Inhibition of Simian Virus 40 Large T Antigen Helicase Activity by Fluoroquinolones Syed Hamid Ali1*, Anil Chandraker 2 and James a Decaprio 3

1_0131_Ali.qxp 1/2/07 12:36 Page 1

Antiviral Therapy 12:1–6 Inhibition of Simian virus 40 large T antigen helicase activity by fluoroquinolones Syed Hamid Ali1*, Anil Chandraker 2 and James A DeCaprio 3

1Department of Biological and Biomedical Sciences, The Aga Khan University, Karachi, Pakistan 2Transplantation Research Center, Brigham and Womens Hospital, Boston, MA, USA 3Department of Medical Oncology, Dana-Farber Cancer Institute, Boston, MA, USA

*Corresponding author: Tel: +92 21 486 4435; Fax: +92 21 493 4294/493 2095; E-mail: [email protected]

Background: Fluoroquinolones represent a potent group trovafloxacin, ciprofloxacin and ofloxacin, were tested of antibiotics that inhibit bacterial DNA replication by for their ability to inhibit viral DNA replication. targeting the essential bacterial enzymes gyrase and Results: We show here that all four quinolones tested topoisomerase IV. Inhibition of gyrase activity by were effective in the inhibition of SV40 plaque formation quinolones involves the interaction of these drugs with and DNA replication in CV1-P cells. In addition, we found the helicase component of bacterial gyrase. DNA that each of these quinolones was inhibitory to the heli- tumour viruses also encode helicases that are essential case activity of SV40 large tumour antigen. for their DNA replication in the host. Conclusions: Fluoroquinolones and their derivates may Methods: In this study we have evaluated the effect of therefore be useful in the treatment and/or prevention of fluoroquinolones on viral DNA replication using the DNA infection by SV40-homologous human DNA viruses that tumour virus Simian virus 40 (SV40) as our model. Four encode helicase activity for their survival. different fluoroquinolones, namely, levofloxacin,

Introduction

Simian virus 40 (SV40) large tumour antigen (T Ag) targets of (fluoro)quinolones [9]. In contrast with plays a key role in SV40-mediated transformation as another group of drugs, coumarins, which inhibit well as in viral replication. T Ag possesses several gyrase by targeting its ATPase subunit GyrB [10], biochemical activities including helicase, ATPase and quinolones block DNA replication by interacting with DNA-binding activities that are essential for viral the DNA-binding subunit of gyrase, GyrA, thus replication. In addition, Large T Ag serves to recruit converting gyrase into a cellular poison [11]. This cellular proteins including replication protein A, DNA quinolone poisoning of gyrase is effected by freezing of polymerase and topoisomerase I that participate in the gyrase–DNA complex into a ternary viral DNA replication [1]. Human DNA polyoma gyrase–quinolone–DNA complex, which leads to the viruses BKV and JCV encode SV40-homologous disruption of replication and double-stranded DNA T Ags that play a similar role in viral replication [2,3]. breaks, eventually resulting in cytotoxicity [12,13]. BK and JC viruses are thought to contribute to the To test whether SV40 could be inhibited by fluoro- pathogenesis of several human diseases including quinolones, the ability of levofloxacin, trovafloxacin, haemorrhagic cystitis, BK nephropathy and progres- ciprofloxacin and ofloxacin to inhibit viral DNA repli- sive multifocal leukoencephalopathy. cation was tested. We show here that all four In clinical trials, fluoroquinolones have been found quinolones tested were effective in the inhibition of to be effective against DNA [4] as well as RNA [5,6] SV40 plaque formation and DNA replication in CV1-P viruses. The ability of fluoroquinolones to alter cells. In addition, we found that each of these protein–DNA binding in vitro has shown promise for quinolones was inhibitory to the helicase activity of the therapy against HIV as well [7]. It has been previously SV40 T Ag. In view of what we report here, fluoro- reported that the fluoroquinolones nalidixic acid and quinolone drugs may be effective in treating infections oxolinic acid can inhibit replication and the cytopathic by viruses that encode helicase activity for their effect of BK virus in cell culture [8]. Type II topoiso- survival. The results presented here could be useful in merase, gyrase and topoisomerase IV are the cellular designing novel quinolone-related antiviral drugs.

SH Ali et al.

Materials and methods Results and discussion

SV40 plaque assay Previous in vitro and in vivo studies have shown that CV1-P cells were routinely grown in Dulbecco’s quinolones can be effective against several DNA and minimum essential medium (Gibco BRL, Gaithersburg, RNA viruses [6,8,22]. To test if fluoroquinolones could MD, USA) with 10% Fetal Clone (Hyclone, Logan, UT, inhibit the growth of SV40, we tested four fluoro- USA). The cells were grown to 85% confluency in quinolones, namely, levofloxacin, trovafloxacin, 35-mm wells. The monolayer was then infected with ciprofloxacin and ofloxacin. SV40 was grown in CV1- SV40 for 3–5 h in complete medium containing 10% P cells in the absence or presence of these quinolones. Fetal Clone, using 20 μl of viral suspension per 35-mm All four quinolones tested were able to inhibit the cyto- well. The cells were fed on alternate days with fresh pathic effect of SV40 polyoma virus on CV1-P mono- complete medium containing the desired quinolone and layers in the viral plaque assay (Figure 1). allowed to grow for 8–10 days or until the cytopathic Comparatively, trovafloxacin proved to be the most effect (CPE) became visible. The monolayers were then potent inhibitor of the viral CPE, showing inhibition at fixed in ethanol, stained with crystal violet and 10 μM. levofloxacin was found to be inhibitory at photographed. 60 μM, whereas ciprofloxacin and ofloxacin inhibited the CPE at 1–2 mM. Quantitation of viral DNA To rule out the possibility that the four quinolone The monolayers of CV1-P cells were infected with drugs were inhibitory to the viral replication due to SV40 and incubated in the presence of a quinolone, their deleterious effects on CV1-P cells, the cells were as described above. The wells were fed every 48 h studied in the presence of these drugs alone. with complete medium containing the quinolone. Monolayers of CV1-P cells were grown for 5 days in The cells were lysed after various time intervals and the presence of the minimum concentration of each of the viral DNA was extracted using Qiagen miniprep the four quinolones that was found sufficient to inhibit columns (Qiagen, Inc., Valencia, CA, USA). The puri- viral DNA replication. Daily cell counts and cell-cycle fied viral DNA was then run on an agarose gel, trans- analysis by flow cytometry analysis (FACS) revealed ferred to a nitrocellulose membrane and Southern that CV1-P cells were unaffected by quinolone treat- blotted, using a 32P-labelled PfMl1-BsgI fragment ment. from the SV40 genome as probe. The DNA thus To study the effect of these drugs on the replication detected on Southern blots was then quantitated on of SV40 DNA, the experiment was set up in the same a phosphoimager. way as the SV40 plaque assay. Following SV40 infection, viral DNA from the CV1-P cells was extracted at Helicase assay various time intervals and quantified on Southern The substrate was prepared by annealing the single- blots. All four drugs were found to inhibit the SV40 stranded, circular M13mp18 plasmid with a 32P- DNA replication (Figure 2). At 48 h, no effect of labelled 35-base-long oligonucleotide that was quinolones was observed on the DNA replication (data complementary to an M13p18 sequence. The not shown). After 72 or 96 h, however, a clear inhibi- substrate was purified using commercially available tion of the SV40 DNA replication by quinolones was columns and stored at 4°C. To perform the helicase observed. As with the CPE assay, trovafloxacin was assay, 0.5 μg purified T Ag (CHIMERx, Milwaukee, found to have the strongest inhibitory effect at 10 μM. WI, USA) was incubated with the substrate at 37°C Levofloxacin and ciprofloxacin showed a comparable for 45 min. The reaction mixture contained 40 mM level of inhibition, whereas ofloxacin was to found to

Tris-HCl pH 7.6, 5 mM MgCl2, 2 mM dithiothreitol be the least potent of the four quinolones (Figure 2). In (DTT), 2 mM ATP, 25 mM KCl, 2% glycerol and the plaque assays and for the DNA replication experi- bovine serum albumin (BSA) (100 μg/ml). The heli- ments, the quinolone nalidixic acid was routinely used case assay was performed in the presence or absence as a positive control. Since nalidixic acid is known to of the quinolone. The reaction was stopped by inhibit the DNA replication of DNA tumour viruses adding a STOP buffer containing 50 mM ethylenedi- [8], we were able to successfully use it as a positive aminetetraacetic acid (EDTA), 2% sodium dodecyl control to validate our methodology (data not shown). sulphate (SDS), 40% glycerol and 0.1% To further investigate the mechanism of action of the bromophenol blue. The mixture was loaded onto an quinolones on SV40 replication, we sought to deter- 8% polyacrylamide gel containing 15% glycerol. The mine the effect of these drugs on the helicase activity of gel was allowed to run at 150 V, terminated after SV40 T Ag. In this assay, the ability of purified T Ag to 90 min, and dried. The signal was detected and release a 35-base complementary oligonucleotide from quantitated using a phospho-imager. M13mp18 plasmid was determined in the presence of

Fluoroquinolone inhibition of SV40 T helicase

Figure 1. Inhibition of SV40 CPE by quinolones

Levofloxacin

mM 0.01 0.02 0.04 0.06 0.08

Ciprofloxacin

mM 0.01 0.05 0.1 0.2 0.2 Ofloxacin

mM 0.01 0.05 0.1 0.2 0.2

Trovafloxacin

0.02

mM 0.001 0.01 0.02 0.05 0.05

Virus + + + + –

Monolayers of CV-1P cells were infected with Simian virus 40 (SV40) and incubated in the presence of the indicated concentrations of quinolones until cytopathic effect (CPE) was observed (approximately 10 days). When CPE became apparent, the cells were fixed, stained and photographed. Concentrations of antibiotics are given in mM. The far right-hand panels in each experiment show the cells treated with the highest dose of the drugs alone, to evaluate any deleterious effects that the drugs may have on the growth of cells. No such effects were observed on the cells.

Antiviral Therapy 12:1 3 1_0131_Ali.qxp 1/2/07 12:36 Page 4

SH Ali et al.

Figure 2. Inhibition of SV40 DNA replication by quinolones A 72 h 900 800 700 600 500 400 300 200 SV40 DNA 100 0 0.0 0.01 0.02 0.05 0.1 0.2 0.5 0.1 0.2 0.5 0.02 0.05 0.1 mM

Trovafloxacin Ciprofloxacin Ofloxacin Levofloxacin B

96 h 1,400 1,200 1,000 800 600 400

SV40 DNA 200 0 0.0 0.01 0.02 0.05 0.1 0.2 0.5 0.1 0.2 0.5 0.02 0.05 0.1 mM

Trovafloxacin Ciprofloxacin Ofloxacin Levofloxacin

CV-1P cells were infected and incubated in the presence of quinolones as in Figure 1. The cells were harvested after either (A) 72 h or (B) 96 h following infection. The viral DNA was extracted, resolved by electrophoresis and Southern blotted using a fragment of Simian virus 40 (SV40) genome as the probe. The viral DNA thus detected was then quantified. The figure shows a representative Southern blot. Each bar represents an average of two independent samples. Concentrations of antibiotics are given in mM.

various concentrations of each of the four quinolones. shown). Four other quinolones, namely, levofloxacin, Once again, all of the four quinolones were found to be trovafloxacin, ciprofloxacin and ofloxacin, were also inhibitory to the helicase activity of T Ag. able to produce the same inhibitory effect on SV40 Ciprofloxacin, levofloxacin and ofloxacin significantly CPE and DNA replication (Figures 1 and 2). The (P≤0.02) inhibited T Ag’s helicase activity (as deter- quinolone drugs, in general, therefore appear to have a mined by the Student’s t-test) at 0.5, 1.0 and 2.0 mM, similar inhibitory effect on the homologous DNA respectively, whereas trovafloxacin was effective at tumour viruses BKV and SV40. Since quinolones can 50 μM (Figure 3). Trovafloxacin was therefore found act on type II isomerases, the in vitro inhibition of viral to be the most potent of the four quinolones in CPE and DNA replication we observed could involve inhibiting the SV40 CPE and DNA replication, as well the interaction of quinolones with host topoisomerases, as the T Ag’s helicase activity. which SV40 T Ag binds and utilizes for viral replica- In this study we show that four quinolone drugs, tion. To find out if the observed inhibition of SV40 known for their ability to inhibit prokaryotic gyrase/ DNA replication by quinolones was a direct effect on topoisomerase IV, can also efficiently inhibit SV40 the viral replication proteins, we used purified SV40 T growth and DNA replication, possibly by targeting the Ag in a helicase assay with or without the quinolone in helicase activity of SV40 T Ag. The work that question. It was found that quinolones, in fact, inhib- prompted this study was reported by Portolani et al. ited the helicase activity of T Ag, known to be crucial [8], in which they demonstrated the inhibition of BK for SV40 replication. Our finding that quinolones can viral DNA replication by two quinolones, nalidixic inhibit a helicase independently may lead to new acid and oxolinic acid. For our study we used SV40 as insights into the mechanism of their action. our model, since it is closely related to BK virus. Using In our experiments, the in vitro inhibition of T Ag SV40, we were indeed able to reproduce the viral DNA helicase activity took place at a concentration consid- replication inhibition by nalidixic acid (data not erably higher than that which was sufficient for viral

Fluoroquinolone inhibition of SV40 T helicase

Figure 3. Inhibition of SV40 T Ag helicase activity by quinolones

% Oligo released 10

T Ag, µg 0.0 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5

Drug, mM 0.0 0.0 0.1 1.0 0.1 0.5 1.0 2.0 0.01 0.05

Levofloxacin Ciprofloxacin Ofloxacin Trovafloxacin

Purified Simian virus 40 (SV40) tumour antigen (T Ag) was incubated with a double-stranded substrate, with or without the quinolone, as described in Materials and methods. The radiolabelled oligonucleotide released by T Ag was separated by electrophoresis and quantified. The results are expressed as percentage of the oligonucleotide released from the total substrate used in each in sample. The figure shows a representative gel and its quantification. Concentrations of antibiotics are given in mM. Each bar represents mean of triplicates ± standard deviation. Oligo, oligonucleotide.

plaque formation inhibition. A similar observation has mutations in the DNA sequence of quinolone-resistant been made in studies that compared quinolone inhibi- bacterial strains [14]. The nature of helicase–DNA (or tion of bacterial growth with in vitro inhibition of helicase–quinolone–DNA) interactions, on the other bacterial gyrase activity [14]. It should be noted, hand, still remains elusive. A comparison between how however, that the duration of the in vitro SV40 heli- quinolone might interfere with helicase–DNA versus case assay was much shorter than that for SV40 CPE gyrase–DNA complex would be difficult to draw at this or DNA replication assays. Whereas the helicase point. It is tempting to speculate, however, that assays were carried out for less than 1 h, the CPE and quinolone–helicase–DNA frozen complexes may be replication assays were completed in days. It is similar to the gyrase–quinolone–DNA complexes. possible therefore that a lower concentration of There are conflicting reports on how the replication quinolones, ineffective over a period of few hours, is poison, gyrase–quinolone–DNA complex, is assem- ultimately sufficient against the virus when used bled. There are reports demonstrating that the consistently over longer periods. In the DNA replica- quinolone norflavin traps gyrase–DNA intermediates tion assays as well, moderate quinolone concentra- by binding not the enzyme but the DNA [20]. Another tions exhibited a delayed inhibitory effect; whereas the group, however, has shown evidence that the binding replicated viral DNA could be detected in CV1-P cells of norflavin to the gyrase–DNA intermediate is stimu- after 48 h, no inhibition by quinolones was observed lated by the formation of gyrase–DNA complex [21]. before 72 h of treatment. Considering our current understanding of the mecha- Gyrase decatenates and untwines the DNA double nism of gyrase action, and of the inhibition of gyrase by helix by creating double-stranded breaks in the helix, quinolones, it can be speculated that a related mecha- utilizing ATP as the energy source [15]. Helicase, on the nism may be operative in the quinolone inhibition of other hand, unwinds the DNA double helix by melting SV40 T Ag helicase. the H-bonds and bringing the single DNA strands Keeping in view the above-mentioned findings on apart, hydrolysing ATP in the process [16]. The crystal coumarins [10] and quinolones [11], it appears unlikely structures of some gyrases [17] as well as some hexam- that quinolones inhibit the ATPase activity of SV40 eric helicases, including SV40 T Ag, have been solved T Ag helicase. It may be speculated that, as is the case [18,19]. In comparison with that of the helicase, the in the inhibition of gyrase GyrA, quinolone may inhibit DNA–gyrase interaction is relatively well-defined in T Ag helicase activity by disrupting T Ag–DNA inter- terms of the interacting residues of gyrase and DNA. action. This disruption may involve a direct binding of The interactions involved in the ternary the quinolone to the DNA that T Ag uses as substrate. gyrase–quinolone–DNA have also been deduced from Alternatively, T Ag may form a complex with the DNA the study of gyrase crystal structure [17] and of the substrate that is then converted into a frozen

Antiviral Therapy 12:1 5 1_0131_Ali.qxp 1/2/07 12:36 Page 6

SH Ali et al.

DNA–quinolone–T Ag intermediate. Subsequent 10. Sugino A, Higgins NP, Brown PO, Peebles CL, Cozzarelli NR. Energy coupling in DNA gyrase and the mechanism of experiments could be performed to study the binding action of novobiocin. Proc Natl Acad Sci USA 1978; intermediates formed by DNA, quinolone and T Ag. 75:4838–4842. 11. Kreuzer KN, Cozzarelli NR. Escherichia coli mutants ther- mosensitive for deoxyribonucleic acid gyrase subunit A: Acknowledgements effects on deoxyribonucleic acid replication, transcription, and bacteriophage growth. J Bacteriol 1979; 140:424–435. This work was supported in part by National Institutes 12. Froelich-Ammon SJ, Osheroff N. Topoisomerase poisons: harnessing the dark side of enzyme mechanism. J Biol of Health grants RO1CA093804 and PO1CA50661. Chem 1995; 270:21429–21432. Publication costs were paid for with a grant from the 13. Reece RJ, Maxwell A. DNA gyrase: structure and function. University Research Council, Aga Khan University. Crit Rev Biochem Mol Biol 1991; 26:335–375. 14. Levine C, Hiasa H, Marians KJ. DNA gyrase and topoisomerase IV: biochemical activities, physiological roles during References chromosome replication, and drug sensitivities. Biochim Biophys Acta 1998; 1400:29–43. 1. Simmons DT. SV40 large T antigen functions in DNA replication and transformation. Adv Virus Res 2000; 15. Wang JC. Moving one DNA double helix through another 55:75–134. by a type II DNA topoisomerase: the story of a simple molecular machine. Q Rev Biophys 1998; 31:107–144. 2. Caputo A, Barbanti-Brodano G, Wang E, Ricciardi RP. Transactivation of BKV and SV40 early promoters by BKV 16. Marians KJ. Crawling and wiggling on DNA: structural and SV40 T-antigens. Virology 1986; 152:459–465. insights to the mechanism of DNA unwinding by helicases. Structure Fold Des 2000; 8:R227–235. 3. Tavis JE, Frisque RJ. Altered DNA binding and replication activities of JC virus T-antigen mutants. Virology 1991; 17. Morais Cabral JH, Jackson AP, Smith CV, Shikotra N, 183:239–250. Maxwell A, Liddington RC. Crystal structure of the 4. Josephson MA, Williams JW, Chandraker A, Randhawa breakage-reunion domain of DNA gyrase. Nature 1997; PS. Polyomavirus-associated nephropathy: update on 388:903–906. antiviral strategies. Transpl Infect Dis 2006; 8:95–101. 18. Li D, Zhao R, Lilyestrom W, et al. Structure of the replica- 5. Kojima H, Kaita KD, Hawkins K, Uhanova J, Minuk GY. tive helicase of the oncoprotein SV40 large tumour antigen. Use of fluoroquinolones in patients with chronic hepatitis Nature 2003; 423:512–518. C virus-induced liver failure. Antimicrob Agents 19. Singleton MR, Sawaya MR, Ellenberger T, Wigley DB. Chemother 2002; 46:3280–3282. Crystal structure of T7 gene 4 ring helicase indicates a 6. Negro F, Male PJ, Perrin L, Giostra E, Hadengue A. mechanism for sequential hydrolysis of nucleotides. Cell Treatment of chronic hepatitis C with alpha-interferon plus 2000; 101:589–600. ofloxacin in patients not responding to alpha-interferon 20. Shen LL, Pernet AG. Mechanism of inhibition of DNA alone. J Hepatol 1998; 29:369–374. gyrase by analogues of nalidixic acid: the target of the 7. Richter SN, Palu G. Inhibitors of HIV-1 Tat-mediated drugs is DNA. Proc Natl Acad Sci USA 1985; 82:307–311. transactivation. Curr Med Chem 2006; 13:1305–1315. 21. Shen LL, Mitscher LA, Sharma PN, et al. Mechanism of 8. Portolani M, Pietrosemoli P, Cermelli C, et al. Suppression inhibition of DNA gyrase by quinolone antibacterials: a of BK virus replication and cytopathic effect by inhibitors cooperative drug–DNA binding model. Biochemistry 1989; of prokaryotic DNA gyrase. Antiviral Res 1988; 28:3886–3894. 9:205–218. 22. Richter S, Parolin C, Palumbo M, Palu G. Antiviral proper- 9. Higgins PG, Fluit AC, Schmitz FJ. Fluoroquinolones: struc- ties of quinolone-based drugs. Curr Drug Targets Infect ture and target sites. Curr Drug Targets 2003; 4:181–190. Disord 2004; 4:111–116.

Accepted for publication 29 August 2006