Acute Toxicity Testing In Vitro and the Classification and Labelling of Chemicals

The Report and Recommendations of ECVAM Workshop 161,2

Reprinted with minor amendments from ATLA 24, 499-510

Hasso Seibert3, Michael Balls4, Julia H. Fentem4, Vera Bianchi5, Richard H. Clothier6, Paul J. Dierickx7, Björn Ekwall8, Michael J. Garle6, Maria José Gómez-Lechón9, Laura Gribaldo4, Michael Gülden3, Manfred Liebsch10, Eva Rasmussen11, Roland Roguet12, Ravi Shrivastava13 and Erik Walum14
3Institut für Toxikologie, Christian-Albrechts Universität, Weimarer Str. 8 Haus 3, 24106 Kiel, Germany; 4ECVAM, JRC Environment Institute, 21020 Ispra (Va), Italy; 5Department of Biology, University of Padova, via Trieste 75, 35121 Padova, Italy; 6Department of Human Morphology, University of Nottingham Medical School, Nottingham NG7 2UH, UK; 7Division of Toxicology, Institute of Hygiene & Epidemiology, J. Wytsmanstraat 14, 1050 Brussels, Belgium; 8Department of Pharmaceutical Biosciences, Division of Toxicology, Uppsala University, 75124 Uppsala, Sweden; 9Unidad de Hepatologia Experimental, Centro de Investigacion, Hospital Universitario La Fe, Avda de Campanar 21, 46009 Valencia, Spain; 10ZEBET, Bundesinstitut für gesundheitlichen Verbraucherschutz und Veterinär-medizin (BgVV), Diedersdorfer Weg 1, 12277 Berlin, Germany; 11Institute of Toxicology, Danish National Food Agency, 19 Morkhoj Bygade, 2860 Soborg, Denmark; 12Central Department of Product Safety, Recherche Avancée, L'Oréal, 93601 Aulnay-sous-Bois, France; 13VITRO-BIO, Biopôle, Clermont-Limagne, 63360 Saint Beauzire, France; 14Pharmacia AB, Biopharmaceuticals, 112 87 Stockholm, Sweden

1ECVAM - The European Centre for the Validation of Alternative Methods. 2This document represents the agreed report of the participants as individual scientists.

Address for correspondence: Professor Michael Balls, ECVAM, TP 580, JRC Environment Institute, 21020 Ispra (Va), Italy.

Address for reprints: ECVAM, TP 580, JRC Environment Institute, 21020 Ispra (VA), Italy

Preface

This is the report of the sixteenth of a series of workshops organised by the European Centre for the Validation of Alternative Methods (ECVAM). ECVAM's main goal, as defined in 1993 by its Scientific Advisory Committee, is to promote the scientific and regulatory acceptance of alternative methods which are of importance to the biosciences and which reduce, refine or replace the use of laboratory animals. One of the first priorities set by ECVAM was the implementation of procedures which would enable it to become well-informed about the state-of-the-art of non-animal test development and validation, and the potential for the possible incorporation of alternative tests into regulatory procedures. It was decided that this would be best achieved by the organisation of ECVAM workshops on specific topics, at which small groups of invited experts would review the current status of various types of in vitro tests and their potential uses, and make recommendations about the best ways forward (1).

The workshop on Acute Toxicity Testing In Vitro was held in Angera, Italy, on 18-22 April 1994, under the co-chairmanship of Hasso Seibert (Christian-Albrechts Universität, Kiel, Germany) and Michael Balls (ECVAM, Ispra, Italy).

Introduction

Acute systemic toxicity testing involves assessment of the general toxic effects of a single dose of a chemical or product, or, in some cases, the effects of multiple doses given within 24 hours. The test material can be administered by various routes (for example, orally, by inhalation or by injection). Acute systemic tests include lethal dose tests (for example, the classical and modified LD50 tests [2, 3], the acute toxic class method [4-6], and up-and-down methods [7, 8]) and non-lethal tests (for example, the fixed dose procedure [9, 10]).

Of the 2,842,361 laboratory animals used in Great Britain in 1994, 473,064 (16.6%) were used in toxicity testing, of which nearly half (232,696) were used in acute and sub-acute whole body dose-ranging tests, lethal toxicity tests, or clinical sign toxicity tests (11). The animals used were mainly mice (77,542), rats (72,380) or fish (42,768).

The most important general use of acute systemic toxicity test data is as a basis for classifying and labelling chemicals in relation to their manufacture, transport and use. Acute systemic toxicity test data are also used in a number of other ways; for example, for defining modes of toxic action, setting dose levels for other tests, and for providing medical advice in cases of poisoning (based on the toxic effects observed). For the latter purposes, an acute systemic tolerance study can be used instead of an LD50 test.

The use of lethal dose tests has been widely criticised, not least by toxicologists themselves, and much effort has been put into the development of alternative methods which could reduce the numbers of animals required, reduce the suffering caused to any animals necessarily used, or replace animal procedures altogether (8, 12). The question of whether, and if so, to what extent, non-animal tests or testing strategies could significantly reduce or even replace the need for in vivo acute systemic toxicity tests is still unresolved. This topic has been the subject of many review papers (for example, 13-25).

Many studies have shown good correlations between in vitro cytotoxicity data obtained with undifferentiated cell lines and LD50 data (for example, 13, 14, 16, 17, 23, 25). However, acute systemic toxicity can be caused by a variety of mechanisms, many of which would not be operative in such basal cytotoxicity tests, so more sophisticated approaches are also necessary. The problem was put in this way by William Parish, while chairing a discussion in the acute oral toxicity testing section of a meeting held to consider the Second Report of the FRAME (Fund for the Replacement of Animals in Medical Experiments) Toxicity Committee (26): "When you have a chemical which is generally cytotoxic, like sodium hydroxide, then some sort of in vitro test is feasible. But, when you are dealing with substances which require very specific receptors, or which are absorbed internally and modified to become toxic metabolites, you cannot use the general tissue culture system. When you are given such a substance, you will not know how to test it." Correlations between sets of in vitro and in vivo data, however good they may be, cannot be used as a basis for deciding on the relevance and reliability of an in vitro test result for one new chemical for which in vivo data are not available.

One of the main purposes of the workshop was to face up to the difficulties raised by Dr Parish. This required a realistic reappraisal of what had been achieved with regard to the development of in vitro tests for cytotoxicity. It is apparent that a very large number of tests for basal cytotoxicity have been developed, but that insufficient attention has been paid to how the data they provide can be applied, particularly as a means of making decisions or predictions. In addition, more account needs to be taken of biokinetic factors, which affect the interactions between cells and test materials, and which influence attempts to interpret the test data in relation to potential toxic effects in vivo.

Types of Cytotoxicity

Chemicals can have three main types of toxic effect at the cellular level, if cytotoxicity is defined as the adverse effects of interference with structures and/or properties essential for cell survival, proliferation and/or function. These effects can involve, for example, the integrity of membranes and the cytoskeleton, metabolism, the synthesis and degradation or release of cellular constituents or products, ion regulation, and cell division. General (basal) cytotoxicity involves one or more of the above mentioned structures or processes, when all of the cell types studied show similar sensitivities. Selective cytotoxicity occurs where some types of differentiated cells are more sensitive to the effects of a particular toxicant than others, for example, as a result of biotransformation, binding to specific receptors, or uptake by specific mechanisms. Cell-specific function toxicity occurs when the toxicant affects structures or processes which may not be critical for the affected cells themselves, but which are critical for the organism as a whole. For example, such toxicity can involve effects on cell-cell communication, via the synthesis, release, binding and degradation of cytokines, hormones and transmitters, or on specific transport processes.

All three types of effect can result in acute systemic toxicity in vivo. In addition, toxicity may result from chemicals interfering with extracellular processes. Any non-animal test or testing strategy must take all of these possibilities into account. The development of such test systems would be considered by some to be an unrealistic ambition, because a very wide variety of cell types and endpoint measurements would be required. Another view is that the design of manageable test systems capable of dealing with the main types of chemicals and mechanisms of toxicity might be possible. This was the basis for the discussions which took place at the workshop.

A prerequisite to the development of any toxicity test system must be an appreciation of the concepts of active concentration and of meaningful dose. It must not be assumed that what is added to the culture medium will be available to the cells in a simple concentration-related manner. Nor must it be forgotten that almost all test materials will exert cytotoxic effects when added at sufficiently high concentrations. Thus, it is essential in any scheme that a quantitative comparison be made of the concentrations of test materials which exert basal cytotoxic effects versus selectively cytotoxic or cell-specific function effects.

Cytotoxicity Testing Schemes

The participants in the workshop considered a number of evaluation schemes and testing strategies which had been proposed, including the MEIC (Multicentre Evaluation of In Vitro Cytotoxicity) project, the ECITTS (ERGATT/CFN Integrated Toxicity Testing Scheme) approach, and proposals from the University of Kiel and the University of Nottingham (27). Individuals associated with all four of these initiatives were participants in the workshop. A number of developments have taken place since the workshop was held, which have also been taken into account in this report.

The MEIC project was organised under the auspices of the Scandinavian Society for Cell Toxicology and ran from 1989 until 1996 (28-31). It was a collective voluntary effort to evaluate the relevance and reliability of in vitro cytotoxicity tests as alternatives to, or supplements to, animal tests for acute systemic toxicity, chronic systemic toxicity, target organ toxicity, skin irritancy, or other forms of general toxicity. Fifty reference chemicals were selected, to be tested by participants in their own laboratories according to their own test procedures. These chemicals were selected primarily because sufficiently good human data on acute systemic toxicity were available for them. One of the main conclusions of the project, when the application of 68 methods to the first 30 MEIC chemicals was evaluated, was that similar results were obtained, regardless of the cell type employed and regardless of whether cell viability or cell proliferation endpoint measurements were used. The managers of the project concluded that the findings strongly supported the basal cytotoxicity concept, that is, that the majority of chemicals ultimately cause toxicity in humans via general cytotoxic mechanisms (31).

The ECITTS approach is built on the idea that sets of test batteries for predicting different types of local and systemic toxicity can be combined into integrated testing schemes, in ways which would be more efficient than animal-based investigations (32). It was recognised that basal cytotoxicity testing and an evaluation of biokinetic parameters would be vital in all applications of the ECITTS approach, but that subsequent testing would depend on the test material concerned and the specific area(s) of main interest (for example, developmental toxicity, immunotoxicity, nephrotoxicity, neurotoxicity). The basal cytotoxicity data were seen as essential to the interpretation of more specific effects on potential target cells and tissues (33), while protein binding and biotransformation were seen as essential components of the biokinetics evaluation. A pilot study on the ECITTS approach has been supported by ECVAM (33), and a further, more extensive, ECVAM-supported study is now in progress.

Seibert and his colleagues at the Institute of Toxicology, University of Kiel, have developed an approach to testing for acute systemic toxicity (34) which involves the use of a continuous cell line (Balb/c 3T3 cells) and differentiated mammalian cells (primary cultures of rat hepatocytes and of rat skeletal muscle cells, and bovine spermatozoa). The Kiel group emphasises the importance of comparative cell toxicology (involving different endpoints, tissues and species [35]), and of incorporating physicochemical data (for example, on lipophilicity [36]).

Two other ECVAM workshop reports are relevant to the outcome of the discussions which took place at the workshop; namely, the report of ECVAM Workshop 8, on the integrated use of alternative approaches for predicting toxic hazard (37), and the report of ECVAM Workshop 15, on the use of biokinetics and in vitro methods in toxicological risk evaluation (38).

Inclusion of Biokinetic Parameters

The ultimate toxicity of a chemical in vivo is strongly influenced by the time-dependent processes of intake, uptake (absorption), distribution in the organism, biotransformation, and elimination. These factors determine the biokinetics (toxicokinetics) of a compound in a particular species or individual. How, then, can biokinetic considerations be incorporated into testing for acute toxicity in vitro?

Typically, the in vitro effects of chemicals on cells are investigated and the critical concentrations of chemicals producing these effects are determined (for example, EC50 or EC10 values). These critical concentrations characterise the potencies of chemicals to produce a given effect in vitro under particular conditions (39). To find out whether the effects measured and the potencies determined in vitro are likely to be of any relevance for assessing the acute toxicities of chemicals in vivo, in vitro potency data have to be compared to in vivo acute toxic potency data for a sufficiently large and sufficiently representative set of chemicals.

The appropriate in vivo data for comparison with the critical concentrations determined in vitro are the critical concentrations at the target sites of acute toxic effects in vivo. This is the concentration to which the ultimate toxicity of a compound is related. This concentration, however, is very seldom known. The concentration at the target site is related to, but need not be identical to, that in the circulating blood plasma. Due to various factors, target site concentrations may deviate markedly from plasma concentrations in many cases. Nevertheless, critical plasma concentrations are certainly a reasonable basis for conducting in vivo/in vitro comparisons. Plasma concentrations depend upon the dose administered and the route of administration, and on various biokinetic factors.

Human blood/plasma/serum concentrations associated with toxicity or death are known for some chemicals. The MEIC reference chemicals were selected on the basis of such knowledge (28, 29). However, the in vivo acute toxic potencies of chemicals are more frequently characterised by LD50 values. Thus, the in vivo potency data which are typically available are critical body doses, which are expressed in units of dosages (for example, as mg/kg body weight). In vitro potencies are given in units of concentration (for example, as µg/ml of the culture medium). Obviously, these quantities should not be compared directly, because of the importance of biokinetic factors. Nevertheless, due to the lack of the necessary biokinetic data, it has been a common practice to directly compare LD50 and EC50 values as a means of evaluating the relevance of in vitro data for the prediction of acute toxicity in vivo.

Linear regression analysis is often performed to describe the relationship between such EC50 and LD50 values (25); that is, a linear relationship between EC50 and LD50 values is assumed. However, such a linear relationship would only be expected if: a) all of the reference chemicals were mechanistically similar (for example, if they produced acute toxicity in vivo by a similar mechanism, and the effects determined in vitro were related to this mechanism); and b) all of the reference chemicals demonstrated similar biokinetics (for example, if similar conversion factors related the acute toxic concentrations at the target site to the acute toxic body doses).

For a population of toxicodynamically and toxicokinetically unrelated test chemicals, the result of the linear regression analysis will inevitably indicate a big scatter. Nevertheless, linear regression analysis may reveal a statistically significant and positive linear correlation between EC50 and LD50 values. Such a result must be interpreted with caution, since: a) linear regression lines are often largely dictated by data for a few chemicals with either very high or very low potencies; and b) if the data points for the moderately toxic chemicals are considered alone, there would often appear to be little correlation between the EC50 and LD50 values.

The correlative approach to the evaluation of relevance might be successful if it were possible to select chemicals on the basis of mechanistic and biokinetic similarities. Then the linear correlation might also be meaningful, and the established correlative relationship between in vivo and in vitro data for such a class of chemicals could be used to predict the acute toxicity of a new test chemical in that particular mechanistic and biokinetic class. Nevertheless, it was recognised that it is often difficult to classify chemicals according to their biokinetics and mechanisms of action.

There was general agreement that biokinetic factors have to be considered before performing in vitro/in vivo comparisons, in order to make the in vivo and in vitro data more comparable and the resulting comparison more meaningful.

Biotransformation

Chemicals can produce their toxic effects either directly, or following biotransformation in the liver or in other tissues. The possibility that toxic metabolites may be formed must be considered. One of the major limitations to using cell lines for cytotoxicity studies is their poor metabolic competence, as a result of which they are typically able to detect only those chemicals which are directly toxic to the cell at the concentrations attainable within the practical constraints of the methodology used. Therefore, the toxicities of compounds which must be metabolised in order to exert their ultimate toxic effects are not accurately predicted by these in vitro tests for basal cytotoxicity.

Although many tissues can contribute significantly to the metabolic modification of xenobiotics, the liver is the predominant organ involved in biotransformation. To facilitate their elimination from the organism, lipophilic xenobiotics are chemically modified in the liver in a series of biotransformation reactions (known as Phase I and Phase II reactions), which increase their polarities and solubilities. If this does not occur, lipophilic compounds can accumulate in other body tissues, which can result in toxicity. The compound or its metabolites generated during Phase I reactions are frequently, but not always, conjugated with endogenous molecules (glucuronic acid, glutathione, sulphate, amino acids) to produce derivatives which are much more soluble and more easily excreted. Such Phase II reactions are frequently limited by the availability of the endogenous compounds required for conjugation with either the parent compound or its Phase I metabolite(s). A major endogenous protective system is the glutathione redox cycle. Glutathione is present at high concentrations in most mammalian cells, and it acts as a nucleophilic scavenger of chemicals and their metabolites.

Although biotransformation generally represents a detoxification process, there are many examples in which metabolites which are produced during Phase I reactions are more reactive and toxic than the parent compounds; this often results in target organ toxicity. The action of the reactive metabolite depends, among other factors, on its half-life. In addition, Phase II conjugates can also be responsible for the toxic effects of certain chemicals. Ultimately, the toxicity of a compound depends on the balance between the Phase I and Phase II metabolic pathways.

The large biotransformation capability of the liver permits an efficient elimination of toxic compounds, but can also make the liver itself a major target organ for toxicity. A large number of chemicals are known to induce liver toxicity. Compounds absorbed from the alimentary canal, after oral or deep rectal administration, may undergo a first-pass effect in the liver and subsequent enterohepatic recirculation. Both of these processes can increase hepatic exposure to the chemicals or their metabolites.

Due to the metabolic and toxicological importance of the liver, many liver-derived in vitro experimental systems have been developed. These include the use of homogenates, subcellular fractions (for example, microsomes), liver slices, freshly isolated hepatocytes (in suspension), primary monolayer cultures of hepatocytes, and immortalised hepatocyte and hepatoma cell lines (40). The key requirement for biotransformation studies is the use of preparations which maintain appropriate and sufficient metabolic competence. Liver homogenates and subcellular fractions lack those Phase II enzymes which are not membrane-bound, and the use of hepatoma cells for metabolism and toxicity studies is limited by the fact that these cells do not express many biotransformation activities. There is also variable stability in the expression of both Phase I and Phase II enzyme activities in freshly isolated hepatocytes and in primary cultures of hepatocytes from various animal species and from humans.

The detection of selective toxicity requires a comparison of the toxicities of the same chemical in different cell types, including hepatocytes. Furthermore, a means of detecting the effects of toxic metabolites on target cells may be required. This can be undertaken by: a) co-culturing the competent metabolising cells with the target cells; or b) exposing hepatocytes (as the metabolising cells) to the test compound, and then culturing the target cell in the resulting conditioned culture medium. There was general agreement that the co-culture of hepatocytes with target cells, as described by Ericsson & Walum (41) and by Voss & Seibert (42, 43), appears to be the most promising approach, since it enables the detection of: a) hepatocyte-specific cytotoxicity; b) interference with specific (non-vital) functions of hepatocytes; and c) metabolism-mediated effects on target cells.

Distribution In Vitro and In Vivo

There are various approaches to taking biokinetic factors into account in the extrapolation of critical body doses from critical concentrations, and vice versa.

Physiologically-based biokinetic (PBBK) modelling would certainly be the most reliable approach (37, 38). However, it is very complex and requires a great deal of knowledge about in vivo target organs and about various toxicokinetic factors for the particular chemical of interest, and cannot be regarded as a suitable method for the toxicity assessment of large numbers of chemicals. Another approach is to use a few, carefully selected, in vivo biokinetic parameters, such as the fraction absorbed from the intestine and the apparent volume of distribution, to estimate body doses from critical in vitro concentrations and to estimate plasma concentrations from body doses (29, 36, 38).

However, for most chemicals apart from pharmaceuticals, information on biokinetics in vivo is lacking. If in such cases, acute toxic doses are to be predicted, non-animal data must be relied upon. The fraction absorbed from the intestine could be estimated from knowledge about the general relationships between physicochemical properties of chemicals and their absorption in the gastrointestinal tract, or from experimental data obtained in vitro (38). The latter could be derived, for example, by using two-compartment systems comprising epithelia-like monolayers of human colon carcinoma cells (for instance, Caco-2 or HT-29). Non-animal data on specific chemicals, and parameters defining the composition/ compartmentalisation of particular in vitro test systems and of the in vivo animal model, can be used as the basis for converting in vitro effective concentrations into equivalent body doses. To do this, it is necessary to have, as a minimum, information on: a) various physicochemical characteristics of the chemical (for example, pKa, lipophilicity, volatility); b) quantitative estimates of protein binding; and c) the basic characteristics of the in vitro system (for example, cell concentration, cell protein concentration, ratio of cell-medium volumes, medium albumin concentration). In addition, a mathematical model is needed, which permits the calculation of equivalent body doses.

The development of such an approach is only at its early stages. Gülden et al. (36) have described the first steps, which take into account the equilibrium distribution of xenobiotics governed by their lipophilicities and the relative volumes of the lipid and aqueous phases. The relationship between the volumes of both compartments differs markedly in cell culture systems compared to the mammalian body. Two basic assumptions underly this extrapolation method. Firstly, ignoring any other biokinetic factors (including plasma protein and tissue protein binding), both the in vitro and the in vivo systems (model body) were considered to be simple two-compartment systems, with the total dose of a given substance being equally distributed between the water and the lipid compartments according to the octanol/water partition coefficient (Kow). Secondly, it was assumed that a model body dose is equivalent to an in vitro effective nominal concentration (for example, the EC50 value), in which case it results in the same concentration in the water compartment of the model body as in the in vitro system.

It is evident that this approach alone is not sufficient. Studies need to be conducted on other factors which could be important, especially for passive distribution, and ways in which these could be determined and included.

A Proposed Tier Testing Scheme for Acute Toxicity

Based on the experience of the workshop participants with various existing in vitro test systems, there was general agreement that, at the moment, as a practical approach to testing for acute systemic toxicity in vivo, three principal steps for testing in vitro can be proposed.

Stage 1

Basal cytotoxicity should be determined by using cell proliferation inhibition as the endpoint. A rapidly dividing, transformed, undifferentiated cell line is best suited for this purpose. The cells should be exposed during the exponential growth phase, and the exposure time should be at least three times as long as the doubling time of the cells employed.

During the workshop, there was much discussion about the additional value of measuring cytolethal concentrations in non-dividing cells at this stage; some participants felt that cytolethality could be more relevant than growth inhibition for predicting acute systemic toxicity in vivo.

Stage 2

A test should be performed which enables the detection of hepatocyte-specific cytotoxicity and defines the role of biotransformation in the cytotoxic effects of the test chemical. In principle, this could be achieved by using co-cultures of metabolically competent hepatocytes with the proliferating cells used in Stage 1 as target cells. Such a system, consisting of microcarrier-attached rat hepatocytes co-cultured with proliferating 3T3 cells as the target cells in a two-compartment system, has been described previously by Voss & Seibert (42, 43). Similar systems are now commercially available (44). Another possible two-compartment system could involve the use of filter inserts in culture dishes to separate the "metabolism" and "response" systems, or application of the roller chamber technique described by Ericsson and Walum (45). A third possibility would be to expose hepatocytes to the test compound and then to culture the target cells in the hepatocyte-conditioned medium. Further effort should be invested in the development of such systems, to find out which one is best suited for testing in this context, based on practicability and the ability to interpret the results obtained in terms of likely effects in vivo.

Stage 3

The test system used at Stage 3 should provide information concerning selective cytotoxicity (other than any hepatocyte-specific cytotoxicity, which would have already been detected at Stage 2), as well as an indication of any interference with important specific, but non-vital, cell functions. Many possible in vitro test systems could be included at this stage, employing differentiated cells from various potential target organs and systems, for instance, cells from the nervous system, heart or kidney (for example, 46). In principle, the same line of argument is valid for the cell-specific endpoints which should be used to detect chemical interference with cell functions of vital importance in vivo.

A promising example of a Stage 3 test is the contractility inhibition assay in primary cultures of rat skeletal muscle cells. Microscopically visible contractions of these cells are triggered by spontaneous electrical activity in their excitable membranes. Spontaneous contractility was shown to be responsive to the actions of model chemicals known: a) to change the resting membrane potential and the threshold of excitation; b) to alter the function of different ion channels; c) to inhibit Na+/K+-ATPase; and d) to have so-called membrane "fluidising" or "stabilising" activities (47, 48). Thus, there is evidence that this test system permits the detection of those neurotoxic and cardiotoxic compounds which act by interference with electrically excitable membranes. An important question for the future is which additional cell types and specific endpoints have to be included as Stage 3 tests.

In Vitro Testing and the Classification of Chemicals

As outlined above, there was general agreement that a key step in the application of in vitro test data is relating nominal in vitro effective concentrations to in vivo effective doses. This means that important biokinetic factors have to be taken into account, and such parameters should be included in the mathematical prediction model. This is true for all in vitro data, independent of the stage of testing. Taking all of this into account, a testing scheme for the classification and labelling of chemicals according to their acute toxicities has been proposed (Figure 1), which is similar to the tier scheme outlined in the section above.

Figure 1: Proposed Testing Scheme for the Classification and Labelling of Chemicals According to Their Potential Acute Toxicities

As the first step, a test for basal cytotoxic activity would be conducted as described earlier. The result would be converted to an "equivalent effective body dose" by means of a prediction model based on physicochemical data and basic assumptions about toxicokinetic parameters in vivo and in vitro. If the result were positive, that is, if it indicated that the compound should be classified as "very toxic", no further testing would need to be done. If not, Stage 2 testing would have to be performed. Again, if the result were positive (that is, classification in the highest toxicity class was indicated), testing would be stopped at this stage. If not, Stage 3 testing would have to be performed. Finally, the chemical would be classified as "very toxic", "toxic", "harmful", or "no label", according to the lowest EC50 value determined at any of the three testing levels.

If the results indicated that the chemicals should be assigned to the lowest toxicity class (that is, "no label"), a limited in vivo confirmatory study might need to be carried out. These animal experiments would be expected to confirm that no major underestimation of toxic potency had occurred; this could be the case if the particular chemical acted by a mechanism not covered by the in vitro test battery, or if toxicokinetics in vivo exerted an unexpected influence.

Conclusions and Recommendations

  1. The question of whether, and if so, to what extent, non-animal tests or testing strategies could significantly reduce or even replace the need for in vivo acute systemic toxicity tests is still unresolved. According to the current view of many regulatory and industrial toxicologists, knowledge of the in vivo toxic effects resulting from acute poisoning is a prerequisite for trying to ensure that workers and consumers are adequately protected. In vitro methods for acute toxicity testing could, however, be used in a tier testing scheme to reduce the number of animals used (and to reduce animal suffering). Such an approach is a natural progression of recent attempts to refine in vivo acute toxicity tests by the use of sequential dosing methods (such as the acute toxic class and up-and-down procedures). In these in vivo tests, use of the minimum number of animals possible depends upon the correct choice of the starting dose; this could be optimised by conducting appropriate in vitro tests prior to any animal tests which were then considered to be necessary.
  2. Many studies have shown good correlations between in vitro cytotoxicity data obtained with undifferentiated cell lines and LD50 data. However, acute systemic toxicity can be caused by a variety of mechanisms, many of which would not be operative in such basal cytotoxicity tests, so more sophisticated approaches are also necessary.
  3. Chemicals can have three main types of toxic effect at the cellular level, namely, general (basal) cytotoxicity, selective cytotoxicity, and cell-specific function toxicity. Since all three types of effect can result in acute systemic toxicity in vivo, any non-animal test or testing strategy must take them all into account.
  4. A prerequisite to the development of any toxicity test system must be an appreciation of the concepts of active concentration and of meaningful dose. It must not be assumed that what is added to the culture medium will be available to the cells in a simple concentration-related manner. Nor must it be forgotten that almost all test materials will exert general cytotoxic effects when added at sufficiently high concentrations. It is essential in any scheme that a quantitative comparison be made of the concentrations of test materials which exert basal cytotoxic effects versus selectively cytotoxic or cell-specific function effects.
  5. The ultimate toxicity of a chemical in vivo is strongly influenced by the time-dependent processes of intake, uptake (absorption), distribution in the organism, biotransformation, and elimination. These factors determine the biokinetics (toxicokinetics) of a compound in a particular species. Ways must be found of evaluating such factors, so that they can be taken into account when predictions of in vivo toxicity are based on in vitro data.
  6. Non-animal data on specific chemicals, and parameters defining the composition/compartmentalisation of particular in vitro test systems and of the in vivo animal model, can be used as the basis for converting in vitro effective concentrations into equivalent body doses. To do this, it is necessary to have, as a minimum, information on: a) various physicochemical characteristics of the chemical (for example, pKa, lipophilicity, volatility); b) quantitative estimates of protein binding; and c) the basic characteristics of the in vitro system (for example, cell concentration, cell protein concentration, ratio of cell/medium volumes, medium albumin concentration).
  7. The detection of selective toxicity requires a comparison of the toxicities of the same chemical in different cell types, including hepatocytes. Furthermore, a means of detecting the effects of toxic metabolites on target cells may be required.
  8. A three-stage tier scheme for in vitro testing for acute systemic toxicity is proposed:

    Stage 1. Basal cytotoxicity should be determined by using cell proliferation inhibition as the endpoint.

    Stage 2. A test should be performed which permits the detection of hepatocyte-specific cytotoxicity and defines the role of biotransformation in the cytotoxic effects of the test chemical.

    Stage 3. The test system used at Stage 3 should provide information concerning selective cytotoxicity (other than any hepatocyte-specific cytotoxicity, which would have already been detected at Stage 2), as well as an indication of any interference with important specific, but non-vital, cell functions. Many possible in vitro test systems could be included at this stage, employing differentiated cells from various potential target organs and systems, for example, cells from the nervous system, heart or kidney. In principle, the same line of argument is valid for the cell-specific endpoints which should be used to detect chemical interference with cell functions of vital importance in vivo.

  9. A scheme for the classification and labelling of chemicals is proposed (Figure 1). As the first step, a test for general cytotoxicity activity would be conducted. The result would be converted to an "equivalent effective body dose" by means of a prediction model based on physicochemical data and basic assumptions about toxicokinetic parameters in vivo and in vitro. If the result were positive, that is, if it indicated that the compound should be classified as "very toxic", no further testing would need to be done. If not, Stage 2 testing would have to be performed. Again, if the result were positive (that is, classification in the highest toxicity class was indicated), testing would be stopped at this stage. If not, Stage 3 testing would have to be performed. Finally, the chemical would be classified as "very toxic", "toxic", "harmful", or "no label", according to the lowest EC50 value determined at any of the three testing levels. If the results indicated that the chemicals should be assigned to the lowest toxicity class (that is, "no label"), a limited in vivo confirmatory study might need to be carried out.
  10. A feasibility study should be conducted on the likely practicability, relevance, and reliability of this scheme, including an evaluation of the state of development and validation of the component tests and of non-animal test systems for determining essential biokinetic parameters.

References

  1. Anon. (1994). ECVAM News & Views. ATLA 22: 7-11.
  2. Trevan, J.W. (1927). The error of determination of toxicity. Proceedings of the Royal Society (London), Series B 101: 483-514.
  3. Anon. (1987). OECD Guideline for Testing of Chemicals, No. 401. Acute Oral Toxicity. 7 pp. Paris: OECD.
  4. Diener, W., Siccha, L., Mischke, U., Kayser, D. & Schlede, E. (1994). The biometric evaluation of the acute toxic class method (oral). Archives of Toxicology 68: 599-610.
  5. Schlede, E., Mischke, U., Diener, W. & Kayser, D. (1995). The international validation study of the acute toxic class method (oral). Archives of Toxicology 69: 659-670.
  6. Anon. (1996). OECD Guideline for Testing of Chemicals, No. 423. Acute Oral Toxicity -- Acute Toxic Class Method. 16 pp. Paris: OECD.
  7. Bruce, R.D. (1985). An up-and-down procedure for acute toxicity testing. Fundamental and Applied Toxicology 6: 151-157.
  8. Lipnick, R.L., Cotruvo, J.A., Hill, R.N., Bruce, R.D., Stitzel, K.A., Walker, A.P., Chu, I., Goddard, M., Segal, L., Springer, J.A. & Myers, R.C. (1995). Comparison of the up-and-down, conventional LD50, and fixed-dose acute toxicity procedures. Food and Chemical Toxicology 33: 223-231.
  9. van den Heuvel, M.J., Clark, D.G., Fielder, R.J., Koundakjian, P.P., Oliver, G.J.A., Pelling, D., Tomlinson, N.J. & Walker, A.P. (1990). The international validation of a fixed-dose procedure as an alternative to the classical LD50 test. Food and Chemical Toxicology 28: 469-482.
  10. Anon. (1992). OECD Guideline for Testing of Chemicals, No. 420. Acute Oral Toxicity - Fixed Dose Method. 7 pp. Paris: OECD.
  11. Anon. (1995). Statistics of Scientific Procedures on Living Animals in Great Britain 1994. 52 pp. London: HMSO.
  12. Balls, M. & Fentem, J.H. (1993). The on-going process to replace the LD50 test. Humane Innovations and Alternatives 7: 544-547.
  13. Ekwall, B. (1983). Correlation between cytotoxicity in vitro and LD50 values. Acta Pharmacologica et Toxicologica 52 (Suppl. II): 80-99.
  14. Clothier, R.H., Hulme, L.M., Smith, M. & Balls, M. (1987). Comparison of the in vitro toxicities of 59 chemicals. Molecular Toxicology 1: 571-577.
  15. Fry, J.R., Garle, M.J. & Hammond, A.H. (1988). Choice of acute toxicity measures for comparison of in vivo/in vitro toxicity. ATLA 16: 175-179.
  16. Phillips, J.C., Gibson, W.B., Yam, J., Alden, C.L. & Hard, G. (1990). Survey of the QSAR and in vitro approaches for developing non-animal methods to supersede the in vivo LD50 test. Food and Chemical Toxicology 28: 375-394.
  17. Fry, J.R., Garle, M.J., Hammond, A.H. & Hatfield, A. (1990). Correlation of acute lethal potency with in vitro cytotoxicity. Toxicology In Vitro 4: 175-178.
  18. Babich, H. & Borenfreund, E. (1991). Cytotoxicity and genotoxicity assays with cultured fish cells: a review. Toxicology In Vitro 5: 91-100.
  19. Balls, M. & Clothier, R.H. (1992). Cytotoxicity assays for intrinsic toxicity and irritancy. In In Vitro Methods of Toxicology (ed. R.R. Watson), pp. 37-52. Boca Raton, FL: CRC Press.
  20. Balls, M. & Fentem, J.H. (1992). The use of basal cytotoxicity and target organ toxicity tests in hazard identification and risk assessment. ATLA 20: 368-388.
  21. Halle, W. & Spielmann, H. (1992). Two procedures for the prediction of acute toxicity (LD50) from cytotoxicity data. ATLA 20: 40-49.
  22. Seibert, H., Gülden, M., Kolossa, M. & Schepers, G. (1992). Evaluation of the relevance of selected in vitro toxicity test systems for acute systemic toxicity. ATLA 20: 240-245.
  23. Shrivastava, R., Delomenie, C., Chevalier, A., John, G., Ekwall, B., Walum, E. & Massingham, R. (1992). Comparison of in vivo acute lethal potency and in vitro cytotoxicity of 48 chemicals. Cell Biology and Toxicology 8: 157-170.
  24. Guzzie, P.J. (1994). Lethality testing. In In Vitro Toxicology (ed. S.C. Gad), pp. 57-86. New York: Raven Press.
  25. Garle, M.J., Fentem, J.H. & Fry, J.R. (1994). In vitro cytotoxicity tests for the prediction of acute toxicity in vivo. Toxicology In Vitro 8: 1303-1313.
  26. Parish, W. (1991). The future for acute oral toxicity testing. In Animals and Alternatives in Toxicology - Present Status and Future Prospects (ed. M. Balls, J. Bridges & J. Southee), pp. 19-20. Basingstoke: Macmillan and New York: VCH Publishers.
  27. Fentem, J., Fry, J., Garle, M., Gülden, M., Seibert, H., Voss, J-U., Wassermann, O., Perchermeier, M. & Wiebel, F. (1993). An International Evaluation of Selected In Vitro Toxicity Test Systems for Predicting Acute Systemic Toxicity, 143 pp. Report prepared for the European Commission (DGXI), contract no. B92/B4-3063/14086.
  28. Bondesson, I., Ekwall, B., Hellberg, S., Rombert, L., Stenberg, K. & Walum, E. (1989). MEIC - a new international multicenter project to evaluate the relevance to human toxicity of in vitro cytotoxicity tests. Cell Biology and Toxicology 5: 331-348.
  29. Ekwall, B., Bondesson, I., Castell, J.V., Gómez-Lechón, M.J., Hellberg, S., Hógberg J., Jover, R., Ponsoda, X., Romert, L., Stenberg, K. & Walum, E. (1989). Cytotoxicity evaluation of the first ten MEIC chemicals: acute lethal toxicity in man predicted by cytotoxicity in five cellular assays and by oral LD50 tests in rodents. ATLA 17: 83-100.
  30. Clemedson, C., McFarlane-Abdulla, E., Andersson, M., Barile, F.A., Calleja, M.C., Chesné, C., Clothier, R., Cottin, M., Curren, R., Daniel-Szolgay, E., Dierickx, P., Ferro, M., Fiskesjö, G., Garza-Ocañas, L., Gómez-Lechón, M.J., Gülden, M., Isomaa, B., Janus, J., Judge, P., Kahru, A., Kemp, R.B., Kerszman, G., Kristen, U., Kunimoto, M., Kärenlampi, S., Lavrijsen, K., Lewan, L., Lilius, H., Ohno, T., Persoone, G., Roguet, R., Romert, L., Sawyer, T.W., Seibert, H., Shrivastava, R., Stammati, A-L., Tanaka, N., Torres-Alanis, O., Voss, J-U., Wakuri, S., Walum, E., Wang, X., Zucco, F. & Ekwall, B. (1996). MEIC evaluation of acute systemic toxicity. Part I. Methodology of 68 in vitro toxicity assays used to test the first 30 reference chemicals. ATLA 24, Suppl. I: 251-272.
  31. Clemedson, C., McFarlane-Abdulla, E., Andersson, M., Barile, F.A., Calleja, M.C., Chesné, C., Clothier, R., Cottin, M., Curren, R., Daniel-Szolgay, E., Dierickx, P., Ferro, M., Fiskesjö, G., Garza-Ocañas, L., Gómez-Lechón, M.J., Gülden, M., Isomaa, B., Janus, J., Judge, P., Kahru, A., Kemp, R.B., Kerszman, G., Kristen, U., Kunimoto, M., Kärenlampi, S., Lavrijsen, K., Lewan, L., Lilius, H., Ohno, T., Persoone, G., Roguet, R., Romert, L., Sawyer, T.W., Seibert, H., Shrivastava, R., Stammati, A-L., Tanaka, N., Torres-Alanis, O., Voss, J-U., Wakuri, S., Walum, E., Wang, X., Zucco, F. & Ekwall, B. (1996). MEIC evaluation of acute systemic toxicity. Part II. In vitro results from 68 toxicity assays used to test the first 30 reference chemicals and a comparative cytotoxicity analysis. ATLA 24, Suppl. I: 273-311.
  32. Walum, E., Balls, M., Bianchi, B., Blaauboer, B., Bolcsfoldi, G., Guillouzo, A., Moore, G.A., Odland, L., Reinhardt, C.A., Spielmann, H. (1992). ECITTS: an integrated approach for the application of in vitro test systems for the hazard assessment of chemicals. ATLA 20: 406-428.
  33. Forsby, A., Pilli, F., Bianchi, V. & Walum, E. (1995). Determination of critical cellular neurotoxic concentrations in human neuroblastoma (SH-SY5Y) cell cultures. ATLA 23: 800-811.
  34. Seibert, H., Gülden, M. & Voss, J-U. (1994). An in vitro toxicity testing strategy for the classification and labelling of chemicals according to their potential acute lethal potency. Toxicology In Vitro 8: 847-850.
  35. Seibert, H., Gülden, M., & Voss, J-U. (1994). Comparative cell toxicology: the basis for in vitro toxicity testing. ATLA 22: 168-174.
  36. Gülden, M., Seibert, H. & Voss, J-U. (1994). Inclusion of physicochemical data in quantitative comparisons of in vitro and in vivo toxic potencies. ATLA 22: 185-192.
  37. Barratt, M.D., Castell, J.V., Chamberlain, M., Combes, R.D., Dearden, J.C., Fentem, J.H., Gerner, I., Giuliani, A., Gray, T.J.B., Livingstone, D.J., Provan, W.Mc., Rutten, A.A.J.J.L., Verhaar, H.J.M. & Zbinden, P. (1995). The integrated use of alternative approaches for predicting toxic hazard. The report and recommendations of ECVAM workshop 8. ATLA 23: 410-429.
  38. Blaauboer, B.J., Bayliss, M.K., Castell, J.V., Evelo, C.T.A., Frazier, J.M., Groen, K., Gülden, M., Guillouzo, A., Hissink, A.M., Houston, J.B., Johanson, G., de Jongh, J., Kedderis, G.L., Reinhardt, C.A., van de Sandt, J.J.M. & Semino, G. (1996). The use of biokinetics and in vitro methods in toxicological risk evaluation. The report and recommendations of ECVAM workshop 15. ATLA 24: in press.
  39. Walum, E., Nordin, M., Beckman, M. & Odland, L. (1993). Cellular methods for identification of neurotoxic chemicals and estimation of neurotoxicological risk. Toxicology In Vitro 7: 321-326.
  40. Blaauboer, B.J., Boobis, A.R., Castell, J.V., Coecke, S., Groothuis, G.M.M., Guillouzo, A., Hall, T.J., Hawksworth, G.M., Lorenzon, G., Miltenburger, H.G., Rogiers, V., Skett, P., Villa, P. & Wiebel, F.J. (1994). The practical applicability of hepatocyte cultures in routine testing. The report and recommendations of ECVAM workshop 1. ATLA 22: 231-241.
  41. Ericsson, A.C. & Walum, E. (1988). Differential effects of allyl alcohol on hepatocytes and fibroblasts demonstrated in roller chamber co-cultures. ATLA 15: 208-213.
  42. Voss, J-U. & Seibert, H. (1991). Microcarrier-attached rat hepatocytes as a xenobiotic-metabolising system in co-cultures. Cell Biology and Toxicology 7: 387-399.
  43. Voss, J-U. & Seibert, H. (1992). Toxicity of glycols and allyl alcohol evaluated by means of co-cultures of microcarrier-attached rat hepatocytes and Balb/c 3T3 mouse fibroblasts. ATLA 20: 266-270.
  44. Guyomard, C., Chesné, C. & Guillouzo, A. (1994). Immobilized frozen hepatocytes: a cell kit for in vitro analysis of hepatic drug metabolism and cytotoxicity testing. Cell Biology and Toxicology 10: 445.
  45. Ericsson, A.C., & Walum, E. (1984). Cytotoxicity of cyclophosphamide and acrylamide in glioma and neuroblastoma cell lines co-cultured with liver cells. Toxicology Letters 20: 251-256.
  46. Walum, E., Varnbo, I. & Peterson, A. (1986). A multiple cell-culture toxicity test system based on neuroblastoma C1300 cells and differentiated primary cultures of brain, muscle, heart, and liver cells. Food and Chemical Toxicology 24: 567-568.
  47. Gülden, M. (1993). In vitro toxicity screening using cultured rat skeletal muscle cells. I. Surfactants and mitochondrial poisons. Toxicology In Vitro 7: 25-34.
  48. Gülden, M., Seibert, H. & Voss, J-U. (1994). In vitro toxicity screening using cultured rat skeletal muscle cells. II. Agents affecting excitable membranes. Toxicology In Vitro 8: 197-206.