Inhibition of Dengue Virus by Targeting Viral NS4B Protein ^▿

Cells, viruses, and compounds.

A549 human alveolar epithelial cells were maintained in F-12 medium containing 10% fetal bovine serum (FBS) and 1% penicillin/streptomycin. BHK-21 (baby hamster kidney) and Vero (African green monkey) cells were cultured in Dulbecco modified Eagle medium (DMEM) supplemented with 10% FBS and 1% penicillin/streptomycin. C6/36 (mosquito) cells were grown in RPMI 1640 medium containing 10% FBS and 1% penicillin/streptomycin. A549, BHK-21, and Vero cells were incubated at 37°C, and C6/36 cells were maintained at 28°C.

The viruses used in this study included DENV serotype 1 (DENV-1) (West Pacific strain), DENV-2 (New Guinea C and TSV01 strains), DENV-3 (H87 strain), DENV-4 (H241 strain), WNV (strain 3356), YFV (17D vaccine strain), Western equine encephalomyelitis virus (WEEV) (strain Cova746), Chikungunya virus (CHIKV) (Ross strain), and vesicular stomatitis virus (VSV) (New Jersey serotype). The sources of these viruses were reported previously (30, 31). The synthesis of compound NITD-618 will be reported elsewhere. The compound was dissolved in 90% dimethyl sulfoxide (DMSO) as a stock for in vitro experiments.

HTS using a luciferase reporter replicon of DENV-2.

A549 cells containing a luciferase reporter replicon of DENV-2 (New Guinea C strain) were used for HTS. Renilla luciferase and puromycin acetyltransferase genes were engineered in the replicon to replace viral structural proteins, as described previously (24). Briefly, A549 DENV-2 replicon cells were seeded at a density of 750 cells per well in a 1,536-well microplate. After incubation at 37°C with 5% CO₂ overnight, the cells were treated with compounds. Luciferase activities were measured after 48 h of incubation using Promega's EnduRen live-cell substrate. Following luciferase activity measurement, Promega's CellTiter-Glo reagent was added to each well to determine the cytotoxicity of the compounds. The compounds were screened at a single concentration of 5 μM.

CFI assay.

The cell-based flavivirus immunodetection (CFI) assay was performed as described previously (37). In brief, 2 × 10⁴ A549 cells per well in a 96-well plate were infected with DENV-2 (New Guinea C strain) at a multiplicity of infection (MOI) of 0.3 in the presence of 2-fold serial dilutions of the compound. At 48 h postinfection (p.i.), the infected wells were washed, fixed, and immunoblotted with primary antibody (4G2; American Type Culture Collection) and horseradish peroxidase (HRP)-conjugated anti-mouse IgG (Sigma) to determine the yield of viral envelope protein. A dose-response curve was plotted using Prism 5 software (GraphPad Software, La Jolla, CA), and the effective concentration of compound to reduce envelope protein production by 50% (EC₅₀) was calculated using nonlinear regression analysis.

Viral-titer reduction assay.

A viral-titer reduction assay was performed for the following viruses: DENV-1, DENV-2, DENV-3, DENV-4, CHIKV, WNV, YFV, WEEV, and VSV. Approximately 2 × 10⁴ cells (A549, BHK-21, or Vero) or 8 × 10⁴ cells (C6/36) were seeded per well of 96-well plates. At 24 h postseeding, A549 cells were infected with DENV-2 (MOI, 0.3; New Guinea C strain) or CHIKV (MOI, 0.1); Vero cells were infected with WNV, YFV, WEEV, or VSV (MOI, 0.1); and BHK-21 cells were infected with DENV-1 to -4 (MOI, 0.3). The infected cells were immediately treated with series dilutions of compound. Because of the difference in replication kinetics among different viruses, culture fluids were collected at different time points postinfection: CHIKV and VSV infections at 16 h p.i.; WNV, YFV, and WEEV infections at 42 h p.i.; DENV-1 and DENV-2 infections at 48 h p.i.; and DENV-3 and DENV-4 infections at 72 h p.i. C6/36 cells were infected with wild-type (WT) or mutant DENV-2 (strain TSVO1; MOI, 0.3), and culture fluids were collected at 72 h p.i. The viral titers of all samples collected were quantified using a plaque assay (30).

Cytotoxicity assay.

Cell viability was measured using the Cell Counting Kit-8 (CCK-8) (measuring cellular dehydrogenase activity; Dojindo Molecular Technologies) according to the manufacturer's protocol. A549, BHK-21, or Vero cells were seeded at 5 × 10³ cells per well in a 96-well plate. After 24 h of incubation at 37°C in 5% CO₂, the cells were treated with 2-fold serial dilutions of the compound. At 48 h posttreatment, 4 μl of CCK-8 solution was added to each well. After another 90 min of incubation at 37°C, the absorbance was measured at 450 nm using a microplate reader (Tecan).

Antiviral assay in K562 cells.

A human myelogenous leukemia cell line, K562, was used to study the antiviral activity of NITD-618 against a luciferase reporter DENV-2 (New Guinea C strain) (46). K562 cells were seeded at 1 × 10⁴ cells per well in a 96-well plate. After overnight incubation at 37°C in 5% CO₂, the cells were infected with the reporter DENV-2 at an MOI of 1.0 in the presence of serial dilutions of NITD-618. At 48 h p.i., Renilla luciferase activity was quantified using ViviRen Live Cell Substrate (Promega), after which cytotoxicity was measured by the intracellular level of ATP using a CellTiter-Glo luminescent-cell viability assay (Promega) according to the manufacturer's protocol.

Time-of-addition assay.

Approximately 2 × 10⁵ A549 cells were seeded per well of a 24-well plate. At 24 h postseeding, the cells were infected with DENV-2 (New Guinea C strain; MOI, 2.0) for 1 h at 4°C. Subsequently, the viral inocula were removed, and the cells were washed three times with cold phosphate-buffered saline (PBS) to remove unabsorbed viruses. At 0, 2, 4, 6, 8, 10, 12, 14, 16, 18, or 20 h p.i., 5 μM compound was added to the infected cells. As negative controls, 0.9% DMSO was added to the infected cells at 0, 10, and 20 h postinfection. At 24 h p.i., culture fluids were collected, and the viral titer was determined by plaque assay (30).

Selection and sequencing of a DENV-2 replicon resistant to NITD-618.

BHK-21 cells containing DENV-2 replicon (RepPAC-2A-EGFP; New Guinea C strain) were used to select for a replicon resistant to NITD-618. RepPAC-2A-EGFP contained two reporters (a puromycin acetyltransferase [PAC] and enhanced green fluorescent protein [EGFP]) separated by foot-and-mouth disease virus (FMDV) 2A protein (24). The RepPAC-2A-EGFP BHK-21 cells were first cultured in medium containing 6 μM NITD-618 for 3 days (P₀). The cells were then passaged in medium containing 6 μM NITD-618 and 10 μg/ml puromycin for three rounds (P₁ to P₃; 3 days per passage round). The cells were further passaged in medium containing 12 μM NITD-618 and 20 μg/ml puromycin for three rounds (P₄ to P₆; 3 days per passage round). Cells from P₆ were subjected to fluorescence-activated cell sorter (FACS) analysis using a FACSAria flow cytometer (BD Biosciences). The top 10% of the cells with the strongest GFP signal were recovered and expanded. The foci of the sorted cells were cloned and expanded in medium containing 12 μM NITD-618 and 20 μg/ml puromycin. Individually cloned cells were tested for compound sensitivity using FACS analysis. Data analysis was performed using FlowJo software (version 8.2; Tree Star Inc., Ashland, OR). Total cellular RNA of the cloned cells was extracted using RNeasy kits (Qiagen). Viral replicon cDNA was amplified using SuperScript One-Step reverse transcription (RT)-PCR with Platinum Taq (Invitrogen). The RT-PCR products were gel purified and subjected to DNA sequencing.

Construction of recombinant plasmids.

Recombinant plasmids were constructed by using an infectious cDNA clone of DENV-2 (pACYC TSVFL; strain TSV01), a cDNA clone of a Renilla luciferase replicon of DENV-2 (pACYC TSV replicon; strain TSV01), and one shuttle vector, TA TSV-F (containing cDNA from nucleotide 5426 to the 3′ end of the DENV-2 genome; GenBank accession number AY037116). The construction of plasmids pACYC TSVFL and pACYC TSV replicon and the shuttle vector TA TSV-F was reported previously (45). NS4B mutations and an NS5stop mutation (the first amino acid [Gly] of NS5 was mutated to a stop codon) were first engineered into the shuttle vector TA TSV-F using a QuikChange II XL site-directed mutagenesis kit (Stratagene). The fragment containing the mutation(s) was then engineered into the infectious cDNA clone and the replicon cDNA clone using XhoI (nucleotide position 5426) and ClaI (immediately downstream of the 3′ end of the viral genome). The primer pairs used for mutagenesis are listed in Table 1. All constructs were verified by DNA sequencing.

Transient replicon assay.

A Renilla luciferase replicon of DENV-2 was in vitro transcribed using a T7 mMessage mMachine kit (Ambion, Austin, TX) from cDNA plasmids linearized with ClaI, as described previously (6). BHK-21 cells were electroporated with 10 μg of replicon RNA using a GenePulser Xcell system (Bio-Rad, Hercules, CA) and an established protocol (6). For transfection of Vero, A549, and C6/36 cells, 8 × 10⁶ cells in 0.8 ml cold Ingenio electroporation solution (Mirus Bio, Madison, WI) were electroporated with 10 μg of replicon RNA in a 4-mm cuvette at settings of 450 V and 25 μF with three pulses at 3-s intervals. The transfected cells (BHK-21, Vero, and A549) were seeded in a 12-well plate (3 × 10⁵ cells per well). At various time points posttransfection (p.t.), the cells were washed once with PBS and lysed in 200 μl 1× lysis buffer (Promega). The plates containing the lysis buffer were sealed with Parafilm and stored at −80°C. Once samples for all time points had been collected, 20 μl of cell lysates was transferred to a 96-well plate and assayed for luciferase signals in a Clarity luminescence microplate reader (BioTek). For testing compound inhibition of the replicon, the transfected A549 (2 × 10⁴ cells per well), BHK-21 (2 × 10⁴ cells per well), or C6/36 (8 × 10⁴ cells per well) cells were seeded in a 96-well plate. Immediately after seeding, the cells were treated with serial dilutions of NITD-618 or 0.9% DMSO. At 48 h p.t. (A549 cells and BHK-21 cells) or 96 h p.t. (C6/36 cells), the cells were washed once with PBS, lysed in 20 μl 1× lysis buffer, and assayed for luciferase activity as described above.

Production of recombinant viruses, SIA, and IFA.

Recombinant viruses were produced as described previously (45). Briefly, BHK-21 cells were transfected with 10 μg of WT or mutant genome length RNA containing the NS4B P104L, A119T, or P104L plus A119T mutation. The electroporated cells were incubated at 37°C for 24 h and then at 30°C for an additional 96 h. On day 5 p.t., culture fluids were harvested, aliquoted, and stored at −80°C. The titers of recombinant viruses were quantified by plaque assay using BHK-21 cells (30). For specific infectivity assay (SIA), 1 ml of a series of 1:10 dilutions of the transfected cells was seeded onto confluent BHK-21 cell monolayers (6 × 10⁵ cells per well seeded in a six-well plate 2 days in advance). The seeded cells were allowed to attach to the plates for 5 h before addition of 3 ml RPMI 1640 medium containing 0.8% methyl cellulose (Aquacide II; Calbiochem) and 2% FBS. After 5 days of incubation at 37°C with 5% CO₂, the cells were fixed in 3.7% formaldehyde (Sigma) and stained with 1% crystal violet. An immunofluorescence assay (IFA) was performed as described previously (34), using anti-E monoclonal antibody 4G2 and Alexa Fluor 488 goat anti-mouse IgG as primary and secondary antibodies, respectively.

Trans complementation analysis.

Two types of trans complementation experiment were performed. For trans complementation of an NS4B lethal mutant, BHK-21 cells with or without the WT replicon of DENV-2 (TSV01) were transfected with equal amounts (10 μg) of WT or mutant genome length RNA. At 96 h p.t., the transfected cells were assayed for viral E protein expression using IFA, as described above. For trans complementation between the WT replicon and compound-resistant virus, A549 cells were transfected with 10 μg of WT luciferase replicon RNA of DENV-2. The transfected cells were resuspended in DMEM containing 10% FBS at a density of 2 × 10⁵ cells/ml; 100 μl of the cell suspension was immediately seeded in a 96-well white polystyrene microplate (Greiner, Germany). At 18 h p.t., the culture medium was removed and the cells were infected with WT or P104L plus A119T mutant DENV-2 (MOI = 10) or incubated with 50 μl DMEM supplemented with 2% FBS (as a no-virus infection control). At 20 h p.t., the inocula were removed and the cells were washed twice with PBS and treated with serial dilutions of NITD-618 in fresh medium containing 2% FBS. At 48 h p.t., the cells were lysed in 20 μl 1× lysis buffer, and luciferase activity was measured. A two-way analysis of variance (ANOVA) test was performed to indicate statistical significance among different experimental groups.

Identification of NITD-618.

We performed an HTS using an A549 cell line containing a luciferase replicon of DENV-2. The HTS was performed in a 1,536-well format with a signal-to-noise ratio of 5 and a Z′ value of 0.4. Approximately 1.8 million compounds with diverse structures were screened at a single concentration of 5 μM. Compounds inhibiting ≥50% of luciferase activity were defined as “hits.” The HTS had a hit rate of 0.96%. The screening allowed us to identify compound NITD-618 (Fig. 1A), which suppressed 85% of the replicon luciferase activity at 5 μM (data not shown). The antiviral activity of NITD-618 was validated using a DENV CFI assay. The CFI assay is an ELISA-based test that measures the amount of viral E protein in cells infected with DENV-2 (Fig. 1B). The compound reduced viral E protein production in a dose-responsive manner, with an EC₅₀ of 1.0 μM (Fig. 1B). Next, the anti-DENV activity was further confirmed using a viral-titer reduction assay. As shown in Fig. 1C, NITD-618 inhibited virus production, with an EC₅₀ of 1.6 μM. A cytotoxicity assay showed that the compound was not cytotoxic up to 40 μM in Vero, BHK-21, and A549 cells (Fig. 1D), suggesting that the observed antiviral activity was not due to compound-mediated cytotoxicity. The antiviral activity of NITD-618 was further validated using K562 cells (a human myelogenous leukemia cell line) infected with a luciferase reporter DENV-2. The compound showed an EC₅₀ of 1.8 μM with no detectable cytotoxicity up to 20 μM in K562 cells (Fig. 1E).

Specific inhibition of DENV.

We examined the antiviral spectrum of NITD-618. The compound was tested in a viral-titer reduction assay using three other serotypes of DENV (DENV-1, -3, and -4), two closely related flaviviruses (WNV and YFV), two plus-strand RNA alphaviruses (CHIKV and WEEV), and a negative-strand RNA rhabdovirus (VSV). NITD-618 was active against three other serotypes of DENV, with EC₅₀s of 1.5, 1.6, and 4.1 μM against DENV-1, -3, and -4, respectively (Fig. 2). In contrast, the compound (up to 40 μM) did not suppress titers of any other viruses, including the closely related WNV and YFV. We did not test the compound at concentrations of >40 μM due to potential cytotoxicity (Fig. 1D). These results demonstrate that NITD-618 selectively inhibits all four serotypes of DENV.

Suppression of viral RNA synthesis.

A time-of-addition experiment was performed to determine the stage of inhibition in a viral infection cycle. A549 cells were synchronously infected with DENV-2. NITD-618 (5 μM) was added to the infected cells at various time points after infection, and viral yields were quantified at 24 h postinfection. The inhibition of virus production by NITD-618 gradually diminished when the compound was added at time points later than 12 h p.i. (Fig. 3A), suggesting that the compound blocks a late stage(s) of the viral life cycle.

Since NITD-618 was identified using replicon screening, the compound should inhibit viral translation and/or RNA synthesis. To differentiate between the two steps, we performed a transient replicon assay using a luciferase replicon of DENV-2. The luciferase replicon was electroporated into A549 cells. The transfected cells were immediately treated with NITD-618 (12 μM) or DMSO (0.9% as a negative control) and assayed for luciferase activity at various time points posttransfection. As shown in Fig. 3B, the compound showed no effect on luciferase signals at 2 and 4 h p.t. (representing translation of input replicon RNA). In contrast, the compound suppressed luciferase activities at >15 h p.t. (indicating replicon RNA synthesis). At 24, 36, and 48 h p.t., NITD-618 blocked the luciferase signals to <1% of those of the DMSO control group. The results indicate that NITD-618 inhibits viral RNA synthesis.

To examine whether the compound inhibits replication through suppression of known viral enzymes, we tested NITD-618 in the DENV protease (2), NTPase (40), methyltransferase (4), and RdRp (25) assays. None of the enzymatic activities were inhibited by the compound at up to 20 μM (data not shown).

Selection and characterization of resistant replicon cells.

To identify the target of NITD-618, we selected resistant replicons of DENV-2 by using a flow cytometry-based cell-sorting method (33). BHK-21 cells containing a GFP replicon of DENV-2 (Fig. 4A) were continuously passaged in medium with increasing concentrations of NITD-618 (Fig. 4B). After 7 rounds of passaging, the cells with the top 10% of EGFP signal were pooled through live-cell sorting. Individual cells were cloned, expanded, and examined for compound sensitivity by FACS analysis. Figure 4C shows the compound inhibition result for one of the selected replicon cell lines (clone 1). Upon compound treatment, the GFP signals from WT replicon cells were effectively suppressed, with an EC₅₀ of 0.4 μM. In contrast, the GFP signals from selected replicon cells were not reduced but increased in the presence of the compound. Microscopic analysis showed that the EGFP fluorescence from the selected replicon cells was not suppressed by the compound, whereas the EGFP signal from the WT replicon cells was completely inhibited by 20 μM compound (Fig. 4D). Similar resistance phenotypes were observed for two other independently selected replicon cells (clones 2 and 3) (data not shown). These results demonstrate that resistant replicons could be selected in cell culture.

We sequenced the complete genomes of three independently selected resistant replicons. The mutations are summarized in Table 2. Replicon from clone 1 accumulated two mutations. (i) A C→U change at nucleotide position 7136 of the viral genome resulted in a Pro→Leu substitution at amino acid position 104 of the NS4B protein (P104L). This mutation was recovered from all three independently selected replicons. (ii) A G→A mutation at nucleotide position 7180 led to an Ala→Thr change at amino acid position 119 of the NS4B protein (A119T). This mutation was recovered from two of the three selected replicons (clones 1 and 2). It should be noted that the P104L and A119T NS4B mutations were not recovered in replicons passaged without compound treatment (negative controls).

Sequence alignment of NS4B showed that P104 and A119 are absolutely conserved among all four serotypes of DENV but are different among other flaviviruses (Fig. 4E). Notably, the mutated 104Leu residue (recovered from the resistant DENV-2 replicon) exists as the WT amino acid in the JEV, St. Louis encephalitis virus (SLEV), and WNV NS4B proteins. Such differences at amino acid 104 of NS4B may account for the selectivity of NITD-618 in inhibiting DENV (Fig. 2). According to the NS4B topology proposed by Miller and colleagues (21), amino acids P104 and A119 are located in the third transmembrane domain and are embedded in the ER membrane (Fig. 4F).

Replicon analysis of P104L and A119T mutations.

The functions of NS4B mutations were analyzed using a luciferase replicon of DENV-2. We generated three mutant replicons containing mutation P104L alone or A119T alone or the combined mutations P104L plus A119T. Initially, we examined the effects of these mutations on viral replication in the absence of compound. A549 cells were transfected with equal amounts of replicon RNAs and assayed for luciferase activities at various time points after transfection. None of the mutations affected the translation of input RNA at 2, 4, and 6 h p.t. (Fig. 5A). In contrast, luciferase signals at ≥24 h p.t. showed differences in replication efficiencies among various replicons in the following order: A119T > P104L plus A119T > WT ≥ P104L. To examine whether the observed effects on viral replication were cell type specific, we compared the replicon replication in two other cells lines. Figures 5C and D show the luciferase signals from replicon-transfected BHK-21 (at 48 h p.t.) and C6/36 (at 96 h p.t.) cells, respectively. In BHK-21 cells, the replication levels of all three mutant replicons were slightly reduced compared with the WT (Fig. 5C). In C6/36 cells, the A119T mutant replicated to a level comparable to that of the WT replicon, whereas the P104L and P104L plus A119T mutants replicated much less efficiently than the WT replicon (Fig. 5D). Taken together, the results demonstrate that NS4B mutations (P104L and A119T) affect viral replication in a host species-dependent manner.

Next, we compared compound inhibition between the WT and mutant replicons. The replicon-transfected cells (A549, BHK-21, and C6/36) were treated immediately after transfection with various concentrations of NITD-618 and assayed for luciferase activities at the indicated time points. In A549 (Fig. 5A and B) and BHK-21 (Fig. 5C) cells, mutation P104L or A119T alone conferred partial resistance; the double mutation P104L plus A119T enhanced the resistance in an additive manner. In C6/36 cells (Fig. 5D), the single mutation P104L and the double mutation P104L plus A119T conferred partial resistance, whereas mutant A119T did not show any resistance. Interestingly, the compound inhibited the WT replicon in A549 cells more efficiently that in BHK-21 and C6/36 cells. Nevertheless, these results indicate that both NS4B mutations play roles in conferring resistance in mammalian cells, whereas only mutation P104L (but not mutation A119T) functions in resistance in mosquito cells.

Recombinant virus analysis of P104L and A119T mutations.

To confirm the above replicon results, we validated the functions of NS4B mutations in the context of genome-length RNA. Three recombinant mutant viruses (P104L, A119T, and P104L plus A119T) were prepared using an infectious cDNA clone of DENV-2. Three mutant viruses produced plaques with sizes comparable to that of the WT virus on BHK-21 cells (Fig. 6A). Interestingly, mutant viruses P104L and P104L plus A119T produced opaque plaques. Sequencing of the recombinant viruses confirmed that the engineered mutations were retained in the recovered viruses without any other changes (data not shown).

Next, we examined the sensitivity of mutant viruses to NITD-618 inhibition using a viral-titer reduction assay. In A549 cells (Fig. 6B), the single mutation P104L or A119T alone conferred resistance; the double mutation P104L plus A119T enhanced the resistance. For example, treatment with 5 μM NITD-618 reduced the WT virus titer by ≥2.4 × 10³-fold to an undetectable level, whereas treatment with the same concentration of NITD-618 reduced the titers of the P104L, A119T, and P104L plus A119T viruses by 19-, 140-, and 4.6-fold (fold reduction = viral titer without compound treatment/viral titer with compound treatment), respectively. In C6/36 cells (Fig. 6C), the resistance phenotype was much less pronounced, only mutants P104L and P104L plus A119T showed resistance, and the resistance was clearly observed when treated with 20 μM NITD-618. Collectively, the results demonstrate that, in A549 cells, (i) both P104L and A119T mutations in NS4B are responsible for resistance and (ii) the double mutation P104L plus A119T improves resistance in an additive manner, whereas in C6/36 cells, only P104L is responsible for resistance.

Effects of other P104 substitutions on viral replication and resistance.

To examine whether amino acid substitutions other than P104L of NS4B could confer resistance to NITD-618, we generated a panel of mutant luciferase replicons, each of which contained a single substitution: P104A, P104V, P104E, or P104R. We first compared the replication efficacies of the WT and mutant replicons in the absence of compound treatment. BHK-21 cells were selected for this experiment because the cell line is incompetent in producing interferon (29). Equal amounts of replicon RNAs were transfected into BHK-21 cells, and luciferase activities were measured at various time points posttransfection. As shown in Fig. 7A, P104R completely abolished viral RNA synthesis; P104E reduced viral RNA replication; and P104A, P104V, and P104L slightly attenuated luciferase activity at 24 h posttransfection.

Next, we compared drug sensitivities among the mutant replicons by incubating the replicon-transfected cells with NITD-618. A549 cells were used for this experiment because the cell line showed higher sensitivity than BHK-21 and C3/36 cells in compound inhibition (Fig. 5B to D). Luciferase activities at 48 h p.t. showed that mutant replicons were less sensitive to compound inhibition than the WT replicon; the sensitivity to compound inhibition was on the order of WT > P104V > P104A > P104L (Fig. 7B). The mutant P104E replicon was not included in the inhibition assay because the mutation was not stable (see details below). As a positive control, NITD-008, a known nucleoside inhibitor of flavivirus (41), suppressed both WT and mutant replicons with equal potency (data not shown).

To further analyze the effect of P104 substitution on viral replication, we engineered individual mutation of P104A, P104V, P104E, or P104R into the genome length RNA of DENV-2. Equal amounts of genome length RNAs were transfected into BHK-21 cells and analyzed for their abilities to express viral E protein using an IFA (Fig. 8A). The transfected cells were directly seeded onto monolayers of BHK-21 cells for plaque development (Fig. 8B). (i) For P104A and P104V mutants, the transfected cells generated slightly fewer IFA-positive cells than the WT RNA-transfected cells. Both P104A and P104V RNAs produced infectious viruses with a homogeneous plaque morphology. Sequencing analysis showed that the recovered viruses retained the P104A and P104V mutations without any other changes (Fig. 8C). (ii) For the P104E mutant, the transfected cells yielded far fewer IFA-positive cells than the WT-RNA-transfected cells. The plaques of P104E virus were smaller than those of the WT virus (Fig. 8C). Sequencing of the viruses derived from culture fluids of the transfected cells showed the engineered P104E mutation. Passaging of the recovered viruses on Vero cells for six rounds (4 days per round) generated viruses with mixed plaque morphologies (Fig. 8D). Sequencing of the plaque-purified viruses showed that viruses with big plaques contained a P104E-to-P104G reversion whereas viruses with small plaques accumulated a P104E-to-P104T change (Fig. 8D). These results suggest that the P104E mutation attenuates viral replication and that the P104E mutation could revert to P104G or P104T. (iii) For the P104R mutant, the transfected cells did not generate any IFA-positive cells (Fig. 8A) or plaques (Fig. 8B). Passaging the supernatant from transfected cells on Vero cells for 4 days recovered virus with clear plaque morphology; sequencing of the recovered virus showed a 104R-to-104I reversion (Fig. 8D). These data indicate that the P104R mutation is lethal for viral replication.

We performed full-genome sequencing of the following revertant viruses: P104E-derived virus exhibiting big plaques, P104E-derived virus exhibiting small plaques, and P104R-derived virus. The sequencing results showed that, besides the NS4B mutations presented in Fig. 8D, mutations outside NS4B had accumulated in those revertant viruses (data not shown). Experiments are ongoing to define the functions of the recovered mutations.

Trans complementation of NS4B.

Since NS4B P104R is lethal for viral replication, we asked whether the lethal phenotype could be trans complemented. To address this question, we transfected WT and mutant NS4B P104R genome length RNAs into BHK-21 cells containing the DENV-2 replicon. Since the replicon does not contain viral structural genes, expression of E protein in the transfected cells was used to monitor the replication of genome length RNA. As shown in Fig. 8E, WT genome length RNA yielded E protein-expressing IFA-positive cells. However, the IFA-positive cells were fewer than when naive BHK-21 cells were transfected with the WT RNA (compare Fig. 8A and E); the difference in the number of IFA-positive cells was most likely due to exclusion of homologous virus replication in the replicon-containing cells (47). Interestingly, the NS4B P104R genome length RNA generated IFA-positive signals in the replicon BHK-21 cells (Fig. 8E). As a negative control, transfection of NS5stop RNA (containing a substitution for the first amino acid of NS5 by a stop codon) into the replicon BHK-21 cells did not generate any IFA-positive cells. As expected, no IFA-positive cells were detected when naive BHK-21 cells (without replicon) were transfected with the NS4B P104R or NS5stop RNA (Fig. 8E). The results demonstrate that NS4B could be trans complemented.

Next, we examined whether coinfection of compound-sensitive and compound-resistant viruses would affect antiviral efficacy. To explore this possibility, we transfected A549 cells with WT DENV-2 luciferase replicon. The transfected cells were infected with WT or P104L plus A119T resistant virus. The cells were then treated with NITD-618 and assayed for luciferase activity (derived from the WT replicon). Figure 9A depicts the experimental procedure. The results showed that coinfection with WT virus did not change the sensitivity of the WT replicon to NITD-618 inhibition (Fig. 9B). In contrast, coinfection of P104L plus A119T resistant virus reduced the compound inhibition of the WT replicon (Fig. 9B); such reduction of WT replicon compound inhibition was statistically significant (Fig. 9C). These results indicate that trans complementation of NS4B could affect drug sensitivity when a cell is coinfected with drug-sensitive and drug-resistant viruses.