Charge movements in intact amphibian skeletal muscle fibres in the presence of cardiac glycosides

Abstract

1. Intramembrane charge movements were examined in intact voltage-clamped amphibian muscle fibres following treatment with cardiac glycosides in the hypertonic gluconate-containing solutions hitherto reported to emphasise the features of q_γ at the expense of q_β charge.

2. The application of chlormadinone acetate (CMA) at concentrations known selectively to block Na⁺-K⁺-ATPase conserved the steady-state voltage dependence of intramembrane charge, contributions from delayed (q_γ) charging transients, and their inactivation characteristics brought about by shifts in holding potential.

3. The addition of either ouabain (125, 250 or 500 nm) or digoxin (5 nm) at concentrations previously reported additionally to influence excitation-contraction coupling similarly conserved the steady-state charge-voltage relationships, Q(V), in fully polarised fibres to give values of maximum charge, Q_max, transition voltage, V*, and steepness factor, k, that were consistent with a persistent q_γ component as reported on earlier occasions (Q_max≈ 25-27 nC μF⁻¹, V*≈ -45 to -50 mV, k≈ 7-9 mV).

4. In both cases shifts in holding potential from -90 to -50 mV produced a partial inactivation that separated steeply and more gradually voltage-dependent charge components in agreement with previous characterisations.

5. However, charge movements that were observed in the presence of either digoxin or ouabain were monotonic decays in which delayed (q_γ) transients could not be distinguished from the early charging records. These features persisted despite the further addition of chlormadinone acetate over a 10-fold concentration range (5-50 μm) known to displace ouabain from the Na⁺-K⁺-ATPase.

6. Ouabain (500 nm) restored the steady-state charge movement that was previously abolished by the addition of 2.0 mm tetracaine in common with previous results of using ryanodine receptor (RyR)-specific agents.

7. Perchlorate (8.0 mm) restored the delayed ‘on’ relaxations and increased the prominence of the ‘off’ decays produced by q_γ charge following treatment with cardiac glycosides. This was accompanied by a negative (≈10-15 mV) shift in the steady-state charge-voltage relationship but an otherwise conserved maximum charge, Q_max, and steepness factor, k, in parallel with previously reported effects of perchlorate following treatments with RyR-specific agents.

8. The features of cardiac glycoside action thus parallel those of other agents that act on RyR-Ca²⁺ release channels yet influence the kinetics but spare the steady-state properties of intramembrane charge.

Charge movements are thought to represent the electrical signature for conformational changes that take place within intramembrane voltage sensors at least some of which trigger calcium release in the process of excitation- contraction coupling in skeletal muscle (for review see Huang, 1993). It is thought that the latter process is initiated by the influence of voltage change on the conformation of dihydropyridine receptors (DHPRs) that are located within the membranes of the transverse tubules (Rios & Brum, 1987; Huang & Peachey, 1989; Huang, 1990; Chen & Hui, 1991; Anderson & Meissner, 1995). This in turn triggers a release of intracellularly stored Ca²⁺ by ryanodine receptor (RyR)-Ca²⁺ release channels that are located geometrically close to the transverse tubules within the terminal cisternal membranes (for reviews see Meissner, 1994; Franzini-Armstrong & Protasi, 1997).

There has been recent interest in the relationship between the conformational processes that may take place in the RyR and the charge movements that may partly reflect changes within the triggering DHPR (Huang, 1996, 1997, 1998a,b). These have particularly concerned the kinetics and steady-state properties of one particular charging component, q_γ, in view of its delayed and complex time course at membrane potentials close to the cellular activation threshold (Adrian & Peres, 1979; Huang, 1981; Hui, 1983). The steady-state distribution of the q_γ component additionally shows consistent parallels with the steep voltage dependence of the Ca²⁺ release process (Huang, 1982; Hui, 1983; Hui & Chandler, 1990, 1991; Jong et al. 1995; Pape et al. 1996). Thus, there have been suggestions that the Ca²⁺ release process might itself be responsible for some of the measured steady-state charge (Csernoch et al. 1991) or for some of its kinetic properties (Jong et al. 1995; Pape et al. 1996). Alternatively, the striking time course of the q_γ component of the charge movement might result from an allosteric or mechanical coupling between a DHPR situated within the tubular membrane field and some of the RyRs that fall outside it. Thus, pharmacological modifications of the RyR reciprocally influenced the kinetics but preserve the steady-state properties, including the total available quantity of the DHPR q_γ component of the charge movement (Huang, 1996, 1997, 1998a,b).

However, biochemical and anatomical evidence has demonstrated that amphibian skeletal muscle possesses more than one RyR type of which only the skeletal muscle isoform (RyR-I or RyR-α: Olivares et al. 1991; Maruyama & Ogawa, 1992) might be directly coupled to the DHPR voltage sensor (for review see Franzini-Armstrong & Protasi, 1997). Such a finding prompted explorations for mechanisms by which the remaining RyRs might be recruited into and amplify Ca²⁺ release. These have included suggestions for a Ca²⁺-induced Ca²⁺ release process (e.g. Klein et al. 1990; Jacquemond et al. 1991; Hollingworth et al. 1992; Csernoch et al. 1993) or a more direct slaving of such Ca²⁺ release channels by the directly coupled skeletal muscle type RyRs (e.g. Pape et al. 1995)

Recent reports suggest that cardiac glycosides might provide a useful means of investigating changes in the membrane charging processes through their action on such non-skeletal muscle-type RyRs. Cardiac glycosides with known positive inotropic actions on the mammalian heart enhanced calcium release from cardiac SR vesicles by increasing the open probabilities of the cardiac RyR-Ca²⁺ release channel when applied in nanomolar concentrations (Rardon & Wasserstrom, 1990; McGarry & Williams, 1993). In contrast, skeletal muscle-type RyRs were not affected even by micromolar concentrations of such agents. Yet, Sarkozi et al. (1996) reported that cardiac glycosides selectively enhanced the inactivating phase but spared the steady phase of sarcoplasmic reticular (SR) calcium release in response to applied depolarizing steps in amphibian skeletal muscle. Both ouabain and digoxin similarly increased Ca²⁺ release rates in isolated sarcoplasmic reticular vesicles from amphibian skeletal muscle to levels comparable to those observed with caffeine or ryanodine (Rousseau et al. 1988). In contrast, that study did not observe major actions of cardiac glycosides on the intramembrane charge. Taken together such findings supported the conclusion that the inactivating component of such Ca²⁺ release (see Pizarro et al. 1992) involved non-skeletal muscle-type ryanodine receptors.

The present paper describes more detailed studies of the effect of cardiac glycosides upon the characteristics of individual components of the intramembrane charge itself. The same agents, ouabain, digoxin and chlormadinone acetate (CMA), were applied at concentrations and under conditions that were similar to those used in the earlier report (Sarkozi et al. 1996). However, the remaining experimental conditions closely paralleled those adopted on previous occasions that investigated the effect of pharmacological RyR modification on the characteristics of the intramembrane charge in intact voltage-clamped amphibian fibres (Huang, 1996, 1997, 1998a,b). The actions of cardiac glycosides shown previously to influence RyR function were compared with the consequences of adding CMA. The latter agent is considered exclusively to affect Na⁺-K⁺-ATPase, but not the RyR, nor to exert any inotropic effects, and therefore could be applied under similar conditions to provide useful controls (LaBella et al. 1979). The findings agreed with the earlier reports in that such manoeuvres conserved steady-state charge. However, ouabain and digoxin, expected to influence non-skeletal muscle RyRs (Rardon & Wasserstrom, 1990; McGarry & Williams, 1993), but not CMA, reported only to affect Na⁺-K⁺-ATPase, modified the time course of the q_γ charge. Such effects persisted even in the presence of high CMA concentrations that would displace ouabain from such Na⁺-K⁺-ATPase. However, the cardiac glycoside actions were reversed by perchlorate, which has been reported to enhance interactions between the DHPR and RyR (Luttgau et al. 1983; Gomolla et al. 1983; Huang, 1986, 1987, 1998a). Finally, ouabain itself restored intramembrane charge previously abolished by the RyR antagonist tetracaine in parallel with previously reported effects with the twitch potentiators perchlorate and caffeine (see Huang, 1998a,b). Taken together, these findings demonstrate that the actions of cardiac glycosides parallel those of other agents that are directed towards the RyR-Ca²⁺ release channel in their influence upon the properties of the q_γ intramembrane charge. They implicate those RyRs influenced by cardiac glycosides in such feedback effects upon the charge movement.

METHODS

Sartorius muscles were dissected from cold-adapted frogs (Rana temporaria: Blades Biological, Kent, UK) previously killed by concussion followed by decapitation and pithing (UK Schedule 1 Home Office regulations) in cold (4 °C) Ringer solution. They were mounted in a temperature-controlled chamber and stretched to give centre fibre sarcomere lengths of 2.2-2.4 μm. The Ringer solution was then replaced with the following isotonic solution at the same temperature: 120 mm tetraethylammonium gluconate, 2.0 mm MgCl₂, 2.5 mm RbCl, 800 μm CaCl₂, 1.0 mm 3,4-diaminopyridine, 2 × 10⁻⁷m tetrodotoxin and 3.0 mmN-2-hydroxyethylpiperazine-N′-2-ethanesulphonic acid (Hepes) buffered to pH 7.0. After 15 min, this was further replaced by a similar solution containing 500 mm sucrose. The final test solutions that finally included the adopted test agents were introduced 15 min prior to beginning electrophysiological studies under similarly cooled (2-4 °C) conditions. The pharmacological agents were applied at the same concentrations as in earlier studies: ouabain (125, 250 or 500 nm), digoxin (5 nm), tetracaine (2 mm) and perchlorate (8.0 mm; all from Sigma Chemical Co., Poole, UK).

The three-microelectrode voltage clamp used 3-5 MΩ KCl (3 m) glass recording electrodes positioned at l= 375 μm (voltage control electrode, V₁) and 2l= 750 μm (second voltage electrode, V₂) from the pelvic ends of superficial muscle fibres. The shielded potassium citrate (2 m) current injection electrode I₀ was inserted at 5l/2 = 940 μm. Electrical signals from the clamp voltage V₁, voltage difference (V₁ - V₂), and injected current I₀(t) were filtered through low pass (1.0 kHz) 3-pole Butterworth filters, then sampled every 200 μs using a 12-bit CED-Model 1401-plus interface. The latter was controlled by a Tandon IBM-compatible computer that executed the entire process of pulse generation, data acquisition, and on-line and off-line analysis and display through user-written Signal version 1.6 scripts (Cambridge Electronic Design, Cambridge, UK). Five sweeps, spaced by intervals of 20 s, were averaged into each test or control record. Test transients were elicited by voltage steps of 124 ms duration made from -90 mV to different potentials. The control pulses superimposed +50 mV steps at a time point 300 ms following introduction of 610-ms-long prepulse steps from a -90 mV holding potential to a level of -140 mV. These regularly bracketed successive sets of four to six test acquisitions. Steady-state values of V₁(t), V₂(t) and the injected current I₀(t) were all attained well before the end of such control steps. These were accordingly determined directly from the experimental records without the sloping baseline corrections required in some cut fibre preparations. Fibre length constants, λ, internal longitudinal resistances, r_i, and membrane resistances of unit fibre length, r_m, were then computed on-line from each individual averaged record. Calculations of fibre diameters, d, and specific membrane resistances, R_m, assumed an internal sarcoplasmic resistivity, R_i, of 391 Ω cm in 2.5 times hypertonic solution at 2 °C, and a temperature coefficient of 0.73. The calculated cable constants made it possible to assess fibre stability and condition over time. The membrane current as a function of time t through unit fibre surface area I_m(t) was then calculated from the equation:

Charge movements were derived by subtracting test traces from records obtained from control records constituted from the depolarizing and hyperpolarizing responses not only to the control steps but also their associated prepulses. This effectively doubled the level of signal averaging, with an enhancement of signal-to-noise ratio. The derived traces were then appropriately scaled to the ratio of the respective amplitudes of the test to the control voltage steps. Similar scalings and subtractions were also applied to the averaged records of the test and the control voltage (V₁) steps to further verify that the derived charge movements indeed reflected non-linear contributions to fibre electrical properties. In addition, any small change in linear membrane properties that might have taken place over the course of each set of test voltage steps was also corrected for. Thus, the final control records with which test/control comparisons were made were constituted from a weighted mean of the two bracketing control records, each weighted using the position of the relevant test average within the bracketed test sequence. The resulting charge movements are plotted at high gain to best discriminate the presence or absence of delayed (q_γ) charge movement even at the expense of a full display of the earlier exponential decays.

All the results of the steady-state determinations are expressed as means + standard errors of the mean (s.e.m.). The steady-state charge-voltage data were described using a single two-state Boltzmann relationship between the steady-state charge movement, Q(V), the maximum charge, Q_max, the transition voltage, V*, and the steepness factor, k:

Some of the data were fitted to the sum of two rather than one Boltzmann function as described by the differential sensitivities of Q_β and Q_γ charge to sustained depolarization to a holding potential of -50 mV (Huang, 1994b, 1996):

All these functions were optimized using a Levenberg-Marquadt algorithm (Origin, MicroCal, MA, USA) that performed successive least-squares minimisations of the values of each of the parameters a_j of a generalized non-linear function y(x) (see also Huang, 1996). This was performed simultaneously over all the experimentally obtained mean values y(x_i) as obtained at each ith test voltage x_i. The successive iterations minimized the values of chi-square, χ², in a weighted fit derived from the mean and standard errors of the data points y_i.

RESULTS

Both the steady-state and kinetic properties of charge movements were examined using cardiac glycoside concentrations that were comparable to those hitherto employed upon intramembrane charge and the release of intracellularly stored Ca²⁺ in an earlier study (Sarkozi et al. 1996). All electrophysiological studies were performed within 50 min of the addition of cardiac glycosides to the extracellular solutions. The remaining experimental conditions and pulse protocols paralleled those adopted in recent studies on the effect of RyR-specific agents on such charge movements (Huang, 1996, 1998a,b). Thus, fibres were studied in 3,4-diaminopyridine and tetraethylammonium-containing solutions that largely replaced extracellular Cl⁻ with gluconate and most of the Ca²⁺ with Mg²⁺. This specific choice of experimental solutions reduced q_β relative to q_γ charge (Hui & Chen, 1992), reduced Cl⁻ contributions to the overall membrane leak conductance and minimized any time-dependent Ca²⁺ and K⁺ currents. The ON charging currents that were obtained under such conditions decayed to stable baselines and are accordingly displayed following only simple direct current corrections based on the final 20 ms of the ON records. Intervening time-dependent currents, particularly the inward current phases reported to follow the slow q_γ currents (see Csernoch et al. 1991) were not observed.

Cardiac glycoside treatment conserves the steady-state dependence of intramembrane charge upon test potential in fully polarised fibres

Figure 1 plots the charge-voltage data that were obtained in the presence of 5 μm chlormadinone acetate (CMA; open squares) and progressively increasing concentrations of ouabain; the relevant experimental values are plotted as means ±s.e.m. The applied concentration of chlormadinone acetate (5.0 μm) has been reported both to inhibit Na⁺-K⁺-ATPase and to displace ouabain from such Na⁺-K⁺-ATPase in the absence of any inotropic effect in cardiac muscle (LaBella et al. 1979; Sarkozi et al. 1996). Ouabain was applied at concentrations (125, 250 and 500 nm, respectively) previously reported either to leave Ca²⁺ transients unaltered (≈100 nm), or to potentiate both mechanical activity (Fujino & Fujino, 1982) and Ca²⁺ transients in skeletal muscle (250-500 nm; Sarkozi et al. 1996) in addition to its known effects on Na⁺-K⁺-ATPase activity. The results obtained in the presence of chlormadinone acetate therefore provided useful controls for assessment of the consequences of ouabain action on RyR function.

Figure 1. Cardiac glycoside treatment conserves the steady-state intramembrane charge.

Effects of chlormadinone acetate (CMA; 5.0 μm) and increasing concentrations (125, 250 and 500 nm) of ouabain on charge-voltage curves in fully polarised intact fibres (V_H= -90 mV) and after partial depolarisation of the holding potential (V_H= -50 mV; diamonds and hexagons). Five fibres exposed to 5.0 μm CMA (squares and diamonds): temperature = 5.7 ± 0.04 °C; R_i= 388.3 ± 0.49 Ω cm, λ= 1.93 ± 0.099 mm, r_i= 10089 ± 1731 kΩ cm⁻¹, d= 72.7 ± 5.32 μm, r_m= 372.3 ± 70.71 kΩ cm, R_m= 8.13 ± 0.917 kΩ cm², C_m= 9.8 ± 0.86 μF cm⁻². Four fibres exposed to 125 nm ouabain (inverted triangles): temperature = 4.6 ± 0.07 °C; R_i= 402.8 ± 0.93 Ω cm, λ= 1.62 ± 0.160 mm, r_i= 8755 ± 1811 kΩ cm⁻¹, d= 81.0 ± 7.32 μm, r_m= 208.4 ± 18.19 kΩ cm, R_m= 5.28 ± 0.684 kΩ cm², C_m= 10.3 ± 1.13 μF cm⁻². Four fibres exposed to 250 nm ouabain (triangles): temperature = 6.2 ± 0.06 °C; R_i= 383.0 ± 0.63 Ω cm, λ= 1.82 ± 0.183 mm, r_i= 10778 ± 1545 kΩ cm⁻¹, d= 69.8 ± 5.88 μm, r_m= 331.0 ± 22.82 kΩ cm, R_m= 7.33 ± 0.900 kΩ cm², C_m= 10.4 ± 1.74 μF cm⁻². Six fibres exposed to 500 nm ouabain (circles and hexagons): temperature = 6.2 ± 0.06 °C; R_i= 383.0 ± 0.63 Ω cm, λ= 1.99 ± 0.124 mm, r_i= 9341 ± 894 kΩ cm⁻¹, d= 73.6 ± 4.47 μm, r_m= 350.4 ± 18.42 kΩ cm, R_m= 8.14 ± 0.618 kΩ cm², C_m= 9.9 ± 1.56 μF cm⁻².

Steady-state results obtained following such interventions using cardiac glycosides closely resembled earlier reports of an available charge that increased with progressive depolarisation from the -90 mV holding level. This charge- voltage curve then crossed an inflexion at test potentials between -45 and -50 mV before increasing to a maximum value of around 25-27 nC μF⁻¹ between test voltages of -30 and 0 mV (Huang, 1996). Figure 1 thus demonstrates that pharmacological interventions involving either agent preserved the steady-state charge, the steepness of its voltage dependence and the position of the charge-voltage curves along the voltage axis.

Table 1 thus summarises results of the least-squares minimisations of the steady-state charge-voltage data that were obtained at each ouabain or chlormadinone concentration to an equation for a single two-state Boltzmann system. It confirms that introduction of either agent conserved the steady-state values of the maximum charge, Q_max, to values between 25 and 27 nC μF⁻¹, values if anything that were slightly higher than those obtained on earlier occasions that used comparable gluconate-containing extracellular solutions (around 20 nC μF⁻¹: Huang, 1994a,b, 1996). Similarly neither chlormadinone acetate nor ouabain affected the steepness factors, k, which remained at around 7-9 mV. These were thus conserved to the strong voltage sensitivities similarly reported in intact fibres studied in gluconate-containing solutions in the absence of such agents (7-9 mV: Huang, 1996, 1997, 1998a,b; Jong et al. 1995). Such a finding contrasts with the steepness factors of 16-17 mV reported by Sarkozi et al. (1996) in cut fibres. Transition potentials in the presence of ouabain fell between -45 and -50 mV, if anything corresponding to small positive shifts in transition voltages, V*, in contrast to the negative shifts reported in an earlier study (Sarkozi et al. 1996). These overall similarities to findings obtained in untreated fibres whether in Vaseline gap or microelectrode voltage clamp preparations (Hui & Chandler, 1990, 1991; Huang, 1994a,b, 1996; Jong et al. 1995) strongly suggest that both q_β and q_γ charge contributions persisted under the present pharmacological conditions.

Table 1.

The action of chlormadinone acetate and ouabain on the steady-state properties of the intramembrane charge

Conditions	No. of fibres (n)	Qmax (nC μF_1)	k (mV)	V * (mV)
CMA (5.0 μM)	5	25.4 ± 0.38	7.5 ± 0.49	−52.4 ± 0.52
Ouabain (125 nM)	4	26.7 ± 0.80	9.4 ± 0.62	−45.9 ± 0.79
Ouabain (250 nM)	4	27.2 ± 0.74	8.9 ± 0.82	−46.5 ± 0.93
Ouabain (500 nM)	6	27.2 ± 0.55	8.2 ± 0.49	−49.2 ± 0.62

The action of chlormadinone acetate (5.0 μM) and ouabain (125, 250 and 500 nM) on the steady-state properties of the intramembrane charge as a whole as described by a single two-state Boltzmann function given by an equation of the form Q(V) =Q_max/{1 + exp[(-V- V*)/k]} in fibres studied in gluconate-containing solutions. Values of the maximum charge, Q_max, the transition voltage, V*, and the steepness factor, k, are given as means ± S.E.M.

Figure 1 shows the results of a shift in holding potential from -90 to -50 mV whilst continuing to impose the test voltage steps from a level of -90 mV at a time 300 ms after interposition of a 610 ms long prepulse. This produced an inactivation of the charge movement, in the presence of either ouabain or CMA, similar to that reported on an earlier occasion in similar solutions (Huang, 1994b). Such a manoeuvre left considerably shallower charge- voltage curves and a reduced maximum charge. It thus separated a steeply voltage-sensitive intramembrane component, on earlier occasions attributed predominantly to q_γ charge, from the more gradual voltage dependence of the q_β charge that remained. In the presence of chlormadinone acetate, shifting the holding voltage from -90 to -50 mV left values of Q_max= 11.5 ± 0.46 nC μF⁻¹, V *= -41.1 ± 2.16 mV and k= 19.6 ± 1.38 mV (means ±s.e.m.; n= 5 fibres). In the presence of 500 nm ouabain the corresponding values were similar: Q_max= 12.9 ± 0.85 nC μF⁻¹, V*= -49.5 ± 3.45 mV and k= 17.1 ± 2.74 mV (n= 5 fibres). The kinetic and steady-state properties of charge in such depolarised fibres in the presence of cardiac glycosides thus closely resembled records that have been described in earlier studies in which such a procedure selectively inactivated q_γ whilst sparing q_β charge in gluconate-containing solutions (Huang, 1994b).

It was further possible to derive an indication of the properties of the inactivated, mainly q_γ charge, by comparing the charge-voltage data before and after such inactivation as has been performed on previous occasions (Huang, 1994b, 1996). These predicted a voltage dependence, in particular values of the slope factor, k, that were consistent with the higher voltage sensitivity reported for the q_γ system on earlier occasions. Thus the derived q_γ charge-voltage curve for fibres in chlormadinone acetate gave: Q_max= 13.9 ± 0.50 nC μF⁻¹, V*= -51.9 ± 1.03 mV and k= 4.7 ± 0.94 mV; fibres in 125 nm ouabain gave: Q_max= 16.2 ± 0.42 nC μF⁻¹, V*= -46.3 ± 0.64 mV and k= 6.3 ± 0.59 mV. The derived features for q_γ charge in the presence of 250 nm and 500 nm ouabain similarly were: Q_max= 17.0 ± 0.42 nC μF⁻¹, V*= -46.6 ± 0.70 mV and k= 5.6 ± 0.61 mV, and Q_max= 17.9 ± 0.45 nC μF⁻¹, V*= -49.0 ± 0.67 mV and k= 5.7 ± 0.55 mV, respectively.

Delayed (q_γ) charge transfers persist following addition of chlormadinone acetate

The conservation of steady-state properties that were described above could next be contrasted against the presence or absence of any kinetic effects shown by such cardiac glycosides. These concerned the visibility or otherwise of delayed (q_γ) kinetic components that would normally contribute ‘humps’ or ‘shoulders’ to the charge movements (Adrian & Peres, 1979; Huang, 1981; Hui, 1983). Figure 2A displays typical charge movements obtained from voltage-clamped muscle fibres that were studied in the presence of 5 μm chlormadinone acetate reported earlier to influence Na⁺-K⁺-ATPase activity (see above; LaBella et al. 1979). The fibres were subject to applied voltage clamp steps to a series of progressively depolarised test levels, V_T, from a fixed holding potential of -90 mV. These steps were incremented in close, 5 mV, or even 2-3 mV, intervals in order to examine closely for the existence of and changes in steeply voltage-sensitive delayed q_γ charge movement components hitherto reported to contribute delayed capacity transients. Such derived charging records are plotted at high gain specifically to emphasise such current components even if at the expense of earlier but larger q_β component decays.

Figure 2. Chlormadinone acetate conserves delayed q_γ charge movement.

Charge movements in a fibre treated with CMA before (A) and after (B) shifts of the holding voltage from -90 to -50 mV. A and B: temperature = 5.8 °C; R_i= 387.6 Ω cm, λ= 1.98 mm, r_i= 16559 kΩ cm⁻¹, d= 54.6 μm, r_m= 650.8 kΩ cm, R_m= 11.16 kΩ cm², C_m= 7.69 μF cm⁻².

Whereas the smaller voltage steps elicited relatively little charging current, larger voltage steps that imposed depolarisations to levels as positive as V_T= -50 mV elicited the simple exponential current decays that have been previously attributed to q_β charge (Huang, 1982). Still larger voltage excursions gave rise to the delayed shoulders in the charging transients that have been attributed to q_γ charge transfer (Adrian & Peres, 1979; Huang, 1982; Hui, 1983). The appearance of such ON currents around threshold voltages between -45 and -50 mV was accompanied by an increase in the initial amplitudes of the corresponding OFF tail currents. The waveforms of such ON shoulders were sharply voltage sensitive: small further depolarisation to around -40 mV gave rise to more rapid current decays whence the two component transients merged. Beyond this, voltage steps close or positive to -35 mV left simple exponential decays in which individual charge movement components could not be distinguished.

Figure 2B goes on to demonstrate that shifts in holding potential from -90 to -50 mV in chlormadinone-treated muscle fibres selectively inactivated intramembrane charge in close agreement with the q_γ inactivation reported on earlier occasions in similar solutions when such pharmacological agents were absent (Huang, 1994b). It displays typical charge movements in response to depolarising steps to potentials, V, between -80 and 0 mV. The test steps were imposed 300 ms after the fibre was returned to a prepulse level of -90 mV from a shifted holding potential of -50 mV. The resulting charge movements resembled simple monotonic decays but were significantly reduced in amplitude and voltage dependence. This procedure left the considerably shallower charge- voltage curves shown in Fig. 1, and a reduced maximum charge around 10 nC μF⁻¹. This suggested kinetic, steady state, and inactivation properties of charge in CMA-treated fibres that closely resembled corresponding features defined in earlier studies that established the differential features of q_γ and q_β charge (Huang, 1994b).

Ouabain treatment produces a merger of q_β and q_γ waveforms

Figure 3A-C shows the effect of a range of ouabain concentrations upon the intramembrane charge, in muscle fibres subject to similarly incremented test voltage steps. The ouabain was applied at concentrations previously reported either to leave Ca²⁺ transients unaltered (≈125 nm), or to potentiate both mechanical activity (Fujino & Fujino, 1982) and Ca²⁺ transients (250 and 500 nm: Sarkozi et al. 1996) in addition to inhibiting Na⁺-K⁺-ATPase activity. The lowest concentration (125 nm) left some evidence of delayed q_γ contributions to charging currents. The higher concentrations (250 and 500 nm) left only rapid decays but the same total steady-state charge transfer. Thus the smaller voltage steps up to -50 mV elicited the expected decays of q_β charge; larger voltage steps would have been expected to elicit delayed q_γ currents, but these were significantly less prominent even at the lowest applied ouabain concentration (Fig. 3A). The higher ouabain concentrations further accelerated the time courses of q_γ currents with the result that their contributions could not be separated from the remaining q_β current decays.

Figure 3. Ouabain modifies the kinetics of the intramembrane charge.

Charging kinetics in the presence of progressively higher (125 (A), 250 (B) and 500 nm (C)) concentrations of ouabain. A, fibre in 125 nm ouabain: temperature = 6.5 °C; R_i= 379.1 Ω cm, λ= 1.79 mm, r_i= 10678 kΩ cm⁻¹, d= 67.2 μm, r_m= 340.2 kΩ cm, R_m= 7.19 kΩ cm², C_m= 12.11 μF cm⁻². B, fibre in 250 nm ouabain: temperature = 6.0 °C; R_i= 385.1 Ω cm, λ= 1.49 mm, r_i= 18717 kΩ cm⁻¹, d= 51.2 μm, r_m= 415.9 kΩ cm, R_m= 6.69 kΩ cm², C_m= 12.92 μF cm⁻². C, fibre in 500 nm ouabain: temperature = 7.0 °C; R_i= 373.2 Ω cm, λ= 1.27 mm, r_i= 6069.6 kΩ cm⁻¹, d= 88.5 μm, r_m= 97.4 kΩ cm, R_m= 2.71 kΩ cm², C_m= 5.44 μF cm⁻².

Figure 3B and C thus displays charge movements from fully polarised fibres exposed to 250 or 500 nm ouabain and again subject to graded depolarising test steps. It displays the results of exploring a full potential range between -90 and 0 mV in progressive 5 mV increments. The small depolarising steps transferred little charge. Larger voltage steps yielded progressively more prominent charge transfers, but nowhere did these show significant contributions from delayed transients even at the relatively high display magnifications adopted here. Both the ON and OFF charge movements thus consisted exclusively of monotonic decays to leave DC baselines without evidence of any appreciable delayed outward current. The features of such charge movements thus paralleled findings in similar intact fibre preparations after their treatment with other RyR-specific agents such as ryanodine and daunorubicin, or caffeine (Huang, 1996, 1998a,b). The findings contrast with recent reports from single cut fibres that reported that cardiac glycosides exerted relatively little effect upon the charge movement and therefore attributed its effects to changes in a Ca²⁺-activated release of intracellularly stored Ca²⁺ (Sarkozi et al. 1996).

The kinetic effects of ouabain on intramembrane charge persist in the presence of chlormadinone acetate

LaBella et al. (1979) pointed out that chlormadinone displaces ouabain from its binding site on Na⁺-K⁺-ATPase, inhibits the latter's activity, yet exerts no positive inotropic effect. Sarkozi et al. (1996) went on to demonstrate that the drug influenced neither resting [Ca²⁺] nor Ca²⁺ transients in cut skeletal muscle fibres. Furthermore, when ouabain (250 nm) was applied together with CMA, the effects were essentially the same as when ouabain was applied alone. Figure 4 displays typical charge movements following the addition of 250 nm ouabain to the extracellular solution surrounding intact fibres that had already been treated with 5 μm (A) or 50 μm CMA (B). Closely incremented steps to test voltages between -80 and 0 mV then gave simple exponential current decays for both ON and OFF charge movements that progressively increased in amplitude with depolarisation until charge saturation was reached between -30 and 0 mV under both conditions. Again it was not possible to distinguish delayed transients that have been identified with q_γ charge transfer (Adrian & Peres, 1979; Huang, 1982; Hui, 1983).

Figure 4. The actions of ouabain on charge movements persist in CMA-treated fibres.

Charge movements in ouabain (250 nm)-treated fibres in the presence of 5 (A) or 50 μm CMA (B) known to displace ouabain from Na⁺-K⁺-ATPase and charge-voltage curves obtained in fibres treated with both ouabain (250 nm) and 0 μm (triangles), 5 μm (inverted triangles) and 50 μm CMA (hexagons) (C). A, fibre in the presence of 5.0 μm CMA and 250 nm ouabain: temperature = 4.4 °C; R_i= 405.0 Ω cm, λ= 1.92 mm, r_i= 6422 kΩ cm⁻¹, d= 89.6 μm, r_m= 235.5 kΩ cm, R_m= 6.63 kΩ cm², C_m= 10.81 μF cm⁻². B, fibre in the presence of 50 μm CMA and 250 nm ouabain: temperature = 4.4 °C; R_i= 405.0 Ω cm, λ= 1.56 mm, r_i= 8757 kΩ cm⁻¹, d= 76.7 μm, r_m= 213.7 kΩ cm, R_m= 5.15 kΩ cm², C_m= 9.72 μF cm⁻². C, four fibres exposed to 250 nm ouabain (triangles): temperature = 6.2 ± 0.06 °C; R_i= 383.0 ± 0.63 Ω cm, λ= 1.82 ± 0.183 mm, r_i= 10778 ± 1545 kΩ cm⁻¹, d= 69.8 ± 5.88 μm, r_m= 331.0 ± 22.82 kΩ cm, R_m= 7.33 ± 0.900 kΩ cm², C_m= 10.4 ± 1.74 μF cm⁻². Four fibres in 5 μm CMA and 250 nm ouabain (inverted triangles): temperature = 4.5 ± 0.07 °C; R_i= 403.4 ± 0.95 Ω cm, λ= 2.66 ± 0.431 mm, r_i= 5328 ± 1194 kΩ cm⁻¹, d= 103.0 ± 9.85 μm, r_m= 333.5 ± 48.48 kΩ cm, R_m= 11.24 ± 2.502 kΩ cm², C_m= 8.9 ± 1.10 μF cm⁻². Five fibres in 50 μm CMA and 250 nm ouabain (hexagons): temperature = 4.6 ± 0.07 °C; R_i= 402.8 ± 0.93 Ω cm, λ= 1.88 ± 0.162 mm, r_i= 7504 ± 1743 kΩ cm⁻¹, d= 89.3 ± 9.83 μm, r_m= 242.0 ± 31.80 kΩ cm, R_m= 6.53 ± 0.769 kΩ cm², C_m= 9.8 ± 0.93 μF cm⁻².

Figure 4C demonstrates that such manoeuvres similarly conserved the overall steady-state charge and its voltage dependence. It compares charge-voltage plots that were obtained in fibres treated with ouabain alone (250 μm: triangles), with the results of further adding increasing CMA concentrations (5 and 50 μm) in such ouabain-treated fibres. All three cases gave similar Boltzmann parameters. Thus, the four fibres treated with 250 nm ouabain alone gave: Q_max= 27.2 ± 0.74 nC μF⁻¹, k= 9.0 ± 0.82 mV and V*= -46.5 ± 0.93 mV. The corresponding parameters in four fibres following addition of 5 μm CMA were: Q_max= 26.0 ± 0.59 nC μF⁻¹, k= 8.3 ± 0.65 mV and V*= -45. 5 ± 0.74 mV. Increasing the CMA concentration to 50 μm in five fibres gave: Q_max= 25.3 ± 0.35 nC μF⁻¹, k= 8.9 ± 0.41 mV and V*= -45.6 ± 0.47 mV. Thus, values of maximum charge Q_max, transition potential, V*, and the steepness factors, k, remained close in all three experimental conditions.

Similar effects of 5 nm digoxin on charge movements

Sarkozi et al. (1996) reported that digoxin, like ouabain, also potentiated the rate of calcium release, particularly the early inactivating component from the sarcoplasmic reticulum, following application of voltage steps. Figure 5 summarises experiments that complete the comparison between the present and the earlier study. They demonstrated that digoxin exerts similar effects as ouabain upon the time course of the intramembrane charge movement that was obtained in response to depolarising test steps that explored the entire potential range between -80 and 0 mV in progressive 5 mV increments (Fig. 5A). Thus the ON and the OFF charge movements again appeared only as monotonic decays that increased with voltage excursion until charge saturation: delayed q_γ components were absent. The currents similarly decayed completely to clearcut DC baselines. Figure 5B summarises the result of a charge- voltage plot based on six fibres studied in the presence of 5 nm digoxin that gave similar characteristics as in the previous experiments of: Q_max= 24.1 ± 0.26 nC μF⁻¹, k= 9.9 ± 0.33 mV and V*= -44.8 ± 0.38 mV.

Figure 5. Similar effects of digoxin on charge movements.

Typical charge movements obtained in the presence of 5 nm digoxin (A) and steady-state charge-voltage curves (B). A, fibre in the presence of 5 nm digoxin: temperature = 6.0 °C; R_i= 385.1 Ω cm, λ= 1.90 mm, r_i= 10354 kΩ cm⁻¹, d= 68.8 μm, r_m= 374.5 kΩ cm, R_m= 8.10 kΩ cm², C_m= 9.6 μF cm⁻². B, six fibres in 5 nm digoxin: temperature = 5.3 ± 0.39 °C; R_i= 393.7 ± 4.80 Ω cm, λ= 1.87 ± 0.22 mm, r_i= 8960 ± 2284 kΩ cm⁻¹, d= 82.2 ± 8.22 μm, r_m= 294.1 ± 62.45 kΩ cm, R_m= 7.15 ± 1.553 kΩ cm², C_m= 15.2 ± 2.85 μF cm⁻².

Ouabain antagonises the steady-state and kinetic effects of tetracaine

Recent studies have reported that the RyR-specific agents perchlorate and caffeine antagonise tetracaine in its inhibitory effect on q_γ charge movement compatible with either competitive actions between these agents upon a common RyR or separate actions upon reciprocally interacting intramembrane DHPRs and RyRs (Huang, 1998a,b). Figure 6 demonstrates results of similar experiments that explored for interactions between the actions of ouabain and an otherwise fully effective, 2.0 mm, tetracaine concentration. Figure 6A confirms that tetracaine abolished the delayed q_γ transient to leave exponential q_β decays of reduced amplitude when applied by itself (see Huang, 1982). Figure 6B displays charging records that were obtained from fibres that were exposed to both 500 nm ouabain and 2.0 mm tetracaine following successively larger depolarizing steps from the -90 mV holding potential. In common with previous findings in tetracaine-treated fibres with either 8 mm perchlorate or 0.2 mm caffeine (Huang, 1998a,b), ouabain restored the delayed q_γ transients to families of such records. The restored q_γ currents (Fig. 6B) appeared at more positive test potentials (-30 to -20 mV) than in control fibres in common with such earlier findings.

Figure 6. Ouabain antagonises the effects of tetracaine on charge movements.

Charge movements obtained in a tetracaine-treated fibre before (A) and following (B) addition of 500 nm ouabain. C, ouabain restores steady-state charge in tetracaine-treated fibres. A and B, fibre studied using both 2 mm tetracaine and 500 nm ouabain: temperature = 3.2 °C; R_i= 420.6 Ω cm, λ= 2.24 mm, r_i= 6614 kΩ cm⁻¹, d= 90.0 μm, r_m= 333.2 kΩ cm, R_m= 9.42 kΩ cm², C_m= 8.73 μF cm⁻². C, four fibres studied using both 2 mm tetracaine and 500 nm ouabain: temperature = 3.5 ± 0.26 °C; R_i= 416.4 ± 3.42 Ω cm, λ= 2.66 ± 0.197 mm, r_i= 34842 ± 3037 kΩ cm⁻¹, d= 99.8 ± 3.46 μm, r_m= 2621.3 ± 230.67 kΩ cm, R_m= 23.09 ± 2.735 kΩ cm², C_m= 9.6 ± 1.73 μF cm⁻².

Such effects of ouabain reflected a restoration of steady-state charge previously reduced by tetracaine treatment. Figure 6C displays the corresponding effects of ouabain upon the corresponding steady-state charge-voltage curves obtained from tetracaine-treated fibres. The data were described in terms of their approximations to two-state Boltzmann functions (see Jong et al. 1995; Huang, 1996). By itself, tetracaine more than halved the available intramembrane charge and sharply reduced its voltage dependence to give Q_max= 9.9 ± 0.35 nC μF⁻¹, V*= -59.1 ± 1.60 mV and k= 10.9 ± 1.42 mV (n= 4 fibres). However, the introduction of 500 nm ouabain restored the maximum charge (to give Q_max= 24.2 ± 0.77 nC μF⁻¹, V*= -38.4 ± 1.31 mV and k= 12.3 ± 0.96 mV; n= 4); a comparison of the resulting charge- voltage curves directed at a separation of individual steady-state q_β and q_γ Boltzmann terms yielded steepness values that would be compatible with a persistent q_γ charge: Q_max= 13.0 ± 0.95 nC μF⁻¹, V*= -30.5 ± 2.18 mV and k= 6.7 ± 1.80 mV, albeit with the same positive shift in the charge-voltage curves as described in the earlier experiments (Huang, 1998a,b). These findings concerning the effects of ouabain thus paralleled recent reports on the interactions between perchlorate and caffeine at low (0.2 mm) concentrations with tetracaine at the level of the intramembrane charge (Huang, 1998a,b).

Perchlorate restores the delayed kinetics of q_γ charge movement previously modified by ouabain

The actions of cardiac glycosides on the kinetics of the intramembrane charge thus resemble those of other agents thought to act on the RyRs to which they might be somehow coupled. Figure 7 displays typical charge movements from fibres exposed to 500 nm ouabain following a further addition of 8.0 mm perchlorate. Maximally effective concentrations (8.0 mm) of perchlorate restored all the kinetic and the steady-state properties of charge altered by prior treatment with ouabain. The charge movements regained the pattern shown in fibres that were exposed to perchlorate alone (Huang, 1998a). Thus delayed ON ‘hump’ currents reappeared and did so with relatively small depolarising steps to test voltages around -60 mV. They appeared at test voltages between -60 and -45 mV even when significant early q_β decays were absent and were accompanied by prolonged OFF tails (Fig. 7A). Finally, both the early q_β and the later q_γ current components were distinct in records obtained through a wide voltage range between test potentials of -40 and -20 mV. As in previous work, perchlorate shifted the position of the steady-state charge-voltage relationship (Fig. 7B) in the negative direction and enhanced the voltage sensitivity of the intramembrane charge, giving steady-state values compatible with contributions from steeply voltage-dependent q_γ charge: Q_max= 24.1 ± 0.18 nC μF⁻¹, V*= -57.5 ± 0.30 mV and k= 6.3 ± 0.29 mV (Huang, 1986, 1987, 1998a).

Figure 7. Perchlorate obliterates the effects of ouabain on intramembrane charge.

Charge movements in (500 nm) ouabain-treated fibres following treatment with 8 mm perchlorate (A) and a comparison of charge-voltage curves obtained in ouabain-treated fibres in the absence (○) and presence (•) of perchlorate (B). A, fibre in the presence of perchlorate: temperature = 4.2 °C; R_i= 407.6 Ω cm, λ= 1.63 mm, r_i= 6140 kΩ cm⁻¹, d= 91.9 μm, r_m= 162.6 kΩ cm, R_m= 4.70 kΩ cm², C_m= 8.7 μF cm⁻². B, four fibres studied using perchlorate: temperature = 4.2 ± 0.01 °C; R_i= 407.6 ± 0.001 Ω cm, λ= 1.78 ± 0.112 mm, r_i= 5972 ± 1207 kΩ cm⁻¹, d= 98.5 ± 11.57 μm, r_m= 180.9 ± 30.29 kΩ cm, R_m= 5.34 ± 0.568 kΩ cm², C_m= 7.3 ± 0.93 μF cm⁻².

DISCUSSION

Excitation-contraction coupling in skeletal muscle is thought to involve a triggering of intracellularly stored Ca²⁺ release through terminal cisternal RyR channels following initial activation of the DHPR voltage sensor. The latter transition has often been identified with a steeply voltage-sensitive intramembrane q_γ charge movement (Huang, 1982; Hui, 1983). The present study investigated the effects of cardiac glycosides on its steady state and kinetic properties. It was prompted by recent evidence that the cardiac glycosides digoxin and ouabain might influence excitation-contraction coupling through acting upon some RyR release channels (Fujino & Fujino, 1982), and that they might do so even in amphibian skeletal muscle. This is in addition to their known effects upon Na⁺-K⁺-ATPase activity. Thus, nanomolar concentrations of cardiac glycosides known to exert positive cardiac inotropic actions also enhanced calcium fluxes across the membranes of cardiac SR vesicles by increasing the open probabilities of cardiac RyR-Ca²⁺ release channels (Rardon & Wasserstrom, 1990; McGarry & Williams, 1993). Yet even micromolar concentrations did not so influence skeletal muscle-type (RyR-I) ryanodine receptors. However, biochemical and anatomical evidence suggests that amphibian skeletal muscle possesses multiple RyR subtypes (Olivares et al. 1991; Maruyama & Ogawa, 1992) of which only the skeletal muscle isoform may be directly coupled to the DHPR voltage sensor (for review see Franzini-Armstrong & Protasi, 1997). These findings prompted questions as to whether the remaining RyRs were also recruited into the Ca²⁺ release process, and if so, whether this requires Ca²⁺-induced Ca²⁺ release (e.g. Klein et al. 1990; Jacquemond et al. 1991; Hollingworth et al. 1992; Csernoch et al. 1993) or their more direct slaving to the mechanically coupled skeletal muscle-type RyRs (e.g. Pape et al. 1995).

In the latter connection Sarkozi et al. (1996) reported that digoxin increased both the developed tension and the early inactivating, but not the steady, component of Ca²⁺ release following application of depolarising voltage clamp steps in cut amphibian skeletal muscle. Such effects occurred at concentrations similar to the levels that were effective upon isolated cardiac-type RyR-Ca²⁺ release channels. Cardiac glycosides produced no accompanying changes in observed resting cytosolic [Ca²⁺], the voltage threshold for the appearance of measurable Ca²⁺ increase or sarcoplasmic reticular Ca²⁺ content. Intracellularly and extracellularly applied ouabain or digoxin all gave similar effects. Thus, although 50-100 nm extracellular ouabain little influenced Ca²⁺ transients, 250 nm ouabain reproduced all the effects on excitation-contraction coupling seen with digoxin. Studies of Ca²⁺ fluxes through heavy sarcoplasmic reticular vesicles suggested that such submicromolar digoxin and ouabain concentrations promoted opening of Ca²⁺ release channels to give maximum rates of release that were potentially comparable to those elicited by high caffeine concentrations. However, unlike the situation following caffeine treatment (Klein et al. 1990), cardiac glycosides permitted the membrane potential to exert a continued control on Ca²⁺ release: they did not alter the parameters that described the [Ca²⁺] decline which followed termination of the depolarizing pulse.

However, Sarkozi et al. (1996) did not observe major alterations in the intramembrane charge following such treatment with cardiac glycosides. They accordingly suggested that the inactivating component of such Ca²⁺ release involved non-skeletal muscle-type RyRs that resembled the cardiac RyR isoform in its specific sensitivity to cardiac glycosides. However, other evidence suggests that modifications in the Ca²⁺ release process influence intramembrane charge movement whether through direct feedback effects or alterations in cytosolic [Ca²⁺]. These led to suggestions that the q_γ charge is itself a consequence of Ca²⁺ release (Csernoch et al. 1991). Alternatively, changes in cytosolic [Ca²⁺] might profoundly influence the kinetics of an intramembrane q_γ charge transfer, even if it is primarily driven by the tubular potential (Huang, 1994a,b; Jong et al. 1995; Pape et al. 1996). Finally, pharmacological RyR modification exerts reciprocal effects upon intramembrane charge in a manner compatible with reciprocal allosteric or mechanical couplings between the RyR and the DHPR q_γ charge accounting for the intricacies in q_γ charging kinetics (Huang, 1996, 1997, 1998a,b).

The studies described here complement the earlier report of Sarkozi et al. (1996) in their detailed examination of intramembrane charge in the presence of the same cardiac glycosides at similar concentrations. They studied intact rather than cut voltage-clamped amphibian muscle fibres in the hypertonic gluconate-containing solutions hitherto reported selectively to emphasise q_γ at the expense of q_β charge. Earlier work had used such conditions to examine the effects of pharmacological RyR modification upon intramembrane charge (Huang, 1996, 1998a,b). The present studies additionally investigated the full depolarising voltage range at closer voltage intervals than used previously (Sarkozi et al. 1996) and so could examine for detailed alterations in both charging kinetics and steady-state charge (see Huang, 1996). Finally, they investigated the possible interactions between the actions of cardiac glycosides and those of the known RyR-specific reagents tetracaine and perchlorate (Huang, 1997, 1998a).

Some of the present findings agree with the results of the earlier explorations. Thus, both digoxin and ouabain generally conserved the steady-state properties of intramembrane charge as defined by the single two-state Boltzmann functions used previously to assess the persistence of the steeply voltage-dependent q_γ system in fibres studied in gluconate-containing solutions (Jong et al. 1995; Huang, 1996). If anything, there was a slightly increased rather than decreased total charge, Q_max, in fibres exposed to cardiac glycosides (27 rather than 20 nC μF⁻¹: Huang, 1994a,b, 1996, 1998a,b). Steepness factors, k, were also conserved. However, their values suggested the stronger voltage sensitivities similarly reported in intact fibres studied in gluconate-containing solutions in the absence of such agents (7-9 mV: Jong et al. 1995; Huang, 1996, 1997, 1998a,b) rather than the steepness factors of 16-17 mV reported by Sarkozi et al. (1996). Similarly, cardiac glycosides produced small positive shifts in transition voltages, V*, in contrast to the negative shifts reported on the earlier occasion (Sarkozi et al. 1996). These quantitative results were corroborated by a two-component Boltzmann analysis which separated steeply and gradually voltage-sensitive components of intramembrane charge consistent with a persistence of the individual q_β and q_γ species demonstrated in earlier analyses (Huang, 1994a,b, 1996, 1998a,b).

The present studies then extended the earlier findings to examine kinetic as opposed to just steady-state effects of digoxin and ouabain. The influences of these agents were compared with those of chlormadinone acetate (CMA). The latter is thought exclusively to affect the Na⁺-K⁺-ATPase, but not the RyR, and not to exert inotropic effects. Experiments using CMA thus offered useful controls that could be compared against the results of treatment with the remaining cardiac glycosides (LaBella et al. 1979). Thus, the steady-state voltage dependence of overall charge was similar in the presence of any of the glycosides CMA, ouabain or digoxin. Furthermore, CMA-treated fibres showed distinct q_β and persistent delayed q_γ contributions. In contrast, effective concentrations of ouabain as used by Sarkozi et al. (1996) resulted in faster q_γ transients that consequently tended to fuse with and become indistinguishable from the q_β contribution. The resulting overall records were simple monotonic decays that closely resembled the charging currents following other pharmacological procedures known to modify the RyR-Ca²⁺ release channel that similarly preserved total charge but altered q_γ kinetics (see below; Huang, 1996, 1997, 1998a,b). Digoxin produced similar effects. Such effects of ouabain persisted even at concentrations of CMA (5 and 10 μm) known to be sufficient to displace it from the Na⁺-K⁺-ATPase.

Further experiments demonstrated that the actions of such cardiac glycosides interacted directly with those of other agents that are known to influence the RyR. They were prompted by two lines of earlier evidence. First, tetracaine blocks sarcoplasmic reticular RyR-Ca²⁺ channels in lipid bilayers, prevents Ca²⁺ release in triad preparations and inhibits contractile activation in skinned fibres when applied at millimolar concentrations (Bull & Marengo, 1993; Xu et al. 1993). It also reduced both the sustained phase of cisternal Ca²⁺ release observed in voltage-clamped fibres (Pizarro et al. 1992) and abolished q_γ charge in a study that attributed these actions to changes in a RyR coupled to the voltage sensor (Huang, 1997). Perchlorate (8.0 mm; Huang, 1998a) and caffeine (0.2 mm; Huang, 1998b) antagonised such actions and restored the delayed q_γ charge transfer. The present experiments demonstrate a similar interaction. Thus, ouabain (250 nm) similarly restored such delayed q_γ currents, the maximum charge, Q_max, and the steepness of the overall charge-voltage relationship following their abolition by tetracaine. Second, perchlorate antagonises the effect of the RyR-specific agents ryanodine, daunorubicin and micromolar tetracaine in their modifications of the kinetics of q_γ charge whilst conserving its pharmacological and steady-state identity (Huang, 1996, 1997, 1998a). In the present experiments, perchlorate similarly restored the ‘hump’ waveforms shown by such q_γ charge in ouabain-treated fibres. All these manoeuvres conserved both the total charge and its voltage sensitivity.

Finally, the effects of ouabain and digoxin upon intramembrane charge reported here also resembled those of caffeine, which similarly converted delayed transfers of q_γ charge into simple exponential decays (Huang, 1998b). Caffeine is thought to act directly upon RyR-Ca²⁺ release channels even in fully polarised fibres when applied at sufficiently high (e.g. 5 mm) concentrations (Luttgau & Oetliker, 1968; Kovacs & Szucs, 1983; Rousseau et al. 1988; Klein et al. 1990; Shirokova & Rios, 1996; Huang, 1998b). Nevertheless these findings are compatible with direct actions upon a RyR that, unlike the voltage sensor itself, falls outside the tubular field. RyR modification then would not be expected directly to alter the total quantity or voltage dependence of the steady-state intramembrane charge. Nevertheless it could well influence its kinetics should reciprocal interactions exist between them (Huang, 1996, 1998b). Such direct interactions are consistent with the close morphological associations of four DHPRs with each RyR-I reported in skeletal muscle triads. Biochemical evidence similarly suggests allosteric links between DHPRs and RyR-I or RyR-α isoforms in mammalian and amphibian skeletal muscle triads, respectively, but not with the RyR-II isoforms in cardiac muscle (for reviews see Olivares et al. 1991; Maruyama & Ogawa, 1992; Meissner, 1994; Anderson & Meissner, 1995).

However, the existing reports summarised above indicate that the influence of the cardiac glycosides studied here upon RyRs do not extend to the RyR-I isoform with which the dihydropyridine receptor voltage sensor may thus be directly coupled (Rardon & Wasserstrom, 1990; McGarry & Williams, 1993). Nevertheless, they still permit effects upon other RyR isoforms that are also present in amphibian skeletal muscle (e.g. Olivares et al. 1991; Maruyama & Ogawa, 1992). Such modifications might influence the kinetics of intramembrane charge indirectly through a modification of their Ca²⁺ release (Jong et al. 1995; Pape et al. 1996). Alternatively, the close anatomical relationships between the various RyRs and the DHPR (for reviews see Meissner, 1994; Franzini-Armstrong & Protasi, 1997) permit a hypothesis in which transitions in such RyR isoforms are directly slaved to those in the RyR-Is during excitation- contraction coupling (e.g. Pape et al. 1995). This scheme would be compatible with earlier suggestions that the similar effects of caffeine on the kinetics of the intramembrane charge reflect a dissociation of the reciprocal allosteric coupling between the RyR-I and the DHPR (Huang, 1998b). Caffeine treatment might thus drive the RyRs into a gating mode that permits Ca²⁺ release from intracellular stores even in fully polarised fibres (Luttgau & Oetliker, 1968; Kovacs & Szucs, 1983; Rousseau et al. 1988; Klein et al. 1990; Shirokova & Rios, 1996; Huang, 1998b). Binding of cardiac glycosides to the remaining RyR isoforms might then promote similar transitions if the two different RyR subtypes present also were reciprocally associated. This would also explain why caffeine and cardiac glycosides exert similar effects upon Ca²⁺ release (Sarkozi et al. 1996) and upon the kinetics of the intramembrane charge, and their interactions with tetracaine action, as described here.

In contrast, the actions of perchlorate would then reflect its promotion of a contrasting gating mode in which there was an enhanced direct coupling between RyR-Is and DHPR voltage sensors that in turn would restore the delayed kinetics in the q_γ charge transfers despite cardiac glycoside treatment. Such prolonged charge transfers have thus been associated only with excitation- contraction coupling in vertebrate skeletal muscle that may involve direct allosteric contacts between voltage sensors and Ca²⁺ release channels (Luttgau et al. 1983; Gomolla et al. 1983; Huang, 1986, 1987, 1998a). They have never been reported in other systems such as cardiac and invertebrate striated muscle, where Ca²⁺ release may be triggered more indirectly (Gilly & Scheuer, 1984; Bean & Rios, 1989; Gyorke & Palade, 1992; Cannell et al. 1995).