The mechanism of S100A8/A9 heterodimer promoting atherosclerotic thrombosis

Abstract

S100A8/A9 heterodimer is an inflammatory mediator that promotes atherosclerotic thrombosis (AT) and can also be used as a biomarker of the disease. However, its origin and mechanism of action remain largely unknown. This study aimed to elucidate the role of S100A8/A9 heterodimer produced by endothelial cells stimulated by neutrophil extracellular traps (NETs) in AT. The effects of NETs on human umbilical vein endothelial cells (HUVECs) were studied in a co-culture system. We detected the reactive oxygen species (ROS) activity, pyroptosis and S100A8/A9 heterodimer production of HUVECs, as well as the NF-κB signaling pathway activation. The effects of S100A8/A9 heterodimer on platelet activation and underlying mechanisms were investigated in washed human platelets, and then validated in FeCl₃-injured carotid thrombosis model in rats. NETs stimulated HUVECs were found to produce ROS, then activating the NF-κB signaling pathway, inducing pyroptosis and the production of S100A8/A9 heterodimer. Additionally, S100A8/A9 heterodimer was found to induce the activation of platelets through activation of epidermal growth factor receptor (EGFR) signaling pathway. Subsequently, S100A8/A9 heterodimer mediated platelet activation thus inducing thrombus formation via the EGFR/PI3K/AKT and EGFR/p38 MAPK axes in platelets. Our results suggest that S100A8/A9 heterodimer produced by endothelial cells stimulated by NETs induces platelet activation by regulating the EGFR/PI3K/AKT and EGFR/p38 MAPK axes, thus promoting thrombosis. Our data provide new insights into the pathogenesis and therapeutic targets for AT.

Supplementary Information

The online version contains supplementary material available at 10.1007/s10238-025-02016-z.

Introduction

Various arterial thrombotic diseases, such as acute coronary syndrome and ischemic stroke, are mainly caused by atherosclerotic thrombosis (AT) [1]. AT is characterized by unpredictable rupture or corrosion of atheromatous plaques, leading to thrombosis [2]. Vascular endothelial cell dysfunction, platelet activation and infiltration at the site of vascular injury, and the resulting occlusive thrombus formation are key steps in AT [3]. To date, the mechanisms by which platelets become overactive in atherosclerosis (AS) are still not fully understood. Since current anti-platelet strategies also affect hemostasis, thus limiting their clinical application [4, 5]. Therefore, it is essential to clarify the mechanisms that promote platelet activation in AS and develop new treatments.

AS is a chronic sterile inflammatory disease of the arterial wall, the development of which is associated with innate and adaptive immune disorders [6]. Neutrophils are the most abundant innate immune cells that may aggravate AS [7]. Neutrophils are extremely rich in proteins and cytokines on their cell membranes or inside cells, and they respond rapidly to various stimuli by releasing neutrophil extracellular traps (NETs) through activation mediated by receptors [8]. Histones and DNA are the main components of NETs scaffolds, and a variety of enzymes are attached to the scaffolds, including cathepsin G, neutrophil elastase (NE) and myeloperoxidase (MPO) [9]. NETs have the function of inducing vascular endothelial damage [10], which may be the initial factor of AT. Numerous studies have shown that NETs can promote the development of AT [11, 12]. However, the mechanism by which NETs affect endothelial cells and then promote thrombosis is still unclear.

S100A8 and S100A9 belong to a class of damage-related pattern molecules that are mainly secreted by immune cells activated by damaged or necrotic tissue [13]. S100A8/A9 heterodimer is a valuable biomarker for inflammatory diseases [14, 15], and S100A8/A9 heterodimer is an early marker for detecting acute coronary syndrome [16]. Extracellularly, S100A8/A9 heterodimer can exert immunomodulatory functions by binding to receptors [17, 18]. There are many receptors expressed on the platelet including EGFR [19], and we found that there is an interaction between S100A9 and EGFR. Therefore, we propose for the first time that S100A8/A9 heterodimer may interact with the EGFR receptor on platelets, which subsequently activates downstream signaling pathways and promotes platelet activation. Through experimental investigations, we elucidated that the endothelial cells stimulated by NETs may be the cellular origin of S100A8/A9 heterodimer, and S100A8/A9 heterodimer can induce platelet activation through the EGFR/PI3K/AKT and EGFR/p38 MAPK axes, thus promoting thrombosis. Our data provide new insights into the pathogenesis and therapeutic targets for AT.

Materials and methods

Cell culture

Fresh blood was drawn via venipuncture from healthy volunteers (male and female, 20–30 years old). Ethical approval was from the ethical committees of Harbin Medical University (Number: YJSKY2022-394). Neutrophils were isolated as described in our previous study [20]. Isolated neutrophils were cultured in RPMI 1640 medium (Sigma, Saint Louis, MO, USA, Cat: R8758) with or without 10% fetal bovine serum (FBS, BD Biosciences, San Jose, CA, USA, Cat:11011-8611). Platelets were isolated as described in our previous study [21]. Isolated platelets were resuspended in Tyrode`s buffer (Solarbio, Beijing, China, Cat: T1422).

The HUVEC cell line was purchased from the Chinese academy of sciences cell bank. Cells were grown in DMEM/F12 medium (Sigma, Saint Louis, MO, USA, Cat: D0697) supplemented with 10% FBS and 1% antibiotics (Sigma, Saint Louis, MO, USA, Cat: V900929) at 37 °C in 5% CO₂ cell incubator.

Immunofluorescence (IF) assays

1 × 10⁵ treated or untreated HUVECs were seeded on poly-L-lysine (Sigma, Saint Louis, MO, USA, Cat: P3150) coated coverslips in a 24-well plate, with approximately 60% confluence at the start of the experiment. Then the cells were fixed with 4% paraformaldehyde (Solarbio, Beijing, China, Cat: P1110), permeabilized with 0.5% Triton X-100 (Solarbio, Beijing, China, Cat: T8200), blocked with 50% goat serum (ZSGB-BIO, Beijing, China, Cat: ZLI-9056) and incubated with primary antibodies against NE (1:100, Abcam, Cambridge, Cambridgeshire, United Kingdom, Cat: ab14188), NF-κB p65 (1:100, Affinity Biosciences, Cincinnati, OH, USA, Cat: AF2006), GSDMD (1:100, Affinity Biosciences, Cincinnati, OH, USA, Cat: AF4012) or IL-1β (1:100, Affinity Biosciences, Cincinnati, OH, USA, Cat: AF4006) at 4 °C overnight. Followed by Alexa Fluor 594 (1:200, Immunoway, Plano, TX, USA, Cat: RS23410) or 488 conjugated secondary antibodies (1:200, Immunoway, Plano, TX, USA, Cat: RS23240). To label DNA, cells were counterstained with DAPI (5 µg/mL, Invitrogen, Carlsbad, CA, USA, Cat: 62248). The images were captured from a laser confocal microscope (LSCM, Carl Zeiss AG, Oberkochen, Badenburg, Germany) or a fluorescence microscope (ECLIPSE Ti, Nikon, Tokyo, Japan).

The thrombi were fixed in 4% paraformaldehyde immediately after removing and embedded in paraffin. 4 μm paraffin-embedded tissue sections were deparaffinized, rehydrated, and processed for antigen retrieval. The sections were then incubated with primary antibody against GPIbα (1:100, Affinity Biosciences, Cincinnati, OH, USA. Cat: DF8519) followed by Alexa Fluor 488 conjugated secondary antibody. The sections were counterstained with DAPI and visualized under a fluorescence microscope.

Microarray expression profiling

3 × 10⁵ HUVECs were seeded in a 6-well plate, with approximately 80% confluence at the start of the experiment. Then, HUVEC samples from NETs-treated and untreated groups were collected, and total RNA was extracted using a TRIzol reagent (Sigma, Saint Louis, MO, USA, Cat:322806). Transcriptomic microarray sequencing was performed by Agilent Technologies Inc (Shanghai, China). Differentially expressed genes (DEGs) with a fold change > 2 and P < 0.05 between the control and NETs-treated groups were selected for further analysis. The raw relative expression levels of the selected DEGs were normalized and illustrated using a heatmap.

Western blot assays

2 × 10⁶ HUVECs were seeded in 150 × 25 mm culture dishes, with approximately 80% confluence before the start of the experiment. Then, treated or untreated cells were harvested and lysed with RIPA buffer (Beyotime, Shanghai, China, Cat: P0013B) containing PMSF (1 mM, Beyotime, Shanghai, China, Cat: ST506) and a phosphatase inhibitor cocktail (1 mM, Roche, Basel, Switzerland, Cat: 4906845001). 30 µg of proteins was subjected to SDS-PAGE and then transferred to PVDF membranes (Millipore, Massachusetts, USA). After being blocked with 5% BSA (Solarbio, Beijing, China, Cat: SW3015), the membranes were incubated with primary antibodies overnight at 4 °C, followed by horseradish-peroxidase-conjugated secondary antibodies (1:5000, ZSGB-BIO, Beijing, China, Cat: SP9001). Primary antibodies including anti-phospho-PI3K (CST, Massachusetts, USA, Cat: #17366), anti-PI3K (CST, Massachusetts, USA, Cat: #4292), anti-phospho-AKT (CST, Massachusetts, USA, Cat: #4060), anti-AKT (CST, Massachusetts, Cat: #9272), anti-phospho-p38 (CST, Massachusetts, USA, Cat: #4511), anti-p38 (Proteintech Group, Chicago, IL, USA, Cat: #9212) and anti-β-actin (ZSGB-BIO, Beijing, China, Cat: ZB-5301) were used at a dilution of 1:1000. Specific bands were determined by ECL (Seven, Beijing, China, Cat: SW134-01) using ChemiDocTM MP imaging system (Bio-Rad, Berkeley, CA, USA).

Determination of intracellular ROS

2 × 10⁶ HUVECs were seeded in 150 × 25 mm culture dishes, with approximately 80% confluence before the start of the experiment. Then, HUVECs were treated with indicated treatments. Then the cells were loaded with the fluorescence probe 2,7-dichlorodihydrofluorescein diacetate (DCFH-DA, 10 µM, Beyotime, Shanghai, China, Cat: S1105S). After 3 washes, DCF relative fluorescence intensity in cells was quantified with flow cytometry (Becton Dickinson FACSCanto™ II, USA). Data were obtained in BD FACSDiva Software and analyzed in FlowJo X Software.

Caspase-1 activity assay

Caspase-1 activity in HUVECs was determined using a Caspase-1 activity assay kit (Beyotime, Shanghai, China, Cat: C1101) according to the manufacturer’s instructions. Briefly, 2 × 10⁶ HUVECs were seeded in 150 × 25 mm culture dishes, with approximately 80% confluence before the start of the experiment. After treatment, HUVECs were lysed, centrifuged and the supernatants were taken as the test samples. Then, the standard solution or the samples was pipetted into a 96-well plate, and the Ac-YVAD-pNA was added. After being incubated at 37 °C for 1–2 h, the optical density was obtained with a microplate reader at a wavelength of 405 nm. Results were calculated according to the standard curve.

Reverse transcription-quantitative polymerase chain reaction(RT-qPCR)

3 × 10⁵ HUVECs were seeded in a 6-well plate, with approximately 80% confluence at the start of the experiment. After treatment, total RNA was extracted with a TRIzol reagent. The cDNAs were synthesized using a Transcriptor First Strand cDNA Synthesis Kit (Roche, Basel, Switzerland, Cat: 04896866001) according to the manufacturer’s instructions. RT-qPCR was performed using a Bio-Rad system (Bio-Rad, Berkeley, CA, USA). The primer sequences were shown in Table 1.

Table 1.

Primer sequences

Names	Sequences (5′-3′)
S100A8	F: 5’-AAAGCCTTGAACTCTATCA-3’
	R: 5’-ACTGAGGACACTCGGTCT-3’
S100A9	F: 5’-GGTCATAGAACACATCATGGAGG- 3’
	R: 5’-GGCCTGGCTTATGGTGGTG-3’
GAPDH	F: 5’-CATGTTCGTCATGGGTGTGAA-3’
	R: 5’-GGCATGGACTGTGGTCATGAG-3’

Enzyme-linked immunosorbent assay (ELISA)

S100A8/A9 heterodimer, TXB₂ or 6-keto-PGF1α level was detected by ELISA double antibody sandwich method using a Human S100A8/A9 heterodimer (Bo Shen, Jiangsu, China, Cat: BS-E8588M2), Human TXB₂ (Bo Shen, Jiangsu, China, Cat: GOY-01E6630), Rat TXB₂ (Bo Shen, Jiangsu, China, Cat: ZK-R3210) or Rat 6-keto-PGF1α (Bo Shen, Jiangsu, China, Cat: BJ-R6419) kit according to the manufacturer’s instructions. The optical density was obtained with the Bio-Rad microplate reader at a wavelength of 450 nm. Results were calculated according to the standard curve.

Transmission electron microscopy (TEM)

Platelets with or without treatment were harvested and fixed in 2.5% glutaraldehyde (Yuanye Bio-Technology, Shanghai, China, Cat: R20510) at 4℃ overnight. The precipitation was washed with PBS and fixed with OsO₄ at 4℃ for 2 h. Then the precipitation was dehydrated with acetone of gradient concentration. The specimen was embedded in EPON 812 and cured for 24 h to prepare ultrathin sections of 70 nm. The sections were stained with both uranyl acetate and lead citrate and then observed with TEM (Hitachi, Tokyo, Japan).

Platelet adhesion test (PAdT)

For platelet adhesion test, the experiments were performed as described earlier with minor modifications [22]. Briefly, the slides were coated with collagen (50 µg/mL, Solarbio, Beijing, China, Cat: C8062) at 4 °C overnight. The slides were washed with PBST and then blocked with 50% goat serum. Platelets (2 × 10⁷) with or without treatment were seeded on coverslips for 1 h at 37 °C. After 3 washes to remove nonadherent platelets, the platelets were fixed with 4% paraformaldehyde and washed with PBST. The images were captured from a LSCM.

Platelet aggregation test (PAgT)

Platelet rich plasma (PRP) was prepared as described [21]. Then platelet aggregating agents, adenosine diphosphate (ADP, 10 µM, Helena Laboratories, Texas, USA, Cat: AG001K), arachidonic acid (AA, 500 µg/mL, Helena Laboratories, Texas, USA, Cat: 5369) or epinephrine (EPI, 500 µM, Helena Laboratories, Texas, USA, Cat: AG003K) was added to PRP. The platelet aggregation ratios were real-time monitored with turbidimetry in a Helena Platelet Aggregation System (Helena Laboratories, Texas, USA) under continuous stirring at 250 g at 37 ℃.

Detection of CD62P expression on platelets

Treated or untreated platelets were collected and 20 µL of isotype contrast IgG labeled with PE (BioLegend, San Diego, CA, USA, Cat: #403004) or anti-human CD62P antibody labeled with PE (BioLegend, San Diego, CA, USA, Cat: #304905) was added. After incubated at room temperature in the dark for 20 min, the platelets were fixed with 1% paraformaldehyde. The positive rate of CD62P on platelets were quantified with flow cytometry. Data were obtained in BD FACSDiva Software and analyzed in FlowJo X Software.

Platelet spreading on immobilized fibrinogen

Platelet spreading on immobilized fibrinogen was performed as previously described [23]. Briefly, the slides were coated with fibrinogen (50 µg/mL, Solarbio, Beijing, China, Cat: F8050) at 4 °C overnight. Then the slides were blocked with 50% goat serum. Platelets (2 × 10⁷) with or without treatment were seeded on coverslips for specified time intervals at 37 °C. Platelets were then fixed with 4% paraformaldehyde, permeabilized with 0.5% Triton X-100 and incubated with phalloidin labeled with Alexa Fluor 488 (1 µg/mL, Beyotime, Beijing, China, Cat: C2201S) to label the cytoskeleton of platelets. The images were captured from a LSCM.

Clot Retraction

For clot retraction, the experiments were performed as described earlier with minor modifications [23]. Briefly, platelets (3 × 10⁸/mL) were incubated with corresponding treatments. After that, 400 µL of suspension of platelets was pipetted into the cuvettes and then 0.2 U/mL Thrombin (Solarbio, Beijing, China, Cat: T8021) and 400 µg/mL Fibrinogen were added. The suspension of platelets was evenly mixed and incubated at 37 °C. The images at specified time intervals of 0, 20, 40 and 60 min were captured with a digital camera.

FeCl₃-injured carotid artery thrombosis model

SD rats (male, ~ 250 g) were purchased from Harbin Medical University. Ethical approval was from the ethical committees of Harbin Medical University (Number: YJSDW2022-211). The study protocol was implemented in accordance with international guidelines on animal experiments. In selected rats (n = 6), the tail vein injection of recombinant S100A8/A9 heterodimer (Sino Biological, Beijing, China, Cat: CT002-H0822B) at the appropriate dose of 20 µg was administered 30 min before surgery. Then the FeCl₃-injured carotid artery thrombosis model was conducted [20]. Briefly, rats were anesthetized via i.p.-injection with 2% sodium pentobarbital (0.2 mL/100 g), and the right carotid artery was isolated. A 1cm² patch of filter paper pre-soaked in 10% FeCl₃ was attached on the surface of the outer membrane. The sections were stained with a hematoxylin-eosin (HE, Beyotime, Shanghai, China, Cat: C0105S) or a Masson’s trichrome kit (Solarbio, Beijing, China, Cat: C0189S) according to standard procedures. Histological analysis was performed using a light microscope.

Immunohistochemical (IHC) assay

After deparaffinization, the sections were incubated with 0.3% H₂O₂ solution to block endogenous catalase. Antigens were retrieved and primary antibodies against GPIbα, phospho-PI3K (1:100, Immunoway, Plano, TX, USA, Cat: YP0224), phospho-AKT (1:100, Immunoway, Plano, TX, USA, Cat: YM0224) and phospho-p38 (1:100, Immunoway, Plano, TX, USA, Cat: YP0338) were used for IHC staining. The sections were then incubated with horseradish-peroxidase-conjugated secondary antibodies. Finally, the slides were stained with 3,3’-diaminobenzidine (DAB substrate kit, ZSGB-bio, Beijing, China, Cat: ZLI-9017) and counterstained with hematoxylin. Histological analysis was performed using a light microscope.

Statistical analysis

Data were presented as mean ± SEM in at least three independent experiments. The data obtained were first examined by analysis of variance (ANOVA). Then, Dunnet`s post hoc test for comparisons between control and experimental groups, and Tukey`s post hoc test for multiple group comparisons were performed. Statistical analysis was performed using SPSS (version 22.0, SPSS, Chicago, IL, USA) and Graphpad Prism (version 8.0.2, Graphpad Prism, Chicago, IL, USA) statistical software. P < 0.05 was considered statistically significant.

Results

NETs induce pyroptosis and increase S100A8/A9 heterodimer expression in HUVECs

Neutrophils were isolated from EDTA-anticoagulant peripheral blood of healthy volunteers by the technique of dextran sedimentation (Fig. 1A). The viability of the neutrophils as assessed by Trypan blue stain was (95.63 ± 0.65)%. The purity of the neutrophils as assessed by Giemsa stain was (98.33 ± 0.31)%. Next, NETs were induced with phorbol myristate acetate (PMA, 50 nM, Sigma, Saint Louis, MO, USA, Cat: 524400) and confirmed by IF (Fig. 1B). We performed a transcriptomic analysis by microarray gene expression profiling to examine the effects of NETs on HUVECs. The heatmap analysis revealed that S100A8 and S100A9, NF-κB, NLRP3 and IL-1β were the significantly upregulated genes after NETs treatment, indicating S100A8/A9 release and a potential role in pyroptosis regulation (Fig. 1C). To explore the potential effect of NETs on HUVECs, we co-cultured neutrophils and BSA, NETs (100 µg/mL) or NETs (100 µg/mL) and DNase I (100 U/ml, Sigma, Saint Louis, MO, USA, Cat: AMPD1) with HUVECs for 24 h. In PCR (Fig. 1D and E) and ELISA analysis (Fig. 1F), S100A8/A9 heterodimer, which functions as alarmins in inflammation and blood coagulation process [24] was found significantly elevated in the NETs group, and this effect was abolished in the NETs and DNase I group. Pyroptosis was traced by detecting GSDMD enrichment on the cell membrane through IF staining (Fig. 1G).

Fig. 1.

NETs induce pyroptosis and increase S100A8/A9 heterodimer expression in HUVECs (A) Giemsa and Trypan blue staining of isolated neutrophils from healthy volunteers. (B) Representative images of IF staining of the confirmation of NETs, NETs were confirmed by neutrophil elastase (NE) binding on the DNA filaments. (C) Heatmap showing the DEGs with the most significant upregulated genes after NETs treatment. (D, E) The RNA level of S100A8 (D) and S100A9 (E) in HUVECs co-cultured with BSA and neutrophils, NETs or NETs and DNase I. (F) S100A8/A9 heterodimer secretion detected with ELISA in HUVECs co-cultured with BSA and neutrophils, NETs or NETs and DNase I. (G)Representative images of IF staining of IL-1β (Green) and GSDMD (Red) in HUVECs co-cultured with BSA and neutrophils, NETs or NETs and DNase I. n = 3 independent experiments. Significant results are presented as ^**P<0.01, or ^***P<0.001.

NETs induce pyroptosis and increase S100A8/A9 heterodimer expression in HUVECs via ROS/NF-κB/Caspase-1 axis

To evaluate the effect of NETs on ROS generation in HUVECs, we performed flow cytometry and found that intracellular ROS levels in HUVECs incubation with NETs (100 µg/mL) for 24 h were significantly elevated and targeting degradation of NETs inhibited this effect (Fig. 2A). As reported, NETs can induce cell pyroptosis [25]. Because ROS/NF-κB/Caspase-1 axis plays a crucial role in mediating pyroptosis [26], we proposed that NETs might trigger pyroptosis of HUVECs via ROS/NF-κB in a Caspase-1 dependent manner. To further investigate the involvement of the ROS/NF-κB/Caspase-1 axis in NETs induced cell pyroptosis, we detected the activation of NF-κB and Caspase-1 activity. IF staining results demonstrated that NETs enhanced nuclear translocation of NF-κB (Fig. 2B). We also found that NETs (100 µg/mL) incubation for 24 h increased the phosphorylation levels of NF-κB (Fig. 2C). Moreover, our data revealed that the activity of Caspase-1 increased under NETs (100 µg/mL) incubation for 24 h, and this effect was abrogated with NETs digestion with DNase I (Fig. 2D). In addition, we searched for transcription factors for the S100A8 and S100A9 genes using the Gene Transcription Regulation Database (GTRD). Our predictions revealed that NF-κB is a transcription factor for both S100A8 and S100A9 genes indicating that NF-κB could bind to the promoter of S100A8/A9 heterodimer, which might result in S100A8/A9 heterodimer over expression in HUVECs.

Fig. 2.

NETs induce pyroptosis and increase S100A8/A9 heterodimer expression in HUVECs via ROS/NF-κB/Caspase-1 axis (A) ROS activity in HUVECs co-cultured with BSA and neutrophils, NETs, NETs and DNase I or NETs and a ROS scavenger (DPI, 10 μM, MedChem Express, Monnouth Junction, NJ, USA, Cat: HY-100965). LPS was used as positive control. (B) Representative images of IF staining of NF-κB p65 in HUVECs co-cultured with BSA and neutrophils, NETs, NETs and DNase I or NETs and DPI. (C) Expression of p-p65 and p65 in HUVECs co-cultured with BSA and neutrophils, NETs, NETs and DNase I or NETs and DPI. (D) Caspase-1 activity in HUVECs co-cultured with BSA and neutrophils, NETs, NETs and DNase I or NETs and DPI. n = 3 independent experiments. Significant results are presented as ^*P<0.01, ^**P<0.001, or ^***P<0.001.

S100A8/A9 heterodimer promotes platelet adhesion and aggregation via EGFR/PI3K/AKT and EGFR/p38 MAPK axes

Resting platelets circulate in the bloodstream and become activated when the vessel wall is damaged and exposed to subendothelial collagen [27]. The roles of S100A8/A9 heterodimer and EGFR signaling pathway in platelet shape, adhesion and aggregation were investigated using recombinant S100A8/A9 heterodimer (Sino Biological, Beijing, China, Cat: CT002-H0822B) or specific inhibitors EGFR inhibitor (AG1478, 10 µM, MedChem Express, Monnouth Junction, NJ, USA, Cat: HY-13524) and PI3K inhibitor (LY294002, 10 µM, MedChem Express, Monnouth Junction, NJ, USA, Cat: HY-10108). TEM results showed that, compared with the vehicle group, S100A8/A9 heterodimer treatment group had larger volume, more pseudopodia, and more high-density electron nuclei were observed in the α-granules. AG1478 and LY294002 treatment alleviated the morphological changes of platelets (Fig. 3A). The result of platelet adhesion test in vitro indicated that the number of platelets adsorbing on the collagen surface increased largely in S100A8/A9 heterodimer treatment group, and the pharmacological preconditioning of AG1478 and LY294002 markedly inhibited this effect (Fig. 3B). S100A8/A9 heterodimer alone could not induce platelet aggregation in washed human platelets [28], but enhanced platelet aggregation induced by ADP, AA or EPI. As expected, while in the presence of AG1478 and LY294002, the enhancing effect of S100A8/A9 heterodimer on platelet aggregation was abolished (Fig. 3C). Next, platelets were harvested and lysed from the reaction cells of the platelet aggregometer, and proteins were extracted for subsequent assays. We found that the phosphorylation of p38 MAPK and PI3K were suppressed by AG1478, and the phosphorylation of AKT was abrogated by LY294002, suggesting that p38 MAPK and PI3K/AKT are downstream signaling molecules of the EGFR pathway (Fig. 3D and E).

Fig. 3.

S100A8/A9 heterodimer promotes platelet adhesion and aggregation via EGFR/PI3K/AKT and EGFR/p38 MAPK axes (A) Representative images of the morphological changes observed by TEM in washed human platelets treated with vehicle (DMSO), 10 µg/mL S100A8/A9 heterodimer, AG1478 or LY294002. (B) Representative images of the platelet adhesion test of washed human platelets treated with vehicle (DMSO), 1 µg/mL S100A8/A9 heterodimer, AG1478 or LY294002. The statistics are shown in the histogram below. (C) Platelet aggregation induced by ADP, AA or EPI in PRP treated with vehicle (DMSO), 1 µg/mL S100A8/A9 heterodimer, AG1478 or LY294002. The statistics are shown in the histogram below. (D) Expression of p-p38 and p38 in the platelets harvested from the reaction cells of the platelet aggregometer. The statistics are shown in the histogram below. (E) Expression of p-PI3K, PI3K, p-AKT and AKT in the platelets harvested from the reaction cells of the platelet aggregometer. The statistics are shown in the histogram below. n = 3 independent experiments. Significant results are presented as ^*p<0.05, ^**P<0.01, or ^***P<0.001.

S100A8/A9 heterodimer promotes platelet αⅡbβ3 activation and TXB₂ secretion via EGFR signaling pathway

Integrin αⅡbβ3 is the most abundant receptor on the platelet surface and regulates platelet functional responses by mediating outside-in signal transduction [29], and the activation of integrin αⅡbβ3 is crucial for the growth of occlusive thrombus [30]. In the present study, we found that the activation of platelet integrin αⅡbβ3 measured by CD62P expression on platelet surface was potentiated by S100A8/A9 heterodimer in washed human platelets stimulated with ADP, and this effect was abolished by AG1478 and LY294002 (Fig. 4A). We further examined platelet spreading, which is an early-phase inside-out signaling event, and found that S100A8/A9 heterodimer promoted human platelet spreading on immobilized fibrinogen for a certain time, while this effect was abolished by AG1478 and LY294002(Fig. 4B). In agreement with this effect, we found that clot retraction, which is a late-phase outside-in signaling event was dramatically accelerated by S100A8/A9 heterodimer in the time frame elevated, while this effect was abolished by AG1478 and LY294002 (Fig. 4C). Platelet recruitment is often accompanied by local accumulation of TXA₂ [31]. TXA₂ is synthesized and released by platelets at sites of vascular injury, thus enhancing platelet activation. TXB₂ is the stable metabolite of TXA₂ [32]. We found that S100A8/A9 heterodimer increased TXA₂ synthesis in washed human platelets stimulated by ADP as measured by TXB₂. Similarly, this effect was abolished by AG1478 and LY294002 (Fig. 4D).

Fig. 4.

S100A8/A9 heterodimer promotes platelet αⅡbβ3 activation and TXB₂ secretion via EGFR signaling pathway (A) Activity of CD62P in washed human platelets treated with Vehicle (DMSO), ADP, ADP + S100A8/A9, ADP + S100A8/A9 + AG1478 or ADP + S100A8/A9 + LY294002. The statistics are shown in the histogram below. (B) Representative phalloidin-stained images of washed human platelets treated with vehicle (DMSO), 1 µg/mL S100A8/A9 heterodimer, AG1478 or LY294002 spreading on immobilized fibrinogen for every 20 min. The statistics are shown in the histogram below. (C) Clot retraction of human platelets treated with vehicle (DMSO), 1 µg/mL S100A8/A9 heterodimer, AG1478 or LY294002. The statistics are shown in the histogram below. (D) TXB₂ secretion detected with ELISA in human platelets treated with Vehicle (DMSO), ADP, ADP + S100A8/A9, AG1478 or LY294002. n = 3 independent experiments. Significant results are presented as ^*P<0.05, ^**P<0.01, or ^***P<0.001. Non-significant results are presented as ns.

S100A8/A9 heterodimer enhances carotid artery thrombus formation in vivo

To exclude the effects of S100A8/A9 heterodimer on in vivo thrombosis, SD rats aged 10 weeks after intravenous injection of S100A8/A9 heterodimer (Sino Biological, Beijing, China, Cat: CT002-H0822B) were studied. We detected the occlusion time of rat common carotid artery by laser speckle contrast imaging (LSCI) technique and found that S100A8/A9 heterodimer shorten the carotid artery occlusion time prominently (Fig. 5A). To evaluate the effect of S100A8/A9 heterodimer on the histopathologic changes in the rat carotid artery, we performed HE and Masson trichrome staining. HE staining revealed that the vascular intima was intact, and there was no thrombus in the lumen in the sham surgery group. While in other groups, the vascular intima was injured, and mixed thrombus was observed. Compared with the model group, thrombus occupying area was clearly increased in the S100A8/A9 group. Masson staining showed that, compared with the sham surgery group, irregular distribution of blue-stained collagen fiber was observed in the other groups, and the content of collagen fiber was more abundant in the S100A8/A9 group (Fig. 5B). GPIbα is a specific marker of platelets and can be used to evaluate platelet infiltration [33]. IF and IHC staining proved that the expression of GPIbα in the S100A8/A9 group was significantly higher than that in the other groups (Fig. 5C). IHC staining also demonstrated that S100A8/A9 heterodimer could increase the downstream PI3K, AKT and p38 MAPK phosphorylation, which was consistent with our in vitro experiments (Fig. 5D). In addition, we found that the ratio of TXB₂/6-keto-PGF1α in the S100A8/A9 group were obviously higher than that of other groups (Fig. 5E). Taken together, our data provide new insights into the pathogenesis and therapeutic targets for AT (Fig. 6).

Fig. 5.

S100A8/A9 heterodimer enhances carotid artery thrombus formation in vivo (A) Representative images of carotid artery blood flow change at different time points in rats under indicated treatments. Color change from red to blue represented a change of blood perfusion from 100% to 0%. The statistics are shown in the histogram right. (B) Representative images of HE and Masson staining of the carotid artery in rats under indicated treatments. (C) Representative images of IHC and IF staining of GPIbα in rats under indicated treatments. (D) Representative images of IHC staining of p-p38, p-PI3K and p-AKT in rats under indicated treatments. (E) TXB₂ and 6-keto-PGF1α secretion detected with ELISA in serum of rats under indicated treatments. n = 6 independent experiments. Significant results are presented as ^*p<0.05, ^**P<0.01 or ^***P<0.001. Non-significant results are presented as ns.

Fig. 6.

The role of S100A8/A9 heterodimer produced by endothelial cells stimulated by NETs in thrombosis is shown. Neutrophil extracellular traps (NETs) induce pyroptosis and increase S100A8/A9 heterodimer expression in HUVECs via ROS/NF-κB/Caspase-1 axis and S100A8/A9 heterodimer activates platelets via the EGFR/PI3K/AKT and EGFR/p38 MAPK signaling pathways, driving thrombus formation.

Discussion

Thrombotic diseases are the leading cause of death worldwide and their incidence has been increasing in recent years, posing a serious threat to human health. Atherosclerotic thrombosis is known as the main cause of thrombotic diseases that occurs suddenly, commonly without any warning signs [34]. Atherosclerotic vascular disease involves multiple processes, including endothelial cell dysfunction, intimal injury, and enhanced immune responses, which together lead to the formation of atherosclerotic plaques [35]. When unstable plaques rupture or erode, the necrotic core is exposed to the circulation, leading to the activation of tissue factor and the recruitment of inflammatory cells and platelets, leading to thrombus formation.

Neutrophils are highly differentiated cells that participate in inflammatory responses through a range of coordinated functions, including chemotaxis and respiratory burst [36, 37]. AS is a chronic inflammatory process in which neutrophils are activated and NETs, the mesh-like structures are released into the bloodstream. As reported, elevated levels of NETs components such as cell-free DNA, nucleosomes, and NE in plasma are positively correlated with cardiovascular events [38]. In our previous study, we found that anti-β₂GPⅠ/β₂GPⅠ-induced NETs promoted platelet aggregation thus enhancing thrombosis, and this effect was abrogated by DNase I [20]. However, the role of NETs in AT has not been fully elucidated. For this purpose, we isolated human neutrophils to induce and isolate NETs for further investigations. After IF identification, the isolated NETs contained DNA filaments and NE, indicating that we successfully isolated PMA-induced NETs in vitro.

Endothelial cell dysfunction is a key event in cardiovascular diseases [39]. Normal endothelial cells express numerous molecules with antiplatelet, anticoagulant and fibrinolytic properties, which contributes to limiting the occurrence of thrombosis and ischemic events [40]. However, when the plaque ruptures, platelets attach to the surface of damaged endothelial cells, eventually forming a platelet-rich thrombus [41]. NETs have been reported to be closely related to vascular inflammation, endothelial cell damage, and thrombosis in T2DM [42]. Excessive ROS produced by endothelial cells have usually been implicated in several inflammatory diseases [43, 44]. Increased intracellular ROS can conversely aggravate vascular endothelial damage [45], which is the main initiator of the coagulation cascade and play a major role in arterial thrombosis [46]. In the present study, we found that NETs can induce pyroptosis of endothelial cells, and the possible mechanism might relate to pyroptosis induced by the ROS/NF-κB/Caspase-1 axis activation. This process results in massive amplification of the inflammatory process. Meanwhile, S100A8/A9 heterodimer, an inflamed mediator was found remarkably elevated.

S100A8 and S100A9 belong to the S100 family of calcium-binding proteins. S100A8/A9 heterodimers are the main forms of S100A8 and S100A9 in serum and can function as damage-related pattern molecules by binding to pattern recognition receptors. Clinical studies have found that S100A8/A9 heterodimer is an early marker for detecting acute coronary syndrome [16], and in the PROVE IT-TIMI 22 trial [47], the likelihood of recurrence of cardiovascular events increased with each level increase in the quartiles of S100A8/A9 heterodimer. In one study in mice, S100A9^−/− mice had reduced neointimal thickening and thrombus area after femoral artery injury, and reduced monocyte and neutrophil infiltration were observed at the lesion site [48]. In vitro experiments confirmed that exogenous S100A8/A9 heterodimer can damage vascular endothelial cells, recruit neutrophils and activate monocytes to promote thrombosis [49]. However, relatively few studies have examined the interaction between S100A8/A9 heterodimer and platelets, the cells in human peripheral blood that most directly contribute to thrombosis. In order to clarify the effect and mechanism of S100A8/A9 heterodimer on platelets, we previously performed HitPredict protein interaction prediction, and the results suggested that there is an interaction between S100A9 and EGFR.

The most well-known effectors of EGFRs are PI3Ks and MAPKs, which have shown to be common and key signaling mediators in platelet activation [50, 51]. The PI3K/AKT signaling pathway is an important axis for inducing platelet granule release and is crucial for platelet activation and stable platelet adhesion induced by intracellular signal transduction pathway of GPIb-IX, integrin αⅡbβ3 and ITAM [52]. LDL has been reported to activate the p38 MAPK signaling pathway in platelets and increase platelet reactivity, which may promote the progression of AT [53]. What is more, our previous studies have demonstrated that p38 MAPK may be a therapeutic target for anti-β₂GPI/β₂GPI-related thrombotic events [21]. In this study, we elucidated the mechanisms by which S100A8/A9 heterodimer promote platelet adhesion, aggregation, granules release, and integrin αⅡbβ3 “inside-out” and “outside-in” activation through EGFR/PI3K/AKT and EGFR/p38 MAPK signaling pathways, thus promoting platelet activation. The previous sections have studied in depth the cellular origin and mechanism of action of S100A8/A9 heterodimer in vitro. In line with those observations, using a FeCl₃-injured carotid thrombosis model in rats and recombinant S100A8/A9 heterodimer, we showed that S100A8/A9 heterodimer induced the phosphorylation of PI3K, AKT and p38 MAPK, which would thus promote platelet infiltration at the lesion site and thrombus formation.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

JF and YL designed this study; JF performed the experiments; HZ was involved in data curation and formal analyses; JF and YQ wrote the manuscript; YL critically revised the manuscript for important intellectual content, approved the final manuscript version to be published and agreed to be accountable for all aspects of the work. JF and HZ checked and approved the authenticity of all the raw data. All authors read and approved the final manuscript.

Funding

This work was supported by the National Natural Science Foundation of China (grant no. 82270889 and 81974108 to YL).

Data availability

No datasets were generated or analysed during the current study.

Declarations

Competing interests

The authors declare no competing interests.

Ethics approval

Ethical approval was obtained from the Ethics Committee of Harbin Medical University (Number: YJSKY2022-394 and Number: YJSDW2022-211).