Dual Effect of Acid pH on Purinergic P2X3 Receptors Depends on the Histidine 206 Residue*

Whole cell patch clamp investigations were carried out to clarify the pH sensitivity of native and recombinant P2X3 receptors. In HEK293 cells permanently transfected with human (h) P2X3 receptors (HEK293-hP2X3 cells), an acidic pH shifted the concentration-response curve for α,β-methylene ATP (α,β-meATP) to the right and increased its maximum. An alkalic pH did not alter the effect of α,β-meATP. Further, a low pH value increased the activation time constant (τon) of the α,β-meATP current; the fast and slow time constants of desensitization (τdes1, τdes2) were at the same time also increased. Finally, acidification accelerated the recovery of P2X3 receptors from the desensitized state. Replacement of histidine 206, but not histidine 45, by alanine abolished the pH-induced effects on hP2X3 receptors transiently expressed in HEK293 cells. Changes in the intracellular pH had no effect on the amplitude or time course of the α,β-meATP currents. The voltage sensitivity and reversal potential of the currents activated by α,β-meATP were unaffected by extracellular acidification. Similar effects were observed in a subpopulation of rat dorsal root ganglion neurons expressing homomeric P2X3 receptor channels. It is suggested that acidification may have a dual effect on P2X3 channels, by decreasing the current amplitude at low agonist concentrations (because of a decrease in the rate of activation) and increasing it at high concentrations (because of a decrease in the rate of desensitization). Thereby, a differential regulation of pain sensation during e.g. inflammation may occur at the C fiber terminals of small DRG neurons in peripheral tissues.

High proton concentrations have been registered in inflamed tissue (down to pH 5.4), after surgical interventions (down to pH 5.5), in fracture-related hematomas (down to pH 4.7), in cardiac ischemia (down to pH 5.7), and in and around malignant tumors (1)(2)(3)(4)(5). Therefore, local acidosis is considered to contribute to pain experienced in these states (5)(6)(7)(8)(9)(10)(11). It is also known that continuous administration of low pH buffered solutions into human skin evokes instant pain and hyperalgesia to mechanical stimulation (12). Electrophysiological experiments in rat skin nerve preparations showed that pathophysiologically relevant high proton concentrations produce a selective nonadapting excitation of nociceptors and a significant sensitization to mechanical stimulation (13). Thus, it has been proposed that local acidosis may play a major role in pain and hyperalgesia (7).
Hydrogen ions are able to excite dorsal root ganglion (DRG) 2 neurons via the activation and/or modulation of inward cationselective currents, including the acid-sensing ion channels (ASICs) (14), the transient receptor potential vanilloid receptor 1 (TRPV1) (15,16), and P2X receptors (17,18). P2X receptors represent a family of ligand-gated cationic channels that open in response to the binding of ATP, possess two transmembrane domains, intracellular N and C termini, a large extracellular loop, and assemble as homo-or heterotrimers (19 -21). In peripheral tissues, large quantities of ATP may leave the intracellular space in response to tissue trauma, tumors, inflammation, migraine, or visceral distension (22). The resulting P2X receptor activation and the subsequent depolarization of the sensory cell membrane initiate action potentials that are perceived centrally as pain. Although sensory neurons express all known P2X subunits, the homomeric P2X 3 and the heteromeric P2X 2/3 receptors occur in these cells at the highest density (23,24).
Whereas agonist-induced currents through recombinant P2X 2 and P2X 2/3 receptor channels (17,18) were potentiated by acidification, the same change in pH depressed currents through P2X 3 receptor channels (18). The failure to use more than a single concentration of ␣,␤-meATP did not allow to construct concentration-response relationships for the P2X 3 receptor subunit. Bullfrog DRG and rat nodose ganglion neurons are known to possess native P2X 1 , P2X 2 , and P2X 3 receptors. Application of ATP onto these cells evoked non-desensitizing currents resembling P2X 2 or P2X 2/3 receptor activation (19). These currents were enhanced by acidification, suggesting that extracellular protons can regulate the function of ATPgated receptor channels in bullfrog DRG and rat nodose ganglion neurons (25,26).
The aim of the present study was to investigate the effect of acidification on recombinant human P2X 3 receptors (hP2X 3 R) transfected into human embryonic kidney (HEK) 293 cells. It was found that a change of pH from the normal 7.4 to 5.8 markedly decelerated both the kinetics of channel opening and the subsequent speed of desensitization, and thereby depressed currents caused by low, but facilitated currents caused by high agonist concentrations. Our findings demonstrate a dual, agonist concentration-dependent effect of acid pH at homomeric P2X 3 receptors which may allow a differential modulation of pain sensation in response to small or large quantities of ATP released by noxious stimulation.
Site-directed Mutagenesis and Transfection Procedures-The human P2X 3 receptor cDNA (GenBank TM accession number NM-00255) was subcloned per PstI and EcoRI restriction sites into pIRES2-EGFP vector from Clontech Laboratories (Mountain View, CA) for independent expression of P2X 3 and EGFP, creating the pIR-P2 plasmid. All P2X 3 receptor mutants were generated by introducing point mutations into the pIR-P2 construct, using the QuikChange site-directed mutagenesis kit from Stratagene (La Jolla, CA) according to the instruction manual. HEK293 cells were plated in 35-mm plastic dishes 1 day before transient transfection. 0.5 g of plasmid DNA per dish was combined with 10 l of Polyfect reagent from Qiagen (Hilden, Germany) and 100 l of Optimem (Invitrogen, Karlsruhe, Germany). After 10 min of incubation, the lipid-DNA complexes were introduced to the cells. Approximately 18-h post-transfection, the medium was replaced with Optimem to remove residual Plasmid DNA. Electrophysiological recordings took place 48 -72 h after transient transfection during maximum protein expression. P2X 3 receptor-positive cells were detected by viewing EGFP fluorescence with a fluorescence microscope.
Preparation of DRG Neuronal Cultures-One-day-old Wistar rats (own breed) were used in the study. The animals were killed under CO 2 and decapitated to obtain cell cultures of DRG neurons. The isolation and culturing procedures of thoracic and lumbar DRG cells have been described in detail previously (28,29). DRG cells were plated at a density of 3 ϫ 10 4 cells onto 35-mm plastic dishes coated by poly-L-lysine (25 g/ml) (Sarstedt). They were kept in Dulbecco's modified Eagle's medium, 35 mM total glucose, 2.5 mM L-glutamine, 15 mM HEPES, 50 g/ml gentamicin, 5% fetal bovine serum (Invitrogen), 30 ng/ml nerve growth factor, 10 g/ml insulin, 5.5 g/ml transferrin, and 5 ng/ml selenium (Sigma). Primary cultures of rat DRG neurons were maintained for 2-4 days in a humidified atmosphere (37°C, 5% CO 2 ) before experimentation.
Whole Cell Patch Clamp Recordings-Whole cell patchclamp recordings were performed 2-6 days after the splitting of permanently transfected HEK293 cells, and 2-4 days after the plating of rat DRG neurons, at room temperature (20 -22°C), using an Axopatch 200B patch-clamp amplifier (Molecular Devices, Union City, CA). Patch pipettes (3- 5 M⍀) for  both HEK293 cells and DRG neurons were filled with intracellular solution of the following composition (in mM): 135 CsCl, 2  MgCl 2 , 20 HEPES, 11 EGTA, 1 CaCl 2 , 1.5 Mg-ATP, and 0.3 Li-GTP, pH was adjusted with CsOH. The external solution consisted of (in mM) 140 NaCl, 5 KCl, 2 MgCl 2 , 2 CaCl 2 , 10 HEPES, and 11 glucose, pH was adjusted with NaOH. The pH of all solutions was routinely checked before and during experiments. All recordings were made at a holding potential of Ϫ70 mV. Data were filtered at 2 kHz with the in-built filter of the Axopatch 200B, digitized at 5 kHz, and stored on a laboratory computer using a Digidata 1200 interface and pClamp 10.0 software (Molecular Devices).
Drugs were dissolved in external solution and applied locally to single cells, using a rapid solution change system (SF-77B Perfusion Fast-Step, Warner Instruments, Hamden, CT; 10 -90% rise time of the junction potential at an open pipette tip was 1-4 ms). Concentration-response curves for the P2X 3 receptor currents at different pH values were constructed by applying every 5 min increasing concentrations of ␣,␤-methylene ATP (␣,␤-meATP). To analyze P2X 3 currents from the same cells at different pH values, ␣,␤-meATP was applied at a given concentration six times with 5-min intervals. After the recording of two ␣,␤-meATP-induced inward currents of the same size at pH 7.4, the pH was changed 2.5 min after the second ␣,␤-meATP administration. Two ␣,␤-meATP currents were recorded at the altered pH value and two further ones at pH 7.4 again. When recording from DRG cells, amiloride (200 M) was given to the bath solution to block ASICs. This concentration of amiloride did not affect P2X 3 currents either in HEK293-hP2X 3 cells or in DRG neurons (see "Results"). Concentration-response curves established to determine the agonistic effects of ␣,␤-meATP were fitted to mean data points using Equation 1, where I is the observed current, I max the extrapolated maximal current, [A] the ␣,␤-meATP concentration in M, EC 50 the concentration of ␣,␤-meATP that produces 50% of I max and nH the Hill coefficient (slope value) (SigmaPlot, SPSS, Erkrath, Germany). In experiments investigating the kinetics of P2X 3 currents, decay phases of the curves were fitted by the following biexponential function in Equation 2, using the in-built function of the pClamp 10.0 software (Molecular Devices), where A 1 and A 2 are the relative amplitudes of the first and second exponential, des1 and des2 are the desensitization time constants, and P is plateau. The onset time constants ( on(10 -90%) ) were calculated from the individual recordings, under the assumption that despite the relatively slow local application they give a rough approximation of the kinetics of channel opening. The effect of a pH change on the receptor activation and desensitization was found to be concentrationindependent and was expressed as the mean Ϯ S.E. of the time constant ratios at each concentration investigated. In experiments examining the recovery of P2X 3 receptors from desensitization, HEK293 cells were stimulated repetitively with ␣,␤-meATP at 3 M (5-s pulses, each) with a progressive increase in the interpulse intervals. The recovery from desensitization was fitted using the following Equation 3 (30), where I max and I min are the start and finish levels during recovery, t is the time, t 50 is the time to regain 50% of maximally recovered currents, and b is the slope factor giving the change in time per e-fold change in recovery.
Materials and Drugs-␣,␤-Methylene ATP lithium salt (␣,␤-meATP) and amiloride hydrochloride were both purchased from Sigma. The drugs were prepared as a concentrated stock solution in distilled water and were diluted to the final concentration in external medium.
Data Analysis-Data were analyzed off-line using pClamp 10.0 software (Molecular Devices). Figures show mean Ϯ S.E. values of n experiments. Student's t test or one way ANOVA followed by Bonferroni's post hoc test were used for statistical analysis. A probability level of 0.05 or less was considered to reflect a statistically significant difference. 3 Currents in HEK293-hP2X 3 Cells-Application of various concentrations of ␣,␤-meATP to HEK293 cells permanently transfected with hP2X 3 Rs (HEK293-hP2X 3 cells) induced inward current responses (Fig. 1A). Acidification of the extracellular solution had 3-fold effects on the concentration-response curves for ␣,␤-meATP. It enhanced the maximal current amplitudes (I max ), increased the EC 50 values, and raised the Hill coefficient (nH; Fig. 1Ba and Table 1). In contrast, the alkalinization of the extracellular solution caused no major change in the I max , the EC 50 value or the nH ( Fig. 1Bb; Table 1).

Effect of Extracellular Protons on P2X
To further investigate the effect of protons on P2X 3 receptors, we applied two selected agonist concentrations (1 and 4.64 M) onto HEK293-hP2X 3 cells and examined, whether a change in the pH affects P2X 3 currents in the same cell. As shown in Fig. 2, the current amplitudes evoked by ␣,␤-meATP at 1 M (approximate EC 50 ) were inhibited by shifting the pH from 7.4 to 5.8 by 43.7 Ϯ 7.6% (n ϭ 10, p Ͻ 0.05) whereas at 4.64 M (high concentration of this agonist), the current amplitudes were facilitated by 28.9 Ϯ 14.0% (n ϭ 8, p Ͻ 0.05) ( Fig. 2A). Both the increase and the decrease of the current responses to these ␣,␤-meATP concentrations were reversible on returning to the original extracellular pH of 7.4. In contrast, a change of the pH from 7.4 to 8.0 did not modify the ␣,␤-meATP currents at either concentration (1.4 Ϯ 6.3%, n ϭ 6 and 1.8 Ϯ 8.6%, n ϭ 6, at 1 and 4.64 M, respectively) (Fig. 2B).
The original tracings in Fig. 2, A and B show that acidification, but not alkalinization itself induces rapidly declining inward currents. These currents may be due to the activation of ASICs described to be present in HEK293 cells (31). As documented in Fig. 2, Aa and Ab, these currents completely desensitized within milliseconds, and therefore they are not expected to alter the currents evoked by ␣,␤-meATP 2.5-min later.

receptors permanently transfected into HEK293 cells at various pH levels
The parameters were determined from the concentration-response curves in Fig. 1B and from the recovery curves in Fig. 4B  Nonetheless, to reliably exclude a possible interference of ASIC currents with ␣,␤-meATP-induced currents, the same experiments as shown in Fig. 2A were performed in the presence of amiloride, a selective inhibitor of ASIC channels (14). In fact, amiloride (200 M) abolished the early current induced by a change of pH from 7.4 to 5.8 (96.5 Ϯ 2.7% inhibition, n ϭ 5, p Ͻ 0.05%), but had no effect on the response to ␣,␤-meATP (1 M) at a pH of 5.8 (40.3 Ϯ 9.6% inhibition, n ϭ 5, p Ͼ 0.05%; compare with Fig. 2Ac). It is noteworthy that the lower concentration (1 M) of ␣,␤-meATP evoked a current response at a pH of 5.8, which desensitized much slower than that observed at the control pH value of 7.4 ( Fig. 2Aa; see below). Effect of Protons on P2X 3 Receptor Activation and Desensitization-Next, we investigated whether different proton concentrations affect the activation and desensitization of P2X 3 channels. The current decay in the presence of ␣,␤-meATP, which represents the onset of desensitization, was fit-ted by a biexponential equation (see "Materials and Methods") and had a first fast and a second slow time constant ( des1 , des2 ). Of these time constants, des1 was found to be the more dominant one, because it constituted concentration-independently 84.9 Ϯ 1.4% of the current decay, as calculated from the contribution of the relative amplitudes of the first and second exponential to the total current amplitude (A 1 /A 1 ϩA 2 , see Equation 2 under "Materials and Methods"; n ϭ 39). The activation time constants ( on ) and both desensitization time constants ( des1 , des2 ) inversely correlated with the ␣,␤-meATP concentration (Fig. 3, Ca-Cc). These data are in accordance with earlier observations showing that the kinetics of activation, desensitization, and recovery from desensitization, but not deactivation of P2X receptors, inversely correlate with the agonist concentration (32)(33)(34).
Lowering the pH to 5.8 increased both the activation time constant and the two desensitization time constants at all ago-  nist concentrations applied (Fig. 3, A, B, and Ca-Cc). Increasing the pH value up to 8.6 did not change the activation and desensitization time constants appreciably. The normalized shift of concentration time constant curves at different pH values is shown in Fig. 3D. As shown in Fig. 3Cb, the pH-dependent amplification of des2 (but not des1 ; Fig. 3Ca) was attenuated at higher agonist concentrations. We calculated the contribution of des2 to the current decay at different pH and agonist concentrations and found that des2 was independent of these parameters and constituted only 16.3 Ϯ 3.4% (n ϭ 38) of the total current decay. It is interesting to note that the concentration versus on curves were less steep than the concentration versus des1 curves. Thus at lower concentrations of ␣,␤-meATP the current amplitude was determined mostly by the activation time constant, whereas at higher concentrations of ␣,␤-meATP the current amplitude was determined mostly by the first time constant of desensitization (Fig. 3, Ca and Cc).
Effect of Protons on the Recovery from Desensitization-After rapid desensitization, P2X 3 receptors stay in the desensitized state for minutes and cannot fully open in response to further agonist application. We next examined the availability of the receptors after different time intervals i.e. how fast they recover from this desensitized state and, moreover, whether changing the proton concentration modulates this process. HEK293-hP2X 3 cells were stimulated repetitively with ␣,␤-meATP (10 M, for 5 s), by progressively increasing the interpulse intervals (Fig. 4). During the application of ␣,␤-meATP, P2X 3 receptors completely desensitized. The recovery of P2X 3 receptors from desensitization exhibited a sigmoidal time course and was best fitted with Equation 3 described under "Materials and Methods." The time course of recovery was expressed as t 50 , namely the time to regain 50% of the maximally recovered current amplitudes. Lowering the pH to 5.8 significantly accelerated the recovery (Table 1; analysis of variance, p Ͻ 0.05, n ϭ 7-8), whereas increasing the pH to 8.0 had no effect on the speed of recovery ( Table 1).
The Role of Histidine Residues in pH Sensitivity-Protons modulated P2X 3 receptors in the range between pH 7.4 and 5.8. The only amino acid with a pK a close to this range is histidine. The pK a of free histidine is 6.0 (35), but in proteins its pK a can range from 5.0 to 8.0 (36) making histidines apparent candidates for the proton-binding site. To test whether histidines are involved in the modulation of P2X 3 receptors by protons, we substituted single histidine residues by alanines in the extracellular domain of the human P2X 3 receptor (H45A and H206A).
The wild-type (WT) P2X 3 receptors transiently transfected into HEK293 cells responded to acidification in a manner similar to that reported for the permanently transfected receptor (compare Tables 1 and 2). In fact, lowering the pH from 7.4 to 5.8 shifted the concentration-response curve for ␣,␤-meATP to the right and increased the I max (Fig. 5, Aa and Ab). We investigated the effect of replacing two histidine residues in the extracellular loop of the P2X 3 receptor by the uncharged amino acid alanine (H45A, H206A) on the pH sensitivity of the ␣,␤-meATP effect (Fig. 5B and Table 2). The most important finding is that the H206A mutation abolished the increase of the maximal response, the shift of the concentration-response curve to the right, and the changes in all three kinetic parameters ( des1 , des2 , and on ) by acidification to a pH of 5.8 (Fig. 5, Ba and C). In contrast, the single mutation H45A, accentuated the pH-sensitivity of the receptor; both the increase of the maximal response and the rightward shift of the concentrationresponse curve were more pronounced than in the case of the

receptors transiently transfected into HEK293 cells at various pH levels
The parameters were determined from the concentration-response curves in Fig. 5 WT receptor (Fig. 5Bb). The effects of acidification on the on , des1 , and des2 were similar to that observed with the WT receptor (Fig. 5, Ca, Cb, Cc). The double mutant caused a mixture of these effects in that although a rightward shift of the concentration-response curve by the acidic pH was still present, the increase in I max and a change in the kinetic parameters of desensitization were abolished (Fig. 5, Bc and C).
Effect of Intracellular pH-To study whether the site of proton action is extra-or intracellular, HEK293-hP2X 3 cells were microdialyzed with intracellular (IC) solutions buffered either at pH 7.4 or 5.8 via the patch-clamp pipette (Fig. 6). The amplitude of currents evoked by ␣,␤-meATP (1 M) at an extracellular pH of 7.4 did not depend on the IC pH (IC pH 7.4: 1.3 Ϯ 0.2 nA, n ϭ 10; IC pH 5.8: 1.3 Ϯ 0.3 pA, n ϭ 9; p Ͼ 0.05) (compare Fig. 6, Aa and Ab). Moreover, the amplitude of the current evoked by 1 M of ␣,␤-meATP was inhibited by a shift in the pH of the extracellular solution from 7.4 to 5.8 to a similar extent, irrespective of whether the pH of the IC solution was 7.4 or 5.8 (by 69 Ϯ 5 and 74 Ϯ 5%, respectively, p Ͼ 0.05; Fig. 6B).
Voltage Sensitivity-To test whether protons alter the ion permeation ratio of P2X 3 receptor channels, the effect of protons on the reversal potential of ␣,␤-meATP current was also investigated. 10 M of ␣,␤-meATP was applied over a range of holding potentials (Ϫ80 to ϩ40 mV, 20 mV increments) at pH 7.4 and pH 5.8. Fig. 6C shows the voltage-current relationship for ␣,␤-meATP-activated currents at different pH values. Changing the pH did not alter the reversal potential of the currents activated by ␣,␤-meATP (1 Ϯ 2 mV at pH 7.4 and Ϫ1 Ϯ 2 mV at pH 5.8; p Ͼ 0.05; n ϭ 6 -7).
Effect of Extracellular Protons on P2X 3 Currents in DRG Cells-Application of ␣,␤-meATP to small diameter (20 -35 m) rat DRG neurons evoked inward currents with different kinetic properties. Out of the total number of 42 neurons, 38% responded with rapidly desensitizing currents (fast-type).   Another 28% of the neurons responded with slowly desensitizing currents (slow type) and 34% could be classified into the intermediate type, with two-phase kinetics consisting of consecutive rapidly and slowly desensitizing components. The fasttype neurons were suggested to contain homomeric P2X 3 receptors, the slow-type neurons heteromeric P2X 2/3 receptors, and the intermediate-type neurons a mixture of P2X 3 and P2X 2/3 receptors (37)(38)(39). Only experiments with fast-type responses to ␣,␤-meATP were evaluated in the experiments below.

DISCUSSION
The present experiments demonstrated that protons are able to modulate homomeric P2X 3 receptors, constituting about 40% of the receptor population of rat DRG neurons. Lowering the pH slowed down both the kinetics of channel opening and the rate of desensitization to ␣,␤-meATP, facilitated the recovery from the desensitized state, and had a dual effect on the current amplitudes. Whereas the action of low agonist concentrations was inhibited by protons, that of high agonist concentrations was facilitated. Site-directed mutagenesis of histidine 206 at the extracellular loop of the P2X 3 receptor to alanine abolished the acid sensitivity. These results indicate that pH modulation of the hP2X 3 receptor is mediated by protonation of H206 and gives further explanation how local acidosis could attenuate or facilitate pain.
H ϩ ions are known to differentially modulate various P2X receptors. Acidification inhibited P2X 1 receptor currents (18) by increasing the EC 50 value for ATP without a concomitant change in the maximum ATP effect (40,41). P2X 4 receptor currents were also found to be inhibited by acidification (18,42). The EC 50 value of ATP was uniformly increased under these conditions, although some of the authors reported that the maximum effect of ATP remained constant (18), while other ones found it to become smaller (42). In contrast, protons potentiated both recombinant and native P2X 2 receptor currents by reducing the EC 50 value without altering the maximum effect of ATP (17,18,43).
Bullfrog DRG and rat nodose ganglion neurons are known to express homomeric P2X 1 , P2X 2 , and P2X 3 , as well as heteromeric P2X 2/3 receptors (44,45). Application of ATP onto these neurons evoked non-desensitizing currents which resembled P2X 2 or P2X 2/3 receptor activation and were enhanced by acidification (25,26,46). Lowering the pH reduced the EC 50 value, but did not alter the maximum effect of ATP for evoking these currents.
Only little is known about the modulation of homomeric P2X 3 receptors by protons. In an early study, recombinant rat homomeric P2X 3 receptors expressed in HEK293 cells and activated by a stable concentration of ATP (1 M) were found to be depressed by acidification to pH 6.3 (18). In another study, homomeric rat P2X 3 receptors expressed in Xenopus oocytes were activated by ATP; acidification to pH 5.5 inhibited the current amplitudes. This inhibition was due to a 15-fold increase in the EC 50 value for ATP without a concomitant change in the maximum ATP effect (40).
Here, we report that acidification to pH 5.8 caused a 1.6-fold increase in the Hill coefficient and a 2.2-fold increase in the EC 50 value for ␣,␤-meATP to activate human homomeric P2X 3 receptor channels. Besides these effects we also found that such a change in the extra-, but not intracellular pH considerably increased the maximum ␣,␤-meATP effect (I max ). Finally, acidification decreased the desensitization rate of the P2X 3 receptor and accelerated the speed of recovery from its desensitized state.
Protons are able to modulate P2X 3 currents by a number of possible mechanisms. These include: 1) altering the affinity of the agonist to bind to the receptor, 2) altering the efficacy of the agonist to activate the channel, 3) altering the rate of receptor desensitization, and 4) altering the ion permeation ratio.
First, acidification shifted the concentration-response curve to the right indicating that protons reduce the affinity of the agonist to its receptor. However, the macroscopic EC 50 value does not necessarily reflect a change in the microscopic affinity (i.e. the rate constants of association and dissociation) because it depends also on the effects of pH on the gating rate constants (opening, closing, desensitization, and recovery) (47). To answer the question more precisely, concentration dependence should be analyzed on the single channel level. However, the fast single channel kinetics of the receptor (48) makes it difficult to perform a proper analysis.
Second, lowering extracellular pH increased the maximal ␣,␤-meATP-evoked current indicating that protons act by increasing the efficacy of ␣,␤-meATP to activate the P2X 3 receptor. Just as an increased EC 50 value does not prove a decreased microscopic affinity (see above), an increased apparent efficacy does not reflect accurately how microscopic rates of opening and closing are affected, because apparent efficacy is also determined by (microscopic) agonist affinity, and by desensitization recovery rates. Third, the decay rate of P2X 3 receptors was clearly decreased by lowered pH, indicating that the channel desensitizes at a slower rate. This may be one of the causes of decreased apparent affinity and increased apparent efficacy. The amplitude of the P2X 3 receptor current is determined by on and des1 time constants, which both depend on the concentration of the agonist. Because the concentration versus on curves were less steep in our experiments than the concentration versus des1 curves, the amplitude of the current at low agonist concentrations is primarily determined by on while at high agonist concentrations primarily by des1 . This may explain that acidification at the same time can depress and facilitate ␣,␤-meATP effects. The inhibition of the response at low agonist concentration may be due to an effect on on , while the facilitation of the response at high concentrations may be due to an effect on des1 . Finally, protons did not alter the reversal potential of currents activated by ␣,␤-meATP, indicating that protons did not affect the ion selectivity of the channel.
All pH-dependent effects were abolished when the histidine 206 of the extracellular loop was mutated to alanine. Histidine residues at the extracellular loop of P2X 2 (H319) and P2X 4 (H286) receptors were found to play a role in the modulation of these subunits (42,49). His 206 is also located in the extracellular loop somewhat closer to the second transmembrane domain than in P2X 2 or P2X 4 . The pK a of free histidine is 6.0 (35) (in proteins its pK a can range from 5.0 to 8.0 (36)). Hence, our results suggest that protonation of the histidine residue 206 changes the energy balance of the channel and thereby slows down the transition into both open and desensitized states.
It is interesting to note that a bimodal action of protons on ATP currents has been reported for native P2X 2 receptors in rat PC12 cells (50). These authors found that acidification potentiated the effect of low ATP concentrations and attenuated the effect of high ATP concentrations. They explained these results by a facilitated binding of the agonist to resting as well as open receptors, which causes the potentiation at low concentrations and the increased fading (resulting in depression) at high concentrations. They suggested, but did not prove the role of extracellular histidines in this phenomenon. Our data concerning the behavior of P2X 3 receptors are fully compatible with these observations, although the effects observed by us were just of the opposite direction.
In conclusion, we have demonstrated a dual effect of acidification on P2X 3 channels in response to agonist application. The current amplitude was decreased at low agonist concentrations (due to a decrease in the rate of activation) and increased at high agonist concentrations (due to a decrease in the rate of desensitization). A balance between these two opposing influences may determine the function of P2X 3 receptor channels. Hence, the effect of low ATP concentrations may be alleviated during inflammation and the accompanying acidification, whereas the effects of large local concentrations of ATP caused by a massive tissue damage may be potentiated by a decrease of pH. This may result in a differential regulation of pain sensation by the C fiber terminals of small DRG neurons in peripheral tissues.