CD11b/CD18 (Mac-1) is a novel surface receptor for extracellular double-stranded RNA to mediate cellular inflammatory responses

Abstract

During viral infection, extracellular dsRNA is a potent signaling molecule that activates many innate immune cells, including macrophages. TLR3 is a well-known receptor for extracellular dsRNA, and internalization of extracellular dsRNA is required for endosomal TLR3 activation. Preserved inflammatory responses of TLR3-deficient macrophages to extracellular dsRNA strongly support a TLR3-independent mechanism in dsRNA-mediated immune responses. The present study demonstrated that CD11b/CD18 (Mac-1 [macrophage-1 Ag]), a surface integrin receptor, recognized extracellular dsRNA and induced macrophage immune responses. CD11b deficiency reduced inflammatory cytokine induction elicited by polyinosinic:polycytidylic acid (poly I:C; a synthetic dsRNA) in mouse sera and livers, as well as in cultured peritoneal macrophages. dsRNA-binding assay and confocal immunofluorescence showed that Mac-1, especially the CD11b subunit, interacted and colocalized with poly I:C on the surface of macrophages. Further mechanistic studies revealed two distinct signaling events following dsRNA recognition by Mac-1. First, Mac-1 facilitated poly I:C internalization through the activation of PI3K signaling and enhanced TLR3-dependent activation of IRF3 in macrophages. Second, poly I:C induced activation of phagocyte NADPH oxidase in a TLR3-independent, but Mac-1-dependent, manner. Subsequently, phagocyte NADPH oxidase-derived intracellular reactive oxygen species activated MAPK and NF-κB pathways. Our results indicate that extracellular dsRNA activates Mac-1 to enhance TLR3-dependent signaling and to trigger TLR3-independent, but Mac-1-dependent, inflammatory oxidative signaling, identifying a novel mechanistic basis for macrophages to recognize extracellular dsRNA to regulate innate immune responses. This study identifies Mac-1 as a novel surface receptor for extracellular dsRNA and implicates it as a potential therapeutic target for virus-related inflammatory diseases.

Figure 1. Mac-1-deficient mice revealed impaired immune response after poly I:C-injection

Wild-type (WT) or CD11b^−/− mice were intraperitoneally injected with 5mg/kg poly I:C. After 2h, blood and livers were collected for analysis of inflammatory cytokine induction. The amount of TNF-α, IL-12p40, and IFN-β in serum was detected by ELISA (A), and the mRNA level of TNF-α, IFN-β, IL-6, and IL-12p40 in livers was measured by RT-PCR (B). Data are shown as means ± SEM from 5 to 7 pairs of WT and CD11b^−/− mice. *p<0.01, compared with corresponding poly I:C-injected WT mice.

Figure 2. Mac-1 deficiency impaired immune responses to poly I:C in macrophages

Poly I:C was added to macrophages prepared from wild-type (WT) or CD11b^−/− mice with the indicated dose range. After 2 h, total RNA was extracted and used in real-time PCR (A). The release of TNF-α and IL-12p40 from cultured macrophages was determined by ELISA at 4h or 24h after the treatment (B). Nitric oxide production was measured at 24 h after the treatment (C). Data are shown as means ± SEM from three independent experiments in triplicate. *p<0.01, compared with corresponding vehicle-treated controls. #p<0.01, compared with corresponding poly I:C-treated WT groups.

Figure 3. Poly I:C bound to Mac-1 receptor on the cell surface

DsRNA-binding assay was performed by using poly I:C-coated agarose beads, unconjugated beads, or polyC-coated beads. Only poly I:C-coated beads can pull down the CD11b subunit of Mac-1 in either primary macrophages or Raw 264.7 macrophage cells (A). Surface binding assay further delineated the interaction between poly I:C and Mac-1 on cell surface (B–E). Flow cytometry showed that wild-type (WT) macrophages had higher surface binding of FITC-labeled poly I:C than CD11b^−/− macrophages, and the poly I:C binding was significantly inhibited by fibrinogen (1 μM) in WT macrophages and RAW264.7 cells but not CD11b^−/− macrophages (B, C). The confocal experiments revealed that more Cy3-labeled poly I:C bound to the surface of WT macrophages than bound to the surface of CD11b^−/− macrophages, and the binding of Cy3-labeled poly I:C was inhibited by fibrinogen only in WT macrophages (D). CD11b immunostaining showed that Mac-1 co-localized with surface-bound Cy3-poly I:C (E). Data are shown as means ± SEM from three independent experiments in triplicate. *p<0.01, compared with corresponding poly I:C-treated groups; #p<0.01, compared with corresponding poly I:C-treated WT groups.

Figure 4. Mac-1 facilitates the internalization of poly I:C

Cultured peritoneal macrophages from wild-type (WT) and CD11b^−/− mice were incubated with Cy3-labeled poly I:C (10 μg/ml) at 37°C for 15 or 30min or Cy3-labeled poly C (10 μg/ml) for 30min. Confocal imaging shows decreased intracellular Cy3 fluorescence in CD11b^−/− macrophages compared with WT macrophages after these cells were treated with Cy3-labeled poly I:C but not with Cy3-labeled poly C (A). Macrophages from WT and CD11b^−/− mice were incubated with FITC-labeled poly I:C (10 μg/ml) at 37°C for 30min. After the fluorescence of extracellular surface-bound FITC-labeled poly I:C was quenched by Trypan blue (1 mg/ml), the fluorescence density of the internalized FITC-labeled poly I:C was measured by flow cytometry and was present after the subtraction of nonspecific autofluorescence in untreated control cells (B). After pretreatment with normal IgG (2.5 μg/ml) or anti-CD11b antibody (2.5 μg/ml) for 15 min, macrophages from WT and CD11b^−/− mice were incubated with Cy3-labeled poly I:C for 30min. Confocal imaging showed reduced fluorescence density in antibody treatment group (C). The pretreatment of WT macrophages for 15 min with PI3K inhibitor wortmannin attenuated the uptake of Cy3-labeled poly I:C (D). Western blot analysis on poly I:C-challenged WT and CD11b^−/− macrophages at different time points. Impaired AKT phosphorylation was observed in CD11b^−/− macrophages (E). Data are shown as means ± SEM from three independent experiments in triplicate. *p<0.01, compared with corresponding vehicle-treated controls; #p<0.01, compared with corresponding poly I:C-treated WT groups.

Figure 5. Mac-1 altered poly I:C-induced downstream signaling pathway

Peritoneal macrophages from wild-type (WT) and CD11b^−/− mice were treated with 50 μg/ml poly I:C. Nuclear fraction was extracted from these cells 60 min or 120 min after the treatment. The phosphorylation and the nuclear translocation of IRF3 protein were examined by using phospho-IRF-3 (Ser396) antibody. Antibody specific for nuclear marker HDAC2 was included to monitor loading errors (A). MAPK pathway activation was determined by the phosphorylation of its major kinases p38 and JNK at indicated time points (0, 15, 30, or 60 min) in the whole cell lysates (B). The activation of NFκB was indicated by the phosphorylation of p65 and Iκbα (an inhibitory cofactor of NFκB) and the degradation of Iκbα in the whole cell lysis (C). Peritoneal macrophages from WT mice were pretreated with the JNK inhibitor SP600125 and the NFκB inhibitor Compound A for 30 min, followed by vehicle or poly I:C challenge. Secreted TNF-α and IL-12p40 were determined by ELISA after 24h. Both inhibitors significantly reduced poly I:C-stimulated cytokine production (D). All immunoblot lanes are representative results from three independent experiments. Data are shown as means ± SEM from three independent experiments in triplicate. *p<0.05, compared with corresponding WT controls. #p<0.05, compared with corresponding poly I:C-treated WT macrophages.

Figure 6. Mac-1-dependent activation of phagocyte NADPH oxidase (NOX2) enhanced poly I:C-elicited immune response

Macrophages from wild-type (WT), CD11b^−/−, and gp91^−/− mice were treated with vehicle, poly I:C (50 μg/ml), or PMA (50 nM). Production of extracellular superoxide was measured by SOD-inhabitable reduction of WST-1 (A). Intracellular ROS (iROS) was determined by using the fluorescent DCFH-DA probe (B). Secreted TNF-α and IL-12p40 were detected in WT and gp91^−/− macrophages treated with 50 μg/ml poly I:C for 24 h (C). WT macrophages were pretreated with vehicle or apocynin (0.1 or 0.25 mM) for 15 min, and secreted TNF-α and IL-12p40 were measured after 24h treatment with 50 μg/ml poly I:C (D). Western blot analysis showed impaired phosphorylation of p38, JNK, and p65 in gp91^−/− macrophages challenged with 50 μg/ml poly I:C (E). After stimulation with poly I:C (50 μg/ml) or PMA (50 nM), TLR3^−/− and WT (TLR3^+/+) macrophages produced similar amount of extracellular superoxide (F). TLR3^−/− and WT (TLR3^+/+) macrophages were pretreated with vehicle or apocynin (0.1 mM) for 15 min, and secreted TNF-α and IL-12p40 were measured 24h after the cells were treated with 50 μg/ml poly I:C (G). Data are shown as means ± SEM from three independent experiments in triplicate. *p<0.01, compared with corresponding vehicle-treated controls; #p<0.01, compared with the corresponding treatment groups in WT.

Figure 7. Model illustration of the role of Mac-1 in dsRNA-mediated signaling in the macrophage

The membrane receptor TLR3 recognizes and binds internalized extracellular dsRNA only in acidified endosomes, and then activated TLR3 induces type I interferon production via interferon regulatory factor 3 (IRF3) and proinflammatory cytokine generation via the NFκB pathway. The present study identifies Mac-1 (CD11b/CD18 or CR3) as a novel PRR on the surface of macrophages. Mac-1 senses and binds extracellular dsRNA to facilitate the endocytosis of extracellular dsRNA thereby amplifying the TLR3-dependent signaling or to activate oxidative enzyme NOX2 to produce ROS thereby activating MAPK and NFκB pathways to induce proinflammatory cytokine production in a TLR3-independent manner.