Catalytic antibodies in arrhythmogenic cardiomyopathy patients cleave desmoglein 2 and N-cadherin and impair cardiomyocyte cohesion

Figures

• 

  Fig. 1. Detection of antibodies against ICD proteins in human left ventricular tissue by immunofluorescence analysis. Human left ventricular tissue was incubated with respective AC-IgGs and control IgGs as indicated in the figure. N-CAD was used as a marker for ICDs, and presence of anti-ICD proteins was defined when there was an overlap of IgG staining with N-CAD (white arrows). The yellow arrow in control IgG shows no positive staining of ICDs. Anti-DSG2 antibody was used to show the presence of DSG2 in the cardiac tissue used. AC patients were grouped into DPC (harboring desmoplakin gene mutations) and ARVC (harboring plakophilin2 gene mutations). Images represent immunofluorescence analysis of three repeats. Scale bar: 10 µm

• 

  Fig. 2. Dissociation assays in HL-1 cells and murine cardiac slices. A Dissociation assays were performed in HL-1 cells (a cartoon illustrates the method) treated with (B) DPC patient IgGs and (C). ARVC patient IgGs for 24 h. The increase in the number of fragments correlates to a decrease in cell cohesion and vice versa. In each experiment, duplicate or triplicate wells were incubated with respective IgGs, and at least a minimum of three biological repeats were performed. Bar graph represents fold change compared to healthy control IgG (IgG). Data are presented as mean ± SEM. One-way ANOVA Holm–Šídák's multiple comparisons test was performed. D Dissociation assay in murine cardiac slice cultures (a. cartoon illustrates the method) obtained from wild-type mice treated with E AC1 (from DPC group as in B) and F AC15 (from ARVC group as in C). N = 6–8 mice. Bar graph represents fold change compared to healthy control IgGs (IgG). Data are presented as mean ± SEM. Paired t-test. *Indicates p < 0.05 **indicates p < 0.005

• 

  Fig. 3. In vitro DSG2 and N-CAD cleavage assays. A Human DSG2-Fc protein (200 ng/lane) or B N-CAD-Fc protein (200 ng/lane) was incubated with respective IgGs from AC patients (AC1-AC15), water (H2O), healthy control IgG (IgG), PV-IgG and AK23 (murine pemphigus monoclonal anti-DSG3 antibody, as a negative control) for 4 h, without (C) and with protease inhibitor (I, cOmplete™), and Western blot analysis was performed. A representative image of three experimental repeats is shown. #Indicates the cleaved fragments. *Indicates mouse IgG heavy chain

• 

  Fig. 4. Homophilic DSG2 and N-CAD interaction probabilities measured by AFM. A Schematic presentation of an AFM interaction experiment in cell-free conditions. Recombinant DSG2/N-CAD extracellular domain-containing proteins tagged with Fc fragments were covalently linked via a PEG linker to the AFM tip and the mica sheet. A laser is directed to the cantilever tip and reflected onto a photodetector. Each interaction event leads to a deflection of the cantilever, detected by the laser reflected on the photodetector, and a force–distance curve of a specific binding event was produced, together with a topography image and adhesion events. Quantification of DSG2 binding frequency expressed in relative interaction probability. B Desmoplakin-mutated AC patients (DPC), C plakophilin 2-mutated patients (ARVC). D Quantification of N-CAD binding frequency expressed in relative interaction probability in DPC. Each data point represents 1000 curves analyzed across two areas (10 µm × 10 µm). Data are presented as mean ± SEM.*p ≤ 0.05 compared to IgG, and NS is not significant compared to IgG. One-way ANOVA with Holm–Šídák's multiple comparisons test was performed. N = 3–5

• 

  Fig. 5. Protein expression analysis and dissociation assays in HL-1 cells treated with AC-IgGs. A Western blot of lysates prepared from HL-1 cells treated with respective IgGs for 24 h. B Western blot quantification of pp38MAPK for the blots shown in A. Ratio between pp38MAPK and total p38MAPK was normalized to healthy control IgG (IgG), and means represents ± SEM. N = 3–4. C Western blots of Triton-X-100 lysates prepared after incubating HL-1 cells with mentioned IgG samples for 24 h. A supernatant fraction represents non-cytoskeleton bound proteins, and a pellet fraction represents cytoskeleton bound proteins. F/R (Forskolin/Rolipram) treatments served as a positive experimental control for the activation of p38MAPK. D Western blot quantification of pp38MAPK in the supernatant fractions of the blots shown in B. Ratio between pp38MAPK and total p38MAPK was normalized to IgG-control (IgG), and means represents ± SEM. N = 4–6. *Indicates p < 0.05, **indicates p < 0.005 compared to healthy control IgGs (IgG); #indicates p < 0.05 compared to control (Ctrl) and NS is not significant. One-way ANOVA with Holm–Šídák's multiple comparisons test was performed between IgG samples, and a Student’s…

• 

  Fig. 6. Catalytic antibodies mediated cardiomyocyte cohesion in AC patients. In normal healthy individuals, the cadherin family proteins, desmoglein 2 (DSG2) and N-cadherin (N-CAD) proteins are properly localized to the ICD via the plaque proteins DSP and PKP2 and armadillo proteins PG, α-catenin (α), β-catenin (β) and YAP (Y), respectively. In AC patients, having mutations in either DSP or PKP2 genes (the corresponding proteins are represented with dashed boxes), presence of catalytic antibodies (CA) that can cleave DSG2 or N-CAD could destabilize the cadherin-mediated adhesion (1), which could lead to the activation of p38MAPK (2), which further can lead to desmosome destabilization. Cleavage of cadherin proteins combined with p38MAPK activation could eventually result in reduced cardiomyocyte cohesion (3) and thereby a reduction in the mechanical strengths of the cardiomyocyte. Cartoon created with BioRender.com