New Trends in Genetic Risk Assessment

New Trends in Genetic Risk Assessment is based on the Fifth International Round Table of the Rhône-Poulenc Santé Foundation, held in Nice 1987. The conference was an attempt to review the latest theories and mechanisms stipulated for the various aspects of genotoxicity; it was above all an open forum to discuss the new trends, the new tests, and the new battery of tests for assessing the genetic risk of chemicals and especially drugs. This volume is actually not a proceedings of the meeting but a monograph specially edited to report the reviews which were presented and the discussions which took place; it was designed to provide a better understanding of the knowledge obtained in the most recent years and to help in the practical choice of approaches or tests for the prediction of the various forms of genotoxicity. The main subjects of this review concern molecular analysis of mutagenesis, detection of DNA damage, gene mutation, clastogenesis, aneuploidy, and germ cells. Each theme is preceded by a short overview summarizing the state of the art and the contributions of each author. Finally, special attention was given to the personal views of some leading toxicologists as to the battery of tests presently available or recommended.

• Language: English
• Publisher: Academic Press
• Release date: Oct 22, 2013
• ISBN: 9781483265674

Book preview

Cordier

Introduction

1

The Role of Genotoxicity in Drug Safety Evaluation

E. MOUSTACCHI,     Institut Curie, Biologie, 26 rue d’UInn, 75231 Paris Cedex 05, France

Publisher Summary

Health hazards from mutational events involve somatic as well as germinal tissues and can lead to various somatic diseases, to teratogenic effects, and to heritable disorders. Although the direct evidence of mutational origins of somatic diseases in man is limited, inferences from experimental results in other organisms give strong support to the view that some diseases are indeed originated by somatic mutations. Although the direct evidence of mutational origins of somatic diseases in man is limited, inferences from experimental results in other organisms give strong support to the view that some diseases are indeed originated by somatic mutations. The relationship between such mutations and cancer is almost undisputed and has received strong support from the recent analysis of oncogene activation, which indicates that the specific alterations of DNA and chromosome rearrangements are intimately involved with the carcinogenic process. Mutational processes and genetic end-points revealed by standard short-term assays cover only a fraction of the series of genetic and/or epigenetic alterations involved in carcinogenesis.

The Rhône-Poulenc Santé Company is to be congratulated for having chosen an interesting subject for this conference and for convening it at a very timely period.

It is generally assumed that health hazards from mutational events involve somatic as well as germinal tissues and can lead to various somatic diseases, to teratogenic effects and to heritable disorders. Although direct evidence of mutational origins of somatic diseases in man is limited, inferences from experimental results in other organisms give strong support to the view that some diseases are indeed originated by somatic mutations. The relationship between such mutations and cancer is almost undisputed and received strong support from recent analysis of oncogene activation which indicated that specific alterations of DNA and chromosome rearrangements are intimately involved with the carcinogenic process. According to estimations of Doll and Peto (1981) from epidemiological data, at least 75% of cancer cases in the USA are caused by environmental factors, especially nutritional habits, and consequently might be prevented by identification of mutagenic agents in the environment and by avoiding human exposure to such agents.

The development and use during the 1970s of short-term test systems for the screening of chemicals for mutagenic activity before they are tested in long-term animal assays have been of great importance in this context. Instead of determining production of cancer or of genetic anomalies, these short-term tests measure a variety of biological end-points that may be related to postulated mechanisms of carcinogenesis. These tests are valuable for their rapidity and low cost, and their use of diverse organisms ranging from bacteriophages and bacteria to whole animals. However, the use of short-term assays to assess human genetic risk entails a number of difficulties which have led to much discussion of how to apply and interpret them, and the predictive value of short-term assays has been questioned.

In this short introduction, my purpose is twofold. Firstly I would like to summarize briefly the conceptual difficulties encountered in the use of the test systems, over 100 of which are presently available. Secondly, it seems to be urgent to introduce some notions, derived from the new developments in our understanding of the molecular aspects of the carcinogenic processes, into the discussion of the end-points chosen in short-term testing. Actually, our aim would possibly be to open new strategies for the future development of genetic toxicology in general.

Before summarizing the difficulties encountered in short-term test interpretations, I would like to recall the sequence of events from the induction of lesions to the genetic end-points analysed (Fig. 1.1). It can be seen that the production of lesions in DNA can be modulated by a number of cellular factors, including the enzymatic metabolic activation of those compounds which do not act directly on DNA, detoxification, and the presence or absence of oxygen. Moreover, several genetic and physiological parameters can then interfere with the processing of the DNA lesions. Among these parameters the position of cells in the cycle, the role of repair mechanisms, the degree of ploidy and the structure of chromatin, etc., have been identified as playing important roles in the final expression of damage. The end-points examined in short-term tests include point mutations (forward and reverse mutations), all types of chromosomal and chromatid alterations (breaks, gaps, translocations, micronuclei, etc.), recombination (gene conversion, crossing-over, sister chromatid exchanges), cellular transformation in vitro, non-disjunction, etc. The lethal effect of compounds (cytotoxicity) is generally assessed in conjunction with genetic effects detected among survivors.

Fig. 1.1 Schematic representation of the sequence of events taking place between exposure to physical or chemical agents which interact with DNA and expression of damage.

Let us now examine the difficulties encountered in the use of short-term tests. Although they are familiar to all of you I think that it may be useful to summarize them clearly again.

1) The use of short-term assays for prediction of human carcinogenicity comprises an extrapolation across the whole series of events in carcinogenesis. In practice, it implies the comparison of mutagenicity with animal carcinogenicity, while animal carcinogenicity is already an approximation to measure human carcinogenicity. This correlation between animal and human is not obvious. Species and tissue differences in susceptibility related to the series of events preceeding induction of mutations (i.e., biotransformation with the formation of reactive metabolites, reaction of such metabolites with DNA, structure of the chromatin, DNA repair or persistence of lesions, etc.) have been widely documented in recent years. Moreover, after induction of mutations, the modulations in cell selection controlled by immunological, endocrinological and growth factors are likely to play a role in animal carcinogenicity.

The direct measurements of mutagenicity and clastogenicity in human T-lymphocytes, as will be discussed in Part III, constitute interesting attempts to overcome this difficulty in extrapolation.

2) The high correlation between the Ames Salmonella/microsomal assay and animal cancer data reported about ten years ago has not been fully substantiated for all groups of chemicals. Recent cancer data by the National Toxicology Program in the United States has comprised more randomly selected chemicals than the ones used in previous comparisons with Salmonella and other tests, and notable deviations from previously determined correlations between mutagenicity and carcinogenicity have been revealed for certain chemicals or groups of chemicals (Zeiger and Tennant, 1986).

3) Although the Ames Salmonella/microsomal assay has been of great importance for various aspects of mutagenicity testing, it is widely recognized that a single assay cannot constitute an adequate reflection of mutagenic alterations occuring in mammalian cells in vivo (see for discussion Ashby, 1986). Therefore various batteries of in vitro and in vivo test systems have been recommended. The addition of test systems, however, implies the problem of interpreting contradictory test results. This has usually been done in a more or less subjective way. It is clear that a more systematic and objective use of the acquired knowledge from short-term testing with different assay systems and different types of chemicals should provide a better basis for risk evaluation. The systematic collection and evaluation of available mutagenicity data, particularly through the Gene Toxprogram, have opened up new possibilities of applying modern computer analysis to make a more efficient use of this massive background experience. The most advanced computer system specifically designed for this purpose will be described in Chapter 5 of this volume (see also Ennever and Rosenkranz, 1987).

4) The fact that there is, grossly speaking, a qualitative correlation between mutagenicity in short-term tests and carcinogenity may be related to an association between different effects on DNA rather than to a clear-cut causative correlation. Examination of not only the qualitative but also the quantitative relationship between short-term data on mutagenicity and animal cancer bioassay data is essential not only for the mechanistic understanding of the overall role of mutagenesis in the carcinogenic process but also for practical purposes and legislative decisions about a given compound. Such quantitative estimates are generally missing, and in some cases for which quantitative comparisons of chemicals were made there was a lack of association between potency in mutagenesis and cancer induction (see for instance Bartsch et al., 1983). Moreover, the standard procedures with mammalian cells has been generally employed to measure effects due to acute doses of mutagens. Assessment of the effects of low doses, which actually correspond to the usual human exposure, is then carried out by means of extrapolation, the validity of which has been debated.

5) In short-term tests, negative results in certain assays are often due to the fact that some carcinogens do not cause mutations in prokaryotes and eukaryotes. Such substances include arsenic, asbestos, diethylstilboestrol and chlorinated compounds, all of which are classified on the basis of epidemiological studies as human carcinogens. In other words, certain cancers (skin cancer in humans exposed to inorganic arsenic, or cancer of the vagina in females exposed in utero to the synthetic oestrogen hormone diethylstilboestrol, etc.) may arise by mechanisms other than gene mutations or chromosomal rearrangements.

All of these points lead to the concept that mutational changes constitute only part of the complex sequence of events in carcinogenesis. In recent years, the analysis of viral induction of cancer, of oncogene activation, and of hereditary forms of malignancy has revealed a number of points of importance which should be stressed at this point.

1) It is clear that alterations at the level of DNA (point mutations) are important in carcinogenesis.

2) In addition, almost all conceivable types of chromosomal alterations seem to be involved in carcinogenesis. The batteries of short-term tests take into account these two end-points.

3) Tumour formation usually implies genetic alterations not only at the initiation stage but also various chromosomal alterations and other mutational events at later stages. It is not known whether these events occur independently of each other, whether they are triggered by a common mechanism, or whether they constitute a cascade of interrelated events.

4) Genetic alterations associated with carcinogenesis may exhibit a frequency far higher than could be expected on the basis of established mutation rates, as in, for instance, the remarkably high frequency of cancer in the offspring of male mice treated with X-irradiation and chemicals as reported by Nomura (1986). In this context, it is of interest to note that by using somatic cell hybrids containing a single human chromosome carrying gene markers, and by using doses of mutagenic agents so low that little cell killing occurs, the true efficiency of mutagenesis for low doses of ionizing radiation is more than 200 times greater than that obtained by conventional methods (Waldren et al., 1986).

In conclusion of the first part of this chapter, it appears that mutational processes and genetic end-points revealed by standard short-term assays cover only a fraction of the series of genetic and/or epigenetic alterations involved in carcinogenesis. What are the other conceivable possibilities which are not sufficiently explored by the actual test-systems?

1) The discovery of oncogenes and the molecular analysis of their activated forms are of primary importance. Indeed, the induction of rat mammary tumours by nitrosomethylurea is associated in more than 80% of the tumours with a point mutation in the H-ras gene at the codon for the twelfth amino acid. This has been attributed to the formation of O⁶-methylguanine which in turn causes transition of the G:C pair to A:T (Sukumar et al., 1983). More recently, Vousden et al. (1986) demonstrated that in vitro modification of plasmids containing the c-Ha-rasI protooncogene by ultimate carcinogens generates a transforming oncogene by point mutations at the twelfth or sixty-first codon. It is of interest that mutations uncommon in bacterial systems (i.e, G:C-to-C:G and A:T-to-C:G transversions) were detected in activated oncogenes, suggesting that the information obtained from mutagenesis studies in prokaryotes is not totally applicable to human systems. I feel that activation of oncogenes will be used as an end-point of short-term assays in the not-too-distant future. Along the same line of thought, it is clear that the development of shuttle vectors containing mammalian genes and allowing molecular analysis — as will be discussed here by Lehman et al. (Chapter 4), or as they are studied in France by Sarasin — correspond to this same general preoccupation.

2) The activation of ras oncogenes is essentially associated with a qualitative mutational change of the gene product. Other forms of oncogene activation involve a quantitative change of gene expression which has been shown to occur by at least four different mechanisms: (a) integration of a retroviral transcriptional promoter which replaces the normal promoter; (b) increase of oncogene transcription by means of enhancers; (c) amplification of c-oncogenes; (d) chromosomal rearrangements that place an oncogene in the vicinity of an immunoglobulin gene, the promoter of which increases transcription of the oncogene.

Amplification of oncogenes seems to be a predominant mechanism, and recent experimental data indicate that amplification — especially following exposure to physical and chemical agents — is a much more widespread phenomenon than initially thought. The best-studied manifestation of amplification comes from cytological observations of homogeneously stained regions and double minutes which are connected with resistance to drugs. (It should be pointed out in the present context that various neoplasms often exhibit homogeneously stained regions and double minutes.) Moreover, agents — including carcinogens — which act on DNA synthesis and the nucleotide pool have been found to enhance the frequency of amplification, as indicated by the induction of methotrexate-resistant cells. A synergistic effect of these agents and the tumour promoter TPA on amplification of the dihydrofolate reductase gene has furthermore been shown. The function of gene amplification includes transposition and insertion of unstable genetic elements: amplification in situ, release of amplified DNA, and transposition of such released sequences to new sites has been recorded in specific cases. The two major models accounting for gene amplification — i.e., the unequal exchange of repeated sequences of DNA with the formation of circular episomal elements which could be transposed, and the onion-skin model (i.e. multiple rounds of synthesis of DNA sequences) — both require recombination events.

It is clear that future developments in our understanding of the gene amplification and transposition processes in response to environmental injury will have an impact on genetic toxicology.

3) Chemicals may act by causing an imbalance of the nucleotide pool, and such imbalances can have profound effects on the genome. Haynes and Kunz (1985) have pointed out that essentially all known forms of genetic alterations, from point mutations to oncogenic transformation, have been observed subsequent to induced disturbance of the deoxyribonucleotide triphosphate pool in appropriate in vivo assay systems. While some of these effects could be due to cell selection in connection with cell cycle parameters, it is clear that this is not the case for all of them.

4) There are several reasons to suspect that recombinogenic events are fundamental in carcinogenicity. It has already been emphasized above that amplification and transcription require some form of genetic recombination. On the other hand, genes responsible for hereditary cancer forms, such as retinoblastoma or Wilm’s tumour formation etc., function as recessive genes, and cell fusion experiments are in favour of a recessive nature of oncogenes. In other words, homozygosity appears to be required for expression of neoplastic transformation, and it is easy to visualize the importance of genetic recombination in reaching the homozygous state. With the exception of induction of sister chromatid exchange in mammalian cells in vitro and induction of gene conversion in yeast or in Drosophila, which have been used as standard indications of genotoxicity in general, there is a fundamental lack of systematic data on chemical induction of recombination in higher organisms. I guess that the development of a new bacterial test, as done recently by Radman’s group, in which mutations, SOS function and recombination are measured concomitantly will be very useful.

5) One end-point that may very well be of great importance to both tumour formation and, more generally, to genetic diseases in man is non-disjunction. There is a great need for validation of short-term methods for detecting non-disjunction; Chapters 15 and 16 will summarize what can be expected in this area in the future.

6) Finally, special attention will be given to germ cells. It is not necessary to elaborate here on the importance in genetic toxicology of such studies; obviously there are still enormous gaps in our knowledge of mutation rates in these target tissues.

Restriction fragment length polymorphism of DNA, DNA sequencing, development of transfection assays, gradient denaturation gels, heteroduplex DNA, subtractive hybridization with synthetic oligonucleotides, and the use of RNase A to cleave C: A mismatches in RNA:DNA heteroduplexes have been identified as methods for possible practical use in monitoring human populations for mutation rates. On the other hand, the efforts which are actually developed for the detection of DNA lesions, making use of new and very sensitive methods, merit following with attention. For practical screening of mutagens, these new techniques will be assimilated only gradually into testing protocols. Our wish is that a permanent interaction between scientists will contribute to the opening of these new avenues in genetic toxicology.

References

Ashby, J. Mutagenesis. 1986; 1:3–16.

Bartsch, H., Terracini, R., Malaveille, C., Tomatis, L., Warendorf, J., Brunn, G., Dodet, B. Mutation Res. 1983; 110:181–219.

Doll, R., Peto, R. J. Natl. Cancer Inst. 1981; 66:1192–1265.

Ennever, F.K., Rosenkranz, H.S. Mutagenesis. 1987; 2:39–44.

Haynes, R.H., Kunz, B.A. Muhammed A., Von Borstel R.C., eds. Basic and applied Mutagenesis. Plenum Publishing Corp.: New York, 1985; 147–156.

Nomura, T. Genetic Toxicology of Environmental Chemicals. In: Ramel C., Lambert B., Magnusson J., eds. Genetic Effects and Applied Mutagenesis. New York: Alan R. Liss; 1986:13–20. part B

Sukumar, S., Notario, V., Martin-Zanca, D., Barbacid, M. Nature. 1983; 306:658–661.

Vousden, K.H., Bos, J.L., Marshall, C.J., Phillips, D.H. Proc. Natl. Acad. Sci. USA. 1986; 83:1222–1226.

Waldren, C., Correll, L., Sognier, M.A., Puck, T.T. Proc. Natl. Acad. Sci. USA. 1986; 83:4839–4843.

Zeiger, E., Tennant, R.W. Genetic Toxicology of Environmental Chemicals. In: Ramel C., Lambert B., Magnusson J., eds. Genetic Effects and Applied Mutagenesis. New York: Alan R. Liss; 1986:75–84. [part B].

Discussion

A. Leonard

Dr Moustacchi said that the water-soluble salts of arsenic are known to produce cancer in humans but are devoid of carcinogenic properties in animals. In fact, I do not share that opinion, partly because we are unable to discover the mutagenic properties in short-term tests, but mainly because water-soluble arsenic salts are promoters and probably act predominantly by inhibition of DNA repair. The distribution of cancers produced by water-soluble arsenic salts shows that they occur mainly in countries where there is a lot of UV irradiation — sunny countries and so on — and that the cancers produced are melanomas. It is possible that for such cancers these salts play a role by inhibition of the repair of UV damage produced in DNA.

E. Moustacchi

Obviously, I did not have time to go into such details. It is a matter of the definition of initiatior and promoter.

H. Tuchmann-Duplessis

Dr Moustacchi gave us a remarkable overview of the problem. She mentioned early in her presentation the modulation system, and at the end pointed out that information on mutation of germ cells could be very useful. Can she comment slightly more? There is the feeling that the modulation system in germ cells is remarkable, and is a kind of protection. Therefore, it is not very likely that heritable mutation will occur, especially in humans.

E. Moustacchi

We say that human oocytes are generally more resistant than mouse — than animal — oocytes. However, the fact that they are more resistant, which means that they will escape the cytotoxic effect of a given drug, means that they can afford to have mutations. There is thus a balance between cytotoxicity and mutation. Surely, it is clear that animals and humans are not the same in terms of oocyte response.

F. Roe

Thank you for an absolutely beautiful dissection of one half of the problem. The other half of the problem is that there needs to be an exactly comparable dissection of the animal tumours that occur in carcinogenicity studies. This is because for the most part they are obviously not models of human cancer. In rats, 90% or more are examples of endocrine disturbance. These arise out of a background of hyperplasia, and there is no evidence of any genetic damage at the hyperplasia stage. Any genetic damage which occurs in these lesions occurs late. This means that everything Dr Moustacchi has said on this initiation/promotion basis does not seem to be relevant to what is actually seen under the microscope with the rat