Sixty-three patients newly diagnosed with primary GC and 13 healthy donors (HDs) attending the Second Affiliated Hospital of Harbin Medical University between October 2019 and April 2021 were enrolled in this study. GC diagnoses were performed based on pathological examinations. Pathological tumor-node-metastasis (TNM) staging and histological classification of GC were performed according to the 7^th American Joint Committee on Cancer (AJCC) guidelines[31]. Patients who were < 18 years old or pregnant, those on antitumor or anticoagulant treatment before surgical treatment, or those with underlying complications such as endocrine, cardiovascular, hematological and other cancers or infectious disease were all excluded. The main clinical characteristics of the patients with GC and HDs are shown in Table 1. We extracted tumor and adjacent normal tissues from GC patients who consented to this study in writing. Blood samples of preoperative patients were obtained at first diagnosis before any clinical treatment, and blood samples of postoperative patients were obtained from 1 mo after surgery before adjuvant therapy to avoid errors caused by postoperative stress. The protocol for this study was approved by the Ethics Committee of the Second Affiliated Hospital of Harbin Medical University (No. KY2016-032).

Fresh whole venous blood was collected into tubes containing 3.2% sodium citrate using 21-gauge needles. The patients underwent overnight fasting before blood collection. The blood was centrifuged at 150 g for 20 min at room temperature (RT) to obtain platelet-rich plasma (PRP), within 1 h of collection. Clean top PRP layer was collected in a new tube, diluted with platelet wash buffer (TBD, Tianjin, China), and centrifuged at 460 g for 20 min at RT to obtain clean platelets and platelet-free plasma (PFP)[32]. The platelets were resuspended in prewarmed modified Tyrode’s buffer (137 mmoL/L NaCl, 2.8 mmoL/L KCl, 1.0 mmoL/L MgCl₂, 12 mmoL/L NaHCO₃, 0.4 mmoL/L Na₂HPO₄, 5.5 mmoL/L glucose, pH 7.4; Solarbio, Beijing, China). Neutrophils were isolated based on the density gradient centrifugation, using the whole blood neutrophil isolation kit (TBD). After lysing erythrocytes based on the red blood cell lysis buffer (150 mmoL/L NH₄CL, 10 mmoL/L KHCO₃, 0.2 mmoL/L Na₂EDTA, pH 7.4; TBD), the purity (> 96%) and viability (> 96%) of neutrophils were assessed using Wright–Giemsa staining and Trypan blue staining, respectively.

The human metastatic GC KATO-III and MKN-45 cell lines, human primary GC AGS cell line, human gastric mucosal epithelial cell (GES-1) line, human umbilical cord endothelial cell (HUVEC) line, and mouse forestomach squamous carcinoma (MFC) cell line were purchased from PROCELL (Wuhan, China). All cell lines were characterized using short tandem repeat (STR) profiling. The GES-1, AGS, MKN-45 and MFC cells were cultured in RPMI 1640 (Gibco, USA), KATO-III cells were cultured in IMDM (Gibco), and HUVECs were cultured in DME/F12 (HyClone, Logan, UT, USA). All media were supplemented with 10% fetal bovine serum (FBS; Gibco) and 1% penicillin–streptomycin solution (Beyotime, Beijing, China). Incubation was performed at 37°C under 5% CO₂ in a humidified environment. To prepare conditioned medium (CM), the cells were cultured to 90% confluence in medium supplemented with 10% FBS, washed three times using 1 × PBS, and recultured for 48 h in medium without FBS. The supernatant was centrifuged at 1500 g for 10 min at 4°C to remove cell debris. The CM was collected and stored at -80°C until use.

The animal protocol was designed to minimize pain or discomfort to the animals. Wild-type male C57BL/6 mice (7–9 wk old, weighing 20–26 g) were purchased from the Animal Experimental Center of Harbin Medical University, and all procedures were approved by the Animal Care and Use Committee of the Second Affiliated Hospital of Harbin Medical University (No. KY2016-032). The animals were individually housed and maintained under standard conditions (12 h/12 h light/dark cycle, 22 ± 1 °C, 50% humidity) and provided with a conventional laboratory diet and an unrestricted supply of drinking water. In the subcutaneous tumor models, mice were injected with MFC cells (2 × 10⁷ cells/mL) subcutaneously into the right axilla, and the tumor volumes were measured every 3 d from 7 d after injection with MFC cells and allowed to reach 1000 mm³ (usually 28–35 d). The volume was calculated by measuring the length (L) and width (W) of the tumor: Tumor volume = π/6 × L × W². Thereafter, a murine model of deep vein thrombosis (DVT) was developed as previously described[33]. The mice were anesthetized by intraperitoneal injection of 2,2,2-tribromoethanol (Sigma, St . Louis, MO, USA), and the intestines were removed to expose the IVC after entering the abdominal cavity through a median abdominal incision. The IVC was carefully separated below the left renal vein plane. After 5-0 (1 mm) sutures passed through the IVC, 3-0 (2 mm) sutures were placed at the parallel part of the IVC as the blocking line. The IVC was ligated and 3-0 sutures were carefully extracted. This procedure has been shown to decrease the vascular lumen by approximately 90%. The other branches of the IVC were ligated to the level of the iliac vein. Thereafter, the abdominal incision was closed, mice were sacrificed after 6 or 48 h, and thrombi that had formed in the IVC were harvested. The IVC stenosis mice in the treatment groups were injected with DNase I (50 μg/mouse; Roche, Switzerland) intraperitoneally every 12 h until the time of death. All animals were killed for tissue collection. Blood (300 μL/mouse) was drawn from the periorbital eye plexus and stabilized with 0.5 mmoL/L EDTA. Plasma was obtained as described above. Thrombi of control and tumor-bearing mice were fixed in 4% paraformaldehyde for 24 h, embedded in Optimal Cutting Temperature (OCT) compound (SAKURA, Torrance, CA, USA), and cryosectioned at a thickness of 4 μm for immunofluorescence staining.

Plasma cell-free DNA (cf-DNA), MPO-DNA and citH3-DNA complexes were quantified using capture ELISA as previously described[34]. The quantification of cf-DNA was performed using the Quant-iT PicoGreen dsDNA assay kit (Invitrogen, Carlsbad, CA, USA). For detection of NET-DNA complexes, 5 μg/mL anti-MPO (ab90810; Abcam, Cambridge, UK) or anti-citH3 antibody (ab5103; Abcam) was coated onto 96-well plates overnight at 4°C. After blocking in 1% bovine serum albumin (BSA), the plasma from GC patients or healthy individuals was added per well and incubated at RT for 2 h. After washing five times with PBST, Quant-iT PicoGreen dsDNA Reagent was added. The values were then read with a fluorometer with a filter setting of 480 nm/520 nm excitation/emission wavelengths.

Fibrin formation of platelets and ECs was detected by turbidity as previously described[35]. To assess the fibrin formation in the setting of platelet or EC monolayers, cell monolayers in 96-well plates were stimulated by NETs with or without DNase I, activated protein C (APC), and sivelestat treatment alone or together for 4 h, followed by two washes with Hank’s balanced salt solution (137 mmoL/L NaCl, 5.3 mmoL/L KCl, 4.17 mmoL/L NaHCO₃, 0.33 mmoL/L Na₂HPO₄, 5.5 mmoL/L glucose, pH 7.4) and then 150 μL PFP from HDs was cocultured with cells at 37°C for 2 min, followed by the addition of 50 μL prewarmed 25 mmoL/L CaCl₂. The fibrin formation was tested using turbidity measurements and quantified as the maximum value by measuring the OD at 405 nm every 10 s for 30 min on a SpectraMax 340 PC plate reader. To assess the fibrin formation in the plasma of mice, 150 μL PFP from control, tumor-bearing mice, or DNase I infused tumor-bearing mice was cultured at 37°C for 2 min, followed by addition of 50 μL prewarmed 25 mmoL/L CaCl₂. To detect the level of TAT complex, a human and mouse TAT complex ELISA kit (Jingkbio, Shanghai, China) was used as previously described[36].

Circulating NETs were measured using flow cytometry. Here, whole blood from HDs and GC patients was diluted with 1 × PBS and incubated in the dark at RT for 30 min with FITC-conjugated-citH3 (eBioscience, San Diego, CA, USA) and PE-conjugated MPO (eBioscience) antibodies. For the assessment of phosphatidylserine (PS) and P-selectin expression on platelets, platelets (2 × 10⁶ cells) isolated from the blood of HDs and GC patients were incubated with FITC-conjugated lactadherin (Haematologic, Essex Junction, VT, USA), APC-conjugated CD62P (Biolegend, San Diego, CA, USA), and PerCP-conjugated CD41 (Biolegend) antibodies.

Tumor and paratumor tissues of GC patients were fixed in 4% paraformaldehyde for 24 h, embedded in OCT compound (SAKURA), and cryosectioned into slices of 4-μm thickness. The samples were cultured overnight at 4°C with primary rabbit anti-histone H3 (1:500, ab5103; Abcam) and mouse anti-MPO (1:500, ab90810; Abcam) antibodies, and washed three times with PBS before incubation for 1 h at RT with Alexa Fluor 594-conjugated goat anti-rabbit (1:200; Proteintech, China) and Alexa Fluor 488-conjugated goat anti-mouse (1:200; Proteintech) secondary antibodies. The tissues were stained for 5 min at RT in the dark with 4’,6-diamidino-2-phenylindole (DAPI) and anti-fade mounting medium (Solarbio, Beijing, China). Thrombi in the IVC of tumor models or control mice were stained with primary rabbit anti-histone H3 (1:500, ab5103; Abcam) and rat anti-Ly6G (1:500, Novus, St. Charles, MO, USA), and the specimens were incubated with the Alexa Fluor 488-conjugated goat anti-rabbit (1:200; Proteintech) and Alexa Fluor 594-conjugated goat anti-rat (1:200; Proteintech) secondary antibodies as previously described[37]. The tissue images were captured using a confocal microscope (LSM 800; Zeiss, Germany).

Neutrophils (5 × 10⁵ cells) isolated from HDs were seeded and incubated in glass-based poly-L-lysine-coated 24-well plates for 1 h at 37°C under 5% CO₂. Thereafter, cell suspensions of KATO-Ⅲ, MKN-45, AGS, and GES-1 (2 × 10⁵ cells) or CM from GC cells were cocultured with neutrophils for 4 h at 37°C under 5% CO₂. To detect and quantify NETs, the samples of the CM group were incubated with primary rabbit anti-histone H3 and mouse anti-MPO antibodies and then fluorescent secondary antibodies. For the samples of cell-cell contact groups, NETs were stained with Sytox Green (Solarbio) for 10 min in the dark at RT. All images were captured using a confocal microscope (LSM 800; Zeiss) and analyzed with ImageJ software (National Institutes of Health, Bethesda, MD, USA).

Cell-free NETs were isolated from neutrophils of GC patients as previously described, with modifications[38]. Neutrophils (10⁷ cells/mL) were cultured for 4 h at 37°C under 5% CO₂ in medium supplemented with 500 nM PMA (HY-18739; MedChemExpress, Monmouth Junction, NJ, USA). The supernatant was discarded, and ice-cold 1 × PBS was added to wash down the cell layer of neutrophils to obtain the NET medium and centrifuged at 1500 g for 10 min at 4°C to remove cell debris. Thereafter, 1.5 mL supernatant (sterile DNA–protein complex) was centrifuged at 15 000 g for 15 min at 4°C. The resultant pellets were suspended in ice-cold 1 × PBS, followed by DNA concentration measurement in the medium obtained using spectrophotometry (Biospec-nano, Japan). An adequate DNA concentration in the medium should range between 50 and 100 μg/mL. The medium containing the NETs was stored at -80°C for subsequent experiments.

Platelet activation and adhesion assays were performed as previously described[22]. Glass-based wells of 24-well plates were coated with cell-free NETs after overnight incubation in the corresponding medium at 4°C in a humidified chamber. For controls, 1% denatured BSA was used. Denatured BSA was prepared by heating the solution in PBS without calcium or magnesium to 80°C for 3 min and immediately placing it on ice until cool; it was then stored at 20°C until use. Platelet suspensions (10⁷ cells/mL) were then seeded in the wells, cultured for 1 h at 37°C under 5% CO₂, fixed for 15 min at RT with 4% paraformaldehyde, and washed three times using 1 × PBS before 20 min permeabilization using 0.1% Triton-X 100. The platelets were then incubated for 30 min with Alexa Fluor 594-conjugated phalloidin primary antibody (1:300; Thermo Fisher, Waltham, MA, USA). To assess PS and P-selectin expression, the platelet suspension was incubated for 1 h at 37°C under 5% CO₂, and the cells were stimulated with NETs for 1 h, followed by FITC-conjugated lactadherin (Hematologic) staining for 30 min. Platelets were washed twice with PBS before fixation for 15 min at RT with 4% paraformaldehyde, permeabilized using 0.1% Triton-X 100 for 20 min, and then stained with primary rabbit anti-P-selectin (1:200; Proteintech) and mouse anti-CD41 (1:500; Novus) antibodies. Images were captured using a confocal microscope (LSM 800; Zeiss) and analyzed with ImageJ software (National Institutes of Health).

HUVECs were incubated with cell-free NETs or PBS in 24-well plates for 4 h when the cells grew to a confluent monolayer. The cells were fixed in 4% paraformaldehyde for 15 min at RT, washed three times using 1 × PBS, and blocked for 1 h using 10% goat serum with 1% BSA solution in PBS. For detection of tissue factor (TF) expression, ECs were incubated overnight at 4 °C with rabbit anti-TF (1:500, ab228968; Abcam) and mouse anti-CD31 (1:500, ab9498; Abcam) primary antibodies. The cells were washed with PBS and reincubated for 1 h at RT with Alexa Fluor 594-conjugated (Proteintech) goat anti-rabbit and Alexa Fluor 488-conjugated (Proteintech) goat anti-mouse secondary antibodies. For detection of intercellular junctions of cells, cells were incubated overnight at 4°C with rabbit anti-VE-cadherin (1:500, ab33168; Abcam) primary antibody, followed by Alexa Fluor 488-conjugated (Proteintech) goat anti-rabbit secondary antibody, and were further incubated with Alexa Fluor 594-conjugated phalloidin primary antibody (1:300; Thermo Fisher, Waltham, MA, USA). They were stained with DAPI and fixed with mounting medium (Solarbio) for 5 min at RT in the dark. The cells were observed and photographed using a confocal microscope. The photos were analyzed with ImageJ software.

For inhibition assays, platelets and HUVECs were cocultured with cell-free NETs for 1 h at 37°C in a humidifier chamber in the presence of DNase I (100 U/mL, Roche), APC (100 nM, HY-P1918; MedChemExpress), and sivelestat (100 nM, HY-17443; MedChemExpress) alone or together. DNase I cleaves NET DNA, whereas APC and sivelestat disrupt histones and NE functions, respectively.

Data are expressed as the mean ± SD. Data were analyzed to assess distribution normality. For normally distributed data, statistical significance was analyzed using Student’s t-tests and one-way analysis of variance. For non-normally distributed data, statistical significance was analyzed using the Mann–Whitney test and Kruskal–Wallis test. All analyses were performed using GraphPad Prism version 8.0 and SPSS 16.0 statistical software. Spearman’s correlation was used to evaluate the association between two variables. P < 0.05 was considered statistically significant.

The levels of plasma cf-DNA, citH3-DNA,and MPO-DNA complexes in GC patients (n = 63) and HDs (n = 13), which reflect the concentration of NETs, were measured using capture ELISA. The levels of NET markers were significantly higher in patients with stage II/III/IV GC than in HDs (Figure 1A–C). There was also a significant difference in preoperative and postoperative plasma NET marker levels in GC patients (Figure 1D–F). The levels of NET markers positively correlated with those of serum D-dimer (cf-DNA: r = 0.5595, P < 0.0001; citH3-DNA: r = 0.5469, P < 0.0001; MPO-DNA: r = 0.5479, P < 0.0001), suggesting that NETs were associated with hypercoagulation and VTE development in GC patients. The levels of NETs (MPO⁺/citH3⁺ neutrophils) in the circulation in GC patients and HDs were measured using flow cytometry. Circulating NETs were higher in the blood of patients with either GC stage (II–IV) than in their HD counterparts (Figure 1G and H). Furthermore, based on MPO and citH3 levels, immunofluorescence staining revealed that NETs were significantly higher in the tumor microenvironment than in the paratumor tissue of the same patients (Figure 1I–M).

Open in New Tab   Full Size Figure   Download Figure

Figure 1 NETs are accumulated in samples of GC patients. A–C: Plasma levels of NET markers cf-DNA, citH3-DNA, and MPO-DNA in GC patients and healthy individuals were measured by ELISA. Healthy individuals, n = 13; stage I, n = 14; II, n = 15; III, n = 22; IV, n = 12; D-F: Comparison of cf-DNA, citH3-DNA and MPO-DNA in the plasma of patients with GC preoperatively and postoperatively by ELISA. n = 51; G and H: The rate of activated neutrophils in the circulating environment of GC patients and healthy individuals was measured by flow cytometry with APC-MPO and FITC-citH3 staining. Each group: n = 6; I and J: NET accumulation was detected by confocal microscopy with MPO and citH3 staining in paratumor and tumor samples from the same GC patient. Magnification 20×; scale bars: 50 μm. Red-citH3, Green-MPO and Blue-DAPI; K and L: Magnified (40×) part of MPO and citH3 colocation in paratumor and tumor samples from the same patient. Scale bars: 10 μm. Red-citH3, Green-MPO and Blue-DAPI; M: The percentage of area coverage of citH3 expression was defined as the rate of red area in total area and analyzed with ImageJ software; all values are the mean ± SD. ^aP < 0.05; ^bP < 0.01; ^cP < 0.001; ^dP < 0.0001. NET: Neutrophil extracellular trap; GC: Gastric cancer; cf-DNA: Cell-free DNA; MPO: Myeloperoxidase; citH3: Citrullinated histone H3.

To assess whether GC cells directly stimulated NET formation, we analyzed the expression of NETs in a coculture of GC cell lines and neutrophils. Immunofluorescence analysis revealed that compared with GES-1 cells, the rate of NET formation was significantly higher in GC cells (Figure 2A). However, the formation of NETs was greater in the metastatic GC cell line than in the nonmetastatic GC cell line (Figure 2B). We measured NET formation when normal neutrophils were cocultured with CM from KATO-III, MKN-45, AGS and GES-1 cells. Immunofluorescence analysis further revealed that the CM of KATO-III and MKN-45 cells exerted greater neutrophil activation for NET formation than the CM of AGS cells. However, the CM of GES-1 cells had no effect on NET formation (Figure 2C and D). Overall, these findings demonstrate that GC cells stimulate NET formation through both intercellular contact and noncontact mechanisms.

Open in New Tab   Full Size Figure   Download Figure

Figure 2 GC cells can stimulate neutrophils to form NETs. A and B: Control neutrophils were cocultured with normal gastric mucosal epithelial cells (GES-1) or GC cells (AGS, MKN-45 and KATO-III) and NET formation was measured by confocal microscopy with cell-impermeable Sytox-Green staining. Magnification 20×; scale bars: 50 μm. Green: Neutrophils; C and D: Control neutrophils were cocultured with CM from GES-1 or GC cells, and stained with MPO and citH3. The percentage of NET-releasing cells was defined as the ratio of the calculated NET releasing neutrophils to the total number of neutrophils. Magnification 20×; scale bars: 50 μm. Red-citH3, Green-MPO, and Blue-DAPI. All values are the mean ± SD. ^aP < 0.05; ^bP < 0.01; ^cP < 0.001; ^dP < 0.0001. GC: Gastric cancer; GES-1: Gastric mucosal epithelial cells; NET: Neutrophil extracellular trap; CM: Conditioned medium; MPO: Myeloperoxidase; citH3: Citrullinated histone H3.

To examine the effect of NETs on platelet activation, we measured the levels of PS and P-selectin expression in these cells. Flow cytometry revealed that compared to HDs, the expression of PS and P-selectin was significantly higher on platelets of patients with GC (Figure 3). In addition, platelets isolated from HDs were cocultured with NET medium or PBS before analyzing PS and P-selectin expression. We found that NETs stimulated PS and P-selectin expression on platelets by confocal microscopy (Figure 4A–C). Flow cytometry also demonstrated this hypercoagulable phenotype of platelets which was stimulated by NETs (Figure 4D and E). In the inhibition assay, we added DNase I, APC and sivelestat alone or together to cleave DNA, histones and NE, which were the most functional factors of NETs. At the highest concentration of NETs (0.5 μg DNA/mL), flow cytometry revealed that DNase I, APC, sivelestat, or a combination of the three inhibited 60.2%, 47.1%, 41.9% and 83.2% of PS expression, respectively (Figure 4D), and 55.2%, 47.0%, 41.0%, and 91.76% of P-selectin expression on platelets, respectively (Figure 4E).

Open in New Tab   Full Size Figure   Download Figure

Figure 3 Platelets are activated in patients with GC. A and C: Flow cytometry showed the rate of P-selectin-positive platelets from each stage GC patients and healthy individuals; B and D: Rate of phosphatidylserine-positive platelets from each stage GC patients and healthy individuals. All values are the mean ± SD. ^dP < 0.0001. PS: Phosphatidylserine; GC: Gastric cancer. CD62P: P-selectin; Lact: Lactadherin.

Open in New Tab   Full Size Figure   Download Figure

Figure 4 NETs contribute to hypercoagulation of platelets. A: PS exposure and P-selectin expression were measured when isolated platelets were cocultured with BETs (μg, DNA/mL) or in the presence of DNase I, activated protein C, and sivelestat alone or together by confocal microscopy. Magnification 63×; scale bars: 10 μm. Red-platelets, Green-Lactadherin, and Blue-P-selectin; B and C: PS exposure and P-selectin expression are indicated as MFI. MFI was defined as the ratio of total fluorescence intensity to the area; D and E: The rates of PS-positive platelets and P-selectin-positive platelets were detected by flow cytometry. All values are the mean ± SD. ^cP < 0.001; ^dP < 0.0001. PS: Phosphatidylserine; GC: Gastric cancer; DNase I: Deoxyribonuclease I; APC: Activated protein C; MFI: Mean fluorescence intensity; Lact: Lactadherin; CD41: Platelet; NET: Neutrophil extracellular trap.

Previous studies have shown that NETs promote thrombosis in murine late-stage breast cancer models and DVT models[19,39]. However, whether NETs derived from GC neutrophils have the ability to stimulate platelet adhesion under static conditions is unknown. To determine the effect of NETs on platelet adhesion, platelets isolated from HDs were seeded in NET-coated wells to measure the effects of NETs on the adherence of platelets to blood vessels. Confocal microscopy revealed that NETs enhanced adherence of platelets to glass slides (Figure 5A and B), indicating that NETs induce the development of thrombosis. Moreover, the results revealed that fibrin formation and TAT complex levels were increased when control plasma was cocultured with platelets activated by NETs (Figure 5C and D).

Open in New Tab   Full Size Figure   Download Figure

Figure 5 NETs promote platelet adhesion and prothrombotic state. A: Isolated platelets were incubated on glass slides which were coated with 1% dBSA, NETs (μg DNA/mL), or NETs pretreated with DNase I, APC, and sivelestat alone or together, followed by the F-actin components of platelets with 594-phalloidin staining. Magnification 63×; scale bars: 10 μm. Red-platelets; B: The percentage of area coverage of platelet adhesion was defined as the rate of red area in the total area and analyzed with ImageJ software; C: Isolated platelets were cocultured with different concentrations of NETs for 30 min with or without DNase I, APC and sivelestat treatment alone or together, and plasma fibrin formation was tested using turbidity measurements and monitored OD at 405 nm; D: TAT complex level of activated platelets was analyzed by ELISA. All values are the mean ± SD. ^aP < 0.05; ^bP < 0.01; ^cP < 0.001; ^dP < 0.0001. dBSA: Denatured bovine serum albumin; NETs: Neutrophil extracellular traps; DNase I: Deoxyribonuclease I; APC: Activated protein C; TAT: Thrombin–antithrombin.

Inhibition assays revealed that digestion of NET DNA using DNase I modulated the adhesion of platelets on glass surfaces. Even so, a few platelets still adhered to the NET-coated well pretreated with DNase I (Figure 5A). This suggested that other protein components other than NET-DNA participated in the adhesion of platelets. NETs treated with DNase I, APC, sivelestat, or a combination of the three inhibited 67.1%, 56.6%, 38.9%, and 91.8% of platelet adhesion, respectively (Figure 5B). The degree of platelet adhesion in the combination group was comparable to that of the controls. We found that DNase I, APC, sivelestat, or a combination of the three reduced fibrin formation by 81.1%, 73.9%, 64.3%, and 90.7%, respectively (Figure 5C), and inhibited 78.9%, 57.4%, 51.2%, and 91.9% of TAT complex level, respectively (Figure 5D), at the highest concentration of NETs. Taken together, these findings demonstrate that NETs play a role in the development of thrombosis.

To detect the effect of NETs on EC thrombogenicity, HUVECs were cocultured with NET medium. Confocal microscopy revealed that NETs destroyed the normal intercellular junctions between ECs (Figure 6A and C). NET treatment upregulated expression of TF on the surface membrane of ECs (Figure 6B and D). In addition, plasma fibrin formation and TAT complex levels were significantly increased when control plasma was incubated with EC monolayers activated by NETs (Figure 6E and F). Further inhibition assays were performed to assess the effect of NETs on ECs after pretreatment with DNase I, APC, sivelestat, or a combination of the three. We found that NET treatment after incubation with DNase I, APC, sivelestat, or all three drugs returned 47.5%, 36.3%, 33.4%, and 86.5%, respectively, of VE expression on ECs (Figure 6C), whereas similar treatment inhibited 60.4%, 44.8%, 38.6%, and 95.5%, respectively, of TF expression (Figure 6D). Additionally, we found that NET treatment after incubation with the above inhibitors inhibited 59.7%, 54.4%, 54.0%, and 91.5% of fibrin formation level, respectively (Figure 6E), and inhibited 51.2%, 35.6%, 25.9%, and 84.3% of TAT complex level, respectively (Figure 6F). Taken together, these findings suggest that NETs promote hypercoagulation of ECs; thus, inhibiting NET function can protect against venous injury.

Open in New Tab   Full Size Figure   Download Figure

Figure 6 NETs drive hypercoagulation of ECs. A: ECs were cocultured with NETs (μg DNA/mL) or PBS in the presence of DNase I, APC, or sivelestat alone or together for 4 h and analyzed by confocal microscopy. The intercellular junctions of ECs were stained with VE–cadherin and phalloidin. Magnification 63×; scale bars: 10 μm. Red-phalloidin, Green-VE, and Blue-DAPI; B: EC activation was stained with CD31 and TF. Magnification 63×; scale bars: 10 μm. Red-TF, Green-CD31, and Blue-DAPI; C and D: VE–cadherin expression and TF expression on ECs were detected by confocal microscopy and analyzed with ImageJ software (expression indicated as MFI). MFI was defined as the ratio of total fluorescence intensity to the area; E and F: EC monolayers were stimulated with various concentrations of NETs for 4 h, followed by determination of fibrin formation by turbidity measurement at 405 nm, and the TAT complex level was detected by ELISA. All values are the mean ± SD. ^aP < 0.05; ^bP < 0.01; ^cP < 0.001; ^dP < 0.0001. NETs: Neutrophil extracellular traps; ECs: Endothelial cells; DNase I: Deoxyribonuclease I; APC: Activated protein C; TF: Tissue factor; TAT: Thrombin–antithrombin; MFI: Mean fluorescence intensity.

Based on these findings in vitro and the pivotal role of NETs in thrombosis, we hypothesized that GC-induced NETs can also promote thrombosis in vivo. Here, in a mouse IVC flow stenosis model, tumor-bearing mice demonstrated more capacity to form thrombi and showed heavier weight and longer length of thrombi compared to control mice (Figure 7A–D). In the 6 h models, three of nine of control mice showed thrombi, whereas seven of nine of tumor-bearing mice formed thrombi. In the 48 h models, all mice demonstrated thrombi in the IVC. In addition, confocal images of thrombi formed in the tumor-bearing mice after 48 h of IVC stenosis showed that NETs were significantly accumulated compared to control mice (Figure 7E–H). The thrombi of control mice included some neutrophils (Ly6G+) but were not activated to form NETs (Figure 7E and G). Furthermore, tumor-bearing mice showed higher fibrin formation and TAT complex levels than control mice (Figure 7I and J).

Open in New Tab   Full Size Figure   Download Figure

Figure 7 Tumor-bearing mice show a greater ability to form thrombi by inferior vena cava flow restriction. A and B: The values for weight and length of thrombi present in control, tumor-bearing mice, DNase I infused tumor-bearing mice at 6 h after surgery. Each group, n = 9; C and D: Values for weight and length of thrombi present in mice at 48 h after surgery. Each group, n = 5; E and F: Confocal imaging of thrombi derived from control mice and tumor-bearing mice with Ly6G and citH3 staining. Magnification 10×; scale bars: 200 μm. Red-Ly6G, Green-citH3, and Blue-DAPI; G and H: Magnified (40×) part of thrombi derived from control mice and tumor-bearing mice. Scale bars: 50 μm. Red-Ly6G, Green-citH3, and Blue-DAPI; I and J: Fibrin formation levels in the plasma of control, tumor-bearing mice, or DNase-I-infused tumor-bearing mice were detected by turbidity measurement at 405 nm, and TAT complex levels were detected by ELISA. Each group, n = 8. All values are the mean ± SD. ^aP < 0.05; ^bP < 0.01; ^cP < 0.001; ^dP < 0.0001. DNase I: Deoxyribonuclease I; TAT: Thrombin–antithrombin.

In the inhibition assay, we infused DNase I into mice immediately after IVC stenosis and examined thrombosis after 6 or 48 h of surgery. We found that treatment with DNase I significantly inhibited thrombi formation in tumor-bearing IVC stenosis mice (Figure 7A–D). Furthermore, fibrin formation and TAT complex levels in tumor-bearing mice were significantly decreased by DNase I treatment (Figure 7I and J). These data suggest that NETs play a role in thrombosis in vivo, which was induced by GC, and inhibiting NETs by DNase I had a protective effect on thrombosis in this mouse model.

The development of venous thromboembolism (VTE) is associated with high mortality among gastric cancer (GC) patients. Neutrophil extracellular traps (NETs) have been reported to correlate with the prothrombotic state in some diseases, but it was rarely reported in GC patients.

Cancer cells exert a procoagulant activity (PCA) in their microenvironment, which is related to activation of the coagulation system. However, the molecular mechanism underlying PCA in GC patients is poorly understood.

The present study aimed to investigate the effect of NETs on the development of cancer-associated thrombosis in GC patients.

The levels of NETs in blood and tissue samples of patients were analyzed by ELISA, flow cytometry, and immunofluorescence staining. NET generation and hypercoagulation of platelets and endothelial cells (ECs) in vitro were observed by immunofluorescence staining. NET PCA was determined by fibrin formation and thrombin–antithrombin complex assays. Thrombosis in vivo was measured in a murine model induced by flow stenosis in the inferior vena cava (IVC).

NETs were likely to form in blood and tissue samples of GC patients compared with healthy individuals. In vitro studies showed that GC cells and their conditioned medium, but not gastric mucosal epithelial cells, can stimulate NET release from neutrophils. Furthermore, NETs induced hypercoagulable state of platelets and ECs. In a model of IVC stenosis, tumor-bearing mice showed a stronger ability to form thrombi, and NETs abundantly accumulated in the thrombi of tumor-bearing mice compared with control mice. Notably, the combination of deoxyribonuclease, activated protein C, and sivelestat markedly abolished the PCA of NETs.

Our findings demonstrate that GC-induced NETs strongly increase the risk of VTE development both in vitro and in vivo. NETs are potential therapeutic targets in the prevention and treatment of VTE in GC patients.

The treatment strategies can consider the combination of traditional anticoagulant drugs and NETs inhibiting drugs, so as to reduce the risk of cancer associated thrombosis in patients with GC and improve the clinical treatment effect.