Skip to main content
NCBI home page
British Journal of Pharmacology logoLink to British Journal of Pharmacology
. 2003 Oct 27;140(6):1027–1034. doi: 10.1038/sj.bjp.0705531

ATP analogues with modified phosphate chains and their selectivity for rat P2X2 and P2X2/3 receptors

Valeria Spelta 1,3, Abdelaziz Mekhalfia 2, Dominik Rejman 2, Mark Thompson 2, G Michael Blackburn 2, R Alan North 1,*
PMCID: PMC1574118  PMID: 14581175

Abstract

  1. Heteromeric P2X2/3 receptors are much more sensitive than homomeric P2X2 receptors to αβ-methylene-ATP, and this ATP analogue is widely used to discriminate the two receptors on sensory neurons and other cells.

  2. We sought to determine the structural basis for this selectivity by synthesising ADP and ATP analogues in which the αβ and/or βγ oxygen atoms were replaced by other moieties (including –CH2–, –CHF–, –CHCl–, –CHBr–, –CF2–, –CCl2–, –CBr2–, –CHSO3–, –CHPO3–, –CFPO3–, –CClPO3–, –CH2–CH2–, –C≡C–, –NH–, –CHCOOH–).

  3. We tested their actions as agonists or antagonists by whole-cell recording from human embryonic kidney cells expressing P2X2 subunits alone (homomeric P2X2 receptors), or cells expressing both P2X2 and P2X3 subunits, in which the current through heteromeric P2X2/3 receptors was isolated.

  4. ADP analogues had no agonist or antagonist effect at either P2X2 or P2X2/3 receptors. All the ATP analogues tested were without agonist or antagonist activity at homomeric P2X2 receptors, except βγ-difluoromethylene-ATP, which was a weak agonist.

  5. At P2X2/3 receptors, βγ-imido-ATP, βγ-methylene-ATP, and βγ-acetylene-ATP were weak agonists, whereas αβ,βγ- and βγ,γδ-bismethylene-AP4 were potent full agonists. βγ-Carboxymethylene-ATP and βγ-chlorophosphonomethylene-ATP were weak antagonists at P2X2/3 receptors (IC50 about 10 μM).

  6. The results indicate (a) that the homomeric P2X2 receptor presents very stringent structural requirements with respect to its activation by ATP; (b) that the heteromeric P2X2/3 receptor is much more tolerant of αβ and βγ substitution; and (c) that a P2X2/3-selective antagonist can be obtained by introduction of additional negativity at the βγ-methylene.

Keywords: P2X receptors, electrophysiology, phosphate chain

Introduction

The P2X3 receptor subunit is of particular interest because its expression is largely restricted to primary afferent neurones (Chen et al., 1995; Lewis et al., 1995). Immunohistochemistry with antibodies directed against a C-terminal epitope showed that the subunit was expressed by a subset of cells in the nodose and dorsal root ganglia, as well as on their central and peripheral terminals (Vulchanova et al., 1997,1998; Bradbury et al., 1998; see the review by Dunn et al., 2001). This subset has small diameter cell bodies and also expresses the receptor for the lectin B4. On the basis of this distribution, and earlier evidence that ATP could elicit pain (Bleehen & Keele, 1977), the P2X3 subunit was considered likely to play a key role in the signalling of pain sensation (reviewed by Dunn et al., 2001; see North, 2002). The breeding of mice with a disrupted P2X3 subunit gene has substantiated the role of the receptor in some forms of pain, but also indicated a wider role in primary afferent signalling (Cockayne et al., 2000; Souslova et al., 2000). In particular, the P2X3 subunit is critical for the primary afferent signals arising from bladder distension, where ATP released from urothelium appears to be the initiating stimulus (Ferguson et al., 1997; Vlaskovska et al., 2001).

P2X receptors form as multimers of several, perhaps three, subunits (reviewed by North, 2002). Although the composition of the receptors in native tissues is not known with any certainty, there is ample evidence that a heteromeric channel containing P2X2 and P2X3 subunits is abundant in primary afferent neurons. This comes from electrophysiological experiments, in which the effects of ATP are compared with those of its analogue αβ-methylene-ATP (αβmeATP) (Figure 1). In most small/medium-sized primary afferent neurons from dorsal root ganglia and trigeminal ganglia, and all from the nodose ganglia, both ATP and αβmeATP elicit an inward current that shows little desensitisation during a few seconds (Cook et al., 1997; Rae et al., 1998; Thomas et al., 1998; Virginio et al., 1998b; Grubb & Evans, 1999; Li et al., 1999; Ueno et al., 1999; Burgard et al., 2000; Labrakakis et al., 2000; Liu et al., 2001). This corresponds to the properties observed when P2X2 and P2X3 subunits are coexpressed, and not to the properties of the currents when either P2X2 or P2X3 subunits are expressed separately (Lewis et al., 1995). The supporting observations are the findings that many small dorsal root ganglion cells, and all nodose ganglion cells, express immunoreactivity for both P2X2 and P2X3 subunits, and the finding that P2X2 and P2X3 subunits will coimmunoprecipitate in heterologous expression systems (Radford et al., 1997; Torres et al., 1999). Moreover, the finding that the current is sustained at the heteromeric P2X2/3 receptor, but desensitises rapidly (100 ms) at the homomeric P2X3 receptor, leads one to argue that the signal to the heteromeric P2X2/3 receptor is likely to be more important in functional terms.

Figure 1.

Figure 1

Open in a new tab

Nomenclature used for the ATP analogues (see Table 1). ATP: X=Y=Z=O. αβmeATP: X=−CH2−, Y=Z=O.

Receptors containing P2X3 subunits can also be differentiated from other forms with the antagonist 2′,3′-O-(2,4,6-trinitrophenol)-ATP (TNP-ATP) (Virginio et al., 1998b; Burgard et al., 2000; North & Surprenant, 2000). This compound has nanomolar affinity as a blocker of P2X3 and P2X2/3 receptors, but blocks homomeric P2X2 receptors only at concentrations some 1000 times higher. TNP-ATP has been used to demonstrate that individual sensory neurons in the nodose ganglion express more than one kind of P2X receptor (Thomas et al., 1998). When αβmeATP is used as agonist, the TNP-ATP inhibition curve is monophasic, and the concentration causing half-maximal inhibition (IC50) is the same as that for heterologously expressed P2X2/3 heteromers. But when ATP is used as the agonist, the inhibition curve is biphasic; the two IC50 values correspond well to those for heteromeric P2X2/3 channels and for homomeric P2X2 channels. TNP-ATP has also been used to provide further evidence for involvement of a P2X3 subunit-containing receptor in visceral pain sensation. The abdominal constriction induced in mice by intraperitoneal acetic acid was inhibited by TNP-ATP; on a molar basis, TNP-ATP was comparable to morphine (Honore et al., 2002). However, TNP-ATP has certain drawbacks. First, it blocks P2X1 receptors almost as effectively as P2X3 or P2X2/3 receptors. Secondly, it seems not to work well in some intact tissues, perhaps because of rapid degradation (Lewis et al., 1998). On the other hand, the metabolically stable antagonist of P2X3-containing subunits recently reported by Jarvis et al. (2002) (A-317491: S-5-[(3-phenoxy-benzyl)-(1,2,3,4-tetra-hydro-naphthalen-1-yl)-carbamoyl-benzene-1,2,4-tricarboxylic acid) is effective in several models of chronic pain. Both these classes of P2X3 receptor blockers retain unmodified phosphate chains.

Although ATP activates all P2X receptor subunits, αβmeATP is effective only at those containing a P2X1 or P2X3 subunit. This discrimination on the basis of a single –O– to –CH2– substitution suggested that it may be possible to develop ATP analogues with modified phosphate chains that might be useful to separate activity at the P2X2 and P2X2/3 subunits. In the present study, we have synthesised several such analogues and compared their activity as agonists and antagonists at heterologously expressed homomeric P2X2 and heteromeric P2X2/3 receptors.

Methods

Synthesis of ATP analogues

αβ-substituted analogues of ADP were generally prepared using the method of Poulter (Davisson et al., 1987). They were phosphorylated to give αβ-substituted analogues of ATP using either p-nitrobenzyl phosphoromorpholidate (Blackburn & Langston, 1991) or equivalent methods. αβ,βγ-Bis-substituted analogues of ATP were also prepared using the Poulter methodology with appropriate bis-substituted analogues of tripolyphosphate (Davisson et al., 1987). βγ-substituted ATP analogues were prepared by condensation of AMP with the appropriate analogues of pyrophosphate using either diphenyl phosphorochloridate or AMP morpholidate (Blackburn et al., 1984,1985; Liu et al., 1999). αβ,βγ-Bis-substituted analogues of adenosine 5′-tetraphosphate (AP4) were prepared from the corresponding ATP analogues by phosphorylation with either p-nitrobenzyl phosphoromorpholidate or inorganic phosphate and carbonyl di-imidazole (Blackburn, unpublished results). βγ,γδ-Bis-substituted analogues of AP4 were prepared by condensation of AMP morpholidate with the corresponding analogues of tripolyphosphate (Blackburn, unpublished results). Adenosine 5′-phosphoromorpholidate was also condensed with phosphonoacetic acid and with β-phosphono-α-alanine to provide adenylyl phosphonoacetic acid and adenylyl β-phosphono-α-alanine, respectively (Blackburn, unpublished results). The final products were purified by ion-exchange medium pressure liquid chromatography. Nucleotides were isolated as their triethylammonium salts and converted into their sodium salts for use by ion exchange. Purity was uniformly in excess of 98%, as gauged by proton and phosphorus nuclear magnetic resonance and by electrospray mass spectrometry. The identity of all compounds was established by high-resolution mass spectrometry and by phosphorus and proton nuclear magnetic resonance. ATP, αβmeATP, βγ-me-D-ATP and βγ-me-L-ATP were purchased from Sigma.

Cell culture and electrophysiological recording

HEK293 cells stably expressing the rat P2X2 and rat P2X2/3 receptors were used. Generation of stable cell lines has been previously described (Evans et al., 1995; Kawashima et al., 1998). Cells were plated onto 13 mm glass coverslips and maintained in Dulbecco's modified Eagle's medium, supplemented with 10% heat-inactivated fetal calf serum and 2 mM L-glutamine at 37°C in a humidified 5% CO2 incubator. Culture media, sera and all cell culture reagents were obtained from Life Technologies (Paisley, U.K.)

Whole-cell recordings were performed at room temperature 24–72 h after the passage of stable cell lines, using an EPC9 patch-clamp amplifier (HEKA Elektronik, Lambrecht, Germany). Experiments were performed at a holding potential of −60 mV. Patch electrodes had resistances of 4–7 MΩ and were filled with (mM): 145 NaCl or NaF, 10 HEPES, 10 EGTA. The extracellular solution was (mM): 147 NaCl, 2 KCl, 1 MgCl2, 2 CaCl2, 10 HEPES and 13 glucose. Osmolarity and pH values of both solutions were maintained at 300–315 mOsm l−1 and 7.3, respectively. ATP and its analogues were applied by an array of glass inlet pipes using the RSC 200 (Biologic Science Instruments, Grenoble, France). ATP was used as the control agonist in all experiments on P2X2 receptors (EC90 30 μM); αβmeATP was used as the control agonist in all experiments on P2X2/3 receptors (EC50 5 μM). Agonists were applied for 3 s duration at 2 min interval, until evoked currents were stable (±5%); then an ATP analogue was applied to test for its agonist or antagonist activity. In the first case, the ATP analogue (at 100 μM) was applied for 3 s duration and the current evoked was measured. The peak current amplitude was expressed as a percentage of the amplitude obtained under the control agonist. Cumulative concentration–response curves were constructed by applying sequentially increasing concentrations at 2 min intervals. All results are expressed as the percentage response to a maximal concentration of control agonist, which was applied 2 min after the last ATP analogue concentration. The protocol for testing for antagonist activity was as follows. When the agonist-induced current amplitude was stable, the ATP analogue was applied for 30 s prior to the agonist application (for screening) or only during the agonist application (for more detailed dose–response studies). The peak current amplitude was expressed as a percentage of the amplitude obtained under control conditions. Antagonist concentration–inhibition curves were obtained using a fixed agonist concentration (αβmeATP 5 μM) and increasing antagonist concentrations.

Data analysis

Agonist concentration–current curves were fitted with E=([A]n/([A]n+EC50n)), where E is the current as a fraction of the current evoked by a maximal concentration of the control agonist, EC50 is the concentration of agonist required for a half-maximal response and n is the Hill coefficient. Inhibition curves were fitted with I/I0=1/(1+(IC50/[B])−n), where I and I0 are peak currents in the presence and absence of the antagonist (concentration [B]), IC50 is the concentration of antagonist required to block the control current by 50% and n is the Hill coefficient. The data in the figures are presented as mean±s.e.m. from individual cells and the pooled data were fitted using Kaleidagraph (Synergy Software, Reading, PA, U.S.A.).

Results

In all, 22 ATP and 12 ADP derivatives with modified triphosphate and diphosphate chains were synthesised and tested. The oxygen bridge in the αβ, βγ and/or γδ position (Figure 1) was replaced with halomethylene, acetylene, imido groups and branched methylene groups bearing a negatively charged function, or more bulky substituents (Table 1 ).

Table 1.

Structures of ATP and ADP analogues used

Compound X Y Z Name P2X2 P2X2/3
ATP O O OH ATP Agonist Agonist
αβmeATP CH2 O OH αβ-Methylene-ATP None Agonist (EC50≈3 μM)
1 CF2 O OH αβ-Difluoromethylene-ATP None Weak agonist (≈100 μM)3
2 CBr2 O OH αβ-Dibromomethylene-ATP None Very weak agonist2
3 O CH2 OH βγ-Methylene-L-ATP None Very weak agonist2
4 O CH2 OH βγ-Methylene-D-ATP None Weak agonist (≈200 μM)3
5 O NH OH βγ-Imido-ATP None Weak agonist (≈40 μM)3
6 O C≡C OH βγ-Acetylene-ATP None Weak agonist (≈30 μM)3
7 O CH2CH2 OH βγ-Ethylene-ATP None None
8 O CHBr OH βγ-Bromomethylene-ATP None Very weak agonist2
9 O CHCl OH βγ-Chloromethylene-ATP None Very weak agonist2
10 O CF2 OH βγ-Difluoromethylene-ATP Weak agonist Weak agonist (≈40 μM)3
11 O CBr2 OH βγ-Dibromomethhylene-ATP None None
12 O CCl2 OH βγ-Dichloromethylene-ATP None None
13 O CHCOOH OH βγ-Carboxymethylene-ATP None Antagonist
14 O CHSO3H OH βγ-Sulphonomethylene-ATP None None
15 O CHPO3H2 OH βγ-Phosphonomethylene-ATP None Very weak agonist2
16 O CFPO3H2 OH βγ-Flurophosphonomethylene-ATP None Very weak agonist2
17 O CClPO3H2 OH βγ-Chlorophosphonomethylene-ATP None Antagonist
18 CH2 CH2 OH αβ,βγ-Bismethylene-ATP None Very weak agonist2
19 CCl2 CCl2 OH αβ,βγ-Bisdichloromethylene-ATP None Agonist
20 CH2 CH2 OPO3H2 αβ,βγ-Bismethylene-AP41 None Agonist
21 O CH2 CH2PO3H2 βγ,γδ-Bismethylene-AP41 None Agonist
22 O CCl2 CCl2PO3H2 βγ,γδ-Bisdichloromethylene-AP41 No None
ADP O OH   ADP None None
23 CHBr OH   αβ-Dibromomethylene-ADP None None
24 CBr2 OH   αβ-Dibromomethylene-ADP None None
25 CF2 OH   αβ-Difluoromethylene-ADP None None
26 CH2CH2 OH   αβ-Ethylene-ADP None None
27 CH=CH OH   αβ-Etheno-(E)-ADP None None
28 C≡C OH   αβ-Acetylene-ADP None None
29 CHSO3H OH   αβ-Sulphonomethylene-ADP None None
30 CHPO3H2 OH   αβ-Phosphonomethylene-ADP None None
31 CHCOOH OH   αβ-Carboxymethylene-ADP None None
32 CHCOOBz OH   αβ-Benzyloxycarbonylmethylene-ADP None None
33 O CH2COOH   CarboxymethylphosphonoAMP None None
34 O OCH2CH(NH2)COOH   P2-(O3-L-serinyl)-ADP None None
Open in a new tab
1

AP4=adenosine tetraphosphate.

2

Very weak agonist: analogue at 100 μM evokes a current that is <10% that caused by αβmeATP (5 μM).

3

Approximately EC50 for partial agonists.

ADP analogues

None of the ADP derivatives (compounds 23–34, tested at 100 μM; Table 1) had any agonist activity either at the P2X2 or P2X2/3 receptors (n=2–3 each). Likewise, none of these compounds reduced the current elicited by ATP at P2X2 receptors, or by αβmeATP at P2X2/3 receptors.

ATP analogues

At P2X2 receptors, βγ-difluoromethyleneATP (10) was the only compound tested that showed any agonist activity. It was a weak agonist, with the maximal response at 300 μM being about 50% of the maximal ATP effect. This compound (10)(EC50=72±2.6 μM, n=5) was also about 10-fold less potent than ATP (EC50=8.5±0.25 μM, n=4).

In contrast, eight ATP analogues showed agonist properties at the heteromeric P2X2/3 receptor. These were αβ-difluoro-methylene-ATP (1), βγ-methylene-D-ATP (4), βγ-imido-ATP (5), βγ-acetylene-ATP (6), βγ-difluoromethylene-ATP (10), αβ,βγ-bisdichloromethylene-ATP (19), αβ,βγ-bismethylene-AP4 (20) and βγ, γδ-bismethylene-AP4 (21); the last two can also be regarded as adenosine tetraphosphate analogues. The only triphosphate modification that retained full agonist activity was the replacement of both hydrogen atoms by chlorine at both the αβ and βγ bridges (αβ,βγ-bisdichloromethylene-ATP); however, this compound was still about 10-fold less potent than αβmeATP (EC50: 22±2.7 μM, n=8). The two methylene-substituted tetraphosphate analogues were also full agonists, and αβ,βγ-bismethylene-AP4 was the only compound examined that was more potent than αβmeATP (Figure 2). Compound 20 (EC50=1.8±0.3 μM, n=4) gave a larger peak current than αβmeATP, whereas compound 21 (EC50=11.8±0.7 μM, n=4) gave a peak current of amplitude similar to that of ATP (Figure 3). The other four compounds gave lower maximal currents, and were less potent than αβmeATP. The EC50's (in μM, number of observations in parentheses) were αβmeATP: 3.3±0.4 (4); 4: 176±45 (6); 5: 36.4±4.6 (5) (Figure 3); 6: 31.2±1.4 (7) (Figure 3) and 10: 41.6±1.6 (6).

Figure 2.

Figure 2

Open in a new tab

Full agonists at PX2/3 receptors. Currents evoked by αβ,βγ-bisdichloromethylene-ATP (19), αβ,βγ-bismethylene-AP4 (20) and βγ,γδ-bismethylene-AP4 (21) at P2X2/3 receptors. (a) Representative current traces; period of agonist application indicated by bars above traces. (b) Concentration–response curves. Currents are plotted as the per cent of that evoked by 30 μM αβmeATP.

Figure 3.

Figure 3

Open in a new tab

Weak agonists at P2X2/3 receptors. Currents evoked by D-βγ-methylene-ATP (4), βγ-imido-ATP (5) and βγ-acetylene-ATP (6) at P2X2/3 receptors. (a) Representative current traces; period of agonist application indicated by bars above traces. (b) Concentration–response curves. Currents are plotted as the per cent of that evoked by 30 μM αβmeATP.

Cells transfected with both P2X2 and P2X3 plasmids express both homomeric P2X2 and heteromeric P2X2/3 receptors. As βγ-difluoromethylene-ATP (10) also had agonist activity at homomeric P2X2 receptors, it is not possible to say whether it has an additional action at heteromeric P2X2/3 receptors. The response to ATP of homomeric P2X2 receptors is almost fully blocked by increasing the extracellular calcium concentration to 10 mM (Virginio et al., 1998a). We found that the current evoked by βγ-difluoromethylene-ATP (10) was similarly blocked by 10 mM calcium at cells expressing homomeric P2X2 receptors (results not shown), but it was not affected by 10 mM calcium in the doubly transfected cells (P2X2 and P2X3); this implies that βγ-difluoromethylene-ATP (10) is an agonist also at the heteromeric P2X2/3 receptor (Figure 4).

Figure 4.

Figure 4

Open in a new tab

Difluoromethylene-substituted ATP analogues are agonists at both P2X2 and P2X2/3 receptors. (a) Cells transfected with P2X2 and P2X3 subunits. Currents evoked by αβ-difluoromethylene-ATP (100 μM)(left) and βγ-difluoromethyleneATP (100 μM)(right). The currents are indicated by open arrows. These are superimposed on currents evoked by αβmeATP and ATP in the same cells. Left, αβmeATP (5 μM). Right, αβmeATP (30 μM) and ATP (30 μM). (b) Cells transfected only with P2X2 subunits. βγ-DifluoromethyleneATP is an agonist, though much less potent than ATP. (c) Cells transfected with P2X2 and P2X3 subunits. The solution contained 10 mM calcium to inhibit currents at homomeric P2X2 receptors. βγ-difluoromethyleneATP showed agonist activity, which must therefore be at P2X2/3 receptors.

At P2X2/3 receptors, βγ-carboxymethylene-ATP (13) and βγ-chlorophosphonomethylene-ATP (17) caused concentration-dependent inhibitions of the αβmeATP-evoked currents (Figure 5). Compound 17 (IC50=14.4±12.8 μM, n=2–4) was about equipotent with 13 (IC50=16.7±1.2 μM, n=4–6). None of the ATP derivatives displayed any antagonist activity on cells expressing P2X2 receptors.

Figure 5.

Figure 5

Open in a new tab

Antagonists at P2X2/3 receptors. Inhibition of αβmeATP-evoked current by βγ-carboxymethylene-ATP and βγ-chlorophosphonomethylene-ATP. (a) Representative current traces; the currents shown were evoked by applying meATP in the presence of increasing concentrations of the antagonist. The first two traces show the control response in the absence of antagonist. (b) Antagonist concentration–inhibition curves. Currents are plotted as the per cent of that evoked by 30 μM meATP in the absence of antagonist.

Discussion

The results with ATP, ADP and αβmeATP confirm a body of work on these compounds at P2X2 and P2X2/3 receptors (North, 2002). At reasonable concentrations (<100 μM), we can say that ATP activates both, ADP activates neither, and αβmeATP is selective for P2X2/3 receptors. In other words, P2X3 subunits are tolerant of methylene in place of oxygen in the αβ position, whereas P2X2 subunits are not. The complete lack of activity of any ADP analogue at either receptor (Table 1) confirms the essential requirement for a triphosphate as distinct from a diphosphate chain.

With respect to the homomeric P2X2 receptor, it is remarkable that virtually every modification resulted in complete loss of activity. The only exception was the addition of two fluorine atoms onto the βγ carbon, which gave a compound (10) with weak partial agonist action; introduction of larger chlorine and bromine at this position was not tolerated. This finding indicates a key requirement for compactness and/or electronegativity at this position. Presumably, the βγ oxygen atom itself plays a critical role through hydrogen bonding to residues in the binding pocket of the P2X2 subunit. This is also the case for the case of the αβ oxygen, where substitution by methylene results in a complete loss of activity.

Although the P2X2/3 receptor was more tolerant than the P2X2 receptor of replacement of the phosphate chain oxygen atoms, no compounds were found with agonist activity as large as αβmeATP (a direct comparison with ATP itself is difficult because ATP might activate both homomeric P2X2 and heteromeric P2X2/3 on the same cells). At the αβ position, activity was reduced by introduction of fluorine atoms, and completely abolished by the larger halogen bromine. At the βγ position, the most effective analogues (but still about 10 times higher EC50 than αβmeATP) were those that only modestly increased the size at this position (5: –NH–; 6: –C≡C–; or 7: –CF2–), whereas larger substitutions caused complete loss of activity (8: –CHCl–; 9: CHBr–; 11: CBr2–; 12: –CCl2–). The stereochemistry of the ribose moiety is obviously critical for activity; the D-isomer of βγmeATP was weakly able to induce currents in cells expressing the P2X2/3 receptor (EC50 about 200 μM; Figure 2), whereas its L-enantiomer was essentially without activity (Table 1).

Replacement of both oxygen atoms by methylene reveals interactive effects. αβ-Methylene-ATP is fully active as an agonist, βγ-methylene-ATP (4) is about 100 × less effective, and the doubly substituted αβ,βγ-bismethylene-ATP (18) is essentially without activity. However, simple extension of the phosphate chain restores activity to αβ,βγ-bismethylene-ATP (20); indeed, the corresponding AP4 analogue is the only compound found that was more potent than αβmeATP itself (Figure 3). The oxygen atom in the γδ position presumably contributes considerably to this, because βγ,γδ-bismethylene-AP4 (21) was some ten-fold less effective than the αβ,βγ substituted analogue. This could be because the partial charge on the oxygen atom of the γδ bridge can substitute to a degree for that of the –OH group of the γ phosphate that normally contributes to the binding site. We do not understand why the introduction of chlorine atoms into the βγ,γδ-bis-substituted analogue reduces activity, whereas the same changes in the αβ,βγ-bis-substituted compounds increase activity.

The most useful compounds with respect to further elucidation of the function of sensory P2X2/3 receptors will be selective antagonists. Our results indicate that it may be possible to find such compounds by relatively minor modification of the phosphate chain. Only two of the analogues tested had antagonist activity, and these inhibited the current evoked by αβmeATP with IC50 of about 10 μM. The antagonism was quickly reversible. The common feature of the two antagonists (βγ-carboxymethylene-ATP (13) and βγ-chlorophosphonomethylene-ATP (17); Table 1; Figure 4) is the additional negative charge on the βγ methylene bridge. However, this ‘supercharging' feature alone is not the full explanation, because the antagonist properties of βγ-chlorophosphonomethylene-ATP were not shared by βγ-fluorophosphonomethylene and βγ-phosphonomethylene (Table 1). Presumably, the chlorine atom contributes a steric effect that favourably positions the carboxylate, and that is not seen with fluorine or hydrogen. The effectiveness of the carboxylate attachment at the βγ-bridge suggests key positive charges at the binding site; the pattern of negative moieties is a common feature of other competitive P2X2/3 receptor antagonists suramin, pyridoxal-5-phosphate-6-azo-2′4′-disulphonic acid and TNP-ATP (Spelta et al., 2002). The same features are observed in a recently described non-nucleotide receptor antagonist at P2X3 and P2X2/3 receptors (A-317491); this is a benzene ring bearing three carboxylates and one substituted carboxamide (Jarvis et al., 2002).

It would be desirable to be able to interpret the present results in the context of the structure of the ATP-binding site on the P2X receptor protein. Unfortunately, there is no relatedness in the amino-acid sequence of the P2X receptor and any other protein. Although some regions of the protein have been suggested to be involved in ATP binding (e.g. the conserved K–X–K–G beginning at Lys69 in the P2X2 subunit; Jiang et al., 2000), none of the ‘classical' nucleotide-binding motifs have been detected (Traut, 1994; Ramakrishnan et al., 2002). Among proteins that bind but do not hydrolyse ATP, there are several structures that show hydrogen bond formation between the bridging oxygen atoms and residues in the binding pocket (e.g. αβ oxygen with NH3+ in ATP:co(I)rrinoid adenyltransferase; Bauer et al. (2001), and with oxygens of the C-terminus of the S1 subunit in pertussis toxin, Hazes et al., 1996). The present finding would certainly be consistent with a key role for such an interaction in the binding of ATP to P2X receptors, in which case the residue(s) involved might be expected to differ between the P2X1 and P2X3 subunits (where αβmeATP is effective) and the P2X2 and other subunits (where it is not).

There have been only limited previous studies on the effects of phosphate chain-modified ATP analogues at P2 receptors. P2Y receptors, in general, are not activated or blocked by αβmeATP, βγmeATP and βγimidoATP (Jacobson et al., 1999; Jacobson & Knutsen, 2001). We have found that replacement of the anhydride βγ oxygen with a difluoromethylene unit retained activity at the P2X2/3 receptor. This compound has previously been shown to be a partial agonist in the rabbit ear artery (Leff et al., 1993), where it presumably activates P2X1 receptors. The βγ-difluoromethylene substitution is a feature of ARL67156 (6-N,N-diethyl-βγ-dibromomethylene-D-ATP), which is widely used as a blocker of ectoATPase. Futhermore, ARL67085 is a βγ-dichloromethylene derivative (2-propylthio-D-βγ-dichloromethylene-ATP), which acts as an antagonist with good selectivity for the platelet P2Y receptor (P2Y12, formerly P2T) (Ingall et al., 1999).

Acknowledgments

We thank Annmarie Surprenant and Lin-Hua Jiang for advice, and Daniele Estoppey for cell culture. This work was supported by The Wellcome Trust.

Abbreviations

AP4

adenosine 5′-tetraphosphate

PPADS

pyridoxal-5-phosphate-6-azo-2′,4′-disulphonic acid derivative

TNP-ATP

2′,3′-O-(2,4,6-trinitrophenol)-ATP

αβmeATP

αβ-methylene-ATP

References

  1. BAUER C.B., FONSECA M.V., HOLDEN H.M., THODEN J.B., THOMPSON T.B., ESCALANTE-SEMERENA J.C., RAYMENT I. Three-dimensional structure of ATP:corrinoid adenosyltransferase from Salmonella typhimurium in its free state, complexed with MgATP, or complexed with hydroxycobalamin and MgATP. Biochemistry. 2001;40:361–374. doi: 10.1021/bi002145o. [DOI] [PubMed] [Google Scholar]
  2. BLACKBURN G.M., ECKSTEIN F., KENT D.E., PERRÉE T.D. Isopolar vs isosteric phosphonate analogues of nucleotides. Nucleosides Nucleotides. 1985;4:165–167. [Google Scholar]
  3. BLACKBURN G.M., LANGSTON S.P. Novel P1,P2-substituted phosphonate analogues of 2′-deoxyadenosine and thymidine 5′-triphosphates. Tetrahedron Lett. 1991;32:6425–6428. [Google Scholar]
  4. BLACKBURN G.M., KENT D.E., KOLKMANN F. The synthesis and metal binding characteristics of novel, isopolar phosphonate analogues of nucleotides. J. Chem. Soc. Perkin Trans. 1984;15:1119–1125. [Google Scholar]
  5. BLEEHEN T., KEELE C.A. Observations on the algogenic actions of adenosine compounds on the human blister base. Pain. 1977;3:365–371. doi: 10.1016/0304-3959(77)90066-5. [DOI] [PubMed] [Google Scholar]
  6. BRADBURY E.J., BURNSTOCK G., MCMAHON S.B. The expression of P2X3 purinoreceptors in sensory neurons: effects of axotomy and glial-derived neurotrophic factor. Mol. Cell. Neurosci. 1998;12:256–268. doi: 10.1006/mcne.1998.0719. [DOI] [PubMed] [Google Scholar]
  7. BURGARD E.C., NIFORATOS W., VAN BIESEN T., LYNCH K.J., KAGE K.L., TOUMA E., KOWALUK E.A., JARVIS M.F. Competitive antagonism of recombinant P2X2/3 receptors by 2′, 3′-O-(2,4,6-trinitrophenyl) adenosine 5′-triphosphate (TNP-ATP) Mol. Pharmacol. 2000;58:1502–1510. doi: 10.1124/mol.58.6.1502. [DOI] [PubMed] [Google Scholar]
  8. CHEN C.-C., AKOPIAN A.N., SIVILOTTI L., COLQUHOUN D., BURNSTOCK G., WOOD J.N. A P2X purinoceptor expressed by a subset of sensory neurons. Nature. 1995;377:428–431. doi: 10.1038/377428a0. [DOI] [PubMed] [Google Scholar]
  9. COCKAYNE D.A., HAMILTON S.G., ZHU Q.M., DUNN P.M., ZHONG Y., NOVAKOVICS S., MALMBERG A.B., CAIN G., BERSON A., KASSOTAKIS L., HEDLEY L., LACHNIT W.G., BURNSTOCK G., MCMAHON S.B., FORD A.P. Urinary bladder hyporeflexia and reduced pain-related behaviour in P2X3-deficient mice. Nature. 2000;407:1011–1015. doi: 10.1038/35039519. [DOI] [PubMed] [Google Scholar]
  10. COOK S.P., VULCHANOVA L., HARGREAVES K.M., ELDE R., MCCLESKEY E.W. Distinct ATP receptors on pain-sensing and stretch-sensing neurons. Nature. 1997;387:505–508. doi: 10.1038/387505a0. [DOI] [PubMed] [Google Scholar]
  11. DAVISSON V.J., DAVIS D.R., DIXIT V.M., POULTER C.D. Synthesis of nucleotide 5′-diphosphates from 5′-O-tosyl nucleosides. J. Org. Chem. 1987;52:1794–1801. [Google Scholar]
  12. DUNN P.M., ZHONG Y., BURNSTOCK G. P2X receptors in peripheral neurons. Prog. NeuroBiol. 2001;65:107–134. doi: 10.1016/s0301-0082(01)00005-3. [DOI] [PubMed] [Google Scholar]
  13. EVANS R.J., LEWIS C., BUELL G., VALERA S., NORTH R.A., SURPRENANT A. Pharmacological characterization of heterologously expressed ATP-gated cation channels (P2X-purinoceptors) Mol. Pharmacol. 1995;48:78–183. [PubMed] [Google Scholar]
  14. FERGUSON D.R., KENNEDY I., BURTON T.J. ATP is released from rabbit urinary bladder epithelial cells by hydrostatic pressure changes – a possible sensory mechanism. J. Physiol. 1997;505:503–511. doi: 10.1111/j.1469-7793.1997.503bb.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. GRUBB B.D., EVANS R.J. Characterization of dorsal root ganglion neuron P2X receptors. Eur. J. Neurosci. 1999;11:49–54. doi: 10.1046/j.1460-9568.1999.00426.x. [DOI] [PubMed] [Google Scholar]
  16. HAZES B., BOODHOO A., COCKLE S.A., READ R.J. Crystal structure of the pertussis toxin-ATP complex: a molecular sensor. J. Mol. Biol. 1996;258:661–671. doi: 10.1006/jmbi.1996.0277. [DOI] [PubMed] [Google Scholar]
  17. HONORE P., MIKUSA J., BIANCHI B., MCDONALD H., CARTMELL J., FALTYNEK C., JARVIS M.F. TNP-ATP, a potent P2X3 receptor antagonist, blocks acetic acid-induced abdominal constriction in mice: comparison with reference analgesics. Pain. 2002;96:99–105. doi: 10.1016/s0304-3959(01)00434-1. [DOI] [PubMed] [Google Scholar]
  18. INGALL A.H., DIXON J., BAILEY A., COOMBS M.E., COX D., MCINALLY J.I., HUNT S.F., KINDON N.D., TEOBALD B.J., WILLIS P.A., HUMPHRIES R.G., LEFF P., CLEGG J.A., SMITH J.A., TOMLINSON W. Antagonists of the platelet P2T receptor: a novel approach to antithrombotic therapy. J. Med. Chem. 1999;42:213–220. doi: 10.1021/jm981072s. [DOI] [PubMed] [Google Scholar]
  19. JACOBSON K.A., HOFFMANN C., KIM Y.-C., CAMAIONI E., NANDANAN E., JANG S.Y., GUO D.-P., JI X.-D., VON KUGELGENI., MORO S., ZIGANSHIN A.U., RYCHKOV A., KING B.F., BROWN S.G., WILDMAN S.S., BURNSTOCK G., BOYER J.L., MOHANRAM A., HARDEN T.K. Molecular recognition in P2 receptors: ligand development aided by molecular modelling and mutagenesis. Prog. Brain Res. 1999;120:119–134. doi: 10.1016/s0079-6123(08)63550-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. JACOBSON K.A., KNUTSEN L.J.S.P1 and P2 purine and pyrimidine receptor ligands Purinergic and Pyrimidinergic signalling I, Handbook of Experimental Pharmacology 2001Berlin: Springer; 129–175.ed. Abbracchio, M.P. & Williams, M. pp [Google Scholar]
  21. JARVIS M.F., BURGARD E.C., MCGARAUGHTY S., HONORE P., LYNCH K., BRENNAN T.J., SUBIETA A., VAN BIESEN T., CARTMELL J., BIANCHI B., NIFORATOS W., KAGE K., YU H., MIKUSA J., WISMER C.T., ZHU C.Z., CHU K., LEE C.H., STEWART A.O., POLAKOWSKI J., COX B.F., KOWALUK E., WILLIAMS M., SULLIVAN J., FALTYNEK C. A-317491, a novel potent and selective non-nucleotide antagonist of P2X3 and P2X2/3 receptors, reduces chronic inflammatory and neuropathic pain in the rat. Proc. Natl. Acad. Sci. U.S.A. 2002;99:17179–17184. doi: 10.1073/pnas.252537299. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. JIANG L.H., RASSENDREN F., SURPRENANT A., NORTH R.A. Identification of amino acid residues contributing to the ATP-binding site of a purinergic P2X receptor. J. Biol. Chem. 2000;275:34190–34196. doi: 10.1074/jbc.M005481200. [DOI] [PubMed] [Google Scholar]
  23. KAWASHIMA E., ESTOPPEY D., FAHMI D., VIRGINIO C., REES S., SURPRENANT A., NORTH R.A. A novel and efficient method for the stable expression of heteromeric ion channels in mammalian cells. Recept. Channels. 1998;5:53–60. [PubMed] [Google Scholar]
  24. LABRAKAKIS C., GERSTNER E., MACDERMOTT A.B. Adenosine triphosphate-evoked currents in cultured dorsal root ganglion neurons obtained from rat embryos: desensitization kinetics and modulation of glutamate release. Neuroscience. 2000;101:1117–1126. doi: 10.1016/s0306-4522(00)00373-0. [DOI] [PubMed] [Google Scholar]
  25. LEFF P., WOOD B.E., O'CONNOR S.E., MCKECHNIE K. Quantitative analysis of the agonist and antagonist actions of some ATP analogues at P2X-purinoceptors in the rabbit ear artery. Br. J. Pharmacol. 1993;108:490–496. doi: 10.1111/j.1476-5381.1993.tb12830.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. LEWIS C., NEIDHART S., HOLY C., NORTH R.A., BUELL G., SURPRENANT A. Coexpression of P2X2 and P2X3 receptor subunits can account for ATP-gated currents in sensory neurones. Nature. 1995;377:432–434. doi: 10.1038/377432a0. [DOI] [PubMed] [Google Scholar]
  27. LEWIS C.J., SURPRENANT A., EVANS R.J. 2′,3′-O-(2,4,6-trinitrophenyl) adenosine 5′-triphosphate (TNP-ATP) – a nanomolar antagonist at rat mesenteric artery P2X ion channels. Br. J. Pharmacol. 1998;124:1463–1466. doi: 10.1038/sj.bjp.0702001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. LI C., PEOPLES R.W., LANTHORN T.H., LI Z.-W., WEIGHT F.F. Distinct ATP-activated currents in different types of neurons dissociated from rat dorsal root ganglion. Neurosci. Lett. 1999;263:57–60. doi: 10.1016/s0304-3940(99)00114-7. [DOI] [PubMed] [Google Scholar]
  29. LIU M., KING B.F., DUNN P.M., RONG W., TOWNSEND-NICHOLSON A., BURNSTOCK G. Coexpression of P2X3 and P2X2 receptor subunits in varying amounts generates heterogeneous populations of P2X receptors that evoke a spectrum of agonist responses comparable to that seen in sensory neurons. J. Pharmacol. Exp. Ther. 2001;296:1043–1050. [PubMed] [Google Scholar]
  30. LIU X., BRENNER C., GURANOWSKI A., STARZYNSKA E., BLACKBURN G.M. New tripodal ‘supercharged' analogues of adenosine nucleotides: inhibitors for the Fhit Ap3A hydrolase. Angew. Chem. 1999;38:1244–1247. doi: 10.1002/(SICI)1521-3773(19990503)38:9<1244::AID-ANIE1244>3.0.CO;2-0. [DOI] [PubMed] [Google Scholar]
  31. NORTH R.A. The molecular physiology of P2X receptors. Physiol. Rev. 2002;82:1013–1067. doi: 10.1152/physrev.00015.2002. [DOI] [PubMed] [Google Scholar]
  32. NORTH R.A., SURPRENANT A. Pharmacology of P2X receptors. Annu. Rev. Pharmacol. Toxicol. 2000;40:563–580. doi: 10.1146/annurev.pharmtox.40.1.563. [DOI] [PubMed] [Google Scholar]
  33. RADFORD K.M., VIRGINIO C., SURPRENANT A., NORTH R.A., KAWASHIMA E. Baculovirus expression provides direct evidence for heteromeric assembly of P2X2 and P2X3 receptors. J. Neurosci. 1997;17:6529–6533. doi: 10.1523/JNEUROSCI.17-17-06529.1997. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. RAE M.G., ROWAN E.G., KENNEDY C. Pharmacological properties of P2X3-receptors present in neurones of the rat dorsal root ganglia. Br. J. Pharmacol. 1998;124:176–180. doi: 10.1038/sj.bjp.0701803. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. RAMAKRISHNAN C., DANI V.S., RAMASARMA T. A conformational analysis of Walker motif A [GXXXXGKT(S)] in nucleotide-binding and other proteins. Prot. Eng. 2002;15:783–798. doi: 10.1093/protein/15.10.783. [DOI] [PubMed] [Google Scholar]
  36. SPELTA V., JIANG L.-H., SURPRENANT A., NORTH R.A. Kinetics of antagonist actions at rat P2X2/3 heteromeric receptors. Br. J. Pharmacol. 2002;135:1524–1530. doi: 10.1038/sj.bjp.0704591. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. SOUSLOVA V., CESARE P., DING Y., AKOPIAN A.N., STANFA L., SUZUKI R., CARPENTER K., DICKENSON A., BOYCE S., HILL R., NEBENUIS-OOSTHUISEN D., SMITH A.J., KIDD E.J., WOOD J.N. Warm-coding deficits and aberrant inflammatory pain in mice lacking P2X3 receptors. Nature. 2000;407:1015–1017. doi: 10.1038/35039526. [DOI] [PubMed] [Google Scholar]
  38. THOMAS S., VIRGINIO C., NORTH R.A., SURPRENANT A. The antagonist trinitrophenyl-ATP reveals co-existence of distinct P2X receptor channels in rat nodose neurones. J. Physiol. 1998;509:411–417. doi: 10.1111/j.1469-7793.1998.411bn.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. TORRES G.E., EGAN T.M., VOIGT M.M. Hetero-oligomeric assembly of P2X receptor subunits. Specificities exist with regard to possible partners. J. Biol. Chem. 1999;274:6653–6659. doi: 10.1074/jbc.274.10.6653. [DOI] [PubMed] [Google Scholar]
  40. TRAUT T.W. The functions and consensus motifs of nine types of peptide segments that form different types of nucleotide-binding sites. Eur. J. BioChem. 1994;222:9–19. doi: 10.1111/j.1432-1033.1994.tb18835.x. [DOI] [PubMed] [Google Scholar]
  41. UENO S., TSUDA M., IWANAGA T., INOUE K. Cell type-specific ATP-activated responses in rat dorsal root ganglion neurons. Br. J. Pharmacol. 1999;126:429–436. doi: 10.1038/sj.bjp.0702319. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. VIRGINIO C., NORTH R.A., SURPRENANT A. Calcium permeability and block at homomeric and heteromeric P2X2 and P2X3 receptors, and receptors in rat nodose neurones. J. Physiol. 1998a;510:27–35. doi: 10.1111/j.1469-7793.1998.027bz.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. VIRGINIO C., ROBERTSON G., SURPRENANT A., NORTH R.A. Trinitrophenyl substituted nucleotides are potent antagonists selective for P2X1, P2X3 and heteromeric P2X2/3 receptors. Mol. Pharmacol. 1998b;53:969–973. [PubMed] [Google Scholar]
  44. VLASKOVSKA M., KASAKOV L., RONG W., BODIN P., BARDINI M., COCKAYNE D.A., FORD A.P., BURNSTOCK G. P2X3 knock-out mice reveal a major sensory role for urothelially released ATP. J. Neurosci. 2001;21:5670–5677. doi: 10.1523/JNEUROSCI.21-15-05670.2001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. VULCHANOVA L., RIEDL M.S., SHUSTER S.J., BUELL G., SURPRENANT A., NORTH R.A., ELDE R. Immunohistochemical study of the P2X2 and P2X3 receptor subunits in rat and monkey sensory neurons and their central terminals. Neuropharmacology. 1997;36:1229–1242. doi: 10.1016/s0028-3908(97)00126-3. [DOI] [PubMed] [Google Scholar]
  46. VULCHANOVA L., RIEDL M.S., SHUSTER S.J., STONE L.S., HARGREAVES K.M., BUELL G., SURPRENANT A., NORTH R.A., ELDE R. P2X3 is expressed by DRG neurons that terminate in inner lamina II. Eur. J. Neurosci. 1998;10:3470–3478. doi: 10.1046/j.1460-9568.1998.00355.x. [DOI] [PubMed] [Google Scholar]

Articles from British Journal of Pharmacology are provided here courtesy of The British Pharmacological Society

RESOURCES