Exploiting Mechanistic Differences between Drug Classes to Define ...

THEJOURNAL OF BIOLOGICAL CHEMISTRY

Vol. 268, No. 19, Issue of July 5, pp. 14394-14398, 1993 Printed in U.S.A.

0 1993 by The American Society for Biochemistry and Molecular Biology, Inc

Exploiting Mechanistic Differences between Drug Classes toDefine Functional DrugInteraction Domains on TopoisomeraseI1 EVIDENCE THAT SEVERAL DIVERSE DNA CLEAVAGE-ENHANCING AGENTSSHARE OF ACTION ON THE ENZYME*

A COMMON SITE

(Received for publication, February 1, 1993, and in revised form, March 24, 1993)

Anita H. CorbettS, Dorothy Hong, and Neil Osherofft From the Department of Biochemistry, Vanderbilt University School of Medicine, Nashville, Tennessee 37232-0146

TOfully understand the mechanism of action of topoisomerase 11-targeted agents, the effects of these drugs on the catalytic cycleof the enzyme must bewell characterized. The present study utilized a nonturnover DNA catenation assay to determine the effects of several drugs (etoposide, genistein, CP-l15,953, amsacrine, and novobiocin) on the DNA strand passage event mediatedby topoisomerase 11. With the exception of etoposide, all of the drugs inhibited the DNA strand passage step of the topoisomerase I1 catalytic cycle. A series of drug competition experiments that exploited this mechanistic difference was used to determine relationships between drug interaction domains on the enzyme. While the inclusion of etoposide in nonturnover DNA catenation assays reversed the inhibition of strand passage induced by genistein, CP-115,953, and amsacrine, it had no effect on the inhibition induced by novobiocin. These results stronglysuggest that etoposide can displace other DNA cleavage-enhancing agents from the enzyme-DNA complex. Therefore, it is concluded that the interaction domain of etoposide overlaps those of several DNA cleavage-enhancing drugs but, consistent with previous observations (Robinson, M. J., Corbett, A. H., and Osheroff, N. (1993) Biochemistry 32,3638-3643), is distinct from that of novobiocin.

(1, 2). In addition to the DNA cleavage-enhancing agents, topoisomerase I1 is also the target for coumarin-based drugs that act by disrupting interactions between the enzyme and its ATP cofactor (5-8). Although these latter compounds are not used clinically as antineoplastics, they are cytotoxic in nature and have been used as antimicrobial agents (9). In contrast to the antineoplastic drugs, the ATPase-inhibiting agents kill cells by blocking the essential physiological functions of topoisomerase 11. Despite the clinical importance of many topoisomerase IItargeted drugs, little isunderstood concerning the interactions of these compounds with the enzyme.DNA complex. Early mutagenesis studies with DNA gyrase, the prokaryotic counterpart of topoisomerase I1 (9, lo), suggested that DNA cleavage-enhancing drugs interact primarily with the enzyme subunit that contains the active site tyrosine residue ( g y r A ) (9, 11,12), while ATPase-inhibiting drugs interact primarily with the enzyme subunit that contains the consensus nucleoside triphosphate recognition sequence ( g y r B )(9). Later mutagenesis studies with both the prokaryotic and eukaryotic type I1 topoisomerases made such conclusions less obvious. Indeed, two mutations that confer resistance to DNA cleavage-enhancing drugs have been mapped to the B subunit of gyrase (9, 12, 13). Furthermore,the majority of resistance mutations identified in eukaryotic topoisomerase I1 are located in the gyrB homology region of the enzyme (14-17). Generalizations concerning the interaction domains on eukaryotic topoisomerase I1 for specific drug classes have been Topoisomerase I1 is the target for a number of structurally further confounded by the fact that many mutant enzymes diverse drugs (1, 2). Many of these compounds are highly display distinctly different drug resistance profiles. For exactive antineoplasticagents that are used for the clinical ample, while the CEM/VM-l enzyme displays resistance to treatment of human cancers (1, 2). The one common feature all classes of DNA cleavage-enhancing drugs examined to shared by all clinically relevant topoisomerase 11-targeted date (18), the HLGO/AMSA enzyme shows resistance only to antineoplastic drugs is the ability to stabilize covalent enintercalative topoisomerase 11-targetedagents (19,20).Morezyme-cleavedDNA complexes that arereaction intermediates over, the VpmR-5 enzyme displays a broad drug resistance in the catalytic cycle of the type I1 enzyme (1-3). This ability pattern but is highly sensitive to quinolone-based compounds to enhance topoisomerase 11-mediated DNA cleavage converts (21,22).l Thus,in the absence of corroborative evidence, it is the enzyme into a physiological poison (4), which in turn not clear which (if any) DNA cleavage-enhancing drugs share triggers a chain reaction that eventually leads to cell death a common interaction domain on the type I1 enzyme. In an attempt to clarify relationships between drug inter* This work was supported by National Institutes of Health Grant action domains on topoisomerase 11, a novel enzymological GM33944 and American Cancer Society Research Grant NP-812. The costs of publication of this article were defrayed in part by the approach that exploits mechanistic differences between drug payment of pagecharges. Thisarticlemustthereforebehereby classes has been developed (8).These mechanistic differences marked “advertisement” in accordance with 18 U.S.C. Section 1734 are used as the basis for a series of competition experiments solely to indicate this fact. that categorize drug interaction domains on the type I1 en$TraineeunderNationalInstitutes of HealthGrant 5 T32 zyme. While thisapproach does not define sites of drug CA09582. 5 Supported by American Cancer Society Faculty ResearchAward binding on topoisomerase 11, it can readily distinguish FRA-370. To whom correspondence should be addressed Dept. of whether two different compounds interact with the enzyme Biochemistry, 621 Light Hall, Vanderbilt University School of Medicine,Nashville, T N 37232-0146. Tel. 615-322-4338; Fax: 615-3224349.

D. M. Sullivan, M. D. Latham, M. J. Robinson, P. R. McGuirk, T. D. Gootz, and N. Osheroff, unpublished result.

14394

Functional Drug Interaction Domains Topoisomerase on

II

14395

at sites that overlap or are distinct from one another. Since mined by quantitatingthe accumulation of high molecular mass the method utilized in the present study is based on drug catenenes (which remained at thegel origin) or the loss of supercoiled plasmid DNA. function, interaction domainselucidated by this approach are Topoisomerase 11-mediated Pre- and Post-strand Passage DNA termed functional interaction domains. Cleauage-DNA cleavage assays were performed as previously deA previous mechanistic study that focused on the DNA scribed (27, 28). All samples contained 100 nM topoisomerase I1 and cleavage and ATP hydrolysis steps of the topoisomerase I1 5 nM negatively supercoiled pBR322 DNA in 20 r1 of assay buffer. catalytic cycle (8) concluded that the functional interaction Reactions that monitored the DNA cleavage/religation equilibrium prior to strand passage were carried out in the absence of domain for novobiocin (a drug classically defined as an ATP- established a nucleotide triphosphate cofactor. Reactions that monitored the ase inhibitor (5-9)) was distinct from those of several DNA equilibrium established following strand passage included 1 mM cleavage-enhancing drugs. To extend this analysis, the pres- APP(NH)P. In all cases, reaction mixtures were incubated a t 30 "C ent work determined the effects of topoisomerase 11-targeted for 6 min. Cleavage products were trapped (29,30) by the addition of agents on the DNA strand passage step of the enzyme's SDS (1% final concentration) followed by EDTA (15 mM final catalytic cycle. Results indicate that several DNA cleavage- concentration). Topoisomerase I1 was digested with proteinase K as enhancing drugs as well as novobiocin are potent inhibitors described above and final products were resolved by electrophoresis in 1% agarose gels in 40 mM Tris-acetate, pH 8.3, 2 mM EDTA. of this essential reaction step. In contrast, etoposide showed Double-stranded DNA cleavage was followed by monitoring the aclittle ability to block the DNA strand passage event. On the cumulation of linear DNA. basis of subsequentdrug competition experiments,it was DNA Intercalation-A topoisomerase I unwinding assay was emconcluded that the functional interaction domain for etopo- ployed to monitor the intercalation of amsacrine into DNA (22). side on topoisomerase I1 overlaps those of a number of other Reactions were carried out in 20 j11 of assay buffer that contained 5 nM relaxed pBR322 plasmid DNA and 10 units of topoisomerase I. DNA cleavage-enhancing agents. EXPERIMENTALPROCEDURES

Topoisomerase I1 was purified from the nuclei of Drosophila melanogaster Kc tissue culture cells by the procedure of Shelton et al. (23). Negatively supercoiled plasmid pBR322 DNA was isolated from Escherichia coli DH1 by a Triton X-100 lysis procedure followed by double banding in cesium chloride/ethidium bromide gradients (24). Etoposide was purchased from Bristol Laboratories as a sterile 20 mg/ml solution in 2 mg/ml citric acid, 30 mg/ml benzyl alcohol, 80 mg/ml polysorbate 8O/Tween 80, 650 mg/ml poly(ethy1ene glycol) 300,30.5% (v/v) ethanol. Amsacrine (NSC-249992)was the generous gift of Dr. Yves Pommier (NCI) and was dissolved as a 10 mM solution in dimethyl sulfoxide. The quinolone CP-115,953 was synthesized a t Pfizer Central Research by the procedure of Gilligan et al. (25) and was the generous gift of Dr. T. D. Gootz and Dr. P. R. McGuirk. The quinolone was dissolved as a 25 mM solution in 0.1 N NaOH and diluted to a 5 mM stock with 10 mM Tris-HC1, pH 8.0. Genistein was obtained from ICN and was dissolved as a 10 mM solution in dimethyl sulfoxide. Novobiocin was obtained from Sigma and was dissolved as a 10 mM solution in water. SDS and proteinase K were from E. Merck Biochemicals; Tris, ethidium bromide, and APP(NH)P*were obtained from Sigma; histone H1 was from Boehringer Mannheim; and calf thymus topoisomerase I was from Bethesda Research Laboratories. All other chemicals were analytical reagent grade. Nonturnouer Topoisomerase II-mediated DNA Catenation-Reaction mixtures contained 100 nM topoisomerase 11 and 5 nM negatively supercoiled pBR322 DNA in 20 pl of assay buffer (10 mM Tris-HC1, pH 7.9, 50 mM NaC1, 50 mM KCl, 5 m M MgC12, 0.1 mM EDTA, and 2.5% glycerol) that contained 6rg/ml histone H1 (aDNA condensing agent) and 1 mM APP(NH)P (26). Samples were incubated at room temperature for various times up to 20 s, and reactions were stopped by the addition of EDTA (25 mM final concentration) followed by SDS (1%final concentration). Topoisomerase I1 and histones were digested by incubation with proteinase K (60 pg/ml final concentration) for 45 min at 45 "C. Final products were mixed with 2.5 rcl of loading buffer (60% sucrose, 0.05% bromphenol blue, 0.05% xylene cyano1 FF, and 10 mM Tris-HC1, pH 7.9) and subjected to electrophoresis in 1%agarose gels in 100 mM Tris-borate, pH 8.3, 2 mM EDTA. Following electrophoresis, gels were stained in an aqueous solution (1 pg/ml) of ethidium bromide. DNA bands were visualized by transillumination with ultraviolet light (300 nm) and photographed through Kodak 23A and 12 filters with Polaroid type 665 positive/negative film. DNA bands were quantitated by scanning negatives with an E-CApparatus modelEC910 densitometer in conjunction with Hoefer GS-370 Data System software. Under the conditions employed, the intensity of bandsin the negative was proportional to the amount of DNA present. Control assays always contained an amount of drug diluent equivalent to that present in drug-containing reactions. Levels of DNA catenation were deter-

* The abbreviations used are: APP(NH)P, adenyl-5"yl P,y-imidodiphosphate; ECw, effective drug concentration required to inhibit DNA strand passage 50%.

Following a 15-min incubation at 37 "C, samples were extracted with phenol/chloroform and subjected to agarose gel electrophoresis in Tris-acetate buffer as described above. RESULTS

Effects of Topoisomerase 11-targeted Drugs on the DNA Strand Passage Step of the Enzyme's Catalytic Cycle-A previous study found that novobiocin as well as a number of DNA cleavage-enhancing agents (including genistein, amsacrine, CP-115,953, quercetin, andquercitrin) inhibited the ATP hydrolysis activity of topoisomerase I1 (8). In order to extend this mechanistic analysis of drug action, the effects of several compounds on the DNA strand passage event of the enzyme were examined. This reaction step was chosen as the focus of the present work since it is the catalytic event that is triggered by the initial binding interaction of topoisomerase I1 with its ATP cofactor (3). A nonturnover DNA catenation assay was utilized to monitor the DNA strand passage step of the topoisomerase 11 catalytic cycle (26). (This is the reaction step in which a DNA helix is translocated through the double-stranded break made in asecond helix (3)).The nonturnover DNA catenation assay employs stoichiometric levels of the type I1 enzyme (relative to molecules of plasmid DNA) in conjunction with APP(NH)P, a nonhydrolyzable ATP analog that induces DNA strand passage but does not allow enzyme turnover (7, 27, 28). Thus, each molecule of topoisomerase I1 is capable of catalyzing only a single DNA strand passage event. This assay monitors all of the reaction steps catalyzed by the enzyme up to and including DNA strand passage (3, 26). However, since none of the drugs utilized in the present study impairs DNA cleavage (the reaction step that immediately precedes strand passage (3)) mediated by the Drosophila enzyme (8, 22, 31, 321, decreases in reaction rates can be attributed primarily to effects on the DNA strand passage event. Finally, DNA catenation(ratherthan relaxation) wasused to monitor strand passage, since a single round of catalysis dramatically alters the electrophoretic mobility of supercoiled DNA (Fig. 1, inset). The effects of novobiocin on the topoisomerase 11-mediated DNA strand passage event areshown in Fig. 1. As determined by a series of nonturnover DNA catenation time courses carried out over a range of drug concentrations, novobiocin appears to be a potent inhibitorof the enzyme-mediated DNA strand passage event (left panel).To confirm this conclusion, the ability of novobiocin to block the transitionfrom the preto the post-strand passage topoisomerase 11-DNA cleavage

Functional Drug Interaction Domains Topoisomerase on

14396

Time (s)

+

11

0

+

0

Novo APP

500

1

[Drug1 (VM)

FIG.1. Effects of novobiocin on the topoisomerase 11-mediated DNA strand passage event.A series of nonturnover DNA catenation time courses for the inhibition of DNA strand passage by novobiocin is shown in the left panel. Experiments were carried out or in the presence of 20 p M (A),50 p M (O), in the absence of drug (0) 100 p M (e),200 pM (A), or 2-50p M (m) novobiocin. Results are typical of at least two independent experiments. Theinset shows an agarose gel of a 20-s nonturnover DNA catenation assay. Lane I , DNA standard: lane 2, no drug: lane 3, 250 p~ novobiocin. The positions of supercoiled ( I ) , nicked circular (11).and catenated ( c a t ) DNA are indicated. The effect of novobiocin on the transition from pre- to post-strand passage DNA cleavage is shown in the right panel. Relative levels of DNA cleavage are indicated by the solid columns. The level of pre-strand passage DNA cleavage obtained in the absence of drug was setto 1.0. Assays were carried out in theabsence of novobiocin and APP(NH)P ( N o Drug); in the presence of 250 p t ~ novobiocin (Novo):in the presence of 1 mM APP(NH)P ( A P P ) ;in the presence of250 p~ novobiocin following a 10-s preincubation with 1 mM APP(NH)P ( A P P + Novo); or in the presence of 1 mM APP(NH)P following a 10-s preincubation with 2.50 p~ novobiocin (Novo + APP). Data represent the averages of four independent experiments. Standard deviations are indicated bv the error bars.

250

FIG.2. Effects of topoisomerase 11-targeted drugs on the enzyme-mediated DNA strand passage event. Drug titrations are shown tor novobiocin (O),genistein ( O ) ,CP-llR,953 (A),amsacrine ( 0 , and etoposide (m). The amount of DNA strand passage observed in the absence of drug was set to 100%. Results are typical of at least two independent 20-s nonturnover DNA catenation assays.

-

100

8

a,

a

0

250

50(

[Etoposide] (pM) complex was determined (rightpanel). Thistransitionis straightforward tofollow since the stability of the post-strand FIG.3. Effects of etoposide on novobiocin- and genisteinpassage complex (and hence the level of DNA cleavage that induced inhibition of DNA strand passage. Reaction mixtures and the occurs following DNA strand passage) is -5-fold higher than contained 250 p~ novobiocin (0)or 250 p~ genistein (0) that of its pre-strand passage counterpart (27, 28). When indicated concentrations of etoposide. Results are plotted relative to novobiocin was added to assay mixtures following a preincu- the level of DNA strand passage observed in the absenceof drug (set to 1005) and aretypical of three independent 20-s nonturnover DNA bation with APP(NH)P, little effect was observed. However, catenation a s s a y . when thedrug was addedpriortotheATP analog, the transition from pre-topost-strand passage was virtually ing Drugs but Is Distinct from That of Novobiocin-A series abolished. Taken together, the data in Fig. 1 provide strong of competition experiments was carried out to define funcevidence that novobiocin blocks the DNA strand passage tional drug interaction domains on topoisomerase 11. Since event of topoisomerase 11. etoposide had onlya minimal effect onthe DNA strand The effects of several topoisomerase 11-targeted drugs on passage event of the enzyme compared to those of the other the enzyme-mediated DNA strand passage event are shown compounds examined, its ability to block the inhibitorypropin Fig. 2. With the exception of novobiocin, all of the drugs erties of the other drugs was determined. If etoposide shares employed are classically defined as DNA cleavage-enhancing a common interaction domain with other drugs, high etopoagents (1, 2). Genistein was approximatelyequipotentto side concentrations shoulddisplace theseagents from the novobiocin at inhibiting DNA strand passage (EGO= 40 pM enzyme. DNA complex and consequentlyshouldalleviate drug), while the quinolone CP-115,953 and amsacrine were drug-induced inhibition of the DNA strand passage event. intermediate in their potency (E& =: 400 to 500 p M ) . In Conversely, if the interaction domainfor etoposide is distinct contrast, etoposide displayed a minimal ability to impair the from those of other agents, it should be unable to reverse the DNA strand passage event of topoisomerase 11. Even at a drug-induced inhibition. concentration of 500 p~ etoposide, less than 15% inhibition Results of drug competition experiments are shown in Figs. was observed. 3 and 4. In all cases, etoposide and other drugs were added The above findings correlate with the previously reported simultaneously toassay mixtures. Even at a concentration of effects of drugs on the ATPase activity of the enzyme (8). 500 p ~ etoposide , showed no ability to reverse the inhibition Therefore, it is likely that drug-induced inhibition of ATP of DNA strand passage induced by 250 P M novobiocin. This hydrolysis results from alterations in the initial interaction result is consistent with the previous findings that etoposide of topoisomerase I1 with its high energy cofactor rather than was unable toreverse the inhibitionof enzyme-mediated ATP from a decrease in the catalytic rateof ATP turnover. hydrolysis by novobiocin and that novobiocin was unable to T h e Functional Interaction Domain for Etoposide on Topo- block the etoposide-promoted enhancement of DNA cleavage isomerase I I Overlaps Those of Other D N A Cleavage-enhanc- by topoisomerase I1 (8).Thus, the present data confirm the

Functional Drug Interaction Domains Topoisomerase on

11

14397

conclusion that the functional interaction domain for etopo- sidedoes notinterferewithamsacrine.DNA binding and side on the enzyme is distinct from that of novobiocin (8). makes itlikely that the resultsof DNA strand passage experIn marked contrast to the resultsnovobiocin, with etoposide iments reflecta competition for drug binding sites on the dramatically reversed inhibition of the DNA strand passage enzyme rather than onDNA. Taken together with the results event induced by 250 p~ genistein (Figs. 3 and 4). Further- of DNA strand passage assays, it is concluded that etoposide more, etoposide also reversed inhibition of the DNA strand shares a common interaction domain on topoisomerase I1 passage event observed in the presenceof 500 pM CP-115,953 with the other DNA cleavage-enhancing agents examined. or amsacrine (Fig. 4). These findings strongly suggest that DISCUSSION etoposide can efficientlydisplace genistein, CP-115,953, or amsacrine from the topoisomerase11. DNA complex. Numerous studieshave demonstrated that drugs inhibit the Since several of the above drugs havebeen shown to interact catalytic DNA strand passage activity of topoisomerase I1 (1, with nucleicacids (33, 34),it is not known whetherthe 2). However, due to the fact that all of the assays employed reversal of inhibition by etoposide is due toa competition for require ATP (which allows enzyme turnover), it was imposdrug binding sites on the enzyme or on itsDNA substrate. To sible to ascribe the effects of drugs on overall catalytic activity examine these possibilities, the ability of etoposide (which is to any specific step of the enzyme's catalytic cycle. For exnonintercalative in nature (33)) todisplace amsacrine (which ample, while inhibition by any given agent could have resulted is intercalative in nature (34)) from DNA was determined. A from an inhibition of the DNA strand passage event, it also topoisomerase I DNA unwinding assaywas employed forthis could haveresulted from adecreased rate of post-strand experiment (22) (Fig. 5). In the presence of the intercalative passage DNA religation (27), ATP hydrolysis (8), or enzyme drug, anet negative supercoilingof relaxed DNAwas observed turnover. Therefore, a nonturnover DNA catenation assay following treatment with the type I enzyme (lane 3 ) . Con- (which replaces ATP with APP(NH)P) was employed to versely, as expected, no supercoiling was observed in the specifically examine the effects of topoisomerase 11-targeted presence of etoposide (lune 4 ) . As seen in lane 5, a %fold agentsonthe DNA strand passage step of the enzyme's molar excess of etoposide (compared to the 1:l drug ratio catalytic cycle. employed in DNAstrand passage competition assays)did not The agents employed in this work are representative of a decrease the supercoiling observed in the presence of topoi- broad spectrum of mechanistically and structurally diverse somerase I and amsacrine. This finding indicates that etopo- drug classes including coumarins (novobiocin), demethylepipodophyllotoxins (etoposide), isoflavones (genistein), quinolones (CP-115,953), and anilinoacridines (amsacrine).Of the drugs examined, all but etoposide inhibited the critical enzyme-mediated DNA strand passage ( i e . helix translocation) event. A similar result was found previously for the effects of drugs on the ATP hydrolysis reaction of topoisomerase I1 (8). Thus, thesetwo studies demonstrate that many antineoplastic agents affectmore than just the DNAcleavage/religation equilibria of the enzyme. How mechanistic differencesbetween drug classes ultimately impact the physiological actions of these agents has yet to bedetermined. However, it was possible to exploit " No Drug Novo Gen AMSA 9 5 3 mechanistic differencesbetweenetoposide and other comFIG.4. Summary of the effects of etoposide on the druginduced inhibition of enzyme-mediated DNA strand passage. pounds in order todefine functional drug interaction domains The graph compares the amountof DNA strand passage observed in on topoisomerase 11. Results of a series of drug competition the presence of the indicated drug (stippledcolumns) orinthe experiments provide the first biochemical evidence that the presence of the indicated drug plus500 y~ etoposide (solid columns). site of action of etoposide on the enzyme overlaps those of a Results are shown for reactions that contained no drug( N o Drug) or number of other DNA cleavage-enhancing agents. Furtherthat contained 250 y M novobiocin (Novo), 250 y M genistein (Gen), 500 y M amsacrine ( A M S A ) ,or 500 pM CP-115,953 (953). The level more, the presentwork confirms theprevious conclusion that topoisomerase I1 for of DNA strand passage obtained in the absence of drug was set to the functional interaction domain on 100%. Results are the averages of three independent20-s nonturnover etoposide is distinct from that of novobiocin (8). DNA catenation assays. Standard deviations are indicated by the A number of mutagenesis studies have identified a region error bars. in thegyrB homology domain of topoisomerase I1 that appears to be important for the interaction of the enzyme witha variety of DNA cleavage-enhancing drugs (14-17). However, the use of resistance-conferring mutations to determinerelationships between drug binding sites in the enzyme.DNA complex has been complicated by the fact thatmany mutant type I1 topoisomerases display different and often contradictory profiles of drug resistance (18-22).' While the enzymological technique employed above does not identify residues that are involved in topoisomerase I1 .drug binding, it does 1 2 3 4 5 providea novel method for definingrelationships between FIG. 5. Effect of etoposide on amsacrine.DNA binding. An drug interaction domains on theenzyme. Since the strengths agarose gel is shown. Lane 1, supercoiledDNA standard lane 2, of the mutagenesis and enzymological techniques arecomplerelaxed DNAstandard; lanes3-5,DNA incubated with topoisomerase mentary, when used inconjunction, they should create a I in the presence of 250 PM amsacrine, 500 y~ etoposide, or both 250 powerful approach to refining our understanding of how anyM amsacrine and 500 PM etoposide, respectively. The positions of supercoiled (FI)and nicked circular ( F I I )DNA are indicated. Results tineoplastic agents convert topoisomerase I1 into a cellular are typical of three independent experiments. poison.

14398

Functional Drug Interaction Domains

Acknowledgments-We are grateful to Dr. P. R. McGuirk and Dr. T. D. Gootz (Pfizer Central Research) for providing CP-115,953, to Y. Pommier (NCI) for providing amsacrine, to C. Brewer for assistance in preparing topoisomerase I1 and plasmid DNA, to J. Rule for expert photography, and toS. Heaver for the conscientious preparation of the manuscript. REFERENCES 1. Liu, L. F. (1989) Annu. Reu. Biochem. 58,351-375

2. Schneider, E.,Hsiang, Y.-H., and Liu, L. F. (1990) Ado. Pharmacol. 2 1 , 149-183 3. Osheroff, N., Zechiedrich, E. L., and Gale, K. C. (1991) BioEssays 1 3 , 269275 4. Kreuzer, K. N., and Cozzarelli, N. R.(1979) J. Bacteriol. 140,424-435 5. Gellert, M.,, Mizuuchi, K., O'Dea, M. H., and Nash,H. A. (1976) Proc. Natl. Acad. Scr. U. S. A. 73,3872-3876 6. Sugino, A., Higgins, N. P., Brown, P. O., Peebles, C. L., and Cozzarelli, N. R. (1978) Proc. Natl. Acad. Sci. U. S. A. 7 5 , 4838-4&12 7. Osheroff, N., Shelton, E. R., and Brutlag, D. L. (1983) J. Biol. Chem. 2 5 8 , 9536-9543 8. Robinson, M. J., Corbett, A. H., and Osheroff, N. (1993) Biochemistry 3 2 , 3638-3643 9. Reece, R. J., and Maxwell, A. (1991) CRC Crit. Reu. Biochem. Mol. Biol. 26,335-375 10. Caron, P. R., and Wang, J. C. (1993) Molecular Biology of DNA Topoisomermes and Its Application to Chemotherapy (Andoh, T., Ikeda, H., and Oguro, M., eds) pp. 1-18, CRC Press, Inc., Boca Raton, FL 11. Yoshida, H., Bogaki, M., Nakamura, M., Yamanaka, L. M., and Nakamura, S. (1991) Antimicrob. Agents Chemother. 3 5 , 1647-1650 12. Maxwell, A. (1992) J. Antimicrob. Chemother. 30,409-414 13. Yamagishi, J., Yoshida, H., Yamayoshi, M., and Nakamura, S. (1986) Mol. & Gen. Genet. 204,367-373 14. Bugg, B. Y., Danks, M. K., Beck, W. T., and Suttle, D. P. (1991) Proc. Natl. Acad. Sci. U. S. A. 88,7654-7658

on Topoisomerase I1 15. Hinds, M., Deisseroth, K., Mayes, J., Altschuler E. Jansen, R., Ledley, F. D., and Zwelling, L. A. (1991) Cancer Res. 51: 47'29-4731 16. Chan, V. T.W., Ng, S., Eder, J. P., and Schnipper, L. E. (1993) J. Biol. ChPm. "_, 26R. 21tW-01fi.5 ""_ 17. Lee, M.-S., Wang, J. C., and Beran, M. (1992) J . Mol. Biol. 2 2 3 , 837-843 18. Danks, M. K., Schmidt, C. A., Cirtain, M. C., Suttle, P. D., and Beck, W. T. (1988) Biochemistm 27. RR61-8869 ~ ~ . .. ~ . ~ 19. Zwelling, L. A., Hinds,