Portal de Periódicos da CAPES

Catalysis of ATP Hydrolysis by Two NH2-terminal Fragments of Yeast DNA Topoisomerase II

1999; Elsevier BV; Volume: 274; Issue: 31 Linguagem: Inglês

10.1074/jbc.274.31.21688

1083-351X

Stéphane Olland, James C. Wang,

Synthesis and bioactivity of alkaloids

Catalysis of ATP hydrolysis by two NH2-terminal fragments of yeast DNA topoisomerase II was studied in the absence and presence of DNA, and in the absence and presence of inhibitor ICRF-193. The results indicate that purified Top2-(1–409), a fragment containing the NH2-terminal 409 amino acids of the yeast enzyme, is predominantly monomeric, with a low level of ATPase owing to weak association of two monomers to form a catalytically active dimer. The ATPase activity of Top2-(1–409) is independent of DNA in a buffer containing 100 mm NaCl, in which intact yeast DNA topoisomerase II exhibits robust DNA-dependent ATPase and DNA transport activities. Purified Top2-(1–660), a fragment containing the NH2-terminal 660 amino acid of the yeast enzyme, appears to be dimeric in the absence or presence of DNA, and the ATPase activity of the protein is significantly stimulated by DNA. These results are consistent with a model in which binding of an intact DNA topoisomerase II to DNA places the various subfragments of the enzyme in a way that makes the intramolecular dimerization of the ATPase domains more favorable. We believe that this alignment of subfragments is mainly achieved through the binding of the enzyme to the DNA segment within which the enzyme makes transient breaks. The ATPase activity of Top2-(1–409) is inhibited by ICRF-193, suggesting that the bisdioxopiperazine class of DNA topoisomerase II inhibitors directly interacts with the paired ATPase domains of the enzyme. Catalysis of ATP hydrolysis by two NH2-terminal fragments of yeast DNA topoisomerase II was studied in the absence and presence of DNA, and in the absence and presence of inhibitor ICRF-193. The results indicate that purified Top2-(1–409), a fragment containing the NH2-terminal 409 amino acids of the yeast enzyme, is predominantly monomeric, with a low level of ATPase owing to weak association of two monomers to form a catalytically active dimer. The ATPase activity of Top2-(1–409) is independent of DNA in a buffer containing 100 mm NaCl, in which intact yeast DNA topoisomerase II exhibits robust DNA-dependent ATPase and DNA transport activities. Purified Top2-(1–660), a fragment containing the NH2-terminal 660 amino acid of the yeast enzyme, appears to be dimeric in the absence or presence of DNA, and the ATPase activity of the protein is significantly stimulated by DNA. These results are consistent with a model in which binding of an intact DNA topoisomerase II to DNA places the various subfragments of the enzyme in a way that makes the intramolecular dimerization of the ATPase domains more favorable. We believe that this alignment of subfragments is mainly achieved through the binding of the enzyme to the DNA segment within which the enzyme makes transient breaks. The ATPase activity of Top2-(1–409) is inhibited by ICRF-193, suggesting that the bisdioxopiperazine class of DNA topoisomerase II inhibitors directly interacts with the paired ATPase domains of the enzyme. The type II DNA topoisomerases are enzymes that catalyze the ATP-dependent transport of one DNA double helix through another (1Brown P.O. Peebles C.L. Cozzarelli N.R. Proc. Natl. Acad. Sci. U. S. A. 1979; 76: 6110-6119Crossref PubMed Scopus (116) Google Scholar, 2Liu L.F. Liu C.C. Alberts B.M. Cell. 1980; 19: 697-707Abstract Full Text PDF PubMed Scopus (389) Google Scholar). All forms of life are known to contain one or more of these enzymes, and they participate in vital cellular processes including chromosome condensation and segregation, and the maintenance of genome stability (reviewed in Ref. 3Wang J.C. Annu. Rev. Biochem. 1996; 65: 635-692Crossref PubMed Scopus (2054) Google Scholar). The biological importance of these enzymes is further illustrated by the identification of a large number of natural toxins and antibacterial and antitumor agents that target these enzymes (4Chen A.Y. Liu L.F. Annu. Rev. Pharmacol. Toxicol. 1994; 34: 191-218Crossref PubMed Scopus (777) Google Scholar, 5Maxwell A. Trends Microbiol. 1997; 5: 102-109Abstract Full Text PDF PubMed Scopus (315) Google Scholar, 6Nitiss J.L. Beck W.T. Eur. J. Cancer. 1996; 32A: 958-966Abstract Full Text PDF PubMed Scopus (181) Google Scholar, 7Andoh T. Ishida R. Biochim. Biophys. Acta. 1998; 1400: 155-171Crossref PubMed Scopus (243) Google Scholar). The transport of one DNA duplex through another by a type II DNA topoisomerase is strongly dependent on ATP (8Mizuuchi K. O'Dea M.H. Gellert M. Proc. Natl. Acad. Sci. U. S. A. 1978; 75: 5960-5963Crossref PubMed Scopus (198) Google Scholar, 9Liu L.F. Liu C.C. Alberts B.M. Nature. 1979; 281: 456-461Crossref PubMed Scopus (162) Google Scholar, 10Miller K.G. Liu L.F. Englund P.T. J. Biol. Chem. 1981; 256: 9334-9339Abstract Full Text PDF PubMed Google Scholar). Extensive biochemical and structural studies of these enzymes in recent years have led to a molecular model on this dependence (Refs. 11Roca J. Wang J.C. Cell. 1994; 77: 609-616Abstract Full Text PDF PubMed Scopus (246) Google Scholar and 12Berger J.M. Gamblin S.J. Harrison S.C. Wang J.C. Nature. 1996; 379: 225-232Crossref PubMed Scopus (739) Google Scholar; for a recent review, see Ref. 13Wang J.C. Q. Rev. Biophys. 1998; 31: 107-144Crossref PubMed Scopus (296) Google Scholar). In this model, the enzyme is a dimer with two pairs of jaws that form two molecular ports (gates). The pair of ATPase domains of the enzyme form one port of the enzyme, termed the N-gate because of the NH2-terminal location of the ATPase domain in the single polypeptide eukaryotic enzyme. The second port of the protein is formed by regions closer to the COOH termini of the pair of polypeptides in a homodimeric eukaryotic enzyme, and hence has been denoted the C-gate. In Fig. 1, an enzyme molecule bound to a duplex DNA segment, termed the gate- or G-segment, is sketched to illustrate the locations of of the various parts of the enzyme-DNA complex. It is believed that the primary event following the binding of ATP to a G-segment bound enzyme is the intramolecular dimerization of the pair of ATPase domains to close the N-gate. The closure of the N-gate in turn triggers a conformational cascade in the enzyme-DNA complex. First, the enzyme-linked broken DNA ends, formed by transesterification between a pair of enzyme tyrosyl residues and a pair of DNA phosphoryl groups in the G-segment, are pulled apart to open a gate in the DNA. The T-segment is then transported through the transiently opened DNA gate into the large hole between the G-segment and the C-gate. Following these steps, the enzyme-linked ends of the G-segment retract toward each other to close the DNA gate, and this retraction in turn reduces the size of the large hole and forces out the T-segment through the transiently opened C-gate. Finally, the N-gate reopens after ATP hydrolysis and/or the release of the hydrolytic products, and the enzyme is set for another round of reaction. There is strong evidence that the binding of ATP to a type II DNA topoisomerase leads to intramolecular dimerization of the pair of ATPase domains (14Wigley D.B. Davies G.J. Dodson E.J. Maxwell A. Dodson G. Nature. 1991; 351: 624-629Crossref PubMed Scopus (484) Google Scholar, 15Ali J.A. Jackson A.P. Howells A.J. Maxwell A. Biochemistry. 1993; 32: 2717-2724Crossref PubMed Scopus (308) Google Scholar, 16Kampranis S.C. Maxwell A. J. Biol. Chem. 1998; 273: 22606-22614Abstract Full Text Full Text PDF PubMed Scopus (53) Google Scholar, 17Roca J. Berger J.M. Harrison S.C. Wang J.C. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 4057-4062Crossref PubMed Scopus (161) Google Scholar), which brings together key residues in both halves of the enzyme for the catalysis of ATP hydrolysis (14Wigley D.B. Davies G.J. Dodson E.J. Maxwell A. Dodson G. Nature. 1991; 351: 624-629Crossref PubMed Scopus (484) Google Scholar). It is also well established that the ATPase activity of a type II DNA topoisomerase is stimulated by the presence of DNA (8Mizuuchi K. O'Dea M.H. Gellert M. Proc. Natl. Acad. Sci. U. S. A. 1978; 75: 5960-5963Crossref PubMed Scopus (198) Google Scholar, 9Liu L.F. Liu C.C. Alberts B.M. Nature. 1979; 281: 456-461Crossref PubMed Scopus (162) Google Scholar, 10Miller K.G. Liu L.F. Englund P.T. J. Biol. Chem. 1981; 256: 9334-9339Abstract Full Text PDF PubMed Google Scholar). How the binding of a DNA stimulates the ATPase activity is, however, less clear. In the simplest view, the binding of an enzyme to a DNA G-segment facilitates the dimerization of its pair of ATPase domains and thus enhances its ATPase activity. Alternatively, the ATPase activity of the enzyme may depend on its interactions with the G- as well as the T-segment. The involvement of the T-segment in the catalysis of ATP hydrolysis by a type II DNA topoisomerase has been implicated in a number of studies. First, a mutation replacing Arg-286 of the Escherichia coligyrase B-subunit by glutamine was found to have no effect on the DNA-independent ATPase activity of gyrase but abolish the stimulation of the ATPase activity by DNA (18Tingey A.P. Maxwell A. Nucleic Acids Res. 1996; 24: 4868-4873Crossref PubMed Scopus (75) Google Scholar). The location of Arg-286 in the crystal structure of a 43-kDa NH2-terminal fragment ofE. coli GyrB protein suggests that this residue may be involved in T-segment binding, and hence the lack of DNA-dependent stimulation of the mutant gyrase ATPase may reflect a role of the T-segment in the catalysis of ATP hydrolysis (18Tingey A.P. Maxwell A. Nucleic Acids Res. 1996; 24: 4868-4873Crossref PubMed Scopus (75) Google Scholar). Second, stimulation of the E. coli gyrase ATPase by DNA was found to depend on the length of the DNA (19Maxwell A. Gellert M. J. Biol. Chem. 1984; 259: 14472-14480Abstract Full Text PDF PubMed Google Scholar): for each enzyme molecule, a single DNA molecule longer than 100 bp 1The abbreviations used are: bp, base pairs.1The abbreviations used are: bp, base pairs. would suffice but two DNA molecules shorter than 70 bp would be required. It was therefore suggested that ATPase stimulation involved DNA binding to two separate sites in the enzyme, plausibly for the binding of the G and T segments. A long DNA fragment could presumably form a loop to contact both sites, but two short DNA fragments would be needed to do the same (19Maxwell A. Gellert M. J. Biol. Chem. 1984; 259: 14472-14480Abstract Full Text PDF PubMed Google Scholar). As the authors pointed out, however, interpretation of these results is not straightforward (18Tingey A.P. Maxwell A. Nucleic Acids Res. 1996; 24: 4868-4873Crossref PubMed Scopus (75) Google Scholar, 19Maxwell A. Gellert M. J. Biol. Chem. 1984; 259: 14472-14480Abstract Full Text PDF PubMed Google Scholar). Because of allosteric changes in the various reaction steps of a type II DNA topoisomerase, it is uncertain whether the R286Q mutation affects T-segment binding or influences an allosteric conformational change triggered by G-segment binding. The DNA length dependence of the gyrase ATPase is suggestive of two DNA-binding sites, but both sites could be involved in G-segment binding (18Tingey A.P. Maxwell A. Nucleic Acids Res. 1996; 24: 4868-4873Crossref PubMed Scopus (75) Google Scholar). When a similar experiment on the dependence of the ATPase activity on DNA length was carried out with human DNA topoisomerase IIα, the rate of ATP hydrolysis at a molar ratio of two DNA molecules per enzyme dimer showed a sharp spike at a DNA length of about 340 bp, illustrating the complexity of this dependence (20Hammonds T.R. Maxwell A. J. Biol. Chem. 1997; 272: 32696-32703Abstract Full Text Full Text PDF PubMed Scopus (49) Google Scholar). Several studies were also carried out with various type II DNA topoisomerase fragments. Early studies with the GyrB protein ofE. coli showed little effect of DNA on its ATPase in the absence of the GyrA protein (8Mizuuchi K. O'Dea M.H. Gellert M. Proc. Natl. Acad. Sci. U. S. A. 1978; 75: 5960-5963Crossref PubMed Scopus (198) Google Scholar, 19Maxwell A. Gellert M. J. Biol. Chem. 1984; 259: 14472-14480Abstract Full Text PDF PubMed Google Scholar, 21Sugino A. Higgins N.P. Brown P.O. Peebles C.L. Cozzarelli N.R. Proc. Natl. Acad. Sci. U. S. A. 1978; 75: 4838-4842Crossref PubMed Scopus (324) Google Scholar). The ATPase activity of a 43-kDa NH2-terminal fragment of GyrB was also found to be unaffected by DNA (15Ali J.A. Jackson A.P. Howells A.J. Maxwell A. Biochemistry. 1993; 32: 2717-2724Crossref PubMed Scopus (308) Google Scholar, 22Ali J.A. Orphanides G. Maxwell A. Biochemistry. 1995; 34: 9801-9808Crossref PubMed Scopus (89) Google Scholar). More recently, study of a renatured 50-kDa fragment of human DNA topoisomerases IIα containing amino acids 1–435 of the intact enzyme showed that it differed from the 43-kDaE. coli GyrB fragment in two major aspects. First, whereas the 43-kDa GyrB fragment is monomeric, the 50-kDa human enzyme fragment is dimeric. Second, unlike the GyrB ATPase, the ATPase activity of the 50-kDa human protein is stimulated by 5–10-fold in the presence of DNA. This stimulation was attributed to a plausible involvement of the T-segment (23Gardiner L.P. Roper D.I. Hammonds T.R. Maxwell A. Biochemistry. 1998; 37: 16997-17004Crossref PubMed Scopus (30) Google Scholar). In the present study, we report measurements of the ATPase activities of two NH2-terminal fragments of yeast DNA topoisomerase II containing amino acid residues 1–409 and 1–660, both of which can be overexpressed in a soluble form and thus required no unfolding and renaturation prior to ATPase measurements. The effect of an inhibitor of eukaryotic DNA topoisomerase II, a bisdioxopiperazine ICRF-193, on the ATPase activities of these fragments is also reported. We infer from these results that DNA stimulates the ATPase activity of a type II DNA topoisomerase by facilitating intramolecular dimerization of the pair of ATPase domains of the enzyme. The binding of an enzyme to a DNA G-segment is likely to bring various structural domains of the enzyme into positions, such that the ATPase domains are correctly aligned to contact each other upon ATP binding, to form a pair of active sites for the hydrolysis of ATP. Whether a DNA T-segment is involved in ATP hydrolysis is unclear; it appears that ATP hydrolysis by the enzyme does not involve direct contact between DNA and the ATPase domains of the enzyme. We also show that a DNA topoisomerase II-targeting antitumor agent ICRF-193 inhibits the ATPase of yeast Top2-(1–409), indicating that bisdioxopiperazines of the ICRF-193 class (7Andoh T. Ishida R. Biochim. Biophys. Acta. 1998; 1400: 155-171Crossref PubMed Scopus (243) Google Scholar) are likely to interact primarily with the dimer form of the ATPase domain of eukaryotic DNA topoisomerase II. This conclusion differs from the one drawn from previous experiments with a 50-kDa human DNA topoisomerase II NH2-terminal fragment (23Gardiner L.P. Roper D.I. Hammonds T.R. Maxwell A. Biochemistry. 1998; 37: 16997-17004Crossref PubMed Scopus (30) Google Scholar). The expression vector pESP-1 (Stratagene) was used in the construction of plasmids for the overexpression of the NH2-terminal fragments ofSaccharomyces cerevisiae DNA topoisomerase II inSchizosaccharomyces pombe. The coding sequence of Top2-(1–409) was synthesized by the polymerase chain reaction, using 5′-GGGAATTCCATATGTCAACTGAACCGGTAAGCGCC-3′ as the 5′ primer and 5′-TACGGATCCTAGCTAGCTTCTTCATTTGCGTCGGCAATTTCAAACATTCT-3′ as the 3′ primer. The use of these primers introduced anNdeI site at the ATG start of the TOP2 gene and an NheI and a BamHI site near the other end of the polymerase chain reaction product (positions of these sites are underlined in the primer sequences). The polymerase chain reaction-amplified DNA was cloned in between the NdeI and BamHI sites of pUC19, and a synthetic oligonucleotide was used to introduce a hexahistidine tag between the NheI and BamHI sites of the cloned DNA segment such that the open reading frame contained codons for the first 409 amino acid residues ofS. cerevisiae DNA topoisomerase II plus Gly-Ser-(His)6-Stop (codons for Gly-Ser were introduced by the addition of the NheI site GCTAGC). The NdeI to BamHI fragment was then mixed with two segments derived from pESP-1, a BstEII to BamHI fragment and aBstEII to NdeI fragment, and treated with DNA ligase to give the desired plasmid for overexpression of Top2-(1–409). The overexpression clone for Top2-(1–660) was similarly constructed, using the oligonucleotide 5′-TACGGATCCTAGCTAGCCCCGGGTTCGTATTGTCTCAGCCATTCTTTACG-3′ instead of the 3′ primer described above for the synthesis of the Top2-(1–409) coding region by polymerase chain reaction. In the final construct, the first 660 codons of S. cerevisiae are followed by Ser-(His)6-Stop. Similar clones were also constructed for induced expression of the proteins in E. coli. Although overexpression was readily achieved, the proteins were mostly in an insoluble form and were not used in the studies reported here. Transfection, growth, and harvest ofS. pombe cells were carried out according to the supplier of the pESP-1 vector. In the preparation of each protein, cells from 10 liters of culture were pelleted by centrifugation, resuspended in 150 ml of a lysis buffer (20 mm Tris·HCl, pH 8, 500 mm NaCl, 5 mm imidazole, 0.25% Triton X-100, and 1 μg/ml each of leupeptin and pepstatin), and disrupted by blending with about equal volume of glass beads in a "Bead-beater" (Biospec Products). The lysate was cleared by centrifugation, and loaded on about 15-ml bed volume of Ni-resin (Novagen). After washing successively with 15 column volumes of the lysis buffer and 15 volumes of a washing buffer (20 mm Tris·HCl, pH 8, 100 mm NaCl, and 20 mm imidazole), the protein was eluted with a buffer containing 20 mm Tris·HCl, pH 8, 100 mm NaCl, and 200 mm imidazole. The eluate (20 ml) was immediately loaded on a heparin column (3-ml bed volume) pre-equilibrated with 20 mm Tris·HCl, pH 8, 150 mm NaCl, and 5% glycerol, rinsed with 15 volumes of the same buffer and eluted with 20 mm Tris·HCl, pH 8, 400 mm NaCl, and 5% glycerol. Peak fractions of the overexpressed protein were pooled, concentrated by centrifugation in a Centricon 30 concentrator (Amicon) to about 20 mg/ml in a storage buffer containing 20 mm Tris·HCl, pH 8, 400 mm NaCl, and 15% glycerol. The concentrated protein was divided into small aliquots, flash-frozen in liquid nitrogen, and stored at −80 °C. The typical yield from a 10-liter culture was about 15 mg of Top2-(1–409) or 8 mg of Top2-(1–660). In the spectrophotometric assay using an ATP regeneration system that quantitatively couples NADH oxidation to the conversion of ADP back to ATP (24Morrical S.W. Lee J. Cox M.M. Biochemistry. 1986; 25: 1482-1494Crossref PubMed Scopus (123) Google Scholar), each reaction mixture (1 ml) contained 20 mm Tris·HCl, pH 7.4, 8 mmMgCl2, 2.5 mm dithiothreitol, 0.5 mg/ml bovine serum albumin, 1 mm ATP, 0.1 mm NADH, 2 mm phosphoenolpyruvate, 3 units of pyruvate kinase, 4 units of lactate dehydrogenase, and varying concentrations of NaCl and DNA as specified in the individual experiments. Reactions were initiated by mixing 0.9 ml of a reaction mixture without ATP and 0.1 ml of 10 mm ATP in the assay buffer, both pre-equilibrated to 30 °C. Decrease in NADH concentration was monitored by absorbance at 340 nm in a Cary 50 spectrophotometer equipped with a thermostated cell holder (Varian). Plasmids for the overexpression of two NH2-terminal fragments of S. cerevisiae DNA topoisomerase II in the fission yeast S. pombe were constructed as described under "Experimental Procedures." The 47-kDa protein fragment Top2-(1–409) contains Met-1 to Glu-409 of the budding yeast enzyme, corresponding to the first 390 amino acid residues of E. coli GyrB. This fragment is known to be generated by cleavage of the intact enzyme with SV8 endoprotease (25Lindsley J.E. Wang J.C. Proc. Natl. Acad. Sci. U. S. A. 1991; 88: 10485-10489Crossref PubMed Scopus (117) Google Scholar). Interestingly, cleavage after Glu-409 is blocked by binding of a nonhydrolyzable ATP analog to the intact enzyme, indicating that the conformation of the peptide around this site is allosterically altered by the intramolecular dimerization of the ATPase domains (25Lindsley J.E. Wang J.C. Proc. Natl. Acad. Sci. U. S. A. 1991; 88: 10485-10489Crossref PubMed Scopus (117) Google Scholar). The 76-kDa fragment Top2-(1–660) contains Met-1 to Gly-660 of the budding yeast enzyme, corresponding to full-length E. coli GyrB protein (26Caron P.R. Wang J.C. Adv. Pharmacol. 1994; 29B: 271-297Crossref PubMed Scopus (98) Google Scholar). For convenience of purification, a hexahistidine tag was fused to the COOH terminus of each of the protein fragments. Both tagged fragments overexpressed well and were readily purified to near homogeneity (Fig. 2). Panel A of Fig. 3 depicts the ATPase activity of yeast Top2-(1–409) in the absence of DNA. In either an assay buffer containing 100 mm NaCl, or the same buffer with 10 instead of 100 mm NaCl, the rate of ATP hydrolysis showed a parabolic dependence on protein concentration. Plotting the rate (in nanomolar of ATP hydrolyzed per second) as the square of protein concentration (in μm2) produced a straight line in both buffers, with a slope of 6.4 and 22.7, respectively, in the 100 and 10 mm NaCl medium (Fig. 3B). These results indicate a monomer-dimer equilibrium of the 409-amino acid fragment and that the ATPase activity of the monomer is much lower than that of the dimer. The parabolic dependence of the ATPase activity on protein concentration also suggests that the monomer is the predominant species over the entire concentration range shown, otherwise the ATPase activity would switch to a linear dependence on protein concentration at the higher concentration end of the curves. The higher ATPase activity of yeast Top2-(1–409) in the 10 mm NaCl buffer can be attributed to either an increase in the monomer-dimer association constant, or an increase of the intrinsic ATPase activity of the dimer, when the salt concentration is lowered; we feel that the former interpretation is the more likely one. Fig. 4 depicts the DNA dependence of the ATPase activity of yeast Top2-(1–409). As shown in panel A, in the 100 mm NaCl buffer the rate of ATP hydrolysis was found to be independent of DNA concentration. In the 10 mmNaCl buffer, however, an increase in the ATPase activity of yeast Top2-(1–409) upon the addition of DNA was observed. At a protein concentration of 1 μm in the 10 mm salt buffer, the ATPase activity increased steeply with increasing DNA concentration, reaching a plateau at a DNA to protein ratio of 40 bp per protein monomer (Fig. 4, panel A). This steep rise in the ATPase activity of Top2-(1–409) with increasing DNA concentration can be attributed to a shift of the protein monomer to dimer equilibrium in favor of the dimer formation in the low salt buffer, owing to the preferential binding of the dimeric species to DNA. The preferential binding of dimeric Top2-(1–409) to DNA in the 10 mm salt buffer is consistent with the linear rather than parabolic dependence of the rate of ATP hydrolysis on enzyme concentration (Fig. 4, panel B). The source and form of DNA showed little effect in the enhancement of ATPase; supercoiled plasmid, linear plasmid, and sonicated salmon sperm DNA were equally effective. No stimulation of the ATPase activity of Top2-(1–409) was seen, however, with tRNA. Results of the ATPase activity of yeast Top2-(1–660) are shown in Fig. 5. Surprisingly, in the absence or presence of DNA, the rate of ATPase hydrolysis is linearly dependent on protein concentration. Thus it appears that even in the absence of DNA and in the presence of ATP the 660-amino acid fragment is dimeric in solution, whereas Top2-(1–409) is largely monomeric under the same conditions. Amino acid residues 409–660 of yeast DNA topoisomerase II constitute the "B′-subfragment" of the enzyme (Fig. 1), which has been observed to form a dimer interface in the crystal structure of a 92-kDa fragment of the yeast enzyme (12Berger J.M. Gamblin S.J. Harrison S.C. Wang J.C. Nature. 1996; 379: 225-232Crossref PubMed Scopus (739) Google Scholar). The presence of the B′-subfragment in yeast Top2-(1–660) may therefore favor dimerization of the fragment. Despite the dimeric behavior of this fragment under the ATPase assay conditions, the addition of DNA to a ratio of 55 bp per protein monomer was found to stimulate the ATPase, by about 4- and 7-fold, respectively, in the 100 and 50 mm salt buffer (Fig. 5, panel B). The nature of this stimulation will be discussed in a later section. When the ATPase activity of yeast Top2-(1–660) at a fixed protein concentration of 0.4 μm was measured as a function of DNA concentration, in the 100 mm salt buffer a largely monotonic increase was seen with increasing DNA concentration, reaching a plateau at higher DNA concentrations (Fig. 6, lower curve). In contrast, in the 50 mm salt medium the increase was not monotonic: a sharp increase with increasing DNA concentration was followed by a drop at still higher DNA concentrations (Fig. 6, upper curve). A number of bisdioxopiperazines with antitumor activities, including ICRF-159 and ICRF-193, are believed to stabilize the closed-clamp conformation of eukaryotic DNA topoisomerase II (7Andoh T. Ishida R. Biochim. Biophys. Acta. 1998; 1400: 155-171Crossref PubMed Scopus (243) Google Scholar, 27Roca J. Ishida R. Berger J.M. Andoh T. Wang J.C. Proc. Natl. Acad. Sci. U. S. A. 1994; 91: 1781-1785Crossref PubMed Scopus (334) Google Scholar). Mammalian cell lines resistant to these compounds have been isolated, and two drug-resistant mutants have been mapped to the ATPase domain of the enzyme (7Andoh T. Ishida R. Biochim. Biophys. Acta. 1998; 1400: 155-171Crossref PubMed Scopus (243) Google Scholar, 28Sehested M. Wessel I. Jensen L.H. Holm B. Oliveri R.S. Kenwrick S. Creighton A.M. Nitiss J.L. Jensen P.B. Cancer Res. 1998; 58: 1460-1468PubMed Google Scholar). These results suggest that the bisdioxopiperazine-binding pocket of the enzyme is located in a dimerized pair of ATPase domains. Studies of interactions between ICRF-159 and Drosophilia DNA topoisomerase II lacking the ATPase domain raised the possibility, however, that the bisdioxopiperazines bind outside this domain (29Chang S. Hu T. Hsieh T.S. J. Biol. Chem. 1998; 273: 19822-19828Abstract Full Text Full Text PDF PubMed Scopus (40) Google Scholar). To test further the mechanism of inhibition of eukaryotic DNA topoisomerase II by the bisdioxopiperazines, we examined the effect of ICRF-193 on the ATPase activity of yeast Top2-(1–409) and Top2-(1–660). In the experiment shown in Fig. 7, the DNA-dependent ATPase activity of yeast Top2-(1–409) was monitored in a spectrophotometric assay in which the formation of ADP from ATP hydrolysis was enzymatically coupled to NADH oxidation (24Morrical S.W. Lee J. Cox M.M. Biochemistry. 1986; 25: 1482-1494Crossref PubMed Scopus (123) Google Scholar). Curve 1 shows the DNA-dependent ATPase activity of the fragment in the absence of ICRF-193. As expected, the concentration of NADH decreased linearly with time until all NADH was exhausted at the end of the reaction. The rate of ATP hydrolysis was calculated to be 178 nm s−1, from the slope of the linear region of the plot, at a total Top2-(1–409) concentration of 0.85 μm (in monomers). In the presence of 0.1 mmICRF-193, however, the kinetics of NADH consumption or ATP hydrolysis by Top2-(1–409) in the presence of DNA is complex (curve 2in Fig. 7). In the first several minutes following the mixing of the various reagents including the drug, the rate of ATP hydrolysis in the presence of the drug was essentially the same as that in the absence of the drug, as indicated by the convergence of curves 1 and 2 of Fig. 7 near the start of the reaction. At longer times, however, the slope of curve 2 progressively decreased from 178 nm s−1 to about 40 nms−1, signifying a gradual reduction in the DNA-dependent ATPase activity of yeast Top2-(1–409) in the presence of ICRF-193. These data indicate that ICRF-193 slowly inhibits the ATPase activity of Top2-(1–409). Preincubation of the protein with ICRF-193 in the absence of ATP did not alter this slow kinetics of inhibition (results not shown), in agreement with previous results that the drug specifically interacts with the "closed-clamp" conformation of DNA topoisomerase II, in which the N-gate is closed by the binding of ATP to the ATPase domains of the dimeric enzyme (27Roca J. Ishida R. Berger J.M. Andoh T. Wang J.C. Proc. Natl. Acad. Sci. U. S. A. 1994; 91: 1781-1785Crossref PubMed Scopus (334) Google Scholar). The relatively slow inhibitory effect of ICRF-193 was also observed when similar experiments were carried out with Top2-(1–660) in the absence or presence of DNA (data not shown). Together, these experiments indicate that the bisdioxopiperazine inhibits the ATPase activity of yeast DNA topoisomerase II, and that the primary site of interaction is most likely within the 409-amino acid ATPase domain. The reversibility of the inhibition of DNA topoisomerase II ATPase activity by ICRF-193 was tested in view of the slow time course of inhibition. Yeast Top2-(1–660), at a concentration of 13.1 μm, was first incubated in the 100 mmsalt reaction buffer containing 3 mm ATP and 0.1 mm ICRF-193 at 30 °C for 1 h. The mixture was then split into two: one was dialyzed at 4 °C against the same buffer containing 0.1 mm ICRF-193 and no ATP, and the other the same buffer without either drug or ATP. The sample dialyzed against 0.1 mm ICRF-193 was then diluted to a final Top2-(1–660) concentration of 0.26 μm into the ATPase assay mixture containing DNA, ATP, and 0.1 mm ICRF-193, and incubated at 30 °C. As shown in Fig. 8, the NADH concentration in this sample decreased linearly with time, with a slope corresponding to an ATP hydrolysis rate of 82 nms−1 (curve 1). The NADH concentration of the other sample initially decreased with the same slope as that of curve 1, but the rate of NADH oxidation accelerated at longer times, approaching a slope corresponding to an ATP hydrolysis rate of 380 nm s−1 after a period of about 25 min (curve 2 of Fig. 7). These results demonstrate that the inhibition of the ATPase activity of the yeast enzyme by ICRF-193 is reversible. The slow rate of ATP hydrolysis near the beginning of the reaction suggests that the protein-bound ICRF-193 remained bound during dialysis at 4 °C, but gradually dissociated during the first 20 min of incubation at 30 °C in the presence of ATP. In the above section, we describe results on the ATPase activities of two NH2-terminal fragments of yeast DNA topoisomerase II, one containing the first 409 and the other the first 660 amino acids of the intact enzyme. Both fragments overexpressed well inS. pombe cells and were readily purified to near homogeneity in a soluble form. For intact yeast DNA topoisomerase II, the ATP-dependent transport of one DNA double helix through another is optimal in a buffer containing more than 100 mm monovalent salt (30Goto T. Laipis P. Wang J.C. J. Biol. Chem. 1984; 259: 10422-10429Abstract Full Text PDF PubMed Google Scholar). It is therefore significant that in the 100 mm salt buffer the ATPase activity of the 47-kDa yeast DNA topoisomerase II fragment Top2-(1–409) is not increased by DNA: under conditions that are optimal for catalysis of ATP hydrolysis by yeast DNA topoisomerase II, interaction between DNA and the ATPase domains of the enzyme spanning amino acids 1–409 appears to be insignificant. The ATPase activity of Top2-(1–409) is significantly affected by the presence of DNA, however, in a buffer with a much reduced monovalent ion concentration. The dependence of the rate of ATP hydrolysis on protein concentration, which is a quadratic function in the absence of DNA, becomes a linear one in the presence of DNA in the low salt buffer, suggesting that the presence of DNA shifts the monomer-dimer equilibrium of the protein fragment toward dimerization. While here it might be tempting to invoke a role of the DNA T-segment in the stimulation of ATP hydrolysis, for the reasons below we believe that such an inference is problematic. First, in this low salt buffer intact yeast DNA topoisomerase II is inactive in transporting a T-segment through a G-segment. Second, these results, while supporting the preferential binding of the dimeric form to DNA in the low salt buffer, provide no information on how the DNA is interacting with the dimer or whether the DNA is binding to the dimer in a way expected for a T-segment. Third, because of the linear and quadratic dependence of ATPase on Top2-(1–409) concentration in the presence and absence of DNA, respectively, the apparent enhancement of ATPase activity by DNA may be misleading. In the 10 mm NaCl buffer and at a Top2-(1–409) concentration close to 5 μm, the presence of DNA at a ratio of about 40 bp of DNA per enzyme monomer stimulates ATPase by less than 2-fold; at even higher protein concentrations, the ATPase activity of the enzyme in the presence of DNA is probably lower than that in the absence of DNA, if the two curves in Fig. 2 can be extrapolated to higher protein concentrations (the limited solubility of the protein in the low salt medium makes it impractical, however, to test this prediction). Fourth, the measured level of the DNA-dependent ATPase of Top2-(1–409) is rather low; the slope of the linear plot gives a rate of 0.2 ATP hydrolyzed per second per Top2-(1–409) or 0.4 ATP hydrolyzed per second per pair of Top2-(1–409), which is comparable in magnitude to the DNA-independent ATPase of intact yeast DNA topoisomerase II but much lower than the turnover number of 5–20 ATP/s per intact dimeric yeast enzyme in the presence of DNA (31Lindsley J.E. Wang J.C. J. Biol. Chem. 1993; 268: 8096-8104Abstract Full Text PDF PubMed Google Scholar). These considerations, together with the observed lack of ATPase stimulation by DNA in the 100 mm salt medium, led us to favor the interpretation that direct interaction between DNA and the ATPase sites of a type II DNA topoisomerase is not important in the catalysis of ATP hydrolysis. For the 76-kDa yeast Top2-(1–660), the rate of ATP hydrolysis is substantially enhanced by DNA, even though the linear dependence of the ATPase activity on protein concentration suggests that under the assay conditions Top2-(1–660) in the absence of DNA is already in a dimeric form. The intrinsic ATPase activity of Top2-(1–660) in the 100 mm salt buffer, estimated from the data in Fig. 4, is about 0.06 ATP hydrolyzed per second per Top2-(1–660) monomer or 0.12 ATP hydrolyzed per second per Top2-(1–660) dimer; the presence of excess DNA raises the rate by about 5-fold to about 0.3 ATP hydrolyzed per second per Top2-(1–660) monomer or 0.6 ATP hydrolyzed per second per Top2-(1–660) dimer. The magnitude of this increase appears significant, especially in view of a total lack of stimulation of the ATPase of Top2-(1–409) by DNA in the same buffer. Structural and biochemical studies suggest that the B′-subfragment in Top2-(1–660) is involved in the binding of the DNA by full-length yeast DNA topoisomerase II (12Berger J.M. Gamblin S.J. Harrison S.C. Wang J.C. Nature. 1996; 379: 225-232Crossref PubMed Scopus (739) Google Scholar, 32Li W. Wang J.C. J. Biol. Chem. 1997; 272: 31190-31195Abstract Full Text Full Text PDF PubMed Scopus (26) Google Scholar, 33Berger J.M. Fass D. Wang J.C. Harrison S.C. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 7876-7881Crossref PubMed Scopus (89) Google Scholar, 34Liu Q. Wang J.C. Proc. Natl. Acad. Sci. U. S. A. 1999; 96: 881-886Crossref PubMed Scopus (65) Google Scholar), and interaction between this subfragment and DNA in the 100 mm salt buffer is presumably responsible for the observed stimulation of the ATPase activity by DNA. The stimulation of the ATPase activity of Top2-(1–660) by DNA also suggests that the DNA-bound Top2-(1–660) dimer may differ in conformation from the unbound dimer, such that the pair of ATPase domains in a DNA-bound Top2-(1–660) can more easily come into contact to form the active ATPase pockets. It has recently been suggested that the pair of B′-subfragments in yeast DNA topoisomerase II may undergo large displacements during a reaction cycle (33Berger J.M. Fass D. Wang J.C. Harrison S.C. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 7876-7881Crossref PubMed Scopus (89) Google Scholar, 34Liu Q. Wang J.C. Proc. Natl. Acad. Sci. U. S. A. 1999; 96: 881-886Crossref PubMed Scopus (65) Google Scholar), and therefore a pair of B′-subfragments in the unbound and DNA-bound fragments may assume different positions relative to other parts of the fragments. The coordinated alignment of various structural domains of a DNA-bound full-length or NH2-terminal fragment of DNA topoisomerase II is probably the key to efficiently catalysis of ATP hydrolysis. Presumably, the relative positions of various structural domains are also modified by additional interactions, such as interactions between the protein and ATP or its hydrolytic products, transesterification between the enzyme and the DNA G-segment, and interactions between the protein and a DNA T-segment. Mutations replacing residues involved in DNA cleavage and rejoining, for example, have been shown to affect the ATPase activity of the enzyme (35Morris S.K. Harkins T.T. Tennyson R.B. Lindsley J.E. J. Biol. Chem. 1999; 274: 3446-3452Abstract Full Text Full Text PDF PubMed Scopus (25) Google Scholar). Interpretation of the effect of DNA on the ATPase activity of the 76-kDa yeast Top2-(1–660) in the 50 mm salt medium is complicated by the rise and fall of the ATPase activity with increasing DNA concentration. The maximal rate occurs around a stoichiometric ratio of about 40 DNA base pairs per Top2-(1–660) monomer or 80 DNA base pairs per Top2-(1–660) dimer. This type of rise and fall in ATPase activity with increasing DNA concentration was previously observed with human and yeast type II DNA topoisomerases (20Hammonds T.R. Maxwell A. J. Biol. Chem. 1997; 272: 32696-32703Abstract Full Text Full Text PDF PubMed Scopus (49) Google Scholar), and a likely interpretation is that the occupation of certain weaker DNA-binding sites of the enzymes may hinder its ATPase activity. In the case of the 76-kDa Top2-(1–660) fragment, a contributing factor of this rise or fall might be that both the dimeric and monomeric form of Top2-(1–660) can bind DNA. Dimerization of the ATPase domains of a pair of monomers, bound to two separate sites along the DNA, would be dependent on both their relative orientation and their separation along the DNA. At high DNA concentrations, monomer binding might be significant and the larger separation between even a properly oriented pair of bound monomers would make dimerization less favorable. Finally, our results on the inhibition of ATP hydrolysis by yeast Top2-(1–409) and Top2-(1–660) by ICRF-193 add strong evidence that the bisdioxopiperazines are interacting primarily with the ATPase domains rather than the B′ fragments. Because the bisdioxopiperazines appear to interact with the dimerized domains (11Roca J. Wang J.C. Cell. 1994; 77: 609-616Abstract Full Text PDF PubMed Scopus (246) Google Scholar), the action of these compounds is presumably manifested after the binding of ATP. We are uncertain about whether these compounds inhibit the ATPase activity of a type II DNA topoisomerase by interfering with the binding, hydrolysis, or the release of the hydrolytic products of ATP. It is interesting to note that even at a very high concentration of ICRF-193, there is a significant residual rate of ATP hydrolysis. It is plausible that by trapping the closed-clamp form of a type II DNA topoisomerase, a bisdioxopiperazine may slow down one or more of these steps. We thank R. Ishida and T. Andoh for the sample of ICRF-193.