Prevention of cardiomyopathy in mouse models lacking the - complex

Ronald D. Cohn, … , Sally Prouty, Kevin P. Campbell

J Clin Invest. 2001;107(2):R1-R7. https://doi.org/10.1172/JCI11642.

Article

Cardiomyopathy is a multifactorial disease, and the -glycoprotein complex has been implicated in the pathogenesis of both hereditary and acquired forms of the disease. Using mouse models of cardiomyopathy made by ablating for components of the sarcoglycan complex, we show that long-term treatment with verapamil, a calcium channel blocker with vasodilator properties, can alleviate the severe cardiomyopathic phenotype, restoring normal serum levels for cardiac I and normal morphology. Interruption of verapamil treatment leads again to vascular dysfunction and acute myocardial necrosis, indicating that predilection for cardiomyopathy is a continuing process. In contrast, verapamil did not prevent cardiac muscle pathology in dystrophin-deficient mdx mice, which neither show a disruption of the sarcoglycan complex in nor vascular dysfunction. Hence, our data strongly suggest that pharmacological intervention with verapamil merits investigation as a potential therapeutic option not only for patients with sarcoglycan mutations, but also for patients with idiopathic cardiomyopathy associated with myocardial ischemia not related to atherosclerotic coronary artery disease.

Find the latest version: https://jci.me/11642/pdf Prevention of cardiomyopathy in mouse Rapid Publication models lacking the smooth muscle See related Commentary, pages 153–154. sarcoglycan-sarcospan complex

Ronald D. Cohn,1 Madeleine Durbeej,1 Steven A. Moore,2 Ramón Coral-Vazquez,1 Sally Prouty,1 and Kevin P. Campbell1

1Howard Hughes Medical Institute, Department of Physiology and Biophysics, Department of Neurology, and 2Department of Pathology, University of Iowa College of Medicine, Iowa City, Iowa, USA Address correspondence to: Kevin P. Campbell, Howard Hughes Medical Institute, University of Iowa College of Medicine, 400 EMRB, Iowa City, Iowa 52242, USA. Phone: (319) 335-7867; Fax: (319) 335-6957; E-mail: [email protected]; Website: http://www.physiology.uiowa.edu/campbell.

Received for publication October 25, 2000, and accepted in revised form November 28, 2000.

Cardiomyopathy is a multifactorial disease, and the dystrophin-glycoprotein (LGMD) types 2C-F, and recently, an complex has been implicated in the pathogenesis of both hereditary and increasing number of patients with acquired forms of the disease. Using mouse models of cardiomyopathy made sarcoglycanopathies have been report- by ablating genes for components of the sarcoglycan complex, we show that ed to exhibit cardiomyopathy (refs. 3, long-term treatment with verapamil, a calcium channel blocker with vasodila- 9–12; R.D. Cohn, unpublished obser- tor properties, can alleviate the severe cardiomyopathic phenotype, restoring vation). Interestingly, in vivo studies of normal serum levels for cardiac and normal cardiac muscle mor- coronary blood flow in phology. Interruption of verapamil treatment leads again to vascular dys- patients with sarcoglycanopathies sug- function and acute myocardial necrosis, indicating that predilection for car- gested coronary dysfunction in these diomyopathy is a continuing process. In contrast, verapamil did not prevent patients, which might be related to cardiac muscle pathology in dystrophin-deficient mdx mice, which neither defects in vascular smooth muscle (13). show a disruption of the sarcoglycan complex in vascular smooth muscle nor We have recently characterized a vascular dysfunction. Hence, our data strongly suggest that pharmacological unique sarcoglycan-sarcospan (SG- intervention with verapamil merits investigation as a potential therapeutic SSPN) complex expressed in vascular option not only for patients with sarcoglycan mutations, but also for patients smooth muscle composed of ε-sarco- with idiopathic cardiomyopathy associated with myocardial ischemia not glycan (which replaces α-sarcoglycan), related to atherosclerotic coronary artery disease. β-, γ-, and δ-sarcoglycan, and sarcospan J. Clin. Invest. 107:R1–R7 (2001). (14–16). The significant impact of the distinct tissue distribution of the on the pathogenetic Introduction pathogenesis of certain DGC-associat- mechanisms has been specifically Dilated cardiomyopathy, one of the ed cardiomyopathies (5, 6). Moreover, shown in animal models for LGMD leading causes of heart failure in the mutations in the human δ-sarcoglycan type 2E and 2F, which are deficient for United States, is a multifactorial dis- have been characterized in β- and δ-sarcoglycan (Sgcb- and Sgcd- ease that includes both hereditary and patients with familiar and sporadic null, respectively) (5, 6). Our data sug- acquired forms (1). Experimental data cases of dilated cardiomyopathy with- gested that loss of the SG-SSPN com- have demonstrated that several hered- out significant involvement of the plex in vascular smooth muscle of itary forms of dilated cardiomyopathy (7). these mice caused perturbation of vas- are related to defects of the extrasar- The DGC is a multisubunit complex cular function with the presence of vas- comeric myocyte (2) and that provides a mechanical link cular constrictions leading to intermit- imply that familial cardiomyopathy between the and tent ischemic-like events, which can result from defective transmission the cytoskeleton (for review see ref. 8). initiated the development of car- of mechanical force generated in the The DGC is comprised of the diomyopathy. Furthermore, it was (2). Recent evidence is accu- cytoskeletal dystrophin, syn- shown that acute, treadmill mulating that hereditary and acquired trophins, and ; the sar- exercise–induced myocardial necrosis forms of cardiomyopathy can also be colemmal localized (α was rescued by administration of a caused by alterations within the dys- and β subunits); the sarcoglycans (α, β, vasodilator in Sgcd-null mice (5). These trophin-glycoprotein complex (DGC) γ, and δ subunits) and sarcospan. findings prompted us to explore the (3–6). Functional defects of vascular Mutations in any of the subunits (α, β, pathogenetic significance of vascular smooth muscle associated with muta- γ, δ) of the sarcoglycan subcomplex dysfunction and intermittent ischemia tions of the β- and δ-sarcoglycan genes have been shown to be associated with and possible therapeutic options to have revealed new insights into the limb-girdle muscular dystrophy prevent cardiomyopathy in genetic

The Journal of Clinical Investigation | Volume 107 R1 mouse models with different muta- tration of 1 mg/ml. These mice con- tions within the DGC. sumed an average dose of 3.5 mg/day Here we demonstrate, for the first (19). A group of six wild-type mice also time to our knowledge, that long-term received the same amount of vera- treatment with verapamil, a L-type cal- pamil orally in their drinking water. cium channel blocker with vasodilator Age-matched groups of Sgcb- and Sgcd- properties and clinical relevance, abol- null mice (n = 6 each) were not treated ishes vascular constrictions and effec- and served as controls for the study. tively prevents the development of After 16 weeks of oral verapamil treat- severe cardiomyopathy, as shown by ment, mice were analyzed for morpho- normal serum levels of cardiac tro- logical alterations and used for perfu- ponin I and cardiac muscle morpholo- sion studies. In a subgroup of Sgcb- gy in an animal model for the common and Sgcd-null mice (n = 4 each) vera- Figure 1 Levels of cTnI in verapamil-treated versus LGMD type 2E, deficient for β-sarco- pamil administration was stopped at untreated Sgcb- and Sgcd-null mice. Note the glycan. Various early studies have 16 weeks, and 14 days later mice were significant elevation of cTnI in the untreated shown beneficial effects of verapamil in analyzed for morphological alteration group of Sgcb- and Sgcd-null mice. Data are the cardiomyopathic hamster that car- and used for perfusion studies. All represented as mean plus or minus SD. ASgcb- ries a deletion within the δ-sarcoglycan studies were performed in accordance null treated versus untreated P < 0.002; gene (17, 18). Here we also demon- with the guidelines of the animal care BSgcd-null treated versus untreated P < 0.001. strate beneficial effects of verapamil on facility at the University of Iowa (Iowa the cardiomyopathic phenotype in a City, Iowa, USA). more genetically defined mouse model Histopathology studies. Histopathology (cTnI) was measured by an enzyme deficient for δ-sarcoglycan. Interrup- studies were performed as described sandwich immunoassay (ACS:180; tion of verapamil treatment leads again previously (5, 6). After embedding the Ciba-Corning Diagnostics Corp., Med- to perturbation of vascular function tissue of treated and untreated mdx, field, Massachusetts, USA) that uses and myocardial cell necrosis, indicating Sgcb-, and Sgcd-null mice (n = 6 each) in two cTnI-specific mAb’s with inde- that intermittent ischemic-like events paraffin, several hematoxylin and eosin pendent epitopes for cTnI. All assays as a predilection for cardiomyopathy and Masson’s trichrome–stained sec- were performed by technicians blind- are an ongoing process. Interestingly, tions (4 µm) throughout the entire ven- ed to the source of the serum. The nor- verapamil treatment has no beneficial tricle and lower part of the atrium were mal range for troponin I levels in the effects on the cardiomyopathic pheno- prepared to characterize cardiac muscle serum is less than 0.2 ng/ml. type of dystrophin-deficient mdx mice, pathology. All analyses were performed which neither show a disruption of the by five different people blinded to the Results SG-SSPN complex in vascular smooth treatment groups of each mouse. The Normal serum levels of cTnI in verapamil muscle nor vascular dysfunction. total and damaged myocardial areas of treated Sgcb- and Sgcd-null mice. Treat- Hence, our data reveal new com- the six different slices of each heart were ment of Sgcb- and Sgcd-null mice was pelling evidence for the pathogenetic traced and measured by an image-ana- started at 8 weeks of age, before the mechanism whereby cardiac muscle of lyzing system (Scion, Frederick, Mary- existence of any cardiac lesions. To Sgcb- and Sgcd-null mice is prone to land, USA). Subsequently, the percent- monitor cardiac muscle damage dur- intermittent ischemic-like events, age of these analyses was calculated, ing verapamil treatment we measured which consequently leads to cardio- and statistical analysis was performed cTnI levels in the sera of our mice at 8, myopathy. Furthermore, pharmaco- using the unpaired Student’s t test. 16, and 24 weeks of age (Figure 1). logical intervention with verapamil Immunohistological techniques were Throughout ontogeny, cTnI is not warrants investigation as a potential performed as described previously (5, expressed in skeletal muscle, and therapeutic option to prevent severe 6). Rabbit polyclonal Ab’s against dys- unlike creatine kinase isoenzyme MB, cardiomyopathy caused by intermit- trophin (rabbit 31), (rabbit it has not been detected in other tissues tent ischemic damage. 56), β-sarcoglycan (goat 26) and apart from the heart at any develop- δ-sarcoglycan (rabbit 214), β-dystrogly- mental stage or in diseased skeletal Methods can COOH-terminal peptide (rabbit muscle (20). Thus, the presence of cTnI Study design. Beginning at the age of 8 83), and sarcospan (rabbit 235) were in the circulation above the reference weeks, before the existence of any car- described previously (5, 6, 16). Microfil limit is highly specific for myocardial diac lesions, Sgcb- and Sgcd-null mice (Flowtech Inc., Carver, Massachusetts, injury. At 8 weeks of age cTnI values (n = 13 each) received oral supplemen- USA) perfusion studies were performed were normal in both verapamil-treated tation of verapamil (American Regent as described previously (5, 6). and nontreated Sgcb- and Sgcd-null Laboratories, Shirley, New York, USA) Determination of cardiac troponin I. mice, which is in accordance with in their drinking water. Treatment in Serum samples from mice were histopathological observations (data mdx mice started at 16 weeks of age obtained by retro-orbital and/or tail not shown). At 16 weeks of age, cTnI (n = 6 each). Verapamil was dissolved bleeding according to institutional levels were significantly elevated in in 10% dextrose solution at a concen- guidelines. Cardiac-specific troponin I untreated Sgcb- and Sgcd-null mice

R2 The Journal of Clinical Investigation | Volume 107 Figure 2 Morphological analysis of cardiac muscle of Sgcb- and Sgcd-null mice. The untreated mice (V–) show extensive areas of fibrosis in the heart (left panels). In contrast, most of the cardiac muscle of verapamil-treated mice (V+) is free of any morphological alterations (right panels). Bar, 1.5 mm and 75 µm, respectively. (a) Sgcb-null mice. (b) Sgcd-null mice.

(11.5 ± 1.4 and 15.3 ± 1.3 ng/ml, respec- old Sgcb- and Sgcd-null mice. No Verapamil has no beneficial effect on car- tively). These values correlate with the regions of acute or remote necrosis diomyopathy in mdx mice. Preliminary histological findings of severe ongoing were observed. In contrast, histological morphological observations of the focal myocytolytic cell damage evaluation of cardiac muscle from age- cardiomyopathic phenotype in dys- observed in 16-week-old Sgcb- and Sgcd- matched, untreated Sgcb- and Sgcd-null trophin-deficient mdx mice suggested null mice (5, 6). At 24 weeks of age, mice showed the expected severe mor- that the cardiomyopathy in these mice cTnI levels decreased to 2.8 ± 0.6 and phological signs of cardiomyopathy is not related to vascular dysfunction 4.2 ± 0.41 ng/ml in untreated Sgcb-null with extensive fibrosis, scarring, and and intermittent ischemic-like events. and Sgcd-null mice, respectively. Again, focal deposition of calcium (Figure 2, To test this we decided to study the these values are in accordance with pre- a and b). However, no significant ben- effects of verapamil supplementation vious histological findings of few areas eficial effect of verapamil treatment on the development of cardiomyopa- of acute ongoing necrosis at this stage was observed in Sgcb- and Sgcd-null thy in mdx mice. At the beginning of of the disease, where changes like fibro- mice when treatment was started in treatment at 16 weeks of age, no sig- sis and calcification become more the case of already existing severe car- nificant elevations of cTnI level were prominent (5, 6). In marked contrast, diac muscle fibrosis, indicating the detected in treated and untreated mdx cTnI values were normal or only slight- potential preventative effect of vera- mice. However, at 24 weeks (5.8 ± 0.4 ly elevated in the treated group of Sgcb- pamil (data not shown). Morphomet- ng/ml and 5.4 ± 1.2 ng/ml, respective- and Sgcd-null mice at 16 and 24 weeks ric analysis of cardiac muscle, deter- ly) and 30 weeks (7.2 ± 0.5 ng/ml and of age (0.4 ± 0.09 and 0.53 ± 0.05 mining total and damaged area of 7.8 ± 0.6 ng/ml, respectively) of age, ng/ml; 0.3 ± 0.06 and 0.2 ± 0.04), treated versus untreated mice, con- both treated and untreated mdx mice respectively (Figure 1). Evaluation of firmed the significant beneficial effect showed elevation of cTnI levels, sug- cTnI level in wild-type mice revealed of verapamil on the cardiac muscle of gesting that cardiac muscle damage normal values at different ages (<0.2 the treated mice (data not shown). Ver- was present in both groups of mice ng/ml), but elevated in dystrophin- apamil did not have any deleterious (Figure 3a). Histological analysis of utrophin–deficient mice, which devel- effects in wild-type mice, and no seri- cardiac muscle from treated and op a severe cardiomyopathy (9.8 ± 0.05 ous side effects were observed in any of untreated mdx mice displayed severe ng/ml; n = 3). Our data thus indicated the treated mice. Histological analysis fibrotic alterations and areas of acute that no significant cardiac muscle of skeletal muscle (diaphragm and ongoing necrosis, despite supplemen- pathology occurred during verapamil quadriceps) did not reveal any signifi- tation of verapamil (Figure 3b). Eval- treatment in Sgcb- and Sgcd-null mice. cant differences between treated and uation of cTnI levels in mice at 36 Verapamil prevents morphological alter- untreated mice (data not shown). This weeks of age, where only a few areas of ations of cardiac muscle in Sgcb- and Sgcd- is most likely due to the fact that acute necrosis can be observed, null mice. Evaluation of multiple histo- treatment with verapamil was started showed only mild elevations of cTnI logical cross sections from different at 8 weeks of age. Since the severe alter- levels. The lack of beneficial effects of levels of the entire heart showed at the ations of muscular dystrophy start as verapamil on the cardiomyopathic end of the verapamil-treatment trial early as 2 weeks of age, it is not sur- phenotype in dystrophin-deficient that the cardiomyopathic phenotype prising that verapamil had no preven- mdx mice prompted us to study the was effectively prevented in 24-week- tative effect in skeletal muscle damage. expression of the SG-SSPN complex

The Journal of Clinical Investigation | Volume 107 R3 Figure 3 Verapamil does not prevent cardiomyopathy in dystrophin-deficient mdx mice. (a) Levels of cTnI in verapamil-treated versus untreated mdx mice. Both groups of mice show elevations of cTnI levels in the serum. (b) Histological analysis of hearts from mdx mice after treatment with verapamil reveals areas of fibrosis in both groups of mice. (c) Immunohistochemical expression of β-sarcoglycan, δ-sarcoglycan, sarcospan, and β- is preserved in vascular smooth muscle of coronary arteries of mdx mice. In contrast, the SG-SSPN complex is absent in vascular smooth muscle of Sgcb-null mice. Note that utrophin is expressed at similar levels in wild-type, mdx, and Sgcb-null mice. β-SG, β-sarcoglycan; δ-SG, δ-sarcoglycan; β-DG, β-dystroglycan; DYS, dystrophin; and UTR, utrophin. Bar, 20 µm. in vascular smooth muscle of coro- mice by using Microfil perfusion as an Predilection for cardiomyopathy is an nary arteries in these mice. Interest- in vivo approach and compared it with ongoing process in Sgcb- and Sgcd-null mice. ingly, immunohistochemical analysis age-matched untreated mice (Figure 4). Next we tested whether the potential revealed that despite the loss of dys- As suggested by the immunohisto- for vasoconstriction as a predisposi- trophin in vascular smooth muscle, chemical results obtained in coronary tion for cardiomyopathy was an ongo- no alteration in the expression pattern arteries of mdx mice, even in the pres- ing process in Sgcb- and Sgcd-null mice. of the SG-SSPN complex was ence of cardiac muscle damage, neither We interrupted administration of ver- observed (Figure 3c). Our data strong- treated nor untreated mdx mice exhib- apamil in Sgcb- and Sgcd-null mice after ly suggest that the molecular mecha- ited any signs of vascular perfusion 16 weeks of treatment. Two weeks nism responsible for the development abnormalities, demonstrating that the later we analyzed these mice for car- of cardiomyopathy in mdx mice is dif- observed vascular dysfunction is not diac muscle morphology and vascular ferent from the mechanism of vascu- secondarily related to preexisting car- function. All mice developed focal lar dysfunction demonstrated in Sgcb- diac muscle damage. In contrast, myocytolytic lesions in the heart, a and Sgcd-null mice. Furthermore, the numerous microvascular constrictions phenomenon that is usually observed data substantiate the specificity of the associated with aneurysms were detect- around 14–18 weeks of age in untreat- beneficial effects of verapamil on the ed in coronary arteries of untreated ed mice. Furthermore, in vivo perfu- cardiomyopathic phenotype in Sgcb- Sgcb- and Sgcd-null mice. Remarkably, sion revealed numerous microvascular and Sgcd-null mice. long-term oral administration of vera- constrictions and vessel walls with a Abolition of vascular dysfunction in vera- pamil not only prevented morphologi- serrated contour (Figure 5). Determi- pamil-treated Sgcb- and Sgcd-null mice. cal alterations of cardiac muscle in nation of cTnI in the serum of these One rationale for using verapamil as a Sgcb- and Sgcd-null mice, but also abol- mice revealed elevated levels after potential therapeutic agent to prevent ished the observed vascular dysfunc- interruption of verapamil treatment cardiomyopathy in our animal models tion in coronary arteries. The coronary (8.5 ± 0.1 ng/ml and 11.2 ± 0.1 ng/ml, was given primarily by its potential to artery bed of treated mice was com- respectively), which confirmed the act as a vasodilator. Therefore, we ana- prised of smoothly tapered lumens histopathological findings. These data lyzed the vascular function of vera- without any evidence of vascular con- indicate that vascular dysfunction and pamil-treated mdx, Sgcb-, and Sgcd-null strictions (Figure 4). the necrotizing process as a predilec-

R4 The Journal of Clinical Investigation | Volume 107 tion for cardiomyopathy is an ongoing process that most likely continues for the life of Sgcb- and Sgcd-null mice. Thus, a long-term treatment with ver- apamil would be necessary to prevent heart disease in these mice.

Discussion An increasing number of patients have been reported to develop cardiomy- opathy due to mutations of either β-, γ-, or δ-sarcoglycan (9–12). In our study we monitored cardiac muscle damage in vivo via determination of cTnI. So far, measurement of cTnI has been used exclusively as a marker for myocardial infarcts in adult patients with ischemic heart disease (20). Here, we were able to demonstrate for the first time to our knowledge in genetic mouse models with muscular dystro- phy and cardiomyopathy, that evalua- tion of cTnI could indeed serve as an excellent marker to determine cardiac muscle damage in patients with mus- cular dystrophy due to its high speci- Figure 4 ficity as an indicator for myocardial Perfusion of coronary arteries of verapamil-treated and untreated mice. Transillumination necrosis. Evaluation of cTnI levels in of Microfil-perfused coronary arteries in untreated mice shows multiple constrictions patients with different forms of mus- (arrows) with pre- and poststenotic dilations, as well as narrow vessels with a serrated, rather than a smooth, contour (upper and middle left panels). In contrast, coronary vessels of treat- cular dystrophy could potentially ed Sgcb- and Sgcd-null mice (upper and middle right panels) reveal smoothly tapered vessels reveal further insights into the onset without signs of constriction and/or focal narrowing. Note the normal vascular perfusion of and extent of cardiac muscle lesions at treated and untreated mdx mice (lower panel). Bar, 40 µm. different stages of the disease. Because cTnI is a marker of acute myocardial damage, this would be particularly in the β- or δ-sarcoglycan gene (Figure tection from intermittent ischemic important for early stages of car- 6). Disruption of the SG-SSPN complex damage. Treatment with verapamil in diomyopathy where only small lesions in vascular smooth muscle leads to per- older mice did not have beneficial in the heart might be present. Studies turbation of vascular function as effects on cardiac muscle pathology, from our laboratory showed that even demonstrated by multiple vascular con- indicating that verapamil in fact serves in the presence of rather severe mor- strictions. In addition, loss of the as a preventative treatment option. phological damage in cardiac muscle SG-SSPN complex and a unique Interestingly, verapamil did not have of Sgcd-null mice, only about one-third ε-sarcoglycan–containing complex in any beneficial effects on the cardiomy- of mice demonstrated significant cardiomyocytes (see below) causes the opathic phenotype in dystrophin-defi- physiological dysfunction as observed cardiac muscle to be prone to intermit- cient mdx mice, which do not display by transthoracic echocardiography tent ischemic-like events that eventual- vascular dysfunction. Hence, our data (K.P. Campbell et al., unpublished ly lead to focal myocardial necrosis and substantiate the concept of intermit- observations). Although a direct phys- cardiomyopathy. Identification of this tent ischemic-like events as the patho- iological comparison between mice novel mechanism prompted us to genetic mechanism for the develop- and is difficult to make, it explore the possibility of pharmacolog- ment of cardiomyopathy in genetic should be taken into account that ele- ical intervention as a treatment option mouse models with targeted ablation of vation of cTnI might indicate early for cardiomyopathy in these genetic the β- or δ-sarcoglycan gene. and minor cardiac muscle damage mouse models. Here we demonstrate, Vascular smooth muscle tone and that would precede the onset of physi- we believe for the first time, that phar- myocardial contractility both depend ological abnormalities of cardiac mus- macological intervention using long- on calcium entry. Verapamil, a first- cle function. term oral treatment with verapamil is generation voltage-gated L-type calci- The current study and our data able to prevent severe cardiomyopathy um channel antagonist, has been reported previously suggest a novel in mice with disruption of the smooth shown to be nearly equipotent in pathogenetic mechanism for cardiomy- muscle sarcoglycan complex by allevia- reducing vascular smooth muscle tone opathy associated with mutations with- tion of vascular constrictions and pro- and inhibiting the membrane calcium

The Journal of Clinical Investigation | Volume 107 R5 mice, which neither show perturbation of the SG-SSPN complex in vascular smooth muscle nor any signs of vascu- lar dysfunction. Moreover, mdx mice did not develop the characteristically focal myocytolytic lesions observed in the sarcoglycan-deficient mice (5, 6) but showed a rather diffuse distribution pattern of histopathological abnormal- ities that further support the hypothe- sis that that cardiac muscle pathology of mdx mice is not related to a primary perturbation of vascular function. However, biochemical isolation of the DGC of cardiac muscle from mdx mice revealed a dissociation of the sarcogly- can and dystroglycan complexes (M. Figure 5 Durbeej and K.P. Campbell, unpub- Interruption of verapamil treatment leads to vascular dysfunction and myocardial necrosis. lished observations), suggesting that Analysis of cardiac muscle morphology and vascular perfusion 10 days after interruption of destabilization of the DGC due to loss 16 weeks of verapamil therapy exhibits the development of focal necrotic lesions and microvas- of dystrophin within the cardiac mus- cular constrictions in Sgcb- and Sgcd-null mice. Bar, 75 µm and 40 µm, respectively. cle per se predisposes mdx mice to devel- opment of cardiomyopathy. The equipotent potential of vera- influx into cardiomyocytes (for review patients with mutations within the pamil to improve calcium metabolism see ref. 21). The beneficial use of vera- δ-sarcoglycan and/or β-sarcoglycan of cardiomyocytes raises the possibili- pamil has been shown extensively in a gene. Furthermore, it will be interest- ty that a direct interaction of vera- variety of diseases such as hypertension ing to study the molecular basis of how pamil on the calcium metabolism of and cardiac arrhythmias, as well as in the absence of the SG-SSPN complex cardiomyocytes had some additional humans and animal models with car- in vascular smooth muscle affects vas- beneficial effect in Sgcb- and Sgcd-null diomyopathy (17, 18, 22, 23). Interest- cular function and whether the benefi- mice. Calcium overload causes myo- ingly, the cardioprotective efficacy of cial effect of verapamil on the vascula- cardial cell death and various mecha- verapamil was attributed to increased ture can be interpreted as a specific nisms such as ischemia, catecholamine transmural blood flow related to dysfunction of calcium metabolism in exposure, and membrane defects of vasodilation of the coronary artery bed vascular smooth muscle. cardiomyocytes may give rise to (17, 22) as well as to the potential of The pathogenetic mechanism and the increased cellular calcium concentra- verapamil to inhibit membrane calci- specificity of the verapamil action in tions. In fact, recent biochemical data um influx into cardiomyocytes (18). these mice was further supported by from our laboratory obtained from The current data reveal new com- the lack of beneficial effects of vera- Sgcb- and Sgcd-null mice suggest pelling evidence that perturbed vascu- pamil in dystrophin-deficient mdx defects of the cardiac membrane in lar function and intermittent ischemic- like events play a significant role in the pathogenesis of cardiomyopathy in Sgcb- and Sgcd-null mice. Abolition of microvascular constrictions over an extended time period is able to protect the cardiac muscle of Sgcb- and Sgcd- null mice from ischemic damage, which would consequently lead to the development of cardiomyopathy. Inter- estingly, histological examination of coronary arteries in recently reported patients with cardiomyopathy due to mutations in the δ-sarcoglycan gene Figure 6 Pathogenesis of cardiomyopathy in LGMD 2E/2F. The flow chart represents the current under- did not reveal any obvious coronary standing of the pathogenesis of cardiomyopathy due to primary mutations within the β- or δ- artery dysfunction (7). Future studies sarcoglycan gene. Absence of SG-SSPN complex in vascular smooth muscle leads to vascular will be needed to more carefully evalu- dysfunction in the form of microvascular spasms. In addition, loss of the SG-SSPN complex ate (e.g., functional studies) the possi- in cardiac muscle renders the heart susceptible to intermittent ischemic-like events, which even- ble involvement of coronary arteries in tually leads to the development of focal myocardial necrosis and severe cardiomyopathy.

R6 The Journal of Clinical Investigation | Volume 107 addition to the observed vascular dys- (24–27). Identification of novel muta- muscular dystrophy due to sarcoglycan gene mutations. Muscle Nerve. 22:473–479. function, due to disruption of a sec- tions and characterization of the asso- 12. Barresi, R., et al. 2000. Disruption of heart sarco- ond ε-sarcoglycan–containing com- ciated phenotypes would eventually glycan complex and severe cardiomyopathy plex in skeletal and cardiac muscle of warrant treatment trials with vera- caused by beta sarcoglycan mutations. J. Med. Genet. 37:102–107. Sgcb- and Sgcd-null mice (ref. 6; M. pamil in these patients. 13. Gnecchi-Ruscone, T., et al. 1999. Cardiomyopa- Durbeej and K.P. Campbell, unpub- thy in Duchenne, Becker and sarcogly- canopathies: a role for coronary dysfunction? lished observations). It is therefore Acknowledgments Muscle Nerve. 22:1549–1556. possible that a membrane abnormali- We would like to thank all members of 14. Durbeej, M., and Campbell, K.P. 1999. Biochem- ty renders the cardiomyocytes more the Campbell Laboratory for the criti- ical characterization of the epithelial dystrogly- can complex. J. Biol. Chem. 274:26609–26616. susceptible to intermittent ischemia cal reading of the manuscript, fruitful 15. Straub, V., et al. 1999. epsilon-sarcoglycan related to microvascular constrictions. discussions, and supply of critical replaces alpha-sarcoglycan in smooth muscle to form a unique dystrophin-glycoprotein complex. Prevention of vascular constrictions reagents. R.D. Cohn was supported by J. Biol. Chem. 274:27989–27996. and possible intervention with calci- the Deutsche Forschungsgemein- 16. Barresi, R., Moore, S.A., Stolle, C.A., Mendell, J.R., um homeostasis by verapamil may schaft (Co 241-1). This work was also and Campbell, K.P. 2000. Expression of γ-sarco- glycan in smooth muscle and its interaction with lead to myocyte protection and the supported by the Muscular Dystrophy the sarcoglycan-sarcospan complex. J. Biol. Chem. absence of cardiomyopathy. Association (K.P. Campbell). K.P. 275:38554–38560. Taken together, the successful treat- Campbell is an Investigator of the 17. Kobayashi, A., et al. 1987. Effects of verapamil on experimental cardiomyopathy in the Bio 14.6 Syr- ment of cardiomyopathy with vera- Howard Hughes Medical Institute. ian hamster. J. Am. Coll. Cardiol. 10:1128–1138. pamil in mouse models with muta- 18. Factor, S.M., Cho, S.H., Scheuer, J., Sonnenblick, E.H., and Malhotra, A. 1988. Prevention of hered- β δ 1. Schocken, D.D., Arrieta, M.I., Leaverton, P.E., and tions in the - or -sarcoglycan gene itary cardiomyopathy in the Syrian hamster with Ross, E.A. 1992. Prevalence and mortality rate of chronic verapamil therapy. J. Am. Coll. Cardiol. implies that efforts toward drug ther- congestive heart failure in the United States. J. Am. 12:1599–1604. apy can be of tremendous benefit in Coll. Cardiol. 20:301–306. 19. Dong, R., Liu, P., Wee, L., Butany, J., and Sole, M.J. 2. Chen, J., and Chien, K.R. 1999. Complexity in preventing certain forms of hereditary 1992. Verapamil ameliorates the clinical and simplicity: monogenic disorders and complex car- pathological course of murine myocarditis. J. Clin. cardiomyopathy. This particularly diomyopathies. J. Clin. Invest. 103:1483–1485. Invest. 90:2022–2030. 3. Nigro, G., Comi, L.I., Politano, L., and Bain, R.J. accounts for cardiomyopathy as a seri- 20. Bodor, G.S., Porterfield, D., Voss, E.M. Smith, S., 1990. The incidence and evolution of cardiomy- and Apple, F.S. 1995. Cardiac troponin-I is not ous complication in patients with opathy in Duchenne muscular dystrophy. Int. J. expressed in fetal and healthy or diseased adult LGMD types 2E and 2F, where no Cardiol. 26:271–277. human skeletal . Clin. Chem. 4. Badorff, C., et al. 1999. Enteroviral protease 2A treatment options are currently avail- 41:1710–1715. cleaves dystrophin: evidence of cytoskeletal dis- 21. Little, W.C., and Cheng, C.P. 1994. Vascular ver- able. Because of the preventative bene- ruption in an acquired cardiomyopathy. Nat. Med. sus myocardial effects of calcium antagonists. 5:320–326. ficial effects of verapamil, screening Drugs. 47(Suppl.):41–45. 5. Coral-Vazquez, R., et al. 1999. Disruption of the for mutations within the β- and δ- 22. Udelson, J.E., et al. 1989. Verapamil prevents sarcoglycan-sarcospan complex in vascular silent myocardial perfusion abnormalities during sarcoglycan genes should be extended smooth muscle: a novel mechanism for car- exercise in asymptomatic patients with hyper- diomyopathy and muscular dystrophy. Cell. to a wider group of patients with trophic cardiomyopathy. Circulation. 98:465–474. 79:1052–1060. inherited cardiomyopathy. This is fur- 6. Durbeej, M., et al. 2000. Disruption of the β-sarco- 23. Gistri, R., et al. 1994. Effect of verapamil on ther supported by recent studies, glycan gene reveals a complex pathogenetic mech- absolute myocardial blood flow in hypertrophic anism for LGMD 2E. Mol. Cell. 5:141–151. which showed that familiar and spo- cardiomyopathy. Am. J. Cardiol. 74:363–368. 7. Tsubata, S., et al. 2000. Mutations in the human 24. Cannon, R.O., III, et al. 1985. Myocardial radic cases of dilated cardiomyopathy delta-sarcoglycan gene in familial and sporadic ischemia in patients with hypertrophic car- δ dilated cardiomyopathy. J. Clin. Invest. caused by mutations within the - diomyopathy: contribution of inadequate 106:655–662. vasodilator reserve and elevated left ventricular sarcoglycan gene can occur even in the 8. Cohn, R.D., and Campbell, K.P. 2000. The molec- filling pressures. Circulation. 71:234–243. absence of significant skeletal muscle ular basis of muscular dystrophy. Muscle Nerve. 25. Dunn, R.F., et al. 1982. Comparison of thallium- 23:1456–1471. problems (7). Therefore, screening for 201 scanning in idiopathic dilated cardiomyopa- 9. Nigro, V., et al. 1996. Autosomal recessive limb- thy and severe coronary artery disease. Circulation. mutations should also include girdle muscular dystrophy, LGMD2F, is caused by 66:804–810. a mutation in the delta-sarcoglycan gene. Nat. patients without a primary skeletal 26. Opherk, D., et al. 1983. Coronary dilatory capac- Genet. 14:195–198. ity in idiopathic dilated cardiomyopathy: analy- muscle involvement but who have 10. Moreira, E.S., et al. 1998. A first missense muta- sis of 16 patients. Am. J. Cardiol. 51:1657–1662. been reported to have an idiopathic tion in the delta sarcoglycan gene associated with cardiomyopathy associated with a severe phenotype and frequency of limb-girdle 27. Pasternac, A., and Bourassa, M.G. 1983. Patho- muscular dystrophy type 2F (LGMD2F) in Brazil- genesis of chest pain in patients with cardiomy- myocardial ischemia in the absence of ian sarcoglycanopathies. J. Med. Genet. 35:951–953. opathies and normal coronary arteries. Int. J. Car- atherosclerotic coronary artery disease 11. Melacini, P., et al. 1999. Heart involvement in diol. 3:273–280.

The Journal of Clinical Investigation | Volume 107 R7