|Home | About | Journals | Submit | Contact Us | Français|
P2X receptors (P2XR) act as ligand-gated, cation-selective ion channels. A common characteristic of all seven P2X family members is a conserved consensus sequence for protein kinase C (PKC)-mediated phosphorylation in the intracellular N-terminus of the receptor. Activation of PKC has been shown to enhance currents through P2X3R, however the molecular mechanism of this potentiation has not been elucidated. In the present study we show that activation of PKC can enhance adenosine triphosphate (ATP)-mediated Ca2+ signals ~2.5-fold in a DT-40 3KO cell culture system (P2 receptor null) transiently overexpressing P2X3R. ATP-activated cation currents were also directly studied using whole cell patch clamp techniques in HEK-293 cells, a null background for ionotropic P2XR. PKC activation resulted in a ~8.5-fold enhancement of ATP-activated current in HEK-293 cells transfected with P2X3R cDNA, but had no effect on currents through either P2X4R- or P2X7R-transfected cells. P2X3R-transfected HEK-293 cells were metabolically labeled with 32PO4− and following treatment with phorbol-12-myristate-13-acetate (PMA) and subsequent immunoprecipitation, there was no incorporation of 32PO4− in bands corresponding to P2X3R. Similarly, in vitro phosphorylation experiments, utilizing purified PKC catalytic subunits failed to establish phosphorylation of either P2X3R or P2X3R-EGFP. These data indicate that PKC activation can enhance both the Ca2+ signal as well as the cation current through P2X3R, however it appears that the regulation is unlikely to be a result of direct phosphorylation of the receptor.
Extracellular ATP can initiate signal transduction by activating purinergic receptors of the P2X and P2Y subtypes. P2XR are ligand-gated, cation selective ion channels; whereas P2Y receptors (P2YR) are G-protein coupled receptors (GPCR). Seven genes have been identified as coding for P2XR (P2X1-P2X7R) . The functional channel is formed by three P2X subunits . Each subunit contains two transmembrane spanning domains, with both the N- and C-terminus facing the cytoplasm, and the large extracellular loop containing the ATP binding site .
The N- and C-termini of P2XR provide putative unique areas of regulation/modulation by intracellular factors. One conserved feature of all members of the P2XR family is a consensus sequence for PKC mediated phosphorylation in the N-terminal domain. This region consists of the amino acids Thr-X-Arg/Lys, or T-X-R/K (X = amino acid). The importance of this site in the N-terminus of P2X2R was first implicated by studies showing that the fast-desensitizing kinetics of a C-terminal truncated receptor could be transformed into one with slow-desensitizing kinetics following treatment with PKC activators . This study also showed that mutations in the PKC consensus sequence exhibited rapid desensitization kinetics and that residue Thr18 was phosphorylated in wild-type P2X2R .
Expression of P2X3R were first described in sensory neurons [4, 5]. Inflammatory mediators, including the phospholipase C coupled ligands, substance P and bradykinin together with phorbol ester treatment to directly activate PKC have also been shown to augment P2X3R currents . However, this particular study was unable to determine whether PKC-mediated phosphorylation occurred directly, presumably at the N-terminal PKC site, or alternatively that the effect was mediated via phosphorylation of an unknown accessory protein . In fact, recent studies have even suggested the possibility of an external PKC site on the receptor [7, 8]. Therefore the goal of this study was specifically to determine whether P2X3R are directly phosphorylated by PKC.
DT-40 3KO cells were used as a null background for P2XR . These cells lack all three (3KO) inositol 1,4,5-trisphosphate receptors (InsP3R) thus avoiding any possible Ca2+ response as a consequence of P2YR activation. DT-40 3KO cells were kindly provided by Dr. Kurosaki (Kansai Medical University, Japan)  and maintained as previously described [11–14]. DT-40 3KO cells provided a convenient system for digital Ca2+ imaging studies since multiple cells from a single experimental run could be averaged. HEK-293 cells could not be used for Ca2+ imaging studies since they express endogenous P2YR that are activated by extracellular ATP . DT-40 3KO cells were loaded with the Ca2+ sensitive dye Fura-2 AM (2 μM, TEFLABS, Austin, TX) by incubation for 15 minutes at room temperature (RT). Subsequently, cells were removed from the Fura-2 AM containing solution, and resuspended in a physiological saline solution used for imaging experiments that contained (mM): 137 NaCl, 0.56 MgCl2, 4.7 KCl, 1 Na2HPO4, 10 HEPES, 5.5 glucose, 1.26 CaCl2, pH 7.4. Rapid solution changes were performed utilizing an electronic solenoid controlled perfusion system and gravity fed reservoirs (Warner Instruments, Hamden, CT). Imaging was performed using an inverted epifluorescence Nikon microscope with a 40X oil immersion objective lens (numerical aperture, 1.3). Fura-2 loaded cells were excited alternately with light at 340 and 380 nm using a monochrometer-based illumination system and the emission at 510 nm captured using a high speed, digital CCD camera (TILL Photonics, Pleasanton, CA). The fluorescent ratio of 340 nm/380 nm was calculated and all data is presented as the change in ratio units. Images were acquired at a rate of 1 Hz with an exposure of 20 ms. All imaging experiments were performed at RT, essentially as previously described [16, 17]. Traces are from a single cell, representative of multiple individual cells in a particular experimental run and n represents the number of experimental runs, with at least 3 cells per experimental run.
Human P2X3R cDNA, kindly provided by R.A. North (University of Manchester, UK), was transiently transfected into DT-40 3KO cells using a Nucleofector System (Amaxa, Gaithersburg, MD) following the instructions provided. Specifically, 5 × 106 cells were resuspended in 100 μL of Cell-line Nucleofector Kit T solution and were co-transfected with 5 μg of the P2X3R cDNA and 1 μg of pHcRed 1-N1 cDNA (red fluorescent protein for visualization of positively transfected cells) using Nucleofector program B-23. Immediately after transfection, 500 μL of media was added to the Nucleofector cuvette, followed by transfer of the cells to a single well of a 6 well culture plate which contained 1.5 mL of media. Experiments were performed 24 hours after transfection.
Human P2X3R cDNA was amplified by PCR. Hind III and Sal I restriction sites were incorporated into the oligonucleotides used for PCR amplification. The PCR products were restriction enzyme-digested and ligated into pEGFP-N3 at the Hind III and Sal I sites (BD Biosciences Clontech, San Jose, CA). This construct was verified by sequencing and creates a fusion protein with EGFP at the C-terminus of the human P2X3R.
Human P2X3R, human P2X3R-EGFP, rat P2X4R, rat P2X7R (rat P2XR cDNA also kindly provided by R.A. North), human 3HA-M3R (obtained from the UMR cDNA resource center, available on the World Wide Web at www.cdna.org), or rat S1−/S2+ InsP3R type I cDNA (kindly provided by Dr. S. Joseph, Thomas Jefferson University) was transiently transfected into HEK-293 cells using Lipofectamine 2000 (Invitrogen, Carlsbad, CA) following the instructions provided. Specifically, 5 × 104 cells were grown on 25 mm cover slips in 6 well culture plates and were co-transfected with 1 μg of the P2XR cDNA and 100 ng of pHcRed 1-N1 cDNA as described previously .
ATP-activated cation currents were recorded at a sampling rate of 1 kHz using an Axopatch 200A patch clamp amplifier (Axon Instruments, Union City, CA), Axon digital interface, and pCLAMP version 9.0 software under whole cell patch clamp conditions. To measure ATP-activated currents in HEK-293 cells, cells were perfused with an extracellular solution containing (mM): 140 NaCl, 5 CsCl, 1.2 MgCl2, 1 CaCl2, 10 HEPES-CsOH, 10 D-glucose, pH 7.4. Internal patch solution contained (mM): 140 Cs-acetate, 1.22 MgCl2, 10 HEPES-CsOH, 0.1 EGTA, 10 NaCl, 0.0365 CaCl2, pH 7.2. Intervals of 2–3 minutes were allowed between patch rupture and stimuli to allow for equilibration with the patch pipette solution. HEK-293 cells were held at a holding potential of −30 mV. Experiments were performed at RT.
Immunoprecipitated samples were prepared from either mock-, InsP3R type I-, 3HA-M3R-, P2X3R-, or P2X3R-EGFP-transfected HEK-293 cells. The media was removed and cells were washed in ice-cold PBS, followed by resuspension in 400 μL of ice-cold RIPA (RadioImmunoPrecipitation Assay) lysis buffer that contained (mM): 50 NaF, 2 EDTA, 150 NaCl, 0.01 Na-Phosphate, 1% nonidet P-40 (NP-40), 1% Na-deoxycholate, 0.1% SDS, and 1 Complete EDTA-free protease inhibitor mixture tablet (Roche, Mannheim, Germany). Samples were left on ice with gentle agitation for 30 minutes to solubilize. Immunoprecipitating primary antibodies, polyclonal α-P2X3R from Alomone Labs (Jerusalem, Israel), polyclonal α-InsP3R type I from Calbiochem (San Diego, CA), monoclonal α-GFP from Roche (Mannheim, Germany), and monoclonal α-HA.11 from Covance (Princeton, NJ) were used at a 1:100 dilution and incubated for 2 hours at 4°C with rotation. Protein A or G sepharose (70 μL or 50 μL) was added to each sample and rotated at 4°C for 1 hour, then washed and centrifuged seven times, before final resuspension in 1X SDS loading buffer. Samples were resolved on 7.5% SDS-PAGE and transferred as described previously . Polyclonal α-P2X3R (1:200 dilution) and α-InsP3R type I (1:750 dilution) primary antibodies (see above) were used following the manufacturers instructions for immunoblotting. Monoclonal α-GFP (1:3000 dilution) and α-HA.11 (1:1000 dilution) primary antibodies (see above) were used following the manufacturers instructions for immunoblotting. Proteins were visualized as previously described .
Mock-, InsP3R type I-, or P2X3R-transfected HEK-293 cells were metabolically labeled by incubating for 3 hours with 150 μCi/mL 32PO4− (PerkinElmer, Boston, MA) in a phosphate-free DMEM (Invitrogen, Carlsbad, CA). Following incubation, cells were treated with or without 100 nM PMA for 10 minutes at RT. Cells were washed once in an ice cold TBS solution that contained (mM): 20 Tris, 138 NaCl, pH 7.6. Cells were then resuspended in 400 μL of ice-cold RIPA lysis buffer and immunoprecipitations were performed as mentioned above. Samples were resolved on 7.5% SDS-PAGE and then the gel was placed on a gel dryer for 1 hour at 80°C (Bio-Rad, Hercules, CA). The dried gel was then placed in a 20 × 25 cm phosphor screen (Amersham Biosciences, Piscataway, NJ) for 22–72 hours before visualizing using a Molecular Dynamics PhosphorImager. Dried gels were then rehydrated for 30 minutes in SDS-PAGE running buffer that contained (mM): 25 mM Tris, 192 Glycine, 0.1% SDS, pH 8.3. Rehydrated gels were transferred to nitrocellulose and proteins were visualized as previously described in the immunoblotting section.
P2X3R, P2X3R-EGFP, 3HA-M3R, or InsP3R type I were immunoprecipitated (see above protocol) from mock-, P2X3R-, P2X3R-EGFP-, 3HA-M3R-, or InsP3R type I transiently transfected HEK-293 cells and transferred from ice cold RIPA lysis buffer after three washes to PKC phosphorylation buffer containing 20 mM Tris-HCl, 10 mM MgCl2, and Complete, EDTA-free protease inhibitor cocktail tablets (Roche, Mannheim, Germany), pH 7.5. Samples were briefly centrifuged and washed three times at 4°C in PKC phosphorylation buffer. After the final wash, all remaining buffer was removed and samples were then resuspended in PKC phosphorylation buffer containing 20 μM ATP, 8 μl of [γ-32P]-ATP (40 μCi) (PerkinElmer, Boston, MA), and either 10 ng of a catalytically active PKC fragment from rat brain (Calbiochem, San Diego, CA), which does not require Ca2+ or phosphatidylserine for its activity , were added or omitted as indicated (+ or − PKC). Final volume was 400 μl. Samples were mixed gently and incubated at 30°C for 15 minutes. Reactions were quenched by adding 1.3 mL of ice-cold PKC phosphorylation buffer containing 1 mM ATP. Samples were briefly centrifuged at 4°C and washed twice with 1.5 mL of phosphorylation buffer containing 1 mM ATP, finally resuspended in 1X SDS loading buffer. Samples were resolved on 7.5% SDS-PAGE and then the gel was dried. The dried gel was then placed in a 20 × 25 cm phosphor screen (Amersham Biosciences, Piscataway, NJ) for 60 hours before visualizing using a Molecular Dynamics PhosphorImager. Dried gels were then rehydrated and proteins were visualized as described above.
Statistical significance was determined using either a paired or unpaired t test as indicated. Data from several cells in a particular experimental run were averaged, and experimental averages were used to calculate the mean ± S.E. Two-tailed p values of less than 0.05 were considered statistically significant.
DT-40 3KO cells are a useful screening tool for P2XR studies since activity from multiple cells can be recorded in a single experimental run and unlike most cells lines, including HEK-293 cells, they have no endogenous P2YR . To verify that DT-40 3KO cells represent a null background for P2R, Ca2+ signaling events were first evaluated in mock-transfected DT-40 3KO cells following ATP stimulation . No Ca2+ signals were evoked by 200–500 μM ATP or 100 nM PMA in mock-transfected cells (Fig. 1A). However, cyclo-piazonic acid (CPA) treatment, to inhibit the sarcoendoplasmic reticulum Ca2+-ATPase (SERCA) and promote Ca2+ leak from intracellular stores, confirmed the viability of the measurement (Fig. 1A). In P2X3R-transfected DT-40 3KO cells, stimulation with 200 μM ATP caused a small reproducible Ca2+ transient (Fig. 1B, P2X3R control; second response 85 ± 16% of initial stimulation, n = 6). Importantly, after treatment of P2X3R-expressing cells with 100 nM PMA for 4–5 minutes, subsequent reapplication of 200 μM ATP caused a significant ~2.5-fold enhancement of the Ca2+ signal over control values (Fig. 1C, P2X3R PMA treatment; second response 268 ± 42% of initial stimulation, n = 12, p = 0.008). Note treatment with CPA resulted in a similar Ca2+ response in either mock- or P2X3R-transfected DT-40 3KO cells. When desensitization of the second ATP-induced Ca2+ signal in the absence of PMA was considered, this potentiation reached ~3.2-fold (Control; second response 85 ± 16% of initial stimulation versus PMA treatment; second response 268 ± 42% of initial stimulation). These data are summarized in Fig. 1D.
Quantifying the magnitude of the PMA-induced enhancement on P2X3R-mediated Ca2+ signaling was complicated by the initial small responses observed in DT-40 3KO cells. One possible reason for the small Ca2+ response is that these indirect measurements of channel activity rely on a spatially averaged global signal from the Ca2+ indicator, which might underestimate the signal if locally defined. In addition, the fact that the receptor desensitizes very rapidly may compound this problem. As a direct measurement of channel activity, we therefore used an electrophysiological approach to study the regulation of P2X3R in isolation following expressing in P2XR null HEK-293 cells [1, 19–24].
Figure 2 shows membrane currents recorded in the whole cell configuration of the patch clamp technique in response to extracellular ATP stimulated at a holding potential of −30 mV. The pipette solution and holding potential were chosen to isolate inwardly directed cation currents, predominately carried by Na+, as we have reported previously . No inward currents were evoked by 25 μM ATP in mock-transfected cells (Fig. 2A). Stimulation of P2X3R-transfected cells with 1 μM ATP resulted in an inward current, repeated stimulation after 4–5 min revealed a measurable desensitization of the P2X3R activity (Fig. 2B, P2X3R control; second response 72 ± 7% of initial stimulation, n = 3). In spite of this desensitization, treatment with PMA (4 min) in P2X3R expressing cells significantly enhanced the maximum amplitude of the inward current ~8.5-fold (Fig. 2C, P2X3R PMA treatment; second response 857 ± 199% of initial stimulation, n = 6, p = 0.03). When desensitization of the second ATP response in the absence of PMA was considered, this potentiation reached almost 12-fold (Control; second response 72 ± 7% of initial stimulation versus PMA treatment; second response 857 ± 199% of initial stimulation). These data are summarized in Fig. 2D.
One possible mechanism underlying this observation is as a result of a phosphorylation event occurring on the PKC consensus site in the N-terminal domain, conserved among all P2XR family members. Table I shows the amino acid alignment of the N-terminal regions of all seven rat P2XR . It should be noted that human P2X3R is identical to rat P2X3R in this region. The underlined residues represent the PKC consensus site, where the highlighted Thr (T) is presumably the putative site phosphorylated . Given the similarities in this region among P2XR, we next determined if treatment with PMA would also enhance signaling through both P2X4R and P2X7R.
Stimulation of P2X4R-transfected HEK-293 cells with 25 μM ATP elicited a robust inward current, which was unaffected by PMA treatment (Fig. 3A, P2X4R PMA treatment; second response 74 ± 2% of initial stimulation, n = 4). Stimulation of P2X7R-transfected HEK-293 cells with 100 μM ATP evoked inward currents of similar magnitude to those stimulated by 25 μM ATP in P2X4R-transfected HEK-293, consistent with reported EC50 values for each receptor (Fig. 3B) . Similarly, PMA treatment did not significantly enhance the current (Fig. 3B, P2X7R PMA treatment; second response 72 ± 7% of initial stimulation, n = 10). Currents through both P2X4R and P2X7R in the presence of PMA were similar to control values that we have previously reported under identical conditions . These data are summarized in Fig. 3C.
Our results have shown that following PKC activation, P2X3R channel activity is significantly augmented in two different cell lines following transient expression of the receptor. The simplest explanation of the underlying mechanism for this enhancement would be a PKC-mediated phosphorylation event occurring at the N-terminal region of the receptor. Contrary to this idea, it should be noted that PMA treatment can enhance P2X3R, but not P2X4R or P2X7R inward currents, yet all three receptors share a similar N-terminal PKC phosphorylation site. These findings might suggest that the conserved N-terminal PKC site might not be a universal substrate for phosphorylation/regulation by PKC. To determine if receptor phosphorylation was occurring on the P2X3R directly, we performed both intact cell and in vitro phosphorylation experiments after immunoprecipitation with a P2X3R antibody.
Immunoprecipitation followed by immunoblotting with a P2X3R antibody was performed to verify purification of the receptor. In lysates from P2X3R-transfected HEK-293 cells, a single major band was identified by a P2X3R-specific antibody, which was absent in mock-transfected cells (Fig. 4A). The size of the immunoreactive band was larger than the predicted size (44 kDa in human), as reported previously [26, 27]. Where indicated, the lower band marked (*) near the 50 kDa marker in samples depicts the band corresponding to the heavy chain of the antibody used for immunoprecipitation.
As a positive control for PKC-mediated phosphorylation, immunoprecipitation followed by immunoblotting with a InsP3R type I antibody was also performed. This control was chosen because this ligand-gated ion channel is also involved in Ca2+ signaling and has been shown to be a substrate for PKC [28, 29]. Lysates from InsP3R type I-transfected HEK-293 cells revealed a single distinct band of the predicted size (313 kDa), which was not evident in mock-transfected cells (Fig. 4B).
To determine if activation of endogenous PKC results in P2X3R phosphorylation, we first metabolically labeled mock-, InsP3R type I-, or P2X3R-transfected cells with 32PO4−. Subsequently, protein that had incorporated 32PO4− was detected by autoradiography. Following treatment with PMA and immunoprecipitation with either InsP3R type I or P2X3R antibodies, there was an enhanced labeling of a single band in InsP3R type I-transfected samples compared to control (Fig. 5A). This demonstrates that InsP3R (type I) are directly phosphorylated, presumably by a PKC-mediated process. In contrast, under identical conditions, no detectable signal was observed at the appropriate molecular weight for P2X3R (Fig. 5C). All mock-, InsP3R type I-, or P2X3R-transfected samples shown were run on the same gel, thus we next confirmed expression of transfected protein by rehydrating the gel, transferring to nitrocellulose, and immunoblotting with either InsP3R type I or P2X3R antibodies. Immunoblotting with a InsP3R type I antibody showed that both InsP3R type I-transfected samples contained similar amounts of protein and confirmed the identity based on the size of the phosphorylated protein detected in Figure 5A (Fig. 5B). Immunoblotting with a P2X3R antibody demonstrated similar expression of P2X3R in P2X3R-transfected samples, that was absent in mock-transfected samples (Fig. 5D). These data confirm that the P2X3R was present in Figure 5C. It is formally possible, however, that low level phosphorylation of the P2X3R might be below the sensitivity of this detection system. In addition, it is possible that 100 nM PMA stimulation is not sufficient to promote phosphorylation of the receptor, even though functional effects are readily evident. To address these issues, we next attempted to directly phosphorylate the P2X3R directly using active catalytic PKC subunits in vitro.
To determine whether the P2X3R can be directly phosphorylated by PKC, the receptor was purified by immunoprecipitation, followed by treatment with a catalytically active PKC fragment from rat brain and [γ-32P]-ATP in vitro. This PKC fragment has been shown previously to phosphorylate PKC substrates such as the δ2 glutamate receptor . Again, as a positive control, InsP3R type I-transfected HEK-293 cells were included. After a 15 minute incubation there was no detectable incorporation of 32P in either mock or P2X3R immunoprecipitated samples between the 50 and 75 kDa markers, even after substantial overexposure (Fig. 6A). Following rehydration and transferring this gel to nitrocellulose, immunoblotting with a P2X3R antibody confirmed the presence of the P2X3R protein in P2X3R- but not mock-transfected cells (Fig. 6B, bottom). InsP3R type I-transfected cells showed an increase in 32P incorporation after 15 minutes of PKC and [γ-32P]-ATP treatment at the correct molecular weight of the receptor. The presence of the InsP3R type I was confirmed after rehydrating the gel, transferring to nitrocellulose, and immunoblotting these samples with a InsP3R type I antibody (Fig. 6B, top). This increase in 32P incorporation was not present in the lane lacking PKC addition. These results demonstrate that P2X3R do not appear to be a substrate for PKC-mediated phosphorylation.
A possibility exists that the P2X3R was not being efficiently immunoprecipitated from our samples, since immunoblotting for P2X3R only resulted in a modest signal. To address this issue we generated a C-terminal EGFP tagged human P2X3R. This construct was confirmed to be functional by electrophysiological recordings and retained enhancement by PMA treatment (data not shown). Epitope tagging this receptor with EGFP has two distinct advantages; it allows for better resolution from the heavy chain of the antibody used for immunoprecipitation. Secondly, it also allows us to utilize a α-GFP antibody for a more efficient immunoprecipitation. To determine whether the P2X3R-EGFP can be directly phosphorylated by PKC, the receptor was purified by immunoprecipitation, followed by treatment with a catalytically active PKC fragment from rat brain and [γ-32P]-ATP in vitro. In addition, InsP3R type I- and 3X HA-tagged human M3R- (3HA-M3R) transfected HEK-293 cells were included as positive controls. Human M3R has been utilized previously as a positive control for P2X1R phosphorylation , and it is similar in size to P2X3R-EGFP. After a 15 minute incubation there was no detectable incorporation of 32P in either mock or P2X3R-EGFP immunoprecipitated samples between the 75 and 100 kDa markers, even after substantial overexposure (Fig. 7A and Fig. 7B). Following rehydration and transferring this gel to nitrocellulose, immunoblotting with a GFP antibody confirmed the presence of the P2X3R-EGFP protein in P2X3R-EGFP- but not mock-transfected cells (Fig. 7C, right). 3HA-M3R- and InsP3R type I-transfected cells showed an increase in 32P incorporation after 15 minutes of PKC and [γ-32P]-ATP treatment at the correct molecular weight of each receptor (Fig. 7A and Fig. 7B). The human M3 receptor has been shown previously to migrate as a diffuse band running at approximately 97-110 kDa due to post-translational modification of the receptor . The presence of both receptors was confirmed after rehydrating the gel, transferring to nitrocellulose, and immunoblotting these samples with either a α-HA or α-InsP3R type I antibodies (Fig. 7C, left and center). This increase in 32P incorporation was not present in the lanes lacking PKC treatment. These results demonstrate that P2X3R-EGFP do not appear to be a substrate for PKC-mediated phosphorylation. This additional information reinforces the notion that the regulation of P2X3R by PKC is likely indirect and thus presumably involves the phosphorylation of an unknown accessory protein.
The present study demonstrates that PKC activation can significantly enhance both the Ca2+ signal as well as the cation current through P2X3R in different cell lines. More importantly, this is the first study that has specifically addressed whether the P2X3R is directly phosphorylated after PKC activation. Utilizing two different techniques, activation of PKC failed to increase the phosphorylation of the P2X3R.
There is no consensus in the literature regarding the regulation of P2XR by PKC. It was first reported that phorbol ester treatment caused a mutant P2X2R with truncated C-terminus to convert from one exhibiting rapid desensitizing currents to slow desensitizing currents . In addition, P2X2R were shown to be phosphorylated at Thr18 using an antibody that recognizes a phosphothreonine-proline motif, whereas the P2X2R mutant K20T was not recognized . These authors showed that P2X2R are constitutively phosphorylated, thus giving these receptors their characteristic slow rate of desensitization; however they did not demonstrate whether the truncated C-terminus mutant P2X2R was phosphorylated at this N-terminal site following phorbol ester treatment. It should be noted that the particular antibody used to detect PKC-mediated phosphorylation would not be of use in the present study because it specifically recognizes a motif absent in P2X3R.
Ennion and Evans demonstrated that disruption of the N-terminal PKC site in P2X1R (T18A) caused reduced peak current amplitude as well as rapid desensitization kinetics . The authors speculated that PKC-mediated phosphorylation/dephosphorylation of P2X1R was a mechanism for regulating channel function . In a further study, the P2X1R mutants T18A, T18N, P19V, and R20T were made and the authors found that all mutations except P19V significantly reduced the current . They also indicated that the wild-type P2X1R was phosphorylated, whereas the mutant R20T was not, after immunoblotting with an antibody that recognizes a phosphothreonine-proline motif . However, Evans and colleagues later showed that activation of PKC, mediated either by GPCR stimulation (metabotropic glutamate receptor 1α, P2Y1R, or P2Y2R) or PMA, still enhanced current through point mutants with the N-terminal PKC site disrupted (R20I, R20A, and T18V). These data suggested PKC activation does not involve P2X1R phosphorylation in this region . This prediction was confirmed after radiolabelling phosphorylated proteins in the presence or absence of PMA, where P2X1R were basally phosphorylated and no enhanced phosphorylation was observed after PMA treatment. These data in total suggest that the mechanism of PKC-mediated P2X1R regulation likely involves the phosphorylation of an accessory protein .
P2X3R had been shown to be positively enhanced by Gq-coupled inflammatory mediators such as substance P and bradykinin. These effects could be mimicked by phorbol ester treatment and blocked by inhibitors of protein kinases . The most plausible explanation for this enhancement was either an N-terminal PKC-mediated phosphorylation or an unidentified protein that was phosphorylated and controls activity of P2X receptors . However, there have been recent reports that P2X3R are regulated by ecto-PKC activity . P2X3R have seven PKC consensus sequences, including the conserved N-terminal intracellular PKC site as well as a C-terminal intracellular PKC site. Mutation of the C-terminal intracellular PKC site did not inhibit PKC-mediated P2X3R potentiation . However, P2X3R also have five external PKC consensus sequences. The authors argued that since PKC activators such as PMA and DAG-lactone can transverse the cell membrane, that it gives no specificity as to the intracellular or extracellular location of the phosphorylation site. External PKC site mutants T134A and S178A both abolished the UTP-induced potentiation of the current through P2X3R, yet mutants T196A and S269A had no effect . Interestingly, activity of the internal PKC site mutant T13A, was not enhanced by UTP in this study, although a previous report has shown that this same mutation, P2X3T13A was still enhanced by PKC activation . The substitution of the same four external PKC sites to the negatively charged Asp residue (T134D, S178D, T196D, and S269D), to mimic phosphorylation, all prevented the potentiation by UTP . To summarize, adding a negative charge to T196 and S269, blocked potentiation by UTP, but the alanine mutants (T196A and S269A) were still enhanced by UTP, which renders the importance of these PKC sites unclear . Regardless, the substrate of the putative ecto-protein kinase is not resolved, in particular, it remains to be determined whether phosphorylation occurs at extracellular receptor sites or an accessory membrane protein(s) .
The present study reports a previously unreported finding; namely that PKC-mediated enhancement of P2X3R channel activity does not seem to involve direct channel phosphorylation. These results suggest that the PMA-induced potentiation of P2X3R signaling involves an unknown accessory protein. In support of this contention, we also have shown that two other P2XR family members, which also possess N-terminal PKC sites, are not regulated by PKC activation, suggesting these N-terminal PKC sites are not important for PKC-mediated regulation. While it is possible that PKC-mediated phosphorylation of P2X3R or P2X3R-EGFP was below the level of detection in both our intact cell and in vitro systems, this seems unlikely as the InsP3R type I and 3HA-M3R, included as a positive controls were phosphorylated after phorbol ester treatment or after incubation with active PKC catalytic subunits as previously reported [28, 29, 31]. Furthermore, P2X1R are positively modulated by PKC activation, however it does not involve direct phosphorylation of the receptor . It is interesting to note that these two P2X receptors belong to the same sub-group of P2XR based on their rapid desensitization properties and high sensitivity to ATP and selective agonist α,β-methylene ATP [3, 35]. In support of our findings, Vial and Evans were unable to successfully measure phosphorylation of the P2X3R. However, in this case they attributed this finding to insufficient expression levels of P2X3R achieved in HEK-293 cells in their experiments . In our transiently overexpressing HEK-293 cell culture system the expression of the human P2X3R or P2X3R-EGFP was unlikely to be an issue, since the receptor was readily detected by immunoblotting using either a P2X3R- or GFP-specific antibody (Fig. 4A and Fig. 7C).
One other possibility explaining why PKC-mediated phosphorylation was not detected could be that the number of PKC sites was significantly different in the InsP3R type I and 3HA-M3R when compared to the P2X3R-EGFP. After PKC-mediated phosphorylation followed by thermolysin digestion and two-dimensional phosphopeptide analysis, the InsP3R type I has been shown to produce one major phosphopeptide and two minor phosphopeptides , suggesting that the InsP3R type I has potentially 3 different PKC sites. The number and location of the active PKC phosphorylation sites in the human M3R has not been well defined, however the number of active PKC sites in human M1R has been determined to be 2–3 PKC sites . This number of sites seem reasonable for comparison to potential P2X3R-EGFP PKC-mediated phosphorylation. In addition both proteins are similar in size (Figure 7C). Furthermore, a search of the human 3HA-M3R protein shows it contains 17 PKC consensus sequences compared to 9 PKC consensus sequences in the human P2X3R-EGFP using the search criteria of [ST]X[RK] and including all regions of the receptor. If the assumption is made that the stoichiometry of phosphorylation is similar, one would expect to measure approximately half the amount of phosphorylation in the P2X3R-EGFP compared to the 3HA-M3R. Thus, it is reasonable to suggest, that since we can robustly detect 3HA-M3R phosphorylation (Figure 7A and and7B),7B), that P2X3R-EGFP phosphorylation should also be evident.
There is an emerging notion that P2XR activities are modulated by a range of protein kinases [17, 37]. Furthermore, it appears that protein kinases can selectively modulate signaling through different P2XR. Recent studies from our laboratory have revealed that mechanisms that increase cellular adenosine 3′,5′-cyclic monophosphate (cAMP), activating PKA, can significantly enhance both the Ca2+ signal and the cation current through P2X4R, however raising cAMP has no effect on P2X7R signaling . It remains to be determined if this regulation involves direct receptor phosphorylation or not. Additionally, P2X2R have also been shown to be regulated by PKA .
Given the preponderance of evidence, our results pose a novel and interesting question, what is the identity of the unknown accessory protein that modulates P2X3R? It is intriguing that this protein is conserved in at least two different cell lines of chicken and human origin (DT-40 and HEK-293) and future studies should be directed to ascertain the identity of this seemingly ubiquitous accessory protein and also address whether it can associate and regulate other P2XR. The current study helps support the notion that accessory proteins play a significant role in both the regulation and modulation of P2XR. In support of this contention, a heat shock protein, HSP90 was recently shown to interact with the P2X7R and modulate the receptor when phosphorylated .
In summary, PKC modulation of P2X3R represents a mechanism resulting in augmented intracellular Ca2+ signaling. These results support the emerging consensus that protein kinases can regulate P2XR signaling and therefore represent a point of convergence between individual signaling systems. However, importantly, it does not appear that the P2X3R is subject to a direct PKC-mediated phosphorylation event. This is nevertheless likely an important mechanism for selectively modulating P2XR and has broad implications for the fidelity of downstream P2XR signaling.
This work was supported in part by Grants R01-DK-054568 and R01-DE014756 from the National Institutes of Health (to D.I.Y.) and the NIDCR, National Institutes of Health Training Grant T32-DE07202 (to D.A.B.). We would like to thank Dr. R.A. North for providing the P2XR constructs and Drs. Jim Melvin, Matthew Betzenhauser, Larry Wagner, and Trevor Shuttleworth for helpful discussion during the course of this study.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.