Dyspedic mouse skeletal muscle expresses major elements of the triadic junction but lacks detectable ryanodine receptor protein and function.

The ry153 dyspedic mouse contains two disrupted alleles for ryanodine receptor type 1 (skeletal isoform of ryanodine receptor; Ry1R) resulting in perinatal death. In the present study, whole skeletal muscle homogenates and sucrose gradient-purified junctional sarcoplasmic reticulum from neonatal wild-type and dyspedic mice were assayed for biochemical and functional markers. Equilibrium binding experiments performed with 1-120 nM [3H]ryanodine reveal saturable high and low affinity binding to membrane preparations from wild-type mice, but not to preparations from dyspedic mice. Binding experiments performed with [3H]PN200 show a 2-fold reduction in [3H]PN200 binding capacity in dyspedic muscle, compared to age-matched wild-type muscle, with no change in receptor affinity. The presence or absence of proteins known to be critical for normal ryanodine receptor/Ca2+ channel complex function was assessed by Western blot analysis. Results indicate that FKBP-12, DHPRα1, triadin, calsequestrin, SERCA1 (sarco(endo)plasmic reticulum Ca2+ ATPase), and skeletal muscle myosin heavy chain are present in both dyspedic and wild-type muscle. Only wild-type membranes showed immunoreactivity toward Ry1R antibody. Neither dyspedic nor wild-type mouse muscle showed detectable immunoreactivity toward Ry2R or Ry3R antibodies, even after sucrose gradient purification of sarcoplasmic reticulum. These results indicate that proteins critical for ryanodine receptor function are expressed in dyspedic skeletal muscle in the absence of Ry1R. Ca2+ transport measurements show that membranes from wild-type controls, but not dyspedic mice, release Ca2+ upon exposure to ryanodine. Dyspedic mice and cells derived from them serve as excellent homologous expression systems in which to study how Ry1R structure relates to function.

The dyspedic mouse contains two disrupted alleles (ry1 53 / ry1 53 ) for ryanodine receptor type 1 (skeletal isoform of ryanodine receptor; Ry 1 R) 1 resulting in a birth lethal defect. Skeletal muscle from dyspedic mice lack excitation-contraction (E-C) coupling (1) 2 but maintains many ultrastructural details of the triadic junction. Two important differences have been observed with thin section transmission and freeze fracture scanning microscopy of dyspedic muscle. First, dyspedic muscle lacks the regularly spaced array of junctional feet which span the gap between the t-tubule and SR membranes, significantly reducing the gap size (3). 2 Second, dyspedic fibers lack the tetradic arrangement of dihydropyridine receptors which is characteristic of normal fibers (3). It is not yet known if dyspedic muscle lacking Ry 1 R expression results in altered expression of other key triadic proteins involved in modulating SR Ca 2ϩ transport.
In addition to Ry 1 R, separate genes encode two other ryanodine receptors, namely the cardiac (Ry 2 R) and "brain" (Ry 3 R) isoforms (4 -8). Ry 1 R is predominantly expressed in skeletal muscle and cerebellar Purkinje cells (9,10). Ry 2 R is predominately expressed in cardiac tissue, where it functions as a Ca 2ϩ -induced Ca 2ϩ release channel, but is also widely expressed in brain tissues (7,9). In comparison, Ry 3 R is widely expressed at low levels in many non-muscle cells, in mammalian brain, and in very low levels in heart and skeletal muscle (11)(12)(13). However, its functional significance is presently unclear.
The development of Ry 1 R-null (dyspedic) mice (1) 2 represents a significant advance in the goal to understand the molecular mechanisms regulating the function of Ry 1 R. Recent studies have shown that dyspedic mice die perinatally and lack skeletal E-C coupling (1). 2 Dyspedic skeletal muscle fibers have also been found to have 30-fold less L-type Ca 2ϩ entry current than control fibers, despite the presence of comparable levels of immobilization-resistant charge movement. Significantly, both voltage-dependent Ca 2ϩ entry current and SR Ca 2ϩ release are concomitantly restored in cultured dyspedic myotubes by micronuclear injection of Ry 1 R cDNA (14). Despite the loss of E-C coupling, dyspedic muscle fibers in culture exhibit small Ca 2ϩ fluxes in response to caffeine and adenine nucleotide (14,15). The pharmacological responses in cultured dyspedic muscle myotubes have been attributed to enhanced expression of Ry 3 R mRNA (15). However, there has been no direct evidence of Ry 3 R protein expression in dyspedic muscle.
The present study demonstrates for the first time that dys-pedic muscle 1) exhibits a 2-fold reduction in [ 3 H]PN200 binding capacity with a concomitant change in immunoreactive DHPR␣1 subunit expression, 2) lacks detectable levels of specific high or low affinity [ 3 H]ryanodine-binding sites, 3) lacks immunoreactivity toward Ry 1 R, Ry 2 R, or Ry 3 R antibody, and 4) lacks a ryanodine-sensitive Ca 2ϩ efflux pathway across SR membranes, compared with skeletal muscle from wild-type littermates. However, the complement of other triadic proteins known to be critical for normal E-C coupling (triadin, FKBP-12 (12-kDa FK506-binding protein), calsequestrin, SERCA1), and skeletal myosin heavy chain are shown to be expressed in dyspedic muscle.

EXPERIMENTAL PROCEDURES
Dyspedic Mice-Dyspedic mice were obtained as described elsewhere. 2 Briefly, a 9-kilobase EcoRI-Tth111I genomic fragment which contains part of the skeletal Ry 1 gene was screened from a 129/Sv mouse genomic library and was used to construct a targeting vector. A neomycin-resistance gene (Neo) was used to disrupt transcription of the mRNA and select for homologous recombinants. A herpes simplex virus thymidine kinase gene was used to reduce the number of non-homologously targeted clones. Both were under the control of a phosphoglycerate kinase promoter and were inserted at the KpnI and EcoRI sites, respectively. Approximately 2.5 ϫ 10 7 J1-ES cells were electroporated with the targeting vector and subjected to double selection with G418 and FIAU. Homologously targeted ES clones were injected into blastocysts from C57BL/6 mice and transferred to pseudopregnant females. Chimeric males were bred to C57BL/6 females, and germline transmission was verified. Mutant mice homozygous for the Ry 1 targeted allele were obtained from heterozygous matings and verified by PCR and Southern blot.
Tissue Preparation-Crude and sucrose gradient-purified membrane fractions were prepared from wild-type (ry 1 ϩ/ϩ) and dyspedic (ry 1 -/-) neonates under identical conditions. Immediately after birth, wild-type animals were euthanized by cervical dislocation. Tail tissue from each sample (dyspedic and wild-type) was removed for PCR analysis. The legs were removed from each animal at the hip or shoulder, trimmed of feet, skinned, and frozen in liquid nitrogen. Wild-type or dyspedic tissues were then pooled and prepared in the following manner. Legs were thoroughly homogenized on ice using a Polytron at high speed in 30-s bursts in homogenization buffer consisting of 300 mM sucrose, 25 mM Hepes, pH 7.1, 200 M phenylmethylsulfonyl fluoride (prepared immediately before use), and 10 g/ml leupeptin. Bone fragments were removed by gentle centrifugation at 50 ϫ g for 2 min at 4°C. The homogenate was centrifuged at 110,000 ϫ g for 90 min at 4°C. The resulting pellet was resuspended in 10% sucrose, 25 mM Hepes, pH 7.1, to a protein concentration of approximately 3 mg/ml, snap-frozen in liquid nitrogen, and stored at Ϫ80°C until used.
In some experiments, enrichment of the heavy SR membrane fraction from normal and dyspedic homogenates was achieved by sucrose gradient sedimentation. Crude membrane homogenates (3-5 ml) were layered onto a discontinuous sucrose gradient (2 ml 27%, 3 ml 32%, 3 ml 34%, 3 ml 38%, and 2 ml 45%) and centrifuged at 70,000 ϫ g for 16 h at 4°C. Membrane fractions removed from the 38 -45% sucrose step interface were pooled, diluted to 10% sucrose with a 25 mM Hepes pH 7.1 buffer, homogenized, and pelleted at 110,000 ϫ g for 90 min at 4°C. The pellets were resuspended in 10% sucrose, 25 mM Hepes, pH 7.1, to a protein concentration of approximately 1 mg/ml, snap-frozen in liquid nitrogen, and stored at Ϫ80°C until used.
The following tissue preparations were used as reference standards for Western blot analysis. Rabbit fast skeletal muscle junctional SR was isolated according to Saito et al. (16), rat cardiac SR membranes were isolated according to Feher et al. (17), and avian SR was isolated from pectoralis muscles according to Airey et al. (18). A whole membrane homogenate was prepared from rat testicles following the procedure described for wild-type mouse muscle, with the exception that the 50 ϫ g centrifugation step was replaced by a 30-min, 8000 ϫ g centrifugation step. Protein was quantitated in all preparations by the method of Lowry (19) using bovine serum albumin as a standard.
Radioligand Binding Assay-High and low affinity binding of [ 3 H]ryanodine (84 Ci/mmol, DuPont NEN) to 0.2 mg/ml skeletal muscle microsomes was measured in the presence of 250 mM KCl, 15 mM NaCl, 20 mM Hepes, pH 7.1, 50 M Ca 2ϩ , 1 nM [ 3 H]ryanodine, and 0.5-120 nM unlabeled ryanodine (Calbiochem). The reaction was initiated by the addition of tissue and allowed to equilibrate at 37°C for 3 h. Nonspecific binding was assessed in the presence 260 nM unlabeled ryanodine.
Separation of bound and free ligand was performed by filtration through Whatman GF/B glass fiber filters using a Brandel (Gaithersburg, MD) cell harvester. Filters were washed with three volumes of 0.5 ml of ice-cold wash buffer containing 20 mM Tris-HCl, 250 mM KCl, 15 mM NaCl, 50 M CaCl 2 , pH 7.1, and placed into vials containing 5 ml of scintillation mixture (Ready Safe, Beckman). Binding of radioligand to muscle membranes was determined by scintillation spectrometry. K d and B max values were derived from Scatchard analysis of the binding data.
Specific binding of 0.06 -5.0 nM [ 3 H]PN200 (83 Ci/mmol, DuPont NEN) to the ␣1 subunit of the L-type Ca 2ϩ channel (the dihydropyridine receptor) was measured in the presence of 140 mM NaCl, 15 mM KCl, 20 mM Hepes, pH 7.0, and 0.2 mg/ml skeletal muscle homogenate. The reaction was initiated by the addition of tissue and allowed to equilibrate at 22°C for 30 min in the dark. Paired nonspecific controls were measured in the presence of 10 M nifedipine. Separation of bound and free ligand was performed as described for high affinity [ 3 H]ryanodine binding except that the filters were washed with 5 ϫ 2 ml of wash buffer. K d and B max values were derived from Scatchard analysis of the binding data.
Electrophoresis and Immunoblot Analysis-Constituent proteins from membrane preparations were resolved on 3-10% gradient, 4 -20% gradient, or 7% isocratic gels by the method of Laemmli (20). Gels were either stained with silver (Silver Stain Plus, Bio-Rad) or Stains All (Sigma), or transferred onto polyvinylidene difluoride (PVDF) membranes (Immobilon-P, Millipore) for subsequent Western blot analysis. For Western blots, proteins were electroblotted (Mini Trans-Blot, Bio-Rad) overnight at 30 V, followed by a 60-min fast transfer at 100 V. Nonspecific antibody binding was blocked by incubating blots for 1 h at 37°C in TTBS buffer solution (20 mM Tris-HCl, 500 mM NaCl, 0.5% Tween 20, pH 7.5) with the addition of 5% bovine serum albumin or 5% nonfat dry milk. Specific binding of the primary antibody of interest was performed by incubating the blots for 1 h at 37°C in TTBS buffer in the presence of 1% bovine serum albumin and antibody. Resulting immunoblots were labeled with horseradish peroxidase-conjugated goat antimouse (Sigma) or donkey anti-rabbit (Amersham) secondary antibody for 1 h at 37°C and then visualized using either colorimetric (TMB, Vector Laboratories) or chemiluminescent (ECL, Amersham) techniques. In some cases, exposed films from ECL were quantitated using a densitometer (model PC9310, Shimadzu). Nonspecific binding of secondary antibodies to membrane preparations was minimized by performing a dilution series in the absence of primary antibody. Antibodies were purchased or generously provided as follows: Ry  Transport Measurements-Ca 2ϩ flux measurements using microsomal membranes from wild-type or dyspedic mouse skeletal muscle were performed fluorometrically (SLM AB-2, SLM-Aminco). Briefly, 50 g of skeletal muscle microsomal membranes were equilibrated to 37°C in a buffer consisting of 92 mM KCl, 7.5 mM Na 4 P 2 O 7 , 20 mM MOPS, pH 7.0, 0.01% NaN 3 , 2-10 M Ca 2ϩ , and 0.5 M fluo-3. A coupled enzyme system (20 g/ml creatine phosphokinase and 5 mM phosphocreatine) was present to maintain ATP concentrations. Loading of Ca 2ϩ was initiated by the addition of 1 mM MgATP. Transport of Ca 2ϩ into or out of microsomal membranes was determined by following changes in the fluo-3 fluorescence intensity (excitation at 500 nm, emission at 530 nm). The presence of Na 4 P 2 O 7 in the transport buffer maintained a linear dye response at added Ca 2ϩ concentrations up to 8 M. Ca 2ϩ efflux was initiated by the addition of 20 or 200 M ryanodine. Ca 2ϩ accumulation by microsomal stores was verified by addition of 2 g/ml 4-bromo-A23187. Linearity of fluo-3 emission with increasing Ca 2ϩ concentration was verified after each experiment by adding known aliquots from a National Bureau of Standards Ca 2ϩ stock.

Silver Stain of Dyspedic and Wild-type Control Mouse Skeletal
Muscle-Wild-type and dyspedic neonatal mouse skeletal muscle proteins, resolved on 3-10% Laemmli gels and visualized by silver stain, are shown in Fig. 1. Lanes containing either dyspedic (dys) or wild-type (w-t) preparations exhibit a similar pattern of staining and density of protein bands, with the major exception of a band in wild-type lanes corresponding in size to the Ry 1 R protomer found in junctional rabbit skeletal muscle SR (jsr lane, marked with arrow).
The Ry 1 R, Ry 2 R, and Ry 3 R Isoforms Are Not Detectable in Dyspedic Mouse Skeletal Muscle-The equilibrium binding of [ 3 H]ryanodine to neonatal wild-type and dyspedic skeletal muscle microsomes was investigated under conditions that stabilize high affinity binding of the alkaloid to ryanodine receptors (21). Fig. 2 shows that membranes prepared from dyspedic skeletal muscle do not possess specific high affinity [ 3 H]ryanodine-binding sites, whereas preparations of newborn wild-type muscle exhibit 148 Ϯ 16 fmol/mg of high affinity [ 3 H]ryanodine binding with a K d of 6.9 Ϯ 1.3 nM (mean B max from three independent preparations, each performed in duplicate). Scatchard analysis of data from control muscle membranes is best fit using a two-site model, with a second site having lower affinity (K d ϭ 20.3 nM) and B max of 305 Ϯ 94 fmol/mg of protein.
Experiments designed to measure the presence of binding sites having lower affinity were assessed using [ 3 H]ryanodine concentrations from 250 -1600 nM (specific activity 1.2 Ci/mmol) and revealed that while wild-type preparations were fully saturated, dyspedic preparations failed to show any specific binding (data not shown).
The binding data presented above suggest that dyspedic muscle either lacks general expression of ryanodine receptor isoforms or, if these proteins (e.g. Ry 3 R) are expressed, they fail to measurably recognize the alkaloid. We directly examined whether or not dyspedic muscle expresses any of the known ryanodine receptor isoforms by Western blot analysis. Proteins from wild-type and dyspedic skeletal muscles were resolved by SDS-PAGE and transferred overnight onto PVDF membranes as described under "Experimental Procedures." Blots were probed using antibodies selective for either Ry 1 , Ry 2 , or Ry 3 receptors. Specific antibody labeling was visualized using a highly sensitive chemiluminescent (ECL) technique.
Blots probed with a Ry 1 R-selective mouse monoclonal antibody (22) stain positive for the ϳ560-kDa protomer in a protein concentration-dependent manner with preparations from wildtype neonatal mice and with rabbit junctional SR (Fig. 3, top panel, w-t and jsr lanes). In comparison, dyspedic skeletal muscle preparations lack any detectable immunoreactivity with the Ry 1 R antibody (Fig. 3, top, dys lanes), as does rat cardiac SR (Fig. 3, top, crd lane).
As expected, Western blot analysis using a Ry 2 R-selective monoclonal antibody (23) shows an absence of reactivity with both dyspedic and wild-type skeletal muscle preparations (Fig.  3, middle panel, dys and w-t lanes). However, the antibody strongly recognizes the Ry 2 R protein found in rat cardiac preparations, which is included as a positive control (Fig. 3, middle, crd lane).
Western blot analysis using a Ry 3 R-selective polyclonal antibody (8) does not recognize any protein associated with either the dyspedic or wild-type membrane preparations (Fig. 3, bottom panel, dys and w-t lanes), nor does it recognize any proteins associated with rabbit fast skeletal muscle junctional SR (Fig.  3, bottom, jsr lane). However, the antibody strongly recognizes the Ry ␤ receptor isoform found in avian pectoralis muscle, which possesses Ͼ80% sequence homology with the mammalian Ry 3 R "brain" isoform ( Fig. 3, bottom, avi lanes). The anti-FIG. 1. SDS-PAGE of wild-type and dyspedic skeletal muscle proteins reveals the presence and absence of Ry 1 R, respectively, by silver stain. SDS-PAGE was performed with wild-type and dyspedic mouse skeletal muscle preparations according to the method of Laemmli using 3-10% gradient gels. Proteins were visualized using the Silver Stain Plus kit (Bio-Rad). jsr, 4 g of rabbit skeletal junctional SR; dys, 4, 8, and 12 g of dyspedic mouse microsomal membranes, respectively; w-t, 4, 8, and 12 g of wild-type mouse microsomal membranes, respectively; std, protein standards. Molecular mass markers (indicated by arrows along left side of figure) are 170 kDa (reduced ␣ 2 -macroglobulin), 116 kDa (␤-galactosidase), 85 kDa (fructose 6-phosphate), 55 kDa (glutamate dehydrogenase), 39 kDa (aldolase), and 26 kDa (triosephosphate isomerase).

FIG. 2. Dyspedic skeletal muscle membranes lack detectable high affinity binding sites for [ 3 H]ryanodine.
Specific binding of [ 3 H]ryanodine to membrane preparations from wild-type and dyspedic mice were performed as described under "Experimental Procedures" in the presence of 1 nM [ 3 H]ryanodine and 0.5-120 nM unlabeled ryanodine. Nonspecific binding was determined in the presence of 260 nM unlabeled ryanodine. Scatchard analysis (inset) of specific binding found in wild-type preparations was best fit by a two-site model. The high affinity site gave a B max of 148 Ϯ 16 fmol/mg of protein and a K d of 6.9 Ϯ 1.3 nM; the lower affinity site gave a B max of 305 Ϯ 94 fmol/mg of protein and a K d of 21 Ϯ 1.8 nM (pooled data from three independent experiments performed in duplicate). In contrast, specific ryanodine binding was not detected with dyspedic mouse skeletal muscle (three independent preparations). body also recognizes Ry 3 R protein in rat testicular tissue (Fig.  3, bottom, tst lane), which has been shown to express very high levels of Ry 3 R mRNA (8). Sequential labeling experiments with Ry 1 R and Ry 3 R antibodies reveal that the Ry 3 R protomer found in avian muscle and rat testicular tissue migrates with a smaller apparent size than Ry 1 R.
Expression of Ry 1 R and Ry 3 R proteins was further investigated using SR membrane fractions purified by sucrose density gradient centrifugation. Gradient fractions isolated at the 38 -45% sucrose interface and within the 45% sucrose layer were analyzed by Western blot using Ry 1 R-and Ry 3 R-selective antibodies. Fig. 4 , top panel, shows that lanes containing either whole membrane fractions (memb), 38 -45% sucrose gradient fractions (38 -45%), or the 45% sucrose fraction (45%) react with the Ry 1 R-selective antibody only in preparations obtained from wild-type muscle, whereas dyspedic muscle preparations completely lack immunoreactivity toward this antibody. In comparison, no immunoreactive protein could be detected with the Ry 3 R polyclonal antibody in the purified fractions from either the wild-type or dyspedic muscle preparations (Fig. 4,  bottom panel). Multiple bands labeled by the Ry 1 R-selective antibody in the 38 -45% sucrose gradient lane of wild-type muscle (Fig. 4, top) are most likely proteolytic fragments of the ryanodine receptor with M r estimated to be 250,000 and 300,000, which is reflective of the fragmentation pattern of Ry 1 R induced by trypsin (24) or calpain (25) digestion. The sedimentation density of the vesicles could be related to the degree of proteolysis of Ry 1 R, especially since Ry 1 R fragmentation was not observed within the 45% sucrose gradient fraction.
The results presented here and elsewhere (1, 3, 15) 2 clearly demonstrate that dyspedic mouse skeletal muscle does not express the skeletal isoform (Ry 1 R) of the ryanodine receptor. In the present study, specific antibodies fail to detect the presence of Ry 1 R, Ry 2 R or, more importantly, Ry 3 R protein in the neonatal muscle, even when the highly sensitive ECL technique is used in conjunction with sucrose gradient purification of SR. Furthermore, analysis of high and low affinity [ 3 H]ryanodine binding fails to support the presence of any known ryanodine receptor proteins in these preparations. Since dyspedic and wild-type skeletal muscle preparations are from newborn mice, SR volume (26) and frequency of SR/t-tubule junctions are much less than those found in adult mouse muscle (26 -28). This morphological difference, along with results from the Western blots presented here, suggests that neonatal mouse skeletal muscle either does not express Ry 3 R protein or expresses it at levels below detection limits.
The question of whether dyspedic muscle expresses an alternate ryanodine receptor isoform (Ry 2 R or Ry 3 R) is important, since responses to Ca 2ϩ , caffeine, and ryanodine have been observed in cultured dyspedic myotubes (1,14,15). However, the total Ca 2ϩ released in response to these ligands was observed to be 10 -15-fold less than that seen in normal myotubes, and the time course for release was significantly slower (1,14). Furthermore, cultured dyspedic myotubes have been reported to lack responsiveness to a second addition of caffeine when ryanodine is introduced between caffeine additions (14,15). Analysis of RNA from cultured myotubes using reverse transcription-polymerase chain reaction have revealed increased Ry 3 R mRNA expression in cultured dyspedic myotubes

FIG. 3. Dyspedic muscle membranes do not express detectable levels of Ry 1 R, Ry 2 R, or Ry 3 R proteins by Western blot analysis.
Constituent proteins from dyspedic and wild-type muscle membranes were resolved by SDS-PAGE using 3-10% gradient gels and transferred overnight onto PVDF membranes. Secondary antibody was visualized using enhanced chemiluminescent (ECL, Amersham) methods. Lane markers are as follows: jsr, 1 g of rabbit skeletal SR; avi, 15 g of avian pectoralis SR; memb, 15 g of whole muscle homogenate from either dyspedic (dys) or wild-type (w-t) muscle; 38 -45%, 15 g of purified membranes from corresponding to pooled sucrose gradient fractions from either dyspedic or wild-type muscle; 45%, 15 g of particulate fraction from bottom of 45% sucrose layer from either dyspedic or wild-type muscle. The multiple bands labeled in the 38 -45% w-t lane are proteolytic fragments generated during the purification process. (15). Based on these lines of evidence, the Ca 2ϩ fluxes seen in dyspedic muscle cells have been attributed to up-regulation of Ry 3 R protein as a consequence of Ry 1 R deletion (15). In support of this hypothesis, Conti et al. (12) recently described a differential distribution of the Ry 3 R gene product in various adult mammalian skeletal muscles using Western blot analysis and in situ hybridization. In that study, bovine hind limb, Type II rat skeletal muscle, and diaphragm from the mouse, rabbit, and cow were all found to differentially express Ry 3 R protein.
Their results indicate that expression of Ry 3 R protein differs dramatically among muscle types within a single species and suggest that these differences may also be species-specific. Additionally, Ry 3 R protein expression was found to be concentrated to the terminal cisternae of SR in bovine diaphragm muscle. However, the significance of Ry 3 R expression in skeletal muscle remains unclear. In the Ry 3 -null mouse, the protein does not appear to be essential for E-C coupling or normal muscle development (12), but the maximum Ca 2ϩ -induced Ca 2ϩ release response obtained at high Ca 2ϩ in permeabilized muscle fibers is reduced by ϳ25% (12).

DHPR Protein Expression Is Diminished in Dyspedic Mouse Skeletal Muscle-The functional importance of DHP receptors
in E-C coupling has been amply demonstrated in studies of the dysgenic mouse, which has been shown to lack expression of DHPR␣1 subunit protein and DHPR function (29). Like dysgenic muscle (30), dyspedic muscle lack tetrads found in normal skeletal muscle. When cultured dysgenic myotubes are transfected with cDNA coding for the DHPR␣1 subunit, E-C coupling is restored and the DHP receptors align as tetrads revealing the same structures as seen in wild-type muscle (31). Electron microscopic studies of dyspedic and dysgenic skeletal muscle has revealed that neither Ry 1 R nor DHPR␣1 expression is the signal required for close apposition of the t-tubule and SR membranes and the subsequent co-localization of the remaining triadic cytoskeletal elements (3). Interestingly, while disruption of Ry 1 R gene expression does not preclude the close proximity of the triadic membranes, it does prevent DHPR tetrad formation (3).
Recently, Nakai et al. (14) reported that whole cell L-type Ca 2ϩ currents in cultured dyspedic myotubes are reduced ϳ30fold when compared to wild-type controls. Micronuclear injection of Ry 1 R cDNA into dyspedic myotubes reconstituted L-type inward Ca 2ϩ current to ϳ40% of control myotubes and restored E-C coupling with no concomitant change in immobilizationresistant charge movement (Q max ). Their results indicate that E-C coupling is restored in dyspedic myotubes upon expression of Ry 1 R, and that Ry 1 R expression does not significantly alter the level of expression and/or targeting of DHPR ␣1 subunit protein to the surface membrane, but rather is important in conveying reciprocal regulation for DHPR Ca 2ϩ channel function. However, since Q max is normalized to membrane surface area, the absolute change in the amount of DHPR expression in dyspedic muscle remains unclear.
To more directly ascertain levels of DHPR expression in dyspedic skeletal muscle, the specific binding of [ 3 H]PN200 (0.06 -5.0 nM) is compared in whole membrane preparations from dyspedic and wild-type skeletal muscle. Fig. 5A shows that compared to those from wild-type muscle, the density of specific [ 3 H]PN200-binding sites is reduced ϳ50% in dyspedic muscle preparations without a significant change in K d (Fig.  5B). Three independent experiments using preparations from different animals reveal that the density of [ 3 H]PN200-binding sites is consistently reduced in the dyspedic preparations compared to wild-type (168 Ϯ 7 fmol/mg of protein and 331 Ϯ 29 fmol/mg of protein, respectively). Western blot analysis with a DHPR␣1-selective antibody (32) reveals that while both dys-pedic and wild-type muscle preparations exhibit immunoreactivity at ϳ170 kDa, dyspedic preparations stain to a lesser degree compared to wild-type lanes containing equal amounts of whole membrane-bound protein (Fig. 6, DHPR␣1 blot). Consistent with the [ 3 H]PN200 binding data shown in Fig. 5, densitometric analysis of ECL radiograms reveals that the immunoreactive band at ϳ170 kDa corresponding to DHPR␣1 is approximately 50% less dense in the dyspedic preparations, regardless of the amount of protein loaded on the gel (data not shown).
The observation of reduced DHPR␣1 expression in dyspedic skeletal muscle compared to age-matched wild-type muscle could reflect either 1) a lower density of DHP receptors within the t-tubule membrane, or 2) a decrease in t-tubule surface area, both of which would lower the total DHPR capacity of dyspedic membranes. It is unlikely that DHP receptor density is reduced in t-tubule membranes, since Q max is normalized to membrane surface area and Nakai et al. reported no significant difference in Q max between dyspedic and wild-type myotubes (14). It is more likely that the lower density of [ 3 H]PN200- binding sites can be accounted for by a decrease in t-tubule membrane surface area, since t-tubules are significantly less extensive in dyspedic muscle compared to wild-type muscle. 3 In the report by Nakai et al. (14), L-type Ca 2ϩ currents in cultured dyspedic myotubes were restored to ϳ40% of those seen in cultured wild-type myotubes by microinjection of Ry 1 cDNA. This increase could be attributed to 1) an increase in t-tubule membrane surface area (since DHP receptor density is unaltered by Ry 1 expression), or 2) to the establishment of a requisite communication between Ry 1 R and DHPR. The present results show that DHPR␣1 protein levels are reduced ϳ2fold in dyspedic muscle compared to wild-type muscle. This difference cannot account for the dramatic 30-fold reduction in L-type Ca 2ϩ current seen between wild-type and dyspedic myotubes, and as stated above, is more likely the result of a less developed t-system. This result indicates that deletion of Ry 1 R protein expression eliminates some form of reciprocal (retrograde) regulation on the DHPR complex, and therefore, it is unlikely that the increase in L-type Ca 2ϩ current seen in cultured dyspedic myotubes in which Ry 1 expression has been restored is due to an increase in membrane surface area. This interpretation is supported by the observations that 1) the L-type Ca 2ϩ current in dyspedic myotubes prior to microinjection is much less than 50% of wild-type myotubes, and 2) under ideal conditions (100% expression of Ry 1 R in dyspedic myotubes) the L-type Ca 2ϩ current would not be expected to return to 100% of that seen in wild-type myotubes, since dyspedic myotubes contain significantly less t-tubule membrane.

Triadic Proteins Critical for E-C Coupling Are Expressed in
Dyspedic Mouse Skeletal Muscle-Whether or not Ry 1 R deletion alters the expression of other key triadic proteins normally found in skeletal muscle has not been addressed. In addition to DHPR␣1, several proteins localized at the muscle triad have been shown to directly or indirectly modulate the function of Ry 1 R. Therefore, the presence of FKBP-12, triadin, calsequestrin, SERCA1, and myosin heavy chain in dyspedic mouse skeletal muscle membranes was examined by Western blot analysis.
A high affinity interaction between Ry 1 R and FKBP-12 (12-kDa FK506-binding protein) has been shown to be essential for stabilizing the full conductance gating behavior of the SR channel (33)(34)(35). Blots stained with a monoclonal antibody directed against FKBP-12 reveal the presence of this protein (M r ϳ14,000) in both dyspedic and wild-type muscle membranes (Fig. 6, FKBP-12 blot, dys and w-t lanes). Interestingly, the amount of FKBP-12 associated with dyspedic membrane preparations was consistently observed to be greater than the corresponding wild-type preparations. The underlying reason for this increase in FKBP-12 in dyspedic membranes is unknown. Whether FKBP-12 is elevated in dyspedic muscle membranes as a direct result of loss of muscle function or is associated with infiltration of immunocompetent cells or erythrocytes (36) remains to be determined. However, the latter possibility is unlikely since microscopic examination of muscle does not reveal evidence of an inflammatory response or increased numbers of red cells in dyspedic muscle. In agreement with previous reports (33,37), purified rabbit junctional SR preparations contain FKBP-12 (Fig. 6, FKBP-12 blot, jsr lane). 3 C. Franzini-Armstrong, personal communication.
Triadin, a 95-kDa protein initially identified by Caswell et al. (38) and cloned by Campbell and co-workers (39), appears to form an association with Ry 1 R and calsequestrin (40 -42). Blots probed with a monoclonal antibody directed toward triadin (39) and visualized by ECL reveal the presence of this protein in both dyspedic and wild-type skeletal muscle (Fig. 6, triadin blot, dys and w-t lanes, respectively), as well as in rabbit junctional SR (Fig. 6, triadin blot, jsr lane). The low density of labeling seen in the dys and w-t lanes, compared to the junctional SR lane, reflects the lower density of this protein found in the whole muscle preparation.
In addition to its role in enhancing the Ca 2ϩ loading capacity of SR, calsequestrin appears to elicit a signal that is communicated to Ry 1 R during Ca 2ϩ release (43). The presence of calsequestrin in dyspedic and wild-type skeletal muscle was probed using Stains All as described by Campbell et al. (44). Proteins were resolved on 3-10% Laemmli gels and stained for 48 h with Stains All, followed by de-staining for 1-2 h to resolve an intensely blue band at M r ϳ60,000. The calsequestrin gel shown in Fig. 6 reveals the presence of a blue band corresponding to the location of calsequestrin in lanes containing either rabbit junctional SR (jsr), dyspedic (dys), or wild-type (w-t) muscle proteins. Note that the lane containing rabbit junctional SR stains a comparatively broad band, reflecting the higher density of this protein found in the purified preparation.
The presence of two additional proteins critical for normal muscle function was probed by Western blot analysis. The SERCA1 and myosin blots in Fig. 6 show that antibodies selective for SERCA1 (2) and myosin heavy chain recognize their respective targets in both wild-type and dyspedic membrane preparations. Since the relative density of protein labeling by these antibodies reflects differences in protein content within these preparations, SERCA1 stains to a much higher degree in purified junctional SR preparations (SERCA1 blot, jsr lane) than in the whole muscle preparations (SERCA1 blot, dys and w-t lanes). In comparison, antibody specific for myosin heavy chain reveals a higher density of myosin in the dyspedic and wild-type preparations (myosin blot, dys and w-t lanes, respectively) since these are whole membrane preparations.
Results presented here using Western blot analysis reveal that FKBP-12, triadin, and calsequestrin are all expressed in dyspedic mouse skeletal muscle preparations. Interestingly, while FKBP-12 expression is increased and DHPR expression is decreased in dyspedic mouse skeletal muscle as compared to wild-type membranes, the pattern of expression of key triadic proteins remains and suggests that the molecular components required to form a functional triadic complex will also be present in myogenic cell lines produced using the Ry 1 R gene targeting approach.

Dyspedic Mouse Skeletal Muscle Microsomes Do Not Exhibit Ryanodine-induced Ca 2ϩ Release-[ 3 H]Ryanodine binding and
Western blot analysis reveal a lack of ryanodine receptor expression in dyspedic muscle. As a functional correlate, Ca 2ϩ flux measurements were performed and ryanodine-induced Ca 2ϩ release was assayed using microsomal membranes from dyspedic and wild-type muscle. Ca 2ϩ transport across isolated microsomal membranes was assessed fluorometrically with the dye fluo-3. In the presence of 7.5 mM sodium pyrophosphate in the buffer, Ca 2ϩ additions of 1-8 M produced linear responses from the dye (Fig. 7A). Addition of MgATP in the presence of a regenerating system initiates active accumulation of Ca 2ϩ into membrane vesicles (Fig. 7, B and C, insets). Under the experimental conditions used, wild-type and dyspedic membranes could be loaded with similar amounts of Ca 2ϩ (100 and 74 nmol/mg of protein, respectively). Addition of 20 M ryanodine to Ca 2ϩ -loaded microsomes from wild-type muscle results in release of approximately 40% of the intravesicular Ca 2ϩ , and this effect is fully blocked by pretreatment with 10 M ruthenium red (Fig. 7B, traces a and b, respectively). Ryanodineinduced Ca 2ϩ release from control microsomes can also be FIG. 7. Ryanodine stimulates Ca 2؉ release from wild-type control, but not dyspedic mouse microsomal preparations. Ca 2ϩ release was measured from actively loaded dyspedic mouse or wild-type mouse skeletal muscle microsomes using the fluorescent dye fluo-3 as described under "Experimental Procedures." After loading, Ca 2ϩ release was initiated by the addition of either 20 or 200 M ryanodine. A, Ca 2ϩ aliquots (1 M) were added to the complete transport buffer in the presence of 50 g of wild-type membranes and 4-bromo-A23187 to verify the linearity of the fluo-3 dye response over the ranges used. blocked by prior addition of 1 M neomycin (data not shown). Addition of 200 M ryanodine to wild-type vesicles elicited a biphasic effect on Ca 2ϩ transport, initially inducing Ca 2ϩ release followed by Ca 2ϩ reaccumulation (Fig. 7B, trace c). This result is in agreement with results commonly obtained with rabbit junctional SR and reflects a sequential action of ryanodine at its binding sites. In contrast, membrane microsomes obtained from dyspedic muscle do not release Ca 2ϩ in response to 200 M ryanodine (Fig. 7C). The absence of a transient response to 200 M ryanodine in the dyspedic preparation is further evidence that these muscle microsomes do not contain measurable ryanodine-sensitive Ca 2ϩ effluxes.
The results presented above indicate that while dyspedic mouse skeletal muscle does not express Ry 1 R, it does express the balance of the major triadic elements critical for E-C coupling. These findings, along with the Ca 2ϩ flux measurements presented above, indicate that the mouse model is an ideal system with which to examine Ca 2ϩ regulation in skeletal muscle using a homologous expression system and transgenic approaches.