Functional and Biochemical Evidence for Heteromeric ATP-gated Channels Composed of P2X1 and P2X5Subunits
1999; Elsevier BV; Volume: 274; Issue: 22 Linguagem: Inglês
10.1074/jbc.274.22.15415
ISSN1083-351X
AutoresKhanh‐Tuoc Lê, Éric Boué‐Grabot, Vincent Archambault, Philippe Séguéla,
Tópico(s)Bipolar Disorder and Treatment
ResumoThe mammalian P2X receptor gene family encodes two-transmembrane domain nonselective cation channels gated by extracellular ATP. Anatomical localization data obtained by in situ hybridization and immunocytochemistry have shown that neuronal P2X subunits are expressed in specific but overlapping distribution patterns. Therefore, the native ionotropic ATP receptors diversity most likely arises from interactions between different P2X subunits that generate hetero-multimers phenotypically distinct from homomeric channels. Rat P2X1 and P2X5 mRNAs are localized within common subsets of peripheral and central sensory neurons as well as spinal motoneurons. The present study demonstrates a functional association between P2X1 and P2X5subunits giving rise to hybrid ATP-gated channels endowed with the pharmacology of P2X1 and the kinetics of P2X5. When expressed in Xenopus oocytes, hetero-oligomeric P2X1+5 ATP receptors were characterized by slowly desensitizing currents highly sensitive to the agonist α,β-methylene ATP (EC50 = 1.1 μm) and to the antagonist trinitrophenyl ATP (IC50 = 64 nm), observed with neither P2X1 nor P2X5 alone. Direct physical evidence for P2X1+5co-assembly was provided by reciprocal subunit-specific co-purifications between epitope-tagged P2X1 and P2X5 subunits transfected in HEK-293A cells. The mammalian P2X receptor gene family encodes two-transmembrane domain nonselective cation channels gated by extracellular ATP. Anatomical localization data obtained by in situ hybridization and immunocytochemistry have shown that neuronal P2X subunits are expressed in specific but overlapping distribution patterns. Therefore, the native ionotropic ATP receptors diversity most likely arises from interactions between different P2X subunits that generate hetero-multimers phenotypically distinct from homomeric channels. Rat P2X1 and P2X5 mRNAs are localized within common subsets of peripheral and central sensory neurons as well as spinal motoneurons. The present study demonstrates a functional association between P2X1 and P2X5subunits giving rise to hybrid ATP-gated channels endowed with the pharmacology of P2X1 and the kinetics of P2X5. When expressed in Xenopus oocytes, hetero-oligomeric P2X1+5 ATP receptors were characterized by slowly desensitizing currents highly sensitive to the agonist α,β-methylene ATP (EC50 = 1.1 μm) and to the antagonist trinitrophenyl ATP (IC50 = 64 nm), observed with neither P2X1 nor P2X5 alone. Direct physical evidence for P2X1+5co-assembly was provided by reciprocal subunit-specific co-purifications between epitope-tagged P2X1 and P2X5 subunits transfected in HEK-293A cells. Ionotropic ATP receptors constitute a unique class of neurotransmitter-gated ion channels generated from the assembly of P2X subunits having two transmembrane-spanning domains and a protein architecture similar to the one of the amiloride-sensitive sodium channels (1Buell G. Collo G. Rassendren F. Eur. J. Neurosci. 1996; 8: 2221-2228Crossref PubMed Scopus (240) Google Scholar, 2North R.A. Barnard E.A. Curr. Opin. Neurobiol. 1997; 7: 346-357Crossref PubMed Scopus (426) Google Scholar). Functional characterization studies of the seven mammalian cloned P2X subunits heterologously expressed as homomeric channels allowed to classify them in three groups according to their properties of desensitization and to their sensitivity to the agonist α,β-methylene ATP (αβm-ATP) 1The abbreviations used are: αβm-ATP, α,β-methylene ATP; His6, hexahistidine; TNP-ATP, trinitrophenyl-ATP; NTA, nitrilotriacetic acid1The abbreviations used are: αβm-ATP, α,β-methylene ATP; His6, hexahistidine; TNP-ATP, trinitrophenyl-ATP; NTA, nitrilotriacetic acid: (i) rapidly desensitizing and αβm-ATP-sensitive receptors including P2X1 and P2X3 (3Valera S. Hussy N. Evans R.J. Adami N. North R.A. Surprenant A. Buell G. Nature. 1994; 371: 516-519Crossref PubMed Scopus (894) Google Scholar, 4Chen C.-C. Akopian A.N. Sivilotti L. Colquhoun D. Burnstock G. Wood J.N. Nature. 1995; 377: 428-431Crossref PubMed Scopus (911) Google Scholar, 5Lewis C. Neidhart S. Holy C. North R.A. Buell G. Surprenant A. Nature. 1995; 377: 432-435Crossref PubMed Scopus (889) Google Scholar), (ii) moderately desensitizing and αβm-ATP-insensitive receptors including P2X4 and P2X6 (6Bo X. Zhang Y. Nassar M. Burnstock G. Schoepfer R. FEBS Lett. 1995; 375: 129-133Crossref PubMed Scopus (258) Google Scholar, 7Buell G. Lewis C. Collo G. North R.A. Surprenant A. EMBO J. 1996; 15: 55-62Crossref PubMed Scopus (375) Google Scholar, 8Séguéla P. Haghighi A. Soghomonian J.-J. Cooper E. J. Neurosci. 1996; 1: 448-455Crossref Google Scholar, 9Soto F. Garcia-Guzman M. Gomez-Hernandez M.J. Hollmann M. Karschin C. Stuhmer W. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 3684-3688Crossref PubMed Scopus (304) Google Scholar, 10Wang C. Namba N. Gonoi T. Inagaki K. Seino S. Biochem. Biophys. Res. Commun. 1996; 230: 8063-8067Google Scholar, 11Collo G. North R.A. Kawashima E. Merlo-Pich E. Neidhart S. Surprenant A. Buell G. J. Neurosci. 1996; 16: 2495-2507Crossref PubMed Google Scholar, 12Garcia-Guzman M. Soto F. Laube B. Stuhmer W. FEBS Lett. 1996; 388: 123-127Crossref PubMed Scopus (103) Google Scholar), and (iii) nondesensitizing as well as αβm-ATP-insensitive receptors including P2X2, P2X5, and P2X7 (11Collo G. North R.A. Kawashima E. Merlo-Pich E. Neidhart S. Surprenant A. Buell G. J. Neurosci. 1996; 16: 2495-2507Crossref PubMed Google Scholar, 12Garcia-Guzman M. Soto F. Laube B. Stuhmer W. FEBS Lett. 1996; 388: 123-127Crossref PubMed Scopus (103) Google Scholar, 13Brake A.J. Wagenbach M.J. Julius D. Nature. 1994; 371: 519-523Crossref PubMed Scopus (837) Google Scholar, 14Surprenant A. Rassendren F. Kawashima E. North R.A. Buell G. Science. 1996; 272: 735-738Crossref PubMed Scopus (1476) Google Scholar). Results from Northern blots and in situ hybridization data (11Collo G. North R.A. Kawashima E. Merlo-Pich E. Neidhart S. Surprenant A. Buell G. J. Neurosci. 1996; 16: 2495-2507Crossref PubMed Google Scholar) have indicated that the six neuronal P2X subunits genes are transcribed in specific but overlapping populations in the central and peripheral nervous system (1Buell G. Collo G. Rassendren F. Eur. J. Neurosci. 1996; 8: 2221-2228Crossref PubMed Scopus (240) Google Scholar, 11Collo G. North R.A. Kawashima E. Merlo-Pich E. Neidhart S. Surprenant A. Buell G. J. Neurosci. 1996; 16: 2495-2507Crossref PubMed Google Scholar). This strongly suggests that neuronal P2X subunits belonging to different functional groups might co-assemble into heteromultimeric channels. All P2X subunits have been detected in peripheral sensory ganglia, reinforcing the view that synaptically or lytically released ATP could play an important signaling role in sensory pathways (1Buell G. Collo G. Rassendren F. Eur. J. Neurosci. 1996; 8: 2221-2228Crossref PubMed Scopus (240) Google Scholar, 11Collo G. North R.A. Kawashima E. Merlo-Pich E. Neidhart S. Surprenant A. Buell G. J. Neurosci. 1996; 16: 2495-2507Crossref PubMed Google Scholar, 15Cook S.P. Vulchanova L. Hargreaves K.M. Elde R. McCleskey E.W. Nature. 1997; 387: 505-508Crossref PubMed Scopus (393) Google Scholar). Rat P2X3 subunits have been reported to be exclusively expressed in small to medium-sized isolectin B4-positive nociceptive neurons in nodose, trigeminal, and dorsal root ganglia (4Chen C.-C. Akopian A.N. Sivilotti L. Colquhoun D. Burnstock G. Wood J.N. Nature. 1995; 377: 428-431Crossref PubMed Scopus (911) Google Scholar, 5Lewis C. Neidhart S. Holy C. North R.A. Buell G. Surprenant A. Nature. 1995; 377: 432-435Crossref PubMed Scopus (889) Google Scholar, 15Cook S.P. Vulchanova L. Hargreaves K.M. Elde R. McCleskey E.W. Nature. 1997; 387: 505-508Crossref PubMed Scopus (393) Google Scholar). A significant proportion of sensory neurons are thought to express hetero-oligomeric P2X2+3 receptors based on their sustained response to αβm-ATP applications (5Lewis C. Neidhart S. Holy C. North R.A. Buell G. Surprenant A. Nature. 1995; 377: 432-435Crossref PubMed Scopus (889) Google Scholar). However, recent immunocytochemistry results have demonstrated that P2X2 and P2X3 subunits in rat dorsal root ganglia are rarely co-localized at the level of central primary afferents in the dorsal horn of the spinal cord, despite their high degree of co-localization in somata, indicating different subunit-specific subcellular targetings (16Vulchanova L. Riedl M.S. Shuster S.J. Buell G. Surprenant A. North R.A. Elde R. Neuropharmacology. 1997; 36: 1229-1242Crossref PubMed Scopus (377) Google Scholar). Altogether, these data suggest that physiologically relevant associations of neuronal P2X subunits, giving rise to phenotypes that are not mediated by the previously described P2X2+3 (5Lewis C. Neidhart S. Holy C. North R.A. Buell G. Surprenant A. Nature. 1995; 377: 432-435Crossref PubMed Scopus (889) Google Scholar, 17Radford K.M. Virginio C. Surprenant A. North R.A. Kawashima E. J. Neurosci. 1997; 17: 6529-6533Crossref PubMed Google Scholar) or P2X4+6 (18Lê K.-T. Babinski K. Séguéla P. J. Neurosci. 1998; 18: 7152-7159Crossref PubMed Google Scholar) receptors, remain to be discovered. Rat P2X5 subunits mRNAs have the most restricted distribution in the P2X family, but in situ hybridization studies have indicated that P2X1 and P2X5mRNAs are co-localized in primary sensory neurons as well as within subsets of large motoneurons in the ventral horn of the spinal cord (1Buell G. Collo G. Rassendren F. Eur. J. Neurosci. 1996; 8: 2221-2228Crossref PubMed Scopus (240) Google Scholar,11Collo G. North R.A. Kawashima E. Merlo-Pich E. Neidhart S. Surprenant A. Buell G. J. Neurosci. 1996; 16: 2495-2507Crossref PubMed Google Scholar). We report here the characterization of a novel heteromeric P2X receptor with hybrid properties generated by co-expression and co-assembly of P2X1 with P2X5 subunits inXenopus laevis oocytes and transfected HEK-293A cells, further strengthening arguments for a diversity of native ATP-gated channels and purinergic phenotypes in mammalian neurons. Full-length wild-type rat P2X1 and P2X5 cDNAs were obtained through polymerase chain reaction amplification using A10 smooth muscle cells (ATCC No. CRL 1476) and adult rat spinal cord reverse transcribed-cDNA templates, respectively. Reactions were performed with exact match oligonucleotide primers based upon published primary sequences (3Valera S. Hussy N. Evans R.J. Adami N. North R.A. Surprenant A. Buell G. Nature. 1994; 371: 516-519Crossref PubMed Scopus (894) Google Scholar, 11Collo G. North R.A. Kawashima E. Merlo-Pich E. Neidhart S. Surprenant A. Buell G. J. Neurosci. 1996; 16: 2495-2507Crossref PubMed Google Scholar, 12Garcia-Guzman M. Soto F. Laube B. Stuhmer W. FEBS Lett. 1996; 388: 123-127Crossref PubMed Scopus (103) Google Scholar) using Pfu DNA polymerase (Stratagene) to minimize artifactual mutations. Epitope-tagged P2X subunits with carboxyl-terminal hexahistidine motif (His6) or Flag peptide were constructed as reported previously (18Lê K.-T. Babinski K. Séguéla P. J. Neurosci. 1998; 18: 7152-7159Crossref PubMed Google Scholar). Briefly, anXhoI-XbaI stuffer cassette containing in-frame Flag or His6 epitopes followed by an artificial stop codon was ligated to P2X1 and P2X5 cDNAs previously mutated to replace their natural stop codon with aXhoI restriction site. P2X1-Flag, P2X1-His6, P2X5-Flag, and P2X5-His6 were then subcloned directionally into the HindIII-XbaI sites of pcDNAI vector (Invitrogen, San Diego, CA) compatible with CMV-driven heterologous expression in HEK-293A cells and Xenopus laevis oocytes. RT-PCR products as well as mutant epitope-tagged subunits were subjected to automatic dideoxy sequencing (Sheldon Biotechnology Center, Montreal). cDNA transfections of epitope-tagged P2X subunits were performed in mammalian cells. HEK-293A cells (ATCC No. CRL 1573) were cultured in Dulbecco's modified Eagle's medium and 10% heat-inactivated fetal bovine serum (Wisent, St. Bruno, Canada) containing penicillin and streptomycin. Cells reaching 30–50% confluency were used for transient cDNA transfections with the calcium phosphate method with 10 μg of supercoiled plasmid cDNA per 106 cells. Transfected HEK-293A cells used for Western blots were then lifted in Hanks' modified calcium-free medium with 20 mm EDTA, pelleted at low centrifugation, and homogenized in 10 volumes of 10 mmHEPES buffer and 0.3 m sucrose, pH 7.40, containing protease inhibitors phenylmethylsulfonyl fluoride (0.2 mm) and benzamidine (1 mm). Membranes from cell lysates were solubilized with 1% Triton X-100 (Sigma) for 2 h at 4 °C and pelleted at 14000 × g for 5 min, and remaining membrane proteins within supernatants were used for Western blots. Solubilized proteins were incubated with 25 μl of equilibrated Ni-NTA resin (Qiagen, Hilden, Germany) for 2 h at 4 °C under agitation. Then Ni-NTA beads were washed six times in Tris-buffered saline containing 25 mm imidazole and 1% Triton X-100. Bound proteins were eluted from His6-binding resin with 500 mm imidazole, diluted 1:1 (v/v) with SDS-containing loading buffer. Samples were then loaded onto 10–12% SDS-PAGE and transferred to nitrocellulose. Immunostainings were performed with M2 murine monoclonal antibodies (10 μg/ml) (Sigma) or chicken anti-Flag polyclonal antibodies (1:200) (Aves) followed by incubations with corresponding species-specific peroxidase-labeled secondary antibodies (1:5000–1:20,000) for visualization by enhanced chemiluminescence (Amersham Pharmacia Biotech). Electrophysiological recordings were performed in Xenopus oocytes. Ovary lobes were surgically retrieved from X. laevis frogs under deep tricaine (Sigma) anesthesia. Oocyte-containing lobes were then treated for 3 h at room temperature with type II collagenase (Life Technologies, Gaithersburg, MD) in calcium-free Barth's solution under vigorous agitations. Stage V–VI oocytes were then chemically defolliculated before nuclear micro-injections of 5–10 ng of cDNA coding for each P2X channel subunit. Following 2–5 days of incubation at 19 °C in Barth's solution containing 1.8 mm calcium chloride and 10 μg/ml gentamicin (Sigma), elicited P2X currents were recorded in two-electrode voltage-clamp configuration using an OC-725B amplifier (Warner Instruments). Responsive signals were low pass filtered at 1 kHz, acquired at 500 Hz using a Macintosh IIci computer equipped with an NB-MIO-16XL analog-to-digital card (National Instruments). Recorded traces were post-filtered at 100 Hz in Axograph (Axon Instruments). Agonists, antagonists, and P2X co-factors (10 μm zinc chloride, pH 6.40 and pH 8.40) were prepared at room temperature in Ringer's perfusion solution containing 115 mm NaCl, 2.5 mm KCl, 1.8 mm CaCl2, and 10 mm HEPES buffered at pH 7.40. Solutions were perfused onto oocytes at a constant flow rate of 10–12 ml/min. Dose-response curves were fitted to the Hill sigmoidal equation, and EC50 as well as IC50 values were determined using the software Prism 2.0 (Graphpad Software, San Diego, CA). To assess the presence of P2X1+5 heteromers inXenopus oocytes co-injected with both subunits, we tested the expression of inward currents during prolonged applications (5–10 s) of 50 μm αβm-ATP, exploiting the fact that homomeric P2X5 ATP-gated channels are almost insensitive to this agonist when applied at concentrations below 100 μm(Fig. 1) (11Collo G. North R.A. Kawashima E. Merlo-Pich E. Neidhart S. Surprenant A. Buell G. J. Neurosci. 1996; 16: 2495-2507Crossref PubMed Google Scholar, 12Garcia-Guzman M. Soto F. Laube B. Stuhmer W. FEBS Lett. 1996; 388: 123-127Crossref PubMed Scopus (103) Google Scholar). Whereas homomeric P2X1 receptors desensitize strongly in the first seconds of agonist application, a slowly desensitizing response induced by 50 μm αβm-ATP was observed in oocytes co-injected with P2X1 and P2X5 subunits at a 1:1 cDNA molar ratio (Fig. 1). This hybrid phenotype was the unambiguous trademark of the expression of heteromeric P2X1+5 receptors. Oocytes expressing P2X1+5 receptors showed robust 50 μm αβm-ATP-induced whole-cell currents with amplitudes in the range of 3–15 μA at Vh = −50 mV after 2–5 days of post-injection time, similar to currents recorded from oocytes expressing P2X1 alone (Fig. 1). P2X1+5 receptors slowly desensitized during agonist application but showed complete recovery in 2 min (Fig.2), a noticeable difference with homomeric P2X5 receptors that do not desensitize in heterologous systems (Fig. 1) (11Collo G. North R.A. Kawashima E. Merlo-Pich E. Neidhart S. Surprenant A. Buell G. J. Neurosci. 1996; 16: 2495-2507Crossref PubMed Google Scholar, 12Garcia-Guzman M. Soto F. Laube B. Stuhmer W. FEBS Lett. 1996; 388: 123-127Crossref PubMed Scopus (103) Google Scholar). However, P2X1+5receptors (Fig. 2, B and D) recovered significantly faster than P2X1 receptors, the latter recovering less than 50% of their initial response after 5 min of washout (Fig. 2, A and C). We noticed slight differences in the rate of desensitization of P2X1+5receptors between oocytes (Fig. 2). These variations of phenotype could be because of the expression of populations of heteromeric channels with different stoichiometries, a cell-dependant variable that is not controlled in these experiments of co-injection. The kinetic properties of P2X2 receptors have been shown to be modulated by protein kinase A activity (19Chow Y.-W. Wang H.-L. J. Neurochem. 1998; 70: 2606-2612Crossref PubMed Scopus (42) Google Scholar). Thus it is possible that inter-individual differences in the levels of endogenous kinase activity present in oocytes could have some impact on the properties of desensitization of P2X1+5 receptors. Furthermore, the correlation between the number of P2X5 subunits and the kinetic properties of the oligomeric complex, which has been reported to be a trimer for homomeric P2X1 channels (20Nicke A. Bäumert H.G. Rettinger J. Eichele A. Lambrecht G. Mutschler E. Schmalzing G. EMBO J. 1998; 17: 3016-3028Crossref PubMed Scopus (479) Google Scholar), is not yet known. P2X1+5 receptors were challenged with ATP, αβm-ATP, and ADP at various concentrations for comparison with the pharmacology of homomeric P2X1 and P2X5 receptors. We measured EC50 values for P2X1+5 heteromers of 0.4 ± 0.2 μm for ATP, 1.1 ± 0.6 μm for αβm-ATP and 13 ± 4 μm for ADP (Fig.3). These EC50 values were not significantly different from those obtained with homomeric P2X1 receptors in the same experimental conditions: 0.7 ± 0.1 μm for ATP, 2.4 ± 1 μm for αβm-ATP, and 47 ± 9 μm for ADP (Fig. 3), in good agreement with previously published data (3Valera S. Hussy N. Evans R.J. Adami N. North R.A. Surprenant A. Buell G. Nature. 1994; 371: 516-519Crossref PubMed Scopus (894) Google Scholar). Differences in the apparent Hill coefficient n H(cooperativity index) of ADP activation between P2X1(n H = 4.9 ± 2.3) and P2X1+5(n H = 1.6 ± 0.8) (Fig. 3 C) could be because of the fact that we record from a heterogeneous population of P2X1-containing receptors with varying stoichiometries. The amplitudes of peak currents from P2X5-expressing oocytes were too small to carry out complete dose-response curve experiments with these agonists (Fig. 1). No significant differences were observed between P2X1+5 and P2X1 receptors during co-applications of extracellular zinc ions (10 μm), protons (pH 6.4), or alkaline solutions (pH 8.4) with sub-saturating concentrations of ATP (0.1 μm) (data not shown). Our results suggest that P2X1 subunits confer their high αβm-ATP sensitivity to the P2X1+5heteromers. Another specific pharmacological property of P2X1 subunits, the potent inhibitory effect of trinitrophenyl-ATP (TNP-ATP) (20Nicke A. Bäumert H.G. Rettinger J. Eichele A. Lambrecht G. Mutschler E. Schmalzing G. EMBO J. 1998; 17: 3016-3028Crossref PubMed Scopus (479) Google Scholar), is observed in the heteromeric receptors (Fig. 4 A). In conditions of co-application of TNP-application of TNP-ATP and αβm-ATP without pre-incubation, we measured an IC50 of 64 ± 14 nm on P2X1+5 and 200 ± 120 nm on homomeric P2X1receptors (Fig.4 B). This subunit association is therefore reminiscent of the association between P2X2 and P2X3 in which P2X3 is the pharmacologically dominant component both for αβm-ATP sensitivity (5Lewis C. Neidhart S. Holy C. North R.A. Buell G. Surprenant A. Nature. 1995; 377: 432-435Crossref PubMed Scopus (889) Google Scholar, 17Radford K.M. Virginio C. Surprenant A. North R.A. Kawashima E. J. Neurosci. 1997; 17: 6529-6533Crossref PubMed Google Scholar) and blockade by TNP-ATP (21Virginio C. Robertson G. Surprenant A. North R.A. Mol. Pharmacol. 1998; 53: 969-973PubMed Google Scholar, 22Thomas S. Virginio C. North R.A. Surprenant A. J. Physiol. 1998; 509: 411-417Crossref PubMed Scopus (94) Google Scholar).Figure 4Potent blockade of P2X1and P2X1+5 receptors-mediated responses by the antagonist TNP-ATP . A, representative P2X1+5 currents in conditions of inhibition. B, sensitivity of P2X1 and P2X1+5 responses to TNP-ATP, co-applied with 1 μm αβm-ATP. Peak currents were normalized to the response elicited by application of 10 μm αβm-ATP alone (mean ± S.E. from 5 to 8 oocytes per point). Membrane potentials were held at −100 mV.View Large Image Figure ViewerDownload Hi-res image Download (PPT) To demonstrate direct associations between P2X1 and P2X5 subunits that underlie their assembly in hybrid heteromers, we assayed their physical interaction by co-purification of epitope-tagged subunits in transfected HEK-293A cells. Purification of P2X5-His6 on nickel-binding resin in nondenaturing conditions (see "Experimental Procedures" for details) allowed the detection of co-transfected P2X1-Flag in Western blots (Fig. 5, lane C). Reciprocally, P2X1-His6 was shown to co-assemble with P2X5-Flag (Fig. 5, lane D). Positive controls included pseudo-homomeric receptors composed of P2X1-His6 + P2X1-Flag or P2X5-His6 + P2X5-Flag (Fig. 5,lanes A and B). Technical controls of transfections with one P2X subunit only or with sham-transfected HEK-293A cells were negative (data not shown). Peripheral sensory neurons have been reported to express ATP-gated channels with a slow rate of desensitization and a high sensitivity to αβm-ATP characterized by EC50 in the low micromolar range (Ref. 5Lewis C. Neidhart S. Holy C. North R.A. Buell G. Surprenant A. Nature. 1995; 377: 432-435Crossref PubMed Scopus (889) Google Scholar, and references therein). This sensory phenotype was thought to be exclusively accounted for by the co-assembly of P2X2 and P2X3 subunits into heteromeric P2X2+3 receptors (5Lewis C. Neidhart S. Holy C. North R.A. Buell G. Surprenant A. Nature. 1995; 377: 432-435Crossref PubMed Scopus (889) Google Scholar, 17Radford K.M. Virginio C. Surprenant A. North R.A. Kawashima E. J. Neurosci. 1997; 17: 6529-6533Crossref PubMed Google Scholar). Alternatively, we propose from our results that slowly desensitizing and αβm-ATP-elicited responses could be mediated by hybrid P2X1+5 heteromeric receptors endowed with the pharmacology of P2X1 and the kinetics of P2X5. Our data suggest to use TNP-ATP as a specific antagonist of P2X1-containing ATP-gated channels. In spinal motoneurons where P2X3 is absent, complete blockade of slowly desensitizing P2X responses by 1 μmTNP-ATP would indicate the expression of P2X1+5 heteromeric channels. Using subunit-specific polyclonal antibodies, Vulchanova et al. (23Vulchanova L. Arvidsson U. Rield M. Wang J. Buell G.N. Surprenant A. North R.A. Elde R. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 8063-8067Crossref PubMed Scopus (365) Google Scholar) described a strong P2X1 immunoreactivity in the laminae I-II of spinal cord, corresponding to presynaptic labeling of central axon terminals from dorsal root ganglia sensory neurons. As P2X2 and P2X3 subunits do not appear to co-assemble in heteromeric channels in these primary afferents (16Vulchanova L. Riedl M.S. Shuster S.J. Buell G. Surprenant A. North R.A. Elde R. Neuropharmacology. 1997; 36: 1229-1242Crossref PubMed Scopus (377) Google Scholar), a presynaptic localization of P2X1+5 receptors would provide sensory axon terminals with high sensitivity to ATP and slowly desensitizing voltage-independent calcium entry that could play a modulatory role in the release of central neurotransmitters glutamate or substance P (24Gu J.G. MacDermott A.B. Nature. 1997; 389: 749-753Crossref PubMed Scopus (420) Google Scholar). The effects of presynaptic P2X1+5receptors on the release of sensory transmitters can now be experimentally challenged with application of the blocker TNP-ATP at low concentrations. In the central nervous system, an important role for purines in motor systems is deduced both from the distribution of several P2X subunits mRNA within cranial and spinal motor nuclei (11Collo G. North R.A. Kawashima E. Merlo-Pich E. Neidhart S. Surprenant A. Buell G. J. Neurosci. 1996; 16: 2495-2507Crossref PubMed Google Scholar) and from the powerful cellular effects of extracellular ATP on motor outflow (25Funk G.D. Kanjhan R. Walsh C. Lipski J. Comer A.M. Parkis M.A. Housley G.D. J. Neurosci. 1997; 17: 6325-6337Crossref PubMed Google Scholar). More specifically, a subset of large projection motoneurons in lamina IX of rat spinal cord has been characterized by the co-expression of P2X1 and P2X5 subunits (11Collo G. North R.A. Kawashima E. Merlo-Pich E. Neidhart S. Surprenant A. Buell G. J. Neurosci. 1996; 16: 2495-2507Crossref PubMed Google Scholar). We propose from their functional properties that highly agonist-sensitive P2X1+5 receptors might provide a specific excitatory function to the motor control by allowing a sustained entry of extracellular calcium within motoneurons in response to minute amounts of released ATP. We gratefully acknowledge Kazimierz Babinski for the cloning of the rat P2X5 receptor subunit.
Referência(s)