Plakophilin-2 Haploinsufficiency Causes Calcium Handling Deficits and Modulates the Cardiac Response Towards Stress

Abstract

Human variants in plakophilin-2 (PKP2) associate with most cases of familial arrhythmogenic cardiomyopathy (ACM). Recent studies show that PKP2 not only maintains intercellular coupling, but also regulates transcription of genes involved in Ca²⁺ cycling and cardiac rhythm. ACM penetrance is low and it remains uncertain, which genetic and environmental modifiers are crucial for developing the cardiomyopathy. In this study, heterozygous PKP2 knock-out mice (PKP2-Hz) were used to investigate the influence of exercise, pressure overload, and inflammation on a PKP2-related disease progression. In PKP2-Hz mice, protein levels of Ca²⁺-handling proteins were reduced compared to wildtype (WT). PKP2-Hz hearts exposed to voluntary exercise training showed right ventricular lateral connexin43 expression, right ventricular conduction slowing, and a higher susceptibility towards arrhythmias. Pressure overload increased levels of fibrosis in PKP2-Hz hearts, without affecting the susceptibility towards arrhythmias. Experimental autoimmune myocarditis caused more severe subepicardial fibrosis, cell death, and inflammatory infiltrates in PKP2-Hz hearts than in WT. To conclude, PKP2 haploinsufficiency in the murine heart modulates the cardiac response to environmental modifiers via different mechanisms. Exercise upon PKP2 deficiency induces a pro-arrhythmic cardiac remodeling, likely based on impaired Ca²⁺ cycling and electrical conduction, versus structural remodeling. Pathophysiological stimuli mainly exaggerate the fibrotic and inflammatory response.

Keywords: arrhythmogenic cardiomyopathy; calcium handling; cardiac pressure overload; exercise; fibrosis; inflammation; plakophilin-2; second hit.

Figure 1

Remodeling of proteins involved in calcium signaling pathways in the PKP2-Hz mouse. Representative western blots (left) and average densitometry (right; n = 5 for all groups) of ankyrinB (AnkB), calsequestrin-2 (Casq2), Ca_v1.2, and SERCA2a, measured from wildtype and PKP2-Hz ventricular lysates of three-month-old mice. Left upper panels represent western blots, bottom panels according to ponceau staining used for quantification (mean ± SEM, * p < 0.05).

Figure 2

Absence of cardiac remodeling in three and six months old PKP2-Hz mice. (A) Hematoxylin and eosin staining of wildtype and PKP2-Hz heart sections, at three and six months of age; scale bar, 100 µm. (B) Representative pictures of Picrosirius red staining (left panel) in the inner myocardium of wildtype and PKP2-Hz murine hearts of both ages; scale bar, 500 µm. Quantification of collagen abundance (right panel) in overview slides of whole heart sections (WT 3 months; n = 7, PKP2-Hz three months; n = 7, WT six months; n = 6, PKP2-Hz six months; n = 6, mean ± SEM). (C) Relative mRNA expression of Col1α1, Col1α2, MM9, Timp1, NF-κß, and IL-6 assessed by RT-qPCR in three (left) and six month old (right), wildtype and PKP2-Hz ventricular tissue (three months; n = 6 for all groups, six months; n = 4 for all groups, mean ± SEM, * p < 0.05). (D) Quantification of heart weight (HW) to body weight (BW) ratio (WT three months; n = 8, PKP2-Hz three months; n = 8, WT six months; n = 10, PKP2-Hz six months; n = 10, mean ± SEM, * p < 0.05). (E) Percentage of Langendorf-perfused hearts within each group susceptible for sustained ventricular arrhythmias by epicardial pacing (WT three months; n = 8, PKP2-Hz three months; n = 8, WT six months; n = 10, PKP2-Hz six months; n = 10). (F) Representative western blots (bottom) and average densitometry of N-cadherin (Ncad), connexin 43 (Cx43), and plakoglobin (PKG; top, WT; n = 6, PKP2-Hz n = 5 for Cx43 and PKG, WT; n = 4, PKP2-Hz n = 5 for Ncad), measured from wildtype and PKP2-Hz ventricular lysates, six months of age. Upper panels represent western blots, bottom panels according ponceau staining used for quantification (mean ± SEM, * p < 0.05). (G) Immunofluorescence staining for Cx43 (green) and Ncad (red) in wildtype (WT) and PKP2-Hz (PKP2) heart sections of three (3M) and six months (6M) of age; Scale bar, 100 µm.

Figure 3

Pro-arrhythmic cardiac remodeling in PKP2-Hz mice exposed to exercise training. (A) Representative pictures of Picrosirius red staining (left panel) in the inner myocardium of wildtype and PKP2-Hz murine heart sections, of mice exposed to exercise training; scale bar, 500 µm. Quantification of collagen abundance (right panel) in overview slides of whole heart sections (WT; n = 5 and PKP2-Hz; n = 5, mean ± SEM). (B) Hematoxylin and eosin staining of wildtype and PKP2-Hz heart sections, of mice exposed to exercise; scale bar, 100 µm. (C) Quantification of heart weight (HW) to body weight (BW) ratio in both groups (WT; n = 5 and PKP2-Hz; n = 5, mean ± SEM). (D) Representative western blots (left panel) and average densitometry (right panels, n = 5 for all groups) of AnkyrinB (ANKB), Ca_v1.2, and calsequestrin-2 (Casq2) measured from wildtype (WT) and PKP2-Hz ventricular lysates of mice exposed to exercise training. Left upper panels represent western blots, bottom panels according ponceau staining used for quantification (mean ± SEM, * p < 0.05). (E) Immunofluorescence staining for Cx43 (red) and Ncad (green) in right ventricular heart sections of WT and PKP2-Hz mice; scale bar, 100 µm. Upper right panel shows magnified shots of according merged pictures, indicated by the white boxes. Lateral Cx43 expression is more pronounced in PKP2-Hz hearts exposed to exercise training, indicated by white arrows. (F) Left ventricular fractional shortening examined via echocardiography in hearts of wildtype (WT) and PKP2-Hz mice after one-month of training (WT; n = 5 and PKP2-Hz; n = 5, mean ± SEM). (G) Conduction (CV) velocity measured by epicardial mapping on the left ventricle (LV) and right ventricle (RV) in longitudinal (left) and transverse (right) directions in hearts of wildtype (WT) and PKP2-Hz mice (WT; n = 5 and PKP2-Hz; n = 5, mean ± SEM, * p < 0.05). (H) Percentage of Langendorf-perfused hearts, within each group, susceptible for sustained ventricular arrhythmias by epicardial pacing (WT; n = 5 and PKP2-Hz; n = 5).

Figure 4

Pro-fibrotic cardiac remodeling in PKP2-Hz trans aortic constriction (TAC) operated mice. (A) Hematoxylin and eosin staining of wildtype and PKP2-Hz heart sections of Sham and TAC operated mice; scale bar, 100 µm. (B) Quantification of heart weight (HW) to body weight (BW) ratio in all four groups (WT Sham; n = 5, PKP2-Hz sham; n = 8, WT TAC; n = 8, and PKP2-Hz TAC; n = 9, mean ± SEM, ** p < 0.01). (C) Left ventricular fractional shortening examined by echocardiography in wildtype (WT) and PKP2-Hz mice after eight weeks of TAC/Sham surgery (WT Sham; n = 5, PKP2-Hz sham; n = 8, WT TAC; n = 8, and PKP2-Hz TAC; n = 9, mean ± SEM, *** p < 0.001). (D) QRS duration (left) and ventricular repolarization (QT) time corrected for heart rate (QTc; right) examined via electrocardiograms in wildtype (WT) and PKP2-Hz mice after eight weeks of surgery (WT Sham; n = 5, PKP2-Hz sham; n = 8, WT TAC; n = 8, and PKP2-Hz TAC; n = 9, mean ± SEM, *** p < 0.001). (E) Percentage of Langendorf-perfused hearts, within each group, susceptible for sustained ventricular arrhythmias by epicardial pacing (WT Sham; n = 5, PKP2-Hz sham; n = 8, WT TAC; n = 8, and PKP2-Hz TAC; n = 9). (F) Immunofluorescence staining for Cx43 (green) and Ncad (red) in wildtype and PKP2-Hz (PKP2) heart sections of mice exposed to Sham (SH) or TAC surgery; scale bar, 100 µm. (G) Representative pictures of Picrosirius red staining (left panel) in the inner myocardium of all four groups; scale bar, 500 µm. Quantification of collagen abundance (right panel) in overview slides of whole heart sections (WT Sham; n = 5, PKP2-Hz sham; n = 8, WT TAC; n = 8, and PKP2-Hz TAC; n = 9, mean ± SEM, * p < 0.05). (H) Quantification of collagen abundance (right panel) in the 0.2 mm subepicardial layer, examined in overview slides of whole heart sections (WT Sham; n = 5, PKP2-Hz sham; n = 8, WT TAC; n = 8, and PKP2-Hz TAC; n = 9, mean ± SEM).

Figure 5

Exaggerated cardiac fibrotic and inflammatory response in PKP2-Hz mice exposed to autoimmune myocarditis. (A) Representative pictures of Picrosirius red staining (left panel) of the outer myocardial layer in wildtype and PKP2-Hz murine heart sections, of mice after three weeks of experimental auto-immune myocarditis (3wk EAM) and six weeks of experimental auto-immune myocarditis onset (6wk EAM; scale bar, 750 µm). (B) Quantification of collagen abundance in the entire heart (left panel) and 0.2 mm subepicardial region (right panel) of wildtype (WT) and PKP2-Hz (PKP2) mice after three weeks experimental autoimmune myocarditis (EAM) and six weeks EAM onset or without treatment (untreated; WT 3wk EAM; n = 9, PKP2 3wk EAM, n = 11, WT 6wk EAM; n = 16, PKP2 3wk EAM, n = 15, WT untreated; n = 5, and PKP2 untreated, n = 4, mean ± SEM, * p < 0.05). (C) Hematoxylin and eosin (H&E) staining in wildtype and PKP2-Hz heart sections exposed to EAM; scale bar, 500 µm. (D) Quantification of damaged area in the 0.2 mm subepicardial layer of wildtype (WT) and PKP2-Hz (PKP2) mice hearts exposed to EAM or without any treatment, quantification was performed on H&E stained heart sections (WT 3wk EAM; n = 9, PKP2 3wk EAM, n = 11, WT 6wk EAM; n = 16, PKP2 6wk EAM, n = 15, WT untreated; n = 5, and PKP2 untreated, n = 4, mean ± SEM, * p < 0.05, *** p < 0.001). (E) Representative examples of Connexin 43 (Cx43) immunohistochemistry labeling in the subepicardial region of WT and PKP2-Hz heart sections exposed to EAM; scale bar, 100 µm. (F) Representative examples (right) of myosin (red) and vimentin (green) immunofluorescence labeling and quantification of non-myosin area (left) in wildtype and PKP2-Hz heart sections of mice exposed to EAM; scale bar, 100 µm. The non-myosin area was examined in the epicardial layer of both ventricles, areas were normalized by the length of the epicardium (WT 3wk EAM; n = 8, PKP2 3wk EAM, n = 11, WT 6wk EAM; n = 16, and PKP2 6wk EAM, n = 15, mean ± SEM, * p < 0.05, ** p < 0.01). (G) Quantification of Ly6C/6G (neutrophil) coverage area (left) in whole heart and 0.2 mm subepicardial regions of wildtype and PKP2-Hz mice after three weeks EAM onset (WT whole heart; n = 8, PKP2 whole heart, n = 11, WT subepicard; n = 9, and PKP2 subepicard, n = 11, mean ± SEM, * p < 0.05). Representative example (right) of the neutrophil positive area in both groups; scale bar, 500 µm.