key: cord-0761482-4mrxdp6z
authors: Marchiano, Silvia; Hsiang, Tien-Ying; Khanna, Akshita; Higashi, Ty; Whitmore, Leanne S.; Bargehr, Johannes; Davaapil, Hongorzul; Chang, Jean; Smith, Elise; Ong, Lay Ping; Colzani, Maria; Reinecke, Hans; Yang, Xiulan; Pabon, Lil; Sinha, Sanjay; Najafian, Behzad; Sniadecki, Nathan J.; Bertero, Alessandro; Gale, Michael; Murry, Charles E.
title: SARS-CoV-2 infects human pluripotent stem cell-derived cardiomyocytes, impairing electrical and mechanical function
date: 2021-02-13
journal: Stem Cell Reports
DOI: 10.1016/j.stemcr.2021.02.008
sha: c73a1706b7aba0828995c610a469f56104a69b1a
doc_id: 761482
cord_uid: 4mrxdp6z

COVID-19 patients often develop severe cardiovascular complications, but it remains unclear if these are caused directly by viral infection or are secondary to a systemic response. Here we examine the cardiac tropism of SARS-CoV-2 in human pluripotent stem cell-derived cardiomyocytes (hPSC-CMs) and smooth muscle cells (hPSC-SMCs). We find that that SARS-CoV-2 selectively infects hPSC-CMs through the viral receptor ACE2, whereas in hPSC-SMCs there is minimal viral entry or replication. After entry into cardiomyocytes, SARS-CoV-2 is assembled in lysosome-like vesicles and egresses via bulk exocytosis. The viral transcripts become a large fraction of cellular mRNA while host gene expression shifts from oxidative to glycolytic metabolism and up-regulates chromatin modification and RNA splicing pathways. Most importantly, viral infection of hPSC-CMs progressively impairs both their electrophysiological and contractile function, and causes widespread cell death. These data support the hypothesis that COVID-19-related cardiac symptoms can result from a direct cardiotoxic effect of SARS-CoV-2.

With over 100 million people affected worldwide, the outbreak of Severe Acute Respiratory Syndrome-Coronavirus 2 (SARS-CoV-2) has already left its permanent mark on human history (Hopkins, 2020; Zhu et al., 2020) . SARS-CoV-2 belongs to the family of coronaviridae, a large group of single-stranded enveloped RNA viruses reported for the first time in humans in the 1960s (Andersen et al., 2020; Corman et al., 2018; Cui et al., 2019) . Besides being long recognized as one of the common cold viruses, coronaviruses took center stage in infectious disease medicine following the outbreaks of SARS-CoV in 2003 and of Middle East Respiratory Syndrome-coronavirus (MERS-CoV) a decade later. Coronaviruses became thus recognized as highly pathogenic for humans, with a symptomatology that focuses on the respiratory system while often also involving extra-respiratory organs (Alhogbani, 2016; Nishiga et al., 2020; Oudit et al., 2009; Zhou et al., 2020) . Indeed, even though the lungs represent the main target, cardiovascular complications (including worsening of pre-existing conditions and onset of new disorders), were not only reported for SARS-CoV and MERS-CoV, but are also significantly contributing to the mortality of COVID-19 patients during the ongoing pandemic (Alhogbani, 2016; Oudit et al., 2009; Shi et al., 2020; Zhou et al., 2020) .

The most common cardiovascular complications observed after SARS-CoV-2 infection are myocardial injury (including cases with and without classic coronary occlusion), arrhythmias, and heart failure (Baggiano et al., 2020; Nishiga et al., 2020; Ojha et al., 2020; Ruan et al., 2020; Shi et al., 2020; Wang et al., 2020; Xu et al., 2020; Zhou et al., 2020) . In particular, myocardial injury, characterized by elevated serum levels of cardiac troponin I and/or electrocardiogram (ECG) abnormalities, has been independently associated with increased mortality in COVID-19 patients . Moreover, as reported also for SARS-CoV (Madjid et al., 2007) , SARS-CoV-2 can trigger acute coronary syndrome even in the absence of systemic inflammation (Nishiga et al., 2020; Wang et al., 2020) . Retrospective studies show that J o u r n a l P r e -p r o o f hospitalized COVID-19 patients develop cardiac arrhythmias, including ventricular tachycardia and atrial fibrillation (Bhatla et al., 2020; Malaty et al., 2020; Wang et al., 2020; Zylla et al., 2021) . Progressive left ventricular dysfunction and overall symptoms that resemble heart failure have also been observed in a significant number of patients Huang et al., 2020a; Wang et al., 2020; Zhou et al., 2020) . At the beginning of the outbreak this symptomatology was reported mostly in critically ill COVID-19 patients . A sizable number of more recent studies has reported that cardiac symptoms are observed also in mild and even asymptomatic cases of COVID-19 (Arentz et al., 2020; Huang et al., 2020b; Inciardi et al., 2020; Puntmann et al., 2020; Rajpal et al., 2021) .

The mechanisms behind cardiac disease reported for COVID-19 are still unclear (Nishiga et al., 2020; Zhou et al., 2020) . Upon lung infection, the uncontrolled release of inflammatory cytokines, termed "cytokine storm", could induce multi-organ damage, ultimately leading to organ failure and worsening of pre-existing cardiovascular disorders (Driggin et al., 2020; Inciardi et al., 2020; Tay et al., 2020) . Moreover, COVID-19 is associated with coagulopathies, which also can induce ischemic heart damage (Driggin et al., 2020; Nishiga et al., 2020; Ranucci et al., 2020) . Lastly, SARS-CoV-2 could directly mediate heart injury by entering cardiomyocytes or other cardiac stromal and/or vascular cells via binding of the viral spike glycoprotein to its extracellular receptor, angiotensin I converting enzyme 2 (ACE2) (Baggiano et al., 2020; Hoffmann et al., 2020) . This protein is expressed in different tissues of the human body, including cardiomyocytes and cardiac pericytes, and its primary function is to counterbalance the renin-angiotensin-aldosterone system (Chen et al., 2020a; Hikmet et al., 2020; Li et al., 2020; Verdecchia et al., 2020) .

Several studies detected SARS-CoV-2 genome in the heart and/or reported signs of viral myocarditis in COVID-19 infected individuals, including asymptomatic cases (Bradley et al., 2020; Dolhnikoff et al., 2020; Lindner et al., 2020; Rajpal et al., 2021) . Moreover, in vivo and in J o u r n a l P r e -p r o o f vitro studies utilizing both human adult cardiomyocytes and human pluripotent stem cell-derived cardiomyocytes (hPSC-CMs) have shown that SARS-CoV-2 can infect cardiomyocytes, indicating that SARS-CoV-2 could exhibit cardiac tropism (Bojkova et al., 2020; Chen et al., 2020b; Sharma et al., 2020; . However, whether SARS-CoV-2 infection of human cardiomyocytes leads to a direct impairment of cardiac function is still unresolved.

Furthermore, whether other cardiac cell types are also susceptible to SARS-CoV-2 remains unclear.

In this study we examine the mechanisms behind COVID-19-related cardiac symptoms using hPSC-CMs and hPSC-derived smooth muscle cells (hPSC-SMCs), established models for cardiovascular disease research (Bertero et al., 2019b; Cheung et al., 2012; Serrano et al., 2019; Yang et al., 2018) . SARS-CoV-2 specifically infects and propagates within hPSC-CMs, a process that appears to exquisitely rely on ACE2 and to both involve direct membrane fusion and entry through the endo-lysosomal pathway. Pathways involved in RNA splicing and chromatin accessibility are significantly upregulated after infection, whereas pathways involved in oxidative metabolism and mitochondrial function are downregulated. SARS-CoV-2 infection results in disruption of the contractile cytoskeleton, electrical and mechanical dysfunction, and eventual cell death. These findings provide evidence for a direct viral cytopathic pathway involving cardiac myocytes in the context of COVID-19-related cardiac disease.

infection is thought to depend on expression of both the viral receptor ACE2 and various host proteases Millet and Whittaker, 2015; Shang et al., 2020; Zumla et al., 2016) . We found that ACE2 is transcriptionally activated during cardiac differentiation of both RUES2 embryonic stem cell-derived cardiomyocytes (hESC-CMs; Fig. 1A ) and WTC11c induced pluripotent stem cell-derived cardiomyocytes (hiPSC-CMs; Supplementary Fig. 1A ).

Single-cell RNA-sequencing analysis detected ACE2 mRNA in ~9% of hESC-CMs, indicating low and/or transitory expression (Fig. 1B) . A larger fraction of cells expressed moderate to high levels of endosomal cysteine proteases CTSB (cathepsin B; ~71.0%) and CTSL (cathepsin L; ~46.0%). Detection of these factors is relevant because they can cleave the spike glycoprotein leading to endomembrane fusion-mediated release of SARS-CoV-2 genome inside the cytoplasm (Kang et al., 2020; Millet and Whittaker, 2015; Ou et al., 2020; Yang and Shen, 2020) . Importantly, these viral processing factors were often co-expressed with ACE2

( Supplementary Fig. 1B) . Although viral entry can also be mediated by TMPRSS2 Shulla et al., 2011) , this transmembrane serine protease was not detectable in hESC-CMs ( Supplementary Fig. 1C ), as also reported for the adult human heart (Litvinukova et al., 2020) . Interestingly, the lipid phosphatase, PIKFYVE, another endosomal viral processing factor, and FURIN, a membrane-bound serine protease, were also broadly expressed in hESC-CMs ( Supplementary Fig. 1C ), overall suggesting that the mechanism of SARS-CoV-2 entry in cardiomyocytes might be different from the TMPRSS2-dependent one reported for lung epithelial cells Schneider, 2020; Shang et al., 2020; Xia et al., 2020b) .

Despite the relatively low levels of mRNA, ACE2 protein was clearly detectable by western blot in hPSC-CMs derived from multiple lines (RUES2 female hESCs, H7 female hESCs, and WTC11c male hiPSCs), reaching levels comparable to those of VERO cells, a primate kidney epithelial line with established SARS-CoV-2 tropism (Figs. 1C and 1D). Emphasizing the specific tropism of SARS-CoV-2 for hPSC-CMs, ACE2 was expressed at very low levels in hESC-SMCs of varying embryonic origin (lateral mesoderm-or neural crest-derived, all differentiated from H9 female hESCs; Fig. 1C and Supplementary Fig. 1D ). Collectively, hPSC-CMs express proteins that may render them susceptible to SARS-CoV-2 infection (Bojkova et al., 2020; Sharma et al., 2020; . To investigate whether hPSC-CMs are permissive to SARS-CoV-2 replication, we quantified extracellular viral particles and intracellular viral RNA (by plaque assay and RT-qPCR, respectively). The one-step growth curve after 5 MOI infection indicated that viral replication occurred steadily from 24-72 HPI, followed by a precipitous decline as the cells died (Fig. 2C ).

The multi-step growth curve (0.1 MOI infection) confirmed that SARS-CoV-2 replicated inside hPSC-CMs (Fig. 2D) , with a marked increase in viral particles and RNA at 48 and 72 HPI (at which point the experiment was stopped). In agreement with morphological observations, H7derived cardiomyocytes were more permissive to SARS-CoV-2 replication than WTC11c- 

Distinctly from hPSC-CMs, hESC-SMCs exposed to SARS-CoV-2 did not show any cytopathic effects even at the highest MOI at 72 HPI ( Supplementary Fig. 2F ). Accordingly, extracellular viral particles and intracellular viral RNA in hPSC-SMCs were more than two orders of magnitude lower than the ones observed for hPSC-CMs ( Supplementary Fig. 2G ). These findings highlight the specific tropism of SARS-CoV-2 for hPSC-CMs, and exclude that cytopathic effects observed in cardiomyocytes may be attributable to toxic contaminants in the viral preparation.

To investigate the role of ACE2 during cardiomyocyte infection by SARS-CoV-2, we generated . We also observed dilated membrane-bound tubular structures, likely representing the endoplasmic reticulum Golgi intermediate compartment (ERGIC), in close proximity to membrane-enclosed viral particles ( Fig. 3D ). In addition, we observed a variety of vesicles containing viruses (Figs. 3D-F). Most notable were large vesicles with electron dense content packed with mature virus particles (Fig.   3E ). In some of these vesicles, we also observed lipid droplets and multilamellar bodies, consistent with lysosomes, whereas others were smooth walled vesicles (Figs. 3E, F) (Ghosh et al., 2020; Snijder et al., 2020) . Finally, exocytosis of virions was readily identifiable on the cell surface (Figs. 3G-I). In summary, these electron microscopic studies demonstrated viral entry via both direct fusion and endocytosis, replication in subcellular membrane structures, and "hijacking" of lysosomal vesicles for the bulk release of mature virions by exocytosis. (Banerjee et al., 2020) . In addition, we observed that genes involved in mitochondrial function and energy production were downregulated (Fig. 4F, Supplementary Fig. 4B and Supplementary Tables 2 and 4), indicating that SARS-CoV-2 might promote a shift toward a glycolytic metabolism by suppressing mitochondrial oxidative phosphorylation, which could also favor its replication (Ajaz et al., 2021; Icard et al., 2021) .

Upon viral infection, pathogen-associated molecular patterns (PAMPs) initiate the early immune response via host pattern recognition receptors (PRRs). After virus uncoating, RIG-I-like receptors (RLRs) bind to the uncapped and double stranded viral RNA in the cytosol and trigger innate immune activation, leading to the production of type I and type III interferons and the interferon-induced antiviral response (Loo and Gale, 2011) . Upregulated pathways after SARS-CoV-2 infection in hPSC-CMs included those involved in viral defense (Fig. 4E) . To more finely clarify the underlying kinetics, we analyzed the interferon response from 2 to 72 HPI by RT-qPCR. We found that interferon transcripts (INFB1 and IFNL1) were markedly upregulated at 48 and 72 HPI in both cell lines, with a stronger effect in the more sensitive H7 cardiomyocytes ( Fig. 4G and Supplementary Fig. 4C ). The interferon-stimulated genes IFIT1 and IFITM1 were also upregulated at the latest time point. These results indicate that SARS-CoV-2 induces innate immune activation and interferon response in hPSC-CMs, similar to other cell types . Fig. 5A ). This outcome may reflect a decreased efficiency of viral propagation as hPSC-CMs were plated at high density to ensure robust assessment of electrophysiological properties. Nevertheless, viral RNA and viral particles could still be detected at 0.1 MOI, with the highest levels at 5 MOI ( Fig. 5B and Supplementary Fig.5B) , showing that the infection occurred also in these highly dense cultures. Representative propagation maps are shown in Figure 5C and Supplementary Figure 5C , while representative field potential recordings showcasing changes in spike amplitude and frequency are included in Figure 5D and Supplementary Figure   5D . Remarkably, SARS-CoV-2 infection rapidly resulted in reduced beating rate, lower depolarization spike amplitude, and decreased electrical conduction velocity (Figs. 5E and Supplementary Figs. 5E, F) . In H7 hESC-CMs we also observed a time-dependent increase in the field potential duration (FPD) both in spontaneously beating and electrically paced cultures ( Fig. 5F ; similar measurements could not be reliably obtained from WTC11c hiPSC-CMs due to the limited amplitude of the repolarization wave after SARS-CoV-2 infection). Overall, abnormalities in the generation and propagation of electrical signals were significant even in the absence of extensive cell death, suggesting that SARS-CoV-2 infection in cardiomyocytes could directly create a substrate for arrhythmias (Bhatla et al., 2020; Zylla et al., 2021) .

We then evaluated the contractile properties of hPSC-CMs using three-dimensional engineered heart tissues (3D-EHTs), following their contractile behavior through magnetic field sensing (Bielawski et al., 2016) (Figs. 6A-B) . For these experiments we focused on WTC11c hiPSCs since 3D-EHTs from H7 hESC-CMs proved to beat spontaneously at too high a frequency (> 2 Hz) to enable accurate measurements of contractile behavior (i.e. the tissue had a tetanic-like contraction with minimal relaxation between beats at this frequency). We infected 3D-EHTs from WTC11c hiPSC-CMs with 10 MOI (to facilitate infection within the non-vascularized, cell-dense J o u r n a l P r e -p r o o f tissue), and analyzed their contraction for a week. This 3D-model experienced viral replication comparable to that of 2D cultures, highlighting once again the cardiac tropism of SARS-CoV-2 ( Fig. 6C) . The maximal twitch force in infected tissues decreased as early as 72 HPI ( Supplementary Fig. 6A) , and the contractions continued to subside to less than 25% of the force measured at the baseline at 144 HPI (Figs. 6D, E and Supplementary Videos 1 and 2) .

Cardiomyocyte density progressively decreased while cells also became more rounded (i.e. dedifferentiated) and less aligned with the longitudinal axis of the 3D-EHTs (Fig. 6F and Supplementary Fig. 6B ). This could collectively contribute to the loss of force production.

Infected 3D-EHTs also showed decreased expression of the sarcomeric genes MYL2 and MYH6, which may be correlated to the loss of sarcomere organization (Fig. 6G) . Overall, the significant impairment in the contractile properties of 3D-EHTs demonstrates that the mechanical function of cardiomyocytes is impacted by SARS-CoV-2 infection in vitro, and suggest that similar mechanisms could contribute to whole-organ cardiac dysfunction also in patients .

A rapidly increasing number of reports acknowledge cardiovascular involvement as a prevalent complication observed in COVID-19 patients, but discriminating between direct versus indirect effects is still an open challenge (Nishiga et al., 2020; Shi et al., 2020; Tay et al., 2020) . In this study, we show that SARS-CoV-2 has the ability to directly infect cardiomyocytes, to impair both their electrophysiological and contractile properties, and to eventually induce cell death.

In agreement with earlier reports, we find that cardiomyocytes (but not smooth muscle cells) express ACE2, making them susceptible to SARS-CoV-2 infection. Our experiments in ACE2 knockout hPSC-CMs formally demonstrate the key role of this factor for SARS-CoV-2 entry in this cell type. Interestingly, mRNA levels of ACE2 are heterogeneous within hPSC-CMs from the same culture, and the resulting protein is differentially abundant in hPSC-CMs from different genetic backgrounds. The cytopathic effects of SARS-CoV-2 infection also strongly vary between cardiomyocytes derived from different hPSC lines. ACE2 expression and SARS-CoV-2 susceptibility may be similarly heterogeneous in vivo, both across different regions of the heart and within different subjects (Litvinukova et al., 2020) . This might partially explain the discrepancy between the strong prevalence of heart damage in COVID-19 patients and the limited evidence for viral particles in the heart found by autopsy examinations. We suggest that future analyses should aim to sample various regions of the heart and focus on those patients that had shown the strongest cardiac symptoms.

A puzzling observation is that the cytopathic effects of SARS-CoV-2 infection expands to virtually the entire monolayer of hPSC-CMs, even though single cell RNA-seq indicates that many cardiomyocytes do not express detectable levels of ACE2. One possible explanation is that ACE2 transcription is episodic and/or still meaningful low levels that are below the sensitivity of sc-RNA-seq (either way resulting in sufficient protein levels to allow SARS-CoV-2 entry). Alternatively, cytotoxic stimuli triggered by SARS-CoV-2 might spread to the adjacent J o u r n a l P r e -p r o o f cells via gap junctions or through the supernatant as toxic cytokines and/or other danger signals. Finally, the fusogenic properties of the SARS-CoV-2 spike protein may mediate membrane fusion not only between viral and host membranes but also within cardiomyocytes, leading to inter-cellular viral spreading (Schneider, 2020) . Spike protein fusogenicity is secondary to its proteolytic cleavage by host proteases like furin (Millet and Whittaker, 2015) , which is expressed in hPSC-CMs. Noticeably, we observed direct fusion between the virus and hPSC-CMs, a process that may be mediated by proteolytically-primed spike proteins, as in the case of MERS-CoV (Millet and Whittaker, 2014; Xia et al., 2020a; Xia et al., 2020b) . Future studies are needed to clarify whether any of these non-mutually exclusive mechanisms are involved in the strong susceptibility of hPSC-CMs to SARS-CoV-2 infection, and whether similar events are recapitulated in vivo.

We found that hPSC-CMs are also extremely permissive to viral replication. Among other types of ultrastructural features, we detected the presence of double-membrane vesicles. These are used by viruses both to concentrate their building materials for efficient replication, and to evade the immune surveillance by "hiding" viral factors that can trigger the PAMP pathway (Wolff et al., 2020) . Accordingly, we observed activation of interferon-responsive genes only at late time points of SARS-CoV-2 infection. The interferon response, which is part of the innate immune response activation (Loo and Gale, 2011) , usually occurs within hours from viral infection. The fact that cardiomyocytes infected with SARS-CoV-2 show a delayed response may facilitate viral replication to high levels . Furthermore, RNA-seq showed that SARS-CoV-2 infection affects pathways involved in RNA regulation. SARS-CoV-2 can impair RNA splicing to evade the intracellular innate immune response (Banerjee et al., 2020) , which may be the case also in hPSC-CMs. Moreover, oxidative phosphorylation and mitochondrial function are severely downregulated in the infected cells. SARS-CoV-2 may shift cellular metabolism to J o u r n a l P r e -p r o o f promote glycolytic metabolic activity in support of viral replication (Icard et al., 2021) , providing yet another way to boost its replication also in hPSC-CMs.

The presence of highly replicating virus severely affects both the morphology and the function of hPSC-CMs. Infected cardiomyocytes lose cytoskeletal organization, become packed with different types of vesicles, and show broad alterations of gene expression. Using a multielectrode array system, which has been validated to detect potential arrhythmogenic properties of novel drugs (Blinova et al., 2018) , we identified several electrophysiological abnormalities induced by SARS-CoV-2 infection. Prolongation of FPD is particularly noticeable.

This measurement reflects the interval between membrane depolarization and repolarization, and as such represents an in vitro surrogate of the QT interval measured by an electrocardiogram. It is well known that prolongation of the QT interval is pro-arrhythmogenic (Chiang and Roden, 2000) . Thus, FPD prolongation in SARS-CoV-2 infected hPSC-CMs may be an in vitro surrogate phenotype mirroring the arrhythmias observed in ~20% of COVID-19 patients (Malaty et al., 2020) . Last but not least, we found marked impairment in both contractile function and histological organization in 3D-EHTs infected with SARS-CoV-2. If similar effects were to occur in the hearts of some COVID-19 patients, this could contribute to cardiac dysfunction. Overall, hPSC-CMs on MEAs and/or organized in 3D-EHTs may represent valuable scalable platforms to identify active compounds that may provide therapeutic value.

Collectively, our results support the notion that, independent of inflammation or coagulopathy, SARS-CoV-2 can cause direct functional heart damage by either inducing cell death and/or by impairing electro-mechanical functions. One limitation of this study is our reliance on hPSC-CMs, which are well-known for their functional immaturity (Guo and Pu, 2020; Karbassi et al., 2020; Marchiano et al., 2019) . While the in vitro systems we used have been successfully leveraged to model electrophysiological and contractile alterations due to drugs or inherited mutations (Blinova et al., 2018; Paik et al., 2020) , their application to modeling COVID-19 still J o u r n a l P r e -p r o o f requires further validation. Nevertheless, a recent report by Dolhnikoff et al. identified coronaviral particles in the cytoplasm of cardiomyocytes, endothelial cells, and fibroblasts by electron microscopy in the heart of an 11 year-old child who died from multi-system inflammatory syndrome in children (MIS-C) following COVID-19 infection (Dolhnikoff et al., 2020) . This indicates that in vivo cardiomyocytes with substantially greater maturity than used here are susceptible to SARS-CoV-2 infection. COVID-19 patients are commonly treated with steroids to control systemic inflammation. However, our data suggests that treatments aimed to control the direct damage of SARS-CoV-2, not only by preventing infection but also by preventing viral replication or rescuing cardiac function, should also be taken into consideration to prevent long-term cardiovascular complications.

Experimental procedures Cell culture. RUES2 hESCs and WTC11c hiPSCs were maintained and differentiated using small molecules modulators of the WNT pathway (Bertero et al., 2019a) . H7 hESCs were differentiated in suspension culture format by collaborators at the Center for Applied Technology Development at the City of Hope in California. H9 hESCs were maintained and differentiated into LM-SMCs and NC-SMCs as previously described (Bargehr et al., 2016; Serrano et al., 2019) . ACE2 KO clones were generated using CRISPR/Cas9 ribonucleprotein complexes (Synthego).

Bulk RNA-seq datasets from RUES2 hESC-CMs had been previously generated and analyzed (Bertero et al., 2019a) . Bulk mRNA-seq data from infected cardiomyocytes were generated by constructing mRNA-seq libraries using the KAPA mRNA HyperPrep Kit (Kapa Biosystems). Libraries were sequenced on an Illumina NovaSeq. For single cell-RNA-seq analysis, a single cell suspension was generated from RUES2 hESC-CMs and single cell RNA-seq was performed using the Chromium NextGEM Single Cell 3' kit (10X Genomics). For quantitative real-time reverse transcription PCR (RT-qPCR), RNA from infected cardiomyocytes was harvested using Trizol reagent. cDNA was obtained with M-MLV reverse transcriptase (Invitrogen), and RT-qPCR was performed with SYBR Select Master Mix (Applied Biosystems). Primers are reported in Supplementary Table 5 .

Western blot. hPSC-CMs were lysed using RIPA Buffer. Samples were run on mini-PROTEAN TGX precast gels (Bio-Rad) and then transferred on PVDF membranes. Primary and secondary antibodies (Supplementary Experimental Procedures) were incubated in blocking buffer, and fluorescent signals were acquired using with a GelDoc Imager (Bio-Rad).

Level 3 (BSL-3) facility at the University of Washington in compliance with the BSL-3 laboratory J o u r n a l P r e -p r o o f safety protocols (CDC BMBL 5 th ed.) and the recent CDC guidelines for handling SARS-CoV-2.

were obtained from BEI Resources (NR-52281) and the University of Texas , respectively, and propagated in VERO cells (USAMRIID). Viral preparations and culture supernatant from SARS-CoV-2-infected cardiomyocytes were titered using a plaque assay. Viral infection. SARS-CoV-2 wild-type or expressing mNeonGreen protein was diluted to the desired MOI in DMEM and incubated on hPSC-CMs or hPSC-SMCs for 1 h at 37 °C (noninfected [mock] controls were incubated with DMEM only). Cells were then washed with DPBS and cultured in the appropriate maintenance media.

Immunofluorescence. hPSC-CMs were fixed with 4% paraformaldehyde (PFA) and permeabilized using 0.25% Triton X-100 (Sigma-Aldrich) before staining with primary and secondary antibodies (Supplementary Experimental Procedures).

Electron Microscopy. hPSC-CMs were fixed with Karnovsky's fixative. Heavy metal impregnation was performed as detailed elsewhere (Deerinck, 2010) . Thin sections were viewed using a JEOL 1230 transmission electron microscope.

Electrophysiological analysis with MEA. CytoView MEA plates (Axion BioSystems) were coated with Matrigel and hPSC-CMs were plated as previously described (Bertero et al., 2019b) . SARS-CoV-2 effects on hPSC-CMs electrophysiology were recorded using Maestro Pro system and analyzed with Cardiac Analysis Software v3.1.8 (all from Axion BioSystems).

Post-suspended, fibrin-based 3D-EHTs were generated with hPSC-CMs and HS27a stromal cells (ATCC) at a 1:10 ratio. Twitch force was recorded by tracking the movement of magnets embedded in the flexible posts, as previously described (Bielawski et al., 2016) . For immunostaining, 3D-EHTs were arrested in diastole, fixed in 4% PFA and embedded in TissueTek O.C.T. before sectioning and staining. 

Mitochondrial metabolic manipulation by SARS-CoV-2 in peripheral blood mononuclear cells of patients with COVID-19

Acute myocarditis associated with novel Middle east respiratory syndrome coronavirus

The proximal origin of SARS-CoV-2

Characteristics and Outcomes of 21 Critically Ill Patients With COVID-19 in Washington State

A patient with rapid worsening dyspnoea during Covid-19 pandemic

SARS-CoV-2 Disrupts Splicing, Translation, and Protein Trafficking to Suppress Host Defenses

Embryological Origin of Human Smooth Muscle Cells Influences Their Ability to Support Endothelial Network Formation

Dynamics of genome reorganization during human cardiogenesis reveal an RBM20-dependent splicing factory

Chromatin compartment dynamics in a haploinsufficient model of cardiac laminopathy

COVID-19 and cardiac arrhythmias

Real-Time Force and Frequency Analysis of Engineered Human Heart Tissue Derived from Induced Pluripotent Stem Cells Using Magnetic Sensing

International Multisite Study of Human-Induced Pluripotent Stem Cell-Derived Cardiomyocytes for Drug Proarrhythmic Potential Assessment

SARS-CoV-2 infects and induces cytotoxic effects in human cardiomyocytes

Histopathology and ultrastructural findings of fatal COVID-19 infections in Washington State: a case series

The ACE2 expression in human heart indicates new potential mechanism of heart injury among patients infected with SARS-CoV-2

SARS-CoV-2 Infected Cardiomyocytes Recruit Monocytes by Secreting CCL2. Res Sq

Generation of human vascular smooth muscle subtypes provides insight into embryological origin-dependent disease susceptibility

The long QT syndromes: genetic basis and clinical implications

Hosts and Sources of Endemic Human Coronaviruses

Origin and evolution of pathogenic coronaviruses

Enhancing Serial Block-Face Scanning Electron Microscopy to Enable High Resolution 3-D Nanohistology of Cells and Tissues

SARS-CoV-2 in cardiac tissue of a child with COVID-19-related multisystem inflammatory syndrome

End-Stage Heart Failure With COVID-19: Strong Evidence of Myocardial Injury by 2019-nCoV

Cardiovascular Considerations for Patients, Health Care Workers, and Health Systems During the COVID-19 Pandemic

beta-Coronaviruses Use Lysosomes for Egress Instead of the Biosynthetic Secretory Pathway

Cardiovascular Implications of Fatal Outcomes of Patients With Coronavirus Disease 2019 (COVID-19)

Cardiomyocyte Maturation: New Phase in Development

The protein expression profile of ACE2 in human tissues

SARS-CoV-2 Cell Entry Depends on ACE2 and TMPRSS2 and Is Blocked by a Clinically Proven Protease Inhibitor

Cronavirus disease 2019 (COVID-19) Situation map

Clinical features of patients infected with 2019 novel coronavirus in Wuhan

Cardiac Involvement in Patients Recovered From COVID-2019 Identified Using Magnetic Resonance Imaging

The key role of Warburg effect in SARS-CoV-2 replication and associated inflammatory response

Cardiac Involvement in a Patient With Coronavirus Disease 2019 (COVID-19)

Inhibition of PIKfyve kinase prevents infection by Zaire ebolavirus and SARS-CoV-2

Cardiomyocyte maturation: advances in knowledge and implications for regenerative medicine

Activation and evasion of type I interferon responses by SARS-CoV-2

Physiological and pathological regulation of ACE2, the SARS-CoV-2 receptor

Association of Cardiac Infection With SARS-CoV-2 in Confirmed COVID-19 Autopsy Cases

Cells of the adult human heart

Immune signaling by RIG-I-like receptors

Influenza epidemics and acute respiratory disease activity are associated with a surge in autopsy-confirmed coronary heart disease death: results from 8 years of autopsies in 34,892 subjects

Learn from Your Elders: Developmental Biology Lessons to Guide Maturation of Stem Cell-Derived Cardiomyocytes

Host cell entry of Middle East respiratory syndrome coronavirus after two-step, furin-mediated activation of the spike protein

Host cell proteases: Critical determinants of coronavirus tropism and pathogenesis

COVID-19 and cardiovascular disease: from basic mechanisms to clinical perspectives

Cardiac Magnetic Resonance Imaging in Coronavirus Disease 2019 (COVID-19): A Systematic Review of Cardiac Magnetic Resonance Imaging Findings

Characterization of spike glycoprotein of SARS-CoV-2 on virus entry and its immune crossreactivity with SARS-CoV

SARS-coronavirus modulation of myocardial ACE2 expression and inflammation in patients with SARS

Patient and Disease-Specific Induced Pluripotent Stem Cells for Discovery of Personalized Cardiovascular Drugs and Therapeutics

Outcomes of Cardiovascular Magnetic Resonance Imaging in Patients Recently Recovered From Coronavirus Disease 2019 (COVID-19)

Cardiovascular Magnetic Resonance Findings in Competitive Athletes Recovering From COVID-19 Infection

The procoagulant pattern of patients with COVID-19 acute respiratory distress syndrome

Clinical predictors of mortality due to COVID-19 based on an analysis of data of 150 patients from Wuhan, China

SARS-CoV-2 direct cardiac damage through spike-mediated cardiomyocyte fusion. Res sq

A Novel Human Pluripotent Stem Cell-Derived Neural Crest Model of Treacher Collins Syndrome Shows Defects in Cell Death and Migration

Cell entry mechanisms of SARS-CoV-2

Human iPSC-Derived Cardiomyocytes Are Susceptible to SARS-CoV-2 Infection

Association of Cardiac Injury With Mortality in Hospitalized Patients With COVID-19 in Wuhan, China

A transmembrane serine protease is linked to the severe acute respiratory syndrome coronavirus receptor and activates virus entry

A unifying structural and functional model of the coronavirus replication organelle: Tracking down RNA synthesis

The trinity of COVID-19: immunity, inflammation and intervention

The pivotal link between ACE2 deficiency and SARS-CoV-2 infection

Clinical Characteristics of 138 Hospitalized Patients With 2019 Novel Coronavirus-Infected Pneumonia in

Double-Membrane Vesicles as Platforms for Viral Replication

The role of furin cleavage site in SARS-CoV-2 spike protein-mediated membrane fusion in the presence or absence of trypsin

Inhibition of SARS-CoV-2 (previously 2019-nCoV) infection by a highly potent pancoronavirus fusion inhibitor targeting its spike protein that harbors a high capacity to mediate membrane fusion

An Infectious cDNA Clone of SARS-CoV-2

Clinical Characteristics and Risk Factors of Cardiac Involvement in COVID-19

Novel Adult-Onset Systolic Cardiomyopathy Due to MYH7 E848G Mutation in Patient-Derived Induced Pluripotent Stem Cells

A Human Pluripotent Stem Cell-based Platform to Study SARS-CoV

Tropism and Model Virus Infection in Human Cells and Organoids

Targeting the Endocytic Pathway and Autophagy Process as a Novel Therapeutic Strategy in COVID-19

Clinical course and risk factors for mortality of adult inpatients with COVID-19 in Wuhan, China: a retrospective cohort study

A Novel Coronavirus from Patients with Pneumonia in China

Coronaviruses -drug discovery and therapeutic options

Predictors and Prognostic Implications of Cardiac Arrhythmias in Patients Hospitalized for COVID-19

Statistical analyses of intraexperimental variability as for panel B (F) Representative field potential traces in SARS-CoV-2-infected H7 hESC-CMs at 72 HPI, and quantification of field potential duration corrected by beat rate (FPDc) in spontaneous and paced experiments. Mean ± SEM of 8 and 6 wells for spontaneously beating and paced cells

Figure 6 -Force production of 3D-EHTs made from hiPSC-CMs progressively declines during SARS-CoV-2 infection. (A) Schematic representation of the magnetic sensing system: the 3D-EHT is suspended between two posts (one rigid, one flexible). The magnet is localized inside the flexible post, and the post movement during 3D-EHT contraction is recorded by giant magnetoresistance (GMR) sensor located at the bottom of the dish (converting displacement into voltage changes). (B) Representative picture of a 3D-EHT

CoV-2 viral RNA and particles detected in 3D-EHTs infected at 10 MOI at specified time points

Representative time course analysis of twitch force in 3D-EHTs from WTC11c hiPSC-CMs after SARS-CoV-2 infection at 10 MOI. Data are shown as mean ± SEM for 2 mock controls and 4 infected 3D-EHTs. (E) Representative twitch traces of 3D-EHT at 10 MOI at different time points. (F) Immunofluorescent images of 3D-EHTs sections at different time points. Scale bar: 20 µm (G) RT-qPCR of sarcomeric genes in 3D-EHTs

The authors declare no competing interests.J o u r n a l P r e -p r o o f