COVID19 Impairs Cardiac Function via Endothelial H19 and IL6 Signaling

Cardiac complications occur in 40% of hospitalized patients with COVID-19, and individuals with COVID-19 are at increased risk for cardiovascular postacute sequela.¹ Although direct cardiac SARS-COV-2 invasion is a plausible underlying mechanism, evidence shows little or no presence of SARS-COV-2 genome in premortem or postmortem cardiac biopsies,² in line with the low incidence of virus particles found in the circulation.³ Cardiac microvascular endothelial cells (CMECs), which form the capillaries adjacent to cardiomyocytes in the heart, are important in regulating cardiac function.⁴ In the present study, we show that COVID-19 reduces the expression of long noncoding RNA H19-a known anti-inflammatory transcript in endothelium,⁵ leading to cardiac endothelial dysfunction and subsequent impairment of myocardial contractility, independent of viral invasion to the heart.

Meet the First Author, see p 544

We discovered that H19 expression was significantly lower in cardiac biopsies from patients with COVID-19 (Table S1) as compared to non-COVID-19 controls (Figure [A], left), despite the undetected levels of SARS-COV-2 genome (data not shown). Using single-cell resolution in situ hybridization, we found that H19 was enriched in cardiac endothelium (Figure [A], middle, left) as it was expressed along the endothelial-specific marker vascular endothelial cadherin (VE-cadherin) (Figure [A], middle, right), and confirmed the reduced expression of H19 in cardiac endothelium in COVID-19 heart biopsies (Figure [A], middle, left). H19 is important for endothelial function as its silencing (siH19) in primary human CMECs induced endothelial stress, shown as irregular and increased adherens junction area stained by VE-cadherin, and reduced endothelial NOS3 (nitric oxide synthase 3) expression (Figure [B]). In line with our previous study,⁵ H19 silencing induced STAT3 (signal transducer and activator of transcription 3) activation and increased level of IL (interleukin)-6 (Figure [B]), indicating proinflammatory activation of cardiac endothelial cells upon H19 depletion.

Figure.

Dysregulation of H19 in COVID-19 induces cardiac endothelial dysfunction, leading to impairment of cardiomyocyte (CM) function. A, Left, Quantitative real-time PCR (qRT-PCR) analysis of H19 in postmortem cardiac biopsies from control (non-COVID-19, n=8) vs patients with COVID-19 (n=14 donors, unpaired t test). Middle, Single-cell resolution in situ hybridization on tissues (SCRINSHOT) of H19 (green, arrows, left) and sequential slides of immunofluorescent (IF) staining of endothelial marker vascular endothelial cadherin (VE-cadherin) (yellow, right, lining the cardiac microvasculature [dashed line]) in the cardiac biopsies (DAPI: blue; cTNT: green, right). Right, Staining overview. Top, IF of VE-cadherin (Cell Signaling; No. 2500), cTNT (Abcam; No. ab8295), and DAPI of low (left) and high magnification (right) of a cardiac tissue. Bottom, SCRINSHOT negative (no H19 probe) and positive control (H19 probe added) from sequential slides of the same cardiac biopsy. Most representative images were shown. B, IF of VE-cadherin (green; DAPI: blue) and adherens junction area in human cardiac microvascular endothelial cells (CMECs) upon siRNA-mediated H19 silencing (siH19) vs siRNA control (sictr); qRT-PCR of NOS3 (nitric oxide synthase 3); Western blot of phospho-STAT3 (Y705, No. 9145; Cell Signaling) over total STAT3 (Cell Signaling; No. 4904); and ELISA of IL-6 (interleukin-6) (Thermo Fisher Scientific; No. EH2IL6) in CMECs (n=4 biological replicates/different donors, Mann-Whitney U test). C, Coculture of CMECs and adult rat ventricular CMs. CMECs were transfected with siH19 or sictr for 48 hours and subsequently cocultured with CMs. Measurement of CM contraction (% sarcomere shortening) and relaxation (return velocity and tau) upon field stimulation using video-based MultiCell system (CytoCypher; n=6 independent experiments, 30–40 CMs were measured per condition per experiment, ANOVA followed by Tukey multiple comparison test). D, qRT-PCR of H19 in CMECs treated with plasma from patients with COVID-19 from ward or intensive care unit (ICU) (n=5 plasma donors per condition, Kruskal-Wallis test). Coculture of CMs with CMECs pretreated for 6 hours with the collected plasma (the plasma was washed out, and the medium was refreshed before CMECs were cocultured with CMs). Measurement of CM contraction and relaxation after coculture (n=5 independent experiments, Kruskal-Wallis test). E, qRT-PCR of IL-6 in cardiac biopsies from control (non-COVID-19, n=8) vs patients with COVID-19 (n=14, unpaired t test). Measurement of CM function after coculture with CMECs (+CMEC) transfected with siH19, in the presence or absence of tocilizumab (tcz) administered in the CM compartment (n=6 independent experiments, ANOVA). F, Measurement of CM function after coculture with CMECs (+CMEC) pretreated with plasma from patients with COVID-19 before and after the patients received tocilizumab (n=5 independent experiments, Kruskal-Wallis test). Data are presented as mean±SD. Differences were considered significant when P was <0.05.

To underline the functional effect of endothelial H19 silencing on cardiomyocyte contractility, we cocultured CMECs devoid of H19 with cardiomyocytes (Figure [C]) and measured cardiomyocyte contractile function.⁴ Silencing of H19 in CMECs impaired the beneficial effects of endothelial cells on cardiomyocyte contraction (shown as reduced percentage of shortening) and relaxation (shown as reduced return velocity and longer constant time of relaxation, tau), as compared with CMECs with no H19 silencing (siRNA control; Figure [C]).

As patients with COVID-19 are characterized by hyperinflammatory response indicated by elevated levels of circulating proinflammatory cytokines, such as TNFα (tumor necrosis factor alpha), IL-1, and IL-6, we wondered whether this proinflammatory state impairs cardiac endothelial control of cardiomyocyte function. Interestingly, COVID-19 plasma reduced endothelial H19 levels (Figure [D]). We then pretreated CMECs with plasma from hospitalized patients with COVID-19 (Table S1) and subsequently cocultured them with cardiomyocytes (Figure [D], experiment setup). Interestingly, COVID-19 plasma impaired the ability of CMECs to enhance cardiomyocyte contraction and relaxation (Figure [D]), the effects that were more pronounced in the intensive care unit (ICU) than in the ward patient group. These indicate that the proinflammatory state induced in patients with COVID-19 affects cardiac endothelial H19 levels, leading to cardiac endothelial dysfunction, which impairs cardiomyocyte function.

In line with increased endothelial IL-6 level upon H19 silencing in CMECs (Figure [B]), we found increased IL-6 expression in cardiac biopsies of patients with COVID-19 (Figure [E]), indicating that IL-6 can act as an endothelial-derived factor released upon H19 reduction that impairs cardiomyocyte function. Interestingly, using the IL-6 receptor inhibitor tocilizumab in our coculture experiments (Figure [E], experiment setup), we showed that inhibition of IL-6 pathway on cardiomyocytes can restore cardiomyocyte contractile function upon coculture with CMECs devoid of H19 (Figure [E]), underlining the role of endothelial-derived IL-6 in mediating the detrimental effect of H19 depletion on cardiomyocyte function.

Lastly, to further translate these findings to the clinical situation, we treated CMECs with plasma samples collected from ICU patients with COVID-19 before and after the patients received tocilizumab treatment (Figure [F]). We demonstrated that exposure to the plasma post-tocilizumab treatment restored endothelial enhancement of cardiomyocyte function (Figure [F]), as compared with the exposure to the plasma pre-tocilizumab treatment, further underlining the contribution of IL-6 in the COVID-19–induced cardiac contractile impairment.

In conclusion, we demonstrated that H19-a cardiac endothelium-enriched long noncoding RNA—is downregulated in the heart of patients with COVID-19. H19 silencing in CMECs induces cardiac endothelial dysfunction and impairs cardiomyocyte function. Exposure of CMECs to COVID-19 plasma reduces H19 levels and impairs endothelial regulation of cardiomyocyte function. Mechanistically, H19 silencing induced cardiac endothelial STAT3 activation with increased IL-6 expression, which was also observed in the cardiac biopsies from patients with COVID-19, suggesting that reduced H19 level causes endothelial-derived IL-6 release, inducing impairment of cardiomyocyte function. Interestingly, the IL-6 receptor blocker tocilizumab restored impairment of cardiomyocyte function upon coculture with CMECs devoid of H19. Moreover, exposure to plasma from patients with COVID-19 post-tocilizumab treatment rescued the impairment of endothelial enhancement of cardiomyocyte function. Our findings provide the first proof that COVID-19 induces dysregulation of long noncoding RNA H19, which results in cardiac endothelial dysfunction, leading to impairment of cardiac function—a novel mechanism underlying the pathogenesis of cardiac sequela in COVID-19. Further, we showed that the IL-6 pathway interference with tocilizumab can rescue the blunted cross talk between cardiac endothelium and cardiomyocytes, highlighting the potential of this pathway as a target therapy in patients with COVID-19 with cardiac complications.

Article Information

Acknowledgments

The authors acknowledge Valentijn Jansen, MSc, for excellent technical assistance.

Author Contributions

R.P. Juni and R.A. Boon designed the study; R.P. Juni, P.G. Phelp, D.I. Bink, V. Kremer, A.S. van Bergen, K.B. Rombouts, P.E. Gonzalez, K.J. Harber, and D.A.F. Heister performed the experiments; P.A.J. Krijnen and H.W.M. Niessen provided cardiac biopsies from COVID-19 patients; Amsterdam UMC COVID-19 Biobank provided plasma samples from COVID-19 patients; R.P. Juni, P.G. Phelp, D.I. Bink, and V. Kremer analyzed the data; R.P. Juni made the figures; R.P. Juni drafted the manuscript; R.P. Juni, P.G. Phelp, D.I. Bink, A.S. van Bergen, D.W.D. Kuster, P.A.J. Krijnen, H.W.M. Niessen, K.K. Yeung, J. Van den Bossche, and R.A. Boon revised the manuscript; all authors approved the final version of the manuscript.

Data Availability

No next generation sequencing, proteomics or similar high throughput data was generated for this study. The data that was generated for this study is available upon request.

Sources of Funding

This work was supported by the European Union (ERC, project number 101002599), the Netherlands Organisation for Scientific Research (NWO Vidi), the European Union (Horizon 2020 Grant No. 825670), and the Deutsche Forschungsgemeinschaft (TRR267).

Disclosures

None.