Stress (Takotsubo) Cardiomyopathy -- A Novel Pathophysiological Hypothesis to Explain Catecholamine-induced Acute Myocardial Stunning : The Hypothesis

The Hypothesis

Surges in catecholamine levels are an evolutionary response to sudden shock, fright or danger. Proposed mechanisms for catecholamine-mediated stunning in stress cardiomyopathy include epicardial spasm, microvascular dysfunction, hyperdynamic contractility with midventricular or outflow tract obstruction, and direct effects of catecholamines on cardiomyocytes. We hypothesize that stunning is the result of epinephrine- mediated effects on cardiomyocytes.

Stimulus Trafficking

At physiological and elevated concentrations, norepinephrine, released from the sympathetic nerves, acts predominantly via the β₁-adrenoceptors (β₁ARs) on ventricular cardiomyocytes, exerting positive inotropic and lusitropic responses. These effects are the result of β₁AR coupling to the G_s protein family, which increases intracellular cyclic AMP levels through adenyl cyclase. Elevated cyclic AMP concentrations activate protein kinase A (PKA), which phosphorylates several downstream intracellular targets, resulting in an increased contractile response. Epinephrine also binds β₁ARs and activates this response, but it has a higher affinity for the β₂-adrenoceptor (β₂AR). Humans have a higher concentration of β₂ARs in the ventricular myocardium than other mammals. The ratio of β₁ARs:β₂ARs in normal human ventricular myocardium is approximately 4:1.^[26] Studies with transgenic mice that overexpress human β₂ARs have enabled the pharmacology of the human β₂AR in the ventricular cardiomyocyte to be studied.^[27] At epinephrine concentrations in the normal physiological range, epinephrine binding to β₂ARs activates the G_s protein–adenyl cyclase– PKA pathway, resulting in a positive inotropic response. At higher 'supraphysiological' concentrations, epinephrine stimulates a negative inotropic effect on myocyte contraction.^[28] This change in response results from a switch in β₂AR coupling, from G_s protein signaling to G_i protein signaling,^[29] a process called stimulus trafficking (Figure 2). PKA-mediated phosphorylation of the β₂AR, resulting from intense activation of the β₁AR–G_s protein and β₂AR–G_s protein pathways, is thought to initiate the switch in signal trafficking from β₂AR–G_s protein to β₂AR–G_i protein coupling.^[30,31]

Figure 2. (click image to zoom) Inotropic effects of epinephrine and norepinephrine. (A) Effects of epinephrine and norepinephrine on ventricular myocardium from transgenic mice overexpressing the human β2-adrenoceptor (TG4 mice), and the effects of PTX. (B) Simulations of the ventricular effects of epinephrine and PTX treatment. Abbreviation: PTX, pertussis toxin. Permission obtained from the American Society for Pharmacology and Experimental Therapeutics © Heubach JF et al. (2004) Mol Pharmacol 65: 1313–1322.

Although the bell-shaped concentration– response curve for the function of epinephrine on the human β₂AR has been demonstrated only in the transgenic mouse model, there is evidence for potential β₂AR–G_i protein interactions in human atrial^[32] and ventricular^[33] muscle. Stimulation of β₂AR–G_i protein signaling pathways has been shown to produce a negative inotropic effect on human ventricular myocytes,^[34] although the effect was much more pronounced in cells from a failing human heart, in which G_i protein signaling is increased, than in cells from a healthy heart.^[35] After the surge in epinephrine levels has cleared from the circulation, β₂ARs coupled to G_i proteins either switch back to G_s protein coupling or are internalized and degraded, enabling cardiomyocytes to recover their inotropic function. This sequence of events would explain the reported recovery of ventricular function in individuals with stress cardiomyopathy.

The mechanism of the negative inotropic effect mediated by β₂AR–G_i protein interaction is still under debate. The G_i protein pathway can activate the p38 mitogen-activated protein (MAP) kinase pathway, which exerts a negative inotropic effect.^[36–38] Alternatively, β₂AR–G_i protein signaling could upregulate the sodium– calcium ion exchanger,^[39] inhibit l-type calcium channel currents^[40] or act through other as yet unidentified pathways. At first, these responses seem counterintuitive to the evolutionary need for catecholamine-induced increases in cardiac output. High levels of β₁AR-mediated G_s protein pathway activation induce apoptotic pathways in the cardiomyocyte. The switch to G_i protein signaling via the β₂AR at high epinephrine concentrations might, therefore, have a protective role. β₂AR–G_i protein coupling also activates the phosphoinositide 3 kinase–protein kinase B (Akt) pathway through the G_iβγ subunit, which has an antiapoptotic effect.^[41] This action would counteract the proapoptotic effect of excessive β₁AR–G_s protein pathway activation^[42] and act as a physiological balance to prevent excessive catecholamine-mediated damage. Patchy apoptosis could still occur before, or despite, the activation of G_i-protein-dependent pathways, explaining the elevation of troponin levels and necrosis seen in patients following stress cardiomyopathy. The scale of myocardial injury is probably reduced compared with the scenario of unopposed activation mediated by β₁AR–G_s protein and β₂AR–G_s protein pathways.

Negative inotropism mediated by epinephrine, β₂AR and G_i protein interactions can explain the propensity for apical suppression with basal sparing in stress cardiomyopathy. Sympathetic stimulation of adrenoceptors in the ventricular myocardium is achieved through two routes: local release of norepinephrine by sympathetic nerve endings, directly innervating the myocardium; and diffusion of circulating catecholamines into the myocardium from the coronary circulation. In normal human hearts, the density of sympathetic nerve endings, as identified by tyrosine hydroxylase during autopsy, is approximately 40% higher in the basal myocardium than in the apical myocardium.^[43] Sympathetic innervation of the myocardium in the canine left ventricle shows a similar pattern, with the highest density of nerve endings found at the base, decreasing to the lowest levels at the apex.^[44] In the normal physiological setting, the majority of norepinephrine is released from nerve terminals, with circulating norepinephrine released from the adrenal medulla making a minimal contribution. The innervation pattern, therefore, is inconsistent with the region of greatest dysfunction found in stress cardiomyopathy.

Provided that perfusion is balanced, circulating catecholamines have a global effect on the myocardium. The magnitude of the effect will depend on the local density of adrenoceptors in different regions of the myocardium. Mori et al. demonstrated that the canine heart has a higher concentration of β-adrenoceptors in the apical myocardium, with the concentration gradient decreasing from apex to base (455 vs 341 fmol/mg protein).^[44] This difference in distribution resulted in a greater contractile response to catecholamine challenge in the apical myocardium than in the basal myocardium. The proposed explanation for this difference is that the density of β-adrenoceptors in the apical myocardium is increased to compensate for the decrease in direct sympathetic innervation, to maintain a balanced responsiveness of the ventricle to sympathetic drive. This difference implies that the apex might be more sensitive than the basal myocardium to circulating catecholamines and that, under conditions of stress, the circulating catecholamine is predominantly epinephrine. Mori et al. did not, however, differentiate between β₁ARs and β₂ARs in their study, and this greater density of adrenoceptors at the apex has not been assessed in the human ventricle. Their observation is supported by findings from models of heart failure induced by either acute or chronic catecholamine infusion.^[45,46] These models demonstrate increased myocardial fibrosis in the apical ventricular myocardium, indicating that the apical myocardium has a raised sensitivity to circulating catecholamines and in particular the β-adrenoceptor agonist isoprenaline used in these studies. An increasing density of β₂ARs from the base to the apex could explain the regional difference in response to high catecholamine levels, with circulating epinephrine having a greater influence on apical, relative to basal, function (Figure 3).

Figure 3. (click image to zoom) Schematic representation of the regional differences in response to high catecholamine levels, explaining stress cardiomyopathy.

We do not discount the role of norepinephrine in stress cardiomyopathy, given the systemic activation of the sympathetic nervous system in response to sudden shock. Norepinephrinemediated coronary vasospasm might have an additional role, and there is evidence of coronary or inducible spasm at provocation in some patients.^[47] This response is unsurprising following a massive surge in norepinephrine levels, although other investigators have found no evidence of epicardial or microcirculatory flow abnormalities in stress cardiomyopathy.^[48] Coronary vasospasm could impose a secondary ischemic insult, superimposed on the primary epinephrine-induced apical stunning.

Sex-related Differences in Stress Cardiomyopathy

Many unanswered questions regarding stress cardiomyopathy remain. One of the most puzzling concerns why there is an apparent increased incidence in females, who comprise over 90% of reported cases. Sex-related differences in the response of the adrenal medulla to sudden high-intensity sympathetic discharge and differing pharmacokinetics of epinephrine release could explain the increased rate in women. Of interest, basal plasma epinephrine levels are lower in women than in men.^[49] This difference could reflect reduced synthesis, increased degradation or reduced basal release with more potential stores for sudden release. Estrogens have cardioprotective effects against acute injury through a variety of complex mechanisms.^[50,51] Stress activates early gene expression in both the central nervous system and the ventricular myocardium in rodent models,^[52,53] the myocardial changes in gene expression being mediated by activation of both α-adrenoceptors and β-adrenoceptors. Estrogen reduces these changes in gene expression, protecting against the apical ventricular dysfunction observed in this rodent model of stress cardiomyopathy induced by conscious immobilization.^[54] Chronic (but not acute) exposure of the rat ventricular myocardium to estrogen reduces the enhanced expression of β₁ARs that occurs in response to their activation by catecholamines and ischemia– reperfusion injury.^[55] Furthermore, oophorectomy increases the expression of β₁ARs, an effect that is reversed by estrogen supplementation.^[56] Beyond the myocardium, greater vascular β₂AR-mediated sensitivity has been demonstrated in women than in men.^[57] Estrogens could, therefore, influence the β₁AR:β₂AR signaling ratio in women in favor of the protective effects of β₂AR–G_i protein signaling following surges in catecholamine levels. This protection would occur at the mechanical cost of negative inotropism in the regions with the highest density of β-adrenoceptors, namely the apical myocardium. Perhaps men who lack this protective 'dampening' effect on β₁AR–G_s protein signaling develop more-intense acute cardiotoxicity mediated by β₁AR–G_s protein signaling following surges in catecholamine levels, resulting in a fatal event rather than stress cardiomyopathy.

Atypical Stress Cardiomyopathy

The number of reports of patients with an atypical or 'inverted Takotsubo' pattern of disease is increasing, with basal ventricular suppression, sometimes extending to the midventricular myocardium, but with apical sparing.^{[22,25,58,59]} The mechanism described above is derived from a scenario in which the entire ventricular myocardium is exposed to the same high concentration of circulating epinephrine released from the adrenal medulla. In this scenario, the gradient of the density of β₂ARs explains the predominance of apical suppression. Other factors can also influence the epinephrine concentration in the myocardium, however, and these could account for the different local concentration gradients and phenotypes observed. For example, conversion of norepinephrine to epinephrine by phenylethanolamine N-methyltransferase in the ventricular myocardium has been demonstrated in rabbits,^[60] and this would occur at points of highest sympathetic innervation.