INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

THE MEMBRANOUS LABYRINTH separates two fluids of differing composition. The luminal compartment is filled with endolymph, a K⁺-rich, Na⁺-low, and hyperosmotic fluid, whereas the surrounding spaces are filled with perilymph whose composition resembles that of an extracellular fluid, and a positive transepithelial potential is recorded between the endolymphatic and perilymphatic compartments. Both the high K⁺ concentration of the endolymph and the positive transepithelial potential are involved in the transduction mechanisms at the apical side of the sensory cells. Endolymph composition is maintained by a K⁺ transport localized in nonsensory epithelia of the inner ear, i.e., the dark cells in the vestibule and the strial marginal cells in the cochlea (for review, see Ref. 27).

Evidence has been provided for regulating roles of extracellular ATP and UTP on both neurotransmission and endolymph homeostasis in inner ear organs (2, 22, 28). ATP has been found in perilymphatic and endolymphatic fluids of the cochlea (24), and it regulates the transduction processes of sound and motion stimuli and controls the functions of the secretory epithelial cells (20, 28). For UTP actions, a UTP-sensitive phosphoinositidase C activity has been identified in ampulla from Rana ridibunda semicircular canal (6) and it was reported that UTP decreases the vectorial transport of K⁺ toward the endolymphatic space in the gerbil vestibule and cochlea (20). It has been assumed that the phosphoinositidase C stimulation after UTP-sensitive P2(Y) receptor occupancy triggers direct activation of protein kinase C through diacylglycerol-dependent and inositol trisphosphate- and Ca²⁺-independent mechanisms (22, 23).

The aim of the present work was to characterize the pharmacological properties of the pyrimidine nucleotide-sensitive P2(Y) receptors expressed in ampulla of R. ridibunda semicircular canal by making direct [³H]UTP binding experiments on single ampullas isolated by microdissection and screening the abilities of unlabeled structural nucleotide analogs more specific for the P2(X) ligand-gated ion channels or P2(Y) receptors linked to G protein signal pathways (13, 14) to inhibit radioligand binding and to stimulate phosphoinositidase C in myo-[³H]inositol-loaded ampullas. The binding study was based on previous findings performed on rat lung membranes demonstrating that [³H]UTP-labeled membranous binding sites recognize specifically pyrimidine nucleotides (17). Our results suggest that the population of ampullary [³H]UTP-labeled binding sites is heterogeneous and contains a minor class of high-affinity UTP binding sites whose functions remain to be determined and a major class of P2(Y)-like receptors exhibiting moderate affinity for UTP and triggering phosphoinositidase C activation.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Products Used

Other chemicals were provided from the following sources: adenosine, AMP, cAMP, ADP, ATP, adenosine 5'-O-(3-thiotriphosphate) (ATPS), 2'- and 3'-O-4-(benzoylbenzoyl)-ATP (Bz-ATP), ,-methylene-ATP (,-Me-ATP), diadenosine tetraphosphate (Ap₄A), cytidine, CTP, dTTP, UMP, UDP, UTP, bacitracin, BSA, DIDS, reactive blue 2 (basilen blue E-3G), Coomassie blue, ouabain, and sodium azide from Sigma (St. Louis, MO); pyridoxal-phosphate-6-azophenyl-2',4'-disulfonic acid (PPADS), and 2-methylthio-ATP (2-MeS-ATP) from Research Biochemicals International (Natick, MA); suramin from Miles (Naperville, IL); hexokinase, creatine kinase, and creatine phosphate from Boehringer (Mannheim, Germany); urethan from Prolabo (Paris, France); and Dowex resin AG₁-X₈ (200-400 mesh, formate form) from Bio-Rad Laboratories (Richmond, CA).

Animals

Microdissection of Ampullas from Frog Semicircular Canals

[³H]UTP Binding Studies

Binding assays. Assays were performed using a microtechnique reported earlier with slight modifications (7). Ampullas were first washed at 4°C in a large volume of medium B devoid of divalent cations (in mM: 40 HEPES-NaOH, pH 7.5; 0.25 EDTA-NaOH, pH 7.5; 82 NaCl, 3 KCl, 0.33 Na₂HPO₄, 0.44 NaH₂PO₄, 5 glucose, and 10 sodium acetate). For each single determination, one ampulla was transferred onto a 10-µl droplet of medium C (medium B containing 0.1 mM ouabain, 10 mM sodium azide, 50 nM [³H]UTP, and various amounts of unlabeled UTP analogs or antagonists) put on the hollow of a siliconized bacteriological slide, and samples were tightly covered with a petroleum jelly-coated slide to obtain a watertight seal. The binding reaction was carried out for 4 h at 4°C and stopped by adding 200 µl chilled medium B. The tissue sample was removed, washed five times in 200 µl of medium B at 4°C, and a 5-µl droplet containing the ampulla was sucked out from the last rinse, whereas an equivalent volume of this rinse was used as blank reference (the duration of washing was ~1 min). Then, samples were treated with 10 µl 1 N NaOH and 200 µl 30% Coomassie blue for protein content determinations using BSA as a standard. The latter was 2.50 ± 0.07 µg protein/ampulla (mean ± SE of mean data from 33 different experiments with 35-65 individual determinations). Protein assay media and ampullas were removed, transferred in counting vials, and counted twice for 20 min by liquid scintigraphy.

Calculations. Specific binding capacities were computed by

(1)

where RL is the specific binding measured; X, Y, and B are the radioactivities measured for total binding, nonspecific binding, and blank samples; P₁ and P₂ are the protein contents of ampullas used for total and nonspecific binding determinations, and SRA is the specific radioactivity of the radioligand. Binding capacities were expressed as 10¹⁵ moles [³H]UTP bound per microgram protein (fmol [³H]UTP bound/µg protein).

(2)

(3)

(4)

(5)

Phosphoinositidase C Studies

Enzymatic assays. Assays were performed using a microtechnique described earlier (6). Ampullas were labeled in 40 µl medium D [medium A containing 1 mM cytidine and 0.32 nmol myo-[³H]inositol (~1 MBq)] for 2.5 h at 27.5°C in a humid atmosphere and then extensively washed in a large volume of medium A at room temperature to eliminate extracellular myo-[³H]inositol.

Measurements of [³H]inositol phosphates by anion exchange chromatography. Radioactivities associated to free inositol, glycerophosphoinositol (GroPIns), and total inositol phosphates (InsP_s) were separated as reported earlier (4). Samples were applied to columns containing 0.25 g Dowex resin AG₁-X₈ and free inositol, GroPIns, and InsP_s were, respectively, eluted with 8 ml of the following solutions: 1) 3 mM HEPES-NaOH, pH 7.0, 2) 30 mM ammonium formate, and 3) 1 M ammonium formate and 0.1 M formic acid.

Calculations. Results were expressed as ratios between radioactivities measured in a given inositol-containing pool and in all labeled inositol-containing cellular pools (the latter was: 23,254 ± 3,865 cpm/ampulla; means ± SE of results from 19 independent experiments).

(6)

(7)

(8)

(9)

Control experiments. Accurate screening of receptor stereospecificity for recognition of structural UTP analogs could be hampered by impurities contained in commercial nucleotide preparations (18, 19) or by potential ectoenzyme-catalyzed interconversion between triphospho- and diphosphonucleotides (11, 12). These ectonucleotidase activities have been shown to occur in guinea pig cochlea (29). Thus control experiments were run to check the possible effects of an UTP contaminant on phosphoinositidase C response to UDP by pretreatment of 1 mM UDP with 10 IU/ml hexokinase and 25 mM glucose in medium A for 1 h at 37°C (18) and, conversely, to investigate the effects of a UTP-regenerating system (0.1% creatine kinase and 20 mM creatine phosphate) on UTP-stimulated enzyme activity.

Statistical Analysis

Generally, for each experiment, six to eight concentrations of a given UTP analog were tested in five replicates. Therefore, each dose-dependent inhibition binding curve or dose-dependent enzyme activation curve was drawn from data obtained with 30-40 ampullas microdissected from five to seven frogs.

For each analog tested, the 95% confidence interval variation range of pK_B value for binding and of pK_a value for agonist-induced phosphoinositidase C activation or pK_i value for antagonist-induced inhibition of UTP-stimulated enzyme activity (pK_B = log K_B, pK_a = log K_a, and pK_i = log K_i in which K_B, K_a, and K_i values are expressed as molar concentrations) were calculated by computerized analysis of the corresponding dose-dependent inhibition binding curves or dose-response curves (GraphPad Prism Software; sigmoidal dose-response fit with variable slope). It has been assumed at the 95% probability level that two analogs bind to ampullas, stimulate phosphoinositidase C, or inhibit UTP-induced enzyme activation with differing apparent affinities only when the lower limit of the pK_B, pK_a, and/or pK_i variation range of the best nucleotide is higher than the upper limit of the pK_B, pK_a, and/or pK_i variation range of the worst analog.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

[³H]UTP Binding Studies

Ampullas isolated from frog semicircular canals bound specifically [³H]UTP. Thus, with the use of 50 nM radioligand, total, nonspecific, and specific binding capacities were, respectively, 4.01 ± 0.25, 0.67 ± 0.07, and 3.34 ± 0.26 fmol [³H]UTP bound/µg protein (means ± SE of mean data from 33 independent experiments).

Results depicted in Fig. 1 summarize the main characteristics of UTP binding to frog ampullas measured under equilibrium conditions. Specific binding of [³H]UTP was competitively inhibited by the corresponding unlabeled nucleotide through a concentration range greater than two orders of magnitude. Unlabeled UTP interacted with the population of labeled binding sites with the following kinetic parameters (mean data from 2 experiments): apparent dissociation constant K_B = 0.8 µM and Hill coefficient nH = 0.7. Fitting binding data in Scatchard coordinates generated a curvilinear plot that, after analysis, yields two straight lines, suggesting the presence of two classes of receptors: a minor class of high-affinity binding sites [maximal binding capacity (R_T1) = 45 fmol UTP bound/µg protein; dissociation constant of the first class of binding sites (K_D1) = 0.4 µM] and a major class of moderate-affinity binding sites [maximal binding capacity (R_T2) = 365 fmol UTP bound/µg protein; dissociation constant of the second class of binding sites (K_D2) = 10 µM].

View larger version (15K): [in this window] [in a new window]   Fig. 1.   Dose-dependent inhibition of [3H]UTP binding to frog ampullas by unlabeled UTP. Results were obtained in 1 experiment, and mean data from 2 independent experiments are summarized in Table 1. Ampullas were incubated for 4 h at 4°C with 50 nM [3H]UTP and increasing amounts of unlabeled UTP. A: binding capacities (means ± SE of 5 replicates performed in the same experiment) were corrected for nonspecific binding measured in presence of 0.1 mM unlabeled UTP; arrow, UTP concentration leading to half-maximal occupancy of binding sites (KB). B: fitting data as log[(RL0/RL)  1] vs. log[UTP] in which RL0 and RL were binding capacities observed in absence or presence of UTP (Eq. 3) generated a linear plot (y = 0.64x + 3.84, r = 0.99); apparent dissociation constant KB = 0.78 µM and Hill coefficient nH = 0.64 for UTP binding were computed from Eqs. 3 and 5. C: Scatchard plot of binding data. Binding capacities were corrected for dilution of specific radioactivity, and ratios [RL]/[L](1/µg protein) between concentrations of bound ligand [RL] (fmol UTP bound · µg protein1 · µl1) and free nucleotide [L] (fmol UTP/µl) were plotted as a function of [RL]. Equations of linear regression lines: y = 2.2 × 103 × + 0.119, r = 0.99 (solid line) and y = 8.2 × 105 × + 0.037, r = 0.99 (dotted line). For each class of binding sites, the dissociation constant for UTP binding (KD) was calculated from slope of the corresponding plot and the maximal binding capacity (RT) from y-intercept [(RT1/KD1) + (RT2/KD2)] of the solid line and x-intercept (RT1 + RT2) of the dotted line. Computations give KD1 = 0.45 µM and RT1 = 39 fmol UTP bound/µg protein and KD2 = 12.2 µM and RT2 = 411 fmol UTP bound/µg protein.

The stereospecificity of labeled binding sites for recognition of a series of unlabeled structural UTP analogs and antagonists was assessed on the basis of competition experiments similar to those illustrated in Fig. 2, and all results are summarized in Table 1. Data demonstrate that most of the drugs tested inhibited [³H]UTP binding to the same extent as unlabeled UTP did, indicating that they interacted with the entire population of [³H]UTP-labeled binding sites but with differing potencies as regards 1) their apparent dissociation constants and 2) their Hill coefficients (UMP, UDP, ADP, UTP, Bz-ATP, AMP, ,-Me-ATP, uridine, DIDS, reactive blue 2, and suramin bound with negative cooperativity phenomena, whereas dTTP, ATPS, 2-Me-ATP, Ap₄A, and PPADS interacted with labeled binding sites according to Michaelian kinetics).

View larger version (20K): [in this window] [in a new window]   Fig. 2.   Dose-dependent inhibitions of [3H]UTP binding to frog ampullas by unlabeled structural UTP analogs and antagonists. For each analog, results were obtained in a separate experiment and mean data from 2 independent experiments are summarized in Table 1. Ampullas were incubated for 4 h at 4°C with 50 nM [3H]UTP and increasing amounts of either unlabeled pyrimidine nucleotides [UMP; UDP; CTP; uridine; A], unlabeled purine nucleotides [ADP; ATP; 2-methylthio-ATP (2-MeS-ATP); B], or unlabeled antagonists [pyridoxal-phosphate-6-azophenyl-2', 4'-disulfonic acid (PPADS); suramin; C]. Binding capacities (means ± SE of 5 replicates performed in the same experiment) were corrected for nonspecific binding determined in presence of 0.1 mM unlabeled UTP and expressed as % of control values measured in absence of inhibitors [, 3.56 ± 0.39 fmol [3H]UTP bound/µg protein (means ± SE of mean data from 9 experiments)]. Fitting data as log[(RL0/RL)  1] vs. log[I] in which RL0 and RL were binding capacities observed in absence or presence of inhibitors I (Eq. 3) generated linear plots: UMP, y = 0.46x + 3.39, r = 98; UDP, y = 0.65x + 4.19, r = 0.99; CTP, y = 0.89x + 4.24, r = 0.99; uridine, y = 0.62x + 2.05, r = 0.99; ADP, y = 0.61x + 3.68, r = 0.99; ATP, y = 0.91x + 4.36, r = 0.99; 2-MeS-ATP, y = 1.19x + 4.95, r = 0.98; PPADS, y = 0.89x + 3.60, r = 0.98; suramin, y = 0.48x + 1.46, r = 0.99. Apparent dissociation constants and Hill coefficients for binding of unlabeled nucleotides and antagonists were computed from Eqs. 3 and 4.

Phosphoinositidase C Studies

View larger version (19K): [in this window] [in a new window]   Fig. 3.   Dose-dependent UTP-induced changes in labeled inositol-containing cellular pools in ampullas from frog semicircular canals. Results were obtained in 1 experiment, and mean data from 5 independent experiments are summarized in Table 2. Myo-[3H]inositol-labeled ampullas were incubated for 30 min at 26°C with the indicated amounts of UTP before measurements of radioactivities incorporated in cellular free inositol, total () phosphoinositides, and total inositol phosphates (InsPs) and glycerophosphoinositol (GroPIns). Data (means ± SE of 5 replicates performed in the same experiment) were expressed as % of total radioactivity found in all inositol-containing cellular pools. Arrow, UTP concentration leading to half-maximal activation of InsPs production. Apparent activation constant and Hill coefficient for UTP-induced enzyme activation were computed from x-intercept and slope of Hill plot of UTP-stimulated InsPs reduction curve (y = 0.52x + 2.64, r = 0.99).

Results summarized in Table 2 illustrate the sensitivity of phosphoinositidase C to structural UTP analogs. It is worth noting that the nucleotides tested were able to stimulate phosphoinositidase C with differing potencies as regards 1) their apparent activation constants [K_a values for UDP and UTP are far lower than those of UMP, the selective P2Y_Ap4A agonist Ap₄A (14), and the potent marker of P2X₇ receptors Bz-ATP], 2) their Hill coefficients (UTP and Ap₄A stimulated enzyme activity with negative cooperativity phenomena, UDP, and Bz-ATP according to Michaelian kinetics, and UMP with slight positive cooperativity phenomena), and 3) their intrinsic activities or magnitude of the maximal responses observed (UMP and Bz-ATP exhibited partial agonistic potencies, whereas uridine was devoid of activity).

Data from a competition experiment between UMP and UTP [increases in [³H]InsP_s production () above basal generation, 1.08 ± 0.22% of total radioactivity found in all inositol-containing cellular pools; means ± SE of 7 replicates] show clearly that 1) the enzyme activation elicited by a liminal concentration of UMP (50 µM UMP, 0.27 ± 0.20) did not impair the response to a submaximal amount of UTP ( 5 µM UTP, 1.48 ± 0.78 vs. 50 µM UMP + 5 µM UTP, 1.52 ± 0.81; not significant, Student's t-test), 2) the increases in InsP_s production due to submaximal concentrations of UMP (1 mM UMP, 1.40 ± 0.23) and UTP were fully additive [the response observed in the presence of UMP and UTP did not differ from the sum of both activations (experimental 1 mM UMP + 5 µM UTP, 2.46 ± 0.1 vs. computed , 2.88 ± 0.81; not significant, Student's t-test)], and 3) the maximal enzyme activations induced by saturating amounts of UMP and UTP (10 mM UMP, 2.71 ± 0.26; 0.1 mM UTP, 3.47 ± 0.57) were not additive. The increase in InsP_s production measured in presence of UMP and UTP was significantly lower than the sum of both activations (experimental 10 mM UMP + 0.1 mM UTP, 2.84 ± 0.51 vs. computed , 6.18 ± 0.63; P < 0.01, Student's t-test). These observations indicate that UMP and UTP stimulate the same pool of phosphoinositidase C with differing affinities.

As expected, chemicals revealing antagonistic potencies in other systems (13, 32) failed to increase basal InsP_s production: the latter was not modified by 0.2 mM reactive blue 2, 0.5 mM DIDS, or 5 mM PPADS and was decreased by ~50% in presence of 10 mM suramin (results not shown). The four antagonists tested inhibited UTP-induced phosphoinositidase C activation according to pure competitive inhibition kinetics (Fig. 4 and Table 2). They increased the UTP concentration leading to half-maximal enzyme activation, did not reduce the maximal enzyme activation, and diminished in a dose-dependent and saturable fashion the enzyme response to 5 µM UTP either with negative cooperativity phenomena (DIDS and PPADS), according to Michaelian kinetics (reactive blue 2), or with slight positive cooperativity phenomena (suramin). In addition, reactive blue 2, DIDS, and PPADS enhanced the Hill coefficient for UTP-stimulated InsP_s production; ni values (see Eq. 7) were, respectively, 0.8, 1.2, and 1.3 for UTP-induced enzyme activations measured in presence of 0.2 mM reactive blue 2, 0.5 mM DIDS, and 5 mM PPADS, whereas ni was equal to 0.6 in presence of 10 mM suramin.

View larger version (15K): [in this window] [in a new window]   Fig. 4.   Inhibition induced by DIDS of UTP-stimulated phosphoinositidase C activity in ampullas from frog semicircular canals. Results illustrated in A and B were obtained in separate experiments, and mean data from 2 sets of independent experiments are summarized in Table 2. Myo-[3H]inositol-labeled ampullas were incubated for 30 min at 26°C with either the indicated amounts of UTP in absence () or presence of 0.5 mM DIDS (; A) or 5 µM UTP and increasing concentrations of DIDS (B). [3H]InsPs productions (means ± SE of 5 replicates performed in the same experiment) were calculated as % of total radioactivity found in all inositol-containing cellular pools, corrected for basal activities, and further expressed as % of activation induced by a saturating amount of UTP (UTPmax) measured in the same experiment. Arrows, UTP concentrations leading to half-maximal activations of InsPs production (A) and DIDS concentration eliciting half-inhibition of 5 µM UTP-induced enzyme activation (B). Equations of linear regression lines of Hill plots of dose-response curves A: (UTP  DIDS), y = 0.61x + 3.52, r = 0.99, and (UTP + 0.5 mM DIDS), y = 1.22x + 6.09, r = 0.95. Fitting data in B (DIDS + 5 µM UTP) as log[(V0/V)  1] vs. log[DIDS] in which V0 and V were enzyme activations observed in absence or presence of antagonist (equation 8) generated a linear plot (y = 0.66x + 2.94, r = 0.98). Apparent inhibition constant and Hill coefficient for DIDS-induced inhibition of UTP-stimulated enzyme activity were computed from Eqs. 7-9.

Finally, data from control experiments show that the pretreatment of UDP with hexokinase did not modify kinetic parameters for UDP-induced phosphoinositidase C stimulation. Apparent activation constants and Hill coefficients were K_a = 0.61 and 0.56 µM, and nH = 0.90 and 1.01 for control and hexokinase-treated samples, respectively. Indeed, fitting basal, submaximal, and maximal UDP-stimulated enzyme activities of hexokinase-treated samples against the corresponding activities of control samples generated a linear plot (y = 1.08x + 0.17, r = 0.97) whose slope and y-intercept do not differ, respectively, from unity and zero (P < 0.05; Student's t-test). In addition, the presence in the incubate of a UTP-regenerating system did not impair basal, 5 µM UTP-, and 0.1 mM UTP-stimulated phosphoinositidase C activities. The corresponding [³H]InsP_s productions (% of total radioactivity found in all inositol-containing cellular pools; means ± SE of 5 replicates) were 1.8 ± 0.2 and 2.0 ± 0.1 (basal), 7.9 ± 0.9 and 8.3 ± 1.2 (5 µM UTP), and 9.8 ± 1.0 and 10.9 ± 0.7 (0.1 mM UTP) for samples assayed in absence or presence of the UTP-regenerating system, respectively.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

The experiments described above provide evidence for the presence of specific binding sites for [³H]UTP in ampulla from R. ridibunda semicircular canal and confirm the presence of a UTP-sensitive phosphoinositidase C system in this structure (6).

Despite the minute amounts of tissue used, the validity of the binding microassay employed is supported by the following observations: 1) [³H]UTP binding was saturable with incubation time, 2) [³H]UTP binding was competitively inhibited by unlabeled UTP and a series of structural nucleotide analogs, and 3) unrelated nucleotide chemicals acting as antagonists in other systems (13, 32) reduced UTP binding. On the one hand, it should be stressed, that the maximal UTP binding capacity of frog ampulla was similar to those reported for binding of adenosine 5'-O-(2-thiodiphosphate) (ADPS) to the same structure (7) and to cultured bovine aortic endothelial cells (31) but was about two orders of magnitude higher than the density of ADPS receptors in avian erythrocyte membranes (10) and of UTP receptors in rat lung membranes (17). On the other hand, it is worth noting that UTP bound to frog ampullas with an apparent dissociation constant close to that of ADPS (7) but higher than the K_B values for UTP binding to rat lung membranes (17) and for ADPS binding to avian erythrocyte membranes and to cultured bovine aortic endothelial cells (10, 31). Thus the discrepancies found between amphibia and higher vertebrates might reflect some zoological stereospecificities in nucleotide sensitivity of target cells, as observed earlier for neurohypophysial hormone binding to renal receptors from various amphibian and mammalian species (1, 5, 16).

It seems unlikely that the differences observed between K_B values for nucleotide bindings and between K_a values for phosphoinositidase C stimulations resulted from ecto-ATPase and ecto-nucleotidase activities (11, 12, 29) because binding assays were performed under experimental conditions known to reduce enzymatic degradation of nucleotides, i.e., at 4°C and in absence of divalent cations (32, 33), and the presence in the phosphoinositidase C assay medium of a triphosphonucleotide-regenerating system did not modify enzyme responses to UTP and ATP (6). Moreover, the presence of contaminants in nucleotide preparations could hamper determinations of kinetic parameters for nucleotide binding and/or enzyme activation (17, 18). However, the pretreatment of UDP with hexokinase did not impair K_a and nH values for UDP-induced phosphoinositidase C stimulation. These results suggest that the differences found between pK_B values for UDP and UTP binding and pK_a values for UDP- and UTP-induced phosphoinositidase C activations could not result from contaminations of UDP by UTP.

Data of binding and phosphoinositidase C studies suggest strongly that the population of ampullary [³H]UTP-labeled binding sites is heterogeneous because Scatchard plot of UTP binding may be analyzed as a combination of a minor class of high-affinity binding sites and a major class of moderate-affinity binding sites and no correlation occurs between the pK_a (or pK_i) values of analogs for phosphoinositidase C activation [reported in this study and earlier (6)] and their corresponding pK_B values for binding (y = 0.33x + 2.21, r = 0.37; not significant). Among all chemicals tested, only the agonists UDP, ATPS, ATP, 2-MeS-ATP, and Ap₄A and the antagonists DIDS, PPADS, and reactive blue 2 bound to ampullas, stimulated InsP_s production, and/or inhibited UTP-induced phosphoinositidase C activation with similar apparent affinities. Moreover, results of competition enzyme experiments showed that enzyme responses to submaximal amounts of UMP and UTP were additive, whereas those to saturating concentrations of both nucleotides were not. This latter observation demonstrates that UMP and UTP stimulate the same pool of phosphoinositidase C with differing apparent affinities and suggests that the high-affinity UMP receptors do not promote enzyme activation. Thus, taken together, results of binding and enzyme studies indicate that some [³H]UTP-labeled binding sites exhibit functional properties differing from those of the biological nucleotide receptors triggering phosphoinositidase C activation.

The [³H]UTP-labeled binding sites uncoupled to phosphoinositidase C might correspond to those revealing high affinity for UTP for the following evidence: 1) UTP interacted with the minor class of labeled binding sites with a K_D value 10 times lower than its K_a value for activation of InsP_s production, 2) the weak P2(Y) agonist UMP (26) bound to ampullas with the highest apparent affinity and negative cooperativity phenomena, whereas it stimulated phosphoinositidase C with a low affinity and slight positive cooperativity phenomena, 3) dTTP and CTP interacted with apparent affinities far higher than their corresponding affinities for enzyme stimulation (6), 4) uridine bound with a low affinity but did not increase InsP_s production, and 5) the antagonist suramin bound with a K_B value 10 times lower than its K_i value for inhibition of UTP-induced phosphoinositidase C activation. So, this population of high-affinity UTP binding sites might represent either 1) nonspecific nucleotide binding proteins unrelated to P2 receptors, 2) P2(Y)-like receptors promoting their biochemical effects through phosphoinositidase C-independent pathways as shown for P2(Y) receptor-mediated phospholipase D stimulation in Madin-Darby canine kidney-D1 cells (3), 3) a desensitized form of P2(Y)-like receptors revealing high affinities for pyrimidine nucleotides as observed for acetylcholine receptors from Torpedo marmorata (15), or 4) enzymes implied in the degradation of extracellular nucleotides, but differing from the ecto-nucleotidases found in the guinea pig cochlear perilymphatic compartment that hydrolyze substrates with Michaelis constant values in the millimolar range (29). These UTP binding sites might be clearance nucleotide receptors, as reported for atrial natriuretic peptide (ANP) binding to rat kidney clearance ANP receptors defined as opposed to biological ANP receptors triggering guanylate cyclase stimulation (21). But these hypotheses call for additional investigations.

On these grounds, it may be postulated that the major class of moderate-affinity UTP binding sites correspond to the physiological nucleotide P2(Y)-like receptors involved in phosphoinositidase C activation processes. Finally, it should be stressed that reactive blue 2 diminished UTP-induced phosphoinositidase C activation according to competitive inhibition kinetics, whereas it acted as a pure noncompetitive inhibitor for decrease of ATP-stimulated enzyme activity (6). This observation provides further evidence for the coexpression in frog ampulla of both purine P2Y₁-like and pyrimidine P2(Y)-like receptors linked to phosphoinositidase C activation (6, 7).

Perspectives

The expression of P2(Y)-like receptors in frog ampulla raises obvious questions about their cellular localization(s) and biological function(s). Physiological experiments performed in nonsensory tissues of the gerbil inner ear suggest that pyrimidine nucleotide P2(Y) receptors are involved in the regulation of K⁺ secretion (20). So, the biological significance of the P2(Y)-like receptors found in amphibian semicircular canal remains to be defined in further electrophysiological experiments.