Spatial Relationship of the C-Terminal Domains of Dystrophin and ß-Dystroglycan in Cardiac Muscle Support a Direct Molecular Interaction at the Plasma Membrane Interface

From the National Heart and Lung Institute, Imperial College School of Medicine (S.S., S.R., N.J.S.), London, England, and the School of Neuroscience (M.J.C), University of Newcastle on Tyne (England).

Correspondence to Professor N.J. Severs, Department of Cardiac Medicine, National Heart and Lung Institute, Imperial College School of Medicine, Royal Brompton Hospital, Sydney St, London SW3 6NP, England. E-mail n.severs{at}ic.ac.uk

	Abstract

Top Abstract Introduction Materials and Methods Results Discussion References

 
Abstract—Dystrophin and ß-dystroglycan are components of a complex of at least nine proteins (the dystrophin-glycoprotein complex) that physically link the membrane cytoskeleton in skeletal and cardiac muscle, through the plasma membrane, to the extracellular matrix. Mutations in the dystrophin gene, which result in an absence or a quantitative or qualitative alteration of dystrophin, cause a subset of familial dilated cardiomyopathies as well as Duchenne and Becker muscular dystrophy. Biochemical studies on isolated skeletal muscle molecules indicate that dystrophin is bound to the glycoprotein complex via ß-dystroglycan, with the C-terminus of ß-dystroglycan binding to the cysteine-rich domain and first half of the C-terminal domain of dystrophin. Ultrastructural labeling has demonstrated a close spatial relationship between dystrophin and ß-dystroglycan in intact skeletal muscle, but no previous ultrastructural labeling studies have examined the dystrophin/ß-dystroglycan interaction in cardiac muscle. In the present study, we have applied complementary immunoconfocal microscopy and double immunogold fracture-label, a freeze-fracture cytochemical technique that allows high-resolution visualization of labeled membrane components in thin section and in platinum-carbon replicas, to investigate the spatial relationship between dystrophin and ß-dystroglycan in rat cardiac muscle. When immunogold probes of two different sizes for the two proteins were used, "doublets" representing side-by-side antibody labeling were demonstrated in en face views at the level of the plasma membrane. The results support the conclusions that dystrophin and ß-dystroglycan directly interact at the cytoplasmic face of the rat cardiac muscle plasma membrane.

Key Words: dystrophin • ß-dystroglycan • cytoskeleton • freeze-fracture cytochemistry • confocal microscopy

	Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

 
Identification of the gene defects causing muscular dystrophy and related diseases has opened new perspectives on pathogenetic mechanisms of skeletal and heart muscle degeneration. The product of the Duchenne/Becker muscular dystrophy gene, dystrophin, is a large subsarcolemmal cytoskeletal protein that forms an integral part of an oligomeric membrane-associated protein complex referred to as the dystrophin-glycoprotein complex.^{1 2 3 4 5} This complex, which has been extensively studied in skeletal muscle, comprises a series of tightly associated cytoplasmic, transmembrane, and extracellular components.^{1 5 6 7} The transmembrane components include four proteins of the sarcoglycan subcomplex linked laterally to ß-dystroglycan (formerly designated 43-Dag).^{8 9 10} ß-Dystroglycan is bound to -dystroglycan, a peripheral membrane protein on the extracellular side of the membrane, which in turn binds to laminin-2.¹¹ Dystrophin is a long, hinged molecule linked to the cytoplasmic side of the transmembrane complex as well as to a further set of proteins, the syntrophins, toward its C-terminal domain and to F-actin at its N-terminus.^{12 13 14} By these multiple protein-protein interactions, the internal membrane cytoskeleton, mechanically coupled to the extracellular matrix, is strategically placed to strengthen the plasma membrane during muscle contraction.^{3 8 9 15 16}

Lack of dystrophin due to mutations in the dystrophin gene is thought to reduce the capacity of the plasma membrane to withstand the mechanical forces imposed by repeated muscle contraction, leading to damage, progressive necrosis, and degeneration.^{1 4 16 17 18 19 20 21} In the case of Duchenne muscular dystrophy and a subset of familial dilated cardiomyopathies, the consequences to the patient of this deficiency are ultimately fatal. Apart from defects in dystrophin, mutations in the genes for -, ß-, -, and -sarcoglycans are associated with limb girdle muscular dystrophy types 2D, 2E, 2C, and 2F, respectively,^{22 23 24 25} and a mutation in the gene for -sarcoglycan leads to an autosomal recessive cardiomyopathy in the hamster.²⁶ In addition, deficiency of merosin (laminin-2), a component of the basal lamina, is associated with classical congenital muscular dystrophy in humans and dy^2J dystrophy in mice.²⁷

An understanding of the molecular pathogenesis of these diseases and the mechanism by which plasma membrane integrity is maintained in healthy muscle depends on detailed knowledge of the precise nature of the interactions between the individual components of the dystrophin-glycoprotein complex. Biochemical studies on skeletal muscle initially suggested that dystrophin links to ß-dystroglycan via a 59-kD dystrophin-associated protein, 59-DAP,² but subsequent studies indicated the interaction between dystrophin and ß-dystroglycan to be direct.^{28 29} By applying multiple immunogold labeling of skeletal muscle sections, Wakayama et al³⁰ succeeded in providing a direct demonstration of a strikingly close association between the C-terminal domains of dystrophin and ß-dystroglycan at the ultrastructural level and also of these two proteins with -sarcoglycan,³¹ in keeping with the current biochemical evidence from studies on the isolated molecules.

However, compared with the extensive investigations conducted in skeletal muscle, few studies have examined the interactions between components of the dystrophin-glycoprotein complex in cardiac muscle. The members of the complex have been shown by immunofluorescence microscopy to colocalize with dystrophin and with laminin in cardiac muscle,^{32 33} suggesting an overall organization similar to that of skeletal muscle, but the resolution of this approach does not permit the study of the interactions between individual components of the complex. In the present study, we therefore set out to apply high-resolution double–immunogold labeling techniques to permit simultaneous visualization of the C-terminal domains of dystrophin and ß-dystroglycan in rat cardiac muscle. A key feature of our approach was to apply an ultrastructural labeling technique, fracture-label, which permits visualization of the spatial relationship between the two labeled components in planar views of the cell at the level of the plasma membrane.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

 
Sources of Tissue
Fifteen male Sprague-Dawley rats of 300 g body weight were used. The animals were preanesthetized by intraperitoneal injections of Hypnorm (0.315 mg/mL fentanyl citrate and 10 mg/mL fluanisone at 0.5 mL/kg body weight) 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 and then removed from the animal for further processing. The procedures were conducted according to the Animals (Scientific Procedures) Act, 1986, under license from the Home Office. For cryosectioning (and subsequent immunoconfocal microscopy), half-ventricle slices were frozen in isopentane cooled with liquid nitrogen; for fracture-label electron microscopy, tissue blocks of 3 to 5 mm³ were cryoprotected with 30% glycerol (PBS-buffered) for 2 hours before the mounting and freezing procedures described below.

Antibodies and Detection Systems
Two primary antibodies were used for labeling dystrophin, a mouse monoclonal antibody (DY8/6C5) and a rabbit polyclonal antibody (P1583), both raised against the last 17 amino acids of the dystrophin C-terminal domain. For labeling ß-dystroglycan (formerly termed 43-Dag), we used a mouse monoclonal antibody (43 Dag1/8D5) produced against the last 16 amino acids of the C-terminal domain.⁴ Western blots confirmed the specificity of these antibodies, with DY8/6C5 and P1583 recognizing a single band at 400 kD (ie, dystrophin) and 43 Dag1/8D5 recognizing a single band at 43 kD (ie, ß-dystroglycan) in cardiac and skeletal muscle homogenates (Fig 1). In control experiments on labeling specificity and resolution, we also used a rabbit polyclonal antibody (HJ) raised against amino acid residues 131 to 142 of connexin43 to label gap junctions.^{34 35}

View larger version (45K): [in this window] [in a new window]   Figure 1. The specificity of the antibodies for ß-dystroglycan and dystrophin is demonstrated in these Western blots. Homogenates of rat left ventricle and skeletal muscle were run on 6.5% and 12% SDS-polyacrylamide gels and transferred to polyvinylidene difluoride membranes. The membrane from the 6.5% gel was incubated with mouse monoclonal anti–ß-dystroglycan (43 Dag1/8D5), and that from the 12% gel was incubated with mouse monoclonal dystrophin (DY8/6C5), followed in both cases by alkaline phosphatase–conjugated secondary antibodies. Prominent bands are observed for ß-dystroglycan at 43 kD and for dystrophin at 400 kD.

The secondary antibody/detection systems used for immunoconfocal microscopy were (1) Cy3-conjugated goat anti-mouse (Chemicon International) and (2) goat anti-rabbit FITC (Dako). For fracture-label immunogold electron microscopy, the detection systems were (1) biotinylated goat anti-mouse immunoglobulin followed by 10-nm gold-streptavidin (Amersham Life Sciences) and (2) goat anti-rabbit 5-nm gold complexes (British BioCell Intl).

Immunofluorescent Labeling and Confocal Microscopy
For immunoconfocal microscopy, 10-µm cryosections were cut at -25°C and thaw-mounted onto poly-L-lysine–coated glass slides. The sections were treated with 0.3% Triton X-100 for 15 minutes to improve permeability to the reagents and blocked with 0.5% BSA at room temperature. For single labeling, the sections were then incubated overnight with primary antibody followed by the appropriate anti-mouse or anti-rabbit secondary antibody. The anti–ß-dystroglycan antibody was used at a concentration of 1:50; both anti-dystrophin antibodies were used at 1:1000. The mouse monoclonal primary antibodies DY8/6C5 (anti-dystrophin) and 43 Dag1/8D5 (anti–ß-dystroglycan) were followed by Cy3 anti-mouse antibody (1:500) for 1 hour; the rabbit polyclonal antibody P1583 (anti-dystrophin) was followed by anti-rabbit FITC (1:20) for 1 hour. For double labeling, sections were exposed sequentially to (1) polyclonal anti-dystrophin overnight, (2) monoclonal anti–ß-dystroglycan for 4 hours, (3) anti-rabbit FITC for 1 hour, and (4) Cy3 anti-mouse (1:500) for 1 hour. All sections were mounted using Citifluor mounting medium. The following controls were run in parallel: (1) omission of primary antibody, (2) switching of detection systems in single-labeling experiments (eg, using mouse monoclonal followed by anti-rabbit secondary antibodies), and (3) reversing the order of primary antibodies in the double-labeling procedure.

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 Cy3 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 according to the procedure of Pinto da Silva et al,³⁶ using a modification designed to increase the incidence of fractures along the plasma membranes.³⁷ The principle of this modification is to mount slices of tissue between a pair of supports such that the long axes of the myocytes are oriented to lie parallel with the supports; on freeze-fracture, the fracture is then readily generated in the same plane as the lateral surfaces of the myocytes. Samples of the perfuse-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 manually fractured under liquid nitrogen using a precooled razor blade 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. All the above solutions were buffered with PBS.

Primary antibody treatment for single and double labeling of dystrophin and ß-dystroglycan was carried out as described for confocal microscopy and was followed by corresponding mouse or rabbit secondary antibodies coupled to colloidal gold (concentration, 1:50; 1 hour at room temperature). Gold markers of 10-nm diameter were used in single-labeling experiments, and markers of 10- and 5-nm diameter were used in double-labeling experiments. In double-labeling experiments, each primary antibody treatment was followed by its corresponding secondary detection system [eg, the following sequence: (1) polyclonal dystrophin C-terminal domain antibody, (2) anti-rabbit gold, (3) monoclonal ß-dystroglycan, and (4) biotinylated anti-mouse/streptavidin gold]. The experiments were conducted using 5-nm gold followed by 10-nm gold, and vice versa. The following controls were run in parallel: (1) omission of primary antibody and (2) one primary with both secondaries. Cross-fractured cells in positively labeled samples served as internal controls. As a further control to the specificity and resolution of detection within the membrane, samples were double-labeled for ß-dystroglycan and the gap-junctional protein, connexin43, for examination by the SDS-digested freeze-fracture replica technique (see next section).

The fracture-label specimens were rinsed in PBS, postfixed in 2.5% glutaraldehyde for 30 minutes, further rinsed in PBS, and processed for examination by 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.³⁸ The replicas were carefully cleaned in sodium hypochlorite such that the biological material was removed without dislodging the gold label. After they were rinsed in water, the cleaned replicas were mounted on grids for electron microscopy. Thin sections and replicas of fracture-label specimens were examined in a Philips EM301 electron microscope.

SDS–Freeze-Fracture Replica Labeling: Detection of ß-Dystroglycan and Connexin43
SDS-digested freeze-fracture replicas were prepared and labeled as described by Fujimoto.³⁹ In this freeze-fracture cytochemical technique, conventional freeze-fracture replicas are first prepared; the biological material is then digested using SDS. The SDS removes the bulk of the biological material, leaving a fine layer of proteins adherent to the replica, which may then be localized in the plane of the membrane using immunogold labeling. Small samples of unfixed rat ventricle were briefly (<15 minutes) treated with 30% glycerol, mounted in gold well holders, and rapidly frozen in liquid nitrogen slush. The frozen specimens were fractured by knife, and platinum/carbon replicas were prepared at -120°C and in a vacuum better than 10^-6 mbar in a Balzers BAF 400T freeze-fracture unit. The replicas were floated off their holders in PBS and transferred to 2 mL of 2.5% SDS (Sigma Chemical Co) containing 10 mmol/L Tris-HCl and 30 mmol/L sucrose, pH 8.3. SDS digestion was carried out overnight at room temperature. The replicas were washed for 1 hour with four changes of PBS and, before labeling, blocked with 1% BSA at room temperature for 30 minutes. Double labeling was then carried out with anti–ß-dystroglycan antibody overnight, followed by anti-connexin43 (1:500) for 1 hour. The specimens were then washed three times in PBS, followed by a 1-hour incubation in a mixture of goat anti-mouse 15-nm gold and goat anti-rabbit 10-nm gold. Thorough rinsing (five washes in PBS) was followed by a 3-minute postfixation in 1% glutaraldehyde. The labeled replicas were finally floated on distilled water and picked up on copper 460 mesh grids for electron microscopical examination.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

 
Immunolabeling studies were conducted by three comple-mentary techniques: (1) immunofluorescence confocal laser scanning microscopy, (2) thin-section examination of im-munogold-labeled freeze-fractured specimens (fracture-label thin-section technique), and (3) platinum-carbon replicas of immunogold-labeled freeze-fractured specimens (fracture-label/replica technique).

Immunoconfocal Localization
Antibodies against the dystrophin and ß-dystroglycan C-termini gave clear localization patterns by confocal microscopy (Figs 2A and 2B). When viewed individually, both the dystrophin and ß-dystroglycan C-terminal domains were labeled in a uniform manner along the cell surface with intense and continuous fluorescence except for conspicuous gaps at the intercalated disks (Fig 2A and 2B). No preferential labeling of costameres was apparent, as reported previously for dystrophin in rat cardiac muscle.³⁷ Apart from cell surface labeling, positive but less pronounced signal was observed in the form of discontinuous striations penetrating into the cell, which represent transverse tubules. To determine the spatial relationship between the two labeled components, visualization was performed using dual-channel imaging of double-labeled sections (Fig 2C). These images revealed complete coincidence of signal, indicating that at the resolution attainable by confocal microscopy, the dystrophin and ß-dystroglycan C-termini were colocalized. The pattern of label observed in both double- and single-labeling preparations was constant irrespective of antibody incubation time and concentration.

View larger version (68K): [in this window] [in a new window]   Figure 2. Immunoconfocal localization of dystrophin and ß-dystroglycan by single- and dual-channel scanning of double-label preparations. Independent immunolocalization of dystrophin and of ß-dystroglycan is shown in panels A and B, respectively. A continuous distribution of both proteins is seen at the peripheral plasma membrane of the cells in both longitudinal and transverse sectional views. Weaker labeling penetrating deep into the cells is seen marking the positions of T tubules. In panel C, by combining the two images, a precise superimposition is apparent (yellow), indicating colocalization of dystrophin and ß-dystroglycan at the resolution of confocal microscopy. This colocalization is apparent both at the cell surface plasma membrane and in the T-tubular system within the cells. Bar=25 µm.

Immunogold Fracture-Label Electron Microscopy: Thin-Section Examination
To gain more detailed information on the spatial relationship between the dystrophin and ß-dystroglycan C-termini and to determine whether these two components are colocalized at the ultrastructural level, single- and double-labeled specimens were examined by immunogold fracture-label and thin-section electron microscopy. Thin sectioning revealed that labeling for both dystrophin and ß-dystroglycan C-termini, as observed in single-label experiments, was uniformly distributed along the cell surface membrane, with no preferential association to specific bands of the sarcomeres of subjacent myofibrils (ie, no concentration at costameres) (Fig 3A). In both cases, labeling was predominantly associated with the plasma membrane P half (protoplasmic side), with only a trace of label present on the E half (extracellular side) (Fig 3B). Cross fractures revealed that no labeling of either protein was detectable within the cell.

View larger version (136K): [in this window] [in a new window]   Figure 3. Thin-section fracture-label specimens showing immunogold localization of ß-dystroglycan. A, Single labeling for ß-dystroglycan is shown in a cell in which the fracture has followed the plasma membrane, leaving the P half of the membrane and the protoplasmic side of the cell for viewing. The labeling is continuous at the level of the plasma membrane, with no apparent specificity for any region of the sarcomere. B, In views of the plasma membrane E half, ß-dystroglycan labeling is sparse (arrowheads). This aspect of the plasma membrane is readily identified: part of the cross-fractured cytoplasm of the same cell remains visible to the right of the field (asterisk). Similar results were obtained on single labeling for dystrophin. Bars=250 nm.

Results for double immunogold labeling of the dystrophin and ß-dystroglycan C-termini examined by the thin-section fracture-label technique are illustrated in Figs 4A and 4B. Two distinct patterns of labeling that were found to be directly dependent on the detection procedure were observed. When the dystrophin was detected first using 10-nm gold and ß-dystroglycan was detected second using 5-nm gold, very little 5-nm gold was present, with 81% of the label comprising the larger gold (Fig 4A). However, when the dystrophin was detected first using 5-nm gold and ß-dystroglycan was detected second using 10-nm gold, both sizes of gold label were present in the proportion of 2:3 (small gold:large gold) (Fig 4B). In these preparations, much of the 10-nm and 5-nm gold labels appeared widely and individually distributed, but in some instances there was an extremely close association between the two in the form of 10-nm/5-nm gold "doublets" (Fig 4B inset). This last result provides a possible explanation for the seemingly contradictory results obtained when switching the order of the two secondary gold labels. If, as suggested in Fig 4B, the dystrophin and ß-dystroglycan C-terminal domains are extremely closely associated, then steric hindrance might contribute to the labeling patterns observed. Use of secondary 10-nm gold as the first detection step might well block accessibility of probes for subsequent labeling of ß-dystroglycan with 5-nm gold. On the other hand, use of 5-nm gold for the first detection step, although also potentially interfering with subsequent detection of the second epitope, might be expected to do so to a lesser extent.

View larger version (130K): [in this window] [in a new window]   Figure 4. Double-label preparations in which the distribution of ß-dystroglycan and dystrophin are visualized simultaneously. A, Dystrophin has been labeled first with 10-nm gold followed by ß-dystroglycan labeled second with 5-nm gold. B, The two sizes of the gold label have been used in the reverse order (ie, dystrophin/5-nm gold followed by ß-dystroglycan/10-nm gold). In both images, the smaller gold label is indicated by arrowheads as an aid to identification. Note that the incidence of 5-nm gold is higher when the small gold is applied before (B) rather than after (A) the large gold. Doublets consisting of closely apposed small and large gold markers (inset, arrow pairs) are present with the former protocol, although many individually distributed large and small gold markers are also present. Bars=250 nm.

At first sight, such an explanation might appear inconsistent with the relative sparsity of closely associated 10-nm and 5-nm gold complexes (Fig 4B inset) in the dystrophin/5-nm gold followed by ß-dystroglycan/10-nm gold procedure. However, in the two-dimensional views provided by thin sections, such associations, even if relatively common, would be seen only in occasional instances of favorable viewing angle (Fig 5). To test this hypothesis requires viewing of the label from above the plasma membrane, and it was for this reason that we applied the platinum/carbon replica method of visualization for fracture label.

View larger version (19K): [in this window] [in a new window]   Figure 5. Diagram illustrating the technical problems of viewing doublets comprising one small and one large gold particle in thin sections. If two closely associated components are labeled with two sizes of gold, when viewed from position A, the large gold particle conceals the presence of the small gold particle. The same would apply from a range of viewing positions around position A and, similarly, if a small gold particle is situated in front of the large gold particle. Only when the pair of gold particles falls along a plane at right angles to the position of viewing are both gold particles clearly visualized (B).

Immunogold Fracture-Label Electron Microscopy: Platinum/Carbon Replica Examination
Replicas of the plasma membrane after single labeling for ß-dystroglycan revealed a widely dispersed scattered distribution of label (Fig 6), consistent with the thin-section views (Fig 4A). Some plasma membrane views revealed domains devoid of label that differed in texture from the major expanses of membrane viewed (Fig 6A). These domains are attributed to the reorganization of the fractured surface on exposure to aqueous media at the thawing stage. It should be emphasized that the presence of these structures did not significantly interfere with obtaining clear extensive views of the label distribution pattern (Fig 6B).

View larger version (103K): [in this window] [in a new window]   Figure 6. ß-Dystroglycan localization at the level of the plasma membrane as viewed in fracture-label replicas. The ß-dystroglycan is widely and irregularly distributed over the exposed surface, showing no preferential association with the positions of the sarcomeric bands of underlying myofibrils. In panel A the labeling distribution is less uniform than in panel B, with distinct patches devoid of label (asterisks). Bars=500 nm.

Replicas of specimens processed for double labeling of dystrophin and ß-dystroglycan C-terminal domains (Fig 7) revealed a strikingly high incidence of colocalized pairs of 5- and 10-nm gold markers (doublets), as predicted above. This effect was apparent both in samples exposed to 5-nm gold probes first (Fig 7A) and in those exposed to 10-nm gold probes first (Fig 7B). Apart from the presence of the 5-nm/10-nm gold "doublets," the former group showed a greater incidence of individually dispersed 5-nm gold markers, whereas the latter showed a higher incidence of individually dispersed 10-nm gold markers.

View larger version (114K): [in this window] [in a new window]   Figure 7. Double immunogold localization of ß-dystroglycan and dystrophin in fracture-label preparations examined in platinum-carbon replicas. In panel A, dystrophin was labeled first using 5-nm gold followed by ß-dystroglycan labeled second using 10-nm gold (circled), whereas in panel B, dystrophin was labeled first using 10-nm gold followed by ß-dystroglycan labeled second using 5-nm gold. Doublets are common with both procedures but especially when the 5-nm marker is used first (panel A). Of the remaining label, 5-nm gold predominates in panel A, and 10-nm gold predominates in panel B. Inset at the top left of panel B shows detail of doublets (arrow pairs) from the boxed area from the same panel. Bars=250 nm.

Immunogold Labeling Controls
Thin sections and replicas of controls in which the primary antibodies were omitted consistently revealed no or very little gold labeling. When samples were exposed to an individual primary antibody followed by both secondaries, only labeling with the relevant secondary antibody was observed. Interaction of the secondary antibody complexes with one another was excluded using this approach. Cross-fractured myocytes served as internal negative controls, showing virtually no labeling, as illustrated in Fig 8A. Adjacent regions of plasma membrane fractures in the same samples revealed extensive labeling as described above, confirming specificity of the primary antibody for membrane components. That the labeling within membrane domains was specific and of high spatial resolution was confirmed by the SDS-digested freeze-fracture replica double-labeling experiments (Fig 8B). These controls demonstrated specific labeling of gap-junctional plaques using connexin43 antibody, with ß-dystroglycan in a dispersed distribution confined to surrounding (nonjunctional) membrane. The SDS-digested freeze-fracture labeling technique was used for these experiments rather than the fracture-label technique to provide a sufficiently clear visualization of the gap-junctional plaques.

View larger version (112K): [in this window] [in a new window]   Figure 8. A, Fracture-label replica of a cross-fractured myocyte from the same experiment as in Fig 7. This example serves as a negative control, revealing that no label is present where the fracture plane has exposed the myofibrils within the cells. Positive label is thus specifically associated with plasma membrane fractures. M, A, and Z/I indicate sarcomeric bands of the myofibril. B, Double immunogold localization of ß-dystroglycan and connexin43 in SDS-digested freeze-fracture replicas. This example serves as a positive control, revealing connexin43 labeling of gap-junctional plaques (gj), with ß-dystroglycan labeling confined to the nonjunctional plasma membrane, with no label present in the cytoplasm (Cyt). Bars=250 nm.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

 
Interaction between dystrophin and the transmembrane dystroglycan/sarcoglycan complex is critical to the coupling of the cytoskeleton to the extracellular matrix in muscle cells. By applying double immunogold electron microscopy with complementary dual-channel scanning immunoconfocal microscopy, the present study sought to establish the spatial relationship between the C-terminal domains of dystrophin and ß-dystroglycan at the plasma membrane of cardiac muscle. To achieve this goal, we applied the fracture-label technique,³⁶ one of a highly effective but underused set of approaches in freeze-fracture cytochemistry.⁴⁰ The fracture-label technique permits high-resolution visualization of labeled membrane components both in sections and in platinum-carbon replicas of the fractured tissue interface. For the purpose of the present study, the en face replica approach proved particularly informative, providing unique information on the spatial relationship of simultaneously immunogold-localized C-terminal domains of dystrophin and ß-dystroglycan in a plane parallel with the plasma membrane surface, an aspect that cannot readily be reconstructed from sectional views.

A key feature of the fracture-label technique is that cytochemical labeling is carried out after membranes in the sample have been split by freeze fracture and the sample has been thawed; thus, in principle, the label has direct access to the entire face-on aspects of suitably fractured plasma membranes. Both integral membrane components and associated peripheral proteins are rendered accessible for labeling by this procedure, since on contact with aqueous media at the thawing stage, the fractured half-membrane leaflets become reorganized into a discontinuous bilayer, thereby exposing associated cytoplasmic or extracellular components.⁴¹ This feature makes the fracture-label technique particularly well suited to gaining access to proteins such as dystrophin and the cytoplasmic domain of ß-dystroglycan, which lie just beneath the inner surface of the plasma membrane. Because of this location, these components are detected in association with the plasma membrane P half rather than the E half, as observed in the present study and reported in an earlier fracture-label study localizing the C-terminal domain of skeletal muscle dystrophin.⁴²

Current models of the dystrophin-glyoprotein complex based on biochemical studies conducted on skeletal muscle now agree that the component of the transmembrane dystroglycan complex with which dystrophin interacts is ß-dystroglycan.^{7 8 9 21 43 44} The dystrophin molecule has four domains, the N-terminal domain (linked to F-actin), a large triple helical spectrin-like domain, a cysteine-rich domain, and the C-terminal domain. A number of studies have identified the cysteine-rich domain and the first half of the C-terminus as the site that interacts with a cytoplasmic segment of the C-terminal portion of the ß-dystroglycan molecule.^{13 14} Recent in vitro experiments using dystrophin and ß-dystroglycan from skeletal muscle and brain have specifically delineated the 15 C-terminal amino acids of ß-dystroglycan as a unique binding site for the second half of hinge 4 and the cysteine-rich domain of dystrophin.²⁸ Evidence that these deductions from the study of isolated molecules apply to the intact skeletal muscle tissue comes from ultrastructural cytochemical localization of the C-terminal domains of dystrophin and ß-dystroglycan (previously termed 43-Dag) in sections; double immunogold labeling revealed a significant incidence of doublets, ie, pairs of closely associated gold markers containing one of each size of the gold particles used to differentiate the two components.³⁰ In skeletal muscle, similar doublets have been shown using antibodies against the rod domain of dystrophin and the C-terminal domain of ß-dystroglycan, and both these components have been shown to be closely associated with -sarcoglycan.³¹

Our demonstration of similar gold marker doublets, representing side-by-side antibody labeling of the C-terminal domains of dystrophin and ß-dystroglycan, supports the conclusion that these two components directly interact at the cardiac muscle cell plasma membrane in a manner similar to that in skeletal muscle. Our ability to demonstrate this close association depended on two critical factors, the specific epitopes to which our antibodies were directed and the ability to observe the plasma membrane from above. We reasoned that an antibody directed against a site immediately adjacent to rather than directly at the putative site of interaction would avoid complete blockage of the second labeling step through steric hindrance. Thus, we used an antibody to the C-terminal of dystrophin rather than to the cysteine-rich and hinge 4 region. With the optimal sequence of applying the different-sized gold markers, doublets similar to those reported in skeletal muscle by Wakayama et al³⁰ were demonstrable on sections, but their incidence was low. A key factor in visualizing a doublet on section is a favorable viewing angle, in which the two sizes of gold stand side by side, at right angles to the direction of view. Application of the fracture-label replica technique enabled us to overcome this limitation by making it possible to look at the doublets from above.

Recently, it has been shown that the second half of the C-terminus of dystrophin, specifically those amino acid residues encoded by exons 73 to 74, bind ₁- and ß₁-syntrophin, which are cytoplasmic proteins of the glycoprotein complex.⁴⁵ The epitope that was labeled in the present study, encoded by exon 79, lies beyond the binding site, which suggests that the tip of dystrophin folds back toward the dystroglycan binding site or that the syntrophins are of a shape that does not prevent near apposition of the tips of the molecules of dystrophin and ß-dystroglycan.

Because of the key role of the dystrophin/glycoprotein complex in integration of the internal cytoskeleton with the extracellular matrix, the organization and interaction of these components is relevant to the understanding of mechanisms of mechanical dysfunction at the cellular level in cardiomyopathic diseases. In Duchenne muscular dystrophy, the loss of dystrophin is as complete in cardiac muscle as it is in skeletal muscle,^{32 46} but cardiomyopathy, though detectable by imaging or electrical studies, is only rarely clinically apparent.^{47 48 49} In Becker muscular dystrophy, by contrast, where reduced levels or semifunctional forms of dystrophin are expressed, clinical symptoms affecting skeletal muscle are usually comparatively mild, but cardiac involvement can be more severe than in Duchenne dystrophy.^{47 48} In a subset of familial dilated cardiomyopathies that show X-linked inheritance, mutations affecting both cardiac and skeletal muscle dystrophin result in rapidly progressive and fatal heart failure with no or only relatively minor clinical signs of skeletal muscle involvement.^{50 51 52 53} Since it has now been established that, in addition to dystrophin, defects in the dystrophin-associated proteins contribute to the pathogenesis of muscular dystrophies,^{22 23 24 25 26 54} the potential roles of these proteins in mechanical dysfunction of the diseased heart deserve further investigation.

	Acknowledgments

 
This study was supported by grant PG/97121 from the British Heart Foundation (Dr Severs). Dr Cullen is supported by the Wellcome Trust (grant 046045). We thank Dr Louise Anderson (University of Newcastle on Tyne) for the gift of monoclonal antibodies, Dr Henry Klamut (Ontario Cancer Institute, Toronto, Canada) for the polyclonal antibody P1583, and Dr Steven Coppen for his help with the Western blots.

Received July 23, 1997; accepted October 8, 1997.

	References

Top Abstract Introduction Materials and Methods Results Discussion References

 

1. Ervasti JM, Ohlendieck K, Kahl SD, Gaver MG, Campbell KP. Deficiency of a glycoprotein component of the dystrophin complex in dystrophic muscle. Nature. 1990;345:315–319.[Medline] [Order article via Infotrieve]
2. Ervasti JM, Campbell KP. Membrane organisation of the dystrophin-glycoprotein complex. Cell. 1991;66:1121–1131.[Medline] [Order article via Infotrieve]
3. Campbell KP. Three muscular dystrophies: loss of cytoskeleton-extracellular matrix linkage. Cell. 1995;80:675–679.[Medline] [Order article via Infotrieve]
4. Ibraghimov-Beskrovnaya O, Evarsti JM, Leveille CJ, Slaughter CA, Sernett SW, Campbell KP. Primary structure of dystrophin-associated glycoproteins linking dystrophin to the extracellular matrix. Nature. 1992;355:696–702.[Medline] [Order article via Infotrieve]
5. Yoshida M, Ozawa E. Glycoprotein complex anchoring dystrophin to sarcolemma. J Biochem. 1990;108:748–752.[Abstract/Free Full Text]
6. Campbell KP, Kahl SD. Association of dystrophin and an integral glycoprotein. Nature. 1989;338:259–262.[Medline] [Order article via Infotrieve]
7. Ohlendieck K. Towards an understanding of the dystrophin-glycoprotein complex: linkage between the extracellular matrix and the membrane cytoskeleton in muscle fibers. Eur J Cell Biol. 1996;69:1–10.[Medline] [Order article via Infotrieve]
8. Brown RH Jr. Dystrophin-associated proteins and the muscular dystrophies: A glossary. Brain Pathol. 1996;6:19–24.[Medline] [Order article via Infotrieve]
9. Tinsley JM, Blake DJ, Zuellig RA, Davies KE. Increasing complexity of the dystrophin-association protein complex. Proc Natl Acad Sci U S A. 1994;91:8307–8313.[Abstract/Free Full Text]
10. Cullen MJ, Walsh J, Nicholson LV. Immunogold localisation of the 43-kDa dystroglycan at the plasma membrane in control and dystrophic human muscle. Acta Neuropathol. 1994;87:349–354.[Medline] [Order article via Infotrieve]
11. Deyst KA, Bowe MA, Leszyk JD, Fallon JR. The -dystroglycan-ß-dystroglycan complex. J Biol Chem. 1995;270:25956–25959.[Abstract/Free Full Text]
12. Ervasti JM, Campbell KP. A role for the dystrophin-glycoprotein complex as a transmembrane linker between laminin and actin. J Cell Biol. 1993;122:809–823.[Abstract/Free Full Text]
13. Suzuki A, Yoshida M, Yamamoto H, Ozawa E. Glycoprotein-binding site of dystrophin is confined to the cysteine-rich domain and the first half of the carboxy-terminal domain. FEBS Lett. 1992;308:154–160.[Medline] [Order article via Infotrieve]
14. Suzuki A, Yoshida M, Hayashi K, Mizuno Y, Hagiwara Y, Ozawa E. Molecular organisation at glycoprotein-complex-binding site of dystrophin: three dystrophin-associated proteins bind directly to the carboxy-terminal portion of dystrophin. Eur J Biochem. 1994;220:283–292.[Medline] [Order article via Infotrieve]
15. Ganote CE, Armstrong S. Ischaemia and the myocyte cytoskeleton. Cardiovasc Res. 1993;27:1387–1403.[Free Full Text]
16. Hoffman EP. Clinical and histopathological features of abnormalities of the dystrophin-based membrane cytoskeleton. Brain Pathol. 1996;6:49–61.[Medline] [Order article via Infotrieve]
17. Matsumura K, Tome FMS, Ionasescu V, Ervasti JM, Anderson RD, Romero NB, Simon D, Recan D, Kaplan J, Fardeau M. Deficiency of dystrophin-associated proteins in Duchenne muscular dystrophy patients lacking COOH terminal domains of dystrophin. J Clin Invest. 1993;92:866–871.
18. Ohlendieck K, Campbell KP. Dystrophin-associated proteins are greatly reduced in skeletal muscle from mdx mice. J Cell Biol. 1991;115:1685–1694.[Abstract/Free Full Text]
19. Ohlendieck K, Matsumura K, Ionasescu VV, Towbin JA, Bosch EP, Weinstein SL, Sernett SW, Campbell KP. Duchenne muscular dystrophy: deficiency of dystrophin-associated proteins in the sarcolemma. Neurology. 1993;43:795–800.[Abstract/Free Full Text]
20. Petrof BJ, Shrager JB, Stedman HH, Kelly AM, Sweeney HL. Dystrophin protects the sarcolemma from stresses developed during muscle contraction. Proc Natl Acad Sci U S A. 1993;90:3710–3714.[Abstract/Free Full Text]
21. Nawrotzki R, Blake DJ, Davies KE. The genetic basis of neuromuscular disorders. Trends Genet. 1996;12:294–298.[Medline] [Order article via Infotrieve]
22. Nigro V, de Sa Moreira E, Piluso G, Vainzof M, Belsito A, Politano L, Puca AA, Passos-Bueno MR, Zatz M. Autosomal recessive limb-girdle muscular dystrophy LGMD2F, is caused by a mutation in the -sarcoglycan gene. Nat Genet. 1996;1:49–58.
23. Jung D, Leturcq F, Sunada Y, Duclos F, Tomé FMS, Moomaw C, Merlini L, Azibi K, Chaouch M, Slaughter C, Fardeau M, Kaplan JV, Campbell KP. Absence of gamma-sarcoglycan (35DAG) in autosomal recessive muscular dystrophy linked to chromosome 13q12. FEBS Lett. 1996;381:15–20.[Medline] [Order article via Infotrieve]
24. Vainzof M, Passos-Bueno MR, Canovas M, Sa Moreira E, Pavanello RCM, Marie SK, Anderson LVB, Bonnemann CG, Mc Nally EM, Nigro V, Kunkel LM, Zatz M. The sarcoglycan complex in the six autosomal recessive limb-girdle muscular dystrophies. Hum Mol Genet. 1996;5:1963–1969.[Abstract/Free Full Text]
25. Barresi R, Confalonieri V, Lanfossi M, Di Blasi C, Torchiana E, Mantegazza R, Jarre L, Nardocci N, Boffi P, Tezzon F, Pini A, Cornelio F, Mora M, Morandi L. Concomitant deficiency of ß-and -sarcoglycans in 20 -sarcoglycan (adhalin)-deficient patients: immunohistochemical analysis and clinical aspects. Acta Neuropathol. 1997;94:28–35.[Medline] [Order article via Infotrieve]
26. Nigro V, Okazaki Y, Belsito A, Matsuda Y, Politano G., Nigro C, Ventura C, Abbondanza AM, Molinari D, Acampora M, Nishimura Y, Hayashizaki Y, Puca GA. Identification of the Syrian hamster cardiomyopathy gene. Hum Mol Genet. 1997;6:601–607.[Abstract/Free Full Text]
27. Williamson RA, Henry MD, Daniels KJ, Hrstka RF, Lee JC, Sunada Y, Ibraghimov-Beskrovnaya O, Campbell KP. Dystroglycan is essential for early embryonic development: disruption of Reichert's membrane in Dag1-null mice. Hum Mol Genet. 1997;6:831–841.[Abstract/Free Full Text]
28. Jung D, Yang B, Meyer J, Chamberlain JS, Campbell KP. Identification and characterisation of the dystrophin anchoring site on ß-dystroglycan. J Biol Chem. 1995;270:27305–27310.[Abstract/Free Full Text]
29. Rosa G, Ceccarini M, Cavaldesi M, Zini M, Petrucci T. Localisation of the dystrophin binding site at the carboxy terminus of ß-dystroglycan. Biochem Biophys Res Commun. 1996;223:272–277.[Medline] [Order article via Infotrieve]
30. Wakayama Y, Shibuya S, Takeda A, Jimi T, Nakamura Y, Oniki H. Ultrastructural localisation of the C-terminus of the 43-kd dystrophin-associated glycoprotein and its relation to dystrophin in normal murine skeletal myofiber. Am J Pathol. 1995;146:189–196.[Abstract]
31. Inoue M, Wakayama Y, Murahashi M, Shibuya S, Takahiro J, Kojima H, Oniki H. Electron microscopic observations of triple immunogold labelling for dystrophin, ß-dystroglycan and adhalin in human skeletal myofibres. Acta Neuropathol. 1996;92:569–575.[Medline] [Order article via Infotrieve]
32. Arahata K, Ishiura T, Tsukahara T, Suhara Y, Eguchi C, Ishihara T, Nonaka I, Ozawa E, Sugita H. Immunostaining of skeletal and cardiac muscle surface membrane with antibody against Duchenne muscular dystrophy peptide. Nature. 1988;333:861–863.[Medline] [Order article via Infotrieve]
33. Klietsch R, Ervasti JM, Arnold W, Campbell KP, Jorgensen AO. Dystrophin-glycoprotein complex and laminin colocalize to the sarcolemma and transverse tubules of cardiac muscle. Circ Res. 1993;72:349–360.[Abstract/Free Full Text]
34. Green CR, Severs NJ. Distribution and role of gap junctions in normal myocardium and human ischaemic heart disease. Histochemistry. 1993;99:105–120.[Medline] [Order article via Infotrieve]
35. Severs NJ. Cardiac muscle cell interaction: from microanatomy to the molecular make-up of the gap junction. Histol Histopathol. 1995;10:481–501.[Medline] [Order article via Infotrieve]
36. Pinto da Silva P, Barbosa MLF, Aguas AP. A guide to fracture-label: cytochemical labeling of freeze-fractured cells. In: Koehler JK, ed. Advanced Techniques in Biological Electron Microscopy. New York, NY: Springer Verlag; 1986;3:201–227.
37. Stevenson S, Rothery S, Cullen MJ, Severs NJ. Dystrophin is not a specific component of the cardiac costamere. Circ Res. 1997;80:269–280.[Abstract/Free Full Text]
38. Pinto da Silva P, Kachar B, Torrisi MR, Brown C, Parkison C. Freeze-fracture cytochemistry: replicas of critical point-dried cells and tissues after fracture-label. Science. 1981;213:230–233.[Abstract/Free Full Text]
39. Fujimoto K. Freeze-fracture replica electron microscopy combined with SDS digestion for cytochemical labeling of integral membrane proteins: application to the immunogold labeling of intercellular junctional complexes. J Cell Sci. 1995;108:3443–3449.[Abstract]
40. 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.
41. Pinto da Silva P, Parkison C, Dwyer N. Freeze-fracture cytochemistry: thin sections of cells and tissues after labeling of fracture faces. J Histochem Cytochem. 1981;29:917–928.[Abstract]
42. Squarzoni S, Sabatelli P, Maltarello MC, Cataldi A, Di Primio R, Maraldi NM. Localization of dystrophin COOH-terminal domain by the fracture-label technique. J Cell Biol. 1992;118:1401–1409.[Abstract/Free Full Text]
43. Ahn AH, Kunkel LM. The structural and functional diversity of dystrophin. Nat Genet. 1993;3:283–291.[Medline] [Order article via Infotrieve]
44. Campbell KP, Crosbie RH. Utrophin to the rescue. Nature. 1996;384:308–309.[Medline] [Order article via Infotrieve]
45. Suzuki A, Yoshida M, Ozawa E. Mammalian 1-and ß1-syntrophin bind to alternative splice-prone region of the dystrophin COOH terminus. J Cell Biol. 1995;128:373–381.[Abstract/Free Full Text]
46. Hoffman EP, Hudecki MS, Rosenberg PA, Pollina CM, Kunkel LM. Cell and fiber-type distribution of dystrophin. Neuron. 1988;1:411–420.[Medline] [Order article via Infotrieve]
47. Emery AEH. Duchenne Muscular Dystrophy. Oxford, England: Oxford Medical Publications; 1993:24. Oxford Monographs on Medical Genetics.
48. Carter LF, Rubin SA. The molecular and cellular biology of heart failure. Curr Opin Cardiol. 1994;9:264–271.[Medline] [Order article via Infotrieve]
49. Hunsaker RH, Fulkerson PK, Barry FJ, Lewis RP, Lefier CV, Unverferth DV. Cardiac function in Duchenne's muscular dystrophy: result of 10-year follow-up study and non-invasive test. Am J Med. 1982;73:235–238.[Medline] [Order article via Infotrieve]
50. Franz WM, Cremer M, Herrmann R, Grünig E, Fogel W, Scheffold T, Goebel HH, Kircheisen R, Kübler W, Voit T, Katus HA. X-linked dilated cardiomyopathy: novel mutation of the dystrophin gene. Ann N Y Acad Sci. 1995;752:470–491.[Abstract]
51. Muntoni F, Wilson L, Marrosu MG, Cianchetti C, Mestroni L, Ganau A, Dubowitz V, Sewry C. A mutation in the dystrophin gene selectively affecting dystrophin expression in the heart. J Clin Invest. 1995;96:693–699.
52. Towbin JA, Hejtmancik JF, Brink P, Gelb B, Zhu XM, Chamberlain JS, McCabe ERB, Swift M. X-linked dilated cardiomyopathy: molecular genetic evidence of linkage to Duchenne muscular dystrophy (dystrophin) gene at the Xp21 locus. Circulation. 1993;87:1854–1865.[Abstract/Free Full Text]
53. Yoshida K, Ikeda S, Nakamura A, Kagoshima M, Takeda S, Shoji S, Yanagisawa N. Molecular analysis of the Duchenne muscular dystrophy gene in patients with Becker muscular dystrophy presenting with dilated cardiomyopathy. Muscle Nerve. 1993;16:1161–1166.[Medline] [Order article via Infotrieve]
54. Brown RH. Dystrophin-associated proteins and the muscular dystrophies. Annu Rev Med. 1997;48:457–466.[Medline] [Order article via Infotrieve]