An NF-κB-microRNA regulatory network tunes macrophage inflammatory responses

Abstract

The innate inflammatory response must be tightly regulated to ensure effective immune protection. NF-κB is a key mediator of the inflammatory response, and its dysregulation has been associated with immune-related malignancies. Here, we describe a miRNA-based regulatory network that enables precise NF-κB activity in mouse macrophages. Elevated miR-155 expression potentiates NF-κB activity in miR-146a-deficient mice, leading to both an overactive acute inflammatory response and chronic inflammation. Enforced miR-155 expression overrides miR-146a-mediated repression of NF-κB activation, thus emphasizing the dominant function of miR-155 in promoting inflammation. Moreover, miR-155-deficient macrophages exhibit a suboptimal inflammatory response when exposed to low levels of inflammatory stimuli. Importantly, we demonstrate a temporal asymmetry between miR-155 and miR-146a expression during macrophage activation, which creates a combined positive and negative feedback network controlling NF-κB activity. This miRNA-based regulatory network enables a robust yet time-limited inflammatory response essential for functional immunity.MicroRNAs (miR) are important regulators of gene transcription, with miR-155 and miR-146a both implicated in macrophage activation. Here the authors show that NF-κB signalling, miR-155 and miR-146a form a complex network of cross-regulations to control gene transcription in macrophages for modulating inflammatory responses.

Conflict of interest statement

The authors declare no competing financial interests.

Figures

Fig. 1

miR-155 is required for myeloproliferation and extramedullary haematopoiesis in aged miR-146a−/− mice. a, b Twelve-month-old WT, miR-155−/−, miR-146a−/− and double knock out (DKO) mice were analysed for serum IL-6 levels (a), spleen mRNA levels of IL-1b, IL-6 and TNF (b). c–f Twelve-month-old WT, miR-155−/−, miR-146a−/−, DKO and LyzM-Cre miR-146afl/fl mice were analysed for spleen weight (c), relative macrophage percentage in peripheral blood (d) and spleen (e), as well as extramedullary haematopoiesis by HSC quantification in spleen (f). N = 10 (a, b); N = 7 (c, d) per group, from at least two independent experiments and are represented as mean ± SEM. *p < 0.05, **p < 0.01 and ***p < 0.001 using one-way ANOVA

Fig. 2

miR-155 expression is required for the elevated acute inflammatory response in miR-146a deficient mice. a–c Eight to ten-week-old WT, miR-155−/−, miR-146a−/− and double knock out (DKO) mice were infected with Listeria monocytogenes. Seventy-two hours after infection, mice were analysed for colony formation units (CFUs) in spleen (a) and liver (b), as well as serum IL-6 levels (c). Eight to ten-week-old WT, miR-155−/−, miR-146a−/− and DKO mice received 1 mg/kg LPS every 24 h for 3 days via IP injections. Serum IL-6 levels (d) as well as peripheral blood macrophage activation was quantified 4 h after each injection by of CD11b+ MHC-II high (e) and CD11b+ CD80+ percentage (f). Arrows indicate time of LPS injections. g–h Time course expression profile of miR-155 of WT and miR-146a−/− (g) and miR-146a of WT and miR-155−/− BMMs (h) after LPS stimulation (100 ng/mL) were quantified by qPCR. N > 7 (a–d), or N > 9 (e–h) per group from two independent experiments and are represented as mean ± SEM. *p < 0.05, **p < 0.01 and ***p < 0.001 using one-way ANOVA (a–d) or two-way ANOVA (d–h)

Fig. 3

MiR-155 overexpression overrides miR-146a repression of NF-κB, leading to NF-κB activation and myeloproliferation. a, b Mice reconstituted with bone marrow transduced with GFP and control (MG), miR-155, miR-146a or both miR-155 and miR-146a (dmiR) were analysed 4 months after reconstitution for (a) CD11b+ peripheral blood macrophages and (b) spleen weight. c–e Mice reconstituted with bone marrow from NF-κB reporter mice, transduced with control (MG), miR-155, miR-146a or dmiR were analysed 3 months after reconstitution. (c) Basal NF-κB activity of CD11b+ peripheral blood macrophages. (d) Mice received 1 mg/kg LPS every 24 h for 3 days via IP injections. Peripheral blood macrophage NF-κB activation was quantified 4 h after each injection by GFP expression, as well as serum IL-6 levels (e). Dashed line represent ELISA detection sensitivity. N = 7 per group from three (a–c) or two (d, e) independent experiments and are represented as percentage (a) or mean ± SEM. *p < 0.05, **p < 0.01, using one-way ANOVA (b, c) or two-way ANOVA (d–e)

Fig. 4

Molecular characterization of miR-146a and miR-155 KO BMMs reveal SHIP1 and SOCS1 as the major miR-155 targets. mRNA from BMMs of WT, miR-155−/−, miR-146a−/− and double knock out (DKO) mice before and 8 h after LPS stimulation was subjected to RNA sequencing. a Heat map, indicating the number of differentially expressed genes between samples of all strains before (top) and 8 h after LPS stimulation (bottom). b Enriched functional annotations for genes upregulated in stimulated miR-146a−/− BMMs compared to WT 8 h after LPS stimulation. c qPCR quantification of potential miR-146a and miR-155 targets in BMMs 24 h after LPS stimulation. d Protein from BMMs of WT, miR-155−/−, miR-146a−/− and dKO mice was subjected to western blot analysis 24 h after LPS stimulation. e–h Mice were reconstituted with bone marrow from WT, miR-155−/−, miR-146a−/− or dKO mice transduced with GFP expressing control (MG), SHIP-1 or SOCS-1 shRNA vectors. e Reconstitution competitiveness of SOCS1 and SHIP1 attenuated WT bone marrow cells compared to control as measured by GFP expressing CD45+ cells. f Ki67 and Hochst proliferative state of CD45+ WT cells transduced with SOCS1 shRNA compared to control. g CD11b+ GFP+ peripheral blood macrophage percentage of WT cells transduced with control, SHIP1 shRNA or SOCS1 shRNA. h CD11b+ peripheral macrophage percentage comparison between WT, miR-155−/−, miR-146a−/− and dKO mice transduced with SHIP1 shRNA vector. (a, b) N = 10, (c, d) N = 8 from three independent experiments, (e–h) N > 7 per group from two independent experiments. c, e, g, h Presented mean ± SEM. *p < 0.05, **p < 0.01, using one-way ANOVA

Fig. 5

SHIP1 downregulation restores miR-146a−/− phenotype in DKO macrophages. WT, miR-155−/−, miR-146a−/− and double knock out (DKO) bone marrow cells transduced with control (MG) (left) or SHIP1 shRNA (right) were reconstituted into recipient mice. Three months after reconstitution, the mice received 1 mg/kg LPS every 24 h for 3 days via IP injections. Peripheral blood macrophage CD11b+ percentage (a, b), MHC-II high— (c, d) and CD80+ (e, f) were compared for all strains by FACS analysis. Arrows indicate time of LPS injections. N > 7 per group from two independent experiments, represented as mean ± SEM. *p < 0.05, using two-way ANOVA

Fig. 6

Mir-155 acts as an amplifier in suboptimal stimulation. WT, miR-155−/−, miR-146a−/− and double knock out (DKO) BMMs were stimulated with low levels of LPS (10 ng/mL) (a) or Pam3CSK4 (5 ng/mL) (b). mRNA levels of IL-1β (left) and IL-6 (right) were then quantified using qRT-PCR. c TNF and phospho-p65 (P-p65) protein expression of BMMs from WT (grey), miR-155−/− (blue), miR-146a−/− (red) and dKO (green) 4 h after stimulation with low levels of LPS (10 ng/mL) as measured by FACS using intracellular staining. N = 4 per group from at least two independent experiments, represented as mean ± SEM. a, b *p < 0.05, using two-way ANOVA

Fig. 7

MiR-155 and miR-146a form a combined positive and negative auto regulatory loop to control precise NF-κB activity during inflammatory stimuli. The dynamic expression of miR-155, miR-146a, and the NF-κB targets IL-1β and IL-6 at different time points after LPS (a) and Pam3CSK4 (b) stimulation. The dynamic expression of IL-6 (c) and IL-1β (d) mRNA in WT BMMs transfected with miR-146a specific anti miR (red) or control (black) 2 h post LPS stimulation (100 ng/mL). e A suggested model for the miR-146a and miR-155 regulatory loop during inflammatory stimulation. a, bN = 4 per group from two independent experiments. *p < 0.05, using one-way ANOVA (c, d)