Original Article Fluoroquinolones Inhibit HCV by Targeting Its Helicase

Antiviral Therapy 2012; 17:467–476 (doi: 10.3851/IMP1937)

Original article Fluoroquinolones inhibit HCV by targeting its helicase

Irfan A Khan1,2, Sammer Siddiqui1, Sadiq Rehmani1, Shahana U Kazmi2, Syed H Ali1,3*

1Department of Biological and Biomedical Sciences, Aga Khan University, Karachi, Pakistan 2Department of Microbiology, University of Karachi, Karachi, Pakistan 3Department of Microbiology, Dow University of Health Sciences, Karachi, Pakistan

*Corresponding author e-mail: [email protected]

Background: HCV has infected >170 million individuals of 12 different fluoroquinolones. Afterwards, Huh-7 and worldwide. Effective therapy against HCV is still lacking and Huh-8 cells were lysed and viral RNA was extracted. The there is a need to develop potent drugs against the virus. extracted RNA was reverse transcribed and quantified by In the present study, we have employed two culture models real-time quantitative PCR. Fluoroquinolones were also to test the activity of fluoroquinolone drugs against HCV: a tested on purified NS3 protein in a molecular-beacon- subgenomic replicon that is able to replicate independently based in vitro helicase assay. in the cell line Huh-8 and the Huh-7 cell culture model Results: To varying degrees, all of the tested fluoroqui- that employs cells transfected with synthetic HCV RNA to nolones effectively inhibited HCV replication in both produce the infectious HCV particles. Fluoroquinolones have Huh-7 and Huh-8 culture models. The inhibition of HCV also been shown to have inhibitory activity against certain NS3 helicase activity was also observed with all 12 of the viruses, possibly by targeting the viral helicase. To tease out fluoroquinolones. the mechanism of the antiviral activity of fluoroquinolones, Conclusions: Fluoroquinolones inhibit HCV replication their effect on HCV NS3 helicase protein was also tested. possibly by targeting the HCV NS3 helicase. These drugs Methods: Huh-7 cells producing the HCV virion as well hold promise for the treatment of HCV infection. as Huh-8 cells were grown in the presence or absence

Introduction

The burden of HCV infection is substantial world- Fluoroquinolones have been found effective against wide, with an overall prevalence of 2.2–3.0% [1]. The vaccinia virus and papovaviruses [6,7]. Such studies current recommended treatment for HCV combines prompted a series of efforts to synthesize modified fluo- pegylated interferon (IFN)-a2a and a broad spectrum roquinolones to enhance their antiviral activities [6]. antiviral drug, ribavirin. Such a regimen has been Various fluoroquinolones with broad-spectrum antivi- shown to be only moderately effective in combating ral activities were identified and tested in clinical trials HCV infection, with considerable variation across the with varying outcomes [6,8,9]. On the basis of these HCV genotypes [2,3]. Hence, more effective thera- and our own studies on fluoroquinolone-mediated inhi- peutic options need to be identified, which should bition of simian virus 40 [10], we hypothesized that preferably incorporate a short treatment duration and fluoroquinolones might possess an inhibitory activity low toxicity profile. against HCV. Quinolones form a major group of antibacterial Recently, the availability of efficient virus culture drugs, active against a broad spectrum of microorgan- systems for HCV [11,12] has made it possible to screen isms. These drugs target the bacterial enzymes DNA for antiviral drugs against this virus. To test antiviral gyrase and topoisomerase IV [4], specifically interact- activity of fluoroquinolones against HCV, two cell cul- ing with the DNA-binding unit of gyrase called Gyr A. ture systems were employed in the current study: first, Such an interaction leads to impaired bacterial DNA the human hepatoma (Huh)-7 cell line, transfected with replication and double-stranded DNA breaks [5]. The a chimeric HCV RNA, capable of producing complete versatility of the quinolone molecule has led to the dis- HCV virions and second, the Huh-8 cell line, harbour- covery of various quinolone-based bioactive agents that ing an autonomously replicating HCV subgenomic also exhibit antiviral activities [6]. replicon [11,13]. A fluorescence-based helicase assay

AVT-11-OA-2098_Khan.indd 467 10/04/2012 11:36:25 IA Khan et al.

[14] was also used to test the effect of fluoroquinolones centrifuge tube. To this mixture was added 500 µl on the activity of NS3 helicase. Here, we show that all HEPES buffer, pH 7.0 (50 mM HEPES, 280 mM NaCl

of the 12 fluoroquinolones used in this study inhibited and 1.5 mM Na2HPO4) dropwise on a vortex mixer HCV production and HCV RNA replication, as well as for 30 s. After incubation for 30 min, the mixture was the activity of HCV NS3 helicase. added to Huh-7 cells plated in 60×15 mm cell culture dishes [15]. After 3–4 h, the medium was aspirated and Methods the cells were replenished with fresh complete medium [10,12]. After 48 h, the expression of transfected Plasmids and cell lines RNA in the cells was confirmed by immunostain- Full-length FL-J6/JFH genotype 2a/2a chimeric clone, ing, using a monoclonal antibody against HCV NS3 Huh-7 (gift from Dr Charles M Rice) and Huh-8 har- protein (AbCam, Cambridge, UK), and by detecting bouring an autonomous Con1/SG-Neo genotype 1b HCV RNA in a PCR, using primers against HCV NS3 subgenomic replicon with neomycin phosphotransfer- gene: forward (5′-CATGGCATGCATGTCGG-3′) and sase (NPTII) selectable marker were used. reverse (5′-CGATGTAAGGGAGGTGTGAGG-3′). The transfected cells were then allowed to grow for Cell culture 2–3 weeks until cell death became apparent. The Huh-7 cells were maintained in complete growth supernatant containing HCV virion was then collected medium (Dulbecco’s modified Eagle medium contain- and filtered through a 0.22 mm filter. The supernatant ing 10% heat-inactivated fetal bovine serum, 0.1 mM was then stored at 4°C, and used to infect fresh Huh-7 non-essential amino acids and 1% penicillin/strepto- cells. The success of infection was confirmed by immu- mycin solution). All reagents were purchased from Inv- nostaining and PCR, as described above. itrogen (Carlsbad, CA, USA). Huh-8 cells, carrying the self-replicating subgenomic replicon, were maintained Determination of antiviral activity of fluoroquinolones in the same growth medium, additionally containing Twelve fluoroquinolones (ofloxacin, 8-quinolinol, G418 (0.75 mg/ml; Sigma–Aldrich, St Louis, MO, 8-hydroxyquinoline, cinoxacin, enoxacin, enrofloxacin, USA) [11,13]. fleroxacin, flumiquine, lumifloxacin, norfloxacin, balofloxacin and difloxacin) were obtained from Sigma– FL-J6/JFH in vitro transcription Aldrich [10]. The desired fluoroquinolones were added FL-J6/JFH plasmid was linearized with XbaI (New to Huh-7 or Huh-8 cells at concentrations of 0.01 mM, England BioLabs, Ipswich, MA, USA), purified using 0.1 mM, 1.0 mM or 10 mM. After 4 days (96 h), total gel filtration columns (Promega, Madison, WI, USA) RNA was isolated from both the cell lines and HCV and used for in vitro transcription of HCV genomic RNA levels were quantitated in triplicate in a quantita- RNA with T7 Express RiboMAX (Promega). RNA tive real-time PCR (qPCR), using primers against the transcription reaction mixture (20 ml) contained 2 ml HCV NS3 gene (described above). As an internal con- enzyme mix T7 Express (T7 RNA polymerase, recom- trol, the house-keeping gene b-globin was also quanti- binant RNasin ribonuclease inhibitor and recombinant tated in all samples. HCV RNA levels were corrected inorganic pyrophosphatase), 10 ml RiboMAX Express for the levels of b-globin RNA in all samples. T7 2X buffer, 110 units RQ1 RNase-Free DNase, 1 mg linearized DNA template and Nuclease-Free water up Time-point assay to determine effects of to 20 ml (Promega). Following a 0.5 h synthesis at 37°C, fluoroquinolones on cellular and viral RNA replication the synthesized RNA was DNase-treated with RQ1 Both Huh-7 and Huh-8 cells were split in six-well RNase-Free DNase (Promega). RNA concentration tissue culture plates (Corning Inc., Corning, NY, was measured by spectrophotometry and aliquots were USA), grown overnight and then treated with 1.0 mM stored at -80°C until use. or 10 mM fluoroquinolones. Total RNA was isolated from treated and untreated cells after 24, 48, 72 and HCV RNA transfection of Huh-7 cells 96 h. HCV RNA was quantitated in triplicate by For RNA transfection, cells were washed with amplifying the HCV NS3 gene in a qPCR. b-Globin phosphate-buffered saline (PBS), trypsinized, re- messenger RNA (mRNA) levels were measured as well suspended in complete growth medium and plated to assess the inhibitory effects of fluoroquinolones on overnight, aiming for 80–90% confluency. The fol- cellular RNA production. lowing day, cells were placed in fresh medium 2–3 h prior to transfection. Transfection was achieved using HCV RNA analysis and quantitation

a CaCl2 transfection protocol [15]. FL-J6/JFH RNA Total RNA was isolated from Huh-8 or HCV-infected transcripts (2–3 mg) were mixed with 50 ml 2.5M Huh-7 cells using an AquaPure RNA Isolation kit

CaCl2 and 450 ml sterile distilled water in a 1.5 ml (Bio-Rad Laboratories, Hercules, CA, USA). For

AVT-11-OA-2098_Khan.indd 468 10/04/2012 11:36:25 Inhibition of HCV by fluoroquinolones

complementary DNA (cDNA) synthesis, 2 mg RNA with 14,000× rpm for 15 min, dissolved in 1.5 ml of 20 mM 0.5 mg random primers was added in 15 ml sterile water Tris (pH 8), 50 mM NaCl, 1 mM EDTA, 0.1 mM DTT and heated at 70°C for 5 min. The reaction was placed (gel filtration buffer), and centrifuged at 14,000× rpm on ice and 2 ml 10 mM dNTPs mix, 2 ml 5×M-MLV RT for 15 min (fraction III). Fraction III was loaded onto reaction buffer containing 250 mM Tris-HCl (pH 8.3), a 100 ml Sephacryl S-200 high-resolution gel filtra-

375 mM KCl, 15 mM MgCl2, 50 mM DTT and 0.5 ml tion column, which was equilibrated and eluted with 200 units M-MLV RT enzyme (Promega) and 0.5 ml 25 gel filtration buffer, after measuring absorbance; frac- units RNAse inhibitors (Promega) were added in a 25 ml tions containing the helicase were pooled (fraction IV). reaction and the reactions were incubated at 25°C for Fraction IV was loaded onto a 2 ml DEAE Sepharose 10 min, 37°C for 60 min and at 85°C for 5 min in an column and 1 ml fractions eluted with a NaCl gradi- Eppendorf (Hamburg, Germany) thermal cycler. ent from 10 to 500 mM. Combined fractions were dia- cDNA levels were quantitated in triplicate by amplify- lysed into gel filtration buffer (20 mM Tris [pH 8], 50 ing the HCV NS3 gene in a qPCR. The 25 ml reaction mM NaCl, 1 mM EDTA and 0.1 mM DTT) containing mixture contained 5 ml of cDNA, 0.5 ml (10 pmol/ml) 25% glycerol for storage (fraction V). Purity of NS3 of forward and reverse primers (sequences given above), protein was monitored by examining eluates from each and 12.5 ml of Maxima™ SYBR Green qPCR master purification step, using SDS–PAGE. Protein concentra- mix (Fermentas, Life Sciences, Burlington, ON, Canada). tion was determined by measuring the absorbance at Reactions were run on the Chromo 4 Real-time thermo 280 nm [18,19]. cycler detection system (Bio-Rad Laboratories), using the thermal cycle 5 min at 95°C, 35 cycles of 40 s at 95°C, NS3 helicase assay 40 s at 55°C and 40 s at 72°C. Results were normal- NS3 helicase activity was measured in a molecular ized against the b-globin mRNA levels for each sample. beacon assay, as described in [14]. DNA oligonucleo- b-Globin qPCR was carried out in a similar approach tides modified with Cyanine 5 (Cy5), and Black Hole using 5´-ACACAACTGTGTTCACTAGC-3´ as the for- Quencher® (BHQ) were synthesized as described in [14] ward and 5´-CAACTTCATCCACGTTCACC-3´ as the (Integrated DNA Technologies, Coralville, IA, USA). reverse primer [16]. The following thermal cycle was The oligonucleotides were dissolved in DNAse/RNAse used: 5 min at 95°C, 35 cycles of 40 s at 95°C, 40 s at free water (Fisher Biotech, Fair Lawn, NJ, USA) and the 55°C and 40 s at 72°C. concentrations were determined from the extinction coefficients provided. Helicase substrates were pre- HCV NS3 helicase protein expression and purification pared by combining these single strands at a 1:1 molar HCV NS3 cloned in pET33a vector, His-Hel-His, was ratio to a final concentration of 20 mM in 10 mM Tris a gift from Dr David Frick [17]. The cloned protein HCl pH 8.5. The mixture of two strands was placed at expresses the HCV genotype 1a helicase fragment with 95°C and then allowed to anneal by gradual cooling a His-tag on both the C- and N-termini. Escherichia to room temperature for 1 h. Each helicase reaction coli BL21 cells that have T7 polymerase under the con- contained 25 mM KCl, 2% glycerol, 4 mM DTT, 40

trol of a lac promoter were used to express this NS3 mM Tris pH 7.6, 5 mM MgCl2, 1 mM ATP, 500 nM helicase. The cells containing the NS3 construct were helicase enzyme and 5 nM nucleic acid substrate. The grown to an absorbance of 0.5 at 600 nm, and induced fluoroquinolones was added at concentrations of 0.01, with 1 mM isopropyl-b-d-thioglactopyranoside. After 0.1, 1.0 or 10 mM, and the reactions were carried out 3 h, cells were harvested by centrifugation, washed in in 100 ml, in triplicate, in black ‘Half-volume’ 96-well Tris-buffered saline (pH 7.5) and resuspended in 10 ml polystyrene plates (Thermo Fisher Scientific, Vantaa, of 20 mM Tris (pH 8), 500 mM NaCl, 5 mM imidazole Finland). Fluorescence was measured as arbitrary units and 10 mg/ml lysozyme (lysis buffer). The suspension (au) in each well after 30 min incubation at 37°C. Data was kept on ice for 30 min and then the extract was were collected using a Chameleon fluorescence spec- cleared by centrifugation at 14,000× rpm (fraction I). trophotometer (Hidex, Turku, Finland). Cy5-labelled Fraction I was loaded onto a 5 ml Ni-nitrilotriacetic substrates were measured at 595/675 nm [14]. The acid column (NiNTA; Promega) previously charged results were expressed as the percentage of inhibi-

with NiSO4 and equilibrated in buffer A. The col- tion relative to the fluorescence observed without any umn was washed with buffer A (20 mM Tris [pH 8], added fluoroquinolone. 500 mM NaCl and 40 mM imidazole) and eluted with buffer B (20 mM Tris [pH 8], 500 mM NaCl and 500 Results mM imidazole). Fractions containing the helicase protein were analysed on an SDS gel and combined (frac- Antiviral activity of fluoroquinolones in Huh-8 cells

tion II). (NH4)2SO4 was added to 60% saturation to Twelve fluoroquinolones were screened for their anti- fraction II, and the precipitate thus formed was spun at viral activity, using Huh-8 cell culture. Huh-8 cells

Antiviral Therapy 17.3 469

AVT-11-OA-2098_Khan.indd 469 10/04/2012 11:36:26 IA Khan et al.

Figure 1. Inhibition of HCV Con1/SG-Neo genotype 1b subgenomic replicon at different concentrations of fluoroquinolones

A 0.01 µM 0.1 µM 100 100

80 80

60 60

40 40 HCV replicon, % HCV replicon, % HCV replicon,

20 20

0 0 Fluoroquinolone Fluoroquinolone

1 µM Noroxacin 100 Ooxacin Flumiquine 80 Enrooxacin Cinoxacin 60 Enoxacin Fleroxacin 40 8-Quinolinol

HCV replicon, % HCV replicon, 8-Hydroxyquinoline 20 Lumioxacin Balooxacin 0 Dioxacin Fluoroquinolone

(A) Real-time PCR results of HCV RNA levels in Huh-8 cells treated with various fluoroquinolones at 0.01 mM, 0.1 mM and 1.0 mM for 4 days. RNA copy number per mg of total RNA was expressed as a percentage relative to the level for the untreated controls. Huh-8 cells treated with 1 mM fluoroquinolones showed a significant decrease in HCV RNA levels by day 4. Each bar represents the mean of triplicates ±sem. (B) Time-dependent inhibition of HCV Con1/SG-Neo subgenomic replicon by fluoroquinolones. Lines with shaded triangles and squares represent HCV replicon RNA levels in Huh-8 cells treated with 1 mM and 10 mM fluoroquinolones, respectively. The topmost lines with empty triangles and squares represent b-globin levels in the presence of fluoroquinolones at 1 mM and 10 mM, respectively. The reduction of the HCV RNA level was normalized against the messenger RNA level of human b-globin.

were grown at three different concentrations of each compounds were, respectively, 8.2, 6.3, 1.65, 3.3, 4.4, fluoroquinolone, namely 0.01, 0.1 and 1 mM, for 96 h 4.3 and 3.9 mM. (Figure 1A); following that, total RNA was extracted To study the fluoroquinolone-mediated inhibition of and HCV RNA was quantitated using real-time qPCR, HCV in a time-dependent manner, the above experi- as described in Methods. As shown in Figure 1A, inhi- ment was performed at two concentrations (1 mM and bition of HCV RNA replication was clearly evident at 10 mM) of each fluoroquinolone, and the HCV RNA 0.1 mM in most cases (Figure 1A) and became most was quantitated at 24, 48, 72 and 96 h. Additionally, to pronounced at 1 mM. Enoxacin, fleroxacin, 8-quin- exclude the cytotoxic effect of these fluoroquinolones, olinol, 8-hydroxyquinoline, difloxacin, balofloxacin RNA of a house-keeping host cell gene, b-globin, was and lumifloxacin inhibited the HCV RNA to 30% also measured at each of the four time points (Figure or less at a concentration of 1 mM (Figure 1A). The 1B). At both the concentrations tested, HCV RNA rep-

half-maximal effective concentrations (EC50) for these lication was found to be inhibited in a time-dependent

AVT-11-OA-2098_Khan.indd 470 10/04/2012 11:36:26 Inhibition of HCV by fluoroquinolones

Figure 1. Continued

B 120 O oxacin 120 8-Quinolinol 120 8-Hydroxyquinoline

100 100 100

80 80 80

60 60 60

40 40 40 HCV replicon, % HCV replicon, HCV replicon, % HCV replicon, HCV replicon, % HCV replicon, 20 20 20

0 0 0 24 48 72 96 24 48 72 96 24 48 72 96 Time, h Time, h Time, h

120 Enoxacin 120 Cinoxacin 120 Enro oxacin

100 100 100

80 80 80

60 60 60

40 40 40 HCV replicon, % HCV replicon, HCV replicon, % HCV replicon, HCV replicon, % HCV replicon, 20 20 20

0 0 0 24 48 72 96 24 48 72 96 24 48 72 96 Time, h Time, h Time, h

120 Lumi oxacin 120 Fleroxacin 120 Flumiquine

100 100 100

80 80 80

60 60 60

40 40 40 HCV replicon, % HCV replicon, HCV replicon, % HCV replicon, HCV replicon, % HCV replicon, 20 20 20

0 0 0 24 48 72 96 24 48 72 96 24 48 72 96 Time, h Time, h Time, h

120 Nor oxacin 120 Balo oxacin 120 Di oxacin

100 100 100

80 80 80

60 60 60

40 40 40 HCV replicon, % HCV replicon, HCV replicon, % HCV replicon, HCV replicon, % HCV replicon, 20 20 20

0 0 0 24 48 72 96 24 48 72 96 24 48 72 96 Time, h Time, h Time, h

Antiviral Therapy 17.3 471

AVT-11-OA-2098_Khan.indd 471 10/04/2012 11:36:27 IA Khan et al.

Figure 2. Inhibition of FL-J6/JFH HCV RNA transcript replication at different concentrations of fluoroquinolones

A 0.01 µM 0.1 µM 100 100

80 80

60 60

40 40 HCV RNA, % HCV RNA, %

20 20

0 0 Fluoroquinolone Fluoroquinolone

Noroxacin 100 1 µM Ooxacin Flumiquine 80 Enrooxacin Cinoxacin 60 Enoxacin Fleroxacin 40 8-Quinolinol

HCV RNA, % 8-Hydroxyquinoline 20 Lumioxacin Balooxacin 0 Dioxacin Fluoroquinolone

(A) Real-time PCR results of HCV RNA levels in Huh-7 cells treated with various fluoroquinolones (0.01 µM, 0.1 µM and 1 µM) for 4 days. For treatment and analysis, the same procedures as in the previous experiments were used. RNA copy number per µg of total RNA was expressed as a percentage relative to the level for the untreated controls. Huh-7 cells treated with 1 mM fluoroquinolones showed a significant decrease in HCV RNA levels by day 4. Each bar represents mean of triplicates ±sem. (B) Time-dependent inhibition of the FL-J6/JFH HCV RNA transcript by fluoroquinolones. Lines with shaded triangles and squares represent HCV replicon RNA levels in Huh-7 cells treated with 1 mM and 10 mM fluoroquinolone, respectively. The topmost lines with empty triangles and squares represent b-globin levels in the presence of fluoroquinolones at 1 mM and 10 mM, respectively. The reduction of the HCV RNA level was normalized against the messenger RNA level of human b-globin.

manner by all 12 fluoroquinolones. Moreover, no such 2, inhibition by most fluoroquinolones tested was evi- inhibition was observed for the b-globin RNA, confirm- dent at a concentration of 0.1 mM, and was clearly pro- ing that the inhibitory effect of fluoroquinolones was nounced in all cases at 1 mM (Figure 2A). In this experi- specific to HCV RNA. ment, drugs that inhibited the HCV RNA to 30% or less were fleroxacin, difloxacin, ofloxacin, 8-quinolinol,

Antiviral activity of fluoroquinolones in Huh-7 cells enoxacin, lumifloxacin and flumiquine. The EC50 for The same 12 fluoroquinolones used above were also these compounds were, respectively, 3.8, 2.5, 2.4, 2.2, tested on HCV-producing Huh-7 cells. These cells were 1.0, 1.9 and 1.7 mM. grown in the presence of 0.01, 0.1 or 1 mM fluoroqui- Time-dependent inhibition by the 12 fluoroquinolone, and the effect of these drugs on HCV virion nolones was also observed in the Huh-7 cultures (Fig- production was analysed by quantifying the amount of ure 2B). The results were similar to those observed in HCV RNA as described in Methods. As shown in Figure Huh-8 cells. In these experiments as well, the levels of

AVT-11-OA-2098_Khan.indd 472 10/04/2012 11:36:28 Inhibition of HCV by fluoroquinolones

Figure 2. Continued

B 120 O oxacin 120 8-Quinolinol 120 8-Hydroxyquinoline

100 100 100

80 80 80

60 60 60

40 40 40 HCV replicon, % HCV replicon, HCV replicon, % HCV replicon, HCV replicon, % HCV replicon, 20 20 20

0 0 0 24 48 72 96 24 48 72 96 24 48 72 96 Time, h Time, h Time, h

Enoxacin Cinoxacin Enro oxacin 120 120 120

100 100 100

80 80 80

60 60 60

40 40 40 HCV replicon, % HCV replicon, HCV replicon, % HCV replicon, HCV replicon, % HCV replicon, 20 20 20

0 0 0 24 48 72 96 24 48 72 96 24 48 72 96 Time, h Time, h Time, h

Lumi oxacin 120 120 Fleroxacin 120 Flumiquine

100 100 100

80 80 80

60 60 60

40 40 40 HCV replicon, % HCV replicon, HCV replicon, % HCV replicon, HCV replicon, % HCV replicon, 20 20 20

0 0 0 24 48 72 96 24 48 72 96 24 48 72 96 Time, h Time, h Time, h

Nor oxacin Balo oxacin Di oxacin 120 120 120

100 100 100

80 80 80

60 60 60 HCV replicon, % HCV replicon, 40 40 % HCV replicon, 40 HCV replicon, % HCV replicon, 20 20 20

0 0 0 24 48 72 96 24 48 72 96 24 48 72 96 Time, h Time, h Time, h

Antiviral Therapy 17.3 473

AVT-11-OA-2098_Khan.indd 473 10/04/2012 11:36:28 IA Khan et al.

b-globin RNA remained unchanged at either of the two and 2B). The molecular-beacon-based helicase assay concentrations (1 mM and 10 mM) of the fluoroquinolo- showed that HCV helicase activity was efficiently nes studied, confirming that the inhibition of HCV by inhibited by fluoroquinolones in a concentration- fluoroquinolones was not a reflection of a cytotoxic dependent manner at 0.01, 0.1, 1.0 and 10 mM of the effect on Huh-7 cells. 12 drugs tested (Figure 3). With slight differences, all 12 fluoroquinolones Inhibition of HCV NS3 helicase showed a consistent level of potency between the HCV helicase activity was measured in the presence of Huh-8 and Huh-7 models. Enoxacin, fleroxacin, 0.01, 0.1, 1.0 and 10 mM fluoroquinolone. The results 8-quinolinol, 8-hydroxyquinoline, difloxacin and were expressed as the percentage of inhibition by the lumifloxacin were found to be highly potent (exhibit- fluorquinolone, using the fluoroquinolone-minus con- ing 30% or more inhibition of HCV RNA) in both trol as reference. Fluoroquinolones inhibited the activ- the Huh-8 and Huh-7 culture models. However, balo- ity of NS3 helicase in a linear fashion. At 0.1 mM, drugs floxacin was found to be potent in Huh-8 model but that inhibited the HCV helicase activity to 30% or less was not found to be as effective in the Huh-7 cells, were 8-quinolinol, enoxacin, fleroxacin, flumiquine and whereas flumiquine and ofloxacin were found to be difloxacin. The half-maximal inhibitory concentrations remarkably effective in Huh-7 but not in Huh-8 cells. for these compounds were, respectively, 4.1, 5.9, 4.8, A plausible reason for this observation might be the 9.9 and 5.2 mM. difference in the viral genotypes in the two systems, since Huh-8 and Huh-7 cells express HCV genomes Discussion originating from genotypes 1b and 2a, respectively. HCV-positive patients are known to respond to the A panel of 12 fluoroquinolones was screened to assess same antiviral drugs differently [22]; genotype-specific the antiviral activity of these drugs against HCV and effects of anti-HCV drugs have not been thoroughly HCV NS3 helicase using two different cell culture analysed so far. Although fluoroquinolones share the models and one in vitro assay. The two culture mod- basic quinolone carboxylic acid template with a flu- els used convincingly showed the inhibition of HCV orine atom at C6 [23], we speculate that functional replicon and HCV viral replication by all 12 fluoro- groups specific to different types of fluoroquinolone quinolones. The in vitro helicase assay also showed might render differential specificity for various HCV inhibition of HCV NS3 helicase activity by all the genotypes. Fluoroquinolones have been reported to fluoroquinolones tested. modulate gene expression [24] and cytokine produc- As stated previously, fluoroquinolones constitute an tion [25], as well as to effect degradation of mitochon- important class of broad-spectrum antibacterial drugs, drial DNA [26] in mammalian cells. Although we did targeting the bacterial type II topoisomerase family not observe any cytotoxic effect of fluoroquinolones – gyrase and topoisomerase IV [4]. These drugs spe- on Huh-7 or Huh-8 cells (Figures 1B and 2B), cel- cifically block the topoisomerase catalytic activity by lular effects of these drugs should also be taken into forming a frozen quinolone–gyrase–DNA complex [20], account when interpreting their anti-HCV activity. inhibiting bacterial DNA replication. Previously, vari- The observation that quinolones bind to the heli- ous in vitro studies and clinical trials have demonstrated case component of bacterial gyrases [24] led to the the antiviral activity of fluoroquinolones against vari- studies that tested whether quinolones could also ous DNA and RNA viruses [6,21]. However, sufficient bind to viral proteins with analogous properties. data regarding the antiviral activity of fluoroquinolones Subsequently, studies suggested that ofloxacin has against HCV are lacking. We believe our study is the an inhibitory activity against vaccinia virus topoi- first to screen a variety of fluoroquinolones systemati- somerase, leading to a decreased synthesis of viral cally for their effectiveness to inhibit HCV replication, DNA and RNA [21]. The ability of fluoroquinolo- and show that viral helicase is targeted by these drugs. nes to inhibit HIV replication in in vitro systems by In both the culture models we used, at a concentra- a protein–DNA interaction has also been reported tion of 1 mM, all 12 fluoroquinolones showed at least [27]. Assays performed with SV40 T antigen helicase 60% inhibition of HCV RNA replication (Figures 1A have shown an in vitro inhibitory effect of fluoroqui- and 2A) with eight of the drugs showing a 30% or nolones on the viral helicase activity [16]. Although more inhibitory effect (Figures 1A and 2A). No inhibi- research is in progress to find potential targets for tion of the house-keeping gene b-globin was observed antiviral therapy against HCV, NS3 helicase has not at the same (1 µM) or 10× higher (10 µM) concentra- been a focus of such studies so far. In this study, we tion of the drugs, confirming that the inhibitory effect show evidence that fluoroquinolones might inhibit of the fluoroquinolones was specific to HCV and was HCV RNA replication by targeting the HCV NS3 not a reflection of toxicity to the host cells (Figures 1B helicase. Molecular dynamics studies of the HCV

AVT-11-OA-2098_Khan.indd 474 10/04/2012 11:36:28 Inhibition of HCV by fluoroquinolones

Figure 3. Results from molecular-beacon-based helicase assay

Ooxacin 8-Quinolinol 8-Hydroxyquinoline 100 100 100

80 80 80

60 60 60

40 40 40

Inhibition, % 20 Inhibition, % 20 Inhibition, % 20

0 0 0 0.01 0.1 1 10 0.01 0.1 1 10 0.01 0.1 1 10

100 Cinoxacin 100 Enoxacin 100 Enrooxacin

80 80 80

60 60 60

40 40 40 Inhibition, % Inhibition, % Inhibition, % 20 20 20

0 0 0 0.01 0.1 1 10 0.01 0.1 1 10 0.01 0.1 1 10 Fleroxacin Flumiquine Lomioxacin 100 100 100

80 80 80

60 60 60

40 40 40 Inhibition, % Inhibition, % Inhibition, % 20 20 20

0 0 0 0.01 0.1 1 10 0.01 0.1 1 10 0.01 0.1 1 10

Noroxacin Balooxacin Dioxacin 100 100 100

80 80 80

60 60 60

40 40 40 Inhibition, % 20 Inhibition, % 20 Inhibition, % 20

0 0 0 0.01 0.1 1 10 0.01 0.1 1 10 0.01 0.1 1 10

Bars represent HCV NS3 levels treated with 0.01 mM, 0.1 mM, 1.0 mM and 10 mM fluoroquinolone. Most of the tested fluoroquinolones showed 50–90% reduction of HCV NS3 activity at 1.0 mM and 10 mM concentrations of drug. Each bar represents a mean of triplicates ±sem.

NS3 helicase–fluoroquinolone complex will reveal with IFN-a [28,29]; however, later trials recruiting the mechanistic details of fluoroquinolone-mediated a larger number of patients failed to reproduce these inhibition of HCV RNA replication. results convincingly [30,31]. Fluoroquinolones that Previous studies assessing the efficacy of ofloxacin and have tested as highly potent in our experiment have not ciprofloxacin in patients with chronic hepatitis C have been tested on patients yet. These drugs could be tested produced conflicting results. Two pilot studies initially in clinical trials to find out whether the same effect suggested a potential benefit of combining ofloxacin recapitulates in HCV-infected patients. Additionally,

Antiviral Therapy 17.3 475

AVT-11-OA-2098_Khan.indd 475 10/04/2012 11:36:29 IA Khan et al.

16. Oniangue-Ndza C, Aus dem Siepen M, Lohmann V, et al. insight into the specific interactions between the drugs In vitro replicative properties of replicons constructed using and NS3 helicase will help in designing novel, more sequence variants of the hepatitis C virus strain AD78 that caused a single-source outbreak of hepatitis C. Virus Res potent, drugs for use against HCV. 2009; 142:1–9. 17. Frick DN. The hepatitis C virus NS3 protein: a model RNA helicase and potential drug target. Curr Issues Mol Biol Disclosure statement 2007; 9:1–20. 18. Lam AM, Keeney D, Eckert PQ, Frick DN. Hepatitis C The authors declare no competing interests. virus NS3 ATPases/helicases from different genotypes exhibit variations in enzymatic properties. J Virol 2003; 77:3950–3961. References 19. Frick DN, Rypma RS, Lam AM, Gu B. The nonstructural 1. Lavanchy D. The global burden of hepatitis C. Liver Int protein 3 protease/helicase requires an intact protease 2009; 29 Suppl 1:74–81. domain to unwind duplex RNA efficiently. J Biol Chem 2004; 279:1269–1280. 2. Hadziyannis SJ, Sette H, Jr., Morgan TR, et al. Peginterferon-alpha2a and ribavirin combination therapy 20. Witvrouw M, Daelemans D, Pannecouque C, et al. Broad- in chronic hepatitis C: a randomized study of treatment spectrum antiviral activity and mechanism of antiviral duration and ribavirin dose. Ann Intern Med 2004; action of the fluoroquinolone derivative K-12. Antivir Chem 140:346–355. Chemother 1998; 9:403–411. 3. Fried MW, Shiffman ML, Reddy KR, et al. Peginterferon 21. Ikeda S, Yazawa M, Nishimura C. Antiviral activity and alfa-2a plus ribavirin for chronic hepatitis C virus infection. inhibition of topoisomerase by ofloxacin, a new quinolone N Engl J Med 2002; 347:975–982. derivative. Antiviral Res 1987; 8:103–113. 4. Higgins PG, Fluit AC, Schmitz FJ. Fluoroquinolones: structure 22. Tsutsumi M, Takada A, Takase S, Sawada M. Effects of and target sites. Curr Drug Targets 2003; 4:181–190. combination therapy with interferon and ofloxacin on chronic type C hepatitis: a pilot study. J Gastroenterol 5. Froelich-Ammon SJ, Osheroff N. Topoisomerase poisons: Hepatol 1996; 11:1006–1011. harnessing the dark side of enzyme mechanism. J Biol Chem 1995; 270:21429–21432. 23. Andersson MI, MacGowan AP. Development of the quinolones. J Antimicrob Chemother 2003; 51 Suppl 1:1–11. 6. Richter S, Parolin C, Palumbo M, Palu G. Antiviral properties of quinolone-based drugs. Curr Drug Targets 24. Liguori MJ, Blomme EA, Waring JF. Trovafloxacin-induced Infect Disord 2004; 4:111–116. gene expression changes in liver-derived in vitro systems: comparison of primary human hepatocytes to HepG2 cells. 7. Portolani M, Pietrosemoli P, Cermelli C, et al. Suppression Drug Metab Dispos 2008; 36:223–233. of BK virus replication and cytopathic effect by inhibitors of prokaryotic DNA gyrase. Antiviral Res 1988; 9:205–218. 25. Ogino H, Fujii M, Ono M, et al. In vivo and in vitro effects of fluoroquinolones on lipopolysaccharide-induced pro- 8. Koukoulaki M, Apostolou T, Hadjiconstantinou V, inflammatory cytokine production. J Infect Chemother Drakopoulos S. Impact of prophylactic administration of 2009; 15:168–173. ciprofloxacin on BK polyoma virus replication. Transpl Infect Dis 2008; 10:449–451. 26. Lawrence JW, Claire DC, Weissig V, Rowe TC. Delayed cytotoxicity and cleavage of mitochondrial DNA in 9. Josephson MA, Williams JW, Chandraker A, Randhawa PS. ciprofloxacin-treated mammalian cells. Mol Pharmacol Polyomavirus-associated nephropathy: update on antiviral 1996; 50:1178–1188. strategies. Transpl Infect Dis 2006; 8:95–101. 27. Manns MP, McHutchison JG, Gordon SC, et al. 10. Ali SH, Chandraker A, DeCaprio JA. Inhibition of Peginterferon alfa-2b plus ribavirin compared with Simian virus 40 large T antigen helicase activity by interferon alfa-2b plus ribavirin for initial treatment of fluoroquinolones. Antivir Ther 2007; 12:1–6. chronic hepatitis C: a randomised trial. Lancet 2001; 11. Lindenbach BD, Evans MJ, Syder AJ, et al. Complete 358:958–965. replication of hepatitis C virus in cell culture. Science 2005; 28. Komatsu M, Ishii T, Ono T, et al. Pilot study of ofloxacin 309:623–626. and interferon-alpha combination therapy for chronic 12. Lindenbach BD, Meuleman P, Ploss A, et al. Cell culture- hepatitis C without sustained response to initial interferon grown hepatitis C virus is infectious in vivo and can administration. Can J Gastroenterol 1997; 11:507–511. be recultured in vitro. Proc Natl Acad Sci U S A 2006; 29. Negro F, Male PJ, Perrin L, Giostra E, Hadengue A. 103:3805–3809. Treatment of chronic hepatitis C with alpha-interferon plus 13. Blight KJ, Kolykhalov AA, Rice CM. Efficient initiation ofloxacin in patients not responding to alpha-interferon of HCV RNA replication in cell culture. Science 2000; alone. J Hepatol 1998; 29:369–374. 290:1972–1974. 30. Kojima H, Kaita KD, Hawkins K, Uhanova J, Minuk GY. 14. Belon CA, Frick DN. Monitoring helicase activity with Use of fluoroquinolones in patients with chronic hepatitis C molecular beacons. Biotechniques 2008; 45:433–442. virus-induced liver failure. Antimicrob Agents Chemother 2002; 46:3280–3282. 15. Kingston RE. Introduction of DNA into Mammalian Cells. In Asubel FM, Brent R, Kingston RE, et al. (Editors). 31. Belon CA, Frick DN. Helicase inhibitors as specifically Current Protocols in Molecular Biology. John Wiley & targeted antiviral therapy for hepatitis C. Future Virol 2009; Sons 2003. 4:277–293.

Accepted 12 July 2011; published online 25 October 2011

AVT-11-OA-2098_Khan.indd 476 10/04/2012 11:36:29