P2Y receptor-mediated inhibition of voltage-activated Ca²⁺ currents in PC12 cells

Materials

Na-UTP, Na-UDP, Na₂-ATP, Na-ADP, Li₃-ADPβS, Li-GTP, Li₃-GDPβS, Li₄-GTPγS, reactive blue 2 (basilene blue E3G), adenosine-2‘-phosphate-5’-phosphate (A2P5P) and adenosine-3’-phosphate-5’-phosphate (A3P5P), 2‘-& 3‘-O-(4-benzoylbenzoyl)-ATP, creatine phosphokinase type III and creatine phosphate were purchased from Sigma (Vienna, Austria); 2-methylthio-ADP and -ATP, pyridoxal-phosphate-6-azophenyl-2’,4’-disulphonic acid tetrasodium (PPADS), N⁶-cyclopentyladenosine, 2-p-(2-carboxyethyl)phenethylamino-5’-N-ethylcarboxamidoadenosine (CGS 21680), 8-cyclopentyl-1,3-dipropylxanthine (DPCPX), and xanthine amine congener (XAC) from RBI (Natick, MA, USA); the Sp-isomer of adenosine-5’-O-(1-thiotriphosphate) (ATPαS) from Roche Diagnostics (Vienna, Austria); tetrodotoxin from Latoxan (Rosans, France). Suramin was a gift of Bayer, Austria.

Cell culture

PC12 cells were obtained from the European Collection of Cell Cultures (ECACC, Salisbury, UK), plated onto collagen-coated (Biomedical Technologies Inc., Stoughton, MA, USA) culture dishes (NUNC, Roskilde, Denmark) and were kept in OptiMEM (Life Technologies, Vienna, Austria) supplemented with 0.2 mm l-glutamine (HyClone, Aalst, Belgium), 25.000 IU/L penicillin and 25 mg/L streptomycin (Sigma, Vienna, Austria), 5% fetal calf serum and 10% horse serum (both Life Technologies, Vienna, Austria). Once per week, cell cultures were split, and the medium was exchanged twice weekly. Prior to the recording of I_Ca, PC12 cells were detached from culture dishes by incubation in Ca²⁺-free buffer and replated at low density. In order to induce neuronal differentiation, PC12 cells were exposed to recombinant human β-nerve growth factor (NGF, 50 ng/mL, R & D Systems Inc., Wiesbaden, Germany) for 5–6 days.

Electrophysiology

Whole-cell currents were recorded at room temperature (20–24 °C) from PC12 cells 24–48 h after replating at low density, according to pulished procedures (Boehm et al., 1996), using an Axopatch 200B amplifier and the Pclamp 6.0 hard- and software (Axon Instruments, Foster City, CA, USA). Currents were low-pass filtered at 5 kHz, digitized at 1–5 kHz, and stored on an IBM compatible computer. Traces were analysed off-line by the Clampfit program (Axon). Patch electrodes were pulled (Flaming-Brown puller, Sutter Instruments, Novato, CA, USA) from borosilicate glass capillaries (Science Products, Frankfurt/Main, Germany) and filled with a solution consisting of (mM): CsCl, 130; tetraethylammonium chloride, 20; CaCl₂, 0.24; glucose, 10; HEPES, 10; EGTA, 5; Mg-ATP, 2; and Li-GTP, 2; adjusted to pH 7.3 with KOH, to yield tip resistancies of 2–3 MΩ. Where indicated, 2 mm GTP was replaced by either 2 mm GDPβS or by 0.2 mm GTPγS. The external bathing solution consisted of (mM): NaCl, 120; tetraethylammonium chloride, 20; KCl, 3; MgCl₂, 2; CaCl₂, 5; glucose, 20; HEPES, 10, adjusted to pH 7.3 with KOH. Drugs were applied via a DAD-12 drug application device (Adams & List, Westbury, NY, USA) which permits a complete exchange of solutions surrounding the cells under investigation within less than 100 ms (see, e.g. Boehm, 1999b). To evaluate the effects of adenosine and nucleotide receptor antagonists (e.g. suramin, PPADS, reactive blue 2, A2P5P, A3P5P, ATPαS, benzoyl-ATP), neurons were first exposed to these agents for at least 30 s, and then ADP or other nucleotides were applied together with antagonist. Afterwards, nucleotides were tested in the absence of antagonists.

Unless indicated otherwise, whole-cell Ca²⁺ currents were elicited by 30 ms depolarizations from a holding potential of −80 mV to 0 mV at a frequency of 2–4 per min. To investigate the voltage-dependence of the inhibition of I_Ca by ADP, a 50-ms depolarization to 0 mV was preceded by a 50-ms predepolarization to +50 mV and a subsequent 10 ms repolarization to −80 mV (Fig. 2C). Currents through P2X receptors were evoked by the direct application of nucleotides at a membrane potential of −80 mV at a frequency of 2 min. In order to enable complete equilibration of the cytoplasm with the internal solution of the pipette containing diverse drugs (Pusch & Neher, 1988), current recordings were started more than 5 min after the establishment of the whole-cell configuration, unless indicated otherwise.

Calculations

I_Ca was quantified by measuring peak current amplitudes during the depolarization to 0 mV. To account for time-dependent changes in Ca²⁺ current amplitudes (see Fig. 2A), drug effects were evaluated by evoking currents in the presence of test drugs (B) and by comparing them to control currents recorded before (A) and after (washout, C) the application of the drugs, according to the equations: 200 × B/(A + C) = percentage of control current, or 100 − [200 × B/(A + C)] = percentage inhibition (see, e.g. Boehm et al., 1996). Currents through P2X receptors were also quantified by determination of peak current amplitudes.

All data represent arithmetic means ± SEM; n represent numbers of single neurons. Concentration–response curves were fitted to experimentally obtained data by the ALLFIT programme (DeLean et al., 1978) which provides estimates ( ± SEM) for half maximal concentrations, as well as minimal and maximal effects, and determines the qualities of fitted results. Differences between single concentration–response curves were determined by simultaneous fitting with shared parameters and subsequent calculation of the F-statistic on the resulting ‘extra sum of squares’ (DeLean et al., 1978). Significance of differences between single data points was evaluted by the Mann–Whitney test.

ADP inhibits I_Ca in untreated as well as NGF-treated PC12 cells

Depolarization of PC12 cells, whether treated with NGF or not, from −80 mV to 0 mV elicited I_Ca with rapid activation kinetics, the time constants of activation were: 2.2 ± 0.2 ms in untreated (n = 7) and 2.5 ± 0.3 ms in NGF-treated (n = 5) PC12 cells. As reported previously (Bouron et al., 1999), whole cell I_ca were significantly larger in NGF-treated cells than in untreated cells, but this rise in current amplitudes was accompanied by an increase in cell capacitance, so that current densities in NGF-treated cells were not significantly larger than in untreated cells (Fig. 1A and C). In the presence of 100 µm ADP, I_Ca amplitudes were reduced by 30% in untreated and by 60% in NGF-treated PC12 cells, and this effect was entirely reversible upon removal of the nucleotide (Fig. 1A and C). Apart from reducing current amplitudes, ADP markedly slowed activation kinteics which resulted in a biexponential activation time course, the fast components of activation had time constants of: 1.6 ± 0.2 ms in untreated (n = 7) and of 2.4 ± 0.7 in NGF-treated (n = 5) cells, whereas the second slow components had time constants of 51.2 ± 32.8 and 71.4 ± 65.8 ms, respectively.

In addition to 100 µm ADP, PC12 cells were exposed to the same concentration of ATP, UDP and UTP. In the presence of such high concentrations of ATP, Ca²⁺ currents could no longer be measured, since ATP acts at P2X receptors and thereby elicits rapidly activating inward currents at negative membrane potentials, an effect that is not seen in the presence of ADP (Fig. 1B). Subsequent to the application of 100 µm ATP, I_Ca amplitudes were irreversibly diminished (not shown), most likely due to transmembrane Ca²⁺ entry caused by the opening of the ATP-gated ion channels. UDP and UTP, both at 100 µm, failed to alter I_Ca (Fig. 1C).

The inhibitory action of ADP is mediated by GTP binding proteins

Receptor-dependent inhibition of VACCs most commonly involves GTP binding proteins (e.g. Hille, 1994; Zamponi & Snutch, 1998). To test for a role of G-proteins, electrodes were filled with solutions containing 2 mm GTP, 2 mm GDPβS, or 0.2 mm GTPγS, and I_Ca recordings were started one minute after formation of the whole-cell configuration. With GTP and GDPβS, I_Ca displayed significant ‘run-up’, whereas accelerated ‘run-down’ was observed with GTPγS. With GTP in the electrode, the inhibitory action of 100 µm ADP was constant for more than 15 min, whereas the effect was lost during this period of time when GTP was replaced by GDPβS. With GTPγS, the rapid run-down occluded the inhibition by ADP.

To obtain evidence as to which types of G-proteins might be involved in the inhibitory action of ADP, PC12 cells were treated with pertussis toxin (100 ng/mL for 24 h) and I_Ca was recorded in the absence and presence of 100 µm ADP; in these cells, ADP failed to alter I_Ca. In sister cultures, not treated with pertussis toxin, I_Ca was reduced in the presence of ADP by about 40% (Fig. 2B).

Upon activation, G-proteins may inhibit VACCs through a direct interaction of βγ subunits with the ion channels, an effect that is voltage-dependent and, thus, attenuated by depolarizing prepulses (Zamponi & Snutch, 1998; Kammermeier et al. 2000). Accordingly, predepolarization of PC12 cells to +50 mV prior to the activation of I_Ca attenuated the inhibitory action of ADP and prevented the agonist-induced slowing of activation kinetics (Fig. 2C).

The inhibitory action of ADP does not involve adenosine receptors

Adenine nucleotides are rapidly degraded by a cascade of nucleotidases to generate adenosine as the final product. Therefore, we tested adenosine receptor agonists and antagonists to find out whether the nucleoside might contribute to the action of ADP. However, in cells which clearly showed an inhibition of I_Ca by ADP, neither the A₁ adenosine receptor agonist cyclopentyladenosine (1 µm, Fig. 3C) nor the A₂ receptor agonist CGS 21680 (5 µm; Fig. 3A and B) caused significant alterations in I_Ca amplitudes. Furthermore, the inhibitory action of 100 µm ADP was not altered in the presence of the A₁ receptor antagonist DPCPX (3 µm) or the nonselective adenosine receptor antagonist XAC (1 µm; Fig. 3D).

Effects of P2 receptor antagonists on the inhibition of I_Ca by ADP

The results shown above suggest that it was the nucleotide itself and not the degradation product, adenosine, that caused the inhibition of I_Ca. To prove that ADP acted via a nucleotide receptor, we investigated the effects of several nonselective P2 receptor antagonists on the inhibition of I_Ca by ADP. Suramin and PPADS, both at concentrations of up to 100 µm, did not alter the reduction of I_Ca amplitudes by 100 µm ADP (Fig. 4B). However, reactive blue 2 at 30 µm virtually abolished the inhibitory action of this concentration of ADP (Fig. 4A and B). In addition, reactive blue 2 per se enhanced I_Ca amplitudes (Fig. 4A) by 16.5 ± 5.4% (n = 5; P < 0.05), whereas suramin and PPADS had no such effect (not shown).

Agonist pharmacology of the P2 receptor mediating the inhibition of I_Ca by ADP

The data presented so far indicates that the action of ADP on I_Ca was mediated by a G-protein coupled nucleotide receptor, i.e. a P2Y purinoceptor. To further characterize this receptor, the following nucleotides were applied at concentrations between 1 nm and 300 µm: ATP, ADP, ADPβS, 2-methylthio-ATP and 2-methylthio-ADP. To account for differences in the inhibitory actions of nucleotides in between different cells, the effects of these nucleotides were compared with the effect of 100 µm ADP in the very same cell. Concentration–response curves for the reduction of I_Ca amplitudes obtained in this manner revealed that the action of ADP was half maximal at 34.1 ± 9.1 µm. ATP could be tested at concentrations of up to 30 µm only (higher concentrations produced large P2X receptor-mediated inward currents, see 1, 6); at these concentrations, ATP inhibited I_Ca to about the same extent as ADP. Assuming that ATP might cause the same maximal inhibition as ADP (see Fig. 6C), the action of ATP was calculated to be half maximal at 25.0 ± 7.5 µm (Fig. 5A). ADPβS caused the same extent of maximal inhibition as ADP, but was significantly more potent than ADP with half maximal inhibition of I_Ca at 8.8 ± 1.5 µm (P < 0.05 vs ADP). The most potent agonists were 2-methylthio-ADP and -ATP, the inhibition being half-maximal at 38.4 ± 14.4 nm and 104.2 ± 47.4 nm, respectively; these half maximal concentrations were not significantly different from each other (P > 0.1).

P2X rexceptors do not contribute to the inhibition of I_Ca

At least two of the agonists used above, namely ATP and 2-methylthio-ATP, may activate not only P2Y, but also P2X receptors (Boehm, 1999a; Ralevic & Burnstock, 1998), and several P2X receptor subunits are expressed in PC12 cells (Arslan et al. 2000). To exclude that activation of P2X receptors might contribute to the inhibition of I_ca by these agonists, 100 µm ATP was applied in the absence and presence of the P2 antagonists suramin or PPADS (both at 100 µm), respectively; these antagonists did not attenuate the inibition of I_Ca by ADP (see Fig. 4) but they are known to block ATP-evoked currents through P2X receptors (e.g. Boehm, 1999a). ATP applied alone elicited inward currents with amplitudes in the range of 0.1–1 nA, which were reduced by suramin or PPADS by ≥ 90% (Fig. 6A and C). Nevertheless, in the presence of these antagonists, ATP (100 µm) did reduce I_Ca to the same extent as ADP (100 µm), and the action of ADP in the presence of suramin or PPADS was not different from its inhibitory effect when applied alone (Fig. 6B and C). To corroborate that activation of P2X receptors was not required for the inhibition of I_Ca, 10 µm ATP or 2-methylthio-ATP were used. At these concentrations, the adenine triphosphates did not cause significant inward currents, but clearly reduced I_Ca (Fig. 6C) to about the same relative extent as indicated in the concentration–response in Fig. 5. These results prove that P2X receptor activation was not a prerequisite for the inhibition of I_Ca.

The receptor mediating inhibition of I_Ca is P2Y1-like but not P2Y1

The agonist pharmacology described in Fig. 5 is compatible with those of cloned P2Y1 receptors (e.g. Leon et al., 1997; Ralevic & Burnstock, 1998; Filippov et al. 2000). Therefore, ADP (100 µm) was applied in presence of 100 µm of the selective P2Y1 receptor antagonists A2P5P and A3P5P (Boyer et al., 1996) which, however, failed to alter the inhibitory action of the nucleotide (Fig. 7A). In contrast, the inhibition of I_Ca by 100 µm ADP was significantly attenuated by 100 µm benzoyl-ATP (Fig. 7A) which blocks not only P2Y1, but also P2Y1-like, receptors (Vigne et al., 1999). A2P5P, A3P5P, and benzoyl-ATP per se did not alter I_Ca amplitudes. ATPαS has been shown to block P2Y1-like (Hechler et al., 1998a), but not P2Y1 receptors, at which it rather appears to be an agonist (Vöhringer et al. 2000). At 100 µm, this nucleotide elicited inward currents (presumably by activation of P2X receptors), and subsequently reduced I_Ca in an irreversible manner, as did ATP (see above). At 30 µm, however, ATPαS had no such effect, but attenuated the inhibitory action of 30 µm ADP on I_Ca (Fig. 7A).

ATP preparations may contain amounts of ADP sufficient to activate P2Y receptors. To find out whether the actions of ATP and 2-methylthio-ATP were caused by respective diphosphate contaminations, ATP and 2-methylthio-ATP containing solutions were treated with an ATP regenerating system provided by creatine phosphokinase (20 U/mL) and 10 mm creatine phosphate for at least 90 min, which has been shown to virtually eliminate diphosphate contaminations (Hechler et al., 1998b). In addition, ADP may be generated from ATP by ectonucleotidases. For this reason, ATP and 2-methylthio-ATP were applied to PC12 cells together with creatine phosphokinase (20 U/mL) and 10 mm creatine phosphate. This procedure did not significantly alter the maximal inhibition achieved by 2-methylthio-ATP, but raised its half maximal concentration to 250.2 ± 78.8 nm (P < 0.05 vs 2-methylthio-ATP in the absence of creatine phosphokinase and creatine phosphate; Fig. 7B). The fitted concentration–response curves for ATP in the absence and presence of the ATP regenerating system, respectively, were not significantly different from each other, most likely because concentrations of up to 30 µm only could be applied (see Fig. 6). Nevertheless, at this concentration, ATP produced significantly less inhibition of I_Ca in the presence than in the absence of creatine phosphokinase and creatine phosphate (Fig. 7B).

P2Y, but not P2X or P1, receptors cause inhibition of VACCs in PC12 cells

The superfamily of receptors for purines and pyrimidines comprises representatives of two principal classes of neurotransmitter receptors, i.e. ligand-gated ion channels and G-protein-coupled receptors (Fredholm et al., 1994; Ralevic & Burnstock, 1998). Several of the results presented above show that the nucleotide-induced inhibition of I_Ca was mediated by G-proteins, and not by ATP-gated ion channels, i.e. P2X receptors. (i) Under conditions that prevented the activation of ionotropic purine receptors, the nucleotides used did reduce I_Ca. (ii) Intracellular application of the transphosphorylation-resistant GDP analogue GDPβS, which traps G-proteins in an inactive conformation, abolished the inhibitory effect of ADP on I_Ca; this occurred in a time-dependent manner which reflects, on one hand, the time that is needed for dialysis of the cytoplasm with the GDPβS-containing soution (Pusch & Neher, 1988), and on the other hand, the time that is required to exchange GDPβS for GDP within the nucleotide binding pocket of the G-protein α-subunits (Nanoff et al., 1994). (iii) Intracellular application of GTPγS, which traps G-proteins in an active conformation, slowly reduced I_Ca amplitudes and thereby increasingly occluded the inhibitory effect of ADP. (iv) Treatment of the cells with pertussis toxin which prevents the receptor-dependent activation of G_i and G_o proteins entirely abolished the inhibition of I_Ca by ADP. (v) Depolarizing prepulses attenuated the inhibitory action of ADP on I_Ca. This voltage-dependent type of I_Ca inhibition has been shown to be mediated by G-protein βγ subunits, whereas a voltage-independent inhibition of I_Ca in sympathetic neurons involves G-protein α-subunits (Kammermeier et al. 2000). Thus, the inhibitory action of ADP on I_Ca can be assumed to be mediated by βγ subunits of inhibitory G-protein heterotrimers. Taken together, these results indicate that ADP acted via a G-protein-coupled receptor, but not via P2X receptors.

The group of G-protein coupled receptors for purines and pyrimidines consists of receptors for adenosine (P1 receptors) and of receptors for nucleotides (P2Y receptors; Fredholm et al., 1994; Ralevic & Burnstock, 1998). In cells that showed a pronounced inhibitory effect of ADP, agonists at adenosine receptors did not alter I_Ca. Furthermore, the inhibition of I_Ca by ADP was not altered by adenosine receptor antagonists, but was largely attenuated by the P2 receptor antagonist, reactive blue 2. Thus, ADP diminished I_Ca via a G-protein-coupled nucleotide receptor, i.e. via a P2Y receptor, and not via a P1 (adenosine) receptor.

ADP inhibits VACCs in PC12 cells via P2Y1-like receptors

Currently, four different subtypes of rat P2Y receptors have been characterized by molecular means: P2Y1, −2, −4 and −6 (Ralevic & Burnstock, 1998). Amongst these, the P2Y1 receptor is the only one that is not activated by uridine nucleotides, and uridine nucleotides did not affect I_Ca in PC12 cells. In contrast, ADP and ATP at micromolar concentrations equipotently inhibited I_Ca. Therefore, we tested additional agonists that are known to activate P2Y1 receptors, such as 2-methylthio-ADP and -ATP and ADPβS, for their capacity to inhibit I_Ca. ADPβS was about four-fold more potent than ADP in inhibiting I_Ca, and the 2-methylthio-derivatives were at least 300-fold more potent than ADP. These pharmacological characteristics resemble those of cloned P2Y1 receptors (Ralevic & Burnstock, 1998). In some, but not all, experimental systems, adenosine triphosphates were found to be antagonists at heterologously expressed P2Y1 receptors, and this fact may be occluded by diphosphate contaminations in commercially available ATP preparations (e.g. Hechler et al., 1998b; Palmer et al., 1998; Filippov et al. 2000). However, an ATP regenerating system, that is known to minimize ADP contaminations, reduced the potency, but did not prevent the agonistic acitivity, of ATP and 2-methylthio-ATP.

So far, the results are compatible with the ADP-induced inhibition of I_Ca in PC12 cells being mediated by P2Y1 receptors. However, the selective P2Y1 receptor antagonists A2P5P and A3P5P (Boyer et al., 1996; Filippov et al. 2000) failed to antagonize the effect of ADP. In contrast, ATPαS an antagonist at P2Y1-like receptors, which inhibit adenylyl cyclase activity in platelets (Hechler et al., 1998a), but not at P2Y1 receptors themselves (Hechler et al., 1998a; Vöhringer et al. 2000), attenuated the inhibitory action of ADP. Similarly, benzoyl-ATP, which blocks P2Y1 as well as cyclase-inhibiting P2Y1-like receptors (Vigne et al., 1999), also attenuated the inhibition of I_Ca by ADP. Additional data also argue against a role of P2Y1 receptors in the modulation of I_Ca by ADP: (i) Suramin and PPADS are antagonists at recombinant P2Y1 receptors (e.g. Leon et al., 1997; King et al., 1998), but did not alter the inhibition of I_Ca by ADP; (ii) In heterologous expression systems, P2Y1 receptors are activated by ADP in submicromolar, rather than micromolar (see above), concentrations, and 2-methylthio-ADP is about 10-fold, not 1000-fold (as above), more potent than ADP (Leon et al., 1997; Palmer et al., 1998; Filippov et al. 2000); (iii) PC12 cells have been reported not to express P2Y1 receptors (Arslan et al. 2000), and a RT-PCR analysis performed in the PC12 clone we used detected transcripts for P2Y2, −4, and −6, but none for P2Y1 (Moskvina and Boehm, unpublished observation). Thus, the receptor mediating the inhibition of I_Ca by ADP is a P2Y1-like receptor, but not P2Y1 itself.

Apart from P2Y1 receptors, the human P2Y11 receptor is activated by adenine, but not uridine, nucleotides (Communi et al., 1997). Hence, a homologous receptor in the rat might be involved in the effects described above. However, ADP and 2-methylthio-ADP are only weak agonists at P2Y11. Furthermore, this receptor stimulates the formation of cyclic AMP and is thus linked to stimulatory, rather than inhibitory, G-proteins (Communi et al., 1997).

Significance of the P2Y receptor-mediated inhibition of VACCs

Sympathetic neurons, chromaffin cells, and PC12 cells store and release not only catecholamines and peptides, but also ATP (von Kügelgen & Starke, 1991; Winkler, 1993). Hence, upon vesicle exocytosis, endogenous ATP may act on receptors located in the membrane of those cells which the nucleotide has been released from. In addition, nucleotidases rapidly hydrolyse released ATP (Todorov et al., 1997) and degradation products, such as ADP and adenosine, may also act on these receptors. Indeed, in chromaffin cells VACCs were found to be under control of an autocrine modulator that was suggested to be ATP (Currie & Fox, 1996), although the receptors involved remained unknown.

Transmitter release relies on transmembrane Ca²⁺ entry via VACCs, and the receptor-dependent inhibition of Ca²⁺ channels was suggested to be the only mechanism mediating the autoinhibitory modulation of sympathetic transmitter release via presynaptic α₂-adrenoceptors (Boehm & Huck, 1996). Given that the receptors described above are also expressed in sympathetic neurons, these P2Y1-like receptors may underly the ATP-dependent presynaptic autoinhibition previously described for the sympathetic nervous system (Fuder & Muth, 1993; von Kügelgen et al., 1993; Boehm, 1999a).