Study on ATP concentration changes in cytosol of individual cultured neurons during glutamate-induced deregulation of calcium homeostasis.

For the first time, simultaneous monitoring of changes in the concentration of cytosolic ATP ([ATP]c), pH (pHc), and intracellular free Ca2+ concentration ([Ca2+]i) of the individual neurons challenged with toxic glutamate (Glu) concentrations was performed. To this end, the ATP-sensor AT1.03, which binds to ATP and therefore enhances the efficiency of resonance energy transfer between blue fluorescent protein (energy donor) and yellow-green fluorescent protein (energy acceptor), was expressed in cultured hippocampal neurons isolated from 1-2-day-old rat pups. Excitation of fluorescence in the acceptor protein allowed monitoring changes in pHc. Cells were loaded with fluorescent low-affinity Ca2+ indicators Fura-FF or X-rhod-FF to register [Ca2+]i. It was shown that Glu (20 µM, glycine 10 µM, Mg2+-free) produced a rapid acidification of the cytosol and decrease in [ATP]c. An approximately linear relationship (r(2) = 0.56) between the rate of [ATP]c decline and latency of glutamate-induced delayed calcium deregulation (DCD) was observed: higher rate of [ATP]c decrease corresponded to shorter DCD latency period. DCD began with a decrease in [ATP]c of as much as 15.9%. In the phase of high [Ca2+]i, the plateau of [ATP]c dropped to 10.4% compared to [ATP]c in resting neurons (100%). In the presence of the Na+/K+-ATPase inhibitor ouabain (0.5 mM), glutamate-induced reduction in [ATP]c in the phase of the high [Ca2+]i plateau was only 36.6%. Changes in [ATP]c, [Ca2+]i, mitochondrial potential, and pHc in calcium-free or sodium-free buffers, as well as in the presence of the inhibitor of Na+/K+-ATPase ouabain (0.5 mM), led us to suggest that in addition to increase in proton conductivity and decline in [ATP]c, one of the triggering factors of DCD might be a reversion of the neuronal plasma membrane Na+/Ca2+ exchange.

[Mathematical modelling of delayed calcium deregulation in brain neurons caused by hyperstimulation of glutamate receptors].

Mathematical model presented in this paper is based on the data obtained in experimental studies of the mechanism of glutamate-induced deregulation of Ca2+ homeostasis and mitochondrial depolarization in cultured nerve cells. According to our hypothesis the secondary Ca2+ increase during a prolonged glutamate challenge is mainly due to a profound mitochondrial depolarization which stops mitochondrial Ca2+ uptake in the face of continuous Ca2+ influx into the cell. It is supposed that a progressive decrease in mitochondrial NADH during glutamate exposure greatly enhances sensitivity of mitochondria to intracellular Ca-neurotoxicity. A system of equations developed in this work made it possible to provide a satisfactory simulation of most of experimental events observed in the present study.

Probing of NMDA channels with fast blockers.

Using whole-cell patch-clamp techniques, we studied the interaction of open NMDA channels with tetraalkylammonium compounds: tetraethylammonium (TEA), tetrapropylammonium (TPA), tetrabutylammonium (TBA), and tetrapentylammonium (TPentA). Analysis of the blocking kinetics, concentration, and agonist dependencies using a set of kinetic models allowed us to create the criteria distinguishing the effects of these blockers on the channel closure, desensitization, and agonist dissociation. Thus, it was found that TPentA prohibited, TBA partly prevented, and TPA and TEA did not prevent either the channel closure or the agonist dissociation. TPentA and TBA prohibited, TPA slightly prevented, and TEA did not affect the channel desensitization. These data along with the voltage dependence of the stationary current inhibition led us to hypothesize that: (1) there are activation and desensitization gates in the NMDA channel; (2) these gates are distinct structures located in the external channel vestibule, the desensitization gate being located deeper than the activation gate. The size of the blocker plays a key role in its interaction with the NMDA channel gating machinery: small blockers (TEA and TPA) bind in the depth of the channel pore and permit the closure of both gates, whereas larger blockers (TBA) allow the closure of the activation gate but prohibit the closure of the desensitization gate; finally, the largest blockers (TPentA) prohibit the closure of both activation and desensitization gates. The mean diameter of the NMDA channel pore in the region of the activation gate localization was estimated to be approximately 11 A.

The relationship between mitogen-induced changes in the cytoplasmic pH and free Ca2+ concentration in rat thymocytes.

The relationship between pHi and [Ca]i signals generated in rat thymocytes by the mitogen Con A has been investigated. It is shown that the mitogen-induced [Ca]i rise is dependent on Na+/H+ exchange or some other Na(+)-sensitive process. This conclusion is based on the following findings: (i) [Ca]i response to Con A weakens upon decreasing the concentration of extracellular Na+, or inhibiting Na+/H+ exchange; (ii) agents that alkalinize the cytoplasm (the phorbol ester TPA, the Na+/H+ ionophore monensin and NH4Cl) cause an increase in [Ca]i (Klip, A., Rothstein, A. and Mack, E. (1984) Biochem. Biophys. Res. Commun. 124, 14-22; Grinstein, S. and Goetz, J.D. (1985) Biochim. Biophys. Acta 819, 267-270); (iii) The effects of Con A, TPA and monensin on [Ca]i are not additive. The last observation suggests that all these agents activate the same Na+/H+ (Na+ and/or H+)-dependent system of Ca2+ transport. It is found that the pH i and [Ca]i responses in rat thymocytes are sensitive to changes in the intracellular levels of cyclic nucleotides, ATP and in temperature. These regulatory effects on the ionic signals are different for Con A, TPA and monensin. In particular, both the stimulation of Na+/H+ antiport and the [Ca]i rise brought about by Con A or TPA are inhibited upon elevating the cellular cAMP. In contrast, the monensin-induced [Ca]i signal is almost independent of cAMP but is highly sensitive to changes in cGMP and temperature. Reducing the ATP level eliminates both the pHi and [Ca]i responses to Con A but not to monensin. These different characteristics of [Ca]i signals elicited by the mitogen and the Na+/H+ ionophore indicate that these agents use different mechanisms to activate the Na+/H(+)-dependent Ca2+ transporting system. A [Ca]i response to monensin has been obtained in some other cell types, namely, in lymphoblastoid Raji cells, Ehrlich ascites tumor cells and also in platelets.

Fluctuation analysis of Na+ channels modified by batrachotoxin in myelinated nerve.

(1) Single myelinated nerve fibers of Rana esculenta were treated with the steroidal alkaloid batrachotoxin, and Na+ currents and Na+-current fluctuations were measured near the resting potential under voltage-clamp conditions. Between test pulses the fibres were held at hyperpolarizing membrane potentials. (2) The spectral density of Na+-current fluctuations was fitted by the sum of a 1/f component and a Lorentzian function. The time constant tau c = 1/(2 pi fc) obtained from the corner frequency fc of the Lorentzian function approximately agreed with the activation time constant tau m of the macroscopic currents. (3) The conductance gamma of a single Na+ channel modified by batrachotoxin was calculated from the integral of the Lorentzian function and the steady-state Na+ current. At the resting potential V = 0 we obtained gamma - 1.6 pS, higher gamma-values of 3.2 and 3.45 pS were found at V = --8 and --16 mV, respectively. (4) The conductance of a modified Na+ channel is significantly lower than the values 6.4 to 8.85 pS reported in the literature for normal Na+ channels. Hence, our experiments are in agreement with the view that batrachotoxin acts in an 'all-or-none' manner on Na+ channels and creates a distinct population of modified channels.

Voltage dependence of intramembrane charge movement and conductance activation of batrachotoxin-modified sodium channels in frog node of Ranvier.

Sodium current and sodium channel intramembrane gating charge movement (Q) were monitored in voltage-clamped frog node of Ranvier after modification of all sodium channels by batrachotoxin (BTX). BTX caused an approximately threefold increase in steepness of the Q vs. voltage relationship and a 50-mV negative shift in its midpoint. The maximum amount of intramembrane charge was virtually identical before and after BTX treatment. BTX treatment eliminated the charge immobilization observed in untreated nodes after relatively long depolarizing pulses and slowed the rate of OFF charge movement after a pulse. After BTX treatment, the voltage dependence of charge movement was the same as the steady-state voltage dependence of sodium conductance activation. The observations are consistent with the hypothesis that BTX induces an aggregation of the charged gating particles associated with each channel and causes them to move as a unit having approximately three times the average valence of the individual particles. Movement of this single aggregated unit would open the BTX-modified sodium channel.

Changes of cytoplasmic pH in frog nerve fibers during K(+)-induced membrane depolarization.

Frog sciatic nerve and its thin bundles were loaded with fluorescein diacetate in order to monitor changes in cytoplasmic pH (pHi) caused by high K+ depolarization. Isosmotic substitution of external Na+ by K+ at pHo 7.3 led to a steady concentration-dependent (20-120 mM K+) decrease in pHi. Elevation of pHo from 7.3 to 8.5 prevented or even reversed these pHi changes, indicating their strong dependence on transmembrane H+ fluxes. The depolarization-induced intracellular acidification could not be prevented or decreased by any of the following treatments: removal of external Ca2+; application of the Ca2+ antagonists Ni2+ and Co2+; blockade of K+ channels by TEA; addition to the external solution of Zn2+, a blocker of putative voltage-sensitive H+ channels. By contrast, blockade of Na+ channels by 1-3 microM TTX prevented the effect of high K+ concentrations on pHi. It is concluded that the decrease in pHi induced by a prolonged membrane depolarization in frog nerve fibers is mainly due to an enhanced H+ influx through non-inactivating Na+ channels.

Effects of repetitive stimulation, veratridine and ouabain on cytoplasmic pH in frog nerve fibres: role of internal Na+.

Changes of cytoplasmic pH (pHi) in frog nerve fibres during repetitive stimulation have been measured using the fluorescent pH indicator dye fluorescein diacetate (FDA). Under control conditions repetitive (10-50 Hz) stimulation caused only a very small decrease in pHi (by 0.015-0.06 pH units). Modification of Na+ channels by veratridine (VER, 10 microM) greatly increased this stimulus-evoked (SE) internal acidification. Blockade of the Na(+)-K+ pump by ouabain (0.5 mM) enhanced the effects VER and prevented pHi recovery after the termination of repetitive stimulation. A similar inhibition of post-stimulatory recovery of pHi was observed after replacement of external Na+ with Li+, which is not accepted by the Na(+)-K+ pump instead of Na+. These data suggest that SE intracellular acidification in nerves results from or is closely associated with an increase in [Na+]i. Treatments that promote Na+ influx and accumulation of Na+ inside the fibre enhance reduction of pHi. Li+ can be substituted for Na+ in this process.

Blockade of ADP-induced Ca2+-signal and platelet aggregation by lipoxygenase inhibitors.

Stimulation of platelets results in the liberation of arachidonic acid (AA) which is further metabolized via the cyclooxygenase or lipoxygenase (LPG) pathway. We have examined the effect of inhibition of LPG on (i) the ADP-induced increase of cytoplasmic Ca2+ concentration and (ii) platelet aggregation. Lipoxygenase inhibitors, nordigidroguaiaretic acid (NDGA) and BW-755C, both suppressed ADP-induced Ca2+-signals and aggregation in a dose-dependent manner, with an IC50 value of 1 2 microM for NDGA. Qualitatively the same effect was obtained with 4-bromophenylacyl bromide, the inhibitor of phospholipases A2 and C. By contrast, cyclooxygenase inhibitor indomethacin had only a negligible effect on Ca2+-signals and suppressed only the second phase of ADP-induced aggregation. It is concluded that the LPG pathway of AA metabolism in platelets might play a crucial role in ADP-induced Ca2+-signal generation and platelet aggregation.

Blockade of NMDA channels in acutely isolated rat hippocampal neurons by the Na+/Ca2+ exchange inhibitor KB-R7943.

Neurons acutely isolated from the CA1 region of rat hippocampal slices using the 'vibrodissociation' method were voltage-clamped in the whole-cell configuration. The currents through NMDA channels were elicited by application of 100 microM aspartate (ASP) in a Mg2+-free solution in the presence of 3 microM glycine. The compound KB-R7943, (2-[2-[4-(4-nitrobenzyloxy)phenyl]ethyl]isothiourea methanesulfonate) known as a Na+/Ca2+ exchange inhibitor was able to block effectively the NMDA channels. At a holding potential of -100 mV, the measurement of the concentration dependence of the stationary current blockade revealed the existence of two populations of NMDA channels differing by a high (IC50 = 0.8 microM) and low (IC50 = 11 microM) affinity for KB-R7943. The Hill coefficients indicated that one blocking molecule can bind to NMDA channels which have a high affinity for KB-R7943 and at least two blocking molecules can bind to the NMDA channels which have a low affinity for KB-R7943. When applied externally, KB-R7943 can bind to the low-affinity NMDA channels irrespective of whether or not these channels are activated by the agonist. The KB-R7943-induced blockade of the NMDA channel was partly voltage-dependent. Within the framework of the Woodhull model, the apparent value of delta calculated for the voltage-dependent binding of KB-R7943 was in the range of 0.26-0.41. The blocking action of KB-R7943 on NMDA channels did not depend either on ASP or glycine concentrations which indicated that the binding sites for KB-R7943 and those for the agonist and the coagonist did not overlap.

Bepridil-induced blockade of NMDA channels in rat hippocampal neurones.

Neurones isolated from the CA1 region of rat hippocampal slices by the "vibrodissociation" method were voltage-clamped in the whole-cell configuration. The currents through N-methyl-D-aspartate (NMDA) channels were recorded in response to the rapid application (solution exchange time <30 msec) of 100 microM aspartate (ASP) in a Mg2+-free solution in the presence of 3 microM glycine. When added to the ASP solution, bepridil (BPD) caused a concentration-dependent decrease in both peak and stationary currents due to an uncompetitive open-channel blockade of NMDA channels. At -100 mV, the half-blocking concentration (IC50) for the stationary current was 14.01 +/- 0.17 microM (n = 10). The blocking and unblocking time constants were 7.4 +/- 0.3 x 10(3)/M/sec and 0.12 +/- 0.02/sec, respectively. Membrane hyperpolarization enhanced the BPD block. The equilibrium dissociation constant behaved as an exponential function of the membrane potential and increased e-fold every 37 mV.

The effect of yohimbine on sodium and gating currents in frog Ranvier node membrane.

The indolalkylamine alkaloid yohimbine induced two phenomenologically-different types of sodium current (INa) inhibition in the voltage-clamped frog node of Ranvier, a tonic and a phasic ('use-dependent') block. The latter developed during a repetitive membrane stimulation with short (5 ms) depolarizing pulses at frequencies at 1 to 10 Hz. Unlike repetitive pulsing, a single-long lasting (1 s) depolarizing step did not produce a phasic block. Turning on a hyperpolarizing prepulse (50 ms to E = -123 mV) immediately before each test pulse produced a gradual unblocking of Na channels, while a depolarizing prepulse (to -86 mV) enhanced the phasic block. Yohimbine blocked the outward INa much more strongly than the inward ones. Reduction of external Na+ ions concentration from 112 to 55 mM caused a shift in the voltage-department of yohimbine block to more negative voltages, which coincided with the shift of INa reversal potential. Sodium current inhibition produced by yohimbine was accompanied by partial depression of the intramembrane charge movements ('ON-response'). Modification of Na channels by batrachotoxin made the Na channels resistant to both tonic and phasic blocking action of yohimbine. The features of the yohimbine-induced block suggest an interaction of the drug with open Na channels. The current-dependence of yohimbine block indicates an electrostatic interaction between Na+ ion and charged (protonated) form of yohimbine within the channel lumen and suggests the localization of the receptor at the inner mouth of the channel. Binding of yohimbine to the channel receptor promotes the inactivation of this channel. Comparison of the effects of yohimbine on NA and gating currents with those of local anesthetics leads us to suggest that these drugs share a common receptor.

Use-dependent blockade of sodium channels by local anaesthetics and antiarrhythmic drugs. Effects of chloramine-T and calcium ions.

Voltage clamp studies were carried out of the effects of chloramine-T(CT) and external Ca++ on the blocking interactions of local anaesthetics (LAs) and antiarrhythmic drugs (lidocaine, tetracaine, N-propyl ajmaline, compound KC 3791) with Na+ channels in frog Ranvier nodes. The results obtained provided direct evidence for the notion that: LAs interact preferentially with inactivated Na+ channels and stabilize their inactivated conformation ("drug-induced slow inactivation": SI); and SI underlies the cumulative inhibition of INa during repetitive membrane stimulation. Normal inactivation is not indispensable, but plays an auxiliary role in the mechanism of cumulative inhibition of INa by drugs interacting with open Na+ channels. This block results mainly from accumulation of the channels in the resting blocked state (due to the inability of charged drugs to leave the channel via a "hydrophobic pathway"). The contribution of the blockade-inactivated state to this type of block may depend on some properties of the drug and the holding membrane potential. The problem of the location of the binding site responsible for LA-induced SI requires further investigation in view of the fact that in the myocardium, along with LA, the lipid-insoluble tetrodotoxin (TTX) induces a pronounced SI.

Effects of KC 3791 on sodium and potassium channels in frog node of Ranvier.

Effects of a new antiarrhytmic compound KC 3791 on sodium (INa) and potassium (IK) currents were studied in frog myelinated nerve fibres under voltage clamp conditions. When applied externally to the node of Ranvier, KC 3791 (KC) at concentrations of 10(-5)-10(-4) mol.l-1 produced both tonic and cumulative (use-dependent) inhibition of INa. An analysis of the frequency-, voltage- and time dependence of cumulative block by KC suggested that this block resulted from a voltage-dependent interaction of the drug with open Na channels. The progressive decrease in INa during repetitive pulsing was due to accumulation of Na channels in the resting-blocked state: closing of the activation gate after the end of each depolarizing pulse stabilized the KC-"receptor" complex. To unblock these channels a prolonged washing of the node had to be combined with a subsequent repetitive stimulation of the membrane; this suggested that channel could not become cleared of the blocker unless the activation gate has opened. KC also proved to be capable of blocking open K channels at outwardly directed potassium currents (IK). This block increased during membrane depolarization. Unblocking of K channels after the end of a depolarizing pulse proceeded much faster than unblocking of Na channels under identical conditions. Cumulative inhibition of outward IK during high-frequency membrane stimulation was therefore readily reversible upon a decrease in pulsing frequency.

Blockade of sodium and potassium channels in the node of Ranvier by ajmaline and N-propyl ajmaline.

The inhibition of sodium and potassium currents in frog myelinated fibres by ajmaline (AM) and its quaternary derivative, N-propyl ajmaline (NPA), depends on voltage-clamp pulses and the state of channel gating mechanisms. The permanently charged NPA and protonated AM interact only (or mainly) with open channels, while unprotonated AM affects preferently inactivated Na channels. Inhibition of Na currents by NPA and AM does not depend on the current direction and Na ion concentration in external or internal media. In contrast only the outward potassium currents can be blocked by NPA and AM; the inward potassium currents in high K+ ions external media are resistant to the blocking action of these drugs. The voltage dependence of ionic current inhibition by charged drugs suggests the location of their binding sites in the inner mouths of Na and K channels. Judging by the kinetics of current restoration after cessation of pulsing, the drug-binding site complex is much more stable in Na than in potassium channels. Batrachotoxin and aconitine, unlike veratridine and sea anemone toxin, decrease greatly the affinity of Na channel binding sites to NPA and AM. The effects of NPA and AM are compared with those of local anesthetics and other amine blocking drugs.

A study of properties of batrachotoxin modified sodium channels.

A further analysis of the effects of the steroidal alkaloid batrachotoxin (BTX) on sodium channels in frog node of Ranvier has been carried out under voltage-clamp conditions. The main properties of modified channels as compared with those of normal ones are as follows: The rate of channel closing is drastically decreased, whereas that of opening is changed slightly if at all; The steady-state voltage dependence of channel activation is shifted towards more negative potentials by 60-70 mV; Currents through modified channels do not show a decay during maintained depolarization as it is typical for normal channels. However modified channels retain the ability to partial inactivation as shown by experiments with depolarizing prepulses; Sodium against potassium selectivity beyond--20 mV suggesting either nonhomogeneity of the modified channels as for their kinetic and selectivity properties or potential-dependence of ionic selectivity for each channel; The selectivity sequence determined from peak current reversal potential measurements is as follows: H: Na :NH4:K = 528:1:0.47: :0.19; The effective pK value of proton block is decreased by about 0.4; 7) The sensitivity of the channels to tetrodotoxin (TTX) block is practically unchanged.

The possible role of intracellular Ca2+-stores in the rhythm-inotropic relationship of frog heart muscle (a simulation study).

The hypothesis that intracellular calcium stores play an essential role in determining force-frequency relationships of frog myocardium was tested quantitatively. A simplified mathematical model of excitation-contraction coupling in frog heart muscle was developed and its behaviour under various patterns of stimulation was analysed by means of computer simulation. The model represents a system of ordinary differential equations for individual fluxes within the cell Ca2+-recirculation system and includes a one-compartmental intracellular pool as opposed to the two-compartmental structure of the mammalian sarcoplasmic reticulum (Kaufmann et al. 1974). The behaviour of the model is consistent with available experimental data concerning the basic rhythm-inotropic characteristics of amphibian myocardium and offers some evidence in favour of the basic concept. Within the framework of the proposed model the staircase phenomena in amphibia were accounted for and the impact of different intracellular Ca-movements on the resulting contractile response and rhythm-inotropic phenomena was elucidated.

Blocker studies of the functional architecture of the NMDA receptor channel.

Blockade of ion channels passing through the NMDA receptors of isolated rat hippocampus pyramidal neurons with tetraalkylammonium compounds, 9-aminoacridine, and Mg2+ was studied using patch-clamp methods in the whole-cell configuration. Currents through NMDA channels were evoked by application of 100 microM aspartate in magnesium-free medium containing glycine (3 microM) to neurons. Analysis of the kinetics, charge transfer, and relationships between the extent of suppression of stationary currents on the one hand and membrane potential, agonist concentration, and blocker concentration on the other showed that blockers had different effects on the closing, desensitization, and agonist dissociation of NMDA channels. The size of the blocker was found to be the decisive factor determining its action on the gating functions of NMDA channels: larger blockers prevented closure and/or desensitization of the channel; smaller blockers only had partial effects on these processes, while the smallest blockers had no effect at all. These experiments showed that the apparent affinity of the blocker for the channel (1/IC50) depended not only on the microscopic equilibrium dissociation constant (Kd), but also on the number of blocker binding sites, their mutual influences, and, of particular importance, the interaction of the blocker with the gating structures of the channel. These data led us to propose hypotheses relating to the geometry of the NMDA channel and the structure of its gating mechanism. The channel diameter at the level of activated gates was estimated to be 11 A.

The leading role of mitochondrial depolarization in the mechanism of glutamate-induced disruptions in Ca2+ homeostasis.

Data obtained in studies of the nature of the correlation which we have previously observed [10,17] between mitochondrial depolarization and the level of disruption of Ca2+ homeostasis in cultivated brain neuronsare summarized. Experiments were performed on cultured cerebellar granule cells loaded with Fura-2-AM or rhodamine 123 to measure changes in cytoplasmic Ca2+ and mitochondrial potential during pathogenic treatments of the cells. Prolonged exposure to 100 microM glutamate induced a reversible increase in [Ca2+]i, which was accompanied by only a small degree of mitochondrial depolarization. A sharp increase in this mitochondrial depolarization, induced by addition of 3 mM NaCN or 300 microM dinitrophenol (DNP) to the glutamate-containing solution, resulted in further increase in [Ca2+]i, due to blockade of electrophoretic mitochondrial Ca2+ uptake. Prolonged exposure to CN- or DNP in the post-glutamate period maintained [Ca2+]i at a high level until the metabolic inhibitors were removed. In most cells, this plateau was characterized by low sensitivity to removal of external Ca2+, demonstrating that the mechanisms of Ca2+ release from neurons were disrupted. Addition of oligomycin, a blocker of mitochondrial ATP synthase/ATPase, to the solution containing glutamate and CN- or DNP eliminated the post-glutamate plateau. Parallel experiments with direct measurements of intracellular ATP levels ([ATP]) showed that profound mitochondrial depolarization induced by CN- or DNP sharply enhanced the drop in ATP due to glutamate, while oligomycin significantly weakened this effect of the metabolic inhibitors. Analysis of these data led to the conclusion that blockade of mitochondrial Ca2+ uptake and inhibition of ATP synthesis resulted from mitochondrial depolarization and plays a key role in the mechanism disrupting [Ca2+]i homeostasis after toxic exposure to glutamate.

[Action of batrachotoxin on the membrane sodium channels of neuroblastoma cells]. / Deistvie batrakhotoksina na natrievye kanaly membrany kletok neiroblastomy.

Currents through normal and batrachotoxin (BTX)-modified sodium channels of dialyzed neuroblastoma cells were measured under voltage clamp conditions. BTX is shown to induce a shift of voltage range of activation toward more negative potentials by 25-40 mV and the appearance of steady-state sodium conductance. BTX-modified sodium channels retain the ability to partial inactivation. It is evidenced by partial decay of the current during maintained depolarization and by dependence of current size and kinetics on prepulses. BTX induces changes in channel selectivity. Permeability ratios determined from reversal potential measurements are: Na : NH4 : K = 1 : 0.70 : 0.29 and 1 : 0.35 : 0.11 for BTX-modified and normal channels, respectively.