Enhancement of Mast Cell Degranulation Mediated by Purinergic Receptors’ Activation and PI3K Type δ

Mast cells, by virtue of their wide organ distribution, play important roles in both localized and systemic inflammatory processes and the host’s immune response (1). Mast cells degranulation, which follows the activation of the high-affinity IgE receptor (FcεRI), is associated with the release of cytokines, lipids, and chemical mediators of inflammation, including histamine, that can result in edema and pruritus via the type I allergic response (2, 3). In human subjects, these responses can not only decrease quality of life but may also be life-threatening. The function of human mast cells (HMCs) in general and cultured human lung mast cells (HLMCs) in particular during physiological and pathophysiological conditions have been the subject of multiple studies (4–9).

Previously, we have reported that extracellular adenine nucleotides can either enhance or suppress allergic stimulation-induced degranulation (i.e., histamine release) in HLMCs (10, 11). Thus, we have hypothesized that the exacerbation of allergic symptoms could be mediated by the release of intracellular adenine nucleotides from cells associated with inflammatory processes. Indeed, it is well documented that ATP is released from different types of cells under inflammatory conditions (12, 13).

Purinergic receptors (PRs) are divided into two superfamilies: the adenosine receptors, P1R (consists of four receptors: A1AdoR, A2aAdoR, A2bAdoR, and A3AdoR); and P2R, the purine and pyrimidine receptors. The P2Rs consist of two families: purinergic P2Y receptor (P2YR), seven-transmembrane-domain G-protein–coupled receptors (metabotropic); and PR P2X ligand-gated ion channel (P2XR), trans-cell membrane cationic channels (ionotropic), of which eight and seven receptors have been cloned, respectively, heretofore (14, 15). P2X7R plays a critical role in inflammatory processes (16, 17). P2YRs are expressed by HMCs (17, 18); however, thus far, only one receptor (i.e., P2Y14R) has been proposed to be linked to histamine release (19).

In terms of endogenous extracellular nucleotides, ATP is the only ligand that activates both P2XR and P2YR. ATP is also the most potent agonist at the metabotropic purinergic P2Y11 receptor (P2Y11R) site; other P2YR subtypes are sensitive to other nucleotides, such as ADP or uridine-derived nucleotides (20, 21). It has been documented that adenosine 5'-(3-thio)triphosphate (ATPγS), a stable, nonhydrolyzable P2Y11R agonist, mimics endogenous ATP’s effects mediated by P2Y11R (22–25).

The purpose of this study was to investigate the mechanism of enhancement of a weak allergic stimulation (WAS)-induced degranulation in HMCs after the activation of P2YR in LAD2 cells (26–28). These studies may provide some insight as to why brief, minimal exposures to allergens (e.g., grass, mice, cat) can at times lead to profound clinical reactions.

LAD2 cells isolated from the human bone marrow was a gift from Dr. Arnold Kirshenbaum (National Institutes of Health, National Institute of Allergy and Infectious Diseases, Bethesda, MD) (29). These cells were grown in StemPro-34 media (Invitrogen, Carlsbad, CA), supplemented with 2 mM l-glutamine, 100 IU/ml penicillin and 50 μg/ml streptomycin (Meiji Seika, Tokyo, Japan), and 100 ng/ml human stem cell factor (Wako, Osaka, Japan). LAD2 cells were sensitized with anti–4-hydroxy-3-nitrophenylacetyl Ab (anti–NP-IgE, 1 μg/ml) (Bio-Rad, Hercules, CA) or untreated (control) for 24 h. Cells were washed twice with PBS and once with calcium-free Tyrode’s solution. WAS challenge was carried out in Tyrode’s solution (2 μM Ca²⁺) using Ag (NP-BSA) (0.13 μg/ml) (Biosearch Technologies, Hoddesdon, UK), for 20 min at 37°C, which was just above the threshold for degranulation. Cells were then sedimented and supernatants collected. Cells were similarly challenged with the addition of P1R and P2R subtype-selective agonists (i.e., adenosine, ATP, ADP, UTP, UDP, 2Me-S-ATP, and ATPγS). In other experiments, cells were pretreated with the Compound 15e (Fujifilm Wako, Osaka, Japan), a PI3K inhibitor, and then Ag or control challenged in the absence and presence of ATPγS. Degranulation was quantified by release of the granule component β-hexosaminidase (β-Hex) into the supernatant. This was determined by measurement of absorbance at 405 nm, after the addition to the supernatant of N-acetyl-β-glucosaminide (4 mM), a substrate of β-Hex, and incubation at 37°C for 90 min followed by addition of glycine (0.2 M). Total β-Hex was determined by treatment of the residual cell pellet with 1% Triton-X.

For adherence, LAD2 cells were cultured on fibronectin (40 µg/ml)-treated coverslips in StemPro-34 medium. Cells were loaded with fura 2-AM (2 µM), and fluorescence was monitored at an emission wavelength of 510 nm. [Ca²⁺]_i was expressed as the ratio of the 510 nm fluorescence intensity excited at 340 and 380 nm (I₃₄₀/I₃₈₀). To evaluate calcium mobilization caused by various stimuli, we challenged cells treated with WAS or untreated (control) with a calcium ionophore A23187 (Cayman Chemical, Ann Arbor, MI), nucleotides, or a nucleoside in Tyrode’s solution (2 μM Ca²⁺). In experiments designed to examine calcium flux, cells treated with anti–NP-IgE or untreated (control) were exposed to calcium ionophore A23187, nucleotides, or a nucleoside in Tyrode’s solution (2 μM Ca²⁺) for 20 min, followed by stimulation with NP-BSA at 37°C for 30 min or no stimulation.

Proteins were extracted from LAD2 cells (from different cell stocks based on the difference of the seeding time and culture periods) by radioimmunoprecipitation assay buffer containing both protease inhibitors (GE Healthcare, Uppsala, Sweden) and phosphatase inhibitors (Roche, Mannheim, Germany). The extracts were centrifuged at 34,500 × g for 10 min, and the supernatants were run on polyacrylamide gels. Abs for P2X7R, P2Y11R, ectonucleotide triphosphate dephosphorylase 1 (NTPD-1 = CD39), ecto-5'-nucleotidase (CD73), PI3K, phosphoinositide-dependent kinase-1 (PDK-1), protein kinase B (Akt), p-PI3K, p-PDK-1, p-Akt, and GAPDH were purchased from Abcam (Cambridge, UK), GeneTex (Irvine, CA), and Cell Signaling (Danvers, MA). HRP-conjugated secondary Abs (Rockland, Gilbertsville, PA) were used to detect the first Ab signals by H₂O₂-induced color development on polyvinylidene difluoride membranes. Western blotting images were quantitated by BioRad ChemiDoc Touch system (Hercules, CA) and ImageJ (National Institutes of Health, Bethesda, MD).

The short hairpin RNA (shRNA), including the “loop” and constructs for PI3K(δ) knockdown (K/D), were constructed as follows: top strand, 5'-CACCGGTAATTGAACCAGTAGGCACGAATGCCTACTGGTTCAATTAC-3'; and bottom strand, 5'-AAAAGTAATTGAACCAGTAGGCATTCGTGCCTACTGGTTCAATTACC-3'.

These oligo DNAs were annealed to form dsDNA for insertion into pENTR/U6 (Invitrogen, Carlsbad, CA), a shRNA plasmid. The plasmid was amplified in the One Shot TOP10 Chemically Competent E. coli (Invitrogen, Carlsbad, CA). The plasmid was then extracted from the E. coli and purified for RNA interference. The small interfering RNA (siRNA) system for P2Y11R mRNA (Santa Cruz Biotechnology, Santa Cruz, CA) was also used in LAD2 cells for the K/D of the P2Y11R expression. The pENTR/U6 without K/D construct and the control miRNA (Santa Cruz Biotechnology, Santa Cruz, CA) were used as controls for shRNA and siRNA procedures, respectively.

The shRNA plasmid was transfected into LAD2 cells using a MicroPorator (Digital Bio, Seoul, Korea) at 1800-V, 20-ms pulse. After transfection, the cells were cultured in StemPro-34 without antibiotics for 3 days before use in experiments. LAD2 cells were cotransfected with plasmids containing either the shRNA constructs for PI3K(δ) mRNA K/D or enhanced GFP (eGFP) expression as previously described (30). The eGFP expression was used for the transfection’s success rate and availability of the cells. LAD2 cells transfected only with the eGFP plasmid were used as control for effects of non-K/D in both Western blot analysis and functional assays. The siRNA transfection reagent for the P2Y11 siRNA and the control siRNA were also used (Santa Cruz Biotechnology, Santa Cruz, CA).

To assess the breakdown of ATP during the incubation periods, we collected the assay medium to quantify the net contents of ATP every 5 min. The collected aliquots of the medium were analyzed by use of the luciferin-luciferase procedure kit (Cayman Chemical, Ann Arbor, MI).

Data were analyzed by one-way ANOVA followed by Tukey–Kramer post hoc tests to reduce any false-positive or type I errors. A p value <0.05 was considered to be statistically significant.

LAD2 cells express three P1Rs (A2aAdo, A2bAdo, and A3Ado), two P2XRs (P2X1 and P2X7), and five P2YR subtypes (P2Y1, P2Y6, P2Y11, P2Y12, and P2Y14) (Supplemental Fig. 1). The ability of multiple selective PR agonists (adenosine, ATP, ADP, UTP, UDP, 2Me-S-ATP, and ATPγS) at physiological concentrations (1–100 μM) to induce degranulation in nonsensitized LAD2 cells was assessed. None of the P2YR agonists tested significantly induced degranulation (Fig. 1A). In sensitized cells, WAS-induced release was significantly larger versus the spontaneous release of the nonstimulated controls. Moreover, ATPγS markedly enhanced net WAS-induced degranulation in sensitized LAD2 cells (7-fold; n = 4; p < 0.01) (Fig. 1B). Even higher concentrations (1000 μM) of ATPγS alone did not induce degranulation in nonsensitized LAD2 cells (Fig. 1C), whereas the same high concentration of BzATP, a P2X7R agonist, which is known to activate many types of mast cells, induced degranulation in nonsensitized LAD2 cells (Fig. 1C). The potency of a high concentration (1 mM) of BzATP’s effect on degranulation was the same as that of the calcium ionophore (A23187, 2 μM), which is known to induce degranulation in LAD2 mast cells (31, 32). Because ATPγS is a known agonist at P2Y11R, the earlier data strongly suggested that activation of the P2Y11R was mediating the WAS degranulation-enhancing effects. Also, the lack of WAS degranulation-enhancing effects of UTP and ADP excluded their associated receptors, P2Y1R, P2Y6R, P2Y12R, and P2Y14R. In addition, P2Y11R is coupled to both G_q/11 and G_s proteins that are known to be linked to induction and suppression, respectively, of mast cell degranulation. Furthermore, the effect by ATPγS (100 μM) on WAS degranulation enhancement was suppressed in a concentration-dependent manner by NF157, a selective P2Y11R antagonist (Fig. 1D).

Strong allergic stimulation (see spike, Fig. 2A), but not WAS (no spike, Fig. 2B), induced intracellular Ca²⁺ mobilization in sensitized LAD2 cells. ATPγS (10–1000 μM) alone induced a moderate intracellular Ca²⁺ mobilization (spike, Fig. 2C). However, ATPγS did not positively or significantly affect ensuing WAS-induced Ca²⁺ mobilization (no spike, Fig. 2D), but markedly enhanced degranulation (Fig. 1B). Based on the results of degranulation assays (Fig. 1) and intracellular Ca²⁺ mobilization (Fig. 2) assays, as well as evidence for P2Y11R-mRNA (Supplemental Fig. 1) and its protein (Supplemental Fig. 2) expression we concluded that P2Y11R is responsible for the enhancement of WAS-induced degranulation.

The effects of inhibition of PI3K were examined in sensitized LAD2 cells. LAD2 cells were cotransfected with plasmids that contained either the shRNA constructs for PI3K(δ) mRNA K/D or contained eGFP alone (expression >85%) (Fig. 3A1, 3A2). PI3K expression was suppressed to less than 50% versus the cells without the PI3K(δ) mRNA K/D (Fig. 3A3–5). P2Y11R mRNA K/D by siRNA similarly inhibited P2Y11R protein expression (Supplemental Fig. 2). Both the PI3K(δ) and P2Y11R K/Ds virtually abolished the enhancing effects of ATPγS (100 μM) on WAS-induced degranulation (Fig. 3B). There was no significant difference in the effects of ATPγS on WAS-induced degranulation in the control shRNA-treated, control siRNA-treated, and nontreated LAD2 cells (data not shown). We also confirmed that the PI3K inhibitor compound 15e suppressed ATPγS’s effect on WAS-induced degranulation enhancement in a concentration-dependent manner (Fig. 3C).

The effects of ATPγS on the phosphorylation of key kinases related to intracellular PI3K(δ)’s activation cascades were determined. ATPγS (100 μM) enhanced the phosphorylation of PI3K(δ) and Akt in LAD2 cells (Fig. 4A). Comparison of signal levels in the images of PI3K(δ), p-PI3K(δ), PDK-1, p-PDK-1, Akt, and p-Akt are shown in (Fig. 4B.These data indicate that both PI3K(δ) and Akt were phosphorylated by ATPγS and are further upregulated by WAS, especially in the case of Akt. However, PDK-1, which is known to be a link between PI3K(δ) and Akt, was not phosphorylated.

LAD2 cells express ectonucleoside triphosphate dephosphorylase-1 (CD39) (Fig. 5A), but not ecto-5'-nucleotidase (CD73) (Fig. 5B), suggesting extracellular nucleotides could be degraded to AMP, but not to adenosine. Thus, administrated or endogenous ATP could be broken down to AMP (by CD39), but not to adenosine (absent CD73).

When added to the tissue culture medium of LAD2 cells during the experimental period (20–30 min), ATP (100 μM) was metabolized to a half of its initial content (Fig. 7).

We have previously shown that extracellular ATP is a powerful enhancer of low-level IgE-mediated degranulation in HLMCs (11, 17). The intracellular signal transduction mechanism underlying this effect has not been fully delineated. To our knowledge, this study is the first to examine the intracellular mechanisms of the enhancement of IgE-mediated degranulation induced by the activation of a P2YR in HMCs. The present data indicate that P2Y11R is linked to degranulation responses in LAD2 cells. This conclusion is based on the following: (1) the mRNA and protein expression of P2Y11R (Supplemental Figs. 1 and 2); (2) the lack of enhancing effect of purines and pyrimidines that target other P2YRs expressed by LAD2 cells; (3) the strong enhancing effects of ATPγS, which acts as an agonist at the P2Y11R sites (23, 33, 34); and (4) the marked inhibition of ATPγS’s effects by NF157, a selective P2Y11R antagonist (35, 36).

In agreement with Gao et al. (19), we report that P2Y11 shows the most robust expression of the P2YR in LAD2 cells. We also confirmed that P2R agonists, including ATPγS, do not by themselves trigger degranulation. In contrast with our studies where we examined the effects of P2YR agonists on WAS, Gao et al. studied the effects on maximal Ag-induced degranulation and proposed a role for P2Y14. This raises the possibility that different P2YRs may play distinct roles in varying degrees of Ag activation. Notably, however, their examination of ATPγS was limited to a single concentration of 10 μM. We required 100 μM to see the enhancement of WAS (Fig. 1).

Downregulation of FcεRI caused by the internalization of the IgE receptor is manifested by certain types of mast cell (37, 38). Accordingly, the possibility that the expression of P2Y11R in LAD2 cells could be changed should be considered. However, no change in the expression levels of P2Y11R mRNA or protein was observed after addition of ATPγS or Ag under the experimental conditions (data not shown).

P2Y11R-mediated effects of ATPγS involve the PI3K/Akt pathway, but not the induction of intracellular Ca²⁺ mobilization after WAS. This is in congruence with previous reports on PI3K and mast cell activation (39, 40). It is well known that increased intracellular levels of free Ca²⁺ plays a key role in mast cell activation leading to degranulation (4, 41). IgE receptor (FcεRI) activation is usually associated with intracellular Ca²⁺ mobilization. However, intracellular Ca²⁺-independent steps involving degranulation in mast cells have also been reported (42–45). We show here that the enhancement of WAS-induced degranulation mediated by P2Y11R activation can be independent of Ca²⁺ mobilization (Fig. 2); but, in contrast, it involves the PI3K/Akt pathway. This is supported by the dose-dependent inhibition of the ATPγS’s enhancing effects by compound 15e, a PI3Kδ inhibitor (46, 47), and marked attenuation of these effects in PI3Kδ K/D LAD2 cells. Although PDK-1 is known as one of the key elements of the PI3K/Akt pathway (48–50), the present results regarding phosphorylation of the intracellular signal proteins do not support involvement of PDK-1 in ATPγS enhancement of WAS-induced degranulation in LAD2 cells.

Aside from enhancement of allergic degranulation of mast cells (11, 17), multiple proinflammatory effects of ATP in the lungs have been described. We have shown that aerosolized ATP directly triggers a central pulmonary–pulmonary vagal reflex resulting in bronchoconstriction and cough (51). We have also suggested that ATP released from activated platelets plays a mechanistic role in syncope and bradycardia associated with pulmonary embolism (51, 52). Furthermore, these findings are in congruence with the expression of PRs in the lungs mediating inflammatory reactions (53, 54). In addition, AMP has been used as a selective bronchoconstrictor in the assessment of airway hypersensitivity in patients with asthma versus those without asthma (52, 55). Holgate and colleagues (56–58) have shown that aerosolized AMP-induced bronchoconstriction in human subjects is mediated by adenosine acting on lung mast cells triggering the release of histamine. However, under in vitro conditions, AMP can act as a nucleotide and not a nucleoside (i.e., adenosine). In this study, we have shown that LAD2 cells lack CD73 required to degrade AMP to adenosine (Fig. 5).

Accordingly, using primary cultured HLMCs from human pulmonary parenchymal tissues, we have previously shown that AMP had no effect on mast cell degranulation (11, 17). Thus, it seems that in vivo, ATP acts directly on HLMCs, but AMP’s action in vivo depends on its degradation to adenosine by ectoenzymes.

Physiological agonist/agonist receptor systems require a given nondegraded agonist present at its cognate receptor site. The sensitivity to purinergic agonists at PR sites is thus influenced by the expression and activity of ectonucleotidases on the cell surface (59, 60). It follows that the “right” purinergic agonist at the “right” purinergic cell surface receptor in the “right” permissive ectonucleotidases environment must “line up” in aggregate to result in effective WAS-induced enhancement mediated by P2Y11R’s activation in LAD2 cells. This also means that higher cell surface expression of ectonucleotidases (e.g., CD39) might reduce allergic exacerbations because of the enhanced degradation of ATP to AMP and the latter’s lack of effect in the case of HLMC degranulation (11, 61). As mentioned earlier, many reports deal with phenotypes and tissue of origin–dependent diversity of mast cells’ responses to extracellular nucleotides (9, 62).

We have previously reported that the half-life of extracellularly ATP/ADP (1 mM) was 14.88 min in another human cell line expressing ectonucleotidases CD39, CD73, and alkaline phosphatase (30). Recent studies have emphasized the role of ectonucleotidases in the magnitude of ATP’s effects in pulmonary and other organ disorders (63, 64). It seems that many physiological factors interact to either maintain or degrade extracellular ATP in vivo, depending on localized physiological conditions, such as pH. It is plausible that data obtained in vitro may be optimized to achieve therapeutic advantages in clinical settings.

Because ATPγS is a known stable agonist at P2Y11R, previously published data strongly suggested that P2Y11R mediated the observed WAS-enhancing effects. Also, the lack of WAS degranulation-enhancing effects of UTP and ADP excluded P2Y1R, P2Y6R, P2Y12R, and P2Y14R. Further supporting the role of P2Y11R is its coupling to both G_q/11 and G_s proteins (14). In this study, ATP is metabolized, but ATPγS is nonhydrolyzable (Fig. 6). However, the breakdown observed over nucleotide preincubation periods before Ag challenge (20 min) was only 50% (reduced from 100 to 50 μM); the latter concentration is still high enough for the enhancing effect of ATP. This observation may suggest that the breakdown product(s) of ATP (e.g., AMP, ADP) may exert opposing dose-dependent inhibitory effects on IgE-mediated activation of mast cells as we have previously shown for adenosine in HLMCs (11, 17).

P2Y11R is coupled to G_q/11 but also to G_s protein (14) so that stimulation of P2Y11R could increase intracellular cAMP levels via the G_s pathway. Accumulation of intracellular cAMP could activate protein kinase A or the Epac (exchange protein directly activated by cAMP) system via the activation of adenylyl cyclase (65). However, sustained elevations of cAMP are well known to suppress mast cell degranulation (66). There are several reports that Epac1/2 can link to intracellular Ca²⁺ mobilization (67), but Epac1/2 do not cause mast cell activation (68). Our results suggest that the effect of ATPγS on WAS-induced degranulation in LAD2 cells is not related to the cAMP/Epac pathways (Supplemental Fig. 3) but is mediated by the P2Y11R-G_q/11 signaling, including the PI3K system and not G_s protein. Thus, the enhancing effect of ATPγS on WAS-induced degranulation in sensitized LAD2 is probably not directly related to intracellular Ca²⁺ mobilization.

The results of this study indicate that P2Y11R is expressed in HMCs. Its activation is linked to marked enhancement of WAS-induced degranulation (Fig. 7). Induction of intracellular Ca²⁺ mobilization is not required for the P2Y11R-mediated effects on WAS-induced degranulation, which is linked to PI3K(δ) and Akt activation and not PDK-1. These results indicate that in the clinical setting given a favorable physiological milieu, even a weak allergic trigger can lead to development of much heightened expression of IgE-mediated reactions and worsening of allergic symptoms.

We thank Prof. Arnold Kirshenbaum for kindly supplying the LAD2 cells.

The authors have no financial conflicts of interest.