Extracellular ATP acts as a signaling molecule exerting effects on a range of biological functions including immune regulation, apoptosis, cellular proliferation, and neurotransmission (Khakh and North, 2006). P2X receptors are cationic channels gated by extracellular ATP, of which seven subtypes have been identified that assemble as either homo- or heterotrimeric receptors (North, 2002; Barrera et al., 2005; Nicke et al., 2005; Ormond et al., 2006). Heteromerization can change both the functional and pharmacological properties of P2X receptors (Lewis et al., 1995; King et al., 2000). One member of this family, P2X₇ receptor, is believed to be unique among P2X receptors in only forming homomeric assemblies (Torres et al., 1999).

Activation of P2X₇ receptors, which are expressed in a range of immune cells, can result in the release of IL-1β, IL-18, tumor necrosis factor -α, and matrix metalloproteinase 9, activation of the stress-activated protein kinase/c-Jun NH₂-terminal kinase pathway, membrane blebbing, and apoptotic or necrotic cell death (Humphreys et al., 2000; Perregaux et al., 2000; Wilson et al., 2002; Gu and Wiley, 2006). This receptor represents an important target in inflammatory diseases such as arthritis, neuropathic pain, and stroke (Dell'Antonio et al., 2002; Labasi et al., 2002; Chessell et al., 2005). In immune cells such as macrophages, monocytes, and microglia, P2X₇ receptors are coexpressed with another member of the P2X family also important in neuropathic pain, the P2X₄ receptor (Bowler et al., 2003; Xiang and Burnstock, 2005). The role of this receptor in immune cells is not as well understood; however, it has considerably higher affinity for ATP than the P2X₇ receptor, and its up-regulation in spinal cord microglia as a result of peripheral nerve injury contributes to allodynia type hypersensitivity (Tsuda et al., 2003).

Coexpression of P2X₄ and P2X₇ is not restricted to immune cells. They are also present together in endothelial and epithelial cells, and a recent electrophysiological study of P2X receptors that are present in ciliated airway epithelia and are believed to be important for mucociliary clearance reported currents with a novel combination of both P2X₇ and P2X₄ receptor characteristics (Ma et al., 2006). By comparing amino acid sequences, it was found that P2X₄ is more homologous to P2X₇ (∼40%) than are the other P2X receptor subtypes; however, a previous coimmunoprecipitation study with heterologously expressed receptors failed to provide evidence for the formation of stable P2X_4/7 complexes (Torres et al., 1999).

To understand how receptors transmit an ATP signal, we need to define the subunit identity of the physiological receptors. The formation of a P2X_4/7 heteromer could provide an important mechanism for the modulation of P2X₇ receptor signaling and so has important consequences for P2X₇ as a therapeutic target and for its physiological roles in a range of diseases and immune cell function. We sought to evaluate a molecular basis for its existence using biochemical, functional, and pharmacological methods.

Materials and Methods

Antibodies and Reagents. The following primary antibodies were used: rabbit polyclonal anti-P2X₂ subunit (0.6 μg/ml; Alomone Labs, Jerusalem, Israel), rabbit polyclonal anti-P2X₄ subunit (6 μg/ml; Alomone Labs), rabbit polyclonal anti-P2X₇ subunit (1.5 μg/ml; Alomone Labs), mouse monoclonal anti-hemagglutinin (HA) (0.8 μg/ml; Roche, Indianapolis, IN), rabbit polyclonal anti-EE (1 μg/ml; Bethyl Laboratories, Cambridge, UK), anti-LAMP-1 (0.8 μg/ml; Santa Cruz Biotechnology, Santa Cruz, CA). FITC- or Cy3-conjugated goat anti-mouse or anti-rabbit IgG secondary antibodies (1:250; Jackson ImmunoResearch Laboratories, Inc., West Grove, PA) were used for immunofluorescence. Horseradish peroxidase-conjugated anti-mouse or anti-rabbit secondary antibodies (1:10,000; GE Healthcare, Chalfont St. Giles, Buckinghamshire, UK, and Perbio Science, Cramlington, UK) or Rabbit TrueBlot-Horseradish Peroxidase (HRP) anti-rabbit IgG (eBioscience, San Diego, CA) were used for Western blotting. Complete Protease Inhibitor cocktail (Roche), n-dodecyl-b-d-maltoside (DDM; Melford Laboratories, Ipswich, Suffolk, UK), anti-rabbit IgG beads (eBioscience), and BCA Protein assay kit (Pierce, Rockford, IL) were used for coimmunoprecipitation experiments. Sulfo-succinimidyl 2-(biotinamido)-ethyl-1,3′-dithiopropionate was from Pierce. Unless otherwise stated, all others reagents were obtained from Sigma (St. Louis, MO) or Invitrogen (Carlsbad, CA).

DNA Constructs. The construction and characterization of P2X₄ and P2X₂ receptors with enhanced green fluorescent protein fused to the C terminus (P2X₄-EGFP, P2X₂-EGFP) has been described previously (Bobanovic et al., 2002). In brief, to generate cDNA encoding P2X₄ with EGFP fused to the C terminus, the rat cDNA (a kind gift from Dr. P. P. A. Humphrey) was amplified by polymerase chain reaction using oligonucleotide primers to introduce a Kozak initiation sequence (Kozak, 1987) to remove the stop codon and to introduce NheI and SacII sites at the 5′ and 3′ ends, respectively. Amplification products were then cloned into the pEGFP-N1 vector (BD Biosciences Clontech, Palo Alto, CA). Other constructs used include wild-type P2X₄ and P2X₄ with a HA tag at the C terminus. These sequences were subcloned into the pEGFP-N1 vector so that the enhanced green fluorescent protein (EGFP) sequence was excised. The C353W and S341W point mutations were made using the QuikChange II site-directed mutagenesis kit (Stratagene, La Jolla, CA) in P2X₄-EGFP. The sequences of all amplified regions were verified using automated DNA sequencing (Department of Genetics, University of Cambridge, Cambridge, UK). Other constructs used include wild-type rat P2X₇ and P2X₂, P2X₂ tagged at its N terminus with HA, pEGFP-N1, and DsRed-ER (BD Biosciences Clontech). Glu-Glu (EE)-tagged P2X₇ in pcDNA3 was a kind gift from A. Surprenant.

Cell Culture and Transfection. Normal rat kidney (NRK) cells and human embryonic kidney (HEK) 293 cells were maintained in Dulbecco's modified Eagle's medium containing 10% fetal bovine serum and 100 U/ml penicillin/streptomycin at 37°C and 5% CO₂. Transient transfections of NRK cells were carried out using Lipofectamine 2000 (Invitrogen) according to the manufacturer's instructions. For transfection of one well of a 12-well plate, 1 μg of plasmid DNA was used. Transient transfections of HEK293 cells were carried out using the modified calcium phosphate method as described previously (Bobanovic et al., 2002). The amount of DNA used to form a precipitate was 3 μg (in 100 μl of CaCl₂/100 μl of 2× HBS), and this was added to cells (200 μl/well) for 6 h. For cotransfection experiments, equal amounts of DNA were used, and we also included 0.5 μg of pEGFP-N1 vector for coexpression of EGFP with nonfluorescent constructs.

Bone marrow-derived macrophages (BMDMs) were obtained from 5- to 6-week-old male CD-1 mice. Mice were killed, the femur was excised, and the epiphyses were removed before flushing out the bone marrow. Cells were washed and resuspended in RPMI 1640 media supplemented with 20% fetal calf serum, 100 U/ml penicillin/streptomycin, and 30% L929 cell-conditioned media. Cells were cultured for 7 days before use and were treated for 48 h with lipopolysaccharide solution (1 μg/ml) before membrane protein fraction isolation.

Cell Biology and Immunofluorescence Protocols. Both NRK and HEK293 cells were plated onto poly(d-lysine)-treated coverslips. All cells were used 24 h after transfection. The basic protocol for total staining of the receptors was as follows. Cells were fixed in 3% paraformaldehyde (PFA) and 4% sucrose in PBS (1.5 mM NaH₂PO₄, 8 mM Na₂HPO₄, and 145 mM NaCl, pH 7.3) for 10 min at 4°C. If required, permeabilization was done using 0.1% Triton X-100 in PBS for 10 min at 4°C. Nonspecific sites were blocked using PBS containing 4% normal goat serum and 3% bovine serum albumin (blocking buffer). Antibodies were diluted to their final concentration in blocking solution. Primary antibodies were applied for 2 h at room temperature. Cells were rinsed once in blocking buffer and three times for 5 min with PBS, and then secondary antibodies were applied for 2 h at room temperature. Finally, cells were washed five times for 5 min with PBS and mounted onto slides with Vectashield (Vector Laboratories, Burlingame, CA) as a mounting medium. In some experiments, 100% methanol was used for 10 min at -20°C to fix and permeabilize the cells.

Image Analysis. Fluorescence was visualized using a Zeiss Axiovert LSM510 confocal microscope using a 63× oil immersion objective (Carl Zeiss Inc., Thornwood, NY). For FITC-Cy3 anti-FLAG double labeling, FITC and Cy3 were excited at 7 and 60% of 488 and 543 laser power, respectively. For each experiment, images were collected using identical acquisition parameters and analyzed using Image J. Pixel values were on an 8-bit scale (2⁸ = 256; 0–255).

Biotinylation. Cells were washed with ice-cold PBS and incubated with 1 mg/ml sulfo-succinimidyl 2-(biotinamido)-ethyl-1,3′-dithiopropionate for 20 min at 4°C. Excess biotin was quenched with PBS containing 50 mM glycine. Cells were solubilized with lysis buffer (20 mM Tris, pH 8.0, 150 mM NaCl, 1 mM EDTA, 2 mM EGTA, 1% NP-40, 1 mM phenylmethylsulfonyl fluoride, 50 mM NaF, 1 mM sodium orthophosphate, 20 mM β-glycerophosphate, 10 mM sodium pyrophosphate, and protease inhibitors) and incubated on ice for 30 min, after which time they were sonicated and cleared by centrifugation. The majority of the supernatant was incubated with immobilized NeutrAvidin biotin binding protein beads (Pierce) on a rotating rack for 2 h at 4°C to precipitate biotinylated proteins. The remaining supernatant was kept to assess total protein in each sample. Beads containing precipitated biotinylated proteins were spun for 1 min at 10,000 rpm at 4°C and washed at least three times. The protein was eluted from the beads by incubation in 20 μl of the Laemmli buffer. Proteins were separated by SDS-PAGE by loading on 7.5% polyacrylamide gels and detected by immunoblotting. The P2X₄ receptor was detected using a rabbit polyclonal anti-P2X₄ antibody (1:500). The P2X₇ receptor was detected using a rabbit polyclonal anti-P2X₇ antibody (1:100). Immunoreactive bands were visualized using appropriate horseradish peroxidase-conjugated secondary antibodies followed by enhanced chemiluminescence detection. All blots shown in figures are typical of at least two and, in most cases, four similar results.

Membrane Protein Fraction Isolation. To obtain total membrane protein fractions for immunoprecipitation assays, transfected HEK293 cells or BMDMs were washed three times with HBS-EDTA buffer (50 mM HEPES pH 7.4, 100 mM NaCl, and 2 mM EDTA), scraped off, and then collected by gentle centrifugation. The cell pellet was resuspended in ice-cold hypotonic buffer (10 mM Tris-HCl pH 7.0, 2 mM EDTA, 1 mM phenylmethylsulfonyl fluoride, and Protease Inhibitors cocktail) and incubated for 20 min on ice. Cells were mechanically disrupted by passing the solution through a needle, and then the extract was centrifuged at high speed (14,000 rpm, 15 min). The pellet containing the membrane-derived protein fraction was solubilized using 1% DDM in HBS buffer for 1 h on ice, and the solution was ultracentrifuged at 30,000 rpm for 1 h. The membrane protein fraction was collected from the supernatant and subjected to the BCA protein assay.

• Download figure
• Open in new tab
• Download powerpoint

Fig. 1.

Coexpression of P2X₇ and P2X₄ increased P2X₄ at the plasma membrane. A, confocal images of HEK293 cells transfected with either P2X₄ or P2X₇. P2X₄ receptors (green) were colocalized with LAMP-1 (red), and P2X₇ receptors (green) were colocalized with an ER marker, DsRed-ER. Cells were fixed with 3% PFA and permeabilized with 0.1% Triton X-100 in PBS for 10 min at 4°C. B, NRK cells transfected with either P2X₇ or P2X₄ alone (top) or cotransfected with P2X₇ and P2X₄-EGFP (bottom) were fixed with 100% methanol. Methanol fixation increased the detection of P2X₇ at the plasma membrane. In cotransfected cells, there was overlap between P2X₄ and P2X₇ at the plasma membrane. Scale bars, 10 μm. C, transfected NRK cells were incubated with biotin (1 mg/ml) for 20 min at 4°C to label surface proteins and were then solubilized, and surface proteins were precipitated with streptavidin beads. Surface expression of P2X₄ increased in the presence of P2X₇ (n = 4), although totals show equivalent P2X₄ expression. D, surface expression of P2X₇ was unchanged in the presence of P2X₄; in both cases, the proportion biotinylated was ∼10% of the total P2X₇. The first lane shows no detectable expression of P2X₇ in untransfected cells.

Immunoprecipitation. Total membrane protein extracts were preabsorbed with anti-rabbit IgG-beads for 30 min. The precleared membrane protein extracts were then incubated with 2 to 5 μg of anti-P2X₄ or anti-P2X₇ antibody in HBS buffer containing 1% DDM for 2 h at 4°C. Anti-rabbit IgG beads were subsequently added to the samples, and the mixture was further incubated for 1 h at 4°C. The protein beads complexes were washed four times with HBS buffer containing 1% DDM, and proteins were eluted by boiling for 5 min in 40 μl of Laemmli buffer. Samples were analyzed by SDS-PAGE and were probed by Western blot using the corresponding primary antibody and Rabbit TrueBlot-HRP anti-rabbit IgG as secondary antibody. When the immunoprecipitation was performed using a monoclonal anti-HA antibody, protein G beads were used instead to isolate the complexes, and an HRP-anti-mouse secondary antibody was used for detection in Western blots.

Electrophysiological Recordings. Standard whole-cell recordings were performed at room temperature using an Axopatch 200A amplifier (Molecular Devices, Sunnyvale, CA). Patch pipettes (3–6 MΩ) were pulled from thick-walled borosilicate glass (GC150F-10; Harvard Apparatus, Inc., Holliston, MA). ATP-induced responses were measured at -30 mV, and different extracellular (EC) solutions were used. These included normal Na⁺ EC solution (140 mM NaCl, 5 mM KCl, 2 mM CaCl₂, 1 mM MgCl₂, 10 mM d-glucose, and 10 mM HEPES, pH 7.3); Na⁺ EC solution with low divalent (151 mM NaCl, 0.3 mM CaCl₂, 10 mM HEPES, and 10 mM glucose, pH 7.3); Cs⁺ EC solution with low divalent (151 mM CsCl, 0.3 mM CaCl₂, 10 mM HEPES, and 10 mM glucose, pH 7.3); Cs⁺ EC solution with 1.58 mM Ca²⁺ (151 mM CsCl, 1.58 mM CaCl₂, 10 mM HEPES, and 10 mM glucose, pH 7.3); Cs⁺ plus Na⁺ EC solution with low divalent (121 mM CsCl, 30 mM NaCl, 0.3 mM CaCl₂, 10 mM HEPES and 10 mM glucose, pH 7.3). When using Na⁺ EC solution, the intracellular solution had the following composition (70 mM K₂SO₄, 10 mM KCl, 1 mM MgCl₂, 10 mM HEPES, and 75 mM sucrose, pH 7.3). When using a Cs⁺ EC solution, the composition of the intracellular solution was as follows (151 mM CsCl, 10 mM HEPES, and 0.1 mM EGTA, pH 7.3).

Whole-cell currents were low pass-filtered at 2 kHz and digitized at 10 kHz. Agonists were applied locally using a Picospritzer II (Parker Hannifin, Mayfield Heights, OH). To ensure delivery of drug, 0.05% (w/v) fast green was used. (Local applications of 1% fast green induced no response.) To visualize cells expressing P2X receptors without an EGFP tag, cells were cotransfected with EGFP. Cells expressing EGFP or EGFP-tagged P2X subunits were observed under a microscope with an epifluorescence attachment (Nikon, Tokyo, Japan). Untransfected cells and cells expressing EGFP alone were found to have no inward current in response to application of agonists. Acquisition was performed using HEKA Pulse Version 8.30, and data were subsequently analyzed using IgorPRO Version 3.16. Statistical analyses were performed with Student's unpaired t test.

Results

Coexpression of P2X₄ and P2X₇ Increased the Surface Expression of P2X₄. We examined the distribution of P2X₄ and P2X₇ when expressed either individually or together in HEK293 and NRK cells using immunocytochemical labeling. We showed previously that P2X₄ receptors undergo rapid and constitutive endocytosis and reside predominantly within intracellular compartments (Bobanovic et al., 2002; Royle et al., 2002). Figure 1 shows P2X₄ colocalized with the lysosomal marker, Lamp-1, whereas P2X₇ receptors containing a Glu-Glu tag at the C terminus (P2X₇-EE) were predominantly colocalized with the ER marker, DsRed-ER (Fig. 1A). Using methanol rather than PFA to fix the cells, enhanced antibody labeling of P2X₇ was performed at the plasma membrane (Fig. 1B). When P2X₄ and P2X₇ were coexpressed in NRK cells, P2X₄ was still predominantly within endolysosomes, and there was very little labeling of P2X₇-EE in these compartments, suggesting that these receptors are P2X₄ homomers. At the plasma membrane, however, there was overlap in the distribution of the two receptors. To look at any change in the surface expression of P2X₄ and P2X₇, surface proteins were biotinylated and analyzed by Western blot (Fig. 1, C and D). The biotinylated fraction of P2X₄ increased ∼2-fold in the presence of P2X₇, although total P2X₄ levels did not change, suggesting that association with P2X₇ stabilizes P2X₄ at the plasma membrane. In contrast, there was no change in the surface expression of P2X₇ with and without P2X₄.

Association of P2X₄ and P2X₇ Receptor Subunits. To test whether P2X₇ receptor subunits can associate with P2X₄ receptor subunits, we performed immunoprecipitation experiments using HEK293 cells cotransfected with P2X₇ and P2X₄. Membrane proteins were solubilized using 1% DDM because this was more effective at solubilizing P2X₇ than 1% Triton X-100, NP-40, or CHAPS (data not shown). After immunoprecipitation of the receptor complex using anti-P2X₄ antibody, P2X₇ was detected by immunoblotting with anti-P2X₇ antibody (Fig. 2A) but only when P2X₄ was also present. We also coexpressed P2X₄-HA with P2X₇ and were able to coimmunoprecipitate P2X₇ with anti-HA antibody (Fig. 2B). This gel shows that although two bands were detected for P2X₇ in the membrane fraction, which presumably represent fully and partially glycosylated forms, only the higher band was detected after coimmunoprecipitation. This suggests that the complex formed by P2X₄ and P2X₇ is not a misaggregate that is retained in the ER but instead is trafficked along the secretory pathway to the trans-Golgi network and presumably from there to the plasma membrane.

• Download figure
• Open in new tab
• Download powerpoint

Fig. 2.

P2X₇ and P2X₄ subunits physically associate both when overexpressed in HEK293 cells and in native tissue, as shown by coimmunoprecipitation. A, HEK293 cells were transfected with P2X₄ and P2X₇ or with P2X₇ alone, and membrane protein complexes were immunoprecipitated with the anti-P2X₄ antibody, separated by SDS-PAGE, and immunoblotted with anti-P2X₄ (left) and anti-P2X₇ (right) polyclonal antibodies. The total membrane protein fractions (input, 2.5 μg) were also blotted with the same antibodies for comparison. B, cells were cotransfected with P2X₄-HA and P2X₇ or with P2X₇ alone, and membrane proteins were immunoprecipitated with an anti-HA antibody, showing that the complex can be isolated via a non–P2X-specific antibody. Blots were performed with anti-P2X₇ (left) and anti-HA (right) antibodies. C, membrane protein fractions isolated from mice BMDMs were subjected to immunoprecipitation with the anti-P2X₄ antibody, and the isolated complex was resolved by SDS-PAGE and blotted with both anti-P2X₄ (left) and anti-P2X₇ (right) antibodies, demonstrating the physical association between P2X₄ and P2X₇ in native tissue.

Having shown that overexpressed P2X₄ and P2X₇ associate with one another, we next tested whether the endogenous receptors in bone marrow-derived macrophages could be coimmunoprecipitated. We performed the coimmunoprecipitation with anti-P2X₄ and the immunoblot with anti-P2X₇ and were able to detect a clear band running at the appropriate size for P2X₇ (Fig. 2C). Thus, native P2X₄ and P2X₇ receptors in mouse macrophages associate to form part of the same complex.

A Functional Interaction between P2X₄ and P2X₇ Receptors. To investigate the functional significance of the interaction between P2X₄ and P2X₇ subunits, we used two nonfunctional P2X₄ receptor mutants, C353W and S341W, which were characterized previously in Xenopus laevis oocytes (Silberberg et al., 2005). When expressed alone in oocytes, neither mutant produced a current in response up to 300 μM ATP, although their surface expression was equivalent to the wild-type (wt) P2X₄ receptor, indicating that the mutations interfered with channel functional and not maturation or trafficking. When coexpressed with wt P2X₄, the C353W mutant dramatically reduced the currents, whereas the S341W mutant had no inhibitory effect (Silberberg et al., 2005). We obtained very similar results using the EGFP-tagged mutants, expressed alone or with wt P2X₄ in HEK293 cells (Fig. 3A). Both mutants expressed individually were nonfunctional in response to 30 to 100 μM ATP, and when coexpressed with wt P2X₄, the C353W mutant reduced the peak current amplitude by ∼75%, whereas the S341W mutant produced a small but not significant potentiation. We next compared the effect of these mutants on the current carried by the wt P2X₇ receptor (Fig. 3B). The P2X₄C353W mutant inhibited the P2X₇ receptor currents evoked by 1 mM ATP^4- by greater than 50%, but the currents recorded from cells coexpressing the S341W mutant with P2X₇ were slightly increased compared with P2X₇ alone. Similar results were obtained with 100 μM ATP^4- and with the P2X₇ receptor preferred agonist, BzATP (Fig. 3C). We compared the BzATP dose-response relationship for P2X₇ alone and P2X₇ coexpressed with the S341W mutant, and the EC₅₀ values were similar (11 ± 1 and 9.1 ± 0.8 μM, respectively) (Fig. 3D). In contrast, the wt P2X₄ receptor produced very little response to BzATP up to concentrations of 100 μM. Neither the S341W nor the C353W mutant altered the surface expression of P2X₇, as measured by biotinylating surface proteins (Fig. 3E), indicating that functional inhibition of P2X₇ by the C353W mutant was not caused by a reduction in the number of receptors reaching the plasma membrane. If we assume that association of the C353W mutant with P2X₇ abolishes receptor function, as suggested for receptors formed from wt P2X₄ and the C353W mutant, then we can estimate that there was a >2-fold reduction in the number of functional homomeric P2X₇ receptors at the plasma membrane in the presence of the C353W mutant. This suggests that more than half of the surface P2X₇ receptor subunits were associated with the P2X₄C353W mutant. In contrast to the inhibition of P2X₇ by P2X₄C353W, there was no change in P2X₂ receptor currents upon coexpression with either P2X₄C353W or S341W (Fig. 3F).

• Download figure
• Open in new tab
• Download powerpoint

Fig. 3.

Inhibition of P2X₇ receptor function by coexpression with a dominant-negative P2X₄ mutant. A, whole-cell patch-clamp recordings were carried out on HEK293 cells expressing P2X₄ alone or together with either the P2X₄C353W or P2X₄S341W mutants. The mutants were also expressed alone. Mutants were tagged with EGFP to enable the transfected cells to be detected. Inward currents, evoked by 30 μM MgATP at a holding potential of -30 mV, were measured in normal Na⁺ EC solution. The histogram shows the normalized, mean peak current densities (mean ± S.E.M., n > 8). B, similar experiments were carried out, coexpressing P2X₇ and the P2X₄C353W and S341W mutants. Currents were evoked by applying 1 mM ATP^4- in Na⁺ EC solution with a low divalent cation concentration at a holding potential of -30 mV. Histogram on the right shows a comparison of the normalized peak current densities (mean ± S.E.M., n > 10). C, similar results were obtained using either ATP^4- (100–1000 μM) or BzATP (10–100 μM). D, concentration-response curves for BzATP in HEK293 cells expressing P2X₇, P2X₇ and P2X₄ S341W, or P2X₄ alone. The data were fit with a Hill equation, and the EC₅₀ values obtained were 11 ± 1 μM BzATP for P2X₇ and 9.1 ± 0.8 μM BzATP for P2X₇ with P2X₄ S341W (n = 4–7), and the corresponding Hill slopes were 1.7 ± 0.2 and 2.0 ± 0.3, respectively. E, the surface expression of P2X₇, as measured by biotinylation of surface proteins at 4°C, was unchanged in the presence of the two P2X₄ mutants. F, coexpression of the P2X₄ mutants with P2X₂ did not inhibit P2X₂-receptor mediated currents evoked by 30 μM ATP (mean ± S.E.M., n > 10). *, p < 0.05; **, p < 0.01; and ***, p < 0.001.

Having demonstrated an interaction between the P2X₄ mutants and wt P2X₇, we tested how coexpression of wt P2X₄ and P2X₇ affected the responses to BzATP and MgATP (Fig. 4, A and B). MgATP activated large currents in cells expressing P2X₄ alone, and BzATP activated large currents in cells expressing the P2X₇ receptor. In cells coexpressing these receptors, summation of the response produced by each receptor expressed individually would suggest two independent pools of homomeric receptors. The currents evoked by both agonists, however, were significantly reduced compared with what one would predict for a simple summation, suggesting a functional interaction between the two receptors. This was not the case for P2X₂ and P2X₇ (Fig. 4C), and the apparent lack of a functional interaction between these two subtypes was consistent with their inability to coimmunoprecipitate (Supplementary Fig. S1).

• Download figure
• Open in new tab
• Download powerpoint

Fig. 4.

P2X₄ but not P2X₂ functionally interacts with P2X₇. Whole-cell currents were recorded from HEK293 cells expressing the constructs shown. Normalized mean current densities are for n > 8 for each condition and V_h = -30 mV. A, inward currents were evoked by 10 μM BzATP in Na⁺ EC solution with low divalent. B and C, inward currents were evoked by 200 μM MgATP in Na⁺ EC solution with 1.58 mM Ca²⁺.

Coexpression of P2X₄ with P2X₇ Confers Ivermectin Sensitivity to BzATP-Evoked Currents. To further test whether or not an interaction between P2X₄ and P2X₇ alters the functional properties of the receptors, we compared some of their pharmacological properties when expressed individually and together. P2X₇ receptors have been shown to be sensitive to EC Na⁺ (Ma et al., 2006). Substituting Cs⁺ for Na⁺ in the EC solution slowed the activation and deactivation kinetics of P2X₇ receptor currents and enhanced the difference in its time course compared with P2X₄ receptor currents (Fig. 5). It also potentiated the amplitude of both BzATP-evoked P2X₇ receptor currents and currents recorded from cells coexpressing P2X₄ and P2X₇ (Supplementary Fig. S2), whereas there was minimal activation of P2X₄ receptors using 3 μM BzATP (Fig. 5A). Ivermectin (IVM) is an allosteric modulator of P2X₄ receptors that augments currents by stabilizing the agonist-induced open state (Priel and Silbergberg, 2004). We tested its effects on BzATP-evoked currents, and whereas there was no effect on currents from cells transfected with P2X₇ alone, in cells coexpressing P2X₄ and P2X₇, the currents were potentiated >2-fold after prior incubation with IVM. This suggests that association of P2X₄ with P2X₇ confers IVM sensitivity to the receptor, although this interpretation is confounded by the finding that BzATP is a much more effective agonist at homomeric P2X₄ receptors after IVM treatment. The time course of the P2X₄ receptor currents, however, differed from those recorded from cells coexpressing the two receptors. To remove any component of the whole-cell current that was mediated by P2X₄ homomeric receptors, a similar experiment was performed but using the P2X₄S341W mutant coexpressed with P2X₇. This mutant expressed alone was nonfunctional with and without IVM pretreatment but conferred IVM sensitivity to BzATP-evoked currents recorded from cells coexpressing this mutant with P2X₇ (Fig. 5B). Similar results were obtained with Na⁺ EC solution (Fig. 5C) and using MgATP as the agonist, although the currents were much smaller (Fig. 5D).

• Download figure
• Open in new tab
• Download powerpoint

Fig. 5.

P2X_4/7 receptor currents were potentiated by IVM. Whole-cell patch-clamp recordings from HEK293 cells expressing the constructs indicated with and without preincubation with 3 μM IVM for 5 min. V_h =-30 mV. All histograms represent peak current densities (mean ± S.E.M., n > 10). A and B, recordings were made in Cs⁺ EC solution with low divalent cations in response to 3 μM BzATP. C, BzATP-activated currents were recorded in Na⁺ EC solution with low divalent cations. D, currents were recorded in Na⁺ EC solution containing 1.58 mM Ca²⁺ and in response to 200 μM MgATP.

TNP-ATP and Brilliant Blue G Inhibit P2X_4/7 Receptor Currents. We next tested the effects of the P2X₄ receptor antagonist, TNP-ATP, and the P2X₇ receptor antagonist Brilliant Blue G (BBG) (Virginio et al., 1998; Jiang et al., 2000; Tsuda et al., 2003). TNP-ATP (2 μM) inhibited MgATP-activated P2X₄ receptor currents by ∼70% but had no significant effect on the amplitude of BzATP-evoked P2X₇ receptor currents. In contrast, it significantly inhibited BzATP-evoked currents from cells coexpressing P2X₄ and P2X₇ (Fig. 6A). At this concentration of BzATP, P2X₄ homomeric receptor currents are not expected to make a significant contribution to the whole-cell current amplitude; however, to rule out the possibility that there was a significant P2X₄ receptor-mediated component, the experiment was repeated using the S341W mutant. Currents recorded from cells coexpressing P2X₄ S341W and P2X₇ were reduced by ∼70% after incubation with TNP-ATP.

Finally, we tested the effects of BBG, which almost completely abolished BzATP-activated P2X₇ receptor currents but had very little effect at P2X₄ receptors (Fig. 6B). It substantially reduced currents recorded from cells coexpressing P2X₇ and P2X₄ S341W, although these were significantly less sensitive as judged by the amplitude of the BzATP-evoked response after a 10-min incubation with 1 μM BBG.

Discussion

The P2X₇ receptor represents an important therapeutic target in a number of diseases such as stroke and arthritis pain. ATP-mediated responses in native tissues are frequently ascribed to P2X₇; however, they often display characteristics that are shared by other P2X receptors (North, 2002; Inoue, 2006). A molecular basis for these responses has not been established because the P2X₇ receptor is not believed to associate with any other P2X subunits. Our results provide evidence that the P2X₇ receptor can associate with another P2X subunit, which has an emerging role in pain and inflammation, namely P2X₄ (Tsuda et al., 2003; Inoue et al., 2004; Guo and Schluesener, 2005). This interaction was demonstrated not only in HEK293 cells overexpressing both receptors but also endogenously for native receptors present in primary cultures of BMDMs. Our results differ from those of an earlier study, which failed to coimmunoprecipitate P2X₄ and P2X₇ from HEK293 cells (Torres et al., 1999). A possible explanation is the use of different detergents; we used DDM, which often preserves protein activity better than other detergents, including NP-40. The association with P2X₇ affected the trafficking properties of P2X₄, increasing its stability at the plasma membrane, although it was still predominantly located within intracellular compartments. Coexpression with a dominant-negative mutant of P2X₄ (C353W) knocked down P2X₇ receptor currents by >50% without reducing its surface expression, suggesting that more than half of the surface P2X₇ subunits were in complexes associated with the C353W mutant. By using a nonfunctional but non–dominant-negative P2X₄ mutant, we provide evidence that a heteromer formed from P2X₄ and P2X₇ has properties in common with both of the parent homomeric receptors. A question that remains to be answered is whether or not P2X₄ and P2X₇ subunits coassemble to form heterotrimeric structures with a common central conduction pore. In light of our findings, however, the prevailing hypothesis of P2X₇ receptors as unique members within the P2X family forming exclusively stable homotrimers seems unlikely to be correct.

• Download figure
• Open in new tab
• Download powerpoint

Fig. 6.

Inhibition of P2X_4/7 receptor currents by TNP-ATP and BBG. A, whole-cell currents recorded in Na⁺ EC solution, with and without preincubation with 2 μM TNP-ATP for 8 min. V_h =-30 mM. Histogram shows the peak current amplitudes (mean ± S.E.M., n >10). B, representative currents recorded from HEK293 cells expressing either P2X₇ or with P2X₄ S341W EGFP in response to 30 μM BzATP in Na⁺ ES solution with low divalent. BBG (1 μM) was applied for 10 min. V_h = -30 mM. Histogram shows mean current densities ± S.E.M. The data for P2X₄ was obtained using 30 μM MgATP (n > 4) instead of BzATP.

The identification of novel functional properties that cannot be attributed to the parent homomeric receptors is a well-established approach for demonstrating heteromerization between different members of the same family of receptor. In this study, to distinguish between homomeric and heteromeric receptor currents, we took advantage of a P2X₄ receptor mutant (S341W), which, although nonfunctional when expressed alone, had been shown previously to traffic to the surface in the normal way and to produce no inhibition of wt P2X₄ receptor currents (Silberberg et al., 2005). When coexpressed with wt P2X₇, this mutant produced a small potentiation of BzATP- and ATP^4--evoked currents, and we have made the assumption that currents with P2X₄-like pharmacological properties were mediated by a heteromer composed of P2X₄S341W and P2X₇. Our conclusions are that the heteromeric receptors are preferentially activated by BzATP compared with MgATP, they are allosterically modulated by IVM, and they are inhibited by both TNP-ATP and BBG. In Cs⁺ EC solution, the IVM-sensitive component of the whole-cell current was larger in amplitude and with slower activation and deactivation kinetics than in Na⁺ EC solution, suggesting that the P2X_4/7 complex is inhibited by EC Na⁺ similar to P2X₇ homomeric receptors. The increase in receptor diversity as a result of functional heteromeric and homomeric P2X receptors being expressed in cells such as macrophages and microglia, which predominantly express P2X₄ and P2X₇, clearly has implications for the development of new therapies that target purinergic receptors for the treatment of neuropathic pain.

P2X₄ and P2X₇ are coexpressed in epithelial and endothelial cells and in immune cells (Bowler et al., 2003; Xiang and Burnstock, 2005; Ma et al., 2006). In airway epithelia, P2X receptor agonists stimulate Cl^- transport across nasal mucosa and are involved in the regulation of ciliary beat (Zsembery et al., 2004; Hayashi et al., 2005). Manipulation of both of these processes may be of therapeutic benefit for patients with cystic fibrosis, and defining the subunit identity and functional properties of the native receptors is important if these receptors are to be targeted for cystic fibrosis therapy. Ma et al. (2006) described the pharmacological features of the P2X receptor in airway ciliated cells, and several properties are similar to what we report here for the heterologously coexpressed receptors. The native receptor currents were inhibited by BBG and EC Na⁺ and augmented by IVM. Based on this pharmacological profile, they hypothesized that the P2X receptor in ciliated cells is an assembly of P2X₄ and P2X₇ subunits, which is supported by our results.

The association between P2X₄ and P2X₇ may alter downstream signaling pathways, for example activation of the mitogen-activated protein kinase cascade (Donnelly-Roberts et al., 2004), phosphatidylserine translocation (Dutot et al., 2006), and coupling to ABC transporters (Marty et al., 2005). The hemichannel pannexin-1 was shown recently to form a complex with P2X₇ and to play an important role in coupling activation of the receptor to increased membrane permeability to large molecules such as ethidium and to IL-1β synthesis and release (Pelegrin and Surprenant, 2006). Whether pannexin-1 can functionally couple with P2X_4/7 receptors remains to be established. Our results provide a molecular basis for the existence of P2X_4/7 receptors, and further studies are required to elucidate the precise nature of the interaction between the constitutive subunits. These data also present functional significance for the heteromeric assembly between P2X₄ and P2X₇, and we foresee the importance of this interaction in purinergic receptor-mediated signaling of pain in health and disease.