Antiviral activity of luteolin against porcine epidemic diarrhea virus in silico and in vitro

Abstract

Background: Porcine epidemic diarrhea virus (PEDV) mainly causes acute and severe porcine epidemic diarrhea (PED), and is highly fatal in neonatal piglets. No reliable therapeutics against the infection exist, which poses a major global health issue for piglets. Luteolin is a flavonoid with anti-viral activity toward several viruses.

Results: We evaluated anti-viral effects of luteolin in PEDV-infected Vero and IPEC-J2 cells, and identified IC₅₀ values of 23.87 µM and 68.5 µM, respectively. And found PEDV internalization, replication and release were significantly reduced upon luteolin treatment. As luteolin could bind to human ACE2 and SARS-CoV-2 main protease (Mpro) to contribute viral entry, we first identified that luteolin shares the same core binding site on pACE2 with PEDV-S by molecular docking and exhibited positive pACE2 binding with an affinity constant of 71.6 µM at dose-dependent increases by surface plasmon resonance (SPR) assay. However, pACE2 was incapable of binding to PEDV-S1. Therefore, luteolin inhibited PEDV internalization independent of PEDV-S binding to pACE2. Moreover, luteolin was firmly embedded in the groove of active pocket of Mpro in a three-dimensional docking model, and fluorescence resonance energy transfer (FRET) assays confirmed that luteolin inhibited PEDV Mpro activity. In addition, we also observed PEDV-induced pro-inflammatory cytokine inhibition and Nrf2-induced HO-1 expression. Finally, a drug resistant mutant was isolated after 10 cell culture passages concomitant with increasing luteolin concentrations, with reduced PEDV susceptibility to luteolin identified at passage 10.

Conclusions: Our results push forward that anti-PEDV mechanisms and resistant-PEDV properties for luteolin, which may be used to combat PED.

Keywords: Drug resistant mutant; Luteolin; Mpro; PEDV; Porcine ACE2; Pro-inflammatory cytokine; Spike.

Fig. 1

The evaluation of luteolin cytotoxicity on Vero and IPEC-J2 cell lines. A-B The luteolin cytotoxicity on Vero (A) and IPEC-J2 (B) cell lines by CCK-8 assays. C-D The CC50 values were calculated based on CCK-8 data of Vero (C) and IPEC-J2 (D) cell lines by nonlinear regression analysis with GraphPad Prism v.9.0.0

Fig. 2

Anti-viral activity of Luteolin on PEDV infected Vero, and IPEC-J2 cells. A The anti-PEDV effects of luteolin in Vero cells by using absolute RT-PCR assay. B IC₅₀ values in Vero cells were calculated based on RT-PCR data. C Anti-PEDV effects of luteolin by using RT-PCR assay in IPEC-J2 cells. D IC₅₀ values in IPEC-J2 cells were calculated based on RT-PCR data. E–G The anti-PEDV effects of luteolin in Vero cells by using immunofluorescence assay (E), western blot (F) and TCID₅₀ assays (G). Vero cells infected with PEDV and cultured with different concentrations of luteolin (0-50 μM). Densitometry quantification immunoblot analysis results of PEDV-N presented relative to those of GAPDH using Image J software. Values represent the mean ± standard deviation from three independent experiments. ns: non-significant; *P < 0.05; **P < 0.01; ***P < 0.001

Fig. 3

The effect of luteolin in different viral replication steps. The effects of luteolin on PEDV attachment (A), internalization (B), replication (C), release (D), and inactivation (E) in Vero cells (RT-PCR assay data). Values represent the mean ± standard deviation from three independent experiments. ns: non-significant; *P < 0.05; **P < 0.01; ***P < 0.001

Fig. 4

Luteolin inhibits PEDV replication independent of PEDV-S binding to pACE2. A Molecular docking data show interactions between PEDV-S and pACE2. B Molecular docking data show luteolin interactions with pACE2. C Luteolin binds to the same position on the PEDV-S surface as the PEDV-S/pACE2 binding interface. D Luteolin binding curve with pACE2; 5 µM–80 µM luteolin and 200 nM pACE2 were used. E Binding curve of pACE2 with PEDV-S1. pACE2, captured on a COOH chip, did not bind different PEDV-S1 concentrations (12.5–200 nM)

Fig. 5

Binding mode between Mpro and luteolin. A Binding patterns between Mpro (PDB: 7W6M) and luteolin. Red box: the candidate binding pocket in Mpro; B Hydrophilic-hydrophobic interactions between luteolin and Mpro in the binding pocket; C Key Mpro residues form binding interactions with the luteolin ligand. The binding free energy between Mpro and luteolin was − 7.36 kcal/mol. D Luteolin effects on Mpro activity. The IC₅₀ was determined in GraphPad. Data are shown as the mean ± standard deviation from three independent experiments

Fig. 6

Luteolin effects on pro-inflammatory cytokine levels. Overview of differentially expressed genes (DEGs) in PEDV and luteolin + PEDV groups. A The heat map shows DEGs convergence in different groups. B Volcano Plot showing DEPs. Cut off ratio of 1.5-fold and a P-value ≤ 0.05. C-D DEGs analysis between luteolin + PEDV and PEDV groups using GO (C) and KEGG (D) databases. E RT-PCR analysis of IL-6, IL-1β, TNF-a, MCP1, HO-1, and Nrf2 expression levels in different groups. Values represent the mean ± standard deviation from three independent experiments. ns: non-significant; *P < 0.05; **P < 0.01; ***P < 0.001

Fig. 7

Luteolin-resistant PEDV is generated after 10 passages. A Scheme showing the selection of a luteolin-resistant PEDV mutant virus. P1–P3 passages were subjected to 12.5 μM luteolin, P4–P6 to 25 μM luteolin, and P7–P10 to 50 μM luteolin. B Resistance analysis. Vero cells were infected with P10 mock-treated or luteolin treated virus at MOI = 0.05 for 1 h followed by 48 h incubation in media plus 50 μM luteolin or DMSO (negative control). Viral titers were quantified by TCID₅₀ assays and resistance quantified by comparing viral titers between luteolin-treated and luteolin untreated infections. Values represent the mean ± standard deviation from three independent experiments. ns: non-significant; *P < 0.05; **P < 0.01; ***P < 0.001