Identification of cepharanthine as an effective inhibitor of African swine fever virus replication

Abstract

African swine fever virus (ASFV) causes highly contagious swine disease, African swine fever (ASF), thereby posing a severe socioeconomic threat to the global pig industry and underscoring that effective antiviral therapies are urgently required. To identify safe and efficient anti-ASFV compounds, a natural compound library was screened by performing an established cell-based ELISA in an ASFV-infected porcine alveolar macrophage (PAM) model. In total, 6 effective anti-ASFV compounds with low cytotoxicity were identified. Cepharanthine (CEP), a bisbenzylisoquinoline alkaloid, was the most potent inhibitor effect with an IC₅₀ of 0.3223 μM. To further investigate the mechanism through which CEP inhibits ASFV replication, transcriptome profiles were generated in PAMs treated with CEP and/or infected with ASFV. ASFV infection dramatically altered immune response-associated gene expression. CEP treatment upregulated the expression of cholesterol biosynthesis-related genes, regardless of infection status. According to time-of-addition experiments, CEP primarily exerts its antiviral effect during the early stages of ASFV infection, specifically by inhibiting viral entry. Transcriptomic analysis suggested that CEP blocks ASFV entry through the clathrin-mediated endocytosis pathway by increasing EHD2 gene expression in macrophages. Disrupting EHD2 with small interfering RNA promoted ASFV entry into clathrin-positive vesicles. Finally, the protective effect of CEP in vivo was evaluated using ASFV-infected pigs. CEP could provide partial protection against ASFV infection, as indicated by an increase in survival time from 9.67 days to 16.67 days. Our findings imply that CEP exhibits potential antiviral activity against ASFV infection in PAMs, positioning it as a promising therapeutic strategy for ASF.

Keywords: African swine fever virus; EHD2; cepharanthine; cholesterol biosynthesis; transcriptome analysis.

Figure 1.

Screening of antiviral compounds against ASFV in the Selleck natural compounds library. (A) Depiction of the screening process of anti-ASFV compounds. A total of 803 compounds were screened in ASFV infected PAMs (1 MOI) using a cell-based Elisa assay. Two replicates were set for each compound at a concentration of 10 μM, and those with high cytotoxicity were excluded in the second round of screening. (B) The screening results of anti-ASFV compounds. Each dot represents a compound, ranging by infection ratio. (C) The 50% cytotoxicity concentrations (CC₅₀) of candidate anti-ASFV compounds. PAMs (2 × 10⁵ cells/well) were seeded in a 96 well plate overnight, treated with the indicated concentration of DMSO or compounds for 24 h, and then treated with CCK-8 agent for an additional 2 h. The OD₄₅₀ value was measured for cell viability detection. (D) The Half Maximal Inhibitory Concentration (IC₅₀) of candidate anti-ASFV compounds. PAMs were treated with the indicated concentration of DMSO or compounds for 2 h, followed by a cell-based ELISA assay. Data were calculated through nonlinear regression analysis.

Figure 2.

CEP inhibits viral replication during the ASFV entry stage. (A) The chemical structures of CEP. PAMs were pretreated with CEP at the indicated concentration and then infected with or without ASFV (1 MOI). At 24 hpi, the cells were collected for analysis. (B-C) Viral p72 mRNA expression level (B) and intracellular virus titer (C) were measured at 24 hpi in ASFV-infected cells incubated with DMSO or CEP (4 μM). (D) Samples were analyzed by Western blotting assay; ASFV-p72 protein levels were detected with a specific ASFV-p72 monoclonal antibody. (E) PAMs were infected with ASFV (1, 2, or 5 MOI) and then applied using the same method as given in (B). (F) PAMs were treated with DMSO (top panel), 2 μM CEP (middle panel), or 4 μM (bottom panel) CEP for 2 h and then infected with ASFV (1 MOI). After 24 h, the cells were fixed and subjected to an indirect immunofluorescence assay. ASFV-p72 (Green) was detected in each group, and representative images are shown (Scale bar = 50 μm). (G-H) Time-of-addition analysis of CEP against ASFV. (G) Overall scheme for PAMs infection and treatment with CEP (4 μM); the black line refers to the ASFV infection period and the gray line refers to the CEP treatment period. (H) Protein samples from each indicated group were analyzed for Western blotting analysis. (I) PAMs were treated as described in the ASFV entry assay. After treatment, the cells were fixed for IFA experiments. Viral proteins were localized using monoclonal antibodies against ASFV-p30 and ASFV-I73R, and the nuclei of the cells were stained with DAPI (Scale bar = 25 μm), the images were acquired by the Leica STELLARIS confocal microscopy. Representative images are shown.

Figure 3.

Summary of transcript abundance. (A) Sample treatment process for each group. (B) Principal component analysis (PCA) results comparing 12 transcriptomes from PAMs infected with or without ASFV and treated with or without CEP. (C) Graph showing the number of DEGs between different groups. Up-regulated DEGs are denoted in black and down-regulated DEGs are indicated in gray. (D-E) Transcriptome analysis of ASFV-infected PAMs. (D) Volcano plot of 214 DEGs with q value < 0.05 in the Virus-Mock group comparison. The red color represents upregulated genes in the virus-infected cell with Log₂FC > 1 and blue represents downregulated genes with Log₂FC < −1. (E) Top upregulated pathways enriched from 182 upregulated genes in the virus-infected cells shown in the bubble plot. The bubble color scales the enrichment score. The size of the bubble scales the count of the enriched genes. The X-axis depicts the enrichment score, while the Y-axis shows the name of the KEGG-enriched pathway.

Figure 4.

Transcriptional response of PAMs after CEP treatment. (A-B) Volcano plot (A) and heatmap (B) of a total of 329 DEGs. The red color represents upregulated genes in CEP-treated cells with Log₂FC > 1, and the blue color represents downregulated genes with Log₂FC < −1. The relative expression in the Z-score mode is shown in the heatmap. Red cells represent upregulated genes, and blue cells represent downregulated genes. (C-D) GO enrichment analysis of CEP-Mock DEGs. (C) Top upregulated pathways enriched from 110 upregulated genes. (D) Top downregulated pathways enriched from 219 downregulated genes in CEP-treated cells are shown in the bubble plot. The bubble color scales the P-value. The size of the bubble scales the count of the enriched genes. The X-axis is equal to the enrichment score, and Y-axis is the name of the GO enriched pathway. (E) Heatmap of pathways of cholesterol biosynthetic process, ATP binding, and innate immune response. Red cells represent upregulated genes; blue cells represent downregulated genes. (F) Network visualization of pathways enriched by 110 upregulated genes in CEP treated cells analysis. (G) Protein-protein network visualization of the gene component.

Figure 5.

ASFV infection poorly alters the gene transcription status of CEP-treated PAMs. (A) Venn diagrams show an overlap of DEGs in CEP-Mock and (CEP + Virus)-Mock comparison. (B) Graph depicting the number of DEGs in (CEP + Virus)-CEP. (C-D) Bar diagram displaying the top enrichment results of the same downregulated DEGs (C) and the same upregulated DEGs (D) in CEP-Mock and (CEP + Virus)-Mock comparison. (E) Heatmap showing genes of metabolism of the cholesterol pathway in the Mock, Virus, CEP, and (CEP + Virus) group. Yellow cells represent upregulated genes, while blue cells represent downregulated genes. (F) RT-qPCR showed that the expression of SREBF2, HMGCR, HMGCS1, FASN, and SQLE genes was significantly higher in the CEP-treated group with or without ASFV infection (*P < 0.05, **P < 0.01, ***P < 0.001).

Figure 6.

EHD2 is a target for CEP to inhibit ASFV entry. (A-C) CEP treatment partially reverses virus infection genes transcription. (A) GSEA of expression data on ASFV up/downregulated genes in (CEP + Virus)-Virus. (B) Diagrams showing percentages of ASFV up/downregulated genes reversed by CEP treatment. (C) Heatmap showing genes of clathrin mediated endocytosis pathway in the Mock, Virus, CEP, and CEP + Virus group. Yellow cells represent upregulated genes, while blue cells represent downregulated genes. (D-F) CEP can upregulate intracellular EHD2 expression. PAMs were treated with DMSO or CEP (4 μM) and infected with or without ASFV (1 MOI). Samples were harvested at 24 hpi and subjected to (D) Western blotting analysis with an EHD2 antibody, and RT-qPCR analysis with specific ASFV-p72 primers (E) or EHD2 primers (F). (G) Brief diagram of EHD2 structure. (H-I) Full-length or truncated EHD2 were cloned into an EGFP vector and transfected into WSL cells for 36 h and then infected with ASFV (5 MOI) for an additional 2 h. The samples were collected after washing off the virions on the cell surface. (H) Protein samples were collected for Western blotting analysis with EHD2 antibody. The arrow indicates the protein size of the EHD2 in each group. (I) Intracellular virus titer of each samples with EHD2 overexpression was detected. (J-K) PAMs were seeded on 12-well plates with 10⁶ cells/well, cultured overnight, and then transfected with 50 nM of targeting EHD2 siRNA or nontargeting scrambled siRNA by RNAi transfection reagent for 72 h, followed by treatment with 4 μM CEP or DMSO for 2 h, infection with ASFV (1 MOI) for 24 h, washing off the unincorporated virions, and collection of the protein and DNA samples for further analysis. (J) Intracellular EHD2 and GAPDH protein level were measured by Western-blotting. (K) Intracellular ASFV copies were measured. (L-M) 293 T cells (L) or WSL cells (M) were transfected with HA-Cav1 and empty vector or Flag-EHD2, and whole cell lysates were immunoprecipitated (IP) with HA antibody or control IgG. The immunocomplexes were then used to detect Flag-EHD2, HA-Cav1, and LC3 by immunoblotting with the indicated antibodies. (N-S) WSL cells were transfected with empty vector or Flag-EHD2 (1 μg) (N) and nontargeting scrambled siRNA or targeting EHD2 siRNA (50 nM) (O) for 72 h, followed by infection with ASFV (5 MOI) for 6 h. Next, the cells were fixed and stained with Caveolin-1 antibody (ab2910) and Alexa 555-conjugated goat anti-rabbit IgG antibodies (red), ASFV-p30 antibody and Alexa 647-conjugated donkey Anti-Mouse IgG antibody (green), EHD2 antibody (11440-1-AP) and Alexa 488-conjugated goat anti-rabbit IgG antibody (cyan), The nuclei were stained with DAPI (blue). The images were acquired with a Leica STELLARIS confocal microscopy (Scale bar = 25 μm). (P) and (R) Graph shows the percentage of Cav1-ASFV-p30 colocalized cells within EHD2-postived cells (P) or in EHD2-silenced cells (R). Approximately 200 cells were counted in each group. Transfection efficiency was detected by Western-blotting of intracellular Flag (Q) or EHD2 (S) protein level. (T) Schematic diagram showing that CEP can upregulate EHD2 to inhibit ASFV replication.

Figure 6.

EHD2 is a target for CEP to inhibit ASFV entry. (A-C) CEP treatment partially reverses virus infection genes transcription. (A) GSEA of expression data on ASFV up/downregulated genes in (CEP + Virus)-Virus. (B) Diagrams showing percentages of ASFV up/downregulated genes reversed by CEP treatment. (C) Heatmap showing genes of clathrin mediated endocytosis pathway in the Mock, Virus, CEP, and CEP + Virus group. Yellow cells represent upregulated genes, while blue cells represent downregulated genes. (D-F) CEP can upregulate intracellular EHD2 expression. PAMs were treated with DMSO or CEP (4 μM) and infected with or without ASFV (1 MOI). Samples were harvested at 24 hpi and subjected to (D) Western blotting analysis with an EHD2 antibody, and RT-qPCR analysis with specific ASFV-p72 primers (E) or EHD2 primers (F). (G) Brief diagram of EHD2 structure. (H-I) Full-length or truncated EHD2 were cloned into an EGFP vector and transfected into WSL cells for 36 h and then infected with ASFV (5 MOI) for an additional 2 h. The samples were collected after washing off the virions on the cell surface. (H) Protein samples were collected for Western blotting analysis with EHD2 antibody. The arrow indicates the protein size of the EHD2 in each group. (I) Intracellular virus titer of each samples with EHD2 overexpression was detected. (J-K) PAMs were seeded on 12-well plates with 10⁶ cells/well, cultured overnight, and then transfected with 50 nM of targeting EHD2 siRNA or nontargeting scrambled siRNA by RNAi transfection reagent for 72 h, followed by treatment with 4 μM CEP or DMSO for 2 h, infection with ASFV (1 MOI) for 24 h, washing off the unincorporated virions, and collection of the protein and DNA samples for further analysis. (J) Intracellular EHD2 and GAPDH protein level were measured by Western-blotting. (K) Intracellular ASFV copies were measured. (L-M) 293 T cells (L) or WSL cells (M) were transfected with HA-Cav1 and empty vector or Flag-EHD2, and whole cell lysates were immunoprecipitated (IP) with HA antibody or control IgG. The immunocomplexes were then used to detect Flag-EHD2, HA-Cav1, and LC3 by immunoblotting with the indicated antibodies. (N-S) WSL cells were transfected with empty vector or Flag-EHD2 (1 μg) (N) and nontargeting scrambled siRNA or targeting EHD2 siRNA (50 nM) (O) for 72 h, followed by infection with ASFV (5 MOI) for 6 h. Next, the cells were fixed and stained with Caveolin-1 antibody (ab2910) and Alexa 555-conjugated goat anti-rabbit IgG antibodies (red), ASFV-p30 antibody and Alexa 647-conjugated donkey Anti-Mouse IgG antibody (green), EHD2 antibody (11440-1-AP) and Alexa 488-conjugated goat anti-rabbit IgG antibody (cyan), The nuclei were stained with DAPI (blue). The images were acquired with a Leica STELLARIS confocal microscopy (Scale bar = 25 μm). (P) and (R) Graph shows the percentage of Cav1-ASFV-p30 colocalized cells within EHD2-postived cells (P) or in EHD2-silenced cells (R). Approximately 200 cells were counted in each group. Transfection efficiency was detected by Western-blotting of intracellular Flag (Q) or EHD2 (S) protein level. (T) Schematic diagram showing that CEP can upregulate EHD2 to inhibit ASFV replication.

Figure 7.

CEP extended the survival time of ASFV-infected pigs in animal challenge experiments. (A) Schematic illustration of the animal experiment. Landrace pigs (42 days old) were randomly assigned into the ASFV-/CEP- group (n = 2), ASFV group (n = 3), and CEP + ASFV group (n = 3). CEP (4.44 mg/kg) or DMSO was injected into the neck muscle at −2, −1, 0, and 1 d. Next, the drug was injected every 48 h before the virus challenge. At 0 d, 50 TCID₅₀ ASFV was inoculated intramuscularly (i.m.) into the ASFV and CEP + ASFV group pigs. The CEP + ASFV group pigs were co-housed with those from the ASFV group from day 0 until the termination of the experiment. (B) Rectal temperatures of animals after inoculation with ASFV. (C) The survival rates of animals after inoculation with ASFV. Statics were calculated using GraphPad Prism 9.0 software. (D) Tissue lesions of pigs in each group. (E) Major organ tissues samples including the lung, heart, and spleen were sectioned (H&E) and histopathologically analyzed (20 μm scale). (F-G) Total nucleic acids were extracted from the tissues or swabs after animal death, and viral loading in the nucleic acids of each sample was detected. (F) Viral loading in the heart, liver, spleen, lungs, kidneys, and mesenteric lymph nodes. (G) Viral loading in the oral, nasal, and anal swabs.