Chem.-Biol. Interactions, 79 (1991) 31--40 Elsevier Scientific Publishers Ireland Ltd.

A NEW

ACTION FOR TOPOISOMERASE

31

INHIBITORS*

ROBERT M. ZUCKER and KENNETH H. ELSTEIN

ManTech Environmental Technology, Inc., P.O. Box 12313, Research Triangle Park, NC 27709 (U.S.A.) (Received December 17th, 1990) (Revision received March 6th, 1991) (Accepted March 6th, 1991)

SUMMARY

Topoisomerases are known to aid DNA replication by breaking and resealing supercoiled DNA. Consequently, cells exposed to topoisomerase inhibitors before or during the S (DNA synthetic) phase of the cell cycle undergo abnormal DNA replication and become irreversibly blocked in the G2 (pre-mitosis) phase. We report that following a 4-h exposure to topoisomerase II inhibitors, murine erythroleukemic cells (MELC) do not form mitotic figures but exhibit a timedependent progression into G2 (4N DNA) and > G2 (up to 8N DNA) stages of the cell cycle. Following exposure to the topoisomerase I inhibitor camptothecin, recovering MELC also exhibit >G2 polyploidy, but to a considerably lesser degree: mitotic figures are present and a subpopulation of cells resumes cycling. However, both topo I and topo II inhibitors induce maximal percentages of > G2 cells when synchronized MELC are in the GJM phase at the time of exposure. This suggests that, in addition to their S-phase action, topoisomerase inhibitors can interfere with chromosome condensation during G2 and, in so doing, induce polyploidy.

Key words: Topoisomerase inhibitors -- Polyploidy -- Chromosome condensation --

Flow cytometry -- G2 block

Correspondence to: Robert M. Zucker, ManTech Environmental Technology, Inc., P.O. Box 12313, Research Triangle Pk, NC 27709, U.S.A. *The research described in this article has been funded wholly or in part by the Health Effects Research Laboratory, U.S. Environmental Protection Agency under contract to ManTech Environmental Technology, Inc. It has been reviewed and approved for publication as an E P A document. Approval does not signify that the contents necessarily reflect the views and policies of the Agency nor does mention of trade names or commercial products constitute endorsement or recommendation for use. 0009-2797/91/$03.50 © 1991 Elsevier Scientific Publishers Ireland Ltd. Printed and Published in Ireland

32 INTRODUCTION

The topoisomerases disentangle DNA during the S phase of the cell cycle by transiently breaking either one (topo I) or both (topo II) DNA strands [1,2]. A number of effective chemotherapeutic agents inhibit topoisomerase activity by stabilizing these otherwise temporary strand breaks [1,2] which, subsequently, can block cycling cells in the tetraploid G2 phase [3,4] and/or induce 'unbalanced growth' (i.e., unregulated RNA and protein synthesis) [5]. However, recent evidence that these agents also perturb premitotic and mitotic events, together with the observation that more topoisomerase is present during the GJM phase than during S [6,7], suggests that topoisomerase may not act solely in S phase. For example, studies in yeast indicate that topo II plays an essential role in the proper condensation and separation of chromosomes [8--10], while the addition of the topo II inhibitor VM-26 to BHK cells synchronized in G2 results in inhibition of chromosome condensation, histone phosphorylation, and nuclear envelope breakdown [11]. We report that during recovery from exposure to topo I or topo II inhibitors, murine erythroleukemic cells (MELC) become blocked not only in the G2 phase (4N), but in polyploid stages exhibiting up to 8N DNA. Moreover, no mitotic figures are present in cells exposed to topo II inhibitors. These observations suggest that, under certain exposure conditions, topoisomerase inhibitors not only perturb the S phase, but directly or indirectly inhibit chromosome condensation in G2 and allow continued (redundant) DNA synthesis. MATERIALS AND METHODS

Cell culture MELC (T3CL2, obtained from Dr. Clyde Hutchison, University of North Carolina, Chapel Hill, NC) were maintained in logarithmic growth in RPMI 1640 (No. 320-1875, GIBCO, Grand Island, NY), supplemented with 10% Fetal Bovine Serum (GIBCO No. 230-6140) and 25 mM HEPES (No. H3375, Sigma Chemical Co., St. Louis, MO) at 37°C in an atmosphere of 5% CO2 in air. Drug exposure MELC were exposed to the topo II inhibitor adriamycin (ADR: 0.5 #g/ml) or the topo I inhibitor camptothecin (CPT: 0.5 ~g/ml) for 4 h, then washed and recultured in drug-free medium for 72 h. Both ADR (Sigma No. D1515, St. Louis, MO) and CPT (No. 94600-J/33, National Cancer Institute Drug Development Branch, Bethesda, MD) were dissolved in DMSO (Sigma No. D5879). The doses used were established through preliminary studies. Synchronization To determine whether cell cycle stage at time of exposure affected the percentage of induced polyploid cells, MELC (2 × 105/ml) were exposed for 2 h to the topo II inhibitors VM-26 (0.25 ~g/ml, NCI No. 122819-I/12) or ADR (0.5 ~g/ml) or the topo I inhibitor CPT (1.0 ~g/ml) at 2-h intervals following release from syn-

33 chronization induced by a double-thymidine block [12]. By so doing, the majority of cells were in either the early S (T = 0 h), S (T = 2 h), G2/M (T -- 4 h), or M and G0/G~ (T = 6 h) phase(s) at the time of exposure. Cells then were washed of the drug and recultured overnight prior to analysis.

Flow cytometry Flow cytometric analyses were made with an Ortho 50H cytofluorograph (Becton-Dickinson, Westwood, MA) equipped with an analytical flow cell (30H) and a 100 mW, 488 nm-line argon-ion laser (Model 75, Lexel, Inc., Fremont, CA). Flow rate was maintained at 200 particles/s while collecting 10 a particles (cells/nuclei) per sample. Data were collected, stored, and manipulated via an Ortho 2150 computer system. The reported data are representative of a series of replicated experiments that were all internally consistent. Viability assay Viability was assessed by the carboxyfluorescein diacetate (CFDA)/propidum iodide (PI) assay [13] as the percentage of cells exhibiting green (CF) fluorescence (derived from the esterase-catalyzed hydrolysis of nonfluorescent CFDA), but not red (PI) fluorescence (resulting from the intercalation of PI into DNA). Bromodeoxyuridine incorporation The ability of exposed cells to synthesize DNA was assessed by bromodeoxyuridine (BrdU) incorporation. After a 4-h exposure to ADR or VM-26, recovering MELC were inoculated with BrdU 30 min prior to harvest. Cells were then washed, fixed in 70% ethanol, denatured with 4 N HC1, and stained indirectly with FITC-conjugated IgG according to the protocol of Dolbeare et al. [14]. Nuclear analysis Nuclei were isolated from PBS-washed ME LC by detergent-mediated cytolysis and stained with PI to quantify DNA content [15]. The relative percentage of nuclei in the G0/G1, S, and GJM phases of the cell cycle were estimated using Multicycle (Phoenix Flow Systems, San Diego, CA), a PC-based software package employing the mathematical model of Dean and Jett [16]. Those nuclei located in the region beyond the Gaussian distribution of G2 were considered to be > 4N polyploid (> G2). Growth and volume measurements Drug effects on growth and on nuclear volume were measured with a Coulter Counter (Model ZBI) and Channelyzer (Model 20; Coulter Electronics, Hialeah, eL). Cytology For examination of changes in morphology and mitotic index, cytological samples were prepared by a cytocentrifugation technique [15].

34 RESULTS

MELC exposed (4 h) to the topo II inhibitor ADR exhibit a time-dependent accumulation in the G2 phase of the (DNA) cell cycle and in stages beyond G2 containing up to 8N DNA ('> G2') (Fig. 1). Moreover, 85--90% of cells containing > 4N but < 8N DNA incorporate BrdU, indicating that this > G2 polyploidy is the result of continued DNA synthesis. However, no mitotic figures are present, cells do not proliferate, and by 72 h viability decreases to 77% in ADR-exposed cells and to 47% in VM-26-exposed cells (similar growth/cell cycle data not shown). Polyploidy induction also occurs in CPT-exposed MELC, but by 24 h post-washout, mitotic figures are present (1%, vs. 4% in control) and a subpopulation of exposed cells resumes cycling, subsequently diluting viability and cell cycle data (Fig. 1). lOO

lO

%---PI E x ~ l . ~ " ' - ~ . ~ .

ADR

80

8

60

/

(1)

4o

~J

20 /

4

%>G 2 ~ - - - - - - - - ~

-I.--

(

E

o; ~

O

6

%G2/

i

,

~

,

A

.

/

P. ~row._, {h

CPT

.%PI Ex c I . ' ~ ' - - - . o ~

L

80

(1) El--

60

o

6

%G2 •

40

4

20

2

~ 0

0

.~-----~=, 24

O ~_

.Growth,

.

48

0 72

Time Posf-WGsh (hr) Fig. 1. The effects of a 4-h exposure to 0.5 #g/ml CPT or 0.5 #g/ml ADR on MELC viability, proliferation (growth), and progression into G 2 and > G 2 phases over a 72-h recovery period. Both agents induce a time-dependent increase in the percentage of polyploid cells. By 24 h, a subpopulation of CPT-exposed cells resumes cycling; however, even by 72 h ADR-exposed ceils do not proliferate, but begin to lose viability, suggesting that the blocks are terminal. (Note that the slight decrease in the % G 2 of ADR-exposed cells at 72 h is due both to the increase in % > Gz and to pyknotic nuclei registering in < G 2 channels.) Over the same 72-h period, control cells remain 98% viable, exhibit an 83-fold increase in cell number, and maintain a normal cell cycle distribution (e.g. see Table I).

35 TABLE I EFFECT OF TOPOISOMERASE INHIBITORS ON THE CELL CYCLE AND MEAN NUCLEAR VOLUME OF ASYNCHRONOUS MELC FOLLOWING RECOVERY FROM A 2-h EXPOSURE Compared to either topo II inhibitor, the topo I inhibitor CPT induces a similar, but attenuated, perturbation of the MELC cell cycle. Moreover, the large percentage of CPT-exposed cells in the G0/GI and S phases partially offsets the increase in the mean nuclear volume of G2 and >G2 nuclei. Cell cycle analysis

Control 0.25 t~g/ml VM-26 0.5 #g/ml ADR 1.0 t~g/ml CPT

Nucl. vol.

%G0/G1

%S

%G2/M

% > G2

23.9 0.3 0.3 17.1

63.2 6.4 11.3 26.0

11.2 76.6 72.9 47.5

1.7 16.7 15.5 9.4

1.0 3.6 3.1 2.1

To assess whether the cell cycle stage at time of exposure affects the percentage of cells progressing beyond G2, asynchronous and thymidine-synchronized MELC were exposed to topo inhibitors for 2 h prior to washout and reculturing overnight in drug-free medium. Following recovery, - 9 0 % of asynchronous MELC exposed to either topo II inhibitor accumulate in G2 and > G2 phases and exhibit increased nuclear volume (Table I). Exposure of thymidine-synchronized MELC to the topo II inhibitors results in a phase-dependent increase in the relative percentage of recovering MELC containing >4N DNA. While exposure during G2/M traverse (T = 4 h) induces maximal >G2 polyploidy, exposure during S-phase traverse (T = 0, 2 h) predominantly induces G2-phase blocks (Fig. 2). CPT-exposed cells also exhibit phase specificity, with the maximal percentage of > G2 cells resulting from exposure during GJM traverse (T --- 4 h); however, in both synchronous and asynchronous populations, a considerably larger percentage of CPT-exposed cells progress into the G1 and S ('< G2') phases and are capable of cycling (see also Fig. 1). Moreover, this percentage is further increased in cells exposed during M- or G0/G,-phase traverse (T = 6 h), suggesting that these stages are more resistant to the toxic actions of CPT. DISCUSSION

MELC exposed to the topo II inhibitors VM-26 or ADR exhibit a timedependent increase in the percentage of cells progressing beyond the G2 phase of the cell cycle. By 72 h post-exposure, viability (measured by propidium iodide exclusion) decreases, mitotic figures are absent, and cells do not proliferate, suggesting that the induced polyploidy is terminal. MELC exposed to the topo I inhibitor CPT also exhibit increased percentages of polyploid cells, but by 24 h post-exposure, mitotic figures are present and a subpopulation of cells resumes cycling. In addition to their role in the S (DNA synthetic) phase, the topoisomerases

36 100

Profile at Exposure

ADR (0.5 t t g / m l )

80 60

v) O r EL . bE

o

i-°~.

40 20 o~

vM-26 (0.25 ~g/mO ~

z

CPT ( 1 . 0 . g / m l )

80' [

) G2

--J'~.~

60

G2

40 /

2O 0

0

2

XG2

4

_

6

~~"~ 2

~

~.

4

Time Post-Release (h.r) Fig. 2. Cell cycle data illustrating the phase-specific induction of > 4N polyploid cells by the topo I! inhibitors VM-26 and ADR and the topo I inhibitor CPT. 'Profile at Exposure' depicts the timedependent changes in the cell cycle distribution following release from thymidine-induced synchronization. Subsequent graphs depict the percentage of recovering cells blocked in G~ or >G~ polyploid phases as well as the sum of those in Gj and S phases ('< G~').

are known to act in G 2 and M. Studies using fission yeast demonstrate that topo II is required for normal condensation and separation of sister chromatids [8--10]. Inhibition of chromosome separation ('chromosome stickiness') also was observed in grasshopper neuroblasts exposed to relatively high concentrations of the topo II inhibitor m-AMSA during prophase [ 17]. That topo II also may play a role in chromosome condensation/separation in mammalian cells is supported by the observations that its levels are greatest in HeLa [6] and BALB/c 3T3 cells [18] during G2/M traverse, and that topo II inhibitors block Chinese hamster ovary cells at the G2 transition point preceding chromosome condensation [19] and prevent induction of chromosome condensation in rat liver and thymus nuclei [20]. We observe an occasional mitotic figure ( G2 polyploid cells is maximal when MELC are exposed to topo II inhibitors while traversing the G2/M phases is consistent with a perturbation of a topo II role in chromosome condensation [9,11,20] (Fig. 3). Exactly how MELC incapable of mitosis can undergo a second round of DNA synthesis is unclear. In cells capable of normal chromosome condensation/separation, the cell cycle is regulated by the cyclical synthesis and degradation of cyclin, a protein whose concentration increases during DNA synthesis, whose binding to p34 ¢dc2 kinase (to form M-phase promoting factor [MPF]) triggers mitosis, and whose subsequent degradation terminates mitosis [22--24] and allows the next cell cycle to begin. If cyclin concentration were to decrease in cells incapable of chromosome condensation/separation, it is conceivable that DNA synthesis could resume once cyclin concentration reached post-mitosis levels. The common link in this uncoupling of mitosis regulation from chromosome condensation/separation may be p34 cdc2 kinase, which not only triggers mitosis via MPF activation, but, by phosphorylating relevant proteins, may regulate chromosome condensation [25], nuclear envelope disassembly, and mitotic spindle construction [26]. That this may be the mechanism of topo II inhibitor action is supported by the observations that the topo II inhibitor VP-16 markedly

Microtubule Inhibitor

polyploidy

polyploidy

4n ~,~8n DNA -~ [cyclln]~?~ strand

oond.n.,io.,...,.,ion

Topo II

~, • • / / :o.,,n:,,

/ G2phase'",i

': :i........

\ ......

=)

Topos

,.-~ (uNAsynllleSlS! ,

I & II

~ 2rl i~4n UNA

DNA unwinding

~

~,,

\

/ ",

/ 'y

D=-

Fig. 3. The MELC cycle ( - 12 h). 'G 1' denotes the phase preceding DNA synthesis; 'S,' DNA synthesis; 'G,~,' the phase following DNA synthesis but preceding mitosis; an2 'M,' mitosis. Mitosis is believed to be triggered by the cyclical decrease in the concentration of the protein cyclin. Topoisomerase is known to act in S phase (to untangle DNA for replication) and may function in DNA repair [37]. Our data support the hypothesis that, in mammalian cells, it also acts in G~/M to aid in condensation/separation of replicated chromatin. The changes in cyclin concentration depict a possible mechanism to explain redundant DNA synthesis.

38 reduces p34 cdc2 kinase activity [25] and that topoisomerase II itself may require phosphorylation to be active in mitosis [7]. Indeed, Roberge et al. Ill} demonstrated in BHK cells that VM-26 blocked histone H1 and H3 phosphorylation, inhibited H1 kinase (which, apparently, has MPF-like growth-promoting capacity [27]), and not only prevented chromosomal condensation but decondensed previously condensed chromosomes. Thus, topo II inhibitors may induce > 4N polyploidy by depressing kinase activity and subsequently uncoupling essential mitotic events from the cyclical mechanisms that regulate them. This also is supported by the observation that DNA and RNA synthesis continue in yeast incapable of cytokinesis [28] and that yeast released from cdc2 blocks exhibit polyploidization [29]. Such a mechanism of polyploidy induction differs from that of the DNA inhibitors (e.g. aphidicolin, hydroxyurea), which reversibly block cells in S phase and induce a second round of DNA synthesis from which cells can spontaneously enter mitosis [30,31]; or that of the microtubule inhibitors (e.g. taxol, vinblastine), which induce polyploidy through a mechanism that blocks cells in mitosis [32]. However, we do observe a similar, but attenuated, response in CPTexposed cells, which suggests that either this mechanism is not specific to topo II inhibitors or that CPT can partially inhibit topo II activity. Indeed, using synchronized BHK cells, Roberge et al. [11] observed that VM-26 inhibited chromosome condensation if added during any stage of G,~, but CPT was effective only when applied during early G,~. This is consistent with our observation of increased percentages of cycling (< G,~) cells in MELC exposed to CPT during late G2/M or M and Go/G1 phases. In addition to the marked differences between the topo I and topo II inhibitors, we also observe some disparity between the two topo II inhibitors. For example, ADR induces a substantial percentage of > G2 cells when exposure occurs during S-phase traverse (Fig. 2). Moreover, at a lower dose (0.05 ~g/ml), ADR induces more > G2 cells in MELC exposed during S-phase traverse (% > G~ = 42) than in MELC exposed during G2/M (% > G2 = 22). In contrast, VM-26 exhibits G2/M-phase specificity at high concentrations (0.25--1.0 ~g/ml) and recovery at lower ones. However, these discrepancies are not inconsistent with the hypothesis of a G2/M role for topo II. Unlike VM-26, which does not bind to DNA [33], ADR intercalates DNA and persists at reduced concentrations following washout [34]. As such, the S phase, during which the DNA is least condensed, may be more susceptible to damage than the G2/M phase, particularly when exposed to an intercalating agent. Indeed, ADR-induced strand breaks are more persistent [35] and, consequently, are as much as 30-fold more toxic than those induced by other topo II inhibitors [36]. One can, therefore, speculate that at the lower dose, S-phase damage and/or the presence of the intercalated agent may prevent chromosome condensation/separation, yet allow redundant DNA synthesis; while at the high dose, the S-phase damage may be so severe that the majority of cells do not progress to the GJM-phase transition point preceding induction of > 4N polyploidy. However, exposure during G2/M traverse (when the DNA is less open and,

39

presumably, less susceptible) induces polyploidy only at the higher ADR concentration. Certainly, other explanations may exist. In summary, our findings suggest that inhibition of topoisomerase activity results in the induction of > G2 polyploid blocks, possibly as a consequence of the uncoupling of chromosomal condensation/separation processes from mitotic regulatory mechanisms. The relative percentages of cells blocked in >4N polyploid stages is a function of the type and concentration of topo inhibitor and the cell cycle stage at time of exposure, with terminal G2 blocks predominant in cells exposed during S phase, and > G2 polyploidy resulting from those exposed during GJM traverse. ACKNOWLEDGEMENT

We thank Dr. Robert Kavlock and Dr. John Rogers of the Developmental Toxicology Division of HERL/EPA and Dr. Richard Scearce of ManTech Environmental Technology, Inc. for their support throughout the course of this study and Dr. Edward Massaro for his helpful comments. REFERENCES 1 2 3 4

5 6

7

8

9 10 ll

12

13

R. D'Arpa and L.F. Liu, Topoisomerase-targeting antitumor drugs, Biochim. Biophys. Acta, 989 (1989) 163--177. L.F. Liu, DNA topoisomerases -- enzymes that catalyze the breaking and rejoining of DNA, CRC Crit. Rev. Biochem., 15 (1983) 1--24. R.A. Tobey, Different drugs arrest cells at a number of distinct stages in G2, Nature, 254 (1975) 245--247. P.N. Rao, G2 arrest induced by anticancer drugs, in: H. Busch, S.T. Crooke and Y. Daskal (Eds.), Effects of Drugs on the Cell Nucleus, Academic Press, Inc., New York, 1979, pp. 475--490. D.W. Ross, Unbalanced cell growth and increased protein synthesis induced by chemotherapeutic agents, Blood Cells, 9 (1983) 57--68. E. Estey, R.C. Adlakha, W.N. Hittelman and L.A. Zwelling, Cell cycle stage-dependent variations in drug-induced topoisomerase II mediated DNA cleavage and cytotoxicity, Biochemistry, 26 (1987) 4338--4344. M.M,S. Heck, W.N. Hittelman and W.C. Earnshaw, Communication: In vivo phosphorylation of the 170-kDa form of eukaryotic DNA topoisomerase II, J. Biol. Chem., 264 {1989) 15161--15164. T. Uemura, H. Ohkura, Y. Adachi, K. Morino, K. Shiozaki and M. Yanagida, DNA topoisomerase II is required for condensation and separation of mitotic chromosomes in S. pombe, Cell, 50 (1987) 917--925. C. Holm, T. Goto, J.C. Wang and D. Botstein, DNA topoisomerase II is required at the time of mitosis in yeast, Cell, 41 (1985) 553--563. C. Holm, T. Stearns and D. Botstein, DNA topoisomerase II must act at mitosis to prevent chromosome nondisjunction and chromosome breakage, Mol. Ceil Biol., 9 (1989) 159--168. M. Roberge, J. Th'ng, J. Hamaguchi and E.M. Bradbury, The topoisomerase II inhibitor VM-26 induces marked changes in histone H1 kinase activity, histones H1 and H3 phosphorylation, and chromosome condensation in G2 phase and mitotic BHK cells, J. Cell Biol., 111 (1990} 1753--1762. R.M. Zucker, N.C. Wu, A. Krishan and M. Silverman, Synchronization of murine erythroleukemic cells: Nuclear volume measurements for monitoring cell cycle traverse, Exp. Cell Res., 123 (1979) 383--387. R.M. Zucker, K.H. Elstein, R.E. Easterling and E.J. Massaro, Flow cytometric analysis of the cellular toxicity of tributyltin, Toxicol. Lett., 43 (1988) 201--218.

40 14

15 16 17

18 19

20

21 22 23 24

25 26 27 28 29 30 31 32 33

34

35 36

37

F. Dolbeare, H.G. Gratzner, M.G. Pallavicini and J.W. Gray, Flow cytometric measurement of total DNA content and incorporated bromodeoxyuridine, Proc. Natl. Acad. Sci. U.S.A., 80 (1983) 5573--5577. R.M. Zucker, K.H. Elstein, R.E. Easterling and E.J. Massaro, Flow cytometric analysis of the mechanism of methylmercury cytotoxicity, Am. J. Pathol., 137 (1990) 1187--1198. P.N. Dean and J.H. Jett, Mathematical analysis of DNA distributions derived from flow cytometry, J. Cell Biol., 60 (1974) 523--527. M.E. Gaulden, Hypothesis: some mutagens directly alter specific chromosomal proteins (DNA topoisomerase II and peripheral proteins) to produce chromosome stickiness, which causes chromosome aberrations, Mutagenesis, 2 (1987) 357--365. K-C. Chow and W.E. Ross, Topoisomerase-specific drug sensitivity in relation to cell cycle progression, Mol. Cell. Biol., 7 (1987) 3119--3123. R. Rowley and K. Kort, Novobiocin, nalidi×ic acid, etoposide, and 4'-(9-acridinylamino) methanesulfon-m-anisidide effects on G2 and mitotic chinese hamster ovary cell progression, Cancer Res., 49 (1989) 4752--4757. J. Newport and T. Spann, Disassembly of the nucleus in the mitotic extracts: membrane vesicularization, lamin disassembly, and chromosome condensation are independent processes, Cell, 48 (1987) 219--230. K.W. Lanks and J.M. Lehman, DNA synthesis by L929 cells following doxorubriein exposure, Cancer Res., 50 (1990) 4776--4778. A.W. Murray and M.W. Kirschner, Cyclin synthesis drives the early embryonic cell cycle, Nature, 339 (1989) 275--280. A.W. Murray and M.W. Kirschner, Dominoes and clocks: The union of two views of the cell cycle, Science, 246 (1989) 614--621. G. Draetta, F. Luca, J. Westendorf, L. Brizuela, J. Ruderman and D. Beach, cdc2 protein kinase is complexed with both cyclin A and B: Evidence for proteolytic inactivation of MPF, Cell, 56 (1989) 829--838. S. Moreno and P. Nurse, Substrates for p34Cd('2: In vivo veritas? Cell, 61 (1990) 549--551. R.B. Lock and W.E. Ross, Inhibition of p34 ¢dc2 kinase activity by etoposide or irradiation as a mechanism of G2 arrest in Chinese hamster ovary cells, Cancer Res., 50 (1990) 3761--3766. R.J. Inglis, T.A. Langan, H.R. Matthews, D.G. Hardie and E.M. Bradbury, Advance of mitosis by histone phosphokinase, Exp. Cell Res., 97 (1976) 418--425. T. Uemura and M. Yanagida, Mitotic spindle pulls but fails to separate chromosomes in type II DNA topoisomerase mutants: uncoordinated mitosis, EMBO J., 5 (1986) 1003--1010. F. Cross, J. Roberts and H. Weintraub, Simple and complex cell cycles, Annu. Rev. Cell Biol., (1989) 341--395. A.B. Hill and R.T. Schimke, Increased gene amplification in L5178Y mouse lymphoma cells with hydroxyurea-induced chromosomal aberrations, Cancer Res., 45 (1985) 5050-5057. R.T. Schimke, Minireview: Gene amplification in cultured cells, J. Biol, Chem., 263 (1988) 5989--5992. J.R. Roberts, D.C. Allison, R.C. Donehower, and E.K. Rowinsky, Development of polyploidization in taxol-resistant human leukemia cells in vitro, Cancer Res., 50 (1990) 710--716. J. Markovits, Y. Pommier, D. Kerrigan, J.M. Covey, E.J. Tilchen and K.W. Kohn, Topoisomerase II-mediated DNA breaks and cytotoxicity in relation to cell proliferation and the cell cycle in NIH3T3 fibroblasts and L1210 Leukemia cells, Cancer Res., 47 (1987) 2050--2055. M.C. Willingham, M.M. Cornwell, C.O. Cardarelli, M.M. Gottesman and I. Pastan, Single cell analysis of daunomycin uptake and efflux in multidrug-resistant and -sensitive KB cells: Effects of verapamil and other drugs, Cancer Res., 46 (1986) 5941--5946. W.E. Ross, Commentary: DNA topoisomerases as targets for cancer therapy, Biochem. Pharmacol., 34 (1985) 4191--4195. R.J. Epstein and P.J. Smith, Estrogen-induced potentiation of DNA damage and cytotoxicity in human breast cancer cells treated with topoisomerase II-interactive antitumor drugs, Cancer Res., 48 (1988) 297--303. C.S. Downes and R.T. Johnson, DNA topoisomerases and DNA repair, Bioessays, 8 (1988) 179--184.