SARS-CoV-2 infection of human lung epithelial cells induces TMPRSS-mediated acute fibrin deposition

COVID-19, caused by SARS-CoV-2, leads to acute respiratory distress syndrome (ARDS) characterized by diffuse alveolar damage (DAD) and fibrin microthrombi in the lungs. Despite standard thromboprophylaxis with low molecular weight heparin, mortality remains high. While DAD is observed in other respiratory infections, the mechanism of SARS-CoV-2-induced DAD is poorly understood. Understanding this mechanism is crucial for developing effective therapies. The classical coagulation pathway involves a cascade of serine proteases culminating in thrombin activation and fibrin formation. SARS-CoV-2 infection triggers an inflammatory response, but the direct contribution of the virus to fibrin deposition is unclear. This study investigates the role of SARS-CoV-2 infection in fibrin deposition using primary human bronchial epithelial cells (NHBE).

Previous research has shown extensive DAD and fibrin microthrombi in the lungs of COVID-19 patients, contributing to reduced oxygen intake and the need for ventilation. Studies have observed thrombotic structures in pulmonary arteries, veins, and microvasculature. Despite the use of anticoagulants like heparin, mortality persists. While DAD is a general feature of ARDS and is seen in other infections, the precise mechanism of SARS-CoV-2-induced DAD is unknown. Several mechanisms have been proposed, such as dysfunctional thrombosis via the classical coagulation pathway, inflammation-induced TGF-β driven myofibroblast activation, and involvement of neutrophil extracellular traps. The classical coagulation pathway involves sequential activation of serine proteases leading to thrombin activation and fibrin formation. SARS-CoV-2 infection activates host inflammatory and antiviral responses, but the direct contribution to fibrin deposition remains unclear.

The study used NHBE cells, permissive to SARS-CoV-2 infection, as a model system. They were infected with both replication-incompetent SARS-CoV-2 pseudoviruses (pSARS-2) and replication-competent field strains. A turbidity-based fibrin clotting assay was used to measure fibrinogen cleavage and fibrin polymerization. Mass spectrometry-based proteomics analysis was performed on bronchoalveolar lavage fluid (BALF) from healthy individuals, acute COVID-19 patients, and recovered patients. ELISA was used to quantify fibrinogen, prothrombin, and total IgG concentrations. The impact of various SARS-CoV-2 variants on fibrin clot formation was assessed. The role of thrombin was investigated using protease and thrombin-specific inhibitors. Mass spectrometry identified proteins in infected and uninfected NHBE culture supernatants. Bovine prothrombin depletion and human prothrombin supplementation experiments were conducted. Transcriptome analysis using RNAseq examined gene expression in infected cells. Tissue factor knockdown was performed using CRISPR/Cas9. The potential involvement of TMPRSS family serine proteases was investigated. Fibrin clot formation in the presence of COVID-19 BALF was examined ex vivo. Confocal and scanning electron microscopy were used to visualize fibrin fibers.

Elevated prothrombin and fibrinogen levels were found in acute COVID-19 BALF compared to healthy and recovered samples. SARS-CoV-2 infection of NHBE cells, but not Vero or HEK293T cells, induced fibrin clot formation. This cell-mediated fibrin deposition was observed across various SARS-CoV-2 variants. The process required thrombin but was independent of tissue factor and other classical coagulation factors. Thrombin inhibitors (hirudin, dabigatran, argatroban) suppressed infection-induced fibrin clotting. Mass spectrometry revealed thrombin presence in infected NHBE supernatants. Bovine prothrombin depletion and human prothrombin supplementation confirmed the thrombin-dependency. Transcriptome analysis showed upregulation of type I interferon and antiviral responses but not significant tissue factor upregulation. Tissue factor knockdown did not affect fibrin clotting, indicating its independence from the classical pathway. TMPRSS family proteases (ST14 and TMPRSS11D), expressed in NHBE and HSAEC cells, were implicated in prothrombin activation. Metalloproteinase inhibitors reduced fibrin clot formation, suggesting TMPRSS shedding. Recombinant matriptase and HAT cleaved prothrombin and promoted fibrin clot formation. Transfection of ACE2-293T cells with ST14 or TMPRSS11D enabled fibrin clot formation upon infection. Acute COVID-19 BALF, but not healthy or recovered BALF, supported fibrin clot formation in the presence of infected NHBE cells. The study indicates a novel cell-mediated mechanism of fibrin deposition in COVID-19, independent of the classical coagulation pathway.

This study reveals a novel, non-classical mechanism of acute fibrin deposition in SARS-CoV-2-infected lungs. The findings highlight the limitations of current plasma-targeted anticoagulation therapies, as the mechanism is cell-mediated and independent of the classical coagulation cascade. The identification of TMPRSS family proteases (ST14 and TMPRSS11D) as key players in prothrombin activation offers potential targets for therapeutic intervention. Direct targeting of infected lung cells with thrombin inhibitors, potentially via nebulization, may be a more effective approach than systemic anticoagulation.

SARS-CoV-2 infection in lung epithelial cells triggers a cell-mediated fibrin deposition mechanism via activation of TMPRSS family proteases, independent of the classical coagulation pathway. This discovery challenges current anticoagulation strategies and suggests that direct targeting of infected lung cells with thrombin inhibitors could be a more effective therapeutic approach. Future research should focus on developing a suitable animal model to validate these findings in vivo and to further explore the therapeutic potential of targeted thrombin inhibition.

The study primarily utilizes in vitro and ex vivo models. The findings require in vivo validation, which is currently hampered by the lack of suitable animal models that replicate the human ARDS-like lung pathology. The variability in fibrin clot formation observed among acute COVID-19 BALF samples suggests other factors may influence the process. Further research is needed to fully understand these factors and their interactions.