Molecular dissection of cholinergic binding sites: how do snakes escape the effect of their own toxins?

Snakes have evolved a novel binding site demonstrating selective biorecognition. The snake nicotinic acetylcholine receptor is sensitive to acetylcholine while resistant to the effect of the lethal neurotoxins secreted in their own venom. By subjecting recombinant binding sites to point mutagenesis, biochemical analyses and NMR spectroscopy the binding characteristics of three cholinergic ligands have been measured. The amino acid residue at position 189 has been found to be of particular importance to toxin binding.     

Antibodies against preselected peptides to map functional sites on the acetylcholine receptor

Rabbit immune sera and mouse monoclonal antibodies were raised against the synthetic peptide Tyr-Cys-Glu-Ile-Ile-Val matching in sequence residues 127-132 of the alpha-subunit of all nicotinic acetylcholine receptors sequenced so far. Representative cholinergic ligands did not interfere with the binding of these antibodies to the receptor from Torpedo marmorata, indicating that this sequence is not part of the binding sites for cholinergic ligands. The applicability of antigenic sites analysis to the mapping of functional sites on receptor proteins is discussed.     

A monoclonal antibody interfering with binding and response of the acetylcholine receptor

Employing a monoclonal antibody raised against the receptor protein, we have probed the mechanism of ligand interaction of the nicotinic acetylcholine receptor from Torpedo marmorata. Antibody WF6 specifically binds to alpha-subunits of the receptor with a stoichiometry of one molecule per receptor monomer. At saturating concentrations, WF6 blocks half of the binding sites for acetylcholine, all of the binding sites for alpha-neurotoxins, and none of the binding sites for representative cholinergic antagonists (with the exception of alpha-toxins) at the receptor. In the presence of saturating concentrations of antibody WF6, acetylcholine (or its agonists) cannot induce T1+ influx into Torpedo membrane vesicles. Rapid oversaturation of the receptor by agonist also cannot overcome this blockade of channel gating. The observed competition patterns of WF6 and representative cholinergic ligands with the receptor are evidence for separate binding sites for groups of ligands and for a network of allosterically linked effector regions at the receptor. The blockade by saturating concentrations of WF6 of the agonist-induced channel gating supports the conclusion that two molecules of agonist are required to activate the receptor-integral ion channel.     

Studies on nicotinic acetylcholine receptors in mammalian brain. Interaction of solubilized protein with cholinergic ligands

Binding of alpha-bungarotoxin, labeled with 125I, has been studied in detergent extracts and affinity purified acetylcholine receptor from rat cerebral cortex. Binding to detergent extracts is saturable and appears to be due to one class of binding sites present at a level of 0.27 pmol/mg of protein. The association constant is 2 X 10(7) liters mol-1 . Competition with cholinergic ligands indicates that toxin binding to both detergent solubilized and affinity purified receptor retains its nicotinic nature. Values for the ligand concentrations required to produce 50% inhibition of extent and rate of toxin binding are presented.     

Organization of ligand binding sites at the acetylcholine receptor: a study with monoclonal antibodies

We have studied 20 monoclonal antibodies directed against both the solubilized and the membrane-bound receptor from Torpedo marmorata. We find the following: (i) Six of the antibodies compete with cholinergic ligands for receptor binding and, hence, are directed against the ligand binding regions. (ii) Of these six antibodies, two cross-react with receptor from Electrophorus electricus, rat myotubes, and chicken sympathetic ganglia. These two antibodies therefore define a preserved structure within the ligand binding regions. The other four antibodies bind to structures not common between the receptor preparations tested. (iii) From competition binding studies using internally 3H-labeled antibodies, nine nonoverlapping antigenic regions were defined at the surface of the receptor. Three of these regions overlap with the ligand binding regions. Since two of these three regions do not overlap with each other, two structurally distinct ligand binding regions must exist at the receptor. (iv) From competition binding studies with representative cholinergic ligands, the antibodies directed against the ligand binding regions can be subdivided into three groups: one group competes with all ligands tested; the second group competes with all ligands except the bismethonium compounds; the third group competes with all ligands except the bismethonium compounds and tubocurarine. The results are summarized in a model of the organization of ligand binding sites at the receptor: There are two ligand binding regions differing in their antigenic properties. Furthermore, either there exists separate sites for distinct groups of ligands within each of these binding regions or some ligands produce conformational changes of the receptor that reversibly abolish some antigenic sites. In any case, the cholinergic ligands must interact with the receptor by more and/or other structural determinants than are provided by the structure of acetylcholine.     

Muscarinic and Nicotinic Cholinergic Binding Sites in the Terminal Abdominal Ganglion of the Cricket (Acheta domesticus)

Cricket (Acheta domesticus) terminal abdominal ganglia (TG) contain high concentrations (approximately 2 pmol/mg protein) of muscarinic and nicotinic cholinergic binding sites, based on the capacity of TG to bind specifically the labelled ligands L-[3H]quinuclidinyl benzilate ([3H]QNB) and [125I]alpha-bungarotoxin ([125I]alpha-BGT) with high affinity. For both ligands, binding is saturable and reversible. Competitive displacement experiments indicate that the [3H]QNB and [125I]alpha-BGT binding sites probably represent pharmacologically distinct classes of putative TG acetylcholine receptors (AChRs). Results from physiological recording and autoradiographic localization experiments demonstrate that a portion of the putative nicotinic AChRs is localized in synaptic regions of the well-characterized cercal sensory-giant interneuron pathway in the TG, where they are likely to serve as functional synaptic AChRs. Unlike nicotinic ligands, muscarinic agents do not appear to be pharmacologically active in this pathway. Therefore, in the insect CNS, putative muscarinic and nicotinic AChRs coexist at high density, but can be pharmacologically distinguished from one another on the basis of criteria derived from both ligand binding and physiological methods.     

A monoclonal antibody to a synthetic fragment of rabies virus glycoprotein binds ligands of the nicotinic cholinergic receptor

Rabies virus glycoprotein and snake venom curaremimetic neurotoxins share a region of high homology (30-45 for neurotoxins and 190-203 for the glycoprotein) in the regions that are believed to be responsible for binding the nicotinic acetylcholine receptor. Monoclonal antibodies raised to the 190-203 synthetic fragment of rabies virus glycoprotein were immobilized on a high performance affinity chromatography column and were able to bind neurotoxins. Toxins were displaced from the affinity column by elution at acidic pH and by affinity competition with acetylcholine at neutral pH. Furthermore, the affinity column proved to be useful for the purification of cholinergic ligands. Overall, these results indicate that the paratope of our monoclonal antibodies could behave as an 'internal image' of the nicotinic cholinergic receptor acetylcholine binding site.     

Characterization of Curaremimetic Neurotoxin Binding Sites on Cellular Membrane Fragments Derived from the Rat Pheochromocytoma PC 12

Studies were conducted on the properties of 125I-labeled alpha-bungarotoxin binding sites on cellular membrane fragments derived from the PC12 rat pheochromocytoma. Two classes of specific toxin binding sites are present at approximately equal densities (50 fmol/mg of membrane protein) and are characterized by apparent dissociation constants of 3 and 60 nM. Nicotine and d-tubocurarine are among the most potent inhibitors of high-affinity toxin binding. The affinity of high-affinity toxin binding sites for nicotinic cholinergic agonists is reversibly or irreversibly decreased, respectively, on treatment with dithiothreitol or dithiothreitol and N-ethylmaleimide. The nicotinic receptor affinity reagent bromoacetylcholine irreversibly blocks high-affinity toxin binding to PC12 cell membranes that have been treated with dithiothreitol. Two polyclonal antisera raised against the nicotinic acetylcholine receptor from Electrophorus electricus inhibit high-affinity toxin binding. These detailed studies confirm that curaremimetic neurotoxin binding sites on the PC12 cell line are comparable to toxin binding sites from neural tissues and to nicotinic acetylcholine receptors from the periphery. Because toxin binding sites are recognized by anti-nicotinic receptor antibodies, the possibility remains that they are functionally analogous to nicotinic receptors.     

Presynaptic Nicotinic Cholinergic Receptors Labeled by [3H]Acetylcholine on Catecholamine and Serotonin Axons in Brain

Nicotinic cholinergic receptor binding sites labeled by [3H]acetylcholine were measured in the cerebral cortices, thalami, striata, and hypothalami of rats lesioned by intraventricular injection of either 6-hydroxydopamine or 5, 7-dihydroxytryptamine. In addition, [3H]acetylcholine binding sites were measured in the cerebral cortices of rats lesioned by injection of ibotenic acid into the nucleus basalis magnocellularis. [3H]Acetylcholine binding was significantly decreased in the striata and hypothalami of both 6-hydroxydopamine- and 5,7-dihydroxytryptamine-lesioned rats. There was no change in binding in the cortex or thalamus by either lesion. Ibotenic acid lesions of the nucleus basalis magnocellularis, which projects cholinergic axons to the cortex, did not alter [3H]acetylcholine binding. These results provide evidence for a presynaptic location of nicotinic cholinergic binding sites on catecholamine and serotonin axons in the striatum and hypothalamus.     

Functional significance of aromatic amino acids from three peptide loops of the alpha 7 neuronal nicotinic receptor site investigated by site-directed mutagenesis.

Three aromatic amino acids, Tyr92, Trp148 and Tyr187 belonging to three separate domains of the alpha 7-subunit of neuronal nicotinic acetylcholine receptor were mutated to phenylalanine, and the electrophysiological response of the resulting mutant receptors analyzed in the Xenopus oocyte expression system. All mutations significantly decreased the apparent affinities for acetylcholine and nicotine, and to a lesser extent, those for the competitive antagonists dihydro-beta-erythroidine and alpha-bungarotoxin. Other properties investigated, such as the voltage dependency of the ion response as well as its sensitivity to the open channel blocker QX222, were not significantly changed, indicating that the mutations affected selectively the recognition of cholinergic ligands by the receptor protein. The maximal rates for the rapid desensitization process were slightly modified, suggesting that the contribution of Tyr92, Trp148 and Tyr187 to the binding area might differ in the various conformations of the nicotinic receptor. Other mutations at nearby positions (S94N, W153F, G151D and G82E) did not affect the properties of the electrophysiological response. These data point to the functional significance of Tyr92, Trp148 and Tyr187 in the binding of cholinergic ligands and ion channel activation of the nicotinic receptor, thus supporting a multiple loop model [(1990) J. Biol. Chem. 265, 10430-10437] for the ligand binding area.     

Expression of soluble ligand- and antibody-binding extracellular domain of human muscle acetylcholine receptor alpha subunit in yeast Pichia pastoris. Role of glycosylation in alpha-bungarotoxin binding

The N-terminal extracellular domain (amino acids 1-210; halpha-(1-210)) of the alpha subunit of the human muscle nicotinic acetylcholine receptor (AChR), bearing the binding sites for cholinergic ligands and the main immunogenic region, the major target for anti-AChR antibodies in patients with myasthenia gravis, was expressed in the yeast, Pichia pastoris. The recombinant protein was water-soluble and glycosylated, and fast protein liquid chromatography analysis showed it to be a monomer. halpha-(1-210) bound (125)I-alpha-bungarotoxin with a high affinity (K(d) = 5.1 +/- 2.4 nm), and this binding was blocked by unlabeled d-tubocurarine and gallamine (K(i) approximately 7.5 mm). Interestingly, (125)I-alpha-bungarotoxin binding was markedly impaired by in vitro deglycosylation of halpha-(1-210). Several monoclonal antibodies that show partial or strict conformation-dependent binding to the AChR were able to bind to halpha-(1-210), as did antibodies from a large proportion of myasthenic patients. These results suggest that the extracellular domain of the human AChR alpha subunit expressed in P. pastoris has an apparently near native conformation. The correct folding of the recombinant protein, together with its relatively high expression yield, makes it suitable for structural studies on the nicotinic acetylcholine receptor and for use as an autoantigen in myasthenia gravis studies.     

Phosphatidylcholine Biosynthesis in the Neuroblastoma-Glioma Hybrid Cell Line NG 108-15: Stimulation by Phorbol Esters

Studies were conducted on curaremimetic neurotoxin binding to the nicotinic acetylcholine receptor present on membrane fractions derived from the human medulloblastoma clonal line, TE671. High-affinity binding sites (KD = 2 nM for 1-h incubation at 20 degrees C) and low-affinity binding sites (KD = 40 nM) for 125I-labeled alpha-bungarotoxin are present in equal quantities (60 fmol/mg membrane protein). The kinetically determined dissociation constant for high-affinity binding of toxin is 0.56 nM (k1 = 6.3 X 10(-3) min-1 nM-1; k-1 = 3.5 X 10(-3) min-1) at 20 degrees C. Nicotine, d-tubocurarine, and acetylcholine are among the most effective inhibitors of high-affinity toxin binding. The quantity of toxin binding sites and their affinity for cholinergic agonists is sensitive to reduction, alkylation, and/or oxidation of membrane sulfhydryl residues. High-affinity toxin binding sites that have been subjected to reaction with the sulfhydryl reagent dithiothreitol are irreversibly blocked by the nicotinic receptor affinity reagent bromoacetylcholine. High-affinity toxin binding is inhibited in the presence of either of two polyclonal antisera or a monoclonal antibody raised against nicotinic acetylcholine receptors from fish electric tissue. Taken together, these results indicate that curaremimetic neurotoxin binding sites on membrane fractions of the TE671 cell line share some properties with nicotinic acetylcholine receptors of peripheral origin and with toxin binding sites on other neuronal tissues.     

Separate sites of low and high affinity for agonists on Torpedo californica acetylcholine receptor

We have studied alkylation of the membrane-bound acetylcholine receptor (AcChR) from Torpedo californica electric organ by the cholinergic agonist bromo-acetylcholine (BrAcCh). Following reduction of the AcChR with dithiothreitol (DTT) under strictly controlled conditions, a single class of binding sites was covalently labeled by BrAcCh. The extent of alkylation was dependent on the concentration of both DTT and BrAcCh and reached a maximum when a number of sites equivalent to the number of alpha-bungarotoxin (alpha-BTx) binding sites were labeled. The reaction with BrAcCh was completely inhibited by saturating concentrations of alpha-BTx. On the contrary, complete alkylation of the AcChR with [3H]BrAcCh consistently inhibited only approximately 50% of alpha-BTx binding. The effects of DTT reduction and subsequent BrAcCh alkylation on the cation-gating properties of the AcChR were investigated in rapid kinetic experiments. DTT reduction resulted in a slight decrease in the maximum cation flux and a small shift in the effective dissociation constant to higher acetylcholine (AcCh) concentration. The flux response was completely inhibited by maximal alkylation of the membrane vesicles by BrAcCh. A low-affinity binding site for AcCh, which is likely to be important in AcChR activation, has been revealed for T. californica AcChR by studying the effects of cholinergic ligands on the fluorescence of a probe, 4-[(iodoacetoxy)ethylmethylamino]-7-nitro-2,1,3-benzoxadiazole (IANBD), covalently bound to the AcChR protein. Maximal labeling by BrAcCh did not affect the binding of AcCh to the low-affinity binding site, as monitored by changes in the fluorescence of this probe. This low-affinity binding site must therefore be distinct from the site labeled by BrAcCh. The results strongly support the notion that the nicotinic AcChR contains multiple binding sites for cholinergic ligands.     

Analysis of α-Bungarotoxin Binding in the Goldfish Central Nervous System

Abstract: We report here the equilibrium, kinetic, and pharmacological analysis of α-¹²⁵I-bungarotoxin (α-¹²⁵I-Bgt) binding to a Triton x-100-solubilized goldfish brain synaptosomal fraction. In addition, a refined analysis of equilibrium binding to a particulate synaptosomal fraction is presented. Equilibrium binding from both particulate and soluble fractions revealed an apparent heterogeneity of binding sites. Kinetic analysis of the soluble receptor revealed linear association kinetics and nonlinear dissociation kinetics. The dissociation curve suggested the presence of at least two rate constants. Potential sources of the binding heterogeneity found in both the equilibrium binding and dissociation kinetics experiments are (1) multiple receptor species, (2) multiple ligand species, and (3) different, possibly interconvertible, states of a single receptor type. No evidence for the first two alternatives was found. Support for the third alternative was obtained by observing the effect of cholinergic ligands on α-¹²⁵I-Bgt dissociation. Carbamylcholine and d -tubocurarine increased the apparent proportion of rapidly dissociating sites, suggesting that the two binding affinities can be interconverted and may arise from a single receptor type. Evidence concerning the identity of the α -Bgt binding protein as a nicotinic acetylcholine receptor is discussed.     

Novel Photoactivatable Agonist of the Nicotinic Acetylcholine Receptor of Potential Use for Exploring the Functional Activated State

Abstract: The nicotinic acetylcholine receptor (AChR) exhibits at least four different conformational states varying in affinity for agonists such as acetylcholine (ACh). Photoaffinity labeling has been previously used to elucidate the topography of the AChR. However, to date, the photosensitive probes used to explore the cholinergic binding site photolabeled only closed or desensitized states of the receptor. To identify the structural modifications occurring at the ACh binding site on allosteric transition associated with receptor activation, we have investigated novel photoactivatable 4-diazocyclohexa-2,5-dienone derivatives as putative cholinergic agonists. Such compounds are fairly stable in the dark and generate highly reactive carbenic species on irradiation. In binding experiments using AChRs from Torpedo marmorata, these ligands had affinities for the ACh binding site in the micromolar range and did not interact with the noncompetitive blocker site (greater than millimolar affinity). Irreversible photoinactivation of ACh binding sites was obtained with the ligand 1b (up to 42% at 500 µM) in a protectable manner. In patch-clamp studies, 1b was shown to be a functional agonist of peripheral AChR in TE 671 cells, with the interesting property of exhibiting no or very little desensitization even at high concentrations.     

Muscarinic Receptor Subclasses in the Chick Embryo Retina: Influence of Corticosterone Treatment

The present study was performed on retinas of chick embryos receiving at day 8 of incubation an intracerebral injection of 0.02 microgram of corticosterone. We had previously shown with the use of [3H]quinuclidinylbenzilate [( 3H]QNB) that such treatment induced the appearance of two muscarinic binding sites in the treated retinas, whereas only one was detectable in the controls. In the present study we investigated muscarinic cholinergic receptor subclasses with agonist and antagonist binding. Agonist binding was studied by varying the concentrations of carbachol and acetylcholine (10(-9) M-10(-5) M) in the presence of a constant concentration (0.2 nM) of [3H]QNB. Two subpopulations of receptors were revealed, a high- and a low-affinity receptor, in both treated and control retinas. However, in the hormone-treated retinas, the two subpopulations significantly differed from the controls in their affinity and in their relative percentage among the total receptor population. Moreover, using pirenzepine, an antagonist known to have the capacity to distinguish between muscarinic cholinergic subclasses, two receptor subpopulations were found to be present in the hormone-treated retinas but a single one in the controls. It is suggested that hormone treatment can either induce the appearance of a new subclass of muscarinic cholinergic receptors or favor the maturation of a population of retinal cells having these receptors. Pirenzepine binding in retinas from intact embryos of 7, 9, and 11 days of incubation revealed one receptor subpopulation. Thus, these findings are more consistent with the hypothesis that corticosterone effects the target cells, either inducing changes in muscarinic receptor and/or modifying the receptor environment.     

Ultracryotomy for the study of unfixed membranes

Plasma membranes from chromaffin cells of bovine adrenal medullae and from chicken macrophages were isolated on a urografin density gradient, frozen and sectioned without previous chemical fixation. Their receptor binding sites were localized by specific labelling. The sections were then post-fixed in the presence of K2Cr2O7 to produce positive staining of the membrane proteins. Chromaffin cell membranes formed single vesicles. The nicotinic acetylcholine receptor (localized using a monoclonal antibody against its cholinergic binding site) was always found in patches on the surface of vesicles, whose profiles corresponded to thickened bilayers. Macrophage membrane vesicles were agglutinated. The mannose receptor (localized using the ligand, mannosylferritin) was randomly distributed within the electron-dense coat of the agglutinated vesicles or on electron-dense caps involved in agglutination. The binding sites of both receptors were intact, as revealed by their being recognized by a monoclonal antibody against their cholinergic binding sites and by the active binding of the mannosylated ligand which was inhibited by mannan. The distribution of the receptors on the vesicles reflected their distribution on the cell surface.     

Nicotinic Acetylcholine Receptors have Ligand-Specific Attachment Point Patterns

Abstract

Employing a panel of synthetic peptides as representative structural elements of the nicotinic acetylcholine receptor from Torpedo electric organ, we recently identified three sequence regions of the receptor (α55–74, α134–153 and α181–200) serving as subsites for the binding of high molecular weight antagonists of acetylcholine (Conti-Tronconi et al. 1990). The relative binding affinities to these subsites of α-bungarotoxin and three competitive antibodies varied in a ligand-specific fashion. Employing a set of homologous synthetic peptides differing from α181–200 by the exchange of single amino acid residues along the sequence, we now find that ligand binding crucially depends on the presence of particular amino acids within the subsite while others influence binding only marginally if at all. The existence of ligand-specific attachment points may account for the wide range in binding and kinetic parameters, pharmacological specificity and distinct mean open times of the receptor-integral cation channel observed for cholinergic ligands.     

Characterization of the binding of alpha-bungarotoxin to bacterially expressed cholinergic binding sites

Bacterially expressed cDNA fragments of the alpha-subunit of the nicotinic acetylcholine receptor previously have been shown to bind alpha-bungarotoxin (Gershoni, J. M. (1987) Proc. Natl. Acad. Sci. U. S. A. 84, 4318-4321). Here, a novel system has been developed in which totally synthetic alpha-bungarotoxin binding sites are expressed in Escherichia coli transformants. The amino acid sequences, alpha 184-200 and alpha 184-196 of the Torpedo californica alpha-subunit of the nicotinic acetylcholine receptor were expressed as trpE fusion proteins via the expression vector pATH2 and a method for the enrichment of these fusion proteins is described. Quantitative analysis of toxin binding to the recombinant binding sites demonstrates that they bind toxin with affinities of KD = 2.5 X 10(-7) and 4.7 X 10(-6) M, respectively. Furthermore, the pharmacological profile of alpha 184-200 qualitatively reflects that of the intact receptor. These data not only indicate that the area of alpha 184-200 is an essential element of the cholinergic binding site but that residues alpha 197-200 contribute a point of contact between the receptor and alpha-bungarotoxin.     

Autoradiographic distribution of high affinity muscarinic and nicotinic cholinergic receptors labeled with [3H]acetylcholine in rat brain

The relative distribution of muscarinic and nicotinic cholinergic receptors labeled with [3H]acetylcholine was determined using autoradiography. [3H]Acetylcholine binding to high affinity muscarinic receptors was similar to what has been described for an M-2 distribution: highest levels of binding occurred in the pontine and brainstem nuclei, anterior pretectal area and anteroventral thalamic nucleus, while lower levels occurred in the caudate-putamen, accumbens nucleus and primary olfactory cortex. Nicotinic receptors were labeled with [3H]acetylcholine to the greatest extent in the interpeduncular nucleus, several thalamic nuclei, medial habenula, presubiculum and superior colliculus, and to the least extent in the hippocampus and inferior colliculus. By using autoradiography to localize cholinergic binding sites throughout the brain it was observed that the distributions of high affinity muscarinic and nicotinic sites labeled with the endogenous ligand, [3H]acetylcholine are different from each other and are different from distributions of muscarinic and nicotinic sites labeled with muscarinic and nicotinic antagonists.