INTERNATIONAL PROGRAMME ON CHEMICAL SAFETY
ENVIRONMENTAL HEALTH CRITERIA 57
PRINCIPLES OF TOXICOKINETIC STUDIES
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, 1986
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.
ISBN 92 4 154257 8
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 1986
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
PRINCIPLES OF TOXICOKINETIC STUDIES
1. INTRODUCTION
2. ANALYTICAL METHODS
2.1. General considerations
2.2. Techniques
2.2.1. Methods of isolation
2.2.2. Methods of purification, identification,
and quantification
2.2.3. Methods for assay quality assurance
2.3. Specificity of analytical techniques
2.4. Data evaluation
2.4.1. Assay accuracy and precision
2.4.2. Assay dynamic range
3. ABSORPTION
3.1. General introduction
3.1.1. Simple diffusion
3.1.2. Filitration
3.1.3. Specialized transport systems
3.2. Gastrointestinal absorption
3.2.1. General considerations
3.2.2. In vivo methods
3.2.2.1 Measurement
3.2.2.2 The site of gastrointestinal absorption
3.2.3. In vitro methods
3.2.3.1 Isolated loops
3.2.3.2 Isolated cells and vesicles
3.3. Pulmonary absorption
3.3.1. General considerations
3.3.2. In vivo methods
3.3.2.1 Methods for the pulmonary exposure
of intact animals
3.3.2.2 Methods for the pulmonary exposure
of anaesthetized animals
3.3.3. In vitro methods
3.3.3.1 Perfused lungs
3.3.3.2 Fluid-filled lung lobes
3.3.3.3 Isolated cells
3.4. Dermal absorption
3.4.1. General considerations
3.4.2. In vivo methods
3.4.2.1 Methods for dermal exposure
3.4.3. In vitro methods
3.4.3.1 Isolated skin preparations
3.4.3.2 Different cell populations
3.5. Other routes of exposure
3.5.1. General considerations
3.5.2. The intravenous (iv) route
3.5.3. Intraperitoneal (ip) absorption
3.5.4. Intramuscular (im) administration
3.5.5. Subcutaneous (sc) administration
4. DISTRIBUTION
4.1. General considerations
4.2. Invasive methods
4.2.1. Qualitative methods
4.2.1.1 Autoradiographic methods
4.2.2. Quantitative methods
4.2.2.1 Radiometric methods
4.2.2.2 Chemical methods
4.3. Non-invasive methods
5. BINDING
5.1. General considerations
5.2. Methods for assessing reversible binding
5.2.1. Extracellular sites
5.2.2. Intracellular sites
5.3. Methods for assessing irreversible binding
6. METABOLISM
6.1. General considerations
6.2. Important enzymatic pathways in xenobiotic
metabolism
6.2.1. Phase I reactions
6.2.1.1 Oxidation reactions
6.2.1.1.1 Cytochrome P-450 monooxy-
genase system (EC 1.14.14.1)
6.2.1.1.2 Microsomal flavin-containing
monooxygenase (EC 1.14.13.8)
6.2.1.1.3 Cooxidation by prosta-
glandin H synthase
(EC 1.14.99.1)
6.2.1.1.4 Miscellaneous peroxidative
pathways
6.2.1.1.5 Alcohol dehydrogenase (EC
1.1.1.1) and aldehyde
dehydrogenase (EC 1.2.1.3)
6.2.1.1.6 Monoamine oxidase (EC 1.4.3.4)
6.2.1.2 Reduction reactions
6.2.1.2.1 Cytochrome P-450-
dependent reactions
6.2.1.2.2 Flavoprotein-dependent reactions
6.2.1.2.3 Carbonyl reductases
6.2.1.3 Hydrolysis reactions
6.2.1.3.1 Epoxide hydrolase (EC 3.3.2.3)
6.2.1.3.2 Carboxylesterases/amidases
6.2.2. Phase II reactions
6.2.2.1 UDP-glucuronosyltransferase (EC 2.4.1.17)
6.2.2.2 Sulfotransferases
6.2.2.3 Mercapturic acid biosynthesis
6.2.2.3.1 Glutathione S-transfer-
ases (EC 2.5.1.18)
6.2.2.3.2 Cysteine conjugate betalyase/
thiomethylation
6.2.2.4 Amino acid N-acyltransferases
6.2.2.5 N-acetyltransferases (EC 2.3.1.5)
6.2.2.6 N- and O-methyltransferases
6.3. Modulation of important metabolic pathways
6.3.1. Physiological factors
6.3.1.1 Age
6.3.1.2 Genetic factors
6.3.1.3 Sex hormones
6.3.1.3.1 Sex-linked differences
6.3.1.3.2 Pregnancy
6.3.1.4 Thyroid hormones
6.3.1.5 Corticoid hormones
6.3.1.6 Pituitary hormones
6.3.1.7 Immune system
6.3.2. Environmental factors
6.3.2.1 Enzyme induction
6.3.2.2 Inhibition
6.3.3. Pathological factors
6.3.3.1 Liver disease
6.3.3.1.1 Acute viral hepatitis
6.3.3.1.2 Chronic hepatitis and cirrhosis
6.3.3.2.3 Obstructive jaundice
and cholestasis
6.3.3.2 Kidney disease
6.3.3.3 Diabetes
6.4. Sampling procedures for parent compounds and
metabolites in vivo
6.4.1. Non-invasive procedures
6.4.2. Invasive procedures
6.5. Experimental systems
6.5.1. Systems with intact cellular structure
6.5.1.1 Intact animals
6.5.1.2 Isolated organs
6.5.1.3 Freshly isolated cells
6.5.1.4 Organs and cells in culture
6.5.2. Cell-free systems
6.5.2.1 Subcellular fractions of tissue
homogenate
6.5.2.2 Purified enzymes and/or reconstituted
enzyme systems
6.5.3. Intestinal microflora
6.6. Methods for assessing chemically reactive
metabolites in vitro
7. EXCRETION
7.1. General considerations
7.2. Important excretory mechanisms
7.2.1. Diffusion and filtration
7.3. Sites of excretion
7.3.1. Kidney
7.3.1.1 Glomerular filtration
7.3.1.2 Tubular secretion
7.3.1.3 Tubular reabsorption
7.3.2. Liver-biliary excretion
7.3.2.1 Enterohepatic circulation
7.3.3. Other excretory sites
7.4. Modulation by physiological, environmental, and
pathological factors
7.4.1. Urinary excretion of xenobiotics
7.4.1.1 pH and urine volume
7.4.1.2 Inhibition and stimulation by
xenobiotics
7.4.1.3 Age differences
7.4.1.4 Species differences
7.4.1.5 Renal dysfunction
7.4.2. Biliary excretion
7.4.2.1 Species and age differences
7.4.2.2 Effects of physiological compounds
7.4.2.3 Effects of xenobiotics
7.4.2.4 Hepatic disease and regeneration
7.5. Methods for assessing excretion
7.5.1. Whole animals
7.5.2. In vitro preparations
7.5.2.1 Isolated organs
7.5.2.2 Intestinal preparations
7.5.2.3 Slices of renal cortex
7.5.2.4 Other kidney preparations
7.5.2.5 Purified membrane preparations
8. KINETIC MODELS
8.1. General considerations
8.2. Dose-independent kinetics
8.2.1. One-compartment model
8.2.1.1 Single dose
8.2.1.2 Repeated dosing
8.2.2. Two-compartment model
8.2.2.1 Single dose
8.2.2.2 Repeated dosing
8.3. Kinetics of metabolites in the presence of
parent compound
8.4. Non-linear kinetics
8.5. Physiological kinetic models
8.6. Modulation of kinetics
9. TOXICOKINETIC METHODOLOGY IN THE ASSESSMENT OF
HUMAN EXPOSURE
9.1. General considerations
9.2. Analysis of parent compounds or metabolites
9.2.1. Toxicokinetics and sampling strategy
9.2.2. Dermal absorption
9.2.3. Specimens in use
9.3. Effect monitoring
9.4. Monitoring of exposure to carcinogens
9.4.1. Urinary mutagenicity
9.4.2. Alkylation or arylation of proteins,
peptides, amino acids, and nucleic acids
9.4.3. Chromosomal damage
9.5. Preanalytical error
9.5.1. Physiological and environmental sources
of variation
9.5.2. Variation associated with speciment
collection and storage
10. ASSESSMENT OF TOXICOKINETIC STUDIES
10.1. General considerations
10.2. Analytical data
10.3. Absorption data
10.4. Distribution data
10.5. Reversible binding data
10.6. Metabolism data
10.7. Excretion data
10.8. Kinetic model data
10.9. Human data
REFERENCES
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.
IPCS TASK GROUP ON TOXICOKINETICS
a,c Professor A. Aitio, Institute of Occupational Health,
Helsinki, Finland
a,c Professor E.A. Bababunmi, Department of Biochemistry,
University of Ibadan, College of Medicine, Ibadan,
Nigeria (Vice-Chairman)
a,b Dr J.R. Bend, National Institute of Environmental Health
c,e Sciences, Research Triangle Park, North Carolina,
USA (Rapporteur)
b Professor H. Bräunlich, Institute of Pharmacology and
Toxicology, Friedrich Schiller University of Jena,
Jena, German Democratic Republic
c Professor I. Darmansjah, Department of Pharmacology,
University of Indonesia, Medical School, Jakarta,
Indonesia
a,b,c Dr E. Dybing, Department of Toxicology, National
Institute of Public Health, Oslo, Norway (Chairman)
b Professor H. Hoffmann, Academy of Sciences, Central
Institute of of Microbiology and Experimental
Therapy Jena, Jena, German Democratic Republic
a,b Professor W. Klinger, Institute of Pharmacology and
c,e Toxicology, University of Jena, Jena, German
Democratic Republic
b,d,e Dr D. Müller, Institute of Pharmacology and
Toxicology, University of Jena, Jena, German
Democratic Republic
a,c,e Professor S.D. Nelson, Department of Medicinal Chemistry,
University of Washington, Seattle, Washington, USA
a,c,e Professor O.G. Nilsen, Department of Pharmacology and
Toxicology, University of Trondheim, Trondheim,
Norway
e Dr Y. Ohno, Division of Medical Chemistry, National
Institute of Hygienic Sciences, Tokyo, Japan
e Dr A. Takahashi, Division of Medical Chemistry,
National Institute of Hygienic Sciences, Tokyo, Japan
a,b Dr A. Tanaka, Division of Medical Chemistry, National
c,e Institute of Hygienic Sciences, Tokyo, Japan
Secretariat
Dr J. Parizek, International Programme on Chemical Safety,
World Health Organization, Geneva, Switzerland (Secretary)
-------------------------------------------------------------------
a Planning meeting, Oslo, 15-18 August 1984.
b Preparatory meeting, Jena, 3-7 March 1985.
c Task Group meeting, Oslo, 8-13 September 1985.
d Representing the Institute of Pharmacology and Toxicology,
Jena, at the Task Group meeting, Oslo, 8-13 September 1985.
e Contributed in drafting sections of the document.
1. INTRODUCTION
Since the publication of EHC 6: Principles and Methods for
Evaluating the Toxicity of Chemicals, Part I which included a
section dealing with chemobiokinetics, substantial progress has
been made in studies on toxicokinetics and recognition of their
role in evaluating and elucidating the toxicity of chemicals. For
this reason, it was felt that a special monograph should be
prepared on the subject to assist both those conducting the
relevant studies and those using them in evaluations. This
includes the readers who use the IPCS documents for the evaluation
of the effects of specific chemicals.
The outline of the document and the strategies used for its
development were agreed on at a planning meeting convened for the
International Programme on Chemical Safety by the National
Institute of Public Health, Oslo, Norway, 15-18 August 1984.
Dr E. Dybing was elected Chairman of the meeting,
Professor E.A. Bababunmi, Vice-Chairman, and Dr J.R. Bend,
Rapporteur. DR DYBING agreed to guide the group of experts
throughout the whole development of the monograph.
A second preparatory meeting was hosted for the International
Programme on Chemical Safety by the Institute of Pharmacology and
Toxicology of the Friedrich Schiller University of Jena, German
Democratic Republic, 3-7 March 1985.
The text of the whole document was finalized at the Task Group
meeting hosted by the National Institute of Public Health, Oslo,
Norway, 8-13 September 1985.
The International Programme on Chemical Safety wishes to
acknowledge the work of the Chairman (DR E. DYBING) and members of
the Task Group, and of all who contributed to the preparation of
the document including: Professor S.D. Nelson who drafted section 2
(which was later reviewed by Dr W.R. Porter, Abbott Laboratories,
North Chicago, Illinois, USA); Professor W. Klinger who drafted
sections 3, 4, and 5; Dr R. Bend who drafted section 6;
Dr D. Müller who drafted section 6.3; Dr A. Tanaka who drafted
section 7, and whose co-workers Dr A. Takahashi and Dr Y. Ohno
drafted section 7.4 and 7.5.2 respectively; Professor O.G. Nilsen
who drafted section 8 (which was later reviewed by Professor K. S.
Pang, Faculty of Pharmacy, University of Toronto, Toronto, Canada);
and Professor A. Aitio who drafted section 9. The members of the
Task Group drafted section 10.
The financial support is gratefully acknowledged of: the
National Institute of Public Health, Oslo, Norway, for the planning
meeting and the Task Group meeting in Oslo; and the Ministry of
Public Health of the German Democratic Republic, Berlin, and the
Institute of Pharmacology and Toxicology of the Friedrich Schiller
University of Jena for the preparatory meeting in Jena.
2. ANALYTICAL METHODS
2.1. General Considerations
The primary hypothesis underlying studies of the metabolism
and toxicokinetics of a chemical is that the adverse biological
effects of the substance are correlated with its concentration in
the tissues of the organism (microorganisms, plants, or animals,
including human beings) in which the toxic effect is observed.
Consequently, it is always necessary to both identify and quantify
the toxic chemicals that might be present in a sample. The methods
that have been used to accomplish these tasks cover the whole range
of modern analytical techniques. Some generally useful methods
will be discussed in this section, but the primary emphasis will
be on the criteria used to judge the acceptability of proposed
methods and the mathematical techniques used to evaluate the data
obtained.
It is nearly always possible to devise several acceptable
methods to identify and quantify chemicals of interest. Indeed, it
is highly desirable to use more than one method, at times. If two
or more methods yield essentially the same results, confidence in
each method is increased. Several possible analytical methods
should always be considered before beginning a new study. The
criteria that should be used for the final selection of a method
are assay specificity, sensitivity, speed, simplicity, and cost.
In toxicokinetic studies, chemicals will nearly always be
present in biological samples in very low concentrations. After
all, only chemicals of relatively low toxicity would be
administered to test animals at the g/kg body weight level.
Toxicological studies are more concerned with the chemicals that
are toxic at the µg/kg body weight level. Indeed, certain
chemicals produce health hazards at even lower dose levels,
especially after long-term exposure. These extremely toxic
substances pose an analytical challenge of the greatest magnitude,
and individuals engaged in research on such extremely toxic
substances must devote much of their time to developing analytical
skills at the forefront of technology.
Because of the extremely low concentration of chemicals of
toxicological interest in samples of biological origin, the
analytical process usually consists of three stages (Smith &
Stewart, 1981). In the first stage, the toxic chemical is
separated from the bulk of the biological material present in the
sample (i.e., the sample matrix) and concentrated for further
processing. This sample-processing step is similar for most
toxicological investigations, since the primary problems
encountered are related to the nature of the biological materials
to be removed (proteins, lipids, salts, etc.) and not to the nature
of the toxic chemical. Once the toxic chemical has been isolated
from the sample matrix, additional separation steps are usually
required in order to obtain material of sufficient purity to permit
identification and quantification. These steps are usually
different for various toxic chemicals, as the methods used depend
heavily on the chemical and physical properties of the chemical
being investigated, as well as on the requirements of the specific
instruments or techniques used for identification and
quantification. Finally, selected biological, chemical, or
physical methods must be used to identify the toxic chemical
unambiguously and to determine the exact amount present in the
sample. Obviously, this complex process may be subject to many
errors that must be guarded against. General problems associated
with each of these stages will be discussed in section 2.2. The
most serious error that might arise is lack of specificity in the
assay method; the analytical result may not be due to the test
chemical in the sample, but may result from interfering sustances
in the sample or from substances inadvertently introduced during
the complex stages of sample preparation. This problem will be
addressed in section 2.3. Whichever analytical methods may be
employed to isolate, identify, and quantify the toxic chemical,
certain general methods of assuring assay validity and
statistically correct treatment of the data obtained can be used.
These will be discussed in sections 2.2 and 2.4.
When a new chemical (not previously or not sufficiently studied
for toxicological effects) is being tested, several analytical
methods may be required during the course of a study. These
methods can be classified according to their intended sensitivity
and specificity. For example, early in the course of the study of
a new chemical, relatively general approaches can be tried in order
to develop a satisfactory method for preliminary studies without
investing large amounts of time or money. As the study progresses,
it may become evident that these methods are not specific enough
for the measurements that are required, or that the methods are not
capable of providing all of the types of information desired. More
specialized methods may then have to be developed, once it is
apparent that investing more time and money in the project is
appropriate. Ultimately, the importance of the chemical may
dictate that extensive environmental monitoring and detailed
studies of the mechanism(s) of toxicity are required. These
studies will only be possible if highly specialized analytical
methods are used. Reliability, ruggedness, and cost effectiveness
during use may be major considerations in the selection of such
methods, which are usually costly and time-consuming to develop.
A combination of analytical methods may be necessary for other
reasons. In any toxicokinetic study, the identity and purity of
the chemical used in the test must be assured. Analytical methods
capable of detecting undesirable impurities will be required, as
well as methods to assure that the chemical is of uniform potency
from batch to batch. Additional methods will be required to
monitor the stability and uniformity of the form in which the test
substance is administered to the organisms used in the
toxicokinetic studies. Finally, methods suitable to identify and
quantify the test substance in toxicokinetic studies must be
employed. It is unlikely that a single analytical method will be
of use for all of these purposes, though sometimes only minor
modifications of a procedure may be required to accomplish most of
the tests. The necessity of using several different analytical
approaches in such studies is frequently overlooked by scientists
trained in areas other than analytical chemistry.
2.2. Techniques
2.2.1. Methods of isolation
In order to identify and quantify a toxic chemical in a sample
of biological origin, it is first necessary to separate the
chemical from the bulk of the biological components of the sample.
The traditional separation methods in analytical chemistry have all
been used for this purpose in the past, including precipitation,
distillation, centrifugation, liquid-liquid extraction, liquid-
solid extraction, adsorption, ultra-filtration, and complexation.
These techniques have been developed extensively by scientists in
the fields of biochemistry, pharmacognosy, clinical chemistry, and
agricultural chemistry. The literature in these fields is a rich
source of information for the scientist interested in metabolism.
Many of the techniques required for the preparation of samples for
further processing are standard, and textbooks in pharmaceutical
analysis, pesticide residue analysis, clinical chemistry, or
forensic toxicology should be consulted for examples of useful
techniques (Henry et al., 1974; Sunshine, 1975; Tietz, 1976; Bauer
et al., 1978; Connors, 1982; Schirmer, 1982; Baselt, 1984;
Kratochvil et al., 1984).
It is usually necessary to desorb the toxic chemical from the
macromolecular components of the sample and to remove proteins,
lipids, etc., that may complicate further analysis. Careful thought
should be given to the method of sample preparation; it must be
sufficiently gentle to allow recovery of the toxic chemical without
degradation, yet sufficiently powerful to ensure the recovery of
all, or nearly all, of the toxic chemical in the sample. Any
method that is used must be evaluated to make certain that recovery
of the test chemical is consistent and reasonably complete. Such
an evaluation presupposes that methods are available to identify
and quantify the test chemical and to separate it from substances
that might interfere with the measurement process. Consequently,
though sample preparation is the first step in a complete assay
method, it is usually the last step developed when a new method is
tried.
Sample preparation methods are usually evaluated by "spiking"
experiments. A known quantity of the test chemical is deliberately
added to a sample known not to contain the substance. After
careful and thorough mixing, the sample is processed by the
proposed technique and the amount of test chemical recovered is
measured. Ideally, this should be identical to the amount added.
Recovery of more than the amount added is a strong indication
of sample contamination or the unsuspected presence of the test
chemical in the sample prior to the "spiking" experiment. Another
possibility could be lack of specificity. If the "spiking"
experiment is repeated with different amounts of added test
chemical, it may be possible to estimate the amount of chemical
apparently present in the sample originally. However, it may be
that the chemical was not actually present initially, but that the
sample became contaminated during processing, or, that substances
present in the sample interfered with the measuring process. In
these cases, the process requires further investigation.
"Spiking" experiments may also reveal significant losses of the
test chemical during processing. If this is the case, the
processing method will have to be modified in such a way as to
reduce the losses.
The "spiking" experiments may show good recovery of the test
chemical, but this could be the result of the fortuitous balancing
of sample losses with interfering substances. Measurement of the
test chemical recovered in these experiments by two or more
methods should yield consistent results. Failure to do so
indicates that the processing method requires further evaluation.
When a sample processing technique seems to be functioning
adequately, it is still necessary to verify, by repeated
experiments, that the results of the method are reproducible within
the limits of precision required. A frequently used method to
improve precision is the thorough mixing into the original sample
of a known amount of a standard substance that can be isolated
along with the test chemical. This internal standard is then used
to obtain a relative measure of the amount of test chemical through
appropriate calibration experiments. Such internal standards
should not be present in the original sample, should show similar
recovery to the test chemical, and should be detectable by the same
analytical technique.
From the foregoing remarks, it can be seen that validation of
the sample processing stage may require elaborate experimentation.
Despite the fact that most sample processing techniques are
standard, the conditions used in actual experimentation require
careful control. Furthermore, validation of the processing
technique for one type of sample matrix (e.g., plasma) does not
guarantee that the method will work with a different sample matrix
(e.g., urine or tissue homogenate). It will be necessary to
revalidate the sample preparation technique for each type of sample
matrix encountered.
Sometimes results that seem anomalous will occur during the
practical application of an analytical method. It may be that
unexpected interference with the method has occurred through
contamination of the sample by residues extracted from glassware,
pipettes, sample vial caps, etc., or through a failure of the
sample processing technique to remove certain substances. Such
anomalous results merit investigation. The specific steps that
must be taken to remedy defects in the sample processing stage will
depend on both the nature of the chemical under study and the
nature of the processing procedure.
In recent years, techniques of sample preparation based on the
column chromatography process of adsorption, partition,
ultrafiltration, or complexation have become popular. A relatively
short column is employed, and elution solvents are selected to
elute the compound of interest as a class with other compounds of
similar chemical and physical properties. Thus, it may be possible
to recover the test compound and its degradation products or
metabolites as a mixture, which can then be separated further.
Such sample clean-up techniques are only useful for separating the
test compound from substances with grossly different chemical and
physical properties. They may, in fact, also separate the test
compound from metabolites or degradation products that are to be
measured. This potential problem illustrates the necessity of also
validating sample processing techniques for suspected metabolites.
This may not always be possible, unless authentic samples of the
suspected metabolites are available for testing.
A variant of the "digital chromatography" technique, which is
especially useful in preliminary studies, makes use of commercially
available thin-layer chromatography plates that incorporate a
"spotting zone" about 3 cm in height on a 20 cm high plate.
Samples of up to 50 ml of plasma, urine, or 25% tissue homogenates
can be applied to a 1 cm wide segment of the "spotting zone".
After thorough drying, the toxic chemical can be eluted from the
biological materials with a suitable solvent, such as methanol.
The plate is developed only to the beginning of the conventional
adsorbant layer, and this process is repeated several times. As a
result, the toxic chemical is concentrated in a narrow band at the
beginning of the adsorbant layer, and conventional elution
techniques can then be used to separate it from other materials
that might also have been concentrated along with it. This
technique is especially easy to use if a radioisotope labelled form
of the toxic chemical is available. Excess unlabelled material can
then be added to the sample to help dissociate the test substance
from macromolecular binding sites. The added unlabelled material
also serves to reduce overall decomposition of the labelled test
substance during analysis, by a dilution effect. This technique
can also be used for monitoring the uniformity and stability of the
dosage, and may be applicable for determining the purity of the
bulk test chemical as well. Covalent binding of the labelled
chemical to tissue components can readily be detected using this
technique, since any bound material should remain in the "spotting
zone." These methods seem to be generally useful in the study of
toxicokinetics of many substances, at least as a first approach.
Of course, validation experiments must be performed with this
method, as with any other.
2.2.2. Methods of purification, identification, and quantification
Once a toxic chemical has been separated from the biological
components of the sample matrix, it will normally be necessary to
further purify it by additional separation procedures. The degree
of further purification required depends entirely on the
selectivity and specificity of the measurement techniques to be
used to identify and quantify the toxic chemical. If a very
general and nonselective measurement method is used, the burden of
assuring adequate specificity, for the assay procedure as a whole,
will fall on the purification stage. On the other hand, if the
toxic chemical possesses unique biological, chemical, or physical
properties, these may be exploited to develop highly selective
measurement methods for unambigously identifying and quantifying
the substance. If this is the case, very little, if any,
purification may be required. For this reason, purification
procedures are normally developed simultaneously with the
measurement procedures that will be employed. Up-to-date reviews
of the various methods, as applied to specific groups of chemicals,
can be found in the Application Reviews of Analytical Chemistry.
The types of purification procedures that have been used in the
past have varied widely. Since the measurement of the metabolites
or degradation products of the toxic chemical is usually required,
methods that separate these products from the parent compound in
such a way that they can be measured simultaneously are desirable.
This goal is usually achieved by adopting a multiple, continous
purification process, such as a chromatography system.
Most chromatography systems exploit differences in adsorption,
partition, molecular size or shape, or complexation properties to
effect a separation of chemicals that may otherwise have very
similar chemical and physical properties. The behaviour of a
chemical in a given chromatographic system can usually be predicted
adequately on the basis of knowledge of its chemical structure,
thus allowing a reasonable choice of chromatographic technique
(Lyman et al., 1981). The fact that only small amounts of the test
chemical are present in samples from toxicokinetic studies is an
advantage when selecting chromatographic techniques, as a wider
range of possible systems can be evaluated.
The mobile phase used in a chromatography system can be either
a gas, liquid, or a supercritical fluid. Because the hardware for
gas chromatography (GC) has reached a higher state of technological
improvement than the hardware for liquid or supercritical fluid
chromatography, it is generally more reliable (Freeman, 1979;
Jorgenson, 1984). The separating power of gas chromatography
equipment has been increased greatly through the use of capillary
columns (Jennings, 1980). GC systems are usually less costly to
acquire, maintain, and operate than high performance liquid
chromatography (HPLC) systems. If a toxic chemical is volatile, or
can be made volatile by simple chemical modifications, GC is
usually preferred to HPLC. GC equipment is also more readily
interfaced to the mass spectrometer, a measuring device that can
provide highly reliable identification of the chemicals being
studied as well as quantification of the material identified.
GC/MS systems provide a purification and measurement package that
can be used in a wide variety of metabolic and toxicokinetic
investigations (Watson, 1976; Millard, 1978; Message, 1984).
However, highly-skilled operators are required to operate and
maintain the equipment. Despite the high initial cost of such
systems, and despite the need for highly skilled personnel to
operate them, it can be argued that such equipment is cost-
effective in the long run, if a wide variety of different toxic
chemicals is to be studied.
Some toxic chemicals, or their metabolites, are not
sufficiently volatile to be separated from contaminants by GC, and
a liquid supercritical fluid mobile phase must be used instead. In
recent years, HPLC systems of increasing sophistication have become
available. Such systems are now often the first choice because of
the flexibility in the choice of mobile and stationary phases for
the chromatographic portion of the system as well as the wide
variety of measurement devices that can be incorporated as
detectors. Detectors that measure ultraviolet, visible, or
infrared absorption spectra, fluorescene, atomic absorption,
refractive index, radioactivity, electrochemical properties, etc.,
have been found to be useful (Hadden et al., 1971; Tsuji &
Morozowich, 1978; Snyder & Kirkland, 1979; Krstulovic & Brown,
1982). Interfacing an HPLC system to the mass spectrometer is not
yet as well-established a technique as GC/MS, but progress in this
area continues to be made (Cooks et al., 1983).
Despite the attractiveness of GC and HPLC techniques, thin-
layer chromatography (TLC) is still very useful and is especially
powerful when combined with radioisotope tracer methods for
establishing the identity and quantity of a toxic chemical and its
metabolites. Significant advantages of this combination are that
nearly every toxic substance of interest can be studied using this
technology, only one analytical instrument (a radiation counter) is
required, labour and material costs are low, and the level of skill
required to perform TLC/radiochemical assays is modest. Mass
balance experiments, an important requirement in validation
studies, are easily performed. The major drawbacks of this
technique are the necessity of having a radioisotope labelled
chemical, and that the procedures involved are slow and tedious
compared to GC or HPLC methods. However, for the scientist on a
limited budget, working in a area with low labour costs and a
shortage of personnel with advanced technical training,
TLC/radioisotope methods are especially attractive. The amount of
radioactive material required is generally quite small, and it is
used under experimental conditions that are unlikely to pose a
health hazard. Wastes can usually be disposed of by incineration
without releasing significant amounts of radioactivity into the
atmosphere. However, it is essential that radioisotope use be
supervised by individuals with adequate training in radiation
safety.
GC, HPLC, and TLC/radioisotope methods are commonly used for
studying toxicokinetics of various chemicals. Of these methods,
TLC/radioisotope technology is the easiest method to implement, the
method most likely to give useful results quickly, and the method
least affected by the level of skill of the technicians employed.
Although it rapidly becomes onorous if a large number of samples
need to be analysed over long periods of time, the TLC/radioisotope
method remains a good choice for early investigation of new
compounds.
In some cases, isolation, purification, and quantification of a
substance from a matrix can be carried out in only one stage. One
of the major developments of the past decade has been the
application of immunochemistry to the measurement of hormones,
drugs, toxins, and enzymes (Monroe, 1984). Radioimmunoassay (RIA)
is based on the existence of an equilibrium between an antigen (the
component to be measured), an antibody, and a corresponding
antigen-antibody complex in a system that includes trace amounts of
radioactively labelled antigen. A simple separation step is
required in order to determine the amount of labelled antigen that
is bound to the antibody or free in solution. Newer enzyme
immunoassay (EIA) techniques such as EMIT and ELISA do not require
radioactivity and EMIT does not require a separation step. These
methods involve coupling an enzyme with an antibody or antigen.
All of these techniques can be highly sensitive with the ability to
measure concentrations of chemicals in the range of ng - µg/litre.
Whether RIA or EIA techniques are used, the major disadvantage of
the methods is the need for specific antibodies. Development of
monoclonal antibody techniques and the combining of liquid
chromatographic separation with immunoassays will no doubt decrease
problems associated with specificity. As production techniques for
antibodies are made more efficient, EIA techniques will be used
more extensively because highly trained technicians and special
instrumentation are not required.
2.2.3. Methods for assay quality assurance
After the chemical under study and its metabolites have
been separated from the sample matrix and purified, they must
be identified and quantified. Too often, in toxicokinetic
studies, it is not demonstrated that the methods of separation,
identification, and quantification used are adequate for the
particular study.
The method used to validate a particular purification and
measurement process (i.e., to show that the process in fact
measures what it purports to measure) depends, to some extent, on
the techniques that are used. Each scientist must become familiar
with his instruments and techniques, and must learn the problems
and pitfalls of these techniques and the way in which these can be
overcome. In addition to having the appropriate background
knowledge, there are certain other general precautions that can be
taken, and guidelines have been established by several
organizations for proper assay quality control (APA, 1972; Federal
Register, 1978; ASTM, 1979, 1981; ACS Committee on Environmental
Improvement, 1980; Digne, 1984). The following list summarizes the
more important requirements:
(a) Assure that the bulk chemical used in tests is of known
purity and free of undesirable contaminants. A biological
assay of a highly purified sample of the bulk chemical
should demonstrate the same toxicological response as the
bulk chemical.
(b) Assure that dosage forms used in biological experiments
are of uniform content and are stable under conditions of
actual use. A stability-indicating assay of the potency
of experimental dosage forms is required. Such assays in
turn require validation (i.e., the assay must show a loss
in potency in an artifically degraded sample consistent
with the amount of degradation induced).
(c) Assure that materials used as standards are of known
potency and purity. Additional analytical tests may be
required, and it may be necessary to repeat these tests at
regular intervals.
(d) Assure that experimental work is conducted as planned,
that any modifications are documented, and that adequate
records of the experiments are maintained as they are
conducted. Unusual observations should be noted and
investigated.
(e) Assure that adequate tests to challenge the ability of the
analytical procedures used to discriminate against
potentially interfering substances have been conducted.
These tests are discussed in section 2.3.
(f) Assure that uniform or controlled conditions have been,
or can be, maintained throughout the duration of the
experiment. Verify experimentally that different analysts
or different instruments yield data of the same quality
with respect to assay sensitivity, precision, and accuracy.
These problems are discussed in section 2.4.
(g) Assure that the biological samples are stable during the
period prior to analysis, and that the analytical
procedure does not itself cause the substance being
analysed to degrade. If it is impossible to avoid some
degradation, determine experimental conditions to assure
that such degradation will be consistent throughout the
period of experimentation.
(h) Assure that samples are analysed in a statistically valid,
randomized fashion. Ideally, the analyst should not know
the identity of the samples until after the analysis has
been completed. If it is necessary to assay samples in a
particular order, make certain that this information is
carefully documented.
(i) Assure that the right samples are being analysed. Guard
against faulty labelling or record-keeping.
(j) Assure that instruments are properly maintained and in a
good working order.
(k) Assure that instruments are calibrated properly. Verify
that calibration is performed at appropriate intervals.
(l) Assure that experimental samples and standards, and
"reagent blanks" or control samples are processed
identically. Verify that any "reagent blank" samples are
appropriately selected.
(m) Assure that the results obtained are reproducible and
accurate. There are many ways to achieve this aim, among
which are periodic reanalysis of experimental samples
previously assayed in the laboratory (internal control),
and the comparison of results obtained from the analysis
of samples that have been assayed in other laboratories
either by similar or different methods (external control).
If the preceding guidelines are adhered to, the quality of the data
produced from toxicokinetic experiments should be controllable,
with the result that interpretations made from such data are more
likely to be grounded on fact.
2.3. Specificity of Analytical Techniques
Assay specificity is perhaps the most serious problem
encountered, since the chemicals being measured are present in such
low concentrations in the samples that it is potentially possible
for many other components in the sample to copurify with the
chemicals of interest. If this should occur, then these
contaminants may interfere with either the identification or the
quantification process or both. Data obtained using measurement
techniques of too general use should be regarded with suspicion.
For example, most chemicals absorb light somewhere in the
ultraviolet region of the spectrum, especially at low wavelengths.
An assay method that depends on the measurement of the absorption
of ultraviolet radiation at low wavelengths for quantification is
unlikely to be very specific. A purification technique of great
power must therefore be used to minimize the potential for the
occurrence of interference in the measuring process.
The use of "reagent blanks," i.e., samples that are supposedly
free of the chemical being assayed, is helpful in demonstrating the
lack of interfering substances. Such "reagent blanks" must be
selected with care, as they should be otherwise identical to the
samples that will be analysed for the toxic chemical, except for
the absence of the test chemical. Because of the genetic diversity
of biological test organisms, it is difficult to select an
appropriate "reagent blank."
Although "blanks" provide some assurance that no instrument
response will be obtained in the absence of the test chemical, a
better approach is to select an instrument or bioassay that
responds to some biological, chemical, or physical property of the
test chemical that is not shared with many other substances. If a
bioassay is used, the organisms used in the assay should ideally
respond only to the test chemical, or to a limited number of other
chemicals that can be shown not to copurify with the test chemical.
"Information rich" methods of analysis, such as mass spectrometry,
infrared photometry, nuclear magnetic resonance spectrometry, or
batteries of monoclonal antibodies provide the user with greater
assurance that he or she is measuring the chemical under test, and
not some interfering substance. Unfortunately, at present, only
mass spectrometry and specific immunoassays have adequate
sensitivity to be used routinely in toxicokinetic studies.
It is advisable to use a second analytical technique to
confirm, at least on a spot-check basis, that the usual method is
producing correct results. The second method should ideally be as
different as possible, at both the purification stage and the
measurement stage. If both assay methods agree, there is less
likelihood that interfering substances are present, since, if they
were present, they would have to interfere, to the same extent,
with both assays. This is so unlikely that a confirming assay is
often used as a check for assay specificity. Of course, the
confirming assay may be invalid if it is too similar to the
original method.
Sometimes it is possible to increase the information content of
the instrumental signal. For example, if ultraviolet
spectrophotometry is being used to monitor the effluent from a
liquid chromatography column, measurements at a number of
wavelengths could be obtained. If an interfering substance is
present in some samples, but not in others, or if an interfering
substance is present in different amounts in each sample, the ratio
of ultraviolet absorbances at two or more wavelengths will change
(unless the interfering substance has an absorbance spectrum
identical to that of the test chemical). This extra data can be
used for quality assurance purposes without complicating the normal
data processing procedures.
"Spiking" experiments can sometimes be used to verify the
absence of interfering substances. A single sample is split into
several subsamples, some of which are "spiked" with known,
different amounts of authentic test chemical. All of the
subsamples are then analysed and the amounts found in the variously
"spiked" samples are extrapolated to zero "spike" level. The
results should agree with the amounts actually found for the
unspiked subsamples.
Finally, it is helpful to repeat the purification process on a
large scale, using a representative sample. Fractions
corresponding to the chemical and its metabolites can then be
collected and studied by a variety of other analytical techniques,
in order to confirm their identities using all of the available
methods for structure elucidation that can be employed. If these
procedures are followed, it will be reasonably certain that the
assay is specific for the toxic chemical and its metabolites. In
practice, however, "suspicious" results will occasionally be
observed, and, in this case, the sample material should be
reassayed, preferably by a second method. In this way, other
sources of interference may be detected during the course of
experimentation.
2.4. Data Evaluation
2.4.1. Assay accuracy and precision
If the precautions noted previously are taken, the assay method
should yield accurate results. Nevertheless, however accurate the
data may be, they may still not be sufficiently precise to be
useful. In the context of analytical methods, "accuracy" refers to
how closely the average value reported for the assay of a sample
agrees with the actual amount of substance being assayed in the
sample, whereas "precision" refers to the amount of scatter in the
measured values around the average result. If the average assay
result does not agree with the actual amount in the sample, the
assay is said to be "biased", i.e., lacks specificity; bias can
also be due to low recovery (section 2.3). Assuring accuracy
requires all of the elaborate experimentation discussed previously,
while assuring precision requires repeating the experiment a number
of times and then statistically analysing the averaged results.
However, this improvement is achieved at great cost, as every
halving of the degree of scatter in the average requires
quadrupling the number of measurements. High precision can be
better achieved through improved experimental designs (Cochran &
Cox, 1957; Aarons, 1981; Mitchell & Garden, 1982).
Before proceeding further, it is useful to discuss the level of
precision that is really needed in toxicokinetic experiments. The
overall level of precision that can be achieved will always be
limited by the intrinsic genetic variability of the organisms used
in the experiments. This can be reduced, to a certain extent, by
clever experimental design, but it nevertheless remains a major
limitation. In the usual situation, statistical theory can be used
to demonstrate that, if other sources of variation total less than
about 30% of the biological component of variation, their overall
effect will be small. Thus, if the biological variation expected
in the experiment will cause, on average, about a 20% relative
scatter in the data, it is only necessary to reduce the relative
scatter of the analytical procedures to less than 6% to achieve
adequate precision. This is well within the capabilities of GC,
HPLC, TLC, or immunoassay methods used in conjunction with
appropriate detectors and conventional laboratory sample handling
procedures.
It can also be achieved by bioassay methods if sufficient
replications of an appropriately designed assay are conducted.
Bioassays require careful experimental design guided by appropriate
statistical considerations in order to achieve high precision
(Finney, 1978). It is almost always necessary to compare each
unknown sample to an authentic standard, using a procedure in which
several dilutions of both sample and standard are tested
simultaneously, in randomly selected test organisms, under
identical conditions. A relative potency is then calculated by
plotting the responses obtained for the various dilutions of both
the standard and unknown sample and calculating an average ratio of
sample to standard potency using properly weighted nonlinear
regression analysis. With the advent of low-cost digital
computers, there is no reason for failing to use the correct
computational procedures. Once a computer program has been
obtained to perform potency ratio computations, it can be used for
all future bioassay work. The use of a calibration curve, which is
standard practice for chemical or physical analytical data, is
almost never permissible in bioassay work.
Data obtained from assays in which the final measurements are
based on chemical reactivity or observation of a change in a
physical property are usually analysed by means of a calibration
curve. Calibration curves must be monotonic (an increase in the
concentration of the test chemical must always either increase or
decrease the measured property or process). Although computation
is simplified if calibration curves are straight lines, this is by
no means essential (Hubaux & Voss, 1970; Garden et al., 1980;
Kurtz, 1983; Moler et al., 1983; Schwartz, 1983). On the other
hand, excessive curvature is undesirable, because the assay
sensitivity is proportional to the rate of change in measured
response to the rate of change in test chemical concentration, and
this should be reasonably constant for the assay method to be
useful. Usable calibration curves can nearly always be well-
approximated by a cubic polynomial function. Statistical theory
can be used to select the best arrangement of standards to maximize
assay precision. Assay precision using data interpolated from a
calibration curve is affected by the precision of the measurements
of the unknown samples and the measurements of the standards (which
ought to be the same), and the precision with which the parameters
of the calibration curve can be determined. The last depends on
the design of the calibration experiments as well as on the
measurement precision. The concentrations for the standards should
cover a wider range than that of the unknown samples. If the
calibration curve is a straight line, then the average
concentration of the standards should be close to the anticipated
average concentration of the unknowns. The total number of
standards analysed should be greater than, or equal to, the number
of replicate measurements on a single unknown, but the best
precision will be obtained when the number of standards assayed is
equal to the number of replicate determinations for each unknown
sample. Replicating the measurements on unknown samples is usually
the best way to improve precision, since it is likely that the
number of standards used will greatly exceed the number of
replicate measurements on unknown samples.
Whatever type of assay may be performed, it is essential that
both the assay design and the data evaluation be conducted
according to correct statistical principles. If unsure how to
proceed, a statistican should be consulted before any experimental
work is begun. Salient references that should be consulted include
Youden & Steiner (1975), Montag (1982), Caulcutt & Boddy (1983),
Bolton (1984), and Delaney (1984).
2.4.2. Assay dynamic range
Although quality assurance and good experimental design will
help to improve assay accuracy and precision, it is also necessary
that the assay method be usable over a sufficiently wide range of
concentrations for the toxic chemical and its metabolites.
The lower limit of usefulness for an analytical method has been
perceived in different ways; frequently, the term "sensitivity" has
been used to indicate the ability of an analytical method to
measure small amounts of a substance accurately and with requisite
precision. This, however, is a misnomer; sensitivity correctly
refers to the slope of the calibration curve. The lower limit at
which the test chemical can be distinguished from a "reagent blank"
has been called the limit of detection for the analytical method.
This is the lowest level at which the chemical being analysed can
be identified. It is more useful to define a lower limit at which
adequate precision can be obtained (Long & Winefordner, 1983;
Oppenheimer et al., 1983); this may be called the limit of
quantification. Typically, the limit of quantification occurs at a
concentration fifteen to twenty times higher than the limit of
detection.
These concepts have limited use in practice; in toxicokinetic
studies, it is the range of concentrations over which a reasonably
uniform level of precision can be obtained that is important. This
range of concentrations may be called the dynamic range of the
analytical method. Typical assays useful for toxicokinetic studies
should be able to measure the test chemical over a 100-fold range
of concentrations. This is not often easy to achieve without
sacrificing some precision. If the dynamic range is not wide
enough, many invesitigators dilute and reassay samples that are too
concentrated, but if this is done, it is necessary to demonstrate
that the dilution process does not introduce additional error.
3. ABSORPTION
3.1. General Introduction
Absorption is the process(es) by which an administered
substance enters the body (OECD, 1981). For the purposes of this
document, absorption will be equated with the appearance of the
chemical in the circulation. The rate and extent of absorption of
the administered substance can be estimated by various methods,
with and without reference groups (i.e., a test group in which the
substance is administered via another route that ensures complete
availability of the dose). These methods include:
(a) determination of the amount of test substance and/or
metabolites in urine, bile, faeces, and exhaled air,
and that remaining in the carcass;
(b) comparison of a biological response (e.g., acute
toxicity studies) between test and control and/or
reference groups;
(c) comparison of the amount of dose excreted renally in
test and reference groups; or
(d) determination of the area under the plasma steady
state curve of the test substance and/or metabolites
and comparison with data from a reference group
(OECD, 1981).
The rate of absorption can be determined from the plasma
concentration time-curve of the test substance (sections 8.2.1.1,
8.2.1.2).
The skin is the main barrier that separates mammals including
man from environmental chemicals. However, chemicals are absorbed
via the skin and may produce damage. The major routes by which
toxicants enter the body are via the lungs, the gastrointestinal
tract, and the skin. Once the chemical has entered the blood-
stream, it may exert its toxic action directly in the blood or in
any target tissue or organ to which the circulatory system
transports the chemical.
Often, a chemical must pass through many "barriers" before
reaching its target. These barriers include the many membranes of
the cells of the skin, the layers in the lung and gastrointestinal
tract, the capillary cell, the cells of the tissue and organs where
the chemical exerts its damaging effect, and the cells of the
organs that eliminate the chemicals, mainly the liver and the
kidneys.
All cell membranes are similar: they consist of a bimolecular
layer of lipid molecules coated on each side with a protein layer,
branches of which penetrate the lipid bilayer or even extend right
through it. At physiological temperatures, the lipids of the
membranes (mainly phosphatidylcholine, cholesterol) have a
quasifluid character, determined by the structure and relative
proportion of unsaturated fatty acids. The higher the
concentration of unsaturated fatty acids, the higher is the
fluidity. Most foreign chemicals cross body membranes by simple
diffusion. The rate and extent of this diffusion or absorption is
influenced by many factors, summarized in Table 1.
Table 1. Factors influencing the rate and extent of absorption of
a chemicala
-------------------------------------------------------------------
Properties of the Morphology and dimension of the absorbing
organism body surface, perfusion of the absorbing
area, distribution and elimination processes,
general factors (e.g., nutritional status,
age, disease)
Characteristics of Relative molecular mass
the chemical Physical state
- conformation
- aggregation
- dispersion
Charge
- acid or base characteristics
Stability
Reactivity
Solubility in various solvents
Characteristics of Dose/concentration, duration of contact with
exposure the absorbing surface
Exogenous factors Formulation
- vehicle
- additives
Interaction with other chemicals
Physical conditions (e.g., temperature,
radiation)
-------------------------------------------------------------------
a Modified from: Scheler (1980).
3.1.1. Simple diffusion
Most organic molecules possess a certain degree of
lipophilicity and cross membranes by diffusion through the lipid
moiety. The rate of transfer depends on the lipid solubility,
which can be characterized by the lipid/water partition coefficient
(most frequently determined are the oil/water or octanol/water
partiton coefficents), and the concentration gradient across the
membrane.
Chemicals exist in solution in ionized and/or non-ionized
forms. The charged (ionized) form is generally less able to
penetrate cell membranes and, thus, diffusion is dependent on the
lipid-soluble non-ionized form of the substance. The dissociation
constant or the negative logarithm of the dissociation constant
(pKa) and the pH of the medium determine the degree of
dissociation. When the pKa of the chemical and the pH of the
medium are equal, 50% of the chemical exists in the ionized form.
Note: the pKa alone does not indicate whether a compound is an
acid or a base, because a basic chemical can have a pKa greater
than 7 and an acidic chemical a pKa lower than 7. The degree of
dissociation can be calculated according to the Henderson-
Hasselbach equation:
(anion-) x (H+)
for acids: K = ------------------
(non-ionized acid)
(cation+) x (OH-)
for bases: K = ------------------
(non-ionized base)
These equations can be transformed to:
non-ionized
for acids: pKa - pH = log -----------
ionized
ionized
for bases: pKa - pH = log -----------
non-ionized
Consequently, organic acids are more likely to cross membranes by
diffusion when they are in an acidic medium, and organic bases,
when in an alkaline medium.
3.1.2. Filtration
Very small hydrophilic compounds can pass through aqueous
channels or pores. This passage is called filtration, because it
involves the bulk flow of water due to hydrostatic or osmotic
forces. The size and number of these channels or pores differ
considerably in various membranes from 4 to 40 A (kidney
glomerulus). Such pores permit chemicals with a relative molecular
mass ranging from 100 - 200 to up to 60 000 to pass.
3.1.3. Specialized transport systems
These systems are important for nutrients and endogenous
substances (sugars, amino acids, amines, etc.), but less important
for most xenobiotics; they are relevant only for xenobiotics, such
as amines or organic anions, that are very similar to endogenous
substrates. Active transport is characterized by: a) the
requirement of energy, or energy-producing metabolism; b) a certain
selectivity with respect to the structure of the chemicals
transported; c) a limited capacity so that the transport system can
be saturated and a transport maximum is exhibited; and d) transport
of a chemical proceeding against electrochemical or concentration
gradients. Active transport is important for the pulmonary uptake
of, e.g., paraquat, a herbicide structurally similar to endogenous
diamines such as putrescine and cadaverine, and for the organic
anions phenol red and chromoglycate, as well as for the elimination
of the chemicals by the kidneys (tubular secretion of weak acids
and bases by acidic and basic carriers) and by the liver (biliary
secretion of weak acids and bases and neutral compounds). The
central nervous system also has two transport systems for the
active transport at the chorioid plexus, one for organic acids and
one for organic bases.
Facilitated diffusion (without energy requirement and without
movement of the chemical against a gradient) can account for the
uptake of acidic chemicals into liver parenchymal cells (Müller &
Klinger, 1975; Klaassen, 1980; Scheler, 1980; Klaassen & Watkins,
1984).
Phagocytosis and pinocytosis play an important role in the
uptake of particulate matter, e.g., in the lungs (alveolar
macrophages), in the subcutis (leukocytes, histiocytes), in the
liver (Kupffer cells), and in the gastrointestinal tract
(epithelial cells).
3.2. Gastrointestinal Absorption
3.2.1. General considerations
The gastrointestinal tract is a major site of absorption. In
principle, the contents of the gastrointestinal tract must be
considered exterior to the body and absorption can take place along
its full length, including the mouth and rectum. In general,
gastric juice is acidic and the intestinal contents, almost
neutral. Thus, a chemical will be absorbed predominantly in the
part of the gastrointestinal tract where it exists in the most
lipid-soluble form. Even if only a small percentage is present in
the non-ionized, diffusible form, e.g., weak acids in the
intestine, the very large surface area (villi, microvilli), long
contact time, and high concentration gradient (quick removal of
absorbed material due to a high perfusion rate) generally provide
high absorption rates, as equilibrium is always obtained. The
mechanisms by which some lipid-insoluble compounds are absorbed are
not clear. Some metals are absorbed by specialized transport
systems (e.g., thallium, cobalt, and manganese by the iron-system,
lead by the calcium-system). Special attention must be paid to the
stability of chemicals in the acidic stomach (e.g., esters can be
hydrolysed), to the enzymes in the stomach and intestine (peptides
will be split), and to the microbial flora, which have high
hydrolytic and reductive capacity. Moreover synthetic reactions
can take place, e.g., nitrosamine formation from secondary amines
in the stomach. Absorption from the gastrointestinal tract can be
modified by many factors such as nutrients (e.g., fat, milk),
fibres in the diet, by starvation, by alcohol, and by drugs that
alter stomach emptying time and/or motility of the gastrointestinal
tract.
3.2.2. In vivo methods
3.2.2.1. Measurement
Measurement of the chemical and its metabolites in the blood
stream, as well as in the urine, gives insight regarding the site,
extent, and rate of absorption. Bioavailability is the percentage
of a given dose that appears after absorption and distribution with
the portal blood through the liver in the circulating blood stream.
It comprises absorption rate (the percentage of a given dose that
disappears from the lumen of the gut) and the first pass-effect
(the percentage of the absorbed chemical that is filtered out from
the portal blood by accumulation and/or biotransformation in the
liver). Chemicals may be administered into the mouth cavity
(without swallowing of the material), into the stomach via a
gastric tube, or directly into the duodenum via a catheter. If a
chemical is absorbed from the mouth cavity or from the lower part
of the rectum it does not pass the liver. If it is administered
directly into the duodenum it is not exposed to the acidic
environment in the stomach. Bioavailability may be determined by
investigating the parent compound and its metabolites, mainly in
the urine, or in the urine plus faeces. The rate of uptake into
the circulating blood stream can be determined by investigating the
time-course of the concentration of the chemical in the blood
(section 8.2.1.1).
3.2.2.2. The site of gastrointestinal absorption
The site of gastrointestinal tract absorption can also be
determined in animals with sections of the gut separated by
ligatures yet retaining the vascular and nervous connections with
the body (suitable predominantly for small animals) or with
intestinal loops prepared from various segments of the
gastrointestinal tract with two external orifices (suitable only
for such animals as the cat, rabbit, dog, pig) (Giraldez et al.,
1984). Whereas in situ sections can be prepared in anaesthetized
animals only for a duration of a few hours, with all the possible
influences of the anaesthesia on circulation, peristalsis etc., the
so-called chronic intestinal loops are not filled with the
physiological intestinal contents, the microflora may be altered
and irritation and inflammation may influence absorption.
Absorption can be investigated by determining the blood levels and
excretion of the chemical and of its metabolites; luminal and
vascular perfusion techniques, including lymph collection, can also
be applied (Csąky, 1984; Windmueller & Spaeth, 1984).
3.2.3. In vitro methods
3.2.3.1. Isolated loops
Various methods with isolated loops [uneverted, everted loops
(sac preparations), loops consisting mainly of mucosa, intestinal
sheets and isolated villi] have been developed (Csąky, 1984;
Windmueller & Spaeth, 1984). Most investigations have been
carried out with various segments of the gastrointestinal tract and
the rate of the diffusion process can be determined under optimal
oxygenation and nutrition conditions in both directions (outside
in, inside out). Thus, simple diffusion can be determined under
conditions in which the influence of perfusion and the liver are
excluded. However, the whole wall of the gastrointestinal tract
is taken as the diffusion barrier in this model system and this is
not true in vivo.
3.2.3.2. Isolated cells and vesicles
The absorption of chemicals can also be determined in isolated
cells from the liver and gut and in isolated membrane vesicles
(Csąky, 1984; Murer & Hildmann, 1984). In these cases, only the
cell membrane is a barrier to diffusion. Time courses of uptake,
possible facilitated and directed diffusion (the latter determined
comparing both types of membrane vesicles), storage within the
cells, etc., can be studied. As the survival time of the cells and
vesicles is limited and various functions decline at different
rates, the use of freshly isolated cells and vesicles is
restricted. Cultured cells can also be used, but changes in cell
characteristics (differentiation, special functions) during
cultivation must be taken into account.
3.3. Pulmonary Absorption
3.3.1. General considerations
Chemicals that are absorbed by the lungs are usually gases
(e.g., carbon monoxide, nitrogen oxides, ozone, sulfur oxides),
vapours of volatile liquids (e.g., benzene, halogenated alkanes
such as carbon tetrachloride), or aerosols (such as silica and
asbestos). The site of deposition of an aerosol depends on the
size (diameter) of the particles and their charge. Deposition in
the nasopharyngeal segment is followed by absorption through the
epithelium of this region and, after swallowing of the secreta,
through that of the gastrointestinal tract. Particles that are
deposited in the tracheal, bronchial, and bronchiolar regions are
cleared by the upward ciliary movement of the mucous layer. This
movement, normally rapid and efficient, is affected by smoking,
coughing, and sneezing. The mucous may be swallowed and the
contents absorbed in the gastrointestinal tract. Gases and
vapours, as well as very small particles, reach the alveolar zone,
where absorption takes place rapidly due in part to a very large
surface area (80 - 100 m2 in adults) and a very thin diffusion
barrier (4 µm). Particles with an aerodynamic diameter smaller
than 2 µm may not be deposited but exhaled. Depending on polarity,
gases and vapours (e.g., aldehydes, ammonia) may also be absorbed
through the mucous membranes of the upper respiratory tract, and
they may exert a considerable effect at this site of absorption.
Equilibrium is established rapidly between gases in the
alveolar air and their concentration in the blood, which depends on
the solubility of the gas or vapour in the blood; this varies
widely. The more soluble a chemical is in blood, the more must be
dissolved to reach equilibrium and the more time is required to
reach an equilibrium with body water. In the case of a gas with
low solubility, only a small percentage of the total gas in the
lung is removed by the blood during respiration and the blood
becomes saturated. Increasing respiration rate does not change the
rate of absorption, whereas increasing cardiac output markedly
enhances this parameter. However, it should be realized that
respiration rate and cardiac output generally change concomitantly.
For a gas with high solubility, almost all of the gas is
transferred to the blood during each respiration, and saturation
may not be reached. Increasing the rate of respiration will
increase the absorption rate considerably, whereas cardiac output
is of minor importance. The absorption of low-solubility gases is
thus perfusion limited and that of high-solubility gases is
respiration limited. Thus, the absorption rate of gases and
vapours with intermediate solubility is influenced by both the
respiration and cardiac output rates. These rules also hold true
for liquid aerosols. Particles are removed from the alveolar
surface through phagocytosis by the aid of macrophages. The
macrophages migrate to the distal end of the bronchiolar system and
are then removed by ciliary movement. Unphagocytozed particles are
also removed by ciliary movement. In addition, both free and
phagocytozed particles reach the lymphatic system, where they can
remain for long periods (Witschi & Brain, 1985).
3.3.2. In vivo methods
3.3.2.1. Methods for the pulmonary exposure of intact animals
Methods have been developed for the short-term and especially
the long-term exposure of intact animals to gases, vapours, or
defined aerosols. In principle, open or closed circuit systems
(the latter are much more complicated) can be used for whole-body
exposures. In such systems, the animal in the exposure chamber can
move freely for periods of hours, days, weeks, or months, with or
without exposure-free intervals.
Nose-only exposure, which avoids dermal and oral uptake and
reduces contamination, requires restriction or even anaesthesia of
the animal and can be performed for only short periods (several
hours). These exposure periods can then be repeated. Blood (and
tissue at the end of the study), urine, and faeces sampling can be
carried out at intervals (Ther, 1965; Covert & Frank, 1980; Kennedy
& Trochimowicz, 1983).
3.3.2.2. Methods for the pulmonary exposure of anaesthetized
animals
Pulmonary exposure of anaesthetized animals can be performed by
cannulation of the trachea, with or without artifical respiration,
in an open or a closed system. Absorption can be measured by
monitoring blood levels, and elimination by the sampling of urine
via a catheter in the urinary bladder. The generation and
administration of aerosols are complicated (Covert & Frank, 1980;
Kennedy & Trochimowicz, 1983), whereas exposure to distinct
concentrations of gases or vapours can be performed in an open
system without major technical problems (Ther, 1965). However, it
should be realized that such open systems involve exposure hazards.
As an alternative, any unabsorbed chemical remaining in the lungs
can be assayed after injection of small volumes of the chemical in
a vehicle 1 - 2 mm above the bifurcation of the trachea of the rat,
after killing the animal at the end of the absorption period
(Schanker, 1978). In all cases, no differentiation can be made
between absorption through the tracheobronchial tract and/or the
alveolar epithelium.
3.3.3. In vitro methods
3.3.3.1. Perfused lungs
Isolated, ventilated, and perfused lungs have been used to
investigate the tissue uptake of chemicals from the circulation
(blood, plasma, or plasma substitutes) (Schanker, 1978).
3.3.3.2. Fluid-filled lung lobes
Isolated, fluid-filled lung lobes perfused with blood, plasma,
or plasma substitutes can also be used to assess the permeability
of the alveolar epithelium and the capillary wall. The estimation
of permeability coefficients and of apparent membrane pore radii is
possible, but may differ considerably from those in normal,
ventilated organs (Schanker, 1978).
3.3.3.3. Isolated cells
Freshly isolated lung cells (Bend et al., 1985) can be used to
study of the kinetics of uptake. As with isolated liver,
intestinal, or kidney cells, the survival time is limited and
various functions decline at different rates. Up to the present,
these isolated lung cells have been used predominantly for
metabolism studies (Bend et al., 1985).
3.4. Dermal Absorption
3.4.1. General considerations
The skin is a relatively good barrier for lipid- and water-
soluble substances. Nevertheless, many substances can be absorbed
in sufficient quantities to produce acute or chronic systemic
effects (Klaassen, 1980; Scheler, 1980). For example, poisons have
been developed for use in chemical warfare and these are readily
absorbed through intact skin. Most chemicals pass through the
epidermal cells, which constitute the major part of the skin
surface. The cells of the sweat glands, the sebaceous glands, and
the hair follicles seem to be of much less importance. A chemical
must cross a large number of cells (outer layer of horny,
keratinized epidermal cells, spiny and germinal cells, corium) and
many membranes to reach the systemic circulation. This diffusion
barrier is rate limiting for overall absorption. The rate of
diffusion of nonpolar chemicals is related to their lipid
solubility and inversely related to their relative molecular mass;
polar substances may diffuse through the protein filaments. The
structure, function, and also the permeability of the skin vary
from one region of the body to another, and hydration, abrasion, or
removal of the stratum corneum distinctly enhance permeability,
sometimes by a factor of 10 or more. Damage to skin by acids,
bases, and skin irritants followed by inflammation drastically
increases skin permeability for a wide range of chemicals.
Moreover, many solvents increase permeability. Dimethyl sulfoxide
(DMSO) is one of the best known of these solvents.
3.4.2. In vivo methods
3.4.2.1. Methods for dermal exposure
Methods for dermal exposure must take into account species
differences and microlesions that appear after shaving in
furbearing species. If a definite area of skin or even the whole
animal is exposed to a chemical, the solvent used, water uptake by
skin, etc., can influence the absorption rate, much more than the
physical and chemical properties of the substance under
consideration. After various lengths of exposure, blood samples
can be assayed or the appearance of the chemical and of its
metabolites can be monitored in urine. The skin can also be
analysed after washing and cleaning. Precautions should be taken
to avoid inhalation exposure.
3.4.3. In vitro methods
3.4.3.1. Isolated skin preparations
Isolated skin preparations of adult animals, with or without
fur (e.g., nude mice, pigs), as well as of newborn animals (e.g.,
rats and mice without fur), can be used as a diaphragm in a
diffusion chamber. The rate of diffusion can be determined without
the influence of perfusion. Again, the influence of water and
solvents may play the most important role.
3.4.3.2. Different cell populations
Different cell populations can also be obtained from skin and
can be used for the study of absorption (i.e., uptake processes).
In the epidermis, cell to cell connections play a much greater role
than those in organs such as liver and kidney. Thus, possible
artefacts must be taken into account.
3.5. Other Routes of Exposure
3.5.1. General considerations
In general, chemicals enter the interior of the body from the
exterior by crossing the epithelial barriers of the skin, lung, or
gastrointestinal tract. Penetration via a lesion (e.g., wound) is
the exception. A parenteral route is often chosen to study the
action of a chemical in order to control dose, concentration, etc.,
and to avoid uncertainties of bioavailability (Scheler, 1980).
3.5.2. The intravenous (iv) route
The intravenous (iv) route introduces the chemical in solution
directly into the blood stream, avoiding the process of absorption.
3.5.3. Intraperitoneal (ip) administration
In general, after intraperitoneal (ip) administration of a
chemical, absorption is facilitated by the large surface of the
peritoneal cavity. The chemical mainly enters the liver by the
portal circulation; thus, first pass effects must be considered.
3.5.4. Intramuscular (im) administration
In general, the chemical is readily absorbed after
intramuscular (im) administration, because of the good perfusion of
muscular tissue, but it must also pass several membranes.
3.5.5. Subcutaneous (sc) administration
After subcutaneous (sc) administration, absorption is
relatively slow. Changes in the perfusion by vasoactive compounds
(e.g., vasoconstriction by sympathomimetics, vasodilatation by
local anaesthetics) as well as peculiarities in the formulation
(solvent, microcrystals, etc.) can strongly influence the rate of
absorption.
4. DISTRIBUTION
4.1. General Considerations
Distribution is the process(es) by which an absorbed substance
and/or its metabolites circulate and partition within the body.
Two approaches can be used for the analysis of distribution
patterns:
(a) the qualitative approach using information obtained
by whole-body autoradiographic techniques; and
(b) the quantitative approach using information obtained
by sacrificing animals at different times after
exposure and determining the concentration, and
amount of, the test substance and/or metabolites in
tissues and organs (Klaassen, 1980; OECD, 1981).
The distribution of a chemical in the blood usually occurs
rapidly. Distribution to the different tissues and organs is
determined by the blood flow through the capillary walls, and
interstitial and cell barriers, by the concentration gradient of
the free, unbound chemical, and by the affinity to binding sites in
the tissues and organs. Thus, all considerations on diffusion
given in section 3, are also valid for distribution.
The concentration of a chemical in blood (blood level) is
dependent on the dose, the rate of elimination, and the volume of
distribution (section 8.2.1.1). The greater the volume of
distribution, the lower the blood level. But, as the chemical may
be found in various tissues and organs in different concentrations
according to its hydro- and lipophilicity, this distribution volume
is a fictitious space. There are very few chemicals that are
distributed in plasma (e.g., trypan blue, indocyanine green),
plasma plus interstitial = extracellular water (chloride, inulin,
thiosulfate), or in extracellular plus intracellular = total body
water (antipyrine). Chemicals with high lipophilicity occur in
much higher concentrations in fat and many are accumulated in the
liver and kidney. The apparent volume of distribution is an
estimation of the magnitude of the distribution volume, based on
the blood level, if the chemical were distributed equally in the
body after one single exposure. Only the circulatory system is a
distinct, closed "compartment" where chemicals are distributed
rapidly. Distribution to the various tissues and organs is usually
markedly delayed; this second type of distribution volume is termed
a "peripheral compartment" and there is an equilibrium of the free
chemical between the so-called rapid, or central, and the slow or
peripheral compartment. As the free chemical is eliminated (Fig.
1), the chemical from the peripheral compartment is slowly released
into the circulation (rapid or central compartment).
In contrast to the fictitious volumes of distribution and
compartments, some organs and tissues are morphologically defined
compartments due to special diffusion barriers or to storage
capacities. Many chemicals do not enter the central nervous system
(CNS) as readily as other organs and tissues, because there are
special diffusion barriers: the capillary endothelial cells of the
CNS have very tight junctions with very few, if any, pores, and the
capillaries are surrounded by specialized glial cells termed
astrocytes. This so-called blood-brain barrier is not absolute,
but it does exclude many chemicals. The characteristics of the
blood-brain barrier vary in different brain areas. Chemicals with
high lipid solubility readily enter the CNS, but hydrophilic
compounds (e.g., ions) are almost totally excluded. Similarly, an
efficient blood-testicular barrier exists. In pregnancy, the
embryo and fetus are readily reached by most chemicals, which pass
the placenta by simple diffusion. The number of layers that
represent a diffusion barrier in the placenta varies with the
species and within one species with the state of gestation.
Usually the more lipid-soluble chemicals will cross the placenta
more readily and an equilibrium between the concentrations of the
free chemical in maternal and fetal compartments (e.g., maternal
and fetal plasma water) will be reached sooner or later. In
general, the placental-fetal unit belongs to the peripheral
compartment of the mother.
4.2. Invasive Methods
4.2.1. Qualitative methods
4.2.1.1. Autoradiographic methods
Autoradiographic methods have been developed to study the
distribution of chemicals after labelling with radioactive isotopes
(3H, 14C, 35S, etc.) with alpha-, beta-, and gamma-radiation
(Amlacher, 1974; Rogers, 1979, 1985; Nagata, 1984), in the whole
body of experimental animals. By killing the animals at different
times after administration or by taking biopsies, the time curves
of distribution in the various tissues and organs as well as the
distribution patterns at equilibrium can be obtained.
Autoradiographic methods can also be used for distribution studies
at the tissue and cell levels.
4.2.2. Quantitative methods
4.2.2.1. Radiometric methods
After administration of a radiolabelled chemical, blood and
tissue samples are taken at intervals, either after killing the
animals or by obtaining biopsies, and the activities determined by
different techniques, depending on the radiolabel and radiation
type. In this way, time distribution curves as well as
distribution patterns can be obtained.
4.2.2.2. Chemical methods
If sensitive and specific chemical detection methods are
available (section 2), distribution-time curves and patterns can be
studied after killing the animals or by taking biopsies at
different intervals after administration.
4.3. Non-Invasive Methods
Non-invasive methods are desirable in studying the distribution
of chemicals in expensive non-rodents, and especially in man.
Limited information can be obtained through the detection of
chemicals and their metabolites in saliva, breath, and urine (using
radiometric, chemical, or stable isotope methods), as well as
through whole-body scanning, after administration of radiolabelled
substances (Breimer & Danhof, 1980; Posti, 1982; Walther et al.,
1983; Zylber-Katz et al., 1984; Kretschko & Berg, 1985; Krumbiegel
et al., 1985a,b). However, developments in this field, such as
positron imaging, X-ray fluorescence, neutron activation, and
magnetic pneumography, deserve special attention. They include
positron imaging, X-ray fluorescence, neutron activation, and
magnetic pneumography.
5. BINDING
5.1. General Considerations
Many proteins including albumin, glycoproteins, lipoproteins,
and specialized proteins for metals, bind chemicals (Davison, 1971;
Vallner, 1977). Plasma albumin, which binds the greatest variety
of foreign chemicals, i.e., acidic, basic, and neutral substances,
has been thoroughly investigated. The main binding force is
hydrophobic bonding, but other reversible bonds such as hydrogen,
van der Waal's, and ionic bonds are also involved. As hydrophobic
bonding plays the most important role, the extent of binding is
directly correlated with lipophilicity. The high relative
molecular mass of albumin (about 68 000) prevents the bound
chemical from crossing the capillary walls, and this fraction is
not directly available for extravascular distribution. But, as
unbound (free) compound leaves the vessels by diffusion, bound
compound dissociates from the protein and the relative degree of
dissociation remains constant. This process continues until the
free chemical in the extravascular water equilibrates with the
free chemical in the plasma-water. Plasma-albumin binding of
chemicals does not influence active processes such as carrier-
mediated transport in the tubular cell in the kidney.
Depending on the structure and character of a chemical (acidic,
basic, neutral), different parts of the albumin molecule serve as
the binding sites; the binding capacity (number of binding sites)
is limited and therefore saturable. Consequently, competition
between chemicals for the same binding site(s) occur. The
competing chemicals are generally similar in structure, but not
always, and competition phenomena depend on the concentrations and
the affinities of the particular compounds. Competitive release
has been observed with certain drugs, the sudden high concentration
of the free drug leading to potent pharmacological or toxic
responses.
Protein binding of chemicals is also observed in tissues or
organs. The liver and the kidneys have a high capacity to bind
certain chemicals. In both organs, substances are bound to so-
called carriers, responsible for active transport, and also to
specific binding proteins. Special binding proteins for organic
anions (glutathione transferases) and for metals (metallothioneins)
have been described. Different types of binding are observed in
bone with toxic metals (lead, strontium), halogens (fluoride) and
other compounds (e.g., tetracyclines). Certain metals can replace
calcium cations and fluoride and tetracyclines can replace
hydroxylanions in the hydroxyapatite lattice structure of bone by
an exchange adsorption reaction, a typical surface phenomenon. The
chemicals diffuse into the hydration shell of the hydroxyapatite
crystals, which have a large surface area. Fluoride and
radioactive metals may cause damage at the site of storage, others
are not toxic to the bone, but may serve as a depot. Release by
ionic exchange, pH changes, or osteoclastic activity can lead to an
increased concentration in plasma resulting in toxic reactions, if
enough chemical is mobilized.
5.2. Methods for Assessing Reversible Binding
5.2.1. Extracellular sites
The binding of chemicals to serum, plasma, or albumin (crude or
purified fractions) is estimated by various techniques including
ultrafiltration techniques, (Sephadex)-gel filtration, or
equilibrium dialysis (Davison, 1971; Vallner, 1977). Assay of the
chemicals can be accomplished by various analytical methods
(section 2); radiometric methods are used when radiolabelled
chemicals are studied.
The following physical methods for studying chemical-protein
binding can be carried out only in very special cases:
(a) ultraviolet and visible absorption spectroscopy of
free and bound chemical;
(b) fluorescence spectroscopy;
(c) optical rotatory dispersion and dichroism; or
(d) nuclear magnetic resonance.
5.2.2. Intracellular sites
The same methods used for extracellular sites, especially
ultrafiltration techniques, can be used in the study of reversible
binding to cell fractions (nuclei, nuclear membranes, mitochondria,
smooth and rough endoplasmic reticulum, lysosomes, cytosol, special
membranes) and others such as synaptosomes.
5.3. Methods for Assessing Irreversible Binding
Irreversible binding of chemicals and especially of reactive
metabolites to macromolecules (DNA, RNA, proteins, and lipids)
plays an important role and is studied predominantly using
radiolabelled chemicals (Pohl & Branchflower, 1981). The term
"irreversible binding" (or "covalent binding") is used when the
bound proportion does not change after isolation, dilution, and
washing procedures; frequently this binding is also stable during
denaturation. Only recently have other techniques, such as mass
spectroscopy, X-ray crystallography, and immunochemical analysis,
been applied to assess the nature of the irreversible binding
(Berlin et al., 1984a).
6. METABOLISM
6.1. General Considerations
In this section, metabolism refers to the process or processes
by which an administered xenobiotic chemical is structurally
altered in the body by either enzymatic or nonenzymatic reactions.
In this context, the terms biotransformation and metabolic
transformation are used interchangeably with metabolism.
Xenobiotic denotes a relatively small (relative molecular mass <
1000), non-nutrient chemical that is foreign to the species in
which biotransformation is being studied, though certain compounds
biosynthesized by some species (e.g., alkaloids, glycosides) are
xenobiotics in others.
The major role of biotransformation is to convert poorly
excretable lipophilic compounds to more polar entities that can be
readily excreted in the urine and/or the bile. In the absence of
metabolism, such xenobiotics accumulate in the mammalian body,
increasing the potential for a toxic response. Examples of such
compounds are certain polychlorinated biphenyl (PCB) and
polychlorinated dibenzofuran (PCDF) congeners (Masuda et al.,
1985). On the other hand, biotransformation is less likely in
xenobiotics that have high water/oil partition ratios (hydrophilic
compounds), which are rapidly excreted in urine.
Two or more sequential enzymatic reactions are routinely
required to convert lipophilic xenobiotics to metabolites that are
efficiently excreted. Williams (1959) classified the pathways
involved into phase I and phase II reactions. Oxidation,
reduction, and hydrolysis are termed phase I reactions, whereas
conjugation and synthesis are phase II reactions. Normally, one or
more phase I reactions precede phase II metabolism. Initially,
xenobiotic metabolism was associated with detoxication. However,
it is now known that both phase I and phase II reactions function
in metabolic activation processes as well. Many different types of
compounds are converted to their ultimate toxic chemical species
during metabolism; a few of the best studied examples include
acetaminophen, 2-acetylaminofluorene, aflatoxin B1, benzo( a)pyrene,
carbon tetrachloride, diethylnitrosamine, dimethylnitrosamine, and
4-ipomeanol.
Historically, most in vitro studies of xenobiotic metabolism
were conducted on mammalian liver, because this organ generally
contains a high concentration of biotransformation enzymes, is
relatively large, and consists of few cell types. Compared with
the liver, extrahepatic tissues do not normally play a major
quantitative role in the biotransformation of foreign compounds,
though there are exceptions. Interest in the xenobiotic-
metabolizing enzymes of extrahepatic tissues has increased
markedly during the last decade, largely in an attempt to
understand the relationships between metabolism and target organ or
cell toxicity (Gram, 1980; Rydström et al., 1983; Bend & Serabjit-
Singh, 1984). It is now known that, for certain compounds, the
relative ability of a tissue or cell type to metabolically activate
a chemical versus the ability of the same tissue or cell type to
detoxify the reactive/toxic metabolite(s) can determine the
severity and location of the lesion. The best example of this
phenomenon is the lung-specific toxin 4-ipomeanol, which is
biotransformed to a metabolite by the cytochrome P-450 (P-450)
monooxygenase (EC 1.14.14.1) system. The metabolite formed causes
highly selective necrosis of the nonciliated bronchiolar
epithelial (Clara) cells of several species (Boyd, 1980).
Endoplasmic reticulum and P-450-dependent enzyme activity are also
concentrated in Clara cells (Devereux et al., 1981, 1982) compared
with other lung cells such as the alveolar type II cell and the
alveolar macrophage. Thus, the susceptibility of Clara cells to
4-ipomeanol is related to their ability to rapidly oxidize it to a
toxic metabolite. It may also be related to their inability to
maintain intracellular glutathione (GSH) at a concentration
sufficient to detoxify the electrophilic metabolite formed. For a
recent review of the metabolic activation of xenobiotics to toxic
metabolites, see Guengerich & Liebler (1985).
The contribution of the intestinal flora to the in vivo
metabolism and toxicity of xenobiotics should not be forgotten
(Scheline, 1980; Goldman, 1982), and can be especially important
for chemicals that undergo enterohepatic circulation.
The remainder of this section will deal with the enzymes known
to be important in xenobiotic metabolism, their modulation by
physiological, environmental, and pathological factors, non-
invasive methods for studying xenobiotic metabolism in vivo, and
in vitro preparations that are used to study organ-specific and
cell-specific biotransformation.
6.2. Important Enzymatic Pathways in Xenobiotic Metabolism
6.2.1. Phase I reactions
6.2.1.1. Oxidation reactions
Oxidation is the first step in the metabolism of most
xenobiotics, and there are several different enzyme systems that
oxidize chemicals in mammals. The most important pathways are
discussed here.
6.2.1.1.1. Cytochrome P-450 monooxygenase system (EC 1.14.14.1)
This is the most important enzyme system involved in the phase
I metabolism of xenobiotics, primarily because of its great
versatility (for reviews, see Sato & Omura, 1978; Johnson, 1979;
Coon & Persson, 1980; Lu & West, 1980; Wislocki et al., 1980; Wolf,
1982; Estabrook, 1984; Johnson et al., 1985; Levin et al., 1985).
In addition to most xenobiotics, it metabolizes several classes of
endogenous compounds, including fatty acids, prostaglandins,
steroids, and vitamins. The P-450 system is membrane bound and has
two major components, cytochrome P-450, a haemoprotein, and NADPH-
cytochrome P-450 reductase, a flavoprotein that contains both FMN
and FAD prosthetic groups. This enzyme also reduces cytochrome c
and is known as NADPH cytochrome c reductase; its role in
xenobiotic metabolism has been reviewed by Masters (1980). A major
reason for the wide variety of substrates that are oxidized by this
system is that there are many forms or isozymes of P-450. These
differ in substrate specificity, as well as in their response to
the administration of enzyme inducers, such as phenobarbital (PB)
or 3-methylcholanthrene (3-MC). Detailed protein purification and
characterization studies have shown that there are at least 10
distinct P-450 proteins in rat liver (Guengerich et al., 1982;
Levin et al., 1985). This is not the only reason for substrate
diversity, however, for single purified P-450 isozymes in
reconstituted monooxygenase systems catalyse many different
reactions.
The overall oxidation of a substrate, RH, can be summarized by
the following equation (1), where NADPH, one reduced form of
nicotinamide adenine dinucleotide, is shown as the required
cofactor:
RH + O2 + NADP + H+ + ----> ROH + H2O + NADP+ (1)
The sequence of events that occurs during a monooxygenase
reaction is now well understood (Estabrook, 1984). P-450 contains
one molecule of iron-protoporphyrin IX as its prosthetic group.
Normally, in microsomal preparations, this iron is in the ferric
(Fe+3) state. Oxidized P-450 first reacts with a molecule of
substrate to form an enzyme-substrate complex. Next, one electron
is donated to this complex from NADPH via NADPH-cytochrome P-450
reductase, which converts RH-P-450(Fe+3) to RH-P-450(Fe+2). The
P-450(Fe+2)-substrate complex then reacts with molecular oxygen to
form an oxycytochrome P-450 ternary complex. This complex may then
accept one additional electron from NADPH via NADPH-P-450
reductase, or from NADH by cytochrome b5 reductase (EC 1.6.2.2) to
form the equivalent of a two-electron reduced complex of
haemoprotein, oxygen, and substrate, which dissociates to yield
oxidized substrate, P-450(Fe+3) and water. The reaction is termed a
monooxygenation because one atom of atmospheric oxygen is
transferred to the substrate; the other is incorporated into water.
A simplified reaction scheme is given in Fig. 2.
Some of the reactions catalysed by the P-450 monooxygenase
system include aliphatic hydroxylation, epoxidation, aromatic
hydroxylation, heteroatom ( N-, O-, S-) dealkylation, oxidative
deamination, nitrogen oxidation, oxidative desulfuration, oxidative
dehalogenation, and oxidative denitrification (Wislocki et al.,
1980).
There are 2 types of P-450-dependent monooxygenase systems in
mammals. The most important one for xenobiotic metabolism, and the
one described above, is associated with the endoplasmic reticulum
(microsomal fraction) of the liver and many extrahepatic tissues
(lung, kidney, placenta, small intestine, skin, adrenal, testis,
ovary, eye, pancreas, mammary gland, aorta walls, brain, nasal
epithelial membranes, colon, salivary glands, prostate, heart,
lymph nodes, spleen, thymus, and thyroid) (Gram, 1980; Bend &
Serabjit-Singh, 1984). The other P-450 monooxygenase system is
associated with the mitochondria of steroid-metabolizing tissues
(adrenal, ovary, testis) and contains an FAD-flavoprotein and an
iron sulfur protein that facilitate electron transfer from NADPH to
P-450 (Estabrook et al., 1973). The mitochondrial P-450 system is
normally involved with the metabolism of endogenous compounds
including cholesterol, cholecalciferol, and deoxycorticosterone,
and has a much higher degree of substrate specificity than the
microsomal system (Estabrook, 1984).
Although the P-450 system is concentrated in the endoplasmic
reticulum of hepatocytes, it is also present at lower
concentrations in the nuclear membrane, in the Golgi apparatus, and
in the plasma membrane (Stasiecki & Oesch, 1980). Presumably, this
is also true for extrahepatic tissues.
There are several parameters associated with the microsomal
P-450 system that can be readily assayed. The total content of
microsomal P-450, and of the related microsomal haemoprotein
cytochrome b5, can be measured by the method of Omura & Sato
(1964a,b). P-450 is quantified from the difference spectrum
generated from carbon monoxide-saturated and reduced versus reduced
microsomal preparations and b5 from the difference spectrum between
oxidized and reduced microsomes. However, it must be remembered
that, because of the diversity of P-450 isozymes present in
microsomes, changes in total P-450 content seldom correlate with
changes in monooxygenase activity.
Microsomal NADPH-P-450 reductase activity can also be measured
(Gigon et al., 1969), though it is much easier to quantify the
enzyme activity as NADPH-cytochrome c reductase activity (Masters,
1980).
Studies of the biotransformation of a xenobiotic, and the rate
of this process, using the P-450 system in microsomal or 10 000 x g
supernatant preparations of liver and/or a variety of extrahepatic
tissues (Burke & Orrenius, 1979), can be an important step in
evaluating the toxicity of a chemical. In general, it is best to
perform initial studies with hepatic microsomes (or 10 000 x g
supernatant preparations), in which the P-450 is present in the
highest concentration. However, since the different isozymes of
P-450 are not distributed equally in different tissues and cells,
the ratio of oxidative detoxication and toxication reactions can
vary from tissue to tissue. Thus, in some cases, it will be more
appropriate to study a toxication pathway in microsomes from an
extrahepatic tissue. In studies of this type, it is necessary to
buffer (to approximately pH 7.4) the microsomal preparation
(105 000 x g pellet) and to add NADPH or an NADPH-generating system
(e.g., NADP, glucose-6-phosphate, and glucose-6-phosphate
dehydrogenase), as cofactor, prior to incubation. Since an
individual xenobiotic is frequently oxidized by several different
pathways, it may be necessary to separate and to chemically
identify the various metabolites formed in order to understand the
nature of the toxication pathway(s) (section 2).
6.2.1.1.2. Microsomal flavin-containing monooxygenase (EC 1.14.13.8)
There is also a P-450-independent monooxygenase localized in
the endoplasmic reticulum of mammalian liver and extra-hepatic
tissues, which can be detected in virtually all nucleated cells.
This flavin-containing monooxygenase (FMO) was originally isolated
from pig liver (Ziegler & Mitchell, 1972) and has recently been
purified from rat, mouse, and rabbit liver (Kimura et al., 1983;
Tynes et al., 1985) and mouse and rabbit lung (Williams et al.,
1984b; Tynes et al., 1985). The enzyme contains the coenzyme
FAD and requires NADPH as a cofactor; its purification,
characterization, kinetic mechanism, and catalytic properties have
been reviewed by Ziegler (1980, 1984) and Poulsen (1981). The
physiological substrate for the enzyme is believed to be
cysteamine, which is oxidized to cystamine.
Although the FMO does not catalyse oxidation reactions at
carbon as does the P-450 system, it can oxidize certain nitrogen-
containing (Ziegler, 1984), sulfur- and selenium-containing
(Poulsen, 1981; Ziegler, 1984), and phosphorus-containing
xenobiotics (Smyser & Hodgson, 1985). Tertiary amines (e.g.,
dimethylaniline, chlorpromazine, imipramine) are oxidized to
fairly stable amine oxides by this enzyme and almost all
N,N-disubstituted alkylamines and arylamines that do not have
negatively charged functional groups are substrates (equation 2).
FMO
R3N -----> R3N+ -O- (2)
Secondary amines, such as N-benzylamphetamine and
desimipramine, are rapidly N-oxygenated through this enzyme. The
initial oxidation product is the corresponding hydroxylamine, which
is further oxidized to a nitrone (equation 3).
FMO FMO
RCH2NHCH2R -----> RCH2N(OH)CH2R -----> RCH=N+ (O-)CH2R (3)
Aliphatic nitrones readily decompose in aqueous solution to
produce aldehydes and primary hydroxylamines.
Only primary amines that readily form imine tautomers, such as
2-naphthylamine, are substrates for the FMO. Consequently, the
N-oxygenation of most primary xenobiotic aromatic amines is
catalysed by the P-450 system.
Other nitrogen-containing substrates for the FMO are
1,2-disubstituted hydrazines and N-substituted aziridines (Prough,
1973).
Since there is a large number of drugs and environmental
pollutants that contain sulfur, it is of considerable interest that
FMO preferentially catalyses the oxidation of sulfur atoms in
compounds that contain both sulfur and nitrogen (Poulsen, 1981).
FMO is apparently a more universal sulfur oxidase than P-450.
Thus, it catalyses the oxidation of: alkyl and aryl thiols, free of
anionic groups initially, to the disulfide; alkyl and aryl
disulfides to sulfinic acids; cyclic disulfides to sulfoxides;
thiocarbamides or thioureas to formamidine sulfonic acids, through
intermediate sulfenic and sulfinic acids; and thioether containing
organophosphates or carbamates to their corresponding sulfoxides
(Poulsen, 1981; Ziegler, 1984; Tynes & Hodgson, 1985). The
oxidation of thiocyanates, carbodithoic acids, and dithiocarbamates
is also catalysed by this enzyme.
It is apparent that FMO is an important enzyme for the
metabolism of several classes of xenobiotics. However, the
relative importance of FMO in toxication/detoxication is still
being evaluated. Because some compounds are susbstrates for both
the FMO and the P-450 monooxygenases, it may be necessary to
determine which one of these pathways (or both) is responsible for
the oxidative metabolism of a given toxicant. Fortunately,
several methods are now available to distinguish between the two
systems. For example, only the P-450 system is induced following
the administration of compounds such as PB or 3-MC, P-450-dependent
activity can be selectively inhibited by carbon monoxide, by
lipophilic primary alkylamines, or by specific antibodies to
NADPH-P-450 reductase (Ziegler, 1984; Tynes & Hodgson, 1985).
Trypsin proteolysis of mouse liver microsomes, under appropriate
conditions, also removes NADPH-P-450 reductase activity leaving FMO
activity intact (Tynes & Hodgson, 1985).
Not much is known about the number of different forms or
isozymes of FMO and their endogenous regulation. However, recent
reports have demonstrated the presence of the enzyme in the lung of
rabbit and mouse in forms that are immuno-chemically and
catalytically different from those in the liver (Williams et al.,
1984b; Tynes et al., 1985). Moreover, the rabbit pulmonary isozyme
is induced during pregnancy, whereas that in liver is not (Williams
et al., 1984b).
It is worth reemphasizing that P-450 isozymes catalyse several
reactions not catalysed by FMO. These include aliphatic
hydroxylation, aromatic hydroxylation, epoxide formation, and
N-, O-, and S-dealkylation reactions that occur via rearrangement
of initial alpha-carbon hydroxylation products (i.e., FMO does not
catalyse oxidation at carbon).
6.2.1.1.3. Cooxidation by prostaglandin H synthase (EC 1.14.99.1)
Marnett et al. (1975) originally demonstrated, in preparations
of sheep seminal vesicles, that several organic chemicals,
including benzo( a)pyrene, are oxidized during the prostaglandin H
synthase-catalysed conversion of arachidonic acid to
prostaglandins. Over the last 5 years, many in vitro studies have
focused on this metabolic pathway because of its potential
toxicological importance. This reaction converts certain
xenobiotics to electrophilic metabolites (Marnett, 1981, Marnett &
Eling, 1983; Krauss & Eling, 1984). Prostaglandin H synthase
catalyses 2 distinct enzymatic reactions; its cyclooxygenase
activity converts arachidonic acid to prostaglandin (PG)G2, a
hydroperoxyendoperoxide, and its peroxidase activity reduces (PG)G2
to (PG)H2, a hydroxyendoperoxide. Xenobiotic oxidation is
catalysed by the hydroperoxidase of prostaglandin H synthase, and
the reaction is called cooxidation (Marnett, 1981; Marnett & Eling,
1983). The reduction of (PG)G2 by prostaglandin H synthase
requires the donation of single electrons and these can come from
the substrate that is cooxidized, though this is not always the
case (Marnett & Eling, 1983). Many xenobiotics that mediate toxic
responses, including acetaminophen, 2-aminofluorene, 2-amino-4-(5-
nitrofuryl)thiazole, diethylstilbestrol (DES), benzo( a)pyrene
7,8-dihydrodiol, 7,8-dihydrobenzo( a)pyrene, and 4-phenetidine, are
known substrates for this reaction, and 2-aminofluorene and benzo
( a)pyrene 7,8-dihydrodiol are converted to potent mutagens by
it (Krauss & Eling, 1984).
Prostaglandin H synthase activity is high at several
extrahepatic sites that are low in P-450 monooxygenase activity.
These include skin, kidney medulla, lung of certain species,
platelets, and the endothelial cells lining blood vessels. Thus,
it is possible that prostaglandin H synthase complements and/or
serves as an alternative to, the P-450 monooxygenases for the
metabolic activation of certain carcinogens; the relative
importance of these 2 pathways for the formation of ultimate
carcinogens in vivo is still not known. However, this one electron
oxidation pathway cannot be ignored when reactions for the
formation of toxic xenobiotics are being considered.
Ram seminal vesicles are very rich in prostaglandin H synthase
and buffered incubation mixtures containing substrate, arachidonic
acid, and ram seminal vesicle microsomes are excellent for testing
for prostaglandin H synthase-dependent cooxidation (Marnett &
Eling, 1983; Krauss & Eling, 1984).
6.2.1.1.4. Miscellaneous peroxidative pathways
Phenols and arylamines are excellent substrates for peroxidase-
catalysed one electron oxidation, and complex product mixtures
often result from the interactions of the free radicals produced.
A few examples of the role of peroxidases (EC 1.11.1.7) in toxicity
are described below.
Certain arylamines cause methaemoglobinaemia, which is
currently believed to be due to their oxidation by oxyhaemoglobin
in the erythrocyte (Eyer, 1983). Haemoglobin in erythrocytes has
also been shown to activate styrene to metabolites that cause
s