Orally Efficacious Broad-Spectrum Ribonucleoside Analog Inhibitor of Influenza and Respiratory Syncytial Viruses

Focusing on the dual-active compounds only, two candidates were excluded based on excessive toxicity in dose-response potency and toxicity counterscreens (see Fig. S1 in the supplemental material), while the remaining candidate (dual 1, EIDD-1931 or N⁴-hydroxycytidine [NHC]) (Fig. 1B) showed active concentrations in the nanomolar to low-micromolar range and selectivity indices (SIs) (SI = CC₅₀ [50% cytotoxic concentration]/EC₅₀ [50% effective concentration]) of ≥89 against a broad panel of RSV, IAV, and IBV laboratory strains and isolates (Table 1). This group included clinical RSV isolates cultured from nasal wash specimens (Fig. 1C); IAVs of human, avian, and swine origins representing both group 1 and 2 hemagglutinins (HAs) (Fig. 1D and Table 1); highly pathogenic H5N1 and emerging H7N9 avian IAVs (AIVs) (Fig. 1E); and IBVs representing both circulating lineages (35), Victoria and Yamagata (Fig. 1F). A subset of avian influenza viruses was tested in ovo (Table 1), demonstrating NHC potency in primary cells. Consistent with the high initial SI values in cultured cells, embryo development in the treated chicken eggs was visually unaffected by the compound (Fig. S2). Whereas T-705 was inactive in primary HBTECs, the antiviral activity of NHC against both RSV and influenza virus was unchanged in these disease-relevant human primary cells compared to immortalized cell lines (Fig. 1G). Drug combination testing of NHC and the current standard of care (SOC) against IAV infection, oseltamivir, in cultured cells identified an extended plateau area of medium-level antiviral synergy (HSA model), while no significant increase in cytotoxicity was noted in the presence of the drug combination (Fig. S3). These data demonstrate the activity of NHC against a panel of respiratory viruses associated with influenza-like diseases. Combined with the previously reported antiviral activity of NHC in cell culture against some Flaviviridae, Coronaviridae, and Togaviridae family members (36–39), these data establish the broad-spectrum activity of the compound against different positive- and negative-strand RNA virus families and, in the case of IAV infection, spotlight the potential for synergistic combination with the current SOC.

Based on the reversal of NHC-mediated inhibition of a chikungunya virus replicon through an excess of exogenous cytidine and uridine, the compound was thought to behave as a pyrimidine analog in cellula (37). We first confirmed through time-of-addition variation studies that NHC blocks a postentry step in the influenza virus and RSV replication cycles (Fig. 2A). Plasmid-based minigenome reporter assays validated in both cases NHC interference with viral RdRp activity (Fig. 2B). However, an in vitro activity assay using host cell polymerase α did not reveal inhibition of the cellular polymerase by the active 5′-triphosphate form of the compound, NHC-TP (Fig. 2C). As previously observed for positive-strand RNA virus replicons, the inhibition of RSV and influenza virus by NHC can be reversed through the addition of exogenous pyrimidines but not purines (Fig. 2D). To determine whether NHC-TP is accepted by negative-strand RNA virus RdRps as a substitute for CTP or acts as a chain terminator, we employed a biochemical assay of RSV polymerase activity that uses purified viral phosphoprotein (P) and large protein (L) RSV polymerase components, a nucleoside triphosphate (NTP) mix, and a synthetic oligonucleotide template, resulting in a 23-mer product (40). Without CTP in the NTP mix, polymerization is stalled at the first guanidine residue in the template, releasing 14-mer amplicons. The addition of NHC-TP partially restored the generation of full-length 23-mer amplicons, demonstrating that the RSV RdRp complex accepts NHC in place of cytidine and that the compound does not act as an obligatory chain terminator (Fig. 2E). Rather, we noted significantly increased frequencies of C-to-U, G-to-A, and, in the case of RSV only, A-to-G transition mutations in viral RNA after single-cycle infection of cells in the presence of 10 μM NHC, followed by subcloning of individual viral RNA-derived cDNA amplicons and Sanger sequencing (Fig. 2F). Relative mutation frequencies were increased approximately 7- to 10-fold (Table 2), but no specific mutation hot spots were detected in the fragment analyzed. This observation is consistent with the postulated ability of NHC to base pair as either cytosine or uracil (41), which may drive the replicating viruses into error catastrophe (36, 42).

Previous attempts to induce robust resistance of two positive-strand RNA viruses, Venezuelan equine encephalitis virus (VEEV) and bovine viral diarrhea virus, from inhibition by NHC were unsuccessful (36, 43). To resistance profile negative-strand RNA virus-derived RdRps, we incubated RSV-A2 and IAV-WSN in the presence of gradually increasing NHC concentrations for 10 consecutive passages. However, no robust resistance (>5-fold increase in EC₅₀s) emerged, indicating a universally high barrier against viral escape from NHC, independent of the target virus examined.

In preparation for in vivo testing, we determined anabolic, pharmacokinetic (PK), and pharmacodynamic (PD) profiles and lung tissue distributions of the compound in primary HBTECs and mice. Liquid chromatography-tandem mass spectrometry (LC-MS/MS) analysis after exposure of HBTECs to 20 μM NHC demonstrated effective conversion to active NHC-TP, represented by a concentration plateau at approximately 4 nmol/10⁶ cell over the 24-h time period examined. In contrast, steady-state levels of free prodrug and NHC–5′-monophosphate (NHC-MP) remained flat at a low ∼2 pmol/10⁶ cells in this period (Fig. 3A). Subsequent washout revealed a high metabolic stability of NHC-TP in the HBTECs, resulting in a calculated half-life exceeding 4 h (Fig. 3B).

For species consistency with our mouse efficacy models of RSV and IAV infection, we determined plasma PK of NHC and lung levels of both NHC and the active antiviral agent NHC-TP in mice. Intraperitoneal (i.p.) and oral (p.o.) administration of different dose levels ranging from 10 to 50 mg/kg of body weight (i.p.) and 50 to 500 mg/kg (p.o.) resulted in dose-dependent increases in overall exposure (area under the concentration-time curve [AUC]) to the prodrug (Fig. 3C) and peak plasma concentrations (C_max) after oral dosing exceeding 40 μM (Fig. 3D). Exposure levels corresponded to a dose-dependent oral bioavailability of 36 to 56% (Table 3). The dose dependency of overall prodrug exposure extended to respiratory tissue (Fig. 3E), but peak NHC-TP concentrations in lung saturated above oral dose levels of approximately 150 mg/kg (Fig. 3F). While this C_max plateau suggests an anabolism bottleneck of NHC in mouse respiratory tissue at higher prodrug levels, lung tissue distribution assessment revealed sustained concentrations of active NHC-TP of approximately 10 nmol/g lung tissue for over 8 h after administration at the 150- and 500-mg/kg levels. These results highlight the strong potential of NHC for clinical use against influenza-like diseases.

Based on the PK/PD profiles, we selected as the starting point for all experiments oral doses of 100 mg/kg (below the NHC-TP lung tissue plateau level) and 400 mg/kg (comfortably within the NHC-TP tissue saturation range) and a twice-daily (b.i.d.) dosing regimen for in vivo efficacy testing against RSV and IAV in mice. For testing against RSV, BALB/cJ mice were infected intranasally (i.n.) with 10⁵ PFU each of recombinant RSV-A2-L19F, which is based on the A2 strain but contains an F protein derived from RSV isolate line 19 that increases pathogenicity in the mouse model and better reproduces key features of human RSV disease, such as high lung virus loads, extensive mucus production, and pronounced respiratory distress (44). Treatment was initiated 2 h prior to infection and continued until lung virus titers were determined at 5 days postinfection. At both dose levels, RSV loads were significantly reduced by more than 1 order of magnitude compared to those in vehicle-treated animals (Fig. 4A). However, we did not detect an appreciable difference in progeny titers between the dose groups, presumably reflecting that even at the lower dose used, the NHC-TP concentration in lung tissue approaches saturation levels. Lung histopathology and periodic acid-Schiff (PAS) staining demonstrated complete suppression of excessive mucin production in the 400-mg/kg group and a partial reduction in animals dosed with 100 mg/kg (see Fig. S4 in the supplemental material). Exploratory dosing at 30 and 50 mg/kg failed to significantly lower lung virus loads (Fig. S5), but oral doses of as low as 30 mg/kg were sufficient to completely ameliorate the severe respiratory distress experienced by vehicle-treated animals (Fig. 4B). These results demonstrate the effective inhibition of RSV replication by NHC in vivo at higher dose concentrations. The data furthermore suggest that even minimal pharmaceutical interference with RSV replication can translate into major changes in RSV disease markers in the mouse model.

To test NHC efficacy against seasonal influenza viruses, we infected BALB/cJ mice i.n. with 10³ PFU of mouse-adapted A/Puerto Rico/8/34 (H1N1) (PR8) and assessed lung viral loads at 6 days postinfection as the primary efficacy milestone. In addition, virus-induced hypothermia and selected proinflammatory cytokines were monitored. Based on the experience with anti-RSV efficacy, we focused on the 100- and 400-mg/kg dose levels only for these experiments. After prophylactic dosing as described above, we again observed significant reductions in lung virus loads in NHC- versus vehicle-treated animals (Fig. 4C). All infected animals experienced virus-induced hypothermia, but symptoms were significantly alleviated in mice treated with high-dose NHC (Fig. S6). To put the antiviral impact of NHC in perspective to the SOC, we orally administered NHC (400-mg/kg dose level only in this experiment) or oseltamivir to PR8-infected mice and determined lung virus load profiles for each treatment group (Fig. 4D). Consistent with previous observations (45), oseltamivir delayed viral replication in the first 2 days after infection, but peak lung titers in the oseltamivir group were not significantly different from those in vehicle-treated animals. In contrast, treatment with NHC caused a significant and sustained reduction in the virus load. Reverse transcription-quantitative PCR (RT-qPCR)-based quantitation of the virus-triggered induction of selected proinflammatory cytokines, gamma interferon (IFN-γ) and interleukin-6 (IL-6), revealed a statistically significant decrease in mean induction levels in the 400-mg/kg group (Fig. 4E).

Since PR8 reaches plateau lung titers within 24 h after infection (Fig. 4D), the time window provided by the mouse model to assess the postexposure efficacy of NHC is narrow. To gain first insight into the effect of delayed dosing onset, we initiated treatment at 6 h postinfection, again administering the compounds at the 400-mg/kg dose level only and continuing with a b.i.d. regimen. Lung virus loads were determined at 3 and 6 days postinfection and showed a significant reduction compared to those in vehicle-treated control animals at both time points (Fig. 4F). In keeping with the results obtained after prophylactic dosing, virus-induced hypothermia was significantly reduced in the postexposure NHC-treated animals also (Fig. S6), establishing a comparable benefit of NHC for the management of influenza virus infection independent of whether treatment followed a prophylactic or postexposure regimen.

Having obtained a proof of concept for oral NHC efficacy against PR8 in mice, we employed a highly pathogenic AIV (HPAIV)/mouse model to examine the potential of the compound to strengthen preparedness against highly pathogenic IAVs that constitute a high pandemic threat. Mice were infected i.n. with 6 PFU of A/Vietnam/1203/2004 (H5N1) (A/Vietnam), and treatment was initiated prophylactically at the 400-mg/kg dose level only and continued as described above. Due to the pronounced neuropathogenicity of A/Vietnam in the model (46), both lung and central nervous system (CNS) viral loads were determined at day 6 postinfection in comparison with vehicle- and SOC-treated animals (Fig. 4G). Lung virus titers were again significantly reduced in drug-treated animals compared to the vehicle-only group, and virus was undetectable in the CNS in four out of five animals that had received NHC. This inhibitory effect on A/Vietnam lung virus loads was equivalent to that observed in oseltamivir-treated animals, but NHC more efficiently prevented HPAIV dissemination to the CNS, since three of the five animals in the oseltamivir treatment group presented with detectable virus in brain tissue.

Since influenza virus does not transmit efficiently in mice (47), we employed the well-established guinea pig IAV transmission model (48) to evaluate the effect of NHC on virus spread. Similar to uncomplicated human influenza, virus replication in guinea pigs is limited predominantly to the upper respiratory tract. Source animals were infected i.n. with 10⁴ PFU of A/Netherlands/602/2009 pH1N1 (NL/09) virus and cohoused with uninfected contact guinea pigs starting at 24 h postinfection. Treatment of the source group was initiated prophylactically at the 100- and 400-mg/kg dose levels and continued b.i.d. as described above. Source animals showed a significant, dose-dependent reduction in shedding titers determined from nasal lavage specimens that exceeded 2 orders of magnitude (Fig. 4H). Vehicle-treated animals furthermore efficiently passed the virus to contact guinea pigs. Transmitted virus was first detected at day 4 postinfection, and titers peaked at approximately 10⁶ PFU/ml of nasal wash fluid on day 6 after the initiation of the study. In contrast, low-dose NHC treatment of source guinea pigs delayed transmission by approximately 1 day, and titers remained below 10⁴ PFU/ml of nasal wash fluid throughout the study (Fig. 4I). We noted an even more pronounced inhibitory effect on virus spread when source animals were treated at the 400-mg/kg dose level. Weak transmission was detected in this group only at day 8 and remained limited to two of the four contacts. The other contact animals remained virus-free for the duration of the study. These studies demonstrate the oral efficacy of NHC against major pathogens associated with influenza-like illnesses in different animal models.

Human embryonic kidney cells (293T; ATCC CRL-3216), Madin-Darby canine kidney (MDCK) cells (ATCC CCL-34), HEp-2 cells (ATCC CCL-23), and baby hamster kidney cells (BHK-21; ATCC CCL-10) stably expressing T7 polymerase (BSR-T7/5) were maintained at 37°C and 5% CO₂ in Dulbecco's modified Eagle's medium (DMEM) supplemented with 7.5% fetal bovine serum (FBS). HEp-2 cells are listed in the ICLAC database, version 8.0, of commonly misidentified cell lines, but their use is justified, as these cells are accepted and widely used for studies involving respiratory syncytial virus (RSV). GeneJuice transfection reagent (Invitrogen) was used for all transfection reactions. Normal primary human bronchial tracheal epithelial cells (HBTECs) (purchased from LifeLine Cell Technology [catalog no. LM-0050], passages 1 to 3) were grown in BronchiaLife cell culture medium (LifeLine Cell Technology). Immortalized cell lines used in this study are routinely checked for microbial contamination (in approximately 6-month intervals). HBTECs were tested for microbial contamination on 25 July 2017 by LifeLine Cell Technology. HBTEC passage numbers 2 and 3 were used for this study.

Influenza viruses A/WSN/33 (WSN) (H1N1), WSN-nanoLuc, A/California/7/2009 (H1N1), A/Georgia/M5081/2012 (H1N1), A/Netherlands/602/2009 (H1N1), A/Panama/2007/99 (H3N2), A/Wisconsin/67/2005 (H3N2), A/Aichi/2/68 (H3N2), A/swine/Ohio/sw10-132/2010 (H3N2), B/Yamagata/16/88, and B/Brisbane/60/08 were propagated in MDCK cells for 2 days at 37°C. Influenza viruses A/duck/Alberta/35/76 (H1N1), A/swine/Spain/53207/2004, and A/chicken/Potsdam/178-4/83 (H2N2) were propagated in 10-day-old embryonated chicken eggs for 2 days at 37°C. Influenza viruses A/Vietnam/1203/2004 (H5N1) and A/Anhui/1/2013 (H7N9) were propagated in 9-day-old embryonated chicken eggs at 37°C for 24 h. All experiments using live, highly pathogenic H7N9 avian influenza viruses were reviewed and approved by the institutional biosafety program at the University of Georgia, were conducted in biosafety level 3 enhanced containment and followed guidelines for the use of select agents approved by the CDC. Viruses were titrated by standard plaque assays, hemagglutination assays (80), 50% tissue culture infective dose (TCID₅₀) assays, or TCID₅₀-hemagglutination (TCID₅₀-HA) assays in MDCK cells. For TCID₅₀-HA assays, 10-fold serial dilutions of virus samples in eight replicates each were propagated for 48 h on MDCK cells in a 96-well plate format, followed by the transfer of culture supernatants to suspensions of chicken red blood cells and scoring of individual wells based on hemagglutination activity. Clinical RSV isolates were collected from patient nasal wash specimens in 2010, cultured on primary rhesus monkey kidney (RhMK) cells, and amplified once on HEp-2 cells prior to use in this study. Recombinant RSV and RSV isolates were grown in HEp-2 cells and titrated by a plaque or immunoplaque assay in HEp-2 cells.

Progeny virions were collected from cell culture supernatants (IAV) or released from infected cells through one freeze-thaw cycle (RSV) and subjected to a clearance centrifugation (4,000 rpm for 20 min at 4°C). Virions were diluted in DMEM, purified through a 20% to 60% one-step sucrose gradient in TNE buffer (1 mM Tris [pH 7.2], 100 mM NaCl, 10 mM EDTA) (30,000 rpm for 120 min at 4°C), and harvested from the gradient intersection. Purified virus stocks were stored in aliquots at −80°C.

MDCK cells were injected into barcoded white-walled/clear-bottomed 384-well plates by using a MultiFlo automated dispenser (BioTek) equipped with dual 10-μl peristaltic pump manifolds and incubated for 5 h at 37°C and 5% CO₂. Compounds were added to a final concentration of 5 μM (20 nl/well) by using a high-density pin tool (V&P Scientific), followed by coinfection with recRSV-A2-L19F-fireSMASh (multiplicity of infection [MOI] = 0.1) and recIAV-nanoLuc (MOI = 0.02) at 10 μl/well by use of a MultiFlo dispenser unit and incubation for 48 h at 37°C and 5% CO₂. The final vehicle (dimethyl sulfoxide [DMSO]) concentration was 0.05%. The reporter gene activity was recorded at 48 h postinfection with an H1 synergy multimode plate reader (BioTek), and compounds showing ≥80% inhibition of both RSV and IAV were pursued as hit candidates. The MScreen software package was used for library management and campaign analysis.

For automated dose-response testing, 3-fold serial dilutions were prepared in 96-well plates in three replicates each by using a Nimbus liquid handler (Hamilton). Target cells, as specified, were seeded into white-wall clear-bottom 96-well plates (8 × 10³ cells/well), and the serial dilutions were transferred by using the liquid handler, followed by infection with IAV-WSN-nanoLuc (MOI = 0.02) or RSV-A2-L19F-fireSMASh (MOI = 0.1). Reporter signals were recorded with the H1 synergy plate reader as described above. To determine cell viability, the PrestoBlue substrate (5 μl/well; Life Technologies) was added after 48 h of incubation of compound-exposed uninfected cells at 37°C, and top-read fluorescence (excitation at 560 nm, emission at 590 nm, and instrument gain of 85) was recorded after incubation for 45 min at 37°C by using the H1 synergy plate reader. Raw data for all automated dose-response assays were analyzed according to the formula % inhibition = (X_Sample − X_Min)/(X_Max − X_Min) × 100, with X_Min representing the average of data from four positive-control (1 mg/ml cycloheximide) wells and X_Max representing the average of data from four negative-control (DMSO) wells included on each plate. Four-parameter variable-slope regression modeling was applied to determine 50% active (EC₅₀) and toxic (CC₅₀) concentrations, using the nonlinear regression function in the Prism software package (GraphPad). For manual dose testing of nonreporter viruses, target cells were seeded into 12-well plates (1.5 × 10⁵ cells/well) and, at approximately 90% confluence, infected with the test virus in the presence of serial compound dilutions. Progeny virus titers were determined at 36 to 48 h postinfection, depending on the virus strain analyzed, and viral titers were determined as described above.

Inhibition of human DNA polymerase α was assayed in a 96-well format in wells containing reaction buffer [50 mM Tris-HCl (pH 8.7), 10 mM MgCl₂, 0.4 mg/ml bovine serum albumin (BSA), 1 mM dithiothreitol (DTT), 15% glycerol, 0.05 mM dCTP, 0.05 mM dTTP, 0.05 mM dATP, 10 μCi [α-³²P]dGTP (800 Ci/mmol)], 20 μg activated calf thymus DNA, and NHC-TP in a range of different concentrations. Aphidicolin served as a polymerase inhibitor reference. Reactions were carried out for 30 min at 37°C, followed by transfer to filter plates, precipitation with 10% trichloroacetic acid, and repeated washing with 5% trichloroacetic acid and 95% ethanol. The incorporation of [α-³²P]GTP was measured after filter drying using a Microbeta scintillation counter.

Serum-pathogen-free (SPF) freshly fertilized chicken eggs were purchased from Hy-Line and incubated at 37°C with 55 to 60% humidity for 10 to 11 days. Eggs were candled and disinfected with 70% ethanol, and NHC in sterile phosphate-buffered saline (PBS) was administered to a final concentration of approximately 10 μM directly into the allantoic fluid 2 h prior to infection by using a 22-gauge needle. The average volume of the allantoic fluid was considered to be 50 ml. Eggs were sealed and incubated for 2 h, followed by infection with 10 HA units of A/Swine/Spain/53207/2004 (H1N1). After a 48-h incubation, eggs were cooled to 4°C, and virus was harvested from the allantoic fluid and titrated by using standard HA assays and turkey red blood cells.

Reporter activities were determined in the presence of 3-fold serial dilutions of NHC starting from 20 μM for RSV and 100 μM for IAV; treatment was initiated immediately after transfection. Luciferase activities in cell lysates were measured with a Synergy H1 microplate reader (BioTek) in the top-count mode using a Dual-Glo luciferase assay system (Promega). Inhibitory concentrations were calculated through four-parameter variable-slope regression modeling.

HEp-2 cells were infected with RSV-A2-fireSMASh at an MOI of 0.1, and MDCK cells were infected with IAV WSN-nanoLuc at an MOI of 0.05. At the specified time points relative to infection, NHC, GRP-71271, AL-8176, or CL-309623 was added to the culture media to a final concentration of 10 μM. T-705 was added to a final concentration of 50 μM, and equivalent volumes of DMSO served as vehicle controls. Reporter gene expression was measured at 24 h (IAV-WSN-nanoLuc) or 48 h (RSV-A2-fireSMASh) postinfection, and the obtained values were expressed relative to the values for the vehicle-treated samples.

HEp-2 cells were infected with RSV-A2-fireSMASh at an MOI of 0.1, and MDCK cells were infected with IAV-WSN-nanoLuc at an MOI of 0.05. At the time of infection, NHC was added to a final concentration of 10 μM NHC alone or in combination with 0.1 to 300 μM exogenous nucleosides (Sigma-Aldrich). Volume equivalents of DMSO served as vehicle controls. Reporter gene expression was quantified at 24 h (IAV-WSN-nanoLuc) or 48 h (RSV-A2-fireSMASh) postinfection. Values are expressed relative to the values for the vehicle-treated samples.

RSV large polymerase subunit (L) and phosphoprotein (P) complexes were expressed from a baculovirus vector and purified by affinity chromatography: L-P complexes were eluted from a Ni-nitrilotriacetic acid (NTA) column with 250 mM imidazole in a solution containing 50 mM NaH₂PO₄ (pH 7.5), 150 mM NaCl, and 0.5% NP-40, followed by dialysis against a solution containing 150 mM NaCl, 20 mM Tris-HCl (pH 7.4), 1 mM DTT, and 10% glycerol. L-P hetero-oligomers were mixed in Mg²⁺ buffer with a 25-mer RNA oligonucleotide template containing essential RSV promoter sequences and rNTPs (ribonucleotide triphosphate), including 0.07 μM [α-³²P]UTP tracer but lacking CTP. NHC, NHC-TP, or CTP was added as specified. Predominant initiation at the +3 position results in up to 23-mer radiolabeled amplicons, which were subjected to denaturing gel electrophoresis, followed by autoradiography.

HEp-2 cells were infected with RSV-A2-mKate at an MOI of 0.1, and MDCK cells were infected with IAV-WSN at an MOI of 0.05, followed by growth in the presence of 10 μM NHC for 24 h; equivalent volumes of DMSO were used as vehicle controls. Total RNA was extracted by using the ZR viral RNA kit (Zymo Research), and cDNA of the viral message was synthesized with SuperScript III reverse transcriptase (Thermo Scientific) and oligo(dT) primers. The PB1 segment of IAV-WSN or an ∼1,500-nucleotide fragment of the RSV-A2-mKate L open reading frame (ORF) was amplified by PCR and subcloned into the pUC19 vector. For each virus and treatment condition, at least 10 distinct subclones (equaling in aggregate at least approximately 7,500 nucleotides each) were Sanger sequenced by using universal M13 primers. Data were analyzed with the Sequencer package, and mutation frequencies were expressed per 5,000 nucleotides for IAV-WSN and per 10,000 nucleotides for RSV-A2. Fisher's exact test was used for statistical analyses.

Both dose escalation and fixed-dose adaptation strategies were applied. For dose escalation, MDCK cells were infected with IAV-WSN at an MOI of 0.01, and HEp-2 cells were infected with RSV-A2-mKate at an MOI of 0.1. The viruses were passaged in the presence of increasing concentrations of NHC, starting at 250 nM. Dose concentrations were doubled at virus passage up to a final concentration of 10 μM. Six independent passage lines each per target virus were advanced simultaneously for a total of 10 passages, each entailing virus harvest from infected cells, dilution, and reinfection of fresh cell populations in the presence of the compound or vehicle. Virus titers declined significantly toward the end of the series, and no resistant variants emerged. For fixed-dose adaptation, MDCK cells were infected with IAV-WSN or B/Brisbane/60/08 at an MOI of 0.01, and HEp-2 cells were infected with RSV-A2-mKate at an MOI of 0.1. The viruses were passaged in the presence of EC₉₉ equivalents of NHC for the respective target viruses, as described above, for a total of 10 passages. Progeny virus titers again declined toward the end of the cycle, and no drug-resistant virus population could be cultivated.

To screen for cellular uptake and anabolism of NHC to NHC-TP, HBTECs were grown in the presence of 20 μM NHC for 0, 1, 2, 3, 4, 6, 16, and 24 h. Cells were washed with PBS and lysed with 70% methanol, and clarified samples were stored at −20°C until analysis. To determine the stability of NHC-TP and other anabolites, HBTECs were grown in the presence of 20 μM NHC for 24 h, medium was changed to drug-free BronchiaLife cell culture medium, and cells were incubated for an additional 0, 0.5, 1, 2, 3, and 6 h. Metabolites were extracted with 70% methanol as described above, and samples were kept at −20°C until analysis. NHC anabolites were quantitated by using internal standard-based LC-MS/MS on an Agilent 1200 system (Agilent Technologies) equipped with a SeQuant ZIC-pHILIC column (The Nest Group). Mass spectrometry analysis was performed on a QTrap 5500 mass spectrometer (AB Sciex) using negative-mode electrospray ionization (ESI) in the multiple-reaction-monitoring (MRM) mode. Data analysis was performed by using Analyst software (AB Sciex).

Female CD-1 mice (6 to 8 weeks of age) distributed randomly into groups were dosed p.o. with NHC in 240 mM citrate buffer, followed by blood and lung tissue sampling. Plasma was purified from heparinized blood, and tissue samples were snap-frozen prior to 70% acetonitrile extraction. Drug concentrations were determined by using ¹³C₅-labeled internal standards for NHC and NHC-TP. Mass spectrometry was performed as detailed above. For calibration, standard curves were prepared in blank plasma (concentration range, 25 to 30,000 ng/ml) and blank tissue lysate (concentration range, 1.49 to 1,490 ng/ml). Quality control samples of 30, 500, and 900 ng/ml in blank plasma were analyzed at the beginning of the analysis of each sample set. Calibration in each matrix showed linearity with R² values of >0.99.

Female BALB/cJ mice (5 to 6 weeks of age) were obtained from the Jackson Laboratory or Envigo, rested for 1 week, assigned to groups randomly, anesthetized by intraperitoneal injection of a ketamine-xylazine solution, and infected i.n. with 1 × 10⁵ PFU of recRSV-A2-L19F. NHC was administered orally in 240 mM citrate buffer or an equivalent volume of the vehicle at 2 h preinfection, and dosing continued b.i.d. for up to 8 days. For virus load titration, lungs were extracted and homogenized, homogenates were serially diluted and transferred to HEp-2 cells, and cells were overlaid at 1 h postinfection with minimum essential medium (MEM) containing 10% FBS, penicillin G, streptomycin sulfate, an amphotericin B solution, and 0.75% methylcellulose. At 6 days postinfection, cells were fixed with methanol, and plaques were visualized by immunodetection. To quantify mucin expression, mice were euthanized at 8 days postinfection, and heart-lung tissue was harvested and fixed in 10% formalin. Lung tissue sections embedded in paraffin blocks were stained with periodic acid-Schiff (PAS) stain, and slides were digitally scanned by using a Zeiss Mirax Midi microscope (Carl Zeiss Microimaging). To determine respiratory distress noninvasively, a rodent pulse oximeter (MouseOx; Starr Life Sciences Corp., Oakmont, PA) was applied to the mouse's thigh, and arterial O₂ saturation, heart rate, pulse rate, pulse distension, and breath distension were measured every 0.1 s for a 1- to 5-min overall period. The mean breath distension for each treatment group was calculated based on mean values of all measurements for each animal in which all target parameters were present.

Female BALB/cJ mice (6 to 8 weeks of age) were received from Envigo and housed in ABSL-2 (animal biosafety level; for infection with IAV-PR8) or ABSL-3 (for infection with HPAIV) facilities in HEPA-filtered microisolator caging. Mice were rested for 1 week, weighed, assigned to groups randomly, and infected with 10³ PFU of IAV-PR8 or 6 PFU of A/Vietnam/1203/04 (H5N1), as specified, in PBS. Treatment was initiated at 2 h preinfection (prophylactic dosing) or 6 h postinfection (therapeutic dosing) and continued for up to 6 days b.i.d. Compounds or equivalent volumes of the vehicle were administered orally in a 240 mM citrate buffer formulation. Animal clinical signs were tracked daily, and animals were euthanized at the indicated time points or when humane endpoints were reached. Lung and, where indicated, brain tissue were removed, homogenized, and clarified by centrifugation, and aliquots were frozen at −80°C, as outlined above, until virus titration was performed.

Relative IFN-γ and IL-6 induction levels present in mouse lung tissue were determined by semiquantitative real-time PCR analysis. Total RNA was extracted from lung tissue 3 days after infection of animals with IAV-PR8 or mock infection (representing day 0). Infected animals were treated orally with NHC or equivalent volumes of the vehicle (citrate buffer). RNA was reverse transcribed with SuperScript III reverse transcriptase, and the resulting cDNAs were subjected to real-time PCR using Fast SYBR green master mix (Applied Biosystems). Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) mRNA served as an internal control, and mRNA induction levels were normalized to the average of the values obtained for mock-infected animals. Each biological repeat was determined in duplicate, and relative changes in transcription levels were calculated according to fold changes determined by the 2^−ΔΔC_T method.

Female Hartley strain guinea pigs weighing 250 to 300 g were obtained from Charles River Laboratories. The animals were assigned to groups randomly, and prior to intranasal inoculation, nasal lavage, or CO₂ euthanasia, the guinea pigs were sedated with a mixture of ketamine and xylazine (30 mg/kg of body weight and 4 mg/kg, respectively). Inoculation and nasal lavage were performed with PBS as the diluent/collection fluid in each case. Oral treatment of infected donor animals with NHC or the vehicle (240 mM citrate buffer with Ora-Sweet [Paddock Laboratories]) was initiated at 2 h preinfection and continued b.i.d. until the end of day 3 postinfection. Following inoculation and recovery from sedation, donor guinea pigs were housed in Caron 6040 environmental chambers (fitted with the optional dryer package) set to 10°C and 20% relative humidity. Twenty-four hours after inoculation of the donor animals, exposed guinea pigs were introduced into the donor animal cages. Conditions of 10°C and 20% relative humidity were maintained throughout the exposure period, which ended on day 8 postinoculation. Titers of virus shedding in nasal lavage fluids of source and contact animals were determined through plaque assays.

To assess experimental variation and the statistical significance of differences between sample means, one-way or two-way analysis of variance (ANOVA) was carried out in combination with Tukey's, Dunnett's, or Sidak's post hoc test, as specified in the figure legends, using the Prism software package (GraphPad). Fisher's exact test was used for statistical analyses of mutation frequencies. Results for individual biological replicates are shown for all in vivo efficacy experiments. When appropriate, experimental uncertainties are identified by error bars, representing standard deviations (SD).

All animal work was performed in compliance with the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health (81). Mouse work at Georgia State University was approved by the GSU Institutional Animal Use and Care Committee (IACUC) under protocol A17019; mouse and guinea pig work at Emory University was approved by the Emory IACUC under protocol no. DAR-2003089-ENTRPR-N and DAR-2002738-ELMNTS-A, respectively; and mouse studies with HPAIV at the University of Georgia were approved by the UGA IACUC under protocol no. A2017 05-009.