Proc. Nati. Acad. Sci. USA Vol. 82, pp. 940-944, February 1985 Neurobiology Demonstration and affinity labeling of a stereoselective binding site for a benzomorphan opiate on acetylcholine receptor-rich membranes from Torpedo electroplaque (cholinergic receptor/ion channel/photoaffinity labeling/N-aliyl-N-normetazocine/phencyclidine) ROBERT E. OSWALD, NANCY N. PENNOW, AND JAMES T. MCLAUGHLIN Department of Pharmacology, New York State College of Veterinary Medicine, Cornell University, Ithaca, NY 14853 Communicated by R. H. Wasserman, October 3, 1984 ABSTRACT The interaction of an optically pure benzo- Benzomorphan opiates have been shown to inhibit the morphan opiate, (-)-N-allyl-N-normetazocine [(-)-ANMC], binding of [3H]PCP to central nervous system synaptic mem- with the nicotinic acetylcholine receptor from Torpedo electro- branes (15, 16) and to modify the activity of serum cholines plaque was studied by using radioligand binding and affinity terase (17). Some opiate derivatives, in particular the benzo- labeling. The binding was complex with at least two specific morphans, can inhibit the binding of [3H]perhydrohistrioni- components having equilibrium dissociation constants of 0.3 cotoxin and [3H]PCP to the Torpedo AcChoR (18, 19). In the jiM and 2 jiM. The affinity of the higher affinity component present studies, a radioactive optically pure benzomorphan, was decreased by carbamoylcholine but not by a-bungaro- (-)-[3H]N-allyl-N-normetazocine {(-)-[3H]ANMC}, was toxin. The effect of carbamoylcholine was not blocked by a- used to measure reversible binding to and affinity labeling of bungarotoxin. In comparison, the affinity of [3Hlphencycli- the Torpedo electroplaque AcChoR. Specific binding was dine, a well-characterized ligand for a high-affinity site for shown to be complex with at least one component having noncompetitive blockers on the acetylcholine receptor, is in- lower affinity in the presence rather than the absence of cho- creased by carbamoylcholine and the increase is blocked by a- linergic effectors. Binding to the high-affinity site was inhib- bungarotoxin. The binding of (-)-[3HJANMC was inhibited ited by PCP with a mechanism that is not competitive. a- by a number of other benzomorphans, with (-) isomers being Bungarotoxin (a-BuTx) had no effect on the decrease in af- 4- to 5-fold more potent than (+) isomers. Phencyclidine inhib- finity induced by cholinergic effectors. UV-catalyzed its the binding of (-)-[3H]ANMC to its high-affinity site by a affinity labeling (20) resulted in the labeling of the 8 subunit mechanism that is not competitive. UV-catalyzed affinity la- and, in some cases, the a subunit. these results may indicate beling indicated that the high-affinity-binding site for (-)- a unique binding site for benzomorphans on the AcChoR. [3H]ANMC is at least partially associated with the 8 subunit. Tryptic degradation of the Torpedo marmorata 8 chain sug- MATERIALS AD METHODS gested that (-)-ANMC labeled a 16,000-dalton COOH-termi- AcChoR-rich membranes were prepared from the electro- nal portion of the subunit. In contrast, 5-azido-[3Hltrimethiso- plaque of Torpedo californica, Torpedo marmorata, and quin, a photoaffinity label of the high-affinity site for noncom- Torpedo nobiliana as described (21). The binding of (-)- petitive blockers, labels a 47,000-dalton N1H2-terminal [3H]ANMC was measured by either a centrifugation (10) or a fragment of the 8 subunit. These results suggest that (-)- filtration assay (22, 23) in a total volume of 125 a.l. The buff- [3HJANMC binds to sites completely distinct from the binding er used in all cases was 50 mM 3-(N-morpholino)propanesul- sites for acetylcholine. The high-affinity-binding site for (-)- fonic acid (Mops)/NaOH and } mM EGTA, pH 7.5 (Mops/ ANMC and that for phencyclidine and 5-azidotrimethisoquin EGTA). are allosterically coupled but are regulated differently and are Cross-linking of (-)-[3H]ANMC and 5-azido-[3H]trimethi- probably physically distinct. soquin (5A[3H]T) induced by irradiation with UV light was performed as described (20, 24) except that Mops/ The nicotinic acetylcholine receptor (AcChoR) from Torpe- EGTA was the buffer. Trypsin degradation was performed do electroplaque is a multisubunit protein that mediates ion as described by Wennogle et al. (25). flux across the cell membrane in response to the binding of Benzomorphan opiates (ANMC, metazocine, pentazo- acetylcholine (AcCho). The receptor complex consists of cine, cyclazocine), etorphine, naloxone, and (-)- four different subunits of 50,116 (a), 53,681 (,B), 56,279 ('y), [3H]ANMC (49.5 Ci/mmol; 1 Ci = 37 GBq) were supplied by and 57,565 (8) daltons (measured by sequence analysis; ref. the National Institute on Drug Abuse. Phenazocine, theba- 1), existing in a stoichiometry of 2:1:1:1 (2-4). These sub- ine, and oxymorphone were gifts of A. E. Jacobson (Nation- units migrate in NaDodSO4/polyacrylamide gels with appar- al Institutes of Health). A. Gero (Hanemann University) ent molecular masses of 40,000, 50,000, 60,000, and 66,000 generously supplied metazocine, pentazocine, cyclazocine, daltons, respectively. A high-affinity-binding site for AcCho 5-ethyl-2'-hydroxy-2-methyl-6,7-benzomorphan, and 5-(m- is associated, at least in part, with each of the a subunits hydroxyphenyl)-2-methylmorphan. PCP was obtained from (reviewed in ref. 5). In addition, a series of compounds, U.S. Pharmacopeial Convention (Rockville, MD), and known collectively as noncompetitive blockers, has been [3H]PCP (48 Ci/mmol), [3H]AcCho (90 mCi/mmol), and 125I1 shown to bind to the AcChoR (6-13). These compounds, labeled a-BuTx (1251-a-BuTx) were purchased from New which include phencyclidine (PCP), histrionicotoxin, chlor- England Nuclear (Boston, MA). 5A[3H]T wag a gift of J. P. promazine, and some aminated local anesthetics, bind to one Changeux (Institut Pasteur, Paris). Live T. californica were high-affinity site per AcChoR monomer (reviewed in ref. 14) purchased from Pacific Biomarine (Venice, CA), and frozen as well as to a variable number of low-affinity sites associat- T. nobiliana were purchased from Biofish Associates ed with the membrane lipids (10, 11). Abbreviations: AcCho, acetylcholine; AcChoR, AcCho receptor; The publication costs of this article were defrayed in part by page charge (-)-ANMC, (-)-N-allyl-N-normetazocine; 5A[3H]T, 5-azido- payment. This article must therefore be hereby marked "advertisement" [3H]trimethisoquin; a-BuTx, a-bungarotoxin; MPTA, [4-(N-malei- in accordance with 18 U.S.C. §1734 solely to indicate this fact. mido)phenyl]trimethylammonium; PCP, phencyclidine. 940 Downloaded by guest on September 28, 2021 Neurobiology: Oswald et aL Proc. NatL. Acad. Sci USA 82 (1985) 941 (Georgetown, MA). Ptirified T. marmorata membranes were a gift of B. Holton and A. Ribera (Institut Pasteur). RESULTS Equilibrium Bindihg of (-)-[3H]ANMC to AcChoR-Rich Membrane Fragments. As illustrated in Fig. 1, the specific binding of (-)-[3H]ANMC to AcChoR-rich membrane frag- ments from T. californica is saturable. (Identical results were obtained with T. marmorata and T. nobiliana.) The curve drawn through the experimental points represents a nonlinear least squares fit of the data to a two-site model. 0 The equilibrium dissociation constant (Kd) of the high-affini- L.O ty component is 0.36 ± 0.07 MM and that of the low-affinity ~0 component is 2.2 ± 0.4 AM. The high-affinity component o 1.0 comprises between 40% and 50% of the specific binding. The mo total concentration of sites in the preparation shown in Fig. 1 was 0.1 MM, which is within 10% of the number of AcCho binding sites measured by [3H]AcCho binding and a-BuTx binding. This suggests that two binding sites for (-)-ANMC may exist on each 250,000-dalton AcChoR monomer. Both components were lost following heating to 100'C for 3 min. Fig. 2 shows a comparison of the binding of (-)- [3H]ANMC and of [3H]PCP in the presence and absence of carbamoylcholine and a-BuTx. The affinity and number of binding sites for (-)-[3H]ANMC are unaffected by the pres- ence of a-BuTx; however, carbamoylcholine decreases the affinity of the high-affinity component without changing the number of binding sites (Fig. 2A), with an IC50 of 40 ± 8 AM. This effect is not blocked by a-BuTx. Under identical condi- Bound, cpm x 10O3 tions, a-BuTx blocks completely the agonist-induced in- FIG. 2. Binding of (-)-[3H]ANMC (A) and [3H]PCP (B) to Ac- crease in affinity for [3H]PCP and inhibits completely the ChoR-rich membrane fragments (200 nM in I251-a-BuTx sites, 0.31 binding of 125I-a-BuTx. d-Tubocurarine and decamethonium nmol of 125I-a-BuTx sites per mg of prdtein) in the presence or ab- also decrease the affinity of (-)-[3H]ANMC (IC50 values of sence of 0.2 mM carbamoylcholine (Carb) and/or 2 AM a-BuTx us- 47 ± 10 AM and 7.6 ± 1.1 MM, respectively). The effects of ing the centrifugation assay. Assays were performed following a 45- d-tubocurarine and decamethonium are not inhibited by a- min incubation at 25°C. The data are plotted a§ a Scatchard transfor- BuTx. This is in contrast to the binding of [3H]PCP, which mation of the specific binding determined following subtraction of an increased in the of nonspecific binding measured using either 0.2 mM nonradioactive shows affinity presence carbamoyl- (-)-ANMC (A) or 0.2 mM nonradioactive PCP (B). (A) The Kd for choline (EC50 = 0.15 ,uM), d-tubocurarine (EC50 = 0.08 (-)-[3H]ANMC in the presence of carbamoylcholine was 1.9 ,M, and, in the absence of carbamoylcholine, the Kd values were 0.3 AM and 2 ,M. a-BuTx had no effect on binding either in the presence or 3 A absence of carbamoylcholine. (B) The Kd for [3H]PCP was 0.11 ,M 45 S in the presence of carbamoylcholine, 1.0 AM in the absence of car- bamoylcholine, 1.6 AM in the presence of a-BuTx, and 1.1 MM in 2 0 the presence of both carbamoylcholine and a-BuTx.