Topoisomerase II.etoposide interactions direct the formation of drug-induced enzyme-DNA cleavage complexes.

Topoisomerase II is the target for several highly active anticancer drugs that induce cell death by enhancing enzyme-mediated DNA scission. Although these agents dramatically increase levels of nucleic acid cleavage in a site-specific fashion, little is understood regarding the mechanism by which they alter the DNA site selectivity of topoisomerase II. Therefore, a series of kinetic and binding experiments were carried out to determine the mechanistic basis by which the anticancer drug, etoposide, enhances cleavage complex formation at 22 specific nucleic acid sequences. In general, maximal levels of DNA scission (i.e. Cmax) varied over a considerably larger range than did the apparent affinity of etoposide (i.e. Km) for these sites, and there was no correlation between these two kinetic parameters. Furthermore, enzyme·drug binding and order of addition experiments indicated that etoposide and topoisomerase II form a kinetically competent complex in the absence of DNA. These findings suggest that etoposide· topoisomerase II (rather than etoposide·DNA) interactions mediate cleavage complex formation. Finally, rates of religation at specific sites correlated inversely with Cmax values, indicating that maximal levels of etoposide-induced scission reflect the ability of the drug to inhibit religation at specific sequences rather than the affinity of the drug for site-specific enzyme-DNA complexes.

Topoisomerase II is the target for several highly active anticancer drugs that induce cell death by enhancing enzyme-mediated DNA scission. Although these agents dramatically increase levels of nucleic acid cleavage in a site-specific fashion, little is understood regarding the mechanism by which they alter the DNA site selectivity of topoisomerase II. Therefore, a series of kinetic and binding experiments were carried out to determine the mechanistic basis by which the anticancer drug, etoposide, enhances cleavage complex formation at 22 specific nucleic acid sequences. In general, maximal levels of DNA scission (i.e. C max ) varied over a considerably larger range than did the apparent affinity of etoposide (i.e. K m ) for these sites, and there was no correlation between these two kinetic parameters. Furthermore, enzyme⅐drug binding and order of addition experiments indicated that etoposide and topoisomerase II form a kinetically competent complex in the absence of DNA. These findings suggest that etoposide⅐ topoisomerase II (rather than etoposide⅐DNA) interactions mediate cleavage complex formation. Finally, rates of religation at specific sites correlated inversely with C max values, indicating that maximal levels of etoposide-induced scission reflect the ability of the drug to inhibit religation at specific sequences rather than the affinity of the drug for site-specific enzyme-DNA complexes.
As a prerequisite for its DNA passage reaction, topoisomerase II generates transient double-stranded breaks in the nucleic acid backbone (3)(4)(5). In order to maintain the integrity of the cleaved genetic material during this process, the enzyme forms a proteinaceous bridge that spans the nucleic acid break. This bridge is anchored by covalent phosphotyrosyl bonds established between the active site residues of the homodimeric enzyme and the newly created 5Ј-DNA termini (15)(16)(17)(18). Because the covalent topoisomerase II-cleaved DNA complex (referred to as the cleavage complex) is normally a short-lived intermediate in the catalytic cycle of the enzyme, it is tolerated by the cell. However, when present in high concentrations, cleavage complexes become potentially toxic, promoting frameshift mutations, permanent double-stranded DNA breaks, illegitimate recombination, and apoptosis (6, 8, 19 -23).
The cytotoxic potential of topoisomerase II has been exploited clinically by the development of anticancer drugs that generate high levels of covalent enzyme-DNA cleavage complexes (19,(22)(23)(24)(25)(26). Because rapidly proliferating cells contain high concentrations of topoisomerase II (27)(28)(29)(30)(31), aggressive malignancies are most susceptible to these agents (19,25,29,32,33). Etoposide, which is the most widely prescribed chemotherapeutic agent currently used for the treatment of human cancers (26,34,35), is targeted to topoisomerase II (36). This drug increases topoisomerase II-mediated DNA breakage primarily by inhibiting the ability of the enzyme to religate cleaved nucleic acid molecules (37,38). Etoposide binds to DNA in a nonintercalative manner in the absence of topoisomerase II but does so with relatively low affinity compared with many anticancer drugs targeted to the enzyme (39).
Although topoisomerase II cleaves DNA at preferred sequences, little is understood regarding the mechanism by which the enzyme identifies (i.e. selects) its sites of action. Furthermore, the distribution of DNA sites cleaved by topoisomerase II is altered dramatically in the presence of anticancer agents, and the levels of scission at drug-induced sites vary significantly (26). On the basis of previous sequence analysis studies, it appears that etoposide is predisposed to induce topoisomerase II-mediated DNA cleavage at sites that are immediately 3Ј to cytosine residues (26,40,41). However, the molecular interactions that underlie this base preference are unknown.
To address the basis for cleavage site selection by topoisomerase II in the presence of drugs, a combination of kinetic and binding experiments were employed to determine the mechanism by which etoposide enhances cleavage complex formation at 22 specific sites. Results indicate that etoposide⅐topoisomerase II (rather than drug⅐DNA) interactions mediate cleavage complex formation and that maximal levels of etoposide-induced scission reflect the ability of the drug to inhibit religation at specific sequences rather than the affinity of the drug for site-specific enzyme⅐DNA complexes. These conclusions are consistent with the recently proposed "positional poison model" 1 for the action of DNA lesions and topoisomerase II-targeted anticancer drugs.

EXPERIMENTAL PROCEDURES
Drosophila melanogaster topoisomerase II was purified from embryonic Kc cells as described previously by Shelton et al. (43). Saccharomyces cerevisiae topoisomerase II was overexpressed and purified from yeast cells as described by Elsea et al. (44) except that the initial phosphocellulose chromatography was replaced by chromatography on hydroxylapatite (43). pBR322 plasmid DNA was purified as described previously (45). Etoposide and ellipticine were purchased from Sigma and were stored at Ϫ20°C as a 10 or 20 mM stock, respectively, in dimethyl sulfoxide. Tris-HCl and urea were purchased from Sigma; SDS was from Merck; proteinase K was from U. S. Biochemical Corp.; restriction endonucleases, calf intestine alkaline phosphatase, and polynucleotide kinase were from New England BioLabs; [␥-32 P]ATP (6000 Ci/mmol) was from Amersham Corp.; and [ 3 H]etoposide (ϳ1 Ci/mmol) was from Moravek. All other chemicals were analytical reagent grade.
DNA Labeling and Purification-A uniquely end-labeled 564-bp 2 DNA substrate (residues 375-939 in pBR322) was prepared as follows. pBR322 plasmid DNA (50 g) was digested with restriction endonuclease EagI and dephosphorylated with calf intestine alkaline phosphatase. The DNA was then phosphorylated with T4 polynucleotide kinase using 10 M [␥-32 P]ATP, digested with restriction endonuclease BamHI, and subjected to electrophoresis on a 5% nondenaturing polyacrylamide gel. The uniquely end-labeled 564-bp product was detected by shadowing with UV light, excised, and eluted overnight at 37°C in 2.5 M ammonium acetate. Finally, the DNA was ethanol precipitated and resuspended in water.
DNA Cleavage Reactions-Levels of topoisomerase II-mediated DNA cleavage were assayed using a protocol similar to that described by Knab et al. (46). Cleavage reactions contained 1.4 nM (25 ng) labeled DNA, 3.5 nM (60 ng) Drosophila topoisomerase II, etoposide (at concentrations ranging from 1 to 100 M), or a dimethyl sulfoxide solvent control (final concentration, 1% (v/v)) in a total of 50 l of assay buffer (10 mM Tris-HCl, pH 7.9, 50 mM NaCl, 50 mM KCl, 5 mM MgCl 2 , 0.1 mM EDTA, and 2.5% glycerol (v/v)). Following a 10-min incubation at 30°C, cleavage complexes were trapped by the addition of SDS (1% final concentration), and topoisomerase II was digested for 30 min at 45°C with proteinase K (final concentration, 80 g/ml) in the presence of EDTA (final concentration, 15 mM). DNA cleavage products were ethanol precipitated twice, dried, and resuspended with 40% formamide, 8.4 mM EDTA, 0.02% bromphenol blue, and 0.02% xylene cyanol FF. Samples were subjected to electrophoresis in 8% sequencing gels (46), fixed in 10% methanol/10% acetic acid for 30 min, and dried. Reaction products were analyzed using a Molecular Dynamics PhosphorImager. DNA sequence ladders of the cleavage substrate were generated by the dideoxynucleotide sequencing method (47).
Order of addition cleavage experiments were performed as described above, except that etoposide was preincubated in assay buffer with either topoisomerase II or DNA in the absence of the missing component for 2 min at 30°C. During this period, the missing component was also incubated at 30°C. Cleavage was initiated by mixing the two samples, and reaction mixtures were incubated for 10 s to 10 min. Cleavage reactions were terminated and analyzed as described above.
DNA Religation-Rates of DNA religation were determined as outlined previously (38). Briefly, cleavage/religation equilibria were established for 10 min at 30°C as described above. Religation was initiated by shifting reaction mixtures to 55°C (the temperature at which the religation reaction is favored). Reactions were terminated at various times (20 s to 2 min) and analyzed as described above.
Etoposide⅐Topoisomerase II Binding-Interactions between etoposide and topoisomerase II in the absence of DNA were monitored by one of two techniques. Due to the large amounts of enzyme required for these techniques, overexpressed yeast topoisomerase II (44) was utilized for these experiments.
In the first technique, etoposide⅐topoisomerase II binding was determined by the ability of the drug to compete with the anticancer agent ellipticine for interactions with the enzyme. Competition was quantitated by monitoring the decrease in the intensity of ellipticine fluorescence. Steady-state fluorescence spectroscopy was performed as outlined previously (48). For all fluorescence experiments, ellipticine was excited at 326 nm, and emission light was monitored at 420 nm. The emission polarizer was fixed at the magic angle (54.7°), and a 420 nm interference band pass filter was employed to separate fluorescence from scattered light. Data were acquired at 25°C. Samples contained 1 M ellipticine, 100 nM enzyme, and 0 -500 nM etoposide in a final volume of 500 l of 20 mM HEPES, pH 7.9, 100 mM NaCl, 5 mM MgCl 2 , and 0.1 mM EDTA and were incubated for 6 min prior to fluorescence measurements. All chemicals were ultrapure grade to minimize nonspecific fluorescence. The buffer background was subtracted from intensities for binding calculations.
In the second technique, etoposide⅐topoisomerase II binding was monitored by a nitrocellulose filter assay. Assays were carried out by a modification of the method described by Higgins and Cozzarelli (49). Nitrocellulose filters (0.45 m, Millipore) were presoaked in assay buffer. Binding reactions contained 5 M topoisomerase II and 2 M [ 3 H]etoposide in 80 l of assay buffer. Reactions were incubated at room temperature for 6 min and rapidly applied to the center of nitrocellulose filters at a flow rate of ϳ5 ml/min. Filters were washed three times with 1 ml of ice-cold assay buffer, dried, and quantified by liquid scintillation counting.

RESULTS
Topoisomerase II-targeted anticancer drugs greatly alter the DNA cleavage sites utilized by the enzyme (19, 40, 41, 50 -55). Although sequence analyses of drug-induced DNA scission have identified the spectra of sites associated with different agents, the features that dictate the nucleotide specificity of topoisomerase II-targeted drugs remain an enigma. Therefore, a series of kinetic and binding experiments were utilized to determine the mechanism by which the anticancer drug, etoposide, stimulates topoisomerase II-DNA cleavage complex formation at specific sequences.
Kinetics of Site-specific Topoisomerase II-mediated Cleavage Complex Formation in the Presence of Etoposide-As a first step toward elucidating the mechanistic basis for the sequence selectivity of topoisomerase II in the presence of anticancer drugs, relationships between the affinity of etoposide for specific enzyme-DNA cleavage complexes and levels of drug-induced scission in those complexes were examined. This was accomplished by analyzing the kinetics of site-specific DNA cleavage mediated by topoisomerase II in the presence of etoposide.
A uniquely end-labeled 564-bp BamHI-EagI fragment of pBR322 plasmid was used as a DNA substrate and cleavage was monitored over a concentration range of etoposide that spanned 2 orders of magnitude (1-100 M). As seen in Fig. 1, etoposide enhanced topoisomerase II-mediated cleavage at numerous sites in this nucleic acid substrate. Twenty-two sites of cleavage (numbered in order from the 5Ј end of the 32 P-labeled strand) were analyzed. The corresponding nucleotide sequence position in pBR322 is listed for each site in Table I. Cleavage sites located beyond (i.e. further from the labeled DNA terminus) site 22 were not included in the analysis due to decreased resolution of these reaction products, as well as an apparent decrease in cleavage complexes formed at these sites observed at high drug concentrations (presumably due to additional scission events at sites nearer to the labeled DNA terminus).
Previous studies by Pommier et al. (26,40) indicate that etoposide preferentially induces DNA cleavage mediated by murine topoisomerase II at sequences with a cytosine residue immediately 5Ј to the scissile bond. Although the present data were not subjected to rigorous statistical analysis, a similar preference was observed for cleavage with the Drosophila type II enzyme (Table I). Of the 22 sites that were characterized in the present study, 17 contained a cytosine either 5Ј to the point of cleavage on the strand that was analyzed or 5Ј to the predicted point of cleavage four bases downstream on the complementary strand. Six of the sequences characterized contained cytosine residues 5Ј to the point of cleavage on both strands. Therefore, the sequence specificity of etoposide with Drosophila topoisomerase II appears to be similar to that of the mammalian enzyme.
Levels of etoposide-induced topoisomerase II-mediated DNA cleavage at specific sites were analyzed by Eadie-Hofstee plots (the kinetic equivalent of Scatchard binding plots). Representative data for DNA scission at four sites are shown in Fig. 2. For each site, C max (the theoretical maximal cleavage induced by etoposide) is represented by the intercept with the x axis, and the apparent K m value (the kinetic affinity of etoposide) is represented by Ϫ1/slope of the line (56). C max and apparent K m values for all DNA sites examined are given in Fig. 3 and Table I. Under the conditions utilized, the C max values (which are expressed as the percentage of the initial DNA substrate cleaved at a given site) for etoposide-induced scission varied by more than 60-fold for the 22 sites analyzed. In contrast, apparent K m values varied by only 12-fold. Furthermore, much of this variation in drug affinity was due to sites 2-5, which appear to comprise a region on the DNA that is a poor substrate for drug-induced topoisomerase II-mediated cleavage. The sites in this cluster displayed apparent K m values for etoposide that were 3-5 times higher than any other sites examined and C max values that were among the lowest determined. The poor ability of etoposide to induce scission within this cluster indicates that specific DNA sequences and/or structures may exert a dominant negative effect on the assembly of cleavage complexes and preclude drug action.
However, for the other 18 sites examined, the apparent K m values for etoposide varied no more than 3-fold, despite the wide range (ϳ60-fold) of C max values. Moreover, no correlation existed between these two sets of kinetic constants. These results indicate that for the overwhelming majority of DNA sites examined, maximal levels of etoposide-induced scission were independent of the affinity of the drug for cleavage complexes formed at these sites. Because the nucleotide sequences of the sites examined displayed little similarity (other than the predisposition for a cytosine 5Ј to the scissile bond) (Table I), this finding implies that complex formation is not driven by the affinity of etoposide for specific sequences in the duplex DNA.
The above conclusion is supported by two additional lines of evidence. First, there is a strong correlation between the kinetic specificity (i.e. C max /K m ) of etoposide and C max values (for all sites examined) but none with apparent K m values (Fig. 3). Second, although maximal levels of scission spanned nearly 2 orders of magnitude, the time required to achieve DNA cleav- age equilibrium for every site analyzed varied less than 2-fold (not shown). The time to equilibrium value encompasses the interval required to form the noncovalent ternary topoisomerase II⅐drug⅐DNA complex followed by the time needed to induce cleavage within this complex. Because DNA cleavage equilibrium is established in less than 5 s in the absence of drugs (57), the longer times (ϳ60 s) required to attain equilibrium in the presence of etoposide presumably reflect the period required to form the noncovalent ternary complex. Therefore, if the intrinsic affinity of etoposide for specific sites on the DNA substrate differed substantially, equilibrium times likely would have differed accordingly.
Binding of Etoposide to Topoisomerase II-The data presented above indicate that the specificity of cleavage complex formation is not due to interactions between etoposide and individual nucleic acid sequences. Taken together with the relatively low binding affinity of etoposide for DNA (39), these data imply that etoposide enters cleavage complexes primarily through interactions with topoisomerase II. To address this possibility, two independent experimental approaches were employed to assess etoposide⅐enzyme binding in the absence of DNA. Due to the high levels of topoisomerase II required for the techniques discussed below, overexpressed yeast enzyme was utilized for these studies.
The first approach took advantage of a fluorescence-based assay previously used to quantitate binding of the anticancer agent, ellipticine, to topoisomerase II (48,58). Because the excitation spectrum of etoposide overlaps that of the enzyme (precluding a direct measurement of drug binding), etoposide⅐ topoisomerase II interactions were monitored by the ability of the drug to displace ellipticine bound to the enzyme.
The binding of ellipticine to the enzyme, even in the absence of nucleic acids, (K D Ϸ 160 nM) is accompanied by a large increase in the intrinsic fluorescence intensity of the drug ( em ϭ 420 nm) (Fig. 4). As determined by the decrease in fluorescence intensity at 420 nm, etoposide effectively competed with ellipticine for binding to topoisomerase II in the absence of DNA (Fig. 4). This result indicates that etoposide has direct interactions with the enzyme. Furthermore, consistent with enzymological studies of several topoisomerase II-targeted anticancer drugs (59,60), this finding suggests that etoposide and ellipticine share an interaction domain on the enzyme.
The second approach utilized a nitrocellulose filter binding assay to monitor [ 3 H]etoposide⅐topoisomerase II binding. Although this assay does not represent a true equilibrium system, it has been used previously to determine interactions between DNA and eukaryotic topoisomerase II (61)(62)(63). At enzyme and drug concentrations of 5 and 2 M, respectively, ϳ 1 ⁄3 of the etoposide was bound to topoisomerase II (not shown). Furthermore, [ 3 H]etoposide binding decreased ϳ90% when 200 M nonradioactive etoposide was included in assay mixtures. Along with the fluorescence data, these results provide strong evidence that etoposide binds to topoisomerase II in the absence of DNA.
Etoposide⅐Topoisomerase II Interactions Enhance the Rate of DNA Cleavage Complex Formation-Although the above binding studies demonstrate a direct association between etoposide and topoisomerase II, they do not establish the relevance of this interaction in promoting enzyme-mediated DNA scission. Therefore, to address this critical issue, an order of addition experiment was carried out to determine the effects of drug⅐enzyme interactions on the rate of DNA cleavage complex formation. In this experiment, etoposide was incubated with either topoisomerase II or DNA prior to cleavage, and scission was initiated by the addition of nucleic acid or enzyme (as appropriate). The 564-bp fragment of pBR322 that was used for the kinetic analysis was employed as the DNA substrate for this study.
Results of a typical time course experiment are shown in Fig.  5. Quantitation of cleavage at four specific sites (which were chosen to represent DNA sequences encompassing the range of drug-induced C max values in this substrate) is also shown. In all cases, the initial velocity of cleavage complex formation was ϳ2-4-fold greater when etoposide was preincubated with topoisomerase II as opposed to DNA. These findings strongly suggest that the etoposide⅐topoisomerase II complex established in the absence of DNA represents a kinetically competent intermediate in the pathway of cleavage complex formation. Together with the kinetic data presented above, these studies lead to the conclusion that etoposide enters cleavage complexes primarily through interactions with topoisomerase II rather than DNA.
Effects of Etoposide on Site-specific Topoisomerase II-mediated DNA Religation-Although etoposide⅐topoisomerase II interactions appear to direct DNA cleavage complex formation (consistent with the relatively low variation in apparent K m values observed for etoposide at most cleavage sites (Table I)), maximal levels of scission vary dramatically at different nucleic acid sequences. This variation does not result from alterations in the rates of cleavage complex formation, because the time to attain DNA cleavage equilibrium was similar for all sites examined (see above).
Previous studies indicate that etoposide enhances topoisomerase II-mediated DNA breakage primarily by inhibiting religation of cleaved nucleic acids (37,38). This was confirmed for those drug-induced sites in the present DNA substrate that also displayed significant cleavage in the absence of drug (thus allowing direct comparisons with or without the drug). For example, the apparent first order rate of religation for site 18 of Fig. 1 was decreased Ͼ30-fold in the presence of 50 M etoposide, from ϳ1.3 min Ϫ1 in drug-free assays to ϳ0.04 min Ϫ1 .
Given the mechanistic basis of etoposide action, the wide range (ϳ60-fold; Table I) of C max values for specific sites may reflect the ability of etoposide to inhibit religation in a sequence-specific manner. Therefore, the effects of etoposide on topoisomerase II-mediated DNA religation were determined for all 22 cleavage sites that were subjected to kinetic analysis. Consistent with the wide range in C max values, apparent first order rates of religation varied Ͼ30-fold for the sites examined (Table I).
As seen in Fig. 6, sites with higher C max values generally displayed lower rates of religation. For the 16 sites with C max values Յ1.5%, a strong linear inverse correlation (r Ϸ 0.86) was observed between rates of religation and levels of maximal cleavage (Fig. 6, inset). This inverse correlation indicates that the ability of etoposide to inhibit religation at these nucleic acid sequences to a large extent dictates the levels of DNA cleavage complex formed.
For the six sites with C max values Ͼ1.5%, the linear relationship with rates of religation no longer held. Although these sites displayed some of the slowest religation rates observed, maximal levels of cleavage could not be accounted for solely on the basis of drug-induced inhibition of religation. In these cases, it is proposed that levels of noncovalent topoisomerase II⅐drug⅐DNA complexes (i.e. ternary complexes) present at these sequences are higher than those found at lesser sites. This increased concentration of enzyme would contribute to the increased maximal scission observed, allowing levels of cleavage to exceed those predicted on the basis of religation rates alone. DISCUSSION Although topoisomerase II-targeted anticancer drugs induce cell death by stimulating enzyme-mediated DNA scission (19,(22)(23)(24)(25)(26), little is understood regarding the nucleotide sequence selectivity of these agents. Therefore, to address this critical issue, a series of kinetic and binding experiments were carried out in order to determine the mechanism by which etoposide enhances cleavage complex formation at specific nucleic acid sequences. Results indicate that interactions between topoisomerase II and etoposide appear to direct cleavage complex formation, and enzyme⅐drug complexes are kinetically competent intermediates in this pathway. Furthermore, maximal levels of drug-induced scission are determined largely by the ability of etoposide to inhibit religation at specific sequences rather than the kinetic affinity of the drug for cleavage complex formation at those sites.
In some cases, levels of cleavage complexes formed at specific sites exceeded those predicted on the basis of religation rates. Thus, it is possible that higher levels of noncovalent ternary complexes are formed at these sites. Although the nucleic acid sequence elements that potentially promote increased levels of enzyme⅐drug⅐DNA complex formation are not obvious, it should be noted that the six sites that supported the highest levels of maximal cleavage were located in two small clusters. Thus, it may be that strong sites located in close proximity to one another act synergistically to promote higher levels of enzyme binding.
The kinetics of etoposide-enhanced DNA cleavage at specific sites were analyzed by Eadie-Hofstee plots. Topoisomerase IImediated DNA scission appears to be amenable to this type of analysis despite the fact that cleaved DNA is never released by the enzyme. Most likely this is because all components of the cleavage reaction are in equilibrium with one another. Consequently, the level of DNA cleavage, which is a direct measurement of cleavage complex concentration, reflects the rates of all forward reactions (i.e. binding and cleavage) minus the rates of all backward reactions (i.e. DNA religation and dissociation) and hence is a value that can be equated to the velocity of the FIG. 6. Correlation of rates of religation with maximal cleavage levels. Rates of religation were plotted versus C max values. The inset shows a least squares linear fit to the data for C max Յ1.5% (r ϭ 0.86). The data represent the averages of two to four independent experiments. FIG. 7. Pathway of etoposide-induced DNA cleavage complex formation. A pathway is shown depicting the formation of topoisomerase II-DNA cleavage complexes in the presence of etoposide (the structure of which is shown at the top). It is based on the premise 1 that drugs enhance topoisomerase II-mediated nucleic acid scission predominately by altering the structure of DNA (shown in red) within the cleavage overhang (indicated by vertical arrows). It is proposed that the primary pathway for the formation of the noncovalent enzyme⅐drug⅐DNA ternary complex is through etopo-side⅐topoisomerase II interactions (left side). Although etoposide may bind DNA prior to ternary complex formation (right side), these nonspecific interactions are proposed to be repository in nature. Finally, once the ternary complex is formed at specific sites, enhanced cleavage complex formation results from the ability of etoposide to inhibit DNA religation at those sequences (bottom). cleavage reaction at steady-state.
Recently, a model for the mechanistic basis of DNA cleavage enhancement by topoisomerase II poisons has been proposed. 1 This model, termed the positional poison model, is based on the finding that apurinic and apyrimidinic sites 1 as well as base mismatches (64) stimulate enzyme-mediated scission, but only if these DNA lesions are located within the four-base stagger that lies between the points of cleavage on the two strands of the double helix. Therefore, it was postulated that topoisomerase II-active compounds enhance scission by altering the structure of DNA within the cleavage stagger. Because lesions have a fixed location within the sequence of DNA, interactions with the enzyme are not required in order to position them within the site of cleavage.
Utilizing a photoaffinity labeled derivative of the anticancer agent amsacrine, Freudenreich and Kreuzer found that (like lesions) this drug was located at the junction of the four-base cleavage stagger during topoisomerase II-mediated DNA scission (65). However, in contrast to DNA lesions, topoisomerase II-targeted anticancer drugs are not fixed at specific locations within the DNA. Therefore, these agents must be localized to the cleavage stagger by specific interactions with the enzyme, the DNA, or both. Because many topoisomerase II-targeted drugs bind to DNA in a relatively sequence-independent fashion (Ref. 19 and references therein), it was hypothesized that interactions with the enzyme are likely required to recruit these agents to the cleavage stagger. 1 The results of the present study provide strong support for this hypothesis.
On the basis of the data presented above and the positional poison model, 1 the following pathway is proposed for the formation of etoposide-induced topoisomerase II cleavage complexes (Fig. 7). Although etoposide is capable of binding DNA (39), it is proposed that the primary pathway for the formation of the noncovalent ternary complex is through the drug⅐enzyme interactions shown at the left. Implicit in this pathway is the postulate that sites of ternary complex formation are dictated primarily by the specificity of the enzyme (rather than the drug) for those sites. If drug⅐DNA interactions play a significant role in this process (right pathway), they are largely repository in nature, serving to increase the local concentration of etoposide near the sites of action of the enzyme. Once the ternary complex is formed, the ability to shift the equilibrium from this noncovalent interaction to a covalent cleavage complex (bottom) is determined largely by the ability of etoposide to inhibit religation at the specific nucleic acid sequence to which the enzyme is bound. Thus, levels of etoposide-induced DNA scission at a specific topoisomerase II cleavage site are determined by the ternary complex concentration at that site coupled with the effect of drug on the rate of religation.
The nucleic acid specificity imposed by topoisomerase IItargeted anticancer agents may play a significant role in determining the efficacy of drug action against specific malignancies (19,26,42). The present study contributes to our understanding of the factors that govern this important aspect of drug mechanism.