Author manuscript, published in "The Journal of Biological Chemistry 2005;280(47):39302-8" DOI: 10.1074/jbc.M506566200 TRIADIN (TRISK 95) OVER-EXPRESSION BLOCKS EXCITATION-CONTRACTION COUPLING IN RAT SKELETAL MYOTUBES

To identify the function of triadin in skeletal muscle, adenovirus-mediated overexpression of Trisk 95 or Trisk 51, the two major skeletal muscle isoforms, was induced in rat skeletal muscle primary cultures, and the physiological behavior of the modified cells was analyzed. Overexpression did not modify the expression level of their protein partners ryanodine receptor, dihydropyridine receptor, and the other triadin. Caffeine-induced calcium release was also unaffected by triadin overexpression. Nevertheless, in the absence of extracellular calcium, depolarization-induced calcium release was almost abolished in Trisk 95 overexpressing myotubes (T95 myotubes), and not modified in Trisk 51 overexpressing myotubes (T51 myotubes). This was not because of a modification of dihydropyridine receptors, as depolarization in presence of external calcium still induced a calcium release, and the activation curve of dihydropyridine receptor was unchanged, in both T95 and T51 myotubes. The calcium release complex was also maintained in T95 myotubes as Trisk 95, ryanodine receptor, dihydropyridine receptor, and Trisk 51 were still co-localized. The effect of Trisk 95 overexpression on depolarization-induced calcium release was reversed by a simultaneous infection with an antisense Trisk 95 adenovirus, indicating the specificity of this effect. Thus, the level of Trisk 95 and not Trisk 51 is important on regulating the calcium release complex, and an excess of this protein can lead to an inhibition of the physiological function of the complex.

In skeletal muscle cells, excitation-contraction (E-C) 2 coupling is the process by which depolarization of the plasma membrane produces a large transient release of Ca 2ϩ from the sarcoplasmic reticulum, which in turn triggers contraction. Dihydropyridine receptors (DHPRs), the L-type Ca 2ϩ channels localized in the T-tubule membrane, serve as the voltage sensors for E-C coupling (1,2) and activate ryanodine receptors (RyRs), the intracellular Ca 2ϩ release channels of the sarcoplasmic reticulum membrane (3,4). In skeletal muscle, entry of Ca 2ϩ through DHPR is not required for E-C coupling (5). Instead, a voltage-dependent conformational change in the II-III loop of the skeletal muscle DHPR ␣1 subunit (Cav1.1) activates the skeletal muscle ryanodine receptor, RyR1. Activation of RyR1 results in a release of Ca 2ϩ from the sarcoplasmic reticulum intracellular stores into the cytosol, which in turn activates the cellular contractile apparatus to initiate muscle contraction. In cardiac muscle, the entry of external calcium through DHPR is required to induce activation of RyR and release of internal Ca 2ϩ via RyR, a mechanism known as calcium-induced calcium release. Calcium-induced calcium release is not the initial mechanism responsible of Ca 2ϩ release in skeletal muscle, but it can contribute to the amplification of the signal, and both calcium-induced calcium release and conformational coupling are involved in skeletal muscle contraction, to different extents (6).
Triadin is a transmembrane protein co-localized to the junctional sarcoplasmic reticulum membrane with RyR and was originally detected in rabbit skeletal muscle as a 95-kDa molecular mass protein (7)(8)(9). Several isoforms of triadin are expressed in rat skeletal muscle, and we have previously cloned two of these isoforms (10), both expressed at equivalent amounts in rat skeletal muscle. One is the rat homolog of the 95-kDa triadin identified in rabbit skeletal muscle, named Trisk 95, and the second one, Trisk 51, is a truncated form of Trisk 95 (461 amino acids instead of 687 amino acids) almost identical for the major part of the protein but with a unique C-terminal end of 6 amino acids. In cardiac muscle, multiple isoforms of triadin have also been identified (11,12), but the major if not only isoform expressed is CT1, a 32-kDa triadin isoform (13).
Trisk 95 is composed of a short N-terminal cytosolic domain, a single transmembrane domain, and a large C-terminal domain located in the lumen of the sarcoplasmic reticulum (9,14). The intraluminal domain, composed of a high density of positively and negatively charged amino acid residues, interacts with RyR and calsequestrin (15)(16)(17). Trisk 51 shares this same structure but has a smaller intraluminal domain.
The function of triadins in skeletal muscle is not precisely established for the moment. In skeletal muscle, triadin has been shown to inhibit the calcium channel activity of purified RyR (18,19). These experiments were the first to identify a function of triadin in skeletal muscle, and they led to the conclusion that triadin could regulate RyR function via inhibition of the channel. At that time, the presence of Trisk 51 in skeletal muscle had not been observed. To obtain new information concerning the respective function of Trisk 95 or Trisk 51 in skeletal muscle E-C coupling, we induced adenovirus mediated Trisk 95 or Trisk 51 overexpression in primary culture of skeletal muscle, and we then studied the behavior of the calcium release complex in these triadin-modified cells. We investigated intracellular Ca 2ϩ transients elicited by depolarizationinduced calcium release (K ϩ depolarization) or by direct RyR calcium channel activation (caffeine or 4-chloro-m-cresol, 4-CmC). Here, we show that the depolarization-induced Ca 2ϩ release was significantly decreased in the Trisk 95 overexpressing myotubes, and this effect was reversed by infection with an antisense virus. Trisk 51 overexpression has no effect, suggesting an isoform-specific action on E-C coupling. These results are discussed in terms of possible physiopathological consequences in some myopathies.

EXPERIMENTAL PROCEDURES
Antibodies-Rabbit or guinea pig antibodies directed against the specific C-terminal end of Trisk 95 or Trisk 51 were described previously (10). A polyclonal antibody directed against RyR purified from pig skeletal muscle was previously characterized (20). The sheep anti-␣1 subunit of DHPR was from Upstate Biotechnology.
Western Blot-The presence of RyR, ␣1 subunit of DHPR, Trisk 95, or Trisk 51 in different samples was assayed by Western blot (10), using a chemiluminescent reagent (Western lightning Chemiluminescence reagent plus, PerkinElmer Life Science). After electrophoretic separation on a 5-15% acrylamide gel, the proteins were electrotransferred to Immobilon P (Millipore). The secondary antibodies were labeled with horseradish peroxidase (Jackson ImmunoResearch laboratories).
Viruses and Infection-In this study, we used adenoviral-mediated gene transfer to modify Trisk 95 or Trisk 51 expression in primary cultures of rat skeletal muscle. The viruses were engineered and produced by the Gene Vector Production Network, at Genethon III (Evry, France). Four type 5 adenoviruses were used in this study, a control virus (AdV-DsRed) with the cDNA of the red fluorescent protein (DsRed), AdV-Trisk 95, an adenovirus with the full-length sequence of rat skeletal muscle Trisk 95 (10), AdV-AS-Trisk 95, an adenovirus with the nucleotides 1765-2064 of Trisk 95 in antisense orientation, and AdV-Trisk 51, an adenovirus with the full-length sequence of rat skeletal muscle Trisk 51 (10). All the transgenes were under the control of a CMV promoter. Cells were infected by overnight incubation with the chosen virus, at a multiplicity of infection 40, in proliferation medium, and then differentiation was induced by a change to differentiation medium. Control cells were infected with the control virus (AdV-DsRed). Trisk 95-overexpressing myotubes (T95-myotubes) consisted of cells infected with AdV-Trisk 95, and Trisk 51-overexpressing myotubes (T51-myotubes) with AdV-Trisk 51. For double infection, AdV-Trisk 95 (multiplicity of infection 40) and AdV-AS-Trisk 95 (multiplicity of infection 60) were simultaneously added to the cell culture. Calcium imaging studies were performed after 2-3 days of infection/differentiation. The infection of COS-7 cells was performed in the same way, except that the cells were kept in proliferation medium before and after infection (Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum and 1% penicillin-streptomycin).
Immunofluorescence Labeling-Myotubes (2-3 days differentiation) grown on plastic dishes coated with laminin were washed with PBS and fixed for 15 min at room temperature with 4% paraformaldehyde in PBS. After two washes with PBS for 5 min, the cells were permeabilized with 0.5% Triton X-100 in PBS for 5 min then saturated in PBS supplemented with 0.1% Triton X-100, 0.5% bovine serum albumin, and 2% goat serum for 30 min at room temperature. Cells were incubated with primary antibodies at room temperature for 2 h. After three washes with PBS, cells were incubated for 30 min at room temperature with fluorescent secondary antibodies, then washed three times with PBS, and the coverslips were mounted on microscope slides with anti-fading medium (DakoCytomation fluorescent mounting medium, DakoCytomation). For double/triple labeling, the two/three primary antibodies (from different species) and later the two/three secondary antibodies were added at the same time. The secondary antibodies were either labeled with Alexa-488, Alexa-546 (both from Molecular Probes), or Cy5 (from Jackson ImmunoResearch laboratories). The samples were analyzed by confocal laser scanning microscopy using a Leica TCS-SP2 operating system. The objectives used were Leica 40ϫ oil immersion (NA 1.25) or Leica 100ϫ oil immersion (NA 1.96). Fluorochromes were excited and collected sequentially (400 Hz line by line) by using the 488 nm line of an argon laser for Alexa-488, the 543 nm line of an helium-neon laser for Alexa-546, and the 633 nm line of an helium-neon laser for Cy5 excitation. Fluorescence emission was collected from 498 to 541 nm for Alexa-488, from 554 to 625 nm for Alexa-546, and from 640 nm to 730 nm for Cy5.
Immunoprecipitation-Rat skeletal muscle microsomes were solubilized in presence of CHAPS, and immunoprecipitation was performed with the chosen antibody as described previously (20). The immunoprecipitated proteins were then analyzed by Western blotting.
Intracellular Calcium Measurements-Changes in intracellular calcium were measured using the calcium-dependent fluorescent dye Fluo 4-AM (Molecular Probes). Myotubes were incubated for 1 h at room temperature with Fluo-4-AM 10 M, in Krebs buffer (136 mM NaCl, 5 mM KCl, 2 mM CaCl 2 , 1 mM MgCl 2 , 10 mM HEPES, pH 7.4). Uptake of the dye was facilitated by the addition of 0.02% pluronic F-127 acid (Sigma). After loading, myotubes were washed for 1 h at 37°C to allow de-esterification of the dye. Calcium imaging was performed in Krebs buffer. To obtain a calcium-free Krebs solution, CaCl 2 was left out, while 1 mM EGTA and 50 M lanthanum were added. In the depolarization solution, NaCl was replaced by KCl (140 mM final concentration). The other agonists, caffeine and 4-CmC, were prepared as stock solutions in Krebs buffer and used at final concentrations of 10 mM and 500 M, respectively.
Fluorescence was measured by confocal laser scanning microscopy, using a Leica TCS-SP2 operating system, in the xyt mode. Fluo-4 was excited by using the 488 nm wavelength, and its fluorescence was collected from 500 to 570 nm. The fluorescence images were collected every 1.6 s, for 2-4 min, and then analyzed frame by frame. Fluorescence curves are expressed as a function of time, as ⌬F/F, where F represents the baseline fluorescence immediately prior to depolarization, and ⌬F represents the fluorescence changes from baseline. Data are given as mean Ϯ S.E., and n represents the number of myotubes in each condition.
Patch Clamp-Whole-cell currents through voltage-gated calcium channels were recorded by means of the patch clamp technique (21). After removal of the culture medium, the cells were perfused with a medium containing triethylammonium-Cl 145 mM, CaCl 2 2 mM, MgCl 2 2 mM, HEPES 10 mM, pH 7.4 (TEAOH). Patch pipettes were pulled from thick wall borosilicate glass capillaries (1.5 mm outer diameter ϫ 0.86 mm inner diameter, Clark Electromedical Instruments, UK). When filled with the following intracellular solution, CsCl 120 mM, EGTA 10 mM, MgCl 2 2 mM, Na 2 -ATP 3 mM, HEPES 10 mM, pH 7.2 (CsOH), they had a resistance of ϳ2-5 megohms. Electrical signals were stimulated and recorded by means of the pClamp software (version 6.0, Axon Instruments). Voltage-gated calcium channel currents were filtered at 1-2 kHz and sampled at 5-10 kHz on an Axoclamp 200B amplifier (Axon Instruments). Myotubes were stimulated at a frequency of 0.2 Hz NOVEMBER 25, 2005 • VOLUME 280 • NUMBER 47 from a holding potential of Ϫ60 mV. Electrophysiological recordings were done at room temperature. Data are expressed as mean Ϯ S.E., with n being the number of myotubes tested.

RESULTS
Trisk 95 and Trisk 51 Are Associated with RyR-We have previously shown that two triadin isoforms are expressed in rat skeletal muscle, Trisk 95, the 95-kDa isoform, and Trisk 51, the 51-kDa isoform (10). To determine their association with RyR, immunoprecipitations were performed. These immunoprecipitations were done on solubilized rat skeletal muscle microsomes using antibodies specific for Trisk 51 (Fig. 1,  lane 2) or Trisk 95 (lane 3), and the immunoprecipitated proteins were analyzed by Western blot with anti-RyR antibodies. The presence of RyR was observed among the proteins immunoprecipitated by both antibodies, whereas it is absent from the proteins immunoprecipitated with non-immune serum (lane 4), as recently observed (22). This indicates that both triadin isoforms are involved in the calcium release complex and are, directly or not, associated with RyR. Thus, both isoforms could regulate RyR function.
Trisk 95 or Trisk 51 Overexpression Does Not Modify Expression of the Other Partners in the Complex-Trisk 95 or Trisk 51 overexpression was induced by adenoviral infection. Satellite cells were infected and induced to differentiate, and after 3 days of infection/differentiation myotubes were formed, and the cells were collected. The expression of both triadin isoforms in differentiated myotubes has been described previously (10). The infection of satellite cells with AdV-T95 induced 15-fold Trisk 95 overexpression ( Fig. 2A), whereas Trisk 51 expression was not significantly modified (Fig. 2B). In a similar way, the infection of satellites cells with AdV-T51 induced 40-fold overexpression of Trisk 51 (Fig. 2B), whereas Trisk 95 expression was not significantly modified ( Fig. 2A). When overexpressed, both Trisk 95 and Trisk 51 appeared as large and diffused bands, similar to the ones observed in rat skeletal muscle (lane 1), whereas the bands in young myotubes are more precisely defined and sharp, as previously observed (10). The expression level of RyR and DHPR was evaluated in control or in triadin overexpressing myotubes. Neither Trisk 95 overexpression nor Trisk 51 overexpression modified the expression level of RyR or DHPR (data not shown). Thus, Trisk 95 or Trisk 51 overexpression did not induce modifications in the expression level of the other proteins of the calcium release complex.
RyR Calcium Channel Function in Triadin-modified Cells-To determine whether RyR calcium channel function is preserved in T95myotubes and in T51-myotubes, we performed Fluo-4 confocal microscopy calcium imaging. We measured the calcium release induced by RyR activators 10 mM caffeine (23) or 500 M 4-CmC (24). The addition of caffeine (Fig. 3) or 4-CmC on cultured myotubes induced a massive calcium release from the internal stores in all of the myotubes. Compar-ison of the intensity of calcium signals induced by caffeine (Fig. 3) showed no significant differences among control myotubes, T95-myotubes, or T51-myotubes. Caffeine induced an increase in the relative fluorescence of 2.07 Ϯ 0.15 in control myotubes, whereas the increase in the relative fluorescence was of 1.92 Ϯ 0.12 in T95-myotubes and 2.02 Ϯ 0.19 in T51-myotubes, indicating a similar RyR functionality (non-significant difference, Student's t test). Thus, Trisk 95 or Trisk 51 overexpression did not modify the function of RyR as a calcium channel.
Depolarization-induced Calcium Release in Triadin-modified Cells-Depolarization of the plasma membrane induces activation of the DHPR and results both in (i) activation of RyR and internal calcium release, and (ii) influx of external calcium via DHPR if external calcium is present (25,26). Depolarization was induced in the absence of external calcium (Fig. 4A) on control, T95-myotubes, or on T51-myotubes by changing to 140 mM KCl. It resulted, as expected (27)(28)(29), in an increase in intracellular calcium concentration followed by a return to the basal level in control myotubes as well as in T51-myotubes. Nevertheless, in T95-myotubes, this effect was almost abolished, and depolarization induced only a minor calcium release (Fig. 4A). The relative increase in fluorescence was significantly reduced (p Ͻ 0.0001, Student's t test) from 1.54 Ϯ 0.21 in control myotubes to 0.30 Ϯ 0.06 in T95-myotubes.  To test the functionality of DHPR and to check that Trisk 95 overexpression had not led to an inhibition of DHPR, the same depolarization experiment was performed, but this time in the presence of external calcium (Fig. 4B). As expected, in control myotubes, as well as in T51myotubes, the depolarization induced a larger calcium transient (2.08 Ϯ 0.22 in presence of external calcium instead of 1.54 Ϯ 0.21 in the absence of calcium for control myotubes, and 2.06 Ϯ 0.18 in presence of calcium compared with 1.60 Ϯ 0.18 in the absence of calcium for T51myotubes). The same effect was observed in T95-myotubes, and depolarization induced a larger calcium transient than in the absence of extracellular calcium, but the ⌬F/F remained significantly lower than in control myotubes (1.01 Ϯ 0.10 in T95 myotubes and 2.08 Ϯ 0.22 in control myotubes; p Ͻ 0.0001, Student's t test). These results suggest that (i) the DHPR still functions as a permeating Ca 2ϩ channel in Trisk 95 overexpressing myotubes, (ii) the depolarization-induced calcium release was absent in these myotubes, and only the calcium induced calcium release component is observed. Thus, overexpression of Trisk 95 abolishes only the "skeletal-type" excitation-contraction coupling (the direct coupling between DHPR and RyR), and the calcium-induced calcium release is not affected. The overexpression of Trisk 51 did not induce any modification in the depolarization-induced calcium release.
Electrophysiological Analysis of DHPR in T95-myotubes and T51-myotubes-As the depolarization in absence of external calcium (skeletal-type E-C coupling) was abolished when Trisk 95 was overexpressed, we wanted to know whether Trisk 95-overexpression could have shifted the maximal activation potential of the DHPR. If so, then the addition of 140 mM KCl, which results in a depolarization to ϳ0 mV may not be adapted to the optimal activation of DHPR. A whole-cell patch clamp analysis was performed on control, T95-myotubes, and T51-myotubes. Depolarizing steps were delivered to the myotube membrane via the patch pipette, and the resulting DHPR calcium current was recorded. The relationship between the depolarization and the maximal intensity of the recorded current is represented in Fig. 5. The curve corresponding to T95-overexpressing myotubes is superimposed with the curves obtained in control myotubes and in T51-myotubes. There is no significant modification of DHPR activation in T95-overexpressing myotubes compared with control myotubes or T51-myotubes. Thus, the absence of Ca 2ϩ release in T95-overexpressing myotubes upon membrane depolarization (Fig. 4A) is not the result of a shift in the activation potential of DHPR, with a subsequent reduced activation of DHPR upon membrane depolarization.
Co-localization of the Members of the Calcium Release Complex-Overexpression of Trisk 95 did not modify the expression of the other members of the calcium release complex (RyR, DHPR, Trisk 51) nor did it modify the function of RyR and of DHPR independently. We then hypothesized that Trisk 95 overexpression might induce a dissociation  's t test). B, the same KCl depolarization was realized in presence of 2 mM external calcium. In this condition, the mean fluorescence variation was 2.08 Ϯ 0.22 in control myotubes, 1.01 Ϯ 0.10 in T95 myotubes, and 2.06 Ϯ 0.18 in T51 myotubes. The asterisk (*) indicates a statistically significant difference (p Ͻ 0.0001 Student's t test).

FIGURE 5. I-V relationship of DHPR in control or triadin overexpressing myotubes.
Voltage changes from a holding potential of Ϫ60 mV were imposed to myotubes through the patch pipette, and the resulting calcium current was recorded. The maximum amplitude of this current was plotted as a function of membrane depolarization. Data presented are mean Ϯ S.E., with n being the number of myotubes tested. NOVEMBER 25, 2005 • VOLUME 280 • NUMBER 47 of the members of the calcium release complex, and so we used immunostaining to check the co-localization of RyR, DHPR, Trisk 95, and Trisk 51. Double or triple immunofluorescent labeling was performed on T95-overexpressing myotubes, with antibodies directed against RyR, DHPR, T95, or T51 (Fig. 6). In T95-myotubes, DHPR, T95, and RyR presented a similar punctuated pattern (Fig. 6, a-c), as observed for normal myotubes (30,31). Most of the dots co-localized for the three antibodies, indicating that the co-localization was preserved in T95myotubes. The overexpressed Trisk 95 was normally targeted to the calcium release complex. In Fig. 6, d and e, double immunofluorescent labeling was performed with antibodies against DHPR and Trisk 51 on T95-myotubes, and both proteins were still co-localized (f). Thus, not only is Trisk 95 correctly targeted in T95-myotubes, but also its overexpression does not disrupt the calcium release complex.

Trisk 95 Blocks Excitation-Contraction Coupling
Restoration of Skeletal Excitation-Contraction Coupling by Trisk 95-antisense Virus-An antisense-Trisk 95 adenovirus (AdV-AS-T95) has been developed to block Trisk 95 expression. The ability of this antisense virus to block the overexpression of Trisk 95 induced by infection with AdV-T95 was first checked in COS-7 cells. The presence of Trisk 95 was analyzed by Western blot either in AdV-T95-infected cells or in AdV-T95 ϩ AdV-AS-T95-infected cells. The result is presented in Fig. 7A, a. Simultaneous infection of cells with AdV-AS-T95 and AdV-T95 led to a complete block of Trisk 95 expression. This indicates that AS-T95 virus indeed has the ability to block Trisk 95. This effect is not an unspecific effect resulting from double infection, as it is not observed in the case of double infection AdV-T51 ϩ AdV-AS-T95 (Fig. 7A, b). The infection with AdV-AS-T95 does not block overexpression induced by AdV-T51.
Rat skeletal muscle primary cultures were infected simultaneously with AdV-T95 and AdV-AS-T95 viruses, and calcium imaging analysis was performed. The same KCl-induced depolarization was realized on AdV-T95 ϩ AdV-AS-T95-infected cells in absence of external calcium and is presented on Fig. 7B, with the previous experiments in gray (shadowed). This double infection in myotubes restored the depolarizationinduced calcium release. The relative fluorescence increase induced by KCl depolarization shifted from 0.30 Ϯ 0.06 in T95-myotubes to 1.19 Ϯ 0.14 in double infected myotubes (significantly different, p Ͻ 0.0001, Student's t test). The depolarization had a similar effect in control myotubes and in T95 ϩ AS-T95 myotubes (1.54 Ϯ 0.21 compared with 1. 19 Ϯ 0.14, not significantly different, p ϭ 0.17, Student's t test). This  indicates that the reduction in skeletal E-C coupling is specific to Trisk 95 expression.

DISCUSSION
To study the respective function of Trisk 95 and Trisk 51 in the calcium release complex, Trisk 95 or Trisk 51 have been overexpressed in rat skeletal myotubes via recombinant adenoviruses. The overexpressions are specific to the targeted protein (Trisk 95 or Trisk 51), and the expression level of other proteins in the calcium release complex is unchanged (DHPR, RyR, Trisk 51, or Trisk 95). This situation, due to an acute modification of a single protein, has already been observed with adenoviral-mediated overexpression of FKBP12.6 in cardiomyocytes (32), which did not induce any modification in RyR2 expression level. This contrasts with transgenic mice overexpressing single proteins of the cardiac calcium release complex, which results in a compensatory modification of the other proteins of the complex, as observed for cardiac calsequestrin overexpression (33), cardiac junctin overexpression (34), and cardiac triadin CT1 overexpression (35). In the case of transgenic mice with overall modifications of different proteins in addition to the targeted one, the attribution of a function to a single protein is more difficult to achieve.
Trisk 95 or Trisk 51 overexpression had no effect on direct stimulation of RyR (caffeine-or 4-CmC-induced calcium release). Therefore RyR function was not modified by Trisk 95 or Trisk 51 overexpression, and in addition, the amount of releasable calcium was equivalent with or without T95/T51 overexpression. On the contrary, when Ca 2ϩ release was induced by depolarizing the plasma membrane, control and T51myotubes were different from Trisk 95-overexpressing myotubes. In the absence of external calcium, a situation in which only the skeletal E-C coupling was assayed, the ability of plasma membrane depolarization to induce Ca 2ϩ release was largely altered in T95-overexpressing myotubes compared with control myotubes or T51-overexpressing myotubes. In the presence of external calcium, a situation in which both skeletal and cardiac E-C coupling were challenged, plasma membrane depolarization induced a smaller Ca 2ϩ release in T95 myotubes than in control or T51 myotubes. Calcium-induced calcium release was still functioning in T95-myotubes, indicating that depolarization activated DHPR and induced an influx of calcium into the cell, which activated RyR and resulted in internal calcium release. Nevertheless, as the skeletal E-C coupling was blocked, the resulting effect of depolarization in presence of external calcium was reduced compared with control myotubes.
Another technique, patch clamp, was used to control the functionality of DHPR. Calcium currents via the DHPR were measured at different membrane potentials. The maximum current for each membrane depolarization (I-V curve) showed no difference among control, T95-myotubes, and T51-myotubes. Thus, we concluded that neither Trisk 95 nor Trisk 51 overexpression induced any major modification in DHPR function.
As both DHPR and RyR are functional, but no longer functionally coupled, we hypothesized that T95 overexpression could have induced a dissociation of both channels, which we have previously shown to be associated in skeletal muscle (20), or that T95 overexpression could have induced a dissociation of Trisk 51 from the calcium release complex, and therefore the observed effect should be attributed to a depletion of the complex in Trisk 51. Thus, double or triple immunolabeling was performed, and we observed that the overexpressed Trisk 95 was correctly targeted and remained perfectly co-localized with RyR, DHPR, and Trisk 51. The amount of Trisk 95 associated with RyR is not known, but it can be increased 15-fold without dissociation of Trisk 95 from the calcium release complex.
Therefore, we imagine that both proteins RyR and DHPR are still co-localized and associated, but by steric hindrance, the overexpressed Trisk 95 should block the conformational change induced in RyR by DHPR upon membrane depolarization. Another explanation could be an effect of Trisk 95 overexpression on DHPR charge movement, which is responsible for the skeletal type of E-C coupling. This overexpression could result in a faster inactivation of DHPR, then to a reduced stimulation of RyR, and consequently to a reduction in calcium release. Such an effect of triadin should be indirect, because the cytoplasmic domain of triadin is too short to make a contact with the DHPR in the adjacent T-tubule membrane (14).
All of the triadin isoforms are issued from an alternative splicing of the same triadin gene (36); nevertheless, the expression pattern of triadin is different in cardiac and skeletal muscles. Trisk 95 and Trisk 51 are mainly, if not exclusively, expressed in skeletal muscle. Therefore the lack of effect of these two skeletal isoforms on the "cardiac part" of E-C coupling is probably to be related to the fact that they are not expressed in cardiac muscle and they act only in a skeletal environment. What is more surprising is the major effect of Trisk 95 overexpression on skeletal E-C coupling compared with the lack of effect related to Trisk 51 overexpression. Both proteins are present in equivalent amount in rat skeletal muscle (10), both are associated with RyR, and they both have the same important RyR-interaction domains previously identified, the N-terminal cytoplasmic domain (19) and the KEKE region (37). Nevertheless, as these two proteins do not have the same function, this may suggest that the functional regulation of RyR by triadin does not involve these interaction domains. The KEKE is probably necessary for the interaction RyR/triadin, but it is probably not sufficient for functional regulation. The regulation of E-C coupling via Trisk 95 should be related to a specific part of Trisk 95 and should somehow involve amino acids 455-687.
We have recently observed that three triadin isoforms are expressed in a similar amount in skeletal muscle: Trisk 95, Trisk 51, and Trisk 32 (22). Trisk 32 is a shorter triadin isoform, which is identical to the major cardiac triadin isoform CT1. Nevertheless, we observed that Trisk 32 is not associated to RyR (immunolocalization and immunoprecipitation), but to the IP 3 -receptor, the other intracellular calcium channel. Even with the presence of the KEKE motif, Trisk 32 cannot interact with RyR, and therefore, Trisk 32 probably cannot regulate calcium release via RyR in skeletal muscle. On the contrary, in cardiac muscle, CT1 overexpression has recently been obtained by adenoviral infection (38). The authors concluded that CT1 overexpression stimulated cardiac excitation-contraction coupling, and that this effect was mediated by the KEKE motif in region 200 -224. Thus, as Trisk 32 is equivalent to CT1, its function is most probably different in cardiac and in skeletal muscle, and its function has to be identified in skeletal muscle.
The effect of Trisk 95 overexpression on E-C coupling could be reversed by simultaneous infection with an antisense virus that blocked T95 overexpression. Thus, this effect is indeed correlated to the amount of Trisk 95, and one could imagine that a similar effect could be obtained in pathological situations. Either the total amount of Trisk 95 is an important factor, or most probably the relative amount of Trisk 95 compared with RyR is the key factor of a functional calcium release complex. Indeed, we have recently observed in a pathological situation a drastic reduction of RyR expression in a patient affected with Multimini Core Disease (39). In fact, reduction of RyR compared with triadin could be equivalent to increase in triadin level compared with RyR.
Triadin expression levels should be tested in this patient for any modi-