Article Text

Original article
Desmosomal protein gene mutations in patients with idiopathic dilated cardiomyopathy undergoing cardiac transplantation: a clinicopathological study
  1. Pablo Garcia-Pavia1,
  2. Petros Syrris2,
  3. Clara Salas3,
  4. Alison Evans2,
  5. Jesus G Mirelis1,
  6. Marta Cobo-Marcos1,
  7. Carlos Vilches4,
  8. Belen Bornstein5,
  9. Javier Segovia1,
  10. Luis Alonso-Pulpon1,
  11. Perry M Elliott2
  1. 1Cardiomyopathy Unit, Heart Failure and Heart Transplant Section, Department of Cardiology, Hospital Universitario Puerta de Hierro, Madrid, Spain
  2. 2Inherited Cardiovascular Disease Unit, The Heart Hospital, University College London, London, UK
  3. 3Department of Pathology, Hospital Universitario Puerta de Hierro, Madrid, Spain
  4. 4Department of Immunology, Hospital Universitario Puerta de Hierro, Madrid, Spain
  5. 5Department of Biochemistry, Hospital Universitario Puerta de Hierro, Madrid, Spain
  1. Correspondence to Dr Pablo Garcia-Pavia, Cardiomyopathy Unit, Department of Cardiology, Hospital Universitario Puerta de Hierro, Manuel de Falla 2, 28222 Madrid, Spain; pablogpavia{at}


Backgroud Idiopathic dilated cardiomyopathy (DCM) is the most frequent indication for orthotopic heart transplantation. It has been suggested that mutations in genes encoding desmosomal proteins, more typically associated with arrhythmogenic right ventricular cardiomyopathy, are a cause of DCM.

Objectives To determine the frequency of desmosomal protein gene mutations in heart transplant recipients and their families and to examine histopathological characteristics of explanted organs from mutation carriers.

Methods 89 unrelated patients aged 47.9±13.5 years (80% male) transplanted for end-stage DCM underwent genetic screening of five desmosomal genes (PKP2, DSP, DSC2, DSG2 and JUP). The findings were correlated with clinical features and histological characteristics in explanted hearts.

Results Pathogenic mutations were identified in 12 patients (13%). Five additional patients (6%) had genetic variants of unknown significance. The clinical phenotype of patients with pathogenic mutations was indistinguishable from that observed in patients without mutations. Evaluation of 76 relatives from 14 families with sequence variants (11 with pathogenic mutations and three with variants of unknown effect) identified 38 mutation carriers, four of whom had an overt DCM phenotype. Evidence of co-segregation of mutations with DCM phenotype was found in five families. Histological evaluation of explanted hearts did not show any specific features in patients with pathogenic mutations.

Conclusions Mutations in desmosomal genes are frequent in patients with advanced DCM undergoing cardiac transplantation. These findings emphasise the importance of familial evaluation and genetic counselling in patients with end-stage DCM and pose important challenges for current histopathological criteria for arrhythmogenic right ventricular cardiomyopathy.

  • Idiopathic dilated cardiomyopathy
  • arrhythmogenic right ventricular cardiomyopathy
  • desmosomal proteins
  • cardiac transplant
  • heart transplant
  • genetics
  • cardiomyopathy dilated

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


Idiopathic dilated cardiomyopathy (DCM), defined as dilation and systolic impairment of the left or both ventricles in the absence of hypertension, coronary artery disease or valvular abnormalities,1 is the commonest cause of heart failure in the young2 and the most frequent indication for orthotopic heart transplantation.3 Family studies in predominantly stable outpatient populations suggest that up to 48% of patients have a familial predisposition to disease.4 5 However, the genetic basis of disease of most patients is still unknown,6 7 particularly in those with advanced disease in whom there are few genetic or family studies.

A recent study has suggested that at least 5% of patients with chronic stable DCM harbour mutations in genes encoding desomosomal proteins.8 The primary aim of this study was to determine the frequency of pathogenic mutations in desmosomal protein genes in heart transplant recipients and their families and to compare and contrast the histopathological characteristics of explanted organs and clinical features of patients with and without mutations.


Study population

Ninety-five patients undergoing heart transplantation for DCM were randomly selected from a cohort of 187 patients transplanted for DCM at our institution between September 1993 and December 2007 (39% of transplants over that period). Only those patients fulfilling the WHO/International Society and Federation of Cardiology Task Force clinical criteria for DCM9 at the time of heart transplantation (ie, left ventricular end-diastolic diameter >27 mm/m2 and an ejection fraction <40% or fractional shortening <25% in the absence of abnormal loading conditions, coronary artery disease, congenital heart lesions and other systemic diseases) were included. Patients with significant coronary artery disease at pathological examination of the explanted heart were excluded.

Pretransplant assessment comprised physical examination, 12-lead ECG, echocardiography, 24 h ambulatory ECG monitoring, 6 min walk test, upright exercise testing, biventricular radionuclide ventriculography and cardiac catheterisation. Additional studies including endomyocardial biopsy, electrophysiological study and cardiac MRI were performed only if there was a clinical indication to do so.

Patient records were reviewed retrospectively by two physicians blinded to the genetic results (PG-P, JGM). Clinical data and results from the first pre-transplant evaluation at our unit were collected. No patient had myocarditis (focal or diffuse inflammatory lymphocyte infiltration in the interstitium with myocyte damage/necrosis) on pretransplant endomyocardial biopsy.

Based on patient history and family pedigree analysis, DCM was defined as familial if one or more relatives (in addition to the proband) had DCM during life or at post-mortem examination or had experienced unexplained sudden cardiac death before the age of 35 years.10

Genetic evaluation

Since September 1993 all patients placed on the heart transplant waiting list at our centre have been asked to provide a blood sample for genetic analysis. DNA from these samples is extracted and stored at −70°C. In this study, DNA samples were amplified by PCR as described previously11–13 using primers designed to amplify the coding exons and the flanking intronic sequences of five arrhythmogenic right ventricular cardiomyopathy (ARVC)-related desmosome-encoding genes: plakophillin-2 (PKP2), desmoplakin (DSP), desmocollin-2 (DSC2), desmoglein (DSG2) and plakoglobin (JUP). Following PCR amplification, direct sequencing of amplicons was performed on an ABI PRISM 3130 DNA analyser using BigDye Terminator chemistry (V.3.1, Applied Biosystems, Warrington, UK). Primer sequences and PCR conditions are available on request.

For every sequence variant detected in patients with DCM, a cohort of 200 ethnically-matched control subjects was screened using the same methods. Every sequence variant found was cross-referenced to the international ARVC database.14 Patients were classified as carriers of pathogenic mutations if they had a genetic variant that was reported in the database as pathogenic, a novel sequence variant not found in controls that predicts a premature truncation, frameshift or abnormal splicing, a novel missense mutation that affects a conserved amino acid residue and co-segregated with disease on familial evaluation, or a variant classified in the international database as a variant of unknown effect which co-segregated with disease on subsequent family screening.

Patients with sequence variants classified in the international database as variants of unknown effect or with novel missense mutations not found in controls without corroborative family screening data were considered as carriers of mutations of uncertain significance.

The likelihood of pathogenic effect of missense sequence variants in desmosomal genes was determined by four in silico prediction methods: the Grantham score, PolyPhen, PolyPhen-2 and SIFT (see references w1–w4 in online supplement). The Grantham score method uses a formula for the difference between amino acids according to their physicochemical properties with high values indicating radical mutations. PolyPhen predicts the possible effect of amino acid substitutions based on physiochemical differences, evolutionary conservation and the proximity of the substituted amino acid to important known or predicted functional protein domains. SIFT predicts the functional importance of amino acid changes based on the alignment of orthologous and/or paralogous protein sequences. A missense sequence change is most likely to be pathogenic when it is classified as ‘probably damaging’ by PolyPhen or ‘not tolerated’ by SIFT. Conservation of amino acid residues in desmosomal genes was determined by Homologene ( by multiple alignment of orthologues in various species including Homo sapiens, Pan troglodytes, Mus musculus, Rattus norvegicus, Gallus gallus, Bos taurus.

Family screening

All relatives of patients with pathogenic mutations or variants of uncertain significance were offered clinical and genetic evaluation after genetic counselling. Clinical evaluation included physical examination, ECG and an echocardiogram. In accordance with our unit's genetic testing policy, genetic screening was not offered to relatives <16 years of age if they were asymptomatic and clinical screening tests were completely normal.

Family screening was considered positive if one or more relatives had DCM and the same genetic defect as the proband.

Pathological examination of explanted hearts

Gross (non-formalin-fixed) and microscopic examinations of explanted hearts were performed as described previously15 by a histopathologist (CS) blinded to the genetic and clinical data. Blocks of the anterior and posterior free wall of the left ventricle, right ventricle and interventricular septum were removed systematically for histopathological study. Specimens were also taken from areas showing macroscopic abnormality. Tissue samples were stained with H&E, orcein and Masson's trichrome. The presence of fibrosis and fatty tissue was evaluated in three preparations per ventricle. The average extent of fibrosis and fatty tissue from epicardium to endocardium in both ventricles was recorded (<10%, 10–25%, 25–50%, 50–75% and >75%). Isolated epicardial fatty infiltration was excluded from the analysis. Other microscopic abnormalities including disorganised myocardial cells, degenerative myocytes and inflammatory infiltrates were noted.

Statistical analysis

Continuous variables are expressed as mean±SD values. Discrete variables are shown as percentages. Differences between means were compared using the Student t test and the Mann–Whitney U test for normally distributed and non-normally distributed continuous data, respectively. χ2 and Fisher exact analysis were used to test for associations between dichotomous variables. Probability values reported are two-sided and values <0.05 were considered statistically significant. All data were analysed using the SPSS software V.5.0.


Following a review of case notes, six of the 95 patients were excluded for the following reasons: family relation (brother) to another selected patient (n=1); past history of chemotherapy (n=1); history of toxic oil syndrome (n=1); DCM associated with muscular dystrophy known to be caused by a dystrophin mutation (n=1); and dilated phase of restrictive cardiomyopathy (n=2). The final study cohort therefore comprised 89 patients (mean age at transplantation 49.2±13.1 years, range 12.3–68.6, 80% male). All patients were Caucasians. Their clinical characteristics are summarised in table I in the online supplement. Forty-three patients (48%) were evaluated as inpatients, 10 (11%) on the intensive care unit. Two patients were removed from the heart transplant waiting list due to improvement in their functional class but were subsequently relisted and transplanted 248 and 378 days later, respectively. Before transplant, none of the patients fulfilled current diagnostic criteria for ARVC16 17 or had a family history of ARVC.

Sixteen patients (18%) had a known family history of DCM and seven (8%) had a family history of sudden cardiac death (only one in a relative aged <35 years). From history and pedigree analysis, 17 patients (19%) fulfilled familial DCM criteria.10

Genetic analysis

Twelve patients (13%) had pathogenic mutations and five patients (6%) had genetic variants of unknown pathogenicity. None of these genetic abnormalities were found in the 200 control patients.

Patients with pathogenic mutations

Family evaluation was possible in 11 of the 12 patients with pathogenic mutations. Seven patients had mutations that had been described previously in patients with ARVC14; two had novel mutations in DSG2 and DSP that predicted premature truncation of the transcribed protein; three patients (D2, D11 and D18) had missense mutations in DSP (D11 and D18 had the same mutation previously classified as an unknown variant and D2 had two novel mutations) that co-segregated with DCM in families D2 and D18. Degree of conservation through species, current classification in ARVC international database and predicted pathogenicity at in silico analysis of missense mutations found are shown in table II in the online supplement.

Co-segregation of mutations with DCM in another relative was demonstrated in five families (D2, D5, D12, D16, and D18; see family trees in online supplement). Three patients (D1, D12 and D17) with pathogenic mutations also carried sequence variants of unknown significance.

Tables 1 and 2 show the clinical course, genetic, pathological and family data for patients with pathogenic mutations. No significant differences in clinical and echocardiographic characteristics were found between mutation carriers and patients without pathogenic genetic variants (table 3).

Table 1

Clinical and genetic characteristics of probands with genetic variants in desmosomal genes

Table 2

Clinical, familial and histological characteristics of probands with genetic variants in desmosomal genes

Table 3

Clinical, electrocardiographic, echocardiographic and histological characteristics of patients with and without desmosomal protein gene mutations

Patients with genetic variants of unknown effect

Five patients (6%) had genetic variants of unknown pathogenicity. Four patients (D6, D7, D15 and D20) had previously unreported missense mutations that were not found in controls. Another patient (D13) had a previously described mutation in PKP2 classified as a genetic variant of unknown effect in the ARVC/D mutation database.14 All these missense mutations affect amino acid residues conserved through species (see table III in online supplement). Family evaluation was either negative (D6, D7, D20) or unavailable (D13, D15).

Tables 2 and 3 show the clinical course, genetic, pathological and familial information for patients with variants of unknown effect.

Family evaluation

Of a total of 17 families with genetic abnormalities, 14 (82%) agreed to participate in the study. From a total of 85 relatives contacted, 76 (89%) agreed to be screened. Thirty-eight relatives (50%) were carriers of the same genetic abnormalities as those found in their families (27 had pathogenic mutations and 11 had variants of unknown significance). Five relatives were found to have impaired left ventricular function. In one of them (patient D5.2) this was caused by coronary artery disease. In the remaining four relatives with DCM, coronary artery disease was excluded by coronary angiography. None of them fulfilled old or current ARVC criteria (see table IV in online supplement). The penetrance of DCM among mutation carriers was therefore 14% (four affected individuals from 27 mutation carriers). Echocardiographic and ECG characteristics of unaffected relatives were indistinguishable between those who were carriers of genetic abnormalities and those who were not, except for T wave inversion in lead III which was found more frequently among mutation carriers (see table V in online supplement).

Histopathological findings

Fifty-eight explanted hearts (65%) showed fatty (n=35; 39%) and/or fibrotic infiltration (n=45; 50%) in the right ventricle and 72 (81%) had fat (n=7; 8%) and/or fibrosis (n=71; 80%) in the left ventricle. The percentage of fibrofatty change was highly variable with only 12 (14%) and 12 (14%) patients having more than 25% of their right ventricle and left ventricle walls occupied, respectively. Twenty explants (23%) had lymphocyte infiltrates and 39 (44%) had signs of active myocyte degeneration.

Among probands with mutations, findings in explanted hearts were very heterogeneous (table 2) with some showing extensive fibrofatty infiltration and others a complete absence of fat and fibrotic tissue (figure 1). Patients with mutations of uncertain effect also exhibited a wide range of fibrofatty infiltration (table 2).

Figure 1

Microscopic examinations from patients (A) D12, (B) D19 and (C) D17. (A) Masson's trichrome stain (×25) showing fibrofatty infiltration of the right ventricle. (B) Masson's trichrome stain (×200) showing extensive fibrotic replacement of myocytes in the left ventricle. (C) Haematoxylin-eosin stain (×200) showing absence of fibrofatty infiltration with normal disposition of myocytes in the left ventricle.

Differences between both groups in the amount of left ventricular and right ventricular walls occupied by fat and/or fibrosis were not statistically significant (table 3).


This study shows that a substantial proportion of patients undergoing heart transplantation for end-stage DCM have mutations in genes coding for desmosomal proteins. Importantly, clinical findings prior to transplantation and histological evaluation of explanted hearts in mutation carriers were similar to those observed in non-mutation carriers. A wide spectrum of histological findings was noted among mutation carriers and patients with mutations of uncertain significance. Some patients exhibited extensive fibrofatty infiltration while others showed near normal histology.

Role of desmosomal genes in DCM

Desmosomes are secondary mechanical intercellular junctions present in abundance in epithelial tissues and the myocardium. Together with the adherens and gap junctions, the desmosomes connect myocardial cells and maintain the mechanical and electrical integrity of the heart.19 The desmosomes have a complex structure that includes adhesion molecules of the cadherin family (desmogleins and desmocollins) and proteins of the plakin and catenin families (desmoplakin, plakophilins, and plakoglobin) which link intermediate filaments of the cytoskeleton to the desmosomal cadherins.20 Some desmosomal proteins (particularly plakoglobin) act as nuclear signalling molecules via Wnt signalling pathways.21 A link between mutations in desmosomal protein genes and ARVC was first suggested by the discovery of homozygous mutations in plakoglobin (Naxos disease) and DSP (Carvajal syndrome).22 Subsequent studies demonstrated mutations in these and other desmosomal proteins in patients with the more common (but phenotypically more variable) autosomal dominant form of the disease.23 While the defining feature of ARVC is involvement of the right ventricle, it was recognised quite early that the left ventricle (in particular the posterolateral wall) can also be involved and that progressive dilation of both ventricles can lead to an end stage indistinguishable from DCM. However, until recently DCM has been thought to be a distinct clinical entity unrelated in most instances to ARVC. The findings in this study suggest that the rigid distinction between DCM and ARVC is to some extent arbitrary. For example, many of the mutations reported in this study have been reported in patients with a pure ARVC phenotype, but none of the patients had clinical features that fulfil recently modified criteria for the diagnosis of ARVC.17 Although biventricular fibrosis was common, histological analysis of explanted hearts revealed fibrofatty change in both ventricles in only two patients (2%) of the total study cohort and, interestingly, there were no differences in conventional histology in patients with and without pathogenic mutations.

Previous studies in transplant populations

Very few studies have examined genetic characteristics in heart transplant recipients.24 Monserrat et al25 undertook clinical and echocardiographic evaluation in relatives of 43 heart-transplanted patients with DCM and found that DCM was familial in 11 probands (25.6%) and possibly familial in a further 11 individuals (25.6%). The pattern of inheritance was autosomal dominant in most families and some relatives had other genetically determined cardiac diseases such as hypertrophic cardiomyopathy. Karkkainen et al24 investigated the prevalence of lamin A/C mutations (so far the most frequent gene associated with familial DCM) among 66 heart transplant recipients and found mutations that explained DCM in six cases (9%). Family screening in our study revealed two relatives (D12.3 and D16.2) with previously unknown (and untreated) DCM. This, along with the demonstration that 50% (7/14) of the screened families and 19% of the entire cohort had evidence of familial DCM, illustrates the importance of genetic counselling and the provision of clinical screening to the relatives of patients with end-stage DCM.

Clinical implications

Although the genetic basis of DCM has been known for more than 25 years26 and mutations in more than 40 genes have been associated with the condition,6 7 the use of genetic screening in daily practice has been limited by the low yield of current testing strategies. When systematic genetic screening has been applied to large DCM cohorts, a significant number of mutations (around 5%) has been found only in two genes (LMNA and MYH7).27 28 More recently, screening of 14 genes associated with DCM in a cohort of 312 patients with DCM has shown that approximately 27% of DCM probands had possible or likely disease-causing genetic variants.27 29 30

While our study suggests that the addition of desmosomal protein gene mutations to the standard panel of tests offered to patients with DCM has the potential to increase the efficiency of genetic testing, low clinical penetrance and emerging data showing that variants in desmosomal genes are relatively common in some populations31 means that caution should be employed in the interpretation of mutation analysis in this context. Compound heterocigosity (multiple mutations in ≥2 genes) is increasingly identified in other inherited cardiomyopathies, and it is highly probable that this phenomenon will affect the clinical expression of a disease like DCM where so many genes have been linked with the condition. As it is possible that some genetic variants merely increase susceptibility to disease or modify the response to other genetic, epigenetic or environmental factors, testing for desmosomal protein gene mutations should always be performed in conjunction with careful clinical phenotyping and familial evaluation.


Desmosomal protein gene mutations are common in patients with end-stage DCM but are not associated with a specific clinical or histological phenotype. These findings emphasise the importance of genetic counselling for patients undergoing cardiac transplantation and pose important challenges for current histopathological criteria for ARVC.


We are indebted to the patients and their families for their participation in the study. We also gratefully acknowledge Mrs Ana Briceño for her excellent technical assistance and Drs Francisco Hernández, Inés Garcia, Manuel Sánchez and Patricia Avellana for their valuable help in approaching the families.


Supplementary materials

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

    Files in this Data Supplement:


  • Funding This work was partially supported by the Instituto de Salud Carlos III (grant PI08/0978) and the Spanish Ministry of Health (Red Cooperativa de Insuficiencia Cardiaca (REDINSCOR) (RD06/03/0018). Some of this work was undertaken at UCLH/UCL who received a proportion of funding from the Department of Health's NIHR Biomedical Research Centres funding scheme.

  • Competing interests None.

  • Ethics approval This study was conducted with the approval of the Hospital Universitario Puerta de Hierro and complies with the principles of the declaration of Helsinki.

  • Provenance and peer review Not commissioned; externally peer reviewed.