Altered membrane proteins and permeability correlate with cardiac dysfunction in cardiomyopathic hamsters

Am J Physiol Heart Circ Physiol 278: H1362 H1370, 2000. Altered membrane proteins and permeability correlate with cardiac dysfunction in cardiomyopathic hamsters YASUHIRO IKEDA, 1 MARYANN MARTONE, 2 YUSU GU, 1 MASAHIKO HOSHIJIMA, 1 ANDREA THOR, 2 SAM S. OH, 1 KIRK L. PETERSON, 1 AND JOHN ROSS, JR. 1 1 Division of Cardiology, Department of Medicine, and 2 Department of Neurosciences, University of California, San Diego, La Jolla, California 92093-0613 Ikeda, Yasuhiro, Maryann Martone, Yusu Gu, Masahiko Hoshijima, Andrea Thor, Sam S. Oh, Kirk L. Peterson, and John Ross, Jr. Altered membrane proteins and permeability correlate with cardiac dysfunction in cardiomyopathic hamsters. Am J Physiol Heart Circ Physiol 278: H1362 H1370, 2000. A mutation in the -sarcoglycan (SG) gene with absence of -SG protein in the heart has been identified in the BIO14.6 cardiomyopathic (CM) hamster, but how the defective gene leads to myocardial degeneration and dysfunction is unknown. We correlated left ventricular (LV) function with increased sarcolemmal membrane permeability and investigated the LV distribution of the dystrophindystroglycan complex in BIO14.6 CM hamsters. On echocardiography at 5 wk of age, the CM hamsters showed a mildly enlarged diastolic dimension (LVDD) with decreased LV percent fractional shortening (%FS), and at 9 wk further enlargement of LVDD with reduction of %FS was observed. The percent area of myocardium exhibiting increased membrane permeability or membrane rupture, assessed by Evans blue dye (EBD) staining and wheat germ agglutinin, was greater at 9 than at 5 wk. In areas not stained by EBD, immunostaining of dystrophin was detected in CM hamsters at sarcolemma and T tubules, as expected, but it was also abnormally expressed at the intercalated discs; in addition, the expression of -dystroglycan was significantly reduced compared with control hearts. As previously described, -SG was completely deficient in CM hearts compared with control hearts. In myocardial areas showing increased sarcolemmal permeability, neither dystrophin nor -dystroglycan could be identified by immunolabeling. Thus, together with the known loss of -SG and other SGs, abnormal distribution of dystrophin and reduction of -dystroglycan are associated with increased sarcolemmal permeability followed by cell rupture, which correlates with early progressive cardiac dysfunction in the BIO14.6 CM hamster. -sarcoglycan; dystrophin-dystroglycan complex; heart failure; Evans blue dye NUMEROUS STUDIES on muscular dystrophy have focused on dystrophin with its associated membrane-glycoprotein complex and their pathophysiological role (see Ref. 3 for review). The transmembrane dystrophin-glycoprotein complex (DGC) serves as a link between the intracellular cytoskeleton and the extracellular matrix, The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. connecting intracellular F-actin, dystrophin, the transmembrane sarcoglycans, and the dystroglycans ( -, -subunits) with extracellular laminin- 2. The sarcoglycan complex of dystrophin-associated glycoproteins is composed of four subunits (,,, ), sarcospan, the dystrobrevins, and the syntrophins (3, 24). Although the precise function of these components has not yet been elucidated, the main current hypothesis regarding the primary role of the DGC holds that it has a mechanical function to strengthen the plasma membrane during muscle contraction (3). Also, several lines of data support a role for dystrophin in regulating certain signal transduction pathways, along with the syntrophins and/or calmodulin (24). The function of the sarcoglycan complex is not known, although the sarcoglycan-sarcospan complex is assumed to serve as a molecular stabilizer of the DGC because mutations in any sarcoglycan subunit lead to a concomitant loss or reduction of all four sarcoglycans and sarcospan, manifested in various types of muscular dystrophy or cardiomyopathy (24). In addition, the molecular function of each subunit can differ (4); for example, deficiencies of - and -sarcoglycan are associated with a cardiomyopathy, whereas -sarcoglycan deficiency is not (5). The BIO14.6 cardiomyopathic (CM) hamster, an autosomal recessive strain, recently has been found to have a mutation in the -sarcoglycan gene (19). Sakamoto et al. (22) subsequently reported that two other strains of cardiomyopathic hamsters (UMX7.1 and TO-2) also possess a mutation in the -sarcoglycan gene. In addition, -sarcoglycan was reported to be involved in type F limb girdle muscular dystrophy (18). Deficiency of -sarcoglycan leads to a concomitant loss of the other sarcoglycan subunits (,, ), sarcospan, and -dystroglycan (25). Holt et al. (9) have recently shown that transfer of the normal human -sarcoglycan gene can restore membrane structure and function in skeletal muscle of the BIO14.6 hamster. The BIO14.6 hamster develops severe cardiomyopathy with hypertrophic features and heart failure but manifests a comparatively mild skeletal muscle phenotype. In dystrophin-deficient mdx mice, apoptosis appears to play a role in the onset of skeletal muscle degeneration (16, 27), whereas in one strain of CM hamster with heart failure (CHF147) we (8) have found that apoptosis plays a minor role compared with necrosis in the early deterioration of left ventricular function during the first 4 wk of life. Others (20) have reported a H1362 0363-6135/00 $5.00 Copyright 2000 the American Physiological Society http://www.ajpheart.org

CARDIAC DYSTROPHIN COMPLEX IN CARDIOMYOPATHIC HAMSTER H1363 correlation between the degree of muscle activity and sarcolemmal damage in skeletal muscle of mdx mice. However, the correlation between changes in sarcolemmal permeability and left ventricular (LV) function has not been quantified. Therefore, we investigated changes in cardiac function and membrane permeability, as well as the subcellular distribution of dystrophin and -dystroglycan, during early progression of the cardiomyopathy in the BIO14.6 CM hamster. MATERIALS AND METHODS Animals. Four-week-old normal male hamsters (F1B strain) and age-matched male BIO14.6 CM hamsters were obtained from BIO Breeders (Fitchburg, MA) and housed under controlled temperature and humidity conditions during the serial echocardiographic study. The experimental protocol was approved by the Animal Subjects Committee of the University of California San Diego, and all procedures were conducted in accordance with institutional guidelines. Echocardiography. Ten normal and eleven BIO14.6 CM hamsters underwent serial transthoracic echocardiographic examinations at 4, 6, 8, 10, 12, and 14 wk of age to assess LV size and function. The echocardiographic methods have been described in detail in the rat (10) and mouse (26). The CM hamsters showed normal cardiac function at 4 wk of age and subsequently gradually manifested LV dysfunction; we selected 5 and 9 wk of age to correlate changes of membrane permeability and immunostaining with echocardiographic findings. Briefly, after anesthesia with pentobarbital sodium (65 mg/kg ip for normal hamsters and 55 mg/kg ip for BIO14.6 hamsters), the anterior chest was shaved, and small needle electrodes for recording the electrocardiogram were inserted into the upper two extremities and lower left extremity. Using the left lateral position, we placed the echocardiographic probe on the surface of the chest and left ventricular endsystolic and end-diastolic dimensions (LVSD, LVDD, respectively), and the thicknesses of the LV posterior wall and interventricular septum (PWT, IVST) were measured from the LV M-mode tracing. Ejection time was estimated from the aortic Doppler flow signal at the LV outflow tract. Variables derived from the echocardiographic data, including percent fractional shortening (%FS) of the LV diameter, the percent thickening of the posterior wall (%WT), mean circumferential fiber shortening rate (VCF) corrected by the R-R interval, and the estimated LV mass index, were calculated as previously described (26). In vivo assessment of membrane permeability. To assess changes in sarcolemmal membrane permeability of cardiac myocytes, in vivo Evans blue dye (EBD) distribution was assessed. EBD is a nontoxic, low-molecular-weight dye (mol wt 960) that binds to serum albumin when infused intravenously and allowed to recirculate (16). EBD cannot cross the sarcolemma of intact myocytes, but if the membrane is ruptured, it is found in the intracellular space where it binds to intracellular proteins (16). Intracellular staining by EBD therefore identifies increased membrane permeability or sarcolemmal rupture. We applied this method to assess membrane permeability of cardiac myocytes in five normal and five CM hamsters at 5 and 9 wk of age. The hamsters were anesthetized with an intraperitoneal injection of pentobarbital sodium (65 mg/kg). Through a small incision in the left inguinal region, the femoral vein was exposed and cannulated with PE-50 tubing (Becton-Dickinson, Parsippany, NJ). Two percent EBD (Sigma, St. Louis, MO) solution was dissolved in PBS (ph 7.2) and sterilized by passage through a membrane filter with a pore size of 0.2 µm. EBD was then injected intravenously (50 µl/10 g body wt) and the skin was sutured; the hamsters were then allowed to recover. Forty eight hours after EBD injection, an overdose of pentobarbital sodium (100 mg/kg) was given, and the heart was quickly removed and arrested in modified Krebs-Hensleit buffer containing high potassium chloride (10 mg/ml). The LV was excised and divided with a cut parallel to the atrioventricular groove, and the apical two-thirds were embedded with OCT compound (Miles, Elkhart, IN) and frozen in isopentane chilled with liquid nitrogen. Consecutive sections were made from each frozen heart with the use of a cryostat (Microm, Heidelberg, Germany) with a chamber temperature of 25 C. Several pairs of cross-sectional 10-µm-thick adjacent sections were randomly selected from the consecutive sections. To visualize the sarcolemma and transverse tubules, sections were treated with 4% paraformaldehyde for 30 min and then stained with wheat germ agglutinin conjugated to fluorescein isothiocynate (WGA-FITC) and coverslipped with anti-fading media (Gelvatol, Air Products and Chemicals, Allentown, PA). This section was observed under a fluorescent microscope (HFX-DX, Nikon, Japan) equipped with blue and green activation filters (488 and 546 nm, respectively); different barrier filters (520 and 590 nm, respectively) were used to view and photograph WGA and EBD staining separately. Adjacent sections were stained with hematoxylineosin (HE), and observed through bright-field optics. Because heart muscle has a strong autofluorescence in the green and red fluorescent channels, control sections were made from EBD-free hearts treated in the same fashion and compared with EBD-stained sections. With the use of 100 magnification, 10 microscopic fields were randomly selected from each section, photomicrographs were taken, and the myocardial regions that were unstained and stained with EBD were scanned using a Hewlett-Packard PhotoSmart Photo Scanner. EBD-positive areas were estimated using SigmaScan Pro (Jandel Scientific Software, San Rafael, CA). The percent area of stained myocardium was calculated from each field and averaged. Antibodies. Mouse monoclonal antibodies against the COOH-terminal domain of human dystrophin (clone Dy8/ 6C5) and clone MANDRA-1 (kindly supplied by Dr. G. W. Morris), the rod domain of human dystrophin (clone Dy4/ 6D3), -dystroglycan (clone 43DAG1/8D5), and -sarcoglycan (clone Ad1/20A6) were obtained from Novocastra Laboratory (Newcastle, UK). The polyclonal anti-human -sarcoglycan antibody was kindly provided by Dr. V. E. Nigro. Secondary antibodies for light microscopy were donkey or goat antimouse FITC or Cy5 (Dako) and nanogold (Nanoprobes) conjugated to the goat anti-mouse IgG and followed by a silver enhancement solution (Intense M, Amersham Life Sciences, Arlington Heights, IL). Secondary antibodies for electron microscopy were nanogold and silver enhancement solution (Amersham). Preparation of cell lysates, immunoblotting. Normal and BIO14.6 hamster heart tissues at 6 wk were homogenized by a polytron grinder and lysed on ice for 30 min in 2 ml of ice-cold RIPA buffer (50 mm Tris, ph 7.4, 150 mm NaCl, 1% Nonidet P-40, 0.1% SDS, 0.1% deoxycholic acid, 1 mm phenylmethylsulfonyl fluoride, and a protease inhibitor cocktail purchased from Sigma). Cell lysates were collected by centrifugation at 13,000 g for 10 min at 4 C. Protein concentration was determined by the Bio-Rad protein assay, using BSA as standard. Cell lysates were mixed with 6 protein sampling buffer (final concentration of buffer: 50 mm Tris, ph 6.8, 10% glycerol, 2% SDS, 100 mm dithiothreitol, 0.0001% bromophenol), loaded on 4 20% Tris-glycine gel, and SDS-PAGE performed at 15 ma constant current. Proteins were then

H1364 CARDIAC DYSTROPHIN COMPLEX IN CARDIOMYOPATHIC HAMSTER transferred from SDS-PAGE gels to nitrocellulose membranes at 20 V overnight. Membranes were blocked in a western buffer (0.05% Tween-20 in Tris-buffered saline, ph 7.4) with 5% skim milk for 1 h, and then incubated with primary antibodies for 2 h at room temperature. After extensive washing with Western buffer, the membranes were incubated for 1 h with horseradish-conjugated goat antimouse or rabbit IgG secondary antibody diluted at 1:1,000 in the Western buffer with 0.5% milk. The horseradish peroxidase-conjugated protein complex was detected by enhanced chemiluminescence according to the manufacturer s protocol (ECL kit, Amersham). Light microscopic immunohistochemistry. Heart tissues (n 5) were frozen in isopentane precooled with liquid nitrogen and sectioned at a thickness of 10 µm on a cryostat. After air-drying for 30 min, sections were incubated for 30 min in 1% BSA/3% normal donkey or goat serum (depending on the secondary antibody). Sections were incubated with primary antibody (dystrophin, -dystroglycan, and -sarcoglycan) for 60 min, fixed with 4% paraformaldehyde for 30 min, and then labeled with secondary antibody conjugated to FITC or Cy5. Omission of the primary antibody and omission of both primary and secondary antibodies provided negative controls. The subcellular structure of the dystrophin-dystroglycan complex was visualized with a laser scanning confocal microscope (MRC-1024, Bio-Rad). Measurement of immunolabeling intensity. For measurement of immunolabeling intensity, images were taken as eight-bit-intensity digital image (intensity value 0 255) from each channel using the same laser excitation and photo-multiamplifier conditions used for laser scanning confocal microscopy in BIO14.6 and normal hamsters. To compare intensities of membrane immunolabeling, six cross-sectional areas were randomly chosen from multiple sections of each heart (n 5). To determine the intensity of sarcolemma and T tubules, areas showing an intensity over the threshold (set at 25 30) were automatically detected by SigmaScan Pro (Jandel Scientific), and the average membrane intensity (arbitrary units) was computed as the average of those areas. Electron microscopic immunohistochemistry of dystrophin. Tissue sections were fixed with 2% formaldehyde 0.001% glutaraldehyde (n 2) and treated with a denaturing buffer, 6 M guanidine hydrochloride in 50 mm Tris HCl. Sections were then incubated with blocking buffer (1% BSA/3% normal goat serum/1% cold fish gelatin in 0.1 M glycine/pbs) and subsequently incubated with the primary antibody for 2 h, washed six times with PBS, and then incubated with secondary mouse IgG conjugated to 1-nm immunogold (Nanogold, Molecular Probes). After several washes in PBS, sections were fixed with 2.5% glutaraldehyde, incubated with a silver enhancement solution (Intense M), followed with 0.05% gold chloride solution, then osmicated with 1% OsO 4, dehydrated, embedded into Ducarpan (ACM Resin), and then polymerized by baking. Polymerized thick sections were further sectioned onto an electron microscopy grid at a thickness of 0.1 µm. Thin sections were stained with 1% uranyl acetate and 1% lead solution (1% lead acetate, 1% lead nitrate) and observed on an electron microscope (JEM 100-CX, JEOL, Tokyo). Statistics. Echocardiographic values are represented as means SD and other data as means SE. To compare echocardiographic data, a two-way ANOVA was used. The factors for ANOVA were the hamster strain and ages. When overall variances associated with those factors were significantly different, multiple comparison tests between normal and BIO14.6 CM hamsters, or between 4 wk and the other ages were performed using the Student-Newman-Keuls test. A P value 0.05 was regarded as significant. To estimate significant differences between two groups of data such as EBD-stained areas in normal and BIO14.6 hamsters, unpaired t-tests were used with a P value 0.05 considered significant. RESULTS Echocardiographic findings. The body weight in CM hamsters was smaller but increased significantly with age from 4 to 14 wk in both normal and CM hamsters. The LV mass index, estimated by echocardiography divided by the body weight, tended to be smaller in BIO14.6 hamsters than in normals (difference significant only at 10 14 wk of age, data not shown). As shown in Fig. 1, the LVDD increased significantly with age in both strains of hamsters, but in the CM hamsters LVDD increased substantially more and was significantly larger than in normal hamsters between 8 and 14 wk. The LV systolic function, measured as %FS and corrected VCF, gradually decreased with age in CM hamsters and was significantly reduced compared with normal hamsters between 6 and 14 wk (Fig. 1; significant between 8 and 14 wk compared with 4 wk of same strain). The thickness of the posterior LV wall increased with age in normal hamsters but was significantly reduced in CM hamsters compared with normals after 8 wk (Fig. 1). All of these findings are consistent with progression of the cardiomyopathy. The heart rates were not significantly different between CM and normal hamsters at baseline (448 54 vs. 465 45 beats/min, respectively) and did not differ at subsequent time points. In vivo assessment of membrane permeability. As shown in Fig. 2, at low and high magnification myocytes with increased sarcolemmal permeability showed two types of intracellular staining with EBD. In one pattern (type A), sarcolemmal and T tubular structures could be identified by WGA staining, and such blocks of cells exhibited a diffuse, uniform pattern of intracellular EBD staining (Fig. 2, E and H, asterisks) without abnormalities on the adjacent HE-stained sections (Fig. 2B). In the second pattern (type B), the structure of the sarcolemma and T tubules could not be identified by WGA staining, and EBD-stained areas were irregular and less intense (Fig. 2, E and H, pound sign). In the adjacent HE-stained sections, clusters of necrotic myocytes with infiltration by mononuclear or polymorphonuclear cells were observed (Fig. 2, A and B). In normal tissues (Fig. 2C), there was no intracellular staining with EBD, as shown in Fig. 2, F and I. The percentage of the entire LV cross-sectional area stained by EBD (including both staining patterns) in normal and CM hamsters is shown in Fig. 3. In normal hamsters, no myocytes were stained with EBD at 5 wk (rare myocyte noted at 9 wk), whereas in CM hamsters 1.42 0.88% of myocytes were EBD positive at 5 wk with a significantly further increase to 5.17 1.64% at 9 wk. Because cardiac myocytes exhibit autofluorescence under the fluorescent microscope, we also observed sections unstained with EBD in which the autofluorescent effect on EBD positivity was negligible and similar

CARDIAC DYSTROPHIN COMPLEX IN CARDIOMYOPATHIC HAMSTER H1365 results were obtained. Heart sections from advanced CM hamster hearts (3 5 mo) show calcified areas in the myocardium, which exhibit strong emission at red fluorescent channels, but this effect was negligible at 5 and 9 wk. Immunoblotting and immunohistochemistry by light microscopy. On immunoblot analysis, the expression of dystrophin protein was not altered. However, -dystroglycan was reduced ( 9% reduction compared with normals), and - and -sarcoglycan proteins were completely deficient (Fig. 4). To assess the subcellular distribution of the DGC and its spatial relationship to damaged myocardial areas, we compared immunoblotting and immunohistochemical myocardial labeling of dystrophin, -dystroglycan, and -sarcoglycan at 5, 9, and 14 wk of age in normal and CM hamster hearts. -Sarcoglycan was considered representative of all the sarcoglycans, which have been shown to be absent or markedly reduced in the BIO14.6 hamster (25). As shown in Fig. 5, E and F, -sarcoglycan was found to be distributed in the sarcolemma in normal hamster hearts, whereas it was not detected in the BIO14.6 hamsters even at 5 wk, suggesting that the sarcoglycan complex was lost consequent to the deficiency of -sarcoglycan. The transmembrane component of the DGC, -dystroglycan, was localized at the sarcolemma and T tubules as a continuous staining pattern both in normal and CM hamster as shown in Fig. 5, A D. However, in CM hamster hearts the intensity of -dystroglycan was reduced by 19% (47.03 3.51 vs. 37.14 1.96 arbitrary units, P 0.05) compared with normal hearts (Fig. 5). Fig. 1. Time course of echocardiographic changes. LVDD, left ventricular diastolic dimension; %FS, percent LV fractional shortening; VCFc, corrected LV circumference fiber shortening; PWT, LV posterior wall thickness. Values are means SD; #P 0.05 vs. 4 wk; *P 0.05 vs. normal hamsters. Dystrophin was identified in the sarcolemma and T tubules both in normal and CM hamsters, and the intensity of dystrophin in the CM hamster hearts was comparable to that in normals. However, in CM hamster hearts dystrophin also was strongly expressed at the intercalated discs as shown in Fig. 6, D and F (arrows), but it was not found at that location in normal hearts. There was no staining of -dystroglycan at the intercalated discs of CM hamster hearts. The distribution of dystrophin and dystroglycan in cardiomyocytes exhibiting increased sarcolemmal permeability was assessed in sections stained for both EBD and WGA. The membrane lipid and glycoprotein components showed staining by WGA, whereas there was no specific staining of dystrophin and -dystroglycan in these areas, suggesting that the DGC was completely lost, despite some preservation of cellular architecture on WGA and HE staining in these areas. Electron microscopic immunohistochemistry of dystrophin. Because dystrophin was abnormally localized at the intercalated disc in CM hamster hearts, confirmation by electron microscopic immunolabeling of dystrophin was performed. In normal hamster hearts dystrophin was clearly distributed at the sarcolemma, frequently at the invagination site of the T tubules at the Z lines (image not shown), but was not observed at the intercalated discs. In the CM hamster hearts, dystrophin was observed in association with the sarcolemma and with the sarcolemma and the intercalated discs as observed in Fig. 6.

H1366 CARDIAC DYSTROPHIN COMPLEX IN CARDIOMYOPATHIC HAMSTER Fig. 2. Representative images of Evans blue dye (EBD) stained area. A, B: hematoxylin-eosin (HE) staining of area adjacent to wheat germ agglutinin (WGA) and EB sections. C: HE staining of normal field. Top half of A shows a necrotic area, and B is a high magnification of rectangular area in A. D, E: representative images of EBD intracellular staining or adjacent section on some regions. G, H: WGA staining of adjacent section of some regions to clarify sarcolemmal membrane structure of the cardiac myocytes. F, I: EBD/WGA staining from normal field, showing no intracellular EBD staining. In B, E, and H the * marks a typical type A EBD-stained myocyte; # indicates type B EBD-stained myocytes. Scale bar in A, C, D, F, G, I 25 µm; bar in B, E, H 10 µm. DISCUSSION Relationship of membrane damage to cardiac dysfunction. By quantifying the area of in vivo intracellular staining through the use of EBD in 5- and 9-wk-old Fig. 3. Quantification of EBD-positive area. Percent EBD-stained area of entire transverse LV sections at 5 and 9 wk in normal and BIO14.6 cardiomyopathic (CM) hamster hearts (n 5, each group). *P 0.05 vs. normal hamsters; #P 0.05 vs. 5 wks. hamster hearts, an increase of regions with enhanced membrane permeability and cell damage could be observed between 5 and 9 wk. Because the development of LV dysfunction also occurred during this period, increased membrane permeability could be the initiating event leading to myocyte damage and the progression of LV dysfunction. HE staining of type B areas of WGA and EBD staining showed fracture and loss of the myocardial cell cytoskeleton as well as large numbers of inflammatory cells, findings consistent with necrosis, whereas areas with type A staining displayed relatively mild changes in myocardial structure and few inflammatory cells. Although the fate of type A EBD-stained myocytes is uncertain, apoptosis has been reported to have a significant role in mdx skeletal muscle that stained positive with EBD (16), occurring mainly in the early period before necrosis was evident (27). However, our preliminary results in the CHF147 hamster (a strain derived from the UMX7.1 CM hamster) indicated that the contribution of apoptosis to myocardial cell damage is

CARDIAC DYSTROPHIN COMPLEX IN CARDIOMYOPATHIC HAMSTER H1367 Fig. 4. Western blots of dystrophin (DYS), -dystroglycan ( -DG), -sarcoglycan ( -SG), and -sarcoglycan ( -SG). Lane 1, normal hamster heart; lane 2, BIO14.6 CM hamster heart. minor compared with that resulting from necrosis early during the onset of heart dysfunction (8). Therefore, it can be hypothesized that the continuous contraction of the heart, which is enhanced in rate and strength during exercise, may promote progressively more sarcolemmal rupture with increasing age in hearts with cardiac -sarcoglycan deficiency. The absence of dystrophin and -dystroglycan immunolabeling in type A- positive EBD-stained areas in which WGA staining of membrane lipid and glycoproteins were preserved is consistent with increased sarcolemmal permeability rather than cell rupture and suggests that loss of the DGC from the sarcolemma predisposes to cell rupture. Other pathological mechanisms must also be considered, including the possibility that microvascular spasm described by Factor et al. (7) also may be involved in causing necrosis in blocks of myocardial cells. In this connection, the -sarcoglycan protein has been reported to be absent in vascular smooth muscle (6, 14). Distribution of dystrophin and -dystroglycan in CM hamster hearts. It was recently reported (25) that all four sarcoglycans and -dystroglycan are lost, whereas dystrophin and -dystroglycan are preserved in the BIO14.6 hamster heart. In the present study, we correlated serial changes in cardiac function with the extent and type of myocardial damage and further investigated the subcellular distribution of dystrophin and -dystroglycan in the young BIO14.6 hamster heart. In CM hamsters as young as 5 wk of age, we confirmed the absence of -sarcoglycan, but unlike the previous study Fig. 5. Immunohistochemical labeling of -dystroglycan and -sarcoglycan. A: cross-sectional image; C: high magnification of -dystroglycan labeling in normal hamster at 5 wk. B: cross-sectional image; D: high magnification of -dystroglycan labeling in BIO14.6 hamster at 5 wk. E: immunolabeling of -sarcoglycan in normal hamster at 5 wk. F: immunolabeling of -sarcoglycan in the BIO14.6 CM hamster at 5 wk. Scale bar in A, B 30 µm; bar in C, D 8 µm; bar in E, F 20 µm.

H1368 CARDIAC DYSTROPHIN COMPLEX IN CARDIOMYOPATHIC HAMSTER Fig. 6. Immunohistochemical labeling of dystrophin. A: cross-sectional image; C: longitudinal image; and E: highmagnification longitudinal image in normal hamster at 5 wk of age. B: crosssectional image; D: longitudinal image; and F: high-magnification longitudinal image in the BIO14.6 hamster at 5 wk. Arrows in D and F indicate abnormal labeling of dystrophin at the intercalated discs in CM hamster (Dy8/6C5). Scale bar in A D 30 µm; bar in E, F 8 µm. Downloaded from http://ajpheart.physiology.org/ by 10.220.33.6 on December 5, 2017 (25), we found -dystroglycan to be reduced in the sarcolemma and T tubules. The use of a laser scanning confocal microscope to quantify -dystroglycan may have allowed detection of the mild reduction in cardiac muscle. We also examined the expression level of -dystroglycan in skeletal muscle and found no significant difference between normal and CM hamsters (data not shown), suggesting a possible contributing factor to the relatively more severe cardiac phenotype compared with that of skeletal muscle observed in BIO14.6 hamsters. In CM hearts, dystrophin in the sarcolemma and T tubules was comparable to that in normal hamster hearts, but in addition it was strongly expressed at the intercalated discs at 5 and 14 wk. In skeletal muscle, dystrophin reportedly is found only at the sarcolemma, whereas in cardiac muscle it is also distributed in the transverse tubules but has never been reported to be present at the intercalated discs (13, 17). Decreased content of dystrophin protein has been reported previously on immunoblot analysis in BIO14.6 CM hamster hearts (11, 17), although this finding was evident in

CARDIAC DYSTROPHIN COMPLEX IN CARDIOMYOPATHIC HAMSTER H1369 animals older than those in our study. Our immunohistochemical data clearly indicate that dystrophin is preserved at 5 wk of age compared with -dystroglycan. Iwata et al. (11) reported that the subcellular distribution of dystrophin in the BIO14.6 CM hamster was not different from that in normals by using a rod domain anti-dystrophin antibody (Dy4/6D3). We used the same clone of anti-dystrophin antibody but found that it did not react as specifically with dystrophin in hamster tissue with immunoblotting and immunocytochemistry as the carboxy-terminal domain antibodies (Dy8/6C5 and MANDRA-1). Both light microscopic data and electron microscopic observations support our finding that dystrophin is abnormally distributed at the intercalated discs in the CM hamster hearts. Utrophin is a homolog of dystrophin localized to the intercalated discs in heart muscle; utrophin can replace the function of dystrophin in the dystrophin-deficient mdx mouse (21). Therefore, there is a possibility that the anti-dystrophin antibody used in the present study might have cross-reacted to some extent with utrophin localized at the intercalated discs, even if it exhibits a single band on Western immunoblotting, because the molecular weights are not distinguishable. However, we used specific monoclonal antibodies to dystrophin, and labeling by the antibody at the intercalated discs was absent in normal hamster hearts where utrophin is abundant (21). Therefore, cross-reactivity of the antidystrophin antibody with utrophin is a highly unlikely explanation for our finding. Several lines of evidence suggest that alterations are present at the intercalated discs in the CM hamster. Kawaguchi et al. (12) reported decreased expression of desmin and increased expression of vinculin at the intercalated discs in CM hamsters. Luque et al. (15) described decreased expression of the gap junction protein connexin43 on immunohistochemical analysis in CM hamsters; connexin43 turns over several times a day (1), and it is possible that its expression might be affected by excessive dystrophin distribution at the intercalated discs. However, when and how the abnormal distribution of dystrophin occurs and its functional significance remain to be investigated. Two possible hypotheses might be considered. Dystrophin is reported to have interactions not only with the dystrophinassociated proteins but also with other cytoskeletal proteins such as the integrins (28), caveolin-3 (23), and aciculin (2) in cultured skeletal myocytes. Aciculin, a cell adhesion regulatory protein, is localized mainly at the adherens junctions in cardiac myocytes. Because dystrophin is also localized at the intercalated discs near the adherens junction, we can speculate that in the CM hamster heart the cytoskeletal function of dystrophin may be partially mediated through adherens junction proteins, such as aciculin. Alternatively, the -sarcoglycan deficiency may cause abnormal dissociation of dystrophin protein into the cytosol, where it aggregates at the intercalated discs due to protein-protein interactions, without playing any functional role. In summary, together with the known absence of the sarcoglycans and reduced -dystroglycan, abnormal distribution of dystrophin may contribute to the observed loss of sarcolemmal integrity, which was associated with early progression of cardiac dysfunction in the CM hamster. NOTE ADDED IN PROOF Since this work was submitted, a study has been published (see Ref. 4a) on mice deficient in -sarcoglycan showing disruption of the sarcoglycan-sarcospan complex in vascular smooth muscle, with evidence for a role of vascular dysfunction in the development of cardiomyopathy in that murine model. The authors are grateful to Tom Deerinck for excellent suggestions about electron microscopy immunolabeling, to Farid Abdel-Wahhab for technical assistance, and to Pamela Alford for manuscript preparation. This study was supported by National Institutes of Health Specialized Center of Research in Heart Failure Grant HL-53773, an endowed chair awarded by the American Heart Association, California Affiliate, San Diego County Division (J. Ross), and by the Richard D. Winter Fund. Laser scanning confocal microscopic images and electron microscopic images were taken at the National Center for Microscopy & Imaging Research at San Diego, supported by Grant RR-04050. Address for reprint requests and other correspondence: J. Ross, Jr., Dept. of Medicine, Univ. of California, San Diego, 9500 Gilman Dr., 0613B, La Jolla, CA 92093-0613 (E-mail: jross@ucsd.edu). Received 28 June 1999; accepted in final form 19 October 1999. REFERENCES 1. Beardslee MA, Laing JG, Beyer EC, and Saffitz JE. Rapid turnover of connexin43 in the adult rat heart. Circ Res 83: 629 635, 1998. 2. Belkin AM and Burridge K. Association of aciculin with dystrophin and utrophin. J Biol Chem 270: 6328 6337, 1995. 3. Campbell KP. Three muscular dystrophies: loss of cytoskeletonextracellular matrix linkage. Cell 80: 675 679, 1995. 4. Chan YM, Bonnemann CG, Lidow HGW, and Kunkel LM. Molecular organization of sarcoglycan complex in mouse myotubes in culture. J Cell Biol 143: 2033 2044, 1998. 4a.Coral-Vasquez R, Cohn RD, Moore SA, Hill JA, Weiss RM, Davisson RL, Straub V, Barresi R, Bansal D, Hrstka RF, Williamson R, and Campbell KP. Disruption of the sarcoglycan-sarcospan complex in vascular smooth muscle: a novel mechanism for cardiomyopathy and muscular dystrophy. Cell 98: 465 474, 1999. 5. Duclos F, Straub V, Moore SA, Venzke DP, Hrstka RF, Crosbie RH, Durbeej M, Lebakken CS, Ettinger AJ, van der Meulen J, Holt KH, Lim LE, Sanes JR, Davidson BL, Faulkner JA, Williamson R, and Campbell KP. Progressive muscular dystrophy in alpha-sarcoglycan-deficient mice. J Cell Biol 142: 1461 1471, 1998. 6. Ettinger AJ, Feng G, and Sanes JR. Ipsiron-sarcoglycan, a broadly expressed homologue of the gene mutated in limb girdle muscular dystrophy 2D. J Biol Chem 272: 32534 32538, 1997. 7. Factor SM, Minase T, Cho S, Dominitz R, and Sonnenblick E. Microvascular spasm in the cardiomyopathic Syrian hamster: a preventable cause of focal myocardial necrosis. Circulation 66: 342 354, 1982. 8. Gu Y, Ryoke T, Martone M, Oh S, Knowton K, and Ross J Jr. The role of apoptosis in the early transition to heart failure in the cardiomyopathic hamster (Abstract). FASEB J: A520, 1999. 9. Holt KH, Lim LE, Straub V, Venzke DP, Duclos F, Anderson RD, Davidson BL, and Campbell KP. Functional rescue of the sarcoglycan complex in the BIO 14.6 hamster using deltasarcoglycan gene transfer. Mol Cell 1: 841 848, 1998.

H1370 CARDIAC DYSTROPHIN COMPLEX IN CARDIOMYOPATHIC HAMSTER 10. Hongo M, Sentianin EM, Tanaka N, Mao L, McKirnan MD, Clark RG, Won W, Chien KR, and Ross J Jr. ANG II blockade followed by growth hormone as adjunctive therapy after experimental myocardial infarction. J Card Fail 4: 213 224, 1998. 11. Iwata Y, Nakamura H, Fujiwara K, and Shigekawa M. Altered membrane-dystrophin association in the cardiomyopathic hamster heart muscle. Biochem Biophys Res Commun 190: 589 595, 1993. 12. Kawaguchi N, Fujitani N, Scaper J, and Onishi S. Pathological changes of myocardial cytoskeleton in cardiomyopathic hamster. Mol Cell Biochem 144: 75 79, 1995. 13. Klietsch R, Ervasti JM, Arnold W, Campbell KP, and Jorgensen AO. Dystrophin-glycoprotein complex and laminin colocalize to the sarcolemma and transverse tubules of cardiac muscle. Circ Res 72: 349 360, 1993. 14. Lim LE, Duclos F, Broux O, Bourg N, Sunada Y, Allamand V, Meyer J, Richard I, Moomaw C, Slaughter C, Tomé FMS, Fardeau M, Jackson CE, Beckmann JS, and Campbell KP. Beta-sarcoglycan: characterization and role in limb-girdle muscular dystrophy linked to 4q12. Nat Genet 11: 257 265, 1995. 15. Luque EA, Veenstra RD, Beyer EC, and Lemanski LF. Localization and distribution of gap junctions in normal and cardiomyopathic hamster heart. J Morphol 222: 203 213, 1994. 16. Matsuda R, Nishikawa A, and Tanaka H. Visualization of dystrophic muscle fibers in mdx mouse by vital staining with Evans blue: evidence of apoptosis in dystrophin-deficient muscle. J Biochem (Tokyo) 118: 959 964, 1995. 17. Meng H, Leddy JJ, Frank J, Holland P, and Tuana BS. The association of cardiac dystrophin with myofibrils/z-disc regions in cardiac muscle suggests a novel role in the contractile apparatus. J Biol Chem 271: 12364 12371, 1996. 18. Nigro V, de Sá Moreira E, Piluso G, Vainzof M, Belsito A, Politano L, Puca AA, Passos-Bueno MR, and Zatz M. Autosomal recessive limb-girdle muscular dystrophy, LGMD2F, is caused by a mutation in the delta-sarcoglycan gene. Nat Genet 14: 195 198, 1996. 19. Nigro V, Okazaki Y, Belsito A, Piluso G, Matsuda Y, Politano L, Nigro G, Ventura C, Abbondanza C, Molinari AM, Acampora D, Nishimura M, Hayashizaki Y, and Puca GA. Identification of the Syrian hamster cardiomyopathy gene. Hum Mol Genet 6: 601 607, 1997. 20. Petrof BJ, Shrager JB, Stedman HH, Kelly AM, and Sweeney HL. Dystrophin protects the sarcolemma from stresses developed during muscle contraction. Proc Natl Acad Sci USA 90: 3710 3714, 1993. 21. Pons F, Nicholsoon, LV, Robert A, Voit T, and Leger JJ. Dystrophin and dystrophin-related protein (utrophin) distribution in normal and dystrophin-deficient skeletal muscles. Neuromuscul Disord 3: 507 514, 1993. 22. Sakamoto A, Ono K, Abe M, Jasmin G, Eki T, Murakami Y, Masaki T, Toyo-oka T, and Hanaoka F. Both hypertrophic and dilated cardiomyopathies are caused by mutation of the same gene, delta-sarcoglycan, in hamster: an animal model of disrupted dystrophin-associated glycoprotein complex. Proc Natl Acad Sci USA 94: 13873 13878, 1997. 23. Song KS, Scherer PE, Tang Z, Okamoto T, Li S, Chafel M, Chu C, Kohtz DS, and Lisanti MP. Expression of caveolin-3 in skeletal, cardiac, and smooth muscle cells. Caveolin-3 is a component of the sarcolemma and co-fractionates with dystrophin and dystrophin-associated glycoproteins. J Biol Chem 271: 15160 15165, 1996. 24. Straub V and Campbell KP. Muscular dystrophies and the dystrophin-glycoprotein complex. Curr Opin Neurol 10: 168 175, 1997. 25. Straub V, Duclos F, Venzke DP, Lee JC, Cutshall S, Leveille CJ, and Campbell KP. Molecular pathogenesis of muscle degeneration in the delta-sarcoglycan-deficient hamster. Am J Pathol 153: 1623 1630, 1998. 26. Tanaka N, Dalton N, Mao L, Rockman HA, Peterson KL, Gottshall KR, Hunter JJ, Chien KR, and Ross J Jr. Transthoracic echocardiography in models of cardiac disease in the mouse. Circulation 94: 1109 1117, 1996. 27. Tidball JG, Albrecht DE, Lokensgard BE, and Spencer MJ. Apoptosis precedes necrosis of dystrophin-deficient muscle. J Cell Sci 108: 2197 2204, 1995. 28. Yoshida T, Pan Y, Hanada H, Iwata Y, and Shigekawa M. Bidirectional signaling between sarcoglycans and the integrin adhesion system in cultured L6 myocytes. J Biol Chem 273: 1583 1590, 1998.