A novel desmocollin-2 mutation reveals insights into the molecular link between desmosomes and gap junctions

Background Cellular adhesion mediated by cardiac desmosomes is a prerequisite for proper electric propagation mediated by gap junctions in the myocardium. However, the molecular principles underlying this interdependence are not fully understood. Objective The purpose of this study was to determine potential causes of right ventricular conduction abnormalities in a patient with borderline diagnosis of arrhythmogenic right ventricular cardiomyopathy. Methods To assess molecular changes, the patient's myocardial tissue was analyzed for altered desmosomal and gap junction (connexin43) protein levels and localization. In vitro functional studies were performed to characterize the consequences of the desmosomal mutations. Results Loss of plakoglobin signal was evident at the cell junctions despite expression of the protein at control levels. Although the distribution of connexin43 was not altered, total protein levels were reduced and changes in phosphorylation were observed. The truncation mutant in desmocollin-2a is deficient in binding plakoglobin. Moreover, the ability of desmocollin-2a to directly interact with connexin43 was abolished by the mutation. No pathogenic potential of the desmoglein-2 missense change was identified. Conclusion The observed abnormalities in gap junction protein expression and phosphorylation, which precede an overt cardiac phenotype, likely are responsible for slow myocardial conduction in this patient. At the molecular level, altered binding properties of the desmocollin-2a mutant may contribute to the changes in connexin43. In particular, the newly identified interaction between the desmocollin-2a isoform and connexin43 provides novel insights into the molecular link between desmosomes and gap junctions.

2 formalin, embedded in paraffin, cut in 5 μm sections and stained with hematoxylin and eosin for histological examination. Unstained slide-mounted sections of paraffin-embedded tissue were immunostained and myocardial specimens obtained at autopsy from three age-matched individuals with no clinical history or pathological evidence of heart disease were subjected to the same staining protocols and served as controls 2 . Briefly: slide-mounted sections were deparaffinised, placed in citrate buffer (10 mmol/l, pH 6.0), and heated in a microwave oven until boiling for 10 min to enhance specific immunostaining. After being cooled to room temperature, the tissue sections were simultaneously permeabilised and blocked by incubating them in PBS containing 0.1 % Triton X-100 and 3 % normal goat serum. The sections were then incubated with the primary mouse monoclonal antibodies (mAB), anti-PG, anti-Ncadherin (both Sigma) or anti-Cx43 (Zymed), overnight at 4 °C, brought to room temperature, washed three times in PBS, and incubated with indocarbocyanine-conjugated goat anti-mouse or anti-rabbit IgGs (Jackson Immunoresearch) for 2 h at 25 °C.
Immunostained preparations were analyzed by laser-scanning confocal microscopy (Sarastro Model 2000, Molecular Dynamics) as previously described 2 .
For high resolution Western blotting of different Cx43 species the chemiluminescent signal (SuperSignal West Pico Chemiluminescent Substrate, Thermo Scientific) was recorded with the Molecular Imager Chemidoc XRS system using QuantityOne software (both Biorad).
For quantification of Cx43 in the patient sample, blots for Cx43 and desmin (loading control) were run in triplicates. Densitometric analysis was performed (ImageJ software), and the ratio of Cx43 to desmin expressed relative to the control sample (control set to 1). Data are visualised as mean +/-standard deviation. Student's t-test was performed and p < 0.05 considered being significant.

Generation of cDNA constructs
The A517V change was introduced into a full length DSG2 GFP fusion construct 5 with the QuikChange mutagenesis kit (Stratagene) according to the manufacturer's instructions using the following primer pair: 5'-gtatgtgaatgttactgtagaggacctggatggac-3', 5'gtccatccaggtcctctacagtaacattcacatac-3'. Generation of wild-type (WT) full length DSC2a pCDNA3 has been described elsewhere 6 . The Q851fsX855 mutation was introduced as above using following primers: 5'-gaaaatcacaagcatgcccagactatgtcctgacatataa-3' and 5'-tatatgtcaggacatagtctgggcatgcttgtgattttc-3'. The pCDNA3 constructs coded for the human DSC2 proteins without any additional amino acids; these proteins could be detected with DSC2 antibodies (which do not cross-react with endogenous mouse or rat DSC2 protein).

Indirect immunofluorescence and confocal microscopy
Transfected cells were fixed with 4 % paraformaldehyde in phosphate buffered saline (PBS) for 10 min at room temperature, and subsequently permeabilized with 0.2 % Triton X-100 in PBS for 5 min. Cells were blocked with 10 % normal goat serum (Sigma) in PBS prior to incubation with the primary antibodies (mAB anti-DSC2/3, mAB anti-PG, pR anti-alphaactinin, pR anti-DSP, Serotec, pR anti-beta-catenin, Sigma), diluted in 1 % bovine serum albumin in PBS, for 1 h at room temperature. After washing with 0.1 % Tween-20 in PBS, cells were incubated with Goat-anti-Mouse-antibody or Goat-anti-Rabbit-Antibody conjugated to fluorescent dyes (Jackson ImmunoResearch, following combinations: antirabbit Cy2 with anti-mouse Cy3). DAPI (Sigma) was used to visualize nuclei. After being washed, the cells were mounted in Tris-buffered glycerol with n-propyl gallate (Sigma) as anti-fading agent. Specimens were analyzed and documented on a confocal microscope Leica SPE (Leica), using a 63x/1.30 oil immersion lens and solid phase lasers.
Protein-loaded beads were incubated with pre-cleared adult rat heart lysate (350 µg total protein) or COS-1 cell lysate (transfected with human Cx43 cytoplasmic tail construct The same GST fusion constructs (together with DSC2a A897fsX900, aa 716-900) were also cloned into mammalian expression vector pEBG as above for expression as GSTfusion proteins in COS-1 cells 6 . The cells were transfected with 7.5 micrograms of each pEBG constructs alone in 100 mm dishes (to assay binding to endogenous Cx43).
Alternatively, the pEBG constructs were transfected together with the Cx43-YFP or DSP-NTP-GFP, using 5 microgams of DNA each. After 48 h, the transfected COS-1 cells were harvested, lysed in GST-pulldown buffer for 20 min on ice, and extracts clarified by centrifugation at 50,000 g at 4°C for 15 minutes. GST-pulldown assays were performed by 6 addition of glutathione-sepharose for 1 h on ice and washed, eluted and analyzed as above, using anti-Cx43, anti-GFP and goat polyclonal anti-GST (GE Healthcare) antibodies.
Subsequently, the extract was clarified by centrifugation at 50,000 g at 4°C for 15 minutes.
Immunoprecipitations were performed using 1 µg anti-DSC2/3 or anti-Cx43 in the presence of 1 % bovine serum albumin. Immuno-complexes were purified with Protein G sepharose (Sigma), washed three times with IP-buffer and the bound proteins were eluted with two-fold SDS-sample buffer and detected by Western blotting. For comparison, diluted lysate samples (1 % of total input) were run as lysate controls on the same gel.
To assess DSG2-DSC2 interactions, COS-1 cells were transfected with construct combinations as indicated and lysed in IP-buffer supplemented with 60 mM n-octyl-beta-Dglucoside. GFP proteins were precipitated with 1.5 µg anti-GFP as above. Figure S1:

Supplementary Figures
Pedigree of the family: grey symbols indicate individuals with a borderline diagnosis of ARVC (see Table S1). Slanted bar indicates deceased individual (sudden cardiac death at the age of 19); squares indicate males; circles indicate females; plus (+) and minus (−) signs indicate the presence or absence of the listed mutations; arrow: index patient; question mark: the father was not available for clinical or genetic evaluation.
One daughter of the index patient had died suddenly at the age of 19 years. She was negative for both DSC2 and DSG2 changes. Instead, a heterozygous KCNQ1 R591H missense change was identified, suggestive for sudden cardiac death caused by long QT syndrome 7 . Neither the index patient nor her living daughter were found to carry this KCNQ1 mutation. Hence, the sudden cardiac death in the family was unrelated to the desmosomal mutations.
8 Figure S2:    Table   Table S1: Clinical evaluation of the index patient and her living daughter Summary of clinical evaluation and genetic screening of the index patient and her living daughter (see Figure S1), based on the revised Task Force Criteria 8 . Diagnostic and genetic features suggestive for ARVC are printed in bold. (M -major, m -minor criteria). The summary of the number of major and minor diagnostic criteria is given, suggesting a borderline diagnosis of ARVC in both individuals.