David B. Wainscott

Comparisons of hallucinogenic phenylisopropylamine binding affinities at cloned human 5-HT2A, 5-HT2B and 5-HT2C receptors

Since the classical hallucinogens were initially reported to produce their behavioral effects via a 5-HT2 agonist mechanism (i.e., the 5-HT2 hypothesis of hallucinogen action), 5-HT2 receptors have been demonstrated to represent a family of receptors that consists of three distinct subpopulations: 5-HT2A, 5-HT2B, and 5-HT2C receptors. Today, there is greater support for 5-HT2A than for 5-HT2C receptor involvement in the behavioral effects evoked by these agents. However, with the recent discovery of 5-HT2B receptors, a new question arises: do classical hallucinogens bind at 5-HT2B receptors? In the present study we examined and compared the binding of 17 phenylisopropylamines at human 5-HT2A, 5-HT2B, and 5-HT2C receptors. Although there was a notable positive correlation (r>0.9) between the affinities of the agents at all three populations of 5-HT2 receptors, structural modification resulted only in small differences in 5-HT2B receptor affinity such that the range of affinities was only about 50-fold. As with 5-HT2A and 5-HT2C receptor affinity, there is a significant correlation (r>0.9, n=8) between 5-HT2B receptor affinity and human hallucinogenic potency. Nevertheless, given that 5-HT2A and 5-HT2A/2C antagonists – antagonists with low affinity for 5-HT2B receptors – have been previously shown to block the stimulus effects of phenylisopropylamine hallucinogens, it is likely that 5-HT2A receptors play a more prominent role than 5-HT2B and 5-HT2C receptors in mediating such effects despite the affinity of these agents for all three 5-HT2 receptor subpopulations.

Pharmacologic Characterization of the Cloned Human Trace Amine-Associated Receptor1 (TAAR1) and Evidence for Species Differences with the Rat TAAR1

The hemagglutinin-tagged human trace amine-associated receptor1 (TAAR1) was stably coexpressed with rat Gαs in the AV12-664 cell line, and receptor activation was measured as the stimulation of cAMP formation. After blockade of endogenously expressed α2- and β-adrenoceptors with 2-[2-(2-methoxy-1,4-benzodioxanyl)]-imidazoline hydrochloride (2-methoxyidazoxan, RX821002) and alprenolol, respectively, the resulting pharmacology was consistent with that of a unique receptor subtype. β-Phenylethylamine (β-PEA), the putative endogenous ligand, gave an EC50 of 106 ± 5 nM in the assay. For a series of β-PEA analogs used to explore the pharmacophore, small substituents at ring positions 3 and/or 4 generally resulted in compounds having lower potency than β-PEA, although several were as potent as β-PEA. However, small substituents at ring position 2 resulted in a number of compounds having potencies as good as or better than β-PEA. A number of nonselective antagonists known to share affinity for multiple monoaminergic receptors were evaluated for their ability to inhibit β-PEA stimulation of the human TAAR1. None had an IC50 <10 μM. For comparison, the rat TAAR1 receptor was expressed in the AV12-664 cell line. A number of agonist compounds had significantly different relative potencies between the rat and human TAAR1, demonstrating a significant species difference between the rat and human TAAR1. The TAAR1 receptor exhibits a pharmacologic profile uniquely different from those of classic monoaminergic receptors, consistent with the structural information that places them in a distinct family of receptors. This unique pharmacologic profile suggests the potential for development of TAAR-selective agonists and antagonists to study their physiologic roles.

Receptor subtype and density determine the coupling repertoire of the 5-ht2 receptor subfamily

The 5-Hydroxytryptamine (5-HT)2C receptor (originally known as the 5-HT1C receptor) is a member of the 5-HT2 subfamily of G protein coupled receptors, which is known to couple to phospholipase C. Within the 5-HT2 subfamily, only the 5-HT2C receptor also coupled to inhibition of forskolin-stimulated cAMP production when expressed at high density (12 pmol/mg membrane protein) in stably transformed AV12 cells. The 5-HT2C receptor coupled with high efficacy to both phospholipase C as measured by IP3 (inositol 1,4,5-trisphosphate) production and to inhibition of forskolin-stimulated cAMP production (EC50 = 2.98 nM +/- 0.9 and IC50 = 47.99 nM +/- 10.25 respectively). The 5-HT2A and 5-HT2B receptors, while coupling to phospholipase C with high affinity (EC50s of 19.24 nM +/- 6.44 and 1.24 nM +/- 0.136 respectively), did not decrease adenylyl cyclase activity. The 5-HT2C receptor actions in both systems showed the expected pharmacology for the 5-HT2C receptor, e.g., mesulergine antagonized the effects of 5-HT and spiperone did not. Preincubation of cells with PTX showed that the G protein coupling of the 5-HT2C receptor to phospholipase C is PTX insensitive, while the G protein coupling to inhibition of adenylyl cyclase is PTX sensitive, even to concentrations as low as 20 ng/ml of PTX. PTX pretreatment of the 5-HT2C bearing cells also unmasked a small stimulatory effect on adenylyl cyclase. When expressed at low density the 5-HT2C receptor potentiated forskolin-stimulated cAMP production by 2 fold while still maintaining its ability to enhance PI hydrolysis. A more modest potentiation of cAMP production was noted with low density expression of the 5-HT2B receptor. Thus the ability of the 5-HT2C receptor to interact with several effectors through at least two different G proteins is, in part, receptor subtype specific but also influenced by receptor density.

Preclinical pharmacological profile of the selective 5-HT1F receptor agonist lasmiditan

Introduction: Lasmiditan (also known as COL-144 and LY573144; 2,4,6-trifluoro-N-[6-[(1-methylpiperidin-4-yl)carbonyl]pyridin-2yl]benzamide) is a high-affinity, highly selective serotonin (5-HT) 5-HT1F receptor agonist. Results: In vitro binding studies show a Ki value of 2.21 nM at the 5-HT1F receptor, compared with Ki values of 1043 nM and 1357 nM at the 5-HT1B and 5-HT1D receptors, respectively, a selectivity ratio greater than 470-fold. Lasmiditan showed higher selectivity for the 5-HT1F receptor relative to other 5-HT1 receptor subtypes than the first generation 5-HT1F receptor agonist LY334370. Unlike the 5-HT1B/1D receptor agonist sumatriptan, lasmiditan did not contract rabbit saphenous vein rings, a surrogate assay for human coronary artery constriction, at concentrations up to 100 µM. In two rodent models of migraine, oral administration of lasmiditan potently inhibited markers associated with electrical stimulation of the trigeminal ganglion (dural plasma protein extravasation, and induction of the immediate early gene c-Fos in the trigeminal nucleus caudalis). Conclusions: Lasmiditan presents a unique pyridinoyl-piperidine scaffold not found in any other antimigraine class. Its chemical structure and pharmacological profile clearly distinguish it from the triptans. The potency and selectivity of lasmiditan make it ideally suited to definitively test the involvement of 5-HT1F receptors in migraine headache therapy.

Human 5-HT1F receptor-stimulated [35S]GTPγS binding: correlation with inhibition of guinea pig dural plasma protein extravasation

To determine the potency and efficacy of 5-HT1F receptor ligands, a [35S]GTPgammaS binding assay was developed and optimized for the human 5-HT1F receptor. Compounds which are known to be effective in the abortive treatment of migraine were tested for efficacy and potency in this assay. Naratriptan, sumatriptan, zolmitriptan, and rizatriptan all had agonist activity. The 5-HT1F receptor ligand LY334370 (4-fluoro-N-[3-(1-methyl-4-piperidinyl)-1H-indol-5-yl]-benzamide) was the most potent compound tested with an EC50 of 2.13 +/- 0.15 nM. LY302148 (5-fluoro-3-[1-[2-(1-methyl-1H-pyrazol-4-yl)ethyl]-4-piperidinyl]-1H-ind ole), methysergide, LY306258 (3-dimethylamino-2,3,4,9-tetrahydro-1H-carbazol-6-ol), dihydroergotamine (DHE), L-694,247 and CP-122,288 were also investigated for potency and efficacy. There was a statistically significant correlation between the pEC50 for the stimulation of [35S]GTPgammaS binding and the pID50 for the inhibition of trigeminal nerve-stimulated dural plasma protein extravasation in the guinea pig. In the course of these studies, it was found that the purportedly selective 5-HT1D receptor antagonist GR127935 inhibited 5-HT1F receptor-stimulated [35S]GTPgammaS binding with a Ki of 39.6 +/- 9.5 nM. These studies demonstrate that 5-HT1F receptor-mediated stimulation of [35S]GTPgammaS binding in a clonal cell system is a reproducible, high throughput assay that is predictive of an in vivo model of 5-HT1F receptor activation.

Neurogenic dural protein extravasation induced by meta-chlorophenylpiperazine (mCPP) involves nitric oxide and 5-HT2B receptor activation

The compound m-chlorophenylpiperazine (mCPP), which is known to trigger migraine-like head pain in some subjects, was evaluated for its ability to induce dural plasma protein extravasation (PPE) in guinea pigs. Intravenous mCPP dose-dependently increased PPE. This effect was inhibited by non-selective 5-HT2 receptor antagonists (methysergide, LY53857, LY215840), by a peripherally restricted 5-HT2 receptor antagonist (xylamidine) and by a 5-HT2B selective receptor antagonist (LY202146). These data suggests that peripheral 5-HT2B receptors mediate mCPP-induced PPE. The nitric oxide synthase inhibitor L-NAME and 5-HT1 agonist sumatriptan also blocked mCPP-induced PPE, suggesting a role for nitric oxide (NO) and the trigeminal system, respectively. NO release has been linked to activation of the 5-HT2B receptor on the vascular endothelium. However, LY202146 did not block PPE induced by electrical stimulation of the trigeminal ganglion. These data are consistent with activation of peripheral 5-HT2B receptors initiating PPE and the theory that selective 5-HT2B antagonists might be effective prophylactic therapies for migraine.

[3H]Rauwolscine: an antagonist radioligand for the cloned human 5-hydroxytryptamine2B (5-HT2B) receptor

Abstract In previous reports, [3H]5-HT has been used to characterize the pharmacology of the rat and human 5-HT2B receptors. 5-HT, the native agonist for the 5-HT2B receptor, has a limitation in its usefulness as a radioligand since it is difficult to study the agonist low-affinity state of a G protein-coupled receptor using an agonist radioligand. When using [3H]5-HT as a radioligand, rauwolscine was determined to have relatively high affinity for the human receptor (Ki human = 14.3 ± 1.2 nM, compared to Ki rat = 35.8 ± 3.8 nM). Since no known high affinity antagonist was available as a radioligand, these studies were performed to characterize [3H]rauwolscine as a radioligand for the cloned human 5-HT2B receptor expressed in AV12 cells. When [3H]rauwolscine was initially tested for its usefulness as a radioligand, complex competition curves were obtained. After testing several α2-adrenergic ligands, it was determined that there was a component of [3H]rauwolscine binding in the AV12 cell that was due to the presence of an endogenous α2-adrenergic receptor. The α2-adrenergic ligand efaroxan was found to block [3H]rauwolscine binding to the α2-adrenergic receptor without significantly affecting binding to the 5-HT2B receptor and was therefore included in all subsequent studies. In saturation studies at 37° C, [3H]rauwolscine labeled a single population of binding sites, Kd = 3.75 ± 0.23 nM. In simultaneous experiments using identical tissue samples, [3H]rauwolscine labeled 783 ± 10 fmol of 5-HT2B receptors/mg of protein, as compared to 733 ± 14 fmol of 5-HT2B receptors/mg of protein for [3H]5-HT binding. At 0° C, where the conditions for [3H]5-HT binding should label mostly the agonist high affinity state of the human 5-HT2B receptor, [3H]rauwolscine (Bmax = 951 ± 136 fmol/ mg), again labeled significantly more receptors than [3H]5-HT (Bmax = 615 ± 34 fmol/mg). The affinity of [3H]rauwolscine for the human 5-HT2B receptor at 0° C did not change, Kd = 4.93 ± 1.27 nM, while that for [3H]5-HT increased greatly (Kd at 37° C = 7.76 ± 1.06 nM; Kd at 0° C = 0.0735 ± 0.0081 nM). When using [3H]rauwolscine as the radioligand, competition curves for antagonist structures modeled to a single binding site, while agonist competition typically resulted in curves that best fit a two site binding model. In addition, many of the compounds with antagonist structures displayed higher affinity for the 5-HT2B receptor when [3H]rauwolscine was the radioligand. Typically, ∼ 85% of [3H]rauwolscine binding was specific binding. These studies display the usefulness of [3H]rauwolscine as an antagonist radioligand for the cloned human 5-HT2B receptor. This should provide a good tool for the study of both the agonist high- and low-affinity states of the human cloned 5-HT2B receptor.

Mutations of transmembrane IV and V serines indicate that all tryptamines do not bind to the rat 5-HT2A receptor in the same manner

Two mutations of the rat serotonin 5-HT2A receptor were made, expressed and examined for their ability to bind and be stimulated by certain tryptamines as well as their ability to bind antagonists. Mutation of Ser207 to an Ala (S207A) resulted in no substantial changes in binding of either 5-HT2A antagonists or agonists. In contrast, mutation of Ser239 to an Ala (S239A) resulted in significant changes in the 5-HT2A receptor with some but not all agonists and antagonists examined. Specifically, 5-HT had decreased affinity for the S239A mutated 5-HT2A receptor, showing over a 10-fold decrease in receptor-binding displacement, while still being capable of stimulating IP3 formation. However, the agonists tryptamine, 5-methoxytryptamine (5-MeOT), and N-1-isopropyl-5-methoxytryptamine; and the antagonists ketanserin, LY 86057, and LY 53857 were significantly less affected by a S239A mutation. These results suggest that while 5-HT might have a direct interaction with the Ser239 of the 5-HT2A receptor, tryptamine and 5-MeOT interact with this receptor in a different manner.

A novel class of 5-HT2a receptor antagonists: Aryl aminoguanidines

Local delivery of serotonin (5-HT) produces a rapid edematous response in soft tissues via increased fluid extravasation which is prevented by 5-HT2 antagonists such as ketanserin or mianserin. Here we report the effects of a new class of aminoguanidine 5-HT2 antagonists, with relative selectivity for 5-HT2A receptors which are potent inhibitors of 5-HT-induced paw edema in the rat. Radioligand binding studies with 125I DOI on human 5-HT2A and 5-HT2C receptors and with 3H-5-HT on human 5-HT2B receptors demonstrated that, LY314228, and LY320954 displayed some selectivity for the 5-HT2A receptor. When compared to binding at other 5-HT2 receptor subtypes, LY314228 had an 18.6-fold greater affinity for the 5-HT2A site over the 5-HT2B site, and 2.6 fold greater at the 5-HT2C site. LY320954 displayed similar preference for 5-HT2A sites. Both compounds also inhibited 5-HT-induced paw swelling in rats, with ED50s of 6.4 and 4.8 mg/kg (for LY314228 and LY320954, respectively). These studies offer evidence for a novel class of pharmacophores for the 5-HT2 receptor family which show greater relative affinities for the 5-HT2A receptor subclass.

Positive Allosteric Modulation of the Glucagon-like Peptide-1 Receptor by Diverse Electrophiles

Therapeutic intervention to activate the glucagon-like peptide-1 receptor (GLP-1R) enhances glucose-dependent insulin secretion and improves energy balance in patients with type 2 diabetes mellitus. Studies investigating mechanisms whereby peptide ligands activate GLP-1R have utilized mutagenesis, receptor chimeras, photo-affinity labeling, hydrogen-deuterium exchange, and crystallography of the ligand-binding ectodomain to establish receptor homology models. However, this has not enabled the design or discovery of drug-like non-peptide GLP-1R activators. Recently, studies investigating 4-(3-benzyloxyphenyl)-2-ethylsulfinyl-6-(trifluoromethyl)pyrimidine (BETP), a GLP-1R-positive allosteric modulator, determined that Cys-347 in the GLP-1R is required for positive allosteric modulator activity via covalent modification. To advance small molecule activation of the GLP-1R, we characterized the insulinotropic mechanism of BETP. In guanosine 5′-3-O-(thio)triphosphate binding and INS1 832-3 insulinoma cell cAMP assays, BETP enhanced GLP-1(9–36)-NH2-stimulated cAMP signaling. Using isolated pancreatic islets, BETP potentiated insulin secretion in a glucose-dependent manner that requires both the peptide ligand and GLP-1R. In studies of the covalent mechanism, PAGE fluorography showed labeling of GLP-1R in immunoprecipitation experiments from GLP-1R-expressing cells incubated with [3H]BETP. Furthermore, we investigated whether other reported GLP-1R activators and compounds identified from screening campaigns modulate GLP-1R by covalent modification. Similar to BETP, several molecules were found to enhance GLP-1R signaling in a Cys-347-dependent manner. These chemotypes are electrophiles that react with GSH, and LC/MS determined the cysteine adducts formed upon conjugation. Together, our results suggest covalent modification may be used to stabilize the GLP-1R in an active conformation. Moreover, the findings provide pharmacological guidance for the discovery and characterization of small molecule GLP-1R ligands as possible therapeutics.