INTRODUCTION

Top Abstract Introduction Methods Results Discussion References

EXTRACELLULAR ATP MEDIATES many biological responses ranging from smooth muscle contraction to synaptic transmission in the autonomic and central nervous systems (4, 6). The biological effects of ATP are less well defined in hemopoietic cells, although extracellular ATP has been shown to stimulate granule exocytosis from mast cells and to induce apoptosis in thymocytes (5, 41). These diverse actions of ATP are mediated by the interactions of this nucleotide with P2 purinergic receptors, which can be broadly classified into those coupled via G proteins to phospholipase C (the P2Y class) and those for which the receptor is a ligand-gated ion channel (the P2X and P2Z classes). Many studies of the P2Z receptor subtype have been in hemopoietic cells such as macrophages and mast cells, which coexpress P2Z and either P2Y₂ or P2Y₁ receptors (1, 6, 8, 14, 31). To distinguish between effects mediated by these two P2 receptors, an agonist specific for the P2Z subtype such as 3'-O-(4-benzoyl)benzoyl-adenosine 5'-triphosphate (BzATP) is generally used, although this agonist is a weak agonist for P2Y₂ receptors (2) and a partial agonist for some P2X receptors (31). Activation of the P2Z receptor of human lymphocytes produces multiple membrane effects, including ionic fluxes, stimulation of phospholipase D, and activation of a membrane protease. In both normal and leukemic lymphocytes, the P2Z purinoceptor is part of or closely coupled to an ionic channel that shows strong selectivity for the divalent cations Ca²⁺ and Ba²⁺, and, because divalent cation fluxes are enhanced in KCl or other Na⁺-free media, it is likely that Na⁺ exerts a weak inhibitory effect on the channel as well as being a permeant ion (3, 23, 26, 38-40). Anion permeability of this channel has been reported to be negligible (22). The P2Z purinoceptor shows an absolute specificity for the ATP⁴ species, since addition of Mg²⁺ in molar excess over nucleotide leads to closure of the channel in <1 s (39). Suramin is the best known inhibitor of P2 receptors but is only weakly active against the P2Z receptor, with an IC₅₀ value of 60 µM (39). In contrast, isoquinoline sulfonamides such as KN-62 are potent inhibitors of the human P2Z receptor, with an IC₅₀ of 13 nM (11), whereas amiloride analogs such as 5-(N, N-hexamethylene)-amiloride (HMA) have intermediate antagonist potency (40).

Recently the genes for both the rat and human P2Z receptors were cloned and expressed in HEK-293 cells (27, 32). Both have homology to the six described members of the P2X family of receptors, and it is proposed that P2Z be termed P2X₇. All members of this family have protein structures with two putative transmembrane segments linked by a long extracellular loop and intracellular NH₂- and COOH-termini. However, the number of P2Z/P2X₇ monomers required to form an oligomeric channel through the membrane is unknown. The P2Z/P2X₇ receptor has a long COOH-terminal tail that confers the property of permeation by large fluorescent cationic dyes as well as smaller cations through this channel (27, 32). A recent study of Xenopus oocytes injected with macrophage mRNA and stimulated with BzATP showed two conductance states. There was a small, rapidly activated inward current carried by monovalent cations, which was followed 1 min later by a large current carried by organic cations (N-methyl-D-glucamine or Tris) in addition to the monovalent cations (25). The detailed kinetics of this transition have not been examined.

Cells often coexpress several P2 receptor subtypes, so multiple pathways may be available for ATP-stimulated fluxes. However, competitive experiments with BzATP and ATP in lymphocytes from patients with chronic lymphocytic leukemia show that the permeability responses are dominated by the P2Z/P2X₇ receptor (10). At this receptor, the natural ligand ATP is a partial agonist, producing only 70% of the maximal response in ethidium cation influx observed for the full agonist BzATP. Several other ATP analogs such as 2-methylthioadenosine 5'-triphosphate (2-MeS-ATP) and adenosine 5'-O-(3-thiotriphosphate) (ATPS) are partial agonists that stimulate P2Z-mediated responses in lymphocytes but with even lower maximal activities than BzATP or ATP, such that the rank order of agonist efficacy is BzATP > ATP > 2-MeS-ATP > ATPS (10). In this study, the increase in permeability of the P2Z channel was measured using time-resolved flow cytometry with ethidium as the permeant ion and compared with the uptake of isotopic ¹³³Ba²⁺. The results suggest that partial agonists, including ATP, produce a biphasic opening of the P2Z channel for which the second increment in permeability develops slowly over 0.5-1.5 min.

	METHODS

Top Abstract Introduction Methods Results Discussion References

Materials. Ficoll-Paque (density 1.077) was obtained from Pharmacia (Uppsala, Sweden). ATP, BzATP, ethidium bromide, BaCl₂, phorbol 12-myristate 13-acetate, and BSA were from Sigma Chemical (St. Louis, MO). The amiloride analog HMA and KN-62 {1-[N,O-bis(5-isoquinoline sulfonyl)N-methyl-L-tyrosyl]-4-phenylpiperazine} were from Research Biochemicals (Natick, MA), and ¹³³Ba²⁺ (0.5 mCi/ml, 3 mCi/mg) was from DuPont (Boston, MA). HMA was kept dessicated at room temperature in the dark, since traces of moisture lead to loss of activity. Stock solutions of HMA and KN-62 were prepared in dry DMSO. HMA solutions were prepared immediately before use, whereas stocks of KN-62 were stored at 4°C for up to 1 mo. Di-n-butyl phthalate and di-iso-octyl phthalate were from British Drug Houses (Poole, UK). HEPES was from Boehringer Mannheim (Mannheim, Germany). Streptolysin O was from Wellcome Reagents. YO-PRO-1 iodide was from Molecular Probes (Eugene, OR).

Source of cells. Peripheral blood lymphocytes were obtained from six patients with B-cell chronic lymphocytic leukemia whose cells showed permeability responses in our previous studies (38-40). Liver tissue was freshly dissected from diagnostic samples taken intraoperatively, whereas peripheral blood monocytes were obtained from peripheral blood drawn from normal subjects.

Lymphocyte preparation. Venous blood from patients was diluted with an equal volume of HEPES-buffered saline (pH 7.4; 10 mM HEPES, 0.1% BSA, 5 mM D-glucose, 145 mM NaCl, and 5 mM KCl). Mononuclear cells were separated by density gradient centrifugation over Ficoll-Paque and washed twice, and monocytes were depleted by adhesion to plastic. Cytocentrifuge preparations showed that >98% of cells were small mature lymphocytes. Immunophenotype analysis of these lymphocytes showed 94 ± 4% B cells (CD5⁺, CD19⁺) and 3 ± 2% T cells (CD3⁺).

Monocyte/macrophage preparation. Mononuclear cells from normal subjects were prepared as above, and monocytes were allowed to adhere to plastic flasks for 1 h at 37°C. Nonadherent cells were then washed away, and adherent cells were cultured for 7 days in RPMI 1640 medium with 20% FCS plus interferon- (100 U/ml).

RT-PCR analysis. Total RNA was isolated from cells using RNA isolation reagent (Advanced Biotechnologies) according to the manufacturer's recommendations. For cDNA synthesis, 1 µg of total RNA was used in the reverse transcription reaction with 0.5 µg of primer oligo(dT) with 200 units SUPERSCRIP II RT (Life Technologies), 25 units RNase inhibitor, 2 mM dNTPs, 50 mM Tris · HCl (pH 8.3), 75 mM KCl, 3 mM MgCl₂, and 10 mM dithiothreiotol in a total volume of 29.5 µl. Primers for RT-PCR were designed from the human P2X₇ mRNA sequence in the area of the COOH-terminal tail for bp 1425-1780 (GenBank accession no. Y09561), using the computer program PRIMER (version 0.5; S. E. Lincoln, M. J. Daly, and E. S. Lander, unpublished). The forward primer was bp 1425-1446 (5'-ACTCCTAGATCCAGGGATAGCC-3'), and the reverse primer was bp 1780-1759 (5'-TCACTCTTCGGAAACTCTTTCC-3'). For PCR amplification, 30-120 ng of cDNA were used as template DNA in a final volume of 25 µl, containing 0.75 units Taq DNA polymerase, 2 mM MgCl₂, 100 µM dNTPs (dATP, dGTP, dCTP, and dTTP), 1× Taq DNA polymerase buffer, and 10 pmol of each human P2X₇ (hP2X₇) primer. The reaction solution above was overlaid with one drop of mineral oil, and the PCR was carried out with a thermocycling program as follows: initial denaturation at 95°C for 5 min, and then 35 cycles of 95°C for 30 s, 55°C for 30 s, and 72°C for 1 min; the final extension step was carried out at 72°C for 10 min. The RT-PCR products were electrophoresed in a 2% agarose gel at 100 V for 1 h. The size of the fragment was determined with a 100-bp DNA molecular weight ladder track.

Uptake of ¹³³Ba²⁺. Ba²⁺ is a good surrogate for Ca²⁺ and once inside the cell is neither pumped nor sequestered by transport ATPases (39). Ba²⁺ uptake was measured over 60 s using ¹³³BaCl₂ (final concentration, 0.2 mM). At time 0, a prewarmed stock solution of ¹³³Ba²⁺ (0.4 mM and 1 µCi/ml) was added in equal volumes to prewarmed lymphocytes (2.5 × 10⁷ cells/ml, in 150 mM KCl with HEPES, pH 7.4) at 37°C or 24°C. ATP (1 mM) was added either 10 min before or simultaneously with the ¹³³Ba²⁺ isotope. Aliquots of 0.8 ml taken at time points between 0 and 60 s were immediately mixed with 0.2 ml of ice-cold 50 mM MgCl₂ (in KCl-HEPES medium) that had been previously layered over 250 µl of oil mixture (di-n-butyl phthalate and di-iso-octyl phthalate, 7:3 vol/vol) and then centrifuged at 8,000 g for 30 s. The supernatants and the oil were aspirated, and the cell pellets were counted in a Wallac Wizard 3 automatic gamma-counter. The uptake was converted to picomoles per 10⁷ cells using the relative specific activity of the ¹³³Ba²⁺ stock.

Ethidium and YO-PRO-1 measurements by flow cytometry. Flow cytometry was used to quantitate the uptake of ethidium bromide (39). Cells were suspended at 10⁶ cells/ml in medium containing 150 mM KCl buffered with 10 mM HEPES (pH 7.4), and either 1) the cells were incubated for 10 min at 37 or 24°C with or without ATP (0.5 mM) before addition of 25 µM ethidium or 2) ATP was added 30 s after ethidium. Samples in which ethidium uptake was measured were kept stirred and with the temperature controlled at 37°C (unless otherwise stated) using a Time Zero module (Cytek, Fremont, CA). Histograms (256 channels) of lymphocyte-associated fluorescence signals were collected over consecutive 6-s intervals for 5 min using a Coulter Elite flow cytometer (Coulter, Hialeah, FL) with argon laser excitation at 488 nm. Fluorescent emission was collected using a 590-nm long-pass filter. The mean channel of fluorescence intensity was then calculated for each of the histograms collected for each of the 6-s intervals and plotted against time. Uptake of YO-PRO-1 (5 µM) was measured by an identical procedure.

Data presentation. Data are expressed as means ± SE.

	RESULTS

Top Abstract Introduction Methods Results Discussion References

Lymphocyte P2Z is the hP2X₇ receptor. Total RNA from human lymphocytes from six different patients was subjected to RT-PCR using primers specific for the long COOH-terminal tail of the hP2X₇. Agarose gel electrophoresis showed a single oligonucleotide product, which migrated at the predicted 356-bp position (Fig. 1). Single bands at the same position were also observed for RT-PCR products from RNA of macrophages and liver, which also express P2X₇ (27, 32). One feature of hP2X₇ heterologously expressed in HEK-293 cells is that it allows ATP-induced permeation of the fluorescent dye YO-PRO-1 (375.5 Da) (27, 32). Human lymphocytes incubated with ATP (500 µM) took up the divalent YO-PRO-1 cation at a rate comparable to that for the smaller, monovalent ethidium cation (data not shown).

View larger version (46K): [in this window] [in a new window]   Fig. 1.   Agarose gel electrophoresis of RT-PCR products from liver, lymphocyte, and macrophage cDNA using human P2X7 (hP2X7) purinoceptor-specific primers; 2% agarose gel electrophoresis reveals a single 356-bp product following amplification by PCR using hP2X7-specific primers and cDNA from lymphocytes of 6 different patients (lanes 1-4, 9, and 10), macrophages (lanes 5 and 6), liver (lanes 7 and 8), and H2O (lane 11). M, molecular weight markers (100-bp ladder).

ATP-induced influx of ¹³³Ba²⁺. The Ba²⁺ permeability of lymphocytes at 37°C was compared immediately after ATP addition and after a 10-min preincubation with ATP. Figure 2A shows that the ¹³³Ba²⁺ uptake into lymphocytes incubated with ATP commenced within 10 s and continued throughout 60 s for each condition. The influx of ¹³³Ba²⁺ was greater for cells preincubated with ATP than for cells to which coaddition of ATP and ¹³³Ba²⁺ was made at time 0. Under both conditions, the basal ¹³³Ba²⁺ influx was negligible. The ¹³³Ba²⁺ influx under both conditions was also measured at 24°C (Fig. 2B), and again the ATP-induced ¹³³Ba²⁺ influx was greater for cells preincubated with ATP than for cells to which ATP and ¹³³Ba²⁺ were added simultaneously. The relative difference between the two conditions was greater at 24°C than at 37°C.

View larger version (24K): [in this window] [in a new window]   Fig. 2.   Extracellular ATP-induced isotopic Ba2+ influx into fresh leukemic lymphocytes measured either at 37°C (A) or at 24°C (B). Cells were suspended at 1.2 × 107 cells/ml in HEPES-buffered isotonic KCl medium with no added Ca2+. Either ATP (1.0 mM) was added to suspension 10 min before 133Ba2+ (0.2 mM) or ATP and 133Ba2+ were added simultaneously at time 0.

ATP-induced uptake of ethidium measured by flow cytometry. We have previously shown that ethidium is a large organic permeant ion for the lymphocyte P2Z/P2X₇-operated ion channel and that uptake of this cation can be readily followed by time-resolved flow cytometry (10, 39). Lymphocytes were incubated at 37 and 24°C for 10 min, either with or without ATP (0.5 mM). In cells preincubated with ATP, influx was initiated by addition of 25 µM ethidium bromide, and uptake of this dye was compared with that into cells to which ethidium cation was added just before ATP (Fig. 3). Cells preincubated with ATP showed a greater influx of ethidium than cells for which ATP was added at time 0. However, the uptake of ethidium was linear with time only for those cells that had been preincubated with ATP. In contrast, a delay phase was observed before ethidium uptake commenced in cells exposed to ATP at time 0, and the delay was more prominent and prolonged at 24°C than at 37°C (Fig. 3, A and B). At 12°C, the ATP-induced uptake of ethidium was insignificant whether or not cells were preincubated for 30 min with ATP at this temperature (data not shown).

View larger version (19K): [in this window] [in a new window]   View larger version (17K): [in this window] [in a new window]   Fig. 3.   Extracellular ATP-induced uptake of ethidium cation into fresh lymphocytes measured either at 37°C (A) or at 24°C (B). Cells were suspended at 106 cells/ml in HEPES-buffered KCl medium with no added Ca2+ and incubated with or without ATP (0.5 mM) at designated temperature. In cells with ATP preexposure, an addition of ethidium bromide (Et; 25 µM) was made as indicated. In cells without ATP preexposure, additions of ethidium cation (Et+) and ATP were made in quick succession as indicated. Mean channel cell-associated fluorescence intensity was measured at 6-s intervals.

Partial agonists produce ethidium uptake only after a delay. The time course of ethidium uptake into lymphocytes was compared for ATP and its three analogs BzATP, 2-MeS-ATP, and ATPS added at maximal effective concentrations immediately after 25 µM ethidium addition. Ethidium influx commenced immediately after addition of the most efficacious agonist, BzATP (50 µM), at a rate that was linear with time (Fig. 4A). In contrast, a delay was observed before uptake of ethidium commenced after the addition of ATP (500 µM), 2-MeS-ATP (500 µM), and ATPS (2.0 mM). The delay before maximal ethidium influx was reached was estimated for each agonist by extrapolation of the line of maximal slope back to its intersection with the line of basal ethidium influx and measurement of the time interval between agonist addition and this intercept. Figure 4B shows the inverse relationship between this estimate of delay time and the maximal ethidium influx achieved for each agonist (J) relative to that of BzATP (J_max, taken as 100%). All three partial agonists for the P2Z/P2X₇ receptor allowed ethidium cation entry, but only after a delay, which became longer as the efficacy of the partial agonist became less.

View larger version (21K): [in this window] [in a new window]   View larger version (14K): [in this window] [in a new window]   Fig. 4.   Agonist-induced ethidium cation uptake at maximal concentrations. A: cells (106 cells/ml) suspended in HEPES-buffered KCl medium at 37°C were incubated with 50 µM 3'-O-(4-benzoyl)benzoyl-adenosine 5'-triphosphate (BzATP), 500 µM ATP, 500 µM 2-methylthioadenosine 5'-triphosphate (2-MeS-ATP), or 2 mM adenosine 5'-O-(3-thiotriphosphate) (ATPS) added 30 s after 25 µM ethidium. Control cells were incubated with 25 µM ethidium alone. Mean channel cell-associated fluorescence was measured at 6-s intervals by flow cytometry. Results are from 1 experiment representative of 3 or 4. B: maximal rates of ethidium uptake for each agonist (J) calculated as a percentage of response to 50 µM BzATP (Jmax) are plotted against length of delay phase of uptake. Line of best fit through data points from 4 experiments was calculated by least squares analysis. Correlation between J/Jmax and length of delay phase was 0.96 (Pearson's r; P < 0.001).

Delay phase with low BzATP concentrations. Although no delay phase was observed for BzATP concentrations (50 µM) producing maximal responses, a delay in ethidium uptake became apparent when submaximal concentrations of BzATP were added 30 s after ethidium (Fig. 5). Furthermore, no delay was observed when cells were incubated with BzATP (10-20 µM) for 10 min before the addition of ethidium (data not shown).

View larger version (22K): [in this window] [in a new window]   Fig. 5.   BzATP-stimulated ethidium cation uptake. Cells (106 cells/ml) suspended in HEPES-buffered KCl medium at 37°C were incubated with BzATP added 30 s after ethidium, and mean channel cell-associated fluorescence was measured at 6-s intervals by flow cytometry. Control cells were incubated with 25 µM ethidium alone. Results are from 1 experiment representative of 3.

P2Z/P2X₇ inhibitors lengthen the delay phase. We have previously shown that the isoquinoline derivative KN-62 is a potent antagonist of the P2Z/P2X₇ receptor and inhibits ATP-stimulated ethidium uptake with an IC₅₀ of 13.1 ± 2.6 nM and with complete inhibition at 500 nM (11). However, KN-62 is a less potent inhibitor of BzATP, and even 2,000 nM KN-62 blocked only 70% of the ethidium influx (Fig. 6). In addition, Fig. 6 shows that KN-62 added 10 min before ethidium (25 µM) and BzATP (50 µM) not only inhibited ethidium influx through the P2Z channel but lengthened the delay before significant ethidium uptake was detected.

View larger version (20K): [in this window] [in a new window]   Fig. 6.   KN-62 delays BzATP-stimulated ethidium cation uptake. Fresh lymphocytes were suspended at 106 cells/ml in buffered KCl medium and preincubated with or without KN-62 for 10 min at 37°C, followed by additions of 25 µM ethidium and 50 µM BzATP as indicated. Mean channel cell-associated fluorescence intensity was measured at 6-s intervals by flow cytometry. Results are from 1 experiment representative of 4.

View larger version (19K): [in this window] [in a new window]   Fig. 7.   Inhibition of ATP-induced uptake of ethidium cation by the amiloride analog 5-(N, N-hexamethylene)-amiloride (HMA). Fresh lymphocytes were suspended at 106 cells/ml in buffered KCl medium without Ca2+ at 37°C. Additions of HMA (0-30 µM) were made, followed immediately by ethidium bromide (25 µM) and ATP (0.5 mM) as indicated. Mean channel cell-associated fluorescence intensity was measured at 6-s intervals.

	DISCUSSION

Top Abstract Introduction Methods Results Discussion References

Our study has shown that the P2Z responses of human lymphocytes (3, 10, 11, 22, 38-40) are mediated by the hP2X₇ receptor that was recently cloned from a human monocyte library (27) and is the latest member of the P2X family of receptors. RT-PCR analysis shows an identically sized product for human lymphocytes and monocytes, and the rank order of agonist efficacy and the upper limit of permeant ion size are also the same for the P2Z/P2X₇ channel in its native lymphocyte and for HEK-293 cells heterologously expressing the cDNA for hP2X₇.

The major finding of this study is the slow further increase in permeability that develops after an initial opening of the P2Z/P2X₇ receptor channel induced by ATP and other partial agonists. The data in Fig. 2 show that ATP induces an immediate ¹³³Ba²⁺ influx into cells at both 37 and 24°C, which is in line with electrophysiological data showing rapid (<1 s) channel opening following ATP addition (3, 25, 31, 34). This influx occurs only via the P2Z/P2X₇ ion channel, since inhibitors of L-type voltage-gated Ca²⁺ channels are without effect on ATP-induced Ba²⁺ influx (39). ¹³³Ba²⁺ influx into lymphocytes was always greater for cells preincubated with ATP than for cells to which simultaneous addition of ATP and ¹³³Ba²⁺ were made (Fig. 2), suggesting a second increase in permeability to Ba²⁺ had developed within 2 min at 37°C. This second permeability increase could be conveniently studied with the larger ethidium cation, since the large size (314 Da) of this cation used at low concentrations (25 µM) revealed the slow kinetics of channel enlargement. Previous studies by Gomperts, Tatham, and colleagues (13, 33, 34) measured uptake of ethidium cation by the P2Z/P2X₇ ion channel of mast cells using a fluorometric technique that relies on the enhancement of fluorescence after ethidium cation enters the cell and binds to nucleic acids, a step that eliminates any backflux of the permeant ion. We have adapted this method to flow cytometry, which offers unique advantages for studying permeability states of the P2Z/P2X₇ ion channel. Time-resolved flow cytometry generates the mean fluorescence intensity of 3,000 cells per 6-s interval while mainly cell-associated fluorescence is measured, since very little of the bulk medium is illuminated in the laser light path. Thus this technique allows a sensitive measurement of the initial rates of ethidium cation uptake, which are essentially unidirectional because of binding of permeant ions to nucleic acids. Figure 3, A and B, shows that the rapidly activated P2Z/P2X₇ channel is initially impermeable to ethidium cation, but within the first 1 min after addition of ATP a steady increase in ethidium influx was observed. This delay and slow increase in the rate of ethidium cation uptake contrast with the immediate uptake of ¹³³Ba²⁺ observed following ATP addition and suggest that the initial P2Z/P2X₇ channel, once opened, is slowly modified to accept larger permeant ions. Uptake of ethidium after ATP addition was strongly temperature dependent (Fig. 3), and no uptake occurred at 12°C. A strong temperature dependence has also been noted previously for uptake of Lucifer yellow through the P2Z/P2X₇ channel of murine macrophages, with no dye entering at 18°C (30). A recent patch-clamp study of human B lymphocytes showed that P2Z/P2X₇ unitary channel currents could be carried by Na⁺ or Cs⁺ but not by choline ions (22). However, the reduced temperature (22°C) and short time frame (~2 min) of these measurements may not have allowed sufficient time to elapse for channel enlargement.

We and others have previously shown that the influx of ethidium stimulated by maximally effective agonist concentrations follows a rank order BzATP > ATP > 2-MeS-ATP > ATPS (10, 32). These data together with the results of competitive experiments show that the natural ligand ATP is a partial agonist for the P2Z/P2X₇ receptor, whereas BzATP is a full agonist (10). Two other less potent agonists for the P2Z/P2X₇ receptor, 2-MeS-ATP and ATPS, are also partial agonists, and all three partial agonists stimulate the uptake of ethidium only after a delay phase of 30-90 s (Fig. 4A). Partial agonism is a well-recognized feature of many receptor-operated channels at which the partial agonist, added at maximally effective concentrations, produces less permeant ion flux (J) than does the full agonist (J_max). This lower maximal response (J/J_max) is a measure of the intrinsic efficacy of the partial agonist, and it is striking that J/J_max shows an inverse relationship to an estimate of the duration of the delay phase (Fig. 4B). It is generally accepted that partial agonists are unable to induce the same change in receptor conformation as full agonists and thus maintain a lower proportion of the channels in the open conformation (21).

Although the full agonist BzATP (50 µM) produced an immediate influx of ethidium with no sign of delay, lower BzATP concentrations (10-20 µM) stimulated ethidium uptake only after significant delays of up to 30 s (Fig. 5). Such low concentrations of the full agonist also maintain a lower proportion of the channels in the open state. A similar effect was observed with the inhibitors KN-62 and HMA, both of which slowed agonist-induced ethidium influx with an associated increase in the delay before significant ethidium entry could be detected (Figs. 6 and 7). Taken together, these data support a model proposed by Tatham and Lindau (34) and by Nuttle and Dubyak (25) in which agonists for the P2Z/P2X₇ receptor activate immediate opening of an ion channel (P2Z_open), followed by a slower transition to a permeability pathway that passes larger cations up to the size of ethidium and YO-PRO-1 cations (P2Z_pore). Thus in which the proportion of channels in the open conformation is determined by the efficacy of agonist as well as agonist concentration. Our data show that the rate of formation of the larger "pore" is slow (k₂  k₁), so that the model predicts the second step (or series of steps) occurs at a rate (k₂ × P2Z_open) that is largely determined by the abundance of the open channels in the membrane. Amiloride analogs, even at high concentrations (30-100 µM), block only 70-85% of the flux of monovalent or divalent cations through the P2Z/P2X₇ channel, despite having inhibitor constants of ~2-10 µM (25, 40). A similar and incomplete inhibition of ethidium influx by the amiloride analog HMA was observed in this study (Fig. 7). Even more striking was the finding that the most potent inhibitor described to date, KN-62, could only inhibit 75% of the BzATP-induced ethidium influx (Fig. 6) although KN-62 could completely abolish the ATP-induced ethidium influx (11). One explanation is that the two permeability states of the channel have different sensitivity to inhibitor (25). However, it is equally possible that a fraction of the P2Z/P2X₇ pores, once formed, develop firm interactions with other components of the plasma membrane (k₂ becomes smaller) and that these complexes resist the action of inhibitors. Recently it has been reported that calmidazolium inhibits agonist-induced ionic currents through the hP2X₇ channel by up to 90% with little effect of inhibitor on YO-PRO-1 cation uptake (37). Although this finding supports the suggestion that conformational changes accompany channel-to-pore transition (32), confirmatory findings are needed in a cell type expressing native hP2X₇.

There is electrophysiological evidence that the macrophage P2Z/P2X₇ receptor ion channel has variable conductance states. Thus single channel recordings of patch-clamped membranes showed a number of conductance sublevels ranging from 3.5 to 15 pS with Ba²⁺ as the permeant cation (22, 23). Multiple conductance states for receptor-operated single channels have also been documented for the nicotinic ACh receptor (15) and for -aminobutyric acid-operated Cl channels (7, 9). However, slow changes in channel conductance have not been reported at 37°C over the time frame (1-5 min) of the present ethidium uptake studies. There is good evidence for phosphorylation of ion channel subunits (20), and it is possible that the phosphorylation state may influence the channel flux. Thus the permeability of the P2Z/P2X₇ receptor channel of mouse lacrimal acinar cells or mast cells is potentiated by stimulation of cAMP-dependent protein kinase activity (24, 29), although there is conflicting evidence (35). Figure 6 shows that the inhibitor KN-62 at 10 nM induces a delay before BzATP-induced uptake of ethidium is observed. Although KN-62 inhibits Ca²⁺/calmodulin-activated kinase II with an IC₅₀ of 0.9 µM (36), phosphorylation of the channel by this enzyme is unlikely to be involved, since KN-62 blocks the receptor at a concentration too low to inhibit this kinase. Another possibility is that oligomerization of open P2Z/P2X₇ monomeric channels might occur on ATP binding and modulate the permeability pathway to allow larger cations to permeate. Aggregation of channel monomers to form a larger oligomer pore has been suggested previously for the P2Z/P2X₇ channel of mast cells (34) and by Nuttle and Dubyak (25) for the P2Z/P2X₇ channel of macrophages. There is evidence that the long COOH-terminal tail of the P2Z/P2X₇ receptor is required for permeation through the channel by large fluorescent dyes (32), although the molecular mechanisms involved are unclear.

There is increasing evidence for protein-protein interactions in the plasma membrane between receptor/channel monomers that may allow their recruitment and assembly into larger complexes. Thus the nicotinic ACh receptor of Torpedo membranes undergoes ligand-induced dimerization (19), whereas the T-cell receptor for antigen undergoes spontaneous oligomerization on engagement of peptide-major histocompatibility complex molecules (28). Recently the postsynaptic density protein PSD-95 has been shown to mediate the clustering of both N-methyl-D-aspartate receptors (18) and K⁺ channels (17) via specialized domain interactions. However, it is not known whether oligomerization of these channels allows the permeation of larger molecules. Whatever the mechanism, the use of time-resolved flow cytometry allows the slow second-phase permeability changes of the P2Z/P2X₇ channel to be studied, with results that suggest that the formation of the P2Z/P2X₇ "pore pathway" occurs at a rate that is proportional to the fraction of P2Z/P2X₇ receptors that are held in the open channel state.