Elsevier

Neuropharmacology

Volume 40, Issue 3, March 2001, Pages 334-344
Neuropharmacology

S 14506: novel receptor coupling at 5-HT1A receptors

https://doi.org/10.1016/S0028-3908(00)00162-3Get rights and content

Abstract

S 14506 is chemically related to the inverse agonist at 5-HT1A receptors, spiperone, but S 14506 behaves as one of the most potent agonists known at these receptors, both in vitro and in vivo. In hippocampal membranes, the specific binding of [ 3H]-S 14506 (Kd=0.79±0.2 nM; Bmax=400±32 fmol/mg protein) to 5-HT1A receptors resembled that of an antagonist in that it was increased by GppNHp, whereas GppNHp reduced the binding of the classic agonist [ 3H]-8-OH-DPAT (Kd=1.5±0.5 nM; Bmax=303±20 fmol/mg protein). Manganese, magnesium and calcium reduced the binding of [ 3H]-S 14506 to 5-HT1A receptors whereas the binding of [ 3H]-8-OH-DPAT was increased. Further, sodium markedly reduced the binding of [ 3H]-8-OH-DPAT, without affecting the binding of [ 3H]-S 14506. [ 3H]-S 14506 also bound with high affinity to h 5-HT1A receptors stably expressed in membranes of CHO cells (Kd=0.13±0.05 nM; Bmax=2.99±0.60 pmol/mg protein): the Bmax was double that of [ 3H]-8-OH-DPAT. GppNHp strongly decreased [ 3H]-8-OH-DPAT binding but scarcely changed [ 3H]-S 14506 binding; calcium, magnesium and manganese had little effect on [ 3H]-S 14506 binding in CHO cells. Antagonists (WAY 100635, WAY 100135) and inverse agonists (spiperone and metitepine) displaced [ 3H]-S 14506 binding with high affinity and Hill slopes close to unity, whereas agonists (5-HT and 5-CT) displayed low affinity with low Hill slopes: partial agonists (buspirone, ipsapirone) showed intermediate properties. In fusion proteins of h 5-HT1A receptors with Giα1 the compound potently increased high-affinity GTPase, with a steeper Hill slope than for 5-HT, which may indicate positive cooperativity. The maximum response for S 14506 in these assays was equivalent to 5-HT, indicating it to be a full agonist.

In molecular modelling studies, using a three-site model of the 5-HT1A receptor, S 14506 spanned between the 5-HT recognition site and the “arginine switch” (DRY microdomain) postulated to activate the interaction of the receptor with the G protein. Thus it is possible to synthesise ligands at G-protein-coupled receptors which are highly potent agonists, but which are structurally related to inverse agonists and show some features of antagonist/inverse agonist binding.

Introduction

The presence of diverse functional consequences of the specific interaction of drugs at single receptors may be underestimated. Agonists have been hypothesised to allow the possibility of “differential trafficking” by specifically activating certain subsets of G proteins after induction of particular conformational states of a receptor (Kenakin, 1995) and some evidence has been obtained that agonists at the 5-HT1A receptor may induce differential coupling to different G proteins (Gettys et al., 1994), which might lead to different pharmacological profiles for the different agonists concerned. The rich diversity of the different coupling pathways activated by 5-HT1A receptors has been reviewed by Raymond et al. (1999). Nevertheless, drugs that are very close structural congeners [e.g., the partial agonists at 5-HT1A receptors: buspirone, ipsapirone, tandospirone, gepirone, Fig. 1(A)] would be expected to interact with the receptor in a very similar way (Jacoby et al., 1999), which may explain why the pharmacology is essentially the same for these drugs and why there has been no major improvement over the first drug, buspirone. However, contrasting pharmacological profiles have been shown in binding experiments from agonists, antagonists or inverse agonists acting at 5-HT1A receptors (McLoughlin and Strange, 2000).

5-HT1A receptors are important targets for drug action in that presynaptic 5-HT1A receptors in the raphe nuclei control 5-HT release throughout the brain, and postsynaptic 5-HT1A receptors in, for example, the hippocampus are important for memory and stress modulation. However, with the possible exception of the weak anxiolytic, buspirone, the therapeutic potential of this target has not been realised. This may be because both anxiogenic and anxiolytic effects follow from stimulating different sites within the 5-HT1A receptor (Spedding et al., 1998).

Spiperone is a potent inverse agonist at 5-HT1A receptors (Barr and Manning, 1997; Newman-Tancredi et al., 1997a, Newman-Tancredi et al., 1997b; Kellett et al., 1999; McLoughlin and Strange, 2000). S 14506, although an analogue of spiperone [Fig. 1(B)], has been reported to be a highly potent agonist at 5-HT1A receptors in behavioural tests, and to be a potent anxiolytic (Colpaert et al., 1992). S 14506 fully substitutes for the 5-HT1A receptor cue of 8-hydroxy-2-(di-n-propylamino)tetralin (8-OH-DPAT) in drug discrimination paradigms in the rat, with an EC50 of 9 μg/kg subcutaneous. (Schreiber et al., 1995). However, the binding of [ 3H]-S 14506 to 5-HT1A receptors in brain membranes is atypical (Lima et al., 1997). Furthermore, a distictive feature of the drug was the steep Hill slope (nH=1.4) in displacing [ 3H]-8-OH-DPAT binding (Newman-Tancredi et al., 1998; Assie et al., 1999), which may be associated with positive cooperativity (Forsén and Linse, 1995; Assie et al., 1999). We have assessed the binding characteristics and agonist properties of this drug in an attempt to explain its exceptional potency and potential positive cooperativity. In order to assess whether S 14506 showed any selective activation of 5-HT1A receptors coupled to Gi or Go, 5-HT1A receptor–G protein (Gi1α or Go1α) fusions were used (Milligan, 2000). It is known that 5-HT1A receptors couple to Gi and Go proteins, and it has recently been shown that the receptors couple to Gz proteins in the hypothalamus, in both in vitro and in vivo (Serres et al., 2000). Differential coupling is therefore possible at this receptor (Raymond et al., 1999).

Because of the unique effects observed with S 14506 we have also modelled the compound in the 5-HT1A receptor using the three-site model developed to explain the unique pharmacology of RS-30199, an anxiolytic agent acting at 5-HT1A receptors (Spedding et al., 1998; Jacoby et al., 1999). The three-site model allows the interaction of drugs with different sites within the same receptor, and permits the explanation of several distinct structure/activity relationships between chemical series. This principle has been established for many years in ion channnel pharmacology, where drugs may act at different sites, which interact allosterically, at the alpha subunit of calcium channels. Binding at different sites confers distinct structure/activity, allosteric interactions between the sites, and distinct pharmacological and clinical profiles (Spedding, 1985; Spedding and Paoletti, 1992). Binding at three (or more) sites in G-protein-coupled receptors (GPCRs) provides an analagous rationale for distinct pharmacology, and is an extension of Kenakin's (1995) proposition of differential coupling. The transduction mechanisms of GPCRs are becoming clear at a molecular level. An Asp–Arg–Tyr (DRY in standard one-letter code) domain located at the bottom of transmembrane helix 3 (TM3) is a common feature of rhodopsin-like GPCRs and is considered crucial for G protein activation; activation is dependent on an arginine shift out of a lipophilic pocket at the intracellular side of the receptor (Oliveira et al., 1994). We report that S 14506 is the first drug which may span from the agonist binding site to activate the DRY domain, producing a unique and powerful activation mechanism. Preliminary reports of some of this work have been communicated (Dacquet et al., 1994, Dacquet et al., 1996).

Section snippets

Recombinant h 5-HT1A receptors

Membranes were prepared from recombinant Chinese hamster ovary (CHO) cells stably expressing the human 5-HT1A receptor (Newman-Tancredi et al., 1992, Newman-Tancredi et al., 1997a, Newman-Tancredi et al., 1997b). Cells grown in suspension culture were harvested by centrifugation and homogenised in buffer [20 mM N-(2-hydroxyethyl)piperazine-N′-(2-ethanesulfonic acid) (HEPES), pH 7.5 and 5 mM MgSO4] using a Kinematica Polytron. The homogenate was centrifuged at 50,000g for 30 min and the membrane

Hippocampal membranes

[ 3H]-S 14506 binding to 5-HT1A receptors from rat hippocampal membranes was saturable and its displacement by S 14506 was of high affinity. The Kd for [ 3H]-S 14506 binding was 0.79±0.2 nM (Bmax=400±32 fmol/mg protein). S 14506 displaced 100% of specific [ 3H]-S 14506 binding with a pKi of 9.1±0.2, nH=1.03±0.07. [ 3H]-8-OH-DPAT specific binding was characterised as having a Kd of 1.5±0.5 nM and a Bmax of 303±20 fmol/mg protein. However, the specific binding of [ 3H]-S 14506 was increased in a

Discussion

S 14506 is one of the most potent agonists at 5-HT1A receptors thus far reported. The compound is highly potent in both in vitro and in vivo conditions (Colpaert et al., 1992; Goudie and Leathley, 1994; Schreiber et al., 1994, Schreiber et al., 1995; Lima et al., 1997; Newman-Tancredi et al., 1998; Assie et al., 1999). [ 3H]-S 14506 is a highly potent radioligand and S 14506 was the most potent agent in displacing [ 3H]-S 14506 binding. Spiperone, a close structural analogue and an inverse

References (65)

  • D.A. Horstman et al.

    An aspartate conserved among G-protein coupled receptors confers allosteric regulation of α2-adrenergic receptors by sodium

    Journal of Biological Chemistry

    (1990)
  • J.A. Javitch et al.

    Constitutive activation of the β2 adrenergic receptor alters the orientation of its sixth membrane-spanning segment

    Journal of Biological Chemistry

    (1997)
  • T. Kenakin

    The nature of agonist–receptor efficacy. II. Agonist trafficking of receptor signals

    TiPS

    (1995)
  • M.A. Kjelsberg et al.

    Constitutive activation of the alpha 1B-adrenergic receptor by all amino acid substitutions at a single site. Evidence for a region which constrains receptor activation

    Journal of Biological Chemistry

    (1992)
  • K. Konvicka et al.

    A proposed structure for transmembrane segment 7 of G protein-coupled receptors incorporating an Asn–Pro/Asp–Pro motif

    Biophysics Journal

    (1998)
  • O.H. Lowry et al.

    Protein measurement with the Folin phenol reagent

    Journal of Biological Chemistry

    (1951)
  • G. Milligan

    Insights into ligand pharmacology using receptor–G protein fusion proteins

    Trends in Pharmacological Sciences

    (2000)
  • A.T. Newman-Tancredi et al.

    Agonist and inverse agonist efficacy at human recombinant serotonin 5-HT1A receptors as a function of receptor G-protein stoichiometry

    Neuropharmacology

    (1997)
  • A.T. Newman-Tancredi et al.

    Agonist and antagonist actions of antipsychotic agents at 5-HT1A receptors: a [35S]GTPγS binding study

    European Journal of Pharmacology

    (1998)
  • L. Oliveira et al.

    A common step for signal transduction in G protein-coupled receptors

    Trends in Pharmacological Sciences

    (1994)
  • R. Schreiber et al.

    The potent activity of the 5-HT1A receptor agonists, S 14506 and S 14671, in the rat forced swin test is blocked by novel 5-HT1A receptor antagonists

    European Journal of Pharmacology

    (1994)
  • T.W. Schwartz et al.

    Schwartz and Rosenkilde reply: dogmas versus data in 7TM ‘locks’ for ‘keys’

    Trends in Pharmacological Sciences

    (1996)
  • M. Spedding

    Calcium antagonist subgroups

    Trends in Pharmacological Sciences

    (1985)
  • M. Spedding et al.

    RS-30199, an atypical anxiogenic 5-HT1A agonist, inhibits sexual activity in the male rat

    Neuropharmacology

    (1998)
  • P. Strange

    7TM receptors — locks and keys

    Trends in Pharmacological Sciences

    (1996)
  • H. Sundaram et al.

    Characterization of recombinant human serotonin 5-HT1A receptors expressed in Chinese hamster ovary cells: [3H]spiperone discriminates between G-protein coupled and uncoupled forms

    Biochemical Pharmacology

    (1993)
  • W.N. Tian et al.

    Precoupling of Gi/Go-linked receptors and its allosteric regulation by monovalent cations

    Life Sciences

    (1993)
  • G. Vriend

    WHATIF: a molecular modeling and drug design program

    J. Mol. Graph.

    (1990)
  • R.J. Ward et al.

    An Asp79Asn mutation of the α2A-adrenoceptor interferes equally with agonist activation of individual Giα-family G protein subtypes

    FEBS Letters

    (1999)
  • D.S. Bahia et al.

    Hydrophobicity of residue 351 of the G-protein Gi1α determines the extent of activation by the α2A-adrenoceptor

    Biochemistry

    (1998)
  • T. Branchek

    Site-directed mutagenesis of serotonin receptors

    Med. Chem. Res.

    (1993)
  • P.K. Chanda et al.

    Identification of residues important for ligand binding to the human 5-hydroxytryptamine 1A serotonin receptor

    Molecular Pharmacology

    (1993)
  • Cited by (20)

    • Functional binding assays for estimation of the intrinsic efficacy of ligands at the 5-HT<inf>1A</inf> receptor: Application for screening drug candidates

      2014, Journal of Pharmacological and Toxicological Methods
      Citation Excerpt :

      Looking forward for explaining the residual difference between the two methods, we hypothesized that the differences in the composition of the incubation medium used could be relevant. To test this hypothesis, we performed a GTP-shift assay for 5-HT in the presence of millimolar concentrations of Ca2 + and Mn2 + because these divalent cations are reported to increase the affinity of agonists (Hall et al., 1985; Milligan et al., 2001) and they were used in our experiments with [3H]-8-OH-DPAT in the Ki ratio assay (Table 1). The GTP shift measured by this new IC50 ratio (calculated as in Table 4) was now equal to 63.5 ± 16.4 (mean ratio from 4 independent experiments) as a consequence of the higher affinity of 5-HT for the high-affinity sites in the absence of GTP when adding the divalent cations (from 23 nM to 8.7 nM).

    • Dimer-based model for heptaspanning membrane receptors

      2005, Trends in Biochemical Sciences
    View all citing articles on Scopus
    1

    Present address: Novartis Pharma AG, WSJ-507.2.51, CH-4002 Basel, Switzerland.

    View full text