PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
 
J Immunol. Author manuscript; available in PMC 2014 February 1.
Published in final edited form as:
PMCID: PMC3552145
NIHMSID: NIHMS425513

Regulation of TLR2-mediated tolerance and cross-tolerance through IRAK4 modulation by miR-132/-212

Abstract

Innate immune response is the first defense against pathogens via recognition by various conserved pattern recognition receptors, such as Toll-like receptors (TLRs), to initiate a rapid and strong cytokine alarm. TLR signaling-mediated cytokine production must be properly regulated to prevent pathological conditions deriving from overproduction of cytokines. In this report, the role of specific microRNAs in TLR-signaling pathway was investigated to reveal the cross-interaction and -regulation in the MyD88 pathway. In peptidoglycan (PGN)/TLR2-stimulated THP-1 monocytes, PBMCs, and primary macrophages showed rapid and dramatic miR-132 and miR-212 (miR-132/-212) upregulation. This newly identified response appeared earlier in time than the characteristic miR-146a response in lipopolysaccharide (LPS)-TLR4 stimulation. The rapid induction of miR-132/-212 was transcription factor CREB-dependent and the sustained expression of miR-132/-212 was responsible for inducing tolerance to subsequent PGN challenge. Cross-tolerance was observed by TLR5 ligand flagellin and heat-killed or live bacteria resulting from miR-132/-212 upregulation. Mechanistically, IRAK4 was identified and validated as a target of miR-132/-212 by luciferase reporter assay and seed-sequence mutagenesis of the reporter. Transfection of miR-132 or miR-212 alone mimicked PGN tolerance in monocytes while transfected specific miRNA inhibitors tampered the tolerance effect. During bacterial infection, PGN-mediated TLR2-signaling induces miR-132/-212 to downregulate IRAK4, an early component in the MyD88-dependent pathway, while LPS/TLR4-induced miR-146a downregulates downstream components of the same MyD88-dependent pathway. The identification of miR-132/-212 and miR-146a together to prevent damaging consequences from the overproduction of proinflammatory cytokines by targeting a common signaling pathway is significant and will provide insights into future design and development of therapeutics.

Keywords: innate immunity, lipopolysaccharide, microRNA, peptidoglycan, TLR ligands

INTRODUCTION

Innate immune system is the primary defense mechanism that is ignited shortly after pathogenic invaders are detected by various conserved pattern recognition receptors. One such class of well-documented receptors is toll-like receptors (TLRs). These receptors are efficient in detecting signature molecules including lipopeptides, peptidoglycan (PGN), lipopolysaccharide (LPS), flagellin, and nucleic acids (1). Upon detection, the subsequent signaling events elicit a core set of stereotyped responses, including cytokines, chemokines, and adhesion molecules, most notably through activation of NF-kB transcription factor, which leads the immune system to sense and react to infection. In contrast, pathological dysregulation of this process is a hallmark of inflammatory damage, autoimmune diseases, and possibly cancer (2). Therefore, innate immune response involving TLR signaling cascades must be tightly regulated by elaborate mechanisms to control its onset and termination. It is acknowledged that TLR4 signaling events have been extensively studied both in vivo and in vitro in terms of endotoxin tolerance, which limits the pathogenic effects of LPS (37). Other microbial components such as PGN (a potent TLR2 agonist) are also involved in priming of innate immune cells (8, 9). To explain this tolerance mechanism, a number of negative regulatory controllers have been proposed (10). These include soluble decoy receptors for TLR4, IRAKM, A20, TRIM30α, and splice variants of signal-transduction proteins such as MyD88-s (1115). However, at this time, there is no consensus on the molecular mechanisms involved to resolve inflammation.

MicroRNAs (miRNAs), short noncoding RNA, have emerged recently as key regulators of gene expression acting at the posttranscriptional level (16). MiRNAs have been shown to be critical in many biological processes, ranging from development to differentiation and including regulation of the mammalian immune system (17, 18). A few miRNAs are induced in innate immune cells in response to cognate TLR ligands, with a consensus emerging that miR-146a, miR-155, and miR-21 are important to negatively regulate the activation of inflammatory pathways in myeloid cells (15, 18, 19). Although miR-146a regulation of IRAK1 and TRAF6 adaptor molecules has been shown to play a major role in endotoxin tolerance and cross-tolerance, cytokine response is not extinguished completely, suggesting the possible involvement of other miRNAs in this intricate process (20, 21). It is well established that recruitment of adaptor kinases are the prime factor for triggering a TLR signaling cascade. Upon TLRs activation, IL-1 receptor-associated kinase 4 (IRAK4) is known to be recruited to MyD88, forming a helical assembly of the MyD88-IRAK4-IRAK2/1 complex that further activates TRAF6 and eventually leads to NF-κB activation for inflammatory gene transcription (22). Thus, IRAK4 should be the pivotal adaptor kinase used by all TLR signaling (except TLR3). In this connection, compared to IRAK1 knockdown, the knockdown of IRAK4 renders immune cells much less responsive to TLR agonists (23). Phenotypically similar to mice lacking MyD88, IRAK4 knockout mice show severe impairment of IL-1 and TLR signaling (24). Based on these reports, regulation of adaptor kinases might be an important molecular mechanism for maintaining cytokine response in a controlled manner. Although IRAK1 and TRAF6 are known to be regulated by miR-146a (25), no such miRNA-mediated regulation of IRAK4 has been documented. IRAK4 has been found to be a putative target of miR-132 and miR-212 by bioinformatics analysis using TargetScan (TargetScan.org), but this has not been experimentally validated. Accordingly, in a very recent review, it is still unknown whether signaling molecules in TLR pathways are targeted by miR-132 and miR-212 (26).

Mature miR-132 and miR-212, sharing the same seed sequence, are processed from a single non-coding gene transcript regulated primarily by the cyclic AMP-response-element-binding (CREB) transcriptional factor (27, 28). The function for these miRNAs has been described in a few studies. miR-132 has been shown to regulate neuronal morphogenesis and the dendritic plasticity of cultured neurons (27, 29). miR-132 may also be responsible for limiting inflammation in the mouse brain by targeting acetylcholinesterase (AChE) (30). miR-132 can also modulate inflammation induced by early stage Kaposi's sarcoma-associated herpesvirus (KSHV) infection (31). miR-212 can interfere with the craving for cocaine in mice (32) and acts as a tumor suppressor (33). To date no detailed expression kinetic of miR-132 or miR-212 has been described in response to innate immune ligands associated with TLR ligand-induced tolerance.

Our study shows the first evidence that the exposure of innate immune cells to PGN, Pam, flagellin, or whole bacteria induces rapid expression of mature miR-132 and miR-212. This report highlights the importance of investigating their mechanistic role in innate immunity in the context of TLR2 ligand-induced tolerance, which can modulate innate immune system.

Materials and methods

Reagents

Ultrapure TLR-grade lipopolysaccharide (Salmonella enterica serotype Minnesota Re595), lipoteichoic acid (Staphylococcus aureus) were from Sigma-Aldrich (St. Louis, MO). Peptidoglycan (Escherichia coli 0111:B4), synthetic bacterial lipoprotein Pam3CSK4CysSerLys4, LPS from Porphyromonas gingivalis ATCC 33277, poly(I:C), and recombinant flagellin (Salmonella typhimurium) were from InvivoGen (San Diego, CA). siGENOME SMARTpool siRNA for IRAK4, CREB, and Lamin A/C (LMNA) were from Dharmacon (Lafayette, CO). All miRNA-mimics, non-specific (NS) miRNA-mimic negative control, and inhibitors (anti-miRNA inhibitor) were from Ambion (Austin, TX). Antibodies to human IRAK1, IRAK4, CREB, pCREB, NF-kB p65 (sc-372), and ERK 1/2 (sc-135900) were from Santa Cruz Biotechnology (Santa Cruz, CA). Antibodies to p300 was from Abcam (ab14984, Cambridge, MA). Kinase inhibitors PD98059 and U0126 were from Calbiochem (San Diego, CA).

Cell culture and innate immune ligand stimulation

Human THP-1, HEK293, and murine RAW264.7 cells were obtained from the American Type Culture Collection (ATCC, Manassas, VA). All cells were maintained in either RPMI or DMEM media containing 10% (v/v) FBS (Mediatech, Manassas, VA), and 100 U/ml penicillin-streptomycin (Mediatech). For analysis of THP-1 monocyte response to microbial ligand in vitro, log phase cells were seeded at 5×105 cells/ml in a 24-well plate. Unless otherwise mentioned, cells were stimulated with the following agonists: 1000 ng/ml of LPS from S. enterica (LPS Se, TLR4 ligand), Pam3CSK4CysSerLys4 (Pam, TLR2/TLR1 ligand), Peptidoglycan (PGN, TLR2 ligand), LPS from P. gingivalis (LPS Pg, TLR2 ligand), lipoteichoic acid (LTA, TLR2 ligand), and 300 ng/ml recombinant flagellin (TLR5 ligand). TLR ligands were reconstituted in endotoxin-free water and used at concentrations as reported before (20).

For analysis of human peripheral blood mononuclear cells (PBMCs) response, mononuclear cells were isolated from whole blood by a Ficoll gradient (GE Healthcare, Uppsala, Sweden) and were grown in RPMI media as described above. PBMCs (106 cells/ml) were then stimulated for 6–24 h with PGN (0–5 μg/ml).

Mouse macrophages were from, 3-mo-old female C57BL/6 mice, which were intraperitoneally injected with 0.5 ml of 4% sodium thioglycollate 3 d earlier. Peritoneal cells (~90% macrophages) were harvested by lavaging the peritoneal cavity and grown in DMEM as described above. After 5 h, cells were washed with growth medium and then stimulated for 6–24 h with PGN (0–10 μg/ml).

TLR ligands-induced tolerance

PGN-, Pam- or flagellin-induced tolerance and/or cross-tolerance experiments were performed using a THP-1 monocyte model, adapted from methods described previously (9, 20) with minor modifications. Briefly, before starting tolerance assays, THP-1 cells were cultured until they were in log phase and reached a density of 106 cells/ml. In all experiments, trypan blue exclusions were performed to verify cell viability and cells that were >95% viable were considered for study. THP-1 cells were then transferred to fresh complete medium at 5×105 cells/ml. Cells were incubated with a low dose of PGN (100 or 500 ng/ml) or Pam (100 ng/ml) for 18 h. In some tolerance assays, THP-1 cells were primed for 18 h with flagellin (200 ng/ml) or heat-killed (HK) Tannerella forsythia at multiplicity of infection (MOI) 100. After two washes with tissue culture-grade PBS, primed cells were re-stimulated with various ligands or cultured without stimulation. For bacteria-induced tolerance study, heat-killed P. gingivalis FDC 381, Treponema denticola ATCC 35404, and T. forsythia ATCC 43037 were prepared as described previously (34). For human PBMCs and RAW264.7 tolerance assays, cells were primed with PGN (500 ng/ml) or Pam (100 ng/ml) for 18 h, followed by challenge with PGN and LPS (1 μg/ml). For the mouse primary macrophages tolerance assay, primary cells were primed with PGN (100 ng/ml), followed by washing and challenging with PGN, Pam, or LPS. In all cases, after 3–24 h secondary challenge, supernatants were harvested and stored at −80°C until assays were performed for secreted inflammatory mediators.

RNA extraction and real-time RT-PCR

Total RNA from microbial ligands-treated and untreated THP-1 cells were prepared using the mirVana miRNA isolation kit (Ambion). For miRNA analysis, 6.7 ng RNA of each sample was used for quantitative stem-loop reverse transcription and real-time PCR (qRT-PCR). Quantification of expression of mature miRNAs was performed using the TaqMan microRNA RT kit, TaqMan Universal PCR Master Mix, and TaqMan miRNA assay primers of interest for human or mouse miRNAs (Applied Biosystems, Carlsbad, CA). For gene expression analysis, cDNA was prepared with the High Capacity cDNA Reverse Transcription kit (Applied Biosystems) and individual mRNA was monitored with the following inventoried Taqman assays (Applied Biosystems): human IRAK1, IRAK4, CREB, and p300 with 33 ng total RNA per reaction. The cycle threshold (Ct) values, corresponding to the PCR cycle number at which fluorescence emission reaches a threshold above baseline emission, were determined, and miRNA expression values were calculated using human RNU44 or mouse SnoRNA202 (Applied Biosystems) as an endogenous reference following the 2−ΔΔCt method (35). mRNA for gene expression values were quantified in the same way after normalization to mammalian 18S rRNA. Standard curves for mature miR-132 and miR-146a were prepared by qRT-PCR analysis using synthetic mature miRNA (IDT, Coralville, IA). The Ct values were determined by qRT-PCR analysis of the total RNA from PGN (2500 ng/ml) treated cells and then converted to miR-132 and miR-146a copy numbers using the standard curve.

Transient transfection

miR-132, miR-212, and miR-146a functional analyses were performed by transfecting synthetic mimic or inhibitor (40 nM) in THP-1 monocytes using lipofectamine 2000 (Invitrogen, Carlsbad, CA) as described (20, 21). THP-1 cells were transfected with siRNA targeting IRAK4, CREB, or LMNA using the same above protocol. For luciferase assays, HEK293 cells were plated in 24-well plate at 105 cells/well and transfected 24 h later by 3% lipofectamine reagent.

Luciferase reporter assay

Complete 3'UTR of IRAK4 was subcloned downstream of firefly luciferase coding sequence in pMiRTarget vector (IRAK4-wt, Origene Technologies, Rockville, MD) and then this reporter (50 ng) and renilla luciferase reporter (0.1 ng) were co-transfected together with 100 nM of miR-132-mimic, miR-212-mimic, or miR-146a-mimic into HEK293 cells for 48 h. A mutated version of this construct (IRAK4-mut) carrying 4-bp substitutions in the putative miR-132/-212 seed sequence target site was obtained by using the site-directed mutagenesis kit (Agilent Technologies, Santa Clara, CA). Reporter luciferase activities were measured using the Dual-Luciferase kit (Promega, Madison, WI) 48 h after transfection. To determine the functional regulation of IRAK4 3'UTR during PGN stimulation, IRAK4-wt or IRAK4-mut together with the renilla luciferase reporter were co-transfected into THP-1 cells for 6 h. Cell lysates were harvested 12 h after PGN stimulation (2 μg/ml) for the measurement of luciferase activities.

ELISA

Cytokine concentrations in cell culture supernatants were measured by ELISA using OptEIA cytokine kits (BD Biosciences, Franklin Lakes, NJ) and the DuoSet Development system (R&D Systems, Minneapolis, MN) following the manufacturers' instructions.

Western blot analysis

PGN-, Pam- or flagellin-primed and unprimed THP-1 cells (5 ×106/condition) were collected 2 h after secondary challenge with ligands and then lysed with lysis buffer containing Complete Protease Inhibitors Cocktail (Roche, Indianapolis, IN) as described1. THP-1 cell lysate pre-treated with or without PD98059 (MEK1 inhibitor, 50 μM) and U0126 (MEK1/2 inhibitor, 0.5 μM), were similarly prepared. Soluble lysates were quantitated for protein concentration (Bio-Rad Bradford protein assay), separated by 10% SDS-PAGE, and electrotransferred to a polyvinylidene difluoride membrane. The membranes were blocked for 1 h at room temperature with 5% nonfat milk in PBS/0.05% Tween 20 (PBS-T) and were probed with rabbit polyclonal antibody anti-IRAK1 (1:300), anti-IRAK4 (1:300), and mouse monoclonal anti-tubulin (1:5000, Sigma-Aldrich). The membranes were then washed with PBS-T and incubated for 1 h with goat anti-rabbit or anti-mouse IgG-HRP (1:5000, Southern Biotech, Birmingham, AL). After washing with PBS-T, reactive protein bands were visualized by SuperSignal Pico chemiluminescent reagent (Pierce, Rockford, IL). Similarly, CREB and pCREB were visualized using specific antibodies at 1:300 dilutions. Integrated density of protein bands in the scanned image of Western blot film was analyzed using IMAGE-J software and normalized to tubulin or CREB in each lane, and is presented relative to results obtained with the control sample (percentage of fraction), which was set as 1.0.

Ethics Statement

Experiments involving mice were approved by the institutional animal care and use committee of the University of Florida. The protocol for blood collection from healthy controls was approved by the Institutional Review Board (IRB) and this study meets and is in compliance with all ethical standards in medicine and informed consent was obtained from all patients according to the Declaration of Helsinki.

Statistical analysis

Student's t-test (two-tailed) was used to compare data between groups, except where mentioned otherwise. Prism for Windows, version 5.0 (GraphPad Software, San Diego, CA) was used, and P < 0.05 was considered statistically significant.

Results

TLR ligand-induced TNF-α secretion and kinetics of miR-132 and miR-212 expression in innate immune cells

Emerging data have shown that microbial ligands are recognized through specific receptors found on innate immune cells. Beside LPS, PGN is known to be a potent inducer of a diverse array of inflammatory mediators, including TNF-α both in vitro and in vivo. To monitor TNF-α production in in vitro tissue culture after PGN stimulation, the most commonly employed acute monocytic leukemia cell line THP-1 was used. TNF-α was detected in culture supernatants of PGN-treated THP-1 monocytes starting at 2 h and peaked at 8 h post-stimulation, followed by gradual decrease thereafter (Figure 1A). The progressive changes of TNF-α secretions showed a dose- and time-dependent pattern, similar to LPS stimulation (21). The highest dose of PGN (5 μg/ml) induced the highest level of TNF-α, up to 12 ng/ml at 8 h. The kinetics of TNF-α mRNA expression was also measured by quantitative real-time PCR (qRT-PCR) and was consistent with the kinetics of the secreted TNF-α protein (data not shown).

Figure 1
PGN induction of TNF-α and miRNA expression kinetics in monocytes/macrophages. (A) Dose-response and time-course analysis of TNF-α secretion in culture supernatant by THP-1 monocytes stimulated with 0–5 μg/ml PGN for 2–48 ...

In our recent studies, LPS-stimulated THP-1 monocytes showed a continuous expression of miR-146a, which was demonstrated as critical in LPS tolerance and cross-tolerance (20, 21). Similar to induction by LPS, an up to 25-fold increase of miR-146a expression was noted at 48 h with the highest concentration of PGN tested in THP-1 cells (Figure 1B); for comparison, LPS stimulation of THP-1 cells often showed >100-fold increase of miR-146a (21). LPS stimulated THP-1 monocytes are known to produce other miRNAs, including miR-132 and miR-155 (25). Interestingly and in sharp contrast, miR-132 and miR-212 showed more impressive 110- and 65-fold increased expression in PGN-treated cells, respectively, at their peak (Figure 1B). PGN treatment resulted in an early increase in miR-132 and miR-212 levels at 4 h reaching 60- and 43-fold at the 5 μg/ml PGN dose, respectively, while the change in levels for miR-146a was only 7-fold (Figure 1B). Although miR-132 and miR-212 are derived from the same primary transcript, the reason for the difference in the lower fold induction in miR-212 compared to miR-132 is not clear. However, a similar lower expression in miR-212 compared to miR-132 has also been observed in KSHV-stimulated endothelial cells (31). Other miRNAs miR-155 (Figure 1B) and miR-16 (data not shown) showed little or no change under the same conditions although both miRNAs are expressed in relatively abundant levels in THP-1 cells. The exact copy numbers of both miR-132 and miR-146a in the PGN-treated cells (same data as in Figure 1B) were determined using synthetic miR-132 and miR-146a as standards for the qRT-PCR assay (Supplementary Figure S1A). Each unstimulated THP-1 cell showed only ~2 copies of miR-132 and ~30 copies of miR-146a. Upon PGN stimulation, each THP-1 cell could produce ~100 copies of miR-132 and ~ 600 copies of miR-146a at 12 h (Supplementary Figure S1A). In contrast, LPS-stimulated THP-1 monocytes showed significantly lower miR-132 (10-fold) and miR-212 (3-fold) at 12 h (Supplementary Figure S1B) than when stimulated with PGN. The above results indicate that PGN is a potent inducer for both miR-132 and miR-212, which were focused for further investigation.

To understand the physiological relevance of induction of miR-132, a range of other cell types were used. Human PBMCs were treated with PGN from 6 to 24 h. PBMCs showed dose-dependent production of TNF-α up to 3 ng/ml (Figure 1C) and upregulation of miR-132 up to 6-fold at 12 h (Figure 1D). A similar increase of miR-132 expression (~5-fold at 24 h) was also observed in mouse macrophage cell line RAW264.7 in response to PGN, Pam3CSK4 (Pam), and LPS (Figure 1E). Similar results were also obtained in mouse primary peritoneal macrophages, which showed high levels of TNF-α production up to 10 ng/ml peaking at 6 h (Figure 1F) and similar miR-132 expression (5-fold miR-132 upregulation peaking at 6 h) in response to PGN (Figure 1G). It is noted that the expression fold change of miR-132 in these cells varied and were lower than in THP-1 monocytes. A similar variation of miR-146a expression has been observed in various cells (15). The variation may be attributed to the difference in cell types (monocytes vs macrophages) and cell population (homologous vs mixed cell). To further support the ability of PGN to induce miR-132/-212 expression, effects of a common synthetic PGN derivative Pam was also carefully examined in this study. Similar to PGN, Pam induced TNF-α production (Supplementary Figure S1C) and miR-132 and miR-212 expression (Supplementary Figure S1D,E). The fold change for miR-132 was notably higher than those of miR-146a (Supplementary Figure S1D) or miR-155 (Supplementary Figure S1E). LPS from P. gingivalis (TLR2 ligand) also showed similar patterns of miR-132 and miR-212 expression (data not shown). The capacity of TLR3 and TLR5 ligand to induce miR-132 and miR-212 were also examined. TLR3 ligand poly(I:C) (25 μg/ml) did not induce any of these miRNAs (data not shown), indicating that innate immune responses that involve these miRNA expression are associated with the MyD88-dependent pathway rather than the TLR3-TRIF pathway. TLR5 agonist flagellin stimulation induced a rapid production of TNF-α (Supplementary Figure S1F) and, at 8 h, also showed greater increases of miR-132 (55-fold) and miR-212 (27-fold) than miR-146a (Supplementary Figure S1G). Together these findings suggest the capacity of PGN, Pam, and flagellin to induce significant levels of miR-132 and miR-212 in THP-1 monocytes, human PBMCs, and mouse primary macrophages. To gain a better understanding on the dynamic nature and robust expression kinetics, their biological significance was investigated in the subsequent studies.

CREB-dependent rapid induction of miR-132

In the previous time-course experiment, miR-132 was rapidly induced compared to miR-146a (Figure 1B). miR-132 has been shown to be regulated by a key transcription factor, CREB (27). CREB is activated in THP-1 monocytes by phosphorylation at Ser 133 (pCREB) within minutes after exposure to PGN or Pam. Notably, at 15 min, PGN induced a 4.17-fold increase in pCREB compared to untreated control (UTX) by Western blot analysis (Figure 2A), consistent with previous reports (27, 36). Pam also showed similar induction in pCREB (Figure 2A). In contrast, TLR3 ligand poly(I:C) stimulation of THP-1 cells did not induce pCREB (data not shown) and little or no miR-132 induction was observed as mentioned above. It is noted that although the timing of PGN-induced pCREB was correlated with the initiation of miR-132/-212 expression, at 4 h the level of pCREB was back to baseline untreated level while miR-132/-212 expression continued to increase until 12 h (Figure 1B). Although the reason is not clear, pCREB might be only required to initiate the transcription of miR-132/-212. A similar observation was reported in KSHV-infected endothelial cells; the initiation of pCREB correlated with miR-132 induction but the level of pCREB decreased while the expression of miR-132 was maintained at similar levels (31).

Figure 2
Rapid induction of miR-132 and miR-212 in PGN-/Pam-stimulated THP-1 monocytes is mediated through CREB-dependent machinery. (A) Immunoblot analysis demonstrating phosphorylation of CREB (Ser 133, pCREB) within 15 min after PGN (top) or Pam (bottom) stimulation ...

The requirement of CREB in the PGN-induced miR-132/-212 expression was further investigated. In THP-1 cells with siRNA-mediated knockdown of CREB, the levels of CREB mRNA was reduced by ~50% and protein by 70% as demonstrated by qRT-PCR (Figure 2B) and Western blot analysis (Figure 2C), respectively. Transfection of unrelated control siRNA for Lamin A/C did not affect CREB mRNA level (Figure 2B). Knockdown of CREB resulted in significantly reduced (P < 0.01) miR-132 expression in THP-1 cells after treatment with PGN or Pam and thus provided evidence for the involvement of CREB in the induction of miR-132 (Figure 2D). Transfection of control siRNA for Lamin A/C did not affect miR-132 level after PGN treatment (Figure 2D). CREB phosphorylation is mediated by mitogen- or stress-activated protein kinases (MAPK or SAPK), which are known to be activated by PGN in THP-1 monocytes (36). U0126 and PD98059 inhibit ERK (extracellular signal-regulated kinase) activation (31) and result in inhibition of CREB phosphorylation after PGN treatment in THP-1 monocytes (Figure 2E). Strikingly, these synthetic inhibitors actively blocked transcription of both pri-miR-132 by ~80% and pri-miR-212 (data not shown), despite a higher PGN concentration, whereas the pri-miR-146a level was unaltered (Figure 2F); mature miR-132 and miR-212, but not miR-146a, were also significantly reduced (Figure 2G). Thus, U0126 and PD98059 blocked miR-132/-212 transcription, while miRNA processing activity appeared unaffected as evidenced by the unaltered expression in miR-146a (Figure 2G). This data confirming PGN- or Pam-induced higher CREB phosphorylation in THP-1 monocytes is consistent with its regulatory role in miR-132/-212 expression. Since CREB regulation of miR-132 has already been established (27, 28), our data serve only to substantiate their relationship in our THP-1 cell model system.

miR-132 and miR-212 may account for PGN- and Pam-induced tolerance

Previously, PGN-induced homologous and heterologous tolerance has been shown in vitro and in vivo (8, 9). In this report, to evaluate this tolerance phenomenon in greater depth, the most commonly employed THP-1 cell model was used again. THP-1 monocytes were primed with PGN (500 ng/ml) or Pam (100 ng/ml) for 18 h followed by challenge exposure with various TLR agonists (1000 ng/ml). After 3 h, the TNF-α protein level was assessed by ELISA (Figure 3A). As expected, TNF-α production was reduced by 80–90% in tolerized cells compared to untolerized controls, after challenge with a panel of inflammatory ligands, including LPS Se (S. enterica), PGN, Pam, and LPS Pg (P. gingivalis). Analysis of TNF-α mRNA by qRT-PCR showed profoundly lower expression in tolerized cell (Figure 3B), consistent with the data at protein level. Beside TNF-α, production of IL-1β, IL-6, and IL-8 was also assessed in PGN- or Pam-tolerized conditions (Supplementary Figure S2A–C) from the same supernatant used in Figure 3A. PGN-tolerized THP-1 monocytes showed significant IL-1β reductions, by 20% against LPS Se challenge, 60% against PGN, and 50% against Pam, but not against LPS Pg (Supplementary Figure S2A). A similar pattern of IL-1β reduction was apparent in Pam-tolerized cells with clear effect observed in challenge with PGN or Pam, but not LPS Se or LPS Pg (Supplementary Figure S2A). IL-6 was clearly diminished across all conditions by ~50–70% in PGN- or Pam-tolerized THP-1 cells against the same or different ligands (Supplementary Figure S2B). For IL-8, PGN- or Pam-tolerized THP-1 monocytes showed 20–30% reduction after PGN, Pam, or LPS Pg challenge, whereas no effects were observed against LPS Se challenge (Supplementary Figure S2C). Thus, PGN- or Pam-tolerized cells showed uniform hyporesponsiveness for TNF-α and IL-6, but variable hyporesponsiveness for IL-1β and IL-8, when challenged with different inflammatory ligands. Since the levels of miRNA examined are not different (Figure 3C), the variable hyporesponsiveness for IL-1β and IL-8 has to be explained by other means, such as other unidentified miRNA or additional regulatory factors other than IRAK4.

Figure 3
High levels of miR-132 and miR-212 may account for TLR2 ligand-induced tolerance via downregulation of IRAK4. (A) TNF-α production by THP-1 monocytes primed with or without (unprimed) PGN or Pam (500 ng/ml) for 18 h and then challenged with ligands ...

Having examined the tolerance in PGN- or Pam-primed THP-1 monocytes, we then investigated if there was any association of miR-132/-212 expression with the observed tolerance. miRNA expression in the tolerized THP-1 cells were examined compared to the control (Figure 3C). As expected, miR-132 and miR-212 in the PGN-tolerized sample showed ~40- and ~20-fold higher expression, respectively, over 18 h of initial incubation plus 3 h of challenge compared with untolerized controls; in contrast, miR-146a showed ~10-fold increase while miR-155 had little or no changes in expression over the same time period (Figure 3C). Analysis of TNF-α and miR-132 expression in the PGN-tolerized human PBMCs showed similar results to that obtained in THP-1 monocytes (Figure 3D,E). Similar to PGN-primed PBMCs, which can show cross-tolerance to multiple TLR ligands (3), Pam-primed PBMCs showed a reduction of TNF-α and IL-6 against PGN, Pam, and LPS challenges, and this was negatively correlated with miR-132 expression (Supplementary Figure S2D–F). PGN-primed mouse primary macrophages showed a similar reduction of TNF-α and increase in miR-132 level after challenge with PGN, Pam, or LPS (Figure 3F,G). Murine macrophage RAW264.7 cells showed similar PGN-induced tolerance and cross-tolerance with LPS (Supplementary Figure S2G), similar to previously described (9); in this case, there was ~5-fold increase in miR-132 compared to control unprimed cells (Supplementary Figure S2H). In addition to PGN or Pam, the capacity of flagellin to induce tolerance was also examined. Flagellin-primed THP-1 monocytes after challenge with flagellin, PGN, or Pam showed reduced TNF-α production and increase in miR-132 and miR-212 expression (Figure 4A–C). Furthermore, PGN- or Pam-primed THP-1 monocytes showed cross-tolerance to flagellin challenge (Figure 4E) and this was consistent with the high level of miR-132 expression (Figure 4F).

Figure 4
Flagellin-induced tolerance contributed by high level of miR-132 and miR-212 expression. (A) TNF-α production by THP-1 cells primed with or without flagellin (200 ng/ml) for 18 h and then challenged with flagellin (300 ng/ml), PGN (1000 ng/ml), ...

Next, we investigated changes in adaptor kinase expression, which has been implicated in hyporesponsiveness. IRAK4 is an important candidate based on its indispensable role in the MyD88-dependent TLRs pathway and we identified it to be a putative target of miR-132/-212 based on TargetScan prediction for miRNA targets (37). To find such potential link between IRAK4 and miR-132/-212, IRAK4 expression was analyzed by immunoblot, in which a moderate reduction of ~32–76% was observed in all cases when normalized to tubulin level and compared to unprimed control, no matter priming was with PGN or Pam (Figure 3H). A similar decrease of IRAK4 protein was evident in PGN-treated RAW264.7 cells (up to 64% reduced at 8 h, Supplementary Figure S2I), in flagellin-treated THP-1 cells (up to 41% reduced at 6 h, Figure 4D), and in THP-1 monocytes treated with PGN or Pam for 18 h (up to 90% reduced, Figure 4G). To investigate the probable cause for the variations in IRAK4 protein reduction, IRAK4 expression under increasing priming concentrations of PGN or Pam were examined. The data showed that increasing priming concentrations from 0.01 to 1 μg/ml yielded a greater reduction in IRAK4 protein, as evidenced by immunoblot analysis (Figure 5G,H). Notable tolerance was demonstrated by significant reduction of TNF-α at priming with 0.5 or 1 μg/ml PGN (Figure 5A) or Pam (Figure 5B). With the increase of priming dose, miR-132/-212 expression was also increased as expected and priming with 1 μg/ml PGN or Pam caused a greater than 90% reduction (P < 0.01) of TNF-α, which correlated with the higher miR-132 and miR-212 levels that were not observed at lower PGN/Pam priming concentrations (Figure 5A–F). In our recent report, elevated miR-146a expression depended on continuous exposure to LPS or PGN (20). Similarly, in this report, after 12 h of PGN withdrawal, cells started to regain PGN responsiveness and were almost completely recovered from tolerance after 22 h (Figure 6A). At this point, miR-132 and miR-212 expression was significantly lower (P < 0.01) compared to 18 h continuous PGN priming plus 5 h PGN challenge (Figure 6B) indicating the importance of their presence at high levels to keep cytokine levels under control.

Figure 5
Dose-dependent priming effect of PGN and Pam to induce efficient tolerance are inversely correlated to the levels of miR-132 and miR-212 expression. (A,B) TNF-α levels by THP-1 monocytes primed with 0–1 μg/ml PGN (1'PGN) or Pam ...
Figure 6
Reduction in miR-132 and miR-212 expression in PGN-tolerized THP-1 cells inversely correlated with TNF-α production. (A) TNF-α production by THP-1 cells cultured with (tolerized) or without (untolerized) PGN (100 ng/ml) for 18 h and then ...

Together these data suggest the likely role of miR-132/-212 in PGN- or Pam-induced tolerance, based on the inverse correlation with proinflammatory cytokine production and the repressed levels of IRAK4. Another promising concept that can be introduced is that bacterial components seem to cause priming of miR-132/-212, which renders innate immune cells hyporesponsive to subsequent challenge. The half-lives (t1/2) of miR-132 and miR-212 estimated from this experiment was ~9–10 h and ~10–12 h, respectively, based on the simple difference in fold change between the data points (Figure 6B). The t1/2 of these miRNAs may in part determine the programed control in how long the PGN-induced tolerance takes effect. It is acknowledged that the actual t1/2 is likely shorter than these estimated t1/2 due to continuous miRNA synthesis. In any case, to ensure that physiological changes in the level of IRAK4 was indeed detectable in THP-1 cells when stimulated with PGN, the same RNA samples obtained at different time after PGN stimulation used in Figure 1B were also analyzed for IRAK4 mRNA expression kinetics (Figure 3I). IRAK4 mRNA levels peaked at 2 h and came back down at 4 h and at 8, 12, 24 h below the baseline at time 0. This data support our notion that at 4 h, the increase in miR-132 and miR-212 (Figure 1C) correlated with the decrease in IRAK4 mRNA level and was consistent with the mechanism of miRNA-mediated mRNA degradation (38). As shown in a subsequent section, a detailed analysis was performed to confirm the specificity of miR-132/-212 in regulating the IRAK4 mRNA.

Upregulation of miR-132 and miR-212 alone can mimic PGN/Pam priming to induce tolerance

Earlier in this report, the dramatic induction of miR-132 and miR-212 by PGN or Pam was shown (Figure 1B). To monitor the direct consequence of miR-132 and miR-212 expression in TLR2 ligand-induced tolerance, THP-1 cells were transfected with 40 nM of miR-132- or miR-212-mimic alone or in combination with miR-146a-mimic for 24 h. After 8 h challenge of transfected THP-1 cells (>95% viable) with PGN, TNF-α production decreased 58% by miR-132-mimic, 47% by miR-212-mimic, 50% by miR132- plus miR-212-mimic, 63% by miR-146a-mimic transfection (positive control), 70% by miR-132- plus miR-146a-mimic transfection, and 67% by miR-212- plus miR-146a-mimic transfection compared to mock transfected control (Figure 7A). Transfection with NS control or unrelated miR-375-mimic did affect the level of TNF-α production (Figure 7A). A similar reduction of TNF-α was observed after challenge with Pam in the same transfected THP-1 monocytes (data not shown). It is noted that the combination of miR-132-mimic (20 nM) and miR-212-mimic (20 nM) was not more effective than either miR-132-mimic (40 nM) or miR-212-mimic (40 nM) alone; this is not unexpected as the two miRNAs have the same seed sequence and likely bind to the same IRAK4 3'UTR site (see next section). miR-146a-mimic in combination with either miR-132-mimic (70% reduction) or miR-212-mimic (67% reduction) showed only slightly increased reduction of TNF-α production compared to individual miRNA-mimic alone suggesting this may be a limitation of the experimental system or that the contribution of these miRNAs to PGN-tolerance is only up to 70%. To corroborate such a role of miR-132 or miR-212, a parallel experiment in THP-1 cells was performed by substituting the miRNA-mimics with miRNA inhibitors (Figure 7B). TNF-α production increased by 36–54% in miRNA inhibitors transfected cells compared to the mock transfected control (Figure 7B). TNF-α production was not significantly affected when cells were transfected with unrelated miR-375 inhibitor. Similar results were obtained in a parallel experiment in THP-1 cells transfected with miRNA mimics and inhibitors but challenged with Pam instead of PGN (data not shown). The function of the miR-132 and miR-212 inhibitors was documented in transfection with up to 90% reduction of these miRNAs (P < 0.01) compared with mock transfected THP-1 monocytes (Figure 7C). To address the contribution of miR-132 and miR-212 in PGN-tolerance, their respective miRNA inhibitors were transfected to THP-1 cells for 24 h prior to the tolerance assay. Normalizing the TNF-α production of the untransfected unprimed and PGN-primed cells (1'PGN) to 100% (untolerized) and 0% (tolerized), respectively, the effect of each miRNA inhibitor alone on the increase in TNF-α production ranged from 28% to 37% (Figure 7D). Of note, miR-132 or miR-212 inhibitor plus miR-146a inhibitor showed more pronounced increase of TNF-α (61–64%) than each inhibitor alone suggesting a cooperative effect. Transfection of unrelated miR-375 inhibitor did show significant effect compared to control without miRNA inhibitor. These data support the dominant role of PGN- or Pam-induced miR-132 and/or miR-212 to mediate tolerance.

Figure 7
Overexpression of miR-132 or miR-212 alone can mimic TLR2 ligand priming while specific miRNA inhibitors tamper PGN-induced tolerance. (A) Reduction in TNF-α production by THP-1 cells transfected for 24 h with miRNA mimics or non-specific (NS) ...

IRAK4 is the molecular target of miR-132 and miR-212

In the above experiments, the levels of IRAK4 showed a negative correlation with miR-132/-212. According to the TargetScan algorithm, IRAK4 shares the same single 3'UTR binding site for both miR-132 and miR-212 (Figure 8A). The interaction between IRAK4 and miR-132/-212 was analyzed by luciferase assay using the pMir-3'IRAK4 vector containing the 3'UTR of IRAK4 cloned downstream of a firefly luciferase reporter. HEK293 cells were co-transfected with pMir-3'IRAK4 vector (reporter IRAK4-wt), RL luciferase control vector, and miR-132-mimic, miR-212 mimic, or miR-146a mimic. Luciferase expression was significantly reduced with miR-132 mimic (~50%) and miR-212-mimic (~25%), while no significant modulation was observed with miR-146a-mimic or for the reporter with 4 nt of the seed sequence mutated (IRAK4-mut, Figure 8B). To further verify the direct regulation of IRAK4 by miR-132/-212, the expression of IRAK4 mRNA and protein was assayed in THP-1 cells, which showed 35–40% reduction after miR-132- and miR-212-mimic transfection compared to mock or miR-146a-mimic transfection (Figure 8C,D). The specificity for IRAK4 repression was demonstrated as miR-132-mimic and miR-212-mimic had no apparent effect on IRAK1, TRAF6, and lamin A/C mRNA (Figure 8C). Immunoblot analysis showed no change in NF-kB p65 and ERK 1/2 levels in miR-132-mimic transfected cells compared to NS control transfected cells (Figure 8D, right panel). Conversely, IRAK4 mRNA was moderately increased after blocking of mature miR-132/-212 using respective miRNA inhibitors in THP-1 monocytes (Figure 8E). Figure 8F further shows the functional regulation of IRAK4 3'UTR in THP-1 cells during PGN stimulation as the IRAK-wt reporter showed 40% reduction in luciferase activity compared to IRAK-mut control and only after PGN stimulation. Having confirmed IRAK4 as a molecular target of miR-132 and/or miR-212 using various approaches, the next question was whether silencing of IRAK4 would affect cytokine secretion. siRNA targeting IRAK4 (siIRAK4) showed >50% reduction at the mRNA level (Figure 8G) and 69% decrease in protein by immunoblot analysis (Figure 8H). With the silencing of IRAK4, TNF-α, IL-6, and IL-8 (data not shown) were markedly reduced (40–60%) after PGN, Pam, or LPS challenge (Figure 8I); these reductions in cytokines are similar to those in previous reports with IRAK4 knockdown (23, 39). The overall findings is consistent with IRAK4 being targeted by miR-132 and/or miR-212, a promising mechanism to prevent excessive cytokine production.

Figure 8
IRAK4 mRNA is a molecular target of miR-132 and miR-212 post-transcriptional silencing in TLR signaling. (A) Sequence alignment of miR-212 (top) and miR-132 (bottom) with putative target site in 3'UTR of IRAK4 (IRAK4-wt) and mutant construct (IRAK4-mut). ...

Bacteria-induced miR-132 and miR-212 contribute to resistance to bacterial infection

In the preceding experiments, PGN- or Pam-induced tolerance or cross-tolerance was observed against purified TLR2, TLR4, and TLR5 ligands. However, in nature, hosts are exposed to whole bacteria that usually display more than one type of TLR ligands. Thus, it is important to determine whether miR-132/-212 are induced during bacterial infection. In this study, heat-killed (HK) P. gingivalis and T. forsythia significantly induced miR-132 (P < 0.01) and miR-212 to a similar extent (data not shown), whereas HK T. denticola had much milder effect (Figure 9A). When infected with live T. forsythia, THP-1 cells showed a similar miR-132 expression whereas P. gingivalis and T. denticola showed relatively low levels of induction (Figure 9B left). As expected, in the same infected monocytes, miR-212 showed a similar expression pattern as miR-132 (Figure 9B right). In contrast, miR-146a expression was significantly lower than miR-132 or miR-212 (data not shown), similar to our previous results (40). HK T. forsythia-primed THP-1 cells showed significant reduction (P < 0.01) of TNF-α after challenge with HK bacteria and various ligands (Figure 9C) and a similar reduction was observed in live T. forsythia-infected THP-1 monocytes (data not shown). HK T. forsythia-tolerized THP-1 cells showed a significant increase (P < 0.001) of miR-132 (Figure 9D) compared to untolerized controls. In line with higher miR-132 and miR-212 expression, HK and live P. gingivalis- and T. forsythia-treated THP-1 monocytes showed substantial reduction (50–60%) of IRAK4 protein expression vs control (Figure 9E). These data were consistent with the above findings in PGN- and Pam-tolerized THP-1 monocytes. Of note, none of these bacteria induced miR-132 and miR-212 to a similar extent (especially T. denticola), suggesting their expression might be bacteria- and ligand-specific. In summary, higher expression of miR-132/-212 plays an important role in providing tolerance or cross-tolerance in the in vitro THP-1 cell model against various TLRs ligands, as well as bacterial infection.

Figure 9
Heat-killed bacterial stimulation- or live bacterial infection-induced miR-132 and miR-212 contribute to resistance to recurrent bacterial challenge. (A,B) qRT-PCR analysis of miR-132 and miR-212 expression in THP-1 monocytes stimulated for 2–48 ...

Discussion

Emerging results indicate that TLR activation affects the expression of a few key miRNAs (26, 41, 42). Recently, an increased expression of miR-146a in response to LPS in THP-1 monocytes was described (21), while a relatively smaller amount of miR-132 induction was noted in the same condition, although no detailed expression analysis of miR-132 and miR-212 by other TLR ligands has been documented. Of note, induction of miR-212 by innate immune ligand has not been described previously. In this study, miR-132/-212 was selected for further study because of its unusual high fold change in expression. Note that the PGN-induced miR-132/-212 (this study) and LPS-induced miR-146a responses (20, 21) are not typical of the “fine-tuner” character described for miRNAs in TLR signaling (26). Other studies have characterized miRNAs as “rheostats” that make fine-scale adjustments to protein output (43, 44). The induced expression of miR-132/-212 suggests their role in a more elaborated programmed feedback mechanism.

The rapid and high levels of miR-132/-212 induction by PGN, Pam, or flagellin stimulation is analogous to miR-146a induction by LPS. In this report, miR-132 and miR-212 expression kinetics following several microbial ligands were examined in THP-1 monocytes, human PBMCs and mouse primary macrophages. Unlike LPS, TLR2/TLR5 ligands triggered a sharp increase in miR-132/-212 expression at earlier time-points compared to miR-146a in THP-1 monocytes (Figure 10A, PGN priming). miR-132 remained at significantly high levels over 48 h. Based on these observations, we can conclude that compared to miR-146a, miR-132/-212 have earlier expression kinetics, which may be very important during acute infection. It is noted that although miR-132/-212 showed higher fold changes than miR-146a (~80-fold vs. 15-fold, respectively, at 12 h), copy number of miR-132 (~100 copies/cell) was lower than miR-146a (~500 copies/cell). The fact is that untreated THP-1 cells have a lower basal level of miR-132 than miR-146a. Upon PGN or Pam stimulation, TNF-α production was negatively correlated with miR-132/-212 expression similar to previous observations on kinetics of miR-146a and TNF-α after LPS stimulation (21). Similar to LPS-induced tolerance mediated by miR-146a (21), the higher fold change of miR-132/-212 induced by TLR2/TLR5 ligands might represent the early response to infection, followed by the miR-146a response as reported for TLR4 stimulation (20, 21).

Figure 10
PGN-induced miR-132 and miR-212 in TLR signaling. (A–B) A schematic summary of PGN-mediated induction of miR-132 and miR-212 targeting IRAK4 in tolerized and untolerized THP-1 monocytes. (C) LPS, PGN, and flagellin from bacterial infection bind ...

miR-132/-212 induction by PGN, Pam, or flagellin stimulation is controlled by CREB activation. miR-132/-212 is located on chromosome 17p13 and transcriptionally activated by CREB in neurons. Its rapid upregulation has been observed in KSHV-infected cells through phosphorylation of CREB (31). In this report, LPS induced a lower level of miR-132/-212 compared to PGN/Pam stimulation. Although both LPS and PGN have been reported to interact with CD14 (45, 46), differences in binding affinity to different receptors (TLR4 vs TLR2) and other unknown factors may all contribute to differences in the specificity of the dominant miRNA induction (miR-146a vs. miR-132/-212). PGN-induced phosphorylation of CREB was not due to endotoxin contamination (36); this demonstration of PGN specificity is consistent with the fact that only ultra-pure ligands were used in this study. Moreover, PGN and LPS induce differential activation of MAP kinases, with LPS strongly inducing all three families of kinases (ERK, JNK, and p38), whereas PGN only induces ERK and JNK without affecting p38 (36). Inhibition of activation of these kinases by U0126 and PD98059, reduced both in primary and mature miR-132/-212, while miR-146a was not diminished, indicating their specificity without affecting other general miRNA processing function. CREB-mediated expression of miR-132/-212 induced by PGN or Pam is a novel finding and this opens a new horizon to evaluate their kinetics in depth in innate immunity regarding the mechanism of cross-tolerance. CREB-regulated rapid miR-132 induction may serve as an anti-apoptotic response in macrophages. Thus, CREB activity is important in innate immunity against certain bacteria, such as Salmonella spp., Shigella spp., and Yersiniae spp., which inhibit survival signals and induce apoptosis of macrophages as a mechanism to evade the host immune response (47).

miR-132 and miR-212 play important roles in TLR2 ligand-induced tolerance and cross-tolerance. Although our data and other reports suggest that PGN is a potent trigger for cytokine production, tolerance induced by this gram-positive bacteria cell wall component has not been studied as extensively as LPS tolerance. The ability of PGN to induce heterologous tolerance shown in this study is congruent with previous findings (8, 9). Accordingly, monocytes primed with PGN or Pam showed hyporesponsiveness to TLR2 ligands (PGN, Pam, LPS from P. gingivalis) or TLR5 ligand flagellin. Moreover, flagellin-primed cells showed tolerance to itself, PGN, or Pam with notably higher miR-132/-212 expression. Thus, these new findings support that tolerance or cross-tolerance is linked to miR-132/-212 overexpression. The dominant effect of miR-132 and miR-212 alone, or in combination with miR-146a, in tolerance was verified by transfection experiments with the corresponding miRNA-mimics; all miRNA mimics showed significantly less TNF-α response to either PGN or Pam. On the other hand, knockdown of miR-132 and miR-212 expression using miRNA inhibitors, also known as antagomirs (48), alone or in combination with knockdown of miR-146a in THP-1 cells tampered the tolerance effect with an increased inflammatory response to TLR2 ligand. Taken together, miR-132/-212 plays an important regulatory role in cytokine production, PGN-induced tolerance, and cross-tolerance. The high levels of miR-132/-212 may also contribute to other regulation downstream of IRAK4. miR-132 may also targets p300 as reported in an earlier study in neuronal cells (31). Since p300 is a known regulator of the innate immune response, and specifically binds the TNF promoter (49), miR-132 may also exert its regulation via p300. However, in THP-1 cells used in the present study, expression of p300 is very low and no significant change detected in LPS, PGN, or Pam stimulated cells as minotored by both qRT-PCR and Western blot (Supplementary Figure S3).

Regulation of IRAK4 by miR-132/-212 is the major mechanism for PGN-induced tolerance. miR-132 and miR-212 share an identical seed sequence and thus, would be expected to regulate a similar subset of target genes. However, miR-132 and miR-212 can be employed for similar (50) or distinct functions in different cell types (30, 33). In this report, IRAK4 mRNA was validated as a molecular target for miR-132 and miR-212 with apparently different degree of efficiency. In experiments to evaluate the effect of IRAK4 knockdown on PGN-stimulated tolerance, cytokine response was not eliminated, but significantly reduced, suggesting that other IRAK family members may compensate in part during infection. This mechanism seems biologically relevant since IRAK4-deficient mice were viable against S. typhimurium infection (39). Figure 10A,B outlines the model that demonstrates how PGN-induced miR-132/-212 play an important role in the response to microbes or its components at early stages of infection and limits the overstimulation of proinflammatory cytokines by suppressing IRAK4. Low-dose PGN-primed THP-1 cells (10 ng/ml, panel A) produce TNF-α rapidly and continue to do so for 4 to 6 h. As soon as regulatory miR-132 starts to increase, then TNF-α production decreases. At 18 h post priming, a profound difference between miR-132 expression and TNF-α secretion is established due to the negative effect on IRAK4 by upregulated miR-132, which leads to tolerance (Figure 10A). Unlike the untolerized control (Figure 10B), tolerized cells do not respond to even high dose of PGN challenge (Figure 10A).

Our findings fully support the dominant role of miR-132/-212 in in vitro PGN- and related ligand-induced tolerance. Beside purified ligands, whole heat-killed or live P. gingivalis and T. forsythia stimulation showed significant expression of miR-132/-212 in monocytes. Subsequently, T. forsythia-primed monocytes showed significant reduction of TNF-α after challenge with various ligands due to the reduction of IRAK4 by higher miR-132/-212 expression. It is interesting to speculate that miR-132/-212 induced by bacteria or its components may play a role in immune-inflammatory diseases, such as periodontitis, by affecting IRAK4 or other targets like MMP-9 as shown in a report on the critical role of miR-132/-212 for epithelial-stromal interactions via targeting MMP-9 (50).

The important biological significance in TLR ligands-induced tolerance is that it is a part of innate immune response to TLR ligands (danger signals) and macrophage/monocyte response by producing proinflammatory cytokines, such as TNF-α. The affected cells have done their job by sounding the cytokine “alarm” and TLR ligands-induced tolerance is a mechanism to dampen cytokine production in a programed manner. As the alarm is rung, further stimulation with the same TLR ligand (tolerance) or different TLR ligands (cross-tolerance), even at a high dosage, does not generate a strong cytokine response and thus prevent overproduction of proinflammatory cytokines that are capable to induce tissue damages. In TLR signaling, the current understanding from structural studies is that binding of TLR ligand to receptor activates the formation of myddosome which involves the helical assembly of the MyD88-IRAK4-IRAK2/IRAK1 complex (22). Thus IRAK4 is recruited to MyD88 earlier than IRAK2/IRAK1, both targets of miR-146a (18, 21, 25). As discussed above, the miR-132/-212 response to TLR2 ligand appears earlier acting on IRAK4, while miR-146a affects IRAK2/1 and TRAF6 somewhat later.

The most interesting part is that both miR-146a and miR-132/-212 target key adaptors of the MyD88-dependent TLR signaling pathway – like a one-two punch - to ensure that the blockage of this pathway will lead to tolerance state to inhibit further stimulation and limiting cytokine production. The speculation of a one-two punch is illustrated in Figure 10C where bacterial infection releases LPS, PGN, and flagellin, which first induces TLR signaling via the MyD88-dependent pathway leading to the activation of NF-κB and production of TNF-α and other cytokines. In the THP-1 cell model, there is approximately 2 hr delay in the appearance of miR-132/-212 (this study) and miR-146a (20, 21). These two miRNA sets serve as a one-two punch to the MyD88-dependent pathway by specific translational repression of IRAK4 and IRAK2/1, which are the critical components for the formation of the myddosome complex. This consideration is appropriate as in any given bacterial infection, the host cell is likely presented by different TLR ligands and the cooperative response by these miRNAs on the MyD88-dependent pathway is potentially advantageous to ensure cross-tolerance is achieved to prevent overproduction of proinflammatory cytokines.

MyD88-dependent TLR pathways use IRAK4 and patients deficient in IRAK4 failed to respond to IL-1, IL-18, and six TLRs (TLR1-5 and TLR9) as expected (1). Accordingly, the upregulated miR-132/-212 in PGN-mediated tolerance is likely to affect other pattern recognition receptor activity in innate immunity. As all TLRs, with the exception of TLR3, use IRAK4 and the MyD88-dependent pathway, they all are likely to be regulated by miR-132/-212 in a comparable manner. Detection and activation of immune cells in response to PGN can also occur through alternate pattern recognition receptors including NOD1 and NOD2 (51, 52), CD14, and a family of peptidoglycan recognition proteins (53) and it remains possible that these receptor signaling pathways are involved in the mechanism of PGN-induced tolerance. As a consequence, PGN-induced tolerance associated with the upregulated miR-132/-212 may have a broader role in regulating TNF-α by TLR pathways.

Innate immune response to the invading microorganism in animals may be influenced by miR-132/-212. PGN-tolerant mice were significantly resistant to both gram-positive (S. aureus) and gram-negative (Pseudomonas aeruginosa) bacteria (8). Overexpression of miR-132/-212 associated with bacteria or PGN tolerance is likely to have important consequences in host innate immunity responding to myriad bacterial infections. More extensive studies on the expression kinetics are needed to fully explore the role of these miRNAs, especially in terms of its half-life in PGN-tolerant animals. In vivo investigations, such as the phenotypic analysis of mice with targeted deletion of miR-132/-212, maybe necessary to fully elucidate the role of these miRNAs in innate immunity. Note that miR-132/-212 have been reported as dysregulated in cancer (33, 54, 55) and overexpression of miR-132 has been shown in such inflammatory diseases as rheumatoid arthritis and osteoarthritis (56). Therefore, miR-132/-212 expression may be associated with inflammation and tumorigenesis, and, given its role in innate immunity, might be an important link between inflammation and cancer.

In summary, a series of evidence provides mechanistic insights into the function of miR-132/-212 in TLR2/TLR5 ligand-induced tolerance, which operates as a negative regulatory feedback mechanism to prevent uncontrolled inflammatory reaction potentially comparable to that observed in sepsis. The major milestone of this study has been highlighted using various cell types, including human and murine monocytes/macrophage cell lines, murine primary macrophages, and human PBMCs. In these cells, miR-132 and/or miR-212 were upregulated in response to purified ligands, as well as bacteria. It should be noted that such key components of MyD88-pathway as IRAK2/1 and TRAF6 are regulated by miR-146a and our report shows that IRAK4 is targeted by miR-132/-212 via mRNA degradation and/or translational repression. These miRNAs have now been shown to work in what appear to be a complementary fashion in response to the various ligand stimulations or bacterial infections. Thus, further investigations of the modulation of the levels of miR-132/-212 alone and/or in combination with miR-146a may be very important as these are interesting targets for therapeutic intervention for boosting or limiting TLR activation.

Supplementary Material

Acknowledgements

We thank Dr. Lakshmyya Kesavalu, Departments of Periodontology and Oral Biology, for providing the bacteria strains P. gingivalis, T. denticola, and T. forsythia used in this study.

This work was supported in part by a grant from the Lupus Research Institute, the National Institutes of Health grant AI47859, and the Andrew J. Semesco Foundation, Ocala, FL. MAN was supported by NIAMS Rheumatology training grant T32 AR007603. PRDG was supported by NIDCR training grant T90/R90 DE007200.

Abbreviations

ELISA
enzyme-linked immunosorbent assay
IRAK1
IL-1 receptor-associated kinase 1
IRAK4
IL-1 receptor-associated kinase 4
KSHV
Kaposi's sarcoma-associated herpesvirus
LPS
lipopolysaccharide
miRNA
microRNA
PAMPs
pathogen-associated molecular patterns
PBMCs
peripheral blood mononuclear cells
PGN
peptidoglycan
pre-miR
precursor miRNA
qRT-PCR
quantitative real-time PCR
TLR
toll-like receptor
TNF-α
tumor necrosis factor-alpha
TRAF6
TNF receptor-associated factor 6

References

1. Akira S, Takeda K. Toll-like receptor signalling. Nat Rev Immunol. 2004;4:499–511. [PubMed]
2. Cook DN, Pisetsky DS, Schwartz DA. Toll-like receptors in the pathogenesis of human disease. Nat.Immunol. 2004;5:975–979. [PubMed]
3. de Vos AF, Pater JM, van den Pangaart PS, de Kruif MD, van 't Veer C, van der Poll T. In vivo lipopolysaccharide exposure of human blood leukocytes induces cross-tolerance to multiple TLR ligands. J. Immunol. 2009;183:533–542. [PubMed]
4. del Fresno C, Garcia-Rio F, Gomez-Pina V, Soares-Schanoski A, Fernandez-Ruiz I, Jurado T, Kajiji T, Shu C, Marin E, Gutierrez del Arroyo A, Prados C, Arnalich F, Fuentes-Prior P, Biswas SK, Lopez-Collazo E. Potent phagocytic activity with impaired antigen presentation identifying lipopolysaccharide-tolerant human monocytes: demonstration in isolated monocytes from cystic fibrosis patients. J. Immunol. 2009;182:6494–6507. [PubMed]
5. Foster SL, Hargreaves DC, Medzhitov R. Gene-specific control of inflammation by TLR-induced chromatin modifications. Nature. 2007;447:972–978. [PubMed]
6. Liew FY, Xu D, Brint EK, O'Neill LA. Negative regulation of toll-like receptor-mediated immune responses. Nat Rev Immunol. 2005;5:446–458. [PubMed]
7. Medvedev AE, Kopydlowski KM, Vogel SN. Inhibition of lipopolysaccharide-induced signal transduction in endotoxin-tolerized mouse macrophages: dysregulation of cytokine, chemokine, and toll-like receptor 2 and 4 gene expression. J. Immunol. 2000;164:5564–5574. [PubMed]
8. Murphey ED, Sherwood ER. Pretreatment with the Gram-positive bacterial cell wall molecule peptidoglycan improves bacterial clearance and decreases inflammation and mortality in mice challenged with Pseudomonas aeruginosa. Microbes Infect. 2008;10:1244–1250. [PMC free article] [PubMed]
9. Nakayama K, Okugawa S, Yanagimoto S, Kitazawa T, Tsukada K, Kawada M, Kimura S, Hirai K, Takagaki Y, Ota Y. Involvement of IRAK-M in peptidoglycan-induced tolerance in macrophages. J. Biol. Chem. 2004;279:6629–6634. [PubMed]
10. Carpenter S, O'Neill LA. Recent insights into the structure of Toll-like receptors and post-translational modifications of their associated signalling proteins. Biochem. J. 2009;422:1–10. [PubMed]
11. Bowie AG. TRIM-ing down Tolls. Nat.Immunol. 2008;9:348–350. [PubMed]
12. Jacinto R, Hartung T, McCall C, Li L. Lipopolysaccharide- and lipoteichoic acid-induced tolerance and cross-tolerance: distinct alterations in IL-1 receptor-associated kinase. J. Immunol. 2002;168:6136–6141. [PubMed]
13. Kobayashi K, Hernandez LD, Galan JE, Janeway CA, Jr., Medzhitov R, Flavell RA. IRAK-M is a negative regulator of Toll-like receptor signaling. Cell. 2002;110:191–202. [PubMed]
14. Li L, Cousart S, Hu J, McCall CE. Characterization of interleukin-1 receptor-associated kinase in normal and endotoxin-tolerant cells. J. Biol. Chem. 2000;275:23340–23345. [PubMed]
15. Sheedy FJ, Palsson-McDermott E, Hennessy EJ, Martin C, O'Leary JJ, Ruan Q, Johnson DS, Chen Y, O'Neill LA. Negative regulation of TLR4 via targeting of the proinflammatory tumor suppressor PDCD4 by the microRNA miR-21. Nat.Immunol. 2010;11:141–147. [PubMed]
16. Filipowicz W, Bhattacharyya SN, Sonenberg N. Mechanisms of post-transcriptional regulation by microRNAs: are the answers in sight? Nat.Rev.Genet. 2008;9:102–114. [PubMed]
17. Baltimore D, Boldin MP, O'Connell RM, Rao DS, Taganov KD. MicroRNAs: new regulators of immune cell development and function. Nat.Immunol. 2008;9:839–845. [PubMed]
18. Hou J, Wang P, Lin L, Liu X, Ma F, An H, Wang Z, Cao X. MicroRNA-146a feedback inhibits RIG-I-dependent Type I IFN production in macrophages by targeting TRAF6, IRAK1, and IRAK2. J. Immunol. 2009;183:2150–2158. [PubMed]
19. Ruggiero T, Trabucchi M, De Santa F, Zupo S, Harfe BD, McManus MT, Rosenfeld MG, Briata P, Gherzi R. LPS induces KH-type splicing regulatory protein-dependent processing of microRNA-155 precursors in macrophages. FASEB J. 2009;23:2898–2908. [PubMed]
20. Nahid MA, Satoh M, Chan EKL. Mechanistic role of microRNA-146a in endotoxin-induced differential cross-regulation of TLR signaling. J. Immunol. 2011;186:1723–1734. [PMC free article] [PubMed]
21. Nahid MA, Pauley KM, Satoh M, Chan EKL. miR-146a is critical for endotoxin-induced tolerance: Implication in innate immunity. J. Biol. Chem. 2009;284:34590–34599. [PMC free article] [PubMed]
22. Lin SC, Lo YC, Wu H. Helical assembly in the MyD88-IRAK4-IRAK2 complex in TLR/IL-1R signalling. Nature. 2010;465:885–890. [PMC free article] [PubMed]
23. De Nardo D, Nguyen T, Hamilton JA, Scholz GM. Down-regulation of IRAK-4 is a component of LPS- and CpG DNA-induced tolerance in macrophages. Cell. Signal. 2009;21:246–252. [PubMed]
24. Suzuki N, Suzuki S, Duncan GS, Millar DG, Wada T, Mirtsos C, Takada H, Wakeham A, Itie A, Li S, Penninger JM, Wesche H, Ohashi PS, Mak TW, Yeh WC. Severe impairment of interleukin-1 and Toll-like receptor signalling in mice lacking IRAK-4. Nature. 2002;416:750–756. [PubMed]
25. Taganov KD, Boldin MP, Chang KJ, Baltimore D. NF-kappaB-dependent induction of microRNA miR-146, an inhibitor targeted to signaling proteins of innate immune responses. Proc. Natl. Acad. Sci. U. S. A. 2006;103:12481–12486. [PubMed]
26. O'Neill LA, Sheedy FJ, McCoy CE. MicroRNAs: the fine-tuners of Toll-like receptor signalling. Nat Rev Immunol. 2011;11:163–175. [PubMed]
27. Vo N, Klein ME, Varlamova O, Keller DM, Yamamoto T, Goodman RH, Impey S. A cAMP-response element binding protein-induced microRNA regulates neuronal morphogenesis. Proc. Natl. Acad. Sci. U. S. A. 2005;102:16426–16431. [PubMed]
28. Remenyi J, Hunter CJ, Cole C, Ando H, Impey S, Monk CE, Martin KJ, Barton GJ, Hutvagner G, Arthur JS. Regulation of the miR-212/132 locus by MSK1 and CREB in response to neurotrophins. Biochem. J. 2010;428:281–291. [PubMed]
29. Wayman GA, Davare M, Ando H, Fortin D, Varlamova O, Cheng HY, Marks D, Obrietan K, Soderling TR, Goodman RH, Impey S. An activity-regulated microRNA controls dendritic plasticity by down-regulating p250GAP. Proc. Natl. Acad. Sci. U. S. A. 2008;105:9093–9098. [PubMed]
30. Shaked I, Meerson A, Wolf Y, Avni R, Greenberg D, Gilboa-Geffen A, Soreq H. MicroRNA-132 potentiates cholinergic anti-inflammatory signaling by targeting acetylcholinesterase. Immunity. 2009;31:965–973. [PubMed]
31. Lagos D, Pollara G, Henderson S, Gratrix F, Fabani M, Milne RS, Gotch F, Boshoff C. miR-132 regulates antiviral innate immunity through suppression of the p300 transcriptional co-activator. Nat Cell Biol. 2010;12:513–519. [PubMed]
32. Hollander JA, Im HI, Amelio AL, Kocerha J, Bali P, Lu Q, Willoughby D, Wahlestedt C, Conkright MD, Kenny PJ. Striatal microRNA controls cocaine intake through CREB signalling. Nature. 2010;466:197–202. [PMC free article] [PubMed]
33. Incoronato M, Garofalo M, Urso L, Romano G, Quintavalle C, Zanca C, Iaboni M, Nuovo G, Croce CM, Condorelli G. miR-212 increases tumor necrosis factor-related apoptosis-inducing ligand sensitivity in non-small cell lung cancer by targeting the antiapoptotic protein PED. Cancer Res. 2010;70:3638–3646. [PubMed]
34. Nahid AM, Sugii S. Binding of porcine ficolin-alpha to lipopolysaccharides from Gram-negative bacteria and lipoteichoic acids from Gram-positive bacteria. Dev. Comp. Immunol. 2006;30:335–343. [PubMed]
35. Livak KJ, Schmittgen TD. Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) method. Methods. 2001;25:402–408. [PubMed]
36. Gupta D, Wang Q, Vinson C, Dziarski R. Bacterial peptidoglycan induces CD14-dependent activation of transcription factors CREB/ATF and AP-1. J. Biol. Chem. 1999;274:14012–14020. [PubMed]
37. Friedman RC, Farh KK, Burge CB, Bartel DP. Most mammalian mRNAs are conserved targets of microRNAs. Genome Res. 2009;19:92–105. [PubMed]
38. La Spada AR, Taylor JP. Repeat expansion disease: progress and puzzles in disease pathogenesis. Nature reviews. Genetics. 2010;11:247–258. [PubMed]
39. Brown KL, Falsafi R, Kum W, Hamill P, Gardy JL, Davidson DJ, Turvey S, Finlay BB, Speert DP, Hancock RE. Robust TLR4-induced gene expression patterns are not an accurate indicator of human immunity. Journal of translational medicine. 2010;8:6. [PMC free article] [PubMed]
40. Nahid MA, Rivera M, Lucas A, Chan EKL, Kesavalu L. Polymicrobial Infection with Periodontal Pathogens Specifically enhances miR-146a in ApoE−/− Mice during Experimental Periodontal Disease. Infect. Immun. 2011;79:1597–1605. [PMC free article] [PubMed]
41. Taganov KD, Boldin MP, Baltimore D. MicroRNAs and immunity: tiny players in a big field. Immunity. 2007;26:133–137. [PubMed]
42. Nahid MA, Satoh M, Chan EKL. MicroRNA in TLR signaling and endotoxin tolerance. Cellular & molecular immunology. 2011;8:388–403. [PMC free article] [PubMed]
43. Baek D, Villen J, Shin C, Camargo FD, Gygi SP, Bartel DP. The impact of microRNAs on protein output. Nature. 2008;455:64–71. [PMC free article] [PubMed]
44. Selbach M, Schwanhausser B, Thierfelder N, Fang Z, Khanin R, Rajewsky N. Widespread changes in protein synthesis induced by microRNAs. Nature. 2008;455:58–63. [PubMed]
45. Dziarski R, Tapping RI, Tobias PS. Binding of bacterial peptidoglycan to CD14. J. Biol. Chem. 1998;273:8680–8690. [PubMed]
46. Gupta D, Kirkland TN, Viriyakosol S, Dziarski R. CD14 is a cell-activating receptor for bacterial peptidoglycan. J. Biol. Chem. 1996;271:23310–23316. [PubMed]
47. Hsu LC, Park JM, Zhang K, Luo JL, Maeda S, Kaufman RJ, Eckmann L, Guiney DG, Karin M. The protein kinase PKR is required for macrophage apoptosis after activation of Toll-like receptor 4. Nature. 2004;428:341–345. [PubMed]
48. Krutzfeldt J, Rajewsky N, Braich R, Rajeev KG, Tuschl T, Manoharan M, Stoffel M. Silencing of microRNAs in vivo with `antagomirs'. Nature. 2005;438:685–689. [PubMed]
49. Tsai EY, Falvo JV, Tsytsykova AV, Barczak AK, Reimold AM, Glimcher LH, Fenton MJ, Gordon DC, Dunn IF, Goldfeld AE. A lipopolysaccharide-specific enhancer complex involving Ets, Elk-1, Sp1, and CREB binding protein and p300 is recruited to the tumor necrosis factor alpha promoter in vivo. Mol. Cell. Biol. 2000;20:6084–6094. [PMC free article] [PubMed]
50. Ucar A, Vafaizadeh V, Jarry H, Fiedler J, Klemmt PA, Thum T, Groner B, Chowdhury K. miR-212 and miR-132 are required for epithelial stromal interactions necessary for mouse mammary gland development. Nat. Genet. 2010;42:1101–1108. [PubMed]
51. Dziarski R. Peptidoglycan recognition proteins (PGRPs) Mol. Immunol. 2004;40:877–886. [PubMed]
52. Girardin SE, Boneca IG, Carneiro LA, Antignac A, Jehanno M, Viala J, Tedin K, Taha MK, Labigne A, Zahringer U, Coyle AJ, DiStefano PS, Bertin J, Sansonetti PJ, Philpott DJ. Nod1 detects a unique muropeptide from gram-negative bacterial peptidoglycan. Science. 2003;300:1584–1587. [PubMed]
53. Mullaly SC, Kubes P. The role of TLR2 in vivo following challenge with Staphylococcus aureus and prototypic ligands. J. Immunol. 2006;177:8154–8163. [PubMed]
54. Anand S, Majeti BK, Acevedo LM, Murphy EA, Mukthavaram R, Scheppke L, Huang M, Shields DJ, Lindquist JN, Lapinski PE, King PD, Weis SM, Cheresh DA. MicroRNA-132-mediated loss of p120RasGAP activates the endothelium to facilitate pathological angiogenesis. Nat. Med. 2010;16:909–914. [PMC free article] [PubMed]
55. Wada R, Akiyama Y, Hashimoto Y, Fukamachi H, Yuasa Y. miR-212 is downregulated and suppresses methyl-CpG-binding protein MeCP2 in human gastric cancer. Int. J. Cancer. 2010;127:1106–1114. [PubMed]
56. Murata K, Yoshitomi H, Tanida S, Ishikawa M, Nishitani K, Ito H, Nakamura T. Plasma and synovial fluid microRNAs as potential biomarkers of rheumatoid arthritis and osteoarthritis. Arthritis Res Ther. 2010;12:R86. [PMC free article] [PubMed]