Photoaffinity labeling of serotonin-binding proteins

NCS-MPP (4-(2'-methoxy-phenyl)-1-[2'-(N-2"-pyridyl)-p-isothiocyanobenz amido]-ethyl-piperazine): a high affinity and irreversible 5-HT1A receptor ligand

A novel irreversible 5-HT1A receptor binding ligand, NCS-MPP (4-(2'- methoxy-phenyl)-1-[2'-(N-2"-pyridyl)-p-isothiocyanobenzamido]- ethyl-piperazine), based on the new 5-HT1A receptor antagonist p-MPPI (4-(2'-methoxy-phenyl)-1-[2'-(N-2"-pyridyl)-p-iodobenzamido]-ethyl -piperazine ), was synthesized, and its binding characteristics were evaluated using in vitro homogenate binding with rat hippocampal membranes. The Ki value of NCS-MPP was estimated to be 1.8 +_ 0.2 nM using analysis of concentration-dependent inhibition for the binding of [125I]p-MPPI to 5-HT1A receptors. NovaScreen of NCS-MPP showed low to moderate binding affinities to alpha-1, alpha-2-adrenergic and 5-HT2 receptors, with Ki values of 350, 420, and 103 nM, respectively. These data strongly suggest that the ligand bound to 5-HT1A receptors with high affinity and high selectivity. Irreversible inhibition of [125I]p-MPPI binding by NCS-MPP following a 5 min incubation at room temperature was concentration dependent; the inhibition increased to 50% at a concentration less than 10 nM, and became more pronounced (90%) at 400 nM. Under similar assay conditions, NCS-MPP was significantly less efficient in irreversibly inhibiting agonist ligand [125I]8-OH-PIPAT binding to 5-HT1A receptors at lower concentrations (<10nM). After pretreatment of membranes with a low concentration of NCS-MPP (2nM), there was an apparent loss of [125I]p-MPPI binding sites, as expected, but no change in the binding affinity (Kd) was observed. However, the significant increase in Kd at a higher concentration of NCS-MPP (50 nM) indicated that there may be a secondary alkylation site, which may not be directly involved in p-MPPI binding to receptors; nevertheless, it would lead to an increased Kd value. The availability of an irreversible ligand, NCS-MPP, may provide a useful tool for studies of 5-HT1A receptors in the central nervous system.

Photoaffinity labeling of the two forms of serotonin binding protein: peptide mapping of the binding sites

Serotonin binding protein (SBP) is a vesicular protein found in neurectoderm-derived cells that store 5-hydroxytryptamine (5-HT, serotonin), such as central and peripheral serotonergic neurons and paraneurons (parafollicular cells of the thyroid). 5-HT is stored as a complex with SBP in vivo. Two forms of the protein are found. These differ in molecular mass: one is 45 kDa and the other 56 kDa. It has been suggested that the 56-kDa form of SBP may be the precursor of the 45-kDa form. To study the relationship between these two proteins, we have used a covalently bound radiolabeled probe to analyze their binding domains. A photoaffinity reagent, N-(4-azido-2-nitrophenyl)-5-hydroxytryptamine (NAP-5-HT), was synthesized and characterized by nuclear magnetic resonance spectroscopy, mass spectra, and UV-visible absorption spectra. A 1 M excess of NAP-5-HT inhibited the binding of [3H]5-HT to SBP by 50%. NAP[3H]5-HT was also synthesized and attached to both high- and low-affinity binding sites on both forms of SBP. The high-affinity constants for 45-kDa and 56-kDa proteins were 0.8 nM and 0.02 nM, respectively, whereas the low-affinity constants were 0.3 microM and 0.15 microM. When the high-affinity site of partially purified SBP was photoaffinity-labeled with the reagent, two covalently labeled proteins (45 kDa and 56 kDa) were found by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE). Inhibition of the labeling of both proteins by 50% was observed in the presence of a 15-fold molar excess of 5-HT.(ABSTRACT TRUNCATED AT 250 WORDS)

Identification of 5-hydroxytryptamine1A receptor proteins in bovine frontal cortex

5-Hydroxytryptamine1A (5-HT1A) receptor proteins were identified by a novel approach in which photoaffinity labeling technique was used in conjunction with affinity column chromatography. 5-HT1A receptors were solubilized from bovine frontal cortical membranes with 0.3% digitonin and 0.1% Nonidet P-40, and bound effectively to 1-[2-(4-aminophenyl)ethyl]-4-(3-trifluoromethylphenyl)piperazine (PAPP)-coupled Affi-Gel 10 in a time-dependent manner. PAPP was shown previously to be a selective ligand for the 5-HT1A receptor. Two protein bands with molecular masses of approximately 55,000 and 38,000 daltons revealed on sodium dodecyl sulfate-polyacrylamide gel electrophoresis were eluted from the affinity column with either 1 mM 5-HT or 1 microM [3H]1-[2-(4-azidophenyl)ethyl]-4-(3-trifluoromethyl-phenyl)piperazine ([3H]p-azido-PAPP). [3H]p-Azido-PAPP is a selective photoaffinity labeling probe for the 5-HT1A receptor. The intensity of these two protein bands and the incorporation of [3H]p-azido-PAPP into these two proteins decreased significantly when the solubilized fraction was preincubated with excess 5-HT or PAPP (saturating all 5-HT1A receptors) prior to affinity column chromatography. These results suggest strongly that these two proteins are related to the 5-HT1A receptor protein. The isoelectric points of the photolabeled 5-HT1A receptor proteins were 6.0 and 6.5.

Purification and reconstitution of serotonin receptors from bovine brain

An affinity-chromatography column was used to isolate and purify 5-hydroxytryptamine (serotonin, 5-HT) receptors from bovine brain frontal cortex. The affinity ligand lysergic acid ethylamidoethylbromide was synthesized and coupled to an agarose matrix via a thioether bond. Receptors in the crude cortical membrane fragments were solubilized using 3-[(3-cholamidopropyl)-dimethylammonio]-1-propanesulfonate (CHAPS), affinity purified, and reconstituted into lipid vesicles. [3H]5-HT binding analysis indicates a single class of high-affinity binding site (Kd, 16.9 nM) that was reconstituted. 5-Methoxytryptamine, a competitor for high-affinity serotonin sites, inhibited this binding and showed a Ki of 27.4 nM. Ketanserin, a high-affinity ligand for 5-HT2 type receptors, was ineffective in displacing [3H]5-HT binding at concentrations up to 4 microM indicating a 5-HT1 receptor as the primary receptor type isolated. The average specific activity of 359 pmol/mg in the reconstituted fractions is an enrichment of 1062-fold over crude membrane fragments. Sodium dodecyl-sulfate electrophoresis showed the presence of four proteins in the reconstituted vesicles with approximate relative Mr values of 63,000, 70,000, 81,000, and 94,000.

Identification of the 5-HT1A receptor binding subunit in rat brain membranes using the photoaffinity probe [3H]8-methoxy-2-[N-n-propyl, N-3-(2-nitro-4-azidophenyl)aminopropyl]aminotetralin

The synthesis of a tritiated derivative of the 5-HT1A photoaffinity probe 8-methoxy-2-[N-n-propyl, N-3-(2-nitro-4-azidophenyl)aminopropyl]aminotetralin ([3H]8-methoxy-3'-NAP-amino-PAT) allowed the use of this probe for attempting the irreversible labeling of specific binding sites in rat brain membranes. Sodium dodecyl-sulfate-polyacrylamide gel electrophoresis of proteins solubilized from hippocampal microsomal membranes that had been incubated with 20 nM [3H]8-methoxy-3'-NAP-amino-PAT under UV light revealed a marked incorporation of 3H label into a 63-kilodalton protein termed PI. As expected of a possible correspondence between PI and 5-HT1A receptor binding sites, 3H labeling by the photoaffinity probe could be prevented by selective 5-HT1A ligands such as 8-hydroxy-2-(di-n-propylamino)tetralin, ipsapirone, buspirone, and gepirone and by N-ethylmaleimide, but not by the 5-HT2 antagonist ketanserin, noradrenaline- and dopamine-related drugs, monoamine oxidase inhibitors, and chlorimipramine. Furthermore, the regional and subcellular distributions of PI were identical to those of specific 5-HT1A binding sites. These results indicated that the binding subunit of the 5-HT1A receptor is a 63-kilodalton protein with a functionally important sulfhydryl group(s).

Solubilization of serotonin1a and serotonin1b binding sites from bovine brain

Serotonin1 (5-hydroxytryptamine1, 5-HT1) binding sites have been solubilized from bovine brain cortex using a mixture of 0.1% Nonidet P-40 and 0.3% digitonin in a low-salt buffer containing 0.1% ascorbic acid. The affinity of [3H]5-HT for the soluble cortical binding sites (2.1 nM) is identical to its affinity at membrane-bound binding sites (2.1 nM). [3H]8-Hydroxy-2-(di-n-propylamino)tetralin ([3H]DPAT), a selective 5-HT1a radioligand, also binds to soluble cortical binding sites with high affinity (1.8 nM) comparable with its affinity in the crude membranes (1.7 nM). A significant correlation exists in the rank order potency of serotonergic agents for [3H]5-HT binding and for [3H]DPAT binding to crude and soluble membranes. The density of [3H]DPAT binding sites relative to the [3H]5-HT sites in the solubilized cortical membranes (35%) corresponds well with the proportion of 5-HT1a sites in the crude membranes determined by spiperone displacement (33%), suggesting that both the 5-HT1a and 5-HT1b binding sites have been cosolubilized. [3H]5-HT binding in the soluble preparations was inhibited by GTP, suggesting that a receptor complex may have been solubilized. [3H]Spiperone-specific binding was not detectable in this preparation, suggesting that 5-HT2 sites were not cosolubilized.

Photoaffinity labelling and solubilization on the central 5-HT1A receptor binding site

Two complementary approaches, covalent labelling and solubilization, have been used to study the biochemical properties of the central 5-HT1A receptor binding site. We have first designed a photoaffinity ligand containing the structure of 8-OH-DPAT, a potent and specific agonist of 5-HT1A sites. Thus, 8-methoxy-2[N-n-propyl,N-3-(2-nitro-4-azido-phenyl)- aminopropyl]aminotetralin or 8-methoxy-3'-NAP-amino-PAT, was found to displace, in the dark, [3H]8-OH-DPAT from 5-HT1A sites in rat hippocampal membranes with an IC50 of 6.6 nM. Under two cumulative UV irradiations (366 nm, for 20 min at 4 degrees C), 8-methoxy-3-'-NAP-amino-PAT (30 nM) blocked irreversibly 55-60% of 5-HT1A binding sites. This blockade was specific of 5-HT1A sites since the other serotoninergic sites, 5-HT1B, 5-HT2 and also the presynaptic 5-HT3 sites were not affected by the treatment. In addition, the binding of [3H]Spiperone and [3H]7-OH-DPAT to striatal dopamine sites remained unchanged under similar photolysis conditions. The tritiated derivative of the photoaffinity ligand (92 Ci/mmol) was then synthesized for the identification of the covalently bound protein(s). SDS-PAGE of solubilized membranes irradiated in the presence of 20 nM 3H-8-methoxy-3'-NAP-amino-PAT allowed the detection of a 63 kD protein whose labelling appeared specific. Thus, 3H-incorporation into the 63 kD band could be prevented by microM concentrations of 5-HT, 8-OH-DPAT and other selective 5-HT1A ligands such as isapirone. In contrast, the 5-HT2 antagonist ketanserin, norepinephrine and dopamine-related ligands (including 7-OH-DPAT) were ineffective. Direct solubilization of 5-HT1A receptor binding sites was also attempted from rat hippocampal membranes. The best results were obtained using CHAPS (10 mM) plus NaCl (0.2 M), which led to 50% recovery of 5-HT1A sites in the 100,000 g supernatant. The pharmacological properties and sensitivity to N-ethyl-maleimide and GppNHp of soluble sites appeared near identical to those of membrane-bound 5-HT1A sites.

Irreversible blockade of central 5-HT1A receptor binding sites by the photoaffinity probe 8-methoxy-3'-NAP-amino-PAT

A new photoaffinity ligand derived from the potent 5-HT agonist, 8-OH-DPAT, has been synthesized. In the dark, this compound, 8-methoxy-2-(N-n-propyl,N-3-(2-nitro-4-azidophenyl)aminopropyl) aminotetralin or 8-methoxy-3'-NAP-amino-PAT, displaced [3H]8-OH-DPAT and [3H]5-HT bound to 5-HT1A and 5-HT1 sites in hippocampal membranes with IC50 values of 6.6 and 18.1 nM respectively. The apparent affinity of 8-methoxy-3'-NAP-amino-PAT for the 5-HT1A binding sites was at least 20 times higher than for the other 5-HT receptor sites (5-HT2 and 5-HT3) or the dopamine-related [3H]spiperone and [3H]7-OH-DPAT binding sites. Under UV irradiation (lambda = 366 nm), 8-methoxy-3'-NAP-amino-PAT produced an irreversible blockade of 5-HT1A sites which could be prevented by prior site occupancy by a saturating concentration (10 microM) of reversible 5-HT ligands such as 5-HT itself, 8-OH-DPAT or LSD. The blockade of 5-HT1A binding sites was concentration-dependent, and two successive irradiations of rat brain membranes in the presence of 30 nM 8-methoxy-3'-NAP-amino-PAT were found to be more efficient that a single exposure to 100 nM of the photosensitive ligand. Thus, a 55-60% irreversible blockade of 5-HT1A binding sites was achieved following 2 cumulative irradiations of hippocampal membranes with 30 nM 8-methoxy-3'-NAP-amino-PAT. Under such conditions, cortical 5-HT2 receptor binding sites as well as striatal 5-HT3 and dopamine-related binding sites remained unaltered.

In situ molecular sizes of the various types of 5-HT binding sites in the rat brain

The radiation inactivation technique has been used to estimate the molecular weights of 5-HT binding sites in various regions of the rat brain. Using 3H-5-HT or 3H-8-OH-DPAT as the ligand, the same molecular weight of 55,000-60,000 daltons was calculated for the postsynaptic 5-HT1A and 5-HT1B sites in the hippocampus and cerebral cortex. Studies with 3H-ketanserin as the selective ligand indicated a molecular weight in the same range for the post-synaptic 5-HT2 binding site in the cerebral cortex. In contrast, a higher value (67,000 daltons) was found for the presynaptic 5-HT3 site selectively labelled by 3H-8-OH-DPAT in the striatum and cerebral cortex. The curvilinear pattern of the radiation-induced inactivation of 5-HT1A and 5-HT1B binding sites suggested that both sites belong to complex polymeric structures. In contrast, the 5-HT2 and 5-HT3 sites may correspond to less cooperative structures since simple monoexponential inactivation curves were observed upon irradiation.

Para-azidoclonidine: a novel photoaffinity ligand for the alpha 2-receptor

Reversible and irreversible interactions of the photoreactive clonidine analogue p-azidoclonidine (PAZC) with brain alpha 2-adrenergic receptors were examined. In the absence of light, PAZC displayed selective, high affinity, competitive interactions with sites labeled by the alpha 2-agonist 3H-p-aminoclonidine (3H-PAC). Reversible binding characteristics resembled those of other alpha 2-agonists. Preincubation of bovine frontal cortex membranes with 100 nM PAZC followed by ultraviolet irradiation and thorough washing decreased alpha 2-receptor density 42% relative to controls receiving irradiation alone. The loss of receptors could be prevented by inclusion of 500 nM phentolamine in the preincubation medium. Alpha 1- and beta-adrenergic receptors were relatively unaffected. PAZC is a potential photoaffinity ligand for the alpha 2-adrenergic receptor.

5-HT-1a and 5-HT-1b selectivity of two phenylpiperazine derivatives: evidence for 5-HT-1b heterogeneity

The ability of m-trifluoromethylphenylpiperazine (TFMPP) and an N-substituted derivative, LY 165163 (p-NH2-PE-TFMPP), to discriminate 5-HT-1 binding sites labelled by [3H]5-HT is compared in rat corpus striatum and rat cortex. TFMPP displays at least a 30 fold selectivity for the 5-HT-1b subtype. Furthermore, TFMPP reveals heterogeneity within the 5-HT-1b binding sites. TFMPP displays a Kd of 6 nM for approximately two-thirds of the 5-HT-1b binding sites and a Kd of 273 nM for the remaining one third of the 5-HT-1b sites. p-NH2-PE-TFMPP, on the other hand, discriminates the 5-HT-1 sites in a manner similar to spiperone, displaying a 110 fold selectivity for the 5-HT-1a sites. p-NH2-PE-TFMPP displays a Kd of about 3 nM for the 5-HT-1a sites. p-NH2-PE-TFMPP does not discriminate subtypes within the 5-HT-1b binding sites. The significance of the selectivity of these two compounds as well as structurally related compounds is discussed.

Irreversible blockade of central 5-HT binding sites by 8-methoxy-2'-chloro-PAT

We have synthesized 8-methoxy-2-(N-2'-chloropropyl, N-propyl) aminotetralin (8-methoxy-2'-chloro-PAT), an alkylating agent derived from the potent 5-HT agonist, 8-hydroxy-2-(N,N-dipropyl)-aminotetralin (PAT). As expected for an irreversible ligand, the blockade of 3H-PAT or 3H-5-HT binding to post-synaptic 5-HT1 (A and B) sites in rat hippocampal membranes pretreated with 8-methoxy-2'-chloro-PAT could not be prevented by extensive washing of membranes. Prior occupancy of 5-HT1 sites by 5-HT or PAT prevented any subsequent irreversible blockade by the alkylating agent. Similar irreversible blockade by 8-methoxy-2'-chloro-PAT was found on 3H-PAT binding to striatal membranes suggesting that presynaptic 5-HT binding sites (see Gozlan et al., Nature, Lond. 305, 140, 1983) were sensitive also to the alkylating agent. In contrast, the modifying agent N-ethylmaleimide (NEM) reduced markedly 3H-PAT binding to postsynaptic hippocampal 5-HT1 sites, but did not alter 3H-PAT binding to striatal presynaptic 5-HT sites. Although 8-methoxy-2'-chloro-PAT bound irreversibly to different classes of 5-HT binding sites (5-HT1A, 5-HT1B, presynaptic sites), it can be considered a selective alkylating agent, since it exerted no action on 3H-spiperone binding to 5-HT2 sites, 3H-muscimol binding to GABA sites, or 3H-flunitrazepam binding to benzodiazepine sites.

A trifluoromethylphenyl piperazine derivative with high affinity for 5-hydroxytryptamine-1A sites in rat brain

The inhibition of [3H]5-hydroxytryptamine [( 3H]5-HT) binding in rat brain by 1-[2-(3-bromoacetamidophenyl)ethyl]-4-(3-trifluoromethylphenyl) piperazine (BrAcTFMPP) and that by spiperone were compared. Spiperone inhibition of [3H]5-HT binding in cortex was consistent with displacement from two sites with dissociation constants (KD) of 24 nM (5-HT-1A site) and 19 microM (5-HT-1B site) for spiperone. BrAcTFMPP also discriminated two subpopulations of [3H]5-HT binding sites with dissociation constants of 0.5 nM and 146 nM for the compound. The proportion of high-affinity sites for each compound represented about 35% of the specific [3H]5-HT binding. In the presence of 1 microM spiperone, a concentration that saturates the 5-HT-1A sites while having a minimal effect on 5-HT-1B sites, BrAcTFMPP displaced [3H]5-HT from a single site with a KD for BrAcTFMPP of 145 nM. The inhibition of [3H]5-HT binding by spiperone in the presence of 30 nM BrAcTFMPP was best fit by a single-site model with a KD of 21 microM for spiperone. In corpus striatum, 5-HT-1A sites, as defined with spiperone, represented 15% of the specific [3H]5-HT binding and 30 nM BrAcTFMPP also blocked about 15% of the binding. A significant difference between spiperone and BrAcTFMPP was their affinity for 5-HT-2 receptors. BrAcTFMPP (KD = 41 nM) had an 80-fold lower affinity for these sites than spiperone (KD = 0.5 nM). Thus, BrAcTFMPP and spiperone discriminate the same two subpopulations of [3H]5-HT binding sites and BrAcTFMPP displays a high affinity and a selectivity for 5-HT-1A sites versus both 5-HT-1B and 5-HT-2 sites.

Interaction of N-(2-nitro-4-azidophenyl)-serotonin with two types of monoamine oxidase in rat brain

The effects of N-(2-nitro-4-azidophenyl) serotonin (NAP-5-HT) on types A and B monoamine oxidase (MAO) in rat brain cortex were studied. In the dark this compound acted as a competitive inhibitor for both types A and B MAO (Ki values of 0.19 microM and 0.21 microM for types A and B MAO, respectively). Upon photolysis, NAP-5-HT became an irreversible inhibitor for only type B MAO. A 50% inhibition was obtained by irradiation of the enzyme in the presence of 35 nM NAP-5-HT. Furthermore the inhibition of type B MAO could be protected by including its substrate phenylethylamine during the irradiation. Under the same photolytic conditions photodependent inhibition of type A MAO by NAP-5-HT was not clearly observed. These results provide further evidence that there is a fundamental difference in the active site of the two types of MAO in brain. NAP-5-HT may be a useful photoaffinity probe for characterizing the active site of type B MAO.