Presynaptic dopamine receptors: Insensitivity to kainic acid and the development of supersensitivity following chronic haloperidol

Summary The effects of kainic acid lesions and chronic haloperidol treatment on rat striatal dopaminergic presynaptic receptors were studied. Following the -butyrolactone-induced inhibition of dopaminergic impulse flow, and after dopa decarboxylase inhibition, dopa accumulation and its reversal by dopamine agonists was measured in vivo.³H-apomorphine (a dopamine receptor ligand with purported presynaptic specificity) was used for in vitro binding experiments. Presynaptic dopamine receptors, as assessed by both methods, were unaffected by intrastriatal kainic acid injection 5–6 days before sacrifice. Seven days after termination of chronic haloperidol treatment (28 days, 0.5 mg/kg/day s.c.) both an increased apomorphine response using the dopa accumulation method and an increase in³H-apomorphine binding were observed, indicating the development of presynaptic dopamine receptor supersensitivity.     

Selective labeling of apomorphine receptors by 3H-LSD.

There are at least two types of dopamine receptors: the 3H-dopamine or 3H-apomorphine receptor (with high or nM affinity for dopamine), and the 3H-neuroleptic receptor (with low or microM affinity for dopamine). While 3H-LSD can label the 3H-neuroleptic receptor, this study was done in order to label the 3H-apomorphine/dopamine receptor site. In the presence of excess phentolamine, serotonin and spiperone (to preclude binding to alpha-adrenergic, serotonergic and neuroleptic receptors, respectively) similar concentrations of dopaminergic drugs inhibited the binding (to calf caudate) of 3H-LSD and 3H-apomorphine. This is compatible with the concept that the 3H-apomorphine/dopamine receptor and the 3H-neuroleptic/dopamine receptor are separate.     

Dopamine receptor mediated inhibition by pergolide of electrically-evoked 3H-dopamine release from striatal slices of cat and rat: Slight effect of ascorbate

Summary The dopamine receptor agonist pergolide inhibited the calcium-dependent, electrically evoked overflow of tritium from slices of the striatum of cat or rat prelabelled with ³H-dopamine. This inhibition of tritium overflow by nanomolar concentrations of pergolide was antagonized by the benzamide neuroleptic S-sulpiride (0.1 M). In millimolar concentrations, l- ascorbate had slight or no effects on this dopamine receptor mediated inhibition, in striatal slices of either the cat or the rat. Since these same concentrations of ascorbate have been reported to completely block the specific binding of ³H-2-amino-6,7-dihydroxy-1,2,3,4-tetrahydronaphthalene (ADTN) and of ³H-apomorphine to presumed dopamine receptors, the present results suggest a dissociation between the characteristics of ³H-ADTN and ³H-apomorphine binding and the dopamine autoreceptor. Previous contradictory results concerning the existence of inhibitory dopamine receptors which modulate depolarization-evoked overflow of dopamine from the striatum of the rat are thus apparently not due to a species difference nor to the use of ascorbate, but rather to differences in experimental conditons.     

Presynaptic subsensitivity as a possible basis for sensitization by long-term dopamine mimetics

A possible cellular basis for dopaminergic sensitization by long-term dopamine mimetics was examined in rat brain striatum. Long-term apomorphine or amphetamine administration (10 mg/kg/day for 14 days) resulted in a decrease in the specific binding of 3H-apomorphine, but no change in 3H-haloperidol binding. Long-term apomorphine treatment also enhanced the cataleptogenic action of haloperidol, with many rats being spontaneously cataleptic after apomorphine withdrawal. It is suggested that the reduced 3H-apomorphine binding signifies less presynaptic receptors. This permits less autoregulation and enhanced dopamine agonist action, possibly accounting for the dopaminergic sensitization by long-term agonists.     

Species variation in dopamine receptor binding

Binding of 3H-spiroperidol, 3H-apomorphine and 3H-ADTN (2-amino-6,7-dihydroxytetrahydronaphthalene) associated with dopamine receptors has been evaluated in corpus striatal membranes of calf, rat and human brains. Substantial species differences are apparent for numberous agonists and antagonists in competing for receptor binding. In general, dopamine receptor antagonists are more potent in rat and agonists more potent in calf. In competing for 3H-spiroperidol binding sulpiride, molindone and metaclopramide show the most pronounced species differences, being 3--10 times more potent in rat and human than in calf. In all three species agonists compete for 3H-spiroperidol binding with Hill coefficients less than one while antagonists inhibit 3H-spiroperidol binding with Hill coefficients of about 1.0. Conversely, 3H-apomorphine and 3H-ADTN binding in all three species is inhibited by antagonists with Hill coefficients less than 1.0 while agonists display Hill coefficients of about 1.0. In general agonists are more potent in competing for binding of 3H-apomorphine and 3H-ADTN than 3H-spiroperidol. However, a small component of dopamine, apomorphine and ADTN inhibition of 3H-spiroperidol binding displays very high affinity (IC50 about 1 nM). In human amygdala 3H-spiroperidol appears to label serotonin receptors predominantly.     

Neuroleptic receptors: stereoselectivity for neuroleptic enantiomers.

In order to identify a pair of neuroleptic enantiomers with the highest stereoselective interaction with neuroleptic/dopamine receptors, the effects of eight pairs of neuroleptic enantiomers were tested on the specific binding of 3H-spiperone to crude homogenates of calf caudate nucleus. The ratios of the Ki values were: (+)-butaclamol/(-)-butaclamol = 3000; dexclamol/(-)-analogue = 151; (+)-isobutaclamol/(-)-isobutaclamol = 146; (-)-CTC/(+)-CTC= 109; (-)-centbutindole/(+)-centbutindole = 20; S(+)-octoclothepin/R(-)-octoclothepin = 11. Thus, the neuroleptic receptor is highly stereoselective for the rigid butaclamol derivatives, but much less so for the flexible neuroleptics. The 3H-apomorphine binding site, however, had a stereoselectivity ratio of only 7 for isobutaclamol, further suggesting that the high affinity sites (i.e. nM) for 3H-neuroleptic binding and for 3H-apomorphine binding are different.     

Cardiac glycosides,calcium and the release of neurotransmitter from peripheral noradrenergic nerves

Summary The best experimental conditions for a selective binding of³H-apomorphine to dopamine receptors on cryostat sections were first selected by liquid scintillation quantification of the bound radioactivity. In the corpus striatum, a specific binding occurred with a half-maximal saturation concentration of about 1 nM and a maximal capacity of 180 fmol/mg of slice protein, both values in agreement with previous binding data on either membranes or slices incubated in a physiological medium. Inhibition with domperidone was clearly biphasic, indicating two classes of sites corresponding to the D-2 and D-3 sites as previously defined on membranes.When³H-apomorphine was used at low concentrations (0.8–1.5 nM), a condition ensuring a preferential labelling of D-2 sites, rather well contrasted autoradiographic pictures were generated. The major dopaminergic projection fields in telencephalon (caudate-putamen, nucleus accumbens, olfactory tubercles) were visualised as well as other catecholaminergic regions such as the superficial gray layer of superior colliculi. Within the striatum, differences in density of these sites were observed in three perpendicular planes and confirmed by a computer densitometric image analysis. Labelling of areas of origin of the cerebral dopaminergic neurons in substantia nigra or ventral tegmental area were also observed.When a higher concentration of³H-apomorphine (3.5 nM) was used in the presence of domperidone, another, but autoradiographically less distinct subclass of sites (D-3 sites) was demonstrated.     

Interaction of novel hybrid compounds with the D3 dopamine receptor: Site-directed mutagenesis and homology modeling studies

The dopamine D3 receptor has been implicated as a potential target for drug development in various complex psychiatric disorders including psychosis, drug dependence, and Parkinson's disease. In our overall goal to develop molecules with preferential affinity at D3 receptors, we undertook a hybrid drug development approach by combining a known dopamine agonist moiety with a substituted piperazine fragment. In the present study, three compounds produced this way with preferential D3 agonist activity, were tested at D3 receptors with mutations in the agonist binding pocket of three residues known to be important for agonist binding activity. At S192A and T369V, the hybrid agonist compounds produced an interaction profile in [³H]spiperone binding assays similar to that of the parent 5-OH-DPAT and 7-OH-DPAT molecules. The loss of affinity at the S192A mutant was most prominent for 5-OH-DPAT and its corresponding hybrid compound D237. D110N did not show any radioligand binding. Homology modeling indicated that 7-OH-DPAT-derived D315 uniquely shares H-bonding with Tyr365 which produced favorable interaction and no loss of H-bonding in the S192A mutant, suggesting that agonist activity may not be solely controlled by residues in the binding pocket.     

Allosteric nature of P2X receptor activation probed by photoaffinity labelling

BACKGROUND AND PURPOSE

In P2X receptors, agonist binding at the interface between neighbouring subunits is efficiently transduced to ion channel gating. However, the relationship between binding and gating is difficult to study because agonists continuously bind and unbind. Here, we covalently incorporated agonists in the binding pocket of P2X receptors and examined how binding site occupancy affects the ability of the channel to gate.

EXPERIMENTAL APPROACH

We used a strategy for tethering agonists to their ATP-binding pocket, while simultaneously probing ion channel gating using electrophysiology. The agonist 2′,3′-O-(4-benzoylbenzoyl)-ATP (BzATP), a photoaffinity analogue of ATP, enabled us to trap rat homomeric P2X2 receptor and a P2X2/1 receptor chimera in different agonist-bound states. UV light was used to control the degree of covalent occupancy of the receptors.

KEY RESULTS

Irradiation of the P2X2/1 receptor chimera – BzATP complex resulted in a persistent current that lasted even after extensive washout, consistent with photochemical tethering of the agonist BzATP and trapping of the receptors in an open state. Partial labelling with BzATP primed subsequent agonist binding and modulated gating efficiency for both full and partial agonists.

CONCLUSIONS AND IMPLICATIONS

Our photolabelling strategy provides new molecular insights into the activation mechanism of the P2X receptor. We show here that priming with full agonist molecules leads to an increase in gating efficiency after subsequent agonist binding.     

Comparison of two putatively selective radioligands for labeling central nervous system beta-adrenergic receptors: inadequacy of [3H]dihydroalprenolol

[3H]Dihydroalprenolol ([3H]DHA) has been used extensively in receptor binding studies to measure beta-adrenergic receptors in the central nervous system. Usually, nonspecific binding has been defined by high concentrations of the beta-adrenergic receptor agonist isoproterenol or antagonists such as alprenolol or propranolol. Scatchard plots of such "specific" [3H]DHA saturation data in rat cerebral cortex membranes are linear. However, computer analysis demonstrated that the competition curves of these drugs for 2.0 nM [3H]DHA binding are biphasic, with a continuous inhibition of [3H]DHA binding in the concentration range usually used to determine nonspecific binding. These data indicate that another saturable high affinity site was being labeled by the radioligand and that the definition of nonspecific binding with any of these unlabeled drugs is not satisfactory. We used the nonlinear, least squares, curve-fitting program LIGAND to analyze total [3H]DHA binding, allowing the program to mathematically define nonspecific binding as a function of 3H-ligand concentration. Significantly lower Bmax (-44%) and Kd (-58%) values for beta-adrenergic receptors were found, indicating that under normal experimental procedures (defining [3H]DHA non-specific binding with these nonradioactive drugs) a second binding site was being labeled. We found that [3H]DHA binding to this site could be inhibited by drugs such as RU24969, a 5-hydroxytryptamine1A (5HT1A) and 5HT1B receptor subtype-selective agonist, and CGS12066B, a 5HT1B receptor subtype-selective agonist, which were able to compete for 15-20% of [3H]DHA binding in the nanomolar concentration range, whereas drugs that are selective for other serotonin receptor subtypes inhibited [3H]DHA binding only at much higher concentrations. Another beta-adrenergic receptor antagonist radioligand, [3H]CGP-12177, was found to be more selective for beta-adrenergic receptors. Alprenolol competition curves for [3H]CGP-12177 binding were monophasic and saturation curves, with nonspecific binding defined either by 10 microM alprenolol or by LIGAND, yielded Bmax values close to those obtained with [3H]DHA when its nonspecific binding was defined by LIGAND. [3H]DHA cannot be considered a suitable radioligand to quantify central nervous system beta-adrenergic receptors in the manner in which it has been typically used.     

Effects of bunitrolol on adrenergic and serotonergic receptors

To assess the importance of anti-adrenergic and anti-serotonergic activities of bunitrolol for its efficacy as an antihypertensive and antianginal agent, effects of this substance on the binding of adrenergic and serotonergic agents to the respective receptors of the rat brain, rat heart, dog brain, and/or dog aorta were examined using the radioligand binding assay methods. In addition, the pA2 values of bunitrolol as an antagonist against the positive chronotropic and inotropic actions (beta 1-adrenoceptor) of isoproterenol were also determined by pharmacological methods using the isolated guinea pig atria. To assess the specificity, pA2 values were also obtained in the isolated trachea (beta 2-adrenoceptor) using isoproterenol as an agonist and in the isolated aorta from the guinea pig and the rat using phenylephrine as an agonist (alpha 1-adrenoceptor). A strong inhibition by bunitrolol of 3H-dihydroalprenolol (3H-DHA) binding to beta-adrenoceptors was observed, while the inhibition of 3H-prazosin binding to alpha 1-adrenoceptors, 3H-serotonin binding to 5HT1-receptors. 3H-p-aminoclonidine binding to alpha 2-adrenoceptors, and 3H-ketanserin binding to 5HT2-receptors were found to be very weak. The rank order of antagonistic potencies of bunitrolol against the adrenergic receptors as assessed with pA2 values were beta 1 greater than beta 2 much greater than alpha 1. From these two different types of experiments, it is clear that the antihypertensive and antianginal effects of bunitrolol are mainly due to its beta-blocking actions, with the alpha 1-blocking action of this drug playing a minor role.     

Nigrostriatal lesions enhance striatal 3H-apomorphine and 3H-spiroperidol binding.

Following unilateral 6-hydroxydopamine nigrostriatal lesions in rats, the binding of both 3H-apomorphine and 3H-spiroperidol in the striatum is increased. In rats with incomplete lesions or at early time points after lesion, binding is not significantly different from control levels.     

Dopamine receptor of the porcine anterior pituitary gland. Solubilization and characterization

In the anterior pituitary gland, dopamine controls the release of prolactin from the mammotrophs. The dopamine receptors in the porcine gland have been shown to exist in two different affinity states of equal proportion, one bearing high affinity for agonists and labeled by 3H-agonist-ligands and the other displaying low affinity for agonists. Both forms of the receptor can be labeled by 3H-antagonist-ligands. Dopamine receptors from porcine anterior pituitary membranes can be solubilized with retention of their ability to interact with specific dopaminergic ligands. Treatment of membrane preparations with 1% digitonin resulted in the solubilization of 20-25% of the specific binding sites labeled by [3H] spiroperidol with a specific activity of about 100 fmoles/mg. The receptor was a glycoprotein as assessed by the interaction of these binding sites with agarose-immobilized lectin. [3H]Spiroperidol binding in solubilized preparations was saturable, of high affinity (KD = 570 pM), and to a single class of stereoselective binding sites. Agonist competition for [3H]spiroperidol binding indicated that, whereas the solubilized receptor retained its dopaminergic specificity, the high-affinity interactions of the receptor with agonists present in membranes and sensitive to guanine nucleotides were lost in solubilized preparations. Thus, the KD values calculated from the agonist competition curves for [3H]spiroperidol corresponded to the agonist affinities for the low-affinity state of the receptor documented in membranes. However, high-affinity agonist binding and its sensitivity to guanine nucleotides were preserved when the membrane-bound receptor was prelabeled with the agonist [3H]N-n-propylnorapomorphine prior to solubilization. These results suggest that a component that confers agonist high-affinity binding and guanine nucleotide responsiveness to the receptor is lost during solubilization unless a stable complex is formed with the agonist prelabeled receptor prior to solubilization.     

Transducer abstraction: a novel approach to the detection of partial agonist efficacy in radioligand binding studies

The properties of the ternary complex model (TCM) of drug action at G protein-coupled receptors (GPCRs) were examined, using theoretical computer simulations, with regard to the predicted effects of the presence of a fixed concentration of one agonist on the competition binding profile of another. Subsequently, the binding properties of the full muscarinic acetylcholine receptor (mAChR) agonists acetylcholine (ACh) and carbachol (CCh), and the partial agonists pilocarpine and McN-A-343, were investigated in competition experiments against [(3)H]N-methylscopolamine using homogenate preparations from Chinese hamster ovary cells, stably expressing the human M(1) or M(2) mAChR. At the M(2) mAChR, all agonists displayed biphasic binding curves and were readily modulated by the non-hydrolyzable GTP analogue, Gpp(NH)p, in accordance with previously established experimental observations. In contrast, agonist binding at the M(1) mAChR showed no significant change in the presence of Gpp(NH)p, even in the case of a full agonist. This phenomenon precludes using the "GTP-shift" to assess agonist efficacy at the M(1) mAChR. When the ACh competition curves were reconstructed in the presence of graded concentrations of either a full or a partial agonist, a significant redistribution of the fraction of the high-affinity state recognized by ACh was observed. However, when the procedure was repeated using the antagonist, atropine, no significant effect on the fraction of either the high or low affinity ACh binding components at the mAChR was observed. Taken together, these results indicate that changes in the profile of full agonist binding isotherms, when constructed in the presence of a partial agonist, may be more sensitive indicators of partial agonist efficacy than regular assays that directly measure partial agonist binding.     

Muscarinic cholinergic receptors in the embryonic chick heart: interaction of agonist, receptor, and guanine nucleotides studied by an improved assay for direct binding of the muscarinic agonist [3H]cismethyldioxolane

Muscarinic agonist binding has been studied by the indirect method of competition between binding of agonist and 3H-antagonist. Studies of 3H-agonist binding have either been complicated by high levels of nonspecific binding or have been carried out at low concentrations of agonist, which bind only to high affinity sites. A new assay for binding of the muscarinic agonist, [3H]cismethyldioxolane, allows measurement of binding at concentrations up to 1000 nM with a high degree of specificity (92% at 18 nM, 50% at 1000 microM). Using this new binding method, it is possible not only to directly establish the validity of those observations made using competition binding studies, but also to obtain new insight into the mechanism of interaction of receptor agonist and guanine nucleotide. Specifically, data have been presented which demonstrate: 1) that binding of agonist to the high and low affinity forms of the muscarinic receptor involves two independent parallel reactions, and, in the absence of guanine nucleotides, agonist binding alone does not mediate the interconversion of the receptor from one affinity state to another, thus suggesting that high affinity receptors are present in chick heart membranes and do not require agonist binding for their formation; 2) that the interaction of agonist and guanine nucleotide with the receptor-guanine nucleotide-binding protein complex involves formation of an intermediate state in which both agonist and guanine nucleotide are bound to the receptor-guanine nucleotide-binding protein complex; and, finally, 3) that the order of binding of agonist and guanine nucleotide during formation of this intermediate state is random.     

Detection of multiple H3 receptor affinity states utilizing [3H]A-349821, a novel, selective, non-imidazole histamine H3 receptor inverse agonist radioligand

1. A-349821 is a selective histamine H3 receptor antagonist/inverse agonist. Herein, binding of the novel non-imidazole H3 receptor radioligand [3H]A-349821 to membranes expressing native or recombinant H3 receptors from rat or human sources was characterized and compared with the binding of the agonist [3H]N--methylhistamine ([3H]NMH). 2. [3H]A-349821 bound with high affinity and specificity to an apparent single class of saturable sites and recognized human H3 receptors with 10-fold higher affinity compared to rat H3 receptors. [3H]A-349821 detected larger populations of receptors compared to [3H]NMH. 3. Displacement of [3H]A-349821 binding by H3 receptor antagonists/inverse agonists was monophasic, suggesting recognition of a single binding site, while that of H3 receptor agonists was biphasic, suggesting recognition of both high- and low-affinity H3 receptor sites. 4. pKi values of high-affinity binding sites for H3 receptor competitors utilizing [3H]A-349821 were highly correlated with pKi values obtained with [3H]NalphaMH, consistent with labelling of H3 receptors by [3H]A-349821. 5. Unlike assays utilizing [3H]NMH, addition of GDP had no effect on saturation parameters measured with [3H]A-349821, while displacement of [3H]A-349821 binding by the H3 receptor agonist histamine was sensitive to GDP. 6. In conclusion, [3H]A-349821 labels interconvertible high- and low-affinity states of the H3 receptor, and displays improved selectivity over imidazole-containing H3 receptor antagonist radioligands. [3H]A-349821 competition studies showed significant differences in the proportions and potencies of high- and low-affinity sites across species, providing new information about the fundamental pharmacological nature of H3 receptors.     

Agonist and antagonist interactions with D1 dopamine receptors: agonist-induced masking of D1 receptors depends on intrinsic activity.

The effects of agonist and antagonist compounds on the equilibrium binding of the D1 antagonist ligand [3H]SCH 23390 were examined in membranes from the striatum of the rat. The antagonist SK&F 83566 interacted with D1 receptors in the manner of a competitive antagonist, causing a decrease in the affinity of the binding of [3H]SCH 23390, without altering the maximum number of binding sites (Bmax). The interaction of agonist compounds with the D1 receptor appeared to be more complex. The drug SK&F 75670, a weak partial agonist, also acted competitively at D1 sites. However, agonists with moderate (SK&F 38393, CY 208-243) or full (dopamine) intrinsic activity to stimulate adenylate cyclase, interacted with D1 binding sites in a mixed competitive/non-competitive manner, causing both a decrease in ligand affinity and a decrease in Bmax. The benzazepine analogue, which also has full agonist activity, SK&F 82958, only caused a reduction in Bmax. Furthermore, there was a positive relationship between the intrinsic activity of agonists and the magnitude of the reductions in Bmax which they induced. In the presence of the GTP analogue, Gpp(NH)p, CY 208-243 no longer caused an apparent reduction in the number of receptors. These data suggests that the apparent loss of D1 receptors, induced by agonists, may result from an interaction with a guanine-nucleotide sensitive, high affinity agonist binding site and that the degree of interaction with this site depends on the intrinsic D1 activity of the agonist.     

Bupropion: a new antidepressant drug, the mechanism of action of which is not associated with down-regulation of postsynaptic beta-adrenergic, serotonergic (5-HT2), alpha 2-adrenergic, imipramine and dopaminergic receptors in brain

The present experiments were undertaken to determine: (1) whether bupropion had any direct effects on receptors present in rat brain; (2) whether the drug could down-regulate postsynaptic beta-adrenergic, alpha 2-adrenergic, serotonergic, imipramine and dopaminergic receptors after chronic administration, as had been demonstrated for tricyclic antidepressants, monoamine oxidase (MAO) inhibitors, electroconvulsive therapy (ECT) and "atypical" antidepressants. Bupropion was found to be weak or inactive when its affinity for 14 different receptors present in brain was assessed by binding assays. The drug failed to desensitize beta-adrenergic receptors in the cerebral cortex of the rat as determined by [3H]dihydroalprenolol binding, after being administered at 25 mg/kg (i.p.) once a day for 6 weeks, or after being administered by the intraperitoneal route to rats at doses as large as 150 mg/kg per day for 4 days. When administered at doses of 37.5, 75 and 150 mg/kg per day for 21 days, the drug had no effect on beta-adrenergic, alpha 2-adrenergic, imipramine or serotonergic (5-HT2) receptors in the brain of the rat as determined by Scatchard analysis of the binding data. These data show that the antidepressant activity of bupropion is not associated with a down-regulation of receptors in the CNS commonly implicated in the mechanism of action of antidepressant drugs. Bupropion also produced a dose-dependent tendency to decrease the activity of norepinephrine-stimulated adenylate cyclase in slices of cerebral cortex obtained from rats treated chronically with the drug. However, the decrease was highly variable, was most obvious in tissues obtained from rats receiving large, non-pharmacologically relevant doses (150 mg/kg per day) of the drug and was statistically significant at only one of three concentrations of the agonist that produced maximal stimulation of the enzyme.     

Differential allosteric modulation by amiloride analogues of agonist and antagonist binding at A(1) and A(3) adenosine receptors

The diuretic drug amiloride and its analogues were found previously to be allosteric modulators of antagonist binding to A(2A) adenosine receptors. In this study, the possibility of the allosteric modulation by amiloride analogues of antagonist binding at A(1) and A(3) receptors, as well as agonist binding at A(1), A(2A), and A(3) receptors, was explored. Amiloride analogues increased the dissociation rates of two antagonist radioligands, [3H]8-cyclopentyl-1,3-dipropylxanthine ([3H]DPCPX) and [3H]8-ethyl-4-methyl-2-phenyl-(8R)-4,5,7,8-tetrahydro-1H-imidazo[2,1-i]purin-5-one ([3H]PSB-11), from A(1) and A(3) receptors, respectively. Amiloride and 5-(N,N-dimethyl)amiloride (DMA) were more potent at A(1) receptors than at A(3) receptors, while 5-(N,N-hexamethylene)amiloride (HMA) was more potent at A(3) receptors. Thus, amiloride analogues are allosteric inhibitors of antagonist binding at A(1), A(2A), and A(3) adenosine receptor subtypes. In contrast to their effects on antagonist-occupied receptors, amiloride analogues did not affect the dissociation rates of the A(1) agonist [3H]N(6)-[(R)-phenylisopropyl]adenosine ([3H]R-PIA) from A(1) receptors or the A(2A) agonist [3H]2-[p-(2-carboxyethyl)phenyl-ethylamino]-5'-N-ethylcarboxamidoadenosine ([3H]CGS21680) from A(2A) receptors. The dissociation rate of the A(3) agonist radioligand [125I]N(6)-(4-amino-3-iodobenzyl)adenosine-5'-N-methyluronamide ([125I]I-AB-MECA) from A(3) receptors was decreased significantly by amiloride analogues. The binding modes of amiloride analogues at agonist-occupied and antagonist-occupied receptors differed markedly, which was demonstrated in all three subtypes of adenosine receptors tested in this study. The effects of the amiloride analogues on the action of the A(3) receptor agonist were explored further using a cyclic AMP functional assay in intact CHO cells expressing the human A(3) receptor. Both binding and functional assays support the allosteric interactions of amiloride analogues with A(3) receptors.