Introduction

Extracellular purine (ATP, ADP) and pyrimidine (UTP, UDP) nucleotides mediate their effects via P2 receptors (Ralevic and Burnstock, 1998). P2 receptors consist of ligand-gated intrinsic ion channels, P2X receptors, and G-protein-coupled P2Y receptors. Stimulation of P2X receptors (P2X1-P2X7) causes Ca2+ and Na+ influx from the extracellular space with accompanying plasma membrane depolarization (Khakh and North, 2006). The P2Y receptors are composed of eight subtypes in mammals: P2Y1, P2Y2, P2Y4, P2Y6, P2Y11, P2Y12, P2Y13, and P2Y14. The P2Y1, P2Y12, and P2Y13 receptors are preferentially activated by ADP whereas P2Y6 is activated by UDP (Abbracchio et al., 2006). The P2Y11 receptor prefers ATP as an agonist (Communi et al., 1997) whereas the P2Y2 (Lustig et al., 1993) and P2Y4 receptors are equally sensitive to ATP and UTP (Communi et al., 1995). The P2Y14 receptor is activated by the nucleotide sugar UDP-glucose (Brunschweiger and Muller, 2006). The P2Y1, P2Y2, P2Y4, P2Y6, and P2Y11 receptors are coupled to Gq, promoting PLC-catalyzed generation of an inositol phosphate (IP3) and subsequent release of intracellular calcium (Ralevic and Burnstock, 1998). The P2Y12, P2Y13, and P2Y14 receptors are coupled to Gi/o, inhibiting adenyl cyclase (Hollopeter et al., 2001).

Recent studies have suggested that activated immune cells secrete ATP into the extracellular space, to release cytokines by activation of P2 receptors (Di Virgilio and Solini, 2002). Extracellular ATP is known to be a potential mediator of inducible iNOS expression (Ohtani et al., 2000) and TNF-α release in rat microglia, via P2X7 receptors (Hide et al., 2000). The presence of the P2X7 receptor in microglia is consistent with involvement of this receptor in the inflammatory process [e.g., in the production of proinflammatory cytokines, nitric oxide, and reactive oxygen species (ROS)] and cytotoxicity (Di Virgilio et al., 1999; Parvathenani et al., 2003). In addition, ATP evokes the production of IL-6, a mediator of inflammation, in the microglial cell line MG-5 (Shigemoto-Mogami et al., 2001), and induces maturation and release of IL-1β by activating the IL-1β-converting enzyme/caspase1 (Sanz and Di Virgilio, 2000). Further, recent studies have reported that P2X4 receptors are involved in pain transmission (Tsuda et al., 2003) and that during microglial activation, P2Y12 receptors, which mediate the movement of microglial projections, are down-regulated (Haynes et al., 2006), but P2Y6 receptors are up-regulated and trigger phagocytosis (Koizumi et al., 2007). Together, these results suggest that P2 receptors play a critical role in brain inflammation by regulating the production of inflammatory mediators and microglial activation.

Our previous study demonstrated that ATP released from LPS-activated microglia, and metabolites of ATP, induced IL-10 expression in a process involving P2 receptors, in an autocrine fashion, indicating that P2 receptors also have a function that counteracts the effects of proinflammatory mediators by participating in the production of the anti-inflammatory cytokine IL-10 (Seo et al., 2004). Therefore, we sought to determine which subtype of P2 receptors is responsible for the modulation of IL-10 expression in ATP-stimulated microglia.

Materials and Methods

Microglial cell culture

Microglial cultures were prepared from the brains of 3 day-old Sprague-Dawley rats as described previously (Kim et al., 2002). Briefly, whole brains were dissected into small cubes, incubated in D-PBS (JBI, Daegu, Korea) containing 0.1% trypsin and 40 µg/ml DNase I for 15 min at 37℃, and dissociated into single cells by gentle pipetting. Dissociated cells were suspended in DMEM (JBI) containing 5% horse serum, 5 mg/ml glucose, 100 U/ml penicillin and 100 µg/ml streptomycin, and plated on poly-D-lysine-coated T-75 culture flasks, and incubated at 37℃ in incubator with 5% CO2/95% air atmosphere. After 2-4 weeks of growth in flasks, microglia floating in the medium were collected and grown in separate 6- or 96-well plates.

RT-PCR analysis

To determine which subtypes of P2X and P2Y receptors are expressed by rat microglia, microglial cells (1 × 106 cells/well) were plated into 6-well plates in DMEM (JBI) containing 5% horse serum (JBI). Total RNA was extracted using Trizol (Life Technologies, Rockville, MD). Total RNA from each sample was subjected to DNase I treatment and then processed for the first strand cDNA synthesis using Moloney murine leukemia virus (M-MLV) reverse transcriptase (Life Technologies). 5 µl of each cDNA products was amplified by PCR using specific sense and antisense primers designed from the cDNA sequences for P2 receptors: P2X1, 5'-AGAGGCACTACTACAAGCAGAA-3' (sense) and 5'-GGTAAGGCTGTGGGAAAGA-3' (antisense); P2X2, 5'-GAATCAGAGTGCAACCCCAA-3' (sense) and 5'-TCACAGGCCATCTACTTGAG-3' (antisense); P2X3, 5'-TTAAGTTTGCTGGACAGGAT-3' (sense) and 5'-GTTCCCATATACCAGCACAT-3' (antisense); P2X4, 5'-TCCCTTCTGCCCCATATTCC-3' (sense) and 5'-TTCATCTCCCCCGAAAGACC-3' (antisense); P2X5, 5'-CGACCTGGTACTTATCTACCTC-3' (sense) and 5'-ACGTTCACAATGGCATTC-3' (antisense); P2X6, 5'-GCCCAGAGCATCCTTCTGTTC-3' (sense) and 5'-CGTGGCTGTATGTCCCCATC-3' (antisense); P2X7, 5'-AACAGTGCCATTCTGACC-3' (sense) and 5'-GCCACCTCTGTAAAGTTCTC-3' (antisense); P2Y1, 5'-GGCAGGCTCAAGAAGAAGAAT-3' (sense) and 5'-ATCACACATTTCTGGGGTCTG-3' (antisense); P2Y2, 5'-AGCTCTTTAGCCATTTTGTG-3' (sense) and 5'-CGGAAGGAGTAATAGAGGGT-3' (antisense); P2Y4, 5'-ACTGTCTTTGCTGTCTGCTT-3' (sense) and 5'-AGACAGCTATTAGCACTGGC-3' (antisense); P2Y6, 5'-GTCTACCGTGAGGATTTCAA-3' (sense) and 5'-CTAGGTATCGCTGGAAGCTA-3' (antisense); P2Y12, 5'-CTCCACCACCTACATGTTTC-3' (sense) and 5'-AAGAGGATGCTGCAGTAGAG-3' (antisense); P2Y13, 5'-TGCACTTTCTCATCCGTGGT-3' (sense) and 5'-GGCAGGGAGATGAGGAACAT-3' (antisense)

Measurement of IL-10

To evaluate the effects of ATP on IL-10 production, microglial cells (3 × 104 cells/well) were plated into 96-well plates in DMEM containing 5% horse serum. The amount of IL-10 in the supernatant was measured by ELISA. To assess which purinergic receptor was involved in the microglial IL-10 production, microglia cells were treated with P2 receptor agonists: ATP, ADP, adenosine 5'-O-(3-thiotriphosphate) (ATP-γ-S), adenosine 5'-O-(2-thiodiphosphate) (ADP-β-S), 2-methylthio-ATP (2-meSATP), 2-methylthio-ADP (2-meSADP), α,β-methylene ATP (α,β-meATP), 2',3'-(benzoyl-4-benzoyl)-ATP (BzATP), UTP, UDP, or dATP (Sigma, St. Louis, MO); P2 receptor antagonists: trinitrophenyl-substituted ATP (TNP-ATP), adenosine 5'-triphosphate 2',3'-acylic dialcohol (oxidized ATP; oATP), 2'-deoxy-N6-methyladenosine-3',5'-bisphosphate (MRS2179), 2-methylthioadenisine 5'-monophosphate (2-meSAMP), 5'-O-thiomnophosphate (5'-AMPS) (Sigma, St. Louis, MO); Ca2+ chelators: EGTA (Sigma, St. Louis, MO), bisaminophenoxyethane tetraacetic acid-acetoxymethyl ester (BAPTA-AM; Calbiochem, San Diego, CA); IP3 inhibitor, Xes-C (Sigma, St. Louis, MO); adenylate cyclase inhibitor, SQ22536 (Sigma, St. Louis, MO); PKA inhibitor, H-89 (Sigma, St. Louis, MO); Gi protein inhibitor, pertussis toxin (PTX; Sigma, St. Louis, MO). For IL-10 assay, we used Cytosets kit for rat IL-10 (Biosource, Camarillo, CA) according to the manufacturer's protocol.

Statistical analysis

All statistical comparisons in this study were performed using one-way ANOVA with the Tukey-Kramer multiple comparisons test, and data are expressed as mean ± SEM of triplicate samples. A value of P < 0.05 was considered statistically significant.

Results

Characterization of ATP-(or ATP-γS)-induced IL-10 release, and ADP-(or ADP-βS)-induced IL-10 release

To characterize IL-10 expression by ATP-stimulated microglia, microglial cells were treated with various concentrations (1, 10, 100, 1,000 µM) of ATP, ADP, ATP-γS (a hydrolysis-resistant analog of ATP), or ADP-βS (a hydrolysis-resistant analog of ADP). We found that the patterns of IL-10 production were dose-dependent and bell-shaped (Figure 1). Interestingly, the concentrations of ATP and ATP-γS that showed maximal IL-10 release were different. ATP-induced IL-10 release peaked at an ATP concentration of 100 µM (811.51 ± 29.59 pg/ml IL-10) and was sustained to 1,000 µM (750.15 ± 5.66 pg/ml). On the other hand, ATP-γS-induced IL-10 release peaked at an ATP-γS concentration of 10 µM (930.65 ± 30.94 pg/ml) but dropped to 480.88 ± 18.52 pg/ml at 100 µM (P < 0.01). In the case of ADP, ADP-induced or ADP-βS-induced IL-10 release peaked at a concentration of 100 µM, but treatment with 1,000 µM ADP (IL-10 release of 186.27 ± 20.70 pg/ml) or 1,000 µM ADP-βS (IL-10 release of 475.10 ± 30.96 pg/ml) appeared to induce less IL-10 release than did treatment with 100 µM ADP (485.26 ± 20.33 pg/ml) or 100 µM ADP-βS (721.43 ± 35.20 pg/ml) (P < 0.01). These results indicate that 100 µM ATP-γS or 1,000 µM ADP-βS inhibit IL-10 production by affecting distinct subtypes of the P2 receptor involved in IL-10 expression. We found no loss of cell viability in the presence of either 100 µM ATP-γS or 1,000 µM ADP-βS (data not shown).

Figure 1
figure 1

Characterization of ATP (or ATP-γS)- or ADP (or ADP-βS)-induced IL-10 release. Microglial cells (3 × 104 cells/well) were treated with (A) ATP or ATP-γ-S, (B) ADP or ADP-β-S at the indicated concentrations. The amount of IL-10 was measured 24 h after treatment with ATP (ATP-γ-S) or ADP (ADP-β-S). Data shown are mean ± SEM of triplicate samples. The figure shows one representative of five independent experiments. *P < 0.01 compared to 10 µM ATP-γS or 100 µM ADP-βS.

Full size image

Expression of mRNAs encoding the P2X and P2Y receptors

To determine which subtypes of P2X and P2Y receptors are expressed by rat microglia, mRNA was isolated from microglial cells and analyzed by RT-PCR. Amplified PCR products of the expected sizes were obtained for P2X1 (434 bp), P2X3 (272 bp), P2X4 (489 bp), and P2X7 (358 bp) receptor mRNAs (Figure 2A). Similarly, amplified PCR products of the expected sizes were obtained for the P2Y1 (411 bp), P2Y2 (244 bp), P2Y4 (149 bp), P2Y6 (325 bp), P2Y12 (168 bp), and P2Y13 (185 bp) receptor mRNAs from microglial cell total mRNA (Figure 2B). A recent study reported that microglia express various receptors for ATP including both P2X receptors (P2X3, P2X4, P2X5, P2X7) and P2Y receptors (P2Y1, P2Y2, P2Y4, P2Y6, P2Y12, P2Y13) (Light et al., 2006). At this point, we could not test the expression of P2Y11 receptor, because rat P2Y11 receptor has not been cloned.

Figure 2
figure 2

Expression of P2X and P2Y receptors mRNA. RT-PCR analysis of P2 receptor mRNA expression in microglial cells was done with primers specific for distinct P2 receptors subtypes. cDNA products were analyzed by 1.5% agarose gel electrophoresis. A representive gel with ethidium bromide-stained cDNA fragments of the P2 receptors.

Full size image

Effects of P2 receptor agonists on the release of IL-10 from microglial cells

We next examined the effects of various concentrations (1, 10, 100, 300, 1,000 µM) of agonists (2-meSATP, 2-meSADP, α,β-meATP, BzATP, UTP, UDP, dATP) of the microglia-expressed P2 receptors (identified by RT-PCR) on the release of IL-10 from microglia. The agonists 2-meSADP, BzATP, and dATP increased the release of IL-10 in a dose-dependent manner, but 2-meSATP, α,β-meATP, UTP, and UDP did not (Figure 3). 2-meSADP and BzATP showed the bell-shaped paradigm which has maximal IL-10 release at 100µM. On the other hand, dATP-induced IL-10 release was sustained as agonist concentrations rose to 1 mM. The rank order of agonist potency was BzATP = dATP > 2-meSADP. Because 2-meSADP is an agonist for both P2Y1 and P2Y12 receptors, and both P2X7 agonist BzATP and dATP are potent P2Y11 receptor agonists, these results indicate that purinergic modulation of IL-10 expression may be mediated by P2X7, P2Y1, P2Y11, and P2Y12 receptors.

Figure 3
figure 3

Effects of the P2 receptor agonist on the release of IL-10 from microglial cells. Microglial cells (3 × 104 cells/well) were treated with 2-meSATP, 2-meSADP, α,β-meATP, BzATP, UTP, UDP, and dATP at the indicated concentrations. The amount of IL-10 was measured 24 h after treatment. Data shown are mean ± SEM of triplicate samples. The figure shows one representative of five independent experiments. *P < 0.01 compared to control.

Full size image

Effects of P2 receptor antagonists on the release of IL-10 from microglial cells

To determine which subtype of P2 receptor was responsible for the modulation of IL-10 expression, we then examined the effects of the specific antagonist of the P2X7, P2Y1, P2Y11, or P2Y12 receptors, and an antagonist of the P2X1, P2X3, and P2X4 receptors. ATP-γS (100 µM)-induced IL-10 release was restored by TNP-ATP (an antagonist of the P2X1, P2X3, and P2X4 receptors) (Figure 4A). 2-meSADP (100 µM)-induced IL-10 release was inhibited by MRS2179 (a P2Y1 antagonist), but 2meSAMP (a P2Y12 antagonist) up-regulated 2-meSADP (100 µM)-induced IL-10 release (Figure 4B). On the other hand, BzATP (300 µM)-induced IL-10 release was not restored by oATP (a P2X7 antagonist) (Figure 4C). BzATP (100 µM)-induced IL-10 release was inhibited by 5'-AMPS (a P2Y11 antagonist) (Figure 4D). These results indicate that both the P2Y1 and P2Y11 receptors are major receptors involved in IL-10 expression, and that the P2X1, P2X3, P2X4 receptor group, or the P2Y12 receptor, negatively modulates the P2Y11 receptor or the P2Y1 receptor, respectively.

Figure 4
figure 4

Effects of the P2 receptor antagonist on the release of IL-10 from microglial cells. Microglial cells (3 × 104 cells/well) were pretreated with TNP-ATP, MRS2179, 2-meSAMP, oATP, or 5'-AMPS for 30 min at the indicated concentrations, then treated with ATP-γ-S (100 µM), 2-meSADP (100 µM), and BzATP (100 or 300 µM). The amount of IL-10 was measured 24 h after treatment. Data shown are mean ± SEM of triplicate samples. The figure shows one representative of five independent experiments. *P < 0.01 compared to the agonist alone.

Full size image

Effects of Ca2+ chelators and the inhibitors of IP3, adenylate cyclase, and PKA on IL-10 release from microglial cells

We investigated the effects of BAPTA-AM (a membrane permeable Ca2+ chelator) and Xes-C (an IP3 inhibitor) on the IL-10 production. As expected, ATP-γS (100 µM), BzATP (300 µM), dATP (1 mM), and 2-meSADP (100 µM)-induced IL-10 expression was down-regulated with BAPTA-AM or Xes-C treatment, indicating that intracellular Ca2+ release contributes the IL-10 production. However, IL-10 production by ATP-γS (100 µM) or BzATP (300 µM) was not affected when the cells were preincubated in EGTA-containing Ca2+-free medium, indicating that Ca2+ influx through P2X1, P2X3, P2X4, P2X7 receptor group did not inhibit the IL-10 production (Figure 5A). As can be seen in Figure 5B, ATP-γS (10 µM), BzATP (100 µM), and dATP (1 mM)-induced IL-10 production was down-regulated by SQ22536 (an adenylate cyclase inhibitor) or H-89 (a PKA inhibitor), and 2-meSADP (100 µM)-induced IL-10 production was more up-regulated by pertussis toxin (PTX; a Gi protein inhibitor). These results suggest that intracellular Ca2+ release and/or cAMP-activated PKA are the main contributors to extracellular ATP-(or ADP)-mediated IL-10 expression.

Figure 5
figure 5

Effects of the calcium chelators and inhibitor of IP3, adenylate cyclase, or PKA on IL-10 release from microglial cells. Microglial cells (3 × 104cells/well) were pre-treated with EGTA (500 µM, 20 min), BAPTA-AM (50 µM, 1 h), Xes-C (100 nM, 20 min), PTX (500 ng/ml, 1 h), SQ22536 (10 µM, 20min), or H-89 (10 µM, 20 min), then treated with ATP-γ-S, BzATP, 2-meSADP, and dATP at the indicated concentrations. The amount of IL-10 was measured 24 h after treatment. Data shown are mean ± SEM of triplicate samples. The figure shows one representative of three independent experiments. *P < 0.01 compared to agonist alone.

Full size image

Discussion

In this study, we presented evidences for crosstalk between P2 receptors in a situation where P2Y1 and P2Y11 receptors are major receptors involved in extracellular ATP-(or ADP)-mediated IL-10 production, and P2X1, P2X3, P2X4 receptor group, or the P2Y12 receptor, negatively modulate the P2Y11 receptor or the P2Y1 receptor, respectively.

The P2Y1 and P2Y11 receptors are coupled to Gq, promoting PLC-catalyzed generation of an IP3, and subsequent release of intracellular calcium (Ralevic and Burnstock, 1998). Therefore, we investigated the effects of BAPTA-AM and Xes-C (an IP3 inhibitor) on IL-10 production. As expected, the IL-10 expression induced by ATP-γS (100 µM), BzATP (300 µM), dATP (1 mM), or 2-meSADP (100 µM) was down-regulated by BAPTA-AM or Xes-C treatment, indicating that intracellular Ca2+ release contributes to IL-10 production (Figure 5A). In accord with these results, previous study demonstrated that Ca2+ and calmodulin (CaM) stimulated adenylate cyclase (Simonds, 1999), and that prostaglandin E2 (PGE2), which was shown to accumulate cAMP in microglia, enhanced the LPS-induced IL-10 expression by microglia (Patrizio et al., 1996; Aloisi et al., 1999).

Previous work has shown that agonists for the P2Y11 receptor are ATP-γS, BzATP, ATP, and ADP-βS (Communi et al., 1999). Indeed, exposure to these nucleotides increased the release of IL-10 (Figures 1 and 3), and IL-10 release mediated by ATP-γS (10 µM), BzATP (100 µM), and dATP (1 mM) was down-regulated by an adenylate cyclase inhibitor, SQ22536, and a PKA inhibitor, H-89 (Figure 5B). These results are consistent with previous findings that P2Y11 receptors activate adenylate cyclase and contribute to cAMP formation (Torres et al., 2002). Recent studies showed that ATP-γS enhanced LPS-induced IL-10 production in human monocyte-derived dendritic cells (Marteau et al., 2004), and that ATP-γS and BzATP inhibited the production of TNF-α, IL-8, and MIP-1β in human mast cells through a Gs-coupled receptor (Feng et al., 2004). At this point, we can not rule out the involvement of other P2Y11 receptor subtypes in our study, as BzATP (100 µM)-induced IL-10 release was only partially inhibited by 5'-AMPS (a P2Y11 antagonist) (Figure 4D), and P2Y11 receptor-transcripts have not been found in rats and mice (von Kügelgen, 2006). To date, the P2Y11 receptor is the only cloned ATP-binding P2Y family member known to stimulate Gs proteins (Communi et al., 1999).

Data from work with agonists indicated that the P2X receptors did not contribute to IL-10 production (Figure 3), but TNP-ATP (an antagonist of the P2X1, P2X3, and P2X4 receptors) up-regulated ATP-γS (100 µM)-induced IL-10 production (Figure 4A). These results suggest that there is reciprocal crosstalk between the P2Y11 and P2X receptors, through which plasma membrane depolarization by Na+ and Ca2+ entry from the extracellular space negatively modulates IL-10 expression. Previous study showed that depolarization induced by P2X receptor-mediated Na+ influx inhibited store-operated channels (SOC)-mediated Ca2+ entry resulting from P2Y activation (Wang et al., 2000). Therefore, down-regulation of IL-10 production by 100 µM ATP-γS may be mediated by P2X1, P2X3, and P2X4 receptors-mediated inhibition of Ca2+ entry through SOC.

Unlike the Gq-coupled P2Y1 receptor, the ADP-selective P2Y12 and P2Y13 receptors both use Gi proteins to inhibit adenylate cyclase (Communi et al., 2001). In accord with these data, our results showed that 100 µM 2-meSADP-induced IL-10 production was up-regulated by 2meSAMP (a P2Y12 antagonist) and PTX (a P2Y12 receptor inhibitor), but down-regulated by MRS2179 (a P2Y1 antagonist) (Figure 4B and 5B), indicating reciprocal crosstalk between P2Y1 and P2Y12 receptors. Crosstalk between the downstream pathways of P2Y1 and P2Y12 receptors was demonstrated in the regulation of platelet aggregation and astrocytes cell death (Hardy et al., 2004; Mamedova et al., 2006).

In conclusion, our study provides evidence that extracellular ATP-(or ADP)-mediated IL-10 production in microglia is mediated via P2Y1 and P2Y11. Further, IL-10 production is modulated by crosstalk between P2Y1 and P2Y12 receptors, and P2Y11 and P2X receptors.