Identification of niclosamide as a novel antiviral agent against porcine epidemic diarrhea virus infection by targeting viral internalization

Abstract

Porcine epidemic diarrhea virus (PEDV), an enteropathogenic coronavirus, has catastrophic impacts on the global pig industry. However, there remain no effective drugs against PEDV infection. In this study, we utilized a recombinant PEDV expressing renilla luciferase (PEDV-Rluc) to screen potential anti-PEDV agents from an FDA-approved drug library in Vero cells. Four compounds were identified that significantly decreased luciferase activity of PEDV-Rluc. Among them, niclosamide was further characterized because it exhibited the most potent antiviral activity with the highest selectivity index. It can efficiently inhibit viral RNA synthesis, protein expression and viral progeny production of classical and variant PEDV strains in a dose-dependent manner. Time of addition assay showed that niclosamide exhibited potent anti-PEDV activity when added simultaneously with or after virus infection. Furthermore, niclosamide significantly inhibited the entry stage of PEDV infection by affecting viral internalization rather than viral attachment to cells. In addition, a combination with other small molecule inhibitors of endosomal acidification enhanced the anti-PEDV effect of niclosamide in vitro. Taken together, these findings suggested that niclosamide is a novel antiviral agent that might provide a basis for the development of novel drug therapies against PEDV and other related pathogenic coronavirus infections.

Keywords: Antiviral; Coronavirus; Endocytosis; Host-targeted antivirals; Niclosamide (NIC); Porcine epidemic diarrhea virus (PEDV); Virus entry.

Fig. 1

Schematic of high-throughput screening of PEDV inhibitors from an FDA-approved drug library. Vero cells were pre-treated with drugs or DMSO as a control for 2 ​h and then infected with PEDV-DR13-Rluc virus in the presence of drugs. At 24 hpi, the luciferase activity of PEDV-DR13-Rluc in each group was determined by using the Luciferase Reporter Assay System. (A) ORF3 gene of recombinant PEDV DR13 strain is replaced by Renilla luciferase gene. (B) Time schedules of drug treatments and assay operations. (C) Inhibition rates of PEDV replication after different treatments indicated. It is the results of one representative experiment with eight replicates per drug. NIT: nitazoxanide; AZI: azithromycin; NIC: niclosamide; IVE: ivermectin.

Fig. 2

Evaluation of cytotoxicity and anti-PEDV efficacy of four identified drugs. Dose-dependent curves showed viability of Vero cells with serial dilution concentrations of Ivermectin (A), Nitazoxanide (B), Azithromycin (C), and Niclosamide (D). Dose-dependent curves showed the anti-PEDV efficacy of serially diluted Ivermectin (E), Nitazoxanide (F), Azithromycin (G), and Niclosamide (H). Selectivity index (SI) of identified anti-PEDV drugs (I). Error bars represent standard errors from three independent experiments. The IC₅₀ and CC₅₀ were determined by a best-fit Log(dose)-response curve-fitting in GraphPad Prism 8.

Fig. 3

Inhibition of VSV, PRV and PEDV replications by NIC. Microscopic images of Vero and LLC-PK1 cells infected with indicated viruses treated with different concentrations of NIC. At the same time, infected cells and the culture media were collected and progeny virus titer was determined by TCID₅₀ assay. (A) VSV infected cells were treated with different concentrations of NIC; (B) PRV infected cells were treated with different concentrations of NIC; (C) PEDV infected cells were treated with different concentrations of NIC. The experiment was performed three times independently. Differences were considered significant at (∗) 0.01 <P ​< ​0.05, (∗∗) 0.001<P ​< ​0.01, (∗∗∗) P ​< ​0.001.

Fig. 4

Effect of NIC on viral progeny production of various PEDV strains. (A–E) Vero cells were infected with DR13-GFP, CV777, HNXX, HW or HB strains with NIC treatment at different concentrations (0, 0.5, 1, 1.5 ​μmol/L). (F–J) LLC-PK1 cells were infected with DR13-GFP, CV777, HNXX, HW or HB strain with NIC treatment at different concentrations (0, 0.5, 0.8, 1 ​μmol/L). The PEDV titers in the culture supernatant were measured by TCID₅₀ assay. The experiment was performed three times independently. Differences were considered significant at (∗) 0.01 <P ​< ​0.05, (∗∗) 0.001<P ​< ​0.01, (∗∗∗) P ​< ​0.001.

Fig. 5

Effects of NIC on viral RNA synthesis of multiple PEDV strains. (A–E) Vero cells were infected with DR13-GFP, CV777, HNXX, HW or HB strains with NIC treatment at different concentrations (0, 0.5, 1, 1.5 ​μmol/L). (F–J) LLC-PK1 cells were infected with DR13-GFP, CV777, HNXX, HW or HB strain with NIC treatment at different concentrations (0, 0.5, 0.8, 1 ​μmol/L). The genomic RNA level of PEDV was then measured by RT-qPCR. The experiment was performed three times independently. Differences were considered significant at (∗) 0.01 <P ​< ​0.05, (∗∗) 0.001<P ​< ​0.01, (∗∗∗) P ​< ​0.001.

Fig. 6

Effects of NIC on S protein expression of multiple PEDV strains. Vero cells were infected with DR13-GFP, CV777, HNXX, HW or HB strains with NIC treatment. At 24 hpi, the N protein expression of DR13-GFP, CV777, HNXX, HW or HB (MOI ​= ​0.1) in Vero (A) or LLC- PK1 cells (B) was determined by IFA. Images were representative of results obtained from three independent experiments; scale bars ​= ​100 ​μm.

Fig. 7

Antiviral effects of NIC on PEDV infections under different kinds of drug-addition approach. (A) Schematics of NIC addition experiments. Vero cells were incubated with virus (MOI ​= ​0.1) for 1 ​h. Different concentrations of NIC were added prior to infection (−2 ​h) as well as at 0 ​h or 1 ​h post infection. (B) Antiviral effects of NIC on PEDV DR13-GFP proliferation under four kinds of treatments. (C) Antiviral effects of NIC on PEDV HW proliferation under four kinds of treatments. The viral titers in the culture supernatant were measured by TCID₅₀ assay. The experiment was performed three times independently. Differences were considered significant at (∗) 0.01 <P ​< ​0.05, (∗∗) 0.001<P ​< ​0.01, (∗∗∗) P ​< ​0.001.

Fig. 8

Time-of-drug addition study of NIC treatment on PEDV infections. Vero cells pretreated with DMSO or NIC (1 ​μmol/L) were infected with PEDV. At the indicated times post infection, NIC was added to a final concentration of 1 ​μmol/L (A). At 16 hpi, virus-infected cells were fixed, and virus infectivity was determined by quantifying the percentage of cells expressing GFP proteins by using FCM (B, C). The experiment was performed three times independently. Differences were considered significant at (∗) 0.01 <P ​< ​0.05, (∗∗) 0.001<P ​< ​0.01, (∗∗∗) P ​< ​0.001.

Fig. 9

The antiviral effect of NIC on different infection steps of PEDV. (A) NIC treatment schemes. Red bars represent PEDV infection, blue bars represent NIC treatment, and vertical bars represent the cell collection. (B–E) Vero cells were infected with PEDV and treated with NIC at indicated time points (A), which represented the stage of viral attachment (B), internalization (C), replication (D) or release (E), respectively. At 16 hpi, the samples were prepared for determining viral RNA copies by RT-qPCR, and viral titers by TCID₅₀ or plaque assay. (F) Inactivation assay. To determine whether NIC directly inactivates PEDV viral particles, NIC at the concentration of 1 ​μmol/L was incubated with PEDV for 1 ​h at 37 ​°C, then the mixtures were added to Vero cells. The infectious titers were determined by TCID₅₀ and PFU at 24 hpi.

Fig. 10

Synergistic anti-PEDV efficacy of NIC with BafA1 or CQ. (A) The chemical structures of NIC, BafA1 and CQ. (B) Vero cells incubated with or without of BafA1 (0.5 ​μmol/L, 1 ​μmol/L) or CQ (15 ​μmol/L, 25 ​μmol/L) were treated with 0.5 ​μmol/L NIC, and then infected with PEDV. At 16 hpi, the samples were prepared for TCID₅₀ assay. The experiment was performed three times independently. Differences were considered significant at (∗) 0.01 <P ​< ​0.05, (∗∗) 0.001<P ​< ​0.01, (∗∗∗) P ​< ​0.001.