Mincle suppresses Toll‐like receptor 4 activation

Short abstract

C‐type lectin receptor Mincle involvement on the modulation of TLR4 inflammatory responses.

Abstract

Regulation of Toll‐like receptor responses is critical for limiting tissue injury and autoimmunity in both sepsis and sterile inflammation. We found that Mincle, a C‐type lectin receptor, regulates proinflammatory Toll‐like receptor 4 signaling. Specifically, Mincle ligation diminishes Toll‐like receptor 4–mediated inflammation, whereas Mincle deletion or knockdown results in marked hyperresponsiveness to lipopolysaccharide in vitro, as well as overwhelming lipopolysaccharide‐mediated inflammation in vivo. Mechanistically, Mincle deletion does not up‐regulate Toll‐like receptor 4 expression or reduce interleukin 10 production after Toll‐like receptor 4 ligation; however, Mincle deletion decreases production of the p38 mitogen‐activated protein kinase‐dependent inhibitory intermediate suppressor of cytokine signaling 1, A20, and ABIN3 and increases expression of the Toll‐like receptor 4 coreceptor CD14. Blockade of CD14 mitigates the increased sensitivity of Mincle^−/− leukocytes to Toll‐like receptor 4 ligation. Collectively, we describe a major role for Mincle in suppressing Toll‐like receptor 4 responses and implicate its importance in nonmycobacterial models of inflammation.

Abbreviations

BMDC

Bone marrow–derived dendritic cell

BMDM

Bone marrow–derived macrophage

ITAM

immunoreceptor tyrosine‐based activation motif

Mal

MyD88 adaptor‐like

MyD88

myeloid differentiation primary response gene 88

PAMP

pathogen‐associated molecular pattern

SHP

small heterodimer partner

SOCS

suppressor of cytokine signaling

siRNA

small interfering RNA

Syk

spleen tyrosine kinase

TANK

TRAF family member–associated NF‐κB activator

TDB

trehalose‐6,6‐dibehenate

TRIF

Toll/IL‐1R domain‐containing adapter‐inducing IFN‐β

TNFAIP3

TNF‐α–induced protein 3

TRAF

TNFreceptor‐associated factor

WT

wild‐type

Introduction

TLRs are a family of pattern recognition receptors in innate immune cells, which directly link environmental stimuli to inflammatory responses. TLRs bind specific molecular motifs derived from microbes called PAMPs, activate innate immune cells, and trigger cytokine production, leading to the clearance of invading pathogens [1]. In addition to binding PAMPs, TLRs are activated by ligating nonpathogenic “danger molecules,” including by‐products of inflammatory injury and cellular necrosis, which are collectively called damage‐associated molecular patterns [2, 3]. As such, TLR ligation also has a central role in perpetuating sterile inflammation and tissue damage in numerous nonpathogenic contexts.

Although TLR‐mediated inflammation is important in host defense, regulation of TLR responses is critical in limiting tissue damage, septic injury, and autoimmunity. Suppression of TLR signaling occurs via 3 distinct mechanisms: 1) dissociation of TLR‐dependent signaling complexes, 2) degradation of TLR‐associated adaptor proteins, and 3) regulation of transcription of soluble inflammatory mediators. TANK, SHP1, and A20 (also known as TNFAIP3) can each inhibit TRAF6 ubiquitination [4]. As such, mice deficient in any of these proteins exhibit more robust proinflammatory responses to TLR ligands and progress to organ failure in diverse TLR‐dependent disease models [5, 6–7]. ITAM‐coupled receptors, including DAP12 and β2 integrins, cross‐regulate TLR responses through increased IL‐10 expression, as well as via Syk‐mediated activation of the E3 ubiquitin ligase, Cblb, which degrades MyD88 and TRIF [8, 9, 10–11]. As such, β2‐integrin–deficient mice have a decreased ability to phosphorylate p38 MAPK after TLR4 ligation, which leads to enhanced TLR responses [10, 12, 13]. SOCS proteins, including SOCS1 and SOCS3, are also negative regulators of TLR activation by promoting the degradation of Mal or TRAF proteins [14, 15]. Finally, Iκβ can regulate transcription of proinflammatory cytokines and chemokines by binding NF‐κβ in the cytoplasm and preventing its translocation into the nucleus [16].

Mincle is a type II transmembrane receptor and member of the C‐type lectin receptor family of pattern recognition receptors. Mincle is required for the innate immune response to mycobacterial and fungal pathogens and is expressed on innate immune cells, including macrophages, dendritic cells, and neutrophils [17, 18–19]. Mincle contains an extracellular carbohydrate‐recognition domain and associates with ITAM‐containing FcRγ [20]. Upon Mincle ligation, the CARD9 adaptor protein is recruited and induces inflammatory responses via the phosphorylation of Syk [21]. This signaling cascade leads to the production of an array of cytokines, including TNF‐α and IL‐6. In this study, we investigated whether Mincle critically regulates TLR4 signaling.

MATERIALS AND METHODS

Animals and in vivo models

Male C57BL/6 (H‐2K^b) mice were purchased from the Jackson Laboratory (Bar Harbor, ME, USA). Mincle^−/− mice on a C57BL/6 background and were obtained from the MMRRC Repository (San Diego, CA, USA). Animals were bred in house in a single facility and 6–8‐wk‐old, male mice were used in experiments. For measurement of cytokine responses, mice were treated with a single dose of LPS (Escherichia coli O111:B4; 10 mg/kg; Sigma‐Aldrich, St. Louis, MO, USA) or TNF‐α (2 µg/mouse; PeproTech, Rocky Hill, NJ, USA). For survival experiments, mice were treated i.p. with 35 mg/kg of LPS in split doses. The New York University School of Medicine Institutional Animal Care and Use Committee approved all procedures. Core temperature was measured at various time points using a rectal thermometer for rodents (Braintree Scientific, Braintree, MA, USA). Serum cytokine levels were determined using a cytometric bead array according to the manufacturer's protocol (BD Biosciences, Franklin Lakes, NJ, USA). Serum CD14 was measured using Quantikine ELISA (R&D, Minneapolis, MN, USA).

Cellular isolation and activation

Single‐cell suspensions of murine splenocytes were made by manual disruption of whole spleen. BMDCs were generated as described [22]. Briefly, bone marrow aspirates were cultured for 8 d in complete RPMI (RPMI 1640 with 10% FBS, 2 mM l‐glutamine, and 1% penicillin–streptomycin) supplemented with granulocyte macrophage CSF (20 ng/ml; PeproTech). BMDMs were generated as described [23] and grown in DMEM/F‐12 medium (with 10% FBS and 1% penicillin–streptomycin) supplemented with recombinant macrophage CSF (100 U/ml; ProSpec, Rehovot, Israel). Both BMDC and BMDM were harvested on day 7–8 of culture. The Raji cell line was obtained from ATCC (Manassas, VA, USA). In vitro cellular activation was accomplished using LPS‐EB Ultrapure (100 ng/ml to 10 µg/ml), TDB (0.1–10 µg/ml; InvivoGen, Carlsbad, CA, USA), recombinant murine TNF‐α (20 ng/ml; PeproTech), or recombinant IL‐10 (10 ng/ml; R&D Systems). In selected experiments, cells were additionally treated with mAbs directed against CD14 (10 µg/ml, 4C1; BD Biosciences) or Mincle (5 µg/ml, 6G5; InvivoGen).

Flow cytometry and cytokine analysis

For flow cytometry, cellular suspensions were pretreated with FC blocking reagent (BioLegend, San Diego, CA, USA) followed by incubation with fluorescently conjugated mAbs directed against CD45 (30‐FII), CD11c (N418), MHC II (M5/114.14.2), CD4 (GK1.5), F/480 (BM8), Gr1 (RB6‐8C5), CD11b (M1/70), CD14 (Sa14‐2), CD19 (HIB19), and TLR4 (MTS510; all from BioLegend) or Mincle (anti‐mouse:V‐14; Santa Cruz Biotechnology, Dallas, TX, USA; anti‐human: AT163E; Abcam, Cambridge, MA, USA). For intracellular staining, cells were incubated for 4–6 h with Brefeldin A (1:1000) before permeabilization of cells and staining with fluorescently conjugated TNF‐α (6B8; BioLegend). For cytokine analysis in cell culture supernatant, cellular suspensions were incubated for 24 h, unless otherwise specified, and supernatant was analyzed by cytometric bead array (BD Biosciences).

Western blotting

For Western blotting, total protein was isolated from tissue or cell suspension by homogenization in radioimmunoprecipitation assay buffer with complete protease inhibitor cocktail and phosphatase inhibitor cocktail (Roche Molecular Diagnostics, Pleasanton, CA, USA). Protein quantification was determined by the Bradford protein assay, and samples were equally loaded onto 10% polyacrylamide gels (NuPage, Grand Island, NY, USA), electrophoresed at 200 V, electrotransferred to polyvinylidene difluoride membranes, and probed with mAbs to ERK1/2, p42/44 ERK, p‐JNK, TRAF6, Mal, Mincle, Syk, p‐Syk, p‐p38, p38, A20, ABIN3, SOCS1, SHP‐1, Hes‐1, TANK, β‐actin, (all from Cell Signaling Technology, Beverly, MA, USA). Blots were developed by ECL (Thermo Fisher Scientific, Asheville, NC, USA). Quantification was performed by densitometry, with all values normalized to corresponding β‐actin levels.

PCR

For PCR analysis, total RNA was isolated using the RNeasy Mini Kit (Qiagen, Germantown, MD, USA) and cDNA was made using the high‐capacity reverse transcription kit (Applied Biosystems, Grand Island, NY, USA). RT‐PCR was then performed on a Stratagene Mx3005P QPCR System (Agilent Technologies, Santa Clara, CA, USA) using the mouse inflammation cytokines and receptors PCR array (Qiagen).

siRNA knockdown of Mincle

Raji cell lines were electroporated at 275 V and transfected with 2 µg ready‐made psiRNA‐hMincle and psiRNA‐LUCGL3 control DNA plasmids (InvivoGen, San Diego, CA, USA). Cells were subsequently cultured in complete RPMI 1640 supplemented with 200 µg/ml Zeocin for transfection selection. Knockdown of Mincle in these cells line was confirmed by Western blotting and flow cytometry.

Statistics

Data are presented as means ± se. Survival was measured according to the Kaplan‐Meier method. Statistical significance was determined by the Student's t test and the log‐rank test using GraphPad Prism 6 (GraphPad Software, La Jolla, CA, USA). P < 0.05 was considered significant.

RESULTS

Mincle deletion leads to exaggerated TLR4 responses in vitro

To investigate the potential regulatory effects of Mincle on TLR4 responses, we treated WT and Mincle^−/− splenocytes in vitro with the TLR4 ligand LPS. Mincle^−/− splenocyte suspensions exhibited exaggerated cytokine responses to TLR4 ligation compared with WT ( Fig. 1A ). Splenocyte composition was similar in WT and Mincle^−/− mice (Supplemental Fig. 1A). Similarly, using BMDCs (Fig. 1B–E) and BMDMs (Fig. 1F–H), we found that Mincle deletion resulted in increased cytokine responses to TLR4 ligation. Similar results were obtained using a range of LPS doses (Supplemental Fig. 1B).

Figure 1.

Mincle deletion results in exaggerated TLR4 responses. (A) Splenocytes derived from WT or Mincle^−/− mice (n = 3) were stimulated with TLR4 ligand (LPS, 10 µg/ml) and tested for expression of activating cytokines in cell culture supernatant after 24 h. (B) CD11c⁺ BMDCs derived from WT or Mincle^−/− mice were stimulated with PBS or LPS and tested for expression of TNF‐α by intracellular cytokine analysis. (C–E) WT and Mincle^−/− BMDCs were stimulated with PBS or LPS and tested for expression of TNF‐α (C), IL‐6 (D), and MCP‐1 (E) in cell culture supernatant after 24 h. (F–H) WT or Mincle^−/− BMDMs were stimulated with PBS or LPS and tested for expression of TNF‐α (F), IL‐6 (G), and MCP‐1 (H). Experiments were performed >3 times in triplicate. *P < 0.05.

Mincle inhibition or knockdown leads to exaggerated TLR4 responses

To confirm the suppressive effects of Mincle on TLR4 activation, we administered a neutralizing Mincle mAb (6G5) to WT splenocyte cultures coincident with treatment with TLR4 ligand. Similar to our findings associated with Mincle deletion, Mincle blockade resulted in significantly elevated TNF‐α expression in response to LPS ( Fig. 2A ). Similarly, siRNA‐mediated knockdown of Mincle in human Raji cells (Fig. 2B and C) resulted in higher proinflammatory responses to TLR4 ligation, including IL‐6, which was produced at low levels of uncertain physiologic significance, and IL‐8, which is a prominent cytokine released by human B cells (Fig. 2D and E). Collectively, these data suggest that Mincle negatively regulates TLR4 activation.

Figure 2.

Mincle inhibition or knockdown results in exaggerated TLR4 responses. (A) Splenocytes derived from WT or Mincle^−/− mice were stimulated with TLR4 ligand (LPS, 10µg/ml) after pretreatment with a neutralizing Mincle mAb (5 µg/ml) or isotype control and tested for expression of TNF‐α in cell culture supernatant at 24 h. (B and C) Knockdown of Mincle in Raji cells was confirmed by flow cytometry based on mean fluorescence intensity (B) and Western blotting (C). Density analysis based on triplicates is shown. (D and E) Raji cell siRNA Mincle transfectants and controls were tested for expression of IL‐6 (D) and IL‐8 (E) in 24‐h cell culture supernatant after stimulation with LPS. Experiments were repeated 3 times with similar results. *P < 0.05, **P < 0.01.

Mincle deletion results in exaggerated TLR4‐mediated inflammation in vivo

Based on our in vitro findings, we postulated that Mincle may similarly suppress TLR4 responses in vivo. Accordingly, serum levels of proinflammatory cytokines, including IL‐6, IFN‐γ, MCP‐1, and TNF‐α, were substantially higher and more sustained in LPS‐treated Mincle^−/− animals compared with similarly treated WT mice ( Fig. 3A–D ). Macrophages (Fig. 3E), dendritic cells (Fig. 3F), inflammatory monocytes (Fig. 3G), and neutrophils (Fig. 3H) all exhibited higher activation in Mincle^−/− mice as compared with WT mice after in vivo LPS administration. Furthermore, Mincle^−/− mice became profoundly more hypothermic than their WT counterparts in response to LPS treatment (Fig. 3I), and 100% of Mincle^−/− mice died within 96 h of initiating LPS treatment, whereas all WT mice survived (Fig. 3J). Taken together, these data indicate that Mincle deletion leads to exaggerated LPS‐mediated endotoxic responses in vivo.

Figure 3.

Mincle^−/− mice are more susceptible to endotoxic shock. WT and Mincle^−/− mice were treated with LPS i.p. and (A–D) tested for serum levels of proinflammatory cytokines at timed intervals. Splenic macrophages (E), dendritic cells (F), inflammatory monocytes (G), and neutrophils (H) were tested for expression of TNF‐α at 6 h and 24 h by intracellular cytokine analysis. (I) The changes in core body temperature from baseline in LPS‐treated mice were measured using a rectal probe. *P < 0.05, **P < 0.01, ***P < 0.001. (J) Survival was assessed according to the Kaplan‐Meier method (n = 5 mice per group; P < 0.001). In vivo experiments were repeated twice with similar results. H, hours.

Mincle ligation negatively regulates TLR4 activation

Because Mincle deletion, knockdown, or blockade resulted in exaggerated TLR4‐mediated inflammation, we postulated that Mincle ligation may attenuate TLR4 signaling. Splenocytes were treated with low doses of the synthetic Mincle ligand TDB and activating doses of LPS, separately or in combination. Consistent with our hypothesis, we found that coligation of Mincle mitigated LPS‐mediated IL‐6 production; however, LPS‐induced TNF‐α expression was not significantly affected ( Fig. 4A ). Further, analysis by Western blotting revealed diminished expression of p‐JNK and TRAF6 over time after costimulation with both Mincle and TLR4 ligands, as compared with TLR4 ligation alone (Fig. 4B).

Figure 4.

Mincle ligation inhibits TLR4 activation. (A) Splenocytes were stimulated with LPS (100 ng/ml), TDB, or costimulated with LPS + TDB and tested for expression of IL‐6 and TNF‐α in cell culture supernatant after 24 h. (B) Cells were similarly stimulated with LPS and TDB, alone or in combination, for the indicated time intervals and tested for expression of β‐actin, TRAF6, and p‐JNK by Western blotting. Density plots based on triplicates are shown. Experiments were repeated 3 times with similar results. *P < 0.05, ***P < 0.001.

Mincle does not regulate non‐TLR4–mediated sterile inflammation

To test whether the regulatory effects of Mincle are specific to TLR4 responses, we investigated whether Mincle mitigates inflammatory signaling in alternate models of sterile inflammation. However, WT and Mincle^−/− leukocytes exhibited similar cytokine responses in vitro after stimulation with TNF‐α (Supplemental Fig. 2A). Further, in vivo administration of TNF‐α elicited similar patterns of MCP‐1 elevation in both Mincle^−/− and WT animals (Supplemental Fig. 2B).

Mincle deletion results in higher activation of TLR4‐dependent signaling pathways

To understand the pattern and mechanism of enhanced TLR4‐mediated inflammation in the context of Mincle deletion, we investigated changes in the temporal activation of signaling mechanisms downstream of TLR4 ligation. As expected, we found earlier and more sustained activation of canonical MAPK signaling intermediates in LPS‐treated Mincle^−/− splenocytes compared with WT splenocytes, including higher p‐Erk, p‐JNK, and TRAF6 ( Fig. 5A ). We observed similar effects in LPS‐stimulated WT and Mincle^−/− BMDMs (Supplemental Fig. 3). Collectively, these data suggest that Mincle deletion leads to increased activation of proinflammatory signaling pathways after TLR4 ligation.

Figure 5.

Mincle modulates expression of proinflammatory and inhibitory signaling pathways. (A and B) WT or Mincle^−/− splenocytes were stimulated with LPS for timed intervals and tested for expression of p‐ERK, ERK, p‐JNK, and TRAF6 (A) and phosphorylated and nonphosphorylated Syk and p38, SOCS1, A20, ABIN3, Mal, SHP‐1, Hes‐1, and TANK (B) by Western blotting. β‐actin was used as a loading control. (C) Density plots based on triplicates are shown. *P < 0.05.

Mincle deletion results in diminished Syk and p38 activation and reduced expression of inhibitory proteins in response to LPS

We postulated that Mincle may suppress TLR4 responses by inducing the up‐regulation of inhibitory signaling intermediates. Specifically, we investigated whether Mincle‐dependent Syk activation may induce p38 phosphorylation, which can result in up‐regulation of the inhibitory intermediates SOCS1, A20, and ABIN3 [4]. Accordingly, we found markedly reduced phosphorylation of both Syk and p38 in LPS‐stimulated Mincle^−/− splenocytes compared with WT splenocytes (Fig. 5B). Moreover, SOCS1 and A20 were each substantially reduced at baseline and after LPS treatment in Mincle^−/− cells compared with WT cells (Fig. 5B). ABIN3 was also less‐robustly up‐regulated at early time points in LPS‐treated Mincle^−/− splenocytes (Fig. 5B). p38‐mediated expression of SOCS1 can suppress TLR4 signaling by inducing the degradation of Mal protein, which critically regulates inflammatory signaling via TRAF6 [24, 25]. Accordingly, we found more sustained expression of Mal in LPS‐treated Mincle^−/− splenocytes (Fig. 5B). However, there were no appreciable differences between WT and Mincle^−/− leukocytes in expression of SHP‐1, Hes‐1, and TANK (Fig. 5B) [4].

Mincle does not regulate TLR4 responses through IL‐10

Because IL‐10 can down‐regulate inflammatory responses, we studied the potential for Mincle to also mitigate TLR4 responses by promoting TLR4‐mediated IL‐10 production. However, we did not find a reduction in IL‐10 expression at baseline or after LPS stimulation in Mincle^−/− BMDCs ( Fig. 6A ), BMDMs (Fig. 6B), or splenocytes (Fig. 6C) compared with those from WT mice. Conversely, IL‐10 was actually elevated in LPS‐treated Mincle^−/− BMDMs and BMDCs. Similarly, we did not find diminished IL‐10 production in vivo in LPS‐treated Mincle^−/− mice compared with WT animals (Fig. 6D). Moreover, Mincle^−/− BMDCs exhibited a similar reduction in LPS‐mediated IL‐6 production as compared with WT BMDCs after pretreatment with recombinant IL‐10 (Fig. 6E). Taken together, these findings suggest that Mincle does not regulate TLR4 signaling by promoting IL‐10 expression or modulating cellular responsiveness to IL‐10.

Figure 6.

Mincle deletion does not suppress IL‐10 expression or alter its inhibitory function after LPS stimulation. BMDC (A), BMDM (B), and splenocyte (C) concentrates from WT or Mincle^−/− mice were stimulated with LPS (10 µg/ml) and tested for IL‐10 production in culture supernatant after 24 h. (D) WT and Mincle^−/− mice were injected with LPS i.p (10 mg/kg), and serum IL‐10 levels were measured at serial time intervals (n = 3/group). (E) WT and Mincle^−/− BMDCs were pretreated in culture with PBS or recombinant IL‐10 (10 ng/ml) 1 h before overnight stimulation with PBS or LPS (100 ng/ml). Cell culture supernatant was tested for IL‐6. Each experiment was repeated at least twice in triplicates with similar results. ns, not significant; *P < 0.05.

Mincle cross‐regulates TLR4 expression

To investigate whether Mincle regulates TLR4 expression, we tested the effects of Mincle ligation on TLR4 expression levels directly. We found that Mincle ligation of BMDMs using TDB decreased TLR4 expression ( Fig. 7A ). Interestingly, TLR4 ligation with LPS similarly lowered Mincle expression, which suggests reciprocal regulation between the 2 receptors (Fig. 7B). However, Mincle deletion did not up‐regulate TLR4 expression (Fig. 7C). Hence, modulation of receptor expression does not appear to contribute to the inhibitory effects of Mincle on TLR4 signaling in the absence of direct Mincle ligation.

Figure 7.

Mincle and TLR4 reciprocally cross‐regulate each other after receptor ligation. (A) WT BMDM were treated with PBS or TDB and tested for TLR4 expression. Mean fluorescence intensities (MFIs) are shown. (B) Similarly, BMDMs were stimulated with PBS or LPS (10 µg/ml) and tested for Mincle expression. (C) Expression of TLR4 was tested in splenic dendritic cells, macrophages, and neutrophils from WT and Mincle^−/− mice. Representative data and average MFIs from experiments repeated 3 times are shown. ns, not significant; *P < 0.05; **P < 0.01.

Mincle deletion is associated with increased expression of CD14

We further tested whether Mincle also suppresses TLR4 signaling via inhibition of expression of the TLR4 coreceptor CD14. LPS did not induce differential expression of soluble CD14 in vivo (Supplemental Fig. 4). However, we found markedly increased expression of CD14 in Mincle^−/− BMDMs and splenocytes compared with those from WT mice at baseline and after TLR4 ligation by flow cytometry and PCR, respectively ( Fig. 8A and B ). Moreover, blockade of CD14 abrogated the enhanced TLR4‐mediated inflammatory responses in Mincle^−/− splenocytes but had comparatively minor effects in WT splenocytes (Fig. 8C and D). For example, pretreatment with a CD14 inhibitor before LPS stimulation, resulted in a nearly 20‐fold reduction in Mincle^−/− cellular TNF‐α production as compared with only a 3‐fold reduction in WT TNF‐α levels. These data imply that Mincle may regulate TLR4 responses by suppressing CD14 expression.

Figure 8.

Mincle regulates TLR coreceptor expression. BMDMs (A) or splenocyte suspensions (B) derived from WT or Mincle^−/− mice were treated with PBS or LPS (10 µg/ml) and tested for expression of CD14 by flow cytometry (A) (fluorescence intensities are shown) and PCR (B). (C and D) Splenocytes derived from WT and Mincle^−/− mice were treated with PBS or a neutralizing CD14 mAb (10 µg/ml) and LPS, either alone, or in combination. Expression of TNF‐α (C) and IL‐6 (D) were tested in cell culture supernatant. The relative decrease in cytokine production upon CD14 blockade in Mincle^−/− leukocytes compared with WT leukocytes was calculated. Experiments were repeated twice with similar results. ***P < 0.001.

DISCUSSION

We have shown that Mincle suppresses TLR4‐mediated inflammatory responses. We found that various Mincle^−/− leukocytes, including macrophages and dendritic cells. produce higher levels of proinflammatory cytokines, such as TNF‐α and IL‐6, and that Mincle^−/− mice are more susceptible to LPS‐induced inflammation in vivo. Mincle deletion is associated with elevated TRAF6, p‐ERK, and p‐JNK expression in response to LPS but decreased activation of the p38‐dependant inhibitory proteins SOCS1, A20, and ABIN3, as well as increased expression of the TLR4 coreceptor CD14.

Mincle has been widely shown to enhance inflammation in various pathogen‐mediated contexts, particularly in response to mycobacterial and fungal pathogens [18, 26, 27–28]. Mincle is an essential receptor for TDM [29]. There are data showing that Mincle^−/− immune cells produce lower levels of inflammatory cytokines in response to mycobacterial byproducts. For example, Ishikawa et al. [30] showed that Mincle‐deficient macrophages have depleted cytokine responses to TDM. Others have shown that Mincle^−/− mice produce less TNF‐α and IL‐6 in response to Malassezia spp. and bacillus Calmette‐Guérin, respectively [18, 26]. However, our data showing increased inflammatory responses in Mincle^−/− mice do not conflict with the current literature, given the different models and stimulants used. Furthermore, there is some evidence to suggest an anti‐inflammatory role for Mincle. For example, Devi et al. [31] showed that siRNA knockdown of Mincle in human THP‐1 macrophages resulted in increased production of proinflammatory cytokines and concluded that Mincle modulates inflammation to Helicobacter pylori infection. Another study [32] showed that Mincle suppressed antifungal immunity through IL‐12. Our study specifically implicates Mincle as a potential regulator of TLR4‐mediated responses, which has not yet been clearly elucidated, although Mincle is known to be up‐regulated in response to LPS [26, 27], and Mincle inflammation in adipose tissues may be modulated through the TLR4–NF‐κβ pathway [33].

Our findings have significant clinical relevance because regulation of TLR4 responses is critical for physiologic homeostasis. Uncontrolled TLR4‐mediated inflammatory signaling is linked to autoimmune diseases [1, 34]. Similarly, deficiency in regulators of TLR4 signaling in pathogenic contexts can result in overwhelming sepsis. However, knowledge of regulatory mechanisms governing TLR signaling, in general, is limited. Mice deficient in either A20 or the adaptor protein TANK, both of which control TLR‐induced polyubiquitination of TRAF6, have exaggerated inflammatory responses [5, 35, 36]. For example, TANK‐deficient mice develop severe autoimmune glomerulonephritis [5]. ABIN‐3 and A20 also protect against LPS‐induced mortality by inhibiting NF‐κβ activation [37]. Recently, studies have indicated a role for the adaptor protein DAP12 in regulation of TLRs [38]. BMDMs deficient in DAP12 have increased TLR‐mediated inflammatory cytokine production, and DAP12^−/− mice are more susceptible to endotoxic shock [8]. Similarly, mice lacking the CD11b integrin have heightened TLR‐mediated inflammatory responses [11]. The above studies correlate with our findings that mice with genetic deletion of Mincle, an ITAM‐containing receptor similar to DAP12 and CD11b integrin, also have heightened TLR‐medicated responses. Thus, the current work demonstrates that Mincle is a potent addition to this short list of known TLR modulators. Notably, a clinical study [39] found that a polymorphism in the Mincle receptor is protective against rheumatoid arthritis. Based on the current findings, it is plausible that regulation of TLR4 responses is the mechanism of protection against this autoimmune disease. However, definitive determination requires more exact clinical investigation in patients.

We explored multiple mechanisms by which Mincle may regulate TLR4 signaling. We hypothesized that Mincle‐TLR cross‐regulation was Syk dependent, since several studies have shown that Syk activation can suppress TLR responses through degradation of MyD88 and TRIF [40, 41]. Accordingly, we found reduced phosphorylation of Syk in our LPS‐stimulated Mincle^−/− leukocytes. Importantly, we showed that Mincle^−/− leukocytes exhibit decreased activation of p38 MAPK after TLR4 activation, despite higher expression of p‐ERK. There are multiple mechanisms by which p38 MAPK activation can inhibit TLR responses. Specifically, p38 mitigates TLR responses through MSK1 and MSK2, which phosphorylate the transcription factors CREB and ATF1, leading to IL‐10 production [42, 43]. However, we found that Mincle^−/− leukocytes did not exhibit reduced IL‐10 production in response to TLR ligation, implying this is an unlikely mechanism. In fact, Mincle^−/− BMDMs and BMDCs produced higher levels of IL‐10 in response to LPS in vitro. Notably, we did find lower expression of SOCS1, A20, and ABIN3 in Mincle^−/− leukocytes after LPS stimulation. Previous studies have shown that p38‐driven A20 and ABIN3 can inhibit key proinflammatory TLR4 signaling intermediates, including TRAF6 [1, 37]. Finally, p38‐dependant expression of SOCS1 can suppress TLR4 signaling by mediating degradation of Mal, which we found to be expressed at higher levels at later time points in Mincle^−/− leukocytes [3, 44, 45–46]. Mal has a crucial role in MyD88‐dependent signaling specific to TLR4 ligation by regulating NF‐κβ transcription via TRAF6 [24, 25]. Therefore, decreased activation of p38 MAPK and SOCS1 and diminished degradation of Mal may explain, in part, why Mincle deletion or blockade results in higher proinflammatory responses to TLR4.

Another important mechanistic finding relating to our work is that Mincle deletion results in increased levels of the TLR coreceptor CD14 on BMDMs and splenocytes. As such, CD14 blockade decreased LPS‐induced cytokine responses to a markedly greater extent in Mincle^−/− vs. WT leukocytes, although it is possible this is due to higher starting levels of the receptor. Additionally, we did not find any differences in the serum levels of soluble CD14 between Mincle^−/− and WT mice after LPS stimulation, suggesting that our findings are not related to reduced shedding of the receptor. These data suggest that modulation of CD14 expression is an important mechanism by which Mincle deletion increases TLR4 activation. This is significant because CD14 is required for pathogen‐induced endocytosis of TLR4, which is mediated through Syk [47]. Additionally, as a coreceptor for TLR4, CD14 augments sterile and pathogenic inflammatory responses. For example, unregulated expression of CD14 has been linked to acute liver inflammation and inflammatory bowel disease in humans [48, 49, 50–51]. Furthermore, blockade of CD14 ameliorates endotoxemia [52, 53]. Therefore, our findings suggest that Mincle may also alter TLR4 responses through CD14.

Besides modulating coreceptor expression, there are additional possible mechanisms by which Mincle may regulate TLR4 signaling. For example, IL‐10 is released by Th2‐polarized T cells and by myeloid cells. TLR4 ligation can be a potent stimulus for IL‐10 release from macrophages and dendritic cells. Nevertheless, IL‐10 can down‐regulate inflammatory responses to TLR4 ligation by suppressing NF‐κB signaling [54]. However, we found that IL‐10 expression was not suppressed in Mincle^−/− APC or splenocyte concentrates and, in fact, was elevated, and recombinant IL‐10 did not differentially suppress LPS‐mediated IL‐6 production in Mincle^−/− cells compared with WT cells, suggesting that IL‐10 is not a likely mechanism for Mincle regulation of TLR4 responses.

Another plausible mechanism of Mincle suppression of TLR4 is regulation of receptor expression. Interestingly, we found that Mincle ligation decreases TLR4 expression. Further, TLR4 ligation with LPS similarly lowered Mincle expression. However, Mincle deletion alone—in absence of receptor ligation—did not alter TLR4 expression. Taken together, these data suggest reciprocal regulation between the respective 2 receptors is an unlikely mechanism for the enhanced responses to LPS in the context of Mincle deletion.

Collectively, our study implicates a novel role for Mincle in the governing of physiologic homeostasis through suppression of TLR4 responses. This represents an important addition to a growing list of known regulators of TLR4 activation and provides insight into potential strategies for experimental therapeutics in sterile inflammatory disease and sepsis.

AUTHORSHIP

S.H.G. designed the research studies, conducted the experiments, acquired and analyzed the data, and contributed to writing manuscript. S.K.M. and A.K.V. performed experiments and analyzed data. A.O. performed experiments. J.B., M.D., R.B., L.S., H.L.P., D.D., A.T.H., M.H., and V.M., contributed to experimental design and planning and performed experiments. G.M. was principal investigator with oversight of experimental design, data analysis and interpretation, and manuscript production.

DISCLOSURES

The authors declare no conflicts of interest.

Supporting information

Supplementary data

Supplementary data