Benzo[c]phenanthrene adducts and nogalamycin inhibit DNA transesterification by vaccinia topoisomerase.

Vaccinia DNA topoisomerase forms a covalent DNA-(3'-phosphotyrosyl)-enzyme intermediate at a specific target site 5'-C(+5)C(+4)C(+3)T(+2)T(+1)p downward arrow N(-1) in duplex DNA. Here we study the effects of position-specific DNA intercalators on the rate and extent of single-turnover DNA transesterification. Chiral C-1 R and S trans-opened 3,4-diol 1,2-epoxide adducts of benzo[c]phenanthrene (BcPh) were introduced at single N2-deoxyguanosine and N6-deoxyadenosine positions within the 3'-G(+5)G(+4)G(+3)A(+2)A(+1)T(-1)A(-2) sequence of the nonscissile DNA strand. Transesterification was unaffected by BcPh intercalation between the +6 and +5 base pairs, slowed 4-fold by intercalation between the +5 and +4 base pairs, and virtually abolished by BcPh intercalation between the +4 and +3 base pairs and the +3 and +2 base pairs. Intercalation between the +2 and +1 base pairs by the +2R BcPh dA adduct abolished transesterification, whereas the overlapping +1S BcPh dA adduct slowed the rate of transesterification by a factor of 2700, with little effect upon the extent of the reaction. Intercalation at the scissile phosphodiester (between the +1 and -1 base pairs) slowed transesterification by a factor of 450. BcPh intercalation between the -1 and -2 base pairs slowed cleavage by two orders of magnitude, but intercalation between the -2 and -3 base pairs had little effect. The anthracycline drug nogalamycin, a non-covalent intercalator with preference for 5'-TG dinucleotides, inhibited the single-turnover DNA cleavage reaction of vaccinia topoisomerase with an IC50 of 0.7 microM. Nogalamycin was most effective when the drug was pre-incubated with DNA and when the cleavage target site was 5'-CCCTT/G instead of 5'-CCCTT/A. These findings demarcate upstream and downstream boundaries of the functional interface of vaccinia topoisomerase with its DNA target site.

Poxvirus DNA topoisomerase I is important for virus replication (1) and a potential target for drug therapy of smallpox, in light of its unique DNA recognition specificity, compact structure, and distinctive pharmacological sensitivities compared with human topoisomerase I (2)(3)(4)(5)(6). Poxvirus topoisomerases are exemplary type IB family members; they cleave and rejoin one strand of the DNA duplex through a transient DNA-(3Ј-phosphotyrosyl)-enzyme intermediate. Vaccinia topoisomerase cleaves duplex DNA at a pentapyrimidine target sequence, 5Ј-(T/C)CCTTp2 (3). (The Tp2 nucleotide is defined as the ϩ1 nucleotide.) Topoisomerases encoded by other genera of poxviruses recognize the same DNA target sequence (6 -10). Available structural and biochemical studies suggest that the assembly of a catalytically competent topoisomerase active site is triggered by recognition of the DNA target (11,12).
Early studies using nuclease footprinting, modification interference, modification protection, analog substitution, and UV crosslinking techniques suggested that vaccinia topoisomerase makes contact with several nucleotide bases and the sugar-phosphate backbone of DNA within and immediately flanking the CCCTT element (13)(14)(15)(16)(17)(18)(19). Recent studies have focused on delineating the features of the DNA interface that affect the kinetics of transesterification. Modifications at the scissile phosphodiester have illuminated the chemical mechanism of topoisomerase IB, the roles of the individual amino acids in either transition-state stabilization or general acid catalysis, and the parameters affecting the stability of the covalent topoisomerase-DNA intermediate (20 -24). Positionspecific interference by modifications at remote phosphates on the scissile and nonscissile strands has provided an atomicresolution map of the DNA backbone contacts required for active site assembly (12).
Another approach taken to define the interface between vaccinia topoisomerase and its cleavage site exploits position-specific covalent polycyclic aromatic hydrocarbon (PAH) 1 diol epoxide-DNA adducts (25,26). For example, to probe the DNA minor groove surface, 7,8-diol 9,10-epoxide adducts of benzo-[a]pyrene (BP) were introduced at the exocyclic N 2 -amino group of single deoxyguanosine (dG) positions within the nonscissile 3Ј-G ϩ5 G ϩ4 G ϩ3 A ϩ2 A ϩ1 N Ϫ1 N Ϫ2 strand of a suicide cleavage substrate for vaccinia topoisomerase (25). The transopened BP dG adducts fit into the minor groove without perturbing helix conformation or base pairing, and the C-10 R and S diastereomers are oriented in opposite directions within the minor groove (27,28). A sharp margin of interference effects was observed, whereby ϩ5 and Ϫ2 BP dG modifications were well tolerated, but ϩ4, ϩ3, and Ϫ1 BP dG adducts were severely deleterious. The stereoselective effects at the Ϫ1 nucleoside (the R diastereomer interfered, whereas the S diastereomer did not) delineated at high resolution the downstream border of the minor groove interface (25). It was inferred that BP dG inhibition of transesterification is likely caused by steric exclusion of essential constituents of the topoisomerase from the DNA minor groove.
PAH diol epoxide-DNA adducts have also been exploited to probe the effects of position-specific intercalation, by introducing trans-opened 7,8-diol 9,10-epoxide adducts of BP at the exocyclic N 6 -amino group of deoxyadenosine (dA) positions within the nonscissile strand (26). The R and S BP dA adducts intercalate from the major groove on the 5Ј and 3Ј sides of the modified base, respectively, and perturb local base stacking (29 -33). R and S BP dA modifications at ϩ1A reduced the transesterification rate by a factor of 700 to 1000 without affecting the yield of the covalent topoisomerase-DNA complex. BP dA modifications at ϩ2A reduced the extent of transesterification and elicited rate decrements of 200-fold and 7000-fold for the S and R diastereomers, respectively (26). In contrast, BP dA adducts at the Ϫ2 position had no effect on the extent of the reaction and relatively little impact on the rate of cleavage. The BP dA interference effects demarcated the Ϫ1 base pair as the "downstream" margin of the functional interface between DNA and vaccinia topoisomerase that can be affected significantly by BP intercalation. The "upstream" margin remained undefined because the effects of intercalating PAH adducts at the guanine positions of the cleavage target site were not tested.
Here we extend the use of defined PAH diol epoxide-DNA adducts to probe the effects of position-specific intercalation by benzo[c]phenanthrene (BcPh) at all purines of the topoisomerase target sequence. BcPh exemplifies the sterically hindered, nonplanar "fjord-region" class of PAHs. BcPh has been studied extensively in light of the highly tumorigenic and mutagenic properties of metabolically activated BcPh diol epoxides, which react at the benzylic C1 position by trans addition of guanine N 2 and adenine N 6 in DNA to form the covalent trans 1S and 1R BcPh dG and BcPh dA adducts depicted in Fig. 1. Structures of duplex DNAs containing single BcPh dG adducts have established that the aromatic ring systems intercalate from the minor groove on opposite sides of the modified dG base depending on their stereochemical configuration (34). The S diastereomer intercalates on the 5Ј side of the dG, whereas the R diastereomer intercalates on the 3Ј side (Fig. 2). Structures of DNAs containing single BcPh dA adducts show that the hydrocarbon intercalates from the major groove, such that the S diastereomer is on the 3Ј side of the modified dA base and the R diastereomer is on the 5Ј side ( Fig. 2; Refs. 35 and 36). The intercalated BcPh dG and BcPh dA adducts do not disrupt base pairing, but they do cause a buckling of the modified base pair and the unmodified base pair flanking the hydrocarbon (Fig. 2). An attractive feature of BcPh adducts with respect to studies of vaccinia topoisomerase is that they provide structural probes for intercalative interference effects at all of the purine bases of the nonscissile strand of the 5Ј-CCCTT/3Ј-GGGAA target site and for effects of intercalation at positions immediately 3Ј of the scissile phosphodiester.
In light of these results, we tested the effects of the anthracycline drug nogalamycin, which intercalates preferentially at 5Ј-TG dinucleotides in duplex DNA (37)(38)(39). We find that nogalamycin inhibits the single-turnover DNA cleavage reaction of vaccinia topoisomerase and that its potency is higher when the cleavage target site is 5Ј-CCCTT2G versus 5Ј-CCCTT2A. We surmise that nogalamycin can inhibit the vaccinia enzyme by intercalating between the ϩ1 and Ϫ1 base pairs.

EXPERIMENTAL PROCEDURES
Synthesis of BcPh Diol Epoxide-adducted Oligonucleotides-Oligonucleotide 18-mers containing trans-opened S and R BcPh diol epoxide adducts ( Fig. 1) were synthesized from appropriately protected 5Ј-Odimethoxytrityl-3Ј-phosphoramidites (BcPh dA and BcPh dG phosphoramidites) as described (40). Oligonucleotides containing BcPh dG were synthesized using purified S and R diastereomers of the BcPh dG phosphoramidites (41). BcPh dA adducts were incorporated as their mixed S/R diastereomers (42). The resulting S and R BcPh dA-adducted oligonucleotides were purified and resolved by HPLC. Absolute configurations of the BcPh dA adducts were assigned to the separated oligonucleotides after enzymatic hydrolysis to form the nucleoside adducts, which were identified by their CD spectra (43,44). Details of the high performance liquid chromatography purification of the adducted oligonucleotides are provided in Supplemental Table S1. The compositions of the oligonucleotides were confirmed by matrix-assisted laser desorption ionization-mass spectrometry analysis.
DNA Cleavage Substrates-The CCCTT-containing scissile strands were 5Ј 32 P-labeled by enzymatic phosphorylation in the presence of [␥-32 P]ATP and T4 polynucleotide kinase. The labeled oligonucleotides were gel-purified and hybridized to standard or modified nonscissile strand oligonucleotides at a 1:4 molar ratio of scissile:nonscissile strand. Annealing reaction mixtures containing 0.2 M NaCl and oligonucleotides as specified were heated to 80°C and then slow-cooled to 22°C. The hybridized DNAs were stored at 4°C. The structures of the annealed duplexes and the sequences of the component strands are depicted in the figures.
Vaccinia Topoisomerase-Recombinant vaccinia topoisomerase was produced in Escherichia coli BL21 by infection with bacteriophage CE6 (2) and then purified to apparent homogeneity from the soluble bacterial lysate by phosphocellulose and Source S-15 chromatography steps. Protein concentration was determined by using the dye-binding method (Bio-Rad) with bovine serum albumin as the standard.
Drugs-Nogalamycin was obtained from the Open Chemical Repository, Developmental Therapeutics Program, National Cancer Institute. Nogamycin was obtained from Amersham Biosciences, courtesy of Dr. Paul Aristoff. The drugs were dissolved in Me 2 SO at a concentration of 10 mM. The stock solutions were stored at Ϫ20°C and diluted freshly in Me 2 SO for each experiment to attain the desired concentrations.
Single-turnover DNA Cleavage-Reaction mixtures containing (per 20 l) 50 mM Tris-HCl (pH 7.5), 0.3 pmol of CCCTT-containing DNA and 75 or 150 ng (2 or 4 pmol) of vaccinia topoisomerase were incubated at 37°C. Aliquots (20 l) were withdrawn at the times specified and quenched immediately with SDS (1% final concentration). The products were analyzed by electrophoresis through a 10% polyacrylamide gel containing 0.1% SDS. Free DNA migrated near the dye front. Covalent complex formation was revealed by transfer of radiolabeled DNA to the topoisomerase. The extent of covalent complex formation was quantified by scanning the dried gel using a Fujifilm BAS2500 imager. A plot of the percentage of input DNA cleaved versus time established the end point values for cleavage. The data were then normalized to the end point values (defined as 100%), and the cleavage rate constants (k cl ) were calculated by fitting the normalized data to the equation, 100 Ϫ %cleavage (norm) ϭ 100e Ϫkt .
Single-turnover Religation by the Suicide Intermediate-Cleavage reaction mixtures (20 l) containing 0.3 pmol of the 18-mer/30-mer DNA (5Ј 32 P-labeled on the 18-mer scissile strand and containing a G:C base pair at position Ϫ1) and 2 pmol of topoisomerase were incubated at 37°C for 30 s to form the suicide intermediate. The reaction mixtures were adjusted to 10% Me 2 SO and 0, 5, 10, or 20 M nogalamycin and incubated for 5 min at 37°C. Religation was initiated by the simultaneous addition of NaCl to 0.5 M and a 5Ј hydroxyl-terminated 18-mer acceptor strand d(GTTCCGATAGTGACTACA) to a concentration of 15 pmol/22 l (i.e. a 50-fold molar excess over the input DNA substrate). The reaction was quenched after 15 s by adding an equal volume of buffer containing 2% SDS, 76% formamide, and 20 mM EDTA. The samples were heat-denatured and then analyzed by electrophoresis through a 17% polyacrylamide gel containing 7 M urea in TBE (90 mM Tris/64.6 mM boric acid/2.5 mM EDTA, pH 8.3). Religation of the covalently bound 12-mer strand to the 18-mer acceptor DNA will yield a 5Ј 32 P-labeled 30-mer strand transfer product. The extent of religation is expressed as the percent of the input labeled 18-mer strand recovered as 30-mer.

Position-specific BcPh Interference Effects on DNA Transesterification-Oligodeoxynucleotide 18-mers containing a single
were synthesized and then annealed to a 5Ј 32 P-labeled 34-mer scissile strand to form suicide cleavage substrates for vaccinia topoisomerase (Fig. 3). The cleavage transesterification reaction results in covalent attachment of the 32 P-labeled 12-mer 5Ј-pCGTGTCGCCCTTp to the enzyme via Tyr-274. The unlabeled 22-mer 5Ј-OH-leaving strand dissociates spontaneously from the protein-DNA complex. Loss of the leaving strand drives the reaction toward the covalent state, so that the reaction can be treated kinetically as a first-order unidirectional process. The reaction of excess topoisomerase with the unmodified control substrate attained an end point at which 90% of the DNA was converted to covalent topoisomerase-DNA complex; the reaction was complete within 20 s. The extent of transesterification after 5 s was 55% of the end point value. From this datum, we calculated a single-turnover cleavage rate constant (k cl ) of 0.27 s Ϫ1 (Fig. 3).
We found that the S and R BcPh modifications at position ϩ5G and the R BcPh modification at ϩ4G slowed the transesterification rate to 0.13, 0.07, and 0.07 s Ϫ1 , respectively, without significantly affecting the yield of covalently bound DNA at the reaction end point (61-65%). The k obs for cleavage of these BcPh dG-modified substrates did not increase when the concentration of topoisomerase in the reaction mixture was varied over a 4-fold range (not shown), indicating that the modestly slowed cleavage rates (2-to 4-fold compared with the unmodified control DNA) were not caused by slow noncovalent binding of topoisomerase to these substrates.
In contrast, the S BcPh adduct at ϩ4G, the R and S BcPh adducts at ϩ3G, and the S and R adducts at ϩ2A virtually abolished the transesterification reaction. The extents of DNA cleavage after a 24-h reaction in enzyme excess were in the range of 1-4% of the input suicide substrate (Fig. 3). Although the accumulation of the topoisomerase-DNA complex on these modified DNAs was time-dependent, the reaction clearly did not attain a useful end point for the purpose of calculating a cleavage rate constant. The end point did not increase when the concentration of topoisomerase was doubled, implying that the reaction was not limited by the noncovalent binding step. Rather, we surmise that the majority of the topoisomerase binding events on the ϩ4S, ϩ3R, and ϩ3S BcPh dG and ϩ2S and ϩ2R BcPh dA substrates were nonproductive with respect to transesterification, and that there was not a free equilibrium between productive and nonproductive binding modes (at least not within the 24-h period that the reactions were monitored).
The S and R BcPh dA modifications at ϩ1A reduced k cl to 0.0001 and 0.0006 s Ϫ1 , respectively, without significantly affecting the end point (67-71% cleavage). Thus, ϩ1 BcPh dA adducts elicited rate decrements of 2700-fold and 450-fold for the S and R diastereomers, respectively. Note that the interference effect was greater when the ϩ1 BcPh dA adduct was intercalated on the 3Ј side of the modified adenine base facing away from the scissile phosphodiester (Fig. 3). The BcPh dA adducts at the Ϫ2 position also displayed a strong orientation bias with respect to rate effects. The S diastereomer, which faces toward the scissile phosphodiester, reduced the cleavage rate constant to 0.003 s Ϫ1 , a 90-fold decrement compared with the unmodified control DNA, whereas the R diastereomer at Ϫ2 had little impact on the rate (0.12 s Ϫ1 ). Neither of the Ϫ2 BcPh dA adducts affected the cleavage reaction end point.
To address whether BcPh substitutions altered the site of cleavage within the 34-mer scissile strand, the products of the cleavage reactions with unmodified DNA, and the ϩ5R, ϩ5S, ϩ4R, ϩ1R, Ϫ2S, and Ϫ2R BcPh diol epoxide-modified DNAs were digested with proteinase K in the presence of SDS to remove the covalently linked topoisomerase, and the radiolabeled DNA reaction products were then analyzed by denaturing polyacrylamide gel electrophoresis. Reaction of topoisomerase with the unmodified control substrate results in the appearance of a cluster of radiolabeled species migrating faster than the input 32 Plabeled 34-mer strand, which consists of the 12-mer 5Ј-pCGT-GTCGCCCTTp linked to one or more amino acids of the topoisomerase. The same cluster was produced by proteinase K digestion of the covalent complex formed by reaction of topoisomerase with the BcPh dA and BcPh dG substrates (data not shown). Thus, the site of covalent complex formation was unchanged by the BcPh modifications. Any shift in the cleavage site, and hence the size of the covalently bound oligonucleotide, would have been readily detected by an altered mobility of the array of labeled oligonucleotide-peptide complexes.
Effects of Nogalamycin on DNA Transesterification by Vaccinia Topoisomerase-The exquisite sensitivity of vaccinia topoisomerase to position-specific intercalators has implications for the discovery of poxvirus-specific topoisomerase inhibitors and/or poisons as candidate antipoxviral drugs. DNA intercalating agents have been widely studied as inhibitors of other DNA topoisomerases; indeed, intercalating drugs that target topoisomerase II are mainstays of cancer chemotherapy. Existing compounds that display some activity against microbial or eukaryotic cellular topoisomerases are a reasonable starting point for screening for inhibition or poisoning of the poxvirus topoisomerase.
The anthracycline drug nogalamycin intercalates in duplex DNA with sequence selectivity for 5Ј pyrimidine-purine dinucleotides 5Ј-TG and 5Ј-CG (45). Nogalamycin consists of a core 4-ring chromophore with a nogalose sugar and a methyl ester attached to the A ring and a bicyclic aminoglucose sugar fused to the D ring (Fig. 4). X-ray and NMR structures of drug-DNA complexes have shown that the dumbbell-shaped nogalamycin molecule intercalates with its nogalose sugar in the minor groove and its aminoglucose sugar in the major groove (37-39, 46, 47). The intercalated chromophore causes buckling of the flanking base pairs. Nogalamycin poisons mammalian DNA topoisomerase I in vitro, resulting in the accumulation of covalent topoisomerase-DNA complexes at a subset of topoisomerase cleavage sites (48,49). Nogalamycin does not poison mammalian DNA topoisomerase II; rather, it globally inhibits DNA cleavage by topoisomerase II (48). Here we tested the effects of nogalamycin on the rate and extent of single-turnover cleavage by vaccinia topoisomerase (Fig. 5). In these experiments, the suicide substrates were made by annealing a 5Ј 32 P-labeled 18-mer scissile strand to an unlabeled 30-mer strand. We tested two different cleavage site sequences, 5Ј-CCCTTpG and 5Ј-CCCTTpA, that differed with respect to the Ϫ1 base pair immediately 3Ј of the scissile phosphodiester. The DNA was preincubated for 10 min with increasing concentrations of nogalamycin, and the DNA-drug mixtures were then reacted for 15 s with vaccinia topoisomerase, which was present in excess over the 32 P-labeled DNA. The use of a short reaction time afforded reasonable sensitivity to drug effects on the rate of topoisomerase transesterification. In control reactions containing 10% Me 2 SO and no drug, 70 -80% of the input DNA was converted to covalent topoisomerase-DNA complex in 15 s. Nogalamycin elicited a concentration-dependent inhibition of transesterification with both substrates. The instructive finding was that the potency of nogalamycin as a DNA cleavage inhibitor was acutely dependent upon the dinucleotide sequence at the scissile phosphodiester (Fig. 5). When the cleavage site was 5Ј-TpG (the preferred dinucleotide sequence for nogalamycin intercalation), the IC 50 for cleavage inhibition by nogalamycin was 0.7 M and the extent of covalent complex formation at 10 M drug was 1/40 of the drug-free control value, whereas when the cleavage site was 5Ј-TpA, the IC 50 was increased to 5 M nogalamycin, and there was substantial residual activity at 10 M drug.
The order of addition of the reaction components also had a profound effect on nogalamycin inhibition of vaccinia topoisomerase (Fig. 6A). In one series of reactions, nogalamycin was preincubated with the 5Ј-CCCTTpG DNA substrate for 10 min prior to initiating the transesterification reaction by the addition of topoisomerase. In another reaction series, nogalamycin was preincubated with topoisomerase for 10 min prior to initiating the transesterification reaction by adding the DNA. In both series, the DNA cleavage reactions were quenched after 15 s. The striking result was that nogalamycin had to be exposed to the DNA substrate prior to adding topoisomerase to exert its inhibitory effects. For example, 10 M nogalamycin reduced cleavage by 98% when added to DNA first, but only 19% when added to topoisomerase first (Fig. 6A). Taken together, the data in Figs. 5 and 6A suggest that nogalamycin exerts its effect on the cleavage reaction of vaccinia topoisomerase by virtue of binding to the DNA, not to the topoisomerase.
Nogamycin is an analog of nogalamycin that lacks the methyl ester on the A ring (Fig. 4). Nogamycin inhibited DNA cleavage by vaccinia topoisomerase in a concentration-dependent fashion when the drug was preincubated with the DNA substrate (IC 50 1.5 M and IC 90 10 M), but was less effective when the order of addition was reversed, so that nogamycin was preincubated with the topoisomerase and the reactions were initiated by adding DNA (Fig. 6B).
A kinetic analysis of single-turnover cleavage of DNA preincubated with 10 M drug is shown in Fig. 6C. The apparent cleavage rate constants (and reaction endpoints) were 0.006 s Ϫ1 (89% of input DNA cleaved) and 0.0005 s Ϫ1 (77% of input DNA cleaved) for nogamycin and nogalamycin, respectively. It is noteworthy that the impact of nogalamycin on the rate of DNA cleavage is virtually identical to that elicited by intercalation of BcPh between the ϩ1 and Ϫ1 bases flanking the scissile phosphodiester.
The topoisomerase catalytic cycle includes two transesterification reactions, cleavage and religation. Religation entails the attack of the DNA 5Ј-OH on the covalent intermediate, leading to expulsion of the Tyr-274 leaving group and restoration of the DNA phosphodiester backbone. The effect of nogalamycin on the religation reaction was studied under single-turnover conditions by assaying the ability of pre-formed suicide intermediate to transfer the covalently held 5Ј 32 P-labeled 12-mer strand to a 5Ј OH-terminated 18-mer strand to form a 30-mer product (5,50). The reaction mixtures containing the suicide intermediate were adjusted to 10% Me 2 SO and 0, 5, 10, or 20 M nogalamycin and incubated for 5 min at 37°C. Religation was initiated by the simultaneous addition of NaCl to 0.5 M and a 50-fold molar excess of a 5Ј hydroxyl-terminated 18-mer acceptor strand d(GTTCCGATAGTGACTACA) complementary to the 5Ј single-stranded tail of the suicide intermediate. The religation reaction was quenched after 15 s. The extent of religation (expressed as the percent of input 32 P-labeled DNA converted to 30-mer) was 77, 76, 75, and 75% in reactions containing 0, 5, 10, and 20 M nogalamycin, respectively (data not shown). These results suggest that nogalamycin interferes selectively with the forward cleavage reaction of vaccinia topoisomerase.

Position-specific Intercalation Interference by Covalent BcPh
Diol Epoxide Adducts-By introducing R and S BcPh adducts at all purine positions of the nonscissile strand sequence 3Ј-GGGAATA complementary to the 5Ј-CCCTTAT target site, we have defined the upstream and downstream margins of the functional interface of vaccinia topoisomerase with DNA containing an intercalated PAH. We observe a discrete and abrupt upstream boundary for BcPh interference effects, whereby transesterification was unaffected by BcPh intercalation between the ϩ6 and ϩ5 base pairs, slowed modestly by intercalation between the ϩ5 and ϩ4 base pairs, and virtually abolished by BcPh intercalation between the ϩ4 and ϩ3 base pairs. It is noteworthy that identical 4-fold rate effects were seen for the ϩ5S and ϩ4R BcPh adducts, which intercalate between ϩ5G and ϩ4G from different orientations. Identical drastic inhibition of cleavage occurs when the ϩ4S and ϩ3R BcPh dG adducts intercalate from different directions between ϩ4G and ϩ3G bases.
The profound suppression of cleavage persists when BcPh insinuates between ϩ3G and ϩ2A, whether it intercalates from the upstream side via the minor groove (ϩ3S BPhG) or from the downstream side via the major groove (ϩ2S BcPh dA). It was striking that the ϩ2R BcPh dA adduct suppressed the yield of the topoisomerase-DNA intermediate, whereas ϩ1S BcPh dA adduct, which insinuates into the same dinucleotide step, affected only the rate of cleavage. We presume that subtle differences in space occupancy by the intercalated ring systems or their induced DNA distortions are responsible for the different effects of the ϩ2R and ϩ1S BcPh dA adducts on topoisomerase cleavage. This orientation disparity for the BcPh dA adducts is reminiscent of the orientation-biased effects of benzo[a]pyrene intercalation into the same dinucleotide step (26).
Placement of the BcPh adduct progressively closer to, and then past, the cleavage site resulted in a gradual alleviation of the interference effects. BcPh intercalation at the scissile phosphodiester (between the ϩ1 and Ϫ1 base pairs) slowed transesterification to 0.0006 s Ϫ1 . The identical rate cleavage constant was reported previously when benzo[a]pyrene was intercalated into the same niche from the ϩ1A base (26). BcPh intercalation between the Ϫ1 and Ϫ2 base pairs slowed cleavage by two orders of magnitude, but intercalation between the Ϫ2 and Ϫ3 base pairs had little effect. Thus the Ϫ2 base pair is the distal margin of the BcPh intercalation interference footprint.
The simplest interpretation of these data is that BcPh inhibition of transesterification is a consequence of disruption of contacts between the topoisomerase and the base pairs of the CCCTT target site that are critical to trigger DNA cleavage. This view is consistent with NMR data showing that single R and S BcPh adducts cause spreading and buckling of the base pairs immediately flanking the intercalated BcPh moiety (Fig.  2). Alternatively, the BcPh adducts may inhibit indirectly by perturbing the phosphodiester backbone.
A previous study by Pommier et al. (51) addressed the effects of S and R BcPh dG modifications of a scissile strand guanine base at a Tp2G cleavage site for human topoisomerase I (equivalent to the Ϫ1 base in our numbering system). They found that the S BcPh dG diastereomer, which intercalates at the scissile phosphodiester (ϩ1/Ϫ1), and the R diastereomer, which intercalates on the 3Ј side of the guanine (between the Ϫ1 and Ϫ2 base pairs in our numbering), completely suppressed cleavage at the T2G site and, instead, triggered cleavage at a new site two nucleotides upstream. The propensity of human topoisomerase I to switch cleavage sites when confronted with an unfavorable DNA lesion makes it difficult to obtain a quantitative picture of interference effects on transesterification by the human enzyme. Nonetheless, the suppressive effects of BcPh intercalation at ϩ1/Ϫ1 and Ϫ1/Ϫ2 steps are qualitatively concordant for the human and vaccinia topoisomerases. As shown previously (25,26), and confirmed here for BcPh, the cleavage site of vaccinia topoisomerase was unaltered by PAH modifications.
Intercalation Interference by Nogalamycin-Having demonstrated the potency of covalently fixed intercalators as inhibitors of DNA cleavage by vaccinia topoisomerase, we tested the effects of a well studied soluble intercalator, nogalamycin, which binds DNA in a sequence-dependent fashion. The inhibition of vaccinia topoisomerase by nogalamycin is concentration-dependent, requires preincubation of drug with the DNA, and is enhanced by the presence of a guanine at the N base flanking the CCCTTp2N cleavage site. Given that TG is a preferred target site for nogalamycin intercalation, we posit that nogalamycin inhibits by intercalating between the ϩ1 and Ϫ1 base pairs and thereby mimicking the action of the covalent ϩ1R BcPh dA adduct, which elicits an effect similar to nogalamycin on the rate of single-turnover cleavage.
The order of addition effect on nogalamycin inhibition suggests that (i) the drug acts by binding to DNA rather than to topoisomerase, and (ii) the binding of topoisomerase to DNA is faster than the binding of nogalamycin to DNA. The latter point is consistent with experimental data (52,53) and is sensible in light of the steric problems involved in inserting the bulky sugar appendages of the dumbbell-shaped nogalamycin molecule between the base pairs of duplex DNA. Once intercalated, nogalamycin dissociates very slowly from DNA (54). The kinetics of dissociation depend upon the DNA sequence complexity. Nogalamycin is released with a half-life of 50 min from the alternating copolymer poly(dG-dT)⅐ poly(dA-dC), which consists exclusively of tandem TG dinucleotides that are preferred intercalation sites. Nogalamycin dissociates from calf thymus DNA with three apparent kinetic components, with half-lives of 5, 17, and 72 min, respectively (54).
Nogalamycin suppresses the forward cleavage transesterification reaction, but has little apparent effect upon religation by the suicide intermediate. Thus, nogalamycin can be classified as a poxvirus topoisomerase inhibitor rather than a poxvirus topoisomerase poison. The poisoning effect of nogalamycin on mammalian DNA topoisomerase I has been studied thoroughly by Liu and coworkers (48,49). The covalent trapping of mammalian topoisomerase on DNA elicited by nogalamycin occurs only at a subset of topoisomerase cleavage sites. Detailed analysis of one such nogalamycin-stimulated cleavage site attributes the poisoning effect to the intercalation of nogalamycin at a TG step located six nucleotides upstream (5Ј) of the scissile phosphodiester (49). However, it has been shown that nogalamycin is also a concentration-dependent inhibitor of DNA cleavage by mammalian topoisomerase I at many other target sites in duplex DNA (48,49). Thus, the outcome of nogalamycin intercalation with respect to the mammalian topoisomerase will depend upon the sequence context and the position of the intercalated drug relative to the cleavage site. From a pharmacological perspective, the topo I poison mechanism dominates cytotoxicity, which allows for drug efficacy even if the topoi-somerase is nonessential for cell growth (55) and even if only a fraction of the potential topoisomerase-DNA complexes can be trapped by the drug.
In the case of vaccinia topoisomerase, we can reasonably surmise that nogalamycin is not acting as a poison on the model CCCTTpG substrate employed herein for the analysis of the cleavage and religation reactions. We do not exclude the possibility that nogalamycin can have a modest slowing effect upon religation at this cleavage site, because the religation reaction is fast (k rel ϭ 1 s Ϫ1 ) and the sensitivity of the 15-s single-turnover religation assay is such that a 10-fold rate decrement would not be detected. Also, we do not exclude the possibility that nogalamycin could poison vaccinia topoisomerase at a subset of target sites in viral DNA that have an especially suitable flanking sequence context. Because nogalamycin is not likely to intercalate at subtoxic concentrations within the conserved pentapyrimidine (T/C)CCTT target site, and given that only a fraction of poxvirus topoisomerase cleavage sites in vivo will be flanked by a TpG dinucleotide, we suspect that the inhibitory mode of nogalamycin action will not provide for antiviral efficacy predicated on specific targeting of the poxvirus topoisomerase. Nonetheless, initial studies show that nogalamycin inhibits vaccinia virus replication in cell culture; 2 it will be of interest to identify a molecular target for its antiviral action.
In conclusion, position-specific intercalation is a powerful means to block DNA cleavage by poxvirus topoisomerase and may, in principle, afford useful diffusible small-molecule inhibitors, provided that the compounds can be directed to the CCCTT target sequence, and small molecule poisons, provided that the compounds can be directed to the covalent topoisomerase-DNA intermediate.