Ataluren: An Investigational Dystrophin Restoration Drug ...
Dystrophin is Not a Specific Component of the Cardiac Costamere
-
Upload
leftrright -
Category
Documents
-
view
9 -
download
0
description
Transcript of Dystrophin is Not a Specific Component of the Cardiac Costamere
+
Circulation Researchcircres.ahajournals.org
Circulation Research. 1997; 80: 269-280doi: 10.1161/01.RES.80.2.269
Articles
Dystrophin Is Not a Specific Component of the Cardiac Costamere
Shirley Stevenson, Stephen Rothery, Michael J. Cullen, Nicholas J. Severs
Author Affiliations
Correspondence to Prof N.J. Severs, Cardiac Medicine, Imperial College School ofMedicine at National Heart and Lung Institute, Royal Brompton Hospital, Sydney Street,London SW3 6NP, England. E-mail [email protected]
Abstract
Dystrophin is a key component of the subsarcolemmal skeleton of muscle cells, and lack
of dystrophin is the direct cause of Duchenne muscular dystrophy. In skeletal muscle,
dystrophin is reported to be localized specifically at costameres, transversely oriented
riblike subsarcolemmal plaques that mechanically couple the contractile apparatus to the
extracellular matrix. Costameres are characteristically rich in vinculin and are prominent
in cardiac as well as skeletal muscle. To define the precise spatial relationship between
dystrophin in relation to the costamere in cardiac muscle, we applied high-resolution
single- and double-immunolabeling techniques, under a range of preparative conditions,
with visualization of vinculin (as a costamere marker) and dystrophin by confocal
microscopy and by the freeze-fracture cytochemical technique, fracture label.
Immunoconfocal visualization revealed dystrophin as a continuous uniform layer at the
cytoplasmic surface of the peripheral plasma membrane of the rat cardiac myocyte at
both costameric and noncostameric regions. The pattern of labeling was reproducible
with three different antibodies and was independent of time and antibody concentration.
Platinum/carbon replicas and thin sections of fracture-label specimens permitted
high-resolution visualization of the distribution of dystrophin in plan views of the freeze-
fractured plasma membrane and in relation to the sarcomeric banding patterns of the
underlying myofibrils. These results confirmed no preferential association of dystrophin
with costameres or with any region of the sarcomeres of underlying myofibrils in rat
cardiac tissue. We conclude that in contrast to skeletal muscle, dystrophin in cardiac
muscle is not exclusively a component of the costamere.
Key Words:
costamere
dystrophin
vinculin
confocal microscopy
freeze-fracture cytochemistry
Dystrophin, the 427-kD protein product of the Duchenne/Becker muscular dystrophy
gene, is a major component of the subsarcolemmal skeleton of muscle cells. The
subsarcolemmal skeleton acts as a scaffold at the cytoplasmic surface of the plasma
membrane, linking the intracellular cytoskeleton to the extracellular matrix. Dystrophin is
tightly associated with a series of transmembrane proteins, the sarcoglycan and
dystroglycan complexes, which link externally to laminin, a component of the basal
lamina. The interaction of dystrophin with the cytoskeleton within the cell is
mediated via binding to F-actin. Lack of dystrophin due to mutations in the dystrophin
gene leads to a gradual but remorseless degeneration of skeletal and cardiac muscle
with, in the case of Duchenne muscular dystrophy, fatal consequences for the patient.
Since the exact function of dystrophin is not yet understood, the precise cellular
mechanism initiating myofiber necrosis has yet to be identified. However, because of its
position linking the cytoskeleton to the extracellular matrix, the most accepted current
hypothesis regarding the role of the dystrophin/glycoprotein complex is that it has a
mechanical function, strengthening the plasma membrane during contraction of the
muscle.
The subcellular distribution of dystrophin has been extensively studied in skeletal
muscle, where numerous studies have demonstrated its localization at the cytoplasmic
surface of the plasma membrane. Initial immunofluorescence studies in skeletal muscle
1 2 3 4 5
Dystrophin Is Not a Specific Component of the Cardiac C... http://circres.ahajournals.org/content/80/2/269.full
1 of 12 02/27/2013 08:15 PM
reported a homogeneous distribution beneath the plasma membrane. However,
immunogold localization at the electron microscopic level has suggested a lattice-like
organization, and a series of more recent immunofluorescence studies,
including visualization by confocal microscopy, have reported a regular, nonuniform
lattice-like arrangement in which dystrophin is predominantly localized at transversely
oriented riblike subsarcolemmal plaques called costameres. Costameres,
which are found in both skeletal and cardiac muscle, anchor the myofibrils to the plasma
membrane, maintain their spatial organization, and serve as sites of mechanical
coupling between the contractile apparatus and the extracellular matrix. Originally
defined by the presence of their high vinculin content, costameres typically contain a
range of other proteins, including spectrin, integrins, and desmin.
Compared with these investigations of skeletal muscle, fewer studies have investigated
dystrophin organization in cardiac muscle, and most have suggested a continuous
uniform distribution at the surface plasma membrane, similar to that described in the
earlier investigations of skeletal muscle. The organization of dystrophin
specifically in relation to simultaneously identified costameres has not previously been
investigated in detail in cardiac muscle. In order to define the precise spatial relationship
between dystrophin in relation to the costamere in cardiac muscle, the present study set
out to apply high-resolution single- and double-immunolabeling techniques, under a
range of preparative conditions, for simultaneous visualization of vinculin and dystrophin
by confocal microscopy and freeze-fracture cytochemistry. The results demonstrate that
dystrophin in rat cardiac muscle is not uniquely distributed at costameres but is
continuously and uniformly distributed at the cytoplasmic surface of the peripheral (ie,
nonintercalated disk) plasma membrane.
Materials and Methods
Sources of Tissue
Fifteen male Sprague-Dawley rats (≈300 g body wt) were used. The animals were
preanesthetized by an intraperitoneal injection of Hypnorm (0.315 mg/mL fentanyl citrate
and 10 mg/mL fluanisone, at 0.5 mL/kg body wt) and then anesthetized with
intraperitoneal Hypnovel (2.0 mg/kg midazolam hydrochloride) before retrograde
perfusion fixation via a catheter in the abdominal aorta. After initial perfusion with
heparinized PBS, the hearts were perfused with 2% paraformaldehyde (PBS-buffered,
pH 7.4) for 15 minutes. The procedures were conducted according to the Animals
(Scientific Procedures) Act, 1986, under license from the Home Office. The fixed heart
was removed, and half-ventricle slices were frozen in isopentane cooled with liquid
nitrogen for frozen sectioning. For fracture-label electron microscopy, tissue blocks of 3
to 5 mm were cryoprotected with 30% PBS-buffered glycerol for 2 hours before
mounting and freezing as described below.
Antibodies and Detection Systems
The following three primary antibodies were used for dystrophin labeling: (1) Dy8/6C5, a
mouse monoclonal raised against the last 17 amino acids of the COOH terminal domain,
(2) P1583, a rabbit polyclonal raised against the same sequence of the COOH terminal
domain, and (3) Dy4/6D3, a mouse monoclonal raised against a fusion protein
containing a 208–amino acid sequence in the region of exons 26 to 29 (ie, an area near
the NH2 terminus of the rod domain; for convenience, referred to here as NH2 terminus
antibody). The monoclonals were a gift from Dr Louise Anderson (University of
Newcastle Upon Tyne); the polyclonal antibody was a gift from Dr Henry Klamut (Ontario
Cancer Institute, Toronto, Canada). Western blots confirmed that the antibodies
recognized a single band of >400 kD (ie, dystrophin) in cardiac and skeletal muscle (Fig
1⇓). For vinculin (costamere marker) and α-actinin, standard commercially available
mouse monoclonal antibodies were used (Sigma Chemical Co).
Figure 1.
Western blots demonstrating
specificity of antibodies for
dystrophin. Homogenates of rat
skeletal muscle and left and right
ventricular myocardium were run
on 6.5% SDS–polyacrylamide
gels, transferred to polyvinylidene
difluoride membranes, and
incubated with mouse monoclonal
6 7 8 9 10
11 12 13 14
15 16 17 18
19 20 21
22
23 24 25
9 26 27
3
Dystrophin Is Not a Specific Component of the Cardiac C... http://circres.ahajournals.org/content/80/2/269.full
2 of 12 02/27/2013 08:15 PM
View larger version:In this page In a new windowDownload as PowerPoint Slide
View larger version:
anti-dystrophin antibodies,
followed by alkaline phosphatase–
conjugated secondary antibodies.
A, Results with antibody Dy8/6C5
(COOH terminus). B, Results with Dy4/6D3 (NH2 terminus). Prominent
immunoreactive bands corresponding to dystrophin are observed at ≈400
kD.
The secondary antibody/detection systems used for immunoconfocal microscopy were
(1) biotinylated goat anti-mouse immunoglobulin and Texas red–streptavidin (Amersham
Life Sciences) and (2) goat anti-rabbit FITC (Dako). For fracture-label immunogold
electron microscopy, the detection systems were (1) biotinylated goat anti-mouse
immunoglobulin used with 10-nm gold–streptavidin complexes (Amersham Life
Sciences) and (2) goat anti-rabbit 5 nm gold complexes (British BioCell International).
Immunolabeling for Confocal Microscopy
For immunoconfocal microscopy, 10-μm cryosections were cut at −25°C and
thaw-mounted onto poly-L-lysine–coated glass slides. They were treated with 0.3%
Triton X-100 for 15 minutes to improve permeability to the reagents, followed by 0.5%
bovine serum albumin (as blocking agent) at room temperature. The sections were then
incubated for single labeling with anti-vinculin antibody (1:50) overnight or with
anti-dystrophin antibody used at concentrations of 1:10, 1:50, 1:100, 1:500, and 1:1000
for 30 minutes, 1 hour, 2 hours, 3 hours, 4 hours, or overnight. For double labeling,
sections were exposed sequentially to (1) the polyclonal anti-dystrophin, followed by the
monoclonal anti-vinculin, or (2) the polyclonal anti-dystrophin (COOH terminus), followed
by the monoclonal anti-dystrophin (NH2 terminus) overnight and then by biotinylated
anti-mouse/streptavidin–Texas red (1:250) and anti-rabbit–FITC (1:20) for 1 hour each.
The following controls were run in parallel: (1) omission of primary antibody, (2)
switching of detection systems (eg, using mouse monoclonal followed by anti-rabbit
secondary antibody), and (3) reversing the order of primary antibodies in the double-
labeling procedure. The sections were mounted with Citifluor mounting medium.
Confocal Laser Scanning Microscopy
The immunolabeled sections were examined by confocal laser scanning microscopy
using a Leica TCS 4D equipped with an argon/krypton laser and fitted with the
appropriate filter blocks for the detection of fluorescein and Texas red fluorescence.
Double-labeled samples were imaged using simultaneous dual-channel scanning. Both
single optical sections and projection views from sets of 10 consecutive single optical
sections taken at intervals between 0.6 and 1 μm were examined. All specimens were
examined within 24 hours of immunolabeling.
Fracture-Label Electron Microscopy
Fracture-label electron microscopy was carried out by following a procedure modified
from that described by Pinto da Silva et al. A mounting and fracturing technique was
developed to increase the incidence of suitably fractured plasma membranes (Fig 2⇓).
This involved manual freeze-fracturing of adhesive-mounted tissue in which the long
axes of the myocytes were oriented parallel with the plane of the metal mounts on either
side. Samples of the perfusion-fixed glycerinated rat hearts were sandwiched between
small squares of Thermanox coverslips using cyanoacrylate adhesive (Perma Bond C,
R.S. Components) and rapidly frozen in liquid nitrogen slush (ie, liquid nitrogen cooled to
its melting point). The sandwich was fractured by using a blade under liquid nitrogen and
allowed to thaw in precooled 2% paraformaldehyde in 30% glycerol for 5 minutes. The
thawed specimens were rinsed in 30% glycerol to remove excess fixative and
deglycerinated by passage through 1 mmol/L glycylglycine in 30% glycerol for 5 minutes
followed by pure 1 mmol/L glycylglycine for a further 5 minutes.
Figure 2.
Fracture-label procedure
developed to obtain a high
incidence of fractures that follow
the myocyte plasma membrane.
Slices of myocardium are
mounted using cyanoacrylate
adhesive between pairs of
28
Dystrophin Is Not a Specific Component of the Cardiac C... http://circres.ahajournals.org/content/80/2/269.full
3 of 12 02/27/2013 08:15 PM
In this page In a new windowDownload as PowerPoint Slide
View larger version:
Thermanox coverslips (Perma
Bond C, R.S. Components). The
myocardial sample is prepared
such that the long axes of the myocytes lie parallel with the coverslips.
Tissue sandwiches are frozen in subcooled liquid nitrogen and fractured with
a precooled razor blade. When biological samples are fractured in this way,
membranes may be split in the same manner as in conventional freeze-
fracture electron microscopy. Labeling is done after the freeze-
fractured samples have been thawed, permitting labeling of the membrane
halves created by freeze fracture. The labeled specimens are then examined
by thin sectioning or as platinum/carbon replicas.
Primary antibody treatment for single and double labeling of vinculin and dystrophin
(both COOH and NH2 termini) was carried out as described for confocal microscopy,
followed by the corresponding secondary antibody–gold complex (1:50, 1 hour at room
temperature). Gold markers of 10 nm were used with the monoclonal antibodies, and
markers of 5 nm were used with the polyclonal antibodies throughout. In double-labeling
experiments, each primary antibody treatment was followed by its corresponding
secondary detection system (eg, the following sequence: [1] dystrophin COOH-terminal
antibody, [2] anti-rabbit/5 nm gold, [3] monoclonal vinculin, and [4] biotinylated
anti-mouse/streptavidin–10 nm gold). Separate experiments in which specimens were
single-immunogold–labeled for α-actinin followed the same detection procedure as used
for vinculin. All specimens were rinsed in PBS, postfixed in 2.5% glutaraldehyde for 30
minutes, further rinsed in PBS, and processed for thin sectioning or platinum/carbon
replication. Specimens for thin sectioning were postfixed in OsO4, dehydrated through a
graded series of ethanols, and embedded in Araldite. Semithin and ultrathin sections
were cut at right angles to the fracture plane using a Riechert E ultramicrotome. For
replication, the specimens were partially dehydrated (to 70% ethanol), dried, and
mounted fracture side up on the stage of a Balzers BAF 400T unit, and platinum-carbon
replicas were prepared at ambient temperature. The replicas were carefully cleaned in
sodium hypochlorite such that the biological material was removed without dislodging
the gold. Sections and replicas were examined using a Philips 301 electron microscope.
Quantitative Analysis of Dystrophin Immunogold Labeling
To examine the distribution of plasma membrane dystrophin in relation to the Z/I-band
and A-band regions of the underlying contractile apparatus, 15 to 20 thin-section
micrographs of plasma membrane P-face fractures (magnification, ×36 000) from each
of four separate fracture-label runs of specimens labeled with P1583 and Dy4/6D3
anti-dystrophin antibodies were analyzed. The number of gold particles per unit length of
the plasma membrane overlying the Z/I-band region and the A-band region was
determined using VIDS III image analysis software (Analytical Measuring Systems).
Statistical significance was assessed using the nonparametric Mann-Whitney U test.
Results
Immunoconfocal Localization
Confocal microscopy of single-labeled sections consistently revealed characteristic
patterns of vinculin and dystrophin localization (Figs 3⇓ and 4). Vinculin was distributed
in a prominent punctate pattern around the outer circumference of the cells, in the
classical positions of the costameres (Fig 3A and 3B⇓⇓). This pattern of vinculin
distribution was so sharply defined that it was readily visible in low-magnification survey
views (Fig 3A⇓). Vinculin was particularly conspicuous at the transversely oriented
portions of the intercalated disks, corresponding to the positions of the fascia adherens
junctions but absent from the longitudinal segments of intercalated disk membrane. In
higher magnification views (Fig 3B⇓), less prominent striations of vinculin
immunoreactivity, in register with the costameres, were visible. This immunolabeling
extended as finger-like projections deep into the cell, as confirmed by serial optical
sectioning, and was identified as being associated with transverse tubules.
Figure 3.
Immunolocalization of vinculin by
confocal microscopy. A,
Low-magnification survey view. B,
High-magnification view of boxed
area in panel A. Note localization
29 30 31
29
Dystrophin Is Not a Specific Component of the Cardiac C... http://circres.ahajournals.org/content/80/2/269.full
4 of 12 02/27/2013 08:15 PM
In this page In a new windowDownload as PowerPoint Slide
View larger version:In this page In a new windowDownload as PowerPoint Slide
View larger version:In this page In a new windowDownload as PowerPoint Slide
of vinculin as prominent rows of
spots at the peripheral cell surface
(arrowheads in panel B),
representing costameres. Transversely running segments of the intercalated
disk (d), corresponding to the position of the fasciae adherentes, show
strong signal. Weaker vinculin labeling is apparent at transverse tubules
(indicated by dotted lines in panel B). Bars=25 μm (A) and 5 μm (B).
Dystrophin, by contrast, appeared uniformly distributed over the cell surface; a punctate
pattern was never observed (Fig 4⇓). The dystrophin label was intense and continuous,
apart from gaps at the end-on abutments between the cells corresponding to fascia
adherens junctions of the intercalated disk membrane. The continuous pattern was
consistent irrespective of the antibody concentration or period of incubation and was
confirmed in three dimensions by taking serial optical sections through the tissue slice.
As with vinculin, higher magnification views disclosed less pronounced labeling within
the cell in the form of discontinuous striations or regular punctate patterns (Fig 4B⇓).
This signal was weaker than that found in the corresponding position for vinculin and
was demonstrated by serial optical sectioning to be organized as finger-like projections
within the cell. That this dystrophin labeling was transverse tubular rather than Z-band–
associated was further confirmed by its being quite distinct from the immunolabeling
pattern for α-actinin (not illustrated).
Figure 4.
Immunolocalization of dystrophin
by confocal microscopy. A,
Low-magnification survey view. B,
High-magnification view of boxed
area in panel A. Note prominent
uniform labeling at the peripheral
plasma membrane. Weaker
signal, in the form of punctate
striations representing transverse tubules, is apparent within the cell (B). In
this example, antibody Dy8/6C5 (COOH terminus) was used; all
anti-dystrophin antibodies gave the same result. Bars=25 μm (A) and 5 μm
(B).
The spatial relationship between the distributions of vinculin and dystrophin were
dramatically apparent when the two components were simultaneously visualized by
dual-channel imaging of double-label preparations, as illustrated in Fig 5⇓. In this
composite figure, the immunolabeling patterns for vinculin and dystrophin are presented
separately and simultaneously in longitudinal and transversely sectioned myocytes. As
with the corresponding single-labeling experiments, vinculin reveals a clear punctate
distribution at the cell surface, whereas dystrophin shows a continuous pattern (Fig 5A
through 5C⇓), apparent in both longitudinal and transversely sectioned cells.
Simultaneous viewing clearly demonstrated that dystrophin does not localize specifically
to the vinculin-rich costameres but has a widespread distribution close to the cell
surface, present both at the vinculin-rich punctae and in the intervening regions of
membrane. The interior labeling for dystrophin associated with transverse tubules was
found to coincide with that of vinculin, although the signal for the latter was the more
intense.
Figure 5.
Simultaneous confocal
visualization of vinculin (vinc) and
dystrophin (dys) by dual-channel
imaging of double-labeled
preparations. Immunolabeling for
vinc and dys is shown
independently in panels A and B,
respectively, in the longitudinal
section (left column) and
transverse section (right column).
Note clear punctate cell surface
labeling for vinc and continuous
Dystrophin Is Not a Specific Component of the Cardiac C... http://circres.ahajournals.org/content/80/2/269.full
5 of 12 02/27/2013 08:15 PM
View larger version:In this page In a new windowDownload as PowerPoint Slide
labeling for dys. Combination of these images in panel C clearly shows that
vincu and dys signal does not colocalize at identical sites in the peripheral
plasma membrane. Note punctate patterns of alternating red and yellow
fluorescence (arrowhead). These represent dys only (red) in the regions
between costameres and dys plus vinc (yellow) in costameres. Intercalated
disks (d) are rich in vinc (A) but lack dys (B). Labeling for dys within the cell
(transverse tubules) coincides with that of vin. Bar=25 μm.
Immunogold Fracture-Label Electron Microscopy
Thin-section examination of fracture-labeled specimens confirmed that vinculin is
specifically localized at the plasma membrane overlying the Z disks of the superficial
myofibrils, in a position corresponding to the costamere (Fig 6A⇓). No vinculin was
detectable in the intervening membrane areas. The label was apparent only on those
fractured fragments of tissue containing the P half of the plasma membrane (ie, the
half-membrane leaflet attached to the protoplasm ). E halves (ie, half-membrane
leaflets attached to the extracellular space) were unlabeled. Dystrophin, by contrast, was
uniformly distributed at the level of the plasma membrane, with no preferential
association with any region of the sarcomere (Fig 6B⇓). This pattern of distribution was
confirmed with three different anti-dystrophin antibodies used independently and
simultaneously for double labeling (Fig 6B⇓). In both cases, dystrophin labeling was
predominantly but not exclusively associated with the plasma membrane P half (gold
label sixfold more abundant on the P half than the E half). Cross-fractured myofibrils (ie,
within the cell, below the level of the plasma membrane) revealed no detectable
dystrophin at the Z/I-band region or any other part of the myofibril (Fig 6C and 6D⇓⇓).
Proteins such as α-actinin, known to be present at the Z disk, were readily detectable
with the same approach (Fig 7⇓).
Figure 6.
Immunogold localization of
vinculin and dystrophin in
fracture-label specimens
examined by thin sectioning. A,
Localization of vinculin in a cell in
which the fracture has followed
the plasma membrane. Gold label
(arrowheads) occurs specifically
at the level of the plasma
membrane in regions in register
with the Z bands (Z) of the
underlying myofibrils. B,
Distribution of dystrophin along a
fractured plasma membrane localized by dual labeling with rabbit polyclonal
P1583(COOH terminal) using 5 nm gold (arrowheads) and mouse
monoclonal Dy4/6D3 (NH2 terminus) using 10 nm gold. Labeling is
continuous along the plasma membrane, showing no preferential association
with any sarcomeric region of the underlying myofibril (positions of A, I, and
Z bands indicated). C and D, Dystrophin labeling of cross-fractured myocytes
(ie, the fracture has passed through the contractile apparatus and other
cytoplasmic components within the cell). There is no labeling for dystrophin
at the Z bands or other regions of the sarcomere, demonstrating that
dystrophin is exclusively a membrane-associated protein. Arrows in panel C
show labeling where, after cross-fracturing the upper myocyte, the fracture
has followed the plasma membrane of a neighboring cell. Bars=200 nm.
Figure 7.
Immunogold localization of α-actinin in thin-sectioned fracture-label
specimens in which the myocytes have been cross-fractured. Labeling is
specifically associated with the Z bands (Z) within the myofibril (arrows,
panel A). In the example in panel B, the fracture has fortuitously traveled
along a Z band, which is heavily labeled. Bars=250 nm.
32
Dystrophin Is Not a Specific Component of the Cardiac C... http://circres.ahajournals.org/content/80/2/269.full
6 of 12 02/27/2013 08:15 PM
View larger version:In this page In a new windowDownload as PowerPoint Slide
View larger version:In this page In a new windowDownload as PowerPoint Slide
View larger version:In this page In a new windowDownload as PowerPoint Slide
Replicas of fracture-labeled specimens provided en face views of dystrophin distribution
at the cell surface (Fig 8⇓). In favorable views, the positions of the costameres were
discernible as transverse elevations at the surface, resulting from “sinking” of
surrounding areas during specimen drying. The lack of any preferential association of
dystrophin with the costameres was confirmed in these preparations (Fig 8⇓), as was
the absence of label in cross-fractured specimens in which the myofibrils had been
exposed (Fig 9⇓).
Figure 8.
Immunogold localization of
dystrophin in fracture-label
specimens examined using
platinum-carbon replicas. These
examples show plasma
membrane fractures, revealing the
spatial distribution of dystrophin in
the plane of the membrane using
Dy4/6D3 antibody against the NH2
terminus (A) and Dy8/6C5
antibody against the COOH
terminus localized with 10 nm
gold–secondary antibody
complexes (B). Drying of the specimens before replication causes shrinkage,
leaving mitochondria and costameres (c) standing proud at the cell surface.
Dystrophin appears widely distributed in these en face views of the
membrane and has no association with the costameres. Bars=500 nm.
Figure 9.
Replica of a dystrophin-labeled
cross-fractured myocyte from the
same experiment as in Fig 8⇑. In
examples like this, where the
fracture plane reveals the
myofibrils within the cell, no
labeling for dystrophin is seen. M,
A, and Z/I indicate sarcomeric
bands of the myofibril; mi
indicates mitochondria. Bar=500
nm.
Fracture-labeled tissue processed for double-labeling confirmed that vinculin and
dystrophin have quite distinctive distributions (Fig 10⇓). As in the single-labeling
experiments, dystrophin labeling was observed along the entire lengths of plasma
membrane profiles, whereas vinculin labeling was confined to clusters at the Z disk.
Figure 10.
Double labeling for vinculin and dystrophin (polyclonal antibody P1583) as
viewed in thin sections of fracture-label preparations. In fractures that follow
the plasma membrane, vinculin (10 nm gold) occurs only in line with the Z
band (Z) of the underlying myofibril, whereas dystrophin (5 nm gold) occurs
Dystrophin Is Not a Specific Component of the Cardiac C... http://circres.ahajournals.org/content/80/2/269.full
7 of 12 02/27/2013 08:15 PM
View larger version:In this page In a new windowDownload as PowerPoint Slide
continuously along the plasma
membrane (panels A and B). In
the lower magnification example
in panel A, the larger gold
(vinculin) is indicated with large
arrowheads; the small gold
(dystrophin), with small
arrowheads. Bars=250 nm.
Quantitative analysis of specimens labeled with the two anti-dystrophin antibodies
(P1583 and Dy/6D3) revealed no significant differences in the extent of labeling over
A-band versus Z/I-band plasma membrane regions (A band, median 7.86 gold
particles/μm; Z/I band, median 7.14 gold particles/μm; P>.05; n=83).
Discussion
In the present investigation, we have applied simultaneous dual-channel scanning
immunoconfocal microscopy and complementary double-immunogold electron
microscopy to investigate the spatial relationship between dystrophin and costameres in
cardiac muscle. For localization at the electron microscopic level, we elected to use the
technique of fracture label, one of a range of methods in freeze-fracture
cytochemistry. In fracture label, cytochemical labeling is performed immediately after
samples have been freeze-fractured and thawed. Because membranes are split along
their hydrophobic interior when fractured at low temperature, the label has unrestricted
access to the entire face-on aspects of plasma membranes of cells within the tissue
sample. A further feature of the technique, of particular relevance to the present study, is
that both integral membrane components and their associated peripheral proteins on the
cytoplasmic side of the membrane are rendered accessible for labeling. This happens
because upon exposure to aqueous media at the thawing stage, the fractured
half-membrane leaflets become reorganized into a discontinuous bilayer, thereby
exposing underlying cytoplasmic or extracellular components. Fracture label is thus
particularly well suited to the investigation of proteins (such as dystrophin and vinculin)
that are closely associated with the plasma membrane. Our observation that in
fracture-label the dystrophin antibodies labeled the plasma membrane P half rather than
the E half indicates that in cardiac muscle, both the carboxy- and amino-terminal
domains of dystrophin are closely associated with the protoplasmic side of the
membrane, as reported in an earlier fracture-label study in skeletal muscle (in which the
carboxy-terminal domain was localized to the P half ) and as widely depicted in current
models of the dystrophin-glyoprotein complex.
The key question we sought to address was whether dystrophin in cardiac muscle is
specifically associated with costameres, as reported in skeletal muscle, or whether some
other distinctive arrangement characterizes cardiac muscle. Costameres were originally
defined in both skeletal and cardiac muscle by the presence of their high vinculin
content. The intense punctate immunofluorescent labeling of vinculin we
observed at the plasma membrane in the present study is fully consistent with these
earlier observations. Immunogold fracture label confirmed that these patches of vinculin
are localized in the characteristic position of the costamere, at the level of the plasma
membrane overlying the Z bands of the superficial myofibrils. Other features of vinculin
distribution observed in the present study, ie, the presence of high concentrations of
vinculin at the fascia adherens junctions of the intercalated disk and vinculin associated
with the transverse tubular system penetrating into the cell, accord with the established
literature. These comparisons confirm that immunolabeling of vinculin, under the
conditions applied in the present study, provided a reliable means for the identification of
costameres.
In skeletal muscle, the current consensus from immunofluorescence studies is that
dystrophin has a nonuniform distribution at the cytoplasmic surface of the plasma
membrane in the form of dense transversely oriented bands at the I/Z-band level (ie, at
costameres) linked by finer longitudinally oriented strands, in a pattern that lies in
register with α-actinin and mirrors that of spectrin and vinculin. Although, in
guinea pig muscle, a few small patches of dystrophin label apparently may occur in the
absence of vinculin, the two proteins are predominantly colocalized, and current
models specifically depict dystrophin as a component of the skeletal muscle
costamere.
28
30
31
33
3 4 34
19 20 21 22 35
20 36 37
15 16 17
17
34 38
Dystrophin Is Not a Specific Component of the Cardiac C... http://circres.ahajournals.org/content/80/2/269.full
8 of 12 02/27/2013 08:15 PM
The results of the present study, however, indicate that such models are not universally
applicable to cardiac muscle. Instead of having a costameric distribution pattern, our
observations indicate that dystrophin appears uniformly distributed at the cytoplasmic
surface of the general plasma membrane in rat cardiac muscle. Compared with skeletal
muscle, relatively few studies have previously investigated dystrophin organization in
cardiac muscle, and none has applied double labeling for vinculin and dystrophin at both
the immunoconfocal and immunoelectron microscopic levels to allow the precise spatial
localization of dystrophin in relation to the costamere to be defined. Previous reports on
cardiac muscle variously describe a continuous or punctate pattern of dystrophin
distribution at the surface plasma membrane, and the absence or presence of
dystrophin at transverse tubules, although there is general agreement on the lack of
dystrophin at the adherens junctions of the intercalated disks. Our present
observations on the surface plasma membrane and transverse tubules are in close
agreement with those of Frank et al and Klietch et al ; although in contrast to the
findings of Meng et al, we find no evidence for the presence of dystrophin within the Z
disks of myofibrils within the cell either by confocal microscopy or fracture-label
techniques.
Whereas the demonstration of uniform continuous labeling at the peripheral plasma
membrane strongly suggests that dystrophin is ubiquitous at this site, it does not exclude
the possibility of local differences in dystrophin abundance. It might be hypothesized, for
example, that if dystrophin had a preferential, though nonexclusive, association with
costameres, the ability to detect such a relationship would depend critically on
preparative conditions. Our demonstration by confocal microscopy that the continuous
labeling pattern was demonstrable in rat cardiac myocytes by use of three different
antibodies and a wide range of antibody concentration and incubation periods and, in
particular, that the continuous labeling was apparent at extremely low antibody
concentration and very brief incubations; indicates that, if major local differences exist,
they are exceptionally difficult to detect by immunofluorescence. In line with these
findings, quantitative analysis of the immunogold results demonstrated no significant
difference in the extent of dystrophin labeling in costamere regions versus noncostamere
regions of the plasma membrane.
From the point of view of organization of the membrane skeleton, there seems to be no
fundamental necessity for dystrophin and vinculin to coexist at the same plasma
membrane sites. Immunoconfocal studies on smooth muscle have shown that
dystrophin and vinculin are organized in distinct, entirely separate alternating domains.
Taking this and other published findings together with our present results, the current
evidence suggests that the relationship between dystrophin and vinculin may vary in a
characteristic and distinctive manner in each muscle type. Whereas in skeletal muscle
dystrophin is largely colocalized to the same domains as vinculin, the situation is
reversed in smooth muscle, with dystrophin being confined specifically to nonvinculin
zones. Cardiac muscle shows yet another distinctive pattern, with dystrophin distributed
throughout both the vinculin and nonvinculin domains. Such muscle type–specific
features in dystrophin distribution may reflect subtly different roles for dystrophin in
myocardium and skeletal muscle that could in turn influence the relative susceptibility of
these muscle types to dysfunction in myopathic diseases characterized by deficiencies
in dystrophin expression. In Duchenne muscular dystrophy, the loss of dystrophin is as
complete in cardiac muscle as it is in skeletal muscle, but clinically apparent
cardiomyopathy, though common, does not normally become evident until relatively late,
and in only 10% of cases is death attributable to cardiac failure. Comparable but
less severe cardiac abnormalities are apparent in most forms of the clinically milder
Becker muscular dystrophy, in which reduced levels or semifunctional forms of
dystrophin are expressed. That dystrophin is ultimately critical to cardiac function,
however, is demonstrated by the linkage of mutations in the dystrophin gene to a subset
of familial dilated cardiomyopathies that show X-linked inheritance. Here, mutations
specifically affecting dystrophin expression in the heart result in rapidly progressive and
fatal heart failure with no or only relatively minor clinical signs of skeletal muscle
involvement. These findings point to the potential importance of further more
detailed investigation of the expression of dystrophin and components of the
dystroglycan complex in human cardiomyopathies.
Acknowledgments
This study was supported by grants FS/94044 and PG/93136 from the British Heart
Foundation. We thank Steven Coppen for help with the Western blots.
Received September 10, 1996.Accepted November 25, 1996.
7 9 26 27 39 40
26 40
41
42
7 43
44 45 46
45 46
47 48
47 48 49 50
Dystrophin Is Not a Specific Component of the Cardiac C... http://circres.ahajournals.org/content/80/2/269.full
9 of 12 02/27/2013 08:15 PM
References
Ervasti JM, Campbell KP. A role for the dystrophin-glycoprotein complex as atransmembrane linker between laminin and actin. J Cell Biol . 1993;122:809-823.
Abstract/FREE Full Text
1.
Tinsley JM, Blake DJ, Zuellig RA, Davies KE. Increasing complexity of thedystrophin-association protein complex. Proc Natl Acad Sci U S A .1994;91:8307-8313. Abstract/FREE Full Text
2.
Campbell KP. Three muscular dystrophies: loss of cytoskeleton-extracellular matrixlinkage. Cell . 1995;80:675-679. CrossRef Medline
3.
Ohlendieck K. Towards an understanding of the dystrophin-glycoprotein complex:linkage between the extracellular matrix and the membrane cytoskeleton in musclefibers. Eur J Cell Biol . 1996;69:1-10. Medline
4.
Brown RH Jr. Dystrophin-associated proteins and the muscular dystrophies: aglossary. Brain Pathol . 1996;6:19-24. Medline
5.
Zubrzycka-Gaarn EE, Bulman DE, Karpati G, Burghes AHM, Belfall B, Kalmut HJ,Talbot J, Hodges RS, Ray PN, Worton RG. The Duchenne muscular dystrophygene product is localised in sarcolemma of human skeletal muscle. Nature .1988;333:466-469. CrossRef Medline
6.
Arahata K, Ishiura T, Tsukahara T, Suhara Y, Eguchi C, Ishihara T, Nonaka I,Ozawa E, Sugita H. Immunostaining of skeletal and cardiac muscle surfacemembrane with antibody against Duchenne muscular dystrophy peptide. Nature .1988;333:861-863. CrossRef Medline
7.
Bonilla E, Samitt C, Miranda A, Hays A, Salviati G, Dimauro S, Kunkel LM,Hoffman E, Rowland L. Duchenne muscular dystrophy: deficiency of dystrophin atmuscle cell surface. Cell . 1988;54:447-452. CrossRef Medline
8.
Byers TJ, Kunkel LM, Watkins SC. The subcellular distribution of dystrophin inmouse skeletal, cardiac and smooth muscle. J Cell Biol . 1991;115:411-421.
Abstract/FREE Full Text
9.
Hoffman EP, Kunkel LM. Dystrophin abnormalities in Duchenne/Becker musculardystrophy. Neuron . 1989;2:1019-1029. CrossRef Medline
10.
Watkins SC, Hoffman EP, Slayter HS, Kunkel LM. Immunoelectron microscopiclocalisation of dystrophin in myofibres. Nature . 1988;333:863-866. CrossRef
Medline
11.
Cullen MJ, Walsh J, Nicholson LV, Harris JB. Ultrastructural localisation ofdystrophin in human muscle by using gold immunolabelling. Proc R Soc Lond BBiol Sci . 1990;240:197-210. Medline
12.
Cullen MJ, Watkins SC. Ultrastructure of muscular dystrophy: new aspects.Micron . 1993;24:287-307.
13.
Wakayama Y, Shibuya S, Jimi T, Takeda A, Oniki H. Size and localisation ofdystrophin molecule: immunoelectron microscopic and freeze etching studies ofmuscle plasma membranes of murine skeletal myofibers. Acta Neuropathol .1993;86:567-577. CrossRef Medline
14.
Straub V, Bittner RE, Le´ger JJ, Voit T. Direct visualization of the dystrophinnetwork on skeletal muscle fiber membrane. J Cell Biol . 1992;119:1183-1191.
Abstract/FREE Full Text
15.
Porter GA, Dmytrenko GM, Winkelmann JC, Bloch RJ. Dystrophin colocalises withβ spectrin in distinct subsarcolemmal domains in mammalian skeletal muscle. JCell Biol . 1992;117:997-1005. Abstract/FREE Full Text
16.
Masuda T, Fujimaki N, Ozawa E, Ishikawa H. Confocal laser microscopy ofdystrophin localization in guinea pig skeletal muscle fibers. J Cell Biol .1992;119:543-548. Abstract/FREE Full Text
17.
Minetti C, Beltrame F, Marcenaro G, Bonilla E. Dystrophin at the plasmamembrane of human fibres shows a costameric localisation. Neuromuscul Disord .1992;2:99-109. CrossRef Medline
18.
Pardo JV, Siliciano JD, Craig SW. A vinculin-containing cortical lattice in skeletalmuscle: transverse lattice elements (“costameres”) mark sites of attachmentbetween myofibrils and sarcolemma. Proc Natl Acad Sci U S A .1983;80:1008-1012. Abstract/FREE Full Text
19.
Pardo JV, Siliciano JD, Craig SW. Vinculin is a component of an extensive networkof myofibril-sarcolemma attachment regions in cardiac muscle fibres. J Cell Biol .1983;97:1081-1088. Abstract/FREE Full Text
20.
Ganote CE, Armstrong S. Ischaemia and the myocyte cytoskeleton: review andspeculation. Cardiovasc Res . 1993;27:1387-1403. FREE Full Text
21.
Danowski BA, Imanaka-Yoshida K, Sanger JM, Sanger JW. Costameres are sitesof force transmission to the substratum in adult rat cardiomyocytes. J Cell Biol .1992;118:1411-1420. Abstract/FREE Full Text
22.
Craig SW, Pardo JV. Gamma actin, spectrin, and intermediate filament proteins23.
Dystrophin Is Not a Specific Component of the Cardiac C... http://circres.ahajournals.org/content/80/2/269.full
10 of 12 02/27/2013 08:15 PM
colocalise with vinculin at costameres, myofibril to sarcolemmal attachment sites.Cell Motil Cytoskeleton . 1983;3:449-462. CrossRef Medline
Shear CR, Bloch RJ. Vinculin in subsarcolemmal densities in chicken skeletalmuscle: localisation and relationship to intracellular and extracellular structures. JCell Biol . 1985;101:240-256. Abstract/FREE Full Text
24.
Nelson WJ, Lazarides E. Expression of the β-subunit of spectrin in nonerythroidalcells. Proc Natl Acad Sci U S A . 1983;80:363-367. Abstract/FREE Full Text
25.
Frank JS, Mottino G, Chen F, Peri V, Holland P, Tuana BS. Subcellular distributionof dystrophin in isolated adult and neonatal cardiac myocytes. Am J Physiol CellPhysiol . 1994;267:C1707-C1716. Abstract/FREE Full Text
26.
Peri V, Ajdukovic B, Holland P, Tuana BS. Dystrophin predominantly localizes tothe transverse tubule /Z-line regions of single ventricular myocytes and exhibitsdistinct associations with the membrane. Mol Cell Biochem . 1994;130:57-65.
CrossRef Medline
27.
Pinto da Silva P, Barbosa MLF, Aguas AP. A guide to fracture-label: cytochemicallabeling of freeze-fractured cells. In: Koehler JK, ed. Advanced Techniques inBiological Electron Microscopy. New York, NY: Springer Verlag; 1986;3:201-227.
28.
Pinto da Silva P, Kachar B, Torrisi MR, Brown C, Parkison C. Freeze-fracturecytochemistry: replicas of critical point-dried cells and tissues after fracture-label.Science . 1981;213:230-233. Abstract/FREE Full Text
29.
Severs NJ. Freeze-fracture cytochemistry: an explanatory survey of methods. In:Severs NJ, Shotton DM, eds. Rapid Freezing, Freeze Fracture, and Deep Etching.New York, NY: Wiley-Liss Inc; 1995:173-208.
30.
Pinto da Silva P, Parkison C, Dwyer N. Freeze-fracture cytochemistry: thin sectionsof cells and tissues after labeling of fracture faces. J Histochem Cytochem .1981;29:917-928. Abstract/FREE Full Text
31.
Branton D, Bullivant S, Gilula NB, Karnovsky MJ, Moor H, Muhlethaler K,Northcote DH, Packer L, Satir B, Satir P, Speth V, Staehelin LA, Steere RL,Weinstein RS. Freeze-etching nomenclature. Science . 1975;190:54-56.
FREE Full Text
32.
Squarzoni S, Sabatelli P, Maltarello MC, Cataldi A, Di Primio R, Maraldi NM.Localization of dystrophin COOH-terminal domain by the fracture-label technique. JCell Biol . 1992;118:1401-1409. Abstract/FREE Full Text
33.
Ahn H,A., Kunkel LM. The structural and functional diversity of dystrophin. NatGenet . 1993;3:283-291. CrossRef Medline
34.
Belkin AM, Ornatsky OI, Glukhova MA, Koteliansky VE. Immunolocalisation ofmeta-vinculin in human smooth and cardiac muscle. J Cell Biol . 1988;107:545-553.
Abstract/FREE Full Text
35.
Geiger B, Tokuyasu KT, Dutton AH, Singer SJ. Vinculin, an intracellular proteinlocalised at specialised sites where microfilament bundles terminate at cellmembranes. Proc Natl Acad Sci U S A . 1980;77:4127-4131. Abstract/FREE Full Text
36.
Schaper J, Froede R, Hein St, Buck A, Hashizume H, Speiser B, Friedl A, BleeseN. Impairment of the myocardial ultrastructure and changes of the cytoskeleton indilated cardiomyopathy. Circulation . 1991;83:504-514. Abstract/FREE Full Text
37.
Taylor RG, Geesink GH, Thompson VF. Is z-disk degradation responsible forpostmorten tenderisation. J Animal Sci . 1995;73:1351-1367.
Abstract/FREE Full Text
38.
Uchino M, Araki S, Mike T, Teramoto H, Nakamura T, Yasutake T. Localisation andcharaterisation of dystrophin in muscle biopsy specimens from Duchenne musculardystrophy and various neuromuscular disorders. Muscle Nerve .1989;12:1009-1016. CrossRef Medline
39.
Klietsch R, Ervasti JM, Arnold W, Campbell KP, Jorgensen AO. Dystrophin-glycoprotein complex and laminin colocalize to the sarcolemma and transversetubules of cardiac muscle. Circ Res . 1993;72:349-360. Abstract/FREE Full Text
40.
Meng HP, Leddy JJ, Frank J, Holland P, Tuana BS. The association of cardiacdystrophin with myofibrils/Z-disc regions in cardiac muscle suggests a novel role inthe contractile apparatus. J Biol Chem. 1996;271:12364-12371.
Abstract/FREE Full Text
41.
North AJ, Galazkiewicz B, Byers TJ, Glenney JR Jr, Small JV. Complementarydistributions of vinculin and dystrophin define two distinct sarcolemma domains insmooth muscle. J Cell Biol . 1993;120:1159-1167. Abstract/FREE Full Text
42.
Hoffman EP, Hudecki MS, Rosenberg PA, Pollina CM, Kunkel LM. Cell andfiber-type distribution of dystrophin. Neuron . 1988;1:411-420. CrossRef Medline
43.
Hunsaker RH, Fulkerson PK, Barry FJ, Lewis RP, Lefier CV, Unverferth DV.Cardiac function in Duchenne's muscular dystrophy: result of 10-year follow-upstudy and non-invasive test. Am J Med. 1982;73:235-238. CrossRef Medline
44.
Emery AEH. Duchenne Muscular Dystrophy. Oxford, England: Oxford Medical45.
Dystrophin Is Not a Specific Component of the Cardiac C... http://circres.ahajournals.org/content/80/2/269.full
11 of 12 02/27/2013 08:15 PM
Publications; 1993;24:109-114. Oxford Monographs on Medical Genetics.
Carter LF, Rubin SA. The molecular and cellular biology of heart failure. Curr OpinCardiol . 1994;9:264-271. Medline
46.
Towbin JA, Hejtmancik JF, Brink P, Gelb B, Zhu XM, Chamberlain JS, McCabeERB, Swift M. X-linked dilated cardiomyopathy: molecular genetic evidence oflinkage to Duchenne muscular dystrophy (dystrophin) gene at the Xp21 locus.Circulation . 1993;87:1854-1865. Abstract/FREE Full Text
47.
Muntoni F, Wilson L, Marrosu MG, Cianchetti C, Mestroni L, Ganau A, Dubowitz V,Sewry C. A mutation in the dystrophin gene selectively affecting dystrophinexpression in the heart. J Clin Invest . 1995;96:693-699.
48.
Yoshida K, Ikeda S, Nakamura A, Kagoshima M, Takeda S, Shoji S, YanagisawaN. Molecular analysis of the Duchenne muscular dystrophy gene in patients withBecker muscular dystrophy presenting with dilated cardiomyopathy. Muscle Nerve .1993;16:1161-1166. CrossRef Medline
49.
Franz WM, Cremer M, Herrmann R, Gru¨nig E, Fogel W, Scheffold T, Goebel HH,Kircheisen R, Ku¨bler W, Voit T, Katus HA. X-linked dilated cardiomyopathy: novelmutation of the dystrophin gene. Ann N Y Acad Sci . 1995;752:470-491. Medline
50.
Articles citing this article
Combined Deficiency of Dystrophin and {beta}1 Integrin in the CardiacMyocyte Causes Myocardial Dysfunction, Fibrosis and Calcification
Circ. Res.. 2008;102:1109-1117,
Abstract Full Text PDF
Challenges and opportunities in dystrophin-deficient cardiomyopathygene therapy
Hum Mol Genet. 2006;15:R253-R261,
Abstract Full Text PDF
Natriuretic peptide receptor-B in adult rat ventricle is predominantlyconfined to the nonmyocyte population
Am. J. Physiol. Heart Circ. Physiol.. 2002;282:H2117-H2123,
Abstract Full Text PDF
Distinct Patterns of Dystrophin Organization in Myocyte Sarcolemmaand Transverse Tubules of Normal and Diseased Human Myocardium
Circulation. 2000;101:2586-2594,
Abstract Full Text PDF
Co-localization of Dystrophin and {beta}-Dystroglycan Demonstrated inEn Face View by Double Immunogold Labeling of Freeze-fracturedSkeletal Muscle
J Histochem Cytochem. 1998;46:945-953,
Abstract Full Text PDF
Spatial Relationship of the C-Terminal Domains of Dystrophin and-Dystroglycan in Cardiac Muscle Support a Direct Molecular Interactionat the Plasma Membrane Interface
Circ. Res.. 1998;82:82-93,
Abstract Full Text PDF
Natriuretic peptide receptor-B in adult rat ventricle is predominantlyconfined to the nonmyocyte population
Am. J. Physiol. Heart Circ. Physiol.. 2002;282:H2117-H2123,
Abstract Full Text PDF
Dystrophin Is Not a Specific Component of the Cardiac C... http://circres.ahajournals.org/content/80/2/269.full
12 of 12 02/27/2013 08:15 PM