Article Text

PDF

Original article
Turmeric curcumin inhibits entry of all hepatitis C virus genotypes into human liver cells
  1. Anggakusuma1,
  2. Che C Colpitts2,
  3. Luis M Schang2,
  4. Heni Rachmawati3,
  5. Anne Frentzen1,
  6. Stephanie Pfaender1,
  7. Patrick Behrendt1,5,
  8. Richard J P Brown1,
  9. Dorothea Bankwitz1,
  10. Joerg Steinmann4,
  11. Michael Ott5,6,
  12. Philip Meuleman7,
  13. Charles M Rice8,
  14. Alexander Ploss8,9,
  15. Thomas Pietschmann1,
  16. Eike Steinmann1
  1. 1Institute of Experimental Virology, Twincore Centre for Experimental and Clinical Infection Research, Hannover, Germany
  2. 2Departments of Biochemistry and Medical Microbiology and Immunology, Li Ka Shing Insitute of Virology, University of Alberta, Edmonton, Alberta, Canada
  3. 3School of Pharmacy, Institut Teknologi Bandung, Bandung, Indonesia
  4. 4Institute of Medical Microbiology, University Hospital Essen, Essen, Germany
  5. 5Department of Gastroenterology, Hepatology and Endocrinology, Medical School Hannover, Hannover, Germany
  6. 6Translational Research Group, Twincore Centre for Experimental and Clinical Infection Research, Hannover, Germany
  7. 7Center for Vaccinology, Ghent University, Ghent University hospital, Ghent, Belgium
  8. 8Laboratory of Virology and Infectious Diseases, Center for the Study of Hepatitis C, The Rockefeller University, New York, New York, USA
  9. 9Department of Molecular Biology, Princeton University, Princeton, New Jersey, USA
  1. Correspondence to Dr Eike Steinmann, Institute of Experimental Virology, Twincore Center for Experimental and Clinical Infection Research (a joint venture between the Medical School Hannover (MHH) and the Helmholtz Centre for Infection Research (HZI)), Feodor-Lynen-Straße 7-9, Hannover 30625, Germany; eike.steinmann{at}twincore.de

Abstract

Objective Hepatitis C virus (HCV) infection causes severe liver disease and affects more than 160 million individuals worldwide. People undergoing liver organ transplantation face universal re-infection of the graft. Therefore, affordable antiviral strategies targeting the early stages of infection are urgently needed to prevent the recurrence of HCV infection. The aim of the study was to determine the potency of turmeric curcumin as an HCV entry inhibitor.

Design The antiviral activity of curcumin and its derivatives was evaluated using HCV pseudo-particles (HCVpp) and cell-culture-derived HCV (HCVcc) in hepatoma cell lines and primary human hepatocytes. The mechanism of action was dissected using R18-labelled virions and a membrane fluidity assay.

Results Curcumin treatment had no effect on HCV RNA replication or viral assembly/release. However, co-incubation of HCV with curcumin potently inhibited entry of all major HCV genotypes. Similar antiviral activities were also exerted by other curcumin derivatives but not by tetrahydrocurcumin, suggesting the importance of α,β-unsaturated ketone groups for the antiviral activity. Expression levels of known HCV receptors were unaltered, while pretreating the virus with the compound reduced viral infectivity without viral lysis. Membrane fluidity experiments indicated that curcumin affected the fluidity of the HCV envelope resulting in impairment of viral binding and fusion. Curcumin has also been found to inhibit cell-to-cell transmission and to be effective in combination with other antiviral agents.

Conclusions Turmeric curcumin inhibits HCV entry independently of the genotype and in primary human hepatocytes by affecting membrane fluidity thereby impairing virus binding and fusion.

  • Antiviral Therapy
  • Chronic Viral Hepatitis
  • Drug Development
  • HCV

Statistics from Altmetric.com

Significance of this study

What is already known about this subject?

  • Globally, an estimated 160 million people are chronically infected with hepatitis C virus (HCV).

  • Individuals undergoing organ liver transplantation for complications of HCV infection pose a particular clinical problem: graft re-infection with HCV occurs in nearly all cases and long-term outcomes are unsatisfactory.

  • Affordable antiviral strategies targeting the early stages of infection are urgently needed to prevent the recurrence of HCV infection.

What are the new findings?

  • Curcumin potently inhibits entry of all major HCV genotypes.

  • Other curcumin derivatives, but not tetrahydrocurcumin, exert antiviral activity, suggesting the importance of α,β-unsaturated ketone groups.

  • Membrane fluidity experiments indicate that curcumin affects the fluidity of the HCV envelope resulting in impairment of viral binding and fusion.

  • Curcumin also inhibits cell-to-cell transmission and is effective in combination with other antiviral agents.

How might it impact on clinical practice in the foreseeable future?

  • This study paves the way for clinical evaluation of curcumin as a potential entry inhibitor for preventing graft re-infection in HCV-induced organ liver transplantation.

Introduction

Globally, an estimated 160 million people are chronically infected with hepatitis C virus (HCV)1 and are therefore at a high risk of developing severe liver damage including hepatic steatosis, fibrosis, cirrhosis and hepatocellular carcinoma.2 ,3 Owing to its error-prone RNA replication, HCV is a highly variable virus, and, based on phylogenetic analyses, viral isolates are grouped into seven major genotypes with variable prevalence across the world.4 The current standard therapy, a combination of pegylated interferon-α (PEG-IFN-α) and ribavirin, is associated with serious side effects and does not achieve an effective response in every patient.5 New HCV protease inhibitors tailored to genotype 1 have recently been approved by the US Food and Drug Administration (FDA) and shown to improve treatment options for genotype 1-infected patients. Organ liver transplantation (OLT) for complications of HCV infection poses a particular clinical problem: graft re-infection with HCV occurs in nearly all cases and long-term outcomes are unsatisfactory.6 The pharmacological repression of immune function after OLT results in enhanced viral replication. Moreover, some immunosuppressants have been found to mediate proviral effects, as we have shown for glucocorticoids.7 Prevention of donor liver re-infection, as routinely achieved in the case of hepatitis B, is a major clinical goal, but will likely require efficient pharmacological means of preventing viral entry into hepatocytes. Several compounds or peptides have been reported to block HCV entry.8–13 However, none of these has yet been approved for clinical use.

HCV is a hepatotropic RNA virus that establishes a chronic infection in the majority of cases. The HCV genome, 9.6 kb in size, encodes a single polyprotein which is cleaved by cellular and viral proteases into 10 different proteins: the structural protein core, E1, E2, the ion channel p7 and the non-structural proteins NS2, NS3, NS4A, NS4B, NS5A and NS5B. The variability of HCV permits immune evasion and thus facilitates viral persistence. The rapidly evolving HCV RNA is also a challenge for development of specific antiviral therapies active against all viral genotypes and for prevention of drug resistance. Recently, cell culture systems based on the Japanese fulminant hepatitis 1 (JFH1) strain of HCV genotype 2a have been developed. These viruses are infectious both in vitro and in vivo and therefore permit the study of the complete HCV replication cycle.14–16 The availability of these full-length cell-culture-derived HCV (HCVcc) systems allows the identification and characterisation of new antiviral agents or therapies acting on early and later stages of HCV infection.

Curcumin, a natural non-steroidal polyphenolic compound, is the principal curcuminoid derived from the rhizome of the popular spice turmeric (65–80%), in addition to smaller amounts of two structurally related derivatives that lack either one (desmethoxycurcumin, 15–25%) or both (bisdesmethoxycurcumin, 5–15%) methoxy groups in the phenyl rings.17 Curcumin has been used for centuries in Asian traditional medicine to cure a wide range of gastrointestinal disorders. It has also been shown to possess a wide spectrum of biological and pharmacological properties, with no associated toxicities.18 ,19 The anti-inflammatory, antiangiogenic and antineoplastic properties of curcumin have been demonstrated to improve treatment outcomes against various gastrointestinal tract inflammations and malignancies in both in vitro and in vivo studies.20 The hepatoprotective ability of curcumin has also been shown to alleviate the severity of hepatic inflammation in mice with steatohepatitis21 and reduce liver damage, cholangitis and biliary fibrosis in Mdr2−/− mice.22 In this study, curcumin was identified as a novel, pan-genotypic inhibitor of HCVcc cell entry without interfering with genome replication or virus production. These preclinical studies pave the way for clinical evaluation of curcumin as a potential entry inhibitor for preventing graft re-infection in HCV-induced OLT.

Materials and methods

Compounds and reagents

Curcumin, demethoxycurcumin, bisdemethoxycurcumin, tetrahydrocurcumin (THC) and epigallocatechin-3-gallate (EGCG) were purchased from Sigma-Aldrich (Seelze, Germany). Human PEG-IFN-α was purchased from IntronA, Essex Pharma (München, Germany). Ciclosporin A was provided by Novartis (Basel, Switzerland) and boceprevir by Institute Pasteur Korea (Seongnam, Korea). Octadecyl rhodamine B chloride (R18) was obtained from Invitrogen (Grand Island, New York, USA). Cholesterol and 1,6-diphenyl-1,3,5-hexatriene (DPH) was obtained from Sigma-Aldrich (Oakville, Ontario, Canada).

Plasmids and viruses

Plasmids pFK-Jc1 and pFK-JFH115 ,23 and intergenotypic chimeras with core-NS2 of genotype 1–7 with untranslated regions and NS3-NS5B of JFH1 have been described previously.24–28 Monocistronic Renilla luciferase reporter virus genomes designated H77c/1a/R2a, J4/1b/R2a, JcR2a, J8/2b/R2a, S52/3a/R2a, ED43/4a/R2a, SA13/5a/R2a, HK6a/6a/R2a and QC69/7a/R2a were described recently.10

Cell culture

Huh-7.5 cells were cultured as described previously.29 Primary human hepatocytes (PHHs) were purchased from Primacyt (Schwerin, Germany) and resuspended in hepatocyte plating medium (500 mL Dulbecco's modified Eagle's medium, high glucose; 10% fetal bovine serum (FBS)) and plated at a concentration of 2.5×105 cells/cm2. The viability of Huh-7.5 cells and PHHs was measured by MTT colorimetric assay (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide; Sigma-Aldrich).

Preparation of retroviral pseudo-particles

Murine leukaemia virus (MLV)-based pseudo-types bearing envelope glycoprotein of vesicular stomatitis virus (VSV-G) or amphotropic murine leukaemia virus (A-MLV) or HCV E1-E2 of H77 or J6CF isolates were generated as described recently.9

Quantification of HCV infection and replication

Huh-7.5 cells were infected or electroporated with reporter virus genome as previously described.9 HCV infection and RNA replication rate were quantified by measuring luciferase activity.10 X-tail reverse transcriptase (RT)-PCR experiments were performed as described.30

Additional methods are posted as online supplementary information.

Results

Curcumin inhibits HCV infection, and the α,β-unsaturated ketone groups are important for the antiviral activity

The main curcuminoid ingredient in turmeric is curcumin, in addition to smaller amounts of two structurally related derivatives that lack either one (desmethoxycurcumin) or both (bisdesmethoxycurcumin) methoxy groups in the phenyl rings. To test whether curcuminoids in turmeric, including tetrahydrocurcumin (THC), the main metabolised form of curcumin that lacks α,β-unsaturated ketone groups (figure 1A), would influence HCV entry, we assessed the effect of all curcuminoids and THC on HCVcc infection. Huh-7.5 cells were infected with Jc1 reporter virus in the presence of the compounds, washed with phosphate-buffered saline (PBS) 4 h later, and the infection measured 48 h after that. All curcuminoids inhibited HCV infection in a similar dose-dependent manner (figure 1B), whereas THC did not (figure 1B), suggesting that the α,β-unsaturated ketone groups, but not the methoxy group, are critical for the anti-entry curcumin activity.

Figure 1

Curcumin and the other curcuminoids are equally active in inhibiting hepatitis C virus infectivity, and the α,β-unsaturated ketone groups in curcumin are important for the antiviral activity. (A) Structure of curcumin derivatives and its main metabolite, tetrahydrocurcumin (THC). (B) Huh-7.5 cells were inoculated with JcR2a reporter viruses in the presence of increasing concentrations of curcumin, desmethoxycurcumin, bis-desmethoxycurcumin or THC. The inoculum was removed 4 h later and then monolayers were washed three times with phosphate-buffered serum and overlaid with fresh medium containing no curcumin. Infected cells were lysed 2 days later and Renilla luciferase activity was determined. A representative experiment of three independent repetitions with SD of the means is shown. DMSO, dimethyl sulfoxide.

Curcumin inhibits entry of all HCV genotypes but does not repress HCV RNA replication, assembly or release

Next, we evaluated the antiviral effects of curcumin on all HCV genotypes and used Renilla luciferase reporter viruses and non-reporter viruses. The addition of curcumin during infection of naive Huh-7.5 cells with HCVcc resulted in dose-dependent inhibition of infectivity with no observed cytotoxicity (figure 2A). The half-maximal inhibitory concentration (IC50) for curcumin was approximately 8.46±1.27 µM (2.94±0.37 µg/mL; n=3). The antiviral activity of curcumin against all major HCV genotypes was comparable, indicating that curcumin inhibits HCV independently of viral genotype or subtype (figure 2A). The same inhibition was observed for a novel genotype 1a strain TN (figure 2B) and HCV wild-type viruses without reporter (figure 2C,D). To determine the effect of curcumin on HCV replication or virion assembly and release, Huh-7.5 cells were transfected with full-length virus RNA from both genotype 1a and 2a HCVcc, and then increasing doses of curcumin were added 4 h later. Replication efficiency was assessed 48 h after transfection. In parallel, naïve cells were infected with the supernatant from the electroporated and curcumin-treated cells. Curcumin had only a negligible effect on HCV RNA replication of genotype 1a and 2a viruses (see online supplementary figure S1A,C) and none on infectious virion production of both genotypes (see online supplementary figure S1B,D). Together, these data suggest that curcumin is a novel inhibitor of the HCV entry pathway and that this molecule does not affect RNA replication or assembly/release of infectious particles.

Figure 2

Curcumin reduces the infectivity of all hepatitis C virus (HCV) genotypes in a dose-dependent manner. Huh-7.5 cells were inoculated with Renilla luciferase (RLuc) reporter viruses (A) or TN-strain HCVcc (B) in the presence of increasing concentrations of curcumin. The inoculum was removed 4 h later, and monolayers were washed three times with phosphate-buffered saline (PBS) and overlaid with fresh medium containing no curcumin. Infected cells were lysed 2 days later, and reporter virus or TN-strain HCVcc infection were determined by RLuc activity or X-tail quantitative reverse transcriptase PCR. The cell viability was measured by MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) assay. (C) Huh-7.5 cells were inoculated with non-reporter viruses including TN-strain HCVcc in the presence of curcumin. The inoculum was removed 4 h later, and then monolayers were washed three times with PBS and overlaid with fresh medium containing no curcumin. Infected cells were fixed 2 days later and stained for expression of HCV NS5A or core and NS5A in TN-strain HCVcc infection by indirect immunofluorescence (green). Next, the focus forming units (FFUs) were quantified by counting the number of infected cells (D). A representative experiment out of three independent repetitions with SD of the means is shown. DAPI, 4′,6-diamidino-2-phenylindole; DMSO, dimethyl sulfoxide.

Curcumin inhibits cell entry into human hepatoma cells and primary hepatocytes

To further test whether curcumin targets HCV entry, we used HCVpp. These are retroviral capsids with HCV glycoproteins incorporated into the viral envelope. In this model, only the early steps of virus entry (ie, virus binding, uptake and fusion) are HCV specific. In this system, curcumin inhibited cell entry mediated by HCVpp bearing both genotype 1a and 2a envelope glycoproteins, in addition to retroviral particles pseudo-typed with VSV-G and A-MLV envelope glycoproteins (figure 3A). Infectivity of HCVpp carrying E1E2 proteins of the strains H77 (genotype 1a) and J6 (genotype 2a) were similarly reduced, confirming that HCV inhibition by curcumin is not genotype or isolate specific. PHHs more closely resemble the hepatocytes that are the main reservoir for HCV within the infected host. As PHHs do not support robust replication of HCV genomes in vitro, PHHs were infected in the presence or absence of curcumin with HCVpp of strain H77 carrying a green fluorescent protein reporter gene. Entry of HCVpp into PHHs was strongly inhibited by curcumin in a dose-dependent manner (figure 3B). Thus, curcumin is also able to block HCV entry into its natural target cells.

Figure 3

Curcumin inhibits hepatitis C virus (HCV) entry into Huh-7.5 cells and primary human hepatocytes (PHHs). (A) Huh-7.5 cells were infected with pseudo-particles with envelope glycoprotein of vesicular stomatitis virus (VSV-G), amphotropic murine leukaemia virus (A-MLV), HCV genotype 2a (J6) or 1a (H77) envelope glycoprotein in the presence of different concentrations of curcumin during infection. The inoculum was removed 4 h later, and monolayers were washed and then overlaid with fresh medium containing no curcumin. Infected cells were lysed 2 days later, and luciferase activity was determined. A representative experiment of three independent repetitions with SDs of the means is shown. (B) PHHs were infected with HCV genotype 1a pseudo-particles containing a green fluorescent protein (GFP) reporter as described in (A). After overnight incubation, the inoculum was removed, the cells were washed, and fresh medium without curcumin was added. Infected cells were fixed 3 days later, and the GFP-positive cells were determined by fluorescence activated cell sorter (FACS) analysis. A representative experiment of two independent repetitions is shown. DMSO, dimethyl sulfoxide.

Curcumin does not downregulate HCV receptors and impedes HCV cell entry by directly acting on the virus

To test if curcumin treatment affects the expression of known essential HCV (co-)receptors, after 4 h of curcumin treatment, we assessed the expression of CD81 and scavenger receptor class B type 1 (SR-BI) by using fluorescence-activated cell sorter (FACS) analysis. Expression of claudin-1 (CLDN1), occludin (OCLN) and Niemann-Pick C1-Like 1 (NPC1L1) was assessed by western blotting. Expression levels of all tested (co-) receptors were unaltered, indicating that curcumin does not act through their downregulation (figure 4A).

Figure 4

Curcumin treatment does not affect the expression levels of crucial cellular entry factors and inhibits hepatitis C virus efficiently only when present during infection. (A) Huh-7.5 cells were treated with curcumin at the indicated concentrations for 4 h and the levels of hCD81 and hSR-BI surface expression were measured by fluorescence-activated cell sorter (FACS) analysis, whereas hClaudin-1, hOccludin, and hNPC1L1 expression were measured by western blotting. Hep56.1D mouse hepatic cells were used as negative control. (B) Huh-7.5 cells were pre-treated before JcR2a infection, co-treated during infection, or post-treated after infection with various concentrations of curcumin for 4 h. Each step in each given treatment was always followed by cell washing with phosphate-buffered saline to remove non-specific effects of the compound. After each final step, the cells were incubated in fresh medium without curcumin, and Renilla luciferase activity was determined 48 h after infection. A representative experiment of three independent repetitions with SDs of the means is shown. DMSO, dimethyl sulfoxide.

We next tested the mechanism whereby curcumin inhibits HCV entry. We assessed its antiviral activity when administered at different time points during the early phase of infection. We administered curcumin before (pre-treatment), during (co-treatment) or after (post-treatment) the infection of Huh-7.5 cells with reporter virus. Pre-treatment with curcumin for 4 h before virus inoculation did not protect the cells from the infection (figure 4B), confirming that curcumin neither affects the susceptibility of the target cells nor the functionality of crucial HCV (co-)receptors. Curcumin added 4 h after treatment likewise could only mildly inhibit infectivity. In contrast, curcumin treatment during virus inoculation resulted in maximum inhibition. In summary, curcumin appears to directly impair the ability of the virus itself to enter the target cells.

Curcumin reduces the infectivity of HCV particles without affecting virion integrity

HCV particles display a broad-density distribution, which is thought to be the result of a non-homogeneous association of virions with lipoproteins. Therefore, we tested whether the antiviral activity of curcumin is a result of disruption of virion–lipoprotein association or another modification of the viral particle resulting in an altered density profile. We incubated HCVcc particles in the presence of curcumin or THC at 25 µM for 4 h at 37°C, followed by centrifugation through an iodixanol density gradient. The collected fractions were assayed for viral infectivity and HCV core protein. Curcumin, but not THC, significantly reduced HCV infectivity (figure 5A) without major changes in density distribution (figure 5B), indicating that curcumin acts on HCV through a mechanism that does not alter physical integrity or lipoprotein association of the viral particle. Conversely, we also evaluated the effect of curcumin when HCV density fractions were first separated by density gradient centrifugation, followed by individual incubation of each fraction with curcumin and assay for their entry capacity. The antiviral activity of curcumin was more pronounced, especially against virus particles with lower density (figure 5C). Curcumin as a hydrophobic compound probably has better access to interact with the virus particles that are tightly associated with lipids. To further support this theory, we determined the amount of core protein resistant to proteolysis. This assay tests for disruption of the lipid envelope, as core protein enveloped by an intact membrane should be protected from digestion by the protease. Triton X-100 was used as the lysis control. There were no differences in protease digestion of the core between control and curcumin-treated particles (figure 5B), indicating that the antiviral effect of curcumin is not due to viral particle rupture or lysis.

Figure 5

Curcumin inhibits the infectivity of hepatitis C virus (HCV) particles without affecting the density distribution or disrupting the lipid envelope of the particle. HCV Jc1 virions were incubated in the presence of dimethyl sulfoxide (DMSO), curcumin or tetrahydrocurcumin (THC) at 37°C in cell-free conditions. After 4 h, an iodixanol step-gradient centrifugation was performed. HCV infectivity (A) and core protein distribution (B) were determined for each of 10 gradient fractions harvested from the bottom. Values are plotted against the density of the respective fraction, as measured by refractometry. An experiment, representative of three independent repetitions, is shown with SD of the means. (C) HCV Jc1 virions were separated first by density gradient, then similarly harvested as in the previous method, followed by individual incubation of each fraction with curcumin, and then assayed for infectivity. (D) HCV Jc1 virions were exposed to either curcumin or DMSO for 4 h in cell-free conditions and then treated with phosphate-buffered saline (PBS) or proteinase K for 1 h on ice (fraction 1 and 2). As a control, 2.5% Triton X-100 was added to the third fraction in the presence of proteinase K. The abundance of core protein was determined by western blot. A representative experiment of three independent repetitions is shown.

Curcumin impairs the ability of HCV to attach and fuse to target cells by decreasing the fluidity of viral envelopes

Curcumin has been reported to affect membrane fluidity.31 We therefore tested the effects of curcumin and THC on HCV envelope fluidity using the DPH fluorescence polarisation method and cholesterol as a control.31–33 When membrane fluidity decreases—for example, by the addition of cholesterol—the polarisation of DPH fluorescence increases. The addition of curcumin to HCV envelopes increased polarisation to the same level as the addition of cholesterol (figure 6A). In contrast, and consistent with its lack of effect on virion infectivity, THC did not affect membrane fluidity.

Figure 6

Curcumin affects the membrane fluidity of viral envelopes and inhibits hepatitis C virus (HCV) binding and fusion. (A) Membrane fluidity was assessed by 1,6-diphenyl-1,3,5-hexatriene (DPH) fluorescence polarisation. Like cholesterol, curcumin decreases the fluidity of HCV lipid envelopes, shown by the increase in DPH polarisation (average±range; n=2). Data are normalised to the polarisation of DPH in virions exposed to dimethyl sulfoxide (DMSO) vehicle control (set as 1). (B) Binding of HCV virions to target cells after being exposed to curcumin or tetrahydrocurcumin (THC) in the absence (mean±SD; n=4) or presence of serum (mean±SD; n=2). R18-labeled HCV virions were exposed to the test compound for 10 min at 37°C, and then cooled to 4°C before being added to prechilled Huh-7.5 cell monolayers. Cells and virions were then incubated at 4°C for 1 h, then washed three times with cold phosphate-buffered saline. Cells and attached virions were lysed, and lysates were examined for R18 fluorescence. Binding is expressed as a percentage relative to the R18 fluorescence of DMSO vehicle-treated virions after standardisation to the total amount of fluorescence detected for each sample. FBS, fetal bovine serum. (C) Fusion of HCV virions to target cells measured by fluorescence dequenching of HCV exposed to curcumin or THC (average±range; n=2). R18-labelled HCV virions were exposed to DMSO vehicle, curcumin or THC for 10 min at 37°C, and then cooled to 4°C before being added to prechilled Huh7.5 cells. Alternatively, virions were allowed to attach to cells before curcumin treatment. Fusion was triggered by increasing the temperature and lowering the pH.

Virion binding and fusion to target cells require appropriate membrane fluidity, both for the movement of viral glycoproteins to allow higher-affinity binding and for the lipid rearrangements required for fusion. We therefore postulated that the reduction in membrane fluidity leads to inhibition of virion binding and fusion. To test this possibility, R18-labelled HCV virions pre-exposed to curcumin, THC or dimethyl sulfoxide (DMSO) vehicle were adsorbed on to Huh-7.5 cell monolayers at 4°C in the presence or absence of serum. Unbound virions were washed away and the R18 fluorescence bound to cells was quantified. HCV binding to target cells is expressed as a percentage relative to the R18 fluorescence of a DMSO vehicle-treated control. Curcumin inhibited the binding in a dose-dependent manner (figure 6B), with IC50 of 18.2 µM (no FBS) or 13.3 µM (in 5% FBS). We then tested the effects of curcumin on the fusion of HCV to Huh-7.5 cells. HCV JFH-1 virions labelled at self-quenching concentrations of R18 were exposed to curcumin, THC or DMSO vehicle before mixing with Huh-7.5 cells. Fluorescence was dequenched by ∼18% for HCV virions treated with DMSO vehicle, but by only 8% for HCV virions treated with 20 μM or 200 μM curcumin, in the range of the background dequenching at neutral pH in these assays. Treatment with curcumin after virions were already attached to cells at 4°C also inhibited fusion to background levels, consistent with its effects on fluidity (figure 6C). THC, in contrast, did not inhibit HCV fusion, although it somewhat delayed it.

Curcumin inhibits HCV cell-to-cell transmission

HCV can also be transmitted by cell-to-cell spread. This transmission is probably important in vivo and was reported to be refractory to neutralisation by E2 monoclonal antibodies and to occur in a CD81-independent manner.34 ,35 Therefore, we tested whether curcumin would also block cell-to-cell transmission of HCV in tissue culture. Huh-7.5 cells were infected with HCVcc (H77/JFH1 chimera) at a high multiplicity of infection and confirmed that >99% of the cells were HCV positive by immunofluorescence against NS5A (data not shown). These ‘donor cells’ (marked cells with green cytoplasm but without red nuclei in figure 7A) were co-cultured with Huh-7.5 ‘acceptor’ cells, which carry a reporter protein allowing detection of HCV infection by relocation of a red fluorescent protein (tagRFP) from a mitochondrial (ie, uninfected) to a nuclear (ie, infected) localisation.36 Co-cultured cells were overlaid with medium containing increasing amounts of curcumin and 1% agarose to prevent cell-free diffusion of virus. Cell-to-cell transmission was inhibited by curcumin (figure 7A,B), indicating that curcumin supplementation also prevents HCV cell-to-cell spread in vitro.

Figure 7

Curcumin inhibits hepatitis C virus (HCV) cell-to-cell spread. A complete H77/JFH1-infected population of Huh-7.5 donor cells were co-cultured with uninfected Huh-7.5 target cells harbouring a tagRFP-NLS-IPS reporter in the presence or absence of curcumin, under a 1% agarose overlay. Red fluorescence is localised in the cytoplasm in uninfected cells (tagRFP). Upon HCV infection, tagRFP-NLS-IPS is cleaved and the red fluorescent reporter relocates to the nucleus. Nuclei were co-stained with DAPI (blue) and the presence of HCV polyproteins was proved by NS5A indirect immunofluorescence staining (green). The number of tagRFP-positive nuclei and the percentage of infected cells in increasing amounts of curcumin are presented as evaluated at 120 h of co-culture. A representative experiment of three independent repetitions with mean values and ranges are presented. DAPI, 4′,6-diamidino-2-phenylindole; DMSO, dimethyl sulfoxide; RFP, red fluorescent protein.

The combination of curcumin with other antivirals promotes better viral clearance

We next tested whether curcumin could be combined with other commercially available antiviral agents. We used PEG-IFN-α, boceprevir (inhibitor of HCV protease, NS3/4A), ciclosporin A (inhibitor of HCV host cofactor, cyclophilin A) and EGCG (another natural compound that also inhibits HCV entry). We tested these antiviral agents in the presence of curcumin and found that the addition of curcumin can increase the efficacy of these agents to inhibit HCV infection. In detail, the combination of curcumin at 15 µM increased the antiviral effect about 10-fold compared with a single treatment of those antiviral agents (figure 8A–D). In the case of boceprevir combined with curcumin, HCV infection was already cleared to background level at 1 µM boceprevir. These data indicate that curcumin could be used efficiently in combination therapy with other antiviral agents without any adverse effect.

Figure 8

Curcumin is effective in combination with commercial hepatitis C virus inhibitors. Huh-7.5 cells were inoculated with JcR2a reporter viruses in the presence of the indicated doses of curcumin. The inoculum was removed 4 h later and monolayers were washed three times with phosphate-buffered saline and then overlaid with fresh medium containing the indicated doses of (A) pegylated interferon-α (PEG-IFNα), (B) protease inhibitor boceprevir or (C) cyclosporin A. In the case of epigallocatechin-3-gallate (EGCG), co-treatment was conducted (D). Infected cells were lysed 48 h later, and Renilla luciferase activity was determined. A representative experiment of two independent repetitions with SD of the mean is shown. DMSO, dimethyl sulfoxide.

In vivo studies of curcumin in mice

To assess the potency of curcumin in an in vivo setting, we conducted experiments using genetically humanised mice for HCV infection, which allows the study of viral entry only.37 The experimental setup was the following: 1 day before HCV infection, human entry receptors required for HCV entry were delivered into 18 Rosa26-Fluc mice by intravenous injection of recombinant adenoviruses encoding the four essential HCV receptors. Subsequently, we distributed the mice into three groups: untreated-control (n=8), carrier (n=5) or curcumin-treated (n=5). The mice in treated groups were pre-treated with either carrier (3% ethanol in PBS) or 20 mg/kg curcumin in carrier, 1 h before and during 5×106 TCID50 CRE-recombinase-expressing HCVcc (HCV-CRE) infection. Bioluminescence was measured 96 h later. No significant differences among groups were observed (figure 9A). This result suggests that curcumin's antiviral activity by modulating membrane fluidity may only be achieved when the compound is present at the organ target at higher concentration.

Figure 9

(A) Effect of curcumin treatment of hepatitis C virus (HCV) infection in genetically humanised mice. The Rosa26-Fluc mice transiently expressing all four human entry factors were pre-treated with either carrier (3% ethanol in phosphate-buffered saline) or 20 mg/kg curcumin in carrier, 1 h before and during 5×106 TCID50 HCV-CRE infection. Bioluminescence was measured 96 h later. (B) Pilot pharmacokinetic study of nano-formulated curcumin. BALB/c mice were distributed into four groups (n=2) with different delivery methods (oral or intraperitoneal (IP)) and formulations (normal or nanocrystal-curcumin). Blood samples were taken at 0, 0.5, 1, 2, 4 and 8 h after the delivery and analysed.

It has been reported that curcumin has a low bioavailability and rapid clearance. In an attempt to study if the bioavailability of curcumin could be enhanced through its formulation, we also generated curcumin nanocrystals and performed a pilot pharmacokinetic study. As can be seen in figure 9B, the nano-formulated curcumin has a higher plasma concentration, especially at the 0.5 h time point, reaching 95.38±14.99 ng/mL with oral delivery or 126.06±14.17 ng/mL with intraperitoneal delivery compared with normal curcumin. Two hours after the delivery, detectable amounts of curcumin given in nano-formulation can still be observed in the blood plasma.

Discussion

The entry step of the HCV lifecycle is critical for initiation, maintenance and dissemination of viral infection in vivo, and represents an attractive target for therapeutic intervention. Most currently available anti-HCV treatments target later stages of the viral lifecycle, such as viral RNA replication, and are intended for use in chronically infected patients. Consequently, the discovery of novel antivirals to block HCV cell entry is an area of intense research, with the aim of restricting universal re-infection of the donor liver by circulating virions in the setting of liver transplantation for HCV-associated end stage liver disease. In this study, we identified curcumin as a novel entry inhibitor of all major HCV genotypes and of cell-to-cell spread between neighbouring cells. Viral entry into both hepatoma cell lines as well as PHHs was inhibited by the presence of the compound without affecting the viability of the cells. Using reporter replicon, it has been reported that curcumin can inhibit HCV subgenomic RNA replication.38 ,39 However, the activity of the reporter gene was inhibited by only 50% at 15 µM, and RNA levels reduced to only ∼55% at 20 µM. Consistently, we observed only minor inhibitory effects on both genotype 1 and 2 full-length HCV RNA replication (see online supplementary figure S1) compared with >90% inhibition of entry at 20 µM. Curcumin also did not affect viral assembly/release of both genotypes.

Several other agents have been reported to inhibit HCV entry through various mechanisms.8 ,11 ,40 ,41 Furthermore, antibodies targeting the glycoproteins or cellular receptors, such as CD81, SR-BI and CLDN1, have been shown to block viral entry and control spread in vitro and in vivo.42 A recent study by Lupberger et al13 identified the epidermal growth factor receptor (EGF-R) as a host factor required for HCV entry and suggested the EGF-R inhibitor, erlotinib, as a possible anti-HCV agent. The discovery of the new HCV entry receptor, Niemann-Pick C1-like 1 (NPC1L1), has also suggested the NPC1L1 antagonist, ezetimibe, as a potential blocker for HCV uptake.43 The small molecule compound, ITX-5061, was reported to disrupt the interaction of E2 and SR-BI and is currently entering a phase Ib study in humans.44 Compared with the HCV entry inhibitors mentioned above, curcumin remains a highly attractive antiviral candidate because of its broad potency as a chemopreventive agent (especially against various gastrointestinal and liver cancers), high safety profile in humans, and substantial number of formulation, preclinical and clinical studies.17 ,21 ,22 ,45 ,46 Moreover, there is an increasing demand for antivirals with greater accessibility and affordability in developing countries, which suffer a substantial healthcare burden from HCV infection. Curcumin, used worldwide as a spice and food supplement, is an attractive compound for future clinical development and investigation as a low-cost antiviral.9 ,10

In this study, we discovered the mechanisms whereby curcumin inhibits entry. As indicated by the NMR data obtained by Barry et al31 using model synthetic lipid bilayers composed of a single phospholipid, we postulated that the property of curcumin as a membrane fluidity modulator is important for its antiviral activity, to the point that it over-rides the potential pro-fusiogenic effects of its ability to promote negative curvature. Using a DPH polarisation assay, we found that curcumin also reduces the fluidity of the HCV envelope. The viral envelope, which is mainly derived from portions of host phospholipid membranes, is a liquid or disordered under physiological conditions. The fluidity of the membrane results from anisotropic rotation of phospholipid acyl chains and the flip-flop movement of membrane molecules.47 ,48 Under some circumstances, such as during excessive cholesterol incorporation, the membrane bilayer liquid-disordered phase can switch to the liquid-ordered (rigid) phase.49 ,50 Curcumin rigidifies membranes by deep penetration into the membrane in a transbilayer orientation, anchored by hydrogen bonding to the phosphate group of lipids in a manner analogous to cholesterol. However, curcumin also promotes the formation of the highly curved inverted hexagonal phase by lipid bilayers, and negative curvature favours membrane fusion.31 Using self-quenched-fluorescence-labelled HCV virions, we demonstrated that curcumin actually inhibits HCV fusion and also binding. We conclude that the decrease in envelope fluidity over-rides its potential pro-fusiogenic properties. Other small compounds sharing similar properties have also been reported to inhibit HCV or other viruses.49 ,51 We passaged the highly infectious Jc-1 clone in the presence of curcumin for several weeks. However, no mutations conferring resistance to curcumin were apparent in viruses isolated from these experiments (data not shown).

Curcumin is a natural compound isolated from plants. Several phase I clinical trials indicate that curcumin is well tolerated in doses as high as 12 g per day.17 Despite its broad spectrum of potentially beneficial pharmacological activities, curcumin has not yet been approved as a therapeutic agent, mainly because of its low bioavailability at physiologically achievable concentrations. In this study, we also tested the possibility of using curcumin in vivo. However, because of its low bioavailability, no antiviral activity was observed and further improvements in its bioavailability would be necessary for its application in vivo. Many studies have attempted to improve the bioavailability of curcumin using novel delivery strategies, such as nanoparticles, liposomes and defined phospholipid complexes. Nanoparticle formulation of curcumin, as has been shown in this study, can significantly increase its bioavailability. Another previous report has also shown that nanocrystal and nanocarrier systems of curcumin are promising delivery systems, which have substantially improved its bioavailability.52 Incorporation of curcumin into liposomes also strongly enhances its stability and may be even more effective.31 The administration of a curcumin adjuvant, piperine, has also been shown to increase the bioavailability of curcumin in a human study.53 In addition, curcumin analogues with enhanced biological activity and bioavailability, with no increased toxicity, have been synthesised.54

In summary, turmeric curcumin inhibits HCV entry independently of the infecting genotype in vitro, in addition to demonstrating entry inhibition to PHHs. We show here that curcumin affects virion membrane fluidity without disrupting virion integrity, and hence impairs HCV binding and fusion ability. This novel inhibitor may provide a new low-cost, well-tolerated addition to conventional HCV therapies for clearing chronic infection. Moreover, the ability of curcumin to block the entry step of the viral lifecycle makes it an attractive new candidate for facilitating prevention of graft re-infection during liver transplantation of chronically infected HCV patients.

Acknowledgments

We are grateful to Takaji Wakita for the JFH1 isolate and Jens Bukh for the TN and all chimeric-HCV isolates. We thank Gary Eitzen for the use of his fluorimeter for the binding, fusion and polarisation assays. We also thank Benno Wölk and all members of the Institute of Experimental Virology, Twincore, for helpful support, suggestions and discussions. We are grateful to Bridget Donovan, Tamar Friling, Rachael Labitt and Kevin Vega for excellent technical help.

References

View Abstract

Supplementary materials

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

    Files in this Data Supplement:

Footnotes

  • Contributors Conceived and designed the experiments: AK, LMS, TP and ES. Performed the experiments: AK, CCC, HR, AF, SP, PB, RJPB, DB, JS and AP. Analysed the data: AK, CCC, LMS, MO, PM, TP and ES. Contributed reagents/materials/analysis tools: HR, MO, PM and CMR. Wrote the manuscript: AK, LMS and ES.

  • Funding TP was supported by grants by the Helmholtz Association SO-024 and the DFG (PI 734/2-1 and SFB 900, Teilprojekt A6). ES was supported by an intramural young investigator award from the Helmholtz Centre for Infection Research and the DFG (STE 1954/1-1). PM was supported by grants from the Research Foundation—Flanders (#3G052112), the Ghent University (GOA #01G01712) and the Belgian state (IAP-HEPRO-2). CCC was supported by AI-HS and NSERC. LMS is a BWF Investigator in the Pathogenesis of Infectious Disease. This work was supported in part by the BWF and the CIHR. AP and CMR were supported by US National Institute for Allergy and Infectious Disease grants R01AI072613 and R01AI099284. AP is a recipient of a Liver Scholar Award from the American Liver Foundation.

  • Competing interests TP has received consulting fees from Biotest AG and from Janssen Global Services, LLC, and CMR has equity in Apath, LLC, which holds commercial licenses for the Huh-7.5 cell line, HCV cell culture system, and the fluorescent cell-based reporter system used to detect HCV infection. AP serves as a consultant for Apath, LLC.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Data sharing statement The curcumin adaptation study is mentioned as data not shown in the discussion section. These unpublished data are available to everybody on request.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Linked Articles