Looking deeper into takotsubo's heart

This editorial refers to ‘Metabolic alterations in a rat model of takotsubo syndrome' by Godsman et al., https://doi.org/10.1093/cvr/cvab081. Sato and Uchida1 first described a previously unrecognized, acute cardiac condition in an elderly woman in 1990. Due to the characteristic shape of the significantly dilated end-diastolic appearance of the left ventricle, they coined it ‘takotsubo’ after the Japanese word for a fisherman’s octopus trap due to its physical resemblance. For quite some time, it was considered a rare syndrome only affecting Japanese patients and associated with spontaneous recovery and a very good prognosis. It was only several years later that takotsubo syndrome (TTS) was recognized as mainly affecting post-menopausal women of any ethnicity.2 In a majority of patients, the left ventricular (LV) systolic dysfunction is transient with spontaneous recovery within days or weeks. However, as many as one in six patients have a poor trajectory, developing cardiogenic shock, thromboembolism, decompensated heart failure, and death. Proposed Metabolic pathway of the cardiac myocyte involved in experimental takotsubo syndrome. As takotsubo is triggered by physical or psychological stress in most patients, activation of the sympathetic nervous system was hypothesized to be the major underlying pathophysiological mechanism. Indeed, increased levels of catecholamines, such as norepinephrine and epinephrine,3 as well as increases in plasma endothelin levels4 have been described in the acute setting. As these vasoactive mediators lead to marked coronary constriction and an increased oxygen consumption of myocytes, it is remarkable that troponin levels are generally only minimally elevated. Indeed, patients with TTS clinically present like patients with ST-segment elevation myocardial infarction (STEMI) with angina pectoris with or without, dyspnoea, and often with similar electrocardiographic (ECG) changes. Also, at least the classical apical ballooning type is associated with significant regional wall motion abnormalities of the LV myocardium associated with pronounced increases in the plasma levels of natriuretic peptides. Given these abnormalities, one would also have expected huge increases in troponin levels as is notable in patients with STEMI. Thus, the surge of vasopressor hormones alone is unlikely to explain the acute clinical course of TTS. Besides coronary microvascular vasoconstriction, protective mechanisms, including myocardial stunning, must be considered as such patients do not develop large or even moderately sized infarcts. These observations indicate that transient metabolic and functional alterations must occur in cardiac myocytes during the acute phase of this condition. Early studies into the pathomechanisms of TTS have identified metabolic alterations. Mitochondrial β-oxidation of fatty acids accounts for a majority of myocardial energy requirements in normal aerobic conditions.5 It has been hypothesized that in TTS a metabolic switch would allow down-regulation of the cardiomyocyte contractile apparatus and spare metabolites for other essential cellular functions, thereby protecting the cardiomyocyte from the supply demand mismatch during a catecholamine storm and protecting the myocardium from cellular necrosis and arrhythmias. 18F-Fluorodeoxyglucose positron emission tomography (FDG-PET) studies have demonstrated reduced 18FDG uptake in regions of myocardial dysfunction in patients with TTS.6 This transient reduction in glucose metabolism may be secondary to a reduction in the calcium-mediated translocation of intracellular glucose transporter type 4 (GLUT-4) receptors to sarcolemmal sites or due to a reduction in glycolytic pathway enzymatic activity. 123Iodine-beta-methyl-p-iodophenylpenta-decanoic acid dual-isotope myocardial single-photon emission computed tomography imaging has also shown significant reduction in myocardial fatty acid metabolism in TTS.5 Catecholamine stressed, induced pluripotent stem cells (iPSC)-derived cardiomyocytes from patients with TTS evidenced cytosolic lipid accumulation with an increase in lipid importers CD36 and carnitine-palmitoyl transferase-1C (CPT1C).7 This was supported by TTS mouse models demonstrating lipid accumulation possibly due to inhibition of ApoB lipoprotein-mediated lipid export.8 In contrast to these findings are discordant PET results reporting increased FDG uptake rather than reductions in patients with TTS as others have suggested.9 Using 31Phospor-Cardia Magnetic Resonance spectroscopy, a significantly reduced phosphocreatine (PCr):adenosine triphosphate (ATP) ratio was noted in the acute phase of TTS compared with controls. Although the ratio improved during convalescence, it remained significantly lower than in controls. This highlights the reduction in cardiac energetic status in TTS and indicates impaired mitochondrial energy production.10 Further studies have attempted to explain this ‘energy drain’ in TTS by an increase in nitric oxide (NO) signalling and postulated a role of nitrosative stress in the energy deficit.11 As a consequence of elevated catecholamines and β2-adrenoceptor coupling NO signalling, NO synthase is activated and NO signalling increased. The elevated NO levels in the presence of increased superoxide (O2−) production (as a result of catecholamine stress) led to increased peroxynitrite (ONOO−) anion levels. This instigates redox stress which may contribute to DNA damage and promote both poly (ADP-ribose) polymerase 1 activation and NAD+/ATP depletion. The ATP depletion in turn results in PCr depletion and inhibits glycolysis.11 Of note, Borchert et al. created iPSC-derived cardiomyocytes from patients with TTS and replicated its cardiac phenotype. Short-term catecholamine exposure in vitro induced stress via β1- and in part β2-adrenergic signalling, including activation of cyclic adenosine monophosphate (cAMP) and cAMP-dependent protein kinase A. However, active contractile force was reduced and sensitivity to isoprenaline-stimulated inotropy enhanced and electrical activity and lipid accumulation altered.7 We have to remember that myocardial energy production occurs continuously as ATP, the major energy source is not stored, but rather produced as contraction ensues. Notably, the myocardium is able to utilize different substrates for mitochondrial ATP production including lipids, carbohydrates, amino acids, and ketone bodies. Normally, mitochondria utilize fatty acetyl-coenzyme A and pyruvate which are the primary metabolites of fatty acid and carbohydrate metabolism, respectively. The enzyme CPT1 represents the rate-limiting step for uptake of long-chain acetyl-coenzyme A. Pyruvate is then oxidized by pyruvate dehydrogenase and eventually enters the Krebs cycle. In their paper published in this issue of Cardiovascular Research, Godsman and colleagues12 further investigate myocardial mechanisms involved in TTS in vivo using isoprenaline injections in rodents.8 In female rats treated with catecholamines, LV function and myocardial metabolism were studied at Day 3 and at recovery at Day 7. They observed multiple changes at almost all levels of the primary metabolic pathways of the cardiac myocyte, specifically demonstrating significant alterations in glucose and fatty acid metabolism as well as in the Krebs cycle. As normally fatty acids are the primary source of energy in the heart, they hypothesize that a metabolic switch to glucose may occur under acute conditions, in particular as during the sympathetic surge glucose becomes abundant in the circulation. Indeed, using 18FDG-PET imaging, they found glucose uptake to be increased, and fatty acids and acetyl-carnitine levels to be decreased. This was predominantly due to increased protein levels of GLUT-1 and of the enzymatic activity of hexokinase-2, but not GLUT-4 transporters. As pyruvate levels decreased and hexokinase increased, glucose metabolism also switched from the classical pyruvate production to glucose-6-phosphate (G6P) with consequent elevation of G6P and fructose-6-phosphate levels in the myocardium. This reflects alternate fates of glucose metabolites including possible stimulation of the pentose phosphate pathway, as shown in figure 1. Of note, deficiency of G6P dehydrogenase (G6PD), the first and rate-limiting enzyme of the pentose phosphate pathway, worsens ischaemia–reperfusion injury in mice. This finding is therefore of particular interest as in the acute phase of the TTS when huge wall motion abnormalities are not matched with myocardial necrosis as reflected by only mildly elevated troponin levels. On the other hand, as expected with the switch to glucose metabolism, long-chain fatty acids and acyl-carnitines levels were decreased, while the β-oxidation enzymes CPT1b-RNA and 3-ketoacyl coenzyme thiolase were, probably in a compensatory fashion, increased as was the CPT1 regulator malonyl-coenzyme A. These findings are somewhat in contrast to our studies in TTS iPSC-cardiomyocytes in which a dose-dependent increase in lipid droplets occurred following stimulation with epinephrine.7 As indicated above, expression of the lipid importer CD36 was increased in TTS iPSC-cardiomyocytes both under basal conditions and upon treatment with epinephrine. Also, the lipid translocator CPT1C exhibited lower expression levels in TTS iPSC-cardiomyocytes (rather than the higher expression observed in the rat model).7 Finally, in the study of Godsman et al. the Krebs cycle was also altered as evidenced by reduced levels of intermediates such as α-ketoglutarate and succinyl-carnitine as was the cellular ATP reporter dihydroorotate. The NAD(P)H and FAD were also increased, which resolved by Day 7. Surprisingly, there were no differences in mitochondrial respiratory function, sarcomere shortening, or Ca2+ transients of isolated mitochondria, implying preserved structural integrity of both mitochondria and cardiomyocyte. The authors conclude that in the acute phase of TTS, dysregulation of glucose and lipid metabolic pathways occurs with decreases in final glycolytic and β-oxidation metabolites and reduced availability of Krebs intermediates. This energetic deficit is associated with defective Ca2+ handling and inflammation, but preservation of sarcomeric and mitochondrial integrity. This is a very important and detailed study highlighting differences in various parts of the metabolic pathways in TTS and hypothesizing potential mechanisms to account for the energy deficit seen. Although the study is comprehensive and important using an in vivo model of TTS, there are some important drawbacks. First of all, the rat model has obviously some inherent limitations. As a consequence, not all findings are congruent with what has been observed in TTS iPSC-cardiomyocytes obtained from patients with the disease; in particular as regards the protein and mRNA expression of various enzymes and transporters. The observed differences in enzymatic activity and metabolic intermediaries between different myocardial regions (e.g. base to apex) are also conflicting with previous findings and do not readily explain the regional variation of myocardial contractility characteristic of TTS. Finally, this study remains largely descriptive in nature which limits the findings substantially. For instance, it would have been interesting and technically feasible to genetically silence the pentose phosphate pathway (i.e. the rate-limiting enzyme G6PD) which may be protective under these conditions to provide further mechanistic insight. Furthermore, the importance of the metabolic switch could have been proved by silencing the GLUT-1 transporter prior to catecholamine stimulation. Nevertheless, the authors should be congratulated for their comprehensive findings and these observations set the stage for such further mechanistic studies that are desperately needed to better understand the myocardial recovery in TTS and to advance possible therapeutic strategies. Conflict of interest: The authors do not have any conflicts of interests related to this manuscript. Outside this manuscript, T.