This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Duan, W. Articles by Hurley, L. H. Search for Related Content PubMed PubMed Citation Articles by Duan, W. Articles by Hurley, L. H.

College of Pharmacy, The University of Texas, Austin, Texas 78712 [W. D., A. R., M-Y. K., Q. Z., O. Y. F., L. H. H.]; Institute for Drug Development, San Antonio, Texas 78245 [D. S., S. Y. R., E. I., D. D. V. H.]; Department of Biology, Georgetown University, Washington, DC 20057 [D. N.]; and Arizona Cancer Center, Tucson, Arizona 85724 [H. H., D. D. V. H., L. H. H.]

	Abstract

Top Abstract Introduction Materials and Methods Results Discussion References

 
In this study we have identified a new structural motif for a ligand with G-quadruplex interaction that results in biological effects associated with G-quadruplex-interactive compounds. Fluoroquinolones have been reported to possess weak telomerase inhibitory activity in addition to their better known bacterial gyrase poisoning. Starting with a fluoroquinobenzoxazine, which has modest potency in a human topoisomerase II assay, we have designed a more potent inhibitor of telomerase that has lost its topoisomerase II poisoning activity. This fluoroquinophenoxazine (FQP) interacts with G-quadruplex structures to inhibit the progression of Taq polymerase in a G-quadruplex polymerase stop assay. In addition, we demonstrate by ¹H NMR studies that this compound interacts with telomeric G-quadruplex structures by external stacking to the G-tetrad with both the unimolecular fold-over and the parallel G-quadruplex structures. A photocleavage assay confirms the FQP interaction site, which is located off center of the external tetrad but within the loop region. Molecular modeling using simulated annealing was performed on the FQP-parallel G-quadruplex complex to determine the optimum FQP orientation and key molecular interactions with the telomeric G-quadruplex structure. On the basis of the results of these studies, two additional FQP analogues were synthesized, which were designed to test the importance of these key interactions. These analogues were evaluated in the Taq polymerase stop assay for G-quadruplex interaction. The data from this study and the biological evaluation of these three FQPs, using cytotoxicity and a sea urchin embryo system, were in accord with the predicted more potent telomeric G-quadruplex interactions of the initial lead compound and one of the analogues. On the basis of these structural and biological studies, the design of more potent and selective telomeric G-quadruplex-interactive compounds can be envisaged.

	Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

 
Fluoroquinolones are best known therapeutically as bacterial gyrase inhibitors (1, 2) but more recently have also been demonstrated to be active against eukaryotic topoisomerase II, either as catalytic inhibitors or topoisomerase II poisons (3), or as telomerase inhibitors (4). One such group of compounds, the quinobenzoxazines, typified by A-62176 (Fig. 1B), has served as a starting point for the design of more potent topoisomerase II inhibitors using a structure-based approach (5). The extended aromatic conjugation system of A-62176 suggested to us that quinobenzoxazines may also intercalate with the more expansive system of G-quadruplex DNA and thereby be telomerase inhibitors in parallel with other G-quadruplex-interactive compounds, such as the anthraquinones (6-10), cationic porphyrins (11-13),⁹ piperazines (14), ethidium compounds, (15), and a pentacyclicacridinium compound (16). After we started this project, an independent report that bacterial quinolones may be weak telomerase inhibitors (4) also supported the idea that a ring-extended fluoroquinolone would be a more potent telomerase inhibitor.

View larger version (20K): [in this window] [in a new window]   Fig. 1 A, G-tetrad and G-quadruplexes. Panel a, four guanine residues forming a planar structure G-tetrad through Hoogsteen hydrogen bonding. Panel b, a parallel G-quadruplex model. Panel c, an intermolecular antiparallel G-quadruplex model. Panel d, an intramolecular basket G-quadruplex model. Each parallelogram in panels b–d represents a G-tetrad. B, structures of the topoisomerase II-interactive compound (A-62176) and G-quadruplex-interactive compounds QQ58, QQ27, QQ28, and TMPyP4.

G-quadruplex-interactive compounds have aroused more interest recently because it has been demonstrated that these compounds not only inhibit telomerase in cell-free (6, 9) and in in vitro systems (36) but also cause telomere shortening and cell crisis in cancer cells (14).¹⁰ A number of reviews on G-quadruplex and their targeting by small molecules have appeared recently (37-45).

In this contribution we demonstrate that QQ58 (Fig. 1B), a FQP¹¹ that lacks bacterial gyrase or topoisomerase II poisoning activity, is able to bind to G-quadruplex structures and, by doing so, inhibit telomerase. Structural insight into the interactions of this compound with both intermolecular and intramolecular G-quadruplex structures has been obtained by one-dimensional ¹H NMR molecular modeling and a photocleavage assay. We also demonstrate that this compound induces chromosomal effects and slowing of the division of sea urchin embryos typical of G-quadruplex-interactive compounds. On the basis of molecular modeling studies, two additional analogues of this compound were designed and synthesized (QQ27 and QQ28; Fig. 1B), and their interactions with G-quadruplex structures and biological effects were examined. The maintenance or loss of biological potency paralleled their interactions with the G-quadruplex structure, and this was predicted on the basis of molecular modeling.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

 
Synthesis of Extended Fluoroquinolones (QQ58, QQ27, and QQ28 in Figure 2)
General   All of the melting points were recorded on a Thomas-Hoover capillary melting point apparatus and are uncorrected. ¹H and ¹³C data were obtained on a Varian Unity 300 MHz NMR spectrometer. The chemical shifts are relative to the trace proton, carbon, or fluorine signals of the deuterated solvent. Coupling constants, J, are reported in Hz and refer to apparent peak multiplicity rather than coupling constants. Mass spectroscopic experiments were performed by the Mass Spectroscopy Center at The University of Texas, Austin, TX. Elemental analysis of C, H, and N was done by Desert Analytics, Tucson, AZ. Flash column chromatography was performed on silica gel 60, 230–400 mesh, purchased from Spectrum. All of the starting materials were obtained from commercial sources unless otherwise specified.

View larger version (21K): [in this window] [in a new window]   Fig. 2 Condition: (a) (i) Ac2O, CH(OC2H5)3, (ii) 3-amino-4-hydroxydibenzofuran hydrobromide; (b) THF, NaH; (c) pyridine, (R) t-Boc-3-aminopyrrolidine; (d) OH-/H+; (e) OH-; (f) CF3COOH.

Ethyl 1,2-difluoro-4-oxo-4H-pyrido[3,2,1-k,l]-(3-amino-2-hydroxydibenzofuranyl)[1,2-g]phenoxazine-5-carboxylate (3)   Ethyl 3-(2-hydroxydibenzofuran-3-amino)-2-[(2,3,4,5,-tetrafluorophenyl)carbonyl] prop-2-enoate (2.32 g; 4.9 mmol) and NaH (60% in mineral oil; 0.45 g; 11.3 mmol) were mixed with freshly distilled tetrahydrofuran and stirred at -78°C for 15 min. The mixture was allowed to warm to room temperature gradually and then heated at 65°C for 30 min. The excess NaH was quenched by addition of 30 ml of methanol. The solution was evaporated to dryness, and the product was purified through flash column chromatography, yielding a yellow powder (1.84 g; 89%): mp >229°C (decomposed); ¹H NMR (CDCl₃) 9.01 (s, 1H), 7.90 (d, 1H, J = 7.8 Hz), 7.80 (dd, 1H, J = 7.8, 2.1 Hz), 7.39 (t, 1H, J = 6.9 Hz), 4.48 (q, 2H, J = 7.2 Hz), 1.49 (t, 3H, J = 7.2 Hz); ¹⁹F NMR -133.6 (dd, 1F, J = 23, 10 Hz), -1561.5 (dd, 1F, J = 21, 9 Hz); and MS (CI): m/z 434 (M + H).

Ethyl (R)-1-{3-[(tert-butoxy)carbonylamino]pyrrolidin-1-yl}-2-fluoro-4-oxo-4H-pyrido[3,2,1-k,l]-(3-amino-2-hydroxydibenzofuranyl)[1,2-g]phenoxazine-5-carboxylate (4)   Ethyl 1,2-difluoro-4-oxo-4H-pyrido[3,2,1-k,l](3-amino-2-hydroxydibenzofuranyl)[1,2-g]phenoxazine-5-carboxylate (0.232 g; 0.54 mmol) and (R)-3-(tert-butoxycarbonylamino)pyrrolidine (0.302 g; 1.62 mmol) were dissolved in 30 ml of pyridine, and the residue was purified using flash column chromatography (dichloromethane:ethylacetate = 2:3), yielding a yellow powder (0.27 g, 84%): mp: >261°C (decomposed). ¹H NMR 8.90 (s, 1H), 7.90 (d, 1H, J = 7.5 Hz), 7.64 (m, 3H), 7.57 (s, 1H), 7.51 (t, 1H, J = 6.9 Hz), 7.39 (t, 1H, J = 6.9 Hz), 5.35 (s, br, 1H), 4.44 (q, 2H, J = 7.2 Hz), 3.70 (m, 1H), 3.61 (m, 1H), 3.40 (m, 2H), 2.29 (m, 1H), 2.00 (m, 1H), 1.49 (s, 9H), 1.46 (t, 3H, J = 7.2 Hz); ¹⁹F -119.8 (d, 1F, J = 13 Hz); and MS (CI): m/z 600 (M+1).

(R)-1-(3-aminopyrrolidin-1-yl)-2-difluoro-4-oxo-4Hpyrido[3,2,1-k,l]-(3-amino-2-hydroxydibenzofuranyl)[1,2g]phenoxazine-5-carboxylic Acid HCl Salt (5; QQ58)   Ethyl (R)-1-{3-[(tert-butoxy)carbonylamino]pyrrolidin-1-yl}-2-fluoro-4-oxo-4H-pyrido[3,2,1-k,l]-(3-amino-2-hydroxydibenzofuranyl)[1,2-g]phenoxazine-5-carboxylate (0.228 g; 0.526 mmol) was mixed with 10 ml of ethanol and 3 ml 1 N KOH. The mixture was refluxed for 30 min. HCl (9 ml of 2 N) and 6 ml of ethanol were then added to the mixture, and it was refluxed for 4 h. After the reaction mixture was cooled down slowly, a yellow powder was obtained by filtration, washing with water and ethanol, and drying (250 mg, 95%): mp: >250°C (decomposed); ¹H NMR (DMSO-d₆) 9.24 (s, 1H), 8.49 (m, 1H), 8.12 (s, 1H), 8.05 (d, 1H, J = 7.5 Hz), 7.69 (d, 1H, J = 8.4 Hz), 7.54 (t, 1H, J = 7.2 Hz), 7.47 (t, 1H, J = 13.5 Hz), 7.40 (t, 1H, J = 7.5 Hz), 3.80–4.05 (m, 5H), 2.31 (m, 1H), 2.08 (m, 1H); ¹⁹F NMR -119.1 (d, 1F, J = 15 Hz); and MS (CI): m/z 472 (M + H). Anal. (C₂₆H₁₈N₃FO₅.HCl.7/4H₂O) C, H, N (Cal. C 62.09 H3.96 N8.35; Found C62.21 H3.75 N 8.13).

Ethyl (R)-1-(3-aminopyrrolidin-1-yl)-2-fluoro-4-oxo-4H-pyrido[3,2,1-k,l]-(3-amino-2-hydroxydibenzofuranyl)[1,2-g]phenoxazine-5-carboxylate (6; QQ28)   To a mixture of ethyl (R)-1-{3-[(tert-butoxy)carbonylamino]pyrrolidin-1-yl}-2-fluoro-4-oxo-4H-pyrido[3,2,1-k,l]-(3-amino-2-hydroxydibenzofuranyl)[1,2-g]phenoxazine-5-carboxylate (50 mg; 0.008 mmol) and methylene chloride (2 ml) was added trifluoroacetic acid (0.5 ml) at room temperature. After addition, the reaction mixture was stirred at room temperature for 2 h. Solvent was removed to give a residue and to this residue was added methanol (5 ml) and triethylamine (0.5 ml), and the yellow solid was collected to give a yellow powder (30 mg; 72%): mp: >210°C (decomposed); ¹H NMR (DMSO-d₆) 8.79 (s, 1H), 8.05 (s, 1H), 8.03 (s,1H), 7.81 (s, 1H), 7.61 (d, 1H, J = 8.0 Hz), 7.47 (t, 1H, J = 8.0 Hz), 7.36 (t, 1H, J = 8.0 Hz), 7.21 (d, 1H, J = 8.0 Hz), 4.23 (q, 2H, J = 6.8 Hz), 3.78 (m, 2H), 3.67 (m,1H), 3.54 (m, 2H), 2.00 (m, 1H), 1.68 (m, 1H), 1.32 (t, 3H, J = 6.8 Hz); MS (FAB): 500.6 (M + H); HRMS (FAB) calc. for C₂₈H₂₃FN₃O₅ 500.1622, found 500.1623; Anal. (C₂₈H₂₂FN₃O₅.H₂O) C, H, N (Cal. C64.98 H4.67 N8.12: Found C65.10 H4.81 N7.90).

(R)-1-{3-[(tert-butoxy)carbonylamino]pyrrolidin-1-yl}-2-fluoro-4-oxo-4H-pyrido[3,2,1-k,l]-(3-amino-2-hydroxydibenzofuranyl)[1,2-g]phenoxazine-5-carboxylic acid (7;QQ27)   Ethyl (R)-1-{3-[(tert-butoxy)carbonylamino]pyrrolidin-1-yl}-2-fluoro-4-oxo-4H-pyrido[3,2,1-k,l]-(3-amino-2-hydroxydibenzofuranyl)[1,2-g]phenoxazine-5-carboxylate (100 mg; 0.17 mmol) was dissolved in methanol (6 ml), a solution of NaOH (37 mg; 0.67 mmol) in water (3 ml) was added, and the reaction mixture was stirred at room temperature for 2 h. Methanol was evaporated off, clear solution was neutralized with 10% HCl to pH = 2.0, and precipitate was collected and dried with a vacuum pump to give a yellow powder (91 mg, 95%): MP: 235–237°C; ¹H NMR (DMSO-d₆) 9.15 (s, 1H), 8.40 (s,1H), 8.04 (d, 1H, J = 7.5 Hz), 7.94 (s, 1H), 7.66 (d, 1H, J = 8.2 Hz), 7.51 (t, 1H, J = 7.5 Hz), 7.39 (m, 2H), 4.12 (m, 1H), 3.95 (m, 1H), 3.89 (m, 1H), 3.77 (m, 1H), 3.56 (m, 1H), 2.11 (m,1H), 1.89 (m, 1H),1.41 (m,1H); MS (FAB): 572.6 (M + H); HRMS (FAB) (M + H) calc. for C₃₁H₂₇FN₃O₇ 572.1838, found 572.3338; Anal. (C₃₁H₂₆FN₃O₇.7/4H₂O) C, H, N (Cal. C61.73 H4.63 N7.13; Found C61.97 H4.50 N6.96).

Drug and Oligonucleotide Preparation   Drug solutions were prepared as 5 mM stock solutions in DMSO and stored at -20°C. These stock solutions were diluted to working concentrations in distilled water immediately before use. Oligonucleotides were synthesized on a PerSeptive Biosystems Expedite 8909 automatic DNA synthesizer. The samples for NMR experiments were purified by reverse phase high-performance liquid chromatography on a C18 column (Dynamax-300A) and dialyzed extensively against 10 mM KCl or 20 mM NaCl solutions followed by deionized water. Solid supports and phosphoramadites were purchased from Glen Research and PerSeptive Biosystems. Oligonucleotides for polymerase extension assay and photocleavage reactions were purified by denaturing PAGE, diluted to required concentrations, and dispensed into small aliquots.

Decatenation Assay   Kinetoplast DNA (0.25 µg) was incubated with various concentrations of QQ58 or A-62176 in 10 µl of reaction buffer [50 mM Tris-HCl (pH 8.0), 120 mM KCl, 10 mM MgCl₂, 0.5 mM ATP, and 0.5 mM DTT) for 10 min. Two units of human topoisomerase II (TopoGEN Inc.) were added to the mixture, which was then incubated at 37°C for 30 min. The reaction was terminated with 0.1 volume of stop buffer (5% sarkosyl, 0.025% bromophenol blue, and 50% glycerol). The decatenation products were analyzed on 1% agarose gels run with 0.5 µg/ml ethidium bromide.

Preparation and End-labeling of Oligonucleotides for Topoisomerase II Cleavage Reaction   DNA oligomers 5'-d[CGATGGGGAAGATCGGGCTCGTATACATTGATACGGGGCTCATGAGCGCTTGTTTCGGCG]-3' (A1) and 5'-d[CGCC-GAAACAAGCGCTCATGAGCCCCGTATCAATGTATACGAGCCCGATCTTCCCCATCG]-3' (A2; Ref. 46) were synthesized as above. The 5' end-labeled single-strand oligonucleotide was obtained by incubating A1 with T4 polynucleotide kinase and [-³²P]ATP at 37°C for 1 h. Labeled DNA was purified with a Bio-Spin 6 chromatography column (Bio-Rad) after inactivating T4 polynucleotide kinase by heating for 8 min at 70°C. The labeled strand was then annealed with the complementary strand (A2) and purified on an 8% native polyacrylamide gel.

Topoisomerase II Cleavage Reaction   The labeled double-stranded DNA was incubated with 20 units of human topoisomerase II in 20 µl of reaction buffer [30 mM Tris-HCl (pH 7.6), 3 mM ATP, 15 mM ß-mercaptoethanol, 8 mM MgCl₂, and 60 mM NaCl] at 30°C for 10 min in the presence of various concentrations of QQ58 or A-62176. Reactions were terminated by adding SDS to 1% of the final concentration, and topoisomerase II was removed by proteinase K digestion (100 µg/ml) at 42°C for 1 h followed by phenol-chloroform extraction and ethanol precipitation. Samples were loaded onto a 12% denaturing sequencing gel. The dried gels were exposed on a phosphor screen. Imaging and quantification were performed using a PhosphorImager (Storm 820) and ImageQuant 5.1 software from Molecular Dynamics.

Telomerase Assay   Telomerase assays were run as described previously (47) using a primer extension assay in which a 5'-biotinylated primer consisting of three telomeric repeats was used, and the incorporation of ³²P-labeled GTP was measured. In brief, reaction mixtures (20 µl) contained 4 µl of cell extracts (60 µg total cell protein), 50 mM Tris-OAc (pH 8.5), 50 mM K-OAc, 1 mM MgCl₂, 5 mM BME, 1 mM spermidine, 1 µM 5'-biotinylated telomere primer (TTAGGG)₃, 1.2 µM [-³²P]-dGTP (800 Ci/mmol), 1 mM dATP, and 1 mM dTTP, and were incubated at 37°C for 30 min. Reactions were terminated by adding 20 µl of Streptavidin-Dynabeads, which bind selectively to the desired targets (5'-biotinylated primer), forming a magnetic bead-target complex. This complex was separated from the suspension using a magnet (Dynal MPC) and washed several times with washing buffer (2 M NaCl) to eliminate [-³²P]-dGTP background. Telomerase reaction products were separated from the magnetic beads by protein denaturation with 5.0 M guanidine-HCl at 90°C for 30 min. After recovery of these products, analysis was performed by 8% denaturing PAGE. Telomerase activities were quantified by densitometric analyses of an autoradiogram using ImageQuant software. All of the reactions were carried out in amber Eppendorf tubes under subdued lighting to avoid photocleavage.

Polymerase Stop Assay   The DNA primer 5'-d[TAATACGACTCACTATAG]-3' and template DNA 5'-d[TCCAACTATGTATAC(TTGGGG)₄TTAGCGGCACGCAATTGCTATA-GTGAGTCGTATTA]-3' were synthesized and purified as before (48). About 200 ng of the primer DNA was 5'-end-labeled with ³²P using T4 polynucleotide kinase and subsequently purified by denaturing PAGE. The assay was carried out as described previously (48). Briefly, 24 nM of labeled primer and 12 nM of template DNA were annealed in a 1 x reaction buffer [10 mM Tris-HCl (pH 8.0) and 5 mM KCl] by heating to 95°C and slowly cooled to room temperature. MgCl₂, dNTP, and Taq DNA polymerase (BoeringherManheim) were added, and the mixture was incubated at 55°C for 20 min. The polymerase extension was stopped by adding 2 x stop buffer (10 mM EDTA, 10 mM NaOH, 0.1% xylene cyanol, and 0.1% bromphenol blue) and loaded onto a 12% sequencing gel. After electrophoresis, the gels were dried and visualized on a PhosophorImager and quantitated using ImageQuant software.

Photo-mediated Strand Cleavage Reaction   The sequence of the single-stranded DNA (G-quadruplex) used was 5'-d[CATGGTGGTTT(GGGTTA)₄CCAC]-3'. About 400 ng of G-quadruplex DNA was 5'-end-labeled with ³²P using T4 polynucleotide kinase (New England Biolabs) and subsequently purified by denaturing PAGE. This 5'-labeled G-quadruplex DNA was stored in Tris-EDTA buffer [10 mM Tris-HCl, 1 mM EDTA (pH 7.5)] at a concentration of 5 ng/µl and 3000 cpm/µl. For each photocleavage reaction, 30 µl of G-quadruplex DNA was mixed with 30 µl of a 200 mM KCl solution, placed in a water bath at 95°C for 4 min, and slowly cooled to room temperature. A control set was run in distilled water instead of KCl solution. QQ58 (6 µl) at varied concentrations was added to each sample and transferred to a 24-well Titretek microtiter plate (ICN). This plate was placed on top of a Pyrex glass shield and irradiated for 2 h with an 85-W xenon lamp placed under the Pyrex glass. (Pyrex glass was used to filter the UV light under 300 nm, thereby eliminating DNA damage caused directly by UV irradiation.) During the irradiation, the Titretek plate was rotated three times to eliminate light heterogeneity. Reactions were terminated by the addition of 10 µg of calf thymus DNA, followed by phenol-chloroform extraction and ethanol precipitation. The resulting samples were subjected to treatment with 0.1 M piperidine. The samples were then loaded onto a 16% sequencing gel and electrophoresed. The dried gels were exposed to a phosphor screen. Imaging and quantification were performed using a PhosophorImager and ImageQuant software.

NMR Spectroscopy   NMR experiments were performed on a Varian UNITY plus 500 MHz spectrometer. All of the titration experiments were carried out at 30°C in a 90% H₂O/10% D₂O solution containing 150 mM KCl, 25 mM KH₂PO₄, and 1 mM EDTA (pH 7.0). A standard 1–1 echo pulse sequence with a maximum excitation centered at 12 ppm was used for water suppression. Thirty-two scans were acquired for each spectrum with a relaxation delay of 2 s. Two-dimensional Nuclear Overhauser Effect Spectroscopy spectra of exchangeable protons were collected at 30°C in time proportional phase incrementation mode with a mixing time of 200 ms using 2048 and 1024 complex points in t₂ and t₁ experiments, respectively. NMR data were processed and analyzed using the FELIX program (Molecular Simulations Inc.). Nearly all of the nonexchangeable DNA protons of the 5'-d[A(G₃T₂A)₃G₃]-3' and 5'-d[TAG₃T₂A]₄-3' in a free form or in a complex with ligand, except for several H5'/H5'' pairs, were assigned as described (49).

Model Building and Molecular Dynamics Simulations   The initial coordinates used in the model building process were those published for the NMR-based model of the 1:1 PIPER:5'-d[TAG₃T₂A]₄-3' complex (49) and the antiparallel human telomeric sequence d[AG₃(T₂AG₃)₃] (50). The necessary replacements and the addition of hydrogens were carried out using INSIGHT II (Ref. 51; Molecular Simulations Inc.), and position was refined by energy minimization (1000 cycles of steepest descent and 2 x 1000 cycles of conjugate gradient) while constraining the positions of heavy atoms. Finally, the entire structure was subjected to conjugate gradient minimization until convergence was reached, at which time constraints were gradually removed. Molecular dynamics simulations (100 ps) at 300 K and mechanics (2 x 1000 cycles of conjugate gradient minimization) were performed using DISCOVER with the consistent force field. This structure subsequently served as the starting structure for additional energy refinement, docking, molecular dynamics, and complex formation.

To explore the molecular interactions of QQ58 and its analogues, QQ27 and QQ28, with the G-quadruplex structures, molecular models were built and energy minimized using a protocol similar to that described for G-quadruplex structures. The obtained low energy structures were manually docked into the intercalation site, and energies were computed using the AFFINITY (52) program. To clarify the orientation of the ligands in the intercalation site, the electrostatic potentials at the van der Waals surface of the G-quadruplex were determined by using solvent surface calculations. The SA docking with 100 fs/stage duration (50 SA stages) was then performed to find the most favorable orientation. Thus, orientation with low intermolecular potential energy was obtained while moving ligand and ligand-binding G-tetrad residues. The resulting ligand-G-quadruplex complex trajectories were energy minimized using 1000 cycles of conjugate gradient minimizer, and the interaction energies were computed.

Cytotoxicity Assay   Human breast carcinoma MDA-MB-231 and prostate carcinoma DU-145 cell lines were purchased from American Tissue Culture Collection (Rockville, MD) and cultured according to the supplier’s instructions. Exponentially growing cells (1–2 x 10³ cells) in 0.1 ml of medium were seeded on day 0 in a 96-well microtiter plate. On day 1, 0.1-ml aliquots of medium containing graded concentrations of the investigational compound were added to the cell plates. At days 7–10, the cell cultures were incubated with 50 µl of 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl-tetrazolium bromide (1 mg/ml in Dulbecco’s PBS) for 4 h at 37°C. The resulting formazan precipitate was solubilized with 200 µl of 0.04 M HCl in isopropyl alcohol (53). For determination of the IC₅₀ values, the absorbance readings at 570 nm were fitted to the four-parameter logistic equation.

Sea Urchin Embryo Culture   Lytechinus pictus sea urchins (Marinus Inc., Long Beach, CA) were maintained at 15°C in refrigerated aquaria containing Instant Ocean artificial sea water. Spawning, fertilization, drug treatment, and embryo processing were done as described (54). Briefly, 10 min after insemination, the fertilized eggs were allowed to settle, and the supernatant was aspirated and replaced with fresh artificial sea water. The embryos were cultured at 18°C. Stock solutions (10 mM) of the tested agents were prepared in DMSO. Twenty min after fertilization, the agents were added to 1% embryo suspensions to a final concentration of 1 µM. Equivalent amounts of DMSO were added to control egg suspensions. At measured times after insemination (10 and 24 h), the embryos were pelleted by centrifugation. Video images of the embryos were captured with a Zeiss standard research microscope interfaced with a Javelyn video camera and a Panasonic time-lapse video recorder. The nuclei were stained by the Feulgen reaction, and the chromatin was visualized and photographed with an Olympus BH2 photomicroscope equipped with fluorescence optics, as described elsewhere (55).

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

 
Chemistry
The synthetic procedure (Fig. 2) was based on a published method for the synthesis of the quinobenzoxazines, with some modification (56). The synthesis was started with ethyl 2,3,4,5-tetrafluorobenzoylacetate (Fig. 2, 1), which was prepared according to published procedure (5). Treatment of 2,3,4,5-tetrafluorobenzoylacetate with triethyl orthoformate in acetic anhydride followed by 2-hydroxy-3-amino-dibenzofuran hydrobromide generated the enamino keto acid intermediate (Fig. 2, 2). This intermediate was treated with sodium hydride at -78°C and then heated to reflux to complete the double annulation (Fig. 2, 3). Regioselective nucleophilic substitution of fluoride with 3-(R)-(tert-butoxycarbonylamino)pyrrolidine was carried out to give the key intermediate (Fig. 2, 4). Final target molecules 5, 6, and 7 (Fig. 2) were prepared by hydrolysis and deprotection of 4, hydrolysis of 4, and deprotection of 4, respectively, in overall yields of 59, 59, and 44.7%, respectively.

Comparison of the Effects of A-62176 and QQ58 on Topoisomerase II and Telomerase Activity
On the basis of our previous insights into structural requirements (extended planarity, cationic interactions) for ligand binding with G-quadruplex structures, the fluoroquinolones seemed to be good candidates for structural modification. Our objective was to convert a topoisomerase II poison (A-62176) to a G-quadruplex-interactive compound. As this work progressed, a paper was published (4) that demonstrated that quinolones were telomerase inhibitors, suggesting to us that quinolones might interact with G-quadruplex structures. Our first objective was to compare the topoisomerase II inhibitory activities and G-quadruplex interactions of A-62176 and QQ58 to determine whether it would be possible to separate these activities.

Whereas A-62176 Is Both a Topoisomerase II Poison and a Catalytic Inhibitor, QQ58 Is Only a Catalytic Inhibitor   Experiments were designed to compare the effects of A-62176 and QQ58 on the activity of human topoisomerase II. The ability to cleave and religate double-stranded DNA is central to the physiological functions of topoisomerase II. Kinetoplast DNA is a massive network consisting of thousands of interlocked, closed circular DNA molecules called minicircles. Topoisomerase II introduces a double-strand break in DNA to release one minicircle from the network, which is called a decatenation reaction (57, 58). The in vitro effects of A-62176 and QQ58 were determined by using a decatenation assay. As shown in Fig. 3, A and B, QQ58 had about a 5-fold lower decatenation activity than A-62176. To additionally characterize the properties of A-62176 and QQ58, the contrasting effect on the DNA cleavage reaction of human topoisomerase II of QQ58 was determined. As shown in Fig. 3, C and D, the intensity of the topoisomerase II-mediated cleavage at site A decreased as the concentration of QQ58 was increased, whereas the DNA cleavage was initially enhanced at low concentrations of A-62176 and then decreased at high concentrations. These results show that the two fluoroquinolones act differently on human topoisomerase II: QQ58 acts as a catalytic inhibitor that inhibits the activity of topoisomerase II without trapping the cleaved complex, and A-62176 is a topoisomerase II poison that interferes with the breakage-rejoining reactions of topoisomerase II by trapping the enzyme covalent reaction intermediate, known as the cleaved complex.

View larger version (35K): [in this window] [in a new window]   Fig. 3 A, inhibition of decatenation of human topoisomerase II by QQ58 and A-62176. Catenated KDNA was treated with 2 units of human topoisomerase II in the presence of QQ58 or A-62176 followed by SDS/proteinase K and then analyzed on agarose gel. Lane 1 contains catenated KDNA only. Lane 2 contains KDNA and topoisomerase II without compounds. Lanes 3–7 contain 2, 5, 10, 20, and 50 µM of QQ58, respectively. Lanes 8–12 contain 0.5, 1, 2, 5, and 10 µM of A-62176, respectively. B, intensity of decatenated KDNA was quantitated from the gel using ImageQuant software. C, autoradiogram of a 12% denaturing polyacrylamide gel showing the topoisomerase II-mediated cleavage pattern in the presence of QQ58 or A-62176. The cleavage reaction was done in a reaction buffer as described in "Materials and Methods." Lane 3 contains DNA only. Lane 4 contains DNA and topoisomerase II without QQ58 or A-62176. Lanes 5–10 and 11–16 contain 0.2, 0.4, 1, 2, 5, and 10 µM QQ58 and A-62176, respectively. D, strand breakage products indicated by the arrow in C were quantitated using a PhosphorImager and ImageQuant software. The intensity of the DNA cleavage produced by QQ58 or A-62176 was determined from the volume of the bands normalized by the total radioactivity in each lane.

View larger version (30K): [in this window] [in a new window]   Fig. 4 Inhibition of human telomerase by A-62176, QQ58, and TMPyP4. A, autoradiogram showing telomerase inhibition with increasing concentrations of A-62176, QQ58, and TMPyP4. Lanes 1–4, 5–8, and 9–12 contain 12.5, 25, 50, and 100 µM of A-62176, QQ58, and TMPyP4, respectively. B, graphical representation of data from A.

View larger version (38K): [in this window] [in a new window]   Fig. 5 Inhibition of Taq polymerase with increasing concentrations of QQ58. A, cartoon of the assay. B, autoradiogram of the sequencing gel showing enhanced DNA synthesis pausing at the G-quadruplex site with increasing concentrations of QQ58 (Lanes 1–7). The free primer, the pause site, and the full-length product are indicated. C, graphical representation of the quantification of the sequencing gel shown in B, showing the concentration of QQ58 to the ratio of intensity of the bands obtained for the pausing site/total intensity per lane.

View larger version (21K): [in this window] [in a new window]   Fig. 6 Effect of QQ58 on the 1H NMR spectra of G-quadruplex structures. A, one-dimensional 1H NMR titration of d[TAG3T2A]4 with QQ58 to obtain a 1:1 QQ58:DNA ratio. The imino proton and aromatic proton regions of the 500 MHz NMR spectra are shown with increasing amounts of ligand added at 30°C. G3, G4, and G5 represent the imino proton resonances of the free DNA, whereas G3*, G4*, and G5* represent the imino proton resonances of the ligand-bound DNA. B, one-dimensional 1H NMR titration of d[AG3(T2AG3)3] with QQ58 to obtain a 1:1 QQ58:DNA ratio. The imino proton regions of the 500 MHz NMR spectra are shown with increasing amounts of ligand added at 30°C. The arrow points to the imino proton resonance of G22, which shifts on addition of QQ58.

Photo-mediated DNA Cleavage Reactions Confirm the Proximity of QQ58 to the External G-Tetrad   To additionally characterize the binding mode of QQ58 to the intramolecular G-quadruplex structures, a photo-mediated cleavage reaction of the QQ58-G-quadruplex complex was carried out. A DNA sequence that forms an intramolecular G-quadruplex under conditions standardized previously was used (12). The photo-reactive property of the FQPs leads to strand cleavage in the QQ58-G-quadruplex complex in the presence of light of a specific wavelength. This in turn gives important information about the drug localization in the complex. The photocleavage pattern produced by QQ58 on the intramolecular G-quadruplex is shown in Fig. 7A. With increased concentrations of the drug, there is enhanced cleavage at G1, G6, and G12, which correspond to specific guanines involved in the G-quadruplex. These guanines, G1, G6, and G12, correspond to the residues that form the lowermost G-tetrad (Fig. 7B), with G12 and G6 being the most reactive residues. This photocleavage data confirms that the drug is bound external to this G-tetrad. On the basis of the NMR studies and the photocleavage experiments, a cartoon (Fig. 7B) is proposed for the binding of QQ58 with the sequence that forms an intramolecular G-quadruplex structure (Ref. 12; Fig. 7B). G12 in this structure corresponds to G22 in the previous two-dimensional ¹H NMR study, providing confirmation of the location of QQ58 in the intramolecular fold-over G-quadruplex structure.

View larger version (41K): [in this window] [in a new window]   Fig. 7 QQ58 photo-mediated strand cleavage reaction of intramolecular G-quadruplex. A, autoradiogram of a photocleavage sequencing gel with increasing concentrations of QQ58. Lanes 1 and 2 are the AG/TC Maxam-Gilbert sequencing reactions, and Lanes 4–8 are increasing concentrations of QQ58 (1–25 µM). The enhanced cleavage sites are indicated by G1, G6, and G12. G7 is in the same tetrad but is not cleaved excessively. B, diagrammatic representation of the proposed external binding of QQ58 to the G-quadruplex below the lower-most G-tetrad based on the photocleavage results with the ligand.

View larger version (50K): [in this window] [in a new window]   Fig. 8 Molecular modeling of the d[AG3(T2AG3)3]-QQ58 (A) and d[TAG3T2A]4-QQ58 (B and C) complexes. A, structure of the d[AG3(T2AG3)3]-QQ58 complex in the final model after energy minimization. The G-quadruplex structure is shown as a stick model with compound QQ58 in color by atom type Corey-Pauling Koltun model. The proposed intercalation sites (yellow) are labeled. B, structure of the d[TAG3T2A]4-QQ58 complex in the final model after energy minimization. Coloring is as in A. C, stereo representation of the proposed binding mode of QQ58 in complex with parallel-stranded G-quadruplex shown in color by atom type. The white dotted lines show the hydrogen bonding distances in Å. The G-quadruplex complex is clipped to afford a better view of the intercalation site.

The Polymerase Stop Assay Demonstrates That QQ28 Has Only a Minimal Stabilizing Effect on G-Quadruplex Structure, Whereas QQ27 Has an Intermediate Effect   On the basis of molecular modeling results, we predicted that the steric bulk of the N-Boc group on the aminopyrrolidine of QQ28 should interfere with intercalation into the G-quadruplex structure and prevent any ionic interaction between the positively charged amino group of QQ28 and the anionic backbone of DNA. However, the ethyl ester of QQ58 should not dramatically affect the binding of QQ27. A polymerase stop assay was used to evaluate the comparative abilities of QQ58, QQ27, and QQ28 to stabilize the intramolecular fold-over G-quadruplex structure. The results shown in Fig. 9, A and B, demonstrate that, as predicted, QQ58 and QQ27 have approximately the same potency in inhibition of the Taq polymerase, whereas QQ28 was much less potent.

View larger version (27K): [in this window] [in a new window]   Fig. 9 Differential effects of QQ27, QQ28, and QQ58 on DNA synthesis arrest by G-quadruplex structure. A, autoradiogram of the sequencing gel showing Taq polymerase primer extension arrest at the G-quadruplex site in the presence of QQ58, QQ27, and QQ28. Lanes G, C, A, and T are DNA sequencing lanes. B, quantitation of the sequencing gel shown in A. The drug concentrations are plotted against the normalized ratios of the band intensity of the G-quadruplex pause site and the band intensity of the full-length product.

In this study, we evaluated the in vivo effects of 1 µM of QQ28, QQ27, and QQ58 in developing sea urchin embryos. Preliminary dose-dependence tests have shown that these concentrations are subtoxic and compatible with the assays. Antiproliferative effects of the agents were assessed by examining the morphology of the embryos cultured for 10 and 24 h in the presence of the three quinolones. Photomicrographs (Fig. 10A) show that at 10 h all of the embryos are still within fertilization coats. The control group and QQ28-treated embryos appear normal with well-defined blastocoels. Cleavage is delayed in QQ27- and QQ58-treated embryos, so there are fewer cells, and a blastocoel has not yet formed. At 24 h (Fig. 10B) all of the embryos have hatched and are swimming, which suggests lack of major toxicity of the examined compounds. The control group and QQ28-treated embryos appear normal with well-defined blastocoels into which PMCs have ingressed at the vegetal pole (seen as the thicker layer of cells in the bottom of the embryos). Invagination of the vegetal pole has begun in the control group and the QQ28-treated embryos. Because of the cleavage delay in the QQ27- and QQ58-treated embryos, the embryos are smaller and there are fewer cells. Additionally, the blastocoels appear to be abnormally filled with cells, and there does not appear to be a concentration of PMCs at a well-defined vegetal pole. The antiproliferative effect of QQ58 is apparently greater than that of QQ27, and this difference is more noticeable at 10 h than at the later time point at 24 h.

View larger version (71K): [in this window] [in a new window]   Fig. 10 Effects of FQPs on proliferation dynamics of early sea urchin embryos. The embryos were cultured in the presence of FQPs or DMSO (control) for 10 h (A) or 24 h (B), pelleted by centrifugation, and video micrographed as described in "Materials and Methods." A, 10 h: the thin surrounding coat elevated away from the surface of the embryo is the fertilization coat (FC), which is present in all embryos. The thick layer below the FC is the cellular layer of the early blastula embryo. At 10 h, there are 400–500 cells in it. These layers are fully formed in control and QQ28-treated embryos. There are fewer and larger cells in the QQ27- and QQ58-treated embryos, so the layer is thicker and not fully formed. The cellular layer surrounds the blastocoel, the fluid-filled cavity of the early embryo into which cells of the cellular layer will move during the formation of the larval gut. B, 24 h: no FCs are present in any of the embryos. This demonstrates that toxicity is not found at 1 µM because to hatch out of the FCs the embryos must synthesize a specific protease, hatching enzyme that degrades the FCs. Hatching occurs at 15 h in these embryos. In addition, all embryos produce cilia and begin to swim at the normal time, shortly before hatching, thus demonstrating their viability, although there are fewer cells in the QQ27- and QQ28-treated cells. In control and QQ28-treated embryos, development is normal. There is a thin cellular layer composed of 1000 cells that surrounds the blastocoel. Cells from the cellular layer have moved into the blastocoels. They are the PMCs. At 7:00 in the control embryo and at 9:00 in the QQ28-treated embryo the cellular layer has begun to invaginate to form the larval gut. At this stage it can be labeled the VP. In the QQ27- and QQ58-treated embryos, cell division has proceeded since 10 h, and there are obviously more cells. However, the embryos lack definitive blastocoels because cells seem to abnormally fill the blastocoels. Well-organized PMCs and obvious VPs are not present, although the thickened, flattened cellular layer in the QQ58-treated embryo at 3:00 could be a VP.

View larger version (68K): [in this window] [in a new window]   Fig. 11 Chromosomal effects of FQPs in early sea urchin embryos. The embryos were cultured in the presence of FQPs or DMSO (control) for 10 h and pelleted by centrifugation. The nuclei were stained by Feulgen reaction and the chromatin was visualized as described under "Materials and Methods." Chromosomal bridges are marked with arrows.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

View larger version (20K): [in this window] [in a new window]   Fig. 12 Evolution of QQ58 from norfloxacin and effects of QQ58 structural modification.

It seems likely that QQ58, like A-62176, will bind nonspecifically by an intercalative mechanism to duplex DNA, i.e., similar to that in which divalent cations are not involved. An important objective is to design G-quadruplex-interactive compounds belonging to the FQP group that clearly differentiate between duplex and G-quadruplex DNA.

Molecular Recognition of the G-Quadruplex Structure by QQ58
Despite the contrasts in the molecular targets (duplex topoisomerase II versus G-quadruplex structures) of A-62176 and QQ58, there are common molecular features of both molecules that appear critical for recognition of unique features of the two different receptors. These are the amphoteric nature of both molecules, the intramolecular separation distance of the positive and negative charges, and the planar nature of both the quinobenzoxazine and FQP molecules. For the FQP molecule QQ58, the carboxylic acid and amino group of the pyrrolidine are both involved in donor-acceptor interactions with N7 of guanine in the external tetrad and the anionic backbone of the G-quadruplex structure, respectively. The charge separation of the amphoteric groups on QQ58 is precisely that required for in-register interaction with the complementary groups on the G-quadruplex structure. This observation provides the insight into why even norfloxacin shows some modest telomerase inhibition. However, because norfloxacin and A-62176 have minimal stacking interactions within the G-tetrad-thymine step of the G-quadruplex structures, there is presumably insufficient - interaction to adequately stabilize the complex. The relative importance of the charge-charge interactions and the stacking is illustrated by the comparative modeling and biological studies of QQ58, QQ27, and QQ28. Whereas steric disruption of the two in-register donor-acceptor pairs in the QQ28-G-quadruplex interactions almost completely eliminates G-quadruplex binding and biological activity, the sterically tolerated ethylester of QQ27 still permits one of two H-bonds and may partially make up for the post-electrostatic interactions by increasing the van der Waals interactions, thus permitting residual G-quadruplex binding and biological activity that is almost as good as QQ58. Most important, our QQ58-G-quadruplex model, which is based on ¹H NMR and photocleavage data, along with molecular modeling, accounts fully for the modulation of activity of QQ27 and QQ28 and also provides an important starting point for the design of more potent compounds.

Biological Effects of the FQPs
The three quinolones QQ28, QQ27, and QQ58 examined in this study exhibit different modes of interaction with G-quadruplexes and effects on topoisomerase II activity. Interestingly, QQ28, which neither interacts with G-quadruplex nor poisons topoisomerase II, showed the lowest cytotoxicity in two model human tumor cell lines in vitro and had no major antiproliferative effect on developing sea urchin embryos in vivo. The G-quadruplex-interactive compounds QQ58 and QQ27, which do not poison topoisomerase II, showed increased cytotoxicity in human tumor cells and sea urchin embryos. In both systems, the antiproliferative effects of QQ58 were more potent than those of QQ27.

Although QQ27 and QQ58 significantly delayed growth of the embryos at the early stages, the development of the embryos was not largely impaired, which is consistent with a lack of major cytotoxic effects. Under subtoxic conditions, we were able to observe chromosome end fusions induced only by these compounds. The chromosomal abnormality exhibited specifically by embryos treated with QQ27 and QQ58 is an apparent end-to-end adherence of anaphase and telophase chromosomes. The end-to-end adhesions are manifested in the form of anaphase chromosomes that often extend from pole to pole of the mitotic spindle. These effects are very similar to the chromosomal fusions that result from a mutation in the telomerase template in Tetrahymena (66) and a dominant negative mutation in the telomere binding protein TRF2 in HTC75 cells (67).

The end-to-end adherence of chromosomes is observed in embryos treated with structurally dissimilar telomere- and telomerase-interactive agents designed to target very different sites in the telomere-telomerase complex. We have found that the G-quadruplex-interactive porphyrin TMPyP4 showed the most potent chromosome destabilizing properties with no significant inhibition of DNA synthesis (54). Because the isomeric porphyrin TMPyP2, which does not show stacking interactions with G-quadruplex, has not demonstrated similar effects in sea urchins and other systems examined to date, the telomere effects of TMPyP4 might be related to stabilization and/or de novo formation of G-quadruplex in cells. This specific chromosomal effect has also been induced by an oligonucleotide telomere mimic but not the scrambled sequence (54), and by other agents, of which the mechanisms of action might involve regulation of telomere maintenance (62, 63).

By analogy, the chromosome destabilizing activity of QQ27 and QQ58 suggests that these compounds might also affect telomere integrity. The results suggest that the G-quadruplex interactions are linked with the ability of FQP to induce chromosome destabilization. Poisoning of topoisomerase II does not apparently play a role in the process, because QQ58 and QQ27, which do not poison topoisomerase II, stimulated end-to-end adherence of chromosomes. We have demonstrated¹⁵ that the quinolones can stabilize, but do not stimulate, de novo formation of G-quadruplex. Thus, the present results imply that G-quadruplex-like structures may well exist in vivo. Additional investigation of this issue is warranted.

	Acknowledgments

 
We thank the other members of the National Cooperative Drug Design Group team in Austin and San Antonio for valuable discussions. We also thank Dr. David Bishop, College of Pharmacy, The University of Arizona, Tucson, AZ for preparing, proofreading, and editing the final version of the manuscript, figures, schemes, and tables.

	Footnotes

 
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

¹ Supported by NIH Grants CA88310 and CA67760.

² These authors contributed equally to this paper.

³ Present address: College of Pharmacy, The University of Arizona, Tucson, AZ 85721.

⁴ Present address: Monsanto Co., St. Louis, MO 63167.

⁵ Present address: Department of Biomolecular Sciences, University of Manchester Institute of Science and Technology, Manchester M60 1QD, England.

⁶ Present address: Cancer Metastasis Research Center, Yonsei University College of Medicine, Seoul, Korea.

⁷ Present address: College of Pharmacy and Department of Chemistry, The University of Arizona, Tucson, AZ 85721 and Arizona Cancer Center, 1515 North Campbell Avenue, Tucson, AZ 85724.

⁸ To whom requests for reprints should be addressed, at Arizona Cancer Center, 1515 N. Campbell Ave., Tucson, AZ 85724. Phone: 520-626-5622; Fax: 520-626-5623; E-mail: hurley{at}pharmacy.arizona.edu.

⁹ D-F. Shi, R. T. Wheelhouse, D. Sun, and L. H. Hurley. Quadruplex-interactive agents as telomerase inhibitors: synthesis of porphyrins and structure-activity relationship for the inhibition of telomerase. J. Med. Chem., in press, 2001.

¹⁰ M. A. Shammas, R. B. Batchu, J. Y. Wang, L. H. Hurley, R. J. S. Reis, and N. Munshi. Telomerase inhibition and cell growth arrest following porphyrin treatment of multiple myeloma cells, submitted for publication.

¹¹ The abbreviations used are: FQP, fluoroquinophenoxazine; NMR, nuclear magnetic resonance; CI, confidence interval; SA, simulated annealing; PIPER, N,N'-bis[2-(-1-piperidino)ethyl]-3,4,9,10-perylenetetracarboxylic diimide; PMC, primary mesenchyme cell; VP, vegetal plate.

¹² A. Rangan, unpublished observations.

¹³ O. Fedoroff, unpublished observations.

¹⁴ E. Izbicka, manuscript in preparation.

¹⁵ A. Rangan, unpublished observations.

Received 8/17/01; revised 9/21/01; accepted 10/ 2/01.

	References

Top Abstract Introduction Materials and Methods Results Discussion References

 

1. Wang, J. C. DNA topoisomerases.Annu. Rev. Biochem. , 65:635 –692,1996 .[CrossRef][Medline]
2. Reece, R. J., and Maxwell, A. DNA gyrase: structure and function.CRC Crit. Rev. Biochem. Mol. Biol. , 26:335 –375,1991 .
3. Nitiss, J. L. Investigating the biological functions of DNA topoisomerases in eukaryotic cells.Biochim. Biophys. Acta , 1400:63 –81,1998 .[Medline]
4. Yamakuchi, M., Nakata, M., Kawahara, K., Kitajima, I., and Maruyama, I. New quinolones, ofloxacin and levofloxacin, inhibit telomerase activity in transitional cell carcinoma cell lines.Cancer Lett. , 119:213 –219,1997 .[CrossRef][Medline]
5. Zeng, Q., Kwok, Y., Kerwin, S. M., Mangold, G., and Hurley, L. H. Design of new topoisomerase II inhibitors based upon a quinobenzoxazine self-assembly model.J. Med. Chem. , 41:4273 –4278,1998 .[CrossRef][Medline]
6. Sun, D., Thompson, B., Cathers, B. E., Salazar, M., Kerwin, S. M., Trent, J. O., Jenkins, T. C., Neidle, S., and Hurley, L. H. Inhibition of human telomerase by a G-quadruplex-interactive compound.J