Cardiorenal Tissues Express SARS-CoV-2 Entry Genes and Basigin (BSG/CD147) Increases With Age in Endothelial Cells

Visual Abstract

. In addition, basigin (BSG) (also known as CD147 or EMMPRIN) is a second but putative receptor by which SARS-CoV-2 may enter cells (2,3). For viral entry by ACE2, it is thought that the SARS-CoV-2 spike protein is primed, and ACE2 cleaved, by the cellular serine proteases transmembrane serine protease 2 (TMPRSS2) (1) and ADAM metallopeptidase domain 17 (ADAM17). Intracellular processing of SARS-CoV-2 spike protein is thought to involve the lysosomal cysteine proteases cathepsin B (CTSB) and cathepsin L (CTSL), which can also substitute for TMPRSS2 in some cells (1). FURIN cleaves viral enveloping proteins, providing another putative priming step for the spike protein of SARS-COV-2 (4).
For viral entry via BSG, less is known regarding specific receptor and viral processing partners for SARS-CoV-2. Indeed, firm evidence for BSG as a stand-alone receptor for SARS-CoV-2 remains the subject of investigation, with a recent study noting no "direct" binding of SARS-CoV-2 spike protein to BSG (5). However, for SARS-CoV (6), human immunodeficiency virus (7), and the measles virus (8), respectively, peptidylprolyl isomerase A (PPIA; also known as cyclophilin A) and peptidylprolyl isomerase B (PPIB; also known as cyclophilin B), which are natural ligands for BSG, incorporate into virus and facilitate binding to BSG. Similarly, cyclophilins are required for infection via BSG in malaria. In this case, PPIB forms a complex with the malaria pathogen (Plasmodium falciparum merozoites) and BSG to facilitate infection of red blood cells (9).
Initial infection with SARS-CoV-2 occurs via the respiratory epithelium; high gene expression of ACE2 and TMPRSS2 in nasal epithelium (10,11) has been taken to imply that the nose is a primary entry point for the virus (10). ACE2 and TMPRSS2 are also coexpressed in bronchial epithelium (10)(11)(12). However, where COVID-19 progresses to severe disease, the lung and other organs are also affected. The emerging pattern of severe and fatal COVID-19 includes pneumonia with acute respiratory distress syndrome, cytokine storm, widespread vasculopathy, thrombosis, renal failure, hypertension, and endothelial dysregulation seen across multiple vascular beds and organ systems (13,14). Although hypertension and thrombosis are common features after 15), the important question as to whether COVID-19 is an independent risk factor for cardiovascular disease in the acute setting and during the recovery period is a concern and remains to be established.
This secondary thrombotic and vascular clinical syndrome of severe COVID-19 suggests that SARS-CoV-2 infects not only respiratory epithelium but also the endothelium, disrupting barrier function and allowing access to cardiovascular tissues and other organs of the body (16). This idea is supported by reports showing that SARS-CoV-2 can infect endothelial cells in vitro (17) and that coronaviruses including SARS-CoV-2 can progress to systemic infection (18,19),  were obtained from Genotype-Tissue Expression from cardiorenal tissues (aorta, coronary artery, atrial appendage [AA], left ventricle [LV], kidney cortex, and whole blood; red columns) and other tissues (lung, blue columns; spleen, gray columns; and colon, green columns). Data for each tissue corrected for batch effects using ComBat-seq and expressed as individual points and mean AE SEM. Tissues are ranked in order of expression for each gene. A heat map showing expression of angiotensin-converting enzyme 2 (ACE2) and basigin (BSG) pathways and associated viral processing proteases in each tissue was generated (K). Data are colored by gene, whereby black is the lowest expressing tissue and red is the highest expressing tissue. SARS-CoV-2 ¼ severe acute respiratory syndrome coronavirus-2.  Although some studies report expression profiles of ACE2 and TMPRSS2 in epithelial cells (10,12) and immune cells (11,12), expression patterns of a wider range of host SARS-CoV-2 entry and processing genes in these cells were recently reported (12). However, the relative expression levels of SARS-CoV-2 entry and processing genes in vessels and in endothelial cells have not been fully established. Finally, the impact of age on the expression of these genes in a cardiovascular setting is incompletely understood.
Here we have used publicly available gene expression data to determine the relative expression of key SARS-CoV-2 host entry and processing genes in human cardiorenal tissues, including aorta, coronary artery, heart (atria and left ventricle), whole blood, and the kidney and for comparison the colon, spleen, and lung. We went on to investigate gene expression in endothelial cells and, for comparison, airway (nasal and bronchial) epithelium and leukocytes (peripheral blood mononuclear cells [PBMCs]). We used blood outgrowth endothelial cells as a model because, as they are obtained from blood samples of living donors, datasets across age ranges have been created. cardiorenal tissues including the aorta, coronary artery, heart (atrial and appendage), left ventricle, kidney (cortex), and whole blood; and 2) "other tissues," including lung, colon, and spleen. We performed principal-component analysis on gene expression data from each tissue of interest. We observed that the major variation in gene expression was due to type of death (Hardy score) (Supplemental Figure 1) and so corrected for this. We normalized the gene expression data for each tissue separately using ComBat-seq (29), with Hardy score as a batch. After and CTSL (J) were obtained from online databases from human blood outgrowth endothelial cells (endothelial cells; red columns), peripheral blood mononuclear cells (PBMCs) (gray columns), and epithelial cells (nasal and bronchial; blue columns). The data were aligned and analyzed using Partek Flow and corrected for batch effects using ComBat-seq and expressed as individual data points and mean AE SEM. Cells were ranked in order of expression each gene. A heat map showing expression of ACE2 and BSG pathways and viral processing proteases in each cell type was generated (K). Data are colored by gene, whereby black is the lowest expressing cell type and red is the highest expressing cell type. Abbreviations as in Figure 1.   with very low levels present in blood (Figures 1 and 3).
Of the 2 putative SARS-CoV-2 receptors, BSG was highly expressed across all tissues, with higher levels seen in most cardiorenal tissues than in the lung or spleen. ACE2, across all tissues, was expressed in relatively low levels. However, cardiorenal tissues including kidney, heart, and blood vessel (coronary artery) expressed higher levels of CTSB and CTSL were more highly expressed in arteries than in kidney, heart, or blood. S p l e e n . None of our studied genes were altered with age in the spleen (Figure 3).

EFFECT OF AGE ON EXPRESSION OF SARS-CoV-2 ENTRY GENES IN ENDOTHELIAL CELLS, AIRWAY
CELLS, AND LEUKOCYTES. In endothelial cells BSG, but not other genes, was increased in samples from adults >40 years of age (Figures 3 and 5), and levels showed a positive linear correlation with age ( Figure 5). In contrast, only ACE was fractionally (but statistically significantly) reduced in nasal epithelium between age categories (Figure 3), and this did not linearly correlate with age (Supplemental Figure 4).
No genes were altered in bronchial epithelium and PBMCs with age ( Figure 3).
Although not the primary outcome of our study, which was age, we also investigated the effect of sex Moreover, we found that both ACE2 and TMPRSS2 were enriched in nasal epithelium, with low levels in bronchial epithelium and PBMCs, which is in agreement with recent work from others (nasal vs. bronchial epithelium [10,11] and versus PBMCs [11]). followed by ACE2 in airway cells. However, our focus was on the cardiovascular system and kidney.
Importantly, we confirm that endothelial cells express ACE2 and TMPRSS2, although at lower levels than nasal epithelium but higher levels than bronchial epithelial cells (ACE2) and PBMCs (ACE2 and TMPRSS2). These findings suggest that SARS-COV-2 could infect endothelial cells via the ACE2 pathway.
In cells in which TMPRSS2 is low, SARS-CoV-2 can gain access by using CTSL and/or CTSB (1). In our study both CTSL and CTSB were found to be enriched Severe COVID-19 is exceptionally rare in children.
In adults, the strongest risk factor for severe disease and death is age, with those younger than 40 years being at very low risk; the risk for severe COVID-19 and death increases proportionally after the age of 40 (24). Of the genes we studied, several candidates, including ACE2, were affected by age, but with the exception of BSG in endothelial cells and PPIB and FURIN in aorta, expression was reduced in those >40 years of age. We found consistent age-related reductions in ACE2 in whole blood, aorta, and the colon. Our findings are in line with those published by Chen et al. (33), who also reported a negative correlation between ACE2 and age in a range of tissues including colon and blood. Moreover, our work corroborates earlier studies showing that ACE2 (protein) declines with age in mouse aorta (41). Other studies in rats also showed that ACE2 declines with age in the Ahmetaj-Shala et al. lung and kidney (42,43). It should be noted, however, that Chen et al. (34) found no effect of age on ACE2 expression across a similar selection of tissues and that Santesmasses et al. (44) found that ACE2 expression increased with age in the lung. We also found a trend for ACE2 to increase in the lung, but this did not reach statistical significance in our study.
Key differences between the studies include the analytic approaches applied, the number of tissues selected, and the age groups used.