J. Mol. Biol.

(1991)

222,

909-924

Distribution of Topoisomerase II Cleavage Sites in Simian Virus 40 DNA and the Effects of Drugs Yves Pommiert, Laboratory

Giovanni

Capranicot,

Ann Orr and Kurt W. Kohn

of Molecular Pharmacology, Developmental Therapeutics Division of Cancer Treatment, National Cancer Institute National Institutes of Health, Bethesda, MD, U.S.A. (Received

3 June

1991; accepted 20 August

Program

1991)

The distributions of DNA cleavage sites induced by topoisomerase II in the presence or absence of specific drugs were mapped in the simian virus 40 genome. The drugs studied were 5-iminodaunorubicin, amsacrine (m-AMSA), teniposide (VM-26) and 2-methyl-ghydroxyellipticinium; each produced a distinctive pattern of enhanced cleavage. Consistently intense cleavage, both in the presence and in the absence of drugs, occurred in the nuclear matrix-associated region. Since topoisomerase II is a major constituent of the nuclear matrix, and cleavage complexes include a covalent link between topoisomerase II and DNA, the findings suggest that topoisomerase II may function to attach DNA t’o the nuclear matrix. Cleavage usually occurred on both DNA strands with the expected four base-pair 5’ stagger, and strong sites tended to occur within A/T runs such as have been associated with binding to the nuclear scaffold. Intense cleavage was present also in the replication termination region, but was absent from the vicinity of the replication origin. Cleavage intensities were found to change with time in a manner that depended both on the site and on the drug, suggesting that topoisomerase II can move along the DNA from a kinetically preferred site to a thermodynamically preferred site. Keywords:

topoisomerase II, simian virus 40, replication, topoisomerase inhibitors

1. Introduction

(m-AMSAt) and teniposide (VM-26) have been used to map enzyme binding sites in a variety of DNA fragments and cellular genes (Darby et al., 1986; Riou et al., 1986; Rowe et al., 1986; Udvardy et al., 1986; Yang et aE., 1985; Zwelling et al., 1988). Topoisomerase II is preferentially associated with the nuclear matrix of interphase cells and with the nuclear scaffold of metaphase chromosomes (Berrios et al., 1985; Earnshaw et aZ., 1985; Gasser & Laemmli, 1986). The latter is consistent with the essential role of the enzyme in chromosome condensation (Newport & Spann, 1987; Roberge et aZ., 1990; Wright & Schatten, 1990) and separation of daughter DNA molecules at the end of DNA replication (intermolecular DNA strand passage: DiNardo et al., 1984; Holm et al., 1985; Uemura et al., 1987).

DNA topoisomerase II is a nuclear enzyme that introduces a cut in a DNA duplex, catalyzes the passage of an intact DNA duplex through the gap and then reseals the break. The 5’ termini of DNA double-strand breaks are covalently linked to a tyrosine residue of the enzyme and are staggered by four base-pairs (Liu, 1989; Wang, 1985, 1987). Topoisomerase II-mediated DNA breaks can be detected after protein denaturation with sodium dodecylsulfate (SDS) and/or alkali, and can be enhanced by a variety of anticancer drugs (Liu, 1989; Pommier & Kohn, 1989). Consequently, topoisomerase IT inhibitors such as amsacrine

t Author to whom reprint requests should be addressed at Building 37, Room 5C27, National Institutes of Health, Bethesda, MD 20892, U.S.A. $ Present address: Istituto Nationale per lo Studio la Cura dei Tumori, via G. Venezian 1, 20133 Milan. Italy.

enhancers,

t Abbreviations used: m-AMSA, 4’(9acridinylamino)methanesulfon-m-anisidide; (VM-26), 4’-demethylepipodophyllotoxin-9-(4,6,-Othionylidine-fl-n-glucopyranoside; MAR, associated region; bp, base-pair(s).

et

teniposide nuclear

matrix-

909 ~w)22~2836/91/24090916

$03.00/O

0

1991 Academic

Press Limited

910

Y. Pommier et al.

Topoisomerase II, along with topoisomerase I, is also a DNA relaxing enzyme (intramolecular DNA strand passage), which may remove DNA supercoiling and torsional tension as they arise during transcription and DNA replication (Brill & Sternglanz, 1988; Brill et al., 1987; Kim & Wang, 1989; Wu et al., 1988). Simian virus 40 (SV40) uses host DNA topoisomerases for replication (Avemann et al., 1988: Hsiang et al., 1989; Shin & Snapka, 1990; Snapka, 1986; Wold et al., 1989; Yang et al., 1987) and contains a nuclear matrix-associated region (MAR) located in the early transcribed region of the virus (between positions 4100 and 4300: Prives et al., 1986). We have shown that the SV40 MAR is preferentially cleaved by mouse topoisomerase II in the absence of drug or in the presence of doxorubicin derivatives (Capranico et al., 199Ob;Pommier et al., 1990). In the present study, we have carried out a more detailed mapping of the cleavage sites produced by purified mouse leukemia (Ll210) DNA topoisomerase II and extended the study to compare several different classesof drugs.

2. Materials

and Methods

(a) Enzymes and chem2:calx m-AMSA and 5iminodaunorubicin were obtained from the Drug Synthesis and Chemistry Branch, National Cancer Institute, Bethesda, MD. VM-26 was obtained from Bristol-Myers Co., Wallingford, CT, and 2-methyl-ghydroxyellipticinium was a gift from Dr J. B. LePecq, RhBne-Poulenc Santi, Ivry/Seine. France. m-AMSA and VM-26 stock solutions were made in dimethylsulfoxide at 10 mM. Further dilutions were made in distilled water. 5-Iminodaunorubicin and 2-methyl-9-hydroxyellipticinium stock solutions were at 10 mM in water. DNA, restriction endonucleases, phage T4 polynucleotide kinase, calf intestine phosphatase, agarose and polyacrylamide/his were purchased from Bethesda Research Laboratories (Gaithersburg, MD) or from New England Biolabs (Beverly, MA). [y-32P]ATP was purchased from New England Nuclear Research Products (Boston, MA). DNA topoisomerase II was purified from mouse leukemia L1210 cell nuclei as described (Minford et al., 1986), and stored at -70°C in 40% (v/v) glycerol, @35 M-NaCl, 5 m&r-MgCl,, 1 mM-EGTA. 1 mivr-KH,Po,, @2 mMdithiothreitol, @l mM-phenylmethylsulfonyl fluoride, (pH 64). The purified enzyme yielded a single 170,000 M, band after silver staining of SDS/polyacrylamide gels (Minford et al., 1986; Pommier et aE., 1986). (b) Preparation

of end-labeled DNA fragments

DNA fragments were 5’.end-labeled as described (Capranico et al., 1990; Fesen & Pommier, 1989). Briefly, native SV40 DNA was first linearized with one of the restriction enzymes shown in Fig. 1, then the DNA 5’ termini were dephosphorylated with calf alkaline phosphatase and labeled with [y-“P]ATP (asterisks, Fig. 1) using T4 polynucleotide kinase. Next, labeled DNA was subjected to a second enzyme digestion in order to generate uniquely 5’-end-labeled fragments of different length. which were separated and isolated by agarose gel electrophoresis and electroelution. DNA fragments were purified by extraction with phenol/chloroform and pre-

Figure 1. Restriction enzyme labeling sites used for mapping and sequencing topoisomerase II cleavage sites in SV40 DNA. Asterisks indicate the DNA labeled 5’ termini and arrows above the genomic position scale indicate the directions from the labeling site that were analyzed. Coding regions are indicated under the genomic position scale (broken lines with arrows indicate the direction of transcription: arrows pointing to the left. early genes; arrows pointing to the right, lates genes (AG. agnoprotein)). The matrix associated region (MAR) is indicated by the shaded box between positions 4100 and 4400. Position 0 is at the replication origin: replication terminates near position 2630.

cipitation with ethanol between each step and at t,he end of the labeling procedure. (c) Topoisomeraue

II-induced

DNA cleavage reactions

SV40 DNA fragments were equilibrated with OI without drug in 0.01 M-Tris.HCl. (pH 7.5), QO.5 ~w-KC’I. 5 mi%%-MgCl,. @I mM-EDTA, I mM-ATP. 15 pg bovine serum albumin/ml for 5 min before addition of topoisomerase II (40 to 70 ng in 20 pl final reaction volume). Reactions were stopped by adding SDS to a final concentration of 1y/, (w/v) and proteinase K to 100 fig/ml, followed by incubation for 1 h at 42°C. For agarose gel analysis. 3 ~1 of 10 x loading buffer (@30/; (w/v) bromophenol blue. 167, (w/v) Ficoll. @Ol M-Na,HPO,) was added to each sample. which was then heated at 65°C for 1 to 2 min before loading ont,o an agarose gel made in TBE buffer (89 mM-Tris-borate (pH 8), 2 mM-EDTA: Capranico et al., 199Oa: Fesen PC Pommier, 1989; Pommier et al., 1990). Agarose gel electrophoresis was at 2 V/cm overnight. The gels were dried on Whatman 3MM paper sheets and autoradiographed with Kodak XAR-5 film. For DNA sequence analysis, samples were precipitated with ethanol and resuspended in 2.5 ~1 of loading buffer (80% (v/v) formamide. 10 mM-NaOH, 1 mM-EDTA, @l T/b xylene cyanol, 0.1 ?10 bromophenol blue). Samples were heated to 90°C and immediately loaded onto DNA sequencing gels (8 “/b polyacrylamide: 29 : 1 (w/w) acrylamide to bisacrylamide) containing 7 M-urea in TBE buffer. Electrophoresis was at 1500 V (60 W) for 2 to 3 h. (d) Qenomic mappkg

of DNA hrc>alcs

The genomic localization of drug-induced topoisomerase II-mediated DNA breaks was determined as described (Capranico et al., 1990a,b; Fesen & Pommier. 1989; Pommier et aZ., 1990). Briefly, autoradiography films were scanned with a DU-8B Beckman spectrophotometer set at’ 555 nm, and the position and absorbance values were transmitted to a computer. 32P-end-labeled

Topoisomerase

Al

911

II Cleavage Sites in SV40 DNA

23456h 5148 4973 4268 3530

1270 2008

2027 1904

3511 3634

1584 1375

3954 4163

947 a31

4591 4707

564

4974

125

170

Size (bp)

Genomic position

Figure 2. Double-strand breaks induced by mouse L1210 DNA topoisomerase II in SV40 DKA. Reactions were performed in the presence of 1 mmATP at 37°C for 30 min using BanI-labeled DNA (Fig. 1). Reactions were stopped by adding SDS and proteinase K and were analyzed by agarose gel electrophoresis. Lane 1, control DNA; lane 2, topoisomerase II without drug; lanes 3, 4, 5 and 6, topoisomerase II with 10 PM-m-AMSA, 10 PM-VM-26, 1 pm-2-methyl-9-hydroxyellipticinium or 2 PM-5-iminodaunorubicin, respectively. First column of numbers on the right, size of the lambda fragments (bp, lane 1); 2nd column; corresponding positions in SV40 DNA. MAR, nuclear matrixassociated region.

HindIII-EcoRI digested lambda DNA was usually run as DNA markers in 4 lanes/gel to check the uniformity of DNA migration throughout the gel. Regression lines of the logarithm of the fragment size (in base-pairs) versus the migration distance of each fragment from a reference line were determined for the DNA markers. Regression

coefficients were consistently near 999. Each autoradiography lane was analyzed by using the same reference line and the size of each DNA fragment induced by topoisomerase II was computed. The reproducibility of fragment size determinations in different gels was usually within 50 bp.

912

Y. Pommier

et al.

3. Results (a) Double-strand cleavage patterns the SV40 genome

in

The RV40 genome was mapped for double-strand cleavage induced by topoisomerase IT in the presence or absence of several types of drugs (Figs 2 and 3). In the absence of drug. cleavage occurred most prominently in the nuclear matrix-associated region (MAR) and to a lesser extent in several other regions. m-AMSA intensified the cleavage in all regions affected by the enzyme alone. and stimulated cleavage in other regions that were not detectably affected in the absence of drug. VM-26, 2methyl-9-hydroxyellipticinium and 5-iminodaunorubicin each produced a distinctive pattern of enhanced cleavage. Some peaks in the cleavage patterns coincided for different drugs or with peaks generated by the enzyme alone. However, the most consistent region of cleavage that was prominent in all cases was the MAR. Among the three types of drugs examined, the anthraaycline 5-iminodaunorubicin produced the greatest selectivity of enhanced cleavage, most’ prominently in the MAR and near nucleotide 1500. The regions of the replicat,ion origin and the early t,ranscription enhancers, nucleotides 1 to 250, exhlbited relatively little cleavage in the presence or absence of any of the drugs tested (Fig. 3).

t. 1

m-AMSA

VM-26

Enzyme

MAR

alone r

t ?

(b) Cleavage sites in the nuclear associated region

matrix-

Sites of cleavage in the MAR were determined at, the nucleotide level in both the coding and the noncoding strands of the early message. In the coding strand. cleavage was observed in t.he absence of drug at positions 4309,4268,4251? 4247.4208,4143. 4129 and 4118 (Fig. 4). These sites were cleaved very rapidly and exhibited equal int,ensities at one and 30 minutes of reaction. VM-26 exhibited the highest frequency of cleavage sites among the drugs tested. Cleavage enhancements by this drug appeared within one minute, but most sites increased progressively in int’ensity at the three and 30 minute time points. With m-AMSA. some sites increased progressively in intensity (posit*ions 4331 1 4317,4311, 4265 and 4169), while others were strong at one minute and then decreased in intensity (positions 4334, 4286, 4268, 4221, 4214, 4203, 4197, 4143 and 4106). In the non-coding strand of the MAR, cleavage in the absence of drug usually occurred at the sites expected from the dyad symmetry of double-strand cleavage by topoisomerase II (Figs 5 and 6). the mapping of t,opoFigure 6 summarizes isomerase II-induced DNA cleavage sites on both strands of the SV40 MAR. Arrows and numbers correspond to the base covalent’ly linked to the enzyme. Cleavage sites corresponding to doublestrand breaks should be staggered by four bp with a 5’-overhang (Liu et al., 1983; Sander &, Hsieh, 1983).

0

1000

2000 Genomic

3000 position

4000

5000

Figure 3. ToI)oisomsrsse II cleavage intjrnsity Inaps in SV40 DXA. Each map was assembled from agarosr rlectrophoresis scans of 3 restriction fragments. 5’-~ntllabeled at Hrtnl. L4ccT or TagI. (‘leavagr intensities ww c,omputed as described in Ma,terials and M&hods. I)rup treatments from weI’< 2 pm.top to bottom iminodaunoruhincin. IO ~w~I-AMSA. 10 p-VM-26. 01 none. Incubations w-rrr for 30 min at, 37 ‘(I

A11 sites on one strand had a corresponding sit,e on the other strand. except, for one weak sit,r on the upper strand at position 4240. Furt,hertnore. most intense sites on one strand were paired with sites of comparable int,ensity on the other strand. This finding is consistent with the possibility that, topoisomerase TT cleavage in the absence of drug wa,s double-stranded or that single-stranded cleavage. if present. would result from alternative cleavage at corresponding DNA double-&and &es. Symmetrical cleavage of the caomplementary strands also owurred for most of the sites stimulated by m-AMSA (Fig. 5 and data not shown). The distribution of the m-AMSA-stimulated sites within t,he MAR region was non-random and showed evidence of clustering with a periodic interval of approximately 65 nucleotides (Fig. 7).

Topoisomerase

c P130

13301

913

II Cleavage Sites in SV40 DNA

4

4268