Search

The olfactory system is remarkably sensitive to airborne odor molecules, but precisely how very low odor concentrations bordering on just a few molecules per olfactory sensory neuron can trigger graded changes in firing is not clear. This report reexamines signaling in olfactory sensory neurons in light of the recent account of NaV1.5 sodium channel-mediated spontaneous firing. Using a model of spontaneous channel activity, the study shows how even submillivolt changes in membrane potential elicited by odor are expected to cause meaningful changes in NaV1.5-dependent firing. The results suggest that the random window currents of NaV1.5 channels may underpin not only spontaneous firing in olfactory sensory neurons but the cellular response to odor as well, thereby ensuring the robustness and sensitivity of signaling that is especially important for low odor concentrations., (Copyright © 2016 the American Physiological Society.)

Olfactory sensory neurons (OSNs) fire spontaneously as well as in response to odor; both forms of firing are physiologically important. We studied voltage-gated Na(+) channels in OSNs to assess their role in spontaneous activity. Whole cell patch-clamp recordings from OSNs demonstrated both tetrodotoxin-sensitive and tetrodotoxin-resistant components of Na(+) current. RT-PCR showed mRNAs for five of the nine different Na(+) channel α-subunits in olfactory tissue; only one was tetrodotoxin resistant, the so-called cardiac subtype NaV1.5. Immunohistochemical analysis indicated that NaV1.5 is present in the apical knob of OSN dendrites but not in the axon. The NaV1.5 channels in OSNs exhibited two important features: 1) a half-inactivation potential near -100 mV, well below the resting potential, and 2) a window current centered near the resting potential. The negative half-inactivation potential renders most NaV1.5 channels in OSNs inactivated at the resting potential, while the window current indicates that the minor fraction of noninactivated NaV1.5 channels have a small probability of opening spontaneously at the resting potential. When the tetrodotoxin-sensitive Na(+) channels were blocked by nanomolar tetrodotoxin at the resting potential, spontaneous firing was suppressed as expected. Furthermore, selectively blocking NaV1.5 channels with Zn(2+) in the absence of tetrodotoxin also suppressed spontaneous firing, indicating that NaV1.5 channels are required for spontaneous activity despite resting inactivation. We propose that window currents produced by noninactivated NaV1.5 channels are one source of the generator potentials that trigger spontaneous firing, while the upstroke and propagation of action potentials in OSNs are borne by the tetrodotoxin-sensitive Na(+) channel subtypes., (Copyright © 2014 the American Physiological Society.)

Coupling of olfactory sensory neurons (OSNs) in the olfactory epithelium of Necturus maculosus was demonstrated by dye-transfer with Lucifer yellow CH; however, the incidence of dye-transfer was low. Immunocytochemistry and Western blot analysis indicated that connexin 43, a gap junction channel subunit, was widely expressed by cells in the olfactory epithelium. Electrical coupling by presumptive gap junctions was assessed using electrophysiological recordings, heptanol block, tracer-uptake through hemi-junctions, and tracer-injection into tissue whole-mounts. Coupling, which involved pairs of OSNs only, was detected in approximately 3-10% of the OSN population; there was no evidence that OSNs were coupled into extended neural syncitia. These results suggest that coupling of OSNs by gap junctions is unlikely to have a general role in olfactory responses by mature (odor responsive) OSNs. Instead, the incidence of inter-neuronal coupling was small, similar to the fraction of immature OSNs, suggesting a possible role of gap junctions in the continual turnover and development of OSNs or possibly their senescence.

Odor transduction mediated by the adenylyl cyclase/cAMP pathway has been well studied, but it is still uncertain whether this pathway mediates the transduction of all odors in vertebrates. We isolated olfactory sensory neurons from the salamander Necturus maculosus and used calcium imaging with the indicator dye fura-2 to examine olfactory responses elicited by amino acids. The properties of approximately two-thirds of the odor responses suggested they were mediated by the adenylyl cyclase/cAMP pathway, but one-third of the responses were not mimicked by cAMP analogs nor blocked by inhibition of adenylyl cyclase, suggesting that these odor responses were mediated differently. Responses that were unaffected by inhibition of adenylyl cyclase were blocked by neomycin, an inhibitor of phospholipase C, implying that they were transduced by activation of phospholipase C. Some cells which responded to more than one amino acid appeared to employ both pathways, but each was used to transduce different odors. In addition, many responses that were mediated by the adenylyl cyclase/cAMP pathway were enhanced following inhibition of phospholipase C, suggesting that the phospholipase C pathway has a role not only in odor transduction, but also in the modulation of olfactory responses.

The terminal nerve is an anterior cranial nerve that innervates the lamina propria of the chemosensory epithelia of the nasal cavity. The function of the terminal nerve is ambiguous, but it has been suggested to serve a neuromodulatory role. We tested this hypothesis by exposing olfactory receptor neurons from mudpuppies (Necturus maculosus) to a peptide, gonadotropin releasing hormone (GnRH), that is found in cells and fibers of the terminal nerve. We used voltage-clamped whole-cell recordings to examine the effects of 0. 5-50 micrometer GnRH on voltage-activated currents in olfactory receptor neurons from epithelial slices. We found that GnRH increases the magnitude, but does not alter the kinetics, of a tetrodotoxin-sensitive inward current. This increase in magnitude generally begins 5-10 min after initial exposure to GnRH, is sustained for at least 60 min during GnRH exposure, and recovers to baseline within 5 min after GnRH is washed off. This effect occurred in almost 60% of the total number of olfactory receptor neurons examined and appeared to be seasonal: approximately 67% of neurons responded to GnRH during the courtship and mating season, compared with approximately 33% during the summer, when the sexes separate. GnRH also appears to alter an outward current in the same cells. Taken together, these data suggest that GnRH increases the excitability of olfactory receptor neurons and that the terminal nerve functions to modulate the odorant sensitivity of olfactory receptor neurons.

Whole-cell membrane currents were recorded from olfactory receptor neurons from the neotenic salamander Necturus maculosus. Cyclic nucleotides, released intracellularly by flash photolysis of NPE-caged cAMP or NPE-caged cGMP, activated a transient chloride current. The chloride current could be elicited at constant voltage in the absence of extracellular Ca2+ as well as in the presence of 3 mM intracellular Ca2+, suggesting that the current did not require either voltage or Ca2+ transients for activation. The current could be elicited in the presence of the protein kinase inhibitors H-7 and H-89, and in the absence of intracellular ATP, indicating that activation was independent of protein kinase A activity. These results suggest that Necturus olfactory receptor neurons contain a novel chloride ion channel that may be directly gated by cyclic nucleotides.

Odors are powerful stimuli that can focus the attention, elicit behaviors (or misbehaviors) and even resurrect forgotten memories. These actions are directed by the central nervous system, but they depend upon the initial transduction of chemical signals by olfactory receptor neurons. Electrophysiological recordings suggest that the responses of olfactory receptor neurons to odors are more diverse than was initially believed, being mediated by effects on several different conductances. Both excitatory and inhibitory responses are produced by these effects and some, if not all, odors can affect more than one component of the membrane conductance. The extent of this diversity is reviewed here, and its impact on our understanding of odor discrimination is discussed.

Figure 1 summarizes our current concept of a signaling mechanism to explain agonist-induced Ca2+ entry in the pancreatic acinar cell. We propose that cGMP can modulate Ca2+ entry under conditions of internal Ca2+ store depletion and that the NO signaling system may be involved in coupling Ca2+ depletion to cGMP formation. The finding that Ca2+ entry after Ca2+ store depletion can occur with no elevation in [Ca2+]i37 raises the possibility that alternative signaling pathways may converge to stimulate cGMP formation or that additional messengers may activate plasmalemmal Ca2+ entry mechanisms in parallel.

Odors affect the excitability of an olfactory neuron by altering membrane conductances at the ciliated end of a single, long dendrite. One mechanism to increase the sensitivity of olfactory neurons to odorants would be for their dendrites to support action potentials. We show for the first time that isolated olfactory dendrites from the mudpuppy Necturus maculosus contain a high density of voltage-activated Na+ channels and produce Na-dependent action potentials in response to depolarizing current pulses. Furthermore, all required steps in the transduction process beginning with odor detection and culminating with action potential initiation occur in the ciliated dendrite. We have previously shown that odors can modulate Cl- and K+ conductances in intact olfactory neurons, producing both excitation and inhibition. Here we show that both conductances are also present in the isolated, ciliated dendrite near the site of odor binding, that they are modulated by odors, and that they affect neuronal excitability. Voltage-activated Cl- currents blocked by 4,4'-diisothiocyanatostilbene-2,2' disulfonic acid and niflumic acid were found at greater than five times higher average density in the ciliated dendrite than in the soma, whereas voltage-activated K+ currents inhibited by intracellular Cs+ were distributed on average more uniformly throughout the cell. When ciliated, chemosensitive dendrites were stimulated with the odorant taurine, the responses were similar to those seen in intact cells: Cl- currents were increased in some dendrites, whereas in others Cl- or K+ currents were decreased, and responses washed out during whole-cell recording. The Cl- equilibrium potential for intact neurons bathed in physiological saline was found to be -45 mV using an on-cell voltage-ramp protocol and delayed application of channel blockers. We postulate that transduction of some odors is caused by second messenger-mediated modulation of the resting membrane conductance (as opposed to a specialized generator conductance) in the cilia or apical region of the dendrite, and show how this could alter the firing frequency of olfactory neurons.

In the pancreatic acinar cell, hormonal stimulation causes a rise in the intracellular free Ca2+ concentration by activating the inositol 1,4,5-trisphosphate-mediated release of Ca2+ from intracellular stores (Berridge, M. J., and Irvine, R. F. (1989) Nature 341, 197-205). The released Ca2+ is, for the most part, extruded from the cell, necessitating a mechanism for Ca2+ entry and reloading of intracellular Ca2+ stores (Putney, J. W., Jr. (1990) Cell Calcium 11, 611-624; Rink, T. J. (1990) FEBS Lett. 268, 381-385). However, neither the mechanism of depletion-activated Ca2+ entry nor the signal that activates it is known. We report here that a sustained inward current of depletion-activated Ca2+ entry can be measured in pancreatic acinar cells using patch-clamp recording methods. Furthermore, the current can be blocked by an inhibitor of guanylyl cyclase, can be reactivated by 8-bromo-cGMP after inhibition, and can be activated in the absence of Ca2+ depletion by perfusing the cell with cGMP, but not cAMP. Intracellular perfusion with 1,3,4,5-inositol tetrakisphosphate did not activate an inward current, whereas perfusion with 2,4,5-inositol trisphosphate did activate an inward current. We conclude that cGMP may be an intracellular messenger that regulates depletion-activated Ca2+ entry.

Bovine serum albumin greatly enhanced the cholinergic response mediated by neuronal nicotinic acetylcholine receptors in chick ciliary ganglion neurons. The enhancement exceeded 5-fold in some experiments (mean +/- standard error, 3.26 +/- 0.43-fold) and was rapid, was dose dependent, and occurred without changes in the unitary conductance or the mean open time of the acetylcholine receptor channel. This lack of detectable change in permeation or kinetic properties suggests that bovine serum albumin might increase acetylcholine responses by increasing the number of functional receptors. The enhancement appears to be specific to the albumin molecule, because activity could not be removed by detergent extraction, gel filtration, or dialysis. Acetylcholine responses in these cells are known to be enhanced by a cAMP-dependent mechanism that converts existing acetylcholine receptors from a nonfunctional to a functional state. We found that the enhancement by bovine serum albumin occurred without an increase in cAMP and that pretreatment with membrane-permeable cAMP analogs prevented any additional enhancement of the cholinergic response by bovine serum albumin. These observations are consistent with a cAMP-dependent modulation of the enhancement produced by bovine serum albumin or a convergence of the two enhancement mechanisms onto a single pathway.

Odors are transduced by processes that modulate the membrane conductance of olfactory receptor neurons. Olfactory neurons from the aquatic salamander, Necturus maculosus, were acutely isolated without enzymes and studied with a resistive whole-cell method to minimize loss of soluble intracellular constituents. 55 of 224 neurons responded to the test compound taurine at concentrations between 10 nM and 100 microM. Four different conductance changes were elicited by taurine: an increased Cl- conductance (33%), an increased nonselective cation conductance (15%), a decreased Cl- conductance (15%), and a decreased K+ conductance (15%); in addition, responses too small to be characterized were elicited in some neurons. In most cases, taurine appeared to modulate only a single conductance in any particular cell. Modulation of each conductance was dose dependent, and each response ran down quickly in the normal whole-cell mode, presumably due to washout of a diffusible component in the transduction pathway. Modulation of taurine-sensitive conductances caused either inhibitory or excitatory responses. A similar diversity of responses in vivo would produce a complex pattern of electrical activity that could encode the identity and characteristics of an odor.

A PCR-based strategy for amplifying putative receptors involved in murine olfaction was employed to isolate a member (OR3) of the seven-transmembrane-domain receptor superfamily. During development, the first cells that express OR3 appear adjacent to the wall of the telencephalic vesicle at embryonic day 10. The OR3 receptor is uniquely expressed in a subset of olfactory cells that have a characteristic bilateral symmetry in the adult olfactory epithelium. This receptor and its specific pattern of expression may serve a functional role in odor coding or, alternatively, may play a role in the development of the olfactory system.

Olfactory receptor neurons were isolated without enzymes from the mudpuppy, Necturus maculosus, and tested for chemosensitivity. The cells responded to odorants with changes in firing frequency and alterations in excitability that were detected with tight-seal patch electrodes using on-cell and whole-cell recording conditions. Chemosensitive cells exhibited two primary response characteristics: excitation and inhibition. Both types of primary response were observed in different cells stimulated by mixtures of amino acids as well as by the single compound L-alanine, suggesting that there may be more than one transduction pathway for some odorants. Using the normal whole-cell recording method, the chemosensitivity of competent cells washed out rapidly; a resistive whole-cell method was used to record odorant responses under current-clamp conditions. In response to chemical stimulation, excitability appeared to be modulated in several different ways in different cells: odorants induced hyperpolarizing or depolarizing receptor potentials, elicited or inhibited transient, rhythmic generator potentials, and altered excitability without changing the membrane potential or input resistance. These effects suggest that olfactory transduction is mediated through at least three different pathways with effects on four or more components of the membrane conductance. Polychotomous pathways such as these may be important for odor discrimination and for sharpening the "odor image" generated in the olfactory epithelium.

The dependence of acetylcholine receptor mean single-channel conductance on temperature was studied at garter snake twitch-muscle endplates using fluctuation analysis. In normal saline under conditions where most of the endplate current was carried by Na+, the channel conductance increased continuously from near 0 degrees C to approximately 23 degrees C with a Q10 of 1.97 +/- 0.14 (mean +/- SD). When 50% of the bath Na+ was replaced by either Li+, Rb+, or Cs+, the Q10 did not change significantly; however, at any temperature the channel conductance was greatest in Cs-saline and decreased with the ion sequence Cs greater than Rb greater than Na greater than Li. The results were fit by an Eyring-type model consisting of one free-energy well on the extracellular side of a single energy barrier. Ion selectivity appeared to result from ion-specific differences in the well and not in the barrier of this model. With a constant barrier enthalpy for different ions, well free-energy depth was greatest for Cs+ and graded identical to the permeability sequence. The correlation between increased well depth (i.e., ion binding) and increased channel conductance can be accounted for by the Boltzmann distribution of thermal energy.

Excitatory postsynaptic currents (EPSCs) have been studied in voltage-clamped bullfrog sympathetic ganglion B cells. The EPSC was small, rose to a peak within 1-3 ms, and then decayed exponentially over most of its time-course. For 36 cells at --50 mV (21-23 degrees C), peak EPSC size was --6.5 +/- 3.5 nA (mean +/- SD), and the mean decay time constant tau was 5.3 +/- 0.9 ms. tau showed a small negative voltage dependence, which appeared independent of temperature, over the range --90 to --30 mV; the coefficient of voltage dependence was --0.0039 +/-0.0014 mV-1 (n = 29). The peak current-voltage relationship was linear between --120 and --30 mV but often deviated from linearity at more positive potentials. The reversal potential determined by interpolation was approximately --5 mV. EPSC decay tau had a Q10 = 3. The commonly used cholinesterase inhibitors, neostigmine and physostigmine, exhibited complex actions at the ganglia. Neostigmine (1 X 10(-5)M) produced a time-dependent slowing of EPSC decay without consistent change in EPSC size. In addition, the decay phase often deviated from a single exponential function, although it retained its negative voltage dependence. With 1 x 10(-6) M physostigmine, EPSC decay was slowed by the decay phase remained exponential. At higher concentrations of physostigmine, EPSC decay was markedly prolonged and was composed of at least two decay components. High concentrations of atropine (10(-5) to 10(-4) M) produced complex alterations in EPSC decay, creating two or more exponential components; one decay component was faster and the other was slower than that observed in untreated cells. These results suggest that the time-course of ganglionic EPSC decay is primarily determined by the kinetics of the receptor-channel complex rather than hydrolysis or diffusion of transmitter away from the postsynaptic receptors.

Ionic conductances present in the dendritic region of the cerebellar Purkinje neuron were studied using the single-channel and whole-cell recording methods. Several types of voltage-sensitive K+ channels including a Ca2+ activated K+ channel were found to be a prominent components of the dendritic membrane. All patches studied contained K+ channel types and most patches contained more than one K+ channel type. In cell attached recordings, K+ channel activity was associated with the late phase of spontaneous action potentials suggesting a functional relationship. These data demonstrate that voltage-sensitive ion channels contribute to dendritic excitability and suggest that the transduction and integration of synaptic signals may involve both active and passive ionic conductances.

The iontophoretic release of drugs from micropipettes into free (Ringer's) solution was described using an ion-selective microelectrode assay method. This characterization, with a temporal resolution of 20 ms, showed that the equilibrium rate of drug transport was not linearly proportional to release current; the departure from linearity was increased by backing current and the result was demonstrated with analytically derived drug release functions. The general relation between the drug transport rate and release current was independent of the specific drug or pipette resistance; no functional relation was observed that might quantitatively predict this dependence without prior use of the assay. The diffusion coefficients at 25 degrees C in frog Ringer's of the drugs used in this study, all neuromuscular agonists, were determined: all values X 10(6) cm2/s; acetylcholine 6.11 +/- 0.30; carbamylcholine 7.44 +/- 0.34; 3-(m-hydroxyphenyl) prophyltrimethyl ammonium 5.79 +/- 0.13.

Biochemical and immunological studies indicate that the GABAA receptor contains at least two types of subunit. Here we report that coexpression of two GABAA receptor subunit clones (alpha and beta) in Xenopus oocytes yields receptors with many biophysical properties of native GABAA receptors. These include ion selectivity, multiple single-channel conductance states, voltage-dependent gating and rectification, and complex desensitization kinetics. Furthermore, the receptors are competitively inhibited by bicuculline and display the expected allosteric and agonist effects of the barbiturate pentobarbital. The expressed receptors, however, appear to be activated by one molecule of GABA instead of two and fail to show potentiation by benzodiazepines. This implies that an additional factor(s) or subunit(s) is required for the reconstitution of a fully functional GABAA receptor.

A method is described for producing large numbers of isolated olfactory receptor neurons from adult mouse nasal epithelium. The dissociated neurons and other cell types isolated from nasal epithelium retain their morphology and can be identified visually. The neurons were judged to be intact and viable by trypan blue dye exclusion, the presence of olfactory marker protein (OMP), and a variety of electrophysiological measurements indicating the presence of substantial membrane potentials, low levels of intracellular Ca2+, and the ability to fire action potentials. The receptor neurons and other cell types produced by this method are amenable to the patch-clamp technique and to immunohistochemical studies.

The temporal relationships among junctional acetylcholine receptor single-channel currents have been examined to probe the mechanism of channel activation. We have presented an analytical approach, termed single-channel ensemble analysis, that allows one to estimate the kinetic transition rate constants for channel-opening and closing as well as the rate of leaving the specific doubly-liganded, closed state from which opening occurs. This approach may be applied to data produced by any number of independent channels as long as the probability of channel opening is low, a condition that is experimentally verifiable. The method has been independently validated using simulated single-channel data generated by computer from one or 100 hypothetical channels. Typical experimental values for the transition rate constants estimated from acetylcholine-activated single channels at the garter snake neuromuscular junction were: opening = 1,200 s-1, closing = 455 s-1, back rate for leaving the doubly-liganded, closed state = 3,200 s-1 at a transmembrane potential of -92 mV at room temperature. Each of these three rate constants was voltage dependent, with the closing rate decreasing e-fold for 173 mV of hyperpolarization, the opening rate increasing e-fold for 78 mV, and the unbinding rate increasing e-fold for 105 mV. The channel-closing rate was agonist dependent, being greater at all potentials for channels activated with carbamylcholine than for channels activated with acetylcholine. However, the single-channel conductance and reversal potential were the same for these two agonists.

Acetylcholine receptors in slow fiber neuromuscular junctions of garter snake (sp. Thamnophis) produced synaptic responses that were more complicated than those observed from twitch fibers. Although the slow fiber miniature end plate currents decayed monoexponentially with time, both the current fluctuations spectrum and the voltage jump end plate current required two temporal components for good theoretical fits. This behavior was accurately accounted for by a generalized version of the three-state kinetic model by del Castillo and Katz. Application of the model allowed not only the rate of channel closing to be estimated, but also the rate of channel opening (from the closed state with acetylcholine bound) and the apparent rate of acetylcholine unbinding from the receptor. The results suggest that at the peak of the miniature end plate current local receptor saturation occurs.

The design and construction of a novel apparatus to fire polish patch-clamp recording pipets is described. The device positions the polishing filament in the field of view of the high-power polishing objective and uses the mechanical stage of the microscope to hold the electrode, eliminating the need for a micromanipulator to hold either the filament or electrode.

A developmental change in the ionic basis of the inward current of action potentials has been observed in many excitable cells. In cultured spinal neurones of Xenopus, the timing of the development of the action parallels that seen in vivo. In vitro, as in vivo, neurones initially produce action potentials of long duration which are principally Ca-dependent; after 1 day of development the impulse is brief and primarily Na-dependent. At both ages, however, both inward components are present and the mechanism underlying shortening of the action potential is unknown. One possibility is that the outward currents change during development. Using the patch-clamp technique, we have recorded single K+-channel currents in membrane patches isolated from the cell bodies of cultured embryonic neurones. The unitary conductance of one class of K+ channels was approximately 155 pS and depolarization increased the probability of a channel being open. Neither conductance nor voltage dependence seemed to change with time in culture; in contrast, the Ca2+-sensitivity of this K+ channel increased. In younger neurones, Ca2+-sensitivity was greatly reduced or absent, whereas in more mature neurones, the activity of this channel was Ca-dependent. Such a change could account for the shortening of the action potential duration by increasing the relative contribution of outward currents.

The alpha and beta subunits of the gamma-aminobutyric acidA (GABAA) receptor were expressed individually in Xenopus oocytes by injection of RNA synthesized from their cloned DNAs. GABA-sensitive chloride channels were detected several days after injection with any one of three different alpha RNAs (alpha 1, alpha 2, and alpha 3) or with beta RNA. The channels induced by each of the alpha-subunit RNAs were indistinguishable, they had multiple conductance levels (10, 19, 28, and 42 picosiemens), and their activity was potentiated by pentobarbital and inhibited by picrotoxin. The beta channels usually expressed poorly but showed similar single channel conductance levels (10, 18, 27, and 40 picosiemens), potentiation by pentobarbital and inhibition by picrotoxin. The finding that both alpha and beta subunits, examined separately, form GABA-sensitive ion channels with permeation properties and regulatory sites characteristic of the native receptor suggests that the amino acid sequences that confer these properties are within the homologous domains shared by the subunits.

Previous studies have shown that the number of functional acetylcholine receptors (AcChoRs) on chicken ciliary ganglion neurons in culture is considerably smaller than the total number of AcChoRs detected on the neurons by labeled receptor probes. Here we use patch-clamp recording to show that a cAMP-dependent process enhances the AcCho response of the neurons by a mechanism likely to involve an increase in the number of functional AcChoRs. The increase occurs without requiring protein synthesis and without involving a detectable increase in the total number of AcChoRs on the cell surface measured with a labeled receptor probe. The results imply that the neurons have functional and nonfunctional pools of AcChoRs and that functional receptors can be recruited from intracellular receptors or from nonfunctional receptors on the cell surface by a cAMP-dependent process. A cAMP-dependent regulation of the number of functional neurotransmitter receptors would provide a reversible mechanism by which cell-cell interactions could modulate synaptic transmission in the nervous system.

1. Slow muscle synaptic responses were modelled kinetically in an attempt to define the mechanism by which slow fibre acetylcholine-operated channels differ from those in twitch fibres. 2. Three kinetically distinguishable states were necessary. 3. All applicable three-state kinetic schemes were considered in an attempt to identify the simplest description of the data. Experimental tets eliminated several models. Two models were not tested because they contained an excessive number of adjustable parameters. 4. The data were not fitted by kinetic schemes which postulated (i) channels which opened with one as well as two bound agonist molecules, (ii) channels which became blocked after opening, or (iii) separate populations of synaptic and extrasynaptic channels. 5. The three-state kinetic model of del Castillo & Katz (1957) accurately described all the data. This sequential model relates a closed channel state with no agonist bound to its receptors, an intermediate state (also closed) with agonist bound, and an open channel state. It is the same model which has been used to describe synaptic responses in twitch fibres. 6. The variation which allows this model to describe both twitch and slow fibre synaptic responses is the lifetime of the intermediate state. In twitch fibres the intermediate state lifetime is undetectably brief by electrophysiological methods. However, in slow fibres this lifetime appears to be 1-2 msec, varying with voltage. 7. Three of the four transition rates in this three-state kinetic scheme may be estimated by fitting the model to the data. These are the channel opening rate, the channel closing rate and the rate at which closed channels lose their bound agonist molecules. The latter two rates appear to depend exponentially on voltage. The channel opening rate was not detectably voltage-sensitive.

Single-channel currents from acetylcholine receptor channels of garter snake neuromuscular junctions were recorded using the patch-clamp technique. Low concentrations of acetylcholine or carbamylcholine induced populations of single current events whose amplitudes and durations had unimodal distributions. The probability with which channel opening transitions occurred was time dependent, so that it was more probable for channels to open during the several hundred microseconds following a closing transition than during any later equivalent interval. The time-dependent distributions of duration and opening-transition probability were fitted by a sequential, reversible kinetic model in which the agonist binding steps occur before, and separately from, channel activation. This description allowed estimates to be obtained of both the opening (approximately 750s-1) and closing (approximately 500s-1) transition rates of these channels and of the mean lifetimes of the open- (approximately 2 ms) and the closed-channel state (approximately 200 mus) to which the open state was reversibly related.

1. End-plate currents produced by nerve-released acetylcholine and iontophoretically applied acetylcholine and carbachol have been recorded from voltage-clamped frog cutaneous pectoris neuromuscular junctions made visible with Nomarski differential interference contrast optics. 2. The effectiveness of agonists - that is, the end-plate conductance change produced by a given dose-has been determined as a function of post-junctional membrane potential. 3. As the post-junctional membrane potential is made more negative, nerve-released acetylcholine becomes less effective whereas iontophoretically-applied agonists become more effective. 4. This voltage dependence of agonist effectiveness is mediated neither by end-plate current iontophoresis of agonist into the cleft nor through electric field effects on the esterase. 5. Influences of membrane potential on the opening and closing of end-plate channel gates can account quantitatively for the voltage-dependent effectiveness of both nerve-released and iontophoretically applied agonist.

Previous studies have shown that mudpuppy taste receptor cells respond to sour taste stimuli (weak acids) with depolarizing receptor potentials or action potentials that are blocked by the K+ channel blocker tetraethylammonium. Voltage-clamp recordings from isolated taste cells indicated that taste receptor cells exhibit a variety of voltage-dependent conductances and that acids reduce a voltage-dependent K+ current. Since taste stimuli are restricted to the apical surface of the intact tongue, only 1-2% of the taste receptor cell surface is exposed to chemical stimuli. Thus, modification of a K+ conductance would be an effective transduction mechanism in receptor cells only if the majority of K+ channels were located on the apical membrane. We have used a combination of "loose-patch" and whole-cell recording methods to map the distribution of voltage-sensitive K+ and Na+ channels on dissociated Necturus maculosus taste cells. We report here that the K+ conductance is approximately equal to 50-fold greater on apical membrane than on basolateral membrane, whereas the Na+ conductance is distributed evenly. The marked nonuniformity of the voltage-sensitive K+ conductance, together with the block of this conductance by sour stimuli, indicates that K+ current modulation is the mechanism of sour taste transduction.

1. Two different types of acetylcholine receptor channels can be detected on the post-junctional membrane of slow muscle fibres in garter snakes. Here they are designated T-type and S-type channels. Only T-type channels can also be found at twitch fibre neuromuscular junctions. 2. The physiological properties of slow fibre T-type channels are similar to those of acetylcholine receptor channels in end-plates of twitch fibres in these animals. 3. S-type channels had a smaller conductance than T-type channels (32 vs. 49 pS), but a similar reversal potential near 0 mV. 4. Both S- and T-type channels were found together in most patches of slow fibre end-plate membrane, but some patches displayed just one type or the other. 5. The activity of both S- and T-type channels desensitized in the presence of micromolar concentrations of acetylcholine. S-type channels desensitized less rapidly and less completely than did T-type channels. 6. Desensitized channels of both types recovered and produced bursts of activity, then went silent again. During a burst, channels did not appear to change type. 7. The activation of channels of either type was not correlated with activity in channels of the other type. 8. The open-duration distribution of S-type channels required two exponential components to be well fitted, with time constants in the range of 1-2 ms and 3-10 ms. In contrast, the open-duration distribution of T-type channels was a single exponential with a time constant similar in magnitude to the slower S-type component. 9. Desensitization-resistant S-type acetylcholine receptor channels could allow slow muscle end-plates to retain their sensitivity to acetylcholine during periods of heavy use. Under non-desensitizing conditions, differences in the decay properties of slow fibre end-plate currents compared to those in twitch fibres can be explained by the activation kinetics of S-type channels.

1. Frog neuromuscular junctions were viewed with Nomarski optics and voltage clamped. Agonist was applied ionophoretically and agonist concentrations were measured using a micro-electrode sensitive to quaternary amines. 2. The dose-response relationship was studied using the agonists carbamylcholine, suberyldicholine and hydroxyphenyl-propyltrimethylammonium. 3. With all of these agonists, it appeared that the ACh receptor could be active when either one or two agonist binding sites were occupied. The receptor was much more likely to be active when both sites were occupied. Agonist dissociation constants and receptor activation probabilities were estimated by non-linear regression techniques for several possible receptor activation schemes.

1. Synaptic responses to acetylcholine were recorded from garter snake (sp. Thamnophis) neuromuscular junctions with the voltage clamp. 2. In the following respects the responses were identical in twitch and slow fibres: exponential miniature end-plate current decay, reversal potential (approximately -5 mV), permeant ionic species (Na+, K+, Ca2+, but not Cl-), two component miniature end-plate current decay in the presence of procaine, insensitivity to tetrodotoxin, alpha-bungarotoxin and leiurus toxin. 3. The responses differed in several important ways: miniature end-plate current decay rate was only half as sensitive to voltage at slow fibre end-plates as at twitch; while all twitch fibre current fluctuation spectra were single component, about 60% of the slow fibre spectra were not. The latter could be fitted with two Lorentzian components. In all cases the mean end-plate current was proportional to total induced noise variance. 4. Two mechanisms which might account for the response differences were identified and tested in a fashion independent of specific molecular kinetics. The different responses do not arise from separate ion-selective channels for the permeant ions at slow fibre end-plates and seem unlikely to derive from separate populations of synaptic and extrasynaptic channels.

Olfactory receptor neurons isolated from embryonic, neonatal, and adult mice were studied using the patch-clamp technique. Several distinct types of ion channels were characterized in patches of membrane from the neuronal soma and the dendritic knob of receptor neurons, including a 130-pS Ca++-activated K+ channel with voltage-dependent kinetics, an 80-pS Ca++-activated K+ channel with voltage-insensitive kinetics, a 25-pS K+ channel with properties similar to inward rectifiers, and a 40-pS K+ channel that was activated and then inactivated by rapid depolarization. Evidence of large-conductance (greater than 200 pS) Cl- channels, which were Ca++ insensitive and increasingly active at depolarizing membrane potentials, and voltage-activated Ca++ channels (16 pS) was also obtained. From K+ channel activity recorded from cell-attached patches, the intracellular [Ca++] was inferred to be below 0.1 microM, and the membrane potential was inferred to be approximately -50 mV. The receptor neurons had high input resistances, and action potentials could be elicited by picoampere amounts of depolarizing current. The receptor neurons responded to applied odorant molecules and to forskolin with increases in membrane conductance. These results provide a description of the membrane properties of olfactory receptor neurons and a basis for understanding their electrical activity and response to odorants.