The alpha 1a and alpha 1b-adrenergic receptor subtypes: molecular mechanisms of receptor activation and of drug action

In this chapter we summarize some aspects of the structure-functional relationship of the alpha 1a and alpha 1b-adrenergic receptor subtypes related to the receptor activation process as well as the effect of different alpha-blockers on the constitutive activity of the receptor. Molecular modeling of the alpha 1a and alpha 1b-adrenergic receptor subtypes and computational simulation of receptor dynamics were useful to interpret the experimental findings derived from site directed mutagenesis studies.     

Amino acids of the human alpha1d-adrenergic receptor involved in antagonist binding

Computer simulations of the human alpha(1d)-adrenergic receptor (alpha(1d)-AR) based on the crystal structure of rhodopsin have been combined with experimental site-directed mutagenesis to investigate the role of residues in the transmembrane domains in antagonist binding. Our results indicate that the amino acids Asp176 in the third transmembrane domain (TMD), Glu237 in TMD IV, and Ser258 in TMD V of alpha(1d)-AR were directly involved in prazosin and tamsulosin binding. The Asp176Ala mutant did not exhibit any affinity for [(3)H]prazosin and neither did it show agonist-stimulated inositol phosphates (IP) formation. On the other hand, the Glu237Ala and Ser258Ala mutant alpha(1d)-AR showed increased binding affinity for [(3)H]prazosin. Competition binding experiments showed that prazosin affinity had increased to 5-fold and 3-fold in the Glu237Ala and Ser258Ala mutants, respectively, versus wild-type; and tamsulosin affinity only increased in the Ser258Ala mutant (2-fold vs wild-type). It seems that these two residues constrain the receptor by interaction with other residues and this disruption of the interaction increased the receptor's binding affinity towards antagonists. However, the Glu237Ala and Ser258Ala mutant receptors retained the ability to stimulate the formation of myo-[(3)H]inositol but had activities lower than that of the wild-type receptor. The present results provide direct evidence that these amino acid residues are responsible for the interactions between alpha(1d)-AR and the radioligand [(3)H]prazosin as well as tamsulosin.     

Identification of subtypes of [3H]prazosin-labelled alpha 1 receptor binding sites in rat brain

Detailed antagonist competition curves for [3H]prazosin-labelled binding sites in rat cerebral cortex membranes reveal anomalous binding characteristics. Dihydroergocryptine and indoramine compete in a steep, monophasic manner while WB4101 and phentolamine exhibit shallow competition curves. Computer-assisted analysis of binding data indicate that both WB4101 and phentolamine discriminate identical subpopulations of [3H]prazosin binding sites, which each comprise approximately 50% of specific [3H]prazosin binding. These data suggest the presence of subtypes of [3H]prazosin-labelled alpha 1 adrenergic receptors in rat cerebral cortex.     

Phenoxybenzamine is more potent in inactivating alpha 1- than alpha 2-adrenergic receptor binding sites

The effects of phenoxybenzamine on alpha 1- and alpha 2-adrenergic receptor binding sites were examined directly using radioligand binding assays of the antagonist drugs [125I]BE 2254 and [3H]rauwolscine, respectively, in homogenates of rat cerebral cortex. Treatment with phenoxybenzamine caused an irreversible, dose-dependent decrease in the density of both alpha 1- and alpha 2-adrenergic receptors. Phenoxybenzamine was approximately 250-fold more potent at decreasing alpha 1-adrenergic receptor density than it was at decreasing alpha 2-adrenergic receptor density.     

Different internalization properties of the alpha1a- and alpha1b-adrenergic receptor subtypes: the potential role of receptor interaction with beta-arrestins and AP50

The internalization properties of the alpha1a- and alpha1b-adrenergic receptors (ARs) subtypes transiently expressed in human embryonic kidney (HEK) 293 cells were compared using biotinylation experiments and confocal microscopy. Whereas the alpha1b-AR displayed robust agonist-induced endocytosis, the alpha1a-AR did not. Constitutive internalization of the alpha1a-AR was negligible, whereas the alpha1b-AR displayed significant constitutive internalization and recycling. We investigated the interaction of the alpha1-AR subtypes with beta-arrestins 1 and 2 as well as with the AP50 subunit of the clathrin adaptor complex AP2. The results from both coimmunoprecipitation experiments and beta-arrestin translocation assays indicated that the agonistinduced interaction of the alpha1a-AR with beta-arrestins was much weaker than that of the alpha1b-AR. In addition, the alpha1a-AR did not bind AP50. The alpha1b-AR mutant M8, lacking the main phosphorylation sites in the receptor C tail, was unable to undergo endocytosis and was profoundly impaired in binding beta-arrestins despite its binding to AP50. In contrast, the alpha1b-AR mutant DeltaR8, lacking AP50 binding, bound beta-arrestins efficiently, and displayed delayed endocytosis. RNA interference showed that beta-arrestin 2 plays a prominent role in alpha1b-AR endocytosis. The findings of this study demonstrate differences in internalization between the alpha1a- and alpha1b-AR and provide evidence that the lack of significant endocytosis of the alpha1a-AR is linked to its poor interaction with beta-arrestins as well as with AP50. We also provide evidence that the integrity of the phosphorylation sites in the C tail of the alpha1b-AR is important for receptor/beta-arrestin interaction and that this interaction is the main event triggering receptor internalization.     

Demonstration of alpha 1A- and alpha 1B-adrenoceptor binding sites in human brain tissue

Radioligand binding studies suggest that alpha 1-adrenoceptor recognition sites are heterogeneous. Several adrenergic agents discriminate between two adrenoceptor binding sites designated alpha 1A and alpha 1B. In the present study we demonstrate for the first time that these two subtypes exist in the human brain. 5-Methyl-urapidil and (+)-niguldipine, which have previously been shown to be alpha 1A-selective, inhibited [3H]prazosin binding to cortical membranes in a biphasic manner. The irreversible alpha 1B-ligand, chloroethylclonidine, preferentially eliminated the binding sites with low affinity for (+)-niguldipine. In contrast, BE 2254 and unlabelled prazosin displaced the radioligand in a monophasic manner. The IC50 values for prazosin were not affected by pretreatment of the membranes with chloroethylclonidine. Our data on human brain membranes are in excellent agreement with recent findings in rat tissues and suggest that the alpha 1-adrenoceptor subtypes in human brain are similar to those in rat tissues.     

Interaction of subtype-selective antagonists with alpha 1-adrenergic receptor binding sites in rat tissues.

(+)-Niguldipine inhibited specific 125I-BE 2254 binding more potently in membrane preparations from rat tissues enriched in the alpha 1A subtype (hippocampus and vas deferens) than those with the alpha 1B subtype (liver and spleen). Inhibition curves for (+)-niguldipine were better fit by a two-site model in most tissues, although Kl values for each site varied markedly between tissues. The potency of this lipophilic drug was highly dependent on tissue concentration, probably accounting for most of this variability. Pretreatment of membranes with chloroethylclonidine (CEC) to inactivate the alpha 1B subtype did not completely eliminate the low affinity sites for (+)-niguldipine, particularly in heart. Saturation analysis showed that (+)-niguldipine competitively inhibited both alpha 1A and alpha 1B subtypes. However, substantial non-competitive inhibition was also observed in several tissues. Analysis of inhibition curves for 5-methylurapidil gave similar proportions of alpha 1A and alpha 1B receptor sites as were calculated for (+)-niguldipine in various tissues. Although (+)-niguldipine and 5-methylurapidil revealed variable proportions of low affinity sites in CEC-pretreated hippocampus and heart, this was not observed with inhibition curves for WB 4101 and phentolamine. These results are generally consistent with the previously defined alpha 1A and alpha 1B subtypes. 5-Methylurapidil currently appears to be the best antagonist for discriminating these subtypes; (+)-niguldipine shows similar selectivity but is complicated by a high lipophilicity. However, the persistence of low affinity sites for 5-methylurapidil and (+)-niguldipine after CEC pretreatment and the noncompetitive effects of (+)-niguldipine in some tissues raise the possibility of an additional subtype(s) of alpha 1-adrenergic receptors in rat tissues.     

5-HT3 receptor binding sites are on capsaicin-sensitive fibres in the rat spinal cord

Specific binding sites for [3H]zacopride were found in the dorsal part of the rat spinal cord, particularly in the superficial layers of the dorsal horn. These binding sites had the same pharmacological profile as 5-HT3 receptors in membranes from the rat entorhinal cortex or from NG 108-15 neuroblastoma-glioma cells. Administration of capsaicin (50 mg/kg s.c.) to neonatal rats to induce degeneration of unmyelinated primary sensory fibres resulted in a significant decrease in [3H]zacopride specific binding (-50%) in the dorsal zone of the spinal cord of 4 month-old rats. This decrease was as pronounced as the decrease in [3H]bremazocine and [3H]naloxone binding to opiate receptors. These data support the presynaptic location of 5-HT3 receptors, at least in part, on capsaicin-sensitive primary afferent fibres in the rat spinal cord.     

Disposition and exposure of the fibrinogen receptor antagonist XV459 on alphaIIBbeta3 binding sites in the guinea pig

The disposition of XV459, a potent, selective GP IIb/IIIa antagonist, has been examined following intravenous administration of XP280, the benzenesulphonate salt, and 3H-SA202, the trifluroacetic acid salt, to male guinea pigs. A liquid chromatography-mass spectrometry (LC-MS) method was developed and validated for XV459 quantitation in guinea pig plasma with an LLOQ of 0.1 ng/mL. Intravenous infusions (30 min) of XP280 at doses of 0.5 and 2.0 microg/kg were administered to guinea pigs which were sequentially sacrificed at 0.5, 1, 1.5, 4, 8, 12, 24, 48 and 72 h postinitiation of infusion. Maximum total (unbound and GP IIb/IIIa displaced) XV459 plasma concentration of approximately 3.5 microg/mL was obtained at the 2.0 microg/kg dose. Pooling individual concentration-time data yielded a systemic clearance of 1.42 mL/min/kg, Vss of 0.24 L/kg, and a terminal half-life of 2.8 h in the guinea pig at the 0.5 microg/kg dose. The 2.0 microg/kg dose yielded XV459 exposure that was less than proportional to the previous dose. Similar behaviour has been observed in human trials. Cumulative (up to 72 h) urinary and faecal recovery of total radioactivity was 66.4 and 11.2%, respectively. The time course of spleen, marrow and whole blood radioactivity profiles was similar, suggesting that XV459 was not preferentially sequestered on non-plasma GP IIb/IIIa binding sites. Tissue to blood ratios of 20.7 and 8.3 for the spleen and bone marrow, respectively, indicate that increased (relative to blood) exposure was evident for sites containing the GP IIb/IIIa receptor. In vitro studies confirmed the similarity of XV459 binding to both resting and activated platelets in the guinea pig and humans. Given the comparability of dissociation rate constants and IC50s based on in vitro platelet aggregation, human dosimetry estimates should assume similar partitioning of radiolabelled XV459 as in the guinea pig. These results suggest that the guinea pig may indeed be an appropriate animal model for pharmacokinetic and distribution studies with DMP754; in conjunction with recent pharmacological findings with GP IIb/IIIa antagonists, our results suggest that the guinea pig may be the rodent species of choice for preclinical studies with some other GP IIb/IIIa antagonists.     

Nonadrenergic [3H]idazoxan binding sites are physically distinct from alpha 2-adrenergic receptors

We have recently demonstrated that the alpha 2-adrenergic radioligand [3H]idazoxan also labels additional sites that do not recognize catecholamines but bind with high affinity several chemically distinct drugs previously assumed to be highly selective for alpha 2-adrenergic receptors [Mol. Pharmacol. 35:324-330 (1989)]. We now have used three approaches to distinguish the nonadrenergic [3H]idazoxan sites from alpha 2-adrenergic receptors. (a) No nonadrenergic [3H]idazoxan binding sites were found in COS-7 cells transfected with the genes for the two known alpha 2-adrenergic receptor subtypes. (b) The ratio of alpha 2-adrenergic and nonadrenergic [3H]idazoxan sites in human platelet membranes varied considerably between various donors. (c) Highly purified platelet plasma membranes were enriched for alpha 2-adrenergic receptors but did not contain any nonadrenergic [3H]idazoxan binding sites. We conclude that the nonadrenergic [3H]idazoxan binding sites are not co-expressed with alpha 2-adrenergic receptors and at least in human platelets may be located in an intracellular compartment.     

Functions of acidic transmembrane residues in human melanocortin-3 receptor binding and activation

The melanocortin-3 receptor (MC3R) is an important regulator of energy homeostasis, inflammation, and cardiovascular function. Inactivating mutations in MC3R gene are associated with childhood obesity. How MC3R binds to its ligands has rarely been studied. In the present study, we systematically mutated all ten acidic residues in transmembrane (TM) domains and measured the cell surface expression levels as well as ligand binding and signaling properties of these mutants. Our results showed that of the 19 mutants stably expressed in HEK293 cells, all were expressed on the cell surface, although some mutants had decreased levels of cell surface expression. We showed that with the superpotent analog [Nle(4), D-Phe(7)]-alpha-melanocyte stimulating hormone (MSH), E92, E131, D154, D158, D178, and D332 are important for ligand binding. D121 and D332 are important for binding and signaling. Further experiments using other ligands such as D-Trp(8)-gamma-MSH, alpha-MSH and gamma-MSH showed that different ligands induce or select different conformations. In summary, we showed that acidic residues in TMs 1 and 3 are important for ligand binding whereas the acidic residues in TMs 2 and 7 are important for both ligand binding and signaling.     

Neuroleptic drug interactions with norepinephrine alpha receptor binding sites in rat brain.

Alpha adrenergic receptor sites in mammalian brain tissue can be labeled by the binding of [³H]WB-4101 (2-([2′,6′-dimethoxy] phenoxyethylamino) methyl benzodioxan), a potent α-adrenergic antagonist. Numerous neuroleptic drugs of phenothiazine, butyrophenone and thioxanthene classes are potent in competing for [³H]WB-4101 binding, with affinities resembling those of classic α-antagonists such as phentolamine and phenoxybenzamine. The potencies of neuroleptics in competing for WB-4101 binding sites correlate closely with their potencies in antagonizing norepinephrine and epinephrine induced lethality in rats, confirming that affinity for WB-4101 binding sites predicts α-receptor antagonism in vivo. The relative affinities of neuroleptics for WB-4101 binding sites and for dopamine receptors as labeled by [³H]haloperidol provides an index of the relative propensities of these drugs for eliciting autonomic side effects such as orthostatic hypotension and sedation.     

Specific binding sites for 125I-endothelin-1 in the porcine and human spinal cord

Specific binding sites for ¹²⁵I-endothelin-1 (¹²⁵I-ET-1) in the spinal cord were investigated using quantitative receptor autoradiographic and chemical cross-linking methods. The binding sites were highly concentrated in porcine and human spinal cord areas corresponding anatomically to the dorsal horn (Rexed's laminae I–III), an area around the central canal (lamina X) and the principal part of the intermediolateral nucleus (IMLp). The localization of the binding sites differed from those of ¹²⁵I-ω-conotoxin GVIA (¹²⁵I-CgTx) and ¹²⁵I-Bolton-Hunter substance P (¹²⁵I-BH-SP), with the exception that the IMLp shared ¹²⁵I-ET-1 with ¹²⁵I-CgTx and ¹²⁵I-BH-SP binding sites. Specific ¹²⁵I-ET-1 binding sites in the areas examined were characteristically single and of high affinity. There were no differences between the potencies of unlabeled ET family peptides, ET-1, ET-2, ET-3 and sarafotoxin S6b at inhibiting ¹²⁵I-ET-1 binding to the areas. Chemical cross-linking studies showed that ¹²⁵I-ET-1 and ¹²⁵I-ET-3 mainly bound to a protein with molecular mass of 43 kDa in the porcine and human thoracic spinal cord membranes. The present finding shows the neuronal significance of this newly discovered peptide in the spinal cord.     

A model of the adrenergic beta-2 receptor and binding sites for agonist and antagonist.

From the known experimental data on adrenergic ligands and beta-2 receptor a model for the ligand binding sites was proposed and built. Agonist and antagonist have overlapping but different binding sites. The model correlates well with the structural information known from a series of adrenergic ligands.     

Positive inotropic effects of imidazoline derivatives are not mediated via imidazoline binding sites but alpha1-adrenergic receptors

Imidazoline-binding sites are non-adrenergic receptors and classified into I11/I2 subtypes. There is strong evidence that I1-binding sites, located in the rostro-ventrolateral medulla, are involved in regulation of blood pressure. However, less is known about the peripheral participation of I1-binding sites in cardiovascular reactions. Therefore, the aim of this study was to investigate whether specific imidazoline derivatives influence myocardial contractility and whether imidazoline binding sites are expressed in rat heart. Agmatine, clonidine and idazoxan failed to alter inotropy in left atria within the whole concentration range tested (1 nM - 100 microM), whereas cirazoline (1- 100 microM) and moxonidine (100 microM) increase inotropy by about 20-30%. After preincubation with the alpha1-adrenoceptor antagonist prazosin, the cirazoline and moxonidine stimulated inotropy was antagonized, indicating more an alpha1-adrenergic and less an imidazoline binding site mediated mechanism. Radioligand-binding studies in membranes of left ventricles using [3H]-clonidine to specify I1-binding yielded KD values of 12.7 microM, confirming the functional results of an absence of I1-binding sites in ventricles of rats. However, the existence of low affinity I2-binding sites determined by [3H]-idazoxan labeling could not be excluded since a KD of 0.5 microM was calculated and since competition studies with guanabenz (Ki = 0.1 microM), clonidine (Ki = 58.1 microM) and moxonidine (Ki = 129 microM) confirmed the specificity of the I2-binding.     

Specific 125I-endothelin-1 binding sites in the atrioventricular node of the porcine heart

The quantitative receptor autoradiographic method we used revealed that specific 125I-endothelin-1 binding sites are highly concentrated in the atrioventricular node of the porcine heart. 125I-Endothelin-1 binding to the atrioventricular node, interatrial and interventricular septa was displaced by unlabeled endothelin-1 with Kd values of 53 pM, 2.03 nM and 3.48 nM, respectively. Knowledge of the existence of specific 125I-endothelin-1 binding sites in the porcine heart helps with understanding the physiology of endothelin, a possible endogenous Ca2+ channel agonist.     

Estradiol and guanine nucleotide modulation of dopamine receptor agonist and antagonist binding sites in 7315a pituitary tumors

The agonist high- and low-affinity states of the dopamine (DA) receptor were investigated with apomorphine competition for [3H]spiperone binding to DA receptors in 7315a tumors grown in intact female rats, while the antagonist site was investigated with saturation of [3H]spiperone binding. Such as for the intact pituitary, the antagonist binding site density in 7315a tumors was not affected by NaCl and/or Gpp(NH)p, and its binding affinity was increased in the presence of NaCl. The DA receptor in 7315a tumors existed in high- and low-affinity agonist states, and the two apomorphine sites had similar affinities in tumoral and intact tissue. However, the proportion of the high affinity state was slightly lower in the 7315a tumor compared to intact tissue. Tumor (7315a) growth in ovariectomized rats was slower than in intact animal; chronic 17 beta-estradiol treatment inhibited growth of these tumors. Prolactin (PRL) concentration and density of DA receptors were higher in tumors grown in ovariectomized than in intact female rats, whereas both decreased after 23 days of 17 beta-estradiol treatment. Estradiol treatment decreased the affinity of the high- and the low-apomorphine binding sites, whereas their proportions remained unchanged. Thus, changes of DA receptors and 7315a tumor growth seem to be related; however, their relationship is complex.     

Specific binding sites for 125I-endothelin-1 in the porcine and human spinal cord.

Specific binding sites for 125I-endothelin-1 (125I-ET-1) in the spinal cord were investigated using quantitative receptor autoradiographic and chemical cross-linking methods. The binding sites were highly concentrated in porcine and human spinal cord areas corresponding anatomically to the dorsal horn (Rexed's laminae I-III), an area around the central canal (lamina X) and the principal part of the intermediolateral nucleus (IMLp). The localization of the binding sites differed from those of 125I-omega-conotoxin GVIA (125I-CgTx) and 125I-Bolton-Hunter substance P (125I-BH-SP), with the exception that the IMLp shared 125I-ET-1 with 125I-CgTx and 125I-BH-SP binding sites. Specific 125I-ET-1 binding sites in the areas examined were characteristically single and of high affinity. There were no differences between the potencies of unlabeled ET family peptides, ET-1, ET-2, ET-3 and sarafotoxin S6b at inhibiting 125I-ET-1 binding to the areas. Chemical cross-linking studies showed that 125I-ET-1 and 125I-ET-3 mainly bound to a protein with molecular mass of 43 kDa in the porcine and human thoracic spinal cord membranes. The present finding shows the neuronal significance of this newly discovered peptide in the spinal cord.