Glycyrrhizin Inhibits PEDV Infection and Proinflammatory Cytokine Secretion via the HMGB1/TLR4-MAPK p38 Pathway

Abstract

Our previous study showed that glycyrrhizin (GLY) inhibited porcine epidemic diarrhea virus (PEDV) infection, but the mechanisms of GLY anti-PEDV action remain unclear. In this study, we focused on the anti-PEDV and anti-proinflammatory cytokine secretion mechanisms of GLY. We found that PEDV infection had no effect on toll-like receptor 4 (TLR4) protein and mRNA levels, but that TLR4 regulated PEDV infection and the mRNA levels of proinflammatory cytokines. In addition, we demonstrated that TLR4 regulated p38 phosphorylation but not extracellular regulated protein kinases1/2 (Erk1/2) and c-Jun N-terminal kinases (JNK) phosphorylation, and that GLY inhibited p38 phosphorylation but not Erk1/2 and JNK phosphorylation. Therefore, we further explored the relationship between high mobility group box-1 (HMGB1) and p38. We demonstrated that inhibition of HMGB1 using an antibody, mutation, or knockdown decreased p38 phosphorylation. Thus, HMGB1 participated in activation of p38 through TLR4. Collectively, our data indicated that GLY inhibited PEDV infection and decreased proinflammatory cytokine secretion via the HMGB1/TLR4-mitogen-activated protein kinase (MAPK) p38 pathway.

Keywords: HMGB1; MAPK p38; PEDV infection; TLR4; glycyrrhizin; proinflammatory cytokine.

Figure 1

Porcine epidemic diarrhea virus (PEDV) had no effect on expression of toll-like receptor 4 (TLR4). Vero cells were infected with PEDV (0.1 multiplicity of infection (MOI)) for different lengths of time (4, 8, 12, 24, or 36 h). (A) Levels of TLR4 protein were analyzed by Western blotting. (B) Fold changes in the TLR4/actin ratio were plotted using ImageJ. (C) Levels of TLR4 mRNA were analyzed by qRT-PCR. Bars represent standard deviations.

Figure 2

TLR4 inhibitor TAK-242 (TAK) inhibited PEDV infection and increased levels of proinflammatory cytokines. Vero cells were treated with different concentrations of TAK for 2 h and then infected with PEDV (0.1 MOI) in the presence of different concentrations of TAK for 24 h. (A) porcine epidemic diarrhea virus nucleocapsid (PEDV-N) expression was analyzed by Western blotting. Actin was used as a loading control. (B) Immunofluorescence of PEDV-N (green) detected in infected Vero cells (blue is 4’,6-diamidino-2-phenylindole (DAPI)). (C) Levels of PEDV open reading frame (ORF3) RNA in infected cells were determined by qRT-PCR. (D) Viral titers in supernatants after TAK treatment were measured using a plaque formation assay. (E) Levels of mRNAs encoding proinflammatory cytokines were analyzed by qRT-PCR. p-values less than 0.05 were considered statistically significant (** p < 0.01). Bars represent standard deviations.

Figure 3

PEDV infection affected the activation of mitogen-activated protein kinase (MAPK) p38, extracellular regulated protein kinases1/2 (ERK1/2), and c-Jun N-terminal kinases (JNK). Vero cells were infected with PEDV (0.1 MOI) at 4, 8, 12, 24, and 36 h post-infection (h.p.i.). The cells were collected after different lengths of time for Western blotting. An equal amount of protein was subjected to Western blotting analysis. (A) Levels of phosphorylated and total MAPK p38, ERK1/2, or JNK were analyzed by Western blotting. Beta-actin was used as a loading control. (B) Levels of phospho-p38/total p38 were plotted using ImageJ. (C) Levels of phospho-JNK/total JNK were plotted using ImageJ. (D) Fold changes in the phospho-Erk/total Erk ratio were plotted using ImageJ. p-values less than 0.05 were considered statistically significant (** p < 0.01). Bars represent standard deviations.

Figure 4

MAPK p38 inhibitor SB202190 (SB) inhibited PEDV infection and increased levels of proinflammatory cytokine production. Vero cells were treated with different concentrations of SB for 2 h and then infected with PEDV (0.1 MOI) in the presence of different concentrations of TAK for 24 h. (A) PEDV-N levels were analyzed by Western blotting. Beta-actin was used as a loading control. (B) Immunofluorescence of PEDV-N (green) detected in infected Vero cells (blue is DAPI). (C) Levels of PEDV ORF3 RNA in infected cells were determined by qRT-PCR. (D) Viral titers in supernatants after SB treatment were measured using a plaque formation assay. (E) Levels of mRNAs encoding proinflammatory cytokines were analyzed by qRT-PCR. p-values less than 0.05 were considered statistically significant (** p < 0.01). Bars represent standard deviations.

Figure 5

TLR4 was an upstream modulator of MAPK p38 during PEDV infection. Vero cells were treated with different concentrations of TAK for 2 h and then infected with PEDV (0.1 MOI) in the presence of different concentrations of TAK for 24 h. (A) Levels of phosphorylated and total MAPK p38, ERK1/2, or JNK were analyzed by Western blotting. Βeta-actin was used as a loading control. (B) Levels of phospho-p38/total p38 were plotted using ImageJ. (C) Levels of phospho-JNK/total JNK were plotted using ImageJ. (D) Fold changes in phospho-Erk/total Erk ratio were plotted using ImageJ. p-values less than 0.05 were considered statistically significant (** p < 0.01). Bars represent standard deviations.

Figure 6

High mobility group box-1 (HMGB1) binding to TLR4 induced activation of p38 MAPK. (A–D) Vero cells were treated with different concentrations of glycyrrhizin (GLY) for 2 h and were then infected with PEDV (0.1 MOI) in the presence of different concentrations of GLY for 24 h. (A) Levels of phosphorylated and total MAPK p38, ERK1/2, or JNK were analyzed by Western blotting. Βeta-actin was used as a loading control. Levels of phospho-p38/total p38 (B), phospho-JNK/total JNK (C), and phospho-Erk/total Erk (D) were plotted using ImageJ. (E,F) Vero cells were treated with different doses of an anti-HMGB1 antibody for 2 h and then infected with PEDV (0.1 MOI) in the presence of anti-HMGB1 antibody for 24 h. (E) Levels of phosphorylated MAPK p38 and total MAPK p38 were analyzed by Western blotting. (F) Fold change in phospho-p38/total p38 ratio were plotted using ImageJ. (G,H) Vero cells were transfected with control pCAGGS (name of vector) plasmid (PCA) or plasmids encoding mutant HMGB1 (C45S, C106S, or C45S/C106S) for 24 h, and then infected with PEDV (0.1 MOI) for 24 h. (G) Levels of phosphorylated MAPK p38, total MAPK p38, and wild-type or mutant HMGB1 were analyzed by Western blotting. (H) Fold change in phospho-p38/total p38 ratio were plotted using ImageJ. (I,J) Vero cells were transfected with small interfering HMGB1 (siHMGB1) to silence HMGB1 expression for 24 h. SiNC (an irrelevant small interfering RNA (siRNA)) was used as a negative control. The cells were infected with PEDV for 24 h. (I) Levels of total and phosphorylated MAPK p38 and of HMGB1 were analyzed by Western blotting. (J) Levels of phospho-p38/total p38 were plotted using ImageJ. p-values less than 0.05 were considered statistically significant (* p < 0.05 and ** p < 0.01). Bars represent standard deviations.

Figure 7

GLY prevented binding of HMGB1 to TLR4 to inhibit PEDV infection, dependent on the HMGB1/TLR4-MAPK p38 pathway.