Mitochondrial Dysfunction in Atrial Cardiomyocytes Associates with Atrial Fibrillation

A. Gassa (Köln)1, S. C. R. Erlhöfer (Köln)2, M. Limmer (Köln)3, V. Siebert (Köln)3, N. Mader (Köln)4, L. Conradi (Köln)4, D. Steven (Köln)2, R. Wiesner (Köln)5
1Universitätsklinikum Köln Klinik und Poliklinik für Herz- und Thoraxchirurgie, Herzzentrum Köln, Deutschland; 2Herzzentrum der Universität zu Köln Elektrophysiologie Köln, Deutschland; 3Institut für vegetative Physiologie Universität zu Köln Köln, Deutschland; 4Universitätsklinikum Köln Klinik und Poliklinik für Herzchirurgie Köln, Deutschland; 5Universitätsklinikum Köln Institut für Vegetative Physiologie Köln, Deutschland

Introduction: Organs from elderly individuals show a mosaic pattern, with rare cells presenting severe mitochondrial dysfunction due to the stochastic accumulation of mitochondrial DNA deletions and duplications (indels). This mechanism is regarded as one of several contributing factors to organ failure associated with physiological ageing. As atrial fibrillation represents an age-related cardiac disorder, it was hypothesized that an increased prevalence of dysfunctional cardiomyocytes of this type would be detected in patients presenting with cardiac arrhythmia. This hypothesis was supported by previous findings demonstrating that induction of a comparable genetic mosaic in transgenic mice—characterized by the accumulation of indels throughout the myocardium—results in the development of severe ventricular arrhythmias. 

Methods: Fresh atrial tissue slices obtained from patients over 65 years of age undergoing cardiac surgery were subjected to histochemical staining for cytochrome c oxidase (COX) and succinate dehydrogenase (SDH) enzymatic activities. This dual staining allows the identification of COX-deficient cardiomyocytes (COX⁻ cells), which appear blue due to SDH activity in the absence of COX function, as COX requires three essential subunits encoded by mitochondrial DNA (mtDNA). The proportion of COX⁻ cells was quantified using ImageJ software.

Results: The proportion of COX-negative (COX⁻) cells was 0.01% ± 0.01 in patients in sinus rhythm and 0.05% ± 0.04 in patients with atrial fibrillation (AF) (p < 0.05), consistent with observations in the mouse model. Comparison between AF subtypes revealed 0.07% ± 0.08 COX⁻ cells in persistent AF and 0.11% ± 0.14 in paroxysmal AF (p = 0.13), indicating no significant difference. Analysis of ventricular tissue from both groups also showed no variation in COX⁻ cell frequency.

Discussion: The age-related accumulation of stochastically distributed COX-negative (COX⁻) cells in the human atrium is postulated to contribute significantly to the development of atrial fibrillation. Comprehensive analysis of the cellular proteome in laser-microdissected cardiomyocytes, including membrane ion transporters and ion channels, is expected to elucidate the mechanisms by which these cells generate intermittent action potentials.