Abstract
The adrenergic receptors play a key role in the modulation of sympathetic nervous system activity as well as a site of action for many therapeutic agents. The α1-adrenergic receptor subtypes (α1A-, α1B-, α1D) are the prime mediators of smooth muscle contraction and hypertrophic growth, but their characterization in both binding and function have lagged the other adrenergic family members. Although they are derived from a related ancestral gene and all nine adrenergic receptor family members bind the endogenous ligands, epinephrine and norepinephrine, with roughly similar affinities, there are major differences in the mode of binding, second messenger utilization, and physiological effects of the α1-subtypes compared with β- or α2-subtypes. Here, we review the recent literature on aspects of its binding pocket and how it differs from the β-adrenergic receptor paradigms. We also review the signaling components and aspects of its function and provide new insights into its roles in smooth muscle, growth, neurological, and cardiovascular function.
Classification
Adrenergic receptors (α1A-, α1B-, α1D-, α2A-, α2B-, α2C-, β1, β2, β3) are members of the G-protein-coupled receptor (GPCR) superfamily of membrane proteins that mediate the actions of the endogenous catecholamines, norepinephrine and epinephrine. Similar to rhodopsin, these proteins are proposed to traverse the membrane in seven transmembrane (TM)-spanning α-helical domains linked by three intracellular and three extracellular loops. Since their original classification of adrenergic receptors into stimulatory and inhibitory receptors and subdivision into α1- and α2-ARs, it became apparent that there was heterogeneity in α1-ARs. Indeed, prior to the cloning of any receptor subtypes, numerous reports provided functional evidence of α1-AR heterogeneity. McGrath was the first to suggest subdividing the α1-ARs into α1A- and α1B-ARs. Morrow and Creese noted that the inhibition curves for a series of agonists and antagonists to displace [3H]prazosin were biphasic. Since these initial studies, the α1A-subtype was pharmacologically classified to have higher binding affinity for agonists, such as methoxamine and oxymetazoline, and antagonists, such as 5-methylurapidil, (+)niguldipine, and WB4101. In contrast, the α1B-AR subtype had lower binding affinity for the above ligands (reviewed in Minneman, 1988).
After these initial pharmacological studies, the first cDNA cloned was the hamster α1B-AR. This cDNA had all of the pharmacological properties of the tissue-characterized α1B-AR and has never been questioned in its classification. The next receptor cloned was called the α1C-AR and was thought to represent a novel subtype. However, it was later reclassified to be the tissue-type α1A-AR. The confusion was centered on its inability to localize its mRNA to tissues known to express the α1A-AR. The next cDNA cloned was initially termed the α1A-AR but also later was reclassified to a novel subtype called the α1D-AR (reviewed in Hieble et al., 1995). In this case, the confusion was due to an incomplete pharmacological profile. With the discovery of the α1D-AR, its binding and functional properties were compared with the previously known tissue subtypes. The α1D-AR has a binding profile much like the α1B-AR (reviewed inGraham et al., 1996). Recently, an α1D-AR-selective drug has become available (Table 1) (Saussy et al., 1996). Since this subtype was never classified from tissue studies, its functional role has remained largely unexplored. Contraction of large caliber-type arteries have been found to be controlled by the α1D-AR (Piascik et al., 1995, 1997) but other tissue types need to be characterized. A summary of current characteristics of the three α1-AR subtypes is shown in Table 1.
Historically, the cloning of the β-AR by Dixon and its homology to rhodopsin initiated the field to recognize that all GPCRs are encoded by genes with similar features. There is conservation of particular sequence(s) and spacing between key functional amino acids, especially in the transmembrane domains and where the G-proteins are predicted to bind and activate. This has suggested the viewpoint that the entire GPCR family arose from a single ancestral gene. The genomic structure of the α1-ARs have been reported, and all three subtypes have a large intron after the TM6 domain (reviewed in Graham et al., 1996). Phylogenetically, α1-ARs are considered the closest neighbor to the rhodopsin family because of the presence of introns in the coding regions and their high sequence homology. Polymorphisms have been reported in the α1-ARs but have not as of yet been linked to functional consequences, unlike members of the β-ARs (reviewed inBuscher et al., 1999).
Binding Pocket
α1-AR Agonist Binding and Differences from the β-AR.
Early studies with the β-AR suggested that the agonist binding pocket was constituted by the interaction of the protonated amine of the catecholamine that hydrogen bonds to an aspartate residue in TM3. Both agonist and antagonist binding are more severely affected by mutation of the aspartate in TM3 of the β-AR (Strader et al., 1989) than in the α1-AR (Porter et al., 1996). It can be argued that mutants made of this residue are highly unstable and very little surface expression occurs, making binding analysis difficult. The decrease in agonist affinity of the α1-AR aspartate mutants could also have been less than expected due to their constitutive activity that results in increases in agonist affinity. We have shown that this aspartate is also part of the activation process in which disruption of the salt bridge by the agonist between this aspartate and a lysine residue in TM7 initiates activation but does not account for full activation (Porter et al., 1996). This activation mechanism also appears to be conserved in the δ-opioid receptor (Befort et al., 1999).
There are also differences in how the serines in TM5 of the β2- and α1-ARs interact with the catechol hydroxyls. It was proposed that Ser204 of the β2-AR forms a hydrogen bond with themeta-hydroxyl group of the catecholamine whereas Ser207 forms a hydrogen bond with the para-hydroxyl group. Both of these serines contribute equally (about 50%) and are required for full efficacy (Strader et al., 1989). In the α1A-AR, we found that Ser188 (corresponds to Ser203 in the β-AR) interacts with the meta-hydroxyl whereas Ser192 (corresponds to Ser207) interacts with the para-hydroxyl. We also found that only Ser188 is involved in the activation with minimal contribution from Ser192. In modeling these interactions, the catechol ring appears to dock in a planar orientation in the α1-AR but is tilted by 120 degrees relative to the skewed orientation in the β-AR (Hwa and Perez, 1996).
The aromatic interactions of the phenyl ring of the agonist and the phenylalanine residue in TM6 are also different in the α1- than the β2-AR. In the β2-AR, Phe290 (equivalent to Phe311 in the α1B-AR) was suggested to be involved in an aromatic interaction with the agonist (Strader et al., 1989). In contrast, a detailed study in the α1B-AR using multiple techniques demonstrated that only Phe310 (equivalent to Phe289 in the β2-AR) is critically involved in both binding and activation (Chen et al., 1999).
Stereoselectivity in the β2-AR has been attributed to an interaction of the β-hydroxyl group of the agonist and Asn293 in TM6 (Wieland et al., 1996). This study ruled out Ser165 in TM4, which was previously postulated to impart stereoselectivity but could not be expressed (Strader et al., 1989). However, Asn or a homologous residue that is capable of hydrogen bonding is not present in TM6 of the α1-AR, although stereoselectivity of compounds is conserved. Thus, the actual residue responsible for stereoselectivity in the α1-AR has yet to be determined.
Although all of the above residues have been previously regarded as the completed agonist binding pocket, we have recently reported the unexpected finding of two additional phenylalanine residues involved in the binding of agonists (Waugh et al., 2000). Phe163 in TM4 and Phe187 in TM5 of the α1A-AR are involved in agonist-specific binding interactions. Mutation of both of these residues contributes to a 150-fold loss of affinity for the endogenous agonist. Interestingly, the β2-AR does not conserve these aromatic residues, further illustrating inherent differences in the agonist binding pocket between these two receptors.
The difference in catechol ring orientation as concluded from Hwa and Perez (1996) actually explains many potential differences in β2- versus α1-AR agonist binding. Modeling studies of Phe310 in TM6 of the α1B-AR favors interactions with the planar form of the agonist (Chen et al., 1999). Optimization of binding interactions in TM4 and 5 were achieved when the agonist was also in a planar orientation (Waugh et al., 2000). A planar orientation would also require a different residue responsible for stereoselectivity, explaining the nonconserved role of Asn293 in TM6 of the β2-AR.
All of the agonist interactions reviewed have been applicable to nonselective agonist binding in general and not to the intricacy of subtype-selective ligand interactions. The α1A-AR displays a 10- to 100-fold higher binding affinity for the agonists compared with the α1B-AR subtype. There are two residues responsible for this selectivity, Ala204 in TM5 and Met313 in TM6 of the α1B-AR (Hwa et al., 1995). Although the interactions were direct, the mechanism of this selectivity was due to the particular packing interactions of these two residues as mismatched combinations of the α1B- and α1A-AR residues resulted in constitutive activity.
A summary of all the residues identified in agonist binding is shown in Fig. 1. This molecular model of the α1A-AR indicates that the agonist binding pocket is formed from residues in TM3 through TM6. Residues in TM7 have not been identified to be involved in agonist binding. This is also unlikely since the carbon chain of the protonated amine tends to be limited to small aliphatic chains. Increased bulk at this position tends to convert α1-AR agonists into antagonists.
Antagonist Binding.
Our knowledge, however, of how antagonists bind to the α1-AR is much more limited. Mutagenesis studies in our laboratory have identified that the subtype-selectivity of two α1A-AR-selective antagonists, phentolamine and WB4101, is conferred by interactions with three consecutive residues (Gln177, Ile178, Asn179) of the second extracellular loop (Zhao et al., 1996). Similar results were also obtained in the 5-hydroxytryptamine1D receptor (Wurch et al., 1998) and the δ-opioid receptor (Varga et al., 1996). This can be reconciled in the structure of rhodopsin (Palczewski et al., 2000) in which the second extracellular loop folds down into the binding pocket of retinal. A phenylalanine residue (Phe86) at the surface of TM2 in the α1A-AR accounts for the α1A- versus α1D-selectivity of dihydropyridine antagonists such as niguldipine (Hamaguchi et al., 1996). However, all of the above studies involved residues involved in selectivity and with a limited set of antagonists. We recently reported two conserved phenylalanine residues near the extracellular surface of TM7 involved in nonselective binding for almost all α1-antagonists (Waugh et al., 2001). This study represents the first report of a common site of antagonist binding for members of the adrenergic receptor family. Interestingly, these two residues also altered imidazoline binding such as cirazoline. It has often been regarded in the past that imidazoline agonists bind in a different manner than phenethylamine agonists, and this study supports that hypothesis. It appears that imidazolines bind much like an antagonist, and this may explain their partial agonist properties.
Cellular Localization
Classically, GPCRs have been thought of as being expressed predominantly on the cell surface where they are accessible to water-soluble ligands. However, recent data have suggested that the cellular localization of GPCRs is more complicated than previously envisioned. In transfected fibroblasts, the α2C-AR was detected in intracellular compartments as well as on the cell surface, whereas the α2A-AR was found exclusively on the cell membrane (Daunt et al., 1997). Results obtained over the last several years have shown that there are major differences in the subcellular distribution of the α1-AR subtypes. Fonseca et al. (1995) developed a peptide antibody against a sequence in the C-terminal tail of the hamster α1B-AR. Immunocytochemistry with HEK 293 cells stably transfected with the α1B-AR showed that this receptor is localized predominantly on the cell membrane. This result was confirmed byHirasawa et al. (1997) who used α1B-AR/GFP fusion proteins to demonstrate cell membrane localization of the α1B-AR in COS-7 cells. Using similar protocols, these authors also demonstrated that GFP/α1A-AR constructs were expressed in intracellular compartments. Intracellular α1A-AR expression was confirmed using antibodies directed against a FLAG-tagged epitope inserted into the α1A-AR (Hirasawa et al. (1997). Intracellular expression of the α1-ARs was also observed in a unique series of studies from the laboratory of McGrath. These workers used BODIPY-FL-labeled prazosin to image α1-AR subtypes in cultured prostate smooth muscle cells and fibroblasts stably transfected with each subtype. These authors noted intracellular expression for each of the three subtypes in the fibroblast cell lines. These authors estimate that in smooth muscle cells, 40% of the total α1-AR population is expressed intracellularly (McGrath et al., 1999; Mackenzie et al., 2000).
We studied the cellular distribution of the α1-AR subtypes in stably transfected fibroblasts as well as cultured vascular smooth muscle cells. Using either commercially available antibodies or the antibody developed byFonseca et al. (1995), we observed that α1B-AR was expressed predominantly on the cell surface (Hrometz et al., 1999). In agreement with the work of Hirasawa et al. (1997), we noted an intracellular localization of the α1A-AR. We also detected a significant degree of cell surface expression of the α1A-AR in both fibroblasts and vascular smooth muscle cells (Hrometz et al., 1999; McCune et al., 2000). Surprisingly, we detected very little cell surface expression of the α1D-AR. Indeed most of the α1D-AR immunoreactivity was detected intracellularly in a perinuclear orientation. To assess cellular localization in a manner that does not require immunostaining, we transfected HEK 293 cells with cDNA encoding α1-AR/GFP fusion proteins (D. Chalathorn, D. F. McCune, S. E. Edelmann, M. L. Garcia, G. Tsujimoto, and M. T. Piascik, submitted for publication). Living cells were then visualized in real time with laser scanning confocal microscopy. We detected α1B-AR/GFP fluorescence predominantly on the cell surface whereas the α1A-AR was on the cell surface but was also intracellular. The α1D-AR was detected mainly intracellularly (Table 1). The cumulative weight of the localization data indicate that the α1-ARs are expressed, to one degree or another, in intracellular compartments. The experimental challenge now is to assess the functional significance of this localization and whether the receptors can be attacked therapeutically.
Constitutively Active α1-ARs.
The fact that constitutively active mutations have been engineered into recombinant forms of the α1B-AR and α1A-AR is well known. Also, artificial overexpression of many wild-type receptors also can cause constitutive activity, depending upon the amount of overexpression and the cell type that it is expressed in. However, there is good evidence that the native α1D-AR is constitutively active. Gisbert et al. (2000) provided evidence that in the rat aorta, α1D-AR is constitutively active with regard to contractile activity. Garcia-Sainz and Torres-Padilla (1999) showed constitutive activity with regard to calcium transients in stably transfected Rat-1 fibroblasts. In fibroblasts, the α1D-AR displays constitutive activity with respect to inositol phosphate formation and the activation of extracellular signal-regulated kinase (ERK) (McCune et al., 2000). The native constitutive activity of the α1D-AR would also explain its higher binding affinity for agonists than the α1A-AR subtype with no corresponding higher binding affinity for antagonists. The observation of constitutive activity may shed some light on the relationship between α1D-AR cellular localization and functional responses. The large degree of intracellular localization of the α1D-AR in unstimulated cells may be because the receptor is continuously internalized due to its constitutively active nature.
Cellular Signaling Pathways
The α1-ARs utilize a variety of second messenger pathways to modulate cellular function. This topic has been recently and extensively reviewed (see Garcia-Sainz et al., 2000). Studies with many cell types demonstrate that all α1-ARs activate phospholipases C and A2 (Perez et al., 1993). In addition to mobilizing intracellular calcium, the α1-ARs have also been shown to activate calcium influx via voltage-dependent and independent calcium channels (Minneman, 1988). Additionally, these receptors signal through both pertussis toxin-sensitive G-proteins (Perez et al., 1993) and G proteins of the Gqfamily (Wu et al., 1992). Minneman and associates studied the coupling of the α1-AR subtypes and noted that there were marked differences in the ability of α1-ARs to generate intracellular second messengers (Table 1). In particular, these authors noted that the α1A-AR was the most efficiently coupled to calcium release and inositol phosphate production whereas the α1D-AR was poorly coupled to intracellular signaling cascades (Theroux et al., 1996), suggesting potential differences in the functional outcomes of α1-AR activation.
In addition to modulating pathways that link the α1-ARs to calcium movements and smooth muscle contraction, the α1-ARs are also intimately involved in the regulation of growth promoting responses via the mitogen-activated protein kinase (MAPK) family. There are at least three major subfamilies of MAPK, including the ERKs, c-Jun N-terminal kinases (JNKs), and the p38 kinases. The activity of all MAPK family members is regulated through a series of phosphorylation events. MAPKs, in turn, phosphorylate numerous nuclear transcription factors and other cytosolic proteins making these enzymes key regulators of cellular growth (reviewed in Widmann et al., 1999). α1-AR-stimulated MAPK signaling pathways potentially contribute to increased DNA synthesis and cell proliferation in human vascular smooth muscle cells (Hu et al., 1999). α1-AR stimulation also activates 46-kDa JNK, 54-kDa JNK, and p38 kinases in Rat-1 fibroblasts but uses at least partly different pathways to do so (Alexandrov et al., 1999). Coupling of the three α1-AR subtypes to MAPK pathways were also studied in stably transfected PC12 cells. These studies show that α1A-ARs activate all three MAPK pathways, α1B-ARs activate ERKs and p38 but not JNKs, and α1D-ARs only activate ERKs (Zhong and Minneman, 1999). Therefore, the activation of MAPKs may be highly dependent upon the particular α1-AR subtype but also the particular cell line or tissue that it is expressed in.
GPCR signaling is also tightly regulated by a series of cellular proteins that promote receptor desensitization and internalization. Agonist occupation promotes receptor phosphorylation by a series of GPCR kinases. The phosphorylated receptor exhibits high affinity for the arrestins, which in turn prevent further interaction between the receptor and G-proteins. The β-arrestins promote internalization via clathrin-coated pits (Gagnon et al., 1998). The agonist-dependent phosphorylation of the α1-AR as well as the characteristics of homologous and heterologous desensitization have been recently and extensively reviewed by Garcia-Sainz and coworkers (2000). A majority of the studies on the α1-AR subtypes have been with the α1B-AR. For example, key domains involved in α1B-AR phosphorylation and desensitization have been identified (Diviani et al., 1997; Wang et al., 2000).
Much less is known about the phosphorylation, internalization, and desensitization of the other α1-ARs. Vazquez-Prado et al. (2000) demonstrated that following agonist activation, the α1B-AR is more extensively phosphorylated than the α1A-AR. Furthermore, these authors demonstrated that the phosphorylation of this receptor is mediated by protein kinase C activation. Yang and coworkers (1999) used fibroblasts stably transfected with each of the α1-ARs to show that the increase in inositol phosphates mediated by the α1A- and α1B-ARs could be desensitized whereas the increase mediated by the α1D-AR was refractory to agonist-mediated desensitization (Table 1).
We used α1-AR/GFP fusion proteins and real time imaging of transiently transfected HEK 293 cells to show that there were marked differences in the ability of phenylephrine to promote internalization of these receptors (D. Chalathorn, D. F. McCune, S. E. Edelmann, M. L. Garcia, G. Tsujimoto, and M. T. Piascik, submitted for publication). The α1B-AR underwent rapid internalization following exposure to phenylephrine. However, the rate of cell surface α1A-AR internalization was slower than that seen for the α1B-AR, whereas the α1D-AR was unaffected by phenylephrine treatment. Cotransfection of a dominant negative form of β-arrestin-1, blocked agonist-mediated internalization, indicating that β-arrestin-1 participates in agonist-mediated internalization of the α1-ARs.
α1-ARs in Neurological Function
Although α1-ARs are the most abundant adrenergic receptor in the brain, their role in the CNS is the least understood. The α1-ARs are considered a postsynaptic and not a presynaptic receptor, but like other postsynaptic receptors in the brain, can cause modulation of the release of neurotransmitters. α1-ARs are generally considered stimulatory in nature. In the somatosensory areas of the cortex, α1-AR activation has been found to increase the excitation seen after administration of glutamate or acetylcholine (Mouradian et al., 1991). α1-ARs can also directly enhance neurotransmitter release from glutamate terminals that innervate layer V pyramidal cells of the prefrontal cortex (Marek and Aghajanian, 1999). α1-ARs also cause excitatory responses in subcortical areas such as the medial and lateral geniculate nuclei, the reticular thalamic nucleus, dorsal raphe, and spinal motor neurons. It appears that this activation is largely due to a decrease in resting potassium conductance and not to an increase in calcium (McCormick et al., 1991). α1-ARs may modulate activation of the pyramidal neurons in the neocortex and may play a role in attention and memory (as reviewed in Sirvio and MacDonald, 1999). α1-ARs may effect many brain functions via non-neuronal mechanisms since they are also localized to glial cells. The activation of α1-ARs has been found to increase calcium transients in hippocampal astrocytes and Bergmann glial (Kulik et al., 1999).
Although the localization of the different α1-AR subtypes in the brain has been performed, the results are not definitive. Binding or autoradiography studies using [125I]HEAT or [3H]WB4101 have been reported but these studies indicate that essentially all areas of the brain seem to have α1-AR receptors (Unnerstall, 1987). The distribution of α1-AR subtypes in the brain has also been determined by hybridization of mRNAs (Domyancic and Morilak, 1997). The three α1-AR subtypes have differential localizations in the brain. These studies are considered to be more accurate but are not quantitative and mRNAs may be transported. To date, no reliable immunohistochemistry experiments have been reported due to the lack of highly specific antibodies to the receptors. Therefore, although we know of general areas in the CNS that α1-ARs are likely to be expressed, localization of the specific subtypes is not definitive.
α1-ARs and Locomotion.
Although CNS control of locomotion is thought to be primarily due to dopaminergic-striatal pathways, recent data also suggest the involvement of the α1-ARs. The α1A-AR subtype has been found to be predominant in rat spinal motoneuron activity (Wada et al., 1997). Patch clamp recordings on the rat substantia nigra reticulata indicate that phenylephrine increased whereas prazosin blocked the spontaneous firing of the reticulata cells (Berretta et al., 2000). This increase in excitability may have a significant impact on the final output of the basal ganglia and motor-related behaviors. There also appears to be an inverse relationship between dopamine receptors and the α1-AR in the brain. Electrophysiological studies have shown that prazosin administered systemically can decrease the burst firing and can regularize the firing pattern of dopaminergic neurons located in the ventral tegmental area (Grenhoff and Svensson, 1993).
We have recently described a neurodegenerative phenotype in a transgenic mouse model that systemically overexpresses the α1B-AR (Zuscik et al., 2000). These mice display symptoms and degenerative histology that share similarities with a Parkinsonian degenerative disease called multiple system atrophy (MSA). The main features of MSA include Parkinsonism, autonomic failure, and cerebellar ataxia with corresponding cerebellar degeneration. The mice also display a grand mal seizure disorder that appears to be a multifocal epilepsy. Seizures are not typically seen in MSA but this may be a mouse manifestation of the disease as the α1B-AR is more extensively expressed in the mouse cortex than it is in humans. This model may provide clues to the roles of the α1B-AR subtype in neurotransmission, locomotion behavior, and the mechanism of some neurodegenerative diseases.
The α1-ARs and the Regulation of Smooth Muscle Contraction
Early attempts to characterize the role for the three α1-AR subtypes in smooth muscle contraction was made confusing by the reliance on chloroethylclonidine as a diagnostic reagent (reviewed in Guarino et al., 1996). It is now clear that chloroethylclonidine can inactivate, to one degree or another all the α1-ARs (Xiao and Jeffries, 1998). Therefore, data relying on this ligand are probably not valid. A better understanding of the precise regulatory functions of the α1-ARs has also been impaired by a dearth of highly selective reagents that allow an adequate characterization of the contribution of each subtype. Therefore, most data cannot exclude participation of the other subtypes.
Previous work has shown that all multiple α1-AR mRNA and receptor proteins are expressed on peripheral arteries from rats (Piascik et al., 1995, 1997; Guarino et al., 1996) and humans (Rudner et al., 1999). The next fundamental question is the extent to which each of the α1-AR contributes to the contraction of vascular smooth muscle. Based on a series of studies using multiple techniques, we proposed that despite expression of multiple α1-ARs on vascular smooth muscle, a single receptor is responsible for mediating the contraction of the blood vessel and that the dominant contractile α1-AR is different in different vascular beds (Piascik et al., 1995, 1997; Hrometz et al., 1999). This hypothesis is supported by numerous studies from different laboratories using a variety of different experimental preparations, which consistently demonstrate that in a given artery one α1-AR is responsible for mediating agonist-induced increases in contractile function. For example the α1A-AR has been shown to mediate the contraction of the renal and caudal arteries (Lachnit et al., 1997; Piascik et al., 1997; Hrometz et al., 1999). The α1D-AR has been shown to regulate the contraction of the aorta, femoral, iliac, and the superior mesenteric artery (Piascik et al., 1997; Hrometz et al., 1999). There is little direct evidence for a role of the α1B-AR as a mediator of contractile function in blood vessels.
These data from isolated vessels show that there is diversity in the vascular tree regarding the α1-AR, which modulates vascular smooth muscle contraction. Although these data increase our understanding of the regulation of individual vascular beds by the α1-ARs, the contribution of each subtype to the integrated control of systemic arterial blood pressure cannot be ascertained from these studies. To address this issue, several laboratories have genetically engineered lines of transgenic mice. Cotecchia's laboratory reported that deletion of the α1B-AR in mice had no effect on resting arterial blood pressure (Cavalli et al., 1997). Also, the pressor response to phenylephrine was only modestly affected in α1B-AR knockout mice. As studies with isolated blood vessels failed to observe a significant role in the contraction of isolated blood vessels, it is not surprising to note little effect of α1B-AR gene deletion on systemic arterial blood pressure. As an alternative approach, we examined the regulation of systemic arterial blood pressure in mice systemically overexpressing the wild-type and constitutively active α1B-AR mutations (Zuscik et al., 2001). Overexpression of the α1B-AR did not result in an increase in blood pressure. This also argues that the α1B-AR does not mediate increases in systemic arterial blood pressure. Indeed we noted that basal blood pressure was reduced in transgenic animals. The mechanism of the hypotension is likely due to autonomic failure, which is commonly seen in the MSA phenotype previously described in these mice (see α1-ARs in Neurological Function). The mechanism is also not likely because of changes in vasodilatation since ex vivo preparations of the mesenteric arteries (1st order) from both the overexpressing as well as the knockout mouse models failed to show any changes in contractile function. All in all, these studies indicate that the α1B-AR is not a significant player in vasoconstriction. Simpson's laboratory examined blood pressure regulation in α1A-AR knockout mice (Rokosh et al., 2000). These authors noted a decrease in mean arterial blood pressure and a decrease in the pressor response to phenylephrine. These results are in agreement with data from in vitro studies that indicate a role for the α1A-AR in the regulation of vascular smooth muscle contraction. Knockout of the α1D-AR indicated an impaired vasopressor response without a change in basal blood pressure (G. Tsujimoto, personal communication). This result suggests that the α1D-AR is also involved in vascular smooth muscle contraction but perhaps not as dominant as the α1A-AR.
The α1-ARs and the Regulation of Growth Promoting Responses
The α1-ARs are known regulators of hypertrophic growth responses. These receptors are linked to cellular growth responses by the MAPKs. A large volume of data has shown that the α1A-AR is the primary mediator of hypertrophic responses in neonatal cardiomyocytes, and these data have been recently reviewed in detail (Varma and Deng, 2000). The role of the other α1-ARs has been less well studied. The α1D-AR has been shown to promote ERK activity and hypertrophic growth in rat aortic smooth muscle cells (Xin et al., 1997).
The strongest evidence for a role of the α1B-AR in mediating hypertrophic growth has been obtained in studies with transgenic mice. Mice overexpressing a cardiac-targeted, constitutively active α1B-AR exhibit significant cardiac hypertrophy (Milano et al., 1994). Cardiac-specific overexpression of the wild-type α1B-AR leads to contractile dysfunction without hypertrophy (Grupp et al., 1998). Similarly, Akhter and coworkers (1997) found that these mice also exhibited decreased contractile responsiveness to isoproterenol and decreased cardiac adenylyl cyclase activity. It is somewhat surprising that overexpression of the wild-type α1B-AR did not produce hypertrophy. Regardless, it appears that tonic unregulated activation of the α1B-AR can lead to contractile dysfunction and/or cardiac hypertrophy. We also initiated studies with our transgenic mice that systemically overexpresses the wild-type or a constitutively active α1B-AR. Our mice exhibited enlarged hearts and had echocardiographic evidence of hypertrophy characterized by an increased thickness of the interventricular septum and posterior wall and an increased isovolumetric relaxation time (Zuscik et al., 2001). This suggests that developmentally regulated overexpression of the wild-type α1B-AR can induce cardiac pathophysiology at much lower levels of receptor expression than that required for the myosin heavy chain promoter. We also have evidence that in the hearts of these transgenics, there is an increase in the activity of the MAPKs, ERK, and JNK indicating a role for these kinases in mediating hypertrophy (D. F. McCune, D. Chalathorn, M. L. Garcia, S. E. Edelmann, D. M. Perez, and M. T. Piascik, manuscript in preparation). We have also used high-density oligonucleotide arrays to determine gene expression profiles associated with cardiac hypertrophy from an α1B-AR-induced etiology (Zuscik et al., 2001). We find that the growth responses tend to be dominated by Src-related receptors and signaling pathways. There is also an inflammatory/autoimmune component to the hypertrophy, a finding that is commonly associated in some heart failure models but not typically found in hypertrophy.
In summary, our knowledge of the binding pocket and specific functions linked to a particular α1-AR subtype have increased greatly within the last few years. However, there is still much to be deciphered, which will be possible with the manufacturing of ligands with greater selectivity and further in-depth analysis of the transgenic and knockout models. Of particular importance is the role of the three α1-AR subtypes outside of the cardiovascular system.
Footnotes
-
Original work of the authors cited here was supported by National Institutes of Health Grants HL38120 (to M.T.P.) and HL61438 (to D.M.P.), the Southeast Affiliate of the American Heart Association (to M.T.P.), an American Heart Established Investigator Award (to D.M.P.), and an unrestricted research grant from Glaxo Wellcome (to D.M.P.).
- Abbreviations:
- GPCR
- G-protein-coupled receptor
- AR
- adrenergic receptor
- TM
- transmembrane
- CNS
- central nervous system
- [125I]HEAT
- (±)-β-([125I]iodo-4-hydroxyphenyl)-ethyl-aminomethyl-tetralone
- GFP
- green fusion protein
- MSA
- multiple system atrophy
- MAPK
- mitogen-activated protein kinase
- ERK
- extracellular signal-regulated kinase
- JNK
- c-Jun N-terminal kinases
- Received January 31, 2001.
- Accepted March 14, 2001.
- The American Society for Pharmacology and Experimental Therapeutics