chickens exposed to
high levels of ochratoxin A (LD₅₀ values) (Peckham et al., 1971).
The same authors reported that ochratoxin B, the other naturally
occurring ochratoxin, was not highly toxic for chickens (LD₅₀, 54
mg/kg) or other animals.

Groups of chicks (40 birds per group) were fed diets containing
ochratoxin A levels in the range of 0-8 mg/kg for 3 weeks (Huff et
al., 1974). The group receiving the highest dose (8 mg/kg feed)
showed decreased packed blood cell volume, haemoglobin
concentration, and serum-iron and serum-transferrin saturation
percentages. In a similar study, Chang et al.(1979) observed that
lymphocytopenia developed at all dose levels. In the same study,
decreased bone strength, as measured physically by resistance to
fracturation, was observed at feed levels of 2-4 mg ochratoxin A/kg
(Huff et al., 1980). When groups of chicks were fed ochratoxin A
at levels of 0, 2, or 4 mg/kg for 20 days, concentrations of serum-
immunoglobulins (IgA, IgG, IgM) were reduced to 57-66% of normal
values in the toxin-exposed groups (Dwivedi & Burns, 1984).

In B6C3F₁ female mice, groups of 6-7 animals were administered
a total of 0, 20, 40, or 80 mg ochratoxin A/kg body weight ip on
alternate days over an 8-day period. A dose-related decrease in
thymic mass was observed as well as myelotoxicity, indicated by
bone marrow hypocellularity, due to decreased marrow pluripotent
stem cells, and granulocyte-macrophage progenitors (Boorman et al.,
1984).

When determining the LD₅₀ in mice following intraperitoneal
injection, synergistic effects were observed when ochratoxin A was
combined with citrinin as well as with penicillic acid (Sansing et
al., 1976). An additive effect was observed between ochratoxin A
and citrinin in terms of embryotoxicity in chicken embryos (Vesela
et al., 1983). In beagle dogs, when combined oral doses of
ochratoxin A (0.1-0.2 mg/kg body weight) and citrinin (5-10 mg/kg)
were injected intraperitoneally, synergism was observed with regard

to severity of clinical disease and mortality (Kitchen et al.,
1977a,b). Increased toxicity in terms of LD₅₀ values and
pathological changes was observed in rats, when the ochratoxin A
was given orally combined with either of the drugs biscoumacetate
or phenylbutazone, apparently because of displacement of the toxin
from binding sites on plasma-proteins (Galtier et al., 1980). The
toxic effects of ochratoxin A on the renal epithelial cells of the
monkey were demonstrated in in vitro studies in the form of
abnormal mitotic cells (Steyn et al., 1975).

I.4.2.2 Teratogenicity

Intraperitoneal injection of pregnant mice with ochratoxin A at
5 mg/kg body weight on one of gestation days 7-12 resulted in
increased prenatal mortality, decreased fetal weight, and various
fetal malformations, including exencephaly and anomalies of the
eyes, face, digits, and tail (Hayes et al., 1974). When a
combination of ochratoxin A (2 or 4 mg/kg body weight) and T-2
toxin (0.5 mg/kg body weight) was injected intraperitoneally in
mice on gestation day 8 or 10, ochratoxin A exacerbated the
incidence of T-2-induced gross malformations (tail and limb
anomalies); increased fetocidal effects were also noted (Hood et
al., 1978). Mice were given 3-5 mg ochratoxin A/kg body weight
intraperitoneally or orally on gestation days 8, 9, and 10 or 15,
16, and 17 (Szczech & Hood, 1981). Cerebral necrosis was found in
most fetuses from dams treated on days 15-17, but no cerebral
necrosis developed after treatment on days 8-10, when ochratoxin A
is overtly teratogenic. Pups of mice given 1.25 and 2.25 mg
ochratoxin A/kg orally on gestation days 15, 16, and 17 were tested
for surface righting, swimming, and pivoting (Poppe et al., 1983).
The results of all 3 tests indicated that a delay in development
had occurred; no dose-related pathological alterations were found.

Pregnant mice were administered 1-2 mg ochratoxin A/kg body
weight, and/or 5-20 mg zearalenone/kg body weight or 0.125 mg
diethylstilboestrol/kg body weight, orally, on day 9 of pregnancy,
either individually or in combination, and the offspring were
examined on day 19 (Arora et al., 1983). Teratogenic effects
produced by ochratoxin A, such as exencephaly, open eyelids, and
microphthalmia, were reduced or absent when the toxin was given in
combination with one of the 2 non-steroidal estrogenic substances,
zearalenone or diethylstilboestrol.

Rats were treated orally with ochratoxin A at 0.25, 0.50, 0.75,
1, 2, 4, or 8 mg/kg body weight on gestation days 6-15 (Brown et
al., 1976). Maternal toxicity was not observed below 4 mg/kg body
weight, but an increased incidence of fetal resorptions was
observed from 0.75 mg/kg body weight. All fetuses from dams given
0.25-0.75 mg/kg body weight weighed significantly less than control
fetuses, and fetuses from dams given 0.75 or 1 mg/kg body weight
were stunted. Reduced litters and decreased fetal weight were
observed in rats administered 5 mg ochratoxin A/kg body weight,
orally, on gestation day 8 (Moré et al., 1978).

Subcutaneous administration of ochratoxin A to rats (1.75 mg/kg
body weight) on gestation days 5-7 resulted in the highest number
of malformations, including hydrocephaly, omphalocele, and
anophthalmia as well as a shift in position of the oesophagus
(Mayura et al., 1982); lower doses (0.5 and 1 mg ochratoxin A/kg
body weight) did not have any teratogenic effects, and higher doses
(5 mg/kg) caused all fetuses to be resorbed. In a subsequent study
by the same authors (Mayura et al., 1983), using the same
ochratoxin A exposure conditions, it was shown that a diet low in
protein (10% of protein concentration in normal rat feed) enhanced
the teratogenic action of ochratoxin A in the rat. The combined
action of ochratoxin A exposure and a low protein diet also
resulted in decreases in mating and fertilization rates (22% and
39%, respectively) compared with the control group.

When rats were exposed to ochratoxin A and citrinin (another
nephrotoxic mycotoxin produced by species of the Aspergillus and
Penicillium genera) either singly or combined, enhanced teratogenic
effects were observed in terms of gross malformations, visceral
anomalies, and skeletal defects following combined oral
administration of 1 mg ochratoxin A/kg body weight and 30 mg
citrinin/kg body weight (Mayura et al., 1984). Maternal deaths
(22-40%) occurred after the administration of the combined dose on
days 5, 6, 7, and 14 of gestation, whereas administration of
individual toxins did not cause any maternal deaths and only
minimal malformations no matter which gestation day they were
administered.

Increased prenatal mortality and malformations, including
hydrocephaly, micrognathia, and heart defects, were observed in
hamsters injected intraperitoneally with ochratoxin A at doses of
5-20 mg/kg body weight on one of gestation days 7-9 (Hood et al.,
1976).

I.4.2.3 Mutagenicity

Ochratoxin A did not have any effects in a Bacillus subtilis
Rec- assay, measuring DNA damage when tested at 20 and 100 µg/plate
(Ueno & Kubota, 1976). Ochratoxin A was not mutagenic to
Salmonella typhimurium TA 1535, TA 1537, TA 1538, TA 98, or TA 100
at doses of up to 500 µg/plate, with or without exogenous metabolic
activation (Kuczuk et al., 1978; Wehner et al., 1978b). No
increase was observed in genetic changes at the ade 2 locus of
Saccharomyces cerevisiae after treatment with 50 or 100 µg/plate
ochratoxin A, with or without exogenous metabolic activation
(Kuczuk et al., 1978). Ochratoxin A did not induce mutations to
8-azaguanine resistance in C3H mouse mammary carcinoma cells (FM3A)
treated with doses of 5 or 10 µg/litre (Umeda et al., 1977).

When rats were orally exposed to ochratoxin A every 48 h for 12
weeks (corresponding to a feed level of 4 mg/kg), single-strand
breaks of DNA in renal and hepatic tissue (the only tissues
investigated) were more pronounced than those in control animals
(Kane et al., 1986a,b).

Contradictory results have been obtained by testing ochratoxin
A in the Salmonella assay (SOS chromo test without metabolic
activation) (Ueno et al., in press).

I.4.2.4 Carcinogenicity

Kanizawa & Suzuki (1978) have indicated that ochratoxin A is a
hepatic and renal carcinogen in male mice. A group of 10 male ddY
mice were fed a diet containing 40 mg ochratoxin A/kg for 44 weeks.
A group of 10 untreated controls were fed the basal diet. All
survivors were killed after 49 weeks. Hepatic cell tumours were
found in 5 out of 9 treated mice; no tumours were found in the 10
controls. Solid renal cell tumours were found in 2 out of 9
treated mice and none in the 10 controls. Cystic renal adenomas
were found in 9 out of 10 treated mice compared with none in the 10
controls.

Two groups of 50 male and 2 groups of 50 female B6C3F₁ mice
were fed diets containing 1 or 40 mg ochratoxin A/kg, respectively;
one group (control) was fed the basal diet. All survivors were
killed after 24 months. Eleven out of 49 male mice in the 40 mg/kg
group had renal carcinomas; 24 out of 49 male mice in the 40 mg/kg
groups showed renal adenomas. All male mice in the 40 mg/kg group
had microscopic evidence of nephropathy. A few females in the 40
mg/kg group exhibited nephropathic changes but no carcinomas or
adenomas. Compound-related lesions were absent in the controls
and the 1 mg/kg groups (Bendele et al., 1985). In a test for the
development of hyperplastic liver nodules in rats, ochratoxin A was
characterized as having both an initiating and a promoting activity
(Imaida et al., 1982).

Seven groups of 16 male ddY mice each were fed a diet
containing 50 mg ochratoxin A/kg for various periods ranging from 0
to 30 weeks followed by feeding of the basal diet until the end of
70 weeks (Kanizawa, 1984). After 15 weeks of ochratoxin A
exposure, 3 out of 15 animals had renal cell tumours; after 20
weeks, 1 out of 14 mice had renal cell tumours and 2 had hepatomas;
after 25 weeks of exposure, 2 out of 15 mice had renal cell tumours
and 5 had hepatomas; and after 30 weeks of exposure, 4 out of 17
mice had renal cell tumours and 6 had hepatomas; these tumours were
not observed in the controls. The nature of the renal cell tumours
was not further defined, but such tumours are mostly malignant. In
addition, pulmonary tumours were found in all groups, including the
controls, the incidence ranging from 20 to 73%. In a second study,
the effects of a combination of ochratoxin A and citrinin were
elucidated. Groups of 20 male ddY mice were fed diets containing
25 mg ochratoxin A/kg in combination with 100 or 200 mg
citrinin/kg for 70 weeks. Control groups were fed diets that did
not contain any toxins or the individual toxins at the
concentrations indicated above. In addition, one group was fed 25
mg ochratoxin A/kg feed for the first 25 weeks followed by 200 mg
citrinin/kg feed for the remaining period of time; another group
was exposed to the 2 toxins in the reverse order. Exposure to
citrinin alone did not produce any tumours. Exposure to ochratoxin
A alone resulted in renal cell tumours in 6 out of 20 mice, and in

hepatomas in 8 mice. Exposure to ochratoxin A and 100 mg
citrinin/kg feed did not result in any renal cell tumours, but 10
out of 19 mice had hepatomas. Exposure to ochratoxin A and 200 mg
citrinin/kg feed resulted in renal cell tumours in 10 out of 18
mice, and in hepatomas in 7 mice. Exposure to one toxin followed
by the other toxin did not produce any renal cell tumours, but
hepatomas were observed in less than 20% of the mice.

On the basis of these studies, IARC concluded that there was
limited evidence of carcinogenicity for animals, and inadequate
evidence of carcinogenicity for human beings (IARC, 1987a).

Groups of 80 F 344/N rats of each sex were administered 0, 21,
70 or 210 µg ochratoxin A/kg in corn oil by gavage, 5 days per
week, for 103 weeks (Boorman, 1988). In the male rats, renal
carcinomas were found in 16 out of 51 animals dosed with 70 µg/kg
and in 30 out of 50 animals dosed with 210 µg/kg; no carcinomas
were found in lower dose groups. In the female rats, renal
carcinomas were less common, as 1 out of 50 animals dosed with 70
µg/kg and 3 out of 50 animals dosed with 210 µg/kg had carcinomas;
no carcinomas were found in the lower dose groups. Renal adenomas
were found in all groups of male rats, with increasing frequencies
associated with increased doses. In the female groups, renal
adenomas were only found in the two highest dose groups. In the
female rats, fibroadenomas in the mammary gland were found in 45-
56% in the treated groups, a significantly higher percentage than
in the control group.

I.4.2.5 Biochemical effects and mode of action

Ochratoxin A is an inhibitor of tRNA synthetase and protein
synthesis in several microorganisms ( Bacillus subtilis, B.
stearothermophilus, Streptococcus faecalis, yeasts) as well as in
rat hepatoma cells (Konrad & Roschenthaler, 1977; Bunge et al.,
1978; Heller & Roschenthaler, 1978; Creppy et al., 1979a,b). The
competitive inhibitor effect of ochratoxin A on tRNA synthetase and
protein synthesis in rat hepatoma cells can be prevented by the
addition of phenylalanine in the cell culture medium at a molar
ratio of phenylalanine: ochratoxin A of 1.7:1 (Creppy et al.,
1979b). This observation suggested the possibility of preventive
measures for ochratoxin A-induced disease. Thus, the acute
intraperitoneal effect of ochratoxin A (LD₁₀₀) in mice was
prevented by concomitant injection of phenylalanine (Creppy et al.,
1980; Moroi et al., 1985).

However, in the study on the reproduction of porcine
nephropathy previously mentioned (Krogh et al., 1974a), the molar
ratio of phenylalanine: ochratoxin A in the feed exceeded 4600:1,
implying that, in the field situation, phenylalanine does not
prevent ochratoxin A from inducing the development of nephropathy.

Ochratoxin A in the concentration range studied (20-1667 µmol/
litre) caused a 47-50% inhibition of macrophage migration (Klinkert
et al., 1981); this effect could be prevented by the simultaneous

addition of phenylalanine. In BALB/c mice, a dose of ochratoxin A
as low as 0.005 µg/kg body weight was able to suppress the immune
response to sheep erythrocytes (Haubeck et al., 1981); the effect
could be prevented by the simultaneous addition of phenylalanine.
Studying the same effect in Swiss mice, Prior & Sisodia (1982) were
unable to show any suppression of the immune response to sheep
erythrocytes, even after daily injection of 5 mg ochratoxin A/kg
for 50 days. (4R)-4-Hydroxyochratoxin A at a dose of 1 µg/kg body
weight caused an 80% reduction in the number of cells producing IgM
and a 93% reduction in cells producing IgG in BALB/c mice compared
with 90% and 92%, respectively, for ochratoxin A (Creppy et al.,
1983).

Female B6C3F₁ mice (6 per group) were administered ochratoxin A
in amounts of 0.34, 6.7, or 13.4 mg/kg body weight or ochratoxin B
(13.4 mg/kg body weight) 6 times during 12 days. Ochratoxin A
inhibited the natural killer cell activity at all dose levels, and
increased the growth of transplantable tumour cells without
affecting T-cell or macrophage-mediated anti-tumour activity
(Luster et al., 1987). Ochratoxin B did not influence immune
function. The inhibition by ochratoxin A of natural killer cell
activity appeared to be caused by reduced production of basal
interferon.

Ochratoxin A affects the carbohydrate metabolism in rats.
Thus, a single oral dose of ochratoxin A at 15 mg/kg body weight
caused a decrease in the glycogen level in the liver and an
increase in the heart glycogen level, 4 h later (Suzuki & Satoh,
1973).

In a more extensive study on rats, the decrease in liver
glycogen level, 4 h after a single oral dose of ochratoxin A at 15
mg/kg body weight, was associated with an increase in serum glucose
levels and a decrease in liver glucose-6-phosphate (Suzuki et al.,
1975). At the same time, the liver glycogen synthetase (EC
2.7.1.37) activity decreased and the liver phosphorylase (EC
2.4.1.1) activity increased. Three daily oral doses of ochratoxin
A at 5 mg/kg body weight caused a decrease in the liver glycogen
concentration, which was measured on the fourth day. The decrease
was attributed to inhibition of the active transport of glucose
into the liver, suppression of glycogen synthesis from glucose, and
acceleration of glycogen decomposition.

During in vitro studies on rat liver mitochondria, it was
observed that ochratoxin A inhibited the respiration of whole
mitochondria by acting as a competitive inhibitor of transport
carrier proteins located in the inner mitochondrial membrane
(Meisner & Chan, 1974). Further studies with mitochondrial
preparations revealed that the mitochondrial uptake of ochratoxin A
was an energy-using process that resulted in depletion of
intramitochondrial adenosine triphosphate (ATP), and that
ochratoxin A inhibited intramitochondrial phosphate transport
resulting in deterioration of the mitochondria (Meisner, 1976).
This might explain the degeneration of liver mitochondria observed
by Purchase & Theron (1968) in rats exposed orally to a single dose

of ochratoxin A at 10 mg/kg body weight. These authors observed
accumulation of glycogen in the cytoplasm of the rat liver cells
microscopically. This was in contrast to the previously discussed
observations of Suzuki et al. (1975), who found a decrease in
glycogen levels.

In a study on mice, Sansing et al. (1976) found that ochratoxin
A, administered intraperitoneally at 6 mg/kg body weight, inhibited
orotic acid incorporation into both liver and kidney RNA, 6 h after
toxin injection. In this respect, ochratoxin A acted synergistically
with another nephrotoxic mycotoxin, citrinin.

When neonatal rats were exposed orally to a single dose of 1 mg
ochratoxin A/kg or of 25 mg citrinin/kg or both doses within 24 h
of birth, a synergistic effect of the 2 mycotoxins was observed on
cytochrome P-450, NADPH-dependent dehydrogenase, and NADPH-
cytochrome c reductase (Siraj et al., 1981).

In rats fed 2 mg ochratoxin A/kg body weight per day for 2
days, renal gluconeogenesis from pyruvate was decreased by 26%, and
renal phosphoenol-pyruvate carboxy kinase (PEPCK) (EC 4.1.1.32)
activity was reduced by 55%, whereas hepatic PEPCK was unchanged
(Meisner & Selanik, 1979). In a subsequent study on rats, it was
found that even lower dose levels (0.3-0.5 mg ochratoxin A/kg body
weight) caused a 50% reduction in PEPCK activity (Meisner &
Meisner, 1981). A number of other enzymes located in the proximal
tubule of the nephron were unaffected. When longer exposure
periods (8-12 weeks) were used, and 145 ng ochratoxin A/kg body
weight were administered orally to groups of 3 rats each, increased
urinary excretion and corresponding renal tubular depletion of the
following enzymes were observed: gamma-glutamyl transferase,
alkaline phosphatase, leucine aminopeptidase, lactate
dehydrogenase, and N-acetyl-beta- D-glucosaminidase (Kane et al.,
1986a). Renal PEPCK in pigs was also sensitive to ochratoxin A
with a feed level of 100 µg/kg causing a significant decrease in
PEPCK activity (Meisner & Krogh, 1982).

In a study on pigs fed ochratoxin A at 0, 0.2, or 1 mg/kg for 5
weeks, enzyme activities were measured in renal biopsies, collected
1, 3, and 5 weeks after initiation (Krogh et al., 1988). After one
week, the activities of renal PEPCK and gamma-glutamyltranspeptidase
were decreased by 40%. The dose-related decrease in the activity
of PEPCK and gamma-glutamyltranspeptidase was accompanied by a
dose-related increase in renal impairment, as measured by the
reduction of Tm_PAH/C_in, suggesting that these enzymes are sensitive
indicators of ochratoxin-induced porcine nephropathy. Thus, the
renal biopsy-based measurements of enzyme activities might prove
diagnostically useful in ochratoxin-induced disease in human
beings.

Ochratoxin A reduced the total renal mRNA concentration in male
Sprague-Dawley rats and certain mRNA species, notably PEPCK, were
reduced to a greater extent than the bulk of the RNA pool (Meisner
et al., 1983).

I.5 EFFECTS ON MAN

I.5.1 Ochratoxin A, Balkan endemic nephropathy, and tumours of the
urinary system

This topic has been reviewed by Krogh (1979, 1983) who called
attention to the striking similarities in the changes in renal
structure and function induced experimentally in animals by the
administration of ochratoxin A, and the clinical and pathological
features of a localized endemic disease known as Balkan endemic
nephropathy. So far, the disease has been observed only in rural
populations of Bulgaria, Romania, and Yugoslavia, but information
on the present magnitude of the problem was not available to the
Task Group.

The Balkan endemic nephropathy is a chronic disease that
predominantly affects women and progresses slowly up to death
(Hrabar et al., 1976, Chernozemsky et al., 1977). Age-specific
incidence rates are highest above the age of 40. Younger cases
occur in the 10-19-year-old age group, and the mean age of new
patients is in the early 50s (Stoyanov et al., 1978).

The disease is characterized by an extreme geographical
clustering with a tendency for familial aggregation of cases
(Nicolov et al., 1978). However, sporadic cases occur outside
endemic areas. Age-adjusted incidence rates of 555 per 100 000
population in females and 322 in males over a ten-year period have
been recorded from a population sample of 147 000 in an endemic
area of Bulgaria (Stoyanov et al., 1978). In one of several
endemic regions in Yugoslavia, the prevalence varied from 3% to 8%
(Hrabar et al., 1976).

Autopsy has shown that kidneys are notably reduced in size.
The histological lesions are interstitial fibrosis, tubular
degeneration, and hyalization of glomeruli in the more superficial
part of the cortex (Heptinstall, 1974). Impairment of tubular
function, indicated by a decrease in Tm_PAH, is a prominent and
early sign (Dotchev, 1973).

A high incidence of tumours of the urinary system is strongly
correlated with the prevalence of Balkan endemic nephropathy
(Ceovic et al., 1976; Chernozemsky et al., 1977; Nicolov et al.,
1978). In one instance in Bulgaria, 46.6% of patients with tumours
of the urinary system were also affected by endemic nephropathy.
Among the tumours of the urinary system, cancers of the renal
pelvis and ureters are more frequently associated with endemic
nephropathy than urinary bladder tumours.

The relative risk for developing cancer of the renal pelvis and
ureters is 88:1 in patients with nephropathy compared with controls
in non-endemic areas. The relative risk for the development of any
tumour of the urinary system is only 28:1 in the same sample
(Stoyanov et al., 1978).

Over the past 2 decades, investigations have been carried out
to verify a variety of etiological assumptions with unconvincing
results (review by Puchlev, 1973, 1974). However, the assumption
of the possible role of ochratoxin A, on the basis of similarities
with the animal disease, has received increasing epidemiological
support.

In Yugoslavia, surveys indicated that the contamination of
foodstuffs (grains, maize, pork meat) with ochratoxin A occurred in
12.8% of samples in an area where the prevalence of endemic
nephropathy was 7.3%, compared with only 1.6% of contaminated
samples in areas free of the disease (Krogh et al., 1977). The
concentration of ochratoxin A in maize was 5-90 µg/kg and that in
pork meat, 5 µg/kg (Krogh et al., 1977) with levels of up to 27
µg/kg in pig kidneys (Pepeljnjak et al., 1982). Similarly, studies
in Bulgaria have revealed that 16.7% of beans and 27.3% of maize
from an endemic area were contaminated with ochratoxin A compared
with 7.1% and 9%, respectively, from a non-endemic area (Petkova-
Bocharova & Castegnaro, 1985).

In a subsequent survey of home-produced foodstuffs (cereal and
bread) from the same endemic area in Yugoslavia over a 5-year
period, a mean contamination of 8.7% was found with pronounced
annual variations, which probably reflect climatic conditions
during the crop harvesting periods (Pavlovic et al., 1979).

Surveys on the presence of ochratoxin A in blood samples are
difficult to compare, as different analytical methods have been
used with different levels of sensitivity. Prevalences of 16.6%
and 5.9% with a one-year interval were reported in the same endemic
village in Yugoslavia (Hult et al., 1982). It is not possible to
determine whether this difference reflects annual variations in the
content of the blood, or the result of a less sensitive analytical
method in the second instance. However, higher prevalences of
ochratoxin A and higher concentrations in blood are generally
present in people from endemic areas, especially in persons
suffering from Balkan nephropathy.

A survey in Yugoslavia reported a prevalence in the blood of
16.6% in an endemic village and 6% in a non-affected one (Hult et
al., 1982). In Bulgaria, reported rates were 17.7% in an endemic
area and 7.7% in a non-endemic one (Petkova-Bocharova et al.,
1988). Table 9 illustrates the trend towards higher concentrations
in endemic situations and in patients. The similarity of results
in healthy families from affected villages and healthy persons from
unaffected villages (groups III and IV) suggests than an
environmental or a behavioural determinant plays a role at the
household level.

Thus, available epidemiological information seems to indicate
that Balkan endemic nephropathy is associated with consumption
patterns involving foodstuffs contaminated with ochratoxin A and
with a higher frequency of positive blood samples of ochratoxin A.
However, the association does not permit the establishment of a

causal relationship. Cross sectional surveys, such as those
reported in the literature so far, are probably not the appropriate
means to determine this relationship.

The results of experimental animal studies suggest that Balkan
nephropathy, a chronic condition, may require a long latency period
between exposure and the onset of symptoms, or more likely a
prolonged exposure or repeated exposure over a long period of time.

Investigations based on individual exposure time sequences
together with follow-up cohort surveys could provide a clue to the
causal role of ochratoxin A in Balkan endemic nephropathy. In view
of the lack of this information, such a causal relationship cannot
be established or rejected.

Table 9. Ochratoxin A in blood samples from people in endemic and
non-endemic areas in Bulgaria^a
-------------------------------------------------------------------
Group^b No. of No. of Mean concentration
persons ochratoxin A ± S.D. (µg/kg)
assayed positive cases
(%)
-------------------------------------------------------------------
I. Persons with 61 16 (26.3) 20.3 ± 9.7
UST and/or EN

II. Healthy 63 10 (15.8) 14.5 ± 7.6
persons from
families with UST
and/or EN cases

III. Healthy 63 7 (11.1) 12.5 ± 3.5
persons from
families in
endemic villages

IV. Healthy 60 7 (11.6) 15.0 ± 4.2
persons from
unaffected villages
in endemic areas

V. Healthy 65 5 (7.7) 10.0
persons from
villages in
non-endemic areas
-------------------------------------------------------------------
^a From: Petkova-Bocharova et al. (1988).
^b UST: urinary system tumours; EN: endemic nephropathy.
The differences between group I and groups III and IV are
statistically significant ( P <0.002). Difference between
groups I and V is statistically significant ( P <0.001).

I.6 EVALUATION OF THE HUMAN HEALTH RISKS

Human exposure, as demonstrated by the occurrence of ochratoxin
A in food and in the blood, has been observed in various countries
in Europe. The Task Group was not aware of attempts to detect
ochratoxin A in human blood in other parts of the world.

The causal role of ochratoxin A in porcine nephropathy has been
established, based on studies of field cases as well as
reproduction of the disease with ochratoxin A. Using the porcine
model, it has been postulated that Balkan endemic nephropathy may
result from exposure to ochratoxin A. Available epidemiological
information indicates that Balkan nephropathy may be associated
with the consumption of foodstuffs contaminated by this toxin.
Since the publication of Environmental Health Criteria 11, in 1979,
epidemiological studies on the concentration of ochratoxin A in
human blood in affected and non-affected areas, have provided
additional support for the relationship between Balkan nephropathy
and exposure to ochratoxin A.

It has been shown that both the prevalence of ochratoxin A in
the blood and the blood concentrations are higher in residents in
endemic areas. However, a direct causal relationship cannot be
established on the basis of indirect evidence provided by the above
retrospective studies alone. Neither can it be excluded in view of
the long latency period between the exposure and the onset of
symptoms.

Ochratoxin A has been demonstrated to be carcinogenic to the
renal tubular epithelium in male mice and both sexes of rats. A
highly significant relationship has been observed between Balkan
nephropathy and tumours of the urinary tract, particularly with
tumours of the renal pelvis and ureters. However, there are no
published data to establish a direct causal role of ochratoxin A in
the etiology of such tumours.

II. TRICHOTHECENES

II.1 PROPERTIES AND ANALYTICAL METHODS

II.1.1 Physical and chemical properties

The sesquiterpenoid trichothecenes possess the tetracyclic
12,13-epoxytrichothecene skeleton. A total of 148 trichothecenes,
83 non-macrocyclic and 65 macrocyclic, have been isolated from
fungal cultures and plants (Drove, 1988). They can be conveniently
divided into 4 categories according to similarity of functional
groups (Ueno, 1977). The first class is characterized by a
functional group other than a ketone at C-8 (type A). This is the
largest category containing members such as T-2 toxin and
diacetoxyscirpenol (DAS). The second category of trichothecenes
has a carbonyl function at C-8 (type B) typified by 4-deoxynivalenol
(DON) and nivalenol (NIV). The third category is characterized by
a second epoxide group at C-7,8 or C-9,10 (type C), and the fourth
contains a macrocyclic ring system between C-4 and C-15 with two
ester linkages (type D). The structures of representative
trichothecenes of each category are illustrated below.

II.1.1.1 Physical properties

The trichothecenes are colourless, mostly crystalline solids
that have been well characterized by physical and spectroscopic
techniques (Cole & Cox, 1981). The type A trichothecenes are

soluble in moderately polar solvents, such as chloroform, diethyl
ether, ethyl acetate, and acetone, whereas the more polar type B
trichothecenes require higher polarity solvents, such as aqueous
methanol or aqueous acetonitrile. Some physical properties of the
main trichothecenes are summarized in Table 10.
Table 10. Some physical properties of main trichothecenes
-------------------------------------------------------------------------------
Trichothecenes Molecular Relative Melting (alpha)²⁰_D References
formula molecular point
mass (°C)
-------------------------------------------------------------------------------
T-2 toxin C₂₄H₃₄O₉ 466 151-152 +15 Bamburg et
al. (1968b)
HT-2 toxin C₂₂H₂₆O₈ 424 - -