Suramin Inhibits Chikungunya Virus Replication by Interacting with Virions and Blocking the Early Steps of Infection

Abstract

Chikungunya virus (CHIKV) is a mosquito-transmitted alphavirus that can cause a debilitating disease that is primarily characterized by persistent joint pain. CHIKV has been emerging globally, while neither a vaccine nor antiviral medication is available. The anti-parasitic drug suramin was previously shown to inhibit CHIKV replication. In this study we aimed to obtain more detailed insight into its mechanism of action. We found that suramin interacts with virions and can inhibit virus binding to cells. It also appeared to inhibit post-attachment steps of the infection process, likely by preventing conformational changes of the envelope glycoproteins required for fusion and the progression of infection. Suramin-resistant CHIKV strains were selected and genotyping and reverse genetics experiments indicated that mutations in E2 were responsible for resistance. The substitutions N5R and H18Q were reverse engineered in the E2 glycoprotein in order to understand their role in resistance. The binding of suramin-resistant viruses with these two E2 mutations was inhibited by suramin like that of wild-type virus, but they appeared to be able to overcome the post-attachment inhibitory effect of suramin. Conversely, a virus with a G82R mutation in E2 (implicated in attenuation of vaccine strain 181/25), which renders it dependent on the interaction with heparan sulfate for entry, was more sensitive to suramin than wild-type virus. Using molecular modelling studies, we predicted the potential suramin binding sites on the mature spikes of the chikungunya virion. We conclude that suramin interferes with CHIKV entry by interacting with the E2 envelope protein, which inhibits attachment and also interferes with conformational changes required for fusion.

Keywords: CHIKV; E2 envelope protein; alphavirus; antiviral; attachment; drug repurposing; fusion; suramin.

Figure A1

Docking grids used in molecular modeling of the trimeric complex of E1-E2 heterodimers. 36 Å docking grids (inner-box 10 Å and outer-box 46 Å) were prepared as described in the text of Appendix A.

Figure 1

The effect of suramin on the early events of Chikungunya virus (CHIKV) infection (a) The binding of ³⁵S-labelled CHIKV (1 × 10⁴ CPM) to Vero E6 cells in the presence or absence of suramin was determined at 4 °C by scintillation counting of remaining radioactivity in cellular lysates obtained after extensive washing (average +/− SD; n = 3). (b) Binding of fluorescently (DiD)-labeled CHIKV to suramin-treated BS-C-1 cells, analyzed by fluorescent microscopy, in the presence of increasing compound concentrations. (c) Fusion of pyrene-labeled CHIKV in a bulk fusion assay with liposomes, triggered by lowering the pH, in the presence of increasing suramin concentrations (n = 5 and 3, for untreated and treated samples, respectively). (d) Binding of ³H-labeled suramin (5 × 10⁵ CPM) after a 1-h incubation at 37 °C to CHIKV (purified virus was used to exclude interference by serum proteins). The control used in this assay was culture medium from uninfected cells that was treated the same way as when purifying virus (n = 3). The data represent the means ± the SD and significant differences are indicated with * (**** p < 0.001, *** p < 0.005, ** p < 0.01, * p < 0.05 and ns as not significant).

Figure 2

Characterization and suramin sensitivity of reverse engineered CHIKV variants (a) Mutations identified in suramin-resistant CHIKV mutants (Table 1) were reverse-engineered (individually or in combinations) into infectious cDNA clone CHIKV LS3. Plaque morphology (in the absence of suramin) and EC₅₀ (mean, n = 8) for suramin as determined by CPE reduction assay (curves in graph above table) are shown for each of the recombinant viruses. The values were determined from two independent experiments performed in quadruplicate. (b,c) Replication kinetics of CHIKV mutants S4, S5, S9 and wt virus were compared during infection of Vero E6 cells in the absence (b) or presence of 0.2 mM suramin (c). At several time-points p.i., culture supernatants were harvested and infectious virus titers were determined by plaque assay (n = 2). (d) Side-by-side comparison of the 36 h p.i. titers of various mutants and wt virus grown in the absence (N.T., not treated) or presence of 0.2 mM suramin (n = 2). All experiments were performed in Vero E6 cells and the data represent mean ± the SD.

Figure 3

The effect of E2 mutations on suramin-resistance and the early steps of infection (a) Virus uptake and infectivity were determined based on a PRNT-like assay. Approx. 100 PFU of wt CHIKV and mutants S7, S8, S9 and G82R were incubated with Vero E6 cells for 1h in the presence or absence of increasing suramin concentrations. Afterwards the inoculum was removed, the monolayers were washed with PBS and overlay medium was added. After a three-day incubation, the cells were fixed and plaques were stained and counted. (b) The plaque number reduction assay was used to analyze synchronized attachment of wt CHIKV and variant S9 at 4 °C, in the presence and absence of suramin. After binding for 1 h in the cold, the inoculum and suramin were removed and replaced with overlay medium without suramin, and the rest of the procedure was performed as described under (a). The data represent the means ± the SD (n = 3).

Figure 4

Molecular docking of suramin to a mature CHIKV spike (a) The top view of a full wt E1-E2 heterotrimer (PDB ID 3J2W). The E2 proteins are represented as blue ribbon, the E1 as purple ribbon and the fusion loop as orange ribbon; the N5 and H18 residues are represented with carbon atoms in blue and green, respectively, and residue G82 with carbon atoms in red, belong to E2. Suramin is represented in yellow (3 molecules, carbon atoms and molecular surface). The inset (black rectangle) shows a clearer magnified view of the spike core and the positioning of suramin with respect to the residues N5, H18 and G82. (b) Electrostatic potential (Coulombic surface coloring) of the heterotrimer of wt CHIKV. The black rectangle marks the N-terminal domain of one E2 protein, where the positive charges are found. (c) Electrostatic potential (Coulombic surface coloring) of the heterotrimer of the N5R/H18Q mutant, CHIKV S9. The black rectangle highlights the N-terminal domain of E2 showing an increase in positive charges (blue molecular surface). For presentation purposes, the transmembrane and C-terminal segments of the E1 and E2, which interact with capsid proteins seen in (a), were removed.