Introduction

The global pandemic caused by SARS-CoV-2 is increasing morbidity and mortality worldwide. While the virus primarily targets the respiratory system during the course of the disease other organs such as the heart are infected as well. Nevertheless, several questions remain concerning that can best be answered by investigating myocardial tissue directly. This includes the open question about (a) possibilities of viral replication within the myocardial tissue, (b) localization of the viral particles within the distinct cell types in the heart as well as (c) systematic changes of gene expression associated with SARS-CoV-2 cardiac infection. Therefore, we examined in the largest autopsy study so far the changes of myocardial tissue in patients died with confirmed SARS-CoV-2 infection.

Material and Method

Consecutively, 95 deceased patients with diagnosed SARS-CoV-2 infection have been autopsied. Two tissue specimens from the left ventricle were collected and either snap frozen or fixed in formalin. Total RNA was isolated and used for gene expression analysis using TaqMan and to determine the virus load and virus replication in cardiac tissue. For 3’mRNA sequencing analysis, ten patients with cardiac infection were compared to 10 age-matched patients without cardiac infection. To visualize the virus within cardiac tissue, RNA scope, immune fluorescence staining and electron microscopy were performed on tissue sections.

Results:

In 46 out of 95 patients, no SARS-CoV-2 was detected in cardiac tissue, whereas in 41 patients more than 1000 copies per µg RNA, which we deemed as clinical significant, were determined. Additionally, in 28 out of these 41 patients with cardiac infection, virus replication was detectable. To investigate the localization of virus particles we performed fluorescent co-staining. In the majority of cases, virus particles are localized in the interstitium or in interstitial cells.

During autopsy, comorbidities were diagnosed, odds ratios for cardiac comorbidities revealed no association between comorbidities and cardiac infection or replication.

For 3’ mRNA sequencing, 10 patients with high virus load in cardiac tissue were selected and compared to 10 age-matched patients without cardiac infection. 23,335 genes were annotated whereof 824 genes were identified as differentially expressed. Out of these 824 genes, 394 genes were significantly up-regulated and 430 genes significantly down-regulated in cardiac tissue with high virus load of SARS-CoV-2. All differentially expressed genes were further restricted by FDR < 0.05 resulting in the top 19 differentially expressed genes. Next, Gene Ontology enrichment analysis was performed to identify significant GO terms. We selected all GO terms which include at least 3 of the top 19 genes. Interestingly, 15 out the top 19 genes are annotated in GO terms linked to the immune system. The majority of genes in the GO terms linked to the immune system were upregulated.

Conclusion:

In this analysis of autopsy cases, viral genome and in some of these cases viral replication within the myocardium could be documented and visualized. Using 3’mRNA sequencing, a response to this infection could be determined in cases with high virus load vs. no virus infection.  Gene Ontology enrichment analysis further emphasizes the SARS-CoV-2 infection of the cardiac tissue since GO terms linked to the immune systems were significantly regulated.