| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
From the Montreal Neurological Institute and Department of Neurology and Neurosurgery* and the Respiratory Division, McGill University Health Center and Meakins-Christie Laboratories, McGill University, Montreal, Quebec, Canada; Genzentrum, Munich, Germany; and the Department of Neurology and Friedrich-Baur-Institute, Ludwig-Maximilians-University, Munich, Germany
 |
Abstract |
|---|
 Top Abstract Materials and MethodsResultsDiscussionReferences |
|---|
-sarcoglycan13
can also cause skeletal myopathy. As part of our studies of gene therapy for inherited muscular dystrophies, we have studied the expression in skeletal muscle of the Coxsackie and adenovirus receptor (CAR),14,15 a highly conserved protein that belongs to the immunoglobulin superfamily. We have previously shown that CAR is expressed in myoblasts and is diffusely distributed on the plasma membrane of immature and regenerating myofibers.16 CAR expression is, however, down-regulated during skeletal muscle maturation, and in adult muscle, CAR is normally restricted to the sarcolemma at the neuromuscular junction.17 The decrease in CAR levels parallels the decreased transducibility of mature versus immature skeletal muscle by adenoviral vectors of subgroup C.16 Therefore, to test whether increased CAR expression in mature muscle would overcome the observed low efficiency of adenoviral transduction, we generated transgenic mice in which CAR is regulated by the muscle creatine kinase (MCK) promoter.18 The level of extrasynaptic CAR expression obtained in adult mice hemizygous for the CAR transgene was sufficient to induce increased susceptibility to transduction by adenoviral vectors18 and did not cause any overt skeletal muscle pathology. However, when CAR transgenic mice were bred to homozygosity, they developed an early, severe myopathy that was associated with an up-regulation of caveolin-3 levels and simultaneous deficiencies in dystrophin and dysferlin. Although murine models of caveolin overexpression12 and of primary dystrophin deficiency19 or dysferlin deficiency20 exist, they each have relatively mild phenotypes. Until this study there had been no report of a pathological condition exhibiting conspicuous simultaneous alterations in these key muscle proteins. Physiological dysfunction related to excess CAR and/or simultaneous deficiency of dystrophin and dysferlin therefore constitute a previously unrecognized scenario for the production of an exceptionally severe myopathy of skeletal muscle.
|
Materials and Methods |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
The transgenic mice that express full-length murine CAR (mCAR1 isoform) under the control of the 1.35-kb MCK promoter were described previously.18 The CAR transgenics were maintained on a C3H or C57BL/6 background. The colony was housed in a specific pathogen-free facility. The skeletal muscle phenotype was similar in both backgrounds. Homozygosity was determined by Southern blot of tail genomic DNA and quantitation of normalized signal from the CAR transgene by PhosphorImage analysis (Molecular Dynamics, Sunnyvale, CA). All animal experimentation was approved by the Institutional Animal Care Committee and conformed to the guidelines of the Canadian Council of Animal Care.
Microscopic Analysis of Muscle
Hindlimb muscles (gastrocnemius, tibialis anterior, quadriceps, and soleus) were dissected, mounted using tragacanth gum, and frozen in liquid nitrogen-cooled isopentane. Fresh frozen muscles were sectioned at 4 µm, applied to slides, and stored at –80°C until use. Standard histological and histochemical staining techniques were performed for the following reactions: hematoxylin and eosin (H&E), modified trichrome, acetylcholine esterase, acid phosphatase, myofibrillar ATPase (preincubation at pH 9.4 and 4.6), NADH tetrazolium reductase, and periodic acid-Schiff.
Assessment of Sarcolemmal Damage
Evans blue dye, a vital dye that is unable to penetrate the sarcolemma of normal muscle fibers, has been used to evaluate sarcolemmal integrity in mouse models of muscular dystrophy.21 Accordingly, we used Evans blue dye penetration into the muscle fiber cytoplasm as an index of sarcolemmal disruption resulting from mechanical stress. Evans blue dye [5 µg/µl solution in phosphate-buffered saline (PBS); 5 µl/g body wt] was injected into the jugular vein of anesthetized CAR homozygote (n = 3) and wild-type mice (n = 3). To allow sufficient time for the circulating Evans blue dye to enter damaged membranes at reproducible levels, the mice were euthanized at least 3 hours later. At the end of each experiment, hindlimb muscles (gastrocnemius, tibialis anterior, quadriceps, and soleus) from wild-type and homozygous mice were dissected, mounted using tragacanth gum, and frozen in liquid nitrogen-cooled isopentane. Transverse cryosections of frozen tibialis anterior, quadriceps, and gastrocnemius muscles were fixed in methanol and visualized by fluorescence microscopy under green light. Slides were analyzed on a Leica microscope-based imaging system using OpenLab imaging software (Quorum Technologies, St. Catharines, ON, Canada).
Antibodies
The production, purification, and characterization of the polyclonal antibody (Ab 2240) against CAR N-terminal extracellular domain has been described in detail previously.16
The rabbit polyclonal antibody raised against the C-terminal intracellular domain of the human 46-kd CAR isoform (RP291) that cross-reacts with the mouse homolog mCAR1 was a kind gift of Dr. Kerstin Sollerbrant (Ludwig Institute for Cancer Research, Stockholm Branch, Karolinska Institutet, Stockholm, Sweden) and was produced as described.22
The polyclonal anti-dystrophin antibody was described previously.23
All other antibodies were purchased commercially: anti-desmin, anti-
-sarcoglycan, and anti-ß-dystroglycan were from NovoCastra (Vector Laboratories, Burlington, ON, Canada); anti-dysferlin was from Vector Laboratories; anti-caveolin-3 and anti-neuronal nitric-oxide synthase (anti-nNOS) were from Transduction Laboratories (BD Biosciences, Mississauga, ON, Canada).
Immunolocalization
Hindlimb muscles (gastrocnemius, tibialis anterior, quadriceps, and soleus) from wild-type hemizygous and homozygous mice were dissected, mounted using tragacanth gum, and frozen in liquid nitrogen-cooled isopentane. Fresh frozen muscles were sectioned at 4 µm, applied to slides, and stored at –80°C until use. Sections from wild-type and transgenic mice were placed on the same slide to ensure identical experimental conditions. Slides were thawed at room temperature, washed in PBS, and blocked in 4% bovine serum albumin/5% goat serum (and Fab2 fragment when required) for 20 minutes. Sections were incubated for 1 hour with primary antibody at 23°C, PBS-washed, blocked, and incubated for an additional hour with Biotin-SP-conjugated mouse anti-rabbit IgG or goat anti-mouse IgG (Vector Laboratories) in blocking solution. For indirect immunofluorescence, sections were then PBS-washed, blocked, and incubated with Cy-2- or Cy-3-conjugated streptavidin (Jackson ImmunoResearch Laboratories, West Grove, PA) for 15 minutes followed by an overnight PBS wash at 4°C before mounting with Immu-mount (ThermoShandon, Pittsburgh, PA). For immunohistochemical studies at the level of light microscopy, sections were then PBS-washed, blocked, and incubated with a mixture of avidin DH and biotinylated-horseradish peroxidase for 20 minutes using the Vectastain Elite ABC kit according to the manufacturer’s instructions (Vector Laboratories). Sections were then PBS-washed, blocked, incubated with the 3,3'-diaminobenzidine substrate system for 5 minutes, and quickly rinsed in PBS followed by multiple ethanol washes before being cleared with xylene and mounted with Permount (Fisher Scientific, Pittsburgh, PA). Antibodies were used at the following dilutions: polyclonal antibodies RP291 at 1:100, 2240 at 1:40; monoclonal antibodies dystrophin and dysferlin at 1:40; desmin at 1:60, ß-dystroglycan at 1:30, -sarcoglycan at 1:40, caveolin-3 at 1:40, and nNOS at 1:100. Sections incubated in the absence of primary or secondary antibodies were used as immunostaining controls. Neuromuscular junctions were revealed with Alexa 488-conjugated -bungarotoxin using a 1:50 dilution (Molecular Probes, Eugene, OR). Slides were analyzed on a Leica microscope-based imaging system using OpenLab imaging software (Quorum Technologies) or Zeiss microscope-based imaging system using Remote Capture imaging software (Canon Inc., Lake Success, NY). Immunostained sections from wild-type and transgenic mice were photographed at the same exposure.
Tissue Extract Preparation and Immunoblotting
Hindlimb muscles from wild-type control, CAR+/–, and CAR+/+ mice were minced with razor blades and homogenized in lysis buffer [50 mmol/L Tris-HCl, pH 6.8, 2% sodium dodecyl sulfate (SDS), 10% glycerol, and protease inhibitors (ethylenediaminetetraacetic acid-free cocktail tablet) (Roche Diagnostics, Laval, QC, Canada)]. The protein concentration was determined using the Bio-Rad protein assay (Bio-Rad, Mississauga, ON, Canada). Samples were standardized to 1 µg/µl, and equal concentrations (15 µg per lane) were electrophoresed on an SDS-polyacrylamide gel of appropriate percentage, depending on protein size to be examined, and electrotransferred for 1 to 2 hours at 100 V to nitrocellulose membrane (Amersham Biosciences, Oakville, ON, Canada). Membranes were blocked with 4% nonfat dry milk, 0.9% NaCl, 50 mmol/L Tris-HCl, pH 7.5, and 0.1% Tween 20 for 1 hour and incubated with primary antibody in blocking buffer for 2 to 3 hours at room temperature, washed, and probed with horseradish peroxidase-conjugated goat anti-mouse or goat anti-rabbit antisera (Kirkegaard & Perry Laboratories, Gaithersburg, MD) at a 1/4000 dilution. Blots were developed using the ECL chemiluminescence system (Amersham Biosciences). Antibodies were used at the following dilutions: anti-CAR (ab2240) and dystrophin at 1:500, dysferlin (Vector Laboratories) at 1:250, ß-dystroglycan (NovoCastra, Vector Laboratories) at 1:50, desmin (NovoCastra) at 1:50, caveolin-3 (Transduction Laboratories) at 1:1000, and nNOS (Transduction Laboratories) at 1:1000.
Northern Blot Analysis
Analysis of total RNA from CAR homozygous (n = 3) and wild-type (n = 2) mice was conducted by transfer to nylon membranes after electrophoretic separation through a 6.6% formaldehyde agarose gel according to previously published procedures.16 The probes consisted of the following: a cDNA probe Cf56a for dystrophin,24 a cDNA probe for human dysferlin corresponding to nucleotides 2154 to 2903 (GenBank accession number NM_003494), and a cDNA probe for caveolin-3 corresponding to nucleotides 1 to 1214 (GenBank accession number BC024383). Hybridizations were performed in a solution consisting of 50% formamide, 7% SDS, 0.25 mol/L NaHPO4, 0.25 mol/L NaCl, and 1 mmol/L ethylenediaminetetraacetic acid with ZetaProbe nylon membrane (Bio-Rad Laboratories, Hercules, CA) buffers for 18 to 24 hours at 44°C. Membranes were rinsed three times for 15 minutes in 0.1x standard saline citrate and 0.1% SDS at 44°C followed by a fourth rinse for 1 hour in the same solution at 65°C. Equal loading of samples was verified by hybridization with a cDNA probe for the housekeeping gene glyceraldehyde-3-phosphate dehydrogenase.25
Activation of MAPK during Muscle Contraction
The phosphorylation level of MAPK was determined after eccentric contractions performed on tibialis anterior of CAR hemizygous and wild-type littermate mice, based on the protocols of Ryder and colleagues26 and Wretman and colleagues.27 Each contraction involved supramaximal stimulation at 120 Hz for a total of 300 ms; the muscle was held at Lo (the length at which maximal twitch force is achieved) during the initial 100 ms (isometric component) and then lengthened through a distance of 25% of Lo during the last 200 ms (eccentric component). Peak muscle length was maintained for an additional 100 ms after cessation of the stimulation, followed by a return to Lo during the next 100 ms. A total of 30 such contractions were imposed on the muscle, each being separated by a 30-second recovery period. Muscles were then frozen immediately in liquid nitrogen, ground, and homogenized in lysis buffer [50 mmol/L Tris-HCl, pH 6.8, 2% SDS, 10% glycerol, 25 mmol/L NaF, 2 mmol/L sodium orthovanadate, 40 mmol/L ß-glycerophosphate, and protease inhibitors (ethylenediaminetetraacetic acid-free cocktail tablet; Roche Diagnostics)]. Muscle homogenates containing 10 µg of proteins were separated by 12% SDS-polyacrylamide gel electrophoresis and transferred to nitrocellulose membrane. Membrane was first incubated with a monoclonal phospho-p44/42 antibody (Cell Signaling Technology, Mississauga, ON, Canada) in 7.5% bovine serum albumin and visualized by enhanced chemiluminescence. The membrane was then stripped (0.2 mol/L glycine, pH 2.8) and reincubated using the polyclonal p-44/42 antibody (Cell Signaling Technology). Bands were quantitated using the GeneGnome Imaging system (Syngene, Frederick, MD).
In Vivo Measurement of Muscle Contractility Parameters
Isometric force determination was performed as described previously28 on CAR+/– hemizygotes and wild-type littermates at 4 months of age. Mice were anesthetized with ketamine (130 mg/kg) and xylazine (20 mg/kg) and immobilized in the supine position on a surgical platform. The skin of the anterior region of the lower hind limb was opened from the ankle to just above the knee to expose the tibialis anterior fully. Two 27.5-gauge needles were used to secure the knee and ankle to the platform. The distal tendon of the tibialis anterior was isolated and tied with 4-0 nylon suture to the lever arm of a force transducer/length servomotor system (model 305B dual mode; Cambridge Technology, Watertown, MA), which was mounted on a mobile micrometer stage to allow fine incremental adjustments of muscle length. Exposed portions of the tibialis anterior were kept moist with a 37°C isotonic saline drip, and the tibialis anterior was stimulated directly via an electrode placed on the belly of the muscle. Supramaximal stimuli with a monophasic pulse duration of 2 ms were delivered using a computer-controlled electrical stimulator (model S44; Grass Instruments, Quincy, MA). Muscle force and length signals were displayed on a storage oscilloscope (Tektronix, Beaverton, OR) and simultaneously acquired to a computer (Labdat/Anadat software; RHT-InfoData, Montreal, QC, Canada) via an analogue-to-digital converter at a sampling rate of 1000 Hz. After adjusting the tibialis anterior to optimal muscle length (Lo, the length at which maximal twitch force is achieved), two twitch stimulations were performed. The mean value of force produced from these two contractions was considered as maximal isometric twitch force. Isometric tetanic force was then measured by stimulating the muscle at 10, 30, 60, 90, and 120 Hz for 300 ms. Maximal isometric tetanic force was usually achieved at a stimulation frequency of 120 Hz. Muscle cross-sectional area was determined by dividing muscle weight by its length and tissue density (1.06 g/cm3), which then allowed specific isometric force (ie, force normalized to muscle cross-sectional area) to be calculated.
|
Results |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
The CAR homozygous animals manifested symptoms of muscle weakness. When suspended from the tail, CAR+/+ mice (Figure 1B)
invariably kept their hindlimbs adducted, a behavior recapitulated in other murine models of muscle disease such as the dystrophin- and dysferlin-deficient mice.29,30
This is in marked contrast to the reflex of normal mice (Figure 1A)
that extend and outstretch the hindlimbs on suspension.
|
The pattern of localization of CAR in normal mature skeletal muscle fibers is comparable with that of many other molecules that occur exclusively at the neuromuscular junction such as utrophin and NCAM. CAR is uniformly expressed throughout the sarcolemma of normal immature muscle, but within a few weeks of birth, CAR expression is down-regulated16
and becomes confined to the neuromuscular junction.17
In the hemizygous transgenic mice, expression of full-length CAR resulted in widespread, fairly uniform distribution of the protein on the sarcolemmal surface and restored susceptibility to adenoviral transduction to the same extent as, or even higher than, that obtained in neonate skeletal muscle.18
Increased levels of ectopic CAR as seen in the CAR homozygous mice (CAR+/+) (Figure 2)
caused a severe and necrotizing myopathy manifesting as early as 4 weeks after birth and resulting in premature death of these animals by 
6 months of age.
|
CAR homozygotes displayed a striking reduction (50%) in muscle fiber area compared with wild-type controls. The CAR+/+ muscle revealed widespread necrosis, phagocytosis, and active regeneration (Figure 3D)
. Numerous centrally nucleated muscle fibers of differing caliber were seen (Figure 3D)
, as well as fibers with irregular areas of reduced or absent oxidative enzyme activity (data not shown). There were scattered areas occupied by atrophic polygonal fibers and small clusters of adipocytes, implying muscle fiber loss (Figure 3D)
. There were also groups of fibers of the same histochemical type, with complimentary contours, in the CAR homozygous muscle (Figure 3E)
compared with the flawless checkerboard pattern of wild-type littermates (Figure 3B)
. Although sparse CD4-positive mononuclear inflammatory cells were seen in the endomysial space and macrophages were prevalent in the vicinity of necrotic fibers, a cellular response such as that observed in inflammatory myopathies was absent in the CAR+/+ animals (data not shown). To assess the integrity of the myofibers, CAR homozygotes and wild-type control animals were subjected to Evans Blue dye uptake as described in the Materials and Methods section. Intracellular accumulation of this vital dye, a hallmark of disrupted plasma/sarcolemmal integrity, was visible in CAR+/+ (Figure 3F)
but not in wild-type controls (Figure 3C)
. In normal human and murine nonregenerating muscle, CAR expression is confined to the neuromuscular junction.17,31
No abnormalities in gross morphological structure of the neuromuscular junction could be detected by either -bungarotoxin binding (Figure 4, C and D)
or acetylcholinesterase staining of CAR+/+ endplate regions (Figure 4, E and F)
compared with normal controls (intramuscular nerves of the CAR homozygote were intact and appeared normal).
|
|
To understand how ectopic expression of CAR, a cell surface protein, results in the severe myopathological condition displayed by the homozygous animals, we performed histochemical and biochemical analyses of CAR and key proteins associated with diseases involving plasmalemmal defects, namely dystrophin, dysferlin, caveolin-3, and representative members of the dystrophin-glycoprotein complex (Figures 5 and 6)
. To assess initially the expression levels and pattern of localization of selected skeletal plasmalemmal proteins, cryostat sections of CAR homozygous and wild-type muscle (gastrocnemius, quadriceps, tibialis anterior, and soleus) were subjected to immunohistochemistry (Figure 5)
, and the results were further validated by immunofluorescence analysis of tissue sections obtained from additional mice (Figure 6)
. We observed extrasynaptic sarcolemmal as well as cytoplasmic CAR immunoreactivity in greater than 85% of the muscle fibers (Figure 5A)
of the homozygous mice compared with controls (Figure 5B)
. CAR homozygotes displayed a striking reduction in both dystrophin (Figures 5C and 6A)
and -sarcoglycan immunoreactivity (Figure 5E)
compared with controls (Figure 5, D and F
; Figure 6A
). Conversely, levels of caveolin-3, a scaffolding protein and primary component of caveolae, were markedly increased in the CAR+/+ mice (Figures 5G and 6A)
compared with the wild-type littermates (Figures 5H and 6A)
. This increase occurred through transcriptional up-regulation of the caveolin-3 gene in the CAR transgenics because the homozygotes expressed an approximately threefold increase in caveolin-3 transcript as measured by Northern blots (Figure 6B)
and protein levels determined by Western blots (Figure 6C)
.
|
|
. Although both dystrophin and dysferlin transcripts were detected at normal levels by Northern blot analysis (Figure 6B)
, dystrophin and dysferlin protein levels were virtually undetectable in the CAR homozygotes compared with wild-type littermates (Figure 6C)
and CAR hemizygous mice (data not shown). The instability of the dystrophin-glycoprotein complex that occurs in the absence of dystrophin in Duchenne muscular dystrophy and mdx muscle leads to diminished ß-dystroglycan levels. Unexpectedly, ß-dystroglycan levels were not diminished but rather were increased in the CAR homozygotes (Figure 6, A and C)
. Contractile Characteristics
Caveolin-3 is a key regulator of signaling in cardiac and skeletal muscle.36
Muscle contraction and exercise have been shown to activate several members of the mitogen-activated protein kinase (MAPK) family, including the extracellular signal-regulated kinases 1/2 (erk1/2; p42/p44).27,37-41
To assess whether changes in caveolin-3 levels had a functional consequence on the integrity of a signaling pathway known to be regulated by caveolins,42,43
MAPK signaling was studied in the skeletal muscle of hemizygous CAR mice. Hemizygotes were chosen to prevent any confounding effect of overt pathology on the subsequent analysis of MAPK activation. The tibialis anterior muscle of hemizygous CAR mice and their wild-type littermates were subjected to eccentric contractions, homogenates were prepared from the contracted muscles and the phosphorylation status of p42 and p44 was determined by Western blot analysis.27
Although overall p42 and p44 levels were similar in the two groups, CAR hemizygotes demonstrated a blunted response with greatly diminished phosphorylation of p42 and p44 (Figure 7)
.
|
.
|
|
Discussion |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
In the hemizygous CAR mice, the inability to down-regulate ectopically expressed CAR in adult muscle fibers did not lead to overt pathology or affect their lifespan. The myopathy observed in the CAR transgenics was therefore dose-dependent. Only animals expressing CAR at higher levels, as in the homozygous lines, were severely affected, indicating that there is a threshold level above which CAR expression is detrimental to skeletal muscle. This is analogous to the situation observed on overexpression in mice of other normally abundant constituents of the plasmalemma such as caveolin-3 and -sarcoglycan, which also cause skeletal myopathy.12,13
Interestingly, the levels of CAR achieved in the homozygotes are similar to the CAR levels observed at birth in normal rodent brain (K.-C. Huang and J.N., unpublished data), indicating a context dependency for cellular dysfunction.
Molecular and Physiological Consequences of CAR Ectopic Expression
The particular susceptibility of skeletal muscle to alterations in CAR expression is evident even in the hemizygotes, in which subtle but significant physiological changes could be documented (Figures 7 and 8)
. The significant reduction of the normalized maximal twitch and tetanic force in hemizygous CAR+/– animals, in the absence of microscopic pathology of muscle including a normalcy of the intramuscular nerves, suggests either an abnormality of excitation-contraction coupling or a disturbance of the contractile machinery itself. Such changes are considered myopathic rather than neuropathic and point to an underlying pathology in the phenotypically normal hemizygous animals. The CAR homozygotes showed nonspecific myopathic features including necrosis and regeneration. The severity of the myopathy in the CAR homozygotes is probably attributable to a constellation of changes reflected by the conspicuous reduction in dystrophin, dysferlin, and nNOS and the substantial increase of caveolin-3 and ß-dystroglycan. Even though a transcriptional up-regulation of caveolin-3 was detected (Figure 6B)
, posttranscriptional regulation of other proteins (eg, dystrophin, dysferlin) must occur, because their transcript levels were unchanged although their protein levels were substantially decreased (Figure 6B)
. The severe myopathy observed in the CAR homozygotes may therefore be, at least in part, a consequence of inappropriate protein-protein interactions, secondary to extrasynaptic (and possibly cytoplasmic) expression of CAR. The C termini of the predominant CAR isoforms contain PDZ domain-binding sites that mediate CAR-induced cell adhesion.44-47
Several lines of CAR transgenics have been produced to examine the role of the high-affinity receptor in adenovirus-mediated gene transfer.48-51
Like the transgenics described here, most of these lines have been engineered to express CAR under the control of tissue-specific promoters, mainly in lymphocytes.48,50,51
We could not generate any transgenic lines that expressed full-length CAR ubiquitously, using the cytomegalovirus promoter/enhancer (N.L. and H.L., unpublished). The only transgenic lines with ubiquitous expression were produced with a construct expressing a CAR molecule with a truncation of the entire cytoplasmic domain.49
Hence, the C-terminal domain may be required for induction of pathological changes. Because CAR seems to be strictly regulated during development, the persistent expression of a cell adhesion molecule such as CAR may be deleterious in certain tissues. It is easily conceivable that overexpression of a protein containing a PDZ domain-binding motif may disturb existing protein complexes by competing for binding and/or induce the association of abnormal, nonphysiological complexes. This is a likely scenario in these CAR+/+ mice in view of the previously discussed normal transcript levels for both dystrophin and dysferlin but in the absence of accumulation of the corresponding proteins (Figure 6)
.
Protein-Protein Interactions
In our hands, CAR does not co-immunoprecipitate with caveolin-3 or ß-dystroglycan, suggesting its interactions with these proteins are either indirect or not sufficiently robust to withstand co-precipitation. Of interest, a number of these proteins have been reported to interact with each other: caveolin-3 interacts with nNOS,32,33
ß-dystroglycan,52
and dysferlin,34
whereas dystrophin binds ß-dystroglycan.53,54
One possibility is that CAR may, indirectly, potentiate the interaction of caveolin-3 with ß-dystroglycan,52
leading to perturbations of the potential interdependent interactions that exist among caveolin, ß-dystroglycan, and dystrophin. Both caveolin and dystrophin have WW-like domains that bind a PPXY motif at the extreme C terminus of ß-dystroglycan.52,53
Increased expression of caveolin could therefore competitively displace ß-dystroglycan from dystrophin and thereby destabilize the dystrophin glycoprotein complex. Additionally, the increased accumulation of ß-dystroglycan may lead to a disruption of the interaction of caveolin-3 with dysferlin,34
leading to the observed reduction in dysferlin levels despite the presence of dysferlin transcript (Figure 6)
.
Relation to Human and Mouse Myopathies
The constellation of the general myopathological features observed in CAR+/+ animals, including necrosis, phagocytosis, and atrophy of muscle fibers, are highly nonspecific and may be observed in diverse human muscle diseases.55 In making comparisons with human myopathies, two features of the CAR+/+ histochemical profile deserve special mention. One is that in human states of primary dystrophin deficiency (Duchenne or Becker muscular dystrophies), there is a decrease of the sarcolemmal ß-dystroglycan56 but an increase of extrasynaptic utrophin.57 By contrast, in the CAR+/+ transgenics, in which there is substantial dystrophin deficiency, ß-dystroglycan is up-regulated, and extrasynaptic utrophin is not seen. Secondly, a substantial de novo up-regulation of caveolin-3 has not been described in human muscle disease.
Experimental up-regulation of caveolin-3 has been produced in caveolin-3 transgenic mice, in which caveolin-3 is ubiquitously expressed under the control of the relatively strong hybrid chicken ß-actin promoter/cytomegalovirus enhancer, resulting in a threefold to fivefold increase in expression in heart and skeletal muscle relative to endogenous levels.12 Skeletal muscle pathology develops within 3 to 4 weeks of age, and the phenotype is retained at least until 15 months, the last reported date of analysis. These mice have virtually no dystrophin and a threefold to fourfold reduction in ß-dystroglycan; this is accompanied by inhibition of NOS activity (without an alteration of nNOS or endothelial NOS levels). The CAR homozygotes, with a manifest decrease in nNOS and complete absence of dystrophin, have some similarities with caveolin-3-overexpressing mice. However, they also have many divergent features, the most important being a persistence, if not an increase, of ß-dystroglycan accumulation, as well as a dramatic reduction of dysferlin levels.
There is no report of dysferlin down-regulation in caveolin-3 transgenics, and dysferlin-null mice express normal levels of caveolin-3.20 On the other hand, some patients suffering from limb-girdle muscular dystrophy type 1C (LGMD1C) have a mutated caveolin-3 gene do show secondary reduction in dysferlin immunostaining.58-60 The converse is also observed in that patients who have dysferlin mutations (LGMD2B and Miyoshi myopathy) may have variable caveolin-3 levels.61
Dystrophin- or dysferlinopathies share a common feature of pathophysiology involving sarcolemmal instability attributable to either the compromised cytoskeletal-extracellular linkage normally maintained by dystrophin and its associated glycoprotein complex or an undermined plasmalemmal integrity caused by impaired dysferlin-mediated repair mechanisms. In the CAR homozygotes, the increase in caveolin-3 and the concomitant decreases in dystrophin and dysferlin presumably precipitate a cascade that involves, among other things, down-regulation of nNOS and aberrant MAPK signaling. Although it is presently unclear how persistent overexpression of CAR brings about such dramatic changes in the accumulation of dystrophin, dysferlin, caveolin-3, and ß-dystroglycan in mature skeletal muscle, our observations are consistent with the more severe pathological phenotype and shortened lifespan of the CAR homozygotes in comparison with either mdx, caveolin-3-overexpressing,12 or dysferlin-null mice.20,29 Although the transgenic CAR model does not have a human counterpart, it provides a valuable example of possible interactions of clinically important molecules of the muscle cell surface.
|
Acknowledgements |
|---|
|
Footnotes |
|---|
Supported by the Canadian Institutes for Health Research (MOP-53071) and the Muscular Dystrophy Association (USA).
C.A.S., N.L., P.C.H., and J.N. contributed equally to this article.
N.L. was a postdoctoral fellow of Aktion Benni & Co. e.V.; J.N. is a Killam scholar; J.N. and B.J.P. are Scholars of the Fonds de la Recherche en Santé du Québec.
Accepted for publication August 29, 2006.
|
References |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  M. Mirza, C. Petersen, K. Nordqvist, and K. Sollerbrant Coxsackievirus and Adenovirus Receptor Is Up-Regulated in Migratory Germ Cells during Passage of the Blood-Testis Barrier Endocrinology, November 1, 2007; 148(11): 5459 - 5469. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
