- Email: [email protected]
Genotoxicity testing strategies
Toxic'. in Vitro Vol. 8, No. 4, pp. 871-872, 1994
~
Pergamon
0887-2333(94)E0086-9
Copyright © 1994ElsevierScienceLtd Printed in Great Britain.All rights reserved 0887-2333/94$7.00+ 0.00
GENOTOXICITY TESTING STRATEGIES B. M. ELLIOTT Zeneca Central Toxicology Laboratory, Alderley Park, Macclesfield, Cheshire SKI0 4TJ, UK Abstract~enotoxicity tests contribute to the assessment of whether a chemical has the potential to cause somatic or germ-cell effects in animals (i.e. the potential to induce cancer or heritable mutation). Such genotoxicity testing is usually undertaken in a stepwise approach; first an assessment in vitro, to determine intrinsic genotoxic activity, and secondly, an evaluation in vivo to determine whether any such activity is expressed in the whole animal. While this core principle is now generally accepted, there are still different opinions as to when, and how many, in vitro or in vivo assays should be conducted. Such strategies would be simplified and harmonized if testing was carried out based on the knowledge that there was a precedent for that strategy being required to detect a carcinogen or mutagen rather than the possibility that it might.
Introduction Genotoxicity tests contribute to the toxicological assessment of whether a chemical has the potential to induce cancer or heritable mutation in animals. The principle underlying these tests is whether the chemical or a metabolite interacts with and damages DNA. There is a large number of tests available. These vary in complexity and cover a range of cell types and genetic endpoints. Until recently, in some quarters there has been an apparent comparability in the way assays as disparate as the Ames test (an in vitro bacterial assay) and the mouse specific locus assay (involving hundreds of mice) are considered to give an assessment of genotoxicity (gene mutation) (Berry and Litchfield, 1985). Such an equivalence would render testing strategies valueless. However, the suggestion by Bridges (1974), and the promotion by Ashby (1986), of a stepwise approach to genotoxicity testing has allowed the development of both effective and efficient hierarchical testing strategies to assess the likely genotoxicity of a chemical to the whole animal. This approach, together with the results of international trials designed to assess the value and reliability of the plethora of available genotoxicity assays for their ability to detect carcinogens and discriminate non-carcinogens (Ashby et al., 1985 and 1988; de Serres and Ashby, 1981), has led to the current focused testing strategies that use a limited n u m b e r of validated assays interpreted in a coherent and stepwise manner.
A stepwisegenotoxicitytestingstrategy The core principle of the current strategies generally accepted by genetic toxicologists is that an initial assessment is made using in vitro assays, which are designed to be sensitive and to detect an intrinsic genotoxic activity. If clear evidence of genotoxicity is seen in one or more of these assays, an assessment is made in vivo in order to determine whether this
intrinsic genotoxic activity is expressed in the whole animal. Since the in vivo assays are designed to assess the relevance of the in vitro results to the animal, it is recommended that the route of exposure used should be relevant for potential h u m a n exposure. To do otherwise undermines the unique role of the animal studies in this testing stategy. They must be allowed to superimpose the effects of absorption, distribution, metabolism, and excretion, together with the cellular processes such as D N A repair, onto the established intrinsic activity of the chemical. This then provides a toxicologically relevant assessment of any genotoxic effects in animals. The available data indicate that a thorough evaluation can be achieved in vitro by using the Ames test together with an in vitro cytogenetic assay in mammalian cells, and in vivo by using the bone marrow micronucleus assay and rat liver unscheduled D N A synthesis assay (Fig. 1). These provide an effective screen for genotoxic carcinogens. These principles are reflected in recent genotoxicity testing guideline updates both within (Kirkland, 1993) and outside (Clayson and Grant, 1993) Europe. In such a strategy, once genotoxicity has been determined in vitro, further in vitro assays are of limited value. What is required is the evaluation in vivo. The use of chemical structure-activity considerations can be invaluable in the interpretation of any activity seen in the in vitro assays (Tennant and Ashby, 1991). Thus an aromatic amine structure based on 4-aminobiphenyl that gives a positive response in an Ames test using Salmonella strain TA98 ( + S - 9 ) will not benefit from further in vitro evalua t i o n - i t is genotoxic in vitro; the evaluation should move immediately to assess any effects in the whole animal. There is no value in conducting further in vitro assays simply because they are present in a test strategy. The above strategy involves in vivo investigations in somatic cells and defines the genotoxicity of the test 871
872
B. M. ELLIOTT
Step I. IN VITRO ASSESSMENT FOR INTRINSIC ACTIVITY Ames test Cytogenetic assay in mammalian cells if positive:
Step 2. IN VIVO ASSESSMENT FOR EXPRESSED ACTIVITY Bone-marrow micronucleus test Rat liver unscheduled DNA synthesis assay Fig. I. A stepwise strategy for genotoxicity testing. material in vivo. If the material is negative in the in vivo somatic cell assays, this is considered to indicate a lack of genotoxicity in both somatic and germ-cell tissues. The justification for this is that a number of reviews (Adler and Ashby, 1989; Brusick, 1980; Holden, 1982) have shown that those chemicals reliably regarded as germ-cell mutagens in rodents are detected in in vivo somatic cell assays. In fact there are a number of somatic cell mutagens that are not germ-cell mutagens; thus germ-cell mutagens are a subset of somatic cell mutagens. Therefore, on the basis of these observations, a chemical identified as non-genotoxic in in vivo somatic cell assays will not show genotoxic activity in germ cells and further evaluation is unnecessary. This hierarchical testing strategy, based on experience with established carcinogens and germ-cell mutagens, allows the detection of both somatic cell and germ-cell genotoxins with the most effective use of resources and the minimum use of animals. For it to operate effectively, c o m m o n sense and good scientific practice must be applied. Thus testing in vitro should be to sufficient protocols for a thorough investigation, yet not to excessive stringency (e.g. marked cytotoxicity, significant p H changes, precipitating concentrations). Such situations may yield equivocal or positive responses which are of little or no relevance biologically yet may be viewed as warranting follow-up in vivo studies by a rigid application of the testing strategy. This may not be relevant. Similarly, any extension of testing outside the core strategy should be based on established precedents that it will allow an increased detection of genotoxins. For example, the value of conducting an in vitro mammalian cell gene mutation assay following negative Ames test and in vitro cytogenetic assays is not clear. For the in vivo assays, a rodent bone-marrow metaphase cytogenetic study can be considered equivalent to a rodent bone-marrow micronucleus study
and both are not therefore required on the same compound. Similarly, the existing database shows that germ-cell assays are not required unless clear evidence of somatic cell genotoxicity (e.g. bone marrow micronucleus) is found. Clearly, such a testing strategy must be applied on a case-by-case basis. However, to use this caveat as a reason to extend into assays for which there is no clear precedent for their ability to add significantly to the overall toxicological evaluation, is to undermine the progress made in the last decade while making inefficient use of resources and/or unnecessary use of animals.
REFERENCES
Adler I. D. and Ashby J. (1989) The present lack of evidence for unique rodent germ-cell mutagens. Mutation Research 212, 55~56. Ashby J. (1986) The prospects for a simplified and internationally harmonized approach to the detection of possible human carcinogens and mutagens. Mutagenesis l, 3-16. Ashby J., de Serres F. J., Draper M., lshidate M., Margolin B. H., Matter B. and Shelby M. D. (1985) Evaluation of short term tests for carcinogens. Report of the IPCS collaborative study on in vitro assays. Progress in Mutation Research 5.
Ashby J., de Serres F. J., Shelby M. D., Margolin B. H., Ishidate M. and Becking G. C. (1988) Evaluation of short term tests for carcinogens. Report of the IPCS collaborative study on in vivo assays. Cambridge University Press, Cambridge. Berry D. J. and Litchfield M. H. (1985) A review of the current regulatory requirements for mutagenicity testing. Progress in Mutation Research 5, 727 740. Bridges B. (1974) The three-tier approach to mutagenicity screening and the concept of radiation equivalent dose. Mutation Research 26, 335-340. Brusick D. (1980) Principles o f Genetic Toxicology. p. 113. Plenum Press, New York. Clayson D. B. and Grant D. L. (1993) The assessment of mutagenicity. Health protection branch mutagenicity guidelines (Canada). Environmental and Molecular Mutagenesis 21, 15-37. de Serres F. J. and Ashby J. (1981) Evaluation of short term tests for carcinogens. Report of the International Collaborative Programme. Progress in Mutation Research I, Holden H. E. (1982) Comparison of somatic and germ cell models for cytogenetic screening. Journal o f Applied Toxicology 2, 19(~200. Kirkland D. J. (1993) Genetic toxicology testing requirements: official and unofficial views from Europe. Environmental and Molecular Mutagenesis 21, 8 14. Tennant R. W. and Ashby J. (1991) Classification according to chemical structure, mutagenicity to Salmonella and level ofcarcinogenicity of a further 29 chemicals tested for carcinogenicity by the US National Toxicology Program. Mutation Research 257, 209-22?.
~
Pergamon
0887-2333(94)E0086-9
Copyright © 1994ElsevierScienceLtd Printed in Great Britain.All rights reserved 0887-2333/94$7.00+ 0.00
GENOTOXICITY TESTING STRATEGIES B. M. ELLIOTT Zeneca Central Toxicology Laboratory, Alderley Park, Macclesfield, Cheshire SKI0 4TJ, UK Abstract~enotoxicity tests contribute to the assessment of whether a chemical has the potential to cause somatic or germ-cell effects in animals (i.e. the potential to induce cancer or heritable mutation). Such genotoxicity testing is usually undertaken in a stepwise approach; first an assessment in vitro, to determine intrinsic genotoxic activity, and secondly, an evaluation in vivo to determine whether any such activity is expressed in the whole animal. While this core principle is now generally accepted, there are still different opinions as to when, and how many, in vitro or in vivo assays should be conducted. Such strategies would be simplified and harmonized if testing was carried out based on the knowledge that there was a precedent for that strategy being required to detect a carcinogen or mutagen rather than the possibility that it might.
Introduction Genotoxicity tests contribute to the toxicological assessment of whether a chemical has the potential to induce cancer or heritable mutation in animals. The principle underlying these tests is whether the chemical or a metabolite interacts with and damages DNA. There is a large number of tests available. These vary in complexity and cover a range of cell types and genetic endpoints. Until recently, in some quarters there has been an apparent comparability in the way assays as disparate as the Ames test (an in vitro bacterial assay) and the mouse specific locus assay (involving hundreds of mice) are considered to give an assessment of genotoxicity (gene mutation) (Berry and Litchfield, 1985). Such an equivalence would render testing strategies valueless. However, the suggestion by Bridges (1974), and the promotion by Ashby (1986), of a stepwise approach to genotoxicity testing has allowed the development of both effective and efficient hierarchical testing strategies to assess the likely genotoxicity of a chemical to the whole animal. This approach, together with the results of international trials designed to assess the value and reliability of the plethora of available genotoxicity assays for their ability to detect carcinogens and discriminate non-carcinogens (Ashby et al., 1985 and 1988; de Serres and Ashby, 1981), has led to the current focused testing strategies that use a limited n u m b e r of validated assays interpreted in a coherent and stepwise manner.
A stepwisegenotoxicitytestingstrategy The core principle of the current strategies generally accepted by genetic toxicologists is that an initial assessment is made using in vitro assays, which are designed to be sensitive and to detect an intrinsic genotoxic activity. If clear evidence of genotoxicity is seen in one or more of these assays, an assessment is made in vivo in order to determine whether this
intrinsic genotoxic activity is expressed in the whole animal. Since the in vivo assays are designed to assess the relevance of the in vitro results to the animal, it is recommended that the route of exposure used should be relevant for potential h u m a n exposure. To do otherwise undermines the unique role of the animal studies in this testing stategy. They must be allowed to superimpose the effects of absorption, distribution, metabolism, and excretion, together with the cellular processes such as D N A repair, onto the established intrinsic activity of the chemical. This then provides a toxicologically relevant assessment of any genotoxic effects in animals. The available data indicate that a thorough evaluation can be achieved in vitro by using the Ames test together with an in vitro cytogenetic assay in mammalian cells, and in vivo by using the bone marrow micronucleus assay and rat liver unscheduled D N A synthesis assay (Fig. 1). These provide an effective screen for genotoxic carcinogens. These principles are reflected in recent genotoxicity testing guideline updates both within (Kirkland, 1993) and outside (Clayson and Grant, 1993) Europe. In such a strategy, once genotoxicity has been determined in vitro, further in vitro assays are of limited value. What is required is the evaluation in vivo. The use of chemical structure-activity considerations can be invaluable in the interpretation of any activity seen in the in vitro assays (Tennant and Ashby, 1991). Thus an aromatic amine structure based on 4-aminobiphenyl that gives a positive response in an Ames test using Salmonella strain TA98 ( + S - 9 ) will not benefit from further in vitro evalua t i o n - i t is genotoxic in vitro; the evaluation should move immediately to assess any effects in the whole animal. There is no value in conducting further in vitro assays simply because they are present in a test strategy. The above strategy involves in vivo investigations in somatic cells and defines the genotoxicity of the test 871
872
B. M. ELLIOTT
Step I. IN VITRO ASSESSMENT FOR INTRINSIC ACTIVITY Ames test Cytogenetic assay in mammalian cells if positive:
Step 2. IN VIVO ASSESSMENT FOR EXPRESSED ACTIVITY Bone-marrow micronucleus test Rat liver unscheduled DNA synthesis assay Fig. I. A stepwise strategy for genotoxicity testing. material in vivo. If the material is negative in the in vivo somatic cell assays, this is considered to indicate a lack of genotoxicity in both somatic and germ-cell tissues. The justification for this is that a number of reviews (Adler and Ashby, 1989; Brusick, 1980; Holden, 1982) have shown that those chemicals reliably regarded as germ-cell mutagens in rodents are detected in in vivo somatic cell assays. In fact there are a number of somatic cell mutagens that are not germ-cell mutagens; thus germ-cell mutagens are a subset of somatic cell mutagens. Therefore, on the basis of these observations, a chemical identified as non-genotoxic in in vivo somatic cell assays will not show genotoxic activity in germ cells and further evaluation is unnecessary. This hierarchical testing strategy, based on experience with established carcinogens and germ-cell mutagens, allows the detection of both somatic cell and germ-cell genotoxins with the most effective use of resources and the minimum use of animals. For it to operate effectively, c o m m o n sense and good scientific practice must be applied. Thus testing in vitro should be to sufficient protocols for a thorough investigation, yet not to excessive stringency (e.g. marked cytotoxicity, significant p H changes, precipitating concentrations). Such situations may yield equivocal or positive responses which are of little or no relevance biologically yet may be viewed as warranting follow-up in vivo studies by a rigid application of the testing strategy. This may not be relevant. Similarly, any extension of testing outside the core strategy should be based on established precedents that it will allow an increased detection of genotoxins. For example, the value of conducting an in vitro mammalian cell gene mutation assay following negative Ames test and in vitro cytogenetic assays is not clear. For the in vivo assays, a rodent bone-marrow metaphase cytogenetic study can be considered equivalent to a rodent bone-marrow micronucleus study
and both are not therefore required on the same compound. Similarly, the existing database shows that germ-cell assays are not required unless clear evidence of somatic cell genotoxicity (e.g. bone marrow micronucleus) is found. Clearly, such a testing strategy must be applied on a case-by-case basis. However, to use this caveat as a reason to extend into assays for which there is no clear precedent for their ability to add significantly to the overall toxicological evaluation, is to undermine the progress made in the last decade while making inefficient use of resources and/or unnecessary use of animals.
REFERENCES
Adler I. D. and Ashby J. (1989) The present lack of evidence for unique rodent germ-cell mutagens. Mutation Research 212, 55~56. Ashby J. (1986) The prospects for a simplified and internationally harmonized approach to the detection of possible human carcinogens and mutagens. Mutagenesis l, 3-16. Ashby J., de Serres F. J., Draper M., lshidate M., Margolin B. H., Matter B. and Shelby M. D. (1985) Evaluation of short term tests for carcinogens. Report of the IPCS collaborative study on in vitro assays. Progress in Mutation Research 5.
Ashby J., de Serres F. J., Shelby M. D., Margolin B. H., Ishidate M. and Becking G. C. (1988) Evaluation of short term tests for carcinogens. Report of the IPCS collaborative study on in vivo assays. Cambridge University Press, Cambridge. Berry D. J. and Litchfield M. H. (1985) A review of the current regulatory requirements for mutagenicity testing. Progress in Mutation Research 5, 727 740. Bridges B. (1974) The three-tier approach to mutagenicity screening and the concept of radiation equivalent dose. Mutation Research 26, 335-340. Brusick D. (1980) Principles o f Genetic Toxicology. p. 113. Plenum Press, New York. Clayson D. B. and Grant D. L. (1993) The assessment of mutagenicity. Health protection branch mutagenicity guidelines (Canada). Environmental and Molecular Mutagenesis 21, 15-37. de Serres F. J. and Ashby J. (1981) Evaluation of short term tests for carcinogens. Report of the International Collaborative Programme. Progress in Mutation Research I, Holden H. E. (1982) Comparison of somatic and germ cell models for cytogenetic screening. Journal o f Applied Toxicology 2, 19(~200. Kirkland D. J. (1993) Genetic toxicology testing requirements: official and unofficial views from Europe. Environmental and Molecular Mutagenesis 21, 8 14. Tennant R. W. and Ashby J. (1991) Classification according to chemical structure, mutagenicity to Salmonella and level ofcarcinogenicity of a further 29 chemicals tested for carcinogenicity by the US National Toxicology Program. Mutation Research 257, 209-22?.