Neurotensin, bradykinin and somatostatin inhibit cAMP production in neuroblastoma N1E115 cells via both pertussis toxin sensitive and insensitive mechanisms

Heterogeneity of ghrelin/growth hormone secretagogue receptors. Toward the understanding of the molecular identity of novel ghrelin/GHS receptors

Ghrelin is a gastric polypeptide displaying strong GH-releasing activity by activation of the type 1a GH secretagogue receptor (GHS-R1a) located in the hypothalamus-pituitary axis. GHS-R1a is a G-protein-coupled receptor that, upon the binding of ghrelin or synthetic peptidyl and non-peptidyl ghrelin-mimetic agents known as GHS, preferentially couples to G(q), ultimately leading to increased intracellular calcium content. Beside the potent GH-releasing action, ghrelin and GHS influence food intake, gut motility, sleep, memory and behavior, glucose and lipid metabolism, cardiovascular performances, cell proliferation, immunological responses and reproduction. A growing body of evidence suggests that the cloned GHS-R1a alone cannot be the responsible for all these effects. The cloned GHS-R1b splice variant is apparently non-ghrelin/GHS-responsive, despite demonstration of expression in neoplastic tissues responsive to ghrelin not expressing GHS-R1a; GHS-R1a homologues sensitive to ghrelin are capable of interaction with GHS-R1b, forming heterodimeric species. Furthermore, GHS-R1a-deficient mice do not show evident abnormalities in growth and diet-induced obesity, suggesting the involvement of another receptor. Additional evidence of the existence of another receptor is that ghrelin and GHS do not always share the same biological activities and activate a variety of intracellular signalling systems besides G(q). The biological actions on the heart, adipose tissue, pancreas, cancer cells and brain shared by ghrelin and the non-acylated form of ghrelin (des-octanoyl ghrelin), which does not bind GHS-R1a, represent the best evidence for the existence of a still unknown, functionally active binding site for this family of molecules. Finally, located in the heart and blood vessels is the scavenger receptor CD36, involved in the endocytosis of the pro-atherogenic oxidized low-density lipoproteins, which is a pharmacologically and structurally distinct receptor for peptidyl GHS and not for ghrelin. This review highlights the most recently discovered features of GHS-R1a and the emerging evidence for a novel group of receptors that are not of the GHS1a type; these appear involved in the transduction of the multiple levels of information provided by GHS and ghrelin.

Interactions between neurotensin receptors and G proteins

Three neurotensin (NT) receptors have been cloned to date, two of which, NTS1 and NTS2, belong to the family of seven transmembrane domain receptors coupled to G proteins (GPCRs). NTS1 and NTS2 may activate multiple signal transduction pathways, involving several G proteins. However, whereas NT acts as an agonist towards all NTS1-mediated pathways, this peptide may exert either agonist or antagonist activities, depending on the NTS2-mediated pathway in question. Studies on these receptors reinforce the concept of independence between multiple signals potentially mediated through a single GPCR, generating a wide diversity of functional responses depending on the host cell and the ligand.

Identification and Functional Characterization of a 5-Transmembrane Domain Variant Isoform of the NTS2 Neurotensin Receptor in Rat Central Nervous System

The present study demonstrated that alternative splicing of the rat nts2 receptor gene generates a 5-transmembrane domain variant isoform (vNTS2) that is co-expressed with the full-length NTS2 receptor throughout the brain and spinal cord, as evidenced by reverse transcription-PCR. The vNTS2 polypeptide is 281 amino acids in length, which is 135 amino acids shorter than the full-length isoform. Immunohistochemical and radioligand binding studies revealed that the HA-tagged recombinant vNTS2 receptor is poorly targeted to plasma membranes in transfected COS-7 cells. Binding studies also showed that the truncated receptor displayed a 5000-fold lower affinity for neurotensin (NT) than its full-length counterpart (IC(50) of 10 mum and 2 nm, respectively). Yet NT binding induced efficient internalization of receptor-ligand complexes in vNTS2-transfected cells. Furthermore, it produced a rapid (<5 min) activation of the mitogen-activated protein kinases (ERK1/2) pathway, indicating functional coupling of the variant receptor. This activation is sustained (>1 h) and is also produced by the NTS2 agonist levocabastine. Western blotting experiments suggested that vNTS2 is not expressed in monomeric form in the rat central nervous system. However, it does appear to form a variety of multimeric complexes, including homodimers and heterodimers, with the full-length NTS2. Indeed, co-immunoprecipitation studies in dually transfected cells demonstrated that the two receptor isoforms can form stable associations. Taken together, the present results indicated that the rat vNTS2 is a functional receptor that may play a role in NT signaling in mammalian central nervous system.

Bradykinin B2 and GPR100 receptors: a paradigm for receptor signal transduction pharmacology

The aim of the present report was to investigate the ligand selectivity of the human orphan G-protein-coupled receptor GPR100 (hGPR100), recently identified as a novel bradykinin (BK) receptor, as compared with that of the human B(2) receptor (hB(2)R) stably transfected in Chinese hamster ovary cells. BK was able to inhibit the cAMP production induced by forskolin with a potency 100-fold lower at the hGPR100 (pEC(50) = 6.6) than that measured at the hB(2)R (pEC(50) = 8.6). Both effects were inhibited by the B(2) receptor antagonist Icatibant (1 microM). The nonpeptide B(2) receptor agonist FR190997 (8-[2,6-dichloro-3-[N-methylcarbamoyl)cinnamidoacetyl]-N-methylamino]benzyloxy]-2-methyl-4-(2-pyridylmethoxy)quinoline) did inhibit the forskolin-induced cAMP production (pEC(50) = 7.7) at the hB(2)R, whereas it was not able to exert any effect at the hGPR100. The human insulin-like peptide relaxin 3 did inhibit the cAMP production at the hGPR100 (pEC(50) = 7.3) at a greater extent than BK, and was devoid of any effect at the hB(2)R. FR190997 and relaxin 3 responses at the hB(2)R and hGPR100, respectively, were not inhibited by Icatibant (1 microM). These data indicate FR190997 and relaxin 3 as selective agonists for hB(2)R and hGPR100, respectively, and support the concept that different agonists may specifically bias the conformational states of a receptor to result in a final common G protein coupling, which is differentially recognized by antagonists.

Neurotensin induces tyrosine hydroxylase gene activation through nitric oxide and protein kinase C signaling pathways

The regulation of tyrosine hydroxylase (TH) represents an effective means to control the level of catecholamines, because TH is the major limiting enzyme of monoamine biosynthesis. The neuropeptide neurotensin (NT) is a neuromodulator of dopaminergic systems, and a direct interaction between NT and TH expression has been demonstrated in vivo and in vitro. In the present work, the molecular mechanisms and signaling pathways responsible for TH gene activation have been explored. In N1E-115 cells, NT agonist induced a TH protein level increase, correlating with a significant increase in TH mRNA abundance. This cellular response was the result of TH promoter activation, via c-fos and Jun D binding at the AP-1 responsive element. Using selective protein kinase C and nitric oxide synthase inhibitors, we demonstrate, by quantitative reverse transcription-polymerase chain reaction, gel shift, and protein assays, that TH gene activation by NT agonist requires both protein kinase C stimulation and nitric oxide production. The two pathways exert distinct roles; whereas nitric oxide synthase inhibitors blocked c-fos expression, protein kinase C inhibitors blocked that of Jun D. The requirement for two distinct and concomitant pathways by NT demonstrates a very fine level of control of specificity on TH gene activation.

Regulation of I(Cl,swell) in neuroblastoma cells by G protein signaling pathways

Guanosine 5'-O-(3-thiotriphosphate) (GTPgammaS) activated the I(Cl,swell) anion channel in N1E115 neuroblastoma cells in a swelling-independent manner. GTPgammaS-induced current was unaffected by ATP removal and broadly selective tyrosine kinase inhibitors, demonstrating that phosphorylation events do not regulate G protein-dependent channel activation. Pertussis toxin had no effect on GTPgammaS-induced current. However, cholera toxin inhibited the current approximately 70%. Exposure of cells to 8-bromoadenosine 3',5'-cyclic monophosphate did not mimic the effect of cholera toxin, and its inhibitory action was not prevented by treatment of cells with an inhibitor of adenylyl cyclase. These results demonstrate that GTPgammaS does not act through Galpha(i/o) GTPases and that Galpha(s)/Gbetagamma G proteins inhibit the channel and/or channel regulatory mechanisms through cAMP-independent mechanisms. Swelling-induced activation of I(Cl,swell) was stimulated two- to threefold by GTPgammaS and inhibited by 10 mM guanosine 5'-O-(2-thiodiphosphate). The Rho GTPase inhibitor Clostridium difficile toxin B inhibited both GTPgammaS- and swelling-induced activation of I(Cl,swell). Taken together, these findings indicate that Rho GTPase signaling pathways regulate the I(Cl,swell) channel via phosphorylation-independent mechanisms.

Somatostatin enhances neurite outgrowth in PC12 cells

The rat pheochromocytoma cell line PC12 forms neurites in response to nerve growth factor (NGF), and it was also reported to extend processes in the presence of somatostatin (somatotropin release-inhibiting factor, SRIF), a neuroactive peptide that seems to act as a morphogenetic factor in the developing nervous system. In the present study, we re-evaluated the effects of SRIF on PC12 cell differentiation. Our results indicate that SRIF alone is ineffective in promoting neurite outgrowth. Instead, SRIF or its analogue, octreotide (a SRIF agonist on the receptor subtypes 2, 3 and 5), potentiates neurite extension induced by NGF. These results suggest that SRIF enhances neurite formation in PC12 cells without directly promoting neurite outgrowth. SRIF potentiation of NGF-induced neurite outgrowth persists at least in part in the presence of pertussis toxin (PTX), suggesting the involvement of PTX-insensitive G-proteins. In addition, protein kinase-dependent pathways are likely to mediate SRIF effects on NGF-induced differentiation.

Mechanisms of regulation of neurotensin receptors

Since its discovery in 1973, the neuropeptide neurotensin has been demonstrated to be involved in the control of a broad variety of physiological activities in both the central nervous system and in the periphery. Pharmacological studies have shown that the biological effects elicited by neurotensin result from its specific binding to cell membrane neurotensin receptors that have been characterized in various tissue and in cell preparations. In addition, it is now well documented that most of these responses are subject to rapid desensitization. Such desensitization results in transient responses to sustained peptide applications, or to tachyphylaxis during successive stimulations in the same conditions. More recently, desensitization of neurotensin signalling was investigated at the cellular and molecular levels. In cultured cells, regulation at the second messenger level, receptor internalization, and receptor down-regulation processes have been reported. These are proposed to play a critical role in the control of cell responsiveness to neurotensin. This review aims to compile recent data on the different biochemical processes involved in the regulation of the neurotensin receptor and to discuss the physiological consequences of this regulation in vivo.

Neurotensin induces calcium oscillations in cultured amacrine cells

The peptide, neurotensin, is found in a class of amacrine cells synapsing chiefly with other amacrine cells in the chicken retina (Li & Lam, 1990; Watt et al., 1991). To investigate the possible effects of neurotensin, we have used Ca2+ imaging to measure cytosolic Ca2+ concentrations in cultured chick amacrine cells. Following a delay of about 2 min, neurotensin (300 nM) induced oscillations in Ca2+ concentration that typically had a period of 2 min and peak values of about 300 nM when averaged over the cell body. The phospholipase C inhibitors U-73, 112 and 4'-bromophenacyl bromide terminated oscillations induced by neurotensin but the protein kinase inhibitors H7 and staurosporine did not inhibit oscillations, increasing their frequency instead. In the absence of external Ca2+, neurotensin induced only a single Ca2+ transient, much briefer than when external Ca2+ was present. Together these results suggest that neurotensin activates phospholipase C, thereby producing IP3 that triggers Ca2+ release from an internal store. Although this released Ca2+ contributes to periodic Ca2+ peaks, the majority of cytosolic Ca2+, even in the first peak, comes from Ca2+ influx across the plasmalemma.

Multiple mechanisms of bradykinin-induced contraction in rat and guinea pig smooth muscles in vitro

Bradykinin caused graded contraction in the guinea pig ileum, trachea and urinary bladder and rat uterus and vas deferens in vitro. The order of potency (EC50, nM) was: ileum (3) > uterus (5) > trachea (15) > vas deferens (41) > urinary bladder (52) and the maximal responses (percentage to 80 mM KCl) were: 152 +/- 8 (ileum), 122 +/- 6 (uterus), 97 +/- 3 (urinary bladder), 75 +/- 5 (trachea) and 33 +/- 3 (vas deferens). Responses to bradykinin in guinea pig ileum and urinary bladder and rat vas deferens and uterus were markedly attenuated in Ca(2+)-free medium with or without EGTA or by nicardipine, whereas those in guinea pig trachea depended almost exclusively on intracellular Ca2+ sources which were sensitive to ryanodine. Treatment of the animals with pertussis toxin only inhibited bradykinin-induced contraction of the rat uterus. Furthermore, the protein kinase C inhibitors, H7 (5-isoquinolinysulfonyl-2-methyl-piperazine) and staurosporine, antagonized in a graded manner bradykinin responses in guinea pig ileum and trachea and rat vas deferens, indicating the possible dependence on activation of protein kinase C mechanisms, while responses of the rat uterus rely on coupling by a pertussis toxin-sensitive G protein. Thus, bradykinin acting at B2 receptors may induce contractions in several smooth muscles from rat and guinea pig through activation of multiple second messenger pathways.

Neurotensin binding to human embryonic lung fibroblasts

Several neuropeptides have been shown to regulate the function of cells involved in immune response and inflammation. Neurotensin is a 13 amino acid neuropeptide localized primarily to the nervous system and gut. Neurotensin also stimulates mast cell degranulation and enhances phagocytic and cytolytic capability of macrophages, suggesting that this peptide regulates inflammatory and immune responses. Fibroblasts play an important role in inflammation and tissue healing, and these processes may be regulated by several neuropeptides that have been shown to bind to fibroblasts. However neurotensin receptors have not been identified on fibroblasts. Human embryonic lung fibroblasts (HELF) were examined for binding and biological effects of neurotensin. 125I-neurotensin binding to adherent and confluent human embryonic lung fibroblasts (HELF), plated in 12mm diameter wells was specific and saturable. Computer-assisted resolution of the binding data demonstrated two classes of binding sites: a high affinity, low capacity site (Kd = 1.6 x 10(-11) M, 19.5 x 10(7) sites/well), and a low- affinity, high-capacity site (Kd = 10(-8) M, 4 x 10(9) sites/well). Neurotensin stimulated immediate, transient, dose-dependent increases of cytosolic calcium in HELF (threshold dose: 10(-11) M), suggesting release of calcium from intracellular stores. The novel finding of neurotensin receptors on fibroblasts provides further support for this neuropeptide's role as a regulator of inflammatory and immune responses.

In vitro and in vivo evidence of neurotensin release from preganglionic axon terminals in the stellate ganglion of the cat

We have previously shown that the neurotensin (NT) store in preganglionic axon terminals of the cat stellate ganglion (SG) is reversibly depleted by prolonged preganglionic stimulation. The present study addresses the questions of whether the preganglionic axon terminals release NT in response to depolarizing stimuli in vitro and whether in vivo NT is released by the tonic firing of the sympathetic preganglionic neurons. Slices of the SG of the anaesthetized cat, maintained in oxygenated Ringer solution, released NT. The efflux increased when the K concentration was increased from 5 to 25 or 45 mM or when veratridine was added to the medium. In Ca-free medium, efflux was suppressed. The effect of veratridine was blocked by tetrodotoxin (TTX). In awake, freely moving cats, in which TTX was applied for 4 days to the preganglionic input of the right SG, the NT content of this ganglion doubled by comparison with the left SG. Since NT accumulates proximal to a ligature on the preganglionic input of the SG, the increased NT content is likely to result from suppression of action potential-dependent release while influx into the terminals persists. This result suggests that the steady state of the NT store in sympathetic preganglionic terminals is the result of a steady influx from the soma balanced by action potential-dependent loss, presumably release.

Pertussis toxin pretreatment affects opiate/nonopiate and stress-induced analgesia differently

Intracerebroventricular injection of pertussis toxin (PTX, 1 microgram/rat) six days before the hot plate test abolished analgesia induced by central morphine. The toxin did not affect analgesia evoked by central neurotensin or ASU 1-7 eel calcitonin. PTX pretreatment also attenuated footshock-induced analgesia (FSIA) delivered to all four paws. When the shock was restricted to the front paws, PTX consistently lowered postshock tail flick latencies, but did not reduce analgesia resulting from shock delivered to the hind paws. It thus appears that PTX-sensitive G-proteins are an essential transduction step needed to initiate the molecular events underlying opiate analgesia evoked by either morphine or shock. In contrast, the signal transduction mechanism subsequent to the stimulation of neurotensin or calcitonin receptors, and to the nonopiate FSIA, appears not to involve PTX-sensitive G-proteins.

Glucose transport stimulation by bradykinin in Swiss 3T3 fibroblasts: a pertussis toxin-sensitive mechanism operates without involvement of arachidonic acid and cyclic AMP

The possible role of cAMP and/or arachidonic acid (and metabolites) in the stimulation of glucose transport elicited by bradykinin in Swiss 3T3 fibroblasts was investigated with particular attention to the part of this effect inhibitable by pertussis toxin. Application of the membrane permeant cAMP analog 8-BrcAMP modified neither basal nor stimulated transport observed after bradykinin, insulin, or the combination of the two, indicating that [cAMP]i fluctuations are probably not involved. In contrast, arachidonic acid, which is released by the cells exposed to bradykinin, was able to markedly stimulate glucose transport, however, only at relatively high concentrations (EC50 approximately 30 microM). The stimulation by arachidonic acid was insensitive to pertussis toxin and was largely inhibited by both the cyclooxygenase blocking drug, indomethacin, and the [Ca2+]i clamping at the resting level (by ionomycin administered in a Ca2(+)-free incubation medium). Neither of the last treatments affected the glucose transport activated by bradykinin to a great extent. Moreover, the bradykinin-induced arachidonic acid release was unaffected by pertussis toxin and markedly inhibited by two treatments ineffective on glucose transport, the blockade of [Ca2+]i increases elicited by the peptide and the administration of the phospholipase A2 blocker, quinacrine. These results exclude that glucose transport stimulation by bradykinin is mediated intracellularly via arachidonic acid release. Since the involvement of Ca2+ and diacylglycerol can also be ruled out by present and previous results, this effect of the peptide appears to be independent of the generation of known second messengers and might be triggered by the direct interaction of a pertussis toxin-sensitive G protein with the glucose transporter in the plane of the plasma membrane.