Arbidol: a broad-spectrum antiviral that inhibits acute and chronic HCV infection
© Boriskin et al. 2006
Received: 02 June 2006
Accepted: 19 July 2006
Published: 19 July 2006
Skip to main content
© Boriskin et al. 2006
Received: 02 June 2006
Accepted: 19 July 2006
Published: 19 July 2006
Arbidol (ARB) is an antiviral compound that was originally proven effective for treatment of influenza and several other respiratory viral infections. The broad spectrum of ARB anti-viral activity led us to evaluate its effect on hepatitis C virus (HCV) infection and replication in cell culture. Long-term ARB treatment of Huh7 cells chronically replicating a genomic length genotype 1b replicon resulted in sustained reduction of viral RNA and protein expression, and eventually cured HCV infected cells. Pre-treatment of human hepatoma Huh7.5.1 cells with 15 μM ARB for 24 to 48 hours inhibited acute infection with JFH-1 virus by up to 1000-fold. The inhibitory effect of ARB on HCV was not due to generalized cytotoxicity, nor to augmentation of IFN antiviral signaling pathways, but involved impaired virus-mediated membrane fusion. ARB's affinity for membranes may inhibit several aspects of the HCV lifecycle that are membrane-dependent.
There are presently limited therapeutic options for patients with chronic hepatitis C, especially those who have failed interferon (IFN) based modalities. The HCV replicon system, originally described by Lohmann and colleagues , has proven to be an effective in vitro model for pre-clinical evaluation and large-scale screening of new anti-HCV compounds (reviewed in . In addition to the testing of novel anti-HCV compounds, the replicon system has also facilitated the characterization of existing compounds that show antiviral activity against other viruses [3, 4]. Drugs that target viral proteins, such as the NS3-4a protease, are presumed to be less toxic and more specific. However, it is now clear that the use of these compounds can lead to drug-resistant viral variants, at least in vitro . Another group of antivirals with broad-spectrum activity impact cellular metabolic pathways such as interferon production, or interfere with cellular functions or critical steps in virus-cell interactions, possibly exerting higher cell toxicity with little, if any, virus resistance. Other broad-spectrum antiviral agents target rate-limiting events in viral replication cycle such as envelope protein glycosylation, processing and folding , or viral-cell membrane fusion during viral uncoating or assembly (reviewed in [6, 7]).
One example of the latter group compounds is arbidol (ARB; ethyl-6-bromo-4-[(dimethylamino)methyl]-5-hydroxy-1-methyl-2-[(phenylthio)methyl]-indole-3-carboxylate hydrochloride monohydrate), a Russian-made broad-spectrum antiviral that had been shown to inhibit the fusion of influenza A and B viruses within endosomes . An acidic environment (pH 5.0) is a strict prerequisite for influenza virus-induced fusion, and even a slight increase in pH abolishes the fusion process . ARB has also been shown to exert antiviral activity against other pH-dependent viruses, such as hepatitis B virus , and rhinovirus 14 (reviewed in ). Since ARB is a weak base, it might elevate endosomal pH and abrogate virus-endosome fusion. Since optimal HCV envelope protein fusion with cell membranes requires low pH and the fusion process may occur within endosomes [11–14], ARB might potentially exert an antiviral activity on HCV as well. However, since ARB was shown to inhibit various pH-independent viruses such as the human respiratory syncytial virus and the parainfluenza virus 3 , ARB could affect membrane interactions that are necessary for virus replication. Alternatively, ARB has been claimed to stimulate the production of interferon [7, 8], a well-known and potent inhibitor of HCV replication . Taken together, these observations prompted us to investigate ARB activity on HCV lifecycle, and more specifically on HCV replication. We therefore studied the effects of ARB on HCV RNA and protein expression using primary productive HCV infection or persistent non-productive HCV infection in cultured replicon cells. We report that ARB inhibits acute and chronic HCV replication independently of activation of innate antiviral signaling pathways. Instead, we show that ARB antiviral activity towards HCV is due to a direct effect of ARB on virus-cell membrane interactions.
The advent of HCV replicon cultures [1, 23] and in particular, productive replicon infection systems [17, 24, 25], has given tremendous opportunities for preclinical assessment of anti-HCV compounds. As a result, a number of specific viral inhibitors are already in various phases of clinical trials (reviewed in ). Among the non-specific, broad-spectrum antivirals, a few well-known over-the-counter drugs have shown inhibitory activity against HCV in replicon cell cultures [3, 4]. This latter group of antivirals can now be extended to include ARB, which exerts anti-viral activity against acute and chronic HCV replication.
Based on its chemical structure, ARB may be a pro-drug which becomes chemically converted into an active drug by cellular metabolic processes. The pro-drug nature of ARB may explain its relatively high CC50 values, assuming that the actual ARB metabolite concentration is lower than that of the ARB pro-drug. The carboxylic acid ester moiety contained in the structure of ARB may be a substrate for hydrolysis in vivo that leads to the intracellular accumulation of ARB. The fact that ARB displayed prophylactic activity when administered 24–48 hours before primary infection, and over several weeks of treatment in persistent HCV infection, might indicate a prerequisite for ARB accumulation in intracellular compartments before antiviral activity is observed. Clearly, additional studies of ARB and various chemical derivatives are warranted. Nonetheless, the sustained effect of ARB on persistent HCV could be of clinical significance since chronic infection comprises about 75–85% of all hepatitis C cases .
Our results suggest that the inhibitory effect of ARB on HCV is not mediated by stimulation of type 1 IFN signaling pathways. In fact, ARB inhibited antiviral signaling in FL-Neo and Huh7 cells, an effect which might be attributable to ARB disruption of membrane interactions required for signal transduction. Instead, ARB's anti-HCV action appears to be attributable to an inhibitory effect on viral fusion (Pécheur EI, Boriskin Y, Lavillette D, Roberts, M, Cosset FL, Polyak SJ, manuscript in preparation). In brief, at 1–6 μM concentration, ARB completely blocked the fusion of HCV pseudoparticles from multiple genotypes with liposomal membranes in a strictly controlled pH environment of 5.0. This observation makes it unlikely that ARB inhibits virus replication by increasing the endosomal pH like other weak bases such as chloroquine . Moreover, since HCV fusion takes place within much wider range (6.3 – 5.0) compared to strictly HA conformation-sensitive influenza virus fusion , if ARB did induce a basic pH shift, it is predicted that the shift would be minor and affect the HCV fusion process minimally. One model for ARB's anti-fusion activity is that ARB, because of its chemical structure, has a propensity for cell membranes. If ARB intercalates into membranes and adopts a consistent orientation, the formation of an "ARB cage" could lead to excessive stabilization of cell membranes, and thereby prevent HCV fusion. Alternatively, at the virus level, ARB might block the un-coating of the membrane during the fusion process. It is also conceivable that ARB could inhibit other aspects of the HCV life cycle that are dependent on cell membranes. For example, the HCV replication complex associates with endoplasmic reticulum membranes to form membranous webs . The web is formed via the association of HCV non-structural proteins with ER membranes . Thus, ARB-induced inhibition of HCV non-structural protein interactions with organelle membranes might also contribute to suppression of HCV replication. However, at least in terms of acute infection with JFH-1, the predominant mechanism of action of ARB is likely to be inhibition of fusion, since ARB only suppressed JFH-1 infection when given 24–48 hours prior to infection. On the other hand, FL-Neo cells were cured of HCV replication following several weeks of ARB treatment. Since genomic length replicons do not produce infectious virus, in these cells, ARB inhibition of the HCV replication complex association with membranes could be the predominant mode of HCV suppression. Additional studies are required to sort out these possibilities.
In conclusion, ARB inhibits HCV acute and chronic HCV infection and replication. The inhibitory effect appears to be due to the interaction of ARB with membranes and to subsequent ARB-induced membrane alterations, but the nature of the membrane modification(s) require further study.
Huh7 and Huh7.5.1 human hepatoma cells were cultured in Dulbecco's modified Eagle medium (DMEM) containing 9% fetal calf bovine serum, 1% penicillin-streptomycin-fungizone and 1% nonessential amino acids (all reagents were from Invitrogen, Carlsbad, CA). FL-Neo cells are a stable human hepatoma Huh7-derived cell line harboring autonomously replicating genomic length genotype 1b HCV replicon with adaptive mutations in NS3 (P1496L) and NS5A (S2204I). Huh7 and FL-Neo cells were obtained from Apath, LLC (St Louis, MO), and Huh7.5.1 cells were obtained from Francis Chisari . FL-Neo cells were cultured in Huh7 medium supplemented with 0.4 mg/ml of G418 (Calbiochem, San Diego, CA). They were passaged with a 1:10 split and maintained subconfluent in the course of each experiment. During short- or long-term ARB treatment, FL-Neo cells were cultured in medium without G418. Control cultures consisting of FL-Neo cells grown in the absence of ARB were always run in parallel with ARB treated cultures, and for the same duration. All cell lines were checked for mycoplasma using MycoAlert assay (Cambrex Bio Science, Rockland, ME) and found to be mycoplasma-free.
ARB was a powdered free base formulation and was dissolved to completion in 0.5 ml of 96-proof ethanol at 37°C for 10 min followed by dilution in 4.5 ml of sterile distilled water. The final ethanol content in cell culture medium was always less than 10-6 M. For each experiment a freshly prepared stock of ARB was used. JFH-1 viral stock preparation, cell infection and titration was performed exactly as described [17, 25]. JFH-1 plasmid was kindly provided by Takashi Wakita. RIG-N, a constitutively active mutant of RIG-I, was kindly provided by Michael Gale.
We used the luminescence ATP detection assay system (ATPlite, Perkin Elmer, Boston, MA) as described by the manufacturer. Huh7, Huh7.5.1 or FL-Neo cells were grown overnight in black 96-well view plates (104 cells per well, 6 wells per drug concentration). After 24 h incubation with the compound the wells were washed twice with 0.2 ml of phosphate-buffered saline (PBS) followed by addition of 0.1 ml of PBS and 50 μl of lysis solution (provided with the kit) to each well. The microplate was shaken for 5 min at 600 rpm on an orbital shaker to allow cell lysis and ATP stabilization. 50 μl of the substrate solution was then added, and the microplate was shaken for 5 min at 600 rpm. Luminescence was measured on a TopCount NXT microplate scintillation & luminescence counter (Packard; Perkin Elmer) after a 10 min dark adaptation. The 50% cytotoxic concentration (CC50) was determined from the dose-response curve and defined as the drug concentration that caused a 50% signal reduction compared to that of untreated cultures.
Cells grown in Costar 6-well plates (0.2 × 106 cells per well) were lysed in 0.1 ml of RIPA buffer (50 mM tris-HCl, pH 7.2, 150 mM NaCl, 0.1% SDS, 0.1% Na deoxycholate, 1% Triton X-100, 17.4 μg/ml PMSF). The protein lysates were quantified using the BSA Protein Assay (Pierce Biotechnology, Rockford, IL). Before gel loading each sample was adjusted to contain 10 μg of protein per gel well. Samples were mixed with equal volume of double-strength reducing loading buffer and heated at 95°C for 7 min before subjected to electrophoresis on a 4–20% tris-glycine gel (Invitrogen). Separated proteins were transferred to a 0.45 μm nitrocellulose membrane (Pierce) using a semi-dry transfer system. After the transfer, the membrane was blocked in Superblock buffer (Pierce) and incubated with primary mouse antibody, either anti-NS5A (Biodesign International, Saco, ME.; 1:1000 dilution) or anti-Core (Affinity Bioreagents, Golden, CO; 1:1000 dilution) for 1 hr at room temperature. Both proteins were detected on the same blot after sequential treatment with anti-NS5A, then with anti-Core antibody, and four washes with PBS – 0.2% Tween 20 (PBST) in between antibody treatment. The secondary antibody was HRP-conjugated anti-mouse immunoglobulin G (IgG) (Pierce, 1:10000 dilution). To control for comparable gel loading, the same blot was stripped with Restore stripping buffer (Pierce), and was re-blocked in Pierce Superblock buffer and incubated overnight with goat polyclonal IgG antibody to glyceraldehyde-3-phosphate dehydrogenase (GAPDH; Santa Cruz Biotechnology, Santa Cruz, CA) diluted 1:1000 in PBST. The blot was washed as above and incubated with bovine anti-goat IgG-HRP (Santa Cruz Biotechnology) diluted 1:10000 in PBST, for 1 hr at room temperature. Protein bands were detected using chemiluminescence LumiGlo reagents (Cell Signaling, Danvers, MA) and visualized after exposing the membrane against X-ray film. Quantification of bands was performed using ImageJ software .
Huh7 or FL-Neo cells were seeded in Costar 6-well clusters at 2.0 × 105 cells per well. Sub-confluent treated or untreated cultures were lysed in 0.6 ml/well of RLT buffer (Qiagen) containing 1% beta-mercaptoethanol. Total cellular RNA was isolated using the RNeasy kit (Qiagen, Valencia, CA) according to manufacturer's instructions. The RNA integrity was verified by visualizing ribosomal RNAs on 1.2% agarose gel, and total RNA concentration was determined using RediPlate™ 96 Ribogreen RNA quantitation Kit (Molecular Probes, Invitrogen). Ten ng of RNA were added to wells of a 384 well plate containing the EZ RT-PCR master mix (Perkin Elmer). Samples were run on an ABI HT7900 real time RT-PCR machine. HCV RNA was quantitated by real time RT-PCR, as described [31, 32]. For each run, dilutions of HCV plasmid DNA (precisely quantitated using Invitrogen PicoGreen DNA quantitation kit) ranging from 0–107 copies per tube, were run in triplicate to generate a standard curve, which served as a reference to calculate HCV RNA copy number based on the cycle threshold (Ct). The HCV RNA copy number was expressed as copies per 10 ng total cellular RNA. Negative controls included reactions lacking template as well as RNA from Huh7 cells, which were always negative for HCV RNA.
The HCV pseusoparticle (HCVpp) system was described in details elsewhere [14, 22]. The measure of fusion between pseudoparticles and liposomes was based upon a lipid mixing assay, as described earlier . Briefly, liposomes of phosphatidylcholine/cholesterol labeled with octadecylrhodamine (R18) (65:30:5 mol/mol) were incubated at 37°C with HCVpp harboring the E1 and E2 glycoproteins of HCV genotype 1b isolates AY734975  and AF333324, in the absence or presence of 1 or 6 μg/ml ARB. Lipid mixing was initiated by decreasing the pH to 5.0 with diluted HCl, and kinetics were recorded over a 10-min period of time on an SLM Aminco 8000 spectrofluorimeter, at λexc 560 nm and λem 590 nm.
hepatitis C virus
IFN stimulated response element
Japanese fulminant hepatitis-1
reverse transcriptase polymerase chain reaction
retinoic acid inducible gene I
sodium dodecyl sulphate-polyacrylamide gel electorphoresis
signal transducer and activator of transcription
We thank A.M. Schuster and I.A. Leneva for the gift of ARB, Jeff Krise for discussing ARB structure, François Penin for very helpful discussions, and Dimitri Lavillette for HCVpp. We also thank Apath, LLC, Francis Chisari, Michael Gale, and Takashi Wakita for reagents and technical advice, and Jessica Wagoner, and Michael Austin for technical assistance. YSB was partially supported by the Fulbright Visiting Scholar Program. SJP is supported by NIH grants RO1 DK62187 and U19 A1066328, and EIP is supported by a grant from the ANRS (Agence Nationale de Recherche contre le SIDA et les Hépatites virales).
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.