Is agonist self-inhibition at the nicotinic acetylcholine receptor a nonspecific action?

Agonist concentration-response relationships at nicotinic postsynaptic receptors were established by measuring 86Rb+ efflux from acetylcholine receptor rich native Torpedo membrane vesicles under three different conditions: integrated net ion efflux (in 10 s) from untreated vesicles, integrated net efflux from vesicles in which most acetylcholine sites were irreversibly blocked with alpha-bungarotoxin, and initial rates of efflux (5-100 ms) from vesicles that were partially blocked with alpha-bungarotoxin. Exposure to acetylcholine, carbamylcholine, suberyldicholine, phenyltrimethylammonium, or (-)-nicotine over 10(8)-fold concentration ranges results in bell-shaped ion flux response curves due to stimulation of acetylcholine receptor channel opening at low concentrations and inhibition of channel function at 60-2000 times higher concentrations. Concentrations of agonists that inhibit their own maximum 86Rb+ efflux by 50% (KB values) are 110, 211, 3.0, 39, and 8.9 mM, respectively, for the agonists listed above. For acetylcholine and carbamylcholine, KB values determined from both 10-s and 15-ms efflux measurements are the same, indicating that the rate of agonist-induced desensitization increases to maximum at concentrations lower than those causing self-inhibition. For all partial and full agonists studied, Hill coefficients for self-inhibition are close to 1.0. Concentrations of agonists up to 8 times KB did not change the order parameter reported by a spin-labeled fatty acid incorporated in Torpedo membranes. We conclude that agonist self-inhibition cannot be attributed to a general nonspecific membrane perturbation. Instead, these results are consistent with a saturable site of action either at the lipid-protein interface or on the acetylcholine receptor protein itself.     

Single channel gating events in tracer flux experiments. II. Flux amplitude analysis

Measurement of tracer ion flux from or into a collection of closed membrane structures (CMS) constitutes a broadly applicable technique for studying ion channel gating by specialized gating molecules in biological membranes. The amplitudes for the flux process reflect the overall change in tracer content due to flux during a period in which channels on at least some of the CMS were open. In practice, the attainment of a time-invariant, finite overall tracer content, indicating a cessation of flux, need not imply that flux has reached completion, i.e., that the CMS internal and external tracer concentrations have fully reached equilibrium. Less than maximum flux amplitudes arise when binding of control ligands leads to an inhibition or inactivation of the channel gating molecules prior to a complete equilibration of tracer. Analysis of the dependence of the flux amplitudes on control ligand concentration permits determination of characteristic parameters of the CMS that may vary with the methods of preparation (e.g., the distributions of CMS size and CMS content of gating units). Knowledge of these parameters in turn permits evaluation of the mean single channel flux amplitude contribution, which is functionally dependent on the rate constant ratio (k'eff/ki), where k'eff and ki are, respectively, the effective rate constants for tracer flux and for gating unit inactivation.     

High acetylcholine concentrations cause rapid inactivation before fast desensitization in nicotinic acetylcholine receptors from Torpedo.

By using both a 3 to 4 ms quenched-86Rb+ flux assay and native acetylcholine receptor (AChR) rich electroplaque vesicles on which 50-60% of acetylcholine activation sites were blocked with alpha-BTX, we determined apparent rates of agonist-induced inactivation in AChR from Torpedo under conditions where measured flux response was directly proportional to initial 86Rb+ influx rate. Inactivation kinetics with acetylcholine in both the activating range (10 microM-10 mM) and the self-inhibiting range (15-100 mM) were measured at 4 degrees C. In the presence of 10 microM-1 mM acetylcholine, inactivation is characterized by a single exponential rate constant, kd (fast desensitization). Plots of kd vs. acetylcholine concentration display maximum kds [kd(max)] of 6.6-8.0 s-1, half-maximal kd at 102 +/- 16 microM, and a Hill coefficient of 1.6 +/- 0.3, closely paralleling the initial ion flux response of AChR. Thus, fast desensitization probably occurs from a doubly-liganded preopen state or the open channel state. In the self-inhibiting acetylcholine concentration range, inactivation is biphasic. A "rapid inactivation" phase is complete within 30 ms, followed by fast desensitization at a rate close to kd(max). Both the rate and extent of rapid inactivation increase with acetylcholine concentration, indicating that acetylcholine binds to its self-inhibition site with apparent kon approximately equal to 10(3) M-1s-1 and koff approximately equal to 40 s-1. This slow kon suggests either hindered access to the inhibitory allosteric site or that a fast binding step is followed by a slower conformational change leading to channel inhibition. Overall, our data suggest that acetylcholine binds preferentially to its inhibitory site when the receptor is in the open-channel conformation and that fast desensitization can occur from all multiple-liganded states.     

Apparent cooperative effects in acetylcholine receptor-mediated ion flux electroplax membrane preparations.

The kinetics of acetylcholine receptor-mediated flux of ²²sodium ions from microsacs has been measured in the presence of activators (carbamylcholine and decamethonium) and an inhibitor (d-tubocurarine) of neural transmission. The dependence of the first-order rate constant, k_obs, for ²²sodium ion efflux on either decamethonium or carbamylcholine concentration does not exhibit cooperativity. The apparent cooperativity observed by Kasai and Changeux in dose-response curves for ²²sodium flux from the same preparation is adequately accounted for by the contribution which efflux from non-excitable microsacs, the main component of the preparation, makes to the measurements. d-Tubocurarine was found to be a non-competitive inhibitor of decamethonium-activated ²²sodium efflux. The results of the kinetic measurements are in agreement with equilibrium measurements of the interaction of decamethonium with the same microsac preparation, i.e. adherence to a classic Langmuir binding isotherm and separate binding sites for activators and inhibitors of neural activity. The results indicate a direct relationship between ligand binding and receptor-mediated ion flux. How these two processes contribute to electrophysiological measurements is not apparent.     

Effects of pH on acetylcholine receptor function

Summary We have examined the effects of changing extracellular pH on the function of nicotinic acetylcholine receptors fromTorpedo californica using ion flux and electrophysiological methods. Agonist-induced cation efflux from vesicles containing purified, reconstituted receptors showed a monotonic dependence on external hydrogen ion concentration with maximal fluxes at alkaline pH and no agonist-induced efflux at pH's less than 5. A similar pH dependence was measured for the peak agonist-activated membrane currents measured in microelectrode voltage-clampedXenopus oocytes induced to expressTorpedo receptor through mRNA injection. Half-maximal inhibition occurred at a similar pH in both systems, in the range of pH 6.5–7.0. Single-channel currents fromTorpedo ACh receptors measured in patch-clamp recordings were also reduced in amplitude at acid pH with an apparent pK _a for block of <5. Measurements of channel kinetics had a more complicated dependence on pH. The mean channel open time determined from patch-clamp measurements was maximal at neutral pH and decreased at both acid and alkaline pH's. Thus, both channel permeability properties and channel gating properties are affected by the extracellular pH.     

Alpha 1-adrenergic receptor activation mobilizes cellular Ca2+ in a muscle cell line

Regulation of cellular Ca2+ movements by alpha 1-adrenergic receptors has been studied using 45Ca2+ flux techniques in monolayer cultures of intact BC3H-1 cells. Unidirectional 45Ca2+ efflux from BC3H-1 cells reveals multiphasic kinetics, with a major fraction of cellular Ca2+ residing in a slowly exchanging intracellular compartment. Stimulation of alpha 1-adrenergic receptors by the agonist phenylephrine substantially increases 45Ca2+ unidirectional efflux, accompanied by a far smaller increase in 45Ca2+ influx. The selective enhancement of 45Ca2+ unidirectional efflux upon alpha 1-adrenergic receptor activation results in a net 30-40% decline in total cell Ca2+ content, measured either by radioisotopic equilibrium techniques or by atomic absorption spectroscopy. The relatively large pool of Ca2+ responsive to alpha-adrenergic stimulation is not displaced by La3+ but can be depleted with the Ca2+ ionophore A-23187. These results indicate that alpha 1-adrenergic receptor activation predominantly mobilizes Ca2+ from intracellular stores, together with a much smaller increase in transmembrane Ca2+ permeability. This interpretation is supported by comparative 45Ca2+ flux studies using a sister clone of BC3H-1 cells possessing surface nicotinic acetylcholine receptors but no alpha 1-adrenergic receptors. Agonist stimulation of the cholinergic receptor opens a well characterized transmembrane ion permeability gate. Cholinergic receptor activation greatly enhances the observed 45Ca2+ unidirectional influx relative to efflux, leading to net elevation of cellular Ca2+ content as Ca2+ moves down its inwardly directed concentration gradient.     

Acetylcholine-induced receptor-controlled ion flux investigated by flow quench techniques

Using a quench flow technique with membrane vesicles, the acetylcholine receptor-controlled transmembrane ion flux and the inactivation of the receptor with acetylcholine were measured in the msec time region. The ion flux was followed by influx of radioactive tracer ion and the inactivation was followed by an ion flux assay of receptor pre-incubated with ligand. The measurements covered a concentration range to complete saturation of the $\text{active}$ state of the receptor with ligand, and were consistent with a minimal model previously proposed on the basis of experiments with carbamylcholine. The ion translocation rate at saturation with acetylcholine is about twice that at saturation with carbamylcholine and this reflects a more favored channel opening equilibrium for acetylcholine.     

Cocaine, phencyclidine, and procaine inhibition of the acetylcholine receptor: characterization of the binding site by stopped-flow measurements of receptor-controlled ion flux in membrane vesicles

Noncompetitive inhibition of acetylcholine receptor-controlled ion translocation was studied in membrane vesicles prepared from both Torpedo californica and Electrophorus electricus electroplax. Ion flux was measured in the millisecond time region by using a spectrophotometric stopped-flow method, based on fluorescence quenching of entrapped anthracene-1,5-disulfonic acid by Cs+, and a quench-flow technique using 86Rb+. The rate coefficient of ion flux prior to receptor inactivation (desensitization), JA, was measured at different acetylcholine and inhibitor concentrations, in order to assess which active (nondesensitized) receptor forms bind noncompetitive inhibitors. The degree of inhibition of JA by the inhibitors studied (cocaine, procaine, and phencyclidine) was found to be independent of acetylcholine concentration. The results are consistent with a mechanism in which each compound inhibits by binding to a single site that exists with equal affinity on all active receptor forms. Mechanisms in which the inhibitors bind exclusively to the open-channel form of the receptor are excluded by the data. The same conclusions were reached in cocaine experiments at 0-mV and procaine experiments at -25-mV transmembrane voltage in T. californica vesicles. It had been previously shown that phencyclidine, in addition to decreasing JA (by binding to active receptors), also increases the rate of rapid receptor inactivation (desensitization) and changes the equilibrium between active and inactive receptors (by binding better to inactivated receptor than to active receptor in the closed or open conformations). These effects were not observed with cocaine or procaine. Here it is shown that despite these differential effects on inactivation, cocaine and phencyclidine bind to the same inhibitory site on active receptors (in E. electricus vesicles).(ABSTRACT TRUNCATED AT 250 WORDS)     

Actions of avermectin B1a on the gamma-aminobutyric acidA receptor and chloride channels in rat brain

The interaction of avermectin B1a (AVM) with the gamma-aminobutyric acid (GABA) receptor of rat brain was studied using radioactive ligand binding and tracer ion flux assays. Avermectin potentiated the binding of [3H]flunitrazepam and inhibited the binding of both [3H]muscimol and [35S]t-butylbicyclophosphorothionate to the GABAA receptor. Inhibition of muscimol binding by AVM suggested competitive displacement. Two kinds of 36chloride (Cl) flux were studied. The 36Cl efflux from preloaded microsacs was potentiated by AVM and was highly inhibited by the Cl-channel blocker 4,4'-diisothiocyano-2,2'-stilbenedisulfonic acid (DIDS). However, it was not potentiated by GABA nor was it sensitive to the convulsants picrotoxin or bicuculline. On the other hand, 36Cl-influx measurement in a different microsac preparation of rat brain was very sensitive to GABA and other GABA-ergic drugs. Avermectin induced 36Cl influx into these microsacs in a dose-dependent manner, but to only 35% of the maximal influx induced by GABA. The AVM-induced 36Cl influx was totally blocked by bicuculline. It is suggested that AVM opens the GABAA-receptor Cl channel by binding to the GABA recognition site and acting as a partial receptor agonist, and also opens a voltage-dependent Cl channel which is totally insensitive to GABA but is very sensitive to DIDS.     

Postsynaptic effects of crotoxin and of its isolated subunits.

Crotoxin is a potent neurotoxin from the venom of Crotalus durissus terrificus. It is composed of two subunits: a basic phospholipase A2 with low toxicity (component B) and an acidic protein seemingly devoid of intrinsic biological activity (component A). Crotoxin and its isolated phospholipase subunit block the depolarisation caused by cholinergic agonists on the isolated electroplaque from Electrophorus electricus. The other component, which is inactive when applied alone, enhances the pharmacological activity of the phospholipase when the two components are used together. Crotoxin also blocks the increase of 22Na+ efflux caused by carbamylcholine from excitable microsacs prepared from Torpedo marmorata electric organ. Crotoxin therefore acts postsynaptically, but does not interfere with the binding of alpha-toxin from Naja nigricollis to the nicotinic cholinergic receptor site. Instead, like local anesthetics, it stabilizes a desensitized form of the acetylcholine receptor characterized by its high affinity for agonists. The phospholipase component B binds in a non-saturable manner to receptor-rich membranes. In contrast, component A does not bind to acetylcholine receptor-rich membranes, but completely prevents the non-saturable binding of component B. When the two components are applied together, a saturable binding of the latter is observed with the acetylcholine receptor-rich membranes.     

Distribution of Na+ channels and ankyrin in neuromuscular junctions is complementary to that of acetylcholine receptors and the 43 kd protein

We have used immunogold electron microscopy to study the organization of the acetylcholine receptor, 43 kd protein, voltage-sensitive Na+ channel, and ankyrin in the postsynaptic membrane of the rat neuromuscular junction. The acetylcholine receptor and the 43 kd protein are concentrated at the crests of the postsynaptic folds, coextensive with the subsynaptic density. In contrast, Na+ channels and ankyrin are concentrated in the membranes of the troughs and in perijunctional membranes, both characterized by discontinuous submembrane electron-dense plaques. This configuration of interspersed postsynaptic membrane domains enriched in either Na+ channels or acetylcholine receptors may facilitate the initiation of the muscle action potential. Furthermore, the results support the involvement of ankyrin in immobilizing Na+ channels in specific membrane domains, analogous to the proposed involvement of the 43 kd protein in acetylcholine receptor immobilization.     

Functional properties of acetylcholine receptors coexpressed with the 43K protein in heterologous cell systems.

The nicotinic acetylcholine (ACh) receptor is an integral membrane protein which mediates synaptic transmission at the skeletal neuromuscular junction. A key event in the development of the neuromuscular junction is the formation of high density aggregates of ACh receptors in the postsynaptic membrane. Receptor clustering has been attributed, in part, to their association with a peripheral membrane protein of Mr 43,000 (43K protein). We have addressed whether the association of the 43K protein can alter the single channel properties of the ACh receptor, and thus influence neuromuscular transmission at developing synapses, by expressing ACh receptors with and without the 43K protein in heterologous expression systems. We found that coexpression of the 43K protein with the receptor did not significantly alter either its single channel conductance or its mean channel open time. This was true in oocytes and also in COS cells where it was possible to localize 43K-induced clusters by fluorescence microscopy and to record from those clustered receptors. These data are in agreement with previous single channel studies which have shown that the properties of diffusely distributed and clustered receptors in native muscle cells from both mice and Xenopus do not differ.     

A second pathway of activation of the Torpedo acetylcholine receptor channel.

We have studied the interaction of the reversible acetylcholine esterase inhibitor (-)physostigmine (D-eserine) with the nicotinic acetylcholine receptor (nAChR) from Torpedo marmorata electric tissue by means of ligand-induced ion flux into nAChR-rich membrane vesicles and of equilibrium binding. We find that (-) physostigmine induces cation flux (and also binds to the receptor) even in the presence of saturating concentrations of antagonists of acetylcholine, such as D-tubocurarine, alpha-bungarotoxin or antibody WF6. The direct action on the acetylcholine receptor is not affected by removal of the methylcarbamate function from the drug and thus is not due to carbamylation of the receptor. Antibodies FK1 and benzoquinonium antagonize channel activation (and binding) of eserine, suggesting that the eserine binding site(s) is separate from, but adjacent to, the acetylcholine binding site at the receptor. In addition to the channel activating site(s) with an affinity of binding in the 50 microM range, there exists a further class of low-affinity (Kd approximately mM) sites from which eserine acts as a direct blocker of the acetylcholine-activated channel. Our results suggest the existence of a second pathway of activation of the nAChR channel.     

An isotopic rubidium ion efflux assay for the functional characterization of nicotinic acetylcholine receptors on clonal cell lines

An isotopic rubidium ion efflux assay has been developed for the functional characterization of nicotinic acetylcholine receptors on cultured neurons. This assay first involves the intracellular sequestration of isotopic potassium ion analog by the ouabain-sensitive action of a sodium-potassium ATPase. Subsequently, the release of isotopic rubidium ion through nicotinic acetylcholine receptor-coupled monovalent cation channels is activated by application of nicotinic agonists. Specificity of receptor-mediated efflux is demonstrated by its sensitivity to blockade by nicotinic, but not muscarinic, antagonists. The time course of agonist-mediated efflux, within the temporal limitations of the assay, indicates a slow inactivation of receptor function on prolonged exposure to agonist. Dose-response profiles (i) have characteristic shapes for different nicotinic agonists, (ii) are described by three operationally defined parameters, and (iii) reflect different affinities of agonists for binding sites that control receptor activation and functional inhibition. The rubidium ion efflux assay provides fewer hazards but greater sensitivity and resolution than isotopic sodium or rubidium ion influx assays for functional nicotinic receptors.     

Acetylcholine receptor inhibition by d-tubocurarine involves both a competitive and a noncompetitive binding site as determined by stopped-flow measurements of receptor-controlled ion flux in membrane vesicles

The issue of whether d-tubocurarine, the classical acetylcholine receptor inhibitor, inhibits the receptor by a competitive or noncompetitive mechanism has long been controversial. d-Tubocurarine, in this study, has been found to be both a competitive (KC = 120 nM) and a noncompetitive (KNC = 4 microM) inhibitor of receptor-mediated ion flux at zero transmembrane voltage in membrane vesicles prepared from Electrophorus electricus electroplax. A spectrophotometric stopped-flow method, based on fluorescence quenching of entrapped anthracene-1,5-disulfonic acid by Cs+, was used to measure both the rate coefficient of ion flux prior to receptor inactivation (desensitization) and the rate coefficient of the rapid inactivation process. Inhibition by d-tubocurarine of the initial rate of ion flux decreased with increasing acetylcholine concentration, consistent with competitive inhibition, but the inhibition by micromolar concentrations of d-tubocurarine could not be overcome with saturating concentrations of acetylcholine, consistent with noncompetitive inhibition. A minimum mechanism is proposed in which d-tubocurarine competes for one of the two acetylcholine activating sites and also binds to a noncompetitive site. The present data do not distinguish between one or two competitive sites, although one successfully accounts for all of the data. By variation of the acetylcholine concentration, the two types of sites could be studied in isolation. Binding of d-tubocurarine to the noncompetitive site does not change the rate of rapid receptor inactivation, whereas binding of d-tubocurarine to the competitive site decreases the rate of rapid inactivation by displacing acetylcholine, in agreement with the observation that d-tubocurarine does not inactivate (desensitize) the E. electricus receptor by itself.(ABSTRACT TRUNCATED AT 250 WORDS)     

Control of acetylcholine receptor mobility and distribution in cultured muscle membranes. A fluorescence study

The molecular control of the distribution and motion of acetylcholine receptors in the plasma membrane of developing rat myotubes in primary cell culture was investigated by fluorescence techniques. Acetylcholine receptors were marked with tetramethylrhodamine-labeled α-bungarotoxin and lateral molecular motion in the membrane was measured by the fluorescence photobleaching recovery technique. Three types of experiments are discussed: (I) The effect of enzymatic cleavages, drugs, cross-linkers, and physiological alterations on the lateral motion of acetylcholine receptors and on the characteristic distribution of acetylcholine receptors into patch and diffuse areas. (II) Observation of the distribution and/or motion of fluorescence-labeled concanavalin A receptors, lipid probes, cell surface protein, and stained cholinesterase in acetylcholine receptor patch and diffuse areas. (III) The effect of a protein synthesis inhibitor and electrical stimulation on membrane incorporation of new acetylcholine receptors.Some of the main conclusions are: (a) acetylcholine receptor lateral motion is inhibited by concanavalin A plant lectin and by anti-α-bungarotoxin antibody, but marginally enhanced by treatment with a local anesthetic; (b) patches are stabilized by an immobile cellular structure consisting of molecules other than the acetylcholine receptors themselves; (c) this structure is highly selective for acetylcholine receptors and not for other cell membrane components; (d) acetylcholine receptor patch integrity and diffuse area motion are independent of direct metabolic energy requirements and are sensitive to electrical excitation of myotube; (e) lipid molecules can move laterally in both acetylcholine receptor patches and diffuse areas; and (f) acetylcholine receptor lateral motion in diffuse areas and immobility in patch areas are not altered by specific agents which are known to affect extrinsic cell surface proteins, or cytoplasmic microfilaments and microtubules.     

Muscarinic acetylcholine receptor produced in recombinant baculovirus infected Sf9 insect cells couples with endogenous G-proteins to activate ion channels.

Following the infection of insect ovarian cells (Sf9) with recombinant bearing the cDNA coding for the rat muscarinic acetylcholine (ACh) receptor subtype m3, ionic flux across the membrane in response to the application of ACh was examined electrophysiologically. We show that ACh activates potassium currents. The response is abolished when cells are treated with pertussis toxin. No ACh-induced currents are observed from uninfected cells or cells infected with virus which do not contain the cDNA coding for ACh receptors in its genome. The characteristics of single channel currents show time-dependent changes following the application of ACh. Initially, ACh activates brief channel currents with a conductance of about 5 pS. The conductance level of channels gradually increases in steps to 10 pS and then to 20 pS and 40 pS. At the same time, channel open probability also increases. Thereafter, additional channels appear, opening and closing independently of, or at times in synchrony with, the original channel.     

Acetylcholine receptor: channel-opening kinetics evaluated by rapid chemical kinetic and single-channel current measurements.

A combination of rapid chemical kinetic (quench-flow) and single-channel current measurements was used to evaluate kinetic parameters governing the opening of acetylcholine-receptor channels in the electric organ (electroplax) of Electrophorus electricus. Chemical kinetic measurements made on membrane vesicles, prepared from the E. electricus electroplax, using carbamoylcholine (200 microM-20 mM) at 12 degrees C, pH 7.0, and in the absence of a transmembrane voltage, yielded values for K1 (dissociation constant for receptor activation), phi (channel closing equilibrium constant), J (specific reaction rate for ion flux), and alpha max (maximum inactivation rate constant) of 1 mM, 3.4, 4 x 10(7) M-1 s-1, and 12 s-1, respectively. The single-channel current recordings were made with cells also from the E. electricus electroplax, at the same temperature and pH as the chemical kinetic measurements, using carbamoylcholine (50 microM-2 mM), acetylcholine (500 nM), or suberyldicholine (20 nM). Single-channel current measurements indicated the presence of a single, unique open-channel state of the E. electricus receptor, in concurrence with previous, less extensive measurements. The rate constant for channel closing (kc) obtained from the mean open time of the receptor channel is 1,100 s-1 for carbamoylcholine, 1,200 s-1 for acetylcholine, and 360 s-1 for suberyldicholine at zero membrane potential; and it decreases e-fold for an 80 mV decrease in transmembrane voltage in each case. The decrease in mean open times of the receptor channel that is associated with increasing the carbamoylcholine concentration is interpreted to be due to carbamoylcholine binding to the regulatory (inhibitory) site on the receptor. An analysis of data obtained with carbamoylcholine showed that the closed times within a burst of channel activity fit a two-exponential distribution, with a concentration-independent time constant considered to be the time constant for carbamoylcholine to dissociate from the regulatory site, and a carbamoylcholine concentration-dependent, but voltage-independent, time constant interpreted to represent the rate constant for channel opening (k0). An analysis of the mean closed time data on the basis of the minimum model gives values for K1 and k0 of 0.6 mM and 440 s-1, respectively, with carbamoylcholine as the activating ligand. The values obtained for K1, phi (= kc/k0), J, and alpha from the single-channel current measurements using electroplax are in good agreement with the values obtained from the chemical kinetic measurements using receptor-rich vesicles prepared from the same cells.(ABSTRACT TRUNCATED AT 400 WORDS)     

Reaction of tetraethylammonium with the open and closed conformations of the acetylcholine receptor ionic channel complex

The effect of tetraethylammonium (TEA) bromide on the neurally and iontophoretically evoked endplate current (EPC) of frog sartorius muscle was investigated using voltage-clamp and noise analysis techniques, and its binding to the acetylcholine (ACh) receptor ionic channel complex was determined on the electric organ of Torpedo ocellata. TEA (250-500 microM) produced an initial enhancement followed by a slow decline in the amplitude of the endplate potential and EPC, but caused only depression in the amplitude of the miniature endplate potential and current. In normal ringer's solution, the EPC current-voltage relationship was approximately linear, and the decay phase varied exponentially with membrane potential. Upon addition of 50-100 microM TEA, the current-voltage relationship became markedly nonlinear at hyperpolarized command potentials, and with 250-2000 microM TEA, there was an initial linear segment, an intermediate nonlinear segment, and a region of negative conductance. The onset of nonlinearity was dose-dependent, undergoing a 50 mV shift for a 10-fold increase in TEA concentration. The EPC decay phase was shortened by TEA at hyperpolarized but not depolarized potentials, and remained a single expotential function of time at all concentrations and membrane potentials examined. These actions of TEA were found to be independent of the sequence of polarizations, the length of the conditioning pulse, and the level of the initial holding potential. TEA shifted the power spectrum of ACh noise to higher frequencies and produced a significant depression of single channel conductance. The shortening in the mean channel lifetime agreed closely with the decrease in the EPC decay time constant. At the concentrations tested, TEA did not alter the EPC reversal potential, nor the resting membrane potential, and had little effect on the action potential duration. TEA inhibited the binding of both [3H] ACh (Ki = 200 microM) and [3H]perhydrohistrionicotoxin (Ki = 280 microM) to receptor-rich membranes from the electric organ of Torpedo ocellata, and inhibited the carbamylcholine-activated 22Na+ efflux from these microsacs. It is suggested that TEA reacts with the nicotinic ACh-receptor as well as its ion channel; the voltage-dependent actions are associated with blockade of the ion channel. The results are compatible with a kinetic model in which TEA first binds to the closed conformation of the receptor-ionicchannel complex to produce a voltage-depdndent depression of endplate conductance and sudsequently to its open conformation, giving rise to the shortening in the EPC decay and mean channel lifetime.