Cells and reagents

Huh7 and Huh7.5 cells (a gift from Marco Binder lab; Heidelberg University; Germany) were grown in DMEM, high glucose, with GlutaMAX (Gibco; France) supplemented with 10% fetal calf serum (FCS; Biosera; France) and 100 IU/mL penicillin/streptomycin (Gibco; France). Cells were grown at 37°C and 5% CO₂. 6-Aminonicotinamide (6-AN), Nicotinamide (NAM), Nicotinamide Riboside (NR), FK866, STF-118804 were obtained from MedChemExpress (CliniSciences; France).

Human hepatocyte derived from chimeric mice (HepaSH) cultures

Freshly purified cells were obtained from Biopredic International, seeded upon reception in glycogen-coated well plates as recommended by the supplier and cultured for 4 days in William-E medium (Gibco 12551032) containing 100μg/mL Penicillin/Streptomycin, 2mM L-Glutamine, 1% Non-Essential Amino Acids, Hydrocortisone, 5μg/mL Insulin, 2ng/mL EGF, 5% SVF, 2% DMSO and 10mM HEPES, before infection.

In vitro RNA transcription

Plasmids containing the subgenomic replicon HCV-JFH1-R2A or HCV-Jc1 full length were obtained from Prof. Ralf Bartenschlager (Department of Molecular Virology, Heidelberg, Germany) and RNA synthetized as previously described [25]. Briefly, plasmids containing the subgenomic replicon and the plasmid containing Jc1 strain were linearized with XbaI and AseI enzymes, respectively. Linearized DNA was purified by phenol/chloroform extraction, precipitated with isopropanol and resuspended in RNase-free water. In vitro transcription was performed with T7 RNA polymerase (T7 Express Large Scale RNA Production System, Promega). RNA was extracted with acidic phenol and chloroform, precipitated with isopropanol and resuspended in RNase-free water.

Electroporation of in vitro-transcribed RNA

HCV-RNA was electroporated into Huh7 cells as previously described [26]. Briefly, cells were trypsinized and washed twice with PBS. 4x10⁶ cells were resuspended in 400 µL of cytomix (120 mM KCl, 150 mM CaCl₂, 10 mM K₂HPO₄/KH₂PO₄, 25 mM HEPES, 2 mM EDTA, 5 mM MgCl₂, pH 7.6) and electroporated in presence of 5 µg of HCV-RNA with the following conditions: 975 µF and 270 V with a Gene Pulser system (Bio-Rad) in a cuvette with a gap width of 0.4 cm (Bio-Rad). Immediately after electroporation, cells were resuspended in 20 mL of complete medium and seeded in culture plates suitable for subsequent experiments.

Luciferase assay

When electroporated with subgenomic luciferase-replicon, cells were seeded in white opaque 96-well microplates (200 µL of cell suspension per well). 72 h post-electroporation, 100 µL of medium were removed and luciferase activity was assessed by addition of 50 µL of Luciferase Assay System mix (Promega). After 10 min incubation, luminescence was quantified using a Tristar5 luminometer (Berthold, Freiburg, Germany) for 1s, and expressed as relative light units (RLU). Luciferase activity from cells harvested 4 h after electroporation was used to determine transfection efficiency.

Virus stock production and titration

Jc1 and Jc1-E2-FLAG virus stocks were generated as previously described [6]. Briefly, master stocks were generated by electroporation of Huh7.5 cells with in vitro transcripts of the full-length HCV genomes as described above. Immediately after electroporation cells were resuspended in 12 mL of complete medium and seeded in two 10 cm diameter dishes. 6 h post-electroporation medium was replaced with fresh media. 72 h post-electroporation supernatants were harvested, clarified through a 0.45 µm filter and stocked at -80°C before titration. A second harvesting of the culture was performed 72 h later and treated identically. For titration, Huh7.5 cells were seeded in 48-well plates at 2x10⁴ cells/well 24 h before their infection with serial dilutions of virus-containing culture supernatants. At 72 h post-infection, cells were fixed with ice-cold ethanol for 30 min before staining with anti-NS5A antibody (clone 9E10, Sigma-Aldrich) combined with anti-mouse IgG HRP-conjugate antibody.

dsRNA foci staining and quantification

Huh7 cells were seeded in ibidi µ-Slide 8-well chambers 24 h before infection (MOI 1) and treated or not with 100 µM 6-AN ± 500 µM NAM for 72 h. After removing the culture supernatant, cells were washed with PBS and fixed for 15 min at room temperature with 4% paraformaldehyde. They were permeabilized for 30 min at 4 °C using PBS + 0.2 M glycine + 1% Triton X-100, then blocked for 30 min at room temperature with PBS + 0.2 M glycine + 0.5% BSA. After a PBS + 0.2 M glycine wash, cells were incubated for 1 h at room temperature with anti-dsRNA antibody (1:100, clone J2, Jena Bioscience RNT-SCI-10010200) diluted in PBS + 0.2 M glycine. Cells were washed twice with PBS + 0.2 M glycine, incubated with Alexa Fluor 488-conjugated goat anti-mouse IgG (1:2000, Invitrogen A-11029) for 1 h at room temperature, washed again, stained for 5 min with 5 µM Hoechst in PBS, washed, and imaged on a Yokogawa HCS CQ1 confocal system. For dsRNA foci quantification, Z-stacks (11 µm total, 1.8 µm steps) were acquired in random fields (14 fields per condition from two independent slides) using a UPLSAPO40X2 objective (Olympus). Acquisition settings were 100% excitation and 750 ms exposure for the 488 nm channel, and 50% excitation and 500 ms exposure for the 405 nm channel. Images were analyzed in the CQ1 software using the “Dots in Cell body” template in slice mode to automatically segment dsRNA foci. Nuclei segmentation used the following steps: ThresholdToZero (400), Threshold (450), ErosionCircle (0.7 µm), DivideEachRegion, SizeFilter (450–20000 µm³), FillUp, ExcludeEdge. Cell bodies were segmented from green-channel autofluorescence using: MeanImage (5 µm mask), Threshold (150), DivideEachRegion, ExpandRegion3D (nucleus-guided), SizeFilter (20–10,000,000 µm³). dsRNA dots were segmented using: Gaussian filter (0.1 µm mask), Threshold (510), FindMaximumImage (remove size <0.3 µm), DivideEachRegion, ExpandRegion3D (Gaussian input; Threshold 450, 5 iterations), SizeFilter (0.1–1,000,000,000), and DilateRegion (0.1 µm). Low-magnification single-plane images were acquired using a UPLSAPO4X objective (Olympus) with the 488 nm laser (100% excitation, 2000 ms exposure), and stitched with the microscope software to generate mosaics covering nearly the entire well area.

Iodixanol density gradient

10 mL of supernatant from HCV-infected cells were concentrated 10 times on a 100,000 MWCO PES membrane Vivaspin (Sartorius). 1 ml of the resulting concentrated supernatant was layered on the top of a 0–30% continuous preformed iodixanol gradient (Optiprep; Merck). Gradients were centrifuged for 16 h at 32,000 rpm in a SW41 swinging rotor at 4°C using an Optima L-90 K Beckmann centrifuge. Fifteen fractions of 750 µl were collected from the top and were analyzed for virus titer and viral RNA content.

Apolipoprotein B (ApoB), Apolipoprotein E (ApoE) and triglycerides quantifications

ApoB and ApoE concentrations in medium and gradients fractions were determined by ELISA as previously described [27], using the following antibodies: anti-human ApoB mAbs (LDL 20/17), anti-human ApoB mAb (LDL 11) biotin, anti-human ApoE mAb (E276), anti-human ApoE mAb (E887) biotin (MABTECH, Sweden). Triglycerides (TG) were quantified using a specific enzymatic assay (Millipore Sigma-Aldrich).

Intracellular lipid droplet staining

Supernatant of cultures were removed and cells were washed with PBS before fixation with a 4% formaldehyde solution during 15 min at RT. Cells were then washed twice with deionized water before a 5 min incubation with isopropanol 60%. Isopropanol was removed and Oil-Red-O solution (Millipore Sigma-Aldrich) added on cells for 15 min at RT. The Oil-Red-O working solution was prepared extemporaneously by mixing 6 ml of a 0.3% Oil-Red-O w/v isopropanol solution with 4 ml of water. Cells were then extensively washed with water to remove the exceeding dye before nucleus counterstaining with NucBlue Fixed Cell Stain ReadyProbes reagent (ThermoFisher Scientific) and observation with an inverted confocal microscope Zeiss LSM800 (x63 objective). The intracellular lipid droplet contents were analyzed using the IMARIS software.

Quantification of viral RNA and gene expression analysis by RT-qPCR

RNAs were extracted either from cells or supernatants with TRI Reagent (ThermoFischer Scientific), reverse transcribed with High-Capacity RNA-to-cDNA kit (Applied Biosystems), and HCV RNAs were quantified using SYBR Green PCR kit QuantiNova (Qiagen) on an Applied StepOne Real-Time PCR apparatus using the forward primer 5’-TCTGCGGAACCGGTGAGTA-3’ and the reverse primer 5’-TCAGGCAGTACCACAAGGC-3’. For quantification of cellular gene expression, the specific following primers were used: NAMPT (forward primer 5’-GGTTCTGGTGGAGGTTTGCT-3’ and the reverse primer 5’-CTGCTGGCGTCCTATGTAAAGA-3’), NAPRT (forward primer 5’-AGCCACGAATGAAGCTGACCGA-3’ and the reverse primer 5’-CACTGGCTCTTCTGCTAACTGC-3’) and QPRT (forward primer 5’- CTCCAGTGCCCAAAATCCAC-3’ and the reverse primer 5’- CTGACCCTAAAGATGTGTGACC-3’). Expression of RPL13A was used as housekeeping gene (forward primer 5’-AAAAGCGGATGGTGGTTCCT-3’ and the reverse primer 5’-GCTGTCACTGCCTGGTACTT-3’).

Metabolomic analysis

Metabolites extractions and metabolomic analysis were performed on four independent biological replicates. For each condition 2x10⁶ Huh7 cells were seeded in 10 mm dishes and grown for 24 h in 10 mL of culture medium. Supernatant was removed and replaced by 8 mL of fresh culture medium containing or not 6-AN at the final concentration of 100 µM for 48h. Culture medium was removed and metabolites were immediately extracted from the cell monolayer in 2 mL of ice-cold (-20˚ C) 80% MS-grade methanol (Sigma-Aldrich) diluted with sterile pyrogen free water Otec (Aguettant; France). Cell extracts were immediately transferred into tubes and vortexed for 1 min, samples were stored at -80°C. Sample analysis was carried out by MS-Omics (Denmark) as follows. Samples were dried under nitrogen flow and reconstituted in 140 μl MQW. After reconstitution samples were filtered and additionally diluted 10 times in eluent A for semi-polar metabolites analysis and 5 times in eluent A for polar metabolites analysis. The analyses were carried out using a Thermo Scientific Vanquish LC coupled with a high-resolution quadrupole-orbitrap mass spectrometer (Q Exactive HF Hybrid Quadrupole-Orbitrap, Thermo Fisher Scientific). An electrospray ionization interface was used as ionization source. Analysis was performed in negative and positive ionization mode. For semi-polar metabolites, the UPLC was performed using a slightly modified version of the protocol described by C.E. Doneanu et al. [28]. For polar metabolites, the UPLC was performed using a slightly modified version of the protocol described by Hsiao et al. 2018 [29]. Peak areas were extracted using Compound Discoverer 3.2 (Thermo Fisher Scientific) and Skyline [30]. Identification of compounds were performed at four levels; Level 1: identification by retention times (compared against in-house authentic standards), accurate mass (with an accepted deviation of 3ppm), and MS/MS spectra, Level 2a: identification by retention times (compared against in-house authentic standards), accurate mass (with an accepted deviation of 3ppm). Level 2b: identification by accurate mass (with an accepted deviation of 3ppm), and MS/MS spectra, Level 3: identification by accurate mass alone (with an accepted deviation of 3ppm). A total of 1,768 compounds were detected in the samples. Hereof, were 522 annotated on level 3, 55 on level 2b, 68 on level 2a, and 83 on level 1. Compounds analyzed in this study are from Level 1 or 2a. The results from the analyses are presented as log2 values of the ratio between the average of 6-AN-treated vs control samples. Statistical significance was determined with a Student’s t-test for paired samples (two-tailed). MetaboAnalyst (version 6.0) was used for metabolite enrichment analyses.

Real-time monitoring of metabolic phenotype with a Seahorse XF analyzer

Cells were seeded in Seahorse XF 96-well microplates (Agilent), coated with Poly-L-Lysine 0.01% (Sigma-Aldrich), at 8x10³ cells/well in 200 µL of DMEM medium (Gibco) supplemented with 10% FCS (Biosera), 2 mM L-glutamine, 100 U/mL penicillin/streptomycin, treated with 100 µM 6-AN, 500 µM NAM or H₂O alone and incubated at 37°C and 5% CO₂ for 72 h. The assay was initiated by replacing growth medium with prewarmed Seahorse assay medium (XF DMEM pH7.4; 103575–100; Agilent) supplemented with glucose (10 mM) and L-glutamine (2 mM). Cells were washed with 200 µL assay medium and incubated at 37°C for 1 h without CO₂. The medium was replaced by 180 µL of prewarmed assay medium prior measurement of oxygen consumption rate (OCR) and extracellular acidification rate (ECAR) with the Seahorse XFe24 analyzer using the XF Cell Mito Stress Test (Agilent) or XF Glycolytic Rate assay (Agilent). The number of cells was determined at the end of the run after Hoechst 33342 staining and cell counting using Cytation 1 cell imaging reader (Agilent BioTek). Results were normalized by cell count and analyzed using the Seahorse Wave software.

Quantification of cell proliferation

Cellular amounts in culture wells were determined with the CellTiter-Glo Luminescent Cell Viability Assay (G7570; Promega; France), which relies on ATP quantification as a proxy for cell number in a well, or by staining cell nuclei with Hoechst followed by quantification of the fluorescent signal. Huh7 cells were seeded at 8x10³ cells/well in quintuplicates in white 96-well plates for the CellTiter-Glo detection or black plates with clear bottoms for fluorescence-based cell quantification. Cells were incubated for 96 h in 200 µL of culture medium with DMSO alone or indicated drugs (6-AN at 100 µM or NAM at 500 µM). For ATP quantification in culture wells, 100 µL of culture medium were removed first, and 50 µL of CellTiter-Glo reagent (Promega) were added in each well. After 10 min of incubation at RT, luminescence was quantified with Tristar 5 Multimode reader (Berthold; Germany). For staining cell nuclei, Hoechst 33342 (Thermo Scientific) was added to each well at 40 µM and incubated for 30 min at 37°C. Wells were washed with PBS and fluorescence (excitation 350 nm/ emission 461 nm) was measured with a Tristar 5 Multimode reader.

Cell viability

Cytotoxicity was evaluated by CellTox Green cytotoxicity assay (Promega). 8x10³ cells/well were seeded in black 96-well plates and cultured for 72 h in presence of 6-AN (100 µM), 6-AN (100 µM) plus NAM (500 µM), or DMSO alone. Then, cells were labeled with 1/500 dilution of CellTox green dye during 15 min at room temperature shielded from ambient light, before fluorescence measurement at 490 nm (excitation) and 520 nm (emission), using a TECAN M200 microplate reader. Under the same culture conditions, a series of wells were treated with 10% Triton X-100 one hour before the cytotoxicity assay to determine the maximal fluorescent signal corresponding to 100% toxicity for each condition.

Double membrane vesicles assessment by electron microscopy

5x10⁵ of Huh7 cells were seeded on T75 flasks 24 h before infection with HCV Jc1 at a MOI of 1. 6 h after inoculation, culture medium was removed and replaced by fresh medium containing 6-AN (100 µM), 6-AN (100 µM) plus NAM (500 µM), or DMSO alone. After 72 h of culture, cells were harvested, washed twice with PBS before being fixed by incubation for 48h in 4% paraformaldehyde and 1% glutaraldehyde in 0.1 M phosphate buffer pH 7.2 and then washed twice with PBS. Cells were then incubated with 2% osmium tetroxide (Agar Scientific, Stansted, UK) for 1 h. Fixed cells were then fully dehydrated in increasing concentration of ethanol (70%, 90% and 100%) and then propylene oxide (100%). Fixed cells were then impregnated with a 1:1 mixture of propylene oxide/Epon resin (Sigma) and incubated overnight in pure resin. Cells were then embedded in Epon resin (Sigma) and left to polymerize for 48 h at 60°C. Ultrathin sections (80 nm) were cut with an EM UC7 ultramicrotome (Leica Microsystems, Wetzlar, Germany). Contrast staining was performed with 2% uranyl acetate (Agar Scientific) and 5% lead citrate (Sigma), and the samples were then observed with a JEOL JEM-1011 (Tokyo, Japan) transmission electron microscope operated at 100 kV and equipped with an Ametek-GATAN RIO9 CMOS camera.

Bioinformatics and statistical analyses: Significance values were calculated by applying tests indicated in the Figure legends using the GraphPad Prism 10 software (GraphPad Software, USA). P values under 0.05 were considered statistically significant and are indicated within the Figures.

HCV replication and DMV formation depend on NAD(H)

In mammalian cells, NAD(H) and NADP(H) are synthesized de novo either from tryptophan via the kynurenine pathway or from nicotinic acid (NA) via the Preiss-Handler pathway (Fig 1A). NAD(H) is also produced via the salvage pathway using nicotinamide (NAM), a byproduct of NAD(H)-consuming reactions (Fig 1A). 6-aminonicotinamide (6-AN) is an analogue of NAM which competes with NAM as a substrate for nicotinamide phosphoribosyl transferase (NAMPT), the first enzyme of the salvage pathway. The use of 6-AN leads to the formation of 6-aminonicotinamide adenine dinucleotide (6-ANAD) and 6-ANAD phosphate (6-ANADP), which are non-reducible analogues of NAD(H) and NADP(H). The consequence is an inhibition of enzymes using NAD(H) and NADP(H) in redox reactions. Since 6-AN was previously reported to inhibit the replication of vaccinia virus, ZIKV, HBV, and DENV [8,14,15,31], we sought to determine whether this compound could also inhibit HCV replication. Huh7 cells were transfected with a subgenomic replicon (strain JFH1) expressing luciferase as a reporter instead of structural proteins, and then treated with increasing doses of 6-AN for 72 h. As shown in Fig 1B, HCV replication was inhibited in a dose-dependent manner, by 80% at 100 µM and 90% at 500 µM. To assess the viability of cells treated with 6-AN for 72 h, we measured intracellular ATP levels, performed cell count under treatment and conducted a cytotoxicity assay (S1 Fig). As described elsewhere [14], we observed that 6-AN treatment reduced intracellular ATP level, indicating a decreased metabolic activity (S1A Fig). We also observed a decreased cell count by ~30% at 100 µM of 6-AN (S1B Fig), but no impact on cellular integrity in a membrane integrity assay using CellTox Green dye (S1C Fig) showing a delay in cell growth. Thus, 6-AN effectively inhibits HCV replication in Huh7 cells, likely by limiting cellular metabolic activity and in particular ATP availability among other metabolites, while not affecting cellular membrane integrity. This shows that, without killing cells, 6-AN induces a metabolic slowdown limiting viral replication. Interestingly, viral replication was restored in a dose-dependent manner when the culture medium was supplemented with an excess of NAM or nicotinamide riboside (NR), two NAD(H) precursors (Fig 1C and 1D).

Download:

• PNG
  larger image
• TIFF
  original image

Fig 1. HCV replication depends on NAD(H).

(A) Schematic representation of NAD(H), NADP(H) biosynthesis pathways and targets of drugs used in the study. (B-D) Huh7 cells were electroporated with subgenomic replicon and cultured for 72 h with the indicated treatment. Then, Renilla luciferase activity was measured in cell homogenates and viral replication normalized to non-treated control (0 or Ctl.). Are presented means ± SEM. (B) Replicon-electroporated Huh7 cells were cultured in presence of increasing concentrations of 6-AN (0, 50, 100 and 500 µM), n = 3, one-way ANOVA, Bonferroni-Sidák adjusted p-value for multiple comparison. (C) Cells were cultured with or without 100 µM of 6-AN and increasing doses of NAM, n = 4, one-way ANOVA, Bonferroni-Sidák adjusted p-value for multiple comparison to 6-AN treatment. (D) Cells were cultured with or without 100 µM 6-AN and increasing doses of nicotinamide riboside (NR), n = 3, one sample t-test, Bonferroni-Sidák adjusted p-value for multiple comparison to condition control. Are presented means ± SEM (n.s. non significative, *p < 0.01, **p < 0.01, ***p < 0.001). (E) Huh7 cells were infected and treated or not with 6-AN ± NAM, stained for dsRNA and imaged by confocal microscopy. Representative Z-stack maximum intensity projections of cells stained for dsRNA. Closeups and whole fields of view (with stronger brightness and contrast adjustment to facilitate the visualization of foci-containing cells) are shown. (F) Corresponding quantifications of the proportion of infected cells (dsRNA foci-containing cells), dsRNA foci abundance per cell and dsRNA foci volume. Data are presented as mean ± SD, and were analyzed using one way ANOVA, *p < 0.05, ns none significative. (G) Mosaic microscopy images showing wide field of views of the cell monolayers stained for dsRNA. Data aggregate 2 independent experiments. (H) Huh7 cells were infected with HCV at MOI = 1 and cultured with or without 100 µM 6-AN ± 500 µM NAM and prepared for observation by transmission electron microscopy. Red arrows indicate DMVs.

https://doi.org/10.1371/journal.ppat.1014165.g001

As a proxy of viral replication, we investigated the impact of 6-AN on viral replication complexes. We thus infected Huh7 cells with HCV (Jc1 strain) and treated cells for 72 h with either 100 µM 6-AN or 6-AN supplemented with 500 µM NAM. Replication foci were labeled using an anti-dsRNA antibody and visualized with an Alexa Fluor 488–conjugated secondary antibody, while nuclei were counterstained with Hoechst (Fig 1E–1G). Confocal microscopy analysis revealed that the percentage of cells exhibiting detectable replication complexes was significantly reduced following 6-AN treatment, indicating impaired viral spread upon nicotinamide metabolism inhibition (Fig 1F). This effect was reversed by NAM supplementation, which restored the proportion of cells displaying replication foci. Interestingly, the number of replication foci per infected cell was not markedly altered by 6-AN treatment. However, analysis of individual foci volumes showed that they were, on average, smaller in 6-AN–treated cells. These results suggest that while the number of replication complexes per cell remains largely unchanged, their size, and potentially their functional activity, is reduced when nicotinamide metabolism is inhibited. Such a reduction in replication complex size likely contributes to the overall decrease in viral replication observed under these conditions. HCV replicates in membranous structures characterized by double-membrane vesicles (DMVs) that are induced by viral replication. We thus specifically analysed the formation of DMVs by electron microscopy in Huh7 cells infected with HCV. Fig 1H shows representative pictures of infected cells 72 h post-infection, where the presence of DMV structures is indicated by the red arrows. While DMVs are visible in control condition, 6-AN treatment strongly inhibits the formation of these structures. Interestingly, the addition of NAM restored the formation of DMVs. Thus, the inhibition of NAD(H) metabolism by 6-AN prevents the formation of DMVs which are required for HCV replication.

HCV replication depends on the NAD(H) salvage pathway

We then assessed the contribution of the different pathways involved in the synthesis of NAD(H) to HCV replication (Fig 2A). We first verified whether HCV infection modulated the expression of enzymes involved in this pathways. We quantified NAMPT, NAPRT, and QPRT by RT-qPCR in Huh7 cells after 3 days of infection. NAMPT is the rate-limiting enzyme of the nicotinamide salvage pathway, and NAPRT and QPRT are required for the Preiss-Handler and de novo NAD(H) synthesis pathways, respectively (Fig 1A). None of these enzymes was significantly modulated upon infection (S2A–S2C Fig). These results suggest that sensitivity to inhibitors of the nicotinamide metabolic pathway does not appear to be affected by viral infection. Furthermore, these observations are in agreement with transcriptomic data obtained from biopsies of patients infected or not by HCV [32] (GSE84346)), and showing no significant difference in the expression of these three enzymes, in HCV-infected patients compared to non-infected patients (S2D–S2F Fig). Altogether, these observations suggest that sensitivity to inhibitors of the nicotinamide metabolic pathway does not appear to be affected by viral infection. Because the salvage pathway depends on the reaction catalyzed by NAMPT (Fig 1A), we first evaluated a direct inhibitor of this enzyme, namely FK866. As shown in Fig 2A, FK866 suppressed viral replication and similar results were obtained with STF-118804, another inhibitor of NAMPT (Fig 1B). We quantified total NAD in cells treated with the antimetabolite 6-AN or the NAMPT inhibitor FK866. We used a bioluminescent assay measuring total NAD⁺ and NADH with high sensitivity through an enzymatic cycling system in which NAD ⁺ is converted to NADH, enabling reductase-dependent generation of luciferin and subsequent luminescence proportional to total NAD cellular content. Intracellular NAD-dependent luminescence decreased by 24% in 6-AN–treated cells and by 97% in FK866-treated cells (S3 Fig). In both cases, NAM supplementation restored NAD levels. These results are consistent with the distinct mechanisms of action of the two compounds. FK-866 is a potent inhibitor of NAMPT, the rate-limiting enzyme of the nicotinamide salvage pathway, and therefore induces a profound depletion of intracellular NAD. In contrast, 6-AN acts as an antimetabolite that competes with nicotinamide and leads to the synthesis of 6-ANAD. Because it does not inhibit NAMPT directly, 6-AN results in only a modest reduction in NAD levels while still impairing NAD-dependent enzymes through 6-ANAD accumulation. Similarly to 6-AN, FK866 and STF-118804 induced a marked reduction in intracellular ATP levels and a concomitant inhibition of cell proliferation, reflecting a global impairment of cellular metabolism. Both effects were fully reversed by supplementation of the culture medium with NAM (Fig 2C–2F). Importantly, the antiviral activity of FK866 and STF-118804 was also abrogated by the addition of NAM, demonstrating that inhibition of NAMPT is responsible for this phenotype. The resulting metabolic slowdown creates a cellular environment that is probably less permissive to viral replication. We also sought to determine whether inhibition of nicotinamide metabolism could have an effect on infection once it was established, and not only during the initial phases of infection. To do this, we infected the cells at a MOI of 1, let the infection develop for 72 hours before treating the cultures with 6-AN or FK866 supplemented or not with NAM. Although an effect of 6-AN on intracellular RNA levels was observed, no effect of FK-866 was detected (S4A Fig). Furthermore, we observed no effect of the inhibitors on either the quantity or the quality of secreted viral particles (S4B and S4C Fig). These results indicate that nicotinamide metabolism plays a critical role during the establishment phase of HCV infection, while its contribution is limited once the infection is fully established. To determine whether drugs targeting nicotinamide metabolism could alter viral replication in quiescent cells, Huh7 cells were differentiated for 7 days in the presence of 2% DMSO, and were then infected with HCV as previously described ([34]; S5 Fig). In this model, treatment with 6-AN or FK866 reduced intracellular levels of viral RNA (S5A Fig). Furthermore, and despite a limited effect on viral RNA secretion, the production of infectious particles collapsed (S5B and S5C Fig). This inhibitory effect was reversed by adding NAM to the culture medium, confirming the role of nicotinamide metabolism in HCV replication. We have also tested the effect of inhibition of NAD(H) metabolism in HepaSH cells, which are human hepatocytes purified from the liver of immunocompromised mice transplanted with primary human hepatocytes (PHH) [33]. Freshly purified cells were seeded in glycogen-coated well plates and cultured for 4 days prior to infection at a MOI of 1. Three days after infection, intracellular and secreted HCV RNA levels were quantified (S6 Fig); although viral titers remained below the detection threshold, quantifiable levels of extracellular viral RNA were detected. Under these conditions, HCV infection was reduced upon 6-AN treatment, with a more pronounced effect observed on intracellular viral RNA than on secreted RNA. Supplementation with NAM restored extracellular viral RNA levels, while only partially recovering intracellular genome levels. Taken together, and despite the overall low level of infection in this experimental setting, these data support an inhibitory effect of 6-AN on HCV replication in PHH. As NAD(H) can also be synthesized from tryptophan via the de novo biosynthesis pathway (Fig 1A), phthalic acid was used to inhibit Quinolinic acid PhosphoRibosylTransferase (QPRT), an enzyme catalyzing the last reaction leading to the de novo synthesis of nicotinic acid mononucleotide (NaMN) [34,35]. This drug had no effect on HCV replication in Huh7 cells (Fig 2A), suggesting that the de novo biosynthesis pathway is not required for viral replication. We also showed that viral replication is not affected by thionicotinamide adenine dinucleotide (Thio-NAD; S1D Fig), an analogue of NAD(H) inhibiting NADPK, the enzyme converting NAD to NADP [36](Fig 2A). Neither phthalic acid nor Thio-NAD had an effect on intracellular ATP or cell proliferation (S7 Fig). This suggests that conversion of NAD(H) into NADP(H) is not critical in the antiviral effect of 6-AN. 6-AN is also used as an inhibitor of the pentose phosphate (PPP) pathway via the inhibition of glucose-6-phosphate dehydrogenase (G6PD) and 6-phosphogluconate dehydrogenase (6PGD). D-ribose was thus added to the culture medium to determine whether inhibition of the PPP pathway contributes to the restriction of HCV replication by 6-AN. Indeed, phosphorylation of D-ribose by the ribokinase RBKS produces R5P, which enables nucleotide biosynthesis when the PPP pathway is impaired. HCV replication was weakly restored by D-ribose, suggesting that the effect of 6-AN is essentially independent of PPP pathway inhibition (Fig 2H) [14,37]. Furthermore, HY-W107464, a specific inhibitor of G6PD did not inhibit HCV replication (Fig 2I), confirming that PPP inhibition was not involved in HCV inhibition by 6-AN. Overall, these observations indicate that HCV critically depends on a functional NAD(H) salvage pathway for its replication and that NAMPT activity, the limiting enzyme of this pathway, is essential for HCV replication.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 2. HCV replication depends on nicotinamide salvage pathway.

(A-B) Huh7 cells were electroporated with subgenomic replicon and cultured for 72 h with the indicated treatment. Renilla luciferase activity was measured in cell homogenates and viral replication normalized to non-treated control (0 or Ctl.). Are presented means ± SEM. (A) Cells were cultured with or without 100 µM of thio-NAD, 100 µM of phthalic acid, 100 nM of FK866 or 500 µM of NAM, n = 3, one sample t-test, Bonferroni- Sidák adjusted p-value for multiple comparison to Ctl. (B) Cells were cultured with or without 100 nM STF-118804 and increasing doses of NAM. n = 3, one-way ANOVA, Bonferroni-Sidák adjusted p-value for multiple comparison to condition with STF-118804 alone. (C-F) Huh7 cells were cultured with or without 100 nM of FK866 or STF-118804 and increasing doses of NAM (0, 10, 50, 100 or 500 µM). Intracellular ATP amounts (C and E) were determined using CellTiter Glo assay (Promega) and cell proliferation (D and F) after Hoechst staining of nuclei and quantification of fluorescence. Are presented means ± SEM, n = 2. (G) Cells were electroporated with subgenomic replicon and cultured for 72 h with or without 100 nM of FK866 and increasing doses of NAM, n = 3, one sample t-test, Bonferroni-Sidák adjusted p-value for multiple comparison to condition with FK866 alone. (H) Cells were cultured with or without 100 µM 6-AN ± 25 mM of Ribose, n = 3, one-way ANOVA, Bonferroni adjusted p-value for multiple comparison. (I) Cells were cultured with or without 100 nM of G6PDi, n = 3, one sample t-test. n.s. non significative, *p < 0.05, **p < 0.01.

https://doi.org/10.1371/journal.ppat.1014165.g002

Inhibition of the NAD(H) pathway reduces the secretion and specific infectivity of infectious particles

As the inhibition of NAD(H) biosynthesis affects HCV replication as assessed with subgenomic replicon, we wanted to evaluate the impact on the production of new virions. Thus, Huh7 cells were infected with full-length HCV virions and treated cells with 6-AN in the absence or presence of NAM. For technical reasons, we used the strain Jc1-E2Flag [25] at a MOI 1. After 72 h of culture, virus titration showed that 6-AN decreased by 2 logs in the production of infectious particles and this inhibition was reversed by NAM (Fig 3A). The same observations were obtained with the NAMPT inhibitor FK866 (Fig 3D). Surprisingly, in presence of 6-AN, the amount of viral RNA secreted did not decrease as much as the number of infectious particles produced (Fig 3B), suggesting a change in particle infectivity. Indeed, the ratio between infectious particles and secreted viral genomes (FFU/GE), shown in Fig 3C and 3F was markedly reduced for cells treated with 6-AN or FK866 respectively. As the FFU/GE ratio reflects the specific infectivity of the produced particles, this decrease suggests a reduced infective capacity of the virions, or at least a reduction in the number of infectious particles secreted, consistent with the inhibitory effect initially observed on viral replication (Fig 1B–1D) and DMV formation (Fig 1H). The addition of NAM reversed this phenotype as expected (Fig 3C). The same results were obtained with FK866 (Fig 3E and 3F). A comparable inhibitory effect was also observed in Huh7 cells differentiated with DMSO prior to infection (S5 Fig). Thus, inhibition of the NAD(H) salvage pathway reduced not only the overall production of infectious particles but also changed their capacity to propagate infection (Fig 3). In contrast and as expected, Thio-NAD and Phtalic acid had no effect on the production of infectious HCV particles (Fig 3G), confirming the results obtained with the subgenomic replicon (Fig 2A).

Download:

• PNG
  larger image
• TIFF
  original image

Fig 3. Inhibition of NAD(H) metabolism reduces the production of infectious HCV particles.

(A-C) Huh7 cells were infected with HCV Jc1 strain at MOI 1 and cultured 72 h with or without 6-AN (100 µM), in the absence or presence of NAM (500 µM). (A) FFU were determined from cell culture supernatant harvested at 72 h post-infection. (B) Viral genomes within culture supernatants at 72 h post-infection. (C) Specific infectivity of secreted viral particles. Are presented means ± SEM, n = 8, one-way ANOVA for multiple comparison, n.s. non significative, *p < 0.05, **p < 0.01, ****p < 0.0001. (D-F) Huh7 cells were infected with HCV at MOI = 1 and cultured for 72 h with or without 100 nM FK866 or 100 nM FK866 + 500 µM NAM. (D) FFU were determined from cell culture supernatant harvested at 72 h post-infection. (E) Viral genomes within cell culture supernatants at 72 h post-infection. (F) Specific infectivity of viral particles. (D-F) Are presented means ± SEM, n = 3, one-way ANOVA for multiple comparison, n.s. non significative, *p < 0.05, **p < 0.01, ***p < 0.001. (G) Huh7 cells were infected with HCV at MOI = 1 and cultured with or without 100µM 6-AN, 100 µM thio-NAD, 100 µM phthalic acid, 100 nM FK866, or 500 µM NAM. Infectious titer was determined in cell culture supernatants at 72 h post-infection.

https://doi.org/10.1371/journal.ppat.1014165.g003

6-aminonicotinamide inhibits central carbon metabolism

We then determined the effect of 6-AN on the metabolism of Huh7 cells to better understand how inhibition of NAD(H) biosynthesis reduces HCV replication and the production of infectious particles. We first analyzed polar and semi-polar metabolites in Huh7 cells treated or not with 6-AN at 100 µM. Of the 1,768 metabolites detected by LC-MS/MS, 151 were identified with a high level of confidence (annotation level 1 and 2a) (S1 Table). Principal component analysis of experimental replicates based on these 151 metabolites revealed a clear segregation between untreated and 6-AN-treated cells (S8A Fig). The PC1 variation between controls and 6-AN-treated cells suggests that the main differences between the two clusters are captured by a unique set of metabolites explaining the effect. Sixty-eight metabolites were significantly downregulated and 19 were upregulated (Fig 4A; fold change>20% and p-value<0.05). First, as expected, the levels of ATP, NAD⁺ and NAM were reduced after 6-AN treatment. Interestingly, the first metabolites of glycolysis (glucose and glucose-6-phosphate) were increased, while final products such as pyruvate and lactate were decreased (Fig 4B). This may reflect a downregulation in glucose consumption rate and an overall reduced metabolic activity, contributing to the drop in ATP levels observed in the presence of 6-AN. Moreover, the amounts of several metabolites of the TCA cycle were reduced, suggesting that mitochondrial metabolic activity is also affected (Fig 4C).

Download:

• PNG
  larger image
• TIFF
  original image

Fig 4. Metabolome modifications induced by NAD(H) biosynthesis inhibition.

Huh7 cells were cultured for 72 h with or without 6-AN (100 µM) before metabolite extraction. (A) Volcano plot showing variations in the expression of the 151 metabolites identified with high confidence criteria. The x-axis corresponds to Log₂(fold change) and the y-axis corresponds to -Log₁₀(p-value). Horizontal dotted line indicates the p-value threshold of 0.05 and vertical dotted lines indicate variations above the 20% threshold. Metabolites were grouped by color for their association with a given metabolic pathway. (B and C) Relative abundances of metabolites associated to glycolysis and TCA cycle. Means ± SEM obtained from 4 independent experiments are indicated with p-value for Student’s t-test comparison.

https://doi.org/10.1371/journal.ppat.1014165.g004

To confirm the effect of 6-AN on glycolysis and mitochondrial respiration, both activities were measured after a 72 h of treatment, using an extracellular flux analyzer [38]. In cell cultures, glycolysis and mitochondria-derived CO₂ are the two main contributors to extracellular acidification rate (ECAR). We therefore determined first the glycolytic proton efflux rate (glycoPER) to monitor the glycolytic activity of cells [39]. Basal glycolytic activity was measured first, followed by compensatory glycolysis after blockade of complex I and III of the mitochondrial respiratory chain with Rotenone and Antimycin A, respectively. In agreement with LC-MS/MS data, 6-AN treatment induced a strong inhibition of both basal and compensatory glycolysis, which are restored by NAM (Fig 5A and 5B). We then analyzed the oxygen consumption rate (OCR) to determine the effect of 6-AN treatment on mitochondrial respiration (Fig 5C and 5D). In cells treated with 6-AN, basal respiration increased, while the addition of NAM reversed the effect of 6-AN (Fig 5C and 5D). Using sequential addition of inhibitors, we determined as previously described the fraction of O₂ consumption utilized for ATP production, the maximal respiration rate and the spare respiration capacity (Fig 5E). We observed that in presence of 6-AN, the increase in basal respiration correlates with an increase in ATP production through oxidative phosphorylation (Fig 5D and 5E). Thus, the increase in oxidative phosphorylation aims to compensate the glycolytic blockade. We also observed a total loss of the spare respiratory capacity in 6-AN-treated cells (Fig 5F). This mitochondrial parameter reflects the cell’s ability to respond to enhanced energy demand under a physiological stress by stimulating the respiratory chain. In presence of 6-AN, oxidative phosphorylation is increased to its maximum, matching levels reached by addition of the uncoupling agent FCCP. Overall, these results show reduced glycolytic activity, which could limit viral replication, combined with increased mitochondrial respiration as compensatory mechanism for ATP production.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 5. 6-AN inhibits glycolysis and rebalances mitochondrial respiration.

Huh7 cells were treated or not with 100 µM 6-AN ± 500 µM of NAM for 72 h. (A) Quantification of glycolytic activity determined by the GlycoPER (proton efflux rate specific to glycolysis) using the Glycolytic Rate Test. Are presented means ± SEM of 20 replicates. Data are representative of 3 independent experiments. (B) Basal glycolysis and compensatory glycolysis were calculated from the Glycolytic Rate Test presented in A (one-way ANOVA for multiple comparison, n.s. non significative, ****p < 0.0001). (C) Oxygen consumption rate (OCR) was determined using the Mito Stress test before and after addition of oligomycin (Complex V inhibitor), FCCP (uncoupling agent), rotenone (Rot; Complex I inhibitor) plus antimycin A (Anti-A; Complex III inhibitor). (D) Basal OCR and ATP production were calculated from (C), as indicated in E. (E) Graphical representation of the different parameters that can be calculated with the Mito Stress assay from Agilent. (F) Spare respiration capacity and maximal respiration were calculated from C. (C, D and F) Are presented means ± SEM of 5 replicates (one-way ANOVA for multiple comparison, *p < 0.05, **p < 0.01, ***p < 0.0002, ****p < 0.0001). Data are representative of 3 independent experiments.

https://doi.org/10.1371/journal.ppat.1014165.g005

Inhibition of NAD(H) biosynthesis reduces cellular triglycerides level and secretion of apolipoproteins

LC-MS/MS data also showed modulation of several lipid metabolites (Fig 4A), probably a consequence of the overall effect of 6-AN on the cellular energetic metabolism. In particular, a metabolite set enrichment analysis (MSEA) showed that phosphatidylcholine and phosphatidylethanolamine biosynthesis pathways were strongly impacted by 6-AN (S8B Fig). Levels of carnitine-conjugated fatty acids were also reduced, suggesting changes in fatty acid and triglyceride metabolism (Figs 4A and S8B). Given their key role in the formation of LVPs, we investigated the effect of 6-AN on apolipoproteins secretion and intracellular levels of triglycerides. After 72 h of 6-AN treatment, the secretion of ApoB and ApoE, which are associated with LVPs and are essential to their infectivity, was reduced (Fig 6A and 6B). We also observed a reduction in intracellular TG content (Fig 6C). We thus studied the impact of 6-AN treatment on intracellular lipid droplets by confocal microscopy after staining of neutral lipids. The number and morphological parameters of lipid droplets (LDs) were determined after three-dimensional reconstruction from fluorescence microscopy images (Fig 6D–6E). Interestingly, the average LD volume per cell was reduced by 62% in 6-AN–treated cells (Fig 6E), consistent with intracellular TG quantification using a specific biochemical enzymatic assay (Fig 6A). Addition of NAM to the cell culture medium barely restored the volume of the LDs (Fig 6E). Altogether, these data indicate an important impact of NAD(H) metabolism inhibition on hepatocyte neutral lipids metabolism decreasing intracellular TG content and lipoprotein secretion, with a potential impact on the amount and quality of secreted of LVPs.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 6. Inhibition of NAD(H) metabolism decreases intracellular TG storage.

Huh7 cells were treated or not with 100 µM 6-AN for 72h of culture. (A-B) ApoB and ApoE secreted in cells supernatant quantified by ELISA. Are presented means ± SEM (n = 3, student’s t-test, **p < 0.0021, ***p < 0.0002). (C) Quantification of intracellular triglycerides in total cell extracts. Are presented means ± SEM (n = 3, student’s t-test, *p < 0.01). (D) Representative picture of 3D lipid droplets reconstruction using IMARIS software in Huh7 cells after Oil-red-O (yellow) and Hoechst (blue) staining. (E) The mean total volume of lipid droplets per cell was determined by analyzing their intracellular content of 50 cells for each condition with the IMARIS software. Are presented means ± SEM, one-way ANOVA for multiple comparison, n.s. non significative, ****p < 0.0001.

https://doi.org/10.1371/journal.ppat.1014165.g006

Inhibition of NAD(H) biosynthesis alters the composition and buoyant density of HCV secreted particles

Based on the effect of 6-AN on lipid metabolism, we hypothesized that 6-AN could affect LVP production. We therefore infected cells with the Jc1-E2Flag strain, treated them with or without 6-AN, and then separated secreted HCV particles on a density gradient. We then collected fractions of increasing density, and analyzed the specific infectivity of HCV particles in each fraction. In low density fractions (1.00-1.06 g/ml), specific infectivity was reduced by up to 90% by 6-AN, whereas it was unaffected in higher-density fractions (Fig 7A). This effect on low-density viral particles was reversed by NAM. As previously observed in whole culture supernatants, the overall amount of viral RNA secreted and FFU titer were decreased in the presence of 6-AN, regardless the density of the viral particles (Fig 7B and 7C). These results were confirmed when we analyzed the specific infectivity of LVPs produced by infected cells with the Jc1 (S9 Fig). Therefore, in addition to inhibiting viral replication, treatment with 6-AN specifically reduced the production of LVPs.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 7. Inhibition of NAD(H) metabolism reduces LVP infectivity.

(A-C) Huh7 cells were infected with HCV at MOI = 1 and cultured for 72 h with or without 6-AN (100 µM), in the absence or presence of NAM (500 µM). 1ml of ten-fold concentrated cell-culture supernatants were separated on iodixanol gradient (3-40%). Specific infectivity of viral particles (A), HCV RNA content (B) and FFU (C) were determined in each of the collected density fractions. Are presented data representative of 3 independent experiments.

https://doi.org/10.1371/journal.ppat.1014165.g007

Highlights

• NAD(H) pathway inhibitor 6-AN rewires energy and lipid metabolism in hepatocytes
• HCV depends on NAD(H) metabolism to form double-membrane vesicles and replicate
• The production of HCV Lipo-Viro-Particles is altered by NAD(H) pathway inhibitors