Procoagulant phenotype of virus-infected pericytes is associated with portal thrombosis and intrapulmonary vascular dilations in fatal COVID-19
Massimiliano Cadamuro,Alberto Lasagni,Claudia Maria Radu,Arianna Calistri,Matteo Pilan,Clarissa Valle,Pietro Andrea Bonaffini,Adriana Vitiello,Serena Toffanin,Camilla Venturin,Yahima Friòn-Herrera,Sandro Sironi,Maria Grazia Alessio,Giulia Previtali,Michela Seghezzi,Andrea Gianatti,Mario Strazzabosco,Alastair J Strain,Elena Campello,Luca Spiezia,Giorgio Palù,Anna Chiara Frigo,Antonella Tosoni,Manuela Nebuloni,Cristina Parolin,Aurelio Sonzogni,Paolo Simioni,Luca Fabris
DOI: https://doi.org/10.1016/j.jhep.2024.06.014
Abstract:Background & aims: The underlying mechanisms and clinical impact of portal microthrombosis in severe COVID-19 are unknown. Intrapulmonary vascular dilation (IPVD)-related hypoxia has been described in severe liver diseases. We hypothesised that portal microthrombosis is associated with IPVD and fatal respiratory failure in COVID-19. Methods: Ninety-three patients who died from COVID-19 were analysed for portal microvascular damage (histology), IPVD (histology and chest-computed tomography, CT), and hypoxemia (arterial blood gas). Seventeen patients who died from COVID-19-unrelated pneumonia served as controls. Vascular lesions and microthrombi were phenotyped for endothelial (vWF) and pericyte (αSMA/PDGFR-β) markers, tissue factor (TF), viral spike protein and nucleoprotein (SP, NP), fibrinogen, and platelets (CD41a). Viral particles in vascular cells were assessed by transmission electron microscopy. Cultured pericytes were infected with SARS-CoV-2 to measure TF expression and tubulisation of human pulmonary microvascular endothelial cells was assessed upon vWF treatment. Results: IPVD was present in 16/66 patients with COVID-19, with available liver and lung histology, and was associated with younger age (62 vs. 78 years-old), longer illness (25 vs. 14 days), worsening hypoxemia (PaO2/FiO2 from 209 to 89), and an increased requirement for ventilatory support (63% vs. 22%) compared to COVID-19/Non-IPVD. IPVD, absent in controls, was confirmed by chest CT. COVID-19/IPVD liver histology showed portal microthrombosis in >82.5% of portal areas, with a thicker wall of αSMA/PDGFR-β+/SP+/NP+ pericytes compared with COVID-19/Non-IPVD. Thrombosed portal venules correlated with αSMA+ area, whereas infected SP+/NP+ pericytes expressed TF. SARS-CoV-2 viral particles were observed in portal pericytes. In vitro SARS-CoV-2 infection of pericytes upregulated TF and induced endothelial cells to overexpress vWF, which expanded human pulmonary microvascular endothelial cell tubules. Conclusions: SARS-CoV-2 infection of liver pericytes elicits a local procoagulant response associated with extensive portal microthrombosis, IPVD and worsening respiratory failure in fatal COVID-19. Impact and implications: Vascular involvement of the liver represents a serious complication of COVID-19 infection that must be considered in the work-up of patients with long-lasting and progressively worsening respiratory failure, as it may associate with the development of intrapulmonary vascular dilations. This clinical picture is associated with a procoagulant phenotype of portal venule pericytes, which is induced by SARS-CoV-2 infection of pericytes. Both observations provide a model that may apply, at least in part, to other vascular disorders of the liver, featuring obliterative portal venopathy, similarly characterised at the clinical level by development of hypoxemia and at the histological level by phlebosclerosis and reduced calibre of the portal vein branches in the absence of cirrhosis. Moreover, our findings shed light on an overlooked player in the pathophysiology of thrombosis, i.e. pericytes, which may present a novel therapeutic target.