Word and Element Counts:
Word Count: 500
Number of references: 5
Funding Statement: Support was provided solely from institutional and/or departmental sources
Conflicts of Interest: The authors declare no competing interests
Dear Editor,
We enjoyed the article by Cardot-Leccia et al. demonstrating apoptosis and pericyte loss (PL) in alveolar capillaries in COVID + lung [1]. This PL, with capillary and venular wall thickening in the area of DAD and in normal lung, without inflammation, and with intact epithelium and endothelium, is remarkable; we believe it supports our “Epithelial-Endothelial Cross-Talk” hypothesis [2] with additional players ─ pericytes.
As SARS-CoV-2 has minimal cytopathic effects and low immunogenicity [2], we believe that direct epithelial cell-injury is not the usual mechanism by which COVID-19 progresses to ARDS. Instead, we believe, “epithelial-endothelial-pericyte” cross-talk following SARS-CoV-2 infection of pulmonary alveolar epithelial cells (AEC) results in indirect activation of endothelial cells (EC) giving rise to a procoagulant-proinflammatory and profibrotic-phenotype similar to that seen in indirect-ARDS (Supplementary Fig. 1).
Based on our model, SARS-CoV-2 infection of AEC leads to downregulation of ACE2 and increased gene-expression of IL-6 and TNF-α in AEC via SARS-CoV2-ACE2-TACE and SARS-CoV2-PRR-NFκB interactions [3]. IL6-sIL-6Rα, IL-1β, and sTNF-α are then released on the luminal and abluminal sides of epithelial cells. IL6-sIL6R-1α and sTNF-α subsequently act on the EC from the basal side and increases the gene and protein expression of ACE and AT1R in the EC. Increased ACE activity on the apical side of EC increases local production of Ang II and AT1R upregulation results in increased Ang II-AT1R activity. Ang II-AT1R overactivity then increases TACE-TNF-α and ACE-AT1R expression and downregulates ACE2 in the EC, establishing a positive feed-forward pathway [3]. Increased ACE/ACE2 balance along with upregulated IL-6 and TNF-α mediated actions may stimulate a cascade of pathways, resulting in vasoconstriction, PL, endothelial barrier-disruption and cytokine-release syndrome (Fig. 1).
Direct action of Ang II-AT1R on pericytes can increase cytosolic calcium, resulting in pericyte contraction and capillary vasoconstriction. Ang II-AT1R-PKC pathway-mediated closure of intercellular gap-junctions can lead to pericyte-pericyte and pericyte-endothelial uncoupling [4]. Upregulation of Ang-2 and VEGF activity by Ang II-AT1R-MAPK pathways antagonizes protective Tie2-Ang-1 system resulting in PL, EC apoptosis and pathological angiogenesis. Increased TACE-mediated ectodomain shedding of ACE2 and Ang II-mediated reduction of ACE2 expression may relieve repression of integrin signaling rendering pericyte more susceptible to the Ang-2 and in turn, lead to the pericyte apoptosis by the Ang-2/integrin signaling pathway [5]. Ang II/AT1R/ROS regulated phosphorylation of PDGFRβ on pericytes may play a role in the pathobiology. Upregulation of ET-1 causes inappropriate pericyte contraction and reduced vessel diameter.
Our model explains the stage of PL when endothelium and epithelium are intact, perivascular inflammation is minimal and SARS-CoV2 infection of pericytes via ACE2 (even if present) seems unlikely. PL may be the first morphological change progressing to EC loss, leaving behind non-perfusing and constricted acellular capillaries. Capillary non-perfusion can stimulate intussusceptive and sprouting angiogenesis, which might explain findings wedge-shaped areas of reduced perfusion with dilatation of proximal vasculature as well as extensive angiogenesis in COVID-19.
PL and endothelial barrier-disruption can expedite pulmonary epithelium cell-injury and epithelial disruption, allowing hematogenous spread of infection. Notably, previous reports on SARS-CoV2 endothelitis have confirmed SARS-CoV2 directly infecting the EC [1].