ABSTRACT
Vaccinia topoisomerase, a eukaryotic type IB enzyme, catalyzes relaxation of supercoiled DNA by cleaving and rejoining DNA strands through a DNA- (3'
-phosphotyrosyl)-enzyme intermediate. We have performed a kinetic analysis of mutational effects at four essential amino acids: Arg-130, Gly-132, Tyr-136 and Lys-167. Arg-130, Gly-132 and Lys-167 are conserved in all members of the type IB topoisomerase family. Tyr-136 is conserved in all poxvirus topoisomerases. We show that Arg-130 and Lys-167 are required for transesterification chemistry. Arg-130 enhances the rates of both cleavage and religation by 105. Lys-167 enhances the cleavage and religation reactions by 103 and 104, respectively. An instructive distinction between these two essential residues is that Arg-130 cannot be replaced by lysine, whereas substituting Lys-167 by arginine resulted in partial restoration of function relative to the alanine mutant. We propose that both basic residues interact directly with the scissile phosphate at the topoisomerase active site. Mutations at positions Gly-132 and Tyr-136 reduced the rate of strand cleavage by more than two orders of magnitude, but elicited only mild effects on religation rate. Gly-132 and Tyr-136 are suggested to facilitate a pre-cleavage activation step. The results of comprehensive mutagenesis of the vaccinia topoisomerase illuminate mechanistic and structural similarities to site-specific recombinases.
The type IB DNA topoisomerase family includes eukaryotic 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 mechanism, which involves noncovalent binding of the topoisomerase to duplex DNA, cleavage of one DNA strand with concomitant formation of a covalent DNA-(3'-phosphotyrosyl)-protein intermediate, strand passage and strand religation.
The 314 amino acid vaccinia virus topoisomerase is the smallest topoisomerase known and likely constitutes the minimal functional unit of a type IB enzyme (2 ). The cellular type IB enzymes vary in size from 765 to 1019 amino acids (3 ) Our aim is to understand the structural basis for transesterification reaction chemistry via mutational analysis of the vaccinia protein. Four strategies have been employed: (i) random mutagenesis followed by in vivo genetic selection of mutations that adversely affect enzyme activity (4 ,5 ); (ii) serial deletion of amino acids from the C-terminus (6 ); (iii) targeted mutagenesis of a specific class of amino acid side chains irrespective of location within the protein (7 ,8 ); and (iv) comprehensive mutagenesis of specific regions of the enzyme (9 -12 ).
In applying the regional mutagenesis strategy, we previously targeted 103 amino acids within three conserved protein segments: from residues 126-167, 181-207 and 213-274, respectively. Six amino acids in addition to the active site Tyr-274 were defined as essential, i.e., substitution by alanine elicited at least a 10-2 effect on activity in DNA relaxation and the formation of the covalent DNA-protein intermediate. The six essential residues are Arg-130, Gly-132, Tyr-136, Lys-167, Arg-223 and His-265. Three other residues, Ser-204, Lys-220 and Asn-228, were defined as important for catalysis, i.e., alanine substitution slowed the rate of covalent adduct formation by at least one order of magnitude. Six of the essential and important amino acids (Arg-130, Gly-132, Lys-167, Lys-220, Arg-223 and His-265) are strictly conserved in every member of the eukaryotic type IB enzyme family. Tyr-136, Ser-204 and Asn-228 are conserved among the poxvirus-encoded topoisomerases.
In the present study, we present a detailed kinetic analysis of the mutational effects at four essential positions: Arg-130, Gly-132, Tyr-136 and Lys-167. We analyzed the effects of alanine substitutions and conservative substitutions on the rates of DNA strand cleavage and religation under equilibrium and single-turnover conditions. We show that Arg-130 and Lys-167 are essential for both cleavage and religation chemistry and suggest a model for their action at the scissile phosphate. In contrast, Gly-132 and Tyr-136 are required for strand cleavage, but mutations at these positions elicit only mild effects on religation.
Wild type topoisomerase and mutant proteins R130A, R130K, G132S, G132A, Y136S, Y136A, K167A and K167R were expressed in bacteria and purified from soluble bacterial lysates by phosphocellulose column chromatography as described (5 ,9 ,10 ). The topoisomerase polypeptide constituted the major species in each protein preparation, as determined by SDS-PAGE, and the extents of purification were essentially equivalent (5 ,9 ,10 ). The protein concentrations of the topoisomerase preparations were determined by using the dye-binding method (BioRad) with bovine serum albumin as the standard.
A 60mer oligonucleotide containing a centrally placed CCCTT element was 5'-end-labeled by enzymatic phosphorylation in the presence of [[gamma]-32P]ATP and T4 polynucleotide kinase, then gel-purified, and annealed to an unlabeled complementary 60mer strand. Reaction mixtures containing (per 20 [mu]l) 50 mM Tris-HCl (pH 7.5), 0.3 pmol of 60mer DNA duplex and 38-76 ng of topoisomerase were incubated at 37oC. The reactions were initiated by the addition of topoisomerase to prewarmed reaction mixtures. Aliquots (20 [mu]l) were withdrawn at various times and the reaction was quenched immediately by adding SDS to 0.5%. The samples were then digested for 60 min at 45oC with 10 [mu]g of proteinase K. The volume was adjusted to 100 [mu]l and the digests were then extracted with an equal volume of phenol/chloroform. DNA was recovered from the aqueous phase by ethanol precipitation. The pelleted material was resuspended in formamide and the samples were electrophoresed through a 17% polyacrylamide gel containing 7 M urea in TBE (90 mM Tris-borate, 2.5 mM EDTA). The cleavage product, a 32P-labeled 30mer bound to a short peptide, was well-resolved from the input 60mer substrate. The extent of strand cleavage was quantitated by scanning the wet gel using a FUJIX BAS1000 Bio-Imaging Analyzer. A plot of the percent of input DNA cleaved versus time established endpoint values for cleavage. The data were normalized to the endpoint values and kobs for approach to equilibrium was determined by fitting the data to the equation (100 - %Clnorm) = 100e-kt. The cleavage equilibrium constant (Kcl) is defined as the ratio of covalently bound DNA to noncovalently bound DNA at the reaction endpoint under conditions of saturating enzyme and was calculated according to the equation Kcl = %Cl/(100 - %Cl). Enzyme titration experiments established that the conditions used for measuring approach to equilibrium were saturating with respect to input topoisomerase. The experimental values for kobs and Kcl were used to calculate the rate constants for the cleavage and religation steps (kcl and krel) according to equations Kcl = kcl/krel and kobs = kcl + krel.
Equilibrium cleavage reaction mixtures containing (per 18 [mu]l) 0.3 pmol of the 60 bp DNA and 38-76 ng of topoisomerase were incubated at 37oC for a period sufficient to attain equilibrium. At this time (time zero), an aliquot (18 [mu]l) was removed and quenched with SDS. The reaction mixtures were then adjusted to 0.5 M NaCl and aliquots (20 [mu]l) were taken at various times thereafter. The samples were digested with proteinase K and processed for electrophoresis as described in the preceding section. The decrease in the abundance of the covalent complex (expressed as percent of input DNA) was plotted as a function of time. The data were normalized to the time zero and endpoint values and krel was determined by fitting the data to the equation %Clnorm = 100e-kt.
An 18mer CCCTT-containing DNA oligonucleotide was 5'-32P-labeled, then gel-purified and hybridized to a complementary 30mer strand. Cleavage reaction mixtures containing (per 20 [mu]l) 50 mM Tris-HCl (pH 7.5), 0.3 pmol of 18mer/30mer DNA and 38-76 ng of topoisomerase were incubated at 37oC. The cleavage reactions were initiated by the addition of topoisomerase to prewarmed reaction mixtures. Aliquots (20 [mu]l) were withdrawn at various times and the reaction was quenched immediately by adding SDS to 1%. The samples were electrophoresed through a 10% polyacrylamide gel containing 0.1% SDS. Free DNA migrated near the bromophenol blue dye front. 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 5'-32P-labeled oligonucleotide that was transferred to protein) was quantitated by scanning the dried gel. A plot of the percent of input DNA cleaved versus time established endpoint values for cleavage. The data were normalized to the endpoint values and kcl was determined by fitting the data to the equation (100 - %Clnorm) = 100e-kt.
Cleavage reaction mixtures containing (per 14 [mu]l) 0.3 pmol of the 18mer/30mer DNA and 38 ng of topoisomerase were incubated at 37oC to form the suicide intermediate. Religation was initiated by the simultaneous addition of NaCl to 0.5 M and a 5' hydroxyl-terminated 18mer acceptor strand (5'-ATTCCGATAGTGACTACA) to a concentration of 30 pmol/16 [mu]l (i.e., a 100-fold molar excess over the input DNA substrate). Aliquots (16 [mu]l) were withdrawn at the times indicated and quenched immediately in an equal volume of buffer containing 2% SDS, 76% formamide and 20 mM EDTA. A time zero sample was withdrawn prior to addition of the acceptor strand. The samples were heat-denatured and then electrophoresed through a 17% polyacrylamide containing 7 M urea in TBE. Religation of the covalently bound 12mer strand to the 18mer acceptor DNA will yield a 5'-32P-labeled 30mer strand transfer product. The extent of religation (expressed as the percent of the input labeled 18mer strand recovered as 30mer) was plotted as a function of reaction time. The data were normalized to the endpoint values and krel was determined by fitting the data to the equation (100 - %Re norm) = 100e-kt.
Reaction mixtures (20 [mu]l) contained 50 mM Tris-HCl (pH 7.5), 0.3 pmol of 60 bp duplex DNA (5'-32P-labeled on the scissile strand) and topoisomerase as specified. The reaction mixtures were incubated at 37oC for 5 min. Glycerol was added to 10% and the samples were electrophoresed through a 6% native polyacrylamide gel in 0.25* TBE (22.5 mM Tris-borate, 0.6 mM EDTA) at 100 V for 2 h. Topoisomerase-DNA complexes of retarded mobility were visualized by autoradiographic exposure of the dried gel. The extent of protein-DNA complex formation, expressed as [Bound DNA/(Bound DNA + Free DNA)] * 100, was quantitated by scanning the gel with a Bio-Imaging Analyzer.
The 314-amino acid vaccinia topoisomerase consists of three protease-resistant structural domains demarcated by protease-sensitive interdomain segments referred to as the bridge and hinge (Fig. 1 ) (13 ,14 ). The amino acid sequence in the vicinity of the hinge region (vaccinia residues 126-168) is conserved at 21/43 positions in other cellular and viral type IB topoisomerases (Fig. 1 ). Four residues that are essential for DNA relaxation are located within this segment of the vaccinia enzyme: Arg-130, Gly-132, Tyr-136 and Lys-167. The essential residues are shown in shaded boxes in Figure 1 . Our initial studies indicated that replacement of Arg-130 with alanine or lysine abolished activity in DNA relaxation (9 ). Substitution of Gly-132 by Ser caused a 100-fold decrement in relaxation activity, as did replacement of Tyr-136 with either Asp or Ala (5 ,9 ). Replacement of Lys-167 by either Ala or Glu reduced topoisomerase activity by more than two orders of magnitude, but substitution by arginine had less severe effects (10 ). Preliminary studies suggested that all these mutants proteins were defective in formation of the covalent topoisomerase-DNA intermediate, as determined by the yield of covalent adduct formed by the mutant protein on a short DNA duplex containing a single CCCTT <=> cleavage site for the vaccinia topoisomerase. We have since developed substrates and assays suitable for the determination of the kinetic and equilibrium parameters of the transesterification reactions under single-turnover and equilibrium conditions (8 ,11 ,15 ,16 ). This has permitted us to quantitate the catalytic contribution of several functional groups on the enzyme to the estimated 109-1012-fold enhancement of the rate of transesterification by topoisomerase (8 ,11 ,15 ). We have now applied these methods to assess the roles played by vaccinia residues Arg-130, Gly-132, Tyr-136 and Lys-167.
The results presented in this study enhance our understanding of the vaccinia topoisomerase in three respects: (i) they provide a quantitative assessment of the contribution of four essential amino acids located within and immediately flanking the hinge region; (ii) they illuminate a clear distinction between functional groups required for transesterification chemistry in general and those required specifically for the cleavage reaction; and (iii) they suggest a model for how Arg-130, Lys-167 and other essential residues act at the active site to promote catalysis.
Mutational rate effects on the religation reaction can be viewed as the simplest measure of the contribution of a given amino acid residue to transesterification reaction chemistry. This is because single-turnover religation entails no site-recognition step, i.e., the topoisomerase is already bound covalently to the DNA. Based on the effects of the R130A and R130K mutations on single-turnover strand closure by the covalent intermediate, we conclude that Arg-130 enhances the rate of transesterification by a factor of 105. This is quantitatively similar to the 105 contribution of another basic residue, Arg-223, to transesterification chemistry (11 ). The magnitude of the mutational effects at Arg-130 and Arg-223 argues that both arginine residues interact directly with the scissile phosphate. A key distinction between the two essential arginines is that Arg-130 cannot be replaced by lysine (this study), whereas Arg-223 can be changed to lysine with relatively little catalytic cost (11 ).
The religation reaction entails an attack by the 5' hydroxyl moiety of the DNA strand on the DNA-(3'-phosphotyrosyl)-protein intermediate. This is likely to procede through a pentacoordinate phosphorane transition state in which the attacking group (the 5' OH) and the leaving group (Tyr-274) are positioned apically. We speculate that Arg-130 makes bivalent hydrogen bond interactions with the phosphate oxygens in the transition state. The same bidentate contacts would stabilize the transition state during the forward cleavage reaction, which involves attack by Tyr-274 and expulsion of a 5' hydroxyl-terminated DNA. Lysine at position 130 would be less effective in transition state stabilization in this scheme because it would contact only one of the scissile phosphate oxygens. The side chain of Arg-223 might make an essential contact with one phosphate oxygen in the transition state. This monovalent interaction could potentially be sustained by lysine, which may account for the mild effects of the R223K mutation on catalysis. Lys-167 enhances the rates of transesterification by a factor of 103-104 and an arginine at this position can only partially restore function. Lys-167 may interact with the bridging 5' phosphate oxygen of the leaving DNA strand and serve as a general acid to promote its expulsion during the cleavage reaction (16 ).
Vaccinia residues Arg-130, Lys-167 and Arg-223 are conserved in other type IB topoisomerases. Hence, we suspect that these side chains play the same role during catalysis by the cellular and viral enzymes. Although the cellular topoisomerases have not been mutagenized as extensively as the vaccinia enzyme, the limited data available are consistent with a common structural basis for transesterification. For example, Jensen and Svejstrup (18 ) mutated the residues corresponding to Arg-130 and Lys-167 and concluded that they are essential for DNA relaxation. Yet, the deleterious changes they introduced into the human enzyme (Arg -> Gln and Lys -> Glu) did not reveal the underlying structure-function relationships (e.g., Lys -> Glu entails charge inversion). Levin et al. (19 ) reported that a yeast Top1 mutant containing an Arg -> Lys mutation at the position corresponding to vaccinia Arg-130 was defective in relaxing supercoiled DNA.
The biased effects of the Gly-132 and Tyr-136 mutations on cleavage suggest that a post-binding, pre-chemical step may apply during the cleavage reaction, which is not pertinent (or at least not rate-limiting) during religation. Our results sound a note of caution in drawing conclusions about the structural requirements for topoisomerase reaction chemistry on the basis of mutational effects in a strand cleavage assay. The mutational effects at Gly-132 and Tyr-136 on krel are too modest to meet the criteria we have set for essentiality in transesterification.
Although the precise nature of the pre-cleavage steps remains unclear, the location of the Gly-132 and Tyr-136 residues within the interdomain hinge suggests a post-binding alteration of the protein-DNA interface. The hinge is highly susceptible to proteolysis when topoisomerase is free in solution; in fact, Tyr-136 is the principal site of cleavage by chymotrypsin. The hinge becomes protected from proteolysis when vaccinia topoisomerase is bound to DNA (13 ). Protection of the hinge does not require covalent adduct formation, indicating that it occurs prior to transesterification. The simple interpretation is that the hinge makes contact with the DNA ligand and that bound DNA excludes access by proteases. However, additional residues adjacent to the hinge become more accessible to proteolysis in the DNA-bound state, which raises the possibility of a conformational change upon DNA binding (13 ). With this in mind, we speculate that hinge dynamics are involved in activating the noncovalent protein-DNA complex for cleavage, e.g., by properly orienting the catalytically essential residues flanking the hinge (Arg-130 and Lys-167) with respect to the scissile phosphate.
The mutational effects at Tyr-136 suggest an important hydrogen bonding interaction. The aromatic moiety is clearly not critical, because activity is preserved when Tyr-136 is replaced by serine. In addition to the requirement for a hydroxyl at this position, it is apparently critical that the functional group be constrained by the polypeptide backbone. We showed previously that a single interruption of the polypeptide backbone by chymotrypsin cleavage between Tyr-136 and Leu-137 elicited effects similar to those described here for the Y136A mutation (13 ). The amino and carboxyl fragments of the singly nicked wild type topoisomerase remained physically associated; the nicked protein bound to CCCTT-containing DNA, but was incapable of catalyzing strand cleavage.
The importance of the conformation of the hinge during strand cleavage is also suggested by the effects of Gly-132 mutations. Introduction of the methyl group in place of the glycine hydrogen may elicit a local conformational distortion (perhaps affecting nearby essential residue Arg-130) that in turn affects the pre-cleavage activation step. We consider it unlikely that Gly-132 mutations cause a global disruption of protein conformation, given that noncovalent DNA binding is preserved and religation chemistry is affected only modestly by the G132A and G132S changes. Vaccinia residue Gly-132 is invariant in every other type IB topoisomerase.
Integrases and other recombinase family members catalyze the same transesterification reaction as the type IB topoisomerases, whereby attack by an active site tyrosine on one DNA strand results in the formation of a DNA-(3'-phosphotyrosyl)-protein intermediate. Functional similarities between vaccinia topoisomerase and the recombinases are probable, in so far as vaccinia topoisomerase promotes sequence-specific recombination in vivo and catalyzes site-specific resolution of Holliday junctions in vitro (22 -24 ). Although recombinases and type IB topoisomerases have no apparent overall sequence similarity, they may well have converged with respect to the functional groups at their active sites. The recent reports of the crystal structures of the catalytic domains of bacteriophage lambda and HP1 integrases reveal two essential arginines and an essential histidine clustered at the active sites of both proteins (20 ,21 ).
We speculate that the Arg-130, Lys-167, Arg-223 triad of essential residues of the vaccinia topoisomerase is functionally analogous to the conserved Arg-His-Arg triad that defines the recombinase family. Further, we suggest that the catalytically essential His-265 of vaccinia topoisomerase (8 ), which is situated nine amino acids upstream of the active site tyrosine in all six known poxvirus topoisomerases, is the functional equivalent of a conserved histidine moiety present at the identical position relative to the active site tyrosines of lambda integrase, HP1 integrase and XerD recombinase (Fig. 7 ). Indeed, the segment proximal to the active site is remarkably well conserved between the poxvirus topoisomerases and the recombinases; particularly the (V,I,L)-(V,I,L)-G-H-(T,S) motif, which includes vaccinia His-265 (Fig. 7 ). In the crystal structures of lambda and HP1 integrases, the histidine residue corresponding to vaccinia His-265 is situated at the active site pocket (21 ,22 ).
In summary, in this and other recent reports (8 ,11 ), we have defined the catalytic contributions of essential amino acid side chains of vaccinia topoisomerase, most of which are conserved in other type IB enzymes. We propose that Arg-130, Lys-167, Arg-223 and His-265 act directly on the scissile phosphate during catalysis. We speculate that vaccinia topoisomerase is related mechanistically and structurally to the catalytic domains of site-specific recombinases. Confirmation and further clarification of the topoisomerase I reaction mechanism will hinge on crystallization of topoisomerase in the DNA-bound state.
This work was supported by NIH grant GM46330.
*To whom correspondence should be addressed. Tel: +1 212 639 7145; Fax: +1 212 717 3623; Email: s-shuman@ski.mskcc.org
REFERENCES