Improvement of 5-HT3 Receptor Binding Assay: Enhancement of Specific [3HlQuipazine Binding with Triton X-100-Treated Membranes from Rat Cortex

Teruaki Une, Kiyoshi Furukawa and Masanobu Komiya

Department of Pharmacology, Research Laboratories, Dainippon Pharmaceutical Co., Ltd., 33-94, Enokicho, Suita, Osaka 564, Japan

Received April 9, 1991 Accepted July 1, 1991

ABSTRACT-The 5-hydroxytryptamine (5-HT)3 receptor binding assay using [3H]quipazine was examined. It was impossible to obtain specific [3H]quipazine bind ing with the membrane fractions from rat cortex prepared by the usual procedure. When the membranes were pretreated with detergent Triton X-100, the ratio of spe cific [3H]quipazine binding markedly increased, depending upon the concentration of Triton X-100 in the range of 0.01-0.1% (w/v). At a concentration of more than 0.05%, the specific binding reached a maximum of 55 to 60% of the total binding. The specific [3H]quipazine binding to the Triton X-100-treated membranes was re versible and was potently inhibited by several 5-HT3 antagonists, while 5-HT], 5-HT2 receptor antagonists and other receptor-specific ligands had no effect on the binding. Scatchard analysis indicated a single class of binding sites with a Kd of 0.62 nM and Bmax of 97 fmol/mg protein. Thus, the Triton X-100-treated membranes retained the characteristics of 5-HT3 binding sites, making it possible to use [3H]quipazine for a 5 HT3 receptor binding assay with a high ratio of specific binding.

Serotonin (5-HT) receptors have been di their method, because of high percentages of vided into three different subtypes: 5-HT1, 5 non-specific binding. To obtain 5-HT3 recep HT2 and 5-HT3 (1). Recently, the binding tor preparations with a high affinity for sites of 5-HT3 in the central nervous system [3H]quipazine, we performed various addition have been identified by radioligand binding al treatments on the membranes; and inter techniques using tritium-labeled 5-HT3 antago estingly, we found that the specific binding of nists such as [3H]GR65630 (2), [3H] [3H]quipazine increased markedly when the (3), [3H]BRL 43694 (4), [3H]LY278584 (5) membranes were treated with Triton X-100 and [3H]GR67330 (6). (polyethylene glycol p-isooctylphenyl ether). Peroutka and Hamik (7) have reported that In the present paper, we describe a reliable [3H]quipazine, a commercially available radio method for 5-HT3 receptor binding assay with ligand, has a high affinity for 5-HT3 binding [3H]quipazine using Triton X-100-treated mem sites in rat cortical membranes. However, we branes of the rat cortex. have been unsuccessful in attempts to obtain A preliminary account of these findings has highly specific [3H]quipazine binding using been presented at the 63rd Annual Meeting of the Japanese Pharmacological Society (8). scintillation cocktail (Amersham). Specific binding was defined as the excess over blanks MATERIALS AND METHODS taken in the presence of 1 uM BRL 43694. Protein concentration was determined by the Preparation of membranes method of Lowry et al. (9) using bovine serum Male Std-Wistar rats (200-250 g) were de albumin as a standard. capitated. The frontal cortex was dissected and homogenized in 10 volumes of ice-cold 0.32 M sucrose in a Potter-Elvehjem glass The following compounds were used: homogenizer fitted with a Teflon pestle. The [3H]quipazine (Dupont New England Nuclear, homogenate was centrifuged at 1000 X g for Specific Activity = 68.1 Ci/mmol); 10 min and the pellet was discarded. The su creatinine sulfate (Merck); pernatant was centrifuged at 17,200 X g for 10 hydrochloride, 1- (Sigma); clonidine min. The crude mitochondrial pellet was re hydrochloride, , SCH23390, amitrip suspended in 20 volumes of Krebs-HEPES tyline hydrochloride (RBI), buffer and centrifuged at 17,200 X g for 10 maleate (extracted from Sansert, Sandoz); min. The pellet (fraction P2) was resuspended (extracted from Acenaline, Janssen in 20 volumes of Krebs-HEPES buffer con Kyowa); and ICS 205-930, BRL 43694, BRL taining 0.01-0.1% Triton X-100. After a 30 24924, GR38032F, MDL 72222, 2-methyl-5 min incubation at 37°C, the membranes were hydroxytryptamine maleate (synthesized in our washed twice by recentrifuging (50,000 X g, laboratories). 15 min) and resuspending in Triton X-100-free buffer (Krebs-HEPES buffer). The pellet was RESULTS finally suspended in 40 volumes of the same buffer. A Krebs-HEPES buffer consisting of Effect of treatment of rat cortical membranes 25 mM HEPES, 180 mM NaCl, 5 mM KCI, with Triton X-100 on [3H]quipazine binding 2.5 mM CaCl2 and 1.2 mM MgCl2 (pH ad The fraction P2 prepared from rat cortex justed to 7.4) was used. was incubated with 1.0 nM [3H]quipazine. No significant amount of specific [3H]quipazine Binding assay binding could be detected, because the non Krebs-HEPES buffer (500,ul) with or with specific binding of [3H]quipazine was very out the , was added to assay tubes and high. Although the amount and proportion of 100 ul of [3H]quipazine was added at final the specific binding varied in different batchs concentrations of 1.0 nM and 0.3 4.0 nM for of [3H]quipazine, the specific binding was less drug competition and saturation studies, re than 30% of the total binding. To remove spectively. Subsequently, the membrane sus possible endogenous serotonin from the mem pension (0.2 mg protein/400 ul) was added to branes, the membrane preparation (fraction initiate binding. The assay tubes were incu P2) was incubated for 30 min at 37°C in bated for 30 min at 25°C. The incubation was Krebs-HEPES buffer, pH 7.4. The specific terminated by rapid filtration under reduced [3H]quipazine binding increased slightly with pressure through Whatman GF/B filters pre increasing total [3H]quipazine binding. soaked in 0.03% polyethyleneimine. The fil When the membrane preparation was incu ters were immediately washed three times with bated for 30 min at 37°C in Triton X-100 solu 4 ml of ice-cold 50 mM Tris-HC1 buffer (pH tion and washed with Triton X-100-free buffer 7.7). by centrifugation twice, the ratio of specific All experiments were performed in dupli [3H]quipazine binding to the Triton X-100 cate or triplicate. Radioactivity was measured treated membranes markedly increased. As by liquid scintillation counting in 10 ml ACS II shown in Table 1, the increases in the ratio of Table 1. Effect of Triton X-100 treatment on specific [3H]quipazine binding

specific binding were dependent on the con centration of Triton X-100 in the range of 0.01 -0 .1%. The treatment with more than 0.05% Triton X-100 gave maximal specific binding that was 55 60% of the total binding at 1.0 nM [3H]quipazine. In preliminary experiments a remarkable decrease of specific binding was observed when the membranes were treated with more than 0.1% Triton X-100, such as 0.3% (data not shown). These effects of Tri ton X-100 treatment were observed consistent ly with any batch of [3H]quipazine.

Characterization of Triton X-100-treated mem branes The binding of [3H]quipazine to 5-HT3 bind ing sites in Triton X-100-treated membranes Fig. 1. The linearity of the specific binding of 1.0 nM was assessed. [3H]quipazine to the Triton X-100-treated membranes. The amount of the specific [3H]quipazine binding was linear over a range of membrane 0.99. It revealed a homogeneous population of protein (25 500 ,ug protein per tube, Fig. 1). [3H]quipazine binding sites having a Kd value In saturation analysis of [3H]quipazine bind of 0.62 ± 0.01 nM and Bmax value of 97 ± 6.9 ing, the specific binding increased with concen fmol/mg protein. trations ranging from 0.3 to 2 nM [3H]quipazine Kinetic analysis of [3H]quipazine binding and reached a saturation at higher concentra demonstrated that [3H]quipazine associated tions (Fig. 2). Scatchard transformation of rapidly with the specific binding sites (K+t = these data was linear with a coefficient of 7.64 X 106/M/s); and once equilibrium had been reached, the binding was shown to be In drug competition studies, selective 5-HT3 completely reversible by the addition of 1 u M antagonists, zacopride, ICS 205-930, BRL BRL 43694 to initiate dissociation (K_1 = 43694, GR38032F, BRL 24924 and MDL 72222, 6.36 X 10-3/s, Fig. 3). The Kd (= K+1/K1) potently inhibited the specific [3H]quipazine calculated from these rate constants gave a binding at subnanomolar to nanomolar con value of 0.83 nM. This Kd value was in close centrations (Table 2). All competition curves agreement with that obtained from Scatchard were steep and monophasic. Nonselective 5 analysis. HT3 antagonists, metoclopramide and cisa

Fig. 2. The saturation analysis of binding of [3H]quipazine (0.3-4.0 nM) to the Triton X-100 (0.05%) treated membranes. (A) shows the saturation curve for the binding of [3H]quipazine. Symbols represent (0) total, (A) specific and (0) non-specific binding. Non-specific binding was defined with 1 uM BRL 43694. (B) shows the Scatchard transformation of these data. The correlation coefficient was 0.99.

Fig. 3. The association (40) and dissociation (0) induced by BRL 43694 of [3H]quipazine binding to the Triton X-100 (0.05%)-treated membranes. The membranes were incubated with 1.0 nM [3H]quipazine for the indicated times at 25°C. After a 30-min incubation, the dissociation was induced by an addition (arrow) of 1 u M BRL 43694. Table 2. The affinities of various compounds Under such a condition, it was difficult to esti that compete for [3H]quipazine binding to Triton mate the data of 5-HT3 receptor binding X-100-treated membranes accurately. However, in this study, the ratio of specific binding was increased by treatment of the membrane with Triton X-100 in a concen tration dependent manner. Treatment of the membrane with Triton X-100 (0.5-1.0%) in creased the ratio of specific binding up to 55% of the total binding. Recently, McKernan et al. (10) solubilized 5-HT3 receptor sites from membranes prepared from pooled rat cerebral cortex and hippocampus using various deter gents, and they found that the non-specific binding of [3H]quaternized ICS 205-930 to solubilized 5-HT3 receptor sites to 20% of to tal binding. Therefore, it is supposed that de tergents decrease the non-specific binding in the 5-HT3 receptor binding assay. To determine if Triton X-100 treated mem branes retain the characteristics of the 5-HT3 receptor or not, we characterized [3H]quipazine binding to Triton X-100 treated membranes. Ki netic analysis showed that specific [3H]quipazine binding is rapid and reversible, and saturation analysis revealed a single class of binding sites with a Kd of 0.62 ± 0.01 nM and a Bmax of 97.3 ± 6.9 fmol/mg protein. The Bmax value was similar to those reported for pride, moderately inhibited the [3H]quipazine [3H]quaternized ICS 205-930 (11) and [3H]ICS binding with a Ki of 320 nM and 250 nM, re 205-930 (12), but higher than that reported for spectively. The , 2-methyl-5-HT (K; _ [3H]LY278584 (5) in rat cortical membranes. 98 nM) and 5-HT (Ki = 31 nM) also inhibited The Kd value was in agreement with the value the specific binding. reported for [3H]LY278584 (5). Furthermore, In contrast, drugs acting at 5-HT1, 5-HT2 drug competition studies demonstrated that and other transmitter receptors were inactive the specific [3H]quipazine binding is potently on the [3H]quipazine binding. , a inhibited by 5-HT3 receptor agonists and an 5-HT-uptake inhibitor, also failed to inhibit tagonists. The range order of their potencies the binding at concentrations up to 1.0,uM. showed good agreement with those obtained by other 5-HT3 binding assays in cortical DISCUSSION membranes (2, 3, 5, 13). Agents that acted on other neurotransmitter receptors were without The major finding of the present study is appreciable affinity for the [3H]quipazine that treatment of rat cortical membranes with binding sites. These data indicate that Triton Triton X-100 markedly increased the ratio of X-100-treated membranes retain the character specific [3H]quipazine binding to 5-HT3 recep istics of the 5-HT3 binding sites found in the tor sites. The specific binding was only about untreated membranes. 30% of the total binding using rat cortical We observed the increase in the ratio of membranes without Triton X-100 treatment. specific [3H]quipazine binding in the cortical membranes treated with Triton X-100 in all of tion untreated with Triton X-100. [3H]quipazine batches tested. Schmidt et al. In conclusion, the present study indicates (14) reported that a 5-HT uptake inhibitor that Triton X-100 treatment permits highly such as chlorimipramine and amitriptyline inhib specific [3H]quipazine binding (55 60% of the ited [3H]quipazine binding at nanomolar to total binding). The Triton X-100 treatment subnanomolar concentrations, and the degree should be used for the receptor binding assay of these inhibitions differed between batches using [3H]quipazine, to obtain precise, repro of [3H]quipazine. They concluded that the dif ducable parameters in the 5-HT3 receptor ferences in the potencies of 5-HT uptake inhib binding study. itors in different batches were due to the dif ferences in the amount of [3H]quipazine de REFERENCES gradation products remaining in the batches which could label the 5-HT uptake site. In the 1 Bradley, P.B., Engel, G., Feniuk, W., Fozard, J.R., Humphrey, P.P.A., Middlemiss, D.N., present study, amitriptyline even at 1.0 ,uM Mylecharane, E.J., Richardson, B.P. and Saxena, concentration did not inhibit the [3H]quipazine P.R.: Proposals for the classification and nomen binding in Triton X-100-treated membranes clature of functional receptors for 5-hydroxytrypta when tested in two different [3H]quipazine mine. Neuropharmacology 25, 563-576 (1986) batches. Thus it is likely that the treatment of 2 Kilpatrick, G.J., Jones, B.J. and Tyers, M.B.: Triton X-100 reduced the binding of Identification and distribution of 5-HT3 receptors in rat brain using radioligand binding. Nature 330, [3H]quipazine degradation products to the 5 746 748 (1987) HT uptake site. 3 Barnes, N.M., Costall, B. and Naylor, R.J.: It is considered that there are mainly two [3H]Zacopride: ligand for the identification of 5 important factors responsible for the increase HT3 recognition sites. J. Pharm. Pharmacol. 40, in the ratio of specific [3H]quipazine binding. 548 551 (1988) One is the decrease of non-specific binding, 4 Nelson, D.R. and Thomas, D.R.: [3H]-BRL43694 and the other is the removal of endogenous (), a specific ligand for 5-HT3 binding inhibitors (e.g., 5-HT in this case) from the sites in rat brain cortical membranes. Biochem. membrane preparation. In the present study, Pharmacol. 38, 1693-1695 (1989) 5 Wong, D.T., Robertson, D.W. and Reid, L.R.: Triton X-100 treatment decreased non-specific Specific [3H]LY278584binding to 5-HT3 recogni binding of [3H]quipazine in a concentration re tion sites in rat cerebral cortex. Eur. J. Pharmacol. lated manner. This indicates that Triton X-100 166, 107-110 (1989) plays a crucial role in decreasing non-specific 6 Kilpatrick, G.J., Butler, A., Hagan, R.M., Jones, binding, probably by preventing non-specific B.J. and Tyers, M.B.: [3H]GR67330,a very high interactions of [3H]quipazine with lipid mole affinity ligand for 5-HT3 receptors. Naunyn cules in the cortical membranes. Triton X-100 Schmiedebergs Arch. Pharmacol. 342, 22 30 (1990) has been frequently used to remove endoge 7 Peroutka, S.J. and Hamik, A.: [3H]Quipazine nous inhibitors from the membranes employed labeler 5-HT3 recognition sites in rat cortical mem in GABA receptor binding studies (15). The branes. Eur. J. Pharmacol. 148, 297-299 (1988) specific [3H]GABA binding to GABA recep 8 Une, T., Furukawa, K. and Komiya, M.: Im tors is enhanced when endogenous GABA provement of 5-HT3 receptor binding assay: En was removed from membranes by the treat hancement of specific [3H]quipazine binding with Triton X-100-treated membranes in rat cortex. ment with Triton X-100 (16). So it is possible Japan. J. Pharmacol. 52, Supp. I, 230P (1990) that in the present study, Triton X-100 re 9 Lowry, O.H., Rosebrough, N.J., Farr, A.L. and moved endogenous 5-HT from the membrane Randall, R.J.: Protein measurement with the Folin preparation. However, the effect of incubation phenol reagent. J. Biol. Chem. 193, 265-275 can not be excluded, because the specific (1951) binding of [3H]quipazine was increased after 10 McKernan, R.M., Quirk, K., Jackson, R.G. and incubation at 37°C for 30 min in the prepara Ragan, C.I.: Solubilisation of the 5-hydroxy tryptamine3 receptor from pooled rat cortical Pharmacol. 338, 472-475 (1988) and hippocampal membranes. J. Neurochem. 54, 14 Schmidt, A.W., Hurt, S.D. and Peroutka, S.J.: 924 930 (1990) '[3H]Quipazine' degradation products label 5-HT 11 Watling, K.J., Aspley, S., Swain, C.J. and Saun uptake sites. Eur. J. Pharmacol. 171, 141 143 ders, J.: [3H]QuaternisedICS 205-930labels 5-HT3 (1989) receptor binding sites in rat brain. Eur. J. Pharma 15 Napias, C., Bergman, M.O., Van Ness, P.C., col. 149, 397-398 (1988) Greenlee, D.V. and Olsen, R.W.: GABA binding 12 Hoyer, D. and Neijt, H.C.: Identification of sero in mammalian brain: Inhibition by endogenous tonin 5-HT3 recognition sites by radioligand bind GABA. Life Sci. 27, 1001-1011 (1980) ing in NG108-15 neuroblastoma-glioma cells. Eur. 16 Horng, J.S. and Wong, D.T.: y-Aminobutyric J. Pharmacol. 143, 291-292 (1987) acid receptors in cerebellar membranes of rat 13 Peroutka, S.J.: Species variations in 5-HT3 recog brain after a treatment with Triton X-100. J. nition sites labeled by 3H-quipazine in the central Neurochem. 32, 1379-1386 (1978) nervous system. Naunyn Schmiedebergs Arch.