| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Article |
Digestive Diseases Branch, National Institute of Diabetes Digestive and Kidney Diseases, National Institutes of Health, Bethesda, Maryland (R.T.J.); Office of the Director, National Institute on Deafness and Other Communication Disorders, National Institutes of Health, Bethesda, Maryland (J.F.B.); Division of Neuroscience, Oregon Primate Research Center, Beaverton, Oregon (E.R.S.); and Division of Medicine and Pharmacology, University of Illinois at Chicago and Chicago Veterans Administration Medical Center (West Side Division), Chicago, Illinois (R.V.B.)
Abstract
The mammalian bombesin receptor family comprises three G protein-coupled heptahelical receptors: the neuromedin B (NMB) receptor (BB1), the gastrin-releasing peptide (GRP) receptor (BB2), and the orphan receptor bombesin receptor subtype 3 (BRS-3) (BB3). Each receptor is widely distributed, especially in the gastrointestinal (GI) tract and central nervous system (CNS), and the receptors have a large range of effects in both normal physiology and pathophysiological conditions. The mammalian bombesin peptides, GRP and NMB, demonstrate a broad spectrum of pharmacological/biological responses. GRP stimulates smooth muscle contraction and GI motility, release of numerous GI hormones/neurotransmitters, and secretion and/or hormone release from the pancreas, stomach, colon, and numerous endocrine organs and has potent effects on immune cells, potent growth effects on both normal tissues and tumors, potent CNS effects, including regulation of circadian rhythm, thermoregulation; anxiety/fear responses, food intake, and numerous CNS effects on the GI tract as well as the spinal transmission of chronic pruritus. NMB causes contraction of smooth muscle, has growth effects in various tissues, has CNS effects, including effects on feeding and thermoregulation, regulates thyroid-stimulating hormone release, stimulates various CNS neurons, has behavioral effects, and has effects on spinal sensory transmission. GRP, and to a lesser extent NMB, affects growth and/or differentiation of various human tumors, including colon, prostate, lung, and some gynecologic cancers. Knockout studies show that BB3 has important effects in energy balance, glucose homeostasis, control of body weight, lung development and response to injury, tumor growth, and perhaps GI motility. This review summarizes advances in our understanding of the biology/pharmacology of these receptors, including their classification, structure, pharmacology, physiology, and role in pathophysiological conditions.
The unusual name of this family of receptors, bombesin (Bn1), comes from the original terminology used by V. Erspamer and his colleagues to name the first natural ligand described, bombesin, which was an amidated tetradecapeptide isolated from the skin of the European frog Bombina bombina (Erspamer et al., 1970
, 1972) (Fig. 1). They isolated many related peptides from other frog skins, and most were named after the genus of frog from which they were isolated (Erspamer and Melchiorri, 1973; Erspamer, 1988). In terms of their structural similarities they were originally divided into three general groups (Fig. 1): the bombesin group, which all had a carboxyl terminus of Gly-His-Leu-Met-NH2 (bombesin, alytesin, and [pGlu1]bombesin6–14); the ranatensin group, which had a carboxyl terminus of Gly-His-Phe-Met-NH2 (ranatensin, ranatensin R and C, litorin, rodhei-litorin, and [Glu (Ote)2 or (Ome)2]litorin); and the phyllolitorin group, which had a carboxyl-terminal Gly-Ser-Phe/Leu-Met-NH2 (phyllolitorin, [Leu8]phyllolitorin, and [Thr5,Leu8]phyllolitorin) (Erspamer, 1988; Falconieri Erspamer et al., 1988) (Fig. 1). Recent molecular studies show that the occurrence of these peptides in amphibian skins is more complicated than originally thought with both Leu and Phe penultimate forms present in the same frog species in many cases (Nagalla et al., 1996; Spindel, 2006). For example, in the skin of the frog, Bombina orientalis [Leu13]bombesin, [Phe13]bombesin, and [Ser3,Arg10,Phe13]bombesin (SAP bombesin) are found, and each of these three forms is derived from separate genes (Nagalla et al., 1996; Spindel, 2006).
|
Subsequently, in mammals two Bn-like peptides were isolated, gastrin-releasing peptide (GRP) (McDonald et al., 1979) and neuromedin B (NMB) (Minamino et al., 1983). GRP, a 27-amino acid peptide was originally isolated from porcine stomach and shares the same seven carboxyl-terminal amino acids with bombesin (McDonald et al., 1979) accounting for similar biological activity (Fig. 1). The decapeptide of GRP was later isolated from porcine spinal cord and originally called neuromedin C (Minamino et al., 1984b), although it is recommended that a more appropriate name is either GRP-10 or GRP18–27 (Anonymous, 1988). The mammalian equivalent of ranatensin, NMB, was isolated from porcine spinal cord and shown to be a decapeptide (Minamino et al., 1983), which also occurs in precursor forms of 30 and 32 amino acids (Minamino et al., 1985). The carboxyl-terminal seven amino acids are identical in ranatensin, except for the replacement of threonine in NMB for valine in ranatensin at the fifth position from the carboxyl terminus (Fig. 1).
Studies of GRP and NMB immunoreactivity as well as mRNA studies have demonstrated that these peptides and their mRNA are widely distributed in mammals in both the nervous system and peripheral tissues, especially the gastrointestinal tract (Penman et al., 1983; Wada et al., 1990; Battey and Wada, 1991; Spindel et al., 1993; Moody and Merali, 2004). In the alimentary tract GRP-like IR is found primarily in neurons as well as in the submucosal and myenteric plexuses and not in endocrine cells (Penman et al., 1983). With Northern blots the highest levels of mRNA occur in the colon with lower amounts in the stomach and small intestine (Sunday et al., 1988). In the spinal cord GRP-IR was found in both the posterior and anterior horn, and in the CNS GRP-IR and mRNA are widely distributed in neurons with high levels in the hypothamic nuclei, forebrain, and medullary nuclei that participate in autonomic functions, as well as in sensory nuclei (Panula et al., 1982, 1988; Wada et al., 1990; Battey and Wada, 1991; Spindel et al., 1993). NMB-IR and mRNA are found throughout the GI tract, but generally at lower levels than GRP except in the esophagus (Spindel et al., 1993). In general, in the brain and spinal cord, NMB-IR is greater than GRP-IR (Minamino et al., 1984a), and NMB mRNA is most abundant in the olfactory bulb, dentate gyrus, and dorsal root ganglia, whereas GRP mRNA is highest in the forebrain and some hypothamic nuclei (Wada et al., 1990; Battey and Wada, 1991). In most brain regions the NMB mRNA distribution does not overlap with GRP (Wada et al., 1990; Battey and Wada, 1991; Moody and Merali, 2004; Ohki-Hamazaki et al., 2005).
The mammalian bombesin peptides, GRP and NMB, demonstrate a broad spectrum of pharmacological and biological responses. GRP stimulates smooth muscle contraction in both the gastrointestinal tract and urogenital system and has profound effects on GI motility, stimulates release of numerous gastrointestinal hormones/neurotransmitters, stimulates secretion and/or hormone release from the pancreas, stomach, colon, and numerous endocrine organs, has potent effects on immune cells (macrophages, dendritic cells, lymphocytes, and leukocytes) (Ruff et al., 1985; De la Fuente et al., 1991, 1993; van Tol et al., 1993; Del Rio and De la Fuente, 1994; Del Rio et al., 1994; Plaisanci é et al., 1998; Makarenkova et al., 2003), has potent growth effects on both normal tissues and tumors; has potent CNS effects, including regulation of circadian rhythm, thermoregulation; regulation of anxiety and the fear response, regulation of food intake, and behavioral effects and is involved in mediating numerous CNS effects on the GI tract (Tache et al., 1988; Bunnett, 1994; Martinez and Tache, 2000; Jensen et al., 2001; Jensen, 2003; Grider, 2004; Jensen and Moody, 2006). In many tissues the effects of NMB overlap with those of GRP; however, NMB has specific effects in some tissues such as contraction of smooth muscle, growth effects in various tissues (Moody et al., 2000; Matusiak et al., 2005), CNS effects including effects on feeding, thermoregulation; regulation of TSH release, stimulation of various CNS neurons, behavioral effects; and effects on spinal sensory transmission (von Schrenck et al., 1989; Rettori et al., 1992; Ladenheim et al., 1997b; Ohki-Hamazaki, 2000; Merali et al., 2006; Oliveira et al., 2006). GRP and to a lesser extent NMB affects the growth and/or differentiation of a number of important human tumors including colon, prostate, lung, and some gynecologic cancers (Cuttitta et al., 1985; Schally et al., 2000; Jensen et al., 2001; Glover et al., 2003; Jensen and Moody, 2006).
Early studies on the biologic effects of the different bombesin peptides isolated from frog skins, primarily examining their effects on contraction of isolated smooth muscle preparations from various tissues, demonstrated markedly varying potencies, which suggested that more than one subtype of bombesin receptor might exist (Falconieri Erspamer et al., 1988; Regoli et al., 1988; Severi et al., 1991). Binding studies and the development of highly selective antagonists established unequivocally the existence of two different classes of receptors in mammalian tissues mediating the actions of these peptides (Jensen et al., 1978; Moody et al., 1978; Jensen and Gardner, 1981; Coy et al., 1988; von Schrenck et al., 1989, 1990; Ladenheim et al., 1990; Jensen and Coy, 1991; Metz et al., 1992). One class had a high affinity for GRP and a lower affinity for NMB (termed GRP-R, GRP receptor, or GRP-preferring receptor) and the other class had a higher affinity for NMB than for GRP (termed NMB-R, NMB receptor, or NMB-preferring receptor) (Jensen and Gardner, 1981; Moody et al., 1988, 1992; von Schrenck et al., 1989, 1990; Ladenheim et al., 1990, 1992; Wang et al., 1992). Subsequently, two mammalian receptors with high affinity for GRP (Spindel et al., 1990; Battey et al., 1991) or NMB (Wada et al., 1991) have been cloned in addition to a closely related orphan receptor (Gorbulev et al., 1992; Fathi et al., 1993b) and one related receptor from amphibians (Nagalla et al., 1995), which will be discussed in more detail below (Table 1).
|
II. Molecular Basis for Nomenclature
Once the receptors were defined using binding studies, cross-linking studies, and studies of biological activity (Kris et al., 1987; Sinnett-Smith et al., 1988; Tache et al., 1988; von Schrenck et al., 1989; Huang et al., 1990; Ladenheim et al., 1990; Lebacq-Verheyden et al., 1990), an active effort to clone the GRP-R was undertaken by Dr. Eliot Spindel, Oregon Regional Primate Center, and Dr. James Battey, National Institutes of Health. In 1990 using electrophysiological and luminometric Xenopus oocyte expression assays, Spindel et al. (1990) succeeded in cloning the GRP-R from murine Swiss 3T3 cells, which express high levels of this receptor (Rozengurt, 1988). The cDNA for the same receptor was isolated and described by Battey et al. in 1991 by using an enriched library from Swiss 3T3 cells and specific oligonucleotide probes on the basis of information from a partial sequence of the GRP-R in these cells obtained after solubilization and purification using wheat germ agglutinin-agarose and ligand affinity chromatography (Feldman et al., 1990). Pharmacology studies demonstrated that the cloned receptor preferred GRP to NMB and its activation was blocked by specific GRP-preferring receptor antagonists (Rozengurt, 1988; Battey et al., 1991). Subsequently, using low stringency conditions with a mouse GRP-R cDNA probe (Wada et al., 1991), the NMB-R was cloned from a cDNA library made from the rat esophagus, a tissue that had been reported to have a high density of NMB-Rs (von Schrenck et al., 1989, 1990). The structure of the cDNA of the human GRP-R and NMB-R were described from a small cell lung cancer cell line in 1991 (Corjay et al., 1991).
In 1992 a novel receptor was cloned from guinea pig uterus (Gorbulev et al., 1992), which showed the highest amino acid identity to the GRP-R (52%) and the NMB-R (47%). This receptor bound GRP and NMB, but only with relatively low affinities (IC50 of 290 and 20,000 nM, respectively). The human analog of this novel receptor was cloned in 1993 (Fathi et al., 1993b), and expression studies showed that it was specifically activated by bombesin-related peptides but only with low affinity and thus was classified as an orphan receptor. It was termed BRS-3 for bombesin receptor subtype 3 (Fathi et al., 1993b). Subsequent binding studies and signaling studies using synthetic ligands of bombesin with high affinity for hBRS-3 (Mantey et al., 1997) demonstrated that it had low affinity not only for GRP and NMB but also for all known naturally occurring bombesin related peptides (Wu et al., 1996; Mantey et al., 1997; Pradhan et al., 1998; Ryan et al., 1998a,b) and therefore it remains an orphan receptor. Subsequently it was cloned from mouse (Ohki-Hamazaki et al., 1997a), rat (Liu et al., 2002), and sheep (Whitley et al., 1999).
In the search for receptors for bombesin-related peptides in amphibians (Nagalla et al., 1995), clones that had a sequence similar to the mammalian GRP-R and NMB-R were isolated. A clone that encoded for a novel bombesin receptor, which had 61, 56, and 70% amino acid identities to the human GRP-R, NMB-R, and BRS-3, respectively, was isolated (Nagalla et al., 1995). This receptor had the highest affinity for [Phe13]-bombesin, the form most prevalent in frog brain and had lower affinity for GRP and NMB. This receptor was called BB4 for bombesin receptor subtype 4 (Nagalla et al., 1995). Subsequent detailed binding studies and studies of cell signaling confirmed these findings and showed that this receptor had greater affinity for [Phe13]bombesin than any other naturally occurring bombesin-related peptide (Katsuno et al., 1999). At present no mammalian equivalent of this receptor has been described and therefore it is not included in the classification discussed in the following sections. Recently in chickens a receptor was cloned that had high amino acid identity to frog BB4 (fBB4) (70%) as well as to human BRS-3 (69%) and lower for human GRP-R (58%) and human NBR-R (52%) (Iwabuchi et al., 2003). When expressed this receptor had low affinity for GRP and NMB, but it retained high affinity for [D-Phe6,β-Ala11,Phe13,Nle14]bombesin6–14 (Iwabuchi et al., 2003), a synthetic analog which has high affinity for hBRS-3, GRP-R, NMBR, and fBB4 (Mantey et al., 1997; Pradhan et al., 1998). It was proposed that this receptor be termed chBRS-3.5 because of its resemblance to both fBB4 and BRS-3. No mammalian equivalent of this receptor has been described and therefore it is also not included in the following classification.
On the basis of the preceding molecular studies, three classes of mammalian bombesin receptors are proposed for which the nomenclature and a few features are summarized in Table 1. Although the usual International Union of Pharmacology Committee on Receptor Nomenclature and Drug Classification (NC-IUPHAR) nomenclature uses the endogenous mammalian ligand, the substantial historical use of the frog peptide bombesin in the field to describe this system was retained. The BB1 through BB3 receptors will each be dealt with in more detail in the following sections, but a few important points will be covered briefly here. The BB1 receptor was previously referred to as the NMB receptor, NMB-R, or NMB-preferring receptor. This terminology is the same used for this bombesin receptor subclass in the Sigma-RBI Handbook of Receptor Classification and Signal Transduction (Watling, 2007) and is the same as the BB1 in the "BJP Guide to Receptors and Channels" (Alexander et al., 2006). The BB2 receptor was previously referred to as the GRP-R, GRP receptor, or GRP-preferring receptor (Table 1). This terminology is the same used for this bombesin receptor subclass in the Sigma-RBI Handbook of Receptor Classification and Signal Transduction (Watling, 2007) and is the same as the BB2 subclass in the "BJP Guide to Receptors and Channels" (Alexander et al., 2006). The BB3 receptor was previously referred to as the BRS-3 receptor, BRS-3, and bombesin receptor subtype 3 (Table 1). This terminology is the same used for this bombesin receptor classes in the Sigma-RBI Handbook of Receptor Classification and Signal Transduction (Watling, 2007) and is the same as the bb3 receptor in the "BJP Guide to Receptors and Channels" (Alexander et al., 2006). Finally, the amphibian BB4 receptor does not have a mammalian equivalent so is not included in this classification. This receptor was also not classified in the Sigma-RBI Handbook of Receptor Classification and Signal Transduction (Watling, 2007) or the "BJP Guide to Receptors and Channels" (Alexander et al., 2006).
A. Early Studies of the BB1 Receptor
Before the identification of the BB1 in 1989 in rat esophageal muscle tissue sections by direct binding studies using 125I-Bolton-Hunter-labeled NMB and subsequent esophageal muscle strip contraction studies (von Schrenck et al., 1989), there were no early studies that unequivocally established the existence of BB1. Numerous previous studies had demonstrated that the frog peptides ranatensin and litorin, which closely resembled NMB (Minamino et al., 1983), had potent effects on various tissues and especially on smooth muscle contraction, which in some classes had differences from bombesin (Falconieri Erspamer et al., 1988; Regoli et al., 1988). However, these differences were not significant enough to clearly establish the existence of a separate class of BB1 receptors (Minamino et al., 1983; Falconieri Erspamer et al., 1988; Regoli et al., 1988). Although there had been many binding studies to numerous tissues from the late 1970s, in almost all cases 125I-[Tyr4] bombesin or another radiolabeled bombesin analog was used (Moody et al., 1978; Ladenheim et al., 1993b; Shapira et al., 1993). Unfortunately, bombesin has high affinity for both BB1 and BB2, making it more difficult to distinguish subtypes. Numerous classes of selective BB2 receptor antagonists were developed before the cloning of the BB1, and these also confirmed the presence of the BB1 on esophageal smooth muscle (von Schrenck et al., 1990). After the pharmacologic description of BB1 on esophageal muscle and before its cloning in 1991, by use of selective BB2 receptor antagonists or binding studies with radiolabeled NMB and selective agonists or BB2 receptor antagonists, BB1 receptors were demonstrated in the CNS (Ladenheim et al., 1990) and on gastric smooth muscle cells (Severi et al., 1991).
B. Cloned BB1 Receptor and Receptor Structure
The human BB1 receptor is a 390-amino acid protein, and it shows an 89% amino acid identity with the rat BB1 (Corjay et al., 1991). The human BB1 receptor has 55% amino acid identities with the human BB2 (Corjay et al., 1991) and 47% with the human BB3 receptor (Fathi et al., 1993b). The human BB1 receptor has two consensus sites for potential PKC phosphorylation and three potential N-linked glycosylation sites (Corjay et al., 1991). Hydropathy plots yielded results consistent with a seven-transmembrane structure typical for a G protein-coupled receptor (Corjay et al., 1991). The BB1 receptor has been cloned from rat (Wada et al., 1991) (Fig. 2), mouse (Ohki-Hamazaki et al., 1997a), and the frog, B. orientalis (Nagalla et al., 1995). Cross-linking studies demonstrate that the mature human BB1 receptor had a molecular mass of 72 ± 1 kDa and when deglycosylated 43 ± 1 kDa (Benya et al., 1995b). Detailed cross-linking and serial deglycosylation studies using enzymatic digestion in the rat BB1 receptor demonstrated a molecular mass of 63 kDa in the membrane and showed that there were no O-linked carbohydrates, but that the mature BB1 receptor was a sialoprotein (Kusui et al., 1994). However, each of the potential N-linked glycosylation sites was, in fact, glycosylated, with tri-antennary and/or tetra-antennary complex oligosaccharide chains (Kusui et al., 1994).
|
The human BB1 receptor gene is localized at human chromosome 6p21-qter and in the mouse on chromosome 10 (Table 1). Both the human, rat, and mouse genes contained three exons with two introns (Corjay et al., 1991; Wada et al., 1991; Ohki-Hamazaki et al., 1997a; Ohki-Hamazaki, 2000). In the mouse the gene for the BB1 receptor spanned more than 10 kb with exon 1 of the BB1 gene separated from exon 2 by 6 kb, and this in turn is separated from exon 3 by 3 kb (Ohki-Hamazaki et al., 1997a). In human and mouse the first intron of the BB1 gene was located between transmembrane domains 3 and 4 and the second between transmembrane domains 5 and 6 (Corjay et al., 1991; Ohki-Hamazaki et al., 1997a). The first intron interrupted a codon for arginine located immediately COOH terminal to the transmembrane domain 3, and the second intron was located between glutamine and methionine codons in both the mouse and human BB1 gene (Corjay et al., 1991; Ohki-Hamazaki et al., 1997a). The positions of the first and second introns were identical in the mouse and human BB1 receptor gene (Corjay et al., 1991; Ohki-Hamazaki et al., 1997a).
Expression levels of BB1 receptor mRNA have been reported in human, mouse, rat, and monkey (Corjay et al., 1991; Wada et al., 1991; Ohki-Hamazaki et al., 1997a; Sano et al., 2004). In the monkey, in which it was studied in detail, the highest levels of BB1 mRNA are found in the CNS and in the testis (Sano et al., 2004). In the CNS the BB1 receptor was expressed widely in different brain regions including the amygdala, caudate nucleus, hippocampus, hypothalamus, thalamus, brain-stem, spinal cord, and peripheral tissues in addition to the testis and the stomach, which is a similar distribution to that found in rats and mice (Wada et al., 1991; Ohki-Hamazaki et al., 1997a; Ohki-Hamazaki, 2000; Sano et al., 2004). In the rat and mouse, BB1 mRNA is present in high amounts in the olfactory region and esophagus (Wada et al., 1991; Ohki-Hamazaki et al., 1997a). Binding studies and studies of biological activity provide evidence for BB1 on both gastrointestinal and urogenital smooth muscle cells (von Schrenck et al., 1989; Severi et al., 1991; Bitar and Coy, 1992; Kim et al., 1993). Binding studies have confirmed the widespread distribution of BB1 in the brain showing especially high levels in the olfactory tract of the rat (Ladenheim et al., 1990, 1992, 1993a).
Using binding studies and/or assessment of BB1 mRNA, BB1 receptors have been shown to exist on a large number of different tumors (Reubi et al., 2002; Jensen and Moody, 2006) including CNS tumors (glioblastomas) (Wada et al., 1991; Wang et al., 1992), small cell and non-small cell lung cancers (Corjay et al., 1991; Moody et al., 1992, 2000; Toi-Scott et al., 1996; Siegfried et al., 1997; Jensen and Moody, 2006), carcinoids (intestinal, thymic, and bronchial) (Reubi et al., 2002), human ovarian epithelial cancers (Sun et al., 2000b), and pancreatic cancer cell lines (Jensen and Moody, 2006).
1. BB1 Receptor Agonists.
The human BB1 receptor (Moody et al., 1992; Benya et al., 1995b; Reubi et al., 2002) as well as the rat BB1 receptor (von Schrenck et al., 1989, 1990; Wang et al., 1992; Ladenheim et al., 1992, 1993a) has a >100-fold higher affinity for NMB than for GRP (Fig. 1, Table 1). Bombesin and the frog peptides, ranatensin and litorin, also had relatively high affinity for the BB1 receptor (affinities 1- to 10-fold less than those for NMB) (Wang et al., 1992; Mantey et al., 1997; Katsuno et al., 1999) (Tables 1 and 2). The synthetic bombesin analog [D-Phe6,β-Ala11,Phe13,Nle14]-bombesin6–14 (Mantey et al., 1997), which has high affinity for the human BB3 receptor also has a high affinity for the human BB1 receptor as well as the human BB2 receptor and fBB4 (Mantey et al., 1997; Pradhan et al., 1998) (Table 2).
|
2. BB1 Receptor Antagonists.
Whereas the search for high-affinity receptor antagonists for the BB2 receptor has been very successful (section IV.E.1.) (Jensen and Coy, 1991; Jensen et al., 1993; de Castiglione and Gozzini, 1996), results with the BB1 receptor have been much less successful and only a few high-affinity receptor antagonists are available. None of the strategies used for making high-affinity BB2 antagonists were successful with the BB1 receptor, including the synthesis of bombesin or NMB COOH-terminal pseudopeptide analogs, COOH-terminal truncated analogs or [des-Met10]-NMB amides, alkylamides, or esters (Lin et al., 1995). Subsequently, it was discovered that certain substituted somatostatin analogs selectively antagonized the BB1 receptor compared with the BB2 receptor (Orbuch et al., 1993). The most potent analog was cyclo-somatostatin-octa[D-Nal-Cys-Tyr-D-Trp-Lys-Val-Cys-Nal-NH2], which had a 100-fold higher affinity for the BB1 receptor than the BB2 receptor (Ki 230 versus 3000 nM) (Orbuch et al., 1993; Ryan et al., 1999) (Table 2). Unfortunately this analog also interacted with high affinity with somatostatin receptors (IC50 0.80 nM) and µ-opioid receptors (IC50 430 nM) (Orbuch et al., 1993). Substitution of an ornithine for Lys greatly reduced the affinity for somatostatin receptors, and a related analog (BIM-23127) inhibited NMB cell signaling in rat BB1 receptor transfected Rat-1 cells (Lach et al., 1995) and selectively reversed NMB feeding suppression, but had no effect on the action of GRP (Ladenheim et al., 1997b). However, a recent study reported that BIM-23127 also functions as a receptor antagonist of both human and rat urotensin-II receptors (Herold et al., 2003), limiting its utility. Peptoid antagonists of BB1 have been described, including PD 165929 (Eden et al., 1996) and PD 168368 (Ryan et al., 1999), which have high affinity and selectivity for BB1. In a detailed comparison of bombesin receptors from different species, PD 168368 was found to have a similar high affinity (Ki 15–45 nM) for BB1 receptors from each species, a 30- to 60-fold lower affinity for the BB2 receptor from different species, and a >300-fold lower affinity for the BB3 receptor or fBB4 (Ryan et al., 1999) (Table 2). It also inhibited NMB-stimulated cellular signaling in a competitive manner (Ryan et al., 1999) as well as inhibiting NMB-induced proliferation of rat C6 glioblastoma cells (Moody et al., 2000) and NMB stimulation of NCI-H1299 lung cancer cell proliferation (Moody et al., 2000).
F. BB1 Receptor Structural Basis of Receptor Binding/Activation
1. BB1 Receptor Agonist Binding/Activation.
Structure-function studies of NMB demonstrated that the COOH-terminal octapeptide is the minimal peptide length required for BB1 receptor activation and the full decapeptide was required for full affinity for the BB1 receptor (Lin et al., 1996). NMB differs from GRP in the COOH octapeptide, which is the biologically active end (Broccardo et al., 1975; Lin et al., 1996), at three residues: substitution of a leucine in NMB for a histidine in GRP at position 3, a threonine for valine at position 6, and a phenylalanine for leucine at position 9 of NMB from the amino terminus (Minamino et al., 1983; Lin et al., 1996) (Fig. 1). Structure-function studies of all naturally occurring bombesin-related peptides for BB1 and BB2 receptors suggested the presence of the phenylalanine instead of leucine, as the penultimate amino acid from the COOH terminus in NMB was not important for selectivity for the BB1 receptor, Single amino acid substitutions in NMB demonstrated the Leu for His substitution in position 3 was the most important for determining high affinity and selectivity for the BB1 receptor (Lin et al., 1996) (Fig. 1).
A chimeric receptor approach (Fathi et al., 1993a) and homology screening after computer alignment of bombesin receptor family members (Sainz et al., 1998), followed by site-directed mutagenesis studies, have been used to explore the molecular basis of NMB high affinity and selectivity for the BB1 receptor over the BB2 receptor (Fig. 3). A study of BB1/BB2 chimeric receptors (Fathi et al., 1993a) demonstrated that differences in the amino terminus of the two receptors were of minimal importance for high-affinity NMB interaction. High affinity and selectivity for the BB1 receptor were primarily determined by differences in transmembrane (TM) domain 5 (Fathi et al., 1993a) (Fig. 3). Site-directed mutagenesis of the amino acid differences between the BB1 receptor and the BB2 receptor in this region demonstrated that the substitution of an Ile216 instead of Ser in the comparable position of the TM5 of the BB2 receptor was the critical difference accounting for high-affinity NMB interaction with the BB1 and not the BB2 receptor (Fathi et al., 1993a). A second study (Sainz et al., 1998) used a different approach to select potentially important amino acids for NMB selectivity for the BB1 receptor and further study. Using amino acid sequence alignment of bombesin receptor family members and identifying conserved amino acids in members with similar peptide affinities (Akeson et al., 1997), four amino acids were identified that could be important for high-affinity bombesin binding to either the BB1 or BB2 receptor (Akeson et al., 1997) (i.e., in the BB1 receptor, Gln123, Pro200, Arg290, and Ala310, and in the BB2 receptor, Gln121, Pro199, Arg288, and Ala 308). Possible gain-of-affinity mutants were made in the BB3 receptor, which has a low affinity for NMB (Mantey et al., 1997; Ryan et al., 1998a,b), by substituting alone or in combination each of these four BB1 receptor amino acids for the comparable amino acid(s) of the BB3 receptor (Arg127, Ser205, His294, and Ser315) (Fig. 3). It was found that each of these four amino acids is important for determining NMB affinity because the affinities for NMB of the BB3 mutants with these BB1 receptor amino acids substituted one at a time were increased (Sainz et al., 1998). The substitution of all four amino acids for the comparable amino acids in the BB3 receptor, which has a very low affinity for NMB (i.e., Ki 3450 nM), increased the affinity and the potency for NMB, almost up to that seen with the native BB1 receptor (Sainz et al., 1998). This study helped to define the binding pocket for NMB by identifying four amino acids needed for high-affinity NMB interaction in markedly different BB1 regions [transmembrane domain 2 (Gln123), extracellular domain 2 (Pro200), extracellular domain 3 (Arg290), and transmembrane region 7 (Ala310)] (Fig. 3) (Sainz et al., 1998).
|
2. BB1 Receptor Antagonist Binding.
Using a chimeric receptor approach combined with site-directed mutagenesis and receptor modeling, the molecular basis of selectivity of the BB1 receptor antagonist, PD 168368 was studied (Tokita et al., 2001a) (Fig. 3). PD 168368 is a new class of antagonists described as a peptoid, because this group of antagonists are nonpeptide ligands, which were designed using the chemical structure of the mammalian neuropeptide of interest as a stating point (Horwell et al., 1994; Horwell, 1995). This approach has yielded antagonists for cholecystokinin, somatostatin, tachykinins, and bombesin receptors (Boden et al., 1993; Boyle et al., 1994; Horwell et al., 1994; Horwell, 1995; Eden et al., 1996; Tran et al., 1998; Tokita et al., 2001a). However, little is known about the molecular basis of their affinity and whether they resemble peptide or other nonpeptide ligands in the basis of their selectivity and affinity (Tokita et al., 2001a). The receptor extracellular domains were shown not to be important for the selectivity of PD 168369 by studying both loss-of-affinity BB1 receptor chimeras in which the extracellular domains of the BB1 were replaced by those from BB2, one at a time or the reverse study performed by making PD 168368 gain-of-affinity chimeras in the BB2 receptor (Tokita et al., 2001a). Additional PD 168368 loss- and gain-of-affinity chimeric studies made by exchanging the upper transmembrane regions of BB1 and BB2 receptors showed that differences in the upper TM5 were the key determinants of selectivity of PD 168368 (Tokita et al., 2001a). Site-directed mutagenesis studies of the different amino acids between the BB1 receptor and the BB2 receptor in the upper TM5 region demonstrated that the substitution of Tyr at position 220 of BB1 for Phe in the comparable position in BB2 was the critical difference (Tokita et al., 2001a) (Fig. 3). Three-dimensional modeling studies showed the critical Tyr220 was facing the interior of a large binding pocket formed primarily by transmembrane domains 3 to 7 and minimum energy conformation of the ligand showed that it was dominated by a large hydrogen bond-accepting region around the nitrophenyl group (Tokita et al., 2001a). It was concluded that the Tyr220 hydroxyl group of the BB1 receptor was critical for interacting with the nitrophenyl group of PD 168368, probably primarily by hydrogen bonding. This result showed that the binding of this peptoid antagonist was similar to that reported with other nonpeptide antagonists, in that it was primarily dependent on interaction with transmembrane regions (Tokita et al., 2001a).
G. BB1 Receptor Signaling, Activation, and Modulatory Processes (Internalization, Down-Regulation, and Desensitization)
The human BB1 receptor (Moody et al., 1986, 1992, 1995a; Corjay et al., 1991; Benya et al., 1995b), as well as the rat BB1 receptor (Wada et al., 1991; Jones et al., 1992; Wang et al., 1992; Dobrzanski et al., 1993; Lach et al., 1995; Akeson et al., 1997; Tsuda et al., 1997b; Vigne et al., 1997; Hou et al., 1998) is coupled to phospholipase C, resulting in breakdown of phosphoinositides, mobilization of cellular calcium, and activation of protein kinase C. BB1 receptor activation also results in the stimulation phospholipase A2 (Moody et al., 1995a) and phospholipase D by a PKC-dependent and -independent mechanism (Tsuda et al., 1997b) but does not activate adenylate cyclase (Benya et al., 1992). BB1 receptor stimulation also results in activation of tyrosine kinases (Lach et al., 1995; Tsuda et al., 1997b) stimulating tyrosine phosphorylation of p125FAK by a phospholipase C-independent mechanism that requires p21rho and the integrity of the actin cytoskeleton (Tsuda et al., 1997b). BB1 receptor activation also stimulated tyrosine phosphorylation of paxillin and MAP kinase activation (Lach et al., 1995). The native and transfected rat BB1 receptor in BALB 3T3 cells have been shown to behave in a similar manner in their binding and signaling cascades (Benya et al., 1992), demonstrating the usefulness of this cell line for studying BB1 receptor interaction and signaling.
The BB1 receptor is coupled to heterotrimeric guanine-nucleotide binding proteins in both native and BALB 3T3-transfected cells (Benya et al., 1992; Wang et al., 1993). In an Xenopus oocyte assay with the injection of antisense oligonucleotides, G
q was identified as a mediator of the BB1 receptor response (Shapira et al., 1994). With an in situ reconstitution assay with purified G protein subunits, it was found that cells expressing the BB1 receptor activated Gq, but not Gt or Gi/o (Jian et al., 1999). This activation was enhanced by β
dimers with a relative potency of β > β1 2 >> β11. In this study (Jian et al., 1999), these results were contrasted with those for the BB2 receptor, and differences were found in their kinetics of activation and preference for Gq proteins from different sources and for β dimers, demonstrating distinct coupling mechanisms for these two closely related receptors (Jian et al., 1999).
In contrast with the BB2 receptor there have been few studies of BB1 receptor modulatory processes (internalization, down-regulation, or desensitization). Both the human (Benya et al., 1995b) and rat BB1 receptors (Benya et al., 1992, 1994c; Wang et al., 1993) are rapidly internalized with receptor activation of the BB1 receptor. The rat BB1 receptor internalized 60 to 80% of the bound ligand, and human BB1 receptors internalized 70% of the bound ligand. In addition to being rapidly internalized by BB1 receptor-bearing cells, the ligand is rapidly degraded by these cells (Benya et al., 1992; Wang et al., 1993). Protease inhibitors markedly decreased ligand degradation by either rat native or rat BB1 receptor-transfected BALB 3T3 cells (Benya et al., 1992; Wang et al., 1993) with the acid proteinase inhibitor, leupeptin being the most potent followed by bacitracin > chymostatin > phosphoramidon >> bestatin and amastatin. The BB1 receptor also undergoes desensitization, which is mediated by receptor down-regulation and internalization (Benya et al., 1994c). Preincubation for 3 h with 3 nM NMB markedly attenuated the ability of a maximally effective concentration of NMB (1 µM) to subsequently stimulate either native or BB1-transfected BALB 3T3 cells but did not alter the response to other stimulants (Benya et al., 1994c). This desensitization was associated with a rapid decrease in BB1 receptors due to internalization of the receptors. Restoration of receptor number and response recovered over a 6-h period, and it was not dependent on new protein synthesis but was due to receptor recycling, because it was inhibited by the recycling inhibitor, monesin, a monocarboxylic acid cation ionophore (Benya et al., 1994c).
H. BB1 Receptor Function in Various Tissues and in Vivo
One of the main difficulties in assessing the effects of BB1 receptor activation in the CNS as well as in peripheral tissues, especially in older studies, is that bombesin was frequently used as the agonist, and it interacts with both BB1 and BB2 receptor with relatively high affinity. Furthermore, many tissues possess both BB1 and BB2 receptors, and therefore it was difficult to assess whether a particular response was due to activation of the BB1 or BB2 receptors present.
Numerous effects of NMB in both in vivo and in vitro studies have been reported, but it is not clear in many cases which are physiological and which are pharmacological. Studies comparing the potencies of NMB to GRP as well as binding studies or antagonist studies provide evidence that the BB1 receptor can stimulate contraction of urogenital and gastrointestinal smooth muscle (esophageal, gastric, colon, and gallbladder) (Regoli et al., 1988; von Schrenck et al., 1989, 1990; Severi et al., 1991; Kilgore et al., 1993; Parkman et al., 1994; Milusheva et al., 1998), potently inhibit thyrotropin release from the pituitary gland by acting as an autocrine and paracrine regulator (Rettori et al., 1992; Pazos-Moura et al., 1996; Ortiga-Carvalho et al., 2003), and have potent CNS effects including inhibiting food intake independent of BB2 stimulation (Ladenheim et al., 1994, 1996b, 1997b; Merali et al., 1999; Ladenheim and Knipp, 2007) and mediating aspects of the stress and fear responses as well as various behaviors such as spontaneous activity (Merali et al., 2002, 2006).
BB1 receptor knockout mice are now available and have undergone a limited number of investigations for actions of NMB (Ohki-Hamazaki et al., 1999; Oeffner et al., 2000; Yamada et al., 2002b, 2003; Yamano et al., 2002) (Table 1). In these mice the hypothermic effect of NMB was reduced by 50% without a change in the GRP response, supporting a possible BB1 receptor-mediated role in thermoregulation: NMB-mediated gastric smooth muscle contraction was not affected, suggesting this is mediated not through BB1 receptors, and no effect on feeding could be confirmed, although NMB did not have an effect in the control animals (Ohki-Hamazaki et al., 1999). The satiety effects of the BB1 receptor are mediated through peripheral neural pathways different from those mediating the satiety effects of the BB2 receptor, because only the satiety effects of BB1 receptors are inhibited by capsaicin treatment, suggesting the involvement of primary sensory afferent neurons (Ladenheim and Knipp, 2007). Recently, NMB has found to be expressed in human and rodent adipose tissue and to be regulated by changes in energy balance. It was proposed that because of the known anorectic effects of NMB centrally, it may form part of a new adipose tissue-hypothalamic regulating system for food intake (Hoggard et al., 2007). In BB1 receptor knockout mice dysregulation of the thyroid occurred, suggesting that BB1 receptor pathways are significantly involved in both TSH gene regulation and function (Oliveira et al., 2006), dysfunction in response to stress was seen (Yamada et al., 2002b; Yamano et al., 2002), impairment in the modulation of the CNS 5-HT system in response to stress occurred (Yamano et al., 2002), and an impairment of learning and memory was seen (Yamada et al., 2003). The alterations in the CNS 5-HT and stress in these animals is particularly interesting, because the dorsal raphe nucleus is one of the brain regions that has a preponderance of BB1 receptors (Wada et al., 1990; Ladenheim et al., 1992; Pinnock et al., 1994; Merali et al., 2006), which are located on 5-HT neurons, and stimulation of this nucleus by NMB stimulates release of 5-HT, resulting in anxiogenesis (Merali et al., 2006). In a study in rats using BB1 and BB2 receptor agonists and antagonists (Bédard et al., 2007), data were provided to show that both GRP and NMB affect the stress response. NMB affected both anxiety and fear responses, whereas GRP affected only fear responses (Bédard et al., 2007).
Whereas the growth effects of the BB2 receptor in normal and especially in neoplastic tissues have received the most attention, stimulation of the BB1 receptor and/or administration of NMB has been shown to have growth-promoting effects in a number of neoplastic tissues. NMB is an autocrine growth factor for non-small cell lung cancer with 14 of 14 such cell lines possessing BB1 receptors in one study (Siegfried et al., 1997), and in four non-small cell lung cancer cell lines examined in detail NMB was synthesized and released into the media by the tumor cell in 7 to 15 times greater amounts than was GRP (Siegfried et al., 1997). Blockade of the BB2 receptor only partially blocked the proliferative effect of NMB on these cells, demonstrating the importance of BB1 receptor activation for the proliferative effects in these tumor cells (Siegfried et al., 1997). Furthermore, in human colon cancers NMB and the BB1 receptor are coexpressed, and they act in an autocrine growth fashion (Matusiak et al., 2005). Activation of BB1 receptors causes proliferation of rat C6 glioblastoma cells (Moody et al., 1995a), BB1 receptor transfected RAT-1 cells (Lach et al., 1995), small cell lung cancers (Moody et al., 1992), and adrenal zona fasciculata cells (Malendowicz et al., 1996).
At present, no disease has been shown to be caused specifically by alterations in the BB1 receptor. Activation of the BB1 receptor in various human cancers (particularly human small cell lung cancers, non-small cell lung cancers, colon, cancer, and various carcinoid tumors) due to an autocrine growth pathway may have an important effect on their growth (Moody et al., 1992; Moody and Jensen, 1996; Siegfried et al., 1999; Matusiak et al., 2005; Jensen and Moody, 2006). In various studies BB1 receptors were overexpressed by 55% of small cell lung cancers, 67% of non-small cell lung cancers, 46% of intestinal carcinoids, and a proportion of colon cancers, prostate cancers, and CNS tumors such as glioblastomas (Moody et al., 1995a; Reubi et al., 2002; Matusiak et al., 2005; Jensen and Moody, 2006).
Numerous studies (Rettori et al., 1992; Pazos-Moura et al., 1996; Ortiga-Carvalho et al., 2003) including BB1 receptor knockout studies (Oliveira et al., 2006) support the conclusion that NMB plays an important physiological role in the regulation of thyrotropin release, having primarily an inhibitory effect. NMB is produced in the pituitary (Jones et al., 1992), and it is proposed that NMB functions as a tonic inhibitor of TSH secretion, acting as an autocrine/paracrine regulator (Rettori et al., 1992; Oliveira et al., 2006) (Table 1). Conditions with increased TSH release such as hypothyroidism are associated with decreased pituitary NMB levels (Jones et al., 1992; Ortiga-Carvalho et al., 2003), whereas in hyperthyroidism in which the TSH levels are suppressed; there is an increased pitui
,–CONH peptide bond changed to–CH2NH–; pGlu, pyroglutamic acid; Cpa, chlorophenylalanine; NMC, neuromedin C; F5, pentafluoro-.
