The role of microRNAs miR-200b and miR-200c in TLR4 signaling and NF-κB activation

Abstract

Recognition of microbial products by members of the Toll-like receptor (TLR) family initiates intracellular signaling cascades that result in NF-κB activation and subsequent production of inflammatory cytokines. We explored the potential roles of microRNAs (miRNAs) in regulating TLR pathways. A target analysis approach to the TLR4 pathway adaptor molecules identified several putative targets of miR-200a, miR-200b and miR-200c. miRNA mimics were co-transfected with a NF-κB activity reporter plasmid into HEK293 cells stably expressing TLR4 (HEK293-TLR4). Mimics of both miR-200b and miR-200c, but not miR-200a, decreased NF-κB reporter activity in either untreated cells or in cells treated with endotoxin:MD2 as a TLR4 agonist. Transfection of HEK293-TLR4 cells with miR-200b or miR-200c significantly decreased expression of MyD88, whereas TLR4, IRAK-1 and TRAF-6 mRNAs were unaffected. When miR-200b or miR-200c mimics were transfected into the differentiated monocytic THP-1 cell line, the abundance of MyD88 transcripts, as well as LPS-induced expression of the pro-inflammatory molecules IL-6, CXCL9 and TNF-α were diminished. These data define miRNAs miR-200b and miR-200c as factors that modify the efficiency of TLR4 signaling through the MyD88-dependent pathway and can thus affect host innate defenses against microbial pathogens.

Figure 1

Putative miRNA target proteins in the signaling pathways activated after TLR4 ligation or TNF-α exposure. Components of the signaling pathways activated by TLR4 ligation or TNF-α exposure, which ultimately lead to NF-κB activation, are shown. Please note the MyD88-dependent and -independent pathways resulting in NF-κB activation after TLR4 ligation. Potential targets for miR-200a (+) or for miR-200b and miR-200c (*) within the TLR4 and TNF-α pathways, identified through a miRNA query of 3′UTRs, are indicated. Analysis of the IRAK1 3′UTR revealed a 7 bp exact complementary sequence match to the miR-200b and miR-200c seed sequence, even though IRAK1 was not a predicted miRNA200b target in TargetScan version 6.0. Experimentally validated miR-146a targets are indicated.

Figure 2

Effects of miRNA mimics on basal NF-κB activity. HEK293-TLR4 cells were co-transfected with an NF-κB reporter plasmid and Renilla luciferase expression plasmid, and incubated for 48 h in the presence of 50 nM of the indicated miRNA mimics. Data indicate the mean ± SEM ratio of firefly/Renilla luciferase activity in transfected cells, normalized to negative control miRNA mimic (first bar). miR-146a, which has been reported to be a repressor of NF-κB activation, served as a positive control. miR-29b, which has no reported effect on NF-κB activation, served as a negative control. The bars represent the ratios of firefly/Renilla luciferase activity in the absence of activating stimulus. Data were generated from three independent experiments, each with three replicates per condition. *P < 0.05 compared with negative control miRNA, unpaired t-test.

Figure 3

Effect of miRNA mimics on LPS-stimulated NF-κB activation. HEK293-TLR4 cells were co-transfected with an NF-κB reporter plasmid, a transfection control Renilla luciferase expression plasmid, and the indicated miRNA mimic at 50 nM final concentration. The latter included a negative control miRNA mimic, or miRNA mimics for miR-146a (A), miR-200b (B) or miR-200c (C). After 48 h cultivation, cells were stimulated with buffer alone (none), the indicated doses of LOS:MD2 or TNF-α and cultured for an additional 6 h. Data indicate the mean±SEM firefly/Renilla luciferase activity compared with the negative control of uninduced miRNA from three independent experiments, each with three replicates per condition. *P < 0.05, **P < 0.005 comparing the test with the negative control miRNA unpaired t-test. LOS, lipooligosaccharide.

Figure 4

MyD88 luciferase assays validating a direct interaction between miR-200b or miR-200c and MyD88. HEK293-TLR4 cells were transfected with constructs containing the luciferase reporter gene followed by either the 3′UTR of MyD88 or a mutated version of the MyD88 3′UTR containing mutations within the predicted miR-200b and miR-200c seed region. (A) Cells were co-transfected with 25 nM of miRNA mimics for miR-200b, miR-200c or a negative control. Data represent the mean ± SEM Renilla/firefly luciferase ratios. (B) TargetScan predicted interaction between miR-200b and miR-200c and the putative MyD88 3′UTR. Shaded bases represent miRNA seed regions. Mutated bases are underlined. (C) HEK293-TLR4 cells were transfected with miRNA mimics for miR 200b, miR-200c or a negative control. RNA harvested at 48 h post-transfection was used as template for RT-qPCR to measure MyD88 transcript levels. Data are mean ± SEM of three independent experiments with three replicates per condition. *P < 0.05, **P < 0.001.

Figure 5

LPS-stimulated cytokine/chemokine induction is repressed by miR-200b or miR-200c mimics. Repression of MyD88 by miR 200b or miR-200c leads to a decrease in classical macrophage activation markers. miRNA mimics (50 nM) of miR-200b or miR-200c were transfected into differentiated THP-1 cells. (A) Twenty-four hours post-transfection, RNA was harvested and RT-qPCR was performed using TaqMan primers for β-actin and MyD88 with β-actin serving as an internal control. Data represent the mean ± SEM fold change in expression comparing experimental to negative control miRNA mimics from three independent experiments. (B) Forty-eight h post-transfection, cells were treated with 100 ng/ml LPS and RNA was harvested after an additional 3 h. RT-qPCR was performed using TaqMan primers for β-actin, IL-6, CXCL9, or TNF-α. Relative mRNA abundance was determined by normalizing to the b-actin cDNA, then comparing cDNA abundance in miR mimic-transfected compared to the negative control miRNA mimic transfected samples. Data represent the mean ± SEM relative expression from three independent experiments. *P < 0.05, **P < 0.01.