Procoagulant role of neutrophil extracellular traps in patients with gastric cancer

Abstract

Background: Patients with gastric cancer (GC) commonly exhibit a hypercoagulable state that results in significant morbidity and mortality. Recent studies have shown that neutrophil extracellular traps (NETs) trigger coagulation through an intrinsic pathway and contribute to thrombus initiation and progression. In this study, we aimed to determine the procoagulant activity (PCA) of NETs in patients with GC.

Methods: NET formation and their PCAs were assessed in 48 patients with GC and 36 healthy controls using immunofluorescence microscopy of neutrophil markers and extracellular DNA as well as a modified capture ELISA technique, and thrombin-antithrombin complex and clot (fibrin) spectroscopic detection, respectively.

Results: Here we showed that neutrophils isolated from patients with GC displayed significantly enhanced NET formation compared with those from healthy controls; furthermore, plasma or platelets obtained from patients with GC induced control neutrophils to release NETs. In addition, NETs released by GC neutrophils significantly increased the potency of control plasma to generate thrombin and fibrin. Notably, these procoagulant effects were dramatically attenuated by application of DNase I. We further found that spontaneous NET formation in patients with GC was significantly higher than that in controls, increased with tumor- node-metastasis stage elevation, and positively correlated with thrombin-antithrombin complex levels and D-dimers. Additionally, the effect of DNase I on cell-free plasma generation of fibrin was dependent on the concentration of NET formation.

Conclusion: These results suggest that GC creates a systemic environment that primes neutrophils to release procoagulant NETs. Thus, targeting NETs might improve the coagulopathy of patients with GC.

Keywords: NETs; Stomach neoplasm; cell-free DNA; neutrophils; prothrombotic state.

Figure 1

Enhanced neutrophil extracellular trap (NET) formation in neutrophils derived from patients with GC. (A) and (B). Representative microphotographs displaying NETs of neutrophils obtained from healthy controls and from patients with GC. Magnification, × 200. Scale bars, 20 µm. (C) The percentage of extracellular trap-releasing neutrophils and (D) extracellular DNA levels significantly increase in patients with GC compared with healthy controls. *indicates P < 0.001. Results are expressed as means ± standard deviation. CTR, healthy control (n = 36); GC, gastric cancer (n = 48), MPO, myeloperoxidase; CFDNA, cell-free DNA.

Figure 2

The microenvironment in patients with GC primes control neutrophils to release neutrophil extracellular traps. A. Representative microphotographs showing neutrophil extracellular trap generation in control neutrophils treated with plasma and platelets obtained from control individuals or patients with GC, respectively. Original magnification: × 200. Scale bar: 20 μm. B-E. Percentage of extracellular trap-releasing neutrophils and extracellular DNA levels were respectively used for illustrating that the potency of plasma and platelets derived from patients with GC to induce control neutrophils (n = 5) to generate NETs were significantly higher than those from healthy controls. *P < 0.001. Results are expressed as means ± standard deviation. Plasma was respectively derived from 48 patients with GC and 36 healthy controls and platelets were respectively derived from 10 patients and 10 healthy controls. CTR, healthy control; GC, gastric cancer; CFDNA, cell-free DNA.

Figure 3

Procoagulant activity of NETs released by neutrophils derived from patients with GC. A. The inclusion of 20% NETs released by GC neutrophils significantly increased the amount of control plasma-generated TAT complex after recalcification. B, C. The inclusion of 20% NETs released by GC neutrophils significantly shortened the time to peak and increased the peak turbidity of control plasma generation of fibrin. D-F. DNase I significantly reduced the effect of NETs released by neutrophils derived from patients with GC on control plasma generation of thrombin and fibrin. *P < 0.001. Results are expressed as means ± standard deviation. CTR plasma, plasma derived from healthy controls (n = 5); CTR NETs, NETs released by healthy control neutrophils (n = 36); GC NETs, NETs released by neutrophils derived from patients with gastric cancer (n = 48); FGT, fibrin generation test.

Figure 4

Significant increase of NET formation in patients with GC. Myeloperoxidase-deoxyribonucleic acid complex (MPO-DNA), cell-free nucleosomes, and neutrophil elastase were measured with ELISA using plasma samples from healthy donors (CTR, n = 36), individuals with GC including patients in stage I (S-I, n = 7), stage II (S-II, n = 9), stage III (S-III, n = 25) and stage IV (S-IV, n = 7) cancer. Plasma CFDNA was quantified with fluorescent quantification. (A) MPO-DNA complex, (B) CFDNA, (C) nucleosomes, and (D) NE in patients with GC were increased with disease progression. Data are expressed as means ± standard deviation. *P < 0.001. GC, gastric cancer; OD, optical density; MPO, myeloperoxidase; CFDNA, cell-free DNA; NE, neutrophil elastase.

Figure 5

DNase I significantly inhibits fibrin generation in autonomous plasma from patients with stage III/IV GC. Autonomous plasma was incubated without or with DNase I, and fibrin formation was continually monitored after addition of CaCl₂. A, B. The time to peak and peak turbidity of autonomous plasma derived from stage III/IV patients were significantly shorter and higher, respectively, than those from healthy controls. C, D. DNase I significantly prolonged the time to peak and reduced peak turbidity in patients with stage III/IV cancer. Each panel consists of the healthy controls (CTR, n = 36), patients in stage I (S-I, n = 7), stage II (S-II n = 9), stage III (S-III, n = 25) and stage IV (S-IV, n = 7) cancer. Data are represented as means ± SD. *P < 0.001. GC, gastric cancer; FGT, fibrin generation test.