Background: Arrhythmogenic cardiomyopathy (ACM) is a genetically inherited heart muscle disease with incomplete penetrance and variable expressivity. Underlying mutations are found in cell-cell junction proteins such as the desmosomal protein plakophilin-2 (PKP2). Disease severity ranges from mild symptoms to life threatening arrhythmias, fibro-fatty myocardial replacement and sudden death. So far, the cause of this heterogeneity remains unknown, but an interplay of genetic predisposition and external “second-hits” like viral infections are increasingly discussed.

Purpose: Common cardiotropic viruses such as cytomegalovirus may influence onset and progression of the phenotype in genetically predisposed myocardium via recurrent inflammation. Here, we tested this hypothesis and infected a mouse model of Pkp2 (Pkp2^+/-) haploinsufficiency with murine cytomegalovirus (MCMV) to examine onset of cardiomyopathy and associated inflammatory processes.

Methods and Results: At baseline, mice with Pkp2^+/- haploinsufficiency and respective wild-type (Ctr) littermates were investigated at 3 month of age on cardiac structure and function using echocardiography and histology. However, compared to Ctr no cardiac abnormalities were detected. Moreover, Pkp2^+/- hearts were able to compensate for the missing Pkp2 allele at RNA and protein expression level. To test the “second-hit” hypothesis, Pkp2^+/- and Ctr mice were infected with 5x10⁵ plaque forming units MCMV. Examinations were conducted during the acute (6-10 days post infection, dpi), subacute (1 month post infection, mpi) and chronic phase of infection (3 and 6mpi). In response to acute infection, CD8⁺ cytotoxic T-cells infiltrated the myocardium of all infected animals with a peak at 7-8dpi. Apart from the detected immune response, Pkp2 protein expression was significantly decreased in Pkp2^+/- MCMV animals, which however returned to normal levels at 1mpi and remained stable thereafter. Cardiac function was assessed in a time dependent manner by echocardiography and unraveled first changes of decreased ejection fraction at 3mpi, which were further pronounced at 6mpi in Pkp2^+/- MCMV mice, but absent in Ctr MCMV mice. However, left ventricular dilatation or right ventricular dysfunction were not detected at any given time. Interestingly, the observed left ventricular dysfunction could not be explained by fibrotic cardiac remodeling assessed by PSR staining at 6mpi. Therefore, we were questioning whether the involvement of inflammation and immune responses may modulate heart function. First evidence came from immunohistochemistry of anti-CD8⁺ cytotoxic T‑cells which were increased not only during the acute phase, but also 1-6mpi in both MCMV infected groups. Remarkably, those cells appeared even more increased in Pkp2^+/- MCMV heart sections. Further flow cytometry analyses of cardiac CD45⁺ leucocytes at 6mpi confirmed significantly elevated counts of CD8⁺ as well as virus-specific effector T-cells (CD8⁺ KLRG1⁺) compared to mock-infected Ctrs. Interestingly, monocytes/macrophages were significantly increased in Pkp2^+/- MCMV even when compared to Ctr MCMV.