Differential remodeling of the electron transport chain is required to support TLR3 and TLR4 signaling and cytokine production in macrophages

Abstract

Increasing evidence suggests that mitochondria play a critical role in driving innate immune responses against bacteria and viruses. However, it is unclear if differential reprogramming of mitochondrial function contributes to the fine tuning of pathogen specific immune responses. Here, we found that TLR3 and TLR4 engagement on murine bone marrow derived macrophages was associated with differential remodeling of electron transport chain complex expression. This remodeling was associated with differential accumulation of mitochondrial and cytosolic ROS, which were required to support ligand specific inflammatory and antiviral cytokine production. We also found that the magnitude of TLR3, but not TLR4, responses were modulated by glucose availability. Under conditions of low glucose, TLR3 engagement was associated with increased ETC complex III expression, increased mitochondrial and cytosolic ROS and increased inflammatory and antiviral cytokine production. This amplification was selectively reversed by targeting superoxide production from the outer Q-binding site of the ETC complex III. These results suggest that ligand specific modulation of the ETC may act as a rheostat that fine tunes innate immune responses via mitochondrial ROS production. Modulation of these processes may represent a novel mechanism to modulate the nature as well as the magnitude of antiviral vs. inflammatory immune responses.

Figure 1

High, but not low, concentrations of Poly(I:C) are associated with pro-inflammatory cytokine production. BMMs were treated with either 100 ng/mL lipopolysaccharide (LPS), 10 ng/mL or 10 μg/mL Poly(I:C) (PIC) for 18 hours. Supernatant was collected and assessed for antiviral (IFN-α, IFN-β, CXCL10) (a–c) and pro-inflammatory (TNF-α, IL-1β, IL-6) (d–f) cytokine expression. Data represents mean ± SEM of four individual mice (*p < 0.05, **p < 0.01, and ***p < 0.001).

Figure 2

Macrophages activated using higher concentrations of Poly(I:C) are functioning near their maximum glycolytic capacity. BMMs were seeded onto Seahorse XFp miniplates and treated with 100 ng/mL LPS, 10 ng/mL or 10 μg/mL PIC for 18 hours. Glycolytic activity, indicated by the proton efflux rate (PER) was measured using sequential injections of rotenone plus antimycin A (Rot/AA) and 2-deoxyglucose (2-DG) (a), determining the %PER dependent on glycolysis (b) and the ratio of mitochondrial oxygen consumption rate (mitoOCR) to glycolytic PER (c). Data represents mean ± SEM of four individual mice. The levels of significance shown in (a) represent pairwise comparisons against LPS-treated macrophages (*p < 0.05, **p < 0.01, and ***p < 0.001).

Figure 3

Poly(I:C) stimulation is linked to low sustained levels of oxidative phosphorylation (OXPHOS). Macrophages were plated onto Seahorse XFp miniplates and subsequently stimulated with 100 ng/mL LPS, 10 μg/mL or 10 ng/mL PIC for 18 hours. OXPHOS function was assessed via successive Oligomycin (Oligo), Carbonyl cyanide-p-trifluoromethoxyphenylhydrazone (FCCP), and Rot/AA injections (a), quantifying the spare respiratory capacity percentage (SRC%) (b) and ATP production (c). Data represents mean ± SEM of four individual mice. The levels of significance shown in (a) is based on pairwise comparisons to LPS-treated macrophages (*p < 0.05, **p < 0.01, and ***p < 0.001).

Figure 4

Low glucose conditions are associated with increased IRF activation and increased type I IFN production. Macrophages were stimulated with either 100 ng/mL LPS or 10 ng/mL PIC for 18 hours under high glucose (25 mM) or low glucose (0.5 mM) media conditions. Supernatant was collected for assessing antiviral (IFN-α, IFN-β, CXCL10) (a–c) and pro-inflammatory (TNF-α, IL-6) cytokine (d,e) expression. Cell lysates were harvested to quantify p-IRF3 and total IRF3 (f), p-IRF7 and total IRF7 (g) and p-Iκbα and total Iκbα (h) expression via immunoblotting. Data represents mean ± SEM of four individual mice (*p < 0.05, **p < 0.01, and ***p < 0.001). For visualization purposes, the western blot images were cropped, but full-length blots and gel images can be found in Supplemental Fig. S5.

Figure 5

Poly(I:C) activation is linked to altered mitochondrial activity under low glucose conditions. BMMs treated with LPS or PIC for 18 hours under high glucose or low glucose media conditions were characterized for differences in mitochondrial function. Tetramethylrhodamine (TMRM) staining was used to measure, via flow cytometry, mitochondrial membrane potential. (a) Core protein levels of Complexes I-IV of the electron transport chain was quantified via immunoblotting. (b) Data represents mean ± SEM of four individual mice (*p < 0.05, **p < 0.01, and ***p < 0.001). For visualization purposes, the western blot images were cropped, but full-length blots and gel images can be found in Supplemental Fig. S6.

Figure 6

Targeting ETC activity reduces type I IFN-mediated responses during Poly(I:C) activation. LPS- (a) or PIC- (b) stimulated BMMs were co-treated with a panel of ETC inhibitors (Rotenone, Antimycin, Cyanide) to assess the importance of mitochondrial function for antiviral responses. CXCL10 and TNF-α cytokine secretion was measured after 18 hours in high glucose or low glucose media conditions. Data represents mean ± SEM of three individual mice (*p < 0.05, **p < 0.01, and ***p < 0.001).

Figure 7

Poly(I:C) activation promotes mitochondrial ROS production and accumulation. Macrophages treated either with LPS or PIC for 18 hours under high glucose or low glucose media conditions were examined for differences in redox metabolism. Mitochondrial superoxide production was measured using MitoSOX Red^TM (a). Protein levels of antioxidant proteins superoxide dismutase (SOD2) and mitochondrial glutathione peroxidase 4 (mtGPX4) were measured via immunoblotting (b). Cytosolic ROS production was measured using CellROX Orange (c). Hydrogen peroxide levels were quantified using the Cell-based Hydrogen Peroxide Assay kit (d). Data represents mean ± SEM of four individual mice (*p < 0.05, **p < 0.01, and ***p < 0.001). For visualization purposes, the western blot images were cropped, but full-length blots and gel images can be found in Supplemental Fig. S7.

Figure 8

Type I IFN production can be inhibited by altering mtROS generation during Poly(I:C) activation. LPS (a) or PIC (b) stimulated BMMs were co-treated with a panel of mtROS (MT, S3QEL, NAC) modulators to assess the importance of mitochondrial function for antiviral responses. CXCL10 and TNF-α cytokine secretion was measured after 18 hours in high glucose or low glucose media conditions. Data represents mean ± SEM of three individual mice (*p < 0.05, **p < 0.01, and ***p < 0.001).