The junctional SR protein JP-45 affects the functional expression of the voltage-dependent Ca²⁺ channel Ca_v1.1

The sarcoplasmic reticulum (SR) of the skeletal muscle and the transverse tubular membranes provide a conducive environment enriched with key proteins that have been shown to play an essential role in excitation-contraction (E-C) coupling (Endo, 1977; Fleischer and Inui, 1989). E-C coupling occurs in specialized junctions called triads, which are an intracellular synapse formed by the membrane system of the transverse tubule and the terminal cisternae of the SR. The two main Ca²⁺ channels involved in E-C coupling are the dihydropyridine receptor, Ca_v1, which lies on the transverse tubules where it acts as a voltage sensor for E-C coupling, and the ryanodine receptor (RyR), present on the junctional face membrane of the SR, which is the Ca²⁺ release channel (Franzini-Armstrong and Jorgensen, 1994; Meissner, 1994; Ma and Pan, 2003; Sutko and Airey, 1996; Franzini-Armstrong, 1980; Lamb and Stephenson, 1990; Rios and Pizzarro, 1991). Ca_v1 is a hetero-oligomeric complex made up of at least four subunits: α1, β, α₂δ and γ (Leung et al., 1987; Lacerda et al., 1991; Birnbaumer et al., 1998; Catterall, 1995; Snutch and Reiner, 1992). The pore-forming Ca_v1.1 is an integral membrane protein and is indispensable for E-C coupling, whereas the cytosolic β1a and α₂δ subunits have regulatory functions; the role of the γ subunit has yet to be clearly defined (Catterall, 1995; Snutch and Reiner, 1992, Tsien et al., 1991). The β1a subunit associates tightly with the Ca_v1.1 by binding to a region on the I-II loop termed α-interacting domain (AID) (Pragnell et al., 1994). The β1a subunit is important for plasma membrane expression of the Ca_v1.1, whereas a sequence located in the C-terminal domain of the Ca_v1.1 seems to be involved in triad targeting (Chien et al., 1995; Flucher et al., 2000; Flucher et al., 2002), although other domains and/or polypeptides might be involved in proper targeting.

In skeletal muscle, Ca_v1.1 responds to transverse tubule depolarisation by sensing the voltage change and it induces Ca²⁺ release from the SR through a direct interaction with the RyR. Besides the RyR, however, the junctional face membrane contains numerous other proteins that, because of their anatomical location, are deemed to be involved in E-C coupling (Zhang et al., 1997; Costello et al., 1986; Zorzato et al., 2000). In the past few years a number of investigators have begun to define the major and minor structural components of the junctional face membrane (Ito et al., 2001; Takeshima et al., 2000). In previous studies (Zorzato et al., 2000; Anderson et al., 2003), we identified and characterized - at the biochemical and molecular level - JP-45, an integral membrane protein constituent of the SR junctional face membrane in skeletal muscle. We also showed that JP-45 colocalizes with the RyR Ca²⁺-release channel and interacts with Ca_v1.1 and the luminal Ca²⁺-binding protein calsequestrin (Anderson et al., 2003). To gather insight into the functional role of JP-45, we defined the domains involved in the interaction between JP-45 and Ca_v1.1. Our results demonstrate that the cytoplasmic domain of JP-45 interacts directly with the I-II loop, the C-terminal domain of Ca_v1.1 subunit and also with the β1a subunit of Ca_v1. In addition, we show that the interaction between JP-45 and the I-II loop occurs through the AID domain, and can be displaced by β1a. Experimental evidence has demonstrated that the β1a subunit interacts with the Ca_v1.1 subunit via AID (Pragnell et al., 1994; Chen et al., 2004) and that this protein-protein interaction is involved in the insertion of the voltage sensor to its proper membrane compartment (Flucher et al., 2002). To confirm the functional role of JP-45 in vivo, we overexpressed and silenced endogenous JP-45 in C2C12 cells and studied their electrophysiological properties. Based on the results of the present report, JP-45 is involved in the regulation of the functional expression of Ca_v1.1 into the transverse tubular membrane compartment.

In a previous report, we demonstrated that JP-45 interacts with native Ca_v1.1 (Anderson et al., 2003). Here, we describe a series of experiments to identify the domains participating in this interaction. We prepared recombinant GST-fusion proteins covering different coding regions of JP-45 (excluding the transmembrane domain) (see Fig. 1). These proteins were then bound to glutathione-Sepharose beads and incubated with solubilized light microsomal vesicles. Fig. 2A shows that the cytoplasmic N-domain of JP-45, more specifically the region between residues 1-80 (domain 2) interacts with native Ca_v1.1. No interaction was observed between any other JP-45 domains and Ca_v1.1. The physiological relevance of the interaction of fusion proteins was verified by studying the effect of the GST-JP-45 domain 2 on co-immunoprecipitation of Ca_v1.1 with a monoclonal anti-JP-45 antibody (Ab). The interaction between native JP-45 and Ca_v1.1 was competed-out by the presence of soluble GST-JP-45 domain 2 in the co-immunoprecipitation reaction (Fig. 2B).

Extensive structure-to-function relationship studies on Ca_v1 have helped to define functionally relevant domain boundaries not only among the subunits making up the supramolecular complex, but also within the primary structure of each subunit (Nakai et al., 1998; Grabner et al., 1999; Kugler et al., 2004; Van Petegem et al., 2004). We reasoned that the physiological relevance of the interaction between Ca_v1.1 and JP-45 could be better understood if we first identified JP-45-binding sites within the Ca_v1.1. GST-JP-45 domain-2 fusion protein was immobilized on GST-Sepharose and incubated with different His-tagged fusion proteins, covering the cytoplasmic domains of the Ca_v1.1 subunit (Tanabe et al., 1987). Whereas some of the His-tagged fusion proteins display the expected molecular mass, they had in other cases a molecular mass lower than the expected value due to the instability of purified fusion proteins (Fig. 3A, lanes 3 and 7). The protein-protein-interaction experiments depicted in Fig. 3B demonstrate that the I-II loop and the C-terminal domain of the Ca_v1.1 as well as the β1a subunit interact with JP-45, whereas fusion proteins corresponding to the N-terminus - the II-III and III-IV loops of the Ca_v1.1 subunit - do not interact with GST-JP-45 domain-2 fusion protein. To validate these results, we performed pull-down experiments using GST- JP-45 domain 2 as bait and solubilized light microsomal vesicles as ligand, in the presence or absence of increasing concentrations of either I-II loop or recombinant C-terminal-distal His-tag fusion proteins. Inhibition of Ca_v1.1-GST-JP-45 interaction was achieved at 3.6 and 1.25 μM of I-II loop and C-terminal-distal His-tag fusion proteins, respectively. To further confirm the specificity of the pull-down assay, we also investigated these interactions under native conditions, i.e. by performing co-immunoprecipitation of native JP-45 with Ca_v1.1 using the monoclonal anti-JP-45 Ab, in the presence or absence of competing I-II loop or recombinant C-terminal-distal His-tag fusion proteins. As shown in Fig. 3D, the presence of the competing recombinant fusion proteins substantially diminished the interaction between JP-45 and the Ca_v1.1 domains. Pull-down and co-immunoprecipitation assays demonstrated that competition with one of the Ca_v1.1 interacting domains (I-II loop or C-terminal-distal His-tag fusion proteins) was sufficient to partially inhibit the interaction. At a first glance it seems unclear how preventing the Ca_v1.1-JP-45 interaction through a single domain reduces total binding to the Ca_v1.1 to almost zero. A plausible explanation of the experiments is described in Fig. 3C and D; addition of excess I-II loop or C-terminal-distal loop occupies the interacting domain of the N-terminal domain of JP-45. Binding of the 50 aa-long JP-45 binding sites by an excess of the 14 kDa or 30 kDa Ca_v1.1 loops (I-II and C-terminal, respectively) most probably creates a steric hindrance, whereby the interaction of alternative domains within the native solubilized Ca_v1.1 is either abolished or substantially weakened.

Immunoprecipitation experiments revealed that JP-45 interacts with the I-II loop of subunit Ca_v1.1 and with purified recombinant β1a subunit (Fig. 3B). The β1a has been shown to interact strongly with the I-II loop of the α1.1 subunit and to affect its functional expression (Flucher et al., 2000; Chien et al., 1995). In the next set of experiments, we therefore characterized in greater detail the latter protein-protein interaction.

We investigated the role of the β1a subunit on the interaction between JP-45 and Ca_v1.1 by pull-down and co-immunoprecipitation assays (Fig. 4). JP-45 domain-2 fusion protein was immobilized on GST-Sepharose and incubated with the His-tagged Ca_v1.1 subunit I-II loop fusion protein in the presence or absence of increasing concentrations of β 1a fusion protein. As seen in Fig. 4A the presence of β1a interferes with the interaction between the I-II loop and JP-45. Pull-down assays were also performed by using the solubilized light microsomal vesicles as ligand. In the latter case, the association between the GST-JP-45 domain-2 fusion protein and the native Ca_v1.1 subunit was remarkably weaker in the presence of excess β1a-His-tagged fusion protein (Fig. 4B). The interference of the interaction between JP-45 and the Ca_v1.1 subunit by excess β1a, was also confirmed by performing co-immunoprecipitation experiments with anti-JP-45 Abs to pull-down the native solubilized complex (Fig. 4C). These results suggest that the presence of excess β1a disrupts the complex between Ca_v1.1 and JP-45. We cannot discriminate whether JP-45 interacts with the Ca_v1.1 or β1a, subunit, or whether JP-45 interacts with both subunits, thereby forming an oligomeric complex.

The Ca_v1 β1a subunit has been shown to bind strongly to the intracellular loop between transmembrane domain I and II of the Ca_v1.1, a region composed of 18 aa displaying the motif QQ-E-L-GY-WI-E, conserved in different isoforms (Pragnell et al., 1994). This binding domain is also referred to as α-interacting domain (AID). It is thought that the interaction between AID and the β1a subunit is important to determine the stability of this Ca²⁺ channel on the plasma membrane (Birnbaumer et al., 1998; Flucher et al., 2000; Flucher et al., 2002; Chien et al., 1995; Ahern et al., 2003; Bichet et al., 2000). In addition, the interaction between AID and β1a is regarded to the major structural determinant required for the surface expression of Ca_v1.1 (Chien et al., 1995; Beurg et al., 1999; Gregg et al., 1996; Jones et al., 1998). Coexpression of Ca_v1.1 and β1a subunits has been shown to increase the density and kinetics of L-type Ca²⁺ currents, and experimental evidence suggest that the β1a subunit facilitates gating of Ca_v1.1 by increasing the coupling between charge movement and pore opening (Sheridan et al., 2003; Sheridan et al., 2004). These effects of the β1a subunit are due to its interaction with the I-II loop, a domain that also interacts with JP-45. In the next set of experiments we addressed the question of whether the β1a subunit and JP-45 share the AID domain as binding site on the I-II loop. To address this issue, we performed pull-down assays with (1) GST-I-II loop fusion protein, (2) with a GST-fusion protein containing AID flanked by 21 and 11 residues at the N-terminal and C-terminal end, respectively, and (3) with a synthetic biotinylated-AID peptide. For the first two assays, the GST-fusion proteins were used as bait and the His-tagged JP-45 domain-2 fusion protein as ligand. The apparent molecular mass of His-tagged JP-45 domain 2 is larger than the theoretical predicted values (approx. 31 kDa vs 14 kDa, respectively). We are confident that the apparent larger molecular mass of the His-tagged JP-45 domain-2 fusion protein is due its dimerization, because its treatment with DEPC (which destroys the His tag) abolished the binding of the anti-His tag Ab (Fig. 5 panel C). Western blot analysis (Fig. 5A,B) demonstrates that JP-45 domain 2 interacts directly with the AID sequence within the I-II loop. The in vitro interaction of the sequence containing AID and JP-45 occurs at micromolar concentrations of His-tagged JP-45 domain-2 fusion protein. Because of such a high concentration it is probably that, under physiological conditions, the interaction between the β1a subunit and AID prevails over that of JP-45 and AID. Nevertheless, on the basis of the highly organized molecular assembly of the triad membranes, we cannot exclude the possibility that the local concentration of JP-45 in the triadic gap is sufficient to interfere either with the high-affinity interaction between AID and the β1a subunit (Pragnell et al., 1994; Sheridan et al., 2003; Strube et al., 1996). If there are no other proteins involved in the interaction with AID, our data imply that, in the β1a subunit knock-out animal, AID is occupied mainly by JP-45 and that such an interaction might account, at least in part, for the severe phenotype of the mice (Gregg et al., 1996). Strube et al. have shown that mouse skeletal muscle cells with a null-mutation in the cchb1 gene (encoding the β1a subunit of the Ca_v1) show a significant decrease in maximum charge-movement and a shift in the half-activation potential towards more negative potentials (Strube et al., 1996).

To investigate the functional effect of JP-45, we altered the stoichiometry of the JP-45-Ca_v1.1 supramolecular complex either by JP-45 overexpression or by JP-45 gene silencing in differentiated C2C12 myotubes, and then examined Ca_v1.1 activity by measuring charge movement.

The observation that alteration of peak charge movement is tightly linked to the magnitude of JP-45 expression suggests that, the stoichiometry of the JP-45/Ca_v1.1 supramolecular complex is crucial for the proper function of the Ca_v1.1. If this is so, we reasoned that depletion of JP-45 in differentiated myotubes would similarly affect the charge movement of Ca_v1.1.

The mRNA encoding JP-45 was much less abundant in C2C12 myotubes transfected with the plasmid containing JP-45 siRNA than in cells transfected with the control pSHAG vector (Fig. 8A). The presence of residual mRNA for JP-45 could either be due to the presence of a small subpopulation of cells which had not been transfected with the JP-45 siRNA construct, or to the inability of the construct to fully eliminate the transcript of JP-45. The amount of β-actin transcript did not differ significantly between cells transfected with the two constructs (Fig. 8A lower panel). Western blot analysis was performed to confirm depletion of JP-45 in differentiated C2C12 myotubes. Expression of JP-45 in C2C12 cells transfected with JP-45 siRNA is lower than the detection limit of the anti-JP-45 Abs that were raised against the N-terminal-interacting domain of the protein (Fig. 8B). However, we did not see changes in the expression of the housekeeping gene β-actin. Having established a reduction in transcription and expression of JP-45, we next measured Ca_v1.1 charge-movement. C2C12 cells were co-transfected with pFP-N3-DsRed2 and either the JP-45 siRNA pSHAG vector or pSHAG vector. Based on the expression of the DsRed2 reporter, C2C12 cells were identified and charge-movement measurements were performed. Fig. 9A shows that depletion of JP-45 in C2C12 myotubes is associated with a decrease in Q_max, apparently without affecting V_Q1/2 or the steepness of the curves (Table 2). Fig. 9 also shows charge-movement traces recorded in C2C12 cells transfected with pSHAG (B) or JP-45 siRNA (C). The lower Q_max value in JP-45-depleted C2C12 cells might originate from a decrease in the amount of Ca_v1.1 in muscle cell membrane. To verify this, we quantified by western blot analysis the relative amount of the Ca_v1.1 in membranes of C2C12 cells that had been transfected with JP-45 siRNA or pSHAG. The Ca_v1.1 shows two distinct high-molecular mass bands of 175 kDa and 170 kDa (Fig.9 D,E) (Leung et al., 1987). Furthermore, in JP-45-depleted C2C12 myotubes, there is a significant reduction of the immunoreactive band that referes to the Ca_v1.1 (Fig. 9D,E). Expression of the Ca_v1.1, measured as the ratio of the immunopositive band between siRNA-JP-45-transfected and control pSHAG-vector-transfected C2C12 cells, was 0.63±0.10 (mean ± s.e.m., n=4). The fractional decrease of Ca_v1.1 expression is 0.37±0.09 (mean ± s.e.m., n=4) and matches the decrease in charge movement (fractional decrease is 0.34) shown in Table 2. These results indicate that JP-45 is important for proper insertion of the α1.1 subunit into the membrane of muscle cells.

A number of data have been obtained in the last few years concerning structural determinant(s) necessary for the proper functional insertion of Ca_v1.1 into the plasma membrane. Although not all aspects of the functional insertion into the membrane and of the targeting of the Ca_v1.1 are fully understood, the emerging idea is that cooperation of the AID domain in the I-II loop with the C-terminal domain, as well as the tertiary and quaternary structure assembly of the Ca_v1 complex, play a crucial role in the functional expression of the Ca_v1.1 in the cell membrane (Flucher et al., 2002). However, involvement of additional protein-protein interactions between other polypeptides of the triad membrane has been implied for the proper functional insertion of the Ca_v1.1. We suggest that JP-45 is one of these additional polypeptides important for the proper assembly of the Ca_v1.1 macromolecular complex into the plasma membrane.

Here, we have shown that JP-45 domain 2 interacts with the AID sequence and with the C-terminal domain of the Ca_v1.1, two important structural determinants for functional expression of the Ca_v1.1. In addition, we have provided clear evidence that the level of JP-45 expression in C2C12 myotubes affects the functional properties and expression of Ca_v1.1. By altering expression levels of JP-45 in differentiated C2C12 myotubes by overexpressing its cDNA or by JP-45 gene silencing, the stoichiometry of the JP-45-Ca_v1.1 supramolecular complex is dramatically altered. It is tempting to speculate that alteration of the stoichiometry of the complex influences the functional expression of the Ca_v1.1 by different mechanisms. Fig. 10 summarizes a speculative model, which describes the potential mode of action of JP-45 on Ca_v1.1 function and underlines the importance of the correct stoichiometry of JP-45-Ca_v1.1 on the functional expression of Ca_v1.1. Overexpression of JP-45 does not seem to induce major changes in the expression levels of α1.1, however it could enforce occupation of its binding site of the I-II loop. This occupation of the JP-45 binding site within the I-II loop might occur as consequence of a partial dissociation of the Ca_v1 β1a subunit or through interaction with the Ca_v1.1, which does not stably bind β1a subunit (Garcia et al., 2002; Jones et al., 2002). The I-II loop is adjacent to a repeat within the α1.1 subunit domain involved in channel gating. It is tempting to speculate that the effect of the overexpression and binding of JP-45 to the binding site within I-II loop is twofold: (1) Interference with the β1a action on gating currents (Strube et al., 1996, Sheridan et al., 2003). (2) Perturbation of the functional role of the adjacent repeat on channel function. Each outcome, or the combination of both, would confer a restricted conformation on the JP-45-Ca_v1.1 supramolecular complex, leading to a decrease in charge movement. However, as indicated by western blot analysis, depletion of JP-45 affects the total amount of Ca_v1.1 in the cell membrane and, in parallel, also decreases the maximal charge movement. The mechanism leading to the decrease in the amount of Ca_v1.1 in cell membranes of JP-45-depleted cells might have several explanations. First, our results show that JP-45 interacts with the C-terminal domain, a region that has been shown to encompass a sequence involved in membrane targeting. When this sequence does not interact with JP-45, proper membrane targeting of Ca_v1.1 might be impaired. Second, JP-45 is involved in stabilizing the Ca_v1 complex. The functional consequence of both events is a decrease in Q_max, with no changes in the voltage dependence of the charge movement.

JP-45 is expressed at a very early stage of skeletal-muscle development - its transcript appears in 15-day-old embryos (Anderson et al., 2003) - in a developmental phase in which ER-SR membrane transition is not completed. In immature skeletal muscle fibres JP-45 might be localized in the ER-SR membrane network; then, in adult muscle fibres, JP-45 is targeted to the junctional face membrane of the SR. JP-45 might be involved in the retention of the Ca_v1.1 into the ER-SR membrane during the assembly of the Ca_v1.1 complex at an early stage of skeletal-muscle development. The appearance of the β1a subunit interferes with the interaction of JP-45 with the I-II loop of the Ca_v1.1. Such an event, together with an interaction of JP-45 and the C-terminal targeting domain, allows the functional expression of the Ca_v1.1 to the transverse tubular membranes.

pGex plasmids, nitrocellulose, rainbow molecular weight markers, glutathione-Sepharose, and [³²P]dCTP were from Amersham Biosciences; goat anti-α1 subunit polyclonal Abs against the skeletal muscle Ca_v1.1 were from Santa Cruz Biotechnology Inc.; isopropyl-β-D-thiogalactoside (IPTG), chemiluminescence kit, restriction enzymes, fugene transfection reagent, peroxidase-conjugated anti-goat Abs and EDTA-free anti-protease cocktail were from Roche Applied Science; protein-G peroxidase, protein assay determination kit and SDS-PAGE protein standards were from Bio-Rad; Talon metal affinity resin was from BD Bioscience; tricine, anti-poly-histidine, anti-β-actin Abs and peroxidase-conjugated anti-mouse Abs were from Sigma. Jet PEI transfection reagent was from Polyplus-Transfection SAS (Illkirch, France). Protein G plus Agarose was from Santa Cruz Biotechnology, Santa Cruz, CA. The pFP-N3/DsRed2 vector was constructed by in-frame substitution of the sequence encoding EGFP with that of DsRed2 in the pEGFP plasmid (Clonetech). The sequence of the plasmid backbone used for subsequent cloning of JP-45 cDNA was confirmed by sequencing. All other chemicals were reagents of highest grade available.

PCR-amplified cDNA encoding overlapping sequences of mouse skeletal muscle JP-45 were cloned in-frame into the multiple cloning site of pGex5x-3. PCR amplification conditions and primer sequences were as previously described (Anderson et al., 2003), using the following sets of primers (forward and reverse, respectively): domain 1, encompassing residues 1-29 (5′-AGAATTCTATGACTACCAGAGGCCTGG-3′ and 5′-AGTCGACGGCTGGTCCCTCCAGAAAT-3′); domain 2, encompassing residues 1-80 (5′-AGAATTCTATGACTACCAGAGGCCTGG-3′ and 5′-AGTCGACTGTGCTCTCCTTGCCCGCTA-3′); domain 3, encompassing residues 81-125 (5′-AGAATTCTGGCAAAGCGGGAACAA-3′ and 5′-AGTCGACATCTCCCCAGGGCAGGTC-3′); the N-terminus, encompassing residues 1-125 (5′-AGAATTCTATGACTACCAGAGGCCTGG-3′ and 5′-AGTCGACATCTCCCCAGGGCAGGTC-3′); the C-terminus, encompassing residues 149-331 (5′-AGAATTCTCGGGACGCAGTGGCT-3′ and 5′-AGTCGACGTCACGCCCCTTCCCTCGCTT-3′). The EcoRI restriction site was added to facilitate subsequent subcloning. cDNA was amplified in a Perkin Elmer GeneAmp 2400 PCR System under the following conditions: 5 minutes at 95°C, followed by 35 cycles of 40 seconds annealing at 61°C, 45 seconds extension at 72°C, 30 seconds denaturation at 92°C and a final elongation step of 4 minutes at 72°C.

The cDNAs encoding different domains of rabbit skeletal muscle Ca_v1.1 α1.1 subunit (Tanabe et al., 1987) were cloned into the pMR78 expression vector designed to express His-tagged proteins. The α1.1 subunit constructs used include the N-terminus (encompassing residues 1-51), the I-II loop (encompassing residues 335-432), the II-III loop (encompassing residues 654-797), the III-IV loop (encompassing residues 1059-1118), the proximal C-terminus (encompassing residues 1382-1585), the distal C-terminus (encompassing residues 1588-1878) and the full-length Ca_v1.1 β1a subunit. All constructs were checked by direct sequencing. Plasmids were used to transform E.coli DH5α cells and fusion-protein production was induced by the addition of 100 μM of isopropyl-β-D-thiogalactoside (IPTG). Fusion proteins were purified, according to the manufacturer's recommendations, with glutathione-Sepharose for GST-tagged fusion proteins and the Talon metal affinity resins for the His-tagged α1.1 subunit fusion proteins. The protein concentration of the purified proteins was determined with the Bio-Rad protein assay kit and bovine serum albumin as standard (Bradford, 1976). Proteins eluted from the affinity columns were analysed by SDS-PAGE or Tricine-SDS-PAGE (Schagger and von Jagow, 1987) and visualized by either Coomassie Brilliant Blue or stained with anti-poly-histidine Abs.

Light microsomal vesicles derived from rabbit skeletal muscle were prepared as described by Saito et al. (Saito et al., 1984) Membranes were solubilized at a final concentration of 1 mg/ml, for 30 minutes at room temperature in a buffer composed of 1% CHAPS, 200 mM NaCl, 1 mM dithiothreitol, 50 mM Tris-HCl pH 8.5 to which the protease inhibitor cocktail was added. Co-immunoprecipitation experiments of native proteins were performed as previously described with the monoclonal anti-JP-45 Ab (Anderson et al., 2003). To identify the domain(s) of JP-45 interacting with the Ca_v1.1, CHAPS-solubilized light microsomal vesicles were incubated for 60 minutes with glutathione-Sepharose beads to which GST-JP-45 fusion proteins had been bound. Following low-speed centrifugation, the beads were washed three times with PBS; bound proteins were eluted using glutathione elution buffer, separated on a 10% SDS-PAGE and transferred onto nitrocellulose. To identify JP-45-binding domains of Ca_v1.1, the JP-45 domain-2 fusion protein encompassing was immobilized on GST-Sepharose beads and incubated for 60 minutes with purified His-tagged fusion proteins covering various domains of the α1.1 subunit, including the N-terminal domain, the I-II, II-III, III-IV loops, the C-terminal-distal and C-terminal-proximal domains, and the β1a domain in 10 mM HEPES and 150 mM NaCl plus anti-proteases. Beads were processed as described above.

C2C12 cells were washed with PBS, treated with 1% digitonin buffer (1% digitonin, 185 mM KCl, 1.5 mM CaCl₂, 10 mM HEPES pH 7.4) on ice for 1 hour and centrifuged at 10,000 g for 10 minutes at 4°C. The lysate (500 μg total protein) was precleared by adding 0.5 μg of mouse IgG together with 20 μl of Protein G plus Agarose, and incubated for 30 minutes on a rotating device at 4°C. After centrifugation at 1000 g, the supernatant was transferred to a fresh tube on ice. Mouse Ca_v1.1 α1 subunit primary Ab (IIF7) (kindly provided by Kevin P. Campbell, University of Iowa, Howard Hughes Medical Institute, Iowa City, IA) (Leung et al., 1987) was added and incubated overnight at 4°C. Then 20 μl Protein G plus Agarose was added to each tube and incubated for 2 hours. After centrifugation, the pellets were washed with PBS and resuspended in 20 μl of double-strength sample buffer for 30 minutes at room temperature (Murray and Olendieck, 1997). The proteins were separated on a 10% SDS-PAGE gel and subsequently transferred onto PVDF membrane. Non-specific binding was blocked by incubating the membrane in 5% non-fat milk in PBS for 60 minutes at room temperature. Incubation in the primary Ab (1:1000 diluted in blot buffer) was for 2 hours at room temperature after which proteins were washed three times for 5 minutes with PBS. The membrane was incubated with anti-mouse IgG conjugated with horseradish peroxidase (HP) for 60 minutes, washed, and finally incubated in ECL Reagent and visualized in X-ray films. Autoradiograms were scanned and analyzed with KODAK-1D Image Analysis Software (Eastman Kodak Company, Rochester, NY).

Rabbit polyclonal antiserum was generated by immunizing a New Zealand white rabbit with glutathione-Sepharose purified GST-JP-45 fusion protein encompassing the cytoplasmic domain (aa 1-125) of JP-45. Serum was tested for the presence of Ab one month after immunization and subsequently the IgG fraction was purified by protein A column chromatography. Immunodetection of α1.1 subunit His-tagged fusion proteins was carried out with monoclonal anti-poly-histidine Ab, followed by HP-conjugated anti-mouse IgG; immunodetection of α1.1 subunit and JP-45 was carried out as described (Anderson et al., 2003). Immunopositive bands were visualized by chemiluminescence.

The mouse C2C12 muscle cell line was obtained from American Type Culture Collection (ATCC, Rockville, MD), cultured in standard conditions and maintained in growth medium (Dulbecco's modified Eagle's medium, DMEM, supplemented with 20% foetal bovine serum, 100 units/ml penicillin and 100 μg/ml streptomycin). Cells were induced to differentiate by switching the medium to differentiation medium (DMEM supplemented with 2% horse serum, 100 units/ml penicillin and 100 μg/ml streptomycin). For overexpression experiments, C2C12 cells were transfected with the cDNA encoding JP-45, cloned in-frame into a hybrid pFP-N3-DsRed2 vector. Briefly, 2 days after plating, cells were transfected by adding a solution of plasmid DNA (1 μg) and 3 μl FUGENE6 previously mixed for 30 minutes at room temperature, to the tissue culture medium. Electrophysiological measurements were made on differentiated cells 5-6 days after cell transfection. To deplete C2C12 of JP-45, we transfected cells with a JP-45 siRNA vector pSHAG. JP-45 RNA interference oligos were designed with siRNA Wizard software available at InvivoGene webpage. We chose the GCTCAACAAGTGCCTGGTACTGGCCTCGCTG nucleotides of JP-45 coding sequence. Complementary JP-45 RNAi oligos were synthesised according to the instructions of G. Hannon (Cold Spring Harbor), annealed and ligated downstream U6 RNA Pol III promoter of the pSHAG-1 vector as previously described (Treves et al., 2004). The nucleotide sequence of the JP-45 siRNA construct was verified by sequencing. For transfection experiments, C2C12 cells were plated on 12-mm diameter glass coverslips or on 100-mm diameter tissue culture dishes and once they reached 50-60% confluency they were transfected by a combination of CaPO4 and Jet PEI, using a total of 7.5 μg of plasmid DNA (coverslips) or 15 μg plasmid DNA (cell culture dishes). The day after transfection, the medium was changed and the cells were allowed to recover for 24 hours. On day 3, the cells were induced to differentiate by switching the medium to differentiation medium (DMEM supplemented with 2% horse serum, 100 units/ml penicillin and 100 μg/ml streptomycin) and were transfected again as described above to increase transfection efficiency. The next day fresh differentiation medium was added and cells were allowed to differentiate for another 3 days, after which multinucleated myotubes were clearly visible. Total RNA was extracted from transfected cells, converted into DNA as previously described (Treves et al., 2004) and reverse transcriptase (RT)-PCR was carried out with JP-45-specific or β-actin-specific primers. Western bloting was done with an Ab recognizing the N-terminal domain of JP-45.

For charge movement recordings, C2C12 cells were plated on glass coverslips and mounted in a small flow-through Lucite chamber positioned on a microscope stage. Myotubes were continuously perfused with the external solution (see below) using a push-pull syringe pump (WPI, Saratoga, FL.). Cells were voltage-clamped in the whole-cell configuration of the patch-clamp (Hamill et al., 1981; Wang et al., 1999) using an Axopatch-200B amplifier (Axon Instruments/Molecular Devices, Union City, CA). Micropipettes were pulled from borosilicate glasses (Boralex) using a Flaming Brown micropipette puller (P97, Sutter Instrument Co., Novato, CA) to obtain electrode resistance ranging from 2-4 MΩ. The composition of the internal solution (pipette) was (in mM): 140 Cs-aspartate; 5 Mg-aspartate₂, 10 Cs₂EGTA (ethylene glycol-bis(α-aminoethyl ether)-N,N,N′N′-tetraacetic acid), 10 HEPES, pH was adjusted to 7.4 with CsOH. The high concentration of Mg²⁺ in the pipette solution helped to stabilize the preparation for longer periods. The external solution contained (in mM): 145 TEA (tetraethylammonium)-Cl, 10 CaCl₂, 10 HEPES and 0.001 tetrodotoxin. Solution pH was adjusted to 7.4 with TEA.OH. For charge movement recording, Ca²⁺ current was blocked by the addition of 0.5 Cd²⁺ plus 0.3 La³⁺ to the external solution (Hamill et al., 1981; Wang et al., 1999; Wang et al., 2000; Beam and Franzini-Armstrong, 1997).

Whole-cell currents were acquired and filtered at 5 kHz with pClamp 6.04 software (Axon). A Digidata 1200 interface (Axon) was used for A-D conversion. Membrane current during a voltage pulse, P, was initially corrected by analogue subtraction of linear components. The remaining linear components were digitally subtracted on-line using hyperpolarizing control pulses of one-quarter test pulse amplitude (-P/4 procedure) (Delbono, 1992). The four control pulses were applied before the test pulse. We recorded the charge movement corresponding to gating of the L-type Ca²⁺ channel Ca_v1.1. To this end, we used a prepulse protocol consisting of a 2-second prepulse to -30 mV and a subsequent 5-millisecond repolarization to a pedestal potential of -50 mV, followed by a 12.5-millisecond depolarization from -50 mV to 50 mV with 10-mV intervals (Adams et al., 1990). The optimal duration of the prepulse defined as the value at which no further immobilization of charge movement is attained was determined to be 2 seconds, after testing a range of prepulses from 1-6 seconds (Wang et al., 2000). Intramembrane charge movements were calculated as the integral of the current in response to depolarizing pulses (charge on, Q_on) and were expressed per membrane capacitance (Coulombs per Farad). The complete blockade of the inward Ca²⁺ current was verified by the Q_on - Q_off linear relationship. Membrane capacitance was calculated as the integral of the transient current in response to a brief hyperpolarizing pulse from -80 mV (holding potential) to -100 mV.

To correlate the level of transfection with charge movement, we recorded fluorescence intensity arising from pFP-N3-DsRed2 in JP-45-transfected or pFP-N3-DsRed2-transfected cells with a Radiance 2100K1 laser scanning confocal system (Zeiss, Oberkochen, Germany). Fluorescence intensity was acquired in all the areas of the cell by using a krypton laser at 568 nm and recording the emission at 640 nm. The acquisition settings, including iris aperture (used at maximum), laser intensity, exposure, and gain were maintained unmodified from cell to cell to standardize recordings and make the comparison among cells valid. The maximum fluorescent intensity of the whole cell was analyzed in digitized images and expressed in arbitrary units (AU).

Data were analyzed using Student's t-test or analysis of variance (ANOVA). A value of P<0.05 was considered significant. Data are expressed as mean ± s.e.m. with the number of observations (n).

This work was supported by grants from the Schweizereische Stiftung für die Erforschung der Muskelkrankheiten, the Association Française contre les Myopathies, the European Union (Number HPRN-CT-2002-00331) and F.I.R.BRBAUO1ERMX, PRIN05, the Department of Anaesthesia, University Hospital Basel and by grants AG18755, AG13934 and AG15820 from the National Institute of Health/National Institute on Aging, USA.