Identification of structurally re-engineered rocaglates as inhibitors against hepatitis E virus replication

Abstract

Hepatitis E virus (HEV) infections are a leading cause of acute viral hepatitis in humans and pose a considerable threat to public health. Current standard of care treatment is limited to the off-label use of nucleoside-analog ribavirin (RBV) and PEGylated interferon-α, both of which are associated with significant side effects and provide limited efficacy. In the past few years, a promising natural product compound class of eukaryotic initiation factor 4A (eIF4A) inhibitors (translation initiation inhibitors), called rocaglates, were identified as antiviral agents against RNA virus infections. In the present study, we evaluated a total of 205 synthetic rocaglate derivatives from the BU-CMD compound library for their antiviral properties against HEV. At least eleven compounds showed inhibitory activities against the HEV genotype 3 (HEV-3) subgenomic replicon below 30 nM (EC₅₀ value) as determined by Gaussia luciferase assay. Three amidino-rocaglates (ADRs) (CMLD012073, CMLD012118, and CMLD012612) possessed antiviral activity against HEV with EC₅₀ values between 1 and 9 nM. In addition, these three selected compounds inhibited subgenomic replicons of different genotypes (HEV-1 [Sar55], wild boar HEV-3 [83-2] and human HEV-3 [p6]) in a dose-dependent manner and at low nanomolar concentrations. Furthermore, tested ADRs tend to be better tolerated in primary hepatocytes than hepatoma cancer cell lines and combination treatment of CMLD012118 with RBV and interferon-α (IFN-α) showed that CMLD012118 acts additive to RBV and IFN-α treatment. In conclusion, our results indicate that ADRs, especially CMLD012073, CMLD012118, and CMLD012612 may prove to be potential therapeutic candidates for the treatment of HEV infections and may contribute to the discovery of pan-genotypic inhibitors in the future.

Keywords: Amidino-rocaglates; Antiviral treatment; Antivirals; Hepatitis E virus; elF4A inhibitors.

Figure 1:. Screening of rocaglate derivatives using HEV subgenomic replicon.

(A) Schematic representation of assay setup: HEVp6-Gluc was delivered to HepG2 cells via electroporation. Four hours after the transfection, cells were treated with 25 nM of each rocaglate derivative. (B) Supernatants were sampled after 24 and 48 h. HEVp6 replication was measured via reporter luciferase read-out (relative light units [RLU]) and normalized to the respective vehicle treated control (DMSO) while cell viability was monitored via MTT assay. Silvestrol (orange dots) was included as a reference. Candidates selected for downstream analysis are highlighted as purple dots. (C) The screening provided a total of 7 candidates (purple dots) with antiviral properties against HEV and were selected based on fulfillment of both selection criteria at 24 h and 48 h for further analysis. Derivatives excluded from further analysis are represented as grey dots. Error bars indicate standard error of the mean, n = 3.

Figure 2:. Comparison of EC₅₀, CC₅₀ and SI of eleven racemic rocaglates against HEVp6 replication.

HepG2 cells were transfected with HEVp6-Gluc replicon and treated with drugs at concentrations ranging from 0.39 nM to 100 nM. Dose-response curves of eleven derivatives were adjusted to a non-linear fit regression model and calculated with a four-parameter logistic curve from three experiments with three replicates. For each derivative EC₅₀, CC₅₀ and SI values are plotted and ordered according to SI values. Based on selective indices at 48 h, 3 derivatives (black arrows) were selected to test for pan-genotypic inhibition.

Figure 3:. Pan-genotypic inhibition of HEV replication by CMLD012073, CMLD012118, and CMLD012612.

HEV subgenomic replicons HEVp6-Gluc, HEV83–2-Gluc, HEVSar55/S17-Fluc were electroporated into HepG2 cells. Four hours post transfection, cells were treated with (A) CMLD012073, (B) CMLD012118 and (C) CMLD012612 at concentrations ranging from 0.39 nM to 100 nM for 24 h and 48 h. Depicted are non-linear fit response curves representative of three experiments with three replicates for HEVp6 (dark purple lines), HEV83–2 (bright purple lines) and Sar55/S17 (orange lines). Cell viability was monitored by MTT assay (black lines). EC₅₀ and CC₅₀ were calculated using GraphPad Prism 8 software. Error bars indicate standard error of the mean, n = 3.

Figure 4:. Cell viability of amidino-rocaglate treated primary and hepatoma cells.

PPHs and HepG2 cells were treated with (A) CMLD012073, (B) CMLD012118 and (C) CMLD012612 at concentrations ranging from 0.032 nM to 100 nM for 24 h (grey lines), 48 h (bright purple lines) and 72 h (dark purple lines). Depicted are non-linear fit response curves representative of two experiments with two replicates. Cell viability was monitored by MTT assay. CC₅₀ values were calculated using GraphPad Prism 8 software and 72 h values were plotted. n = 2 (except CMLD012118 treatment; n = 1).

Figure 5:. Combination treatment of CMLD012118 and RBV on HEV replication.

Representation the dose-dependent inhibition of transfected HEV replicons after RBV/IFN-α combination treatment with CMLD012118 for 24h and 48 h. Compounds were mixed at different ratios (CMLD012118: 0.39 – 100 nM, RBV: 3.13 – 100 μM; IFN: 31.25–1000 IU/mL). HEV replication was measured via reporter luciferase and normalized to DMSO treated control. Data were analyzed according to the Bliss independence model (Prichard and Shipman, 1990) and plotted as three-dimensional differential surface plots based on three technical replicates. Note that synergistic drug interactions appear as a peak above the plane. Conversely, antagonistic interactions appear as a pit in the plane with a negative value.