In vivo expression of G-protein β₁γ₂ dimer in adult mouse skeletal muscle alters L-type calcium current and excitation–contraction coupling

Abstract

A number of G-protein-coupled receptors are expressed in skeletal muscle but their roles in muscle physiology and downstream effector systems remain poorly investigated. Here we explored the functional importance of the G-protein βγ (Gβγ) signalling pathway on voltage-controlled Ca²⁺ homeostasis in single isolated adult skeletal muscle fibres. A GFP-tagged Gβ₁γ₂ dimer was expressed in vivo in mice muscle fibres. The GFP fluorescence pattern was consistent with a Gβ₁γ₂ dimer localization in the transverse-tubule membrane. Membrane current and indo-1 fluorescence measurements performed under voltage-clamp conditions reveal a drastic reduction of both L-type Ca²⁺ current density and of peak amplitude of the voltage-activated Ca²⁺ transient in Gβ₁γ₂-expressing fibres. These effects were not observed upon expression of Gβ₂γ₂, Gβ₃γ₂ or Gβ₄γ₂. Our data suggest that the G-protein β₁γ₂ dimer may play an important regulatory role in skeletal muscle excitation–contraction coupling.

Introduction

G-protein-coupled receptors (GPCRs) represent a large family of cell plasma membrane receptors that are involved in the transduction of extracellular signals into cellular responses (for review see Milligan & Kostenis, 2006). Activation of GPCRs following extracellular agonist stimulation results in the activation of two intracellular signalling molecules, Gα-GTP and the free Gβγ dimer. Whereas Gα-dependent signalling pathways involve the activation of soluble messenger cascades such as adenylyl cyclases, cyclic GMP phosphodiesterases or phospholipases Cβ, Gβγ-mediated regulation generally occurs through direct binding of the dimer to target effectors. One such class of targeted receptors is represented by voltage-gated calcium channels (for review see McCudden et al. 2005).

In skeletal muscle, no less than 43 GPCRs were found to be expressed at the protein or mRNA level. Signalling activity of these GPCRs is suspected to play a role in the regulation of several aspects of muscle physiology, including plasticity, differentiation, pain, but also more specifically in the control of glucose uptake and of ion transport and membrane excitability (for review see Jean-Baptiste et al. 2005). Early sets of data also suggested that G-proteins may be involved in the regulation of the excitation–contraction (E–C) coupling process. In skeletal muscle, E–C coupling relies on a tight control of sarcoplasmic reticulum (SR) Ca²⁺ release through the type 1 ryanodine receptor (RyR1) by the voltage-sensing dihydropyridine receptor (DHPR)/Ca_v1.1 Ca²⁺ channel. Activation of the Ca_v1.1 subunit in response to transverse (T-) tubule membrane depolarization directly activates RyR1, allowing rapid mobilization of the SR Ca²⁺ store and the rise in myoplasmic Ca²⁺ concentration that triggers contraction (for review see Dulhunty, 2006). A possible functional role of GPCR signalling in the regulation of E–C coupling was suggested by experiments in skinned skeletal muscle fibres in which it was shown that constitutive activation of G-proteins by GTPγS either enhanced calcium- or caffeine-induced calcium release (Villaz et al. 1989) or directly elicited fibre contraction (Di Virgilio et al. 1986; Somasundaram et al. 1991). Furthermore, GTPγS was also reported to increase DHPR L-type Ca²⁺ current and intramembrane charge movement measured under vaseline-gap voltage-clamp conditions in rat and frog cut skeletal muscle fibres (Garcia et al. 1990). However, the issue became controversial as Lamb & Stephenson (1991) found no effect of GTPγS on both depolarization-induced Ca²⁺ release in skinned rat fibres and on intramembrane charge movement in rat cut fibres. Differences in the experimental conditions between these studies are most probably responsible for these discrepancies. Nevertheless, the use of GTPγS under such conditions remains a rather limited approach. First, in both skinned and cut fibre preparations, critical diffusible factors may potentially be washed out of the myoplasmic compartment, possibly precluding proper G-protein signalling. Second, GTPγS will indiscriminately activate all G-protein signalling in the cell, thus complicating interpretations at the mechanistic level. Finally, along the same line, GTPγS is expected to stimulate both Gα- and Gβγ-signalling pathways, thus further hampering the interpretation of experimental results. Interestingly, recent data on cultured myotubes showed that direct extracellular application of the calcitonin gene-related peptide (CGRP) enhanced voltage-gated Ca²⁺ release, L-type calcium currents and charge movements (Avila et al. 2007). This then supports a functional role of GPCRs in regulating Ca²⁺ influx through Ca_v1.1 and thus potentially E–C coupling.

The aim of the present study was to specifically test the role of the G-protein βγ complex in the control of Ca²⁺ homeostasis in skeletal muscle fibres from mouse. We investigated the effects of the functional G-protein β₁γ₂ dimer on the Ca²⁺ channel properties of the DHPR, as well as on those of the voltage-activated Ca²⁺ transients in fully differentiated adult skeletal muscle fibres. Our results suggest that the G-protein β₁γ₂ dimer is targeted to the T-tubule membrane system where it modulates Ca²⁺ homeostasis by inhibiting the ion channel function of the DHPR and the voltage-activated SR Ca²⁺ release process. Our findings thus provide novel evidence for an efficient G-protein-dependent regulatory mechanism of E–C coupling in adult skeletal muscle.

Methods

Ethical approval

All experiments and procedures were performed in accordance with the guidelines of the local animal ethics committee of University Lyon 1, of the French Ministry of Agriculture (87/848) and of the European Community (86/609/EEC). We have read the article by Drummond (2009) and our experiments comply with the policies and regulations of The Journal of Physiology and UK regulations on animal experimentation.

Plasmid cDNAs

Plamid cDNAs encoding for GFP-tagged Gβ_1-4 and Gγ₂ were created and tagged as described by us previously (Arnot et al. 2000; Feng et al. 2001).

In vivo cDNA electroporation

Experiments were performed using 4- to 8-week-old male OF1 mice (Charles River Laboratories, L’Arbresle, France). In vivo cDNA electroporation was performed as previously described (Weiss et al. 2008). Mice were anaesthetized by i.p. injection of 10 μl g⁻¹ of a mixture of 5% ketamine (Merial, Lyon, France) and 2% xylazine (Bayer, Kiel, Germany). A 50 μl aliquot of 10 μg μl⁻¹ of the plasmid mix cDNAs encoding for the GFP-tagged Gβ construct (GFP-Gβ₁, GFP-Gβ₂, GFP-Gβ₃ or GFP-Gβ₄) along with the G-protein γ₂ subunit was injected within the ventral side of the posterior paws through a 29-gauge needle (Terumo, Leuven, Belgium). The paw was then placed between two flat platinum electrodes, and eight pulses of 200 V cm⁻¹ amplitude of 20 ms duration were applied at 1 Hz (ECM 830 Electro Square Porator, BTX).

Isolation and preparation of FDB muscle fibres

Experiments were performed on single skeletal fibres isolated from the flexor digitorum brevis (FDB) muscles 5 days after electroporation. Procedures for enzymatic isolation of single fibres, partial insulation of the fibres with silicone grease and intracellular dye loading were as previously described (Jacquemond, 1997; Collet et al. 1999, 2004; Collet & Jacquemond, 2002). In brief, mice were killed by cervical dislocation after isofluorane (Sigma Aldrich, Saint Quentin Fallavier, France) inhalation, before removal of the muscles. Muscles were treated with 2 mg ml⁻¹ collagenase type-1 (Sigma Aldrich) in Tyrode solution (see Solutions) for 60 min at 37°C. Single fibres were then isolated by triturating the muscles within the experimental chamber. Fibres expressing the Gβ_xγ₂ dimers were identified from the GFP fluorescence. The major part of a single fibre was then electrically insulated with silicone grease (Rhodia Siliconi Italia, Treviolo, Italy) so that whole-cell voltage-clamp could be achieved on a short portion of the fibre extremity. As previously reported, upon expression of exogenous ryanodine receptors (Legrand et al. 2008) one hallmark of the expression pattern of the Gβγ dimers was that it was not homogenous throughout the muscle fibres but most usually restricted within one given portion of the fibres, around a nucleus. Transfected fibres thus had to be handled with silicone so that the fibre region exhibiting GFP fluorescence was left out of the silicone. However, not all GFP-positive fibres were amenable to this procedure, especially when the region of expression was within the central portion of the fibre. More importantly, in most cases the total portion of fibre left out of silicone was not 100% GFP-positive so that any effect due to the expressed Gβγ dimer was likely to be underestimated in the silicone-free portion of the fibre for measurements of whole-cell membrane currents and indo-1 fluorescence.

Confocal imaging

Confocal images of the GFP fluorescence and di-8-ANEPPS (10 μm, Invitrogen) labelling on freshly isolated fibres were obtained on a Zeiss LSM 5 laser scanning confocal microscope. The microscope was equipped with a ×63 oil immersion objective (numerical aperture 1.4). All experiments were performed at room temperature (20–22°C).

Electrophysiology

An RK-400 patch-clamp amplifier (Bio-logic, Claix, France) was used in whole-cell configuration. Command voltage pulse generation and data acquisition were done using WinWCP software (freely provided by Dr Dempster, University of Strathclyde, Glasgow, UK) driving an A/D, D/A converter (National Instruments, Austin, TX, USA). Analog compensation was systematically used to decrease the effective series resistance. Voltage-clamp was performed with a microelectrode filled with an internal-like solution (see Solutions) of 1–3 MΩ resistance. The tip of the microelectrode was inserted through the silicone within the insulated part of the fibre. Membrane depolarization were applied every 30 s from a holding command potential of −80 mV. A 30 s pulse interval allows the total recovery of DHPRs from the inactivated state (Collet et al. 2003).

Measurement of calcium currents

Voltage-activated Ca²⁺ currents were recorded in response to 1-s-long depolarizing steps to various potentials. The linear leak component of the current was removed by subtracting the adequately scaled value of the steady current measured during a 20 mV hyperpolarizing step applied before each test pulse. The voltage dependence of the peak Ca²⁺ current density was fitted with the following modified Boltzman equation:

with I(V) being the peak current density at the command potential V, G_max the maximum conductance, V_rev the reversal potential, V_1/2 the half-activation potential and k the steepness factor. The voltage dependence of the whole-cell Ca²⁺ conductance was calculated using the following modified Boltzman equation:

with G(V) being the Ca²⁺ conductance at the command potential V.

Measurement of intramembrane charge movement

Charge movement currents were measured and analysed according to previously described procedures (Collet et al. 2003; Pouvreau & Jacquemond, 2005). In brief, adequately scaled control current records elicited by 50-ms-long hyperpolarizing pulses of 20 mV from a holding potential of −100 mV were subtracted from the current elicited by test depolarizing pulses of the same duration to levels ranging between −80 and +20 mV. The amount of charge moved during a test pulse was measured by integrating the ‘on’ portion of the corrected test current record. The calculated charge was normalized to the capacitance of the fibre. The steady-state distribution of the normalized charge was fitted with a two-state Boltzmann function: Q(V) =Q_max/{1 + exp[(V_1/2−V)/k], where Q_max corresponds to the maximal available charge, V_1/2 to the voltage of equal charge distribution and k to the steepness factor.

Ca²⁺ transients measurements using indo-1

Prior to voltage-clamp, the indo-1 dye, diluted at a concentration of 0.2 mm in an intracellular-like solution (see Solutions), was dialysed into the fibre cytoplasm through the microelectrode inserted through the silicone, within the insulated part of the fibre. In order to ease intracellular dialysis, the electrode tip was crushed within the silicon-insulated portion of the fibre by pushing it back and forth a few times towards the bottom of the chamber. Intracellular equilibration of the solution was allowed for a period of 30 min before initiating measurements. Indo-1 fluorescence was measured on an inverted Nikon Diaphot epifluorescence microscope equipped with a commercial optical system, allowing the simultaneous detection of fluorescence at 405 nm (F₄₀₅) and 485 nm (F₄₈₅) by two photomultipliers (IonOptix, Milton, MA, USA) upon 360 nm excitation. Background fluorescence at both emission wavelengths was measured next to each fibre tested and was then subtracted from all measurements.

Calibration of the indo-1 response and [Ca²⁺]_intra calculation

The standard ratio method was used with the parameters: R=F₄₀₅/F₄₈₅, with R_min, R_max, K_D and β having their usual definitions. Results were either expressed in terms of indo-1% saturation or in actual free Ca²⁺ concentration (for details of calculation see Jacquemond, 1997; Csernoch et al. 1998). In vivo values for R_min, R_max and β were measured using procedures previously described (Collet et al. 1999; Collet & Jacquemond, 2002). Corresponding mean values were 0.239 ± 0.004 (n= 9), 1.783 ± 0.028 (n= 9) and 2.128 ± 0.014 (n= 13), respectively. No correction was made for indo-1–Ca²⁺ binding and dissociation kinetics. As a consequence, the calculated Ca²⁺ transients do not scrupulously reflect the true time course of change in free calcium. This, however, will not influence any of the conclusions from the present study (for details, see Jacquemond, 1997).

Solutions

The intracellular-like solution contained (in mm): 120 potassium glutamate, 5 Na₂-ATP, 5 Na₂-phosphocreatine, 5.5 MgCl₂, 5 d-glucose and 5 Hepes adjusted to pH 7.2 with KOH. The standard extracellular solution contained (in mm): 140 TEA-methanesulphonate, 2.5 CaCl₂, 2 MgCl₂, 0.002 TTX and 10 Hepes, adjusted to pH 7.2. For measurement of the slow Ca²⁺ current, the extracellular solution also contained 1 mm 4-aminopyridine and 5 CaCl₂, and fibres were incubated with 0.1 mm EGTA-AM for 30 min before starting the experiment. The Tyrode solution contained (in mm): 140 NaCl, 5 KCl, 2.5 CaCl₂, 2 MgCl₂ and 10 Hepes, adjusted to pH 7.2 with NaOH. For measurements of charge movement the extracellular solution contained (in mm): 140 TEA-methanesulphonate, 0.5 CaCl₂, 4 MgCl₂, 1 4-aminopyridine, 1 CdCl₂, 1 CoCl₂, 10 TEA-Hepes and 0.002 tetrodotoxin, pH 7.2. In addition, fibres were pressure-microinjected with a solution containing 50 mm EGTA diluted in the intracellular-like solution.

Statistics

Least-squares fits were performed using a Marquardt–Levenberg algorithm routine included in Microcal Origin (OriginLab, Northampton, MA, USA). Data values are presented as mean ±s.e.m. for n fibres tested, where n is specified in Results. Statistical significance was determined using Student's t test: *P < 0.05; **P < 0.01; ***P < 0.001; NS, statistically not different.

Results

In vivo expression of the GFP-tagged G-protein β₁γ₂ dimer in adult mouse skeletal muscle fibres

Under the conditions of fibre electroporation used in this study (for details see Methods), a restricted number of fibres (typically 10 to 20) were found to express the G-protein β₁–GFPγ₂ dimer out of the total population of fibres isolated from one given transfected FDB muscle. Figure 1 shows an example of the typical expression pattern of the GFP–Gβ₁γ₂ dimer observed under confocal microscopy. The GFP–Gβ₁γ₂ dimer yielded a repetitive double-row pattern perpendicular to the longitudinal axis of the fibre, consistent with the T-tubule membrane system. This was confirmed by the strong co-localization of the GFP fluorescence with the di-8-ANNEPS membrane straining (Fig. 1C–D). These observations indicate the proper localization of the G-protein β₁γ₂ dimer with regard to the excitation–contraction coupling machinery (Tanaka et al. 2000).

Figure 1. Expression of the GFP-tagged Gβ₁γ₂ dimer in adult muscle fibres.

A, confocal images of the GFP fluorescence from a fibre freshly isolated from a muscle transfected with the cDNAs coding for the GFP–Gβ₁ fusion protein along with the Gγ₂ subunit. B, di-8-ANNEPS staining. C, overlay of the two images. D, normalized mean fluorescence GFP and di-8-ANNEPS profile along the x axis of the box region superimposed on the image in C. GFP–Gβ₁ fusion protein yields subcellular localization consistent with the T-tubule membrane. TS, T-tubule spacing; SL, sarcomere length.

In vivo expression of the GFP-tagged G-protein β₁γ₂ dimer reduces L-type calcium current density

To investigate the effect of G-protein β₁γ₂ dimer on the ionic properties of the Ca_v1.1 calcium channel, L-type calcium currents were recorded in G-protein β₁γ₂ dimer-expressing fibres. Representative Ca²⁺ current traces recorded in response to 1 s depolarizing steps to values ranging between −50 mV and +80 mV, from a holding potential of −80 mV, are shown in Fig. 2A in a control fibre (left panel) and in a Gβ₁γ₂ dimer-expressing fibre (right panel). Control fibres were GFP-negative fibres issued from the same muscles. Figure 2B shows the mean peak Ca²⁺ current density as a function of membrane voltage for control (filled circles) and Gβ₁γ₂ dimer-expressing fibres (open circles). The L-type Ca²⁺ current density was significantly reduced in fibres expressing the Gβ₁γ₂ dimer. For instance, in response to a depolarizing pulse to +10 mV, the mean peak Ca²⁺ current density was decreased by 40% (P= 0.0004) in Gβ₁γ₂-expressing fibres (−5.9 ± 1.1 A/F, n= 6) compared to control fibres (−9.8 ± 0.4 A/F, n= 25). The voltage dependence of the L-type Ca²⁺ current activation was also determined in control and in Gβ₁γ₂-expressing fibres (Fig. 2B, inset). The mean half-activation potential remained unaltered (P= 0.26) in Gβ₁γ₂-expressing fibres (−2.3 ± 2.8 mV, n= 6) compared to control fibres (−5.0 ± 0.9 mV, n= 25). The mean corresponding conductance versus membrane potential relationships are presented in Fig. 2C for control fibres (filled circles) and Gβ₁γ₂-expressing fibres (open circles). The maximal conductance (Fig. 2C, inset) was reduced by 35% (P= 0.001) in Gβ₁γ₂-expressing fibres (103 ± 11 S/F, n= 6) as compared to control fibres (159 ± 7 S/F, n= 25). Electroporation itself had no effect on the L-type Ca²⁺ current density (data not shown). We also tested the capability of other G-protein β subunit isoforms to modulate L-type Ca²⁺ currents. Representative current traces from fibres expressing Gβ₂γ₂ (top panel), Gβ₃γ₂ (middle panel) and Gβ₄γ₂ (bottom panel) are shown in Fig. 3. Corresponding mean values for peak Ca²⁺ current densities and conductances as a function of membrane potential are shown in Fig. 3B and C, respectively. No significant alteration of these parameters was observed. For instance, neither the mean half-activation potential of the L-type Ca²⁺ current, nor the maximal conductance was changed. Taken together, these data demonstrate that the L-type Ca²⁺ current is specifically altered by the G-protein β₁γ₂ dimer in adult skeletal muscle fibres. The possibility that absence of effect of Gβ₂γ₂, Gβ₃γ₂ and Gβ₄γ₂ would result from improper localization of the dimers in regard to the DHPR was explored by examination of the corresponding GFP fluorescence patterns under confocal microscopy. Results showed that, in contrast with the GFP-Gβ₁γ₂-expressing fibres where a double-row pattern was reproducibly observed (Fig. 1), the vast majority of fibres expressing any of the three other dimers yielded a single-row pattern clearly inconsistent with the triadic region. This is illustrated in Fig. 3D, which shows an example of the GFP pattern (green x, y frame and profile) from distinct fibres expressing, from top to bottom, GFP–Gβ₂γ₂, GFP–Gβ₃γ₂ and GFP–Gβ₄γ₂, respectively. The corresponding fluorescence of di-8-ANNEPS (red x, y frame and profile) was also imaged in the case of the GFP–Gβ₂γ₂ example given; the GFP banded pattern was clearly out of the triadic region highlighted by di-8 ANNEPS and its frequency was half that of di-8 ANNEPS. This indicates that the proteins were retained within a different cell compartment that may be close to, or correspond to, longitudinal/endoplasmic reticulum, as was suggested for other expressed proteins in previous works (see Gensler et al. 2001; Volpe et al. 2004). We wish though to stress that this was not a completely black-and-white feature because in a restricted number of fibres expressing either GFP–Gβ₂γ₂ or GFP–Gβ₃γ₂, the GFP fluorescence was either extremely diffuse with no clear pattern or yielding a faint triadic-like double-row pattern (not illustrated). We have no interpretation for these discrepancies other than raising the possibility that parameters involved in triadic targeting of a given dimer may vary in certain fibres. Altogether we believe though that the data provide strong support for the fact that proper localization of a given Gβγ dimer may well be a critical condition determining its functional effect on the DHPR.

Figure 2. Expression of G-protein β₁γ₂ dimers reduces L-type calcium current density in skeletal muscle fibres.

A, representative sets of Ca²⁺ current traces recorded from a control (left panel) and from a Gβ₁γ₂ dimer-expressing fibre (right panel) in response to 1 s depolarizing steps to values ranging between −50 mV and +80 mV from a holding potential of −80 mV. B, corresponding mean voltage dependence of the peak Ca²⁺ current density. Inset presents the mean half-maximal activation potential for control and Gβ₁γ₂-expressing fibres. C, corresponding mean values for the voltage dependence of Ca²⁺ conductance in the two populations. Inset presents the mean maximal conductance for control and Gβ₁γ₂-expressing fibres. The maximal conductance was reduced by 35% (P= 0.001) in the Gβ₁γ₂-expressing fibres.

Figure 3. Expression of G-protein β₂γ₂, β₃γ₂ or β₄γ₂ dimers does not alter L-type calcium currents in skeletal muscle fibres.

A, representative sets of Ca²⁺ current traces recorded from a Gβ₂γ₂ (top panel), Gβ₃γ₂ (middle panel) and Gβ₄γ₂ (bottom panel) -expressing fibres in response to 1 s depolarizing steps to values ranging between −40 mV and +80 mV from a holding potential of −80 mV. B, corresponding mean voltage dependence of the peak Ca²⁺ current density. Inset presents the mean half-maximal activation potential for Gβ₂γ₂, Gβ₃γ₂ and Gβ₄γ₂-expressing fibres. C, corresponding mean values for the voltage dependence of Ca²⁺ conductance in the three populations. Inset presents the mean maximal conductance for Gβ₂γ₂, Gβ₃γ₂ and Gβ₄γ₂-expressing fibres. The dashed line corresponds to the values from the control fibres. D, expression pattern of the GFP-tagged Gβ₂γ₂, Gβ₃γ₂ and Gβ₄γ₂ dimers in adult muscle fibres. Each panel shows a region of interest selected from a confocal image of the GFP fluorescence (green) of a fibre expressing the given Gβ subunit. The fibre expressing Gβ₂γ₂ was also stained with di-8-ANNEPS and the corresponding fluorescent image is shown in red. The graph underneath each set of images shows the normalized fluorescence profile of GFP (and di-8-ANNEPS) along the x axis of the boxed region superimposed on each image.

Expression of the G-protein β₁γ₂ dimer slows down activation and inactivation kinetics of the L-type calcium current

To further investigate the functional impact of G-protein β₁γ₂ expression on the Ca_v1.1 channel, the kinetics of L-type Ca²⁺ currents were analysed in control and Gβ₁γ₂-expressing fibres. Figure 4A shows normalized Ca²⁺ current traces obtained in response to 1 s depolarizing steps to +10 mV (left panel) and +30 mV (right panel) from a holding potential of −80 mV, in a control fibre and in a Gβ₁γ₂-expressing fibre. We found that the Ca²⁺ current activation was slowed in fibres expressing the Gβ₁γ₂ dimer as shown by larger time-to-peak values, which was particularly pronounced at the lowest voltage tested (Fig. 4B). For instance, in response to a depolarizing step to +10 mV, the time to peak of the Ca²⁺ current recorded from Gβ₁γ₂-expressing fibres was, on average, 1.5-fold greater than in control fibres (205 ± 30 ms, n= 6, versus 137 ± 6 ms, n= 25, respectively, P= 0.001). This slowing remained significant for depolarizing step levels up to +40 mV. The time course of Ca²⁺ current decay was determined by fitting the decay phase of the current by a single exponential function (Fig. 4A, bold lines). Mean corresponding values for the time constant (τ) are presented in Fig. 4C for control (filled circles) and Gβ₁γ₂-expressing fibres (open circles). As shown in the figure, in response to a depolarizing step to +10 mV, the time constant of Ca²⁺ current decay in Gβ₁γ₂-expressing fibres was, on average, 1.7-fold greater than in control fibres (262 ± 46 ms, n= 6, versus 156 ± 12 ms, n= 25, P= 0.003). This slowing remained significant for depolarizing step levels up to +30 mV. In adult skeletal muscle, the decay of the calcium current is believed to include a contribution from Ca²⁺ depletion from the T-tubules (see for instance Friedrich et al. 1999, 2001). The graph in Fig. 4D shows the dependence of the time constant of Ca²⁺ current decay at +10 mV upon the peak current density, for control and Gβ₁γ₂-expressing fibres. Data were grouped into four arbitrary classes of peak current density. There was a clear correlation between the time constant of current decay and the peak current density, both in control and Gβ₁γ₂-expressing fibres, consistent with the possibility that T-tubule Ca²⁺ depletion plays a role in current decay. Hence, by reducing T-tubule Ca²⁺ depletion, the decrease in Ca²⁺ conductance in Gβ₁γ₂-expressing fibres could be anticipated to contribute to the slowing of Ca²⁺ current decay. However, comparison of the mean values for the time constant of Ca²⁺ current decay between control and Gβ₁γ₂-expressing fibres within groups of fibres yielding a similar Ca²⁺ current density (3/6 and 6/9 A/F) revealed that the rate of Ca²⁺ current decay remained significantly slower in the presence of the dimer, suggesting that Gβ₁γ₂ by itself also contributes to slow down of Ca²⁺ current inactivation. The slowed decay of the Ca²⁺ current in Gβ₁γ₂-expressing fibres is thus probably the consequence of both a reduction of the Ca²⁺ conductance and an alteration of the voltage-dependent inactivation by the dimer. Since Gβ₁γ₂ slows down Ca²⁺ current inactivation, this could be thought to indirectly be the reason for the observed increased time-to-peak values of the current. However, comparison of the Ca²⁺ current time to peak between control and Gβ₁γ₂-expressing fibres for similar values of Ca²⁺ decay time constant showed that Ca²⁺ current activation remained significantly slowed in the presence of the dimer (not shown), suggesting that Gβ₁γ₂ by itself slows down Ca²⁺ current activation independently from any apparent effect due to change in the inactivation rate.

Figure 4. Expression of G-protein β₁γ₂ dimers slows down L-type calcium current activation and inactivation kinetics.

A, normalized Ca²⁺ current traces recorded from a control and from a Gβ₁γ₂ dimer-expressing fibre in response to 1 s depolarizing steps to +10 mV (left panel) and +30 mV (right panel) from a holding potential of −80 mV. The continuous superimposed bold lines correspond to the results from fitting a single exponential function to the inactivating phase of the current. B, mean values for the time to peak of the Ca²⁺ current for control and Gβ₁γ₂ dimer-expressing fibres. Values were significantly increased in the Gβ₁γ₂ dimer-expressing fibres for potential values from 0 mV to +40 mV. C, mean values for the time constant τ of Ca²⁺ current decay. Values were significantly increased in the Gβ₁γ₂ dimer-expressing fibres at potentials ranging from 0 mV to +30 mV. D, dependence of the time constant of Ca²⁺ current decay at +10 mV upon the peak current density, in control and Gβ₁γ₂-expressing fibres. Values for the time constant were grouped into 4 classes of peak current density.

Effect of a strong depolarizing pre-pulse on the regulation of the L-type calcium current by G-protein β₁γ₂ dimer

It is well known that application of strong depolarizing pre-pulses on voltage-gated Ca²⁺ channels under direct G-protein regulation produces either a current facilitation, and/or a change in kinetics of the Ca²⁺ current (Scott & Dolphin, 1990; Ikeda, 1991). Representative Ca²⁺ current traces in response to a 1 s depolarizing step to +20 mV from a holding potential of −80 mV, before (P1) and after (P2) a strong 180 ms depolarizing pre-pulse to +100 mV are shown in Fig. 5A in a control fibre (left panel) and in a Gβ₁γ₂-expressing fibre (right panel). Figure 5B shows an enlarged view of the same traces after normalization to the same peak amplitude. In Fig. 5C the normalized peak Ca²⁺ current amplitude recorded before and after the depolarizing pre-pulse (P2/P1) was plotted as a function of the duration of the pre-pulse in control (filled circles, n= 13) and Gβ₁γ₂-expressing fibres (open circles, n= 6). Results show that there was either no pre-pulse-induced current recovery from G-protein inhibition or that the effect was sufficiently small to be masked by the inactivation of the Ca²⁺ current induced by the depolarizing pre-pulse. The time to peak of the current plotted as a function of the duration of the depolarizing pre-pulse is shown in Fig. 5D. Although not spectacular, there was a clear tendency for the values recorded in Gβ₁γ₂-expressing fibres to get closer to the ones recorded in control fibres as the pre-pulse duration was increased, indicative of some recovery from the process of Gβ₁γ₂-induced slowing of Ca²⁺ current activation. Figure 5E shows the mean values for the time constant of Ca²⁺ current decay in control fibres and in Gβ₁γ₂-expressing fibres as a function of the depolarizing pre-pulse duration. In control cells, there was a slowing of Ca²⁺ current decay upon increasing the pre-pulse duration. This slowing could be related to the reduction of the current amplitude (caused by the pre-pulse-induced current inactivation) which indirectly slows down the Ca²⁺ decay kinetics by reducing T-tubule Ca²⁺ depletion, as stated above. By contrast, in Gβ₁γ₂-expressing fibres, there was a reduction of the Ca²⁺ current decay time constant, the value of which clearly converged towards control values for the longest pre-pulse durations. The inset in Fig. 5E shows the same data, but values for the time constant of Ca²⁺ decay in each fibre were normalized to the value measured in the absence of pre-pulse. Effect of the pre-pulse was clearly opposite in the control fibres and in the Gβ₁γ₂-expressing ones; thus, the extent of reversion of Gβ₁γ₂-induced slowing of Ca²⁺ decay by the pre-pulse is likely to be under-estimated because the effect observed in control fibres should also be present in the Gβ₁γ₂-expressing ones. Data thus clearly support the capability of a depolarizing pre-pulse to reverse the slowing of Ca²⁺ current inactivation induced by Gβ₁γ₂ and thus the likelihood of a direct G-protein regulation of the DHPR Ca²⁺ channel activity. Also important is the fact that depolarizing pre-pulses did not speed up Ca²⁺ current inactivation in control conditions, which suggests that DHPR channels are not under an endogenous tonic Gβγ regulation in control fibres. Overall, although no effect of the pre-pulses was detected on the peak current amplitude, the partial recovery from the slowing of current kinetics is consistent with the effect of the G-protein β₁γ₂ dimer on the L-type Ca²⁺ current to occur, at least partially, via the binding of the Gβ₁γ₂ dimer to the Ca_v1.1 subunit.

Figure 5. A strong depolarizing pre-pulse allows partial recovery from the slowing of L-type calcium current kinetics induced by Gβ₁γ₂.

A, representative Ca²⁺ current traces recorded from a control (left panel) and from a Gβ₁γ₂ dimer-expressing fibre (right panel) in response to a 1 s depolarizing step to +20 mV from a holding potential of −80 mV, before (P1) and after (P2) a strong depolarizing pre-pulse (PP) to +100 mV. B, enlarged view of the Ca²⁺ current traces shown in A after normalization to the same peak amplitude, with (P2) and without (P1) pre-pulse. C, mean normalized values of P2/P1 peak Ca²⁺ current amplitude as a function of the pre-pulse (PP) duration. D, mean values for the time to peak of the Ca²⁺ current recorded in P2 as a function of the pre-pulse duration. The inset shows the corresponding mean values after normalization to the time to peak of the current measured in absence of pre-pulse. E, mean values for the time constant of decay of the Ca²⁺ current recorded in P2 as a function of the pre-pulse duration. The inset shows the corresponding mean values after normalization to the time constant measured in absence of pre-pulse.

G-protein β₁γ₂ dimer expression alters voltage-activated calcium transients

In order to further investigate the possible functional role of G-proteins in the control of excitation–contraction coupling and, from a more general point of view, in the control of intracellular Ca²⁺ homeostasis, Ca²⁺ transients were recorded in control and Gβ₁γ₂-expressing fibres under voltage-clamp conditions. Representative [Ca²⁺] traces from a control (left panel) and from a Gβ₁γ₂-expressing fibre (left panel), in response to successive depolarizing pulses of 5, 10, 20, 50 and 100 ms duration to +10 mV from a holding potential of −80 mV are shown in Fig. 6A. Obviously, the amplitude of the Ca²⁺ transients was dramatically diminished in the Gβ₁γ₂-expressing fibre. Figure 6B–D show mean values from that series of measurements for peak Δ[Ca²⁺], time constant (τ) of [Ca²⁺] decay and final Δ[Ca²⁺] in control (filled circle, n= 10) and Gβ₁γ₂-expressing fibres (open circle, n= 6). Measurements were performed on individual [Ca²⁺] traces calculated from each fibre. The time constant of decay was obtained by fitting a single exponential plus constant function to the [Ca²⁺] decline; the value for the constant from the fit was used to calculate the final Δ[Ca²⁺] (Fig. 6D). In response to a depolarizing step of 20 ms duration, the mean peak Δ[Ca²⁺] was 43% less (P= 0.0003) in Gβ₁γ₂-expressing fibres (0.63 ± 0.10 μm, n= 6) than in control ones (1.11 ± 0.05 μm, n= 10). There was no significant difference in the values for the time constant of [Ca²⁺] decay and final Δ[Ca²⁺] level between control and Gβ₁γ₂-expressing fibres. Mean values for resting [Ca²⁺] also did not statistically differ between the two groups; for instance, the mean baseline level of [Ca²⁺] measured before the first depolarizing pulse was applied was 14.4 ± 1.3 nm and 10.9 ± 2.2 nm, in control and Gβ₁γ₂-expressing fibres, respectively. Taken together, these results indicate that expression of the G-protein β₁γ₂ dimer dramatically alters voltage-induced Ca²⁺ release without affecting the ability of the cells to regulate the Ca²⁺ concentration at rest.

Figure 6. Expression of G-protein β₁γ₂ dimers strongly alters the voltage-activated Ca²⁺ transient in skeletal muscle fibres.

A, representative indo-1 [Ca²⁺] traces from a control (left panel) and from a Gβ₁γ₂ dimer-expressing fibre (right panel) in response to depolarizing steps of increasing duration to +10 mV from a holding potential of −80 mV. The inset in each panel shows the corresponding mean (continuous traces) ±s.e.m. (grey shading) [Ca²⁺] traces. B–D, corresponding mean ±s.e.m. values of peak change in [Ca²⁺], time constant (τ) of [Ca²⁺] decay after the end of the pulse and final change in [Ca²⁺] level, respectively, measured from 10 control fibres and 6 Gβ₁γ₂ dimer-expressing fibres. The peak change in [Ca²⁺] was reduced by 43% (P= 0.0003) (20 ms depolarizing step) in the Gβ₁γ₂-expressing fibres. Values for the final change in [Ca²⁺] level were from the exponential fits. When not visible, errors bars are smaller than the mean dot.

Voltage sensing function of the DHPR is unaffected by the G-protein β₁γ₂ dimer

The question of whether the observed effects of the β₁γ₂ dimer on the Ca²⁺ current and on the Ca²⁺ transient were associated with a reduction of the amount of DHPR in the membrane was of most immediate interest. In order to address it, we measured intramembrane charge movement in control and Gβ₁γ₂-expressing fibres. Figure 7A shows illustrative traces of charge movement currents from a control fibre and from a Gβ₁γ₂-positive fibre in response to 50 ms depolarizing pulses from −100 mV to the indicated values. Altogether we found no qualitative difference between the records obtained in the two sets of fibres. Figure 7B shows the voltage dependence of the corresponding mean values for the ‘on’ charge. There was obviously no reduction of the amount of charge in the Gβ₁γ₂-expressing fibres. Fitting a two-state Boltzmann distribution to the data obtained in each fibre gave mean values for Q_max, V_1/2 and k of 18.4 ± 2.8 nC μF⁻¹, −29.1 ± 4.1 mV and 12.9 ± 1.8 mV in control fibres (n= 8), and 19.9 ± 3.7 nC μF⁻¹, −25.3 ± 3.9 mV and 15.6 ± 1.6 mV in Gβ₁γ₂-expressing fibres (n= 5), respectively. None of the parameters significantly differed between the two sets of fibres. Thus, despite the effect of Gβ₁γ₂ expression on the L-type Ca²⁺ current density and on the peak Ca²⁺ transient, there was no associated reduction of intramembrane charge movement. The molecular mechanism(s) targeted by Gβ₁γ₂ thus has to lie beyond the voltage-sensing steps of the DHPR.

Figure 7. Intramembrane charge movement in Gβ₁γ₂-expressing fibres.

A, representative charge movement records from a control fibre and from a GFP-Gβ₁γ₂-positive fibre, measured in response to 50 ms depolarizing steps to the indicated values of membrane potential. B, mean voltage distribution of the ‘on’ charge in control fibres (n= 8) and in GFP-Gβ₁γ₂-positive fibre (n= 5). Superimposed continuous lines correspond to a two-state Boltzmann distribution calculated using the mean parameters obtained from fits to the individual sets of data in the two groups of fibres (see text for details).

Discussion

We demonstrate that expression of the G-protein β₁γ₂ dimer in intact adult mammalian skeletal muscle fibres alters both the Ca²⁺ channel function of the DHPR and the process of voltage-activated Ca²⁺ release from the SR. These results provide strong support for the ability of Ca_v1.1 proteins to undergo G-protein inhibition. In vivo expression of the G-protein β₁γ₂ dimer in adult mouse muscle was successful and yielded a very distinct subcellular pattern consistent with a specific localization of the dimer in the T-tubule membrane system. This is in accordance with previous biochemical studies showing the presence of the G-protein β subunits in T-tubule membrane preparations purified from rabbit skeletal muscle (Toutant et al. 1988), as well as with immunostaining of endogenous G-protein β subunits, also in rabbit skeletal muscle (Toutant et al. 1990). A clear T-tubule consistent pattern could, however, not be reproducibly observed upon expression of either the β₂γ₂, the β₃γ₂ or the β₄γ₂ dimers, which were all found to also have no effect on the DHPR Ca²⁺ channel function. Although we cannot completely exclude that functional properties specific to the Gβ₁ subtype play a role in the observed regulation of the DHPR, proper targeting in the T-tubule membrane may well be a critical parameter determining the efficiency of Gβ₁. One could then also speculate that the β₁γ₂ dimer acts through a disruptive non-specific effect only related to its over-expression in that particular restricted area of the muscle fibres. However, we previously showed that over-expression of wild-type caveolin-3 (Couchoux et al. 2007) or of the C71W mutant form of caveolin-3 (Weiss et al. 2008), which both localize in the T-tubule membrane, have no effect on the DHPR Ca²⁺ current. We thus believe that the possibility of a non-specific effect is quite unlikely.

Possible mechanism of G-protein β₁γ₂ modulation of the L-type Ca²⁺ current

Previous studies aiming at exploring the role of G-proteins on E–C coupling in adult muscle used GTPγS on either skinned skeletal muscle fibre preparations (Di Virgilio et al. 1986; Lamb & Stephenson, 1991; Somasundaram et al. 1991) or vaseline-gap voltage-clamped cut fibres (Garcia et al. 1990), which provided controversial results. As detailed earlier, the use of GTPγS under such conditions is a somewhat limiting strategy. In this context, the in vivo expression of exogenous functional G-protein β₁γ₂ dimer in intact fully differentiated muscle fibres provides a more defined and specific approach. In particular, it should bypass the possible consequences of activation of multiple endogenous G-protein signalling pathways.

On average, Gβ₁γ₂ positive fibres yielded a decrease in Ca²⁺ current amplitude corresponding to a 35% drop in peak conductance. This is a very substantial alteration, especially considering that it is likely to be heavily underestimated because of the local expression of the Gβ₁γ₂ dimer within the whole-cell voltage-clamped portion of fibre (see Methods). Of importance, we found no alteration of intramembrane charge movement in the Gβ₁γ₂-expressing fibres, demonstrating that the reduction of Ca²⁺ current density was not due to a loss of DHPR in the T-tubule membrane and that the overall effects of Gβ₁γ₂ did not involve an alteration of the voltage-sensing steps of the DHPR function. This also excludes the possibility that inhibition in DHPR current would have resulted from Gβ₁γ₂ increasing expression of Rem/Gem-related proteins, previously shown to inhibit DHPR channel and voltage-sensing function by reducing DHPR expression (Bannister et al. 2008). Conversely, we cannot totally exclude that over-expression of Gβ₁γ₂ could sequestrate a tonic stimulatory effect of Gα subunit since it was demonstrated that Ca_v1.1 channel in cultured muscle cells is modulated by the cAMP-dependent protein kinase (Johnson et al. 1997). Still, there is to our knowledge very limited, if any, indication that this is also true in adult muscle fibres. Furthermore, if such a mechanism took place, we could well expect observing the same phenomenon in Gβ₂γ₂-, Gβ₃γ₂- and Gβ₄γ₂-expressing fibres, which clearly was not the case.

The G-protein βγ dimer is known to directly regulate neuronal voltage-gated Ca²⁺ channels (Ca_v2.1, Ca_v2.2 and Ca_v2.3) (for review, see De Waard et al. 2005; Tedford & Zamponi, 2006) through direct binding of the Gβγ dimer onto various structural elements of the Ca_v2.x subunit. Those include the loop linking transmembrane domains I and II (De Waard et al. 1997; Zamponi et al. 1997), the amino-terminal (Agler et al. 2005) and possibly the carboxy-terminal domain (Qin et al. 1997). Our results showing a pre-pulse-dependent reversal of some of the kinetic effects suggest that Ca_v1.1 L-type channels can also be directly regulated by G-proteins, although we do not presently have biochemical evidence to demonstrate a protein–protein interaction. It should be remembered that complete absence of effects of pre-pulses would, though, not have been sufficient to exclude a direct regulation, as for instance Ca_v3.2 channels were shown to be specifically regulated by the Gβ₂γ₂ dimer through direct binding onto the intracellular loop linking domains II and III of the channel, the regulation being though totally insensitive to a strong depolarizing pre-pulse (Wolfe et al. 2003). Also worth mentioning is the fact that Ca_v1.x channels were initially thought not to be regulated by direct G-protein interaction because they lack the Gβγ binding QXXER motif present in the I–II loop of Ca_v2.x channels. However, more recently, it was demonstrated that the Ca_v1.2 channel is sensitive to G-protein inhibition through a direct binding of the Gβγ dimer onto the amino- and carboxy-terminal domains of the channel (Ivanina et al. 2000).

The so far limited amount of evidence concerning G-protein regulation of Ca_v1.1 L-type channels is probably related to difficulty in heterologous expression of Ca_v1.1 channels. Here, the fully differentiated skeletal muscle fibre preparation allowed detailed investigation of the role of the G-protein βγ dimer on the Ca_v1.1 channel activity. Although we cannot totally exclude that over-expression of Gβ₁γ₂ may have activated another signal transduction pathway to modulate Ca_v1.1 activity, several considerations argue against such a mechanism. First, the effect is specific as only the G-protein β₁ subunit modulated channel activity. This is reminiscent of data obtained with the Ca_v3.2 T-type channel that was shown to be solely mediated by the G-protein β₂ subunit, with other Gβ subunits being ineffective (Wolfe et al. 2003; DePuy et al. 2006). Second, the slowing of activation and inactivation kinetics of L-type Ca²⁺ currents in Gβ₁γ₂-expressing fibres, and its reversal by a depolarizing pre-pulse, fits with the idea of direct G-protein regulation (Marchetti et al. 1986; Scott & Dolphin, 1990; Ikeda, 1991; Zamponi, 2001). Finally, previous biochemical studies suggested that the G-protein β subunit associates with the DHPR, although the molecular determinants involved in this interaction remained unidentified (Hamilton et al. 1991). Altogether, we favour the idea of a direct G-protein regulation of the DHPR Ca²⁺ channel activity.

Possible mechanism of G-protein β₁γ₂ modulation of the excitation–contraction coupling

Functional coupling between the DHPR and RyR1 occurs through direct or indirect interactions of possibly several cytosolic determinants of Ca_v1.1 with RyR1 (for recent review see Bannister, 2007). Among them the II–III loop is established to play a key role in E–C coupling (Tanabe et al. 1990; Lu et al. 1994) and alteration of the corresponding interaction alters both the L-type Ca²⁺ current density and the amplitude of the voltage-activated Ca²⁺ transient (Bannister et al. 2009). Other structural determinants of the Ca_v1.1 subunit postulated to be involved in the regulation of E–C coupling include the I–II loop (via the voltage-gated Ca²⁺ channel β_1a subunit (Gregg et al. 1996; Strube et al. 1996), the carboxy-terminal domain (Slavik et al. 1997; Sencer et al. 2001) and, although probably less critical, the III–IV loop (Leong & MacLennan, 1998; Weiss et al. 2004). As there is no evidence of G-protein βγ dimer interaction with RyR1, the inhibition of voltage-activated Ca²⁺ release observed in the Gβ₁γ₂-expressing fibres is probably not the consequence of a direct regulation of RyR1, but may instead be mediated by the binding of the dimer to the Ca_v1.1 subunit. Moreover, our observation that the expression of G-protein β₁γ₂ dimer alters both the L-type Ca²⁺ current and the voltage-activated Ca²⁺ release is suggestive of a common regulation mechanism. It is well known that the interaction of the II–III loop of the Ca_v1.1 subunit with RyR1 is responsible for a bidirectional coupling: (i) an orthograde coupling characterized by a voltage-induced Ca²⁺ release via RyR1, and (ii) a retrograde coupling manifested by an increase of the L-type Ca²⁺ current via the DHPR (Grabner et al. 1999). Hence, it is tempting to propose that the G-protein β₁γ₂ dimer may control simultaneously the L-type Ca²⁺ current and the voltage-activated Ca²⁺ release by modulating the bidirectional coupling between the DHPR and RyR1.

Potential physiological consequences of Gβ₁γ₂ modulation of Ca_v1.1

Our data raise the intriguing possibility that the Gβ₁γ₂-mediated inhibition of E–C coupling revealed here could be triggered by the activation of specific GPCRs that are coupled selectively to Gβ₁γ₂. This could obviously be of very strong relevance for the regulation of muscle function. Many GPCRs are expressed or presumably expressed in skeletal muscle, some of them being devoid of known agonists (Jean-Baptiste et al. 2005). So far only the calcitonin gene-related peptide (CGRP) was shown to be effective in the muscle E–C coupling context, stimulating both DHPR Ca²⁺ current and Ca²⁺ release (Avila et al. 2007). However, results were obtained on cultured myotubes, i.e immature cells in the process of maturation and differentiation. Under these conditions, it is difficult to assess whether CGRP accelerates the process of maturation/differentiation or if it acutely regulates the E–C coupling process. The time course of the reported CGRP action (several hours), and the fact that effects were attenuated at later time during myotube differentiation actually favoured a maturation effect.

Identification of GPCRs involved in G-protein-mediated inhibition of E–C coupling, and of their respective extracellular agonists thus still represents a challenging task. Finally, a functional association of G-protein βγ dimer with the dystrophin glycoprotein complex (DGC) was shown to be present in skeletal muscle (Xiong et al. 2009). One then could raise the possibility that disruption of the DGC in muscular dystrophies (for review see Durbeej & Campbell, 2002) may release free Gβγ dimer, which in turn could alter Ca²⁺ homeostasis and excitation–contraction coupling and contribute to muscle diseases.

Overall, our data constitute the first description of Ca_v1.1 modulation by G-protein βγ subunits. While further work will be required to ascertain its consequences for E–C coupling, the direct modulation of Ca_v1.1 by this pathway may provide a previously unrecognized means of regulating muscle physiology.

Glossary

Abbreviations

FDB

flexor digitorum brevis

GPCR

G-protein-coupled receptor

DHPR

dihydropyridine receptor

RyR

ryanodine receptor

SR

sarcoplasmic reticulum

Author contributions

N.W. and V.J. designed the study; N.W., C.L., S.P., H.B. and V.J. performed experiments; N.W. and V.J. analysed the data; N.W., B.A., G.W.Z., M.D.W. and V.J. wrote the paper. This study was conducted at the University Claude Bernard, Villeurbanne, France. All authors approved the final version of the manuscript.

Author's present address

N. Weiss: Grenoble Institut des Neurosciences – INSERM U836, Bât. Edmond J. Safra, chemin Fortuné Ferrini, Site santé La Tronche BP170, 38042 Grenoble Cedex 9, France.