Direct binding of a hepatitis C virus inhibitor to the viral capsid protein

Abstract

Over 130 million people are infected chronically with hepatitis C virus (HCV), which, together with HBV, is the leading cause of liver disease. Novel small molecule inhibitors of Hepatitis C virus (HCV) are needed to complement or replace current treatments based on pegylated interferon and ribavirin, which are only partially successful and plagued with side-effects. Assembly of the virion is initiated by the oligomerization of core, the capsid protein, followed by the interaction with NS5A and other HCV proteins. By screening for inhibitors of core dimerization, we previously discovered peptides and drug-like compounds that disrupt interactions between core and other HCV proteins, NS3 and NS5A, and block HCV production. Here we report that a biotinylated derivative of SL209, a prototype small molecule inhibitor of core dimerization (IC(50) of 2.80 µM) that inhibits HCV production with an EC(50) of 3.20 µM, is capable of penetrating HCV-infected cells and tracking with core. Interaction between the inhibitors, core and other viral proteins was demonstrated by SL209-mediated affinity-isolation of HCV proteins from lysates of infected cells, or of the corresponding recombinant HCV proteins. SL209-like inhibitors of HCV core may form the basis of novel treatments of Hepatitis C in combination with other target-specific HCV drugs such as inhibitors of the NS3 protease, the NS5B polymerase, or the NS5A regulatory protein. More generally, our work supports the hypothesis that inhibitors of viral capsid formation might constitute a new class of potent antiviral agents, as was recently also shown for HIV capsid inhibitors.

Figure 1. Scheme for synthesis of Biotinylated SL209.

Figure 2. Inhibition of core dimerization and of virus production by SL209-biotin and evidence for direct binding to core protein.

In an AlphaScreen assay, glutathione-coated donor beads and anti-flag antibody coated acceptor beads (20 µg/ml each) were used to monitor the dimerization of GST- and Flag- tagged core106 proteins. GST-core106 (GC106) and Flag-core106 (FC106) were used at 150 nM each and core106 was used at 1 µM as a reference inhibitor. A: Structures of SL201, SL209, SL231, and SL209-biotin. SL201, is a 513 Da small molecule inhibitor originally identified to inhibit HCV core dimerization and virus production. SL209, is a SAR analogue of SL201. SL231, is a dimer of SL209. SL209-biotin is a biotinlyated derivative of SL209. “*” indicates that structures of SL201, SL209, and SL231 have been previously published in Wei et al 2009 , Strosberg et al 2010 , and Ni et al 2011 . B: Levels of inhibition. Core106 inhibited 91% and small compounds inhibitors: SL201, SL209, SL231 and SL209-biotin used at 15 µM inhibited respectively 66%, 74%, 83%, and 75% of core dimerization. C: Direct binding to GST-core106 (GC106) or Flag-core106 (FC106). In a novel AlphaScreen format GC106/SL209-biotin and FC106/SL209-biotin were mixed in 1∶1000 ratio and incubated. Streptavidin donor beads and glutathione coated acceptor beads at 20 µg/ml were used in the detection of the binding. Free SL209 at 50 µM inhibited 83% of SL209-biotin binding to GC106 and 77% of SL209-biotin binding to FC106. D: Dose response analyses. Inhibition levels were analyzed in a dose response format. The compounds were dosed from 160 µM to 150 pM. The IC₅₀ values for SL209 (right panel) and SL209-biotin (left panel) were calculated as 3.7 µM and 2.8 µM using GraphPad Prism. E: Inhibition of HCV production. Inhibition of HCV production in Huh-7.5 cells by SL209 and SL209-biotin was analyzed by adding serially diluted the compounds (individually) and virus onto naïve Huh-7.5 cells. The supernatants of the cells after 3 days of culture were removed from the initial culture and added to naïve cells cultured for another 3 days. RNA was purified from lysed cells and analyzed by Real-Time RT-PCR. EC₅₀ values were calculated to be 3.2 µM and 4.8 µM for SL209 (right panel) and SL209-biotin (left panel), respectively. EC₉₀ values were calculated to be 33.9 µM and 47.3 µM for SL209 (right panel) and SL209-biotin (left panel), respectively.

Figure 3. Affinity isolation on SL209-biotin captures native and recombinant HCV core and core-associated proteins.

SL209-biotin was immobilized on streptavidin agarose beads and incubated with core106 (10 µg) or core169 (10 µg) proteins or lysates of Huh-7.5 cells infected with HCV J6/JFH-1. The retained proteins were examined by immuno-blotting using anti-core, NS3, NS5A or NS5B antibodies. One-twentieth of the cell lysate used in the assay was immuno-blotted for all proteins as the input control. As a control, the lysates were incubated with Biotin immobilized on streptavidin agarose beads. As an additional control, a non-specific protein, CTD (C-terminal domain of HIV capsid protein, p24) was incubated with immobilized SL209-biotin. A: SL209-biotin captures/co-precipitates recombinant core106 (top panel) and core169 proteins (middle panel) and does not capture non-specific control protein CTD (bottom panel). B: SL209-biotin captures/co-precipitates core, NS3, NS5A, and NS5B from HCV-infected Huh-7.5 cells, but C: not from replicon containing Huh-7.5 cell lysates. D: SL209-biotin does not co-precipitate recombinant NS3 and NS5A proteins. E: Core protein added to replicon-containing Huh-7.5 cell lysates rescues co-precipitation of NS3 and NS5A. F: CTD protein added to replicon-containing Huh-7.5 cell lysates does not rescue the co-precipitation of NS3 and NS5A.

Figure 4. SL209 strongly reduces number of HCV-infected Huh-7.5 cells.

Huh-7.5 cells seeded into 96-well plate were infected with HCV in the presence of increasing amounts of SL209 and incubated for 72 hours. Cells were fixed with methanol and immunohistochemistry was performed to stain cells for expression of the HCV NS5A protein with anti-NS5A (9E10) antibodies followed by rabbit anti-mouse HRP-conjugated secondary antibody. We show here three representative fields containing large number of infected cells of which treatment with SL209 progressively reduces numbers of cells stained with the anti-NS5A antibody. The left field (“no compound”) was treated with vehicle, and the following two fields were treated with 3.125 µM and 25 µM SL209. Wells treated with 25 µM of SL209 showed a substantial reduction in HCV-infected cells. The cells were counter stained with hematoxylin 2 to visualize the nuclei. Magnification is 10×. The data are representative of 4 wells in each experiment and 3 different experiments.

Figure 5. Core is required for labeling by SL209-biotin.

Huh-7.5 cells seeded into chamber slides were infected with HCV in the presence of increasing amounts of SL209-biotin and incubated for 72 hours. Infected cells were probed for expression of core with anti-core antibody and replicon cells were probed for both core and HCV NS5A protein respectively with anti-core (R-308) anti-NS5A (9E10) antibodies. Alexa555-coupled secondary antibody was used to detect bound specific antibody. Huh-7.5 cells are stained with DAPI to localize nuclei, and studied by confocal microscopy. In immunofluorescence studies of HCV-infected cells, treated with 2 µM of SL209-biotin (below the EC₅₀ value of 4.8 µM) and streptavidin-Alexa488 and labeled for core, SL209-biotin can be seen to penetrate the infected cells ( Fig. 5A ). No such labeling could be observed when compound is not present (Fig. 5B) , in uninfected cells ( Fig. 5C ), or in cells expressing HCV subgenomic replicon which produce NS5A but not core ( Fig. 5D ). Magnification was 10× with 3× optical zoom and the images were cropped to display the field of interest.