Differential TLR-ERK1/2 Activity Promotes Viral ssRNA and dsRNA Mimic-Induced Dysregulated Immunity in Macrophages

Abstract

RNA virus-induced excessive inflammation and impaired antiviral interferon (IFN-I) responses are associated with severe disease. This innate immune response, also referred to as "dysregulated immunity" is caused by viral single-stranded RNA (ssRNA)- and double-stranded-RNA (dsRNA)-mediated exuberant inflammation and viral protein-induced IFN antagonism. However, key host factors and the underlying mechanism driving viral RNA-mediated dysregulated immunity are poorly defined. Here, using viral ssRNA and dsRNA mimics, which activate toll-like receptor 7 (TLR7) and TLR3, respectively, we evaluated the role of viral RNAs in causing dysregulated immunity. We observed that murine bone marrow-derived macrophages (BMDMs), when stimulated with TLR3 and TLR7 agonists, induced differential inflammatory and antiviral cytokine response. TLR7 activation triggered a robust inflammatory cytokine/chemokine induction compared to TLR3 activation, whereas TLR3 stimulation induced significantly increased IFN/IFN stimulated gene (ISG) response relative to TLR7 activation. To define the mechanistic basis for dysregulated immunity, we examined cell-surface and endosomal TLR levels and downstream mitogen-activated protein kinase (MAPK) and nuclear factor kappa B (NF-kB) activation. We identified significantly higher cell-surface and endosomal TLR7 levels compared to TLR3, which were associated with early and robust MAPK (p-ERK1/2, p-P38, and p-JNK) and NF-kB activation in TLR7-stimulated macrophages. Furthermore, blocking EKR1/2 and NF-kB activity reduced TLR3/7-induced inflammatory cytokine/chemokine levels, whereas only ERK1/2 inhibition enhanced viral RNA mimic-induced IFN/ISG responses. Collectively, our results illustrate that high cell-surface and endosomal TLR7 expression and robust ERK1/2 activation drive viral ssRNA mimic-induced excessive inflammatory and reduced IFN/ISG response and blocking ERK1/2 activity would likely mitigate viral-RNA/TLR-induced dysregulated immunity.

Keywords: ERK1/2; SARS-CoV-2; TLRs; inflammation; interferon; macrophages.

Figure 1

ssRNA/TLR7 stimulation induces robust inflammatory cytokine/chemokine responses, whereas dsRNA mimics are potent inducers of antiviral IFN and ISG responses. Murine BMDMs were stimulated with viral RNA mimics, namely R848 (ssRNA mimic) and poly I:C (dsRNA mimic), each with 10 ug/mL. mRNA (A) and protein levels (B) of inflammatory cytokines and chemokines and mRNA (C) and protein levels (D) of interferons and ISGs were assessed 8 and 24 h after stimulation, respectively. Data are representative of three independent experiments. Statistical significance was determined using one-way ANOVA with * p < 0.05, ** p < 0.01, *** p < 0.001, and **** p < 0.0001.

Figure 2

ssRNA/TLR7 elicits inflammatory cytokine/chemokine responses via MyD88, whereas dsRNA mimics elicit IFN and ISG response via TRIF signaling. WT and TRIF-/- BMDMs (A) and WT and MyD88-/- BMDMs (B) were stimulated with viral RNA mimics, namely poly I:C (dsRNA mimic) and R848 (ssRNA mimic), each with 10 ug/mL, respectively. mRNA levels of interferons and inflammatory cytokines were assessed 8 h post stimulation. Data are representative of three independent experiments. Statistical significance was determined using one-way ANOVA with * p < 0.05, ** p < 0.01, and **** p < 0.0001.

Figure 3

TLR3 and TLR7 expression in mouse bone marrow macrophages. Expression of cell-surface, intracellular, and mRNA levels of TLR3 and TLR7 in naïve mouse bone marrow macrophages was analyzed by flow cytometry (FACS) (A–C) and qRT-qPCR (D) using standard protocols. Scatter plot graphs show the percentage and mean fluorescent intensity of TLR3- and TLR7-positive cells (B,C) and TLR3 and TL7 mRNA expression in naïve BMDMs (D). Data represent 3–4 technical replicates each from 3–4 biological replicates. Statistical significance was determined using one-way ANOVA with * p < 0.05, *** p < 0.001 and **** p < 0.0001.

Figure 4

RNA mimic stimulation induced differential phosphorylation of MAPK and NF-kB signaling. Murine bone marrow macrophages were treated with ssRNA mimic (R848) and dsRNA mimic (PIC), each with 10 ug/mL, for 10, 30, and 60 min. Cell lysates were collected, and Western blot was performed using mAbs against phosphorylated MAPK (p-ERK1/2 (phospho-ERK1/2), p-P38 (phospho-P38), and p-JNK (phospho-JNK) and NF-kB (p-P65 (phospho-P65)). and their total proteins were denoted as T-ERK, T-P38, T-JNK, and T-P65, respectively (A), and relative quantification of protein levels were measured by densitometry analysis (B). Data are representative of three independent experiments (A) or pooled from three independent experiments (B). Statistical significance was determined using one-way ANOVA with * p < 0.05, ** p < 0.01, *** p < 0.001 and **** p < 0.0001.

Figure 5

Effect of blocking MAPK and NF-kB signaling in viral RNA mimic-induced inflammatory and antiviral cytokine response. Murine BMDMs were treated with specific inhibitors (5 µM) of ERK1/2 (i), p38 (ii), NF-kB (iii), and JNK (iv) pathways for 2 h followed by stimulation with R848 and or Poly I:C, each with 5–10 ug/mL. Western blot showing the inhibitory effect of Trametinib, Losmapimod, BAY, and JNKi on phospho-ERK (p-ERK), phospho-P38 (p-P38), phospho-P65(p-P65), and phospho-JNK(p-JNK), respectively, and their total proteins, denoted as T-ERK, T-P38, T-P65, and T-JNK, respectively. (A) Protein levels of cytokines and interferons (B,C) were assessed at 8 and 24 h post stimulation. Data are representative of 2–3 independent experiments (B,C). Statistical significance was determined using one-way ANOVA with ** p < 0.01, *** p < 0.001 and **** p < 0.0001.

Figure 6

Inhibiting ERK1/2 signaling enhances antiviral interferon response in viral RNA mimic-treated cells. Murine BMDMs were treated with Trametinib (5 µM) inhibitor of ERK1/2 pathway for 2 h, followed by stimulation with R848 and or Poly I:C, each with 5–10 ug/mL. mRNA levels of ISGs with R848 and PIC stimulations, respectively, with and without Trametinib (A,B) were assessed at 8 and 24 h post stimulation. Data are representative of 2–3 independent experiments. Statistical significance was determined using one-way ANOVA with * p < 0.05, ** p < 0.01, *** p < 0.001 and **** p < 0.0001.