Dilated Cardiomyopathy Associated With Deficiency of the Cytoskeletal Protein Metavinculin
Background The cytoskeleton plays an important role in maintaining cell structure and integrity. Defects in cytoskeletal proteins can cripple cell strength and may cause cardiomyopathy. We analyzed heart tissues from subjects with dilated cardiomyopathy for abnormalities in the cardiac cytoskeleton. Metavinculin, a cardiac isoform of the cytoskeletal protein vinculin, connects actin microfilaments to the intercalated disk and membrane costameres of the heart.
Methods and Results Metavinculin and vinculin transcripts and protein were analyzed by polymerase chain reaction (PCR) and Western blotting. Thirty-three human heart specimens were studied, including 5 normal controls, 4 subjects with ischemic cardiomyopathy, 1 with X-linked cardiomyopathy, and 23 with idiopathic dilated cardiomyopathy (IDC). PCR of cardiac cDNA detected absence of the metavinculin transcript in cardiac tissue from a subject with IDC. PCR of genomic DNA showed that the metavinculin exon was present but not utilized in the cardiac transcript. Western blot analysis demonstrated absence of metavinculin protein in the heart from this subject. Immunostaining of cardiac vinculin in this heart showed disorganized intercalated disk structures. Metavinculin deficiency was associated with normal cardiac expression of the cytoskeletal proteins vinculin, α-actinin, and dystrophin. Normal metavinculin expression in the other heart specimens suggests that the defect is specific in the IDC subject identified.
Conclusions These results demonstrate an association between metavinculin deficiency and dilated cardiomyopathy due to a defect in alternative mRNA splicing.
The cytoskeleton is composed of a complex group of proteins that maintain the structural integrity of the cell, impart cell shape, and enable the cell to withstand mechanical stress. Defects in cytoskeletal proteins can cripple cell strength and make individual cells or whole organs susceptible to disease. For example, mutations in the cytoskeletal dystrophin protein complex cause membrane instability, leading to cardiac failure and skeletal myopathy.1 2 Idiopathic dilated cardiomyopathy (IDC) is a disease of the heart muscle characterized by contractile impairment and alteration in cardiomyocyte shape, and cytoskeletal abnormalities may be a common feature of this disease.3 The sarcomere attaches to the cardiomyocyte membrane via nonsarcomeric actin microfilaments4 complexed with the cytoskeletal proteins talin, α-actinin, vinculin, and metavinculin,5 6 which are linked to cadherin or the integrin receptor. Metavinculin is interesting because its expression is restricted to cardiac and smooth muscle.7 Metavinculin is an isoform of vinculin, and both proteins are synthesized from the same gene on chromosome 10 via alternative mRNA splicing of a 3′ (204-bp) exon.7 Metavinculin (123 kD) is larger than vinculin (116 kD) and contains 68 additional amino acids encoded by the extra exon. Vinculin and metavinculin appear to localize to similar cellular structures in cardiac muscle: the intercalated disk and the sarcolemmal attachment sites of nonsarcomeric actin.8 These attachment sites are important in establishing mechanical coupling in the ventricular myocardium.6 9 In this study, we analyzed cytoskeletal protein expression in failing cardiac tissue from patients with IDC and found a subject with a deficiency of cardiac metavinculin.
Human ventricular heart samples were obtained from 5 normal donor hearts intended for transplantation, 4 hearts from patients with ischemic dilated cardiomyopathy, 1 heart from a subject with X-linked cardiomyopathy due to dystrophin deficiency, and 23 hearts from subjects with IDC. All pathological samples were obtained according to institutional guidelines and were quickly frozen in isopentane cooled in liquid nitrogen and stored at −80°C.
Western Blot Analysis
Cardiac muscle tissues were homogenized and analyzed by Western blot as previously described.10 Anti-vinculin (Sigma, 1:400), anti–α-actinin (Sigma, 1:800), and anti-dystrophin antibodies10 were incubated with the membrane for 2 hours at room temperature. The anti-vinculin antibody detects both vinculin and metavinculin isoforms. The membrane was incubated with alkaline phosphatase–conjugated anti-mouse IgG antibody for vinculin and α-actinin (1:3000, Sigma) or anti-rabbit IgG antibody for dystrophin, and the immunoreactive bands were detected by nitro blue tetrazolium and bromochloroindolyl phosphate.10 Protein concentrations of tissue homogenates were determined by BCA protein assay (Pierce). Western blot membranes stained for vinculin and metavinculin were double stained with α-actinin antibody as a cytoskeletal protein control in each lane.
Unfixed frozen heart tissues were cut into 6-μm sections, blocked with 3% BSA in PBS, pH 7.4, and incubated with anti-vinculin (and metavinculin) antibody (1:100 dilution) or anti-cadherin antibody as a control for intercalated disk structures11 for 2 hours at room temperature. The slides were then incubated with rhodamine- or fluorescein-conjugated secondary antibodies (Sigma, 1:50) for 1 hour and mounted with 90% glycerol in PBS. Staining with secondary antibodies alone was used as a control for nonspecific fluorescence. The stained sections were photographed under ultraviolet light with a Nikon microscope.
PCR of Metavinculin and Vinculin Transcripts From cDNA
Total RNA was isolated12 and cDNA was prepared by techniques previously described.10 Polymerase chain reaction (PCR) primers were created from the published human vinculin sequence7 and were designed to detect both vinculin- and metavinculin-specific sequences (Fig 1).⇓ The forward primer 5′-TGAAGCTCGCAAATGGTCCAGCAAG-3′ is upstream of the metavinculin-specific exon 19, and the reverse primer 5′-ACCTCATCTGAGGCCTTGGCGATGT-3′ is downstream of exon 19. These primers are predicted to synthesize a 162-bp product from the vinculin transcript and a 366-bp product from the metavinculin transcript. PCR was performed for 35 cycles (94°C for 30 seconds, 60°C for 30 seconds, 72°C for 2 minutes) as described,10 and results were analyzed on 3% NuSieve agarose gel (FMC Bioproduct) containing 0.2 mg/mL ethidium bromide.
PCR of the Metavinculin Exon in Genomic DNA
Genomic DNA was obtained from frozen human left ventricular heart specimens as described.13 Primers were designed to amplify a 395-bp product from genomic DNA intron sequences flanking metavinculin-specific exon 196 : forward primer 5′-CTTTCTCATCCTAGGCAGGCTTTGG-3′ and reverse primer 5′-GAGTGATGCAATGTGTGAAGGCTGC-3′.
Western Blot Analysis
Cytoskeletal vinculin, α-actinin, and dystrophin content did not vary significantly between different hearts, with the exception of one heart with dystrophin deficiency and X-linked cardiomyopathy that we described previously.14 Metavinculin deficiency was detected in one subject with IDC. Fig 1 displays metavinculin, vinculin, and α-actinin immunostaining in 2 IDC patients, including the index case (IDC 2), 2 patients with ischemic cardiomyopathy, 1 patient with X-linked cardiomyopathy, and 1 of the normal control subjects. Of the 28 human hearts analyzed, only IDC 2 showed no immunoreactivity for metavinculin. Metavinculin protein accounts for ≈20% of total vinculin protein in most heart specimens, as previously reported.8 The relative abundances of vinculin and α-actinin were observed to be similar in all samples examined. Metavinculin deficiency in smooth muscle could not be determined because specimens were not obtained at the time of explant.
PCR analysis of metavinculin transcript (Fig 1) shows that vinculin mRNA (162 nucleotides [nt]) but not metavinculin mRNA (366 nt) is detectable by PCR in the IDC 2 patient. The metavinculin product (366 nt) was clearly detectable in control tissue and all other tissues examined. PCR of genomic DNA from the index case shows that the metavinculin exon was present in the IDC 2 patient (not shown). Sequence analysis of this product (which included some flanking intron sequences) did not reveal a splice junction mutation. However, this exon was not normally spliced into the vinculin transcript and expressed in the heart (Fig 2B).⇓
The subject with metavinculin deficiency (IDC 2) was a 21-year-old previously athletic white man diagnosed with IDC in May 1991. The patient's left ventricular ejection fraction was 11%, and his right ventricular ejection fraction was 14%. A chest radiograph showed cardiomegaly. Coronary angiography revealed normal coronary arteries, and endomyocardial biopsy showed no evidence of inflammatory infiltrates. The family history was remarkable for a great-aunt who died of cardiomyopathy.
We observed cardiac vinculin and metavinculin immunostaining at intercalated disks and costameres in control tissues (Fig 2A). Costameres are the periodically distributed spots along cell margins corresponding to the I bands of sarcomeres around the Z line; they represent the sarcolemmal-sarcomere attachment sites. In the IDC 2 (metavinculin-deficient) patient's heart, intercalated disk staining was discontinuous and irregular (Fig 2C and 2D). Membrane staining of vinculin at the costameres in IDC 2 also appeared irregular (Fig 2C). Vinculin and metavinculin staining in control tissue (and IDC subjects with normal metavinculin expression) demonstrated a smooth, continuous staining pattern at the intercalated disk and intact membrane staining, with fewer intracellular T tubules than seen in the IDC 2 subject. Intercalated disk staining of control heart with anti-cadherin antibodies demonstrated a smooth, continuous pattern (Fig 2B), suggesting that the interrupted cadherin pattern in IDC 2 (Fig 2E) is due to disk disruption and not to relocalization of vinculin at the junctional sites. Comparison of the interrupted staining in IDC 2 (vinculin only) with the smooth staining pattern in control tissue (vinculin and metavinculin) suggests that although vinculin and metavinculin colocalize to the same structures in the heart, there may be slight variation in function.
IDC is a disease of heart muscle failure without evidence of other contributing factors.15 Although the etiologic basis of this disease remains unknown, molecular genetic defects may be responsible for as many as 25% of cases.16 Interestingly, the onset of cardiomyopathy in late adolescence or early adulthood is typical of some forms of genetic cardiomyopathy, including X-linked cardiomyopathy, myotonic dystrophy, and hypertrophic cardiomyopathy.14 15 The potential importance of the cytoskeleton as a target for investigation was demonstrated in one qualitative study that found alterations in myocardial ultrastructure and distribution of vinculin, desmin, and tubulin in patients with dilated cardiomyopathy.3 There are now several reports of defects in cytoskeletal proteins that result in cardiac contractile dysfunction in both hypertrophied17 and dilated hearts.1 2 This report supports the hypothesis that alterations in the cardiac cytoskeleton may contribute to the ventricular remodeling that occurs in some cases of cardiomyopathy.
Our results demonstrate a deficiency of cardiac metavinculin mRNA and protein in a subject with IDC. This cytoskeletal defect was associated with abnormalities in ventricular function, cardiac dilatation, and immunohistological defects at the site of metavinculin absence: the cardiomyocyte membrane and intercalated disk. The molecular defect appears to affect splicing of the metavinculin exon in cardiac tissue. Analysis of tissue from our other patients with either IDC or ischemic dilated cardiomyopathy showed them to have levels of vinculin and metavinculin comparable to normal control heart tissue. Thus, it is unlikely that loss of metavinculin is a nonspecific alteration in the failing heart. It should be recognized, however, that this observation appears to be an infrequent or even rare occurrence. Furthermore, metavinculin deficiency may not be the only (or even the primary) defect causing heart failure.
The unique function of metavinculin in cardiac tissue is unknown. C-terminal peptide residues encoded by the metavinculin exon display complete identity in human, porcine, and avian species.7 18 This latter region contains the only tyrosine residue and is presumed to be important for metavinculin function.7 Because skeletal muscle lacks intercalated disks, it is possible that the localization of metavinculin to this highly specialized structure in the heart is related to a specific function conferred by metavinculin, a function that vinculin alone cannot support. Thus, these data, combined with other recent reports,1 2 3 suggest that the organization and function of the cardiac cytoskeleton may be important in some cases of dilated cardiomyopathy. Analyses of the integrity and expression of cytoskeletal proteins in larger populations are required to test this hypothesis.
This work was supported in part by a VA Research Advisory Group grant to Dr Bies and by the Temple Hoyne Buell Foundation.
- Received August 21, 1996.
- Revision received October 30, 1996.
- Accepted November 1, 1996.
- Copyright © 1997 by American Heart Association
Fadig R, Sunada Y, Waclawik A, Buck S, Lewandoski P, Campbell K, Lotz B. Deficiency of a dystrophin associated glycoprotein (adhalin) in a patient with muscular dystrophy and cardiomyopathy.
Schaper J, Froede R, Hein S, Buck A, Hashizume H, Speiser B, Friedl A, Bleese N. Impairment of the myocardial ultrastructure and changes of the cytoskeleton in dilated cardiomyopathy. Circulation.. 1991;83:504-514.
Alberts B, Bray D, Lewis J, Raff M, Roberts K, Watson J. The cytoskeleton. In: Molecular Biology of the Cell. New York, NY: Garland Publishing; 1994:787-861.
Wang N, Butler JP, Ingber DE. Mechanotransduction across the cell surface and through the cytoskeleton. Science.. 1993;260:1124-1127.
Goncharova EJ, Kam Z, Geiger B. The involvement of adherens junction components in myofibrillogenesis in cultured cardiac myocytes. Development.. 1992;114:173-183.
Koteliansky VE, Ogryzko EP, Zhidkova NI, Weller PA, Critchley DR, Vancompernolle K, Vandekerckhove J, Strasser P, Way M, Gimona M, Small JV. An additional exon in the human vinculin gene specifically encodes metavinculin-specific difference peptide. Eur J Biochem.. 1992;204:767-772.
Belkin AM, Ornatsky OI, Glukhova MA, Koteliansky VE. Immunolocalization of metavinculin in human smooth and cardiac muscles. J Cell Biol.. 1988;107:545-553.
Peters NS, Severs NJ, Rothery SM, Lincoln C, Yacoub MH, Green CR. Spatiotemporal relation between gap junctions and facia adherens junctions during postnatal development of human ventricular myocardium. Circulation.. 1994;90:713-725.
Bies RD, Freidman D, Roberts R, Perryman MB, Caskey CT. Expression and localization of dystrophin in human cardiac Purkinje fibers. Circulation.. 1992;86:147-153.
Maeda M, Taft C, Bush E, Holder E, Neville H, Bailey W, Perryman MB, Bies RD. Identification, tissue specific expression, and subcellular localization of the 71- and 80 kDa forms of myotonic dystrophy kinase protein. J Biol Chem.. 1995;270:20246-20249.
Chirgwin JM, Przybukla AE, MacDonald RJ, Rutter WJ. Isolation of biologically active ribonucleic acid from sources rich in ribonucleases. Biochemistry.. 1979;13:2633-2637.
Ausubel FM, Brent R, Kingston RE, Moore DD, Seidman JG, Smith JA, Struhl K. Current Protocols in Molecular Biology. New York, NY: John Wiley & Sons, Inc: 1995.
Bies RD, Holder E, Maeda M. X-Linked cardiomyopathy: selective deficiency of cardiac dystrophin. J Am Coll Cardiol. 1995;February:277A. Abstract.
Goldstein J, Brown MS. Genetics and cardiovascular disease. In: Braunwald E, ed. Heart Disease. 3rd ed. Philadelphia, Pa: WB Saunders Co; 1988:1633.
Tsutsui H, Ishihara K, Cooper G IV. Cytoskeletal role in the contractile dysfunction of hypertrophied myocardium. Science.. 1993;260:682-687.