A theoretical experimental method to determine the locus of desensitization

Consider a ligand-gated channel with n agonist binding sites which can undergo desensitization. We present a theoretical experimental procedure for pinpointing the principal receptor state from which there is a transition to the desensitized state. The method is based on the observation that the dependence of the slope of the time constant of desensitization vs agonist concentration, at low concentrations, represents the state from which desensitization occurs. In those receptors where desensitization occurs from the open state (or the one immediately preceding it), the method also enables us to determine the number of binding sites.     

A molecular scheme for the reaction between acetylcholine and nicotinic channels.

In outside-out patches of mouse-muscle membrane, embryonic-like channels were activated by pulses of acetylcholine (ACh). On increasing the ACh concentration, the rate of desensitization, 1/tau d, increased linearly with the peak open probability, indicating desensitization from the open state. Desensitization had only one time constant tau d at each ACh concentration. Recovery from desensitization was only approximately 10 times slower than desensitization, whereas the probability of steady-state channel opening, declined to < 0.01 with > 10(-6) M ACh. The peak probability of opening in > 10(-4) M ACh pulse was close to 1. A linear reaction scheme was not compatible with these results. The scheme had to be expanded resulting in a circular scheme with two additional ACh binding steps to desensitized channel states. The approximate rate constants of all reaction steps in the circular scheme could be determined using computer simulations. The model predicted that clusters of channel opening had the average duration tau d at the respective ACh concentration. In cell-attached patches on intact muscle fibers, similar average cluster durations were observed at the respective ACh concentration. This indicates that tau d in the intact muscle fibers has similar values as in outside-out patches.     

Multiphasic desensitization of the GABAA receptor in outside-out patches.

GABAA receptor function was studied in outside-out patches from guinea pig hippocampal neurons using a drug application system with an exchange time of under 1.5 ms. Application of GABA to these patches induced a Cl- conductance that desensitized with prolonged exposure. Increasing GABA concentrations induced larger conductance increases that were associated with more complex patterns of desensitization. Smaller GABA responses desensitized with monophasic kinetics, whereas large responses displayed bi- and triphasic kinetics. Desensitization of the response to 1 mM GABA was triphasic in about 70% of the patches (tau = 15.4, 207, and 1370 ms) and biphasic in about 30% of the patches (tau = 44 and 725 ms). All phases of desensitization reversed at the Cl- equilibrium potential. Over the concentration range from 3 microM to 3 mM, both the rate and the extent of desensitization increased; however, complete desensitization was rarely observed. The increase in desensitization rate was due to an increase in the relative contribution of the faster phases with increasing GABA. The time constants of the three phases were independent of concentration. The different phases are not mediated by separate receptor populations, because double pulse experiments demonstrated interconversion among the fastest phase and the two slower phases. We demonstrate the plausibility of a model in which multiphasic desensitization is a consequence of the faster association rate at higher GABA concentrations.     

Molecular mechanisms and binding site location for the noncompetitive antagonist crystal violet on nicotinic acetylcholine receptors

We investigated the molecular mechanisms and the binding site location for the fluorophor crystal violet (CrV), a noncompetitive antagonist of the nicotinic acetylcholine receptor (AChR). To this end, radiolabeled competition binding, fluorescence spectroscopy, Schild-type analysis, patch-clamp recordings, and molecular dynamics approaches were used. The results indicate that (i) CrV interacts with the desensitized Torpedo AChR with higher affinity than with the resting state at several temperatures (5-37 degrees C); (ii) CrV-induced inhibition of the phencyclidine (PCP) analogue [(3)H]thienylcyclohexylpiperidine binding to the desensitized or resting AChR is mediated by a steric mechanism; (iii) tetracaine inhibits CrV binding to the resting AChR, probably by a steric mechanism; (iv) barbiturates modulate CrV binding to the resting AChR by an allosteric mechanism; (v) CrV itself induces AChR desensitization; (vi) CrV decreases the peak of macroscopic currents by acting on the resting AChR but without affecting the desensitization rate from the open state; and (vii) two tertiary amino groups from CrV may bind to the alpha1-Glu(262) residues (located at position 20') in the resting state. We conclude that the CrV binding site overlaps the PCP locus in the resting and desensitized state. The noncompetitive action of CrV may be explained by an allosteric mechanism in which the binding of CrV to the extracellular mouth of the resting receptor leads to an inhibition of channel opening. Binding of CrV probably increases desensitization of the resting channel and stabilizes the desensitized state.     

Inactivation viewed through single sodium channels

Recordings of the sodium current in tissue-cultured GH3 cells show that the rate of inactivation in whole cell and averaged single channel records is voltage dependent: tau h varied e-fold/approximately 26 mV. The source of this voltage dependence was investigated by examining the voltage dependence of individual rate constants, estimated by maximum likelihood analysis of single channel records, in a five-state kinetic model. The rate constant for inactivating from the open state, rather than closing, increased with depolarization, as did the probability that an open channel inactivates. The rate constant for closing from the open state had the opposite voltage dependence. Both rate constants contributed to the mean open time, which was not very voltage dependent. Both open time and burst duration were less than tau h for voltages up to -20 mV. The slowest time constant of activation, tau m, was measured from whole cell records, by fitting a single exponential either to tail currents or to activating currents in trypsin-treated cells, in which the inactivation was abolished. tau m was a bell-shaped function of voltage and had a voltage dependence similar to tau h at voltages more positive than -35 mV, but was smaller than tau h. At potentials more negative than about -10 mV, individual channels may open and close several times before inactivating. Therefore, averaged single channel records, which correspond with macroscopic current elicited by a depolarization, are best described by a convolution of the first latency density with the autocorrelation function rather than with 1 - (channel open time distribution). The voltage dependence of inactivation from the open state, in addition to that of the activation process, is a significant factor in determining the voltage dependence of macroscopic inactivation. Although the rates of activation and inactivation overlapped greatly, independent and coupled inactivation could not be statistically distinguished for two models examined. Although rates of activation affect the observed rate of inactivation at intermediate voltages, extrapolation of our estimates of rate constants suggests that at very depolarized voltages the activation process is so fast that it is an insignificant factor in the time course of inactivation. Prediction of gating currents shows that an inherently voltage-dependent inactivation process need not produce a conspicuous component in the gating current.     

Measurement of conformational changes accompanying desensitization in an ionotropic glutamate receptor

The canonical conformational states occupied by most ligand-gated ion channels, and many cell-surface receptors, are the resting, activated, and desensitized states. While the resting and activated states of multiple receptors are well characterized, elaboration of the structural properties of the desensitized state, a state that is by definition inactive, has proven difficult. Here we use electrical, chemical, and crystallographic experiments on the AMPA-sensitive GluR2 receptor, defining the conformational rearrangements of the agonist binding cores that occur upon desensitization of this ligand-gated ion channel. These studies demonstrate that desensitization involves the rupture of an extensive interface between domain 1 of 2-fold related glutamate-binding core subunits, compensating for the ca. 21 degrees of domain closure induced by glutamate binding. The rupture of the domain 1 interface allows the ion channel to close and thereby provides a simple explanation to the long-standing question of how agonist binding is decoupled from ion channel gating upon receptor desensitization.     

Mutant cycle analysis of the active and desensitized states of an AMPA receptor induced by willardiines

The halogenated willardiines are agonists at the alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) subtype of glutamate receptors. Although they differ only by the nature of the halogen substituent, they display marked differences in their efficacy to activate the receptor channel opening and in causing desensitization. We have studied the origin of the different agonist properties of the willardiines and in particular the nature of the structural element within the receptor binding domain that is able to distinguish between willardiines at a subatomic resolution of 0.6 A (the difference in radius between F and Br) and allow (S)-5-fluorowillardiine to cause receptor desensitization much more than (S)-5-bromowillardiine. For this purpose, we analyzed, with the thermodynamic mutant cycle method, the active and desensitized states induced by the willardiines in the GluR1 subtype of AMPA receptors and GluR1 mutants in which residues E398, Y446, L646, and S650, within the agonist binding domain, were mutated. The results were used to generate a 3D model of the willardiine docking mode. We suggest that the active and desensitized states of the AMPA-R correspond, respectively, to the open-lobe and closed-lobe conformations of the agonist binding domain.     

Conformational Changes Associated with Proton-dependent Gating of ASIC1a

Acid-sensing ion channels are proton-gated Na⁺ channels expressed predominantly in neurons. How channel structure translates an environmental stimulus into changes in pore permeability remains largely undefined. The pore of ASIC1 is defined by residues in the second transmembrane domain (TM2), although a segment of the outer vestibule is formed by residues of TM1. We used the voltage clamp fluorometry technique to define the role of the region preceding TM2 (pre-TM2) in activation and desensitization of mouse ASIC1a. Oocytes expressing E425C channels labeled with Alexa Fluor 488 C5-maleimide showed a change in the emission of the fluorescent probe in response to extracellular acidification. The time course of the change in fluorescence correlated with activation but not desensitization of E425C channels. The fluorescence emission did not change following extracellular acidification in oocytes carrying an inactivating mutation (W287G/E425C), although these channels were labeled and expressed at the plasma membrane. Our data indicate that pore opening occurs in conjunction with a conformational rearrangement of the pre-TM2. We observed a change in the emission of the fluorescent probe when labeled E425C channels transition from the desensitized to the resting state. The substituted-cysteine-accessibility method was used to determine whether the pre-TM2 has different conformations in the resting and desensitized states. State-dependent changes in accessibility to 2-[(trimethylammonium)ethyl]methanethiosulfonate bromide modification were observed in oocytes expressing K421C, K422C, Y424C, and E425C channels. Our results suggest that the pre-TM2 of ASIC1a undergoes dynamic conformational rearrangements during proton-dependent gating.     

Validation of Alexa-647-ATP as a powerful tool to study P2X receptor ligand binding and desensitization

Ion channel opening and desensitization is a fundamental process in neurotransmission. The ATP-gated P2X1 receptor (P2X1R) shows rapid and long-lasting desensitization upon agonist binding. This makes the electrophysiological investigation of its desensitization process, agonist unbinding, and recovery from desensitization a challenging task. Here, we show that the fluorescent agonist Alexa-647-ATP is a potent agonist at the P2X1R and a versatile tool to directly visualize agonist binding and unbinding. We demonstrate that the long-lasting desensitization of the P2X1R is due to both slow unbinding of agonist from the desensitized receptor and agonist mediated receptor internalization. Furthermore, the unbinding of the agonist Alexa-647-ATP from the desensitized receptor is accelerated in the continuous presence of competitive ligand. Modeling of our data indicates that three agonist molecules are required to drive the receptor into desensitization. Direct visualization of ligand unbinding from the desensitized receptor demonstrates the cooperativity of this process.     

The Structural Mechanism of the Cys-Loop Receptor Desensitization

The cys-loop receptors are neurotransmitter-operated ion channels, which mediate fast synaptic transmission for communication between neurons. However, prolonged exposure to the neurotransmitter drives the receptor to a desensitization state, which plays an important role in shaping synaptic transmission. Much progress has been made through more than half a century’s research since Katz and Thesleff first descried desensitization for muscle nicotinic acetylcholine receptor. In this review, we summarized recent research developments of receptor desensitization. Now, it has been identified that many parts of the receptor, such as the pore domain (including the hinge in the M2–M3 linker), the binding domain, the coupling region, and the intracellular domain, are all involved in the cys-loop receptor desensitization and that uncoupling between the amino-terminal domain and channel lining domain seems to play a central role in desensitization. This uncoupling is mainly governed by the balance between coupling strength and relative tightness of gating machinery and influenced by other parts of the receptor. Agonist binding induces conformational change to overcome the gating barrier to open the channel through the stressed coupling region, which is subsequently broken, causing receptor desensitization. With rapid advancement in structural biology of membrane receptors, final validation of this mechanism is expected to occur in the near future when the high-resolution structure of the desensitized state is available.     

Stability of ligand‐binding domain dimer assembly controls kainate receptor desensitization

AMPA and kainate receptors mediate fast synaptic transmission. AMPA receptor ligand‐binding domains form dimers, which are key functional units controlling ion‐channel activation and desensitization. Dimer stability is inversely related to the rate and extent of desensitization. Kainate and AMPA receptors share common structural elements, but functional measurements suggest that subunit assembly and gating differs between these subtypes. To investigate this, we constructed a library of GluR6 kainate receptor mutants and directly measured changes in kainate receptor dimer stability by analytical ultracentrifugation, which, combined with electrophysiological experiments, revealed an inverse correlation between dimer stability and the rate of desensitization. We solved crystal structures for a series of five GluR6 mutants, to understand the molecular mechanisms for dimer stabilization. We demonstrate that the desensitized state of kainate receptors acts as a deep energy well offsetting the stabilizing effects of dimer interface mutants, and that the deactivation of kainate receptor responses is dominated by entry into desensitized states. Our results show how neurotransmitter receptors with similar structures and gating mechanisms can exhibit strikingly different functional properties.     

Desensitization is a property of the cholinergic binding region of the nicotinic acetylcholine receptor, not of the receptor-integral ion channel.

The reversible acetylcholine esterase inhibitor (-)-physostigmine (eserine) is the prototype of a new class of nicotinic acetylcholine receptor (nAChR) activating ligands: it induces cation fluxes into nAChR-rich membrane vesicles from Torpedo marmorata electric tissue even under conditions of antagonist blocked acetylcholine binding sites (Okonjo, Kuhlmann, Maelicke, Neuron, in press). This suggests that eserine exerts its channel-activating property via binding sites at the nAChR separate from those of the natural transmitter. We now report that eserine can activate the channel even when the receptor has been preincubated (desensitized) with elevated concentrations of acetylcholine. Thus the conformational state of the receptor corresponding to desensitization is confined to the transmitter binding region, leaving the channel fully activatable-albeit only from other than the transmitter binding site(s).     

Desensitization onset and recovery at the potassium-depolarized frog neuromuscular junction are voltage sensitive

The influence of voltage on the time-course of desensitization onset and recovery has been studied at the frog neuromuscular junction. The activation-desensitization sequence was determined from carbachol- induced end-plate currents in potassium-depolarized fibers voltage- clamped either to -40 mV or +40 mV. The time-course of both desensitization onset and recovery developed exponentially, with onset occurring more rapidly than recovery. Desensitization onset was voltage dependent, the onset time constant being 8.3 +/- 1.3 s (11 fibers) at - 40 mV and 19.3 +/- 3.4 s (15 fibers) at +40 mV. Recovery from desensitization was also influenced by voltage. The extent of recovery after 2 min was 80.4 +/- 6.3% in those fibers voltage-clamped to -40 mV and 57.4 +/- 3.6% in those fibers voltage-clamped to +40 mV. The voltage dependence of desenistization onset and recovery did not result from a difference in ability to control voltage at these two levels of membrane potential. These results demonstrate that in the potassium- depolarized preparation the processes controlling both desensitization onset and recovery of sensitivity from the desensitivity from the desensitized state are influenced by membrane voltage.     

Functional consequences of the loss of high affinity agonist binding to gamma-aminobutyric acid type A receptors. Implications for receptor desensitization

We reported previously that tyrosine 62 of the beta2 subunit of the gamma-aminobutyric acid, type A (GABA(A)) receptor is an important determinant of high affinity agonist binding and that recombinant alpha1beta2gamma2(L) receptors carrying the Y62S mutation lack measurable high affinity sites for [3H]muscimol. We have now examined the effects of disrupting these sites on the macroscopic desensitization properties of receptors expressed in Xenopus oocytes. Desensitization was measured by the ability of low concentrations of bath-perfused agonist to reduce the current responses elicited by subsequent challenges with saturating concentrations of GABA. Wild-type receptors were desensitized by pre-perfused muscimol with an IC50 approximately 0.7 microm, which correlates well with the lower affinity sites for this agonist that are measured in direct binding studies. Receptors carrying the beta2 Y62S and Y62F mutations desensitized at slightly higher (2-7-fold) agonist concentrations. However, at low perfusate concentrations, the Y62S-containing receptor recovered from the desensitized state even in the continued presence of agonist. The characteristics of desensitization in the wild-type and mutant receptors lead us to suggest that the major role of the high affinity agonist-binding site(s) of the GABA(A) receptor is not to induce desensitization but rather to stabilize the desensitized state once it has been formed.     

A Charge-inverting Mutation in the “Linker” Region of α-Amino-3-hydroxy-5-methyl-4-isoxazolepropionic Acid (AMPA) Receptors Alters Agonist Binding and Gating Kinetics Independently of Allosteric Modulators

AMPA receptors are gated through binding of glutamate to a solvent-accessible ligand-binding domain. Upon glutamate binding, these receptors undergo a series of conformational rearrangements regulating channel function. Allosteric modulators can bind within a pocket adjacent to the ligand-binding domain to stabilize specific conformations and prevent desensitization. Yelshansky et al. (Yelshansky, M. V., Sobolevsky, A. I., Jatzke, C., and Wollmuth, L. P. (2004) J. Neurosci. 24, 4728–4736) described a model of an electrostatic interaction between the ligand-binding domain and linker region to the pore that regulated channel desensitization. To test this hypothesis, we have conducted a series of experiments focusing on the R628E mutation. Using ultrafast perfusion with voltage clamp, we applied glutamate to outside-out patches pulled from transiently transfected HEK 293 cells expressing wild type or R628E mutant GluA2. In response to a brief pulse of glutamate (1 ms), mutant receptors deactivated with significantly slower kinetics than wild type receptors. In addition, R628E receptors showed significantly more steady-state current in response to a prolonged (500-ms) glutamate application. These changes in receptor kinetics occur through a pathway that is independent of that of allosteric modulators, which show an additive effect on R628E receptors. In addition, ligand binding assays revealed the R628E mutation to have increased affinity for agonist. Finally, we reconciled experimental data with computer simulations that explicitly model mutant and modulator interactions. Our data suggest that R628E stabilizes the receptor closed cleft conformation by reducing agonist dissociation and the transition to the desensitized state. These results suggest that the AMPA receptor external vestibule is a viable target for new positive allosteric modulators.     

The cholesterol dependence of activation and fast desensitization of the nicotinic acetylcholine receptor.

When nicotinic acetylcholine receptors are reconstituted into lipid bilayers lacking cholesterol, agonists no longer stimulate cation flux. The kinetics of this process are difficult to study because variations in vesicle morphology cause errors in flux measurements. We developed a new stopped-flow fluorescence assay to study activation independently of vesicle morphology. When receptors were rapidly mixed with agonist plus ethidium, the earliest fluorescence increase reported the fraction of channels that opened and their apparent rate of fast desensitization. These processes were absent when the receptor was reconstituted into dioleoylphosphatidylcholine or into a mixture of that lipid with dioleoylphosphatidic acid (12 mol%), even though a fluorescent agonist reported that resting-state receptors were still present. The agonist-induced channel opening probability increased with bilayer cholesterol, with a midpoint value of 9 +/- 1.7 mol% and a Hill coefficient of 1.9 +/- 0.69, reaching a plateau above 20-30 mol% cholesterol that was equal to the native value. On the other hand, the observed fast desensitization rate was comparable to that for native membranes from the lowest cholesterol concentration examined (5 mol%). Thus the ability to reach the open state after activation varies with the cholesterol concentration in the bilayer, whereas the rate of the open state to fast desensitized state transition is unaffected. The structural basis for this is unknown, but an interesting corollary is that the channels of newly synthesized receptors are not fully primed by cholesterol until they are inserted into the plasma membrane--a novel form of posttranslational processing.     

Dual effect of acid pH on purinergic P2X3 receptors depends on the histidine 206 residue

Whole cell patch clamp investigations were carried out to clarify the pH sensitivity of native and recombinant P2X(3) receptors. In HEK293 cells permanently transfected with human (h) P2X(3) receptors (HEK293-hP2X(3) cells), an acidic pH shifted the concentration-response curve for alpha,beta-methylene ATP (alpha,beta-meATP) to the right and increased its maximum. An alkalic pH did not alter the effect of alpha,beta-meATP. Further, a low pH value increased the activation time constant (tau(on)) of the alpha,beta-meATP current; the fast and slow time constants of desensitization (tau(des1), tau(des2)) were at the same time also increased. Finally, acidification accelerated the recovery of P2X(3) receptors from the desensitized state. Replacement of histidine 206, but not histidine 45, by alanine abolished the pH-induced effects on hP2X(3) receptors transiently expressed in HEK293 cells. Changes in the intracellular pH had no effect on the amplitude or time course of the alpha,beta-meATP currents. The voltage sensitivity and reversal potential of the currents activated by alpha,beta-meATP were unaffected by extracellular acidification. Similar effects were observed in a subpopulation of rat dorsal root ganglion neurons expressing homomeric P2X(3) receptor channels. It is suggested that acidification may have a dual effect on P2X(3) channels, by decreasing the current amplitude at low agonist concentrations (because of a decrease in the rate of activation) and increasing it at high concentrations (because of a decrease in the rate of desensitization). Thereby, a differential regulation of pain sensation during e.g. inflammation may occur at the C fiber terminals of small DRG neurons in peripheral tissues.     

Voltage-dependent block of anthrax toxin channels in planar phospholipid bilayer membranes by symmetric tetraalkylammonium ions. Single-channel analysis

Previous studies have shown that symmetric tetraalkylammonium ions affect, in a voltage-dependent manner, the conductance of membranes containing many channels formed by the PA65 fragment of anthrax toxin. In this paper we analyze this phenomenon at the single-channel level for tetrabutylammonium ion (Bu4N+). We find that Bu4N+ induces a flickery block of the PA65 channel when present on either side of the membrane, and this block is relieved by large positive voltages on the blocking-ion side. At high frequencies (greater than 2 kHz) we have resolved individual blocking events and measured the dwell times in the blocked and unblocked states. These dwell times have single-exponential distributions, with time constants tau b and tau u that are voltage dependent, consistent with the two-barrier, single-well potential energy diagram that we postulated in our previous paper. The fraction of time the channel spends unblocked [tau u/(tau u + tau b)] as a function of voltage is identical to the normalized conductance-voltage relation determined from macroscopic measurements of blocking, thus demonstrating that these single channels mirror the behavior seen with many (greater than 10,000) channels in the membrane. In going from large negative to large positive voltages (-100 to +160 mV) on the cis (PA65-containing) side of the membrane, one sees the mean blocked time (tau b) increase to a maximum at +60 mV and then steadily decline for voltages greater than +60 mV, thereby clearly demonstrating that Bu4N+ is driven through the channel by positive voltages on the blocking-ion side. In other words, the channel is permeable to Bu4N+. An interesting finding that emerges from analysis of the voltage dependence of mean blocked and unblocked times is that the blocking rate, with Bu4N+ present on the cis side of the membrane, plateaus at large positive cis voltages to a voltage-independent value consistent with the rate of Bu4N+ entry into the blocking site being diffusion limited.     

Effects of Substance P on Nicotinic Acetylcholine Receptor Function in PC 12 Cells

The effects of substance P on the functioning of nicotinic acetylcholine receptors in PC12 cells were examined. Carbachol-stimulated 22Na+ uptake was used to assess the functional state of the nicotinic receptor. We found that incubation of the cells with substance P alone caused a loss of receptor function. Receptors recovered from this effect with a t1/2 of 0.94 +/- 0.10 min. Since receptors recovered from carbachol-induced desensitization at a significantly slower rate (t1/2, 1.77 +/- 0.21 min), it was concluded that the two inactive states are not kinetically equivalent. The effects of substance P on carbachol-induced loss of receptor activity were also examined. Substance P had no effect on a component of carbachol-induced loss of activity that was nonrecoverable (inactivation). However, substance P had several effects on the recoverable loss of activity induced by carbachol (desensitization). Substance P caused a shift to the left in the EC50 for carbachol-induced desensitization at equilibrium. If cells were simultaneously incubated with carbachol and substance P7-11, a low-potency analog of substance P, an increase in the rate of formation of a state of the receptor that was kinetically indistinguishable from the state induced by carbachol alone was observed. However, not all inhibition of nicotinic cholinergic function could be explained by an increased rate of formation of a desensitized receptor and it is concluded that substance P causes both enhanced desensitization and block of the nicotinic receptor-linked channel.     

A tyrosine substitution in the cavity wall of a k channel induces an inverted inactivation

Ion permeation and gating kinetics of voltage-gated K channels critically depend on the amino-acid composition of the cavity wall. Residue 470 in the Shaker K channel is an isoleucine, making the cavity volume in a closed channel insufficiently large for a hydrated K(+) ion. In the cardiac human ether-a-go-go-related gene channel, which exhibits slow activation and fast inactivation, the corresponding residue is tyrosine. To explore the role of a tyrosine at this position in the Shaker channel, we studied I470Y. The activation became slower, and the inactivation faster and more complex. At +60 mV the channel inactivated with two distinct rates (tau(1) = 20 ms, tau(2) = 400 ms). Experiments with tetraethylammonium and high K(+) concentrations suggest that the slower component was of the P/C-type. In addition, an inactivation component with inverted voltage dependence was introduced. A step to -40 mV inactivates the channel with a time constant of 500 ms. Negative voltage steps do not cause the channel to recover from this inactivated state (tau > 10 min), whereas positive voltage steps quickly do (tau = 2 ms at +60 mV). The experimental findings can be explained by a simple branched kinetic model with two inactivation pathways from the open state.