Kathryn W. Schenck

Pharmacological Characterization of the Competitive GLUK5 Receptor Antagonist Decahydroisoquinoline LY466195 in Vitro and in Vivo

The excitatory neurotransmitter glutamate has been implicated in both migraine and persistent pain. The identification of the kainate receptor GLUK5 in dorsal root ganglia, the dorsal horn, and trigeminal ganglia makes it a target of interest for these indications. We examined the in vitro and in vivo pharmacology of the competitive GLUK5-selective kainate receptor antagonist LY466195 [(3S,4aR,6S,8aR)-6-[[(2S)-2-carboxy-4,4-difluoro-1-pyrrolidinyl]-methyl]decahydro-3-isoquinolinecarboxylic acid)], the most potent GLUK5 antagonist described to date. Comparisons were made to the competitive GLUK5/α-amino-3-hydroxy-5-methyl-4-isoxazole propionic acid (AMPA) receptor antagonist LY293558 [(3S,4aR,6R,8aR)-6-[2-(1(2)H-tetrazole-5-yl)ethyl]-decahydroisoquinoline-3-carboxylic acid], other decahydroisoquinoline GLUK5 receptor antagonists, and the noncompetitive AMPA receptor antagonist LY300168 [1-(4-aminophenyl)-4-methyl-7,8-methylenedioxy-5H-2,3-benzodi-azepine]. When characterized electrophysiologically in rat dorsal root ganglion neurons, LY466195 antagonized kainate (30 μM)-induced currents with an IC50 value of 0.045 ± 0.011 μM. In HEK293 cells transfected with GLUK5, GLUK2/GLUK5, or GLUK5/GLUK6 receptors, LY466195 produced IC50 values of 0.08 ± 0.02, 0.34 ± 0.17, and 0.07 ± 0.02 μM, respectively. LY466195 was efficacious in a dural plasma protein extravasation (PPE) model of migraine with an ID100 value of 100 μg/kg i.v. LY466195 was also efficacious in the c-fos migraine model, with a dose of 1 μg/kg i.v. significantly reducing the number of Fos-positive cells in the rat nucleus caudalis after electrical stimulation of the trigeminal ganglion. Furthermore, LY466195 showed no contractile activity in the rabbit saphenous vein in vitro. The diethyl ester prodrug of LY466195 was also efficacious in the same PPE and c-fos models after oral administration at doses of 10 and 100 μg/kg, respectively while having no N-methyl-d-aspartate antagonist-like behavioral effects at oral doses up to 100 mg/kg.

Effects of pinacidil on serotonin-induced contractions and cyclic nucleotide levels in isolated rat aortae: comparison with nitroglycerin, minoxidil, and hydralazine.

Pinacidil is a novel, clinically effective vasodilator used for the treatment of hypertension whose mechanism of action has not been precisely defined. In vitro, pinacidil (ED50 = 0.3 μM) was ∼30-fold less potent than nitroglycerin and 700-fold more potent than minoxidil or hydralazine in relaxing rat aortic strip preparations. Aortic relaxations produced by nitroglycerin and acetylcholine were dramatically antagonized by methylene blue (10-5 M), an inhibitor of soluble guanylate cyclase. In contrast, relaxation to hydralazine or minoxidil was unaffected and relaxation to pinacidil was only modestly inhibited (∼threefold) by methylene blue (10-5 M). Furthermore, aortic relaxation to pinacidil was similar in preparations with and without an intact endothelium. Relaxation induced by pinacidil (10-7-10-4 M) was not associated with any elevation in either cyclic AMP (cAMP) or cyclic GMP (cGMP) levels in vitro, although nitroglycerin (10-6 M) but not minoxidil (10-3 M) or hydralazine (10-3 M) significantly elevated cGMP levels. Thus, pinacidil was a potent relaxant agonist in vitro, in contrast to minoxidil and hydralazine, which were considerably weaker in this regard. Vascular relaxation produced by pinacidil was independent of an intact endothelium and was not associated with elevations in either cAMP or cGMP. These data are consistent with the proposal that the antihypertensive activity of pinacidil is due to nonspecific arterial vasodilation.

Serotonin-induced contraction in canine coronary artery and saphenous vein: Role of a 5-HT1D-like receptor

The identity of the serotonin (5-HT) receptor(s) that mediate(s) contraction in canine coronary artery and saphenous vein remains controversial. Ring segments of endothelium-denuded coronary artery and helical strips of saphenous vein were suspended in organ chambers for measurement of isometric force. 5-HT, alpha Me-5-HT and sumatriptan contracted both coronary artery and saphenous vein and the non-selective 5-HT receptor antagonist 1-naphthylpiperazine (100nM) blocked 5-HT- and sumatriptan-induced contraction in both tissues. The agonist rank order potency for contraction (5-HT > sumatriptan > alpha Me5-HT > 5-MeOT > 5-MeT) was similar in both tissues and was consistent with that for a 5-HT1D receptor. Oligonucleotide primers specific for the 5-HT1D receptor sequence were designed for use in a polymerase chain reaction (PCR). cDNA derived from total RNA or mRNA from canine tissues was used in the PCR. PCR resulted in the amplification of a 632 base pair sequence in both canine coronary artery and saphenous vein; consistent with that expected for the 5-HT1D receptor. Southern blot analysis, with an oligonucleotide probe internal to the sequence amplified by the PCR primers, confirmed that the sequence amplified by PCR was the 5-HT1D receptor. Thus, the 5-HT1D receptor is expressed in canine coronary artery and saphenous vein and taken together with the pharmacological data, supports the possibility that a 5-HT1D-like receptor mediates contraction in these two tissues.

Contractile responses to sumatriptan and ergotamine in the rabbit saphenous vein: effect of selective 5-HT1F receptor agonists and PGF2α

Contractile responses to ergotamine, sumatriptan and the novel 5‐HT1F receptor agonists, LY334370 and LY344864 were examined using the rabbit saphenous vein. Ergotamine (pEC50=8.7±0.06) was 30 fold more potent than 5‐hydroxytryptamine (5‐HT) (pEC50=7.2±0.13) and 300 fold more potent than sumatriptan (pEC50=6.0±0.08) in contracting the rabbit saphenous vein in vitro. The selective 5‐HT1F receptor agonists, LY334370 or LY344864 (up to 10−4 M), did not contract the rabbit saphenous vein. The contractile response to ergotamine in this tissue resulted from activation of both alpha1 and 5‐HT1B/1D receptors based on the observation that prazosin (10−6 M), an α‐adrenoceptor antagonist, and GR127935 (10−8 M) a 5‐HT1B/1D receptor antagonist, dextrally shifted the contractile response to ergotamine. In contrast, prazosin (10−6 M) did not alter contraction to sumatriptan whereas GR127935 (10−8 M) was a potent antagonist (−log KB=10.0) suggesting that sumatriptan‐induced contraction of the rabbit saphenous vein was mediated only by activation of receptors similar or identical to 5‐HT1B/1D receptors. PGF2α (3×10−7 M) produced a modest increase (approximately 5.0–10.0% maximum PGF2α contraction) in saphenous vein force. Precontraction with PGF2α (3×10−7 M) dramatically augmented the potency and maximal contractile response to sumatriptan (pEC50=7.1) and modestly enhanced the contractile potency of ergotamine (pEC50=9.0) in the rabbit saphenous vein. However, PGF2α (3×10−7 M) only unmasked a contraction to the 5‐HT1F receptor agonists when concentrations exceeded 10−5 M, concentrations considerably higher than their 5‐HT1F receptor affinities. LY334370 (10−6 M) pretreatment did not alter contraction to either sumatriptan or ergotamine and a higher concentration (10−5 M) of LY334370 or LY344864 inhibited contraction to sumatriptan. Thus, activation of 5‐HT1F receptors will not induce vascular contraction (either alone or following modest tone with PGF2α) or augment contraction to other contractile agonists in the rabbit saphenous vein.

Selective down regulation of vascular β1 adrenergic receptors after prolonged isoproterenol infusion

Prolonged isoproterenol infusion (400 micrograms/kg/h for 4 days) in rats was previously shown to produce a reduction in the sensitivity of both cardiac and vascular beta-adrenergic receptors without affecting responsiveness to alpha 1 agonists or phosphodiesterase inhibitors in either vascular or cardiac muscle. The present study was designed to determine if the loss in beta receptor responsiveness was similar for both beta 1 and beta 2 vascular receptors. The rat jugular vein was previously shown to relax in response to both norepinephrine and isoproterenol with norepinephrine-induced relaxation being mediated by interaction with beta 1 adrenergic receptors and isoproterenol-induced relaxation being mediated by its interaction with beta 2 vascular receptors. Using this preparation, tissues from isoproterenol-infused rats were approximately threefold less responsive to isoproterenol when compared to responses in tissues from saline-treated rats. Relaxation to norepinephrine in jugular veins from isoproterenol-infused rats was virtually abolished relative to the response in saline-treated animals. These data suggest that beta 1-adrenergic receptors in blood vessels are considerably more susceptible to down regulation than are beta 2-adrenergic receptors. This observation may have importance in both the therapy of congestive heart failure, where down regulation of beta-adrenergic receptors has been observed, and in our understanding of the mechanism for the inotropic effects of beta receptor agonists.

In vitro receptor specificity of the 5HT1A selective phenylpiperazine, LY165163.

LY165163, a ligand reported to be selective for the 5HT1A subtype of serotonin receptor, was examined for its ability to interact with 5HT2 receptors in the rat jugular vein and alpha-receptors in the rat aorta. In these smooth muscle preparations, no agonist activity of LY165163 occurred in concentrations up to 10(-5) M. However, LY165163 was an antagonist of serotonin-induced contractions in the jugular vein and of norepinephrine-induced contractions in the rat aorta. The dissociation constant calculated for LY165163 at 5HT2 receptors in the rat jugular vein was 10(-8) M and at alpha-receptors in the rat aorta was 2 X 10(-7) M. Thus, LY165163 is a relatively potent antagonist at vascular 5HT2 sites and possesses appreciable affinity at alpha-receptors. Based on these data, the multiple receptor interactions of LY165163 must be taken into consideration when utilizing this agent as a probe for the 5HT1A subtype of serotonin receptor.

In vitro activity of LY393558, an inhibitor of the 5-hydroxytryptamine transporter with 5-HT1B/1D/2 receptor antagonist properties

1-[2-[4-(6-fluoro-1H-indol-3-yl)-3,6-dihydro-1(2H)-pyridinyl]ethyl]-3-isopropyl-6-(methylsulphonyl)-3,4-dihydro-1H-2,1,3-benzothiadiazine-2,2-dioxide (LY393558) is a potent inhibitor of [3H]5-hydroxytryptamine ([3H]5-HT) uptake into rat cortical synaptosomes (pIC(50)=8.48+/-0.12). It produces a dextral shift of the 5-HT dose-response curves for the binding of GTPgamma[35S] to human 5-HT(1B) (pK(b)=9.05+/-0.14) and 5-HT(1D) (pK(b)=8.98+/-0.07) receptors and inhibits the contractile response of the rabbit saphenous vein to the 5-HT(1B/D) receptor agonist, sumatriptan (pK(b)=8.4+/-0.2). In addition, it is an antagonist at the 5-HT(2A) (pK(i)=7.29+/-0.19) and 5-HT(2B) (pK(i)=7.35+/-0.11) receptors. Presynaptic autoreceptor antagonist activity was demonstrated by its ability to potentiate the K(+)-induced outflow of [3H]5-HT from guinea pig cortical slices (pEC(50)=7.74+/-0.05 nM) in which the 5-HT transporter had been inhibited by a maximally effective concentration of paroxetine. It is concluded that LY393558 should be an effective antidepressant with the potential to produce an earlier onset of efficacy than selective serotonin uptake inhibitors.

RELATIONSHIP BETWEEN 5-HT2A RECEPTOR MRNA DENSITY AND CONTRACTILITY IN TRACHEA AND AORTA FROM GUINEA PIG AND RAT

The present studies document marked differences in contractile responsiveness to serotonin in trachea and aorta between guinea pig and rat. For example, the guinea pig trachea and rat aorta markedly contract in response to serotonin via activation of 5-HT2A receptors. In contrast, the rat trachea and guinea pig aorta only modestly contract to serotonin. The availability of 5-HT2A receptor selective cDNA clones from brain of both guinea pig and rat permitted molecular probes to be designed and PCR amplification studies initiated to identify and quantify 5-HT2A receptor specific mRNA in these tissues. For trachea, 3-fold higher concentrations of 5-HT2A receptor specific mRNA were found in guinea pig relative to rat trachea. These data are consistent with the more profound contractile response to serotonin in guinea pig versus rat trachea and suggest that differences in tracheal contractility to serotonin correlate with the density of 5-HT2A receptor mRNA. In contrast, although rat aorta contracted more dramatically to serotonin than guinea pig aorta, rat aorta possessed a similar concentration of 5-HT2A receptor specific mRNA as compared to guinea pig aorta. Thus, for the aorta, differences in the concentration of 5-HT2A receptor mRNA are not sufficient to explain the observed differences in contractility between tissues from guinea pig and rat. These studies documenting 5-HT2A receptor mRNA in rat trachea and guinea pig aorta, two tissues that do not markedly contract in response to serotonin indicate that 5-HT2A receptor mRNA although present, has not resulted in a receptor capable of mediating a contractile response in these tissues.

Sumatripan and 5-benzyloxytryptamine: contractility of two 5-HT1D receptor ligands in canine saphenous veins

Sumatriptan and 5-benzyloxytryptamine are ligands with high affinity for 5-HT1D receptors in the caudate nucleus. Both compounds contracted canine saphenous veins, in vitro. Benzyloxytryptamine was less potent as a contractile agonist than sumatriptan which was less potent than serotonin. In high concentrations (greater than 10(-5) M) serotonin-induced contraction resulted, in part, from activation of alpha-adrenoceptors as determined by blockade of contraction with prazosin (10(-6) M) and idazoxan (10(-6) M). Likewise, benzyloxytryptamine but not sumatriptan also activated contractile alpha-receptors in the canine saphenous vein. Furthermore, benzyloxytryptamine antagonized contraction to sumatriptan in an apparently non-competitive fashion. Thus, benzyloxytryptamine, although possessing some alpha-receptor agonist activity, like sumatriptan, can interact with serotonin receptors in canine saphenous veins. Although effects of sumatriptan and benzyloxytryptamine quantitatively differed in canine saphenous veins, both agents showed similar affinity and agonist efficacy at 5-HT1D receptors in brain. These studies may reflect potential differences between the 5-HT1D receptor in brain and the 5-HT1-like receptor in canine saphenous veins.

Lack of Sumatriptan-Induced Aortic Contraction or Relaxation: 5-HT 1B Receptor Protein Detected in Endothelium and Smooth Muscle of Vasa Vasorum but Not Aorta

Since 5-HT1B receptor mRNA was reported in rat aorta, and 5-HT1B receptor activation has been linked to vascular contraction, we explored sumatriptan-induced contractility and immunohistochemical density of 5-HT1B receptor protein in rat aorta. Sumatriptan (up to 10(-4) M), a 5-HT1B/1D receptor agonist, did not contract the endothelial intact or denuded rat aorta, even in the presence of L-NAME or after induction of modest tone with PGF2 alpha (10(-6) M). Sumatriptan also did not relax aortic preparations precontract with PGF2 alpha (3 x 10(-6) M) or phenylephrine (3 x 10(-7) M). Using a polyclonal antibody directed against the 5-HT1B receptor, minimal 5-HT1B receptor protein was detected in either the endothelium or smooth muscle of the rat aorta. However, dense 5-HT1B receptor protein was found in the vascular smooth muscle of the vasa vasorum supplying the aorta. Thus, the 5-HT1B receptor mRNA detected in tissue extracts of the rat aorta most likely reflects 5-HT1B receptor expression in the arterioles of the vasa vasorum. These studies support the link between the 5-HT1B receptor and vascular contraction in that the aorta with low density of 5-HT1B receptors lacked responses to sumatriptan, an agonist thought to contract blood vessels via 5-HT1B/1D receptors. Furthermore, caution must be observed when using 5-HT1B receptor mRNA to suggest receptor protein in tissues since this RT-PCR product may be derived predominantly from small blood vessels.