Biological effects of a bifunctional DNA cross-linker.

Mutation Research 426 Ž1999. 89–94

Biological effects of a bifunctional DNA cross-linker. II. Generation of micronuclei and attached micronuclear-like structures Kyle Kurek a , Lloyd Matsumoto b,) , Gary Gustafson c , Richard Pires d , Umadevi Tantravahi a,e, J. William Suggs d a

b

DiÕision of Biology and Medicine, Brown UniÕersity, ProÕidence, RI 02912, USA Department of Biology, Rhode Island College, 600 Mt. Pleasant AÕenue, ProÕidence, RI 02908-1991, USA c ScriptGen, Boston, MA 02155, USA d Department of Chemistry, Brown UniÕersity, ProÕidence, RI 02912, USA e Women and Infants Hospital, ProÕidence, RI 02905, USA Received 13 June 1996; received in revised form 6 January 1998; accepted 6 January 1999

Abstract Madin–Darby bovine kidney ŽMDBK. cells were treated with the bifunctional DNA cross-linker, L-7, to examine the generation of micronuclei and other nuclear abnormalities. The preceding paper demonstrates that L-7 treatment induces the formation of triradial and quadriradial chromosomes in MDBK cells. These chromosomes are believed to result from interduplex DNA cross-links formed between G–C rich centromeric satellite DNA regions on non-sister chromatids. Treatment produces a majority of centromere-positive micronuclei. In addition, many daughter cells remain attached by chromatin bridges which are sometimes beaded with micronuclei. Up to 15% of cell nuclei become lobular and fused with numerous micronuclear-like structures attached to their membranes. These attached structures are classified as attached micronuclear-like structures ŽAMNLS.. Fluorescence in situ hybridization ŽFISH. using a centromeric satellite sequence was performed on treated cells. Hybridization reveals that intercellular bridges are composed of centromeric sequences and initiate at centromeric foci in daughter cells. Furthermore, the majority of junctions between AMNLS and nuclei contain an enhancement of centromeric signal. The frequency of AMNLS appears dependent on the concentration of L-7 and the duration of treatment. Similar results were found for the generation of cross-linked chromosome products in the previous paper. We suggest that AMNLS result from the abnormal mitotic segregation of cross-linked chromosome products. q 1999 Elsevier Science B.V. All rights reserved. Keywords: DNA cross-linker; Madin-Darby bovine kidney ŽMDBK. cell; Centromeric heterochromatin; Micronucleus; Fluorescence in situ hybridization ŽFISH.

)

Corresponding author. Tel.: q1-401-456-9539; Fax: q1-401-456-8379

0027-5107r99r$ - see front matter q 1999 Elsevier Science B.V. All rights reserved. PII: S 0 0 2 7 - 5 1 0 7 Ž 9 9 . 0 0 0 4 1 - X

90

K. Kurek et al.r Mutation Research 426 (1999) 89–94

1. Introduction Assays for abnormalities in nuclear morphology provide rapid and cost-effective methods for evaluating genetic damage. The micronucleus assay is the most common w1,2x. Micronuclei are cytoplasmic nuclear remnants formed from the lagging of chromosomes or chromosome fragments during anaphase w3x. Frequencies of micronuclei are 1.2% in normal human lymphocytes w4x, but significant increases are observed in cells treated with genotoxic compounds w5–8x and in cells from patients with chromosomal instability syndromes w9x. Fluorescence in situ hybridization ŽFISH. using combinations of centromeric probes w10x, chromosome paints w11x and anti-kinetochore antibodies w12x, provides a system to classify micronuclei based on their contents. Positive micronuclei result from the lagging of whole chromosomes or centromeric fragments, and are common with aneugens. Negative micronuclei result from the lagging of acentric fragments, and are common with clastogens w13x. Morphological analysis is less trivial when more complex forms of genetic damage are present. These include chromosome or chromatid exchanges, multicentric and multiradial chromosomes, alterations in chromosome structure and condensation, and excision-repairable lesions. In addition to micronuclei w10x, these lesions produce binucleate cells, intercellular bridges and nuclear abnormalities. In particular, non-apoptotic nuclear lobulations and attachments are observed w14x. Intercellular chromatin bridges are common with genotoxic agents that alter centromeric structure w15–17x, in irradiated cells w18x and in cells from Roberts syndrome, which contain a hereditary abnormality in centromeric heterochromatin structure w19x. Bridges result from the segregation of attached and tangled chromosomes. Following segregation, bridges may break, persist or force the subsequent fusion of the daughter cells w15,18x. Nuclear lobulations and attachments are rarer than micronuclei in normal cells w20x, but can be found in up to 20% of cells treated with anti-neoplastic agents w15,21x and in Roberts syndrome cells. These nuclear abnormalities are also suggested to result from problems segregating tangled and attached chromosomes, however, more studies are needed to explore this hypothesis.

We have examined the ability of the bifunctional DNA cross-linker, L-7, to induce the formation of micronuclei and other nuclear abnormalities in Madin–Darby bovine kidney ŽMDBK. cells. This cross-linker, while specific for DNA, is unreactive to RNA in any form, single-stranded DNA or protein w22,23x. L-7 treatment produces cross-linked chromosome products, including triradials and quadriradials, which are joined by centromeric heterochromatin Žsee accompanying paper.. In addition to micronuclei, L-7 treatment results in the formation of intercellular chromatin bridges, binucleate cells and lobulated nuclei with attached micronuclear-like structures. The points of attachment of these bridges and membranes are enriched with centromeric heterochromatin. Both chromosome and nuclear aberrations are produced with a dose and time dependence during L-7 treatments. We conclude that the intercellular bridges and attached micronuclear-like structures result from the inseparable cross-linking of the centromeric heterochromatin.

2. Materials and methods Madin–Darby Bovine kidney cells ŽMDBK. w24x were cultured in 1 = reinforced Eagle’s medium ŽREM. supplemented with 5% fetal calf serum and 5% calf serum. Cells were grown on sterile, acidwashed coverslips and treated with 0, 1, 2 or 3 mgrml L-7 for 12 to 72 h. Following treatments, coverslips were washed 3 = in Earle’s Balanced Salt Solution ŽBSS. and either mounted with BSS for photography or prepared for hybridization by treatment with a permeabilizing buffer Ž0.5 M Tris–HCl, 1% NP-40, 0.22 M sucrose, 10 mM MgCl 2 , 0.5 mM CaCl 2 . for 1 min at 228C. Coverslips were fixed using three 5 min baths of 3:1 methanol–acetic acid followed by 1 min in methanol and were then allowed to air dry. Fluorescence in situ hybridization procedures and conditions are described in the accompanying paper. Phase-contrast images of cells were obtained at 400 = magnification using TMAX 400 film with exposures of 1–2 s. Fluorescence images were captured

K. Kurek et al.r Mutation Research 426 (1999) 89–94

at 600 = magnification using Ektachrome 400 film with exposures of 15–20 s. 3. Results L-7 treatment of MDBK cells induces the morphological alterations displayed in Fig. 1. Intercellular bridges are produced which unite heterochromatin domains in daughter cells ŽFig. 1b.. Micronuclei are also common within treated cells ŽFig. 1c,d.. In addition, some nuclei become lobular with micronuclear-like structures adjacent and attached to their membranes ŽFig. 1c,d.. We classify these structures as attached micronuclear-like structures or AMNLS. Fluorescence in situ hybridization ŽFISH. with a centromeric sequence was used to probe treated cells ŽFig. 2.. Quantitation of signal revealed that the majority of L-7 induced micronuclei and AMNLS contain centromeric fragments ŽFig. 2c.. Hybridization also revealed that intercellular bridges are com-

91

posed of chromatin enriched with centromeric sequence. Furthermore, these bridges often originate within foci of centromeric signal in the daughter nuclei ŽFig. 2d.. Binucleate cells appear more frequently after treatment, and the junctions between their nuclei contain an enhancement of hybridization signal ŽFig. 2b.. The junctions between AMNLS and nuclei also possess foci of centromeric signal. Table 1 examines the frequency of AMNLS at different concentrations of L-7 and after different times of cell exposure. After 24 h of treatment, the frequency of AMNLS for 3 mgrml L-7 is twice the value for 1 and 2 mgrml L-7. After 48 h of treatment, no viable cells remain at 3 mgrml L-7, but the frequencies of AMNLS for 1 and 2 mgrml L-7 double and plateau to 72 h. These results mirror the cytotoxic effects of 3 mgrml L-7 and cytostatic effects of 1–2 mgrml L-7 found in growth curves generated during treatments Ždata not shown.. Furthermore, the results indicate both a concentration and time dependence for the frequency of AMNLS.

Fig. 1. Phase-contrast images of MDBK cell morphology following treatment with 1 mgrml L-7 for 48 and 72 h. Ža. Untreated cells. Žb. Daughter nuclei are attached by a bridge spanning heterochromatic regions Žarrows.. Žc and d. Cells containing attached micronuclear-like structures Žlarge arrows.. Micronuclei also appear to be present within cells. In Žc., an attached micronuclear-like structure appears to communicate via a stalk to the parent nucleus, while in Žd. clear boundaries of separation exist between attachments Žsmall arrows..

92

K. Kurek et al.r Mutation Research 426 (1999) 89–94

Fig. 2. Fluorescence in situ hybridization ŽFISH. using the satellite I DNA probe in MDBK cells following treatment with 1 mgrml and 2 mgrml L-7 for 24 h. Ža. Untreated cells. Žb. Binucleate cell with enhancement of hybridization signal at the nuclear junctions Žarrow.. Žc. Binucleate cell containing banded hybridization signal and 5 MN containing from 0–4 signals. Žd. Abnormal mitosis producing bridged daughter nuclei, each containing AMNLS. The points of attachment and bridge contain centromeric signal Žarrows..

A comparison with the frequency of metaphase plates containing cross-linked products ŽTable 1. reveals a similar dependence. These frequencies suggest a re-

lationship between the elaboration of rare multiradial chromosomes and the formation of AMNLS during L-7 treatment of cells.

Table 1 Frequency of attached micronuclear-like structures ŽAMNLS. in MDBK cells after different incubation times and at different concentrations of L-7 Concentration of L-7

1 mgrml L-7

Incubation time

24 h

48 h

72 h

24 h

48 h

72 ha

24 ha

Number of cells ŽNormal. Number of cells ŽAMNLS. Percentage of cells with AMNLS Percentage of metaphase plates with cross-linked productsb

468 32 6.4 8.8

422 78 15.6 16.6

422 78 15.6 16

467 33 6.6 17.7

430 70 14 23.3

432 68 13.6 16.6

426 74 14.8 30.5

a b

2 mgrml L-7

Most cells are dead after 72 h at 2 mgrml and all are dead by 48 h at 3 mgrml L-7. See Table 1 in accompanying paper.

3 mgrml L-7

K. Kurek et al.r Mutation Research 426 (1999) 89–94

4. Discussion Treatment of MDBK cells with L-7 produces micronuclei, intercellular chromatin bridges, and attached micronuclear-like structures ŽAMNLS.. The majority of micronuclei contain centromeric signal, which implies that L-7 is predominantly an aneugen. This is supported by the metaphase analysis in the accompanying paper, where L-7 was found to induce cross-linked chromosome products. The products, which include quadriradial and triradial chromosomes, are formed by exchange of double-stranded DNA between non-sister chromatids w25x. These recombined chromatids can have two or no centromeres and kinetochores, and will segregate abnormally at mitosis w26,27x. Cells from Bloom’s syndrome w28x and Fanconi’s anemia w29x contain such recombined chromosomes and frequently display anaphase lag and micronuclei w9x. Treatment with L-7 results in non-reciprocal exchanges between the centromeres of non-sister chromatids, producing complex multiradial chromosomes with abnormal centromeres Žsee Figs. 3 and 4 in accompanying paper.. The presence of centromere-positive micronuclei likely reflects the inability to segregate these products during mitosis. The intercellular chromatin bridges possess centromeric heterochromatin along the fibers and at anchoring sites. Similar bridges have been observed with Hoechst 33258 w21x, berenil w17x and 5-azacytidine w15x, which result from the decondensation and subsequent tangling of the centromeric heterochromatin. These compounds might prevent resolution of the tangles by inhibiting topoisomerase II function w17x. The L-7 induced cross-linked products have abnormal centromeres with significant decondensation. The failure to resolve these structures would result in unraveling at points of attachment during anaphase to generate chromatin bridges. The AMNLS also reflect abnormal mitotic events. Nuclear lobulations and blebs have been reported in cells treated with Hoechst 33258 w21x, 5-azacytidine w15x and from cells of Roberts syndrome w19x. An electron microscopic analysis of mitotic Roberts cells revealed two classes of lagging chromosomes, those which initiate anaphase but remain behind the rest of the migrating chromosomes and those which fail to initiate and remain at the equatorial plane. The latter

93

class form micronuclei while the former produce nuclear blebs and lobulations w19x. The AMNLS may represent membranes formed around multiradial chromosomes that are partially segregated due to an imbalance of opposing kinetochore forces. Alternatively, cross-linked products may completely inhibit anaphase segregation to produce a lobular, endomitotic nucleus with numerous AMNLS. Although the exact mechanisms cannot be determined from these experiments, the hybridizations confirm the involvement of the centromeric sequences at the points of attachment of AMNLS and binucleate daughter nuclei. Furthermore, the frequencies of both AMNLS and cross-linked chromosome products display similar concentration and time dependence. We conclude by suggesting that the chromatin bridges and AMNLS result from the inability to segregate cross-linked chromosome products generated following L-7 cross-linking of centromeric hetereochromatin of chromatids.

References w1x K. Boller, W. Schmid, Chemical mutagenesis in mammals. The Chinese hamster bone marrow as an in vivo test system. Hematological findings after treatment with trenimon, Humangenetik 11 Ž1970. 35–54. w2x J.A. Heddle, A rapid in vivo test for chromosomal damage, Mutat. Res. 18 Ž1973. 187–190. w3x D. Schiffman, U. De Boni, Dislocation of chromatin elements in prophase induced by diethylstilbestrol: a novel mechanism by which micronuclei can arise, Mutat. Res. 246 Ž1991. 113–122. w4x M. Fenech, S. Neville, Conversion of excision-repairable DNA lesions to micronuclei within one cell cycle in human lymphocytes, Environ. Mol. Mutagen. 19 Ž1992. 27–36. w5x A. Slavotinek, P.E. Perry, A.T. Sumner, Micronuclei in neonatal lymphocytes treated with the topoisomerase II inhibitors amsacrine and etoposide, Mutat. Res. 319 Ž1993. 215–222. w6x B.R. Brinkley, A. Tousson, M.M. Valdivia, The kinetochore of mammalian chromosomes: structure and function in normal mitosis and aneuploidy, in: V.L. Dellarco, P.E. Voytek, A. Hollaender ŽEds.., Aneuploidy: Etiology and Mechanisms, Vol. 36, Plenum, New York, 1985, pp. 243–267. w7x M. Hayashi, T. Sofuni, M. Ishidate Jr., Kinetics of micronucleus formation in relation to chromosomal aberrations in mouse bone marrow, Mutat. Res. 127 Ž1984. 129–137. w8x N.L. Rudd, D.I. Hoar, C.L. Greentree, L.S. Dimnik, U.G.G. Henning, The micronucleus assay in human fibroblasts: a

94

w9x

w10x

w11x

w12x

w13x

w14x

w15x

w16x

w17x

w18x

K. Kurek et al.r Mutation Research 426 (1999) 89–94 measure of spontaneous chromosomal instability and mutagen hypersensivity, Environ. Mol. Mutagen. 12 Ž1988. 3–13. M.P. Rosin, J. German, Evidence for chromosome instability in vivo in Bloom syndrome: increased numbers of micronuclei in exfoliated cells, Hum. Genet. 71 Ž1985. 187–191. N. Titenko-Holland, L.E. Moore, M.T. Smith, Measurement and characterization of micronuclei in exfoliated human cells by fluorescence in situ hybridization with a centromeric probe, Mutat. Res. 312 Ž1994. 39–50. M. Guttenbach, M. Schmid, Exclusion of specific human chromosomes into micronuclei by 5-azacytidine treatment of lymphocyte cultures, Exp. Cell Res. 211 Ž1994. 127–132. K.L. Sternes, B.K. Vig, Micronuclei, kinetochores and hypoploidy: tests with some agents, Mutagenesis 4 Ž1989. 425–431. D.A. Eastmond, J.D. Tucker, Identification of aneuploidy-inducing agents using cytokinesis-blocked human lymphocytes and an antikinetochore antibody, Environ. Mol. Mutagen. 13 Ž1989. 34–43. P.E. Swanson, S.B. Carroll, X.F. Zhang, M.A. Mackey, Spontaneous premature chromosome condensation, micronucleus formation, and non-apoptotic cell death in heated HeLa S3 cells: ultrastructural observations, Am. J. Pathol. 146 Ž1995. 963–971. S. Davidson, P. Crowther, J. Radley, D. Woodcock, CytotoxX icity of 5-aza-2 -deoxycytidine in a mammalian cell system, Eur. J. Cancer 28 Ž1992. 362–368. V. Rodilla, J.A. Pellicer, A. Serrano, J. Petrusa, Possible relationship between micronucleated and binucleated cells induced by cisplatin in cultured CHO cells, Mutat. Res. 291 Ž1993. 35–41. M. Poot, J. Koehler, P.S. Rabinovitch, H. Hoehn, J.H. Priest, Cell kinetic disturbances induced by treatment of human diploid fibroblasts with 5-azacytidine indicate a major role for DNA methylation in the regulation of the chromosome cycle, Hum. Genet. 84 Ž1990. 258–262. H. Sasaki, H. Hayashi, A nuclear thread bridging sister cells

w19x

w20x

w21x

w22x

w23x

w24x

w25x

w26x

w27x w28x w29x

prior to radiation-induced cell fusion, Radiat. Res. 77 Ž1979. 577–585. E.W. Jabs, C.M. Tuch-Muller, R. Casano, J.B. Rattner, Studies of mitotic and centromeric abnormalities in Roberts syndrome: implications for a defect in the mitotic mechanism, Chromosoma 100 Ž1991. 251–261. P.E. Tolbert, C.M. Shy, J.W. Allen, Micronuclei and other nuclear anomalies in buccal smears: methods and development, Mutat. Res. 217 Ž1992. 69–77. B.K. Vig, S.E. Swearngin, Sequence of centromere separation: kinetochore formation in induced laggards and micronuclei, Mutagenesis 1 Ž1986. 461–465. J.D. Farmer Jr., G.R. Gustafson, A. Conti, M.B. Zimmt, J.W. Suggs, DNA binding properties of a new class of linked anthramycin analogs, Nucleic Acids Res. 19 Ž1991. 899–903. J.D. Farmer Jr., S.M. Rudnicki, J.W. Suggs, Synthesis and DNA crosslinking ability of a dimeric anthramycin analog, Tetrahedron Lett. 29 Ž1988. 5101–5108. S.H. Madin, N.B. Darby, Established kidney cell lines of normal adult bovine and ovine origin, Proc. Soc. Exp. Biol. Med. 98 Ž1958. 574–576. M. Charron, R. Hancock, Chromosome recombination and defective genome segregation induced in Chinese hamster cells by the topoisomerase II inhibitor VM-26, Chromosoma 100 Ž1991. 97–102. R.S.K. Chaganti, S. Schonberg, J. German, A manyfold increase in sister chromatid exchange in Bloom’s syndrome lymphocytes, Proc. Natl. Acad. Sci. U.S.A. 71 Ž1974. 4508– 4512. E. Therman, E.M. Kuhn, Mitotic crossing-over and segregation in man, Hum. Genet. 59 Ž1981. 93–100. E.M. Kuhn, E. Therman, Cytogenetics of Bloom’s syndrome, Cancer Genet. Cytogenet. 22 Ž1986. 1–18. T.M. Schroeder, J. German, Bloom’s syndrome and Fanconi’s anemia: demonstration of two distinctive patterns of chromosome disruption and rearrangement, Humangenetik 25 Ž1974. 299–306.