help button home button Endocrine Society Endocrine Reviews
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS

First published online on February 16, 2006
Endocrine Reviews, doi:10.1210/er.2005-0034
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
27/3/260    most recent
Final Manuscript
Author Manuscript
Right arrow Purchase Article
Right arrow View Shopping Cart
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Copyright Permission
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Hillhouse, E. W.
Right arrow Articles by Grammatopoulos, D. K.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Hillhouse, E. W.
Right arrow Articles by Grammatopoulos, D. K.
Right arrowPubmed/NCBI databases
*Substance via MeSH
Endocrine Reviews 27 (3): 260-286
Copyright © 2006 by The Endocrine Society

The Molecular Mechanisms Underlying the Regulation of the Biological Activity of Corticotropin-Releasing Hormone Receptors: Implications for Physiology and Pathophysiology

Edward W. Hillhouse and Dimitris K. Grammatopoulos

The Leeds Institute of Genetics, Health and Therapeutics (E.W.H.), The University of Leeds, Leeds LS2 9NL, United Kingdom; and Department of Endocrinology and Metabolism (D.K.G.), Warwick Medical School, The University of Warwick, Coventry CV4 7AL, United Kingdom

Correspondence: Address all correspondence and requests for reprints to: Edward W. Hillhouse, The Leeds Institute of Genetics, Health and Therapeutics, The University of Leeds, Leeds LS2 9NL, United Kingdom. E-mail: e.w.hillhouse{at}leeds.ac.uk; or Dimitris K. Grammatopoulos, Endocrinology and Metabolism, Warwick Medical School, The University of Warwick, Coventry CV4 7AL, United Kingdom. E-mail: d.grammatopoulos{at}warwick.ac.uk.


    Abstract
 Top
 Abstract
 I. Introduction
 II. CRH-R Agonists: the...
 III. CRH-R Subfamilies
 IV. Agonist-CRH-R Interaction
 V. CRH-R Signaling...
 VI. Regulation of CRH-R...
 VII. Conclusions
 References
 
The CRH receptor (CRH-R) is a member of the secretin family of G protein-coupled receptors. Wide expression of CRH-Rs in the central nervous system and periphery ensures that their cognate agonists, the family of CRH-like peptides, are capable of exerting a wide spectrum of actions that underpin their critical role in integrating the stress response and coordinating the activity of fundamental physiological functions, such as the regulation of the cardiovascular system, energy balance, and homeostasis. Two types of mammal CRH-R exist, CRH-R1 and CRH-R2, each with unique splicing patterns and remarkably distinct pharmacological properties, but similar signaling properties, probably reflecting their distinct and sometimes contrasting biological functions. The regulation of CRH-R expression and activity is not fully elucidated, and we only now begin to fully understand the impact on mammalian pathophysiology. The focus of this review is the current and evolving understanding of the molecular mechanisms controlling CRH-R biological activity and functional flexibility. This shows notable tissue-specific characteristics, highlighted by their ability to couple to distinct G proteins and activate tissue-specific signaling cascades. The type of activating agonist, receptor, and target cell appears to play a major role in determining the overall signaling and biological responses in health and disease.

I. Introduction
II. CRH-R Agonists: the Family of CRH-Related Peptides
III. CRH-R Subfamilies
A. Distribution of CRH-R
B. Splicing pattern and genomic organization

IV. Agonist-CRH-R Interaction
A. Receptor pharmacology
B. Receptor-agonist interaction: the role of the N- and J-domains
C. Receptor-agonist high-affinity interaction: the role of posttranslational modifications and G proteins
D. Receptor-G protein interactions: implications for signaling

V. CRH-R Signaling Characteristics
A. Activation of the cAMP/PKA signaling pathway
B. cAMP-independent signaling pathways: CRH-Rs and activation of MAPKs
C. CRH-Rs and the regulation of the NO/cGMP signaling pathway

VI. Regulation of CRH-R Functional Activity
VII. Conclusions


    I. Introduction
 Top
 Abstract
 I. Introduction
 II. CRH-R Agonists: the...
 III. CRH-R Subfamilies
 IV. Agonist-CRH-R Interaction
 V. CRH-R Signaling...
 VI. Regulation of CRH-R...
 VII. Conclusions
 References
 
STRESS IS AN ancient phenomenon that can be traced back through evolution. The concept of stress, however, is modern, and describes a complex set of adaptive physiological responses to external demand. Survival is therefore dependent upon an adequate response to stressful stimuli. This process involves activation of a number of different but integrated physiological responses involving the autonomic, endocrine, immune, cardiovascular, and reproductive systems, which induce a spectrum of behavioral and homeostatic changes. During evolution, mammals have evolved remarkably similar molecular signals that orchestrate the integrated stress response. The conservation of these molecules is a reflection of their role in survival and adaptation. In the mammalian adaptive response to stress, the hypothalamo-pituitary-adrenal (HPA) axis plays a central role, predominantly via the hypothalamic hormone, CRH or corticotropin-releasing factor (CRF), which regulates the secretion of adrenocorticotropin from the anterior pituitary. In addition, CRH exerts a wide spectrum of actions in the central nervous system (CNS) and the periphery that underpin its critical role in integrating and coordinating the activity of diverse physiological systems. Interestingly, this complex process of stress adaptation is fine-tuned by several CRH-related peptides, namely the urocortins (UCNs) that exert complementary or sometimes contrasting actions.

CRH and the UCNs are expressed in multiple sites in the periphery, where they influence physiological mechanisms via paracrine or autocrine activation of specific receptors expressed on the cell membrane of target cells. These CRH receptors (CRH-Rs) and the mechanisms regulating their activity play a crucial role in mediating the biological effects of the CRH family of peptides. This review will discuss current knowledge relating to the structure, biological function, and molecular mechanisms regulating CRH-R expression and activity.


    II. CRH-R Agonists: the Family of CRH-Related Peptides
 Top
 Abstract
 I. Introduction
 II. CRH-R Agonists: the...
 III. CRH-R Subfamilies
 IV. Agonist-CRH-R Interaction
 V. CRH-R Signaling...
 VI. Regulation of CRH-R...
 VII. Conclusions
 References
 
The first mammalian CRH peptide was isolated from ovine hypothalamic extracts in 1981 (1). This was followed by the discovery of a novel family of mammalian CRH-related ligands, which at present contains three members: UCN-I, UCN-II (or stresscopin-related peptide), and UCN-III (or stresscopin) (2, 3, 4, 5). These peptides appear to stem from an ancestral peptide precursor (6).

CRH, a 41-amino acid peptide, is the principal regulator of the basal and stress-induced pituitary-adrenal axis that activates glucocorticoid and adrenal androgen secretion (7, 8). It displays anxiogenic properties and coordinates the adaptive, behavioral, and physical changes that occur during stress; this has been conclusively demonstrated in transgenic or knockout mouse models (9, 10). Some of these actions of CRH may be mediated via the activation of norepinephrine-secreting neurons and other neurotransmitter systems important for mood regulation (11, 12, 13, 14). CRH also influences appetite via its anorexic properties (15, 16). In addition, CRH is synthesized and produced in multiple peripheral tissues and might be involved in many other biological functions, such as energy balance, metabolism, and regulation of the immune response (17, 18). Interestingly CRH also plays a role in mammalian reproduction and embryo implantation (19). It is synthesized and secreted by the human placenta and might act as a "placental clock" that regulates the onset of human labor, perhaps by modulating myometrial contractility (20, 21). Increased or chronic secretion of CRH leads to anxiety, sleep pattern disturbance, and changes in the cardiovascular, metabolic, and immune functions. CRH has been implicated in the pathophysiology of various disorders; in particular, there is good evidence linking CRH system abnormalities to chronic anxiety disorder, melancholic and atypical depression, chronic pain and fatigue states, sleep disorders, addictive behavior, neurodegeneration, allergic and autoimmune inflammatory disorders, the metabolic syndrome, and gastrointestinal diseases (22, 23, 24, 25, 26, 27).

The UCNs exhibit various degrees of amino acid sequence homology to CRH: UCN-II shows moderate identity with human/rat CRH (h/rCRH) (34%), human UCN-I (43%), and UCN-III (37–40%), whereas UCN-III displays less homology to other members of the CRH family (18–32% identity). They are also expressed in the CNS and peripheral tissues; however, their actions are less well characterized. In the brain, UCN-I expression is most prominent in the Edinger-Westphal nucleus and the lateral superior olive. There is very little neuroanatomical overlap in the distribution of UCN-I and CRH in the brain, suggesting differential functional roles for these peptides (28). UCN-I mRNA or immunoreactivity has also been reported in other brain regions including the cerebellum, hippocampus, neocortex, olfactory system, basal ganglia, amygdala, and the supraoptic, ventromedial (VMH), and paraventricular (PVN) nuclei of the hypothalamus, as well as the laterodorsal tegmental nucleus, the dorsal raphe, the periaqueductal gray, the substantia nigra pars compacta, and ventral tegmental nucleus. In the latter two regions, UCN-I immunoreactivity is located almost exclusively in neurons that coexpress tyrosine hydroxylase, suggesting that within these regions, UCN-I is colocalized with dopamine. Interestingly, several brain regions such as the ventromedial hypothalamus and the amygdala (tissues rich in CRH-R2 receptors that exhibit significant selectivity for the UCNs; see Section III.A) do not contain UCN-I immunoreactive fibers. UCN-I expression has also been demonstrated in peripheral tissues such as the heart, adrenal, skeletal muscle, placenta, skin, immune system, and the gastrointestinal tract (29).

A similar pattern of distribution to that of UCN-I was found for UCN-II mRNA in the mouse and rat CNS (29). UCN-II mRNA is highly expressed in the paraventricular, supraoptic, and arcuate nuclei of the hypothalamus, the locus coeruleus, and motor nuclei of the brain stem and spinal cord. The posterior part of the bed nucleus of the stria terminalis, the lateral septum, and the medial amygdaloid nucleus are important brain sites that express UCN-II mRNA. Histologically, UCN-II mRNA and UCN-II-like immunoreactivity were demonstrated in both the anterior and intermediate lobes of the pituitary, but not detected in the posterior lobe. In the periphery, high levels of UCN-II mRNA have been detected in the heart, adrenal gland, placenta, stomach, skin, ovary, gastrointestinal tract, uterus, uterine smooth and skeletal muscle, and peripheral blood cells (29).

UCN-III displays a distinct distribution from that of CRH, UCN-I, and UCN-II (30). UCN-III-positive neurons were found predominately within the hypothalamus and medial amygdala. In the hypothalamus, UCN-III neurons were observed in the median preoptic nucleus and in the rostral perifornical area lateral to the PVN. The UCN-III fibers were distributed mainly in the hypothalamus and limbic structures. Hypothalamic regions that were innervated prominently by UCN-III fibers included the ventromedial nucleus, medial preoptic nucleus, and ventral premammillary nucleus. Outside the hypothalamus, the densest projections were found in the intermediate part of the lateral septum, posterior division of the bed nucleus stria terminalis, and the medial nucleus of the amygdala. Several major UCN-III terminal fields have been identified, including the lateral septum and the ventromedial hypothalamus, which are known to express high levels of CRH-R2. Thus, these anatomical data strongly support the notion that UCN-III is an endogenous ligand for CRH-R2 in these areas. These results also suggest that UCN-III is likely to mediate physiological functions, including food intake and neuroendocrine regulation (30). In the periphery, immunoreactive UCN-III is expressed in the human adrenals (31), heart, and kidney, particularly the distal tubules (32).

Animal studies using UCN-I-knockout mice suggest that UCN-I might not play a critical role in the HPA axis response to stress or in stress-induced autonomic control (33). However, when administered into the brain via the ventricles, UCN-I displays anorectic and anxiogenic properties similar to CRH. The UCNs are differentially distributed in the brain and periphery and appear to be involved in a rapidly expanding array of physiological mechanisms, particularly with respect to the appetite control and cardiovascular system, where they influence vascularization and angiogenesis (34, 35, 36).

An important component of the CRH/UCNs circuits appears to be a 37-kDa circulating protein capable of binding CRH and UCN-I, the CRH binding protein (CRH-BP) (37). In rodents, CRH-BP mRNA is expressed in the pituitary and the brain, especially the cerebral cortex, amygdala, bed nucleus of the stria terminalis, raphe nuclei, brainstem reticular formation, and olfactory, auditory, trigeminal, and vestibular sensory relay systems. Some of these sites also contain CRH and UCN-I-producing neurons or CRH/UCN target cells, such as the anterior pituitary corticotrophs. In humans, CRH-BP expression has been detected in brain, pituitary, liver, and placenta (38, 39, 40). The sites of overlapping expression of CRH-BP with CRH and UCN-I suggest that CRH-BP might modulate the synaptic or endocrine actions of CRH and/or UCN-I in the CNS and the pituitary. The biochemical properties of CRH-BP are well characterized; CRH-BP binds 40–90% of the total CRH, and CRH-BP levels are approximately 10-fold higher than CRH levels in most human brain regions (41). Despite this, the role of CRH-BP in health and disease remains unknown. What is known, however, is that CRH-BP blocks CRH-mediated ACTH secretion from anterior pituitary cultures (41, 42), suggesting that endogenous CRH-BP may act as a negative regulator of CRH in vivo, possibly playing a role in CRH clearance or degradation. This is supported by studies on transgenic mice, in which the response to constitutively elevated pituitary CRH-BP levels is a compensatory elevation of hypothalamic CRH and vasopressin, to maintain homeostasis in the HPA axis (43). In contrast, CRH-BP-deficient mice exhibit normal HPA axis function (44).


    III. CRH-R Subfamilies
 Top
 Abstract
 I. Introduction
 II. CRH-R Agonists: the...
 III. CRH-R Subfamilies
 IV. Agonist-CRH-R Interaction
 V. CRH-R Signaling...
 VI. Regulation of CRH-R...
 VII. Conclusions
 References
 
CRH and CRH-related peptides exert their actions in target cells via activation of two families of specific high-affinity CRH-R, termed R1 and R2, which are encoded by different genes (45, 46). Both of these genes are found in humans as well as in rodents and other mammals. In humans, CRH-R1 and -R2 are located on chromosomes 17 (17q12-q22) and 7 (7p21-p15), respectively, whereas in mouse, they are located on chromosomes 11 and 6, respectively (47, 48, 49, 50). Interestingly, linkage disequilibrium (LD) and single nucleotide polymorphism studies have implicated the CRH-R1 gene in Parkinson’s disease. This relates to a region of LD associated with late-onset Parkinson’s disease on chromosome 17 that encompasses the CRH-R1 gene. Specific haplotype-tagging single nucleotide polymorphisms, lying in this region of LD, are located in or flank the CRH-R1 gene and are significantly associated with increased risk of neurodegeneration and Parkinson’s disease (51). The CRH-Rs are membrane-bound proteins that belong to the family of seven transmembrane (7 TMD) G protein-coupled receptors (GPCRs) that, upon agonist binding, change their structural conformation and transduce signals across cells mainly through activation of heterotrimeric G proteins, which regulate a diverse network of intracellular systems.

The family of GPCRs is the largest single family of integral membrane proteins. Both CRH-R1 and CRH-R2 belong to the class B1 subfamily of GPCRs ("brain-gut" neuropeptide receptors). These receptors are encoded by 15 genes in humans, and the ligands for these receptors are polypeptide hormones of 27- to 141-amino acid residues. Structurally, CRH-R1 and CRH-R2 are approximately 70% identical at the amino acid level, but exhibit considerable divergence at the N terminus, consistent with their distinct pharmacological properties (see Section IV.A).

CRH-Rs are also found in nonmammalian vertebrate species. Studies on nonmammalian species (for review, see Ref. 52) have identified two receptor homologs closely related to the mammalian CRH-R1 and CRH-R2, from Xenopus laevis and chun salmon (Oncorhynchus keta). Studies of the puffer fish genome showed that Fugu rubripes also encodes two CRH-R homologs with significant similarity to human CRH-R1 and -R2, respectively. Unlike all other species studied, the Ameiurus catfish encodes two distinct mammalian CRH-R1 homologs and a single R2 ortholog. Thus, there has been extensive conservation in the expression of the two types of CRH-Rs in select organs during evolution. The two CRH-R1 homologs found in catfish showed greater similarity to each other than to the CRH-R2 homolog; therefore it is possible that the two catfish CRH-R1-like genes are likely derived from further gene duplication of the ancestral CRH-R1 gene. The CRH family of peptides exhibits significant sequence identity as well as primary- and secondary-structure similarity with the diuretic hormones in insects (53). These hormone peptides are essential for the regulation of fluid secretion by Malpighian tubules in insects. Their actions are also mediated via GPCRs homologous to the mammalian CRH-Rs. Functional studies suggest that CRH-related peptides coevolved with their GPCRs during the evolution of vertebrates and insects (53). Recent genomic analysis shows that both the mosquito (Anopheles gambiae) and fruit fly (Drosophila melanogaster) genomes encode orthologs of the diuretic hormone and diuretic hormone receptors. The two D. melanogaster diuretic hormone receptors showed 51 and 57% similarity to CRH-R1 and CRH-R2, respectively (54). These data provide us with a clear evolutionary trail for the origin of the CRH/CRH-R signaling system from invertebrates to vertebrates. Based on the concept that receptors and their agonists coevolved during evolution and that the evolution of most cell surface receptors in humans can be traced to invertebrates, it is likely that the ancestors of the diuretic hormone and diuretic hormone receptors coevolved and gave rise to diuretic hormone/diuretic hormone receptor signaling in insects and CRH/CRH-R signaling in vertebrates. The latter might have originated as a paracrine signaling system important for osmoregulation. However, during early evolution of vertebrates, the ancestor CRH/CRH-R signaling pathway assumed additional functions that included the regulation of stress responses toward various environmental factors. The presence of multiple highly conserved CRH-like peptides and receptors in vertebrates suggests that gene duplication and the subsequent divergence of the regulatory mechanism of these paralogous genes provided an advantage during vertebrate evolution. Interestingly, CRH has been shown to regulate metamorphosis in response to pond drying in some amphibian species. This is consistent with the role of CRH family peptides as osmoregulators and as a stress transducer between the environment and the physiological responses of an organism. Therefore, the CRH family of peptides and their receptors are phylogenetically ancient developmental signaling molecules that allow developing organisms to coordinate physiological responses to a changing environment.

A. Distribution of CRH-R
CRH-R1 mRNA is widely expressed in mammalian brain and pituitary and is responsible for activation of the POMC gene and ACTH and ß-endorphin release from the anterior pituitary. High levels of CRH-R1 are found in the cerebral cortex, cerebellum, amygdala, hippocampus, and olfactory bulb (55). In human peripheral tissues, CRH-R1 is expressed in a wide range of tissues such as the testis, ovary, endometrium, myometrium, placenta, adrenal, adipose tissue, skin, spleen, heart, and specific cells of the immune system (55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66). CRH-R2 mRNA is expressed in a discrete pattern in the brain, with highest densities in the lateral septal nucleus, bed nucleus of stria terminalis, VMH nucleus, olfactory bulb, and mesencephalic raphe nuclei (55). In addition, CRH-R2 is also widely expressed in peripheral tissues, with high levels in the skin and skeletal, smooth, and cardiac muscle (67, 68).

The distribution of CRH-R1 and CRH-R2 in the CNS and periphery is distinct and suggests diverse physiological functions, as implied by the phenotypes of the CRH-R1 or CRH-R2 knockout mice. Interestingly, mice deficient in CRH-R1 display decreased anxiety-like behavior and have an impaired stress response (69, 70), whereas CRH-R2 mutant mice have increased anxiety-like behavior, an accelerated HPA response to stress, and impaired cardiovascular function (71, 72, 73). Paradoxically, however, the responses to administration of CRH-R2 agonists and antagonists into specific brain regions reveal both anxiolytic and anxiogenic-like roles for CRH-R2 (for review, see Ref. 74). Also, it is important to note that the expression of CRH-R1 and -R2 in the periphery exhibits important species-related differences. Most studies in animal models suggest that the CRH-R2 is the main functional receptor expressed in peripheral organs (35, 36, 76, 77). In contrast, most human peripheral tissues express both CRH-R1 and -R2, indicating a higher level of complexity and more subtle roles for CRH and UCNs in human physiology and pathophysiology. From the available data, it would seem that the expression of both CRH-R1 and CRH-R2 in human peripheral tissues allows these peptides to exert diverse and sometimes contrasting effects. Certainly CRH and the UNCs play a role in physiological processes such as energy balance and metabolism, the maintenance of cardiac function and vascular tone, gastrointestinal motor function, reproduction, pregnancy and parturition, angiogenesis, and vascularization. It seems likely that they exert their actions via the activation of CRH-R1 or CRH-R2 or both (19, 21, 77, 78, 79, 80, 81, 82, 83, 84), thus allowing CRH to fulfil its role as a stress-hormone integrating central and peripheral responses to stressful stimuli.

B. Splicing pattern and genomic organization
One of the more remarkable findings from the Human Genome Project (http://genome.wellcome.ac.uk) was the observation that human chromosomes harbor far fewer genes than were predicted. It appears that the key to the differences between humans and worms lies in the functions of some human genes and the proteins they encode. Furthermore, functional complexity is added by the use of alternative splicing of mRNA to create several distinct proteins from a single gene. Both CRH-R genes exhibit substantial similarity to the glucagon/PTH receptor (PTH-R) gene family that is characterized by the presence of introns within its TMD/cytoplasmic module, highly conserved cysteines in its extracellular domain (ECD), and a highly conserved first intracellular (IC) loop. The exon/intron junctions of the CRH, PTH, and glucagon receptor genes are remarkably similar in alignment following the signal peptide encoded by exons 1 and 2. The CRH-R1 exon/intron junctions are aligned to that of the PTH and glucagon receptors after exons 3, 5, 7, 8, 9, 10, and 12. CRH-R1 is similar to the PTH-R in that amino acids 457–509 are divided into exons 11 and 12, whereas this domain in the glucagon receptor is a single exon. The exon-intron organization of this GPCR family suggests a common evolutionary origin and, unlike other GPCR subfamilies, permits extensive alternate splicing. The evolutionary process has used this property to generate diverse receptors from a single gene, each with differing ligand specificity, binding, and G protein coupling. The fact that B1 GPCRs have multiple coding exons in comparison to other GPCR family members may be due to their being generated as an alternative to gene duplication during evolution. The increasing complexity seen within this GPCR subfamily seems to have evolved by divergence from a simpler regulated gene with increasing complexity of the organism.

1. CRH-R1.
In most mammals, the fully active CRH-R1 receptor protein arises from transcription of all 13 exons present within the CRH-R1 gene sequence. In humans, however, the genomic organization and regulatory control is different, possibly reflecting a higher level of regulation associated with species evolution. In humans, the CRH-R1 gene, which spans over 20 kb, contains 14 exons, and the complete gene product is a 444-amino acid 7 TMD protein receptor, termed CRH-R1ß, that exhibits impaired agonist-binding and signaling properties (45). Excision of exon 6, which encodes for a 29-amino acid insert in the first IC loop (IC1), from the mRNA transcript results in expression of CRH-R1{alpha} mRNA. This appears to be the main functional CRH-R1 receptor variant containing 415 amino acids, which primarily mediates CRH (and UCN-I) actions and is widely expressed throughout the body. Therefore, the CRH-R1ß can be regarded as a "pro-CRH-R1" receptor isoform, with unknown, if any, physiological function. Although lack of CRH-R1ß receptor isoform-specific antibodies has prevented the conclusive demonstration of CRH-R1ß protein expression in native cells, mRNA studies suggest that this CRH-R1 variant is expressed in the pituitary (45) as well as in peripheral tissues such as the myometrium (59), mast cells (85), endometrium (86), and heart (our unpublished data). Thus far, no tissue has been identified expressing CRH-R1ß alone, suggesting that the splicing mechanism is closely related to the mechanism regulating CRH-R1 gene expression in native tissues. One interesting feature of CRH-R1ß is that it is only expressed in humans. In rodents, the exon 6 encoding the IC loop characteristic of CRH-R1ß is absent, and, as mentioned above, the mouse and rat CRH-R1 gene contains only 13 exons (87). This suggests that the CRH-R1ß variant does not perform a crucial physiological role. This type of splicing mechanism, utilizing specific exon/intron splicing sites and excision of exons to generate multiple receptor variants, is not unique to CRH-R1, but is rather shared by other members of the B1 receptor superfamily. For example, a similar calcitonin receptor (CT-R) variant has been identified in a giant cell tumor of bone that contains a similar 16-amino acid insert in IC1 and has reduced signaling properties compared with the fully active receptor variant (88). The insert sequences in IC1 of both CRH-R1ß and CT-R "long" variant contain highly charged amino acids in their insert C termini proximal to the IC1/TMD2 junction, which might be responsible for the impaired ability to transmit signals across the cells. Just why these mRNA variants have been conserved through evolution is unknown.

The CRH-R1 gene appears to be subject to significant alternative splicing, and a growing number of CRH-R1 mRNA splice variants have been described in humans and other species. Although, as mentioned above, their physiological role is questionable, structural analysis has provided us with some useful models for studying the structural determinants of the CRH-R1 functional characteristics in comparison with other members of the B1 family of GPCRs. These variants are generated by various partial or complete exon deletions, some of which are associated with a frame-shift in the open reading frame (63, 89, 90, 92). These variants have been termed R1c-n, and all have exon 6 spliced out together with other deletions. CRH-R1c is missing exon 3, and therefore 40 amino acids are missing from the N terminus, including two regions critical for high-affinity ligand binding (see Section IV.B). Mutations in this region can affect CRH binding (93); thus, it is not surprising that expressed recombinant CRH-R1c has a decreased CRH binding capacity. CRH-R1d, which has been identified in humans and hamsters, has exon 13 (exon 12 in the rodent R1 homolog) deleted, which leads to the loss of 14 amino acids from the C-terminal end of the putative 7 TMD. This might result in either a short 7 TMD where the proximal residues of the C-tail are drawn into the membrane or a 6-TMD receptor variant containing a protein segment that fails to segregate into the membrane lipid bilayer leading to an EC C terminus, similar to the CT-R variant {Delta}e13 (94). Overexpression studies of recombinant CRH-R1d have demonstrated that, although this CRH-R1 variant retains agonist binding characteristics, it has significantly impaired G protein coupling and signaling properties. Given that most sites involved in GPCR signaling through posttranslational modifications and docking of signaling proteins are in the carboxyl-terminal tail, the possible retraction and distortion of this tail induced by the 14-amino acid sequence deletion would be expected to alter signaling and hence function. Similar splice variants arising from the same exon deletions have been described for other members of the B1 subfamily of GPCRs, such as the PTH-R (95) and the type II receptor for vasoactive intestinal peptide and pituitary adenylate cyclase activating polypeptide receptor for vasoactive intestinal peptide (96). Analysis of the nucleotide sequence reveals that there are conserved splicing sites within the 7 TMD (site of exon deletion), shared among the members of this receptor family. The structural features of the receptor splice variants found in humans are shown in Fig. 1Go.


Figure 1
View larger version (40K):
[in this window]
[in a new window]
 
FIG. 1. Schematic representation of exon structure of human CRH-R1 variants and structural differences of potential protein products.

 
The work of Pisarchik and Slominski (63, 92) in human, mouse, and hamster has expanded the list of potential CRH-R1 mRNA splice variants involving the excision of multiple exons, which has profound consequences for the tertiary structure of any translated protein. Currently, eight additional mRNA transcripts have been identified, named R1e-n, each with a unique exon splicing pattern and predicted protein structure. The functional significance of these novel transcripts is uncertain.

The physiological relevance of the CRH-R1 mRNA splice variants has been dismissed as being a result of aberrant transcription. Characterization of the biological activity and signaling properties of some of these CRH-R1 variant receptor proteins is based on heterologous overexpression systems using recombinant protein expression and not native cells. This is due to a lack of suitable methods to demonstrate endogenous CRH-R1 variant expression and to study their physiological roles, if any, in mammalian tissues. Nevertheless, the complex pattern of CRH-R1 alternative splicing that is tissue-specific and physiological process-specific [i.e., expression of myometrial CRH-R1d at term of human pregnancy (90)] and affected by environmental stimuli [i.e., expression of CRH-R1g after UV irradiation (63)] suggests a functional role in modifying CRH actions in target tissues. It is conceivable that the presence of different isoforms could have a substantial influence on cellular response to CRH and CRH-like agonists. Regardless of whether or not these aberrant transcripts are significantly expressed as protein products, the dominant expression of the aberrant transcripts at the expense of the wild-type receptor at transcription would have the end result of reducing levels of functional receptor. Although our current knowledge of the mechanisms regulating CRH-R1 splicing and the specific function of distinct CRH-R1 variants is poor, emerging evidence points toward potentially important biological mechanisms. For example, we have identified a specific mechanism involving steroid hormones, which regulates the ratio of CRH-R1{alpha}/R1ß mRNA expression in human myometrial smooth muscle cells during pregnancy. This may represent a mechanism for regulating tissue responsiveness to CRH (Fig. 2Go) (97). In this example, progesterone alters the transcription of the CRH-R1 gene in such a way that the predominant transcript encodes for CRH-R1{alpha}, thus enhancing tissue responsiveness to CRH, an effect that is blocked by estrogen. Furthermore, some variants of CRH-R1, such as the CRH-R1d, fully retain the ability to bind agonist with high affinity although they are deficient in signaling. These receptor variants might act as "decoy receptors" capable of competing with the full-length receptors for agonist binding and absorbing CRH and CRH-like peptide bioactivity and therefore, change the efficiency of hormonal stimulation. Interestingly, in other receptor systems, such as the CT-R system, similar splice variants containing deletions in the 7 TMD, when coexpressed with the fully active CT-R variant, act as dominant negative regulators by forming heterooligomers within the cell. This process inhibits normal receptor expression at the cell surface, leading to a reduction in the signaling response (98). Similar interactions have been observed with the truncated forms of CRH-R1, e and h, which are capable of binding CRH and influencing tissue responsiveness to CRH. For example, in coexpression experiments, CRH-R1e attenuated and CRH-R1h amplified CRH-R1{alpha} signaling (99). These observations provide tantalizing, although far from conclusive, evidence suggesting that the regulation of transcription and splicing of the CRH-R1 receptor gene might play a major role in determining tissue responsiveness to CRH and UCNs. It should be remembered, however, that there is no evidence that these transcripts are translated into functional protein. This might not be a prerequisite, however, because the dominant expression of the aberrant transcripts at the expense of the wild-type receptor at transcription would have the end result of reducing levels of functional receptor.


Figure 2
View larger version (36K):
[in this window]
[in a new window]
 
FIG. 2. Examples of differential splicing mechanisms regulating CRH-R variant expression. In this set of experiments, quantitative RT-PCR with specific oligonucleotide primers was used. Top, CRH-R1{alpha} and -R1ß mRNA expressed in human myometrial tissue obtained at term before the onset (NL) and during labor (L). Results are presented as maxima of melting curves of CRH-R1{alpha} (89.20 C), and CRH-R1ß (88.56 C) genes. Similar results were obtained from six independent myometrial biopsies. Bottom, CRH-R1{alpha} and -R1ß mRNA expressed in human myometrial cells treated with or without progesterone (P4; 5 µM) for 16 h. These results represent four independent myometrial cell preparations.

 
With these caveats in mind, Pisarchik and Slominski (92) have proposed a division of all potential CRH-R1 variants into four groups according to their potential impact on agonist signaling (Fig. 3Go):


Figure 3
View larger version (43K):
[in this window]
[in a new window]
 
FIG. 3. Proposed classification of potential CRH-R1 variants according to their potential impact on agonist signaling. aa, Amino acids. [Derived from Ref. 92 .]

 
a. Group 1.
The first group includes variants with no frame-shift (ß, c, d, g, and n) but with a variable number of TMDs and intra- or extracellular C terminus (yet to be conclusively demonstrated for some variants) that differ in their agonist binding and G protein-coupling characteristics.

b. Group 2.
The second group consists of mRNA variants missing exons 1, 2, 3, and 4, which are important for agonist binding and therefore appear functionally inert (R1e). Although it is difficult to attribute any direct role to individual members of this group, they might have important functional significance by modulating CRH-R1{alpha} activity.

c. Group 3.
Members of the third group have conserved the original reading frame of CRH-R1 up to the fifth exon, thus retaining an intact CRH-binding domain but having no TMDs. Two members of this group have been identified: CRH-R1j and CRH-R1h. CRH-R1j has exon 5 deleted, whereas CRH-R1h has an insertion of a cryptic exon between exons 4 and 5. If translated, these isoforms could potentially serve as soluble CRH-BPs or modulate agonist binding.

d. Group 4.
The fourth group consists of mRNA variants that have an intact CRH-binding domain and a variable number of TMDs (CRH-R1f, k, and m). Each of these transcripts contains a frame-shift that potentially alters their C terminus, raising questions about their potential for participation in signal transduction.

Studies on the promoter of the human CRH-R1 gene suggest that CRH-R1 agonists such as CRH and UCN-I can positively regulate CRH-R1 gene transcription (100) in agreement with several animal studies (101, 102, 103, 104). The hypothalamic PVN is a site of CRH-induced up-regulation of CRH-Rs. An example is the autoregulation of CRH biosynthesis in the PVN through up-regulation of CRH-R1, showing the mediation of positive effects on the amygdaloid CRH system (101). In rats, CRH-R1 mRNA in the PVN is increased after stress, and this response is attenuated by central CRH blockade. Elevated levels of central CRH may trigger CRH-R1 mRNA transcription in the PVN, hippocampus, and frontal cortex (102), suggesting a positive feedback of CRH on its own receptor, acting as a functional adaptation of the HPA axis in response to stress (103). These animal data are supported by studies on human cells and the human CRH-R1 promoter (100) in which a transcriptional positive feedback effect was shown to be dependent on activation of both protein kinase A (PKA) and protein kinase C (PKC) in the case of CRH, and PKC alone in the case of UCN. Collectively, the data suggest that the CRH-R1 gene is under the influence of both CRH and UCN, acting via distinct signaling pathways to create a positive feedback loop and regulate further the transcription of the receptor. However, this feedback mechanism requires further characterization because there is conflicting evidence suggesting positive and negative effects of CRH on rat anterior pituitary CRH-R1 expression (104, 105). Glucocorticoids are another important regulator of CRH-R1 gene expression because dexamethasone has been shown to act as a negative regulator of CRH-R1 mRNA expression in rodent anterior pituitary and PVN (106, 107, 108, 109).

Interestingly, the human as well as the rat CRH-R1 promoters are TATA-less, suggesting that Sp1 sites might play a crucial role for driving transcription because there are a high number of potential Sp1 elements clustered around the transcriptional start sites (100). Other members of the B1 family of GPCRs share this feature, including the promoters for the CT-R, the glucagon-like peptide receptor-1 and the vasointestinal peptide 1 receptor genes (100). The promoter region of the CRH-R1 also contains sites for Egr 2 (Krox-20) and YY1 (yin-yang 1) transcription factors (members of the zinc finger DNA binding transcription factors), which may also play a role in controlling transcription. The presence of a putative transcription response element, however, should not be taken as proof that a potential binding factor is a regulatory factor.

2. CRH-R2.
Interestingly, the CRH-R2 gene exhibits a completely different splicing pattern compared with CRH-R1, possibly relevant to its distinct role in mammalian physiology. Mammals express three known CRH-R2 variants: CRH-R2{alpha} and -R2ß are found in both human and rodents, and -R2{gamma} has so far been found only in the limbic regions of the human CNS (110, 111, 112), although genomic clones containing the CRH-R2{gamma} subtype have been identified in the olive baboon and chimpanzee. All three variant mRNAs are produced by the use of an alternate 5' exon 1 that splices onto a common set of downstream exons (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12), resulting in R2 variants, with identical transmembrane and C-terminus domains. These variants differ only in their N-terminal ECDs; CRH-R2{alpha} has 34 amino acids at the N terminus, which are replaced by 61 amino acids to form the CRH-R2ß or 20 amino acids to form the CRH-R2{gamma}. The human CRH-R2 gene spans 50 kb and consists of 15 exons. Exon ß1a contains the 5'-untranslated region and the start codon for CRH-R2ß and, together with exon ß1b, contributes to the N-terminal ECD. Exons {gamma}1 and {alpha}1 contain the 5'-untranslated region and start codon for CRH-R2{gamma} and CRH-R2{alpha}, respectively, which contributes to their unique N-terminal ECD. The remainder of the N terminus is encoded on exons 2–7, with exons 7–11 coding for the 7 TMDs and exon 12 coding for the intracellular C terminus, all of which are common to all known functional CRH-R2 receptors (113). The different N termini do not significantly alter agonist binding and signaling properties of the various CRH-related peptides, although the CRH-R2ß is about 10-fold more potent in second messenger activation compared with CRH-R2{alpha} or R2{gamma} (112). These variants, however, do exhibit significant differences in their tissue distribution. The CRH-R2{alpha} and the CRH-R2ß are expressed in both the brain and periphery (114), although CRH-R2{alpha} is mainly localized to the subcortical structures (114), whereas the CRH-R2{gamma} appears to be confined predominantly to the brain (112). In addition, both the CRH-R2{alpha} and CRH-R2ß variants are differentially expressed in heart, skeletal muscle, and myometrium (59, 77). In rodents, the CRH-R2{alpha} is the predominant CRH-R2 variant expressed in the rat brain (112, 114), whereas CRH-R2ß mRNA is widely expressed in peripheral tissues, with highest levels in the skeletal muscle, heart, and skin (67). Interestingly, the CRH-R2ß variant is primarily expressed peripherally in rodents, whereas the CRH-R2{alpha} is the predominant splice variant found in the periphery of humans. Given that different promoters regulate each CRH-R2 variant expression (see below), this diversity could be related to the regulation of expression at different tissues. The CRH-R2 gene appears to have diverged from CRH-R1 by duplication and then by increasing complexity to produce three subtypes with different pharmacology and differential expression. CRH-R2{alpha} is present in amphibians, whereas in rodents a second variant CRH-R2ß is also present, and in primates a third variant CRH-R2{gamma}.

The arrangement of the CRH-R2 gene provides the potential for generating multiple alternate splice forms of mRNA (111, 112, 113). It is predicted that the CRH-R2ß pre-mRNA is differentially spliced, producing multiple alternate splice forms, and if any permutation of exons between ß1a and 2 occurs, at least 12 alternate splice forms can be produced, some with high levels of expression (113). Different types of aberrant mRNA splice variants have also been reported: 1) a truncated CRH-R2{alpha} mRNA isolated from rat amygdala (CRH-R2{alpha}-tr), which encodes for the first three TMDs and a part of the fourth TMD of the CRH-R2{alpha} and binds CRH, but not UCN-I, with almost the same affinity as CRH-R2{alpha} (115); and 2) a soluble CRH-R2{alpha}, in which exon 6 is deleted and translation of this variant produces a predicted 143-amino acid soluble protein. The translated protein includes the majority of the first ECD of the CRH-R2{alpha} followed by a unique 38-amino acid hydrophilic C terminus resulting from a frame-shift produced by deletion of exon 6. The soluble CRH-R2{alpha} variant has high levels of expression in the olfactory bulb, cerebral cortex, and midbrain regions. Experiments using recombinant proteins showed that soluble CRH-R2{alpha} protein inhibits cellular responses to CRH and UCN, supporting a potential role as a biological modulator of CRH and CRH-related peptides (116).

Three distinct promoters and differential splicing (113) control regulation of the human CRH-R2 gene expression. The arrangement of the 5'-flanking region of the gene is similar to that of the CRH-R1 gene in that it lacks a functional TATA or CCAAT box and contains several specificity protein 1 binding elements, which suggests that these elements constitute the minimal promoter. Each of the promoters contains some consensus regulatory sequences that provide some insights into CRH-R2 physiology. For example, the CRH-R2ß and CRH-R2{alpha} promoters contain consensus sequences for binding myogenic transcription factors, such as the myocyte-specific enhancer MEF-2, which may be involved in the regulation of expression in cardiac, skeletal, uterine, and smooth muscle. In contrast, the CRH-R2{gamma} promoter contains a putative regulatory element for the pituitary-specific factor Pit-1a, raising the possibility that the CRH-R2{gamma} is the CRH-R2 variant expressed in the pituitary (117). Putative glucocorticoid-responsive elements are also present in the CRH-R2{alpha} promoter, consistent with studies showing glucocorticoid regulation of CRH-R2 expression in the heart and brain (118, 119). This mechanism appears to be conserved across species because the mouse CRH-R2{alpha} 5' flanking region contains 23 putative half-palindrome glucocorticoid response element sequences within its 2.4-kb sequence (120). This might explain the finding that hypothalamic murine CRH-R2{alpha} gene transcription is inhibited by glucocorticoid administration in vivo and enhanced by adrenalectomy. However, other studies have reported that in the VMH nucleus, the levels of CRH-R2 mRNA are up-regulated by corticosterone (119) and leptin administration (121), whereas, stressful stimuli such as starvation (119), repeated immobilization stress (122), and maternal deprivation (123) decrease the expression level of CRH-R2 mRNA in the rat VMH. The same study suggested that CRH-R2 gene expression is differentially regulated in different hypothalamic nuclei because neither corticosterone administration, starvation, nor adrenalectomy influenced the levels of CRH-R2 mRNA in the hypothalamic PVN (119). Clearly, additional studies are required to clarify these discrepancies and reveal the manner by which the CRH-R2 pathway is involved in physiological responses to stress in normal and transgenic mice models.

Despite the well-established role of CRH-Rs in human physiology and nosological states such as clinical depression, no striking abnormalities in CRH-R expression and/or function in human disease have been observed, with a few notable exceptions. In ACTH-secreting pituitary adenomas, a significant overexpression of the CRH-R1 mRNA has been reported (124), which may contribute to a disturbed receptor regulation. Also, in brains from suicide victims, the mRNA for cerebral cortical CRH-R1, but not CRH-R2, appears to be reduced, possibly secondary to sustained increase of CRH activity (125, 126), supporting the view that only the CRH-R1 subtype is aligned with mood disorders. The human placenta in patients with preeclampsia is another example of an organ where abnormal expression of CRH-R1 and -R2 has been reported and linked to an increased placental CRH production and abnormal regulation of the feto-placental blood flow (127, 128). However, in these cases it remains to be established whether abnormalities in CRH-R expression contribute to the pathogenesis of the disease or are secondary to the illness or the stressors associated with it.


    IV. Agonist-CRH-R Interaction
 Top
 Abstract
 I. Introduction
 II. CRH-R Agonists: the...
 III. CRH-R Subfamilies
 IV. Agonist-CRH-R Interaction
 V. CRH-R Signaling...
 VI. Regulation of CRH-R...
 VII. Conclusions
 References
 
A. Receptor pharmacology
Despite an overall similarity of greater than 70% at the amino acid level, the two types of CRH-R exhibit different pharmacological characteristics, reflecting their unique and sometimes complementary or contrasting roles in specific tissues. The low homology at the extracellular N terminus (approximately 47%) accounts for the distinct ligand-specificity characteristics because the N terminus is primarily responsible for agonist binding (see Section IV.B). CRH-R1 binds CRH as well as UCN-I, but not UCN-II and -III, with equivalent high affinity. However, the CRH-R2 exhibits ligand selectivity and binds all the UCNs with significantly higher binding affinity than CRH, suggesting that these peptides may be its natural ligands. The rank order of mammalian CRH-like peptides for binding affinity at CRH-R1 receptors is: UCN-I > h/rCRH > {alpha}-helical CRH (9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41) (a peptide antagonist); however, at the CRH-R2{alpha} receptors, the binding affinity rank order is: human (h) UCN-I = hUCN-II > astressin = astressin2-B = antisauvagine-30 > hUCN-III > {alpha}-helical CRH (9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41) > h/rCRH >> ovine (o) CRH. oCRH differs from its human homolog by eight amino acid residues and, in cellular systems expressing both types of CRH-R, it is the agonist of choice for the selective stimulation of CRH-R1, because it preferentially binds to CRH-R1. The binding constants of oCRH for CRH-R1 and -R2 differ by two orders of magnitude (129, 130). Recently, a CRH-R1-specific peptide agonist was generated, named cortagine ([Glu21,Ala40] [Svg1–12] x [h/rCRH14–30] x [Svg30–40]), by synthesis of chimeric peptides derived from h/rCRH, oCRH, and sauvagine (Svg; a frog CRH-related agonist) (131), and this might allow better characterization of CRH-R1 and -R2 function.

Characterization of the structure-function relationship of CRH and CRH-related agonists has revealed the presence of three main functional domains (132, 133, 134) (Fig. 4Go). The first domain (residues 1–16) is responsible for both binding and receptor activation. The second domain (residues 17–31) appears to function as a linker providing the appropriate spatial and conformational support for the two binding regions located in domains 1 and 3 and contains the CRH-BP binding site (stretch of amino acid residues 22–25, Ala-Arg-Ala-Glu, of h/rCRH representing the ARAE motif) (135). Finally, the third domain, consisting of residues 32–41, is important for receptor binding. Sequence analysis studies of all the members of the family of CRH-related agonists have revealed a number of differences between the CRH-R2 selective agonists, UCN-II and UCN-III, and the CRH-R nonselective agonists, CRH and UCN-I. A residue in the first domain, proline at position 11 (the numbering of residues is based on h/rCRH sequence), is found only in CRH-R2 selective peptides and might play an important role in determining receptor selectivity because substitution of Pro11 in the hUCN-II sequence with corresponding amino acids, found in the CRH-R nonselective agonists, decreased binding potency to CRH-R2 while increasing CRH-R1 activity. It is likely that the presence of this proline residue in the first domain alters the {alpha}-helix structure of the peptide, because proline is the strongest modifier of {alpha}-helix structure and a promoter of turn motifs. Consequently, the presence of Pro11 in the sequence of UCN-II and UCN-III may indicate that the {alpha}-helix present in the first domain of the nonselective CRH-R peptides is modified through the introduction of a kink or turn motif, a modification that may be important in determining agonist selectivity (136). Indeed, circular dichroism spectroscopy studies support the view that the presence of Pro11 in this region decreases {alpha}-helicity and impairs binding to CRH-R1 (137). Furthermore, in the third domain of the CRH-related peptides, CRH-R2 selective peptides contain alanine residues at positions 35 and 39, whereas CRH-R nonselective peptides contain an invariant arginine at position 35 and an acidic amino acid at position 39. Introducing proline at position 11 and alanine at positions 35 and 39 results in increased CRH-R2 selectivity in CRH-R nonselective peptides, mainly through the loss of CRH-R1 potency. Interestingly, unlike sauvagine and hUCN-I, this substitution in the h/rCRH sequence results in a loss of potency at both the CRH-R1 and -R2 (136), indicating that CRH requires additional substitutions to achieve substantial CRH-R2 selectivity.


Figure 4
View larger version (25K):
[in this window]
[in a new window]
 
FIG. 4. Schematic representation of amino acids within the CRH/CRH-related agonists sequence important for determining CRH-R subtype selectivity. It is now accepted that amino acid residues 32–41 are important for receptor binding, whereas residues 1–16 are responsible for both binding and receptor activation. Residues present in the domain 17–31 appear to function as a linker providing the appropriate spatial and conformational support for the two binding regions. CRH-R2 selective agonists contain a proline at position 11 and alanine residues at positions 35 and 39 (the numbering of residues is based on h/rCRH sequence). In contrast, CRH-R nonselective peptides contain an arginine at position 35 and an acidic amino acid at position 39.

 
B. Receptor-agonist interaction: the role of the N- and J-domains
Like all class B1 GPCRs, the CRH-Rs possess a large ECD that allows recognition and high-affinity binding to the carboxyl-terminal regions of peptide ligands. This interaction alone is not sufficient to stimulate coupling of the receptor to G proteins, and an additional interaction is required between the juxta-membrane domain of the GPCR (the transmembrane helices and intervening loops J-domain) and the first few residues within the amino-terminal portion of the peptide ligand to induce intracellular signal activation. Based on this, the generation of N-terminus-truncated CRH peptides produces high-affinity competitive antagonists for CRH-R [e.g., {alpha}-helical CRH (9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41)]. Additional modifications have produced a number of different antagonist peptides including astressin [cyclo(30–33)-[D-Phe12, Nle21,38, Glu30, Lys33]CRF-(12–41)], a high-affinity antagonist for both CRH-R receptors with enhanced biological stability (138) and no detectable agonist activity for the CRH-R1.

Mass spectrometric analysis of a soluble form of the N terminus of the human CRH-R1 yielded a 1:1 complex with ligand, and analysis of the disulfide bond arrangement revealed bonds between Cys30 and Cys54, Cys44 and Cys87, and Cys68 and Cys102 (139) (Fig. 5AGo). An identical arrangement was also found in the soluble N terminus of the mouse CRH-R2ß (140). This arrangement is similar to that of the N terminus of the PTH-R, suggesting a conserved structural motif in the N-terminal domain of the B1 family of GPCRs (141). Recent studies using nuclear magnetic resonance have provided valuable insights into the three-dimensional (3D) structure of the N-terminal ECD of the mouse CRH-R2ß (142). This technique has identified a short consensus repeat (SCR) in this domain, which is more commonly found in proteins of the complement system. This model proposes that the N terminus captures the C-terminal segment of the ligand, allowing the N terminus to penetrate into the TMD region of the receptor to initiate signaling. Key residues of the SCR in the ECD1 are conserved in the GPCR subfamily, suggesting the importance of this SCR fold in the ECD1s of this subfamily.


Figure 5
View larger version (51K):
[in this window]
[in a new window]
 
FIG. 5. A, Schematic of the disulfide bond arrangement between Cys30-Cys54, Cys44-Cys87, and Cys68-Cys102 in the N terminus of the CRH-R1 receptor, based on mass spectrometric analysis of a soluble form of the N terminus of the human CRH-R1 (139 ). B, Mapping the conserved amino acids, as shown in Ref. 142 , onto the 3D structure of ECD1-CRH-R2ß. Stereo view (I) and surface view (II) of the 3D structure with side chains of the conserved amino acids within the B1 family of GPCRs colored dark blue and similar residues colored light blue. The salt bridge between Asp65 and Arg101 is shown by the green dashed line. Inset, Sequence alignment of the ECD1 of the CRH-R family and B1 GPCR family. Only a representative set of sequences is shown. Conserved cysteines are highlighted in yellow, conserved amino acids throughout the whole B1 family are blue, and amino acids conserved more than 80% throughout the whole B1 family are light blue. Mutagenesis studies for the identification of receptor-ligand interaction are summarized here: magenta stars represent amino acid segments proposed to be involved in hormone binding, and green stars represent amino acid segments that are less important for binding. The ß-sheet secondary structure elements are labeled by an arrow above the sequence. TS, Treeshrew; AmNebu, Ameriurus nebulosus; TuBel, Tupaia belangeri; VIPRI, vasoactive intestinal peptide receptor 1; GLPIR, glucagon-like peptide 1 receptor; PTHR2, PTH receptor 2. [Reproduced with permission from C. R. Grace et al.: Proc Natl Acad Sci USA 101:12836–12841, 2004 (142 ). © National Academy of Sciences USA.]

 
According to studies by Grace et al. (142), the CRH-R2ß N terminus contains two antiparallel ß-sheets comprising residues 63–64 (ß1 strand), 70–71 (ß2 strand), 79–82 (ß3 strand), and 99–102 (ß4 strand). This polypeptide fold is stabilized by three disulfide bonds between residues Cys45-Cys70, Cys60-Cys103, and Cys84-Cys118 and by a central core consisting of a salt bridge involving Asp65-Arg101, sandwiched between the aromatic rings of Trp71 and Trp109. The two ß-sheets, interconnected by this core, form the scaffold flanked by two disordered regions (residues 39–58 and 84–98) (Fig. 5BGo). Furthermore, the core is surrounded by a second layer of highly conserved residues, Thr69, Val80, and Arg82, and conservatively conserved residues, Thr63, Ser74, and Ile67. Other conserved residues, such as Pro72 and Pro83, might be important for ending the ß-strands, as well as Gly77, Asn106, and Gly107 located in the hinge regions of the two ß-sheets, which may be important for their relative orientation. Another cluster of conserved residues is present in the disordered loop between strands ß3 and ß4 (Gly92, Phe93, Asn94, and Thr96). In contrast, the disordered loop from residues 39–58 is highly variable in amino acid sequence.

Astressin B has been used in nuclear magnetic resonance chemical-shift perturbation experiments to identify the CRH-R2ß ligand binding site (142). The largest chemical-shift perturbations were observed in the segments comprising residues 67–69, 90–93, 102–103, and 112–116. These residues are clustered in the cleft region between the tip of the first ß-sheet and the edge of the "palm" of the second ß-sheet. The observed changes in the chemical shifts in the disordered loop region 85–98 suggest that folding occurs after ligand binding. Circular dichroism spectroscopy data support this view and show a conformational change toward a more structured N terminus upon ligand binding (140). Furthermore, the accumulated distribution of positive charges on the "back side" of the structure (Arg47, Arg82, Arg97) is indicative of orientation toward the negatively charged ECDs 2–4 and the transmembrane segment of CRH-R2ß. Based on these observations, it has been proposed that hormone binding and receptor activation occur in two steps. First, the ligand binds with its C-terminal segment to the solvent exposed binding site of the N terminus. In order for an agonist to elicit a signaling response, its N-terminal segment penetrates into the transmembrane segments of the CRH-R, producing activation of the receptor. In contrast, antagonists, like astressin, lacking the first 11 N-terminal residues critical for receptor activation cannot penetrate the transmembrane region and activate the receptor (Fig. 6Go).


Figure 6
View larger version (59K):
[in this window]
[in a new window]
 
FIG. 6. Models depicting modes of action of CRH-R peptide agonists (A and B) or antagonists (C and D). According to this, CRH-R agonists recognize specific amino acid residues to the large ECD (N-domain) that enables high-affinity binding to the carboxyl-terminal regions of peptide ligands. This interaction alone is not sufficient to stimulate coupling of the receptor to G proteins, and an additional interaction is required between the receptor domain containing the transmembrane helices and intervening loops (J-domain) and the amino-terminal portion of the peptide ligand. Penetration of the agonist’s N terminus into the receptor J-domain initiates intracellular signal activation. Generation of N-terminus truncated CRH peptides produces high-affinity competitive antagonists for CRH-R that cannot induce signal transduction. However, recent observations suggest that some of these antagonists (i.e., astressin) can modify CRH-R important biological properties such as receptor endocytosis.

 
The J-domain also appears to participate in agonist/antagonist interactions with CRH-R2. Studies using the subtype 2{alpha} of the rat CRH-R2 (143) revealed that the juxta-membrane receptor domain determines the selectivity of antisauvagine-30 [a synthetic CRH-R2 antagonist (144)], whereas the N-terminal-ECD contributes to selectivity of UCN-III, and both domains contribute to selectivity of UCN-II and astressin2-B [CRH-R2 antagonist (145)]. Therefore, ligands differ in the contribution of receptor domains to their selectivity, and CRH-R2-selective antagonists can bind the J-domain. Unlike the CRH-R1 receptor (see below), the CRH-R2 J-domain stabilizes affinity for agonists like UCN-II by about 30-fold and might act as a CRH-R2/CRH-R1 selectivity determinant. A further weak increase in CRH-R2 affinity for UCN-II is achieved by receptor G protein coupling. Overall, the current data suggest that the J-domain of the CRH-R2 receptor binds ligands more strongly than the J-domain of the CRH-R1 (at the R state).

The location of peptide ligand binding sites on the CRH-R1 receptor has also been investigated extensively by a number of studies using mutant and chimeric mammalian CRH-R1s (93, 146, 148, 149, 150, 151, 152). Current data suggest a model in which: 1) agonists such as CRH bind with moderate affinity to the N-domain (residues 1–118) (100 nM); 2) interaction with the J-domain (residues 110–415) weakly stabilizes CRH binding (<2-fold); and 3) G protein binding strongly stabilizes CRH-J-domain interaction (>1000-fold). Nonpeptide antagonist affinity and the full antagonist effect are provided predominantly if not exclusively by the J-domain. The isolated N-domain binds peptide agonist ligands with affinity similar to that seen in the low-affinity, G protein-uncoupled state (R state) of the whole receptor. In contrast, the isolated J-domain mediates full high-affinity binding of nonpeptide antagonists and nearly full efficacy receptor activation by peptide agonists, as demonstrated by activation of the J-domain by a tethered N-terminal CRH fragment (133). A number of different regions within the N-terminal domain appear to be crucial for the binding of CRH-R1 receptor agonists and peptide antagonists (Fig. 7Go). Examples are the region mapping to amino acids 43–50 and a second amino acid sequence extending from position 76 to 84 of the human CRH-R1 (93). Within the latter sequence, Arg76 and Asn81, but not Gly83, appear to be important for determining receptor affinity for CRH-R1 agonists. A third amino acid cassette extending from position 68–109 of mouse CRH-R1 located in close proximity to the first TMD has also been proposed to play an important role for high-affinity agonist binding. Furthermore, agonists like CRH require interactions with the EC loops and TMD of CRH-R1 for high-affinity binding. Within the J-domain of the CRH-R1, three regions have been mapped as being important for optimal agonist binding. Two of the regions are within the second ECD, amino acids 175–178 and His189 at the junction of EC2 and TM3, whereas the third region is at the junction of EC3 and TM5 and involves three amino acid residues, Val266, Tyr267, and Thr268. Also, studies in the rat CRH-R1 revealed that the third EC loop is involved in ligand binding, especially the cassette Tyr346-Asn348 located in close proximity with 7 TMD. In contrast, the region Glu336-Glu338 containing negatively charged amino acids of EC3, does not appear to participate in ligand binding. These studies raise the possibility that during ligand binding the polar part of this region in the third EC loop interacts directly with the polar N terminus of CRH to develop the CRH high-affinity binding site. They also suggest that different agonists require different binding domains within the receptor binding pocket; for example, although the three regions mentioned above also affect the binding of UCN-I and sauvagine, a fourth region in the EC3, Asp254, has been identified to be important for sauvagine but not CRH or UCN-I binding (149).


Figure 7
View larger version (46K):
[in this window]
[in a new window]
 
FIG. 7. Amino acid regions, present in the N terminus and EC loops, important for agonist and peptide antagonist binding to CRH-R1. A number of different regions within the N-terminal domain appear to be crucial for the binding of CRH-R1 receptor agonists and peptide antagonists. Within the amino acid sequence 76–84 of the hCRH-R1 Arg76, Asn81 but not Gly83 appears to be important for determining receptor affinity for CRH-R1 agonists. Within the region at the junction of EC3/TM5, three amino acid residues appear important, Val266, Tyr267, and Thr268. The same is true for the cassette Tyr346-Asn348 located in close proximity with 7 TMD. These studies suggest that during ligand binding the polar part of this region in the third EC loop interacts directly with the polar N terminus of CRH to develop the CRH high-affinity binding site. Letters in parentheses indicate the species where CRH-R1 was obtained: r, rat; m, mouse; h, human.

 
C. Receptor-agonist high-affinity interaction: the role of posttranslational modifications and G proteins
The predicted CRH-R1 amino acid sequence contains multiple potential N-glycosylation sites: N38, N45, N78, N90, and N98, all of which appear to be N-glycosylated to a significant extent (154). Furthermore, this mechanism exhibits tissue-specific characteristics because the CRH-R1 is differentially glycosylated in different regions of the CNS (155). N-Glycosylation appears to be important for ligand binding, because the nonglycosylated CRH-R1 does not bind the radioligand. Although no single polysaccharide chain appears to be essential for binding, the loss of three or more polysaccharide chains significantly impairs normal ligand binding and CRH-R1 function (156).

As mentioned previously, the N termini of the B1 family of GPCRs contain multiple Cys residues forming multiple disulfide bonds (141). The CRH-R contains six conserved Cys in its N-terminal domain and one Cys in each of the first and second EC loops (Cys188 and Cys258), respectively. Additionally, several other Cys residues are located in the TMDs (Cys128, Cys211, Cys233, and Cys364) and the first IC loop (Cys150). These disulfide bonds appear to be critical for ligand recognition as shown by site-directed mutagenesis experiments (157).

The CRH-R, like all members of the GPCR family, has the intrinsic ability to couple to heterotrimeric GDP/GTP-bound proteins, G proteins, an association that stabilizes the receptor in an active high-affinity conformation. A number of studies have shown that the binding of different agonists to specific CRH-R subtypes exhibits varying degrees of sensitivity to G protein coupling. In particular, the CRH-R2 high-affinity state and binding to agonists, such as sauvagine, CRH, and UCNs, is not significantly altered by the presence or absence of receptor-G protein interaction (RG and R states, respectively) (143). In contrast, although the CRH-R1 N-domain predominantly contributes to peptide agonist affinity, the R-G coupling dramatically enhances agonist affinity via an allosteric effect (146). It has been proposed that the CRH-R1 J-domain determines R-G coupling sensitivity, and the change of agonist affinity produced by this coupling probably reflects a change in receptor conformation. Therefore, it is possible that G protein coupling produces different conformational changes within the J-domain of the CRH-R2 compared with the CRH-R1 receptor. Nonpeptide CRH-R1 antagonists, which bind the J-domain, were able to block peptide agonist binding to RG, whereas the binding of peptide antagonists, predominantly to the N-domain, was unaffected by R-G coupling. A naturally occurring model of these structural/functional relationships is the human-specific CRH-R1 variant, R1ß, which has reduced binding affinity for CRH-R1 agonists although its N- and J-domains are intact (158). Impaired R-G coupling due to the presence of the 29-amino acid insert in the first IC loop probably results in reduced ligand affinity of the CRH-R1.

D. Receptor-G protein interactions: implications for signaling
Binding of an agonist to the receptor induces a conformational change and receptor activation that causes the G protein (G{alpha}-subunit) to undergo an exchange of GDP (inactive state) for GTP (active state) (159, 160). Once GTP is bound to the G{alpha}-subunit, it dissociates from the Gß{gamma} dimer, allowing both species to activate a variety of signaling pathways. The ability of GPCRs to couple to different G protein heterotrimers, especially the G{alpha}-subunit, is critical for activation of downstream signaling cascades and induction of diverse cellular responses. The intracellular pathway activated is largely determined by the interaction of individual GPCRs with different G proteins. Within mammalian cellular systems, more than 20 different G{alpha}-subunits have been identified, which fall into four major classes: G{alpha}i/o, G{alpha}s, G{alpha}q, and G{alpha}12 according to sequence homologies (161). Early studies investigating the effects of CRH on pituitary ACTH release demonstrated that activation of pituitary CRH-Rs leads to a potent cAMP response through stimulation of Gs proteins (162). The structural features of Gs necessary for GPCR coupling, nucleotide binding, and adenylyl cyclase stimulation are primarily contained within the {alpha}-subunit (G{alpha}s), which also possesses an intrinsic GTP phosphohydrolase activity that limits the duration and strength of the signal (163).

It is now well accepted that most physiological functions of CRH in the CNS and the periphery involve CRH-R coupling to G{alpha}s proteins, although other G proteins undoubtedly play a role in some functions (164). Indeed, in certain tissues such as the testis and placenta (165, 166), the effects of CRH appear to be completely G{alpha}s-independent, despite adequate levels of endogenous G{alpha}s protein expression.

Characterization of recombinant CRH-R signaling properties in overexpression cellular systems, showed that both R1 and R2 primarily stimulate the adenylyl cyclase/cAMP pathway (45, 46) via coupling and activation of G{alpha}s proteins (90), including the XL{alpha}s (extra large) protein (167), a large variant of G{alpha}s protein derived from the same GNAS1 gene that is highly expressed in neuroendocrine tissues with particula