J. Mol. Biol. (1990) 215,237-244
Characterization of the Interaction between Topoisomerase II and D N A by Transcriptional Footprinting Bo Thomsen, Christian Bendixen, Kaare Lund, Anni H. Andersen Boe S. Sorensen and Ole Westergaardt Department of Molecular Biology and Plant Physiology University of Aarhus, DK-8000 Aarhus C, Denmark (Received 6 March 1990; accepted 29 M a y 1990) The interaction between calf thymus topoisomerase II and DNA has been characterized using a transcription assay. A highly preferred recognition sequence for topoisomerase II was inserted in either direction downstream from a promoter specific for a bacteriophage RNA polymerase. The presence of topoisomerase I I - D N A complexes on the template provoked blockage of transcription, yielding RNA transcripts terminated 5' to the topoisomerase II binding site. A footprint of topoisomerase II, derived from transcription towards the complex from either side, revealed that eukaryotic topoisomerase II binds a region of 28 base-pairs witch a highly protected central core of 22 base-pairs. The binding region was located symmetrically around the topoisomerase II-mediated cleavage site. In agreement with this result, optimal topoisomerase II-mediated cleavage was observed with a DNA substrate consisting of a 28-met oligonucleotide homologous to the protected region. Stepwise removal of base-pairs from the ends of the 28-met gradually reduced the level of enzyme-mediated cleavage.
1. Introduction
topoisomerase II-mediated cleavage products are 4 bp:~ staggered nicks with the subunits covalently linked to the protruding 5'-phosphoryl ends, leaving recessed 3'-OH ends (Sander & Hsieh, 1983). Moreover, the enzyme cleaves DNA by introducing two co-ordinated single-stranded breaks. During the cleavage reaction the enzyme is able to discriminate between the two strands and a marked strand specificity has been reported (Andersen et al., 1989; Lee et al., 1989a). The reported footprints of eukaryotie topoisomerase II vary in size from 18 bp (Spitzner & Muller, 1988) to 25 bp (Lee et al., 1989b); however, the relationship between the extent of the binding region and the DNA necessary for optimal enzyme-mediated cleavage remains to be elucidated. Interestingly, a number of clinically applicable antitumour agents have been shown to interact specifically with eukaryotic topoisomerase II (Glisson & Ross, 1987). Central to the chemotherapeutic value of these drugs is stabilization of
Eukaryotic DNA topoisomerase II (EC 5.99.1.3) is an essential enzyme that changes DNA topology by concerted breakage and rejoining of both strands in the double helix. Type II topoisomerases purified from a large number of eukaryotic sources are all dimers consisting of two identical subunits of approximately 170,000 M r that appear similar in structural and enzymic properties (for reviews, see Vosberg, 1985; Wang, 1985; Osheroff, 1989a). Topoisomerase II has been shown to be important for DNA replication (Nelson et al., 1986; Brill et al., 1987; Yang et al., 1987), transcription (Brill et al., 1987; Glikin & Blangy, 1986), chromosome condensation and segregation (Newport, 1987; Newport & Spann, 1987; DiNardo et al., 1984; Uemura et al., 1987) and it seems to be a major structural protein in the eukaryotic chromosome (Earnshaw & Heck, 1985; Earnshaw et al., 1985; Berrios et al., 1985; Gasser & Laemmli, 1986; Gasser et al., 1986). The DNA-binding properties of topoisomerase II are poorly characterized, whereas the cleavage reaction has been described more thoroughly. Thus,
~:Abbreviations used: bp, base-pair(s); Ro 15-0216, 2-dimethylamino-4'[(1-methyl-2-nitroimidazol-5-yl) methoxy]acetanilide; mAMSA, amsacrine,4'-(9aeridinylamino)methanesulphon-m-anisidide.
t Author to whom all correspondence should be addressed. 0022-2836/90/180237-08 $03.00/0
237
© 1990AcademicPressLimited
B. Thomsen et al.
238
the covalent topoisomerase II-DNA complex. The drugs presumably act by inhibition of the religation half-reaction of the catalytic cycle, thereby accumulating the cleaved DNA intermediate in which the enzyme is covalently linked to the broken strands (Osheroff, 1989b). Thus, stabilization of the enzyme-DNA cleavable complex with antitumour drugs extends the half-life of the covalent intermediate, resulting in an elevated amount of cleavage products as observed by treatment with strong detergents (for reviews, see Bodley & Liu, 1988; D'Arpa & Liu, 1989). In general, topoisomerase II interacts with sites on DNA that share only minimal nucleotide sequence .homology and the highly degenerate consensus sequences derived from cleavage studies vary in size from 15 to 18 bp (Sander & Hsieh, 1985; Spitzner & Muller, 1988; Spitzner et al., 1990). In spite of the relaxed sequence specificity, preferred cleavage sites have been described (Andersen et al., I989; Sander & Hsieh, 1983). To investigate the DNA-binding properties of the topoisomerase II, we have taken advantage of such a sequence previously characterized as a very strong cleavage site by Sander & Hsieh (1983). The sequence was inserted in either orientation downstream from the SP6 RNA polymerase promoter, thus allowing transcription towards both sides of the topoisomerase II-DNA complex. In order to optimize the interference with the elongation process, the complexes were stabilized using topoisomerase II-targetting drugs. Combination of the two sets of blocked RNA transcripts generated by transcription towards either side of the complex defines the DNA-binding region of topoisomerase II. The method has been shown to be precise to + 1 bp from the site of ligand binding (White & Phillips, 1989; Shiet al., 1988). Furthermore, we have recently made a direct comparison of the SP6 polymerase footprint of topoisomerase I with a conventional micrococcal nuclease footprint and found that the binding region defined by these two techniques differed by only 1 to 2 bp (Bendixen et al., 1990). As the RNA polymerase is devoid of the sequence specificity innate to most nucleases, this method provides improved resolution compared with the traditional methods of nuclease footprinting. The results demonstrate that the topoisomerase II-DNA complex spans 28 bp with a highly protected central core of 22 bp. The finding that a 28-mer oligonueleotide homologous to the protected region is able to support optimal cleavage by topoisomerase II accentuates the functional importance of the defined DNA-binding region. On the basis of these results, a possible spatial arrangement of the topoisomerase I I - D N A complex is discussed.
through the polymin P precipitation step. Further purification was according to Shelton et al. (1983). (b) In vitro transcriptions SP6 RNA polymerase transcriptions in the presence of topoisomerase II and Ro 15-0216 (Hoffman-LaRoche: the amounts are given in the Figure legends) were carried out in 30/d of reaction buffer containing 40 mM-Tris'HCl (pH 7-6), 5 mM-MgCI2, 10 mm dithiothreitol, 0"5 mmspermidine, 0"3 mMeach ATP, GTP and CTP, 50 ~M-UTP (Boehringer), 25/~Ci of [a-32P]UTP (Amersham), 20 fmol of DNA template, 1 unit of RNAsin/#! (Promega), and 0"2 unit of SP6-polymerase//~l (Promega). All transcriptions were performed with linear templates that yield fulllength transcripts of 283 nucleotides and 316 nucleotides in pCB641 and pCB651, respectively. Transcriptions were carried out at 37 °C for 15 rain. Transcriptions were terminated by addition of EDTA (10 mM final concn) followed by extraction with phenol/chloroform and precipitation with ethanol. The transcription products were analysed on 6% (w/v) polyacrylamide sequencing gels. (c) RNA transcript sequencing RNA sequencing reactions were done as above except for the addition of 3'-deoxyribonucleotide 5'-triphosphate analogues (Pharmacia) of the nucleotide being sequenced. The concentration of the 3'-dATP, 3'-GTP and 3'-dCTP was 150 gM and the concentration of 3'-dUTP was 25 gin. (d) Topoisomerase I I-.rnediated D N A cleavage reactions The 5'-end-labelled oligonucleotides (40fmol) were incubated with 100 units~ of topoisomerase II in the absence or presence of 1 mm-Ro 15-0216 in 100-/~l reaction volumes in 10 mm-Tris'HCl (pH 7"5), 25 mM-NaCl, 0"l mm-EDTA, 5 mm-CaCl2 for l0 min at 30°C. SDS and EDTA were added to final concentrations of I% (w/v) and l0 ram, respectively, and incubation was continued for 5 min at 42°C. After treatment with proteinase K (500/~g/ml, 30min at 37°C) and precipitation with ethanol, the cleavage products were redissolved in 50% (v/v) formamide, 5 m~-EDTA and analysed on 12% denaturing polyacrylamide gels. End-labelling was performed using phage T4 polynucleotide kinase (Biolabs) and [?-32P]ATP (ICN).
(e) Oligonucleotides Oligonucleotides were synthesized on a DNA synthesizer model 381A from Applied Biosystems and purified by polyacrylamide gel electrophoresis. (f) Quantification of RNA transcription products on polyacrylamide gels A quantitative measure of the relative band intensity was obtained by densitometric scanning of the autoradiograms using a Shimadzu Chromatoscanner, model CS930.
3. Results 2. Materials and Methods
(a) Purification of topoisomerase I I
(a) Topoisomerase I I - D N A complexes stabilized by Ro 15-0216 interfere with R N A transcription
DNA topoisomerase II was purified from calf thymus glands by the procedure of Shomburg & Grosse (1986)
To determine the size of the topoisomerase II binding region surrounding a strong cleavage site,
Transcriptional Footprinting of Topoisomerase I I
(a) 5 "AAATCTAACAATG'CGCTCATCGTCATCCTC 3 "TTTAGATTGTTACGCGAGTAGCAGTAGGAG
(b)
3° 5"
5" G A G G A T G A C G A T G ' A G C G C A T T G T T A G A T T T
I
3"
I 30 bp
Figure 1. A representation of the in vitro transcription
templates (a) pCB641 and (b) pCB651. The large arrows indicate the transcription initiation sites for SP6-polymerase. The displayed sequences denote the topoisomerase II recognition sequence with the cleavage site marked with small arrows.
an in vitro transcription assay was applied. The transcription vectors pCB641 and pCB651 were constructed by insertion of the H i n d I I I - S a l I fragment spanning from nucleotide positions 29 to 303 of pNC1 (Thomsen et al., 1987) into the H i n d I I I - S a l I sites of the polylinker region of pSP64 and pSP65, respectively. This locates the strong topoisomerase II recognition sequence to position +79 in pCB641 and position +242 in pCB651 relative to the transcription initiation site for the SP6 RNA polymerase. The transcription templates are depicted in Figure 1. Addition of topoisomerase II to the template results in complex formation at the recognition site of the enzyme. The transcriptional footprinting assay relies on blockage of the advancing RNA polymerase at the borders of the region protected by topoisomerase II, thereby generating prematurely terminated transcription products. The two sets of blocked transcripts resulting from transcription towards both sides of the complex demarcates the binding region of topoisomerase II. Figure 2 shows the initial experiment where the transcription templates (20 fmol) were preincubat'ed for two minutes at 37 °C in a transcriptional competent reaction mixture with 100 units of calf thymus topoisomerase II, allowing complexes to form prior to initiation of transcription. RNA polymerase was added and incubation was continued for 15 minutes. Following termination with 10 mMEDTA, the transcription products were analysed by denaturing polyacrylamide gel electrophoresis. The data demonstrated that topoisomerase II alone has no detectable effect on elongation, indicating that bound topoisomerase II per se does not impede elongation (Fig. 2(a) and (b), lane 2). Thus, in order to minimize readthrough resulting from dissociation of topoisomerase II, the complexes were stabilized using the topoisomerase II-targetting agent
239
Ro 15-0216. This drug is a 2-nitroimidazole compound that, analogous to antitumour drugs, causes accumulation of cleavable complexes, thereby strongly enhancing the cleavage frequency. In contrast to the traditional antitumour drugs, e.g. mAMSA, which enhances topoisomerase II-mediated cleavage at multiple sites, the action of Ro 15-0216 is restricted to enhancement of major cleavage sites already present i~:ilthe absence of drugs. Thus, in the transcription experiments advantage was taken of the fact that Ro 15-0216 stimulates the formation of a fivefold higher amount of cleavable complexes at the strong topoisomerase II recognition sequence examined in this report as compared to mAMSA (Sorensen et al., 1990). Employment of Ro 15-0216 retarded the dissociation rate of topoisomerase II from the template so that the complex effectively interfered with the elongation process. Thus, in the 'presence of increasing concentrations of Ro 15-0216, two distinctive blocked RNA species were generated when the RNA polymerase approaches the complex from either side (Fig. 2(a) and (b), lanes 4 to 7). The transcriptional process is unperturbed by Ro 15-0216 alone (lane 31. Moreover, inhibition of transcription by drug-stabilized topoisomerase I I - D N A complexes was not confined to 1%o 15-0216, as blockage of elongation has been observed with the topoisomerase II-targetting antitumour drug mAMSA (data not shown). Considering the fact that the RNA polymerase proceeds up to 1 bp from the ligand binding site (White & Phillips, 1989; Bendixen et al., 1990), these initial experiments demonstrate that the transcriptional footprinting assay is applicable for delimitation of the DNA sequence bound by topoisomerase II.
(b) Topoisomerase I I binds a 28 bp region with a strongly protected core of 22 bp The use of drugs to stabilize the topoisomerase I I - D N A complexes implies that the footprinting data primarily reveal the DNA contacts of the cleavable complexes formed during cycles of DNA cleavage and religation. In order to demarcate the extent of these contacts, the exact locations of the sites of blockage were determined by RNA transcript sequencing using 3'-deoxynucleotide analogues. In Figure 3(a) and (b), the topoisomerase II-blocked transcripts (lane 1) were analysed alongside nucleotide sequencing reactions (lanes 2 to 5). This analysis located the sites of transcription termination to nucleotide positions 9 and 12 relative to the topoisomerase II-linked nucleotides from either transcription direction. On template pCB651 (Fig. l(b)) an additional termination product corresponding to position 8 was generated. Quantification of the amount of RNA transcripts by densitometric scanning of the autoradiographs revealed a two- to threefold higher protection at position 9 as compared with position 12. The footprinting data are summarized in Figure 3(e). Taken
B. Thomsen et al.
240
1
2
J
3
4
5
6
7
1
2
3
4
5
6
7
.
(o)
(b)
Figure 2. Topoisomera~se II-DNA complexes stabilized by Ro 15-0216 block the advancing P~NA polymerase. (a) Transcription of pCB641. (b) Transcription of pCB651. (a) and (b) Lane l, without topoisomerase II, without Ro 15-0216; lane 2, 100 units of topoisomerase II, without Ro 15-0216; lane 3, without topoisomerase II, 2 mM-Ro 15-0216; lanes 4 to 7, 100 units of topoisomerase II and 0"07 mM, 0"33 raM, 0'67 mM, 1"3 mM-Ro 15-0216, respectively. Arrows indicate the blocked transcripts.
together, the data show that the calf thymus topoisomerase II-DNA cleavable complex spans a region of 28 bp located symmetrically around the site of cleavage, with strong protection of the central 22 bp. Similar results have been obtained with topoisomerase II purified from other organisms, e.g. Drosophila (data not shown). (c) Duplex DNA requirements for topoisomerase I I-mediated cleavage To address the question of whether the protected 28 bp region is sufficient to support cleavage with high frequency, oligonucleotides homologous to the protected region were synthesized. As controls, a
35 bp oligomer and four smaller DNA substrates varying in size from 16 bp to 24 bp encompassing the topoisomerase II cleavage site were employed. All cleavage reactions were performed in both the presence and absence of 1 mM-RO 15-0216 in order to probe for drug-induced alterations in cleavage site selection. Each of the 5'-end-labelled oligonucleotide (40 fmol) were incubated at 30°C for ten minutes with 100 units of topoisomerase II in the absence and presence of Ro 15-0216. Following treatment with SDS, the samples were digested with proteinase K and the cleavage products analysed by denaturing polyacrylamide gel electrophoresis. The result of this experiment is shown in Figure 4(a). It appears that topoisomerase II cleaves the 28 bp
Transcriptional Footprinting of Topoisomerase II
1
2
3
4
5
1
2
3
241
4
5
(b)
(a)
I
5 "ATGAAATCTAACAATG,
CGCTCATCGTCATCCTCGGC
3"
!
3 "TACTTTAGATTGTTACGCGAGTAGCAGTAGGAGCCG
5"
l
(c) Figure 3. RNA transcript sequencing of the topoisomerase II-DNA blocked transcript. (a) Transcription of pCB641. (b) Transcription of pCB651. (a) and (b) Lane l, topoisomerase II-DNA blocked transcript; lanes 2 to 5, RNA transcript sequencing using 150/aM-3'-dATP, 25/aM-3'-dUTP, 150/aM-3'-dGTP and 150 ~tM-3'-dCTP, respectively. Arrows indicate the blocked transcripts. (c) A representation of the binding region. Filled bars indicate the highly protected region; the open bars indicate the weakly protected regions.
2
3
4
5
6
Ca)
7
8
9
10
11 12
n."
U
o
o
25
50
100-
16
18 20
24
26
28 30
(b)
Bose-poirs
22
32
34
36
Figure 4. Topoisomerase II-mediated cleavage on oligonucleotides homologous to the protected region. (a) Topoisomerase I I (100 units) was incubated with 40 fmol of aligonucleotides of a defined length and treated with SDS. The cleavage products were analysed on 12% polyacrylamide gels. Lanes 1 and 2, 16-mer; lanes 3 and 4, 18-met; lanes 5 and 6, 20-met; lanes 7 and 8, 24-mer; lanes 9 and 10, 28-mer; lanes 11 and 12, 35-mer. The cleavage reactions were carried out in the absence (odd-numbered lanes) or in the presence of 1 mM-Ro 15-0216 (even-numbered lanes). (b) The relative cleavage frequency of each oligonucleotide plotted against the length of the oligonucleotide. The cleavage frequency of the 35-mer was set to 100%.
1
Transcriptional Footprinting of Topoisomerase I I oligomer homologous to the protected sequence with optimal efficiency as the enzyme cleaved the 28-met (lanes 9 and 10) and the 35-met (lanes l l and 12) to the same extent. Topoisomerase II-mediated cleavage of the 24-mer (lanes 7 and 8), 20-mer (lanes 5 and 6) and 18-mer (lanes 3 and 4) is possible but proceeded with gradually reduced efficiency, whereas cleavage of the 16 bp duplex oligomer (lanes 1 and 2) was below the detection level. The cleavage specificity was unaffected by Ro 15-0216, although the level of enzyme-mediated cleavage was extensively stimulated. The functional importance of the length of the DNA substrate with respect to the cleavage reaction is further underscored by densitometric scanning of the autoradiograph. Thus, Figure 4(b) clearly reveals an abrupt increase in cleavage efficiency when the DNA substrate length exceeds 22 to 24 bp, i.e. approximately the size of the strongly protected DNA-binding region. Collectively, these data demonstrate that the DNA-binding region of topoisomerase II actually encompasses the minimal duplex DNA required for maximal topoisomerase II-mediated cleavage, and that the substrate dependence of the cleavage reactiondoes not change with the addition of Ro 15-0216.
4. Discussion
Considering the central role of the cleavable complex formed during the catalytic cycle of topoisomerase II, it appears important to define the DNA contacts of this intermediate. As topoisomerase II does not give a footprint with chemical probes such as MPE'Fe(II) or dimethyl sulphate (Lee et al., 1989b), an alternative approach was necessary. This was accomplished by the employment of drug-stabilized enzyme-DNA cleavable complexes in transcriptional footprinting assay. Thus, the data demonstrated that bound topoisomerase II per se did not cause termination of transcription, indicating that the interactions between topoisomerase II and DNA in the absence of drugs are too transient to block RNA polymerase movement. However, when topoisomerase II-targetting agents known to stabilize the cleavable complexes was employed, a significant amount of blocked transcripts was generated. RNA sequencing of the complex-blocked transcripts permitted precise determination of the DNA-binding region of topoisomerase II. The data demonstrated that calf thymus topoisomerase II binds a 28 bp region located symmetrically around the cleavage site with strong protection of the central 22 bp. A DNase I footprint has been performed of a Drosophila topoisomerase I I - D N A complex in the presence of ATP~S (Lee et al., 1989b). The non-hydrolysable analogue ATP?S supports DNA strand passage but not enzyme turnover and, as a result, impairs dissociation of topoisomerase II from the DNA (Osheroff, 1986). Thus, this non-dissociable poststrand passage complex is distinct from the
243
enzyme-DNA cleavable complex examined in this study as it is non-covalent in nature. The DNA-binding region of the post-strand passage complex spans approximately 25 bp with the cleavage site located near the centre of the protected region. Taken together with our results, these data indicate that the extent of the DNA-binding region of the post-strand passage complex and the cleavable complex are coincident. Furthermore, the Stokes radius of topoisomerase II has been estimated as 65 A (Sander & Hsieh, 1983}. Considering that the size of the protected region is equivalent to roughly 95 A (1 A =0"l nm), this binding region of eukaryotic topoisomerase II is clearly not sufficient to allow extensive coiling of the DNA around the enzyme as has been demonstrated for the prokaryotic counterpart of topoisomerase II, DNA gyrase (Kirkegaard & Wang, 1981). To test the relevance of the binding region demarcated by transcriptional footprinting with respect to enzyme function, we performed topoisomerase II-mediated cleavages on duplex DNA oligomers varying in size from 16 to 35 bp. This analysis demonstrated that a 28 bp substrate homologous to the binding region of calf thymus topoisomerase II promoted maximal cleavage. Increasing the length of the DNA substrate to 35 bp or even 87 bp (data not shown} did not augment the level of enzymemediated cleavage. Moreover, stepwise removal of base-pairs from both ends of the 28-mer oligonucleotide resulted in a strongly reduced level of cleavage efficiency. The reduced ability of topoisomerase II to form cleavable complexes with DNA substrates less than 28 bp in length indicates the loss of important enzyme-DNA contacts, conceivably resulting in a decreased binding affinity. Alternatively, the missing contacts between topoisomerase II and DNA may influence the cleavage ability of the enzyme. It is, however, noteworthy that topoisomerase II retained its cleavage specificity on all cleavable DNA substrates employed, indicating that the sequence specificity is determined by enzyme-DNA interactions established within the smallest cleavable oligomer, i.e. the 18 bp region in which the cleavage site is centrally located. These data demonstrate that the minimal length of the DNA substrate required for optimal enzymemediated DNA cleavage is identical with the 28 bp binding region of topoisomerase II. This result further corroborates the suggestion that extensive wrapping of DNA around topoisomerase II may not be critical for enzyme function. The observation that the topoisomerase I I - D N A complex provoked the formation of two blocked transcripts when approached from either side is intriguing. An interpretation could be that the topoisomerase II subunit linked to the non-coding strand interferes less with elongation as compared with the subunit linked to the coding strand. According to this idea, the outermost weak blocked transcript was generated when the polymerase encountered the topoisomerase II subunit linked to the non-coding strand, whereas the strong blocked
244
B. Thomsen et al.
3 bp 3" 5"
3 bp Figure 5. Model of the topoisomerase II-DNA cleavable complex. The shaded area represents the homodimeric structure of topoisomerase II. The small circle denotes the covalent linkage between topoisomerase II and the cleaved DNA.
transcript resulted from interference with the topoisomerase II subunit linked to the coding strand. In support of this assumption are results showing that topoisomerase I covalently linked to the non-coding strand had no inhibitory effect on elongation, while complexes on the coding strand efficiently blocked the elongation process (Bendixen et al., 1990). Thus, the generation of two blocked transcripts probably reflects the homodimeric nature of topoisomerase II. The footprinting data might therefore suggest a spatial arrangement of the topoisomerase II-DNA complex in which the two subunits of the enzyme are mutually displaced by 3 bp on the DNA substrate. A schematic of this spatial arrangement of the cleavable complex is depicted in Figure 5. The potency of drug-stabilized topoisomerase II-DNA complexes to impede transcription elongation has important implications. Thus, our observations indicate that the RNA synthesis inhibition resulting from exposure of cells to topoisomerase II-targetting antitumour drugs (Grieder et al., 1974) is caused by physical blockage of the advancing RNA polymerases along the transcription unit. Consistently, several lines of evidence indicate that the cytotoxicity of antitumour agents is mediated primarily by cellular responses to drug-stabilized topoisomerase II-DNA complexes rather than inhibition of catalytic activity (for a review, see D'Arpa & Liu, 1989). This work was supported by contract BI-6-0171-DK with EURATOM (CEC, Brussels), The Danish Cancer Society (88-060), The Danish Natural Science Research Council (11-5724/12-6011), The Danish National Agency of Technology (1985-133/001-85.521), and the Aarhus University Bioregulation Research Centre. References
Andersen, A. H., Christiansen, K., Zechiedrich, E.L., Jensen, P. S., Osheroff, N. & Westergaard, O. (1989). Biochemistry, 28, 6237-6244. Bendixen, C., Thomsen, B., Alsner, J. & Westergaard, O. (1990). Biochemistry, 29, 5613-5619. Berrios, M., Osheroff,N. & Fisher, P. A. (1985). Proc. Nat. Acad. Sci., U.S.A. 78, 2883-2887.
Bodley, A. L. & Liu, L. F. (1988). Biotechnology, 6, 1315-1319. Brill, S. J., DiNardo, S. S., Voelkel-Meiman, K. & Sternglanz, R. (1987). Nature (London), 326, 414-416. D'Arpa, P. & Liu, L. F. (1989). Biochim. Biophys. Acta, 989, 163-177. DiNardo, S., Voelkel, K. & Sternglanz, R. (1984). Proc. Nat. Acad. Sci., U.S.A. 81, 2616-2620. Earnshaw, W. C. & Heck, M. M. S. (1985). J. Cell. Biol. 100, 1716-1725. Earnshaw, W. C., Halligan, B., Crooke, C.A., Heck, M. M. S. & Liu, L.F. (1985). J. Cell. Biol. 100, 1706-1715. Gasser, S. M. & Laemmli, U. K. (1986). EMBO J. 5, 511-518. Gasser, S. M., Laroche, T., Falguet, J., Boy de la Tour, E. & Laemmli, U. K. (1986). J. Mol. Biol. 188, 613-629. Glikin, G. C. & Blangy, D. (1986). EMBO J. 5, 151-155. Glisson, B. S. & Ross, W. E. (1987). Pharmacol. Ther. 32, 89-106. Grieder, A., Maurer, R. & St~helin, H. (1974). Cancer Res. 34, 1788-1793. Kirkegaard, K. & Wang, J. C. (1981). Cell, 23,721-729. Lee, M., Sander, M. & Hsieh, T. (1989a). J. Biol. Chem. 264, 13510-13518. Lee, M. P., Sander, M. & Hsieh, T. (1989b). J. Biol. Chem. 264, 21779-21787. Nelson, W. G., Liu, L. F. & Coffey, D. S. (1986). Nature (London), 322, 187-189. Newport, J. (1987). Cell, 48, 205-217. Newport, J. & Spann, T. (1987). Cell, 48, 219-230. Osheroff, N. (1986). J. Biol. Chem. 261, 9944-9950. Osheroff, N. (1989a). Pharmacol. Ther. 41,223-241. Osheroff, N. (1989b). Biochemistry, 28, 6157-6160. Sander, M. & Hsieh, T. (1983). J. Biol. Chem. 258, 8421-8428. Sander, M. & Hsieh, T. (1985). Nucl. Acids Res. 13, 1057-1072. Shelton, E. R., Osheroff, N. & Brutlag, D. L. (1983). J. Biol. Chem. 258, 9530-9535. Shi, Y., Gamper, H., Houten, B. V. & Hearst, J. E. (1988). J. Mol. Biol. 199, 277-293. Shomburg, U. & Grosse, F. (1986). Eur. J. Biochem. 160, 451-457. Sorensen, B. S., Jensen, P. S., Andersen, A. H., Christiansen, K., Alsner, J., Thomsen, B. & Westergaard, O. (1990). Biochemistry, in the press. Spitzner, J. R. & Muller, M. T. (I988). Nucl. Acids Res. 16, 5533-5556. Spitzner, J. R., Chung, I. K. & Muller, M. T. (1990). Nucl. Acids Res. 18, 1-11. Thomsen, B., Mollerup, S., Bonven, B.J., Frank, R., BlScker, H., Nielsen, O. F. & Westergaard, O. (1987). EMBO J. 6, 1817-1823. Uemura, T., Ohkura, H., Adachi, Y., Morino, K., Shiozaki, K. & Yanagida, M. (1987). Cell, 50, 917-925. Vosberg, H.-P. (1985). Curt. Top. Microbiol. Immunol. 114, 19-102. Wang, J. C. (1985). Annu. Rev. Biochem. 54, 665-697. White, R. J. & Phillips, D. R. (1989). Biochemistry, 28, 6259-6269. Yang, L., Wold, M. S., Li, J. J., Kelly, T. J. & Lui, L. F. (1987). Proc. Nat. Acad. Sci., U.S.A. 84, 950-954.
Edited by P. Chambon
Characterization of the Interaction between Topoisomerase II and D N A by Transcriptional Footprinting Bo Thomsen, Christian Bendixen, Kaare Lund, Anni H. Andersen Boe S. Sorensen and Ole Westergaardt Department of Molecular Biology and Plant Physiology University of Aarhus, DK-8000 Aarhus C, Denmark (Received 6 March 1990; accepted 29 M a y 1990) The interaction between calf thymus topoisomerase II and DNA has been characterized using a transcription assay. A highly preferred recognition sequence for topoisomerase II was inserted in either direction downstream from a promoter specific for a bacteriophage RNA polymerase. The presence of topoisomerase I I - D N A complexes on the template provoked blockage of transcription, yielding RNA transcripts terminated 5' to the topoisomerase II binding site. A footprint of topoisomerase II, derived from transcription towards the complex from either side, revealed that eukaryotic topoisomerase II binds a region of 28 base-pairs witch a highly protected central core of 22 base-pairs. The binding region was located symmetrically around the topoisomerase II-mediated cleavage site. In agreement with this result, optimal topoisomerase II-mediated cleavage was observed with a DNA substrate consisting of a 28-met oligonucleotide homologous to the protected region. Stepwise removal of base-pairs from the ends of the 28-met gradually reduced the level of enzyme-mediated cleavage.
1. Introduction
topoisomerase II-mediated cleavage products are 4 bp:~ staggered nicks with the subunits covalently linked to the protruding 5'-phosphoryl ends, leaving recessed 3'-OH ends (Sander & Hsieh, 1983). Moreover, the enzyme cleaves DNA by introducing two co-ordinated single-stranded breaks. During the cleavage reaction the enzyme is able to discriminate between the two strands and a marked strand specificity has been reported (Andersen et al., 1989; Lee et al., 1989a). The reported footprints of eukaryotie topoisomerase II vary in size from 18 bp (Spitzner & Muller, 1988) to 25 bp (Lee et al., 1989b); however, the relationship between the extent of the binding region and the DNA necessary for optimal enzyme-mediated cleavage remains to be elucidated. Interestingly, a number of clinically applicable antitumour agents have been shown to interact specifically with eukaryotic topoisomerase II (Glisson & Ross, 1987). Central to the chemotherapeutic value of these drugs is stabilization of
Eukaryotic DNA topoisomerase II (EC 5.99.1.3) is an essential enzyme that changes DNA topology by concerted breakage and rejoining of both strands in the double helix. Type II topoisomerases purified from a large number of eukaryotic sources are all dimers consisting of two identical subunits of approximately 170,000 M r that appear similar in structural and enzymic properties (for reviews, see Vosberg, 1985; Wang, 1985; Osheroff, 1989a). Topoisomerase II has been shown to be important for DNA replication (Nelson et al., 1986; Brill et al., 1987; Yang et al., 1987), transcription (Brill et al., 1987; Glikin & Blangy, 1986), chromosome condensation and segregation (Newport, 1987; Newport & Spann, 1987; DiNardo et al., 1984; Uemura et al., 1987) and it seems to be a major structural protein in the eukaryotic chromosome (Earnshaw & Heck, 1985; Earnshaw et al., 1985; Berrios et al., 1985; Gasser & Laemmli, 1986; Gasser et al., 1986). The DNA-binding properties of topoisomerase II are poorly characterized, whereas the cleavage reaction has been described more thoroughly. Thus,
~:Abbreviations used: bp, base-pair(s); Ro 15-0216, 2-dimethylamino-4'[(1-methyl-2-nitroimidazol-5-yl) methoxy]acetanilide; mAMSA, amsacrine,4'-(9aeridinylamino)methanesulphon-m-anisidide.
t Author to whom all correspondence should be addressed. 0022-2836/90/180237-08 $03.00/0
237
© 1990AcademicPressLimited
B. Thomsen et al.
238
the covalent topoisomerase II-DNA complex. The drugs presumably act by inhibition of the religation half-reaction of the catalytic cycle, thereby accumulating the cleaved DNA intermediate in which the enzyme is covalently linked to the broken strands (Osheroff, 1989b). Thus, stabilization of the enzyme-DNA cleavable complex with antitumour drugs extends the half-life of the covalent intermediate, resulting in an elevated amount of cleavage products as observed by treatment with strong detergents (for reviews, see Bodley & Liu, 1988; D'Arpa & Liu, 1989). In general, topoisomerase II interacts with sites on DNA that share only minimal nucleotide sequence .homology and the highly degenerate consensus sequences derived from cleavage studies vary in size from 15 to 18 bp (Sander & Hsieh, 1985; Spitzner & Muller, 1988; Spitzner et al., 1990). In spite of the relaxed sequence specificity, preferred cleavage sites have been described (Andersen et al., I989; Sander & Hsieh, 1983). To investigate the DNA-binding properties of the topoisomerase II, we have taken advantage of such a sequence previously characterized as a very strong cleavage site by Sander & Hsieh (1983). The sequence was inserted in either orientation downstream from the SP6 RNA polymerase promoter, thus allowing transcription towards both sides of the topoisomerase II-DNA complex. In order to optimize the interference with the elongation process, the complexes were stabilized using topoisomerase II-targetting drugs. Combination of the two sets of blocked RNA transcripts generated by transcription towards either side of the complex defines the DNA-binding region of topoisomerase II. The method has been shown to be precise to + 1 bp from the site of ligand binding (White & Phillips, 1989; Shiet al., 1988). Furthermore, we have recently made a direct comparison of the SP6 polymerase footprint of topoisomerase I with a conventional micrococcal nuclease footprint and found that the binding region defined by these two techniques differed by only 1 to 2 bp (Bendixen et al., 1990). As the RNA polymerase is devoid of the sequence specificity innate to most nucleases, this method provides improved resolution compared with the traditional methods of nuclease footprinting. The results demonstrate that the topoisomerase II-DNA complex spans 28 bp with a highly protected central core of 22 bp. The finding that a 28-mer oligonueleotide homologous to the protected region is able to support optimal cleavage by topoisomerase II accentuates the functional importance of the defined DNA-binding region. On the basis of these results, a possible spatial arrangement of the topoisomerase I I - D N A complex is discussed.
through the polymin P precipitation step. Further purification was according to Shelton et al. (1983). (b) In vitro transcriptions SP6 RNA polymerase transcriptions in the presence of topoisomerase II and Ro 15-0216 (Hoffman-LaRoche: the amounts are given in the Figure legends) were carried out in 30/d of reaction buffer containing 40 mM-Tris'HCl (pH 7-6), 5 mM-MgCI2, 10 mm dithiothreitol, 0"5 mmspermidine, 0"3 mMeach ATP, GTP and CTP, 50 ~M-UTP (Boehringer), 25/~Ci of [a-32P]UTP (Amersham), 20 fmol of DNA template, 1 unit of RNAsin/#! (Promega), and 0"2 unit of SP6-polymerase//~l (Promega). All transcriptions were performed with linear templates that yield fulllength transcripts of 283 nucleotides and 316 nucleotides in pCB641 and pCB651, respectively. Transcriptions were carried out at 37 °C for 15 rain. Transcriptions were terminated by addition of EDTA (10 mM final concn) followed by extraction with phenol/chloroform and precipitation with ethanol. The transcription products were analysed on 6% (w/v) polyacrylamide sequencing gels. (c) RNA transcript sequencing RNA sequencing reactions were done as above except for the addition of 3'-deoxyribonucleotide 5'-triphosphate analogues (Pharmacia) of the nucleotide being sequenced. The concentration of the 3'-dATP, 3'-GTP and 3'-dCTP was 150 gM and the concentration of 3'-dUTP was 25 gin. (d) Topoisomerase I I-.rnediated D N A cleavage reactions The 5'-end-labelled oligonucleotides (40fmol) were incubated with 100 units~ of topoisomerase II in the absence or presence of 1 mm-Ro 15-0216 in 100-/~l reaction volumes in 10 mm-Tris'HCl (pH 7"5), 25 mM-NaCl, 0"l mm-EDTA, 5 mm-CaCl2 for l0 min at 30°C. SDS and EDTA were added to final concentrations of I% (w/v) and l0 ram, respectively, and incubation was continued for 5 min at 42°C. After treatment with proteinase K (500/~g/ml, 30min at 37°C) and precipitation with ethanol, the cleavage products were redissolved in 50% (v/v) formamide, 5 m~-EDTA and analysed on 12% denaturing polyacrylamide gels. End-labelling was performed using phage T4 polynucleotide kinase (Biolabs) and [?-32P]ATP (ICN).
(e) Oligonucleotides Oligonucleotides were synthesized on a DNA synthesizer model 381A from Applied Biosystems and purified by polyacrylamide gel electrophoresis. (f) Quantification of RNA transcription products on polyacrylamide gels A quantitative measure of the relative band intensity was obtained by densitometric scanning of the autoradiograms using a Shimadzu Chromatoscanner, model CS930.
3. Results 2. Materials and Methods
(a) Purification of topoisomerase I I
(a) Topoisomerase I I - D N A complexes stabilized by Ro 15-0216 interfere with R N A transcription
DNA topoisomerase II was purified from calf thymus glands by the procedure of Shomburg & Grosse (1986)
To determine the size of the topoisomerase II binding region surrounding a strong cleavage site,
Transcriptional Footprinting of Topoisomerase I I
(a) 5 "AAATCTAACAATG'CGCTCATCGTCATCCTC 3 "TTTAGATTGTTACGCGAGTAGCAGTAGGAG
(b)
3° 5"
5" G A G G A T G A C G A T G ' A G C G C A T T G T T A G A T T T
I
3"
I 30 bp
Figure 1. A representation of the in vitro transcription
templates (a) pCB641 and (b) pCB651. The large arrows indicate the transcription initiation sites for SP6-polymerase. The displayed sequences denote the topoisomerase II recognition sequence with the cleavage site marked with small arrows.
an in vitro transcription assay was applied. The transcription vectors pCB641 and pCB651 were constructed by insertion of the H i n d I I I - S a l I fragment spanning from nucleotide positions 29 to 303 of pNC1 (Thomsen et al., 1987) into the H i n d I I I - S a l I sites of the polylinker region of pSP64 and pSP65, respectively. This locates the strong topoisomerase II recognition sequence to position +79 in pCB641 and position +242 in pCB651 relative to the transcription initiation site for the SP6 RNA polymerase. The transcription templates are depicted in Figure 1. Addition of topoisomerase II to the template results in complex formation at the recognition site of the enzyme. The transcriptional footprinting assay relies on blockage of the advancing RNA polymerase at the borders of the region protected by topoisomerase II, thereby generating prematurely terminated transcription products. The two sets of blocked transcripts resulting from transcription towards both sides of the complex demarcates the binding region of topoisomerase II. Figure 2 shows the initial experiment where the transcription templates (20 fmol) were preincubat'ed for two minutes at 37 °C in a transcriptional competent reaction mixture with 100 units of calf thymus topoisomerase II, allowing complexes to form prior to initiation of transcription. RNA polymerase was added and incubation was continued for 15 minutes. Following termination with 10 mMEDTA, the transcription products were analysed by denaturing polyacrylamide gel electrophoresis. The data demonstrated that topoisomerase II alone has no detectable effect on elongation, indicating that bound topoisomerase II per se does not impede elongation (Fig. 2(a) and (b), lane 2). Thus, in order to minimize readthrough resulting from dissociation of topoisomerase II, the complexes were stabilized using the topoisomerase II-targetting agent
239
Ro 15-0216. This drug is a 2-nitroimidazole compound that, analogous to antitumour drugs, causes accumulation of cleavable complexes, thereby strongly enhancing the cleavage frequency. In contrast to the traditional antitumour drugs, e.g. mAMSA, which enhances topoisomerase II-mediated cleavage at multiple sites, the action of Ro 15-0216 is restricted to enhancement of major cleavage sites already present i~:ilthe absence of drugs. Thus, in the transcription experiments advantage was taken of the fact that Ro 15-0216 stimulates the formation of a fivefold higher amount of cleavable complexes at the strong topoisomerase II recognition sequence examined in this report as compared to mAMSA (Sorensen et al., 1990). Employment of Ro 15-0216 retarded the dissociation rate of topoisomerase II from the template so that the complex effectively interfered with the elongation process. Thus, in the 'presence of increasing concentrations of Ro 15-0216, two distinctive blocked RNA species were generated when the RNA polymerase approaches the complex from either side (Fig. 2(a) and (b), lanes 4 to 7). The transcriptional process is unperturbed by Ro 15-0216 alone (lane 31. Moreover, inhibition of transcription by drug-stabilized topoisomerase I I - D N A complexes was not confined to 1%o 15-0216, as blockage of elongation has been observed with the topoisomerase II-targetting antitumour drug mAMSA (data not shown). Considering the fact that the RNA polymerase proceeds up to 1 bp from the ligand binding site (White & Phillips, 1989; Bendixen et al., 1990), these initial experiments demonstrate that the transcriptional footprinting assay is applicable for delimitation of the DNA sequence bound by topoisomerase II.
(b) Topoisomerase I I binds a 28 bp region with a strongly protected core of 22 bp The use of drugs to stabilize the topoisomerase I I - D N A complexes implies that the footprinting data primarily reveal the DNA contacts of the cleavable complexes formed during cycles of DNA cleavage and religation. In order to demarcate the extent of these contacts, the exact locations of the sites of blockage were determined by RNA transcript sequencing using 3'-deoxynucleotide analogues. In Figure 3(a) and (b), the topoisomerase II-blocked transcripts (lane 1) were analysed alongside nucleotide sequencing reactions (lanes 2 to 5). This analysis located the sites of transcription termination to nucleotide positions 9 and 12 relative to the topoisomerase II-linked nucleotides from either transcription direction. On template pCB651 (Fig. l(b)) an additional termination product corresponding to position 8 was generated. Quantification of the amount of RNA transcripts by densitometric scanning of the autoradiographs revealed a two- to threefold higher protection at position 9 as compared with position 12. The footprinting data are summarized in Figure 3(e). Taken
B. Thomsen et al.
240
1
2
J
3
4
5
6
7
1
2
3
4
5
6
7
.
(o)
(b)
Figure 2. Topoisomera~se II-DNA complexes stabilized by Ro 15-0216 block the advancing P~NA polymerase. (a) Transcription of pCB641. (b) Transcription of pCB651. (a) and (b) Lane l, without topoisomerase II, without Ro 15-0216; lane 2, 100 units of topoisomerase II, without Ro 15-0216; lane 3, without topoisomerase II, 2 mM-Ro 15-0216; lanes 4 to 7, 100 units of topoisomerase II and 0"07 mM, 0"33 raM, 0'67 mM, 1"3 mM-Ro 15-0216, respectively. Arrows indicate the blocked transcripts.
together, the data show that the calf thymus topoisomerase II-DNA cleavable complex spans a region of 28 bp located symmetrically around the site of cleavage, with strong protection of the central 22 bp. Similar results have been obtained with topoisomerase II purified from other organisms, e.g. Drosophila (data not shown). (c) Duplex DNA requirements for topoisomerase I I-mediated cleavage To address the question of whether the protected 28 bp region is sufficient to support cleavage with high frequency, oligonucleotides homologous to the protected region were synthesized. As controls, a
35 bp oligomer and four smaller DNA substrates varying in size from 16 bp to 24 bp encompassing the topoisomerase II cleavage site were employed. All cleavage reactions were performed in both the presence and absence of 1 mM-RO 15-0216 in order to probe for drug-induced alterations in cleavage site selection. Each of the 5'-end-labelled oligonucleotide (40 fmol) were incubated at 30°C for ten minutes with 100 units of topoisomerase II in the absence and presence of Ro 15-0216. Following treatment with SDS, the samples were digested with proteinase K and the cleavage products analysed by denaturing polyacrylamide gel electrophoresis. The result of this experiment is shown in Figure 4(a). It appears that topoisomerase II cleaves the 28 bp
Transcriptional Footprinting of Topoisomerase II
1
2
3
4
5
1
2
3
241
4
5
(b)
(a)
I
5 "ATGAAATCTAACAATG,
CGCTCATCGTCATCCTCGGC
3"
!
3 "TACTTTAGATTGTTACGCGAGTAGCAGTAGGAGCCG
5"
l
(c) Figure 3. RNA transcript sequencing of the topoisomerase II-DNA blocked transcript. (a) Transcription of pCB641. (b) Transcription of pCB651. (a) and (b) Lane l, topoisomerase II-DNA blocked transcript; lanes 2 to 5, RNA transcript sequencing using 150/aM-3'-dATP, 25/aM-3'-dUTP, 150/aM-3'-dGTP and 150 ~tM-3'-dCTP, respectively. Arrows indicate the blocked transcripts. (c) A representation of the binding region. Filled bars indicate the highly protected region; the open bars indicate the weakly protected regions.
2
3
4
5
6
Ca)
7
8
9
10
11 12
n."
U
o
o
25
50
100-
16
18 20
24
26
28 30
(b)
Bose-poirs
22
32
34
36
Figure 4. Topoisomerase II-mediated cleavage on oligonucleotides homologous to the protected region. (a) Topoisomerase I I (100 units) was incubated with 40 fmol of aligonucleotides of a defined length and treated with SDS. The cleavage products were analysed on 12% polyacrylamide gels. Lanes 1 and 2, 16-mer; lanes 3 and 4, 18-met; lanes 5 and 6, 20-met; lanes 7 and 8, 24-mer; lanes 9 and 10, 28-mer; lanes 11 and 12, 35-mer. The cleavage reactions were carried out in the absence (odd-numbered lanes) or in the presence of 1 mM-Ro 15-0216 (even-numbered lanes). (b) The relative cleavage frequency of each oligonucleotide plotted against the length of the oligonucleotide. The cleavage frequency of the 35-mer was set to 100%.
1
Transcriptional Footprinting of Topoisomerase I I oligomer homologous to the protected sequence with optimal efficiency as the enzyme cleaved the 28-met (lanes 9 and 10) and the 35-met (lanes l l and 12) to the same extent. Topoisomerase II-mediated cleavage of the 24-mer (lanes 7 and 8), 20-mer (lanes 5 and 6) and 18-mer (lanes 3 and 4) is possible but proceeded with gradually reduced efficiency, whereas cleavage of the 16 bp duplex oligomer (lanes 1 and 2) was below the detection level. The cleavage specificity was unaffected by Ro 15-0216, although the level of enzyme-mediated cleavage was extensively stimulated. The functional importance of the length of the DNA substrate with respect to the cleavage reaction is further underscored by densitometric scanning of the autoradiograph. Thus, Figure 4(b) clearly reveals an abrupt increase in cleavage efficiency when the DNA substrate length exceeds 22 to 24 bp, i.e. approximately the size of the strongly protected DNA-binding region. Collectively, these data demonstrate that the DNA-binding region of topoisomerase II actually encompasses the minimal duplex DNA required for maximal topoisomerase II-mediated cleavage, and that the substrate dependence of the cleavage reactiondoes not change with the addition of Ro 15-0216.
4. Discussion
Considering the central role of the cleavable complex formed during the catalytic cycle of topoisomerase II, it appears important to define the DNA contacts of this intermediate. As topoisomerase II does not give a footprint with chemical probes such as MPE'Fe(II) or dimethyl sulphate (Lee et al., 1989b), an alternative approach was necessary. This was accomplished by the employment of drug-stabilized enzyme-DNA cleavable complexes in transcriptional footprinting assay. Thus, the data demonstrated that bound topoisomerase II per se did not cause termination of transcription, indicating that the interactions between topoisomerase II and DNA in the absence of drugs are too transient to block RNA polymerase movement. However, when topoisomerase II-targetting agents known to stabilize the cleavable complexes was employed, a significant amount of blocked transcripts was generated. RNA sequencing of the complex-blocked transcripts permitted precise determination of the DNA-binding region of topoisomerase II. The data demonstrated that calf thymus topoisomerase II binds a 28 bp region located symmetrically around the cleavage site with strong protection of the central 22 bp. A DNase I footprint has been performed of a Drosophila topoisomerase I I - D N A complex in the presence of ATP~S (Lee et al., 1989b). The non-hydrolysable analogue ATP?S supports DNA strand passage but not enzyme turnover and, as a result, impairs dissociation of topoisomerase II from the DNA (Osheroff, 1986). Thus, this non-dissociable poststrand passage complex is distinct from the
243
enzyme-DNA cleavable complex examined in this study as it is non-covalent in nature. The DNA-binding region of the post-strand passage complex spans approximately 25 bp with the cleavage site located near the centre of the protected region. Taken together with our results, these data indicate that the extent of the DNA-binding region of the post-strand passage complex and the cleavable complex are coincident. Furthermore, the Stokes radius of topoisomerase II has been estimated as 65 A (Sander & Hsieh, 1983}. Considering that the size of the protected region is equivalent to roughly 95 A (1 A =0"l nm), this binding region of eukaryotic topoisomerase II is clearly not sufficient to allow extensive coiling of the DNA around the enzyme as has been demonstrated for the prokaryotic counterpart of topoisomerase II, DNA gyrase (Kirkegaard & Wang, 1981). To test the relevance of the binding region demarcated by transcriptional footprinting with respect to enzyme function, we performed topoisomerase II-mediated cleavages on duplex DNA oligomers varying in size from 16 to 35 bp. This analysis demonstrated that a 28 bp substrate homologous to the binding region of calf thymus topoisomerase II promoted maximal cleavage. Increasing the length of the DNA substrate to 35 bp or even 87 bp (data not shown} did not augment the level of enzymemediated cleavage. Moreover, stepwise removal of base-pairs from both ends of the 28-mer oligonucleotide resulted in a strongly reduced level of cleavage efficiency. The reduced ability of topoisomerase II to form cleavable complexes with DNA substrates less than 28 bp in length indicates the loss of important enzyme-DNA contacts, conceivably resulting in a decreased binding affinity. Alternatively, the missing contacts between topoisomerase II and DNA may influence the cleavage ability of the enzyme. It is, however, noteworthy that topoisomerase II retained its cleavage specificity on all cleavable DNA substrates employed, indicating that the sequence specificity is determined by enzyme-DNA interactions established within the smallest cleavable oligomer, i.e. the 18 bp region in which the cleavage site is centrally located. These data demonstrate that the minimal length of the DNA substrate required for optimal enzymemediated DNA cleavage is identical with the 28 bp binding region of topoisomerase II. This result further corroborates the suggestion that extensive wrapping of DNA around topoisomerase II may not be critical for enzyme function. The observation that the topoisomerase I I - D N A complex provoked the formation of two blocked transcripts when approached from either side is intriguing. An interpretation could be that the topoisomerase II subunit linked to the non-coding strand interferes less with elongation as compared with the subunit linked to the coding strand. According to this idea, the outermost weak blocked transcript was generated when the polymerase encountered the topoisomerase II subunit linked to the non-coding strand, whereas the strong blocked
244
B. Thomsen et al.
3 bp 3" 5"
3 bp Figure 5. Model of the topoisomerase II-DNA cleavable complex. The shaded area represents the homodimeric structure of topoisomerase II. The small circle denotes the covalent linkage between topoisomerase II and the cleaved DNA.
transcript resulted from interference with the topoisomerase II subunit linked to the coding strand. In support of this assumption are results showing that topoisomerase I covalently linked to the non-coding strand had no inhibitory effect on elongation, while complexes on the coding strand efficiently blocked the elongation process (Bendixen et al., 1990). Thus, the generation of two blocked transcripts probably reflects the homodimeric nature of topoisomerase II. The footprinting data might therefore suggest a spatial arrangement of the topoisomerase II-DNA complex in which the two subunits of the enzyme are mutually displaced by 3 bp on the DNA substrate. A schematic of this spatial arrangement of the cleavable complex is depicted in Figure 5. The potency of drug-stabilized topoisomerase II-DNA complexes to impede transcription elongation has important implications. Thus, our observations indicate that the RNA synthesis inhibition resulting from exposure of cells to topoisomerase II-targetting antitumour drugs (Grieder et al., 1974) is caused by physical blockage of the advancing RNA polymerases along the transcription unit. Consistently, several lines of evidence indicate that the cytotoxicity of antitumour agents is mediated primarily by cellular responses to drug-stabilized topoisomerase II-DNA complexes rather than inhibition of catalytic activity (for a review, see D'Arpa & Liu, 1989). This work was supported by contract BI-6-0171-DK with EURATOM (CEC, Brussels), The Danish Cancer Society (88-060), The Danish Natural Science Research Council (11-5724/12-6011), The Danish National Agency of Technology (1985-133/001-85.521), and the Aarhus University Bioregulation Research Centre. References
Andersen, A. H., Christiansen, K., Zechiedrich, E.L., Jensen, P. S., Osheroff, N. & Westergaard, O. (1989). Biochemistry, 28, 6237-6244. Bendixen, C., Thomsen, B., Alsner, J. & Westergaard, O. (1990). Biochemistry, 29, 5613-5619. Berrios, M., Osheroff,N. & Fisher, P. A. (1985). Proc. Nat. Acad. Sci., U.S.A. 78, 2883-2887.
Bodley, A. L. & Liu, L. F. (1988). Biotechnology, 6, 1315-1319. Brill, S. J., DiNardo, S. S., Voelkel-Meiman, K. & Sternglanz, R. (1987). Nature (London), 326, 414-416. D'Arpa, P. & Liu, L. F. (1989). Biochim. Biophys. Acta, 989, 163-177. DiNardo, S., Voelkel, K. & Sternglanz, R. (1984). Proc. Nat. Acad. Sci., U.S.A. 81, 2616-2620. Earnshaw, W. C. & Heck, M. M. S. (1985). J. Cell. Biol. 100, 1716-1725. Earnshaw, W. C., Halligan, B., Crooke, C.A., Heck, M. M. S. & Liu, L.F. (1985). J. Cell. Biol. 100, 1706-1715. Gasser, S. M. & Laemmli, U. K. (1986). EMBO J. 5, 511-518. Gasser, S. M., Laroche, T., Falguet, J., Boy de la Tour, E. & Laemmli, U. K. (1986). J. Mol. Biol. 188, 613-629. Glikin, G. C. & Blangy, D. (1986). EMBO J. 5, 151-155. Glisson, B. S. & Ross, W. E. (1987). Pharmacol. Ther. 32, 89-106. Grieder, A., Maurer, R. & St~helin, H. (1974). Cancer Res. 34, 1788-1793. Kirkegaard, K. & Wang, J. C. (1981). Cell, 23,721-729. Lee, M., Sander, M. & Hsieh, T. (1989a). J. Biol. Chem. 264, 13510-13518. Lee, M. P., Sander, M. & Hsieh, T. (1989b). J. Biol. Chem. 264, 21779-21787. Nelson, W. G., Liu, L. F. & Coffey, D. S. (1986). Nature (London), 322, 187-189. Newport, J. (1987). Cell, 48, 205-217. Newport, J. & Spann, T. (1987). Cell, 48, 219-230. Osheroff, N. (1986). J. Biol. Chem. 261, 9944-9950. Osheroff, N. (1989a). Pharmacol. Ther. 41,223-241. Osheroff, N. (1989b). Biochemistry, 28, 6157-6160. Sander, M. & Hsieh, T. (1983). J. Biol. Chem. 258, 8421-8428. Sander, M. & Hsieh, T. (1985). Nucl. Acids Res. 13, 1057-1072. Shelton, E. R., Osheroff, N. & Brutlag, D. L. (1983). J. Biol. Chem. 258, 9530-9535. Shi, Y., Gamper, H., Houten, B. V. & Hearst, J. E. (1988). J. Mol. Biol. 199, 277-293. Shomburg, U. & Grosse, F. (1986). Eur. J. Biochem. 160, 451-457. Sorensen, B. S., Jensen, P. S., Andersen, A. H., Christiansen, K., Alsner, J., Thomsen, B. & Westergaard, O. (1990). Biochemistry, in the press. Spitzner, J. R. & Muller, M. T. (I988). Nucl. Acids Res. 16, 5533-5556. Spitzner, J. R., Chung, I. K. & Muller, M. T. (1990). Nucl. Acids Res. 18, 1-11. Thomsen, B., Mollerup, S., Bonven, B.J., Frank, R., BlScker, H., Nielsen, O. F. & Westergaard, O. (1987). EMBO J. 6, 1817-1823. Uemura, T., Ohkura, H., Adachi, Y., Morino, K., Shiozaki, K. & Yanagida, M. (1987). Cell, 50, 917-925. Vosberg, H.-P. (1985). Curt. Top. Microbiol. Immunol. 114, 19-102. Wang, J. C. (1985). Annu. Rev. Biochem. 54, 665-697. White, R. J. & Phillips, D. R. (1989). Biochemistry, 28, 6259-6269. Yang, L., Wold, M. S., Li, J. J., Kelly, T. J. & Lui, L. F. (1987). Proc. Nat. Acad. Sci., U.S.A. 84, 950-954.
Edited by P. Chambon
