Comparative evaluation of the in vitro micronucleus test and the in vitro chromosome aberration test: industrial experience

Mutation Research 392 Ž1997. 45–59

Comparative evaluation of the in vitro micronucleus test and the in vitro chromosome aberration test: industrial experience Beate Miller a , Silvio Albertini b,) , Franziska Locher c , Veronique Thybaud d , Elisabeth Lorge e a

b

Vitamin DiÕision, F. Hoffmann-La Roche Ltd., CH-4070 Basel, Switzerland Pharma DiÕision, Preclinical Research, Department of Toxicology, F. Hoffmann-La Roche Ltd., CH-4070 Basel, Switzerland c Drug Safety, Toxicology, NoÕartis Ltd., CH-4002 Basel, Switzerland d Rhone-Poulenc Rorer, Research and DeÕelopment, Drug Safety, F-94403 Vitry-sur-Seine Cedex, France ˆ e Biologie SerÕier, F-45403 Fleury-les-Aubrais Cedex, France

Abstract Because of its rapidness, simplicity and potential for automation, the measurement of micronucleated cells in vivo is not only equivalent to the analysis of chromosome aberrations, but often even preferred within routine genotoxicity testing. In order to evaluate the correlation between the in vitro micronucleus assay ŽMNT. and the in vitro chromosome aberration test ŽCA., we collected data from four pharmaceutical companies obtained either in Chinese hamster cell lines ŽCHO-K5, CHO-K1, V79. or in human peripheral blood lymphocytes. Among the 57 compounds included in this comparison, 45 compounds gave rise to concordant results in both assays Ž26 compounds negative in both assays; 19 compounds positive in both assays.. The high percentage of concordance, i.e. about 79% is very promising and can be even increased to about 88% by omitting the 3 aneugenic compounds and 2 compounds inducing endoreduplicated chromosomes which were found positive only in the in vitro MNT. The results are remarkable in particular considering that most of the compounds evaluated are ‘standard’ pharmaceutical compounds and thus are at most weak inducers of chromosome damage. Our comparison strongly supports that the in vitro micronucleus test is a suitable alternative to the in vitro chromosome aberration assay. Moreover, the MNT has the potential of not only detecting clastogens but additionally aneuploidy inducing chemicals. Keywords: In vitro micronucleus test; In vitro chromosome aberration test; Routine genotoxicity screening

1. Introduction The chromosomal aberration ŽCA. test has been widely used and recommended by regulatory authorities for the assessment of in vitro chromosome damage. Metaphase analysis is very tedious and timeconsuming. For in vivo testing the micronucleus test

)

Corresponding author.

ŽMNT. in bone marrow of rodents has long been established as a well validated, rapid alternative to the much more labor intensive evaluation of chromosomal aberrations in vivo ŽAshby, 1986.. Although, the evaluation of MN in established cell lines or human lymphocytes has been shown feasible for the detection of chromosome damage ŽEllard et al., 1991; Matsuoka et al., 1993; Vian et al., 1993., it is not generally accepted by regulatory authorities as an alternative system.

1383-5718r97r$17.00 Copyright q 1997 Elsevier Science B.V. All rights reserved. PII S 0 1 6 5 - 1 2 1 8 Ž 9 7 . 0 0 0 4 4 - X

46

B. Miller et al.r Mutation Research 392 (1997) 45–59

Because only few studies comparing MNT and CA in vitro exist ŽMigliore et al., 1987, Matsuoka et al., 1993; Li et al., 1993., we collected data from four pharmaceutical companies, namely F. Hoffmann-La Roche Ltd., Novartis Ltd., Rhone-Poulenc ˆ Rorer and Biologie Servier. Either established cell lines ŽCHO, V79. or human peripheral blood lymphocytes ŽHPBL. were used. The results from the in vitro MNT are compared to data from the in vitro chromosome aberration test ŽCA. and, where available, discussed in the context of other genotoxicity data Ži.e. in vivo MNT and gene mutation in bacteria or mammalian cells..

2. Materials and methods 2.1. Test compounds For reasons of confidentiality compounds under development were coded and classified according to the scheme of Kier et al. Ž1986.. 2.1.1. F. Hoffmann-La Roche Nalidixic acid, norfloxacin, and ciprofloxacin were obtained from Sigma. All other test compounds were synthesized in-house. Ciprofloxacin and enrofloxacin were dissolved in deionized water ŽH 2 O., nalidixic acid and ofloxacin in culture medium and all the other compounds were dissolved in DMSO ŽFluka, final concentration in the medium 1%.. The promutagen cyclophosphamide ŽCP; Serva. and the directly acting mutagen bleomycin ŽBLEO; Lundbeck. were dissolved in water and served as positive controls. 2.1.2. NoÕartis All test compounds were synthesized in-house and dissolved in minimal essential medium ŽMEM, Merck. or dimethylsulfoxide ŽDMSO, Merck.. The final DMSO concentration in the culture medium was 1%. Ethyl methanesulfonate ŽEMS; Aldrich., a direct-acting mutagen, and cyclophosphamide ŽCP; Asta Medica., an indirect-acting mutagen, were used as positive controls. 2.1.3. Rhone-Poulenc Rorer ˆ The test chemicals were Rhone-Polenc Rorer inˆ house compounds. Compounds were dissolved either

in culture medium or in DMSO ŽSigma, final concentration in culture medium: 1%.. Mitomycin C ŽMMC; Ametycine, Laboratoires Choay. or methyl methanesulfonate ŽMMS; Merck. were used as positive controls for direct-acting mutagens in the MNT and CA. Cyclophosphamide ŽEndoxan, Asta Medica. was used as a positive control for indirect-acting mutagens in both assays. 2.1.4. Biologie SerÕier All the test compounds were synthesized in-house and the genotoxicity tests were performed at Biologie Servier or in contract laboratories ŽInstitut Pasteur de Lille, France, Hazelton, UK, HRC, UK.. They were dissolved in RMPI 1640 ŽGibco BRL. or DMSO ŽSigma.. The final DMSO concentration in the culture medium was 1%. The following positive control chemicals were used: as direct-acting mutagens ethyl methanesulfonate ŽEMS; Sigma., mitomycin C ŽMMC; Ametycine, Laboratoire Choay., bleomycin ŽR. Bellon., or 4-nitroquinoline-oxide Ž4NQO; Fluka. and as indirect acting mutagen cyclophosphamide ŽCP; Asta Medica.. 2.2. Cell cultures for the MNT and the CA 2.2.1. F. Hoffmann-La Roche The Chinese hamster cell line CHO-K5 Žoriginally obtained from Dr. T. Hertner, Ciba. was used. Cells were grown in Ham’s F-10 medium with Lglutamine ŽGibco BRL., supplemented with 10% inactivated foetal calf serum ŽGibco BRL., and 1% penicillin Ž10 000 IUrml.rstreptomycin Ž10 000 mgrml. ŽGibco BRL., at 378C in a humidified atmosphere containing 5% CO 2 . The modal chromosome number was 19 and the doubling time was under the described experimental conditions about 14–16 h. 2.2.2. NoÕartis V79 Chinese hamster cells were obtained from M. Fox ŽPaterson Laboratories, Manchester, UK.. The cells were routinely cultured at 378C Žhumidified atmosphere containing 5% CO 2 . as monolayers in Eagle’s minimal essential medium ŽMEM; Gibco BRL. supplemented with 10% fetal calf serum ŽNABI., glutamine Ž2 mM, Gibco BRL. and 0.2% sodium bicarbonate ŽMerck., but without antibiotics. The modal chromosome number was 21 " 2 and the

B. Miller et al.r Mutation Research 392 (1997) 45–59

average generation time was about 12 h. During the experiments, penicillin Ž200 IUrml, Gibco BRL. and streptomycin Ž200 mgrml, Gibco BRL. were present in the culture medium. 2.2.3. Rhone-Poulenc Rorer ˆ Chinese hamster ovary cells ŽCHO-K1. were provided by Labsystem Flow Elkay. Cells were maintained in exponential growth in Ham’s F-12 medium supplemented with 10% fetal calf serum ŽFCS, Labsystem., 1 or 2% solution of penicillin Ž5000 IUrml.rstreptomycin Ž5000 mgrml. ŽLabsystem., and 1 or 2% L-glutamine Ž200 nmolrml, Labsystem., at 378C in a humidified atmosphere containing 5% CO 2 . Their modal chromosome number was 21 " 2 and their average generation time was 12–14 h. 2.2.4. Biologie SerÕier Blood was obtained from healthy male and female volunteers. Peripheral blood was collected by venepuncture in heparinised tubes, and whole blood cultures were cultivated in RMPI 1640 Dutch modification ŽGibco BRL. supplemented with 10% foetal calf serum ŽGibco BRL. and antibiotics ŽGibco BRL.. Mitotic stimulation was obtained by using Phytohaemagglutinin A ŽWellcome, 2% Žvrv. or 7.5–37.5 mlrml.. Precultivation before treatment was done for about 20–27 or about 48 h at 378C. 2.3. Metabolic actiÕation system Rat liver S9 was prepared according to Ames et al. Ž1975.. Liver enzymes were induced either with Aroclor 1254 ŽSiegfried or Analabs Inc.. according to Maron and Ames Ž1983. ŽNovartis, Rhone-Pouˆ lenc Rorer, Biologie Servier. or with phenobarbital ŽSiegfried, diluted in pyrogene-free aqua bidest.. and b-Naphtoflavone Žsuspended in corn oil, Serva. following the method of Matsushima et al. Ž1976. ŽF. Hoffmann-La Roche.. The protein content determined according to Lowry et al. Ž1951. were in the range of 30–40 mgrml. The preparations were shown to be sterile. 2.4. Cytotoxicity assay for the MNT and the CA In the absence of cytotoxicity all compounds were tested up to the limit of 10 mM or 5 mgrml.

47

2.4.1. F. Hoffmann-La Roche As a criterion for toxicity in the MNT in vitro a reduction of the cell density by at least 25% compared to the control cultures combined with an altered morphology of the cells Žround, diffuse shape, floating cells. was defined microscopically. In the CA, cytotoxicity was determined by the mitotic indices Žreduction of the mitotic indices G 50%. and the neutral red assay ŽBorenfreund and Puerner, 1985..

2.4.2. NoÕartis For the in vitro MNT, the maximum concentrations were determined after a preliminary cytotoxicity test using a MTT Ž3-w4,5-dimethylthiazol-2-ylx2,5-diphenyl tetrazolium bromid, Sigma. colorimetric method. Concentrations leading to a reduction in absorption of 25% and 80% were determined. For the main experiment, six concentrations between these two values were calculated. The highest concentration analyzed in the main test should lead to a reduction of the mitotic index by at least 50%. In the CA cytotoxicity was determined by the mitotic indices Žreduction of the mitotic indices G 50%..

2.4.3. Rhone-Poulenc Rorer ˆ For the in vitro MNT, the maximum concentrations were determined after a preliminary cytotoxicity test using a MTT Ž3-w4,5-dimethylthiazol-2-ylx2,5-diphenyl tetrazolium bromide, Sigma. colorimetric method ŽMossmann, 1983.. For the in vitro CA, cytotoxicity was determined by measuring the relative cell growth using a coulter counter. In both cases, when cytotoxic, the highest concentrations tested induced 30–50% reduction in the cytotoxicity measurements.

2.4.4. Biologie SerÕier For the in vitro MNT, the maximum concentrations were determined by evaluating the fraction of binucleated cells. For the in vitro CA, cytotoxicity was determined by a reduction of the mitotic indices. In both cases, when cytotoxic, the highest concentrations tested induced at least a 50% reduction in the cytotoxicity measurements.

48

B. Miller et al.r Mutation Research 392 (1997) 45–59

2.5. In Õitro micronucleus assay (MNT) 2.5.1. Hoffmann-La Roche The MNT in vitro was performed as described in Miller et al. Ž1995.. Cells were cultivated in 2 chamber-slides ŽLab-Tek, Nunc.. Approximately 9000 cells diluted in 1.5 ml of culture medium were seeded in each chamber. After about 24 h, the medium was substituted by 1.5 ml of fresh culture medium Ž10% FCS. containing different concentrations of the test compound. In general, sampling times of 48 h were chosen. The cells were continuously treated for about 48 h. The cells were fixed with methanol ŽFluka. containing 1% acetic acid for 20 min at room temperature. Slides were air dried, then treated with RNAse ŽSigma, 0.01 mgrml in 2 = SSC. for about 3–7 min at 378C. After washing in 2 = SSC and in demineralized water, slides were air dried again, stained with 5% Giemsa ŽFluka. for 15 min and mounted with Eukitt ŽO. Kindler.. If possible, micronuclei were analyzed in at least 1000 cellsrculture of a minimum of two parallel cultures Žexcept for the positive controls.. Micronuclei were identified according to following criteria: clearly surrounded by a nuclear membrane, area less than 13 of the area of the main nucleus, non-refractility, not linked to the main nucleus via nucleoplasmic bridges, location within the cytoplasm of the cell Žsee also Countryman and Heddle, 1976; Schmuck et al., 1988; Fenech, 1993.. Only mononucleated cells with well preserved cytoplasm containing five or less micronuclei were scored in order to exclude apoptosis and nuclear fragmentation. Taking into account variability of the MN rates, as a positive result a dose-related effect showing an increase of the MN frequency over the control by about threefold or higher in at least one dose tested was defined. In order to show reproducibility of the results, at leasttwo independent experiments were performed. 2.5.2. NoÕartis 6-Well plates were each inoculated with 2 ml per well of a 5000 cellsrml cell suspension. The plates were incubated for 24 h. Thereafter, the medium was replaced by treatment medium Ž10% FCS. containing different test compound concentrations or positive controls dissolved in medium or S9 mix. Cells were treated for 3 h Ž"S9. or 20 h ŽyS9. in case of

negative results or no toxicity after short-term treatment. After treatment cells were washed with PBS ŽOxoid. and reincubated with fresh medium for 45 Ž28. h. Fixation occurred approximately 48 h after start of treatment. In the case of negative results with 3 h treatment without S9, a second experiment was carried out using continuous treatment for 20 h ŽyS9.. Before sampling, cells were examined for morphology and growth. Medium was removed and trypsin EDTA ŽGibco BRL. was added to release the cells. Then, cells were diluted with PBS for cytocentrifugation ŽShandon cytospin.. Slides were air dried. Cells were fixed with 70% ethanol ŽMerck. and 0.5 M HCl ŽMerck. for 10 min followed by a hydrolysis with 5 M HCl solution for 40 min and finally rinsed with H 2 O. The slides were stained with SCHIFF reagent ŽMerck.. For each test compound at least four concentrations Ž2 cultures per concentration. and 4000 cells per concentration were analyzed. Criteria for analysis of micronuclei were as follows. Every micronucleus was brought into focus in order to exclude small particles of stain. The diameter had to be less than 13 of the main nucleus, be surrounded by a clear membrane, lie within the cell plasma, but not touching the main nucleus. Cells in mitosis were not counted. A compound was considered positive if the MN frequency was ) 1% and at least 0.6% above the concurrent solvent control value. 2.5.3. Rhone-Poulenc Rorer ˆ Each well of a 6-well plate was seeded with 1 = 10 5 CHO-K1 cells in 2 ml culture medium and cells were grown for 24 h. Then, the culture medium was discarded and cells were exposed for 4 h to at least three concentrations of the compound in 1 ml of culture medium devoid of FCS and in the presence or the absence of S9 mix Ž10% vrv.. The medium was then replaced and cells were allowed to grow for further 44 h. The cells were then stained ŽMerck. for 3 min, followed by with May-Grunwald ¨ Giemsa ŽMerck. diluted 1:3 in water Žvrv. for 5 min. The plates were then rinsed with tap water and air-dried; the bottom of the plate was detached by a punch and mounted on a slide. MN were determined microscopically according to Lasne et al. Ž1984.. Results were expressed by the number of micronucleated cells per one thousand cells. A compound was considered positive if at least a twofold increase

B. Miller et al.r Mutation Research 392 (1997) 45–59

49

in the number of micronucleated cells was observed at one or more concentrations in at least one of the assays conducted.

chromosomal aberrations excluding gaps, exceeding historical control values Žabout 2–6%., were regarded as positive.

2.5.4. Biologie SerÕier After about 20–27 Žprotocol 1. or 48 h Žprotocol 2. of precultivation, cultures were treated with the test compounds or the positive controls for 2–3 h in the presence of metabolic activation or for 20–24 h in the absence of metabolic activation. Normal culture medium containing 10% FCS was used for treatment. Cytochalasin B ŽSigma. Ž6 mgrml Ž1. or 3 mgrml Ž1., dissolved in DMSO. ŽSigma. was added to the cultures from about 44 to 72 h Ž1, 2.. Harvest took place at about 72 h Ž1. or 68 h Ž2. after the start of the cultures. Briefly, mild hypotonic treatment ŽRMPI-water, Gibco BRL, 1:4, q2% FCS. Ž1. or 0.075 M KCl Ž2. ŽMerck. was performed, the cells were fixed with methanolracetic acid Ž3:1. and dropped on glass slides. The cells were stained with Giemsa ŽMerck. 10%. For scoring of MN, 1000 binucleated cells ŽBN. per culture were analyzed on randomized coded slides. A compound was considered positive if at least a two-fold increase in the number of micronucleated cells was observed at one or more concentrations.

2.6.2. NoÕartis Petri dishes of 9 cm diameter were seeded with about 500 = 10 3 V79 cells in 10 ml culture medium Žtwo dishes per concentration.. On the first day after seeding, cells were treated with dilutions of the compound in 5 ml culture medium Ž10% FCS. containing different concentrations of the test compound or positive controls dissolved in medium or S9 mix. Cells were treated for 3 h Ž"S9. or 20 h ŽyS9.. Two hours before the end of the incubation, 0.15 mgrml colcemid ŽGibco BRL. was added. Thereafter, the culture medium was transferred from the dishes to 15 ml tubes and the cells were trypsinized with 2 ml 0.2% trypsin ŽGibco BRL. for 5 min and transferred to the corresponding tubes. Hypotonic treatment Ž7.5 min in prewarmed 1% trisodiumcitrate. ŽMerck., fixation Žin methanolracetic acid, 3:1. ŽMerck. and slide preparation were done according to standard procedures. Slides were stained with Giemsa Ž2%. ŽFluka., mounted in DePeX ŽGurr., coded and analyzed for the presence of chromosomal aberrations according to well known procedures ŽSavage, 1976; Scott et al., 1991.. One hundred metaphases originating from one Petri dish Ž200–400 metaphases per treatment, 21 " 2 chromosomes per cell. were analyzed. The mitotic index was determined by counting 1000 cells originating from one Petri dish Ž2000–4000 cells per concentration. and recording the number of metaphases among them. Chromosomal aberration frequencies of G 5% Žexcluding gaps. were judged as positive Žhistorical control data being about 0–4.75%..

2.6. In Õitro chromosomal aberration assay (CA) 2.6.1. Hoffmann-La Roche Single-cell suspensions of 3–5 = 10 4 cells ŽCHOK5. were seeded onto glass slides in Quadriperm culture dishes ŽHeraeus. and incubated for about 40 h before treatment in culture medium containing 10% FCS. Cells were subsequently exposed to the test compound for 20 h in the absence or for 3 h in the presence of rat liver S9 mix Ž10% Žvrv.. followed by a recovery period of 17 h. Colcemid ŽFluka. at a concentration of 0.2 mgrml was added 2 h before the end of the incubation. Metaphase preparations were carried out in situ as described by Salassidis et al. Ž1991.. The air-dried slides were stained for 10 min with 3% Giemsa ŽFluka. solution ŽpH 6.8. and mounted with Eukitt ŽO. Kindler.. The highest concentration evaluated showed clear cytotoxicity as determined by the mitotic indices and the neutral red assay ŽBorenfreund and Puerner, 1985.. The aberrations were classified according to Savage Ž1976.. Significant increases ŽFisher’s exact test. of

2.6.3. Rhone-Poulenc Rorer ˆ For each concentration, two 80 cm2 flasks were seeded with about 1.5 = 10 6 CHO-K1 cells in 10 ml culture medium and cells were grown for 24 h. After 24 h the medium was discarded and the cells were exposed for 4 h to at least three concentrations of the compound diluted in culture medium devoid of FCS in absence or in presence of a S9 mix Ž10% vrv.. After treatment medium was discarded and the cells were incubated with 10 ml of fresh culture medium for 16 h. Cells were then arrested in metaphase with colcemid ŽBoehringer. Ž10 mgrml. during 2 h and

B. Miller et al.r Mutation Research 392 (1997) 45–59

50

Table 1 Results of test compounds tested in the in vitro micronucleus assay ŽMNT. and in the in vitro chromosomal aberration assay ŽCA. Chemical

Structure or Kier classification

Nalidixid acid

Norfloxacin

Ofloxacin

a

a

Ciprofloxacin

CP 67804

a

a

a

Enrofloxacin

a

MNT in vitro cells used

Conc. range scored Žmgrml.

CA in vitro cells used

CHO-K5 yS9: yŽ180.

yS9: 25–180

CHO-K5 yS9: yŽ540.

yS9: 90–540

CHO-K5 yS9: yŽ400.

yS9: 50–400

CHO-K5 yS9: yŽ600. qS9: yŽ600.

yS9: 200–600 qS9: 100–600

CHO-K5 yS9: qŽ400.

yS9: 50–500

CHO-K5 yS9: qŽ400. qS9: yŽ1200.

yS9: 200–1200 qS9: 200–1200

CHO-K5 yS9: qŽ100.

yS9: 50–300

CHO-K5 yS9: qŽ200. qS9: qŽ400.

yS9: 200–800 qS9: 200–400

CHO-K5 yS9: qŽ10.

yS9: 5–50

CHO-K5 yS9: qŽ10. qS9: qŽ20.

yS9: 10–20 qS9: 20

CHO-K5 yS9: qŽ50.

yS9: 50–200

CHO-K5 yS9: qŽ100. qS9: qŽ500.

yS9: 100–1000 qS9: 500

yS9: 629–1000 V79 qS9: 943-1642 yS9: 60–150 V79 qS9: 80–200 yS9: 6–24 V79 qS9: 30–45 yS9: 150–340 V79 qS9: 270–450 yS9: 5–312.0 V79 qS9: 81.8–196.9 yS9: 10–19.6 V79 qS9: 60–119 yS9: 3301–5000 V79 qS9: 470–2169 yS9: 0.6–2.6 V79 qS9: 3.9–11.3 yS9: 406–625 V79 qS9: 406–625 yS9: 125–625 V79 qS9: 149–625 yS9: 23.8–40.0 V79 qS9: 279–482 yS9: 3129–5000 V79 qS9: 3129-5000

Compound 1

16

V79

Compound 2

5

V79

Compound 3

8, 15, 29

V79

Compound 4

4

V79

Compound 5

2, 3

V79

Compound 6

5, 25

V79

Compound 7

8, 15, 25

V79

Compound 8

6, 29

V79

Compound 9

n.d.

V79

Compound 10

18, 22, 25

V79

Compound 11

18, 14 or 10, 25

V79

Compound 12

26

V79

Result ŽLEDT, HIDT.

yS9: yŽ1000. qS9: yŽ1642. yS9: yŽ150. qS9: yŽ200. yS9: yŽ24. qS9: qŽ45.8 yS9: qŽ200. qS9: yŽ450. yS9: qŽ43.4. qS9: yŽ196.9. yS9: qŽ15.7. qS9: yŽ106. yS9: yŽ5000. qS9: yŽ2169. yS9: qŽ2.0. qS9: yŽ9.4. yS9: yŽ625. qS9: yŽ625. yS9: yŽ418. qS9: yŽ625. yS9: qŽ40.0. qS9: yŽ432. yS9: yŽ5000. qS9: yŽ5000.

Result ŽLEDT, HIDT.

Conc. range scored Žmgrml.

yS9: yŽ1100. qS9: yŽ1500. yS9: yŽ400. qS9: yŽ300. yS9: yŽ25. qS9: yŽ101. yS9: qŽ350.

yS9: 300–1100 qS9: 500–1500 yS9: 200–400 qS9: 150–300 yS9: 5–25 qS9: 19–101 yS9: 200–500

yS9: yŽ300.

yS9: 2–300

yS9: qŽ20. qS9: yŽ105. yS9: yŽ5000. qS9: yŽ5000. yS9: yŽ300. ) qS9: yŽ81. yS9: yŽ151. qS9: yŽ1000. yS9: yŽ366. qS9: yŽ437. yS9: qŽ40. qS9: yŽ437. yS9: yŽ5000. qS9: yŽ4440.

yS9: 14–26 qS9: 45–105 yS9: 250–5000 qS9: 1000–5000 yS9: 0.7–300 qS9: 8–81 yS9: 3–151 qS9: 151–1000 yS9: 43–366 qS9: 168–437 yS9: 24–65 qS9: 306–437 yS9: 2080–5000 qS9: 3942–5000

B. Miller et al.r Mutation Research 392 (1997) 45–59

51

Table 1 Žcontinued. Chemical

Structure or Kier classification

MNT in vitro cells used

Result ŽLEDT, HIDT.

Compound 13

26

V79

Compound 14

5, 26

CHO-K1

Compound 15

7, 28

CHO-K1

Compound 16

16, 24

CHO-K1

Compound 17

8, 25

CHO-K1

Compound 18

5, 8

CHO-K1

Compound 19

8, 25, 29

CHO-K1

Compound 20

24, 29

CHO-K1

Compound 21

2, 6, 7, 8

CHO-K1

Compound 22

2, 8, 25, 29, 30 CHO-K1

Compound 23

8, 25, 30

CHO-K1

Compound 24

5, 8, 30

CHO-K1

Compound 25

2, 8, 29, 30

CHO-K1

yS9: yŽ13.1. qS9: yŽ30.6. yS9: yŽ5000. qS9: yŽ5000. yS9: yŽ1000. qS9: yŽ1000. yS9: yŽ50. qS9: qŽ50. yS9: yŽ4.5. qS9: yŽ7.5. yS9: yŽ8. qS9: qŽ8. yS9: yŽ1000. qS9: yŽ1000. yS9: yŽ1000. qS9: qŽ15.5. yS9: yŽ600. qS9: qŽ300. yS9: yŽ2000. qS9: qŽ1500. yS9: yŽ8. qS9: qŽ20. yS9: yŽ1150. qS9: qŽ862. yS9: yŽ0.0025.

Compound 26

8, 25, 29

CHO-K1

Compound 27

8, 25, 30

CHO-K1

Compound 28

5, 8, 30

CHO-K1

Compound 29 Compound 30

2, 8, 23 5, 8, 30

CHO-K1 CHO-K1

Compound 31

7, 25, 29

CHO-K1

Compound 32 Compound 33

2 2, 8, 22

CHO-K1 CHO-K1

Compound 34

5, 22, 25

CHO-K1

Compound 35

8, 22, 29

CHO-K1

Compound 36

2, 8, 22

CHO-K1

Compound 37

8, 13, 25

CHO-K1

Compound 38

2, 8, 22

CHO-K1

Conc. range scored Žmgrml.

yS9: 8.6–13.1 qS9: 18.7–30.6 yS9: 555–5000 qS9: 555–5000 yS9: 100–1000 qS9: 100–1000 yS9: 12.5–50 qS9: 12.5–50 yS9: 1.5–4.5 qS9: 1.9–7.5 yS9: 2–8 qS9: 2–8 yS9: 100–1000 qS9: 100–1000 yS9: 250–1000 qS9: 15.5–62 yS9: 150–600 qS9: 150–600 yS9: 1000–2000 qS9: 500–2000 yS9: 1–8 qS9: 10–23 yS9: 575–1150 qS9: 575–1150 yS9: 0.0006– 0.0025 qS9: yŽ0.01. qS9: 0.0025–0.01 yS9: yŽ62.5. yS9: 31.5–62.5 qS9: yŽ100. qS9: 3–100 yS9: yŽ100. yS9: 1–100 qS9: yŽ100. qS9: 10–100 yS9: yŽ2750. yS9: 2000–2750 qS9: qŽ2600. qS9: 2500–2700 qS9: qŽ750. qS9: 87.5–750 yS9: qŽ1350. yS9: 1350–1650 qS9: qŽ1000. qS9: 1000–1200 yS9: qŽ0.3. yS9: 0.3–1.2 qS9: qŽ0.3. qS9: 0.3–1.2 qS9: qŽ60. qS9: 30–120 yS9: yŽ12. yS9: 3–12 qS9: yŽ12. qS9: 3–12 yS9: yŽ15. yS9: 5–15 qS9: qŽ15. qS9: 10–20 yS9: qŽ- 1.56. yS9: 1.56–6.25 qS9: qŽ- 3.12. qS9: 3.12–12.5 yS9: yŽ50. yS9: 2.5–50 qS9: qŽ5. qS9: 2.5–10 yS9: yŽ10. yS9: 2.5–10 qS9: yŽ10. qS9: 2.5–10 yS9: yŽ24. yS9: 6–24 qS9: qŽ4.5. qS9: 1.5–4.5

CA in vitro cells used

Result ŽLEDT, HIDT.

Conc. range scored Žmgrml.

V79 CHO-K1 Ža. CHO-K1 Ža. CHO-K1 Ža. CHO-K1 Žb. CHO-K1 Žb. CHO-K1 Ža. CHO-K1 Ža. CHO-K1 Ža. CHO-K1 Ža. CHO-K1 Ža. CHO-K1 Ža. CHO-K1

yS9: yŽ500. qS9: yŽ960. yS9: yŽ2500. qS9: yŽ2500. yS9: yŽ1000. qS9: yŽ1000. yS9: yŽ45. qS9: qŽ45. yS9: yŽ6. qS9: yŽ14. yS9: yŽ10. qS9: qŽ5. yS9: yŽ1250. qS9: yŽ1100. yS9: yŽ1000. qS9: yŽ45. yS9: yŽ500. qS9: yŽ500. yS9: yŽ2000. qS9: qŽ50. yS9: yŽ8. qS9: qŽ18. yS9: yŽ1150. qS9: yŽ575. yS9: yŽ0.03.

yS9: 33–500 qS9: 236–960 yS9: 500–2500 qS9: 500–2500 yS9: 250–1000 qS9: 250–1000 yS9: 1–45 qS9: 1–45 yS9: 3–6 qS9: 7–14 yS9: 2.5–10 qS9: 2.5–10 yS9: 156–1250 qS9: 80–1100 yS9: 500–1000 qS9: 5–45 yS9: 20–500 qS9: 20–500 yS9: 500–2000 qS9: 50–830 yS9: 1–8 qS9: 10–23 yS9: 575–1150 qS9: 57.5–575 yS9: 0.015–0.03

Ža. CHO-K1 Ža. CHO-K1 Ža. CHO-K1 Ža. CHO-K1Ža. CHO-K1 Žb. CHO-K1 Ža. CHO-K1Ža. CHO-K1 Ža. CHO-K1 Ža. CHO-K1 Ža. CHO-K1 Ža. CHO-K1 Ža. CHO-K1 Ža.

qS9: yŽ0.175. yS9: yŽ120. qS9: yŽ110. yS9: yŽ12.5. qS9: yŽ3. yS9: yŽ2750. qS9: qŽ1000. qS9: yŽ750. yS9: qŽ1650. qS9: qŽ850. yS9: yŽ2. qS9: yŽ1. qS9: qŽ1000. yS9: yŽ3. qS9: yŽ8. yS9: yŽ20. qS9: qŽ10. yS9: qŽ0.025. qS9: qŽ0.5. yS9: yŽ50. qS9: qŽ7.5. yS9: yŽ10. qS9: yŽ35. yS9: yŽ20. qS9: qŽ5.

qS9: 0.05–0.175 yS9: 60–120 qS9: 30–110 yS9: 0.8–12.5 qS9: 0.4–3 yS9: 1000–2750 qS9: 1000–2750 qS9: 100–750 yS9: 1250–1650 qS9: 400–850 yS9: 0.5–2 qS9: 0.1–1 qS9: 100–1000 yS9: 1–3 qS9: 2–8 yS9: 7–20 qS9: 0.5–10 yS9: 0.025–0.1 qS9: 0.5–1 yS9: 10–50 qS9: 5–12.5 yS9: 6–10 qS9: 25–35 yS9: 5–20 qS9: 1–10

B. Miller et al.r Mutation Research 392 (1997) 45–59

52 Table 1 Žcontinued. Chemical

Structure or Kier classification

MNT in vitro cells used

Result ŽLEDT, HIDT.

Conc. range scored Žmgrml.

CA in vitro cells used

Result ŽLEDT, HIDT.

Conc. range scored Žmgrml.

Compound 39

2, 6, 29, 30

CHO-K1

Compound 40

2, 5, 8, 22, 25

CHO-K1

Compound 41

2, 8, 22

CHO-K1 CHO-K1

Compound 43

5, 8

CHO-K1

Compound 44

2, 6, 7, 8

CHO-K1

Compound 45

25, 30

CHO-K1

Compound 46

2, 8, 25, 30

CHO-K1

Compound 47

8, 25

CHO-K1

Compound 48

2, 8, 25

HPBL

CHO-K1 Ža. CHO-K1 Ža. CHO-K1 Žb. CHO-K1 Ža. CHO-K1 Ža. CHO-K1 Žb. CHO-K1 Žb. CHO-K1 Žb. CHO-K1 Žb. HPBL

yS9: 15–60 qS9: 15–60 yS9: 2–8 qS9: 2–8 qS9: 1–5

28

yS9: 15–60 qS9: 15–60 yS9: 5–20 qS9: 5–20 yS9: 2–6 qS9: 2–6 yS9: 1250–5000 qS9: 1250–5000 yS9: 2–8 qS9: 2–8 yS9: 2.5–10 qS9: 2.5–10 yS9: 2.5–7.5 qS9: 4.9–19.5 yS9: 5–20 qS9: 5–20 yS9: 6–12 qS9: 6–12 yS9: 10–100 1

yS9: yŽ60. qS9: qŽ30. yS9: yŽ8. qS9: yŽ8. qS9: qŽ5.

Compound 42

yS9: yŽ60. qS9: yŽ60. yS9: yŽ20. qS9: yŽ20. yS9: yŽ6. qS9: qŽ4. yS9: yŽ5000. qS9: yŽ5000. yS9: yŽ8. qS9: yŽ8. yS9: yŽ10. qS9: yŽ10. yS9: yŽ7.5. qS9: yŽ19.5. yS9: yŽ20. qS9: yŽ20. yS9: yŽ12. qS9: yŽ12. yS9: yŽ100.

yS9: yŽ5000. qS9: yŽ5000. yS9: yŽ6. qS9: yŽ6. yS9: yŽ8. qS9: "Ž8. yS9: yŽ5. qS9: yŽ10. yS9: yŽ20. qS9: yŽ20. yS9: yŽ10. qS9: yŽ10. yS9: yŽ131. )

yS9: 1260–5000 qS9: 1260–5000 yS9: 2–6 qS9: 2–6 yS9: 3–8 qS9: 3–10 yS9: 3–5 qS9: 2.5–10 yS9: 15–20 qS9: 15–20 yS9: 6–10 qS9: 6–10 yS9: 66–102 Ž20 h., 94–131 Ž44 h.

qS9: qŽ30. qS9: yŽ300.

qS9: 10–100 qS9: 75–300 Ž2d assay.

qS9. yŽ250.

qS9: 160–250 Ž20 h.,

qS9: yŽ203. yS9: yŽ120. qS9: yŽ120. yS9: yŽ175. qS9. yŽ2250.

qS9: 164–203 Ž2d lab. 2 yS9: 51–120 2 qS9: 51–120 2 yS9: 50–175 qS9: 252–2250 1

yS9: yŽ117. qS9: yŽ180.

yS9: 49–117 qS9: 75–180 2

181–250 Ž44 h. 3

Compound 49

8, 25

HPBL

Compound 50

8, 25

HPBL

Compound 51

8, 25

HPBL

HPBL HPBL

HPBL

yS9: yŽ250. qS9: yŽ125. yS9: "Ž50. qS9: "Ž252. yS9: yŽ250. qS9: yŽ250.

yS9: 16–250 Ž18 h. qS9: 1–125 Ž32 h. 4 yS9: 50–125 Ž28 h. qS9: 252–1800 Ž22 h-28 h. 5 yS9: 50–250 Ž28 h. qS9: 50–250 Ž22–28 h.

a

Compounds published previously in Albertini et al., 1995. y: Negative, q: positive, ": equivocal. LEDT, lowest effective dose tested; HIDT, highest ineffective dose tested; 8, single dose positive; ) , negative for clastogenicity testing, but positive for polyploidy induction. Ža. 100 cells scored; Žb. 200 cells scored. 1, 2, 3, 4, 5: for explanation see Section 2.

mitotic cells were harvested by mitotic shake-off. An additional incubation of non mitotic cells was performed for 24 h. Cells were then arrested in metaphase and harvested as described above. After each harvest, the suspension was centrifuged and the cell pellet was resuspended in hypotonic 0.075 M KCl solution ŽProlabo. for 30 min and fixed in methanolracetic acid Ž3:1. ŽProlabo.. Smears were prepared on ice-cold slides, heat dried and stained

with 5% Giemsa ŽGiemsa.. One or two hundred metaphases were examined per concentration and time point. All types of structural aberrations were recorded but gaps, endoreduplicated chromosomes and polyploid cells were not included in the total incidence of chromosome aberrations. A compound was considered positive if at least one concentration induced a statistically significant increase in the number of cells with structural chromosome aberra-

B. Miller et al.r Mutation Research 392 (1997) 45–59

53

Table 2 Comparison of results from the in vitro MNT to the in vitro CA in the context of results from other genotoxicity tests Compound

MNT in vitro Žcells used.

CA in vitro Žcells used.

Bacterial gene mutation assay

Nalidixic acid Norfloxacin Ofloxacin Ciprofloxacin CP 67804 Enrofloxacin

ŽCHO-K5. y y q q q q

ŽCHO-K5. y y q q q q

q q q q q q

a

Compound 1 Compound 2 Compound 3 Compound 4 Compound 5 Compound 6 Compound 7 Compound 8 Compound 9 Compound 10 Compound 11 Compound 12 Compound 13

ŽV 79. y y q q q q y q y y q y y

ŽV 79. y y y q y q y y) y y q y y

y y y y y y y y y y y q y

b

Compound 14 Compound 15 Compound 16 Compound 17 Compound 18 Compound 19 Compound 20 Compound 21 Compound 22 Compound 23 Compound 24

ŽCHO-K1. y y q y q y q q q q q

ŽCHO-K1. y y q y q y y )) y q q y

yf yf n.d. n.d. yf n.d. n.d. yf n.d. yf yf

Compound 25 Compound 26 Compound 27 Compound 28 Compound 29 Compound 30 Compound 31 Compound 32 Compound 33 Compound 34 Compound 35 Compound 36 Compound 37

ŽCHO-K1. y y y q q q q q y q q q y

ŽCHO-K1. y y y q y )) q y q y q q q y

q y y y y y y y y y q y y

a a a a a

c d b b c d d d b b e d

f f f f f f f g f f f f f

Gene mutations in mammalian cells

MNT in vivo Žmouse.

ŽMLrTK. y y y q q q

n.d. n.d. n.d. n.d. n.d. n.d.

n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d.

n.d. n.d. y q n.d. q n.d. y n.d. y y n.d. n.d.

ŽCHO-K1rHPRT. y y y n.d. n.d. y q y n.d. y y

y y y n.d. y y y y y n.d. y

ŽCHO-K1rHPRT. y y y y y y ))) y " y y q y y

q y y y y y q n.d. y y q n.d. n.d.

B. Miller et al.r Mutation Research 392 (1997) 45–59

54 Table 2 Žcontinued. Compound

MNT in vitro Žcells used.

CA in vitro Žcells used.

Bacterial gene mutation assay

Compound 38 Compound 39 Compound 40 Compound 41 Compound 42 Compound 43 Compound 44 Compound 45 Compound 46 Compound 47

q y y q y y y y y y

q q y q y y " y y y

y y y y y y y y y y

f

Compound 48 Compound 49 Compound 50 Compound 51

ŽHPBL. " y y y

ŽHPBL. y) y " y

y y y y

h

f f f f f f f f g

h h h

Gene mutations in mammalian cells

MNT in vivo Žmouse.

y y y y y y y y y y

n.d. y y y y y y y q y

ŽMLrTK. y q " y

" y y y

y: Negative, q: positive, ": equivocal, n.d.: no data. ) Polyploidy induction. )) Positive, if eudoreduplicated chromosomes are included in the analysis. ))) Compound also tested in MLA and found positive. Bacterial gene mutation assay Žstrains tested.: a TA102; b TA1538, TA100; c TA1535, TA1537, TA1538, TA98, TA100; d TA1535, TA97a, TA98, TA100, TA102; e TA1535, TA97a, TA98, TA100, E. coli WP2 uvrA; f TA1535, TA1537, TA1538, TA98, TA100; g TA1535, TA1537, TA1538, TA98, TA100, E. coli WP2 uvrA; h TA 100, TA1535, TA1538, TA98, TA1537, TA97, E. coli WP2 uvrA.

tions, assessed by the Kastenbaum and Bowman test ŽKastenbaum and Bowman, 1970. in at least one of the assays conducted. 2.6.4. Biologie SerÕier After 48 h Žprotocols 3, 4. or 24 h Žprotocol 5. of precultivation, cultures were treated with the test compound or the positive controls for 2–3 h in the presence of metabolic activation or for 20 and 44 h Ž3., 18 h Ž4. or 24 h Ž5. in the absence of metabolic activation. For the treatment normal culture medium

containing 10% FCS was used. Colchicine ŽSigma. Ž1 Ž3. or 0.25 Ž4. mgrml. or colcemid ŽGibco BRL. Ž0.4 mgrml, protocol 4, 5. dissolved in water was added to the cultures 1–2 h before preparation. Harvest took place at 58 and 92 h Ž3., 66 and 80 h Ž4. or 46 and 52 h Ž5. after starting the cultures. Briefly, hypotonic treatment Ž0.075 M KCl, Merck. at 378C for protocols 3, 5, or 20% HBSS ŽGibco BRL. for protocol 4, was performed. Then the cells were fixed with methanolracetic acid Ž3:1. and dropped on glass slides. The cells were stained with

Table 3 Concordance of results from compounds tested both in the in vitro MNT and the in vitro CA

Concordant results Discordant results

In vitro micronucleus test

Chromosome aberration test

No. of compounds

%

Total

y q y q or "

y q q or " y

26 19 3 9

46.6 33.3 5.3 15.8

79% concordant 21% discordant

57 a b

Compound 39, 44, 50. Compound 3, 5, 8, 20, 21, 24, 29, 31, 48.

a b

B. Miller et al.r Mutation Research 392 (1997) 45–59

Giemsa 10% ŽMerck.. For scoring of chromosomal aberrations 100 metaphases per culture were analyzed on randomized coded slides. For a positive result statistically significant increases ŽFisher’s test Ž3 and 4., Likelihood ratio test for 5. or increases exceeding the normal range were used as criteria.

3. Results and discussion Our presentation of the data evaluation mainly concentrates on the aspect of the suitability of the in vitro micronucleus test as alternative for the generally accepted chromosome aberration assay in vitro. In this paper we do not concentrate on individual data andror protocols, but on the overall comparison between results of the chromosome aberration and the micronucleus assay in vitro. Information on historical control data are given in Table 5. Detailed data for 19 compounds ŽF. Hoffmann-La Roche and Novartis. are provided in an Appendix to this publication. In our evaluation 57 compounds, including six gyrase inhibitors previously published ŽAlbertini et al., 1995., are presented. Due to the confidential nature of the structure of most of the compounds tested Ži.e. under development as chemicals, agrochemicals, or pharmaceuticals., 51 compounds are presented coded. The compounds were classified according to the scheme proposed by Kier et al. Ž1986. into 30 different chemical classes. For a meaningful comparison a dataset must include results from a reasonable proportion of both positive and negative compounds and comprise a broad variety of class compounds. The compounds tested fulfil these criteria ŽTable 1.. Therefore, the comparison between the in vitro MNT and the in vitro CA is made valid on this database. Out of the 57 compounds tested in both, the MNT or CA in vitro ŽTable 2., 31 compounds showed positive results in either one or both test systems. The remaining 26 compounds were negative in both assays. Concordant results between the in vitro MNT and the in vitro CA were obtained for 45 out of 57 compounds Ž79% concordance, see Table 3.. Discordant results were obtained with 12 compounds Ž21%. of the dataset. Taking into account that the vast majority of

55

the tested compounds are weak inducers, the high concordance of the test results is promising. For discordant results between in vitro micronucleus and in vitro chromosome aberration assays several explanations can be put forward: Ž1. different concentration ranges tested, Ž2. different toxicity within both assays, Ž3. different sensitivity of the two assays. In the case of a positive CA in vitro, but a negative MN test in vitro, it has to be recognized, that the CA also detects non-stable aberrations that cannot be seen in the MNT, because chromosomal aberrations are detected in cells during the first mitosis after treatment regardless of the long-term ability of these cells to further divide andror survive. In the case of a positive in vitro MNT and a negative CA, the MN induction could be the result of an aneugenic potential of the compound. Analyzing the discordant results from our database in more detail following facts are obvious Žfor comparison see Tables 1 and 2.. Compound 3 was positive in the MNT in vitro after metabolic activation ŽqS9., however, a significant increase in the MN frequency over the control was obtained only for the highest concentration tested which revealed to be toxic. The compound was negative in the CA as well as in the bacterial gene mutation assay. No information about a possible aneugenic activity is available as well as no in vivo MN data. Compound 5 was positive in the MNT in vitro without S9 after 48 h sampling time, but was negative in the CA. No data about polyploidy-induction and no in vivo MNT data are available. Compound 8 was positive in the MNT in vitro Žwithout S9., but negative in the CA for chromosomal aberrations. It clearly induced polyploidy in the CA in vitro without metabolic activation. The MNT data in vivo are negative. Based on the polyploidy induction and the MN induction in vitro the compound is classified as aneugen. Compound 20 was positive in the MNT in vitro Žwith S9.. It was found positive in CA only when cells with endoreduplicated chromosomes were included in the analysis. The compound was negative in Ames test and the in vivo MNT, but positive in the gene mutation test using mammalian cells ŽCHOK1rHPRT.. Compound 21 was positive in the MNT in vitro

B. Miller et al.r Mutation Research 392 (1997) 45–59

56

Žwith S9., however, only a slight increase in the MN frequency over the control was obtained Ž2-fold increase.. It was negative in the CA as well as in all other tests Žbacterial gene mutation assay, gene mutation assay in mammalian cells ŽCHO-K1rHPRT., MNT in vivo.. Compound 24 was positive in the MNT in vitro Žwith S9.. It was highly toxic in the CA after metabolic activation, and the concentrations found positive in the MNT could not be evaluated in the CA. The compound was negative in all other tests ŽAmes, gene mutation assay ŽCHO-K1rHPRT., MNT in vivo.. Compound 29 was positive in the MNT in vitro Žwith S9.. It was found positive in CA only when cells with endoreduplicated chromosomes were included in the analysis. The compound was negative in all other tests ŽAmes, gene mutation test in mammalian cells ŽCHO-L1rHPRT., in vivo MNT.. Compound 31 was positive in the MNT in vitro Žwith and without S9., but negative in the CA for chromosomal aberrations. The compound is a spindle poison and also clearly positive in the in vivo MNT. Moreover, for a spindle poison the compound is not surprisingly negative in the bacterial gene mutation assay and gene mutation test using mammalian cells ŽCHO-K1rHPRT.. Compound 39 was negative in the MNT without using Cytochalasin B, but positive in the in vitro CA after metabolic activation Ž3.8-fold increase.. Interestingly, it was detected in the MNT using Cytochalasin B, however, the increase was not dramatic

Ž2.1-fold, from about 1% to about 2.1%, data not shown.. The compound was clearly negative in three additional assays Žbacterial gene mutation assay, gene mutation assay in CHO-K1 cells, MNT in vivo.. Compound 44 was negative in the in vitro MNT. It produced equivocal results in the CA Ž1 time positive, two times negative. after metabolic activation; the positive result could not be reproduced. The compound was clearly negative in three additional assays Žbacterial gene mutation assay, gene mutation assay in CHO-K1 cells, MNT in vivo.. Compound 48 was equivocal in the in vitro MNT with S9 using HPBL Žafter two different treatment protocols, see Table 1. It induced polyploidy in the in vitro CA Žwithout S9.. In both gene mutation assays Žbacterial gene mutation and MLrTK assay. no induction of mutants was observed. The induction of polyploidy and micronuclei in vitro indicates an aneuploidy-inducing potential of the compound. In vivo it led to a 2.3-fold increase of micronucleated polychromatic erythrocytes at a lethal dose, specifically 6 h after the treatment, which was not found at a sublethal dose. Compound 50 was negative in the in vitro MNT, but equivocal in the CA. The ML assay showed also equivocal results, whereas the MNT in vivo was negative. The aberrations identified were a low number of exchanges and pulverizations which could be due to apoptotic cell death. In summary, all the compounds classified as aneugens Žcompounds 8, 31, 48. were found positive or equivocal in the in vitro MNT and negative in the

Table 4 Compounds tested in the in vivo micronucleus test compared to the in vitro MNrCA results Number of compounds tested in the MNT and in the CA in vitro 57 concordant MNT yrCA y 26

MNT [rCA [ 19

discordant MNT [ or "rCA y 9

From the dataset above: Number of compounds tested additionally in the MNT in vivo 37 15 11 8 [: 2 a , y: 13 [: 3 b , y: 8 [: 1 c , ":1 c , y:6 a b c

Compounds 25 and 46. Compounds 4, 6, 35. Compounds 31 Žq., 48 Ž"..

MNT yrCA [ or " 3

3 [: 0, y: 3

B. Miller et al.r Mutation Research 392 (1997) 45–59 Table 5 Historical control data for the in vitro MNT in mammalian cell lines and HPBL Treatment conditions CHO-K5 a yS9, Negative control MEM MEMq1% DMSO MEMq2% DMSO HEPES HEPES q1% DMSO

% MN cells"SD 2.2 2.7 4.0 2.7 2.9

" " " "

0.5 0.7 1.3 0.5

Table 5 Žcontinued. Historical control data for the in vitro MNT in mammalian cell lines and HPBL

No. of experiments

Treatment conditions

% MN cells"SD

No. of experiments

15 11 3 2

qS9, Positive control CP 15 mgrml CP 25 mgrml CP 40 mgrml CP 50 mgrml

28.49"15.04 37.23"11.02 36.94"10.11 42.00"11.99

69 81 37 95

1.22" 0.39

16

HPBL

V79

yS9, Positive control BLEO 0.5 mgrml BLEO 1.0 mgrml BLEO 2.0 mgrml VCR 0.04 mgrml COL 0.02 mgrml

11.3 22.8 20.8 8.1 15.0

" 3.9 "10.4 " 6.3 " 0.4 " 1.3

5 16 5 2 2

qS9, Positive control CP 2.0 mgrml CP 4.0 mgrml CP 5.0 mgrml CP 8.0 mgrml

5.5 9.4 17.7 20.3

" 1.4 " 1.2 " 2.3 "10.6

6 3 5 7

yS9, Negative control, 3 h MEM MEMq1% DMSO

0.68" 0.21 17 0.44" 0.17 47

yS9, Negative control, 20 h MEM 0.36" 0.15 13 MEMq1% DMSO 0.45" 0.15 63 qS9, Negative control, 3 h MEM MEMq1% DMSO

0.56" 0.23 27 0.47" 0.19 90

yS9, Positive control EMS 6 mM EMS 15 mM

10.1 " 0.2 31 7.8 " 2.4 109

qS9, Positive control CP 10 mM

10.5 " 3.6 115

CHO-K1 yS9, Negative control yS9, Positive control MMC 0.25 mgrml MMC 0.5 mgrml MMS 40 mgrml MMS 60 mgrml qS9, Negative control

2.21" 0.52 278

16.60" 34.43" 30.05" 31.44"

57

3.32 72 9.06 34 9.17 17 9.53 150

2.70" 0.62 290

Negative controls

a

Only sampling times about 40 to 48 h are included. BLEO, Bleomycin; VCR, vincristine; COL, colchicine; CP, cyclophosphamide; EMS, ethyl methanesulfonate; MMC, mitomycin C; MMS, methyl methanesulfonate.

CA. For the three compounds which were positive only in the CA but not in the MNT, except for compound 24, the observed effects were borderline or equivocal. A comparison of the lowest effective dose tested ŽTable 1. indicates a potentially higher sensitivity of the MNT in vitro compared to the CA in vitro, which might be due to the fact, that more cells are analyzed for the MNT. The overall concordance of the test results of about 79% between the in vitro MNT and the in vitro CA is very promising. Taking into account the weakness of the effects observed with those compounds inducing only chromosomal aberrations but no MN, and the mechanistic explanation for some of the compounds showing induction of MN but no chromosomal aberrations Žaneugens, spindle poisons induction of endoreduplicated chromosomes., the overall concordance for clastogenicity is even higher Žabout 88%.. The high concordance of results between both assays is remarkable since for our experiments using established cell lines no cytochalasin B was used. The cytokinesis-block method using cytochalasin B allows better control of cell division kinetics and is, therefore, regarded to be more sensitive. The aspect of sensitivity should be further evaluated. Nevertheless, our data clearly show that good results and correlations can be obtained even when not using the cytokinesis block method. The high concordance between the MNT and the

58

B. Miller et al.r Mutation Research 392 (1997) 45–59

CA is in agreement with other studies performed recently. In a study investigating 11 clastogenic chemicals or spindle poisons simultaneously, using a conventional chromosomal aberration assay and the in vitro micronucleus test with the Chinese hamster cell line CHL ŽMatsuoka et al., 1993,. all 11 chemicals were positive in both assays Ž100% concordance. and the lowest effective dose tested were in the same concentration range. Li et al. Ž1993. analyzed 30 chemicals in the in vitro MNT in Chinese hamster lung cells – 21 carcinogens, 5 questionable carcinogens and 4 non-carcinogens. All carcinogens were detected in the MNT in vitro, while 18 of these carcinogens were detected in the Ames test. Out of the 22 chemicals that were positive in the chromosomal aberration test in vitro, 20 were also found to be positive in the MNT in vitro. The two negative compounds were caprolactam, which in the CA test was tested up to doses exceeding 10 mM, and tegafur, which was found to be highly cytotoxic. In our study out of the 57 compounds tested in vitro, 37 were in addition evaluated in the MNT in vivo ŽTable 4.. Only three compounds Žcompounds 4, 6 and 35. out of the 19 found positive in both, in vitro MNT and CA, induced an increase in the number of micronucleated erythrocytes in vivo. A higher rate of positive results in the in vitro cytogenetic tests compared to the in vivo assay was already reported by Thomson Ž1986.. According to Thomson Ž1986. out of 181 compounds tested both in vitro and in vivo, 115 were positive in the MNT in vitro whereas 62 turned out to be positive in the in vivo MNT. Interestingly, in our dataset there were two compounds found negative for both the in vitro MNT and in vitro CA, but positive or equivocal in the MNT in vivo in mouse bone marrow Žcompounds 25 and 46.. Compound 25 induced in vitro an irreversible blockade in G1 of the cell cycle and was positive in the Ames test; compound 46 was tested in vitro only up to 20 mgrml due to its poor solubility in vitro. Similarly in the study reported by Thomson Ž1986., two compounds were found positive in vivo only. This finding, contradictory to the statement of Ashby and Tinwell Ž1996. that so far no mutagenic compounds are known that are only detected in the in vivo assay but not in in vitro assays, needs further clarification.

4. Conclusion Our data showing high concordance between the in vitro MNT and the in vitro CA strongly support the conclusion that the in vitro MN test is an acceptable alternative to the in vitro chromosome aberration assay for regulatory purposes. Nevertheless, there is still a need for further standardization of the protocol. When considering routine testing, the in vitro MNT has several advantages over the conventionally used in vitro CA: simplicity of the method, rapidness of the assay, more statistical power Ždue to the fact that more cells are analyzed., and possibility for automated scoring. In addition the MN test has the potential to detect, besides clastogenicity, aneugenicity of compounds, an endpoint not included in the currently recommended in vitro genotoxicity test battery.

References Albertini, S., Chetelat, A.-A., Miller, B., Muster, W., Pujadas, E., ´ Strobel, R., Gocke, E., 1995. Genotoxicity of 17 gyrase- and four mammalian topoisomerase II-poisons in prokaryotic and eukaryotic test systems. Mutagenesis 10, 343–351. Ames, B.N., McCann, J., Yamasaki, E., 1975. Methods for detecting carcinogens and mutagens with the Salmonellarmammalian microsome mutagenicity test. Mutation Res. 31, 347– 364. Ashby, J., 1986. The prospects for a simplified and internationally harmonized approach to the detection of possible human carcinogens and mutagens. Mutagenesis 1, 3–16. J. Ashby, H. Tinwell, The rodent bone marrow micronucleus assay: Contrast between its sensitivity to human carcinogens and its intensitivity to NTP rodent carcinogens. Mutation Res. 352 Ž1996. 181–184. Borenfreund, E., Puerner, J.A., 1985. Toxicity determined in vitro by morphological alterations and neutral red absorption. Toxicol. Lett. 24, 119–124. Countryman, P.I., Heddle, J.A., 1976. The production of micronuclei from chromosome aberrations in irradiated cultures of human lymphocytes. Mutation Res. 41, 321–332. Ellard, S., Mohammed, Y., Dogra, S., Wolfel, C., Doehmer, J., ¨ Parry, J.M., 1991. The use of genetically engineered V79 Chinese hamster cultures expressing rat liver CYP1A1, 1A2 and 2B1 cDNAs in micronucleus assays. Mutagenesis 6, 461– 470. Fenech, M., 1993. The cytokinesis-block micronucleus technique: A detailed description of the method and its application to genotoxicity studies in human populations. Mutation Res. 285, 35–44.

B. Miller et al.r Mutation Research 392 (1997) 45–59 Kastenbaum, M.A., Bowman, K.O., 1970. Tests for determining the statistical significance of mutation frequency. Mutation Res. 9, 527–549. Kier, L.D., Brusick, D.J., Auletta, A.E., Von Halle, E.S., Simmon, V.F., Brown, M.M., Dunkel, V.C., McCann, J., Mortelmans, K., Prival, M., Rao, T.K., Ray, V.A., 1986. The Salmonella typhimuriumrmammalian microsomal assay report of the U.S. Environmental Protection Agency Gene-Tox Program. Mutation Res. 168, 69–240. Lasne, C., Gu, Z.W., Venegas, W., Chouroulinkov, I., 1984. The in vitro micronucleus assay for detection of cytogenetic effects induced by mutagens-carcinogens: comparison with the in vitro sister-chromatid exchange assay. Mutation Res. 130, 273–282. Li, J., Suzuki, Y., Shimizu, H., Fukumoto, M., Okonogi, H., Nagashima, T., Ishikawa, T., 1993. In vitro micronucleus assay of 30 chemicals in CHL cells. Jikeikai Med J 40, 69–83. Lowry, O.H., Rosebrough, N.J., Farr, A.L., Randall, R.J., 1951. Protein measurement with the folin phenol reagent. J. Biol. Chem. 193, 265–275. Maron, D.M., Ames, B.N., 1983. Revised methods for the Salmonella mutagenicity test. Mutation Res. 113, 173–215. Matsuoka, A., Yamazaki, N., Suzuki, T., Hayashi, M., Sofuni, T., 1993. Evaluation of the micronucleus test using a Chinese hamster cell line as an alternative to the conventional in vitro chromosomal aberration test. Mutation Res. 272, 223–236. T. Matsushima, M. Sawamura, K. Hara, T. Sugimura A safe substitute for polychlorinated biphenyls as an inducer of metabolic activation system, in: F.J. de Serres, J.R. Fouts, J.R. Bend, R.M. Philpot ŽEds.., In Vitro Metabolic Activation in Mutagenesis Testing, Elsevier, Amsterdam, 1976, pp. 85–88.

59

Migliore, L., Barale, R., Belluomini, D., Gognetti, A.G., Loprieno, N., 1987. Cytogenetic damage in human lymphocytes by adriamycin and vincristine: A comparison between micronucleus and chromosomal aberration assays. Toxicol. In Vitro 1, 247–254. Miller, B.M., Pujadas, E., Gocke, E., 1995. Evaluation of the micronucleus test in vitro using Chinese hamster cells: results of four chemicals weakly positive in the in vivo micronucleus test. Environ. Mol. Mutagen. 26, 240–247. Mossmann, T., 1983. Rapid colorimetric assay for cellular growth and survival: Application to proliferation and cytotoxicity assays. J. Immunol. Methods 65, 55–63. Salassidis, K., Kulka, U., Schmid, E., Paul, D., Bauchinger, M., 1991. Induction of chromosome aberrations and sister chromatid exchanges by indirectly acting mutagens in immortal mouse and rat hepatocyte lines. Mutagenesis 6, 59–63. Savage, J.R.K., 1976. Classification and relationship of induced chromosomal structural changes. J. Med. Genet. 13, 103–122. Scott, D., Galloway, S.M., Marshall, R.R., Ishidate, M., Brusick, D., Ashby, J., Myhr, B.C., 1991. Genotoxicity under extreme culture conditions. A report from ICPEMc Task Group 9. Mutation Res. 257, 177–204. Schmuck, G., Lieb, G., Wild, D., Schiffmann, D., Henschler, D., 1988. Characterization of an in vitro micronucleus assay with Syrian hamster embryo fibroblasts. Mutation Res. 203, 297– 404. Thomson, E.D., 1986. Comparison of in vivo and in vitro cytogenetic assay results. Environ. Mutagen. 8, 753–767. Vian, L., Bichet, N., Gouy, D., 1993. The in vitro micronucleus test on isolated human lymphocytes. Mutation Res. 291, 93– 102.