Document v6obDdpkXnLN0vLpqrJ0y12bw

How do in vivo mammalian assays compare to in vitro assays in their ability m to detect mutagens? iJ A.T. Natarajan12 and G. Obe 3 'in- 1 Department of Radiation Genetics and Chemical Mutagenesis, State University of Leiden, Sylvius Laboratories. Wassenaarseueg 72. 2333 A L Leiden tThe Netherlands), ' J A. Cohen Institute, Inieruntoersity Institute for Radtopathology and Radiation Protection (IRS). Leiden (The Netherlands), and1 Institute of Genetics, Free University, Arnimallee } - 7. Berlin (West) If.R.G.) of (Received 12 October 1984) mg (Revision received 1 October 198S) 'ICS (Accepted 14 October 198}) lUS len 13 Mutations are detected in organisms by their activation system in in vitro studies (Natarajan et I to phenotypic expression. Depending upon the com al,, 1976). This, however, cannot always fully re _nl ;P3 plexity of the organization of the genetic material, place an in vivo situation, and thus limits detec the primary change leading to a mutation can tion of certain classes of chemicals in vitro. In vary. In prokaryotic systems, such as Salmonella such cases in vivo tests would be appropriate. of and Escherichia colt, single base-pair changes or However, a successful mammalian in vivo test additions and deletions of single bases in the DNA depends on the ability of the active metabolites of can lead to mutations. However, large deletions the mutagen to reach the target cells under study. and duplications cannot be tolerated by these Thus, mutagens with short-lived metabolites organisms, as unrepaired or misrepaired breaks in though very active, may not come out as positive DNA lead to lethality. In eukaryotic systems on in such in vivo tests. the contrary, in which the genetic material is Bacterial tests can be of predictive value for the icjI organized in the form of separate chromosomes abHity of a compound to induce point mutations which contain a complex of DNA with proteins, in but a chemical which predominantly induces DNA ..ed an\ addition to mutations resulting from single base- strand breaks and thus chromosomal aberrations her* pair changes similar to bacteria, those due to small will not be detected in this test easily. Since chro wer 000 and large chromosomal rearrangements can be mosomal aberrations per se contribute to a signifi detected. This basic difference between pro cant proportion of detectable genetic damage in karyotes and eukaryotes obviously makes a unified human it is important that the test systems (bat evo any definition of mutations impossible and makes di tery) employed should include the ability to detect by rect comparisons and extrapolations difficult. chromosomal aberrations. Information is also uch US While most of the directly acting mutagens can needed with respect to tissue-specific responses of ni) readily be detected in both in vitro and in vivo test mutagens in germinal vs. somatic cells. In several systems, it is difficult to detect mutagens which instances a potent mutagen may turn out to be require metabolic activation. This, for instance, negative in a mammal in an assay measuring can be achieved by employing a mammalian liver genetic damage in germ cells simply because the active metabolite does not reach the germ cells due Send nil correspondence to A.T. Natarajan. to the existence of the blood-testis barrier. Though such a chemical may not pose a hazard for future Dedicated to Dr. M.S Swaminathan on the occasion of his 60th birthday. generations it might induce changes in the somatic cells of the exposed individual, leading to somatic 0165-H10/86/S03.50 1986 Elsevier Science Publishers B,V. (Biomedical Division) ^ vVAANING: Ibis material iwf h* <-*( I 190 mutations (cancer). In such cases the mouse spot test can be useful. Accumulating evidence points towards an involvement of specific chromosomal rearrangements in the etiology of different types of cancers in human, such as lymphomas, leukemias. Wilm's tumor etc. (Rawley. 1983). In this review, we have compared the mam malian in vivo test systems against in vitro test systems utilizing pro- and eukaryotic cells. We have described briefly the test systems employed and the genetic end points they can detect. All the chemicals which have been tested in a mammalian in vivo germ cell test, namely, specific locus muta tions, dominant lethals and heritable translo cations, have been compared with regard to their ability to induce genetic changes in in vitro tests. Since the number of chemicals tested with germ cells in mammals is very limited, we have further compared the chemicals which have been studied in the mouse micronucleus test with their ability to induce mutations in bacterial assays. These limited data point towards some of the advantages and limitations of an in vivo mammalian mutagenicity test. In the literature several attempts have been made to compare the results from bacterial tests with rodent tests as well as between rodent tests (Hollstein et al., 1979; de Senes and Ashby, 1981; Jenssen and Ramcl, 1980; IARC, 1980; L.B. Rus sell and Matter, 1980). In another study the results of mammalian male germ cell assays were com pared with those from different shon term assays (ICPEMC Committee 1, 1983). We have used the information from these reports as well as the infor mation generated by different Gene-Tox commit tees. However, for the comparative study reported here, it was difficult to use Gene-Tox reports as several of the committees involved have rejected a majority of published papers in this field as they did not fulfil the strict criteria used to evaluate the results. We have tried to group the tested agents into chemical classes in order to check whether we can detect any definite trends in a preferential ability of the mammalian in vivo test systems to detect (or miss) certain classes of chemicals in comparison to in vitro systems. For this purpose we have used the classification of chemicals adapted by Rinkus and Legator (1979) as well as Vogel et al. (1980). Validated in vitro mutagenicity tests (]) Bacteria The standard bacterial assay to detect mutagens is the Salmonella typhimurium test using reverse mutations in the histidine locus developed by B.N. Ames and his colleagues (Levin et al., 1982). Though the Salmonella test is the most universally used, similar reversion tests using Escherichia coli have also been developed (Brusick et al.. 1980; Mohn et al., 1984). The types of genetic damage detected by the different strains employed in these tests are point mutations such as transition and transversion of single DNA bases and frameshift mutations. In addition there are bacterial repair tests, which are based on the inability of special repair-defective tester strains in repairing damaged DNA, thus leading to killing or to reduced pro liferation in comparison to wild-type strains. This test detects all types of DNA damage. Such sensi tive DNA repair tests have been developed for E. coli and Bacillus subtihs. Most classes of mutagens tested with these strains (E. coli, S. typhimurium) come out positive in the repair test with few excep tions (De Flora et al., 1984). Aromatic amines and polycyclic aromatic hydrocarbons can be detected in the standard Salmonella assay with plasmidcontaining strains (e.g,, TA98) but are not effi ciently detected in most of the other bacterial Systems. Procarbazine and safrole are identified as mutagens in E. coli but not in Salmonella stan dard plate test. (2) Yeast Unlike prokaryotic cells such as bacteria, yeast represents an eukaryotic system. Eukaryotes have a nuclear membrane and chromosomes with the DNA associated in a typical nucleosome structure. Therefore, it is possible to detect chromosomal aberrations in addition to point mutations in this system. Other types of genetical alteration which can be detected with yeast are mitotic recombina tion and ancuploidy resulting from mitotic non disjunction or chromosome breaks. In addition, growing yeast cells have some capacity to activate mutagens because they have the cytochrome P-450 system operating. <5l^?82 gens erse B N. >82). -ally coli 980; iage hese and -hift pair xial iged 3TOrhis nsi' E. ;ens im) ep- iidffirial I as an- ast ave the ire. nal his ich la>nw, ite 50 13) In citro cytogenetics (3.1) Chromosomal aberrations Since most mutagens are also known to induce chromosomal aberrations, the ability of a com pound to induce chromosomal aberrations has been used as an index of mutagenic potential of an agent (Natarajan and Obe. 1982). Chromosomal aberrations can be detected in different types of proliferating cells of plant and animal origin. However, for routine screening mammalian cells are employed, such as established Chinese hamster cells or stimulated human peripheral lymphocytes. The types and the frequencies of aberrations in duced depend on the cell cycle stage treated and the mutagen in question. Agents which directly induce DNA double-strand breaks, such as X-rays and bleomycin induce chromosome type of aberra tions in G|, chromatid type in G, and both types in the S phase of the cell cycle. These agents are considered as S-independent agents. On the other hand mutagens such as alkylating agents and UV light induce chromatid type of aberrations irre spective of the cell cycle stage treated and the cells have to go through an S phase before the aberra tions can be visualized. These agents are classified as S-dependent agents. (3.2) Sister-chromattd exchanges (SCEs) SCEs are the outcome of exchanges of repli cated DNA at homologous loci, a process involv ing DNA double-strand breaks and rejoining of such breaks between sister chromatids. Sister-chro matid differentiation to detect SCEs can be achieved by (a) labeling chromosomes with tritiated thymidine for one cycle followed by growing the cells for another cycle in cold thymidine, and (b) by growing cells for the first or the first and second cell cycles in a medium containing 5-bromodeoxyuridine (BrdUrd). The baseline frequency of SCEs using the BrdUrd method varies between 5 and 10 per cell, while the baseline frequency of chromosomal aberrations is between 0.0 and 2.0 per 100 cells (Vogel and Natarajan, 1982; Natarajan et al., 1984). The majority of the baseline frequencies of SCEs are most probably induced by incorporated BrdUrd. Incorporated BrdUrd itself is known to be mutagenic and there fore an evaluation of a chemical using the SCE test 191 in reality reflects an additive or synergistic effect of two agents (Natarajan et al.. 1984). The SCE formation is an S-dependent event and therefore all agents which require an S phase to induce chromosomal aberrations are efficient inducers of SCEs. More than 300 chemicals have been tested for their ability to induce SCEs (Abe and Sasaki, 1982). Chemicals which induce DNA strand breaks directly are not potent inducers of SCEs. On the other hand, with very few exceptions (3 aminobenzamide and other inhibitors of poly(ADPribosc) synthetase) agents which induce SCEs do also induce chromosomal aberrations (Natarajan and Czukas, 1984; Natarajan et al., 1984). (4) Point mutations in mammalian cells in culture Several systems are available to detect point mutations in mammalian cells. These are usually characterized by resistance to selective agents such as 8-azaguanine, 6-thioguanine (hypoxanthineguanine phosphoribosyi transferase), ouabain (Na+-K+-ATPase), bromodeoxyuridine and trifluorothymidine (thymidine kinase). Ouabain re sistance is a dominant mutation and occurs spon taneously at very low frequency (10-8). HGPRT mutations on the other hand occur at much higher frequencies (10~4). Cytological analyses of HGPRT mutations induced by ionizing radiations, ethyl methancsulfonate as well as those occurring spontaneously indicate that part of the mutations are associated with chromosomal rearrangements involving the X-chromosome. This indicates that HGPRT mutations can arise as point mutations as well as chromosomal aberrations. A similar situa tion seems to be true for the thymidine kinase locus as well. (5) DNA repair tests The basic method is to detect incorporation of tritiated thymidine in the nucleus of the mutagentreated cells which is not in the S phase of its cell cycle. Such an incorporation reflects the amount of repair of damaged DNA and this synthesis is called unscheduled DNA synthesis (UDS), as it occurs outside the main replicative DNA synthe sis. UDS can be measured by nncroautoradiography or by scintillation counting. For this test it is OLI 3383 easier to use non-proliferating cells such as human lymphocytes and rodent hepatocytes. UDS does not reflect a mutagenic event per se, but rather a repairable damage to the DNA. There are several other in vitro tests which are being used in differ ent laboratories, but the tests discussed above are the most frequently used and relatively well vali dated ones. (6) Drosophila melanogasier: Sex-linked recessive lethal mutations Next to the bacterial tests, only in the Drosophila sex-linked recessive lethal test (SLRL) a large number of chemicals have been tested for mutagenicity. The capacity of these flies to activate indirect mutagens like mammals and man makes this system very attractive. The SLRL test mea sures mutations in about 700 loci and these muta tions represent a mixture of point mutations and deletions, the relative proportion of which depends on the mutagen under test. A somatic mutation/mitotic recombination as say (SMART) has been developed recently, by which a variety of genetic events can be evaluated, by measuring the frequencies of single and twin spots in the eye (Vogel, 1985). This assay system appears to be very sensitive and less time consum ing in comparison to the SLRL test. (7) Plant systems Several assay systems are available in plants. These include chromosomal aberrations and SCEs (Vicia faba, Allium cepa). point mutations (Glycine mexicana, Zea mays) as well as somatic recombi nation (Glycine). Though detection of chro mosomal alterations is very easy and quick in plant systems, the detection of point mutations is more difficult and time consuming. For this reason this system has not been introduced in large-scale mutagenicity testing of chemicals. However, in cases where metabolic activation of chemicals mediated through plants is essential (atrazine, maleic hydrazide) plant systems become im portant. In view of the extensive cytological stud ies from the early 1930s, a large number of chemicals has been tested using chromosomal aberrations in plants. Already in a review from 1962, Rieger and Michaehs have reported 218 chemicals tested using root tip menstems of Vicia faba. A very sensitive test to detect low exposures to mutagens has been developed using Tradescantia stamina) hairs, but large-scale testing data are lacking. (8) Mammalian in vivo mutation-detecting assays With the in vivo mutagenesis tests available in mammals, both point mutations (recessive and dominant) and chromosomal aberrations can be detected in somatic as well as germ cells. (8.1) Mouse specific locus test The test involves detection of mutations trans mitted from treated germ cells and recovered in the offspring of the next generation. This test has been developed by W.L. Russell (1951) using coat color markers at 7 different loci, which can be screened simultaneously. Though these mutations are generally considered as point mutations, it is recognized that many of these mutations arise due to deletions of single or multilocus nature. About 30 chemicals have been evaluated with this test. All these tests have been made with male germ cells. Only 2 chemicals have been tested so far using oocytes (L.B. Russell et al,, 1981). A selected list of 10 environmental chemicals tested failed to increase the frequencies of mutations in the male germ cells (W.L. Russell, 1984). (8.2) Mouse spot test This is an in vivo somatic mutation test which was originally developed by L.B. Russell and Major (1957) and later propagated by Fahrig (1978). The test detects the expression of recessive markers which are used as indicators in the specific locus test. In principle, heterozygotes are treated in utero and mutant spots on the skin of the newborn mice are scored. The spots can arise due to different types of genetic events, such as point mutations, duplications, deletions, non-disjunction or somatic recombination. In this test, in addition to mu tations, embryonic and teratogenic effects of test chemicals can be evaluated. About 60 chemicals have been tested using this test so far. orted 218 is of 1 u iu exposures ig Trades,'siing data g assu\ s mailable in esstve and >ns can be tions transrcovered in his test has i using coat ich can be ,, mutations ations. it is ns arise due |e. About "this test, male germ sled so far i. A selected ed failed to in the male n test which .11 and Major : (1978). The -ive markers oecific locus ated in utero rwbom mice to different t mutations, n or somatic tion to mu'fects of test '0 chemicals (8.3) Dominant lethal test The dominant lethal mutation test can be used as a quick in vivo test and represents a mutation in a gamete (egg or sperm) which is lethal to the zygote produced by these gametes. Dominant lethals reduce the litter size due to failure of the fertilized egg to implant or to develop after im plantation. It is believed that most of the domi nant lethals arise due to chromosomal breakage or loss. About 100 chemicals have been evaluated using this test (IARC. 1980). A compilation of 140 chemicals tested for induction of dominant lethals, has recently become available in the literature (Green et al.. 1985) and these results are also included in our comparative tables. (8.4) Heritable translocation test The heritable translocation test is designed to detect partial or full sterility among first-genera tion male offspring of mutagen-treated mice. The partial sterility can be caused by the animal being heterozygous for a reciprocal translocation. This test also gives an estimate of transmissibility of induced chromosomal exchanges, if this test is done parallel to a study directed to estimate the induced translocations in male germ cells. (8.5) In uwo chromosomal aberrations Chromosomal aberrations induced in an animal can be evaluated in all proliferating cells. Bone marrow cells as well as spermatogonial cells are usually studied. When induced breaks are not in cluded in the daughter nucleus the fragments usu ally form a small nucleus, termed as micronucleus appearing in the cytoplasm separated from the main nucleus. In the bone marrow, micronuclei formed in the erythroblasts are retained during maturation in contrast to the main nuclei which are extruded from the cells. Thus, in polychro matic erythrocytes, micronuclei can be recognized and scored. Because of the ease with which it can be performed, this test is widely used. All the cytogenetic tests involving bone marrow and germ cells can be done in one animal if an appropriate protocol is used (Tates and Natarajan, 1976). Be cause aberrations induced by chemicals are usually of chromatid type and the proportion of reciprocal translocations which are transmitted to further cell generations is rather low, it is difficult to get high frequencies of reciprocal translocations detectable at the diakinesis of metaphase I stage of the spermatocytes. In routine testing bone marrow chromosomal aberrations and micronuclei are usu ally employed. A liver micronucleus assay has been developed by Tates et al. (1983), in which the treated animals are partially hepatectomized and the hepatocytes from the regenerating liver are analyzed for the presence of micronuclet; so far. 10 chemicals have been tested in this assay, using rats (Table 4). (8.6) Sisier-chromattd exchanges (SCEs) SCEs can be detected in proliferative bone mar row and spermatogonial cells. In vivo SCEs are scored from cells which have gone through the first cycle in the presence of BrdUrd, due to the fast depletion of BrdUrd from the animal when BrdUrd tablets are implanted in the animal. SCEs can also be detected in lymphocytes from spleen or from peripheral blood which are stimulated in vitro in the presence of BrdUrd (Perry. 1981; Abe and Sasaki, 1982). General discussion (l) Factors limiting the sensitivity of mammalian germ cells to mutagens In view of the large number of animals and high cost as well as time necessary to perform these tests only a limited number of chemicals has been screened using these tests. These chemicals are either established mutagens or suspected mutagens which have direct human relevance. ICPEMC Committee 1 (1983) considered the ques tion as to how many of the in vitro short-term tests have predictive value for mammalian germ cell mutagenicity by using available data on 53 compounds with which one or more of the follow ing tests have been conducted: (1) rodent domi nant lethal test; (2) mouse specific locus test; and (3) mouse heritable translocation test. Though there were inconsistencies in the performance of certain compounds within these 3 tests (e.g. myleran, hycanthone, benzofajpyrene, which were positive in the dominant lethal test but negative in the specific locus test), most of the chemicals were positive in all tests in vitro. We have compared the perfor- 194 table i QUALITATIVE COMPARISON OF IN VIVO MAMMALIAN TESTS TO 4 SHORT-TERM TESTS Chemicals Group 1, azindwes Mitomycin C Tepa Thio-tepa Tremmon Tnethylcnemclanamine In vivo mammalian assays Germ cells Soma tic cells 4 4- 4 4- 4+ 4 44 4' Group II: epoxides Ethylene oxide -''Styrene oxide 4- 4 -7 Group III: kaioalkanes > Captan ~-DDT ^Dieldrin Vinyl chlonde DiehJorvos ?< + ) *) 7 7 7 4> 7 7 7 Group IV1 inorganics Sodium nitrite Cadmium chlonde Sodium bisulfite --+ -- Group V. iV-nitroso compounds V-Nitroso-V-butyl urea N-Nitroso-V-ethyl urea A'-Nitroso-V-methyl urea V* Ethyl*nitroso-urelhane N-Nitroso diethylamine i-- V-Nitroso dimcthylamine 7 4- 44 7 7 4- 4- 4- 44' 4- Croup VI: N-, S,- O-mustards Cyclophosphamide ICR 170 Nitrogen mustard 4- 7 7 47 4- Group VII: purine analogues Caffeine 6- Mercaplopurine 4- 7 4- Group VIII: polycyclic hydrocarbons Benzo( a Jpyrene 4- 7,12*Dimethy1benzanthracene ? 44- Group IX: sulfate, sulfonate rThethyl sulfate Methyl methanesulfonate Ethyl methanesulfonate Iso-propyl methanesulfonate Myleran Hycanthone methylsulfonate 7 4- 4- 44 ?(+) 4 4 4 4 47 Droso phila SLRL 44* 4 44- 4- 4- - 7 NT 4- 4- NT 4- 444- + 444- + NT + 4- 4 4+ + 4 4- Bacteria reverse mutations 4NT 4* 44 44- 44- + 7 - NT 44NT 4 4- 4 4 4 4- 4- 4 4- 4- 44' + 4- Yeast mitotic recombtnation 4NT 4 4 4* NT - 44- 44 4* 4 4 NT 4* 4- 44 4- NT NT 4-- 4 444- NT NT In vitro cytogenc* tics (CA + SCE) 44+ 44 44 - 7 NT NT - 4NT 4- 4- 4NT 4 4- 4 NT 4 4 4 4 4 4 4 4 4 4 + \ TABLE 1 (continued) Chemicals Group X miscellaneous Saccharine Chloramphenicol Dimethyl sulfoxide Methotrexate Nitrofurantoin 'Butylaied hydroxytoluene 1soma2id ''Tnmethyl phosphate -"Ethylencthiourea .''i-Nilro-o-phenylenediamine 2'(2-Furanyl)-3-(5'mtro-2- furanyl)-2 propenamide Procarbazine 1,2-Dimethylhydrazine In vivo mammalian assays Germ cells Somatic cells ?( + ) '( + ) 0 9 9 9 7 + -(+ - 9 + 9 9 o 9 7 9 - 7 + - 7 + 9 Germ eel! data include sperm abnormality. ( + ). positive in dominant lethal assay. Droso phila SLRL 9 9 " NT NT NT + + + + 4+ Bacteria reverse mutations + + + 4+ + -- Yeast mitotic recombination + + 4- NT 4NT NT + 4+ NT 195 In vitro cytogene* tics (CA + SCE) 4* 4NT NT + + NT NT 4NT mance of 4 short-term tests, namely. Drosophila SLRL, yeast mitotic recombination, reverse muta tions in Salmonella and E. coli, and in vitro cyto genetics with the performance of in vivo mouse tests, including both somatic and germ cells. We tried to group the chemicals in different chemical classes in order to sec if there is any relationship between the ability of any one system to detect any particular chemical class. With only a few exceptions all chemicals which are detected as positive in mammalian in vivo systems were also positive in the short-term tests under consideration (Table 1). There is a large number of chemicals in in vivo tests where the available results do not allow to draw firm conclusions about their muta genicity (marked as ? in Table 1). This illustrates the inherent difficulty in using whole mammals in routine mutagenicity testing. If we confine our selves to the data obtained in mammalian germ cells alone, the number of uncertainties is even higher (Table 1). Doing an additional sperm anomaly test does not seem to improve the picture. The inconclusive results obtained in the germ cells are most probably due to the inability of the mutagenically active metabolites to reach the testes in sufficient quantities to be effective. Inconclusive or negative results obtained following treatment with potent mutagens such as MNNG, DEN and DMN illustrate this point. However, it is possible that the germ cells are more resistant to mutagenic insult or are more efficient in repairing damaged DNA. Systematic molecular dosimetric measure ments have to be conducted to solve this problem. An evaluation of induction of histocompatibility gene mutations may prove to be useful, but this system is not fully evaluated yet (Hamasch and Stumpf, 1984). (2) Comparison of mouse in vivo somatic chro mosomal aberrations (as revealed by micronuclei in the bone marrow) with the results obtained with the Salmonella typhimurium assay including those with liver microsome activation Unlike the mouse germ cell assays with the micronucleus test many more chemicals have been screened in view of the ease with which this test can be performed. Jenssen and Ramel (1980) com- 1% TABLE 2 QUALITATIVE COMPARISON OF BACTERIAL TEST TO IN VIVO RODENT MICRONUCLEUS TEST Chemical Croup I arontaiu amines (1) 2-Acetylaminofluorene (2) 2-Aminofluorene (3) Benzidine (4) 2,5-Diaminotoluene (5) 1-Naphthylamine (6) 4-Niiro-0*phenylene diamine (7) 2-Niiro-o-phenylene diamine (8) p-Phenylene diamine (9) 3.3'-5,-5'-Telramethylbenzidine (10) 4-Acelylaminofluorene (11) Dinitrosopentamethylene tetramine (12) 2-Naphthylamme (13) 2*4*Diaminoam$ole (14) O-Toludine (1J) Acnflavine Ames MN test test +4 -4- +4 4- +- 4444 4- -44- *> 44 Group II astndmes (16) Ethylencimine (17) Meiepa (18) Mitomycin C (19) Thio-tepa (20) Trenimon (21) Tnethylenemelanine 4- 4 4- 4 4- 4 4- 4 4- 4 44 Group III: biphenyls (22) 4-Chloromethylbiphenyl (23) Bcnzyichlonde (24) 4-Hydroxymcthylbiphenyl 44 4- Group IV. carbamyl-thtocarbomyis (25) Furyl furamide (26) N-Hydroxyurethane (27) iV-4.5-Niiro-2.furyl-2-thiazolyle formamide (28) Urethane (29) Dimethyl carbamoyl chloride (30) Dimethyl formamide (31) Isopropyl-N-3-chlorophenyl carbamide (32) Ethylenelhiourea 4 4 4 4 4 - - 4 4 4 - 4 4 - - Group V: DNA synthesis inhibitors (33) Aminopterin (34) Cytosine arabinoside (35) 5-Fluorouracil (36) Hydroxyurea (37) 5-Iodo-deoxyuridinc (38) 6-Mercaptopurine (39) Methotrexate -4 -4 44 --- 44 4 Group VI. flavonoids (40) Quercetin (41) Kaempferol (42) Rutin (43) Neohespendin dihydrochaicone 44 44 4- ?4 TABLE 2 (continued) Chemical Group VII kaloalkants and haloa/kenes (44) Captan (45) p.p -DDT (46) 1.2-Dichloroethane (47) Ethylene dichlonde tar (48) Vinyl chloride (49) Lindane (50) Chloroethane (51) Aroclor (52) Chloroform (53) 2.4-D (54) 2,4,5-T Group VIII: hydrazines (55) 1.1-Dimethyl hydrazine (56) Formic acid hydrazide (57) Maleic hydrazide (58) Procarbazine (natulan) (59) Hydrazine sulfate Group IX. inorganic salts (60) Methyl mercury acetate (61) Cadmium chlonde (62) Sodium nitrate (63) Lead acetate (64) Potassium dichromatc (65) Sodium chlonde Group X: lactones (66) /LPropiolactone (67) 0-Butyrolactone Group XI: N-. S-, O-mustards (68) Cyclophosphamide (69) Isophosphamide (70) Nitrogen mustard (71) Quinacrine mustard Group Xtl: nttro-aromattc and nilro-tmtdaioles (72) 4-Nitroquinoline oxide (4NQO) (73) 3-Methyl 4NQO (74) Nitrofurantoin (75) Nitrofurazone (76) Metronimidazole (77) Ronidazole (78) Omidazole (79) Dimetridazole (80) Azathioprinc (81) Niridazole Group XIII: N-nilroso compounds (82) Dibutylmtrosamine (83) N-Nitrosodiethylamine (84) N-Nitrosodimethyl urea (85) iV-Methyl-N'-nitro-Af-aitroeoguaeudine (86) N-fiitroso-carbaryl Ames MN test test 4--4- 4 4 ------ - 44 -" -4 4- -- 4-- 44 -- 4-- 44 44 44 44 44 44 4444- 4- 4- 44 4- 4* 444 44- MN Itfsl - - _ - - _ ft W - + + 4- - TABLE 2 (continued) Chemical (87) ,V-Nitrosoethylene thiourea 188) A-Nnrosomeihomyl phenol (89) ,V-Niirosomorpholine (90) Diphenyl mtrosamine Group XIV phenols (91) 2-Amino-4-mtrophenol (92) m-Amino-phenol (93) p-Amino-pheno! (94) Phenol Group XV: polycyclic hydrocarbons (9S) Aroclor 1254 (96) Benxofajpyrene (97) 9,10-Dimethylanthracenc (98) 3-Methylcholanthrene (99) Pyrene (100) Anthracene (101 (Naphthalene Group XVI, sulfatest sulfonates (102)Busulfan (mylcran) (103)Ethyl methanesulfonate (104) Methyl methanesulfonate (105) Isopropyl methanesulfonate (106)Hycanthone methanesulfonate (107)Propyl methanesulfonate Group XVII: triazenes (108) 1 -<4-Chiorophenyl)-3,3-dimethyltriaiene (109)]*Phenyl-3,3-dimethyltnazene Group XVIII: miscellaneous (110)Trimethylphosphate (111) Ethylene oxide (112)Epichlorohydnne (113)Actinomycin D (114)Adriamycin (115) Bleomycin (116)Chloromycetinsuccinate (117)Cycloheximide (118)Daunomycin (119)Gnseofulvin (12Q)Azobenzene (121)Acranil (122)Ascorbic acid (123)Aioxybenzene (124)Auramine (125) Colchicine (126) Vincristine (127)Vinblastine (128)CA cyclamate (129)Codeine phosphate (130)BHT (butylated hydroxytoluenc) (131) Dimethyl sulfoxide Ames MN test test ++ ++ - +-+ - -4- 4+ 4+ 4+--- + 4+ f + + 4++ ++ + + + 4- - + --- 4++ -- + 4- 4- + -4- 7 4* -+ - 4- 4-*-* - TABLE 2 (continued) Chemical (132) Ethanol (133) Epinephrine (134)Hydrochinone (135)D.L-Ethionine (136) Methionine (137) Diethyl stilbestrol (138) Moca (139)Sucrose (140) DAB sulfonic acid (141) Sodium phenobarbiturate (142) Monosodium glutamate (143)Nitrofurantotn (144)Nitrofurazone (145) Resorcinol (146)Quinacrine dihydrochloride (147)Aflatoxin B (148) Bcnzimadazole (149) Formaldehyde 197 Ames MN test lest - + -- 4* - 4--T 4 -44- 4" 4- 4- 4- 44- + -4- pared micronucleus test and Ames test and found that they had about the same specificity and pre dictive value, while there was a significant dif ference in sensitivity in favor of Ames' test. In their review, no attempt was made to group the tested chemicals in specific categories. In Table 2 we have compared the results comprising 148 chemicals tested with both the micronucleus test and the Ames test. The chemicals have been grouped with regard to their nature and mode of action (Table 2). In Table 3 we have summarized the data in order to see if any distinct differences exist between the two tests with regard to the detection of any specific group of chemicals. Potent directly acting mutagens come out positive in both the tests (aziridines, triazenes, mustards, sulfo nates). Inhibitors of DNA synthesis and DNA base analogues are picked up with the micronucleus test with greater efficiency. This is not unexpected as these chemicals are efficient chro mosome breakers and do not induce point muta tions effectively. Mutagens which require meta bolic activation such as aromatic amines and some iV-nitroso compounds are easily detected in the Salmonella test using S9 activation whereas most of them are not detected in the micronucleus test but can be detected in the liver micronucleus test. 198 TABLE 3 PERFORMANCE OF IN VIVO MICRONUCLEUS TEST AND BACTERIAL TESTS IN DETECTING DIFFERENT GROUPS OF CHEMICALS Chemical classes (1) Aromatic amines (2) Azindines Number com pared Bac terial tests Micro nucleus test 15 + (13) + (3) -(1) -(12) 6 + (6) + <5> (3) Biphenyls 3 + <3) -(3) (4) Carbamyl. thiocarbamyls 8 (S) DNA synthesis inhibitors 7 (6) Flavonoids 4 (11) Haloalkanes 11 + <6) -<2) + (2) -<5) + (3) 7(1) + (4) -<7) + (4) -(4) + <J) -(2) + <3) -(1) + (2) -(9) (8) Hydrazines 5 + (3) + <1> -(2) -(4) (9) Inorganic salts 6 + (2) + 0) -(4) -(5) (10) Lactones (11) Mustards 2 + (1) -0) -(2) 4 + (4) + (4) (12) Nitroaromatics 2 + <2) + <2) (13) ALNitroso compounds 9 + (8) 7(1) + <3) -(6) (14) Nitroimidazoles 6 (15) Phenols 4 (16) Polycyclic hydrocarbons 7 + (5> -(1) M3) -0) + (4) -0) + <D -(4) + <U "(3) + (3) -(4) (17) Sulfates, sulfonates 6 + <) + <S) -(1) (18) Tnazenes 2 + (2) -M2) (19) Miscellaneous 50 + (22) + (18) -(27) -(29) 7(1) 7(3) This again should be due to the inability of the active metabolites to reach the target cells in the bone marrow, whereas hepatocytes are directly involved in the activation of these carcinogens. Nitroimidazoles seem to be ineffective in inducing micronuclei in bone marrow, but effective in in ducing mutations in bacteria. Chemicals which disturb the mitotic sptndle apparatus such as vincristine and colchicine are positive in the micronucleus test and negative in Salmonella. Considering only the results which are either clearly positive or negative we have data for 148 chemicals tested in both these systems. Of these SO were positive and 40 were negative in both the test systems. The two test systems re sponded differently to 58 chemicals. These dis crepancies are not unexpected in view of the fact that we are measuring two different genetic end points, namely, point mutations and chromosomal aberrations and only one tissue in vivo, namely bone marrow cells. Chemicals which induce both these events will be picked up with both the sys tems. On the contrary chemicals which induce chromosomal aberrations preferentially as well as those inducing spindle defects will be picked up by the micronucleus assay and not by the bacterial assay. The reverse is true for chemicals which induce point mutations mainly. In bacterial assays the test chemicals are in direct contact with the target cells at high con centrations whereas in mammalian in vivo assays the concentrations reaching the target cells are limited by various pharmacokinetic parameters. For this reason the sensitivity of this test system is TABLE 4 PERFORMANCE OF LIVER MICRONUCLEUS TEST (1) Dimethyl nitrosaminc (2) Diethyl nitrosamine (3) Ethyl nitrosourea (4) Ethyl methanesulfonate (5) Methyl methanesulfonate (6) Mitomycin C (7) 2-Acetylaminofluorene (8) 4-Acctylaminonuorene (9) Benzc(a Jpyrene (10) Pyrene + + + + + + + - Based on data by Tates (personal communKauoo). I the 1 the etc 11 y gens, .icing n i Fl uidic e are \e in h are ;a for v. Of *.e in is ret dist fact c end .omal amely both e sysnduce ell as ire in i conassays Is are neters. tem is ST relatively low. However, employing an in vitro cytogenetic assay using a mammalian metabolic activation system one can detect most of the chem icals which are capable of producing point muta tions as well as chromosomal aberrations, because cells can be treated with much higher concentra tions of the test chemical than those employed for in vivo experiments. (J) The relevance of chromosomal aberrations m mutagenicity testing It is known that about 6 out of 1000 newborns have a chromosomal abnormality though there is no estimate of point mutations in newborn. Popu lations exposed to mutagens respond with in creased frequencies of chromosomal aberrations and cancer rather than detectable point mutations. This demonstrates the importance of screening chemicals for their ability to-induce chromosome aberrations in vitro and/or in vivo as an indicator of potentiality for mutagenicity and carcinogenic ity in man. This brings us to the question of the choice of a test battery to detect mutagenicity of chemicals. While in general all potent mutagens induce chromosomal aberrations potent chro mosome-breaking agents do not always induce point mutations efficiently e.g., X-rays, bleomycin. In a routine test for mutagenicity it is important to employ an in vitro cytogenetic assay as such an assay would detect chemicals which are not effi cient inducers of point mutations but are of poten tial hazard to man by inducing chromosomal aber rations. Similar conclusions were arrived at by the analysis of the results of a study designed to compare short-term tests for mutagens and carcinogens (Ashby et al., 1985). The number of such chemical agents are increasing rapidly, e.g., benzene, resorcine and several flavoring agents (such as allylisotiiiocyanate, anisaldehyde, benzaldehyde, cinnamaldchyde, citronellal, heliotropin, vanillinc etc.) (Kasamaki et al., 1982). Concluding remarks In a comparative survey of the ability of in vivo mammalian assays to in vitro assays to detect mutagens it is obvious that one is dealing with two related phenomena i.e., point mutations and chro 199 mosomal aberrations intermingled. The limited data available using in vivo mammalian assays (somatic as well as germ cells) indicate that the sensitivity of this system is not very high when compared to in vitro systems. This is in view of inherent characteristics of the in vivo assay, namely the complex activation and deactivation mecha nisms that operate as well as the efficiency with which the active metabolite reaches the target cells under study. Nevertheless, with appropriate mod ifications of the protocol, for a micronucleus assay in bone marrow cells, as well as introduction of liver and germ cell micronucleus assay (Tates et al., 1983), the sensitivity of in vivo assays can be greatly improved. In mutagenicity testing, in vitro and in vivo assays serve different purposes. For preliminary screening, in view of the increased sensitivity, the possibility to use higher concentrations of test chemical, and the short time involved, in vitro tests are ideal. These assays detect the genotoxic potential of the test chemical. However, to study the ability of a chemical to express the genotoxic potential in vivo, whole-animal test systems should be used (Ashby, 1983). A decision tree approach to detect possible human carcinogens has been proposed by Ashby (1983) in which it is clear that in vivo tests have a very important position, not at the detection stage, but at validation stage. It is clear that when a chemical is being considered to be released for large-scale public use, an in vivo mammalian assay should be mandatory. Acknowledgements The project was supported by EEC Chemical Mutagen Programme. We acknowledge with thanks the constructive criticisms of Dr. J. Ashby on this manuscript. References Abe, S., and M. Sasaki (1982) SCE as an index of mutagenesis and/or carcinogenesis, in: A.A. Sandberg (Ed.), Sister Chromatid Exchange, Lias, New York, pp. 461-515. Ashby, J. (1983) The unique rote of rodents in the detection of possible human caicinogaia. Mutation Res., 115, 177-213. Ashby, J., FJ. de Senes, M. Draper, M. lshidate Jr., B.H. Margolin, B.E. Matter and M.D. Shelby (1985) Overview and conclusions of the IPCS collaborative study on in vitro OLI 3331 zoo assay systems, in J Ashby el al. (Eds ). Progress in Mutation Research. V'ol 5. Elsevier. Amsterdam, pp. 117-174, Brusick. D.J., V.F, Simmon, H S. Rosenkranz. VA Ray and R.S. Stafford (1980) An evaluation of the Escherichia co/i WP2 and WP2 UVrA reverse mutation assay. Mutation Res., 76. 169-190. De Flora. S . P. Zanacchi. A Cambirano. C. Bcnnicelli and G Badolati (1984) Genotoxtc activity and potency of 135 compounds in the Ames reversion test and in a bacterial DNA-repair test. Mutation Res.. 133, 161-198 de Serres, FJ. and J Ashby (Eds.) (1981) Evaluation of Short-Term Tests for Carcinogens. Elsevier/North-Holland. Amsterdam, p 827. Fating, R. (1978) The mammalian spot test: a sensitive in vivo method for the detection of genetic alterations in somatic cells of mice, in: A. Hollander and FJ. de Serres (Eds ). Chemical Mutagens. Vol. 5, Plenum. New York, pp. 151--176. Green. S,, A. Anletta. J. Fabncant. R. Kapp. M. Manandhar. Chtng-In-Sheu. J. Springer and B. Whitfield (1983) Current status of bioassays in genetic toxicology. The dominant lethal assay. Mutation Res.. 134. 49-67. Hamasch, D,, and R. Stumpf (1984) Studies on the induction of histocompatibility gene mutations of mice by chemical mutagens and/or virus inducing agents. Mutation Res.. 126, 279-293, Hollstein, M., J, McCann. F A. Angelosanto and W.W. Nichols (1979) Short term tests for carcinogens and mutagens. Mu tation Res., 65, 133-226. IARC (1980) Long-Term and Short-Term Screening. Assays for Carcinogens. A Critical Appraisal. IARC Monographs Suppl. 2, p. 426. International Agency for Research on Cancer. Lyon. ICPEMC Committee 1 (1983) Screening strategy for chemicals that are potential germ-cell mutagens in mammals. Com mittee 1, Final Report. International Commission for Pro tection Against Environmental Mutagens and Carcinogens. Document 100-1982*2.3. Mutation Res.. 114, 117-177. Jenssen, D . and C. Ramel (1980) The micronucleus test as part of a short term mutagenicity test program for the prediction of carcinogenicity evaluated by 143 agents tested, Mutation Res., 75, 191-202. Kasamaki. A., H. Takahasi. N. Tsumura, J. Niwa, T. Fujita and S. Urasawa (1982) Genotoxicity of flavoring agents. Mu tation Res.. 105, 387-392. Levin. D.E. M. Hollstein, M.F. Christman. EA. Schwiers and B.N. Ames (1982) A new Salmonella tester strain (TA 102) with AT base pairs at the site of mutation detects oxidative mutagens, Proc. Natl. Acad. Sd. (U.S.A.), 79, 7445-7449. Mohn, G,, P. Kerklaan and J. EUenberger (1984) Methodolo gies for the direct and animal-mediated determination of various genetic effects in derivatives of strain 343/113 of Escherichia coii K-12 in: BJ. Kilbey et al. (Eds.), Handbook of Mutagenicity Test Procedures. Elsevier, Amsterdam, pp. 189-214. Natarajan. A.T., and I. Czukas (1984) Mechanisms of sister chromatid exchanges, in: Critical Evaluation of Mutagenic ity Tests. MMV Medizin Verlag, Mlinehen. pp. 99-110. Natarajan, AT., and G. Obe (1982) Mutagenicity testing with cultured mammalian cells: Cytogenetic assays, in. J A Heddle (Ed.), Mutagenicity -- New Honzons in Genetic Toxicology. Academic Press. New York. pp. 171-213 Natarajan, A.T., A.D Tates. P.P.W. van Buul, M. Meyers and N. de Vogel (1976) Cytogenetic effects of mutagens/ carcinogens after activation in an in vitro microsomal sys tem. I. Induction of chromosome aberrations and sister chromatid exchanges by diethyl nitrosamme (DEN) and dimethylnilrosamine (DMN) in CHO cells in the presence of rat liver microsomes. Mutation Res.. 37. 83-90 Natarajan. A.T., J.W.I.M. Simons, E.W. Vogel and A.A, van Zeeland (1984) Relationship between cell killing, chro mosomal aberrations, sister chromatid exchanges and point mutations induced by monofunctional alkylating agents in Chinese hamster cells. A correlation with different ethyl ation products in DNA, Mutation Res.. 128. 31-40. Natarajan, A.T., A.A. van Zeeland and T.S.B. Zwanenburg (1983) Influence of inhibitors of poly(ADP-ribose) poly merase on DNA repair, chromosomal alterations and muta tions, in: Proc. Princess Takamatsu Cancer Symposium. Jpn. Sci. Society Press. Tokyo, pp. 227-242. Perry. P.E (1981) Chemical mutagens and sister chromatid exchange, in: FJ. de Series and A. Hollander (Eds.). Chem ical Mutagens, Vol. 6, Plenum, New York, pp, 1-39. Rawley, J. (1983) Human oncogene locations and chromosome aben-ations. Nature (London), 301, 291-292. Rieger. R,, and A. Michaelis (1962) Die Auslosung von Chromosomenaberrationen bei Vicia fata durch chcmische Agenzien, Eine Obersicht, Kulturpflanze. 10. 212-191. Rinkus, SJ., and M.S. Legator (1979) Chemical characteriza tion of 465 known or suspected carcinogens and their correlation with mutagenic activity in Salmonella typhimunum system. Cancer Res., 39, 3289-3318. Russell L.B., and M.H. Major (1957) Radiation induced pre sumed somatic mutations in the house mouse. Genetics. 42, 161-175. Russell, L.B., and B.E Matter (1980) Whole-mammal mutagen icity tests; Evaluation of five methods. Mutation Res,, 75, 279-302. Russell, L.B., P.B. Shelby, E von Halle. W. Sbcriden and L. Valcovic (1981) The mouse specific locus test with agents other than radiations. Interpretation or data and recom mendations for future work. Mutation Res., 86. 329-354. Russell. W.L. (1951) X-Ray induced mutations in mice, Cold Spring Harbor Symp. Quant. Biol., 16, 327-336. Russell. W.L (1984) Dose response repair, and no effect dose level in mouse germ cell mutagenesis, in: Y. Tazima et al. (Eds.), Problems of Threshold in Chemical Mutagenesis, The Environmental Mutagenic Society of Japan, pp. 125-131. Tates, A.D., and A.T. Natarajan (1976) A correlative study on the genetic damage induced by chemical mutagens in bone marrow and spermatogonia of mice, I. CNU-cthanoL Muta tion Res,, 37, 267-278. Tates, A.D., AJJ. Dietrich, N. de VogeL I, Neutcboom end A. Bos (1983) A micronuckus method of detection of mciotic micronuclei in male germ edit of mammals. Mutation Res.. 121, 131-138. ' ith ed- ellC ind n5/ -Vi- -ter jnd van h ro om! ' in hyl- 7urg >olymtamm. land lem- ome h.roi>che prev. 42. agen.. 75. id L. gents :com154. Cold dose cl al. nesis. PP dy on : bone Muta- ind A. .teioiic i Res.. Tates. A.D. I Neuteboom. N. de Vogel and L. den Engelse (1983) The induction of chromosomal damage in rat hepatocvtes and lymphocytes. I. Time dependent changes of clastogemc effects of diethylmtrosamine. dimethyl mtrosamtne and ethyl methanesulfonate. Mutation Res,. 107. 131-151. Vogel. E.W (1985) The Drosophila somatic recombination and mutation assay (SRM) using the whiit-coral somatic eye colour svstem, in. J, Ashby, F.J, de Serres el al (Eds ). Progress in Mutation Research. Vol. 5, Elsevier. Amster dam. pp 313-317. Vogel. E.W . and A.T, Natarajan (1982) The relation between reaction kinetics and mutagenic action of monofunctional alkylating agents in higher eukaryotic systems' interspecies comparisons, in: F.J. de Serres and A. Hollander (Eds.). Chemical Mutagens, Vol. 7, Plenum. New York, pp, 295-336 Vogel. E.W., W.G.H Blijleven, P.M. IClapwijk and J A. Zijlstra (1980) Some current prospectives of the application of Drosophila in the evaluation of carcinogens, in: G.M. Wil liams el al. (Eds.), The Predictive Value of Short-Term Screening Tests in Carcinogenicity Evaluation. Elsevier/ Nonh-Holland Biomedical Press. Amsterdam, pp 125-147.