Direct mechanisms of SARS-CoV-2-induced cardiomyocyte damage: an update

Pathology reports from several studies have identified SARS-CoV-2 viral proteins and genetic material in CMs of patients with COVID-19 [12‐15], even in those with no clinical signs of cardiac involvement [16, 17]. In addition, in vitro studies with human induced pluripotent stem cells (hiPS) and isolated adult CMs and in vivo experiments using animal models have confirmed that CMs are permissive for SARS-CoV-2 infection [14, 15, 18]. Bojkova et al. successfully infected hiPS-derived CMs with SARS-CoV-2 and demonstrated the presence of intracellular double-strand viral RNA and viral spike glycoprotein protein expression [14]. Increasing concentrations of viral RNA were detected in the supernatants of infected CMs, and SARS-CoV-2 infection ability was confirmed in 3D cardiosphere tissue. Overall, these data support the hypothesis that COVID-19-associated myocardial injury can result from the direct infection of CMs and the cardiotoxic effect of SARS-CoV-2 and is not merely a secondary effect derived from hypoxia and systemic inflammation.

A single-cell sequencing study has revealed that the expression of viral ACE2 receptors is noticeable in CMs from normal hearts and has suggested that ACE2 expression in the adult human heart is higher than that in the lung [19]. Furthermore, ACE2 expression was significantly elevated in CMs of patients with heart disease (e.g., dilated cardiomyopathy, hypertrophic cardiomyopathy, non-COVID-19 myocarditis, aortic stenosis, and heart failure) compared to healthy controls [20‐22]. These findings are consistent with the clinical observations and epidemiologic reports suggesting that patients with underlying cardiovascular conditions are more vulnerable and prone to myocardial injury [11, 23].

Type II transmembrane serine protease (TMPRSS2) is the protein involved in the cleavage of many SARS-CoV-2 variants. TMPRSS2 cleaves the S protein into the S1 and S2 subunits, exposing the receptor-binding domain (RBD) of the S1 subunit to facilitate recognition and binding to the ACE2 receptor. Notably, SARS-CoV-2 entry into CMs is mediated by an endosomal-dependent mechanism that does not require TMPRSS2, unlike lung-derived cell lines where TMPRSS2 mediates viral entry [13, 14, 19]. Contrary to TMPRSS2, multiple endosomal proteases, including cathepsins and calpains, are highly expressed at a level similar to that in the lungs in human pluripotent stem cell-derived CMs (hPSC-CMs) and CMs from human autopsy specimens, [14, 19]. Bailey et al. inhibited TMPRSS2 and found no effect on hPSC-CMs infection, whereas an endosomal cysteine protease inhibitor abolished SARS-CoV-2 infection [13]. These data suggest that the endosomal-dependent proteases compensate for S protein priming to mediate SARS-CoV-2 infection in CMs. However, Omicron, a recent variant of SARS-CoV-2, only requires ACE2 to invade the cell without the help of TMPRSS2. This variant is encapsulated in endosome bubble, which drifts into the cells and breaks out [24]. Therefore, cells without or with less TMPRSS2 expression are easily infected by Omicron, resulting in relevant damage, and theoretically, CMs are more vulnerable to virus invasion leading to myocardial damage. In conclusion, the specific effects of Omicron on cardiac injury need to be further explored.

In addition to direct attacks, SARS-CoV-2 may also infect CMs indirectly. Kwon et al. isolated extracellular vesicles from lung epithelial cells that overexpressed SARS-CoV-2 genes and found that hiPSC-CMs were receptive to these viral RNA-harboring extracellular vesicles [25]. Viral genes were detected in hiPSC-CMs, along with inflammatory activation.

Recent reports have identified SARS-CoV-2 particles or components within CMs from autopsy and endomyocardial biopsy samples utilizing RNA sequencing, RNAscope, immunohistochemistry, and transmission electron microscopy analysis Direct viral infection leads to adverse effects on CMs in several aspects, such as cell morphology, electrophysiology, subcellular structures, and cell death [15, 16, 26‐29].

Direct injury to the heart tissue by SARS-CoV-2 could be an underlying cause of heart disease in COVID-19 [30]. Myocardial contractile dysfunction is one of the prominent manifestations of COVID-19-related cardiac complications. The significant reduction in cardiac contractility and subsequent myocardial dysfunction induced by SARS-CoV-2 has received widespread attention. The European Society of Cardiology has published a consensus to elucidate the current knowledge of cardiovascular damage induced by SARS-CoV-2 for alteration in the clinic [31].

The pathogenic mechanisms of SARS-CoV-2 are commonly considered to be multifaceted. One of these mechanisms is based on data indicating that virus infection may directly destroy sarcomeres in CMs. Sarcomeric disruptions and myofibril loss in CMs have been observed occasionally in vivo and in vitro, and both presented similar signatures [14, 16, 32]. Pérez-Bermejo et al. observed that the number of myofibrillar fragments produced by cleaving sarcomeres into individual sarcomeric units increased gradually in the cytoplasm of hiPSC-CMs cytoplasm as the virus exposure time increased, suggesting that the invasion of SARS-CoV-2 exacerbated this non-specific cytopathy. Furthermore, high-content imaging analysis of infected hiPSCs-CMs showed that these fragments and intact sarcomeres possessed identical structures, except for the absence of a dark A-band in the myofibrillar fragments, suggesting the cleavage of thick filaments. Notably, they observed the LKGG↓K sequence in myosin heavy chain protein family members. This sequence is similar to those in viral polyproteins that can be cleaved by the SARS-CoV-2 papain-like protease. This result indicates that the virus-derived proteases, indicating besides processing their target sequences, may possess the capacity to cleave several intracellular proteins and alter the cellular structural integrity [29].

Another pathogenic mechanism of SARS-CoV-2 is related to the fact that the virus damages contractility by disrupting gene expression profiles, particularly of those genes directly involved in sarcomere function. Significant downregulation of genes encoding sarcomeric structural proteins, myosin light chains, and linker of nucleoskeleton and cytoskeleton complex (a subset of proteins that are important for anchoring the nucleus to the actin cytoskeleton) in infected hiPSCs-CMs. Furthermore, study revealed alterations in the expression of genes related to energy production leading to a shift in the energy production mechanism from mitochondrial oxidative phosphorylation to glycolytic metabolism in hiPSC-CMs and hPSC-CMs [18]. The resulting energy reduction can adversely affect contraction at the cellular and tissue levels. Although oxidative stress and inflammation may directly alter gene expression profiles in infected CMs, and other effectors as the non-structure protein 1 (Nsp1) may also play a role. Nsp1 binds to the 40S ribosomal subunit in a manner that its C-terminal domain directly blocks the entry channel of messenger RNA in the ribosome, perturbing the translation activities of host cells and shutting down the translation of host messenger RNA [33]. Collectively, these factors may result in the breakdown of the contractile machinery and the appearance of impaired myocardial contractility.

To evaluate the influence of SARS-CoV-2 infection on cardiac tissue in vivo, Siddiq et al. measured the damage severity of hiPSC-CMs treated with interleukins (ILs) or vehicle in the presence or absence of SARS-CoV-2 infection. ILs treatment alone did not increase the release of troponin-I (a marker for damaged myofibrils) but augmented the cytopathic effect of infection in hiPSC-CMs [34]. These results further emphasize the prominent role of direct SARS-CoV-2 infection in CMs damage and indicate the facilitative role of ILs in vivo. Moreover, researchers have utilized three-dimensional engineered heart tissues (3D-EHTs) to further evaluate the contractile properties of hPSC-CMs at the tissue level and observed that 3D-EHTs infected with SARS-CoV-2 presented weakening contractile forces compared to mock controls [13, 18]. These results shed light on the correlations between the contractile dysfunction at the cellular level induced by direct infection and the prevalent myocardial injury in patients with COVID-19 at the organ-tissue level. However, further investigations are warranted to elucidate the underlying mechanisms thoroughly.

Patients with COVID-19 are more prone to suffering heart damage or irregular heartbeats, indicating the presence of CMs dysfunction induced by SARS-CoV-2 [35]. In addition to mechanical dysfunction, SARS-CoV-2 infected CMs also have impaired electrical functions. Marchiano et al. discovered that SARS-CoV-2 infection resulted in the abnormal generation and propagation of electrical signals in hPSC-CMs, hiPSC-CMs, and human embryonic stem cell-derived CMs (hESC-CMs), which show reduced beating rate, lower depolarization spike amplitude, and decreased electrical conduction velocity. They also observed a time-dependent increase in the field potential duration (FPD) in H7 hESC-CMs in both spontaneously beating and electrically-paced cultures [18]. Anomalous electrical signals manifest even in the absence of extensive cell death, suggesting that there may be some virus-derived components or subsequently generated substrates that disrupt cellular homeostasis, thus causing electrical dysfunction at the cellular and even at the tissue-organ level. Interestingly, a recent study proposed that SARS-CoV-2 infection-induced CM structural alterations lead to electrophysiological abnormalities. Some researchers have identified syncytia formation in SARS-CoV-2 infected cells and tissue, which is relatively rare but not absent compared to other cytopathies [36, 37]. Coronaviruses commonly induce cell–cell fusion because of the fusogenic property of the S protein and its capability to trigger virus-cell membrane fusion [38]. Schneider et al. identified intracellular junctions between CMs formed by highly-concentrated sarcolemmal t-tubule viral S protein in myocardial specimens from a young woman who died of sudden cardiac death and was found to be COVID-19-positive during the postmortem. The S protein likely tends to promote the formation of junctions between adjacent CMs instead of syncytia due to cytoskeletal constraints and viral infectivity [37, 38]. These cell-to-cell conduits are prone to short-circuit electrically excitable myocardium, and give rise to electrophysiological abnormalities. Moreover, marked pathological Ca²⁺ flux, sparks, and tsunami-like waves have been observed in hiPSC-CMs built multinucleated cardiomyotubes, which present significantly prolonged action potential duration, elevated membrane capacitance, delayed afterdepolarizations, and erratic beating frequencies frequency compared to mock controls. Altogether, these studies provide a novel perspective on the pathogenic mechanism of COVID-19-related arrhythmias and reiterate the detrimental effects of direct SARS-CoV-2 infection in CMs.

Another widely recognized alteration induced by SARS-CoV-2 infection is the loss of ACE2 and the subsequent events in CMs. ACE2 is a homolog of ACE and negatively regulates RAAS by degrading AngII to Ang (1–7), whereas ACE metabolizes AngI to AngII. Reduction of ACE2 on the CM membrane is induced by suppressed gene expression, enhanced endocytosis, and cleavage by A disintegrin and metalloproteinase-17 (ADAM-17), a transmembrane protease evoked by the sudden-increased in ILs after infection. Decreased ACE2 undermines its counterbalance effect on the AngII/Ang II type-1 receptor pathway and results in the accumulation of AngII and attenuation of the Ang (1–7)/Mas receptor pathway. Excessive AngII induces overt CM autophagy and apoptosis, which is speculated to play a critical role in developing decompensated hypertensive heart disease [39, 40]. Contrary to AngII, Ang (1–7) and Mas are considered cardioprotective factors. Liu et al. observed that Ang (1–7) prevents chronic atrial ionic remodeling by suppressing the decrease of transient outward potassium current (Ito) and L-type Ca²⁺ current (ICaL), as well as the reduction of Kv4.3 subunit gene expression [41]. In addition, Ang (1–7) is reportedly effective in counteracting the increase in isoproterenol-induced beating rate in CMs, reflecting its potential to attenuate cardiac reactivity in response to acute emotional stress [42]. Dias-Peixoto et al. discovered that CMs from Mas-deficient mice are characterized by reduced peak and slower [Ca²⁺] transients, suggesting that depression of the Mas receptor pathway might disturb the Ca²⁺-handling capacity of CMs [43]. Thus, during SARS-CoV-2 infection, the diminished cardioprotective effects and hyperactive harmful processes can synergize and lead to significant CMs damage in various ways. The different mechanisms of direct myocardial injury induced by SARS-CoV-2 are shown in Fig. 1.