Polychlorinated dibenzo-p-dioxins and dibenzofurans (EHC 88, 1989)

INTERNATIONAL PROGRAMME ON CHEMICAL SAFETY

ENVIRONMENTAL HEALTH CRITERIA 88

POLYCHLORINATED DIBENSO- PARA-DIOXINS AND DIBENZOFURANS

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 Orgnization
Geneva

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.

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

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 POLYCHLORINATED
DIBENZO-PARA-DIOXINS AND DIBENZOFURANS

1. SUMMARY AND RECOMMENDATIONS

1.1. Summary
1.1.1. Sources
1.1.2. Ambient levels and routes of exposure
1.1.3. Toxicokinetics, biotransformation, and
biological monitoring
1.1.4. Health effects
1.1.4.1 Animals
1.1.4.2 Humans
1.1.5. Conclusion
1.2. Recommendations

2. IDENTITY, PHYSICAL AND CHEMICAL PROPERTIES,
ANALYTICAL METHODS
2.1. Identity
2.2. Physical and chemical properties
2.3. Analytical methods
2.3.1. General aspects
2.3.2. Sampling strategy and sampling methods
2.3.3. Extraction procedures
2.3.4. Sample clean-up
2.3.5. Isomer identification
2.3.6. Quantification
2.3.7. Confirmation
2.3.8. Other analytical methods

3. SOURCES OF ENVIRONMENTAL POLLUTION
3.1. Production, synthesis, and use
3.2. Industrial processes
3.3. Contamination of commercial products
3.3.1. Chlorophenoxyacetic acid herbicides
3.3.2. Hexachlorophene
3.3.3. Chlorophenols
3.3.4. Polychlorinated biphenyls (PCBs)
3.3.5. Chlorodiphenyl ether herbicides
3.3.6. Hexachlorobenzene
3.3.7. Rice oil
3.4. Sources of heavy environmental pollution
3.4.1. Industrial accidents
3.4.2. Improper disposal of industrial waste
3.4.3. Heavy use of chemicals
3.5. Other sources of PCDDs and PCDFs in the
environment
3.5.1. Thermal degradation of technical
products
3.5.2. Incineration of municipal waste

3.5.3. Incineration of sewage sludge
3.5.4. Incineration of hospital waste
3.5.5. Incineration of hazardous waste
3.5.6. Metal industry and metal treatment
industry
3.5.7. Wire reclamation
3.5.8. Traffic
3.5.9. Fires and accidents in PCB-filled
electrical equipment
3.5.10. Pulp and paper industry
3.5.11. Incineration of coal, peat, and wood
3.5.12. Inorganic chlorine precursors
3.5.13. Photochemical processes
3.6. Comparison of isomeric pattern and congener
profiles from various sources

4. ENVIRONMENTAL TRANSPORT, DISTRIBUTION, AND
TRANSFORMATIONS

4.1. Environmental transport
4.1.1. Air
4.1.2. Water
4.1.3. Soil and sediments
4.2. Environmental transformation
4.2.1. Abiotic transformation
4.2.2. Biotransformation and biodegradation
4.3. Bioaccumulation
4.4. Levels in biota
4.4.1. Vegetation
4.4.2. Aquatic organisms
4.4.3. Terrestrial animals
4.4.4. Human data
4.4.4.1 Adipose tissue
4.4.4.2 Blood plasma

5. ENVIRONMENTAL LEVELS AND HUMAN EXPOSURE

5.1. Air
5.2. Water and leachate
5.3. Soil and sediment
5.4. Food
5.4.1. Meat and bovine milk
5.4.2. Human milk
5.4.3. Rice
5.5. Yusho and Yu-cheng episodes

6. KINETICS AND METABOLISM OF 2,3,7,8-TETRACHLORODIBENZO-
P-DIOXIN (TCDD) AND OTHER PCDDs

6.1. Uptake, distribution, and excretion
6.1.1. Studies on rats
6.1.2. Studies on mice

6.1.3. Studies on guinea-pigs
6.1.4. Studies on hamsters
6.1.5. Studies on monkeys
6.1.6. Studies on dogs
6.1.7. Studies on cows
6.1.8. In vitro studies
6.2. Metabolic transformation
6.2.1. Studies on mammals
6.2.1.1 In vivo studies
6.2.1.2 In vitro studies
6.3. Transfer via placenta and/or milk
6.4. Matrix effects on the uptake
("bio-availability")

7. EFFECTS OF TCDD AND OTHER PCDDs ON EXPERIMENTAL
ANIMALS AND IN VITRO TEST SYSTEMS

7.1. Acute toxicity
7.1.1. In vivo studies on mammals
7.1.2. In vitro studies on mammalian cells
7.1.3. Studies on birds
7.1.4. Toxicity of metabolites
7.1.5. Modulation of the acute toxicity
7.2. Short-term toxicity
7.2.1. Studies on rats
7.2.2. Studies on mice
7.2.3. Studies on guinea-pigs
7.2.4. Studies on hamsters
7.2.5. Studies on monkeys
7.3. Long-term toxicity
7.3.1. Studies on rats
7.3.2. Studies on mice
7.3.3. Studies on monkeys
7.4. Effects detected by special studies
7.4.1. Wasting syndrome
7.4.2. Hepatotoxicity
7.4.2.1 Morphological alterations
7.4.2.2 Hepatic plasma membrane
function
7.4.2.3 Biliary excretion
7.4.3. Porphyria
7.4.4. Epidermal effects
7.4.4.1 In vivo studies
7.4.4.2 In vitro studies
7.4.5. Effects on the immune system
7.4.5.1 Histopathology
7.4.5.2 Humoral-mediated immunity
7.4.5.3 Cell-mediated immunity
7.4.5.4 Macrophage function
7.4.6. Myelotoxicity
7.4.7. Effects on the intermediary
metabolism

7.4.8. Enzyme induction
7.4.8.1 Studies on rats
7.4.8.2 Studies on mice
7.4.8.3 Studies on guinea-pigs
7.4.8.4 Studies on rabbits
7.4.8.5 Studies on hamsters
7.4.8.6 Studies on cows
7.4.8.7 Studies on chick embryos
7.4.8.8 Studies on cell cultures
7.4.9. Endocrine effects
7.4.10. Vitamin A storage
7.5. Embryotoxicity and reproductive effects
7.5.1. Studies on rats
7.5.2. Studies on mice
7.5.3. Studies on rabbits
7.5.4. Studies on monkeys
7.5.5. Studies on chickens
7.6. Mutagenicity and related end-points
7.6.1. Mutagenicity
7.6.1.1 Studies on bacteria
7.6.1.2 Studies on eukaryotic cells
7.6.1.3 In vivo studies
7.6.2. Interaction with nucleic acids
7.6.3. Cytogenetic effects
7.6.4. Cell transformation
7.7. Carcinogenicity
7.7.1. Long-term animal studies on single
compounds
7.7.2. Long-term animal studies with mixed
compounds
7.7.3. Short-term and interaction studies
7.8. Mechanisms of action
7.8.1. Receptor-mediated effects
7.8.2. Toxicokinetics
7.8.3. Impairment of normal cellular regulatory
systems
7.8.3.1 Endocrine imbalance
7.8.3.2 Body weight regulation
7.8.3.3 Plasma membrane function
7.8.3.4 Impaired vitamin A storage
7.8.4. Lipid peroxidation

8. EFFECTS OF PCDDs ON HUMAN BEINGS - EPIDEMIOLOGICAL
AND CASE STUDIES

8.1. Occupational studies - historical perspective
8.2. General population studies
8.2.1. Missouri, USA
8.2.2. Seveso, Italy
8.2.3. Viet Nam
8.3. Signs and symptoms in humans associated with
TCDD exposure

8.3.1. Skin manifestations
8.3.2. Systemic effects
8.3.3. Neurological effects
8.3.4. Psychiatric effects
8.4. Epidemiological studies
8.5. Human experimental studies

9. TOXICOKINETICS OF PCDFs

9.1. Uptake, distribution, and excretion
9.1.1. Studies with 2,3,7,8-tetrachlorodibenzo-
furan (2,3,7,8-TCDF)
9.1.2. Studies with other PCDFs
9.2. Metabolic transformation
9.3. Transfer via placenta and/or milk

10. EFFECTS OF PCDFs ON ANIMALS

10.1. Acute toxicity
10.1.1. Studies on rats
10.1.2. Studies on mice
10.1.3. Studies on guinea-pigs
10.1.4. Studies on rabbits
10.1.5. Studies on monkeys
10.2. Short-term toxicity
10.2.1. Studies on rats
10.2.2. Studies on mice
10.2.3. Studies on guinea-pigs
10.2.4. Studies on rabbits
10.2.5. Studies on hamsters
10.2.6. Studies on monkeys
10.2.7. Studies on chickens
10.3. Chronic toxicity
10.3.1. Studies on monkeys
10.4. Effects detected by special studies
10.4.1. Immunobiological effects
10.4.1.1 Histopathology
10.4.1.2 Humoral-mediated immunity
10.4.1.3 Cell-mediated immunity
10.4.2. Enzyme induction
10.4.2.1 Studies on rats
10.4.2.2 Studies on mice
10.4.2.3 Studies on chickens
10.4.2.4 Studies on cell cultures
10.4.3. Receptor binding
10.5. Embryotoxicity and reproductive effects
10.6. Mutagenicity
10.7. Carcinogenicity

11. EFFECTS OF PCDFs ON HUMAN BEINGS

11.1. Yusho and Yu-cheng

12. EVALUATION OF HEALTH RISKS FROM THE EXPOSURE TO
CHLORINATED DIBENZO-P-DIOXINS (PCDDs) AND
DIBENZOFURANS (PCDFs)

12.1. Introduction
12.2. Exposure assessment
12.2.1. Sources of contamination
12.2.2. Ambient levels
12.2.3. Routes of exposure
12.2.4. Bioavailability
12.3. Animal data
12.3.1. Toxicokinetics of 2,3,7,8-TCDD
12.3.2. Toxicokinetics of PCDDs and PCDFs,
other than TCDD
12.3.3. Toxic effects 2,3,7,8-TCDD
12.3.4. Toxic effects of PCDDs and PCDFs,
other than TCDD
12.3.5. Review of species differences
12.4. Human health effects
12.4.1. PCDDs
12.4.2. PCDFs
12.4.3. Human body burden and kinetics
12.5. General conclusions

13. RECOMMENDATIONS

14. EVALUATIONS BY INTERNATIONAL BODIES AND THE CONCEPT
OF TCDD EQUIVALENTS

14.1. International evaluations
14.2. Methodologies used in assessment of
risk from PCDDs and PCDFs
14.2.1. Individual congeners
14.2.2. Mixtures of PCDD and PCDF congeners and
isomers - concept of TCDD toxic
equivalents

REFERENCES

FRENCH TRANSLATION OF SUMMARY, EVALUATION, AND
RECOMMENDATIONS

WHO TASK GROUP ON CHLORINATED DIBENZO-p-DIOXINS AND
DIBENZOFURANS

Members

Dr U.G. Ahlborg, Unit of Toxicology, National Institute of
Environmental Medicine, Stockholm, Sweden
Dr J.S. Bellin, Office of Toxic Substances, US Environmental
Protection Agency, Washington, DC, USA
Dr B. Birmingham, Ministry of the Environment, Hazardous Contaminants
Section, Toronto, Ontario, Canada
Professor A.D. Dayan, Department of Health and Social Security,
St Bartholomew's Hospital Medical College, London, United
Kingdom (Chairman)
Dr A. di Domenica, Instituto Superiore di Sanita, Rome, Italy
Dr M. Greenberg, Department of Health and Social Security,
Division of Toxicology and Environmental Protection, London,
United Kingdom
Dr R.D. Kimbrough, United States Department of Health and Human
Services, Center for Disease Control, Atlanta, Georgia, USA
(Now at the US Environmental Protection Agency Washington,
DC, USA)
Dr R. Koch, Department of Toxicology, Institute of Hygiene,
Gera, DDR
Professor C. Rappe, Department of Chemistry, University of
Umea, Umea, Sweden
Dr S. Safe, Texas A and M University, College Station, Texas,
USA
Dr H. Spielmann, Max von Pettenkofer Institute, Bundesgesundheitsamt,
Berlin (West)
Dr J. Vos, National Institute of Public Health and Environmental
Hygiene, Bilthoven, Netherlands

Representatives

Dr A. Berlin, Health and Safety Directorate, Commission of the
European Communities, Luxembourg
Mrs E. Cox, Department of the Environment, London, United
Kingdom
Miss F.D. Pollitt, Department of the Environment, London,
United Kingdom

Secretariat

Dr G.C. Becking, International Programme on Chemical Safety,
World Health Organization, Research Triangle Park, North
Carolina, USA (Secretary)

Secretariat (contd)

Dr H. Hakensson, Unit of Toxicology, National Institute of
Environmental Medicine, Stockholm, Sweden (Temporary
Adviser) (Rapporteur)
Dr E. Johnson, International Agency for Research on Cancer,
World Health Organization, Lyons, France
Dr S. Tarkowski, Regional Office for Europe, World Health
Organization, Copenhagen, Denmark

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.

* * *

A detailed data profile and a legal file can be obtained from the
International Register of Potentially Toxic Chemicals, Palais des
Nations, 1211 Geneva 10, Switzerland (Telephone No. 7988400 -
7985850).

ENVIRONMENTAL HEALTH CRITERIA FOR POLYCHLORINATED DIBENZO-PARA-
DIOXINS AND DIBENZOFURANS

A WHO Task Group on Environmental Health Criteria for
Polychlorinated Dibenzo-para-dioxins and Dibenzofurans met at the
Monitoring and Assessment Research Centre, London, United Kingdom,
from 9 to 13 February, 1987. Dr M. Berlin opened the meeting and
welcomed the members on behalf of the host Institute and on behalf of
the United Kingdom Department of Health and Social Security, who
sponsored the meeting. Dr G.C. Becking addressed the meeting on behalf
of the three cooperating organizations of the IPCS (UNEP, ILO, and
WHO). The Task Group reviewed and revised the draft criteria document
and made an evaluation of the risks for human health and for the
environment from exposure to polychlorinated dibenzo-p-dioxins and
dibenzofurans.

The drafts of this document were prepared by Dr U.G. Ahlborg, Dr
H. Hakensson, and Dr B. Holmstedt, all of the National Institute of
Environmental Medicine, Stockholm, Sweden, and by Professor C. Rappe
of the University of Umea, Umea, Sweden.

The efforts of all who helped in the preparation and finalization
of the document 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. The
United Kingdom Department of Health and Social Security generously
supported the cost of printing.

ABBREVIATIONS

AHH aryl hydrocarbon hydroxylase
ALA aminolevulinic acid
BGG bovine gammaglobulin
BHA butylated hydroxyanisole
BP benzo(a)-pyrene
CMI cell-mediated immunity
DEN diethylnitrosamine
diCDD dichlorinated dibenzo-p-dioxin
diCDF dichlorinated dibenzofuran
DMBA dimethylbenzathraline
ECOD 7-ethoxycoumarin-o-deethylase
EGF epidermal growth factor
EH epoxide hydratase
EI electron impact
EROD 7-ethoxyresurofin-o-deethylase
ETG epidermal transglutaminase
fg femtogram (10^-15g)
GC gas chromatography
heptaCDD heptachlorinated dibenzo-p-dioxin
heptaCDF heptachlorinated dibenzofuran
hexaCDD hexachlorinated dibenzo-p-dioxin
hexaCDF hexachlorinated dibenzofuran
HMI humoral-mediated immunity
HPLC high pressure liquid chromatography
IARC International Agency for Research on Cancer
ip intraperitoneal
IR infrared
LOEL lowest-observed-effect level
MCPA 4-chloro-o-tolyloxyacetic acid
MFO mixed-function oxidase
MS mass spectrometry
MSW municipal solid waste
ng nanogram (10^-9g)
NMR nuclear magnetic resonance
NOEL no-observed-effect level
octaCDD octachlorinated dibenzo-p-dioxin
octaCDF octachlorinated dibenzofuran
PAH polyaromatic hydrocarbons
PCB polychlorinated biphenyl
PCDD polychlorinated dibenzo-p-dioxin
PCDF polychlorinated dibenzofuran
PCDPE polychlorinated diphenylether
PCPY polychlorinated pyrene
PCQ polychlorinated quaterphenyl
pentaCDD pentachlorinated dibenzo-p-dioxin
pentaCDF pentachlorinated dibenzofuran
pg picogram (10-¹²g)
SC subcutaneous
SCE sister chromatid exchange

SD standard deviation
SEM standard error of the mean
SIM selected ion monitoring
TCDD 2,3,7,8-tetrachlorinated dibenzo-p-dioxin
TCDF 2,3,7,8-tetrachlorinated dibenzofuran
TCP trichlorophenol
tetraCDD tetrachlorinated dibenzofuran
tetraCDF tetrachlorinated dibenzofuran
TPA 12-o-tetradecanoylphorbol-13-acetate
triCDD trichlorinated dibenzo-p-dioxin
triCDF trichlorinated dibenzofuran
t3 triiodothyronine
t4 thyroxine
UDPGT UDP-glucuronosyltransferase
UV ultraviolet
2,4-D 2,4-dichlorophenoxyacetic acid
2,4,5-T 2,4,5-trichlorophenoxyacetic acid
3-MC 3-methylcholanthrene

1. SUMMARY AND RECOMMENDATIONS

1.1 Summary

1.1.1 Sources

Polychlorinated dibenzo-p-dioxins (PCDDs) and polychlorinated
dibenzofurans (PCDFs) are two series of tricyclic aromatic compounds
with similar chemical and physical properties; they are ubiquitous in
the environment. They do not occur naturally, nor are they
intentionally produced. There are 75 positional isomers of PCDDs and
135 isomers of PCDFs.

The most important sources of contamination with PCDDs and PCDFs
include:
- contaminated commercial chemical products, such as
chlorinated phenols and their derivatives, and PCBs;
- incineration of municipal, hazardous, and hospital
wastes, and of sewage sludges;
- automobile operation;
- fossil fuel combustion;
- overheating and emissions from fires involving PCBs;
- disposal of industrial wastes resulting from
processes such as the production of chlorophenols and
their derivatives, chlorophenol wood treatment, use
of PCB fluids in electrical equipment, and wastes
from pulp and paper processing.

1.1.2 Ambient levels and routes of exposure

The limited data available indicate that ambient levels of these
compounds are very low in air, soil, and sediment, i.e. fg/m³ in
air, ng/kg in soil and sediment. Levels of PCDDs and PCDFs up to 50
ng/kg have been found in aquatic organisms in the general environment.
Data on contamination of drinking water and commercial food are very
limited.

Exposure to these compounds in the general population probably
occurs mainly through the food-chain.

Some workers engaged in the production, use, and destruction of
materials containing PCDDs and PCDFs and their precursors may receive
high exposure. For these persons, inhalation and dermal contact are
the primary exposure routes of concern.

1.1.3 Toxicokinetics, biotransformation, and biological monitoring

The bioavailability of PCDDs and PCDFs depends on the matrix they
are in and the route of exposure. Data on bioavailability through
inhalation are not available for any species.

The quantity absorbed by humans after any route of exposure is
not known.

Studies on rodents given single or repeated oral doses of
2,3,7,8-TCDD have shown that about half of the administered dose is
absorbed from the gastrointestinal tract. The reported half-lives for
elimination were between 12 and 94 days for rodents. The half-life of
2,3,7,8-TCDD in adipose tissue of the rhesus monkey is about 1 year.

Animal data on the toxicokinetics of PCDDs other than
2,3,7,8-TCDD are limited. The half-life for 2,3,7,8-TCDD has been
reported to be in the range of 2 and 8 days for rats, mice, and
monkeys and more than 20 days for guinea-pigs. Studies on rats have
shown that 2,3,4,7,8-pentaCDF is more highly retained than is
2,3,7,8-TCDD.

Data on the retention of PCDDs and PCDFs in tissues of various
species, exposed to synthetic mixtures or to environmental samples
containing PCDDs and PCDFs, show a high variability in retention time
between congeners with or without chlorine substitution in the 2,3,7,
and 8 positions.

Limited human data indicate half-lives for some 2,3,7,8-
substituted PCDDs and PCDFs in the range of 2-6 years.

The PCDDs and PCDFs are predominately stored in fat, but they are
also excreted in milk and pass through the placenta. They also appear
in the blood and vital organs at lower concentrations.

The tissue distribution in humans is not clear at present,
although it has been suggested that the ratio between fatty tissue and
liver is higher in humans than in rodents.

In human fat, background levels of TCDD up to 20 ng/kg have been
found in the general population, with no known specific exposure, but
higher levels have been reported in some cases without evidence of
disease. None of these populations were randomly sampled. The more
highly chlorinated PCDDs and PCDFs, particularly octaCDD, are also
present in these samples. Average tissue levels of TCCD tend to
increase with age.

1.1.4 Health effects

1.1.4.1 Animals

The toxic and biological effects resulting from exposure to
2,3,7,8-TCDD are dependent on a number of factors, which include the
species, strain, age, and sex of the animals used. The toxic responses
observed in several animal species include body weight loss,
hepatotoxicity, porphyria, dermal toxicity, gastric lesions, thymus

atrophy and immunotoxicity, teratogenicity, reproductive effects, and
carcinogenicity. TCDD induces a wide spectrum of biological effects
including enzyme induction and vitamin A depletion. Not all of these
effects are observed in any single animal species. The most
characteristic toxic effects observed in all laboratory animals are
body weight loss, thymus atrophy, and immunotoxicity. Chloracne and
related dermal lesions are the most frequently noted signs of
2,3,7,8-TCDD toxicosis in humans; dermal lesions are also observed in
rhesus monkeys, hairless mice, and rabbits. In contrast, most rodents
do not develop chloracne and related dermal toxic lesions after
exposure to 2,3,7,8-TCDD. Many of the toxic lesions are noted
primarily in epithelial tissues.

Reproductive effects have been reported in rhesus monkeys and
rats. The lowest-observed-effect levels have been reported to be
approximately 1-2 ng/kg body weight per day. In two cancer studies in
rats, hepatocellular carcinomas were produced at approximate dose
levels of 0.1 µg/kg body weight per day and 0.01 µg/kg body weight per
day. Doses of 0.001 µg/kg body weight resulted in foci or areas of
hepatocellular alteration. The incidence of certain hormone-dependent
tumours was lower than in the control animals.

TCDD does not appear to have mutagenic properties, and is
therefore not likely to be genotoxic. Thus, it is assumed to be
carcinogenic through an indirect mechanism.

Several other PCDDs and PCDFs cause signs and symptoms similar to
those of 2,3,7,8-TCDD, but there is a wide variation with regard to
potency. There are 12 isomers that display higher toxicity, i.e., the
tetra-, penta-, hexa-, and heptaCDDs and CDFs with four chlorine atoms
in the symmetrical lateral positions 2,3,7, and 8. A mixture of two
hexachlorodibenzo-p-dioxins (1,2,3,7,8,9- and 1,2,3,6,7,8-hexaCDD)
has been demonstrated to possess carcinogenic properties in long-term
animal studies, but at higher doses than those used in the study of
TCDD. Dibenzo-p-dioxin and 2,7-diCDD failed to demonstrate
carcinogenic properties. The relative toxic and biological potencies
of PCDDs and PCDFs have been estimated using short-term studies in
rats and mammalian cell cultures.

There are marked species differences in the susceptibility of
animals to the biological and toxic effects elicited by
2,3,7,8-substituted PCDDs and PCDFs. For example, the oral LD50 values
range from 0.6 µg/kg body weight in guinea-pigs, to 5051 µg/kg body
weight in Golden Syrian hamsters for 2,3,7,8-TCDD. The tremendous
variation in species and strain sensitivity to 2,3,7,8-TCDD and
related compounds cannot be explained by the observed toxicokinetic
differences. The toxicity and toxicokinetics of TCDD in monkeys most
closely resemble the effects observed in humans. There is evidence in

inbred mice that the cellular levels of the Ah receptor correlate, in
part, with susceptibility to the biological and toxic effects of these
compounds. The receptor has also been identified in other species
including man. However, interspecies comparison of cellular Ah
receptor levels do not explain fully the differences in sensitivity.

1.1.4.2 Humans

For occupational and accidental exposures to PCDDs and PCDFs, in
spite of many clinical and follow-up studies, no clear-cut persistent
systemic effects have been delineated except for chloracne. Other
effects have been noted, but, apart from chloracne and perhaps minor
functional disorders, none has been persistent.

In some epidemiological studies of people exposed to a mixture of
dioxins, furans, and other chemicals, an increased incidence of cancer
at different sites has been claimed, but a number of factors limits
confidence in the findings.

In the Seveso accident, the only clear-cut adverse health effect
recorded has been chloracne. Chloracne (193 cases) occurred in 1976
and 1977, and 20 of those individuals still had active chloracne in
1984. Many studies have been performed to find possible links between
exposure to Agent Orange and health effects in civilians or military
personnel in Viet Nam. However, the information available to date does
not allow definite conclusions to be drawn with regard to effects on
human reproduction or any other significant health effects.

In the Missouri incident, children who showed acute illness when
the contamination occurred in 1971 are now reportedly in good health.
Furthermore, epidemiological studies in Missouri on populations
exposed to lower concentrations of dioxins over longer periods of time
have so far not revealed any significant health effects. Although no
clinical symptoms were observed, there were indications of an effect
on the cell-mediated immune system.

The only documented intoxications with PCDFs in humans are the
two instances of contamination of rice oil with PCDFs, PCBs, and PCQs,
i.e., Yusho in Japan, 1968, and Yu-cheng in Taiwan, 1979. In total,
several thousand people were acutely intoxicated. From the data it
appears most likely that the causative agent was the PCDFs. The
general symptomatology was similar to that seen in intoxications with
TCDD, with the differences reflecting the intensity of exposure and
the ages and sex of those exposed.

The average daily intake of 2,3,7,8-substituted PCDFs by Yusho
patients was estimated to be 0.1-0.2 µg/kg body weight for a period of
several months, while the lowest dose causing disease was estimated to
be 0.05-0.1 µg/kg body weight per day over a period of 30 days.

1.1.5 Conclusion

PCDDs and PCDFs occur throughout the environment and we all
probably carry a body burden of them. They have sometimes produced
complex toxic effects following occupational and accidental exposure.

Based on the Yusho disease and experiments in sensitive species
of monkeys, and making assumptions about the relative potencies of
PCDDs and PCDFs, man and certain monkeys may have comparable
sensitivity to these compounds. However, the uncertainties related to
the real dose received by humans and the difficulties of assessing
toxic effects other than chloracne in humans prevents a firm
conclusion as to the relative resistance of humans to the toxic
effects of these compounds. Exposure should be reduced to levels as
low as reasonably practicable.

1.2 Recommendations

1. Analytical interlaboratory validation and "round-robin" studies
using standardized quality assurance and quality control procedures
are needed to improve analytical methodology.

Sampling strategy and analytical procedures and data
interpretation should be optimized and standardized before undertaking
surveys.

2. Further information is required about the origins and
environmental distributon and fate of PCDDs and PCDFs.

Further monitoring data, including time trends and determinations
of isomer patterns, are required for environmental levels of PCDDs and
PCDFs, especially for food, ambient air, and sediments.

3. Data should be obtained about the effects of PCDDs and PCDFs on
environmental biota.

4. More information is required on the bioavailability of PCDDs and
PCDFs from different matrices in the environment and from the diet.
Exposure from these sources should be correlated with agricultural and
industrial practices.

5. Simpler and less expensive chemical and biological methods
suitable for screening for the presence of PCDDs and PCDFs should be
developed and validated.

6. Studies to determine the mechanisms of toxicity of PCDDs and
PCDFs are needed to support an evaluation of the differences in
effects between species and to support an extrapolation to man.

7. Further investigation of immunotoxicity is important, including
cytotoxic T-lymphocyte function. Studies of the effects of perinatal
exposure and of the duration of actions on the immune system are
important.

8. Long-term toxicity studies should be carried out, including
multigeneration reproductive studies in different species with three
of the most widespread PCDDs and PCDFs, namely 2,3,4,7,8-pentaCDF,
1,2,3,7,8-pentaCDD, and octaCDD.

9. Because humans are exposed to complex mixtures of PCDDs and
PCDFs, test systems, including human cell culture systems, should be
developed further and validated for evaluating the toxic potency of
these compounds and other mixtures. These systems can be used to study
mechanisms of action, structure activity relationships, and
interactive effects.

10. Investigations to examine the body burden and to correlate it
with clinical effects and laboratory findings are indicated. Follow-up
studies of previously exposed groups are important.

2. IDENTITY, PHYSICAL AND CHEMICAL PROPERTIES, ANALYTICAL METHODS

2.1 Identity

The polychlorinated dibenzo-para-dioxins (PCDDs) and
polychlorinated dibenzofurans (PCDFs) are two series of almost planar
tricyclic aromatic compounds with very similar chemical properties.
The general formulae are given in Fig. 1.

The number of chlorine atoms can vary between 1 and 8. The term
isomers refers to comparisons between compounds with the same
empirical formulae. The term congeners refers to comparisons between
compounds within the same series but with a different number of
chlorine atoms. The number of positional isomers is quite large; in
all there are 75 PCDDs and 135 PCDFs and the number of isomers for a
certain number of chlorine atoms is given in Table 1.

The nomenclature used in this document is based on the system
used by Chemical Abstracts. The Chemical Abstracts System Registry
Numbers (CAS RN) for a few PCDDs and PCDFs that have been cited in the
literature are provided in Table 2.

2.2 Physical and Chemical Properties

A large number of the individual PCDDs have been synthesized by
various methods and characterized, mainly by gas chromatography-mass
spectrometry (GC/MS) (Buser & Rappe, 1980, 1984; Taylor et al., 1985;
Rappe et al., 1985a) but also by using nuclear magnetic resonance
(NMR) or ultraviolet (UV), infrared (IR), (Pohland & Yang, 1972; Kende
et al., 1974), or X-ray analyses (Boer et al., 1973; Slonecker et al.,
1983).

Table 1. Number of PCDD and PCDF isomers

Number Number Number
of chlorine atoms of PCDD isomers of PCDF isomers

1 2 4
2 10 16
3 14 28
4 22 38
5 14 28
6 10 16
7 2 4
8 1 1
75 135

Table 2. CAS RN for some PCDDs and PCDFs

PCDD congener CAS RN PCDF congener CAS RN

2,3,7,8-TetraCDD 1746-01-6 2,3,7,8-TetraCDF 51207-31-9
1,2,3,7,8-PentaCDD 40321-76-4 1,2,3,7,8-PentaCDF 57117-41-6
1,2,3,6,7,8-HexaCDD 57653-85-7 2,3,4,7,8-PentaCDF 57117-31-4
1,2,3,7,8,9-HexaCDD 19408-74-3 1,2,3,4,7,8-HexaCDF 70648-26-9
1,2,3,4,6,7,8-HeptaCDD 35822-46-9 1,2,3,6,7,8-HexaCDF 57117-44-9
1,2,3,4,7,8,9-HeptaCDD 58200-70-7 1,2,3,7,8,9-HexaCDF 72918-38-8
OctaCDD 3268-87-9 2,3,4,6,7,8-HexaCDF 60851-34-5

Pyrolysis of chlorinated phenols yields small amounts of one or
more PCDD isomers. Using this technique all the 22 tetraCDDs have been
prepared (Nestrick et al., 1979; Buser & Rappe, 1980) as well as the
14 pentaCDDs (Buser & Rappe, 1984) and 10 hexaCDDs (Lamparski &
Nestrick, 1981; Buser & Rappe, 1984).

Taylor et al. (1985) have synthesized, separated, and isolated
all the 22 tetraCDD isomers. In Table 3 are listed some other isomers
that have been synthesized and isolated.

The most toxic and most extensively studied representative of the
chlorinated dioxins (PCDDs) is
2,3,7,8-tetrachlorodibenzo-para-dioxin (2,3,7,8-tetraCDD) (Fig. 2).
It is commercially available, as are more than 10 other PCDD
congeners.

The empirical formulae, molecular weights, and some physical
properties of a few PCDDs are given in Table 4.

Table 3. Synthetic method and melting point for some PCDDs

PCDD Synthetic Melting point Reference
Isomer method^a °C

1-Chloro- 1 80-90 Pohland & Yang, 1972
2-Chloro- 1 88-89 Pohland & Yang, 1972
1,3-Dichloro- 1 113.5-114.5 Kende et al., 1974
2,3-Dichloro- 1 163-164 Pohland & Yang, 1972
2,7-Dichloro- 2 209-210 Pohland & Yang, 1972
2,8-Dichloro- 3 150.5-151 Pohland & Yang, 1972
1,2,4-Trichloro- 4 128-129 Pohland & Yang, 1972
2,3,7-Trichloro- 1 157-158 Kende et al., 1974
2,3,7,8-Tetrachloro- 2 305-306 Pohland & Yang, 1972
2,3,7,8-Tetrachloro- 5 305-307 Kende et al., 1974
1,2,3,4-Tetrachloro- 4 188-190 Pohland & Yang, 1972
1,3,7,8-Tetrachloro- 1 193.5-195 Kende et al., 1974
1,3,6,8-Tetrachloro- 2 219-219.5 Pohland & Yang, 1972
1,2,3,4,7-Pentachloro- 5 195-196 Kende et al., 1974
1,2,3,4,7,8-Hexachloro- 5 275 Pohland & Yang, 1972
1,2,4,6,7,9-Hexachloro- 2 238-240 Pohland & Yang, 1972
Octachloro 2 330 Pohland & Yang, 1972

^a Synthetic methods as follows:

1 = Catechol + chlorobenzene
2 = Pyrolysis of chlorphenols
3 = Cyclization of chlorophenoxyphenol
4 = Catechol + chloronitrobenzene
5 = Chlorination of chlorodibenzodioxin

Table 4. Physical properties of some PCDDs

Molecular Molecular Absorption
Compound formulae weight maximum Reference
(chloroform)
(nm)

2,3,7,8-TCDD C₁₂H₄Cl₄O₂ 321.9 310 Pohland &
Yang (1972)

1,2,3,7,8-PentaCDD C₁₂H₃Cl₅O₂ 356.5 308 Gray et al.
(1976)

1,2,3,6,7,8-HexaCDD C₁₂H₂Cl₆O₂ 390.9 316 Gray et al.
(1975)

1,2,3,7,8,9-HexaCDD C₁₂H₂Cl₆O₂ 390.9 317 Gray et al.
(1975)

Although tetraCDD is lipophilic, it is only slightly soluble in
most solvents and very slightly soluble in water (Table 5).

Table 5. Solubility of 2,3,7,8-tetraCDD in various solvents^a

Solvent Solubility at 25 °C
g/litre g/kg

O-Dichlorobenzene 1.8 1.4
Chlorobenzene 0.8 0.72
Perchloroethylene 0.68 0.48
Chloroform 0.55 0.37
Benzene 0.47 0.57
Acetone 0.09 0.11
Dimethylsulfoxide^b < 0.1 < 0.1
Methanol 0.01 0.01
Water 2 x 10^-7 2 x 10^-7

^a From: Crummett & Stehl (1973).
^b DMSO caused detector fouling and a better value could not be obtained.

Table 6. Water solubility of PCDDs^a

Compound Water solubility (g/litre)
20.0 °C 40.0 °C

1,3,6,8-TetraCDD (3.2±0.2) x 10^-7 (3.9±0.4) x 10^-7
1,2,3,7-TetraCDD (4.3±0.1) x 10^-7 (12.7±0.8) x 10^-7
1,2,3,4,7-PentaCDD (1.2±0.1) x 10^-7 (4.6±0.1) x 10^-7
1,2,3,4,7,8-HexaCDD (4.4±0.1) x 10^-9 (19.0±0.1) x 10^-9
1,2,3,4,6,7,8-HeptaCDD (2.4±0.3) x 10^-9 (6.3±0.2) x 10^-9
OctaCDD (0.4±0.1) x 10^-9 (2.0±0.2) x 10^-9

^a From: Friesen et al. (1985).

Marple et al. (1986a) have reanalysed the water solubility of
2,3,7,8-TCDD and found it to be considerably less (12.5-19.2
ng/litre). The log water-octanol partition coefficient (K_ow) has
been determined as 6.64 by Marple et al. (1986b).

Friesen et al. (1985) have determined the water solubility for
some PCDDs other than the 2,3,7,8-TCDD compound and these are given in
Table 6.

Similarly Webster et al. (1985) have determined the log
octanol-water partition coefficients for a number of PCDDs (Table 7).

2,3,7,8-TetraCDD is considered to be a stable compound, but due
to its extreme toxicity its chemistry has not been fully evaluated.
However, it undergoes substitution reactions (Baughman, 1974) as well
as photochemical dechlorination (Crosby et al., 1971; Crosby & Wong,
1977; Gebefugi et al., 1977). Thermally it is very stable and rapid
decomposition of 2,3,7,8-tetraCDD occurs only at temperatures above
750 °C (Stehl et al., 1973). The other PCDDs have been much less
studied; however, octaCDD is completely destroyed by treatment with
hot alkali (Albro & Corbett, 1977).

The first synthesis of 2,3,7,8-tetraCDD was reported by
Sandermann et al. (1957), who used catalytic chlorination of the
unchlorinated dioxin. It has also been prepared in good yields by the
dimerization of 2,4,5-trichlorophenol salts (Buu-Hoi et al., 1971b;
Langer et al., 1973).

In the PCDF series, Mazer et al. (1983) synthesized all the 38
positional tetraCDF isomers. The products were mixtures of isomers,
and each of these isomers could be identified. Later Bell & Gari,
(1985) isolated and characterized all the 38 tetraCDFs, 28 pentaCDFs,
and 16 hexaCDFs.

Table 7. Values for log K_ow for some PCDDs from linear and quadratic plots

log K_ow (linear) log K_ow (quadratic)

Waters Waters Waters Waters
Bondapak Bondapak Bondapak Bondapak
Compound (Woodburn data) (Woodburn data)

Dibenzo-p-dioxin 4.26 4.01 4.34 4.17
1-MonoCDD 4.81 4.52 4.91 4.75
2-MonoCDD 5.33 5.00 5.45 5.29
2,7-DiCDD 6.27 5.86 6.39 6.17
1,2,4-TriCDD 7.36 6.86 7.45 7.11
1,2,3,7-TetraCDD 8.15 7.58 8.19 7.72
1,2,3,4-TetraCDD 8.63 8.02 8.64 8.07
1,3,6,8-TetraCDD 8.70 8.08 8.70 8.12
1,2,3,4,7-PentaCDD 9.48 8.80 9.40 8.64
1,2,3,4,7,8-HexaCDD 10.40 9.65 10.22 9.19
1,2,3,4,6,7,8-HeptaCDD 11.38 10.55 11.05 9.69
OctaCDD 12.26 11.35 11.76 10.07

a From: Webster et al. (1985).

Kuroki et al. (1984) have synthesized 51 congeners of PCDFs by a
structure specific method from chlorophenols and chloronitrobenzenes
or chlorophenols and chlorodiphenyls iodonium salts. The structures
were confirmed by MS and NMR.

Safe & Safe (1984) described the synthesis of 22 PCDF congeners
resulting in quantities of 10-320 mg of purified product. They also
reported NMR data on the compounds synthesized.

Sarna et al. (1984) and Burkhard & Kuehl, (1986) have documented
the octanol/water partition coefficients for some PCDFs (Table 8). The
disagreement for OCDF arises because of uncertainties in the K_ow
values of reference compounds of high K_ow. The partitioning of
organic chemicals between lipid and water is an important determinant
of the bioconcentration potential of a toxicant and has sometimes been
effectively used as an indicator of the preferred degradative in in
vivo pathways.

Table 8. The logarithm of the octanol/water partition coefficients (K_ow) of some PCDFs using HPLC methods

PCDF log K_ow Reference

2,8-dichloro- 5.95 Sarna et al. 1984^a
5.30^b Burkhard & Kuehl, 1986^c

2,3,7,8-tetrachloro- 5.82±0.02 Burkhard & Kuehl, 1986^c

octachloro- 13.37 Sarna et al. 1984^a
8.78 Burkhard & Kuehl, 1986^c

^a Quadratic equation treatment: Biorad Biosil (10 mm) data.
^b Quadratic equation treatment: unspecified "microbore" HPLC column.
^c Sarna et al. (1984) data recalculated from experimental data.

2.3. Analytical Methods

2.3.1 General aspects

The earliest reported method used to detect 2,3,7,8-tetraCDD was
a rabbit skin test (Adams et al., 1941). Test samples were applied to
the inner surface of the ear and to the shaven belly of albino
rabbits, and inflammatory responses were observed. Subsequently, Jones
& Krizek (1962) developed a test based on the recovery and weight of
the keratin formed on the rabbit ear after application of a sample.
These biological methods were non-specific as to isomers and not
sufficiently sensitive to detect low levels of contamination.

In the late 1960s and early 1970s, gas chromatographic methods
were used for the quantification mainly of 2,3,7,8-tetraCDD in
commercial 2,4,5-T formations. The detection level was normally in the
range of µg/g. These analyses were not isomer-specific and the results
could not be confirmed. Ryhage (1964) solved the problem of combining
a gas chromatograph with a mass spectrometer. During the 1970s and
1980s, various types of mass spectrometer and gas chromatograph/mass
spectrometer combinations were used in analytical work. Use of these
more sophisticated instruments allowed for the development of
isomer-specific and validated analyses for the tetraCDDs in the very
late 1970s and for the other PCDDs and PCDFs in the early 1980s.

A number of spectroscopic methods are available for the
laboratory identification of 2,3,7,8-tetraCDD, but their use is highly
restricted, with the exception of mass spectroscopy (MS). Data on
X-ray, infra-red (IR), ultra-violet (UV), nuclear magnetic resonance
(NMR), electron spin resonance (ESR), and mass spectra were obtained
by Pohland & Yang (1972), Baughman (1974), and Slonecker et al.
(1983).

Because of the large number of isomers and congeners, and due to
the extreme toxicity of some PCDD and PCDF isomers, highly sensitive
and specific analytical techniques are required for the measurements.
Detection limits for the analysis of environmental and human samples
should be orders of magnitude lower than the usual detection levels
required for pesticide analysis. A detection level of 1 pg or less
might be required to measure 2,3,7,8-tetraCDD and the other toxic
isomers in a 1-g environmental sample. Analyses at such low levels are
complicated by the presence of a multitude of other interfering
compounds and clean-up procedures are required.

The mono-, di-, and trichloro congeners are not usually included
in these analyses. Such compounds are considered to be much less toxic
than the higher chlorinated congeners and are also much more volatile
and losses may occur during clean-up.

It should be mentioned that the level of sophistication needed in
the analyses for PCDD and PCDFs will depend upon the objectives
thereof. In cases where the objectives were primarily to screen
samples to identify groups of PCDDs and/or PCDFs (in a qualitative or
semiquantitative manner), routine assays and bioassays were adequate.
In other instances, where the objective of the analysis was to
quantify accurately specific PCDD and/or PCDF isomers in the samples,
sophisticated analytical procedures were required. Clearly, both types
of analyses can be useful, depending on the purpose for which the
analytical results are to be used.

Many analytical methods have been developed in recent years for
the analysis of trace amounts of PCDDs and PCDFs in environmental
samples, especially for 2,3,7,8-tetraCDD. The most specific of these
methods are based on MS. There are many requirements to be met by such
an analytical method, including representative sampling and
appropriate storage, efficient extraction, high selectivity in the
clean-up, high specificity in the gas chromatography, high sensitivity
in the detection, safe and reliable quantification, good
reproducibility, useful confirmatory information.

Several review articles discussing methods of analyzing PCDDs and
PCDFs have appeared (McKinney, 1978; Esposito et al., 1980; Rappe &
Buser, 1980; Harless & Lewis, 1982; Karasek & Anuska, 1982; Tiernan,
1983; Crummett et al., 1985). Most of the older methods have been
critically reviewed by a panel of experts assembled by the National
Research Council of Canada (1981).

2.3.2 Sampling strategy and sampling methods

The quality and utility of analytical data depend on the validity
of the sample and the adequacy of the sampling program. The purpose of
sampling is to obtain specimens that represent the situation being
studied. Sampling plans may require that systematic samples be
obtained at specified times and places, or simple random sampling may
be necessary. Generally, the sample should be an unbiased
representation of the environmental situation.

All aspects of a sampling programme should be planned and
documented in detail, and the expected relationship of the sampling
protocol to the analytical result should be defined. A sampling
programme should include reasons for choosing sampling sites, the
number and type of samples, the timing of sample acquisition and the
sampling equipment used. A detailed sampling procedure should include
a description of the sampling situation, the sampling methodology,
labelling of samples, field blank preparation, pretreatment
procedures, and transportation and storage procedures.

The quality assurance programme should include means to
demonstrate that containers or storage procedures do not alter the
qualitative or quantitative composition of the sample. Special
transportation and storage procedures (refrigeration or exclusion of
light) should be described, if they are required.

Because environmental samples are typically heterogeneous, a
sufficiently large number of samples (ten or more) must normally be
analyzed to obtain meaningful data on chemical composition. The number
of individual samples that should be analyzed will depend on the kind
of information required by the investigation. If an average
compositional value is required, a number of randomly selected
individual samples may be obtained, combined, and blended to provide
a homogeneous composite sample from which a sufficient number of
subsamples could be analyzed. If composition profiles, time trends, or
the variability of the sample population are of interest, many samples
need to be collected and analyzed individually.

If field blanks are not available, efforts should be made to
obtain blank samples that best simulate a sample that does not contain
the specific chemical. In addition, measurements should be made to
ascertain whether, and to what extent, any reagent or solvent used may
contribute to or interfere with the analytical results (laboratory and
solvent blanks).

The recovery tests are frequently used and necessary to evaluate
the analytical methodology. Uncontaminated samples from control sites
that have been spiked with the chemical of interest provide the best
information because they simulate any matrix effect. When feasible,
isotopically labelled (¹³C, ³⁷Cl) chemicals spiked into the sample
provide the greatest accuracy since they are subjected to the same
matrix effects. The ¹³C- and ³⁷Cl-labelled compounds can be used
to validate:

(a) sampling (sampling surrogate),
(b) analytical pretreatment (clean-up surrogate),
(c) quantification (internal standard).

Very few laboratories in the world have access to and experience
in working with these complicated analyses.

In order to be able to compare data generated in different
laboratories, the same quantitative standard compounds should be used.
Interlaboratory calibrations or "round-robin" studies have been
performed in very few cases.

2.3.3 Extraction procedures

In this step, the sample is homogenized or digested and extracted
with a suitable solvent or solvent mixture to remove the bulk of the
sample matrix and transfer the PCDD and PCDF residue into the solvent.
Both the selection of the proper solvent and the method of extraction
can be critical in obtaining a satisfactory recovery of PCDDs and
PCDFs from the sample matrix.

Many different procedures for the extraction of PCDDs/PCDFs from
various samples are described. In some cases this involves digestion
or destruction of the matrix. Some of these methods have been
evaluated in the report from the National Research Council of Canada
(1982), while other methods are discussed by Tiernan (1983).

An interlaboratory "round-robin" study involving 13 laboratories
was carried out to evaluate the reliability of data on
2,3,7,8-tetraCDD in fish samples. No significant differences were
found from methods differing in the digestion or extraction procedures
(Ryan et al., 1983b).

In a study described by Albro et al. (1985), eight different
approaches were applied in eight laboratories to quantify four PCDDs
(2,3,7,8-tetraCDD; 1,2,3,7,8-pentaCDD; 1,2,3,4,7,8-hexaCDD; and
octaCDD) and three PCDFs (2,3,7,8-tetraCDF; 2,3,4,7,8-pentaCDF; and
1,2,3,7,8,9-hexaCDF) in spiked samples of an extract from human
adipose tissue. Levels of fortification, unknown to the participating
laboratories, were in the 5-50 ng/kg range, except for octaCDD (up to

500 ng/kg). The results indicated that most of the procedures tested
gave a high degree of qualitative reliability. However, other methods
were not so accurate, a large portion of the reported data consisting
of false positives or false negatives.

Lustenhouwer et al. (1980) studied the extraction of PCDDs and
PCDFs from a fly ash sample. A dramatic difference was found between
different solvents.

2.3.4 Sample clean-up

In the sample clean-up, the PCDDs and PCDFs present in the sample
should be separated from a multitude of other co-extracted and
possibly interfering compounds. The clean-up methods, normally three
steps or more, can vary for different sample matrices. Two different
procedural trends can be recognized:

(a) all PCDD and PCDF isomers can be analyzed in one
single fraction by the containment enrichment
procedure (Norstrom et al., 1982; Stalling et al.,
1983; Tiernan, 1983; Rappe, 1984),

(b) specific isomers are analyzed in different fractions
mainly after normal-phase and reverse-phase high
pressure liquid chromatography (HPLC) separation
(Lamparski et al., 1979; Niemann et al., 1983; Tosine
et al., 1983).

This latter method allows the identification of only a few PCDD
isomers in each fraction, and is mainly used to monitor TCDD and a few
other congeners. For a monitoring program of all PCDDs and PCDFs a
more general method might be preferred.

The method described by Stalling et al. (1983) was originally
designed for the analyses of fish samples. In a "round-robin" study of
fish samples it gave good results (Ryan et al., 1983b). This method
has now been used for the clean-up of other biological samples like
bird muscle, seal fat, turtle fat, and human adipose tissue - blood,
liver, kidney, and milk (Rappe et al., 1983c; Nygren et al., 1986;
Rappe et al., 1986b).

2.3.5 Isomer identification

The purified extracts are used directly for the final analyses
with the aid of a gas-chromatograph/mass spectrometer (GC/MS) equipped
with a glass capillary or a fused-silica column. The column leads
directly into the ion source of the mass spectrometer, which operates
either in the electron impact (EI) or the negative ion-chemical
ionization (NCI) mode. In view of the large variation in toxicological

and biological effects of the PCDD and PCDF isomers, it is imperative
that the isomers, particularly those having high toxicity, be
identified. For an unambiguous isomer identification it is necessary
to have access to all analytical standards within a specific group of
isomers, e.g. all the 22 tetraCDDs and all the 38 tetraCDFs. All the
22 tetraCDDs have been prepared and, using a Silar 10c glass capillary
column, the 2,3,7,8-tetraCDD can be separated from all the other 21
tetra isomers (Buser & Rappe, 1980). Recently all the 14 pentaCDDs and
the 10 hexaCDDs have been prepared. Using the Silar 10c column all the
2,3,7,8- substituted isomers can be separated from all the other
isomers (Buser & Rappe, 1984). The SP 2330 fused silica column can
also be used for this separation (Rappe, 1984).

In the PCDF series, Mazer et al. (1983) have synthesized all the
38 positional tetraCDF isomers. The products were mixtures of isomers,
and each of these isomers could be identified using both an SP 2330
and an SE 54 capillary column. Later, Bell & Gara (1985) isolated and
characterized all tetra-, penta- and hexaCDFs. The SP 2330 column can
separate most of these isomers (Rappe, 1984). The 1,2,3,7,8-pentaCDF
co-elutes with the 1,2,3,4,8-isomer and the 1,2,3,4,7,8- hexaCDF with
the 1,2,3,4,7,9-isomer, but they can be separated on less polar
columns like OV-17 and DB-5.

A very limited number of investigations has been performed using
these complete sets of synthetic standards.

2.3.6 Quantification

Mass selective detection (mass fragmentography) has been used to
quantify trace amounts of PCDDs and PCDFs in the samples by
selectively monitoring M, M + 2, and/or M + 4 ions (SIM). The
quantification is based on peak area measurements and a comparison of
these areas using either isotopically labelled internal standards
(¹³C or ³⁷Cl) or calibration curves of external standards. As a
first approach, it has been generally assumed that with the MS
quantification technique, all isomers of a particular congener of PCDD
or PCDF (e.g. the tetrachloro-isomers) have the same response factors.
However, an investigation of 13 well-defined tetraCDF isomers has
shown a three-fold variation in response factors with the EI mode and
up to a 20-fold variation with the negative ion-chemical ionization
mode. For the higher chlorinated homologues (penta, hexa) the
variation was found to be less (Rappe et al., 1983b).

Fung et al. (1985) have studied the mass spectra of 26 PCDF
congeners. They found that the EI spectra are not particularly isomer
specific, while positive ion-chemical ionization spectra show a
greater degree of isomer distinction.

2.3.7 Confirmation

Quality control and quality assurance programs help to assure
that positive data reported actually refer to specific PCDDs and PCDFs
(Kloepfer et al., 1983). To provide reliable data:

(a) isomer specificity must be demonstrated initially and verified
daily,
(b) the retention time must equal (within 3 seconds) the retention
time for the isotopically labelled congener,
(c) the signal to noise ratio must be 2.5:1 or higher,
(d) the chlorine cluster must be within ± 10% of the theoretical
values, given in Table 9,
(e) correct fragments, e.g., M⁺-COCl ions, must be with correct
chlorine clusters.

For confirmation, mass spectroscopy is the best technique now
available. The EI mass spectral properties of PCDFs and PCDD have been
described (Buser, 1975). The molecular (M⁺) and fragment ions of
PCDDs and PCDFs show the typical, expected clustering due to the
chlorine isotopes (Table 9). The typical fragmentation is M^-COCl⁺,
which is a useful fragment to study.

Buser & Rappe (1978) have shown that observation of low mass ions
can be used for the identification of the substitution pattern of
PCDDs, which can be defined as the number of chlorine atoms on each
carbon ring of the dioxin molecular; the 2,3,7,8-isomer has a 2:2
pattern while 1,2,3,4-tetraCDD has a 4:0 pattern. However, these low
mass ions may not be observed in spectra from environmental or
biological samples.

In the negative ion-chemical ionization mode, the PCDFs have the
base peak due to M^-, and the fragmentation produces the unusual
M^--34 ions (uptake of H and loss of Cl). Fragmentation of PCDDs in
this mode is more conventional via loss of Cl yielding M^--35 ions
(Buser et al., 1985).

Using EI technique and a quadropole instrument, the detection
limits are 1-10 pg for the tetrachloro compounds and up to 10-50 pg
for the octachloro compounds using selected ion monitoring or multiple
ion detection (SIM or MID). Full mass spectra require 0.1-1 ng of
compound (Buser et al., 1985). High resolution instruments can improve
the sensitivity by one order of magnitude.

The negative ion-chemical ionization mode, using methane gas as
reagent, gas provides extremely good sensitivity for all PCDFs (tetra-
to octachloro- compounds) and for the higher chlorinated PCDDs (penta-
to octaCDD). The detection limits are in the 10-100 fg (10^-15g)
range using SIR or MID, which is 1 to 2 orders of magnitude better
than EI (Buser et al., 1985). However, the negative ion-chemical
ionization mode has very poor sensitivity for 2,3,7,8- tetraCDD under
these conditions.

Using low resolution MS instruments, a series of interfering
compounds has been identified (Table 10). Some of this interference
can be eliminated using high resolution MS instruments operating at
8000 - 10 000 daltons. However, compounds with the same empirical
formulae cannot be separated by MS technique; they are normally
eliminated during the clean-up or separated by the gas chromatography
step.

2.3.8 Other analytical methods

Paasivirta et al. (1977) have shown that 2,3,7,8-tetraCDD can be
detected down to the pg level using a glass capillary column and a
⁶³Ni electron-capture detector. Combined with efficient clean-up
procedures, this method has shown to be useful down to a level of 9
ppt (Niemann et al., 1983), although positive samples need
confirmation by mass spectroscopy (MID, SIM).

Other techniques, such as enzyme induction and radioimmunoassay
have been described and discussed by Firestone (1978) and McKinney
(1978). McKinney et al. (1982) have used the radioimmunoassay method
for determining 2,3,7,8-tetraCDD in human fat, and found the reliable
sensitivity at 95% confidence interval to be 100 pg per sample.

An analytical method based on the keratonization response of
epithelial cells in an in vitro system has been described by
Gierthy & Crane (1985b). This method can be an assay for dioxin-like
activity in environmental and biological samples. A positive response
was found for 2,3,7,8-tetraCDD at a concentration of 10-11 mol/litre.

Table 9. Isotopic abundance ratio ("cluster") of polychlorinated dioxins and dibenzofurans

Number of
chlorine M M + 2 M + 4 M + 6 M + 8 M + 10 M + 12 M + 14
atoms

1 100.0 33.7
2 100.0 66.1 11.3
3 100.0 98.4 32.7 3.8
4 76.4 100.0 49.4 11.0 1.0
5 61.2 100.0 65.5 21.6 3.6 0.3
6 51.1 100.0 81.7 35.8 8.9 1.2 0.1
7 43.8 100.0 97.9 53.4 17.6 3.5 0.4
8 33.7 87.6 100.0 65.3 26.8 7.0 1.2 0.1

Table 10. List of molecular ions of polychlorinated compounds present in some human and environmental samples
and possibly interfering in the mass spectral analysis of PCDFs and PCDDs^a

Molecular ions (m/z,m⁺,m^-) (chlorination)
Compounds mono- di- tri- tetra- penta- hexa- hepta- octa- nona- deca-

PCDDs 320 354 388 422 456 - -
PCDFs 304 338 372 406 440 - -

PCBs 290 324 358 392 426 460 494
PCNs 264 298 332 366 400 - -
PCTs 298 332 366 400 434 468 502 536 570
PCDPEs^b 238 272 306 340 374 408 442 476 510
PCPYs^c 36 270 304 338 372 406

^a From: Buser et al. (1985).
^b PCDPEs: Polychlorinated diphenylethers.
^c PCPYs: Polychlorinated pyrenes.

3. SOURCES OF ENVIRONMENTAL POLLUTION

3.1 Production, Synthesis, and Use

PCDDs and PCDFs are not produced commercially. These compounds
are in fact formed as trace amounts of undesired impurities in the
manufacture of other chemicals such as chlorinated phenols and their
derivatives, chlorinated diphenyl ethers, and polychlorinated
biphenyls (PCBs). There is no known technical use for the PCDDs and
PCDFs.

The amount of total PCDDs entering the Canadian environment/year
has been estimated to be about 1500 kg, and 75% of this amount has
been estimated to be due to octaCDD alone (National Research Council
of Canada, 1981). There is no estimation of the amount of PCDFs
entering the environment anywhere in the world.

Although the polychlorinated dioxins and dibenzofurans are not
commercially produced, most of these compounds have been synthesized
for research purposes in small quantities according to the reactions
discussed in section 2.

3.2 Industrial Processes

In addition to the synthetic methods mentioned in section 2,
2,3,7,8-tetraCDD may be formed during the industrial preparation of
2,4,5-trichlorophenol from 1,2,4,5-tetra-chlorobenzene. This
substitution reaction takes place at about 180 °C, and when the
solvent is methanol, the pressure rises to about 7 KPa. The formation
of TCDD is an unwanted side reaction which takes place when the
reaction mixture is heated to 230-260 °C (Milnes, 1971). This reaction
is exothermic, so that even higher temperatures may be attained
resulting in uncontrolled conditions.

In some factories ethylene glycol is used as a solvent in order
to avoid the high pressure. As already pointed out by Milnes (1971),
however, use of this solvent requires special precautions because of
the occurrence of a base-promoted polymerization of ethylene glycol
and decomposition reactions that produce ethylene oxide. These
reactions are also exothermic; they may start spontaneously at
temperatures above 180 °C and proceed rapidly and uncontrollably to
result in the formation of relatively large amounts of TCDD.

After most of the solvent has been removed, the reaction mixture
is acidified; the 2,4,5-trichlorophenol can be separated from
2,3,7,8-tetraCDD by one or two distillations, with the result that
2,3,7,8-tetraCDD is concentrated in the still-bottom residues. Up to
1 mg/g of 2,3,7,8-tetraCDD in such residues has been reported
(Kimbrough et al., 1984). Improper disposal of such residues is
discussed in sections 4.4.2 and 9.

Most of the 2,4,5-trichlorophenol produced is used for the
preparation of herbicides such as 2,4,5-T (including various esters
and salts, and the bactericide hexachlorophene).

PCDDs and PCDFs are both formed as by-products during the
manufacture of chlorinated phenols (2,4-dichloro-, 2,4,6-trichloro-,
2,3,4,6-tetrachloro- and pentachlorophenol). The commercial
chlorophenols are produced by two processes, i.e., by chlorination of
the phenol using various catalysts and by the alkaline hydrolysis of
an appropriate chlorobenzene. Apparently both reactions can lead to
the formation of PCDDs as well as PCDFs, and the level of
contamination is normally much higher here than in the production of
2,4,5-trichloro-phenol (see section 3.3).

PCDDs and PCDFs are also formed during the preparation of
chlorinated diphenyl ether herbicides (Yamagishi et al., 1981) and
hexachlorobenzene (Villeneuve et al., 1974). A series of PCDFs are
formed during the production of PCBs (see section 3.3).

Production equipment is often used for the production of several
different chemicals. In the manufacture of chemicals on such equipment
previously contaminated by PCDDs and PCDFs, both the products and
waste generated can be contaminated. Thus, manufactured
2,4-dichlorophenoxyacetic esters (2,4-D), which otherwise should not
be contaminated by 2,3,7,8-tetraCDD, did indeed contain this dioxin
because the equipment used had been employed previously to produce
2,4,5-T and had not been cleaned properly (Federal Register, 1980).

It should be pointed out that the primary occurrence of TCDD in
the environment is possibly related to the synthesis of
2,4,5-trichlorophenol, the use of products prepared from this compound
(Table 11), and to incinerations reactions. The occurrence of the
other PCDDs and PCDFs is related to the synthesis and use of a variety
of other products (Table 12), some of which are quite common.

The other PCDDs and PCDFs are also formed in a variety of
incineration reactions (see section 4.5).

3.3 Contamination of Commercial Products

3.3.1 Chlorophenoxyacetic acid herbicides

Depending on the temperature control and purification efficiency,
the levels of 2,3,7,8-tetraCDD in commercial products may vary
greatly. For example, the levels of 2,3,7,8-tetraCDD in drums of the
herbicide Agent Orange placed in storage in the USA and in the Pacific

before 1970 were between 0.02 and 47 mg/g. More than 450 samples were
analyzed in this study, and the mean value was 1.98 mg/g (Young et
al., 1983). Since Agent Orange was formulated as a 1:1 mixture of the
butyl esters of 2,4,5-T and 2,4-D, the levels of 2,3,7,8-tetraCDD in
individual 2,4,5-T preparations manufactured and used in the 1960s
could have been as high as 100 mg/g.

In analyses using high-resolution GC-MS, Rappe et al. (1978a)
have reported that in other samples of Agent Orange (as well as in
European and the USA 2,4,5-T formulations from the 1950s and 1960s),
2,3,7,8-tetraCDD was the dominating compound of this group of
contaminants. Only minor amounts of other PCDDs and PCDFs could be
found, primarily lower chlorinated PCDDs, in samples of Agent Orange.

As a result of governmental regulations, efforts were made during
the 1970s to minimize the formation of 2,3,7,8-tetraCDD during 2,4,5-T
production, and now all producers claim that their products contain
less than 0.1 µg 2,3,7,8-tetraCDD/g of product (Rappe et al., 1978a).
At present, the chloro-phenoxy herbicides are not the major source of
PCDDs and PCDFs in the environment.

Sixteen samples of 2,4-D esters and amine salts from Canada were
analyzed for the presence of PCDDs. Eight out of nine esters and four
out of seven amine salts were found to be contaminated, with the
esters showing significantly higher levels (210-1752 ng/g) than the
salts (20-278 ng/g). The tetraCDD observed was the 1,3,6,8-isomer, as
verified by a synthetically prepared authentic standard (Cochrane et
al., 1981). In other studies, it has been found that no tetraCDD other
than the 1,3,6,8-isomer elutes in this window. Hagenmaier et al.
(1986) has reported that, unexpectedly, a German 2,4-D formulation
contained 6.8 ng of 2,3,7,8-tetraCDD/g.

Table 11. Some commercial products that may be contaminated with
2,3,7,8-tetraCDD, depending on the method of preparation

Common name Chemical name

2,4,5-T^a 2,4,5-Trichlorophenoxyacetic acid

2,4,5-T esters^a n-butyl-, butoxy ethyl-, and
iso-octyl-esters of 2,4,5-
trichlorophenoxyacetic acid

2,4,5-T salts^a dimethylamine salts of 2,4,5-
trichlorophenoxyacetic acid

Fenoprop esters of 2-(2,4,5-trichlorophenoxy)-
propanoic acid

Erbon ethyl ester of 2-(2,4,5-trichloro-
phenoxy)-2,2-dichloropropanoic acid

2,4,5-Trichlorophenol 2,4,5-Trichlorophenol
Fenochlorphos O,O-Dimethyl O-2,4,5-trichlorophenyl
phosphonothioate

Trichloronate O-Ethyl 0-2,4,5-trichlorophenyl
ethylphosphonothioate

Hexachlorophene/isobac 20 2,2'-Methylene-bis (3,4,6-trichloro-
phenol)

^a There are numerous trade names for this product.

Table 12. Some commercial products which may be contaminated with PCDDs
other than 2,3,7,8-tetraCDD, and with PCDFs, depending on the method of
preparation

Common name Chemical name

Bifenox Methyl-5-2,4-dichlorophenoxy-2-nitrobenzoate

Chloranil 2,3,5,6-Tetrachloro-2,
5-cyclo-hexadiene-1,4-dione.

2,4-D (esters and salts) 2,4-Dichlorophenoxyacetic acid
and esters and salts

2,4-DB and salts 2,4-Dichlorophenoxybutyric acid and
salts

Dicamba 3,6-Dichloro-2-methoxybenzoic acid

Dicamba, dimethylamine salt 3,6-Dichloro-2-methoxybenzoic acid,
dimethylamine salt

Dicapthon Phosphorothioic acid
o-(2-chloro-4-nitrophenyl)
o,o-dimethyl ester

Dichlofenthion Phosphorothioic acid
o-2,4-dichloro-phenyl
o,o-dialkyl ester

Disul sodium (sesone) 2,4-Dichlorophenoxyethyl sulfate,
sodium salt

2,4-DP 2- 2,4-Dichlorophenoxy propionic acid

HCB Hexachlorobenzene

Nitrofen 2,4-Dichlorophenyl-p-nitrophenyl
ether

PCP and salts Pentachlorophenol and salts

PCB Polychlorinated biphenyls

2,4,6-TCP 2,4,6-Trichlorophenol and salts
2,3,4,6-Tetrachlorophenol and salts

Common name Chemical name

CNP 1,3,5-Trichloro-2-(4-nitrophenoxy)
benzene

NIP 2,4-Dichloro-1-(4-nitrophenoxy)
benzene

X-52 2,4-Dichloro-1-(3-methoxy-4-nitro-
phenoxy) benzene

3.3.2 Hexachlorophene

The bactericide hexachlorophene is prepared from
2,4,5-trichlorophenol, also the key intermediate in the production of
2,4,5-T. Due to additional purification, the level of 2,3,7,8-tetraCDD
in this product is usually < 0.03 mg/kg (Baughman, 1974). Ligon & May
(1986) reported 0.0047 mg/kg of TCDD in one hexachlorophene sample.
However, hexachlorophene also contains about 100 mg/kg of a
hexachloroxanthene, the 1,2,4,6,8,9-substituted isomer (Göthe &
Wachtmeister, 1972).

3.3.3 Chlorophenols

Chlorophenols have been used extensively since the 1950s as
insecticides, fungicides, mold inhibitors, antiseptics, and
disinfectants. In 1978 the annual world production was estimated to be
approximately 200 000 tons. The most important use of 2,4,6-tri-,
2,3,4,6-tetra-, and pentachlorophenol, and their salts, is for wood
preservation. Pentachlorophenol is also used as a fungicide for slime
control in the manufacture of paper pulp and for a variety of other
purposes such as in cutting oils and fluids, for tanning leather, and
in paint, glues, and outdoor textiles. 2,4-Di- and
2,4,5-trichloro-phenol are used for the production of 2,4-D and
2,4,5-T herbicides (phenoxy acids), and 2,4,5-trichlorophenol for the
production of hexachlorophene.

Chlorophenols are produced industrially either by direct
chlorination of phenol or by hydrolysis of chlorobenzenes, the actual
process used depending on the isomer desired. Chlorination of phenol
yields 2,4-di-, 2,4,6-tri-, 2,3,4,6-tetra-, or pentachlorophenol,
while hydrolysis of chlorobenzenes is mainly used for the production
of 2,4,5-tri- and pentachlorophenol (Nilsson et al., 1978).
Chlorophenols may contain a variety of by-products and contaminants,
such as other chlorophenols, polychlorinated phenoxyphenols, and
neutral compounds like polychlorinated benzene and diphenyl ethers

(PCDPEs), PCDDs, and PCDFs. Some of these contaminants may also occur
in chlorophenol derivatives like phenoxy acids, other pesticides, and
hexachlorophene. The possible presence of PCDDs and PCDFs in
commercial products is of special significance because of their
extraordinary persistence and toxicological properties (see sections
7-9). A scientific criteria document for chlorophenols and their
impurities in the Canadian environment has been prepared by Jones
(1981, 1984). Chlorophenols were estimated to be the major chemical
sources of PCDDs and PCDFs in the Canadian environment (Sheffield,
1985).

Buser & Bosshardt (1976) reported on the results of a survey of
the PCDD and PCDF contents of pentachlorophenol (PCP) and PCP-Na from
commercial sources in Switzerland. From the results, a grouping of the
samples into two series can be observed: a first series with generally
low levels (hexaCDD <1 µg/g) and a second series with much higher
levels (hexaCDD >1 µg/g) of PCDDs and PCDFs. Samples with high PCDD
values had also high PCDF values. For most samples, the contents of
the PCDF contaminants were in the order:

tetra- = penta- < hexa- < hepta- < octaCDD/CDF.

The ranges of the combined levels of PCDDs and PCDFs were 2-16 and
1-26 µg/g, respectively, for the first series of samples, and 120-500
and 85-570 µg/g, respectively, for the second series of samples. The
levels of octaCDD and octaCDF were as high as 370 and 300 µg/g,
respectively.

Some PCP-Na samples analyzed showed the unexpected presence of a
tetraCDD (0.05-0.25 µg/g), which was later identified by Buser & Rappe
(1978) as the unusual 1,2,3,4-substituted isomer. Table 13 collects a
number of relevant analyses of these chlorophenol formulations. The
levels of PCDDs and PCDFs are higher than for the phenoxy-acetic acid
herbicides.

It has also been reported that several positional isomers of
PCDDs and PCDFs are present in the chlorophenols. However,
isomer-specific methods have not been used in most of these
investigations, and more research is necessary to identify all the
isomers present for a risk evaluation of these products.

Miles et al. (1985) have analyzed PCP samples for hexaCDDs from
five different manufacturers using an isomer-specific analytical
method. The study included both free PCPs as well as the sodium salts.
Total hexaCDDs in PCPs ranged from 0.66 to 38.5 mg/kg, while in the
sodium salts levels of hexaCDDs between 1.55 and 16.3 mg/kg were
found. The most abundant hexaCDD isomer found in the free PCPs was the
1,2,3,6,7,8 isomer; however, in the sodium salts the 1,2,3,6,7,9- and
1,2,3,6,8,9-hexaCDD pair was the most abundant.

Table 13. Levels of PCDDs and PCDFs in commercial chlorophenols (µg/g)^a

2,4,6- 2,3,4,6- PCP PCP
Trichlorophenol Tetrachlorophenol Sample A Sample B

TetraCDDs < 0.1 < 0.1 < 0.1 < 0.1
PentaCDDs < 0.1 < 0.1 < 0.1 < 0.1
HexaCDDs < 1 < 1 < 1 2.5
HeptaCDDs < 1 10 0.5 175
OctaCDD < 1 2 4.3 500
TetraCDFs 1.5 0.5 < 0.1 < 0.1
PentaCDFs 17.5 10 < 0.1 < 0.1
HexaCDFs 36 70 0.03 < 0.3
HeptaCDFs 4.8 70 0.5 19
OctaCDF < 1 10 1.1 25

^a From: Rappe et al. (1979).

Hagenmaier & Brunner (1987) has reported that 2,3,7,8-tetraCDD
can be found in commercial pentachlorphenol formulation at levels of
0.21-0.56 ng/g, while Hagenmeyer & Brunner (1986) report that
1,2,3,7,8-pentaCDD was found in pentachlorophenol and
Na-pentachlorophenates in concentrations of 0.9-18 ng/g.

3.3.4 Polychlorinated biphenyls (PCBs)

Vos et al. (1970) were able to identify PCDFs (tetra- and
pentaCDFs) in samples of European PCBs (Phenoclor DP-6 and Clophen A
60) but not in a sample of Aroclor 1260. The toxic effects of these
PCB products were found to parallel the levels of PCDFs present. Bowes
et al. (1975) examined a series of Aroclors, as well as the samples of
Aroclor 1260, Phenoclor DP-6, and Clophen A-60 previously analyzed by
Vos et al. (1970). They used packed columns and very few standard
compounds, and reported that the most abundant PCDFs had the same
retention time as 2,3,7,8-tetraCDF and 2,3,4,7,8-pentaCDF. Using a
complete set of PCDF standards and an isomer-specific analytical
method, Rappe et al. (1985d) determined the levels of
2,3,7,8-substituted PCDFs in commercial PCB products (see Table 14).

3.3.5 Chlorodiphenyl ether herbicides

In 1981, Yamagishi et al. reported on the occurrence of PCDDs and
PCDFs in the commercial diphenyl ether herbicides
1,3,5-trichloro-2-(4-nitrophenoxy) benzene (CNP),
2,4-di-chloro-1-(4-nitrophenoxy)benzene (NIP), and
2,4-dichloro-1-(3-methoxy-4-nitrophenoxy)benzene (X-52). The total
tetraCDD found was 14.0 mg/kg in CNP, 0.38 mg/kg in NIP, and 0.03 in
X-52. Very few synthetic standards were used, but the major tetraCDDs
were identified as the 1,3,6,8- and 1,3,7,9-isomers, the expected
impurities in the starting material 2,4,6-trichlorophenol. No
2,3,7,8-tetraCDD could be detected in these samples. In all three
herbicides, total tetraCDF was between 0.3 and 0.4 mg/kg.

3.3.6 Hexachlorobenzene

Hexachlorobenzene was used for the control of wheat bunt and
fungi. Villeneuve et al. (1974), analyzing three commercial
hexachlorobenzene preparations, identified octaCDD and hepta- and
octaCDFs. The levels and identity of the heptaCDF isomers were not
given. Great variation in levels of octaCDDs between the three samples
(0.05-211.9 mg/kg) was noted, as well as in the level of octaCDF
(0.35-58.3 mg/kg).

3.3.7 Rice oil

In 1968 more than 1500 people in southwest Japan were intoxicated
by the consumption of a commercial rice oil accidentally contaminated
by PCBs, PCDFs, and polychlorinated quarterphenyls (Masuda &
Yoshimura, 1982; Masuda et al., 1985). In 1979 a similar episode
occurred in central Taiwan, the number of people involved here
approaching 2000 (Chen et al., 1980, 1981). Both these accidents have
been referred to as Yusho episodes, but now the Taiwan episode has
been renamed Yu-cheng (see section 5.4.4.4).

The total level of PCDFs in the Japanese rice oil was reported to
be 5 µg/g (Nagayama et al., 1976) and 5.6 µg/g (Buser et al., 1978d).
For the rice oil from Taiwan, Chen et al. (1985) reported the PCDFs
levels to be in the range 0.18-1.68 µg/g.

Buser et al. (1978) analyzed the Japanese rice oil using glass
capillary columns. They found about 50-60 PCDF congeners and also
reported that the 2,3,7,8-tetraCDF was the major isomer among the
tetraCDFs. However, it was later shown that in this column system the
2,3,4,8-tetraCDF co-elutes with the 2,3,7,8-isomer, and in fact the
2,3,4,8-isomer was the main constituent in this peak (Chen & Hites,
1983; Masuda et al., 1985). The 2,3,7,8-substituted congeners were
estimated to account for 10-15% of the total amount of PCDFs (Buser et
al., 1978).

Table 14. PCDFs in commercial PCBs (ng/g)^a

TRI- TETRA- PENTA- HEXA- HEPTA-

Total 2378 Total 12348 23478 Total 123479 123678 123789 234678 Total Total
PCB-type 12378 123478

Pyralene 700 53 630 10 T 35 ND ND ND ND ND ND
A1254 63 19 1400 690 490 4000 2500 2100 190 130 10 000 960
A1260 10 13 110 48 56 260 500 120 190 27 1500 1300
A30 500 35 573 14 28 160 50 59 ND ND 220 T
A40 1300 180 2600 96 8 1700 79 68 ND T 310 ND
A50 7400 3300 20 000 760 1100 8000 700 360 18 98 3100 75
A60 770 840 6900 1100 990 8100 1600 330 170 330 6800 2000
T64 47 23 360 97 122 840 520 390 58 41 2600 220
Clophen C 710 54 1200 34 30 270 ND T ND ND T ND

^a From: Rappe et al. (1985d).
T = traces.
ND = not detected.

3.4 Sources of Heavy Environmental Pollution

3.4.1 Industrial accidents

Several industrial accidents occurring during the production of
2,4,5-trichlorophenol have been described in the literature. In most
of these accidents the pollution of 2,3,7,8-tetraCDD has been to
factories with circumscribed occupational exposure (section 9).
However, on 10 July, 1976, a runaway reaction in a factory at Meda
near Seveso in Northern Italy resulted in the escape of a chemical
cloud of trichlorophenol/phenate containing 2,3,7,8-tetraCDD.

The cloud initially covered an area outside the factory 5 km long
and 700 m wide. On the basis of the TCDD levels found in the
contaminated soil samples, it has been estimated that 2-3 kg of TCDD
was released in this accident. About 80% of this amount was deposited
in an area of 15 ha, within a distance of about 500 m from the plant.
The levels of soil contamination in three zones are given in Table 15
(Pocchiari, 1978).

3.4.2 Improper disposal of industrial waste

In 1973, three horse arenas in Missouri, USA, were found to be
contaminated by high levels of 2,3,7,8-tetraCDD; the highest value
reported was about 30 µg/g of soil (Kimbrough et al., 1977). This
contamination resulted from the application, in 1971, of contaminated
waste oil to control dust at these locations. The TCDD had originated
at a hexachlorophene-producing factory in Verona, Missouri. Additional
tri- and tetraCDDs were also found, but the major component was
1,2,4,6,8,9-hexachloroxanthene, a compound which apparently can serve
as a marker for this type of contamination. The xanthene is a normal
by-product of hexachlorophene production and has never been associated
with the production of 2,4,5-tri-chlorophenol or 2,4,5-T derivatives
(Viswanathan & Kloepfer, 1986).

In 1982, numerous sites of potential 2,3,7,8-tetraCDD
contamination were discovered in eastern Missouri. The contamination
originated from the same waste oil from the factory in Verona. The
streets of the entire town of Times Beach, Missouri, had been sprayed.
More than 10 000 soil samples from Missouri were analyzed. In this
state more than 40 hazardous waste sites containing 2,3,7,8-tetraCDD
were identified. Most of these contaminated sites resulted from the
disposal of waste from the same factory in Verona. The highest level
reported in these soil samples was 9648 mg TCDD/g (Viswanathan &
Kloepfer, 1986).

Another location of great concern is Love Canal, Niagara Falls,
USA. Here, Smith et al. (1983) found high levels of 2,3,7,8-tetraCDD
in storm sewer sediments taken from around the Love Canal waste
disposal site. The highest value was 312 ng/g sediment.

Table 15. Distribution of TCDD contamination in the Seveso area on the basis of soil sample analyses^a

Range Number of soil samples
(µg/m²)

Zone A Zone B Surrounding monitored area

< 0.750 32 25 249
0.750 - 4.99 32 53 128
5.0 - 14.99 6 19 2
15.0 - 49.99 18 6 0
50.0 - 499.99 31 0 0
500.0 - 4999.99 18 0 0
> 5000 3 0 0

^a From: Pocchiari (1978).
Zone A: high-level contamination, about 115 ha.
Zone B: low level contamination, about 255 ha.
Surrounding area: about 1400 ha.

3.4.3 Heavy use of chemicals

The Eglin Air Force Base in Northwest Florida, USA, has been used
for the development and testing of aerial spraying equipment for
military defoliation operations. During the period 1962-1970, a
3-km² test area was sprayed with 73 tons of 2,4,5-T. Analyses of
archived samples of the formulations indicated that approximately 2.8
kg of 2,3,7,8-tetraCDD had been applied as a contaminant of the
herbicide. However, one 37-ha test grid received 2.6 kg of this TCDD
from 1962 to 1964. Levels of 10-1500 ng/kg were found in 22 soil
samples (the top 15 cm) collected and analyzed 14 years after the last
application of herbicide to this site (Young, 1983).

3.5 Other Sources of PCDDs and PCDFs in the Environment

3.5.1 Thermal degradation of technical products

The formation of 2,3,7,8-tetraCDD as a result of thermal
reactions of 2,4,5-T and 2,4,5-T derivatives has been the subject of
controversy. Heating 2,4,5-T salts at 400-450 °C for 30 minutes or
longer yielded approximately 1 g of 2,3,7,8-tetraCDD per kg of 2,4,5-T
salt, while no TCDD was identified from the same treatment of 2,4,5-T
acid or esters (Langer et al., 1973; Baughman, 1974). Using a more
sensitive analytical method, Ahling et al. (1977) reported that 0.2-3
mg of 2,3,7,8-tetraCDD was formed per kg of 2,4,5-T esters during

combustion at 500-850 °C. Two reports (Stehl & Lamparski, 1977;
Andersson et al., 1978) have shown that 2,3,7,8-tetraCDD could not be
found after burning samples of spiked or sprayed vegetation at 600 °C.
The combustion gases, soot, particles, and ashes were analyzed and the
detection limit was 4 mg of TCDD/kg 2,4,5-T burned.

Rappe (1978b) have studied the burning of material impregnated
with various salts of chlorophenols. Very carefully purified
2,4,6-tri- and pentachlorophenate were studied, in addition to a
commercial formulation of 2,3,4,6-tetra-chlorophenate. The analytical
method used in this study was not isomer specific, but the following
conclusions can be drawn concerning the formation of PCDDs by thermal
reactions:

(a) the expected dimerization products and the products formed
in the "Smiles rearrangement" are the major PCDDs;
(b) no other thermal isomerization of the PCDDs formed can be
observed;
(c) no formation of higher chlorinated PCDDs can be observed;
(d) octaCDD and other higher chlorinated PCDDs yield lower
chlorinated dioxins in a nonspecific dechlorination
reaction;
(e) a series of PCDFs was also observed.

It has been found that PCBs can be converted to PCDFs under
pyrolytic conditions. The pyrolysis of commercial PCBs in sealed
quartz ampoules in the presence of air yielded about 30 major, and
more than 30 minor, PCDFs. The optimal yield of PCDFs was about 10%,
calculated on the amount of PCB decomposed. Thus, uncontrolled burning
of PCBs can be an important environmental source of hazardous PCDFs.
Therefore, it was recommended (Buser et al., 1978a, 1978d) that all
destruction of PCB-contaminated waste using incinerators must be
carefully controlled. In the temperature range 300-400 °C, the
conversion yield seems to be in the part-per-million range (Morita et
al., 1978).

Buser & Rappe (1979) studied the pyrolysis of 15 individual
synthetic PCB congeners and showed that the formation of PCDFs can
follow several competing reaction pathways. In another study where a
series of chlorobenzenes were pyrolyzed in the same way, Buser (1979)
found that significant amounts (> 1%) of PCDDs and PCDFs were formed.
A complex mixture of isomers of PCDDs and PCDFs was found, suggesting
several reaction routes. Using the same technique as above, Lindahl et
al. (1980) studied the thermal decomposition of polychlorinated
diphenyl ethers. Both PCDDs and PCDFs were formed, involving several
pathways. The temperature range was 500-600 °C and the yields varied
from 0.1 to 4.5%.

Bergman et al. (1984) studied the thermal degradation of two
polychlorinated alkanes containing 59% and 70% chlorine, respectively,
and also a commercial chlorinated paraffin containing 70% chlorine.
Their studies indicated the presence of at least mono- and diCDFs.

Ahling et al. (1978) reported that chlorinated benzenes can be
found in the pyrolysis of PVC.

Direct evidence for the conversion of PVC to PCDDs and PCDFs has
recently been reported by Marklund et al. (1986). They found that
laboratory pyrolysis of PVC resulted in the formation of PCDDs and
PCDFs, mainly hexa- and heptaCDDs, and tetra- to heptaCDFs. In some
cases, the pattern of isomers seemed to be similar to those found in
municipal and hazardous waste incinerators, e.g. the pentaCDFs (Rappe
et al., 1987).

The data discussed in this section are summarized in Table 16.

3.5.2 Incineration of municipal waste

For some time, emissions from municipal incinerators, heating
facilities, and thermal power plants have been the subject of concern.
Whereas previously the emission of dust, smoke, toxic metals, and
noxious gases were of prime concern, the presence of potentially
hazardous organic compounds from these emissions has been recognized
only recently. Lahaniatis et al. (1977) reported the presence of
chlorinated organic compounds (chlorinated aliphatics, benzenes, PCBs,
and pesticides) in fly ash from a municipal incinerator.

Olie et al. (1977) reported the occurrence of PCDDs and PCDFs in
fly ash from three municipal incinerators in the Netherlands. Their
results indicated the presence of up to 17 PCDD peaks, but isomer
identification and quantification was not possible due to the lack of
synthetic standards. Buser & Bosshardt (1978) studied fly ash from a
municipal incinerator and an industrial heating facility, both in
Switzerland. In the former, the level of PCDDs was 0.2 µg/g and of
PCDFs 0.1 µg/g. In the industrial incinerator, the levels were 0.6
µg/g and 0.3 µg/g, respectively.

During the period 1978-1982 a series of papers, reports, and
reviews were published confirming the original findings of Olie et al.
(1977) and Buser & Bosshardt (1978) regarding fly ash. Less data have
been published on the levels of PCDDs and PCDFs in other incineration
by-products, e.g., particulates and flue gas condensate, and in total
flue gas, which are the true emissions (Marklund et al., 1986).

A risk evaluation should be based on the emission levels of PCDD
and PCDF isomers found in isomer-specific analyses using validated
sampling and clean-up methods. However, in many studies non-validated
sampling and analytical methods are used and the results are given in

terms of total levels of tetra-, penta-, hexa-, hepta-, and octaCDDs
and CDFs. The value of such studies is limited, particularly in this
situation where the number of isomers is quite large. More than 30
PCDDs and 60 PCDFs have been found in fly ash samples (Buser et al.,
1978b, 1978c).

In March 1986, a working group of experts convened by the World
Health Organization Regional Office for Europe reviewed the available
data on emissions of PCDDs and PCDFs from municipal solid-waste (MSW)
incinerators. It was found that the origin of these compounds was not
completely understood, but they appear to result from complex thermal
reactions occurring during periods of poor combustion. Because of
their high thermal stability, the PCDDs and PCDFs can be destroyed
only after adequate residence times at temperatures above 800 °C
(WHO/EURO, 1987).

Available data on total emissions of PCDDs and PCDFs from tests
on MSW incinerators range between a few and several thousand ng/Nm³
dry gas at 10% carbon dioxide (CO₂). The working group prepared a
table giving a range of estimated isomer specific emissions for those
isomers of major concern with respect to MSW incinerators operating
under various conditions (Table 17).

The emissions tabulated in column 1 are those which the working
group considered to be achievable in the most modern, highly
controlled, and carefully operated plants in use at the present time.
Such results do not represent what is considered to be achievable by
the use of acid gas cleaning equipment; use of such equipment should
result in much lower values (probably at least one order of
magnitude). The results given in column 1 are not representative of
emissions that might be expected from such plants during start-up or
during occasional abnormal conditions. Emission levels listed in
column 2 were considered by the working group to be indicative of the
higher limit of emissions from modern MSW incinerators. These plants
might experience such emissions during start-up or during occasional
upset conditions. Consequently, the majority of the available
concentration data falls between columns 1 and 2. Some of the data
reviewed has shown that the figures in column 2 should not be
considered an absolute maximum. However, most existing plants, if
carefully operated, will have PCDD and PCDF emisions in the range
between columns 1 and 2.

The highest values for MSW incinerators (column 3) were obtained
by multiplying the values in column 2 by a factor of 5. Column 3
includes emission data that were reported to the working group from
all tests and under all circumstances. Generally, these emission
levels are associated with irregular or unstable operating conditions,
high moisture content of the MSW, low combustion or afterburner
temperatures, less than adequate technologies, etc.

Table 16. Formation of PCDDs and PCDFs by thermal processes

Precursor Conditions Products

2,4,5-T salt Pyrolysis 2,3,7,8-tetraCDD
2,4,5-T (vegetation) Pyrolysis No TCDD
Burning No TCDD
Cl-phenate Burning PCDDs^a + PCDFs
PCBs Pyrolysis PCDFs^b
PCBz^c Pyrolysis PCDFs + PCDDs^d
Cl-Diphenyl ethers Pyrolysis PCDFs + PCDDs
Cl-Alkanes (Paraffins) Pyrolysis PCDFs
PVC Pyrolysis PCDDs + PCDFs

^a = PCDDs formed by dimerization and a non-specific dechlorination.
^b = other products: hexa- and pentaCBs.
^c = polychlorinated benzenes.
^d = other products: PCBs, polychlorinated naphthalenes.

The working group was aware of both lower and higher emission
levels than those included in Table 17. However, it was felt that the
values included in Table 17 were likely to be representative of
emissions from current facilities (WHO/EURO, 1987).

Of special importance is the observation that the emission of
1,2,3,7,8-pentaCDD normally exceeds the emission of 2,3,7,8-tetraCDD
by a factor of three to ten.

3.5.3 Incineration of sewage sludge

Sludge from municipal waste water treatment plants may be
incinerated after being dewatered. The WHO working group (see 3.5.2)
reviewed the available data from municipal sewage sludge (MSS)
incinerators, and found that PCDD and PCDF emissions from this type of
plant were generally lower than emissions from MSW incinerators (see
Table 17, column 4) (WHO, 1986).

3.5.4 Incineration of hospital waste

Doyle et al. (1985) claimed that the incomplete combustion of
certain hospital waste containing halogenated organics could produce
high levels of PCDDs and PCDFs. They found the mean values of total
PCDDs to be 69 ng/m³ and total PCDFs to be 156 ng/m³. No
isomer-specific data seems to be available. Hagenmaier et al. (1986)
reported the analyses of stack gas from 10 hospital waste incineration
plants. The mean value of 2,3,7,8-tetraCCD emitted was 0.28 ng/m³,
the mean of all TCDDs being 20 ng/m³. The mean value for total PCDDs
was 118 ng/m³ and for total PCDFs 434 ng/m³.

Table 17. Estimated range of emissions from municipal solid waste (MSW) and municipal sewage sludge (MSS) incineratorsa

Emissions from MSW combustion Emissions
from MSS
combustion

1 2 3 4
Congeners Achievable with Maximum High Most
modern plants from emissions likely
with no acid average highest
gas cleaning operation emissions

(ng/Nm³, dry, at 10% CO₂)

2,3,7,8-TetraCDD 0.1 1.5 7.5 0.1
1,2,3,7,8-PentaCDD 0.3 14 70 0.3
1,2,3,4,7,8-HexaCDD 0.2 31 155 0.2
1,2,3,6,7,8-HexaCDD 0.6 56 280 0.6
1,2,3,7,8,9-HexaCDD 0.4 20 100 0.4
2,3,7,8-TetraCDF 0.9 10 50 0.9
1,2,3,7,8-/1,2,3,4,8-PentaCDF 2.3 52 260 2.3
2,3,4,7,8-PentaCDF 2.0 40 200 2.0
1,2,3,4,7,8/1,2,3,4,7,9-HexaCDF 1.1 48 240 1.1
1,2,3,6,7,8-HexaCDF 1.3 40 200 1.3
1,2,3,7,8,9-HexaCDF 0.06 52 260 0.06
2,3,4,6,7,8-HexaCDF 2.0 36 180 2.0

^a From: WHO/EURO (1987).

3.5.5 Incineration of hazardous waste

Analyses from a test burn of pentachlorophenol waste have been
reported by Rappe et al. (1983c). PCP is a well known precursor to
octaCDD (section 2). Samples of baghouse ash and bottom ash were
analyzed. In the baghouse ash the total level of octaCDD was only 0.2
µg/g. The major constituents were lower chlorinated PCDDs such as
hepta-, hexa-, penta-, and tetraCDDs. The isomeric distribution was
reported to be very similar to a "normal" fly ash. In both cases
2,3,7,8-tetraCDD was a very minor constituent. The level of PCDD in
the bottom ash was 0.31 µg/g. The baghouse ash was also reported to
contain PCDFs at a total level of 2.5 µg/g. For the tetra- and penta-
chlorinated compounds, equal amounts of PCDDs and PCDFs were reported.

Oberg & Bergstrém (1986) reported on test data from a Swedish
hazardous waste incinerator equipped with a rotary kiln, an
afterburner, and a dry scrubbing unit. Combustion tests were performed
with PCB (Aroclor 1242) as a fluid, and as a contaminant in solid
waste (Aroclor 1016 in capacitors). The results of these tests
indicated no correlation between the amount of PCB incinerated and the
amount of PCDDs and PCDFs found in the emissions.

3.5.6 Metal industry and metal treatment industry

It has been reported by Marklund et al. (1986) that industrial
high-temperature processes like copper smelters and electric arc
furnaces in steel mills have been identified as sources of
environmental contamination by PCDDs and PCDFs. The results are
reported in "TCDD equivalents" according to US EPA (1987). The
emission from the copper smelter contained 11 ng of TCDD
equivalents/Nm³ dry gas and 10% CO₂, while the dust from the steel
mill contained 0.8 ng TCDD equivalents/g dust. Marklund et al. (1986)
also considered the emissions from industrial incinerators to be of
the same magnitude, or even higher, than the emissions from MSW
incinerators.

Southerland et al. (1987) analyzed emissions from various
incinerators within Tier 4 of the USA. The highest levels were found
in a secondary copper smelter, which contained 170 ng of
2,3,7,8-tetraCDD/Nm³ and 3% oxygen. This was by far the highest
level found within the US EPA National Dioxin Strategy.

3.5.7 Wire reclamation

Hryhorczuk et al. (1981) studied a wire reclamation incinerator
in the USA. Using a non-isomer-specific analytical method, they
determined total levels of tetraCDDs and tetraCDFs. Two samples were
analyzed, one from the furnace and one from the stack. The furnace
sample contained 58 ng/kg of total TCDDs and 730 ng/kg of total TCDFs,
whereas the stack sample contained 410 ng/kg of total TCDDs and 11 600
ng/kg of total TCDFs.

3.5.8 Traffic

Marklund et al. (1987) reported a study where automobile exhaust
emissions were analyzed for PCDDs and PCDFs. Two groups of test cars
were utilized: (1) cars equipped with a catalytic converter using
unleaded gasoline with no halogenated scavengers; (2) cars with no
catalytic converter using leaded gasoline (0.15 g/litre) and a
dichloroethane scavenger (0.1 g/litre). Before the test runs, the
motor oil was changed in all cars. No PCDDs and PCDFs could be
identifed in the cars using the unleaded gasoline, while the average
emission from the cars running on leaded gasoline was found to be
30-540 pg/kg of TCDD equivalents. It was assumed that the chlorinated
scavenger (dichloroethane) was the precursor of the PCDDs and PCDFs
formed. It was estimated that the total amount of PCDDs and PCDFs from
cars in Sweden using leaded gasoline with halogenated scavengers is in
the range of 10-100 g TCDD equivalents/year.

3.5.9 Fires and accidents in PCB-filled electrical equipment

In February 1981 a fire in the State Office Building in
Binghamton, New York, USA, caused a transformer to rupture, releasing
soot throughout the building. The dielectric fluid in the transformer
consisted of a mixture of PCB (65%) and chlorinated benzenes (35%).
The soot was found to be highly contaminated with PCDFs (total PCDFs
> 2000 µg/g). The most toxic isomers (2,3,7,8-tetraCDF; 1,2,3,7,8-
and 2,3,4,7,8-pentaCDF; and 1,2,3,4,7,8- and 1,2,3,6,7,8-hexaCDF) were
found to be the major constituents within each group of congeners.
Levels reported were 12 mg/g of 2,3,7,8-tetra CDF, 670 mg/g of total
penta-CDFs, and 965 mg/g of total hexa-CDFs, 46 mg/g of total
hepta-CDFs, and 460 mg/g of octa-CDFs (Rappe, 1984; Rappe et al.,
1985b). In addition, a series of PCDDs were identified, including the
highly toxic 2,3,7,8-tetraCDD, and 1,2,3,7,8-pentaCDD (Rappe et al.,
1983a; Buser & Rappe, 1984). It is assumed that the chlorinated
benzenes were the dioxin precursors.

Between 1981 and 1985, a series of transformer accidents (7 in
all) similar to the one in Binghamton were reported in the USA and
Canada (Rappe et al., 1986a). In January 1985, an explosion followed
by a fire ruptured a transformer in the basement of a residential
complex in Rheims, France. The transformer was filled with PCB (60%)
and trichlorobenzene (40%). Total levels of PCDFs were as high as 2570
µg/m² before clean-up. Only traces of hepta- and octaCDD were found
(Rappe et al., 1985a).

In Europe, between 1981 and 1985, 19 accidents involving indoor
capacitor fires and explosions were reported from Scandinavian
countries (Rappe et al., 1986a). All capacitors were mineral-oil
filled, and contamination of the sites averaged 1-5 µg total
PCDFs/m².

3.5.10 Pulp and paper industry

Large amounts of chlorine or chlorine compounds are used in the
pulp and paper industry for the bleaching of the pulp. Three black
liquor boilers from the craft paper process were included in the US
EPA study of combustion sources. No 2,3,7,8-tetraCDD was found in
these emissions, but low levels of other PCDDs and PCDFs were found in
one of the three incinerators and a yearly emission of 0.25 g was
calculated (Southerland et al., 1987).

Rappe et al. (1987) recently identified both 2,3,7,8-tetraCDD
(170 ng/kg) and 2,3,7,8-tetraCDF (890 ng/kg) in a sample taken in a
sedimentation lagoon at a Swedish paper mill. A series of other PCDDs
and PCDFs was also identified but at lower levels. The isomeric
pattern in this sample differed markedly from other sediment samples,
indicating pulping processes to be a source of environmental pollution
by 2,3,7,8-tetraCDD and 2,3,7,8-tetraCDF (Rappe et al., 1987).

3.5.11 Incineration of coal, peat, and wood

The emissions of PCDDs and PCDFs from coal-fired power plants
(Kimble & Gross 1980), wood stoves (Clement et al., 1985), and peat
burning (Marklund et al., 1986) seem to be very low when calculated
per m³. However, the very high flow rates and the large number of
units could make a significant total contribution. The occurrence of
pentaCDDs and all PCDFs was not discussed in this report.

3.5.12 Inorganic chlorine precursors

It is well known that certain organochlorine compounds are
efficient precursors to PCDDs and PCDFs during pyrolysis. However, it
was proposed by scientists from Dow Chemical Company that PCDDs, and
especially 2,3,7,8-tetraCDD, are ubiquitous and formed as trace level
by-products of any normal combustion (Bumb et al., 1980).
Consequently, dioxins should have been present in the environment
since the advent of fire. This suggests that inorganic chloride can
serve as a useful precursor to the formation of PCDDs and PCDFs. A
recent survey of PCDD levels, in particular from residential wood
combustion units, has been quoted in support of the above. The survey
showed PCDD levels in the ng/kg range (see also section 3.5.11).
However, this hypothesis has been criticized. One of the main
arguments against such a hypothesis is that 2,3,7,8-tetraCDD does not
appear to be formed in coal-fired power plants (Kimble & Gross, 1980;
Junk & Richard, 1981). Another argument is that the Dow studies lack
data on levels of dioxin precursors in the material being burned,
including the air in the flames (Rappe, 1984).

The analyses of historical samples gives additional support to
the theory that organochlorine compounds are more important as
precursors than inorganic chloride. When Czuczwa & Hites (1985)
analyzed sediment core samples from Lake Huron, N. America, the first
indication of PCDDs and PCDFs was found in sediments from 1940. There
was also a good correlation between the trend in the levels of PCDDs
and PCDFs in these sediments and the trend in the production of
chlorinated aromatic compounds (section 5.4).

Schecter et al. (1986a) were unable to detect PCDDs and PCDFs in
human liver and lung tissue from two female Eskimos frozen over one
hundred years ago (see also section 4.4.4).

3.5.13 Photochemical processes

Sundström et al. (1979) studied the formation of 2,3,7,8-tetraCDD
in six re-forestation areas that were sprayed with 2,4,5-T esters.
Leaf samples from the areas were analyzed for 2,4,5-T esters and TCDD.
TCDD was found in one leaf sample only, at levels lower than expected
from the level of dioxin contamination of the herbicide formulations
used.

The photochemical formation of PCDDs and PCDFs has also been
studied in laboratory experiments.

The photochemical dimerization of chlorophenols to PCDDs was
studied by Crosby & Wong (1976). The only PCDD formed in this study
was the octaCDD. Other PCDDs can be formed by photochemical
cyclization of chlorinated o-phenoxyphenols, also called pre-dioxins
(Nilsson et al., 1974). These pre-dioxins are very common impurities
(1-5%) in commercial chlorophenols (Nilsson et al., 1978), but the
cyclization is only a minor reaction pathway; the main reaction being
the dechlorination of the pre-dioxin.

Akermark (1978) studied the formation of 2,3,7,8-tetraCDD from
the appropriate pre-dioxins. He could identify the product, but
claimed the reaction to be very inefficient.

Another photochemical process of potential environmental
importance is dechlorination of the higher chlorinated PCDDs and
PCDFs, e.g., octaCDD and octaCDF. The products formed by photolysis of
octaCDD in organic solvent have now been identified (Buser & Rappe,
1978). By comparison with authentic standards, it was found that the
main tetrachloro isomer was the 1,4,6,9-tetraCDD; the major
pentachloro compound was expected to be the 1,2,4,6,9-isomer, and the
main hexa- and heptachloro compounds were the 1,2,4,6,7,9- (or
1,2,4,6,8,9-) and the 1,2,3,4,6,7,9-isomers, respectively. The
reaction scheme deduced from this data indicates that the chlorine

atoms are removed preferentially from the lateral positions on the
carbon rings. Consequently, the most toxic PCDD isomers, such as
2,3,7,8-tetraCDD, are not likely to be formed from the photolysis of
the higher PCDDs in solution.

Crosby et al. (1973) studied the photolysis of a series of PCBs
dispersed in water. For two isomers, the 2,5-dichloro- and
2,2',5,5'-tetrachlorobiphenyls, small amounts (0.2%) of 2-mono-CDF
could be found among the products (for photo-chemical transformations,
see section 4.2.1).

3.6 Comparison of Isomeric Pattern and Congener Profiles From Various Sources

There is a pronounced difference between technical products and
incineration emissions in both isomeric patterns and congener profiles
of PCDDs and PCDFs. In technical products the number of isomers
present is limited, whereas in incineration emissions most isomers
seem to be present. Rappe (1987) has pointed out the large similarity
qualitatively in isomeric patterns between different incineration
sources.

4. ENVIRONMENTAL TRANSPORT, DISTRIBUTION, AND TRANSFORMATIONS

4.1 Environmental Transport

4.1.1 Air

The PCDDs and PCDFs are believed to be transported in the
atmosphere. The transport of these compounds from stacks and other
stationary point sources, as well as from waste disposal sites and
other area sources, can be predicted from dispersion modelling (SAI,
1980). In the case of the accidental release of a toxic cloud
containing 2,3,7,8-tetraCDD at Seveso, Italy, Cavallaro et al. (1982)
determined the transport pattern and the ground deposition. They
determined that the deposition of 2,3,7-8-tetraCDD from air to soil
should follow an exponential decay pattern in the
Gaussian-distribution along the cross-section of the downwind
direction. Thibodeaux (1983) studied the air transport of
2,3,7,8-tetraCDD at a herbicide production facility in Jacksonville,
Arkansas, USA.

The dispersion modelling has limitations. If possible, the
modelling calculations should be combined with true air measurements.

4.1.2 Water

The solubility of 2,3,7,8-tetraCDD in water has been extensively
studied (see section 2), but much less data are available for the
other PCDDs and PCDFs. However, data from microbiological experiments
indicate that 2,3,7,8-tetraCDD is highly adsorbed to sediments and
biota. Matsumura et al. (1983) suggested that more than 90% of the
2,3,7,8-tetraCDD in an aquatic medium could be present in the adsorbed
state. Rappe et al. (1985c) studied a suspension of soot/dust in the
wash water from a PCB fire. The suspension contained 100 ng/ml of
various PCDFs, but when the soot was settled the water contained no
detectable levels of PCDFs (detection level: 0.1 ng/ml of each
isomer). Most of the PCDDs and PCDFs, if present in waterways, should
be in the sediments or attached to suspended particles.

Thibodeaux (1983) has calculated the amount of 2,3,7,8-tetraCDD
transported by a creek within the contaminated herbicide factory in
Jacksonville, Arkansas, USA. The value was 0.89 g/year as an average
rate, and a maximum of 2.1 g/year.

4.1.3 Soil and sediments

The mobility of 2,3,7,8-tetraCDD and of a dichlorodioxin in soils
has been studied by Helling et al. (1973). Both were found to be
immobile in all soils and, therefore, would not be leached out by
rainfall or irrigation, though lateral transport during surface
erosion of the soil could occur.

The US Air Force conducted studies in an area of north-west
Florida that had been heavily sprayed with the herbicide Agent Orange
between 1962 and 1964 (Young et al., 1975). This herbicide mixture was
contaminated with TCDD (section 3.2). A 7.8-ha test grid received a
total of 40 metric tons of 2,4,5-T between 1962 and 1964. When 15-cm
soil core samples were taken in 1974, they showed TCDD concentrations
ranging from 10 to 710 ng/kg. This study illustrates that significant
levels of TCDD residues remained 10 years after the last herbicide
application. Similar TCDD concentrations were obtained from areas that
had been sprayed between 1962 and 1969 (Bartelson et al., 1975).

In another study Young (1983) measured the concentration of
2,3,7,8-tetraCDD in a soil profile. The samples were collected in 1974
and the data suggested that most of the 2,3,7,8-tetraCDD would be
found in the top 15 cm of the soil profile (Table 18).

The probable media and modes of transport of PCDDs from soils are
the following: (1) to air via contaminated airborne dust particles;
(2) to surface water via eroded soil transported by water; (3) to
groundwater via leaching; (4) to air via volatilization. Movement of
particulate matter containing adsorbed PCDDs and PCDFs has been
considered to be a much more important transport mechanism than
leaching and volatilization because of the low water solubility and
volatility of these compounds (Josephson, 1983). However, the
monitoring of Seveso soil one year after the accident showed that the
highest 2,3,7,8-tetraCDD levels were not present in the topmost soil
layer (0.5 cm), but very often in the second (0.5-1.0 cm) or third
(1.0-1.5 cm) layers. This disappearance of at least a part of the
2,3,7,8-tetraCDD from the topmost soil layer was speculated to be due
to volatilization or vertical movement through the soil (DiDomenico et
al., 1980). Therefore, it appears that volatilization from soil and
leaching to groundwater can be responsible for the transport of PCDDs
and PCDFs from soils under certain conditions, namely, heavy rainfall
on sandy soils. Studies by Young (1983) indicate that the half-life
for 2,3,7,8-tetraCDD in soil is 10-12 years.

Thibodeaux (1983) has calculated the vaporization of
2,3,7,8-tetraCDD from a herbicide plant in Jacksonville, Arkansas,
USA. The vaporization can take place from soil surfaces, from landfill
cells, and from the surface of a pond. In Table 19 a summary of yearly
emission rates from these sources is presented.

It was found that vaporization from the soil surface in the
highly contaminated blow out area was the major contributing source of
emissions from this plant.

Table 18. Concentration of 2,3,7,8-tetraCDD in a soil profile^a, ^b

Depth (cm) 2,3,7,8-tetraCDD (ng/kg)

0 - 2.5 150
2.5 - 5.0 160
5.0 - 10 700
10 - 15 44
15 - 90 ND^c

^a From: Young (1983).
^b The area received 1,069 kg/ha of 2,4,5-T Agent Orange during
1962-1964. The soil samples were collected and analyzed in 1974.
^c None detected (minimum detection limit: 10 ng/kg).

Table 19. Surface source areas and emission rates of 2,3,7,8-tetraCDD^a

Source Area (m²) Emission rate (g/year)

Blow-out area, volatilization 753 120-1200
Blow-out area, entrainment 753 28-37
Rocky Branch Creek, dissolved 0.89-2.1
Reasor-Hill dump 1129 0.1-1.0
Rocky Branch Creek, sediment 0.094-0.22
Cooling water pond 15050 0.015-0.016
Total 150-1240

^a From: Thibodeaux (1983).

Freeman et al. (1986) have developed a model to describe the
vaporization and diffusion through a column of soil of low volatility
organic chemicals like PCDDs and PCDFs. This model has been used to
make predictions on the transport of 2,3,7,8-tetraCDD at a site in
Times Beach, Missouri, USA. The model predicted that the 1983 levels
in this soil would be only 10% of the original loading. The model also
predicted that 57% of the initial amount of 2,3,7,8-tetraCDD was
vaporized through the soil column to the surface in the first year
after the spraying and that most transport of the vapour occurred
during the summer months. The same results were also obtained in
studies reported by Facchetti et al. (1986) and Palausky et al.
(1986).

4.2 Environmental Transformation

4.2.1 Abiotic transformation

Like other PCDDs and PCDFs, 2,3,7,8-tetraCDD is chemically quite
stable, and is not likely to be degraded at a significant rate by
hydrolytic reactions under environmental conditions. Under these
conditions, TCDD seems also to be rather stable to photochemical
degradation (Crosby et al., 1971). The half-life of TCDD of about
10-12 years, as found by Young et al. (1983) for soil, is in agreement
with this observation.

However, three reports on rapid photochemical degradation of
2,3,7,8-tetraCDD under experimental conditions make the situation more
complicated. In a methanol solution, TCDD is fairly easily degraded by
photolysis in the laboratory (Crosby et al., 1971). Other studies
using 2,4,5-T ester formulations with known amounts of TCDD and
exposed to natural sunlight on leaves, soil, or glass plates showed
that most of the TCDD was lost during a single day (Crosby & Wong,
1977). In these two studies, a "hydrogen donor", such as methanol or
2,4,5-T ester, enhanced the photochemical dechlorination (Akermark,
1978); they do not therefore truly reflect environmental conditions,
where the 2,4,5-T ester would be rapidly hydrolyzed on the surface of
the leaves. At Seveso the TCDD was released together with salts of
2,4,5-trichlorophenol, ethylene glycol, and inorganic constituents
(Rappe, 1978b). Like water, none of these is a potent hydrogen donor.

According to Bertoni et al. (1978), the addition of a solution of
ethyl oleate in xylene enhances the breakdown of TCDD in soil by UV
light. Similarly, a cationic surfactant, 1-hexadecylpyridinium
chloride, was also reported to enhance photodecomposition (Botre et
al., 1978).

Another experiment, which might be a good model for the
degradation of TCDD bound to dust particles in the air, has shown that
TCDD adsorbed on silica gel undergoes rapid photo-chemical degradation
(Gebefugi et al., 1977).

In order to explain the longer half-life of 2,3,7,8-tetraCDD in
a model laboratory ecosystem than in an outdoor pond, Matsumura et al.
(1983) speculated that photolysis was the most likely cause. In the
outdoor environment, algae-mediated photosensitization of
2,3,7,8-tetra-CDD may have caused some photodecomposition of this
compound.

An increase in chlorine substitution is expected to decrease the
rate of photodegradation. For example, Crosby et al. (1971) showed
that although complete decompostion of 2,3,7,8-tetraCDD in methanol
occurred in 24 h under UV irradiation, > 80% octaCDD in methanol
remained unreacted during the same period under similar irradiation
conditions.

Although the degree of photolysis may be related to the extent of
chlorination, different chlorine substitution patterns also play a
critical part. In higher chlorinated PCDDs, there appears to be
preferential loss of chlorine from the 2,3,7, and 8 positions (Buser
& Rappe, 1978). Thus, PCDDs with chlorine substitutions in positions
2,3,7, and 8 are likely to be photochemically degraded faster than
compounds not having these positions substituted. For example, the
photolysis half-life of 1,2,3,7,8-pentaCDD has been estimated to be
7.8 h in ⁿ-hexadecane solution under sunlamp irradiation (Nestrick
et al., 1980). Similarly, the photolytic half-lives of
1,2,3,7,8-pentaCDD, 1,2,3,6,7,9-, and 1,2,4,6,7,9-hexaCDD in hexane
solutions under sunlight irradiation have been determined to be 5.4,
17, and 47 hours, respectively (Dobles & Grant, 1979). Nestrick et al.
(1980) reported a half-life value of 6.8 h for 1,2,3,6,7,8-hexaCDD in
n-hexadecane under sunlamp irradiation. The primary intermediates of
the photo-degradation of higher chlorinated PCDDs are probably lower
chlorinated dioxins (Buser & Rappe, 1978), but the pathways of
degradation are not known with certainty (National Research Council of
Canada, 1981).

From these discussions of the photolysis of PCDDs in the presence
of organic hydrogen-donating substrates, it is difficult to predict
the photolytic fate of these compounds in natural aquatic media, where
hydrogen donors may or may not be available. The situation is
complicated further by the fact that a predominant amount of PCDDs in
surface water may be adsorbed or suspended on particles and sediments,
rather than in solution. Moreover, since the penetration of UV light
into natural water may be very limited, photolytic degradation of
PCDDs in water is not likely to be of environmental importance.

Hutzinger et al. (1973) have studied the photochemical
degradation of 2,8-diCDF and octaCDF. They found that a reductive
dechlorination takes place, especially in methanolic solution. The
reaction was much slower when a thin film was exposed to sunlight.

Thermally, 2,3,7,8-tetraCDD is quite stable, rapid decomposition
occurring only at temperatures above 750 °C (Stehl et al., 1973).

4.2.2 Biotransformation and biodegradation

The 2,3,7,8-tetraCDD isomer is very resistant to biodegradation.
Only 5 of about 100 microbial strains with the ability to degrade
persistent pesticides were able to degrade 2,3,7,8-tetraCDD (Matsumura
& Benezet, 1973). Ward & Matsumura (1978) studied the biodegradation
of ¹⁴C-labelled 2,3,7,8-tetraCDD in lake waters and sediments from
Wisconsin, USA, and observed a half-life of 2,3,7,8-tetraCDD in lake
waters containing sediment of 550-590 days. In lake water alone, about
70% of the 2,3,7,8-tetraCDD remained after 589 days. Using an outdoor
pond as a model aquatic ecosystem, and dosing it with ¹⁴C-labelled
2,3,7,8-tetraCDD, Matsumura et al. (1983) estimated the half-life of
2,3,7,8-tetraCDD to be approximately 1 year. Although biodegradation

may have been responsible for part of the degradation, it is almost
impossible to estimate the biodegradation half-life of
2,3,7,8-tetraCDD in aquatic systems from this experiment.

Philippi et al. (1982) detected a polar metabolite of
2,3,7,8-tetraCDD in several microbiological cultures after long-term
incubation. They reported chromatographic and MS data that supported
the conclusion that the metabolite was 1-hydroxy-2,3,7,8-tetraCDD,
although a synthetic standard compound was not available.

Tulp & Hutzinger (1978) reported that in rats, dibenzo-p-dioxin,
1-monoCDD, 2-monoCDD, 2,3-diCDD, 2,7-diCDD, 1,2,4-triCDD, and
1,2,3,4-tetraCDD are metabolized to mono- and di-hydroxy derivatives.
In the case of dibenzo-p-dioxin and both of the two monochloro
isomers, sulfur-containing metabolites were also excreted. Primary
hydroxylation exclusively took place at the 2, 3, 7, or 8 positions in
the molecule. In these studies, no metabolites resulting from fission
of the C-O bonds (ortho, ortho'-dihydroxychlorodiphenyl ethers,
chloro-catechols), or hydroxylated derivatives thereof, were detected.
No metabolites were found from octaCDD.

4.3 Bioaccumulation

The bioaccumulation of 2,3,7,8-tetraCDD has been investigated in
several studies, using several aquatic species and different model
ecosystems. In the experiments in which ¹⁴C-TCDD was introduced into
the model ecosystem in the form of residues on sand, particularly high
values were found in the mosquito (Aedes egypti) larvae, the level
exceeding that found in water by more than 9000 times. Under similar
conditions, the level in brine shrimp (Artemia salina) was 1570 times
higher than that found in water (Matsumura & Benezet, 1973). In the
second study (Isensee & Jones, 1975; Isensee, 1978), ¹⁴C-TCDD was
absorbed, at a broad range of levels, into soil and placed at the
bottom of an aquarium. Five species of organisms were added (though
not simultaneously) 1-30 days after flooding, and exposed for 3-32
days. The correlation between the TCDD level in the water and in the
organisms of each species was highly significant (correlation
coefficient of 0.94 or higher).

Bioaccumulation factors for 2,3,7,8-tetraCDD are given in Table
20 (US EPA, 1985).

4.4 Levels in Biota

4.4.1 Vegetation

When ¹⁴C-labelled 2,3,7,8-tetraCDD was added to soil, both oats
and soya beans accumulated small quantities of TCDD, at all stages of
growth. TCDD was also detected in control plants housed with the
experimental plants after treatment (Isensee & Jones, 1971). A maximum
of 0.15% of the TCDD present in the soil was translocated to the

aerial portion of the oats and the soya beans, but neither the grain
nor the beans harvested at maturity showed any detectable level of
¹⁴C-labelled TCDD. When TCDD was applied to the central leaflet of
3-week-old soya bean plants and 12-day-old oat plants, very little
TCDD was lost from the soya bean leaves in 21 days, but there was a
gradual loss (38% in 21 days) from the oat leaves.

Analyses of vegetation from Seveso, Italy, after the industrial
accident, gave values of up to 50 mg TCDD/kg (Firestone, 1978). In the
following years, when there was no direct contact of the newly grown
vegetation with the aerosol cloud, the levels of dioxin in plants
decreased by several orders of magnitude (Wipf & Schmid, 1983). In
1977 (one year after the accident in Seveso), no traces of TCDD were
found in the flesh of apples, pears, and peaches, or in corn cobs or
kernels, grown near the factory (the detection limit for the analyses
was 1 ng/kg). At the same time about 100 ng TCDD/kg was detected in
the fruit peels. This strongly suggests that the contamination was due
to dust and not from plant uptake. The TCDD level in the soil was
found to be in the order of 10 ng/g, which corresponds to about 1000
µg/m² (Wipf et al., 1982).

Facchetti et al. (1986) studied plants grown in soil spiked with
2,3,7,8-tetraCDD in the range 1-752 ng TCDD/kg. At the end of
cultivation, root samples were collected, carefully washed, and
analyzed. The levels of TCDD in the roots were found to be higher than
the levels found in the soil in which the plants were grown. On the
parts above ground, Facchetti et al. (1986) could not find any
significant increase in the levels of TCDD. However, the TCDD
concentration was found to vary with the location, being higher if the
plants were grown in the vicinity of other pots containing
contaminated soil. The conclusion was drawn that evaporation is the
predominant process for the contamination of the aerial parts of
plants. However, studies by Sacchi et al. (1986) indicated that maize
and bean plants grown in soil contaminated by ³H-2,3,7,8-tetraCDD
accumulated radioactivity in the aerial parts progressively with time
and with soil contamination (Sacci et al., 1986). It was suggested
that the distribution of the TCDD into the leaves occurred via the
transpiration stream.

Very few analyses of sprayed vegetation have been reported. A
rough estimate of 20-1000 ng/kg for 2,3,7,8-tetraCDD contamination can
be made on the basis of the level of 2,4,5-T found in newly sprayed
vegetation and the level of 2,3,7,8-tetraCDD in the spray formulation
used. Higher values could be obtained for Agent Orange. Sundström et
al. (1979) reported data in agreement with this estimate. However, the
analytical technique used in their study was not isomer specific.
Vegetation was sprayed with 2,4,5-T ester contaminated by only 0.06 mg
2,3,7,8-tetraCDD/g. A sample of leaves collected 42-45 days after the
spraying was found to have 170 ng TCDD/kg, somewhat lower than the
expected value 600 ng TCDD/kg, indicating a slow photochemical
breakdown.

Table 20. Measured bioaccumulation factor for 2,3,7,8-TCDD in freshwater aquatic organisms^a

Species Tissue Duration Bioconcentration Reference
(days) factor

Alga 33 3094^b Isensee (1978)
(Oedogonium cardiacum)

Alga 32 2075^c Isensee (1978)
(Oedogonium cardiacum) Yockim et al. (1978)

Snail whole body 33 5471^b Isensee (1978)
(Physa sp.)

Snail whole body 32 3095^c Isensee (1978)
(Physa sp.) 3731 Yockim et al. (1978)

Cladoceran whole body 32 3895^b Isensee (1978)
(Daphnia magna)

Cladoceran whole body 30 7070^c Isensee (1978)
(Daphnia magna) 7125 Yockim et al. (1978)

Catfish whole body 28 4875 Yockim et al. (1978)
(Italurus punctatus)

Mosquitofish whole body 14 4850^c Isensee (1978)
(Gambusia affinis) 4875 Yockim et al. (1978)

^a From: US EPA (1985).
^b Arithmetic mean of several values reported.
^c Tissue concentrations at equilibrium.

Table 21. Levels of TCDDs in fish and shellfish^a

Sample Tissue Concentration of
number type^b 2,3,7,8-TCDD (ng/kg)^c

1 Fish (edible flesh) 480
2 Catfish 40
3 Buffalo fish ND(13)
4 Fish (predator) 230
5 Fish (bottom feeder) 77
6 Catfish 50
7 Buffalo fish ND(7)
8 Catfish ND(7)

^a From: Mitchum et al. (1980).
^b All samples were obtained from the Arkansas River, USA, or
a tributary, the Bayou Meto.
^c These are averages of samples that had detectable levels of
TCDD.
ND = none detected; the number in parenthesis is the measured
detection limited for that sample.

4.4.2 Aquatic organisms

Fish and shellfish taken from areas in South Viet Nam that were
heavily exposed to Agent Orange during military defoliation operations
in the 1960s have been reported to contain 18-810 ng TCDD/kg (Baughman
& Meselson, 1973). The analytical technique of direct-inlet high
resolution MS used in this study is not considered isomer specific and
did not include any GC separation at all.

In two streams associated with the US Air Force test area in
north-west Florida (section 4.1.3), which had been heavily sprayed
with Agent Orange between 1962 and 1964, the silt contained, 10 years
later, 10 and 35 ng TCDD/kg where eroded soil entered the water.
Concentrations of 12 ng TCDD/kg were found in two species of fish from
this stream, the sailfin shiner (Notropis hypselopterus) and the
mosquito fish (Gambusia affinis). The spotted sunfish (Lepomis
punctatus) contained 4 ng TCDD/kg in skin and muscle, 18 ng/kg in the
gonads, and 85 ng/kg in the gut (Young et al., 1976).

Table 22. Analytical results for 2,3,7,8-tetraCDD residues in fish from Saginaw Bay Region, Michigan, USA^a

Species Number of Number of TCDD
samples^b positive detected (ng/kg)^c
samples low high mean

Channel
catfish 8 8 28 695 157 (13)

Carp 14 10 20 153 55 (7)

Yellow
perch 6 3 10 20 13 (5)

Smallmouth
bass 2 2 7 8 8 (6)

Sucker 4 3 4 21 10 (4)

Lake trout 2 0 0 0 0 (5)

^a From: Harless et al. 1982).
^b Mean % recovery for 2.5-10 ng ³⁷Cl4-TCDD added to 5 or 10
g of tissue prior to sample preparation was between 78 and
100%.
^c Corrected for losses in efficiency of sample preparation for
particular species. The numbers in parenthesis indicate the
limit of detection for TCDD.

The levels of TCDD in fish from the Atlantic or from ponds in the
USA in areas sprayed with 2,4,5-T were below the detection levels (1-2
ng/kg) (Baughman, 1974; Shadoff et al., 1977).

Mitchum et al. (1980) reported levels of 400 ng
2,3,7,8-tetraCDD/kg in fish samples collected in Bayou Meto/Arkansas
River, USA, a waterway associated with industrial plants for the
production of 2,4,5-T (Thibodeux, 1983) (see Table 21).

Levels ranging from 4-695 ng TCDD/kg were found in the edible
portion of channel catfish, carp, yellow perch, small-mouth bass, and
suckers from Saginaw Bay, Michigan, USA, near facilities used for the
production of 2,4,5-T herbicides. The highest concentrations were
detected in bottom-feeding catfish and carp, while the lowest
concentrations were detected in bass, perch, and suckers (see Table
22) (Harless et al., 1982).

Rappe et al. (1981) identified a series of tetra- to octaCDFs in
fat samples of a snapping turtle from the Hudson River and of gray
seal from the Baltic Sea. The total levels of PCDFs in these samples
were 3 ng/g and 40 ng/kg, respectively. In both samples the major
PCDFs consisted of the most toxic isomers (2,3,7,8-tetra-;
2,3,4,7,8-penta-; and 1,2,3,4,7,8- and 1,2,3,6,7,8-hexaCDFs).

Norstrom et al. (1982) have analyzed pooled samples of herring
gull eggs collected in 1982 from various parts of the Great Lakes, N.
America. In all samples, 2,3,7,8-tetraCDD was found in levels ranging
from 9 to 90 ng/kg. The identity of the 2,3,7,8-isomer was confirmed
by retention times on three capillary columns. In another study
Stalling et al. (1983) were not able to detect measurable levels of
tetraCDDs and other PCDDs in fish samples from Lake Superior,
N.America (the detection level was 2-5 pg/g). The difference could be
explained by the migration of the herring gulls during the winter. On
the other hand, a series of PCDFs could be identified in the Lake
Superior fish samples, indicating more widespread background levels
for the PCDFs than for the PCDDs. Stalling et al. (1983) found the
total levels of PCDFs in fish samples from Lakes Michigan, Huron, and
Ontario, N. America, to be 12-290 ng/kg. The toxic 2,3,7,8-substituted
PCDDs and PCDFs were present in all samples, the highest levels being
found in samples from Lake Huron, Lake Ontario, and the Tittabawasee
River, which flows into Saginaw Bay. The residue pattern found in the
fish and locally high levels suggest a strong influence by local point
source discharges (Stalling et al., 1983). The data of O'Keefe et al.
(1983) are also in agreement with this theory.

Norstrom et al. (1986) have studied the long-term trends of
2,3,7,8-substituted PCDDs and PCDFs in herring gull eggs in the Great
Lakes. The levels of 2,3,7,8-tetraCDD were found to decline
exponentially in Lake Ontario, with a half-life of 3-4 years, from a
high of 2000-5000 ng/kg in the early 1970s to a level of 80-100 ng/kg
in 1984/1985. The levels of TCDD in Lake Michigan were 249 ng/kg in
1971, 70 ng/kg in 1972, and 10-20 ng/kg in 1984/1985. These levels
have not changed significantly since 1979. This suggests that an
equilibrium between input and removal mechanisms has been established
in this water system for most PCDDs and PCDFs. The same trend is
reported for various fish species in Lake Ontario (Ontario, 1986).

Ryan et al. (1983a) analyzed a series of commercial and sport
fish from the Great Lakes and from the Pacific coast of Canada for
2,3,7,8-tetraCDD (Table 23). The highest levels were found in Lake
Huron and Lake Ontario. In a preliminary study they also reported
finding levels of 2,3,7,8-tetraCDFs and other unidentified tetraCDFs
of 3-200 ng/kg of fish.

The Baltic Sea is an area of interest because this region is
without any known point sources of dioxins. Rappe et al. (1987)
reported on the analyses of two samples of homing salmon and two
samples of pooled herring; one herring sample from the Baltic Sea
(Karlskrona) and the other from the northern part of the Gulf of
Bothnia (Lulea) (Table 24). As expected, the levels in the salmon
muscle were much higher than the levels found in the herrings, but,
unexpectedly, the levels in the herring sample from the Gulf of
Bothnia (Lulea) were somewhat higher than levels found in the sample
from the Baltic Sea (Karlskrona).

An interesting observation is that in the majority of the aquatic
samples only the 2,3,7,8-substituted PCDD and PCDF congeners were
found. However, crustaceans seemed to be an exception from this
general trend. Norström et al. (1988) reported that crab
hepatopancreas from the Canadian Pacific Coast contain other
congeners, e.g. 1,2,4,7,8-pentaCDD and
1,2,3,6,7,9-/1,2,3,6,8,9-hexaCDD. Rappe et al. (1987) collected and
analyzed crab hepatopancreas from three different locations along the
west coast of Sweden. The crab samples from the locations Grebbestad
and Idefjord should represent background levels, while Väröfjord has
a potential point source of dioxins from a pulp mill using chlorine
for bleaching. The results are given in Table 25.

Low background levels of series PCDDs and PCDFs were found in all
samples. In addition, the sample from the Väröfjord also contained
much higher levels of some congeners, especially 2,3,7,8-tetraCDF and
2,3,7,8-tetraCDD. This is another indication that pulp bleaching could
be a potential source of 2,3,7,8-tetraCDD and 2,3,7,8-tetraCDF (see
section 3.5.10).

4.4.3 Terrestrial animals

In a heavily sprayed test area in north-west Florida (Young et
al., 1976), a total of 106 adult and 67 fetuses of beach mice
(Peromysous polionotus) were collected in 1973 and 1974 and
examined (method not specified). Livers from the beach mice contained
from 540-1300 ng TCDD/kg and the pelts 130-140 ng/kg. The visceral
mass of race runners (Cnemidophorus sexlineatus) which were caught
in that area contained 360 ng TCDD/kg and the trunk of the reptiles
contained 370 ng/kg.

At the time of the accident in Seveso, Italy, more than 81 000
animals were inhabiting the contaminated zones. Most were rabbits (25
000), poultry, and other small animals (55 500), with 349 cattle, 233
pigs, 49 horses, 21 sheep, and 49 goats also in the zones. Many of
these animals died and others were killed. A large number of these
animals were analyzed for 2,3,7,8-tetraCDD by a method with a
detection level of 250 ng/kg (Pocchiari et al., 1983). The results are
summarized in Tables 26 and 27.

Table 23. Levels of 2,3,7,8-tetraCDD and PCB in Great Lakes Canadian sport
fish (1980) and smelt (1979)^a

Species Origin TCDD PCB
(ng/kg) (µg/g)

Lake trout^b Lake Ontario 58 7.28
Lake Huron 37 5.03

Rainbow trout^b Lake Ontario 33 1.77

Coho salmon Lake Ontario 28^c 7.39
Pacific Coast ND^d (4) 0.03

Smelt Lake Ontario 11
16
11
Lake Erie ND^d (2)

^a From: Ryan et al. (1983a).
^b Whole fish.
^c Also contained 36 ng hexaCDD/kg (three isomers) and 93 ng
octaCDD/kg.
^d ND = not detected at bracketed detection limit.

Table 24. Levels of PCDDs and PCDFs in fish samples from the Baltic Sea
(pg/g) ^a,^b

Salmon Salmon Herring Herring
Ume River Ume River Karlskrona Lulea
1985 1985 1983 1983

2,3,7,8-TetraCDF 29 12 5.5 3.0
2,3,7,8-TetraCDD 1.9 1.3 < 0.3 < 0.6
1,2,3,7,8-/1,2,3,4,8-PentaCDF 6.9 3.3 1.4 0.9
2,3,4,7,8-PentaCDF 49.0 23.0 6.8 8.8
1,2,3,7,8-PentaCDD 8.8 4.3 1.1 4.7
1,2,3,4,7,8-/1,2,3,4,7,9-HexaCDF 1.1 0.7 0.4 0.3
1,2,3,6,7,8-HexaCDF 1.3 0.8 0.4 0.3
1,2,3,7,8,9-HexaCDF ND ND 0.4 0.2
2,3,4,6,7,8-HexaCDF 1.1 0.6 0.4 0.2
1,2,3,4,7,8-HexaCDD ND 0.4 0.2 ND
1,2,3,6,7,8-HexaCDD 4.6 2.3 ND 8.1
1,2,3,7,8,9-HexaCDD ND ND ND ND
Total HeptaCDFs ND 2.7 0.8 ND
Total HeptaCDDs ND ND ND ND
OctaCDF ND 1.0 ND ND
OctaCDD ND ND ND ND

Table 24.(cont'd) Levels of 2,3,7,8-tetraCDD and PCB in Great
Lakes Canadian sport fish (1980) and smelt (1979)^a

^a From: Rappe et al. (1987).
^b ND indicates a level < 0.1 pg/g.

Harless et al. (1983) reported a study in which 2,4,5-T
containing less than 0.1 mg of 2,3,7,8-tetraCDD/kg was applied at a
rate of 3.4 kg/ha to approximately 3 ha of an enclosed plot (4.5 ha).
Twelve deer were placed in the enclosure prior to the application of
2,4,5-T. One deer died two days later of unknown causes. The remaining
deer were sacrificed prior to, and at specific intervals during, the
course of the 30-day study. The analytical results are summarized in
Table 28.

In another study (Hryhorczuk et al., 1981), samples from a horse
grazing close to a wire reclamation incinerator were analyzed and
found to contain unspecified tetraCDFs (165 ng/kg in the fat, 57 ng/kg
in the liver) and unspecified tetraCDDs (45 ng/kg in the fat and less
than 6 ng/kg in the liver) (compare section 3.5.7).

In order to identify PCDD and PCDF levels in the general
terrestrial background, Nygren et al. (1986) analyzed bovine samples
- fat, liver, and milk - and identified the same 2,3,7,8-substituted
PCDDs and PCDFs as were found in the aquatic samples. However, the
levels were lower and normally close to the detection limit.

4.4.4 Human data

Occupational exposure to 2,3,7,8-tetraCDD can occur during the
production of 2,4,5-trichlorophenol and the subsequent production and
use of 2,4,5-T acid and esters. The first commercial production of
2,4,5-T in the United States was in 1944, and the use of 2,4,5-T
herbicides increased in the 1940s and 1950s. However, the problem of
dioxin contamination in 2,4,5-T was not recognized until 1957 (Kimmig
& Schulz, 1957a, b).

During the normal production of 2,4,5-T, the heaviest exposure to
TCDD is during purification steps. The residues are far more
contaminated than the purified products. Only limited information is
available on the levels of TCDD contamination of products prepared
prior to the 1970s, and absolutely no information is available on the
dioxin levels in the corresponding residues. Consequently it is a
difficult task to estimate the levels of occupational and general
population exposures during the period prior to 1970.

Table 25. Levels of PCDDs and PCDFs in samples of crab hepatopancreas
from the west coast of Sweden^a

Crab Hepatopancreas
Idefjorden Grebbestad Väröfjord
(pg/g) (pg/g) (pg/g)

2,3,7,8-tetraCDF 31 47 590
Total tetraCDFs 90 114 800

2,3,7,8-tetraCDD 17 17 170
Total tetraCDDs 17 17 170

1,2,3,7,8-pentaCDF^b 6 7.6 45
2,3,4,7,8-pentaCDF 44 50 130
Total pentaCDFs 130 150 490

1,2,3,7,8-pentaCDD 13 11 28
Total pentaCDDs 86 76 270

1,2,3,4,7,8-hexaCDF^c 12 16 50
1,2,3,6,7,8-hexaCDF 3 5 10
1,2,3,7,8,9-hexaCDF 3 3 11
2,3,4,6,7,8-hexaCDF 16 18 63
Total hexaCDFs 70 88 280

1,2,3,4,7,8-hexaCDD 8 5 14
1,2,3,6,7,8-hexaCDD 26 18 71
1,2,3,7,8,9-hexaCDD 3 4 7
Total hexaCDDs 154 170 465

Total heptaCDFs 23 28 90

Total heptaCDDs 32 30 85

octaCDF < 1 < 1 < 2

octaCDD < 1 < 1 < 2

^a From: Rappe et al. (1987).
^b Not separated from 1,2,3,4,8-pentaCDF.
^c Not separated from 1,2,3,4,7,9-hexaCDF.

Table 26. TCDD content of the livers of farm animals from Seveso
contaminated zones and surrounding areas (1976-1979)^a

Animal Number TCDD-containing TCDD maximum
of samples samples level (ng/g)

Rabbits^b 698 433 633
Poultry 83 35 24
Cattle 43 21 94
Horses 12 2 88
Pigs 13 0 -
Goats 25 17 1
Cats 1 0 -

^a From: Pocchiari et al. (1983).
^b Figures include rabbits kept in the special test plots on
contaminated ground for experimental purposes.

Table 27. CDD analyses of wildlife from Seveso contaminated zones and
surrounding areas (1976-1979)^a

Animal Tested organs Number of Maximum level
and number of samples TCD-containing of TCDD
samples (ng/g)

Rabbits 6 (liver) 4 13
Field mice 14 (whole body) 14 49
Rats 1 (pool-4 livers) 28
Earthworms 2 (pool) 12
Frogs 1 (liver) 0.2
Snakes 1 (liver) 3

^a From: Pocchiari et al. (1983).

Table 28. Analytical results for 2,3,7,8-tetraCDD residues^a

Sections of 11 No. of deer No. of Concentration range Limit of
deer in study samples positive of TCDD detected detection
analyzed samples (ng/kg)^b range
(ng/kg)^b

Muscle 11 3 12 - 27 0.5 - 5
Adipose tissue 10 8 3 - 12 1 - 3

Table 28. (cont'd) Analytical results for 2,3,7,8-tetraCDD residues^a

Sections of 11 No. of deer No. of Concentration range Limit of
deer in study samples positive of TCDD detected detection
analyzed samples (ng/kg)^b range
(ng/kg)^b

Liver 11 4 2 - 5 0.4 - 4
Bone marrow 5 0 ND 1 - 3

^a From: Harless et al. (1983).
^b Results corrected for efficiency of sample preparation.
ND = not detected.

4.4.4.1 Adipose tissue

Gross et al. (1984) reported a study in which 30 coded samples of
adipose tissue from Viet Nam veterans were analyzed for TCDD. The TCDD
levels found for two of the three heavily exposed men were 99 pg/g and
63 pg/g, which is higher than for the other Viet Nam veterans or for
the controls (all well below 15 pg/g). Only one single isomer of
tetraCDDs was found, and it was assumed that this was the
2,3,7,8-isomer. The data in this study has also been discussed by
Young et al. (1983). These authors concluded that the levels do not
correlate well with known exposure data or with health status.

Rappe et al. (1984) reported the presence of 2,3,7,8-substituted
PCDDs and PCDFs in samples of human adipose tissue from Northern
Sweden. A series of reports presented during the period 1984-1986
confirms these observations and it has been clearly shown that there
is a background of 2,3,7,8-substituted PCDDs and PCDFs in the general
population in the industrialized part of the world. Most of these
reports are lacking data on how the sampled people were selected and
possible exposure to PCDDs and PCDFs. Consequently, these studies
might not be representative. A series of earlier studies failed to
detect these background levels due to insufficiently low detection
levels. The Swedish study included 31 people, of which 18 were exposed
to phenoxy esters and 13 were nonexposed. The group included 17 cancer
patients and 14 non-cancer patients. The different groups were matched
against each other. No difference in the levels, isomer patterns, or
ranges could be found between these subgroups (Nygren et al., 1986).
The mean values for these 31 people are given in Table 29.

Schecter et al. (1986a) reported the mean levels of PCDDs and
PCDFs in 46 samples of adipose tissue collected in Canada and in 8
samples from the USA (see Table 29). The Canadian samples were taken
from people who had died in 1976 from car accidents, drownings,

trauma, and suicide. The samples included all ages and both sexes and
came from all over the country. The USA samples, (1983-1984), were
taken from biopsies from New York State residents during the course of
normal medical procedures, and also from autopsies. Table 29 also
includes the PCDD and PCDF levels in adipose tissue samples from Viet
Nam (Schecter et al., 1986b) and from cancer patients in Japan (Ono et
al., 1986).

It is interesting to note the similarity between isomers present,
levels of isomers, isomeric patterns, and congener profiles in samples
collected from the general population in industrialized countries on
three continents. The profile of the PCDD isomers shows increasing
levels with an increasing number of chlorine atoms; the level of OCDD
is 230-900 pg/g. On the other hand, the profile of PCDFs shows a
maximum for 2,3,4,7,8-penta- or 1,2,3,6,7,8-hexaCDF. The difference in
levels found between samples from South and North Viet Nam may be
explained by spraying during the war in the 1960s and by the
difference in industrial activities between the two parts of the
country.

Four samples of adipose tissue taken from German workers exposed
to TCDD in the early 1950s have also been analyzed (Rappe et al.,
1987). In spite of the fact that these workers were exposed more than
30 years before collecting the samples, enhanced levels of
2,3,7,8-tetraCDD could be identified, but the levels of the other
PCDDs and PCDFs seem to be in the normal range (Table 29, last
column).

Patterson et al. (1986) studied the levels of 2,3,7,8-tetraCDD in
the adipose tissue of 39 exposed people and 57 controls in Missouri,
USA. The exposed group had subgroups of recreational, residential, and
occupational exposure. All persons in both the exposed and control
groups had detectable levels of 2,3,7,8-tetraCDD in their adipose
tissue. Nineteen of the 39 exposed people had measurements higher than
the highest level in the control group and six of the exposed people
had levels greater than 100 ng/kg, which was five times higher than
the highest control (Table 30).

Ryan et al. (1986) analyzed autopsy tissue samples that were
collected from three subjects who died in New York State, USA, from
natural causes. The tissue types were: fat (both abdominal and
subcutaneous), adrenal, bone marrow, liver, muscle, spleen, kidney,
and lungs. As far as could be ascertained, no subjects had known
abnormal exposure to PCDDs or PCDFs, yet these chemicals were found in
all tissues analyzed. The highest concentrations of all PCDDs and
PCDFs were found in adipose tissue. In individual tissues of the three
subjects, the levels of individual congeners detected were within a
narrow range, with much higher levels of the higher chlorinated PCDDs
and PCDFs (e.g. hepta- and octa-CDD). In adipose tissue, the level of
2,3,7,8-tetraCDD was 3.7-8.4 ng/kg and that of 2,3,4,7,8-pentaCDF

5.2-13 ng/kg, while octaCDD ranged between 430 and 700 ng/kg. No major
differences were seen between abdominal and subcutaneous fat samples
or between these two types and perirenal fat when the lower lipid
content of the latter was considered. Smaller concentrations of PCDDs
and PCDFs were measured (on a wet weight basis) in decreasing order:
adrenal, bone marrow, liver, muscle, spleen, kidney, and lungs.

The high levels found in exposed Viet Nam veterans and German
workers indicate a very slow excretion rate or metabolism of TCDD in
humans. This indicates a dramatic difference between man and rodents;
in the latter the half-life of TCDD is reported to be in the range of
a few weeks.

4.4.4.2 Blood plasma

Analysis of blood plasma has been used to evaluate occupational
exposure to PCDDs and PCDFs, which can occur during the production or
use of 2,4,6-tri-, 2,3,4,6-tetra-, and pentachlorophenol. Rappe et al.
(1983) investigated such exposure through the analysis of blood plasma
of exposed workers and unexposed controls. Good correlations were
found between the plasma levels and:

(a) the nature of exposure - dermal contact with liquids resulted in
higher levels than inhalation of contaminated dust;

(b) the duration of exposure - higher levels for longer exposure
times.

The isomers present in the formulations used could also be found in
the blood plasma.

Kochman et al. (1986) studied a group of people exposed to PCBs
and PCDFs after a transformer fire in Rheims, France in 1985. Low
levels (1-25 pg/g) of 2,3,7,8-substituted penta-, hexa-, hepta-, and
octaCDFs were found in the blood plasma of these people, and there was
a slight variation in the T-lymphocytes cells.

Kahn et al. (1986) measured the levels of PCDDs and PCDFs in the
blood plasma from 10 heavily exposed Viet Nam veterans and their 17
controls. The levels of TCDD was much higher in the exposed men than
in their controls.

Table 29. Levels of PCDDs and PCDFs in human adipose tissue (ng/kg wet weight)

Sweden USA/NY^f Canada^f Japan^f N Viet Nam^ff FRG
Isomer n=31^a n=8^b n=46^b n=13^c n=9^d n=15^d n=4^e

2,3,7,8-tetraCDD 3 7.2 6.4 (25) 9 (12) < 2 28 (12) 150
1,2,3,7,8-pentaCDD 10 11.1 10 (46) 15 (13) < 2 15 (14) 19.2
1,2,3,6,7,8-hexaCDD 15 96 81 (46) 70 (12) 11 (6) 100 (15) 77
1,2,3,7,8,9-hexaCDD 4 NA NA 12 (10) NA NA 9.4
1,2,3,4,6,7,8-heptaCDD 97 164 135 (46) 77 (12) 28 (6) 178 (15) 56
octaCDD 414 707 830 (46) 230 (12) 104 (8) 1256 (15) 267

2,3,7,8-tetraCDF 3.9 NA NA 9 (13) NA NA 0.9
2,3,4,7,8-pentaCDF 54 14.3 15 (46) 25 (13) 13 (7) 21 (15) 44
1,2,3,4,7,8-hexaCDF 6 NA NA 15 (11) NA NA 10.0
1,2,3,6,7,8-hexaCDF 5 31.3 16 (34) 14 (11) 13 (7) 58 (15) 6.7
2,3,4,6,7,8-hexaCDF 2 NA NA 8 (3) NA NA 3.8
1,2,3,4,6,7,8-heptaCDF 11 16.5 30 (44) NA 7 (3) 29 (15) 19.5
octaCDF 4 NA NA NA NA NA 1

n = number of tissue samples.
NA = not analyzed.
^a Nygren et al. (1986).
^b Schecter et al. (1986a).
^c Ono et al. (1986).
^d Schecter et al. (1986b).
^e Refers to occupationally exposed workers. Rappe et al. (1987).
^f Mean values of positives. Number of positives within brackets.

Levels below the detection level (< 1.0 ng/kg) not included.

Table 30. Comparison of Levels of 2,3,7,8-Tetrachlorodibenzo-p-dioxin (ng/kg) in Adipose Tissue of Exposed and Control
Groups^a

Exposed

Variable Controls Total Recreational Residential Occupational

Number of subjects 57 39 8 16 15
Arithmetic mean 7.4 79.7 90.8 21.1 136.2
Median 6.4 17.0 23.5 14.5 24.7
Range 1.4-20.2 2.8-750 5.0-577 2.8-59.1 3.5-750
Geometric mean 6.4 21.8 24.8 15.3 29.8
Mean age, years (SD) 52.6 (15.7) 44.3 (13.7) 42.1 (14.7) 39.7 (14.9) 50.3 (9.8)
% Men 35.1 61.5 37.5 43.8 93.3

^a From: Patterson et al. (1986).

5. ENVIRONMENTAL LEVELS AND HUMAN EXPOSURE

5.1 Air

Owing to sampling and analytical problems, very few data are
available on the levels of 2,3,7,8-TCDD and other PCDDs and PCDFs in
normal urban air.

Rappe & Kjeller (1987) reported the levels of PCDDs and PCDFs in
total air samples collected in Hamburg, FRG, and samples of air
particulates collected in Sweden (Table 31). Sample 1 was collected on
the outskirts of Hamburg (representing an urban area), sample 2 in a
traffic tunnel, sample 3 downwind from a MSW incinerator, and sample
4 in the vicinity of a dumpsite and metal refinery. PCDDs and PCDFs
were found in all samples. Lower levels of PCDDs and PCDFs were found
in air particulates than in total air samples (Table 31). Sample 5 was
taken when "clean air" was blowing into a street in Gothenburg,
Sweden, and sample 6 was taken in the same street during an inversion
situation. Samples 7 and 8 were taken at a rural research station
outside Gothenburg when the air was blowing from the sea (sample 7) or
from Gothenburg (sample 8). The isomeric pattern found in these
samples were very similar (Rappe & Kjeller, 1987).

Airborne dust was monitored in 1977 in the Seveso area to
evaluate the possibility that 2,3,7,8-tetraCDD-contaminated particles
might have drifted outside the contaminated areas. A high-volume
sampling technique was used. When pooled particulate samples were
analyzed, levels of 0.17-0.50 pg TCDD/m3 were reported (Wipf et al.,
1982).

The atmospheric concentrations of 2,3,7,8-tetraCDD near two
hazardous waste sites have been monitored. In one study, US EPA (1982)
failed to detect (detection limit: 1-20 pg/m3) any 2,3,7,8-TCDD in the
atmosphere at the Love Canal (New York, USA) area. In another study of
a waste disposal site (near Jacksonville, Arkansas, USA), Thibodeaux
(1983) reported an average concentration of 1100 pg of 2,3,7,8-TCDD/g
in two air particulate samples collected near the disposal site.

Rappe et al. (1985c) analyzed indoor air samples for PCDFs
resulting from fires and explosions in PCB-filled electrical
equipment, and in an industrial situation (locomotive shop) (Table
32).

O'Keefe et al. (1985) analyzed air samples collected in an office
building in Binghamton, New York, USA, after a transformer accident in
the basement in February 1981. The samples were collected after a
primary clean-up and the values are given in Table 33.

Table 31. Levels of PCDDs and PCDFs in samples of total air and air
particulates^a

Total air samples Air particulates
1 2 3 4 5 6 7 8
pg/m³ pg/m³ pg/m³ pg/m³ fg/m³ fg/m³ fg/m³ fg/m³

2,3,7,8-tetraCDF^b 0.04 0.72 0.38 0.18 30 240 5 62
Total tetraCDFs 0.36 6.2 4.9 3.3 320 2000 54 490

2,3,7,8-tetraCDD 0.02 0.06 0.02 0.08 3 9 < 1 5
Total tetraCDDs 0.10 0.22 0.21 1.5 150 350 9 130

1,2,3,7,8-pentaCDF^c 0.04 0.36 0.42 1.0 39 190 7 58
2,3,4,7,8-pentaCDF 0.04 Int^e 0.43 1.2 51 240 6 69
Total pentaCDFs 0.51 4.1 5.0 10 470 2500 85 610

1,2,3,7,8-pentaCDD < 0.02 0.28 0.22 0.6 17 66 5 35
Total pentaCDDs 0.07 1.3 2.4 5.0 200 840 31 280

1,2,3,4,7,8-hexaCDF^d 0.03 0.13 0.27 1.1 23 100 8 38
1,2,3,6,7,8-hexaCDF 0.03 0.15 0.24 1.4 20 78 8 33
1,2,3,7,8,9-hexaCDF < 0.01 < 0.05 < 0.02 0.33 4 17 3 14
2,3,4,6,7,8-hexaCDF < 0.01 < 0.05 0.12 0.80 10 84 7 32
Total hexaCDFs 0.18 1.1 2.2 9.5 180 800 70 310

1,2,3,4,7,8-hexaCDD < 0.08 < 0.17 0.19 1.0 3 19 < 1 7
1,2,3,6,7,8-hexaCDD 0.23 0.66 0.71 2.2 11 46 4 14
1,2,3,7,8,9-hexaCDD < 0.08 < 0.17 0.36 5.2 6 92 5 32
Total hexaCDDs 0.74 2.7 5.3 24. 100 520 32 190

Total heptaCDFs 0.10 1.2 2.0 5. 200 1100 120 500

Total heptaCDDs 0.60 3.4 5.3 15. 380 2900 140 1000

OctaCDF < 0.11 < 1.0 0.78 7.0 150 480 100 440

OctaCDD 0.37 6.4 7.4 40.0 290 1900 64 540

^a From: Rappe & Kjeller (1987).
^b Not separated from 2,3,4,8-tetraCDF.
^c Not separated from 1,2,3,4,8-pentaCDF.
^d Not separated from 1,2,3,4,7,9-hexaCDF.
^e Int = Interferences.

5.2 Water and Leachate

Shadoff et al. (1977) failed to identify 2,3,7,8-tetraCDD in
water from areas in the USA where 2,4,5-T herbicides had been used.

In the 2,4,5-T plant in Jacksonville, Arkansas, USA, Thibodeaux
(1983) could not detect any 2,3,7,8-tetraCDD in the creek water (no
limit of detection levels given).

Since August 1976, a number of tests have been periodically
conducted in Seveso on streams running through the affected area, as
far south as the River Lambro, with consistently negative results.
During the same period sediment samples were taken from Torrents,
Certesa, and Seveso. Positive results of the order of 1 pg/g were
obtained within the first few kilometers downstream from their
confluence, but further downstream results were negative. The
intensive rainfalls after the accident caused the Seveso to repeatedly
overflow its embankments at the point of entry into Milan, thus
depositing silt on adjacent areas. Tests conducted to determine TCDD
in these silts yielded negative findings for the first four floods
while the fifth flood yielded positive findings (pg/g). Since August
1976, the monthly determinations conducted on pipeline and ground
waters have consistently yielded negative results, even when the
analytical detection threshold was as low as 1 pg/litre (parts per
quadrillion) (Pocchiari, 1983).

During 1983 and 1984, the Dow Chemical Company conducted a study
to determine the 2,3,7,8-tetraCDD contamination at its plant in
Midland, Michigan, USA. It was estimated that 0.6 g of
2,3,7,8-tetraCDD was being emitted per year in 2.5x107 m³ of
wastewater effluent (Lamparski et al., 1986).

The Ontario Ministry of Environment has included PCDDs and PCDFs
in its Drinking Water Surveillance Program for the St Clair/Detroit
River area (Ontario, 1986). No 2,3,7,8-tetraCDD has been found in any
sample of raw or treated water. Unspecified congeners of PCDDs and
PCDFs have been found mainly in raw water, octaCDD being the most
frequent congener found. The highest value reported was 1.1 pg
octaCDD/litre raw water in Amherstburg. The octaCDD level in the
treated water was below the detection level of 0.01 pg/litre.

Götz (1986) reported on levels of PCDDs and PCDFs in the oily
leachate from a sanitary landfill in Georgswerder, Hamburg, FRG (table
34).

5.3 Soil and Sediment

The levels of 2,3,7,8-tetraCDD in point source after improper
disposal of industrial waste are discussed in section 3.4.2.

Table 32. Analyses of PCDFs in air samples (pg/m3)^a

total
Sample tetra- 2,3,7,8- penta- hexa- hepta- octa-
CDF tetraCDF CDF CDF CDF CDF

Surahammar < 20 < 2 < 10 < 10 < 10 < 10
(during cleaning)

Surahammar < 10 < 2.5 < 10 < 10 < 10 < 10
(after cleaning)

Railway locomotive 500 50 50 30 20 20
(during cleaning
operations)

^a From: Rappe et al. (1985c).

Table 33. Concentrations of PCDFs in air samples collected on various
floors of a Binghamton, New York (USA) office after primary cleanup

Analytical results (pg/m³)

Floor/sample type^a 2,3,7,8- Total Penta- Hexa-
tetraCDF tetraCDFs CDFs CDFs

3 16 151 43
5 11 126 30 2.0
5 (NE) 20 195 60 8.7
7 11 121 36
9 volatiles 14 140 42
9 particulates 1.8 4.8 4.7
9 (SE) volatiles 13 146 31 3.7
9 (SE) particulates 0.8 3.9 3.2
11 23 76 16
11 (SE + NW) 16 133 19
14 11 92 21
14 (NE) 14 185 13
16 16 118 21
17 volatiles 12 79 24
17 particulates 0.8 3.9 ND^b
17 volatiles 9 59 6.6
17 particulates 0.9 ND^b 2.9

Table 33. cont'd

^a Abbreviations in parentheses designate sampling location on the floor,
e.g., SE = south-east corner. Unless otherwise specified samples were
collected in the north-west corner and analyzed as combined particulates
and volatiles.

^b ND = not detected.

Table 34. Levels (ng/g) of the 2,3,7,8-substituted PCDDs and PCDFs in
leachate from a sanitary landfill^a

Isomer Concentration Isomer Concentration

2,3,7,8-tetraCDD 60 2,3,7,8-tetraCDF 9
1,2,3,7,8-pentaCDD 28 1,2,3,7,8-pentaCDF^b 322
1,2,3,4,7,8-hexaCDD 476 2,3,4,7,8-pentaCDF 261
1,2,3,6,7,8-hexaCDD 1440 1,2,3,4,7,8-hexaCDF 748
1,2,3,7,8,9-hexaCDD 310 1,2,3,6,7,8-hexaCDF 336
1,2,3,7,8,9-hexaCDF 558
2,3,4,6,7,8-hexaCDF 114

^a From: Götz (1986).
^b Overlapping isomer: 1,2,3,4,8-pentaCDF.

Analytical results of the 1976-1977 survey of Zones B and R in
Seveso were discussed by Pocchiari (1983). TCDD levels in Zones B and
R were, in general, considerably lower than those in Zone A. In fact,
most TCDD levels were lower than 50 µg/m² in Zone B and 5 µg/m² in
Zone R. In 1980, a large part of Zone R was remonitored to evaluate
the persistence of TCDD in the soil. This zone had been ploughed and
worked since 1978. A comparison, as well as a statistical evaluation,
of the relevant data indicated a significant decrease (40%) in the
geometric mean level of TCDD in the soil of Zone R.

In 1980 and 1981, soil samples from ten sites in Zone R and five
sites outside Zone R were analyzed using a high resolution GS-MS
system to establish whether other isomers of 2,3,7,8-tetraCDD were
also present. A significant percentage decrease in tetraCDDs could be
accounted for by two isomers (1,3,6,8-tetraCDD and 1,3,7,9-tetraCDD)
present in the majority of the samples tested. These two isomers were
not related to the chemical accident at the factory.

Table 35. PCDD contamination in soil from Zone R in Seveso (1981)
(values in ng/kg)^a

Sample 2,3,7,8^b TCDD^c Penta- Hexa- Hepta- Octa- Total
tetraCDD CDDs CDDs CDDs CDD PCDDs

S1 0.8 0.3 0.4 6.0 1.4 1.7 10.6
S2 3.4 1.0 0.7 9.5 2.1 2.2 18.9
S3 4.0 1.9 1.2 8.2 8.6 27.0 50.9
S4 2.3 0.8 0.6 10.2 2.1 2.0 18.0
S5 < 0.1 < 0.3 0.5 9.5 1.9 1.4 13.7
S6 6.3 1.5 1.1 10.4 2.6 1.3 24.8
S7 1.7 1.0 0.8 12.4 1.9 1.8 19.6
S8 2.2 0.4 0.8 8.8 1.8 0.8 14.8
S9 1.0 2.8 2.3 21.2 9.6 13.5 50.4

^a From: Wipf & Schmid (1983).
^b Probably related to the accident.
^c Total levels of isomers other than 2,3,7,8-tetraCDD,
probably not related to accident.

Wipf & Schmid (1983) reported the presence of PCDDs other than
2,3,7,8-tetraCDD in the soil from Zone R (Table 35). They suggest that
a municipal incinerator and the burning of wood shavings treated with
chlorinated phenols could be the source of the other PCDDs.

Nestrick et al. (1986) reported the levels of 2,3,7,8-tetraCDD in
soil samples collected from industrialized areas of US cities. They
observed a widespread occurrence of 2,3,7,8-tetraCDD in urban soils,
with levels of 1-10 ng/kg, and suggested that local combustion
sources, including MSW and industrial incinerators, were the probable
origin.

McLaughlin & Pearson (1984) measured soil concentrations of PCDDs
and PCDFs in the vicinity of a municipal refuse incinerator in
Ontario, Canada. Urban and rural control locations were also sampled.
All soil samples (14) had detectable quantities of at least one of the
five PCDD congener classes (tetraCDD to octaCCD) tested for, whereas
eight samples contained detectable levels of one or more of the five
PCDF congener groups (tetraCDF to octaCDF). The levels ranged from
non-detectable (0.003 - 0.008 ng/g) to 3.5 ng/g (octaCDD); only one
site had a measurable quantity (0.007 ng/g) of tetraCDD in the soil.
The most abundant PCDD or PCDF congener was octaCDD, which had similar
levels whether samples were taken close to or remote from the
incinerator. Similarly, no concentration gradients, relative to
distance from the incinerator, were apparent for any of the other
PCDDs or PCDFs.

Soil samples near a chemical waste incinerator in Scotland have
also been analyzed for PCDDs and PCDFs, together with samples from
control locations (Edulgee et al., 1986). Detectable levels of all
PCDFs or PCDFs that were examined were found in each of the soil
samples (13). Levels found ranged from 1.2 ng/kg (2,3,7,8-tetraCDD) to
1900 ng/kg (total hexaCDF). No consistent pattern was observed to
differentiate levels found in control samples from levels in soil near
the incinerator.

These studies from widely separate areas of the world support the
suggestion that diffuse combustion sources are the major source of
PCDDs and PCDFs in the soil.

Rappe & Kjeller (1987) analyzed soil samples from various parts
of Europe (Table 36). They represent rural areas (samples 1, 2, and 3)
as well as more industrialized areas (samples 4 and 5). PCDDs and
PCDFs could be identified in all samples. The 2,3,7,8-tetraCDD
concentration was below the detection level in the soil samples from
the non-industrialized areas. Trapped sediments from the archipelago
of Stockholm, Sweden, were also analyzed. The samples were collected
in the inner (sample 6), middle (sample 7), and outer archipelago
(sample 8). Levels decreased with increasing distance from the city of
Stockholm. The isomeric patterns for tetra- and pentaCDF isomers are
very similar to those found for samples of total air and air
particulates (section 6.1). A sediment sample from the mouth of River
Viskan, Sweden, was also analyzed (sample 9). A slight difference in
congener profile was found between this sample and the sediments from
the archipelago of Stockholm.

Czuczwa & Hites (1985) found PCDDs and PCDFs in sediment samples
from several locations in Saginaw River and Bay, and southern Lake
Huron, levels ranging from 100 ng/g in urban areas to 100 ng/kg at
remote sites. Although no isomers were identified, the analytical
profiles in the sediments followed closely those found in combustion
samples, suggesting that combustion is the major source of PCDDs and
PCDFs found in the sediments. Analyses of sediment cores showed a
dramatic increase in the PCDD and PCDF concentrations at a depth
corresponding to approximately the year 1940, and levels remained high
up to the present. There is no good correlation between the trend in
these levels and the trend for coal burning in the United States.
However, the levels in the sediments correlate with the production and
use of chlorinated aromatic compounds within this area of the Great
Lakes.

5.4 Food

5.4.1 Meat and bovine milk

The levels of PCDDs and PCDFs in fish and other seafood are
discussed in section 4.4.2.

Table 36. Levels (pg/g) of PCDDs and PCDFs in samples of sediments and soil^a

Soil samples Sediments
1 2 3 4 5 6 7 8 9

2,3,7,8-tetraCDF^b 2.9 1.6 1.1 34 38 30 17 14 1.6
Total tetraCDFs 9.3 7.7 11 320 370 290 150 120 24

2,3,7,8-tetraCDD < 2.0 < 2.1 < 0.2 2.4 0.8 2.4 2.0 < 2.0 0.2
Total tetraCDDs - - 3.2 55.5 11.2 69 21 23 6.4

1,2,3,7,8-pentaCDF^c 2.5 1.6 0.5 17 31 16 8.6 8.5 1.3
2,3,4,7,8-pentaCDF 0.8 1.0 0.6 23 65 20 14 16 1.7
Total pentaCDFs 14 13 6.7 200 450 260 140 130 30

1,2,3,7,8-pentaCDD < 2.0 < 2.0 < 0.1 18 34 7.6 5.2 5.5 0.9
Total pentaCDDs - - 4.6 220 270 230 99 86 13

1,2,3,4,7,8-hexaCDF^d 3.8 2.2 0.9 30 45 16 10 8.8 1.9
1,2,3,6,7,8-hexaCDF 1.8 1.5 0.4 11 25 12 7.1 5.9 1.2
1,2,3,7,8,9-hexaCDF 1.0 0.9 4.3 110 1100 5 < 1 < 1 2.0
2,3,4,6,7,8-hexaCDF 1.9 1.0 0.7 26 57 16 31 23 1.6
Total hexaCDFs 16 12 11 270 1900 250 220 92 44

1,2,3,4,7,8-hexaCDD < 2 < 2 < 0.1 13 28 1.6 0.8 1.0 1.6
1,2,3,6,7,8-hexaCDD < 2 < 2 < 0.1 19 64 48 2.0 2.0 10
1,2,3,7,8,9-hexaCDD < 2 < 2 < 0.1 6.2 19 2.5 0.9 1.0 4.3
Total hexaCDDs - - 4.7 200 330 49 16 19 64
Total heptaCDFs 22 14 18 260 4500 1300 1500 190 300
Total heptaCDDs < 10 < 10 17 370 1600 5700 1200 880 190
OctaCDF - - 5.7 68 71 39 < 20 < 20 330
OctaCDD - - 14 140 180 3100 510 260 900

^a From: Rappe & Kjeller (1987).
^b Not separated from 2,3,4,8-tetraCDF, except in the case of sample 9.
^c Not separated from 1,2,3,4,8-pentaCDF.
^d Not separated from 1,2,3,4,7,9-hexaCDF.

The US Environmental Protection Agency (US EPA) initiated a
2,3,7,8-tetraCDD monitoring programme of beef fat samples taken from
cattle that had grazed on rangelands known to have been treated with
2,4,5-T. The analytical collaborators in this programme were Dow
Chemical Co. (USA), Wright State University, and Harvard University.
All the laboratories used mass spectroscopic techniques for
quantification. Two different extraction techniques were used. All
three laboratories analyzed control samples taken from cattle that had
grazed on non-treated areas. Some of these control samples were spiked
with known amounts of TCDD. All the controls were prepared by the US
EPA (Firestone, 1978). Good agreement was found between the amounts of
TCDD spiked into the control beef fat samples and the reported levels
found, even down to TCDD levels of 10 ng/kg. The average reported TCDD
level was 10 ng/kg; the amount actually added by the EPA was 9 ng/kg.
Of a total of 34 analyses of controls to which no TCDD was added, in
only one case was there a false positive report of TCDD (O'Keefe et
al., 1977). Of 52 samples of beef fat from 2,4,5-T-treated rangeland,
19 (37%) were reported by one or more laboratories to have TCDD. The
average range of levels reported was 5-66 ng/kg, and the overall
average was 7 ng/kg. If one considers only the 40 beef fat samples
from areas receiving at least 1.1 kg of 2,4,5-T/ha, all 19 positive
samples (48%) belong to this group, and the average reported TCDD
level would be 9 ng/kg. The results indicated a consistent trend,
relating the average reported TCDD level in beef fat to the intensity
of the 2,4,5-T application to rangeland (O'Keffe et al., 1978;
McKinney, 1978).

None of the three collaborating laboratories used the most
selective and sensitive analytical method now known (capillary glass
column gas chromatography - high resolution mass fragmentography).

Kocher et al. (1978) also analyzed specimens of fat taken from
steers that had grazed on rangeland previously treated with 2,4,5-T
herbicides. The limit of detection of TCDD (2.5 times peak to peak
noise) was found to be in the 30-60 pg range (3-6 pg/g in beef fat
using 10 gram samples). None of the sixteen samples analyzed in two of
three studies revealed TCDD. In the third study, the animals were
confined to a fenced pasture sprayed in its entirety with 2,4,5-T
herbicides. The samples from three of the seven animals gave a
positive response at the extremely low level of 3 to 4 ng TCDD/kg,
which is at the detection limit; the highest reported value (without
interfering components) was 13 pg/g. The level of 2,3,7,8-tetraCDD in
the 2,4,5-T used was, however, unknown. Beck et al. (1986) analyzed
seven randomly collected samples of cow's milk from different areas of
the Federal Republic of Germany and one sample from the German
Democratic Republic. The cow's milk was taken from road transport
tankers. In all samples, 2,3,7,8-substituted PCDDs and PCDFs were
found at low levels of pg/g on a fat weight basis. Detection limits

were in the range 0.1-0.3 ng/kg. The levels of PCDDs and PCDFs found
in the milk samples were lower than those measured by Rappe et al.
(1987b) in cow's milk from Switzerland (Table 38). In addition, there
was no evidence of high levels of the higher chlorinated congeners
(e.g. hepta and octa- ) (see Table 37). Levels were much lower than
those in human milk samples (see section 5.4.2).

Rappe et al. (1987b) analyzed PCDDs and PCDFs in six samples of
bovine milk from various locations in Switzerland. In all samples,
2,3,7,8-substituted PCDDs and PCDFs were found at levels of pg/kg in
whole milk (Table 38). However, the levels were lower in commercial
milk samples than in samples collected directly from cows grazing in
the vicinity of incinerators.

When Ryan et al. (1985) analyzed PCDDs and PCDFs in chicken and
pork samples in Canada, the incidence of positives for hexa-, hepta-,
and octaCDD in selected samples of chicken fat was 50, 62, and 46%,
with averages of 27, 52, and 90 ng/kg, respectively. Similar levels of
hexa- and heptaCDFs were also found in some of these samples, but
tetra- and pentaCDDs and tetra-, penta-, and octaCDFs were not
detected. A comparison between the tissue analyses and those of the
wood (treated with pentachlorophenol) used to house the animals showed
a marked similarity (see Table 38), indicating that pentachlorophenol
was the probable source of contamination of the food samples.

Firestone et al. (1986) reported the analyses of various food
items collected in the period beginning in 1979. Low levels (< 300
ng/kg) of 1,2,3,4,6,7,8- and 1,2,3,4,6,7,9-heptaCDD were found in some
samples of chicken, bacon, pork chops, and beef liver. HexaCDD was not
found in any of the foods. Several beef livers had high levels of OCDD
residues, the highest reported value being 3830 ng/kg. No PCDDs (at a
detection limit of 10-40 ng/kg) were found in ground beef.

5.4.2 Human milk

Rappe et al. (1984) reported low levels of 2,3,7,8-substituted
PCDDs and PCDFs in five samples of human milk from the Federal
Republic of Germany (FRG) (Table 39).

Table 39 also indicates the results of analyses of four samples
of human milk from the Umea region of northern Sweden (Rappe, 1985),
92 samples from Rheinland-Westfalia in the FRG (Furst et al., 1987),
30 other samples from the FRG (Beck et al., 1987), and five samples
from the Netherlands or Yugoslavia (Rappe et al., 1987).

Van der Berg et al. (1986) also reported the levels of PCDDs and
PCDFs in human milk samples from the Netherlands, but these levels
were reported on milk basis, not on fat basis, and are therefore not
included in Table 39.

A comparison between the isomers, levels, isomeric pattern, and
congener profiles found in human milk (Table 39) and in adipose tissue
(Tables 29 and 30) shows a remarkable degree of similarity.

5.4.3 Rice

Rice from fields in Arkansas, Louisiana, and Texas, USA, treated
at a maximum rate to give 2.52 kg 2,4,5-T/ha, were analyzed for
possible 2,3,7,8-tetraCDD residues. A specification of 1 µg
2,3,7,8-tetraCDD/g 2,4,5-T was given in the report, but no analytical
data were given for the herbicide. No 2,3,7,8-TCDD was detected in the
rice (detection limit = 2-7 µg/kg) and no 2,3,7,8-TCDD residues
(detection limit = 2-10 µg/kg) were found in 30 samples of rice
purchased in retail stores throughout the USA (Jensen et al., 1983).

5.5 Yusho and Yu-cheng Episodes

In 1968, more than 1500 people in south-west Japan were
intoxicated through consuming commercial rice oil accidentally
contaminated by PCBs, PCDFs, and polychlorinated quarterphenyls
(Masuda & Yoshimura, 1982; Masuda et al., 1985). In 1979, a similar
episode occurred in central Taiwan, the number of persons involved
here approaching 2000 (Chen et al., 1980; Chen et al., 1985). In the
past, both accidents were referred to as Yusho episodes, but now the
Taiwan episode has been renamed Yu-cheng.

The Japanese rice oil contained more than 40 PCDF isomers (tri-
to hexaCDFs) (Buser et al., 1978d), whereas the number of isomers in
the Taiwanese oil seems to have been less (Chen & Hites, 1983). The
toxic 2,3,7,8-substituted PCDFs were middle or minor constituents,
about 10-15% of the total amount of PCDFs (Buser et al., 1978d; Masuda
et al., 1985; Chen & Hites, 1983). The mean total consumption of PCDFs
of the Yusho and Yu-cheng patients has been estimated to be 3.3-3.8
mg/person (Hayabuchi et al., 1979; Chen et al., 1985), or a daily
intake of total PCDFs of 0.9 µg/kg body weight (Hayabuchi et al.,
1979). The average intake of 2,3,7,8-substituted PCDFs was 90-135
ng/kg body weight per day. The smallest amount of total PCDFs causing
chloracne has been estimated to be 0.16 µg/kg body weight per day
(Hayabuchi et al., 1979) or 20-30 ng/kg per day of the
2,3,7,8-substituted congeners.

Analysis of liver samples taken from the Yusho patients about 18
months after the exposure showed a dramatic decrease in the number of
PCDF isomers. Apparently most of the PCDF isomers were metabolized or
excreted during the period between exposure and sampling (Rappe et
al., 1979). A comparison between the PCDF isomers found in the Yusho
oil and the liver samples revealed an interesting relationship. Most
of the isomers retained had lateral positions (2, 3, 7, and 8)
substituted with chlorine; these isomers have the highest toxicity
(Rappe et al., 1979).

Table 37. PCDD and PCDF levels in samples of cow's milk (ng/kg on a fat weight basis)^a

SAMPLE

1 2 3 4 5 6 7 8

2,3,7,8-tetraCDF < 0.1 0.29 0.28 1.1 1.0 1.4 1.4 0.27
2,3,7,8-tetraCDD 0.33 < 0.2 < 0.2 < 0.2 ND < 0.2 ND ND
1,2,3,7,8-pentaCDF ND 0.4 ND 0.26 0.24 ND 0.39 < 0.2
2,3,4,7,8-pentaCDF 1.3 0.91 1.1 1.6 1.5 1.3 2.9 0.8
1,2,3,7,8-pentaCDD 1.0 0.72 ND 0.81 0.6 0.78 1.2 < 0.5
1,2,3,4,7,8-hexaCDF 0.93 0.67 0.70 0.85 < 0.3 0.84 1.9 0.57
1,2,3,6,7,8-hexaCDF 0.73 0.58 0.57 0.85 < 0.3 0.73 2.1 0.41
2,3,4,6,7,8-hexaCDF 0.65 0.48 0.53 0.68 ND 0.64 1.8 0.37
1,2,3,4,7,8-hexaCDD 0.33 0.34 < 0.3 < 0.3 < 0.3 0.36 0.33 < 0.3
1,2,3,6,7,8-hexaCDD 1.3 1.2 1.0 0.82 0.32 1.7 1.9 0.80
1,2,3,7,8,9-hexaCDD < 0.3 0.34 0.36 0.39 ND 0.55 0.48 < 0.3
1,2,3,4,6,7,8-heptaCDF^b < 0.5 < 0.5 < 0.5 < 0.5 < 0.5 < 0.5 < 0.5 < 0.5
1,2,3,4,6,7,8-heptaCDD^b < 2 < 2 < 2 < 2 < 2 < 2 < 2 < 2
octaCDD^b <10 <10 <10 <10 <10 <10 <10 <10
octaCDF^b < 1 < 1 < 1 < 1 < 1 < 1 < 1 < 1

^a From: Beck et al. (1987).
^b Not significantly higher than blanks.
ND = not detectable.

Table 38. PCDF and PCDD content of bovine milk from Switzerland. Results in ng/kg (ppt), whole milk basis^a

Compound Commercial Incinerators
1 2 3 4 5 6

2,3,7,8-tetraCDF < 0.028 < 0.035 < 0.021 < 0.022 < 0.032 < 0.028
1,2,3,7,8-pentaCDF < 0.020 < 0.022 < 0.021 < 0.020 < 0.036 < 0.032
2,3,4,7,8-pentaCDF 0.084 0.066 0.069 0.43 0.22 0.23
1,2,3,4,7,8-hexaCDF < 0.020 < 0.026 < 0.017 0.13 0.06 0.084
1,2,3,6,7,8-hexaCDF 0.028 < 0.018 < 0.021 0.19 0.095 0.059
2,3,4,6,7,8-hexaCDF < 0.020 < 0.018 ND 0.28 0.12 0.049
(< 0.02)
1,2,3,4,6,7,8-heptaCDF < 0.12 ND ND 0.49 0.28 < 0.18
(< 0.13) (< 0.08)
octaCDF < 0.20 ND ND ND ND < 0.52
(< 0.13) (< 0.09) (< 0.16) (< 0.21)

2,3,7,8-tetraCDD ND ND ND 0.049 0.038 0.021
(< 0.012) (< 0.013) (< 0.013)
1,2,3,7,8-pentaCDD ND ND ND 0.25 < 0.086 ND
(< 0.04) (< 0.08) (< 0.06) (< 0.1)
1,2,3,4,7,8-hexaCDD < 0.068 ND ND 0.23 0.14 < 0.14
(< 0.1) (< 0.06)
1,2,3,6,7,8-hexaCDD < 0.068 ND ND 0.29 0.16 < 0.21
(< 0.1) (< 0.06)
1,2,3,7,8,9-hexaCDD < 0.068 ND ND 0.17 < 0.080 < 0.11
(< 0.1) (< 0.06)
1,2,3,4,6,7,8-heptaCDD < 0.064 < 0.066 < 0.064 0.26 < 0.095 0.42
octaCDD < 0.16 < 0.26 < 0.12 0.28 < 0.16 0.59

^a From: Rappe et al. (1987b).
ND = not detectable.

Table 39. Levels of PCDDs and PCDFs found in human milk (ng/kg of fat weight).

Sweden FRG FRG FRG Netherlands Yugoslavia
n=4^a n=5^b n=92^c n=30^d n=3^e n=2^e

2,3,7,8-tetraCDD 0.6 1.9 < 5 3.4 9.7 < 1.0
1,2,3,7,8-pentaCDD 6.5 12.9 10.7 15 44 5.5
1,2,3,4,7,8-hexaCDD 2.5 4.6 8.1 12 25 3.5
1,2,3,6,7,8-hexaCDD 19 17.3 32.7 59 251 15
1,2,3,7,8,9-hexaCDD 6.3 1.6 6.4 11 23 ND
1,2,3,4,6,7,8-heptaCDD 59.5 72.8 49.9 61 130 106
octaCDD 302 434 181 530 744 106

2,3,7,8-tetraCDF 4.2 5.4 2.6 2.5 2.8 < 1.0
1,2,3,7,8-pentaCDF < 1 < 1 1.8 < 1 ND ND
2,3,4,7,8-pentaCDF 21.3 36.4 22.9 20 79 25
1,2,3,4,7,8-hexaCDF 4.7 11.4 8.2 8.5 8.9 3.7
1,2,3,6,7,8-hexaCDF 3.4 10.2 6.6 7.8 10.3 3.6
2,3,4,6,7,8-hexaCDF 1.4 4.3 3.3 3.0 6.4 1.3
1,2,3,4,6,7,8-heptaCDF 7.4 9.2 6.4 8.5 39 ND
octaCDF 3.2 2.4 22.8 < 3 ND ND

ND = not detected; NA = not analyzed; n = number of samples.
^a Rappe (1985).
^b Rappe et al. (1984).
^c Furst et al. (1987).
^d Beck et al. (1987).
^e Rappe et al. (1987) (pooled samples)

Kunita et al. (1984) studied the blood levels of PCDFs in people
involved in the Japanese and Taiwanese episodes. They used a
non-isomer-specific analytical method, and reported higher blood
levels in persons with severe dermal symptoms than in persons with
light symptoms. The levels 2 years after exposure were lower than 0.5
year after exposure; for the six persons studied the levels of total
PCDFs decreased on average by 57% ± 12% (Kunita et al., 1984).

The rate of excretion of these toxic PCDF isomers is very low.
Rappe & Nygren (1984) could detect 2,3,4,7,8-pentaCDF in blood plasma
from Yusho patients when the samples were collected 11 years after
exposure. However, higher levels were found in blood from Yu-cheng
patients one year after exposure; these analyses also showed a 15-20%
reduction in levels in one year (Rappe, 1984).

6. KINETICS AND METABOLISM OF 2,3,7,8-TETRACHLORO-
DIBENZO-P-DIOXIN (TCDD) AND OTHER PCDDs

6.1 Uptake, Distribution, and Excretion

Most of the available toxicokinetic data arise from studies of
gastrointestinal exposure and oral or intraperitoneal administration.
There has been one dermal study on rats, but no studies on exposure
via the respiratory tract. Data on the gastrointestinal absorption,
distribution, and elimination of TCDD in various species are
summarized in Tables 40 and 41. Principle organ depots in all species
studied are the liver and adipose tissue. In addition skin and muscle
have been found to be organ depots in monkeys, and skin in
guinea-pigs. In all species studied clearance of TCDD from the body
follows apparent first-order kinetics.

6.1.1 Studies on rats

Male Sprague Dawley rats were dosed by gavage with 14C TCDD in
acetone: corn oil (1:9), at a concentration of 50 µg/kg body weight
(Piper et al., 1973). Rats lost weight and their physical condition
was generally poor, although no deaths occurred. Approximately 30% of
the administered dose of radioactivity was eliminated in the faeces
during the first 48 h, this being most probably unabsorbed TCDD. The
total faecal TCDD content was equivalent to 53.2% of the dose over a
21-day period. During this time 13.2% was eliminated in the urine and
3.2% in the air. From these results the half-time for the clearance of
TCDD was calculated as 17.4 (± 5.6) days. In another study, groups of
two to three rats were killed at different time intervals after
administration of the same dose of ¹⁴C-TCDD. The liver contained
3.2, 4.5, and 1.3% of the administered dose per gram of tissue 3, 7,
and 21 days, respectively, after dosing. The concentrations in adipose
tissue at the same time intervals were 2.6, 3.2, and 0.4% of the dose
per gram of tissue. Other tissues showed lower concentrations.

In a study by Allen et al. (1975), male Sprague Dawley rats were
given ¹⁴C-TCDD in a single dose of 50 µg/kg body weight by stomach
tube. Groups of five rats were sacrificed 1, 3, 5, 7, 14, and 21 days
after dosing. The dose given resulted in marked liver hypertrophy,
thymic regression, weight loss, and death in 50% of the animals within
25 days. Twenty-five percent of the dose was eliminated within the
first 3 days through the faeces. During the next 18 days, 1 to 2% of
the dose was found in the faeces daily. The total amount in the faeces
during the 21 days following administration was about 52%.

Table 40. Gastrointestinal absorption of TCDD

Species/Strain Vehicle Dose % Absorption Reference
(µg/kg body weight) (mean ± SD)

Rats
Sprague Dawley acetone:corn oil (1:9) 50^a 70 Piper et al. (1973)
Sprague Dawley corn oil 50^a > 75 Allen et al. (1975)
Sprague Dawley diet 7 or 20; for 42 days 50-60 Fries & Marrow (1975)
Sprague Dawley acetone:corn oil (1:24) 1^a 84±11 Rose et al. (1976)
Sprague Dawley acetone:corn oil (1:24) 0.1 or 1.0; 5 days/week 86±12 Rose et al. (1976)
for 7 weeks

Mice
ICR/Ha Swiss ethanol:Tween 80:saline 135^a 27-28 Koshakji et al. (1984)
(1:10:89)

Hamsters
Golden Syrian olive oil 650^a 73.5±22.8 Olson et al. (1980a)

^a Single dose.

Table 41. Elimination of TCDD in different species

Species/Strain Route/Vehicle Dose^a Duration Half-life Eliminated radioactivity Reference
(µg/kg of for elimination (% of administered dose)
body study (days)
weight) (days) Faeces Urine

Rats

Sprague Dawley oral/acetone: 50 21 17.4 ±5.6 53.2 13.2 Piper et al. (1973)
corn oil (1:9)
Sprague Dawley oral/corn oil 50 21 21.3 ±2.9 53.3 4.5 Allen et al. (1975)
Sprague Dawley oral/acetone: 1 22 31 ± 6 NR ND Rose et al. (1976)
corn oil (1:24)
Sprague Dawley oral/acetone: 1^b 49 23.7 NR 3.1 ±0.2 (m) Rose et al. (1976)
corn oil (1:24) 12.5 ±5.1 (f)
Sprague Dawley ip/corn oil 400 7 NR 4.96 ±0.3 0.51 ±0.05 van Miller et al.
(1976)
Mice
C57BL/6 ip/olive oil 10 30 11 59.3 20.5 Gasiewicz et al.
(1983b)
DBA/2 ip/olive oil 10 30 24.4 39.8 16.8 Gasiewicz et al.
(1983b)
B6D2F₁ ip/olive oil 10 30 12.6 56.3 20.5 Gasiewicz et al.
(1983b)
ICR/Ha Swiss oral/ethanol: 135 11 20 78 4 Koshakji et al.
tween 80: (1984)
saline (1:10:89)
C57BL/6 Ah^b/Ah^d ip/emulphor: 0.5 42 9.6 61.8 19.9 Birnbaum (1986)
ethanol:H₂O
(1:1:18)
C57BL/6 Ah_d/Ah_d ip/emulphor: 0.5 42 9.6 55.9 26.8 Birnbaum (1986)
ethanol:H₂O
(1:1:18)
DBA/2 Ah_b/Ah_d ip/emulphor: 0.5 42 10.8 71.5 13.6 Birnbaum (1986)
ethanol:H₂O
(1:1:18)

Table 41 (contd).

Species/Strain Route/Vehicle Dose^a Duration Half-life Eliminated radioactivity Reference
(µg/kg of for elimination (% of administered dose)
body study (days)
weight) (days) Faeces Urine

Mice (contd)

DBA/2 Ah^d/Ah^d ip/emulphor: 0.5 42 10.8 76.1 11.1 Birnbaum (1986)
ethanol:H₂O
(1:1:18)

Guinea-pigs

Hartley ip/olive oil 2 23 33 2 Gasiewicz &
Neal (1979)
NR oral/NR NR 22 22-43 ND ND Nolan et al. (1979)
Hartley ip/olive oil 0.56 45 93.7 ± 15.5 26.2 ±3.6 3.1 ±1.2 Olson (1986)

Hamsters

Golden Syrian oral/olive oil 650 35 15.0 ±2.5 NR NR Olson et al.
(1980a)
Golden Syrian ip/olive oil 650 35 12.0 ±2.0 50.0±2.7 34.6±5.4 Olson et al.
(1980a)
Monkeys

Rhesus (adult) ip/corn oil 400 7 NR 3.75 1.06 van Miller et al.
(1976)
Rhesus (infant) ip/corn oil 400 7 NR 1.26 2 van Miller et al.
(1976)

^a Single dose, unless otherwise stated.
^b 5 days/week for 7 weeks.
ND = not detectable, NR = not reported, m = males, f = females, ip = intraperitoneal.

The authors concluded that the faecal content during the first 3
days represented mainly unabsorbed TCDD and that apparently more than
75% of the administered dose had been absorbed from the
gastrointestinal tract. The radioactivity excreted daily through the
urine ranged between 0.1 and 0.2% of the administered dose during the
initial 12 days. Thereafter, the daily urinary radioactivity excretion
increased from 0.25% of the administered dose to 0.43% by day 21. The
total amount of ¹⁴C excreted through the urine over a 21-day period
was about 4.5% of the dose. On the basis of the daily faecal and
urinary excretion, the authors calculated a half-time of 21.3 (± 2.9)
days. In the rats killed on days 1, 3, 5, 7, 14, and 21 after
administration, the total liver content was 56 (± 5), 54 (± 14), 54 (±
6), 54 (± 8), 45 (± 5), and 24 (± 4)% of the administered dose,
respectively. At all intervals, the levels found in the liver exceeded
those found in other organs. On days 5, 7, and 14 about 90% of the
total radioactivity in the liver was present in the microsomal
fraction.

Fries & Marrow (1975) fed Sprague Dawley male and female rats a
diet containing 7 or 20 µg ¹⁴C-TCDD/kg diet for 42 days. Thereafter
all rats received the control diet for another 30 days. Two animals of
each sex and TCDD dietary level were sacrificed at 14-day intervals.
This treatment resulted in decreased food consumption, decreased
weight gain, and increased relative liver weight among both males and
females. The concentration of TCDD-derived radioactivity in the liver,
which was the principal tissue depot of both males and females, was
directly proportional to the dietary intake of ¹⁴C-TCDD. At the end
of the 42-day feeding period, the ¹⁴C levels in male livers
indicated TCDD contents of 5.8 and 15.9 µg/kg of tissue for the lower
and higher feed concentrations, respectively. The concentrations in
the liver of female rats were similar. Analysis of the liver of rats
killed 14 and 30 days after discontinuing TCDD exposure indicated a
gradual decrease in TCDD liver concentration in both sexes. The
steady-state body burden for TCDD was estimated to be 10-11 times the
daily intake in both sexes. The whole-body and liver half-lives were
calculated to be 12 and 11 days in males and 15 and 13 days,
respectively, in females.

Rose et al. (1976) estimated the absorption of a single non-toxic
dose of 1 µg ¹⁴C-TCDD/kg body weight in male and female Sprague
Dawley rats to be 84 ± 11% of the administered dose. The elimination
of TCDD was followed for 22 days after administration and the faecal
excretion accounted for most if not all of the elimination of TCDD
and/or its metabolites. No radioactivity was detectable in urine and
expired air. Twenty-two days after dosing, mean values of 1.26% and
1.25% of the dose/g liver and adipose tissue, respectively, were
found. Much lower ¹⁴C-activities were found in the thymus, kidney,
and spleen, namely 0.09, 0.06, and 0.02% of the dose/g, respectively.

The whole-body half-life was estimated to be 31 days. Rose et al.
(1976) also followed the fate of 5 daily doses per week of 0.01, 0.1,
and 1.0 mg ¹⁴C-TCDD/kg body weight given for 1, 3, and 7 weeks to
male and female Sprague Dawley rats (Table 42).

According to Kociba et al. (1976), a dose level of 0.01 µg/kg per
day, 5 days/week for 13 weeks, produced no overt toxic effects,
whereas 0.1 or 1.0 µg/kg per day produced adverse effects, including
some deaths in the high dose group. In the study by Rose et al.
(1976), the dose level of 0.01 µg/kg per day resulted in no detectable
¹⁴C, except in liver, fat, and excreta. Thus no kinetic calculations
on a truly non-toxic dose could be performed. Pooling the results for
all rats receiving 0.1 or 1.0 µg/kg per day, the absorbed dose
corresponded to 86 ± 12% of the administered dose, with an individual
variation of 66 to 93%. The overall rate constant for elimination
corresponded to a half-time of 23.7 days, with an individual variation
of 16-37 days. The radioactivity was eliminated primarily in the
faeces, but the percentage of the dose excreted in the urine compared
to that eliminated in the faeces tended to increase with time. At the
dose level of 1.0 µg/kg per day, males excreted in the urine 3.1 (±
0.2)% and females 12.5 (± 5.1)% of the cumulative dose over 7 weeks of
exposure. The one female rat that died during the 7th week excreted
17.8% of the cumulative dose in the urine. Exhaled air was not
examined in these studies. After 7 weeks of exposure, the average body
burdens were 47.7 (± 8.8) and 37.1 (± 7.5)%, respectively, of the
administered dose for the rats given 0.1 and 1.0 mg TCDD/kg per day.
The main tissue depots of ¹⁴C were the liver and adipose tissue.
Radioactivity was also detected in thymus, kidney, and spleen at
levels between 1 and 2% of that in liver. Direct chemical
determination of liver samples confirmed the TCDD-concentrations
calculated by radioactivity measurements. TCDD and/or its metabolites
approached a steady-state body burden calculated to be 21.3 D_o for
rats given a daily dose (D_o, µg/kg body weight) on 5 consecutive
days per week for an infinite number of weeks, within 13 weeks over
the dose range 0.01 to 1.0 µg/kg per day. It was thought unlikely that
the dietary intake of extremely low levels of TCDD would result in the
accumulation of toxic amounts in the rat.

The excretion and distribution of a toxic dose of 400 µg
³H-TCDD/kg body weight was followed in male Sprague Dawley rats (Van
Miller et al., 1976). Within 7 days, about 5.0% of the dose had been
excreted in faeces and 0.5% in urine. The principal tissue depots for
radioactivity were liver, muscle, and skin, which contained 43.0, 4.6,
and 4.4% of the administered dose, respectively.

Table 42. Tissue distribution of TCDD-derived radioactivity in rats given
oral doses of 14_C-TCDD^a

Tissue content of 14_C (µg equivalents of TCDD/kg tissue)^b

Tissue 1 week 3 weeks 7 weeks

Exposure level: 1.0 µg/kg body weight per day

liver 49.5±3.6 110.2±37.1 204.0±52.2
adipose tissue 10.0±3.0 23.5±7.5 61.4±36.7
thymus 0.9±0.2 7.3±4.5 6.9±2.5
kidney 0.9±0.2 1.9±0.9 5.5±5.1
spleen 0.4±0.1 1.6±0.8 1.9±1.1

Exposure level: 0.1 µg/kg body weight per day

liver 3.9±1.1 11.8±2.2 19.8±3.1
adipose tissue 0.9±0.6 2.7±0.8 4.5±0.7
thymus ND 1.0±0.6 0.6±0.2
spleen ND 0.6±0.4 0.3±0.1
kidney ND ND ND

Exposure level: 0.01 µg/kg body weight per day

liver ND 0.8±0.1 1.6±0.5
adipose tissue ND 0.3 (2)^c 0.3±0.1 (4)^c
thymus ND 0.6±0.1 ND
spleen ND 0.6±0.3 ND
kidney ND ND ND

^a From: Rose et al. (1976).
^b The mean ± standard deviation of 3 male and 3 female rats.
^c Indicates the number of animals with detectable levels of
14_C-activity.
ND = not detected.

Kociba et al. (1976) gave Sprague Dawley rats TCDD in daily doses
of 0.01, 0.1, and 1 µg/kg body weight 5 days per week for 13 weeks by
gavage. The liver contained TCDD at a level of 324 (± 53) µg/kg wet
weight in males and 284 (± 21) µg/kg wet weight in females given
repeated TCDD doses of 1 µg/kg body weight per day. For the dose level
of 0.1 µg/kg body weight per day, the liver TCDD levels were 36 (± 4)
and 35 (± 4) µg/kg wet weight for males and females, respectively. A
dose of 0.01 µg/kg body weight per day resulted in TCDD liver levels
of 2.6 ( 0.6) µg/kg wet weight in males and 3.7 (± 0.4) µg/kg wet
weight in females.

After feeding diets with TCDD levels corresponding to daily
dietary intakes of 0.001, 0.01, or 0.1 µg/kg body weight for 2 years,
the average concentrations of TCDD found in the liver of female
Sprague Dawley rats were 0.54, 5.1, and 24.0 µg/kg wet weight,
respectively (Kociba et al., 1978). The corresponding levels in
adipose tissue were 0.54, 1.7, and 8.1 µg/kg wet weight. A comparison
of liver TCDD levels found in rats given comparable daily doses of
TCDD for 13 weeks (Kociba et al., 1976) or 2 years (Kociba et al.,
1978) indicates that with prolonged exposure the liver TCDD content
reaches a plateau. This finding agrees well with the prediction
obtained by mathematical analysis of the data resulting from
experiments involving exposure of a few weeks (Rose et al., 1976).

The proportion of a single oral dose of ³H-TCDD found in the
liver of female Sprague Dawley rats was dependent both on the dose
level and on the vehicle used (Poiger & Schlatter, 1980). Maximal
retention occurred within 48 h after dosing. Increasing retention was
observed up to a dose of 280 ng TCDD/rat. Hepatic retention 24 h after
dosing was higher (36.7% of the dose) if TCDD was given in 50% ethanol
than if it was given as an aqueous suspension of soil (37%, w/w),
where it was 16-24.1% of the dose, or activated carbon (25%, w/w),
where it was < 0.07% of the dose (see section 7.4).

Biliary excretion of radioactivity originating from 3H-labelled
TCDD occurred at a more or less constant rate of 0.5 to 1% of the
administered dose per day for 12 days following the administration of
100 µg TCDD/kg body weight in a female Sprague Dawley rat (Poiger &
Buser, 1983). No severe toxic effects were observed within that time
period.

McConnell et al. (1984) dosed female Sprague Dawley rats orally
with either pure TCDD in corn oil or amounts of contaminated soil
giving similar doses of TCDD. In general TCDD in soil was as potent an
inducer of aryl hydrocarbon hydroxylase (AHH) as pure TCDD in corn
oil. The hepatic concentration of TCDD was 40.8 µg/kg in the corn oil
group, receiving 5 µg TCDD/kg body weight, and 20.3 µg/kg in the soil
group, receiving 5.5 µg TCDD/kg body weight (see section 7.4).

Poiger & Schlatter (1980) studied the dermal absorption of
³H-TCDD in hairless rats of the Naked ex Back-Cross and Holzman
strain (200-250 g). Using the amount of TCDD-derived radioactivity
found in the liver as an indicator of its absorption, they reported
that the permeation of TCDD across the epidermis was highly dependent
on the formulation used. The highest radioactivity in the liver, 14.8%
of the administered dose, was detected when TCDD was applied as a
methanolic solution. The hepatic recovery of the administered dose
observed when TCDD was applied in polyethylene glycol 1500 with and
without 15% water was 14.1 and 1.4%, respectively. Dermal application
of TCDD in vaseline or adsorbed onto soil or activated carbon
decreased the percentage of the dose recovered in the liver to 1.4,
1.7-2.2, and < 0.05%, respectively.

In studies by van den Berg et al. (1983), fly ash and crude or
purified toluene extracts of PCDD- and PCDF-containing fly ash from a
municipal incinerator were mixed with ordinary laboratory diet for
rats. Small portions (2 g) of these diets were fed to male Wistar rats
(300 g) every 24 h for 19 days, at which time the animals were
sacrified. Tetra-, penta-, and hexa-chlorinated PCDDs and PCDFs in the
liver and adipose tissues of these rats were determined. Rats fed the
fly ash containing diet stored PCDDs and PCDFs in their livers at
concentrations that were at least 3 to 5 times lower than in the case
of rats fed comparable amounts of fly ash extracts (for the pentaCDD,
hexaCDF, and hexaCDD isomers, the concentrations were approximately
10-20 times lower). Generally PCDFs showed a higher retention in rat
liver than did the corresponding PCDDs. In the adipose tissue of rats
fed with fly ash extracts, retention was higher for penta- and
hexaCDDs than for the corresponding PCDFs.

In further fly ash studies, male Wistar rats (275 g) were fed for
up to 99 days a diet that included 2.5% HC1-pretreated fly ash
(containing PCDDs and PCDFs) from a municipal incinerator (van den
Berg et al., 1986a). A control group received standard diet. All
congeners retained in the liver of the rats had a 2,3,7,8-chlorine
substitution pattern. With the exception of 2,3,4,7,8-pentaCDF and
2,3,4,6,7,8-hexaCDF, liver retention for each congener was below 10%
of the group dose. The retention percentages of the various congeners
in the liver were almost equal at the time-points studied (34, 59, and
99 days), thus indicating a long half-life of these congeners in rat
liver.

Male Wistar rats fed 22.7 (± 1) µg or 120.7 (± 2.8) µg octaCDD
over a two-week period were found to retain about 1-2% of the given
dose in the liver (Williams et al., 1972). The heart, kidneys, spleen,
lung, skeletal muscle, testes, and urine contained no detectable
levels of octaCDD, but minor amounts were found in the adipose tissue

of the high-dose group. Faeces contained 61% and 37%, respectively, of
the low and high dose given. The presence of a large quantity of
octaCDD in the faeces compared to that in the bile 24-72 h after a
single oral dose of 58 mg octaCDD to bile-cannulated male Wistar rats
(400 g) indicated that the dioxin present in the faeces was mainly
unabsorbed octaCDD (Williams et al., 1972).

After 21 daily doses of 100 mg octaCDD containing 12.6 pg
³⁵S-thio-heptaCDD to male Sprague Dawley rats, the radioactivity was
mainly recovered in the faeces and urine, the percentages of the
ingested radioactive dose being 93 (± 6) and 5.2 (± 0.8)% respectively
(Norback et al., 1975). The high faecal excretion suggests poor
absorption. Of the radio-active body burden, 50% was contained in the
liver. The microsomal fraction contained 96.3 (± 8.2)% of the hepatic
radioactivity.

6.1.2 Studies on mice

In studies by Vinopal & Casida (1973), male white mice (20 g)
were given tritium-labelled TCDD intraperitoneally in a single dose of
130 µg/kg body weight. Three days after TCDD administration to one
mouse, 13% of the administered tritium was recovered in the faeces and
0.3% in the urine, while 32% was found in the liver and 0.3% in the
kidneys. In another study, groups of two to six mice were killed at
various time intervals after similar treatment with the same dose of
³H-TCDD. One and 4 days after dosing, the liver contained about 15%
of the dose, on the 8th day, 26%, on the 11th day, 22%, and on the
15th and 20th days, about 10%. The highest amount of ³H-activity was
found in the microsomal fraction of the liver. Somewhat lower activity
was detected in the mitochondrial fraction and the nuclei. The
supernatant fraction was practically devoid of any radioactivity. On
day 8 when the highest levels were observed, the whole liver
homogenate contained 26.7 (± 4.8)% of the administered dose, the
microsomes 12.6 (± 3.8)%, the nuclei 7.8 (± 1.2)%, the mitochondria
6.2 (± 1.3)%, and the supernatant fraction 0.1 (± 0.0)% of the
administered dose.

Coccia et al. (1981) described the effect of adding different
substances to food on the persistence of TCDD in the liver of male
C57Bl/6 mice. In one of the experiments, the test diet was given
immediately after the administration of a single oral dose of 7.6 µg
³H-TCDD/kg body weight. The hepatic radioactivity 14 days after
dosing was 17.3, 6.3, 13.1, and 14.5% of the administered dose in
animals fed standard chow containing 5% vegetable charcoal, 0.5%
cholic acid, and 4% cholestyramine, respectively. When feeding of the
test diet started 3 days after dosing, the hepatic retention of
³H-TCDD was decreased to a similar extent.

Faecal and urinary excretion (Table 41), along with the formation
of faecal, biliary, and urinary metabolites (see 7.2.1.1), of
³H-TCDD was studied in male C57BL/6, DBA/2, and B6D2F1 mice after a
single intraperitoneal dose of 10 µg ³H-TCDD/kg body weight
(Gasiewicz et al., 1983b). The principal tissue depot in C57BL/6 and
B6D2F1 mice was the liver, followed by the adipose tissue. Most other
tissues examined contained less than 1% of the administered dose. In
DBA/2 mice, the adipose tissue contained more radioactivity than the
liver. This difference may be due to the fact that these three strains
of mice differ in their adipose tissue content, being 5.9, 11.5, and
5.0% of the body weight in C57BL/6, DBA/2, and B6D2F1 mice,
respectively. The estimated half-lives of clearance of 3H-TCDD from
the liver of C57BL/6, DBA/2, and B6D2F1 mice were 17, 27, and 13 days,
respectively, and the corresponding figures for the half-life in the
adipose tissue were 11, 42, and 11 days. The cumulative faecal
elimination 30 days after dosing was 59.3, 39.8, and 56.3% of the
administered dose in C57Bl/6, DBA/2, and B6D2F1 mice, respectively,
and the corresponding figures for urinary elimination were 20.5, 16.8,
and 20.5%.

Birnbaum (1986) studied in mice the distribution and excretion of
a single intraperitoneal dose of 500 ng (45 µCi)³H-TCDD/kg body
weight for up to 42 days after treatment. Two sets of congenic strains
of mice were used, i.e., male C57Bl/6 and female DBA/2 mice, where
within each congenic pair the mice differed only at the Ah locus (or
at a limited number of genes closely linked to the Ah locus). The mice
were bred and phenotyped by zoxazolamine paralysis time. The results,
some of them summarized in Tables 47 and 48, suggested that, at the
dose level studied, the distribution and excretion of TCDD were
primarily governed by the total genetic background rather than by the
allele present at the Ah locus.

When male ICR/Ha Swiss mice (27 - 35 g) were given a single oral
dose of 135 µg ¹⁴C-TCDD/kg body weight, about 71% and 1-2%,
respectively, of the administered dose was eliminated via faeces and
urine within the first 24 h (Koshakji et al., 1984). During the
following 10 days, an additional 7% and 2% of the administered
radioactivity were recovered in faeces and urine, respectively. Based
on the estimated body burden of radioactivity, a whole-body half-life
of 20 days was calculated.

The distribution of ³H-TCDD in the skin of hairless (SKH:HR-1)
mice after a single intraperitoneal dose of 6.3 µg TCDD/kg body weight
was examined for up to 14 days (Puhvel et al., 1986). Most of the
³H-TCDD in skin was localized in the dermis, although the
concentration of ³H-TCDD was consistently higher in the epidermis.

6.1.3 Studies on guinea-pigs

The retention of a single intraperitoneal dose of 2 µg
¹⁴C-TCDD/kg body weight in various tissues of male Hartley
guinea-pigs was determined 1, 3, 5, 7, 11, and 15 days after exposure
(Table 43) (Gasiewicz & Neal, 1979). Three animals died and all
animals lost 24 to 35% of the body weight during the study. The
highest amount of radioactivity per tissue was found in the liver and
skin. The radioactivity in the liver increased with time, concomitant
to the depletion of adipose tissues. Radioactivity in the skin
decreased also with time. No signs of toxicity were seen when the
cumulative excretion of a single i.p dose of 0.5 µg ³H-TCDD/kg body
weight in male Hartley guinea-pigs was studied (Gasiewicz & Neal,
1979). The faecal and urinary excretion of radioactivity was linear
throughout the 23-day study. Approximately 1.4% of the administered
dose was excreted daily during that period, and the faeces contained
94% of the excreted radioactivity.

The microsomal fraction of the liver in male Hartley guinea-pigs
contained 40.7 to 47.4% of the hepatic radioactivity 1 day after a
single ip dose of 0.3, 2.0, or 7.0 µg of ³H- or ¹⁴C-TCDD/kg body
weight (Gasiewicz & Neal, 1979). Corresponding values for the crude
nuclear fraction, the mitochondrial fraction, and the soluble fraction
were 20.1-35.6%, 9.5-12.9%, and 7.6-26.4%, respectively. The
subcellular distribution was similar 1 and 6 days after the low dose
but, following the high dose, more radioactivity was present in the
microsomal fraction and less was recovered in the crude nuclear and
soluble fractions on day 6 after exposure.

Olson (1986) followed the distribution, elimination and
metabolism (see section 6.2.1.1) of a single ip dose of 0.56 µg
³H-TCDD/kg body weight in adult (335 to 625 g) male Hartley
guinea-pigs for 45 days. One of seven animals died on day 27, but the
remaining animals gained weight and exhibited no gross signs of
toxicity. At termination the body composition was normal, and 61% of
the administered radioactivity was recovered in the twelve
investigated tissues at the end of the study. The adipose tissue
contained 36% of the dose; liver, pelt, and skeletal muscle plus
carcass contained each 7% of the dose; the gastrointestinal tract
contained about 2% of the dose, and remaining tissues contained less
than 0.5% of the dose. Urinary and faecal elimination followed
apparent first-order kinetics, with half-lives of 82.5 (± 22.4) and
94.4 (± 14.7) days, respectively.

Table 43. Tissue content of TCDD-derived 14_C (% of dose/g tissue)^a in guinea-pigs following a single
dose of 14_C-TCDD^d

Days after exposure

Tissue 1 3 5 7 11 15

Perirenal adipose 3.2±1.0 4.1±0.4 2.1±0.4 1.3±0.2 2.1±0.2
Epididymal adipose 1.5±0.8 3.8±0.5 3.4±0.7 3.2±0.1 3.9^b 2.5±1.1
Adrenal 1.4±0.3 1.4±0.2 0.9±0.1 1.2±0.3 2.1±0.9 1.7±0.2
Liver 1.1±0.4 1.5±0.4 1.3±0.2 1.1±0.2 2.2±0.2 3.2±0.3
Liver^c 11.4±3.3 15.5±3.3 14.0±2.3 12.0±1.9 21.2±2.3 29.6±2.7
Spleen 0.7±0.3 0.5±0.3 0.2±0.1 0.4±0.2 0.4±0.2 0.5±0.1
Duodenum 0.4±0.2 0.2±0.1 0.2±0.1 0.2±0.1 0.2±0.1 0.3±0.1
Pancreas - - 0.2±0.1 0.5±0.3 0.4±0.3 0.3±0.1
Stomach 0.2±0.1 0.3±0.1 0.1±0.1 0.2±0.1 0.3±0.1 0.3±0.1
Testes 0.2±0.1 0.3±0.1 0.2±0.1 0.3±0.1 0.3±0.1 0.2±0.1
Kidneys 0.3±0.1 0.3±0.1 0.2±0.1 0.4±0.1 0.8±0.4 0.7±0.1
Bone marrow 0.3±0.1 0.5±0.1 0.2±0.1 0.4±0.1 0.4^b 0.2±0.1
Lungs 0.3±0.1 0.2±0.1 0.2±0.1 0.4±0.1 0.5±0.2 0.6±0.1
Skin^c 13.8±0.7 16.3±0.3 15.8±2.4 6.5±0.8 6.5±0.7 6.7±0.6
Brain, heart,
skeletal muscle <0.25

^a Mean ± standard error for three animals, unless indicated otherwise.
^b Mean of two animals.
^c Percentage of dose/tissue.
^d From: Gasiewicz & Neal (1979).

6.1.4 Studies on hamsters

In studies by Olson et al. (1980a), Golden Syrian hamsters
absorbed about 73.5% of a single oral dose of 650 µg ³H-TCDD/kg body
weight, a dose that produced thymic atrophy and body weight loss in
several of the animals. The distribution of radioactitivity in various
tissues 1, 3, 10, and 20 days after administration is given in Table
44. The principal depots were the liver and adipose tissue. A similar
pattern of distribution was obtained when the same dose was given
intraperitoneally. The elimination of radioactivity in faeces and
urine was followed for 35 days after a single intraperitoneal (ip) or
oral dose of 650 µg/kg body weight (Olson et al., 1980a). The
half-life for elimination was 12.0 (± 2.0) days and 15.0 (± 2.5) days
for the ip and oral routes, respectively. Of the excreted
radioactivity 41% occurred in urine and 59% in faeces.

The hepatic retention of PCDDs and PCDFs from dietary intake of
HC1-pretreated fly ash from a municipal incinerator was studied in
male Golden syrian hamsters (van den Berg et al., 1986b). The livers
were analysed for tetra-, penta-, and hexaCDDs and PCDFs after feeding
the diet, which contained 25% fly ash, for 34, 58, and 95 days. No
detectable hepatic retention was observed after 34 days. The highest
retention after 95 days was 8.4% for 2,3,4,7,8-pentaCDF, but the
retention was generally below 5% of the total dose. With the exception
of 2,3,4,6,7-pentaCDF, only 2,3,7,8-substituted PCDDs and PCDFs were
retained. Constant relative concentrations were found for the
2,3,7,8-substituted PCDDs and PCDFs at the time points studied.

6.1.5 Studies on monkeys

Van Miller et al. (1976) gave three adult female rhesus monkeys
and four male infant rhesus monkeys a single intraperitoneal dose of
400 µg 3H-TCDD/kg body weight in corn oil. This dose resulted in a
loss of body weight: 10.8% for adults and 20.7% for infants, and light
microscopic changes in the liver. Over a 7-day period the adult
monkeys excreted 1.06% of the dose in the urine and 3.75% in the
faeces. During the same period the infant monkeys excreted
approximately 2% of the administered dose in urine and about 1.26% in
the faeces. The authors questioned the accuracy of these figures owing
to the difficulty of separating the two types of excreta from infant
monkeys. The total tissue concentrations of radioactivity 7 days after
dosing are given in Table 45. The principal tissue depots of
radioactivity in adult monkeys were adipose tissue, skin, liver, and
muscle; they contained 16.2, 13.1, 10.4, and 8.6% of the administered
dose, respectively. The distribution in infant monkeys was 35.6% in
muscle, 22.7% in skin, and only 4.5% of the dose in the liver.

Table 44. Tissue distribution of TCDD-derived radioactivity in Golden
Syrian hamsters at 1, 3, 10, and 20 days following a single oral
dose of 650 µg ³H-TCDD/kg body weight^a

Tissue content of ³H (% of dose/g tissue)^b

Tissue Day 1 Day 3 Day 10 Day 20

Liver 4.03±1.00 5.32±.82 3.19±.93 0.86±0.09
Liver^b 12.74±3.21 20.44±3.45 9.69±0.99 3.70±0.29
Perirenal adipose 2.93±0.87 3.48±0.56 1.38±0.28 0.32±0.03
Adrenals 1.56±0.52 1.12±0.14 0.47±0.08 0.10±0.01
Pancreas 0.39±0.20 0.61±0.13 0.62±0.26 0.21±0.04
Kidneys 0.60±0.16 0.64±0.11 0.60±0.32 0.12±0.03
Spleen 0.30±0.08 0.24±0.05 0.43±0.26 0.07±0.02
Thymus 0.49±0.14 0.34±0.11 - 0.05±0.02
Skin 0.84±0.26 0.31±0.07 0.56±0.18 0.03±0.01
Stomach 0.34±0.07 0.55±0.09 0.65±0.39 0.16±0.06
Duodenum 0.51±0.13 0.47±0.09 0.55±0.28 0.07±0.02
Jejunum 0.59±0.15 0.71±0.20 0.39±0.16 0.08±0.02
Ileum 0.41±0.12 0.35±0.05 0.37±0.21 0.06±0.02
Colon 0.92±0.27 0.60±0.14 0.34±0.07 0.06±0.01
Caecum 0.39±0.11 0.41±0.10 0.28±0.12 0.05±0.01
Lungs 0.38±0.09 0.37±0.05 0.41±0.25 0.07±0.03
Skeletal muscle 0.20±0.07 0.15±0.05 0.15±0.03 0.04±0.02
Heart 0.14±0.03 0.13±0.02 0.15±0.08 0.03±0.01
Testes 0.10±0.04 0.32±0.13 0.13±0.04 0.03±0.01
Blood 0.12±0.02 0.14±0.03 0.12±0.06 0.02±0.01
Brain 0.03±0.01 0.05±0.01 0.06±0.02 0.01

^a From: Olson et al. (1980a).
^b All values are the mean (± standard error) of four hamsters.
^c Percentage of dose/liver.

Table 45. Tissue distribution of TCDD-derived radioactivity in adult
and infant rhesus monkeys 7 days following a single intraperitoneal
dose of 400 µg ³H-TCDD/kg body weight^a

(% of dose/tissue)
Tissue content of 3H^b

Tissue Adult Infant

Liver 10.4±6.9 4.51±1.60
Brain 0.58±0.34 1.41±1.40
Spleen 0.028±0.013 0.026±0.004
Small intestine 0.87±0.39 1.47±0.64
Large intestine 1.29±0.12 0.64±0.24
Muscle^c 8.62±2.39 35.6±14.4
Skin 13.1±4.9 22.7±8.8
Adipose tissued 16.2±5.8

^a Fom: Van Miller et al. (1976).
^b Mean (± standard deviation) of three adults or four infants.
^c Total muscle was taken as 40% of body weight.
^d Quantities in infant monkey were insignificant. For adult
monkeys, the calculation was based on an estimate of 300 g
mesenteric fat.

The concentration of TCDD in samples of faeces, urine, and fat
were measured by GC-MS at intervals after dosing up to 715 days in an
adult female rhesus monkey (Macaca mulatta) given a single oral
dose of 1 µg TCDD/kg body weight (McNulty et al., 1982b). During the
3 months after dosing the monkey lost 50% of its body weight but then
began to gain weight again. The level of TCDD in faeces was high for
4 days and then fell to very low or undetectable levels. TCDD in urine
was very low at all time points. The apparent half-life of TCDD in
adipose tissue was about 1 year.

There were no significant time-dependent changes in TCDD-derived
radioactivity found in the tissues investigated from marmosets during
the 3-week-period following subcutaneous treatment with 5 µg
¹⁴C-TCDD/kg body weight, except for a minor decrease in levels found
in the adipose tissue (Krowke, 1986). The data thus indicate a long
half-life for TCDD in marmosets.

6.1.6 Studies on dogs

A one-year-old male beagle dog received a total dose of 5.4 µg of
TCDD enterally by direct introduction into the duodenal lumen in four
portions of 1-2 µg, with intervals of 2-7 days between treatments
(Poiger et al., 1982). Severe toxic symtoms preceeded the death of the
animal 17 days after the first dose. The excretion of radioactivity in
the bile reached a maximum on day 1 or day 2 following administration.
Significantly more biliary radioactivity was found after
administration of doses 3 and 4 than after the administration of doses
1 and 2, suggesting that TCDD-administration stimulated its own
metabolism. It was later demonstrated in a 18-months-old male boxer
that TCDD-pretreatment (10 µg/kg body weight) stimulated the biliary
excretion of ³H-labelled TCDD (32.8 ng/kg body weight), whereas
phenobarbital-pretreatment had no effect when compared to the biliary
excretion without pretreatment (Poiger & Schlatter, 1985). All the
experiments were carried out in the same animal, which had enough time
between treatments for the radioactivity to return almost to the
background level.

6.1.7 Studies on cows

The major routes for elimination of ³H-TCDD in Holstein cows
(500-650 kg) after oral doses of 0.05µg (two cows) or 7.5µg (one cow)
TCDD/kg body weight were faeces > milk > urine (Jones et al., 1987).
Fifty percent of the administered dose was eliminated in faeces, the
major part in the first few days after treatment. Three lactating
Holstein cows received commercial technical grade pentachlorophenol
orally by gelatine capsule at a dose rate of 10 mg/kg body weight
twice daily for 10 days and once daily for the following 60 days
(Firestone et al., 1979). One cow served as a control and received
gelatine capsules containing only ground corn. The pentachlorophenol
composite used contained ten PCDD congeners (0.1 to 690 mg/kg) and
eight PCDF congeners (0.9 to 130 mg/kg). Faeces collected on day 28 of
the treatment period contained three hexaCDDs (0.05 to 0.63 µg/kg),
two heptaCDDs (21.3 to 33.1 µg/kg), and octaCDD (290 to 429 µg/kg).
Faeces also contained hexa-, hepta-, and octaCDF. Milk, body fat, and
blood contained only three of the PCDD congeners present in the
pentachlorophenol composite, namely 1,2,3,6,7,8-hexaCDD,
1,2,3,4,6,7,8-heptaCDD, and octaCDD. Milk samples also contained
hexa-, hepta-, and octaCDF. The average concentrations of
1,2,3,6,7,8-hexaCDD, 1,2,3,4,6,7,8-heptaCDD, octaCDD, and octaCDF in
the composite milk fat at the end of the treatment period were 20, 40,
25, and 2 mg/kg respectively. Similar concentrations were found in
body (shoulder) fat at the end of the treatment period (13, 24, and 32
mg/kg, respectively, of 1,2,3,6,7,8-hexaCDD, 1,2,3,4,6,7,8-heptaCDD,
and octaCDD). Levels of dioxins in the blood were approximately 1000
times below the values in milk or body fat. The average daily
excretion of 1,2,3,6,7,8-hexaCDD, 1,2,3,4,6,7,8-heptaCDD, and octaCDD
in the milk during days 40-70 was about 20, 40, and 23 mg,
corresponding to 33, 3, and 0.6% of the daily intake of PCDDs. One

hundred days after the cessation of treatment, the average values for
1,2,3,6,7,8-hexaCDD, 1,2,3,4,6,7,8-heptaCDD, and octaCDD in shoulder
fat and milk fat were 2.5, 6.6, 5.6 mg/kg and 4.3, 6.9, 3.0 mg/kg,
respectively.

6.1.8 In vitro studies

The uptake of ³H-TCDD in human fibroblasts was less efficient
when the TCDD was associated with the high density lipoprotein (HDL)
than with the low density lipoprotein (LDL), and even less when
associated with serum (Shireman & Wei, 1986). From studies in mutant
human fibroblasts lacking the normal LDL cellular receptor, the
authors concluded that the LDL receptor pathway was involved in the
cellular uptake of TCDD. The uptake from LDL was time, temperature,
and concentration dependent.

6.2 Metabolic Transformation

6.2.1 Studies on mammals

6.2.1.1 In vivo studies

The possible existence of a urinary metabolite of TCDD was first
suggested by the finding of Allen et al. (1975) that the radioactivity
in urine of 14C-TCDD treated rats was highest from week 2 to 3. Later
both Poiger & Schlatter (1979) and Ramsey et al. (1982) presented
evidence for the in vivo biotransformation of TCDD in the rat.
Male Sprague Dawley rats were given two, four, or six daily oral doses
of approximately 15 µg ¹⁴C-TCDD/kg body weight, and the bile was
collected for 24 h following the last dose (Ramsey et al., 1982).
Using high pressure liquid chromatography, at least eight metabolites
of TCDD were found in the bile from these rats. Incubation of the bile
with ß-glucuronidase indicated the presence of glucuronide conjugates
among the metabolites. Poiger & Schlatter (1979) incubated similarly
the bile or the dialysate with glucuronidase/arylsulfatase, and the
dichloromethane-extractable radioactivity increased from 1.5% to 75%.
Their results indicate the elimination of TCDD-metabolites in the form
of water-soluble sulfate and glucuronide conjugates. ³H-TCDD
metabolites extracted from the bile of one-year-old beagle dogs with
ethanol (Poiger et al., 1982) were given to bile-duct-cannulated
female Sprague Dawley rats (250 g) as single oral doses of 7.8-20.8 µg
³H-TCDD-metabolites/kg body weight (Weber et al., 1982b). The mean
24-h elimination of radioactivity was 86.7 (± 6.7)% of the dose, 8.2%
occurring in urine, 31.3% in bile, and 46.9% in faeces. A delay in the
excretion of radioactivity in rats whose bile ducts were not
cannulated suggested an enterohepatic circulation in the rat of the
³H-TCDD-metabolites from the dog. The radioactive material in the
rat bile seemed to be conjugated forms of the metabolites from the
dogs. A metabolic breakdown scheme of TCDD in the rat and dog (Fig. 3)
was proposed by Poiger & Buser (1983). The major metabolite seems to
be formed via cleavage of an ether bond.

The metabolic fate of a single ip dose of 10 µg, 500 µCi TCDD/kg
body weight in male C57Bl/6, DBA/2, and B6D2F1 mice was studied by
Gasiewicz et al. (1983b). Samples of urine, bile, and faeces,
collected on days 5 to 8, 8, and 7 after treatment, respectively, were
extracted and analyzed for metabolites of TCDD by HPLC. Unmetabolized
TCDD was detected in faeces but not in urine or in bile. More than 85%
of the total radioactivity eliminated was present as metabolites of
TCDD in all three mouse strains. Metabolites in the bile appear to be
less polar than in urine. Qualitatively the elution profiles for
urine, bile, and faeces from all three strains appeared to be quite
similar.

A dose of ³H-TCDD (0.56 µg/kg body weight) that produced no
gross toxicity was given ip in olive oil to six adult male Hartley
guinea-pigs (Olson, 1986). Metabolites of TCDD were found in organic
extracts of the liver, kidney, perirenal adipose tissue, and skeletal
muscle in amounts corresponding to 13, 4, 8, and 28%, respectively, of
the recovered radioactivity in these organs 45 days after dosing.
These figures suggest that TCDD-metabolites are not efficiently
eliminated from tissues in guinea-pigs. All radioactivity in urine and
bile represented metabolites of TCDD, whereas in faeces most (70-90%)
contained unchanged TCDD. Of the radioactivity administered, 73.4% was
eliminated as unchanged TCDD in faeces and 25.7% as metabolites of
TCDD in urine and faeces. The presence of TCDD in faeces and its
absence in bile suggest that direct elimination of TCDD from the blood
to the intestinal lumen may occur. The HPLC elution profiles for
metabolites were similar, although not identical, for bile, urine, and
tissues. Taken together, the data by Olson (1986) indicate that
metabolism does not appear to have a major role in the ultimate
elimination of TCDD in the guinea-pig.

Olson et al. (1980a) collected bile and urine from Golden Syrian
hamsters that had been treated with a single ip dose of 650 µg
¹⁴C-TCDD/kg body weight 7 days earlier. By means of HPLC, one major
and several minor metabolites of ¹⁴C-TCDD were demonstrated both in
bile and urine. No metabolites of ¹⁴C-TCDD were detectable in liver
and adipose tissue, thus suggesting a rapid clearance of
biotransformed products of TCDD.

A one-year-old beagle dog was cholecystectomized and a Thomas
cannula was implanted about 3 months before the first dose of TCDD
(Poiger et al., 1982, Poiger & Buser, 1983). A total dose of 5.4 mg
was administered enterally in four portions of 1.8, 1.08, 1.08, and
1.44 mg on days 0, 2, 7, and 13. Five phenolic metabolites of TCDD
excreted in dog bile were identified by combined gas
chromatography-mass spectrometry. Severe toxic symptoms preceded the
death of the dog 17 days after the first dose. A metabolic breakdown
scheme of TCDD in the dog (Fig. 3) was proposed by Poiger & Buser
(1983), lateral hydroxylation of TCDD seeming to be the major route of
metabolism.

6.2.1.2 In vitro studies

Although there is evidence of different metabolites of TCDD and
the possibility of its metabolic transformation was suggested as early
as 1975 (Allen et al., 1975; Rose et al., 1976), it was only in 1982
that specific metabolites were identified by Sawahata et al. (1982).
They incubated TCDD with isolated rat hepatocytes at 37 ©C for 8 h,
and the resulting incubation mixture was subjected to HPLC. The major
peak of radioactivity not corresponding to TCDD was incubated with
ß-glucuronidase in order to split the possible glucuronide
conjugate(s) of TCDD or its metabolite(s). They found that
4,5-dichlorocatechol and 4,5-dichloroguaiacol are potential
metabolites of TCDD, but due to the limited amount of material the
identity of these metabolites was not confirmed by gas
chromatography-mass spectrometry. Two other metabolites of TCDD,
namely 1-hydroxy-2,3,7,8-tetrachloro-dibenzo-p-dioxin and
8-hydroxy-2,3,7-trichlorodibenzo-p-dioxin, were isolated by means of
HPLC, and were identified by mass spectrometry.

Primary hepatocytes from Sprague Dawley rats and Hartley
guinea-pigs have been used to study the metabolism of 14C-TCDD
(Wroblewski & Olson, 1985). The overall metabolism was 2.8 times
greater in rats than in guinea-pigs. The metabolism of 14C-TCDD was
increased 3.2-fold in rats pretreated with TCDD (5 µg TCDD/kg body
weight ip 72 h prior to isolation of hepatocytes), but no effect was
found in similarly TCDD-pretreated guinea-pigs, or in phenobarbital-
pretreated rats (80 mg/kg body weight ip for 3 days, beginning 4 days
prior to isolation of hepatocytes). Hepatocytes from TCDD-pretreated
rats metabolized TCDD 9 times more rapidly than similarly pretreated
guinea-pig hepatocytes. TCDD may be metabolized by an inducible form
of cytochrome P448 which is expressed in rats but not guinea-pigs.
These differences in metabolism may play a major role in explaining
the differences in species susceptibility to the acute effects of
TCDD.

6.3 Transfer Via Placenta and/or Milk

The transplacental passage of ¹⁴C-TCDD has been studied by
Khera & Ruddick (1973). Pregnant Wistar rats were given ¹⁴C-TCDD in
a single oral dose of 200 µg/kg body weight on gestation days 16, 17,
or 18 and were killed 6 h after dosing. ¹⁴C-activity was detected in
maternal tissues and also in the fetuses and the placenta. Assuming
that all the ¹⁴C-activity found in the samples was present as
¹⁴C-TCDD, the following levels (ng/gram tissue) were found for
gestation days 16, 17, and 18 respectively: maternal liver 339 (± 15),
339 (± 19), and 275 (± 20); maternal blood 25 (± 11), 19 (± 9), and 10
(± 3); placenta 25 (± 6), 38 (± 4), and 41 (± 3); and fetus 11 (± 3),
15 (± 1), and 16 (± 1). Studies by Moore et al. (1973) indicated that

the passage of TCDD or its metabolites into milk could be of
importance, as TCDD-related effects were observed in sucklings,
nourished by lactating mothers given, after delivery, a single oral
dose of 1 or 3 µg/kg body weight.

The TCDD concentration in livers of pregnant NMRI mice, at a
given dose, was significantly lower than in livers of non-pregnant
mice (Krowke 1986). The concentration of TCDD in the liver of
non-pregnant mice was about 5 times higher than in pregnant mice 7
days after a s.c. dose.

Nau & Bass (1981) studied the transfer of ¹⁴C-TCDD to embryos
and fetuses in NMRI mice. The animals were given a single dose of 5,
12.5, or 25 µg TCDD/kg body weight by gavage or by s.c. or ip
injection at day 16 of gestation to study the transfer of TCDD to the
fetus. The animals were killed two days later and various tissues were
analyzed for radioactivity. No evidence was found to indicate a major
first pass effect following oral administration. Maternal livers
contained the highest levels of TCDD, 4.1 to 10.5% of the radioactive
dose administered, which was about one order of magnitude higher than
in extrahepatic maternal tissues, including placenta. Fetal liver and
extrahepatic tissues contained low levels of radioactivity
corresponding to 0.09 to 1.41% and 0.05 to 0.14%, respectively, of the
dose administered to the dams. More radioactivity was recovered in the
placenta and fetus when TCDD was given either as a single ip dose of
25 µg/kg body weight on day 10 of gestation or as 5 daily ip doses of
5 µg/kg body weight on days 7 to 11, when compared to a single i.p.
dose of 25 µg/kg body weight on gestation day 7. Oral dosing of 30 µg
¹⁴C-TCDD (0.332 µCi/ µg)/kg body weight to pregnant C57Bl/6 mice on
gestation day 11 resulted in 0 to 14% embryomortality on gestation
days 12 to 14 (Weber & Birnbaum, 1985). About 0.03% of the radioactive
dose was contained in the embryo and in the placenta on days 12 to 14
of gestation. The maternal liver contained 67.4, 67.9, and 50.6% of
the administered radioactive dose on gestation days 12, 13, and 14,
during which days the cumulative elimination of radioactivity in urine
and faeces was 2.4 and 53.3% of the dose, respectively.

The transfer of ¹⁴C-TCDD via placenta and milk and the
distribution of the transformed TCDD between various embryonic and
fetal tissues were studied in NMRI mice (Nau et al., 1986). Dams were
given a single dose of 25 µg ¹⁴C-TCDD (45 mCi/mmol)/kg body weight
either orally, subcutaneously, or intraperitoneally. To differentiate
between postnatal and in utero exposure, the experimental design
included cross fostering. Depending on the route of administration,
from 0.02 to 0.07% of the administered radioactivity was found in the
liver of the fetus at birth. The highest levels were noted after ip
administration and the lowest after oral intubation. The corresponding
values one week after birth were 0.05 to 0.20%. The hepatic
radioactivity in the neonate reached a peak 1 week after birth and
then decreased slowly throughout and after lactation. The levels of

TCDD-derived radioactivity in extra-hepatic tissues of the offspring
were approximately one order of magnitude lower than the hepatic
levels. Very little radioactivity was found in the stomach filled with
milk, indicating that TCDD ingested from milk was rapidly absorbed in
the stomach.

Arstila et al. (1981) studied the excretion of TCDD in goat milk
after a subchronic administration of 200 ng TCDD per day for 2 months
in the first experiment and of 400 ng TCDD per day for one month in
the second experiment. The minimal detectable concentration in this
study was declared to be below 5 ng/litre. The maximum concentration
of TCDD in milk in the first experiment was 20.8 (± 6.6) ng/litre and
in the second experiment 19.3 (± 6.6) ng/litre. After 18 weeks feeding
with TCDD the levels had dropped to 4.2 and 3.6 ng/litre,
respectively.

The secretion of TCDD in milk and cream has also been studied in
lactating dairy cows kept on a diet containing 10, 30, 100, 300, or
1000 mg/litre 2,4,5-T, corresponding to TCDD levels of 5, 15, 50, 150,
and 500 ng/litre (Jensen & Hummel, 1982). This resulted in levels of
TCDD in the excreted milk of below detection limit, 3, 10, 16-22, and
42-89 ng/litre, respectively, indicating that about 10-20% of the dose
given was eliminated in the milk. The levels in cream were about ten
times higher than those in milk.

One lactating Holstein cow, receiving commercial technical-grade
pentachlorophenol containing several PCDDs and PCDFs orally in
gelatine capsules at a dose rate of 10 mg/kg body weight twice daily
for 10 days and once daily for the following 60 days, calved 151 days
after treatment was stopped (Firestone et al., 1979). The PCDD content
of blood, body fat (shoulder from the cow and hind quarter of the
calf), and milk fat was determined 14 days later. The detected
congeners were 1,2,3,6,7,8-hexaCDD, 1,2,3,4,6,7,8-heptaCDD, and
octaCDD and their levels were 12, 13, 20 ng/litre and 27, 14, 6
ng/litre in the blood of the cow and calf, respectively. Corresponding
values for body fat were 4.8, 11.1, and 6.1 µg/kg in the cow and 2.3,
1.9, and 0.5 µg/kg in the calf. The milk fat from the cow contained
2.2, 4.4, and 3.3 µg/kg of the respective congeners.

6.4 Matrix Effects on the Uptake ("Bio-availability")

Uptake of TCDD and other PCDD congeners is highly dependent upon
the formulation in which it is applied. Although conflicting results
have been obtained from studies where TCDD in soil was administered
(McConnell et al., 1984; Umbreit et al., 1985, 1986a, 1986b), most
available data support the idea that mixing TCDD with soil or
activated carbon results in the adsorption of TCDD to the soil
particles, thus reducing the availability of TCDD. Contact time of
TCDD with soil seem to influence the availability, probably because
the binding of TCDD to soil particles becomes strengthened (Poiger &

Schlatter, 1980). Studies elucidating the matrix effects of various
soils or activated carbon on the TCDD-responses in several animal
species are summarized in Table 46.

Poiger & Schlatter (1980) showed decreasing hepatic recovery of
TCDD in rats 1 day after dosing using suspensions of ethanol, soil, or
activated carbon as vehicles. About 50% lower hepatic retention of
TCDD was obtained in the rat, 6 days after dosing, when Minker stout
site soil was the vehicle as compared to corn oil (Lucier et al.,
1986). However, in this study hepatic enzyme-induction (AHH and UDPGT)
was similar in the two vehicle groups.

The bioavailability of TCDD from environmentally contaminated
soil samples has been studied in young guinea-pigs after intragastric
administration (McConnell et al., 1984). Groups of six animals each
were given soil samples corresponding to doses of approximately 1, 3,
or 10 µg TCDD per kg body weight. The doses were based on analyses of
soil siftings (60-gauge mesh) from the Times Beach and Minker Stout
sites, which indicated concentrations of 770 and 880 µg TCDD/kg,
respectively. Controls received soil samples in which no TCDD, PCBs,
or PCDFs were detected. For comparison, pure TCDD in corn oil was
given at either 0, 1, or 3 µg/kg. The observation time was 30 days.
LD₅₀ values calculated in this study were 1.75 µg/kg for TCDD in
corn oil, 7.15 µg/kg for Times Beach soil, and 5.50 µg/kg for the
Minker Stout site soil. An exact percentage for bioavailability was
not calculated in this study, but the TCDD content of the livers of
exposed guinea-pigs indicated a highly efficient absorption of TCDD
from soil.

TCDD in contaminated soil from a 2,4,5-T manufacturing plant and
from a metal salvage yard (Newark, New Jersey, USA) had a low
bioavailability (0.5% and 21.3%, respectively) in guinea-pigs (Umbreit
et al., 1985, 1986a). The soils were given as single oral doses in a
10% aqueous suspension in 5% gum acacia. The bioavailability was
judged by the hepatic concentration of TCDD 60 days after dosing in
guinea-pigs receiving site soils, decontaminated soil, and
TCDD-recontaminated soil.

The difference in bioavailability of TCDD from
2,4,5-T-manufacturing site soil in Newark (New Jersey) and from Times
Beach Site soil (Missouri) when given orally to guinea-pigs was
confirmed by Umbreit et al. (1986b).

Bonaccorsi et al. (1984) gave to rabbits seven daily doses of
TCDD in corn oil, TCDD-contaminated Seveso soil, or recontaminated
soil. Taking the hepatic recovery of TCDD in the corn oil group as
100% "bioavailability", the decrease in "bioavailability" of TCDD from
Seveso soil was 68%. The decrease in "bioavailability" of
recontaminated soil varied from 0 to 44% with the doses used.

Table 46. Matrix effects of various soils on TCDD-responses, as related to the estimated bioavailability of TCDD given
orally to different species.

Species/ Vehicle/matrix Dose of Lethality Hepatic Estimated Reference
Length TCDD^c recovery of "bio-
of study TCDD^e availability"
(days)

Rat 50% ethanol 14.7 ng 36.7^f Poigner &
1 day recontaminated soil^a 12.7, 22.9 ng 24.1^f Schlatter
recontaminated soil^b 21.2, 22.7 ng 16.0f (1980)
activated carbon 14.7 ng < 0.07^f

Rat corn oil 1.0, 5.0 7.6 and 40.8 Lucier et al.
6 days Minker Stout site soil 1.1, 5.5 1.8 and 20.3 50% (1986)

Guinea- corn oil 1, 3 1/6, 6/6 1.6, 13.3 McConnell et
pig Times Beach site soil 1.3, 3.8, 12.8 0/6, 1/5, 5/5 < 1.0-34.3 Efficient al. (1984)
30 days Minker Stout site soil 1.1, 3.3, 11.0 0/6, 2/6, 6/6 < 1.0-25.7 absorption from
recontaminated soil 10.0 6/6 45.4 soil (85%)

Guinea- corn oil 6 5/8 Umbreit et al.
pig 2,4,5-T-manufacturing 3, 6, 12 0/8, 0/7, 0/7 0.09 (high dose) 0.5% (1985, 1986a)
60 days site soil (Newark,
New Jersey, USA).
Metal salvage yard soil 0.32 0.23 21.3%
(Newark)
recontaminated soil 6 6/7 18

Table 46 (contd).

Species/ Vehicle/matrix Dose of Lethality Hepatic Estimated Reference
Length TCDD^c recovery of "bio-
of study TCDD^e availability"
(days)

Guinea- 2,4,5-T-Manufacturing 3, 5, 10 0/18, 1/20, 1/18 Umbreit et al.
pig site soil, (Newark) (1986b)
60 days Times Beach site soil 1, 3, 10 2/19, 2/20, 8/14
recontaminated soil 6 19/20

Rabbit corn oil 20, 40, 80^d 0.26-2.7 Bonaccorsi et
7 days Seveso soil 80, 160^d 0.88-2.2 > 32% al. (1984)
recontaminated soil 20, 40, 80^d 0.26-1.5 > 66-100%

^a Contact time = 10-15 h.
^b Contact time = 8 days.
^c µg/kg body weight, unless otherwise stated.
^d ng/kg body weight per day.
^e µg/kg liver, unless otherwise stated.
^f % of dose.

Considerably lower hepatic levels of PCDDs and PCDFs were
observed in rats fed a diet containing fly ash from municipal
incinerators compared with those fed a diet containing extracts from
the same fly ash (van den Berg et al., 1983).

Dietary intake of soot-containing TCDD produced 60% mortality in
male and female guinea-pigs on days 46 and 60, respectively, at which
time the estimated TCDD-consumption was 1.3 µg/kg body weight for
males and 1.9 µg/kg body weight for females (DeCaprio et al., 1983,
see also section 8.2.3). These data thus suggest a high uptake of TCDD
from the soot matrix.

7. EFFECTS OF TCDD AND OTHER PCDDs ON EXPERIMENTAL ANIMALS AND
IN VITRO TEST SYSTEMS

7.1 Acute Toxicity

7.1.1 In vivo studies on mammals

The range of doses required to cause death varies considerably
between species, as well as between strains of species, and with sex,
age, and route of administration within a single strain (Table 47).
More than an 8000-fold difference exists between the dose of TCDD
reported to cause 50% lethality to male Hartley guinea-pigs, the most
sensitive species tested (Schwetz et al., 1973), and the corresponding
dose for male Golden Syrian hamsters (Henck et al., 1981). The rat
seems to be the second most sensitive species, although there is a
more than 200-fold variability in LD₅₀ values between different
strains. The oral LD₅₀ value was 22 µg TCDD/kg body weight for male
Sherman rats (Schwetz et al., 1973), whereas Walden & Schiller (1985)
found LD₅₀ values ranging from 164 to 340 µg TCDD/kg body weight
when male Fisher 334 N rats from three different suppliers were
tested. The Han/ Wistar-strain of rat has been demonstrated to be
particularly resistant to TCDD-exposure (Pohjanvirta & Tuomisto,
1986). Among the five rats per dose group (0, 1500, 2000, 2500, or
3000 mg TCDD/kg body weight) only one animal died within the 39-40
days observation period.

Monkeys (McConnell et al., 1978a), New Zealand rabbits (Schwetz
et al., 1973), C57Bl/6 mice (Chapman & Schiller, 1985; Jones & Greig,
1975; McConnell et al., 1978b; Smith et al., 1981; Vos et al., 1974),
DBA/2-mice (Chapman & Schiller, 1985), and B6D2F1-mice (Chapman &
Schiller, 1985) gave oral LD50 values of 70, 115, 114, 2570, and 296
mg TCDD/kg body weight, respectively. The difference in sensitivity
towards TCDD among various strains of mice has been claimed to depend
on a genetic variability in the Ah and/or hr-locus (see section
7.8.1).

Male Sherman rats were found to be more sensitive to TCDD than
were females (Schwetz et al., 1973), whereas Beatty et al. (1978)
reported that male Sprague Dawley rats were more resistant to TCDD
than were females. Smith et al. (1981) found adult female C57BL/10
mice to be more resistant to TCDD than adult males of the same strain.
No differences in sensitivity to TCDD between sexes were recorded for
guinea-pigs (McConnell et al., 1978b; Silkworth et al., 1982) or
hamsters (Olson et al., 1980b). Thus data on sex differences in
sensitivity to lethal effects of TCDD are conflicting.

Data on the effect of age at exposure to TCDD on the sensitivity
of acute response are scarce, and comparisons are hampered by the
absence of information, or incomplete information, on the age and/or
body weight of the tested animals. However, Beatty et al. (1978) found

that weanling male Sprague Dawley rats were more sensitive to TCDD
than were adult males. A dose of 25 µg TCDD/kg body weight caused,
after 35 days, a cumulative lethality of 62% in weanling Sprague
Dawley rats and 25% in young adults (Christian et al., 1986a). When
weanling and mature adults were exposed to a similar toxic dose of
TCDD (LD62 and LD60, respectively), onset of death occurred 9 days
later in the adults (Christian et al., 1986a).

Schwetz et al. (1973) found LD50 values in rabbits of 115 µg
TCDD/kg body weight after oral exposure, as compared to 275 µg TCDD/kg
body weight after dermal exposure. C57Bl/6-mice seem to be more
sensitive to ip administration of TCDD (Gasiewicz et al., 1983b) than
to oral administration (McConnell et al., 1978b). The LD₅₀ value in
guinea-pigs was increased from 2.5 µg/kg body weight to 19 µg/kg body
weight when the vehicle for the oral administration of TCDD was
changed from corn oil to methyl cellulose (Silkworth et al., 1982).
Umbreit et al. (1985) compared mortality, and time to death, among
guinea-pigs given single oral doses of TCDD in corn oil, TCDD in a
suspension of cleaned soil, from an industrial site, or
TCDD-contaminated soil from the same industrial site. Animals treated
with corn oil, cleaned soil and contaminated site soil survived the
60-day study without any sign of TCDD intoxication. Among animals that
received 6 µg TCDD/kg body weight either in corn oil or mixed with
cleaned soil, only 3/8 and 1/7 survived, respectively. Deaths occurred
between days 9 to 31 and 15 to 25, respectively. These results are
different from those reported by McConnell et al. (1984) where
TCDD-contaminated soils from Times Beach and Minker Stout were highly
toxic to guinea-pigs. LD50 values calculated in this study were 1.75
µg/kg for TCDD in corn oil, 7.15 µg/kg for TCDD in Times Beach soil,
and 5.50 µg/kg for TCDD in Minker Stout site soil. The Minker Stout
site soil was also potent in inducing AHH-activity in female Sprague
Dawley rats. The different results may be due to the vehicle used
and/or to the presence of other substances in the soils that may
potentiate or retard the TCDD-induced toxicity (Umbreit et al.,
1986a). In section 6.4, the results from McConnell et al. (1984) and
Umbreit et al. (1985, 1986a) are further discussed from the viewpoint
of the "bioavailability" of TCDD in soils (Table 46).

Despite similar routes and vehicles for the administration of
TCDD to Golden Syrian hamsters, LD₅₀ values varied between 1157
µg/kg body weight (Olson et al., 1980b) and 5051 µg/kg body weight
(Henck et al., 1981). A possible explanation for this difference could
be the spontaneous occurrence of ileitis observed in the former study,
which might have increased the susceptibility of those hamsters to
TCDD toxicity.

Table 47. Single lethal dose values for TCDD^a

Species/strain Sex/No/ Age/weight Route/vehicle Dose Duration of LD₅₀ Time to
(Reference) group tested observation (µg/kg) death
(µg/kg) (days)

Rats

Porton F/5-12 8-9 weeks/ oral/DMSO 0 90 days NR 40
170-200 g 30
(Greig et al., 48
1973) 75
120
190
300

Porton F/6 9-10 weeks/ oral/arachis 0 90 days NR 40
170-200 g oil 126
(Greig et al., 199
1973) 315
500

Sherman M/5-10 NR oral/corn oil 8 2-8 weeks 22 9-27
acetone (9:1) 16
(Schwetz et al., 32
1973) 63

Sherman F/NR NR oral/corn oil NR 2-8 weeks 45 13-43
acetone(9:1)
(Schwetz et al.,
1973)

Table 47 (contd).

Species/strain Sex/No/ Age/weight Route/vehicle Dose Duration of LD₅₀ Time to
(Reference) group tested observation (µg/kg) death
(µg/kg) (days)

Rats (contd)
Sprague M/6 adult/NR ip/olive oil 4 doses 20 60 NR
Dawley 20-80 60

(Beatty et al.,
1978)
Sprague F/6 adult/NR ip/olive oil 4 doses 20 60 NR
Dawley 10-60

(Beatty et al.,
1978)

Sprague M/6 25 days/NR ip/olive oil 4 doses 20 25 NR
Dawley 5-50

(Beatty et al.,
1978)

Fisher 334N M/7 11-12 weeks/ oral/corn oil 0 30 days 340^b 28^b
230-280 g 75 303^c 26^c
(Walden & 150 164^d 25^d
Shiller, 1985) 225
275
325
375

CD M/7 10-11 weeks/ oral/corn oil 0 30 days 297^d 25^d
350-370 g 75
(Walden & 150
Shiller, 1985) 225
275
325
375

Table 47 (contd).

Species/strain Sex/No/ Age/weight Route/vehicle Dose Duration of LD₅₀ Time to
(Reference) group tested observation (µg/kg) death
(µg/kg) (days)

Rats (contd)
Han/Wistar M/5 NR/300-350 oral/corn oil 1500 39-40 days > 3000 NR
2000
(Pohjanvirta 2500
& Tuomisto, 3000
1986)

Mice

C57BL/6 M/14 3 months/ oral/corn oil 0 2 months 114 15-30
23.6-30.8 g acetone (6:1) 100
(Vos et al., 1974) 150
200

C57BL/6 M/NR 7-15 weeks/ oral/arachis NR 35 days 126 21±1.6
14-30 g oil
(Jones & Greig,
1975)

C57BL/6 M/8 9 weeks/ oral/corn oil NR 30 days 284 22-25
21-25 g
(McConnell et al.,
1978^b)

C57BL/6J M/NR NR ip/olive oil NR 30 days 132 NR

(Gasiewicz et al.,
1983b^d)

Table 47 (contd).

Species/strain Sex/No/ Age/weight Route/vehicle Dose Duration of LD₅₀ Time to
(Reference) group tested observation (µg/kg) death
(µg/kg) (days)

Mice (contd)
C57BL/6J M/10-15 10-12 weeks/ oral/corn oil 95 30 days 182 24
22-32 g 145
(Chapman & 190
Schiller, 1985) 285

C57BL/10 M/5 42-121 days/NR oral/arachis 85 45 days 146 22-38
oil 107
(Smith et al., 135
1981) 170
213

C57BL/10 F/5 42-121 days/NR oral/arachis 85 45 days > 450 22-38
oil 107
(Smith et al., 135
1981) 170
213
269
338
426
536

DBA/2J M/NR NR ip/olive oil NR 30 days 620 NR

(Gasiewicz et al.,
1983b^e)

DBA/2J M/10-15 10-12 weeks/ oral/corn oil 1370 30 days 2570 21
22-32 g 1870
(Chapman & 2610
Schiller, 1985) 3500
4470

Table 47 (contd).

Species/strain Sex/No/ Age/weight Route/vehicle Dose Duration of LD₅₀ Time to
(Reference) group tested observation (µg/kg) death
(µg/kg) (days)

Mice (contd)
B6D2F₁/J M/NR NR ip/olive oil NR 30 days 300 NR

(Gasiewicz et al.,
1983^e

B6D2F₁J M/10-15 10-12 weeks/ oral/corn oil 170 30 days 296 25
22-32 g 220
(Chapman & 265
Schiller, 1985) 325
425
450

Guinea-pigs

Hartley M/NR NR oral/corn oil NR 2-8 weeks 0.6 5-34

(Schwetz et al.,
1973)

Hartley M/NR NR oral/corn oil NR 2-8 weeks 2.1 9-42
acetone (9:1)
(Schwetz et al.,
1973)

Hartley M/6 3-4 weeks/ oral/corn oil NR 30 days 2 17-20
200-250 g
(McConnell et al.,
1978^b)

Table 47 (contd).

Species/strain Sex/No/ Age/weight Route/vehicle Dose Duration of LD₅₀ Time to
(Reference) group tested observation (µg/kg) death
(µg/kg) (days)

Guinea-pigs (contd)
Hartley F/6 NR/500-600 g oral/corn oil 0.1 42 days 2.5 32-42
0.5
(Silkworth et al., 2.5
1982) 12.5
20.0

Hartley F/6 NR/500-600 g oral/methyl- 0.1 42 days 19 12-42
cellulose 0.5
(Silkworth et al., 2.5
1982) 12.5
20.0

Rabbits

New Zealand M,F/NR NR oral/corn oil NR 2-8 weeks 115 6-39
acetone (9:1)
(Schwetz et al.,
1973)

New Zealand M,F/NR NR dermal/acetone 31.6 3 weeks 275 12-22
63
(Schwetz et al., 126
1973) 252
500

New Zealand M,F/5 NR ip/corn oil 31.6 4 weeks NR 6-23
63
(Schwetz et al., 126
1973) 252
500

Table 47 (contd).

Species/strain Sex/No/ Age/weight Route/vehicle Dose Duration of LD₅₀ Time to
(Reference) group tested observation (µg/kg) death
(µg/kg) (days)

Hamsters
Golden Syrian M/5-6 NR/50-80 g ip/olive oil 0 50 days > 3000
500
(Olson et al., 1000
1980^b) 2000
3000

Golden Syrian F/5 NR/50-80 g ip/olive oil 0 50 days > 3000 14-32
500
(Olson et al., 1000
1980^b) 2000
3000

Golden Syrian M/5 NR/50-80 g oral/olive oil 500 50 days 1157 2-47
1000
(Olson et al., 2000
1980^b) 3000

Golden Syrian M/6 NR/70-120 g oral/corn oil 0 55 days 5051 9-43
acetone (9:1) 300
(Henck et al., 600
1981) 1000
3000
6000

Table 47 (contd).

Species/strain Sex/No/ Age/weight Route/vehicle Dose Duration of LD₅₀ Time to
(Reference) group tested observation (µg/kg) death
(µg/kg) (days)

Monkeys

Macaca F/3 juvenile/ oral/corn oil 0 47 days < 70 14-34
mulatta 2.1-2.6 kg 70
350
(McConnell et al.,
1978^a)

Dogs

Beagle M/2 NR oral/corn oil 300 2-8 weeks NA 9-15
acetone (9:1) 3000
(Schwetz et al.,
1973)

Beagle F/2 NR oral/corn oil 30 2-8 weeks NA all
acetone (9:1) 100 animals
(Schwetz et al., sur-
1973) vived

Chickens

Leghorn NR 4-6 weeks/NR oral/NR NR NR 25-50 12-21

(Grieg et al.,
1973)

^a M = male, F = female, NR = not reported, NA = not applicable, ip = intraperitoneal, DMSO = dimethyl sulfoxide.
^b supplied by Harlan.
^c supplied by Frederick.
^d supplied by Charles River.
^e based on unpublished studies by Gasiewicz et al., 1983.

TCDD affects a variety of organ systems in different species. The
organ primarily affected in rodents and rabbits is the liver. In
guinea-pigs atrophy of the thymus and lymphatic tissues seems to be
the main effect, while dermal effects are prominent signs in non-human
primates. Generally it is not possible to specify a single organ whose
dysfunction is responsible for death. Overall, TCDD seems to have a
predilection for causing pathological changes in epithelial tissues,
both cutaneous and internal. This is particularly apparent in
non-human primates (Macaca mulatta), and is note-worthy that the
lesions mimic to some degree the effects in human beings. The
histopathological alterations in tissues include hyperplastic and/or
metaplastic alterations as well as hypoplastic responses. The toxic
responses of various species to TCDD are summarized in Table 48,
adapted from Poland & Knutson (1982). In all animal species studied,
death occurred after a time lapse ranging from several days to more
than one month after exposure. The delay was dependent on dose but not
on species (Table 48). Progressive loss of body weight was a
characteristic sign observed in animals given a lethal dose of TCDD.
The weight loss became manifest usually within a few days after
exposure and resulted in a substantial reduction of the adipose tissue
observed at autopsy. At sublethal doses of TCDD a dose-dependent
decrease in body weight gain occurred. This TCDD-induced wasting
syndrome has been thoroughly investigated in several studies discussed
more fully in section 7.4.1.

The greatest difference between species at necropsy, both in
gross and histological effects, concerns pathological alterations in
the liver. As discussed in detail in section 7.4.2, a dose of TCDD
lethal to guinea-pigs did not result in liver damage comparable to the
liver lesions described in rabbits and rats or to liver changes
observed in mice dying after doses higher than those needed to cause
death in these species. In the hamster, frank liver lesions do not
occur even after fatal doses.

Chloracne-like lesions can be induced by topical application
and/or systemic administration of TCDD in rabbits, non-human primates,
and hairless mice. These lesions are further discussed in section
7.4.4.

Severe thymus atrophy was also found at autopsy in all animal
species given lethal doses of TCDD. Histological examinations revealed
lymphoid cell depletion in thymus cortex, spleen, and lymph nodes.
These consistent findings in TCDD poisoning will be discussed in
detail in section 7.4.5 together with the other lymphoid
tissue-related effects.

Table 48. Specific differences in toxic responses following exposure to 2,3,7,8-TCDD^a, ^b

Monkey Guinea- Cow^c Rat Mouse Rabbit^c Chicken^c Hamster
pig

Hyperplasia and/or metaplasia
Gastric mucosa ++ 0 + 0 0 0
Intestinal mucosa + ++
Urinary tract ++ ++ ++ 0 0
Bile duct and/or gall bladder ++ 0 + ++ 0
Lung: focal alveolar ++
Skin ++ 0 +^d 0 0 ++ 0

Hypoplasia, atrophy, or necrosis
Thymus + + + + + + +
Bone marrow + + ± +
Testicle + + + +

Other responses
Liver lesions + + ++ + ++ + ±
Porphyria 0 0 + ++ + 0
Oedema + 0 0 + ++ +

^a References: monkey (Norback & Allen, 1973; Allen et al., 1977; McConnell et al., 1978a), guinea-pig (McConnell
et al., 1978b; Moore et al., 1979; McConnell, 1980; Turner & Collins, 1983), cow (McConnell, 1980), rat (Kociba et
al., 1978; Kociba et al., 1979a; McConnell, 1980), mouse (Vos et al., 1973; Schwetz et al., 1973; McConnell et al.,
1978b), rabbit (Vos & Beems, 1971; Schwetz et al., 1973), chicken (Allen & Lalich, 1962; Vos & Koeman, 1970;
Norback & Allen, 1973; Schwetz et al., 1973), hamster (Olson et al., 1980b; Henck et al., 1981).

^b Symbols: 0 = lesion not observed, + = lesion observed (number of "+" denote severity), ± = lesion observed to
a very limited extent, blank = no evidence reported in literature.
^c Responses followed exposure to 2,3,7,8-TCDD or structurally related chlorinated aromatic hydrocarbons.
^d Skin lesions in cattle have been observed, but they differ from the skin lesions observed in other species.

There are also substantial interspecies differences in the
effects observed in other organs of animals given lethal doses of
TCDD. Icterus was reported in rats (Buu-Hoi et al., 1972a; Gupta et
al., 1973), hepatic porphyria occurred in mice and rats (see section
7.4.3), and ascites with subcutaneous oedema and hydrothorax appeared
in mice (Jones & Greig, 1975; Vos et al., 1974) and monkeys (Allen et
al., 1977). The accumulation of serous fluid in the pericardial sac
occurred in chickens after a single lethal dose of TCDD (see section
7.1.4). Haemorrhages were frequently observed in many organs following
lethal doses in monkeys (Allen et al., 1977), rats, and guinea-pigs
(Gupta et al., 1973). In mice, death was frequently attributed to
terminal haemorrhages (Vos et al., 1974).

When administered in doses sufficient to cause overt toxicity,
TCDD causes testicular atrophy and degeneration characterized by
reduced spermatogenic activity in mice (McConnell et al., 1978b), rats
(Kociba et al., 1976; Van Miller et al., 1977), and guinea-pigs
(McConnell et al., 1978b). The same symptoms were present in monkeys
fed dioxin-containing toxic fat (Allen & Carstens, 1967; Norback &
Allen, 1973). The decreases in male sex organ weights (seminal
vesicles, ventral prostate, testes, and caput epididymis) were
dose-dependent. ED50 values were around 15 mg TCDD/kg body weight in
Sprague Dawley rats 7 days after TCDD-treatment when compared to
pair-fed control rats (Moore et al., 1985). Shrunken, hyperchromatic
nuclei in the two layers of seminiferous tubules closest to the
basement membrane were observed in testes of young Sprague Dawley rats
90 h after a single ip dose of 5 mg TCDD/kg body weight (Mittler et
al., 1984). Epididymal lesions (Khera & Ruddick, 1973) and decreased
amounts of secretory material within the accessory sex glands (Kociba
et al., 1976) have been reported in TCDD-treated rats. Reduced
prostate weights, both absolute and relative, were found in Han/Wistar
rats at non-lethal doses of TCDD (Pohjanvirta & Tuomisto, 1986).

Reduced relative uterine weight, accompanied by decreased mucosa,
stroma, and glands, occurred in young C57Bl/6 mice dosed with 6 µg
TCDD/kg body weight three times a week for 1 month, but there was no
effect on the ovaries (Gallo et al., 1986). TCDD had no effect on the
uterine weight in 25 day-old Long Evans rats 2 - 10 days after
treatment with single ip doses of 20 or 80 µg/kg body weight (Romkes
et al., 1987). However, the increase in uterine weight induced by
estradiol treatment was counteracted by simultaneous TCDD-treatment.

Proliferative lesions of the gastrointestinal tract have been
found primarily in non-human primates (Allen et al., 1977; McConnell
et al., 1978a) whereas proliferative changes of the transitional
epithelium in the urinary tract have been found in both guinea-pigs
(Gupta et al., 1973; McConnell et al., 1978b) and monkeys (Norback &
Allen, 1973).

Reduced or unaffected spleen weight and slight to moderate loss
of lymphocytes from spleen germinal centers have been common findings
in laboratory animals exposed to sublethal to lethal doses of TCDD
(Gasiewicz et al., 1980; Greig et al., 1973; Kociba et al., 1978;
McConnell et al., 1978a; Olson et al., 1980b; Vos et al., 1973, 1974).
Spleen cellularity in C57BL/6 mice was decreased 14 and 21 days after
treatment with 30 µg TCDD/kg body weight (Chastain & Pazdernik, 1985),
but not in B6C3F1 or DBA/2 mice 7 days after oral doses of up to 10 µg
TCDD/kg body weight (Luster et al., 1984).

The thyroid weight, both absolute and relative to body weight, of
the rat has been found to be increased by TCDD-treatment (Bastomsky,
1977; Potter et al., 1983, 1986b). Degenerative changes in the
epithelial cells of the thyroid gland were observed in rats after 31
daily oral doses of 1 µg TCDD/kg body weight (Gupta et al., 1973) and
7 days after a single ip dose of 150 µg TCDD/kg body weight (Rozman et
al., 1986). However, Potter et al., (1986b) found no consistent
changes in follicle size or colloid content, variation in follicle
size, height of follicular epithelium, or resorption of colloid at the
periphery of follicles in rats one week after an oral dose of TCDD in
the range 6.25 to 100 µg/kg body weight.

Histological changes in the pancreas and in the interscapular
brown adipose tissue have been found in Sprague Dawley rats exposed to
single ip doses of 150 µg TCDD/kg body weight (Rozman et al., 1986).

Lesions of the adrenal glands have been observed in mice and
guinea-pigs treated with (Gupta et al., 1973; McConnell et al.,
1978b).

Acute exposure to lethal doses of TCDD has produced minor
haematological alterations in all species studied. Anaemia was not
observed in mice or guinea-pigs (Zinkl et al., 1973), but a moderate
anaemia and leukocytosis occurred in rats (Buu-Hoi et al., 1972a) and
monkeys (McConnell et al., 1978a). On the contrary, haemoconcentration
was observed in rats after exposure to a somewhat lower but still
lethal dose of TCDD (Greig et al., 1973; Zinkl et al., 1973).
Thrombocytopenia and clotting abnormalities were observed in rats
after acute exposure to lethal levels of TCDD (Weissberg & Zinkl,
1973). Hypocellularity of the bone marrow was found in guinea-pigs
(McConnell et al., 1978b), rhesus monkeys (Allen et al., 1977;
McConnell et al., 1978a), and mice (McConnell et al., 1978b). However,
in these studies decreased bone marrow cellularity, as judged by
histological examination, appeared only at doses high enough to cause
severe toxicity in the experimental animals. More recent studies
(Chastain & Pazdernik, 1985; Luster et al., 1980, 1985) have
demonstrated that collection and enumeration of bone marrow cells in
suspension provides a more sensitive and quantitative method than
histological preparation for assessing bone marrow cellularity.

Decreased bone marrow cellularity was found in adult male C57Bl/6 mice
3 days after exposure to 120 µg TCDD/kg body weight (Chastain &
Pazdernik, 1985), and in female B6C3F1 mice 5 days after exposure to
10 µg TCDD/kg body weight (Luster et al., 1985), but not in female
DBA/2 mice, even at a dose of 50 µg TCDD/kg body weight (Luster et
al., 1985). The myelotoxic potential of TCDD is described in section
7.4.6.

Changes in clinical chemistry, including serum enzyme activities,
serum protein concentrations, and lipid levels, observed in animals
after acute exposure to TCDD, primarily reflect damage to other organ
systems, mainly the liver (McConnell et al., 1978a; McConnell & Moore,
1979; Olson et al., 1980b). Liver-related enzyme activities in serum
are affected in those animal species where liver damage is a prominent
sign of TCDD toxicity. In those animal species where hepatotoxicity is
not as apparent, such as monkeys and guinea-pigs, these enzyme
activities are essentially normal. Also the TCDD-related decrease in
serum albumin seems to be secondary to the hepatotoxic effect, since
the decrease is less evident or non-existent in those animals that
show little liver damage. Generally serum triglycerides and free fatty
acid levels are increased after TCDD exposure, while that of serum
cholesterol is decreased. However, marked species differences exist
and again these effects seem to be secondary to liver damage. For
further details on hyperlipidaemia, see section 7.4.7.

Indicators of renal damage such as blood urea nitrogen,
creatinine, and blood electrolytes are usually within normal limits
after TCDD exposure.

In view of the interspecies differences in the organ distribution
of TCDD and the variability in the effects on various organs in the
different species, the effects of TCDD in non-human primates are of
particular interest. Nine female juvenile rhesus monkeys (Macaca
mulatta) were divided into three groups of three and given TCDD in
a single oral dose of 0, 70, or 350 µg per kg/body weight (McConnell
et al., 1978a). The first indication of a toxic effect of TCDD, at day
3, was weight loss, followed by periorbital oedema, conjunctivitis,
and thickening of the Meibomian glands on day 12. Subsequently, the
eye lashes, facial hair, and toe and finger nails were lost. Monkeys
given the highest dose showed a moderate absolute lymphopenia and
thrombocytopenia. Serum cholesterol levels dropped, while the serum
triglyceride levels increased. Alkaline phosphatase and total
bilirubin were normal, but glutamic oxalic transaminase and aldolase
were increased. A decrease in the albumin fraction of the total serum
protein was noticed. The three monkeys given TCDD in a dose of 350
µg/kg body weight died or became moribund between days 28 and 34 after
administration. One of the monkeys given TCDD in a dose of 70 µg/kg
body weight died 14 days after administration. At autopsy body fat was
almost completely absent in all treated monkeys. Ascites was noticed

in two animals and all monkeys had markedly distended and thickened
bile ducts and gall-bladders. Small focal ulcerated areas in the
fundus of the stomach were observed in two monkeys. Microscopic
examination showed that the Meibomian glands were dilated and filled
with keratinaceous debris. Squamous metaplasia of the glandular
portion, with atrophy of sebaceous cells, was present. Occasional
scattered necrotic hepatocytes were noted in the liver on microscopic
examination. The gastric ulcers that were found extended into the
lamina propria.

Schwetz et al. (1973) reported that no signs of toxicity were
observed in four male mice and two female rats given single oral doses
of respectively 2 and 1 g 2,7-diCDD/kg body weight (purity
99.6-99.8%). For octaCDD (purity 98.86%), oral doses of 1 g/kg body
weight to five female rats and 4 g/kg body weight to four male mice
did not cause any toxic symptoms. A sample containing two different
isomers of hexaCDD (purity > 99%, 89:11) killed 1 of 2 and 0 of 2
male rats when given as single oral doses of 100 and 10 mg,
respectively. The only toxic sign observed among these rats was loss
of body weight. These studies lasted for 2 to 8 weeks.

The toxicities of single oral doses of nine PCDDs, including
TCDD, in C57BL/6J mice and Hartley guinea-pigs were compared by
McConnell et al. (1978b) (Table 49). The purity of the various isomers
tested exceeded 97%. Groups of eight male mice and of six male
guinea-pigs were used for each dose of any of the tested compounds,
and the animals were followed for 30 days after administration. The
toxic effects observed after administration of the different PCDDs
were similar, the only difference among the congeners being the amount
needed to produce a given effect. Progressive decrease of body weight,
more pronounced in guinea-pigs than in mice, was observed after a
lethal dose of any of the congeners tested. Marked reduction in
deposited adipose tissue deposits was a constant finding in animals
given a lethal dose of any of the PCDDs. Reduction of muscle mass and
severe dehydration were observed in guinea-pigs, and ascites,
subcutaneous oedema, and hydrothorax in some of the treated mice.
Decrease of thymus weight was a constant finding in both species,
being more pronounced in mice and in the guinea-pigs that died.
Histological examination of the thymus of the guinea-pigs that died as
early as 5 days after administration of PCDDs revealed scattered
necrosis of lymphocytes throughout the cortex, with concomitant
phagocytosis by macrophages. This was even more apparent in animals
that died 14 days after administration, in which a noticeable decrease
in the thickness of the cortex was observed. In the animals that had
died by day 20 it was difficult to differentiate the cortex from the
medulla, but little evidence of necrosis was present.

Table 49. Estimated single oral LD₅₀ values for some PCDDs^a

Chlorination Guinea-pigs Mice
of PCDDs (mg/kg)^b (mg/kg)^b

2,8 > 300 000 NR
2,3,7 29 444 > 3000
2,3,7,8 2 284
1,2,3,7,8 3 338
1,2,4,7,8 1125 > 5000
1,2,3,4,7,8 73 825
1,2,3,6,7,8 70-100^c 1250
1,2,3,7,8,9 60-100^c > 440
1,2,3,4,6,7,8 > 600 NR

^a From: McConnell et al. (1978b).
^b Spearman-Karber method.
^c Estimated range due to variability in replicates.
NR = not reported.

In guinea-pigs that survived 30 days after a lethal dose of any
of the congeners, the thymus was reduced to one-fourth of its size in
controls. However, thymus histology at this time was often
comparatively normal. A reduction of the lymphoid follicles in the
spleen and of the Peyer's patches in the intestine was observed with
less conspicuous necrosis, which again was not evident in the 30-day
survivors. Striking hypocellularity was found in the sternal bone
marrow in the guinea-pigs that died, but was less obvious in
survivors. Similar thymic and splenic changes were found in mice.
However, bone marrow atrophy occurred only rarely in this species and
then it was less pronounced than in guinea-pigs. In the guinea-pigs
that died, and occasionally in survivors, a marked hyperplasia of the
renal pelvis was observed, invariably extending into the ureter and at
times involving the urinary bladder mucosa. Gastrointestinal
haemorrhages and occasional microscopic dilatation of crypts in the
glandular portion of the gastric mucosa were observed in dead
PCDD-treated animals of both species. Retro-orbital haemorrhages with
exophthalmus and haemorrhages with detachment of the retina were seen
in mice that died after a lethal dose of any of the congeners tested.
Adrenal haemorrhages and moderate atrophy of the zona glomerulosa were
seen in guinea-pigs that died. These changes were not observed in mice
or in surviving guinea-pigs. In guinea-pigs, primarily in animals that
died during the observation period, changes in the spermatogenic
epithelium were observed, with testicular tubules containing only
spermatogonia and Sertoli cells in severely affected animals. Reduced
spermatogenesis, necrotic spermatocytes, and spermatozoa within the

lumen of the testicular tubules and in the epididymis, and
multi-nucleated giant cells within the seminiferous tubules, were
found in the mice that died but not in those that survived. On the
other hand, liver lesions were observed with the same frequency and
degree of involvement in all the mice given the same dose, whether
they died or survived. Minimal liver changes were detected in
guinea-pigs, changes largely confined to central congestion with
occasional degeneration of hepatocytes in dead animals. Fluorescence,
as an indication of porphyria, was found only in mice, particularly in
the liver but also in the incisors, cranial bones, costochondral
junction, and stifle joint. It was dose related and detectable at
doses several times less than the LD₅₀.

Haemolysis and hyperproteinaemia were found in dying animals of
both species. In mice surviving 30 days, but not in guinea-pigs, the
blood a-globulin level was decreased, with a resultant increase in the
albumin/globulin ratio.

With all PCDDs for which it was possible to establish an LD50
values were in the range 10 to 100 times lower in guinea-pigs than in
mice. To produce the greatest toxicity the lateral positions 2,3,7,
and 8 must be fully chlorinated. With 2,3,7-triCDD or 2,8-diCDD the
LD₅₀ values were in the range 1000 to 100 000 times higher than with
TCDD. The addition of a chlorine atom at an ortho-position, e.g.,
1,2,3,7,8-pentaCDD, resulted in only a minor reduction of toxicity. An
additional chlorine atom further reduced the toxic potency but the
LD₅₀ values of 1,2,3,4,7,8-hexaCDD and 1,2,3,6,7,8-hexaCDD remained
comparatively low. The toxicity of 1,2,3,4,6,7,8-heptaCDD was further
reduced. A reduction in the rate of weight gain was observable in
guinea-pigs given doses of this compound exceeding 200 µg/kg body
weight. However, no deaths were observed during the experiment even at
doses as high as 600 µg/kg body weight.

The acute oral toxicities of soot and benzene extracts of soot,
containing PCDDs and PCDFs from a fire in a PCB-containing transformer
(Binghamton, New York, USA) were determined to be 410 mg/kg body
weight and 327 mg equivalents/kg body weight, respectively, in female
Hartley guinea-pigs (Silkworth et al., 1982). The observation period
was 42 days. The test substances were given in 0.75% aqueous methyl
cellulose and the doses used were 250, 500, 750, 1000, and 1250 mg
soot/kg body weight and benzene extracts corresponding to 4, 20, 100,
500, and 1000 mg soot/kg body weight. An extensive investigation,
including pathology, haematology, and serum chemistry alterations, in
groups of six male and female Hartley guinea-pigs, 42 days after
single oral doses of 1, 10, 100, or 500 mg Binghamton soot/kg body
weight in 0.75% methyl cellulose, was reported by Silkworth et al.
(1982). Control animals received 500 mg activated carbon/kg body
weight in the same vehicle. No treatment-related differences were
observed at 1 and 10 mg soot/kg body weight. Decreased body weight
gain occurred in both sexes at 100 and 500 mg/kg, and decreased thymus
weight occurred at 500 mg/kg for males and at 100 and 500 mg/kg for

females. The kidney weight was decreased only in males at 100 and 500
mg/kg. There were no treatment-related alterations in haematological
values. Male guinea-pigs had significantly increased serum
triglyceride levels at 100 and 500 mg/kg and females at 500 mg/kg
only. Elevated aspartate amino transferase (at 100 and 500 mg/kg) and
decreased Y-glutamyltransferase (at 500 mg/kg) were observed in the
serum of female guinea-pigs. The only clearly dose-related
microscopical findings in soot-exposed guinea-pigs were metaplasia of
salivary gland interlobular duct epithelium and goblet cell
hyperplasia of pancreatic interlobular ducts. These lesions occurred
only in males at doses of 100 or 500 mg soot/kg body weight.
Microscopic lesions, which tended to be more frequent and/or severe in
treatment groups than in controls, included bile duct hyperplasia,
hepatocellular cytoplasmic inclusions, vacuolation of the adrenal
cortex, and focal lacrimal adenitis.

7.1.2 In vitro studies on mammalian cells

Over 30 cell types, including primary cultures and cells from
established and transformed cell lines derived from various tissues of
at least six animal species, have been examined for their response to
TCDD (Beatty et al., 1975; Knutson & Poland, 1980a; Niwa et al., 1975;
Yang et al., 1983). The effects studied were viability, growth rate,
and morphological alterations. No toxic effects were observed except
in one rat hepatoma cell line in which Niwa et al. (1975) reported
decreased viability after exposure for 72 h to TCDD at a concentration
of 300 nmol/litre. This concentration is high when compared to the
LD₅₀ in rats and mice. No effect on these cells was observable at 30
nmol/litre.

The biochemical responses found in primary hepatocytes from rats
exposed to 10 µg TCDD/kg body weight in vivo was not present when
the hepatocytes were exposed in vitro to TCDD at 50, 100, or 200
nmol/litre for 48 h (Yang et al., 1983a).

7.1.3 Studies on birds

Chick oedema disease first gained attention in the United States
in 1957. An extensive outbreak among chickens occurred in that year in
Georgia (Firestone, 1973; Sanger et al., 1958; Simpson et al., 1959).
The cause of the disease was traced to the presence in the feed of
toxic components later identified as chlorinated dibenzo-p-dioxins.
TCDD was identified as one of the isomers in the mixture of
chlorinated dibenzo-p-dioxins capable of producing chick oedema
(Flick et al., 1973).

Clinical signs of chick oedema disease consist of dyspnea,
reduced body weight gain, stunted growth, subcutaneous oedema, pallor,
and sudden death. In young chickens gasping was the first noticeable
sign, followed by a waddling gait. Gross inspection of the birds at
autopsy revealed an increased amount of fluid in the pericardial sac

and pale livers with a mottled and irregular granular surface. In
advanced stages the chickens developed a distended abdomen filled with
fluid. Endotheliosis of the vascular system was observed at
microscopic examination, as well as pronounced proliferation of the
endothelium of the glomerular capillaries and necrosis of the liver
cells. Diseased chickens developed pulmonary oedema and perivascular
lymphocyte infiltration, as well as oedema of the cardiac muscle with
interstitial lymphocytic infiltration (Allen, 1964; Allen & Lalich,
1962, McCune et al., 1962; Simpson et al., 1959).

Experimentally, chick oedema has been produced with a single dose
of 25 to 50 µg TCDD/kg body weight (Greig et al., 1973). When mixtures
of tri- and tetraCDDs were fed at dietary concentrations of 0.01
µg/kg, the chickens developed oedema and 83% of them died (Flick et
al., 1972).

Potency for inducing chick oedema was compared for three
different PCDDs: TCDD (purity 91% and > 99%), hexaCDD (purity > 99%,
two isomers), and octaCDD (purity 98.86%) (Schwetz et al., 1973). In
the experiments, 3-day-old white Leghorn cockerels were exposed for 20
to 21 days to one of these congeners at several dose levels. Chick
oedema occurred in birds given oral doses of 1 or 10 µg TCDD/kg per
day, or of 10 or 100 µg hexaCDD/kg per day. Chick oedema was not
observed in chicks maintained on a diet containing 0.1 or 0.5%
octaCDD. The weight of the Bursa of Fabricius was significantly
decreased in 2-week-old white Leghorn cockerels decapitated 2 days
after three daily ip doses of 10 µg TCDD/kg body weight (Sawyer et
al., 1986).

7.1.4 Toxicity of metabolites

Not only has identification of several in vivo metabolites been
achieved, but the acute oral toxicity of TCDD metabolites excreted in
the bile of dogs has been studied in the guinea-pig (Weber et al.,
1982a). 3H-labelled TCDD was administered directly into the duodenal
lumen of two 1-year-old Beagle dogs in four portions of 1 to 2 mg at
time intervals of 2 to 7 days. Excretion of radioactive material from
pooled bile samples collected daily for 4 or 7 days, and thereafter in
pooled samples of 2 or 3 days, was performed with a method yielding
about 50% of the total radioactivity of the bile. The extracts
containing the metabolites were concentrated and dissolved in
1,3-propanediol for administration. Male Pirbright guinea-pigs were
used in a 5-week toxicity study. Five animals in each of four
dose-groups were given a single oral dose via gastric intubation. The
amount of TCDD metabolites was calculated by means of radioactivity
measurements to be 0.6, 6.0, 30.0, or 60.0 µg/kg body weight. Three
animals, one control and two in the high-dose group, died within 48 h
after administration. One death in the high-dose group was due to
gastric perforation during dosing. The other animals which died
exhibited histological lesions which, according to the authors, were
due to material coextracted from bile together with the metabolites.

The light microscopic examination of the liver, spleen, pancreas,
thymus, kidneys, lungs, and adrenals revealed no histological changes,
and no other toxic effects due to the metabolites of TCDD could be
recorded in this study. The authors concluded that TCDD metabolites
from the dog are at least 100 times less toxic to male guinea-pigs
than TCDD itself.

When the metabolites 2-hydroxy-3,7,8-triCDD and
2-hydroxy-1,3,7,8-tetra CDD were given as single ip injections of 100,
1000, or 5000 µg/kg body weight to young male Wistar rats, no decrease
in body weight gain was noted and thymic atrophy was not seen 14 days
later (Mason & Safe, 1986). When compared to TCDD,
1-hydroxy-3,7,8-triCDD was at least 3 orders of magnitude less
effective in inducing hepatic AHH and EROD activities, whereas
2-hydroxy-1,3,7,8-tetra CDD was inactive at all dose levels tested.

7.1.5 Modulation of the acute toxicity

Several studies have been performed which attempt to modify the
acute toxicity of TCDD. Manara et al. (1984) studied the effect of
activated charcoal or cholic acids in the diet on the mortality after
60 days and mean time to death in mice, rats, and guinea-pigs exposed
to TCDD. Dietary levels of charcoal (2.5%), cholic acid (0.25%), and
dehydrocholic acid (0.5%) decreased the mortality induced in C57Bl/6
mice by a single sc dose of 110 mg TCDD/kg body weight from 93% to
53, 21, and 53%, respectively. The mean time to death was prolonged
from 29 days with a normal chow diet to 35-48 days with the above
dietary additives. In CD rats the addition of charcoal (2.5%) and
cholic acid (0.15%) decreased mortality from 80% to 50 and 70%
respectively. The mean time to death was not affected. Charcoal (5%)
added to guinea-pig chow decreased TCDD induced mortality from 64% to
29% and reduced the mean time to death from 29 to 14 days. These
additions to the diet did not prevent body weight loss or liver
enlargement in the mice or guinea-pigs (Manara et al., 1984), but 5%
charcoal in the diet protected against TCDD-induced thymus athrophy in
C57Bl/6 mice 14 days after a single oral dose of 10 mg TCDD/kg body
weight (Manara et al., 1982). The addition of 5% n-hexadecane in the
diet increased the TCDD induced mortality from 60% to 100% and the
mean time to death, within the 50 day study, from 19.8 to 28.8 days in
Sprague Dawley rats treated with a single ip dose of 60 mg TCDD/kg
body weight (Rozman, 1984). N-hexadecane itself did not affect
animal viability.

Increased survival times have been demonstrated in mice receiving
daily injections of triiodothyronine (T3) after TCDD treatment (Neal
et al., 1979) and in rats thyroidectomized before TCDD treatment
(Rozman et al., 1984). Although thyroidectomy increases the mean time
to death, it does not prevent TCDD-related mortality in rats (Rozman
et al., 1985a). Thyroidectomy has been demonstrated to counteract
TCDD-induced decreases in thymus weight and to reduce the spleen

plaque-forming cell response (Pazdernik & Rozman, 1985). A number of
hepatic enzyme activities were induced by TCDD to an equal magnitude
in thyroidectomized and normal rats (Rozman et al., 1985b). It has
been suggested that TCDD-treatment of rats leads to hypothyroidism, a
possible protective mechanism against TCDD toxicity (Pazdernik &
Rozman, 1985). However, available data on changes in T3, thyroxine
(T4), and thyroid-stimulating hormone (TSH) levels in TCDD-treated
rats are not sufficient to state whether the animals are functionally
hypothyroid, euthyroid, or hyperthyroid. Results presented by Potter
et al. (1986b) (see section 8.4.9) suggest that TCDD-treated rats
remain essentially euthyroid and that the altered thyroid status is
neither a major contributor to TCDD toxicity nor a key response to
TCDD exposure.

Daily injections of butylated hydroxyanisole protected against
TCDD-induced lethality in female Sprague Dawley rats, whereas vitamins
E and A, two other antioxidants, did not (Hassan et al., 1985a,b).
Dietary selenium, if given in an optimal dose, provides partial
protection from the lethal effects of TCDD in female Sprague Dawley
rats (Hassan et al., 1985c). None of these treatments could counteract
the TCDD-induced body weight loss (Hassan et al., 1985a,b,c).

7.2 Short-term Toxicity

7.2.1 Studies on rats

Data from the earliest subchronic laboratory study, in which rats
were exposed to daily and/or weekly doses of TCDD, were reported in
four separate papers (published simultaneously) covering general
effects (Harris et al., 1973), pathology (Gupta et al., 1973),
haematological and clinical chemistry changes (Zinkl et al., 1973),
and immuno-biological effects (Vos et al., 1973). Female CD rats were
given TCDD by gavage in daily doses of 0.1, 1, or 10 µg/kg body
weight for 31 days. The body weight of animals exposed to the highest
dose started to decrease within the first week of exposure and 15/16
animals died or became moribund 17 to 31 days after administration of
the compound began. Pathological changes were comparable to those
observed in rats given a single lethal dose, and included severe
thymic atrophy and liver damage, icterus, haemorrhages in various
organs, and the depletion of lymphoid organs. Weight gain was also
reduced at the daily dose level of 1 µg/kg body weight. However, there
were no deaths and in the animals that were killed moderate thymic
atrophy, slight to moderate liver damage, and, in some of the
animals, degenerative changes in the kidneys and in the thyroid gland
were reported. The weight gain was not affected and significant
histopathological changes were not found in rats that received 0.1
µg/kg body weight per day. A decrease in thymic weight that was
significant on day 24, was observed at the lowest exposure level
(Gupta et al., 1973; Harris et al., 1973). Blood samples were
collected 3, 6, 10, 13, 17, 24, and 31 days after administration of

TCDD began. Serum enzyme activities related to liver damage began to
increase after 10 days of exposure and remained high until death
occurred in the 10 mg TCDD/kg per day group. This group of animals
also exhibited increased serum bilirubin levels commencing on day 13.
These parameters were only slightly affected in rats receiving 1 µg
TCDD/kg per day. Thrombocytopenia occurred at all dose levels. After
3 days of treatment with either 1 or 10 µg/kg per day, animals had
depressed platelets counts that remained low throughout the study. In
the low-dose rats, platelets were significantly decreased by day 17.
No significant leukocytopenia or lymphocytopenia occurred in rats at
any dose level. These results are in good agreement with the results
from a more detailed haematological study on female CD rats given
daily oral doses of 10 µg TCDD/kg body weight for 10 and 14 days
(Weissberg & Zinkl, 1973).

When oral doses of 0.02, 1.0, or 5.0 mg TCDD/kg body weight were
given weekly to groups of 10 female CD rats for 6 weeks, all the
animals survived (Harris et al., 1973). However, body weight gain
decreased in the 5.0 µg/kg group during the exposure period, and at
the end of this period the thymus/body weight ratio was approximately
50% of the ratio found in the controls. Liver damage was reported as
slight at this dose level. No effect on body or thymic weight and no
significant histopathological changes were observed in rats given
1 µg/kg body weight or less.

Adult male and female Sprague Dawley rats, in groups of 12, were
given 0, 0.001, 0.01, 0.1, and 1.0 µg TCDD/kg body weight by gavage,
5 days per week for 13 weeks (Kociba et al., 1976). At the end of the
treatment period, five rats of each sex were killed for
histopathological examination, while the remaining animals were
retained for post-exposure observation. Doses of 1 µg TCDD/kg body
weight per day caused five deaths in females, with three occurring
during treatment and two after treatment, and two deaths in males,
both occurring in the post treatment period. Decreased body weights
and food consumption were found at the two highest dose levels both in
males and females. Decreased relative thymus weight and increased
relative liver weight occurred only in the males given the highest
dose but in both the 0.1 and 1.0 µg/kg female groups. Male rats had
significantly depressed haematological values (packed cell volume, red
blood cell count, and haemoglobin) in the two high-dose groups, while
these values were normal in all female rats. Gross, as well as
histological, examination revealed treatment-related effects only in
the high-dose groups with some minor findings in the 0.1 µg/kg group.
Subcutaneous oedema, decreased sizes of testes and uteri, and a
decreased number of corpora lutea were found at necropsy. Histological
findings were limited to lymphoid tissues, liver, and epithelial
linings. The lymphoid tissues, including thymus, were depleted of
lymphocytes. The liver of both male and female rats showed pleomorphic
and multinucleated hepatocytes. Foci of necrosis, with focal
reticuloendothelial aggregations in the areas of parenchymal cell
necrosis, were observed. Hyperplasia of Kupffer cells and an increased
amount of a golden-brown pigment were noted. The hepatic changes were
more pronounced near the periphery of the lobules. Slight hyperplasia

of bile ducts and ductular epithelium was present. The uterus was
lined by cuboidal epithelium in the female rats. The rats given 0.01
mg TCDD/kg did not differ from the controls in any of these
parameters, except for a slight increase in the mean liver to body
weight ratio.

Goldstein et al. (1982) exposed groups of eight female Sprague
Dawley rats to 16 weekly oral doses of 0, 0.01, 0.1, 1.0, and 10.0 µg
TCDD/kg body weight in a study (further discussed in section 8.4.3)
primarily aimed at investigating TCDD-induced porphyria. All animals
given the highest dose died, or were moribund after eight to twelve
doses and were killed. A decrease in body weight gain was seen in this
group within one week of treatment. Decreased body weight gain was
observed also in the 1.0 µg TCDD/kg per week group, but not until
several weeks after the start of treatment. Hepatic porphyria was
found in 7 out of 8 animals receiving weekly doses of 1.0 µg TCDD/kg
body weight, in 1 out of 8 receiving 0.1 µg/kg per week, and in none
of the animals receiving 0.01 µg/kg per week or the lethal dose of
10.0 µg/kg per week. Porphyria was not reversed after six months
recovery from a 16-week exposure to 1.0 µg/kg body weight per week.

Feeding male Wistar rats (110 g) 0, 0.2, or 1.0 µg octaCDD/kg
diet for two weeks, resulting in a total intake of 0, 22.7 (± 1.0) or
120.7 (± 2.8) mg octaCDD, had no effect on body weight gain, feed
consumption, or tissue weights (liver, thymus, testes, heart, and
kidney) (Williams et al., 1972). Congestion of the liver occurred in
the high-dose group.

Daily doses of 100 mg octaCDD for 21 days produced no effects on
appearance, activity, or eating habits in male Sprague Dawley rats,
but slightly increased relative liver weight. A moderate increase in
the hepatic smooth endoplasmic reticulum was noted (Norback et al.,
1975).

7.2.2 Studies on mice

Oral doses of 0.2, 1, 5, or 25 µg TCDD/kg body weight were given
in corn oil to male C57Bl/6 mice weekly for four weeks (Harris et al.,
1973; Vos et al., 1973). One animal of the 25 µg/kg group died after
24 days. Significant weight loss was observed only in the high-dose
group. Thymic atrophy, characterized by nearly complete loss of
cortex, occurred in the 5 and 25 µg TCDD/kg body weight groups.

7.2.3 Studies on guinea-pigs

All 10 female Hartley guinea-pigs that received weekly oral doses
of 1 µg TCDD/kg body weight died, or were killed when moribund,
between days 24 and 32 after the first dose (Gupta et al., 1973;
Harris et al., 1973; Vos et al., 1973). Light microscopic findings of
moribund or dead animals revealed severe atrophy of the cortex of the
thymus with destruction of lymphocytes. There was lymphoid cell

depletion in spleen and lymph nodes. Haemorrhages, mitotic figures,
and loss of lipid vacuoles were observed in the adrenals. Liver
effects were restricted to diffuse single-cell necrosis, predominantly
in the periportal area. Haemorrhages were found in the urinary bladder
and gastrointestinal tract. The lymphocyte count was decreased at all
doses, whereas total leukocyte values were decreased at doses > 0.04
µg/kg. Animals that received eight weekly doses of 0.008, 0.04, or 0.2
µg TCDD/kg body weight all survived. At the 0.2 µg/kg dose level,
decreased body weight gain and decreased relative thymus weight were
observed.

A 90-day feeding study of TCDD in male (250-370 g) and female
(230-340 g) Hartley guinea-pigs was performed by DeCaprio et al.
(1986) and included extensive pathology, haematology, and serum
chemistry on surviving animals. The diets contained 0, 2, 10, 76, or
430 ng TCDD/kg. Animals that received the highest dose exhibited
severe body weight loss, decreased feed consumption, and mortality.
When 60% mortality was reached, on day 46 for males and day 60 for
females, the remaining animals in these groups were sacrificed. The
estimated total TCDD consumption at that time was 1.3 and 1.9 µg
TCDD/kg body weight for males and females, respectively. No
treatment-related mortality was observed at the 76 ng/kg dose level,
which corresponded to a total estimated intake of 0.44 µg TCDD/kg body
weight over the 90 days. Decreased body weight gain and increased
relative liver weight were seen in both sexes, whereas reduced
relative thymus weight occurred in males only. At doses of 2 and 10
ng/kg diet, no dose-related alterations were observed. The only
treatment-related effect on haematology and serum chemistry parameters
was the elevation of serum triglycerides for male guinea-pigs at the
76 ng/kg dose level. The presence of hepatocellular cytoplasmic
inclusion bodies in female guinea-pigs was the only significant
microscopical finding, except for thymus atrophy. Based on this
study, a no-observed-effect level of 0.6 ng TCDD/kg body weight per
day in guinea-pigs was estimated.

DeCaprio et al. (1986) also followed the body weight changes and
mortality during and after feeding male Hartley guinea-pigs (250-360
g) a diet containing 430 ng TCDD/kg. The diet was fed for 11, 21, or
35 days and was then withdrawn during a 79, 69, or 55-day recovery
period. The rate of change of weight per day was the same as that of
animals on a control diet, after an initial weight loss of
approximately 10% during the first 5 days. When animals were fed the
TCDD-diet for 21 and 35 days, a significant mortality, 10% and 70%,
respectively, was apparent. Both body weight gain and absolute body
weight were severely depressed in surviving animals throughout the
study. Animals destined to die generally lost more than 20% of the
original weight, whereas less pronounced weight losses were usually
followed by increases in body weight during the recovery period.

A PCB-containing transformer fire at the State office building,
Binghamton, New York, USA, resulted in contamination of the building
with soot-like material containing various PCDDs and PCDFs (see
section 4.5.10). This soot, mixed with diet, was fed to groups of 10
male (250-350 g) and female (200-350 g) Hartley guinea-pigs for 90
days (0, 0.2, 1.9, 9.3, and 46.3 mg soot/kg diet) or 32 days (231.5 mg
soot/kg diet) (DeCaprio et al., 1983). The total intake of soot during
the study corresponded to approximately 0.3, 3, 13, 67, and 100% of
the LD₅₀ dose of Binghamton soot in guinea-pigs (Silkworth et al.,
1982) (see section 8.1.1.1). A dose-related decrease in body weight
gain occurred at 9.3 and 46.3 mg soot/kg diet. Body weight loss and
decreased feed intake was evident in the animals given 231.5 mg
soot/kg diet. Three male and three female guinea-pigs given 46.3 mg
soot/kg died. Seven animals in the highest-dose group died within 28
to 31 days and the remaining moribund animals were killed on day 32.
Gross pathology revealed no effects at 0.2 mg soot/kg diet. Thymic
atrophy occurred only in males at 1.9 mg soot/kg diet, but in both
sexes at higher doses. The relative spleen weight was significantly
increased at 46.3 mg TCDD/kg in both sexes. Treatment-related
microscopical findings included metaplasia of salivary gland epithelia
(> 1.9 mg soot/kg diet), increased number of goblet cells in
pancreatic ducts (> 46.3 mg soot/kg diet for males), focal lacrimal
gland adenitis (> 9.3 mg soot/kg diet for males and > 46.3 mg
soot/kg diet for females), depletion of haematopoietic cells from the
bone marrow (> 46.3 mg soot/kg diet for females, > 231.5 mg soot/kg
diet for males), and hepatocellular cytoplasmic inclusion bodies (9.3
and 46.3 mg soot/kg diet in both sexes). Fatty infiltration of the
liver, reduced thickness of thymic cortex, and degenerative changes of
the stomach and intestine were observed only in high-dose animals.
Haematological alterations were observed only in animals at the 46.3
mg soot/kg diet dose level, whereas alterations in serum chemistry
values were found also at lower levels. Toxic effects of feeding
Binghamton soot for 90 days were similar to the effects occurring
after acute exposure (Silkworth et al., 1982) (see section 8.1.1.1),
but the effects were seen at a lower total dose after subchronic
exposure than after acute dosing. The effect seen at the lowest level
in this study was thymic atrophy at 1.9 mg soot/kg diet, which was
equivalent to 7.8 ng TCDD/kg body weight per day. A comparison between
the effects of feeding pure TCDD and TCDD-contaminated soot to
guinea-pigs (DeCaprio et al., 1986) demonstrated that pure TCDD
produced less variability of alterations and gave a steeper
dose-response relationship for many effects. Exposure to Binghamton
soot characteristically resulted in salivary gland duct metaplasia and
decreased serum sodium and potassium levels (DeCaprio et al., 1983).

When male guinea-pigs (200 g) were fed a diet containing 2.5%
HCl-pretreated fly ash from a municipal incinerator (Zaanstad, The
Netherlands) for up to 95 days, the animals exhibited progressive
weight loss, hair loss, and increased relative liver weight (van den
Berg et al., 1986b). One animal died on day 76.

Table 50. Studies on long-term exposure (excluding cancer studies) to TCDD in laboratory animals

Species/Strain Sex/number/ Doses tested Treatment Parameters
group^d schedule monitored

Rats
Sprague Dawley^a M/10 0 ng/kg in diet survival
1 ng/kg continuously
5 ng/kg for 65 weeks
50 ng/kg
500 ng/kg
1000 ng/kg
5000 ng/kg
50 000 ng/kg
500 000 ng/kg
1 000 000 ng/kg

Sprague Dawley^b M and F/10 0.001 µg/kg per day in diet extensive
0.01 µg/kg per day continuously histopathology,
0.1 µg/kg per day for 2 years haematology and
clinical chemistry

Mice
Swiss M/38-44 0 µg/kg/week by gavage weekly histopathology
0.007 µg/kg per week for 1 year
0.7 µg/kg per week
7.0 µg/kg per week

Monkeys
Macaca mulatta^c F/8 500 ng/kg continuous in extensive
the diet for histopathology,
9 months haematology and
clinical chemistry

^a Van Miller et al. (1977). ^b Kociba et al. (1978, 1979a,b).
^c Allen et al. (1977). ^d M = male; F = female.

7.2.4 Studies on hamsters

No toxic effects were reported in male Golden Syrian hamsters
(50-70 g) given a diet containing 2.5% HCl-pretreated fly ash from a
municipal incinerator in Zaanstad, The Netherlands for up to 95 days
(van den Berg et al., 1986b).

7.2.5 Studies on monkeys

A cumulative dose of 0.2 µg TCDD/kg body weight, divided into
nine oral doses given three times per week, produced no clinical
toxicity in female rhesus monkeys (Macaca mulatta) (McNulty,
1984). However, clearly toxic signs did occur in those monkeys that
received cumulative doses of 1.0 and 5.0 µg TCDD/kg body weight over
the same time period. The first signs were thickening and reddening of
the eyelids, followed by weight loss, dryness and granularity of the
skin, and loss of hair, and in some cases anaemia, purpura, and
bleeding from the nose and mouth. Animals that died showed squamous
metaplasia of the sebaceous glands, mucous metaplasia of the gastric
mucosa, hyperplasia of biliary ductal epithelium, gingivitis, and
hypoplasia of the bone marrow. The times to death after dose were 65
and 116 days at 5 µg/kg and 130 to 211 days at 1 µg/kg.

7.3 Long-term Toxicity

Chronic toxicity studies performed on laboratory animals exposed
to TCDD are summarized in Table 50. Studies on carcinogenicity are
presented in section 7.7.

7.3.1 Studies on rats

In studies by Van Miller et al. (1977), male Sprague Dawley rats
were maintained in groups of 10 on diets containing 0, 1, 5, 50, 500,
1000, 5000, 50 000, 500 000, and 1 000 000 ng TCDD/kg for 65 weeks and
survival was monitored. At the five highest dose levels, all animals
died before the study was completed. The first deaths in these treated
groups occurred by weeks 31, 31, 3, 2, and 2 of treatment,
respectively. Groups receiving 50, 500, or 1000 ng TCDD/kg in the diet
died from acute toxic effects including severe liver necrosis, bile
duct hyperplasia and oedema, atrophy of spleen and thymus, and
gastrointestinal haemorrhages.

Groups of 50 male and 50 female Sprague Dawley rats were fed
diets providing daily doses of 0.001, 0.01, and 0.1 µg TCDD/kg body
weight for 2 years (Kociba et al., 1978, 1979a,b). Control rats, 86
males and 86 females, received diets to which the vehicle acetone had
been added. The dose levels corresponded to a dietary content of 22,
208, and 2193 ng TCDD/kg feed. Increased mortality was observed in
females given 0.1 µg/kg per day, while no increased mortality was
observed in male rats at this dose or in animals receiving doses of

0.01 or 0.001 µg/kg per day. From month 6 to the end of the study the
mean body weights of males and females decreased at the highest dose
and to a lesser degree in females given 0.01 µg/kg per day. During the
course of the study, subnormal body weights were occasionally also
recorded in the low-dose group, although during the last quarter of
the study the body weights were similar to those of the controls.

Increased urinary coproporphyrin and uroporphyrin were noted in
females, but not in males, given TCDD at a dose rate of 0.01 and 0.1
µg/kg per day. Analyses of blood serum collected at terminal necropsy
revealed increased enzyme activities related to impaired liver
function in female rats given 0.1 µg TCDD/kg per day. Necropsy
examination of the rats surviving TCDD exposure to the end of the
study revealed that liver effects constituted the most consistent
alteration in both males and females. Histopathological examination
revealed multiple degenerative inflammatory and necrotic changes in
the liver that were more extensive in females. Multinucleated
hepatocytes and bile duct hyperplasia were also noted. Liver damage
was dose-related and no effect was observable at the lowest dose
studied.

7.3.2 Studies on mice

Weekly oral doses of 0, 0.007, 0.7, and 7.0 µg TCDD/kg body
weight for 1 year resulted in amyloidosis and dermatitis in male Swiss
mice (Toth et al., 1979). The incidence of these lesions was 0/38,
5/44, 10/44, and 17/43 in the control, low-, medium-, and high-dose
groups, respectively.

7.3.3 Studies on monkeys

In a study by Allen & Carstens (1967) groups of four to five
rhesus monkeys were fed diets containing 0, 0.125, 0.25, 0.5, 1.0, and
10.0% of fat (which had been shown to be toxic to chickens) until
death. The "toxic fat" was later demonstrated to contain various
PCDDs, of which 65% by mass of the total PCDDs present was TCDD
(Norback & Allen, 1973). The survival time became shorter with
increasing doses of "toxic fat". Mean time to death was 445 days for
the low dose and 91 days for the high dose. Decreased food consumption
and progressive body weight loss, compared with controls, were noted.
Both clinical and histological changes near the time of death appeared
similar regardless of dose. The monkeys developed subcutaneous oedema,
progressing from the eyelids and face, ascites, and hydropericardium.
Characteristic skin changes were observed as well as anaemia,
leukopenia, and hypoproteinaemia. The bone marrow was hypoplastic.
Centrilobular necrosis, bile duct hyperplasia, and multinucleated
hepatocytes were found in the liver. In more than half of the
animals, there was marked hypertrophy of the gastric mucosa, with
crypts and mucin-containing cysts penetrating into the submucosa.
Ulcerations in the fundic and pyloric regions were observed.

Allen et al. (1977) fed eight adult female rhesus monkeys a diet
containing 500 ng TCDD/kg for 9 months. There after, surviving animals
were removed from the TCDD diet and were observed for another 4
months. No control animals were included, and so data were compared
with pre-exposure values where possible. During the first 3 months of
exposure animals developed periorbital oedema, acne, and loss of
facial hair and eyelashes. By 6 months these changes were quite
prominent in six out of eight monkeys, and a decrease in haemoglobin
haematocrit was noticed. The animals lost weight even though their
food intake was unaltered. Two animals died within the 9-month
exposure period and three monkeys continued to develop toxic symptoms
and died after 3 months on a TCDD-free diet. The three surviving
animals continued to experience periorbital oedema and loss of hair.
The total intake of TCDD over the 9-month period was calculated to be
2-3 µg/kg body weight. Death was preceded by severe anaemia, a
decreased white blood cell count, and severe thrombocytopenia. Autopsy
findings included haemorrhage into a variety of organs, ascites, and
subcutaneous oedema. Hypertrophy, dilatation, oedema, and hydropic
degeneration of the myocardium were noted in all animals. The biliary
ducts showed marked dilatation. Moderate hyperkeratosis of the skin,
with cystic keratosis of the hair follicles, was noted, and
hypocellularity of the lymphoid tissue and the bone marrow were
observed. The hyperplastic mucous-secreting cells of the gastric
epithelium had invaded the submucosa, and ulceration and mucinous
cysts were common in the modified gastric mucosa. Hypertrophy and
hyperplasia of the epithelial lining of the biliary system were
present. The bronchial epithelium, salivary glands, bile ducts, and
pancreatic ducts showed metaplastic changes. Death was attributed to
complications from the severe pancytopenia. The same pattern of
morphological changes were reported to occur in a similar study
performed by Barsotti et al. (1979).

Similar, though less severe, effects were observed in four adult
female rhesus monkeys fed a diet of 50 ng TCDD/kg for 20 months
(Schantz et al., 1978).

7.4 Effects Detected By Special Studies

7.4.1 Wasting syndrome

TCDD causes a starvation-like or wasting syndrome in several
animal species. In young animals, or following a sublethal dose in
adults, this response is manifested as a cessation of weight gain.
Early studies suggested that acute or chronic treatment with TCDD
decreased food consumption, but insufficiently to account for the
weight loss (Allen et al., 1975, 1977; Greig et al., 1973; Harris et
al., 1973; Kociba et al., 1976; McConnell et al., 1978a,b). To
elucidate whether malabsorption could explain the wasting syndrome,
the transfer of a number of nutrients has been studied with everted
intestinal sacs from TCDD-treated rats. A transient increase in the

serosal transfer of ⁵⁹Fe in Sprague Dawley rats was reported by
Manis & Kim (1979). Absorption of glucose (Ball & Chabra, 1981; Madge,
1977) and lipids (Shoaf & Shiller, 1981) was decreased by TCDD
treatment. The absorption of cobalt, galactose, and proline (Manis &
Kim, 1977) as well as of D-galactose, L-arginine, L-histidine (Madge,
1977), and penicillin (Manis & Apap, 1979) was reported to be
unaffected by TCDD treatment. Leucine transport was depressed in
Sprague Dawley rats 4 h after a single oral dose of 100 µg TCDD/kg
body weight (Ball & Chabra, 1981), whereas no effect was observed in
Fisher rats 7 days after exposure to 80 µg TCDD/kg body weight
(Schiller et al., 1982). Neal et al. (1979) found normal absorption
and intermediary metabolism of glucose, L-alanine, and oleate in
guinea-pigs treated with a single oral dose of 2 µg TCDD/kg body
weight. Apparently there was no generalized impairment of intestinal
absorption. The effects reported may well be secondary to decreased
food consumption which by itself causes structural changes in the
intestine (Steiner et al., 1968) as well as impaired absorption of
nutrients (Esposito et al., 1967).

The connection between the wasting syndrome and the lethal effect
of TCDD has been investigated in pair-feeding and forced nutrition
studies. Courtney et al. (1978) fed TCDD- treated female Wistar rats
a normal pelleted diet ad libitum. Supplementation with water,
electrolyte solution, or liquid diet, administered by gavage, could
not reverse or change the pattern or extent of TCDD-induced weight
loss or mortality.

To bypass gastrointestinal absorption, Gasiewicz et al. (1980)
fed rats intravenously with total parenteral nutrition (TPN). Rats
that had received a single ip dose of 100 µg TCDD/kg body weight
gained weight similarly to their TPN-fed controls, yet still died at
days 13 to 17 following treatment. TCDD-treated rats fed a chow diet
ad libitum lost weight progressively (as compared to pair-fed
controls that maintained their starting weight) and died at days 11 to
20. In TPN-fed TCDD-treated rats, liver damage was more severe and fat
depots were increased as compared to chow-fed TCDD-treated animals.
Similar results were obtained with TPN-fed male Hartley guinea-pigs
treated with a single ip dose of 2 µg TCDD/kg body weight in olive
oil, when compared to TPN-fed control guinea-pigs (Lu et al., 1986).
Similar signs of toxicity were present in TPN- and ad libitum-fed
TCDD-treated guinea-pigs. In contrast to TPN-fed rats (Gasiewicz et
al., 1980), TCDD- treatment in TPN-fed guinea-pigs only produced mild
hepatic changes, including increased liver lipid and cytochrome P 450
content, but no morphological changes (Lu et al., 1986).

Seefeld et al. (1984a) suggested that TPN-fed TCDD- treated rats
might have suffered from overnutrition and, secondary to that,
enhanced hepatotoxicity, as compared to chow-fed, TCDD-treated rats.
These same investigators have presented a heuristic model for the

TCDD-induced wasting syndrome based on the assumption that body weight
in rats is regulated around an internal standard or set point (Keesey
et al., 1976). Prevailing weight at a given age is constantly being
compared to this set point value and if differences occur, feed
consumption is adjusted so as to raise or lower body weight to match
the set point level. If TCDD lowers this setpoint, reduction in food
consumption would result, as the rat attempts to reduce its weight to
a new lower level of regulation determined by the dose of TCDD
administered. This hypothesis has been tested in several experiments
under carefully controlled feeding procedures.

Repeated studies have demonstrated that reduction of feed intake,
due to increased food spillage, is sufficient to account for the loss
of body weight in TCDD-treated Sprague- Dawley rats (Seefeld &
Peterson, 1983, 1984; Seefeld et al., 1984a,b). TCDD-treated rats
maintain and defend their reduced weight level with the same precision
as control rats (fed ad libitum) defend their normal weight level
(Seefeld et al., 1984b). The percentage of the daily feed intake that
is absorbed by the gastrointestinal tract of TCDD-treated and control
rats is similar (Potter et al., 1986a; Seefeld & Peterson, 1984).
Water intake, resting and total oxygen consumption, carbon dioxide
production, respiratory quotient, and spontaneous motoractivity were
decreased in a dose-dependent manner by TCDD-treatment (Potter et al.,
1986a; Seefeld & Peterson, 1983; Seefeld et al., 1984a). Urine output
was unaffected by TCDD-treatment, despite decreased water intake,
whereas urinary excretion of energy and urea were decreased and
urinary ammonia was increased (Potter et al., 1986a).

Hypophagia was the major cause of the loss of adipose and lean
tissue in male Fisher F-344 rats, C57Bl/6 mice, and albino guinea-pigs
when exposed to a calculated LD80 dose of TCDD (Kelling et al., 1985).
Body weight loss followed a similar time course in TCDD-treated and
pair-fed control animals of all three species. Lethalities for
TCDD-exposed rats, mice, and guinea-pigs were 95%, 69%, and 81%,
respectively, compared to lethalities in the appropriate pair-fed
controls of 48, 14, and 64%, respectively (Kelling et al., 1985).
Lethality and body weight loss followed almost identical
time-course-curves in Sprague Dawley rats that received a single oral
dose of 75 µg TCDD/kg body weight and in pair-fed controls (Christian
et al., 1986a). Thus the contribution to lethality made by body weight
loss seems to depend on the species and strain. Weight loss appears to
play a greater role in causing death in Sprague Dawley rats and
guinea-pigs than in Fisher F-344 rats and C57Bl/6 mice. Christian et
al. (1986a) demonstrated differences in organ weights and
histopathology in TCDD-treated and pair-fed animals, despite similar
time-courses and magnitude of body weight loss and lethality. Pair-fed
animals exhibited lesions in the gastro-intestinal tract, which were
absent in TCDD-treated rats, that may have contributed to death. The
hepatic carbohydrate, protein, and lipid metabolism was affected

differently in TCDD-treated and pair-fed Sprague Dawley rats
(Christian et al., 1986b). To distinguish direct effects of TCDD from
effects secondary to hypophagia, the studies by Christian et al.
(1986a,b) and Potter et al. (1986a) were performed with schedule-fed
animals. The reason for this was the finding that the 24-h feeding
pattern for TCDD-treated rats was different from the feeding pattern
for pair-fed controls, although it was similar to that of control rats
fed ad libitum. Decreased feed consumption did not contribute to
weight loss in C57BL/6 mice exposed to TCDD until the animals were
moribund (Chapman & Schiller, 1985).

Besides being typical signs of TCDD toxicity, loss of body weight
and appetite are also prominent signs of thyroid dysfunction (see
section 8.4.9). Serum glucose levels were also decreased by TCDD
independently of hypophagia, whereas the decrease in serum insulin
appeared to result from hypophagia, since it was seen in both
TCDD-treated and pair-fed controls. These results indicate that the
effect of TCDD on thyroid hormones cannot explain the TCDD-induced
decrease in body weight gain.

An interesting biochemical effect in TCDD-induced wasting is the
ability of TCDD to decrease hepatic vitamin A storage in rats (see
section 8.4.10). It has long been known that vitamin A is necessary
for growth and that vitamin A deficiency will result in depressed body
weight gain as well as in reduced food intake. However, the animal
continues to eat and grow though body weight gain is less than normal
(Brown & Morgan, 1948; Coward, 1947; Hayes, 1971; Orr & Richards,
1934; Patterson et al., 1942).

The effect of chemical structure on the ability of several PCDDs
to cause body weight loss in rats has been investigated (Mason et al.,
1986). Of the congeners studied, 2,3,7,8-TCDD was the most active.
Those congeners fully substituted in the 2,3,7, and 8 positions but
containing additional chlorosubstituents in the non-lateral 1,4,6, and
9 positions were less active.

7.4.2 Hepatotoxicity

TCDD produces hepatomegaly, due to hyperplasia and hypertrophy of
parenchymal cells, in all species that have been investigated, even at
sublethal doses. However, there is considerable variation between
species in the extent of this lesion. Other liver lesions are more
species specific.

Liver lesions alone cannot explain lethality following TCDD
administration, though it may be a contributing factor at least in the
rat and rabbit.

The morphological changes in the liver are accompanied by
impaired liver function, characterized by liver enzyme leakage,
increased microsomal monooxygenase activities, porphyria, impaired
plasma membrane function, hyperlipidaemia, and increased regenerative
DNA synthesis.

7.4.2.1 Morphological alterations

In Charles River rats given single oral sublethal doses of TCDD
(5 or 25 µg/kg body weight) a dose-related increase was observed 3
days after dosing in the amount of hepatic smooth endoplasmatic
reticulum (SER) around the periphery of cells, particularly in the
areas around bile canaliculi. The effect progressed by days 6 and 9,
when an increased amount of rough endoplasmatic reticulum (RER) was
also present. By day 28 these changes had returned essentially to
normal levels (Fowler et al., 1973).

The livers of CD rats given high sublethal doses of TCDD (5.0
mg/kg body weight per week) for six weeks showed transient
degenerative changes, followed by megalocytosis, regeneration, and the
occurrence of multinucleated giant hepatocytes (Gupta et al., 1973).
They also showed that the hepatotoxic reaction in rats given lethal
doses of TCDD (10 µg/kg body weight per day for 16-31 days) was
characterized by degenerative and necrotic changes, with the
appearance of mononuclear cell infiltration, multinucleated giant
hepatocytes, increased numbers of mitotic figures, and pleomorphism of
cord cells. These lesions were considered severe enough to be a
contributing factor to death.

Parenchymal cell necrosis was observed by Greig et al. (1973) in
Porton rats 3 weeks after exposure to an LD₅₀ dose of TCDD. The
necrosis, which was located in the centrilobular zone close to the
central vein, became more severe with time.

Jones & Butler (1974) further investigated the time course of the
TCDD-induced liver lesions appearing in the centrilobular zone. They
confirmed the transient degenerative and inflammatory lesions
previously reported (Greig et al., 1973; Gupta et al., 1973). At the
ultrastructural level, consistent changes occurred in the cytoplasm
whereas normal nuclear morphology and division were found throughout
the study. Two weeks after a single oral dose of 200 µg TCDD/kg body
weight to Porton rats, extensive fusion of parenchymal cell plasma
membranes in the centrilobular zone was replaced by a diffuse zone
with islands of normal membrane occurring at intervals. Normal tight
and gap junctions were present in control animals and in periportal
areas of the test animals. These findings suggest that the
multinucleated cells occurring in TCDD-treated rats might form by
coalescence of parenchymal cells. The effect of TCDD on plasma

membranes demonstrates a specific subcellular site of action, which
might be involved in the toxic action of TCDD. This lesion was,
however, not observed until 2 weeks after treatment and thus could not
explain the immediate effects on, for instance, food intake, body
weight gain, and general health.

The time course for liver lesions in male Sprague Dawley rats
(200 g) given a single ip dose of 20 µg TCDD/kg body weight was
followed for up to 32 weeks by Weber et al. (1983). The lesions became
progressively worse up to the 16th week after injection, and
thereafter appeared to regress slowly. The lesions were almost
identical to those reported previously (Fowler et al., 1973; Jones &
Butler, 1974). The histological findings were accompanied by
hyperbilirubinaemia, hypercholesterolaemia, hyperproteinaemia, and
increased serum glutamicoxaloacetic transaminase and serum
glutamic-pyruvic transaminase activities, further indicating decreased
liver function (Greig et al., 1973; Zinkl et al., 1973).

Fewer studies of liver lesions produced by TCDD have been made in
other species than in the rat.

In C57BL/6 mice given single oral doses of 100, 150, or 200 µg
TCDD/kg body weight (Vos et al., 1974) and 250 µg TCDD/kg body weight
(Jones & Greig, 1975), centrilobular degenerative and necrotic changes
were present but multinucleated parenchymal cells were not seen.
Proliferation of the bile ducts and bile duct epithelial cells, as
well as lipid accumulation, have been observed, with a substantial
increase in the hepatic levels of esterified fatty acids and
cholesterol. Only slight damage, hepatocellular swelling, was reported
in CD-1 mice 21 days after a single dose of 50 µg TCDD/kg body weight,
and no histological changes were detected 7 or 35 days after
administration (Gupta et al., 1973).

The guinea-pig, while being very sensitive to TCDD, as indicated
by LD₅₀ data, shows less severe morphological alterations in the
liver than do other species. No manifest liver lesions at the light
microscopical or ultrastructural levels has been found (Gupta et al.,
1973; McConnell et al., 1978b; Moore et al., 1979; Turner & Collins,
1983).

The hamster is very resistant to TCDD toxicity (Table 47) and
exhibits no manifest liver damage even after a fatal dose (Henck et
al., 1981; Olson et al., 1980b). However, Gasiewicz et al. (1986)
found bile duct hyperplasia, numerous inflammatory cells, and
increased number of multinucleated cells in the livers of male Golden
Syrian hamsters 35 days after they received a single ip dose of 500 µg
TCDD/kg body weight in olive oil.

7.4.2.2 Hepatic plasma membrane function

The morphological impairment of hepatic plasma membranes in the
centrilobular parenchymal cells of TCDD-treated Porton rats was
demonstrated histochemically to be preceeded by a loss of ATPase
activity (Jones, 1975). The effect occurred 3 days after treatment in
an area five to six cells deep around the central vein along the
canalicular borders, and became more severe with time. At the end of
the study (day 42 posttreatment) the ATPase activity was completely
abolished around the central vein, including the mid-zonal region, and
encroached on the periportal area in moribund animals. The loss of
ATPase activity was related to the clinical state of the animal. Thus
animals displaying minimal signs of intoxication retained the normal
distribution of ATPase in the periportal zone. In animals killed 9
months after treatment, partial restoration of the normal liver
architecture and the ATPase activity were evident.

Biochemical studies of isolated heptatic plasma membranes from
Holtzman rats treated with 10 or 25 µg TCDD/kg body weight revealed
depressed ATPase activities (Peterson et al., 1979a). The activity of
Na/K-ATPase was depressed to the same extent for both doses from day
2 to 40 after treatment, while a similar depression of Mg-ATPase
activity was observed only in the high-dose group. The Mg-ATPase
activity tended to recover by day 40, whereas Na/K-ATPase activities
did not. A pair-feeding experiment demonstrated that these effects
were independent of the TCDD-induced decrease in food consumption. In
vitro incubation of plasma membranes indicated that ATPase inhibition
did not occur by direct interaction with TCDD. Greig & Osborne (1981)
demonstrated a decrease in K/Mg-ATPase activity, but not in the
Na/K-ATPase or 5'-nucleotidase activities of hepatic plasma membranes
prepared from female Porton rats 6 and 11 days after a dose of 200 µg
TCDD/kg body weight.

Many physiological homeostatic mechanisms are dependent on proper
plasma membrane function and composition. Matsumura et al. (1984)
reported that a single ip dose of 25 µg TCDD/kg body weight to male
Sprague Dawley rats had reduced hepatic plasma membrane
ATPase-activities by 40% 10 days after treatment. The marker enzyme
for putative preneoplastic hepatocytes, glutamyl transpeptidase, was
reduced, while protein kinase (both c-AMP-stimulated and
c-AMP-nonstimulated) was increased (Matsumura et al., 1984). Both
c-AMP-dependent and c-AMP-independent protein kinase, in hepatic
plasma membranes from male Sprague Dawley rats treated with an ip dose
of 25 µg TCDD/kg body weight, were maximally increased on day 20 after
treatment (4.5- and 12-fold, respectively). The induction was
measurable within 2 days after the administration and was still
persistent after 40 days (Bombick et al 1985). Protein kinase C was
significantly increased in hepatic plasma membranes from Sprague
Dawley rats but not from guinea-pigs 10 days after single ip doses of
25 and 1 µg TCDD/kg body weight, respectively (Bombick et al., 1985).
TCDD treatment in vivo also affected the in vitro binding of

concanavalin A, epidermal growth factor, and insulin to their cell
surface membrane receptors.

The binding of glucagon and prostaglandin E were not affected by
a dose of 25 µg TCDD/kg body weight (Matsumura et al., 1984). Studies
with Sprague Dawley rats demonstrated that the TCDD-induced decrease
in epidermal growth factor binding (EGF) was observable within 2 days
after dosing, reached its maximum by day 20, and was still significant
40 days after dosing. The decrease was observable after a single ip
dose of 0.1 µg TCDD/kg body weight (Madhukar et al., 1984). The
relative doses of TCDD needed to suppress EGF binding to 50% of the
control level were 1, 14, and 32 µg/kg body weight for the guinea-pig,
the Sprague Dawley rat, and the Golden Syrian hamster, respectively
(Madhukar et al., 1984). A single ip dose of 115 µg TCDD/kg body
weight had decreased the EGF binding 10 days after treatment by 93.1,
97.8, and 46.0% in C57Bl/6, CBA, and AKR mice, respectively (Madhukar
et al., 1984). The effect of daily sc injections of 2 ng EGF/kg body
weight to newborn Balb/c-mice was compared with the effect of a single
ip dose of 10 mg TCDD/kg body weight given to the dam of newborn
Balb/c-mice within 3 h of delivery (Madhukar et al., 1984). The
parameters studied, all well known in vivo effects of EGF, were:
time for eyelid opening and tooth eruption; hair length and diameter
on day 14; and body weight and thymus weight on day 22. All parameters
were significantly reduced both by TCDD treatment and EGF treatment,
as compared to controls.

Bombick et al. (1984) found that, 10 days after treatment, in
vitro binding of ¹²⁵I-low density lipoprotein (LDL) to its receptor
on hepatic plasma membranes was decreased by 73% in TCDD-treated
guinea-pigs (1 µg TCDD/kg body weight) as compared to pair-fed
controls. Primary hepatocytes from guinea-pigs treated in the same way
had a reduced ability to internalize ¹²⁵I-LDL. The reduction of LDL
receptors on hepatic plasma membranes might be responsible for the
increase in plasma of very low density lipoproteins (VLDL) and LDL
noted in TCDD-treated animals.

Quantitative changes in the protein composition of plasma
membranes following TCDD treatment have been reported (Brewster et
al., 1982; Matsumara et al., 1984). The membranes were isolated from
male Sprague Dawley rats 10 days after an ip injection of 25 µg
TCDD/kg body weight and analyzed by SDS-polyacrylamide gel
electrophoresis. Some small proteins (14 000-30 000 daltons) were
completely abolished by TCDD treatment. The effect was most pronounced
10 to 20 days after treatment.

7.4.2.3 Biliary excretion

The early proliferation of liver cells around bile canaliculi
seen after TCDD treatment (Fowler et al., 1973) was suggestive of an
effect on biliary excretion.

The cumulative biliary excretion of indocyanine green (ICG), an
organic anion, was decreased in a dose-dependent manner by treatment
with 5 or 25 µg TCDD/kg body weight in CD rats (Hwang, 1973). On the
contrary, biliary excretion of the organic anions sulfobromophthalein
and phenol-3,6-dibromophthalein was unaffected by treatment with 10 or
25 µg TCDD/kg body weight in Holtzman rats (Yang et al., 1977).

Biliary excretion of ouabain, a model compound for neutral
non-metabolized substrates such as estradiol, progesterone, and
cortisol, was depressed in male Holtzman rats in a dose-related manner
by a single oral dose of 10 or 25 µg TCDD/kg body weight (Yang et al.,
1977). The effect was detectable two days after treatment, reached a
peak between 10 and 20 days, and recovery was only slight by day 40.
Increased plasma concentration and decreased bilary excretion of
ouabain was also measured in male Sprague Dawley rats 10 days after a
single oral dose (25 µg/kg body weight) of TCDD, or
1,2,3,7,8,9-hexaCDD (Yang et al., 1983b). Two other congeners,
1,2,4,6,7,9-hexaCDD and 1,3,6,8-tetraCDD had no effect on these
parameters.

When hepatocytes from TCDD-treated (10 µg/kg body weight) male
Sprague Dawley rats were incubated with labelled ouabain or procaine
amide ethobromide (PAEB) 10 days post-treatment, both the rate of
uptake and the steady-state concentration of ouabain were decreased,
whereas the uptake of PAEB was unaffected by TCDD (Eaton & Klaassen,
1979). The dose of TCDD was very small relative to ouabain
(approximately 150 µmol/litre), so it is not likely that TCDD exerted
its effect by competing with the drug for transport into bile. These
data suggest that the hepatic membrane transport process for ouabain
may be selectively damaged by TCDD.

Peterson et al. (1979a) observed a positive correlation between
the levels of hepatic plasma membrane ATPase activities, biliary
excretion of ouabain, and bile flow in vivo after TCDD treatment.
However, in a further experiment, using perfused rat liver, Peterson
et al. (1979b) demonstrated that biliary excretion of ouabain and
liver membrane ATPase activities could be decreased independently.
Therefore, ATPase activities cannot be directly responsible for the
reduced ouabain excretion.

7.4.3 Porphyria

Chronic sublethal exposure to TCDD produces an accumulation of
porphyrins in the liver and an increase in urinary porphyrin
excretion. In stages of manifest porphyria, accumulation of porphyrins
occurs not only in the liver but also in the kidney and spleen
(Goldstein et al., 1982). It was demonstrated that mice respond to
four weekly doses of 25 µg TCDD/kg body weight with hepatic porphyria,
accompanied by an increase in aminolevulinic acid (ALA) synthetase
activity, and liver lesions (Goldstein et al., 1973), whereas a single

dose of 5, 25, or 100 µg TCDD/kg body weight did not induce porphyria
nor ALA synthetase activity in the rat (Woods, 1973). The suggested
species difference was later ruled out by Cantoni et al. (1981) and
Goldstein et al. (1976, 1982). Chronic administration of 1 µg TCDD/kg
body weight per week to rats for 16 weeks resulted in hepatic
porphyria (Goldstein, 1976a,b, 1982). In contrast, single oral doses
as high as 30 µg TCDD/kg did not produce porphyria either acutely or
16 weeks later. A 6-month recovery period following the final dose was
not long enough to reverse the porphyria. Urinary porphyrins and
hepatic ALA synthetase activity remained maximally elevated, while
hepatic porphyrin levels decreased during this period. Failure to
demonstrate porphyria in rats after chronic administration of TCDD for
13 weeks (Kociba et al., 1976) or 2 years (Kociba et al., 1978) was
suggested to be the result of unsatisfactory porphyrin analysis
(Goldstein et al., 1982).

To further characterize TCDD-induced porphyria, Cantoni et al.
(1981) performed a 45-week study to follow the pattern of porphyrin
excretion in rats exposed orally to 0.01, 0.1, or 1.0 µg TCDD/kg body
weight/week. They found an increase in the coproporphyrin level in the
initial phase of exposure, which remained the only sign of exposure in
the lowest-dose group. A marked porphyric state appeared only in the
1.0 µg/kg dose group, commencing 8 months after dosing started. At
that time urinary porphyrin excretion was 70 times higher than in
control rats. The excretion pattern was characterized by increased
levels of carboxylated porphyrins.

In attempts to elucidate the mechanism of TCDD-induced porphyria,
the effects of TCDD on the enzymes involved in the synthesis and
catabolism of porphyrins have been studied. TCDD was found to be a
potent inducer of ALA synthetase, the initial and rate-limiting enzyme
in haeme synthesis in the liver of chicken embryos (Poland & Glover,
1973b). Elevated ALA synthetase activity has since been demonstrated
also in mice and rats (Goldstein et al., 1973, 1982; Kociba et al.,
1976, 1978). However, the TCDD-induced increase does not appear after
acute exposure and only after several weeks of chronic exposure to
TCDD. Jones & Sweeny (1980) failed to demonstrate increased ALA
activity in mice exposed to 25 µg TCDD/kg body weight per week for 11
weeks, although porphyria was evident. Thus, induction of ALA
synthetase does not seem to be the primary event in TCDD-induced
porphyria. Elder et al. (1976, 1978) suggested that decreased hepatic
porphyrinogen decarboxylase is the primary event in porphyria induced
by halogenated aromatics. TCDD depresses this enzyme activity in
vivo in the liver of mice (Jones & Sweeny, 1980; Elder & Sheppard,
1982; Cantoni et al., 1984a,b) but not in vitro (Cantoni et al.,
1984b).

A decrease in porphyrinogen decarboxylase activity was present
one week after a single dose of 75 µg/kg body weight and continued to
decrease with time, thus preceding the increase in hepatic porphyrins,
which started to rise during the first 2 weeks after treatment (Smith
et al., 1981). TCDD- induced depression of hepatic uroporphyrinogen
decarboxylase activity occurs also in the newly regenerating liver, as
demonstrated by Smith et al. (1985) in C57Bl/10 mice 10 days after
partial (2/3) hepatectomy. The mice were treated with 75 µg TCDD/kg
body weight orally 4 weeks before hepatectomy. Greig et al. (1984)
demonstrated that pretreatment of five different strains of mice with
12.5 mg Fe²⁺ one week before the administration of 75 µg TCDD/kg
body weight had a synergistic effect on porphyria assessed 5 weeks
after dosing, expressed as increased hepatic porphyrin and decreased
porphyrinogen decarboxylase activity. Iron alone did not increase
hepatic porphyrin levels, nor did it affect hepatic porphyrinogen
decarboxylase activity.

7.4.4 Epidermal effects

Chloracne and associated pathological changes in the skin are
among the most sensitive and widespread responses to TCDD in humans.
Similar skin toxicity is expressed only in a limited number of animal
species, namely rabbits, monkeys, and hairless mice. To characterize
the epidermal response and to elucidate the mechanism(s) of toxicity
to epidermal cells, numerous studies have been performed both in
vivo and in vitro.

7.4.4.1 In vivo studies

The acnegenic activity of TCDD and related compounds has been
tested in the rabbit ear bioassay, first developed by Adams et al.
(1941) for industrial applications. The test substance is applied to
the inner surface of one of the ears, while pure vehicle is applied to
the other. Responses indicative of acnegenic activity include comedo
formation, increased ear thickness, and hyperkeratosis. Mild
irritation, increased ear thickness, slight enlargement of follicular
aperture, slight exfoliation and slight crust formation alone are not
considered indicative of acnegenic activity. Microscopically there is
conversion of sebaceous cells in the hair follicles into
keratin-forming cells. A dose-dependent, positive response was found
in this assay when a total dose of 1, 3, or 10 µg TCDD was applied on
three successive days (Jones & Krizek, 1962). Also Schwetz et al.
(1973) found a dose-dependent acnegenic response in the rabbit-ear
bioassay after repeated applications of 4-40 µg TCDD/ear, five days
per week for four weeks, corresponding to total doses of 80-800 µg. No
response was obtained when the total application was 8 ng. Poiger &
Schlatter (1980) applied a single dose of TCDD in various vehicles on
the inner surface of the rabbit ear and followed the appearance of
inflammation, hyperkeratosis, and chloracne. The minimum dose that
induced skin lesions was around 1 µg TCDD/ear when the vehicle was

acetone, vaseline, or polyethylene glycol 1500 with 15% water. When
TCDD was mixed with soil-water (2:1), or activated carbon-water (1:8)
before application, 2-3 and 160 µg TCDD/ear, respectively, were needed
to induce lesions. Even 160 µg TCDD/ear produced very small changes on
the skin surface when applied as an activated carbon-water paste.

Hairless mice constitute another in vivo model for studies of
epidermal effects of TCDD. Puhvel et al. (1982) studied cutaneous
changes induced by topical application of 0.1 µg TCDD, three times per
week for four weeks, in two strains of hairless mice (Skh:HR-1 and
HRS/J). Epidermal hyperplasia, hyperkeratinization, loss of sebaceous
glands and follicles, and keratin build-up in the dermal cysts were
noted in both strains of mice. Follicular keratosis, considered the
pathognomonic lesion in human chloracne, did not appear within 4 weeks
of application. In the same study, follicular keratosis did develop
after topical application of 2 mg 3,4,3',4'- tetrachlorobiphenyl, five
times per week for 8 weeks, suggesting that follicular keratosis is an
extension of the epidermal response and, thus, not related to
metabolic changes in sebaceous glands. The authors considered hairless
mice a less sensitive model for the chloracnegenic response than the
rabbit ear bioassay. Similar findings were obtained when HRS/J mice
were exposed to TCDD topically applied two to three times per week for
four weeks (Knutson & Poland, 1982; Poland & Knutson, 1982). They
found, with a total applied dose of 1.2 µg TCDD, a moderate to severe
response, including hyperplasia, hyperkeratosis of the interfollicular
epidermis, squamous metaplasia of the sebaceous glands, and
hyperkeratosis within the dermal cysts, but no keratosis in the
sebaceous follicles.

Soot or a benzene extract of soot, containing PCDDs and PCDFs
from a PCB-containing transformer fire (Binghamton, New York, USA),
were applied to 64.5 cm² of the shaved, unabraded dorsal surface of
(3 + 3) and (1 + 1) male and female New Zealand white rabbits (3.5
kg), respectively (Silkworth et al., 1982). The dose applied was 500
mg soot or benzene extract corresponding to 500 mg soot/kg body
weight. Controls, one male plus one female, were exposed to activated
carbon or benzene in corresponding amounts. Exposure lasted for 24 h
and the observation period was 67 days. The soot produced no overt
toxicity, no weight loss, and no histological findings in thymus,
kidney, or skin, but hepatic centrilobular hypertrophy was found in
both sexes. The soot extract gave rise to a reversible skin
inflammation and hepatic centrilobular hypertrophy in the female only.
No weight loss was recorded and the kidney, thymus, and skin were
histologically normal at necropsy.

7.4.4.2 In vitro studies

Keratinocytes, the principal cell type of epidermis, form an in
vitro model for studies of TCDD-induced hyperkeratosis both in human
and animal-derived cell cultures. The response is analogous to
hyperkeratinization in vivo. Newly confluent epidermal cell cultures

exhibit proliferative properties, while the number of basal cells
tends to decrease with increasing time of post-confluency growth.
Thus, with the appropriate selection of culture medium and time of
treatment, different aspects of TCDD toxicity in vivo can be
modelled in vitro.

A TCDD-induced response of in vitro keratinization was first
demonstrated in XB cell cultures, an established keratinocyte cell
line derived from a mouse teratoma, plated at high density to avoid
spontaneous keratinization (Knutson & Poland, 1980b). Keratinization
was dose-related and histologically similar to that which occurs
spontaneously when XB cells are plated at low density. The epidermal
proliferation in XB cells produced by TCDD could not be biochemically
related to the response produced by cholera toxin, epidermal growth
factor, or 12-O-tetradecanoylphorbol-13-acetate, other compounds
known to affect cell proliferation in XB cells (Knutson & Poland,
1984). Late passage XB cells, i.e. XBF cells, show increased cell
density at saturation and a fusiform morphology at high density.
Additionally, they have lost their ability to respond with
keratinization upon TCDD treatment. Exposing XBF cells to TCDD
concentrations in the range 10 to 11 X 10^-8 mol/litre resulted in
normal growth until confluency was reached by day 7. Thereafter
TCDD-treated cultures showed a persistent decrease in cell growth and
cell proliferation as well as changed morphology, whereas viability
was unaffected (Gierthy & Crane, 1984). Reseeding these quiescent XBF
cells, previously exposed to 10^-9 mol/litre TCDD for 14 days,
resulted in normal growth and proliferation until confluency. These
TCDD-pretreated cells maintained their susceptibility to TCDD-induced
changes in cell growth and morphology. Both XB cells (keratinization
assay) and XBF cells (flat-cell-assay) have proved to be useful in
vitro bioassays to measure "dioxin-like" activity of both
environmental samples and of pure isomers (Gierthy et al., 1984;
Gierthy & Crane, 1985a,b). Although XBF cells, a highly transformed
variant of XB cells, seem to be less appropriate as a model for TCDD
action on normal mammalian epithelial cell proliferation and
differentiation, they seem to be more stable and easier to maintain
than XB cells. Several continuous lines of human keratinocytes derived
from neonatal foreskin (Milstone & Lavigne, 1984; Osborne et al.,
1984) or squamous cell carcinomas (SCC) (Rice & Cline, 1984; Willey et
al., 1984; Hudson et al., 1985, 1986) have been shown to respond to
TCDD in nanomolar concentrations with a variety of signs that indicate
alterations in the normal differentiation process. Stimulation of
³H-thymidine incorporation was seen in post-confluent human
epidermal cells derived from neonatal foreskin after exposure to TCDD
(Milstone & Lavigne, 1984). Newly confluent human epidermal cells,
derived from foreskin, responded to exposure for 4 days to 10 nmol
TCDD/litre with decreased DNA synthesis, a decrease in the number of
proliferating basal cells, decreased binding of epidermal growth
factor (EGF), an increase in the number of differentiated cells, and
increased envelope formation, i.e., a decrease in the proliferative

capacity and an increase in the state of differentiation (Osborne &
Greenlee, 1985). The decreases in small (basal) cell number and EGF
binding were dose dependent, with EC50 values for TCDD of 2 and 1
nmol/litre, respectively. The responses were also obtained with TCDF
but not with 2,4-diCDD.

The proliferation and differentiation of epidermal cells is
normally regulated by several growth factors and hormones, e.g., EGF,
vitamin A and hydrocortisone. Mouse hepatoma cells exposed to TCDD for
24 h showed 20% inhibition of EGF binding (Karenlampi et al., 1983).

Hudson et al. (1985, 1986) demonstrated that TCDD decreases, in
a dose-dependent manner, the specific binding as well as the cellular
uptake of EGF in cultures of human epidermal cells. The EC₅₀ dose
for inhibition was 1.8 nmol/litre. A similar inhibitory effect was
obtained by TCDF, while 2,7-diCDD was inactive even at doses 100-fold
greater. Maximal inhibition, almost 60%, of EGF binding in confluent
SCC-12F cells exposed to 100 nmol TCDD/litre was obtained after a
pretreatment period of 72 h. No effect was obtained when TCDD was
added at the same time as EGF; thus TCDD did not compete for
EGF-binding sites, neither did TCDD affect the process of
internalizing the EGF. Further studies of the SCC-12F cell line
revealed data suggesting that TCDD specifically reduces the high
affinity EGF-binding sites in the basal cell population of this cell
line (Hudson et al., 1986).

The addition of 10^-6 mol hydrocortisone/litre to the medium
antagonized the growth inhibition of SCC cells grown in 10^-10 mol
TCDD/litre (Rice & Cline, 1984). Hydrocortisone stimulated several
aspects of keratinocyte differentiation. These stimulatory effects
were abolished in the presence of 10^-8 mol TCDD/litre, although TCDD
alone had no effect on these parameters. The hydrocortisone level in
the medium was unaffected by TCDD. TCDD, and even more so
hydrocortisone, were able to stimulate stratification in SCC cultures
held at confluence for extended periods. This effect was opposed by
vitamin A (Rice & Cline, 1984).

Like TCDD, vitamin A suppressed the stimulation of keratinocyte
differentiation by hydrocortisone. However, vitamin A had no effect on
TCDD-induced growth inhibition or its reversal by hydrocortisone (Rice
et al., 1983; Rice & Cline, 1984).

Epidermal transglutaminase (ETG) activity, the marker enzyme for
terminal differentiation, was increased by treatment of basal
keratinocyte cultures from neonatal BALB mice with 10^-9 mol
TCDD/litre for 5 to 12 days, although morphologically no signs of
terminal differentiation were present. A parallel increase in ETG
activity was present when these cells were grown in medium rich in
Ca²⁺, although these cells did stratify and differentiate (Puhvel et
al., 1984).

7.4.5 Effects on the immune system

TCDD produces a pronounced atrophy of the thymus, spleen, and, to
a lesser, extent the peripheral lymph nodes of most experimental
animals. Since Buu-Hoi et al. (1972a) reported on TCDD-induced thymic
atrophy, many studies in rats, mice, guinea-pigs, and monkeys have
shown that the thymus is one of the organs most severely affected by
TCDD. Lesions in the thymus appear at exposure levels well below those
inducing lesions in other organs. Although there is species variation
in the degree and severity of other organ effects, the effects of TCDD
on lymphoid tissues is consistent in all species. Further
investigations of the effect of TCDD on the immune system, which is a
rapidly proliferating and differentiating organ system containing many
cellular components in a highly organized and regulated network, have
revealed that TCDD affects both the humoral-mediated (section 7.4.5.2)
and the cell-mediated immune response (section 7.4.5.3). Also the
complement system, a key component of the innate immunity, is affected
by TCDD treatment (White et al., 1986). Damage to the thymus and to
the cell-mediated immune system seems to be rather specific in that it
occurs at doses considerably lower than those affecting other immune
functions (Faith & Luster, 1979). Thymic involution is believed to be
a direct effect on the gland, and not secondary to factors such as
undernutrition (van Logten et al., 1980), altered levels of hormones
including corticosteroids (Vos et al., 1973; van Logten et al., 1980),
pituitary hormones (Vos et al., 1973), and thymosin (Vos et al.,
1978a,b), or zinc deficiency (Vos et al., 1978a,b), or to a direct
cytotoxic effect on lymphocytes (Vos et al., 1978a,b). A direct effect
of TCDD on mouse fetal thymus organ cells grown in vitro was
demonstrated at concentrations as low as 10^-10 mol/litre (Dencker et
al., 1985). Within the thymus, lymphocytes in the cortex, i.e., the
immature T-cells, are more severely affected in TCDD-treated animals
than are lymphocytes in the medulla, i.e., the mature T-cells. Thus,
it seems that TCDD impairs the differentiation of thymocytes into
immunocompetent T cells. Greenlee et al., (1985) obtained results
demonstrating a direct effect of TCDD on thymus epithelial (TE) cells.

High levels of Ah receptors (section 7.8.1) have been found in
the thymus (Carlstedt-Duke, 1979; Mason & Okey, 1982; Gasiewicz &
Rucci, 1984). Studies with C57Bl/6 mice (responsive to TCDD), DBA/2
mice (less responsive to TCDD), and B6D2F1 mice, hybrid mice from
crosses between these strains, suggest that TCDD-induced thymic
involution, as well as immunosuppression, segregates with the Ah locus
in these strains of mice (Poland & Glover, 1980; Clark et al., 1983;
Vecchi et al., 1983; Nagarkatti et al., 1984; Dencker et al., 1985).
The most profound and persistent effect of TCDD on the immune system
is found when TCDD is administered during pre- and/or immediate
postnatal life. Contrary to other experimental animals investigated,
rainbow trout (Salmo gairdneri) are relatively resistant to the
immunosuppressive effects of TCDD (Spitzbergen et al., 1986). No
humoral-mediated effects, and only minor cell-mediated effects, were
present at doses which caused clinical toxicity.

7.4.5.1 Histopathology

Lymphoid organs, primarily thymus but also spleen and lymph
nodes, have been found to be affected by TCDD over a wide spectrum of
dose ranges in adult rats, guinea-pigs, and mice (Gupta et al., 1973;
Vos et al., 1973; Vos and Moore, 1974; McConnell et al., 1978b). The
marked reduction in the size of thymus has been referred to as atrophy
(Gupta et al., 1973; Vos & Moore, 1974), regression (Allen et al.,
1975), or involution (Kociba et al., 1976), though these terms do not
clearly represent the pathogenesis of this lesion but are more a
description of the final event. Toxic effects on thymus appeared in
adult guinea-pigs, rats, and mice exposed to eight weekly doses of 0.2
µg TCDD/kg body weight, six weekly doses of 5 µg TCDD/kg body weight
and four weekly doses of 5 µg TCDD/kg body weight respectively (Vos et
al., 1973). The thymus from moribund animals, or from animals that
died from TCDD exposure, showed a dose-dependent decrease in the
number of cortical lymphocytes, markedly smaller thymic lobules, and
loss of demarcation between the cortex and medulla. Guinea-pigs, the
species most severely affected by TCDD, showed large cystic Hassall
bodies, filled with polymorphonuclear leukocytes (Gupta et al., 1973;
Vos et al., 1973). Guinea-pigs that received lethal doses of TCDD
showed scattered necrosis of lymphocytes in the cortical region with
concomitant phagocytosis by macrophages as early as 5 days
post-exposure (McConnell et al., 1978b). The effect was more apparent
at day 14, and by day 20 it was difficult to differentiate the cortex
from the medulla. At day 20 little evidence of necrosis remained,
though karyorrhectic debris and prominent phagocytosis indicated that
this had occurred. Since the thymus from guinea-pigs surviving the
TCDD dose for 30 days was usually normal microscopically (Vos et al.,
1973; McConnell et al., 1978b), it seems that thymic necrosis must be
an early event in the course of the toxic syndrome. Furthermore, in
animals which survive, thymic regeneration seemed to be rapid.
Decreased weight of thymus, loss of cortical cells, and cell necrosis
have also been found in hamsters exposed to TCDD (Gasiewicz et al.,
1986).

Depletion of lymphoid cells in the spleen, intestinal tract, and
various lymph nodes observed in guinea-pigs, rats, and mice (Gupta et
al., 1973; McConnell et al., 1978b) was less extensive than in the
thymus. The major effect in the spleen of rats is the loss of the
T-cell-dependent areas, namely the periarterial lymphoid sheet and the
paracortical areas (Vos & Moore, 1974).

Depressed immunoglobulin levels were reported for 1- and
4-month-old C57Bl/6 mice exposed to four and six weekly doses,
respectively, of 25 µg TCDD/kg body weight (Vos & Moore, 1974).
Feeding 10, 20, 50, or 100 µg TCDD/kg in the diet depressed
dose-dependently the Y-globulin level in 7-week-old Swiss-Webster mice
(Hinsdill et al., 1980).

Vos & Moore (1974) demonstrated a dose-related lymphocyte
depletion of thymus cortex, spleen, and intestinal lymph nodes in
maternally exposed pups of rats and mice. However, no effect on
immunoglobulin levels was observed in 25-day-old rats maternally
exposed (5 µg TCDD/kg body weight) on days 0, 7, and 14. The
developing lymphoid tissues were found to be more sensitive to TCDD
than were the lymphoid tissues of adults or young.

7.4.5.2 Humoral-mediated immunity

The humoral-mediated immunity (Tables 51 and 52) operates through
antibody-producing cells and is transferable by serum. This system
includes classical antibody-mediated protective immunity and immediate
hypersensitivity reactions. Vos et al. (1974) reported a significant
decrease in the alpha-, ß-, and gamma-globulin levels in C57BL/6 mice
given non-toxic doses of TCDD. The effects of TCDD on specific humoral
immunity responses in adult animals are summarized in Table 51.
Feeding levels of 10 µg TCDD/kg body weight or more reduced the
primary and secondary antibody response to both sheep red blood cells
(sRBC) and tetanus toxin in male Swiss-Webster mice (Hinsdill et al.,
1980). Weekly oral doses of 1 or 10 µg TCDD/kg body weight reduced
both the primary and secondary serum antitetanus titres in male New
Zealand rabbits (Sharma et al., 1984). The secondary, but not the
primary, serum tetanus antitoxin level was decreased in Hartley
guinea-pigs given eight weekly doses of 0.2 µg TCDD/kg body weight
(Vos et al., 1973). Results presented by Vecchi et al. (1980, 1983)
show that single doses as low as 1.2 µg TCDD/kg body weight to C57BL/6
mice decreased the number of plaque-forming spleen cells in response
to an injection of the thymus-dependent antigen sRBC. The response was
dose dependent and lasted for at least 42 days. Luster et al. (1985)
found a decrease in anti-sRBC plaque-forming spleen cell production in
B6C3F1 and DBA/2 mice 5 days after single oral doses of 2 and 10 µg
TCDD/kg body weight, respectively. A dose of 30 µg TCDD/kg body weight
was needed to produce a significant antibody response to the
thymus-independent antigen type III pneumococcal polysaccharide (sIII)
in C57Bl/6 mice (Vecchi et al., 1980). With sRBC and the
thymus-independent antigen trinitrophenylated Brucella abortus,
Clark et al. (1981) found a depressed number of spleen plaque-forming
cells in C57BL/6 mice only with a total dose of 40 µg/kg body weight
given as four equal weekly doses. Chastain & Pazdernik (1985)
demonstrated decreased numbers of plaque-forming spleen and bone
marrow cells in response to the thymus-independent antigen
trinitrophenylated lipopolysaccharide (LPS). Spleen and bone marrow
cells were collected from male C57BL/6 mice 7 days after a single ip
dose of 30, 60, 90, or 120 µg TCDD/kg body weight. The decrease
occurred at lower doses in the spleen cell assay but was more
pronounced at higher doses in the bone marrow cell assay. The number
of antibody-producing cells, measured asplaque-forming cells,
following immunization with either sRBC or LPS was reduced in B6C3F1
mice treated with 1 and 5 µg TCDD/kg body weight (Tucker et al.,
1986).

Table 51. Effects of TCDD on humoral-mediated immune responses in adult animals

Species/strain Sex/age/weight^f TCDD exposure Parameter measured^e
(reference) Frequency/route/dose

Mice
Swiss-Webster F/4-7 weeks/NR fed 10, 20, 50, 100, or primary and secondary sRBC^a antibody
500 µg/kg continuously level,
(Hinsdill et al., in the diet for 5 weeks primary and secondary serum tetanus
1980) antitoxin level

C57BL/6J M/6-8 weeks/NR single ip dose of anti-sRBC^a plaque-forming spleen cells
1.2, 6, or 30 µg/kg body anti-SIII^b plaque-forming spleen cells
(Vecchi et al., weight
1980)

C57BL/6J M/6-8 weeks/NR four weekly ip doses of anti-sRBC^a plaque-forming spleen cells
0.1, 1, or 10 µg/kg body anti-TNP-BA^c plaque-forming spleen cells
(Clark et al., weight
1981)

C57BL/6 M/8-10 weeks/NR single ip dose of anti-sRBC^a plaque-forming spleen cells
C3H/HeN 1.2, 6, or 30 µg/kg body
DBA/2 weight
AKR
B6D2F1

(Vecchi et al.,
1983)

C57BL/6 M/6-8 weeks/NR single ip dose of 30, 60, anti-TNP-LPS^d plaque-forming spleen cells
90, or 120 µg/kg body weight anti-TNP-LPS^d plaque-forming bone marrow
(Chastain & cells
Pazdernik, 1985)

Table 51. (contd - 2)

Species/strain Sex/age/weight^f TCDD exposure Parameter measured^e
(reference) Frequency/route/dose

Mice (contd).
DBA/2N F/6-8 weeks/ single oral doses of 5, anti-sRBC^a plaque-forming spleen cells
18-21 g 10, or 50 µg/kg body weight
(Luster et al.,
1985)

B6C3F₁ F/6-8 weeks/ single oral doses of 0.2, anti-sRBC^a plaque-forming spleen cells
18-21 g 1, 2, 5, or 10 µg/kg body
(Luster et al., weight
1985)

Guinea-pigs
Hartley F/NR/256 g eight weekly oral doses of primary serum tetanus antitoxin level
0.008, 0.04, 0.2, or secondary serum tetanus antitoxin level
(Vos et al., 1973) 1.0 µg/kg body weight

Rabbit
New Zealand M/Adult/3 kg eight weekly oral doses of primary and secondary serum tetanus
0, 0.01, 0.1, 1.0, or 10.0 antitoxin level
(Sharma et al., µg/kg body weight
1984)

^a sRBC = sheep red blood cell.
^b SIII = type III pneumoccal polysaccharide.
^c TNP-BA = trinitrophenylated Brucella abortus.
^d TNP-LPS = trinitrophenylated lipolysaccharide.
^e = all parameters measured were decreased except for primary serum tetanus antitoxin level in guinea-pigs
(Vosal., 1973).
^f M = male; F = female; NR = not reported.

Table 52. Effects of TCDD on humoral-mediated immune responses in maternally exposed animals

Species/strain Time of TCDD exposure Route/dose Parameter measured-response
(reference)

Rats
Fisher-344 N Prenatal day 18 and/or oral/5 µg/kg body weight primary and secondary BGG^a
postnatal days 0, 7, and 14 antibody level - no effect

(Faith & Moore,
1977)

Fisher/Wistar Prenatal day 18 and/or oral/5 µg/kg body weight primary and secondary BGG^a
postnatal days 0, 7, and 14 antibody level - no effect

(Faith & Luster,
1979)

Mice
Swiss-Webster four weeks before mating, in the diet/ primary and secondary sRBC^b
throughout gestation and 1, 2.5, 5, 10, or 20 µ/kg antibody level - no effect;

(Thomas & lactation primary anti sRBC^b plaque
Hinsdill, 1979) forming spleen cells - decreased

^a = BGG - bovine gamma-globuline.
^b = sRBC - sheep red blood cell.

Only minor effects on antibody responses have been reported in
rodents maternally exposed to TCDD (Table 52). Single oral doses of 5
µg TCDD to pregnant Fisher-344 N (poor immunological responder) and
Fisher-Wistar rats (good immunological responder) on gestation day 18
and/or postnatal days 0, 7, and 14 did not affect the antibody reponse
to bovine gammaglobulin (BGG) in the offspring (Faith & Moore 1977;
Faith & Luster, 1979).

Dietary exposure of female Swiss-Webster mice to 2.5 or 5 µg
TCDD/kg for 4 weeks before mating and throughout gestation and
lactation resulted in normal antibody production in the offspring but
in a decrease in anti-sRBC plaque-forming spleen cells.

In vitro exposure of B6C3F1 spleen cells to 10^-9 mol
TCDD/litre decreased the production of anti-sRBC plaqueforming cells
(Luster et al., 1984). The ED₅₀ for this effect was found to be 7
nmol/litre when TCDD was present from the first day of culture (Tucker
et al., 1986). Similar activity to that of TCDD was found with
2,3,7-triCDD and 1,2,3,7,8-pentaCDD, whereas 2,8-diCDD and octaCDD
were without effect at concentrations up to 5 x 10^-8 mol/litre
(Tucker et al., 1986).

The doses producing 50% suppression of splenic IgM response to
sRBC in C57Bl/6 mice were 7.1 and 85 µg/kg body weight, respectively,
for 1,2,3,6,7,8-hexaCDD and 1,2,3,4,6,7,8-heptaCDD given as single
oral doses 2 days prior to sRBC challenge (Kerkvliet et al., 1985).
OctaCDD had no effect even at doses of 100 and 500 µg/kg body weight.

Humoral immune responses of the rainbow trout (Salmo
gairdneri) were not significantly impaired even at doses of TCDD
that caused clinical toxicity (Spitsbergen et al., 1986).

7.4.5.3 Cell-mediated immunity

Cell-mediated immunity (CMI) operates through specifically
sensitized lymphocytes and is transferred by these cells. Processes
included in this system are classical cell-mediated protective
immunity (which protects against fungi, bacteria, and viruses),
delayed type hypersensitivity, rejection of tumors and foreign tissues
such as transplants, and graft versus host response. Many assays, both
in vivo and in vitro, have been developed to test CMI functions.
Besides a reduction in the number of immunologically competent cells
after TCDD exposure (Gupta et al., 1973; Vos et al., 1973; Zinkl et
al., 1973) TCDD has been demonstrated to induce a decreased CMI
response in adults (Table 53) and even more in maternally exposed
animals (Table 54). Delayed hypersensitivity response, correlatingwith
decreased host resistance to infectious agents in man, was depressed
in rodents exposed to low levels of TCDD (Vos et al., 1973; Faith &
Moore, 1977; Sharma et al., 1978; Faith & Luster, 1979; Thomas &

Hinsdill, 1979; Hinsdill et al., 1980; Clark et al., 1981).
TCDD-exposure also adversely affects host susceptibility to bacteria,
viruses, tumor cells and endotoxins (Thigpen et al., 1975; Thomas &
Hinsdill, 1979; Hinsdill et al., 1980; Luster et al., 1980; Clark et
al., 1983).

Depressed graft versus host response was reported in 2-month-old
C57BL/6 mice exposed to 4 weekly oral doses of 5 µg TCDD/kg body
weight (Vos et al., 1973), whereas no effect was seen in a subsequent
study on 1- and 4-months old C57BL/6Sch mice (Vos & Moore, 1974). In
the same study, Fisher-344 rats maternally exposed to TCDD showed
decreased graft versus host response and prolonged allograft-rejection
time. The latter effect was also demonstrated in maternally
TCDD-exposed C57BL/6Sch mice (Vos et al., 1973). Proliferative
responses of spleen and/or thymus lymphocytes, stimulated by mitogens
specific for the generation of B-lymphocytes and/or T-lymphocytes from
TCDD-exposed animals, were depressed both in adults (Vos & Moore,
1974; Sharma et al., 1978) and maternally exposed rodents (Vos &
Moore, 1974; Faith & Moore, 1977; Vos et al., 1978; Faith & Luster,
1979; Luster et al., 1980). However, Thomas & Hinsdill (1979) found no
effect on the lymphoproliferative response in offspring from
Swiss-Webster mice fed with up to 5 µg TCDD/kg diet 4 weeks before
mating and throughout gestation and lactation. Depressed
lymphoproliferative response is regarded as an extremly sensitive
indicator of immunotoxicity, rather than as a predictor of immune
dysfunction. Cytotoxic T-cell generation in response to allogeneic
antigens has been demonstrated in male DBA/2, C57Bl/6, and B6D2F1 mice
given four weekly ip injections of 1 ng/kg body weight (Clark et al.,
1981, 1983). At this dose no effects were seen on delayed
hypersensitivity, antibody response, thymus cellularity, or enzyme
induction. The adverse effect of TCDD on CMI function seems to be an
age-related phenomenon in rodents. In order to obtain a complete and
persistent immune suppression, TCDD exposure must occur during
ontogenesis of the immune system. In the initial experiments on the
developing immune system, Vos & Moore (1974) exposed Fisher-344 rats
to 1 mg TCDD/kg body weight on gestation days 11 and 18 and on
postnatal days 4, 11, and 18, or to 5 µg TCDD/kg body weight on
postnatal days 0, 7, and 14. CMI functions adversely affected included
in vitro immune competence of spleen and thymus lymphoid cells,
delayed hypersensitivity reaction, prolonged allograft-rejection times
and reduced graft versus host activity. The immune suppression
demonstrated persisted throughout the study, i.e., for 145 days. The
depression of T-cell-dependent immune functions appeared to occur
without helper-cell function being affected (Faith & Moore 1977).

Attempts to study direct effects of TCDD on lymphocytes in
vitro were previously hampered by the low solubility of TCDD in
physiological buffers (Matsumura & Benezet, 1973). Vos & Moore
(1974)obtained no lymphoproliferative response in unstimulated or PHA-
or concanavalin-A-stimulated rat thymus organ cells and mouse spleen
cells when cultured in the presence of up to 20 ng TCDD/ml. Dencker et

al. (1985) demonstrated that mouse fetal thymus cells cultured in
vitro in the presence of TCDD gave a similar response similarly to
that occurring in vivo, i.e., with a dose-dependent inhibition of
the time-dependent increase in the number of lymphoid cells
(EC₅₀=10^-10 mol TCDD/litre). It could not be determined with
certainty whether the decreased cell number caused by TCDD was due to
reduced cell proliferation or to increased cell death. The
TCDD-induced suppression of mitogen-stimulated lymphoproliferation has
recently been demonstrated to be mediated by thymus epithelial cells
(Greenlee et al., 1985).

7.4.5.4 Macrophage function

The primary pathway of endotoxin detoxification is thought to be
macrophage-dependent. Thus the increased sensitivity to endotoxin
following TCDD treatment (Vos et al., 1978; Thomas & Hinsdill, 1979)
was suggestive of macrophage dysfunction. However, the number of
peritoneal macrophages, as well as their capacity to mediate cytolytic
and cytostatic effects, was not adversely affected by single ip doses
of 1.2, 6 or 30 µg TCDD/kg body weight to male C57Bl/6J mice
(Mantovani et al., 1980), nor was the ability of macrophages to reduce
nitroblue tetrazolium affected by four to five weekly oral doses of 50
µg TCDD/kg body weight in Swiss-Webster mice (Vos et al., 1978a).
Macrophage function does not appear to be altered by TCDD.

7.4.6 Myelotoxicity

TCDD treatment inhibits the bone marrow haematopoiesis in mice,
both in vivo and in vitro, by directly altering colony growth of
stem cells (Luster et al., 1980, 1985; Chastain & Pazdernik, 1985).
The bone marrow granulocyte-macrophage progenitor cell (CFU-GM)
production was reduced in B6C3F1 mice, receiving 1 µg TCDD/kg body
weight, but was unaffected in DBA/2 mice, even at a dose of 50 µg
TCDD/kg body weight (Luster et al., 1985). Chastain & Pazdernik (1985)
used the B-lymphocyte colony-forming unit assay to demonstrate
reductions in spleen and bone marrow B-cell function of C57BL/6 mice
exposed to single ip doses of TCDD in the range 30 to 120 µg TCDD/kg
body weight. Bone marrow B-cells tended to be more suseptible to TCDD
than spleen B-cells. A maternal oral dose of 5 µg TCDD/kg body weight
on the 14th day of gestation and postnatal days 1, 7, and 14 resulted
in a weak normocytic anaemia indicative of depressed erythrogenesis,
decreased bone marrow cellularity, and decreased stem cell
proliferation in B6C3F1 offspring, assessed 7 days after weaning
(Luster et al., 1980). In vitro exposure of B6C3F1 bone marrow cells
to 10^-9 mol TCDD/litre resulted in decreased CFU-GM development and
decreased number of erythrocyte colony-forming units. The decrease in
CFU-GM occurred 1 day post-treatment and remained below control values
until 10 days post-treatment (Luster et al., 1985).

Table 53. Effects of TCDD on cell-mediated immunity responses in adult animals

Species/strain Sex/age/weight^d TCDD exposure Parameter measured-response^d
(reference) Frequency/route/dose

Rats
CD F/NR/185 g six weekly oral doses of delayed hypersensitivity to tuberculin - NE
0.2, 1.0, or 5.0 µg/kg
(Vos et al., 1973) body weight

Mice
C57BL/6 M/6-8 weeks/NR four weekly ip doses of delayed hypersensitivity to sRBC^a - Dec;
0.1, 1.0, or 10.0 µg/kg delayed hypersensitivity to oxazalone - Dec;
(Clark et al., body weight generation of alloantigen-specific cytotoxic
1981) T-cells - Dec

C57BL/6 M/NR/NR four weekly ip doses of resistance to Herpes virus challenge - Dec;
0.001, 0.01, 1.0, or generation of alloantigen-specific cytotoxic
(Clark et al., 10.0 µg/kg body weight T-cells - Dec
1983)

DBA/2 M/NR/NR four weekly ip doses of generation of alloantigen-specific cytotoxic
0.001, 0.01, 0.1, 1.0, or T-cells - Dec
(Clark et al., 10.0 µg/kg body weight
1983)

Swiss-Webster F/4-7 weeks/NR five weeks feeding of diets resistance to Salmonella typhimurium challenge - Dec;
containing 10, 50, or resistance to Listeria monocytogenes challenge - Dec;
(Hinsdill et al., 100 µg/kg body weight contact sensitivity to 2,4-dinitro-1-fluoro-
1980) benzene - Dec

Table 53 (contd - 2)

Species/strain Sex/age/weight^d TCDD exposure Parameter measured-response^d
(reference) Frequency/route/dose

Mice (continued)
C57BL/6J M/6-8 weeks/NR single ip doses of 1.2, number of peritoneal macrophages - Dec;
6 or 30 µg/kg body weight number of splenic natural killer cells - Dec;
(Mantovani macrophage mediated cytolysis - NE;
et al., 1980) macrophage mediated cytostasis - NE

C57BL/6 M/NR/NR four weekly ip doses of generation of allospecific cytotoxic T-cells - Dec
DBA 0.001 µg/kg body weight - NE
B6D2F₁ - Dec

(Nagarkatti
et al., 1984)

CD-1 M/NR/28 two, four, or eight weekly lymphoproliferative response of PHA^b- and PWM^c-
doses of 0.01, 0.1, 1.0, or stimulated splenic cells - Dec
(Sharma & 10.0 µg/kg body weight
Gehring, 1979)

C57BL/6Jfh M/4 weeks/NR four weekly oral doses of resistance to Salmonella bern challenge - Dec;
0.5, 1.0, 5.0, 10.0, or resistance to Herpes virus challenge - NE
(Thigpen et al., 20.0 µg/kg body weight
1975)

C57BL/6 M/2 months/ four weekly oral doses of graft versus host activity - Dec
24.4 g 0.2, 1.0, 5.0, or 25.0
(Vos et al., µg/kg body weight
1973)

C57BL/6Sch M/1 month/NR four weekly oral doses of lymphoproliferative response of PHA^b-stimulated
1.0, 5.0, or 25.0 µg/kg spleen cells - Dec;
(Vos & Moore, body weight graft versus host activity - NE
1974)

Table 53 (contd - 3)

Species/strain Sex/age/weightd TCDD exposure Parameter measured-responsed
(reference) Frequency/route/dose

Mice (continued)
C57BL/6Sch M/4 months/NR six weekly oral doses of lymphoproliferative response PHA-stimulated
1.0, 5.0, or 25.0 µg/kg spleen cells - NE
(Vos & Moore, body weight graft versus host activity - NE
1974

Swiss M/3-4 weeks/NR four or five weekly oral resistance of Listeria monocytogenes - NE;
doses of 50 µg/kg body number of peritoneal macrophages - NE;
(Vos et al., weight macrophage reduction of nitroblue tetrazodium - NE
1978a)

Guinea-pigs
Hartley F/NR/256 g eight weekly oral doses delayed hypersensitivity to tuberculin - Dec
of 0.008, 0.04, 0.2, or
(Vos et al, 1973) 1.0 µg/kg body weight

Rabbits
New Zealand M/adult/3 kg eight weekly oral doses delayed hypersensitivity to tuberculin - Dec
of 0.01, 0.1, 1.0, or
(Sharma et al., 10.0 µg/kg body weight
1984)

^a sRBC = sheep red blood cells.
^b PHA = phytohaemagglutinin.
^c PWM = poke weed.
^d NR = not reported; NE = no effect; Dec = decreased; M = male; F = female.

Table 54. Effects of TCDD on cell-mediated immunity responses in maternally exposed animals

Species/strain TCDD exposure Age when Parameter measured - response^f
(reference) Frequency/route/dose tested^g

Rats
Fisher/Wistar 5 µg/kg body weight on gestation day 18 25 days lymphoproliferative response of
and on postnatal days, 0, 7, and 14, or PHA^a- and ConA^b-stimulated spleen
(Faith & Luster, 5 µg/kg body weight on postnatal days and thymus cells - Dec;
1979) 0, 7, and 14 delayed hypersensitivity to
tuberculin - Dec

Fisher-344 5 µg/kg body weight on gestation day 18 25 days lymphoproliferative response of
and on postnatal days 0, 7, and 14, or PHA^a- and Con^b-stimulated spleen
(Faith & Moore, 5 µg/kg body weight postnatal days 0, 7, and thymus cells - Dec;
1977) and 14 delayed hypersensitivity to
oxazolone - Dec

Fisher 344 1 µg/kg body weight on gestation days 11 25 days lymphoproliferative response of
and 18 and on postnatal days 4, 11 and 18 or PHA^a-stimulated spleen cells and of
(Vos & Moore, 5 µg/kg body weight on postnatal days 0, PHA^a- and ConA^b-stimulated thymus-
1974) 7 and 14 cells - Dec

Mice
B6C3F₁ ^c 1, 5, or 15 µg/kg body weight on NR lymphoproliferative response of
gestation day 14 and on postnatal days 1, mitogen-stimulated spleen cells:
(Luster et al., 7, and 14 PHA^a - Dec;
1980) ConA^b - Dec;
LPS^d - NE;
Macrophage proliferation - NE;
Phagocytizing ability - NE;
Resistance to Listeria monocyto-
genes - Dec;
Resistance to PYB6 tumor cells - Dec

Table 54 (contd).

Species/strain TCDD exposure Age when Parameter measured - response^f
(reference) Frequency/route/dose tested^g

Mice (continued)
Swiss-Webster Feeding 1, 2.5, or 5 µg TCDD/kg 5-6 weeks Contact sensitivity to 2,4-dinitro-
4 weeks before mating, throughout 1-fluorobenzene - Dec;
(Thomas & gestation, and 3 weeks postnatally lymphoproliferative response of
Hinsdill, 1979) PHA^a- and ConA^b-stimulated spleen
and thymus cells - NE;
Resistance to Salmonella typhimurium
endotoxin - Dec; Resistance
to Listeria monocytogenes - NE

C57BL/6Sch 2 or 5 µg/kg body weight on gestation days 23 days Skin graft assay - prolonged skin
14 and 17 and on postnatal days 1, 8, and 15 graft rejection time
(Vos et al., 1974)

Swiss 10 µg/kg body weight on postnatal days 22 days lymphoproliferative response of
1, 4, 8, 11, 15, and 18 PHA^a-, ConA^b- &
(Vos et al., PWM^e-stimulated thymus cells - Dec
1978a)

^a PHA = phytohaemagglutinin.
^b ConA = concanavalin A.
^c B6C3F₁ = progeny to female C57BL/6N and male C3H mice.
^d LPS = lipopolysaccharide.
^e PWM = poke weed.
^f Dec = decreased, NE = no effect.
^g NR = not recorded.

7.4.7 Effects on the intermediary metabolism

Changes in intermediary metabolism have been demonstrated in
TCDD-treated experimental animals. The circulating concentration of
glucose in TCDD-treated rats was decreased relative to ad
libitum-fed (Zinkl et al., 1973; Schiller et al., 1985) and pair-fed
(Gasiewicz et al., 1980; Potter et al., 1983) control rats.
TCDD-treated and pair-fed controls (both groups were schedule fed) had
similar concentrations of serum glucose (Christian et al., 1986b). The
TCDD-induced hypoglycaemia was not caused by altered pancreatic
function, as judged by insulin and glucagon levels (Potter et al.,
1983).

Reduced hepatic glycogen content, compared to the value in
control rats, was reported in Sprague Dawley rats 16 days after a
single ip dose of 20 µg TCDD/kg body weight (Weber et al., 1983).
However, in studies by Christian et al. (1986b), compared to pair-and
schedule-fed control rats, TCDD-treated (75 µg/kg body weight daily)
Sprague Dawley rats had significantly increased hepatic glycogen
levels but unaffected cardiac and muscle (gastrocnemius) levels of
glycogen 2-8 days post-treatment.

TCDD-treated rats maintained a normal overall nitrogen balance,
as judged by urinary urea, creatinine, and ammonia levels, but
exhibited changes in certain plasma protein levels (Christian et al.,
1986b).

Elevated circulating cholesterol levels were found in
TCDD-treated rats (Albro et al., 1978; Poli et al., 1980; Schiller et
al., 1985), whereas circulating free fatty acids and triacylglycerols
were decreased in TCDD-treated rats, when compared to pair-and
schedule-fed control rats (Christian et al., 1986b).

A marked accumulation of hepatic lipid has been found in rats
after single doses of TCDD (Cunningham & Williams, 1972; Gupta et al.,
1973; Albro et al., 1978; Schiller et al., 1985). At a sublethal dose
of TCDD, hepatic triglyceride and free fatty acid levels were elevated
already one day after dosing. Abnormal lipid deposition patterns
persisted for at least 2 months (Albro et al., 1978). The increased
level of triglycerides in the liver (Schiller et al., 1985; Christian
et al., 1986b) of TCDD-treated rats was not accompanied by increases
in cardiac or muscle (gastrocnemius) triacylglycerol levels (Christian
et al., 1986b). Hepatic lipid synthesis in Wistar rats, measured as
the 1-h incorporation of ³H-acetate, was not affected by TCDD
treatment when studied 7 days after exposure to 10 µg TCDD/kg body
weight (Cunningham & Williams, 1972).

Mice (both C57BL/6 and DBA/2 strains) responded to TCDD treatment
with dose-dependent decreases in serum levels of glucose, cholesterol,
and triglycerides, and increases in hepatic triglycerides, whereas
serum glycerol and free fatty acid levels were unaffected (Chapman &
Schiller, 1985).

Hartley guinea-pigs given a lethal ip dose of 2 µg TCDD/kg body
weight had significantly increased circulating levels of cholesterol
esters, triglycerides, and phospholipids but a normal free fatty acid
level 7 days post-treatment. The increase in serum lipids was
accompanied by a pronounced increase in low density lipoproteins,
particularly the very low density lipoprotein fraction (Swift et al.,
1981). The plasma cholesterol and triglyceride levels were elevated
also when TCDD-treated guinea-pigs were compared to pair-fed controls
(Gasiewicz & Neal, 1979).

A TCDD-induced increase in plasma triglyceride levels was also
noted in New Zealand rabbits, fed 20 µg TCDD/kg body weight, whereas
plasma cholesterol was unaffected 12 weeks after dosing (Lovati et
al., 1984). TCDD treatment did not alter the liver lipid levels (free
and esterified cholesterol, triglycerides, and phospholipids), but the
triglyceride level was significantly increased. These results were
valid both for rabbits fed normal chow and those fed a
cholesterol-rich (0.5%) diet. Golden Syrian hamsters exposed to 1000
µg TCDD/kg body weight, orally or ip, had elevated plasma cholesterol
levels until 20 days after exposure but normal levels on day 50,
whereas serum triglyceride levels were normal until day 20 and then
became significantly lower than those of controls (Olson et al.,
1980b).

7.4.8 Enzyme induction

Primarily, TCDD has been found to increase enzyme activities
although observations on enzyme inhibition have also been made. Since
the first reports of enzyme systems as targets for TCDD (Buu-Hoi et
al., 1971a, 1972b; Greig, 1972; Poland & Glover, 1973b,c), enzyme
induction has become the most extensively studied biochemical response
produced by TCDD. The mixed function oxidase system (MFO), capable of
metabolizing both endogenous and foreign lipophilic compounds to more
polar products, has been the most thoroughly investigated one, and
arylhydrocarbon hydroxylase (AHH) and 7-ethoxy-resorufin 0-deethylase
(EROD) are the most frequently assayed enzyme activities in this
system. TCDD has also been reported to affect UDP-glucuronosyl
transferases (UDPGT) (Thunberg et al., 1980, 1984) and
glutathione-S-transferases (GT) (Manis & Apap, 1979), which are
multifunctional enzyme systems involved in conjugating a wide variety
of compounds.

Most studies have been performed with microsomal enzymes, but
TCDD has also been found to have effects on enzymes in the cytosolic
fraction. It seems that TCDD produces organ-specific effects, and
although, quantitatively, hepatic enzyme induction is of more concern
than extrahepatic enzyme effects, the latter may qualitatively be as
important. Studies in different species have revealed that enzyme
induction due to TCDD exposure also is a species-specific phenomenon.

Time course studies have shown that maximal increases in enzyme
activities are reached within 3 to 4 days post-treatment. After a lag
period of about 2 to 3 weeks, enzyme activities begin to return to
normal levels (Hook et al., 1975a,b; Lee & Suzuki, 1980; Lucier et
al., 1973; Poland & Glover, 1973a).

According to Kitchin & Woods (1979), TCDD-induced AHH activity
did not reach the normal level until 6 months after rats were exposed
to a single daily dose of 2 µg/kg body weight.

Hook et al. (1975a) found no apparent dependence on age when
studying AHH induction in CD rats which were 10 to 335 days old at the
time of exposure to 25 µg TCDD/kg body weight.

Several investigators have studied the relative potency of various
PCDDs and PCDFs to induce AHH and/or EROD activities (Bradlaw et al.,
1980; Poland et al., 1976; Bandiera et al., 1984a,b; Sawyer & Safe,
1985; Mason et al., 1986). They found an apparent structure-activity
relationship between the location of the halogen atoms on the
dibenzo-p-dioxin molecule and the ability to induce AHH activity
both in vivo and in vitro. Isomers with halogens at the four
lateral ring positions produced a greater biological response than
those with halogens at three lateral ring positions, while two
lateral halogen atoms seemed to be insufficient to produce a
biological response. TCDD was the most potent enzyme inducer of the
compounds tested.

On a molecular basis TCDD is the most potent MFO-inducing
compound known and MFO induction seems to be the most sensitive
biochemical response produced by this chemical. According to Kitchin
& Woods (1979), induction in the rat takes place after a single daily
dose of only 0.002 µg TCDD/kg body weight. In the guinea-pig (the
animal most sensitive to TCDD toxicity), MFO induction has been
observed, but the induced activities were low even at lethal doses
(Hook et al., 1975a). Neither is there a correlation in cell cultures
between induction of MFO and toxicity. Furthermore, it is known that
metabolites of TCDD are less toxic and more readily excreted than the
parent compound (see section 8.1.5). Thus, TCDD-induced MFO activities
represent a detoxification process rather than one leading to toxic
effects.

However, induction of MFO activities might potentiate the
toxicity of other foreign compounds requiring metabolic transformation
by the MFO system before they can exert their toxic effect. A number
of studies have shown that induction of MFO activities alters the

metabolism of the model xenobiotic, benzo(a)pyrene by increasing the
rate of microsomal metabolism, changing the metabolic profile to more
toxic metabolites, and increasing the extent of covalent binding to
liver microsomes (Berry et al., 1976, 1977; Uotila et al., 1978).
TCDD, applied topically or subcutaneously increased the
carcinogenicity of 3-methyl cholanthrene (MC) in DBA/2 mice (Kuori,
1978), but decreased the carcinogenicity of
7,12-dimethylbenz(a)anthracene (DMBA) in CD-1 mice (DiGiovanni et al.,
1979a). These authors suggested that TCDD induces the MFO system and
thus increases activation of MC to the ultimate carcinogen, as well as
inactivation of DMBA, which would explain these effects.

Furthermore, increased MFO activities might adversely affect
important metabolic pathways of endogenous compounds. The effects of
TCDD on enzyme activities, both MFO and others, involved in such
biological pathways as keratinization, steroid metabolism, lipid
metabolism, plasma membrane function and porphyrin metabolism, are
discussed under separate sections. The minute quantities of TCDD
required for maximal enzyme induction or suppression, the long
duration of the effect, and the stereospecific requirements suggest a
specific interaction of TCDD with a cellular species, possibly at the
gene level. Accordingly considerable research has been directed toward
the study of the genetic regulation of AHH induction by TCDD. A
hepatic cytosolic species that bound TCDD has been suggested as the
receptor for the hepatic AHH activity. Numerous studies of this
cytosolic receptor in several species and tissues have been performed
and it seems that there is a structural gene, the Ah locus, for this
receptor, which is responsible for the expression of various enzyme
activities (see sections 7.8 and 7.8.1).

7.4.8.1 Studies on rats

The effect of TCDD on enzyme activities has been most extensively
investigated in the rat. In the liver, TCDD has been shown to increase
both the content of cytochrome P-450 (Lucier et al., 1973, 1986;
Poland & Glover, 1974a,b; Hook et al., 1975a; Aitio & Parkki, 1978;
Kitchin & Woods, 1979; Madhukar & Matsumura, 1981; Goldstein & Linko,
1984) and cytochrome b5 (Lucier et al., 1973; Hook et al., 1975a), as
well as the microsomal enzyme activities involved in the oxidative
transformation and conjugation of xenobiotics, e.g., aniline
hydroxylase, arylhydrocarbon hydroxylase (AHH), biphenyl hydroxylase,
7-ethoxycoumarin-0-deethylase (ECOD), EROD, and UDPGT. These enzyme
activities have been investigated in a vast number of studies, some of
them quoted in Table 55. Goldstein & Linko (1984) demonstrated that
TCDD induced two isozymes of cytochrome P-450 (P-448) in the liver but
only one of these in extrahepatic tissues of young Sprague Dawley rats
2 days after a single oral dose of 25 µg/kg body weight.

Table 55. Studies demonstrating in vivo induction of mixed function
oxidases and UDP-glucuronosyltransferases in TCDD-exposed strains of rats

Rat
Enzyme activity strain Reference

Aniline hydroxylase SD Beatty et al. (1978)

CD Lucier et al. (1973); Hook et al. (1975a)

Aryl hydrocarbon SD Poland & Glover (1973a, 1974a); Beatty et
hydroxylase al. (1978); Manis & Apap (1979);
Haaparanta et al. (1983); Thunberg et al.
(1984); Lucier et al. (1986); Ahlborg et
al. (1987)

CD Lucier et al. (1973); Hook et al.
(1975a); Kitchin & Woods (1979)

Wistar Nagayama et al. (1983); Keys et al.
(1985); Bannister et al. (1986); Farrell
& Safe (1986); Mason et al. (1986);
Tsyrlov et al. (1986)

Biphenyl hydroxylase CD Hook et al. (1975a,b,); Kitchin & Woods
(1979)

7-Ethoxycoumarin-O- Wistar Aitio & Parkki (1978)
deethylase

7-Ethoxyresurofin-O- SD Haaparanta et al. (1983)
deethylase

CD Kitchin & Woods (1979)

Wistar Keys et al. (1985); Bannister et al.
(1986); Farrell & Safe (1986); Mason et
al. (1986)

UDP-glucuronosyl-
transferase:
p-nitrophenol SD Thunberg et al. (1980, 1984); Ahlborg et
al. (1987)

CD Lucier et al. (1973, 1975a, 1986); Hook
et al. (1975a)

Table 55 (Cont.) Studies demonstrating in vivo induction of mixed
function oxidases and UDP-glucuronosyltransferases in TCDD-exposed
strains of rats

Rat
Enzyme activity strain Reference

UDP-glucuronosyl-
transferase:
p-nitrophenol Wistar Aitio et al. (1979); Thunberg & Híkansson
(cont'd.) (1983)

o-aminophenol Wistar Aitio et al. (1979)

4-methylumbelli-
ferone Wistar Aitio & Parkki (1978)

Microsomal glutathion-s-transferase (GT) did not respond to
TCDD (Aitio & Parkki, 1978; Mukitari et al., 1981; Baars et al.,
1982), but cytosolic GT was induced both by a single dose of 17 µg
TCDD/kg body weight 2 days post-treatment (Manis & Apap, 1979), and by
near lethal or lethal doses 1 and 6 days after dosing (Mukitari et
al., 1981; Baars et al., 1982; Hassan et al., 1983). Glutathione
reductase was also increased, while glutathione peroxidase, both total
and Se-dependent, and the content of reduced glutathione were reduced
by TCDD treatment (Hassan et al., 1983, 1985a,b,c).

The following hepatic enzyme activities involved in drug
metabolism have been reported to be unaffected by TCDD treatment in
the rat: N-and O-demethylation (Lucier et al., 1973; Poland &
Glover, 1973a; Hook et al., 1975a; Beatty et al., 1978; Kitchin &
Woods, 1979; Madhukar & Matsumura, 1981), epoxide hydratase (EH)
(Aitio & Parkki, 1978), ß-glucuronidase (Lucier et al., 1973, 1975)
and NADPH cytochrome c reductase (Poland & Glover, 1974; Aitio &
Parkki, 1978; Kitchin & Woods, 1979; Madhukar & Matsumura, 1981). The
glucuronide conjugation of bilirubin (Aitio et al., 1979), estrone,
and testosterone (Lucier et al., 1975a) by liver microsomes from
TCDD-treated rats was not different when compared to control rats.

Some hepatic enzyme activities not belonging to the MFO system,
which are affected by TCDD treatment include aldehyde dehydrogenase
(Deitrich et al., 1978; Lindahl et al., 1978), delta-aminolevulinic
acid synthetase (see section 7.4.3), DT-diaphorase (Beatty & Neal,
1977), transglutaminase (see section 7.4.4), ornithine decarboxylase
(Nebert et al., 1980; Potter et al., 1982; Farrell & Safe, 1986),
plasma membrane ATPases (see section 7.4.2.2), porphyrinogen

carboxylase (see section 8.4.3), prostaglandin synthetase (see section
7.4.9), enzymes involved in testosterone metabolism (see section
7.4.9), and RNA polymerase (Kurl et al., 1982).

Prenatal and postnatal exposure via milk to TCDD, at doses of 3
µg/kg body weight to pregnant rats on days 5, 10 and 16 of gestation,
induced hepatic AHH and UDPGT activities in the offspring. The effect
was seen 8 days post-partum and persisted for at least 2 weeks. The
inductive effect was due both to exposure to TCDD via milk and to the
activation of an inducing mechanism after birth. Fetal liver AHH was
slightly increased during late gestation, although the UDPGT activity
and the cytochrome P-450 content were not (Lucier et al., 1975b).

Administration of 2.5 µg TCDD/kg body weight to pregnant rats on
day 17 of gestation increased the AHH- and N-hydroxylation activities
and cytochrome P-450 content in the fetal liver on day 20 of gestation
(Berry et al., 1976).

AHH induction due to TCDD has been reported to occur also in the
brain (Hook et al., 1975a), kidney (Poland & Glover, 1973a; Hook et
al., 1975a; Aitio & Parkki, 1978; Potter et al., 1982; Nagayama et
al., 1983), lung (Poland & Glover, 1973a; Hook et al., 1975a; Aitio &
Parkki, 1978; Nagayama et al., 1983), prostate (Lee & Suzuki, 1980;
Haaparanta et al., 1983; Nagayama et al., 1983), thymus (Nagayama et
al., 1983), and intestine (Poland & Glover, 1973a; Hook et al.,
1975a), but intestinal AHH activity was found to be unaffected by 17
and 20 µg TCDD/kg body weight (Aitio & Parkki, 1978; Manis & Apap,
1979). Testicular (Poland & Glover, 1973a; Hook et al., 1975a; Aitio
& Parkki, 1978) and adrenal (Guenthner et al., 1979) AHH activities
were not induced by sublethal doses of TCDD. The O-deethylation
activity in kidney, lung, and prostate was increased, but no effect
was seen on the activity in testes or intestine (Aitio & Parkki, 1978;
Haaparanta et al., 1983). UDPGT activities in kidney, lung, intestine,
and brain were increased, while no effect was seen on testicular UDPGT
(Hook et al., 1975a). Similar results were reported by Aitio & Parkki
(1978), though in their study intestinal UDPGT was not affected.

Renal biphenyl hydroxylation activity has been found to increase
after TCDD treatment, but no effect on this enzyme activity was seen
in lung, intestine, brain, or testes (Hook et al., 1975a). Elevated
levels of cytochrome P-450 were found in prostate (Lee & Suzuki, 1980)
and mammary gland (Rikans et al., 1979), but not in adrenals
(Guenthner et al., 1979). Less testicular cytochrome P-450 was found
after a single dose of 25 µg TCDD/kg body weight (Tofilon & Piper,
1982). The GSH tranferase activity was increased in the lung but not
in kidney, intestine, testes (Aitio & Parkki, 1978), or prostate (Lee
& Suzuki, 1980).

Neither EH nor NADPH cytochrome c reductase were inducible by
TCDD in kidney, lung, intestine, testes (Aitio & Parkki, 1978),
mammary gland (Rikans et al, 1979), or prostate (Lee & Suzuki, 1980).
The ED₅₀ values for hepatic AHH and EROD induction were determined
in immature male Wistar rats 13 days after a single ip dose of
2,3,7-triCDD, TCDD, 1,3,7,8-tetraCDD, 1,2,3,7,8-pentaCDD,
1,2,4,7,8-pentaCDD or 1,2,3,4,7,8-hexaCDD (Mason et al., 1986). The
order of enzyme-inducing capacity was TCDD > 1,2,3,7,8-pentaCDD >
1,2,3,4,7,8-hexaCDD > 1,2,4,7,8-pentaCDD > 2,3,7-triCDD >
1,3,7,8-tetraCDD (Table 56).

7.4.8.2 Studies on mice

Enzyme induction studies in mice have been performed mainly with
two strains genetically separated at the Ah locus, thus making them
responsive (C57Bl/6 (B6)) or non-responsive (DBA/2 (D2)) to induction
of hepatic cytochrome P-450-related enzyme activities by aromatic
hydrocarbons, e.g., 3-methyl-cholanthrene (3-MC). However, the
extraordinary potency of TCDD for enzyme induction revealed increased
hepatic cytochrome P-450 content as well as AHH and O-deethylase
activities both in B6 and D2 mice after sublethal exposure to TCDD
(Poland & Glover, 1974a,b,c; Jones & Sweeney, 1977; Greenlee & Poland,
1978). Studies of MFO induction in five responsive and five
non-responsive strains of mice by Poland & Glover (1974a,b,c) revealed
that there were no consistent differences between the strains when
considering the extent to which TCDD induced AHH, O-deethylase,
N-demethylase and O-demethylase activities in the liver, kidney,
lung, skin, or bowel. The ED₅₀ for hepatic AHH induction was
determined to be 10^-9 mol/kg body weight in the responsive strain
and > 10-8 mol/kg body weight in the non-responsive strain (Poland &
Glover, 1975). Fully induced hepatic AHH activity was obtained both in
responsive (C57Bl/6 and AKR/Qdj) and non-responsive (DBA/2 and DDD)
strains of mice 3 days after an ip dose of 30 µg TCDD/kg body weight
(Nagayama et al., 1985a). Both AHH and EROD were induced in C57Bl/6
mice 7 days after an ip dose of 0.32 µg/kg body weight (Bannister et
al., 1986). Both hepatic AHH and ornithine decarboxylase activities
were similarly induced in C57Bl/6 and DBA/2 mice after a single ip
dose of 100 µg TCDD/kg body weight, but at 2 µg TCDD/kg body weight
these enzymes were induced only in the C57Bl/6 strain (Nebert et al.,
1980). Two daily doses of 0.1 µg TCDD, topically applied, increased
the epidermal AHH activity in two strains of hairless mice (Puhvel et
al., 1982). Contrary to observations in rats, TCDD induces testicular
AHH activity both in B6 and D2 mice 40h after an ip dose of 50 µg/kg
body weight (Mattison & Thorgeirsson, 1978).

Table 56. Structure activity relationships for some PCDDs

PCDD In vitro EC₅₀ values (M)^a In vivo ED₅₀ values (mmol/kg)^bLD₅₀ ^c
Congener

Receptor AHH EROD AHH EROD Body Thymic Guinea-pig
binding weight atrophy µg/kg
body
weight

1- > 1.0x10^-4 > 1.0x10^-4 > 1.0x10^-4
2,8- 3.2x10^-6 > 1.0x10^-4 > 1.0x10^-4 > 300 000
1,2,4- 1.3x10^-5 4.8x10^-5 2.2x10^-6
2,3,6- 2.2x10^-7
2,3,7- 7.1x10^-8 3.6x10^-7 1.4x10^-7 19.6 19.6 98.1 29 400
1,2,3,4- 1.3x10^-6 3.7x10^-6 2.4x10^-6
1,2,3,8- 6.1x10^-7
1,3,7,8- 7.9x10^-7 5.9x10^-7 3.2x10^-7 31.2 77.6 132 100
2,3,6,7- 1.6x10^-7 6.1x10^-8 1.1x10^-8
2,3,7,8- 1.0x10^-8 7.2x10^-11 1.9x10^-10 0.004 0.003 0.05 0.09 2
1,2,3,4,7- 6.4x10^-6 6.6x10^-7 8.2x10^-7
1,2,3,7,8- 7.9x10^-8 1.1x10^-8 1.7x10^-8