(Received for publication, July 23, 1996, and in revised form, November 18, 1996)
From the Department of Molecular Biosciences, School
of Veterinary Medicine, University of California, Davis, California
95616, the § Department of Cardiology, Children's Hospital,
Boston, Massachusetts 02115, and ** Department of Anesthesia,
Brigham and Women's Hospital, Boston, Massachusetts 02115
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,
DHPR1, 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 (ry153/ry153) for ryanodine receptor type 1 (skeletal isoform of ryanodine receptor; Ry1R)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 Ry1R expression results in altered expression of other key triadic proteins involved in modulating SR Ca2+ transport.
In addition to Ry1R, separate genes encode two other ryanodine receptors, namely the cardiac (Ry2R) and "brain" (Ry3R) isoforms (4-8). Ry1R is predominantly expressed in skeletal muscle and cerebellar Purkinje cells (9, 10). Ry2R is predominately expressed in cardiac tissue, where it functions as a Ca2+-induced Ca2+ release channel, but is also widely expressed in brain tissues (7, 9). In comparison, Ry3R is widely expressed at low levels in many non-muscle cells, in mammalian brain, and in very low levels in heart and skeletal muscle (11-13). However, its functional significance is presently unclear.
The development of Ry1R-null (dyspedic) mice (1)2 represents a significant advance in the goal to understand the molecular mechanisms regulating the function of Ry1R. 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 Ca2+ entry current than control fibers, despite the presence of comparable levels of immobilization-resistant charge movement. Significantly, both voltage-dependent Ca2+ entry current and SR Ca2+ release are concomitantly restored in cultured dyspedic myotubes by micronuclear injection of Ry1R cDNA (14). Despite the loss of E-C coupling, dyspedic muscle fibers in culture exhibit small Ca2+ 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 Ry3R mRNA (15). However, there has been no direct evidence of Ry3R protein expression in dyspedic muscle.
The present study demonstrates for the first time that dyspedic muscle
1) exhibits a 2-fold reduction in [3H]PN200 binding
capacity with a concomitant change in immunoreactive DHPR1 subunit
expression, 2) lacks detectable levels of specific high or low affinity
[3H]ryanodine-binding sites, 3) lacks immunoreactivity
toward Ry1R, Ry2R, or Ry3R
antibody, and 4) lacks a ryanodine-sensitive Ca2+ 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.
Dyspedic mice were obtained as described elsewhere.2 Briefly, a 9-kilobase EcoRI-Tth111I genomic fragment which contains part of the skeletal Ry1 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 × 107 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 Ry1 targeted allele were obtained from heterozygous matings and verified by PCR and Southern blot.
Tissue PreparationCrude and sucrose gradient-purified
membrane fractions were prepared from wild-type
(ry1+/+) and dyspedic
(ry1-/-) 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 µM phenylmethylsulfonyl fluoride (prepared
immediately before use), and 10 µg/ml leupeptin. Bone fragments were
removed by gentle centrifugation at 50 × g for 2 min
at 4 °C. The homogenate was centrifuged at 110,000 × 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 80 °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 × 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 × 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 80 °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 × g centrifugation step was replaced by a 30-min, 8000 × g centrifugation step. Protein was quantitated in all preparations by the method of Lowry (19) using bovine serum albumin as a standard.
Radioligand Binding AssayHigh and low affinity binding of [3H]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 µM Ca2+, 1 nM [3H]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 µM CaCl2, 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. Kd and Bmax values were derived from Scatchard analysis of the binding data.
Specific binding of 0.06-5.0 nM [3H]PN200
(83 Ci/mmol, DuPont NEN) to the 1 subunit of the L-type
Ca2+ 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 µM
nifedipine. Separation of bound and free ligand was performed as
described for high affinity [3H]ryanodine binding except
that the filters were washed with 5 × 2 ml of wash buffer.
Kd and Bmax values were
derived from Scatchard analysis of the binding data.
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 anti-mouse (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: Ry1 (34C), generous gift of
Dr. J. Sutko; Ry2 (C3-33), generous gift of Dr. G. Meissner; Ry3, generous gift of Dr. V. Sorrentino; DHPR1 (MA3-920), Affinity BioReagents; triadin (11G12A8), generous gift of
Dr. K. Campbell; FKBP-12, generous gift of Dr. M. Harding, Vertex
Pharmaceuticals; SERCA1 (MA3-911), Affinity BioReagents; myosin heavy
chain, Affinity BioReagents (1130-P; Biocytex Biochemicals, San Diego,
CA).
Ca2+ flux measurements using microsomal membranes from wild-type or dyspedic mouse skeletal muscle were performed fluorometrically (SLM AB-2, SLM-Aminco). Briefly, 50 µg of skeletal muscle microsomal membranes were equilibrated to 37 °C in a buffer consisting of 92 mM KCl, 7.5 mM Na4P2O7, 20 mM MOPS, pH 7.0, 0.01% NaN3, 2-10 µM Ca2+, and 0.5 µM fluo-3. A coupled enzyme system (20 µg/ml creatine phosphokinase and 5 mM phosphocreatine) was present to maintain ATP concentrations. Loading of Ca2+ was initiated by the addition of 1 mM MgATP. Transport of Ca2+ 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 Na4P2O7 in the transport buffer maintained a linear dye response at added Ca2+ concentrations up to 8 µM. Ca2+ efflux was initiated by the addition of 20 or 200 µM ryanodine. Ca2+ accumulation by microsomal stores was verified by addition of 2 µg/ml 4-bromo-A23187. Linearity of fluo-3 emission with increasing Ca2+ concentration was verified after each experiment by adding known aliquots from a National Bureau of Standards Ca2+ stock.
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 Ry1R protomer found in junctional rabbit
skeletal muscle SR (jsr lane, marked with
arrow).
The Ry1R, Ry2R, and Ry3R Isoforms Are Not Detectable in Dyspedic Mouse Skeletal Muscle
The
equilibrium binding of [3H]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 [3H]ryanodine-binding sites, whereas
preparations of newborn wild-type muscle exhibit 148 ± 16 fmol/mg
of high affinity [3H]ryanodine binding with a
Kd of 6.9 ± 1.3 nM (mean
Bmax 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 (Kd = 20.3 nM) and
Bmax of 305 ± 94 fmol/mg of protein. Experiments designed to measure the presence of binding sites having
lower affinity were assessed using [3H]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. Ry3R) 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 Ry1, Ry2, or Ry3 receptors. Specific antibody labeling was visualized using a highly sensitive chemiluminescent (ECL) technique.
Blots probed with a Ry1R-selective mouse monoclonal
antibody (22) stain positive for the ~560-kDa protomer in a protein concentration-dependent manner with preparations from
wild-type 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 Ry1R antibody
(Fig. 3, top, dys lanes), as does rat cardiac SR
(Fig. 3, top, crd lane).
As expected, Western blot analysis using a Ry2R-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 Ry2R protein found in rat cardiac preparations, which is included as a positive control (Fig. 3, middle, crd lane).
Western blot analysis using a Ry3R-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
>80% sequence homology with the mammalian Ry3R "brain" isoform (Fig. 3, bottom, avi lanes).
The antibody also recognizes Ry3R protein in rat testicular
tissue (Fig. 3, bottom, tst lane), which has been
shown to express very high levels of Ry3R mRNA (8).
Sequential labeling experiments with Ry1R and Ry3R antibodies reveal that the Ry3R protomer
found in avian muscle and rat testicular tissue migrates with a smaller
apparent size than Ry1R.
Expression of Ry1R and Ry3R 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 Ry1R- and
Ry3R-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
Ry1R-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 Ry3R polyclonal antibody
in the purified fractions from either the wild-type or dyspedic muscle
preparations (Fig. 4, bottom panel). Multiple bands labeled
by the Ry1R-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
Mr estimated to be 250,000 and 300,000, which is reflective of the fragmentation pattern of Ry1R induced by
trypsin (24) or calpain (25) digestion. The sedimentation density of
the vesicles could be related to the degree of proteolysis of
Ry1R, especially since Ry1R 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 (Ry1R) of the ryanodine receptor. In the present study, specific antibodies fail to detect the presence of Ry1R, Ry2R or, more importantly, Ry3R 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 [3H]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 Ry3R protein or expresses it at levels below detection limits.
The question of whether dyspedic muscle expresses an alternate ryanodine receptor isoform (Ry2R or Ry3R) is important, since responses to Ca2+, caffeine, and ryanodine have been observed in cultured dyspedic myotubes (1, 14, 15). However, the total Ca2+ 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 Ry3R mRNA expression in cultured dyspedic myotubes (15). Based on these lines of evidence, the Ca2+ fluxes seen in dyspedic muscle cells have been attributed to up-regulation of Ry3R protein as a consequence of Ry1R deletion (15). In support of this hypothesis, Conti et al. (12) recently described a differential distribution of the Ry3R 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 Ry3R protein. Their results indicate that expression of Ry3R protein differs dramatically among muscle types within a single species and suggest that these differences may also be species-specific. Additionally, Ry3R protein expression was found to be concentrated to the terminal cisternae of SR in bovine diaphragm muscle. However, the significance of Ry3R expression in skeletal muscle remains unclear. In the Ry3-null mouse, the protein does not appear to be essential for E-C coupling or normal muscle development (12), but the maximum Ca2+-induced Ca2+ release response obtained at high Ca2+ in permeabilized muscle fibers is reduced by ~25% (12).
DHPR Protein Expression Is Diminished in Dyspedic Mouse Skeletal MuscleThe 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 DHPR1 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 Ry1R 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 Ry1R 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
Ca2+ currents in cultured dyspedic myotubes are reduced
~30-fold when compared to wild-type controls. Micronuclear injection
of Ry1R cDNA into dyspedic myotubes reconstituted
L-type inward Ca2+ current to ~40% of control myotubes
and restored E-C coupling with no concomitant change in
immobilization-resistant charge movement (Qmax).
Their results indicate that E-C coupling is restored in dyspedic
myotubes upon expression of Ry1R, and that Ry1R
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
Ca2+ channel function. However, since
Qmax 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 [3H]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 [3H]PN200-binding sites is reduced ~50% in dyspedic
muscle preparations without a significant change in
Kd (Fig. 5B). Three independent experiments using preparations from different animals reveal that the
density of [3H]PN200-binding sites is consistently
reduced in the dyspedic preparations compared to wild-type (168 ± 7 fmol/mg of protein and 331 ± 29 fmol/mg of protein,
respectively). Western blot analysis with a DHPR1-selective antibody
(32) reveals that while both dyspedic and wild-type muscle preparations
exhibit immunoreactivity at ~170 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
[3H]PN200 binding data shown in Fig. 5, densitometric
analysis of ECL radiograms reveals that the immunoreactive band at
~170 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 DHPR1 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
Qmax is normalized to membrane surface area and
Nakai et al. reported no significant difference in
Qmax between dyspedic and wild-type myotubes
(14). It is more likely that the lower density of
[3H]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 Ca2+
currents in cultured dyspedic myotubes were restored to ~40% of
those seen in cultured wild-type myotubes by microinjection of
Ry1 cDNA. This increase could be attributed to 1) an
increase in t-tubule membrane surface area (since DHP receptor density
is unaltered by Ry1 expression), or 2) to the establishment
of a requisite communication between Ry1R and DHPR. The
present results show that DHPR1 protein levels are reduced ~2-fold
in dyspedic muscle compared to wild-type muscle. This difference cannot
account for the dramatic 30-fold reduction in L-type Ca2+
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 Ry1R protein expression eliminates some form of reciprocal (retrograde) regulation on the DHPR
complex, and therefore, it is unlikely that the increase in L-type
Ca2+ current seen in cultured dyspedic myotubes in which
Ry1 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 Ca2+ current in dyspedic
myotubes prior to microinjection is much less than 50% of wild-type
myotubes, and 2) under ideal conditions (100% expression of
Ry1R in dyspedic myotubes) the L-type Ca2+
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.
Whether or not Ry1R
deletion alters the expression of other key triadic proteins normally
found in skeletal muscle has not been addressed. In addition to
DHPR1, several proteins localized at the muscle triad have been
shown to directly or indirectly modulate the function of
Ry1R. 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 Ry1R 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 (33-35). Blots stained with a monoclonal antibody directed against FKBP-12 reveal the presence of this protein (Mr ~14,000) 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).
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 Ry1R 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 Ca2+ loading capacity of SR, calsequestrin appears to elicit a signal that is communicated to Ry1R during Ca2+ 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 Mr ~60,000. 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 Ry1R gene targeting approach.
Dyspedic Mouse Skeletal Muscle Microsomes Do Not Exhibit Ryanodine-induced Ca2+ Release[3H]Ryanodine binding and Western blot
analysis reveal a lack of ryanodine receptor expression in dyspedic
muscle. As a functional correlate, Ca2+ flux measurements
were performed and ryanodine-induced Ca2+ release was
assayed using microsomal membranes from dyspedic and wild-type muscle.
Ca2+ 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, Ca2+
additions of 1-8 µM produced linear responses from the
dye (Fig. 7A). Addition of MgATP in the
presence of a regenerating system initiates active accumulation of
Ca2+ 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 Ca2+ (100 and 74 nmol/mg of protein,
respectively). Addition of 20 µM ryanodine to
Ca2+-loaded microsomes from wild-type muscle results in
release of approximately 40% of the intravesicular Ca2+,
and this effect is fully blocked by pretreatment with 10 µM ruthenium red (Fig. 7B, traces a
and b, respectively). Ryanodine-induced Ca2+
release from control microsomes can also be blocked by prior addition
of 1 µM neomycin (data not shown). Addition of 200 µM ryanodine to wild-type vesicles elicited a biphasic
effect on Ca2+ transport, initially inducing
Ca2+ release followed by Ca2+ 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 Ca2+ in response to 200 µM ryanodine (Fig.
7C). The absence of a transient response to 200 µM ryanodine in the dyspedic preparation is further evidence that these muscle microsomes do not contain measurable ryanodine-sensitive Ca2+ effluxes.
The results presented above indicate that while dyspedic mouse skeletal muscle does not express Ry1R, it does express the balance of the major triadic elements critical for E-C coupling. These findings, along with the Ca2+ flux measurements presented above, indicate that the mouse model is an ideal system with which to examine Ca2+ regulation in skeletal muscle using a homologous expression system and transgenic approaches.
We thank Lili Chen for technical assistance.