A catalytic domain of eukaryotic DNA topoisomerase I.

Eukaryotic type IB topoisomerases catalyze the cleavage and rejoining of DNA strands through a DNA-(3'-phosphotyrosyl)-enzyme intermediate. The 314-amino acid vaccinia topoisomerase is the smallest member of this family and is distinguished from its cellular counterparts by its specificity for cleavage at the target sequence 5'-CCCTT downward arrow. Here we show that Topo-(81-314), a truncated derivative that lacks the N-terminal domain, performs the same repertoire of reactions as the full-sized topoisomerase: relaxation of supercoiled DNA, site-specific DNA transesterification, and DNA strand transfer. Elimination of the N-terminal domain slows the rate of single-turnover DNA cleavage by 10(-3.6), but has little effect on the rate of single-turnover DNA religation. DNA relaxation and strand cleavage by Topo-(81-314) are inhibited by salt and magnesium; these effects are indicative of reduced affinity in noncovalent DNA binding. We report that identical properties are displayed by a full-length mutant protein, Topo(Y70A/Y72A), which lacks two tyrosine side chains within the N-terminal domain that contact the DNA target site in the major groove. We speculate that Topo-(81-314) is fully competent for transesterification chemistry, but is compromised with respect to a rate-limiting precleavage conformational step that is contingent on DNA contacts made by Tyr-70 and Tyr-72.

The eukaryotic type IB DNA topoisomerase family includes topoisomerase I, a ubiquitous nuclear enzyme, and the topoisomerases encoded by vaccinia and other cytoplasmic poxviruses (1). These proteins relax supercoiled DNA via a common reaction pathway, which involves noncovalent binding of the topoisomerase to duplex DNA, cleavage of one DNA strand with concomitant formation of a covalent DNA-(3Ј-phosphotyrosyl)-protein intermediate, strand passage, and strand religation. Our aim is to understand the structural requirements for DNA recognition and transesterification chemistry. Toward that end, we have undertaken a structure-function analysis of the vaccinia topoisomerase.
The 314-amino acid vaccinia enzyme is the smallest topoisomerase known and thus affords a more tractable target for structure-function studies than the cellular type IB enzymes, which range from 765 to 1019 amino acids (2,3). Another attractive feature of the vaccinia topoisomerase is its sequence specificity in transesterification; it forms a covalent adduct at sites containing the sequence 5Ј-(C/T)CCTT2 immediately 5Ј of the scissile phosphodiester (4). Thus far, we have identified key contacts on the DNA target site by enzymatic and chemical footprinting (5)(6)(7), probed the protein side of the protein-DNA interface by limited proteolysis of the topoisomerase in the free and DNA-bound states (8), mapped specific DNA contact points on the enzyme by UV photo-cross-linking (9), and performed targeted mutagenesis of 140 individual amino acid residues (10 -18).
The vaccinia virus topoisomerase consists of three proteaseresistant polypeptide fragments separated by two proteasesensitive interdomain segments, which we have referred to as the bridge and hinge ( Fig. 1) (8). Specific functional groups identified through mutagenesis as being required for transesterification chemistry are situated near the hinge and within the C-terminal domain. These include the active site nucleophile (Tyr-274) and four other residues (Arg-130, Lys-167, Arg-223, and His-265) that are essential for the DNA cleavage and religation steps (11-15, 17, 19). Two other residues (Gly-132 and Tyr-136) are critical for the cleavage reaction, but not for religation (17). None of the essential residues appears to play a role in target site affinity, insofar as alanine substitutions that elicit from 10 Ϫ2 to 10 Ϫ7 decrements in transesterification rate have no significant effect on the noncovalent binding of topoisomerase to CCCTT-containing duplex DNA (17).
The interdomain bridge is defined by trypsin-accessible sites at Arg-80, Lys-83, and Arg-84 (8,20). Residues implicated in noncovalent DNA binding are situated within the bridge and in the N-terminal domain just proximal to the bridge. Tyr-70 and Tyr-72 were identified as the sites of UV cross-linking between topoisomerase and the ϩ4 and ϩ3 bromocytosine-substituted bases, respectively, of the CCCTT element (9). Alanine-scanning mutagenesis of the N-terminal domain suggests that Arg-67, Tyr-70, Tyr-72, and Arg-80 contribute to target site affinity (18). Mutational effects on DNA binding are evinced by inhibition of topoisomerase activity in the presence of magnesium and salt (16,18).
These results have prompted the suggestion (9) that low affinity DNA binding and reaction chemistry are performed by the carboxyl two-thirds of the vaccinia enzyme, the sequence of which is similar to that of the cellular topoisomerases, whereas discrimination of the DNA sequence at the cleavage site is facilitated by the N-terminal domain, which is divergent in sequence and three-dimensional structure between the viral and cellular enzymes (20,21). Here, we demonstrate that a deleted version of vaccinia topoisomerase, Topo-(81-314), that lacks the N-terminal 80-amino acid domain (Fig. 1), is active in relaxing supercoiled DNA, but is exquisitely sensitive to inhibition by salt or magnesium. The capacity of Topo-(81-314) to cleave a CCCTT-containing substrate under nonstringent conditions suggests that the catalytic domain per se can discriminate the target site. We propose a revised model for catalysis whereby the N-terminal domain enhances DNA binding and is required for a pre-cleavage conformational step.

EXPERIMENTAL PROCEDURES
The segment of the vaccinia virus topoisomerase gene encoding amino acids 81-314 was polymerase chain reaction-amplified using a sense-strand oligonucleotide primer that introduced an internal NdeI restriction site (CATATG) with an in-frame methionine codon in lieu of the codon for Arg-80. An NdeI-BglII restriction fragment containing the * This work was supported by National Institutes of Health Grant GM46330. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. truncated topoisomerase gene was cloned into the T7-based expression vector pET3c to yield pET-Topo-(81-314). The entire insert of this plasmid was sequenced to confirm that no unwanted mutations had been introduced during amplification or cloning. The pET-Topo-(81-314) plasmid was transformed into Escherichia coli BL21. Topo-(81-314) expression was induced by infection with bacteriophage CE6 (22). Topo-(81-314) was purified from soluble bacterial lysates by phosphocellulose column chromatography and glycerol gradient sedimentation. The elution and sedimentation profiles of the Topo-(81-314) polypeptide were monitored by SDS-PAGE 1 of the column and gradient fractions. Topo-(81-314) adsorbed to phosphocellulose in lysis buffer (50 mM Tris-HCl (pH 8.0), 10 mM EDTA, 1 mM DTT, 10% sucrose) containing 150 mM NaCl, remained bound during a wash with buffer A (50 mM Tris-HCl (pH 8.0), 2 mM DTT, 1 mM EDTA, 10% glycerol, 0.1% Triton X-100) containing 0.5 M NaCl, and was recovered by step elution with buffer A containing 1.0 M NaCl. An aliquot of the phosphocellulose preparation (250 g) was applied to a 4.8-ml 15-30% glycerol gradient in 50 mM Tris-HCl (pH 8.0), 100 mM NaCl, 2 mM DTT, 1 mM EDTA, 0.1% Triton X-100. The gradient was centrifuged for 30 h at 50,000 rpm in a Beckman SW50 rotor. Fractions (0.2 ml) were collected dropwise from the bottom of the tube.
Full-length wild type topoisomerase and the double alanine-substitution mutant Topo(Y70A/Y72A) were purified from soluble bacterial lysates by phosphocellulose column chromatography and glycerol gradient sedimentation as described above for Topo-(81-314). The protein concentrations of the topoisomerase preparations were determined by using the dye-binding method (Bio-Rad) with bovine serum albumin as the standard.

Relaxation of Supercoiled DNA by Topo-(81-314)-Topo-(81-
314) is a truncated version of vaccinia topoisomerase that lacks the N-terminal structural domain defined by limited proteolysis with trypsin ( Fig. 1). Topo-(81-314) was expressed in bacteria and purified from soluble lysates by phosphocellulose chromatography and glycerol gradient sedimentation. The 27-kDa Topo-(81-314) polypeptide sedimented as a discrete peak (Fig. 2). The N-terminal sequence of this species was determined by automated Edman chemistry after transferring the 27-kDa polypeptide from an SDS-polyacrylamide gel to a polyvinylidene difluoride membrane (8). The amino acid sequence (MNAKRDRIF) confirmed that the protein we purified was indeed Topo-(81-314). A single peak of DNA supercoil relaxation activity cosedimented with the Topo-(81-314) protein (Fig. 2).
The rates of DNA relaxation by Topo-(81-314) and fulllength wild type (WT) topoisomerase were compared at 25 nM input protein (Fig. 3). Screening assays were performed in the absence of added salt or magnesium (Fig. 3, Control). The rate of DNA relaxation by WT topoisomerase under these conditions was proportional to enzyme concentration from 8 nM to 100 nM (data not shown); higher concentrations were not tested. 25 nM WT topoisomerase relaxed 0.3 g of supercoiled pUC19 DNA to completion within 5-10 min. Reaction products of intermediate superhelicity were not observed, suggesting that the WT en-zyme relaxed individual DNA molecules to completion before dissociating and engaging a new DNA. 25 nM Topo-(81-314) relaxed all input supercoiled DNA within 30 min, as evinced by the decay of the supercoiled substrate. However, in contrast to WT, the truncated enzyme accumulated partially relaxed intermediates (e.g. at 10 and 20 min) and did not relax the DNA to completion even after 60 min (Fig. 3). This suggested that Topo-(81-314) acted distributively on supercoiled plasmid DNA.
The DNA relaxation assays were also performed in the presence of 100 mM NaCl or 5 mM MgCl 2 . Either salt or magnesium stimulated the relaxation rate of the WT enzyme ϳ20-fold, such that all supercoils were relaxed in 15-30 s (Fig. 3). Salt and magnesium together have an additive effect on relaxation rate (23); this is not apparent in Fig. 3, because the reaction was complete with either solute alone at the earliest time analyzed. The response of Topo-(81-314) to salt and magnesium was the exact opposite of the WT protein. Relaxation by Topo-(81-314) was slowed severely by 100 mM NaCl; MgCl 2 blocked activity almost completely (Fig. 3). The combination of salt and magnesium was similarly deleterious.
Prior studies indicated that product dissociation is rate-limiting during relaxation by WT enzyme in the absence of salt or magnesium. Salt and magnesium stimulate relaxation by enhancing product off-rate, without affecting the rate of DNA cleavage by the WT topoisomerase (23,24). In the presence of salt plus magnesium, the DNA cleavage step is apparently 1 The abbreviations used are: PAGE, polyacrylamide gel electrophoresis; WT, wild type; DTT, dithiothreitol; bp, base pair(s). rate-limiting. The paradoxical response of Topo-(81-314) to salt and magnesium is strongly suggestive of reduced affinity of the truncated protein for the DNA substrate. Strand cleavage is likely to be rate-limiting in the absence of salt or magnesium. However, in the presence of salt or magnesium, the DNA binding step becomes limiting for Topo-(81-314). The relative rates of DNA cleavage can be gauged crudely by comparing the WT relaxation rate in the presence of salt and magnesium to the relaxation rate of Topo-(81-314) in the absence of added solutes. By this criteria, we estimate that the rate of DNA cleavage by Topo-(1-314) was less than 1% of the WT rate.
Topo-(81-314) Forms a Covalent Adduct on CCCTT-containing DNA-A DNA substrate containing a single CCCTT2 site was used to examine DNA cleavage under single-turnover conditions. The substrate consisted of a 5Ј 32 P-labeled 18-mer scissile strand 5Ј-pCGTGTCGCCCTTATTCCC annealed to a 30-mer strand 3Ј-GCACAGCGGGAATAAGGCTATCACTGA-TGT to produce an 18-mer duplex with a 12-mer 5Ј tail (Fig. 4). Upon formation of the covalent protein-DNA adduct, the distal cleavage product 5Ј-ATTCCC is released and the enzyme remains bound to the 32 P-labeled 12-mer strand pCGTGTCGC-CCTTp. The labeled protein-DNA adduct formed by WT topoisomerase migrated as a discrete species of ϳ40 kDa during SDS-PAGE (Fig. 5, left panel). In reactions containing 25 nM 18-mer/30-mer DNA and 125 nM WT topoisomerase, 95% of the input DNA became covalently bound to protein and the reaction was nearly complete within 10 s at 37°C. The observed cleavage rate constant (k cl ) for the WT topoisomerase was 0.28 s Ϫ1 (14,15). We found that Topo-(81-314) cleaved the 18-mer strand to form a labeled protein-DNA adduct that migrated at ϳ32 kDa during SDS-PAGE (Fig. 5, left panel, ⌬N). The faster electrophoretic mobility of the Topo-(81-314)-DNA adduct was consistent with the smaller size of the Topo-(81-314) polypeptide.
Topo-(81-314) cleaved the 18-mer/30-mer DNA very slowly (Fig. 4). An end point of 87% label transfer was attained at 24 h. The apparent cleavage rate constant (k cl ) was 7 ϫ 10 Ϫ5 s Ϫ1 . Thus, the rate of DNA cleavage by Topo-(81-314) was slower than WT by a factor of 10 Ϫ3. 6 . Because the observed k cl did not increase when the Topo-(81-314) concentration was doubled, we surmise that deletion of the N-terminal domain affects either transesterification chemistry per se or a step in the reaction pathway that occurs after initial binding, but prior to strand scission.
The site specificity of cleavage of the 18-mer scissile strand by Topo-(81-314) was initially compared with that of the WT enzyme by treating the respective covalent adducts with proteinase K in the presence of SDS, then resolving the digestion products by electrophoresis through a 17% polyacrylamide gel containing 7 M urea. Digestion of the wild type covalent adduct yielded a cluster of labeled species migrating faster than the 18-mer input strand, but slower than a free 12-mer strand (data not shown). These digestion products consisted of the 12-mer oligonucleotide 5Ј-pCGTGTCGCCCTTp linked to short peptides of heterogeneous size. Digestion of the Topo-(81-314) covalent adduct yielded an identical cluster of 32 P-labeled DNA-peptide adducts (data not shown). Because any alteration in the site of covalent adduct formation on the 32 P-labeled scissile strand would result in an easily detectable shift in the mobility of the proteinase K digestion products (25), we surmise that Topo-(81-314) transesterified at the same phosphodiester bond as the WT enzyme.
DNA Religation by the Covalent Topo-(81-314)-DNA Intermediate-The religation reaction was studied under singleturnover conditions by assaying the ability of the covalent intermediate to transfer the covalently held 5Ј 32 P-labeled 12mer strand to a 5Ј-hydroxyl-terminated 18-mer strand to form a 30-mer product (26). WT topoisomerase and Topo-(81-314) were first preincubated with the 18-mer/30-mer substrate. DNA analysis by denaturing gel electrophoresis established that virtually all of the input 32 P-labeled 18-mer strand had reacted with the WT topoisomerase (as gauged by disappearance of the 18-mer scissile strand; note that the covalent protein-DNA adduct does not enter the gel), whereas Topo-(81-314) reacted with ϳ70% of the substrate (Fig. 5, right panel, (20 l) were withdrawn at the times indicated and quenched immediately by adding SDS to 1%. The denatured samples were electrophoresed through a 12% polyacrylamide gel containing 0.1% SDS. Covalent complex formation was revealed by transfer of radiolabeled DNA to the topoisomerase polypeptide. The extent of covalent adduct formation (expressed as the percent of the input 5Ј 32 P-labeled oligonucleotide that was transferred to protein) was quantitated by scanning the dried gel using a FUJIX BAS1000 Bio-Imaging Analyzer and is plotted as a function of reaction time. The data were normalized to the end point value of 87% label transfer, and k cl was determined by fitting the data to the equation (100 Ϫ %Cl norm ) ϭ 100e Ϫkt . time 0 lanes). The religation reaction was initiated by adding a 50-fold molar excess of an 18-mer acceptor strand complementary to the 5Ј tail of the covalent intermediate. The reaction mixture was adjusted simultaneously to 0.3 M NaCl. Addition of NaCl during the religation phase promotes dissociation of the topoisomerase after strand closure and completely prevents recleavage of the strand transfer product. Religation by WT topoisomerase to form the 30-mer strand transfer product was complete within 10 s (Fig. 5, right panel). Remarkably, so was the religation reaction catalyzed by Topo-(81-314). This was in stark contrast to the severe rate defect evinced by Topo-(81-314) in the forward cleavage reaction. The religation results show that Topo-(81-314) catalyzed transesterification reaction chemistry at a near-wild type rate. The fact that the strand transfer product generated by Topo-(81-314) was identical in size to that formed by WT topoisomerase provides additional proof that the truncated enzyme cleaved the scissile strand at the CCCTTp2A phosphodiester (Fig. 5). A profound and selective decrement in the rate of single-turnover strand cleavage versus religation should result in a drastic decrease in K eq when Topo-(81-314) reacts with CCCTT-containing DNA under equilibrium conditions (17). WT topoisomerase attains a cleavage-religation equilibrium on a 60-bp duplex containing a centrally placed CCCTT element, such that ϳ20% of the substrate is covalently bound to the enzyme (17). Cleavage of the 60-bp equilibrium substrate by Topo-(81-314) was barely detectable; Ͻ0.1% of the scissile strand became covalently bound (data not shown).
Effects of Salt and Magnesium on DNA Cleavage by Topo-(81-314)-Single turnover cleavage reactions are routinely performed at low ionic strength in the absence of a divalent cation. The findings that DNA relaxation by Topo-(81-314) was inhibited by 0.1 M NaCl and 5 mM MgCl 2 suggested that pre-cleavage binding might be rate-limiting under those more stringent conditions. To address this issue, we examined the effects of salt and magnesium on cleavage of the 18-mer/30-mer substrate. The amounts of covalent adduct formed in the presence of 50, 100, 150, and 200 mM NaCl or 1, 2, 4, 6, and 8 mM MgCl 2 were measured and normalized to the extent of cleavage in unsupplemented control reactions. WT topoisomerase cleavage reactions were quenched after 10 s, whereas Topo-(81-314) reactions were terminated after 6 h. The reaction times were chosen to attain comparable sensitivity for the effects of solution parameters on the cleavage reaction. The salt effects are shown in Fig. 6A; magnesium effects are shown in Fig. 6B. We observed that the WT topoisomerase was unaffected by up to 150 mM NaCl, but was inhibited by 26% at 200 mM NaCl. In contrast, covalent adduct formation by Topo-(81-314) was saltsensitive; Topo-(81-314) was inhibited 29% and 85%, respectively, by 50 and 100 mM NaCl. Topo-(81-314) cleavage activity was abolished at 150 mM and 200 mM NaCl (Fig. 5A). WT topoisomerase was unaffected by magnesium up to 8 mM, whereas Topo-(81-314) was inhibited progressively by 1-8 mM MgCl 2 . 52% inhibition occurred at 1 mM MgCl 2 and 88% inhibition was observed at 6 mM MgCl 2 (Fig. 5B). Susceptibility to salt and magnesium inhibition indicates that deletion of the N-terminal domain results in decreased affinity for the CCCTT-containing DNA substrate.
Noncovalent DNA Binding by Topo-(81-314)-The noncovalent binding of Topo-(81-314) to a 32 P-labeled 60-bp DNA containing a single centrally placed CCCTT site (17) was assessed by native gel electrophoresis (27). The full-sized active site mutant protein Topo(Phe-274) was analyzed in parallel (Fig. 7). The binding reaction mixtures contained no added salt or magnesium. The Phe-274 protein bound to the 60-mer ligand to form a single discrete complex of retarded electrophoretic mobility (indicated by the asterisk in Fig. 7). The extent of complex formation was proportional to input Phe-274 topoisomerase and was near quantitative at a 2:1 molar ratio of protein to DNA. Increasing the concentration of Phe-274 to attain 5:1 and 10:1 molar ratios of protein to DNA resulted in the appearance of at least two more slowly migrating complexes (Fig. 7). We presume that this reflects the binding of one or two more topoisomerase monomers to the 60-bp DNA. We showed previously that Phe-274 topoisomerase binds at nonspecific sites on duplex DNA with 7-10-fold lower affinity than to CCCTT sites (27). The same is true of noncovalent binding by WT topoisomerase (27). Incubation of Topo-(81-314) with the 60-mer ligand resulted in the formation of a somewhat diffuse smear of slowly migrating material (Fig. 7). We estimate from the protein concentrations required to attain a comparable decrease in the residual unbound DNA that Topo-(81-314) bound the 60mer with one-fifth the affinity of the full-sized enzyme.
Limited Proteolysis of Topo-(81-314) with Chymotrypsin-To address the possibility that the altered biochemical properties of recombinant Topo-(81-314) were caused by aberrant protein folding, we probed the structure of Topo-(81-314) by digestion with increasing concentrations of chymotrypsin. At limiting protease concentrations, Topo-(81-314) was cleaved to yield a predominant polypeptide fragment of 20 kDa and a minor fragment of 18 kDa (Fig. 8). The 20-kDa carboxyl species was resistant to digestion by a level of chymotrypsin sufficient to cleave all the input Topo-(81-314). The 18-kDa polypeptide increased in abundance at the higher levels of protease. Sequencing of these polypeptides by automated Edman chemistry after transfer to a polyvinylidene difluoride membrane revealed that the 20-kDa species arose via cleavage between amino acids Tyr-136 and Leu-137 and the 18-kDa species arose via cleavage between residues Leu-146 and Thr-147. These two sites of chymotrypsin cleavage of Topo-(81-314) correspond precisely to the chymotrypsin-accessible sites in the hinge region of the WT topoisomerase (8). This result suggests that recombinant Topo-(81-314) was folded properly.
Substitution of Tyr-70 and Tyr-72 by Alanine Mimics the Effects of Deleting the N-terminal Domain-In light of prior findings that two specific residues within the N-terminal domain (Tyr-70 and Tyr-72) make major groove contacts with the nucleotide bases of the CCCTT target site (9), we considered the hypothesis that loss of these contacts might account for the effects of N-terminal domain removal on DNA binding and cleavage. To address this point, we purified and characterized a mutated version of the full-length vaccinia topoisomerase, Topo(Y70A/Y72A), in which Tyr-70 and Tyr-72 were both replaced by alanine. Previously, we reported the effects of single alanine-substitutions at Tyr-70 and Tyr-72 (18). The Y70A mutation had no effect on the kinetics of DNA relaxation in the presence of 0.1 M NaCl. However, Y70A mutant displayed an aberrant response to magnesium, whereby its rate of relaxation was lower by a factor of 5 in the presence of 5 mM magnesium plus 0.1 M NaCl compared with 0.1 M NaCl alone. The Y72A mutation slowed the rate of DNA relaxation in the presence of 0.1 M NaCl by a factor of 8 (compared with the WT rate), and the combination of NaCl and magnesium nearly abolished DNA relaxation by Y72A.
Here, we analyzed the kinetics of DNA relaxation by the Y70A/Y72A double mutant under the same nonstringent conditions (no salt or magnesium) and enzyme concentration used to detect relaxation by Topo-(81-314). We found that the topoisomerase activity of Y70A/Y72A was comparable to that of Topo-(81-314) with respect to rate and the accumulation of partially relaxed topoisomers (Fig. 9). Y70A/Y72A also displayed the same profound inhibition of relaxation by 100 mM NaCl or 5 mM MgCl 2 that was observed for Topo-(81-314) (Fig.  9). Y70A/Y72A formed a covalent intermediate with the 18mer/30-mer CCCTT-containing DNA under nonstringent conditions; however, the rate of single turnover cleavage was extremely slow (k cl ϭ 3.6 ϫ 10 Ϫ5 s Ϫ1 ) (data not shown). The cleavage rate decrement elicited by the double-alanine replacement was nearly identical to that caused by removal of the entire N-terminal domain. The sensitivity of covalent adduct formation by Y70A/Y72A to inhibition by salt and magnesium paralleled that of Topo-(81-314) (Fig. 6). DISCUSSION The results presented in this study of vaccinia topoisomerase enhance our understanding of the eukaryotic type IB enzyme family as follows. (i) They define a catalytically active domain that is conserved between the cellular and poxvirus enzymes; (ii) they illuminate a clear distinction between structural elements required for transesterification chemistry in general and those required specifically for the cleavage reaction; (iii) they demonstrate that the catalytic domain per se is capable of recognizing the target sequence for transesterification.
The 234-amino acid catalytic domain, Topo-(81-314), performed the same repertoire of reactions as the full-sized vaccinia topoisomerase: relaxation of supercoiled DNA, site-specific DNA transesterification, and DNA strand transfer. Our attempts to isolate a more extensively truncated, enzymatically active N-terminal deletion mutant of vaccinia topoisomerase were unsuccessful. 2 Therefore, we suggest that the vaccinia catalytic domain defines (for now) the minimum functional unit of a type IB topoisomerase. The catalytic domain includes the active site Tyr-274 and the four other conserved residues (Arg-130, Lys-167, Arg-223, and His-265) that we have shown are essential for transesterification reaction chemistry. Essential residues are defined as those at which side chain removal causes at least a 10 Ϫ2 rate decrement in single-turnover religation.
We regard mutational effects on the rate of the religation reaction as the simplest measure of the contribution of any protein moiety to reaction chemistry. This is because singleturnover religation entails no site-recognition step, i.e. the topoisomerase is already bound covalently to the DNA. Although loss of the N terminus is not without consequences, it appears that the N-terminal domain is not directly involved in covalent catalysis, i.e. the rate of religation by Topo-(81-314) was similar to that of WT topoisomerase. Note that, because the rate of DNA religation to the 18-mer acceptor oligonucleotide is too fast to measure accurately by manual assay methods (k rel ϭ 1.3 s Ϫ1 ), we can only construe from the finding that the reactions were complete at 10 s that the rate of transesterification by Topo-(81-314) was within a factor of 10 of the WT rate. These results indicate that the N-terminal domain is nonessential for transesterification chemistry. Previously, we had mutated 22 individual residues of the 80-amino acid N-terminal domain to alanine and found that none of these single substitutions had a significant effect on transesterification rate (14,16,18).
The severe and biased effects of deleting the N-terminal domain on the rate of the forward cleavage reaction suggest that a post-binding, pre-chemical step applies during the cleavage reaction, which is not pertinent during religation. Although the precise nature of the pre-cleavage step remains unclear, the available evidence that residues within the N-terminal domain make major groove contacts with the CCCTT target site (9) suggests a post-binding alteration of the protein-DNA interface. Elimination of the two tyrosine side chains responsible for these major groove interactions (Tyr-70 and Tyr-72) elicits the same effects on relaxation and cleavage as does removal of the entire N-terminal domain. We propose that major groove contacts made by the two tyrosines induce a conformational change in the catalytic domain that activates the cleavage step.
The hinge segment of the catalytic domain is a likely mediator of this proposed conformational step. The hinge is highly susceptible to proteolysis when topoisomerase is free in solution and becomes protected from proteolysis when the enzyme is bound to DNA (8). Residues adjacent to the hinge become more accessible to proteolysis in the DNA-bound state (8). The alteration in the protease sensitivity of the hinge occurs prior to transesterification. Moreover, single alanine substitutions at residues Gly-132 and Tyr-136 within the hinge cause a reduction by more than 2 orders of magnitude in the rate of DNA cleavage, but only a modest effect on religation (17). These effects are strikingly similar to those reported here for deletion of the N terminus. We speculate that N-terminal domain and hinge dynamics may activate cleavage by properly orienting the catalytically essential residues with respect to the scissile phosphate.
The scissile phosphate and six other phosphates that contact the topoisomerase are located on the minor groove of the CCCTT target site (7). Our observation that the catalytic domain retains specificity for cleavage at CCCTT implies that vaccinia topoisomerase (i) interacts with the bases in the minor groove, (ii) senses the target site through indirect readout of backbone conformation, or (iii) recognizes the major groove through contacts intrinsic to the catalytic domain per se. Loss of contacts made by the N-terminal domain resulted in decreased affinity for the target site, which was manifest by salt and magnesium inhibition of single-turnover cleavage and DNA relaxation by Topo-(81-314). The salt and magnesium inhibition curves were essentially identical for Topo-(81-314) and Y70A/Y72A. We surmise that the two tyrosine side chains are largely responsible for the contribution made by the N-terminal domain to target site affinity. Studies of the effects of conservative replacement for Tyr-70 or Tyr-72 point to the aromatic character of these side chains as being important for enhancing DNA binding (18).
The present findings suggest that the distinctive target site specificities of the poxvirus and cellular type IB topoisomerases are not simply a function of structural differences between the domains N-terminal to the conserved catalytic core (20,21). For the vaccinia enzyme, the specificity of cleavage is an intrinsic property of the core domain. This may also be the case for the cellular enzyme, to the extent that a core domain has been defined. Stewart et al. (28) have shown that active human topoisomerase can be reconstituted from a 58-kDa fragment derived from the central portion of the protein (this fragment contains all the residues of the vaccinia enzyme essential for transesterification except the active site tyrosine) plus a Cterminal 6.3-kDa fragment that includes the active site tyrosine. The fragment-reconstituted human enzyme cleaved DNA with the same sequence specificity as the full-sized protein.
However, the cleavage-religation equilibrium of the reconstituted enzyme was skewed toward religation, DNA binding affinity was reduced, and the enzyme relaxed distributively (28). These properties of fragment-reconstituted human topoisomerase I vis à vis the intact enzyme are broadly similar to what we observe for the vaccinia catalytic domain. Nonetheless, the reconstituted human fragments together are still more than twice the size of the catalytic domain of the vaccinia enzyme.
Although mutational analysis of the vaccinia topoisomerase has implicated individual amino acid residues in general transesterification chemistry, in the cleavage reaction specifically, and in noncovalent DNA binding, a complete interpretation of these results will ultimately hinge on the availability of a crystal structure of the enzyme in both the free and DNA-2 C. Cheng, unpublished data. bound states. Efforts in this and other laboratories to crystallize the full-sized vaccinia topoisomerase have been unsuccessful. We suspect this is due to flexibility of the protein at the interdomain bridge and hinge segments. We recently achieved success in crystallizing the catalytically active domain characterized in this paper. 2 The structure of the domain is under refinement and will be reported separately.