Mutagenesis vol.17 no.3 pp.219–222, 2002
Griseofulvin induces mitotic delay and aneuploidy in bone marrow cells of orally treated mice
Francesca Pacchierotti3, Bruno Bassani, Francesco Marchetti1 and Cecilia Tiveron2 Section of Toxicology and Biomedical Sciences, ENEA CR Casaccia, Via Anguillarese 301, 00060 Santa Maria di Galeria, Roma, Italy 1Present
address: Biology and Biotechnology Research Program, Lawrence Livermore National Laboratory, L-448, Livermore, CA, USA
2Present
address: Centro Ricerca Sperimentale, Istituto Regina Elena, Roma,
Italy
Griseofulvin (GF) is a fungicide drug well characterized for its aneugenic activity in vitro. In vivo strong evidence of aneuploidy and polyploidy induction has been obtained in germ cells, especially in oocytes. More controversial are the data on the aneugenicity of griseofulvin in somatic cells. In this paper we provide evidence that GF induces non-disjunction and cell cycle delay in bone marrow cells of orally treated mice. Adult female mice were administered olive oil suspensions of 200, 666 or 2000 mg/kg GF by gavage and killed 18 or 24 h later. To minimize animal-toanimal variation in the absorption and distribution of GF, mice were fasted from the time of GF administration to the time of killing. Two hours before treatment the animals were s.c. implanted with a bromodeoxyuridine tablet to obtain differential chromatid staining and to determine the number of divisions after GF treatment for each metaphase. Mitostatic effects of GF were assessed by the relative proportions of first, second and third generation metaphases and the average generation time (AGT) method. A statistically significant increase with respect to the control AGT value was observed after treatment with 666 and 2000 mg/kg, suggesting that GF, as already shown in meiosis, interfered with cell cycle progression. Hyperploidy was scored in second generation metaphases. Eighteen hours after treatment, the frequencies of hyperploid cells were significantly (P < 0.05) higher in all GF-treated groups than in their matched control group. The effect was not dose-dependent. No further increase in aneuploidy was observed at 24 h, suggesting that cells overcoming mitotic arrest did not have a higher rate of non-disjunction. No induction of polyploidy was demonstrated. We conclude that GF induces mitotic delay and aneuploidy in mouse bone marrow and suggest that the protocol used to formulate the gavage suspensions and the after-treatment fasting of the animals enhanced the bioavailability of GF to bone marrow cells. Introduction Griseofulvin (CAS no. 126-07-8) (GF) is a fungicide drug extensively used to treat dermatomycoses in humans and animals (Knasmu¨ller et al., 1997). Its therapeutic efficacy is due to inhibition of chromosome segregation and cell division through binding to microtubule-associated proteins (Roobol
et al., 1977). It has been shown to be teratogenic (Klein and Beall, 1972) and carcinogenic (Rustia and Shubik, 1978) in rodents. Induction of aneuploidy by non-disjunction and chromosome loss has been reported in human lymphocytes treated in vitro with GF (Kolachana and Smith, 1994; Migliore et al., 1996, 1999). The effect was dose-dependent up to ~15 µg/ml, with higher concentrations resulting in a smaller effect, possibly because of cell toxicity (Raimondi et al., 1989). In agreement with these data, different types of spindle abnormalities have been observed after GF treatment of primary and transformed cell cultures (Kochendo¨rfer et al., 1996), including severe alterations such as multipolarity, monopolarity and fiber disorganization, which are likely to cause mitotic arrest, apoptosis or cell death. GF interferes with chromosome segregation during the first and second meiotic divisions of mouse oocytes, where it induces meiotic delay and aneuploidy in a time- and dosedependent manner (Tiveron et al., 1992; Mailhes et al., 1993; Marchetti and Mailhes, 1994, 1995; Marchetti et al., 1996). As many as 40% of analyzed oocytes and zygotes were hyperploid at the most effective doses tested and polyploid chromosome sets were also significantly increased in GFtreated mice. In contrast, only a doubling of the spontaneous aneuploidy rate was detected in mouse sperm after in vivo treatment with GF (Shi et al., 1999). In addition to technical reasons, including the difficulty of assessing time windowsensitive effects in an asynchronous cell population such as spermatocytes, biological differences in the structure and function of spindle and cell cycle checkpoints between male and female meiosis may account for the higher susceptibility of mammalian oocytes to spontaneous and chemically induced aneuploidy (Eichenlaub-Ritter and Sun, 2001). In spite of the well-characterized in vitro and germ cell in vivo aneugenicity of GF, few and contradictory data exist in somatic cells in vivo (Aardema et al., 1996; Vanhauwaert et al., 2001). Aardema and co-workers reported that oral administration of 2–5 g/kg body wt GF significantly increased the frequency of kinetochore-positive micronuclei in bone marrow erythrocytes 24 and 48 h after treatment (Aardema et al., 1996). On the other hand, a significant increase in micronucleated cells in the gut epithelium but no induction of micronuclei in bone marrow were recently reported by Vanhauwaert et al. (2001) 24 and 48 h after oral treatment of mice with 1 or 1.5 g/kg GF. In this paper we report on the cytogenetic analysis of bone marrow metaphases from mice orally treated with GF. Our results show that GF induces cell cycle delay and nondisjunction. Materials and methods (C57Bl/Cne⫻C3H/Cne) F1 female mice, 8–10 weeks old, were orally treated with 200, 666 or 2000 mg/kg GF (Sigma, St Louis, MO) by gavage. GF was
whom correspondence should be addressed. Tel: ⫹39 06 30484442; Fax: ⫹39 06 30483805; Email:
[email protected] *This paper is dedicated to Pietro Metalli on the occasion of his 70th birthday.
3To
© UK Environmental Mutagen Society/Oxford University Press 2002
219
F.Pacchierotti et al. suspended by magnetic stirring in olive oil at a concentration adjusted to treat the animals with a volume of 0.03 ml/g body wt. Control animals received a comparable volume of olive oil. Each experimental group was composed of four mice. In order to minimize animal-to-animal variation in the absorption and distribution of GF, mice were fasted from the time of GF administration to the time of killing. Two hours before treatment mice were s.c. implanted with a 25 mg agar-coated bromodeoxyuridine tablet (BrdU) (Boehringer, Mannheim, Germany) under anesthesia. BrdU labeling of bone marrow cell DNA was used to evaluate cell cycle kinetics by measuring the relative proportions of first (M1), second (M2) and third (M3) generation metaphases, which could be recognized according to their differential chromatid staining achieved by the Hoechst plus Giemsa technique (Perry and Wolff, 1974). In addition, the identification of second generation metaphases allowed us to selectively estimate aneuploidy in those cells that had divided only once following treatment (Manca et al., 1990). Mice were killed by cervical dislocation 18 or 24 h after GF treatment to account for aneuploidy or polyploidy possibly arising after mitotic delay (Manca et al., 1990). Two hours before killing, animals were i.p. injected with 0.3 ml of 10–3 M colchicine (Carlo Erba, Milano, Italy). Bone marrow metaphases were prepared according to standard techniques and the slides were stained by the Hoechst plus Giemsa method to reveal the DNA which had incorporated BrdU. Slides were coded and randomized before scoring. The occurrence of segregation errors was detected by scoring hyperploid and polyploid metaphases. Hypoploid chromosome sets were not recorded because an unknown proportion of them could result from technical artefacts. Less than unequivocal hyperploid metaphases were recorded only when confirmed by two independent observers. Cell replication kinetics was assessed by the average generation time (AGT) method (Ivett and Tice, 1982). One hundred metaphases/mouse were classified as M1, M2 or M3 according to their staining pattern. Only metaphases with homogeneous staining of all chromosomes were recorded as M1 (undifferentiated chromatids) or M2 (differentiated chromatids). They represented cells which had not yet entered S phase until BrdU incorporation started and, for this reason, were most likely exposed to GF (administered 2 h after BrdU tablet implantation) before metaphase onset. AGT was computed as follows: duration of BrdU administration (colchicine treatment time included)/[(1⫻M1 proportion) ⫹ (2⫻M2 proportion) ⫹ (3⫻M3 proportion)]. Aneuploidy was estimated as the frequency of hyperploid cells in a total of 100 euploid ⫹ hyperploid M2 metaphases per mouse. Polyploid cells, which were recognized by metaphases with a unitary appearance with ⬎70 chromosomes of homogeneous morphology, were recorded with the total of M2 scored during the evaluation of cell cycle kinetics. Mean differences between AGT values in control and treated animals were statistically evaluated by two-tailed Student’s t-test. χ2 analysis was used to test for possible heterogeneity in the frequencies of hyperploidy within experimental groups and for differences in the mean hyperploidy frequencies between the control and treated groups. For homogeneous groups the data from each animal were pooled and comparisons between groups were made using a 2⫻2 χ2 test. Heterogeneous groups were compared using a variance ratio value (F) calculated from the between and within group χ2 values.
Results Animal fasting did not affect cell cycle progression or chromosome segregation in bone marrow cells. In fact, AGT values measured in control mice of the present study (10.6 ⫾ 0.1 at 18 h, 11.7 ⫾ 0.2 at 24 h) were comparable with those measured in previous experiments in animals fed ad libitum, ranging from 10.2 ⫾ 0.1 to 10.4 ⫾ 0.4 at 18 h and from 11.2 ⫾ 0.4 to 12.3 ⫾ 0.4 at 24 h (Pacchierotti et al., 1991). Similarly, the frequency of 0.5% hyperploid metaphases observed in this study did not differ with respect to a mean value of 0.4 ⫾ 0.5% observed in 18 mice of the same strain with unrestricted food access (Pacchierotti et al., 1991). The data on cell cycle progression in solvent- and GFtreated mice are shown in Table I. Eighteen hours after treatment with 666 and 2000 mg/kg GF, the proportion of second generation metaphases decreased from the 90% observed in the controls and 200 mg/kg GF-treated groups to 76 and 72%, respectively. This was confirmed, 6 h later, by the decreased proportion of third generation metaphases, from 26% in controls to 9 and 2% at 666 and 2000 mg/kg, respectively. Consequently, an increase in AGT value was 220
Table I. Cell cycle progression in bone marrow cells after GF treatment M1 (% ⫾ SE) M2 (% ⫾ SE) M3 (% ⫾ SE) AGT (⫾ SE)
Dose (mg/kg) Harvest time 18 0 200 666 2000 Harvest time 24 0 200 666 2000
h 10.5 10.0 24.0 27.8
⫾ ⫾ ⫾ ⫾
2.3 4.3 2.4 7.3
89.5 90.0 76.0 72.2
⫾ ⫾ ⫾ ⫾
2.3 4.3 2.4 7.3
0 0 0 0
3.8 2.5 4.0 4.5
⫾ ⫾ ⫾ ⫾
0.9 0.6 1.2 1.4
70.2 76.5 86.8 93.5
⫾ ⫾ ⫾ ⫾
1.9 5.3 4.4 1.0
26.0 21.0 9.2 2.0
h
a,bTwo-tailed Student’s aP ⬍ 0.01; bP ⬍ 0.05.
10.56 ⫾ 0.13 10.54 ⫾ 0.24 11.37 ⫾ 0.16a 11.66 ⫾ 0.51b ⫾ ⫾ ⫾ ⫾
2.6 5.7 5.5 0.8
11.68 ⫾ 0.20 11.92 ⫾ 0.33 12.70 ⫾ 0.39 13.17 ⫾ 0.13
t-test with respect to matched solvent-treated group:
observed in GF-treated mice, which was statistically significant for the 666 and 2000 mg/kg groups at the 18 h harvest time only. The GF effects on cell cycle progression were best observed at the shortest interval between treatment and killing, because at this time the AGT values were less affected by cell recovery processes and/or lethality. Table II shows the results for GF-induced aneuploidy in mouse bone marrow cells. The statistical analysis of withingroup variance demonstrated a significant heterogeneity among the 666 mg/kg GF-treated mice killed 24 h after GF administration only. Eighteen hours after treatment, the frequencies of hyperploid cells were significantly higher in all GF-treated groups than in their matched control group. The effect was not dose-dependent. No further increase in aneuploidy was observed at 24 h, suggesting that cells overcoming mitotic arrest did not have a higher rate of non-disjunction. Grossly hyperploid metaphases (2n ⫽ 43–50) were slightly more frequent in GF-treated groups, especially 24 h after exposure, but differences from control values were not statistically significant. The frequency of polyploid cells ranged from 0 to 0.6% (one or two polyploid metaphases recorded in some experimental groups) with no relationship with GF treatment or harvest time. Discussion The induction of aneuploidy is the best characterized genotoxic effect of GF in vitro. It has been suggested that aneuploidy induction is responsible for the carcinogenicity of GF (Knasmu¨ ller et al., 1997). In vivo a strong aneugenic activity has been shown in mouse oocytes, including the induction of hyperploidy, polyploidy and meiotic delay (Marchetti et al., 1996). However, in vivo data in somatic cells are controversial: both positive (Aardema et al., 1996) and negative (Leonard et al., 1979; Vanhauwaert et al., 2001) results have been reported on the induction of micronuclei in bone marrow erythrocytes, whereas the induction of micronuclei has been demonstrated in gut epithelium (Vanhauwaert et al., 2001). The induction of SCEs in bone marrow cells of mice that had been i.p. injected with GF (Curry et al., 1984) confirmed that chemical biodistribution did not exclude this compartment. The results of this study show that GF can induce both cell cycle delay and aneuploidy in mouse bone marrow. The significant decrease in metaphases at the second cell cycle observed in our experiments 18 h after GF treatment demonstrated the induction of a cell cycle delay, which confirms the activity of GF in bone marrow cells and agrees with observa-
Griseofulvin-induced mitotic delay and aneuploidy
Table II. Frequency of hyperploidy in bone marrow cells after GF treatment Dose (mg/kg)
Harvest time 18 h 0 200 666 2000 Harvest time 24 h 0 200 666 2000
Analyzed metaphases (2n 艌 40)
2n ⫽ 41 or 42
2n ⫽ 43–50
No.
Percent ⫾ SE
402 401 400 402
2 (0, 0, 1, 1) 12 (0, 3, 3, 6) 11 (2, 2, 2, 5) 11 (1, 2, 3, 5)
0.5 3.0 2.8 2.8
⫾ ⫾ ⫾ ⫾
400 402 403 403
2 4 10 14
0.5 1.3 2.5 3.5
⫾ ⫾ ⫾ ⫾
(0, (0, (0, (2,
0, 1, 1, 2,
1, 1, 3, 4,
1) 2) 6) 6)
No.
Percent ⫾ SE
0.3 1.4a 0.9a 1.0a
2 (0, 0, 0, 2) 0 0 2 (0, 0, 0, 2)
0.5 ⫾ 0.6 0 0 0.5 ⫾ 0.6
0.3 0.8 1.5 1.1a
0 2 (0, 0, 1, 1) 3 (0, 1, 1, 1) 3 (0, 0, 1, 2)
0 0.5 ⫾ 0.3 0.8 ⫾ 0.3 0.8 ⫾ 0.6
Data for individual animals are shown in parentheses. aχ2 test: P ⬍ 0.05 with respect to matched solvent-treated group.
tions on meiotic delay in male (Shi et al., 1999) and female (Mailhes et al., 1993) germ cells. The frequencies of hyperploidy observed 18 h after treatment with 200, 666 and 2000 mg/kg were all statistically higher than the control value and very similar among each other. This result is in line with a non-linear absorption capacity of the duodenum (the main site of absorption), as measured by plasma peak levels at doses between 100 and 10 000 mg/kg GF (Knasmu¨ ller et al., 1997). Also, the frequencies of hyperploidy induced by oral treatment of female mice with 500, 1000 or 1500 mg/kg GF during the first and second meiotic divisions did not differ among each other (Marchetti et al., 1996). The 7-fold increase in hyperloid cells observed in this study is more comparable with that found in male germ cells (Shi et al., 1999) than with the 80-fold increase found in female germ cells (Marchetti and Mailhes, 1994). Two factors that may contribute to the higher sensitivity of female germ cells are: (i) the ability to target the window of maximum sensitivity because of the synchronicity of meiotic progression among oocytes; (ii) a less stringent control of the integrity and function of the spindle by the meiotic checkpoints of female germ cells (Eichenlaub-Ritter and Sun, 2001). We did not obtain evidence of polyploidy induction by GF. Other chemicals, such as the classic spindle poisons colchicine and vinblastine, tested in bone marrow under comparable experimental conditions have been shown to dose-dependently induce polyploid cells (Manca et al., 1990; Pacchierotti et al., 1991). In those cases, the frequency of polyploid cells was higher at longer after-treatment sampling times, suggesting that polyploidy arose by a faulty overcoming of a temporary mitotic arrest. GF is known to bind to tubulin dimers and microtubule-associated proteins and to interfere with microtubule polymerization, but the molecular mechanism of GFinduced spindle disorganization is different from that of colchicine and vinblastine (Knasmu¨ ller et al., 1997). It is conceivable that different mechanisms of spindle poisoning affect the subsequent capacity of cell recovery. This is supported by experiments on sequential harvesting of oocytes treated with GF (Mailhes et al., 1993) or vinblastine (Mailhes and Marchetti, 1994) before the first meiotic division. These experiments showed that oocytes temporarily arrested by GF at meiosis I recovered by an error-free mechanism forming a functioning spindle and completing anaphase I. Conversely, after VBL-induced meiotic arrest spindle function was irrevers-
ibly suppressed, homologous chromosomes disjoined but could not segregate and diploid metaphase II oocytes were produced. It is noteworthy that GF was shown to induce polyploid sperm (Shi et al., 1999) and even oocytes (Marchetti and Mailhes, 1994) when administered a short time before the first meiotic division. We can speculate that in both cases cells had been under pressure to trigger anaphase, either by the synchronously dividing syncytial spermatocyte nuclei or by the approaching ovulation, and that under those circumstances they were unable to assemble a functioning spindle. Our results demonstrate that, under our specific experimental conditions, GF induced non-disjunction and cell cycle delay in mouse bone marrow. A significant increase in the frequency of kinetochore-positive micronuclei in bone marrow erythrocytes of GF-treated mice was reported in an abstract by Aardema et al. (1996). At variance with these results, no evidence of micronucleus induction in mouse bone marrow was reported by Vanhauwaert et al. (2001). Because of the limited information in the experimental design provided by Aardema et al. (1996), it is not possibile to reconcile the differences between the two micronucleus studies. On the other hand, various factors might be responsible for the different outcomes between our study and that of Vanhauwaert et al. (2001). First, the sampling time: if the period of GF sensitivity during the cell cycle is short and close to mitosis, the sampling times used by Vanhauwaert and co-workers to evaluate micronucleus induction (24 and 48 h) might have missed exposure of the most susceptible cells. Furthermore, differences in formulation of the GF suspension and in the availability of food following chemical treatment might have been critical. We used olive oil as the vehicle and fasted the mice from the time of treatment to killing, while Vanhauwaert and co-workers used Methocel–Tween as the vehicle and gave their animals free access to food throughout the experiment. Clinical studies in humans have shown large subject-to-subject variation in the absorption of orally administered GF and have concluded that bioavailability of the drug is influenced by particle size, the vehicle and the fasting condition of the subjects (Bates et al., 1977; Aoyagi et al., 1982; Lin et al., 1982). In addition, studies in mice have shown that removal of food before GF administration resulted in a 4-fold increase in the frequencies of aneuploid oocytes (Marchetti and Mailhes, 1994). It is possible that the protocol we used to formulate 221
F.Pacchierotti et al.
the gavage suspensions (concentrations of GF ranging from 6.7 to 66.7 mg/ml in olive oil maintained under continuous magnetic stirring until administration) and the after-treatment fasting of the animals have enhanced GF absorption and its distribution to bone marrow. Acknowledgement This work was partially supported by EU Project ‘Protection of the European Population from Aneugenic Chemicals’ (contract no. QLK4-CT-200000058).
References Aardema,M.J., Curry,P.T., Gudi,R. and Putman,D.L. (1996) Griseofulvininduced structural and numerical chromosome aberrations in vitro and in vivo. Environ. Mol. Mutagen., 27, s27. Aoyagi,N., Ogata,H., Kaniwa,N. and Ejima,A. (1982) Effect of food on the bioavailability of griseofulvin from microsize and PEG ultramicrosize (GRIS-PEG) plain tablets. J. Pharmacobiodyn., 5, 120–124. Bates,T.R., Pieniaszek,H.J., Sequeira,J.A. and Rasmussen,J.E. (1977) Gastrointestinal absorption of griseofulvin from corn oil-in-water emulsions: effect of amount of corn oil ingested in man. Arch. Dermatol., 113, 302–306. Curry,P.T., Reed,R.N., Martino,R.M. and Kitchin,R.M. (1984) Induction of sister-chromatid exchanges in vivo in mice by the mycotoxins sterigmatocystin and griseofulvin. Mutat. Res., 137, 111–115; Eichenlaub-Ritter,U. and Sun,F. (2001) In vitro assays for detection of aneuploidy and influence of aneugens on spindle formation and chromosome distribution in mammalian oocytes. Mutagenesis, in press. Ivett,J.L. and Tice,R.R. (1982) Average Generation Time: a new method of analysis and quantitation of cellular replication kinetics. Environ. Mutagen., 4, 358. Klein,M.F. and Beall,J.R. (1972) Griseofulvin: a teratogenic study. Science, 175, 1483–1484. Knasmu¨ ller,S., Parzefall,W., Helma,C., Kassie,F., Ecker,S. and SchulteHermann,R. (1997) Toxic effects of griseofulvin: disease models, mechanisms and risk assessment. Crit. Rev. Toxicol., 27, 495–537. Kochendo¨ rfer,U, Stammberger,I., Mayer,D. and Schwanitz,G. (1996) A new possible parameter for the detection of aneuploidy inducing substances: the analysis of qualitative and quantitative abnormalities of the spindle apparatus. Mutat. Res., 361, 55–66. Kolachana,P. and Smith,M.T. (1994) Induction of kinetochore-positive micronuclei in human lymphocytes by the anti-fungal drug griseofulvin. Mutat. Res., 322, 151–159. Leonard,A., Poncelet,F., Grutman,G., Carbonelle,E. and Fabry,L. (1979) Mutagenicity tests with griseofulvin. Mutat. Res., 68, 225–234. Lin,C., Lim,J., DiGiore,C., Gural,R. and Symchowicz,S. (1982) Comparative bioavailability of a microsize and ultramicrosize griseofulvin formulation in man. J. Int. Med. Res., 10, 274–277.
222
Mailhes,J.B. and Marchetti,F. (1994) The influence of postovulatory ageing on the retardation of mouse oocyte maturation and chromosome segregation induced by vinblastine. Mutagenesis, 9, 541–545. Mailhes,J.B., Marchetti,F. and Aardema,M. (1993) Griseofulvin-induced aneuploidy and meiotic delay in mouse oocytes: effect of dose and harvest time. Mutat. Res., 300, 155–163. Manca,A., Bassani,B., Russo,A., Pacchierotti,F. (1990) Origin of aneuploidy in relation to disturbances of cell-cycle progression. I. Effects of vinblastine on mouse bone marrow cells. Mutat. Res., 229, 29–36. Marchetti,F. and Mailhes,J.B. (1994) Variation of mouse oocyte sensitivity to griseofulvin-induced aneuploidy and meiotic delay during the first meiotic division. Environ. Mol. Mutagen., 23, 179–185. Marchetti,F. and Mailhes,J.B. (1995) Variation of mouse oocyte sensitivity to griseofulvin-induced aneuploidy during the second meiotic division. Mutagenesis, 10, 113–121. Marchetti,F., Mailhes,J.B., Bairnsfather,L., Nandy,I. and London,S.N. (1996) Dose–response study and threshold estimation of griseofulvin-induced aneuploidy during female meiosis I and II. Mutagenesis, 11, 195–200. Migliore,L., Cocchi,L. and Scarpato,R. (1996) Detection of the centromere in micronuclei by fluorescence in situ hybridization: its application to the human lymphocyte micronucleus assay after treatment with four suspected aneugens. Mutagenesis, 11, 285–290. Migliore,L., Zotti-Martelli,L. and Scarpato,R. (1999) Detection of chromosome loss and gain induced by griseofulvin, estramustine and vanadate in binucleated lymphocytes using FISH analysis. Environ. Mol. Mutagen., 34, 64–68. Pacchierotti,F., Bassani,B., Leopardi,P. and Zijno,A. (1991) Origin of aneuploidy in relation to disturbances of cell-cycle progression. II. Cytogenetic analysis of various parameters in mouse bone marrow cells after colchicine or hydroquinone treatment. Mutagenesis, 6, 307–311. Perry,P. and Wolff,S. (1974) New Giemsa method for the differential staining of sister chromatids. Nature, 251, 256–258. Raimondi,E., Scariolo,S., De Sario,A. and De Carli,L. (1989) Aneuploidy assay on interphase nuclei by means of in situ hybridization with DNA probes. Mutagenesis, 4, 165–169. Roobol,A., Gull,K. and Pogson,C.I. (1977) Evidence that griseofulvin binds to a microtubule associated protein. FEBS Lett., 75, 149–153. Rustia,M. and Shubik,P. (1978) Thyroid tumors in rats and hepatomas in mice after griseofulvin treatment. Br. J. Cancer, 38, 237–249. Shi,Q., Schmid,T.E. and Adler,I.-D. (1999) Griseofulvin-induced aneuploidy and meiotic delay in male mouse germ cells detected by using conventional cytogenetics and three-color FISH. Mutat. Res., 441, 181–190. Tiveron,C., Marchetti,F., Bassani,B. and Pacchierotti,F. (1992) Griseofulvininduced aneuploidy and meiotic delay in female mouse germ cells. Cytogenetic analysis of metaphase II oocytes. Mutat. Res., 266, 143–150. Vanhauwaert,A., Vanparys,P. and Kirsch-Volders,M. (2001) The in vivo gut micronucleus test detects clastogens and aneugens given by gavage. Mutagenesis, 16, 39–50. Received on September 19, 2001; accepted on January 10, 2002
