Defective mitochondrial calcium uptake underlies arrhythmias in Barth syndrome
Cardiac manifestations in patients with Barth syndrome comprise heart failure and arrhythmias. In previous (yet unpublished) studies, we have identified a mechanism how in another hereditary cardiomyopathy, i.e., hypertrophic cardiomyopathy (HCM), which is characterized by increased Ca2+ affinity of the myofilaments as the primary culprit, arrhythmias are induced by a mitochondrial redox mismatch. This mismatch is caused by an imbalance between ADP-induced oxidation (related to increased ATP consumption at the myofilaments) and Ca2+-induced reduction of the mitochondrial redox state of NAD(P)H/NAD(P)+, which causes a net depletion of the NADPH-coupled antioxidative capacity and the emission of hydrogen peroxide (H2O2) from mitochondria. These reactive oxygen species (ROS) trigger and sustain arrhythmias, where the trigger is related to spontaneous Ca2+ release from the sarcoplasmic reticulum, which induces Ca2+ waves and thereby, spontaneous contractions of isolated cardiac myocytes.
Since Barth syndrome is one (albeit very rare) differential diagnosis for HCM, we analyzed excitation-contraction coupling, mitochondrial redox state and arrhythmias in a mouse model of Tafazzin knock-down (Taz-KD). Compared to wild-type (WT) controls, Taz-KD mice developed a form of heart failure with smaller and not hypertrophied hearts compared to WT, slightly decreased systolic function (ejection fraction reduced from 75% to 55% at 1 year, and global longitudinal strain also reduced) and thereby, reduced cardiac output. This was associated with a gradual and substantial increase of left ventricular mRNA expression of atrial natriuretic peptide (ANP; 15-fold after 1 year) as an index of increased left ventricular filling pressures. Importantly, early (passive) diastolic filling of the ventricles, reported by the presence of an E-wave, was substantially impaired in Taz-KD versus WT mice at 1 year. This was reflected by maintained systolic, but substantially impaired diastolic function of isolated cardiac myocytes (the rate of sarcomere relengthening was doubled). This was explained by increased Ca2+ affinity of the myofilaments, as determined by the pCa/force relationship in skinned cardiac myocytes.
On isolated cardiac mitochondria, respiration was only slightly reduced in Taz-KD versus WT, but mitochondrial Ca2+ uptake was absent due to a selective downregulation of mitochondrial Ca2+ uniporter (MCU) protein, but not mRNA. On isolated cardiac myocytes, this prevented mitochondrial Ca2+ uptake during β-adrenergic stimulation, which is required to activate the Krebs cycle dehydrogenases. Consequently, the redox states of NAD(P)H and FAD were strongly oxidized during this workload transition. At the end of this stimulation protocol, the rate of spontaneous sarcomere shortenings (as an index for cellular arrhythmias) was increased in Taz-KD versus WT myocytes. When pretreating cardiac myocytes with CGP-37157, an inhibitor of the mitochondrial Na/Ca exchanger (NCLX), the redox state of NAD(P)H and FAD became more reduced and cellular arrhythmias were effectively prevented.
In conclusion, in Barth syndrome, the defective cardiolipin biosynthesis leads to a loss of MCU protein, which in turn abolishes mitochondrial Ca2+ uptake during physiological increases in workload, such as during exercise. Furthermore, a (presumably secondary) increase in myofilament Ca2+ sensitivity increases the energetic demand of excitation-contraction coupling and impairs diastolic (more than systolic) function. Together, this results in a mismatch of ADP-induced oxidation and Ca2+-induced regeneration of Krebs cycle-related NADH and NADPH production, where the former may contribute to the energetic deficit and the latter to oxidative stress, which in turn may trigger arrhythmias. Strategies to improve mitochondrial Ca2+ uptake, for instance by inhibiting NCLX, may therefore be beneficial in patients with Barth syndrome. For some conventional Ca2+ antagonists, such as diltiazem, NCLX-inhibiting properties have been described and therefore, this could be explored for potential drug repurposes efforts for Barth syndrome.
Christoph Maack, Jan Dudek and Edoardo Bertero,
University Clinic Wuerzburg