A cell-based reporter assay for inhibitor screening of hepatitis C virus RNA-dependent RNA polymerase

A cell-based reporter assay for inhibitor screening of hepatitis C virus RNA-dependent RNA polymerase
of 11
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  A cell-based reporter assay for inhibitor screening of hepatitis C virusRNA-dependent RNA polymerase  Jin-Ching Lee a,b, * , Chin-kai Tseng a , Kuan-Jen Chen a , Kuo-Jung Huang a , Chun-Kuang Lin a , Ying-Ting Lin a, ** a Department of Biotechnology, College of Life Science, Kaohsiung Medical University, Kaohsiung, Taiwan, ROC  b Graduate Institute of Natural Products, College of Pharmacy, Kaohsiung Medical University, Kaohsiung, Taiwan, ROC  a r t i c l e i n f o  Article history: Received 8 January 2010Received in revised form 3 April 2010Accepted 5 April 2010Available online 9 April 2010 Keywords: Hepatitis C virusRNA-dependent RNA polymeraseLuciferaseHigh-throughput screening a b s t r a c t ThehepatitisCvirus(HCV)NS5B,aRNA-dependentRNApolymerase(RdRp),isanattractivetargetforanti-HCVagents.Themajordisadvantagesofthecommonlyusedpolymeraseinhibitorscreeninginvolvingtheassessmentof  invitro RNAsynthesisarethatitisincapableofdemonstratingthecellularpermeabilityandthe cytotoxicity of compounds. To overcome these limitations, we created the BHK-NS5B-FRLuc reportercell linethat carries stably transfected NS5B andabicistronic reporter gene, (+)FLuc-(  )UTR-RLuc, whichcan be used to simultaneously measure cellular toxicity and intracellular RdRp activity. The (+)FLuc-(  )UTR-RLuc construct comprises the firefly luciferase (FLuc) gene and the  Renilla  luciferase (RLuc) geneinreverseorientationflankedbybothnegativestrandsoftheHCV5 0 -and3 0 -untranslatedregions(UTRs),inwhichFLucandRLucreporterproteinsareregulatedbyhostpolymeraseandfunctionalNS5Bpolymer-ase, respectively. The reporter system was validated with specific agents against NS5B polymerization.Additionally,thisassaywasplacedin96-wellplatesandhada  Z  0 -factorvalueofapproximately0.75,whichisamenableforfacilitatinghigh-throughputscreeningoperations.Notably,incombinationwiththestruc-tured-basedvirtualscreening,animidazolederivativecompoundwasevaluatedasacandidateHCVRdRpinhibitor.Crown Copyright    2010 Published by Elsevier Inc. All rights reserved. Chronic infection of hepatitis C virus (HCV) 1 is one of the maincauses of liver-related morbidity and mortality because it resultsin chronic hepatitis, liver cirrhosis, and hepatocellular carcinomas(HCC) [1], which in turn have caused a serious worldwide healthproblem, with more than 170 million carriers of the virus [2]. Novaccine is currently available for HCV infection and there are no spe-cific drugs for clinical therapy. At present, only the standard of careinvolving the administration of pegylated interferon- a  (IFN- a ) incombination with the nucleotide analogue ribavirin is available forchronic HCV infection [3]. Unfortunately, a sustained virological re-sponse (SVR) is not obtained in approximately 50% of treated pa-tients from among those infected with the HCV genotype 1 strain,which is the most prevalent HCV genotype globally [4]. In additionto their variable effectiveness, peg-IFN-RBV regimens also causeunfavorable side effects, which often lead to the discontinuation of therapy [5]. Therefore, the development of a new, safer, and evenmore effective agent that is active against HCV infection is extremelydesirable. HCV is an enveloped virus belonging to the  Hepacivirus  genuswithin the Flaviviridae family [6]. It has a 9.6-kb positive-sensesingle-strandRNAgenomeconsistingofasinglelongopen-readingframe (ORF) flanked by highly structured 5 0 - and 3 0 -untranslatedregions (UTR), which are crucial for RNA synthesis [7]. The 5 0 -UTR contains an internal ribosome entry site (IRES) that allowscap-independent translation of the sole viral polyprotein, whichcomprises approximately 3000 amino acids and finally processedby both host and virus proteases into 10 individual proteins,including 4 structural proteins (C, E1, E2, and p7) and 6 nonstruc-tural proteins (NS2, NS3, NS4A, NS4B, NS5A, and NS5B) [8]. Amongthe latter, NS5B, a RNA-dependent RNA polymerase (RdRp), andNS3 protease are essential for the viral replication and are there-fore considered as attractive targets for chemotherapeutic inhibi-tion in HCV infection [9], especially in the development of robustand reliable biological assay(s), which are urgently required.To date, RdRp activity assays have been based on the scintilla-tion counting of radioisotope-labeling RNA products synthesized in vitro , usually by using a heteropolymeric RNA template, primer, 0003-2697/$ - see front matter Crown Copyright    2010 Published by Elsevier Inc. All rights reserved.doi:10.1016/j.ab.2010.04.004 *  Correspondence to: Jin-Ching Lee, Department of Biotechnology, KaohsiungMedical University, 100, Shih-Chuan 1st Road, San Ming District, 807 KaohsiungCity, Taiwan, ROC. Fax: +886 7 312 5339. **  Correspondence to: Ying-Ting Lin, Department of Biotechnology, KaohsiungMedical University, 100, Shih-Chuan 1st Road, San Ming District, 807 KaohsiungCity, Taiwan, ROC. Fax: +886 7 312 5339. E-mail addresses: (J.-C. Lee), (Y.-T. Lin). 1  Abbreviations used : ActD, actinomycin D; 2 0 - C  -MeCyt, 2 0 -C-methylcytidine; CsA,cyclosporine A; DMEM, Dulbecco’s modified Eagle’s medium; DMSO, dimethylsulfoxide; HCV, hepatitis C virus; HTS, high-throughput screening; IFN- a , inter-feron- a ; IRES, internal ribosome entry site; MTS, 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2 H  -tetrazolium; RdRp, RNA-dependentRNA polymerase; RT-qPCR, quantitative real-time RT-PCR; UTR, untranslated regions. Analytical Biochemistry 403 (2010) 52–62 Contents lists available at ScienceDirect Analytical Biochemistry journal homepage:  isotopic UTP, and purified RdRp protein [10]. Due to the labor-intensive filtration and washing required as well as the generationof a large amount of radioactive waste, these assays are not suit-able for high-throughput screening (HTS). Furthermore, the cellu-lar permeability and cytotoxicity of tested compounds cannot bedetermined in such a cell-free assay system. To improve this diffi-culty, we developed a cell-based reporter system to measure theactivity of intracellular HCV NS5B polymerase activity withoutusing any tedious processes, like protein purification. Particularly,thecellularpermeabilityandtoxicityoftestedcompoundscaneas-ily be examined simultaneously. The system comprises a biscis-tronic construct containing the firefly luciferase (FLuc) gene andthe Renilla  luciferase(RLuc)geneinthereverseorientation, where-in the RLuc sequence is flanked by the negative strands of the HCV5 0 -and3 0 -UTRs.The5 0 -UTRcontainsanIRESforproteintranslationof the plus strand (+) of RLuc RNA, following conversion by NS5Bpolymerase. In this paper, we describe the characterization of NS5B polymerase activity, determined by measuring luminescentsignals. Stable cell lines were also created for use in susceptibilityassays. Two types of well-characterized NS5B inhibitors includingnucleotide, 2 0 -C-methylcytidine (2 0 - C  -MeCyt), and nonnucleoside,cyclosporine A (CsA) and wedelolactone, were used to specificallyexamine RdRp activity. EC 50  values were comparable to those re-ported previously by different assay systems. Furthermore, in con-trast to several assays designed with monocistronic reporterconstructs, the significant advantage of this dual-reporter analysisis to minimize variation in each sample by normalization of lumi-nescent signals from the first cistron, which greatly improves thequality of activitymeasurementsand also allows for the easy eval-uationof the inhibitoryeffect of compounds. Additionally, a statis-tical analysis was performed on data obtained from experimentsinvolving treatment with the above-noted NS5B inhibitors in a96-well format, with the results indicating that the sensitivityand accuracy of this assay systemare suitable for high-throughputoperations in anti-HCVdrug discovery. Ina structure-based virtualscreening study from a compound library, 43 compounds were se-lected for their anti-NS5B activity in an NS5B cell-based reporterassay, with an anti-HCV replication analysis performed to verifythe hit compounds. Materials and methods Inhibitors 2 0 -C-Methylcytidine was purchased from Toronto ResearchChemicals Inc. and stored as 10mM in 100% dimethyl sulfoxide(DMSO). Cyclosporine A, Wedelolactone (7-methoxy-5,11,12-tri-hydroxy-coumestan), Paclitaxel (Taxol), and actinomycin D werepurchased from Sigma Chemicals Inc. and stored as 10mg/ml,10, 1, and 20mM stock in 100% DMSO, respectively. The final con-centration of DMSO in all reactions was maintained constantly at0.1% in the experiments. The interferon alfa-2a (Roferon-A) waspurchased from Roche Ltd. Cell culture Ava5 cells are the human hepatoma cells (Huh-7) containingHCV subgenomic replicon (kindly provided by Dr. Charles Rice,Rockefeller University, New York, USA) and were maintained inDulbecco’s modified Eagle’s medium (DMEM) with 10% heat-inac-tivated fetal bovine serum, 5% antibiotic–antimycotic, 5% nones-sential amino acids, 1mg/ml G418 and incubated at 37  C with a5% CO 2  supplement. BHK-21 (Baby Hamster Kidney) cells weremaintained in Dulbecco’s modified Eagle’s medium with 10%heat-inactivated fetal bovine serum, 5% antibiotic-antimycotic,5% nonessential amino acid and incubated at 37  C with a 5% CO 2 supplement .  In this study, stably transfected BHK-NS5B-FRLuccells, expressing NS5B and bicistronic reporter RNAs, were main-tained in basal medium with 1mg/ml G418 and 10 l g/mlblasticidin. Construction of HCV NS5B polymerase and p(+)FLuc-(  )UTR-RLuc reporter plasmid The backbone of the HCV NS5B express plasmid is pLenti6/V5-D-TOPO (Invitrogen). The NS5B gene was amplified by PCR frompCon1-SGneo[11]andclonedintothe Bam HIand  Xho IsiteofpLen-ti/V5-D-TOPOvector under the control the cytomegalovirus (CMV)immediate-early gene promoter. The final construct was designedas pLent-NS5B. The reporter plasmid for NS5B activity assay,p(+)FLuc-(  )UTR-RLuc, is represented in Fig. 1A(a). To make thep(+)FLuc-(  )UTR-RLuc construct, the Firefly luciferase (FLuc) and Rellina  luciferase (RLuc) were amplified by PCR from pGL3-basicand pRL-TK (Promega Corporation, Madison, WI), respectively.The FLuc and RLuc genes were cloned into the  Hin dIII and  Eco RI– Eco RV sites of pcDNA3.1 (+) plasmid, in which the FLuc sequencewas sense orientation, termed (+)Fluc, and the RLuc sequencewas antisense orientation, termed (  )RLuc. Subsequently, thePCR-amplified HCV 3 0 -UTR-Ribo element containing the antisense3 0 -UTR of HCV and the hepatitis delta virus (HDV) ribozyme se-quences [12] were digested with  Eco RI and inserted into the  Eco RIsite between the (+)FLuc and the (  )RLuc gene. Finally, the PCR-amplified 5 0 -UTR-Ribo element containing the antisense 5 0 -UTR of HCV and the HDV ribozyme sequences were inserted into theabove plasmid with  Eco RV and  Xho I sites downstream of the RLucgene.The5 0 -UTRis341nucleotides(nt)inlengthand3 0 -UTRis232nt in length based on Con1, a HCV genotype 1b [11]. Establishment of the stable cell line and reporter assay The BHK-21 cells were transfected with pLenti-NS5B in 24-wellcultureplates accordingtothemanufacturer’sinstructions. Briefly,0.5 l gofDNAwasmixedwith2.5 l lofExpress-InTransfectionRe-agent (ThermoFisher ScientificInc.) inOPTI-MEM(Invitrogen) andtransferred onto BHK-21 cells. At 48h posttransfection, cells weresplit into a new 24-well culture plate and maintained in the com-plete DMEM medium containing 10 l g/ml of blasticidin. Afterapproximately 4weeks selection, expended clones, designed BHK-NS5B, were examined for the expression of NS5B by using anti-NS5Bantibody(Abcamplc.,UK). Subsequently,theBHK-NS5Bcellswere transfected with p(+)FLuc-(  )UTR-RLuc in 24-well cultureplates following the transfection method described above. Briefly,0.5 l g of p(+)FLuc-(  )UTR-RLuc was mixed with 2.5 l l of Ex-press-InTransfectionReagentinOPTI-MEM(Invitrogen)andtrans-ferredontoBHK-NS5Bcells. Finally, several stable clones, designedBHK-NS5B-FRLuc, were selected by incubation with completeDMEMcontaining1mg/mlofG418forapproximately4weeks.Re-portergeneexpressioninthesecellswasmeasuredwithaDual-GloLuciferaseAssaySystem(PromegaCorporation)followingtheman-ufacturer’s instructions. The relative activity of NS5B polymerasewas determined by normalizing the level of RLuc against that of FLuc. The 50% effective concentration (EC 50 ) indicated the reducedlevels of RLuc by 50% as compared to 0.1% DMSO-treated controland was measured using linear interpolation.  Analysis of the RLuc RNA generated by HCV NS5B polymerase Various cells including BHK-21, BHK-NS5B, and Ava5 cells weretransfectedwiththep(+)FLuc-(  )UTR-RLucreporterplasmid. After3days of incubation, total cellular RNA was extracted using Trizolreagent (Invitrogen, Carlsbad, CA, USA). The levels of positive- andnegative-stranded bicistronic (+)FLuc-(  )UTR-RLuc RNAs were Intracellular HCV RdRp activity assay/J.-C. Lee et al./Anal. Biochem. 403 (2010) 52–62  53  determined by a reverse-transcription (RT) reaction with specificprimers: primer 1, 5 0 -GCC AGC CCC CGA TTG GGG GC-3 0 corre-sponding to the antisense of HCV 5 0 -UTR; primer 2, 5 0 -ACT TGATCT GCA GAG AGG CC-3 0 corresponding to the sense of HCV 3 0 -UTR. Finally, the RT products were amplified with polymerasechain reaction (PCR) using specific primers corresponding to theRLuc gene. The forward primer sequence was 5 0 -ATG GGC AAATCA GGC AAA TC-3 0 and the reverse primer sequence was 5 0 -CGCAAT ATC TTC TTC AAT ATC AGG-3 0 . Quantification of bicistronic and HCV RNAs BHK-NS5B-FRLuc and Ava5 cells were seeded by 4  10 3 and5  10 4 cells per well, respectively, in 24-well cell culture plates Fig. 1.  (A)Cell-basedHCVRdRpactivityassay.(a)Schematicdiagramofthebicistronicreporterconstruct,p(+)FLuc-(  )UTR-RLuc.Theconstructcontainsthefireflyluciferase(FLuc)geneinthesense(+)orientationandthe Rellina  luciferase(RLuc)geneintheantisense(  )orientation,flankedbythe5 0 -untranslatedregion(5 0 -UTR)and3 0 -UTRoftheHCV-negative strand. Hepatitis delta virus (HDV) ribozyme elements [12] were placed at the end of the 5 0 - and 3 0 -UTRs for exposure of the native structure of HCV UTR following self-cleavage. Thebicistronic gene isunder the control of the cytomegalovirus (CMV) immediate-earlygene promoter. (b) Principle of the RdRpactivity assay. Full-length of (+)FLuc-(  )UTR-RLuc reporter RNA is transcribed by host DNA-dependent RNA polymerase. FLuc expression mediated in a host cap-dependent fashion serves as acontrol for the transcription/translation levels. After self-cleavage of the ribozyme, functional RdRp can recognize the exposed 5 0 - and 3 0 -UTR sequences and carry outreplication. Finally, the (+) stranded RLuc RNA can be translated by the internal ribosome entry site (IRES) within the 5 0 -UTR. The expression level of RLuc reflects thepolymerization activity of HCV RdRp. (B) Characterization of transcription product obtainedby catalysis with HCV RdRp inside cells. (a) RT-PCR analysis of transcripts. In theRTreaction, primer 1andprimer2specificallyanneal tothe 3 0 endofthe oppositelyoriented5 0 -UTR-RLuc-3 0 -UTRRNAforcDNAsynthesis. TheprimersforPCRamplificationcorrespond to the sequence of the RLuc gene. (b) RNA was analyzed in BHK-21 parental cells, the BHK-NS5B cell line stably transfected with NS5B, and the HCV subgenomicreplicon-harboring Ava5 cell line following transfection of p(+)FLuc-(  )UTR-RLuc. The RT-PCR products were separated by agarose electrophoresis. Arrows indicate theexpected 252-bp PCR product from the RLuc gene. (c) Transient reporter assay of NS5B polymerase. BHK-21, BHK-NS5B, and Ava5 cells were seeded at a density of 4  10 3 and 5  10 4 cells per well in 24-well plates and transfected with 0.5 l g of p(+)FLuc-(  )UTR-RLuc plasmid for 3days. Then, cells lysates were subjected to luminescencedetection with the Dual-Glo Luciferase Assay Kit (Promega). Each experiment was performed in triplicate. The error bars reflecting standard deviations are shown.54  Intracellular HCV RdRp activity assay/J.-C. Lee et al./Anal. Biochem. 403 (2010) 52–62  in Dulbecco’s modified Eagle’s medium with 10% heat-inactivatedfetal bovine serum, 5% antibiotic-antimycotic, 5% nonessentialamino acid and incubated at 37  C in the presence or absence of inhibitor for 4days. Total cellular RNA was extracted using Trizolreagent (Invitrogen). The expression of RLuc RNA levels was de-tected by quantitative real-time RT-PCR (RT-qPCR) with primerscorresponding to the RLuc gene: forward primer, 5 0 -ATG GGCAAA TCA GGC AAA T-3 0 ; and reverse primer, 5 0 -CGC AAT ATC TTCTTC AAT ATC AGG-3 0 . The copy number of RLuc in each samplewas normalized to cellular GAPDH (forward primer, 5 0 -GTC TTCACC ACC ATG GAG AA-3 0 ; reverse primer, 5 0 -ATG GCA TGG ACTGTG GTC AT-3 0 ) levels from three independent experiments withthe ABI Step One Real-Time PCR-System (ABI Warrington, UK).ThelevelsofHCVsubgenomicRNAweredetectedbyRT-qPCRwithprimers corresponding to the NS5B gene: forward primer, 5 0 -GGAAAC CAA GCT GCC CAT CA-3 0 ; and reverse primer, 5 0 -CCT CCACGGATAGAAGTTTA-3 0 . Thecopynumberof NS5Bineachsamplewas normalized to cellular GAPDH. Briefly, real-time PCR was car-ried out in a 10- l l reaction volume containing 200ng cDNA, 5 l lPower SYBR Green PCR Master Mix, and 0.4 l M primer pair. ThePCR was conducted by the following conditions: denaturation at95  C for 10min, 40 cycles of amplification at 95  C for 15s,60  C for 1min, and finally, 95  C for 15s, 60  C for 1min, 95  Cfor 15s. Quantification of bicistronic p(+)FLuc-(  )UTR-RLuc plasmid copynumber (PCN) in BHK-NS5B-FRLuc cells To estimate the amount of (+)FLuc-(  )UTR-RLuc gene in stablytransfectedcells, the corresponding plasmidcopynumber was cal-culated using the following equation [13,14]: PCN=(6.02  10 23 copy/mol)  DNA amount (g)/[DNA length (bp)  660 (g/mol/bp)].First, p(+)FLuc-(  )UTR-RLuc plasmid was serially diluted 10-fold,ranging from 1  10 2 to 1  10 6 copies/ l l for the construction of standardcurves.Subsequently,quantitativereal-timePCRwasper-formed with primers corresponding to RLuc gene: forward primer,5 0 -ATGGGC AAATCA GGC AAA-3 0 ; and reverse primer, 5 0 -CGC AATATC TTC TTC AAT ATC AGG-3 0 . The standard curve was generatedbyalinearregressionofthresholdcycle( C  t )valuesagainsttheloga-rithmofinitialtemplatecopynumbers.Real-timePCRefficiency( E  )wascalculated,accordingto E  (%)=(10  1/slop  1)  100%.Thegeno-mic DNA (gDNA) was extracted by the AccuPrep genomic DNAextraction kit (Bioneer, Alameda, CA) according to the manufac-turer’s instructions in a final elution volume of 200 l l in ddH 2 O.The levels of RLuc DNAwere detectedby qPCRwithprimers corre-sponding to RLuc gene. The copy number of RLuc in each samplewasnormalizedtothelevelofthehousekeepinggene,  gapdh .Briefly,real-timePCRwascarriedoutina10- l lreactionvolumecontaining100ng gDNA, 5 l l Power SYBR Green PCR Master Mix, and 0.4 l Mprimer pair. PCR parameters were described above. The identity of eachamplifiedproductwasconfirmedbyDNAsequencinganalysis. Western blotting assay Ava5 and BHK-NS5B-FRLuc cells were seeded in 24-well platesat a density of 5  10 4 cells per well and 4  10 3 cells per well,respectively, and treatedwiththe compoundat various concentra-tions. After 4days of incubation, cells were lysed in RIPA lysis buf-fer and the cell lysates were collected fromthe plates and clarifiedby centrifugation at 13K RPM for 60min at 4  C. Five microgramsoftotalproteinforeachsamplewasresolvedby8%SDS–polyacryl-amide gel and subsequently transferredto a polyvinylidene difluo-ride membrane (PALL, Pensacola, FL, USA). The membrane wasblocked with 5% skim milk in PBS-T for 4h at room temperatureand stained with rabbit polyclonal anti-NS5B antibody (Virostat,ME, USA) and rabbit polyclonal anti-GAPDH antibody (GeneTex,CA, USA), respectively. The signal was detected and counted bythe software Quantity One (Bio-Rad, CA, USA). Cytotoxicity assay Cell viability was determined by the colorimetric 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulf-ophenyl)-2 H  -tetrazolium (MTS) assay (Promega), as describedpreviously [15]. Ava5 and BHK-NS5B-FRLuc cells were seeded in96-well plates at a density of 5  10 3 and 8  10 2 cells per well,respectively, and treatedwiththe compoundat various concentra-tions. TheCellTiter96AQ  ueous  OneSolutionCellProliferationAssay(Promega) was used to determine the cell viability after 4days of incubation. The absorbance was detected at 490nm.  Analysis of the correlation between FLuc activity and cytotoxicity BHK-NS5B-FRLuccellsweretreatedwithseriallydilutedactino-mycin D (0, 0.03, 0.06, 0.12, 0.24, 0.48, and 0.96 l M) or Taxol (0,3.125, 6.25, 12.5, 25, 50, and 100nM) for 3days. Both FLuc activityand cytotoxicity assays were described above. Data analysis wascarried out with the SigmaPlot software (SPSS Inc., Chicago, IL,USA). The experiments were performed in triplicates. Calculation of S/B, S/N, and Z  0  factor  The performance of this developed assay for HTS, signal-to-background (S/B), signal-to-noise (S/N), and  Z  0 factor values wasevaluated using the method of Zhang et al. [16]. BHK-NS5B-FRLuccells were seeded at a density of 8  10 2 cells per well in a single96-wellplateovernight.CsAwasaddedtothe40wellsatavolumeof 100 l l and a final concentration of 2 l g/ml in 0.1% DMSO as abackground set. A 0.1% DMSO solution was added to the other 40wells as a signal set. Four days later, cell lysates were harvestedfromeachwell for detectionof the activities of FLucand RLucwiththe Dual-Glo Luciferase Assay System (Promega Corporation).Finally, the data were collected and S/B, and S/N, and  Z  0 factorswere determined. The  Z  0 factor was calculated according to theequation [16]:  Z  0 =1  [3SD inhibitor +3SD positivecontrol )/|mean inhibitor  mean positivecontrol |],whereSDisthestandarddeviationinthesignal,the subscript ‘‘inhibitor” and ‘‘positive control” correspond to themean signal obtained in the presence of CsA and 0.1% DMSO,respectively. Statistical analysis of robust screening  The luciferasesignal data were presentedas means±SD. Statis-tical comparison of luciferase activities among the various cellclones was carried out with Student’s  t   test.  P   values of less than0.01 were indicated statistically significant. Virtual screening of compounds against HCV RdRp ForenhancingthehitrateofthevirtualscreeningforHCVNS5B,we examined the 26 three-dimensional complex crystals of HCVRdRp obtained from the protein data bank. According to the posi-tion of inhibitor binding sites, the complex crystals were dividedto central or palm part [17] (PDB code: 1YVF, 1Z4U, 2AWZ, 2AX0,2AX1,2FVC, 2GC8,2GIQ, 2IJN, 2QE2,2QE5)andsurroundingorfin-ger and thumb part [17,18] (PDB code: 1NHU,1NHV, 1OS5, 2BRK,2BRL, 2DXS, 2GIR, 2HAI, 2HWH, 2HWI, 2I1R, 2O5D, 3CIZ, 3CJQ,3CJ2,3CJ3,3CJ4,3CJ5).Ineachcrystalcomplex,thecocrystalinhib-itor was removed to proceed with the molecular mechanics opti-mization, MMFF94 [19,20], and then the inhibitor was dockedbackto thesrcinal bindingsites. TheLigandFit in AccelrysDiscov-ery Studio 2.0 package was used as virtual screening and docking Intracellular HCV RdRp activity assay/J.-C. Lee et al./Anal. Biochem. 403 (2010) 52–62  55  tools [21,22]. The spatial RMSD was calculated for the srcinalcrystal pose and the after-docking pose of each cocrystal inhibitor.Striving for the minimal RMSD by extensively adjusting the dock-ing parameters, the best procedure with optimal docking parame-ters was obtained. Then, 0.203Å, the lowest RMSD for the crystalinhibitor of 3CJ2, indicates both the correctness of the quality of the3CJ2crystallizationandour HCVRDRPvirtual screeningproce-dure.Theamountof57,177moleculesfromtheMaybridgeScreen-ing collection [23] was used as the molecular screening library. Toeliminate the false positive, we useda total of 11 scoring functionsin LigandFit as the consensus scoring. After our virtual screening,89 molecules were selected due to their greater than 7 scores inthe total 11 scoring function ranking within the top 25%. Forty-three compounds were purchased for further biological tests. Synergy analysis Huh7 harboring HCV replicon cells, Ava5, were treated withseriallydilutedMSC015(0, 1, 2.5, 5, 10, and 20 l g/ml) in combina-tion with serially diluted IFN (0, 7.5, 15, 30, and 60U/ml) in acheckerboard manner. The inhibition data of multiple drug dose–effect were analyzed using the Median Effect methods [24] withCalcuSyn computer program (Biosoft, Ferguson, MO, USA). Results Principle of the HCV RdRp activity assay and characterization of thebicistronic reporter vector  Theconfigurationof thebicistronicRdRpreportervector, desig-nated as p(+)FLuc-(  )UTR-RLuc (a), and the basic features of theHCV RdRp activity assay (b) are illustrated in Fig. 1A. The bicis-tronic reporter contains the sense strand (+)FLucand the antisensestrand(  )RLuc flankedby5 0 - and3 0 -UTRs of the negativestrandof the HCV genome, which have a specific secondary structure re-quired for RdRp binding [25]. The transcription of the full-lengthbicistronic RNA is under the control of a cytomegalovirus immedi-ate-earlypromoterrecognizedbythecellular Pol IIpolymerase;theinternal region and the 3 0 terminus transcript are processed by theself-cleavage of the hepatitis delta virus (HDV) ribozyme flankingtheendof antisense5 0 -UTR-RLuc-3 0 -UTRRNA, resultinginthesep-aration of (+)FLuc and (  )5 0 -UTR-RLuc-3 0 -UTR transcripts, and theexposure of the primordial UTR structure, which is important forrecognition by HCV RdRp [25,26]. Then, synthesis of (+)5 0 -UTR-RLuc-3 0 -UTR RNA was initiated at the 3 0 -terminal region of the(  )RNA because this region of antisense stranded viral RNA is effi-ciently copied by NS5B [26,27]. As expected, the cap-independenttranslation of RLuc is carried out with HCV IRES located withinthe 5 0 -UTR, with the involved RdRp activity measured by lumines-cent detection. In contrast, the FLuc gene undergoes cap-depen-dent translation (Fig. 1A and b). To determine whether (+)RLucRNA is catalyzed by HCV RdRp, a reverse-transcription reactionwas performed using two specific primers, primer 1 and primer2, that anneal to the 3 0 end of the (  )5 0 -UTR-RLuc-3 0 -UTR and(+)5 0 -UTR-RLuc-3 0 -UTR RNAs, respectively (arrowhead line shownin Fig. 1B (a)), and were used in the subsequent transient transfec-tion of the p(+)FLuc-(  )UTR-RLuc reporter plasmid in cells with orwithout RdRp expression. Then, both cDNAs were amplified by apolymerase chain reaction with the primers corresponding to theRLuc gene. Based on the generation of opposite stranded 5 0 -UTR-RLuc-3 0 -UTR RNAs described above (Fig. 1A, (b)), primer 1 will de-tect transcripts generated by cellular polymerase, in contrast, pri-mer 2 will detect HCV RdRp-generated transcripts. As shown inFig. 1B (b), (  )5 0 -UTR-RLuc-3 0 -UTR RNA was initially transcribedby host polymerase following self-cleavage of the HDV ribozyme(lanes 1, 3, and 5). The expected reverse-transcription PCR productof the 252-bpRLuc cDNAwas observedin BHK-NS5Bcells (lane4),which are stably transfected with HCV NS5B polymerase. This wasnot observed in parental BHK21 cells (lane 2). In this experiment,Ava5 [11], an HCV subgenomic replicon-containing cell line, con-taining a complete HCV replication complex served as the positivecontrol (lanes 5 and 6). In parallel, the activity of FLuc and RLuc inthe cells was determined at 72h posttransfection. As shown inFig. 1B (c) (white bars), there was no significant difference in FLucactivity among these cell lines, indicating that transfection effi-ciency is similar in each cell line. Consistent with the productionof RLuc transcripts fromRT-PCR analysis, only the background sig-nal of RLuc was observed in the parental cell (column 1); in con-trast, higher RLuc activity was observed in NS5B-expressing cellsand Ava5 cells, indicating that the (+) RLuc transcripts can be gen-erated by either NS5B alone or NS5B within the replicative com-plex containing other viral proteins and cellular proteins (graybars in columns 2 and 3). Compared to RLuc activity in Ava5 cells,lowerRLucactivityintheBHK-NS5Bcellsmaybeduetothelackof a complete replication complex, even though RLuc activity is stillsignificantly detectable. Establishment and characterization of the cloned cell lines containing HCV RdRp and reporter construct  AsRLucactivitywas significantlyinducedinthetransient assaydescribed above, stably transfected cell lines were created for fur-ther investigation. Initially, cell colonies were selected in the pres-ence of G418 following transfection of p(+)FLuc-(  )UTR-RLuc intoBHK-NS5B cells. Depending on RLuc activity, which reflects thepolymerization activity of NS5B, 2 individual lines, designated asBHK-NS5B-FRLuc-1 and BHK-NS5B-FRLuc-2, with the highest RLucactivity and optimal growth kinetics were isolated in the presenceofdoubleantibioticselection(blasticidinandG418),withtheircul-tureandthenscaledupforuseinsusceptibilityassays.After4daysofincubationina24-wellplate,thecelllysateswereharvestedandanalyzed for luminometric activity by using the Dual-Glo Lucifer-ase Assay System (Promega). RLuc activity in the reporter cells,constitutively expressing the NS5B protein (Fig. 2A (a), gray bars),was approximately 25- to 40-fold higher than that in the cells,BHK-FRLuc,harboringonlythe(+)FLuc-(  )UTR-RLucreporter.FLucactivity indicated the copy number of the integrated reporter DNAthat was transcribedinto various levels of bicistronic reporter RNAineachindividualclone(whitebars).Absolutequantificationofthereporter plasmid copy number was performed using real-timeqPCRamplificationsof thegenomicDNA(gDNA)fromtwoindivid-ual BHK-NS5B-FRLuc cells with primers corresponding to the RLucgene, in which the levels of the cellular  gapdh  gene serves as theloading control. Corresponding to the linear standard curve con-structed by serially diluted reporter plasmid and the  C  t  values( R 2 =0.996), the total gDNA of BHK-NS5B-FRLuc-1 and BHK-NS5B-FRLuc-2 harbored reporter plasmid copy numbers of 31.3and 81.2, respectively (Fig. 2A (b)). The expression of the HCVNS5B protein in each clone was examined by Western blottingwith an NS5B-specific monoclonal antibody; glyceraldehyde-3-phosphate dehydrogenase (GAPDH) served as the loading control.No significant difference was observed in the amount of NS5Bpro-tein in both clones (Fig. 2B), concluding that the higher lumines-cent signal of the BHK-NS5B-FRLuc-2 clone, compared to that of the BHK-NS5B-FRLuc-1 clone, was because the former cells har-bored high copy numbers of chromosomally integrated reporterDNA. However, we cannot exclude the possibility that the FLucactivity may be influenced by chromosomal integration position.Therefore, BHK-NS5B-FRLuc-2 cells with a larger signal-to-back-ground ratio provide a suitable tool for further drug screening.Based on the principle of RNA polymerization on biscistronic RNA, 56  Intracellular HCV RdRp activity assay/J.-C. Lee et al./Anal. Biochem. 403 (2010) 52–62
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks