Characterisation of the neuropeptide Y receptor that mediates feeding in the rat: a role for the Y5 receptor?

Food intake was measured in freely fed rats following intracerebroventricular administration of neuropeptide Y (NPY) and several of its analogues and antagonists to investigate the hypothesis that the NPY Y5 receptor mediates feeding. Rat NPY (rNPY), rNPY(2-36) and rNPY(3-36) produced similar feeding responses over the dose range 0.7-7.0 nmol. Rat peptide YY (rPYY) was more potent and at least as efficacious as rNPY. [Leu31 Pro34]-rNPY (agonist potency: Y1 > Y5 > Y4 = y6) and human pancreatic polypeptide (hPP) produced flatter dose-response curves, suggesting partial agonism at the receptor(s). rNPY(13-36) (agonist potency: Y2 > Y5) had little activity and rPP was inactive. [D-Trp32]-NPY was a weak orexigenic agent given alone and, consistent with partial agonism, it markedly antagonised the response to porcine NPY (pNPY). Similarly, the receptor antagonist (Y1 > Y4) 1229U91 stimulated feeding slightly, and markedly inhibited rNPY-induced feeding. In contrast to a previous report, BIBP 3226 (70 nmol), another Y1 receptor antagonist, failed to inhibit the response to rNPY. Our data in vivo are inconsistent with findings that hPP, [Leu31 Pro34]-rNPY and [D-Trp32]-rNPY are full agonists at the rat cloned Y5 receptor. Thus, whilst the Y5 receptor may be involved, its participation as the sole receptor mediating the orexigenic action of NPY in the rat remains unproven.

Expression of human M2 muscarinic receptors in Sf9 cells: characterisation and reconstitution with G-proteins.

The gene for the human m2 muscarinic receptor was expressed in Sf9 cells using the baculovirus expression system. As assessed by [3H]NMS binding, Sf9 cells expressed receptor at levels of 3.3 pmoles/mg protein. The receptor was identified on western blots using an anti-muscarinic receptor antibody and was shown to have the pharmacological characteristics of an m2 muscarinic receptor. Membranes from Sf9 cells were examined to identify endogenous G-proteins by immuno-blotting and by ADP-ribosylation, indicating the presence of Gq, and a pertussis-toxin substrate which was not recognised by antibodies raised against the alpha-subunits of Gi1, Gi2, Gi3 or Go. Gsalpha was not detected, neither were there any cholera toxin substrates in Sf9 membranes. Sf9 membranes expressing m2 receptors did not show carbachol-stimulated GTPgammaS binding to endogenous G-proteins; however, when membranes were reconstituted with a mixture of purified Gi and Go, a maximum 8-fold stimulation of GTPgammaS binding was observed in response to carbachol that could be reduced by atropine. These data show that the human muscarinic m2 receptor expressed in Sf9 cells is functional.

5-Hydroxytryptamine1A receptor synthetic peptides. Mechanisms of adenylyl cyclase inhibition.

The 5-hydroxytryptamine1A receptor (5-HT1AR) is a G-protein-coupled receptor negatively coupled to adenylyl cyclase (AC). We have studied the functional domains of 5-HT1AR using synthetic peptides to block or mimic receptor function. The entire second intracellular loop (5-HT1AR-i2) and the carboxyl end of the third intracellular loop (5-HT1AR-i3-C) strongly inhibited forskolin-stimulated AC activity. These effects were not additive with those of 5-HT. Like 5-HT, the peptides 5-HT1AR-i3-C and -i2 weakly inhibited AIF4- and Mn2+ stimulated AC activity. 5-HT1AR binding assays indicated that peptides could interact with the same G-protein pool as the 5-HT1AR. 5-HT1AR-i3-C- and -i2-stimulated [35S]guanosine 5'-O-(thiotriphosphate) binding on Go/Gi proteins. Only 5-HT1AR-i3-C partially adopted an alpha-helical conformation in solution. These data show that different domains in the 5-HT1AR second and third intracellular loops can couple to and activate Gi proteins in order to mediate AC inhibition. Peptide-induced AC inhibition was not sensitive to pertussis toxin as opposed to the 5-HT1AR-mediated effect. Our data show that the 5-HT1AR and the 5-HT1AR peptides activate Gi proteins in a slightly different manner.

Activation of Gi protein by peptide structures of the muscarinic M2 receptor second intracellular loop.

The muscarinic M2 receptor that normally couples via Gi to inhibit adenylyl cyclase was made to couple to Gs by exchange of its third intracellular loop for the comparable domain of the beta 2-adrenoceptor. In HeLa cells transfected with the recombinant M2 beta i-3 cDNA, the chimaeric receptor showed carbachol-mediated activation of adenylyl cyclase (EC50 = 73 nM) that was blocked by atropine, but not by propranolol. The chimaeric receptor was shown to mediate a carbachol-stimulated, Bordetella pertussis toxin-sensitive GTPase activity in membranes of transfected HeLa cells. Interestingly, stimulation of adenylyl cyclase by carbachol was 2-fold higher in transfected cells that had been pretreated with pertussis toxin. These data suggested that the M2 beta i-3 receptor was able to couple to both Gi and Gs, and that the ability to recognise and stimulate Gi did not involve the third cytoplasmic loop of M2. We investigated peptide elements taken from the second intracellular loop of the M2 receptor for their ability to mediate activation of Gi and found that a nine amino acid peptide representing the C-terminal sequence, VKRTTKMAG-NH2 (V9G), was capable of inhibiting forskolin-stimulated adenylyl cyclase by up to 18% and could stimulate high affinity GTPase activity of rat brain membranes by 32%. Further, V9G was shown to cause a doubling of the initial rate of [35S]GTP gamma S binding to purified bovine brain Gi/Go in reconstituted phospholipid vesicles. These data identify a domain on the second intracellular loop of the muscarinic M2 receptor that is involved in the selection of a pertussis toxin-sensitive G protein.

Gi3 does not contribute to the inhibition of adenylate cyclase when stimulation of an alpha 2-adrenergic receptor causes activation of both Gi2 and Gi3.

Agonist occupancy of the alpha 2-C10 adrenergic receptor in a stable clone (1C) of Rat 1 fibroblasts produced by transfection of cells with genomic DNA encoding this receptor causes the activation of both of the pertussis-toxin-sensitive G-proteins Gi2 and Gi3 [Milligan, Carr, Gould, Mullaney & Lavan (1991) J. Biol. Chem. 266, 6447-6455]. An IgG fraction from an antiserum (I3B) which identifies the C-terminal decapeptide of Gi3 alpha only was able to inhibit partially receptor stimulation of high-affinity GTPase activity. An equivalent fraction from an antiserum (AS7) able to identify the C-terminal decapeptide of Gi1 alpha + Gi2 alpha, but not Gi3 alpha, was also able to inhibit partially receptor stimulation of GTPase activity, and the effects of the two antisera were additive. By contrast, agonist-mediated inhibition of forskolin-amplified adenylate cyclase activity was abolished completely by the IgG fraction of antiserum AS7, but was not decreased by treatment with antiserum 13B. Based on the proportion of agonist-stimulated high-affinity GTPase which was prevented by each antiserum and on the measured membrane levels of Gi2 and Gi3, calculations indicated that essentially all of the cellular Gi3, but only 15% of the available Gi2, can be activated by the alpha 2-C10 adrenergic receptor in these cells. These results demonstrate that, although Gi3 is activated by alpha 2-adrenergic agonists in membranes of clone 1C cells, it does not contribute to the transduction of receptor-mediated inhibition of adenylate cyclase.

Alpha 2-C10 adrenergic receptors expressed in rat 1 fibroblasts can regulate both adenylylcyclase and phospholipase D-mediated hydrolysis of phosphatidylcholine by interacting with pertussis toxin-sensitive guanine nucleotide-binding proteins.

The alpha 2-C10 adrenergic receptor from human platelets was expressed permanently in Rat-1 fibroblasts. A series of clones that varied in expression of the receptor from 0 to 3.5 pmol/mg of membrane protein were isolated. We have demonstrated recently in cells of one of these clones (1C) that the alpha 2-C10 receptor interacts directly with two distinct pertussis toxin-sensitive G-proteins, Gi2 and Gi3 (Milligan, G., Carr, C., Gould, G. W., Mullaney, I., and Lavan, B.E. (1991) J. Biol. Chem. 266, 6447-6455). High affinity GTPase activity in membranes of cells from the various clones was stimulated by the addition of the alpha 2-adrenergic agonist UK14304, defining that the receptor coupled productively to the G-protein signaling system. Maximal stimulation of high affinity GTPase activity correlated with the levels of receptor expressed. Clones expressing the receptor also demonstrated agonist-mediated inhibition of adenylylcyclase. Futhermore, the alpha 2-C10 receptor in one clone (1C), but not other clones, promoted a marked stimulation in the generation of water-soluble products derived from phosphatidylcholine. The concentration of UK14304 required to produce half-maximal regulation of GTPase activity (20-30 nM), of forskolin-amplified adenylylcyclase activity (30-40 nM), and of choline generation (30-40 nM) were similar. Transphosphatidylation experiments with cells of clone 1C indicated that the receptor-mediated hydrolysis of phosphatidylcholine was via the action of a phospholipase D. All of these effects were attenuated by pretreatment of the cells with pertussis toxin. Dose-effect curves of pertussis toxin-treatment demonstrated similar effective concentrations of the toxin in causing endogenous ADP-ribosylation of both Gi2 and Gi3, inhibition of receptor-stimulated GTPase activity, and phospholipase D activity. Receptor activation of phospholipase D activity was not dependent upon prior phospholipase C-dependent activation of protein kinase C, as alpha 2-adrenergic stimulation of inositol phosphate production was negligible and the presence of the selective protein kinase C inhibitor RO-31-8220, at concentrations up to 10 microM, had no effect on UK14304-mediated production of phosphatidylbutanol. These results demonstrate that expression of the alpha 2-C10 receptor in a heterologous system can result in receptor regulation of signaling elements that appear not to be primary targets for the receptor in vivo. Such results are important in respect to recent observations that transfection of a single defined receptor into separate cell lines can lead to the regulation of distinct effector systems (Vallar, L., Muca, C., Magni, M., Albert, P., Bunzow, J., Meldolesi, J. and Civelli, O. (1990) J. Biol. Chem. 265, 10320-10326).(ABSTRACT TRUNCATED AT 400 WORDS)

Molecular interaction of the human alpha 2-C10-adrenergic receptor, when expressed in Rat-1 fibroblasts, with multiple pertussis toxin-sensitive guanine nucleotide-binding proteins: studies with site-directed antisera.

DNA encoding the human alpha 2-C-10-adrenergic receptor was transfected into Rat-1 fibroblasts by CaPO4 precipitation, and clones expressing the receptor were isolated and expanded. One clone (1C) expressing high levels of the receptor was studied in order to determine the contacts between this receptor and guanine nucleotide-binding proteins (G proteins) mediating second messenger signaling. The alpha 2-adrenergic agonist UK 14304 stimulated high affinity GTPase activity in membranes from these cells. Incubation of these membranes with Protein A-purified fractions from an antiserum able to identify the carboxyl-terminal decapeptide common to Gi1 alpha and Gi2 alpha was partially able to prevent agonist stimulation of high affinity GTPase activity. Similar results were produced with an antiserum that identifies the carboxyl-terminal decapeptide of Gi3 alpha. In contrast, equivalent fractions of antisera that identify the carboxyl-terminal decapeptides of Go alpha and Gs alpha did not inhibit receptor stimulation of high affinity GTPase activity. Coincubation of the membranes from the cells with Protein A-purified fractions from the anti-Gi1 alpha + Gi2 alpha antiserum and the anti-Gi3 alpha antiserum produced greater inhibition of UK14304-stimulated GTPase activity than did either of the two antisera in isolation. These data show direct interaction of the human alpha 2-C10-adrenergic receptor, when expressed in this clone of Rat-1 fibroblasts, with multiple pertussis toxin-sensitive G proteins and demonstrate that a single receptor has the physical capacity to interact functionally with more than a single pertussis toxin-sensitive G protein in a native membrane. Furthermore, because the two antisera were able to inhibit receptor stimulation of high affinity GTPase activity to similar degrees, the G protein pools identified by these antisera must contribute similar amounts of the total receptor activation of pertussis toxin-sensitive G proteins in these cells.

5-Hydroxytryptamine (5-HT)-induced shape change in human platelets determined by computerized data acquisition: correlation with [125I]-iodoLSD binding at 5-HT2 receptors.

The 5-HT-induced shape change and subsequent aggregation of platelets provides a functional assay for 5-HT2 receptors. In the present study we describe a method to increase aggregometer sensitivity by digital conversion of the voltage produced by changes in light transmittance through a platelet suspension, thereby allowing accurate analyses of the primary shape-change response. The pharmacology of 5-HT-induced shape change was then compared with that of [125I]-iodoLSD binding in human platelets. 5-HT caused a dose-dependent change in platelet shape (maximum response 5 x 10(-6) M, EC50 10(-6) M). Furthermore, there was a significant correlation across a selection of drugs between IC50 values for inhibition of 5-HT-induced shape change and for inhibition of platelet binding of the 5-HT2 receptor ligand [125I]-iodoLSD. These results support the hypothesis that 5-HT-induced shape change and [125I]-iodoLSD binding in human platelets are mediated through the same receptor, and validate the methods of data acquisition described.

The alpha 2B adrenergic receptor of undifferentiated neuroblastoma x glioma hybrid NG108-15 cells, interacts directly with the guanine nucleotide binding protein, Gi2.

In membranes of undifferentiated neuroblastoma x glioma hybrid cell line NG108-15, the apparent specific binding of [3H]yohimbine measured in the presence of 1 microM noradrenaline, was increased substantially by the presence of the poorly hydrolysed analogue of GTP, guanylyl-imidodiphosphate (Gpp[NH]p) or by preincubation of membranes with antibodies against the C-terminal decapeptide of the alpha subunit of the G-protein Gi2. Such an effect was not produced by antibodies against the equivalent region of Go alpha Gi3 alpha or Gs alpha or from non-immune serum. By contrast, total specific binding of [3H]yohimbine was not modified by co-incubation with Gpp[NH]p or by preincubation with the antibodies from any of the anti-G protein antisera. These results demonstrate a direct interaction of the alpha 2B adrenergic receptor of NG108-15 cells with Gi2.

The role and specificity of guanine nucleotide binding proteins in receptor-effector coupling.

Nine distinct alpha subunits of guanine nucleotide binding proteins (G-proteins) have now been identified by cDNA cloning. Each of these functions to allow transduction of information between hormone-activated receptors in the plasma membrane and effector systems which are either ion channels or enzymes which regulate the intracellular concentration of second messengers. As the individual G-proteins are highly similar in primary sequence, it is pertinent to ask what degree of specificity of interaction each of these display with the various receptors and effector systems. Specificity of tissue location defines that the rod and cone transducins (TD1 and TD2, respectively) act as the coupling proteins between rhodopsin and cone opsins and their cyclic nucleotide phosphodiesterase effectors and that G(olf) is the G-protein which tranduces signals from odorant receptors to adenylate cyclase in olfactory sensory neurones. However, many of the other identified G-proteins are co-expressed in a single tissue or cell. Whilst sensitivity to ADP-ribosylation catalysed by bacterial toxins from Bordetella pertussis and Vibrio cholerae has allowed a further subdivision of the G-protein family, this approach is limited as these toxins have multiple G-protein substrates. As the extreme C-terminus of the alpha subunit of each G-protein appears to be a key domain for the interactions of receptors and G-proteins we have generated a series of G-protein-selective antipeptide antisera against this region and then have used these antisera to attempt to interfere with receptor-G-protein coupling. With this approach we have been able to demonstrate that a delta opioid receptor-mediated inhibition of adenylate cyclase in neuroblastoma x glioma, NG108-15, cell membranes is transduced specifically by Gi2 and in the same cell that alpha 2 adrenergic inhibition of Ca2+ currents is transduced by Go. Similar strategies are likely to be of universal significance, for example in the identification of the G-protein (Gp) which regulates the receptor-mediated activation of phosphoinositidase C. Methods to allow pharmacological manipulation of the levels of expression of various G-proteins in the membranes of cells are also discussed. Such approaches are also likely to assist in the identification of G-proteins of defined functions.

Hallucinogenic drugs are partial agonists of the human platelet shape change response: a physiological model of the 5-HT2 receptor.

We have assayed several phenylalkylamine and indolealkylamine hallucinogens, as well as structurally similar nonhallucinogens, for their effect on human platelet shape change, a physiological model for the central serotonergic 5-HT2 receptor. The hallucinogenic drugs lysergic acid diethylamide (LSD-25), N,N-dimethyltryptamine (N,N-DMT), 5-methoxy-N,N-dimethyltryptamine (5-MeODMT), 4-iodo-2,5-dimethoxyphenylisopropylamine (DOI), bufotenine, and mescaline all showed a characteristic 5-HT2 partial agonist effect on platelet shape change. Nonhallucinogens with structural similarity to hallucinogens did not share this profile. Lisuride, methysergide, and lysergic acid showed antagonism of 5-HT-induced shape change, but none were shape change agonists. Other "psychoactive" or mood-altering drugs (cocaine, amphetamine, phencyclidine) showed poor antagonism of 5-HT-induced platelet shape change. This work refines recent ideas that some of the behavioral effects of LSD-type hallucinogens in humans are due to their actions at 5-HT2 receptors and suggests that these hallucinogens are partial 5-HT2 agonists.

Subacute and chronic in vivo lithium treatment inhibits agonist- and sodium fluoride-stimulated inositol phosphate production in rat cortex.

We have investigated the effects of in vivo lithium treatment on cerebral inositol phospholipid metabolism. Twice-daily treatment of rats with LiCl (3 mEq/kg) for 3 or 16 days resulted in a 25-40% reduction in agonist-stimulated inositol phosphate production, compared with NaCl-treated controls, in cortical slices prelabelled with [3H]inositol. A small effect was also seen with 5-hydroxytryptamine (5-HT) 24 h after a single dose of LiCl (10 mEq/kg). Dose-response curves to carbachol and 5-HT showed that lithium treatment reduced the maximal agonist response without altering the EC50 value. This inhibition was not affected by the concentration of LiCl in the assay buffer. Stimulation of inositol phosphate formation by 10 mM NaF in membranes prepared from cortex of 3-day lithium-treated rats was also inhibited, by 35% compared with NaCl-treated controls. Lithium treatment did not alter the kinetic profile of inositol polyphosphate formation in cortical slices stimulated with carbachol. Muscarinic cholinergic and 5-HT2 bindings were unaltered by lithium, as was cortical phospholipase C activity and isoproterenol-stimulated cyclic AMP formation. [3H]Inositol labelling of phosphatidylinositol 4,5-bisphosphate was significantly enhanced by 3-day lithium treatment. The results, therefore, indicate that subacute or chronic in vivo lithium treatment reduces agonist-stimulated inositol phospholipid metabolism in cerebral cortex; this persistent inhibition appears to be at the level of G-protein-phospholipase C coupling.

5-Hydroxytryptamine-stimulated inositol phospholipid hydrolysis in the mouse cortex has pharmacological characteristics compatible with mediation via 5-HT2 receptors but this response does not reflect altered 5-HT2 function after 5,7-dihydroxytryptamine lesioning or repeated antidepressant treatments.

5-Hydroxytryptamine (5-HT; 3 x 10(-8)-1 x 10(-5)M) produced a dose-dependent increase in phosphatidylinositol/polyphosphoinositide (PI) turnover in mouse cortical slices, as measured by following production of 3H-labelled inositol phosphates (IPs) in the presence of 10 mM LiCl. Analysis of individual IPs, in slices stimulated for 45 min, indicated substantial increases in inositol monophosphate (IP1; 140%) and inositol bisphosphate (IP2; 95%) contents with smaller increases in inositol trisphosphate (IP3; 51%) and inositol tetrakisphosphate (IP4; 48%) contents. The increase in IP3 level was solely in the 1,3,4-isomer. This response was inhibited by the nonselective 5-HT antagonists methysergide, metergoline, and spiperone. It was also inhibited by the selective 5-HT2 antagonists ketanserin and ritanserin but not by the 5-HT1 antagonists isapirone, (-)-propranolol, or pindolol. 5-HT-stimulated IP formation was also unaltered by atropine, prazosin, and mepyramine. Lesioning brain 5-HT neurones using 5,7-dihydroxytryptamine (5,7-DHT; 50 micrograms i.c.v.) produced a 210% (p less than 0.01) increase in the number of 5-HT2-mediated head-twitches induced by 5-methoxy-N,N-dimethyltryptamine (2 mg/kg). However, 5,7-DHT lesioning had no effect on 5-HT-stimulated PI turnover in these mice. Similarly, an electroconvulsive shock (90 V, 1 s) given five times over a 10-day period caused an 85% (p less than 0.01) increase in head-twitch responses but no change in 5-HT-stimulated PI turnover. Decreasing 5-HT2 function by twice-a-day injection of 5 mg/kg of zimeldine or desipramine (DMI) produced 50% (p less than 0.01) and 56% (p less than 0.01), respectively, reductions in head-twitch behaviour.(ABSTRACT TRUNCATED AT 250 WORDS)

Relationship of increased food intake and plasma ACTH levels to 5-HT1A receptor activation in rats.

Various putative agonists of the 5-HT1A receptor subtype induce feeding in rats, probably by activating raphé somatodendritic 5-HT autoreceptors. These drugs also produce a marked increase in plasma concentrations of corticotropin (ACTH). In the present experiment we attempted to localize the site of action of 5-HT1A agonists on the secretion of ACTH and examined the relationship between 5-HT1A agonist-induced feeding and ACTH secretion. Rats were injected with either the high affinity 5-HT1A agonist 8-hydroxy-2-(di-n-propylamino) tetralin (8-OH-DPAT) (0.016-1.0 mg/kg, s.c.) or the novel anxiolytics buspirone, gepirone or ipsapirone (2.0-16.0 mg/kg, s.c.), and either had their food intake measured 2 hr post injection or were sacrificed 30-40 min post injection for measurement of plasma ACTH. Plasma ACTH also was measured in rats pretreated with the serotonin synthesis inhibitor, para-chlorophenylalanine (PCPA) for three days (150 mg/kg, i.p. per day) and subsequently injected with 8-OH-DPAT (0.3 mg/kg, s.c.). As previously reported, the 5-HT1A agonists increased both food agonists increased both food intake and plasma ACTH concentrations. After 8-OH-DPAT, ipsapirone and gepirone the amount of food consumed was positively correlated with the concentration of plasma ACTH. No such correlation was evident following buspirone. PCPA pretreatment resulted in near total depletion of brain 5-HT content but had no effect on the ACTH rise induced by 8-OH-DPAT. Therefore, in contrast to the presynaptic site previously proposed for 5-HT1A agonist-induced feeding, the present results suggest a agonist-induced feeding, the present results suggest a postsynaptic location for the 5-HT1A receptor mediating ACTH release.

Serotonergic agonists stimulate inositol lipid metabolism in rabbit platelets.

The metabolism of inositol phospholipids in response to serotonergic agonists was investigated in rabbit platelets. In platelets prelabelled with [3H]-inositol, in a medium containing 10 mM LiCl which blocks the enzyme inositol-1-phosphatase, 5-hydroxytryptamine (5-HT) caused a dose-dependent accumulation of inositol phosphates (IP). This suggests a phospholipase-C-mediated breakdown of phosphoinositides. Ketanserin, a selective 5-HT2 antagonist, was a potent inhibitor of the 5-HT response, with a Ki of 28 nM, indicating that 5-HT is activating receptors of the 5-HT2 type in the platelet. Lysergic acid diethylamide (LSD) and quipazine also caused dose-related increases in inositol phosphate levels, though these were considerably less than those produced by 5-HT. These results show that relatively small changes in phosphoinositide metabolism induced by serotonergic agonists can be investigated in the rabbit platelet, and this cell may therefore be a useful model for the study of some 5-HT receptors.

Inhibition of aggregation in Dictyostelium by EGTA-induced depletion of calcium.

During aggregation of Dictyostelium discoideum, amoebae move towards collecting centres emitting cyclic AMP. Previous work has indicated that Ca2+ can affect the operation of various parts of the chemotactic mechanism but there are contradictory reports about the role or requirement for Ca2+ during aggregation of intact amoebae. In this study we show that there is a requirement for Ca2+ during this aggregation process. Addition of EGTA (7 mM) and in some cases extensive washing with EGTA is needed to see inhibition of aggregation. Considerable stores of Ca2+ present within cells may explain the failure of some previous attempts to see a Ca2+ requirement during aggregation.