Tissue Factor-Enriched Neutrophil Extracellular Traps Promote Immunothrombosis and Disease Progression in Sepsis-Induced Lung Injury

Abstract

Background: Patients with sepsis may progress to acute respiratory distress syndrome (ARDS). Evidence of neutrophil extracellular traps (NETs) in sepsis-induced lung injury has been reported. However, the role of circulating NETs in the progression and thrombotic tendency of sepsis-induced lung injury remains elusive. The aim of this study was to investigate the role of tissue factor-enriched NETs in the progression and immunothrombosis of sepsis-induced lung injury.

Methods: Human blood samples and an animal model of sepsis-induced lung injury were used to detect and evaluate NET formation in ARDS patients. Immunofluorescence imaging, ELISA, Western blotting, and qPCR were performed to evaluate in vitro NET formation and tissue factor (TF) delivery ability. DNase, an anti-TF antibody, and thrombin inhibitors were applied to evaluate the contribution of thrombin to TF-enriched NET formation and the contribution of TF-enriched NETs to immunothrombosis in ARDS patients.

Results: Significantly increased levels of TF-enriched NETs were observed in ARDS patients and mice. Blockade of NETs in ARDS mice alleviated disease progression, indicating a reduced lung wet/dry ratio and PaO2 level. In vitro data demonstrated that thrombin-activated platelets were responsible for increased NET formation and related TF exposure and subsequent immunothrombosis in ARDS patients.

Conclusion: The interaction of thrombin-activated platelets with PMNs in ARDS patients results in local NET formation and delivery of active TF. The notion that NETs represent a mechanism by which PMNs release thrombogenic signals during thrombosis may offer novel therapeutic targets.

Keywords: acute lung injury; immunothrombosis; neutrophil extracellular traps; sepsis; tissue factor.

Figure 1

High levels of neutrophil and neutrophil extracellular traps accumulate in sepsis ARDS patients and correlate with worse outcomes. (A) CT imaging of the lungs of healthy volunteers, sepsis patients, and ARDS patients. (B) Fluorescence imaging of human peripheral blood neutrophils among healthy volunteers, sepsis patients, and ARDS patients for 6 hours. The sections were immunostained with MPO (red) and CitH3 (green), and DAPI (blue) was used to counterstain the nuclei. (C) Ex vivo NET assay of neutrophils in healthy volunteers, sepsis patients, and ARDS patients. (D) Ex vivo MPO-DNA complex ELISA in healthy volunteers, sepsis patients, and ARDS patients. (E) Patient survival curve in the low or high NET group of ARDS patients (n=16). (F) Patient survival curve in the low or high MPO-DNA group. (G) Correlation curve between the MPO-DNA complex and PaO₂/FiO₂. (H) Correlation curve between APACHE II score and MPO-DNA complex. Statistical analysis was performed using the log-rank (Mantel-Cox) test. P < 0.05 (*) and P < 0.05 (^#) compared to baseline healthy controls and sepsis patients were considered statistically significant.

Figure 2

NETs increase in a sepsis-induced lung injury model. (A) Lung tissues were collected from C57BL/6 mice in the control group, sham group, ARDS group, or ARDS+DNase group. H&E staining was performed on paraffin-embedded sections of mouse lungs to detect lung injury. (B) Changes in the lung wet/dry ratio in the four groups (C). Changes in PaO₂ as the processing time increased in the four groups. (D) Immunofluorescence of lung tissue in red (MPO) and blue (DAPI). (E) Fluorescence imaging of mouse peripheral blood neutrophils among the healthy control, sham, ARDS, and ARDS+DNase groups. The sections were immunostained with MPO (red) and CitH3 (green), and DAPI (blue) was used to counterstain the nuclei. (F). Ex vivo plasma DNA. (G) MPO-DNA complex. (H) NET-releasing cell count assay. (I) NET degradation among the healthy control, sham, ARDS, and ARDS+DNase groups (n=6). P < 0.05 (*) and P < 0.05 (^#) compared to baseline healthy controls and the sepsis group were considered statistically significant.

Figure 3

Intracellular TF in polymorphonuclear neutrophils of ARDS patients is associated with NET formation. (A) Fluorescence imaging of lung thrombi in ARDS mice. The sections were immunostained with TF (red) and CitH3 (green), and DAPI (blue) was used to counterstain nuclei. (B) Plasma of healthy controls (HC), sepsis or ARDS patients was cocultured with polymorphonuclear neutrophils (PMNs) in HC. Fluorescence imaging of PMNs in the three groups. The sections were immunostained with TF (red) and CitH3 (green), and DAPI (blue) was used to counterstain the nuclei. (C) An increase in NETs was detected by coculture with different plasma samples (n=6). (D) Western blots for TF expression in the HC, sepsis, ARDS or ARDS+DNase groups (n=6). (E) mRNA fold change for TF expression in HC, sepsis, ARDS or ARDS+DNase groups (n=6). (F) TAT complex progression in the HC, sepsis, ARDS+DNase, ARDS, or ARDS+ anti-TF groups (n=6). P < 0.05 (*) and P < 0.05 (^#) compared to baseline healthy controls and the sepsis group were considered statistically significant.

Figure 4

Platelets activated in ARDS are able to induce NET formation. (A) Fluorescence-activated cell sorting analysis of Annexin V on platelets of HC platelets, sepsis patients, and ARDS patients (n=6). (B) Fluorescence-activated cell sorting analysis of CD62P on platelets of HC platelets, sepsis patients, and ARDS patients (n=6). (C) Platelet/polymorphonuclear neutrophil aggregates observed as double-positive CD61/CD11b per 10000 CD11b-positive events with fluorescence-activated cell sorting analysis in polymorphonuclear neutrophils isolated from HC, sepsis, and ARDS patients (n=6). (D) NET generation by control polymorphonuclear neutrophils treated with platelets isolated from the 3 groups. The sections were immunostained with TF (red) and CitH3 (green), and DAPI (blue) was used to counterstain the nuclei. (E) Percentage of NET-releasing PMNs (n=6). P < 0.05 (*) and P < 0.05 (^#) compared to baseline healthy controls and the sepsis group were considered statistically significant.

Figure 5

Activated platelets for subsequent NET generation in PMNs. (A) Neutrophil extracellular trap formation by control polymorphonuclear neutrophils incubated with control platelets pretreated with plasma obtained from the 3 groups. (B) Fluorescence-activated cell sorting analysis of Annexin V on control platelets treated with plasma obtained from the 3 groups (n=6). (C) Fluorescence-activated cell sorting analysis of CD62P on control platelets treated with plasma obtained from the 3 groups (n=6). (D) Percentage of NET-releasing polymorphonuclear neutrophils of control polymorphonuclear neutrophils treated with platelets and plasma obtained from the 3 groups (n=6). P < 0.05 (*) and P < 0.05 (^#) compared to baseline healthy controls and the sepsis group were considered statistically significant.

Figure 6

Thrombin in plasma from ARDS is responsible for platelet activation and subsequent NET generation. (A) Thrombin levels in plasma obtained from 3 different groups as assessed by TAT complex enzyme-linked immunosorbent assay. Fluorescence-activated cell sorting analysis of Annexin V (B) and CD62P (C) on control platelets treated with plasma from ARDS patients in the presence or absence of thrombin (antithrombin III or dabigatran) and PAR-1 signaling inhibitors. Recombinant thrombin was used as a positive control (n=6). (D) Thrombin levels in control plasma incubated with isolated NET structures from control PMNs stimulated as in ARDS plasma- and ARDS plasma-treated PLTs. (E) The NET increase was tested with a myeloperoxidase-deoxyribonucleic acid complex enzyme-linked immunosorbent assay derived after stimulation of control PMNs with plasma from ARDS and control platelets pretreated with ARDS in the presence or absence of thrombin inhibitors (antithrombin III or dabigatran) (n=6). (F) Anti-tissue factor antibody was used to neutralize tissue factor-mediated thrombin generation. DNase I was used for neutrophil extracellular trap scaffold degradation (n=6). P < 0.05 (*), P < 0.05 (^#) and P < 0.01 (^&) compared to baseline healthy controls and the sepsis group or ARDS group were considered statistically significant.

Figure 7

Two-hit procedures of thrombosis formation initiated by NETs and activated platelets lead to the progression of ARDS. When sepsis-mediated ARDS occurs, neutrophils are activated and then form TF-enriched neutrophil extracellular traps in pulmonary blood vessels (first step), resulting in thrombin generation. Platelets are activated by thrombin and then interact with neutrophils, which lead to formation of TF-enriched NET (second step).Thereby creating a vicious cycle that causes massive thrombosis formation.