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


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


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


4 of 12 02/27/2013 08:15 PM

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


5 of 12 02/27/2013 08:15 PM


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


6 of 12 02/27/2013 08:15 PM




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


7 of 12 02/27/2013 08:15 PM


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


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


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.


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.


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.


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.


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.


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.


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.


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,


Natriuretic peptide receptor-B in adult rat ventricle is predominantlyconfined to the nonmyocyte population

Am. J. Physiol. Heart Circ. Physiol.. 2002;282:H2117-H2123,


Distinct Patterns of Dystrophin Organization in Myocyte Sarcolemmaand Transverse Tubules of Normal and Diseased Human Myocardium

Circulation. 2000;101:2586-2594,


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,


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,


Natriuretic peptide receptor-B in adult rat ventricle is predominantlyconfined to the nonmyocyte population

Am. J. Physiol. Heart Circ. Physiol.. 2002;282:H2117-H2123,



12 of 12 02/27/2013 08:15 PM

Dystrophin is Not a Specific Component of the Cardiac Costamere

Documents

Transcript of Dystrophin is Not a Specific Component of the Cardiac Costamere