Preclinical evaluation of GW406108X, a broad-spectrum therapeutic blocking ULK1/2-mediated autophagy to inhibit enterovirus replication

Figures

• 

  Figure 1. Results from primary screen of EV-D68 on H1299 and RD cells. (A) Example images obtained from the primary screens, where immunofluorescence staining was performed to detect EV-D68 VP1 protein (green) and cell nuclei (blue, DAPI). Scale bars represent 200 μmol/L. (B) Scatterplot of results from the primary screen. A total of 7986 compounds were screened; compounds were considered to be hits if the normalised infection rate below 0.5 (black dotted line). Top 50 hits for each screen are shown in Tables S1 and S2.

• 

  Figure 2. Cytotoxicity and post-infection assays for the seven hits chosen for validation. (A) Cytotoxicity assay for seven compounds, where cell viability was normalised to that of the 0.1% DMSO-treated condition, with calculated CC50 values. (B) Post-infection assay for seven hits, carried out at four non-toxic concentrations, with calculated EC50 values. Triplicates were performed with error bars plotted to show the mean and standard deviation. Results were evaluated with one-way ANOVA and Dunnett’s test for statistical significance. ns signifies P ≥ 0.05, ∗P < 0.05, ∗∗P < 0.01, ∗∗∗P < 0.001.

• 

  Figure 3. Time-of-addition (TOA), time-of-removal (TOR), and entry bypass assays for GW406108X on EV-D68 infection. (A) In TOA, after viral infection, cells were incubated in DMEM (2% FCS) media. At the indicated timepoints, 10 μmol/L of GW406108X was added. In TOR, cells were incubated in 10 μmol/L of GW406108X instead, with wells replaced with fresh media at the same timepoints. All samples were harvested at 12 h.p.i. (B) Entry bypass assay where EV-D68 viral RNA was extracted from live virus and transfected directly into cells (100 ng/well). After 4 h of transfection, the transfection mixture was replaced with GW406108X and 10 μmol/L pleconaril as a negative control. (C) Densitometry analysis with viral protein band intensities normalised to that of β-actin. (D) Representative Western blot on EV-D68-infected cell lysate treated with GW406108X. β-Actin was detected as the housekeeping control. (E) Fold change for strand-specific qPCR calculated via the delta-delta Ct method. Triplicates were performed with error bars plotted to show the mean and standard deviation. Results were evaluated with one-way ANOVA and Dunnett’s test for statistical significance. ns signifies P > 0.05, ∗P ≤ 0.05,…

• 

  Figure 4. Schematic of the construction of replication-competent and replication-defective viral RNA replicons. The pcDNA3.1+ vector was used as the cloning backbone. Three DNA fragments were generated by extension PCR and assembled using overlap PCR, followed by amplification through purification PCR to produce the full-length insert. This insert was cloned into the pcDNA3.1+ vector using the HindIII and SwaI restriction sites. In the replication-defective replicon, the red region indicates a partial deletion within the 3D coding sequence, corresponding to the removal of the last 162 nucleotides, which renders the RNA replication function inactive. Primer sequences are provided in the Materials and Methods (Section 2.13). Created with BioRender.com.

• 

  Figure 5. GW406108X likely targets a host factor involved in viral genomic RNA replication. H1299 cells were transfected with either the (A) viral RNA replication-competent or (B) viral RNA replication-defective plasmid DNA at 50 ng/well. At 24 h post-transfection, the transfection mixture was replaced with GW406108X, 0.1% DMSO, guanidine chloride (GuHCl), or cycloheximide (CHX). Luminescence levels were read after 12 h. (C) Cells were infected with EV-D68 (P0) and treated with either the vehicle control (0.1% DMSO or 1–2.5 μmol/L of GW406108X. At 12 h.p.i., the virus was harvested (P1) and used to infected RD cells again at MOI 1, with either vehicle control (0.1% DMSO) GW406108X treatment. In this way, the same virus was serially passaged in the presence of increasing concentrations of GW406108X, for 18 passages. Plaque assays were performed to quantify viral titres. (D) Post-infection assay demonstrating GW406108X sensitivity of passage 18 viral pools. Passage 18 viral pools were subjected to post-infection treatment with GW406108X and compared to wildtype P0 virus. Triplicates were performed with error bars plotted to show the mean and standard deviation. Results were evaluated with…

• 

  Figure 6. Evaluation of the toxicity and efficiency of siRNA knockdown of host targets of GW406108X. (A) Cytotoxicity assay when reverse transfection of respective siRNAs was performed at 25 and 50 nmol/L for 48 h. (B) Representative Western blots and (C) densitometry analysis of H1299 cell lysate following reverse transfection of siRNA for 48 h and probed with antibodies for the respective host proteins. Replicate Western blots are in Fig. S7. (D) Supernatant of post-infection assay following siRNA knockdown of host targets of GW406108X. (E) Representative Western blots with densitometry analysis of (F) EV-D68 VP1 protein and (G, H) respective host proteins. Replicate Western blots are in Fig. S8. Triplicates were performed with error bars plotted to show the mean and standard deviation. Results were evaluated with one-way ANOVA and Dunnett’s test for statistical significance. ns signifies P > 0.05, ∗P ≤ 0.05, ∗∗P < 0.01, ∗∗∗P < 0.001.

• 

  Figure 7. GW406108X inhibits autophagy under both uninfected and EV-D68-infected conditions. Representative Western blots and densitometry analysis of H1299 cell lysate following GW406108X treatment for 12 h in (A, B) uninfected and (C, D) EV-D68-infected conditions. Triplicates were performed with error bars plotted to show the mean and standard deviation. Results were evaluated with one-way ANOVA and Dunnett’s test for statistical significance. ns signifies P > 0.05, ∗P ≤ 0.05, ∗∗P < 0.01, ∗∗∗P < 0.001.

• 

  Figure 8. Immunofluorescence staining of H1299 cells for LC3 puncta formation. (A) Representative images where blue stains for cell nucleus and green stains for LC3. (B) Quantification of the percentage of FITC-positive cells with LC3 puncta formation. HCQ: hydroxychloroquine. Cells were fixed with methanol 12 h.p.i., and cells treated with HCQ were fixed 24 h post-treatment. (C) Representative images where blue stains for cell nucleus, green stains for LC3, and red for dsRNA. (D) Quantification of the percentage of FITC-positive cells with LC3 puncta formation. H1299 cells were infected with EV-D68 at MOI 25, and then treated with either 5 μmol/L GW406108X or 0.1% DMSO. At several timepoints post-infection, cells were fixed in methanol and stained with LC3 antibody. Scale bars represent 40 μmol/L.

• 

  Figure 9. Electron micrographs showing GW406108X treatment reverses the effect of EV-D68 infection in H1299 cells. H1299 cells were infected with EV-D68 and treated with either 0.1% DMSO or 5 μmol/L GW406108X and fixed at 16 h.p.i. Representative images of cells are shown for each condition. The top row images were taken at 2000× magnification, and the bottom row images were taken at 5000× magnification. Red arrows indicate double-membrane vesicles (DMVs). N: nucleus; Cyt: cytoplasm; M: mitochondria. Scale bars are indicated in each image.

• 

  Figure 10. GW406108X toxicity and efficacy studies in P4 suckling C57BL/6 mice. For toxicity studies, P4 suckling C57BL/6 mice were treated with a dose of 10 mg/kg of GW406108X once a day for six days, and the mice monitored daily for up to 17 days post-treatment. The following results are shown (A) Kaplan–Meier survival curve, (B) body weight, and (C) physiological scoring based on five different criteria: activity, breathing, movement, body weight loss, and dehydration. For efficacy studies, P4 suckling C57BL/6 mice were pre-treated with a dose of GW406108X and intranasally inoculated with 1 × 107 PFU of EV-D68 2 h post-treatment. 5 or 10 mg/kg GW406108X or the vehicle control DMSO was administered via intraperitoneal injection for the following five days, and the mice monitored daily for up to 17 days post-treatment. The following results are shown (D) Kaplan–Meier survival curve, (E) body weight, and (F) physiological scoring based on the same five criteria mentioned in Table S1. (G) Normalised EV-D68 viral titres in the lung were evaluated with multiple unpaired t-test for statistical significance. ns signifies P > 0.05, ∗P ≤ 0.05, ∗∗P < 0.01, ∗∗∗P < 0.001. n = X which refers to the…

• 

  Figure 11. Diagram on the proposed mechanism through which GW406108X inhibits enterovirus replication. During enterovirus infection, the virus hijacks the autophagy pathway to promote its own replication via two ways 1) genome replication organelle formation and 2) non-lytic mature virion release. GW406108X, a ULK1/2 inhibitor, likely acts to reverse this effect by preventing inhibiting formation of the ULK complex, a key component required for the initiation of autophagosome formation. In this way, autophagy is inhibited, thereby preventing virus replication. Created with BioRender.com.