Background. Since the beginning of the COVID-19 pandemic, It has been controversial whether the heart of COVID-19 patients is infected by SARS-CoV-2 and how this affects disease severity. K18 transgenic mice express the SARS-CoV-2 entree receptor human ACE2 in epithelial cells (K18-hACE2-Tg) but not on cardiomyocytes. Here, we established  a mouse model by additional hACE2 expression in cardiomyocytes to answer the fundamental questions whether SARS-CoV-2 can a) infect the heart, b) replicate in the heart and c) impair cardiac function. Doing so we provide first evidence for a yet hidden heart-lung crosstalk in COVID-19

Methods. We infected 8-week-old male and female K18-hACE2-Tg mice with an adeno-associated virus type 9 (AAV9) that expresses hACE2 (AAV9-hACE2/K18-hACE2-Tg) or YFP as a control (AAV9-YFP/K18-hACE2-Tg), allowing SARS-CoV-2 to enter murine cardiomyocytes. These mice were intranasally exposed to SARS-CoV-2. Virus replication was determined by plaque-assays, diseases progression by a standardized disease score (DS: change in body weight, damage to the skin and fur, and impaired spontaneous reaction and breathing, systolic and diastolic cardiac function by echocardiography (ejection fraction and E/e’), immune cell infiltration, and gene expression changes by bulk-RNAseq of heart and lung. Due to the rapid progression of COVID-19, mice needed to be sacrificed for organ harvesting and gene expression analyses 5 days post infection (dpi).

Results. K18-hACE2-Tg exposed to SARS-Cov-2 showed high levels of SARS-CoV-2 replication in the lung and brain but no replication in the heart. The COVID-19 DS markedly increased at 3-4 dpi. AAV9-hACE2/K18-hACE2-Tg showed further worsening of COVID-19 severity (BW changes 3 vs 0 dpi: AAV9-YFP/K18-hACE2-Tg: 98.0 ± 0.61 %; n=9 vs. AAV9-hACE2/K18-hACE2-Tg: 92.1 ± 1.00 %, n=8, p=0.002; DS changes 4 vs 0 dpi: AAV9-YFP/K18-hACE2-Tg: 1.9 ± 0.9; n=9 vs. AAV9-hACE2/K18-hACE-Tg 11.75 ± 1.3, n=8, p=0.0009). Despite hACE2 expression in the heart and exaggerated disease progression, we still could not detect SARS-CoV-2 replication in the heart. Both AAV9-YFP/K18-hACE-Tg and AAV9-hACE2/K18-hACE2-Tg showed regular cardiac performance during disease progression in daily performed echocardiographies without any sign of acute cardiac dysfunction. Immune cell infiltration into the heart was not increased. In contrast, bulk RNA-seq data revealed that AAV9-hACE2/K18-hACE-Tg showed higher amount of specific SARS-CoV-2 virus particles despite the lack of SARS-CoV2 replication. Gene Set Enrichment Analysis (GSEA) revealed upregulation of TNFα, NFκB and apoptosis pathways and downregulation of interferon (IFN) α/γ signaling in the heart. In the lung, GSEA  revealed besides upregulation of TNFα, NFκB and apoptosis and downregulation of IFNα/γ signaling, upregulation of hypoxia and TGFβ pathways.