Step Changes In Membrane Potential Research Articles

Photoswitching alters fluorescence readout of jGCaMP8 Ca2+ indicators tethered to Orai1 channels

We used electrophysiology and Ca2+ channel tethering to evaluate the performance of jGCaMP8 genetically encoded Ca2+ indicators (GECIs). Orai1 Ca2+ channel-jGCaMP8 fusions were transfected into HEK 293A cells and jGCaMP8 fluorescence responses recorded by simultaneous total internal reflection fluorescence microscopy and whole-cell patch clamp electrophysiology. Noninactivating currents from the Orai1 Y80E mutant provided a steady flux of Ca2+ controlled on a millisecond time scale by step changes in membrane potential. Test pulses to -100 mV produced Orai1 Y80E-jGCaMP8f fluorescence traces that unexpectedly declined by ~50% over 100 ms before reaching a stable plateau. Testing of Orai1-jGCaMP8f using unroofed cells further demonstrated that rapid and partial fluorescence inactivation is a property of the indicator itself, rather than channel function. Photoinactivation spontaneously recovered over 5 min in the dark, and recovery was accelerated in the absence of Ca2+. Mutational analysis of residues near the tripeptide fluorophore of jGCaMP8f pointed to a mechanism: Q69M/C70V greatly increased (~90%) photoinactivation, reminiscent of fluorescent protein fluorophore cis-trans photoswitching. Indeed, 405-nm illumination of jGCaMP8f or 8m/8s/6f led to immediate photorecovery, and simultaneous illumination with 405 and 488-nm light blocked photoinactivation. Subsequent mutagenesis produced a variant, V203Y, that lacks photoinactivation but largely preserves the desirable properties of jGCaMP8f. Our results point to caution in interpreting rapidly changing Ca2+ signals using jGCaMP8 and earlier series GECIs, suggest strategies to avoid photoswitching, and serve as a starting point to produce more photostable, and thus more accurate, GECI derivatives.

Investigating the Conformational States of the Rabbit Na +/Glucose Cotransporter

The Na + and voltage-dependence of transient rabbit Na +/glucose cotransporter (rSGLT1) kinetics was studied with the two-electrode voltage-clamp technique and Xenopus laevis oocytes. Using step changes in membrane potential, in the absence of glucose but with 100 or 10 mM Na +, transient currents were measured corresponding to binding/debinding of Na + and conformational changes of the protein. Previously, only a single time constant has been published for rSGLT1. We, however, observed three decay components; a fast ( τ f, 0.5–1 ms) voltage- and Na +-independent decay, and medium ( τ m, 0.5–4 ms) and slow ( τ s, 8–50 ms) voltage- and Na +-dependent decays. Transient currents were simulated and fit using a four-state model to obtain kinetic parameters for the system. The four-state model was able to reconstitute an assortment of experimental data.

In vivo intracellular recording and perturbation of persistent activity in a neural integrator.

To investigate the mechanisms of persistent neural activity, we obtained in vivo intracellular recordings from neurons in an oculomotor neural integrator of the goldfish during spontaneous saccades and fixations. Persistent changes in firing rate following saccades were associated with step changes in interspike membrane potential that were correlated with changes in eye position. Perturbation of persistent activity with brief intracellular current pulses designed to mimic saccadic input only induced transient changes of firing rate and membrane potential. When neurons were hyperpolarized below action potential threshold, position-correlated step changes in membrane potential remained. Membrane potential fluctuations were greater during more depolarized steps. These results suggest that sustained changes in firing rate are supported not by either membrane multistability or changes in pacemaker currents, but rather by persistent changes in the rate or amplitude of synaptic inputs.

Inactivation of Ca2+ Release Channels (Ryanodine Receptors RyR1 and RyR2) with Rapid Steps in [Ca2+] and Voltage

The transient responses of sheep cardiac and rabbit skeletal ryanodine receptors (RyRs) to step changes in membrane potential and cytosolic [Ca2+] were measured. Both cardiac and skeletal RyRs have two voltage-dependent inactivation processes (τ≈1–3s at +40mV) that operate at opposite voltage extremes. Approximately one-half to two-thirds of RyRs inactivated when the bilayer voltage was stepped either way between positive and negative values. Inactivation was not detected (within 30s) in RyRs with Po less than 0.2. Inactivation rates increased with intraburst open probability (Po) and in proportion to the probability of a long-lived, RyR open state (POL). RyR inactivation depended on POL and not on the particular activator (Ca2+ (μM), ATP, caffeine, and ryanodine), inhibitor (mM Ca2+ and Mg2+), or gating mode. The activity of one-half to two-thirds of RyRs declined (i.e., the RyRs inactivated) after [Ca2+] steps from subactivating (0.1μM) to activating (1–100μM) levels. This was due to the same inactivation mechanism responsible for inactivation after voltage steps. Both forms of inactivation had the same kinetics and similar dependencies on Po and voltage. Moreover, RyRs that failed to inactivate after voltage steps also did not inactivate after [Ca2+] steps. The inactivating response to [Ca2+] steps (0.1–1μM) was not RyRs “adapting” to steady [Ca2+] after the step, because a subsequent step from 1 to 100μM failed to reactivate RyRs.

Electrogenic K+ transport by the Na(+)-K+ pump in rat cardiac ventricular myocytes.

1. The involvement of electrogenic reaction steps in K+ transport by the Na+, K(+)-ATPase was determined in rat cardiac ventricular myocytes using whole-cell patch clamp techniques. 2. Under K(+)-K+ exchange conditions and in the presence of extracellular K+ or Tl+ at concentrations that stimulated submaximal levels of steady-state Na+,K(+)-ATPase activity, ouabain-sensitive transient currents were observed during ('on') and after ('off') step changes in membrane potential (Vm). 3. The quantity of charge moved during the transient currents depended, in a saturable manner, on the magnitude of the voltage step. Maximal ouabain-sensitive 'on' and 'off' charges were calculated to be 9.6 +/- 0.9 and 9.1 +/- 0.4 fC pF-1 (n = 4), respectively, with an effective valeney of 0.48 +/- 0.07 (n = 7). 4. Kinetics of the transient currents were independent of Vm and Tl+o at positive potentials, but became more rapid at increasingly negative Vm values in an ion concentration-dependent fashion. 5. Those data demonstrate that electrogenic steps participate in K+ transport by the Na+,K(+)-ATPase and that the electrogenic step is extracellular ion binding. 6. The temperature- and Vm-dependent properties of transient charge movements were compared under K(+) -K+ and Na(+) -Na+ exchange conditions. The data suggest that extracellular K+ and Na+ binding occur at different sites in the enzyme or to different enzyme conformations. The sum of the effective valencies, 1.14 +/- 0.12, demonstrates that the electrogenicity of extra-cellular ion binding can explain the Vm dependence of ion transport by the Na+,K(+)-ATPase.

Effect on the indo‐1 transient of applying Ca2+ channel blocker for a single beat in voltage‐clamped guinea‐pig cardiac myocytes.

1. We used rapid solution changes to investigate the mechanisms which trigger Ca2+ release from the sarcoplasmic reticulum (SR) in guinea-pig ventricular myocytes. We patch-clamped myocytes at 36 degrees C and used indo-1 to monitor intracellular Ca2+. Before each test pulse, we established a standard level of SR Ca2+ load by applying a train of conditioning pulses. 2. We switched rapidly to 32 microM nifedipine (an L-type Ca2+ current (ICa,L) blocker) 8 s before a test pulse, and just after applying nifedipine we applied a ramp depolarization to pre-block Ca2+ channels. We found that ICa,L elicited by the following test pulse was inhibited almost completely (98-99% inhibition). 3. The indo-1 transient elicited by an 800 ms depolarizing pulse showed a rapid initial rise which was inhibited by ryanodine-thapsigargin. This indicated that the rapid rise was due to Ca2+ release from the SR, and therefore provides an index of SR Ca2+ release. 4. In cells dialysed internally with 10 mM Na(+)-containing solution, nifedipine application before a +10 mV test pulse blocked 62% of the rapid initial phase of the indo-1 transient. Calibration curves of indo-1 for intracellular Ca2+ (using a KD of indo-1 for Ca2+ of either 250 or 850 nM, the reported range) indicated that between 67 and 76% of the Ca2+i transient was inhibited by nifedipine. Thus, in cells dialysed with 10 mM Na+ and depolarized to +10 mV, and in the absence of ICa,L, this suggests that another trigger mechanism for SR release is able to trigger between 33 and 24% of the Ca2+i transient. 5. For a given dialysing Na+ concentration, the fraction of indo-1 transient which was inhibited by nifedipine decreased as test potential became more positive. In cells dialysed with 10 mM Na+ and pulsed to +110 mV, 24% of the rapid phase of the indo-1 transient was inhibited by nifedipine (equivalent to between 27 and 37% of the Ca2+i transient). 6. For a given test potential, the fraction of the indo-1 transient which was inhibited by nifedipine decreased as dialysing Na+ concentration increased. In cells dialysed with Na(+)-free solution and pulsed to +10 mV, 84% of the indo-1 transient was inhibited by nifedipine (equivalent to between 88 and 91% of the Ca2+i transient). In contrast, in cells dialysed with 20 mM Na+ and pulsed to +10 mV, 41% of the indo-1 transient was inhibited by nifedipine (equivalent to between 47 and 57% of the Ca2+i transient). 7. Dialysing cells with different Na+ concentrations could lead to a different SR Ca2+ content. We therefore manipulated the conditioning train before each test pulse to change the extent of SR loading. For each dialysing Na+ concentration, we found no change in the degree to which nifedipine blocked the indo-1 transient when SR content was either increased or decreased. 8. The results support the idea that both ICa, L and a second mechanism are able to trigger SR release and the resulting Ca2+i transient. When ICa, L was blocked with nifedipine, the fraction of Ca2+i transient which remained increased with more positive test potential and higher internal Na+. This is consistent with the hypothesis that the second SR trigger mechanism is Ca2+ entry via reverse Na(+)-Ca2+ exchange, elicited by a step change in membrane potential.

Inhibition of Mg2+ current by single-gene mutation in Paramecium

"Eccentric" is a newly-isolated mutant of Paramecium tetraurelia that fails to swim backwards in response to Mg2+. In the wild type, this backward swimming results from Mg2+ influx via a Mg(2+)-specific ion conductance (IMg). Voltage-clamp analysis confirmed that, as suspected, step changes in membrane potential over a physiological range fail to elicit IMg from eccentric. Further electrophysiological investigation revealed a number of additional ion-current defects in eccentric: (i) The Ca2+ current activated upon depolarization inactivates more slowly in eccentric than in the wild type, and it requires longer to recover from this inactivation. (ii) The Ca(2+)-dependent Na+ current deactivates significantly faster in the mutant. (iii) The two K+ currents observed upon hyperpolarization are reduced by > 60% in eccentric. It is difficult to envision how these varied pleiotropic effects could result from loss of a single ion current. Rather, they suggest that the eccentric mutation affects a global regulatory system. Two plausible hypotheses are discussed.

Inositol trisphosphate‐mediated Ca2+ influx into Xenopus oocytes triggers Ca2+ liberation from intracellular stores.

1. Inositol 1,4,5-trisphosphate (InsP3) functions as a second messenger by liberating Ca2+ from intracellular stores and by promoting influx of extracellular Ca2+. We examined whether Ca2+ influx modulates intracellular Ca2+ liberation in Xenopus oocytes by fluorescence monitoring of cytosolic free Ca2+ together with voltage clamp recording of Ca(2+)-activated Cl- membrane currents. Sustained activation of membrane Ca2+ permeability was induced by intracellular injections of a non-metabolizable InsP3 analogue, 3-deoxy-3-fluoro-D-myo-inositol 1,4,5-trisphosphate (3-F-InsP3), and Ca2+ influx was controlled by applying step changes in membrane potential to alter the driving force for Ca2+ entry. 2. Negative-going potential steps evoked intracellular Ca2+ signals comprising two components; an initial transient peak followed by a slower rise. The initial transient grew steeply over a narrow (ca 40 mV) voltage range but then increased little with further polarization, whereas the second component showed a nearly linear voltage dependence. 3. The transient Ca2+ signal continued to rise almost unchanged when Ca2+ influx was interrupted by stepping the potential to more positive values after brief hyperpolarization. In contrast, Ca2+ levels declined monotonically when positive-going steps were applied after longer intervals during the second component of the Ca2+ signal. 4. Large Ca(2+)-dependent transient inward (T(in)) membrane currents were evoked during the rising phase of the initial Ca2+ signal, but little current was associated with the second component of the Ca2+ signal. 5. The T(in) currents evoked by hyperpolarization were mimicked at fixed clamp potential by re-admitting Ca2+ to the bathing solution, and by flash photolysis of caged Ca2+ loaded into the oocyte. 6. T(in) currents were strongly inhibited by prior release of Ca2+ from InsP3-sensitive intracellular stores, and vice versa. Experiments with paired hyperpolarizing pulses and paired photorelease of InsP3 showed that responses to both stimuli recovered with similar time courses. 7. We conclude that the transient Ca2+ signal and associated T(in) current evoked by hyperpolarization arise because Ca2+ entering the oocyte triggers regenerative release of Ca2+ from InsP3-sensitive intracellular stores. Since membrane currents evoked by liberated Ca2+ were much greater than those evoked by Ca2+ entry per se, a major function of InsP3-mediated Ca2+ entry may be to modulate the activity of intracellular Ca2+ stores.

Relaxation kinetics of the Na+/glucose cotransporter.

An important class of integral membrane proteins, cotransporters, couple solute transport to electrochemical potential gradients; e.g., the Na+/glucose cotransporter uses the Na+ electrochemical potential gradient to accumulate sugar in cells. So far, kinetic analysis of cotransporters has mostly been limited to steady-state parameters. In this study, we have examined pre-steady-state kinetics of Na+/glucose cotransport. The cloned human transporter (hSGLT1) was expressed in Xenopus oocytes, and voltage-clamp techniques were used to monitor current transients after step changes in membrane potential. Transients exhibited a voltage-dependent time constant (tau) ranging between 2 and 10 ms. The charge movement Q was fitted to a Boltzmann relation with maximal charge Qmax of approximately 20 nC, apparent valence z of 1, and potential V0.5 of -39 mV for 50% Qmax. Lowering external Na+ from 100 to 10 mM reduced Qmax 40%, shifted V0.5 from -39 to -70 mV, had no effect on z, and reduced the voltage dependence of tau. Qmax was independent of, but tau was dependent on, temperature (a 10 degrees C increase increased tau by a factor of approximately 2.5 at -50 mV). Addition of sugar or phlorizin reduced Qmax. Analyses of hSGLT1 pre-steady-state kinetics indicate that transfer upon a step of membrane potential in the absence of sugar is due to two steps in the reaction cycle: Na+ binding/dissociation (30%) and reorientation of the protein in the membrane field (70%). The rate-limiting step appears to be Na+ binding/dissociation. Qmax provides a measure of transporter density (approximately 10(4)/microns 2). Charge transfer measurements give insight into the partial reactions of the Na+/glucose cotransporter, and, combined with genetic engineering of the protein, provide a powerful tool for studying transport mechanisms.

Effects of cyclic nucleotides and intracellular Ca on voltage-activated ciliary beating in Paramecium

1. Coupling mechanisms between ciliary beating and the membrane potential in Paramecium were investigated under voltage clamp applying intracellular pressure injection of cAMP, cGMP and Ca-EGTA buffer. Ciliary responses following step changes in membrane potential were recorded by high-speed video on magnetic tape. 2. Injections of cAMP and cGMP up to millimolar concentrations caused no detectable changes in the frequency voltage relationship. A minor effect was that the ciliary reorientation towards the anterior cell end (reversal) tended to be inhibited with depolarization up to 10 mV. 3. Injection of Ca2+ into the cell clamped at the resting potential caused a transient anteriad ciliary reorientation and a simultaneous increase in the beating frequency. 4. Injection of EGTA (to buffer Ca2+ below 10−8M) was ineffective in relation to frequency for several minutes. After this time, hyperpolarization- and depolarization activated frequency responses of EGTA-injected cells were increasingly inhibited. The ciliary reorientation following depolarization was not affected by EGTA. 5. A posterior contraction of the cell diameter was noticed upon membrane hyperpolarization. The contraction coincided in time with the increase in beating frequency. 6. The results support the view that the voltage-dependent augmentation of the ciliary beating rate is not directly mediated by an intracellular increase in either cAMP or cGMP. 7. The role of Ca2+ as intracellular messenger in the ciliary and somatic compartments is discussed.

Potassium translocation by the Na+/K+ pump is voltage insensitive.

The voltage dependence of steady and transient changes in Na+/K+ pump current, in response to step changes in membrane potential, was investigated in guinea pig ventricular myocytes voltage clamped and internally dialyzed under experimental conditions designed to support four separate modes of Na+/K+ pump activity. Voltage jumps elicited transient pump currents when the pump cycle was running forward or backward, or when pumps were limited to Na+ translocation, but not when they were made to carry out K+/K+ exchange. This result indicates that K+ translocation involves no net charge movement across the membrane field and is therefore voltage insensitive. The transient pump currents seen during Na+/K+ transport demonstrate that both forward and reverse pump cycles are rate limited not by the voltage-dependent step but by a voltage-independent step, probably K+ translocation. These findings severely constrain kinetic models of Na+/K+ pump activity.

Transient voltage and calcium-dependent outward currents in hippocampal CA3 pyramidal neurons.

Membrane currents activated by step changes in membrane potential were studied in hippocampal pyramidal neurons of region CA3 using the single microelectrode voltage-clamp technique. The transient outward current activated by depolarizing steps appeared to be composed of two transient currents that could be distinguished by differences in voltage sensitivity, time course, and pharmacological sensitivity. The more slowly decaying current was activated by voltage steps positive to -60 mV and declined exponentially with a time constant between 200 and 400 ms. This current inactivated as the holding potential was made more positive over the range of -75 to -45 mV and was 50% inactivated near -60 mV. The more slowly decaying transient current was selectively blocked by 0.5 mM 4-aminopyridine (4-AP) but not by 5-10 mM tetraethylammonium (TEA) or 2-5 mM Mn2+. The second transient current had a much faster time course than the 4-AP-sensitive current, having a duration of 5-20 ms. This very fast transient current was observed during potential steps positive to -45 mV. The fast transient current was inactivated when the holding potential was made positive to -45 mV. The amplitude of the fast transient current was greatly reduced by the application of 4 mM Mn2+ or Ca2+-free artificial cerebrospinal fluid (CSF). The fast transient current appeared to be unaffected by 0.5 mM 4-AP but was greatly reduced by 10 mM TEA. These results suggest that the transient outward current observed during depolarizing steps is composed of at least two distinct transient currents. The more slowly decaying current resembles the A-current originally described in molluscan neurons (9, 32, 42) in voltage sensitivity, time course, and pharmacological sensitivity. The faster transient current resembles a fast, Ca2+-dependent transient current previously observed in bull-frog sympathetic neurons (5, 27).

The effect of tubocurarine competition on the kinetics of agonist action on the nicotinic receptor.

1 The rates at which tubocurarine associates with, and dissociates from, the nicotinic receptor, while exerting its classical competitive effect, are still in doubt. We have investigated this problem by observing the effect of low concentrations of tubocurarine on the re-equilibration rate, following a step change in membrane potential, of the current produced by carbachol in voltage-clamped endplates of frog muscle. 2 It is expected, and observed, that in order to see the effects of competition (as opposed to ion channel block), sufficiently high agonist concentrations must be used so that the relaxation rate becomes faster than that seen at low agonist concentrations. 3 Small concentrations of tubocurarine were found to reduce this relaxation rate, towards a value appropriate to a lower agonist concentration. 4 The results suggest that tubocurarine equilibrates very rapidly with the nicotinic receptor. 5 Some of the possible technical problems of this sort of experiment are discussed. The results are similar to those already published for nicotinic receptors in eel electric tissue.

The units of calcium conduction in Helix neurones.

VOLTAGE-DEPENDENT membrane calcium currents, ICa, occur in muscle1,2, axons3,4 and nerve cell bodies5–9 but are not well characterised because of difficulties in separating them from other ionic currents, and inadequate spatial and temporal control of the clamp voltages used to examine them. As a result, information on the conduction process underlying ICa is limited. Recently two new techniques have been reported6,9,10 which overcome most of the problems associated with experimental investigations of ICa. We have used the suction pipette technique9,10, which combines methods of internal perfusion and voltage clamp, to study ICa in isolated nerve cell bodies of the snail, Helix aspersa. The time course of ICa in response to step changes in membrane potential was readily examined and, in addition, very small fluctuations (noise) in current were also observed. Assuming that the fluctuations represent current contributions from a population of independent two-state (conducting/nonconducting) unit conductances, 〈ICa〉, the average value of the fluctuations in the steady state, and σ2, the variance of the fluctuations can be used to estimate a single unit conductance, γca. Analysis of spontaneous current noise has been used previously to obtain γ values for the Na and K systems in several axon membranes11–14, and γ values for acetylcholine-sensitive conductances at the neuromuscular junction15,16 and in molluscan neurones17, but γCa is unknown. Unit ion conductances of a few picosiemens or greater have been reported and where comparisons can be made, these values are similar to those estimated using quite different techniques such as tetrodotoxin binding18–21. We found γCa to be less than 1 pS however, and we suggest that the low value results from the effect of a Ca2+ binding site in the conductance.

Relaxation experiments using bath‐applied suberyldicholine.

1. The effect of step changes in membrane potential on the end-plate conductance change produced by bath-applied suberyldicholine was studied in voltage-clamped frog muscle fibres. 2. The suberyldicholine-induced conductance increased exponentially from its previous equilibrium level to a new equilibrium level following a step hyperpolarization. 3. For low suberyldicholine concentrations the time constant of this relaxation was independent of the concentration. 4. For low suberyldicholine concentrations the voltage dependence of equilibrium conductance and relaxation time constants was identical. 5. Bungarotoxin pretreatment did not affect the responses beyond a simple reduction in their amplitude. 6. The conductance evoked by high suberyldicholine concentrations was less voltage-sensitive than that evoked by low concentrations. 7. A new model for explaining noise and relaxation data is proposed. This postulates rate-limiting binding steps followed by a voltage-dependent isomerization.

The effects of neuraminidase and protamine chloride on potential clamp parameters of the node of Ranvier (Xenopus laevis).

The effects of neuraminidase and protamine chloride on the ionic currents associated with step changes in membrane potential were studied. Neuraminidase (a hydrolase which cleaves glycosidically bound sialic acid) had negligible effects on the permeability properties. The polycation protamine chloride decreased the sodium as well as the potassium currents. The sodium equilibrium potential decreased during the experiments which indicates a gain in internal sodium concentration. The analysis showed that the effects on the currents were caused by a decrease of the specific permeability changes. No shift along the potential axis of the potential dependent parameters was observed.

Voltage-dependence of drug-induced conductance in frog neuromuscular junction.

Membrane currents from voltage-clamped frog muscle fibers were recorded during iontophoretic application of steady doses of carbachol, acetylcholine, and suberyldicholine to the endplate region. In the presence of these drugs, an exponentially relaxing current was observed after step changes of membrane potential. The time constant of relaxation was found to be voltage-dependent. It was equal to the time constant obtained from the autocorrelation function of drug-induced conductance fluctuations measured under similar conditions. Analysis of instantaneous current at the on- and offsets of voltageclamp pulses showed that there is no shift in equilibrium potential during the pulses.

A quantitative description of end-plate currents.

1. End-plate currents have been studied in glycerol-treated frog sartorius nerve-muscle preparations with the voltage clamp technique.2. The effects of temperature on the decay rate of end-plate currents were investigated over a temperature range from 10 to 30.5 degrees C. The Q(10) for the decay constant of end-plate currents depends somewhat on membrane potential; at - 100 mV the decay constant has a Q(10) of 2.7.3. Peak end-plate current depends non-linearly on membrane potential with a decreasing slope conductance associated with hyperpolarization.4. The ;instantaneous' voltage-current relationship for end-plate channels was determined by causing step changes in membrane potential during end-plate current flow. This relationship appears to be linear.5. The interaction of acetylcholine with its receptor is viewed as being analogous to the first step in enzymic catalysis. On this view, acetylcholine binds to its receptor and induces a conformational change which is responsible for opening end-plate channels. By analogy to the first steps in the catalytic sequence of enzymes, the binding step is very rapid, almost diffusion-limited, and the conformational change is rate-limiting.6. Equations describing this process have been derived. Expressions for the rate constants have also been derived by considering changing dipole moments of the transmitter-receptor complex associated with the conformational change. As the transmitter-receptor complex is in the membrane field, different conformational states have different energies, and the rate of conformational change thus depends on membrane potential. The equations thus derived are shown to account adequately for the time course of end-plate conductance change.

Equivalent circuit of frog atrial tissue as determined by voltage clamp-unclamp experiments.

The equivalent circuit that has been used in the analysis of nerve voltage-clamp data is that of the membrane capacity in parallel with the membrane resistance. Voltage-clamp experiments on frog atrial tissue indicate that this circuit will not suffice for this cardiac tissue. The change in membrane current associated with a step change in membrane potential does not show a rapid spike of capacitive current as would be expected for the simple parallel resistance-capacitance network. Rather, there is a step change in current followed by an exponential decay in current with a time constant of about 1 msec. This relatively slow capacitive charging current suggests that there is a resistance in series with the membrane capacity. A possible equivalent circuit is that of a series resistance external to the parallel resistance-capacitance network of the cell membranes. Another possible circuit assumes that the series resistance is an integral part of the cell membrane. The data presented in this paper demonstrate that the equivalent circuit of a bundle of frog atrial muscle is that of an external resistance in series with the cell membranes.

Voltage clamp experiments in striated muscle fibres.

1. Membrane currents during step depolarizations were determined by a method in which three electrodes were inserted near the end of a fibre in the frog's sartorius muscle. The theoretical basis and limitations of the method are discussed.2. Measurements of the membrane capacity (C(M)) and resting resistance (R(M)) derived from the current during a step change in membrane potential are consistent with values found by other methods.3. In fibres made mechanically inactive with hypertonic solutions (Ringer solution plus 350 mM sucrose) step depolarizations produced ionic currents which resembled those of nerve in showing (a) an early transient inward current, abolished by tetrodotoxin, which reversed when the depolarization was carried beyond an internal potential of about +20 mV, (b) a delayed outward current, with a linear instantaneous current-voltage relation, and a mean equilibrium potential with a normal potassium concentration (2.5 mM) of -85 mV.4. The reversal potential for the early current appears to be consistent with the sodium equilibrium potential expected in hypertonic solutions.5. The variation of the equilibrium potential for the delayed current (V'(K)) with external potassium concentration suggests that the channel for delayed current has a ratio of potassium to sodium permeability of 30:1; this is less than the resting membrane where the ratio appears to be 100:1. V'(K) corresponds well with the membrane potential at the beginning of the negative after-potential observed under similar conditions.6. The variation of V'(K) with the amount of current which has passed through the delayed channel suggests that potassium ions accumulate in a space of between (1/3) and (1/6) of the fibre volume. If potassium accumulates in the transverse tubular system (T system) much greater variation in V'(K) would be expected.7. The delayed current is not maintained but is inactivated like the early current. The inactivation is approximately exponential with a time constant of 0.5 to 1 sec at 20 degrees C. The steady-state inactivation of the potassium current is similar to that for the sodium current, but its voltage dependence is less steep and the potential for half inactivation is 20 mV rate more positive.8. Reconstructions of ionic currents were made in terms of the parameters (m, n, h) of the Hodgkin-Huxley model for the squid axon, using constants which showed a similar dependence on voltage.9. Propagated action potentials and conduction velocities were computed for various conditions on the assumption that the T system behaves as if it were a series resistance and capacity in parallel with surface capacity and the channels for sodium, potassium and leak current. There was reasonable agreement with observed values, the main difference being that the calculated velocities and rates of rise were somewhat less than those observed experimentally.