Single channel kinetics of a glutamate receptor

Aggregated Markov processes incorporating time interval omission

We consider a finite-state-space, continuous-time Markov chain which is time reversible. The state space is partitioned into two sets, termed ‘open' and ‘closed', and it is only possible to observe which set the process is in. Further, short sojourns in either the open or closed sets of states will fail to be detected. We show that the dynamic stochastic properties of the observed process are completely described by an embedded Markov renewal process. The parameters of this Markov renewal process are obtained, allowing us to derive expressions for the moments and autocorrelation functions of successive sojourns in both the open and closed states. We illustrate the theory with a numerical study.

Aggregated Markov processes incorporating time interval omission

We consider a finite-state-space, continuous-time Markov chain which is time reversible. The state space is partitioned into two sets, termed ‘open' and ‘closed', and it is only possible to observe which set the process is in. Further, short sojourns in either the open or closed sets of states will fail to be detected. We show that the dynamic stochastic properties of the observed process are completely described by an embedded Markov renewal process. The parameters of this Markov renewal process are obtained, allowing us to derive expressions for the moments and autocorrelation functions of successive sojourns in both the open and closed states. We illustrate the theory with a numerical study.

Aggregated Markov processes with negative exponential time interval omission

We consider a time reversible, continuous time Markov chain on a finite state space. The state space is partitioned into two sets, termed open and closed, and it is only possible to observe whether the process is in an open or a closed state. Further, short sojourns in either the open or closed states fail to be detected. We consider the situation when the length of minimal detectable sojourns follows a negative exponential distribution with mean μ–1. We show that the probability density function of observed open sojourns takes the form , where n is the size of the state space. We present a thorough asymptotic analysis of f O(t) as μ tends to infinity. We discuss the relevance of our results to the modelling of single channel records. We illustrate the theory with a numerical example.

Aggregated Markov processes with negative exponential time interval omission

We consider a time reversible, continuous time Markov chain on a finite state space. The state space is partitioned into two sets, termed open and closed, and it is only possible to observe whether the process is in an open or a closed state. Further, short sojourns in either the open or closed states fail to be detected. We consider the situation when the length of minimal detectable sojourns follows a negative exponential distribution with mean μ–1. We show that the probability density function of observed open sojourns takes the form, wherenis the size of the state space. We present a thorough asymptotic analysis offO(t) as μ tends to infinity. We discuss the relevance of our results to the modelling of single channel records. We illustrate the theory with a numerical example.

Glutamate receptor channel kinetics: the effect of glutamate concentration

Single channel recordings from the locust muscle D-glutamate receptor channel were obtained using glutamate concentrations ranging from 10(-6) to 10(-2) M. Channel kinetics were analyzed to aid in the development of a model for the gating mechanism. Analysis of channel dwell time histograms demonstrated that the channel possessed multiple open and closed states at concentrations of glutamate between 10(-5) and 10(-2) M. Correlations between successive dwell times showed that the gating mechanism was nonlinear (i.e., branched or cyclic) over the same glutamate concentration range. The glutamate concentration dependence of the channel open probability, and of the event frequency, was used to explore two possible allosteric gating mechanisms in more detail.

Cys-Loop Ligand-Gated Chloride Channels in Dorsal Unpaired Median Neurons of Locusta migratoria

In insects, inhibitory neurotransmission is generally associated with members of the cys-loop ligand-gated anion channels, such as the glutamate-gated chloride channel (GluCl), the GABA-gated chloride channels (GABACl), and the histamine-gated chloride channels (HisCl). These ionotropic receptors are considered established target sites for the development of insecticides, and therefore it is necessary to obtain a better insight in their distribution, structure, and functional properties. Here, by combining electrophysiology and molecular biology techniques, we identified and characterized GluCl, GABACl, and HisCl in dorsal unpaired median (DUM) neurons of Locust migratoria. In whole cell patch-clamp recordings, application of glutamate, GABA, or histamine induced rapidly activating ionic currents. GluCls were sensitive to ibotenic acid and blocked by picrotoxin and fipronil. The pharmacological profile of the L. migratoria GABACl fitted neither the vertebrate GABAA nor GABAC receptor and was similar to the properties of the cloned Drosophila melanogaster GABA receptor subunit (Rdl). The expression of Rdl-like subunit-containing GABA receptors was shown at the molecular level using RT-PCR. Sequencing analysis indicated that the orthologous GABACl of D. melanogaster CG10357-A is expressed in DUM neurons of L. migratoria. Histamine-induced currents exhibited a fast onset and desensitized completely on continuous application of histamine. In conclusion, within the DUM neurons of L. migratoria, we identified three different cys-loop ligand-gated anion channels that use GABA, glutamate, or histamine as their neurotransmitter.

The action of philanthotoxin-343 and photolabile analogues on locust (Schistocerca gregaria) muscle

The effects of philanthotoxin-343 (PhTX-343; tyrosyl-butanoyl-spermine) and photolabile analogues of this synthetic toxin on locust (Schistocerca gregaria) skeletal muscle have been investigated using whole muscle preparations (twitch contractions), single muscle fibres (excitatory postsynaptic currents (EPSCs)) and muscle membrane patches containing single quisqualate-sensitive glutamate receptors (qGluR). Analogues containing an azido group attached to either the butanoyl side-chain of PhTX-343 or as a substitute for the hydroxyl moiety of the tyrosyl residue were about 6 fold more potent antagonists than PhTX-343; those with an azido group located at the distal end of the toxin molecule were generally 2-3 fold less potent than PhTX-343. When these compounds were tested in subdued light, they were reversible antagonists of the muscle twitch, EPSC and qGluR. When a muscle was irradiated with U.V. during application of photolabile toxin combined with either neural stimulation of the muscle or L-glutamate application, antagonism of the twitch, EPSC and qGluR was complete and irreversible.

Burst kinetics of single NMDA receptor currents in cell-attached patches from rat brain cortical neurons in culture

1. The patch-clamp technique was used to record single-channel currents from cell-attached patches on rat brain cortical neurons in culture. The composition of the open and shut intervals during bursts of openings was studied in N-methyl-D-aspartate (NMDA) receptors exposed to 1 microM NMDA and 10 microM glycine at a membrane potential of -70 mV. 2. Open interval histograms were constructed for openings at each position (first, second, third, etc.) during all bursts. Distributions from openings two to five were fitted with similar (two or three) exponential components. The first opening from all bursts tended to be of shorter duration than the other openings. 3. Bursts were sorted according to the number of openings they contained and duration histograms were obtained for the openings at each position (one to five) during bursts of up to five openings. For bursts containing two or more openings, the distribution of open durations obtained at a given position were similar to each other regardless of the number of openings in the burst. 4. In bursts of two or more openings (up to five), duration histograms from the openings at each position in the burst were fitted with two or three exponential components that were similar for each opening. Bursts consisting of a single opening had a different distribution, having a relatively larger component of short duration. 5. Shut intervals during bursts were described by two exponential components with average time constants (and relative areas) (means +/- S.E.M.) of 0.06 +/- 0.01 ms (0.59 +/- 0.02) and 0.64 +/- 0.02 ms (0.41 +/- 0.02). Their distribution was independent of the numbers of openings in the bursts, their position within the burst, and the types of openings (long or short duration) contained within the burst. 6. These results suggest that each opening in bursts of two or more openings is kinetically similar to every other opening regardless of burst length. Analogously, each shut period during a burst was similar to every other. A kinetic model with three open and four closed intraburst states is shown to be consistent with these results for bursts of two or more openings.

The effects of nitrous oxide on a glutamate-gated ion channel and their reversal by high pressure; a single channel analysis

Nitrous oxide reversibly affects the kinetics, but not the conductance, of the qGluR channel of locust muscle. 0.5 atm N2O at 20.5 degrees C was without effect but both 1.5 and 2.7 atm significantly reduced the probability of the channel opening, the frequency of opening and the mean open time, and prolonged the mean closed time. 100 atm helium was without effect on these parameters, but when 98.5 atm He was combined with 1.5 atm N2O they, and the associated dwell time distributions, were restored to normal. 100 atm similarly combined with 2.7 atm N2O exerted a comparable trend which fell short of significance. The results are consistent with nitrous oxide binding to the channel with a significant molar volume increase, which pressure opposes. This suggests that nitrous oxide may cause conformational changes in the channel, and that the pressure reversal of nitrous oxide anaesthesia in animals could be caused by molecular antagonism.

Staphylococcus aureus alpha-toxin-induced pores: channel-like behavior in lipid bilayers and patch clamped cells

The conductance of pores induced by Staphylococcus aureus alpha-toxin in Lettre cells has been compared to that in bilayers composed of synthetic lipids or Lettre cell membrane constituents. Previously described characteristics of toxin-induced conductance changes in lipid bilayers, namely rectification, voltage-dependent closure, and closure at low pH or in the presence of divalent cations (Menestrina, 1986) are displayed also in bilayers prepared from Lettre cell membranes and in patch clamped Lettre cells. It is concluded that endogenous proteins do not affect the properties of alpha-toxin-induced channels significantly and that the relative lack of ion channels in Lettre cells makes them ideal for studies of pore-forming toxins by the patch clamp technique.

Gating of the native and purified cardiac SR Ca(2+)-release channel with monovalent cations as permeant species

The primary aim of this study was to characterize the steady-state gating of the native and the purified cardiac sarcoplasmic reticulum Ca(2+)-release channel using monovalent cations (K+ in the purified, Cs+ in the native) rather than Ca2+ as the permeant ions. The improved resolution of the single-channel events under these conditions has provided a more detailed and accurate description of channel gating than was previously possible. Micromolar cytosolic Ca2+ activates the channel but in the absence of other activating ligands cannot fully open the channel. The relationship between the open probability (Po) and cytosolic free [Ca2+] in both native and purified channels indicates the binding of at least three Ca2+ ions for maximal activation. Lifetime analysis indicates a minimum of three open and five closed states for channels activated solely by Ca2+ and demonstrates that the primary mechanism for the increase in Po is an increase in the frequency of channel opening. Burst analysis also indicates that Ca2+ activates the channel by binding to closed states of the channel to increase the frequency of channel opening. Correlations between successive lifetimes suggest the existence of at least two pathways between the open and closed states. At a given activating [Ca2+], the Po is lower at negative than at positive holding potentials; however, we find no change in the mechanisms of Ca2+ activation at different voltages. Po measurements and lifetime analysis indicate that the gating of the purified channel when activated by Ca2+ is indistinguishable from that of the native channel and indicate that the channels are not modified by the purification procedure.

Gating kinetics of the quisqualate-sensitive glutamate receptor of locust muscle studied using agonist concentration jumps and computer simulations

Outside-out patches excised from extrajunctional membrane of locust muscle were subjected to "concentration jumps" of L-glutamate, using the liquid filament switch technique, to study channel opening and closing rates, desensitization onset, and recovery from desensitization of a quisqualate-sensitive glutamate receptor (qGluR). Based on data obtained from these experimental studies, computer modeling techniques have been used in an attempt to simulate the behavior of qGluR during a concentration jump of L-glutamate. A linear model with three closed states (one unliganded, one monoliganded, and one biliganded), one open state (binding two molecules of L-glutamate), and two desensitization states (the one monoliganded, the other biliganded) leading from the unliganded closed state simulated all of the experimentally observed behavior. The results are discussed in the context of previous equilibrium studies in which desensitization was inhibited with concanavalin A and for which a ten-state model was required to simulate the behavior of qGluR.

Effects of high pressure on the channel gated by the quisqualate-sensitive glutamate receptor of locust muscle and its blockade by ketamine; a single-channel analysis

The effects of high pressure on the channel gating kinetics of the quisqualate-sensitive L-glutamate receptor (qGluR) of locust muscle have been investigated using a megaohm seal patch-clamp technique. Pressure was applied with helium gas and recordings were carried out at 20.5 degrees C with Rb+ as the main charge-carrying cation in the patch pipette. The mean open time of the qGluR channel was unaffected by 10 and 30 MPa, but it was significantly reduced at 50 MPa. A high proportion of brief openings (mean 0.808 ms) was seen at 50 MPa but not at lesser pressures. Also, in contrast to lesser pressures, 50 MPa prolonged the mean closed time and reduced both the frequency and probability of channel opening. 10(-6) M ketamine significantly reduced the mean channel open time, as previously reported. A pressure of 10 MPa which alone had no effect on the qGluR channel, restored the mean open time in the presence of 10(-6) M ketamine to the value obtained in the absence of the anaesthetic. This implies the shortening of qGluR channel open time by ketamine involves a large + delta V and, therefore, probably conformational changes in the channel. However 10 MPa did not restore the distribution of open times to normal.

Single ion channel models incorporating aggregation and time interval omission

We present a general theoretical framework, incorporating both aggregation of states into classes and time interval omission, for stochastic modeling of the dynamic aspects of single channel behavior. Our semi-Markov models subsume the standard continuous-time Markov models, diffusion models and fractal models. In particular our models allow for quite general distributions of state sojourn times and arbitrary correlations between successive sojourn times. Another key feature is the invariance of our framework with respect to time interval omission: that is, properties of the aggregated process incorporating time interval omission can be derived directly from corresponding properties of the process without it. Even in the special case when the underlying process is Markov, this leads to considerable clarification of the effects of time interval omission. Among the properties considered are equilibrium behavior, sojourn time distributions and their moments, and auto-correlation and cross-correlation functions. The theory is motivated by ion channel mechanisms drawn from the literature, and illustrated by numerical examples based on these.

Stochastic models for ion channels: introduction and bibliography

This paper provides an introduction to and overview of the use of stochastic models and statistical analysis in the study of ion channels in cell membranes. An extensive bibliography is included.

Single channel analysis of ketamine interaction with a quisqualate receptor

The interactions of the general anaesthetic ketamine with the quisqualate-sensitive L-glutamate receptor (QUIS-R) of locust muscle have been investigated at the single channel level using a M omega seal patch clamp technique. Low concentrations (10(-10) to 10(-9) M) of ketamine did not significantly alter the kinetics of the QUIS-R channel. Higher concentrations of ketamine decreased the probability of the channel being open, the frequency of channel opening and the channel mean open time, and increased the channel mean closed time. Probability density functions of channel dwell times indicate that during application of greater than 10(-8) M ketamine the distribution of channel openings becomes restricted mainly to brief events. These results are consistent with the view that ketamine blocks the open, and possibly also the closed, channel of locust muscle QUIS-R and that this anaesthetic dissociates only slowly from its blocking site(s).

Quisqualate-sensitive glutamate receptors of the locust Schistocerca gregaria are antagonised by intracellularly applied philanthotoxin and spermine

The effects of intracellularly and extracellularly applied synthetic analogues of delta-philanthotoxin (PhTX-433) and the polyamine spermine on the excitatory postsynaptic current (EPSC) of glutamatergic synapses and single channel currents gated by quisqualate-sensitive glutamate receptors (QUIS-R) on locust leg muscle have been compared. When applied extracellularly all 3 compounds reversibly antagonised the EPSC and the single channel currents. Antagonism was voltage independent, but use (agonist) dependent. Antagonism also occurred when they were injected into muscle fibres, but in this case it was not use dependent. It is proposed that spermine and the two toxins bind to the closed and open channel conformations of QUIS-R at a site near the intracellular opening of the channel gated by this receptor.

Molecular cloning of an invertebrate glutamate receptor subunit expressed in Drosophila muscle

Insects and other invertebrates use glutamate as a neurotransmitter in the central nervous system and at the neuromuscular junction. A complementary DNA from Drosophila melanogaster, designated DGluR-II, has been isolated that encodes a distant homolog of the cloned mammalian ionotropic glutamate receptor family and is expressed in somatic muscle tissue of Drosophila embryos. Electrophysiological recordings made in Xenopus oocytes that express DGluR-II revealed depolarizing responses to L-glutamate and L-aspartate but low sensitivity to quisqualate, alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionate (AMPA), and kainate. The DGluR-II protein may represent a distinct glutamate receptor subtype, which shares its structural design with other members of the ionotropic glutamate receptor family.

Sheep cardiac sarcoplasmic reticulum calcium-release channels: modification of conductance and gating by temperature

1. The gating and conduction properties of single calcium-release channels of sheep isolated cardiac junctional sarcoplasmic reticulum membranes incorporated into planar phospholipid bilayers were investigated under voltage clamp conditions at temperatures between 4 and 32 degrees C. 2. Single channel conductance was reduced linearly when temperature was decreased from 32 to 5 degrees C with a Q10 value of 1.5 between 10 and 20 degrees C. The apparent activation enthalpy for conductance between 32 and 5 degrees C was 6.16 +/- 1.2 kcal/mol. 3. Cooling the channel increased open probability (Po) when activating cytosolic calcium concentrations were varied within the range 0.1-100 microM. At an activating free calcium concentration of 10 microM, channel Po increased from 0.13 +/- 0.05 at 23 degrees C to 0.69 +/- 0.07 at 5-10 degrees C. 4. At sub-activating calcium concentrations (100 pM) or high concentrations of calcium (greater than or equal to 1000 microM), the calcium-release channel remained closed at 23 degrees C. Cooling the channel under these conditions did not increase Po. 5. Lifetime analysis indicates that with calcium as the sole activating ligand, the cooling-induced increase in Po results from an increase in channel open lifetimes with no significant alteration in the frequency of channel opening. At 23 degrees C, the open and closed lifetime distributions of the calcium-activated channel are best described by two and three exponentials respectively. At reduced temperatures (5-10 degrees C), both open and closed lifetime distributions were best described by three exponentials. 6. At sub-activating calcium concentrations, calcium-independent channel openings could be induced by sulmazole (AR-L 115 BS, 0.5-10 mM). At 23 degrees C, with sulmazole as the sole activating ligand, the best fits to both open and closed lifetime distributions were obtained with three exponentials. At reduced temperatures (5-10 degrees C), Po was increased. Gating was characterized by long open events, however the open and closed lifetime distributions were still best described by three exponentials. 7. The net effect of temperature reduction is an increase in calcium current through the channel. This finding is consistent with the suggestion that calcium release from the SR is a major factor in the initiation of rapid cooling contractures of mammalian cardiac muscle preparations.

Glutamatergic channels in locust muscle show a wide time range of desensitization and resensitization characteristics

Outside-out patches of membrane were excised from locust muscle. 10 mM L-glutamate applied to such patches in short pulses elicited rapidly a peak of channel opening, followed by desensitization with time constants between 1 ms and 20 ms for different channel subtypes. Slowly rising glutamate concentrations could produce almost complete desensitization without any channel openings. Desensitization thus proceeds from a closed channel state, prior to channel opening. Resensitization, i.e. recovery from desensitization, was tested by applying pairs of glutamate pulses, varying the pulse interval. Resensitization has a biphasic time course. While in rapidly desensitizing channels more than 50% of resensitization was achieved within a few milliseconds, in slowly desensitizing channels resensitization took seconds.

Glutamate receptor-channel gating. Maximum likelihood analysis of gigaohm seal recordings from locust muscle

Gigaohm recordings have been made from glutamate receptor channels in excised, outside-out patches of collagenase-treated locust muscle membrane. The channels in the excised patches exhibit the kinetic state switching first seen in megaohm recordings from intact muscle fibers. Analysis of channel dwell time distributions reveals that the gating mechanism contains at least four open states and at least four closed states. Dwell time autocorrelation function analysis shows that there are at least three gateways linking the open states of the channel with the closed states. A maximum likelihood procedure has been used to fit six different gating models to the single channel data. Of these models, a cooperative model yields the best fit, and accurately predicts most features of the observed channel gating kinetics.

Divalent cation activation and inhibition of single calcium release channels from sheep cardiac sarcoplasmic reticulum

Single Ca2+ release channels from vesicles of sheep cardiac junctional sarcoplasmic reticulum have been incorporated into uncharged planar lipid bilayers. Single-channel currents were recorded from Ca2(+)-activated channels that had a Ca2+ conductance of approximately 90 pS. Channel open probability increased sublinearly as the concentration of free Ca2+ was raised at the myoplasmic face, and without additional agonists the channels could not be fully activated even by 100 microM free Ca2+. Lifetime analysis revealed a minimum of two open and three closed states, and indicates that Ca2+ activated the channels by interacting with at least one of the closed states to increase the rate of channel opening. Correlations between adjacent lifetimes suggested there were at least two pathways between the open- and closed-state aggregates. An analysis of bursting behavior also revealed correlations between successive burst lengths and the number of openings per burst. The latter had two geometric components, providing additional evidence for at least two open states. One component appeared to comprise unit bursts, and the lifetime of most of these fell within the dominant shorter open-time distribution associated with over 90% of all openings. A cyclic gating scheme is proposed, with channel activation regulated by the binding of Ca2+ to a closed conformation of the channel protein. Mg2+ may inhibit activation by competing for this binding site, but lifetime and fluctuation analysis suggested that once activated the channels continue to gate normally.

Intraburst kinetic properties of the GABAA receptor main conductance state of mouse spinal cord neurones in culture

1. The intraburst kinetic properties of the main conductance state of gamma-aminobutyric acidA (GABAA) receptor channels in excised outside-out patches obtained from somata of mouse spinal cord neurones in cell culture were investigated using the patch clamp single-channel recording technique. 2. At 2 microM-GABA, the burst duration distribution was fitted by four exponential functions with time constants of 0.5, 2.4, 8.3 and 31.8 ms. 3. At 0.5, 1 and 2 microM-GABA, frequency distribution histograms of the number of apparent openings per burst were best fitted by three geometric functions with similar mean numbers (1.1, 1.9 and 3.6) of openings per burst. The proportion of bursts with a mean of 1.1 openings per burst decreased with increased GABA concentration while the proportion of bursts with means of 1.9 and 3.6 openings per burst increased with GABA concentration. 4. Analyses of GABA receptor channel intraburst kinetics were performed at all three GABA concentrations. The results were similar for all concentrations, but detailed results are presented only for 2 microM-GABA. 5. The open time distribution for all intraburst openings was best fitted by three exponential functions with time constants of 0.6, 2.9 and 8.9 ms. 6. Intraburst open time and total open time distributions for bursts with one to five openings were fitted with two or three exponential functions or gamma distributions, respectively. The number of components, time constants and relative areas were similar for both distributions. 7. The distributions of open times for the nth opening within bursts of k openings were similar for bursts containing two to five openings. The distributions of open times for the nth opening of all bursts varied with position within the burst. The distributions shifted to longer openings as the opening number increased from one to five. 8. The distributions of all closings within all bursts or within bursts with two to five openings and of closings relative to position in all bursts could be fitted by two exponential functions with time constants of about 0.20 and 3.1 ms and relative proportions of 0.55 and 0.45 at all GABA concentrations. 9. The total closed time distributions for bursts containing two to four closings, however, were all best fitted with only a single gamma distribution with a time constant of 1.3 ms.(ABSTRACT TRUNCATED AT 400 WORDS)

Rapid activation, desensitization, and resensitization of synaptic channels of crayfish muscle after glutamate pulses

Completely desensitizing excitatory channels were activated in outside-out patches of crayfish muscle membrane by applying glutamate pulses with switching times of approximately 0.2 ms for concentration changes. Channels were almost completely activated with 10 mM glutamate. Maximum activation was reached within 0.4 ms with greater than or equal to 1 mM glutamate. Channel open probability decayed with a time constant of desensitization of 2 ms with 10 mM glutamate and more rapidly at lower glutamate concentrations. The rate of beginnings of bursts (average number of beginnings of bursts per time bin) decayed even faster but approximately in proportion to the glutamate concentration. The dose-response curve for the channel open probability and for the rate of bursts had a maximum double-logarithmic slope of 5.1 and 4.2, respectively. Channels desensitized completely without opening at very low or slowly rising glutamate concentrations. Desensitization thus originates from a closed channel state. Resensitization was tested by pairs of completely desensitizing glutamate pulses. Sensitivity to the second pulse returned rapidly at pulse intervals between 1 and 2 ms and was almost complete with an interval of 3 ms. Schemes of channel activation by up to five glutamate binding steps, with desensitization by glutamate binding from closed states, are discussed. At high agonist concentrations bursts are predominantly terminated by desensitization. Quantal currents are generated by pulses of greater than 1 mM glutamate, and their decay is determined by the duration of presence of glutamate and possibly by desensitization.

Markov, fractal, diffusion, and related models of ion channel gating. A comparison with experimental data from two ion channels

The gating kinetics of single-ion channels are generally modeled in terms of Markov processes with relatively small numbers of channel states. More recently, fractal (Liebovitch et al. 1987. Math. Biosci. 84:37-68) and diffusion (Millhauser et al. 1988. Proc. Natl. Acad. Sci. USA. 85:1502-1507) models of channel gating have been proposed. These models propose the existence of many similar conformational substrates of the channel protein, all of which contribute to the observed gating kinetics. It is important to determine whether or not Markov models provide the most accurate description of channel kinetics if progress is to be made in understanding the molecular events of channel gating. In this study six alternative classes of gating model are tested against experimental single-channel data. The single-channel data employed are from (a) delayed rectifier K+ channels of NG 108-15 cells and (b) locust muscle glutamate receptor channels. The models tested are (a) Markov, (b) fractal, (c) one-dimensional diffusion, (d) three-dimensional diffusion, (e) stretched exponential, and (f) expo-exponential. The models are compared by fitting the predicted distributions of channel open and closed times to those observed experimentally. The models are ranked in order of goodness-of-fit using a boot-strap resampling procedure. The results suggest that Markov models provide a markedly better description of the observed open and closed time distributions for both types of channel. This provides justification for the continued use of Markov models to explore channel gating mechanisms.

Argiopine blocks glutamate-activated single-channel currents on crayfish muscle by two mechanisms

1. The effect of the spider venom argiopine on L-glutamate-activated membrane channels of crayfish muscle was investigated using the patch-clamp technique. 2. When 10(-2) M-glutamate and 10(-9) M-argiopine were contained in the pipette solution of a cell-attached patch, bursts of openings of excitatory channels appeared after formation of the patch. These bursts ceased abruptly after variable periods of time in the range of 5 min. Higher concentrations of argiopine (up to 10(-6) M) blocked more rapidly, approximately in proportion to concentration. 3. The block of excitatory channels could be partially or completely reversed by hyperpolarizing the membrane by up to -190 mV from the resting potential. The time constant of the recovery of channel opening decreased with increasing hyperpolarization and was 2 ms with -160 mV hyperpolarization. Switching back from the hyperpolarized level to the resting potential, the time constant for the resulting block was about 3 s (10(-7) M-argiopine). Potential-dependent block by argiopine with similar characteristics was also observed in outside-out patches. 4. Up to argiopine concentrations of 10(-7) M the kinetics of channel openings and of bursts measured in pre-block periods or during reversal of the block by hyperpolarization were indistinguishable from controls. 5. When the potential-dependent block observed in the presence of 10(-6) M-argiopine and 10(-2) M-glutamate was reversed by hyperpolarization, additional short closings occurred during bursts. This 'flickering block' did not change burst length appreciably, but an additional open time component (tau = 0.1 ms) appeared and the average open time per burst was reduced. 6. At least two reaction steps seem necessary to model the behaviour of the potential-dependent block. The flickering block may be described as intermittent blocking of the channel which does not interfere with the reactions between glutamate and the channel.

A note on single-channel autocorrelation functions

It is shown for time-reversible ion channel gating mechanisms that the sojourn time autocorrelation functions are necessarily nonnegative, decreasing, and convex. It is also shown, again for time-reversible mechanisms, that the lagged moments of all orders are determined by the first three moments. The application of these results to the statistical analysis of single-channel kinetics is briefly discussed.

Chloride channels gated by extrajunctional glutamate receptors (H-receptors) on locust leg muscle

Outside-out patches of extrasynaptic membrane were isolated from leg muscles of locusts. L-Glutamate and its agonists were applied to such patches either continuously or in rapidly switched pulses. When the pipette contained a high chloride concentration, 2.5 x 10(-5) M glutamate triggered single-channel currents (gated by H-receptors) with a conductance of 25 pS which were carried by chloride, in addition to cationic channels (gated by D-receptors). For the chloride channels, the distribution of channel open times had components of about 2 and 12 ms. Pulses of higher glutamate concentrations elicited many superimposed channel openings, and the approximately saturating concentration of 10(-3) M glutamate opened 100-200 channels simultaneously. When the pipette contained low chloride, channel conductance was reduced, and the current voltage relation was shifted towards the now negative chloride equilibrium potential. H-Receptor-gated chloride channels were activated by glutamate, ibotenate and aspartate, but not by GABA, quisqualate, kainate, N-methyl-D-aspartate and carbachol. The currents declined in the continued presence of agonist showing a time constant of desensitization greater than 1 s. Recovery from desensitization after removal of the agonist was tested with double pulses and was found to have a time constant of about 300 ms.

Adjacent interval analysis distinguishes among gating mechanisms for the fast chloride channel from rat skeletal muscle

1. The durations of adjacent open and shut intervals, obtained with the patch-clamp technique from fast Cl- channels in tissue-cultured rat skeletal muscle, were analysed to distinguish among eight previously considered gating mechanisms for the channel which differed in the connections among the states. 2. Open intervals were separated into groups based on the duration of the shut intervals which occurred before or after each open interval. Fitting these conditional open distributions with sums of exponentials indicated that they were described by two exponential components. 3. The time constants of the two components in the conditional open distributions were independent of the adjacent shut interval durations. The observation of invariant time constants is consistent with gating mechanisms in which the rate constants for transitions among the states remain constant with time (discrete Markov process). 4. In contrast to the invariant time constants, the areas of the two components in the conditional open distributions were dependent on the adjacent shut interval durations. The area of the fast open component increased, and the area of the slow open component decreased, as the duration of adjacent shut intervals increased. Thus, it is changes in areas, rather than time constants, which give rise to the observed inverse relationship between the durations of adjacent open and shut intervals. 5. The findings in summary statements 2-4 indicate that at least two open states are connected by independent pathways to different shut states; the open state associated with the fast open component is connected to a shut state (or compound shut state) of longer effective lifetime, and the open state associated with the slow open component is connected to a shut state (or compound shut state) of briefer effective lifetime. 6. Seven of the eight previously considered gating mechanisms were rejected because they did not account for the observed relationships between the durations of adjacent open and shut intervals, when analysed in terms of either conditional open distributions or conditional mean open interval durations. 7. The seven rejected gating mechanisms also did not account for the observed correlations between interval durations, when analysed in terms of correlation coefficients. Adjacent interval and correlation analysis thus provided a means to distinguish among the gating mechanisms.(ABSTRACT TRUNCATED AT 400 WORDS)

Analysis of post-perturbation gating kinetics of single ion channels

Analysis of mean dwell-times as a function of the number of channel openings elapsed since a stepwise perturbation in ion-channel kinetics is shown to provide information concerning the topology of the underlying gating mechanism. The difference between the post-perturbation mean dwell-time and the corresponding equilibrium mean is shown to decay as the sum of Ng-1 geometric terms in k, the number of openings since the perturbation, where Ng is the minimum number of gateway states in the channel gating mechanism. The method is illustrated by consideration of various simple gating schemes. A modification of the method accommodating the presence of channel inactivation or desensitization is described. Application of the method to a delayed-rectifier type K+ channel of NG108-15 cells reveals that Ng greater than or equal to 2, consistent with a branched gating mechanism.

Single channel studies of non-competitive antagonism of a quisqualate-sensitive glutamate receptor by argiotoxin636--a fraction isolated from orb-web spider venom

The effects of purified spider toxin (argiotoxin636) on single glutamate-activated channels in voltage-clamped locust muscle fibres have been examined using a megaohm seal, patch-clamp technique. Four experimental protocols were employed in which the composition of the patch pipette and bathing solutions were varied. Three types of channel behaviour were broadly defined when argiotoxin636 was present either in the patch pipette or in the muscle bath; the type of channel behaviour being dependent upon the concentration of argiotoxin636 and/or the duration of its application. Type I behaviour was characterized by reductions in channel open probability (Po) and channel event frequency (f), by an increase in mean channel closed time (mc) and either no change in mean channel open time (mo) or, infrequently, an increase in this parameter; Type II behaviour was characterized by apparent absence of channel openings. For example, with 10(-12) M argiotoxin636 in the patch pipette Type I behaviour changed to Type II behaviour after approximately 60 s and from Type II behaviour to Type III behaviour after approximately 120 s. The results of this study are consistent with the idea that argiotoxin636 blocks the cation-selective channel gated by excitatory glutamate receptors in insect muscle at the level of the open channel although there remains the possibility that it is also either a closed channel blocker and/or a competitive antagonist. The increase in mo seen in a few recordings during the initial stage of argiotoxin636 antagonism raises the possibility that the toxin interacts allosterically with the glutamate binding sites on the excitatory glutamate receptor.

The use of dwell time cross-correlation functions to study single-ion channel gating kinetics

The derivation of cross-correlation functions from single-channel dwell (open and closed) times is described. Simulation of single-channel data for simple gating models, alongside theoretical treatment, is used to demonstrate the relationship of cross-correlation functions to underlying gating mechanisms. It is shown that time irreversibility of gating kinetics may be revealed in cross-correlation functions. Application of cross-correlation function analysis to data derived from the locust muscle glutamate receptor-channel provides evidence for multiple gateway states and time reversibility of gating. A model for the gating of this channel is used to show the effect of omission of brief channel events on cross-correlation functions.

Properties of ion channels formed by Staphylococcus aureus delta-toxin

The delta-toxin of Staphylococcus aureus has been investigated in terms of its potential to form ion channels in planar lipid bilayers formed at the tip of patch electrodes. Channel formation has been shown to occur for delta-toxin concentrations in the range 0.1 to 2.0 microM. In 0.5 M KCl, two major classes of channels were seen--'small' with conductances of 70-100 pS, and 'large' with a conductance of approx. 450 pS. Current-voltage relationships for lipid bilayers containing several delta-toxin channels revealed both voltage-dependent and independent components to channel gating. Reversal potential measurements showed the channels to be cation selective. In the presence of 3.0 M KCl, the channel gating kinetics were complex, with multiple open and closed states. The results are interpreted in terms of a model for the channel consisting of a hexameric cluster of alpha-helical delta-toxin molecules.

Rapid activation and desensitization by glutamate of excitatory, cation-selective channels in locust muscle

Outside-out patches of membrane were excised from extensor tibiae muscles of locusts. L-Glutamate or its agonists were applied to such patches in short pulses by means of a lipid filament switch. Cationselective, excitatory channels were activated by quisqualate, L-glutamate and aspartate (in decreasing order of effectivity), but not by ibotenate, kainate, N-methyl-D-aspartate and glycine. At high agonist concentrations, channel activation reached a peak within 1 ms. Two kinetic types of channels have been identified: L-channels with on average relatively long and S-channels with short openings. Both types of channel openings showed surprisingly high rates of desensitization, channel activity declining after the initial surge to zero with time constants of about 25 and 3 ms, respectively. The L-channels exhibit open times close to those of channels recorded in M omega-seal studies. The S-channel has not been reported previously.

Single-channel autocorrelation functions: the effects of time interval omission

We present a general mathematical framework for analyzing the dynamic aspects of single channel kinetics incorporating time interval omission. An algorithm for computing model autocorrelation functions, incorporating time interval omission, is described. We show, under quite general conditions, that the form of these autocorrelations is identical to that which would be obtained if time interval omission was absent. We also show, again under quite general conditions, that zero correlations are necessarily a consequence of the underlying gating mechanism and not an artefact of time interval omission. The theory is illustrated by a numerical study of an allosteric model for the gating mechanism of the locust muscle glutamate receptor-channel.

Characterization of a delayed rectifier K+ channel in NG108-15 neuroblastoma X glioma cells: gating kinetics and the effects of enrichment of membrane phospholipids with arachidonic acid

A voltage-sensitive K+ channel with characteristics of the delayed rectifier was studied in NG108-15 cells using the cell-attached patch-clamp technique. The primary conductance of the channel was 18 pS, but occasional openings to a subconductance state were observed. The average latency to first opening of the channel was about 4 msec. Based on about 20,000 channel openings, the open time probability density function (pdf) required at least three exponentials (time constants of about 0.2, 3 and 9 msec) to achieve an adequate fit to the data. The closed time pdf required at least six exponentials to describe the data (time constants ranging from 0.093 to 440 msec). Thus, the channel exists in at least three open and six closed states. The ensemble average describing the inactivation of the channel was well fit by two exponentials with time constants of 170 msec and 4.2 sec. To examine the effect of changes in membrane lipid composition on the properties of the channel, the phospholipids of the cells were enriched with polyunsaturated fatty acids. In patches from 20:4-enriched cells the conductance, mean first latency, and open-time pdf were similar to control cells. However, the open state probability was increased from 0.25 to 0.44 and the mean closed time was decreased from 20 to 9 msec. The closed time pdf exhibited a higher proportion of closing events associated with short time constants, i.e., the probability of the channel closing into a long-lived closed state was decreased. The decay phase of the ensemble average also was changed; the proportion of the curve described by the slower time constant was almost doubled. Thus, the delayed rectifier from NG108-15 cells can exist in at least three open and six closed states, and changes in membrane lipid composition may have subtle effects on the gating kinetics of the channel.

Ionic permeabilities of L-glutamate activated, excitatory synaptic channel in crayfish muscle

Excitatory single channel currents triggered by L-glutamate were measured in outside-out excised patches of crayfish muscle membrane. If an 'intracellular' solution was present in the pipette and normal extracellular solution with added glutamate (10(-3) M) passed the outside of the patch, the single channel currents, i1, had amplitudes of -8 pA at a patch potential of -70 mV. If in the extracellular solution Na+ was replaced by Li+ or Ca2+, the amplitudes of single channel currents were reduced by about 30%. Only about 20% of the channel current amplitude remained on replacement of Na+ by choline. Replacement of Na+ reduced the variance of channel amplitude distributions to the level of the baseline. Presence of Na+ thus induces an additional variance of open channel current. When the proportions of Na+/choline were varied, the resulting channel currents could be separated in Na+, Ca2+ and choline components. The amplitude of the Na+ component, i1,Na, could be described by a constant channel permeability pi Na = 110 10(-15) cm3 s-1 according to the constant field equation. Ba2+ could replace Ca2+ without change in single channel current, while replacement of Ca2+ by Mg2+ reduced the channel currents by 20%. The following permeabilities of the single channel were estimated (in 10(-15) cm3 s-1): pi Na = 110, pi K = 86, pi Ca = 30, pi Mg = 24, pi Ba = 30, pi Li = 84 and pi choline = 11. These permeabilities were obtained inserting ionic concentrations. The respective permeabilities are listed also as calculated on the basis of ionic activities.(ABSTRACT TRUNCATED AT 250 WORDS)

Calcium dependent gating of the L-glutamate activated, excitatory synaptic channel on crayfish muscle

Excitatory, glutamate-activated single channel currents were measured in outside-out patches of crayfish muscle. The open times of single channel openings, and the durations and rates of bursts were evaluated. These kinetic parameters were not appreciably affected by replacement of extracellular Na+ by Li+ or choline. Changes in extracellular Ca2+ concentration Cao also did not influence the duration of single openings. However the mean burst duration decreased for Cao less than 13.5 mM and the rate of bursts declined with a power of almost 2 in low Cao. At Cao less than 1 mM practically no channel openings were observed in presence of glutamate. In order to exclude more rapid desensitization of the glutamate receptors in low Cao as the cause of disappearance of channel openings, glutamate was applied in short pulses with a liquid-filament switch. In 0 Cao also a glutamate pulse did not trigger channel openings. In presence of 13.5mM Cao, the inorganic Ca-channel blockers La3+ and Cd2+ diminished the duration and rate of bursts of channel openings in a similar manner as low Cao. The effects of low Cao and of Cd2+ were tested also on quantal postsynaptic currents, EPSCs, which were recorded through a perfused macro-patch-clamp electrode. At 1.4 mM Cao in the perfused electrode tip, spontaneous EPSCs were reduced at least by a factor of 4, and elicited EPSCs by a factor of 16. Application of Cd2+ had similarly strong effects on the EPSCs. Also the decay of EPSCs was shortened substantially in 1.4 mM Cao or 5 mM Cd2+. The inhibitory Cl(-)-channel of crayfish muscle, activated by glutamate or GABA, also was studied in outside-out patches. The openings of this channel persisted in 0 Cao solutions; the block of channel openings in low Cao thus is a specific property of the excitatory channel. The action of Cao on the excitatory channel may be described as that of a cofactor to glutamate. A possible reaction scheme is proposed.

Acquisition and analysis of fast single channel kinetic data on an Apple IIe microcomputer

We describe a method for acquisition and analysis of single channel kinetic data, that uses a ubiquitous microcomputer, the Apple IIe (and is thus cheap), and is yet sensitive and powerful (analysis of up to 6000 events per run of the program; event detection only limited by the amplifier bandwidth). For data acquisition, the original record is converted into an idealized record of open and closed times, using half amplitude threshold analysis or a two-cursor window discriminator. A Time Interval Meter (TIM) has been developed that measures the durations of the TTL-signal from the window discriminator and stores up to 256 intervals in a FIFO-buffer. Each cycle of the TIM is completed within 1 microsecond. The, largely Pascal, software reads the intervals from the FIFO-buffer and stores them in the main memory of the microcomputer (cycle time 70 microseconds). The package is completed with Pascal programs for statistical analysis, including histogram construction of open and closed times, non-linear fit of up to 3 exponentials and correlation analysis.