Abstract
Adenosine receptor–mediated regulation of monocyte/macrophage inflammatory responses is critical in the maintenance of tissue homeostasis. In this study, we reveal that adenosine potently modulates the expression of NR4A1, 2, and 3 orphan nuclear receptors in myeloid cells, and this modulation is primarily through the adenosine A2a receptor subtype. We demonstrate that A2a receptor activation of NR4A1-3 receptor synthesis is further enhanced in TLR4-stimulated monocytes. After TLR4 stimulation, NR4A receptor–depleted monocyte/macrophage cells display significantly altered expression of cell-surface markers and produce increased inflammatory cytokine and chemokine secretion rendering the cells an enhanced proinflammatory phenotype. Exposure of TLR4 or TNF-α–stimulated monocytes to adenosine analogs directs changes in the expression of MIP-3α and IL-23p19, with NR4A2 depletion leading to significantly enhanced expression of these factors. Furthermore, we establish that nuclear levels of NF-κB/p65 are increased in TLR/adenosine-stimulated NR4A2-depleted cells. We show that, after TLR/adenosine receptor stimulation, NR4A2 depletion promotes significant binding of NF-κB/p65 to a κB consensus binding motif within the MIP-3α proximal promoter leading to increased protein secretion, confirming a pivotal role for NF-κB activity in controlling cellular responses and gene expression outcomes in response to these mediators. Thus, these data demonstrate that during an inflammatory response, adenosine modulation of NR4A receptor activity acts to limit NF-κB–mediated effects and that loss of NR4A2 expression leads to enhanced NF-κB activity and hyperinflammatory responses in myeloid cells.
Introduction
Adenosine is an endogenous purine nucleoside, which elicits profound anti-inflammatory or proinflammatory responses dependent on the local cellular environment and the presence or absence of inflammatory milieu. The accumulation of adenosine occurs through an array of processes including intracellular and extracellular generation, and release from necrotic cells, endothelial cells, and infiltrating neutrophils at sites of cellular stress/inflammation (1). Once produced, adenosine performs its actions through one or more of its four cell membrane G protein–coupled adenosine receptors A1, A2a, A2b, and A3 (2–4). These receptors bear varied affinities for adenosine, with A1, A2a, and A3 exhibiting high affinity, whereas A2b displays low affinity for the nucleoside (1). Furthermore, expression levels of each receptor can fluctuate because of cell type, cell differentiation point, stage of inflammation, and inflammatory mediators present (5).
Monocyte and monocyte-derived macrophage cells orchestrate inflammatory responses by a variety of means, including, but not limited to, secreting numerous cytokine and chemokine mediators, which recruit additional myeloid and other immune cells such as T cells to the site of inflammation (5, 6). Human monocytes express all four adenosine receptors, and it is appreciated that the dominant receptor in these cells is the A2a subtype, which is potently and selectively induced in response to inflammatory mediators such as LPS (7, 8). Thus, adenosine receptor activation can alter monocyte/macrophage cytokine and chemokine production and ultimately shape inflammatory outcome (5).
Selective activation of the A2a receptor in macrophage cells attenuates LPS-driven proinflammatory mediators in vitro and in vivo including TNF-α (7–9). However, it is also established that A2b receptors may also contribute to adenosine-dependent TNF-α suppression (5). Adenosine has further been shown to alter chemokine receptor expression on macrophage cells, thereby affecting their migration ability (10). Proinflammatory and anti-inflammatory responses elicited by monocyte/macrophage cells are equally important, and it is the appropriate regulation or fine-tuning of these responses that brings about inflammatory resolution (6, 11). Hence the combination of malleable immune cells with dynamic adenosine receptor expression levels demonstrates a high degree of cellular plasticity and highlights the central role adenosine receptor–mediated responses play in controlling tissue homeostasis.
The NR4A1-3 orphan nuclear receptors are a subfamily of early response regulators that have emerged as key regulators of inflammatory processes required in inflammatory disease initiation and progression, controlling the magnitude of the inflammatory response (11–17). The NR4A family member, NR4A1, is pivotal in monocyte cell differentiation, polarization, and T cell homeostasis (15–18). NR4A1-deficient monocytes are polarized toward a more proinflammatory phenotype displaying enhanced TLR signaling with increased NF-κB signaling and increased expression of MHC class II differentiation surface marker (16). The second family member, NR4A2, acts as a critical receptor in controlling a feedback mechanism limiting/modulating NF-κB activity in microglia and astrocytes during chronic inflammatory events (13). Thus, adenosine and NR4A receptors are known to control differentiation and inflammatory processes in cells of myeloid origin. We and others have shown that adenosine receptor stimulation can potently alter expression of NR4A family members in human synoviocyte, endothelial, mast, and monocyte cells (19–21). However, the transcriptional effects controlled by NR4A receptors downstream of adenosine signaling in myeloid cells remain unknown and are the object of this study.
In this article, we demonstrate for the first time, to our knowledge, that in monocytic cells, adenosine potently upregulates NR4A1, 2, and 3 gene and protein expression, primarily through the A2a receptor. NR4A depletion significantly alters LPS/TLR4-driven monocyte differentiation leading to amplified proinflammatory cytokine secretion and concomitant changes in cell-surface marker expression. Furthermore, IFN-γ in conjunction with LPS promotes significant induction of classical (M1) proinflammatory phenotypic surface markers, and NR4A2/3 depletion significantly alters this expression pattern. Addition of 5′-N-ethylcarboxamidoadenosine (NECA), a stable adenosine analog, significantly attenuates IFN-γ + LPS–induced surface marker CD80, whereas reducing CCR7 expression. NR4A2 or NR4A3 depletion does not effect this NECA-mediated reduction in CCR7 or CD80 expression. However, NECA potently augments the expression of mediators including MIP-3α and IL-23p19, in primary and immortalized human monocytic cells primed using LPS or TNF-α. In NR4A2-depleted cells, NECA-dependent modulation of these mediators is further enhanced. We establish that NECA-induced NR4A2 activity can act as a feedback mechanism to regulate NF-κB signaling/activity and subsequent expression of target genes, MIP-3α and IL-23p19. Finally, we show that, after adenosine receptor stimulation, NR4A2 depletion promotes significantly enhanced binding of NF-κB/p65 to a κB binding motif on the MIP-3α promoter leading to increased protein secretion. Taken together, this supports the observation that NR4A2 activity is central to maintaining an appropriate feedback mechanism via NF-κB. Thus, NR4A2 functions as an important regulator of adenosine signaling by modulating NF-κB activity leading to subsequent changes in expression of several genes involved in controlling immune cell homeostasis.
Materials and Methods
Cell culture and treatments
Human monocytic THP-1 and murine macrophage RAW 264.7 cells obtained from American Type Culture Collection (ATCC TIB-202 and TIB-71) were cultured in RPMI 1640–GlutaMax media (Life Technologies) and DMEM 6546 (Sigma-Aldrich), respectively, and supplemented with 10% FBS, 100 U/ml penicillin, and 100 μg/ml streptomycin. Stable knockdown cells were maintained in an additional 5 μg/ml puromycin dihydrochloride (Sigma-Aldrich) (21). THP-1 cells were differentiated using 20 ng/ml PMA (Sigma-Aldrich) for 48 h before manipulation.
Cells primed with TNF-α (R&D Systems) or LPS (Sigma-Aldrich) for 18 h followed by addition of adenosine receptor agonists for 2–4 h. Appropriate controls include TNF-α or LPS treatment for 18 h followed by an additional 2–4 h with or without agonists (total 20–22 h; Tocris). NF-κB pathway inhibition was performed using pretreatment for 1 h with 10 μM BAY 11-7082 (NF-κBi; Merckbi). Adenosine receptor–specific antagonists were used by pretreating for 30 min with specific antagonists (Tocris) toward each adenosine receptor (A1, A2a, A2b, and A3) at a final concentration of 0.5 μM.
Primary cell isolation
Peripheral venous blood was collected from healthy volunteers at St. Vincent’s University Hospital, Dublin, Ireland. Institutional Review Board approval was obtained from the local Ethics Committee at St. Vincent’s University Hospital, and written, informed consent was obtained from all volunteers. Twenty milliliters blood was collected into vacutainer tubes (Becton Dickinson), inverted slowly, and allowed to stand for 15 min at room temperature. Blood was layered slowly onto polymorphprep solution (1:1) and centrifuged at 500 × g at 20°C for 35 min with the brakes off. The mononuclear layer was removed and mixed with equal volumes of 0.45% NaCl and centrifuged at 400 × g at 20°C for 10 min. Supernatant was discarded and pellet was resuspended in 12 ml ice-cold water and mixed by inverting gently for 1 min followed by the addition of 12 ml 1.8% NaCl and subsequent centrifugation at 300 × g at 20°C for 5 min. Cells were resuspended in media, counted, and seeded at a density of 2.5 × 105 cells/ml for total RNA isolation.
Real-time quantitative RT-PCR
RNA was extracted from cells using the column-based E.Z.N.A total RNA extraction kit (Omega Bio-Tek, Norcross, GA) followed by cDNA synthesis as previously reported (22). Real-time quantitative RT-PCR was performed using Sybr Green master mix (Applied Biosystems) and an ABI 7300 thermocycler (Applied Biosystems). Primer pair sequences used were human GAPDH (forward [for]: 5′-CGACAGTCAGCCGCATCTT-3′, reverse [rev]: 5′-CCCCATGGTGTCTGAGCG-3′), NR4A1 (for: 5′-GTTCTCTGGAGGTCATCCGCAAG-3′, rev: 5′-GCAGGGACCTTGAGAAGGCCA-3′), NR4A2 (for: 5′-TATTCCAGGTTCCAGGCGAA-3′, rev: 5′-GCTAATCGAAGGACAAACAG-3′), NR4A3 (for: 5′- CCAAGCCTTAGCCTGCCTGTC-3′, rev: 5′-AGCCTGTCCCTTACTCTGGTGG-3′), MIP-3α (for: 5′-CTGGCTGCTTTGATGTCAGT-3′, rev: 5′-CGTGTGAAGCCCACAATAAA-3′), IL-23 subunit-p19 (for: 5′-GAGCCTTCTCTGCTCCCTGAT-3′, rev: 5′-AGTTGGCTGAGGCCCAGTAG-3′), adenosine A2a receptor (for: 5′-AGTTCCGCCAGACCTTCC-3′, rev: 5′-ACCTGCTCTCCGTCACTG-3′), adenosine A2b receptor (for: 5′- CACTGAGCTGATGGACCACTC-3′, rev: 5′- CAGTGACTTGGCTGCATGG-3′), adenosine A1 receptor (for: 5′-CAAGAAGGTGTCGGCCTCC-3′, rev: 5′-CTTGGCGATCTTCAGCTCCT-3′), adenosine A3 receptor (for: 5′-CCCTACAGACGGATCTTGCTG-3′, rev: 5′- TGTTGGGCATCTTGCCTTC-3′) and mouse GAPDH (for: 5′-TGTGTCCGTCGTGGATCTGA-3′, rev: 5′-CCTGCTTCACCACCTTCTTGA-3′), NR4A1 (for: 5′-ATGCCTCCCCTACCAATCTTC-3′, rev: 5′-CACCAGTTCCTGGAACTTGGA-3′), NR4A2 (for: 5′-TCAGAGCCCACGTCGATT-3′, rev: 5′-TAGTCAGGGTTTGCCTGGAA-3′), NR4A3 (for: 5′-TTAACCCATGTCGCTCTGTG-3′, rev: 5′-TGCAGAGCCTGAACCTTGAT-3′), MIP-3α (for: 5′- TCTGCTCTTCCTTGCTTTGGC-3′, rev: 5′-AGTCGTAGTTGCTTGCTTCTGC-3′). Relative expression/abundance levels of target gene transcripts were determined using qBase plus software (Biogazelle; Ghent University, Ghent, Belgium) with GAPDH as a reference target. Results are expressed as fold over untreated control (F.O.C).
PCR array
For gene expression profiling, an in-house PCR array was performed consisting of 82 inflammatory genes and 4 reference genes (GAPDH, ACTB, HPRT1, and GUSB) using pooled mRNA from 3 separate experiments (pooled n = 3). cDNA synthesis was performed with 500 ng total mRNA using the RT2 First Strand Kit (Qiagen, Crawley, U.K.) according to the manufacturer’s instructions. Real-time quantitative RT-PCR was performed containing 1 μl cDNA (after 1:5 dilution), 9 μl water, and 10 μl RT2 SYBR Green/ROX qPCR master mix (SABiosciences, Qiagen) in a 7300 Real Time PCRSystem (Applied Biosystems). Using a Web-based PCR array data analysis tool (SABiosciences, Qiagen), we normalized the cycle threshold values of the target genes against the four reference genes, and the relative abundance of gene transcripts is expressed as log 2^ΔCT.
Chromatin immunoprecipitation assay
A total of 7.5 × 106 THP-1 cells was seeded in a 100-mm cell culture dish to a volume of 10 ml after specific treatments. Chromatin immunoprecipitation assay (ChIP) assay was performed using the EZ-ChIP-chromatin immunoprecipitation kit (Millipore) as described in the manufacturer’s instructions. In brief, cells were fixed by addition of formaldehyde (1% final concentration) for 10 min to cross-link DNA and proteins. Fixation was suspended by addition of a glycine solution. Cells were pelleted at 1200 rpm, washed, and lysed using an SDS buffer (containing protease inhibitor mixture) before chromatin sonication. Twenty micrograms sheared chromatin was precipitated overnight at 4°C with rotation using 5 μg p65 Ab (sc-372; Santa Cruz) and isotype-matched control IgG (sc-2027; Santa Cruz), and immunocomplexes were collected using protein G agarose. After washing, protein/DNA complexes were eluted and cross-links were reversed followed by DNA purification and analysis using end-point PCR for the NF-κB binding motif located at position -79/-93 to the transcription start site (TSS). Primers used for the detection of the NF-κB site have been previously described: for: 5′-TGAGGAAAAAGCAGGAAGTTTT-3′, rev: 5′-GTACACAGAAGGCGTGTTGC-3′ (23). Samples were separated on a 1% agarose gel to visualize the expected 106-bp product. Relative band quantification was assessed using LI-COR Image Studio Lite Version 3.1.
Stable lentiviral knockdown
Stable knockdown of human NR4A2 and human NR4A3 was achieved with the transduction of short hairpin RNA (shRNA) using a Mission Lentiviral packaging mix as per instructions (Sigma-Aldrich) using THP-1 as previously described (21). For controls, cells were transduced with scrambled shRNA (Sigma-Aldrich). In brief, 150 μl THP-1 cells from a stock concentration of 2.5 × 105 cells/ml was pipetted into a single well of a 96-well plate and incubated overnight to yield 70% confluency. Hexadimethrine bromide was added to a final concentration of 8 μg/ml followed by the addition of 10 μl lentiviral particles and incubation for 48 h at 37°C. Puromycin was added to the cells (5 μg/ml) and media changed every 3–4 d until resistant colonies were identified. Western blotting and mRNA analysis were used to measureNR4A2/3 expression.
Western blotting
Nuclear lysates.
Suspension cells were pelleted by centrifugation at 1200 rpm followed by resuspension in into 250 μl ice-cold hypotonic buffer (10 mM HEPES-NaOH buffer, pH 7.9, containing 0.1 mM EDTA, 0.1 mM EGTA, 10 mM KCl, 1 mM DTT, 0.5 mM PMSF, and protease inhibitor mixture [Sigma-Aldrich, Poole, Dorset, U.K.]). Adherent cells were scraped directly into hypotonic buffer, placed on ice for 15 min, and lysed by the addition of 1% (v/v) Nonidet P-40. Samples were then spun at 12,000 × g for 1 min. Pellets were resuspended in high-salt buffer (20 mM HEPES-NaOH buffer pH 7.9 containing 420 mM NaCl, 1.5 mM MgCl2, 0.2 mM EDTA, 25% [w/v] glycerol, and 0.5 mM PMSF) and agitated vigorously for 30 min at 4°C followed by centrifugation at 15,000 × g for 10 min at 4°C. Supernatant was removed and stored at −20°C as nuclear lysate fraction.
Whole cell lysates.
Suspension cells were pelleted by centrifugation at 1200 rpm followed by resuspension into 100 μl radioimmunoprecipitation assay, and adherent cells were scraped into buffer. Cell suspensions were agitated for 30 min at 4°C followed by centrifugation at 15,000 × g for 15 min. Supernatant was removed and stored at −20°C as whole cell lysate fraction.
Protein expression levels were measured by Western blot analysis using specific Abs for NR4A2 and NR4A3 (R&D Systems); NF-κB1 p105/p50, p65, and Lamin A/C (Cell Signaling); and TATA box-binding protein (Abcam).
Densitometric analysis included for Western blot data were determined using LI-COR Image Studio Lite version 3.1. In brief, intensity of proteins of interest (NR4A1, NR4A2, NR4A3, p65, and p50) was quantified using LI-COR program relative to loading control protein (Tata binding protein [TBP] or β-tubulin). Values are displayed as raw arbitrary values.
Flow cytometry
After specific treatment, cells were pelleted at 1200 rpm and resuspended to a concentration of 1 × 106 cells/ml using Opti-MEM (Life Technologies). Fluorescently conjugated Abs (1:200) were then added to 200 μl cells and incubated on ice for 30 min in the dark, alongside unstained isotype IgG control cells. Fluorescently conjugated Abs included CD86 (BV510 conjugated), CD80 (Alexa Fluor 700 conjugated), MHC class II-HLA-DR (allophycocyanin-H7 conjugated), and CCR7 (PE conjugated) supplied by BD Biosciences and used at concentrations as recommended by manufacturer’s instructions. Poststaining cells were washed using 1 ml PBS containing 0.1% sodium azide followed by centrifugation at 1200 rpm. Supernatant was removed and cell pellet was washed and centrifuged again as described earlier. Supernatant was removed and cells were then resuspended in 500 μl PBS containing 0.1% sodium azide. Unstained cells (containing isotype-matched control IgG) were analyzed using the Cyan ADP Analyzer (Beckman Coulter) and gated to identify specific staining from fluorochromes. Unstained cells were confined within gate A as illustrated in pictograms within Figs. 3 and 4. Fluorochrome-conjugated Ab incubated cells were then analyzed on the same Cyan ADP Analyzer (Beckman Coulter) and percentage cells positive for staining was obtained (positive staining is determined by cells migrating toward gate B, C, or D, indicating increased fluorescence obtained by Ab-specific staining within a given treatment).
NR4A depletion alters TLR4-mediated proinflammatory responses and differentiation in monocytes. (A) Undifferentiated THP-1 cells transduced using shRNA directed against scrambled nontarget control, NR4A2, or NR4A3 were subsequently treated with 1 μg/ml LPS for 4 h. Whole cell lysates were prepared followed by Western blot analysis for NR4A2, NR4A3, and loading control β-tubulin. (B) Undifferentiated shRNA transduced THP-1 cells were exposed to 1 μg/ml LPS followed by removal of media at 0, 3, 6, 8, and 24 h and subsequent ELISA analysis for secreted TNF-α. (C) Undifferentiated shRNA transduced THP-1 cells were exposed to 100 ng/ml LPS followed by removal of media at 24 h and subsequent ELISA analysis for secreted MCP-1. (D) Undifferentiated THP-1 cells were pretreated with 10 μM NF-κB inhibitor (NF-κBi) followed by the addition of 1 μg/ml LPS for 2 h. RNA was isolated and RT-PCR analysis was performed for TNF-α, MCP-1, and control gene GAPDH. (E) Undifferentiated shRNA transduced THP-1 cells were exposed to 100 ng/ml LPS for 24 h. After 24 h, THP-1 cells were stained using fluorescently conjugated Abs toward CCR7, MHC class II (HLA-DR), CD86, and CD80, and assessed by flow cytometry as described in 2Materials and Methods. Representative pictograms showing flow-cytometry staining are included alongside flow data. Gate A = negative staining, gates B+C = CCR7/CD86+ staining, and gates C+D = MHC class II (HLA-DR)/CD80+ staining. Densitometric analysis included for Western blot data were determined using LI-COR Image Studio Lite version 3.1 and displayed in arbitrary values above relevant treatments (n = 3). Data are expressed as F.O.C/% cells positive for fluorescently conjugated Abs ± SEM for n = minimum of 3 individual experiments. *p < 0.05, **p < 0.01, ***p < 0.001, treatments compared with untreated control (Un); #p < 0.05, ##p < 0.01, ###p < 0.001, treatments compared displayed here using a bar attachment.
NR4A depletion alters TLR4-mediated proinflammatory responses and differentiation in monocytes. (A) Undifferentiated THP-1 cells transduced using shRNA directed against scrambled nontarget control, NR4A2, or NR4A3 were subsequently treated with 1 μg/ml LPS for 4 h. Whole cell lysates were prepared followed by Western blot analysis for NR4A2, NR4A3, and loading control β-tubulin. (B) Undifferentiated shRNA transduced THP-1 cells were exposed to 1 μg/ml LPS followed by removal of media at 0, 3, 6, 8, and 24 h and subsequent ELISA analysis for secreted TNF-α. (C) Undifferentiated shRNA transduced THP-1 cells were exposed to 100 ng/ml LPS followed by removal of media at 24 h and subsequent ELISA analysis for secreted MCP-1. (D) Undifferentiated THP-1 cells were pretreated with 10 μM NF-κB inhibitor (NF-κBi) followed by the addition of 1 μg/ml LPS for 2 h. RNA was isolated and RT-PCR analysis was performed for TNF-α, MCP-1, and control gene GAPDH. (E) Undifferentiated shRNA transduced THP-1 cells were exposed to 100 ng/ml LPS for 24 h. After 24 h, THP-1 cells were stained using fluorescently conjugated Abs toward CCR7, MHC class II (HLA-DR), CD86, and CD80, and assessed by flow cytometry as described in 2Materials and Methods. Representative pictograms showing flow-cytometry staining are included alongside flow data. Gate A = negative staining, gates B+C = CCR7/CD86+ staining, and gates C+D = MHC class II (HLA-DR)/CD80+ staining. Densitometric analysis included for Western blot data were determined using LI-COR Image Studio Lite version 3.1 and displayed in arbitrary values above relevant treatments (n = 3). Data are expressed as F.O.C/% cells positive for fluorescently conjugated Abs ± SEM for n = minimum of 3 individual experiments. *p < 0.05, **p < 0.01, ***p < 0.001, treatments compared with untreated control (Un); #p < 0.05, ##p < 0.01, ###p < 0.001, treatments compared displayed here using a bar attachment.
NR4As are pivotal for appropriate M1 phenotype differentiation. Undifferentiated scrambled control, NR4A2-, and NR4A3-depleted THP-1 cells were exposed to 100 ng/ml LPS + 10 ng/ml IFN-γ (M1) for 24 h. After 24 h, THP-1 cells were stained using fluorescently conjugated Abs toward CCR7, MHC class II (HLA-DR), CD86, and CD80, and assessed by flow cytometry as described in 2Materials and Methods. Gate A = negative staining, Gates B+C = CCR7/CD86+ staining, and gates C+D = MHC class II (HLA-DR)/CD80+ staining. Representative pictograms showing flow-cytometry staining are included. Data are expressed as % cells positive for fluorescently conjugated Abs ± SEM for n = minimum of 3 individual experiments. *p < 0.05, **p < 0.01, ***p < 0.001, treatments compared with untreated control (Un); #p < 0.05 ##p < 0.01, ###p < 0.001, treatments compared displayed here using a bar attachment.
NR4As are pivotal for appropriate M1 phenotype differentiation. Undifferentiated scrambled control, NR4A2-, and NR4A3-depleted THP-1 cells were exposed to 100 ng/ml LPS + 10 ng/ml IFN-γ (M1) for 24 h. After 24 h, THP-1 cells were stained using fluorescently conjugated Abs toward CCR7, MHC class II (HLA-DR), CD86, and CD80, and assessed by flow cytometry as described in 2Materials and Methods. Gate A = negative staining, Gates B+C = CCR7/CD86+ staining, and gates C+D = MHC class II (HLA-DR)/CD80+ staining. Representative pictograms showing flow-cytometry staining are included. Data are expressed as % cells positive for fluorescently conjugated Abs ± SEM for n = minimum of 3 individual experiments. *p < 0.05, **p < 0.01, ***p < 0.001, treatments compared with untreated control (Un); #p < 0.05 ##p < 0.01, ###p < 0.001, treatments compared displayed here using a bar attachment.
ELISA
Media from THP-1–treated cells were centrifuged briefly to remove cell debris, and supernatant was used for ELISA analysis. MIP-3α protein was measured using a human ELISA kit from RayBiotech as per manufacturer’s instructions.
Promoter studies and analysis
The Genomatix software suite program MatInspector (Genomatix Software GmbH) was used to analyze the human MIP-3α promoter 1000 bp upstream of the TSS. Two stringency tests were applied as per Genomatix software instructions to select binding sites that displayed the best probability as “real” binding motifs for transcription factors (core similarity of 1.0 and matrix similarity > 0.80). The first test is the core similarity test, and the maximum core similarity of 1.0 is only reached when the highest conserved bases of a matrix match exactly in the given sequence. The second test is the matrix similarity test and the score must be >0.80.
Statistical analysis
All data are presented as mean ± SEM and representative Western blots for a minimum of three individual n numbers accompanied by densitometric analysis. Statistical significance was performed using one-way ANOVA followed by Turkey’s post hoc test unless otherwise stated. Where appropriate, a two-tailed Student t test was applied and is highlighted within individual figure legends. An asterisk (*) is used to display significance over untreated (Un) control. A number sign (#) displays significance between given treatments (also indicated by a bar attachment).
Results
Adenosine modulates NR4A1-3 gene expression in human and murine monocyte cells
Adenosine and its stable analog, NECA, have been shown to modulate expression of the NR4A family member, NR4A2, to elicit anti-inflammatory effects downstream of methotrexate treatment in primary synovial cells isolated from rheumatoid arthritis synovial tissue (19). In mast cells, NECA has been shown to potently modulate NR4A1-3 expression (20). However, pathways leading to transcriptional changes downstream of adenosine/NECA signaling and the contribution of NR4A genes to these transcriptional events remain unknown.
Thus, we set out to investigate the effects of adenosine on NR4A1-3 gene expression and activity in myeloid cells, and to establish whether these orphan nuclear receptors can function as transcriptional regulators of inflammatory homeostasis controlled by adenosine. NECA, specific adenosine receptor agonists, and antagonists were used to characterize the effect of adenosine receptor modulation on NR4A1-3 receptor expression in monocytic cell types used in numerous similar studies including primary human PBMCs, THP-1, and murine raw macrophage 264.7 cells (24–28). PMA was further used to differentiate THP-1 monocytes to an alternative macrophage phenotype (PMA-differentiated THP-1 cells) (29).
In PMA-differentiated and -undifferentiated THP-1 cells, treatment with NECA (1–20 μM) results in a dose-dependent induction of NR4A1, 2 and 3 mRNA at 2 h (Fig. 1A, 1B). NR4A induction by NECA (Ki values on individual adenosine receptors A1 [14 nM], A2a [20 nM], A2b [140 nM], and A3 [25 nM]) (30) is maintained at 100 and 10 nM concentrations (Supplemental Fig. 1A). NR4A family members are immediate early genes, and consistently NECA dose-dependent increases in NR4A1-3 mRNA levels were measured as early as 1–2 h (Fig. 1) and returned to basal levels by 6–8 h (data not shown). NECA produces a significant increase in NR4A2 nuclear protein accumulation in undifferentiated THP-1 cells as early as 30 min, which increases up to 4 h (Fig. 1B, insert). A similar trend in NR4A2 protein nuclear accumulation was observed in PMA-differentiated THP-1 cells in response to NECA, increasing up to 3–4 h and returning to basal by 8 h (data not shown). Significant increases were also observed in primary human PBMCs and murine raw mac 264.7 cells after exposure to NECA (Fig. 1C, 1D). Thus, NECA, a pan adenosine receptor agonist, significantly increases NR4A gene expression at both mRNA and protein levels in human and murine cells of myeloid origin.
Adenosine-altered NR4A genes in myeloid cells are concentration and time dependent. (A) THP-1 cells differentiated using 20 ng/ml PMA, and (B) undifferentiated THP-1 cells were exposed to varying concentrations of NECA (a pan adenosine receptor agonist) for 2 h. (B, Right panel) Undifferentiated THP-1 cells were exposed to 20 μM NECA for 30 min and 4 h followed by preparation of nuclear lysates, and subsequent Western blot analysis was performed for both NR4A2 and loading control TBP. (C) Human primary PBMCs were exposed to 20 μM NECA for 1, 2, and 3 h, and (B) murine raw macrophage cells 264.7 were exposed to 1 μM NECA for 2 h. (A–D) RNA was isolated and RT-PCR was performed to assess levels of NR4A1, 2, 3, and control gene GAPDH. Densitometric analysis included for Western blot data were determined using LI-COR Image Studio Lite version 3.1 and displayed in arbitrary values above relevant treatments (n = 3). Data are expressed as F.O.C ± SEM or representative Western blots for n = minimum of 3 individual experiments. *p < 0.05, **p < 0.01, ***p < 0.001 compared with untreated control (Un).
Adenosine-altered NR4A genes in myeloid cells are concentration and time dependent. (A) THP-1 cells differentiated using 20 ng/ml PMA, and (B) undifferentiated THP-1 cells were exposed to varying concentrations of NECA (a pan adenosine receptor agonist) for 2 h. (B, Right panel) Undifferentiated THP-1 cells were exposed to 20 μM NECA for 30 min and 4 h followed by preparation of nuclear lysates, and subsequent Western blot analysis was performed for both NR4A2 and loading control TBP. (C) Human primary PBMCs were exposed to 20 μM NECA for 1, 2, and 3 h, and (B) murine raw macrophage cells 264.7 were exposed to 1 μM NECA for 2 h. (A–D) RNA was isolated and RT-PCR was performed to assess levels of NR4A1, 2, 3, and control gene GAPDH. Densitometric analysis included for Western blot data were determined using LI-COR Image Studio Lite version 3.1 and displayed in arbitrary values above relevant treatments (n = 3). Data are expressed as F.O.C ± SEM or representative Western blots for n = minimum of 3 individual experiments. *p < 0.05, **p < 0.01, ***p < 0.001 compared with untreated control (Un).
Adenosine modulates NR4A genes primarily through the A2a receptor
To determine receptor(s) involved in mediating NECA-induced NR4A gene expression, we examined the contribution of each receptor using specific receptor agonists and antagonists. Adenosine receptor agonists were used at concentration ranges (1 nM to 1 μM) to maximize specific receptor affinity as previously determined by individual Ki values (30, 31).
Using PMA-differentiated THP-1 cells, we observe a significant increase in NR4A1-3 mRNA levels in response to the A2a-specific agonist CGS-21680 (Ki values on individual adenosine receptors A2a [27 nM], A1 [289 nM], A2b [>10,000 nM], and A3 [67 nM]) at concentrations ranging from 1 nM to 100 nM (Fig. 2A). The specific A2b agonist BAY-60-6583 at 100 nM (Ki values on individual adenosine receptors A2b [3–10 nM] and >10,000 nM for A1, A2a, and A3 receptors) displayed modest increases in NR4A1-3, whereas at 10 and 1 nM concentrations, no significant alterations in NR4A mRNA levels was measured (Fig. 2A). Specific agonists for A1 receptors 2-chloro-cyclopentyladenosine (Ki values on individual receptors A1 [0.83 nM], A2a [2270 nM], A2b [18,800 nM], and A3 [38 nM]) and A3 receptors 2-CI-IB-MECA (Ki values on individual receptors A3 [1.4 nM], A1 [220 nM], A2a [5360 nM], and A2b [>10,000 nM], respectively) displayed no significant increases in NR4A1, 2, or 3 mRNA at all concentrations examined (100, 10, or 1 nM; data not shown).
Adenosine alters NR4A genes in myeloid cells primarily through A2a receptor. (A) THP-1 cells differentiated using 20 ng/ml PMA were exposed to 100, 10, and 1 nM CGS-21680 (A2a adenosine receptor–specific agonist) and BAY-60-6583 (A2b adenosine receptor–specific agonist) for 2 h. (B) Human primary PBMCs were exposed to 1 μM NECA for 2 h, 1 μM CGS-21680, and 1 μM BAY-60-6583 for 2 h. (C) Undifferentiated THP-1 cells were primed by exposure to 1 μg/ml LPS for 18 h followed by treatment with 1 μM NECA, 1 μM CGS-21680, and 1 μM bay-60-6583 for a further 2 h. (A–C) RNA was isolated and RT-PCR was performed to assess levels of NR4A1, 2, 3, and control gene GAPDH. (D) Undifferentiated THP-1 cells were primed by exposure to 1 μg/ml LPS for 18 h followed by treatment with 1 μM NECA, 1 μM CGS-21680, and 1 μM Bay-60-6583 for 4 h. Nuclear lysates were prepared and Western blot analysis was performed for NR4A1/2 /3 and loading control TBP. Densitometric analysis included for Western blot data were determined using LI-COR Image Studio Lite version 3.1 and displayed in arbitrary values alongside relevant treatments (n = 3). Data are expressed as F.O.C ± SEM or representative Western blots for n = minimum of 3 individual experiments. *p < 0.05, ***p < 0.001, treatments compared with untreated control (Un); #p < 0.05, ###p < 0.001, treatments compared displayed here using a bar attachment.
Adenosine alters NR4A genes in myeloid cells primarily through A2a receptor. (A) THP-1 cells differentiated using 20 ng/ml PMA were exposed to 100, 10, and 1 nM CGS-21680 (A2a adenosine receptor–specific agonist) and BAY-60-6583 (A2b adenosine receptor–specific agonist) for 2 h. (B) Human primary PBMCs were exposed to 1 μM NECA for 2 h, 1 μM CGS-21680, and 1 μM BAY-60-6583 for 2 h. (C) Undifferentiated THP-1 cells were primed by exposure to 1 μg/ml LPS for 18 h followed by treatment with 1 μM NECA, 1 μM CGS-21680, and 1 μM bay-60-6583 for a further 2 h. (A–C) RNA was isolated and RT-PCR was performed to assess levels of NR4A1, 2, 3, and control gene GAPDH. (D) Undifferentiated THP-1 cells were primed by exposure to 1 μg/ml LPS for 18 h followed by treatment with 1 μM NECA, 1 μM CGS-21680, and 1 μM Bay-60-6583 for 4 h. Nuclear lysates were prepared and Western blot analysis was performed for NR4A1/2 /3 and loading control TBP. Densitometric analysis included for Western blot data were determined using LI-COR Image Studio Lite version 3.1 and displayed in arbitrary values alongside relevant treatments (n = 3). Data are expressed as F.O.C ± SEM or representative Western blots for n = minimum of 3 individual experiments. *p < 0.05, ***p < 0.001, treatments compared with untreated control (Un); #p < 0.05, ###p < 0.001, treatments compared displayed here using a bar attachment.
To further investigate which adenosine receptor(s) are responsible for mediating NR4A1-3 modulation in undifferentiated THP-1 cells, we used selective antagonists toward each receptor. THP-1 cells were pretreated for 30 min with 0.5 μM of each antagonist followed by treatment with NECA, and changes in NR4A3 mRNA, the most potent NECA modulated NR4A family member in these cells, was measured. We observed significant increases (5- to 15-fold) in NR4A3 gene expression when A1, A2b, and A3 receptors were antagonized followed by exposure to NECA for 2 h. In contrast, antagonism of the A2a receptor blocked NECA-induced NR4A3 gene expression (Supplemental Fig. 1B). Consistent with PMA-differentiated THP-1 cells, primary PBMCs display significant increases in NR4A3 gene expression upon exposure to specific A2a receptor agonist CGS-21680 (1 μM) and pan adenosine receptor agonist NECA (1 μM); in contrast, no induction was observed in response to the A2b receptor–specific agonist, BAY-60-6583 (1 μM; Fig. 2B). Primary PBMCs treated with A1 and A3 agonists used at 1 μM displayed no induction of NR4A3 gene expression (data not shown).
To monitor the effects of altering the cellular environment on adenosine-dependent NR4A regulation, we primed monocytes using LPS, an established agonist of TLR4. LPS-primed THP-1 cells with the subsequent addition of the A2a-specific agonist CGS-21680 (1 μM) results in a significant induction in NR4A1-3 mRNA and protein expression levels (Fig. 2C, 2D). In LPS-primed cells, the level of NR4A1-3 induction by CGS-21680 is comparable with the magnitude and pattern mediated by NECA (1 μM; Fig. 2C). Treatment with the A2b-specific agonist BAY-60-6583 (1 μM) after LPS priming had no effect on NR4A family members (Fig. 2C, 2D). Induction of NR4A1-3 gene expression was also observed at lower concentrations (100 nM) of the specific A2a agonist, whereas no changes in mRNA expression levels were measured using 100 nM A2b agonist, BAY-60-6583 (Supplemental Fig. 1C). In LPS-primed THP-1 cells, A1- and A3-specific agonists displayed no alteration in NR4A1-3 mRNA levels at all concentrations tested (1 nM, 10 nM, 100 nM, and 1 μM; data not shown). In LPS-primed cells, NECA and CGS-2160 treatment results in significant, albeit modest, induction of NR4A3 protein levels with minimal effects on NR4A1 levels (Fig. 2D). In contrast, NR4A2 protein levels are significantly and markedly induced, suggesting that NR4A2 may be the major adenosine-regulated member of the NR4A subfamily in these cells (Fig. 2D).
Cell-surface adenosine receptor expression levels significantly impact on a cell’s ability to respond to adenosine receptor ligands (5). We therefore measured adenosine receptor expression during LPS priming and PMA differentiation before adenosine exposure. The A2a receptor is the dominant receptor subtype expressed on these cells during LPS treatment and displays significant mRNA increases within 2 h (7.83 ± 3.07-fold) to 8 h (5.07 ± 0.63-fold; Supplemental Fig. 1D). Adenosine receptor subtypes A1, A2b, and A3 mRNA levels remained unchanged upon LPS exposure (data not shown). The selective modulation of receptor subtype A2a mRNA levels observed in THP-1 cells in response to LPS parallels studies using primary isolated human and murine macrophage cells (7, 8, 32). Significantly, mRNA expression of all four adenosine receptors was increased at 24 h during PMA treatment (Supplemental Fig. 1E), although within these cells, our agonist data suggest that adenosine regulation of NR4A1-3 is primarily A2a receptor driven.
NR4A2 and NR4A3 regulate macrophage inflammatory gene expression and differentiation
To address whether NR4A receptor activity contributes to regulatory macrophage induction by LPS/TLR4 activation, we measured proinflammatory mediator production and surface markers of differentiation in NR4A2- and NR4A3-depleted THP-1 cells. TLR4 stimulation using LPS promotes a significant induction of NR4A2 and NR4A3 protein levels, and this regulation is absent in cells transduced with specific NR4A2 and NR4A3 shRNA (Fig. 3A). TLR4-induced inflammatory macrophage responses resulted in the rapid and potent secretion of TNF-α protein over time (Fig. 3B). LPS-stimulated cells also produce significantly increased levels of MCP-1 protein within 24 h, confirming TLR4-induced inflammatory responses in these cells (Fig. 3C). NR4A2 and NR4A3 depletion significantly potentiate LPS-induced TNF-α and MCP-1 (Fig. 3B, 3C).
To confirm the involvement of the NF-κB pathway in macrophage activation after TLR4 stimulation, we tested NF-κB inhibitor BAY 11-7082 (NF-κBi), and results indicate that pretreatment for 1 h with this inhibitor significantly reduced LPS-induced cytokine and chemokine mRNA changes (Fig. 3D). We further demonstrate that LPS-induced NR4As in macrophage cells are NF-κB dependent, which is consistent with previous studies (Supplemental Fig. 1F) (33). Taken together, these data confirm that, after LPS stimulation, NF-κB regulates levels of NR4A expression, which acts as a feedback mechanism to limit excessive proinflammatory cytokine and chemokine production.
We next determined the phenotypic changes occurring in scrambled and NR4A-depleted THP-1 cells treated with LPS. Using flow-cytometric analysis, we measured a panel of four classical M1 surface markers of differentiation (CCR7, MHC class II [HLA-DR], CD86, and CD80). LPS-treated control cells displayed significant induction of MHC class II (HLA-DR), CD80, and enhanced CCR7 (p = 0.058) and CD86 (p = 0.072) surface marker expression (Fig. 3E). LPS treatment of NR4A2-depleted cells demonstrated significantly enhanced expression of CCR7 and CD80 compared with LPS-treated scrambled cells (Fig. 3E). The expression of surface markers CD86 and MHC class II (HLA-DR) in LPS-treated, NR4A2-depleted cells also displayed higher levels over untreated scrambled control versus LPS-treated scrambled control cells (Fig. 3E). NR4A3 depletion attenuates LPS-induced MHC class II (HLA-DR) expression; in contrast, other surface markers were not significantly altered compared with scrambled control cells (Fig. 3E).
To further drive the classical M1 phenotype and to promote cell-surface marker expression, we treated cells with IFN-γ in combination with LPS (IFN-γ + LPS) (34). In scrambled control cells, IFN-γ + LPS promoted a significant and marked increase in CCR7, MHC class II (HLA-DR), CD86, and CD80 expression (Fig. 4). NR4A2 depletion further significantly enhances CCR7 and CD86 expression; in contrast, NR4A3 depletion did not significantly alter these surface markers and their levels remain comparable with IFN-γ + LPS–treated scrambled control cells (Fig. 4). Intriguingly, IFN-γ + LPS–induced MHC class II (HLA-DR) expression was significantly attenuated in both NR4A2- and NR4A3-depleted cells, whereas CD80 expression remain unchanged under similar conditions (Fig. 4). The addition of NECA together with IFN-γ + LPS–treated cells did not alter MHC class II (HLA-DR) levels or result in changes in CD86 expression, whereas CCR7 surface expression was reduced and CD80 surface expression was significantly attenuated (Fig. 4). Changes observed in CCR7 and CD80 by NECA remained in NR4A2- and NR4A3-depleted cells (Fig. 4).
To further investigate immunomodulatory effects mediated by adenosine, we assessed cytokine and chemokine production by monocytes stimulated in the presence of LPS followed by the addition of NECA. To proceed, we performed analysis using a PCR array, consisting of 82 genes with inflammatory functions, using total RNA isolated from primary human PBMCs that had been primed overnight (18 h) with LPS (1 μg/ml) followed by incubation with NECA for a further 2 h (pooled n = 3). Gene expression analysis revealed that 30 genes in the array were shown to be LPS modulated (>2.0-fold), and 13 of these modulated genes were subsequently altered >1.5-fold (potentiated or repressed) with the addition of NECA (Supplemental Table I).
Fig. 5A shows a “selection” of genes modulated after the addition of NECA to LPS- primed PBMCs. We observe decreased LPS-induced TNF-α mRNA levels as a result of NECA stimulation, which is consistent with previous studies (1). We also show that NECA has the capacity to further enhance LPS-stimulated IL-6 gene expression, consistent with responses in peritoneal macrophage cells (35). Screening reveals that NECA treatment of LPS-primed monocytes results in the modulation of genes involved in immune cell chemotaxis (CCR1 and CCR4) and T cell activation (IL-23p19; Fig. 5A). In this study, we show using primary PBMCs that NECA can diminish significantly endogenous CCR1 and CCR4 mRNA expression (Supplemental Fig. 2A). Further confirmation of gene expression changes measured using the array was verified using isolated primary PBMCs; we verify that NECA significantly attenuates LPS-induced TNF-α expression (Fig. 5B) and potently augments LPS-induced IL-23p19 mRNA levels, compared with NECA stimulation alone (Fig. 5C).
Adenosine alters genes involved in T cell recruitment and activation through A2a receptor. (A) Human primary PBMCs were primed by exposure to 1 μg/ml LPS overnight (18 h) (L) followed by treatment with 20 μM NECA for 2 h (L+N). RNA was isolated from 3 separate n numbers, and a PCR array (consisting of 82 target and 3 reference genes) was performed on n = 3 pooled samples; selected genes from this array are shown including TNF-α, CCR1, CCR4, IL-6, IL-23p19. Genes were made relative to control genes as described in 2Materials and Methods. Bar attachments indicate fold change as a result of NECA addition in each gene. (B–D) Human primary PBMCs (B–D) and THP-1 cells (B) were primed by exposure to 1 μg/ml LPS or 5 ng/ml TNF-α overnight (18 h) followed by treatment with 20 μM NECA for 2 h. (E) Undifferentiated THP-1 cells were exposed to 1 μg/ml LPS for 18 h followed by exposure with 1 μM NECA, 1 μM CGS-21680 (A2a adenosine receptor–specific agonist), and 1 μM BAY-60-6583 (A2b adenosine receptor–specific agonist) for 2 h. (B–E) RNA was isolated and RT-PCR analysis was performed for TNF-α (B), IL-23p19, MIP-3α (C–E), and control gene GAPDH. Data are expressed as F.O.C ± SEM for n = minimum of 3 individual experiments. Appropriate controls were included: NECA alone (2 h) or LPS/TNF-α alone (18 h overnight + 2 h additional time, total 20 h). *p < 0.05, **p < 0.01, ***p < 0.001, treatments compared with untreated control (Un); #p < 0.05, ##p < 0.01, ###p < 0.001, treatments compared displayed here using a bar attachment.
Adenosine alters genes involved in T cell recruitment and activation through A2a receptor. (A) Human primary PBMCs were primed by exposure to 1 μg/ml LPS overnight (18 h) (L) followed by treatment with 20 μM NECA for 2 h (L+N). RNA was isolated from 3 separate n numbers, and a PCR array (consisting of 82 target and 3 reference genes) was performed on n = 3 pooled samples; selected genes from this array are shown including TNF-α, CCR1, CCR4, IL-6, IL-23p19. Genes were made relative to control genes as described in 2Materials and Methods. Bar attachments indicate fold change as a result of NECA addition in each gene. (B–D) Human primary PBMCs (B–D) and THP-1 cells (B) were primed by exposure to 1 μg/ml LPS or 5 ng/ml TNF-α overnight (18 h) followed by treatment with 20 μM NECA for 2 h. (E) Undifferentiated THP-1 cells were exposed to 1 μg/ml LPS for 18 h followed by exposure with 1 μM NECA, 1 μM CGS-21680 (A2a adenosine receptor–specific agonist), and 1 μM BAY-60-6583 (A2b adenosine receptor–specific agonist) for 2 h. (B–E) RNA was isolated and RT-PCR analysis was performed for TNF-α (B), IL-23p19, MIP-3α (C–E), and control gene GAPDH. Data are expressed as F.O.C ± SEM for n = minimum of 3 individual experiments. Appropriate controls were included: NECA alone (2 h) or LPS/TNF-α alone (18 h overnight + 2 h additional time, total 20 h). *p < 0.05, **p < 0.01, ***p < 0.001, treatments compared with untreated control (Un); #p < 0.05, ##p < 0.01, ###p < 0.001, treatments compared displayed here using a bar attachment.
To determine whether adenosine responses can alter gene expression profiles primed with a distinct proinflammatory mediator, we exposed primary PBMCs to TNF-α (5 ng/ml) and subsequently cotreated with NECA for 2 h. In contrast with LPS responses, TNF-α had minimal effect on IL23p19 mRNA levels and the addition of NECA failed to alter IL-23p19 in TNF-α–primed primary PBMCs (Fig. 5C). We next monitored MIP-3α mRNA expression levels, an important mediator of immune cell chemotaxis, and our results indicate that NECA alone induces MIP-3α gene expression, albeit modestly; however, NECA cotreatment dramatically augments MIP-3α in TNF-α–primed PBMCs, whereas displaying minimal effects on robust LPS-dependent responses seen in these cells (Fig. 5D). Taken together, these data suggest that the context and potency of the cellular environment may influence adenosine receptor–mediated regulation of gene expression.
NECA effects on LPS-primed THP-1 cells were next monitored, and results verify a similar pattern of changes in gene expression of TNF-α (Fig. 5B), IL-23p19 (Fig. 5C, 5E), CCR1, and CCR4 (data not shown) as measured in PBMCs after LPS/NECA priming.
Using more selective adenosine receptor agonists, we tested the A2a agonist CGS-21680 (1 μM) and the A2b agonist BAY-60-6583 (1 μM) for their ability to alter expression levels of IL-23p19 and MIP-3α in THP-1 cells primed with LPS. The pattern and magnitude of induction after CGS-21680 treatment is equivalent to NECA, whereas BAY-60-6583 had no effect on gene targets (1 μM; Fig. 5E). NECA or CGS-21680 treatment of THP-1 cells alone showed no significant effects on IL-23p19 and MIP-3α gene expression levels (data not shown). Changes in IL-23p40 expression after adenosine receptor stimulation of control or NR4A-depleted cells primed with LPS were not observed (data not shown).
To further elucidate the downstream mediators involved in NECA-dependent modulation of IL-23p19 and MIP-3α mRNA, we monitored expression levels in THP-1 cells in response to TNF-α (0.1, 1.0, 5.0, and 10 ng/ml; Supplemental Fig. 2B, 2C). Pretreatment of TNF-α–primed cells with NF-κBi (10 μM) before NECA addition indicates that NF-κB activity is involved in mediating these NECA responses, as evident by the significant inhibitory effects after inhibition of this protein complex (NF-κBi) (Supplemental Fig. 2B, 2C).
NR4A depletion disrupts NF-κB/p65 nuclear accumulation leading to altered adenosine regulation of MIP-3α and IL-23p19 expression
In myeloid cells, adenosine A2 receptor stimulation alone or after LPS priming potently upregulates NR4A1-3 gene expression leading to robust changes in NR4A2 protein expression levels (Fig. 2). To ascertain the potential involvement of NR4A2 and NR4A3 in mediating adenosine regulation of gene expression changes, we assessed stably depleted NR4A2 and NR4A3 cells. Consistent with previous studies (1), NECA significantly attenuates LPS-induced TNF-α mRNA (Fig. 5B, Supplemental Fig. 2D). The involvement of NR4A receptors in mediating NECA effects on TNF-α expression was measured. Consistent with changes in TNF-α protein production and secretion (Fig. 3B), NR4A2- and NR4A3-depleted cells display significantly enhanced TNF-α mRNA levels during LPS exposure. However, NECA-dependent reduction of LPS-driven TNF-α is maintained in NR4A2 (1.8-fold) and NR4A3 (2.2-fold)-depleted cells (Supplemental Fig. 2D). The extent of NR4A2 and NR4A3 depletion is evident after NECA ± LPS exposure in scrambled cells compared with stably NR4A-depleted cells (Fig. 6A, Supplemental Fig. 3A, respectively).
NR4A2 is required for appropriate adenosine-mediated NF-κB modulation of T cell regulatory genes MIP-3α and IL-23p19. (A) Undifferentiated THP-1 cells transduced using shRNA directed against scrambled nontarget control (Sc) and NR4A2 (A2) were subsequently treated with 1 μg/ml LPS for 18 h followed by addition of 20 μM NECA for a further 4 h. Whole cell lysates were prepared followed by Western blot analysis for NR4A2 and loading control β-tubulin. (B and C) Undifferentiated Sc and A2 THP-1 cells were exposed to 5 ng/ml TNF-α or 1 μg/ml LPS for 18 h followed by treatment with 20 μM NECA for 2 h and appropriate controls TNF-α/LPS/NECA alone. RNA was isolated and RT-PCR was performed to assess levels of IL-23p19, MIP-3α, and control gene GAPDH. (D) Undifferentiated Sc and A2 THP-1 cells were exposed to 1 μg/ml LPS for 18 h followed by treatment with 20 μM NECA for a further 8 h and appropriate controls LPS/NECA alone; media were collected and analyzed for secreted MIP-3α protein using ELISA. (E) Undifferentiated Sc and A2 THP-1 cells were exposed to 1 μg/ml LPS for 18 h followed by treatment with 20 μM NECA for a further 30 min and 4 h. Nuclear lysates were prepared and Western blot analysis was performed for NR4A2, p65, p50, and loading control TBP. (F) Undifferentiated Sc and A2 THP-1 cells were treated with 1 μg/ml LPS for 18 h followed by treatment with 20 μM NECA for 1 h. ChIP assay was then performed using p65 Ab pull-down, and end-point PCR was run for the κB binding site at position -93/-79 bp upstream of the TSS along with igg control and input. Data are expressed as representative end-point PCR image (n = 2 individual experiments), and relative quantification of bands was calculated from n = 2 using LI-COR Image Studio Lite version 3.1 and displayed in raw arbitrary values above relevant treatments. Controls included were appropriate: NECA alone (2/8 h) or LPS/TNF-α alone (18 h overnight + 2/4/8 h additional time). Densitometric analysis included for Western blot data were determined using LI-COR Image Studio Lite version 3.1 and displayed in arbitrary values above relevant treatments (n = 3). Data are expressed as F.O.C ± SEM or representative Western blots for n = minimum of 3 individual experiments. *p < 0.05, **p < 0.01, ***p < 0.001, treatments compared with untreated control (Un); #p < 0.05, ##p < 0.01, ###p < 0.001, treatments compared displayed here using a bar attachment.
NR4A2 is required for appropriate adenosine-mediated NF-κB modulation of T cell regulatory genes MIP-3α and IL-23p19. (A) Undifferentiated THP-1 cells transduced using shRNA directed against scrambled nontarget control (Sc) and NR4A2 (A2) were subsequently treated with 1 μg/ml LPS for 18 h followed by addition of 20 μM NECA for a further 4 h. Whole cell lysates were prepared followed by Western blot analysis for NR4A2 and loading control β-tubulin. (B and C) Undifferentiated Sc and A2 THP-1 cells were exposed to 5 ng/ml TNF-α or 1 μg/ml LPS for 18 h followed by treatment with 20 μM NECA for 2 h and appropriate controls TNF-α/LPS/NECA alone. RNA was isolated and RT-PCR was performed to assess levels of IL-23p19, MIP-3α, and control gene GAPDH. (D) Undifferentiated Sc and A2 THP-1 cells were exposed to 1 μg/ml LPS for 18 h followed by treatment with 20 μM NECA for a further 8 h and appropriate controls LPS/NECA alone; media were collected and analyzed for secreted MIP-3α protein using ELISA. (E) Undifferentiated Sc and A2 THP-1 cells were exposed to 1 μg/ml LPS for 18 h followed by treatment with 20 μM NECA for a further 30 min and 4 h. Nuclear lysates were prepared and Western blot analysis was performed for NR4A2, p65, p50, and loading control TBP. (F) Undifferentiated Sc and A2 THP-1 cells were treated with 1 μg/ml LPS for 18 h followed by treatment with 20 μM NECA for 1 h. ChIP assay was then performed using p65 Ab pull-down, and end-point PCR was run for the κB binding site at position -93/-79 bp upstream of the TSS along with igg control and input. Data are expressed as representative end-point PCR image (n = 2 individual experiments), and relative quantification of bands was calculated from n = 2 using LI-COR Image Studio Lite version 3.1 and displayed in raw arbitrary values above relevant treatments. Controls included were appropriate: NECA alone (2/8 h) or LPS/TNF-α alone (18 h overnight + 2/4/8 h additional time). Densitometric analysis included for Western blot data were determined using LI-COR Image Studio Lite version 3.1 and displayed in arbitrary values above relevant treatments (n = 3). Data are expressed as F.O.C ± SEM or representative Western blots for n = minimum of 3 individual experiments. *p < 0.05, **p < 0.01, ***p < 0.001, treatments compared with untreated control (Un); #p < 0.05, ##p < 0.01, ###p < 0.001, treatments compared displayed here using a bar attachment.
In THP-1 cells, primed with TNF-α or LPS, NR4A2 depletion significantly potentiates NECA-induced IL-23p19 and MIP-3α mRNA levels compared with scrambled control cells (Fig. 6B, 6C, respectively) leading to increased secretion of MIP-3α (Fig. 6D). NR4A3 depletion does not lead to this effect, and levels of IL-23p19 and MIP-3α mRNA remain relatively unchanged (Supplemental Fig. 3B, 3C). These data suggest that NECA-induced NR4A2 activity may specifically act as a feedback mechanism to limit expression of downstream targets including IL-23p19 and MIP-3α.
To further examine the effects of NR4A depletion on expression levels and nuclear accumulation of NR4A2 and NF-κB family members, we primed THP-1 control (sh-scrambled) and NR4A2-depleted cells with LPS overnight followed by exposure to NECA over 30 min and 4 h, and changes in protein levels were measured (Fig. 6E). In control cells, LPS and NECA cotreatment led to time-dependent increases in nuclear NR4A2 protein levels; this induction was absent in shNR4A2 RNA-transduced cells. With LPS alone and NECA cotreatment of NR4A2-depleted cells, we observed increased p65 protein nuclear accumulation over control cells. LPS treatment also induces significant increases in p50 protein levels; however, no differences in nuclear p50 expression were observed between NR4A2-depleted and control cells, and p50 levels remain unaffected by the addition of NECA (Fig. 6E). Comparative effects on p65 and p50 protein nuclear accumulation were observed in NR4A3-depleted THP-1 cells after LPS treatment (Supplemental Fig. 3D), albeit comparable downstream effects on NF-κB–regulated target gene expression were not measured in NR4A3-depleted cells (Supplemental Fig. 3B, 3C) compared with that of NR4A2-depleted cells (Fig. 6B, 6C).
Furthermore, to determine the effects of NR4A2 depletion on p65 binding to an endogenous κB site, we performed ChIP analysis and primers designed to amplify a specific region within the proximal human MIP-3α promoter (-93/-79). This promoter region contains a κB site previously reported as a binding site for p65 (Supplemental Fig. 3E) (23). In NR4A2-depleted cells, a modest increase in basal p65 binding to the κB site is observed; however, upon LPS stimulation followed by NECA cotreatment, we observe a marked increase in p65 binding to the κB site that is not measured in control cells (Fig. 6F). Importantly, these changes in p65 promoter binding are consistent with the significant production and secretion of the MIP-3a protein levels in NR4A2-depleted cells after LPS/NECA exposure (Fig. 6C), with levels of MIP-3α increased significantly from 482 ± 17.8 to 747 ± 41.7 pg/ml. Importantly, although single LPS treatment can induce robust changes in MIP-3α–secreted protein levels, such significant differences were not measured between control and NR4A2-depleted cells treated with LPS alone compared with LPS/NECA cotreatment. Data discussed throughout have been summarized in Fig. 7.
Summary diagram: NR4A receptors are key regulators of immune homeostasis. Left panel, TLR4 activation using LPS promotes NF-κB activation, nuclear localization followed by induced expression of target genes including NR4A1-3, TNF-α and MCP-1. In addition, TLR4 activation promotes cell-surface marker expression indicative of classical proinflammatory (M1) differentiated macrophage cells, including CCR7, MHC class II (HLA-DR), CD86, and CD80. Depletion of NR4A2 and NR4A3 leads to enhanced production of TNF and MCP-1, demonstrating NR4As are important for limiting overproduction of these mediators (indicated using a horizontal T bar). Depletion analysis further reveals NR4A2 specifically is involved in attenuating overexpression of M1 cell-surface markers after TLR4 stimulation (indicated using a horizontal T bar). Center panel, TLR4 or TNFR stimulation using LPS and TNF-α respectively, adenosine signaling through the A2a receptor (A2aR) induces MIP-3α, IL-23p19 (in a NF-κB–dependent manner), and NR4A1, 2, and 3. NR4A2 depletion reveals it functions as an important regulator of adenosine signaling by modulating NF-κB activity and preventing hyperproduction of MIP-3α and IL-23p19. Right panel, Classical M1 differentiators IFN-γ [binding to the IFNR] + LPS promote induction of surface makers CCR7, MHC class II (HLA-DR), CD86, and CD80. Adenosine receptor (AR) stimulation represses IFN-γ + LPS–induced CD80, whereas NR4A2 or NR4A3 depletion does not modulate this effect. Selective NR4A2 depletion leads to increased production of IFN-γ + LPS–induced CD86 and CCR7, demonstrating NR4A2 is important for limiting hyperproduction of these targets. Furthermore NR4A2 and NR4A3 are positive regulators of IFN-γ + LPS–induced MHC class II (HLA-DR) expression.
Summary diagram: NR4A receptors are key regulators of immune homeostasis. Left panel, TLR4 activation using LPS promotes NF-κB activation, nuclear localization followed by induced expression of target genes including NR4A1-3, TNF-α and MCP-1. In addition, TLR4 activation promotes cell-surface marker expression indicative of classical proinflammatory (M1) differentiated macrophage cells, including CCR7, MHC class II (HLA-DR), CD86, and CD80. Depletion of NR4A2 and NR4A3 leads to enhanced production of TNF and MCP-1, demonstrating NR4As are important for limiting overproduction of these mediators (indicated using a horizontal T bar). Depletion analysis further reveals NR4A2 specifically is involved in attenuating overexpression of M1 cell-surface markers after TLR4 stimulation (indicated using a horizontal T bar). Center panel, TLR4 or TNFR stimulation using LPS and TNF-α respectively, adenosine signaling through the A2a receptor (A2aR) induces MIP-3α, IL-23p19 (in a NF-κB–dependent manner), and NR4A1, 2, and 3. NR4A2 depletion reveals it functions as an important regulator of adenosine signaling by modulating NF-κB activity and preventing hyperproduction of MIP-3α and IL-23p19. Right panel, Classical M1 differentiators IFN-γ [binding to the IFNR] + LPS promote induction of surface makers CCR7, MHC class II (HLA-DR), CD86, and CD80. Adenosine receptor (AR) stimulation represses IFN-γ + LPS–induced CD80, whereas NR4A2 or NR4A3 depletion does not modulate this effect. Selective NR4A2 depletion leads to increased production of IFN-γ + LPS–induced CD86 and CCR7, demonstrating NR4A2 is important for limiting hyperproduction of these targets. Furthermore NR4A2 and NR4A3 are positive regulators of IFN-γ + LPS–induced MHC class II (HLA-DR) expression.
Discussion
The endogenous purine nucleoside adenosine, produced at sites of cellular stress, is involved in both physiological and pathological processes including several diseases characterized by chronic inflammation such as rheumatoid arthritis, chronic obstructive pulmonary disease, inflammatory bowel disease, and ischemia-reperfusion injury (1–5, 36). Monocyte and monocyte-derived macrophage cells represent pivotal players involved in orchestrating inflammatory events through the production of cytokines, chemokines, growth factors, and angiogenic factors that regulate inflammation, angiogenesis, and granulation tissue formation. Importantly, the loss of controlling macrophage activation responses during an inflammatory response can result in injury leading to loss of tissue function (6). It has recently been established that engagement of adenosine receptors on macrophage cells can control the balance between TLR-induced inflammatory and regulatory macrophage differentiation (37). However, what remains to be elucidated are the cell-specific signal transduction events and the transcriptional mediators that act to promote adenosine receptor–mediated outcomes during these responses. In this study, we demonstrate that adenosine controls the nuclear orphan receptor NR4A1-3 subfamily gene expression and activity in myeloid cells. We further demonstrate that during an inflammatory response, adenosine modulation of NR4A2 receptor activity acts to limit NF-κB–mediated effects on gene transcription, and that loss of NR4A2 expression leads to enhanced NF-κB activity and hyperinflammatory responses in myeloid cells (Fig. 7).
Regulation of the inflammatory response is controlled by alternatively activated macrophages, with classically activated (M1) macrophages producing a wide variety of proinflammatory cytokines and chemokines. By contrast, alternatively activated (M2) macrophages participate in anti-inflammatory responses and promote inflammatory resolution and tissue repair (6). Significantly, synergistic interactions between TLR/MyD88 and adenosine A2A receptor signaling have emerged in the transition of macrophages to the M2 phenotype in vivo (38). In this study, we reveal that adenosine modulation of NR4A receptors can be controlled through the adenosine A2a receptor subtype and that A2a receptor activation of NR4A1-3 receptor synthesis is further enhanced in TLR-stimulated macrophages. Macrophage cell phenotype is not only characterized by its secretory capacity of cytokines and chemokines but also expression of surface markers including CD86, CD80, CCR7, and MHC class II (HLA-DR) for M1 and CD163, CD206, and mannose receptor for M2 characterization. Using a prototypical model for M1 macrophage differentiation (IFN-γ + LPS), we reveal adenosine significantly reduced M1-driven CD80 expression whereas attenuating CCR7 surface expression. In dendritic cells, adenosine has been shown to enhance LPS-driven CD80, CD86, CD54, MHC class I, and MHC class II, further revealing cell type and differentiation factor specificity regarding adenosine effects on cell differentiation marker expression (39). Depletion of NR4A2 or NR4A3 does not alter this adenosine-driven attenuation of surface marker expression. Intriguingly, we show that NR4A2 depletion significantly potentiates M1-driven CCR7 and CD86, whereas attenuating MHC class II (HLA-DR) expression. NR4A3 depletion, however, although having no effect on M1-driven CCR7 or CD86 expression, significantly attenuates MHC class II (HLA-DR) expression. Recent studies have shown NR4A1 depletion enhances MHC class II (HLA-DR) surface expression in peritoneal macrophages isolated from ApoE−/− mice fed a western diet (16). In this article, we reveal NR4A2 and NR4A3 may play opposing roles in MHC class II (HLA-DR) expression. However, specific factors driving differentiation may play an important role, as evident by enhanced MHC class II (HLA-DR) surface expression in NR4A2-depleted cells differentiated using LPS alone in the absence of IFN-γ. These data expand recent reports that identify NR4A receptors as effector molecules of TLR and cytokine signaling in human and murine macrophage cells and as important regulators of cellular differentiation (15–18, 40).
Treatment of macrophages with LPS, cytokines, or oxidized lipids triggers the transcriptional induction of all three NR4A receptors, and functional studies primarily demonstrate anti-inflammatory and reparative functions for NR4A1 receptors in these cells (16, 41). The role of NR4A1 receptors as transcriptional regulators of macrophage phenotypic differentiation has recently been explored and verifies that depletion of NR4A1 in monocytes and macrophages results in enhanced TLR-mediated signaling and polarization of macrophages toward a proinflammatory M1 phenotype (16). Our data extend these observations as we observe that depletion of additional family members NR4A2 and NR4A3 in THP-1 cells results in altered macrophage activation. We reveal an enhanced proinflammatory phenotype resulting in increased cytokine and chemokine production in both NR4A2- and NR4A3-depleted cells, whereas each family member displays distinct alterations in the expression of cell-surface differentiation markers.
In the presence of TLR agonists such as endotoxin (LPS), adenosine strongly and synergistically upregulates the expression of the angiogenic growth factor, VEGF, by macrophages, thus acting as an angiogenic switch leading to physiological angiogenesis, inflammatory resolution, and tissue repair (5). Importantly, adenosine A2A receptors have been shown to regulate the processes involved in healing, including the transition of macrophages to the regulatory phase. Macrophages from mice lacking A2A receptors express enhanced inflammatory genes including TNF-α in response to TLR agonists (42) but fail to undergo the angiogenic switch in response to A2A receptor agonists (38, 43). Furthermore, bone marrow–derived macrophage cells lacking the surface enzyme CD39, whose activity is needed to control endogenous adenosine production, are unable to transition to a regulatory state, and cells are rendered hyperinflammatory in response to TLR agonists (37). Interestingly, this study suggests that in response to LPS, adenosine attenuates inflammatory cytokine production via A2bR signaling (37). Notwithstanding the specific A2 receptor involved, these studies collectively suggest that switching monocytes/macrophages to a regulatory phenotype may be a prerequisite for adenosine-dependent resolution of inflammation. We have identified the NR4A subfamily as mediators of adenosine signaling in myeloid cells. Our findings indicate that in cells of myeloid origin, adenosine potently induces NR4A gene and protein expression primarily through the A2a receptor. Within PMA-differentiated monocyte cells, concentrations (1.0 and 10 nM) of the A2a agonist CGS-21680 leads to enhanced NR4A expression. A1 (2-chloro-N6-cyclopentyladenosine) and A3 (2-CI-IB-MECA) receptor–specific agonists, at all concentrations examined, displayed no effect on NR4A gene expression, thus excluding a role for A1 or A3 receptors in modulating NR4A expression in this cell type (data not shown). Of note, however, modest increases in NR4A expression are observed with the A2b BAY-60-6583–specific agonist at ≥100 nM concentrations. From established Ki values, this A2b receptor agonist concentration is specific for the A2b receptor (31); therefore, we cannot exclude A2b receptor involvement in modulating NR4A expression under these cell conditions. In LPS-primed THP-1 cells, modulation of NR4A expression cannot be measured in response to A1- and A3-specific receptor agonists at concentrations tested (data not shown). Furthermore, under these stimulated conditions, the specific A2b receptor agonist BAY-60-6583 displayed no change in NR4A mRNA levels, whereas the specific A2a receptor agonist CGS-21680 leads to significantly enhanced expression. The involvement of the A2a receptor in mediating adenosine effects on NR4A gene expression is consistent with a recent study where mice deficient in A2a receptor (A2aR−/−) exhibit prolonged peritonitis after i.p. thioglycollate injection (44). Further characterization after the isolation of A2aR−/− thioglycollate-elicited peritoneal macrophages (A2aR−/− thio-pMacs) reveals that these cells have lost their ability to upregulate the expression of NR4A1, confirming the essential role of the A2a receptor in this process.
As ligand-independent and constitutively active receptors, activity of the NR4A1-3 transcription factors is tightly controlled at the level of expression, nuclear localization, coactivator/corepressor recruitment, and posttranslational modifications (11, 14). Enhanced NR4A activity is not only controlled by NF-κB activity but NR4A receptors have been shown to modulate NF-κB functions downstream of LPS/TLR4 and TNF-α signaling in a dynamic fashion, either repressing or enhancing NF-κB target gene expression in monocyte/macrophage cells (13, 40, 45, 46). In microglial cells, NR4A2 is required for clearance of NF-κB from target gene promoter during TLR4-mediated inflammatory events in vitro and in vivo (13). These NR4A2-dependent effects lead to the fine-tuning of inflammatory responses after NR4A2 docking with NF-κB/p65 subunits on target genes, promoting the recruitment of the CoREST corepressor complex and subsequent removal of NF-κB/p65 and transcriptional repression of several TLR-induced proinflammatory genes (13). We show that nuclear levels of NF-κB/p65 are enhanced in TLR/adenosine-stimulated, NR4A2-depleted macrophage cells. We further establish that, after LR/adenosine receptor stimulation, NR4A2 depletion results in significant binding of NF-κB/p65 to a κB consensus binding motif within the MIP-3α proximal promoter leading to increased gene transcription and protein secretion. Collectively, these data suggest that adenosine may modulate the inflammatory response of monocyte-macrophage cells, in part by increasing NR4A2 expression/activity, which in turn acts to prevent excessive NF-κB activation leading to the attenuation of inflammation. Importantly, we reveal adenosine receptor–mediated attenuation of TNF-α during TLR4 stimulation is maintained in NR4A-depleted cells, suggesting a greater level of complexity regarding the role of NR4As in specific adenosine-modulated target genes.
Critical to uncovering the contribution NR4A1-3 receptors play in guiding inflammatory responses and inflammatory outcome is the study of distinct models of inflammation-associated disease. A recent report uncovers NR4A1 receptor participation in a murine model of acute myocardial infarction during distinct phases of monocyte/macrophage phenotypic differentiation (41). Monocyte/macrophage NR4A1 activity is required for two critical phases of acute myocardial infarction, which are essential for tissue remodeling and healing. NR4A1 promotes and controls the early monocyte proinflammatory responses, which is required for the initiation of inflammatory resolution and subsequent monocyte-derived, macrophage-dependent reparative/healing phase. Similar molecular mechanisms are used in controlling immune homeostasis in LPS-induced sepsis to permit NR4A1-mediated control of hyperinflammatory responses (47). From such studies it is becoming evident that the local environment influences the differentiation stage/function of immune cells, thus determining the cell-specific contribution NR4A receptors play in guiding the magnitude of inflammatory responses and outcomes (41). Within this study, we report a previously unidentified role of NR4A2 in limiting adenosine-mediated NF-κB/p65 hyperinflammatory responses in TLR4/TNF-α–primed cells. We observe adenosine receptor–mediated attenuation of LPS-induced TNF-α, and the magnitude of this effect is maintained in NR4A2/3-depleted cells where TNF-α production and secretion are significantly elevated. Furthermore, we identify that adenosine receptor stimulation of TLR4-primed monocyte cells promotes expression of MIP-3α and IL-23p19, with levels of MIP-3α protein significantly enhanced in NR4A2-depleted cells, coinciding with enhanced p65 binding to the proximal MIP-3α promoter. MIP-3α and IL23p19 are pivotal mediators produced by activated macrophages and involved in dendritic cell, B cell, and T cell (T regulatory cell [Treg] and Th17) recruitment and activation (48). Interestingly, adenosine has been shown to promote Treg development and inhibit proinflammatory Th17 responses through A2a receptor signaling (49, 50). More recent studies endorse the observation that adenosine-dependent Treg recruitment, via A2bR engagement, helps to limit inflammatory responses in endotoxin-induced pulmonary inflammation (51). Notably, under physiological and pathological conditions, NR4A2 controls Treg induction and suppression of aberrant production of Th1 cytokines (17, 18).
The ability of NR4A1-3 receptors to integrate and limit inflammatory signaling and promote tissue repair makes these receptors potential targets for therapeutic intervention. Although NR4A1-3 activity does not appear to be regulated by endogenous ligands, pharmacological modulation of NR4A activity can be achieved with agents including the antineoplastic agent, 6-mercaptopurine, and a series of 1,1-bis(3-indolyl)-1-(p- substituted phenyl) methanes identified as potential NR4A1/2 agonists (52–54). Recent studies reveal that in microglial cells, 1,1-bis(3-indolyl)-1-(p- substituted phenyl) methanes can increase NR4A2 nuclear localization, stabilization of Co-REST and NCoR2 complexes, and subsequent inhibition of NF-κB/p65 signaling (55). In addition, altering the expression level of these constitutively active receptors may be a feasible approach to modulate NR4A target genes in a cell-specific manner. The in vivo therapeutic potential of targeting NR4A/NF-κB interactions to control immune homeostasis has recently been uncovered (47). A chemical compound n-pentyl 2-[3,5-dihydroxy-2-(1-nonanoyl)-phenyl]acetate targets the ligand-binding domain of NR4A1 and prevents phosphorylation of NR4A1 by TLR-activated p38a. Loss of NR4A1 phosphorylation restores and maintains NR4A1 suppression of a hyperinflammatory response through NR4A1 inhibition of NF-κB/p65 binding to κB sites within promoter regions of target genes. Additional in vivo studies will determine whether the clinical benefits of adenosine on limiting macrophage activation and promoting inflammatory resolution may be mediated through molecular cross-talk with the NR4A orphan nuclear receptor subfamily. In this study, we reveal that selective A2aR agonists act as potent inducers of NR4A expression and activity in myeloid cells leading to contained inflammatory responses, thus elucidating a new regulatory pathway that could be targeted to treat diseases in which monocyte/macrophage hyperactivation plays a principal role.
Footnotes
This work was supported by Science Foundation Ireland Grant IN.1B2613 to E.P.M.
The online version of this article contains supplemental material.
References
Disclosures
The authors have no financial conflicts of interest.