ATPase Domain of Eukaryotic DNA Topoisomerase II
2002; Elsevier BV; Volume: 277; Issue: 8 Linguagem: Inglês
10.1074/jbc.m111394200
ISSN1083-351X
AutoresHu Tao, Harvey J. Sage, Tao‐shih Hsieh,
Tópico(s)Bioactive Compounds and Antitumor Agents
ResumoWe have prepared full-lengthDrosophila and human topoisomerase II and truncation constructs containing the amino-terminal ATPase domain, and we have analyzed their biochemical properties. The ATPase activity of the truncation proteins, similar to that of the full-length proteins, is greatly stimulated by the presence of DNA. This activity of the truncation proteins is also sensitive to the inhibition by the drug bisdioxopiperazine, ICRF-193, albeit at a much lower level than the full-length protein. Therefore, bisdioxopiperazine can directly interact with the NH2-terminal ATPase domain, but the drug-enzyme interaction may involve other domains as well. The ATPase activity of the ATPase domain protein showed a quadratic dependence on enzyme concentration, suggesting that dimerization of the NH2-terminal domain is a rate-limiting step. Using both protein cross-linking and sedimentation equilibrium analysis, we showed that the ATPase domain exists as a monomer in the absence of cofactors but can readily dimerize in the presence of a nonhydrolyzable analog of ATP, 5′-adenylyl-β,γ-imidodiphosphate. More interestingly, both ATP and ADP can also promote protein dimerization. This result thus suggests that the protein clamp, mediated through the dimerization of ATPase domain, remains closed after ATP hydrolysis and opens upon the dissociation of ADP. We have prepared full-lengthDrosophila and human topoisomerase II and truncation constructs containing the amino-terminal ATPase domain, and we have analyzed their biochemical properties. The ATPase activity of the truncation proteins, similar to that of the full-length proteins, is greatly stimulated by the presence of DNA. This activity of the truncation proteins is also sensitive to the inhibition by the drug bisdioxopiperazine, ICRF-193, albeit at a much lower level than the full-length protein. Therefore, bisdioxopiperazine can directly interact with the NH2-terminal ATPase domain, but the drug-enzyme interaction may involve other domains as well. The ATPase activity of the ATPase domain protein showed a quadratic dependence on enzyme concentration, suggesting that dimerization of the NH2-terminal domain is a rate-limiting step. Using both protein cross-linking and sedimentation equilibrium analysis, we showed that the ATPase domain exists as a monomer in the absence of cofactors but can readily dimerize in the presence of a nonhydrolyzable analog of ATP, 5′-adenylyl-β,γ-imidodiphosphate. More interestingly, both ATP and ADP can also promote protein dimerization. This result thus suggests that the protein clamp, mediated through the dimerization of ATPase domain, remains closed after ATP hydrolysis and opens upon the dissociation of ADP. Type II DNA topoisomerases (topo II) 1topo IItopoisomerase(s) IIATPDNH2-terminal ATPase domainDmATPDDrosophila melanogaster ATPDHsATPDHomo sapiens ATPDDmGyrBD. melanogasterGyrB domainAMPPNP5′-adenylyl-β,γ-imidodiphosphate are ubiquitous enzymes that catalyze DNA topological changes by transporting one double strand DNA segment through another (1Brown P.O. Cozzarelli N.R. Science. 1979; 206: 1081-1083Crossref PubMed Scopus (222) Google Scholar, 2Mizuuchi K. Fisher L.M. O'Dea M.H. Gellert M. Proc. Natl. Acad. Sci. U. S. A. 1980; 77: 1847-1851Crossref PubMed Scopus (176) Google Scholar, 3Hsieh T. Brutlag D. Cell. 1980; 21: 115-125Abstract Full Text PDF PubMed Scopus (212) Google Scholar, 4Liu L.F. Liu C.C. Alberts B.M. Cell. 1980; 19: 697-707Abstract Full Text PDF PubMed Scopus (396) Google Scholar) (for a review, see Refs. 5Wang J.C. Annu. Rev. Biochem. 1996; 65: 635-692Crossref PubMed Scopus (2086) Google Scholar and 6Champoux J.J. Annu. Rev. Biochem. 2001; 70: 369-413Crossref PubMed Scopus (2218) Google Scholar). They play essential roles in many aspects of DNA transactions in vivo, including chromosome condensation and segregation, and removal of the supercoils generated during replication and transcription. In addition to such essential functions in the cells, topo II is the target of many widely used antibiotic and anti-tumor drugs (7Andoh T. Ishida R. Biochim. Biophys. Acta. 1998; 1400: 155-171Crossref PubMed Scopus (247) Google Scholar, 8Chen A.Y. Liu L.F. Annu. Rev. Pharmacol. Toxicol. 1994; 34: 191-218Crossref PubMed Scopus (780) Google Scholar, 9Froelich-Ammon S.J. Osheroff N. J. Biol. Chem. 1995; 270: 21429-21432Abstract Full Text Full Text PDF PubMed Scopus (509) Google Scholar, 10Maxwell A. Trends Microbiol. 1997; 5: 102-109Abstract Full Text PDF PubMed Scopus (319) Google Scholar, 11Nitiss J.L. Wang J.C. Mol. Pharmacol. 1996; 50: 1095-1102PubMed Google Scholar). topoisomerase(s) II NH2-terminal ATPase domain Drosophila melanogaster ATPD Homo sapiens ATPD D. melanogasterGyrB domain 5′-adenylyl-β,γ-imidodiphosphate Recent biochemical and structural studies have provided an understanding of the molecular mechanism for eukaryotic topo II (12Berger J.M. Gamblin S.J. Harrison S.C. Wang J.C. Nature. 1996; 379: 225-232Crossref PubMed Scopus (747) Google Scholar, 13Roca J. Wang J.C. Cell. 1992; 71: 833-840Abstract Full Text PDF PubMed Scopus (297) Google Scholar, 14Roca J. Wang J.C. Cell. 1994; 77: 609-616Abstract Full Text PDF PubMed Scopus (252) Google Scholar, 15Roca J. Berger J.M. Harrison S.C. Wang J.C. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 4057-4062Crossref PubMed Scopus (165) Google Scholar) (also reviewed in Refs. 5Wang J.C. Annu. Rev. Biochem. 1996; 65: 635-692Crossref PubMed Scopus (2086) Google Scholar, 6Champoux J.J. Annu. Rev. Biochem. 2001; 70: 369-413Crossref PubMed Scopus (2218) Google Scholar, and 16Wang J.C. Q. Rev. Biophys. 1998; 31: 107-144Crossref PubMed Scopus (304) Google Scholar). Eukaryotic topo II is a homodimer with a primary dimeric interface at its COOH terminus. The enzyme binds to a segment of DNA (G-segment) and generates a reversible double strand break to serve as a gateway for the strand passage. The binding of ATP to the ATPase domain leads to the dimerization of this domain and closure of the NH2-terminal protein clamp (N-gate). This movement in the NH2-terminal clamp can entrap another DNA segment (T-segment) and initiate a series of conformational changes that include the passage of the T-segment through the protein-mediated DNA break, the religation of the DNA gate, and release of the T-segment through the opening of its COOH-terminal dimer interface. Recent rapid kinetic measurements have provided further insights into this process (17Harkins T.T. Lindsley J.E. Biochemistry. 1998; 37: 7292-7298Crossref PubMed Scopus (62) Google Scholar, 18Baird C.L. Harkins T.T. Morris S.K. Lindsley J.E. Proc. Natl. Acad. Sci. U. S. A. 1999; 96: 13685-13690Crossref PubMed Scopus (110) Google Scholar). The two ATP molecules bound to each topo II homodimer are hydrolyzed at very different rates, and strand passage of T-segment occurs after the hydrolysis of the first ATP. The ATPase activity of topo II is reduced by a unique class of catalytic inhibitors, bisdioxopiperazines like ICRF-193 and ICRF-159 (7Andoh T. Ishida R. Biochim. Biophys. Acta. 1998; 1400: 155-171Crossref PubMed Scopus (247) Google Scholar). Bisdioxopiperazines are anti-tumor agents that target eukaryotic topo II in vivo (19Ishida R. Hamatake M. Wasserman R.A. Nitiss J.L. Wang J.C. Andoh T. Cancer Res. 1995; 55: 2299-2303PubMed Google Scholar). A previous study on a member of this drug family, ICRF-193, has shown that in the presence of ATP, the drug inhibits yeast topo II activities by trapping the enzyme in a closed clamp form (20Roca J. Ishida R. Berger J.M. Andoh T. Wang J.C. Proc. Natl. Acad. Sci. U. S. A. 1994; 91: 1781-1785Crossref PubMed Scopus (338) Google Scholar). Using rapid kinetic analysis, Morris et al.(21Morris S.K. Baird C.L. Lindsley J.E. J. Biol. Chem. 2000; 275: 2613-2618Abstract Full Text Full Text PDF PubMed Scopus (50) Google Scholar) have shown that ICRF-193, which does not work as a competitive inhibitor of ATP, can bind to a closed clamp complex but still allow ATP hydrolysis at a much lower rate. Initial analysis of some of the cell lines resistant to this drug indicates that mutations in the ATPase domain of topo II are responsible for the resistance (22Sehested 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, 23Wessel I. Jensen L.H. Jensen P.B. Falck J. Rose A. Roerth M. Nitiss J.L. Sehested M. Cancer Res. 1999; 59: 3442-3450PubMed Google Scholar). A recent study on the ATPase domain of yeast topo II has shown that ICRF-193 can inhibit the ATP hydrolysis, indicating a direct interaction between the drug and ATPase domain (24Olland S. Wang J.C. J. Biol. Chem. 1999; 274: 21688-21694Abstract Full Text Full Text PDF PubMed Scopus (39) Google Scholar). In contrast, a study with the 52-kDa ATPase domain of human topo II did not find any significant inhibition of ATPase activity by ICRF-193 (25Gardiner L.P. Roper D.I. Hammonds T.R. Maxwell A. Biochemistry. 1998; 37: 16997-17004Crossref PubMed Scopus (31) Google Scholar). Analysis of the cytotoxicity of this drug in the yeast model system indicates that cell killing requires more than just the irreversibly closed clamp conformation of intracellular topo II (26Jensen L.H. Nitiss K.C. Rose A. Dong J. Zhou J., Hu, T. Osheroff N. Jensen P.B. Sehested M. Nitiss J.L. J. Biol. Chem. 2000; 275: 2137-2146Abstract Full Text Full Text PDF PubMed Scopus (88) Google Scholar). Our previous study on the core domain of Drosophila topo II, which lacks the ATPase domain, also raised the possibility of the involvement of other topo II domains in the protein-drug interaction (27Chang S., Hu, T. Hsieh T.S. J. Biol. Chem. 1998; 273: 19822-19828Abstract Full Text Full Text PDF PubMed Scopus (40) Google Scholar). Whether this class of drugs can stimulate topo II-mediated DNA cleavage is still another issue remaining to be clarified. Earlier works have demonstrated that these drugs can inhibit the catalytic activities of the enzyme without stabilizing the covalent enzyme-DNA intermediate (28Ishida R. Miki T. Narita T. Yui R. Sato M. Utsumi K.R. Tanabe K. Andoh T. Cancer Res. 1991; 51: 4909-4916PubMed Google Scholar, 29Tanabe K. Ikegami Y. Ishida R. Andoh T. Cancer Res. 1991; 51: 4903-4908PubMed Google Scholar). Recent experiments using a different procedure to trap the topo II-DNA cleavable complex showed that ICRF-193 can promote DNA cleavage mediated by human topo II (30Huang K.C. Gao H. Yamasaki E.F. Grabowski D.R. Liu S. Shen L.L. Chan K.K. Ganapathi R. Snapka R.M. J. Biol. Chem. 2001; 276: 44488-44494Abstract Full Text Full Text PDF PubMed Scopus (79) Google Scholar). To further probe the action of bisdioxopiperazines, especially on the ATPase domain of topo II, we examined the effects of this drug on three types of topo II constructs: the ATPase domain; the GyrB homologous domain, which includes both the ATPase domain and the GyrB′ domain; and the full-length enzyme. We have also monitored the effect of nucleotide cofactors on the dimerization of the ATPase domain to gain insights into the effects of cofactors on the clamp closure at the N-gate. We used PCRs to construct the expression vectors for Drosophila and human NH2-terminal ATPase domains (DmATPD and HsATPD, respectively). The sequences of 5′ and 3′ PCR primers for generating HsATPD were 5′-CCT GGT TTG TAC AAA ATC TTT G-3′ and 5′-GGC AAC CTA GG T TA A TGG TGA TGA TGA TGG TGC TTG TTT AAC TGG ACT TGG GC-3′, respectively. The 5′ primer contains the sequence encoding residues 79–85 of human topo II including a BsrGI digestion site (underlined) that is unique on the yeast expression vector containing the entire human topo II coding sequence, YEpWOB6 (31Wasserman R.A. Austin C.A. Fisher L.M. Wang J.C. Cancer Res. 1993; 53: 3591-3596PubMed Google Scholar). The 3′ primer contains the sequence encoding residue 419–425 of human topo II and His6 tag (italicized) followed by a stop codon (boldface type) and a unique AvrII site (underlined). The sequences of 5′ and 3′ PCR primers for generating DmATPD are 5′ CAT TCCGGT GAC CAT GCA CAA G 3′ and 5′ GGC TTG CAT GC T TA A TGG TGA TGA TGA TGG TGC TTG GCA ATG TCA TTT TGG GCC 3′, respectively. The 5′ primer contains the sequence encoding residues 106–113 of Drosophila topo II including aBstEII site (underlined) that is unique on the yeast expression vector containing the entire Drosophila topo II coding sequence, YGBBXΔ22 (32Wyckoff E. Hsieh T.-s. Proc. Natl. Acad. Sci. U. S. A. 1988; 85: 6272-6276Crossref PubMed Scopus (35) Google Scholar). The 3′ primer contains the sequence encoding residues 397–403 of Drosophila topo II and His6 tag (italicized), followed by a stop codon (boldface type) and a unique SphI site (underlined). The PCR fragments were cloned back to the YEpWOB6 and YGBBXΔ22 vector using the above restriction enzymes. To generate the DmGyrB construct, we used TM10, which is a linker insertion construct containing a 4-residue (His-Ala-Cys-Lys) insertion between residues 668 and 669 ofDrosophila topo II (33Lee M.P. Hsieh T.S. J. Mol. Biol. 1994; 235: 436-447Crossref PubMed Scopus (19) Google Scholar). This insertion site is located at the junction of GyrB and GyrA domains, and there is a uniqueMluI site in the linker sequence. To generate a truncation at this site, the following two oligonucleotides were synthesized: 5′-CGC G CA TCA TCA TCA TCA C TA GCTGAC ATG-3′ and 5′-TCA GCT A GT GAT GAT GAT GAT G-3′. These sequences contain a 5′ MluI cohesive end and a 3′ SphI cohesive end (underlined), a 5-histidine tag (italicized), and a stop codon (boldface type). They were annealed to form the adapter and inserted into TM10 between the MluI site and SphI site to generate the DmGyrB construct. The sequence 5′ to that coding for pentahistidine, derived from part of the linker insertion sequence, codes for His-Ala. Together, they generate His-Ala-His5 at the carboxyl terminus of the DmGyrB protein construct. Both the wild-type and truncation proteins were overexpressed and purified as described previously (27Chang S., Hu, T. Hsieh T.S. J. Biol. Chem. 1998; 273: 19822-19828Abstract Full Text Full Text PDF PubMed Scopus (40) Google Scholar, 34Hu T. Chang S. Hsieh T. J. Biol. Chem. 1998; 273: 9586-9592Abstract Full Text Full Text PDF PubMed Scopus (29) Google Scholar). Purification of the His-tagged truncation proteins was performed as follows. The cleared lysate prepared from an 8-liter culture was passed through a 5-ml Ni2+-nitrilotriacetic acid column (Qiagen) in the presence of 15 mm imidazole, pH 7.0. 250 mm NaCl was also included in all of the buffers throughout the purification. After washing with 50 ml of 30 mm imidazole solution, stepwise concentrations of 60, 120, 240, and 500 mm imidazole were used to elute the protein, with a volume of 10 ml at each step. The His-tagged topo II protein was eluted between the 120 and 240 mm imidazole steps. The peak fractions were combined and further purified by HQ and HE column chromatography (PerSeptive Biosystem) using a procedure modified from our published methods (34Hu T. Chang S. Hsieh T. J. Biol. Chem. 1998; 273: 9586-9592Abstract Full Text Full Text PDF PubMed Scopus (29) Google Scholar). The final peak fractions were dialyzed against the storage buffer containing 10 mm Tris acetate, pH 7.8, 50 mmKAc, 5 mm Mg(Ac)2, and 50% glycerol. Also included in solutions used throughout the steps of purification and storage was a mixture of protease inhibitors (1 mmphenylmethylsulfonyl fluoride, 1 μg/ml leupeptin, and 1 μg/ml pepstatin). Prior to the ATPase assay, the proteins were concentrated and exchanged into a buffer of 10 mm Tris acetate, pH 7.9, using a Microcon centrifugal device (Millipore Corp., Bedford, MA). Two methods were used to measure the ATP hydrolysis. The TLC method was done as described previously (34Hu T. Chang S. Hsieh T. J. Biol. Chem. 1998; 273: 9586-9592Abstract Full Text Full Text PDF PubMed Scopus (29) Google Scholar). The coupled PEP/LDH spectrophotometric assay was performed at 30 °C on an HP 8453 diode array spectrophotometer with a thermostat cuvette cell according to a published procedure (35Lindsley J.E. Wang J.C. J. Biol. Chem. 1993; 268: 8096-8104Abstract Full Text PDF PubMed Google Scholar). In all of the experiments, only the data from the linear range of ATP hydrolysis were used to calculate the reaction rate. We have tested for the optimal conditions for these proteins. Both ATPD proteins have shown the highest activity in a buffer with an ionic strength that is less than 10 mmin monovalent salt and 2–4 mm in divalent cation (data not shown). In most experiments, the standard reaction mixture contained 10 mm Tris acetate (pH 7.9), 5 mm KAc, 2.5 mm Mg(Ac)2, and 1.25 mm ATP. The optimal ionic conditions for ATPase activity of DmGyrB fragment were also determined to be 10 mm Tris acetate (pH 7.9), 25 mm KAc, 10 mm Mg(Ac)2. The ionic strength for DmGyrB is higher than that optimal for the HsATPD and DmATPD but still much lower than that for full-length topo II. One apparent effect of the truncation of the carboxyl portion of topo II is that optimal ATPase activity in these proteins occurs at a much lower ionic strength in the reaction conditions. Samples of HsATPD were dialyzed against 10 mm Tris acetate, pH 7.8, 30 mmpotassium phosphate, and 5 mm Mg(Ac)2. 75-μl samples (1 mg/ml) were successively brought to sedimentation equilibrium at 9,000, 12,000, and 15,000 rpm over a period of 2 days at 20 °C in an analytical ultracentrifuge (XL-A ultracentrifuge; Beckman Instruments). To establish a base line, the samples were subjected then to an overspeed at 50,000 rpm, and, where needed, the data were base line-corrected. Cofactors like ATP, AMPPNP, and ICRF-193 were included in some samples at the concentrations indicated. In each case, the same concentration of cofactor was included both in the sample and in the reference solvent. Absorbance measurements were made routinely at 290 nm, where the absorbance due to the cofactors is low compared with the protein. The data were fitted by theIdeal-1 program (Beckman Instruments) using a value ofv̄ = 0.7384 (partial specific volume of HsATPD), calculated from the amino acid composition, and a value of ρ20 = 1.006 (solvent density), measured by picnometry. Achievement of sedimentation equilibrium was confirmed by time lapse scans and, more importantly, by the agreement of calculated best fit molecular weights from sequential sedimentation scans at each speed. To calculate the dimerization constants, we used two software programs provided by Dr. Allen Minton to process and analyze the sedimentation equilibria data. The best fit weight average molecular weight data from all three speeds were obtained first by using the programMWAVCALC6. The dimerization constants were then obtained by using the program F-MWN50-N to globally fit the molecular weight data from all three speeds. This program uses Marquardt-Levenberg χ2 minimization in a nonlinear least squares model to globally fit the sedimentation data. The S.E. in all of the dimerization constants calculated from such analysis is usually smaller than 10%. We assume that the following three equilibria best describe the reactions of HsATPD in the presence of ADP under the conditions of sedimentation equilibrium experiments. N+N↔K0N2Equation 1 N+A↔K1NAEquation 2 NA+NA↔K2(NA)2Equation 3 N and NA represent the unliganded and liganded monomer of ATPD, respectively, and A represents the cofactor ADP.K 1 defines the association constant of ADP with ATPD. K 0 and K 2 are the dimerization constants for unliganded and liganded monomer, respectively. Inclusion of the equilibrium involving the association of heterotypic monomer, N and NA, does not alter the curve-fitting described below, suggesting that the above equilibria are parsimonious with respect to the data analysis presented here. The apparent dimerization constant K Dapp obtained from sedimentation equilibrium includes all of the dimer and monomer species and is defined as the following. KDapp=(Dimer species)(Monomer species)2=((N)2+(NA)2)(N+NA)2Equation 4 Substituting Equations Equation 1, Equation 2, Equation 3 into Equation 4, we obtain the following. KDapp=K12K2(A)2+K0(1+K1(A))2Equation 5 By curve-fitting of the plot K Dapp versus ADP concentration using Equation 5 with KaleidaGraph (Synergy Software, Reading, PA), we can obtain the equilibrium constants defined in Equations Equation 1, Equation 2, Equation 3. The assay was performed as described in Ref. 25Gardiner L.P. Roper D.I. Hammonds T.R. Maxwell A. Biochemistry. 1998; 37: 16997-17004Crossref PubMed Scopus (31) Google Scholar with the following modifications. The protein was dialyzed into 50 mm HEPES, pH 8.5, 5 mm KCl, 4 mm dithiothreitol, and 4 mmMg(Ac)2. After incubation for 1 h at 25 °C, dimethylsuberimidate was added to a final concentration of 0.2 mg/ml followed by an additional 1-h incubation at 25 °C. The reactions were quenched by adding Tris-glycine, pH 8.0, to 50 mm and analyzed by 6% SDS-PAGE (36Laemmli U.K. Nature. 1970; 227: 680-685Crossref PubMed Scopus (207538) Google Scholar). To study the functions of the ATPase domain and the GyrB homologous domain of eukaryotic topo II, we have constructed two NH2-terminal truncation fragments ofDrosophila topo II and one from human topo II. The truncation end points are based on the domain structures ofDrosophila topo II, as determined by protease-sensitive site and linker insertion analysis (Fig.1 A) (33Lee M.P. Hsieh T.S. J. Mol. Biol. 1994; 235: 436-447Crossref PubMed Scopus (19) Google Scholar). The domain structure of eukaryotic topo II is conserved as shown by similar analysis with yeast topo II (37Lindsley J.E. Wang J.C. Proc. Natl. Acad. Sci. U. S. A. 1991; 88: 10485-10489Crossref PubMed Scopus (118) Google Scholar, 38Jensen S. Andersen A.H. Kjeldsen E. Biersack H. Olsen E.H. Andersen T.B. Westergaard O. Jakobsen B.K. Mol. Cell. Biol. 1996; 16: 3866-3877Crossref PubMed Scopus (52) Google Scholar) and x-ray crystallography (12Berger J.M. Gamblin S.J. Harrison S.C. Wang J.C. Nature. 1996; 379: 225-232Crossref PubMed Scopus (747) Google Scholar). For theDrosophila truncation constructs, the 47-kDa fragment of the ATPase domain (DmATPD) contains the first 406 amino acid residues plus a hexahistidine tag, whereas the 77-kDa fragment of the GyrB homologous domain (DmGyrB) contains the first 668 amino acid residues plus a hexahistidine tag (Fig. 1, B and C). The 46-kDa fragment of ATPase domain of human topo II, HsATPD, contains the first 6 amino acid residues of Saccharomyces cerevisiae topo II followed by Ser-29 to Lys-425 of human topo II and a hexahistidine tag (Fig. 1 D). All three constructs were overexpressed in a protease-deficient yeast strain BCY123 and purified to over 95% of homogeneity (Fig. 2).Figure 2Purification of the truncated topo II. 2 μg each of molecular weight size markers and three truncation topo II proteins were run in a 7% SDS-PAGE and stained with Coomassie Brilliant Blue. DmATPD, HsATPD, and DmGyrB are shown inlanes 2, 3, and 4, respectively.View Large Image Figure ViewerDownload Hi-res image Download (PPT) For both DmATPD and HsATPD, the presence of DNA greatly stimulates their ATPase activity (comparing A and B of Fig.3). In the absence of DNA, both proteins had only a low level of ATPase activity, with a reduction of about 20-fold when compared with the conditions in the presence of DNA and at the identical protein concentrations. The DNA dependence of the proteins has been explored over a range of concentrations from 0 to 300 μm in base pairs. The maximal DNA stimulation for DmATPD and HsATPD happens at a concentration of 100 and 30 μm bp, respectively (data not shown). The optimal DNA concentration in the stimulation of ATPase activity for DmGyrB protein is also estimated to be around 100 μm bp. Interestingly, when the ATPase activities were measured at different protein concentrations, both the DNA-independent and the DNA-dependent ATPase activity of DmATPD and HsATPD have shown a parabolic dependence on the enzyme concentration, suggesting that the dimerization of the subunits is the rate-limiting step in ATP hydrolysis under such conditions (Fig. 3). In contrast, the concentration dependence for the DmGyrB proteins follows an expected linear relationship, just like that for the full-length proteins ofDrosophila topo II or human topo IIα (data not shown). Therefore, dimerization of the GyrB domain of topo II is favored at equilibrium, and this process is not rate-limiting for the ATPase reaction. While all the truncation proteins, similar to the full-length enzymes, exhibited ATPase activities that can be further stimulated by the presence of DNA, their ATPase activities are much reduced in comparison with the holoenzymes. To gain quantitative insight into these rate differences, we have analyzed the ATP hydrolysis rates as a function of substrate concentrations. For HsATPD and DmATPD, these rate curves can be fit with a Michaelis-Menten kinetic equation (Fig.4, A and B). For the holoenzymes and DmGyrB, their rate data can also fit Michaelis-Menten kinetics. While the apparent K m values for these constructs are comparable, ranging between 0.4 and 0.6 mm (data not shown), there is a 10-fold difference in apparent k cat between the full-length proteins and the ATPase domain fragments (TableI). These data suggest that substrate binding may remain unaffected whether the ATPase domain is part of the holoenzyme or exists by itself, assuming that the apparentK m values in this case can give an estimate of the ATP binding affinities. However, protein domains other than the ATPase part can have a major role for the ATP hydrolysis activity.Table IApparent k cat for the topo II proteinsWith DNAWithout DNAs−1Human holoenzyme2.30.1Drosophilaholoenzyme2.00.1Drosophila GyrB0.60.03Human ATPD0.210.02DrosophilaATPD0.350.02k cat is obtained by dividing theV max of ATPase activity by the monomeric concentration of the enzymes. V max values were determined using the data analysis similar to what was shown in Fig. 4. Open table in a new tab k cat is obtained by dividing theV max of ATPase activity by the monomeric concentration of the enzymes. V max values were determined using the data analysis similar to what was shown in Fig. 4. The anti-cancer agent bisdioxopiperazine is a topo II inhibitor and can lock up the N-gate in the presence of ATP (20Roca J. Ishida R. Berger J.M. Andoh T. Wang J.C. Proc. Natl. Acad. Sci. U. S. A. 1994; 91: 1781-1785Crossref PubMed Scopus (338) Google Scholar). It remains unclear, however, whether bisdioxopiperazine only targets the ATPase domain and thereby inhibits its ATPase activity. Whereas an NH2-terminal ATPD of yeast topo II (residues 1–409) expressed in yeast cells can be inhibited by ICRF-193 (24Olland S. Wang J.C. J. Biol. Chem. 1999; 274: 21688-21694Abstract Full Text Full Text PDF PubMed Scopus (39) Google Scholar), another drug in this family, ICRF-159, cannot inhibit the ATPase activity from a similar NH2-terminal fragment (residues 1–439) of human topo II purified from the renatured fraction of a bacterial expression system (25Gardiner L.P. Roper D.I. Hammonds T.R. Maxwell A. Biochemistry. 1998; 37: 16997-17004Crossref PubMed Scopus (31) Google Scholar). The difference could be due to different expression/purification systems and ATPD from different organisms. To further investigate the inhibition of eukaryotic topo II by bisdioxopiperazine drugs, we studied the effect of ICRF-193 on the ATPase activities of both human and Drosophila topo II holoenzymes, DmGyrB, and both human and Drosophila ATPase domain fragments. As shown in Fig. 5, the ATPase activity of both holoenzymes is exquisitely sensitive to the drug, with the IC50 of the human and Drosophilaproteins being around 0.1 and 0.3 μm, respectively. The inhibition of the ATPase domains by ICRF-193 could also be detected, although at much higher drug concentrations, with the IC50of HsATPD and DmATPD being 60 and 30 μm, respectively. For DmGyrB protein, the sensitivity of its ATPase activity to the drug is between those of the holoenzyme and the ATPase domain, with an IC50 of 3 μm. Comparing the result shown in Fig. 5 and the ATPase activity of the different constructs listed in Table I, we have found that the sensitivity of topo II proteins to ICRF-193 correlates with the ATPase activity of these proteins. For the topo II constructs with a higher ATPase activity, there is a concomitant increase in the sensitivity toward bisdioxopiperazine. These results also suggest the bisdioxopiperazine drugs can directly effect the ATPase domain of eukaryotic topo II to inhibit its ATP hydrolysis activity. However, we cannot rule out the possible involvement of the other domains of the enzyme in the action of these drugs, since the drug sensitivity increases when other domains of the holoenzyme are also included in the protein constructs. The closure of the N-gate through dimerization of the NH2-terminal domain plays a critical role in the initiation of the catalytic cycle of topo II reaction. Furthermore, since the mechanism of action of bisdioxopiperazine drugs involves the closure of the N-gate, they may affect the dimerization of ATPD. We have examined the effects of cofactors on the dimerization of the ATPD by monitoring the change in the molecular mass by chemical cross-linking and by sedimentation equilibrium. In the cross-linking experiments, dimethylsuberimidate was used to covalently bridge the dimerized protein (Fig.6). For ATPD from bothDrosophila and human topo II, no dimerized species were detected in the absence
Referência(s)