Associate editor: D.R. Sibleyα1-Adrenergic receptor regulation: basic science and clinical implications
Introduction
Drugs targeting adrenergic receptors (ARs) are some of the most widely utilized therapeutic agents in clinical medicine. Activation of ARs provides enhanced inotropy, chronotropy, bronchodilation, vasconstriction, sedation, and analgesia. Inhibition of ARs results in vasodilation, decreased heart rate, inotropy, and relaxation of prostate smooth muscle. Each of these actions is important in treating clinical diseases, such as congestive heart failure, angina, hypertension, benign prostatic hyperplasia (BPH), acute/chronic pain, anesthetic responses, and asthma (reviewed in Smiley et al., 1998). ARs were first divided into α and β by Ahlquist in his landmark manuscript in the Journal of Physiology in 1948. By 1967, Lands had subdivided βARs into β1 and β2 subtypes, and by the mid-1970s, αARs had been subdivided into α1 and α2. These four subtypes are often referred to as “classic” AR subtypes (Schwinn et al., 1991). By the late 1980s and early 1990s, at least 9 AR subtypes (3 α1ARs, α1a, α1b, α1d; 3 α2ARs, α2a, α2b, α2c; and 3 βARs, β1, β2, β3) had been discovered, cloned, expressed stably in cells, and characterized pharmacologically (Fig. 1) (Mizobe et al., 1996, Schwinn et al., 1991a, Schwinn et al., 1991b; for reviews, see Graham et al., 1996, Guarino et al., 1996, Piascik et al., 1996).
ARs can be described by their pharmacology (Fig. 2) and physiology (Fig. 3). In terms of pharmacology, it is important to note that the endogenous catecholamines norepinephrine (NE) and epinephrine are agonists for all AR subtypes, although with varying affinity. βAR subtypes have different agonist potency series, enabling discrimination of subtypes. αARs have an identical agonist potency series; hence, it was not until selective antagonists (prazosin for α1ARs and idazoxan/yohimbine for α2ARs) were discovered that these two subtypes could be discriminated pharmacologically. βARs generally couple via Gs to stimulate adenylyl cyclase, with resultant increases in cyclic AMP (cAMP) in the cell. βARs also couple to K+ channels, possibly via Go, and the β2AR has been shown to be capable of coupling to Gi in cell culture and animal models Kompa et al., 1999, Xiao et al., 1999. Of note, β3ARs couple via Gs in fat cells, but couple via Gi in human heart (Gauthier et al., 1996). α2ARs, when located presynaptically (autoreceptors), inhibit NE release, and when located postsynaptically, mediate vaso- and venoconstriction. α2ARs are postsynaptic on platelets, the spinal cord, and in multiple sites in the CNS. α2ARs couple to Gi, which inhibits the enzyme adenylyl cyclase, causing decreases in cAMP in the cell (Smiley et al., 1998). α1ARs couple predominantly through Gq, resulting in hydrolysis of membrane phospholipids via phospholipase (PL)Cβ, to yield the second messengers inositol triphosphate (IP3) and diacylglycerol, leading to muscle contraction through mobilization of intracellular Ca2+ (Graham et al., 1996) and activation of protein kinase C (PKC). In addition to coupling to Gq, α1ARs have also been reported to activate pertussis-sensitive G-proteins. In myocytes, this leads to increased inotropy, [Na+-K+]- ATPase activation, modulation of intracellular Ca2+ levels, and cell shortening Mizobe et al., 1996, Steinberg et al., 1985, Terzic et al., 1993. Stimulation of α1ARs has been implicated in the pathogenesis of myocardial hypertrophy Simpson et al., 1982, Simpson & Savion, 1982 and BPH Hieble & Ruffolo, 1996, Schwinn & Price, 1999. Finally, recent evidence suggests that another G-protein, Gh [also known as transglutaminase type II (TGII)], has been shown to mediate α1AR stimulation of PLCδ1, increasing inositol phosphate turnover in TGII-transfected cells Chen et al., 1996, Feng et al., 1999, Feng et al., 1996, Park et al., 1998. TGII is a unique bifunctional protein that can (1) act as a transglutaminase that catalyzes Ca2+-dependent post-translational modification of proteins through formation of isopeptide bonds between glutamine and lysine residues (Folk, 1980) and (2) bind guanine nucleotides in a 1:1 ratio and hydrolyze GTP (Lee et al., 1989). Its role as a G-protein remains controversial, however.
ARs are members of the much larger family of G-protein-coupled receptors (GPCRs), which include, for example, muscarinic cholinergic receptors, serotonin receptors, dopamine receptors, neurokinin receptors, as well as the photoreceptor rhodopsin. Overall, membrane topology of ARs (and GPCRs) includes seven hydrophobic α-helical membranes called transmembrane regions (TM1–7), an extracellular amino-terminus, three extracellular loops, an intracellular carboxyl (COOH)-terminus, three main intracellular loops (IC1–3), plus a fourth IC loop created by palmitoylation of a COOH-terminal cysteine residue Mizobe et al., 1996, Schwinn et al., 1991a, Schwinn et al., 1991b. TM amino acid (aa) identity between AR subtype families (e.g., α vs. β) is ≈45%; this rises to 75% between subtypes within a family (e.g., α1a vs. α1b). Species homologues have even higher TM aa homology, often >90%, although key mutations can have functional consequences (e.g., human and rat α2a have different pharmacology due to one critical aa difference). Genomic organization is conserved within the different subtypes, suggesting that they evolved through gene duplication events Graham et al., 1996, Guarino et al., 1996.
Knockout and transgenic mice have provided researchers with valuable information on the effect of a single gene when deleted or overexpressed, respectively. However, a few caveats need to be kept in mind when discussing results from knockout/transgenic animal experiments. First, since other genes may compensate for the targeted gene, the resultant knockout phenotype may not represent the physiology of the gene of interest. Sometimes, no change in phenotype is seen in transgenic animals by knocking out individual genes due to the functional redundancy of closely related family members. Second, any variance in genetic background of animals used for transgenic/knockout experiments can significantly confound interpretation of the resulting phenotypes. Third, in potentially lethal deletions, surviving animals may represent a phenotype (not directly related to the knockout) that facilitated their survival. In spite of these drawbacks, genetically engineered mice have proved to be powerful tools, occasionally resulting in identification of previously unrecognized functions of specific gene products. Transgenic and knockout mice have been made for most AR subtypes Rohrer et al., 1998, Rohrer & Kobilka, 1998. Results from these studies are summarized in Table 1.
The cDNAs encoding three α1AR subtypes (α1a, α1b, and α1d) have been cloned and pharmacologically characterized Cotecchia et al., 1988, Lomasney et al., 1991, Perez et al., 1991, Schwinn et al., 1995, Schwinn et al., 1990. [Of note, α1AR nomenclature has changed within the last 5 years, with the International Union of Pharmacology adopting α1a, α1b, and α1d in 1995 (Hieble et al., 1995); the α1a used to be called α1c, the α1b name has not changed, and the α1d used to be called α1a, α1d, or α1a/d.] Although a fourth α1AR subtype (termed α1LAR, due to its comparatively lower affinity for the α1AR antagonist prazosin) has been described, recent evidence suggests that this subtype represents the low-affinity state of the α1aAR, and not a distinct receptor (Ford et al., 1997). At first glance, the physiologic role for individual AR subtypes appears to have been eloquently dissected using transgenic/knockout mice (Table 1). However, in the process of cloning cDNAs encoding each of the nine AR subtypes, tissue distribution studies in several animal species resulted in the surprising finding that tissue distribution of specific AR subtypes is species dependent. One of the first sets of experiments that noted this was performed in our laboratory, and they demonstrated that the α1AR subtype present in the liver depends on the species examined (the α1bAR is the only α1AR subtype in rat liver, while the α1aAR predominates in human liver; Price et al., 1993, Price et al., 1994a, Price et al., 1994b). Our laboratory extended these studies to demonstrate extensive species differences in all AR subfamilies (α1, α2, βARs), with αARs>βARs Berkowitz et al., 1994, Berkowitz et al., 1995, Price et al., 1993, Price et al., 1994a, Price et al., 1994b. Table 2 gives examples of α1AR subtype expression in a few rat and human tissues Malloy et al., 1998, Price et al., 1993, Price et al., 1994a, Price et al., 1994b. While many tissue-expression studies initially utilized RNA approaches (RNase protection assays, in situ hybridization), most of these findings have since been confirmed at a protein level (ligand binding and contraction studies), with generally good correlation between RNA and protein Malloy et al., 1998, Rudner et al., 1999.
In many tissues, quantification and localization of α1ARs has been difficult to accurately determine, due to limitations in reagents and the high degree of conservation of the α1AR subtypes. Recent methodologic advances seem to be providing more efficient and accurate means for characterizing receptor expression and pharmacology. Daly et al. (1998) recently described the use of a fluorescent quinazoline antagonist with a high affinity for α1ARs in a simplified model system of rat-1 fibroblasts stably expressing the α1dAR. This hopefully will lead to subtype-specific markers with distinct fluorescent tags to enable discrimination of α1AR subtypes within the same cell population. Subtype-specific antibodies are also being used to quantify cell-surface receptor expression, particularly in the cardiovascular system, where α1ARs are known to mediate many sympathetic responses, such as smooth muscle contraction Hrometz et al., 1999, Villalobos-Molina & Ibarra, 1996, cardiac contractility (Graham et al., 1996), cardiac hypertrophy Simpson et al., 1982, Simpson & Savion, 1982, and hypertension (Veelken & Schmieder, 1996). Hrometz et al. (1999) utilized an antibody approach to examine the correlation between function and α1AR subtype distribution in rat vasculature. Results showed that although all three subtypes were expressed in rat vasculature, it appears that only α1dAR is able to mediate smooth muscle contraction in the femoral artery, whereas α1aAR mediates renal artery contraction. It is important to note that results generated with α1aAR- and α1dAR-selective antibodies in cultured smooth muscle cells (SMCs) were not as clear as the α1bAR results, and further work is necessary to provide unequivocal data using antibody-based approaches. In a human model system, Ricci et al. (1999) used a combination of subtype-selective antibodies and reverse transcriptase-polymerase chain reaction (RT-PCR) (to analyze message levels) to demonstrate that all three subtypes are expressed in peripheral blood lymphocytes, with α1bAR levels predominating. As new reagents become more specific and reliable, we will be able to better ascertain the correlation between α1AR subtype expression and functionality in a given tissue. At a minimum, however, these studies should provide valuable insight into how α1AR subtype expression may be linked to pathologies and, thus, may provide useful markers for cardiovascular disorders.
Given the fact that catecholamine levels increase with age and the putative involvement of α1ARs in age-related pathologies, such as myocardial hypertrophy and BPH, a number of studies have focused on α1AR expression and signaling with age. Our laboratory recently examined human vascular α1AR subtype expression with age, and demonstrated artery-specific increases in α1AR density (Rudner et al., 1999). Xu et al. (1997) recently showed an age-dependent decrease in mRNA for all α1AR subtypes in rat aorta, but no change in mesenteric or pulmonary arteries. Thus, α1AR mRNA levels in the vasculature appear to vary with age in a vessel- and species-specific manner. In the heart, studies suggest that α1AR signaling decreases with age (del Balzo et al., 1990). Thus, a number of groups have sought to determine the molecular basis for signal dampening. Gascon et al. (1993) have shown through ligand-binding studies with the α1AR antagonist prazosin an overall reduction of α1ARs in the rat heart with age, although the equilibrium between high- and low-affinity sites is maintained. Miller et al. (1996) subsequently examined steady-state α1AR mRNA levels in the four chambers of the rat heart, and reported no decrease in α1AR mRNA levels with age, suggesting that dampening of α1AR signaling is post-transcriptional. This is in contrast to a previous study noting decreases in both α1AR mRNA and protein levels with age in the heart (Kimball et al., 1991). Taken together, the above studies suggest that age-associated dampening of α1AR signaling in the heart is mediated by nontranscriptional mechanisms, such as receptor synthesis or degradation, internalization, or effector coupling. Conversely, Simpson and colleagues demonstrated a significant increase in all three α1AR subtypes in adult (10 month) isolated rat cardiomyocytes versus 1- to 2-day-old neonatal isolated rat cardiomyocytes (Stewart et al., 1994). In this same study, cardiac fibroblasts were found to be devoid of expression of all three subtypes, demonstrating potentially important developmental as well as tissue-specific expression of the α1ARs in the heart. The potential ramifications of developmental increases in α1AR cardiomyocyte expression levels remain to be explored.
The direct role of α1ARs in mediating cardiac hypertrophy has been elegantly demonstrated in transgenic mouse models (the reader is referred to a recent review of GPCRs in the heart (Brodde & Michel, 1999). Lefkowitz and colleagues introduced a constitutively active mutant (CAM) of α1bAR into mouse hearts, resulting in myocardial hypertrophy in the absence of an increase in blood pressure (Milano et al., 1994). This provided conclusive evidence that α1AR stimulation can result in myocardial hypertrophy in the absence of increased afterload. However, it is important to note that transgenic hearts expressing the wild-type α1bAR at levels 40-fold higher than normal did not develop hypertrophy, despite enhanced Gq signaling [measured by an 8-fold increase in atrial natriuretic factor (ANF) ventricular mRNA] (Akhter et al., 1997; reviewed in Dorn & Brown, 1999). Probing potential mechanisms for altered myocardial function revealed that dual coupling of the α1bAR exists, as pertussis-toxin-sensitive G-protein-mediated attenuation of adenylyl cyclase was seen in myocardial membranes purified from wild-type α1bAR-expressing animals (Akhter et al., 1997). Additionally, it is not clear if the effect by CAM is mediated directly by α1bARs or indirectly by PLC stimulation, leading to, among other things, induction of α1aAR in the heart. Consistent with the latter hypothesis, responsiveness of cardiac cells to α1aAR stimulation persists, whereas desensitization and down-regulation of most other GPCRs is observed (including the α1b and α1dAR subtypes). This is an intriguing example of differential regulation between receptor family subtypes, with important implications regarding the role of α1aARs in chronically stimulated muscular hypertrophy. Both the Simpson and Woodcock laboratories showed that rat neonatal cardiomyocytes chronically stimulated with either NE or the α1aAR-selective agonist A-61603 resulted in induction of α1aAR message levels, with concurrent α1bAR and α1dAR down-regulation Autelitano & Woodcock, 1998, Rokosh et al., 1996. Taken together with studies performed in our laboratory showing that α1aAR predominates in the human heart, whereas α1aAR and α1bAR subtypes predominate in the rat heart Price et al., 1994a, Price et al., 1994b, these data suggest a key role for α1ARs (particularly α1aARs) in mediating myocardial hypertrophy. Great care must be taken, however, when applying results from different species to humans, as receptor-expression levels are species-dependent, with rat cardiac α1-adrenoceptors roughly 8 times the level in human heart (reviewed in Brodde & Michel, 1999).
Section snippets
Ligand binding
α1ARs, like other members of the GPCR protein family, are single polypeptide chains, ranging from 355 to 511 as that fold into a highly conserved structure of 7 predicted membrane-spanning regions, TM1–7. The seven TM regions fold into a conformation that creates a hydrophilic ligand-binding pocket surrounded by a hydrophobic core for the physiologic agonists epinephrine and NE (reviewed in Piascik et al., 1996). Although there is little aa homology among the different GPCRs, the predicted
Regulation of receptor signaling
While the precise signaling mechanisms through which α1AR-stimulated transcriptional activation occurs are not known, multiple pathways have been implicated. Recent experiments involving transgenic animals overexpressing a Gq inhibitor (thus, a functional Gq knockout) demonstrated a 60–70% reduction in myocardial hypertrophy, suggesting that 30–40% of hypertrophic pathways are PLC/PKC independent (Akhter et al., 1998). From recent studies aimed at elucidating the subcellular mechanisms of α1AR
Clinical aspects of α1-adrenergic receptor pharmacology
α1ARs have wide-ranging physiologic roles in health and disease (Table 4). Although present in diverse organs and tissues throughout the body, α1ARs are of particular importance in the cardiovascular system. In the myocardium, α1ARs mediate myocardial inotropy and hypertrophy, as well as play a role in atrial/ventricular arrhythmias and ischemic preconditioning Anyukhovsky et al., 1994, Anyukhovsky et al., 1997, Li et al., 1997, Tomai et al., 1999. Since α1ARs mediate smooth muscle contraction,
Summary
Recent advances in our understanding of the molecular mechanisms governing α1AR function have been described in this review. A tremendous acceleration in the amount of information regarding the structure and function of α1AR has been realized and will provide the framework for future drug design and potential gene therapy approaches. Elucidation of the precise structure of the ligand-binding pocket for each α1AR subtype will aid in the development and characterization of α1AR subtype-specific
Acknowledgements
The authors would like to recognize support from the following National Institutes of Health grants: AG-00745, AG-13853, HL-49103, and GCRC759 (to D.A.S.).
References (242)
- et al.
Transgenic mice with cardiac overexpression of alpha1B-adrenergic receptors. In vivo alpha1-adrenergic receptor-mediated regulation of beta-adrenergic signaling
J Biol Chem
(1997) - et al.
Acute mesenteric ischemia after cardiopulmonary bypass
J Vasc Surg
(1992) - et al.
Selective activation of alpha1A-adrenergic receptors in neonatal cardiac myocytes is sufficient to cause hypertrophy and differential regulation of alpha1-adrenergic receptor subtype mRNAs
J Mol Cell Cardiol
(1998) - et al.
Differential regulation of the phosphatidylinositol 3-kinase/Akt and p70 S6 kinase pathways by the alpha(1A)-adrenergic receptor in rat-1 fibroblasts
J Biol Chem
(2000) - et al.
The distribution of β3-adrenergic receptor mRNA in human tissue
Eur Pharmacol
(1995) Beta-adrenoceptors in cardiac disease
Pharmacol Ther
(1993)- et al.
α1-Adrenoceptor-induced contractility in rat aorta is mediated by the α1D subtype
Eur J Pharmacol
(1996) Alpha-adrenergic blockers for the treatment of benign prostatic hyperplasia
Urol Clin North Am
(1990)- et al.
Molecular cloning, genomic characterization and expression of novel human alpha1A-adrenoceptor isoforms
FEBS Lett
(1998) - et al.
Sp1-mediated transcriptional activation from the dominant promoter of the rat alpha1B adrenergic receptor gene in DDT1MF-2 cells
J Biol Chem
(1997)