Modeling the Aging Heart: From Local Respiratory Defects to Global Rhythm Disturbances

Abstract

In this issue, Baris et al. (2015) describe cardiac rhythm abnormalities in a mouse model of mitochondrial dysfunction in widely distributed cells of the aging human heart. How do a few metabolically challenged cells disrupt cardiac rhythm? We suggest that these cells provide “crystallization centers” for latent dysfunctional zones to allow arrhythmia emergence.

Disturbances in cardiac rhythm (arrhythmias) cause substantial morbidity and mortality. While the management of cardiac arrhythmias has been revolutionized over the past 50 years, many challenges remain (Nattel et al., 2014). Arrhythmias are particularly prevalent in the elderly (Chow et al., 2012). A characteristic feature of aging hearts is the accumulation of cells with severe mitochondrial dysfunction due to accumulated mitochondrial DNA mutations (Khrapko et al., 1999). In this issue of Cell Metabolism, Baris et al. (2015) report the intriguing finding that focal deficiencies in mitochondrial oxidative metabolism of about 0.5% of cells in the mouse heart promote cardiac arrhythmias during aging. A similar frequency of metabolically deficient cardiomyocytes is observed in the aged human heart (Müller-Höcker, 1989), suggesting that the same phenomena might contribute to cardiac rhythm instabilities in elderly humans.

Unlike the elderly human heart, the heart of aged mice does not typically contain mitochondrially deficient cells. To create a model of mitochondrially deficient cells in human cardiac aging, the authors created mice with cardiac-specific expression of a dominant-negative mutant of Twinkle, the mitochondrial helicase. Twinkle mutants cause an excess of large deletions in mitochondrial DNA, which clonally expand to high fractions in a small number of cells, causing their oxidative phosphorylation to fail. Consequently, the tissue becomes a mosaic with occasional deficient cells (~0.5% of cardiomyocytes in Baris et al., 2015) surrounded by normal cells.

This finding poses a conceptual problem: how could such a low proportion of defective components disrupt the functioning of the entire heart? Hypotheses have been proposed to explain how these rare cells could affect an entire aging organ, as caused by mtDNA mutations in diverse tissues including skeletal muscle and colon. In skeletal muscle, the mitochondrially deficient zones are short; thus while their overall presence is at ~1%, almost every muscle fiber contains such a zone somewhere along its length. Judd Aiken suggested that these deficient zones produce local degeneration and disconnection, weakening the entire fiber (Wanagat et al., 2001). In aged colonic mucosa, ~15% of cells are mitochondrially deficient, constituting entirely deficient colonic crypts that have been proposed to act as penetration points in the old colon (Khrapko and Turnbull, 2014). None of the above hypotheses are, however, pertinent to the heart. What causes arrhythmias in Twinkle mice?

The arrhythmias in the aged Twinkle mice manifested as spontaneous premature ventricular contractions (PVCs) and pauses in rhythm that the authors interpret as atrioventricular block. All indices of global electrical function (reflecting sinoatrial-node pacemaker activity, electrical conduction, and repolarization) were normal, and no enhanced mortality was seen. Rhythm disturbances were only observed with the convergence of two factors: aging and swimming stress. No sustained arrhythmias occurred, making re-entrant mechanisms unlikely and suggesting that the electrical dysfunction was predominantly the result of focal ectopic activity. Focal ectopic beats are initiated by oscillations of cardiomyocyte transmembrane voltage called afterdepolarizations, both early (EADs) and delayed (DADs). Indeed, disruption of intracellular Ca²⁺ regulation, known to occur in the aged heart through mitochondrially derived oxidative modification of sarcoplasmic-reticulum Ca²⁺-release channels (Cooper et al., 2013), causes such oscillations. While aged heart cells are more likely to produce EADs/DADs, how do small numbers of isolated mitochondrially deficient cells enhance this phenomenon?

The generation and propagation of cellular EADs/DADs are controlled by the electrototonic effect of the surrounding normal cells acting as a current sink (termed source-sink mismatch). The increased frequency of PVCs in aged Twinkle mice indicates a diminished likelihood of source-sink mismatch. The mismatch can be diminished in two ways. One is to have a larger number of adjacent cells producing EADs/DADs (increased source). The second is to decrease the electrical load exerted by the surrounding cells (reduced sink).

Arrhythmias were only seen in Twinkle mice under stress (swimming). Intense exercise under a psychologically stressful condition, as occurs with the forced-swimming paradigm, greatly increases myocardial metabolic demands and engages the sympathetic nervous system. Sympathetic enhancement causes Ca²⁺ loading and Ca²⁺ release-channel phosphorylation, both of which increase EAD/DAD likelihood (Chen et al., 2014). In addition, the associated greatly increased metabolic demand challenges mitochondrially deficient cells, causing them to maximize anaerobic metabolism with the production of large quantities of toxic products typically seen with acute myocardial ischemia (Carmeliet, 1999). These toxic metabolites are free to diffuse from the cell of origin and affect a substantial number of neighboring cells, producing significant amplification and, along with sympathetic enhancement, creating a “perfect storm” for EAD/DAD source generation.

In addition, mitochondrially deficient cardiomyocytes in aged Twinkle mutant mice could also contribute by lowering the electrotonic load in the vicinity of EAD/DAD-producing sources. Under stress conditions, these cells likely lose their capability to generate action potentials and thus no longer constitute a current sink to the EAD/DAD-producing cell(s), decreasing the regional source-sink mismatch. Given that electrotonic influences extend up to 1 mm, it is conceivable that a number of mitochondrially deficient myocytes could be found in the neighborhood of an EAD/DAD-generating source, reducing the electrical sink. In this scenario, rare dysfunctional cells might be understood conceptually as “centers of crystallization” for afterdepolarizations. In the absence of such “crystallization centers,” a great majority of emerging afterdepolarization “attempts” would be dampened by electrotonic effects of surrounding cells. Mitochondrially deficient myocytes in the immediate surrounding of “attemptor cells” may significantly increase its chances to propagate and precipitate an arrhythmic event.

Either way, one could conceive of the sporadically distributed mitochondrially deficient cells as “crystallization centers,” quiescent at rest but showing enhanced disturbances under stress conditions that alter their electrical properties and affect their aged (and thus not completely normal) neighbors in ways that generate focal arrhythmias.

In conclusion, Baris et al. (2015) have made an important contribution by creating a mouse model of a previously cryptic change in the aging human heart—the appearance of sporadic mitochondrially deficient cells due to accumulated mitochondrial DNA mutations. They demonstrate that reproducing this property of aging predisposes to stress-induced cardiac rhythm disturbances. More work is needed to understand the detailed mechanisms and importance of these electrical abnormalities, but these observations and the availability of this new model will contribute to solving the important enigma of heart disease in the elderly.