• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of jphysiolThe Journal of Physiology SiteMembershipSubmissionJ Physiol
J Physiol. Apr 1, 2006; 572(Pt 1): 269–280.
Published online Jan 19, 2006. doi:  10.1113/jphysiol.2005.104406
PMCID: PMC1779648

A possible role of the junctional face protein JP-45 in modulating Ca2+ release in skeletal muscle


We investigated the functional role of JP-45, a recently discovered protein of the junctional face membrane (JFM) of skeletal muscle. For this purpose, we expressed JP-45 C-terminally tagged with the fluorescent protein DsRed2 by nuclear microinjection in myotubes derived from the C2C12 skeletal muscle cell line and performed whole-cell voltage-clamp experiments. We recorded in parallel cell membrane currents and Ca2+ signals using fura-2 during step depolarization. It was found that properties of the voltage-activated Ca2+ current were not significantly changed in JP-45–DsRed2-expressing C2C12 myotubes whereas the amplitude of depolarization-induced Ca2+ transient was decreased compared to control myotubes expressing only DsRed2. Converting Ca2+ transients to Ca2+ input flux using a model fit approach to quantify Ca2+ removal, the change could be attributed to an alteration in voltage-activated Ca2+ permeability rather than to altered removal properties or a lower Ca2+ content of the sarcoplasmic reticulum (SR). Determining non-linear capacitive currents revealed a reduction of Ca2+ permeability per voltage-sensor charge. The results may be explained by a modulatory effect of JP-45 related to its reported in vitro interaction with the dihydropyridine receptor and the SR Ca2+ binding protein calsequestrin (CSQ).

A rapid rise in intracellular Ca2+ concentration activates force in skeletal muscle cells (Melzer et al. 1995; Bers, 2001). Ca2+ is stored in the terminal cisternae of the sarcoplasmic reticulum (SR) and buffered by the low-affinity, high-capacity Ca2+ binding protein calsequestrin (CSQ). It is released under the control of the membrane potential of the transverse tubules (TTs), which conduct the action potential from the cell surface into the cell. The released Ca2+ is bound by its target protein troponin C on the actin filament to induce force and is re-sequestered to the SR by the SR transport Ca2+ ATPase.

The Ca2+ mobilization involves opening of ryanodine receptors (RyRs) in the membrane region of the terminal cisternae facing the TTs, termed ‘junctional face membrane’ (JFM). Voltage-dependent activation of dihydropyridine receptors (DHPRs) in the TT membrane leads to activation of the RyRs by conformational coupling across the junctional gap separating the JFM and TT. The α1-subunit of the DHPR serves as the voltage sensor in the Ca2+ release process. It is currently thought that step depolarization in voltage-clamp experiments first rapidly activates a flux of Ca2+ from the SR and then very slowly a Ca2+ inward flux from the TTs (L-type Ca2+ current) (Brum et al. 1987; Friedrich et al. 1999; Szentesi et al. 2001). Both depend on the voltage-sensing properties of the DHPRs (Melzer et al. 1995).

In addition to the main constituents of the Ca2+ signalling process, a number of proteins with still ill-defined functions have been identified, which are associated with the α1-subunit of the DHPR in the TT membrane (Walker & De Waard, 1998; Arikkath & Campbell, 2003) or with the RyRs in the JFM (Caswell et al. 1991; Knudson et al. 1993; Guo & Campbell, 1995; Jones et al. 1995; MacKrill, 1999). One of these proteins is a recently discovered constituent of the JFM with an apparent molecular mass of 45 kDa that has been termed JP-45 (Zorzato et al. 2000). Co-localization of JP-45 and RyR1 has been indicated by an overlapping striation pattern in adult rat muscle fibres (Anderson et al. 2003). The protein contains 332 amino-acid residues and exhibits a single trans-membrane segment. The protein is highly charged, in particular the domain facing the SR lumen. The larger number of positively charged amino-acid residues makes it a basic protein. Co-immunoprecipitation experiments demonstrated in vitro interaction with the DHPR and CSQ (Anderson et al. 2003). This raises the question of whether JP-45 acts as a modulator of voltage-controlled Ca2+ entry or Ca2+ release. So far, no functional data are available and it is not known whether JP-45 exhibits interactions with components of the Ca2+ release system in vivo. One approach to assess possible functional effects on muscle excitation–contraction (EC) coupling is the over-expression of the protein in normal muscle cells.

For the present experiments, we chose the skeletal muscle cell line C2C12. C2C12 myotubes show great similarities in Ca2+ signalling to adult muscle fibres (Schuhmeier & Melzer, 2004; Ursu et al. 2005) and can therefore serve as a model system for mature EC coupling. We injected plasmids encoding fluorescent fusion proteins of JP-45 into the nuclei of C2C12 myotubes, observed the intracellular expression pattern and studied function (i.e. Ca2+ inward current, gating charge movements and Ca2+ release) under voltage-clamp conditions. The results suggest that JP-45 alters the voltage-controlled Ca2+ permeability of the SR.


Cell culture

C2C12 cells, purchased from the American Tissue Culture Collection (ATCC, Manassas, VA, USA), were cultured in growth medium (Dulbecco's modified Eagle's medium, DMEM), supplemented with 10% fetal bovine serum as described by Schuhmeier et al. (2003). To induce myotube formation and differentiation, cells were cultured in collagen-coated flasks containing DMEM supplemented with 2% horse serum. One day prior to experiments, myotubes were transferred from flasks onto collagen- and carbon-coated coverslips using a mild trypsin treatment.

Expression plasmids

The following plasmids were purchased from Clontech BD Biosciences (Heidelberg, Germany): pEGFP-C1, pDsRed2-N3 and pDsRed2-ER. pDsRed2-ER has been designed for fluorescent labelling of the endoplasmic reticulum (ER). A plasmid (pGFP-α1C) encoding the green fluorescent protein (GFP)-tagged α1-subunit of the cardiac L-type Ca2+ channel (CaV1.2) (Grabner et al. 1998) was kindly provided by M. Grabner and B. E. Flucher (Innsbruck Medical University).

The coding sequence of JP-45 (Anderson et al. 2003) was inserted ‘in frame’ into the red fluorescent protein coding vector pDsRed2-N3 and into the enhanced green fluorescent protein coding vector pEGFP-C1 resulting in plasmids encoding a C-terminally DsRed2-tagged JP-45, termed JP-45–DsRed2, and an N-terminally EGFP-tagged JP-45, termed GFP–JP-45. The plasmids were generated in Basel and shipped to Ulm for expression and functional testing.

Nuclear injection of plasmids

DNA solutions in sterile water (Aqua ad iniectabilia, Braun, Melsungen, Germany) were diluted to a final concentration of 0.125 μg μl−1 for pEGFP-C1 and pDsRed2-N3, to 0.25 μg μl−1 for pJP-45–DsRed2 and pEGFP–JP-45 and to 0.5 μg μl−1 for pGFP-α1C (Beam & Franzini-Armstrong, 1997). Injection capillaries (Femtotips, Eppendorf, Hamburg, Germany) were filled with the DNA solution and a commercial microinjection system (Injectman NI 2 micromanipulator and Femtojet pressure delivering device, Eppendorf) was used to inject into myotubes. Injections were performed into one nucleus per myotube by applying a 200-hPa pressure pulse of 0.3-s duration. A successful injection was indicated by a slight swelling and brighter appearance of the nucleus. Cells were placed in Hepes-buffered medium before performing injections; afterwards, the medium was exchanged for normal differentiation medium. Cells were voltage clamped 48 h later.

Confocal imaging

To image cells expressing fluorescent fusion proteins we used a Radiance 2000 confocal scanner (Bio-Rad, Hemel Hempstead, UK) adapted to an Eclipse TE300 inverted fluorescence microscope (Nikon). GFP fluorescence was excited by an Argon laser (488 nm) and filtered by an HQ 515/30 nm emission filter while for the DsRed2 fluorescence, a green He–Ne laser (543 nm) and an E 570 LP emission filter were used. The Argon and He–Ne lasers had nominal powers of 13 and 2.5 mW, respectively, and were used at 10–25% and 60–80% of their maximal power, respectively. Image acquisition was performed using LaserSharp 2000 (Bio-Rad).


The experimental solutions had the following compositions. External solution (mm): tetraethyl-ammonium hydroxide (TEAOH) 130, HCl 127, CaCl2 10, MgCl2 1, 4-aminopyridine 2.5, glucose 5, tetrodotoxin (TTX) 0.00125 and Hepes 10; pH adjusted to 7.4 with HCl. Internal (pipette) solution (mm): CsOH 145, aspartic acid 135, EGTA 5, HCl 2, Hepes 10, CaCl2 0.5 (free Ca2+, 10−5), Na2ATP 0.84, MgATP 4.16 (Mg : ATP ratio (according to supplier), 1.4; free Mg2+, 1), sodium creatine phosphate 5 and K5-fura-2 0.2; pH adjusted to 7.2 with CsOH. The program CalcV22 (Föhr et al. 1993) was used to calculate the free ion concentrations in the internal solution.

Whole-cell patch-clamp experiments were performed at room temperature (20–23°C) as described by Schuhmeier et al. (2003) and Schuhmeier & Melzer (2004). Cells expressing fluorescence from the marker proteins (see below) were further selected for compact shape suitable for single-electrode voltage clamping. Leak resistance and capacitance were determined by applying small positive and negative voltage steps (amplitude, 10 mV; duration, 25 ms) from the holding potential of −90 mV. Correction of residual linear current components (leak and capacitive) not compensated by the analog electronics was performed by using a control pulse from −90 to −110 mV applied 400 ms before each test pulse. Non-linear charge movements were determined by integrating the early transient outward component of the leak and linear capacitance-corrected current traces.

Current recordings I(V) were normalized by the linear capacitance to obtain current densities i(V). The voltage-dependence of the Ca2+ current density iCa(V) was least squares-fitted with eqn (1) and eqn (2):

equation image
equation image

where gCa,max and VCa are maximal normalized conductance and reversal potential, respectively, of the L-type Ca2+ current, f(V) is the voltage dependence of activation, V0.5 is the voltage of half maximal activation and k determines voltage sensitivity.

Fluorimetry and analysis of Ca2+ signals

For the observation of intracellular GFP fluorescence in the patch-clamp experiments, we used an excitation filter BP 470/20, a beam splitter FT 493 and an emission filter BP 505–530 (all Zeiss). For identification of DsRed2 fluorescence in cells, an excitation filter HQ 545/30 (AHF, Tübingen), a beam splitter Q 570 LP (Zeiss) and an emission filter HQ 610/75 (Zeiss) were used. Myotubes were loaded with fura-2 by diffusion from the patch-pipette containing internal solution (see above). Ca2+-dependent fluorescence changes excited at 380 nm (F380) were recorded at 515 nm. After background correction, the F380 recordings were normalized by measurements of fluorescence excited at 360 nm (F360), which preceded each voltage-clamp activation as described by Schuhmeier et al. (2003). The flux of Ca2+ mobilization (Ca2+ input flux) during depolarizations was calculated using a Ca2+-binding model (Schuhmeier & Melzer, 2004). The model contained the indicator dye described by fluorescence ratio (R = F380/F360) at zero dye saturation (Rmin), ratio at full dye saturation (Rmax), rate constants kon,Dye, koff,Dye and dye concentration [Dye]total, a saturating buffer (S) representing EGTA (parameters kon,S, koff,S and [S]total) and an uptake mechanism (rate constant kuptake). [Dye]total and [S]total were set to values of 0.2 mm, and 5 mm, respectively. Traces generated with this model were least squares-fitted to the relaxation phases of consecutive voltage-activated fluorescence transients by optimizing the parameters koff,Dye, kon,S, koff,S and kuptake. For the determination of input flux, the free concentration of myoplasmic Ca2+ and the estimated occupancies of the model compartments were summed and the time derivative calculated. Free Ca2+ concentration was determined using background- and bleaching-corrected fura-2 fluorescence ratio signals and included reversal of the filtering effect resulting from the binding kinetics of the indicator (Klein et al. 1988). For Rmin, Rmax and the dissociation constant of the dye (KDye = koff,Dye/kon,Dye) the following values were used: 2.84, 0.68 and 276 nm, respectively. Differential equations were solved using Euler's method. Recordings were smoothed by a digital filter that adjusts its bandwidth automatically to the signal dynamics (for a detailed description of the procedures see Schuhmeier et al. 2003 and Schuhmeier & Melzer, 2004).

General analysis and statistics

General calculations and non-linear curve fitting were performed using Excel (Microsoft). Data are presented and plotted as means ± s.e.m. (n = number of experiments) for averaged values, if not indicated otherwise. To compare average values of two independent data sets, Student's t test was used to identify statistically significant differences (P = 0.05).


C2C12 myotubes for functional expression studies in EC coupling

In the present study we used C2C12 cells to investigate effects of the junctional face protein JP-45 on voltage-dependent Ca2+ signalling in myotubes. C2C12 myotubes express characteristic muscle proteins (Blau et al. 1983) and show voltage-activated Ca2+ inward current and Ca2+ release flux similar to adult skeletal muscle cells (Schuhmeier et al. 2003; Schuhmeier & Melzer, 2004).

For heterologous expression of proteins in C2C12 myotubes, we initially examined two frequently applied transfection methods, using calcium phosphate precipitation and Fugene 6 (Roche). Subsequently we tried nuclear injection of the cDNA-carrying plasmids. The effectiveness was tested with mammalian expression plasmids encoding the fluorescent proteins EGFP and DsRed2 (pEGFP-C1 and pDsRed2-N3, respectively). Of the three methods tested, nuclear injection led to the highest efficiency. The percentage of injected cells expressing the fluorescent marker varied between 10% and 75%. Expression started earlier (~6 h) compared to cDNA-carrier transfection (~24 h). We therefore used the nuclear injection for all experiments presented in this study.

Examples of myotubes expressing EGFP and DsRed2, respectively, are shown in Fig. 1A and B. Both fluorescent proteins exhibited a rather homogeneous distribution in the cytoplasm, as expected for soluble proteins. Figure 1C shows the result of injecting the plasmid pDsRed2-ER designed for fluorescent labelling of the endoplasmic reticulum (ER) in living cells. The DsRed2-ER subcellular marker is composed of the targeting signal sequence of calreticulin, the DsRed2 fluorescent protein coding sequence and the KDEL retention signal of calreticulin (Barton & MacLennan, 2004). The KDEL sequence is recognized by receptors, present in vesicular tubular clusters of the ER and the Golgi apparatus, that capture soluble ER resident proteins and carry them back to the ER (Pelham, 1998).

Figure 1
Expression of fluorescent proteins in C2C12 myotubes

In our C2C12 myotubes, DsRed2-ER labelled a network distributed throughout the cytoplasm, which was denser in the vicinity of the nucleus (Fig. 1C). A similar ‘net-like’ pattern of fluorescence expression was seen when pEGFP–JP-45 (Fig. 1D) or pJP-45–DsRed2 (Fig. 1E) were injected. This is consistent with JP-45 being expressed in ER membranes as indicated by Anderson et al. (2003) for COS-7 cells transfected with pEGFP–JP-45. To obtain further information on the subcellular distribution of JP-45, we coinjected pEGFP–JP-45 and pDsRed2-ER. At 12 h after injection, green and red fluorescence were found to be colocalized in the vicinity of the injected nucleus (not shown). However, later (at 18 and 48 h), the fluorescent patterns were different from each other (Fig. 1F). Previous studies suggest that the ER is transformed during differentiation into the functionally distinct SR through gradual replacement of generic ER proteins with specific SR proteins (Villa et al. 1993; Sanger et al. 2004). Our observations are consistent with JP-45 being generated in the ER and then being targeted to specialized regions of this compartment, whereas DsRed2-ER is retained in the original ER by the KDEL retrieval system.

Test for functional expression

The imaging results described above indicated strong heterologous protein expression in the C2C12 myotubes after nuclear plasmid injection. In addition, we performed tests with a construct of well-defined function to verify that the degree of fluorescence expression that we achieved leads to sufficient functional expression. For this purpose, we injected pGFP-α1C that encodes a fluorescent fusion protein of the α1-subunit of the cardiac L-type Ca2+ channel (CaV1.2). These channels exhibit rapid activation kinetics so that their current can be easily distinguished from the very slowly activating current generated by the intrinsic CaV1.1 channels (see for example Schuhmeier et al. 2005). Figure 2A shows a C2C12 myotube that had been injected with the pGFP-α1C plasmid 2 days before imaging. The expression of fluorescent GFP-α1C occurred later in time (2 days) than GFP alone (~6 h) due to its much larger size (217 versus 27 kDa) and showed a particulate pattern and a final localization predominantly near the periphery of the cell.

Figure 2
Altered Ca2+ inward current in C2C12 myotubes injected with expression plasmids encoding Cav1.2

The expression of the green fluorescence from GFP-α1C was accompanied by a significant change in the inward current that could be recorded from the myotubes (Fig. 2B and C). In the control cells (Fig. 2B), the endogenous Ca2+ current showed its typical slow activation and no inactivation during the pulse. The currents of the pGFP-α1C-expressing cells (Fig. 2C) were larger, became considerably more rapidly activated and showed some degree of inactivation during the pulse.

Figure 2D shows the mean voltage dependence of current density at the peak of activation in non-injected control cells (○) and in pGFP-α1C-expressing cells (•). Figure 2E shows the voltage dependence of the activation kinetics as determined by a single exponential fit between 5 and 50 ms during the pulse for GFP-α1C-expressing cells (•) and between 20 and 100 ms for control cells (○). On average, at +30 mV (i.e. close to the voltage that produces the maximum inward current) the amplitude was 2.6-fold larger in pGFP-α1C-expressing myotubes and the time constant of activation was 12-fold smaller (data see legend). Thus, Fig. 2 confirms that our procedure allows substantial functional over-expression of even large proteins within the available time frame.

L-type Ca2+ current in JP-45 over-expressing cells

Whole-cell patch-clamp experiments were subsequently carried out on myotubes injected with plasmids encoding a fluorescent JP-45 fusion protein. In parallel to the Ca2+ currents, intracellular fura-2 Ca2+ transients were recorded as described by Schuhmeier & Melzer (2004). We chose DsRed2 as the marker protein (plasmid pJP-45–DsRed2) which exhibits red-shifted fluorescence spectra compared to fura-2. Myotubes injected with pDsRed2-N3, encoding DsRed2 alone, served as controls.

Figure 3A and B shows examples of Ca2+ inward current recordings from two myotubes expressing DsRed2 and JP-45–DsRed2, respectively. Figure 3C compares the mean L-type Ca2+ current densities measured in the two groups of cells (controls, ○; JP-45–DsRed2, •). At none of the potentials could a significant difference be observed. Likewise, none of the parameters determined by fitting eqns (1) and (2) to the data in Fig. 3C (continuous lines) showed a significant difference. The best fit parameters are listed in the figure legend to Fig. 3.

Figure 3
Calcium inward current in JP-45–DsRed2-expressing C2C12 myotubes

Figure 3D shows the voltage dependence of the activation time constant obtained from single exponential fits to the activation phase of the slow inward current for both DsRed2- and JP-45–DsRed2-expressing cells (○ and •, respectively). As with current amplitude, no significant difference at any of the investigated voltages could be found.

JP-45 effects on Ca2+ release

Each of the cells was analysed using a removal model fit procedure (originally described by Melzer et al. 1986). Briefly, this method quantifies Ca2+ binding and reuptake parameters of the cell by fitting model-generated traces to the measured fluorescence ratio traces during the time after repolarization (for more details see Schuhmeier & Melzer, 2004). Table 1 lists the mean values of the free kinetic parameters that were adjusted in the removal model to describe the relaxation phases of Ca2+ transients for four consecutive voltage pulses of different duration and amplitude. The table shows that the mean values estimated here were similar to data obtained in non-injected myotubes by Schuhmeier & Melzer (2004). Notably, the kinetic data obtained for the indicator dye and for the slow saturable model component S that was included to simulate EGTA were not significantly different. A statistically significant difference from the previous analysis was observed in the rate constant describing slow uptake (kuptake) which might be due to a different pump rate in the present population of cells. Because there were no significant differences in the removal properties of test (JP-45–DsRed2) and control cells (DsRed2), we pooled the removal parameters and used the following set of mean values for further analysis: kon,Dye, 175.3 μm−1 s−1; koff,Dye, 48.4 s−1; kon,S, 25.9 μm−1 s−1; koff,S, 3.1 s−1; kuptake, 1.5 × 103 s−1.

Table 1
Best fit results of free parameters in the removal model analysis

Figure 4Ab shows a representative fura-2 fluorescence ratio signal obtained during a single application of a 100-ms depolarizing pulse to +30 mV (Fig. 4Aa). The signal is drawn inverted in sign so that a rise in Ca2+ is upward. The amplitude of the transient was determined by calculating the difference between the average of 20 baseline points and the average of the last six points of the pulse. Figure 4B shows the mean amplitudes at different voltages for both DsRed2- (○) and JP-45–DsRed2-expressing (•) myotubes. In the latter set of cells the amplitude was significantly smaller compared to the controls.

Figure 4
Calcium release in JP-45–DsRed2-expressing C2C12 myotubes

Figure 4Ac shows the conversion of the fluorescence signal to calcium input flux according to Schuhmeier & Melzer (2004) by using the results of the removal analysis. The arrows in Fig. 4Ac indicate the peak of the flux and the plateau, which is reached after the rapid inactivation. Both are plotted as functions of voltage in Fig. 4C and D. As in Fig. 4B, the mean values at +10 mV and above differed significantly in the two groups of cells (except for voltages of +20 and +60 mV).

The sigmoidal voltage dependence in Fig. 4C (peaks) was fitted for each individual experiment using a standard Boltzmann equation described by maximum amplitude (Amax), V0.5 and k. In Fig. 4D (plateau) an additional term (as described by Schuhmeier & Melzer, 2004) was added to correct for the gradual decrease in amplitude at large depolarizations. The continuous activation curves in Fig. 4C and D were calculated using the mean values of the individual fit parameters. Comparing the results for JP-45–DsRed2-expressing myotubes with the controls, there was a significant decrease by 42% for the peak and 44% for the plateau in the parameter describing maximal activation at large depolarizations. All other parameters were not significantly different (see the legend to Fig. 4).

A typical feature of the calculated Ca2+ input flux traces was a slowly declining plateau (Fig. 4Ac and Fig. 5A). In mature muscle, the slope of the plateau has been attributed to progressive SR Ca2+ depletion at constant Ca2+ release permeability (Schneider et al. 1987; Gonzalez & Rios, 1993; Ursu et al. 2005). If the initial SR content is known, one can calculate the changes in SR Ca2+ content from the estimated Ca2+ release flux and correct the flux for SR Ca2+ depletion. This leads to an estimate of the voltage-activated SR Ca2+ permeability. Under the given assumption, the initial SR Ca2+ content can be determined as the value that eliminates the slope in the plateau region when performing the correction. The procedure is demonstrated in Fig. 5 for a series of flux traces obtained at different voltages in a single myotube. At voltages lower than +10 mV, kinetics were slower and the separation between the fast and slow phases of decay (putative inactivation and depletion, respectively) became less obvious in most cells. Therefore, we used the value for SR Ca2+ content determined at +10 mV also for depletion correction at voltages below +10 mV. For correction of the flux traces at +10 mV and above, we used the individually determined initial values for SR Ca2+ content as described for experiments on adult mouse fibres by Ursu et al. (2005).

Figure 5
Estimation of Ca2+ permeability from Ca2+ input flux

Figure 6A and B show the mean results of flux and permeability calculation at +30 mV for all JP-45–DsRed2- and DsRed2- (larger amplitude trace) expressing cells, respectively. The shaded area represents the point-by-point calculated s.e.m. While the fluxes (Fig. 6A) showed different time courses due to the faster relative decline during the plateau of the traces with larger amplitudes, permeabilities (Fig. 6B) exhibited a very similar time course.

Figure 6
Altered Ca2+ permeability in JP-45–DsRed2-expressing C2C12 myotubes

We performed this ‘depletion analysis’ with the experiments of Fig. 4. Because this procedure is sensitive to any noise disturbing the determination of the slope in the plateau, we had to exclude one recording at +20 mV for DsRed2-expressing cells.

The mean values of the estimated SR Ca2+ contents are shown in Fig. 6C. The values of mean initial SR Ca2+ content at different voltage varied between 1.21 and 1.81 mm and showed no significant difference between JP-45-expressing cells and the controls. Figure 6D and E shows the voltage dependence of the corrected peak and plateau permeabilities, respectively. Unlike the SR Ca2+ content, the permeabilities show significant differences: in Fig. 6D, values at −10, 0, +10 and +30 mV are significantly different and in Fig. 6E, all values above −20 mV (except for the +20 mV value) are significantly different. Figure 6F shows the ratio of peak/plateau, a frequently used measure to characterize the overall shape of the traces. It is essentially constant over the voltage range of activation and not significantly different for JP-45–DsRed2-expressing cells and controls.

To check whether the reduction in voltage-activated Ca2+ permeability resulted from a loss of functional voltage sensors, we analysed the non-linear capacitive current at the onset of the depolarization as an estimate of the DHPR gating charge movements. Figure 7A and B shows the mean values of charge density at different voltages for myotubes expressing DsRed2 and JP-45–DsRed2, respectively. Consistent with the DHPR-mediated L-type Ca2+ current data, no significant difference was found between the two groups of experiments at any voltage. The data were fitted with conventional Boltzmann functions resulting in maximal charge density (qmax) values of 10.0 ± 1.3 × 10−3 and 13.0 ± 2.4 × 10−3 C F−1 for DsRed2- and JP-45–DsRed2-expressing cells, respectively. The V0.5 values were 5.1 ± 1.9 and 6.4 ± 2.1 mV, and the k values 13.1 ± 1.8 and 9.9 ± 1.2 mV (n = 12 and n = 11), respectively (comp. eqn (2)). A previous study (Zheng et al. 2002a) obtained mean values of qmax, V0.5 and k of 6.52 C F−1, 18.7 mV and 14.7 mV, respectively, for control C2C12 myotubes. Note that qmax values given in this reference were erroneously scaled up 10-fold (Zheng et al. 2002b and O. Delbono, personal communication).

Figure 7
Intramembrane charge movements and EC coupling transfer functions

Ca2+ release flux or Ca2+ permeability as a function of the corresponding voltage-sensor charge movement is commonly referred to as the transfer function of EC coupling (e.g. Csernoch et al. 1999). Figure 7C and D show transfer functions using the permeability data of Fig. 6 and the corresponding gating charge densities. Consistent with the lower permeabilities, the transfer functions of the JP-45–DsRed2-expressing cells (Fig. 7C and D, •) exhibited considerably lower steepness for both peak (Fig. 7C) and plateau permeability (Fig. 7D). The average slopes were 84.3 ± 27.0 versus 180.9 ± 40.5% ms−1 C−1 F (n = 6 versus n = 11, P = 0.063) and 39.1 ± 12.5 versus 97.0 ± 21.2% ms−1 C−1 F (P = 0.033), respectively.

These results indicate that the differences in the amplitudes of Ca2+ transients and Ca2+ flux observed here (Fig. 4) resulted from differences in the activated permeation pathway and not from differences in either the Ca2+ gradient that drives the flux or in voltage-sensor properties.


In the present study, we searched for possible functional effects of the recently discovered JFM protein JP-45 in myotubes. We focused on the two voltage-controlled fluxes of Ca2+; that is, Ca2+ inward current and Ca2+ release. Myotubes derived from the C2C12 cell line were used for over-expression of a fusion protein consisting of the fluorescent marker DsRed2 and JP-45. A change in Ca2+ release but not in Ca2+ current was observed.

Analysis of Ca2+ removal and Ca2+ release in C2C12 myotubes

To study Ca2+ release, we used an adapted version of the removal model fit approach first described by Melzer et al. (1986). The values of the binding rate constants identified with this analysis were remarkably similar to those of our previous investigation in C2C12 myotubes (Schuhmeier & Melzer, 2004) despite the lower concentration of EGTA used. DsRed2- and JP-45–DsRed2-expressing cells did not differ significantly in any of the estimated rate constants, indicating that removal properties were not affected by JP-45 and suggesting that the observed difference in Ca2+ signal amplitude results from a difference in the Ca2+ release activity rather than removal properties.

Ca2+ release flux is the product of SR permeability and SR Ca2+ content. The method that we used to calculate voltage-activated Ca2+ permeability (Schneider et al. 1987; Gonzalez & Rios, 1993) estimates the SR Ca2+ content prior to the pulse (Ca0,SR) and corrects the release flux for the putative changes in SR Ca2+ content determined from the flux and its slope during the slow phase. The values of Ca0,SR estimated in our experiments for DsRed2- and JP-45–DsRed2-expressing myotubes averaged over the voltage range +10 to +80 mV (1.60 ± 0.23 and 1.47 ± 0.27 mm, respectively) were not significantly different from each other indicating that loading of the SR was not changed.

In voltage-clamped rat and mouse muscle fibres, values of 1.9 and about 3 mm, respectively, have been determined for Ca0,SR using the same method (Shirokova et al. 1996; Ursu et al. 2005). It should be noted that Ca0,SR is expressed in concentration with respect to the myoplasmic water volume, as is the release flux. An estimate of the SR volume in relation to myoplasmic water volume would be required (but is not available for myotubes) to convert the numbers to true concentrations in the SR. On the other hand, permeabilities (in % ms−1) and fractional changes of SR Ca2+ content are independent of such morphometric data and therefore permit a comparison with data obtained from adult muscle fibres. For the peak permeability values in control myotubes, we obtained about 1.4% ms−1 at +50 mV in the present study compared to about 6% ms−1 at the same potential in mouse interosseus fibres (Ursu et al. 2005). We determined that a 100-ms voltage pulse that maximally activates Ca2+ release leads on average to about 50% reduction in the SR Ca2+ content of the control (i.e. DsRed2-expressing) C2C12 myotubes. This compares to about 80% in mature mouse fibres estimated by a very similar analysis procedure (Ursu et al. 2005).

Interpretation of JP-45 effects on Ca2+ signalling in C2C12 cells

The localization of the JFM suggests a contribution of JP-45 to the ‘Ca2+ signalling toolkit’ (Berridge et al. 2003) of skeletal muscle. Possible roles include participation in the formation of the triadic junction, in targeting one of its constituents to the junction or in modifying EC coupling directly. Results of Anderson (2003) indicated colocalization of JP-45 with ryanodine receptors but no direct interaction with the release channel. Instead, in vitro interaction with the α1S subunit of the DHPR has been reported (AA Anderson, X Altafaj, Z Zheng, Z-M Wang, O Delbono, M Ronjat, S Treves, F Zorzato, 2003), making JP-45 an interesting candidate for a possible functional modulator of EC coupling. Recent experimental evidence also suggests an interaction with the β subunit of the DHPR (Anderson et al. unpublished results). Modulation of the DHPR might affect its voltage-sensing properties, the size of the Ca2+ entry flux or the conformational transmission to the RyR. Experiments correlating maximum charge movement with the JP-45–DsRed2 fluorescence level in transfected cells indicated a decrease at high expression levels (AA Anderson, X Altafaj, Z Zheng, Z-M Wang, O Delbono, M Ronjat, S Treves, F Zorzato unpublished results). Our set-up was not equipped for quantifying Ca2+ signals and JP-45–DsRed2 fluorescence simultaneously. Because under the present conditions we found no evidence for altered charge movements, we suspect that the expression levels reached in our case are below the values needed to affect charge movements significantly. Nevertheless, our study showed a reduction in the Ca2+ signals and in the calculated flux of Ca2+ mobilization. The lower effectiveness of voltage-sensor charge movements in activating SR permeability, indicated by the reduced slope of the EC coupling transfer functions (Fig. 7C and D), may have resulted from an inhibiting effect on the transmission from the DHPR caused by the interaction of the N-terminal region of JP-45 with the DHPR (Anderson et al. 2003). Alternatively, it may have resulted from an interaction of the luminal C-terminal end of JP-45 with CSQ, as demonstrated in vitro (Anderson et al. 2003). This interaction might affect the Ca2+ storing properties in the lumen of the SR by altering the buffer capacity of CSQ. Yet, our analysis of Ca0,SR (see above) showed no evidence for a reduced Ca2+ concentration in the SR. In heart muscle, three proteins of the JFM (i.e. RyR, triadin and junctin) form a complex with CSQ that seems to influence the open probability of the RyR as a function of the Ca2+ concentration in the SR lumen (Zhang et al. 1997; Györke et al. 2004). In this complex, CSQ has been proposed to serve as a Ca2+ sensor that modulates the RyR via links formed by triadin and junctin. A similar complex appears to be present in skeletal muscle (Beard et al. 2005) and may serve equivalent regulatory functions that may be altered by JP-45. Even though this indirect mechanism cannot be ruled out, in our opinion it is less likely than a more direct effect of JP-45 on functional domains of the DHPR relevant for Ca2+ release control.

JP-45 is specific for skeletal muscle SR. Therefore based on our results, it seems likely that JP-45 down-regulates the multiprotein EC coupling complex of the terminal cisternae in a manner specific to skeletal muscle. Consequently, loss of function by mutations in JP-45 may lead to skeletal muscle-specific pathological over-activity of the Ca2+ release system. Mutations in RyR1 that lead to increased Ca2+ release are known to cause malignant hyperthermia susceptibility (MHS) or central core disease (CCD). Therefore it seems worthwhile to screen the sequence of human JP-45 for mutations in those families with MHS or CCD with no known RyR1 mutations (Jurkat-Rott et al. 2000; Lyfenko et al. 2004). Further important clues about the role of the protein can be expected from experiments similar to the ones shown here performed on myotubes or adult muscle fibres of genetically altered mice lacking JP-45.

In summary, we demonstrated, using plasmids encoding fluorescent fusion proteins, that efficient and rapid protein expression can be achieved in C2C12 myotubes by nuclear microinjection. Fusion constructs of JP-45, a protein of the skeletal muscle SR junctional face, expressed strong fluorescence in a cellular membrane compartment distinct from the ER portion labelled by DsRed2-ER. Using fura-2 fluorimetry under whole-cell voltage-clamp conditions, we found that JP-45–DsRed2-expressing C2C12 myotubes exhibited a decrease in depolarization-induced Ca2+ transients while Ca2+ inward current density was unaltered. Using a procedure to determine Ca2+ flux and permeability from the Ca2+ transient, the change could be tentatively assigned to an effect on the Ca2+ permeability of the SR rather than its Ca2+ content. Because in vitro studies had demonstrated binding of JP-45 to both the DHPR and calsequestrin, the observed effect on Ca2+ release may be attributed to a modulatory effect of either one of these interactions.


We thank E. Schoch for designing and constructing equipment and A. Riecker and K. Nothelfer for expert technical help. We are grateful to M. Grabner and B. E. Flucher (Innsbruck) and to H. Lerche and S. Maljevic (Ulm) for providing mammalian expression plasmids. We acknowledge a contribution of the European Commission to W.M. for graduate training (HPRN-CT-2002-00331). E.G. received partial funding from Graduiertenkolleg 460 and Landesschwerpunkt Nanoskopie. The work was funded by a grant of the Deutsche Forschungsgemeinschaft (DFG) to W.M. (ME-713/10-3).


  • Anderson AA, Treves S, Biral D, Betto R, Sandona D, Ronjat M, Zorzato F. The novel skeletal muscle sarcoplasmic reticulum JP-45 protein. Molecular cloning, tissue distribution, developmental expression, and interaction with alpha 1.1 subunit of the voltage-gated calcium channel. J Biol Chem. 2003;278:39987–39992. [PubMed]
  • Arikkath J, Campbell KP. Auxiliary subunits: essential components of the voltage-gated calcium channel complex. Curr Opin Neurobiol. 2003;13:298–307. [PubMed]
  • Barton K, MacLennan D. The proteins of sarcotubular system. In: Engel AG, Franzini-Armstrong C, editors. Myology: Basic and Clinical. 3. New York: McGrow Hill; 2004. pp. 307–323.
  • Beam KG, Franzini-Armstrong C. Functional and strutural approaches to the study of EC coupling. In: Emerson C, Sweeney H, editors. Methods in Muscle Biology. San Diego: Academic Press; 1997. pp. 284–306.
  • Beard NA, Casarotto MG, Wei L, Varsanyi M, Laver DR, Dulhunty AF. Regulation of ryanodine receptors by calsequestrin: effect of high luminal Ca2+ and phosphorylation. Biophys J. 2005;88:3444–3454. [PMC free article] [PubMed]
  • Berridge MJ, Bootman MD, Roderick HL. Calcium signalling: dynamics, homeostasis and remodelling. Nat Rev Mol Cell Biol. 2003;4:517–529. [PubMed]
  • Bers DM, editor. Excitation-Contraction Coupling and Cardiac Contractile Force. 2. Dordrecht: Kluwer Academic Publishers; 2001.
  • Blau HM, Webster C, Pavlath GK. Defective myoblasts identified in Duchenne muscular dystrophy. Proc Natl Acad Sci U S A. 1983;80:4856–4860. [PMC free article] [PubMed]
  • Brum G, Stefani E, Rios E. Simultaneous measurements of Ca2+ currents and intracellular Ca2+ concentrations in single skeletal muscle fibers of the frog. Can J Physiol Pharmacol. 1987;65:681–685. [PubMed]
  • Caswell AH, Brandt NR, Brunschwig JP, Purkerson S. Localization and partial characterization of the oligomeric disulfide-linked molecular weight 95,000 protein (triadin) which binds the ryanodine and dihydropyridine receptors in skeletal muscle triadic vesicles. Biochemistry. 1991;30:7507–7513. [PubMed]
  • Csernoch L, Szentesi P, Kovacs L. Differential effects of caffeine and perchlorate on excitation-contraction coupling in mammalian skeletal muscle. J Physiol. 1999;520:217–230. [PMC free article] [PubMed]
  • Föhr KJ, Warchol W, Gratzl M. Calculation and control of free divalent cations in solutions used for membrane fusion studies. Methods Enzymol. 1993;221:149–157. [PubMed]
  • Friedrich O, Ehmer T, Fink RH. Calcium currents during contraction and shortening in enzymatically isolated murine skeletal muscle fibres. J Physiol. 1999;517:757–770. [PMC free article] [PubMed]
  • Gonzalez A, Rios E. Perchlorate enhances transmission in skeletal muscle excitation-contraction coupling. J Gen Physiol. 1993;102:373–421. [PMC free article] [PubMed]
  • Grabner M, Dirksen RT, Beam KG. Tagging with green fluorescent protein reveals a distinct subcellular distribution of L-type and non-L-type Ca2+ channels expressed in dysgenic myotubes. Proc Natl Acad Sci U S A. 1998;95:1903–1908. [PMC free article] [PubMed]
  • Guo W, Campbell KP. Association of triadin with the ryanodine receptor and calsequestrin in the lumen of the sarcoplasmic reticulum. J Biol Chem. 1995;270:9027–9030. [PubMed]
  • Györke I, Hester N, Jones LR, Györke S. The role of calsequestrin, triadin, and junctin in conferring cardiac ryanodine receptor responsiveness to luminal calcium. Biophys J. 2004;86:2121–2128. [PMC free article] [PubMed]
  • Jones LR, Zhang L, Sanborn K, Jorgensen AO, Kelley J. Purification, primary structure, and immunological characterization of the 26-kDa calsequestrin binding protein (junctin) from cardiac junctional sarcoplasmic reticulum. J Biol Chem. 1995;270:30787–30796. [PubMed]
  • Jurkat-Rott K, McCarthy T, Lehmann-Horn F. Genetics and pathogenesis of malignant hyperthermia. Muscle Nerve. 2000;23:4–17. [PubMed]
  • Klein MG, Simon BJ, Szücs G, Schneider MF. Simultaneous recording of calcium transients in skeletal muscle using high- and low-affinity calcium indicators. Biophys J. 1988;53:971–988. [PMC free article] [PubMed]
  • Knudson CM, Stang KK, Jorgensen AO, Campbell KP. Biochemical characterization of ultrastructural localization of a major junctional sarcoplasmic reticulum glycoprotein (triadin) J Biol Chem. 1993;268:12637–12645. [PubMed]
  • Lyfenko AD, Goonasekera SA, Dirksen RT. Dynamic alterations in myoplasmic Ca2+ in malignant hyperthermia and central core disease. Biochem Biophys Res Commun. 2004;322:1256–1266. [PubMed]
  • MacKrill JJ. Protein–protein interactions in intracellular Ca2+-release channel function. Biochem J. 1999;337:345–361. [PMC free article] [PubMed]
  • Melzer W, Herrmann-Frank A, Lüttgau HC. The role of Ca2+ ions in excitation-contraction coupling of skeletal muscle fibres. Biochim Biophys Acta. 1995;1241:59–116. [PubMed]
  • Melzer W, Rios E, Schneider MF. The removal of myoplasmic free calcium following calcium release in frog skeletal muscle. J Physiol. 1986;372:261–292. [PMC free article] [PubMed]
  • Pelham HR. Getting through the Golgi complex. Trends Cell Biol. 1998;8:45–49. [PubMed]
  • Sanger JW, Sanger JM, Franzini-Armstrong C. Assembly of the skeletal muscle. In: Engel AG, Franzini-Armstrong C, editors. Myology: Basic and Clinical. 3. New York: McGrow Hill; 2004. pp. 45–65.
  • Schneider MF, Simon BJ, Szücs G. Depletion of calcium from the sarcoplasmic reticulum during calcium release in frog skeletal muscle. J Physiol. 1987;392:167–192. [PMC free article] [PubMed]
  • Schuhmeier RP, Dietze B, Ursu D, Lehmann-Horn F, Melzer W. Voltage-activated calcium signals in myotubes loaded with high concentrations of EGTA. Biophys J. 2003;84:1065–1078. [PMC free article] [PubMed]
  • Schuhmeier RP, Gouadon E, Ursu D, Kasielke N, Flucher BE, Grabner M, Melzer W. Functional interaction of CaV channel isoforms with ryanodine receptors studied in dysgenic myotubes. Biophys J. 2005;88:1765–1777. [PMC free article] [PubMed]
  • Schuhmeier RP, Melzer W. Voltage-dependent Ca2+ fluxes in skeletal myotubes determined using a removal model analysis. J Gen Physiol. 2004;123:33–51. [PMC free article] [PubMed]
  • Shirokova N, Garcia J, Pizarro G, Rios E. Ca2+ release from the sarcoplasmic reticulum compared in amphibian and mammalian skeletal muscle. J Gen Physiol. 1996;107:1–18. [PMC free article] [PubMed]
  • Szentesi P, Collet C, Sarkozi S, Szegedi C, Jona I, Jacquemond V, Kovacs L, Csernoch L. Effects of dantrolene on steps of excitation-contraction coupling in mammalian skeletal muscle fibers. J Gen Physiol. 2001;118:355–375. [PMC free article] [PubMed]
  • Ursu D, Schuhmeier RP, Melzer W. Voltage-controlled Ca2+ release and entry flux in isolated adult muscle fibres of the mouse. J Physiol. 2005;562:347–365. [PMC free article] [PubMed]
  • Villa A, Podini P, Nori A, Panzeri MC, Martini A, Meldolesi J, Volpe P. The endoplasmic reticulum-sarcoplasmic reticulum connection. II. Postnatal differentiation of the sarcoplasmic reticulum in skeletal muscle fibers. Exp Cell Res. 1993;209:140–148. [PubMed]
  • Walker D, De Waard M. Subunit interaction sites in voltage-dependent Ca2+ channels: role in channel function. Trends Neurosci. 1998;21:148–154. [PubMed]
  • Zhang L, Kelley J, Schmeisser G, Kobayashi YM, Jones LR. Complex formation between junctin, triadin, calsequestrin, and the ryanodine receptor. Proteins of the cardiac junctional sarcoplasmic reticulum membrane. J Biol Chem. 1997;272:23389–23397. [PubMed]
  • Zheng Z, Wang ZM, Delbono O. Charge movement and transcription regulation of L-type calcium channel alpha (1S) in skeletal muscle cells. J Physiol. 2002a;540:397–409. [PMC free article] [PubMed]
  • Zheng Z, Wang ZM, Delbono O. Erratum for Zheng et al. J Physiol. 2002b;540:397–409. J Physiol541 1059. [PMC free article] [PubMed]
  • Zorzato F, Anderson AA, Ohlendieck K, Froemming G, Guerrini R, Treves S. Identification of a novel 45 kDa protein (JP-45) from rabbit sarcoplasmic-reticulum junctional-face membrane. Biochem J. 2000;351:537–543. [PMC free article] [PubMed]

Articles from The Journal of Physiology are provided here courtesy of The Physiological Society
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...