Use of Artificial Intelligence in Improving Outcomes in Heart Disease: A Scientific Statement From the American Heart Association.

A major focus of academia, industry, and global governmental agencies is to develop and apply artificial intelligence and other advanced analytical tools to transform health care delivery. The American Heart Association supports the creation of tools and services that would further the science and practice of precision medicine by enabling more precise approaches to cardiovascular and stroke research, prevention, and care of individuals and populations. Nevertheless, several challenges exist, and few artificial intelligence tools have been shown to improve cardiovascular and stroke care sufficiently to be widely adopted. This scientific statement outlines the current state of the art on the use of artificial intelligence algorithms and data science in the diagnosis, classification, and treatment of cardiovascular disease. It also sets out to advance this mission, focusing on how digital tools and, in particular, artificial intelligence may provide clinical and mechanistic insights, address bias in clinical studies, and facilitate education and implementation science to improve cardiovascular and stroke outcomes. Last, a key objective of this scientific statement is to further the field by identifying best practices, gaps, and challenges for interested stakeholders.

Functional integrity of the T-tubular system in cardiomyocytes depends on p21-activated kinase 1.

p21-activated kinase (Pak1), a serine-threonine protein kinase, regulates cytoskeletal dynamics and cell motility. Recent experiments further demonstrate that loss of Pak1 results in exaggerated hypertrophic growth in response to pathophysiological stimuli. Calcium (Ca) signaling plays an important role in the regulation of transcription factors involved in hypertrophic remodeling. Here we aimed to determine the role of Pak1 in cardiac excitation-contraction coupling (ECC). Ca transients were recorded in isolated, ventricular myocytes (VMs) from WT and Pak1(-/-) mice. Pak1(-/-) Ca transients had a decreased amplitude, prolonged rise time and delayed recovery time. Di-8-ANNEPS staining revealed a decreased T-tubular density in Pak1(-/-) VMs that coincided with decreased cell capacitance and increased dis-synchrony of Ca induced Ca release (CICR) at individual release units. These changes were not observed in atrial myocytes of Pak1(-/-) mice where the T-tubular system is only sparsely developed. Experiments in cultured rabbit VMs supported a role of Pak1 in the maintenance of the T-tubular structure. T-tubular density in rabbit VMs significantly decreased within 24h of culture. This was accompanied by a decrease of the Ca transient amplitude and a prolongation of its rise time. However, overexpression of constitutively active Pak1 in VMs attenuated the structural remodeling as well as changes in ECC. The results provide significant support for a prominent role of Pak1 activity not only in the functional regulation of ECC but for the structural maintenance of the T-tubular system whose remodeling is an integral feature of hypertrophic remodeling.

Methods in cardiomyocyte isolation, culture, and gene transfer.

Since techniques for cardiomyocyte isolation were first developed 35 years ago, experiments on single myocytes have yielded great insight into their cellular and sub-cellular physiology. These studies have employed a broad range of techniques including electrophysiology, calcium imaging, cell mechanics, immunohistochemistry and protein biochemistry. More recently, techniques for cardiomyocyte culture have gained additional importance with the advent of gene transfer technology. While such studies require a high quality cardiomyocyte population, successful cell isolation and maintenance during culture remain challenging. In this review, we describe methods for the isolation of adult and neonatal ventricular myocytes from rat and mouse heart. This discussion outlines general principles for the beginner, but also provides detailed specific protocols and advice for common caveats. We additionally review methods for short-term myocyte culture, with particular attention given to the importance of substrate and media selection, and describe time-dependent alterations in myocyte physiology that should be anticipated. Gene transfer techniques for neonatal and adult cardiomyocytes are also reviewed, including methods for transfection (liposome, electroporation) and viral-based gene delivery.

Functional effects of a tropomyosin mutation linked to FHC contribute to maladaptation during acidosis.

Familial hypertrophic cardiomyopathy (FHC) is a leading cause of sudden cardiac death among young athletes but the functional effects of the myofilament mutations during FHC-associated ischemia and acidosis, due in part to increased extravascular compressive forces and microvascular dysfunction, are not well characterized. We tested the hypothesis that the FHC-linked tropomyosin (Tm) mutation Tm-E180G alters the contractile response to acidosis via increased myofilament Ca(2+) sensitivity. Intact papillary muscles from transgenic (TG) mice expressing Tm-E180G and exposed to acidic conditions (pH 6.9) exhibited a significantly smaller decrease in normalized isometric tension compared to non-transgenic (NTG) preparations. Times to peak tension and to 90% of twitch force relaxation in TG papillary muscles were significantly prolonged. Intact single ventricular TG myocytes demonstrated significantly less inhibition of unloaded shortening during moderate acidosis (pH 7.1) than NTG myocytes. The peak Ca(2+) transients were not different for TG or NTG at any pH tested. The time constant of re-lengthening was slower in TG myocytes, but not the rate of Ca(2+) decline. TG detergent-extracted fibers demonstrated increased Ca(2+) sensitivity of force and maximal tension compared to NTG at both normal and acidic pH (pH 6.5). Tm phosphorylation was not different between TG and NTG muscles at either pH. Our data indicate that acidic pH diminished developed force in hearts of TG mice less than in NTG due to their inherently increased myofilament Ca(2+) sensitivity, thus potentially contributing to altered energy demands and increased propensity for contractile dysfunction.

Ablation of p21-activated kinase-1 in mice promotes isoproterenol-induced cardiac hypertrophy in association with activation of Erk1/2 and inhibition of protein phosphatase 2A.

Earlier investigations in our lab indicated an anti-adrenergic effect induced by activation of p21-activated kinase (Pak-1) and protein phosphatase 2A (PP2A). Our objective was to test the hypothesis that Pak-1/PP2A is a signaling cascade controlling stress-induced cardiac growth. We determined the effects of ablation of the Pak-1 gene on the response of the myocardium to chronic stress of isoproterenol (ISO) administration. Wild-type (WT) and Pak-1-knockout (Pak-1-KO) mice were randomized into six groups to receive either ISO, saline (CTRL), or ISO and FR180204, a selective inhibitor of Erk1/2. Echocardiography revealed that hearts of the Pak-1-KO/ISO group had increased LV fractional shortening, reduced LV chamber volume in diastole and systole, increased cardiac hypertrophy, and enhanced transmitral early filling deceleration time, compared to all other groups. The changes were associated with an increase in relative Erk1/2 activation in Pak-1-KO/ISO mice versus all other groups. ISO-induced cardiac hypertrophy and Erk1/2 activation in Pak-1-KO/ISO were attenuated when the selective Erk1/2 inhibitor FR180204 was administered. Immunoprecipitation showed an association between Pak-1, PP2A, and Erk1/2. Cardiac myocytes infected with an adenoviral vector expressing constitutively active Pak-1 showed a repression of Erk1/2 activation. p38 MAPK phosphorylation was decreased in Pak-1-KO/ISO and Pak-1-KO/CTRL mice compared to WT. Levels of phosphorylated PP2A were increased in ISO-treated Pak-1-KO mice, indicating reduced phosphatase activity. Maximum Ca(2+)-activated tension in detergent-extracted bundles of papillary fibers from ISO-treated Pak-1-KO mice was higher than in all other groups. Analysis of cTnI phosphorylation indicated that compared to WT, ISO-induced phosphorylation of cTnI was blunted in Pak-1-KO mice. Active Pak-1 is a natural inhibitor of Erk1/2 and a novel anti-hypertrophic signaling molecule upstream of PP2A.

Novel bradykinin signaling in adult rat cardiac myocytes through activation of p21-activated kinase.

Although bradykinin (BK) is known to exert effects on the myocardium, its intracellular signaling pathways remain poorly understood. Experiments in other cell types indicated that p21-activated kinase-1 (Pak1), a Ser/Thr kinase downstream of small monomeric G proteins, is activated by BK. We previously reported that the expression of active Pak1 in adult cardiac myocytes induced activation of protein phosphatase 2A and dephosphorylation of myofilament proteins (Ke et al. Circ Res 94: 194-200, 2004). In experiments reported here, we tested the hypothesis that BK signals altered protein phosphorylation in adult rat cardiac myocytes through the activation and translocation of Pak1. Treatment of myocytes with BK resulted in the activation of Pak1 as demonstrated by increased autophosphorylation at Thr423 and a diminished striated localization, which is present in the basal state. BK induced dephosphorylation of both cardiac troponin I and phospholamban. Treatment of isolated myocytes with BK also blunted the effect of isoproterenol to enhance peak Ca(2+) and relaxation of Ca(2+) transients. Protein phosphatase 2A was demonstrated to associate with both Pak 1 and phospholamban. Our studies indicate a novel signaling mechanism for BK in adult rat cardiac myocytes and support our hypothesis that Pak 1 is a significant regulator of phosphatase activity in the heart.

Expression of active p21-activated kinase-1 induces Ca2+ flux modification with altered regulatory protein phosphorylation in cardiac myocytes.

p21-Activated kinase-1 (Pak1) is a serine-threonine kinase that associates with and activates protein phosphatase 2A in adult ventricular myocytes and, thereby, induces increased Ca2+ sensitivity of skinned-fiber tension development mediated by dephosphorylation of myofilament proteins (Ke Y, Wang L, Pyle WG, de Tombe PP, Solaro RJ. Circ Res 94: 194-200, 2004). We test the hypothesis that activation of Pak1 also moderates cardiac contractility through regulation of intracellular Ca2+ fluxes. We found no difference in field-stimulated intracellular Ca2+ concentration ([Ca2+]i) transient amplitude and extent of cell shortening between myocytes expressing constitutively active Pak1 (CA-Pak1) and controls expressing LacZ; however, time to peak shortening was significantly faster and rate of [Ca2+]i decay and time of relengthening were slower. Neither caffeine-releasable sarcoplasmic reticulum (SR) Ca2+ content nor fractional release was different in CA-Pak1 myocytes compared with controls. Isoproterenol application revealed a significantly blunted increase in [Ca2+]i transient amplitude, as well as a slowed rate of [Ca2+]i decay, increased SR Ca2+ content, and increased cell shortening, in CA-Pak1 myocytes. We found no significant change in phospholamban phosphorylation at Ser16 or Thr17 in CA-Pak1 myocytes. Analysis of cardiac troponin I revealed a significant reduction in phosphorylated species that are primarily attributable to Ser(23/24) in CA-Pak1 myocytes. Nonstimulated, spontaneous SR Ca2+ release sparks were significantly smaller in amplitude in CA-Pak1 than LacZ myocytes. Propagation of spontaneous Ca2+ waves resulting from SR Ca2+ overload was significantly slower in CA-Pak1 myocytes. Our data indicate that CA-Pak1 expression has significant effects on ventricular myocyte contractility through altered myofilament Ca2+ sensitivity and modification of the [Ca2+]i transient.

p21-Activated kinase-1 and its role in integrated regulation of cardiac contractility.

We review here a novel concept in the regulation of cardiac contractility involving variations in the activity of the multifunctional enzyme, p21-activated kinase 1 (Pak1), a member of a family of proteins in the small G protein-signaling pathway that is activated by Cdc42 and Rac1. There is a large body of evidence from studies in noncardiac tissue that Pak1 activity is key in regulation of a number of cellular functions, including cytoskeletal dynamics, cell motility, growth, and proliferation. Although of significant potential impact, the role of Pak1 in regulation of the heart has been investigated in only a few laboratories. In this review, we discuss the structure of Pak1 and its sites of posttranslational modification and molecular interactions. We assemble an overview of the current data on Pak1 signaling in noncardiac tissues relative to similar signaling pathways in the heart, and we identify potential roles of Pak1 in cardiac regulation. Finally, we discuss the current state of Pak1 research in the heart in regard to regulation of contractility through functional myofilament and Ca(2+)-flux modification. An important aspect of this regulation is the modulation of kinase and phosphatase activity. We have focused on Pak1 regulation of protein phosphatase 2A (PP2A), which is abundant in cardiac muscle, thereby mediating dephosphorylation of sarcomeric proteins and sensitizing the myofilaments to Ca(2+). We present a model for Pak1 signaling that provides a mechanism for specifically affecting cardiac cellular processes in which regulation of protein phosphorylation states by PP2A dephosphorylation predominates.

Regional differences in spontaneous Ca2+ spark activity and regulation in cat atrial myocytes.

Calcium sparks result from the concerted opening of a small number of Ca2+ release channels (ryanodine receptors, RyRs) organized in clusters in the membrane of the sarcoplasmic reticulum (SR). Calcium sparks represent the elementary events of SR Ca2+ release in cardiac myocytes, and their spatial and temporal summation results in whole-cell [Ca2+]i transients observed during excitation-contraction coupling (ECC). Atrial myocytes generally lack transverse tubules; however, during ECC Ca2+ release is initiated from junctional SR (j-SR) in the cell periphery from where activation propagates inwardly through Ca(2+)-induced Ca2+ release (CICR) from non-junctional SR (nj-SR). Despite the structural differences in the microdomains of RyRs of j-SR and nj-SR, spontaneous Ca2+ sparks are observed from both types of SR, albeit at different frequencies. In cells that showed spontaneous Ca2+ sparks from j-SR and nj-SR, subsarcolemmal (SS) Ca2+ sparks from the j-SR were 3-4 times more frequent than central (CTR) Ca2+ sparks occurring from nj-SR. Subsarcolemmal Ca2+ sparks had a slightly higher amplitude, but were essentially identical in their spatial spread and duration when compared to CTR Ca2+ sparks. Sensitization of RyRs with a low concentration (0.1 mM) of caffeine led to a 107% increase in the frequency of CTR Ca2+ sparks, whereas the SS Ca2+ spark frequency increased by only 58%, suggesting that the nj-SR is capable of much higher Ca2+ spark activity than observed normally in unstimulated cells. The L-type Ca2+ channel blocker verapamil reduced SS Ca2+ spark frequency to 38% of control values, whereas Ca2+ spark activity from nj-SR was reduced by only 19%, suggesting that SS Ca2+ sparks are under the control of Ca2+ influx from the extracellular space. Removal of extracellular Ca2+ eliminated SS Ca2+ sparks completely, whereas Ca2+ sparks from the nj-SR continued, albeit at a lower frequency. In membrane-permeabilized (saponin-treated) atrial myocytes, where [Ca2+] can be experimentally controlled throughout the entire myocyte, j-SR and nj-SR Ca2+ spark frequencies were identical, and Ca2+ sparks could be observed spaced at sarcomeric distances throughout the entire cell, suggesting that all release sites of the nj-SR can become active. Measurement of SR Ca2+ load (10 mM caffeine) revealed no difference between j-SR and nj-SR. The data suggest that in atrial myocytes, which lack a t-tubular system, the nj-SR is fully equipped with a three-dimensional array of functional SR Ca2+ release sites; however, in intact cells under resting conditions, peripheral RyR clusters have a higher probability of activation owing to their association with surface membrane Ca2+ channels, leading to higher spontaneous Ca2+ spark activity. In conclusion, Ca2+ sparks originating from both j-SR and nj-SR are rather stereotypical and show little differences in their spatiotemporal properties. In intact cells, however, the higher frequency of spontaneous SS Ca2+ sparks arises from the structural arrangement of sarcolemma and j-SR membrane and thus from the difference in the trigger mechanism.

Regulation of junctional and non-junctional sarcoplasmic reticulum calcium release in excitation-contraction coupling in cat atrial myocytes.

We have characterized the dependence on membrane potential (V(m)) and calcium current (I(Ca)) of calcium-induced calcium release (CICR) from the junctional-SR (j-SR, in the subsarcolemmal (SS) space) and non-junctional-SR (nj-SR, in the central (CT) region of the cell) of cat atrial myocytes using whole-cell voltage-clamp together with spatially resolved laser-scanning confocal microscopy. Subsarcolemmal and central [Ca(2+)](i) transient amplitudes and I(Ca) had a bell-shaped dependence on V(m), but [Ca(2+)](i) reached a maximum at more negative V(m) (-10 to 0 mV) than I(Ca) (+10 mV). Termination of I(Ca) after a brief depolarization (2.5 to 22.5 ms) immediately interrupted only the SS [Ca(2+)](i) transient, leaving the development of the CT [Ca(2+)](i) transient unaffected. Block of SR function with 20 microM ryanodine and 2 microM thapsigargin, revealed that > 90 % of the control [Ca(2+)](i) transient amplitude was attributable to active SR Ca(2+) release through ryanodine receptors (RyRs). The gain of SR Ca(2+) release was highest in the SS space at negative test potentials and was less pronounced in the CT region. Inhibition of Na(+)-Ca(2+) exchange resulted in prolonged and higher amplitude [Ca(2+)](i) transients, elevated resting [Ca(2+)](i), accelerated propagation of CICR, decreased extrusion of Ca(2+) and an increase in j-SR Ca(2+) load. Increasing the cytosolic Ca(2+) buffer capacity by internal perfusion with 1 mM EGTA limited SR Ca(2+) release to the SS region, indicating that Ca(2+) release from nj-SR is initiated by diffusion of Ca(2+) from the cell periphery and propagating CICR. Junctional-SR Ca(2+) release occurred at discrete sites whose order of activation and amplitude of release varied from beat to beat. In conclusion, during normal excitation-contraction coupling in cat atrial myocytes, only Ca(2+) release from the j-SR is directly activated by Ca(2+) entering via I(Ca). Elevation of SS [Ca(2+)](i) is required to provide the cytosolic Ca(2+) gradient needed to initiate regenerative and propagating CICR from nj-SR.

Local calcium gradients during excitation-contraction coupling and alternans in atrial myocytes.

Subcellular Ca(2+) signalling during normal excitation-contraction (E-C) coupling and during Ca(2+) alternans was studied in atrial myocytes using fast confocal microscopy and measurement of Ca(2+) currents (I(Ca)). Ca(2+) alternans, a beat-to-beat alternation in the amplitude of the [Ca(2+)](i) transient, causes electromechanical alternans, which has been implicated in the generation of cardiac fibrillation and sudden cardiac death. Cat atrial myocytes lack transverse tubules and contain sarcoplasmic reticulum (SR) of the junctional (j-SR) and non-junctional (nj-SR) types, both of which have ryanodine-receptor calcium release channels. During E-C coupling, Ca(2+) entering through voltage-gated membrane Ca(2+) channels (I(Ca)) triggers Ca(2+) release at discrete peripheral j-SR release sites. The discrete Ca(2+) spark-like increases of [Ca(2+)](i) then fuse into a peripheral 'ring' of elevated [Ca(2+)](i), followed by propagation (via calcium-induced Ca(2+) release, CICR) to the cell centre, resulting in contraction. Interrupting I(Ca) instantaneously terminates j-SR Ca(2+) release, whereas nj-SR Ca(2+) release continues. Increasing the stimulation frequency or inhibition of glycolysis elicits Ca(2+) alternans. The spatiotemporal [Ca(2+)](i) pattern during alternans shows marked subcellular heterogeneities including longitudinal and transverse gradients of [Ca(2+)](i) and neighbouring subcellular regions alternating out of phase. Moreover, focal inhibition of glycolysis causes spatially restricted Ca(2+) alternans, further emphasising the local character of this phenomenon. When two adjacent regions within a myocyte alternate out of phase, delayed propagating Ca(2+) waves develop at their border. In conclusion, the results demonstrate that (1) during normal E-C coupling the atrial [Ca(2+)](i) transient is the result of the spatiotemporal summation of Ca(2+) release from individual release sites of the peripheral j-SR and the central nj-SR, activated in a centripetal fashion by CICR via I(Ca) and Ca(2+) release from j-SR, respectively, (2) Ca(2+) alternans is caused by subcellular alterations of SR Ca(2+) release mediated, at least in part, by local inhibition of energy metabolism, and (3) the generation of arrhythmogenic Ca(2+) waves resulting from heterogeneities in subcellular Ca(2+) alternans may constitute a novel mechanism for the development of cardiac dysrhythmias.

Palytoxin disrupts cardiac excitation-contraction coupling through interactions with P-type ion pumps.

Palytoxin is a coral toxin that seriously impairs heart function, but its effects on excitation-contraction (E-C) coupling have remained elusive. Therefore, we studied the effects of palytoxin on mechanisms involved in atrial E-C coupling. In field-stimulated cat atrial myocytes, palytoxin caused elevation of diastolic intracellular Ca(2+) concentration ([Ca(2+)](i)), a decrease in [Ca(2+)](i) transient amplitude, Ca(2+) alternans followed by [Ca(2+)](i) waves, and failures of Ca(2+) release. The decrease in [Ca(2+)](i) transient amplitude occurred despite high sarcoplasmic reticulum (SR) Ca(2+) load. In voltage-clamped myocytes, palytoxin induced a current with a linear current-voltage relationship (reversal potential approximately 5 mV) that was blocked by ouabain. Whole cell Ca(2+) current and ryanodine receptor Ca(2+) release channel function remained unaffected by the toxin. However, palytoxin significantly reduced Ca(2+) pumping of isolated SR vesicles. In current-clamped myocytes stimulated at 1 Hz, palytoxin induced a depolarization of the resting membrane potential that was accompanied by delayed afterdepolarizations. No major changes of action potential configuration were observed. The results demonstrate that palytoxin interferes with the function of the sarcolemmal Na(+)-K(+) pump and the SR Ca(2+) pump. The suggested mode of palytoxin toxicity in the atrium involves the conversion of Na(+)-K(+) pumps into nonselective cation channels as a primary event followed by depolarization, Na(+) accumulation, and Ca(2+) overload, which, in turn, causes arrhythmogenic [Ca(2+)](i) waves and delayed afterdepolarizations.