Early tyrosine phosphorylation events following adenosine A_2A receptor in human neutrophils: identification of regulated pathways

Intracellular signaling pathways targeted by the A2AR modulate neutrophil function.

Abstract

Activation of the adenosine _2A receptor (A_2AR) elevates intracellular levels of cAMP and acts as a physiologic inhibitor of inflammatory neutrophil functions. In this study, we looked into the impact of A_2AR engagement on early phosphorylation events. Neutrophils were stimulated with well-characterized proinflammatory agonists in the absence or presence of an A_2AR agonist {3-[4-[2-[ [6-amino-9-[(2R,3R,4S,5S)-5-(ethylcarbamoyl)-3,4-dihydroxy-oxolan-2-yl]purin-2-yl]amino] ethyl] phenyl] propanoic acid (CGS 21680)}, PGE₂, or a mixture of the compounds RO 20-1724 and forskolin. As assessed by immunoblotting, several proteins were tyrosine phosphorylated; CGS 21680 markedly decreased tyrosine phosphorylation levels of 4 regions (37–45, 50–55, 60, and 70 kDa). Key signaling protein kinases—p38 MAPK, Erk-1/2, PI3K/Akt, Hck, and Syk—showed decreased phosphorylation, whereas Lyn, SHIP-1, or phosphatase and tensin homolog (PTEN) was spared. PGE₂ or the intracellular cAMP-elevating combination of RO 20-1724 and forskolin mostly mimicked the effect of CGS 21680. Together, results unveil intracellular signaling pathways targeted by the A_2AR, some of which might be key in modulating neutrophil functions.

Introduction

Polymorphonuclear leukocytes (neutrophils), the first cell type to migrate out of the circulation to accumulate at inflammatory lesions, perform host defense functions against infections, such as the phagocytosis of invading microorganisms, generation of oxygen-derived reactive agents, and release of proteolytic enzymes and antimicrobial proteins. In response to specific external signals, neutrophils can also produce an array of soluble mediators, including TNF, CCL, and CXCL chemokines and eicosanoids, each of which differentially contributes to the onset, progression, and resolution of a proper inflammatory response [1]. Surface receptors sense the microenvironment and regulate the cellular responses of neutrophils [2]. Tyrosine phosphorylation ranks among the earliest events resulting from receptor activation, which either possesses an intrinsic tyrosine kinase domain or can associate with cytoplasmic tyrosine kinases [3]. Downstream coupling with serine/threonine kinases provides an initial amplification stage that may activate several signal transduction cascades [4] toward the appropriate activation of downstream effector systems. This way, external signals can be linked to virtually all aspects of neutrophil functions, including apoptosis, cell motility, metabolism, phagocytosis, degranulation, and gene expression, among others [5]. Despite their crucial role in host defense, neutrophils have also been associated with a variety of inflammatory diseases and worsening of disease states by promoting acute responses leading to increased tissue injury in chronic conditions, such as emphysema, adult respiratory distress syndrome, and rheumatoid arthritis [6]. Whereas endogenous mechanisms of control are in place to steer the normal activation of neutrophils within appropriate boundaries, their nature remains elusive.

Adenosine is an extracellular messenger with a large spectrum of biologic activities, including a protective role in acute inflammation and the promotion of wound healing. The A_2AR takes part in a nonredundant, physiologic, negative-feedback mechanism that limits and terminates both tissue-specific and systemic inflammatory responses. Mice deficient in the A_2AR developed extensive tissue damage in response to subthreshold concentrations of inflammatory stimuli, highlighting the pivotal importance of A_2AR in terminating inflammation [7]. In neutrophils, A_2AR activation provoked a diminution in adhesion to endothelial cells, phagocytosis, and generation of superoxide anions. Moreover, A_2AR agonists inhibited leukotriene and platelet-activating factor synthesis and stimulated cyclooxygenase-2 expression [8], in addition to having a noticeable inhibitory impact on the in vitro and in vivo generation of TNF and several other neutrophil-derived cytokines [1] and on the expression of regulatory factors, enzymes, and receptors [9]. Much of the anti-inflammatory activities of A_2AR engagement can be recapitulated by a rise in intracellular cAMP concentration [9]. PGE₂, which also increases intracellular cAMP, similarly inhibits several neutrophil inflammatory functions [9, 10]. However, the impact of A_2AR engagement on early, intracellular phosphorylation events in neutrophils remains poorly documented.

In the present study, we stimulated human neutrophils in the absence or presence of A_2AR engagement and analyzed tyrosine phosphorylation patterns by immunoblots for the identification of some intracellular pathways regulated by the A_2AR.

MATERIALS AND METHODS

Antibodies

The anti-p38 MAPK (Cat. #506123) was purchased from EMD Millipore (Billerica, MA, USA). Anti-phosphorylated (Thr180/Tyr182) p38 MAPK (cat. #9216), anti-phosphorylated (Ser473) Akt (cat. #9271), anti-phosphorylated (Tyr525/526) Syk (Cat. #2711), anti- p44/42 MAPK (ERK1/2; 3A7; Cat. #9107), anti-phosphorylated (Ser380/Thr382/383) PTEN (Cat. #9554), and anti-phosphorylated (Tyr1020) SHIP-1 (Cat. #3941) were purchased from Cell Signaling Technology (Danvers, MA, USA). Anti-Syk (4D10; Cat. #SC-1240), anti-Hck (N-30; Cat. #SC-72), and anti-Lyn (Cat. #SC-15) were purchased from Santa Cruz Biotechnology (Dallas, TX, USA). Anti-phosphorylated (Tyr185/187) ERK1/2 (Cat. #44-6806) was purchased from BioSource International (Camarillo, CA, USA). Secondary anti-mouse (#NXA931) and anti-rabbit (#NA934V) antibodies were obtained from Amersham Biosciences (Baie d’Urfé, Québec, Canada).

Neutrophil isolation and incubations

The Université Laval Ethics Committee approved all experiments involving human tissues. Informed consent was obtained in writing. Polymorphonuclear leukocytes were isolated as described earlier [11]. Purified granulocytes (>95% neutrophils, <5% eosinophils) contained <0.1% monocytes. Adenosine deaminase (0.1 U/ml) was added to cell suspensions, 20 min before stimulation, to prevent adenosine accumulation in the extracellular milieu. Anti-inflammatory compounds dissolved in dimethylsulfoxide were added to cell suspensions, 10 min before stimulation, at 37°C. Reactions were stopped by adding 200 μl cell suspension to the same volume of 2× Laemmli sample buffer and boiled for 7 min.

Immunoprecipitations

Samples were processed for immunoprecipitations, essentially as described [12]. In brief, cell lysates were incubated with anti-Syk or anti-Hck or anti-Lyn antibodies previously bound to protein A-Sepharose. Beads were then washed 3 times with cold washing buffer and processed for immunoblots.

Western blots

Samples were loaded onto SDS-PAGE gels, and proteins were transferred to polyvinylidene difluoride membranes, which were blocked, and then incubated with the first antibodies—anti-phosphotyrosine (1/4000, final dilution) or anti-phosphoproteins—or anti-total proteins (1/1000, final dilution) for 1 h at 37°C. Rinsed membranes were incubated with an HRP-labeled secondary antibody for 60 min at 37°C. A chemiluminescence kit (ECL; Amersham Biosciences, Baie d’Urfé, Quebec, Canada) was used according to the manufacturer’s instructions.

Statistical analysis

Where applicable, values are expressed as the means ± sem. Statistical analysis was performed using a 2-tailed Student's t test (paired comparisons). Differences were considered significant when P < 0.05.

RESULTS AND DISCUSSION

Early tyrosine phosphorylation events in stimulated human neutrophils

We stimulated freshly isolated human neutrophils for 2 min with 30 ng/ml IL-1β, 100 ng/ml LPS, 100 nM TNF, 100 nM fMLP, or 1.4 nM GM-CSF and analyzed the resulting tyrosine phosphorylation patterns. As shown in Fig. 1A, agonists elicited different events of tyrosine phosphorylation: LPS in the 95 and 75 kDa regions; TNF in the 95 kDa region; fMLP in the 95, 75, and 65 kDa regions; and GM-CSF in the 150, 95, 85, 70, and 55 kDa regions. IL-1β had little to no impact. A strong doublet of molecular mass <50 kDa was constitutively tyrosine phosphorylated in nonstimulated cells, on which stimulation had no clear effect. The fullest pattern was obtained with neutrophils stimulated simultaneously with all 5 agonists (cocktail), approximating the accumulation of specific stimulations. Most affected regions were ∼150 kDa, between 95 and 75, 70, 65–55, and <40 kDa. In all instances, shorter stimulation times merely resulted in weaker patterns (data not shown). In the course of inflammation, neutrophils shall have to respond to a mixture of signals, such as cytokines, chemoattractants, and particulate agonists, rather than to any single one [4]. The selected agonists each engaged distinct classes of receptors and provoked a most comprehensive pattern of phosphotyrosinated proteins. Among them, GM-CSF interacts with a receptor having tyrosine kinase activity [13]. A surrogate of bacteria-derived formylated peptides, fMLP signals through 7-transmembrane domain receptors FPR1 and FPR1 ligand, linked to heterotrimeric GTP-binding proteins to activate PLC, PLD, PLA₂, PI3K, and downstream, PKC and MAPKs. LPS interacts with circulating LPS-binding protein and binds to TLR4, CD14, and myeloid differentiation factor 2, engaging MyD88-dependent and -independent signaling pathways, leading to MAPK and NF-κB activation [14]. PI3K and AKT are also important factors downstream of MyD88. TNF binds to its ceramide-linked receptors TNFRSF1A/TNFR1 and TNFRSF1B/TNFBR, members of the death domain-containing receptors, and the TNFR-associated factor-interacting receptors, respectively. TNFR1 is a potent activator of gene expression [15]. Finally, IL-1β engages its family of receptors (IL-1R1, IL-1R2, IL1-R1 ligand) in the Ig domain superfamily and triggers a complex sequence of phosphorylation and ubiquitination events [16]. In turn, the multilateral stimulation approach adopted herein solicited a large variety of intracellular signaling pathways, as assessed by phosphorylation events. These pathways included the following: TLR/MyD88/p38-ERK, Src kinases, PI3K/Akt, Jak/Stats, NF-κB, PKA, and PKC among others. Small G protein effectors were extensively solicited, as well as inflammasomes and the mammalian target of rapamycin pathway. This situation, where so many pathways appeared stimulated, appears favorable for uncovering points of modulation following A_2AR engagement.

Figure 1. The impact of adenosine and cAMP-elevating agents on early tyrosine phosphorylation patterns in stimulated human neutrophils.

(A) Tyrosine phosphorylation patterns obtained by Western immunoblotting from neutrophils stimulated 2 min at 37°C, individually with each indicated agonist or mixed (cocktail). Graphs depict the intensity of each band, as determined by densitometric analysis. For every condition, signals different from control were shaded. (B) Neutrophils were preincubated with CGS 21680 (CGS), PGE₂, or a mixture of RO 20-1724 (RO) and forskolin (F), before stimulation with the cocktail. Phosphoprotein regions (Reg) that are particularly affected by adenosine are shown in thick-frame, black-bounded boxes. For each panel, a typical immunoblot is shown; results are compiled from n = 5 independent experiments performed in identical conditions with different donors and expressed relative to the response of cocktail treatment alone (means ± sem; n = 5). Western blot regions: regions 1 and 2 include proteins between 150 and 100 kDa; regions 3–6, 100–75 kDa; regions 7–10, 75–50 kDa; and regions 11–14, 50–37 kDa.

We preincubated the cells with the specific A_2AR agonist, CGS 21680 (1 μΜ) before stimulation with the cocktail of agonists. Phosphoproteins in the regions 7–8, 10, and 14 were most affected (Fig. 1B). Likewise, preincubation with PGE₂ (10 μM) or to a lesser extent, with a combination of the phosphodiesterase IV inhibitor RO 20-1724 (10 µM) and the AC activator forskolin (50 µM) essentially reproduced the effects of the A_2AR agonist on the pattern of tyrosine phosphorylation. When taking into account molecular weights of the affected regions following A_2AR engagement (indicated in Fig. 1B), target proteins recognized for their importance in neutrophil signaling include Syk, p38 MAPK, ERK1/2, Hck, p55, PI3K, and Lyn (Supplemental Table 1) [31–37]. Conversely, when considering neutrophil functions that are modulated by A_2AR engagement and kinases known to be involved in mediating these functions, some of the most studied kinases are the following: Syk, Src, p38 MAPK, ERK1/2, PKC, PI3K/Akt, and JNK (Supplemental Table 2). Thus, we looked into the impact of A_2AR engagement on the phosphorylation status of each one of these kinase families.

Syk and Src kinases

Syk and the Src kinase family members Hck and Lyn signal immediately downstream from different receptors in immune cells [17]. Activated through immunoreceptors—C-type lectins, integrins, and G protein-coupled receptors—these kinases, tyrosine-phosphorylated substrates, mediate a broad range of signaling responses, including cell adherence, degranulation, and phagocytosis [18]. With respective molecular masses of 60, 53/56, and 72 kDa, Hck, Lyn, and Syk stand as potentially good candidates for regulation by A_2AR. We immunoprecipitated lysates from cells stimulated with the cocktail for 2 min in the absence or presence of CGS 21680 or 2 distinct cAMP-elevating agents (PGE₂ or RO 20-1724 + forskolin) with anti-Hck, -Lyn, or -Syk antibodies. Immunoprecipitates were immunoblotted for the detection of kinases in their tyrosine-phosphorylated form. Phosphorylated Hck, Lyn, and Syk were each detectable in nonstimulated neutrophils, and only Syk phosphorylation was increased in an appreciable fashion upon stimulation (Fig. 2A–C). CGS 21680 reduced phosphorylation levels of Hck well below levels observed in nonstimulated cells (Fig. 2A) and also prevented the stimulated phosphorylation of Syk (Fig. 2C; 38 ± 2 and 37 ± 8% for Hck and Syk, respectively). CGS 21680 had no measurable effect on the phosphorylation of Lyn (Fig. 2B). PGE₂ and RO 20-1724 + forskolin qualitatively reproduced the effects observed following the A_2AR engagement. In pilot experiments where neutrophils were stimulated with each of the 5 agonists separately for 2 min, only LPS induced Syk tyrosine phosphorylation within that time frame (Fig. 2D, inset). In LPS-stimulated cells pretreated with CGS 21680, PGE₂ or RO 20-1724 + forskolin prevented the tyrosine phosphorylation of Syk, essentially down to basal levels (Fig. 2D), as observed in cocktail-stimulated cells. These results identify Src kinases and Syk as the first and early targets of A_2AR engagement in TLR4-stimulated neutrophils, in line with previous studies that identified these kinases as playing important roles in innate immune signaling [19].

Figure 2. The impact of adenosine and cAMP-elevating agents on Src kinases and Syk early tyrosine phosphorylation.

Neutrophils were preincubated with CGS 21680, PGE₂, or RO 20-1724 + forskolin and then stimulated 2 min at 37°C with the cocktail (A–C) or LPS (D). Samples were lysed and processed for immunoprecipitation (IP) with anti-Hck (A), anti-Lyn (B), or anti-Syk (C and D) antibodies. Immunoblotting was conducted using the anti-phosphotyrosine (anti-TPO) antibody. Membranes were reprobed using anti-Hck, anti-Lyn, or anti-Syk. In each panel, 1 typical immunoblot is shown and is representative of at least n = 4 independent experiments. Bar graphs depict the phosphotyrosine levels relative to the response obtained in the absence of cAMP-elevating agents, as determined by densitometric analysis of the bands (means ± sem; n = 4). *Significantly different from samples stimulated in the absence of cAMP-elevating agents. GM, GM-CSF; WB, Western blot.

MAPKs

MAPKs are members of the serine/threonine protein kinase family and mediate signal transduction from the cell surface to the nucleus. Three major MAPK cascades exist in mammalian cells: ERK, p38 MAPK, and c-Jun N-terminal kinase [20]. In neutrophils, the ERK and p38 MAPK pathways are activated by chemoattractants, proinflammatory cytokines, LPS, and FcγR ligation [21] and have been reported to participate in many neutrophil functions, such as chemotaxis, adhesion, phagocytosis, granule secretion, and respiratory burst. Phosphorylation levels of p38 MAPK and ERK1/2 are significantly increased after 2 min of stimulation with the cocktail of agonists (Fig. 3A and B). Pretreatment of the cells with CGS 21680 partially prevented p38 MAPK and ERK phosphorylation (30 ± 6 and 44 ± 9%, respectively). Of all agonists used, only fMLP induced a significant phosphorylation of p38 MAPK and ERK1/2 within that time frame (Fig. 3C and D, insets). CGS 21680 markedly reduced fMLP-elicited phosphorylation of p38 MAPK and ERK1/2 (31 ± 6 and 41 ± 11%, respectively; Fig. 3C and D). Preincubation with PGE₂ or RO 20-1724 + forskolin mimicked the effect of CGS 21680 on the stimulated phosphorylation levels of p38 MAPK and ERK1/2. It is known that agonists, such as LPS, TNF, or GM-CSF, induce p38 MAPK or ERK1/2 phosphorylation, albeit at later time points [22, 23]. Accordingly, after 10 min of stimulation, CGS 21680 markedly reduced phosphorylation of p38 MAPK, as well as the TNF- or GM-CSF-elicited phosphorylation of ERK1/2 (Fig. 4A and B).

Figure 3. The impact of adenosine and cAMP-elevating agents on p38 MAPK and ERK1/2 phosphorylation.

Neutrophils were preincubated for 10 min with CGS 21680, PGE₂, or RO 20-1724 + forskolin and stimulated with the cocktail (A and B) or fMLP (C and D) for 2 min. In each panel, 1 typical immunoblot is shown and is representative of at least n = 6 independent experiments. Bar graphs depict the p38 MAPK (A and C) and ERK1/2 (B and D) phosphorylation levels (p-p38 and p-ERK, respectively) relative to the response obtained in the absence of cAMP-elevating agents, as determined by densitometric analysis of the bands (means ± sem; n = 6). *Significantly different from stimulated neutrophils in the absence of cAMP-elevating agents.

Figure 4. The impact of adenosine on p38 MAPK, ERK1/2, and Akt phosphorylation for longer stimulation time periods.

Neutrophils were incubated for 10 min with CGS 21680 and stimulated with LPS or TNF for 15 min or GM-CSF for 10 min. (A) Phospho-p38 (p-p38 MAPK); (B) phospho-ERK (p-ERK); (C) phospho-Akt (p-Akt). In each panel, 1 typical immunoblot is shown and is representative of at least n = 3 independent experiments.

PI3K

PI3Ks are serine-threonine kinases involved in signal transduction and amplification [24]. Tyrosine phosphorylation of p85-binding sites within growth factor receptor cytoplasmic domains and linker molecules results in the activation of PI3K [25], which can phosphorylate phosphatidylinositol 4,5-bisphosphate to generate PIP₃, a second messenger that recruits and activates downstream targets, including Akt (phosphorylated on Ser473). Akt phosphorylation levels increased significantly after 2 min of stimulation (Fig. 5A). Preincubation with CGS 21680 caused a marked decrease in Akt phosphorylation induced by the cocktail (55 ± 5%). Of all agonists, only fMLP induced a significant phosphorylation of Akt after 2 min of stimulation (Fig. 5B, inset). Similar to its effects on cocktail-stimulated cells, CGS 21680 markedly reduced fMLP-elicited phosphorylation of Akt (41 ± 6%). At longer stimulation times, CGS 21680 also considerably reduced TNF- or GM-CSF-elicited phosphorylation of Akt (Fig. 4C).

Figure 5. The impact of adenosine and cAMP-elevating agents on Akt, SHIP-1 and PTEN phosphorylation.

Neutrophils were preincubated with CGS 21680, PGE₂, or RO 20-1724 + forskolin and then stimulated 2 min with the cocktail. In each panel, 1 typical immunoblot is shown and is representative of at least n = 5 independent experiments. Bar graphs depict the Akt (A and B), SHIP-1 (C), and PTEN (D) phosphorylation levels (p-Akt, p-SHIP1, and p-PTEN, respectively) relative to the response obtained in the absence of cAMP-elevating agents, as determined by densitometric analysis of the bands (means ± sem; n = 5). *Significantly different from stimulated neutrophils in the absence of cAMP-elevating agents.

Phosphatases

PI3K-derived generation of PIP₃ can be counter balanced by the action of PTEN or by SHIP. These phosphatases dephosphorylate PIP₃, thereby controlling the activation of Akt [26]. It has been well documented that the tyrosine phosphorylation of SHIP-1 correlates with its activation [27]. PTEN, on the other hand, can be serine/threonine as well as tyrosine phosphorylated [28]. We found the phosphatases SHIP-1 and PTEN to be constitutively phosphorylated in human neutrophils; moreover, neither tyrosine phosphorylation of SHIP-1 nor serine/threonine phosphorylation of PTEN was modified in response to the cocktail (Fig. 5C and D). Of importance, CGS 21680 and cAMP-elevating agents spared their phosphorylation status.

Tyrosine phosphorylation of specific targets ranks among the earliest events following stimulation of receptors, which in turn, engages additional intracellular pathways leading to cellular responses [5]. The initial finding from this study was that A_2AR engagement could have a specific impact on the pattern of proteins bearing phosphorylated tyrosine residues. PGE₂ or a combination of forskolin and RO 20-1724 both had a remarkably similar impact to A_2AR engagement, reinforcing the concept of intracellular cAMP involvement. In stimulated cells, these cAMP-elevating agents had a clear inhibitory effect on the tyrosine phosphorylation of key signaling enzymes, namely Syk, PI3K, ERK, and p38 MAPK. MAPKs and PI3K/Akt play a significant role in cell proliferation, differentiation, and apoptosis, initiated by various types of extracellular stimuli, such as oxidative stress. Activated MAPKs induce production of proinflammatory cytokines, recruitment of leukocytes through the modulation of adhesion molecules, and the generation of reactive oxygen species [29]. PI3Kγ, mainly expressed in leukocytes, is positively linked with leukocyte activation, migration, and superoxide production [30]. The Src kinase family plays important roles in innate immune signaling. Src and Syk kinases might function in a linear signaling pathway in immune cells, whereby Src family kinases phosphorylate ITAM-containing proteins, which in turn, recruit and activate Syk. Engagement of A_2AR decreased the tyrosine phosphorylation levels of all of these kinases but with one exception: Lyn. Of interest, neutrophils deficient in Lyn manifest hyper-responsiveness to the integrin regarding tyrosine phosphorylation and effector functions, including adhesion, respiratory burst, and secondary granule release [19], identifying Lyn as a negative regulator of integrin-dependent functional responses in neutrophils [19]. Moreover, Lyn phosphorylates neutrophil inhibitory receptors, which in turn, recruit Src homology region 2 domain-containing phosphatase 1 (and other phosphatases) to down modulate signaling. As well, A_2AR engagement largely spared the phosphorylation of the phosphatases SHIP-1 and PTEN, also favoring an anti-inflammatory profile. Overall results are schematized in Fig. 6.

Figure 6. Model of early signaling pathways regulated by the A_2AR and cAMP-elevating agents in human neutrophils.

Intracellular levels of cAMP are regulated by 2 enzymatic activities: AC, forming cAMP from ATP, and phosphodiesterase (PDE), which degrades cAMP. CGS 21680, a selective A_2AR agonist, and PGE₂ increase cAMP level after receptor engagement, as a result of AC activation. AC activator forskolin and the phosphodiesterase IV inhibitor RO 20-1724 also increase intracellular cAMP levels, promoting activation of the cAMP-dependent kinase (PKA) and possibly other pathways also. Inflammatory stimuli, such as LPS, fMLP, TNF, GM-CSF, or IL-1β, interact with their receptors on membranes of neutrophils and activate intracellular signaling pathways. Shortly after stimulation, several proteins are tyrosine phosphorylated. Activation the A_2AR with CGS 21680 weakened phosphorylation of key signaling protein kinases: p38 MAPK, Erk1/2, PI3K/Akt, Hck, and Syk. PGE₂ or a mixture of the compounds RO 20-1724 and forskolin mimicked the effect of CGS 21680, supporting a cAMP-dependent effect. CGS 21680 had no effect on Lyn, SHIP-1, or PTEN basal phosphorylation levels.

In conclusion, results show that A_2AR engagement can modulate tyrosine phosphorylation pathways in stimulated neutrophils and identify the kinases Syk, Hck, p38, ERK1/2, and PI3K as initial targets of A_2AR and intracellular cAMP elevation, while the phosphatases SHIP and PTEN are spared. These pathways reinforce the concept of stronger PKA-dependent events and further suggest profound reprogramming of neutrophils by adenosine. Additional work, using single-agonist approaches, time courses, and cellular assays, is being performed to determine which of these pathways are mediators of adenosine anti-inflammatory activities, which will be key to the development of new therapeutic approaches aiming for more efficient control of inflammation.

AUTHORSHIP

M.S.G. contributed to the study design, performed the experiments, interpreted the data, and primarily wrote the article. M.P. performed study design, interpreted the data, and wrote and revised the manuscript.

Glossary

A_2AR

adenosine _2A receptor

AC

adenylate cyclase

CGS 21680

3-[4-[2-[ [6-amino-9-[(2R,3R,4S,5S)-5-(ethylcarbamoyl)-3,4-dihydroxy-oxolan-2-yl]purin-2-yl]amino] ethyl] phenyl] propanoic acid (specific agonist of the adenosine _2A receptor)

FPR

formyl peptide receptor

PIP₃

phosphatidylinositol-3,4,5-trisphosphate

PKA/C

protein kinase A/C, PLA₂/C/D= phospholipase A₂/C/D

PTEN

phosphatase and tensin homolog

TNFRSF

TNFR superfamily

DISCLOSURES

The authors declare no conflicts of interest.