HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) All Versions of this Article: 555/1/137    most recent jphysiol.2003.051730v1 Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Squecco, R. Articles by Francini, F. Search for Related Content PubMed PubMed Citation Articles by Squecco, R. Articles by Francini, F.

Department of Physiological Sciences, University of Florence, Viale GB Morgagni 63, I-50134 Firenze, Italy

	Abstract

Top Abstract Introduction Methods Results Discussion References

 
The dihydropyridine receptors (DHPRs)/L-type Ca²⁺ channels of skeletal muscle are coupled with ryanodine receptors/Ca²⁺ release channels (RyRs/CRCs) located in the sarcoplasmic reticulum (SR). The DHPR is the voltage sensor for excitation–contraction (EC) coupling and the charge movement component q has been implicated as the signal linking DHPR voltage sensing to Ca²⁺ release from the coupled RyR. Recently, a new charge component, q_h, has been described and related to L-type Ca²⁺ channel gating. Evidence has also been provided that the coupled RyR/CRC can modulate DHPR functions via a retrograde signal. Our aim was to investigate whether the newly described q_h is also involved in the reciprocal interaction or cross-talk between DHPR/L-type Ca²⁺ channel and RyR/CRC. To this end we interfered with DHPR/L-type Ca²⁺ channel function using nifedipine and 1-alkanols (heptanol and octanol), and with RyR/CRC function using ryanodine and ruthenium red (RR). Intramembrane charge movement (ICM) and L-type Ca²⁺ current (I_Ca) were measured in single cut fibres of the frog using the double-Vaseline-gap technique. Our records showed that nifedipine reduced the amount of q and q_h moved by 90% and 55%, respectively, whereas 1-alkanols completely abolished them. Ryanodine and RR shifted the transition voltages of q and q_h and of the maximal conductance of I_Ca by 4-9 mV towards positive potentials. All these interventions spared q_ß. These results support the hypothesis that only q_; and q_h arise from the movement of charged particles within the DHPR/L-type Ca²⁺ channel and that these charge components together with I_Ca are affected by a retrograde signal from RyR/CRC.

(Received 16 September 2003; accepted after revision 5 December 2003; first published online 5 December 2003)
Corresponding author F. Francini: Department of Physiological Sciences, University of Florence, Viale G.B. Morgagni, 63, 50134 Florence, Italy.  Email: fabio.francini{at}unifi.it

	Introduction

Top Abstract Introduction Methods Results Discussion References

 
Excitation–contraction (EC) coupling in skeletal muscle is due to the direct interaction between the dihydropyridine receptor (DHPR)/L-type Ca²⁺ channel and its coupled ryanodine receptor/Ca²⁺ release channel (RyR/CRC). There is a reciprocal interaction, or cross-talk, between these two proteins that involves specific regions of the coupled RyR. In addition to ‘receiving’ the orthograde EC coupling signal from the DHPR/L-type Ca²⁺ channel, RyR/CRC seems to ‘answer’ with a retrograde regulation that enhances the Ca²⁺ channel activity of the DHPR (Nakai et al. 1998). The transmission of the retrograde signal from RyR/CRC to the DHPR/L-type Ca²⁺ channel occurs via the II–III loop of the DHPR (Grabner et al. 1999) or an intermediary protein (Lamb & Stephenson, 1996).

Voltage steps applied to skeletal muscle fibres evoke intramembrane charge movement (ICM) and the inward L-type Ca²⁺ current (I_Ca) through DHPR/L-type Ca²⁺ channels of the T-tubular membrane. In normally polarized skeletal muscle fibres of the frog, ICM has been resolved into two components, an early q_ß, and a delayed q (Adrian & Peres, 1979; reviewed by Ríos & Pizarro, 1991; Schneider, 1994). Based on the close correspondence of q with Ca²⁺ release, several authors proposed that this charge component could directly gate the sarcoplasmic reticulum (SR) Ca²⁺ release channel (Huang, 1982, 1990, 1998a; Hui, 1983; Chen & Hui, 1991; Hui & Chen, 1995; Jong et al. 1995), whereas q_ß could be the result of a separate molecule that may not be required for the EC coupling function (Huang & Peachey, 1989; Huang, 1990; Chen & Hui, 1991). Jong et al. (1995) considered that q_ß and q can represent either two distinct species of charge or two transitions with different properties of a single species of charge and that SR Ca²⁺ content or release alters the kinetics but not the amount of q. This assumption is incompatible with previous findings indicating that there is one type of charge, q_ß, and that q is a movement of q_ß caused by Ca²⁺ released by SR (Csernoch et al. 1991; Pizarro et al. 1991).

Subsequently, a charge species with a significantly more positive threshold, q_h, has been described (Shirokova et al. 1995) and, more recently, the presence of all three charge components has also been reported in rat skeletal muscle (Francini et al. 2001). Based on the close parallel of q_h voltage dependence and that of I_Ca activation (Shirokova et al. 1995; Francini et al. 2001) it has been proposed that this high threshold charge may be related to the T-tubular L-type Ca²⁺ channel gating.

In the present study on single muscle fibres the investigation of ICM and I_Ca was extended to positive potentials in order to examine all three kinds of charges, q_ß,q and q_h, and a number of pharmacological interventions were carried out, mainly aimed at interfering with (a) the DHPR/L-type Ca²⁺ channel, by using channel blockers such as Cd²⁺ (Hui, 1991; Francini et al. 1996), nifedipine (Huang, 1990; Chen & Hui, 1991) and 1-alkanols (Oz et al. 2001), and (b) RyR/CRC, by using specific blockers such as ryanodine (García et al. 1991; Huang, 1996) and ruthenium red (RR) (Csernoch et al. 1991), known to suppress Ca²⁺ release from RyR/CRC.

The first idea considered in this work is that manipulations of the RyR should reciprocally influence the behaviour of any DHPR with which it makes allosteric contact (García et al. 1991; Csernoch et al. 1991; Huang, 1996). The second idea is that RyR modifications do influence the kinetics of the q intramembrane charge by converting its complex kinetics into a simple exponential decay (Huang, 1996). The third point that this paper attempts to resolve is a discrepancy in the literature in that Csernoch et al. (1991) and García et al. (1991) report that similar manipulations to those used in this study do not influence the Ca²⁺ current, whereas Nakai et al. (1996, 1998) report that they do observe a retrograde regulation of the Ca²⁺ current by the RyR.

Four major conclusions emerge from this work. (1) We have excluded the possibility that q_ß participates in DHPR/L-type Ca²⁺ channel activity, whether in EC coupling or in Ca²⁺ currents. In this respect we support the suggestions made by Huang & Peachey (1989) as opposed to those of Melzer et al. (1986), which implicate the entire charge in excitation–contraction coupling. (2) We have demonstrated that interventions directed at DHPR/L-type Ca²⁺ channels act on both q and q_h, thereby implicating both of these charges in the channel activity. (3) We have also demonstrated that interventions directed at the RyR influence both q and q_h kinetics. We thus confirm earlier reports (Huang, 1996, 1998a) that manipulation of the RyR reciprocally influences the q system. Finally, (4) we also report the novel finding, that q_h and the Ca²⁺ current are also influenced by the RyR, supporting the findings of Nakai et al. (1998) as opposed to reports by García et al. (1991) and Csernoch et al. (1991) (see above).

	Methods

Top Abstract Introduction Methods Results Discussion References

 
Dissection and voltage-clamp recording

Frogs were cold adapted in a refrigerator at about 4°C and killed by decapitation and pithing the spinal cord. All experiments followed the official guidelines laid down by the European Community Council (directive 86/609/EEC) incorporated into Italian Government Legislation and the study was approved by the Ethical Committee for Animal Experiments of the University of Florence. Segments of single muscle fibres were dissected from the semitendinosus muscle of Rana esculenta. Those with a diameter between 50 and 70 µm were selected for use. The fibre preparation procedure, its mounting in the double-Vaseline-gap chamber and the voltage-clamp system were as described in detail by Francini & Stefani (1989) and Francini et al. (1996).

To reduce K⁺ current contamination, the four agar bridges providing electrical connections with the electronic apparatus were made with the same solution of the corresponding compartment of the chamber. All experiments were performed at 16°C. Temperature was controlled by a Peltier system and a thermistor.

Solutions

Solutions were the same as those used by Francini et al. (1996). The composition of the external solution for ICM and I_Ca recording (control external solution) was (mM): free Ca²⁺, 2; Ca²⁺, 13.5; L-malate^2-, 74.5; Mops^-, 18; TEA⁺, 131.4; TTX, 0.001; 3,4-DAP, 1; Rb⁺, 2.5; 9-anthracene carboxylic acid, 1; SO₄^2-, 1.25. CdCl₂ (1 mM) was added to the control external solution for ICM recording (Cd²⁺-containing external solution). The internal solution contained (mM): EGTA^2-, 60; ATP^2-, 2.94; glutamate^-, 13; TEA⁺, 139; Mg²⁺, 5; glucose, 5; Mops, 20.

The composition of the internal solution was calculated by a computer program developed by Fabiato (1988) using published stability constants (Martell & Smith, 1977). The pCa value of the internal solution was calculated using the manufacture's assay of Ca²⁺ contamination, about 50 x 10^-6M as total Ca²⁺ (Francini et al. 1992). The calculated concentrations were: free Ca²⁺ 10^-10M (pCa 10), free Mg²⁺ 1 mM, MgATP complex 2 mM and free ATP^2- 0.29 mM. All drugs were obtained from Sigma, except for TEA-OH and 9-anthracene carboxylic acid (Aldrich-Chemie, Steinheim, Germany).

Solutions were made in order to minimize all ionic currents. They were Na⁺, K⁺ and Cl^- free and blockers of Na⁺, K⁺ and Cl^- channels were used: TTX (1 µM) in the external solution for Na⁺ channel block; Cd²⁺ for Ca²⁺ channel block; 3,4-diaminopyridine (3,4-DAP) and Rb⁺ in the external solution and tetraethylammonium hydroxide (TEA-OH) in both internal and external solutions for K⁺ channel block; 9-anthracenecarboxylic acid (9-AC) and SO₄^2- for Cl^- channel block. Both external and internal solutions were Ca²⁺ buffered: the former with malate to prevent tubular Ca²⁺ depletion (Francini & Stefani, 1989), the latter with EGTA, also useful for blocking any calcium-dependent channels. The high concentration of EGTA (60 mM) in the internal solution was previously adopted by Francini et al. (1996) to avoid muscle contraction with long pulses (5 s), even at high depolarizing potentials such as 30 mV. All the solutions were designed to have approximately the same osmolarity (250 mosmol l^-1). In all solutions 3-(N-morpholino)propanesulphonic acid (Mops) was neutralized to pH 7.15 ± 0.05 with TEA-OH.

The following chemicals were tested: nifedipine (1.0–30 µM), heptanol (1–3 mM), octanol (1–3 mM), ryanodine (50–100 µM) and ruthenium red (RR, 0.5–3 mM). In all experiments the stimulation protocols were applied first to fibres bathed in Cd²⁺-containing external solution. After that, the fibres were washed extensively with the control external solution. Then, after 10 min, nifedipine, 1-alkanols (heptanol or octanol), or ryanodine were added to this solution whereas RR was added to the internal solution. However, since ryanodine and RR did not block I_Ca, experiments designed to evaluate the effects of these blocking molecules only on ICM were performed in Cd²⁺-containing external solution. The same stimulation protocol was applied 10 min after the addition of the drugs and was repeated twice every 30 min. Each stimulation protocol lasted about 20 min.

Stimulation and recording

An IBM compatible personal computer was used. Digital-to-analog and analog-to-digital conversions were done by a Digidata 1200 computer interface (Axon Instruments, Inc., Burlingame, CA, USA). Stimulation protocols, acquisitions and recordings were made using pCLAMP, version 6.02 (Axon Instruments). Current records were filtered by an eight pole Bessel filter with a cut-off frequency of 10 kHz. Both the currents elicited during (ON) and after (OFF) the voltage pulses were sampled. Owing to the fast electronics of our device the overall settling time of step potentials was about 150 µs (Francini et al. 1996).

Two different stimulation protocols were used. Short voltage steps, 150 ms long, were applied from a holding potential of -100 mV to the desired potential in increments of 10 mV from -120 to -70 mV and in increments of 5 mV for further depolarizations up to 40 mV. This protocol allowed us to explore the voltages at which the different I_ICM components arose and the pulse duration was chosen to include the full relaxation of the I_ICM and to determine the time of onset of I_Ca in the control external solution experiments. The OFF currents were recorded for 160 ms. Test voltage pulses 5 s long were applied from a holding potential of -100 mV to the desired potential in increments of 5 mV from -70 to 40 mV. This protocol allowed us to explore I_Ca voltage dependence and to include its full inactivation. The sampling time for the ON and OFF current was 50 µs. I_ICM and I_Ca were evaluated after subtraction of linear capacitive and leak currents using properly scaled currents obtained in response to hyperpolarizing control pulses. Control voltage pulses, of a duration and sampling time equal to those of the test pulse investigated, were hyperpolarizing steps of 20 mV applied from a holding potential of -130 mV. Five control pulses were applied and the corresponding control current traces were averaged. Since scaling of control current made the test-minus-control traces noisier a synthesized version of control current was used for large voltage test pulses. It was constructed by fitting a multiexponential function plus a constant (accounting for the leak current) to the ON current and the same function but without the constant to the OFF current. Each test protocol was followed by a control protocol. Successive voltage test and control steps were separated by 40 s intervals.

Data analysis

Analysis of the steady-state charge versus voltage data obtained by integrating the ON and OFF ICM current, Q_ICM,ON(V) and Q_ICM,OFF(V), was the same as that previously performed in amphibian and mammalian muscle (Melzer et al. 1986; Hui & Chandler, 1990, 1991; Shirokova et al. 1995; Francini et al. 2001).

The equation used was the sum of three Boltzmann terms that, according to Shirokova et al. (1995) and Francini et al. (2001), represent the amount of q_ß, q and q_h charges moved, Q_ß, Q and Q_h, respectively:

(1)

where Q_ß,max,Q_,max and Q_h,max represent the maximal moveable charge for q_ß,q and q_h; V_ß,V and V_h are the corresponding transition voltages with steepness factors k_ß,k and k_h; c is a constant. Equation (1) was constrained to be zero at the holding potential (-100 mV).

The time course of the macroscopic I_Ca was fitted by the sum of two exponential functions:

(2)

where I_Ca(t) is the current density at time t after the depolarization; I_Ca,a and I_Ca,i are the amplitudes of each component representing the activation and inactivation time course, respectively; C is the steady-state current; _a and _i are the time constants for the two components of the current time course; t_o is the tubular delay that, according to Francini et al. (1996) and Francini et al. (2001), was constrained to be 2 ms.

The following eqn (3) was used to determine the voltage dependence of I_Ca,a. Evaluation of I_Ca,a(V) instead of the peak current I_Ca(V) was preferred because, as we will show below, significant inactivation can overlap the activation phase with a consequent alteration of the apparent activation curve.

(3)

where G_max is the maximal conductance for the I_Ca,a, V_rev, is the apparent reversal potential, V_G is the potential that elicits the half-maximal increase in conductance, and k_G is the steepness factor. In control external solution the inward I_Ca followed and partly overlapped with ICM current when the test voltage pulses applied were suprathreshold for I_Ca activation. Thus, in the ON configuration the time course of the ICM current was extracted after having removed the time course of I_Ca, evaluated using eqn (3), from the original current traces.

Currents and their time integrals were normalized to their linear capacitance measured by control pulses. Data fitting was done using non-linear curve fitting based on the Marquardt-Levenberg algorithm (Sigmaplot 4 and Table Curve 3.10 by Jandel Scientific, CA, USA, and Clampfit 6.02 by Axon Instruments). The goodness of fit was evaluated using the statistical parameters of these programs. Since the sum of a different number of Boltzmann functions was fitted to the data, the best fit was chosen using a test based on the value of the likelihood ratio statistic, LRS, with the same formalism as reported in Hui & Chandler (1990). The improvement of the fit was evaluated by ² statistics. A difference showing P < 0.05 was considered statistically significant.

All data were expressed as mean ±S.E.M.; n is the number of fibres. Data were analysed using one-way ANOVA with the Bonferroni' s correction for multiple comparisons. The alpha values were at P < 0.05 for all tests.

Subsequent data analysis included linear cable analysis of the control records, which yielded information about myoplasmic resistance (r_i), membrane resistance (r_m), membrane capacitance (c_m) and the gap factor of the Vaseline seal defined by r_e/(r_e+r_i), where r_e is the external resistance underneath the Vaseline seal (Irving et al. 1987; Hui & Chandler, 1990; Delbono et al. 1991; Francini et al. 2001).The condition of the fibre was tracked by monitoring the holding current, calculating the cable constants' parameters and by applying, every 20–40 min, a long (1700 ms) depolarizing pulse to 10 mV from a holding potential of -100 mV, to test rundown of I_Ca. The fibre was rejected if one of these parameters exceeded 1% its initial value.

	Results

Top Abstract Introduction Methods Results Discussion References

 
Nifedipine blocks I_Ca and reduces the amount of q and q_h charge moved

ICM currents were first recorded in Cd²⁺-containing external solution, where I_Ca was effectively blocked. Figure 1 displays typical ICM records obtained from the same fibre both in Cd²⁺-containing external solution (A) and in the presence of 15 µM nifedipine (B). Figure 1A shows that all the ON current traces regained the baseline, demonstrating the absence of any significant ionic current. Charge conservation was confirmed by the good equality of the time integral of ICM moved by the ON and OFF pulses, Q_ICM,ON and Q_ICM,OFF (Fig. 1D), even at strong depolarizations. The ON and OFF ICM current time course closely parallels that previously described for frog (Huang, 1990; Hui, 1991) and rat skeletal muscle (Francini et al. 2001). ICM records in Fig. 1A show a very early charge, rapidly peaking at 0.1-0.2 ms. This has not been further considered in this study since it may represent either the movement of charges outside the tubular membrane or gating current of Na⁺ or K⁺ channels (Francini et al. 2001). This charge was followed by a slower decaying component representing q_ß. Another charge is superimposed on q_ß above a voltage threshold of about -40 mV and is represented by a prolongation of ICM current decay (indicated by *). In fact, for voltage pulses up to -50 mV ICM regained the baseline in 30 ms, whereas at -40 mV this time increased sharply to 60-90 ms. The occurrence of the ON current prolongation was accompanied by the prolongation of the corresponding OFF current duration from 10-20 to 30-40 ms. The time course of this ON intramembrane charge component changed for further depolarizations: at -30 mV it appeared as a delayed ‘hump-like’ component (indicated by **) and from -20 mV as a true ‘hump’.

View larger version (33K): [in this window] [in a new window]   Figure 1.  Nifedipine modifies the kinetics of the ICM Representative ICM current traces from a fibre in Cd2+-containing external solution (A) and in the presence of nifedipine added to the control external solution (B). Voltage pulses, 150 ms long, were applied from a holding potential of -100 mV and were increased in 10 mV steps from -120 to -70 mV and in 5 mV steps from -65 up to 40 mV. For clarity, only traces from -70 to 30 mV in increments of 10 mV are depicted in this panel. Pulse potentials are indicated in mV next to the traces. Time zero is the ON onset. The ON current traces are blanked between 90 and 140 ms. The horizontal lines denote zero current for each trace. * and ** on the ON current traces at -40 and -30 mV indicate the appearance of a prolongation of the ICM time course and of a delayed ‘hump-like’ charge component, respectively. This latter appears as a true ‘hump’ for further depolarizations in A, whereas a true delayed ‘hump’ cannot be observed at any potential in B. Moreover, the prolongation of the ICM time course is reduced in B compared to A. C, difference current traces calculated by subtracting ICM current traces measured in the presence of nifedipine (B) from those measured in the original Cd2+-containing external solution (A). D and E, QICM,ON/QICM,OFF plots with superimposed linear regression lines showing the charge conservation; data from A (D) and from B (E). Slope = 1.00094 ± 0.0135 and 1.00084 ± 0.0145 (n= 8) for D and E, respectively. During the experiment the holding current changed from -28 to -30 nA and the re/(re+ ri) value from 0.986 to 0.985, whereas the linear capacitance did not change. Fibre diameter 60 µm, linear capacitance 6.3 nF.

To better emphasize the ICM component reduced by nifedipine we show in Fig. 1C the difference current traces calculated by subtracting ICM current traces measured in the presence of nifedipine (Fig. 1B) from the corresponding traces measured in the original Cd²⁺-containing external solution (Fig. 1A). At potentials up to -40 mV, that is in the voltage range where only q_ß is elicited, there is not any difference between the two sets of traces. In contrast, starting from -40 mV we can observe a delayed ‘hump’ component that represents the amount of charge reduced by nifedipine. The time course of the difference current traces of the suprathreshold ICM components shows that the depressed ICM components have a clear delayed ‘hump’ form. Being the q and q_h charges not completely blocked, the difference current traces roughly show the time course of a fraction of these charges. The real time course of q and q_h charges is observable in the difference current traces performed in the presence of heptanol since it completely block these charges (see below).

The Q_ICM(V) data of the eight fibres investigated were obtained by averaging the time integrals of ON and OFF current traces and are shown in Fig. 2A as means ±S.E.M. (•). Notably, the plot showed three clear changes in steepness. This set of data was best fitted by eqn (1) as the sum of three Boltzmann functions, representing Q_ß, Q and Q_h(superimposed continuous line). The best-fit parameters are listed in Table 1, column 2. In summary, q_ß appeared as a weakly voltage-dependent charge component with a transition voltage occurring at low depolarizations, whereas q and q_h charge components were the charge moved above the voltage threshold. These were steeply voltage dependent and their transition voltages occurred at larger depolarizations. In particular, the bulk of q moved in the voltage range between -40 and -20 mV whereas q_h moved mostly at voltages positive to 0 mV.

View larger version (21K): [in this window] [in a new window]   Figure 2.  Nifedipine reduces the amount of q and qh moved A, QICM(V) plots obtained from fibres bathed in Cd2+-containing external solution, Cd2+ (•) and 10 min after the addition in control external solution of nifedipine, (Nif, ) (n= 8) from experiments as in Figure 1. Continuous lines through the QICM(V) data represent the best fit as the sum of three Boltzmann functions (eqn (1)) for Cd2+ and Nif data. Dashed lines represent the voltage dependence of Qß, Q and Qh, calculated by fitting the sum of three Boltzmann functions to the Cd2+ data. B, amount of charge depressed by nifedipine obtained by subtracting Nif from Cd2+ data (). Continuous line through the data represents the best fit as the sum of two Boltzmann functions (eqn (1)) to QICM(V) plots representing Q and Qh. The parameters of the best fits shown in A and B are listed in Table 1. In both panels, horizontal lines indicate the zero level. Data are mean values ±S.E.M. Error bars are shown when the S.E.M. exceeded the symbol size.

In the presence of nifedipine the Q_ICM(V) plots (Fig. 2A) showed different slopes as observed in Cd²⁺-containing external solution, and once again, the best fit was achieved by the sum of three Boltzmann functions (eqn (1)), reported as a superimposed continuous line to the Q_ICM data. Data obtained in the presence of nifedipine perfectly superimposed those obtained in Cd²⁺-containing external solution up to -50 mV, but beyond this voltage value they were strongly reduced in size. This denoted that the amount of q and/or q_h charge moved was reduced in the presence of nifedipine, further confirming what had already been observed with the difference traces of Fig. 1C.

The best-fit parameters obtained in the presence of nifedipine are listed in Table 1, column 3. In comparison with the results obtained in Cd²⁺-containing external solution the addition of nifedipine strongly reduced the total maximal moveable q and q_h charge by 90 and 54%, respectively, whereas neither the transition voltages nor the steepness factors were affected. The calculated difference between the data obtained in Cd²⁺-containing external solution and those obtained with nifedipine (Fig. 2B, ) enhances this view: the curve is flat up to -50 mV, indicating that nifedipine did not affect q_ß, whereas at more positive potentials q and q_h appear reduced in size. Accordingly, the best fit was obtained by the sum of two Boltzmann functions (eqn (1)). The parameter values listed in column 4 of Table 1 confirmed that most of Q was blocked and that Q_h was strongly depressed. According to the voltage dependence presented in Fig. 2 and Table 1, the delayed ‘hump’ charges observed in the difference records showed in Fig. 1C represent the time course of the reduced q and q_h.

No further significant changes were observed in records obtained after 40 or 70 min of exposure to nifedipine (n= 8).

Heptanol and octanol block I_Ca and the mobilization of q and q_h charges

We then examined the effects of adding 1-alkanols (heptanol and octanol) on ICM. Accordingly, after having recorded ICM in Cd²⁺-containing external solution, the external solution was replaced by the control external solution, and 10 min later heptanol (n= 7) or octanol (n= 6) were added. The subsequent electrophysiological study started after 10 min exposure to 1-alkanols. Heptanol or octanol (1.5–2 mM) successfully blocked I_Ca. All the resulting ON currents decayed to the baseline as in Cd²⁺-containing external solution, without requiring any baseline current correction. Figure 3 shows typical records obtained from the same fibre in Cd²⁺-containing external solution (Fig. 3A) and after the addition of 2 mM heptanol (Fig. 3B). Notably, in the presence of heptanol, modifications in the ICM time course, such as the appearance of a prolonged, delayed ‘hump-like’ or true ‘hump’ charge component, were not detectable. To better highlight the ICM component blocked by heptanol we show in panel C the difference current traces calculated by subtracting ICM current traces measured in the presence of heptanol (Fig. 3B) from those measured in the original Cd²⁺-containing external solution (Fig. 3A). The difference records show that heptanol determines a greater reduction in the amount of ICM than nifedipine. Again, the charge conservation was confirmed by the good equality of the time integral of ICM induced by the ON and OFF pulses both in the presence of Cd²⁺ (Fig. 3D, •) and in the presence of heptanol (Fig. 3E, ).

View larger version (33K): [in this window] [in a new window]   Figure 3.  Neither charge prolongation nor delayed ‘hump-like’ or ‘hump’ charge components can be observed in the presence of heptanol Representative ICM current traces from a fibre in Cd2+-containing external solution (A) and in the presence of heptanol added to the control external solution (B). Same stimulating protocol, set of traces and symbols as in Figure 1. B, charge prolongation, delayed ‘hump-like’ or ‘hump’ charge components cannot be observed at any potential. C, difference current traces calculated by subtracting ICM current traces measured in the presence of heptanol (B) from those measured in the original Cd2+-containing external solution (A). D and E, QICM,ON/QICM,OFF plots with superimposed a linear regression line showing the charge conservation: slope 1.00114 ± 0.0155 (n= 7). Fibre diameter 65 µm, linear capacitance 6.4 nF. During the experiment the holding current changed from -29 to -32 nA and the re/(re+ ri) value from 0.988 to 0.987, whereas the linear capacitance did not change.

View larger version (20K): [in this window] [in a new window]   Figure 4.  Heptanol blocks the amount of q and qh moved A, QICM(V) plots obtained from fibres bathed in Cd2+-containing external solution, Cd2+ (•) and 10 min after the addition in control external solution of heptanol, Hept () (n= 7) from experiments as in Figure 3. Continuous lines through the QICM(V) data represent the best fit as the sum of three Boltzmann functions (eqn (1)) for Cd2+ data or a single Boltzmann function, representing Qß, for Hept data. Dashed lines represent the voltage dependence of Q and Qh calculated by fitting the sum of three Boltzmann functions to Cd2+ data. Dashed line related to Qß perfectly superimposes the Hept data. B, amount of charges depressed by heptanol obtained by subtracting Hept from Cd2+ data (). Continuous line through the data represents the best fit as the sum of two Boltzmann functions (eqn (1)) to QICM(V) plots representing Q and Qh. The parameters of the best fits shown in A and B are listed in Table 2. In both panels, horizontal lines indicate the zero level. Data are mean values ±S.E.M. Error bars are shown when the S.E.M. exceeded the symbol size.

Next, we focused our attention on the effects of ryanodine on I_Ca. To this purpose, we performed experiments on fibres bathed in control external solution that allowed I_Ca to be recorded by using long voltage pulses lasting 5 s. The effects of ryanodine were more evident after 40 min than after 10 min exposure, thus they showed a certain time dependence; however, no further significant changes were observed after 70 min. Figure 5 shows typical I_Ca records performed in control external solution prior to the addition of (A) and after 40 min exposure to ryanodine (B) in the same fibre. These records clearly demonstrate the difference in current density and voltage threshold of I_Ca. The voltage threshold for I_Ca activation was assumed to be the voltage value at which the downward slope of the current traces in the last 500 ms of the voltage steps was observed. In control conditions, I_Ca activates at -50 mV and shows maximal peak size at 0 mV; in contrast, in the presence of ryanodine the corresponding values are -40 and 10 mV. The voltage threshold for I_Ca activation, was -56.2 ± 2.5 mV in control prior to the addition of ryanodine, -49.8 ± 2.5 mV after 10 min exposure and -48.4 ± 2.6 mV after 40 min exposure (n= 6). Thus, ryanodine caused a shift of 8 mV towards more positive potentials in both I_Ca threshold and of I_Ca maximal peak size.

View larger version (25K): [in this window] [in a new window]   Figure 5.  Effects of ryanodine on ICa Current traces obtained through the imposition of long depolarizing test pulses (5 s) from a holding potential of -100 mV. Test voltage pulses ranged from -70 to 30 mV in 5 mV increments. For clarity only the traces from -70 to 30 mV in steps of 10 mV are shown. Some potentials are indicated in mV next to the traces. A, typical experiment performed in a fibre bathed in control external solution. B, the experiment was performed in the same fibre 40 min after the addition of 100 µM ryanodine to this solution. ICa appearance is observed at -50 mV in A and -40 mV in B. Fibre diameter 65 µm, linear capacitance 6.9 nF. From the beginning to the end of the experiment the holding current changed from -27 to -30 nA, re/(re+ri) changed from 0.988 to 0.985, and linear capacitance from 6.1 to 6.3 nF. C, I-V plots for ICa,a evaluated by eqn (2)) before (Control) and 10 (Ry (10 min)) or 40 min (Ry (40 min)) after the addition of ryanodine (n= 6).

Effects of ryanodine on ICM components recorded in Cd²⁺-containing external solution

To better evaluate the action of ryanodine on ICM, we performed another set of experiments where ryanodine was added to the Cd²⁺-containing external solution. Again, the effects of ryanodine were more evident after 40 min with respect to 10 min exposure, thus showing a certain time dependence; however, no further significant changes were observed after 70 min. Figure 6 shows typical current traces recorded before (A) and after 40 min exposure to 100 µM of ryanodine (B). Notably, in the presence of ryanodine the delayed ‘hump-like’ or true ‘hump’ charge components were not clearly discernible. Moreover, the charge prolongation and the appearance of the delayed ‘hump-like’ charge, indicated by * and **, respectively, were shifted by 10 mV towards more positive potentials. A comparable shift of the Q transition voltage was reported by Huang (1996) where ryanodine and daunorubicin transformed charge movements that included delayed q‘hump’ components into simpler decays. The related Q(V) plots are shown in Fig. 7 and the corresponding parameters from eqn (1) are listed in Table 4. It should be noted that ryanodine did not affect any parameter of Q_ß, but caused a positive shift in the transition voltages of the suprathreshold charge components, leaving the maximal moveable charge and the steepness factors almost unchanged. After 10 min exposure to ryanodine the amount of the shift was 4.9 and 3.9 mV for Q and Q_h, respectively, whereas after 40 min the respective values were 8.8 and 7.0 mV.

View larger version (24K): [in this window] [in a new window]   Figure 6.  Ryanodine modifies the kinetics of the ICM ICM time course recorded in Cd2+-containing external solution (A) and 40 min after the addition of ryanodine to this solution (B). Stimulating protocol, set of traces and symbols as in Figure 1. At the end of the experiment in A,100 µM ryanodine was added to the Cd2+-containing external solution. During the experiment the holding current changed from -27 to -30 nA and the re/(re+ ri) value from 0.989 to 0.987. Fibre diameter 70 µm, linear capacitance 7.3 nF. C and D, QICM,ON/QICM,OFF plots with superimposed linear regression lines showing the charge conservation; data from panel A (C) and from panel B (D). Slope = 1.00093 ± 0.0134 and 1.00094 ± 0.0125 (n= 9) for C and D, respectively.

View larger version (21K): [in this window] [in a new window]   Figure 7.  Ryanodine shifts towards positive potentials the transition voltages of Q and Qh QICM(V) plots evaluated in fibres bathed in Cd2+-containing external solution before (•) and 40 min after the addition of ryanodine to this solution (), from experiments as in Figure 6 (n= 9). Continuous lines through the data represent the best fit as the sum of three Boltzmann functions (eqn (1)) to QICM(V) plots. For comparison, the Qß(V) (continuous line) and Q(V) and Qh(V) calculated before (short dashed line) and 40 min after the addition of ryanodine (long dashed line) are reported. The related best-fit parameters are listed in Table 4. Data are the mean values ±S.E.M. Error bars are shown when the S.E.M. exceeded the symbol size.

No further changes were observed in records made after 70 and 100 min exposure to ryanodine.

Furthermore, experiments performed with RR revealed no significant differences with respect to ryanodine either in the ICM time course or in the Q_ICM(V) plot. Once more, the only difference was that the effects observed after 40 and 70 min of exposure to RR corresponded to those observed after 10 and 40 min of ryanodine exposure.

	Discussion

Top Abstract Introduction Methods Results Discussion References

 
The goal of the present study was to clarify the cross-talk between the DHPR/L-type Ca²⁺ channel and RyR/CRC by examining not only the classical q_ß and q charge components of ICM but also the recently described high threshold charge component q_h, as well as I_Ca. To this end we extended our analysis to positive potentials and we carried out a number of pharmacological interventions that sought to interfere with DHPR/L-type Ca²⁺ channel and RyR/CRC. We observed the following differential effects: (a) the nifedipine and 1-alkanols (heptanol and octanol), besides blocking I_Ca, depressed the amount of q and q_h charge moved, without affecting their transition voltages and steepness factors; (b) in contrast, interventions with ryanodine and RR directed at the RyR/CRC shifted the transition voltages of both q and q_h charge movement, as well as I_Ca activation, towards positive potentials, leaving the maximal moveable charge and the steepness factors almost unchanged. Notably, previous work has suggested that pharmacological modifications of the RyR/CRC would be likely to exert reciprocal effects upon the DHPR/L-type Ca²⁺ channel only in the event of a direct coupling between these two molecules. Since this was exactly what we observed in our experiments, the specific contribution of the present work is the demonstration that the q_h charge movement is likely to be involved in such coupling, adding to suggestions that have already implicated the q charge (Huang, 1996, 1998a,b).

Both q and q_h reside in the DHPR/L-type Ca²⁺ channel and represent separate independent processes from q_ß

It is remarkable that our results show that not only q and I_Ca, but also q_h reflect phenomena that can be attributed to conformational changes in the DHPRs/L-type Ca²⁺ channels of the T-tubule system. Conversely, q_ß might be an independent event. This assertion related to q and I_Ca is strongly consistent with a number of previously published results: (i) q was exclusively distributed in the T-tubules, in contrast to q_ß, which could be present in both the surface membrane and the tubules (Huang & Peachey, 1989); moreover, the different degrees of voltage dependence of the blockage of q_ß and q by nifedipine argued against a sequential model where the two charges are tightly coupled (Melzer et al. 1986; Csernoch et al. 1991), rather suggesting that q_ß and q are two distinct species of charge moving independently (Huang, 1990; Chen & Hui, 1991); (ii) I_Ca is completely blocked by the specific DHPR antagonist nifedipine (Francini et al. 1996), supporting the suggestions that it originates entirely from the DHPR/L-type Ca²⁺ channel. There are previous reports that suggest that q_h is also a charge component unrelated to q_ß. The bioactive sphingomyelin derivative sphingosine 1-phosphate produced parallel changes on q, q_h and I_Ca, leaving q_ß almost unchanged. In fact, the q and q_h charge components and I_Ca were shifted towards negative potentials by 10 mV by the addition of sphingosine 1-phosphate to the external solution and this effect was likely to occur via G-proteins coupled with specific receptors of the endothelial differentiation gene (Edg) family of the plasma membrane (Bencini et al. 2003). The results reported in the present work argue in favour of this hypothesis, leading to the conclusion that not only q(but also q_h may represent a separate independent process from q_ß. In fact (a) any intervention directed at the DHPR/L-type Ca²⁺ channel or RyR/CRC completely spared q_ß while having significant effects upon both q and q_h; (b) in control external solution the effects of ryanodine and RR on q and q_h, and on I_Ca, were closely related. In fact, ryanodine and RR produced a similar positive shift in the transition voltages of Q and Q_h (8.8 and 7.0 mV, respectively) and I_Ca (6.0 mV).

Many previous studies report that interventions inhibiting I_Ca also affect ICM. The addition of D600 and nifedipine blocked q but left q_ß almost unchanged (Caputo & Bolanos, 1989; Huang, 1990). In addition, many authors have found that nifedipine decreases charge movement (Lamb, 1986; Rios & Brum, 1987; Lamb & Walsh, 1987) and, according to Huang (1990), 10 µM nifedipine added to intact fully polarized fibres of the frog caused a substantial reduction of q. In the present work we investigated the effect of nifedipine on ICM, also at strong positive potentials, and we found that the differential effect of nifedipine on q_ß and q can be extended to q_h. This latter component, being depressed by 55%, was less sensitive to nifedipine than q, which, in contrast, was substantially blocked. This may indicate that the charged amino acid residues involved in q and q_h mobilization are different, but it also points towards a certain degree of interdependence.

It has been previously reported that local anaesthetics cause differential effects on the ICM components. For instance, tetracaine prevents q mobilization while leaving q_ß unaffected (Huang, 1981, 1982, 1996, 1997; García et al. 1991).

There is evidence that general anaesthetics such as 1-alkanols can affect the transport of ions across the cell membrane, acting on discrete sites within the ion channel protein (Covarrubias et al. 1995). There is also evidence that 1-alkanols may affect both DHPR and RyR Ca²⁺ channels. (a) It is reported that ethanol and higher alcohols (butanol, hexanol, octanol and decanol) inhibit the function of the DHPR/L-type Ca²⁺ channel in purified T-tubule membrane vesicles of the rabbit. It appears that the inhibition of channel function is not related to the DHP binding site on the L-type Ca²⁺ channel and that this inhibition is independent of intracellular Ca²⁺ levels (Oz et al. 2001). (b) In contrast, heptanol and octanol, but not other alcohols inhibit gap-junction channels (Ma et al. 1988) as well as RyR/CRCs of skeletal muscle. Our results showed that heptanol and octanol, besides eliminating I_Ca, completely blocked q and q_h. Moreover, we showed that blocking molecules such as ryanodine and RR that specifically bind to the RyR/CRC allowed the movement of q and q_h, causing a shift in their transition voltages towards positive potentials. If alkanols affected mainly RyR/CRC we should have observed a similar effect. Since this was not the case, we can reasonably suggest that the effects of heptanol and octanol observed in the present experiments are mostly due to their action on a site in the L-type Ca²⁺ channel that, according to Oz et al. (2001), is different from the specific binding site for DHPs. Because of the different sensitivities of q and q_h to nifedipine, our results provide further evidence that both q and q_h are related to L-type Ca²⁺ channels, albeit that they may be due to the movement of different charged amino acid residues involved in different functions (see below).

q as the voltage sensor for EC coupling and q_h as the gating current for the opening of L-type Ca²⁺ channels

It is widely reported in the literature that inorganic Ca²⁺ channel blockers used to abolish I_Ca in skeletal muscle fibres cause particular effects on charge movement components. The use of metal ions causes clear effects on charge movement, inducing substantial changes in kinetics and voltage dependence, without affecting the total amount of charge moved. Hui (1991) found that adding 0.5–1 mM Cd²⁺ to the external solution produced a differential shift in the voltage of Q_ß and Q; the former was shifted towards negative potentials and the latter towards positive potentials. Thus, since the free ion concentration was very low, such a shift could not be attributed to a screening of surface charges. These results obtained in frog skeletal muscle agree with those recently observed in rat skeletal muscle fibres (Francini et al. 2001): the presence of external Cd²⁺, besides abolishing I_Ca, shifted not only Q_ß and Q, but also Q_h charges of 8–10 mV towards more positive potentials and influenced the charging kinetics, making the delayed ‘hump’ less evident. Shirokova et al. (1995) showed that the transition voltage V_Ca determined by tail current experiments was slightly shifted towards negative potentials with respect to V_h, determined after the blockage of I_Ca by Cd²⁺ and lanthanum. The hypothesis that q_h is a kind of charge movement related to I_Ca activation predicts that its transition voltage value must be more negative than that of I_Ca. Considering that Cd²⁺ caused a 8–10 mV shift towards more positive potentials in V_h values relative to that found in control external solution (Francini et al. 2001) and that a V_G value of -6.1 mV (Table 3) was found in control external solution, the V_h value of 1.5 mV (Tables 1 and 4) found in the Cd²⁺-containing external solution must be shifted towards more negative potentials to be correctly compared with V_G. Such a consideration positively relates V_h with V_G. Our experimental results showed that the voltage thresholds for the q and q_h movements and that for I_Ca activation are coincident. In fact, the voltage threshold evaluated in the same fibres for the appearance of the charge prolongation due to the suprathreshold charge components in Cd²⁺-containing external solution was -47.4 ± 3.8 mV, and in control external solution with nifedipine added it was not statistically different, being -46.1 ± 4.8 mV (n= 8). In the presence of nifedipine, most of the charge moved was q_h, with being q substantially depressed. However, there are substantial differences in the voltage dependence of q, q_h and I_Ca. In Cd²⁺-containing external solution, most of q moved at -40 mV, whereas at this potential only a small fraction of q_h moved. The reason for this is that q_h has a larger steepness factor (k_h= 8.1 mV versusk= 3.4 mV) and a more positive transition voltage (V_h= 1.4 mVversusV=-40.0 mV) than q. The fact that the k value was smaller than those of k_h and I_Ca (k_G= 6.8 mV) argued against a sequential model where the two charges are tightly coupled. In contrast, the larger value of k_h than k_G agrees with the hypothesis that q_h is related to I_Ca activation (compare column 2 in Tables 1 and 3). This consideration further leads to the conclusion that q may represent a relatively but not absolutely separate independent process from q_h. In particular, we propose that the q and q_h components are related to the movement of the charged particles of the tubular DHPRs: q may be the electrophysiological manifestation of the voltage sensing component of the EC coupling process and q_h may be the manifestation of the gating of L-type Ca²⁺ channels. The difference current traces from experiments performed in heptanol (Fig. 2C), representing the time course of q and q_h, show that both the charge components are delayed and ‘hump’ shaped. Due to the similar time constant values (see Fig. 5 in Francini et al. 2001), these charges are indistinguishable from one another. The occurrence of the two different charge components can be assessed by the analysis of the Q(V) plots. As a result, it may be speculated that the present findings further show that the kinetics not only of q (Huang, 1982; Pizarro et al. 1991; Huang & Peachey, 1992; Ríos et al. 1993; Shirokova et al. 1994; Jong et al. 1995; Francini et al. 2001), but also of q_h (Francini et al. 2001) and the related I_Ca activation (Francini et al. 1996) reflect multistep mechanisms.

DHPR/L-type Ca²⁺ channel and RyR/CRC cross-talk: the retrograde signal from RyR/CRC modulates either q, the voltage-sensing signal for EC coupling, or q_h, the charge associated with the gating of L-type Ca²⁺ channels

Allosteric interactions of DHPR with RyR/CRC may be responsible for the complex delayed kinetics of q (Huang, 1982, 1983, 1996, 1998a,b). Thus, both the steady-state and kinetic properties of the q charge depend steeply and uniquely upon test voltage, independently of charging history (Huang, 1994), and are sensitive to DHPR-specific agents (Huang, 1990; Hui & Chen, 1992). On the other hand, both agonist and antagonist modifications of the RyR in intact, voltage-clamped muscle fibres have been shown to reciprocally but reversibly influence the kinetic but not the steady-state properties of q charge. These findings are compatible with a direct coupling between the DHPR voltage sensor of the T-tubule membrane and the RyR located in the SR membrane (Huang, 1996, 1998a,b, 2001). Such a tight coupling scheme would predict that the relevant charging phenomenon should persist following imposition of voltage changes, even in experimental systems whose intracellular Ca²⁺ stores were sufficiently depleted to compromise their capacity to release Ca²⁺. Thus, Jong et al. (1995) and Pape et al. (1996) reported that intramembrane charge persisted, albeit with modified kinetics, in Vaseline-gap preparations whose SR Ca²⁺ was altered by using (20 mM) EGTA-containing solutions, including either depleted (0.0 mM) or loaded (1.76 mM) Ca²⁺ concentrations. Changes in q have previously been observed in skeletal muscle fibres in which RyR/CRC was affected by different molecules: RyR antagonists such as ryanodine and daunorubicin, RyR agonists such as caffeine (Huang, 1996; Huang, 1998b), or Ca²⁺-ATPase inhibitors (Chawla et al. 2002) specifically affected q kinetics, although the separate identities of the steady-state q_ß and q remained unchanged. However, Csernoch et al. (1991) proposed that the q current was primarily driven by changes in cytosolic [Ca²⁺] rather than tubular voltage on the basis of physical and pharmacological manipulations that appeared to affect both intramembrane charge and Ca²⁺ release. In conclusion, given that we can observe q‘humps’, even in conditions when Ca²⁺ release is compromised, the balance of the evidence is in favour of an allosteric coupling rather than a scheme in which the driving force for q is the release of Ca²⁺ rather than voltage.

It is likely that the cross-talk might reflect some of the characteristics of q and q_h, and their modification by RyR-specific reagents might represent a consequence of Ca²⁺ release rather than of direct contact between the RyR and the DHPR. To this end, it is important to note that Jong et al. (1995) and Pape et al. (1996) observed that the q charge persisted in fibres in which sarcoplasmic reticular Ca²⁺ was depleted, and Chawla et al. (2002) reported that it was possible to demonstrate q currents in both exponential and ‘hump-like’ forms, even after the SR had been depleted of Ca²⁺ by cyclopiazonic acid.

The present experiments tested the particular predictions of such schemes including the nature of the q_h charge component in cut fibre preparations under conditions of extracellular solution, tonicity and temperature that closely paralleled those consistently adopted in previous experiments (Francini et al. 1992, 1996). Two chemically distinct pharmacological agents, ryanodine and RR, were applied at concentrations known to stabilize the closed state of the RyR/CRC. The effects of the RyR antagonists ryanodine and RR on the kinetics and steady-state properties of ICM agree with the hypothesis of cross-talk between the DHPR/L-type Ca²⁺ channel and RyR/CRC that involves not only q, but also q_h. Indeed, our results showed that both the RyR antagonists cause a shift towards positive potentials in the transition voltages of these charge movements, as well as I_Ca. This directly compares to an earlier study by Gonzalez & Rios (1993) where perchlorate both stabilized the open state of RyR and caused a hyperpolarizing shift in charge movement. This finding was considered by Nakai et al. (1996) as further support for the idea of a reciprocal interaction between RyR and DHPR.

In conclusion, the results reported in this paper agree with the hypothesis that q and q_h arise from the movement of charged particles within the DHPR, whereas q_ß appears as a separate and independent species of charge. Taken together, the above observations have deepened our understanding of the cross-talk hypothesis between the DHPR/L-type Ca²⁺ channel and RyR/CRC, clearly suggesting that retrograde regulation by RyR/CRC modulates either q, the charge associated with the voltage sensing component of EC coupling, or q_h, the charge associated with the gating of L-type Ca²⁺ channels and the resulting I_Ca activation.

	References

Top Abstract Introduction Methods Results Discussion References

 
Adrian RH & Peres A (1979). Charge movement and membrane capacity in frog muscle. J Physiol 289, 83–97.[Abstract/Free